Ansys Commands
Ansys Commands
Ansys Commands
1
]
1
1
1
1
1
1
1
1
( , )
( , )
( , )
1
2
0
GG
1
G
2
G
T
A T
A T
( , )
( , )
( , )
0
1
2
1
2
1
]
1
1
1
1
1
1
1
1
1
1
1
1
2
1
]
1
1
1
1
1
1
1
1
The BSAX command, one of several nonlinear general beam section commands, specifies the relationship
of axial strain and axial force for a beam section. The section data defined is associated with the section
most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are BSM1, BSM2, BSTQ, BSS1, BSS2, BSMD, and BSTE.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
226
BSAX
BSMD, DENS, T
Specifies mass density for a nonlinear general beam section.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
DENS
Mass density.
T
Temperature.
Notes
The BSMD command, one of several nonlinear general beam section commands, specifies the mass density
(assuming a unit area) for a beam section. The value specified is associated with the section most recently
defined (via the SECTYPE command).
Related commands are BSAX, BSM1, BSM2, BSTQ, BSS1, BSS2, and BSTE.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
BSM1, VAL1, VAL2, T
Specifies the bending curvature and moment relationship in plane XZ for beam sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1
Curvature component (
1
).
VAL2
Bending moment component (M
1
).
T
Temperature.
Notes
The behavior of beam elements is governed by the generalized-stress/generalized-strain relationship of the
form:
227
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
BSM1
N
M
M
S
S
A T
I T
I T
J
1
2
1
2
E
1
E
2
E
1
]
1
1
1
1
1
1
1
1
( , )
( , )
( , )
1
2
0
GG
1
G
2
G
T
A T
A T
( , )
( , )
( , )
0
1
2
1
2
1
]
1
1
1
1
1
1
1
1
1
1
1
1
2
1
]
1
1
1
1
1
1
1
1
The BSM1 command, one of several nonlinear general beam section commands, specifies the bending
curvature and moment for plane XZ of a beam section. The section data defined is associated with the section
most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are BSAX, BSM2, BSTQ, BSS1, BSS2, BSMD, and BSTE.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
BSM2, VAL1, VAL2, T
Specifies the bending curvature and moment relationship in plane XY for beam sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1
Curvature component (
2
).
VAL2
Bending moment component (M
2
).
T
Temperature.
Notes
The behavior of beam elements is governed by the generalized-stress/generalized-strain relationship of the
form:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
228
BSM2
N
M
M
S
S
A T
I T
I T
J
1
2
1
2
E
1
E
2
E
1
]
1
1
1
1
1
1
1
1
( , )
( , )
( , )
1
2
0
GG
1
G
2
G
T
A T
A T
( , )
( , )
( , )
0
1
2
1
2
1
]
1
1
1
1
1
1
1
1
1
1
1
1
2
1
]
1
1
1
1
1
1
1
1
The BSM2 command, one of several nonlinear general beam section commands, specifies the bending
curvature and moment relationship for plane XY of a beam section. The section data defined is associated
with the section most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are BSAX, BSM1, BSTQ, BSS1, BSS2, BSMD, and BSTE.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
BSPLIN, P1, P2, P3, P4, P5, P6, XV1, YV1, ZV1, XV6, YV6, ZV6
Generates a single line from a spline fit to a series of keypoints.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1, P2, P3, . . . , P6
Keypoints through which a spline is fit. At least two keypoints must be defined. If P1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI).
The following fields are used only if specified end slopes are desired; otherwise, zero curvature end
slopes will be automatically calculated. The slope vector is parallel to a vector pointing from the origin
of the active coordinate system [CSYS] to the position in space that XV, YV, ZV represents in that system.
XV1, YV1, ZV1
Orientation point of an outward vector tangent to line at P1. Vector coordinate system has its origin at
the keypoint. Coordinate interpretation corresponds to the active coordinate system type, i.e., X is R for
cylindrical, etc. Defaults to zero curvature slope.
XV6, YV6, ZV6
Orientation point of an outward vector tangent to a line at P6 (or the last keypoint specified if fewer
than six specified). Defaults to zero curvature slope.
229
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
BSPLIN
Notes
One line is generated between keypoint P1 and the last keypoint entered. The line will pass through each
entered keypoint. Solid modeling in a toroidal coordinate system is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Splines>Spline thru KPs
Main Menu>Preprocessor>Modeling>Create>Lines>Splines>Spline thru Locs
Main Menu>Preprocessor>Modeling>Create>Lines>Splines>With Options>Spline thru KPs
Main Menu>Preprocessor>Modeling>Create>Lines>Splines>With Options>Spline thru Locs
BSS1, VAL1, VAL2, T
Specifies the transverse shear strain and force relationship in plane XZ for beam sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1
Transverse shear strain component (
1
).
VAL2
Transverse shear force component (S
1
).
T
Temperature.
Notes
The behavior of beam elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
M
S
S
A T
I T
I T
J
1
2
1
2
E
1
E
2
E
1
]
1
1
1
1
1
1
1
1
( , )
( , )
( , )
1
2
0
GG
1
G
2
G
T
A T
A T
( , )
( , )
( , )
0
1
2
1
2
1
]
1
1
1
1
1
1
1
1
1
1
1
1
2
1
]
1
1
1
1
1
1
1
1
The BSS1 command, one of several nonlinear general beam section commands, specifies the transverse
shear strain and transverse shear force relationship for plane XZ of a beam section. The section data defined
is associated with the section most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are BSAX, BSM1, BSM2, BSTQ, BSS2, BSMD, and BSTE.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
230
BSS1
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
BSS2, VAL1, VAL2, T
Specifies the transverse shear strain and force relationship in plane XY for beam sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1
Transverse shear strain component (
2
).
VAL2
Transverse shear force component (S
2
).
T
Temperature.
Notes
The behavior of beam elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
M
S
S
A T
I T
I T
J
1
2
1
2
E
1
E
2
E
1
]
1
1
1
1
1
1
1
1
( , )
( , )
( , )
1
2
0
GG
1
G
2
G
T
A T
A T
( , )
( , )
( , )
0
1
2
1
2
1
]
1
1
1
1
1
1
1
1
1
1
1
1
2
1
]
1
1
1
1
1
1
1
1
The BSS1 command, one of several nonlinear general beam section commands, specifies the transverse
shear strain and transverse shear force relationship for plane XY of a beam section. The section data defined
is associated with the section most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are BSAX, BSM1, BSM2, BSTQ, BSS1, BSMD, and BSTE.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
231
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
BSS2
BSTE, ALPHA, T
Specifies a thermal expansion coefficient for a nonlinear general beam section.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
ALPHA
Coefficient of thermal expansion for the cross section.
T
Temperature.
Notes
The BSTE command, one of several nonlinear general beam section commands, specifies a thermal expansion
coefficient for a beam section. The value specified is associated with the section most recently defined (via
the SECTYPE command).
Related commands are BSAX, BSM1, BSM2, BSTQ, BSS1, BSS2, and BSMD.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
BSTQ, VAL1, VAL2, T
Specifies the cross section twist and torque relationship for beam sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1
Twist component ().
VAL2
Torque component ().
T
Temperature.
Notes
The behavior of beam elements is governed by the generalized-stress/generalized-strain relationship of the
form:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
232
BSTE
N
M
M
S
S
A T
I T
I T
J
1
2
1
2
E
1
E
2
E
1
]
1
1
1
1
1
1
1
1
( , )
( , )
( , )
1
2
0
GG
1
G
2
G
T
A T
A T
( , )
( , )
( , )
0
1
2
1
2
1
]
1
1
1
1
1
1
1
1
1
1
1
1
2
1
]
1
1
1
1
1
1
1
1
The BSTQ command, one of several nonlinear general beam section commands, specifies the cross section
twist and torque relationship for a beam section. The section data defined is associated with the section
most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are BSAX, BSM1, BSM2, BSS1, BSS2, BSMD, and BSTE.
For complete information, see Using Nonlinear General Beam Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>NL Generalized
BTOL, PTOL
Specifies the Boolean operation tolerances.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
PTOL
Point coincidence tolerance. Points within this distance to each other will be assumed to be coincident
during Boolean operations. Loosening the tolerance will increase the run time and storage requirements,
but will allow more Boolean intersections to succeed. Defaults to 0.10E-4.
Command Default
PTOL = 0.10E-4.
Notes
Use BTOL,DEFA to reset the setting to its default value. Use BTOL,STAT to list the status of the present setting.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Settings
233
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
BTOL
BUCOPT, Method, NMODE, SHIFT, LDMULTE
Specifies buckling analysis options.
SOLUTION: Nonlinear Options
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
Method
Mode extraction method to be used (no default; you must specify the method). Currently, Block Lanczos
is the only available method:
LANB
Block Lanczos.
NMODE
Number of modes to extract (defaults to 1).
SHIFT
Shift point about which the eigenvalues are calculated (defaults to 0.0).
When Method = LANB and LDMULTE = CENTER, the SHIFT value will determine the center value about
which the modes will be extracted. A blank value yields a center value equal to zero (default).
LDMULTE
Upper end of the load multiplier range of interest (defaults to +
( ( ) * ( )) l U l
l
N
1
where U(I) is the degree of freedom (displacement, temperature, etc.) of term (I). The following example is
a set of two constraint equations, each containing three terms:
0.0 = 3.0* (1 UX) + 3.0* (4 UX) + (-2.0)* (4 ROTY)
2.0 = 6.0* (2 UX) + 10.0* (4 UY) + 1.0* (3 UZ)
The first unique degree of freedom in the equation is eliminated in terms of all other degrees of freedom
in the equation. A unique degree of freedom is one which is not specified in any other constraint equation,
coupled node set, specified displacement set, or master degree of freedom set. It is recommended that the
first term of the equation be the degree of freedom to be eliminated. The first term of the equation cannot
contain a master degree of freedom, and no term can contain coupled degrees of freedom. The same degree
of freedom may be specified in more than one equation but care must be taken to avoid over-specification
(over-constraint).
The degrees of freedom specified in the equation (i.e., UX, UY, ROTZ, etc.) must also be included in the
model (as determined from the element types [ET]). Also, each node in the equation must be defined on
an element (any element type containing that degree of freedom will do).
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Constraint Eqn
245
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CE
CECHECK, ItemLab, Tolerance, DOF
Check constraint equations and couplings for rigid body motions.
PREP7: Database
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ItemLab
Item indicating what is to be checked:
CE
Check constraint equations only
CP
Check couplings only
ALL
Check both CE and CP
Tolerance
Allowed amount of out-of-balance for any constraint equation or coupled set. The default value of 1.0e-
6 is usually good.
DOF
Specifies which DOF is to be checked. Default is RIGID, the usual option. Other choices are individual
DOF such as UX, ROTZ, etc. or THERM. The THERM option will check the constraint equations or coupled
sets for free thermal expansions, whereas the individual DOFs check under rigid body motions. ALL is
RIGID and THERM.
Notes
This command imposes a rigid body motion on the nodes attached to the constraint equation or coupled
set and makes sure that no internal forces are generated for such rigid body motions. Generation of internal
forces by rigid body motions usually indicates an error in the equation specification (possibly due to nodal
coordinate rotations). The THERM option does a similar check to see that no internal forces are created by
the equations if the body does a free thermal expansion (this check assumes a single isotropic coefficient
of expansion).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
CECMOD, NEQN, CONST
Modifies the constant term of a constraint equation during solution.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
NEQN
Reference number of constraint equation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
246
CECHECK
CONST
New value of the constant term of equation.
Notes
Other terms of the constraint equation cannot be changed during the solution phase, but must be defined
or changed within PREP7 prior to the solution. See the CE command for details.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Modify ConstrEqn
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Modify ConstrEqn
Main Menu>Solution>Load Step Opts>Other>Modify ConstrEqn
CECYC, Lowname, Highname, Nsector, HIndex, Tolerance, Kmove, Kpairs
Generates the constraint equations for a cyclic symmetry analysis
PREP7: Constraint Equations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lowname
Name of a component for the nodes on the low angle edge of the sector. Enclosed in single quotes.
Highname
Name of a component for the nodes on the high angle edge of the sector. Enclosed in single quotes.
Nsector
Number of sectors in the complete 360 degrees.
HIndex
Harmonic index to be represented by this set of constraint equations. If Hindex is -1, generate constraint
equations for static cyclic symmetry. If HIndex is -2, generate constraint equations for static cyclic
asymmetry.
Tolerance
A positive tolerance is an absolute tolerance (length units), and a negative tolerance is a tolerance relative
to the local element size.
Kmove
0
Nodes are not moved.
1
HIGHNAME component nodes are moved to match LOWNAME component nodes exactly.
Kpairs
0
Do not print paired nodes
1
Print table of paired nodes
247
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CECYC
Notes
The analysis can be either modal cyclic symmetry or static cyclic symmetry.
The pair of nodes for which constraint equations are written are rotated into CSYS,1.
Distributed ANSYS Restriction This command is only supported for modal analysis (ANTYPE,MODAL) in
Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
CEDELE, NEQN1, NEQN2, NINC, Nsel
Deletes constraint equations.
PREP7: Constraint Equations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
NEQN1, NEQN2, NINC
Delete constraint equations from NEQN1 to NEQN2 (defaults to NEQN1) in steps of NINC (defaults to
1). If NEQN1 = ALL, NEQN2 and NINC will be ignored all constraint equations will be deleted.
Nsel
Additional node selection control:
ANY
Delete equation set if any of the selected nodes are in the set (default).
ALL
Delete equation set only if all of the selected nodes are in the set.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Del Constr Eqn
CEINTF, TOLER, DOF1, DOF2, DOF3, DOF4, DOF5, DOF6, MoveTol
Generates constraint equations at an interface.
PREP7: Constraint Equations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
TOLER
Tolerance about selected elements, based on a fraction of the element dimension (defaults to 0.25 (25%)).
Nodes outside the element by more than the tolerance are not accepted as being on the interface.
DOF1, DOF2, DOF3, . . . , DOF6
Degrees of freedom for which constraint equations are written. Defaults to all applicable DOFs. DOF1
accepts ALL as a valid label, in which case the rest are ignored (all DOFs are applied).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
248
CEDELE
MoveTol
The allowed "motion" of a node (see Note below). This distance is in terms of the element coordinates
(-1.0 to 1.0). A typical value is 0.05. Defaults to 0 (do not move). MoveTol must be less than or equal
to TOLER.
Notes
This command can be used to "tie" together two regions with dissimilar mesh patterns by generating con-
straint equations that connect the selected nodes of one region to the selected elements of the other region.
At the interface between regions, nodes should be selected from the more dense mesh region, A, and the
elements selected from the less dense mesh region, B. The degrees of freedom of region A nodes are inter-
polated with the corresponding degrees of freedom of the nodes on the region B elements, using the shape
functions of the region B elements. Constraint equations are then written that relate region A and B nodes
at the interface.
The MoveTol field lets the nodes in the previously mentioned region A change coordinates when slightly
inside or outside the elements of region B. The change in coordinates causes the nodes of region A to assume
the same surface as the nodes associated with the elements of region B. The constraint equations that relate
the nodes at both regions of the interface are then written.
Solid elements with six degrees of freedom should only be interfaced with other six degree-of-freedom
elements. The region A nodes should be near the region B elements. A location tolerance based on the
smallest region B element length may be input. Stresses across the interface are not necessarily continuous.
Nodes in the interface region should not have specified constraints.
Use the CPINTF command to connect nodes by coupling instead of constraint equations. Use the EINTF
command to connect nodes by line elements. See also the NSEL and ESEL commands for selecting nodes
and elements. See the Theory Reference for the Mechanical APDL and Mechanical Applications for a description
of 3-D space used to determine if a node will be considered by this command.
As an alternative to the CEINTF command, you can use contact elements and the internal multipoint constraint
(MPC) algorithm to tie together two regions having dissimilar meshes. See Solid-Solid and Shell-Shell Assem-
blies for more information.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Adjacent Regions
CELIST, NEQN1, NEQN2, NINC, Option
Lists the constraint equations.
PREP7: Constraint Equations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
NEQN1, NEQN2, NINC
List constraint equations from NEQN1 to NEQN2 (defaults to NEQN1) in steps of NINC (defaults to 1). If
NEQN1 = ALL (default), NEQN2 and NINC are ignored and all constraint equations are listed.
Option
Options for listing constraint equations:
249
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CELIST
ANY
List equation set if any of the selected nodes are in the set (default). Only externally-generated
constraint equations are listed.
ALL
List equation set only if all of the selected nodes are in the set. Only externally-generated constraint
equations are listed.
INTE
List internally-generated constraint equations that are associated with MPC-based contact. Constraint
equations are listed only if all the nodes in the set are selected.
CONV
Convert internal constraint equations to external constraint equations. Internal constraint equations
are converted only if all of the nodes in the set are selected.
Notes
This command is valid in any processor. However, the INTE and CONV options are only valid in the Solution
processor after a SOLVE command has been issued.
Menu Paths
Utility Menu>List>Other>Constraint Eqns>All CE nodes selected
Utility Menu>List>Other>Constraint Eqns>Any CE node selected
CENTER, NODE, NODE1, NODE2, NODE3, RADIUS
Defines a node at the center of curvature of 2 or 3 nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Number to be assigned to the node generated at the center of curvature.
NODE1, NODE2, NODE3
Three nodes used to calculated the center of curvature, as described under RADIUS.
RADIUS
Used to control the interpretation of NODE1, NODE2 and NODE3:
0
NODE1, NODE2 and NODE3 lie on a circular arc. The program will calculate the center of curvature
(and radius) (default).
0
NODE1 and NODE2 are the endpoints of an arc, and RADIUS is the radius of curvature. The program
will locate the center of curvature on the NODE3 side of the NODE1-NODE2 line if RADIUS > 0, and
opposite to NODE3 if RADIUS < 0.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>At Curvature Ctr
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
250
CENTER
CEQN
Specifies "Constraint equations" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Constraint Eqns
CERIG, MASTE, SLAVE, Ldof, Ldof2, Ldof3, Ldof4, Ldof5
Defines a rigid region.
PREP7: Constraint Equations
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
MASTE
Retained (or master) node for this rigid region. If MASTE = P, then graphical picking of the master and
slave nodes is enabled (first node picked will be the master node, and subsequent nodes picked will be
slave nodes), and subsequent fields are ignored (valid only in GUI).
SLAVE
Removed (or slave) node for this rigid region. If ALL, slave nodes are all selected nodes.
Ldof
Degrees of freedom associated with equations:
ALL
All applicable degrees of freedom (default). If 3-D, generate 6 equations based on UX, UY, UZ, ROTX,
ROTY, ROTZ; if 2-D, generate 3 equations based on UX, UY, ROTZ.
UXYZ
Translational degrees of freedom. If 3-D, generate 3 equations based on the slave nodes' UX, UY,
and UZ DOFs and the master node's UX, UY, UZ, ROTX, ROTY, and ROTZ DOFs; if 2-D, generate 2
equations based on the slave nodes UX and UY DOFs and the master nodes UX, UY, and ROTZ DOFs.
No equations are generated for the rotational coupling.
RXYZ
Rotational degrees of freedom. If 3-D, generate 3 equations based on ROTX, ROTY, ROTZ; if 2-D,
generate 1 equation based on ROTZ. No equations are generated for the translational coupling.
UX
Slave translational UX degree of freedom only.
251
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CERIG
UY
Slave translational UY degree of freedom only.
UZ
Slave translational UZ degree of freedom only.
ROTX
Slave rotational ROTX degree of freedom only.
ROTY
Slave rotational ROTY degree of freedom only.
ROTZ
Slave rotational ROTZ degree of freedom only.
Ldof2, Ldof3, Ldof4, Ldof5
Additional degrees of freedom. Used only if more than one degree of freedom required and Ldof is
not ALL, UXYZ, or RXYZ.
Notes
Defines a rigid region (link, area or volume) by automatically generating constraint equations to relate nodes
in the region. Nodes in the rigid region must be assigned a geometric location before this command is used.
Also, nodes must be connected to elements having the required degree of freedom set (see Ldof above).
Generated constraint equations are based on small deflection theory. Generated constraint equations are
numbered beginning from the highest previously defined equation number (NEQN) plus 1. Equations, once
generated, may be listed [CELIST] or modified [CE] as desired. Repeat CERIG command for additional rigid
region equations.
This command will generate the constraint equations needed for defining rigid lines in 2-D or 3-D space.
Multiple rigid lines relative to a common point are used to define a rigid area or a rigid volume. In 2-D space,
with Ldof = ALL, three equations are generated for each pair of constrained nodes. These equations
define the three rigid body motions in global Cartesian space, i.e., two in-plane translations and one in-plane
rotation. These equations assume the X-Y plane to be the active plane with UX, UY, and ROTZ degrees of
freedom available at each node. Other types of equations can be generated with the appropriate Ldof labels.
Six equations are generated for each pair of constrained nodes in 3-D space (with Ldof = ALL). These
equations define the six rigid body motions in global Cartesian space. These equations assume that UX, UY,
UZ, ROTX, ROTY, and ROTZ degrees of freedom are available at each node.
The UXYZ label allows generating a partial set of rigid region equations. This option is useful for transmitting
the bending moment between elements having different degrees of freedom at a node. With this option
only two of the three equations are generated for each pair of constrained nodes in 2-D space. In 3-D space,
only three of the six equations are generated. In each case the rotational coupling equations are not generated.
Similarly, the RXYZ label allows generating a partial set of equations with the translational coupling equations
omitted.
Applying this command to a large number of slave nodes may result in constraint equations with a large
number of coefficients. This may significantly increase the peak memory required during the process of
element assembly. If real memory or virtual memory is not available, consider reducing the number of slave
nodes.
As an alternative to the CERIG command, you can define a similar type of rigid region using contact elements
and the internal multipoint constraint (MPC) algorithm. See Surface-Based Constraints for more information.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
252
CERIG
CERIG cannot be deleted using CEDELE,ALL and then regenerated in the second or higher load steps if the
LSWRITE and LSSOLVE procedure is used. CERIG writes constraint equations directly into load step files.
Deleting constraint equations (CEDELE,ALL) cannot always maintain the consistency among load steps.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Rigid Region
CESGEN, ITIME, INC, NSET1, NSET2, NINC
Generates a set of constraint equations from existing sets.
PREP7: Constraint Equations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME, INC
Do this generation operation a total of ITIMEs, incrementing all nodes in the existing sets by INC each
time after the first. ITIME must be >1 for generation to occur.
NSET1, NSET2, NINC
Generate sets from sets beginning with NSET1 to NSET2 (defaults to NSET1) in steps of NINC (defaults
to 1). If NSET1 is negative, NSET2 and NINC are ignored and the last |NSET1| sets (in sequence from
maximum set number) are used as the sets to be repeated.
Notes
Generates additional sets of constraint equations (with same labels) from existing sets. Node numbers
between sets may be uniformly incremented.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Gen w/Same DOF
CFACT, RFACTA, IFACTA, RFACTB, IFACTB, RFACTC, IFACTC
Defines complex scaling factors to be used with operations.
POST26: Controls
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
RFACTA
Real portion of the complex scale factor used in place of FACTA.
IFACTA
Imaginary portion of the complex scale factor used in place of FACTA.
RFACTB
Real portion of the complex scale factor used in place of FACTB.
IFACTB
Imaginary portion of the complex scale factor used in place of FACTB.
253
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CFACT
RFACTC
Real portion of the complex scale factor used in place of FACTC.
IFACTC
Imaginary portion of the complex scale factor used in place of FACTC.
Command Default
Use the real factors as described with the operation command.
Notes
Defines complex scale factors to be used with the operations [ADD, PROD, etc.]. If this command is supplied,
these complex factors override any real factors (FACTA, FACTB, FACTC) supplied on the operation commands.
Factors are typically involved in scaling a specified variable, such as in the term FACTA x IA of the ADD
command to scale variable IA before the ADD operation.
When the CFACT command is active, defaults are as follows: 1) if the complex factor is not specified, but
the variable upon which it acts (such as IA) is specified, the factor defaults to 1.0+i0.0; 2) if the variable
upon which the factor operates is not specified, but the factor is specified, the variable defaults to 1.0 so
that the term in the operation becomes the complex factor itself; 3) if neither the factor nor the variable
number is supplied, the term is omitted from the operation. Once the operation (such as the ADD command)
has been processed, the CFACT command becomes inactive and must be specified again if it is to be used.
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Complx ScaleFact
*CFCLOS
Closes the "command" file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
254
*CFCLOS
*CFOPEN, Fname, Ext, --, Loc
Opens a "command" file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to CMD if Fname is blank.
--
Unused field.
Loc
Determines whether existing file will be overwritten or appended:
(blank) --
The existing file will be overwritten.
APPEND --
The file will be appended to the existing file.
Notes
Data processed with the *VWRITE command will also be written to this file if the file is open when the
*VWRITE command is issued.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
255
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*CFOPEN
*CFWRITE, Command
Writes an ANSYS command (or similar string) to a "command" file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Command
Command or string to be written. The standard command form of a label followed by arguments separated
by commas is assumed. Command may be a parameter assignment (e.g., *CFWRITE, A = 5).
Notes
Writes an ANSYS command (or similar string) to the file opened with *CFOPEN. The Command string is not
executed (except that numeric and character parameter substitution and operations (with imbedded *, /, >,
etc. characters) are performed before writing). When used with *GET results and parameter substitution, an
ANSYS command can be created from results and then read back into the ANSYS program (or used elsewhere).
For example, if the command *CFWRITE,BF,NNUM,TEMP,TVAL is used in a do-loop, where TVAL is a parameter
value returned from the *GET operation and NNUM is a specified or returned parameter value, a series of
BF commands, with numerical values substituted for the two parameters, will be written. To create a file
without parameter substitution, use *CREATE.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
/CFORMAT, NFIRST, NLAST
Controls the graphical display of alphanumeric character strings for parameters, components, assemblies,
and tables.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NFIRST
Display the first n characters of the parameter, component, assembly, or table name, up to 32. Defaults
to 32.
NLAST
Display the last n characters of the parameter, component, assembly, or table name, up to 32. Defaults
to 0.
Notes
Use this command to control the length of the character string that is shown in the graphics window for a
parameter, component, assembly, or table name.
The total number of characters (NFIRST + NLAST +3) cannot exceed 32.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
256
*CFWRITE
If NFIRST is greater than zero and NLAST = 0, only the NFIRST characters are displayed, followed by an
ellipsis.
If NFIRST = 0 and NLAST is greater than zero, only the NLAST characters are displayed, preceded by an
ellipsis (...).
If both NFIRST and NLAST are greater than zero, the name will be shown as NFIRST, followed by an ellipsis
(...), followed by NLAST, up to a maximum of 32 characters.
For example, if NFIRST = 6 and NLAST = 3, and the character string is LENGTHOFSIDEONE, then it will appear
in the graphics window as LENGTH...ONE.
If the actual length of the character string is less than the specified combination of NFIRST + NLAST +3,
then the actual string will be used.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Size and Shape
CGLOC, XLOC, YLOC, ZLOC
Specifies the origin location of the acceleration coordinate system.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
XLOC, YLOC, ZLOC
Global Cartesian X, Y, and Z coordinates of the acceleration coordinate system origin.
Notes
Specifies the origin location of the acceleration coordinate system with respect to the global Cartesian system.
The axes of this acceleration coordinate system are parallel to the global Cartesian axes.
A structure may be rotating about the global Cartesian origin [OMEGA, DOMEGA], which may in turn be
rotating about another point (the origin of the acceleration coordinate system), introducing Coriolis effects.
The location of this point (relative to the global Cartesian origin) is specified with this CGLOC command.
For example, if Y is vertical and the global system origin is at the surface of the earth while the acceleration
system origin is at the center of the earth, YLOC should be -4000 miles (or equivalent) if the rotational effects
of the earth are to be included. The rotational velocity of the global Cartesian system about this point is
specified with the CGOMGA command, and the rotational acceleration is specified with the DCGOMG
command.
The rotational velocities and accelerations are mainly intended to include mass effects in a static (ANTYPE,STAT-
IC) analysis. If used in dynamic analyses, no coupling exists between the user input terms and the time history
response of the structure. See Acceleration Effect in the Theory Reference for the Mechanical APDL and
Mechanical Applications for details. Related commands are ACEL, CGOMGA, DCGOMG, DOMEGA, and
OMEGA.
This command is also valid in PREP7.
257
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CGLOC
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Flow Environment>Rotating Coords
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Coriolis Effects
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Coriolis Effects
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Acceleration CS>Delete Accel CS
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Acceleration CS>Set Accel CS
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Coriolis Effects
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Coriolis Effects
Main Menu>Solution>FLOTRAN Set Up>Flow Environment>Rotating Coords
Main Menu>Solution>Loading Options>Acceleration CS>Delete Accel CS
Main Menu>Solution>Loading Options>Acceleration CS>Set Accel CS
CGOMGA, CGOMX, CGOMY, CGOMZ
Specifies the rotational velocity of the global origin.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS FL <> <> DY PP <> EME MFS
CGOMX, CGOMY, CGOMZ
Rotational velocity of the global origin about the acceleration system X, Y, and Z axes.
Notes
Specifies the rotational velocity of the global origin about each of the acceleration coordinate system axes.
The location of the acceleration coordinate system is defined with the CGLOC command. Rotational velocities
may be defined in analysis types ANTYPE,STATIC, HARMIC (full or mode superposition), TRANS (full or mode
superposition), and SUBSTR. See Acceleration Effect in the Theory Reference for the Mechanical APDL and
Mechanical Applications for details. Units are radians/time. Related commands are ACEL, CGLOC, DCGOMG,
DOMEGA, and OMEGA.
The CGOMGA command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for CGOMGA_X, CGOMGA_Y, and CGOMGA_Z input values (*DIM) for full transient and
harmonic analyses.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Flow Environment>Rotating Coords
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Coriolis Effects
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Coriolis Effects
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Coriolis Effects
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Coriolis Effects
Main Menu>Solution>FLOTRAN Set Up>Flow Environment>Rotating Coords
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
258
CGOMGA
CHECK, Sele, Levl
Checks current database items for completeness.
PREP7: Database
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Sele
Specifies which elements are to be checked:
(blank)
Check all data.
ESEL
Check only elements in the selected set and unselect any elements not producing geometry check
messages. The remaining elements (those producing check messages) can then be displayed and
corrected. A null set results if no elements produce a message. Issue ESEL,ALL to select all elements
before proceeding.
Levl
Used only with Sele = ESEL:
WARN
Select elements producing warning and error messages.
ERR
Select only elements producing error messages (default).
Notes
This command will not work if SHPP,OFF has been set. A similar, automatic check of all data is done before
the solution begins.
If the "Check Elements" option is invoked through the GUI (menu path Main Menu> Preprocessor> Meshing>
Check Elems), the CHECK,ESEL logic is used to highlight elements in the following way: good elements are
blue, elements having warnings are yellow, and bad (error) elements are red.
Note
The currently selected set of elements is not changed by this GUI function.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Meshing>Check Mesh>Individual Elm>Select Warning/Error Elements
259
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CHECK
CHKMSH, Comp
Checks area and volume entities for previous meshes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Comp
Name of component containing areas or volumes.
Notes
CHKMSH invokes a predefined ANSYS macro that checks areas and volumes to find out if they were previously
meshed. This macro name will appear in the log file (Jobname.LOG) prior to area and volume meshing
operations initiated through the GUI. This command is not intended to be typed in directly in an ANSYS
session (although it can be included in an input file for use with the /INPUT command).
Menu Paths
This command cannot be accessed from a menu.
CINT, Action, Par1, Par2, Par3, Par4, Par5, Par6, Par7
Defines parameters associated with contour integral calculations
PREP7: Data Tables
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Action
Specifies action for defining or manipulating initial crack data:
NEW
Initiate a new contour integral calculation and assign an ID.
CTNC
Define the crack tip node component.
CENC
Define the crack extension node component, the crack tip node, and the crack extension direction.
TYPE
Define the type of calculation to perform.
DELE
Delete the CINT object associated with the specified ID.
NCON
Define the number of contours to be calculated in the contour integral calculation.
SYMM
Indicate whether the crack is on a symmetrical line or plane.
NORM
Define the crack plane normal.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
260
CHKMSH
Command Specification for Action = NEW
CINT, NEW, Par1
Par1 --
CINT ID number.
Command Specifications for Action = CTNC
CINT, CTNC, Par1, Par2, Par3
Par1 --
Crack tip node component name.
Par2 --
Crack-extension direction calculation-assist node. Any node on the open side of the crack.
Par3 --
Crack fronts end-node crack-extension direction override flag:
0 --
Align the extension direction with the edges attached at the two end nodes of the crack front (default).
1 --
Align the extension direction to be perpendicular to the crack front.
Command Specifications for Action = CENC
CINT, CENC, Par1, Par2, Par3, Par4, Par5, Par6, Par7
Par1 --
Crack extension node component name (CM).
Par2 --
Crack tip node. The crack tip node defaults to the first node of the crack extension node component.
Par3, Par4 --
Coordinate system number (Par3) and the number of the axis that is coincident with the crack direction
(Par4). When these parameters are defined, Par5, Par6 and Par7 are ignored.
Par5, Par6, Par7 --
Global x, y, and z components of the crack extension direction vector. (Par3 and Par4 must be blank.)
Command Specifications for Action = TYPE
CINT, TYPE, Par1
Par1 --
Type of calculation to perform:
JINT --
Calculate J-integral (default).
SIFS --
Calculate stress-intensity factors.
261
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CINT
Command Specifications for Action = DELE
CINT, DELE, Par1
Par1 --
CINT ID (default = ALL).
Command Specifications for Action = NCON
CINT, NCON, Par1
Par1 --
Number of contours to be calculated.
Command Specifications for Action = SYMM
CINT, SYMM, Par1
Par1 --
OFF, 0, or NO --
No symmetry (default).
ON, 1, or YES --
Symmetric about the crack line/plane.
Command Specifications for Action = NORM
CINT, NORM, Par1, Par2
Par1 --
Coordinate system number (default = 0, global Cartesian).
Par2 --
Axis of coordinate system (default = 2, global Cartesian Y-axis).
Notes
Initiate a new J-integral calculation using the Action = NEW parameter. Subsequent CINT commands (with
parameters other than NEW) will define the input required for the contour integral calculations.
The simplest method for contour integral calculation is to define crack information using Action = CTNC;
however, this method limits you to only one node for a given location along the crack front.
To define crack information at multiple locations along the crack front, use Action = CENC. You can issue
CINT,CENC, Par1, etc. multiple times to define the crack extension node component, the crack tip, and the
crack extension directions at multiple locations along the crack front.
Although you can vary the sequence of your definitions, all of the crack tip nodes specified must be at the
crack front, and no crack tip node can be omitted
You can define your crack extension direction directly using either Action = CENC or Action = NORM.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
262
CINT
For more information about using the CINT command, including supported element types and material
behavior, see Numerical Evaluation of Fracture Mechanics Parameters in the Structural Analysis Guide.
Menu Paths
This command cannot be accessed from a menu.
CIRCLE, PCENT, RAD, PAXIS, PZERO, ARC, NSEG
Generates circular arc lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
PCENT
Keypoint defining the center of the circle (in the plane of the circle). If PCENT = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI).
RAD
Radius of the circle. If RAD is blank and PCENT = P, the radius is the distance from PCENT to PZERO.
PAXIS
Keypoint defining axis of circle (along with PCENT). If PCENT = P and PAXIS is omitted, the axis is
normal to the working plane.
PZERO
Keypoint defining the plane normal to circle (along with PCENT and PAXIS) and the zero degree location.
Need not be in the plane of the circle. This value is not required if PAXIS is defined along the Y axis
(that is, a circle in the XZ plane).
ARC
Arc length (in degrees). Positive follows right-hand rule about PCENT-PAXIS vector. Defaults to 360.
NSEG
Number of lines around circumference (defaults to minimum required for 90-maximum arcs, i.e., 4 for
360). Number of keypoints generated is NSEG for 360 or NSEG + 1 for less than 360.
Notes
Generates circular arc lines (and their corresponding keypoints). Keypoints are generated at regular angular
locations (based on a maximum spacing of 90). Arc lines are generated connecting the keypoints. Keypoint
and line numbers are automatically assigned, beginning with the lowest available values [NUMSTR]. Adjacent
lines use a common keypoint. Line shapes are generated as arcs, regardless of the active coordinate system.
Line shapes are invariant with coordinate system after they are generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Arcs>By Cent & Radius
Main Menu>Preprocessor>Modeling>Create>Lines>Arcs>Full Circle
263
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CIRCLE
CISOL, n, ID, node, Cont
Stores J-integral information in a variable.
POST26: Set Up
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
n
Arbitrary reference number or name assigned to this variable. Number must be >1 but </= NUMVAR.
ID
Crack ID number.
node
Crack tip node number.
Cont
Contour number.
Menu Paths
This command cannot be accessed from a menu.
/CLABEL, WN, KEY
Specifies contour labeling.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
KEY
Labeling key:
0 or 1
Label contours with legend or color (default).
-1
No contour labeling.
N
Same as 1 except show alphabetic legend only on every Nth element.
Command Default
Show contour line labels.
Notes
Labels contours for identification with alphabetic legend for vector displays and color for raster displays.
Number of contours is automatically reduced to 9 (or fewer) for clarity. Use /CONTOUR command to increase
(24 maximum for alphabetic labeling; no limit for color labeling).
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
264
CISOL
Menu Paths
Utility Menu>PlotCtrls>Style>Contours>Contour Labeling
/CLEAR, Read
Clears the database.
DATABASE: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Read
File read option:
START
Reread start120.ans file (default).
NOSTART
Do not reread start120.ans file.
Notes
Resets the ANSYS database to the conditions at the beginning of the problem. Sets the import and Boolean
options back to the ANSYS default. All items are deleted from the database and memory values are set to
zero for items derived from database information. All files are left intact. This command is useful between
multiple analyses in the same run, or between passes of a multipass analysis (such as between the substructure
generation, use, and expansion passes). Should not be used in a do-loop since loop counters will be reset.
The start120.ans file will be reread (by default) after the database is cleared, unless Read is set to
NOSTART. Additional commands cannot be stacked (using the $ separator) on the same line as the /CLEAR
command.
Use care when placing the /CLEAR command within branching constructs (for example, those employing
*DO or *IF commands). The command deletes all parameters including the looping parameter for do-loops.
(You can preserve your iteration parameter by issuing a PARSAV command prior to the /CLEAR command,
then following the /CLEAR command with a PARRES command.)
/CLEAR resets the jobname to match the currently open session .LOG and .ERR files. This will return the
jobname to its original value, or to the most recent value specified on /FILNAME with KEY = 1.
This command is valid only at the Begin level.
Menu Paths
Utility Menu>File>Clear & Start New
265
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/CLEAR
CLOCAL, KCN, KCS, XL, YL, ZL, THXY, THYZ, THZX, PAR1, PAR2
Defines a local coordinate system relative to the active coordinate system.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
KCN
Arbitrary reference number assigned to this coordinate system. Must be greater than 10. A coordinate
system previously defined with this number will be redefined.
KCS
Coordinate system type:
0 or CART
Cartesian
1 or CYLIN
Cylindrical (circular or elliptical)
2 or SPHE
Spherical (or spheroidal)
3 or TORO
Toroidal
XL, YL, ZL
Location (in the active coordinate system) of the origin of the new coordinate system (R, , Z for cylindrical,
R, , for spherical or toroidal).
THXY
First rotation about local Z (positive X toward Y).
THYZ
Second rotation about local X (positive Y toward Z).
THZX
Third rotation about local Y (positive Z toward X).
PAR1
Used for elliptical, spheroidal, or toroidal systems. If KCS = 1 or 2, PAR1 is the ratio of the ellipse Y-axis
radius to X-axis radius (defaults to 1.0 (circle)). If KCS = 3, PAR1 is the major radius of the torus.
PAR2
Used for spheroidal systems. If KCS = 2, PAR2 = ratio of ellipse Z-axis radius to X-axis radius (defaults
to 1.0 (circle)).
Notes
Defines and activates a local coordinate system by origin location and orientation angles relative to the
active coordinate system. This local system becomes the active coordinate system, and is automatically
aligned with the active system (i.e., x is radial if a cylindrical system is active, etc.). Nonzero rotation angles
(degrees) are relative to this automatic rotation. See the CS, CSKP, CSWPLA, and LOCAL commands for al-
ternate definitions. Local coordinate systems may be displayed with the /PSYMB command.
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
266
CLOCAL
Menu Paths
This command cannot be accessed from a menu.
CLOG, IR, IA, --, --, Name, --, --, FACTA, FACTB
Forms the common log of a variable
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA
Reference number of the variable to be operated on.
--, --
Unused fields.
Name
Thirty-two character name for identifying the variable on printouts and displays. Embedded blanks are
compressed for output.
--, --
Unused fields.
FACTA
Scaling factor applied to variable IA (defaults to 1.0).
FACTB
Scaling factor (positive or negative) applied to the operation (defaults to 1.0).
Notes
Forms the common log of a variable according to the operation:
IR = FACTB*LOG(FACTA x IA)
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Common Log
267
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CLOG
/CLOG, Fname, Ext, --
Copies the session log file to a named file.
SESSION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path to which the log file is to be copied (248 characters maximum, including
directory). If you do not specify a directory path, it will default to your working directory and you can
use all 248 characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
Notes
This command is valid in any processor, but only during an interactive run.
Menu Paths
This command cannot be accessed from a menu.
CLRMSHLN
Clears meshed entities.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
When you use the GUI method to set the number of elements on specified lines, and any of those lines is
connected to one or more meshed lines, areas, or volumes, ANSYS gives you the option to clear the meshed
entities. (This occurs only when you perform this operation via the GUI; ANSYS does not provide such an
option when you use the command method [LESIZE].)
If you activate the mesh clearing option, the program invokes an ANSYS macro, CLRMSHLN, that clears the
meshed entities. This macro name will appear in the log file (Jobname.LOG). This macro is for the ANSYS
program's internal use only. This command is not intended to be typed in directly in an ANSYS session, al-
though it can be included in an input file for batch input or for use with the /INPUT command.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
268
/CLOG
CM, Cname, Entity
Groups geometry items into a component.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Cname
An alphanumeric name used to identify this component. Cname may be up to 32 characters, beginning
with a letter and containing only letters, numbers, and underscores. Component names beginning with
an underscore (e.g., _LOOP) are reserved for use by ANSYS and should be avoided. Components named
ALL,STAT, and DEFA are not permitted. Overwrites a previously defined name.
Entity
Label identifying the type of geometry items to be grouped:
VOLU
Volumes.
AREA
Areas.
LINE
Lines.
KP
Keypoints.
ELEM
Elements.
NODE
Nodes.
Notes
Components may be further grouped into assemblies [CMGRP]. The selected items of the specified entity
type will be stored as the component. Use of this component in the select command [CMSEL] causes all
these items to be selected at once, for convenience.
A component is a grouping of some geometric entity that can then be conveniently selected or unselected.
A component may be redefined by reusing a previous component name. The following entity types may
belong to a component: nodes, elements, keypoints, lines, areas, and volumes. A component may contain
only 1 entity type, but an individual item of any entity may belong to any number of components. Once
defined, the items contained in a component may then be easily selected or unselected [CMSEL]. Components
may be listed [CMLIST], modified [CMMOD] and deleted [CMDELE]. Components may also be further grouped
into assemblies [CMGRP]. Other entities associated with the entities in a component (e.g., the lines and
keypoints associated with areas) may be selected by the ALLSEL command.
An item will be deleted from a component if it has been deleted by another operation (see the KMODIF
command for an example). Components are automatically updated to reflect deletions of one or more of
their items. Components are automatically deleted and a warning message is issued if all their items are
deleted. Assemblies are also automatically updated to reflect deletions of one or more of their components
or subassemblies, but are not deleted if all their components and subassemblies are deleted.
This command is valid in any processor.
269
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CM
Menu Paths
Utility Menu>Select>Comp/Assembly>Create Component
CMACEL, CM_NAME, CMACEL_X, CMACEL_Y, CMACEL_Z
Specifies the translational acceleration of an element component
SOLUTION: Inertia
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
CM_NAME
The name of the element component (8-character maximum).
CMACEL_X, CMACEL_Y, CMACEL_Z
Acceleration of the element component CM_NAME in the global Cartesian X, Y, and Z axis directions,
respectively.
Notes
The CMACEL command specifies the translational acceleration of the element component in each of the
global Cartesian (X, Y, and Z) axis directions.
You can use the CMACEL command to specify translational, acceleration-based loading on up to 100 element
components.
Components for which you want to specify acceleration loading must consist of elements only. The elements
you use cannot be part of more than one component, and elements that share nodes cannot exist in different
element components. You cannot apply the loading to an assembly of element components.
To simulate gravity (by using inertial effects), accelerate the structure in the direction opposite to gravity.
For example, apply a positive CMACELY to simulate gravity acting in the negative Y direction. Units are
length/time
2
.
You can define the acceleration for the following analyses types:
Static (ANTYPE,STATIC)
Harmonic (ANTYPE,HARMIC), full or mode superposition method
Transient (ANTYPE,TRANS), full or mode superposition method
Substructure (ANTYPE,SUBSTR)
Accelerations are combined with the element mass matrices to form a body force load vector term. Units
of acceleration and mass must be consistent to give a product of force units.
In a modal harmonic or transient analysis, you must apply the load in the modal portion of the analysis.
ANSYS calculates a load vector and writes it to the mode shape file, which you can apply via the LVSCALE
command.
The CMACEL command does not support these elements: FLUID79, FLUID80, FLUID81, and PIPE59.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
270
CMACEL
The CMACEL command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for CMACEL_X, CMACEL_Y, and CMACEL_Z input values (*DIM) as a function of both time
and frequency for full transient and harmonic analyses.
Related commands for inertia loads are ACEL, CGLOC, CGOMGA, DCGOMG, DOMEGA, OMEGA, CMOMEGA,
and CMDOMEGA.
This command is also valid in /PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Gravity>On Components
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Gravity>On Components
/CMAP, Fname, Ext, --, Kywrd, NCNTR
Changes an existing or creates a new color mapping table.
GRAPHICS: Set Up
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
If blank, restore color map.
Ext
Filename extension (8 character maximum).
--
Unused field.
Kywrd
Keyword indicating the disposition of the color map file.
(blank)
Loads existing color map file.
CREATE
Starts CMAP utility and modifies or creates the specified file.
SAVE
Writes the active color map to the specified file, which can be imported into future ANSYS sessions.
NCNTR
Number of contours to be defined by the /CMAP command (max = 128). If no value is specified, CMAP
defaults to 9, even if an existing file is being modified.
Command Default
Use predefined ANSYS color map table.
271
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/CMAP
Notes
Reads the color map file (RGB index specifications) to change from current specifications. Only one color
map may be active at a time.
For 2-D drivers (especially Win32c), modifying the color map can produce anomalies, including legend/contour
disagreement.
Changing the color map in ANSYS with the /CMAP command will change the meaning of the color labels
on the /COLOR command. See /COLOR for other color controls.
This command is valid anywhere.
Menu Paths
Utility Menu>PlotCtrls>Redirect Plots>To GRPH File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL2 File
Utility Menu>PlotCtrls>Redirect Plots>To PSCR File
Utility Menu>PlotCtrls>Style>Colors>Default Color Map
CMATRIX, SYMFAC, Condname, NUMCOND, GRNDKEY, Capname
Performs electrostatic field solutions and calculates the self and mutual capacitances between multiple
conductors.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
SYMFAC
Geometric symmetry factor. Capacitance values are scaled by this factor which represents the fraction
of the total device modeled. Defaults to 1.
Condname
Alpha-numeric prefix identifier used in defining named conductor components.
NUMCOND
Total Number of Components. If a ground is modeled, it is to be included as a component. If a ground
is not modeled, but infinite elements are used to model the far-field ground, a named component for
the far-field ground is not required.
GRNDKEY
Ground key:
0
Ground is one of the components, which is not at infinity.
1
Ground is at infinity (modeled by infinite elements or a Trefftz domain).
Capname
Array name for computed capacitance matrix. Defaults to CMATRIX.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
272
CMATRIX
Notes
To invoke the CMATRIX macro, the exterior nodes of each conductor must be grouped into individual
components using the CM command. Each set of independent components is assigned a component name
with a common prefix followed by the conductor number. A conductor system with a ground must also include
the ground nodes as a component. The ground component is numbered last in the component name se-
quence.
A Ground Capacitance matrix is a matrix relating charge to a voltage vector. A ground matrix can not be
applied to a circuit modeler such as SPICE. The Lumped Capacitance matrix is a matrix formed by a combin-
ation of lumped "arrangements" of voltage differences between conductors. You can use the lumped capa-
citance terms in a circuit modeler to represent capacitances between conductors.
You must enclose all name-strings in single quotes in the CMATRIX command line.
See the Theory Reference for the Mechanical APDL and Mechanical Applications and HMAGSOLV in the Low-
Frequency Electromagnetic Analysis Guide for details.
Menu Paths
Main Menu>Solution>Solve>Electromagnet>Static Analysis>Capac Matrix
CMDELE, Name
Deletes a component or assembly definition.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Name of the component or assembly whose definition is to be removed.
Notes
Entities contained in the component, or the components within the assembly, are unaffected. Only the
grouping relationships are deleted. Assemblies are automatically updated to reflect deletion of their com-
ponents or subassemblies, but they are not automatically deleted when all their components or subassemblies
are deleted.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Layers>Picked Lines
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Utility Menu>Select>Comp/Assembly>Delete Comp/Assembly
273
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMDELE
CMDOMEGA, CM_NAME, DOMEGAX, DOMEGAY, DOMEGAZ, X1, Y1, Z1, X2, Y2, Z2
Specifies the rotational acceleration of an element component about a user-defined rotational axis.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
CM_NAME,
The name of the element component (8 character maximum).
DOMEGAX, DOMEGAY, DOMEGAZ
If the X2, Y2, Z2 fields are not defined, DOMEGAX, DOMEGAY, and DOMEGAZ specify the components of
the rotational acceleration vector in the global Cartesian X, Y, Z directions.
If the X2, Y2, Z2 fields are defined, only DOMEGAX is required. DOMEGAX specifies the scalar rotational
acceleration about the rotational axis. The rotational direction of DOMEGAXis designated either positive
or negative, and is determined by the right hand rule.
X1, Y1, Z1
If the X2, Y2, Z2 fields are defined, X1, Y1, and Z1 define the coordinates of the beginning point of the
rotational axis vector. Otherwise, X1, Y1, and Z1 are the coordinates of a point through which the rota-
tional axis passes.
X2, Y2, Z2
The coordinates of the end point of the rotational axis vector.
Notes
Specifies the rotational acceleration components DOMEGAX, DOMEGAY, and DOMEGAZ of an element com-
ponent CM_NAME about a user-defined rotational axis. The rotational axis can be defined either as a vector
passing through a single point, or a vector connecting two points.
You can use the CMDOMEGA command to specify acceleration based loading on up to 100 rotational element
components.
You can define rotational acceleration and rotational axis with the CMDOMEGA command for STATIC,
HARMIC (full), TRANS (full), and SUBSTR analyses. Rotational velocities are combined with the element mass
matrices to form a body force load vector term. Units are radians/time
2
.
The CMDOMEGA command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for CMDOMEGA_X, CMDOMEGA_Y, and CMDOMEGA_Z input values (*DIM) for full transient
and harmonic analyses.
Related commands are ACEL, CGLOC, CGLOC, OMEGA, CMOMEGA, DCGOMG, DOMEGA.
You can use the CMDOMEGA command in conjunction with any one of the following two groups of com-
mands, but not with both groups simultaneously:
GROUP ONE: OMEGA, DOMEGA.
GROUP TWO: CGOMGA, DCGOMG, CGLOC.
Components for which you want to specify rotational loading must consist of elements only. The elements
you use cannot be part of more than one component, and elements that share nodes cannot exist in different
element components. You cannot apply the loading to an assembly of element components.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
274
CMDOMEGA
In a modal harmonic or transient analysis, you must apply the load in the modal portion of the analysis.
ANSYS calculates a load vector and writes it to the mode shape file, which you can apply via the LVSCALE
command.
See Acceleration Effect in the Theory Reference for the Mechanical APDL and Mechanical Applications for more
information.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>On
Components>By Axis
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>On
Components>By origin
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>On
Components>Pick Kpt
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>On
Components>Pick Kpts
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>On
Components>Pick Node
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>On
Components>Pick Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Angular Accel>On
Component
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>On Components>By
Axis
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>On Components>By
origin
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>On Components>Pick
Kpt
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>On Components>Pick
Kpts
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>On Components>Pick
Node
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>On Components>Pick
Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Angular Accel>On Component
CMEDIT, Aname, Oper, Cnam1, Cnam2, Cnam3, Cnam4, Cnam5, Cnam6, Cnam7
Edits an existing assembly.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Aname
Name of the assembly to be edited.
Oper
Operation label:
275
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMEDIT
ADD
To add more components. The level of any assembly to be added must be lower than that of the
assembly Aname (see CMGRP command).
DELE
To remove components.
Cnam1, Cnam2, Cnam3, . . . , Cnam7
Names of components and assemblies to be added to or deleted from the assembly.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Comp/Assembly>Edit Assembly
CMGRP, Aname, Cnam1, Cnam2, Cnam3, Cnam4, Cnam5, Cnam6, Cnam7, Cnam8
Groups components and assemblies into an assembly.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Aname
An alphanumeric name used to identify this assembly. Aname may be up to 32 characters, beginning
with a letter and containing only letters, numbers, and underscores. Overwrites a previously defined
Aname (and removes it from higher level assemblies, if any).
Cnam1, Cnam2, Cnam3, . . . , Cnam8
Names of existing components or other assemblies to be included in this assembly.
Notes
Groups components and other assemblies into an assembly identified by a name. CMGRP is used for the
initial definition of an assembly. An assembly is used in the same manner as a component. Up to 5 levels
of assemblies within assemblies may be used.
An assembly is a convenient grouping of previously defined components and other assemblies. Assemblies
may contain components only, other assemblies, or any combination. A component may belong to any
number of assemblies. Up to 5 levels of nested assemblies may be defined. Components and assemblies
may be added to or deleted from an existing assembly by the CMEDIT command. Once defined, an assembly
may be listed, deleted, selected, or unselected using the same commands as for a component. Assemblies
are automatically updated to reflect deletions of one or more of their components or lower-level assemblies.
Assemblies are not automatically deleted when all their components or subassemblies are deleted.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Comp/Assembly>Create Assembly
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
276
CMGRP
CMLIST, Name, Key, Entity
Lists the contents of a component or assembly.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Name of the component or assembly to be listed (if blank, list all selected components and assemblies).
If Name is specified, then Entity is ignored.
Key
Expansion key:
0
Do not list individual entities in the component.
1 or EXPA
List individual entities in the component.
Entity
If Name is blank, then the following entity types can be specified:
VOLU
List the volume components only.
AREA
List the area components only.
LINE
List the line components only.
KP
List the keypoint components only
ELEM
List the element components only.
NODE
List the node components only.
Notes
This command is valid in any processor. For components, it lists the type of geometric entity. For assemblies,
it lists the components and/or assemblies that make up the assembly.
Examples of possible usage:
CMLIST - List all selected components.
CMLIST, , EXPA - List all selected components and for each component list the underlying entity ID's.
CMLIST,Name - List the specified component.
CMLIST,Name,EXPA - List specified component along with all underlying entity ID's.
CMLIST, , EXPA,Entity - List all selected components of specified entity type. For each component
also list the underlying entity ID's.
Menu Paths
Utility Menu>List>Components
277
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMLIST
Utility Menu>List>Other>Components
Utility Menu>Select>Comp/Assembly>List Comp/Assembly
CMMOD, Cname, Keyword, Value
Modifies the specification of a component.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Cname
Name of the existing component or assembly to be modified.
Keyword
The label identifying the type of value to be modified.
NAME - Modify the NAME of the component
Value
If Keyword is NAME, then the value is the alphanumeric label to be applied. See the CM command for
naming convention details. If a component named Value already exists, the command will be ignored
and an error message will be generated.
Notes
The naming conventions for components, as specified in the CM command, apply for CMMOD (32 characters,
ALL, STAT and DEFA are not allowed, etc.). However, if you choose a component name that is already
designated for another component, an error message will be issued and the command will be ignored.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
CMOMEGA, CM_NAME, OMEGAX, OMEGAY, OMEGAZ, X1, Y1, Z1, X2, Y2, Z2, KSPIN
Specifies the rotational velocity of an element component about a user-defined rotational axis.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
CM_NAME
The name of the element component (eight-character maximum).
OMEGAX, OMEGAY, OMEGAZ
If the X2, Y2, Z2 fields are not defined, OMEGAX, OMEGAY, and OMEGAZ specify the components of the
rotational velocity vector in the global Cartesian X, Y, Z directions.
If the X2, Y2, Z2 fields are defined, only OMEGAX is required. OMEGAX specifies the scalar rotational ve-
locity about the rotational axis. The rotational direction of OMEGAX is designated either positive or
negative, and is determined by the right hand rule.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
278
CMMOD
X1, Y1, Z1
If the X2, Y2, Z2 fields are defined,X1, Y1, and Z1 define the coordinates of the beginning point of the
rotational axis vector. Otherwise, X1, Y1, and Z1 are the coordinates of a point through which the rota-
tional axis passes.
X2, Y2, Z2
The coordinates of the end point of the rotational axis vector.
KSPIN
The spin-softening key:
0 or OFF
No modification of the stiffness matrix of the element component due to rotation. This value is the
default except in a dynamic analysis when the Coriolis effect is taken into account in the rotating
reference frame (CORIOLIS,ON); in this case, ANSYS applies spin-softening by default.
1 or ON
Decrease radial stiffness of the element component due to rotation (that is, include spin-softening
effects).
Notes
Specifies the rotational velocity components OMEGAX, OMEGAY, and OMEGAZ of an element component
CM_NAME about a user-defined rotational axis. The rotational axis can be defined either as a vector passing
through a single point or a vector connecting two points.
You can use the CMOMEGA command to specify velocity-based loading on up to 100 rotational element
components.
You can define rotational velocity and rotational axis for these analysis types:
Static (ANTYPE,STATIC)
Harmonic (ANTYPE,HARMIC) -- Full
Transient (ANTYPE,TRANS) -- Full
Harmonic Modal Superposition HROPT,MSUP
Transient Modal Superposition TRNOPT,MSUP
Substructuring (ANTYPE,SUBSTR)
Modal (ANTYPE,MODAL)
Rotational velocities are combined with the element mass matrices to form a body force load vector term.
Units are radians/time. Related commands are ACEL, CGLOC, CGLOC, CGOMGA, CMDOMEGA, DCGOMG,
DOMEGA.
You can use the CMOMEGA command in conjunction with either one of the following two groups of com-
mands, but not with both groups simultaneously:
GROUP ONE: OMEGA, DOMEGA.
GROUP TWO: CGOMGA, DCGOMG, CGLOC.
Components for which you want to specify rotational loading must consist of elements only. The elements
you use cannot be part of more than one component, and elements that share nodes cannot exist in different
element components. You cannot apply the loading to an assembly of element components.
279
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMOMEGA
The KSPIN option allows you to adjust the stiffness of a rotating body to account for dynamic mass effects.
The adjustment is called spin-softening and applies to a modal (ANTYPE,MODAL), harmonic (ANTYPE,HARM),
or transient (ANTYPE,TRANS) analysis only. By default, ANSYS does not include the spin-softening adjustment
in the analysis except for dynamic analyses where you have applied the Coriolis effect (CORIOLIS,ON) in the
rotating reference frame. To account for this exception, if the default is requested, the KSPIN value appears
as -1 in all saved files containing loads (such as .CDB, .Snn, and .LDHI). The adjustment approximates the
effects of geometry changes caused by large-deflection circumferential motion in a small-deflection analysis.
The KSPIN option is not intended for a large-deflection static analysis; in such a case, use the NLGEOM,ON
command to account for this effect. The command is usually used in conjunction with prestressing (PSTRES).
If you have applied the Coriolis effect (CORIOLIS) using a stationary reference frame, the CMOMEGA command
takes the gyroscopic damping matrix into account for the elements listed under Stationary Reference Frame
in the notes section of the CORIOLIS command. ANSYS verifies that the rotation vector axis is parallel to
the axis of the element; if not, the gyroscopic effect is not applied. If you issue a CMOMEGA command when
the Coriolis or gyroscopic effect is present, a subsequently issued OMEGA command has no effect.
The overall stiffness matrix should not be reused (KUSE) if CMOMEGA command parameters change between
substeps and the spin-softening key (KSPIN) is active. See Reuse of Matrices or Structure Matrices in the
Theory Reference for the Mechanical APDL and Mechanical Applications for more information.
The CMOMEGA command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for CMOMEGA_X, CMOMEGA_Y, and CMOMEGA_Z input values (*DIM) for full transient and
harmonic analyses.
In a modal harmonic or transient analysis, you must apply the load in the modal portion of the analysis.
ANSYS calculates a load vector and writes it to the mode shape file, which you can apply via the LVSCALE
command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>On
Components>By Axis
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>On
Components>By origin
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>On
Components>Pick Kpt
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>On
Components>Pick Kpts
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>On
Components>Pick Node
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>On
Components>Pick Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Angular Veloc>On
Component
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>On Components>By
Axis
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>On Components>By
origin
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>On Components>Pick
Kpt
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
280
CMOMEGA
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>On Components>Pick
Kpts
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>On Components>Pick
Node
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>On Components>Pick
Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Angular Veloc>On Component
CMPLOT, Label, Entity, Keyword
Plots the entities contained in a component or assembly.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Name of the component or assembly to be plotted.
(blank)
All selected components and assemblies are plotted (default). If fewer than 11 components are se-
lected, then all are plotted. If more than 11 components are selected, then only the first 11 are
plotted.
ALL
All selected components are plotted. If number of selected components is greater than 11, then the
legend showing component names will not be shown.
N
Next set of defined components and assemblies is plotted.
P
Previous set of defined components and assemblies is plotted.
Cname
The specified component or assembly is plotted.
SetNo.
The specified set number is plotted.
Entity
If Label is BLANK or ALL, then the following entity types can be specified:
VOLU
Plot the volume components only.
AREA
Plot the area components only.
LINE
Plot the line components only.
KP
Plot the keypoint components only.
ELEM
Plot the element components only.
281
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMPLOT
NODE
Plot the node components only.
Keyword
For Keyword = ALL, plot the specified component name in the Label field in the context of all entities
of the same type. Not valid if Label field is BLANK or ALL.
Notes
Components are plotted with their native entities. For assemblies, all native entities for the underlying
component types are plotted simultaneously. Although more components can be plotted, the legend displays
only 11 at a time. When more than eleven are plotted, the legend is not displayed.
Possible usage:
CMPLOT,CNAME - Plots the specified component (if selected).
CMPLOT,CNAME, ALL - Plot component in the context of all other selected entity components of the
same type as the component.
CMPLOT - Plot the first eleven selected components.
CMPLOT,ALL - Plot all selected components.
CMPLOT,N or CMPLOT,P - Plot next or previous set of eleven components.
CMPLOT,ALL,Entity - Plot all selected components of type specified in Entity.
CMPLOT, ,Entity - Plot components of type specified in Entity, from the first eleven components.
CMPLOT,N,Entity - Plot components of type specified in Entity, if any, from the next set of eleven
components (substitute P for N to plot from previous set).
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Components>By Name / Set Number
Utility Menu>Plot>Components>Next Set
Utility Menu>Plot>Components>Previous Set
Utility Menu>Plot>Components>Selected Components
CMROTATE, CM_NAME, ROTATX, ROTATY, ROTATZ, X1, Y1, Z1, X2, Y2, Z2
Specifies the rotational velocity of an element component about a user-defined rotational axis
SOLUTION: Inertia
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
CM_NAME
The name of the element component (eight-character maximum).
ROTATX, ROTATY, ROTATZ
If the X2, Y2, Z2 fields are not defined, ROTATX, ROTATY, and ROTATZ specify the components of the
rotational angle vector in the global Cartesian X, Y, Z directions.
If the X2, Y2, Z2 fields are defined, only ROTATX is required. ROTATX specifies the scalar rotational ve-
locity about the rotational axis. The rotational direction of ROTATX is designated either positive or
negative, and is determined by the right hand rule.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
282
CMROTATE
X1, Y1, Z1
If the X2, Y2, Z2 fields are defined, X1, Y1, and Z1 define the coordinates of the beginning point of the
rotational axis vector. Otherwise, X1, Y1, and Z1 are the coordinates of a point through which the rota-
tional axis passes.
X2, Y2, Z2
The coordinates of the end point of the rotational axis vector.
Notes
The CMROTATE command specifies the rotational motion velocity components ROTATX, ROTATY, and
ROTATZ of an element component CM_NAME about a user-defined rotational axis. The rotational axis can
be defined either as a vector passing through a single point or a vector connecting two points.
You can use the CMROTATE command to specify angular rotation-based displacement or angular velocity
on up to 100 rotational element components.
You can define rotational velocity and rotational axis for static analyses (ANTYPE,STATIC) only.
The command sets the constant rotational velocity on the nodes of the element component despite any
deformation at the nodes. This feature is primarily useful for generating sliding contact at frictional contact
interfaces in a brake squeal analysis.
A brake squeal analysis generally involves a partial solution step (PSOLVE) subsequent to a large-deformation
static analysis with the Newton Raphson option set as NROPT = UNSYM. Therefore, CMROTATE is not ap-
plicable for multiple load step solves using the LSSOLVE command.
The CMROTATE command typically applies to brake squeal modeling where surface-to-surface contact occurs
between the brake pad and the rotating disk. The applicable contact elements, therefore, are CONTA173,
CONTA174, and CONTA175.
This command is also valid in PREP7.
Menu Paths
This command cannot be accessed from a menu.
CMSEL, Type, Name, Entity
Selects a subset of components and assemblies.
DATABASE: Components
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Select a new set (default).
R
Reselect a set from the current set.
283
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMSEL
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Also select all components.
NONE
Unselect all components.
Name
Name of component or assembly whose items are to be selected (valid only if Type = S, R, A, or U).
Graphical picking is enabled if Type is blank and Name = PICK (or simply P).
Entity
If Name is blank, then the following entity types can be specified:
VOLU
Select the volume components only.
AREA
Select the area components only.
LINE
Select the line components only.
KP
Select the keypoint components only.
ELEM
Select the element components only.
NODE
Select the node components only.
Notes
Selecting by component is a convenient adjunct to individual item selection (e.g., VSEL, ESEL, etc.). CMSEL,
ALL allows you to select components in addition to other items you have already selected.
If Type = R for an assembly selection [CMSEL,R,<assembly-name>], the reselect operation is performed on
each component in the assembly in the order in which the components make up the assembly. Thus, if one
reselect operation results in an empty set, subsequent operations will also result in empty sets. For example,
if the first reselect operation tries to reselect node 1 from the selected set of nodes 3, 4, and 5, the operation
results in an empty set (that is, no nodes are selected). Since the current set is now an empty set, if the
second reselect operation tries to reselect any nodes, the second operation also results in an empty set, and
so on. This is equivalent to repeating the command CMSEL,R,<component-name> once for each component
making up the assembly.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Joint Elems
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>On Joint Elems
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
284
CMSEL
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Joint Elems
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Force/Moment>On Joint Elems
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Joint Elems
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>On Joint Elems
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Joint Elems
Main Menu>Solution>Define Loads>Delete>Structural>Force/Moment>On Joint Elems
Utility Menu>Select>Comp/Assembly>Select Comp/Assembly
CMSFILE, Option, Fname, Ext, CmsKey
Specifies a list of component mode synthesis (CMS) results files for plotting results on the assembly.
POST1: Special Purpose
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Option
Specifies the command operation:
ADD
Add the specified component results file (Fname) to the list of files to plot. This option is the default.
DELETE
Remove the specified component results file (Fname) from the list of files to plot.
LIST
List all specified component results files.
CLEAR
Clear all previous files added.
ALL
Add all component results (.rst) files from the working directory to the list of files to plot.
Fname
The file name (with full directory path) of the component results file. The default file name is the Jobname
(specified via the /FILNAME command).
Ext
The file name (Fname) extension. The default extension is .rst.
CmsKey
Valid only when adding a results file (Option = ADD or ALL), this key specifies whether or not to check
the specified .rst file to determine if it was created via a CMS expansion pass:
ON
Check (default).
OFF
Do not check.
Command Default
If issued with no arguments, the CMSFILE command uses these defaults:
CMSFILE,ADD,Jobname,rst,ON
285
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMSFILE
The command adds the component results file Jobname.rst.
Notes
The CMSFILE command specifies the list of component mode synthesis (CMS) results files to include when
plotting the mode shape of an assembly.
During postprocessing (/POST1) of a CMS analysis, issue the CMSFILE command to point to component
results files of interest. (You can issue the command as often as needed to include all or some of the com-
ponent results files.) Issue the SET command to acquire the frequencies and mode shapes from substeps
for all specified results files. Execute a plot (PLNSOL) or print (PRNSOL) operation to display the mode shape
of the entire assembly.
When you specify a results file to add to the plot list, the default behavior of the command (CmsKey = ON)
is to first verify that the file is from a CMS analysis and that the frequencies of the result sets on the file
match the frequencies on the first file in the list. If CmsKey = OFF, you can add any .rst file to the list of
files to plot, even if the file was not expanded via a CMS expansion pass.
If CmsKey = ON (default), output from the command appears as: ADD CMS FILE = filename.rst. If
CmsKey = OFF, output from the command appears as: ADD FILE = filename.rst.
If Option = DELETE or CLEAR, you must clear the database (/CLEAR), then re-enter the postprocessor
(/POST1) and issue a SET command for the change to take effect on subsequent plots.
Clearing the database does not clear the list of files specified via the CMSFILE command. Specify Option
= CLEAR to clear the list of files.
Menu Paths
Main Menu>General Postproc>Data & File Opts
CMSOPT, CMSMETH, NMODE, FREQB, FREQE, FBDDEF, FBDVAL
Specifies component mode synthesis (CMS) analysis options.
SOLUTION: Analysis Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
CMSMETH
The component mode synthesis method to use. This value is required.
FIX
Fixed-interface method.
FREE
Free-interface method.
RFFB
Residual-flexible free-interface method.
NMODE
The number of normal modes extracted and used in superelement generation. This value is required
and the minimum is 1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
286
CMSOPT
FREQB
Beginning, or lower end, of frequency range of interest. This value is optional.
ANSYS sets this value to zero if the residual-flexible free-interface method (CMSMETH = RFFB) is specified.
FREQE
Ending, or upper end, of frequency range of interest. This value is optional.
FBDDEF
In a free-interface (CMSMETH = FREE) or residual-flexible free-interface (CMSMETH = RFFB) CMS analysis,
the method to use for defining free body modes:
FNUM
The number (FDBVAL) of rigid body modes in the calculation.
FTOL
Employ a specified tolerance (FDBVAL) to determine rigid body modes in the calculation.
FAUTO
Automatically determine rigid body modes in the calculation. This method is the default.
RIGID
If no rigid body modes exist, define your own via the RIGID command.
FBDVAL
In a free-interface CMS analysis (CMSMETH = FREE), the number of rigid body modes if FBDDEF = FNUM
(where the value is an integer from 0 through 6), or the tolerance to employ if FBDDEF = FTOL (where
the value is a positive real number representing rad/sec). This value is required only when FBDDEF =
FNUM or FBDDEF = FTOL; otherwise, any specified value is ignored.
Command Default
Issuing the CMSOPT command with no arguments is invalid. You must specify at least the CMS method
(CMSMETH) and the number of modes (NMODE). In a free-interface (CMSMETH = FREE) or residual-flexible
free-interface (CMSMETH = RFFB) CMS analysis, the default method for determining rigid body modes is
FAUTO (automatic).
Notes
CMS employs the Block Lanczos eigensolution method in the generation pass.
CMS does not yet support damping matrix reduction. ANSYS sets the matrix generation key to 2 automatically
(SEOPT,SEMATR).
CMS does not support the SEOPT,,,,,RESOLVE command. Instead, ANSYS sets the expansion method for the
expansion pass (EXPMTH) to BACKSUB.
This command is also valid in /PREP7.
For more information about performing a CMS analysis, see "Component Mode Synthesis" in the Advanced
Analysis Techniques Guide.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
287
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CMSOPT
CMWRITE, Fname, Ext, --, --, Fmat
Writes components and assemblies to a file.
DATABASE: Components
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to CM if Fname is blank.
--
Unused field.
Fmat
Format of the output file (defaults to BLOCKED).
BLOCKED
Blocked format. This format allows faster reading of the file.
UNBLOCKED
Unblocked format.
Menu Paths
This command cannot be accessed from a menu.
CNCHECK, Option, RID1, RID2, RINC, PairType, TRlevel
Provides and/or adjusts the initial status of contact pairs.
PREP7: Database
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Option
Option to be performed:
DETAIL
List all contact pair properties (default).
SUMMARY
List only the open/closed status for each contact pair.
POST
Execute a partial solution to write the initial contact configuration to a results file, Jobname.RCN.
ADJUST
Physically move contact nodes to target to close gap or reduce penetration.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
288
CMWRITE
RESET
Reset target element and contact element key options and real constants to their default values.
AUTO
Automatically sets certain real constants and key options to recommended values or settings in order
to achieve better convergence based on overall contact pair behaviors.
TRIM
Trim contact pair (remove certain contact and target elements).
UNSE
Unselect certain contact and target elements.
RID1, RID2, RINC
The range of real constant pair ID's for which Option will be performed. If RID2 is not specified, it
defaults to RID1. If no value is specified, all contact pairs in the selected set of elements are considered.
PairType
Type of contact pairs to be trimmed/unselected/auto-set (used only when Option = TRIM, UNSE, or
AUTO):
ANY
All types (default).
MPC
MPC-based contact pairs (KEYOPT(2) = 2).
BOND
Bonded contact pairs (KEYOPT(12) = 3, 5, 6).
NOSP
No separation contact pairs (KEYOPT(12) = 2, 4).
INAC
Inactive contact pairs (symmetric contact pairs for MPC contact or KEYOPT(8) = 2).
TRlevel
Trimming level (used only when Option = TRIM or UNSE):
(blank)
Normal trimming (default): remove/unselect contact and target elements which are in far-field.
AGGRE
Aggressive trimming: remove/unselect contact and target elements which are in far-field, and certain
elements in near-field.
Notes
CNCHECK provides information for surface-to-surface, node-to-surface, and line-to-line contact pairs (element
types TARGE169, TARGE170, CONTA171, CONTA172, CONTA173, CONTA174, CONTA175, CONTA176, CON-
TA177). All contact and target elements of interest, along with the solid elements and nodes attached to
them, must be selected for the command to function properly.
CNCHECK is available both in PREP7 and in SOLUTION, but only before the first SOLVE, that is, only before
the first load step or the first substep.
The command CNCHECK,POST solves the initial contact configuration in one substep. After issuing this
command, you can postprocess the contact result items as you would for any other converged load step.
289
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CNCHECK
However, only the contact status, contact penetration or gap, and contact pressure will have meaningful
values. Other contact quantities (friction stress, sliding distance, chattering) will be available, but not useful.
If CNCHECK,POST is issued within the SOLUTION processor, the SOLVE command that solves the first load
step of your analysis should appear in a different step, as shown in the following example:
/SOLU
CNCHECK,POST
FINISH
. . .
/SOLU
SOLVE
FINISH
. . .
CNCHECK,POST writes initial contact results to a results file named Jobname.RCN. When postprocessing
the initial contact state, you need to explicitly read results from this file using the FILE and SET,FIRST com-
mands in POST1 to properly read the corresponding contact data. Otherwise, the results file may be read
improperly. The following example shows a valid command sequence for plotting the initial contact gap:
/SOLU
CNCHECK,POST
FINISH
/POST1
FILE,Jobname,RCN
SET,FIRST
PLNSOL,CONT,GAP,0,1
FINISH
. . .
You can issue CNCHECK,ADJUST to physically move contact nodes to the target surface. See Physically
Moving Contact Nodes Towards the Target Surface in the Contact Technology Guide for more information.
Similar to the POST option described above, if CNCHECK,ADJUST is issued within the SOLUTION processor,
the SOLVE command that solves the first load step of your analysis should appear in a different step:
/SOLU
CNCHECK,ADJUST
FINISH
. . .
/SOLU
SOLVE
FINISH
. . .
For performance reasons, ANSYS internally uses a subset of nodes and elements based on the specified
contact regions (RID1, RID2, RINC) when executing CNCHECK,POST or CNCHECK,ADJUST.
You can issue CNCHECK,TRIM or CNCHECK,UNSEL to remove or unselect contact and target elements which
are in far-field (that is, open and not near contact), or even in near field if aggressive trimming logic is used
(TRlevel = AGGRE).
The command CNCHECK,RESET allows you to reset all but a few key options and real constants associated
with the specified contact pairs (RID1, RID2, RINC) to their default values. The following key options and
real constants remain unchanged when this command is issued:
Real constants not af-
fected by RESET
Key options not affected by RE-
SET
Element type
R1, R2 KEYOPT(2), KEYOPT(3) TARGE169, TARGE170
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
290
CNCHECK
R1, R2 KEYOPT(1), KEYOPT(3) CONTA171, CONTA172
R1, R2 KEYOPT(1) CONTA173, CONTA174, CONTA175,
CONTA176, CONTA177
The command CNCHECK,AUTO will automatically change certain default or undefined key options and real
constants to optimized settings or values. The changes are based on overall contact pair behaviors. In gen-
eral, this command improves convergence for nonlinear contact analysis. The tables below list typical KEYOPT
and real constant settings implemented by CNCHECK,AUTO. The actually settings implemented for your
specific model may vary from what is described here. You should always verify the modified settings by is-
suing CNCHECK,DETAIL to list current contact pair properties.
CNCHECK,AUTO Default (0 or
blank)
Description KEY-
OPT
Automatic selection based on DOFs of underlying
elements.
Structural
DOFs
Selects DOF set 1
1- Penalty for rigid-rigid contact [1]. Augmented
Lagrange
Contact algorithm 2
2 - Normal to target surface if KEYOPT(2) > 1. Gauss point Location of contact detec-
tion point
4
1 - Auto CNOF if tiny gap exists. No adjust-
ment
CNOF/ICONT adjustment 5
Use default
range
Contact stiffness vari-
ation
6 1 - Nominal refinement for opening contact or
underlying elements having TB plasticity.
or
2 - Aggressive refinement for opening contact and
underlying elements having TB plasticity.
1 - Automatic bisection for self contact. No control Element level time incre-
ment control
7
2 - Auto selection if KEYOPT(2) > 1. No action Assymetric contact selec-
tion
8
1 - Exclude if KEYOPT(5) = 1, or if ICONT was previ-
ously specified.
Include all Effect of initial penetra-
tion or gap
9
2 - Between iterations except when underlying ele-
ments are superelements. [2]
Between load
steps
Contact stiffness update 10
1. Set to 0 if KEYOPT(2) > 1 for debonding.
2. Set to 0 if underlying elements are superelements, or if KEYOPT(9) = 2 was previously specified.
CNCHECK,AUTO De-
fault
Description Real Con-
stants
Name No.
Set to 5 if KEYOPT(9) = 2 (ramp initial penetration) and KEY-
OPT(10) > 0.
1 Normal penalty stiff-
ness factor
FKN 3
Cut in half if spurious contact is detected or contact searching
is slow.
see
[1]
Pinball region PINB 6
291
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CNCHECK
Calculated as a function of highest conductivity of underlying
element and overall model size.
0 Thermal contact
conductance
TCC 14
Calculated as a function of highest permitivity or lowest res-
istivity of underlying element and overall model size.
0 Electric contact con-
ductance
ECC 19
Calculated as a function of highest emissivity of underlying
element and overall model size.
0 Magnetic contact
permeance
MCC 26
1. PINB default depends on contact behavior (rigid vs. flexible target), NLGEOM,ON or OFF, KEYOPT(9)
setting, KEYOPT(12) setting, and the value of real constant CNOF (see Using PINB).
Menu Paths
Main Menu> Preprocessor> Modeling> Create> Contact Pair
CNVTOL, Lab, VALUE, TOLER, NORM, MINREF
Sets convergence values for nonlinear analyses.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Lab
Valid convergence labels. If STAT, list the status of the currently specified criteria.
Force Convergence Displacement Convergence
Discipline
Label Description Label Label Description Label
Forces F Displacements U Structural
Moments M Rotations ROT
Volume DVOL Hydrostatic Pressure HDSP
Heat Flow HEAT Temperature TEMP Thermal
Fluid Flow FLOW Pressures PRES Fluid
Fluid Force VF Velocity V
Current Flow AMPS Voltage VOLT Electric
Current Flow CURT Electromotive Force EMF
Voltage Drop VLTG Current Flow CURR
Scalar Flux FLUX Scalar Potential MAG Magnetic
Current Segments CSG Vector Potential A
VALUE
Typical reference value for the above label for this analysis. If negative, and if this convergence label
was previously specified explicitly, then convergence based on this label is removed. (A negative VALUE
will not remove a default convergence label.) Defaults to the maximum of a program calculated reference
or MINREF. For degrees of freedom, the reference is based upon the selected NORM and the current
total DOF value. For forcing quantities, the reference is based upon the selected NORM and the applied
loads.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
292
CNVTOL
TOLER
When SOLCONTROL,ON, tolerance about VALUE. Defaults to 0.005 (0.5%) for force and moment, 10E-
4 for DVOL, 0.05 (5%) for displacement when rotational DOFs are not present, and 0.05 (5%) for HDSP.
When SOLCONTROL,OFF, the defaults are 0.001 (0.1%) for force and moment, and 10E-4 for DVOL.
NORM
Specifies norm selection:
2
L2 norm (check SRSS value) (default).
1
L1 norm (check absolute value sum).
0
Infinite norm (check each DOF separately).
MINREF
The minimum value allowed for the program calculated reference value. If negative, no minimum is
enforced. Used only if VALUE is blank. Defaults to 0.01 for force, moment, and volume convergence,
1.0E-6 for heat flow, 1.0E-12 for VOLT and AMPS, 1.0E-6 for HDSP, and 0.0 otherwise. When SOLCON-
TROL,OFF, defaults to 1.0 for force, moment, and volume convergence. The default for heat flow (1.0E-
6), VOLT and AMPS (1.0E-12), and others are independent of the SOLCONTROL setting.
Command Default
For static or transient analysis, check the out-of-balance load for any active DOF using the default VALUE,
TOLER, NORM, and MINREF. Also check the displacement convergence for some problems. For analyses
that include CONTA17x contact elements, displacement/rotation convergence checking is off by default. For
harmonic magnetic analysis, check the out-of-balance of the degrees of freedom.
Notes
The default values given for this command assume SOLCONTROL,ON (the default). See the description of
SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and SOLCONTROL,OFF.
Values may be set for the degrees of freedom (DOF) and/or the out-of-balance load for the corresponding
forcing quantities. When the GUI is on, if a "Delete" operation in a Nonlinear Convergence Criteria dialog
box writes this command to a log file (Jobname.LOG or Jobname.LGW), you will observe that Lab is
blank, VALUE = -1, and TOLER is an integer number. In this case, the GUI has assigned a value of TOLER
that corresponds to the location of a chosen convergence label in the dialog box's list. It is not intended
that you type in such a location value for TOLER in an ANSYS session. However, a file that contains a GUI-
generated CNVTOL command of this form can be used for batch input or with the /INPUT command.
If you have multiple DOF types, issue CNVTOL for each DOF, because the default will be reset for the other
DOFs the first time CNVTOL is issued.
This command is also valid in PREP7.
Convergence norms specified with CNVTOL may be graphically tracked while the solution is in process using
the ANSYS program's Graphical Solution Tracking (GST) feature. Use the /GST command to turn GST on or
off. By default, GST is ON for interactive sessions and OFF for batch runs.
293
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CNVTOL
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Convergence Crit
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Harmonic
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Static
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Transient
Main Menu>Solution>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Solution>Load Step Opts>Nonlinear>Convergence Crit
Main Menu>Solution>Load Step Opts>Nonlinear>Harmonic
Main Menu>Solution>Load Step Opts>Nonlinear>Static
Main Menu>Solution>Load Step Opts>Nonlinear>Transient
/COLOR, Lab, Clab, N1, N2, NINC
Specifies the color mapping for various items.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Apply color to the items specified by the following labels:
AXES
Determines the color (specified in next argument, Clab) that the axes of a graph will be plotted in.
AXNUM
Determines the color (specified in next argument, Clab) that the numbering on the axes of a graph
will be plotted in.
NUM
Discretely numbered items (such as element types, element materials, etc., as shown on the /PNUM
command). Also specify number (1 to 11) in the N1 field. For example, /COLOR,NUM,RED,3 will assign
the color red to all items having the discrete number 3 (material displays would show elements
having material 3 as red).
OUTL
Outline of elements, areas, and volumes. Ex: /COLOR,OUTL,BLUE.
ELEM
Elements. Use N1, N2, NINC fields for element numbers.
LINE
Solid model lines. Use N1, N2, NINC fields for line numbers.
AREA
Solid model areas. Use N1, N2, NINC fields for area numbers.
VOLU
Solid model volumes. Use N1, N2, NINC fields for volume numbers.
ISURF
Isosurfaces (surfaces of constant stress, etc.). This option is particularly useful when capturing frames
for animating a single isosurface value.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
294
/COLOR
WBAK
Window background. Use N1, N2, NINC fields for window numbers. The options that you select using
Lab = PBAK will supersede those applied using Lab = WBAK.
b.c.label
Boundary condition label. Enter U, ROT, TEMP, PRES, V, ENKE, ENDS, SP01 through SP06 or their user-
defined names, VOLT, MAG, A, EMF, CURR, F, M, HEAT, FLOW, VF, AMPS, FLUX, CSG, CURT, VLTG,
MAST, CP, CE, NFOR, NMOM, RFOR, RMOM, PATH. See the /PBC command for boundary condition
label definitions.
GRBAK
Graph background.
GRID
Graph grid lines.
AXLAB
Graph X and Y axis labels.
CURVE
Graph curves (identify curve numbers (1-10) in N1, N2, NINC fields).
CM
Component group. Use N1 field for component name, ignore N2 and NINC.
CNTR
ANSYS contour stress colors. The maximum number of contours available is 128. The number of
colors that can be specified interactively (GUI) is 9. (/CONTOUR, , 9). Any other setting will yield in-
consistent results.
SMAX
Specifies that all stress values above the maximum value entered in /CONTOUR will be displayed in
the color designated in the Clab field. Defaults to dark grey.
SMIN
Specifies that all stress values below the minimum value entered in /CONTOUR will be displayed in
the color designated in the Clab field. Defaults to dark grey.
PBAK
Activates background shading options (see command syntax at end of argument descriptions below).
The options that you select using Lab = PBAK will supersede those applied using Lab = WBAK.
Clab
Valid color labels are:
BLAC (0)
Black
MRED (1)
Magenta-Red
MAGE (2)
Magenta
BMAG (3)
Blue-Magenta
BLUE (4)
Blue
295
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/COLOR
CBLU (5)
Cyan-Blue
CYAN (6)
Cyan
GCYA ((7)
Green-Cyan
GREE (8)
Green
YGRE (9)
Yellow-Green
YELL (10)
Yellow
ORAN (11)
Orange
RED (12)
Red
DGRA (13)
Dark Gray
LGRA (14)
Light Gray
WHIT (15)
White
N1, N2, NINC
Apply color to Lab items numbered N1 to N2 (defaults to N1) in steps of NINC (defaults to 1). If N1 is
blank, apply color to entire selected range. If Lab is CM, use component name for N1 and ignore N2
and NINC. If N1 = P, graphical picking of elements, lines, areas and volumes is enabled; your can assign
colors to the entities via the picker. When picking is enabled, the Lab and Clab fields are ignored.
If Lab = PBAK, the command format is /COLOR, PBAK,Key_On_Off, KEY_TYPE, KEY_INDEX.
The options that you select using Lab = PBAK will supersede those applied using Lab = WBAK.
Key_On_Off
Turns the background colors on and off. Acceptable values are ON (1) and OFF (0).
KEY_TYPE
Determines the type of background. Acceptable values are 0 (smooth shading left to right), 1 (smooth
shading top to bottom), 2 (smooth shading right to left), 3 (smooth shading bottom to top), and -1
(textured image background)
KEY_INDEX
An integer value that corresponds to a background color or texture. If Key_Type is -1, the background
will correspond to values specified in the /TXTRE command. If Key_Type is any other acceptable value,
the background will correspond to the color values listed above under Clab.
Command Default
Use the default color mapping.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
296
/COLOR
Notes
Issue /COLOR,STAT to display the current color mapping. Issue /COLOR,DEFA to reset the default color
mapping.
Note
Color labels may also be reassigned any "color" with the /CMAP command.
This command is valid anywhere.
Menu Paths
Utility Menu>PlotCtrls>Style>Colors>BC Colors
Utility Menu>PlotCtrls>Style>Colors>Component Colors
Utility Menu>PlotCtrls>Style>Colors>Entity Colors
Utility Menu>PlotCtrls>Style>Colors>Graph Colors
Utility Menu>PlotCtrls>Style>Colors>Numbered Item Colors
Utility Menu>PlotCtrls>Style>Colors>Window Colors
/COM, Comment
Places a comment in the output.
SESSION: List Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Comment
Comment string, up to 75 characters.
Notes
The output from this command consists of the comment string. This command is similar to C*** except that
the comment produced by C*** is more easily identified in the output. Parameter substitution within the
comment occurs for every valid expression delimited by percent (%) signs. Enclosing such an expression in
single quotes prevents parameter substitution.
Another way to include a comment is to precede it with a ! character (on the same line). The ! may be placed
anywhere on the line, and any input following it is ignored as a comment. No output is produced by such
a comment, but the comment line is included on the log file. This is a convenient way to annotate the log
file.
This command is valid anywhere.
Menu Paths
This command cannot be accessed from a menu.
297
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/COM
COMPRESS
Deletes all specified sets.
AUX3: Results Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Issue this command to delete all sets specified with the DELETE command.
Menu Paths
This command cannot be accessed from a menu.
CON4, XCENTER, YCENTER, RAD1, RAD2, DEPTH
Creates a conical volume anywhere on the working plane.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XCENTER, YCENTER
Working plane X and Y coordinates of the center axis of the cone.
RAD1, RAD2
Radii of the faces of the cone. RAD1 defines the bottom face and will be located on the working plane.
RAD2 defines the top face and is parallel to the working plane. A value of zero or blank for either RAD1
or RAD2 defines a degenerate face at the center axis (i.e., the vertex of the cone). The same value for
both RAD1 and RAD2 defines a cylinder instead of a cone.
DEPTH
The perpendicular distance (either positive or negative based on the working plane Z direction) from
the working plane representing the depth of the cone. DEPTH cannot be zero (see "Notes" (p. 298) below).
Notes
Defines a solid conical volume with either the vertex or a face anywhere on the working plane. The cone
must have a spatial volume greater than zero. (i.e., this volume primitive command cannot be used to create
a degenerate volume as a means of creating an area.) The face or faces will be circular (each area defined
with four lines), and they will be connected with two areas (each spanning 180). See the CONE command
for an alternate way to create cones.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Primitives>Cone
Main Menu>Preprocessor>Modeling>Create>Volumes>Cone>By Picking
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
298
COMPRESS
CONE, RBOT, RTOP, Z1, Z2, THETA1, THETA2
Creates a conical volume centered about the working plane origin.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RBOT, RTOP
Radii of the bottom and top faces of the cone. A value of zero or blank for either RBOT or RTOP defines
a degenerate face at the center axis (i.e., the vertex of the cone). The same value for both RBOT and
RTOP defines a cylinder instead of a cone.
Z1, Z2
Working plane Z coordinates of the cone. The smaller value is always associated with the bottom face.
THETA1, THETA2
Starting and ending angles (either order) of the cone. Used for creating a conical sector. The sector begins
at the algebraically smaller angle, extends in a positive angular direction, and ends at the larger angle.
The starting angle defaults to 0 and the ending angle defaults to 360. See the Modeling and Meshing
Guide for an illustration.
Notes
Defines a solid conical volume centered about the working plane origin. The non-degenerate face (top or
bottom) is parallel to the working plane but not necessarily coplanar with (i.e., "on") the working plane. The
cone must have a spatial volume greater than zero. (i.e., this volume primitive command cannot be used to
create a degenerate volume as a means of creating an area.) For a cone of 360, top and bottom faces will
be circular (each area defined with four lines), and they will be connected with two areas (each spanning
180). See the CON4 command for an alternate way to create cones.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Cone>By Dimensions
/CONFIG, Lab, VALUE
Assigns values to ANSYS configuration parameters.
SESSION: Run Controls
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
Lab
Configuration parameter to be changed:
NRES
VALUE is maximum number of results sets (substeps) allowed on the result file. Defaults to 1000.
For MFX analyses, defaults to 5000. Minimum is 10.
NORSTGM
Option to write or not write geometry data to the results file. VALUE is either 0 (write geometry
data) or 1 (do not write geometry data). Useful when complex analyses will create abnormally large
files. Default is 0.
299
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/CONFIG
NBUF
VALUE is the number of buffers (1 to 32) per file in the solver. Defaults to 4.
LOCFL
File open and close actions. For VALUE use: 0 for global (default); 1 for local. Applicable to
File.EROT, File.ESAV, and File.EMAT. Typically used for large problems where locally closed
files may be deleted earlier in the run with the /FDELE command.
SZBIO
VALUE is the record size (1024 to 4194304) of binary files (in integer words). Defaults to 16384
(system dependent).
ORDER
Automatic reordering scheme. For VALUE use: 0 for WSORT,ALL; 1 for WAVES; 2 for both WSORT,ALL
and WAVES (default).
FSPLIT
Defines split points for binary files. VALUE is the file split point in megawords and defaults to the
maximum file size for the system.
MXND
Maximum number of nodes. If not specified, defaults to 100 at first encounter. Dynamically expanded
by doubling, even at first encounter, when maximum is exceeded.
MXEL
Maximum number of elements. Default and expansion as for MXND.
MXKP
Maximum number of keypoints. Default and expansion as for MXND.
MXLS
Maximum number of lines. Default and expansion as for MXND.
MXAR
Maximum number of areas. Default and expansion as for MXND.
MXVL
Maximum number of volumes. Default and expansion as for MXND.
MXRL
Maximum number of sets of real constants (element attributes). Default and expansion as for MXND.
MXCP
Maximum number of sets of coupled degrees of freedom. Default and expansion as for MXND.
MXCE
Maximum number of constraint equations. Default and expansion as for MXND.
NOELDB
Option to write or not write results into the database after a solution. When VALUE = 0 (default),
write results into the database. When VALUE = 1, do not write results into the database.
DYNA_DBL
Option to invoke the double precision version of the explicit dynamics solver LS-DYNA. When VALUE
= 0 (default), the single precision version is used. When VALUE = 1, the double precision version is
used.
STAT
Displays current values set by the /CONFIG command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
300
/CONFIG
VALUE
Value (an integer number) assigned to the configuration parameter.
Notes
All configuration parameters have initial defaults, which in most cases do not need to be changed. Where
a specially configured version of the ANSYS program is desired, the parameters may be changed with this
command. Issue /CONFIG,STAT to display current values. Changes must be defined before the parameter
is required. These changes (and others) may also be incorporated into the config120.ans file which is
read upon execution of the program (see The Configuration File in the Basic Analysis Guide). If the same
configuration parameter appears in both the configuration file and this command, this command overrides.
Distributed ANSYS uses the default FSPLIT value, and sets NOELDB = 1 and NORSTGM = 1. The FSPLIT,
NOELDB, and NORSTGM options cannot be changed when using Distributed ANSYS.
The /CONFIG command is not valid for the ANSYS Multiphysics 1, 2, or 3 products.
The ANSYS Multi-field solver (MFS and MFX) does not support /CONFIG,NOELDB,1. The ANSYS Multi-field
solver needs the updated ANSYS database.
Product Restrictions
Available Products Command
Option Lab
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS MXND
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS MXEL
Menu Paths
Utility Menu>List>Status>Configuration
CONJUG, IR, IA, --, --, Name, --, --, FACTA
Forms the complex conjugate of a variable.
POST26: Operations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA
Reference number of the variable to be operated on.
--, --
Unused fields.
Name
Thirty-two character name for identifying the variable on printouts and displays. Embedded blanks are
compressed for output.
301
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CONJUG
--, --
Unused fields.
FACTA
Scaling factor (positive or negative) applied to variable (default to 1.0).
Notes
Used only with harmonic analyses (ANTYPE,HARMIC).
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Complex Conjugat
/CONTOUR, WN, NCONT, VMIN, VINC, VMAX
Specifies the uniform contour values on stress displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
NCONT
Number of contour values. NCONT defaults to 9 for X11 or WIN32 and to 128 for X11c or WIN32C. The
default graphics window display for 3-D devices is a smooth continuous shading effect that spans the
maximum of 128 contours available. Use the /DV3D command to create defined banding for your contour
values (values of 9 and 128 are displayed in smooth shading only). The legend, however, will display
only nine color boxes, which span the full range of colors displayed in the graphics window.
VMIN
Minimum contour value. If VMIN = AUTO, automatically calculate contour values based upon NCONT
uniformly spaced values over the min-max extreme range. Or, if VMIN = USER, set contour values to
those of the last display (useful when last display automatically calculated contours).
VINC
Value increment (positive) between contour values. Defaults to (VMAX-VMIN)/NCONT.
VMAX
Maximum contour value. Ignored if both VMIN and VINC are specified.
Command Default
Nine contour values uniformly spaced between the extreme values, or no contours if the ratio of range to
minimum value (or range to maximum if minimum = 0) is less than 0.001.
Notes
See the /CVAL command for alternate specifications. Values represent contour lines in vector mode, and
the algebraic maximum of contour bands in raster mode.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
302
/CONTOUR
Note
No matter how many contours (NCONT) are specified by /CONTOUR, the actual number of contours
that appear on your display depends also on the device name, whether the display is directed
to the screen or to a file, the display mode (vector or raster), and the number of color planes. (All
these items are controlled by /SHOW settings.) In any case, regardless of whether they are
smoothed or banded, only 128 contours can be displayed. See Creating Geometric Results Displays
in the Basic Analysis Guide for more information on changing the number of contours.
If the current ANSYS graphics are not displayed as Multi-Plots, then the following is true: If the current device
is a 3-D device [/SHOW,3D], the model contours in all active windows will be the same, even if separate
/CONTOUR commands are issued for each active window. For efficiency, ANSYS 3-D graphics logic maintains
a single data structure (segment), which contains precisely one set of contours. The program displays the
same segment in all windows. The view settings of each window constitute the only differences in the
contour plots in the active windows.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Contours>Uniform Contours
/COPY, Fname1, Ext1, --, Fname2, Ext2, --
Copies a file.
SESSION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname1
File name to be copied and its directory path (248 characters maximum for both file name and directory).
If you do not specify a directory path, it will default to your working directory and you can use all 248
characters for the file name.
The file name defaults to the current Jobname.
Ext1
Filename extension (8 character maximum).
--
Unused field.
Fname2
File name to be created and its directory path (248 characters maximum for both file name and directory).
If you do not specify a directory path, it will default to your working directory and you can use all 248
characters for the file name.
Fname2 defaults to Fname1.
Ext2
Filename extension (8 character maximum).
Ext2 defaults to Ext1.
303
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/COPY
--
Unused field.
Notes
The original file is untouched. Ex: /COPY,A,,,B copies file A to B in the same directory. /COPY,A,DAT,,,INP
copies the file A.DAT to A.INP. See the Operations Guide for details. ANSYS binary and ASCII files can be
copied.
Menu Paths
Utility Menu>File>File Operations>Copy
CORIOLIS, Option, --, --, RefFrame, RotDamp
Applies the Coriolis effect to a rotating structure.
SOLUTION: Inertia
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Option
Flag to activate or deactivate the Coriolis effect:
1 (ON or YES)
Activate. This value is the default.
0 (OFF or NO)
Deactivate.
--, --
Unused fields.
RefFrame
Flag to activate or deactivate a stationary reference frame.
1 (ON or YES)
Activate.
0 (OFF or NO)
Deactivate. This value is the default.
RotDamp
Flag to activate or deactivate rotating damping effect.
1 (ON or YES)
Activate.
0 (OFF or NO)
Deactivate. This value is the default.
Notes
The CORIOLIS command is used for analyses in either a rotating or a stationary reference frame, and performs
differently according to the designated RefFrame value. Specific restrictions and elements apply to each
case, as follows:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
304
CORIOLIS
Rotating Reference Frame (RefFrame = OFF):
The command applies the Coriolis effect in the following structural element types: MASS21, SHELL181,
PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL281, PIPE288
and PIPE289. It also applies this effect in the PLANE223, SOLID226, and SOLID227 analyses with structural
degrees of freedom.
In a rotating reference frame, both the coriolis and spin-softening effects contribute to the gyroscopic moment.
Therefore, ANSYS applies spin-softening by default for dynamic analyses. (To deactivate spin-softening, issue
the OMEGA or CMOMEGA command, setting KSPIN = 0.). If a rotational velocity is specified (or), centrifugal
forces will be included.
To include coriolis effects in a large deflection prestressed analysis (NLGEOM, ON), issue both the CORIOLIS
and the OMEGA (or CMOMEGA) commands in the static prestress portion of the analysis.
In a nonlinear transient analysis (ANTYPE,TRANS and NLGEOM, ON), any spinning motion applied through
either the IC of the D commands will include the coriolis effect without having to issue the CORIOLIS
command. Refer to Rotating Structure Analysis in the Advanced Analysis Techniques Guide for more information.
Stationary Reference Frame (RefFrame = ON):
The command activates the gyroscopic damping matrix in the following structural elements: BEAM4, PIPE16,
MASS21, SHELL63, BEAM188, SHELL181, BEAM189, SOLID45, SOLID95, SOLID185, SOLID186, SOLID187, SOL-
ID272, SOLID273, SHELL281, PIPE288, and PIPE289.
The rotating structure must be axisymmetric about the axis of rotation.
Static analysis (ANTYPE, STATIC) does not support Coriolis effects with a stationary reference frame. However,
you can include the gyroscopic effects in a small deflection prestresses analysis (NLGEOM, ON and PSTRES,
ON) by issuing the CORIOLIS command in the static prestressed dynamic analysis.
Rotating damping effect (RotDamp = ON) applies only for the stationary reference frame. Therefore, this
effect is supported only by the elements listed above that generate a gyroscopic damping matrix. Propor-
tional damping must be present in the element (MP,DAMP or BETAD). It is also supported by element
COMBI214 with non zero and axisymmetric damping characteristics (non zero real constants C11=C22 and
C21=C12=0).
For more information about using the CORIOLIS command, see Rotating Structure Analysis in the Advanced
Analysis Techniques Guide and also in the Rotordynamic Analysis Guide. For details about the Coriolis and
gyroscopic effects, see the Theory Reference for the Mechanical APDL and Mechanical Applications.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>Coriolis
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>Coriolis
305
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CORIOLIS
COUPLE
Specifies "Node coupling" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Coupled Sets
COVAL, TBLNO1, TBLNO2, SV1, SV2, SV3, SV4, SV5, SV6, SV7
Defines PSD cospectral values.
SOLUTION: Spectrum Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
TBLNO1
First input PSD table number associated with this spectrum.
TBLNO2
Second input PSD table number associated with this spectrum.
SV1, SV2, SV3, . . . , SV7
PSD cospectral values corresponding to the frequency points [PSDFRQ].
Notes
Defines PSD cospectral values to be associated with the previously defined frequency points. Two table
references are required since values are off-diagonal terms. Unlike autospectra [PSDVAL], the cospectra can
be positive or negative. The cospectral curve segment where there is a sign change is interpolated linearly
(the rest of the curve segments use log-log interpolation). For better accuracy, choose as small a curve segment
as possible wherever a sign change occurs.
Repeat COVAL command using the same table numbers for additional points. This command is valid for
SPOPT,PSD only.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
306
COUPLE
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>PSD>Correlation>Cospectral
Main Menu>Solution>Load Step Opts>Spectrum>PSD>Correlation>Cospectral
CP, NSET, Lab, NODE1, NODE2, NODE3, NODE4, NODE5, NODE6, NODE7, NODE8, NODE9, NODE10, NODE11,
NODE12, NODE13, NODE14, NODE15, NODE16, NODE17
Defines (or modifies) a set of coupled degrees of freedom.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET
Set reference number:
n
Arbitrary set number.
HIGH
The highest defined coupled set number will be used (default, unless Lab = ALL). This option is
useful when adding nodes to an existing set.
NEXT
The highest defined coupled set number plus one will be used (default if Lab = ALL). This option
automatically numbers coupled sets so that existing sets are not modified.
Lab
Degree of freedom label for coupled nodes (in the nodal coordinate system). Defaults to label previously
defined with NSET if set NSET already exists. A different label redefines the previous label associated
with NSET. Valid labels are: Structural labels: UX, UY, or UZ (displacements); ROTX, ROTY, or ROTZ (rota-
tions) (in radians); HDSP (hydrostatic pressure). Thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP (temper-
ature). Fluid labels: PRES (pressure); VX, VY, or VZ (velocities). Electric labels: VOLT (voltage); EMF (electro-
motive force drop); CURR (current). Magnetic labels: MAG (scalar magnetic potential); AX, AY, or AZ
(vector magnetic potentials); CURR (current). Explicit analysis labels: UX, UY, or UZ (displacements).
When Lab = ALL:
Sets are generated for each active degree of freedom (that is, one set for the UX degree of freedom,
another set for UY, etc.), and NSET is incremented automatically to prevent overwriting existing
sets.
Existing sets are not modified. NSET must be a new set number n or NEXT.
The degree of freedom set is determined according to all element types defined and the DOF
command, if used.
Hydrostatic pressure (HDSP) is not included.
ALL is the only label applicable to FLOTRAN.
NODE1, NODE2, NODE3, . . . , NODE17
List of nodes to be included in set. Duplicate nodes are ignored. If a node number is input as negative,
the node is deleted from the coupled set. The first node in the list is the primary (retained) node. If
NODE1 = ALL, NODE2 through NODE17 are ignored and all selected nodes [NSEL] are included in the
307
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CP
set. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only
in the GUI). A component name may also be substituted for NODE1.
Notes
Do not include the same degree of freedom in more than one coupled set. Repeat CP command for addi-
tional nodes.
Coupling degrees of freedom into a set causes the results calculated for one member of the set to be the
same for all members of the set. Coupling can be used to model various joint and hinge effects. A more
general form of coupling can be done with constraint equations [CE]. For structural analyses, a list of nodes
is defined along with the nodal directions in which these nodes are to be coupled. As a result of this coupling,
these nodes are forced to take the same displacement in the specified nodal coordinate direction. The amount
of the displacement is unknown until the analysis is completed. A set of coupled nodes which are not coin-
cident, or which are not along the line of the coupled displacement direction, may produce an applied
moment which will not appear in the reaction forces. The actual degrees of freedom available for a particular
node depends upon the degrees of freedom associated with element types [ET] at that node. For example,
degrees of freedom available with BEAM3 elements are UX, UY, and ROTZ only. For scalar field analysis, this
command is used to couple nodal temperatures, pressures, voltages, etc.
For an explicit dynamic analysis, the only valid DOF labels for coupling are UX, UY, and UZ. Since the rota-
tional DOF (ROTX, ROTY, ROTZ) are not allowed, The CP family of commands should not be used in an explicit
analysis to model rigid body behavior that involves rotations. If CP is used in this manner, it could lead to
nonphysical responses.
A set of coupled nodes which are not coincident, or which are not along the line of the coupled displacement
direction, produce an artificial moment constraint. If the structure rotates, a moment may be produced in
the coupled set in the form of a force couple. This moment is in addition to the real reaction forces and
may make it appear that moment equilibrium is not satisfied by just the applied forces and the reaction
forces. Note, however, that in an explicit dynamic analysis, this artificial moment will not be produced. Rather,
just the applied forces and the reaction forces will satisfy the moment equilibrium in the model. Thus, in an
explicit analysis, the magnitude of nodal displacements for this set of nodes will depend on the distance
from each node to the center of the coupled set, and the direction of displacement will depend on the
resulting moment. This may lead to a nonphysical response in some cases.
Additional sets of coupled nodes may be generated from a specified set. Degrees of freedom are coupled
within a set but are not coupled between sets. No degree of freedom should appear in more than one
coupled set. Such an appearance would indicate that at least two sets were in fact part of a single larger
set. The first degree of freedom of the coupled set is the "prime" degree of freedom. All other degrees of
freedom in the coupled sets are eliminated from the solution matrices by their relationship to the prime
degree of freedom. Forces applied to coupled nodes (in the coupled degree of freedom direction) will be
summed and applied to the prime degree of freedom. Output forces are also summed at the prime degree
of freedom. Degrees of freedom with specified constraints [D] should not be included in a coupled set (unless
the degree of freedom is prime).
If master degrees of freedom are defined for coupled nodes, only the prime degree of freedom should be
so defined. The use of coupled nodes reduces the set of coupled degrees of freedom to only one degree
of freedom.
In FLOTRAN, you can apply periodic boundary conditions using the CP command along with the PERI macro.
Attempts to use the CP command outside the context of the PERI macro may lead to unexpected results.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
308
CP
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Couple DOFs
Main Menu>Preprocessor>Coupling / Ceqn>Cupl DOFs w/Mstr
CPCYC, Lab, TOLER, KCN, DX, DY, DZ, KNONROT
Couples the two side faces of a cyclically symmetric model for loadings that are the same on every segment.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Degree of freedom label for coupled nodes (in the nodal coordinate system). If ALL, use all appropriate
labels. Valid labels are: Structural labels: UX, UY, or UZ (displacements); ROTX, ROTY, or ROTZ (rotations,
in radians).
TOLER
Tolerance for coincidence (based on maximum coordinate difference in each global Cartesian direction
for node locations and on angle differences for node orientations). Defaults to 0.0001. Only nodes
within the tolerance are considered to be coincident for coupling.
KCN
In coordinate system KCN, node 1 of CP + dx dy dz = node 2 of CP.
DX, DY, DZ
Node location increments in the active coordinate system (DR, D, DZ for cylindrical, DR, D , D for
spherical or toroidal).
KNONROT
When KNONROT = 0, the nodes on coupled sets are rotated into coordinate system KCN (see NROTAT
command description). When KNONROT = 1, the nodes are not rotated, and you should make sure that
coupled nodal DOF directions are correct.
Notes
Cyclic coupling requires identical node and element patterns on the low and high sector boundaries. The
MSHCOPY operation allows convenient generation of identical node and element patterns. See Using CPCYC
and MSHCOPY Commands in the Modeling and Meshing Guide for more information.
Although developed primarily for use with cyclically symmetric models, your use of the CPCYC command
is not limited to cyclic symmetry analyses.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Offset Nodes
309
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CPCYC
CPDELE, NSET1, NSET2, NINC, Nsel
Deletes coupled degree of freedom sets.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET1, NSET2, NINC
Delete coupled sets from NSET1 to NSET2 (defaults to NSET1) in steps of NINC (defaults to 1). If NSET1
= ALL, NSET2 and NINC are ignored and all coupled sets are deleted.
Nsel
Additional node selection control:
ANY
Delete coupled set if any of the selected nodes are in the set (default).
ALL
Delete coupled set only if all of the selected nodes are in the set.
Notes
See the CP command for a method to delete individual nodes from a set.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Del Coupled Sets
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
CPINTF, Lab, TOLER
Defines coupled degrees of freedom at an interface.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Degree of freedom label for coupled nodes (in the nodal coordinate system). If ALL, use all appropriate
labels except HDSP. Valid labels are: Structural labels: UX, UY, or UZ (displacements); ROTX, ROTY, or
ROTZ (rotations, in radians), HDSP (hydrostatic pressure). Thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP
(temperature). Fluid labels: PRES (pressure); VX, VY, or VZ (velocities). Electric labels: VOLT (voltage); EMF
(electromotive force drop); CURR (current). Magnetic labels: MAG (scalar magnetic potential); AX, AY, or
AZ (vector magnetic potentials); CURR (current).
TOLER
Tolerance for coincidence (based on maximum coordinate difference in each global Cartesian direction
for node locations and on angle differences for node orientations). Defaults to 0.0001. Only nodes
within the tolerance are considered to be coincident for coupling.
Notes
Defines coupled degrees of freedom between coincident nodes (within a tolerance). May be used, for example,
to "button" together elements interfacing at a seam, where the seam consists of a series of node pairs. One
coupled set is generated for each selected degree of freedom for each pair of coincident nodes. For more
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
310
CPDELE
than two coincident nodes in a cluster, a coupled set is generated from the lowest numbered node to each
of the other nodes in the cluster. Coupled sets are generated only within (and not between) clusters. If
fewer than all nodes are to be checked for coincidence, use the NSEL command to select nodes. Coupled
set reference numbers are incremented by one from the highest previous set number. Use CPLIST to display
the generated sets. Only nodes having the same nodal coordinate system orientations ("coincident" within
a tolerance) are included. Use the CEINTF command to connect nodes by constraint equations instead of
by coupling. Use the EINTF command to connect nodes by line elements instead of by coupling.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Coincident Nodes
/CPLANE, KEY
Specifies the cutting plane for section and capped displays.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Specifies the cutting plane:
0
Cutting plane is normal to the viewing vector [/VIEW] and passes through the focus point [/FOCUS]
(default).
1
The working plane [WPLANE] is the cutting plane.
Command Default
The cutting plane is normal to the viewing vector at the focus point.
Notes
Defines the cutting plane to be used for section and capped displays [/TYPE,,(1, 5, or 7)].
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Hidden-Line Options
CPLGEN, NSETF, Lab1, Lab2, Lab3, Lab4, Lab5
Generates sets of coupled nodes from an existing set.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSETF
Generate sets from existing set NSETF.
311
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CPLGEN
Lab1, Lab2, Lab3, . . . , Lab5
Generate sets with these labels (see CP command for valid labels). Sets are numbered as the highest
existing set number + 1.
Notes
Generates additional sets of coupled nodes (with different labels) from an existing set [CP, CPNGEN]. The
same node numbers are included in the generated sets. If all labels of nodes are to be coupled and the
nodes are coincident, the NUMMRG command should be used to automatically redefine the node number
(for efficiency).
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Gen w/Same Nodes
CPLIST, NSET1, NSET2, NINC, Nsel
Lists the coupled degree of freedom sets.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET1, NSET2, NINC
List coupled sets from NSET1 to NSET2 (defaults to NSET1) in steps of NINC (defaults to 1). If NSET1
= ALL (default), NSET2 and NINC are ignored and all coupled sets are listed.
Nsel
Node selection control:
ANY
List coupled set if any of the selected nodes are in the set (default).
ALL
List coupled set only if all of the selected nodes are in the set.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>List>Other>Coupled Sets>All CP nodes selected
Utility Menu>List>Other>Coupled Sets>Any CP node selected
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
312
CPLIST
CPMERGE, Lab
Merges different couple sets with duplicate degrees of freedom into one couple set.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Degree of freedom label for coupled nodes (in the nodal coordinate system). Valid labels are: Structural
labels: UX, UY, or UZ (displacements); ROTX, ROTY, or ROTZ (rotations) (in radians). Thermal labels: TEMP,
TBOT, TE2, TE3, . . ., TTOP (temperature). Fluid labels: PRES (pressure); VX, VY, or VZ (velocities). Electric
labels: VOLT (voltage); EMF (electromotive force drop); CURR (current). Magnetic labels: MAG (scalar
magnetic potential); AX, AY, or AZ (vector magnetic potentials); CURR (current). Explicit analysis labels:
UX, UY, or UZ (displacements). The degree of freedom set is determined from all element types defined
and the DOF command, if used.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Capacitor
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Curr Cntl CS
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Curr Cntl VS
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Diode
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Curr Src>Constant
Amplitude
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Curr Src>Exponential
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Curr Src>Piecewise
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Curr Src>Pulse
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Curr Src>Sinusoidal
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Vltg Src>Constant
Amplitude
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Vltg Src>Exponential
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Vltg Src>Piecewise
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Vltg Src>Pulse
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Indp Vltg Src>Sinusoidal
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Inductor
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Mass Cond 2D
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Mass Cond 3D
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Mutual Ind
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Resistor
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Strnd Coil
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Vltg Cntl CS
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Vltg Cntl VS
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Wire
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Electric>Zener Diode
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Damper>Linear Rotary
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Damper>Linear Trans
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Damper>Nonlin Rotary
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Damper>Nonlin Trans
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Damper>Slide Film
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Mass
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Spring>Linear Rotary
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Spring>Linear Trans
313
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CPMERGE
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Spring>Nonlin Rotary
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Spring>Nonlin Trans
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Capacitor
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Curr Src>Constant
Amplitude
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Curr Src>Exponential
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Curr Src>Piecewise
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Curr Src>Pulse
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Curr Src>Sinusoidal
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Vltg Src>Constant
Amplitude
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Vltg Src>Exponential
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Vltg Src>Piecewise
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Vltg Src>Pulse
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Indp Vltg Src>Sinusoidal
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Inductor
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Resistor
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Piezoelectric>Wire
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Transducer>ElecMech
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
CPNGEN, NSET, Lab, NODE1, NODE2, NINC
Defines, modifies, or adds to a set of coupled degrees of freedom.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET
Set reference number [CP].
Lab
Degree of freedom label [CP].
NODE1, NODE2, NINC
Include in coupled set nodes NODE1 to NODE2 in steps of NINC (defaults to 1). If NODE1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). If -NODE1, delete
range of nodes from set instead of including. A component name may also be substituted for NODE1
(NODE2 and NINC are ignored).
Notes
Defines, modifies, or adds to a set of coupled degrees of freedom. May be used in combination with (or in
place of ) the CP command. Repeat CPNGEN command for additional nodes.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
314
CPNGEN
CPSGEN, ITIME, INC, NSET1, NSET2, NINC
Generates sets of coupled nodes from existing sets.
PREP7: Coupled DOF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME, INC
Do this generation operation a total of ITIMEs, incrementing all nodes in the existing sets by INC each
time after the first. ITIME must be > 1 for generation to occur.
NSET1, NSET2, NINC
Generate sets from sets beginning with NSET1 to NSET2 (defaults to NSET1) in steps of NINC (defaults
to 1). If NSET1 is negative, NSET2 and NINC are ignored and the last |NSET1| sets (in sequence from
the maximum set number) are used as the sets to be repeated.
Notes
Generates additional sets of coupled nodes (with the same labels) from existing sets. Node numbers between
sets may be uniformly incremented.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Gen w/Same DOF
CQC, SIGNIF, Label
Specifies the complete quadratic mode combination method.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SIGNIF
Combine only those modes whose significance level exceeds the SIGNIF threshold. For single point,
multipoint, or DDAM response (SPOPT,SPRS, MPRS or DDAM), the significance level of a mode is defined
as the mode coefficient of the mode, divided by the maximum mode coefficient of all modes. Any mode
whose significance level is less than SIGNIF is considered insignificant and is not contributed to the
mode combinations. The higher the SIGNIF threshold, the fewer the number of modes combined.
SIGNIF defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0. (This mode combination
method is not valid for SPOPT,PSD.)
Label
Label identifying the combined mode solution output.
DISP
Displacement solution (default). Displacements, stresses, forces, etc., are available.
VELO
Velocity solution. Velocities, "stress velocities," "force velocities," etc., are available.
ACEL
Acceleration solution. Accelerations, "stress accelerations," "force accelerations," etc., are available.
315
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CQC
Notes
Damping is required for this mode combination method. The CQC command is also valid for PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Mode Combine
*CREATE, Fname, Ext, --
Opens (creates) a macro file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Do not use a directory path if file is to be read with the macro Name option of the *USE command.
Ext
Filename extension (8 character maximum).
Ext should not be used if file is to be read with the macro Name option of the *USE command.
--
Unused field.
Notes
See the *USE command for a discussion of macros. All commands following the *CREATE command, up to
the *END command, are written to the specified file without being executed. An existing file of the same
name, if any, will be overwritten. Parameter values are not substituted for parameter names in the commands
when the commands are written to the file. Use *CFWRITE to create a file if this is desired. The resulting
macro may be executed with a *USE command (which also allows parameters to be passed into the macro)
or a /INPUT command (which does not allow parameters to be passed in). Several macros may be stacked
into a library file [*ULIB]. You cannot use *CREATE within a DO loop.
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
316
*CREATE
Menu Paths
Utility Menu>Macro>Create Macro
CRPLIM, CRCR, Option
Specifies the creep criterion for automatic time stepping.
SOLUTION: Nonlinear Options
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS
CRCR
Value of creep criteria for the creep limit ratio control.
Option
Type of creep analysis:
1 or ON
Implicit creep analysis.
0 or OFF
Explicit creep analysis.
Notes
The creep ratio control can be used at the same time for implicit creep and explicit creep analyses. For im-
plicit creep (Option = 1), the default value of CRCR is zero (i.e., no creep limit control), and you are allowed
to specify any value. For explicit creep (Option = 0), the default value of CRCR is 0.1. The maximum value
allowed is 0.25. This command is also valid in PREP7. The CUTCONTROL command can be used for the same
purpose and is the preferred command, when SOLCONTROL is ON.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Creep Criterion
Main Menu>Solution>Load Step Opts>Nonlinear>Creep Criterion
CS, KCN, KCS, NORIG, NXAX, NXYPL, PAR1, PAR2
Defines a local coordinate system by three node locations.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Arbitrary reference number assigned to this coordinate system. Must be greater than 10. A coordinate
system previously defined with this number will be redefined.
KCS
Coordinate system type:
0 or CART
Cartesian
317
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CS
1 or CYLIN
Cylindrical (circular or elliptical)
2 or SPHE
Spherical (or spheroidal)
3 or TORO
Toroidal
NORIG
Node defining the origin of this coordinate system. If NORIG = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI).
NXAX
Node defining the positive x-axis orientation of this coordinate system.
NXYPL
Node defining the x-y plane (with NORIG and NXAX) in the first or second quadrant of this coordinate
system.
PAR1
Used for elliptical, spheroidal, or toroidal systems. If KCS = 1 or 2, PAR1 is the ratio of the ellipse Y-axis
radius to X-axis radius (defaults to 1.0 (circle)). If KCS = 3, PAR1 is the major radius of the torus.
PAR2
Used for spheroidal systems. If KCS = 2, PAR2 = ratio of ellipse Z-axis radius to X-axis radius (defaults
to 1.0 (circle)).
Notes
Defines and activates a local right-handed coordinate system by specifying three existing nodes: to locate
the origin, to locate the positive x-axis, and to define the positive x-y plane. This local system becomes the
active coordinate system. See the CLOCAL, CSKP, CSWPLA, and LOCAL commands for alternate definitions.
Local coordinate systems may be displayed with the /PSYMB command.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Local Coordinate Systems>Create Local CS>By 3 Nodes
CSCIR, KCN, KTHET, KPHI
Locates the singularity for non-Cartesian local coordinate systems.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Number of the local coordinate system in which singularity location is to be changed. Must be greater
than 10.
KTHET
Theta singularity location for cylindrical, spherical, and toroidal systems:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
318
CSCIR
0
Singularity at 180.
1
Singularity at 0 (360).
KPHI
Phi singularity location for toroidal systems:
0
Singularity in phi direction at 180.
1
Singularity in phi direction at 0 (360).
Command Default
Singularities at 180.
Notes
Continuous closed surfaces (circles, cylinders, spheres, etc.) have a singularity (discontinuity) at = 180.
For local cylindrical, spherical, and toroidal coordinate systems, this singularity location may be changed to
0 (360).
An additional, similar singularity occurs in the toroidal coordinate system at = 180 and can be moved
with KPHI. Additional singularities occur in the spherical coordinate system at = 90, but cannot be
moved.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Local Coordinate Systems>Move Singularity
CSDELE, KCN1, KCN2, KCINC
Deletes local coordinate systems.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN1, KCN2, KCINC
Delete coordinate systems from KCN1 (must be greater than 10) to KCN2 (defaults to KCN1) in steps of
KCINC (defaults to 1). If KCN1 = ALL, KCN2 and KCINC are ignored and all coordinate systems are
deleted.
Notes
This command is valid in any processor.
319
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CSDELE
Menu Paths
Main Menu>General Postproc>Surface Operations>Create Surface>Sphere>At Node
Utility Menu>WorkPlane>Local Coordinate Systems>Delete Local CS
CSKP, KCN, KCS, PORIG, PXAXS, PXYPL, PAR1, PAR2
Defines a local coordinate system by three keypoint locations.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Arbitrary reference number assigned to this coordinate system. Must be greater than 10. A coordinate
system previously defined with this number will be redefined.
KCS
Coordinate system type:
0 or CART
Cartesian
1 or CYLIN
Cylindrical (circular or elliptical)
2 or SPHE
Spherical (or spheroidal)
3 or TORO
Toroidal
PORIG
Keypoint defining the origin of this coordinate system. If PORIG = P, graphical picking is enabled and
all remaining command fields are ignored (valid only in the GUI).
PXAXS
Keypoint defining the positive x-axis orientation of this coordinate system.
PXYPL
Keypoint defining the x-y plane (with PORIG and PXAXS) in the first or second quadrant of this coordinate
system.
PAR1
Used for elliptical, spheroidal, or toroidal systems. If KCS = 1 or 2, PAR1 is the ratio of the ellipse Y-axis
radius to X-axis radius (defaults to 1.0 (circle)). If KCS = 3, PAR1 is the major radius of the torus.
PAR2
Used for spheroidal systems. If KCS = 2, PAR2 = ratio of ellipse Z-axis radius to X-axis radius (defaults
to 1.0 (circle)).
Notes
Defines and activates a local right-handed coordinate system by specifying three existing keypoints: to locate
the origin, to locate the positive x-axis, and to define the positive x-y plane. This local system becomes the
active coordinate system. See the CLOCAL, CS, CSWPLA, and LOCAL commands for alternate definitions.
Local coordinate systems may be displayed with the /PSYMB command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
320
CSKP
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Local Coordinate Systems>Create Local CS>By 3 Keypoints
CSLIST, KCN1, KCN2, KCINC
Lists coordinate systems.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN1, KCN2, KCINC
List coordinate systems from KCN1 to KCN2 (defaults to KCN1) in steps of KCINC (defaults to 1). If KCN1
= ALL (default), KCN2 and KCINC are ignored and all coordinate systems are listed.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>List>Other>Local Coord Sys
CSWPLA, KCN, KCS, PAR1, PAR2
Defines a local coordinate system at the origin of the working plane.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Arbitrary reference number assigned to this coordinate system. Must be greater than 10. A coordinate
system previously defined with this number will be redefined.
KCS
Coordinate system type:
0 or CART
Cartesian
1 or CYLIN
Cylindrical (circular or elliptical)
2 or SPHE
Spherical (or spheroidal)
3 or TORO
Toroidal
PAR1
Used for elliptical, spheroidal, or toroidal systems. If KCS = 1 or 2, PAR1 is the ratio of the ellipse Y-axis
radius to X-axis radius (defaults to 1.0 (circle)). If KCS = 3, PAR1 is the major radius of the torus.
321
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CSWPLA
PAR2
Used for spheroidal systems. If KCS = 2, PAR2 = ratio of ellipse Z-axis radius to X-axis radius (defaults
to 1.0 (circle)).
Notes
Defines and activates a local right-handed coordinate system centered at the origin of the working plane.
The coordinate system's local x-y plane (for a Cartesian system) or R- plane (for a cylindrical or spherical
system) corresponds to the working plane. This local system becomes the active coordinate system. See the
CS, LOCAL, CLOCAL, and CSKP commands for alternate ways to define a local coordinate system. Local
coordinate systems may be displayed with the /PSYMB command.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>Path Operations>Define Path>On Working Plane
Main Menu>Preprocessor>Path Operations>Define Path>On Working Plane
Utility Menu>WorkPlane>Local Coordinate Systems>Create Local CS>At WP Origin
CSYS, KCN
Activates a previously defined coordinate system.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Specifies the active coordinate system, as follows:
0 (default)
Cartesian
1
Cylindrical with Z as the axis of rotation
2
Spherical
4 or WP
Working Plane
5
Cylindrical with Y as the axis of rotation
11 or greater
Any previously defined local coordinate system
Notes
The CSYS command activates a previously defined coordinate system for geometry input and generation.
The LOCAL, CLOCAL, CS, CSKP, and CSWPLA commands also activate coordinate systems as they are
defined.
The active coordinate system for files created via the CDWRITE command is Cartesian (CSYS,0).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
322
CSYS
This command is valid in any processor.
CSYS,4 (or CSYS,WP) activates working plane tracking, which updates the coordinate system to follow
working plane changes. To deactivate working plane tracking, activate any other coordinate system (for ex-
ample, CSYS,0 or CSYS,11).
CSYS,5 is a cylindrical coordinate system with Y as the axis. The X, Y and Z axes are radial, theta, and axial
(respectively). However, CSYS,5 has a specific orientation: the local X is in the global +X direction, local Y is
in the global -Z direction, and local Z (the cylindrical axis) is in the global +Y direction. The R-Theta plane is
then the global X-Z plane, as it is for an axisymmetric model.
Menu Paths
Main Menu>General Postproc>Surface Operations>Create Surface>Sphere>At Node
Utility Menu>WorkPlane>Change Active CS to>Global Cartesian
Utility Menu>WorkPlane>Change Active CS to>Global Cylindrical
Utility Menu>WorkPlane>Change Active CS to>Global Spherical
Utility Menu>WorkPlane>Change Active CS to>Specified Coord Sys
Utility Menu>WorkPlane>Change Active CS to>Working Plane
Utility Menu>WorkPlane>Offset WP to>Global Origin
/CTYPE, KEY, DOTD, DOTS, DSHP, TLEN
Specifies the type of contour display.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Type of display:
0
Standard contour display.
1
Isosurface display.
2
Particle gradient display.
3
Gradient triad display.
DOTD
Maximum dot density for particle gradient display (KEY = 2). Density is expressed as dots per screen
width (defaults to 30).
DOTS
Dot size for particle gradient display (KEY = 2). Size is expressed as a fraction of the screen width (defaults
to 0.0 (single dot width)).
DSHP
Spherical dot shape precision for particle gradient display (KEY = 2). (3-D options are supported only
on 3-D devices):
323
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/CTYPE
0
Flat 2-D circular dot.
1
Flat-sided 3-D polyhedron.
n
3-D sphere with n (>1) polygon divisions per 90 of radius.
TLEN
Maximum length of triads for gradient triad display (KEY = 3). Value is expressed as a fraction of the
screen width (defaults to 0.067).
Command Default
Standard contour display.
Notes
Use /CTYPE,STAT to display the current settings. Only the standard contour display [/CTYPE,0) and the
isosurface contour display [/CTYPE,1] are supported by PowerGraphics [/GRAPHICS,POWER].
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Contours>Contour Style
CURR2D
Calculates current flow in a 2-D conductor.
POST1: Magnetics Calculations
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS
Notes
CURR2D invokes an ANSYS macro which calculates the total current flowing in a conducting body for a 2-
D planar or axisymmetric magnetic field analysis. The currents may be applied source currents or induced
currents (eddy currents). The elements of the conducting region must be selected before this command is
issued. The total current calculated by the macro is stored in the parameter TCURR. Also, the total current
and total current density are stored on a per-element basis in the element table [ETABLE] with the labels
TCURR and JT, respectively. Use the PLETAB and PRETAB commands to plot and list the element table
items.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Element Based>Current
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
324
CURR2D
CUTCONTROL, Lab, VALUE, Option
Controls time-step cutback during a nonlinear solution.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Lab
Specifies the criteria for causing a cutback. Possible arguments are
PLSLIMIT
Maximum equivalent plastic strain allowed within a time-step (substep). If the calculated value exceeds
the VALUE, ANSYS performs a cutback (bisection). VALUE defaults to 0.15 (15%).
CRPLIMIT
Maximum equivalent creep ratio allowed within a time step (substep). If the calculated value exceeds
VALUE, ANSYS performs a cutback (bisection). The default depends on whether you are performing
an implicit or explicit creep analysis (see Option argument below). The creep ratio control can be
used at the same time for implicit creep and explicit creep analyses. For an implicit creep analysis,
VALUE defaults to zero (i.e., no creep limit control), and you are allowed to enter any value. For an
explicit creep analysis, VALUE defaults to 0.1 (10%) and the maximum value allowed is 0.25.
DSPLIMIT
Maximum incremental displacement within the solution field in a time step (substep). If the maximum
calculated value exceeds VALUE, ANSYS performs a cutback (bisection). VALUE defaults to 1.0 x 10
7
.
NPOINT
Number of points in a cycle for a second order dynamic equation, used to control automatic time
stepping. If the number of solution points per cycle is less than VALUE, ANSYS performs a cutback
in time step size. VALUE defaults to 13.
This option works well for linear problems. For nonlinear analyses, other factors such as contact
status changes and solution convergence rate can overwrite NPOINT. See Automatic Time Stepping
in the Theory Reference for the Mechanical APDL and Mechanical Applications for more information on
automatic time stepping.
NOITERPREDICT
If VALUE is 0 (default), an internal auto time step scheme will predict the number of iterations for
nonlinear convergence and perform a cutback earlier than the number of iterations specified by the
NEQIT command. This is the recommended option. If VALUE is 1, the solution will iterate (if noncon-
vergent) to NEQIT number of iterations before a cutback is invoked. It is sometimes useful for poorly-
convergent problems, but rarely needed in general.
Bisection is also controlled by contact status change, plasticity or creep strain limit, and other factors.
If any of these factors occur, bisection will still take place, regardless of the NOITERPREDICT setting.
CUTBACKFACTOR
Changes the cutback value for bisection. Default is 0.5. VALUE must be greater than 0.0 and less
than 1.0. This option is active only if AUTOTS,ON is set.
VALUE
Numeric value for the specified cutback criterion. For Lab = CRPLIMIT, VALUE is the creep criteria for
the creep limit ratio control.
Option
Type of creep analysis. Option is valid for Lab = CRPLIMIT only.
325
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CUTCONTROL
1 or ON
Implicit creep analysis.
0 or OFF
Explicit creep analysis.
Notes
The default values given for this command assume SOLCONTROL,ON (the default). See the description of
SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and SOLCONTROL,OFF.
A cutback is a method for automatically reducing the step size when either the solution error is too large
or the solution encounters convergence difficulties during a nonlinear analysis. Should a convergence failure
occur, ANSYS will reduce the time step interval to a fraction of its previous size and automatically continue
the solution from the last successfully converged time step. If the reduced time step again fails to converge,
ANSYS will again reduce the time step size and proceed with the solution. This process continues until
convergence is achieved or the minimum specified time step value is reached.
The CRPLIM command is functionally equivalent to Lab = CRPLIMIT.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Cutback Control
Main Menu>Solution>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Solution>Load Step Opts>Nonlinear>Cutback Control
/CVAL, WN, V1, V2, V3, V4, V5, V6, V7, V8
Specifies nonuniform contour values on stress displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
V1, V2, V3, . . . , V8
Up to 8 contour values may be specified (in ascending order). The 0.0 value (if any) must not be the last
value specified. If no values are specified, all contour specifications are erased and contours are automat-
ically calculated.
Command Default
Nine contour values uniformly spaced between the extreme values.
Notes
This command is similar to the /CONTOUR command. With /CVAL, however, you define the upper level of
each contour band instead of having the contours uniformly distributed over the range. The minimum value
(including a zero value for the first band) for a contour band cannot be specified. If you use both /CONTOUR
and /CVAL, the last command issued takes precedence.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
326
/CVAL
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Contours>Non-uniform Contours
CVAR, IR, IA, IB, ITYPE, DATUM, Name
Computes covariance between two quantities.
POST26: Special Purpose
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previous variable, the previous variable will be overwritten with this result.
IA, IB
Reference numbers of the two variables to be operated on. If only one, leave IB blank.
ITYPE
Defines the type of response PSD to be calculated:
0,1
Displacement (default).
2
Velocity.
3
Acceleration.
DATUM
Defines the reference with respect to which covariance is to be calculated:
1
Absolute value.
2
Relative to base (default).
Name
Thirty-two character name for identifying the variable on listings and displays. Embedded blanks are
compressed upon output.
Notes
This command computes the covariance value for the variables referenced by the reference numbers IA
and IB. If DATUM = 2, the variable referenced by IR will contain the individual modal contributions (i.e.,
the dynamic or relative values). If DATUM = 1, the variable referenced by IR will contain the modal contribu-
tions followed by the contributions of pseudo-static and covariance between dynamic and pseudo-static
responses. File.PSD must be available for the calculations to occur.
Menu Paths
Main Menu>TimeHist Postpro>Calc Covariance
327
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CVAR
/CWD, DIRPATH
Changes the current working directory.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
DIRPATH
The full path name of the new working directory.
Notes
After issuing the /CWD command, all new files opened with no default directory specified (via the FILE,
/COPY, or RESUME commands, for example) default to the new DIRPATH directory.
If issuing the command interactively and the specified directory does not exist, no change of directory occurs
and the command generates an error message. If the command executes via a batch run and the specified
directory does not exist, the batch program terminates with an error.
Menu Paths
Utility Menu>File>Change Directory
/CYCEXPAND, WN, OPTION, Value1, Value2
Graphically expands displacements, stresses and strains of a cyclically symmetric model.
PREP7: Special Purpose
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
WN
The window number to which the expansion applies. Valid values are 1 through 5. The default value is
1. The window number applies only to the AMOUNT argument.
OPTION
One of the following options:
ON
Activates cyclic expansion using the previous settings (if any). If no previous settings exist, this option
activates the default settings.
DEFAULT
Resets cyclic expansion to the default settings.
OFF
Deactivates cyclic expansion. This option is the default.
STATUS
Lists the current cyclic expansion settings.
AMOUNT
The number of repetitions or the total angle.
Value1
NREPEAT
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
328
/CWD
Value2
The number of repetitions. The default is the total number of sectors in 360 degrees.
or
Value1
ANGLE
Value2
The total angle in degrees. The default is 360.
WHAT
A specified portion or subset of the model to expand:
Value1
The component name of the elements to expand. The default is all selected components.
EDGE
Sector edge display key.
1
Suppresses display of edges between sectors even if the cyclic count varies between active
windows.
Caution
Plots with fewer than the maximum number of repetitions may have missing element
faces at the sector boundaries.
0 or OFF
Averages stresses or strains across sector boundaries. This value is the default (although the default
reverts to 1 or ON if the cyclic count varies between active windows).
1 or ON
No averaging of stresses or strains occurs and sector boundaries are shown on the plot.
PHASEANG
The phase angle shift:
Value1
The phase angle shift in degrees. The valid range is 0 through 360. The default is 0. Typically,
this value is the phase angle obtained via the CYCPHASE command.
Command Default
The default /CYCEXPAND command (issuing the command with no arguments) deactivates cyclic expansion
(OPTION = OFF). The default window number (WN) is 1.
Notes
In preprocessing, the /CYCEXPAND command verifies a cyclically symmetric model by graphically expanding
it partially or through the full 360 degrees.
For the postprocessing plot nodal solution (PLNSOL) operation, the command graphically expands displace-
ments, stresses and strains of a cyclically symmetric model partially or though the full 360 degrees by com-
329
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/CYCEXPAND
bining the real (original nodes and elements) and imaginary (duplicate nodes and elements) parts of the
solution.
For the print nodal solution (PRNSOL) operation, the command expands the printed output of displacements
or stresses on a sector-by-sector basis.
Use of the /CYCEXPAND command does not change the database. The command does not modify the
geometry, nodal displacements or element stresses.
The command affects element and result plots only. It has no effect on operations other than plot element
solution (PLESOL), plot nodal solution (PLNSOL) and print nodal solution (PRNSOL). Operations other than
PLESOL, PLNSOL, or PRNSOL work on the unprocessed real and imaginary parts of a cyclic symmetry solution
If you issue a /CYCEXPAND,,OFF command, you cannot then expand the model by simply issuing another
/CYCEXPAND command (for example, to specify an NREPEAT value for the number of repetitions). In such
a case, you must specify /CYCEXPAND,,ON, which activates expansion using the previous settings (if any)
or the default settings.
The command requires PowerGraphics. Turn PowerGraphics on (/GRAPHICS,POWER) if not already active.
(By default, PowerGraphics is active during an interactive session, but not during a batch run.) Any setting
which bypasses PowerGraphics (for example, /PBF) also bypasses cyclic expansion; in such cases, the /CYC-
EXPAND command displays unprocessed real and imaginary results.
The command uses full model graphics (/GRAPHICS,FULL) to compute peak values. Because of this, there
may be slight differences between max/min values obtained with CYCPHASE, and those obtained via
/CYCEXPAND (/GRAPHICS,POWER).
The /CYCEXPAND command does not work with PGR files.
The /CYCEXPAND command is incompatible with the /ESHAPE,1 or /ESHAPE,FAC command.
The /PBC (plot boundary conditions) command is deactivated when cyclic expansion is active (/CYCEX-
PAND,,ON). To view BCs on the basic sector, deactivate cyclic expansion (/CYCEXPAND,,OFF) and issue this
command: /PBC,ALL,,1
In cases involving non-cyclically symmetric loading, all boundary conditions (BCs) will plot when /CYCEXPAND
is active. When examining the plot, however, it is not possible to distinguish between BCs that exist on every
sector and those that exist (via a table referencing SECTOR) only on some sectors or which vary in value.
(Only the basic sector is plotted.)
For magnetic cyclic symmetry analyses, the /CYCEXPAND command produces contour plots but not vector
plots.
To learn more about analyzing a cyclically symmetric structure, see Cyclic Symmetry Analysis in the Advanced
Analysis Techniques Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Cyclic Analysis>Cyc Expansion
Main Menu>Preprocessor>Modeling>Cyclic Sector>Cyc Expansion
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
330
/CYCEXPAND
*CYCLE
Bypasses commands within a do-loop.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Bypasses all commands between this command and the *ENDDO command within a do-loop. The next loop
(if applicable) is initiated. The cycle option may also be conditionally executed [Use the *IF]. The *CYCLE
command must appear on the same file as the *DO command.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
CYCLIC, NSECTOR, ANGLE, KCN, Name, USRCOMP
Specifies a cyclic symmetry analysis.
PREP7: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NSECTOR
The number of sectors in the full 360 degrees, or one of the following options:
STATUS
Indicates the current cyclic status.
OFF
Resets model to normal (non-cyclic) status and removes the duplicate sector if it exists. This option
also deletes automatically detected edge components (generated when USRCOMP = 0).
UNDOUBLE
Removes the duplicate sector if it exists. The duplicate sector is created during the solution (SOLVE)
stage of a modal cyclic symmetry analysis.
Note
The duplicate sector is necessary for displaying cyclic symmetry analysis results during
postprocessing (/POST1).
If you specify a value of STATUS, OFF or UNDOUBLE, the command ignores all remaining arguments.
ANGLE
The sector angle in degrees.
KCN
An arbitrary reference number assigned to the cyclic coordinate system. The default value of 0 specifies
automatic detection.
331
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CYCLIC
Name
The root name of sector low- and high-edge components (line, area, or node components). The default
root name (when USRCOMP = 0) is "CYCLIC". A root name that you specify can contain up to 11 characters.
The naming convention for each low- and high-edge component pair is either of the following:
Name_mxxl, Name_mxxh (potentially matched node patterns)
Name_uxxl, Name_uxxh (potentially unmatched node patterns)
The Name value is the default ("CYCLIC") or specified root name and xx is the component pair ID number
(sequential, starting at 01).
USRCOMP
The number of pairs of user-defined low- and high-edge components on the cyclic sector (if any). The
default value of 0 specifies automatic detection of sector edges; however, the automatic setting is not
valid in all cases. (For more information, see the Notes section below.) If the value is greater than 0, no
verification of user-defined components occurs.
Command Default
The default CYCLIC command (issuing the command with no arguments) detects the number of sectors
(NSECTOR), the sector angle (ANGLE), and the coordinate system (KCN) based upon the existing solid or finite-
element model. The command also detects sector low- and high-edge components in most cases and assigns
the default root name "CYCLIC" to the components.
Notes
You can input your own value for NSECTOR, ANGLE or KCN; if you do so, the command verifies argument
values before executing.
When USRCOMP = 0 (default), the CYCLIC command automatically detects low- and high-edge components
for models comprised of any combination of line, area, or volume elements. If a solid model exists, however,
the command uses only the lines, areas, and/or volumes to determine the low- and high-edge components;
the elements, if any, are ignored.
If you issue a CYCOPT,TOLER command to set a tolerance for edge-component pairing before issuing the
CYCLIC command, the CYCLIC command uses the specified tolerance when performing automatic edge-
component detection.
For 2-D models, autodetection does not consider the CSYS,5 or CSYS,6 coordinate system specification.
Autodetection for 180 degree (two-sector) models is not possible unless a central hole exists.
The CYCLIC command sets values and keys so that, if possible, the area-mesh (AMESH) or volume-mesh
(VMESH) command meshes the sector with matching node and element face patterns on the low and high
edges. (The command has no effect on any other element-creation command.)
Issue the CYCLIC command prior to the meshing command to, if possible, produce a mesh with identical
node and element patterns on the low and high sector edges. Only the AMESH or VMESH commands can
perform automated matching. (Other meshing operation commands such as VSWEEP cannot.) If you employ
a meshing operation other than AMESH or VMESH, you should ensure that node and element face patterns
match, if desired. The CYCLIC command output indicates whether each edge-component pair has or can
produce a matching node pair.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
332
CYCLIC
A cyclic solution (via the SOLVE command) allows dissimilar mesh patterns on the extreme boundaries of
a cyclically symmetric model. The allowance for dissimilar patterns is useful when you have only finite-element
meshes for your model but not the geometry data necessary to remesh it to obtain identical node patterns.
In such cases, it is possible to obtain solution results, although perhaps at the expense of accuracy. A
warning message appears because results may be degraded near the sector edges.
The constraint equations (CEs) that tie together the low and high edges of your model are generated at the
solution stage of the analysis from the low- and high-edge components (and nowhere else). You should
verify that automatically detected components are in the correct locations and that you can account for all
components; to do so, you can list (CMLIST) or plot (CMPLOT) the components.
If you issue the CYCLIC command after meshing and have defined element types with rotational degrees
of freedom (DOFs), ANSYS generates cyclic CEs for rotational DOFs that may not exist on the sector bound-
aries. Issue the CYCOPT,DOF command to prevent unused rotational terms from being generated.
Modal cyclic symmetry analysis is supported by the following eigensolvers:
Block Lanczos (MODOPT, LANB)
PCG Lanczos (MODOPT, LANPCG)
Super Node (MODOPT, SNODE)
To learn more about analyzing a cyclically symmetric structure, see Cyclic Symmetry Analysis in the Advanced
Analysis Techniques Guide.
Distributed ANSYS Restriction This command is only supported for modal analysis (ANTYPE,MODAL) in
Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Modeling>Cyclic Sector>Cyclic Model>Auto Defined
Main Menu>Preprocessor>Modeling>Cyclic Sector>Cyclic Model>Status
Main Menu>Preprocessor>Modeling>Cyclic Sector>Cyclic Model>User Defined
Main Menu>Preprocessor>Modeling>Cyclic Sector>Del Dupl Sector
Main Menu>Preprocessor>Modeling>Cyclic Sector>Reset (OFF)
CYCOPT, OPTION, Value1, Value2, Value3, Value4, Value5, Value6, Value7
Specifies solution options for a cyclic symmetry analysis.
PREP7: Special Purpose
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
OPTION
One of the following options:
BCMULT
Controls whether cyclic sector array parameter names are reused or created new for multiple entities.
Value1
To reuse array parameter names, specify YES, ON, or 1.
333
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CYCOPT
To create new array parameter names (the default behavior), specify NO, OFF, or 0.
COMBINE
For linear static cyclic symmetry analysis with non-cyclically symmetric loading only, expands and
combines all harmonic index solutions and writes them to the results file during the solution phase
of the analysis.
Value1
To enable combining of harmonic index solutions, specify YES, ON, or 1.
To disable combining (the default behavior), specify NO, OFF, or 0.
DEFAULT
Sets the default cyclic solution settings.
DOF
The degrees of freedom to couple from the nodes on the low sector boundary to nodes on the high
boundary:
Value1
The component pair ID number.
Value2, Value3, Value4, . . . , Value7
The constraint-equation/-coupling degree of freedom (DOF) for this pair. Repeat the command
to add other DOFs. The default is constraint-equation/-coupling all applicable DOFs.
FACETOL
Tolerance for inclusion of surface nodes into your basic sector. Autodetect defaults to 15, accom-
modating most sections. Specify a new Value1 only when extreme cut angles or complex model
geometry cause surface nodes to be excluded. See Notes (below) for more information.
HINDEX
The harmonic index solution ranges for modal or buckling cyclic symmetry analyses. The SOLVE
command initiates a cyclic symmetry solution sequence at the harmonic indices specified. (By default,
the SOLVE command solves for all available harmonic indices.) Static and harmonic cyclic symmetry
solutions always use all harmonic indices required for the applied loads.
EVEN / ODD
For low-frequency electromagnetic analysis only, EVEN specifies a symmetric solution and ODD
specifies an antisymmetric solution.
The value you specify is based on the harmonic index: EVEN (default) indicates harmonic index
= 0, and ODD indicates harmonic index = N / 2 (where N is an integer representing the number
of sectors in 360). A value of ODD applies only when N is an even number.
The CYCOPT command with this HINDEX option is cumulative. To remove an option (for example,
EVEN), issue this command: CYCOPT,HINDEX,EVEN,,,-1
Value1, Value2, Value3
Solve harmonic indices in range Value1 through Value2 in steps of Value3. Repeat the
command to add other ranges. The default solves all applicable harmonic indices.
Value4
The only valid value is -1. If specified, it removes Value1 through Value2 in steps of Value3
from the set to solve.
LDSECT
Restricts subsequently defined DOF constraint values, force loads, and surface loads to a specified
sector. The restriction remains in effect until you change or reset it.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
334
CYCOPT
Value1
The sector number. A value other than 0 (default) is valid for a cyclic symmetry analysis with
non-cyclically symmetric loading only. A value of 0 (or ALL) resets the default behavior for cyclic
loading (where the loads are identical on all sectors).
MOVE
Flag specifying whether ANSYS should move high- or low-edge component nodes paired within the
specified tolerance (TOLER) to create precisely matching pairs.
Value1
The flag value, as follows:
0 -- Do not move edge component nodes. This value is the default.
1 or HIGH -- Move the high-edge component nodes to precisely match the low-edge component
nodes.
-1 or LOW -- Move the low-edge component nodes to precisely match the high-edge component
nodes.
STATUS
Lists the solution option settings active for the cyclic model.
TOLER
The tolerance used to determine whether a node on the low edge is paired with a node on the high
edge.
Value1
The tolerance value, as follows:
> 0 -- The absolute distance tolerance for automatic sector-boundary detection and low-/high-
edge component node pairing.
< 0 -- The relative tolerance for automatic sector-boundary detection and low-/high-edge com-
ponent node pairing. In this case, the tolerance is Value1 * Length, where Length is the
length of the diagonal of an imaginary box enclosing the model.
The default tolerance is -1.0E-4 (also represented by Value1 = 0).
USRROT
Flag specifying whether ANSYS should override automatic nodal rotations to edge components and
allow you to apply nodal rotations manually.
Value1
The flag value, as follows:
0 (OFF or NO) -- Allow automatic node rotation. This behavior is the default.
1 (ON or YES) -- Suppress automatic node rotation. If you select this option, you must apply ap-
propriate nodal rotations to all edge component nodes; otherwise, your analysis will yield incorrect
solution results.
LOW -- Suppresses automatic rotation of low-edge component nodes only, allowing you to apply
them manually. (Automatic rotation of high-edge component nodes occurs to produce the
matching edge nodes required for a valid cyclic solution.)
335
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CYCOPT
HIGH -- Suppresses automatic rotation of high-edge component nodes only, allowing you to
apply them manually. (Automatic rotation of low-edge component nodes occurs to produce the
matching edge nodes required for a valid cyclic solution.)
VTSOL
Employ Variational Technology to accelerate your cyclic symmetry analysis.
Value1
To employ Variational Technology (VT), specify YES, ON, or 1 (requires a High Performance
Computing License).
To process without VT (the default behavior), specify NO, OFF, or 0.
Command Default
No defaults are available for the CYCOPT command. You must specify an argument (OPTION) when issuing
the command. Other values which may be necessary depend upon which argument you specify.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Notes
ANSYS solves a cyclically symmetric model (set up via the CYCLIC command during preprocessing) at the
harmonic indices specified via the CYCOPT command.
The CYCOPT,COMBINE option is an alternative to the /CYCEXPAND command and is especially useful for
testing purposes. However, ANSYS recommends specifying COMBINE only when the number of sectors is
relatively small. (The option expands nodes and elements into the full 360 and can slow postprocessing
significantly.
If you issue a CYCOPT,TOLER command to set a tolerance for edge-component pairing before issuing the
CYCLIC command, the CYCLIC command uses the specified tolerance when performing automatic edge-
component detection.
Extreme cut angles on smaller sections can result in surface nodes being excluded from the section. The
default FACETOL value of 15 can be decreased to 10 when you find that your valid surface nodes are being
excluded. Change this value only when the auto-detect function fails to include your surface nodes.
In cases involving non-cyclically symmetric loading (that is, when LDSECT > 0), the underlying command
operations create or modify the required SECTOR tabular boundary condition (BC) data to apply on the ap-
propriate sector. Therefore, it is not necessary to manipulate tables for situations where the applied BC is
not a function of other tabular BC variables such as TIME, X, Y, Z, and so on.
To delete a previously applied load on a specified sector, issue an FDELE command.
Because edge nodes are rotated into the cyclic coordinate system during solution, any applied displacements
or forces on sector edges will be in the cyclic coordinate system.
The CYCOPT command is valid in the preprocessing and solution stages of an analysis.
To learn more about analyzing a cyclically symmetric structure, see Cyclic Symmetry Analysis in the Advanced
Analysis Techniques Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
336
CYCOPT
Menu Paths
Main Menu>Preprocessor>Modeling>Cyclic Sector>Cyclic Options
Main Menu>Solution>Solve>Cyclic Options
CYCPHASE, TYPE, OPTION
Provides tools for determining minimum and maximum possible result values from frequency couplets
produced in a modal cyclic symmetry analysis.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
TYPE
The type of operation requested:
DISP
Calculate the maximum and minimum possible displacement at each node in the original sector
model. Store the values and the phase angle at which they occurred.
STRESS
Calculate the maximum and minimum possible stresses at each node in the original sector model.
Store the values and the phase angle at which they occurred.
STRAIN
Calculate the maximum and minimum possible strains at each node in the original sector model.
Store the values and the phase angle at which they occurred.
ALL
Calculate the maximum and minimum possible displacement, stress and strain at each node in the
original sector model. Store the values and the phase angle at which they occurred.
GET
Places the value of a MAX or MIN item into the _CYCVALUE parameter, the node for that value in
the _CYCNODE parameter, and the phase angle for the value in the _CYCPHASE parameter.
PUT
Put resulting sweep values for printing (via the PRNSOL command ) or plotting (via the PLNSOL
command).
LIST
List the current minimum/maximum displacement, stress and strain nodal values.
STAT
Summarize the results of the last phase sweep.
CLEAR
Clear phase-sweep information from the database.
OPTION
If TYPE = DISP, STRAIN, STRESS or ALL, controls the sweep angle increment to use in the search:
Angle
The sweep angle increment in degrees, greater than 0.1 and less than 10. The default is 1.
If TYPE = PUT, controls which values are placed onto the model:
337
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CYCPHASE
MAX
Put all existing nodal maximum values onto the model. This option is the default.
MIN
Put all existing nodal minimum values onto the model.
If TYPE = GET, controls the values placed into cyclic parameters:
Item
Specifies the type of values on which to operate:
U -- Displacement
S -- Stress
EPEL -- Strain
Comp
Specifies the specific component of displacement, stress or strain for which to get information:
X,Y,Z -- Basic components
XY,YZ,XZ -- Shear components
1,2,3 -- Principal values
EQV -- Equivalent value
SUM -- USUM
MxMn
Specifies whether the requested value information is for the maximum or minimum value:
MAX -- Maximum value.
MIN -- Minimum value.
Command Default
No defaults are available for the CYCPHASE command. You must specify an argument (TYPE) when issuing
the command. Other values which may be necessary (OPTION) depend upon which TYPE argument you
specify.
Notes
When you expand the results of a modal cyclic symmetry analysis (via the /CYCEXPAND or EXPAND com-
mand), ANSYS combines the real and imaginary results for a given nodal diameter, assuming no phase shift
between them; however, the modal response can occur at any phase shift.
CYCPHASE response results are valid only for the first cyclic sector. To obtain the response at any part of
the expanded model, ANSYS recommends using cyclic symmetry results expansion at the phase angle obtained
via CYCPHASE.
The phase angles returned by CYCPHASE contain the minimum and maximum values for USUM, SEQV and
other scalar principal stress and strain quantities; however, they do not always return the true minimum and
maximum values for directional quantities like UX or SX unless the values fall in the first sector.
CYCPHASE does not consider midside node values when evaluating maximum and minimum values, which
may affect DISPLAY quantities but no others. (Typically, ANSYS ignores midside node stresses and strains
during postprocessing.)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
338
CYCPHASE
Issuing CYCPHASE,PUT clears the result values for midside nodes on high order elements; therefore, this
option sets element faceting (/EFACET) to 1. The command reports that midside nodal values are set to
zero and indicates that element faceting is set to 1.
If the sweep values are available after issuing a CYCPHASE,PUT command, the PRNSOL or PLNSOL command
will print or plot (respectively) the sweep values of structure displacement Ux, Uy, Uz, component stress/strain
X, Y, Z, XY, YZ, ZX, principal stress/strain 1, 2, 3 and equivalent stress/strain EQV. The vector sum of displace-
ment (USUM) and stress/strain intensity (SINT) are not valid phase-sweep results.
You can specify any coordinate system via the RSYS command for displaying or printing CYCPHASE results.
However, after CYCPHASE results have been extracted, you cannot then transform them via the RSYS
command. If you try to do so, ANSYS issues a warning message.
The CYCPHASE command is valid in /POST1 and for cyclically symmetric models only.
To learn more about analyzing a cyclically symmetric structure, see Cyclic Symmetry Analysis in the Advanced
Analysis Techniques Guide.
Menu Paths
Main Menu>General Postproc>Cyclic Analysis>Cyclic Phase>Clear
Main Menu>General Postproc>Cyclic Analysis>Cyclic Phase>Get
Main Menu>General Postproc>Cyclic Analysis>Cyclic Phase>List
Main Menu>General Postproc>Cyclic Analysis>Cyclic Phase>Phase Sweep
Main Menu>General Postproc>Cyclic Analysis>Cyclic Phase>Put Phase Results
Main Menu>General Postproc>Cyclic Analysis>Cyclic Phase>Status
CYL4, XCENTER, YCENTER, RAD1, THETA1, RAD2, THETA2, DEPTH
Creates a circular area or cylindrical volume anywhere on the working plane.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XCENTER, YCENTER
Working plane X and Y coordinates of the center of the circle or cylinder.
RAD1, RAD2
Inner and outer radii (either order) of the circle or cylinder. A value of zero or blank for either RAD1 or
RAD2, or the same value for both RAD1 and RAD2, defines a solid circle or cylinder.
THETA1, THETA2
Starting and ending angles (either order) of the circle or faces of the cylinder. Used for creating a partial
annulus or partial cylinder. The sector begins at the algebraically smaller angle, extends in a positive
angular direction, and ends at the larger angle. The starting angle defaults to 0 and the ending angle
defaults to 360. See the Modeling and Meshing Guide for an illustration.
DEPTH
The perpendicular distance (either positive or negative based on the working plane Z direction) from
the working plane representing the depth of the cylinder. If DEPTH = 0 (default), a circular area is created
on the working plane.
339
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CYL4
Notes
Defines a circular area anywhere on the working plane or a cylindrical volume with one face anywhere on
the working plane. For a solid cylinder of 360, the top and bottom faces will be circular (each area defined
with four lines) and they will be connected with two surface areas (each spanning 180). See the CYL5,
PCIRC, and CYLIND commands for alternate ways to create circles and cylinders.
When working with a model imported from an IGES file (DEFAULT import option), you must provide a value
for DEPTH or the command will be ignored.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Areas>Circle>Annulus
Main Menu>Preprocessor>Modeling>Create>Areas>Circle>Partial Annulus
Main Menu>Preprocessor>Modeling>Create>Areas>Circle>Solid Circle
Main Menu>Preprocessor>Modeling>Create>Primitives>Solid Cylindr
Main Menu>Preprocessor>Modeling>Create>Volumes>Cylinder>Hollow Cylinder
Main Menu>Preprocessor>Modeling>Create>Volumes>Cylinder>Partial Cylinder
Main Menu>Preprocessor>Modeling>Create>Volumes>Cylinder>Solid Cylinder
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Create>Volume>Cylinder>Solid Cylinder
CYL5, XEDGE1, YEDGE1, XEDGE2, YEDGE2, DEPTH
Creates a circular area or cylindrical volume by end points.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XEDGE1, YEDGE1
Working plane X and Y coordinates of one end of the circle or cylinder face.
XEDGE2, YEDGE2
Working plane X and Y coordinates of the other end of the circle or cylinder face.
DEPTH
The perpendicular distance (either positive or negative based on the working plane Z direction) from
the working plane representing the depth of the cylinder. If DEPTH = 0 (default), a circular area is created
on the working plane.
Notes
Defines a circular area anywhere on the working plane or a cylindrical volume with one face anywhere on
the working plane by specifying diameter end points. For a solid cylinder of 360, the top and bottom faces
will be circular (each area defined with four lines) and they will be connected with two surface areas (each
spanning 180). See the CYL4, PCIRC, and CYLIND commands for alternate ways to create circles and cylinders.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Areas>Circle>By End Points
Main Menu>Preprocessor>Modeling>Create>Volumes>Cylinder>By End Pts & Z
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Create>Volume>Cylinder>By End Pts &
Z
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
340
CYL5
CYLIND, RAD1, RAD2, Z1, Z2, THETA1, THETA2
Creates a cylindrical volume centered about the working plane origin.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RAD1, RAD2
Inner and outer radii (either order) of the cylinder. A value of zero or blank for either RAD1 or RAD2, or
the same value for both RAD1 and RAD2, defines a solid cylinder.
Z1, Z2
Working plane Z coordinates of the cylinder. If either Z1 or Z2 is zero, one of the faces of the cylinder
will be coplanar with the working plane.
THETA1, THETA2
Starting and ending angles (either order) of the cylinder. Used for creating a cylindrical sector. The sector
begins at the algebraically smaller angle, extends in a positive angular direction, and ends at the larger
angle. The starting angle defaults to 0.0 and the ending angle defaults to 360.0. See the Modeling and
Meshing Guide for an illustration.
Notes
Defines a cylindrical volume centered about the working plane origin. The top and bottom faces are parallel
to the working plane but neither face need be coplanar with (i.e., "on") the working plane. The cylinder must
have a spatial volume greater than zero. (i.e., this volume primitive command cannot be used to create a
degenerate volume as a means of creating an area.) For a solid cylinder of 360, the top and bottom faces
will be circular (each area defined with four lines), and they will be connected with two areas (each spanning
180.) See the CYL4 and CYL5 commands for alternate ways to create cylinders.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Cylinder>By Dimensions
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Create>Volume>Cylinder>By Dimensions
CZDEL, grp1, grp2, grp3
Edits or clears cohesive zone sections.
PREP7: Meshing
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
grp1
Initial group of cohesive zone elements to be deleted.
grp2
Final group of cohesive zone elements to be deleted.
grp3
Increment for selected groups.
341
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CZDEL
Notes
The CZDEL command edits or deletes the interface elements and nodes, along with the associated changes
made to the underlying plane or solid elements created during a previous CZMESH operation.
Each CZMESH operation will create groups of elements and nodes with component names in the format
CZME_EL01 (elements) and CZME_ND01 (nodes). The final number of this format will be the number used
for grp1 and grp2. If grp1 = ALL, all nodes and elements created by the CZMESH command will be deleted.
After using CZDEL, all the user-defined components will be unselected.
The CZDEL command is valid for structural analyses only.
Menu Paths
This command cannot be accessed from a menu.
CZMESH, ecomps1, ecomps2, KCN, KDIR, VALUE, CZTOL
Create and mesh an interface area composed of cohesive zone elements.
PREP7: Meshing
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
ecomps1
Component name or number for the group of plane or solid structural elements adjacent to the interface
being meshed.
ecomps2
Component name or number for the opposing (from ecomps1) group of plane or solid structural elements
adjacent to the interface being meshed.
KCN
Coordinate system number for the separation surface and normal direction. (if ecomps1 and ecomps2
not specified)
KDIR
Direction (x, y, or z) normal to separation surface in the KCN coordinate system (if ecomps1 and ecomps2
not specified).
VALUE
Coordinate value along the KDIR axis at which to locate the interface (if ecomps1 and ecomps2 not
specified).
CZTOL
Optional absolute tolerance about VALUE (if ecomps1 and ecomps2 not specified). Allows nodes oc-
curring slightly above or below the separation to be grouped properly. The following expression represents
the default value:
X Y Z
2 2 2
1000
+ +
Where X, Y, and Z are the dimensions of the model based on nodal locations (that is, X = Xmax
- Xmin).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
342
CZMESH
Notes
CZMESH is used on a mesh with shared nodes at the interface.
If ecomps1 and ecomps2 are specified, the CZMESH command creates/meshes interface elements (INTER202,
INTER203, INTER204, INTER205) along the boundary between the two components or groups of elements.
The elements in each of the components or groups of elements will share nodes with each other and also
with the interface elements. This one-element thick boundary of interface elements will split the body
between the two components or groups of elements.
Subsequent separation (delamination and failure) of the interface zone will result in an increasing displacement
between the nodes (within the interface element) along the cohesive zone elements. Unless otherwise
specified, the CZMESH command analyzes the configuration and geometry of the adjacent structural elements
and provides the appropriate interface element.
The CZMESH operation copies any nodal temperatures you have defined on the split surface of the original
mesh from the original nodes to the newly created coincident duplicate nodes. However, displacements,
forces, and other boundary conditions are not copied.
The CZMESH command is valid for structural analyses only.
Menu Paths
This command cannot be accessed from a menu.
343
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
CZMESH
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
344
D Commands
D, NODE, Lab, VALUE, VALUE2, NEND, NINC, Lab2, Lab3, Lab4, Lab5, Lab6
Defines degree-of-freedom constraints at nodes.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Node at which constraint is to be specified. If ALL, NEND and NINC are ignored and constraints are applied
to all selected nodes (NSEL). If NODE = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI). A component name may also be substituted for NODE.
Lab
Valid degree-of-freedom label. If ALL, use all appropriate labels.
Structural labels: UX, UY, or UZ (displacements); ROTX, ROTY, or ROTZ (rotations); WARP (warping).
Thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP (temperature).
FLOTRAN fluid labels: PRES (pressure); VX, VY, or VZ (velocities); ENKE or ENDS (turbulent kinetic
energy or turbulent kinetic energy dissipation rate); SP01 through SP06 (multiple species mass fractions)
or their user-defined names (MSSPEC); UX, UY, or UZ (Arbitrary Lagrangian-Eulerian formulation mesh
displacements).
Electric labels: VOLT (voltage).
Magnetic labels: MAG (scalar magnetic potential); AX, AY, or AZ (vector magnetic potentials).
For structural static and transient analyses, translational and rotational velocities are also valid loads.
Use these labels: VELX, VELY, VELZ (translational velocities); OMGX, OMGY, OMGZ (rotational velocities).
For structural analyses, HDSP (hydrostatic pressure) is also valid. However, HDSP is not included when
Lab = ALL.
For structural transient analyses, the following acceleration loads are also valid: ACCX, ACCY, ACCZ
(translational accelerations); DMGX, DMGY, DMGZ (rotational accelerations). The velocity and acceleration
loads are not included when Lab = ALL.
If the node is connected to an ELBOW290 element, the following pipe cross-section degree-of-freedom
labels are also valid: SE, SO, SW, SRA, and SRT. (For details, see the ELBOW290 documentation.) The degrees
of freedom are not included when Lab = ALL. To constrain all cross-section degrees of freedom, specify
Lab = SECT.
Value
Degree-of-freedom value or table name reference for tabular boundary conditions. To specify a table,
enclose the table name in percent (%) signs (for example, D,NODE,TEMP,%tabname%). Use the *DIM
command to define a table.
If Value = SUPPORT, you can specify pseudo-constraints when using residual vectors in a modal analysis
(RESVEC,ON) or CMS analysis (CMSOPT,RFFB).
345
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VALUE2
Second degree-of-freedom value (if any). If the analysis type and the degree of freedom allow a complex
input, Value (above) is the real component and VALUE2 is the imaginary component.
NEND, NINC
Specifies the same values of constraint at the range of nodes from NODE to NEND (defaults to NODE),
in steps of NINC (defaults to 1).
Lab2, Lab3, Lab4, Lab5, Lab6
Additional degree-of-freedom labels. The same values are applied to the nodes for these labels.
Notes
The available degrees of freedom per node are listed under "Degrees of Freedom" in the input table for
each element type in the Element Reference. Degrees of freedom are defined in the nodal coordinate system.
The positive directions of structural translations and rotations are along and about the positive nodal axes
directions. Structural rotations should be input in radians. The node and the degree-of-freedom label must
be selected (NSEL, DOFSEL).
If Lab = ENKE and Value = -1, a FLOTRAN flag is set to indicate a moving wall.
If Lab = ENDS and Value = -1, FLOTRAN generalized symmetry conditions are applied. Velocity components
are set tangential to the symmetry surface if the ALE formulation is not activated. They are set equal to the
mesh velocity if the ALE formulation is activated.
In a structural analysis, you can apply only one displacement, velocity, or acceleration load at any degree
of freedom. If multiple loads are specified, the last applied load overrides the previous ones. For example,
the following commands apply loads to node 100:
D,100,UX,Value
D,100,VELX,Value
In this case, the velocity load (VELX) applied in the last command will override the displacement load (UX).
For elements HF119 and HF120, used in high-frequency electromagnetic analysis, the AX degree of freedom
is not an x-component of a vector potential, but rather a tangential component of E (the electric field) on
the element edges and faces. To specify an Electric Wall condition, set AX to zero. For more information,
see the High-Frequency Electromagnetic Analysis Guide.
For elements used in static and low frequency electromagnetic analysis (SOLID117, SOLID236 and SOLID237),
the AZ degree of freedom is not a z-component of a vector potential, but rather the flux contribution on
the element edge. To specify a flux-parallel condition, set AZ = 0. For more information, see 3-D Magneto-
statics and Fundamentals of Edge-based Analysis in the Low-Frequency Electromagnetic Analysis Guide.
In an explicit dynamic analysis, the D command can only be used to fix nodes in the model. The degree-of-
freedom value must be zero; no other values are valid. Use the EDLOAD command to apply a nonzero dis-
placement in an explicit dynamic analysis.
For ELBOW290 cross-section degrees of freedom (Lab = SE, SO, SW, SRA, SRT, or SECT), the D command can
only specify fixed constraints. The degree-of-freedom value must be zero; no other values are valid.
Tabular boundary conditions (VALUE = %tabname%) are available only for the following degree-of-freedom
labels: Electric (VOLT), FLOTRAN (UX, UY, UZ, PRES, VX, VY, VZ, ENKE, ENDS, TEMP, SP01, SP02, SP03, SP04,
SP05, and SP06); structural (UX, UY, UZ, ROTX, ROTY, ROTZ, and velocity and acceleration loads VELX, VELY,
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
346
D
VELZ, OMGX, OMGY, OMGZ, ACCX, ACCY, ACCZ, DMGX, DMGY, DMGZ), and temperature (TEMP, TBOT, TE2,
TE3, . . ., TTOP). All labels are valid only in static (ANTYPE,STATIC) and full transient (ANTYPE,TRANS) analyses.
%_FIX% is an ANSYS reserved table name. When VALUE is set to %_FIX%, ANSYS will prescribe the degree
of freedom to the current displacement value. Alternatively, functions UX(), UY(), etc. may be used (see
*GET for a complete list of available functions). However, note that these functions are not available when
multiframe restart features are invoked. In most cases, %_FIX% usage is efficient and recommended for all
structural degrees of freedom.
When Value = SUPPORT, specify only the minimum number of displacement constraints necessary to prevent
rigid body motion: three constraints (or fewer, depending on the element type) for 2-D models and six (or
fewer) for 3-D models.
If constraints and initial conditions (IC) are applied at the same node, the constraint specification overrides.
This combination is useful when a constraint degree-of-freedom value needs to start with a nonzero value
at time = 0.0. For example, if the constraint degree-of-freedom value is prescribed to be a cosine function,
then specifying an initial condition for the same node and degree of freedom ensures that the initial value
for the constraint degree of freedom at time = 0.0 is same as the cosine function evaluated at time = 0.0. If
initial conditions are not specified, the constraint degree-of-freedom value ramps from zero in the first
substep of the first loadstep.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>PerfEC>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>TimeInt>J-Normal>On
Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>TimeInt>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>J-Normal>On
Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>EdgeMVP>Flux Par'l>On
Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>ScalarPot>Flux Nor-
mal>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>ScalarPot>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux
Par'l>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Node Com-
ponents
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>BasePSD>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>MultiPtBas>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>BasePSD>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>MultiPtBas>On Nodes
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>On Nodes
Main Menu>Solution>Constraints>Apply>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>PerfEC>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>TimeInt>J-Normal>On Nodes
347
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
D
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>TimeInt>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>J-Normal>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>EdgeMVP>Flux Par'l>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>ScalarPot>Flux Normal>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>ScalarPot>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Par'l>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Node Components
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>BasePSD>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>MultiPtBas>On Nodes
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>BasePSD>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>MultiPtBas>On Nodes
DA, AREA, Lab, Value1, Value2
Defines DOF constraints on areas.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
AREA
Area on which constraints are to be specified. If ALL, apply to all selected areas [ASEL]. If AREA = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for AREA.
Lab
Symmetry label (see 2 (p. 350) below):
SYMM
Generate symmetry constraints for non-FLOTRAN models. Requires no Value1 or Value2.
ASYM
Generate antisymmetry constraints for non-FLOTRAN models. Requires no Value1 or Value2.
ANSYS DOF labels (see 1 (p. 349), 2 (p. 350), and 3 (p. 350) below):
UX
Displacement in X direction.
UY
Displacement in Y direction.
UZ
Displacement in Z direction.
ROTX
Rotation about X axis.
ROTY
Rotation about Y axis.
ROTZ
Rotation about Z axis.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
348
DA
HDSP
Hydrostatic pressure.
TEMP, TBOT, TE2, TE3, . . ., TTOP
Temperature.
MAG
Magnetic scalar potential.
VOLT
Electric scalar potential.
AX
Magnetic vector potential in X direction (see 4 (p. 350)).
AY
Magnetic vector potential in Y direction.
AZ
Magnetic vector potential in Z direction (see 1 (p. 349)).
ALL
Applies all appropriate DOF labels except HDSP.
FLOTRAN Standard DOF Labels (see 7 (p. 350)): VX, VY, VZ, PRES, TEMP, ENKE, ENDS
FLOTRAN Species Labels (See 8 (p. 350)): SP01, SP02, SP03, SP04, SP05, SP06
FLOTRAN Arbitrary Lagrangian-Eulerian formulation Mesh Displacement Labels (See 9 (p. 350)): UX, UY,
UZ
Value1
Value of DOF or table name reference on the area. Valid for all DOF labels. To specify a table, enclose
the table name in % signs (e.g., DA,AREA,TEMP,%tabname%). Use the *DIM command to define a table.
If Lab = ENKE and Value1 = -1, a FLOTRAN flag is set to indicate a moving wall.
If Lab = ENDS and Value1 = -1, FLOTRAN generalized symmetry conditions are applied. Velocity com-
ponents are set tangential to the symmetry surface if the ALE formulation is not activated. They are set
equal to the mesh velocity if the ALE formulation is activated.
Value2
For FLOTRAN DOFs:
0
Values are applied only to nodes within the area.
1
Values are applied to the edges of the area as well as to the internal nodes. (See 7 (p. 350).)
For MAG and VOLT DOFs:
Value of the imaginary component of the degree of freedom.
Notes
1. For elements SOLID117, SOLID236, and SOLID237, if Lab = AZ and Value1 = 0, this sets the flux-
parallel condition for the edge formulation. (A flux-normal condition is the natural boundary condition.)
Do not use the DA command to set the edge-flux DOF, AZ to a nonzero value.
349
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DA
2. If Lab = MAG and Value1 = 0, this sets the flux-normal condition for the magnetic scalar potential
formulations (MSP) (A flux-parallel condition is the natural boundary condition for MSP.)
3. If Lab = VOLT and Value1 = 0, the J-normal condition is set (current density (J) flow normal to the
area). (A J-parallel condition is the natural boundary condition.)
4. For elements HF119 and HF120, used in high-frequency electromagnetic analysis, the AX DOF is not
an x-component of a vector potential, but rather a tangential component of E (the electric field) on
the element edges and faces. To specify an Electric Wall condition, set AX to zero. For more information,
see the High-Frequency Electromagnetic Analysis Guide.
5. You can transfer constraints from areas to nodes with the DTRAN or SBCTRAN commands. See the
DK command for information about generating other constraints on areas for non-FLOTRAN models.
6. Symmetry and antisymmetry constraints are generated as described for the DSYM command.
7. For the velocity DOF (VX, VY, VZ), a zero value will override a nonzero value at the intersection of two
areas.
8. You can use the MSSPEC command to change FLOTRAN species labels to user-defined labels. You
must define these labels with the MSSPEC command before using them on the DA command.
9. Tabular boundary conditions (VALUE = %tabname%) are available only for the following degree of
freedom labels: Electric (VOLT), FLOTRAN (UX, UY, UZ, PRES, VX, VY, VZ, ENKE, ENDS, TEMP, SP01, SP02,
SP03, SP04, SP05, and SP06); Structural (UX, UY, UZ, ROTX, ROTY, ROTZ), and temperature (TEMP, TBOT,
TE2, TE3, . . ., TTOP).
10. Constraints specified by the DA command can conflict with other specified constraints. See Resolution
of Conflicting Constraint Specifications\ in the Basic Analysis Guide for details.
11. The DA command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>PerfEC>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>TimeInt>J-Normal>On
Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>TimeInt>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>J-Normal>On
Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>EdgeMVP>Flux Par'l>On
Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>ScalarPot>Flux Nor-
mal>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>ScalarPot>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Nor-
mal>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux
Par'l>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>On
Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>On
Areas
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
350
DA
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>On Areas
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>On Areas
Main Menu>Solution>Constraints>Apply>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>PerfEC>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>TimeInt>J-Normal>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>TimeInt>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>J-Normal>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>EdgeMVP>Flux Par'l>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>ScalarPot>Flux Normal>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>ScalarPot>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Normal>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Par'l>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Areas
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>On Areas
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Areas
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>On Areas
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>On Areas
DADELE, AREA, Lab
Deletes DOF constraints on an area.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
AREA
Area for which constraints are to be deleted. If ALL, delete for all selected areas [ASEL]. If AREA = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). You
can substitute a component name for AREA.
Lab
Valid constraint labels are:
ALL
All constraints.
SYMM
Symmetry constraints.
ASYM
Antisymmetry constraints.
UX
Displacement in X direction.
UY
Displacement in Y direction.
UZ
Displacement in Z direction.
ROTX
Rotation about X axis.
351
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DADELE
ROTY
Rotation about Y axis.
ROTZ
Rotation about Z axis.
VX
Velocity component in X direction.
VY
Velocity component in Y direction.
VZ
Velocity component in Z direction.
PRES
Pressure.
TEMP, TBOT, TE2, TE3, . . ., TTOP
Temperature.
ENKE
Turbulent Kinetic Energy.
ENDS
Energy Dissipation Rate.
MAG
Magnetic scalar potential.
VOLT
Electric scalar potential.
SP01. . . SP06
Multiple Species Mass Fraction.
AX
Magnetic vector potential in X direction (see notes).
AY
Magnetic vector potential in Y direction.
AZ
Magnetic vector potential in Z direction (see notes).
Notes
Deletes the degree of freedom constraints at an area (and all corresponding finite element constraints)
previously specified with the DA command. See the DDELE command for delete details.
If the multiple species labels have been changed to user-defined labels via the MSSPEC command, use the
user-defined labels.
See the DA or the DA commands for details on element applicability.
Warning
On previously meshed areas, all constraints on affected nodes will be deleted, whether or not
they were specified by the DA command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
352
DADELE
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Constraint>On All Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>PerfEC>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>TimeInt>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>Voltage>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Displacement>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Species>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Turbulence>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Velocity>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>EdgeMVP>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>ScalarPot>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Temperature>On Areas
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Delete>On Areas
Main Menu>Solution>Constraints>Delete>On Areas
Main Menu>Solution>Define Loads>Delete>All Load Data>All Constraint>On All Areas
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>PerfEC>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>TimeInt>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>Voltage>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Displacement>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Species>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Turbulence>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Velocity>On Areas
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>EdgeMVP>On Areas
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>ScalarPot>On Areas
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Areas
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Areas
Main Menu>Solution>Define Loads>Delete>Thermal>Temperature>On Areas
DALIST, AREA
Lists the DOF constraints on an area.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
AREA
List constraints for this area. If ALL (default), list for all selected areas [ASEL]. If P1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for AREA.
353
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DALIST
Notes
Lists the degree of freedom constraints on an area previously specified with the DA command.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>DOF Constraints>On All Areas
Utility Menu>List>Loads>DOF Constraints>On Picked Areas
DAMORPH, AREA, XLINE, RMSHKY
Move nodes in selected areas to conform to structural displacements.
PREP7: Morphing
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
AREA
Non-structural area to which mesh movement (morph) applies. If ALL, apply morphing to all selected
areas [ASEL]. If AREA = P, graphical picking is enabled. A component may be substituted for AREA.
XLINE
Lines to be excluded from morphing. If ALL, exclude all selected lines [LSEL] from morphing. If XLINE
= P, graphical picking is enabled. A component may be substituted for XLINE. If XLINE is blank (default),
allow morphing of nodes attached to lines of the selected areas (AREA) which are not shared by unse-
lected areas. See Notes for clarification.
RMSHKY
Remesh flag option:
0
Remesh the selected non-structural areas only if mesh morphing fails.
1
Remesh the selected non-structural areas and bypass mesh morphing.
2
Perform mesh morphing only and do not remesh.
Notes
The selected areas should include only non-structural regions adjacent to structural regions. DAMORPH will
morph the non-structural areas to coincide with the deflections of the structural regions.
Nodes in the structural regions move in accordance with computed displacements. Displacements from a
structural analysis must be in the database prior to issuing DAMORPH.
By default, nodes attached to lines can move along the lines, or off the lines (if a line is interior to the selected
areas). You can use XLINE to restrain nodes on certain lines.
By default (RMSHKEY = 0), DAMORPH will remesh the selected non-structural areas entirely if a satisfactory
morphed mesh cannot be provided.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
354
DAMORPH
If boundary conditions and loads are applied directly to nodes and elements, the DAMORPH command re-
quires that these be removed before remeshing can take place.
Exercise care with initial conditions defined by the IC command. Before a structural analysis is performed
for a sequentially coupled analysis, the DAMORPH command requires that initial conditions be removed
from all null element type nodes in the non-structural regions. Use ICDELE to delete the initial conditions.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Phys Morphing>Areas
DATA, IR, LSTRT, LSTOP, LINC, Name, KCPLX
Reads data records from a file into a variable.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
LSTRT
Start at location LSTRT (defaults to 1).
LSTOP
Stop at location LSTOP (defaults to LSTRT). Maximum location available is determined from data previ-
ously stored.
LINC
Fill every LINC location between LSTRT and LSTOP (defaults to 1).
Name
Eight character name for identifying the variable on the printout and displays. Embedded blanks are
compressed upon output.
KCPLX
Complex number key:
0
Data stored as the real part of the complex number.
1
Data stored as the imaginary part of the complex number.
Notes
This command must be followed by a format statement (on the next line) and the subsequent data records,
and all must be on the same file (that may then be read with the /INPUT command). The format specifies
the number of fields to be read per record, the field width, and the placement of the decimal point (if one
is not included in the data value). The read operation follows the available FORTRAN FORMAT conventions
of the system. See the system FORTRAN manual for details. Any standard FORTRAN real format (such as
355
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DATA
(4F6.0), (F2.0,2X,F12.0), etc.) may be used. Integer (I), character (A), and list-directed (*) descriptors may not
be used. The parentheses must be included in the format. Up to 80 columns per record may be read. Locations
may be filled within a range. Previous data in the range will be overwritten.
Menu Paths
This command cannot be accessed from a menu.
DATADEF
Specifies "Directly defined data status" as the subsequent status topic.
POST1: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>General Postproc>Modify Results
DCGOMG, DCGOX, DCGOY, DCGOZ
Specifies the rotational acceleration of the global origin.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS
DCGOX, DCGOY, DCGOZ
Rotational acceleration of the global origin about the acceleration system X, Y, and Z axes.
Notes
Specifies the rotational acceleration of the global origin about each of the acceleration coordinate system
axes [CGLOC]. Rotational accelerations may be defined in analysis types ANTYPE,STATIC, HARMIC (full or
mode superposition), TRANS (full or mode superposition), and SUBSTR. See Acceleration Effect in the Theory
Reference for the Mechanical APDL and Mechanical Applications for details. Units are radians/time
2
.
The DCGOMG command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for DCGOMG_X, DCGOMG_Y, and DCGOMG_Z input values (*DIM) for full transient and
harmonic analyses.
Related commands are ACEL, CGLOC, CGOMGA, DOMEGA, and OMEGA.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
356
DATADEF
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Coriolis Effects
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Coriolis Effects
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Coriolis Effects
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Coriolis Effects
DCUM, Oper, RFACT, IFACT, TBASE
Specifies that DOF constraint values are to be accumulated.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Oper
Accumulation key:
REPL
Subsequent values replace the previous values (default).
ADD
Subsequent values are added to the previous values.
IGNO
Subsequent values are ignored.
RFACT
Scale factor for the real component. Zero (or blank) defaults to 1.0. Use a small number for a zero scale
factor.
IFACT
Scale factor for the imaginary component. Zero (or blank) defaults to 1.0. Use a small number for a zero
scale factor.
TBASE
Base temperature for temperature difference. Used only with temperature degree of freedom. Scale
factor is applied to the temperature difference (T-TBASE) and then added to TBASE. T is the current
temperature.
Command Default
Replace previous values.
Notes
Allows repeated degree of freedom constraint values (displacement, temperature, etc.) to be replaced, added,
or ignored. Operations apply to the selected nodes [NSEL] and the selected degree of freedom labels
[DOFSEL]. This command also operates on velocity and acceleration loads applied in a structural analysis.
The operations occur when the next degree of freedom constraints are defined. For example, issuing the
command D,1,UX,.025 after a previous D,1,UX,.020 causes the new value of the displacement on node 1 in
the x-direction to be 0.045 with the add operation, 0.025 with the replace operation, or 0.020 with the ignore
357
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DCUM
operation. Scale factors are also available to multiply the next value before the add or replace operation. A
scale factor of 2.0 with the previous "add" example results in a displacement of 0.070. Scale factors are applied
even if no previous values exist. Issue DCUM,STAT to show the current label, operation, and scale factors.
Solid model boundary conditions are not affected by this command, but boundary conditions on the FE
model are affected.
Note
FE boundary conditions may still be overwritten by existing solid model boundary conditions if
a subsequent boundary condition transfer occurs.
DCUM does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Constraints
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Constraints
DCVSWP, Option, Elem, Cnum, Vmax, Vinc1, Vinc2, Gap
Performs a DC voltage sweep on a ROM element.
REDUCED ORDER MODELING: Use Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Option
Sweep option:
GV
Perform voltage sweep up to given voltage Vmax.
PI
Perform a voltage sweep simulation up to the pull-in voltage.
Elem
Element number of the ROM element for the ROM use pass analysis.
Cnum
Number of sweep conductor.
Vmax
Maximum voltage. For the PI option, this voltage should be below the pull-in voltage value.
Vinc1
Voltage increment for Vmax (default = Vmax/20).
Vinc2
Voltage increment for pull-in voltage (default = 1).
Gap
Gap elements option:
0
Create gap elements (COMBIN40) (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
358
DCVSWP
1
Do not create gap elements
Notes
Vinc1 is used to ramp the sweep conductor voltage from 0 to Vmax. Vinc2 is used to increase the sweep
conductor voltage from Vmax to the pull-in value if the PI sweep option is used.
Because ramping the voltage may lead to the unstable region of an electromechanical system, DCVSWP
might not converge when the sweep conductor voltage approaches the pull-in value. To avoid non-converged
solutions, you should use the gap option to create a set of spring-gap elements (COMBIN40). By default,
DCVSWP creates two spring-gap elements with opposite orientations for each active modal displacement
DOF of the ROM element. The gap size is set to the maximum absolute values of the deflection range for
the corresponding mode, as calculated by RMMSELECT or modified using the RMMRANGE command. The
spring constants are set to 1.E5 for all the COMBIN40 elements. Along with the spring-gap elements, DCVSWP
creates a set of constraint equations relating the ROM element modal displacements DOF (EMF) and the
displacement DOF (UX) of the gap elements. Constraining the modal displacements using the spring-gap
elements allows DCVSWP to converge in the pull-in range. The DCVSWP macro has a limit of 900 equilibrium
iterations. If this limit is not sufficient to reach convergence, try the advanced techniques given in Overcoming
Convergence Problems in the Structural Analysis Guide. For more information on gap elements, see Using
Gap Elements with ROM144 in the Coupled-Field Analysis Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>ROM Tools>Voltage Sweep
Main Menu>Solution>ROM Tools>Voltage Sweep
DDELE, NODE, Lab, NEND, NINC, Rkey
Deletes degree-of-freedom constraints.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Node for which constraint is to be deleted. If ALL, NEND and NINC are ignored and constraints for all
selected nodes [NSEL] are deleted. If NODE = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI). A component name may also be substituted for NODE.
Lab
Valid degree of freedom label. If ALL, use all selected labels [DOFSEL]. Structural labels: UX, UY, or UZ
(displacements); ROTX, ROTY, or ROTZ (rotations); WARP (warping). Thermal labels: TEMP, TBOT, TE2, TE3,
. . ., TTOP (temperature). FLOTRAN fluid labels: PRES (pressure); VX, VY, or VZ (velocities); ENKE or ENDS
(turbulent kinetic energy or turbulent energy dissipation); SP01 through SP06 (multiple species mass
fractions) or their user-defined names. Electric label: VOLT (voltage). Magnetic labels: MAG (scalar mag-
netic potential); AX, AY, or AZ (vector magnetic potentials). High-frequency electromagnetic label: AX
(Electric Wall or Magnetic Wall boundary condition).
359
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DDELE
In structural analyses, the following velocity and acceleration load labels are also valid: VELX, VELY, VELZ
(translational velocities); OMGX, OMGY, OMGZ (rotational velocities); ACCX, ACCY, ACCZ (translational
accelerations); DMGX, DMGY, DMGZ (rotational accelerations).
If the node is connected to an ELBOW290 element, the following pipe cross-section degree of freedom
labels are also valid: SE, SO, SW, SRA, and SRT. (For details, see the ELBOW290 documentation.) The degrees
of freedom are not included when Lab = ALL. To constrain all cross-section degrees of freedom, specify
Lab = SECT.
NEND, NINC
Delete constraints from NODE to NEND (defaults to NODE) in steps of NINC (defaults to 1).
Rkey
Ramping key:
OFF
Loads are step-removed (default).
ON or FORCE
Forces on the specified degrees of freedom (Lab) are ramped during the next load step. The forces
are ramped from the reaction forces of the previous load step to zero or currently defined force
values (if any).
Notes
Deleting a constraint is not the same as setting it to zero (which "fixes" the degree of freedom to a zero
value). Deleting a constraint has the same effect as deactivating, releasing, or setting the constraint "free."
The node and the degree of freedom label must be selected [NSEL, DOFSEL].
For elements HF119 and HF120, used in high-frequency electromagnetic analysis, the AX DOF is not an x-
component of a vector potential, but rather a tangential component of E (the electric field) on the element
edges and faces. To specify an Electric Wall condition, set AX to zero. For more information, see the High-
Frequency Electromagnetic Analysis Guide.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Constraint>On All Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>PerfEC>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>TimeInt>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>Voltage>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Displacement>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Species>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Turbulence>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Velocity>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>EdgeMVP>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>ScalarPot>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Node Com-
ponents
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Nodes
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
360
DDELE
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Temperature>On Nodes
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Delete>On Nodes
Main Menu>Solution>Constraints>Delete>On Nodes
Main Menu>Solution>Define Loads>Delete>All Load Data>All Constraint>On All Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>PerfEC>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>TimeInt>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>Voltage>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Displacement>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Species>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Turbulence>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Velocity>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>EdgeMVP>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>ScalarPot>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Node Components
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Temperature>On Nodes
DDOPTION, Decomp, --, --, ContKey
Sets domain decomposer option for Distributed ANSYS.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Decomp
Controls which domain decomposition algorithm to use.
AUTO
Use the default domain decomposition algorithm when splitting the model into domains for Distrib-
uted ANSYS. Currently, the default for all jobs is the GREEDY algorithm.
GREEDY
Use the greedy domain decomposition algorithm.
METIS
Use the METIS graph partitioning domain decomposition algorithm.
-- , --
Unused fields.
ContKey
Tells the domain decomposition algorithm how to handle contact pairs.
DEFAULT
By default, all elements belonging to a single contact pair must belong to a single CPU domain.
However, each contact pair may reside on a different CPU domain unless multiple contact pairs
(along with the required underlying elements) overlap, in which case the entire overlapping set must
reside within a single CPU domain.
GROUP
This option forces all of the contact pairs in the model (along with the required underlying elements)
to be grouped onto a single CPU domain.
361
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DDOPTION
Command Default
The greedy domain decomposition algorithm is used.
Notes
This command controls options relating to the domain decomposition algorithm used by Distributed ANSYS
to split the original model into pieces, with each piece being solved on a different CPU.
The greedy domain decomposition algorithm starts from a single element at a corner of the model. The
domain grows by taking the properly connected neighboring elements and stops after reaching the optimal
size.
The METIS domain decomposition algorithm starts by creating a graph from the finite element mesh. It then
uses a multilevel graph partitioning scheme which reduces the size of the original graph, creates domains
using the reduced graph, and then creates the final CPU domains by expanding the smaller domains from
the reduced graph back to the original mesh.
Typically, when using just a few different element types in the model, the METIS decomposer will result in
less communication and better load balancing. This often leads to decreased overall run times, particularly
when using higher numbers of processors (for example, 8 or more CPUs) or when using cluster configurations
with slower interconnect speeds (for example, 4 machines with 1 CPU per machine using a GigE interconnect).
However, the METIS decomposition method is sometimes not as robust as the greedy algorithm and,
therefore, is not the default.
Menu Paths
This command cannot be accessed from a menu.
DEACT
Specifies "Element birth and death" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Elem Birth/Death
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
362
DEACT
DECOMP, --, Ndomains
Decomposes the model into domains used by Distributed ANSYS
PREP7: Special Purpose
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS
--
Unused field
Ndomains
Sets the number of preconditioner domains for model decomposition.
AUTO
Automatically set the number of domains (default).
N
Set the number of domains to be used by the preconditioner of the solver to N>1.
Command Default
DECOMP,,AUTO
Notes
This command breaks (decomposes) the model into domains for previewing purposes through the /PNUM
and EPLOT commands. See the Distributed ANSYS Guide.
If you specify DECOMP (no arguments), the command defaults to DECOMP,,AUTO. The domains decomposed
here are the domains that are used by a solver's preconditioners only. The larger domains built based on
these preconditioner domains are called CPU domains.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Domain Decomp
Main Menu>Solution>Define Loads>Domain Decomp
DEFINE
Specifies "Data definition settings" as the subsequent status topic.
POST1: Status
POST26: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
363
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DEFINE
Menu Paths
Utility Menu>List>Status>General Postproc>Read Options
Utility Menu>List>Status>TimeHist Postproc>Variables
*DEL, Val1, Val2
Deletes a parameter or parameters (GUI).
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Val1
Val1 can be:
ALL --
Indicates that you want to delete all user-defined parameters, or both all user-defined and all system
parameters, as indicated by the Val2 argument.
(blank) --
Indicates that you want to delete the parameter(s) indicated by Val2.
Val2
Val2 can be:
LOC --
When Val1 is (blank), use Val2 to specify the location of the parameter within the Array Parameters
dialog box. The location number is based on an alphabetically ordered list of all parameters in the
database. Not valid when Val1 is ALL.
_PRM --
When Val1 is ALL, specifying _PRM for Val2 deletes all parameters, including those named with
an initial underbar (_) (except _STATUS and _RETURN). When Val1 is (blank), specifying _PRM for
Val2 deletes only those parameters named with an initial underbar (_) (except _STATUS and _RE-
TURN).
PRM_ --
When Val1 is (blank), specifying PRM_ for Val2 deletes only those parameters named with a trailing
underbar (_). Not valid when Val1 is ALL.
(blank) --
When Val1 is ALL, specifying (blank) for Val2 causes all user-defined parameters to be deleted.
Notes
This is a command generally created by the Graphical User Interface (GUI). It will appear in the log file
(Jobname.LOG) if an array parameter is deleted from within the Array Parameters dialog box.
To delete all user-defined parameters, issue the command *DEL,ALL. To delete only those user-defined
parameters named with a trailing underbar, issue the command *DEL,,PRM_. To delete all user-defined and
all system parameters (except for _STATUS and _RETURN), issue the command *DEL,ALL,_PRM. To delete a
parameter by specifying its location within the Array Parameters dialog box, issue the command *DEL,,LOC.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
364
*DEL
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
DELETE, SET, Nstart, Nend
Specifies sets in the results file to be deleted before postprocessing.
AUX3: Results Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SET
Specifies that sets in the results file are to be deleted.
Nstart
The first set in a results file to be deleted.
Nend
The final set in a results file to be deleted. This field is used only if deleting more than one sequential
sets.
Notes
DELETE is a specification command that flags sets in the results file for deletion. It should be followed by
a COMPRESS command, the corresponding action command that deletes the specified sets.
The DELETE command is valid only in the results file editing processor (ANSYS auxiliary processor AUX3).
Menu Paths
This command cannot be accessed from a menu.
/DELETE, Fname, Ext, --
Deletes a file.
SESSION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to the current Jobname.
Ext
Filename extension (8 character maximum).
--
Unused field.
365
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/DELETE
Notes
For Distributed ANSYS, if Fname is the current Jobname (or blank, which defaults to the current Jobname),
the command will be executed on the master and all slave processors. However, if Fname is anything other
than the current Jobname, this command is ignored on the slave processors and is executed only on the
master processor.
Menu Paths
Utility Menu>File>File Operations>Delete
Utility Menu>PlotCtrls>Redirect Plots>To GRPH File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL2 File
Utility Menu>PlotCtrls>Redirect Plots>To PSCR File
DELTIM, DTIME, DTMIN, DTMAX, Carry
Specifies the time step sizes to be used for this load step.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
DTIME
Time step size for this step. If automatic time stepping is being used [AUTOTS], DTIME is the starting
time substep. If SOLCONTROL,ON and contact elements TARGE169, TARGE170, CONTA171, CONTA172,
CONTA173, CONTA174, CONTA175, CONTA176, or CONTA177 are used, defaults to 1 or 1/20 the total
time span of the load step, depending on the physics of the problem. If SOLCONTROL,ON and none of
these contact elements are used, defaults to 1 time span of the load step. If SOLCONTROL,OFF, defaults
to the previously specified value.
DTMIN
Minimum time step (if automatic time stepping is used). If SOLCONTROL,ON, default determined by
ANSYS depending on the physics of the problem. If SOLCONTROL,OFF, defaults to the previously specified
value (or DTIME, if there is no previously specified value).
DTMAX
Maximum time step (if automatic time stepping is used). If SOLCONTROL,ON, default determined by
ANSYS depending on the physics of the problem. If SOLCONTROL,OFF, defaults to the previously specified
value (or the time span of the load step, if there is no previously specified value).
Carry
Time step carry over key:
OFF
Use DTIME as time step at start of each load step.
ON
Use final time step from previous load step as the starting time step (if automatic time stepping is
used).
If SOLCONTROL,ON, default determined by ANSYS depending on the physics of the problem. If SOLCON-
TROL,OFF, defaults to OFF.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
366
DELTIM
Notes
See NSUBST for an alternative input.
Use consistent values for DTIME and TIME (TIME). For example, using 0.9 for DTIME and 1.0 for TIME
results in one time step because 1.0 (TIME) is divisible by .9 (DTIME) at most once. If your intent is to load
in 10 increments over a time span of 1.0, then use 0.1 for DTIME and 1.0 for TIME.
ANSYS calculates the initial incremental time so that (EndingTime - StartingTime) / DELTIM is an integer,
which may affect the initial incremental time that you specify. For example, if the starting time is 0, the
ending time is 1, and the initial incremental time is 0.4, ANSYS rounds to the nearest integer and adjusts
the time to 0.33333.
For the sake of solution efficiency, ANSYS recommends specifying values for all fields of this command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Basic
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Solution>Analysis Type>Sol'n Controls>Basic
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time - Time Step
DEMORPH, ELEM, DIMN, RMSHKY
Move nodes in selected elements to conform to structural displacements.
PREP7: Morphing
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
ELEM
Non-structural elements to which mesh movement (morph) applies. If ALL, apply morphing to all selected
elements [ESEL]. If ELEM = P, graphical picking is enabled. A component may be substituted for ELEM.
DIMN
Problem dimensionality. Use "2" for a 2-D problem and "3" for a 3-D problem (no default).
RMSHKY
Remesh flag option:
0
Remesh the selected non-structural regions only if mesh morphing fails.
1
Remesh the selected non-structural regions and bypass mesh morphing.
2
Perform mesh morphing only and do not remesh.
Notes
The selected elements should include only non-structural regions adjacent to structural regions. The exterior
nodes of the selected elements will usually be on the boundary of the region which will have node positions
367
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DEMORPH
displaced. For DIMN = 2, elements must lie on a flat plane. The DEMORPH command requires a single domain
grouping of elements be provided (multiple domains of elements are not permitted). Exterior nodes will be
assumed fixed (no nodes will be morphed) unless they coincide with structural nodes having nonzero dis-
placements.
Nodes in the structural regions move in accordance with computed displacements. Displacements from a
structural analysis must be in the database prior to issuing DEMORPH.
By default (RMSHKY = 0), DEMORPH will remesh the selected non-structural regions entirely if a satisfactory
morphed mesh cannot be provided.
If boundary conditions and loads are applied directly to nodes and elements, the DEMORPH command re-
quires that these be removed before remeshing can take place.
Exercise care with initial conditions defined by the IC command. Before a structural analysis is performed
for a sequentially coupled analysis, the DEMORPH command requires that initial conditions be removed
from all null element type nodes in the non-structural regions. Use ICDELE to delete the initial conditions.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Phys Morphing>Elements
DERIV, IR, IY, IX, --, Name, --, --, FACTA
Differentiates a variable.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IY, IX
Reference numbers of variables to be operated on. IY is differentiated with respect to IX.
--
Unused field.
Name
Thirty-two character name for identifying the variable on printouts and displays. Embedded blanks are
compressed for output.
--, --
Unused fields.
FACTA
Scaling factor (positive or negative) applied as shown below (defaults to 1.0).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
368
DERIV
Notes
Differentiates variables according to the operation:
IR = FACTA x d(IY)/d(IX)
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Derivative
DESIZE, MINL, MINH, MXEL, ANGL, ANGH, EDGMN, EDGMX, ADJF, ADJM
Controls default element sizes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MINL
Minimum number of elements that will be attached to a line when using lower-order elements (defaults
to 3 elements per line). If MINL = DEFA, all arguments will be set back to default values. If MINL = STAT,
list status of command (Including on/off status). If MINL = OFF, deactivate default element sizing. If
MINL = ON, reactivate default element sizing.
MINH
Minimum number of elements that will be attached to a line when using higher-order elements. Defaults
to 2 elements per line.
MXEL
Maximum number of elements that will be attached to a single line (lower or higher-order elements).
Defaults to 15 elements per line for h-elements and 6 divisions per line for p-elements. To deactivate
this limit, specify a large number (such as 9999).
ANGL
Maximum spanned angle per lower-order element for curved lines. Defaults to 15 degrees per element.
ANGH
Maximum spanned angle per higher-order element for curved lines. Defaults to 28 degrees per element.
EDGMN
Minimum element edge length. Defaults to no minimum edge length. The MINL or MINH argument can
override this value.
EDGMX
Maximum element edge length. Defaults to no maximum edge length. The MXEL argument can override
this value.
ADJF
Target aspect ratio for adjacent line. Used only when free meshing. Defaults to 1.0, which attempts to
create equal sided h-elements; defaults to 4 for p-elements.
ADJM
Target aspect ratio for adjacent line. Used only when map meshing. Defaults to 4.0, which attempts to
create rectangular h-elements; defaults to 6 for p-elements.
369
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DESIZE
Command Default
Default settings as described for each argument are used.
Notes
DESIZE settings are usually used for mapped meshing. They are also used for free meshing if SmartSizing
is turned off [SMRTSIZE,OFF], which is the default. Even when SmartSizing is on, some DESIZE settings (such
as maximum and minimum element edge length) can affect free mesh density. The default settings of the
DESIZE command are used only when no other element size specifications [KESIZE, LESIZE, ESIZE] exist
for a certain line.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Global>Other
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Default Element
Sizing Options
DESOL, ELEM, NODE, Item, Comp, V1, V2, V3, V4, V5, V6
Defines or modifies solution results at a node of an element.
POST1: Set Up
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
ELEM
Element number for which results are defined or modified. If ALL, apply to all selected elements [ESEL].
NODE
Node of element (actual node number, not the position) to which results are specified. If ALL, specify
results for all selected nodes [NSEL] of element. If NODE = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI). A component name may also be substituted for
NODE.
Item
Label identifying results. Valid item labels are shown in Table 147: DESOL - Valid Item and Component La-
bels (p. 371) below. Some items also require a component label (Comp).
Comp
Component of the item (if required); see Table 147: DESOL - Valid Item and Component Labels (p. 371).
V1
Value (in the element coordinate system) assigned to the database item (and component, if any). If zero,
a zero value will be assigned. If blank, value remains unchanged.
V2, V3, V4, . . . , V6
Additional values (if any) assigned to the remaining components (in the order corresponding to the
Comp list shown below) for the specified Item (starting from the specified Comp label and proceeding
to the right).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
370
DESOL
Notes
The DESOL command defines or modifies solution results in the database at a node of an area or volume
element. For example, DESOL,35,50,S,X,1000,2000,1000 assigns values 1000, 2000, and 1000 to SX, SY, and
SZ (respectively) of node 50 of element 35.
The settings of the POST1 FORCE, SHELL, and LAYER commands, if applicable, further specify which database
items are affected.
For layered composite shells, specify the current element layer (LAYER) before issuing the DESOL command.
All data is stored in the solution coordinate system but is displayed in the results coordinate system (RSYS).
To list the current results, use the PRESOL command.
Modified solution results are not saved automatically. To save separate records of modified results, use either
the RAPPND or LCWRITE command.
Result items are available depending on element type; check the individual element for availability. Valid
item and component labels for element results are:
Table 147 DESOL - Valid Item and Component Labels
Description Comp Item
Element number. ELEM
Component stress. X, Y, Z, XY, YZ, XZ S
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Elastic strain energy density. ELASTIC SEND
Plastic strain energy density. PLASTIC "
Creep strain energy density. CREEP "
Component thermal gradient. X, Y, Z TG
Component thermal flux. X, Y, Z TF
Component pressure gradient. X, Y, Z PG
Component electric field. X, Y, Z EF
Component electric flux density. X, Y, Z D
Component magnetic field intensity. X, Y, Z H
371
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DESOL
Description Comp Item
Component magnetic flux density. X, Y, Z B
Component electromagnetic force. X, Y, Z FMAG
X, Y, or Z structural force. X, Y, Z F
X, Y, or Z structural moment. X, Y, Z M
Heat flow. HEAT
Fluid flow. FLOW
Current flow. AMPS
Magnetic flux. FLUX
X, Y, or Z magnetic current segment component. X, Y, Z CSG
Menu Paths
Main Menu>General Postproc>Define/Modify>Elem Results
DETAB, ELEM, Lab, V1, V2, V3, V4, V5, V6
Modifies element table results in the database.
POST1: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ELEM
Element for which results are to be modified. If ALL, modify all selected elements [ESEL] results. If ELEM
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may also be substituted for ELEM.
Lab
Label identifying results. Valid labels are as defined with the ETABLE command. Issue ETABLE,STAT to
display labels and values.
V1
Value assigned to this element table result in the database. If zero, a zero value will be assigned. If blank,
value remains unchanged.
V2, V3, V4, . . . , V6
Additional values (if any) assigned to consecutive element table columns.
Notes
Modifies element table [ETABLE] results in the database. For example, DETAB,35,ABC,1000,2000,1000 assigns
1000, 2000, and 1000 to the first three table columns starting with label ABC for element 35. Use the PRETAB
command to list the current results. After deleting a column of data using ETABLE,Lab,ERASE, the remaining
columns of data are not shifted to compress the empty slot. Therefore, the user must allocate null (blank)
values for V1, V2...V6 for any ETABLE entries which have been deleted by issuing ETABLE,Lab,ERASE. All
data are stored in the solution coordinate system but will be displayed in the results coordinate system
[RSYS].
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
372
DETAB
Menu Paths
Main Menu>General Postproc>Define/Modify>ElemTabl Data
/DEVDISP, Label, KEY
Controls graphics device options.
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Device function label:
BBOX
Disables display information sorting for PowerGraphics displays. When activated (KEY = 1 or ON),
model rotations and replots are performed without recalculating edge and surface data. This will
speed up the rotation (especially for 2-D displays) of large models, although the display information
will not be resolved as quickly (you will see a bounding box instead of the model during dynamic
rotations). The default is OFF (KEY = 0).
DITHER
Dithering. When turned on (default), dithering smooths transitions in color intensity. Applies only to
Z-buffered displays.
FONT
Font selection for the ANSYS graphics window. When Label = FONT, the command format is:
/DEVDISP,FONT,KEY,Val1,Val2,Val3,VAL4,Val5,Val6, where KEY determines the type of font
being controlled, and values 1 through 6 control various font parameters. Note that these values are
device specific; using the same command input file [/INPUT] on different machines may yield different
results.The following KEY values determine the font information that will be supplied to the appro-
priate driver (e.g., Postscript, X11, Win32, JPEG, ...):
KEY = 1
The command controls the LEGEND (documentation column) font.
KEY = 2
The command controls the ENTITY (node and keypoint number) font.
KEY = 3
The command controls the ANNOTATION/GRAPH font.
UNIX: Values 1 through 4 are used to find a match in the X11 database of font strings. Values 1,
2, and 3 are character strings; value 4 is a nonzero integer:
Val1
Family name (e.g., Courier*New). Substitute an asterisk (*) for any blank character that appears
in a family name. If Val1 = MENU, all other values are ignored, and a font selection menu appears
(GUI must be active).
Val2
Weight (e.g., medium)
Val3
Slant (e.g., r)
373
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/DEVDISP
Val4
Pixel size (e.g., 14). Note that this value does not affect the annotation fonts (KEY = 3). Use the
/TSPEC command to control the pixel size of your annotation fonts.
Val5
unused
Val6
unused
PC: The values are encoded in a PC logical font structure. Value 1 is a character string, and the
remaining values are integers:
Val1
Family name (e.g., Courier*New) Substitute an asterisk (*) for any blank character that appears
in a family name. If Val1 = MENU, all other values are ignored and a font selection menu appears
(GUI must be active). A value containing all blank characters causes ANSYS to use the first available
resource it finds.
Val2
Weight (0 - 1000)
Val3
Orientation (in tenths of a degree)
Val4
Height (in logical units) Note that this value does not affect the annotation fonts (KEY =3). Use
the /TSPEC command to control the height of your annotation fonts.
Val5
Width (in logical units)
Val6
Italics (0 = OFF, 1 = ON)
TEXT
Text size specification for the ANSYS Graphics window. When Label = TEXT, the command format is:
/DEVDISP,TEXT,KEY,PERCENT, where KEY determines the type of text being controlled (1 for LEGEND,
and 2 for ENTITY), and PERCENT specifies the new text size as a percent of the default text size. If
PERCENT = 100, the new text size is precisely the default size. If PERCENT = 200, the new text size
is twice the default text size.
KEY
Control key:
OFF or 0
Turns specified function off.
ON or 1
Turns specified function on.
Command Default
Dithering on.
Menu Paths
It is part of the DISPLAY program.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
374
/DEVDISP
/DEVICE, Label, KEY
Controls graphics device options.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Device function label:
BBOX
Bounding box mode. For PowerGraphics plots involving elements with /SHOW,x11 and /SHOW,win32,
ANSYS generally displays dynamic rotations faster. If KEY = 1 (ON), then a bounding box (not the
elements) encompassing the model is displayed and rotated, rather than the element outlines (ON
is default in preprocessing). When KEY = 0 (OFF), then dynamic rotations may be slower (ANSYS re-
draws the element outlines) for plots involving elements with /SHOW,x11 and /SHOW,win32. OFF
is default in postprocessing. This command is ignored if /EDGE,WN,1 is set for any WN. This is ignored
in POST1 and SOLUTION plots.
For any PowerGraphics plots involving elements, regardless of /SHOW settings, plots will generally
be displayed faster.
VECTOR
Vector mode. In vector mode, areas, volumes, elements, and postprocessing display geometries are
shown as outlines (wireframes). When vector mode is off (default), these entities are shown filled
with color.
DITHER
When dithering is turned on (default), color intensity transitions are smoothed. This selection a
applies only to smooth-shaded images, i.e., Z-buffered [/TYPE], or raster plots with Gouraud or Phong
shading [/SHADE].
ANIM
Select the animation type used on 2-D devices on the PC platform. A KEY value of BMP (or 0) sets
animation mode to ANSYS Animation Controller (default). A KEY value of AVI (or 2) sets animation
mode to AVI movie player file.
FONT
Font selection for the ANSYS graphics window. When Label = FONT, the command format is:
/DEVICE,FONT,KEY,Val1,Val2,Val3,Val4,Val5,Val6 where KEY determines the type of font
being controlled, and values 1 through 6 control various font parameters. Note that these values are
device specific; using the same command input file [/INPUT] on different machines may yield different
results.. The following KEY values determine the font information that will be supplied to the appro-
priate driver (e.g., Postscript, X11, Win32, JPEG, ...):
KEY = 1
The command controls the LEGEND (documentation column) font.
KEY = 2
The command controls the ENTITY (node and keypoint number) font.
KEY = 3
The command controls the ANNOTATION/GRAPH font.
UNIX: Values 1 through 4 are used to find a match in the X11 database of font strings. Values 1,
2, and 3 are character strings; value 4 is a nonzero integer:
375
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/DEVICE
Val1
Family name (e.g., Courier). If Val1 = MENU, all other values are ignored and a font selection
menu appears (GUI must be active).
Val2
Weight (e.g., medium)
Val3
Slant (e.g., r)
Val4
Pixel size (e.g., 14). Note that this value does no affect the annotation fonts (KEY = 3). Use the
/TSPEC command for annotation font size.
Val5
unused
Val6
unused
PC: The values are encoded in a PC logical font structure. Value 1 is a character string, and the
remaining values are integers:
Val1
Family name (e.g., Courier*New) Substitute an asterisk (*) for any blank character that appears
in a family name. If Val1 = MENU, all other values are ignored and a font selection menu appears
(GUI must be active). When this value is blank ANSYS uses the first available resource it finds.
Val2
Weight (0 - 1000)
Val3
Orientation (in tenths of a degree)
Val4
Height (in logical units)
Val5
Width (in logical units)
Val6
Italics (0 = OFF, 1 = ON)
TEXT
Text size specification for the ANSYS Graphics window. Using this label with the /DEVICE command
requires the following form: /DEVICE,TEXT,KEY,PERCENT. KEY = 1 for LEGEND fonts; KEY = 2 for
ENTITY fonts. PERCENT specifies the new text size as a percent of the default text size. If PERCENT
= 100, the new text size is precisely the default size. If PERCENT = 200, the new text size is twice
the default text size.
KEY
Control key:
OFF or 0
Turns specified function off.
ON or 1
Turns specified function on or designates the LEGEND font.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
376
/DEVICE
2
Designates the ENTITY font.
3
Designates the ANNOTATION/GRAPH font.
Command Default
Vector mode off (i.e., raster mode); dithering on.
Notes
This command is valid in any processor.
The /DEVICE,BBOX command is ignored in POST1 and SOLUTION plots. Also, the elements are displayed
and rotated if you use /DEVICE,BBOX,ON and /EDGE,WN,1,ANGLE (effectively ignoring the BBOX option).
Menu Paths
Utility Menu>PlotCtrls>Device Options
Utility Menu>PlotCtrls>Font Controls>Entity Font
Utility Menu>PlotCtrls>Font Controls>Legend Font
Utility Menu>PlotCtrls>Redirect Plots>To PSCR File
/DFLAB, DOF, DispLab, ForceLab
Changes DOF labels for user custom elements.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Argument Descriptions
DOF
Number indicating which DOF is to have its labels changed. For example:
1 = UX,FX
2 = UY,FY
3 = UX,FZ
4 = ROTX,MX
etc.
DispLab
New label (4 character maximum) for the displacement label. The old label is no longer valid.
ForceLab
New label (4 character maximum) for the force label for this degree-of-freedom. The old label is no
longer valid.
Notes
The /DFLAB command is rarely used. It is designed for users who are writing custom elements for ANSYS
and want to use degrees-of-freedom that are not part of the standard ANSYS set.
377
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/DFLAB
Menu Paths
This command cannot be accessed from a menu.
DIG, NODE1, NODE2, NINC
Digitizes nodes to a surface.
PREP7: Digitizing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1, NODE2, NINC
Digitize nodes NODE1 through NODE2 in steps of NINC. NODE2 defaults to NODE1 and NINC defaults
to 1.
Command Default
No surface digitizing.
Notes
Digitizes nodes to the surface defined by the DSURF command. The nodes indicated must be digitized from
the tablet after this command is given. The program must be in the interactive mode and the graphics ter-
minal show option [/SHOW] must be active. The global Cartesian coordinates of the nodes are stored.
Menu Paths
Main Menu>Preprocessor>Create>Nodes>Digitize Nodes>Digitize Nodes
DIGIT
Specifies "Node digitizing" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utilty Menu>List>Status>Preprocessor>Digitize Module
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
378
DIG
*DIM, Par, Type, IMAX, JMAX, KMAX, Var1, Var2, Var3, CSYSID
Defines an array parameter and its dimensions.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
Name of parameter to be dimensioned. See *SET for name restrictions.
Type
Array type:
ARRAY --
Arrays are similar to standard FORTRAN arrays (indices are integers) (default). Index numbers for the
rows, columns, and planes are sequential values beginning with one. Used for 1-, 2-, or 3-D arrays.
ARR4 --
Same as ARRAY, but used to specify 4-D arrays.
ARR5 --
Same as ARRAY, but used to specify 5-D arrays.
CHAR --
Array entries are character strings (up to 8 characters each). Index numbers for rows, columns, and
planes are sequential values beginning with one.
TABLE --
Array indices are real (non-integer) numbers which must be defined when filling the table. Index
numbers for the rows and columns are stored in the zero column and row "array elements" and are
initially assigned a near-zero value. Index numbers must be in ascending order and are used only
for retrieving an array element. When retrieving an array element with a real index that does not
match a specified index, linear interpolation is done among the nearest indices and the corresponding
array element values [*SET]. Used for 1-, 2-, or 3-D tables.
TAB4 --
Same as TABLE, but used to specify 4-D tables.
TAB5 --
Same as TABLE, but used to specify 5-D tables.
STRING --
Array entries are character strings (up to IMAX each). Index numbers for columns and planes are
sequential values beginning with 1. Row index is character position in string.
IMAX
Extent of first dimension (row) (For Type = STRING, IMAX is rounded up to the next multiple of eight
and has a limit of 128). Defaults to 1.
JMAX
Extent of second dimension (column). Defaults to 1.
KMAX
Extent of third dimension (plane). Defaults to 1.
Var1
Variable name corresponding to the first dimension (row) for Type = TABLE. Defaults to Row.
379
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*DIM
Var2
Variable name corresponding to the second dimension (column) for Type = TABLE. Defaults to Column.
Var3
Variable name corresponding to the third dimension (plane) for Type = TABLE. Defaults to Plane.
CSYSID
An integer corresponding to the coordinate system ID Number.
Notes
Up to three dimensions (row, column, and plane) may be defined using ARRAY and TABLE. Use ARR4, ARR5,
TAB4, and TAB5 to define up to five dimensions (row, column, plane, book, and shelf ). An index number is
associated with each row, column, and plane. For array and table type parameters, element values are ini-
tialized to zero. For character and string parameters, element values are initialized to (blank). A defined
parameter must be deleted [*SET] before its dimensions can be changed. Scalar (single valued) parameters
should not be dimensioned. *DIM,A,,3 defines a vector array with elements A(1), A(2), and A(3). *DIM,B,,2,3
defines a 2x3 array with elements B(1,1), B(2,1), B(1,2), B(2,2), B(1,3), and B(2,3). Use *STATUS,Par to display
elements of array Par. You can write formatted data files (tabular formatting) from data held in arrays
through the *VWRITE command.
If you use table parameters to define boundary conditions, then Var1, Var2, and/or Var3 can either specify
a primary variable (listed in Table 148: *DIM - Primary Variables (p. 381)) or can be an independent parameter.
If specifying an independent parameter, then you must define an additional table for the independent
parameter. The additional table must have the same name as the independent parameter and may be a
function of one or more primary variables or another independent parameter. All independent parameters
must relate to a primary variable.
Tabular load arrays can be defined in both global Cartesian (defalut) or local (see below) coordinate systems
by specifying CSYSID, as defined in LOCAL. For batch operations, you must specify your coordinate system
first.
The following constraints apply when you specify a local coordinate system for your tabular loads:
Only Cartesian, cylindrical and spherical coordinate systems are supported
Angle values for Y in cylindrical or spherical coordinate systems must be input in degrees and must be
positive values between 0 and 360 degrees ( 0
360)
Angle values for Z in spherical coordinate system must be input in degrees and must be positive values
between -90 and +90 ( -90
90)
If you are specifying a 4- or 5-D array or table, four additional fields (LMAX, MMAX, Var4, and Var5) are
available. Thus, for a 4-D table, the command syntax would be:
*DIM,Par,Type,IMAX,JMAX,KMAX,LMAX,Var1,Var2,Var3,Var4,CSYSID
For a 5-D table, the command syntax would be:
*DIM,Par,Type,IMAX,JMAX,KMAX,LMAX,MMAX,Var1,Var2,Var3,Var4,Var5,CSYSID
You cannot create or edit 4- or 5-D arrays or tables using the GUI.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
380
*DIM
See Array Parameters for a detailed discussion on and examples for using array parameters.
Table 148 *DIM - Primary Variables
Label for Var1, Var2,
Var3
Primary Variable
TIME Time
FREQ Frequency
X X-coordinate location
Y Y-coordinate location
Z Z-coordinate location
TEMP Temperature
VELOCITY Velocity
PRESSURE Pressure
SECTOR Cyclic sector number
Note
The X, Y, and Z coordinate locations listed above are valid in gobal Cartesian, or local (Cartesian,
cylindrical and spherical) coordinate systems. The VELOCITY label is applicable only to the calculated
fluid velocity in element FLUID116.
When using PRESSURE as a primary variable, the underlying element must have the pressure DOF
associated with it. For example, if you use FLUID141 or FLUID142, apply SF,ALL,CONV,%presh%,20.0
to apply a tabular film coefficient that will be evaluated correctly, based on the pressure. However,
if the underlying element is SURF151 or SURF152, the film coefficient will not change because
there is no underlying pressure.
The frequency label (FREQ) is valid for harmonic analyses only.
If you use table parameters to define boundary conditions, the table names (Par) must not exceed 32
characters.
In thermal analyses, if you apply tabular loads as a function of temperature but the rest of the model is linear
(e.g., includes no temperature-dependent material properties or radiation ), you should turn on Newton-
Raphson iterations (NROPT,FULL) to evaluate the temperature-dependent tabular boundary conditions
correctly.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Parameters>Define/Edit
381
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*DIM
/DIRECTORY, StrArray, FileName, Ext, Dir
Put the file names in the current directory into a string parameter array.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
StrArray
Name of the "string array" parameter which will hold the returned values. String array parameters are
similar to character arrays, but each array element can be as long as 128 characters. If the string para-
meter does not exist, it will be created. The array will be created as: *DIM,StrArray,STRING,64,2,numFile-
Name
FileName
File name (64 characters maximum). Only files matching this name will be returned. The FileName ALL
may match any file name.
Ext
File name extension (8 characters maximum). Only files with an extension matching this name will be
returned. A blank or ALL will match any extension.
Directory
The directory in which the files reside. The default is the current working directory.
Notes
The /DIRECTORY command gets the file names in the current directory and puts them into a string para-
meter array. Each file will be included in the array as a name-extension pair.
Menu Paths
This command cannot be accessed from a menu.
DISPLAY
Specifies "Display settings" as the subsequent status topic.
POST1: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
382
/DIRECTORY
Menu Paths
Utility Menu>List>Status>General Postproc>Plot Results
/DIST, WN, DVAL, KFACT
Specifies the viewing distance for magnifications and perspective.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
DVAL
Distance along the view line from the observer to the focus point (defaults to value producing full-window
display). Distances "too close" to the object will produce excessive magnifications. If DVAL = AUTO, zero,
or blank, the program will calculate the distance automatically. If DVAL = USER, the distance of last display
will be used (useful when last display automatically calculated distance).
KFACT
DVAL interpretation key:
0
Interpret numerical DVAL values as described above.
1
Interpret DVAL as a multiplier on the current distance (DVAL of 2 gives twice the current distance;
0.5 gives half the current distance, etc.).
Command Default
Distance is automatically calculated to produce full window magnification.
Notes
The scale factor is relative to the window shape. For example, for objects centered in a square window and
with parallel projection (no perspective), a distance of
is the largest in-plane vertical or horizontal dimension. See also /AUTO and /USER commands.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Pan, Zoom, Rotate
Utility Menu>PlotCtrls>View Settings>Magnification
383
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/DIST
DJ, ELEM, LABEL, VALUE
Specifies boundary conditions on the components of relative motion of a joint element.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element number or ALL to be specified.
LABEL
Valid labels are:
UX
Displacement in local x direction.
UY
Displacement in local y direction.
UZ
Displacement in local z direction.
ROTX
Rotation about local x axis.
ROTY
Rotation about local y axis.
ROTZ
Rotation about local y axis.
VELX
Linear velocity in local x direction.
VELY
Linear velocity in local y direction.
VELZ
Linear velocity in local z direction.
OMGX
Angular velocity in local x direction.
OMGY
Angular velocity in local y direction.
OMGZ
Angular velocity in local z direction.
ACCX
Linear acceleration in local x direction.
ACCY
Linear acceleration in local y direction.
ACCZ
Linear acceleration in local z direction.
DMGX
Angular acceleration in local x direction.
DMGY
Angular acceleration in local y direction.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
384
DJ
DMGZ
Angular acceleration in local z direction.
VALUE
Value of the label.
Notes
This command is valid for MPC184 joint elements. See DJDELE for information on deleting boundary condi-
tions applied with the DJ command.
You can apply only one displacement, velocity, or acceleration load at any relative degree of freedom. If
multiple loads are specified, the last applied load overrides the previous ones. For example, the following
commands apply loads to element 100:
D,100,UX,Value
D,100,VELX,Value
In this case, the velocity load (VELX) applied in the last command will override the displacement load (UX).
Tabular boundary conditions (VALUE = %tabname%) can be used.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Joint Elems
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Joint Elems
DJDELE, ELEM, LAB
Deletes boundary conditions on the components of relative motion of a joint element.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element number or ALL. ALL (or leaving this field blank) will delete all joint element boundary conditions
specified by LAB.
LAB
Valid labels are:
UX
Displacement in local x direction.
UY
Displacement in local y direction.
UZ
Displacement in local z direction.
ROTX
Rotation about local x axis.
ROTY
Rotation about local y axis.
385
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DJDELE
ROTZ
Rotation about local z axis.
VELX
Linear velocity in local x direction.
VELY
Linear velocity in local y direction.
VELZ
Linear velocity in local z direction.
OMGX
Angular velocity in local x direction.
OMGY
Angular velocity in local y direction.
OMGZ
Angular velocity in local z direction.
ACCX
Linear acceleration in local x direction.
ACCY
Linear acceleration in local y direction.
ACCZ
Linear acceleration in local z direction.
DMGX
Angular acceleration in local x direction.
DMGY
Angular acceleration in local y direction.
DMGZ
Angular acceleration in local z direction.
ALL, or (blank)
Delete all applied boundary conditions.
Notes
This command is valid for MPC184 joint elements. See DJ for information on specifying boundary conditions
on the components of relative motion of a joint element.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Constraint>On Joint
Elems
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Joint Elems
Main Menu>Solution>Define Loads>Delete>All Load Data>All Constraint>On Joint Elems
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Joint Elems
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
386
DJDELE
DJLIST, Elem
Lists boundary conditions applied to joint elements.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Elem
Element number or ALL (or blank). Lists joint element boundary conditions on the specified element(s).
Notes
This command is valid for MPC184 joint elements. See DJ for information on specifying boundary conditions
on joint elements.
Menu Paths
Utility Menu>List>Loads>Joint Element DOF Constraints>On Picked Element
DK, KPOI, Lab, VALUE, VALUE2, KEXPND, Lab2, Lab3, Lab4, Lab5, Lab6
Defines DOF constraints at keypoints.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KPOI
Keypoint at which constraint is to be specified. If ALL, apply to all selected keypoints [KSEL]. If KPOI =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for KPOI.
Lab
Valid degree of freedom label. If ALL, use all appropriate labels except HDSP. Structural labels: UX, UY,
or UZ (displacements); ROTX, ROTY, or ROTZ (rotations); WARP (warping); HDSP (hydrostatic pressure).
Thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP (temperature). Electric labels: VOLT (voltage). Magnetic
labels: MAG (scalar magnetic potential); AX, AY, or AZ (vector magnetic potentials).
VALUE
Degree of freedom value or table name reference for tabular boundary conditions. To specify a table,
enclose the table name in percent signs (%) (e.g., DK,NODE,TEMP,%tabname%). Use the *DIM command
to define a table.
VALUE2
Second degree of freedom value (if any). If the analysis type and the degree of freedom allow a complex
input, VALUE (above) is the real component and VALUE2 is the imaginary component.
KEXPND
Expansion key:
0
Constraint applies only to the node at this keypoint.
1
Flags this keypoint for constraint expansion.
387
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DK
Lab2, Lab3, Lab4, . . . , Lab6
Additional degree of freedom labels. The same values are applied to the keypoints for these labels.
Notes
A keypoint may be flagged using KEXPND to allow its constraints to be expanded to nodes on the attached
solid model entities having similarly flagged keypoint constraints. Constraints are transferred from keypoints
to nodes with the DTRAN or SBCTRAN commands. The expansion uses interpolation to apply constraints
to the nodes on the lines between flagged keypoints. If all keypoints of an area or volume region are flagged
and the constraints (label and values) are equal, the constraints are applied to the interior nodes of the region.
See the D command for a description of nodal constraints.
Tabular boundary conditions (VALUE = %tabname%) are available only for the following degree of freedom
labels: Electric (VOLT), structural (UX, UY, UZ, ROTX, ROTY, ROTZ), and temperature (TEMP, TBOT, TE2, TE3, .
. ., TTOP).
Constraints specified by the DK command can conflict with other specified constraints. See Resolution of
Conflicting Constraint Specifications in the Basic Analysis Guide for details.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>TimeInt>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>ScalarPot>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>BasePSD>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>MultiPtBas>On Key-
points
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>BasePSD>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>MultiPtBas>On
Keypoints
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>On Keypoints
Main Menu>Solution>Constraints>Apply>On Keypoints
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>TimeInt>On Keypoints
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>On Keypoints
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>ScalarPot>On Keypoints
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>BasePSD>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>MultiPtBas>On Keypoints
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>On Keypoints
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>BasePSD>On Keypoints
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>MultiPtBas>On Keypoints
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
388
DK
DKDELE, KPOI, Lab
Deletes DOF constraints at a keypoint.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KPOI
Keypoint for which constraint is to be deleted. If ALL, delete for all selected keypoints [KSEL]. If KPOI
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may also be substituted for KPOI.
Lab
Valid degree of freedom label. If ALL, use all appropriate labels. Structural labels: UX, UY, or UZ (displace-
ments); ROTX, ROTY, or ROTZ (rotations); WARP (warping). Thermal labels: TEMP, TBOT, TE2, TE3, . . .,
TTOP (temperature). FLOTRAN fluid labels: PRES (pressure); VX, VY, or VZ (velocities); ENKE or ENDS (tur-
bulent kinetic energy or turbulent energy dissipation); SP01 through SP06 (multiple species mass fractions)
or their user-defined names. Electric label: VOLT (voltage). Magnetic labels: MAG (scalar magnetic potential);
AX, AY, or AZ (vector magnetic potentials). High-frequency electromagnetic label: AX (Electric Wall or
Magnetic Wall boundary condition).
Notes
Deletes the degree of freedom constraints (and all corresponding finite element constraints) at a keypoint.
See the DDELE command for details.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Constraint>On All KPs
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>TimeInt>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>Voltage>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Displacement>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Velocity>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>ScalarPot>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Key-
points
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Temperature>On Keypoints
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Delete>On Keypoints
Main Menu>Solution>Constraints>Delete>On Keypoints
Main Menu>Solution>Define Loads>Delete>All Load Data>All Constraint>On All KPs
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>TimeInt>On Keypoints
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>Voltage>On Keypoints
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Keypoints
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Displacement>On Keypoints
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Keypoints
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Velocity>On Keypoints
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>ScalarPot>On Keypoints
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Keypoints
389
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DKDELE
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Keypoints
Main Menu>Solution>Define Loads>Delete>Thermal>Temperature>On Keypoints
DKLIST, KPOI
Lists the DOF constraints at keypoints.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KPOI
List constraints for this keypoint. If ALL (default), list for all selected keypoints [KSEL]. If KPOI = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for KPOI.
Notes
Listing applies to the selected keypoints [KSEL] and the selected degree of freedom labels [DOFSEL].
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>DOF Constraints>On All Keypoints
Utility Menu>List>Loads>DOF Constraints>On Picked KPs
DL, LINE, AREA, Lab, Value1, Value2
Defines DOF constraints on lines.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LINE
Line at which constraints are to be specified. If ALL, apply to all selected lines [LSEL]. If LINE = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for LINE
AREA
Area containing line. The normal to the symmetry or antisymmetry surface is assumed to lie on this area.
Defaults to the lowest numbered selected area containing the line number.
Lab
Symmetry label (see 2 (p. 392)):
SYMM
Generate symmetry constraints for non-FLOTRAN models.
ASYM
Generate antisymmetry constraints for non-FLOTRAN models.
ANSYS DOF labels (see 3 (p. 392), 4 (p. 392), and 5 (p. 392)):
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
390
DKLIST
UX
Displacement in X direction.
UY
Displacement in Y direction.
UZ
Displacement in Z direction.
ROTX
Rotation about X axis.
ROTY
Rotation about Y axis.
ROTZ
Rotation about Z axis.
HDSP
Hydrostatic pressure.
WARP
Warping magnitude.
TEMP, TBOT, TE2, TE3, . . ., TTOP
Temperature
VOLT
Electric scalar potential.
AX
Magnetic vector potential in X direction.
AY
Magnetic vector potential in Y direction.
AZ
Magnetic vector potential in Z direction.
ALL
Applies all appropriate DOF labels except HDSP.
FLOTRAN standard DOF labels (see 3 (p. 392)): VX, VY, VZ, PRES, TEMP, ENKE, ENDS
FLOTRAN Species Labels (see 4 (p. 392)): SP01, SP02, SP03, SP04, SP05, SP06
FLOTRAN Arbitrary Lagrangian-Eulerian formulation Mesh Displacement Labels (see 6 (p. 392)): UX, UY,
UZ
Value1
Value of DOF (real part) or table name reference on the line. Valid for all DOF labels. To specify a table,
enclose the table name in % signs (e.g., DL,LINE,AREA,TEMP,%tabname%). Use the *DIM command
to define a table.
If Lab = ENKE and Value1 = -1, a FLOTRAN flag is set to indicate a moving wall.
If Lab = ENDS and Value1 = -1, FLOTRAN generalized symmetry conditions are applied. Velocity com-
ponents are set tangential to the symmetry surface if the ALE formulation is not activated. They are set
equal to the mesh velocity if the ALE formulation is activated.
391
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DL
Value2
For FLOTRAN DOFs:
0
Values are applied only to nodes within the line.
1
Values are applied to the endpoints of the line as well as to the internal nodes (see 3 (p. 392)).
For VOLT DOFs:
Actual value of the imaginary component of the degree of freedom.
Notes
1. You can transfer constraints from lines to nodes with the DTRAN or SBCTRAN commands. See the
DK command for information about generating other constraints at lines.
2. Symmetry and antisymmetry constraints are generated as described on the DSYM command.
3. For the velocity DOF (VX, VY, VZ), a zero value will override a nonzero value at the intersection of two
lines.
4. You can use the MSSPEC command to change FLOTRAN species labels to user-defined labels. You
must define these labels with the MSSPEC command before using them on the DL command.
5. Setting Lab = VOLT and Value1 = 0 applies the J-normal boundary condition (current density vector
(J) flows normal to the line). No input is required for the J-parallel condition because it is the natural
boundary condition.
6. Tabular boundary conditions (VALUE = %tabname%) are available only for the following degree of
freedom labels: Electric (VOLT), FLOTRAN (UX, UY, UZ, PRES, VX, VY, VZ, ENKE, ENDS, TEMP, SP01, SP02,
SP03, SP04, SP05, and SP06); Structural (UX, UY, UZ, ROTX, ROTY, ROTZ), and temperature (TEMP, TBOT,
TE2, TE3, . . ., TTOP).
7. Constraints specified by the DL command can conflict with other specified constraints. See Resolution
of Conflicting Constraint Specifications in the Basic Analysis Guide for details.
8. This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>PerfEC>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>J-Normal>On
Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Voltage>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux
Par'l>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Antisymm
B.C.>...with Area
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>On
Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Symmetry
B.C.>...with Area
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
392
DL
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>On
Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>On Lines
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>...with Area
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>On Lines
Main Menu>Solution>Constraints>Apply>...with Area
Main Menu>Solution>Constraints>Apply>On Lines
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>PerfEC>On Lines
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>J-Normal>On Lines
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Voltage>On Lines
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Par'l>On Lines
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>On Lines
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>...with Area
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>On Lines
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>On Lines
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>...with Area
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>On Lines
DLDELE, LINE, Lab
Deletes DOF constraints on a line.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LINE
Line for which constraints are to be deleted. If ALL, delete for all selected lines [LSEL]. If LINE = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for LINE
Lab
Constraint label:
ALL
All constraints.
SYMM
Symmetry constraints.
ASYM
Antisymmetry constraints.
UX
Displacement in X direction.
UY
Displacement in Y direction.
UZ
Displacement in Z direction.
ROTX
Rotation about X axis.
393
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DLDELE
ROTY
Rotation about Y axis.
ROTZ
Rotation about Z axis.
WARP
Warping magnitude.
VX
Velocity component in X direction.
VY
Velocity component in Y direction.
VZ
Velocity component in Z direction.
PRES
Pressure.
TEMP, TBOT, TE2, TE3, . . ., TTOP
Temperature.
ENKE
Turbulent Kinetic Energy.
ENDS
Energy Dissipation Rate.
VOLT
Electric scalar potential.
SP01SP06
Multiple Species Mass Fraction.
AX
Magnetic vector potential in X direction.
AY
Magnetic vector potential in Y direction.
AZ
Magnetic vector potential in Z direction.
Notes
Deletes the degree of freedom constraints (and all corresponding finite element constraints) on a line previ-
ously specified with the DL command. See the DDELE command for delete details.
Warning
On previously meshed lines, all constraints on affected nodes will also be deleted, whether or
not they were specified by the DL command.
If the multiple species labels have been changed to user-defined labels via the MSSPEC command, use the
user-defined labels.
This command is also valid in PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
394
DLDELE
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Constraint>On All Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>PerfEC>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>Voltage>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Displacement>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Species>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Turbulence>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Velocity>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Displacement>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Temperature>On Lines
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Delete>On Lines
Main Menu>Solution>Constraints>Delete>On Lines
Main Menu>Solution>Define Loads>Delete>All Load Data>All Constraint>On All Lines
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>PerfEC>On Lines
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>Voltage>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Pressure DOF>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Displacement>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Pressure DOF>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Species>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Turbulence>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Velocity>On Lines
Main Menu>Solution>Define Loads>Delete>Magnetic>Boundary>VectorPot>On Lines
Main Menu>Solution>Define Loads>Delete>Structural>Displacement>On Lines
Main Menu>Solution>Define Loads>Delete>Thermal>Temperature>On Lines
DLIST, NODE1, NODE2, NINC
Lists DOF constraints.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
NODE1, NODE2, NINC
List constraints for nodes NODE1 to NODE2 (defaults to NODE1) in steps of NINC (defaults to 1). If ALL
(default), NODE2 and NINC are ignored and constraints for all selected nodes [NSEL] are listed. If NODE1
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may also be substituted for NODE1(NODE2 and NINC are ignored).
Notes
Listing applies to the selected nodes [NSEL] and the selected degree of freedom labels [DOFSEL].
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>DOF Constraints>On All Nodes
395
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DLIST
Utility Menu>List>Loads>DOF Constraints>On Picked Nodes
DLLIST, LINE
Lists DOF constraints on a line.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LINE
List constraints for this line. If ALL (default), list for all selected lines [LSEL]. If LINE = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for LINE.
Notes
Lists the degree of freedom constraints on a line previously specified with the DL command.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>DOF Constraints>On All Lines
Utility Menu>List>Loads>DOF Constraints>On Picked Lines
DMOVE, NODE1, NODE2, NINC
Digitizes nodes on surfaces and along intersections.
PREP7: Digitizing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1, NODE2, NINC
Digitize nodes NODE1through NODE2 in steps of NINC. NODE2 defaults to NODE1 and NINC defaults
to 1.
Notes
Digitizes nodes on undefined surfaces, warped surfaces, and along intersection lines. Two orthogonal views
showing the nodes on a plane in each view are required. No surfaces need be specified. Two coordinates
are determined from the second view and the other coordinate is retained from the first view. Use the DIG
command to first define nodes in one view (as determined from the DSET command). Then reset the view
and use this command to move the nodes to the proper location.
Menu Paths
Main Menu>Preprocessor>Create>Nodes>Digitize Nodes>2-View Digitize
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
396
DLLIST
DMPEXT, SMODE, TMODE, Dmpname, Freqb, Freqe, NSTEPS
Extracts modal damping coefficients in a specified frequency range.
SOLUTION: Analysis Options
MP ME <> <> <> <> <> <> <> <> <> PP <> EME MFS
SMODE
Source mode number. There is no default for this field; you must enter an integer greater than zero.
TMODE
Target mode. Defaults to SMODE.
Dmpname
Array parameter name containing the damping results. Defaults to d_damp.
Freqb
Beginning frequency range (real number greater than zero) or 'EIG' at eigenfrequency of source mode.
'EIG' is valid only if SMODE = TMODE. Note that EIG must be enclosed in single quotes when this command
is used on the command line or in an input file. There is no default for this field; you must enter a value.
Freqe
End of frequency range. Must be blank for Freqb = EIG. Default is Freqb.
NSTEPS
Number of substeps. Defaults to 1.
Notes
DMPEXT invokes an ANSYS macro that uses modal projection techniques to compute the damping force by
the modal velocity of the source mode onto the target mode. From the damping force, damping parameters
are extracted. DMPEXT creates an array parameter Dmpname, with the following entries in each row:
response frequency
modal damping coefficient
modal squeeze stiffness coefficient
damping ratio
squeeze-to-structural stiffness ratio
The macro requires the modal displacements from the file Jobname.EFL obtained from the RMFLVEC
command. In addition, a node component FLUN must exist from all FLUID136 nodes. The computed damping
ratio may be used to specify constant or modal damping by means of the DMPRAT or MDAMP commands.
For Rayleigh damping, use the ABEXTRACT command to compute ALPHAD and BETAD damping parameters.
See Thin Film Analysis for more information on thin film analyses.
The macro uses the LSSOLVE command to perform two load steps for each frequency. The first load case
contains the solution of the source mode excitation and can be used for further postprocessing. Solid
model boundary conditions are deleted from the model. In addition, prescribed nodal boundary conditions
are applied to the model. You should carefully check the boundary conditions of your model prior to executing
a subsequent analysis.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
397
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DMPEXT
Menu Paths
Main Menu>Solution>ThinFilm>DampExtract>Eigenfrequency
Main Menu>Solution>ThinFilm>DampExtract>Frequency Range
DMPRAT, RATIO
Sets a constant damping ratio.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
RATIO
Damping ratio (for example, 2% is input as 0.02).
Command Default
Use damping as defined in the Structural Analysis Guide.
Notes
Sets a constant damping ratio for use in the harmonic response (ANTYPE,HARMIC) analysis (full, reduced,
and modal superposition), the mode superposition transient (ANTYPE,TRANS) analysis, and the spectrum
(ANTYPE,SPECTR) analysis.
Note that for structures with multiple materials, MP,DMPR can be used to specify constant material damping
coefficients for full and modal harmonic analyses. MP,DMPR is not applicable for transient or spectrum
analyses.
In a full and reduced harmonic analysis, beta damping (BETAD) for this command is calculated via Equa-
tion 1523.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Damping
Main Menu>Solution>Load Step Opts>Time/Frequenc>Damping
DNSOL, NODE, Item, Comp, V1, V2, V3, V4, V5, V6
Defines or modifies solution results at a node.
POST1: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Node for which results are specified. If ALL, apply to all selected nodes [NSEL]. If NODE = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NODE.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
398
DMPRAT
Item
Label identifying results, see Table 149: DNSOL - Valid Item and Component Labels (p. 399). Items also require
a component label.
Comp
Component of the item. Valid component labels are shown Table 149: DNSOL - Valid Item and Component
Labels (p. 399) below.
V1, V2, V3, . . . , V6
Value assigned to result. If zero, a zero value will be assigned. If blank, the value remains unchanged.
Additional values (if any) assigned to the remaining components (in the order corresponding to the
Comp list shown below for the specified Item (starting from the specified Comp label and proceeding
to the right).
Notes
DNSOL can be used only with FULL graphics activated (/GRAPHICS,FULL); it will not work correctly with
PowerGraphics activated.
DNSOL defines or modifies solution results in the database at a node. For example,
DNSOL,35,U,X,.001,.002,.001 assigns values 0.001, 0.002, and 0.001 to UX, UY, and UZ (respectively) for node
35. All results that are changed in the database, including the nodal degree of freedom results, are available
for all subsequent operations. All data is stored in the solution coordinate system, but will be displayed in
the results coordinate system [RSYS]. Use the PRNSOL command to list the current results.
Data input by DNSOL is stored in temporary space and does not replace information in the database.
Therefore, data input by this command may be overwritten if a change is made to the selected set of nodes.
Issuing the DNSOL command or its GUI equivalent requires you to place the data type (stress/strain) in the
element nodal records. To get around this requirement, use the DESOL command or equivalent path to add
a "dummy" element stress/strain record.
Result items are available depending on element type; check the individual element for availability. Valid
item and component labels for element results are:
Table 149 DNSOL - Valid Item and Component Labels
Valid Item and Component Labels for Nodal DOF Results
Description Comp Item
X, Y, or Z structural displacement. X, Y, Z U
X, Y, or Z structural rotation. X, Y, Z ROT
Temperature. TEMP[1]
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity. X, Y, Z V
X, Y, or Z magnetic vector potential. X, Y, Z A
Turbulent kinetic energy. ENKE
Turbulent energy dissipation. ENDS
399
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DNSOL
Valid Item and Component Labels for Area and Volume Element Results
Description Comp Item
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity. INT "
Equivalent stress. EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity. INT "
Elastic equivalent strain. EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity. INT "
Thermal equivalent strain. EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity. INT "
Plastic equivalent strain. EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity. INT "
Creep equivalent strain. EQV "
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Elastic strain energy density. ELASTIC SEND
Plastic strain energy density. PLASTIC "
Creep strain energy density. CREEP "
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
Component pressure gradient or vector sum. X, Y, Z, SUM PG
Component electric field or vector sum. X, Y, Z, SUM EF
Component electric flux density or vector sum. X, Y, Z, SUM D
Component magnetic field intensity or vector sum. X, Y, Z, SUM H
Component magnetic flux density or vector sum. X, Y, Z, SUM B
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
400
DNSOL
Valid Item and Component Labels for Area and Volume Element Results
Description Comp Item
Component electromagnetic force or vector sum. X, Y, Z, SUM FMAG
Valid Item Labels for FLOTRAN Nodal Results
Description Item
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Turbulent viscosity coefficient. CMUV
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels TBOT, TE2, TE3, ..., TTOP
instead of TEMP.
Menu Paths
Main Menu>General Postproc>Define/Modify>Nodal Results
*DO, Par, IVAL, FVAL, INC
Defines the beginning of a do-loop.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
The name of the scalar parameter to be used as the loop index. See *SET for name restrictions. Any ex-
isting parameter of the same name will be redefined. There is no character parameter substitution for
the Par field.
401
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*DO
IVAL, FVAL, INC
Initially assign IVAL to Par. Increment IVAL by INC for each successive loop. If IVAL exceeds FVAL
and INC is positive, the loop is not executed. INC defaults to 1. Negative increments and non-integer
numbers are allowed.
Notes
The block of commands following the *DO command (up to the *ENDDO command) is executed repeatedly
until some loop control is satisfied. Printout is automatically suppressed on all loops after the first (include
a /GOPR command to restore the printout). The command line loop control (Par,IVAL,FVAL,INC) must
be input; however, a Use the *IF within the block can also be used to control looping [*EXIT, *CYCLE]. One
level of internal file switching is used for each nested *DO. Twenty levels of nested do-loops are allowed.
Note
Do-loops that include /INPUT, *USE, or an "Unknown Command" macro, have less nesting available
because each of these operations also uses a level of file switching. The *DO, *ENDDO, and any
*CYCLE and *EXIT commands for a do-loop must all be read from the same file (or keyboard).
You cannot use the MULTIPRO or *CREATE commands within a *DO-loop. Picking operations
should also not be used within a *DO-loop.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
DOF, Lab1, Lab2, Lab3, Lab4, Lab5, Lab6, Lab7, Lab8, Lab9, Lab10
Adds degrees of freedom to the current DOF set.
PREP7: Element Type
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab1, Lab2, Lab3, . . . , Lab10
Valid labels are: UX, UY, UZ (structural displacements); ROTX, ROTY, ROTZ (structural rotations); TEMP,
TBOT, TE2, TE3, . . ., TTOP (temperatures); PRES (pressure); VOLT (voltage); MAG (magnetic scalar potential);
AX, AY, AZ (magnetic vector potentials); CURR (current); EMF (electromotive force drop); DELETE.
Command Default
Use degree of freedom set determined from element types.
Notes
The degree of freedom (DOF) set for the model is determined from all element types defined. For example,
if only LINK1 is defined, the set is UX,UY. If LINK1 and BEAM3 are defined, the set is UX,UY,ROTZ. This command
may be used to add to the current set. The ALL label may be used on some commands to represent all labels
of the current degree of freedom set for the model. Issue the DOF command with no arguments to list the
current set. Use the DELETE label to delete any previously added DOFs and return to the default DOF set.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
402
DOF
Product Restrictions
Available Products Command Option
Lab
MP ME ST PR PRN <> <> FL <> <> DY PP <> EME MFS Ux, Uy,Uz, Rx,Ry,Rz
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <> MAG, VOLT
MP <> <> <> <> <> <> <> EM EH <> PP <> EME <> Ax,Ay,Az
MP ME <> <> <> <> <> <> <> <> <> PP <> EME MFS PRES
MP ME <> PR PRN <> <> FL <> <> <> PP <> EME MFS TEMP,TBOT, TE2,TE3
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS CURR,EMF
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS DELETE
Menu Paths
Main Menu>Preprocessor>Element Type>Add DOF
Main Menu>Preprocessor>Element Type>Remove DOFs
DOFSEL, Type, Dof1, Dof2, Dof3, Dof4, Dof5, Dof6
Selects a DOF label set for reference by other commands.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Select a new set of labels.
A
Add labels to the current set.
U
Unselect (remove) labels from the current set.
ALL
Restore the full set of labels.
STAT
Display the current select status.
Dof1, Dof2, Dof3, . . . , Dof6
Used only with Type = S, A, or U. Valid structural labels: UX, UY, or UZ (displacements); U (UX, UY, and
UZ); ROTX, ROTY, or ROTZ (rotations); ROT (ROTX, ROTY, and ROTZ); DISP (U and ROT); HDSP (Hydrostatic
pressure). Valid thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP (temperature). Valid fluid flow labels:
PRES (pressure); VX, VY, or VZ (fluid velocities); V (VX, VY, and VZ); ENKE, ENDS (turbulent kinetic energy,
turbulent energy dissipation); EN (ENKE and ENDS turbulent energies) (FLOTRAN). Valid electric labels:
VOLT (voltage); EMF (electromotive force drop); CURR (current). Valid magnetic labels: MAG (scalar
magnetic potential); AX, AY or AZ (vector magnetic potentials); A (AX, AY and AZ); CURR (current). Valid
structural force labels: FX, FY, or FZ (forces); F (FX, FY, and FZ); MX, MY, or MZ (moments); M (MX, MY,
403
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DOFSEL
and MZ); FORC (F and M). Valid thermal force labels: HEAT, HBOT, HE2, HE3, . . ., HTOP (heat flow). Valid
fluid flow force labels: FLOW (fluid flow). Valid electric force labels: AMPS (current flow), CHRG (electric
charge). Valid magnetic force labels: FLUX (scalar magnetic flux); CSGX, CSGY, or CSGZ (magnetic current
segments); CSG (CSGX, CSGY, and CSGZ).
Command Default
Degree of freedom (and the corresponding force) labels are determined from the model.
Notes
Selects a degree of freedom label set for reference by other commands. The label set is used on certain
commands where ALL is either input in the degree of freedom label field or implied. The active label set
has no effect on the solution degrees of freedom. Specified labels which are not active in the model (from
the ET or DOF command) are ignored. As a convenience, a set of force labels corresponding to the degree
of freedom labels is also selected. For example, selecting UX also causes FX to be selected (and vice versa).
The force label set is used on certain commands where ALL is input in the force label field.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Scale FE Loads>Constraints
Main Menu>Preprocessor>Loads>Define Loads>Operate>Scale FE Loads>Forces
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Constraints
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Forces
Main Menu>Solution>Define Loads>Operate>Scale FE Loads>Constraints
Main Menu>Solution>Define Loads>Operate>Scale FE Loads>Forces
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Constraints
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Forces
DOMEGA, DOMGX, DOMGY, DOMGZ
Specifies the rotational acceleration of the structure.
SOLUTION: Inertia
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
DOMGX, DOMGY, DOMGZ
Rotational acceleration of the structure about the global Cartesian X , Y, and Z axes.
Notes
Specifies the rotational acceleration of the structure about each of the global Cartesian axes. Rotational ac-
celerations may be defined in analysis types ANTYPE,STATIC, HARMIC (full or mode superposition), TRANS
(full or mode superposition), and SUBSTR. See Acceleration Effect in the Theory Reference for the Mechanical
APDL and Mechanical Applications for details. Units are radians/time
2
.
The DOMEGA command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for DOMEGA_X, DOMEGA_Y, and DOMEGA_Z input values (*DIM) for full transient and
harmonic analyses.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
404
DOMEGA
Related commands are ACEL, CGLOC, CGOMGA, DCGOMG, and OMEGA.
In a modal harmonic or transient analysis, you must apply the load in the modal portion of the analysis.
ANSYS calculates a load vector and writes it to the mode shape file, which you can apply via the LVSCALE
command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Accel>Global
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Angular Accel>Global
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Accel>Global
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Angular Accel>Global
*DOWHILE, Par
Loops repeatedly through the next *ENDDO command.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
The name of the scalar parameter to be used as the loop index. There is no character parameter substi-
tution for the Par field.
Notes
*DOWHILE loops repeatedly through the next *ENDDO command as long as Par is greater than zero. The
block of commands following the *DOWHILE command (up to the *ENDDO command) is executed repeatedly
until some loop control is satisfied. Printout is automatically suppressed on all loops after the first (include
a /GOPR command to restore the printout). The command line loop control (Par) must be input; however,
*IF within the block can also be used to control looping [*EXIT, *CYCLE]. One level of internal file switching
is used for each nested *DOWHILE. Twenty levels of nested do-loops are allowed.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
405
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*DOWHILE
DSCALE, RFACT, IFACT, TBASE
Scales DOF constraint values.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
RFACT
Scale factor for the real component. Zero (or blank) defaults to 1.0. Use a small number for a zero scale
factor.
IFACT
Scale factor for the imaginary component. Zero (or blank) defaults to 1.0. Use a small number for a zero
scale factor.
TBASE
Base temperature for temperature difference. For temperatures, the scale factor is applied to the temper-
ature difference (T-TBASE) and then added to TBASE. T is the current temperature.
Notes
Scales degree of freedom constraint values (displacement, temperature, etc.) in the database. If velocity and
acceleration boundary conditions are applied in a structural analysis, they are also scaled by this command.
Solid model boundary conditions are not scaled by this command, but boundary conditions on the FE
model are scaled.
Note
Such scaled FE boundary conditions may still be overwritten by unscaled solid model boundary
conditions if a subsequent boundary condition transfer occurs.
Scaling applies to the previously defined values for the selected nodes [NSEL] and the selected degree of
freedom labels [DOFSEL]. Issue DLIST command to review results.
DSCALE does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Scale FE Loads>Constraints
Main Menu>Solution>Define Loads>Operate>Scale FE Loads>Constraints
/DSCALE, WN, DMULT
Sets the displacement multiplier for displacement displays.
GRAPHICS: Scaling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
406
DSCALE
DMULT
AUTO or 0
Scale displacements automatically so that maximum displacement (vector amplitude) displays as 5
percent of the maximum model length, as measured in the global Cartesian X, Y, or Z directions.
1
Do not scale displacements (i.e., scale displacements by 1.0, true to geometry). Often used with large
deflection results.
FACTOR
Scale displacements by numerical value input for FACTOR.
OFF
Remove displacement scaling (i.e., scale displacements by 0.0, no distortion).
USER
Set DMULT to that used for last display (useful when last DMULT value was automatically calculated).
Command Default
The default value is AUTO or 0 except when:
Large deflection effects are included (NLGEOM,ON); then the default is 1.
It is a spectrum analysis (ANTYPE,SPECTR); then the default is OFF.
The amplitude of a time-harmonic solution is computed using the HRCPLX command (OMEGAT 360);
then the default is OFF.
The amplitude of a complex modal or harmonic solution is stored into the database using the SET
command (KIMG = AMPL); then the default is OFF.
Notes
If Multi-Plots are not being displayed, and the current device is a 3-D device [/SHOW,3D], then the displace-
ment scale in all active windows will be the same, even if separate /DSCALE commands are issued for each
active window. For efficiency, ANSYS 3-D graphics logic maintains a single data structure (segment), which
contains only one displacement scale. The program displays the same segment (displacement scale) in all
windows. Only the view settings will be different in each of the active windows.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Displacement Scaling
DSET, NODE1, NODE2, NODE3, DDEV
Sets the scale and drawing plane orientation for a digitizing tablet.
PREP7: Digitizing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1, NODE2, NODE3
Any three (noncolinear) nodes defining a plane parallel to the drawing. Nodes and actual locations (in
any coordinate system) must have been previously defined.
407
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DSET
DDEV
Digitizing device type number (device dependent).
Notes
Sets drawing scale size and defines the drawing plane orientation for use with a digitizing tablet. Drawings
must be to scale. Views must represent standard orthogonal parallel projections. The three nodes indicated
must be digitized [DIG] from the tablet after this command is issued.
Menu Paths
Main Menu>Preprocessor>Create>Nodes>Digitize Nodes>Set Plane/Device
DSPOPTION, Reord_Option, Memory_Option, Memory_Size, --, --, Solve_Info
Sets memory option for the distributed sparse solver.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Reord_Option
Reordering option:
DEFAULT
Use the default reordering scheme.
PARORDER
Use a distributed memory parallel equation reordering scheme within the DSPARSE solver. This option
can often help the DSPARSE solver achieve greater scalability (typically, at 16 or more cores) by doing
a key solver step in parallel instead of sequentially. However, the parallel reordering results can
sometimes be worse than the default reordering scheme, thus having an adverse effect on the fac-
torization times and resulting in worse overall performance. It is best to only use this option when
trying to maximize performance.
Memory_Option
Memory allocation option:
DEFAULT
Use the default memory allocation strategy for the distributed sparse solver. The default strategy
attempts to run in the INCORE memory mode. If there is not enough physical memory available
when the solver starts to run in the INCORE memory mode, the solver will then attempt to run in
the OPTIMAL memory mode.
INCORE
Use a memory allocation strategy in the distributed sparse solver that will attempt to obtain enough
memory to run with the entire factorized matrix in memory. This option uses the most amount of
memory and should avoid doing any I/O. By avoiding I/O, this option often achieves the best per-
formance of all the memory modes. However, a significant amount of memory is required to run in
this mode, and it is only recommended on machines with a large amount of memory. If the allocation
for in-core memory fails, ANSYS will automatically revert to OPTIMAL out-of-core mode.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
408
DSPOPTION
OPTIMAL
Use a memory allocation strategy in the distributed sparse solver that will attempt to allocate enough
work space to keep the largest front matrix in memory during factorization. As a result, this memory
mode often achieves an optimal balance between I/O and memory usage.
FORCE
This option, when used in conjunction with the Memory_Size option, allows you to force the dis-
tributed sparse solver to run with a specific amount of memory. This option is only recommended
for the advanced user who understands distributed sparse solver memory requirements for the
problem being solved, understands the physical memory on the system, and wants to control the
distributed sparse solver memory usage.
Memory_Size
Initial memory size allocation for the sparse solver in MB. The Memory_Size setting should always be
well within the physical memory available, but not so small as to cause the distributed sparse solver to
run out of memory. Warnings and/or errors from the distributed sparse solver will appear if this value
is set too low. If the FORCE memory option is used, this value is the amount of memory allocated for
the entire duration of the distributed sparse solver solution.
--, --
Unused fields
Solve_Info
Solver output option:
OFF
Turns off additional output printing from the distributed sparse solver (default).
PERFORMANCE
Turns on additional output printing from the distributed sparse solver, including a performance
summary and a summary of file I/O for the distributed sparse solver. Information on memory usage
during assembly of the global matrix (that is, creation of the Jobname.FULL file) is also printed
with this option.
Command Default
Automatic memory allocation is used.
Notes
This command controls options related to the distributed sparse solver in all analysis types where the dis-
tributed sparse solver can be used.
The amount of memory required for the distributed sparse solver is unknown until the matrix structure is
preprocessed, including equation reordering. The amount allocated for the distributed sparse solver is then
dynamically adjusted using the ANSYS memory manager.
If you have a large memory system, you may want to try selecting the INCORE memory mode for larger jobs
to improve performance. Also, when running the distributed sparse solver with many processors on the
same machine or on a machine with very slow I/O performance (e.g., slow hard drive speed), you may want
to try using the INCORE memory mode to achieve better performance. However, doing so may require much
more memory compared to running in the OPTIMAL memory mode.
Running with the INCORE memory mode is best for jobs which comfortably fit within the limits of the
physical memory on a given system. If the distributed sparse solver workspace exceeds physical memory
409
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DSPOPTION
size, the system will be forced to use virtual memory (or the system page/swap file). In this case, it is typically
more efficient to run with the OPTIMAL memory mode (DSPOPTION,,OPTIMAL).
Menu Paths
This command cannot be accessed from a menu.
DSUM, SIGNIF, Label, TD
Specifies the double sum mode combination method.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SIGNIF
Combine only those modes whose significance level exceeds the SIGNIF threshold. For single point,
multipoint, or DDAM response (SPOPT, SPRS, MPRS, or DDAM), the significance level of a mode is defined
as the mode coefficient of the mode, divided by the maximum mode coefficient of all modes. Any mode
whose significance level is less than SIGNIF is considered insignificant and is not contributed to the
mode combinations. The higher the SIGNIF threshold, the fewer the number of modes combined.
SIGNIF defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0. (This mode combination
method is not valid for SPOPT, PSD.)
Label
Label identifying the combined mode solution output.
DISP
Displacement solution (default). Displacements, stresses, forces, etc., are available.
VELO
Velocity solution. Velocities, "stress velocities," "force velocities," etc., are available.
ACEL
Acceleration solution. Accelerations, "stress accelerations," "force accelerations," etc., are available.
TD
Time duration for earthquake or shock spectrum. TD defaults to 10.
Notes
This command is also valid for PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Mode Combine
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
410
DSUM
DSURF, KCN, XSURF, YSURF, ZSURF
Defines the surface upon which digitized nodes lie.
PREP7: Digitizing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Surface is located in coordinate system KCN. KCN may be 0,1,2 or any previously defined local coordinate
system number.
XSURF, YSURF, ZSURF
Input one value to define the surface constant. Input 999 in the other two fields. Interpret fields as R, ,
Z for cylindrical or R, , for spherical or toroidal coordinate systems. XSURF and YSURF default to 999
if KCN = 0.
Command Default
Surface associated with DIG command is the global Cartesian X-Y plane with Z = 0.
Notes
Defines the surface upon which the nodes to be digitized (with the DIG command) actually lie. Surfaces are
defined by a coordinate system number and a coordinate constant [MOVE]. Two coordinates are determined
from the drawing and converted to surface coordinates. The third coordinate is defined from the input surface
constant. If nodes lie on warped or undefined surfaces, use the DMOVE command.
Menu Paths
Main Menu>Preprocessor>Create>Nodes>Digitize Nodes>Define Surface
DSYM, Lab, Normal, KCN
Specifies symmetry or antisymmetry degree-of-freedom constraints on nodes.
SOLUTION: FE Constraints
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lab
Symmetry label:
SYMM
Generate symmetry constraints as described below (default).
ASYM
Generate antisymmetry constraints as described below.
Normal
Surface orientation label to determine the constraint set (surface is assumed to be perpendicular to this
coordinate direction in coordinate system KCN):
X
Surface is normal to coordinate X direction (default). Interpreted as R direction for non-Cartesian
coordinate systems.
411
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DSYM
Y
Surface is normal to coordinate Y direction. direction for non-Cartesian coordinate systems.
Z
Surface is normal to coordinate Z direction. direction for spherical or toroidal coordinate systems.
KCN
Reference number of global or local coordinate system used to define surface orientation.
Notes
Specifies symmetry or antisymmetry degree-of-freedom constraints on the selected nodes. The nodes are
first automatically rotated (any previously defined rotations on these nodes are redefined) into coordinate
system KCN, then zero-valued constraints are generated, as described below, on the selected degree-of-
freedom set (limited to displacement, velocity, and magnetic degrees of freedom) [DOFSEL]. Constraints are
defined in the (rotated) nodal coordinate system, as usual. See the D and NROTAT commands for additional
details about constraints and nodal rotations.
This command is also valid in PREP7.
Symmetry and Antisymmetry Constraints:
Symmetry or antisymmetry constraint generations are based upon the valid degrees of freedom in the
model, i.e., the degrees of freedom associated with the elements attached to the nodes. The labels for degrees
of freedom used in the generation depend on the Normal label.
For displacement degrees of freedom, the constraints generated are:
ASYM SYMM
3-D 2-D 3-D 2-D Normal
UY, UZ, ROTX UY UX, ROTZ,
ROTY
UX, ROTZ X
UX, UZ, ROTY UX UY, ROTZ,
ROTX
UY, ROTZ Y
UX, UY, ROTZ -- UZ, ROTX,
ROTY
-- Z
For velocity degrees of freedom, the constraints generated are:
ASYM SYMM
3-D 2-D 3-D 2-D Normal
VY, VZ VY VX VX X
VX, VZ VX VY VY Y
VX, VY -- VZ -- Z
For magnetic degrees of freedom, the SYMM label generates flux normal conditions (flux flows normal to
the surface). Where no constraints are generated, the flux normal condition is "naturally" satisfied. The ASYM
label generates flux parallel conditions (flux flows parallel to the surface).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
412
DSYM
ASYM SYMM
3-D 2-D 3-D 2-D Normal
AY, AZ AZ AX -- X
AX, AZ AZ AY -- Y
AX, AY -- AZ -- Z
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Nor-
mal>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux
Par'l>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>On
Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>On
Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Normal>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Flux Par'l>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Antisymm B.C.>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Displacement>Symmetry B.C.>On Nodes
DSYS, KCN
Activates a display coordinate system for geometry listings and plots.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Coordinate system reference number. KCN may be 0,1,2 or any previously defined local coordinate system
number.
Note
If a cylinder is displayed in its cylindrical coordinate system (with a 1,0,0 view), it will be un-
rolled (developed) into a flat plane (with theta along the Y direction).
Command Default
Global Cartesian (KCN = 0) display coordinate system.
Notes
Boundary condition symbols, vector arrows, and element coordinate system triads are not transformed to
the display coordinate system. The display system orientation (for the default view) is X horizontal to the
right, Y vertical upward, and Z out of the screen (normal).
Line directions and area directions (/PSYMB,LDIR and /PSYMB,ADIR) are not plotted for DSYS >0.
413
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DSYS
When you create ANSYS 3-D annotation, the coordinates are stored to the database in the DSYS that was
active at the time of creation. Changing the DSYS does not change the annotation coordinate data in the
database.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Change Display CS to>Global Cartesian
Utility Menu>WorkPlane>Change Display CS to>Global Cylindrical
Utility Menu>WorkPlane>Change Display CS to>Global Spherical
Utility Menu>WorkPlane>Change Display CS to>Specified Coord Sys
DTRAN
Transfers solid model DOF constraints to the finite element model.
SOLUTION: Solid Constraints
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Notes
Constraints are transferred only from selected solid model entities to selected nodes. The DTRAN operation
is also done if the SBCTRAN command is issued, and is automatically done upon initiation of the solution
calculations [SOLVE].
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Transfer to FE>Constraints
Main Menu>Solution>Define Loads>Operate>Transfer to FE>Constraints
DUMP, NSTRT, NSTOP
Dumps the contents of a binary file.
AUX2: Binary Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSTRT, NSTOP
Dump file from record NSTRT (defaults to 1) to NSTOP (defaults to NSTRT). If NSTRT = HEAD, dump
only record 1 of the file (NSTOP and the format specification are ignored). If NSTRT = ALL, dump the
entire file.
Notes
Dumps the file named on the AUX2 FILEAUX2 command according the format specified on the FORM
command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
414
DTRAN
Menu Paths
Utility Menu>File>List>Binary Files
Utility Menu>List>Files>Binary Files
/DV3D, Lab, Key
Sets 3-D device option modes.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Mode label:
ACCU
Allows ANSYS to use the accumulation buffer for OpenGL graphics. Activating this feature will provide
faster model rotation when shaded backgrounds are in use. This feature is off by default.
ACTR
Label term to designate the cursor position as the center for automatic dynamic rotational center
capability . The subsequent Key value (see below) turns this capability on and off. This feature is on
by default. (Available for OpenGL displays only)
ANIM
Animation mode. The ANIM option allows you to create animation frames in pixmap mode instead
of display list mode. This may improve large model performance, but it eliminates local manipulation
while animation is in progress. This feature is on by default.
ANTI
Label term to control Anti-aliasing, a smoothing technique for your graph plots. (see below) The
subsequent Key value turns this capability on and off. The default for this feature is off. (Available
for OpenGL displays only).
CNTR
Switches banded contours on (1) or off (0) for your 3D contour display. The default is 1 (ON). Other
contour parameters such as number of contours or the increment and range are defined using the
/CONTOUR command. When either 9 or 128 contours are specified via /CONTOUR, this command
is ignored and a smooth contour is always displayed.
DGEN
Local manipulation degenerate mode. You access the DGEN option to set wire-frame local manipu-
lation mode for 3-D devices (device dependent). This feature is off by default.
DLIST
With DLIST, you can specify whether screen updates and redraws will be performed using the ANSYS
Display List (off ), or the 3-D device's Display List (on). DLIST is on by default for Windows systems,
but off for UNIX. .
DELS
You use DELS to suppress contour display screen overwrites when /NOERASE is active. This prevents
the bleed-through that occurs when you overlay contour plots.
TRIS
Triangle strip mode. Tri-stripping provides faster 3-D display capabilities and is on by default. Some
display enhancements, such as texturing, are adversely affected by tri-stripping. You can turn off tri-
415
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/DV3D
stripping in order to improve these display functions. Be sure to turn tri-stripping on after the desired
output is obtained.
Key
The following key options apply to Lab = ACCU:
0
(OFF) The accumulation buffer is not accessed. (default)
1
(ON) Access to the buffer is enabled.
The following key options apply to Lab = ACTR:
0
(OFF) The cursor position has no effect on the existing rotational center for dynamic operations.
1
(ON) The rotational center for dynamic rotations in OpenGL is determined by the position of the
mouse cursor on (or within 15 pixels of ) the model. Any rotations that are initiated with the cursor
more than 15 pixels from the model will occur about the midpoint of the Z-axis at that point in
space. If the Z-buffer has not been refreshed the Z-axis will have an infinite value, and rotations will
appear to occur about an extremely long Z-axis. This behavior stops when the graphics window is
refreshed or replotted. (default)
Note that when using the GUI in 3-D mode, when ACTR = 1, the Rotational Center option is grayed
out under Utility Menu> PlotCtrls> View Setting because the rotational center is determined strictly
by the position of the mouse cursor.
The following key options apply to Lab = ANIM:
0
Display list animation. The object can be dynamically manipulated while animating. No legend,
countour or annotation items are displayed. (see Notes, below)
1
On UNIX, device-dependent pixmap animation is used. On the PC, bitmap animation is provided
(default). When you animate in this mode, you cannot dynamically manipulate your model (see
Notes, below)..
2
On the PC only, this option provides AVI animation which uses the AVI movie player .
Although you can create animations of multiple ANSYS window schemes, animations created with
OpenGL display lists (/DV3D, ANIM, 0) do not retain the windowing scheme information. You CAN
save multiple windows via the X11/WIN32 drivers, or via the OpenGL driver with /DV3D, ANIM, KEY
in effect (where KEY is not zero).
The following key options apply to Lab = ANTI
0
(OFF) Anti-aliasing is not active (default).
1
(ON) The anti-aliasing technique will be applied to smooth the lines in your displays (valid for OpenGL
only).
The following key options apply to Lab = DGEN:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
416
/DV3D
0
Normal manipulation.
1
Wireframe Manipulation.
The following key options apply to Lab = DLIST:
0
(OFF) The ANSYS Display List is used for plotting and dynamic graphics manipulation (UNIX default).
1
(ON) The local (3-D Device) Display List is used for plotting and dynamic rotation (Windows default).
The following key options apply to Lab = TRIS:
0
(OFF) Tri-stripping is off.
1
(ON) Tri-stripping is on (default).
This command is valid in any processor.
Notes
ANSYS uses display list animation for its 3-D models. This memory resident array method interfaces with the
OpenGL model information to allow the program to efficiently pan, zoom, rotate and dynamically manipulate
your model during animation. The logo, legend, contour and other annotation items are produced in 2-D
and will not appear when /DV3D, ANIM, 0 is in effect. To display these items, use /DV3D, ANIM, 1. All screen
data will be displayed, but manipulation of the model will not be possible.
Menu Paths
Utility Menu>PlotCtrls>Device Options
DVMORPH, VOLU, XAREA, RMSHKY
Move nodes in selected volumes to conform to structural displacements.
PREP7: Morphing
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
VOLU
Non-structural volume to which mesh movement (morph) applies. If ALL, apply morphing to all selected
volumes [VSEL]. If VOLU = P, graphical picking is enabled. A component may be substituted for VOLU.
XAREA
Areas to be excluded from morphing. If ALL, exclude all selected areas [ASEL]. If XAREA = P, graphical
picking is enabled. A component may be substituted for XAREA. If XAREA is blank (default), allow
morphing of nodes attached to areas of the selected volumes (VOLU) which are not shared by unselected
volumes. (See Notes for clarification).
RMSHKY
Remesh flag option:
417
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DVMORPH
0
Remesh the selected non-structural volumes only if mesh morphing fails.
1
Remesh the selected non-structural volumes and bypass mesh morphing.
2
Perform mesh morphing only and do not remesh.
Notes
The selected volumes should include only non-structural regions adjacent to structural regions. DVMORPH
will morph the non-structural volumes to coincide with the deflections of the structural regions.
Nodes in the structural regions move in accordance with computed displacements. Displacements from a
structural analysis must be in the database prior to issuing DVMORPH.
By default, nodes attached to areas can move along the areas. You can use XAREA to restrain nodes on
certain areas.
By default (RMSHKY = 0), DVMORPH will remesh the selected non-structural volumes entirely if a satisfactory
morphed mesh cannot be provided.
If boundary conditions and loads are applied directly to nodes and elements, the DVMORPH command requires
that these be removed before remeshing can take place.
Exercise care with initial conditions defined by the IC command. Before a structural analysis is performed
for a sequentially coupled analysis, the DVMORPH command requires that initial conditions be removed
from all null element type nodes in the non-structural regions. Use ICDELE to delete the initial conditions.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Phys Morphing>Volumes
DYNOPT
Specifies "Dynamic analysis options" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
418
DYNOPT
Menu Paths
Utility Menu>List>Status>Solution>Dynamics Options
419
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
DYNOPT
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
420
E Commands
E, I, J, K, L, M, N, O, P
Defines an element by node connectivity.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
I
Number of node assigned to first nodal position (node I). If I = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI).
J, K, L, M, N, O, P
Number assigned to second (node J) through eighth (node P) nodal position, if any.
Notes
Defines an element by its nodes and attribute values. Up to 8 nodes may be specified with the E command.
If more nodes are needed for the element, use the EMORE command. The number of nodes required and
the order in which they should be specified are described in Chapter 4 of the Element Reference for each
element type. Elements are automatically assigned a number [NUMSTR] as generated. The current (or default)
MAT, TYPE, REAL, SECNUM and ESYS attribute values are also assigned to the element.
When creating elements with more than 8 nodes using this command and the EMORE command, it may
be necessary to turn off shape checking using the SHPP command before issuing this command. If a valid
element type can be created without using the additional nodes on the EMORE command, this command
will create that element. The EMORE command will then modify the element to include the additional nodes.
If shape checking is active, it will be performed before the EMORE command is issued. Therefore, if the
shape checking limits are exceeded, element creation may fail before the EMORE command modifies the
element into an acceptable shape.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Auto Numbered>Thru Nodes
EALIVE, ELEM
Reactivates an element (for the birth and death capability).
SOLUTION: Birth and Death
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
ELEM
Element to be reactivated:
ALL
Reactivates all selected elements (ESEL).
421
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
P
Enables graphical picking of elements. All remaining command fields are ignored. (Valid only in the
ANSYS GUI.)
Comp
Specifies a component name.
Notes
Reactivates the specified element when the birth and death capability is being used. An element can be
reactivated only after it has been deactivated (EKILL).
Reactivated elements have a zero strain (or thermal heat storage, etc.) state.
ANSYS recommends using the element deactivation/reactivation procedure for analyses involving linear
elastic materials only. Do not use element deactivation/reactivation in analyses involving time-dependent
materials, such as viscoelasticity, viscoplasticity, and creep analysis.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Birth & Death>Activate Elem
Main Menu>Solution>Load Step Opts>Other>Birth & Death>Activate Elem
EDADAPT, PART, Key
Activates adaptive meshing in an explicit dynamic analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
PART
Part ID (number) for which adaptive meshing is to be turned on (or off ). Use PART = STAT to list the
current adaptive meshing definitions.
Key
Adaptivity key:
OFF
Do not use adaptive meshing for the specified part ID (default).
ON
Use adaptive meshing for the specified part ID.
Command Default
Adaptive meshing is off for all parts in the model.
Notes
When adaptive meshing (adaptivity) is turned on, the mesh will automatically be regenerated to ensure
adequate element aspect ratios. Adaptive meshing is most commonly used in the analysis of large deform-
ation processes such as metal forming, in which the blank would need to be adaptively meshed.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
422
EDADAPT
Adaptive meshing is only valid for parts consisting of SHELL163 elements. By default, adaptive meshing is
OFF for all parts in the model. To specify adaptive meshing for more than one part in the model, you must
issue the EDADAPT command for each part ID. Use the EDPART command to create and list valid part IDs.
Use the EDCADAPT command to define additional adaptive meshing parameters.
The EDADAPT command is not supported in an explicit dynamic full restart analysis (EDSTART,3). In addition,
a full restart cannot be performed successfully if adaptive meshing was used in the previous analysis.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>Adaptive Meshing>Apply to Part
Main Menu>Solution>Analysis Options>Adaptive Meshing>Status
EDALE, Option, --, AFAC, BFAC, --, DFAC, EFAC, START, END
Assigns mesh smoothing to explicit dynamic elements that use the ALE formulation.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the option to be performed:
ADD
Add smoothing controls (default).
DELETE
Delete smoothing controls.
LIST
List smoothing controls.
--
Unused field.
AFAC
Simple average smoothing weight factor (default = 0).
BFAC
Volume weighted smoothing weight factor (default = 0).
--
Unused field.
DFAC
Equipotential smoothing weight factor (default = 0).
EFAC
Equilibrium smoothing weight factor (default = 0). EFAC is only applicable to PLANE162 elements.
START
Start time for ALE smoothing (default = 0).
423
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDALE
END
End time for ALE smoothing (default = 1e20).
Command Default
The Lagrangian formulation is used for all elements by default.
Notes
Mesh smoothing specified by the EDALE command is only applicable to PLANE162 and SOLID164 elements
that are flagged to use the ALE formulation (KEYPOPT(5) = 1). To activate the ALE formulation, you must
specify at least one smoothing weight factor on this command and the number of cycles between advection
(NADV) on the EDGCALE command. See Arbitrary Lagrangian-Eulerian Formulation in the ANSYS LS-DYNA
User's Guide for more information.
The EDALE command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>ALE Options>Define
Main Menu>Solution>Analysis Options>ALE Options>Delete All
Main Menu>Solution>Analysis Options>ALE Options>List All
EDASMP, Option, ASMID, PART1, PART2, PART3, PART4, PART5, PART6, PART7, PART8, PART9, PART10, PART11,
PART12, PART13, PART14, PART15, PART16
Creates a part assembly to be used in an explicit dynamic analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the part assembly option to be performed.
ADD
Adds a part assembly (default).
DELETE
Deletes a part assembly.
LIST
Lists each part assembly number, and the part numbers that make up each part assembly.
ASMID
User defined part assembly ID number. The part assembly number cannot be the same as any currently
defined part ID number.
PART1, PART2, PART3, . . . , PART16
Part numbers to be included in the assembly (up to 16 different parts).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
424
EDASMP
Command Default
Default for Option is ADD. You must specify ASMID to avoid an error message.
Notes
Several ANSYS LS-DYNA commands (such as EDCGEN, EDPVEL, and EDIS) refer to assembly ID numbers. If
you intend to use assembly ID numbers with these commands, you must first define the assembly ID numbers
using EDASMP.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Assembly Options
EDBOUND, Option, Lab, Cname, XC, YC, ZC, Cname2, COPT
Defines a boundary plane for sliding or cyclic symmetry.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the symmetry plane option to be performed.
ADD
Define a sliding or cyclic symmetry plane.
DELE
Delete a specified sliding or cyclic symmetry plane.
LIST
List defined sliding or cyclic symmetry planes.
Lab
Valid boundary options for defining a symmetry plane. A valid label must always be specified for adding,
deleting, or listing boundary planes.
SLIDE
Sliding symmetry plane.
CYCL
Cyclic symmetry plane.
Cname
Name of existing component [CM] to which boundary symmetry is to be applied or deleted. Component
must consist of nodes. For Option = LIST, a component is not required because all defined symmetry
planes are listed for the specified Lab. For Option = DELE, use Cname = ALL to delete all symmetry
planes currently defined for the specified Lab.
XC, YC, ZC
X, Y, and Z coordinates of the head of the vector defining normal (Lab = SLIDE) or axis of rotation (Lab
= CYCL). The tail of the vector is at the global origin.
425
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDBOUND
Cname2
Name of existing nodal component [CM] for which second cyclic boundary plane is to be applied. Each
node in Cname2 component is constrained to a corresponding node in the first component set. Therefore,
component Cname2 must have the same number of nodes as the Cname component. Cname2 is valid
only for Lab = CYCL.
COPT
Specified constraint option for sliding plane symmetry. COPT is valid only for Lab = SLIDE. Valid COPT
options are:
0
Nodes move on normal plane (default).
1
Nodes move only in vector direction.
Notes
For cyclic symmetry, the node numbers in component Cname2 must differ from the node numbers in Cname
by a constant offset value. In addition, the nodes in Cname2 must have locations which, if given in cylindrical
coordinates, all differ by the same angle from the nodes in Cname. The following figure shows how you
would define components for a cyclic symmetry plane.
Side 1
Segment
S
i
d
e
2
Cname - nodes along side 1
Cname2 - nodes along side 2
Conformable Interface
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>Symm Bndry Plane
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Delete>Symm Bndry Plane>Delete All
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Delete>Symm Bndry Plane>Delete Individ
Main Menu>Solution>Constraints>Apply>Symm Bndry Plane
Main Menu>Solution>Constraints>Delete>Symm Bndry Plane>Delete All
Main Menu>Solution>Constraints>Delete>Symm Bndry Plane>Delete Individ
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
426
EDBOUND
EDBX, Option, BOXID, XMIN, XMAX, YMIN, YMAX, ZMIN, ZMAX
Creates a box shaped volume to be used in a contact definition for explicit dynamics.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the contact box definition option to be performed.
ADD
Adds a contact box definition (default).
DELETE
Deletes a contact box definition.
LIST
Lists each box ID number, and the coordinates that make up each box shaped volume.
BOXID
User defined list ID number.
XMIN
Minimum x-coordinate.
XMAX
Maximum x-coordinate.
YMIN
Minimum y-coordinate.
YMAX
Maximum y-coordinate.
ZMIN
Minimum z-coordinate.
ZMAX
Maximum z-coordinate.
Command Default
Default for Option is ADD. You must specify BOXID to avoid an error message.
Notes
The ANSYS LS-DYNA command EDCGEN allows you to define contact and target volumes using box ID
numbers BOXID1 and BOXID2, respectively. If you use these arguments to define contact volumes, you must
first define their coordinates using the EDBX command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Contact>Define Box
427
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDBX
EDBVIS, QVCO, LVCO
Specifies global bulk viscosity coefficients for an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
QVCO
Quadratic viscosity coefficient (defaults to 1.5).
LVCO
Linear viscosity coefficient (defaults to 0.06).
Notes
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Bulk Viscosity
Main Menu>Preprocessor>Material Props>Bulk Viscosity
Main Menu>Solution>Analysis Options>Bulk Viscosity
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Bulk Viscosity
EDCADAPT, FREQ, TOL, OPT, MAXLVL, BTIME, DTIME, LCID, ADPSIZE, ADPASS, IREFLG, ADPENE, ADPTH,
MAXEL
Specifies adaptive meshing controls for an explicit dynamic analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
FREQ
Time interval between adaptive mesh refinements (default = 0.0). Use FREQ = STAT to list the current
adaptive meshing control settings.
TOL
Adaptive angle tolerance (in degrees) for which adaptive meshing will occur (default = 1e31). If the rel-
ative angle change between elements exceeds the specified tolerance value, the elements will be refined.
OPT
Adaptivity option:
1
Angle change (in degrees) of elements is based on original mesh configuration (default).
2
Angle change (in degrees) of elements is incrementally based on previously refined mesh.
MAXLVL
Maximum number of mesh refinement levels (default = 3). This parameter controls the number of times
an element can be remeshed. Values of 1, 2, 3, 4, etc. allow a maximum of 1, 4, 16, 64, etc. elements,
respectively, to be created for each original element.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
428
EDBVIS
BTIME
Birth time to begin adaptive meshing (default = 0.0).
DTIME
Death time to end adaptive meshing (default = 1e31).
LCID
Data curve number (previously defined on the EDCURVE command) identifying the interval of remeshing
(no default). The abscissa of the data curve is time, and the ordinate is the varied adaptive time interval.
If LCID is nonzero, the adaptive frequency (FREQ) is replaced by this load curve. Note that a nonzero
FREQ value is still required to initiate the first adaptive loop.
ADPSIZE
Minimum element size to be adapted based on element edge length (default = 0.0).
ADPASS
One or two pass adaptivity option.
0
Two pass adaptivity (default).
1
One pass adaptivity.
IREFLG
Uniform refinement level flag (no default). Values of 1, 2, 3, etc. allow 4, 16, 64, etc. elements, respectively,
to be created uniformly for each original element.
ADPENE
Adaptive mesh flag for starting adaptivity when approaching (positive ADPENE value) or penetrating
(negative ADPENE value) the tooling surface (default = 0.0).
ADPTH
Absolute shell thickness level below which adaptivity should begin. This option works only if the adaptive
angle tolerance (TOL) is nonzero. If thickness based adaptive remeshing is desired without angle change,
set TOL to a large angle. The default is ADPTH = 0.0, which means this option is not used.
MAXEL
Maximum number of elements at which adaptivity will be terminated (no default). Adaptivity is stopped
if this number of elements is exceeded.
Command Default
No adaptive meshing.
Notes
The EDCADAPT command globally sets the control options for all part IDs that are to be adaptively meshed
(see the EDADAPT command). Because FREQ defaults to zero, you must input a nonzero value in this field
in order to activate adaptive meshing. You must also specify a reasonable value for TOL since the default
adaptive angle tolerance (1e31) will not allow adaptive meshing to occur.
The EDCADAPT command is not supported in an explicit dynamic full restart analysis (EDSTART,3).
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
429
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCADAPT
Menu Paths
Main Menu>Solution>Analysis Options>Adaptive Meshing>Global Settings
Main Menu>Solution>Analysis Options>Adaptive Meshing>Status
EDCGEN, Option, Cont, Targ, FS, FD, DC, VC, VDC, V1, V2, V3, V4, BTIME, DTIME, BOXID1, BOXID2
Specifies contact parameters for an explicit dynamics analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the contact behavior (dictates the meaning of V1 through V4).
AG
Automatic general contact.
ANTS
Automatic nodes-to-surface contact.
ASSC
Automatic single surface contact.
ASS2D
Automatic 2-D single surface contact.
ASTS
Automatic surface-to-surface contact.
DRAWBEAD
Drawbead contact
ENTS
Eroding nodes-to-surface contact.
ESS
Eroding single surface contact.
ESTS
Eroding surface-to-surface contact.
FNTS
Forming nodes-to-surface contact.
FOSS
Forming one way surface-to-surface contact.
FSTS
Forming surface-to-surface contact.
NTS
Nodes-to-surface contact.
OSTS
One way surface-to-surface contact.
RNTR
Rigid nodes to rigid body contact.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
430
EDCGEN
ROTR
Rigid body to rigid body (one way) contact.
SE
Single edge contact.
SS
Single surface contact.
STS
Surface-to-surface contact.
TDNS
Tied nodes-to-surface contact.
TSES
Tied shell edge-to-surface contact.
TDSS
Tied surface-to-surface contact.
TNTS
Tiebreak nodes-to-surface contact
TSTS
Tiebreak surface-to-surface contact.
Cont
Contact surface identified by a component name [CM] , a part ID number [EDPART], or an assembly ID
number [EDASMP]. If a component name is input, the component must contain nodes that represent
the contact surface (assemblies are not valid for a component name). Alternatively, a part number may
be input that identifies a group of elements as the contact surface, or an assembly number may be input
containing a maximum of 16 parts. The assembly ID number must be greater than the highest number
used for the part ID. Cont is not required for automatic general contact, single edge contact, and single
surface contact options (Option = AG, SE, ASSC, ESS, and SS). For automatic 2-D single surface contact
(ASS2D), Cont must be defined as a part assembly. For eroding node-to-surface contact (ENTS), Cont
must be defined as a nodal component. For eroding single surface contact (ESS) and eroding surface-
to-surface contact (ESTS), Cont must be defined as a part ID or part assembly.
Targ
Target surface identified by a component name [CM] , a part ID number [EDPART], or an assembly ID
number [EDASMP]. If a component name is input, the component must contain nodes that represent
the target surface (assemblies are not valid for a component name). Alternatively, a part number may
be input that identifies a group of elements as the target surface, or an assembly number may be input
containing a maximum of 16 parts. The assembly ID number must be greater than the highest number
used for the part ID. Targ is not defined for automatic general contact, single edge contact, automatic
single surface contact, eroding single surface contact, single surface contact, and automatic 2-D single
surface contact options (Option = AG, SE, ASSC, ESS, SS, and ASS2D). For eroding node-to-surface
contact (ENTS) and eroding surface-to-surface contact (ESTS), Targ must be defined as a part ID or part
assembly.
FS
Static friction coefficient (defaults to 0).
FD
Dynamic friction coefficient (defaults to 0).
DC
Exponential decay coefficient (defaults to 0).
431
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCGEN
VC
Coefficient for viscous friction (defaults to 0).
VDC
Viscous damping coefficient in percent of critical damping (defaults to 0).
V1, V2, V3, V4
Additional input for drawbead, eroding, rigid, and tiebreak contact. The meanings of V1-V4 will vary,
depending on Option. See the table below for V1-V4 definitions.
Additional input for drawbead contact (Option = DRAWBEAD):
V1
Load curve ID giving the bending component of the restraining force per unit draw bead length as
a function of draw bead displacement. V1 must be specified.
V2
Load curve ID giving the normal force per unit draw bead length as a function of draw bead displace-
ment. V2 is optional.
V3
Draw bead depth.
V4
Number of equally spaced integration points along the draw bead (default = 0, in which case ANSYS
LS-DYNA calculates this value based on the size of the elements that interact with the draw bead).
Additional input for eroding contact (Option = ENTS, ESS, or ESTS):
V1
Symmetry plane option. The purpose of this option is to retain the correct boundary conditions in
a model with symmetry.
0
Off (default).
1
Do not include faces with normal boundary constraints (e.g., segments of brick elements on a
symmetry plane).
V2
Erosion/interior node option.
0
Erosion occurs only at exterior boundaries.
1
Interior eroding contact can occur (default).
V3
Adjacent material treatment for solid elements.
0
Solid element faces are included only for free boundaries (default).
1
Solid element faces are included if they are on the boundary of the material subset. This option
also allows erosion within a body and the consequent treatment of contact.
Additional input for rigid contact (Option = RNTR or ROTR):
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
432
EDCGEN
V1
Data curve id for force versus deflection behavior [EDCURVE]. Also specify V2. (No default.)
V2
Force calculation method for rigid contact. (No default.)
1
Data curve gives total normal force on surface versus maximum penetration of any node (only
applicable for Option = ROTR).
2
Data curve gives normal force on each node versus penetration of node through the surface
(Option = RNTR or ROTR).
3
Data curve gives normal pressure versus penetration of node through the surface (only applicable
for Option = RNTR).
V3
Unloading stiffness for rigid contact. This should not be larger than the maximum value used in the
data curve. The default is to unload along the data curve (specified on V1).
Additional input for tiebreak surface-to-surface contact (Option = TSTS). V1 and V2 are used to calculate
the failure criterion:
V1
Normal failure stress. (No default.)
V2
Shear failure stress. (No default.)
Additional input for tiebreak nodes-to-surface contact (Option = TNTS). V1 through V4 are used to
calculate the failure criterion:
V1
Normal failure force. Only tensile failure (i.e., tensile normal forces) will be considered in the failure
criterion. (No default.)
V2
Shear failure force. (No default.)
V3
Exponent for normal force. (Defaults to 2.)
V4
Exponent for shear force. (Defaults to 2.)
BTIME
Birth time for which contact definition will become active (defaults to 0.0).
DTIME
Death time for which contact definition will become inactive (defaults to 1e21).
BOXID1
Contact volume as defined using the EDBX command (valid only when defining contact with parts or
assemblies).
BOXID2
Target volume as defined using the EDBX command (valid only when defining contact with parts or
assemblies).
433
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCGEN
Command Default
No contact defined.
Notes
The frictional coefficient used for contact is determined from FS, FD, and DC, and is assumed to be dependent
on the relative velocity of the surfaces in contact:
c = FD+(FS-FD) e
-DC(v
rel
)
The coefficient for viscous friction, VC, is necessary to limit the friction force to a maximum. A limiting force
is computed:
F
lim
= VC(A
cont
)
where A
cont
is the area of the segment contacted by the node in contact. The suggested value for VC is to
use the yield stress in shear:
VC
o
3
where
o
is the yield stress of the contacted material.
If a part number is input for Cont or Targ, it must represent a valid explicit dynamics part definition. For
example, an elastic material for explicit dynamics requires EX, NUXY, and DENS. If any part of the material
definition is missing for the part, the EDCGEN command will be ignored.
In addition to the contact parameters on this command, you can specify penalty scale factors for the contact
(slave) and target (master) surfaces by using the EDCMORE command.
Duplicate definitions of the same contact type on the same components or parts will cause only one contact
to be defined (previous definitions will be ignored). Duplicate definitions of different contact types on the
same components or parts will cause multiple contact types to be defined.
Use the EDCLIST and EDDC commands to list and delete contact surface specifications. Use the EDPC
command to select and plot contact entities.
The EDCGEN command is not supported in an explicit dynamic full restart analysis (EDSTART,3). Thus, you
cannot add new contact specifications in a full restart. You can use the EDCLIST command to list any contact
specifications that were defined in the previous analysis.
EDCGEN is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Contact>Define Contact
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
434
EDCGEN
EDCLIST, NUM
Lists contact entity specifications in an explicit dynamics analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
NUM
Number identifying contact entity to be listed. Use NUM = ALL to list all contact entities (ALL is the default).
Notes
Lists contact entity specifications previously defined with the EDCGEN command. The listing will include
any contact parameters defined using the EDCMORE command.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Contact>List Entities
EDCMORE, Option, NUM, --, VAL1,VAL2
Specifies additional contact parameters for a given contact definition in an explicit dynamic analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the option to be performed.
ADD
Define contact parameters for the contact entity specified by NUM (default).
DELE
Delete contact parameters (VAL1 and VAL2) for the contact entity specified by NUM. If NUM = ALL,
all contact parameters previously defined by EDCMORE are deleted.
NUM
Contact entity number. This contact entity must have been previously defined with the EDCGEN command.
Use EDCLIST to obtain a list of contact entity numbers.
--
Unused field.
VAL1
Penalty scale factor for slave (contact) surface (SFS); default = 1.
VAL2
Penalty scale factor for master (target) surface (SFM); default = 1.
435
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCMORE
Command Default
For all contract definitions, SFS = 1 and SFM = 1.
Notes
You can use the EDCMORE command to specify two additional contact parameters (SFS and SFM) for a
specific contact definition. These parameters will apply only to the contact entity number entered on the
NUM field. Use the EDCLIST command to obtain a list of contact definitions and their corresponding contact
entity numbers. The listing produced by EDCLIST will include any contact parameters specified with the
EDCMORE command.
When you use the EDDC command to delete a contact definition, any parameters you specified with ED-
CMORE for that contact definition will also be deleted. To delete only the parameters specified by EDCMORE
for a given contact definition, use the command EDCMORE,DELE,NUM.
Note
When you delete a contact definition with the EDDC command, the contact entity numbers will
be renumbered for the remaining contact definitions. Therefore, you should always issue EDCLIST
to obtain a current list of contact entity numbers before adding or deleting contact parameters
with the EDCMORE command.
The EDCMORE command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Contact>Additional Parms
EDCNSTR, Option, Ctype, Comp1, Comp2, VAL1
Defines various types of constraints for an explicit dynamic analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the option to be performed.
ADD
Define a constraint (default).
DELE
Delete the constraint specified by Ctype, Comp1, and Comp2. If Ctype = ALL, all constraints are
deleted.
LIST
List all of the constraints previously defined by the EDCNSTR command.
Ctype
Constraint type. The command format will vary, depending on the Ctype value.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
436
EDCNSTR
ENS
Extra node set added to an existing rigid body.
NRB
Nodal rigid body.
STS
Tie between a shell edge and solid elements.
RIVET
Massless rivet between two noncoincident nodes.
If Ctype = ENS, the command format is EDCNSTR,Option,ENS,Comp1,Comp2
Comp1
Part number of the existing rigid body to which nodes will be added. The rigid body must be previously
defined with the EDMP command. EDMP defines a rigid body based on material reference number
(MAT). You must determine the corresponding part number (EDPART) for input in this field.
Comp2
Component name identifying extra nodes to be added to the rigid body specified by Comp1. Comp2
must be a nodal component and must not be attached to any other rigid body.
If Ctype = NRB, the command format is EDCNSTR,Option,NRB,Comp1, --,VAL1
Comp1
Component name identifying a set of nodes that are to be defined as a rigid body. The component may
consist of nodes from several different deformable parts.
--
This field is not used for Ctype = NRB.
VAL1
Coordinate system ID number (CID) to be used for output of data. The coordinate system must have
been previously defined with the EDLCS command.
If Ctype = STS, the command format is EDCNSTR,Option,STS,Comp1,Comp2
Comp1
Node number of the shell element node that will be tied to solid element nodes that are specified by
Comp2. The Comp1 node must lie along the edge of a shell element and be coincident to at least one
node included in Comp2.
Comp2
Component name consisting of solid element nodes (up to nine nodes) to which the shell element node
will be tied. Comp2 must consist of nodes that are on solid elements, and the nodes must define a line
that will remain linear throughout the analysis. At least one of the nodes in Comp2 must be coincident
with the shell node specified in Comp1.
If Ctype = RIVET, the command format is EDCNSTR,Option,RIVET,Comp1,Comp2
Comp1
Node number of the first node on the rivet.
Comp2
Node number of the second node on the rivet. This node cannot have the same coordinates as the first
node specified by Comp1.
437
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCNSTR
Notes
The EDCNSTR command allows you to define several types of constraints in an explicit dynamic analysis. A
brief description of each constraint type is given below. See Constraints and Initial Conditions in the ANSYS
LS-DYNA User's Guide for more information.
Extra Node Set Added to a Rigid Body (Ctype = ENS)
The ability to add extra nodes to an existing rigid body has many potential applications, including placing
nodes where joints will be attached between rigid bodies, defining nodes where point loads will be applied,
and defining a lumped mass at a specific location. The extra nodes specified by Comp2 may be located
anywhere in the model and may have coordinates outside those of the original rigid body specified by
Comp1.
Nodal Rigid Body (Ctype = NRB)
Unlike typical rigid bodies that are defined with the EDMP command, nodal rigid bodies defined with the
EDCNSTR command are not associated with a part number. This can be advantageous for modeling rigid
(welded) joints in a model. For a rigid joint, portions of different flexible components (having different MAT
IDs) act together as a rigid body. It is difficult to define this type of rigid body with a unique MAT ID (and
corresponding part number). However, the rigid joint can be easily defined using a nodal rigid body.
Shell Edge to Solid Tie (Ctype = STS)
The STS option ties regions of solid elements to regions of shell elements. A single shell node may be tied
to up to nine brick element nodes that define a fiber vector. Solid element nodes constrained in this way
remain linear throughout the analysis but can move relative to each other in the fiber direction.
Rivet between Two Nodes (Ctype = RIVET)
The RIVET option defines a massless rigid constraint between two nodes, similar to spotwelds defined with
the EDWELD command. Unlike a spotweld, however, rivets contain nodes that are noncoincident, and failure
cannot be specified. When a rivet is defined, the distance between the nodes is kept constant throughout
any motion that occurs during a simulation. Nodes connected by a rivet cannot be part of any other constraints
specified in the model.
The EDCNSTR command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Apply>Additional Nodal
Main Menu>Solution>Constraints>Apply>Additional Nodal
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
438
EDCNSTR
EDCONTACT, SFSI, RWPN, IPCK, SHTK, PENO, STCC, ORIE, CSPC, PENCHK
Specifies contact surface controls for an explicit dynamics analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
SFSI
Scale factor for sliding interface penalties. Defaults to 0.1.
RWPN
Scale factor for rigid wall penalties (defaults to 0). If RWPN = 0, rigid bodies interacting with rigid walls
are not considered. If RWPN>0, rigid bodies interact with fixed rigid walls. A value of 1.0 should be op-
timal; however, this may be problem dependent.
IPCK
Initial contact surface penetration checking option:
1
No checking.
2
Full check of initial penetration is performed (default).
SHTK
Shell thickness contact option for surface-to-surface and nodes-to-surface contact (see Notes below):
0
Thickness is not considered (default).
1
Thickness is considered, except in rigid bodies.
2
Thickness is considered, including rigid bodies.
PENO
Penalty stiffness option (options 4 and 5 are useful for metal forming calculations):
1
Minimum of master segment and slave node (default).
2
Use master segment stiffness.
3
Use slave node value.
4
Use area or mass weighted slave node value.
5
Use slave node value inversely proportional to shell thickness. (This may require special scaling and
is not generally recommended.)
STCC
Shell thickness change option for single surface contact:
1
Shell thickness changes are not considered (default).
439
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCONTACT
2
Shell thickness changes are included.
ORIE
Option for automatic reorientation of contact surface segments during initialization:
1
Activate for automated (part ID) input only (default).
2
Activate for manual (nodal component) and automated (part ID) input.
3
Do not activate.
CSPC
Contact surface penetration check multiplier, used if small penetration checking is on (PENCHK = 1 or
2). Defaults to 4.
PENCHK
Small penetration check, used only for contact types STS, NTS, OSTS, TNTS, and TSTS. If the contact surface
node penetrates more than the target thickness times CSPC, the penetration is ignored and the contacting
node is set free. The target thickness is the element thickness for shell elements, or 1/20 of the shortest
diagonal for solid elements.
0
Penetration checking is off (default).
1
Penetration checking is on.
2
Penetration checking is on, but shortest diagonal is used.
Notes
The thickness offsets are always included in single surface, automatic surface-to-surface, and automatic
nodes-to-surface contact. The shell thickness change option must be used [EDSHELL,,,1] and a nonzero
value must be specified for SHTK before the shell thickness changes can be included in the surface-to-surface
contact type. Additionally, STCC must be set to 2 if thickness changes are to be included in the single surface
contact algorithms.
To reset the contact options to default values, issue the EDCONTACT command with no fields specified.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Contact>Advanced Controls
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
440
EDCONTACT
EDCPU, CPUTIME
Specifies CPU time limit for an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
CPUTIME
CPU time limit (in seconds) for the current phase of the analysis (defaults to 0). If CPUTIME = 0, no CPU
time limit is set. CPUTIME values below 0 are not allowed.
Notes
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>CPU Limit
EDCRB, Option, NEQN, PARTM, PARTS
Constrains two rigid bodies to act as one in an explicit dynamics analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the option to be performed:
ADD
Define an equation to constrain two rigid bodies (default).
DELE
Delete the equation (specified by NEQN) that constrains two rigid bodies. If NEQN is blank, all equations
constraining rigid bodies are deleted.
LIST
List constrained rigid bodies specified by NEQN. If NEQN is blank, all constrained rigid bodies are
listed.
NEQN
Equation reference number. Defaults to PARTS. NEQN should be a unique number for each pair of PARTM
and PARTS. If it is not unique, the equation reference number defined last will overwrite any previously
defined NEQN with the same number.
PARTM
PART number [EDPART] identifying the master rigid body. This value is ignored if the DELE or LIST labels
are specified. No default; you must enter a value.
PARTS
PART number [EDPART] identifying the slave rigid body. This value is ignored if the DELE or LIST labels
are specified. No default; you must enter a value.
441
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCRB
Notes
EDCRB is valid only for materials defined as rigid bodies with the EDMP,RIGID command. EDCRB automat-
ically generates a constraint equation to force the specified rigid bodies to behave as a single rigid body.
The slave rigid body takes on the material properties and loading of the master rigid body. Any loads [ED-
LOAD] existing on the slave rigid body are ignored.
To create a single large rigid body from several smaller bodies, use a series of EDCRB commands. With the
first command, specify a master and slave to create the first combined rigid body. Then, using that body as
the master, specify another slave to create a larger rigid body. Continue the process, using the expanding
rigid body as the master and adding slave bodies until you have defined the desired large rigid body. All
slave rigid bodies will take on the material properties and loading of the original master rigid body. Note
that you will need to use different NEQN values for each pair of PARTM and PARTS. This command will be
ignored if you specify the previously-defined master rigid body as a slave rigid body in the same analysis.
To change the master and slave definitions, first use the DELE option to delete all master and slave definitions,
and then use the ADD option to redefine them.
The equation number, NEQN, is a reference number by which the constrained bodies can be identified for
listing and deleting purposes on the EDCRB command. For any other reference to the constrained bodies
(loading, contact definitions, etc.), use the master body part number (PARTM).
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Rigid Body CE
EDCSC, Key
Specifies whether to use subcycling in an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Key
Subcycling key:
OFF
Do not use subcycling (default).
ON
Use subcycling.
Command Default
No subcycling.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
442
EDCSC
Notes
Subcycling can be used to speed up an analysis when element sizes within a model vary significantly. Relatively
small elements will result in a small time step size. When subcycling is on, the minimum time step size is
increased for the smallest elements.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Time Controls>Subcycling
EDCTS, DTMS, TSSFAC
Specifies mass scaling and scale factor of computed time step for an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
DTMS
Time step size for mass scaled solutions (defaults to 0).
TSSFAC
Scale factor for computed time step. Defaults to 0.9; if high explosives are used, the default is lowered
to 0.67.
Command Default
No mass scaling; scale factor for computed time step = 0.9.
Notes
If DTMS is positive, the same time step size will be used for all elements and mass scaling will be done for
all elements. Therefore, positive values should only be used if inertial effects are insignificant.
If DTMS is negative, mass scaling is applied only to elements whose calculated time step size is smaller than
DTMS. Negative values should only be used in transient analyses if the mass increases are insignificant.
In order to use mass scaling in an explicit dynamic small restart analysis (EDSTART,2) or full restart analysis
(EDSTART,3), mass scaling must have been active in the original analysis. The time step and scale factor
used in the original analysis will be used by default in the restart. You can issue EDCTS in the restart analysis
to change these settings.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Time Controls>Time Step Ctrls
443
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCTS
EDCURVE, Option, LCID, Par1, Par2
Specifies data curves for an explicit dynamic analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Label identifying the option to be performed.
ADD
Define a data curve (default). If Option = ADD, Par1 and Par2 must be previously defined array
parameters.
DELE
Delete the specified data curve (LCID). If LCID is blank, all data curves are deleted. Par1 and Par2
are ignored for this option.
LIST
List defined data curve (LCID). If LCID is blank, all data curves are listed. Par1 and Par2 are ignored
for this option.
PLOT
Plot defined data curve (LCID). If Option = PLOT, LCID must be previously defined with an ED-
CURVE command. Otherwise a warning message will report that LCID has not been defined. Par1
and Par2 are ignored for this option.
LCID
Data curve ID number (no default). Must be a positive integer.
Par1
Name of user-defined array parameter that contains the abscissa values of the curve data (e.g., time,
effective plastic strain, effective strain rate, displacement, etc.).
Par2
Name of user-defined array parameter that contains the ordinate values of the curve data (e.g., damping
coefficients, initial yield stress, elastic modulus, force, etc.) corresponding to the abscissa values in Par1.
Note
If the length of Par1 and Par2 are different, the shortest length will be used.
Notes
EDCURVE can be used to define material data curves (e.g., stress-strain) and load data curves (force-deflection)
associated with material models in an explicit dynamics analysis. Material data specified by this command
is typically required to define a particular material behavior (e.g., TB,HONEY), and the LCID number is used
as input on the TBDATA command.
EDCURVE can also be used to define load curves that represent time dependent loads (force, displacement,
velocity, etc.). Par1 must contain the time values, and Par2 must contain the corresponding load values.
The LCID number assigned to the load curve can be used as input on the EDLOAD command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
444
EDCURVE
Note
You cannot update a previously defined data curve by changing the array parameters that were
input as Par1 and Par2. The data curve definition is written to the database at the time EDCURVE
is issued. Therefore, subsequent changes to the array parameters that were used as input on
EDCURVE will not affect the load curve definition. If you need to change the load curve definition,
you must delete the load curve (EDCURVE,DELE,LCID) and define it again.
LCID identifies the data curve. If the value input for LCID is the same as the ID number for a data curve
previously defined by EDCURVE, the previous data will be overwritten. Use EDCURVE,LIST and EDCURVE,PLOT
to check existing data curves.
A starting array element number must be specified for Par1 and Par2. The input for these fields must be
a single column array parameter, or a specific column from a multi-column array parameter. When using
the GUI with multi-column parameters, you must specify the parameter name and starting position for Par1
and Par2 by typing the EDCURVE command in the Input Window. This is because only the parameter name
is available through the dialog box, which pulls in the first position of a single-column array parameter.
If you need to change a curve definition in an explicit dynamic small restart analysis, issue EDSTART,2 first
(to specify the restart), then issue the EDCURVE command. The revised curve must contain the same number
of points as the curve it replaces. This limitation does not apply to a full restart analysis (EDSTART,3).
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Curve Options>Add
Curve
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Curve Options>Delete
Curve
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Curve Options>List
Curve
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Curve Options>Plot
Curve
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Curve Options>Add Curve
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Curve Options>Delete Curve
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Curve Options>List Curve
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Curve Options>Plot Curve
Main Menu>Preprocessor>Material Props>Curve Options>Add Curve
Main Menu>Preprocessor>Material Props>Curve Options>Delete Curve
Main Menu>Preprocessor>Material Props>Curve Options>List Curve
Main Menu>Preprocessor>Material Props>Curve Options>Plot Curve
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Curve Options>Add Curve
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Curve Options>Delete Curve
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Curve Options>List Curve
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Curve Options>Plot Curve
Main Menu>Solution>Loading Options>Curve Options>Add Curve
Main Menu>Solution>Loading Options>Curve Options>Delete Curve
Main Menu>Solution>Loading Options>Curve Options>List Curve
445
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDCURVE
Main Menu>Solution>Loading Options>Curve Options>Plot Curve
EDDAMP, PART, LCID, VALDMP
Defines mass weighted (Alpha) or stiffness weighted (Beta) damping for an explicit dynamics model.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
PART
PART number [EDPART] identifying the group of elements to which damping should be applied. If PART
= ALL (default), damping is applied to the entire model.
LCID
Load curve ID (previously defined with the EDCURVE command) identifying the damping coefficient
versus time curve. If time-dependent damping is defined, an LCID is required.
VALDMP
Constant system damping coefficient or a scale factor applied to the curve defining damping coefficient
versus time.
Notes
Mass-weighted (Alpha) or stiffness-weighted (Beta) damping can be defined with the EDDAMP command.
Generally, stiffness proportional or beta damping is effective for oscillatory motion at high frequencies. This
type of damping is orthogonal to rigid body motion and so will not damp out rigid body motion. On the
other hand, mass proportional or alpha damping is more effective for low frequencies and will damp out
rigid body motion. The different possibilities are described below:
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Global Damping
Mass-weighted or Alpha damping
When PART = (blank) or ALL (default), mass-weighted global damping can be defined in the following two
ways. In this case, the same damping is applied for the entire structure.
When the damping coefficient versus time curve (LCID) is specified using the EDCURVE command,
VALDMP is ignored by LS-DYNA (although it is written in the LS-DYNA input file Jobname.K). The
damping force applied to each node in the model is given by f
d
= d(t)mv, where d(t) is the damping
coefficient as a function of time defined by the EDCURVE command, m is the mass, and v is the velocity.
When the LCID is 0 or blank (default), a constant mass-weighted system damping coefficient can be
specified using VALDMP.
The constant and time-dependent damping, described above, cannot be defined simultaneously. The last
defined global damping will overwrite any previously defined global damping.
Damping defined for a PART
Mass-weighted or Alpha damping
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
446
EDDAMP
When both a valid PART number is specified and the damping coefficient versus time curve (LCID) is specified
using the EDCURVE command, mass-weighted time-dependent damping will be defined for the particular
PART. In this case, VALDMP will act as a scaling factor for the damping versus time curve (if VALDMP is not
specified, it will default to 1). A valid PART number must be specified to define this type of damping. For
example, use PART =1 (and not blank) when the entire model consists of only one PART. Issue the command
repeatedly with different PART numbers in order to specify alpha damping for different PARTS.
Stiffness-weighted or Beta damping
When a valid PART number is specified with LCID = 0 or (blank) (default), a stiffness-weighted (Beta) constant
damping coefficient for this particular PART can be defined by VALDMP. The stiffness-weighted value cor-
responds to the percentage of damping in the high frequency domain. For example, 0.1 roughly corresponds
to 10% damping in the high frequency domain. Recommended values range from 0.01 to 0.25. Values lower
than 0.01 may have little effect. If a value larger than 0.25 is used, it may be necessary to lower the time
step size significantly (see the EDCTS command). Issue the command repeatedly with different PART numbers
in order to specify beta damping for different PARTS. Time-dependent stiffness-weighted damping is not
available in ANSYS LS-DYNA.
The mass-weighted and stiffness-weighted damping, described above, cannot be defined simultaneously
for a particular PART number. The last defined damping for the particular PART number will overwrite any
previously defined mass-weighted or stiffness-weighted damping for this PART.
In order to define the mass-weighted and stiffness-weighted damping simultaneously, you can use the
MP,DAMP command (instead of the EDDAMP,PART, ,VALDMP command) to define stiffness-weighted (Beta)
constant damping coefficient. However, do not use both of these commands together to define stiffness-
weighted (Beta) constant damping coefficient for a particular PART. If you do, duplicate stiffness-weighted
(Beta) constant damping coefficients for this PART will be written to the LS-DYNA input file Jobname.K.
The last defined value will be used by LS-DYNA. Also, note that the MP,DAMP command is applied on the
MAT number, and not on the PART number. Since a group of elements having the same MAT ID may belong
to more than one PART (the opposite is not true), you need to issue the MP,DAMP command only once for
this MAT ID and the stiffness-weighted (Beta) damping coefficients will be automatically defined for all the
PARTs with that MAT ID.
Mass-weighted and stiffness-weighted damping can be defined simultaneously using the EDDAMP command
only when mass-weighted damping (constant or time-dependent) is defined as global damping (EDDAMP,
ALL, LCID, VALDMP) and stiffness-weighted damping is defined for all necessary PARTs (EDDAMP,PART,
,VALDMP).
To remove defined global damping, reissue the EDDAMP, ALL command with LCID and VALDMP set to 0.
To remove damping defined for a particular PART, reissue EDDAMP, PART, where PART is the PART number,
with LCID and VALDMP set to 0. There is no default for the EDDAMP command, i.e., issuing the EDDAMP
command with PART = LCID = VALDMP = 0 will result in an error. Stiffness-weighted damping defined by
the MP,DAMP command can be deleted using MPDELE, DAMP, MAT.
In an explicit dynamic small restart (EDSTART,2) or full restart analysis (EDSTART,3), you can only specify
global alpha damping. This damping will overwrite any alpha damping input in the original analysis. If you
do not input global alpha damping in the restart, the damping properties input in the original analysis will
carry over to the restart.
Damping specified by the EDDAMP command can be listed, along with other explicit dynamics specifications,
by typing the command string EDSOLV$STAT into the ANSYS input window. Beta damping specified by
the MP,DAMP command can be listed by MPLIST, MAT command.
447
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDDAMP
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Damping
Main Menu>Preprocessor>Material Props>Damping
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Damping
EDDBL, KEY
Selects a numerical precision type of the explicit dynamics analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
KEY
Number or name identifying numerical precision to be used.
0 or SINGLE
Select single precision version of LS-DYNA (default).
1 or DOUBLE
Select double precision version of LS-DYNA.
STAT
Check the status of the numerical precision in effect.
Notes
Sets the single or double precision version of LS-DYNA into effect. Please check the availability of the double
precision version of LS-DYNA on your system before using the command. If it is not available, use the com-
mand default.
The double precision version may be up to 20% slower than the single precision version. The results may
also vary based on problem specifications.
In addition to EDDBL,STAT, you can use the GUI dialog box to verify which precision version is currently
chosen. The GUI is based on the database and is updated to reflect changes.
See Double Precision LS-DYNA for more information.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>Double Precision
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
448
EDDBL
EDDC, Option, Ctype, Cont, Targ
Deletes or deactivates/reactivates contact surface specifications in an explicit dynamic analysis.
PREP7: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Option to be performed for contact definition specified by Ctype, Cont, and Targ.
DELE
Delete the specified contact definition (default); valid only in a new analysis.
DACT
Deactivate the specified contact definition; valid only in a small restart.
RACT
Reactivate the specified contact definition (which was previously deactivated); valid only in a small
restart.
Ctype
Contact behavior label (see EDCGEN command for valid labels).
Cont
Component name or part number [EDPART] identifying the contact surface.
Targ
Component name or part number [EDPART] identifying the target surface.
Notes
This command allows you to delete or deactivate/reactivate a particular contact specification that was defined
by EDCGEN. The contact definition is identified by Ctype, Cont, and Targ (Note that Cont and Targ
may not be required for Ctype = AG, SE, ASSC, ESS, and SS). The delete option (Option = DELE) permanently
deletes the contact from the database. Any additional contact parameters defined with the EDCMORE
command for the contact definition identified on this command will also be deleted or deactivated/reactivated.
You cannot delete contact specifications in an explicit dynamic small restart (EDSTART,2). However, you can
use Option = DACT to deactivate a contact definition that is not needed in the small restart. That contact
definition may then be reactivated in a subsequent small restart by using Option = RACT.
To delete or deactivate/reactivate all contact specifications for the entire model, use EDDC,Option,ALL.
The EDDC command is not supported in an explicit dynamic full restart analysis (EDSTART,3). Thus, you
cannot delete, deactivate, or reactivate contact specifications in a full restart that were defined in the previous
analysis.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Contact>Activate Entity
Main Menu>Preprocessor>LS-DYNA Options>Contact>Deactvate Entity
Main Menu>Preprocessor>LS-DYNA Options>Contact>Delete Entity
449
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDDC
EDDRELAX, Option, NRCYCK, DRTOL, DFFCTR, DRTERM, TSSFDR, IRELAL, EDTTL
Activates initialization to a prescribed geometry or dynamic relaxation for the explicit analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Specifies when dynamic relaxation is activated.
ANSYS
Stresses are initialized in ANSYS LS-DYNA to a prescribed geometry for small strains, according to
the solution of an ANSYS (implicit) run. The explicit solution is based on the implicit X,Y,Z displace-
ments and rotations contained in the drelax file (created with the REXPORT command).
DYNA
Dynamic relaxation is on. When you use this option, you can specify some or all of the parameters
NRCYCK, DRTOL, DFFCTR, DRTERM, TSSFDR, IRELAL, and EDTTL. Any parameters that you do
not specify are set to their default values.
OFF
Turn off initialization to a prescribed geometry (Option = ANSYS) or dynamic relaxation (Option
= DYNA).
NRCYCK
Number of iterations between convergence checks for dynamic relaxation option. Default = 250.
DRTOL
Convergence tolerance for dynamic relaxation option. Default = 0.001.
DFFCTR
Dynamic relaxation factor. Default = 0.995.
DRTERM
Optional termination time for dynamic relaxation. Termination occurs at this time, or when convergence
is attained, whichever comes first. Default = infinity.
TSSFDR
Scale factor for computed time step during dynamic relaxation. If zero, the value is set to TSSFAC
(defined on the EDCTS command). After converging, the scale factor is reset to TSSFAC.
IRELAL
Automatic control for dynamic relaxation option based on algorithm of Papadrakakis.
0
Not active (default).
1
Active.
EDTTL
Convergence tolerance on automatic control of dynamic relaxation (default = 0.04).
Notes
Use Option = ANSYS when running an implicit-to-explicit sequential solution to initialize the structure to
a static solution performed earlier by the ANSYS implicit solver. Use Option = DYNA to perform dynamic
relaxation within the LS-DYNA program. Use Option = OFF to turn off previously specified stress initialization
or dynamic relaxation. You must specify the Option you want; there is no default.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
450
EDDRELAX
In LS-DYNA, the dynamic relaxation is performed before the regular transient analysis. The convergence
process of the dynamic relaxation is not written to the ANSYS history file. The ANSYS results files only include
the converged result of the dynamic relaxation, which is the result at time zero in the Jobname.HIS and
Jobname.RST files.
You can restart a dynamic relaxation analysis (EDSTART,2 or EDSTART,3) from a previous transient analysis
or a previous dynamic relaxation analysis. In the restart, you can change or set the convergence criteria with
the EDDRELAX command. Only the load curves that are flagged for dynamic relaxation (PHASE = 1 or 2 on
EDLOAD) are applied after restarting. If you restart the explicit portion of an implicit-to-explicit sequential
solution, you do not need to reissue the REXPORT command because displacement information contained
in the drelax file is already included in the LS-DYNA restart file. If the dynamic relaxation is activated from
a regular transient analysis, LS-DYNA continues the output of data to ANSYS results files. This is unlike the
dynamic relaxation phase at the beginning of the calculation for which only the converged solution is written.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>Dynamic Relax
EDDUMP, NUM, DT
Specifies output frequency for the explicit dynamic restart file (d3dump).
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
NUM
Number of d3dump (restart) files written during the analysis (defaults to 1). When you specify NUM, the
time interval between restart files is TIME / NUM, where TIME is the analysis end-time specified on the
TIME command.
DT
Time interval at which the d3dump (restart) files are written. If NUM is input, DT is ignored.
Command Default
One restart file is written at the end of the analysis.
Notes
You can use NUM or DT to specify the time interval at which d3dump restart files will be written. You should
not specify both quantities; if both are input, NUM will be used. The restart files are written sequentially as
d3dump01, d3dump02, etc.
In LS-DYNA, the restart file output is specified in terms of number of time steps. Because the total number
of time steps is not known until the LS-DYNA solution finishes, ANSYS calculates an approximate number
of time steps for the solution, and then uses NUM or DT to calculate the required LS-DYNA input. This ap-
proximated number of time steps may be different from the total number reached in LS-DYNA after the
solution finishes. Therefore, the number of restart dump files or the output interval may differ slightly from
what you requested using NUM or DT.
451
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDDUMP
In an explicit dynamic small restart (EDSTART,2) or full restart analysis (EDSTART,3), the EDDUMP setting
will default to the NUM or DT value used in the original analysis. You can issue EDDUMP in the restart to
change this setting.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Output Controls>File Output Freq>Number of Steps
Main Menu>Solution>Output Controls>File Output Freq>Time Step Size
EDELE, IEL1, IEL2, INC
Deletes selected elements from the model.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IEL1, IEL2, INC
Delete elements from IEL1 to IEL2 (defaults to IEL1) in steps of INC (defaults to 1). If IEL1 = ALL,
IEL2 and INC are ignored and all selected elements [ESEL] are deleted. If IEL1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for IEL1 (IEL2 and INC are ignored).
Notes
Deleted elements are replaced by null or "blank" elements. Null elements are used only for retaining the
element numbers so that the element numbering sequence for the rest of the model is not changed by
deleting elements. Null elements may be removed (although this is not necessary) with the NUMCMP
command. If related element data (pressures, etc.) are also to be deleted, delete that data before deleting
the elements. EDELE is for unattached elements only. You can use the xCLEAR family of commands to remove
any attached elements from the database.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Main Menu>Preprocessor>Modeling>Delete>Elements
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
EDENERGY, HGEN, SWEN, SIEN, RLEN
Specifies energy dissipation controls for an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
HGEN
Hourglass energy control key:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
452
EDELE
OFF or 0
Hourglass energy is not computed.
ON or 1
Hourglass energy is computed and included in the energy balance (default).
SWEN
Stonewall energy dissipation control key:
OFF or 0
Stonewall energy dissipation is not computed.
ON or 1
Stonewall energy dissipation is computed and included in the energy balance (default).
SIEN
Sliding interface energy dissipation control key:
OFF or 0
Sliding interface energy dissipation is not computed.
ON or 1
Sliding interface energy dissipation is computed and included in the energy balance (default).
RLEN
Rayleigh (damping) energy dissipation control key:
OFF or 0
Rayleigh energy dissipation is not computed.
ON or 1
Rayleigh energy dissipation is computed and included in the energy balance (default).
Notes
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>Energy Options
EDFPLOT, Key
Allows plotting of explicit dynamics forces and other load symbols.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Key
Load symbol plotting key.
ON or 1
Turn display of load symbols on (default).
OFF or 0
Turn display of load symbols off.
453
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDFPLOT
Command Default
Display of load symbols on.
Notes
You must issue EDFPLOT,ON to display explicit dynamics load symbols. The explicit load symbols are erased
automatically upon a subsequent plot command.
An explicit load symbol always indicates a positive load direction (e.g., positive X direction for FX load), even
if the load value is negative. The load symbol does not reflect the load magnitude. You can use standard
ANSYS symbol controls to control the appearance of the load symbol. No load symbol is displayed for tem-
perature loads.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Show Forces
Main Menu>Solution>Loading Options>Show Forces
Utility Menu>PlotCtrls>Symbols
EDGCALE, NADV, METH
Defines global ALE controls for an explicit dynamic analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
NADV
Number of cycles between advection (default = 0).
METH
Advection method.
0
Donor cell + Half Index Shift (first order accurate) (default).
1
Van Leer + Half Index Shift (second order accurate).
Command Default
The Lagrangian formulation is used for all elements by default.
Notes
This command sets global ALE controls in an explicit dynamic analysis. These ALE controls apply to all
PLANE162 or SOLID164 elements in the model that are flagged to use the ALE formulation (KEYPOPT(5) =
1). To activate the ALE formulation, you must specify the number of cycles between advection on this
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
454
EDGCALE
command and at least one smoothing weight factor on the EDALE command. See Arbitrary Lagrangian-
Eulerian Formulation in the ANSYS LS-DYNA User's Guide for more information.
To see the current EDGCALE settings, issue the command EDALE,LIST.
The EDGCALE command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Analysis Options>ALE Options>Define
/EDGE, WN, KEY, ANGLE
Displays only the "edges" of an object.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
KEY
Edge key:
Elements Plots
0
Display common lines between all adjacent element faces.
1
Display only the common lines between non-coplanar faces (that is, show only the edges).
Contour Plots
0
Display only the common lines between non-coplanar faces.
1
Display common lines between all element faces.
ANGLE
Largest angle between two faces for which the faces are considered to be coplanar (0 to 180). Defaults
to 45. A smaller angle produces more edges, a larger angle produces fewer edges.
Command Default
For element plots, display common lines between all adjacent element faces. For contour plots, display only
the common lines between non-coplanar faces.
455
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/EDGE
Notes
The ANGLE field is used in PowerGraphics to determine geometric discontinuities. It is a tolerance measure
for the differences between the normals of the surfaces being considered. Values within the tolerance are
accepted as coplanar (geometrically continuous).
A surface can be displayed as an edge outline without interior detail. This is useful for both geometry and
postprocessing displays. Element outlines are normally shown as solid lines for geometry and displacement
displays. Lines common to adjacent "coplanar" element faces are removed from the display. Midside nodes
of elements are ignored. The /SHRINK option is ignored with the edge option. /EDGE is not supported for
PLESOL and /ESHAPE displays when in PowerGraphics mode [/GRAPHICS,POWER].
The /EDGE command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Edge Options
EDHGLS, HGCO
Specifies the hourglass coefficient for an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
HGCO
Hourglass coefficient value (defaults to 0.1). Values greater than 0.15 may cause instabilities.
Notes
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Hourglass
Ctrls>Global
Main Menu>Preprocessor>Material Props>Hourglass Ctrls>Global
Main Menu>Solution>Analysis Options>Hourglass Ctrls>Global
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Hourglass Ctrls>Global
EDHIST, Comp
Specifies time-history output for an explicit dynamic analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Comp
Name of the component containing nodes or elements for which output is desired. Comp is required.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
456
EDHGLS
Command Default
No time-history output is written.
Notes
The time-history output is written to the file Jobname.HIS. Output is written only for the nodes or elements
contained in Comp. The data is written at time intervals specified on the EDHTIME command. If no time
interval is specified, output is written at 1000 steps over the analysis. (See also the EDOUT command which
controls time-history output in ascii form for an explicit dynamics analysis.)
Use EDHIST,LIST to list the time-history output specification. (The listing will include output requested with
the EDOUT command.) Use EDHIST,DELE to delete the time-history output specification.
Jobname.HIS is a binary file that is read by the ANSYS time-history postprocessor (POST26). If LS-DYNA
output has been requested on the EDWRITE command [EDWRITE,LSDYNA or EDWRITE,BOTH], the file
D3THDT will also be written. D3THDT is a binary file that is read by the LS-POST postprocessor.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Output Controls>Select Component
EDHTIME, NSTEP, DT
Specifies the time-history output interval for an explicit dynamics analysis.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
NSTEP
Number of steps at which output is written to the time-history file, Jobname.HIS, and the ASCII output
files. Defaults to 1000. The time increment between output is TIME / NSTEP, where TIME is the analysis
end-time specified on the TIME command.
DT
Time interval at which output is written to the time-history file, Jobname.HIS, and the ASCII output
files. If NSTEP is input, DT is ignored.
Command Default
Time-history output is written at 1000 steps over the analysis.
Notes
EDHTIME controls the number of steps at which output will be written to the time-history file, Jobname.HIS
(see the EDHIST command), and any ASCII files requested on the EDOUT command. You can use NSTEP
or DT to specify the output interval. You should not specify both quantities; if both are input, NSTEP will
be used.
457
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EDHTIME
In an explicit dynamic small restart (EDSTART,2) or full restart analysis (EDSTART,3), the EDHTIME setting
will default to the NSTEP or DT value used in the original analysis. You can issue EDHTIME in the restart
to change this setting.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Output Controls>File Output Freq>Number of Steps
Main Menu>Solution>Output Controls>File Output Freq>Time Step Size
EDINT, SHELLIP, BEAMIP
Specifies number of integration points for explicit shell and beam output.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
SHELLIP
Number of shell integration points used for output (defaults to 3). For element SHELL163, each integration
point is associated with a layer. SHELLIP must be
_
,
_
,
exp exp
1
where f
n
and f
s
are normal and shear interface forces. Normal interface force f
n
is nonzero for tensile values
only.
You can graphically display spotwelds by issuing the command /PBC,WELD,,1.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Spotweld>Delete
Main Menu>Preprocessor>LS-DYNA Options>Spotweld>Genrlizd Spotwld
Main Menu>Preprocessor>LS-DYNA Options>Spotweld>List
Main Menu>Preprocessor>LS-DYNA Options>Spotweld>Massless Spotwld
EDWRITE, Option, Fname, Ext, --
Writes explicit dynamics input to an LS-DYNA input file.
SOLUTION: Explicit Dynamics
<> <> <> <> <> <> <> <> <> <> DY <> <> <> <>
Option
Sets a flag in the LS-DYNA input file (Fname.Ext) to produce desired output.
ANSYS
Set a flag to write results files for the ANSYS postprocessors (default). The files that will be written
are Jobname.RST and Jobname.HIS (see Notes below).
LSDYNA
Set a flag to write results files for the LS-DYNA postprocessor (LS-POST). The files that will be written
are D3PLOT, and files specified by EDOUT and EDHIST (see Notes below).
BOTH
Set a flag to write results files for both ANSYS and LS-DYNA postprocessors.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
500
EDWRITE
Fname
File name and directory path (80 characters maximum, including directory; this limit is due to an LS-
DYNA program limitation). If you do not specify a directory path, it will default to your working directory.
The file name defaults to Jobname. Previous data on this file, if any, are overwritten.
Ext
Filename extension (8 character maximum).
The extension defaults to K in a new analysis and in a full restart analysis, and to R in a small restart
analysis.
--
Unused field.
Notes
This command writes an LS-DYNA input file for the LS-DYNA solver. EDWRITE is only valid if explicit dynamic
elements have been specified. This command is not necessary if the LS-DYNA solver is invoked from within
ANSYS, in which case Jobname.K (or Jobname.R) is written automatically when the solution is initiated.
(If LS-DYNA is invoked from within ANSYS, use EDOPT to specify desired output.)
If the analysis is a small restart (EDSTART,2), the file that is written will have the name Jobname.R (by
default) and will only contain changes from the original analysis.
If the analysis is a full restart (EDSTART,3), the file that is written will have the name Jobname_nn.K (by
default) and will contain all the information from the database. In a full restart, the jobname is changed to
Jobname_nn (nn = 01 initially, and is incremented for each subsequent full restart.)
A command is included in the LS-DYNA input file to instruct the LS-DYNA solver to write the results files
indicated by Option. By default, LS-DYNA will write the ANSYS results file Jobname.RST (see the EDRST
command). If Jobname.HIS is desired, you must also issue EDHIST.
Option = LSDYNA or BOTH will cause LS-DYNA to write results files for the LS-POST postprocessor. The
D3PLOT file is always written for these two options. If other LS-POST files are desired, you must issue the
appropriate EDHIST and EDOUT commands.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Write Jobname.k
/EFACET, NUM
Specifies the number of facets per element edge for PowerGraphics displays.
POST1: Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NUM
Number of facets per element edge for element plots.
501
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/EFACET
1
Use 1 facet per edge (default for h-elements).
2
Use 2 facets per edge (default for p-elements).
4
Use 4 facets per edge.
Command Default
As stated above.
Notes
/EFACET is valid only when PowerGraphics is enabled [/GRAPHICS,POWER], except that it can be used in
FULL graphics mode for element CONTA174. (See the /GRAPHICS command and element CONTA174 in the
Element Reference for more information.) The /EFACET command is only applicable to element type displays.
/EFACET controls the fineness of the subgrid that is used for element plots. The element is subdivided into
smaller portions called facets. Facets are piecewise linear surface approximations of the actual element face.
In their most general form, facets are warped planes in 3-D space. A greater number of facets will result in
a smoother representation of the element surface for element plots. /EFACET may affect results averaging.
See Contour Displays in the Basic Analysis Guide for more information.
For midside node elements, use NUM = 2; if NUM = 1, no midside node information is output. For non-midside
node elements, NUM should be set to 1. See the PLNSOL and PRNSOL commands for more information.
With PowerGraphics active (/GRAPHICS,POWER), the averaging scheme for surface data with interior element
data included (AVRES,,FULL) and multiple facets per edge (/EFACET,2 or /EFACET,4) will yield differing
minimum and maximum contour values depending on the Z-Buffering options (/TYPE,,6 or /TYPE,,7). When
the Section data is not included in the averaging schemes (/TYPE,,7), the resulting absolute value for the
midside node is significantly smaller.
Caution
If you specify /EFACET,1, PowerGraphics does not plot midside nodes. You must use /EFACET,2 to
make the nodes visible.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>Options for Outp
Main Menu>General Postproc>Plot Results>Contour Plot>Nodal Solu
Utility Menu>List>Results>Options
Utility Menu>Plot>Results>Contour Plot>Nodal Solution
Utility Menu>PlotCtrls>Multi-Plot Contrls
Utility Menu>PlotCtrls>Style>Size and Shape
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
502
/EFACET
EGEN, ITIME, NINC, IEL1, IEL2, IEINC, MINC, TINC, RINC, CINC, SINC, DX, DY, DZ
Generates elements from an existing pattern.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME, NINC
Do this generation operation a total of ITIMEs, incrementing all nodes in the given pattern by NINC
each time after the first. ITIME must be >1 if generation is to occur. NINC may be positive, zero, or
negative. If DX, DY, and/or DZ is specified, NINC should be set so any existing nodes (as on NGEN) are
not overwritten.
IEL1, IEL2, IEINC
Generate elements from selected pattern beginning with IEL1 to IEL2 (defaults to IEL1) in steps of
IEINC (defaults to 1). If IEL1 is negative, IEL2 and IEINC are ignored and the last |IEL1| elements
(in sequence backward from the maximum element number) are used as the pattern to be repeated. If
IEL1 = ALL, IEL2 and IEINC are ignored and use all selected elements [ESEL] as pattern to be repeated.
If P1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the
GUI). A component name may also be substituted for IEL1 (IEL2 and INC are ignored).
MINC
Increment material number of all elements in the given pattern by MINC each time after the first.
TINC
Increment type number by TINC.
RINC
Increment real constant table number by RINC.
CINC
Increment element coordinate system number by CINC.
SINC
Increment section ID number by SINC.
DX, DY, DZ
Define nodes that do not already exist but are needed by generated elements (as though the
NGEN,ITIME,INC,NODE1,,,DX,DY,DZ were issued before EGEN). Zero is a valid value. If blank, DX,
DY, and DZ are ignored.
Notes
A pattern may consist of any number of previously defined elements. The MAT, TYPE, REAL, ESYS, and
SECNUM numbers of the new elements are based upon the elements in the pattern and not upon the current
specification settings.
You can use the EGEN command to generate interface elements (INTER192, INTER193, INTER194, and
INTER195) directly. However, because interface elements require that the element connectivity be started
from the bottom surface, you must make sure that you use the correct element node connectivity. See the
element descriptions for INTER192, INTER193, INTER194, and INTER195 for the correct element node definition.
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Elements>Auto Numbered
503
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EGEN
EINTF, TOLER, K, TLAB, KCN, DX, DY, DZ, KNONROT
Defines two-node elements between coincident or offset nodes.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
TOLER
Tolerance for coincidence (based on maximum Cartesian coordinate difference for node locations and
on angle differences for node orientations). Defaults to 0.0001. Only nodes within the tolerance are
considered to be coincident.
K
Only used when the type of the elements to be generated is PRETS179. K is the pretension node that
is common to the pretension section that is being created. If K is not specified, it will be created by
ANSYS automatically and will have an ANSYS-assigned node number. If K is specified but does not already
exist, it will be created automatically but will have the user-specified node number. K cannot be connected
to any existing element.
TLAB
Nodal number ordering. Allowable values are:
LOW
The 2-node elements are generated from the lowest numbered node to the highest numbered node.
HIGH
The 2-node elements are generated from the highest numbered node to the lowest numbered node.
REVE
Reverses the orientation of the selected 2-node element.
KCN
In coordinate system KCN, elements are created between node 1 and node 2 (= node 1 + dx dy dz)..
DX, DY, DZ
Node location increments that define the node offset in the active coordinate system (DR, D, DZ for
cylindrical and DR, D, D for spherical or toroidal).
KNONROT
When KNONROT = 0, the nodes coordinate system is not rotated. When KNONROT = 1, the nodes belong-
ing to the elements created are rotated into coordinate system KCN (see NROTAT command description).
Notes
Defines 2-node elements (such as gap elements) between coincident or offset nodes (within a tolerance).
May be used, for example, to "hook" together elements interfacing at a seam, where the seam consists of a
series of node pairs. One element is generated for each set of two coincident nodes. For more than two
coincident or offset nodes in a cluster, an element is generated from the lowest numbered node to each of
the other nodes in the cluster. If fewer than all nodes are to be checked for coincidence, use the NSEL
command to select the nodes. Element numbers are incremented by one from the highest previous element
number. The element type must be set [ET] to a 2-node element before issuing this command. Use the
CPINTF command to connect nodes by coupling instead of by elements. Use the CEINTF command to
connect the nodes by constraint equations instead of by elements.
For contact element CONTA178, the tolerance is based on the maximum Cartesian coordinate difference for
node locations only. The angle differences for node orientations are not checked.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
504
EINTF
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Auto Numbered>At Coincid Nd
Main Menu>Preprocessor>Modeling>Create>Elements>Auto Numbered>Offset Nodes
EKILL, ELEM
Deactivates an element (for the birth and death capability).
SOLUTION: Birth and Death
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
ELEM
Element to be deactivated. If ALL, deactivate all selected elements [ESEL]. If ELEM = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for ELEM.
Notes
Deactivates the specified element when the birth and death capability is being used. A deactivated element
remains in the model but contributes a near-zero stiffness (or conductivity, etc.) value (ESTIF) to the overall
matrix. Any solution-dependent state variables (such as stress, plastic strain, creep strain, etc.) are set to
zero. Deactivated elements contribute nothing to the overall mass (or capacitance, etc.) matrix.
The element can be reactivated with the EALIVE command.
ANSYS recommends using element deactivation/reactivation (EKILL/EALIVE) for linear elastic materials only.
For all other materials, validate the results carefully before using them.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Birth & Death>Kill Elements
Main Menu>Solution>Load Step Opts>Other>Birth & Death>Kill Elements
ELBOW, Transkey, TOL, Dof, Cons1, Cons2, Cons3, Cons4
Specifies degrees of freedom to be coupled for end release and applies section constraints to elbow ele-
ments.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Transkey
Pipe-to-elbow transition flag:
OFF
Do not automatically transition pipes to elbows. (This behavior is the default.)
505
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ELBOW
ON
Automatically convert straight PIPE289 elements to ELBOW290 elements where it is beneficial. The
program converts elements in transition regions where curved ELBOW290 elements are connected
to straight PIPE289 elements.
TOL
Angle tolerance (in degrees) between adjacent ELBOW290 elements. The default value is 20. A value of
-1 specifies all selected ELBOW290 elements.
Dof
Degrees of freedom to couple:
ALL
Couple all nodal degrees of freedom (UX, UY, UZ, ROTX, ROTY, and ROTZ). This behavior is the default.
BALL
Create ball joints (equivalent to releasing ROTX, ROTY, and ROTZ).
Cons1, Cons2, Cons3, Cons4
Section degrees of freedoms to constrain. If Cons1 through Cons4 are unspecified, no section constraints
are applied:
SECT
All section deformation
SE
Section radial expansion
SO
Section ovalization
SW
Section warping
SRA
Local shell normal rotation about cylindrical axis t2
SRT
Local shell normal rotation about cylindrical axis t1
Notes
The ELBOW command specifies end releases and section constraints for ELBOW290 elements and converts
straight PIPE289 elements to ELBOW290 elements. The command works on currently selected nodes and
elements. It creates end releases on any two connected elbow elements whose angle at connection exceeds
the specified tolerance. From within the ANSYS GUI, the Picked node option generates an end release and
section constraints at the selected node regardless of the angle of connection (that is, the angle tolerance
[TOL ] is set to -1).
Elbow and pipe elements must share the same section ID in order for the pipe-to-elbow transition to occur.
To list the elements altered by the ELBOW command, issue an ELIST command.
To list the coupled sets generated by the ELBOW command, issue a CPLIST command.
To list the section constraints generated by the ELBOW command, issue a DLIST command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
506
ELBOW
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>End Releases>Elbows>On Selected set
Main Menu>Solution>Load Step Opts>Other>End Releases>Elbows>On Selected set
ELEM
Specifies "Elements" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Elements
ELIST, IEL1, IEL2, INC, NNKEY, RKEY, PTKEY
Lists the elements and their attributes.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IEL1, IEL2, INC
Lists elements from IEL1 to IEL2 (defaults to IEL1) in steps of INC (defaults to 1). If IEL1 = ALL
(default), IEL2 and INC are ignored and all selected elements [ESEL] are listed. If IEL1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for IEL1 (IEL2 and INC are ignored).
NNKEY
Node listing key:
0
List attribute references and nodes.
1
List attribute references but not nodes.
RKEY
Real constant listing key:
0
Do not show real constants for each element.
507
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ELIST
1
Show real constants for each element. This includes default values chosen for the element.
PTKEY
LS-DYNA part number listing key (applicable to ANSYS LS-DYNA only):
0
Do not show part ID number for each element.
1
Show part ID number for each element.
Notes
Lists the elements with their nodes and attributes (MAT, TYPE, REAL, ESYS, SECNUM, PART). See also the
LAYLIST command for listing layered elements.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Elements>Attributes + RealConst
Utility Menu>List>Elements>Attributes Only
Utility Menu>List>Elements>Nodes + Attributes
Utility Menu>List>Elements>Nodes + Attributes + RealConst
*ELSE
Separates the final if-then-else block.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Optional final block separator within an if-then-else construct. See the *IF for details. If a batch input stream
hits an end-of-file during a false *IF condition, the ANSYS run will not terminate normally. You will need to
terminate it externally (use either the UNIX kill function or the Windows task manager). The *ELSE command
must appear on the same file as the *IF command, and all five characters must be input.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
508
*ELSE
*ELSEIF, VAL1, Oper1, VAL2, Conj, VAL3, Oper2, VAL4
Separates an intermediate if-then-else block.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
VAL1
First numerical value (or parameter which evaluates to numerical value) in the conditional comparison
operation. VAL1, VAL2, VAL3, and VAL4 can also be character strings (enclosed in quotes) or parameters
for Oper = EQ and NE only.
Oper1
Operation label. A tolerance of 1.0E-10 is used for comparisons between real numbers:
EQ --
Equal (for VAL1 = VAL2).
NE --
Not equal (for VAL1 VAL2).
LT --
Less than (for VAL1<VAL2).
GT --
Greater than (for VAL1>VAL2).
LE --
Less than or equal (for VAL1
VAL2).
GE --
Greater than or equal (for VAL1
VAL2).
ABLT --
Absolute values of VAL1 and VAL2 before < operation.
ABGT --
Absolute values of VAL1 and VAL2 before > operation.
VAL2
Second numerical value (or parameter which evaluates to numerical value) in the conditional comparison
operation.
Conj
(Optional) Connection between two logical clauses.
AND -
True if both clauses (Oper1 and Oper2) are true.
OR -
True if either clause is true.
XOR -
True if either (but not both) clause is true.
VAL3
(Optional) Third numerical value (or parameter which evaluates to numerical value).
509
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*ELSEIF
Oper2
(Optional) Operation label. This will have the same labels as Oper1, except it uses Val3 and Val4. A
tolerance of 1.0E-10 is used for comparisons between real numbers.
VAL4
(Optional) Fourth numerical value (or parameter value which evaluates to a numerical value).
Notes
Optional intermediate block separator within an if-then-else construct. All seven characters of the command
name (*ELSEIF) must be input. This command is similar to the *IF command except that the Base field is
not used. The *IF, *ELSEIF, *ELSE, and *ENDIF commands for each if-then-else construct must all be read
from the same file (or keyboard).
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
EMAGERR
Calculates the relative error in an electrostatic or electromagnetic field analysis.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
The relative error is an approximation of the mesh discretization error associated with a solution. It is based
on the discrepancy between the unaveraged, element-nodal field values and the averaged, nodal field values.
The calculation is valid within a material boundary and does not consider the error in continuity of fields
across dissimilar materials.
For electrostatics, the field values evaluated are the electric field strength (EFSUM) and the electric flux
density (DSUM). A relative error norm of each is calculated on a per-element basis and stored in the element
table [ETABLE] with the labels EF_ERR and D_ERR. Normalized error values EFN_ERR and DN_ERR are also
calculated and stored in the element table. Corresponding quantities for electromagnetics are H_ERR, B_ERR,
HN_ERR, and BN_ERR, which are calculated from the magnetic field intensity (HSUM) and the magnetic flux
density (BSUM). The normalized error value is the relative error norm value divided by the peak element-
nodal field value for the currently selected elements.
Use the PLETAB and PRETAB commands to plot and list the error norms and normalized error values.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Element Based>Error Eval
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
510
EMAGERR
EMATWRITE, Key
Forces the writing of all the element matrices to File.EMAT.
SOLUTION: Analysis Options
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
Key
Write key:
YES
Forces the writing of the element matrices to File.EMAT even if not normally done.
NO
Element matrices are written only if required. This value is the default.
Notes
The EMATWRITE command forces ANSYS to write the File.EMAT file. The file is necessary if you intend
to perform a subsequent partial solution (PSOLVE) requiring File.EMAT data as a prerequisite, or if you
intend to follow the initial load step with a subsequent inertia relief calculation (IRLF).
File.EMAT is also required in these cases:
For the static solution of a prestressed modal analysis when you use stress-stiffening energy to obtain
better strain energy calculation accuracy.
When you perform a QR damp eigensolution with a constant material damping coefficient (MP,DMPR).
This command is also valid in PREP7.
Menu Paths
This command cannot be accessed from a menu.
EMF
Calculates the electromotive force (emf), or voltage drop along a predefined path.
POST1: Magnetics Calculations
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS
Notes
EMF invokes an ANSYS macro which calculates the electromotive force (emf ), or voltage drop along a pre-
defined path (specified with the PATH command). It is valid for both 2-D and 3-D electric field analysis or
high-frequency electromagnetic field analysis. The calculated emf value is stored in the parameter EMF.
You must define a line path (via the PATH command) before issuing the EMF command macro. The macro
uses calculated values of the electric field (EF), and uses path operations for the calculations. All path items
are cleared when the macro finishes executing.
The EMF macro sets the "ACCURATE" mapping method and "MAT" discontinuity option on the PMAP com-
mand. The ANSYS program retains these settings after executing the macro.
511
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EMF
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>EMF
EMFT
Summarizes electromagnetic forces and torques.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
Use this command to summarize electromagnetic force and torque in both static electric and magnetic
problems. To use this command, select the nodes in the region of interest and make sure that all elements
are selected. If RSYS = 0, the force is reported in the global Cartesian coordinate system. If RSYS 0, force
is reported in the specified coordinate system. However, for torque, if RSYS 0, this command will account
for the shift and rotation as specified by RSYS, but will report only the Cartesian components.
Forces are stored as items _FXSUM, _FYSUM, _FZSUM, and _FSSUM. Torque is stored as items _TXSUM,
_TYSUM, _TZSUM, and _TSSUM.
This command is valid only with SOLID117, PLANE121, SOLID122, SOLID123, SOLID236 and SOLID237 elements.
For any other elements, you must use FMAGSUM.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu> General Postprocessor> Elec&Mag Calc> Summarize Force/Torque
EMID, Key, Edges
Adds or removes midside nodes.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM <> DY PP <> EME MFS
Key
Add or remove key:
ADD
Add midside node to elements (default).
REMOVE
Remove midside nodes from elements.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
512
EMFT
Edges
ALL
Add (or remove) midside nodes to (from) all edges of all selected elements, independent of which
nodes are selected (default).
EITHER
Add (or remove) midside nodes only to (from) element edges which have either corner node selected.
BOTH
Add (or remove) midside nodes only to (from) element edges which have both corner nodes selected.
Notes
This command adds midside nodes to (or removes midside nodes from) the selected elements. For this to
occur, the selected elements must be midside node capable, the active element type [TYPE] must allow
midside node capability, and the relationship between the finite element model and the solid model (if any)
must first be disassociated [MODMSH].
By default, EMID generates a midside node wherever a zero (or missing) midside node occurs for that element.
You can control this and add (or remove) midside nodes selectively by using the Edges argument. Nodes
are located midway between the two appropriate corner nodes based on a linear Cartesian interpolation.
Nodal coordinate system rotation angles are also linearly interpolated. Connected elements share the same
midside node. Node numbers are generated sequentially from the maximum node number.
The EMID command is useful for transforming linear element types to quadratic element types having the
same corner node connectivity (for example, by changing the element type from PLANE42 to PLANE82 [ET]
and then issuing the EMID command). EMID is also useful for transforming elements created outside of the
ANSYS program.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Elements>Add Mid Nodes
Main Menu>Preprocessor>Modeling>Move / Modify>Elements>Remove Mid Nd
EMIS, MAT, EVALU
Specifies emissivity as a material property for the Radiation Matrix method.
AUX12: Radiation Matrix Method
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
MAT
Material number associated with this emissivity (500 maximum). Defaults to 1.
EVALU
Emissivity for this material (0.0 < EVALU
1.0). Enter a very small number for zero.
Command Default
Emissivity value of 1.0 is associated with all materials.
513
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EMIS
Notes
Specifies emissivity as a material property for the Radiation Matrix method. This material property can then
be associated with each element.
Menu Paths
Main Menu>Radiation Opt>Matrix Method>Emissivities
EMODIF, IEL, STLOC, I1, I2, I3, I4, I5, I6, I7, I8
Modifies a previously defined element.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IEL
Modify nodes and/or attributes for element number IEL. If ALL, modify all selected elements [ESEL]. If
IEL = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the
GUI). A component name may also be substituted for IEL.
STLOC
Starting location (n) of first node to be modified or the attribute label. If n, modify element node positions
n, n+1, etc. (n = 1 to 20). For example, if STLOC = 1, I1 refers to the first node, I2, the second, etc. If
STLOC = 9, I1 refers to the ninth node, I2, the tenth, etc. Attributes are also modified to the currently
specified values (use -n to modify only nodes and not attributes). If zero, modify only the attributes to
the currently specified values. If MAT, TYPE, REAL, ESYS, or SECNUM, modify only that attribute to the
I1 value.
I1, I2, I3, . . . , I8
Replace the previous node numbers assigned to this element with these corresponding values. A (blank)
retains the previous value (except in the I1 field, which resets the STLOC node number to zero). For
attributes, replace the existing value with the I1 value (or the default if I1 is zero or blank).
Notes
The nodes and/or attributes (MAT, TYPE, REAL, ESYS, and SECNUM values) of an existing element may be
changed with this command.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Elements>Modify Attrib
Main Menu>Preprocessor>Modeling>Move / Modify>Elements>Modify Nodes
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
514
EMODIF
EMORE, Q, R, S, T, U, V, W, X
Adds more nodes to the just-defined element.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Q, R, S, T, U, V, W, X
Numbers of nodes typically assigned to ninth (node Q) through sixteenth (node X) nodal positions, if
any. If Q = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI).
Notes
Repeat EMORE command for up to 4 additional nodes (20 maximum). Nodes are added after the last nonzero
node of the element. Node numbers defined with this command may be zeroes.
Menu Paths
This command cannot be accessed from a menu.
EMSYM, NSECT
Specifies circular symmetry for electromagnetic sources.
PREP7: Special Purpose
MP ME ST <> <> <> <> <> EM EH DY PP <> EME MFS
NSECT
The number of circular symmetry sections (defaults to 1).
Notes
Specifies the number of times to repeat electromagnetic sources for circular symmetry. Applies to SOURC36
elements and to coupled-field elements with electric current conduction results in the database. Sources
are assumed to be equally spaced over 360 about the global Cartesian Z axis.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Magnetics>Options Only>Copy Sources
Main Menu>Solution>Load Step Opts>Magnetics>Options Only>Copy Sources
515
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EMSYM
EMTGEN, Ncomp, Ecomp, PNcomp, DOF, GAP, GAPMIN, FKN, EPZERO
Generates a set of TRANS126 elements.
PREP7: Elements
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Ncomp
Component name of the surface nodes of a structure which attach to the TRANS126 elements. You must
enclose name-strings in single quotes in the EMTGEN command line.
Ecomp
Component name of the TRANS126 elements generated. You must enclose name-strings in single quotes
in the EMTGEN command line. Defaults to EMTELM.
PNcomp
Component name of the plane nodes generated by the command at an offset (GAP) from the surface
nodes. You must enclose name-strings in single quotes in the EMTGEN command line. Defaults to
EMTPNO.
DOF
Active structural degree of freedom (DOF) for TRANS126 elements (UX, UY, or UZ) in the Cartesian co-
ordinate system. You must enclose the DOF in single quotes.
GAP
Initial gap distance from the surface nodes to the plane. Be sure to use the correct sign with respect to
Ncomp node location.
GAPMIN
Minimum gap distance allowed (GAPMIN real constant) for TRANS126 elements. Defaults to the absolute
value of (GAP)*0.05.
FKN
Contact stiffness factor used as a multiplier for a contact stiffness appropriate for bulk deformation. De-
faults to 0.1.
EPZERO
Free-space permittivity. Defaults to 8.854e-6 (MKS units).
Notes
The EMTGEN command generates a set of TRANS126 elements between the surface nodes of a moveable
structure and a plane of nodes, typically representing a ground plane. The plane of nodes are created by
the command at a specified offset (GAP). Each element attaches to a surface node and to a corresponding
node representing the plane. The created elements are set to the augmented stiffness method (KEYOPT(6)
= 1), which can help convergence. The generated plane nodes should be constrained appropriately for the
analysis.
You can use TRANS126 elements for simulating fully coupled electrostatic structural coupling between a
MEMS device and a plane, if the gap distance between the device and the plane is small compared to the
overall surface area dimensions of the device. This assumption allows for a point-wise closed-form solution
of capacitance between the surface nodes and the plane; i.e. CAP = EPZERO*AREA/GAP, where EPZERO if
the free-space permittivity, AREA is the area associated with the node, and GAP is the gap between the
node and the plane. The area for each node is computed using the ARNODE function in ANSYS. See the
*GET command description for more information on the ARNODE function.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
516
EMTGEN
With a distributed set of TRANS126 elements attached directly to the structure and a plane (such as a ground
plane), you can perform a full range of coupled electrostatic-structural simulations, including:
Static analysis (due to a DC voltage or a mechanical load)
Prestressed modal analysis (eigenfrequencies, including frequency-shift effects of a DC bias voltage)
Prestressed harmonic analysis (system response to a small-signal AC excitation with a DC bias voltage
or mechanical load)
Large signal transient analysis (time-transient solution due to an arbitrary time-varying voltage or
mechanical excitation)
The TRANS126 element also employs a node-to-node gap feature so you can perform contact-type simulations
where the structure contacts a plane (such as a ground plane). The contact stiffness factor, FKN, is used to
control contact penetration once contact is initiated. A smaller value provides for easier convergence, but
with more penetration.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Transducers>Electromechanic>Node to Plane
EMUNIT, Lab, VALUE
Specifies the system of units for magnetic field problems.
PREP7: Materials
MP ME <> <> <> <> <> <> EM EH <> PP <> EME MFS
Lab
Label specifying the type of units:
MKS
Rationalized MKS system of units (meters, amperes, henries, webers, etc.). Free-space permeability
is set to 4 e-7 henries/meter. Free-space permittivity is set to 8.85 e-12 F/m.
MUZRO
User defined system of units. Free-space permeability is set to the value input for VALUE. Other units
must correspond to the permeability units. Relative permeability may be altered to absolute values.
EPZRO
User defined system of units. Free-space permittivity is set to the value input for VALUE. Other units
must correspond to the permittivity units.
VALUE
User value of free-space permeability (defaults to 1) if Lab = MUZRO, or free-space permittivity (defaults
to 1) if Lab = EPZRO.
Command Default
Rationalized MKS system of units (meters, amperes, henries, webers, etc.). Free-space permeability is set to
4 e-7 Henries/meter, free-space permittivity is set to 8.85e-12 Farads/meter.
517
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EMUNIT
Notes
Specifies the system of units to be used for electric and magnetic field problems. The free-space permeability
and permittivity values may be set as desired. These values are used with the relative property values [MP]
to establish absolute property values.
Note
If the magnetic source field strength (H
s
) has already been calculated [BIOT], switching EMUNIT
will not change the values.
For micro-electromechanical systems (MEMS), where dimensions are on the order of microns, see the con-
version factors in System of Units in the Coupled-Field Analysis Guide.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Electromag Units
Main Menu>Preprocessor>Material Props>Electromag Units
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Electromag Units
EN, IEL, I, J, K, L, M, N, O, P
Defines an element by its number and node connectivity.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IEL
Number assigned to element being defined. If IEL = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI).
I
Number of node assigned to first nodal position (node I).
J, K, L, M, N, O, P
Number assigned to second (node J) through eighth (node P) nodal position, if any.
Notes
Defines an element by its nodes and attribute values. Similar to the E command except it allows the element
number (IEL) to be defined explicitly. Element numbers need not be consecutive. Any existing element
already having this number will be redefined.
Up to 8 nodes may be specified with the EN command. If more nodes are needed for the element, use the
EMORE command. The number of nodes required and the order in which they should be specified are de-
scribed in the Element Reference for each element type. The current (or default) MAT, TYPE, REAL, SECNUM,
and ESYS attribute values are also assigned to the element.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
518
EN
When creating elements with more than 8 nodes using this command and the EMORE command, it may
be necessary to turn off shape checking using the SHPP command before issuing this command. If a valid
element type can be created without using the additional nodes on the EMORE command, this command
will create that element. The EMORE command will then modify the element to include the additional nodes.
If shape checking is active, it will be performed before the EMORE command is issued. Therefore, if the
shape checking limits are exceeded, element creation may fail before the EMORE command modifies the
element into an acceptable shape.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>User Numbered>Thru Nodes
*END
Closes a macro file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Closes a file opened with *CREATE. The *END command is an 8-character command (to differentiate it from
*ENDIF). If you add commented text on that same line but do not allow enough spaces between *END and
the "!" that indicates the comment text, the *END will attempt to interpret the "!" as the 8th character and
will fail.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
*ENDDO
Ends a do-loop and starts the looping action.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
One *ENDDO is required for each nested do-loop. The *ENDDO command must appear on the same file as
the *DO command, and all six characters must be input.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
519
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*ENDDO
*ENDIF
Ends an if-then-else.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Required terminator for the if-then-else construct. See the *IF for details. If a batch input stream hits an end-
of-file during a false *IF condition, the ANSYS run will not terminate normally. You will need to terminate it
externally (use either the UNIX kill function or the Windows task manager). The *ENDIF command must
appear on the same file as the *IF command, and all six characters must be input.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
ENDRELEASE, --, TOLERANCE, Dof1, Dof2, Dof3, Dof4
Specifies degrees of freedom to be decoupled for end release.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
--
Unused field.
TOLERANCE
Angle tolerance (in degrees) between adjacent elements. Defaults to 20. Set TOLERANCE to -1 to indicate
all selected elements.
Dof1, Dof2, Dof3, Dof4
Degrees of freedom to release. If Dof1 is blank, WARP is assumed and Dof2, Dof3, and Dof4 are ignored.
WARP
Release the warping degree of freedom (default).
ROTX
Release rotations in the X direction.
ROTY
Release rotations in the Y direction.
ROTZ
Release rotations in the Z direction.
UX
Release displacements in the X direction.
UY
Release displacements in the Y direction.
UZ
Release displacements in the Z direction.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
520
*ENDIF
BALL
Create ball joints (equivalent to releasing WARP, ROTX, ROTY, and ROTZ).
Notes
This command specifies end releases for the BEAM188, BEAM189, PIPE288, and PIPE289 elements. The
command works on currently selected nodes and elements. It creates end releases on any two connected
beam elements whose angle at connection exceeds the specified tolerance. From within the GUI, the Picked
node option generates an end release at the selected node regardless of the angle of connection (angle
tolerance is set to -1).
Use the CPLIST command to list the coupled sets generated by the ENDRELEASE command.
Note
You should exercise due engineering judgement when using this command, as improper use
may result in mechanics that render a solution impossible.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>End Releases>Beams>On Selected set
Main Menu>Solution>Load Step Opts>Other>End Releases>Beams>On Selected set
ENERSOL, NVAR, Item, --, Name
Specifies the total energies to be stored.
POST26: Set Up
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NVAR
Arbitrary reference number assigned to this variable (2 to NV).
Item
SENE
Stiffness energy
KENE
Kinetic energy
AENE
Artificial energy due to hourglass control/drill stiffness
STEN
Artificial energy due to nonlinear stabilization
--
Unused field.
NAME
A 32-character name identifying the item on printouts and displays. Defaults to a 4-character label formed
by the four characters of the Item value.
521
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ENERSOL
Menu Paths
This command cannot be accessed from a menu.
ENGEN, IINC, ITIME, NINC, IEL1, IEL2, IEINC, MINC, TINC, RINC, CINC, SINC, DX, DY, DZ
Generates elements from an existing pattern.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IINC
Increment to be added to element numbers in pattern.
ITIME, NINC
Do this generation operation a total of ITIMEs, incrementing all nodes in the given pattern by NINC
each time after the first. ITIME must be > 1 if generation is to occur. NINC may be positive, zero, or
negative.
IEL1, IEL2, IEINC
Generate elements from the pattern that begins with IEL1 to IEL2 (defaults to IEL1) in steps of IEINC
(defaults to 1). If IEL1 is negative, IEL2 and IEINC are ignored and use the last |IEL1| elements (in
sequence backward from the maximum element number) as the pattern to be repeated. If IEL1 = ALL,
IEL2 and IEINC are ignored and all selected elements [ESEL] are used as the pattern to be repeated.
If IEL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI). A component name may also be substituted for IEL1 (IEL2 and IEINC are ignored).
MINC
Increment material number of all elements in the given pattern by MINC each time after the first.
TINC
Increment type number by TINC.
RINC
Increment real constant table number by RINC.
CINC
Increment element coordinate system number by CINC.
SINC
Increment section ID number by SINC.
DX, DY, DZ
Define nodes that do not already exist but are needed by generated elements
(NGEN,ITIME,INC,NODE1,,,DX,DY,DZ). Zero is a valid value. If blank, DX, DY, and DZ are ignored.
Notes
Same as the EGEN command except it allows element numbers to be explicitly incremented (IINC) from
the generated set. Any existing elements already having these numbers will be redefined.
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Elements>User Numbered
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
522
ENGEN
ENORM, ENUM
Reorients shell element normals or line element node connectivity.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
ENUM
Element number having the normal direction that the reoriented elements are to match. If ENUM = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
Notes
Reorients shell elements so that their outward normals are consistent with that of a specified element.
ENORM can also be used to reorder nodal connectivity of line elements so that their nodal ordering is
consistent with that of a specified element.
For shell elements, the operation reorients the element by reversing and shifting the node connectivity
pattern. For example, for a 4-node shell element, the nodes in positions I, J, K and L of the original element
are placed in positions J, I, L and K of the reoriented element. All 3-D shell elements in the selected set are
considered for reorientation, and no element is reoriented more than once during the operation. Only shell
elements adjacent to the lateral (side) faces are considered.
The command reorients the shell element normals on the same panel as the specified shell element. A
panel is the geometry defined by a subset of shell elements bounded by free edges or T-junctions (anywhere
three or more shell edges share common nodes).
Reorientation progresses within the selected set until either of the following conditions is true:
The edge of the model is reached.
More than two elements (whether selected or unselected) are adjacent to a lateral face.
In situations where unselected elements might undesirably cause case b to control, consider using EN-
SYM,0,,0,ALL instead of ENORM. It is recommended that reoriented elements be displayed and graphically
reviewed.
You cannot use the ENORM command to change the normal direction of any element that has a body or
surface load. We recommend that you apply all of your loads only after ensuring that the element normal
directions are acceptable.
Real constant values are not reoriented and may be invalidated by an element reversal.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Elements>Shell Normals
523
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ENORM
ENSYM, IINC, --, NINC, IEL1, IEL2, IEINC
Generates elements by symmetry reflection.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IINC
Increment to be added to element numbers in existing set.
--
Unused field.
NINC
Increment nodes in the given pattern by NINC.
IEL1, IEL2, IEINC
Reflect elements from pattern beginning with IEL1 to IEL2 (defaults to IEL1) in steps of IEINC
(defaults to 1). If IEL1 = ALL, IEL2 and IEINC are ignored and pattern is all selected elements [ESEL].
If IEL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI). A component name may also be substituted for IEL1 (IEL2 and IEINC are ignored).
Notes
This command is the same as the ESYM command except it allows explicitly assigning element numbers to
the generated set (in terms of an increment IINC). Any existing elements already having these numbers
will be redefined.
The operation generates a new element by incrementing the nodes on the original element, and reversing
and shifting the node connectivity pattern. For example, for a 4-node 2-D element, the nodes in positions
I, J, K and L of the original element are placed in positions J, I, L and K of the reflected element.
Similar permutations occur for all other element types. For line elements, the nodes in positions I and J of
the original element are placed in positions J and I of the reflected element. In releases prior to ANSYS 5.5,
no node pattern reversing and shifting occurred for line elements generated by ENSYM. To achieve the same
results as you did in releases prior to ANSYS 5.5, use the ENGEN command instead.
See the ESYM command for additional information about symmetry elements.
The ENSYM command also provides a convenient way to reverse shell element normals. If the IINC and
NINC argument fields are left blank, the effect of the reflection is to reverse the direction of the outward
normal of the specified elements. You cannot use the ENSYM command to change the normal direction of
any element that has a body or surface load. We recommend that you apply all of your loads only after en-
suring that the element normal directions are acceptable. Also note that real constants (such as nonuniform
shell thickness and tapered beam constants) may be invalidated by an element reversal. See Revising Your
Model in the Modeling and Meshing Guide for more information about controlling element normals.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Reverse Normals>of Shell Elems
Main Menu>Preprocessor>Modeling>Reflect>Elements>User Numbered
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
524
ENSYM
/EOF
Exits the file being read.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Causes an end-of-file exit when encountered on a switched file (see /INPUT, *USE, etc.). Commands are
then read continuing from the file that contained the file switching command (or from the terminal if the
switch was made from the terminal). Use the /EXIT command to terminate an ANSYS run. This command
cannot be used in a do-loop or if-then-else construct.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
EORIENT, Etype, Dir, TOLER
Reorients solid element normals.
PREP7: Meshing
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Etype
Specifies which elements to orient.
LYSL
Specifies that certain solid elements will be oriented (SOLID95 with KEYOPT(1) = 1, SOLID185 with
KEYOPT(3) = 1, SOLID186 with KEYOPT(3) = 1, and SOLSH190). This value is the default.
Dir
The axis and direction for orientation, or an element number. If Dir is set to a positive number (n), then
all eligible elements are oriented as similarly as possible to element n.
NEGX
The element face with the outward normal most nearly parallel to the element coordinate systems
negative x-axis is designated (reoriented) as face 1.
POSX
The element face with the outward normal most nearly parallel to the element coordinate systems
positive x-axis is designated (reoriented) as face 1.
NEGY
The element face with the outward normal most nearly parallel to the element coordinate systems
negative y-axis is designated (reoriented) as face 1. .
POSY
The element face with the outward normal most nearly parallel to the element coordinate systems
positive y-axis is designated (reoriented) as face 1.
525
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EORIENT
NEGZ
(Default) The element face with the outward normal most nearly parallel to the element coordinate
systems negative z-axis is designated (reoriented) as face 1.
POSZ
The element face with the outward normal most nearly parallel to the element coordinate systems
positive z-axis is designated (reoriented) as face 1.
TOLER
The maximum angle (in degrees) between the outward normal face and the target axis. Default is 90.0.
Lower TOLER values will reduce the number of faces that are considered as the basis of element reori-
entation.
Notes
EORIENT renumbers the element faces, designating the face most parallel to the XY plane of the element
coordinate system (set with ESYS) as face 1 (nodes I-J-K-L, parallel to the layers in layered elements). It cal-
culates the outward normal of each face and changes the node designation of the elements so the face
with a normal most nearly parallel with and in the same general direction as the target axis becomes face
1.
The target axis, defined by Dir, is either the negative or positive indicated axis or the outward normal of
face 1 of that element.
All SOLID185 Layered Structural Solid, SOLID186 Layered Structural Solid, and SOLSH190 solid shell elements
in the selected set are considered for reorientation.
After reorienting elements, you should always display and graphically review results using the /ESHAPE
command. When plotting models with many or symmetric layers, it may be useful to temporarily reduce
the number of layers to two, with one layer being much thicker than the other.
You cannot use EORIENT to change the normal direction of any element that has a body or surface load.
We recommend that you apply all of your loads only after ensuring that the element normal directions are
acceptable.
Prisms and tetrahedrals are also supported, within the current limitations of the SOLID95, SOLID185, SOLID186,
and SOLSH190 elements. (Layers parallel to the 4-node face of the prism are not supported.)
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Elements>Orient Normal
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
526
EORIENT
EPLOT
Produces an element display.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Produces an element display of the selected elements. In full graphics, only those elements faces with all of
their corresponding nodes selected are plotted. In PowerGraphics, all element faces of the selected element
set are plotted irrespective of the nodes selected. However, for both full graphics and PowerGraphics, adjacent
or otherwise duplicated faces of 3-D solid elements will not be displayed in an attempt to eliminate plotting
of interior facets. See the DSYS command for display coordinate system issues.
This command will display curvature in midside node elements when PowerGraphics is activated [/GRAPH-
ICS,POWER] and /EFACET,2 or /EFACET,4 are enabled. (To display curvature, two facets per edge is recom-
mended [/EFACET,2]). When you specify /EFACET,1, PowerGraphics does not display midside nodes. /EFACET
has no effect on EPLOT for non-midside node elements.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>CMS>CMS Superelements>By Picking
Main Menu>Solution>Time Controls>Time Step Prediction
Utility Menu>Plot>Elements
EQSLV, Lab, TOLER, MULT, --, KeepFile
Specifies the type of equation solver.
SOLUTION: Analysis Options
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS
Lab
Equation solver type:
SPARSE
Sparse direct equation solver. Applicable to real-value or complex-value symmetric and unsymmetric
matrices. Available only for STATIC, HARMIC (full method only), TRANS (full method only), SUBSTR,
and PSD spectrum analysis types [ANTYPE]. Can be used for nonlinear and linear analyses, especially
nonlinear analysis where indefinite matrices are frequently encountered. Well suited for contact
analysis where contact status alters the mesh topology. Other typical well-suited applications are:
(a) models consisting of shell/beam or shell/beam and solid elements (b) models with a multi-branch
structure, such as an automobile exhaust or a turbine fan. This is an alternative to iterative solvers
since it combines both speed and robustness. Generally, it requires memory comparable to the PCG
solver. When memory is limited, the solver works partly in and out-of-core memory, often without
much increase in CPU time.
When the sparse solver is selected in Distributed ANSYS, the distributed sparse solver is automatically
used instead. See the description for DSPARSE below for details.
527
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EQSLV
JCG
Jacobi Conjugate Gradient iterative equation solver, in-memory option. Available only for STATIC,
HARMIC (full method only), and TRANS (full method only) analysis types [ANTYPE]. Can be used for
structural and multiphysics applications. Applicable for symmetric, unsymmetric, complex, definite,
and indefinite matrices. Recommended for 3-D harmonic analyses in structural and multiphysics
applications. Efficient for heat transfer, electromagnetics, piezoelectrics, and acoustic field problems.
This solver can also be run in shared memory or distributed memory (Distributed ANSYS) mode. In
addition to the limitations listed above, this solver is available only for STATIC and TRANS (full
method) analyses where the stiffness is symmetric, and does not support the fast thermal option
(THOPT) when used in Distributed ANSYS.
Distributed ANSYS requires an HPC license. For more information, see the Distributed ANSYS Guide.
ICCG
Incomplete Cholesky Conjugate Gradient iterative equation solver. Available for STATIC, HARMIC (full
method only), and TRANS (full method only) analysis types [ANTYPE]. Can be used for structural and
multiphysics applications, and for symmetric, unsymmetric, complex, definite, and indefinite matrices.
The ICCG solver requires more memory than the JCG solver, but is more robust than the JCG solver
for ill-conditioned matrices.
QMR
Quasi-Minimal Residual iterative equation solver. Available for the HARMIC (full method only) analysis
type [ANTYPE]. Can be used for high-frequency electromagnetic applications, and for symmetric,
complex, definite, and indefinite matrices. The QMR solver is more stable than the ICCG solver.
PCG
Pre-conditioned Conjugate Gradient iterative equation solver (licensed from Computational Applica-
tions and Systems Integration, Inc.). Requires less disk file space than SPARSE and is faster for large
models. Useful for plates, shells, 3-D models, large 2-D models, p-method analysbes, and other
problems having symmetric, sparse, definite or indefinite matrices for nonlinear analysis. Requires
twice as much memory as JCG (/RUNST can be used to determine the space needed). Available only
for analysis types [ANTYPE] STATIC, TRANS (full method only), or MODAL (with PCG Lanczos option
only). Also available for the use pass of substructure analyses (MATRIX50). The PCG solver can robustly
solve equations with constraint equations (CE, CEINTF, CPINTF, and CERIG). This solver gives you
the option of using either:
the MSAVE command (used to obtain a considerable memory savings when applicable (see
"Notes" (p. 531) below)), and/or
the PRECISION command to specify single precision, thereby reducing the memory required
by between 30% and 70%
The PCG solver can handle ill-conditioned problems by using a higher level of difficulty (see PCGOPT).
Ill-conditioning arises from elements with high aspect ratios, contact, and plasticity.
This solver can also be run in shared memory or distributed memory (Distributed ANSYS) mode.
When used in Distributed ANSYS, this solver preserves all of the merits of the classic PCG solver. The
total sum of memory (summed total over the network or all processors) is about 30% more than the
shared-memory PCG solver. In addition to the limitations listed above, this solver does not support
the PRECISION command or p-elements when used in Distributed ANSYS.
Distributed ANSYS requires an HPC license. For more information, see the Distributed ANSYS Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
528
EQSLV
AMG
Algebraic Multigrid iterative equation solver. Available for STATIC analyses and TRANS (full method
only) analyses. The AMG solver is applicable to symmetric matrices. This solver is not available for
substructure analyses, either generation or use pass, or for models containing p-elements (which
switch to the sparse solver). It is very efficient for single-field structural analyses (where the solution
DOFs are combinations of UX, UY, UZ, ROTX, ROTY, and ROTZ). For applications such as single-field
thermal analyses (where the solution DOF is TEMP), the AMG solver is less efficient. Recommended
for ill-conditioned problems in which the ICCG solver would have difficulty converging; applicable
in both single- and multiprocessor environments. In terms of CPU time when used in a single-processor
environment, the AMG solver performs better than the ICCG solver for ill-conditioned problems, and
it delivers about the same level of performance for ordinary problems. In a multiprocessor environ-
ment, the AMG solver scales better than the ICCG solver on shared memory parallel machines. Also
handles indefinite matrix problems for nonlinear analyses. The AMG solver is part of ANSYS' High
Performance Computing, which is a separately-licensed product. For detailed information on multi-
processor solvers, see Improving ANSYS Performance and High Performance Computing in the ANSYS
Advanced Analysis Techniques Guide. See Starting an ANSYS Session from Command Level in the
Operations Guide for information on how to specify the HPCHPC add-on at ANSYS start-up.
ITER
Automatically chooses an iterative solver that is appropriate for the physics of the problem. The
tolerance for the iterative solver is automatically selected based on the accuracy level you choose
(see TOLER, below). This solver option is available only for electrostatic analyses, steady-state/tran-
sient thermal analyses and linear static/full transient structural analyses without superelements. If
you issue the automatic iterative solver selection command, but the appropriate conditions for the
selection are not met, then the program defaults to the sparse solver. See Using the Automatic Iter-
ative (Fast) Solver Option in the Basic Analysis Guide, as well as Structural Static Analysis in the
Structural Analysis Guide or Steady-State Thermal Analysis in the Thermal Analysis Guide, for correct
usage of this option.
DSPARSE
Distributed sparse direct equation solver. It is mathematically similar to the SPARSE solver, except it
is parallelized for distributed memory architecture. As a result, it can achieve over 6X speedup on a
12-processor system. However, its total memory usage is higher than the non-distributed SPARSE
solver because each core front is held in memory on each processor to achieve the distributed per-
formance. Like the SPARSE solver, this solver is well-suited to ill-conditioned nonlinear problems, as
well as to beam/shell structures. The DSPARSE solver is valid for real-value or complex-value symmetric
and unsymmetric matrices. System configuration affects the performance of this solver. If enough
physical memory is available, running this solver in the in-core memory mode achieves optimal
performance. The ideal configuration when using the out-of-core memory mode is to use a single
processor on multiple machines, spreading the I/O across the hard drives of each machine, assuming
that you are using a high-speed network such as Infiniband. See the DSPARSE solver description in
the Basic Analysis Guide for more details.
This solver is available only with Distributed ANSYS and requires an HPC license. For more information,
see the Distributed ANSYS Guide.
TOLER
Iterative solver tolerance value. Used only with the Jacobi Conjugate Gradient, Incomplete Cholesky
Conjugate Gradient, Pre-conditioned Conjugate Gradient, Quasi-Minimal Residual, and Algebraic Multigrid
equation solvers. For the PCG solver, the default is 1.0E-8. The value 1.0E-5 may be acceptable in many
situations. When using the PCG Lanczos mode extraction method, the default solver tolerance value is
1.0E-4. For the JCG and ICCG solvers with symmetric matrices, and for the AMG solver, the default is
1.0E-8. For the JCG and ICCG solvers with unsymmetric matrices, and for the QMR solver, the default is
529
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EQSLV
1.0E-6. Iterations continue until the SRSS norm of the residual is less than TOLER times the norm of the
applied load vector. For the PCG solver in the linear static analysis case, 3 error norms are used. If one
of the error norms is smaller than TOLER, and the SRSS norm of the residual is smaller than 1.0E-2,
convergence is assumed to have been reached. See Iterative Solver in the Theory Reference for the
Mechanical APDL and Mechanical Applications for details.
Note
When used with the Pre-conditioned Conjugate Gradient equation solver, TOLER can be
modified between load steps (this is typically useful for nonlinear analysis).
If used with the ITER label, you can specify an integer value from 1 to 5, which indicates the accuracy
level for convergence for the automatically-selected iterative solver. A value of 1 is the fastest level, and
5 is the most accurate level. ANSYS chooses the tolerance value for the automatic iterative solver based
on the input accuracy level.
If a Lev_Diff value of 5 is specified on the PCGOPT command (either program- or user-specified),
TOLER has no effect on the accuracy of the obtained solution from the PCG solver; a direct solver is
used when Lev_Diff = 5.
MULT
Multiplier (defaults to 2.5 for nonlinear analyses; 1.0 for linear analyses) used to control the maximum
number of iterations performed during convergence calculations. Used only with the Pre-conditioned
Conjugate Gradient equation solver (PCG). The maximum number of iterations is equal to the multiplier
(MULT) times the number of degrees of freedom (DOF). If MULT is input as a negative value, then the
maximum number of iterations is equal to abs(MULT). Iterations continue until either the maximum
number of iterations or solution convergence has been reached. In general, the default value for MULT
is adequate for reaching convergence. However, for ill-conditioned matrices (that is, models containing
elements with high aspect ratios or material type discontinuities) the multiplier may be used to increase
the maximum number of iterations used to achieve convergence. The recommended range for the
multiplier is 1.0
MULT
3.0. Normally, a value greater than 3.0 adds no further benefit toward conver-
gence, and merely increases time requirements. If the solution does not converge with 1.0
MULT
3.0,
or in less than 10,000 iterations, then convergence is highly unlikely and further examination of the
model is recommended. Rather than increasing the default value of MULT, consider increasing the level
of difficulty (Lev_Diff) on the PCGOPT command.
--
Unused field.
KeepFile
Determines whether files from a SPARSE solver run should be deleted or retained. Applies only to Lab
= SPARSE for static and full transient analyses.
DELE
Deletes all files from the SPARSE solver run, including the factorized file, .LN22, upon FINISH or
/EXIT (default).
KEEP
Retains all necessary files from the SPARSE solver run, including the .LN22 file, in the working dir-
ectory. Use this option to allow the sparse matrix files to be reused in a linear analysis in conjunction
with the KUSE,1 command for a subsequent singleframe restart.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
530
EQSLV
Command Default
The sparse direct solver is the default solver for all analyses, with the exception of Distributed ANSYS analyses
and modal/buckling analyses.
In Distributed ANSYS, the DSPARSE solver is the default solver for static, full transient, and harmonic analyses.
For modal/buckling analyses, there is no default solver. You must specify a solver with the MODOPT or
BUCOPT command.
Notes
The selection of a solver can affect the speed and accuracy of a solution. For a more detailed discussion of
the merits of each solver, see Solution in the Basic Analysis Guide.
If you use MODOPT,LANB or BUCOPT,LANB, the internal solver used is EQSLV,SPARSE. If you use MOD-
OPT,LANPCG, the internal solver used is EQSLV, PCG.
You may only specify the solver type in the first load step. You may, however, modify the solver tolerance
in subsequent load steps for the iterative solvers.
This command is also valid in PREP7.
Product Restrictions
Available Products Command
Option Lab
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS SPARSE
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS JCG
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS ICCG
QMR
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS PCG
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS AMG
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS ITER
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS DSPARSE
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Sol'n Options
Main Menu>Preprocessor>Loads>Fast Sol'n Optn
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Sol'n Controls>Sol'n Options
Main Menu>Solution>Fast Sol'n Optn
531
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EQSLV
ERASE
Explicitly erases the current display.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Similar to a hardware screen erase key. Useful during an "immediate" display to erase the screen without a
replot so that the display continues on a clean screen. This action is automatically included in commands
such as NPLOT and EPLOT.
If the /NOERASE command is active, issuing the erase command will simply clear the display area. Subsequent
replots will provide the display previously generated by the /NOERASE command.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Erase Options>Erase Screen
/ERASE
Specifies that the screen is to be erased before each display.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Command Default
Previous /ERASE or /NOERASE setting will be used. Initially defaults to /ERASE setting.
Notes
Erase occurs with the next display request, but before the display is actually started. /NOERASE can be used
to suppress the automatic screen erase.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Erase Options>Erase Between Plots
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
532
ERASE
EREAD, Fname, Ext, --
Reads elements from a file.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to ELEM if Fname is blank.
--
Unused field.
Notes
This read operation is not necessary in a standard ANSYS run but is provided as a convenience to users
wanting to read a coded element file, such as from another mesh generator or from a CAD/CAM program.
Data should be formatted as produced with the EWRITE command. If issuing EREAD to acquire element
information generated from ANSYS EWRITE, you must also issue NREAD before the EREAD command. The
element types [ET] must be defined before the file is read so that the file may be read properly. Only elements
that are specified with the ERRANG command are read from the file. Also, only elements that are fully attached
to the nodes specified on the NRRANG command are read from the file. Elements are assigned numbers
consecutively as read from the file, beginning with the current highest database element number plus one.
The file is rewound before and after reading. Reading continues until the end of the file.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Read Elem File
EREFINE, NE1, NE2, NINC, LEVEL, DEPTH, POST, RETAIN
Refines the mesh around specified elements.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NE1, NE2, NINC
Elements (NE1 to NE2 in increments of NINC) around which the mesh is to be refined. NE2 defaults to
NE1, and NINC defaults to 1. If NE1 = ALL, NE2 and NINC are ignored and all selected elements are
used for refinement. If NE1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NE1 (NE2 and NINC are ignored).
LEVEL
Amount of refinement to be done. Specify the value of LEVEL as an integer from 1 to 5, where a value
of 1 provides minimal refinement, and a value of 5 provides maximum refinement (defaults to 1).
533
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EREFINE
DEPTH
Depth of mesh refinement in terms of number of elements outward from the indicated elements, NE1
to NE2 (defaults to 0).
POST
Type of postprocessing to be done after element splitting, in order to improve element quality:
OFF
No postprocessing will be done.
SMOOTH
Smoothing will be done. Node locations may change.
CLEAN
Smoothing and cleanup will be done. Existing elements may be deleted, and node locations may
change (default).
RETAIN
Flag indicating whether quadrilateral elements must be retained in the refinement of an all-quadrilateral
mesh. (The ANSYS program ignores the RETAIN argument when you are refining anything other than
a quadrilateral mesh.)
ON
The final mesh will be composed entirely of quadrilateral elements, regardless of the element quality
(default).
OFF
The final mesh may include some triangular elements in order to maintain element quality and
provide transitioning.
Notes
EREFINE performs local mesh refinement around the specified elements. By default, the surrounding elements
are split to create new elements with 1/2 the edge length of the original elements (LEVEL = 1).
EREFINE refines all area elements and tetrahedral volume elements that are adjacent to the specified elements.
Any volume elements that are adjacent to the specified elements, but are not tetrahedra (for example,
hexahedra, wedges, and pyramids), are not refined.
You cannot use mesh refinement on a solid model that contains initial conditions at nodes [IC], coupled
nodes [CP family of commands], constraint equations [CE family of commands], or boundary conditions or
loads applied directly to any of its nodes or elements. This applies to nodes and elements anywhere in the
model, not just in the region where you want to request mesh refinement. If you have detached the mesh
from the solid model, you must disable postprocessing cleanup or smoothing (POST = OFF) after the refine-
ment to preserve the element attributes.
For additional restrictions on mesh refinement, see Revising Your Model in the Modeling and Meshing Guide.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>All
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Elements
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
534
EREFINE
EREINF
Generates reinforcing elements from selected existing (base) elements.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
The EREINF command generates reinforcing elements (REINF264 and REINF265) directly from selected base
elements (that is, existing standard elements in your model). The command scans all selected base elements
and generates (if necessary) a compatible reinforcing element type for each base element. (ANSYS allows a
combination of different base element types.)
Although predefining the reinforcing element type (ET) is not required, you must define the reinforcing
element section type (SECTYPE); otherwise, ANSYS cannot generate the reinforcing element.
The EREINF command does not create new nodes. The reinforcing elements and the base elements share
the common nodes.
Elements generated by this command are not associated with the solid model.
After the EREINF command executes, you can issue ETLIST, ELIST, and EPLOT commands to verify the
newly created reinforcing element types and elements.
Reinforcing elements do not account for any subsequent modifications made to the base elements. ANSYS
recommends issuing the EREINF command only after the base elements are finalized. If you delete or
modify base elements (via EDELE, EMODIF, ETCHG, EMID, EORIENT, NUMMRG, or NUMCMP commands,
for example), remove all affected reinforcing elements and reissue the EREINF command to avoid inconsist-
encies.
Menu Paths
This command cannot be accessed from a menu.
ERESX, Key
Specifies extrapolation of integration point results.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Key
Extrapolation key:
DEFA
If element is fully elastic (no active plasticity, creep, or swelling nonlinearities), extrapolate the integ-
ration point results to the nodes. If any portion of the element is plastic (or other active material
nonlinearity), copy the integration point results to the nodes (default).
YES
Extrapolate the linear portion of the integration point results to the nodes and copy the nonlinear
portion (for example, plastic strains).
535
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ERESX
NO
Copy the integration point results to the nodes.
Command Default
Extrapolate integration point results to the nodes for all elements except those with active plasticity, creep,
or swelling nonlinearities (default).
Notes
Specifies whether the solution results at the element integration points are extrapolated or copied to the
nodes for element and nodal postprocessing. The structural stresses, elastic and thermal strains, field gradients,
and fluxes are affected. Nonlinear data (plastic, creep, and swelling strains) are always copied to the nodes,
never extrapolated. For shell elements, ERESX applies only to integration point results in the in-plane direc-
tions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Output Ctrls>Integration Pt
Main Menu>Solution>Load Step Opts>Output Ctrls>Integration Pt
ERNORM, Key
Controls error estimation calculations.
POST1: Controls
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
Key
Control key:
ON
Perform error estimation (default). This option is not valid for PowerGraphics.
OFF
Do not perform error estimation.
Command Default
Error estimation calculations are performed by default unless PowerGraphics is enabled [/GRAPHICS,POWER].
Notes
Especially for thermal analyses, program speed increases if error estimation is suppressed. Therefore, it might
be desirable to use error estimation only when needed. The value of the ERNORM key is not saved on
file.db. Consequently, you need to reissue the ERNORM key after a RESUME if you wish to deactivate
error estimation again.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
536
ERNORM
Menu Paths
Main Menu>General Postproc>Options for Outp
Utility Menu>List>Results>Options
ERRANG, EMIN, EMAX, EINC
Specifies the element range to be read from a file.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
EMIN, EMAX, EINC
Elements with numbers from EMIN (defaults to 1) to EMAX (defaults to 99999999) in steps of EINC
(defaults to 1) will be read.
Notes
Defines the element number range to be read [EREAD] from the element file. If a range is also implied from
the NRRANG command, only those elements satisfying both ranges will be read.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Read Elem File
ESCHECK, Sele, Levl, Defkey
Perform element shape checking for a selected element set.
PREP7: Database
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Sele
Specifies whether to select elements for checking:
(blank)
List all warnings/errors from element shape checking.
ESEL
Select the elements based on the .Levl criteria specified below.
Levl
WARN
Select elements producing warning and error messages.
ERR
Select only elements producing error messages (default).
Defkey
Specifies whether check should be performed on deformed element shapes. .
537
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESCHECK
0
Do not update node coordinates before performing shape checks (default).
1
Update node coordinates using the current set of deformations in the database.
Notes
Shape checking will occur according to the current SHPP settings. Although ESCHECK is valid in all processors,
Defkey uses the current results in the database. If no results are available a warning will be issued.
This command is also valid in PREP7, SOLUTION and POST1.
Menu Paths
Main Menu>General Postproc>Check Elem Shape>Sel Warning/Error Elements
ESEL, Type, Item, Comp, VMIN, VMAX, VINC, KABS
Selects a subset of elements.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Restore the full set.
NONE
Unselect the full set.
INVE
Invert the current set (selected becomes unselected and vice versa).
STAT
Display the current select status.
The following fields are used only with Type = S, R, A, or U:
Item
Label identifying data, see Table 151: ESEL - Valid Item and Component Labels (p. 540). Some items also
require a component label. If Item = PICK (or simply P), graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI). Defaults to ELEM. If Item = STRA (straightened),
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
538
ESEL
elements are selected whose midside nodes do not conform to the curved line or non-flat area on which
they should lie. (Such elements are sometimes formed during volume meshing [VMESH] in an attempt
to avoid excessive element distortion.) You should graphically examine any such elements to evaluate
their possible effect on solution accuracy.
Comp
Component of the item (if required). Valid component labels are shown in Table 151: ESEL - Valid Item
and Component Labels (p. 540) below.
VMIN
Minimum value of item range. Ranges are element numbers, attribute numbers, load values, or result
values as appropriate for the item. A component name (as specified on the CM command) may also be
substituted for VMIN (VMAX and VINC are ignored).
VMAX
Maximum value of item range. VMAX defaults to VMIN for input values.
For result values, VMAX defaults to infinity if VMIN is positive, or to zero if VMIN is negative.
VINC
Value increment within range. Used only with integer ranges (such as for element and attribute numbers).
Defaults to 1. VINC cannot be negative.
KABS
Absolute value key:
0
Check sign of value during selection.
1
Use absolute value during selection (sign ignored).
Command Default
All elements are selected.
Notes
Selects elements based on values of a labeled item and component. For example, to select a new set of
elements based on element numbers 1 through 7, use ESEL,S,ELEM,,1,7. The subset is used when the ALL
label is entered (or implied) on other commands, such as ELIST,ALL. Only data identified by element number
are selected. Selected data are internally flagged; no actual removal of data from the database occurs. Dif-
ferent element subsets cannot be used for different load steps [SOLVE] in a /SOLU sequence. The subset
used in the first load step will be used for all subsequent load steps regardless of subsequent ESEL specific-
ations.
This command is valid in any processor.
Elements crossing the named path (see PATH command) will be selected. This option is only available in
PREP7 and POST1. If no geometry data has been mapped to the path (i.e., via PMAP and PDEF commands),
the path will assume the default mapping option (PMAP,UNIFORM) to map the geometry prior to selecting
the elements. If an invalid path name is given, the ESEL command is ignored (status of selected elements
is unchanged). If there are no elements crossing the path, the ESEL command will return zero elements se-
lected.
539
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESEL
For selections based on non-integer numbers (coordinates, results, etc.), items that are within the range
VMIN -Toler and VMAX + Toler are selected. The default tolerance Toler is based on the relative values
of VMIN and VMAX as follows:
If VMIN = VMAX, Toler = 0.005 x VMIN.
If VMIN = VMAX = 0.0, Toler = 1.0E-6.
If VMAX VMIN, Toler = 1.0E-8 x (VMAX - VMIN).
Use the SELTOL command to override this default and specify Toler explicitly.
Table 151 ESEL - Valid Item and Component Labels
Valid Item and Component Labels ESEL, Type, Item, Comp, VMIN, VMAX, VINC,
KABS
Description Comp Item
Element number. ELEM
Elements adjacent to element VMIN (VMAX and VINC fields are
ignored). Only elements (of the same dimensionality) adjacent to
ADJ
lateral faces are considered. Progression continues until edge of
model or until more than two elements are adjacent at a face.
Element type number. TYPE
Element name (or identifying number). ENAME
Material number. MAT
Real constant number. REAL
Element coordinate system number. ESYS
LS-DYNA part number (applicable only to ANSYS LS-DYNA) PART
Active elements [EALIVE]. VMIN and VMAX fields are ignored, etc. LIVE
Layer number (only composite elements with a nonzero thickness
for the requested layer number are included). [LAYER]
LAYER
Cross section ID number [SECNUM] SEC
p-elements that are included in convergence checking. PINC
p-elements that are excluded from convergence checking. PEXC
Straightened. See description under Argument Descriptions' Item
field above.
STRA
Element pressure. PRES SFE
Element convection bulk temperature. CONV
Element heat flux. HFLUX
Element (acoustic) fluid-structure interaction flag. FSI
Element (acoustic) impedance. IMPD
Surface shielding properties on high-frequency elements. (Select
logic works only on the value of conductivity, not relative per-
meability.)
SHLD
Element Maxwell force flag. MXWF
Electric surface charge density. CHRGS
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
540
ESEL
Valid Item and Component Labels ESEL, Type, Item, Comp, VMIN, VMAX, VINC,
KABS
Description Comp Item
Element infinite surface flag. INF
Element temperature. TEMP BFE
Element fluence. FLUE
Element heat generation rate. HGEN
Element current density, magnitude only. JS
Element magnetic virtual displacements flag. MVDI
Electric charge density. CHRGD
Selects all elements being crossed by the path with name Lab
(see PATH command). If Lab = ALL then all elements related to
all defined paths will be selected.
Lab PATH
Valid item and component labels for element result values are:
Any user-defined element table label (see ETABLE command). Lab ETAB
Menu Paths
Main Menu>Preprocessor>Modeling>CMS>CMS Superelements>By Picking
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
Utility Menu>Select>Entities
/ESHAPE, SCALE, KEY
Displays elements with shapes determined from the real constants or section definition.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SCALE
Scaling factor:
0
Use simple display of line and area elements. This value is the default.
1
Use real constants or section definition to form a solid shape display of the applicable elements.
FAC
Multiply certain real constants, such as thickness, by FAC (where FAC > 0.01) and use them to form
a solid shape display of elements.
KEY
Current shell thickness key:
0
Use current thickness in the displaced solid shape display of shell elements (valid for SHELL181,
SHELL208, SHELL209, and SHELL281). This value is the default.
541
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/ESHAPE
1
Use initial thickness in the displaced solid shape display of shell elements.
Command Default
Use simple display of line and area elements (SCALE = 0).
Notes
The /ESHAPE command allows beams, shells, current sources, and certain special-purpose elements to be
displayed as solids with the shape determined from the real constants or section types. Elements are displayed
via the EPLOT command. No checks for valid or complete input are made for the display.
Following are details about using this command with various element types:
SOLID65 elements are displayed with internal lines that represent rebar sizes and orientations (requires
vector mode [/DEVICE] with a basic type of display [/TYPE,,BASIC]). The rebar with the largest volume
ratio in each element plots as a red line, the next largest as green, and the smallest as blue.
COMBIN14, COMBIN39, and MASS21 are displayed with a graphics icon, with the offset determined by
the real constants and KEYOPT settings.
BEAM188, BEAM189, PIPE288, PIPE289 and ELBOW290 are displayed as solids with the shape determined
via the section-definition commands (SECTYPE and SECDATA). The arbitrary section option (Subtype
= ASEC) has no definite shape and appears as a thin rectangle to show orientation. The elements are
displayed with internal lines representing the cross-section mesh.
/ESHAPE does not support contour plots. Contour plots are available for these elements in postprocessing
for PowerGraphics only (/GRAPHICS,POWER). To view 3-D deformed shapes for the elements, issue
OUTRES,MISC or OUTRES,ALL for static or transient analyses. To view 3-D mode shapes for a modal or
eigenvalue buckling analysis, expand the modes with element results calculation ON (Elcalc = YES
for MXPAND).
SOURC36, CIRCU124, and TRANS126 elements always plot using /ESHAPE when PowerGraphics is activ-
ated (/GRAPHICS,POWER).
In most cases, /ESHAPE renders a thickness representation of your shell, plane and layered elements more
readily in PowerGraphics (/GRAPHICS,POWER). This type of representation employs PowerGraphics to gen-
erate the enhanced representation, and will often provide no enhancement in Full Graphics (/GRAPHICS,FULL).
This is especially true for POST1 results displays, where /ESHAPE is not supported for most element types
with FULL graphics.
When PowerGraphics is active, /ESHAPE may degrade the image if adjacent elements have overlapping
material, such as shell elements which are not co-planar. Additionally, if adjacent elements have different
thicknesses, the polygons depicting the connectivity between the thicker and thinner elements along
the shared element edges may not always be displayed.
For POST1 results displays (such as PLNSOL), the following limitations apply:
Rotational displacements for beam elements are used to create a more realistic displacement display.
When /ESHAPE is active, displacement plots (via PLNSOL,U,X and PLDISP, for example) may disagree
with your PRNSOL listings. This discrepancy will become more noticeable when the SCALE value is not
equal to one.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
542
/ESHAPE
When shell elements are not co-planar, the resulting PLNSOL display with /ESHAPE will actually be a
PLESOL display as the non-coincident pseudo-nodes are not averaged. Additionally, /ESHAPE should
not be used with coincident elements because the plot may incorrectly average the displacements of
the coincident elements.
When plotting stresses (PLNSOL,S,X) of BEAM4 and BEAM44, the thermal warping stresses are not in-
cluded in the stress display. An example of element temperatures causing warping thermal stresses is
-1, 1, -1, 1, -1, 1, -1, 1.
If you want to view solution results (PLNSOL, etc.) on layered elements (such as SHELL181, SOLSH190,
SOLID185 Layered Solid, SOLID186 Layered Solid, SHELL208, SHELL209, SHELL281, and ELBOW290), set
KEYOPT(8) = 1 for the layer elements so that the data for all layers is stored in the results file.
You can plot the through-thickness temperatures of elements SHELL131 and SHELL132 regardless of
the thermal DOFs in use by issuing the PLNSOL,TEMP command (with PowerGraphics and /ESHAPE
active).
The /ESHAPE,1 and /ESHAPE,FAC commands are incompatible with the /CYCEXPAND command used
in cyclic symmetry analyses.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Size and Shape
ESIZE, SIZE, NDIV
Specifies the default number of line divisions.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SIZE
Default element edge length on surface boundaries (i.e., lines). Divisions are automatically calculated
(rounded upward to next integer) from line lengths. If SIZE is zero (or blank), use NDIV.
NDIV
Default number of element divisions along region boundary lines. Not used if SIZE is input.
Notes
Specifies the default number of line divisions (elements) to be generated along the region boundary lines.
The number of divisions may be defined directly or automatically calculated. Divisions defined directly for
any line [LESIZE, KESIZE, etc.] are retained. For adjacent regions, the divisions assigned to the common line
for one region are also used for the adjacent region. See the MOPT command for additional meshing options.
For free meshing operations, if smart element sizing is being used [SMRTSIZE] and ESIZE,SIZE has been
specified, SIZE will be used as a starting element size, but will be overridden (i.e., a smaller size may be
used) to accommodate curvature and small features.
This command is also valid for rezoning.
543
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESIZE
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Global>Size
Main Menu>Preprocessor>Meshing>Size Cntrls>SmartSize>Adv Opts
ESLA, Type
Selects those elements associated with the selected areas.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of element select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Notes
Selects area elements belonging to meshed [AMESH], selected [ASEL] areas.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
ESLL, Type
Selects those elements associated with the selected lines.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of element select:
S
Select a new set (default).
R
Reselect a set from the current set.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
544
ESLA
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Notes
Selects line elements belonging to meshed [LMESH], selected [LSEL] lines.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
ESLN, Type, EKEY, NodeType
Selects those elements attached to the selected nodes.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of element selected:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
EKEY
Node set key:
0
Select element if any of its nodes are in the selected nodal set (default).
1
Select element only if all of its nodes are in the selected nodal set.
NodeType
Label identifying type of nodes to consider when selecting:
ALL
Select elements considering all of their nodes (default).
ACTIVE
Select elements considering only their active nodes. An active node is a node that contributes DOFs
to the model.
545
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESLN
INACTIVE
Select elements considering only their inactive nodes (such as orientation or radiation nodes).
CORNER
Select elements considering only their corner nodes.
MID
Select elements considering only their midside nodes.
Notes
ESLN selects elements which have any (or all EKEY) NodeType nodes in the currently-selected set of nodes.
Only elements having nodes in the currently-selected set can be selected.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
Utility Menu>Select>Entities
ESLV, Type
Selects elements associated with the selected volumes.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of element selected:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Notes
Selects volume elements belonging to meshed [VMESH], selected [VSEL] volumes.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
546
ESLV
ESOL, NVAR, ELEM, NODE, Item, Comp, Name
Specifies element data to be stored from the results file.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVAR
Arbitrary reference number assigned to this variable (2 to NV [NUMVAR]). Overwrites any existing results
for this variable.
ELEM
Element for which data are to be stored. If ELEM = P, graphical picking is enabled (valid only in the GUI).
NODE
Node number on this element for which data are to be stored. If blank, store the average element value
(except for FMAG values, which are summed instead of averaged). If NODE = P, graphical picking is enabled
(valid only in the GUI).
Item
Label identifying the item. General item labels are shown in Table 152: ESOL - General Item and Component
Labels (p. 548) below. Some items also require a component label.
Comp
Component of the item (if required). General component labels are shown in Table 152: ESOL - General
Item and Component Labels (p. 548) below. If Comp is a sequence number (n), the NODE field will be ig-
nored.
Name
Thirty-two character name for identifying the item on the printout and displays. Defaults to a label
formed by concatenating the first four characters of the Item and Comp labels.
Notes
Valid item and component labels for element (except line element) results are listed in Table 152: ESOL -
General Item and Component Labels (p. 548)
Defines element results data to be stored from a results file [FILE]. Not all items are valid for all elements.
See the input and output summary tables of the Element Reference of each element for the available items.
There are two methods of data access that may be used with the ESOL command. You can access some
simply by using a generic label (Component Name method), while others require a label and number (Se-
quence Number method).
The Component Name method is used to access the General element data (that is, element data which is
generally available to most element types or groups of element types). The General Item and Comp labels
for the Component Name method are shown in Table 152: ESOL - General Item and Component Labels (p. 548)
below.
The Sequence Number method is required for data that is not averaged (such as pressures at nodes, tem-
peratures at integration points, etc.), or data that is not easily described in a generic fashion (such as all
derived data for structural line elements and contact elements, all derived data for thermal line elements,
layer data for layered elements, etc.). A table illustrating the Items (such as LS, LEPEL, LEPTH, SMISC, NMISC,
etc.) and corresponding sequence numbers for each element is shown in Element Library in the Element
Reference.
547
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESOL
Element results are in the element coordinate system, except for layered elements where results are in the
layer coordinate system. Element forces and moments are in the layer coordinate system. Results are obtain-
able for an element at a specified node. Further location specifications can be made for some elements with
the SHELL, LAYERP26, and FORCE commands.
Table 152 ESOL - General Item and Component Labels
General Item and Component Labels ESOL, NVAR,ELEM,NODE,Item,Comp,Name
Description Comp Item
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity. INT "
Equivalent stress. EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity. INT "
Elastic equivalent strain. EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity. INT "
Thermal equivalent strain. EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity. INT "
Plastic equivalent strain. EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1,2,3 "
Creep strain intensity. INT "
Creep equivalent strain. EQV "
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Accumulated equivalent creep strain. CREQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Elastic strain energy density. ELASTIC SEND
Plastic strain energy density. PLASTIC "
Creep strain energy density. CREEP "
Damage variable DMG CDM
Maximum previous strain energy for virgin material LM "
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
548
ESOL
General Item and Component Labels ESOL, NVAR,ELEM,NODE,Item,Comp,Name
Description Comp Item
Gasket component stress (also gasket pressure). X GKS
Gasket component total closure. X GKD
Gasket component total inelastic closure. X GKDI
Gasket component thermal closure. X GKTH
Interface traction (stress) X, XY, XZ SS
Interface Separation. X,XY,XZ SD
Contact status. STAT1 CONT
Contact penetration PENE "
Contact pressure PRES "
Contact friction stress SFRIC "
Contact total stress (pressure plus friction) STOT "
Contact sliding distance SLIDE "
Contact gap distance GAP "
Total heat flux at contact surface FLUX "
Total number of contact status changes during substep CNOS "
Fluid penetration pressure FPRS "
Component thermal gradient or vector sum X, Y, Z, SUM TG
Component thermal flux or vector sum X, Y, Z, SUM TF
Component pressure gradient or vector sum X, Y, Z, SUM PG
Component electric field or vector sum X, Y, Z, SUM EF
Component electric flux density or vector sum X, Y, Z, SUM D
Component magnetic field intensity or vector sum X, Y, Z, SUM H
Component magnetic flux density or vector sum X, Y, Z, SUM B
Component electromagnetic forces or vector sum X, Y, Z, SUM FMAG
Component structural force X, Y, Z F
Component structural moment X, Y, Z M
Heat flow HEAT[2]
Fluid flow FLOW
Current flow AMPS
Magnetic flux FLUX
Component magnetic current segment X, Y, Z CSG
"Stiffness" energy SENE
Elemental energy dissipation due to stabilization STEN
Kinetic energy KENE
Element Joule heat generation JHEAT
Conduction current density for elements that support con-
duction current calculation. Components (X, Y, Z) and vector
sum (SUM)
X, Y, Z, SUM JC
549
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESOL
General Item and Component Labels ESOL, NVAR,ELEM,NODE,Item,Comp,Name
Description Comp Item
Source current density for low-frequency magnetic analyses.
Total current density (sum of conduction and displacement
X, Y, Z JS
current densities) in low frequency electric analyses. Com-
ponents (X, Y, Z)
Total measureable current density in low-frequency electro-
magnetic analyses. (Conduction current density in a low-
X, Y, Z, SUM JT
frequency electric analysis.) Components (X, Y, Z) and vector
sum (SUM)
Magnetics Reynolds number MRE
Volume of volume element VOLU
Body temperatures (calculated from applied temperatures)
as used in solution (area and volume elements only)
TEMP BFE
Fictive temperature TEMP FICT
1. For more information on the meaning of contact status and its possible values, see Reviewing Results
in POST1 in the Contact Technology Guide.
2. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels HBOT, HE2, HE3, . . .,
HTOP instead of HEAT.
Menu Paths
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
ESORT, Item, Lab, ORDER, KABS, NUMB
Sorts the element table.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Label identifying the item:
ETAB
(currently the only Item available)
Lab
element table label:
Lab
Any user-defined label from the ETABLE command (input in the Lab field of the ETABLE command).
ORDER
Order of sort operation:
0
Sort into descending order.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
550
ESORT
1
Sort into ascending order.
KABS
Absolute value key:
0
Sort according to real value.
1
Sort according to absolute value.
NUMB
Number of elements (element table rows) to be sorted in ascending or descending order (ORDER) before
sort is stopped (remainder will be in unsorted sequence) (defaults to all elements).
Command Default
Use ascending element number order.
Notes
The element table rows are sorted based on the column containing the Lab values. Use EUSORT to restore
the original order. If ESORT is specified with PowerGraphics on [/GRAPHICS,POWER], then the nodal solution
results listing [PRNSOL] will be the same as with the full graphics mode [/GRAPHICS,FULL].
Menu Paths
Main Menu>General Postproc>List Results>Sorted Listing>Sort Elems
ESSOLV, Electit, Strutit, DIMN, MORPHOPT, Mcomp, Xcomp, ELECTOL, STRUTOL, MXLOOP, --, RUSEKY, RESTKY,
EISCOMP
Performs a coupled electrostatic-structural analysis.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Electit
Title of the electrostatics physics file as assigned by the PHYSICS command.
Strutit
Title of the structural physics file as assigned by the PHYSICS command.
DIMN
Model dimensionality (a default is not allowed):
2
2-D model.
3
3-D model.
MORPHOPT
Morphing option:
551
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESSOLV
<0
Do not perform any mesh morphing or remeshing.
0
Remesh the non-structural regions for each recursive loop only if mesh morphing fails (default).
1
Remesh the non-structural regions each recursive loop and bypass mesh morphing.
2
Perform mesh morphing only, do not remesh any non-structural regions.
Mcomp
Component name of the region to be morphed. For 2-D models, the component may be elements or
areas. For 3-D models, the component may be elements or volumes. A component must be specified.
You must enclose name-strings in single quotes in the ESSOLV command line.
Xcomp
Component name of entities excluded from morphing. In the 2-D case, it is the component name for
the lines excluded from morphing. In the 3-D case, it is component name for the areas excluded from
morphing. Defaults to exterior non-shared entities (see the DAMORPH, DVMORPH, and DEMORPH
commands). You must enclose name-strings in single quotes in the ESSOLV command line.
ELECTOL
Electrostatic energy convergence tolerance. Defaults to .005 (.5%) of the value computed from the pre-
vious iteration. If less than zero, the convergence criteria based on electrostatics results is turned off.
STRUTOL
Structural maximum displacement convergence tolerance. Defaults to .005 (.5%) of the value computed
from the previous iteration. If less than zero, the convergence criteria base on structural results is turned
off.
MXLOOP
Maximum number of allowable solution recursive loops. A single pass through both an electrostatics
and structural analysis constitutes one loop. Defaults to 100.
--
Unused field.
RUSEKY
Reuse flag option:
1
Assumes initial run of ESSOLV using base geometry for the first electrostatics solution.
>1
Assumes ESSOLV run is a continuation of a previous ESSOLV run, whereby the morphed geometry
is used for the initial electrostatic simulation.
RESTKY
Structural restart key.
0
Use static solution option for structural solution.
1
Use static restart solution option for structural solution.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
552
ESSOLV
EISCOMP
Element component name for elements containing initial stress data residing in file jobname.ist. The
initial stress data must be defined prior to issuing ESSOLV (see INISTATE command).
Notes
ESSOLV invokes an ANSYS macro which automatically performs a coupled electrostatic-structural analysis.
The macro displays periodic updates of the convergence.
If non-structural regions are remeshed during the analysis, boundary conditions and loads applied to nodes
and elements will be lost. Accordingly, it is better to assign boundary conditions and loads to the solid
model.
Use RUSEKY > 1 for solving multiple ESSOLV simulations for different excitation levels (i.e., for running a
voltage sweep). Do not issue the SAVE command to save the database between ESSOLV calls.
For nonlinear structural solutions, the structural restart option (RESTKY = 1) may improve solution time by
starting from the previous converged structural solution.
For solid elements, ESSOLV automatically detects the air-structure interface and applies a Maxwell surface
flag on the electrostatic elements. This flag is used to initiate the transfer for forces from the electrostatic
region to the structure. When using the ESSOLV command with structural shell elements (for example,
SHELL63), you must manually apply the Maxwell surface flag on all air elements surrounding the shells before
writing the final electrostatic physics file. Use the SFA command to apply the Maxwell surface flag to the
areas representing the shell elements; doing so ensures that the air elements next to both sides of the shells
receive the Maxwell surface flag.
If lower-order structural solids or shells are used, set KEYOPT(7) = 1 for the electrostatic element types to
ensure the correct transfer of forces.
Information on creating the initial stress file is documented in the "Loading" chapter in the Basic Analysis
Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Physics>Coupled Solvers>Elec/struc
Main Menu>Solution>Physics>Coupled Solvers>Elec/struc
ESTIF, KMULT
Specifies the matrix multiplier for deactivated elements.
SOLUTION: Birth and Death
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
KMULT
Stiffness matrix multiplier for deactivated elements (defaults to 1.0E-6).
553
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESTIF
Command Default
Use 1.0E-6 as the multiplier.
Notes
Specifies the stiffness matrix multiplier for elements deactivated with the EKILL command (birth and death).
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Birth & Death>StiffnessMult
Main Menu>Solution>Load Step Opts>Other>Birth & Death>StiffnessMult
ESURF, XNODE, Tlab, Shape
Generates elements overlaid on the free faces of existing selected elements.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XNODE
Used only when generating SURF151 or SURF152 elements with KEYOPT(5) = 1. XNODE is the single extra
node number (ID) used with these surface elements. There is no default. XNODE must be specified if
KEYOPT(5) = 1. If XNODE = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A parameter or parametric expression can be substituted for XNODE.
Tlab
Used to generate target and contact elements with correct direction of normals. This label is used only
with TARGE169, TARGE170, CONTA171, CONTA172, CONTA173, CONTA174, CONTA176, and CONTA177
elements.
TOP
Generates target and contact elements over beam and shell elements with the normals the same as
the underlying beam and shell elements (default).
BOTTOM
Generates target and contact elements over beam and shell elements with the normals opposite to
the underlying beam and shell elements.
REVERSE
Reverses the direction of the normals on existing selected target and contact elements.
Shape
Used to specify the element shape for target element TARGE170 (Shape = TRI, LINE, or POINT) or
TARGE169 elements (Shape = POINT).
(blank)
The target element takes the same shape as the external surface of the underlying element (default).
TRI
Generates several triangular facet target elements.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
554
ESURF
LINE
Generates LINE or PARA (parabolic) segments on exterior of selected 3-D elements.
POINT
Generates POINT segments on selected nodes.
Notes
The ESURF command generates elements of the currently active element type overlaid on the free faces of
existing elements. For example, surface elements (such as SURF151, SURF152, SURF153, or SURF154) can be
generated over solid elements (such as PLANE55, SOLID70, PLANE182, or SOLID185, respectively).
Element faces are determined from the selected node set (NSEL) and the load faces for that element type.
The operation is similar to that used for generating element loads from selected nodes via the SF,ALL
command, except that elements (instead of loads) are generated. All nodes on the face must be selected
for the face to be used. For shell elements, only face one of the element is available. If nodes are shared by
adjacent selected element faces, the faces are not free and no element is generated.
Elements created by ESURF are oriented such that their surface load directions are consistent with those of
the underlying elements. Carefully check generated elements and their orientations.
Generated elements use the existing nodes and the active MAT, TYPE, REAL, and ESYS attributes. The ex-
ception is when Tlab = REVERSE. The reversed target and contact elements have the same attributes as
the original elements. If the underlying elements are solid elements, Tlab = TOP or BOTTOM has no effect.
When the command generates a target element, the shape is by default the same as that of the underlying
element. Issue ESURF,,,LINE or ESURF,,,POINT to generate LINE, PARA, and POINT segments. While not re-
commended, you can issue ESURF,,,TRI to generate several facet triangle elements.
The ESURF command can also generate the 2-D or 3-D node-to-surface element CONTA175, based on the
selected node components of the underlying solid elements. When used to generate CONTA175 elements,
all ESURF arguments are ignored. (If CONTA175 is the active element type, the path Main Menu> Prepro-
cessor> Modeling> Create> Elements> Node-to-Surf uses ESURF to generate elements.)
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Inf Acoustic
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Node to Surf
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Generl Surface>Ex-
tra Node
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Generl Surface>No
extra Node
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf to Surf
555
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ESURF
ESYM, --, NINC, IEL1, IEL2, IEINC
Generates elements from a pattern by a symmetry reflection.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
--
Unused field.
NINC
Increment nodes in the given pattern by NINC.
IEL1, IEL2, IEINC
Reflect elements from pattern beginning with IEL1 to IEL2 (defaults to IEL1) in steps of IEINC
(defaults to 1). If IEL1 = ALL, IEL2 and IEINC are ignored and pattern is all selected elements [ESEL].
If IEL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI). A component name may also be substituted for IEL1 (IEL2 and IEINC are ignored).
Notes
Generates additional elements from a given pattern (similar to EGEN) except with a "symmetry" reflection.
The operation generates a new element by incrementing the nodes on the original element, and reversing
and shifting the node connectivity pattern. For example, for a 4-node 2-D element, the nodes in positions
I, J, K, and L of the original element are placed in positions J, I, L, and K of the reflected element.
Similar permutations occur for all other element types. For line elements, the nodes in positions I and J of
the original element are placed in positions J and I of the reflected element. In releases prior to ANSYS 5.5,
no node pattern reversing and shifting occurred for line elements generated by ESYM. To achieve the same results
with ANSYS 5.5 as you did in prior releases, use the EGEN command instead.
It is recommended that symmetry elements be displayed and graphically reviewed.
If the nodes are also reflected (as with the NSYM command) this pattern is such that the orientation of the
symmetry element remains similar to the original element (i.e., clockwise elements are generated from
clockwise elements).
For a non-reflected node pattern, the reversed orientation has the effect of reversing the outward normal
direction (clockwise elements are generated from counterclockwise elements).
Note
Since nodes may be defined anywhere in the model independently of this command, any orient-
ation of the "symmetry" elements is possible. See also the ENSYM command for modifying existing
elements.
Menu Paths
Main Menu>Preprocessor>Modeling>Reflect>Elements>Auto Numbered
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
556
ESYM
ESYS, KCN
Sets the element coordinate system attribute pointer.
PREP7: Meshing
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Coordinate system number:
0
Use element coordinate system orientation as defined (either by default or by KEYOPT setting) for
the element (default).
N
Use element coordinate system orientation based on local coordinate system N (where N must be
greater than 10). For global system 0, 1, or 2, define a local system N parallel to appropriate system
with the LOCAL or CS command (for example: LOCAL,11,1).
Command Default
Use element coordinate system orientation as defined (either by default or by KEYOPT setting) for the element
(default).
Notes
Identifies the local coordinate system to be used to define the element coordinate system of subsequently
defined elements. Used only with area and volume elements. For non-layered volume elements, the local
coordinate system N is simply assigned to be the element coordinate system. For shell and layered volume
elements, the x and y axes of the local coordinate system N are projected onto the shell or layer plane to
determine the element coordinate system. See Element Coordinate Systems for more details. N refers to the
coordinate system reference number (KCN) defined using the LOCAL (or similar) command. Element coordinate
system numbers may be displayed [/PNUM].
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>Default Attribs
Main Menu>Preprocessor>Modeling>Create>Elements>Elem Attributes
ET, ITYPE, Ename, KOP1, KOP2, KOP3, KOP4, KOP5, KOP6, INOPR
Defines a local element type from the element library.
PREP7: Element Type
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
ITYPE
Arbitrary local element type number. Defaults to 1 + current maximum.
Ename
Element name (or number) as given in the element library in Chapter 4 of the Element Reference. The
name consists of a category prefix and a unique number, such as BEAM3. The category prefix of the
557
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ET
name (BEAM for the example) may be omitted but is displayed upon output for clarity. If Ename = 0,
the element is defined as a null element.
KOP1, KOP2, KOP3, . . . , KOP6
KEYOPT values (1 through 6) for this element, as described in the Element Reference.
INOPR
If 1, suppress all element solution printout for this element type.
Notes
Elements types are selected from the element library (and are therefore established as local element types
for the current model) by this command. Information derived from the element type is used for succeeding
commands, so the ET command(s) should be entered early. The Element Reference describes the various
elements which are available in the ANSYS library. For example, if a 2-D plane quadrilateral structural element
type is desired, Ename should be PLANE42 or 42. A special option, Ename = 0, permits this element type
to be ignored during solution without actually removing the element from the model. Ename may be set
to zero only after the element type has been previously defined with a nonzero Ename. The preferred
method of ignoring elements is to use the select commands (such as ESEL).
KOPn are element option keys. These keys (referred to as KEYOPT(n)) are used to turn on certain element
options for this element. These options are listed under "KEYOPT" in the input table for each element type
in the Element Reference. KEYOPT values include stiffness formulation options, printout controls, and various
other element options. If KEYOPT(7) or greater is needed, input their values with the KEYOPT command.
The ET command only defines an element type local to your model (from the types in the element library).
The TYPE or similar [KATT, LATT, AATT, or VATT] command must be used to point to the desired local
element type before meshing.
To activate the ANSYS program's LS-DYNA explicit dynamic analysis capability, use the ET command or its
GUI equivalent to choose an element that works only with LS-DYNA (such as SHELL163). Choosing LS-DYNA
in the Preferences dialog box does not activate LS-DYNA; it simply makes items and options related to LS-
DYNA accessible in the GUI.
Menu Paths
Main Menu>Preprocessor>Element Type>Add/Edit/Delete
ETABLE, Lab, Item, Comp
Fills a table of element values for further processing.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Any unique user defined label for use in subsequent commands and output headings (maximum of
eight characters and not a General predefined Item label). Defaults to an eight character label formed
by concatenating the first four characters of the Item and Comp labels. If the same as a previous user
label, this result item will be included under the same label. Up to 200 different labels may be defined.
The following labels are predefined by ANSYS and cannot be used as user-defined labels: REFL, STAT,
and ERAS. Lab = REFL refills all tables previously defined with the ETABLE commands (not the CALC
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
558
ETABLE
module commands) according to the latest ETABLE specifications and is convenient for refilling tables
after the load step [SET] has been changed. Remaining fields will be ignored if Lab is REFL. Lab = STAT
displays stored table values. Lab = ERAS erases the entire table.
Item
Label identifying the item. General item labels are shown in the table below. Some items also require a
component label. Character parameters may be used. Item = ERAS erases a Lab column.
Comp
Component of the item (if required). General component labels are shown in the table below. Character
parameters may be used.
Notes
Defines a table of values per element (the element table) for use in further processing (see POST1 Element
Table commands). The element table is organized as a "worksheet," with the rows representing all selected
elements, and the columns consisting of result items which have been moved into the table (Item,Comp)
with ETABLE. Each column of data is identified by a user-defined label (Lab) for listings and displays.
After entering the data into the element table, you are not limited to merely listing or displaying your data
[PLESOL, PRESOL, etc.]. You may also perform many types of operations on your data, such as adding or
multiplying columns [SADD, SMULT], defining allowable stresses for safety calculations [SALLOW], or mul-
tiplying one column by another [SMULT]. See the Getting Started with ANSYS in theBasic Analysis Guide for
further details.
There are different types of results data that may be stored in the element table. For example, many items
for an element are inherently single-valued (i.e., there is only one value per element). The single-valued
items include: SERR, SDSG, TERR, TDSG, SENE, TENE, KENE, JHEAT, JS, VOLU, and CENT. All other items are
multivalued (i.e., they vary over the element, such that there is a different value at each node). Since only
one value is stored in the element table per element, an average value (based on the number of contributing
nodes) is calculated for multivalued items. Exceptions to this averaging procedure are FMAG and all element
force items, which represent the sum only of the contributing nodal values.
There are two methods of data access that may be used with the ETABLE command. The method you choose
will depend upon the type of data that you wish to store. Some results can be accessed just with the use
of a generic label (Component Name method), while others require a label and number (Sequence Number
method).
The Component Name method is used to access the General element data (that is, element data which is
generally available to most element types or groups of element types). All of the single-valued items and
some of the more general multivalued items are accessible with the Component Name method. Various
element results depend on the calculation method and the selected results location (AVPRIN, RSYS, LAYER,
SHELL, and ESEL).
Although nodal data is readily available for listings and displays [PRNSOL, PLNSOL] without using the element
table, you may also use the Component Name method to enter these results into the element table for
further "worksheet" manipulation (see the Getting Started with ANSYS in theBasic Analysis Guide for more
details). A listing of the General Item and Comp labels for the Component Name method is shown below.
The Sequence Number method allows you to view results for data that is not averaged (such as pressures
at nodes, temperatures at integration points, etc.), or data that is not easily described in a generic fashion
(such as all derived data for structural line elements and contact elements, all derived data for thermal line
elements, layer data for layered elements, etc.). A table illustrating the Items (such as LS, LEPEL, LEPTH,
559
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ETABLE
SMISC, NMISC, SURF, etc.) and corresponding sequence numbers for each element is shown in the Output
Data section of each element description found in the About This ManualElement Reference.
Some element table data are reported in the results coordinate system. These include all component results
(for example, UX, UY, etc.; SX, SY, etc.). The solution writes component results in the database and on the
results file in the solution coordinate system. When you issue the ETABLE command, these results are then
transformed into the results coordinate system [RSYS] before being stored in the element table. The default
results coordinate system is global Cartesian [RSYS,0]. All other data are retrieved from the database and
stored in the element table with no coordinate transformation.
Use the PRETAB, PLETAB, or ETABLE,STAT commands to display the stored table values. Issue ETABLE,ERAS
to erase the entire table. Issue ETABLE,Lab,ERAS to erase a Lab column.
When the GUI is on, if a Delete operation in a Define Element Table Data dialog box writes this command
to a log file (Jobname.LOG or Jobname.LGW), you will observe that Lab is blank, Item = ERASE, and
Comp is an integer number. In this case, the GUI has assigned a value of Comp that corresponds to the loc-
ation of a chosen variable name in the dialog box's list. It is not intended that you type in such a location
value for Comp in an ANSYS session. However, a file that contains a GUI-generated ETABLE command of
this form can be used for batch input or for use with the /INPUT command.
Table 153 ETABLE - General Item and Component Labels
General Item and Component Labels ETABLE, Lab, Item, Comp
Description Comp Item
Valid Item Labels for Degree of Freedom Results
X, Y, or Z structural displacement. X, Y, Z U
X, Y, or Z structural rotation. X, Y, Z ROT
Temperature. TEMP[1]
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity. X, Y, Z V
X, Y, or Z magnetic vector potential. X, Y, Z A
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy. ENKE
Turbulent energy dissipation. ENDS
Mass fraction of species n, where n = 1 to 6. If a species is given
a user-defined name [MSSPEC], use that name instead of SP0n.
SP0n
Valid Item Labels for FLOTRAN Nodal Results
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
560
ETABLE
General Item and Component Labels ETABLE, Lab, Item, Comp
Description Comp Item
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
Laminar mass diffusion coefficient for species n, where n = 1 to
6.
LMDn
Effective mass diffusion coefficient for species n, where n = 1 to
6.
EMDn
Valid Item and Component Labels for Element Results
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity. INT "
Equivalent stress. EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity. INT "
Elastic equivalent strain. EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity. INT "
Thermal equivalent strain. EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity. INT "
Plastic equivalent strain. EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity. INT "
Creep equivalent strain. EQV "
Swelling strain. EPSW
Component total mechanical strain (excluding thermal) (EPEL +
EPPL + EPCR).
X, Y, Z, XY, YZ, XZ EPTO
Principal total mechanical strain. 1, 2, 3 "
561
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ETABLE
General Item and Component Labels ETABLE, Lab, Item, Comp
Description Comp Item
Total mechanical strain intensity. INT "
Total equivalent mechanical strain. EQV "
Component total strain including thermal (EPEL + EPTH + EPPL +
EPCR).
X, Y, Z, XY, YZ, XZ EPTT
Principal total strain. 1, 2, 3 "
Total strain intensity. INT "
Total equivalent strain. EQV "
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Elastic strain energy density. ELASTIC SEND
Plastic strain energy density. PLASTIC "
Creep strain energy density. CREEP "
Damage variable DMG CDM
Maximum previous strain energy for virgin material LM "
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
Component pressure gradient or vector sum. X, Y, Z, SUM PG
Component electric field or vector sum. X, Y, Z, SUM EF
Component electric flux density or vector sum. X, Y, Z, SUM D
Component magnetic field intensity or vector sum. X, Y, Z, SUM H
Component magnetic flux density or vector sum. X, Y, Z, SUM B
Component electromagnetic forces or vector sum. X, Y, Z, SUM FMAG[5]
Structural error energy. SERR[6]
Absolute value of maximum variation of any nodal stress compon-
ent.
SDSG6
Thermal error energy. TERR6
Absolute value of the maximum variation of any nodal thermal
gradient component.
TDSG6
Component structural force. Sum of element nodal values. X, Y, Z F
Component structural moment. Sum of element nodal values. X, Y, Z M
Heat flow. Sum of element nodal values. HEAT
Fluid flow. Sum of element nodal values. FLOW
Current flow. Sum of element nodal values. AMPS
Magnetic flux. Sum of element nodal values. FLUX
Component magnetic current segment. X, Y, Z CSG
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
562
ETABLE
General Item and Component Labels ETABLE, Lab, Item, Comp
Description Comp Item
"Stiffness" energy or thermal heat dissipation (applies to all ele-
ments where meaningful). Same as TENE.
SENE
Artificial energy of the element. This includes the sum of hourglass
control energy and energy generated by in-plane drilling stiffness
AENE
from shell elements (applies to all elements where meaningful).
The energy is used for comparisons to SENE energy to predict
the solution error due to artificial stiffness. See the Theory Refer-
ence for the Mechanical APDL and Mechanical Applications.
Thermal heat dissipation or "stiffness" energy (applies to all ele-
ments where meaningful). Same as SENE.
TENE
Kinetic energy (applies to all elements where meaningful). KENE
Elemental energy dissipation due to stabilization. STEN
Element Joule heat generation. JHEAT
Source current density for low-frequency magnetic analyses. Total
current density (sum of conduction and displacement current
X, Y, Z, SUM JS
densities) in low frequency electric analyses. Components (X, Y,
Z) and vector sum (SUM).
Total measureable current density in low-frequency electromag-
netic analyses. (Conduction current density in a low-frequency
electric analysis.) Components (X, Y, Z) and vector sum (SUM).
X, Y, Z, SUM JT
Conduction current density for elements that support conduction
current calculation. Components (X, Y, Z) and vector sum (SUM).
X, Y, Z, SUM JC
Magnetics Reynolds number MRE
Element volume. Based on unit thickness for 2-D plane elements
(unless the thickness option is used) and on the full 360 degrees
for 2-D axisymmetric elements.
VOLU
Undeformed X, Y, or Z location (based on shape function) of the
element centroid in the active coordinate system.
X, Y, Z CENT
Body temperatures (calculated from applied temperatures) as
used in solution (area and volume elements only).
TEMP BFE
Element summable miscellaneous data value at sequence number
snum (shown in the Output Data section of each applicable ele-
ment description in the Element Reference).
snum SMISC
Element non-summable miscellaneous data value at sequence
number snum (shown in the Output Data section of each applic-
able element description found in the Element Reference).
snum NMISC
Element surface data value at sequence number snum (shown in
Surface Solution of the Element Reference).
snum SURF
Contact status[4]: STAT[3] CONT[2]
3-closed and sticking
2-closed and sliding
1-open but near contact
0-open and not near contact
563
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ETABLE
General Item and Component Labels ETABLE, Lab, Item, Comp
Description Comp Item
Contact penetration (zero or positive). PENE "
Contact pressure. PRES "
Contact friction stress. SFRIC "
Contact total stress (pressure plus friction). STOT "
Contact sliding distance. SLIDE "
Contact gap distance (0 or negative). GAP "
Total heat flux at contact surface. FLUX "
Total number of contact status changes during substep. CNOS "
Fluid penetration pressure FPRS "
Densities used for topological optimization. This applies to the
following elements: PLANE82, SOLID92, SOLID95.
TOPO
Material cap plasticity model only: Cohesion; hydrostatic compac-
tion yielding stress; I1 at the transition point at which the shear
C0,X0,K0,ZONE,
DPLS,VPLS
CAP
and compaction envelopes intersect; zone = 0: elastic state, zone
= 1: compaction zone, zone = 2: shear zone, zone = 3: expansion
zone; effective deviatoric plastic strain; volume plastic strain.
Material EDP creep model only (not including the cap model):
Equivalent creep stress; equivalent creep strain.
CSIG,CSTR EDPC
Components of Biots effective stress. X,Y,Z,XY,YZ,ZX ESIG
Principal stresses of Biots effective stress. 1, 2, 3 "
Stress intensity of Biots effective stress. INT "
Equivalent stress of Biots effective stress. EQV "
Total porosity (Gurson material model). TPOR DPAR
Porosity due to void growth. GPOR "
Porosity due to void nucleation. NPOR "
Fluid flow flux in poromechanics. X,Y,Z FFLX
Fictive temperature. TEMP FICT
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use labels TBOT, TE2, TE3, . . ., TTOP
instead of TEMP.
2. For the CONT items for elements CONTA171 through CONTA177, the reported data is averaged across
the element.
3. Status notation is for all contact elements except CONTAC12 and CONTAC52. If you are using these
elements, check the output listing in the corresponding element documentation.
4. For MPC-based contact definitions, the value of STAT can be negative. This indicates that one or more
contact constraints were intentionally removed to prevent over-constraint. STAT = -3 is used for MPC
bonded contact; STAT = -2 is used for MPC no-separation contact.
5. When using the EMFT procedure to calculate electromagnetic force (SOLID117, PLANE121, SOLID122,
SOLID123, SOLID236 or SOLID237 elements only), the FMAG sum will be zero or near zero.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
564
ETABLE
6. Some element- and material-type limitations apply. For more information, see the documentation for
the PRERR command.
Menu Paths
Main Menu>General Postproc>Element Table>Define Table
Main Menu>General Postproc>Element Table>Erase Table
ETCHG, Cnv
Changes element types to their corresponding types.
PREP7: Element Type
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
Cnv
Converts the element types to the corresponding type. Valid labels are:
ETI
Explicit to Implicit
ITE
Implicit to Explicit
TTE
Thermal to Explicit
TTS
Thermal to Structural
STT
Structural to Thermal
MTT
Magnetic to Thermal
FTS
Fluid to Structural
ETS
Electrical to Structural
Notes
Changes the currently defined element types to their corresponding types. Elements without a companion
element (listed above) are not switched and should be switched with the ET command to an appropriate
element type or to a null element. The KEYOPT values for the switched element types are reset to zero or
to their default values. You must check these values to see if they are still meaningful. Additionally, if Cnv
= ETI, ITE, or TTE, all real constants are set to zero.
If Cnv = ITE, you will need to choose a material model that corresponds to your previously-defined material
properties. If working interactively, you will be prompted to do so.
Element types LINK8 and LINK10 lack a third node; however, their corresponding companion explicit element
types, LINK160 and LINK167, require a third (orientation) node. If you are using element types LINK8 or
565
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ETCHG
LINK10, you must first convert the element type using this command and then manually define the third
node of LINK160 or LINK167 elements using EMODIF.
Also, if you are converting BEAM4 to BEAM161, you may need to manually define the third node of BEAM161
elements as well. However, BEAM4 allows you to define a third, optional node. If you have defined this third
node on BEAM4, then the conversion to BEAM161 will be completed automatically when you issue this
command. If you did not define the third node on BEAM4, then you must manually define it on BEAM161
using EMODIF.
Element Pairs
ETI -- Explicit to Implicit
167 > 10 165 > 14 163 > 181 160 > 8
166 > 21 164 > 185 161 > 4
ITE -- Implicit to Explicit
185 > 164 21 > 166 10 > 167 4 > 161
181 > 163 14 > 165 8 > 160
TTE --Thermal to Explicit
70 > 164 57 > 163 55 > 162
TTS -- Thermal to Structural
131 > 181 75 > 25 67 > 42 32 > 1
132 > 91/93 77 > 183 68 > 8 33 > 180
151 > 153 78 > 83 69 > 45 35 > 2
152 > 154 87 > 187 70 > 185 55 > 182
157 > 63 90 > 186 71 > 21 57 > 41
STT -- Structural to Thermal
154 > 152 88 > 77 62 > 70 1 > 32
158 > 87 89 > 90 63 > 131 2 > 35
180 > 33 91 > 132 64 > 70 8 > 33
181 > 131 92 > 87 65 > 70 21 > 71
182 > 55 93 > 132 73 > 70 25 > 75
183 > 77 95 > 90 74 > 77 41 > 57
185 > 70 106 > 55 82 > 77 42 > 55
186 > 90 107 > 70 83 > 78 45 > 70
187 > 87 108 > 77 84 > 55/77 56 > 55
153 >151 86 > 70 58 > 70
MTT -- Magnetic to Thermal
120 > 90 117 > 90 96 > 70 53 > 77
119 > 87 97 > 70 62 > 70
FTS -- Fluid to Structural
142 > 45 141 > 42
ETS -- Electrical to Structural
128 > 45/95 123 > 92 121 > 42/82 67 > 42
157 > 63 127 > 92 122 > 45/95 68 > 8
69 > 45
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
566
ETCHG
Menu Paths
Main Menu>Preprocessor>Element Type>Switch Elem Type
ETCONTROL, Eltech, Eldegene
Control the element technologies used in element formulation (for applicable elements).
PREP7: Element Type
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Eltech
Element technology control:
SUGGESTION
ANSYS makes a suggestion for the best element technology before solving. If necessary, mixed u-P
(KEYOPT(6)) will also be included and reset. This behavior is the default.
SET
ANSYS informs you of the best settings and automatically resets any applicable KEYOPT settings;
this will override any manual settings you provided previously.
OFF
Turns automatic selection of element technology off. No suggestions will be issued, and no automatic
resetting is done
Eldegene
Element degenerated shape control:
ON
If element shapes are degenerated, the degenerated shape function is employed and enhanced
strain, simplified enhanced strain, and B-bar formulations are turned off (default).
OFF
If element shapes are degenerated, regular shape functions are still used, and the specified element
technologies (e.g., enhanced strain, B-bar, uniform reduced integration) are still used.
Notes
The command default is ETCONTROL,SUGGESTION,ON.
This command is valid for elements SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
BEAM188, BEAM189, SHELL208, SHELL209, REINF264, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288,
PIPE289, ELBOW290.
For more information, see Automatic Selection of Element Technologies in the Element Reference.
Menu Paths
Main Menu>Preprocessor>Element Type>Elem Tech Control
567
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ETCONTROL
ETDELE, ITYP1, ITYP2, INC
Deletes element types.
PREP7: Element Type
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITYP1, ITYP2, INC
Deletes element types from ITYP1 to ITYP2 (defaults to ITYP1) in steps of INC (defaults to 1). If
ITYP1 = ALL, ITYP2 and INC are ignored and all element types are deleted. Element types are defined
with the ET command.
Menu Paths
Main Menu>Preprocessor>Element Type>Add/Edit/Delete
ETLIST, ITYP1, ITYP2, INC
Lists currently defined element types.
PREP7: Element Type
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITYP1, ITYP2, INC
Lists element types from ITYP1 to ITYP2 (defaults to ITYP1) in steps of INC (defaults to 1). If ITYP1
= ALL (default), ITYP2 and INC are ignored and all element types are listed.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>List>Properties>Element Types
ETYPE
Specifies "Element types" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
568
ETDELE
Menu Paths
Utility Menu>List>Status>Preprocessor>Element Types
EUSORT
Restores original order of the element table.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Changing the selected element set [ESEL] also restores the original element order.
Menu Paths
Main Menu>General Postproc>List Results>Sorted Listing>Unsort Elems
EWRITE, Fname, Ext, --, KAPPND, Format
Writes elements to a file.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to ELEM if Fname is blank.
--
Unused field.
KAPPND
Append key:
0
Rewind file before the write operation.
1
Append data to the end of the existing file.
Format
Format key:
569
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EWRITE
SHORT
I6 format (the default).
LONG
I8 format.
Notes
Writes the selected elements to a file. The write operation is not necessary in a standard ANSYS run but is
provided as convenience to users wanting a coded element file. If issuing EWRITE from ANSYS to be used
in ANSYS, you must also issue NWRITE to store nodal information for later use. Only elements having all of
their nodes defined (and selected) are written. Data are written in a coded format. The data description of
each record is: I, J, K, L, M, N, O, P, MAT, TYPE, REAL, SECNUM, ESYS, IEL, where MAT, TYPE, REAL, and ESYS
are attribute numbers, SECNUM is the beam section number, and IEL is the element number.
The format is (14I6) if Format is set to SHORT and (14I8) if the Format is set to LONG, with one element
description per record for elements having eight nodes of less. For elements having more than eight nodes,
nodes nine and above are written on a second record with the same format.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Write Elem File
*EXIT
Exits a do-loop.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
The command following the *ENDDO is executed next. The exit option may also be conditional [Use the
*IF]. The *EXIT command must appear on the same file as the *DO command.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
/EXIT, Slab, Fname, Ext, --
Stops the run and returns control to the system.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Slab
Mode for saving the database:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
570
*EXIT
MODEL
Save the model data (solid model, finite element model, loadings, etc.) only (default).
SOLU
Save the model data and the solution data (nodal and element results).
ALL
Save the model data, solution data and post data (element tables, path results, etc.)
NOSAVE
Do not save any data on File.DB (an existing DB file will not be overwritten).
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name, defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to DB if Fname is blank.
--
Unused field.
Notes
The current database information may be written on File.DB or a named file. If File.DB already exists,
a backup file (File.DBB) will also be written whenever a new File.DB is written.
This command is valid in any processor. Issuing this command at any point will exit the program.
Menu Paths
Utility Menu>File>Exit
EXP, IR, IA, --, --, Name, --, --, FACTA, FACTB
Forms the exponential of a variable.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA
Reference number of the variable to be operated on.
--, --
Unused fields.
571
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EXP
Name
Thirty-two character name for identifying the variable on the printout and displays. Embedded blanks
are compressed upon output.
--, --
Unused fields.
FACTA
Scaling factor applied to variable IA (defaults to 1.0).
FACTB
Scaling factor (positive or negative) applied to the operation (defaults to 1.0).
Notes
Forms the exponential of a variable according to the operation:
IR = FACTB*EXP(FACTA x IA)
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Exponentiate
EXPAND, Nrepeat, MODAL, HIndex, Icsys, SctAng, --, Phase
Displays the results of a modal cyclic symmetry analysis.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Nrepeat
Number of sector repetitions for expansion. The default is 0 (no expansion).
MODAL
Specifies that the expansion is for a modal cyclic symmetry analysis.
HIndex
The harmonic index ID for the results to expand.
Icsys
The coordinate system number used in the modal cyclic symmetry solution. The default is the global
cylindrical coordinate system (specified via the CSYS command where KCN = 1).
SctAng
The sector angle in degrees, equal to 360 divided by the number of cyclic sectors.
--
This field is reserved for future use.
Phase
The phase angle in degrees to use for the expansion. The default is 0. Typically, the value is the peak
displacement (or stress/strain) phase angle obtained via the CYCPHASE command.
Notes
Issue this command to display the results of a modal cyclic symmetry analysis.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
572
EXPAND
When you issue the EXPAND,Nrepeat command, subsequent SET commands read data from the results
file and expand them to Nrepeat sectors. As long as no entities have been modified, this expansion can
be negated (that is, reverted to single sector) by issuing EXPAND with no arguments. If you modify entities
and wish to return to the partial model, use the Session Editor (see Restoring Database Contents in the Op-
erations Guide).
EXPAND displays the results and allows you to print them, as if for a full model. The harmonic index (auto-
matically retrieved from the results file) appears in the legend column.
When plotting or printing element strain energy (SENE), the EXPAND command works with brick or tet
models only. Element kinetic energy (KENE) plotting or printing is not supported.
EXPAND is a specification command valid only in POST1. It is significantly different from the /EXPAND
command in several respects, (although you can use either command to display the results of a modal cyclic
symmetry analysis):
EXPAND has none of the limitations of the /EXPAND command.
EXPAND changes the database by modifying the geometry, the nodal displacements, and element
stresses as they are read from the results file, whereas the /EXPAND command does not change the
database.
You can print results displayed via EXPAND.
Caution
The EXPAND command creates new nodes and elements; therefore, saving (or issuing the /EXIT,
ALL command) after issuing the EXPAND command can result in large databases.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Cyc Expansion
/EXPAND, Nrepeat1, Type1, Method1, DX1, DY1, DZ1, Nrepeat2, Type2, Method2, DX2, DY2, DZ2, Nrepeat3,
Type3, Method3, DX3, DY3, DZ3
Allows the creation of a larger graphic display than represented by the actual finite element analysis
model.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
Nrepeat1, Nrepeat2, Nrepeat3
The number of repetitions required for the element pattern. The default is 0 (no expansion).
Type1, Type2, Type3
The type of expansion requested.
RECT
Causes a Cartesian transformation of DX, DY, and DZ for each pattern (default).
573
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/EXPAND
POLAR
Causes a polar transformation of DR, D-Theta and DZ for each pattern.
AXIS
Causes 2-D axisymmetric expansion (that is, rotates a 2-D model created in the X-Y plane about the
Y axis to create a 3-D model).
LRECT
Causes a Cartesian transformation of DX, DY, and DZ for each pattern about the current local coordin-
ate system (specified via the CSYS command).
LPOLAR
Causes a polar transformation of DR, D-Theta, and DZ for each pattern about the local coordinate
system (specified via the CSYS command).
Method1, Method2, Method3
The method by which the pattern is repeated.
FULL
Causes a normal repeat of the pattern (default).
HALF
Uses a symmetry transformation for alternate repeats (to produce an image of a complete circular
gear from the image of half a tooth, for example).
DX1, DY1, DZ1, DX2, DY2, DZ2, DX3, DY3, DZ3
The Cartesian or polar increments between the repeated patterns. Also determines the reflection plane.
Reflection is about the plane defined by the normal vector (DX, DY, DZ). If you want no translation,
specify a small nonzero value. For a half-image expansion, the increment DX, DY, or DZ is doubled so
that POLAR,HALF, ,45 produces full images on 90 centers, and RECT,HALF, ,1 produces full images on
2-meter centers.
Notes
You can use the /EXPAND command to perform up to three symmetry expansions at once (that is, X, Y, and
Z which is equal to going from a 1/8 model to a full model). Polar expansions allow you to expand a wheel
section into a half wheel, then into the half section, and then into the whole.
The command displays elements/results when you issue the EPLOT command or postprocessing commands.
The command works on all element and result displays, except as noted below. As the graphic display is
created, the elements (and results) are repeated as many times as necessary, expanding the geometry and,
if necessary, the displacements and stresses.
Derived results are not supported.
The /EXPAND command has the following limitations:
It does not support solid model entities.
POLAR, FULL or HALF operations are meaningful only in global cylindrical systems and are unaffected
by the RSYS or DSYS commands. Cartesian symmetry or unsymmetric operations also occur about the
global Cartesian system.
It does not average nodal results across sector boundaries, even for averaged plots (such as those ob-
tained via the PLNSOL command).
Axisymmetric harmonic element results are not supported for Type = AXIS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
574
/EXPAND
The /EXPAND command differs significantly from the EXPAND command in several respects:
The uses of /EXPAND are of a more general nature, whereas the EXPAND command is intended
primarily to expand modal cyclic symmetry results.
/EXPAND does not change the database as does the EXPAND command.
You cannot print results displayed via /EXPAND.
Menu Paths
Utility Menu>PlotCtrls>Style>Symmetry Expansion>
Utility Menu>PlotCtrls>Style>Symmetry Expansion>Expansion by values
Utility Menu>PlotCtrls>Style>Symmetry Expansion>Modal Cyclic Symmetry
Utility Menu>PlotCtrls>Style>Symmetry Expansion>Periodic/Cyclic Symmetry Expansion
EXPASS, Key
Specifies an expansion pass of an analysis.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Key
Expansion pass key:
OFF
No expansion pass will be performed (default).
ON
An expansion pass will be performed.
Notes
Specifies that an expansion pass of a modal, substructure, buckling, transient, or harmonic analysis is to be
performed.
Note
This separate solution pass requires an explicit FINISH to preceding analysis and reentry into
SOLUTION.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>ExpansionPass
Main Menu>Solution>Analysis Type>ExpansionPass
575
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EXPASS
EXPROFILE, Ldtype, Load, VALUE, Pname, Fname, Fext, Fdir
Exports ANSYS interface loads to a CFX Profile file.
POST1: Special Purpose
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Ldtype
Load type:
SURF
Surface load
VOLU
Volumetric load
Load
Surface loads:
DISP
Displacement
TEMP
Temperature
HFLU
Heat flux
Volumetric loads:
DISP
Displacement
FORC
Force
HGEN
Heat generation
VALUE
Surface or volume interface number.
Pname
Field name in CFX Profile file (32 characters maximum). Defaults to bcploadnumber for a surface load
and subdloadnumber for volumetric load.
Fname
The CFX Profile filename (248 character maximum). Defaults to jobname_bcploadnumber for a surface
load and jobname_subdloadnumber for a volumetric load.
Fext
The Profile file extension (8 character maximum). Defaults to csv.
Fdir
The Profile file directory (8 characters maximum). Defaults to current directory.
Notes
To transfer multiple loads across an interface, specify a unique file name and extension for each load.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
576
EXPROFILE
Force (FORC) and heat generation (HGEN) are per unit volume.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
EXPSOL, LSTEP, SBSTEP, TIMFRQ, Elcalc
Specifies the solution to be expanded for reduced analyses.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
LSTEP, SBSTEP
Expand the solution identified as load step LSTEP and substep SBSTEP.
TIMFRQ
As an alternative to LSTEP and SBSTEP, expand the solution at, or nearest to, the time value TIMFRQ
(for ANTYPE,TRANS or ANTYPE,SUBSTR) or frequency value TIMFRQ (for ANTYPE,HARMIC). LSTEP and
SBSTEP should be blank.
Elcalc
Element calculation key:
YES
Calculate element results, nodal loads, and reaction loads.
NO
Do not calculate these items.
Notes
Specifies the solution to be expanded from analyses that use reduced or mode superposition methods
(ANTYPE,HARMIC, TRANS, or SUBSTR). Use the NUMEXP command to expand a group of solutions.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>By Load Step
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>By Time/Freq
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>By Load Step
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>By Time/Freq
577
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EXPSOL
EXTOPT, Lab, Val1, Val2, Val3, Val4
Controls options relating to the generation of volume elements from area elements.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label identifying the control option. The meanings of Val1, Val2, and Val3 will vary depending on
Lab.
ON
Sets carryover of the material attributes, real constant attributes, and element coordinate system
attributes of the pattern area elements to the generated volume elements. Sets the pattern area
mesh to clear when volume generations are done. Val1, Val2, and Val3 are ignored.
OFF
Removes all settings associated with this command. Val1, Val2, and Val3 are ignored.
STAT
Shows all settings associated with this command. Val1, Val2, Val3, and Val4 are ignored.
ATTR
Sets carryover of particular pattern area attributes (materials, real constants, and element coordinate
systems) of the pattern area elements to the generated volume elements. (See 2 (p. 579).) Val1 can
be:
0
Sets volume elements to use current MAT command settings.
1
Sets volume elements to use material attributes of the pattern area elements.
Val2 can be:
0
Sets volume elements to use current REAL command settings.
1
Sets volume elements to use real constant attributes of the pattern area elements.
Val3 can be:
0
Sets volume elements to use current ESYS command settings.
1
Sets volume elements to use element coordinate system attributes of the pattern area elements.
Val4 can be:
0
Sets volume elements to use current SECNUM command settings.
1
Sets volume elements to use section attributes of the pattern area elements.
ESIZE
Val1 sets the number of element divisions in the direction of volume generation or volume sweep.
For VDRAG and VSWEEP, Val1 is overridden by the LESIZE command NDIV setting. Val2 sets the
spacing ratio (bias) in the direction of volume generation or volume sweep. If positive, Val2 is the
nominal ratio of last division size to first division size (if > 1.0, sizes increase, if < 1.0, sizes decrease).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
578
EXTOPT
If negative, Val2 is the nominal ratio of center division(s) size to end divisions size. Ratio defaults
to 1.0 (uniform spacing). Val3 and Val4 are ignored.
ACLEAR
Sets clearing of pattern area mesh. (See 3 (p. 579).) Val1 can be:
0
Sets pattern area to remain meshed when volume generation is done.
1
Sets pattern area mesh to clear when volume generation is done. Val2, Val3 , and Val4 are
ignored.
VSWE
Indicates that volume sweeping options will be set using Val1 and Val2. Settings specified with
EXTOPT,VSWE will be used the next time the VSWEEP command is invoked. If Lab = VSWE, Val1
becomes a label. Val1 can be:
AUTO
Indicates whether you will be prompted for the source and target used by VSWEEP or if VSWE
should automatically determine the source and target. If Val1 = AUTO, Val2 is ON by default.
VSWE will automatically determine the source and target for VSWEEP. You will be allowed to
pick more than one volume for sweeping. When Val2 = OFF, the user will be prompted for the
source and target for VSWEEP. You will only be allowed to pick one volume for sweeping.
TETS
Indicates whether VSWEEP will tet mesh non-sweepable volumes or leave them unmeshed. If
Val1 = TETS, Val2 is OFF by default. Non-sweepable volumes will be left unmeshed. When
Val2 = ON, the non-sweepable volumes will be tet meshed if the assigned element type supports
tet shaped elements.
Val3 is ignored for Lab = VSWE.
Val1, Val2, Val3, Val4
Additional input values as described under each option for Lab.
Notes
1. EXTOPT controls options relating to the generation of volume elements from pattern area elements
using the VEXT, VROTAT, VOFFST, VDRAG, and VSWEEP commands. (When using VSWEEP, the
pattern area is referred to as the source area.)
2. Enables carryover of the attributes of the pattern area elements to the generated volume elements
when you are using VEXT, VROTAT, VOFFST, or VDRAG. (When using VSWEEP, since the volume
already exists, use the VATT command to assign attributes before sweeping.)
3. When you are using VEXT, VROTAT, VOFFST, or VDRAG, enables clearing of the pattern area mesh
when volume generations are done. (When you are using VSWEEP, if selected, the area meshes on
the pattern (source), target, and/or side areas clear when volume sweeping is done.)
4. Neither EXTOPT,VSWE,AUTO nor EXTOPT,VSWE,TETS will be affected by EXTOPT,ON or EXTOPT, OFF.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Volume Sweep>Sweep Opts
Main Menu>Preprocessor>Modeling>Operate>Extrude>Elem Ext Opts
579
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EXTOPT
EXTREM, NVAR1, NVAR2, NINC
Lists the extreme values for variables.
POST26: Listing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVAR1, NVAR2, NINC
List extremes for variables NVAR1 through NVAR2 in steps of NINC. Variable range defaults to its max-
imum. NINC defaults to 1.
Notes
Lists the extreme values (and the corresponding times) for stored and calculated variables. Extremes for
stored variables are automatically listed as they are stored. Only the real part of a complex number is used.
Extreme values may also be assigned to parameters [*GET].
Menu Paths
Main Menu>TimeHist Postpro>List Extremes
EXUNIT, Ldtype, Load, Untype, Name
Indicates units assumed for an interface load for ANSYS to CFX transfer.
POST1: Special Purpose
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Ldtype
Load type:
SURF
Surface load
VOLU
Volumetric load
Load
Surface loads:
DISP
Displacement
TEMP
Temperature
HFLU
Heat flux
Volumetric loads:
DISP
Displacement
FORC
Force
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
580
EXTREM
HGEN
Heat generation
Untype
Unit type:
COMM
Pre-defined unit
USER
User-specified unit
Name
Commonly used pre-defined unit name or user-specified unit name.
SI
International System of Units (meter-kilogram-second)
FT
English System of Units (feet-pound-second)
In SI units, surface loads are: DISP m, TEMP K, and HFLU W/m
2
and volumetric loads are: DISP m, FORC
N/m
3
, and HGEN W/m
3
.
In English units, surface loads are: DISP ft, TEMP F, and HFLU BTU/sec-ft
2
and volumetric loads are: DISP
ft, FORC pdl/ft
3
, and HGEN BTU/sec-ft
3
.
Notes
This command only indicates the assumed units. It does not perform unit conversions.
A pdl is a poundal, and 1 lbf = 32.174.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
581
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
EXUNIT
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
582
F Commands
F, NODE, Lab, VALUE, VALUE2, NEND, NINC
Specifies force loads at nodes.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
NODE
Node at which force is to be specified. If ALL, NEND and NINC are ignored and forces are applied to all
selected nodes [NSEL]. If NODE = P, graphical picking is enabled and all remaining command fields are
ignored (valid only in the GUI). A component name may also be substituted for NODE.
Lab
Valid force label. Structural labels: FX, FY, or FZ (forces); MX, MY, or MZ (moments). Thermal labels: HEAT,
HBOT, HE2, HE3, . . ., HTOP (heat flow). Fluid labels: FLOW (fluid flow). Electric labels: AMPS (current flow),
CHRG (electric charge). Magnetic labels: FLUX (magnetic flux); CSGX, CSGY, or CSGZ (magnetic current
segments). FLOTRAN labels: FX, FY, or FZ (forces).
VALUE
Force value or table name reference for specifying tabular boundary conditions. To specify a table, enclose
the table name in percent signs (%), e.g., F, NODE,HEAT,%tabname%). Use the *DIM command to define
a table.
VALUE2
Second force value (if any). If the analysis type and the force allow a complex input, VALUE (above) is
the real component and VALUE2 is the imaginary component.
NEND, NINC
Specifies the same values of force at the nodes ranging from NODE to NEND (defaults to NODE), in steps
of NINC (defaults to 1).
Notes
The available force loads per node correspond to the degrees of freedom listed under "Degrees of Freedom"
in the input table for each element type in the Element Reference. If both a force and a constrained degree
of freedom [D] are specified at the same node, the constraint takes precedence. Forces are defined in the
nodal coordinate system. The positive directions of structural forces and moments are along and about the
positive nodal axis directions. The node and the degree of freedom label corresponding to the force must
be selected [NSEL, DOFSEL].
Tabular boundary conditions (VALUE = %tabname%) are available only for the following labels: Fluid
(FLOW), Electric (AMPS), Structural force (FX, FY, FZ, MX, MY, MZ), FLOTRAN (FX, FY, FZ), and Thermal (HEAT,
HBOT, HE2, HE3, . . ., HTOP). Tabular boundary conditions are valid only in static (ANTYPE,STATIC), full tran-
sient (ANTYPE,TRANS), and modal superposition transient analyses.
This command is also valid in PREP7.
583
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppCharge>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppCurrent>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>ImprCurr>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppCurrSeg>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppMagFlux>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>On Node Com-
ponents
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>MultiPtNod>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>NodePSD>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>SinglPtFor>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flow>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppCharge>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppCurrent>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>ImprCurr>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppCurrSeg>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppMagFlux>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>On Node Components
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>MultiPtNod>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>NodePSD>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>SinglPtFor>On Nodes
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flow>On Nodes
/FACET, Lab
Specifies the facet representation used to form solid model displays.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Valid labels:
FINE
Use finer tessellation to increase the number of facets for the display. Provides the best representation
(but decreases speed of operation).
NORML
Use the basic number of facets for the display (default).
COAR
Use a limited number of facets for the display. This option will increase the speed of the operations,
but may produce poor representations for some imported models.
WIRE
Display model with a wireframe representation (fast, but surfaces will not be shown).
Command Default
Basic number of facets.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
584
/FACET
Notes
Specifies the facet (or polygon) representation used to form solid model displays. Used only with the APLOT,
ASUM, VPLOT, and VSUM commands.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Solid Model Facets
FATIGUE
Specifies "Fatigue data status" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>General Postproc>Fatigue Calcs
FC, MAT, Lab1, Lab2, DATA1, DATA2, DATA3, DATA4, DATA5, DATA6
Provides failure criteria information and activates a data table to input temperature-dependent stress
and strain limits.
PREP7: Materials
POST1: Failure Criteria
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
MAT
Material reference number. You can define failure criteria for up to 250 different materials.
Lab1
Type of data.
TEMP
Temperatures. Each of the materials you define can have a different set of temperatures to define
the failure criteria.
EPEL
Strains.
585
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FC
S
Stresses.
Lab2
Specific criteria. Not used if Lab1 = TEMP.
XTEN
Allowable tensile stress or strain in the x-direction. (Must be positive.)
XCMP
Allowable compressive stress or strain in the x-direction. (Defaults to negative of XTEN.)
YTEN
Allowable tensile stress or strain in the y-direction. (Must be positive.)
YCMP
Allowable compressive stress or strain in the y-direction. (Defaults to negative of YTEN.)
ZTEN
Allowable tensile stress or strain in the z-direction. (Must be positive.)
ZCMP
Allowable compressive stress or strain in the z-direction. (Defaults to negative of ZTEN.)
XY
Allowable XY stress or shear strain. (Must be positive.)
YZ
Allowable YZ stress or shear strain. (Must be positive.)
XZ
Allowable XZ stress or shear strain. (Must be positive.)
XYCP
XY coupling coefficient (Used only if Lab1 = S). Defaults to -1.0.
YZCP
YZ coupling coefficient (Used only if Lab1 = S). Defaults to -1.0.
XZCP
XZ coupling coefficient (Used only if Lab1 = S). Defaults to -1.0.
DATA1, DATA2, DATA3, . . . , DATA6
Description of DATA1 through DATA6.
T1, T2, T3, T4, T5, T6
Temperature at which limit data is input. Used only when Lab1 = TEMP.
V1, V2, V3, V4, V5, V6
Value of limit stress or strain at temperature T1 through T6. Used only when Lab1 = S or EPEL.
Notes
The data table can be input in either PREP7 or POST1. This table is used only in POST1. When you postprocess
failure criteria results defined via the FC command (PLESOL, PRESOL, PLNSOL, PRNSOL, PRRSOL, etc.), the
active coordinate system must be the coordinate system of the material being analyzed. You do this using
RSYS, SOLU. For layered applications, you also use the LAYER command. See the specific element document-
ation in the Element Reference for information on defining your coordinate system for layers.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
586
FC
Some plotting and printing functions will not support Failure Criteria for your PowerGraphics displays. This
could result in minor changes to other data when Failure Criteria are applied. See the appropriate plot or
print command documentation for more information .
Menu Paths
Main Menu>General Postproc>Failure Criteria>Add/Edit
Main Menu>General Postproc>Failure Criteria>Temp Variation
Main Menu>Preprocessor>Material Props>Failure Criteria>Add/Edit
Main Menu>Preprocessor>Material Props>Failure Criteria>Temp Variation
FCCHECK
Checks both the strain and stress input criteria for all materials.
PREP7: Materials
POST1: Failure Criteria
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
Notes
Issue the FCCHECK command to check the completeness of the input during the input phase.
Menu Paths
Main Menu>General Postproc>Failure Criteria>Criteria Check
Main Menu>Preprocessor>Material Props>Failure Criteria>Criteria Check
FCDELE, MAT
Deletes previously defined failure criterion data for the given material.
PREP7: Materials
POST1: Failure Criteria
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
MAT
Material number. Deletes all FC command input for this material.
A value of ALL deletes all FC command input for all materials.
Notes
This command is also valid in POST1.
Menu Paths
Main Menu>General Postproc>Failure Criteria>Delete
Main Menu>Preprocessor>Material Props>Failure Criteria>Delete
587
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FCDELE
FCLIST, MAT, --, TEMP
To list what the failure criteria is that you have input.
PREP7: Materials
POST1: Failure Criteria
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
MAT
Material number (defaults to ALL for all materials).
--
Unused field.
TEMP
Temperature to be evaluated at (defaults to TUNIF).
Notes
This command allows you to see what you have already input for failure criteria using the FC commands.
Menu Paths
Main Menu>General Postproc>Failure Criteria>List
Main Menu>Preprocessor>Material Props>Failure Criteria>List
FCUM, Oper, RFACT, IFACT
Specifies that force loads are to be accumulated.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Oper
Accumulation key:
REPL
Subsequent values replace the previous values (default).
ADD
Subsequent values are added to the previous values.
IGNO
Subsequent values are ignored.
RFACT
Scale factor for the real component. Zero (or blank) defaults to 1.0. Use a small number for a zero scale
factor.
IFACT
Scale factor for the imaginary component. Zero (or blank) defaults to 1.0. Use a small number for a zero
scale factor.
Command Default
Replace previous values.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
588
FCLIST
Notes
Allows repeated force load (force, heat flow, etc.) values to be replaced, added, or ignored. Operations apply
to the selected nodes [NSEL]. and the force labels corresponding to the selected force labels [DOFSEL]. The
operations occur when the next force specifications are defined. For example, issuing the command F,1,FX,250
after a previous F,1,FX,200 causes the current value of the force on node 1 in the x-direction to be 450 with
the add operation, 250 with the replace operation, or 200 with the ignore operation. Scale factors are also
available to multiply the next value before the add or replace operation. A scale factor of 2.0 with the previous
"add" example results in a force of 700. Scale factors are applied even if no previous values exist. Issue
FCUM,STAT to show the current label, operation, and scale factors. Solid model boundary conditions are
not affected by this command, but boundary conditions on the FE model are affected.
Note
FE boundary conditions may still be overwritten by existing solid model boundary conditions if
a subsequent boundary condition transfer occurs.
FCUM does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Forces
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Forces
FDELE, NODE, Lab, NEND, NINC
Deletes force loads on nodes.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
NODE
Node for which force is to be deleted. If ALL, NEND and NINC are ignored and forces are deleted on all
selected nodes [NSEL]. If NODE = P, graphical picking is enabled and all remaining command fields are
ignored (valid only in the GUI). A component name may also be substituted for NODE.
Lab
Valid force label. If ALL, use all appropriate labels. Structural labels: FX, FY, or FZ (forces); MX, MY, or MZ
(moments). Thermal labels: HEAT, HBOT, HE2, HE3, . . ., HTOP (heat flow). Fluid labels: FLOW (fluid flow).
Electric labels: AMPS (current flow), CHRG (electric charge). Magnetic labels: FLUX (magnetic flux); CSGX,
CSGY, or CSGZ (magnetic current segments). FLOTRAN labels: FX, FY, or FZ (forces).
NEND, NINC
Delete forces from NODE to NEND (defaults to NODE) in steps of NINC (defaults to 1).
Notes
The node and the degree of freedom label corresponding to the force must be selected [NSEL, DOFSEL].
589
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FDELE
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Forces>On All Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppCharge>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppCurrent>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>ImprCurr>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Flow>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppCurrSeg>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppMagFlux>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Force/Moment>On Node Com-
ponents
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Force/Moment>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>MultiPtNod>On
Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>NodePSD>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>SinglPtFor>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Heat Flow>On Nodes
Main Menu>Solution>Define Loads>Delete>All Load Data>All Forces>On All Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppCharge>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppCurrent>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>ImprCurr>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Flow>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppCurrSeg>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppMagFlux>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Force/Moment>On Node Components
Main Menu>Solution>Define Loads>Delete>Structural>Force/Moment>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>MultiPtNod>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>NodePSD>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>SinglPtFor>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Heat Flow>On Nodes
/FDELE, Ident, Stat
Deletes a binary file after it is used.
SESSION: Files
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Ident
ANSYS file name identifier. Valid identifiers are: EMAT, ESAV, FULL, SUB, MODE, TRI, DSUB, USUB, OSAV,
and SELD. See the Basic Analysis Guide for file descriptions.
Stat
Keep or delete key:
KEEP
Keep this file.
DELE
Delete (or do not write, if not necessary) this file.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
590
/FDELE
Command Default
Keep all files.
Notes
Deletes as soon as possible (or prevents writing) a binary file created by the ANSYS program to save space.
Caution
Deleting files that are necessary for the next substep, load step, or analysis will prevent continuation
of the run.
This command is valid only at the Begin Level.
Menu Paths
Utility Menu>File>ANSYS File Options
FE, NEV, CYCLE, FACT, Title
Defines a set of fatigue event parameters.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NEV
Reference number for this event (within MXEV).
CYCLE
Number of required cycles (defaults to 1). If -1, erase all parameters and fatigue stresses for this event.
FACT
Scale factor to be applied to all loadings in this event (defaults to 1.0).
Title
User defined identification title for this event (up to 20 characters).
Command Default
Event assigned one cycle, unity scale factor, and no title.
Notes
Repeat FE command to define additional sets of event parameters (MXEV limit), to redefine event parameters,
or to delete event stress conditions.
The set of fatigue event parameters is associated with all loadings and all locations. See the FTSIZE command
for the maximum set of events (MXEV) allowed.
591
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FE
Menu Paths
Main Menu>General Postproc>Fatigue>Assign Events
Main Menu>General Postproc>Fatigue>Erase Event Data
FEBODY
Specifies "Body loads on elements" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Body Loads
FECONS
Specifies "Constraints on nodes" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>DOF Constraints
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
592
FEBODY
FEFOR
Specifies "Forces on nodes" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Forces
FELIST, NEV1, NEV2, NINC
Lists the fatigue event parameters.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
NEV1, NEV2, NINC
List event parameters from NEV1 (defaults to 1) to NEV2 (defaults to NEV1) in steps of NINC (defaults
to 1). If NEV1 = ALL, NEV2 and NINC are ignored and all events are listed. Fatigue event parameters
are defined with the FE command.
Menu Paths
Main Menu>General Postproc>Fatigue>List Event Data
FESURF
Specifies "Surface loads on elements" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
593
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FESURF
Menu Paths
Utility Menu>List>Status>Solution>Surface Loads
FILE, Fname, Ext, --
Specifies the data file where results are to be found.
POST1: Set Up
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
If Fname is blank, the extension defaults to RST (for structural, fluid, or coupled-field analyses), to RTH
(for thermal or electrical analyses), to RMG (for magnetic analyses), or to RFL (for FLOTRAN analyses). For
postprocessing reduced structural analyses in POST26, use the RDSP extension for displacements from
transient dynamic analyses or the RFRQ extension from harmonic response analyses. For postprocessing
contact results corresponding to the initial contact state in POST1, use the RCN extension.
--
Unused field.
Command Default
Use the result file with the Jobname as Fname and with the extension corresponding to the analysis type.
Notes
Specifies the ANSYS data file where the results are to be found for postprocessing.
Menu Paths
Main Menu>General Postproc>Data & File Opts
Main Menu>TimeHist Postpro>Settings>File
Utility Menu>File>List>Binary Files
Utility Menu>List>Files>Binary Files
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
594
FILE
FILEAUX2, Fname, Ident, --
Specifies the binary file to be dumped.
AUX2: Binary Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to the current Jobname if Ident is specified.
Ident
ANSYS filename identifier. See the Basic Analysis Guide for file descriptions and identifiers. If not an ANSYS
identifier, Ident will be used as the filename extension.
--
Unused field.
Notes
Specifies the binary file to be dumped with the DUMP command.
Menu Paths
Utility Menu>File>List>Binary Files
Utility Menu>List>Files>Binary Files
FILEAUX3, Fname, Ext, --
Specifies the results file to be edited.
AUX3: Results Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to the current Jobname if Ext is specified.
Ext
Filename extension (8 character maximum).
--
Unused field.
Notes
Specifies the results file to be edited.
595
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FILEAUX3
Menu Paths
This command cannot be accessed from a menu.
FILEDISP, Fname, Ext
Specifies the file containing the graphics data.
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
Notes
Specifies the input file containing the graphics data (defaults to File.GRPH).
Menu Paths
It is part of the DISPLAY program.
FILL, NODE1, NODE2, NFILL, NSTRT, NINC, ITIME, INC, SPACE
Generates a line of nodes between two existing nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1, NODE2
Beginning and ending nodes for fill-in. NODE1 defaults to next to last node specified, NODE2 defaults
to last node specified. If NODE1 = P, graphical picking is enabled and all remaining command fields are
ignored (valid only in the GUI).
NFILL
Fill NFILL nodes between NODE1 and NODE2 (defaults to |NODE2-NODE1|-1). NFILL must be positive.
NSTRT
Node number assigned to first filled-in node (defaults to NODE1 + NINC).
NINC
Add this increment to each of the remaining filled-in node numbers (may be positive or negative). Defaults
to the integer result of (NODE2-NODE1)/(NFILL + 1), i.e., linear interpolation. If the default evaluates to
zero, or if zero is input, NINC is set to 1.
ITIME, INC
Do fill-in operation a total of ITIMEs, incrementing NODE1, NODE2 and NSTRT by INC each time after
the first. ITIME and INC both default to 1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
596
FILEDISP
SPACE
Spacing ratio. Ratio of last division size to first division size. If > 1.0, divisions increase. If < 1.0, divisions
decrease. Ratio defaults to 1.0 (uniform spacing).
Notes
Generates a line of nodes (in the active coordinate system) between two existing nodes. The two nodes may
have been defined in any coordinate system. Nodal locations and rotation angles are determined by inter-
polation. Any number of nodes may be filled-in and any node number sequence may be assigned. See the
CSCIR command when filling across the 180 singularity line in a non-Cartesian system.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>Fill between Nds
FILLDATA, IR, LSTRT, LSTOP, LINC, VALUE, DVAL
Fills a variable by a ramp function.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Define data table as variable IR (2 to NV [NUMVAR]).
LSTRT
Start at location LSTRT (defaults to 1).
LSTOP
Stop at location LSTOP (defaults to maximum location as determined from data previously stored.
LINC
Fill every LINC location between LSTRT and LSTOP (defaults to 1).
VALUE
Value assigned to location LSTRT.
DVAL
Increment value of previous filled location by DVAL and assign sum to next location to be filled (may
be positive or negative.)
Notes
Locations may be filled continuously or at regular intervals (LINC). Previously defined data at a location will
be overwritten.
Menu Paths
Main Menu>TimeHist Postpro>Table Operations>Fill Data
597
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FILLDATA
/FILNAME, Fname, Key
Changes the Jobname for the analysis.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
Name (32 characters maximum) to be used as the Jobname. Defaults to the initial Jobname as specified
on the ANSYS execution command, or to file if none specified.
Key
Specify whether to use the existing log, error, lock, and page files or start new files.
0, OFF
Continue using current log, error, lock, and page files.
1, ON
Start new log, error, lock, and page files (old log and error files are closed and saved, but old lock
and page files are deleted). Existing log and error files are appended.
Notes
All subsequently created files will be named with this Jobname if Key = 0. Use Key = 1 to start new log,
error, lock, and page files. The previous Jobname is typically defined on the ANSYS program execution line
(see the Operations Guide). This command is useful when different groups of files created throughout the
run are to have different names. For example, the command may be used before each substructure pass to
avoid overwriting files or having to rename each file individually.
This command is valid only at the Begin level.
Menu Paths
Utility Menu>File>Change Jobname
FINISH
Exits normally from a processor.
SESSION: Processor Entry
DISPLAY: Action
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Exits any of the ANSYS processors or the DISPLAY program. For the ANSYS processors, data will remain intact
in the database but the database is not automatically written to a file (use the SAVE command to write the
database to a file). See also the /QUIT command for an alternate processor exit command. If exiting POST1,
POST26, or OPT, see additional notes below.
POST1: Data in the database will remain intact, including the POST1 element table data, the path table data,
the fatigue table data, and the load case pointers.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
598
/FILNAME
POST26: Data in the database will remain intact, except that POST26 variables are erased and specification
commands (such as FILE, PRTIME, NPRINT, etc.) are reset. Use the /QUIT command to exit the processor
and bypass these exceptions.
OPT: Current optimization data are written to File.OPT for possible resume later [OPRESU]. See also the
OPSAVE command to write optimization data.
This command is valid in any processor. This command is not valid at the Begin level.
Menu Paths
Main Menu>DesignXplorer>Solution>Solve
Main Menu>Finish
FITEM, NFIELD, ITEM, ITEMY, ITEMZ
Identifies items chosen by a picking operation (GUI).
DATABASE: Picking
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NFIELD
Field number on the command which uses the picking data. (Count the command name as a field, so
that a 2 indicates the first command argument, 3 the second command argument, etc.) The corresponding
field on the command will have a P51X label.
ITEM
Entity number of the entity picked. Negative entity numbers are used to indicate a range of entities. If
the item picked is a coordinate location, then this field represents the X-coordinate. See also the FLST
command.
ITEMY, ITEMZ
Y and Z coordinates of a picked coordinate location. ITEM represents the X coordinate. See also the
FLST command.
Notes
This is a command generated by the GUI and will appear in the log file (Jobname.LOG) if graphical picking
is used. This command is not intended to be typed in directly in an ANSYS session (although it can be included
in an input file for batch input or for use with the /INPUT command).
On the log file, a set of FITEM commands is preceded by one FLST command which defines the picking
specifications for that pick operation. The data listed in the FITEM commands are used by the first subsequent
command containing a P51X label in one of its fields.
Caution
For a given entity type, a list containing an ITEM that is larger than the maximum defined entity,
could deplete the system memory and produce unpredictable results.
This command is valid in any processor.
599
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FITEM
Menu Paths
This command cannot be accessed from a menu.
FJ, ELEM, LABEL, VALUE
Specify forces or moments on the components of the relative motion of a joint element.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
ELEM
Element number or ALL to specify all joint elements.
LABEL
Valid labels:
FX
Force in local x direction.
FY
Force in local y direction.
FZ
Force in local z direction.
MX
Moment about local x axis.
MY
Moment about local y axis.
MZ
Moment about local z axis.
VALUE
Value of the label.
Notes
Valid for MPC184 (joint options in KEYOPT(1)).
See FJDELE for information on deleting forces and moments.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>On Joint Elems
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>On Joint Elems
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
600
FJ
FJDELE, ELEM, LAB
Deletes forces (or moments) on the components of the relative motion of a joint element.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
ELEM
Element number, or ALL. (leaving this blank defaults to ALL)
LAB
Valid labels are:
FX
Force in local x direction.
FY
Force in local y direction.
FZ
Force in local z direction.
MX
Moment about local x axis.
MY
Moment about local y axis.
MZ
Moment about local z axis.
ALL, or (blank)
Delete all valid forces or moments.
Notes
Valid for MPC184 (joint options in KEYOPT(1)).
See FJ for information on specifying forces (or moments).
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Forces>On Joint Elems
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Force/Moment>On Joint Elems
Main Menu>Solution>Define Loads>Delete>All Load Data>All Forces>On Joint Elems
Main Menu>Solution>Define Loads>Delete>Structural>Force/Moment>On Joint Elems
FJLIST, Elem
Lists forces and moments applied on joint elements.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Elem
Element number or ALL (or blank). Lists joint element forces and moments on the specified element(s).
601
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FJLIST
Notes
Valid for MPC184 joint elements. See FJ for information on specifying forces and moments.
Menu Paths
Utility Menu>List>Loads>Joint Element Forces>On Picked Element
FK, KPOI, Lab, VALUE, VALUE2
Defines force loads at keypoints.
SOLUTION: Solid Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KPOI
Keypoint at which force is to be specified. If ALL, apply to all selected keypoints [KSEL]. If KPOI = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for KPOI.
Lab
Valid force label. Structural labels: FX, FY, or FZ (forces); MX, MY, or MZ (moments). Thermal labels: HEAT,
HBOT, HE2, HE3, . . ., HTOP (heat flow). Fluid labels: FLOW (fluid flow). Electric labels: AMPS (current flow),
CHRG (electric charge). Magnetic labels: FLUX (magnetic flux); CSGX, CSGY, or CSGZ (magnetic current
segments).
VALUE
Force value or table name reference for specifying tabular boundary conditions. To specify a table, enclose
the table name in percent signs (%), e.g., FK, KPOI, HEAT,%tabname%). Use the *DIM command to
define a table.
VALUE2
Second force value (if any). If the analysis type and the force allow a complex input, VALUE (above) is
the real component and VALUE2 is the imaginary component.
Notes
Forces may be transferred from keypoints to nodes with the FTRAN or SBCTRAN commands. See the F
command for a description of force loads.
Tabular boundary conditions (VALUE = %tabname%) are available only for the following labels: Fluid
(FLOW), Electric (AMPS), Structural force (FX, FY, FZ, MX, MY, MZ), and Thermal (HEAT, HBOT, HE2, HE3, . . .,
HTOP).
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppCharge>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppCurrent>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>ImprCurr>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppCurrSeg>On Keypoints
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
602
FK
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppMagFlux>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>MultiPtNod>On
Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>NodePSD>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Spectrum>SinglPtFor>On Key-
points
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flow>On Keypoints
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppCharge>On Keypoints
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppCurrent>On Keypoints
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>ImprCurr>On Keypoints
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppCurrSeg>On Keypoints
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppMagFlux>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>MultiPtNod>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>NodePSD>On Keypoints
Main Menu>Solution>Define Loads>Apply>Structural>Spectrum>SinglPtFor>On Keypoints
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flow>On Keypoints
FKDELE, KPOI, Lab
Deletes force loads at a keypoint.
SOLUTION: Solid Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KPOI
Keypoint at which force is to be deleted. If ALL, delete forces at all selected keypoints [KSEL]. If KPOI
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may also be substituted for KPOI.
Lab
Valid force label. If ALL, use all appropriate labels. See the FDELE command for labels.
Notes
Deletes force loads (and all corresponding finite element loads) at a keypoint. See the FDELE command for
details.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Forces>On All KPs
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppCharge>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppCurrent>On Key-
points
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>ImprCurr>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Flow>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppCurrSeg>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppMagFlux>On Keypoints
603
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FKDELE
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Force/Moment>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>MultiPtNod>On
Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>NodePSD>On Key-
points
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Spectrum>SinglPtFor>On Key-
points
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Heat Flow>On Keypoints
Main Menu>Solution>Define Loads>Delete>All Load Data>All Forces>On All KPs
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppCharge>On Keypoints
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppCurrent>On Keypoints
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>ImprCurr>On Keypoints
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Flow>On Keypoints
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppCurrSeg>On Keypoints
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppMagFlux>On Keypoints
Main Menu>Solution>Define Loads>Delete>Structural>Force/Moment>On Keypoints
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>MultiPtNod>On Keypoints
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>NodePSD>On Keypoints
Main Menu>Solution>Define Loads>Delete>Structural>Spectrum>SinglPtFor>On Keypoints
Main Menu>Solution>Define Loads>Delete>Thermal>Heat Flow>On Keypoints
FKLIST, KPOI, Lab
Lists the forces at keypoints.
SOLUTION: Solid Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KPOI
List forces at this keypoint. If ALL (default), list for all selected keypoints [KSEL]. If KPOI = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for KPOI.
Lab
Force label to be listed (defaults to ALL). See the DOFSEL command for labels.
Notes
Listing applies to the selected keypoints [KSEL] and the selected force labels [DOFSEL].
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Forces>On All Keypoints
Utility Menu>List>Loads>Forces>On Picked KPs
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
604
FKLIST
FL, NLOC, NODE, SCFX, SCFY, SCFZ, Title
Defines a set of fatigue location parameters.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Reference number for this location (within MXLOC). When defining a new location, defaults to lowest
unused location. If the specified NODE is already associated with a location, NLOC defaults to that existing
location.
NODE
Node number corresponding to this location (must be unique). Used only to associate a node with a
new location or to find an existing location (if NLOC is not input). If NODE = -1 (or redefined), erase all
parameters and fatigue stresses for this location.
SCFX, SCFY, SCFZ
Stress concentration factors applied to the total stresses. Factors are applied in the global X, Y, and Z
directions unless the axisymmetric option of the FSSECT is used (i.e., RHO is nonzero), in which case the
factors are applied in the section x, y, and z (radial, axial, and hoop) directions.
Title
User-defined title for this location (up to 20 characters).
Notes
Repeat FL command to define additional sets of location parameters (MXLOC limit), to redefine location
parameters, or to delete location stress conditions.
One location must be defined for each node of interest and only one node can be associated with each
location. See the FTSIZE command for the maximum locations (MXLOC) allowed. A location will be automat-
ically defined for a node not having a location when the FSSECT, FSNODE, or FS command is issued.
Automatically defined locations are assigned the lowest available location number, unity stress concentration
factors, and no title.
Menu Paths
Main Menu>General Postproc>Fatigue>Stress Locations
FLANGE, NLOC, LENG, MASS, SIF, FLEX, ARINS, ELEM
Defines a flange in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Node where flange is to be placed (as described below). Defaults to current piping run starting point.
LENG
Length of flange (defaults to larger pipe OD).
605
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLANGE
MASS
Dry mass (weight/gravity) of flange without insulation (defaults to equivalent straight pipe mass). Note
that acceleration [ACEL] must be nonzero for weight to be calculated.
SIF
Stress intensification factor (defaults to 1.0).
FLEX
Bending flexibility factor (defaults to 1.0).
ARINS
Insulation surface area (defaults to equivalent straight pipe insulation area). Units (length
2
) must be
consistent with the smallest unit of the system used (not mixed) regardless of the PUNIT option.
ELEM
Element number to be assigned to flange (defaults to the previous maximum element number (MAXEL)
+ 1).
Notes
Defines a flange (straight pipe element (PIPE16) with adjusted specifications and loadings) at a given location
in a piping run. See the PREP7 RUN command. This command is similar to the VALVE command except for
a different flexibility factor default. The location may be 1) between two adjacent colinear straight pipes, 2)
between an adjacent straight pipe and a different piping component, or 3) at the end of a straight pipe.
For Case 1, two new nodes are generated at the ends of the flange. The two straight pipes are automatically
"shortened" to meet the ends of the flange. The flange specifications and loadings are taken from the cor-
responding two straight pipes.
For Case 2, one new node is generated at one end of the flange. The straight pipe is automatically "shortened"
to meet this end of the flange. The other end of the flange meets the other piping component. The flange
specifications and loadings are taken from the straight pipe.
For Case 3, one new node is generated at the free end of the flange. The other end of the flange meets the
straight pipe. The flange specifications and loadings are taken from the straight pipe.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Define Pipes>Flange
FLDATA, Name, Label, Value
Sets up a FLOTRAN analysis.
PREP7: FLOTRAN Options
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
Name
The name identifying the group of FLOTRAN parameters being defined or controlled on this command.
Label
The label of the specific FLOTRAN parameter being input or controlled. Label determines the meaning
of the Value argument.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
606
FLDATA
Value
The numeric value of an input item, the logical value of a switch (T or F, for example), or an alphanumeric
label, depending on the Label argument.
Notes
The FLDATA command is used to define FLOTRAN-specific input data, solution controls, and output controls.
It is valid only with the FLOTRAN CFD option.
The FLDATA command controls groups of FLOTRAN parameters, and the group name is input as the first
argument, Name.
To give you the ability to link directly to the documentation on any of these groups, we have documented
each group under its own name--from FLDATA1 through FLDATA36. For example, documentation on
FLDATA with the first argument set to SOLU is documented as the FLDATA1 command. You may enter the
command with either name--FLDATA or FLDATA1, and you must remember to input the appropriate first
argument, as shown in the list below.
Option and Output Control Commands
Controls which features of the solution algorithm are activated. FLDATA1,SOLU
Sets iteration and output controls for steady state analyses. FLDATA2,ITER
Sets the convergence monitors for the pressure and temperature
equations.
FLDATA3,TERM
Sets controls for transient analyses based on transient time and con-
vergence monitors or sets time integration method.
FLDATA4,TIME
Sets output controls for transient analyses based on number of time
steps.
FLDATA4A,STEP
Sets output and storage controls. FLDATA5,OUTP
Controls the output of the convergence monitor. FLDATA6,CONV
Property Definition Commands
Specifies the type of fluid property. FLDATA7,PROT
Specifies the NOMI coefficient of the fluid property equation. FLDATA8,NOMI
Specifies the COF1 coefficient of the fluid property equation. FLDATA9,COF1
Specifies the COF2 coefficient of the fluid property equation. FLDATA10,COF2
Specifies the COF3 coefficient of the fluid property equation. FLDATA11,COF3
Sets the property update frequency flag. FLDATA12,PROP
Sets the property variation flag. FLDATA13,VARY
Operating Condition Commands
Specifies the reference temperature. FLDATA14,TEMP
Specifies the reference pressure. FLDATA15,PRES
Specifies the bulk modulus parameter. FLDATA16,BULK
Specifies the specific heat ratio. FLDATA17, GAMM
Algebraic Solver Commands
Selects the algebraic solver. FLDATA18, METH
607
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA
Specifies the number of TDMA sweeps. FLDATA19, TDMA
Specifies the number of conjugate direction search vectors. FLDATA20,SRCH
Specifies the amount of fill-in when preconditioning the coefficient
matrix.
FLDATA20A,PGMR
Specifies the number of fill-ins for the ILU preconditioner. FLDATA20B,PBCGM
Specifies the solver convergence criterion. FLDATA21,CONV
Maximum number of semi-direct iterations. FLDATA22,MAXI
Specifies the solver minimum normalized rate of change. FLDATA23,DELT
Turbulence Commands
Sets the turbulence model and the constants used in the Standard k-
Model and the Zero Equation Turbulence Model.
FLDATA24,TURB
Sets constants for the Re-Normalized Group Turbulence Model (RNG). FLDATA24A,RNGT
Sets constants for the k- Turbulence Model due to Shih (NKE). FLDATA24B,NKET
Sets constants for the Nonlinear Turbulence Model of Girimaji (GIR). FLDATA24C, GIRT
Sets constants for the Shih, Zhu, Lumley Turbulence Model (SZL). FLDATA24D, SZLT
Sets constants for the k- turbulence model. FLDATA24E, SKWT
Sets the turbulent production clip factor for the Shear Stress Transport
(SST) turbulence model.
FLDATA24F, SST1
Sets constants in the k- regime for the Shear Stress Transport (SST)
turbulence model.
FLDATA24G, SST1
Sets constants in the k- regime for the Shear Stress Transport (SST)
turbulence model.
FLDATA24H, SST2
Stability Control Commands
Sets solution and property relaxation factors. FLDATA25, RELX
Sets stability controls. FLDATA26,STAB
Miscellaneous Commands
Controls printing flags. FLDATA27,PRIN
Specifies that variable results are to be replaced. FLDATA28,MODR
Re-initializes a results variable. FLDATA29,MODV
Controls the quadrature orders. FLDATA30,QUAD
Specifies dependent variable caps. FLDATA31,CAPP
Controls restart options. FLDATA32,REST
Specifies the approach to discretize the advection term. FLDATA33,ADVM
Sets modified inertial relaxation factors. FLDATA34,MIR
Specifies tolerances for the lower and upper bound of the volume
fraction.
FLDATA35,VFTOL
Specifies ambient reference values outside of the fluid for the volume
of fluid (VOF) method.
FLDATA36,AMBV
Specifies segregated solution or film coefficient algorithms. FLDATA37,ALGR
Specifies the mass type for a fluid transient analysis. FLDATA38,MASS
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
608
FLDATA
Specifies remeshing parameters for transient fluid flow and fluid-solid
interaction analyses.
FLDATA39,REMESH
Controls activation of thermal stabilization near walls. FLDATA40,WADV
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Advection
Main Menu>Preprocessor>FLOTRAN Set Up>Algorithm Ctrl
Main Menu>Preprocessor>FLOTRAN Set Up>Execution Ctrl
Main Menu>Solution>FLOTRAN Set Up>Advection
Main Menu>Solution>FLOTRAN Set Up>Algorithm Ctrl
Main Menu>Solution>FLOTRAN Set Up>Execution Ctrl
FLDATA1, SOLU, Label, Value
Controls which features of the solution algorithm are activated.
PREP7: FLOTRAN Options
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SOLU
Enter the word SOLU in this field.
FLDATA1,SOLU is the FLDATA command with its first argument set to SOLU. It can be entered into the
program as either FLDATA1,SOLU,Label,Value or FLDATA,SOLU,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Solution algorithm choices:
FLOW
Solves the momentum and pressure equations. Defaults to TRUE (on).
TURB
Activates the turbulence model. Defaults to FALSE (off ).
TEMP
Solves the temperature equation. Defaults to FALSE (off ).
COMP
Uses the compressible algorithm. Defaults to FALSE (off ).
SWRL
Activates the swirl option. Defaults to FALSE (off ).
TRAN
Activates the transient solution algorithm. Defaults to FALSE (off ).
SPEC
Activates multiple species transport. Defaults to FALSE (off ).
IVSH
Activates viscous heating for incompressible flow. Defaults to FALSE (off ).
609
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA1
VOF
Activates volume of fluid method. Defaults to FALSE (off ).
SFTS
Activates surface tension. Defaults to FALSE (off ).
ALE
Activates Arbitrary Lagrangian-Eulerian formulation. Defaults to FALSE (off ).
RDSF
Activates the surface-to-surface radiation solution. Defaults to TRUE (on).
Value
Value controlling Label:
TRUE or T
Turn this feature on.
FALSE or F
Turn this feature off.
Notes
Repeat command to set each Label as required.
The analyst must choose the appropriate features, as FLOTRAN will not determine, for example, whether or
not the case is turbulent or whether or not the compressible algorithm is appropriate.
SWRL should only be activated for axisymmetric cases when there is a velocity component normal to the
axisymmetric plane.
Generally, the use of the compressible algorithm (COMP) is not warranted for Mach numbers less than 0.3.
Density may be assumed to vary via the ideal gas law without activating the compressible option.
If fluid properties are not a function of temperature in non-adiabatic flow problems, it is not necessary to
activate the flow (FLOW) and temperature (TEMP) solutions together. First solve the flow problem and then
restart to solve the temperature equation.
For compressible flow, FLOTRAN automatically includes viscous heating.
When the Arbitrary Lagrangian-Eulerian (ALE) formulation is on , FLOTRAN allows the fluid nodes to move
in a manner that satisfies the displacement boundary conditions.
See also the FLDATA2,ITER, FLDATA3,TERM, FLDATA4,TIME, FLDATA4A,STEP, FLDATA5,OUTP, and
FLDATA6,CONV commands for other Solution and Output Controls.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Solution Options
Main Menu>Solution>FLOTRAN Set Up>Solution Options
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
610
FLDATA1
FLDATA2, ITER, Label, Value
Sets iteration and output controls for steady state analyses.
PREP7: FLOTRAN Options
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
ITER
Enter the word ITER in this field.
FLDATA2,ITER is the FLDATA command with its first argument set to ITER. It can be entered into the
program as either FLDATA2,ITER,Label,Value or FLDATA,ITER,Label,Value where Label and Value
are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Type of iteration control (dictates the meaning of Value):
EXEC
Value is the number of global iterations, defined as the sequential solution of the governing
equations of all the features activated, to be performed during this execution of FLOTRAN if the case
is steady state. Defaults to Value = 10.
APPE
Value is the number of steady state global iterations between appends to the results file (Job-
name.RFL). This feature is used to ensure that restarts can be made from earlier points if unforeseen
difficulties occur before the end of the analysis. The default value implies that results will be saved
only for the initial (0
th
) iteration and the final global iteration. Saving many intermediate results can
produce a large results file. Defaults to Value = 0 (implies FLDATA2,ITER,EXEC).
If you are creating domain files (jobname.pv_0000n) for use by ICEM CFD's PV3 postprocessing
visualization tool, this sets the frequency with which domain files are written. See FLDATA5 for more
details.
OVER
Value is the number of steady state global iterations between overwrites of a temporary set of
results in the results file (Jobname.RFL). (This temporary set of results will itself be overwritten by
the set stored for Label = APPE or at the end of the run.) OVER enables the user to retain the most
up-to-date set of results without having the results file continue to grow in size. Defaults to Value
= 0 (no overwrites will be made).
Value
Number of iterations for Label above.
Notes
Repeat command to set each Label as required.
Sets the number of global iterations to control the length of execution for steady state analyses, the frequency
with which results are added to the results file (Jobname.RFL), and the frequency with which results are
overwritten in the results file. Also sets the frequency with which the domain files, used by the ICEM PV3
postprocessing visualization tool, are written. For more information about domain files, see the
FLDATA5,OUTP,DOMA command.
See also the FLDATA1,SOLU, FLDATA3,TERM, FLDATA4,TIME, FLDATA4A,STEP, FLDATA5,OUTP, and
FLDATA6,CONV commands for other Solution and Output Controls.
611
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA2
This command is accessible in the menu if FLDATA1,SOLU,TRAN,FALSE has been issued (default).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Execution Ctrl
Main Menu>Solution>FLOTRAN Set Up>Execution Ctrl
FLDATA3, TERM, Label, Value
Sets the convergence monitors for the degree of freedom set.
PREP7: FLOTRAN Options
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
TERM
Enter the word TERM in this field.
FLDATA3,TERM is the FLDATA command with its first argument set to TERM. It can be entered into the
program as either FLDATA3,TERM,Label,Value or FLDATA,TERM,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Type of convergence monitor:
PRES
Steady-state run will not terminate until the convergence monitor for pressure falls below Value,
unless the specified number of global iterations has been executed. Defaults to Value = 1.0 x 10
-8
.
TEMP
Steady-state run (with the temperature equation solution activated) will not terminate until the
convergence monitor for temperature falls below Value, unless the specified number of global iter-
ations has been executed. Defaults to Value = 1.0 x 10
-8
.
VX
Steady-state run will not terminate until the convergence monitor for X velocity component falls
below Value, unless the specified number of global iterations has been executed. Defaults to Value
= 1.0 x 10
-2
.
VY
Steady-state run will not terminate until the convergence monitor for Y velocity component falls
below Value, unless the specified number of global iterations has been executed. Defaults to Value
= 1.0 x 10
-2
.
VZ
Steady-state run will not terminate until the convergence monitor for Z velocity component falls
below Value, unless the specified number of global iterations has been executed. Defaults to Value
= 1.0 x 10
-2
.
ENKE
Steady-state run will not terminate until the convergence monitor for turbulent kinetic energy falls
below Value, unless the specified number of global iterations has been executed. Defaults to Value
= 1.0 x 10
-2
.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
612
FLDATA3
ENDS
Steady-state run will not terminate until the convergence monitor for turbulence dissipation falls
below Value, unless the specified number of global iterations has been executed. Defaults to Value
= 1.0 x 10
-2
.
Value
Value of convergence monitor criterion, above.
Notes
Repeat command to set each Label as required.
The convergence monitors appear as output for each degree of freedom for each global iteration. The value
is calculated for each degree of freedom individually by summing the absolute value of the change in
solution between global iterations for all the nodes and dividing it by the sum of the absolute values of the
solution for all the nodes. It is an approximation of the normalized rate of change of the solution between
global iterations.
All specified criteria must be met before the case is terminated.
If a termination criterion for a specific label is set negative, the termination check is ignored for that partic-
ular DOF.
See also the FLDATA1,SOLU, FLDATA2,ITER,FLDATA4,TIME, FLDATA4A,STEP, FLDATA5,OUTP, and
FLDATA6,CONV commands for other Solution and Output Controls.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Execution Ctrl
Main Menu>Solution>FLOTRAN Set Up>Execution Ctrl
FLDATA4, TIME, Label, Value
Sets controls for transient analyses based on transient time and convergence monitors or sets time integ-
ration method.
PREP7: FLOTRAN Options
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
TIME
Enter the word TIME in this field.
FLDATA4,TIME is the FLDATA command with its first argument set to TIME. It can be entered into the
program as either FLDATA4,TIME,Label,Value or FLDATA,TIME,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Type of transient analysis control or time integration method (dictates the meaning of Value):
613
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA4
STEP
Value controls the time step size. If Value is greater than zero, Value is the time step size. If
Value is less than zero, the program chooses the time step, according to the following (Value
defaults to -1):
1
The resulting time step will be small enough to prevent the passage of an arbitrarily small fluid
parcel through more than one element length in a single time step.
2
This choice is only applicable for compressible analyses, and is equivalent to the Courant limit.
The time step is small enough to prevent a pressure signal from propagating through more than
one element during a single time step.
3
This choice is only applicable for compressible analyses. The smaller of the two time steps com-
puted for options -1 and -2 is used.
4
This choice is applicable to "conduction only" cases (FLDATA1,SOLU,FLOW,F). The resulting time
step prevents an arbitrarily small "parcel of heat" from diffusing or conducting through more
than one element within a time step.
ISTEP
Value is the time step size for the first time step in an analysis using a FLOTRAN-calculated time
step size (STEP = -1 through -4).
BC
Value is a flag indicating whether a transient boundary condition should be applied as a step
change (when Value = 0) or as a linear ramp (Value = 1). This label is analogous to the ANSYS
command KBC, except that the default for FLOTRAN is a step change. Tabular boundary conditions
do not support ramping and instead apply their full value regardless of the KBC setting.
NUMB
Value is the number of time steps which will be executed unless the transient end time has been
encountered first.
GLOB
Value is the number of global iterations per time step. The user may elect to set this to a high value
(typically between 30 and 50), and expect the time step to converge to the convergence criterion
set by the FLDATA3,TERM command before completing Value iterations. Default to Value = 10.
TEND
Value is the transient end time. This time is used for the calculation of ramped transient boundary
conditions. The case will execute until this end time if the number of time steps specified is large
enough. Defaults to Value = 1.0 x 10
+6
.
APPE
Value is the results output frequency based on transient time. Every Value seconds in the transient,
results are written to the Jobname.RFL file. If you are creating domain files (jobname.pv_0000n)
for use by ICEM CFD's PV3 postprocessing visualization tool, this sets the frequency with which domain
files are written. See FLDATA5 for more details. Defaults to Value = 1.0 x 10
+6
.
SUMF
Value is the output summary frequency based on transient time. Every Value seconds in the
transient, a results summary is written to the Jobname.PFL file.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
614
FLDATA4
Note
A summary will always be written when a set of results is written to the Jobname.RFL
file. Defaults to Value = 1.0 x 10
+6
.
OVER
Value is the time interval between overwrites of the temporary set of results in the results file
(Jobname.RFL). This feature enables the user to retain the most up-to-date set of results without
having the results file continue to grow in size. Defaults to Value = 0 (no overwrites will be made).
PRES
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for pressure falls below Value. Defaults to Value = 1.0 x
10
-6
.
TEMP
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for temperature falls below Value. Defaults to Value = 1.0
x 10
-6
.
VX
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for X velocity component falls below Value. Defaults to Value
= 1.0 x 10
-2
.
VY
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for Y velocity component falls below Value. Defaults to Value
= 1.0 x 10
-2
.
VZ
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for Z velocity component falls below Value. Defaults to Value
= 1.0 x 10
-2
.
ENKE
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for turbulent kinetic energy falls below Value. Defaults to
Value = 1.0 x 10
-2
.
ENDS
If the maximum number of global iterations per time step has not been reached, the time step will
terminate if the convergence monitor for turbulence dissipation falls below Value. Defaults to
Value = 1.0 x 10
-2
.
NTVF
Value controls the time stepping strategy for Volume of Fluid (VOF) Advection. The value gives the
number of time steps used in the VOF Advection for each time step in the FLOTRAN solution. Defaults
to 1.
METH
Value specifies the time integration method. Allowable choices for Value are:
BACK
Backward method (default).
615
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA4
NEWM
Newmark time integration method.
DELT
Newmark parameter to control integration accuracy and stability. It must be
p
p
p
t
C
k
2
which is used in the update of the turbulent viscosity. Value defaults to 0.09.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
648
FLDATA24
C1
Value is the k- turbulence model C1 constant. It is the multiplier of the shear rate generation term
of the turbulent kinetic energy dissipation rate equation. Value defaults to 1.44.
C2
Value is the k- turbulence model C2 constant. It is the multiplier of the dissipation source term in
the turbulent kinetic energy dissipation rate equation. Value defaults to 1.92.
BUC3
Value is the k- buoyancy model constant. A value of zero means that there is no contribution to
the turbulent kinetic energy dissipation rate equation. The default value of one is appropriate for
stable thermally stratified flows. A value of zero is appropriate for unstable thermally stratified flows
BUC4
Value is the k- multiplier applied to the buoyancy term of the turbulent kinetic energy equation.
A value of 1.0 is appropriate for the calculation of stable thermally stratified flows. Value defaults
to 0.0.
BETA
Value is , the coefficient of thermal expansion
1
T
This term is used in the buoyancy terms of the k- model.
KAPP
Value is the slope parameter of the law of the wall constant. It is the slope of the plot of normalized
shear velocity (u
+
) versus the nondimensionalized distance from the wall (y
+
). See Theory Reference
for the Mechanical APDL and Mechanical Applications for details. Value defaults to 0.4.
EWLL
Value is the law of the wall constant. It is related to the y intercept value for a plot of normalized
shear velocity (u
+
) versus the nondimensionalized distance from the wall (y
+
). See the Theory Reference
for the Mechanical APDL and Mechanical Applications for more details. Value defaults to 9.0.
WALL
Value is the choice of wall conductivity model. The default model is the Van Driest model (Value
= VAND), used most often for high Prandtl number fluids. The second choice is the Spalding model
(Value = SPAL), applicable to low Prandtl number fluids. The third choice is the Equilibrium model
(Value = EQLB). The equilibrium model is also automatically invoked for the wall viscosity by this
command.
VAND
Value is the constant in the Van Driest wall conductivity model. See the Theory Reference for the
Mechanical APDL and Mechanical Applications for details. Value defaults to 26.0.
TRAN
Value is the magnitude of y
+
marking the outer boundary of the laminar sublayer. Used only for
the Equilibrium Wall model. Value defaults to 11.5.
ZELS
Value is the Zero Equation Model length scale (defaults to -1). A negative value means that FLOTRAN
will calculate the value internally.
649
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA24
KS
Value is the local uniform wall roughness in length units. The default value of 0.0 implies a smooth
wall.
CKS
Value is an empirical dimensionless factor between 0.5 and 1.0 that specifies the degree of
nonuniformity of the surface. The default value of 0.5 means that the roughness signified by KS is
uniformly distributed. Higher values increase the roughness losses without changing the flow regime
implied by the value of KS.
Value
Turbulence model parameters values as explained above.
Notes
Sets the turbulence model and constants used in the Standard k- Model and the Zero Equation Turbulence
Model. The other turbulence models are an extension of the Standard k- Model and different values are
applied to some of the Standard k- Model constants and some additional constants are added.
Applicable only if the turbulence model is activated [FLDATA1,SOLU,TURB, TRUE]. The default values are
used most often. See the Theory Reference for the Mechanical APDL and Mechanical Applications for more in-
formation on the turbulence models.
See also the FLDATA24A,RNGT, FLDATA24B,NKET, FLDATA24C,GIRT, and FLDATA24D,SZLT commands for
other turbulence model constants.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Buoyancy Terms
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Parameters
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Wall Parameters
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Buoyancy Terms
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Parameters
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Wall Parameters
FLDATA24A, RNGT, Label, Value
Sets constants for the Re-Normalized Group Turbulence Model (RNG).
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
RNGT
Enter the word RNGT in this field.
FLDATA24A,RNGT is the FLDATA command with its first argument set to RNGT. It can be entered into
the program as either FLDATA24A, RNGT, Label,Value or FLDATA,RNGT,Label,Value where Label
and Value are as described below.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
650
FLDATA24A
Label
Turbulence model parameter label (dictates the meaning of Value):
SCTK
Value is the Schmidt number for the turbulent kinetic energy. The diffusion term in the turbulent
kinetic energy equation is divided by this factor. Value defaults to 0.72.
SCTD
Value is the Schmidt number for the kinetic energy dissipation rate. The diffusion term in the dis-
sipation rate equation is divided by this factor. Value defaults to 0.72.
CMU
Value is the turbulence model constant
t
C
k
2
which is used in the update of the turbulent viscosity. Value defaults to 0.085.
C1
Value is the multiplier of the shear rate generation term of the turbulent kinetic energy dissipation
rate equation. Value defaults to 1.42.
C2
Value is the multiplier of the dissipation source term in the turbulent kinetic energy dissipation
rate equation. Value defaults to 1.68.
BETA
Value is the RNG model constant,
Value defaults to 0.012, which corresponds to a value of 0.4 for the von Karman constant.
ETAI
Value is the asymptotic value of the strain rate parameter eta. Value defaults to 4.38.
Value
Turbulence model parameters values as explained above.
Notes
The Re-Normalized Group Turbulence Model (RNG) is an extension of the Standard k- Model. Different values
are applied to five of the Standard k- Model constants and two additional constants are added. The values
for the RNG Model are applied with the FLDATA24A,RNGT command and are separate from the Standard
k- Model constants.
Applicable only if the Re-Normalized Group Turbulence Model (RNG) is activated [FLDATA24,TURB,MODL,3].
The default values are used most often. See RNG Turbulence Model in the Theory Reference for the Mechan-
ical APDL and Mechanical Applications for more information on the turbulence models.
See also the FLDATA24,TURB command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
651
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA24A
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
FLDATA24B, NKET, Label, Value
Sets constants for the k- Turbulence Model due to Shih (NKE).
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
NKET
Enter the word NKET in this field.
FLDATA24B,NKET is the FLDATA command with its first argument set to NKET. It can be entered into
the program as either FLDATA24B,NKET, Label,Value or FLDATA,NKET,Label,Value where Label
and Value are as described below.
Label
Turbulence model parameter label (dictates the meaning of Value):
SCTK
Value is the Schmidt number for the turbulent kinetic energy. The diffusion term in the turbulent
kinetic energy equation is divided by this factor. It is the
k
constant. Value defaults to 1.0.
SCTD
Value is the Schmidt number for the kinetic energy dissipation rate. The diffusion term in the dis-
sipation rate equation is divided by this factor. It is the
. It is the A
szl1
constant. Value defaults
to 0.666.
SZL2
Value is the denominator constant used in the calculation of C
. It is the A
szl2
constant. Value
defaults to 1.25.
SZL3
Value is the strain rate multiplier. It is the A
szl3
constant. Value defaults to 0.90.
Value
Turbulence model parameters values as explained above.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
654
FLDATA24D
Notes
Sets constants for the Shih, Zhu, Lumley Turbulence Model (SZL), which is an extension of the Standard k-
Model. Different values are applied to two of the Standard k- Model constants and three additional constants
are added. The values for the SZL Model are applied with the FLDATA24D,SZLT command and are separate
from the Standard k- Model constants.
Applicable only if the Shih, Zhu, Lumley Model (SZL) is activated [FLDATA24,TURB,MODL,6]. The default
values are used most often. See SZL Turbulence Model in the Theory Reference for the Mechanical APDL and
Mechanical Applications for more information on the turbulence models.
See also the FLDATA24,TURB command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
FLDATA24E, SKWT, Label, Value
Sets constants for the k- turbulence model.
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SKWT
Enter the word SKWT in this field.
FLDATA24E,SKWT is the FLDATA command with its first argument set to SKWT. It can be entered into
the program as either FLDATA24E,SKWT,Label,Value or FLDATA,SKWT,Label,Value where Label
and Value are as described below.
Label
Turbulence model parameter label (dictates the meaning of Value):
SCTK
Value is the Schmidt number for the turbulent kinetic energy. The diffusion term in the turbulent
kinetic energy equation is divided by this factor. Value defaults to 2.0.
SCTW
Value is the Schmidt number for the specific dissipation rate. The diffusion term in the dissipation
rate equation is divided by this factor. Value defaults to 2.0.
GAMA
Value is the GAMMA factor. Value defaults to 0.5555.
BETA
Value is the BETA factor. Value defaults to 0.075.
Value
Turbulence model parameters values as explained above.
655
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA24E
Notes
Sets constants for the k- Turbulence Model. Values for the k- model are applied with the FLDATA24E,SKWT
command.
Applicable only if the k- model is activated [FLDATA24,TURB,MODL,7]. The default values are used most
often. See Standard k-omega Model in the Theory Reference for the Mechanical APDL and Mechanical Applications
for more information on the turbulence models.
See also the FLDATA24,TURB command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
FLDATA24F, SST1, Label, Value
Sets the turbulent production clip factor for the Shear Stress Transport (SST) turbulence model.
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SST1
Enter the word SST1 in this field.
FLDATA24F,SST1 is the FLDATA command with its first argument set to SST1. It can be entered into
the program as either FLDATA24F,SST1,Label,Value or FLDATA,SST1,Label,Value where Label
and Value are as described below.
Label
Enter the word CLMT in this field.
Value
Turbulent production clip factor. Defaults to 1.0e15 (no clipping).
Notes
Sets the turbulent production clip factor for the Shear Stress Transport (SST) turbulence model. Values for
the k- regime are applied with the FLDATA24F,SST1,CLMT command.
Applicable only if the Shear Stress Transport model is activated [FLDATA24,TURB,MODL,8]. The default values
are used most often. See SST Turbulence Model in the Theory Reference for the Mechanical APDL and Mech-
anical Applications for more information on the turbulence models.
See also the FLDATA24,TURB command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
656
FLDATA24F
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
FLDATA24G, SST1, Label, Value
Sets constants in the k- regime for the Shear Stress Transport (SST) turbulence model.
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SST1
Enter the word SST1 in this field.
FLDATA24G,SST1 is the FLDATA command with its first argument set to SST1. It can be entered into
the program as either FLDATA24G,SST1,Label,Value or FLDATA,SST1,Label,Value where Label
and Value are as described below.
Label
Turbulence model parameter label (dictates the meaning of Value):
SCTK
Value is the Schmidt number for the turbulent kinetic energy. The diffusion term in the turbulent
kinetic energy equation is divided by this factor. Value defaults to 1.176.
SCTW
Value is the Schmidt number for the specific dissipation rate. The diffusion term in the dissipation
rate equation is divided by this factor. Value defaults to 2.0.
GAMA
Value is the GAMMA factor.Value defaults to 0.5532.
BETA
Value is the BETA factor. Value defaults to 0.075.
Value
Turbulence model parameter values as explained above.
Notes
Sets constants in the k- regime for the Shear Stress Transport (SST) turbulence model. Values for the k-
regime are applied with the FLDATA24G,SST1 command.
Applicable only if the Shear Stress Transport model is activated [FLDATA24,TURB,MODL,8]. The default values
are used most often. See SST Turbulence Model in the Theory Reference for the Mechanical APDL and Mech-
anical Applications for more information on the turbulence models.
See also the FLDATA24,TURB command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
657
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA24G
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
FLDATA24H, SST2, Label, Value
Sets constants in the k- regime for the Shear Stress Transport (SST) turbulence model.
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SST2
Enter the word SST2 in this field.
FLDATA24H,SST2 is the FLDATA command with its first argument set to SST2. It can be entered into
the program as either FLDATA24H,SST2,Label,Value or FLDATA,SST2,Label,Value where Label
and Value are as described below.
Label
Turbulence model parameter label (dictates the meaning of Value):
SCTK
Value is the Schmidt number for the turbulent kinetic energy. The diffusion term in the turbulent
kinetic energy equation is divided by this factor. Value defaults to 1.0.
SCTW
Value is the Schmidt number for the specific dissipation rate. The diffusion term in the dissipation
rate equation is divided by this factor. Value defaults to 1.168.
GAMA
Value is the GAMMA factor. Value defaults to 0.4403.
BETA
Value is the BETA factor. Value defaults to 0.0828.
Value
Turbulence model parameter values as explained above.
Notes
Sets constants in the k- regime for the Shear Stress Transport (SST) turbulence model. Values for the k-
regime are applied with the FLDATA24H,SST2 command.
Applicable only if the Shear Stress Transport model is activated [FLDATA24,TURB,MODL,8]. The default values
are used most often. See the Theory Reference for the Mechanical APDL and Mechanical Applications for more
information on the turbulence models.
See also the FLDATA24,TURB command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Turbulence>Turbulence Model
Main Menu>Solution>FLOTRAN Set Up>Turbulence>Turbulence Model
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
658
FLDATA24H
FLDATA25, RELX, Label, Value
Sets solution and property relaxation factors.
PREP7: FLOTRAN Stability
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
RELX
Enter the word RELX in this field.
FLDATA25,RELX is the FLDATA command with its first argument set to RELX. It can be entered into the
program as either FLDATA25,RELX,Label,Value or FLDATA,RELX,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Solution and property relaxation factor labels:
VX
Value is the U Velocity relaxation factor.
VY
Value is the V Velocity relaxation factor.
VZ
Value is the W Velocity relaxation factor.
PRES
Value is the pressure relaxation factor.
TEMP
Value is the temperature relaxation factor.
ENKE
Value is the kinetic energy relaxation factor.
ENDS
Value is the dissipation rate relaxation factor.
EVIS
Value is the effective viscosity relaxation factor.
ECON
Value is the effective conductivity relaxation factor.
DENS
Value is the density update relaxation factor.
VISC
Value is the viscosity update relaxation factor.
COND
Value is the conductivity update relaxation factor.
SPHT
Value is the specific heat relaxation factor.
Value
Value of relaxation parameter as explained above (defaults to 0.5). Defaults to 1.0 for SPHT.
659
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA25
Notes
For stability purposes the solution in FLOTRAN is under-relaxed between global iterations. When the solution
of the equations for a given degree of freedom is completed, the actual set of values used is related to the
calculated set and the previous values as follows (where RELX is the relaxation factor):
new
= (1 - RELX)
previous
+ RELX
calculated
See also the FLDATA18,METH, FLDATA19,TDMA, FLDATA20,SRCH, FLDATA21,CONV, FLDATA22,MAXI, and
FLDATA23,DELT commands for other Solver Stability controls.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Relax/Stab/Cap>DOF Relaxation
Main Menu>Preprocessor>FLOTRAN Set Up>Relax/Stab/Cap>Prop Relaxation
Main Menu>Solution>FLOTRAN Set Up>Relax/Stab/Cap>DOF Relaxation
Main Menu>Solution>FLOTRAN Set Up>Relax/Stab/Cap>Prop Relaxation
FLDATA26, STAB, Label, Value
Sets stability controls.
PREP7: FLOTRAN Stability
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
STAB
Enter the word STAB in this field.
FLDATA26,STAB is the FLDATA command with its first argument set to STAB. It can be entered into the
program as either FLDATA26,STAB,Label,Value or FLDATA,STAB,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Control label (dictates the meaning of Value):
TURB
Turbulence inertial relaxation (Value defaults to 1.0 x 10
15
).
MOME
Momentum inertial relaxation (Value defaults to 1.0 x 10
15
).
PRES
Pressure inertial relaxation (Value defaults to 1.0 x 10
15
).
TEMP
Energy inertial relaxation (Value defaults to 1.0 x 10
20
).
VISC
Artificial viscosity (Value defaults to 0.0).
Value
Inertial relaxation factor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
660
FLDATA26
Notes
These controls are used to make the sets of equations diagonally dominant through inertial relaxation.
Making the matrix equations more diagonally dominant makes them easier to solve. More global iterations
are required for convergence if inertial relaxation is used. See Inertial Relaxation in the Theory Reference for
the Mechanical APDL and Mechanical Applications for details.
Artificial viscosity can be activated to enhance stability in difficult compressible problems as well as in some
incompressible problems. The artificial viscosity is added to the main diagonal and the forcing function of
the momentum equations. The equations become more diagonally dominant locally in regions with high
velocity gradients.
In compressible analyses, the artificial viscosity should be gradually removed as convergence is achieved
since a nonzero value will affect the final solution. This is not necessary for incompressible analyses as the
divergence of velocity should be zero. In practice, values of artificial viscosity should not exceed 1000 times
the effective viscosity.
See also the FLDATA18,METH, FLDATA19,TDMA, FLDATA20,SRCH, FLDATA21,CONV, FLDATA22,MAXI,
FLDATA23,DELT, FLDATA24,TURB, and FLDATA25,RELX commands for other Solver Stability controls.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Relax/Stab/Cap>Stability Parms
Main Menu>Solution>FLOTRAN Set Up>Relax/Stab/Cap>Stability Parms
FLDATA27, PRIN, Label, Value
Controls dependent variable printing.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
PRIN
Enter the word PRIN in this field.
FLDATA27,PRIN is the FLDATA command with its first argument set to PRIN. It can be entered into the
program as either FLDATA27,PRIN,Label,Value or FLDATA,PRIN,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Dependent variable tabulation is being specified for:
VX
U velocity.
VY
V velocity.
VZ
W velocity.
661
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA27
PRES
Pressure.
TEMP
Temperature.
ENKE
Kinetic energy.
ENDS
Turbulent kinetic energy dissipation rate.
DENS
Density.
VISC
Viscosity.
COND
Conductivity.
SPHT
Specific heat.
EVIS
Effective viscosity.
ECON
Effective conductivity.
SFTS
Surface tension coefficient.
Value
Tabulation key:
T
Turn tabulation on.
F
Turn tabulation off (default).
Notes
When set to T, this control produces a tabulation of the values of the particular dependent variable at every
node. It is set at the beginning of a load step. The printout can be large and is contained in the file Job-
name.PFL.
To observe the values of a dependent variable at selected nodes during postprocessing, see the PRNSOL
command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
662
FLDATA27
FLDATA28, MODR, Label, Value
Specifies that variable results are to be replaced.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
MODR
Enter the word MODR in this field.
FLDATA28,MODR is the FLDATA command with its first argument set to MODR. It can be entered into
the program as either FLDATA28,MODR,Label,Value or FLDATA,MODR,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Dependent variables to be replaced:
VX
U velocity.
VY
V velocity.
VZ
W velocity.
PRES
Pressure.
TEMP
Temperature.
ENKE
Kinetic energy.
ENDS
Turbulent kinetic energy dissipation rate.
DENS
Density.
VISC
Viscosity.
COND
Conductivity.
SPHT
Specific heat.
EVIS
Effective viscosity.
ECON
Effective conductivity.
TTOT
Total temperature.
SFTS
Surface tension coefficient.
663
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA28
Value
Variable replace flag:
T
Allow this variable to be re-initialized.
F
Do not allow this variable to be re-initialized (default).
Notes
Results for any of the dependent variables available can be replaced with new values for the next load step.
Only a single value may be specified for the entire solution domain. This is used for re-initialization of a
variable such as temperature, which has diverged while other temperature independent calculations have
been successful.
To re-initialize a variable, first use this command to set the replacement flag to T for the desired variable.
Then use the FLDATA29,MODV command to put in the new variable value. Note that the flag(s) will be
automatically reset to F after the new value has been inserted.
See also the FLDATA29,MODV command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Mod Res/Quad Ord>Modify Results
Main Menu>Solution>FLOTRAN Set Up>Mod Res/Quad Ord>Modify Results
FLDATA29, MODV, Label, Value
Re-initializes a results variable.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
MODV
Enter the word MODV in this field.
FLDATA29,MODV is the FLDATA command with its first argument set to MODV. It can be entered into
the program as either FLDATA29,MODV,Label,Value or FLDATA,MODV,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Dependent variables to be replaced:
VX
U velocity.
VY
V velocity.
VZ
W velocity.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
664
FLDATA29
SP0n
Mass fraction of species n, where n = 1 to 6 (FLOTRAN). If a species is given a user-defined name
[MSSPEC], use that name instead of SP0n.
PRES
Pressure.
TEMP
Temperature.
ENKE
Kinetic energy.
ENDS
Turbulent kinetic energy dissipation rate.
DENS
Density.
VISC
Viscosity.
COND
Conductivity.
SPHT
Specific heat.
EVIS
Effective viscosity.
ECON
Effective conductivity.
TTOT
Total temperature.
SFTS
Surface tension coefficient.
LMDn
Laminar mass diffusion coefficient for species n, where n = 1 to 6.
EMDm
Effective mass diffusion coefficient for species n, where n = 1 to 6.
Value
New value of variable (defaults to 0.0). Specifying FLDATA29,MODV,Label,Value modifies the entire
field variable to a constant value.
Notes
Results for any of the dependent variables available can be replaced with new values for the next load step.
Only a single value may be specified for the entire solution domain. This is used for re-initialization of a
variable such as temperature, which has diverged while other temperature independent calculations have
been successful. Boundary conditions are reapplied upon restart.
Quantities which are not allowed to vary (e.g., CONSTANT properties) should not be modified. Also, modifying
results should not be attempted on the initial run.
665
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA29
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Mod Res/Quad Ord>Modify Results
Main Menu>Solution>FLOTRAN Set Up>Mod Res/Quad Ord>Modify Results
FLDATA30, QUAD, Label, Value
Controls the quadrature orders.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
QUAD
Enter the word QUAD in this field.
FLDATA30,QUAD is the FLDATA command with its first argument set to QUAD. It can be entered into
the program as either FLDATA30,QUAD,Label,Value or FLDATA,QUAD,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Element integral to change quadrature for:
MOMD
Momentum diffusion term (defaults to 0 integration points).
MOMS
Momentum source term (defaults to 0 integration points).
PRSD
Pressure diffusion term (defaults to 1 integration point).
PRSS
Pressure source term (defaults to 1 integration point).
THRD
Thermal diffusion term (defaults to 0 integration points).
THRS
Thermal source term (defaults to 0 integration points).
TRBD
Turbulent diffusion terms (defaults to 0 integration points).
TRBS
Turbulent source terms (defaults to 2 integration points).
Value
Number of integration points.
Notes
Controls the number of integration points used in the evaluation of element integrals. They are set at the
optimum values by default. Values of 0 and 1 correspond to 1 point quadrature, but 0 means an average
value of the diffusion coefficient has been used in the integrals. Values are automatically set to 2 for
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
666
FLDATA30
axisymmetric and polar analyses. Using a value of 2 will improve the accuracy for analyses using distorted
elements. Repeat the FLDATA30 command as required.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Mod Res/Quad Ord>CFD Quad Orders
Main Menu>Solution>FLOTRAN Set Up>Mod Res/Quad Ord>CFD Quad Orders
FLDATA31, CAPP, Label, Value
Specifies dependent variable caps.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
CAPP
Enter the word CAPP in this field.
FLDATA31,CAPP is the FLDATA command with its first argument set to CAPP. It can be entered into
the program as either FLDATA31,CAPP,Label,Value or FLDATA,CAPP,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Capping parameters. The first three are used to set the flags, and the rest are used to specify the cap
values after the flag is set to T:
VELO
Value is the velocity capping flag (Value defaults to F).
TEMP
Value is the temperature capping flag (Value defaults to F).
PRES
Value is the pressure capping flag (Value defaults to F).
UMIN
Value is the minimum value of U velocity allowed (Value defaults to -1.E20).
UMAX
Value is the maximum value of U velocity allowed (Value defaults to +1.E20).
VMIN
Value is the minimum value of V velocity allowed (Value defaults to -1.E20).
VMAX
Value is the maximum value of V velocity allowed (Value defaults to +1.E20).
WMIN
Value is the minimum value of W velocity allowed (Value defaults to -1.E20).
WMAX
Value is the maximum value of W velocity allowed (Value defaults to +1.E20).
TMIN
Value is the minimum value of temperature allowed (Value defaults to 0.0).
667
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA31
TMAX
Value is the maximum value of temperature allowed (Value defaults to +1.E20).
PMIN
Value is the minimum value of pressure allowed (Value defaults to -1.E20).
PMAX
Value is the maximum value of pressure allowed (Value defaults to +1.E20).
Value
Capping flag (T or F) or capping parameter value.
Notes
These parameters are used to limit arbitrarily the values of the dependent variables. Capping helps prevent
divergence in the early stages of analyses. Be careful when using caps to ensure that they have no impact
on the final answers. You should remove capping as convergence is approached.
To use capping, you must first set the flag to T and then set the maximum and minimum caps.
The pressure value calculated by the solution of the pressure equation is capped, not the relaxed value.
Therefore, if you introduce pressure capping upon restarting an analysis, pressure values may still be outside
the caps.
Capping applies to relative values of pressure and absolute values of temperature.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Relax/Stab/Cap>Results Capping
Main Menu>Solution>FLOTRAN Set Up>Relax/Stab/Cap>Results Capping
FLDATA32, REST, Label, Value, Value2, Fname, Ext, --
Controls restart options.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
REST
Enter the word REST in this field.
FLDATA32,REST is the FLDATA command with its first argument set to REST. It can be entered into the
program as either FLDATA32,REST,Label,Value,Value2 or FLDATA,REST,Label,Value,Value2
where Label, Value, and Value2 are as described below. See the FLDATA command for other FLOTRAN
CFD input choices.
Label
Restart option (dictates the meaning of Value and Value2):
NSET
Identifies the restart point by results set number. Value is the results set number in the results file
(Jobname.RFL) from which the analysis will be restarted. Value = 0 or blank indicates that the
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
668
FLDATA32
restart will be from the last results set. Value will be reset to zero if the label ITER, LSTP, or TIME is
subsequently set. Value2 is not used.
ITER
Identifies the restart point by cumulative iteration number. Value is the cumulative iteration number
in the results file (Jobname.RFL) from which the analysis will be restarted. If there are no results
stored for this cumulative iteration, the results set with the next lowest cumulative iteration number
will be used. Value = 0 or blank indicates that the restart will be from the last results set. Value
will be reset to zero if the label NSET, LSTP, or TIME is subsequently set. Value2 is not used.
LSTP
Identifies the restart point by load step and substep numbers. Value and Value2 are the load step
number and substep number in the results file (Jobname.RFL) from which the analysis will be re-
started. Value = 0 or blank indicates that the restart will be from the last results set. Value and
Value2 will be reset to zero if the label ITER, NSET, or TIME is subsequently set.
TIME
Identifies the restart point by transient time. Value is the time point in the results file (Job-
name.RFL) from which the analysis will be restarted. If there are no results stored for this time
point, the results set with the next lowest time point will be used. Value = 0 or blank indicates that
the restart will be from the last results set. Value will be reset to zero if the label NSET, LSTP, or
ITER is subsequently set. Value2 is not used.
RFIL
Specifies whether the CFD data structure restart file (Jobname.CFD) is to be read for the restart.
Useful for large models where the creation of the data structures may take a long time. Value may
be T (true) or F (false) and defaults to F. If the Jobname.CFD file does not exist, it will be created
(if RFIL is set to true). Setting RFIL to true will toggle WFIL to false, and setting WFIL to true will
toggle RFIL to false.
WFIL
Specifies whether the CFD data structure restart file (Jobname.CFD) is to be written. Useful for
overwriting an existing restart file when changes in the model or boundary conditions have occurred.
Value may be T (true) or F (false) and defaults to F. Setting WFIL to true will toggle RFIL to false,
and setting RFIL to true will toggle WFIL to false.
OVER
Specifies whether to overwrite the set of results from which the restart occurs. Value may be -1, 0,
or 1 (defaults to 0). If Value is -1, the previous set of results are overwritten. If Value is 1, the
previous set of results is saved. If Value is 0 (default), the previous results are saved only if the
results were written as a saved (converged) set of results. When this flag is used to change the status
of the previous set of results, ANSYS sets it to 0 so that future sets of results are not affected.
CLEAR
Specifies whether to eliminate from the results file (Jobname.RFL) all results sets stored before
and after the set used for the restart. Value may be T (true) or F (false) and defaults to F. The restart
set is the last set or the set specified with another FLDATA32,REST command. Use a positive value
of NSET, ITER, LSTP, or TIME to create a backup of the results file and use a negative value of NSET,
ITER, LSTP or TIME if you do not desire a backup of the results file (see Notes below).
Value, Value2
Restart point or restart file flag (T or F) as described above.
669
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA32
Fname
File name and directory path of a results file to be used for the restart (248 characters maximum, including
directory). If you do not specify a directory path, it will default to your working directory and you can
use all 248 characters for the file name.
The file name defaults to Jobname. This field is valid only with Label = NSET, ITER, LSTP, or TIME.
Ext
Filename extension (8 character maximum).
The extension defaults to RFL, and is valid only with Label = NSET, ITER, LSTP, or TIME.
--
Unused field.
Notes
If both RFIL and WFIL are set to true through the GUI, the state of WFIL will prevail and a new Jobname.CFD
file will be written.
If the value of NSET, ITER, LSTP, or TIME is positive, the original results file (Jobname.RFL) is moved to
Jobname.RFO, and a new Jobname.RFL is created containing all the results sets stored prior to the restart
point as well as new results. If the value is negative, the Jobname.RFL file will contain the prior results
and the new results but the old file will be destroyed. This latter option is used if the results sets currently
stored beyond the desired restart point are not worth saving.
If restarting from an existing file other than Jobname.RFL, no backup file (Jobname.RFO) is created because
the existing file is not affected.
If a results file name (Fname) is entered for a restart, FLOTRAN interpolates those results onto the current
mesh in the database, regardless of whether or not the mesh has changed. This causes the convergence
monitors to start again from zero. This restart is different than a restart without a file name specification.
However, the results quickly converge to the original solution.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>CFD Restart File
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>Restart/Clear
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>Restart/Iteration
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>Restart/Load step
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>Restart/Set
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>Restart/Time
Main Menu>Preprocessor>FLOTRAN Set Up>Restart Options>Tag set status
Main Menu>Solution>FLOTRAN Set Up>Restart Options>CFD Restart File
Main Menu>Solution>FLOTRAN Set Up>Restart Options>Restart/Clear
Main Menu>Solution>FLOTRAN Set Up>Restart Options>Restart/Iteration
Main Menu>Solution>FLOTRAN Set Up>Restart Options>Restart/Load step
Main Menu>Solution>FLOTRAN Set Up>Restart Options>Restart/Set
Main Menu>Solution>FLOTRAN Set Up>Restart Options>Restart/Time
Main Menu>Solution>FLOTRAN Set Up>Restart Options>Tag set status
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
670
FLDATA32
FLDATA33, ADVM, Label, Value
Specifies the approach to discretize the advection term.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
ADVM
Enter the word ADVM in this field.
FLDATA33,ADVM is the FLDATA command with its first argument set to ADVM. It can be entered into
the program as either FLDATA33,ADVM,Label,Value or FLDATA,ADVM,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Specifies the transport equation.
MOME
Momentum equations.
TURB
Turbulence equations.
PRES
Compressible pressure equation.
TEMP
Energy equation.
Value
Choice of approach to discretize the advection term:
MSU
Monotone Streamline Upwind approach (default for PRES).
SUPG
Streamline Upwind / Petrov-Galerkin approach (default for MOME, TURB, and TEMP).
COLG
Collocated Galerkin (COLG) approach.
Notes
See Using SUPG in the Fluids Analysis Guide for more information on the SUPG approach.
See Derivation of Fluid Flow Matrices in the Theory Reference for the Mechanical APDL and Mechanical Applic-
ations for more information on the advection term.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Advection
Main Menu>Solution>FLOTRAN Set Up>Advection
671
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA33
FLDATA34, MIR, Label, Value
Sets modified inertial relaxation factors.
PREP7: FLOTRAN Stability
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
MIR
Enter the word MIR in this field.
FLDATA34,MIR is the FLDATA command with its first argument set to MIR. It can be entered into the
program as either FLDATA34,MIR,Label,Value or FLDATA,MIR,Label,Value where Label and
Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Modified relaxation factor labels:
MOME
Momentum modified inertial relaxation.
TURB
Turbulence modified inertial relaxation.
TEMP
Energy modified inertial relaxation.
Value
Modified inertial relaxation factor. Value defaults to 0 (modified inertial relaxation off ).
Notes
Value must be a positive real number. A Value between 0.1 and 1.0 is recommended. A larger Value
provides a more robust scheme, but it may yield a slower convergence.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Relax/Stab/Cap>MIR Stabilization
Main Menu>Solution>FLOTRAN Set Up>Relax/Stab/Cap>MIR Stabilization
FLDATA35, VFTOL, Label, Value
Specifies tolerances for the lower and upper bound of the volume fraction.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
VFTOL
Enter the word VFTOL in this field.
FLDATA35,VFTOL is the FLDATA command with its first argument set to VFTOL. It can be entered into
the program as either FLDATA35,VFTOL,Label,Value or FLDATA,VFTOL,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
672
FLDATA34
Label
Tolerance choices:
VOFL
Lower bound tolerance in the VOF advection algorithm (Value defaults to 1.0e-6).
VOFU
Upper bound tolerance in the VOF advection algorithm (Value defaults to 1.0e-6).
LAML
Lower bound tolerance in the solver for laminar flows (Value defaults to 1.0e-2).
LAMU
Upper bound tolerance in the solver for laminar flows (Value defaults to 1.0e-2).
TRBL
Lower bound tolerance in the solver for turbulent flows (Value defaults to 1.0e-1).
TRBU
Upper bound tolerance in the solver for turbulent flows (Value defaults to 1.0e-1).
Value
Tolerance value for Label above.
Notes
Volume fractions less than the lower bound tolerance are treated as 0. Volume fractions greater than 1 minus
the upper bound tolerance are treated as 1.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>VOF Environment>VFRC Tolerance
Main Menu>Solution>FLOTRAN Set Up>VOF Environment>VFRC Tolerance
FLDATA36, AMBV, Label, Value
Specifies ambient reference values outside of the fluid for the volume of fluid (VOF) method.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
AMBV
Enter the word AMBV in this field.
FLDATA36,AMBV is the FLDATA command with its first argument set to AMBV. It can be entered into
the program as either FLDATA36,AMBV,Label,Value or FLDATA,AMBV,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Ambient Reference for which value is being specified:
VX
U velocity.
673
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA36
VY
V velocity.
VZ
W velocity.
TEMP
Temperature.
ENKE
Turbulent Kinetic Energy.
ENDS
Turbulent Dissipation Rate.
Value
Value for the ambient reference as described above.
Notes
PRES is used as a boundary condition at the free surface and for plotting purposes. VX, VY, VZ, TEMP, ENKE
and ENDS are only used for plotting purposes.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>VOF Environment>Ambient Condit'n
Main Menu>Solution>FLOTRAN Set Up>VOF Environment>Ambient Condit'n
FLDATA37, ALGR, Label, Value
Specifies segregated solution or film coefficient algorithms.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
ALGR
Enter the word ALGR in this field.
FLDATA37,ALGR is the FLDATA command with its first argument set to ALGR. It can be entered into
the program as either FLDATA37,ALGR,Label,Value or FLDATA,ALGR,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Algorithm label:
SEGR
Segregated
HFLM
Film Coefficient
Value
If Label = SEGR, Value specifies the segregated solution algorithm:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
674
FLDATA37
SIMPLEF
Original segregated algorithm (default)
SIMPLEN
Enhanced segregated algorithm
If Label = HFLM, Value specifies the film coefficient algorithm:
MATX
Conductivity matrix algorithm (default)
TEMP
Temperature field algorithm
Notes
Settings automatically changed when SIMPLEN is chosen are not automatically reset if SIMPLEF is reselected.
See Coupling Algorithms in the Fluids Analysis Guide for the settings.
The conductivity matrix algorithm uses the thermal conductivity matrix to calculate heat fluxes and film
coefficients. The temperature field algorithm calculates film coefficients directly from thermal gradients.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Algorithm Control
Main Menu>Solution>FLOTRAN Set Up>Algorithm Control
FLDATA38, MASS, Label, Value
Specifies the mass type for a fluid transient analysis.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
MASS
Enter the word MASS in this field.
FLDATA38,MASS is the FLDATA command with its first argument set to MASS. It can be entered into
the program as either FLDATA38,MASS,Label,Value or FLDATA,MASS,Label,Value where Label
and Value are as described below. See the FLDATA command for other FLOTRAN CFD input choices.
Label
Degree of freedom set for which mass type is being specified:
MOME
Momentum equation.
PRES
Pressure equation.
TURB
Turbulent equation.
675
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA38
TEMP
Energy equation.
ALL
Momentum, pressure, turbulent, and energy equations.
Value
Mass type for fluid transient analysis:
LUMP
Lumped mass matrix (default).
CONS
Consistent mass matrix.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Transient Ctrl>Mass Type
Main Menu>Solution>FLOTRAN Set Up>Transient Ctrl>Mass Type
FLDATA39, REMESH, Label, Value
Specifies remeshing parameters for transient fluid flow and fluid-solid interaction analyses.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
REMESH
Enter the word REMESH in this field.
FLDATA39,REMESH is the FLDATA command with its first argument set to REMESH. It can be entered
into the program as either FLDATA39,REMESH,Label,Value or FLDATA,REMESH,Label,Value where
Label and Value are as described below. See the FLDATA command for other FLOTRAN CFD input
choices.
Label
Remeshing fluid element label (dictates the meaning of Value):
ELEM
Value specifies the fluid elements included in remeshing. The following are valid values:
NON
Do not remesh (default).
ALL
Remesh all defined fluid elements if the quality of the worst defined element falls below any
quality requirement. Required element qualities are specified by Label = ARMAX, VOCH, or
ARCH as explained below.
PAR
Remesh defined fluid elements that have a quality below any quality requirement.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
676
FLDATA39
A component name may be specified for Value. All elements grouped into a component name
are remeshed if the quality of the worst element falls below any quality requirement. The com-
ponent name length is up to 8 characters.
XBNE
Value specifies the elements connected to the boundary nodes excluded from remeshing. The
following are valid values:
NON
Do not exclude any elements connected to the boundary nodes (default).
ALL
Exclude all elements connected to the boundary nodes.
FSI
Exclude elements connected to fluid-solid interaction interfaces.
RESIZ
Value is the element size used for remeshing. Value defaults to 0 (the element size at the nearest
boundary is used for remeshing).
REXPN
Value is the area expansion (or contraction) option for remeshing. (This option is the same as
SMRTSIZE,,,EXPND.) This option is used to size internal elements in an area based on the size of the
elements on the area's boundaries. Value is the expansion (or contraction) factor. For example, is-
suing FLDATA39,REMESH,EXPND,2 before meshing an area will allow a mesh with elements that are
approximately twice as large in the interior of an area as they are on the boundary. If Value is less
than 1, a mesh with smaller elements on the interior of the area will be allowed. Value for this
option should be greater than 0.5 but less than 4. Value defaults to 1, which does not allow expan-
sion or contraction of internal element sizes. If Value = 0, the default value of 1 will be used. The
actual size of the internal elements will also depend on RESIZ sizing, if used.
ARMA
Value is the maximum allowable element generalized aspect ratio. Value defaults to 10.
VOCH
Value is the maximum allowable change of element size (area or volume). Value defaults to 3.
ARCH
Value is the maximum allowable element aspect ratio change.Value defaults to 3.
STEP
Value is the element quality checking frequency based on time steps. Every Value time steps, a
quality check takes place. Value defaults to 1 (a quality check at every step).
TIME
Value is the only element quality checking time. A quality check takes place at a time specified by
Value. Value defaults to -1 (a quality check at every time point).
Value
Value as described for Label above.
Notes
See Table 7.2: "Element Qualities" in the Fluids Analysis Guide for definitions of element qualities.
Repeat command to set each Label as required.
677
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLDATA39
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Remesh Ctrl>Element size ctrl
Main Menu>Preprocessor>FLOTRAN Set Up>Remesh Ctrl>Elements for remesh
Main Menu>Preprocessor>FLOTRAN Set Up>Remesh Ctrl>Mesh qualities
Main Menu>Preprocessor>FLOTRAN Set Up>Remesh Ctrl>Remesh frequency
Main Menu>Solution>FLOTRAN Set Up>Remesh Ctrl>Element size ctrl
Main Menu>Solution>FLOTRAN Set Up>Remesh Ctrl>Elements for remesh
Main Menu>Solution>FLOTRAN Set Up>Remesh Ctrl>Mesh qualities
Main Menu>Solution>FLOTRAN Set Up>Remesh Ctrl>Remesh frequency
FLDATA40, WADV, Label, Value
Controls activation of thermal stabilization near walls.
PREP7: FLOTRAN Turbulence
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
WADV
Enter the word WADV in this field.
FLDATA40,WADV is the FLDATA command with its first argument set to WADV. It can be entered into
the program as either FLDATA40,WADV,Label,Value or FLDATA,WADV,Label,Value where Label
and Value are as described below.
Label
Enter the word TEMP in this field.
Value
Value controlling Label:
TRUE or T
Turn this feature on.
FALSE or F
Turn this feature off.
Notes
Thermal oscillations may occur for turbulent heat transfers in near-wall regions when using the SUPG or the
COLG advection scheme with a coarse mesh. Use this command to minimize such spatial oscillations.
For the nodes that lie at the intersection of the wall or solid surface and the inlet, the thermal boundary
conditions for those nodes must be the same as the wall or solid surface, not the inlet.
See the FLDATA33,ADVM command to see various advection discretization schemes.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
678
FLDATA40
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Relax/Stab/Cap>Thermal Stabilization
Main Menu>Solution>FLOTRAN Set Up>Relax/Stab/Cap>Thermal Stabilization
FLIST, NODE1, NODE2, NINC
Lists force loads on the nodes.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
NODE1, NODE2, NINC
List forces for nodes NODE1 to NODE2 (defaults to NODE1) in steps of NINC (defaults to 1). If ALL, list
for all selected nodes [NSEL] and NODE2 and NINC are ignored (default). If NODE1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for NODE1.
Notes
Listing applies to the selected nodes [NSEL] and the selected force labels [DOFSEL].
Caution
A list containing a node number that is larger than the maximum defined node (NODE2), could
deplete the system memory and produce unpredictable results.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Forces>On All Nodes
Utility Menu>List>Loads>Forces>On Picked Nodes
FLLIST, NLOC1, NLOC2, NINC
Lists the fatigue location parameters.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
NLOC1, NLOC2, NINC
List location parameters from NLOC1 (defaults to 1) to NLOC2 (defaults to NLOC1) in steps of NINC
(defaults to 1). If NLOC1 = ALL, NLOC2 and NINC are ignored and all locations are listed.
Menu Paths
Main Menu>General Postproc>Fatigue>List Stress Loc
679
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLLIST
FLOCHECK, Key
Sets up and runs a zero-iteration FLOTRAN analysis.
SOLUTION: FLOTRAN Checkout
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
Key
Determines whether the FLOTRAN analysis is initialized and whether boundary condition status now
becomes old.
0
(Default) Initialization is not performed.
1
Initialization is performed
2
No initialization performed, boundary condition status now becomes old.
Notes
The FLOCHECK command will provide a results summary for the Zeroth iteration (KEY = 1) or the current
iteration (KEY = 0). The results summary provides max/min/average values for each property and DOF. Also,
mass flow boundaries are identified and all thermal energy transfer information is summarized. All input
information is summarized in the Jobname.PFL file (i.e., print file). If you are running the ANSYS program
from the GUI, the FLOTRAN print file is echoed to the output window. The FLOCHECK command aids veri-
fication of boundary condition and property specification.
Initialization [FLOCHECK,1] deletes any existing Jobname.PFL and Jobname.RFL files. Run FLOCHECK
initialization only when you are sure you no longer need the existing results files.
The FLOCHECK,2 command changes boundary conditions to the old state. It has no effect on the Job-
name.RFL file. For information on changing fluid boundary conditions, see Applying Transient Boundary
Conditions in Fluids Analysis Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Flocheck
Main Menu>Solution>FLOTRAN Set Up>Flocheck
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
680
FLOCHECK
FLOTRAN
Specifies "FLOTRAN data settings" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL <> <> DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>General Postproc>FLOTRAN Module
FLREAD, Fname, Ext, --
Reads the residual file written by the FLOTRAN CFD option.
POST1: FLOTRAN Processing
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to RDF (residual file) if Fname is blank.
--
Unused field.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Read Results>FLOTRAN 2.1A
681
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FLREAD
FLST, NFIELD, NARG, TYPE, Otype, LENG
Specifies data required for a picking operation (GUI).
DATABASE: Picking
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NFIELD
Field number on the command which uses the picking data. (Count the command name as a field, so
that a 2 indicates the first command argument, 3 for the second command argument, etc.) The corres-
ponding field on the command will have a P51X label.
NARG
Number of items in the picked list.
TYPE
Type of items picked:
1
Node numbers
2
Element numbers
3
Keypoint numbers
4
Line numbers
5
Area numbers
6
Volume numbers
7
Trace points
8
Coordinate locations (in Global Cartesian coordinates)
9
Screen picks (in X, Y screen coordinates (-1 to 1))
Otype
Data order:
NOOR
Data is not ordered (default).
ORDER
Data is in an ordered list (such as for the E,P51X and A,P51X commands, in which the order of the
data items is significant for the picking operation).
LENG
Length of number of items describing the list (should equal NARG if Otype = NOOR; default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
682
FLST
Notes
Specifies data required for the FITEM command during a picking operation. This is a command generated
by the GUI and will appear in the log file (Jobname.LOG) if graphical picking is used. This command is not
intended to be typed in directly in an ANSYS session (although it can be included in an input file for batch
input or for use with the /INPUT command).
On the log file, FLST will always be followed by one or more FITEM commands which in turn are followed
by the ANSYS command that contains a P51X label in one of its fields. This set of commands should not be
edited.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
FLUXV
Calculates the flux passing through a closed contour.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
FLUXV invokes an ANSYS macro which calculates the flux passing through a closed contour (path) predefined
by PATH. The calculated flux is stored in the parameter FLUX. In a 2-D analysis, at least two nodes must be
defined on the path. In 3-D, a path of nodes describing a closed contour must be specified (i.e., the first and
last node in the path specification must be the same). A counterclockwise ordering of nodes on the PPATH
command will give the correct sign on flux. Path operations are used for the calculations, and all path items
are cleared upon completion. This macro is only available for vector potential formulations.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>Path Flux
FMAGBC, Cnam1, Cnam2, Cnam3, Cnam4, Cnam5, Cnam6, Cnam7, Cnam8, Cnam9
Applies force and torque boundary conditions to an element component.
SOLUTION: Misc Loads
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Cnam1, Cnam2, Cnam3, . . . , Cnam9
Names of existing element components (CM command). Must be enclosed in single quotes (e.g., `Cnam1')
when the command is manually typed in.
683
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FMAGBC
Notes
FMAGBC invokes a predefined ANSYS macro to apply Maxwell and virtual work force and torque boundary
conditions to an element component. These boundary conditions are used for subsequent force and torque
calculations during solution. Magnetic virtual displacements (MVDI = 1) are applied to nodes of elements in
the components, and Maxwell surface flags (MXWF) are applied to air elements adjoining the element
components. Incorrect force and torque calculations will occur for components sharing adjacent air elements.
Companion macros FMAGSUM and TORQSUM can be used in POST1 to summarize the force and torque
calculations, respectively. Torque calculations are valid for 2-D planar analysis only. For 2-D harmonic analysis,
force and torque represent time-average values.
If using elements SOLID117, PLANE121, SOLID122, SOLID123, SOLID236 and SOLID237 (static analyses only),
use EMFT to summarize electromagnetic force and torque. If you do use FMAGSUM, you do not need to
first set either the Maxwell or the virtual work force flags via FMAGBC.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>Comp. Force
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Flag>Comp. Force/Torque
Main Menu>Solution>Define Loads>Apply>Electric>Flag>Comp. Force
Main Menu>Solution>Define Loads>Apply>Magnetic>Flag>Comp. Force/Torque
FMAGSUM, Cnam1, Cnam2, Cnam3, Cnam4, Cnam5, Cnam6, Cnam7, Cnam8, Cnam9
Summarizes electromagnetic force calculations on element components.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Cnam1, Cnam2, Cnam3, . . . , Cnam9
Names of existing element components for which Maxwell or virtual work boundary conditions were
applied in the preprocessor. Must be enclosed in single quotes (e.g., `Cnam1') when the command is
manually typed in.
Notes
FMAGSUM invokes an ANSYS macro that summarizes the Maxwell and virtual work forces. The element
components must have had appropriate Maxwell or virtual work boundary conditions established in the
preprocessor prior to solution in order to retrieve forces (see FMAGBC, SF, BF commands). The forces are
also stored on a per-element basis for the adjacent air layer surrounding the components in the element
table [ETABLE]. Maxwell forces are stored as items FMX_X, FMX_Y, and FMX_Z, and virtual work forces are
stored as items FVW_X, FVW_Y, and FVW_Z. Use the PLETAB and PRETAB commands to plot and list the
element table items.
If using elements SOLID117, PLANE121, SOLID122, SOLID123, SOLID236, and SOLID237, use EMFT to sum-
marize electromagnetic force and torque.
FMAGSUM can also be used to summarize time-average forces from a 2-D harmonic analysis.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
684
FMAGSUM
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Component Based>Force
/FOCUS, WN, XF, YF, ZF, KTRANS
Specifies the focus point (center of the window).
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
XF, YF, ZF
Location of the object to be at the focus point (center of the window) in the global Cartesian coordinate
system. If XF = AUTO, allow automatic location calculation. If XF = USER, use focus location of last display
(useful when last display had auto focus).
KTRANS
Translate key:
0
Interpret numerical XF, YF, ZF values as described above.
1
Interpret XF, YF, ZF values as multiples of half-screens to translate from the current position in the
screen coordinate system. Example: XF of 2.4 translates the display approximately 2.4 half-screens
to the left in the screen X (horizontal) direction.
2
Interpret XF, YF, ZF values as multiples of half-screens to translate from the current position in the
global Cartesian coordinate system. Example: XF of 1.5 translates the display approximately 1.5 half-
screens in the global Cartesian X direction of the model.
Command Default
Focus location is automatically calculated to be at the geometric center of the object (modified for centering
within the window, depending upon the view).
Notes
Specifies the location on (or off ) the model which is to be located at the focus point (center of the window).
For section and capped displays, the cutting plane is also assumed to pass through this location (unless the
working plane is used via /CPLANE). See also /AUTO and /USER commands.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Pan, Zoom, Rotate
Utility Menu>PlotCtrls>View Settings>Focus Point
685
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/FOCUS
FOR2D
Calculates magnetic forces on a body.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
FOR2D invokes an ANSYS macro which calculates magnetic forces on a body that is completely surrounded
by air (symmetry permitted). The calculated forces are stored in the parameters FX and FY. In interactive
mode, a node plot is produced with the integration path highlighted. A predefined closed path [PATH],
passing through the air elements surrounding the body, must be available for this calculation. A counter-
clockwise ordering of nodes on the PPATH command will give the correct sign on the forces. Forces are
calculated using a Maxwell stress tensor approach. The macro is valid for 2-D planar or axisymmetric analysis.
Path operations are used for the calculations, and all path items are cleared upon completion.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>Mag Forces
FORCE, Lab
Selects the element nodal force type for output.
POST1: Controls
POST26: Controls
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Lab
Type of force to be associated with the force items:
TOTAL
Total forces (static, damping, and inertia).
STATIC
Static forces.
DAMP
Damping forces.
INERT
Inertia forces.
Command Default
Use the total forces.
Notes
FORCE selects the element nodal force type for output with the POST1 PRESOL, PLESOL, PRRFOR, NFORCE,
FSUM, etc. commands, the POST26 ESOL command, and reaction force plotting [/PBC]. For example,
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
686
FOR2D
FORCE,STATIC causes item F of the PRESOL command to be the static forces for the elements processed.
Element member forces (such as those available for beams and shells, which are processed by Item and
Sequence number) are not affected by this command. Damping and inertia forces are only available for full
transient and harmonic analyses.
The PRRSOL command is not valid with FORCE. Use the PRRFOR command, which provides the same
functionality as PRRSOL, instead.
In POST26, the ESOL data stored is based on the active FORCE specification at the time the data is stored.
To store data at various specifications (for example, static and inertia forces), issue a STORE command before
each new specification.
The FORCE command cannot be used to extract static, damping, and inertial forces for MPC184 joint elements.
Menu Paths
Main Menu>General Postproc>Options for Outp
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
Utility Menu>List>Results>Options
FORM, Lab
Specifies the format of the file dump.
AUX2: Binary Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Format:
RECO
Basic record description only (minimum output) (default).
TEN
Same as RECO plus the first ten words of each record.
LONG
Same as RECO plus all words of each record.
Command Default
Basic record description (RECO).
Notes
Specifies the format of the file dump (from the DUMP command).
Menu Paths
Utility Menu>File>List>Binary Files
Utility Menu>List>Files>Binary Files
687
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FORM
/FORMAT, NDIGIT, Ftype, NWIDTH, DSIGNF, LINE, CHAR
Specifies format controls for tables.
POST1: Listing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NDIGIT
Number of digits (3 to 32) in first table column (usually the node or element number). Initially defaults
to 7.
Ftype
FORTRAN format types (initially defaults to G):
G
Gxx.yy. xx and yy are described below.
F
Fxx.yy
E
Exx.yy
NWIDTH
Total width (9 to 32) of the field (the xx in Ftype). Initially defaults to 12.
DSIGNF
Number of digits after the decimal point (yy in F or E format) or number of significant digits in G format.
Range is 2 to xx-7 for Ftype = G or E; and 0 to xx-4 for Ftype = F. Initially defaults to 5.
LINE
Number of lines (11 minimum) per page. Defaults to ILINE or BLINE from the /PAGE command.
CHAR
Number of characters (41 to 240, system-dependent) per line before wraparound. Defaults to ICHAR or
BCHAR from the /PAGE command.
Command Default
Program determines format for the data.
Notes
Specifies various format controls for tables printed with the POST1 PRNSOL, PRESOL, PRETAB, PRRSOL,
and PRPATH commands. A blank (or out-of-range) field on the command retains the current setting. Issue
/FORMAT,STAT to display the current settings. Issue /FORMAT,DEFA to reestablish the initial default specific-
ations.
For the POST26 PRVAR command, the Ftype, NWIDTH, and DSIGNF fields control the time output format.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
688
/FORMAT
FP, STITM, C1, C2, C3, C4, C5, C6
Defines the fatigue S vs. N and Sm vs. T tables.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
STITM
Starting item number for entering properties (defaults to 1). If 1, data input in field C1 of this command
is entered as the first item in the list; if 7, data input in field C1 of this command is entered as the seventh
item in the list; etc. If the item number is negative, C1-C6 are ignored and the item is deleted. If -ALL,
the table is erased. Items are as follows (items 41-62 are required only if simplified elastic-plastic code
calculations are to be performed):
1,2,...20
N1, N2, ... N20
21,22,...40
S1, S2, ... S20
41,42,...50
T1, T2, ... T10
51,52,...60
Sm1, Sm2, ... Sm10
61
M (first elastic-plastic material parameter)
62
N (second elastic-plastic material parameter)
C1, C2, C3, . . . , C6
Data inserted into six locations starting with STITM. If a value is already in one of these locations, it will
be redefined. A blank retains the previous value.
Notes
Defines the fatigue alternating stress (S) vs. cycles (N) table and the design stress-intensity value (Sm) vs.
temperature (T) table. May also be used to modify any previously stored property tables. Log-log interpolation
is used in the S vs. N table and linear interpolation is used in the Sm vs. T table. Cycles and temperatures
must be input in ascending order; S and Sm values in descending order. Table values must be supplied in
pairs, i.e., every N entry must have a corresponding S entry, etc. Not all property pairs per curve need be
used. If no S vs. N table is defined, the fatigue evaluation will not produce usage factor results. See the
Structural Analysis Guide for details.
Menu Paths
Main Menu>General Postproc>Fatigue>Property Table>Elas-plas Par
Main Menu>General Postproc>Fatigue>Property Table>Erase Tables
Main Menu>General Postproc>Fatigue>Property Table>S-N Table
Main Menu>General Postproc>Fatigue>Property Table>Sm_T Table
689
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FP
FPLIST
Lists the property table stored for fatigue evaluation.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Menu Paths
Main Menu>General Postproc>Fatigue>Property Table>List Tables
FREQ, FREQ1, FREQ2, FREQ3, FREQ4, FREQ5, FREQ6, FREQ7, FREQ8, FREQ9
Defines the frequency points for the SV vs. FREQ tables.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
FREQ1, FREQ2, FREQ3, . . . , FREQ9
Frequency points for SV vs. FREQ tables. Values must be in ascending order. Log-log interpolation will
be used between frequency points. FREQ1 should be greater than zero. Units are cycles/time.
Command Default
No frequency table.
Notes
Repeat the FREQ command for additional frequency points (100 maximum). Values are added after the last
nonzero frequency. If all fields (FREQ1 -- FREQ9) are blank, erase SV vs. FREQ tables.
Frequencies must be in ascending order.
Spectral values are input with the SV command and interpreted according to the SVTYP command. Applies
only to the SPRS (single-point) option of the SPOPT command. See the PSDFRQ command for frequency
input with the MPRS (multi-point) option.
Use the STAT command to list current frequency points.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Erase Table
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Freq Table
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Erase Table
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Freq Table
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
690
FPLIST
FRQSCL, Scaling
Turns on automatic scaling of the entire mass matrix and frequency range for modal analyses using the
Block Lanczos, PCG Lanczos, or Supernode mode extraction method.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> EH <> PP <> EME MFS
Scaling
Off
Do not use automatic scaling of the mass matrix and frequency range.
On
Use automatic scaling of the mass matrix and frequency range.
Command Default
ANSYS will use automatic scaling if appropriate.
Notes
Use this command to deactivate or force activation of automatic scaling of the entire mass matrix and fre-
quency range for modal analyses where the entire mass matrix is significantly different (i.e., orders of mag-
nitude difference) than the entire stiffness matrix (for example, due to the particular unit system being used).
Where the mass matrix is significantly smaller compared to the stiffness matrix, the eigenvalues will tend to
approach very large numbers (>10e12), making the Block Lanczos, PCG Lanczos, or Supernode mode extraction
method less efficient and more likely to miss modes.
ANSYS uses scaling (if appropriate) by default. However, you can issue FRQSCL,ON to force the entire mass
matrix and frequency range to be scaled to bring the stiffness and mass matrices closer together in terms
of orders of magnitude, improving efficiency and reducing the likelihood of missed modes. The resulting
eigenvalues are then automatically scaled back to the original system. If you are using micro MKS units,
where the density is typically very small compared to the stiffness, you may want to issue FRQSCL,ON to
force scaling on.
If the stiffness and mass are on the same scale, FRQSCL,ON has no effect.
This command is available only for modal analyses using the Block Lanczos, PCG Lanczos, or Supernode
mode extraction method (MODOPT,LANB, LANPCG, or SNODE).
This command is not valid and has no effect when used in conjunction with the MSAVE,ON command in a
modal analysis with the PCG Lanczos mode extraction method.
Menu Paths
This command cannot be accessed from a menu.
691
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FRQSCL
FS, NODE, NEV, NLOD, STITM, C1, C2, C3, C4, C5, C6
Stores fatigue stress components at a node.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE
Node number corresponding to this location. Used only to associate a node with a new location or to
find an existing location.
NEV
Event number to be associated with these stresses (defaults to 1).
NLOD
Loading number to be associated with these stresses (defaults to 1).
STITM
Starting item number for entering stresses (defaults to 1). If 1, data input in field C1 of this command
is entered as the first item in the list; if 7, data input in field C1 of this command is entered as the seventh
item in the list; etc. Items are as follows:
1-6
SX, SY, SZ, SXY, SYZ, SXZ total stress components
7
Temperature
8-13
SX, SY, SZ, SXY, SYZ, SXZ membrane-plus-bending stress components.
C1, C2, C3, . . . , C6
Stresses assigned to six locations starting with STITM. If a value is already in one of these locations, it
will be redefined. A blank retains the previous value (except in the C1 field, which resets the STITM
item to zero).
Notes
Stores fatigue stress components at a node as input on this command instead of from the current data in
the database. Stresses are stored according to the event number and loading number specified. The location
is associated with that previously defined for this node [FL] or else it is automatically defined. May also be
used to modify any previously stored stress components. Stresses input with this command should be con-
sistent with the global coordinate system for any FSNODE or FSSECT stresses used at the same location.
Menu Paths
Main Menu>General Postproc>Fatigue>Store Stresses>Specified Val
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
692
FS
FSCALE, RFACT, IFACT
Scales force load values in the database.
SOLUTION: FE Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
RFACT
Scale factor for the real component. Zero (or blank) defaults to 1.0. Use a small number for a zero scale
factor.
IFACT
Scale factor for the imaginary component. Zero (or blank) defaults to 1.0. Use a small number for a zero
scale factor.
Notes
Scales force load (force, heat flow, etc.) values in the database. Scaling applies to the previously defined
values for the selected nodes [NSEL] and the selected force labels [DOFSEL]. Issue FLIST command to review
results. Solid model boundary conditions are not scaled by this command, but boundary conditions on the
FE model are scaled.
Note
Such scaled FE boundary conditions may still be overwritten by unscaled solid model boundary
conditions if a subsequent boundary condition transfer occurs.
FSCALE does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Scale FE Loads>Forces
Main Menu>Solution>Define Loads>Operate>Scale FE Loads>Forces
FSDELE, NLOC, NEV, NLOD
Deletes a stress condition for a fatigue location, event, and loading.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Delete stresses associated with location NLOC. Defaults to zero.
NEV
Delete stresses associated with event NEV. Defaults to zero.
NLOD
Delete stresses associated with loading NLOD. Defaults to zero.
693
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FSDELE
Notes
Deletes a stress condition stored for a particular fatigue location, event, and loading. Use FE command to
delete all stresses for a particular event or FL command to delete all stresses for a particular location.
Menu Paths
Main Menu>General Postproc>Fatigue>Store Stresses>Dele Stresses
FSLIST, NLOC1, NLOC2, NINC, NEV, NLOD
Lists the stresses stored for fatigue evaluation.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC1, NLOC2, NINC
List stresses from NLOC1 (defaults to 1) to NLOC2 (defaults to NLOC1) in steps of NINC (defaults to 1).
If NLOC1 = ALL, NLOC2 and NINC are ignored and stresses for all locations are listed.
NEV
Event number for stress listing (defaults to ALL).
NLOD
Loading number for stress listing (defaults to ALL).
Notes
Stresses may be listed per location, per event, per loading, or per stress condition. Use FELIST and FLLIST
if only event and location parameters (no stresses) are to be listed.
Menu Paths
Main Menu>General Postproc>Fatigue>Store Stresses>List Stresses
FSNODE, NODE, NEV, NLOD
Calculates and stores the stress components at a node for fatigue.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE
Node number for which stress components are stored.
NEV
Event number to be associated with these stresses (defaults to 1).
NLOD
Loading number to be associated with these stresses (defaults to 1).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
694
FSLIST
Notes
Calculates and stores the total stress components at a specified node for fatigue. Stresses are stored according
to the event number and loading number specified. The location is associated with that previously defined
for this node [FL] or else it is automatically defined. Stresses are stored as six total components (SX through
SYZ). Temperature is also stored along with the total stress components. Calculations are made from the
stresses currently in the database (last *SET or LCASE command). Stresses stored are in global Cartesian
coordinates, regardless of the active results coordinate system [RSYS]. The FSLIST command may be used
to list stresses. The FS command can be used to modify stored stresses.
Menu Paths
Main Menu>General Postproc>Fatigue>Store Stresses>From rst File
FSPLOT, NLOC, NEV, ITEM
Displays a fatigue stress item for a fatigue location and event.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Display stresses associated with location NLOC.
NEV
Display stresses associated with event NEV.
ITEM
Display stresses associated with item number ITEM. Items are as follows:
16
SX, SY, SZ, SXY, SYZ, SXZ total stress components
7
Temperature
813
SX, SY, SZ, SXY, SYZ, SXZ membrane-plus-bending stress components.
Notes
Displays a fatigue stress item as a function of loading number for a particular fatigue location and event.
Menu Paths
Main Menu>General Postproc>Fatigue>Store Stresses>Plot Stresses
695
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FSPLOT
FSSECT, RHO, NEV, NLOD, KBR
Calculates and stores total linearized stress components.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
RHO
In-plane (X-Y) average radius of curvature of the inside and outside surfaces of an axisymmetric section.
If zero (or blank), a plane or 3-D structure is assumed. If nonzero, an axisymmetric structure is assumed.
Use a suitably large number (see the Theory Reference for the Mechanical APDL and Mechanical Applications)
or use -1 for an axisymmetric straight section.
NEV
Event number to be associated with these stresses (defaults to 1).
NLOD
Loading number to be associated with these stresses (defaults to 1).
KBR
For an axisymmetric analysis (RHO 0):
0
Include the thickness-direction bending stresses
1
Ignore the thickness-direction bending stresses
2
Include the thickness-direction bending stress using the same formula as the Y (axial direction )
bending stress. Also use the same formula for the shear stress.
Notes
Calculates and stores the total linearized stress components at the ends of a section path [PATH] (as defined
by the first two nodes with the PPATH command). The path must be entirely within the selected elements
(that is, there must not be any element gaps along the path). Stresses are stored according to the fatigue
event number and loading number specified. Locations (one for each node) are associated with those previ-
ously defined for these nodes [FL] or else they are automatically defined. Stresses are separated into six
total components (SX through SXZ) and six membrane-plus-bending (SX through SXZ) components. The
temperature at each end point is also stored along with the total stress components. Calculations are made
from the stresses currently in the database (last *SET or LCASE command). Stresses are stored as section
coordinate components if axisymmetric or as global Cartesian coordinate components otherwise, regardless
of the active results coordinate system [RSYS]. The FSLIST command may be used to list stresses. The FS
command can be used to modify stored stresses. See also the PRSECT and PLSECT commands for similar
calculations.
Menu Paths
Main Menu>General Postproc>Fatigue>Store Stresses>At Cross Sect
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
696
FSSECT
FSSPARM, PORT1, PORT2
Calculates reflection and transmission properties of a frequency selective surface.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
PORT1
Port number of input port. Defaults to 1.
PORT2
Port number of output port. Defaults to 1.
Notes
FSSPARM calculates reflection and transmission coefficients, power reflection and transmission coefficients,
and return and insertion losses of a frequency selective surface.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Port>FSS Parameters
FSUM, LAB, ITEM
Sums the nodal force and moment contributions of elements.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LAB
Coordinate system in which to perform summation.
(blank)
Sum all nodal forces in global Cartesian coordinate system (default).
RSYS
Sum all nodal forces in the currently active RSYS coordinate system.
ITEM
Selected set of nodes.
(blank)
Sum all nodal forces for all selected nodes (default), excluding contact elements.
CONT
Sum all nodal forces for contact nodes only.
BOTH
Sum all nodal forces for all selected nodes, including contact elements.
Notes
Sums and prints, in each component direction for the total selected node set, the nodal force and moment
contributions of the selected elements attached to the node set. Selecting a subset of nodes [NSEL] and
697
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FSUM
then issuing this command will give the total force acting on that set of nodes (default), excluding surface-
to-surface, node-to-surface, line-to-line, and line-to-surface contact elements (TARGE169, TARGE170, CONTA171,
CONTA172, CONTA173, CONTA174, CONTA175, CONTA176, and CONTA177). Setting ITEM = CONT sums the
nodal forces and moment contributions of the selected contact elements (CONTA171, CONTA172, CONTA173,
CONTA174, CONTA175, CONTA176, and CONTA177). Setting ITEM = BOTH sums the nodal forces for all se-
lected nodes, including contact elements. Nodal forces associated with surface loads are not included. The
effects of nodal coupling and constraint equations are ignored. Moment summations are about the global
origin unless another point is specified with the SPOINT command. This vector sum is printed in the global
Cartesian system unless it is transformed [RSYS] and a point is specified with the SPOINT command. By
default, the sum is done in global Cartesian, and the resulting vector is transformed to the requested system.
The LAB = RSYS option transforms each of the nodal forces into the active coordinate system before summing
and printing. The FORCE command can be used to specify which component (static, damping, inertia, or
total) of the nodal load is to be used. This command output is included in the NFORCE command.
The FSUM command should not be used with axisymmetric elements because it might calculate a moment
where none exists. Consider, for example, the axial load on a pipe modeled with an axisymmetric shell element.
The reaction force on the end of the pipe is the total force (for the full 360 degrees) at that location. The
net moment about the centerline of the pipe would be zero, but the program would incorrectly calculate
a moment at the end of the element as the force multiplied by the radius.
The FSUM command should also not be used with the SOLID117 element.
Using with the NLGEOM Command
If you have activated large deflection (via the NLGEOM,ON command), the FSUM command generates the
following message:
Summations based on final geometry and
will not agree with solution reactions.
The message warns that the moment reactions are incorrect. When computing moment reactions, the
command assumes that the summation of rotations applies; however, it does not apply for large rotations,
which require pseudovector representation to sum the rotations.
In contrast, the results for force reactions will be correct because they depend upon linear displacement
vectors (which can be added).
Menu Paths
Main Menu>General Postproc>Nodal Calcs>Total Force Sum
FTCALC, NLOC, NODE
Performs fatigue calculations for a particular node location.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Location number of stress conditions to be used for fatigue calculation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
698
FTCALC
NODE
Node number (used only for convenience if NLOC is not input).
Menu Paths
Main Menu>General Postproc>Fatigue>Calculate Fatig
FTRAN
Transfers solid model forces to the finite element model.
SOLUTION: Solid Forces
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Notes
Forces are transferred only from selected keypoints to selected nodes. The FTRAN operation is also done if
the SBCTRAN command is issued or automatically done upon initiation of the solution calculations [SOLVE].
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Transfer to FE>Forces
Main Menu>Solution>Define Loads>Operate>Transfer to FE>Forces
FTSIZE, MXLOC, MXEV, MXLOD
Defines the fatigue data storage array.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
MXLOC
Maximum number of fatigue locations (defaults to 5).
MXEV
Maximum number of fatigue events (defaults to 10).
MXLOD
Maximum number of loadings in each event (defaults to 3).
Command Default
5 locations, 10 events, 3 loadings.
Notes
Defines the size and erases the stress conditions for the fatigue data storage array. A stress condition is a
loading (stresses) at a particular location (node) for a particular event. Size is defined in terms of the maximum
699
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FTSIZE
number of locations, events, and loadings. The array size cannot be changed once data storage has begun
(without erasing all previously stored data). If a size change is necessary, see the FTWRITE command.
Menu Paths
Main Menu>General Postproc>Fatigue>Size Settings
FTWRITE, Fname, Ext, --
Writes all currently stored fatigue data on a file.
POST1: Fatigue
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to FATG if Fname is blank.
--
Unused field.
Notes
Data are written in terms of the equivalent POST1 fatigue commands [FTSIZE, FL, FS, etc.] which you
can then edit and resubmit to POST1 (with a /INPUT command).
Once you have created a fatigue data file, each subsequent use of the FTWRITE command overwrites
the contents of that file.
Menu Paths
Main Menu>General Postproc>Fatigue>Write Fatig Data
FVMESH, KEEP
Generates nodes and tetrahedral volume elements from detached exterior area elements (facets).
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEEP
Specifies whether to keep the area elements after the tetrahedral meshing operation is complete.
0
Delete area elements (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
700
FTWRITE
1
Keep area elements.
Notes
The FVMESH command generates a tetrahedral volume mesh from a selected set of detached exterior area
elements (facets). (Detached elements have no solid model associativity.) The area elements can be triangular-
shaped, quadrilateral-shaped, or a mixture of the two.
The FVMESH command is in contrast to the VMESH command, which requires a volume to be input.
The main tetrahedra mesher [MOPT,VMESH,MAIN] is the only tetrahedra mesher that supports the FVMESH
command. The alternate tetrahedra mesher [MOPT,VMESH,ALTERNATE] does not support FVMESH. MESH200
elements do not support FVMESH.
Tetrahedral mesh expansion [MOPT,TETEXPND,Value] is supported for both the FVMESH and VMESH
commands. Tet-mesh expansion is the only mesh control supported by FVMESH.
Triangle- or quadrilateral-shaped elements may be used as input to the FVMESH command. Where quadri-
laterals are used, the default behavior is for the pyramid-shaped elements to be formed at the boundary
when the appropriate element type is specified. See the MOPT,PYRA command for details.
The FVMESH command does not support multiple "volumes." If you have multiple volumes in your model,
select the surface elements for one "volume," while making sure that the surface elements for the other
volumes are deselected. Then use FVMESH to generate a mesh for the first volume. Continue this procedure
by selecting one volume at a time and meshing it, until all of the volumes in the model have been meshed.
If an error occurs during the meshing operation, the area elements are kept even if KEEP = 0.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Tet Mesh From>Area Elements
701
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
FVMESH
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
702
G Commands
GAP
Specifies "Reduced transient gap conditions" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Gap Conditions
GAPF, NVAR, NUM, Name
Defines the gap force data to be stored in a variable.
POST26: Set Up
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NVAR
Arbitrary reference number assigned to this variable (2 to NV [NUMVAR]). Overwrites any existing results
for this variable.
NUM
Number identifying gap number for which the gap force is to be stored. Issue the GPLIST command to
display gap numbers.
Name
Thirty-two character name for identifying the item on the printout and displays (defaults to the name
GAPF).
Notes
Defines the gap force data to be stored in a variable. Applicable only to the expansion pass of the reduced
or mode superposition linear transient dynamic (ANTYPE,TRANS) analysis. The data usually on Fname.RDSP.
703
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Menu Paths
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
GAPFINISH
Exits from the CAD import topology repair stage.
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Data will remain intact in the database, but the database is not automatically written to a file (use the SAVE
command to write the database to a file).
This command is part of the suite of commands available to find gaps and repair the topology of a model
imported from a CAD system through the Default IGES option. These commands are made available if, during
a CAD import operation, ANSYS detects topological defects (small gaps, large holes, or overlapping entities).
You cannot use the GAPMERGE command on a model after issuing the GAPFINISH command; however,
you can continue to repair the model with the geometry repair commands.
Menu Paths
Main Menu>Preprocessor>Modeling>Topo Repair>Finish
GAPLIST, Lab
Lists all joined or disjoined lines in a model (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label defining the type of plot:
OPEN
Lists all disjoined edges. This is the default option.
CLOSED
Lists all seam lines that are currently "sewn" to adjacent areas.
Notes
Disjoined lines are those lines which have only one area attached to them. Typically, these occur around
gaps in the model and should be merged with adjacent disjoined lines using the GAPMERGE command
(topological repair) or ARFILL command (geometry repair).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
704
GAPFINISH
This command is available for both topological and geometric repair for models imported from CAD files
through the Default IGES option. If topological repair is active, you can specify what is listed through the
GAPOPT,OESELE command.
Menu Paths
Main Menu>Preprocessor>Modeling>Geom Repair>Lst Model Gaps>Close Edges
Main Menu>Preprocessor>Modeling>Geom Repair>Lst Model Gaps>Open Edges
Main Menu>Preprocessor>Modeling>Topo Repair>Lst Model Gaps>Close Edges
Main Menu>Preprocessor>Modeling>Topo Repair>Lst Model Gaps>Open Edges
GAPMERGE, Lab, VAL1, VAL2, VAL3
Merges adjacent disjoined lines (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label defining the type of merge operation.
TOLER
Disjoined lines are merged if they are within the tolerance factor specified by the GAPOPT command.
ITER
Disjoined lines are merged through an iterative merging process, where a series of merges occur
starting with the tolerance factor specified in VAL1 until the value specified in VAL2 is reached. The
merges occur at increments specified through VAL3. By default, VAL1 is set to its minimum value
(1), VAL2 is set to its maximum value (10), and VAL3 is set to 1.
Notes
While the Lab = ITER options provides a robust, automatic method for "sewing" gaps, it also requires con-
siderable processing.
This command is part of the suite of commands available to repair the topology of a model imported from
CAD systems through the Default IGES option (see Importing Solid Models in the Modeling and Meshing
Guide). These commands are made available if, during a CAD import operation, ANSYS detects topological
defects (small gaps, large holes, or overlapping entities). You cannot use the topological repair commands
on a model after issuing the GAPFINISH command; however, you can continue to repair the model with
the geometry commands.
Menu Paths
Main Menu>Preprocessor>Modeling>Topo Repair>Mrg Model Gaps>By Tolerance
Main Menu>Preprocessor>Modeling>Topo Repair>Mrg Model Gaps>Iterative
705
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GAPMERGE
GAPOPT, Lab, Value
Sets preferences for the CAD import repair commands.
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label Identifying the preference set by the command. The meaning of Value varies depending on Lab.
TOLER
Set tolerance for the GAPMERGE (used only for manual, not iterative, merging), GAPLIST, and
GAPPLOT commands as a factor, defined by Value, times the default tolerance. By default, Value
is set to its minimum value.
OESELE
Determines what will be displayed by the GAPLIST and GAPPLOT commands. Value can be:
ALL
GAPLIST and GAPPLOT will display all currently existing disjoined edges with both small gaps
and large holes. This is the default option.
MERGE
GAPLIST and GAPPLOT will display all open edges that can be merged with the current tolerance.
REMAIN
GAPLIST and GAPPLOT will display all open edges that will remain after merging is performed
using the current tolerance.
Value
Additional input value described under Lab.
Notes
This command is used for the topological repair of models imported from CAD files through the Default
IGES option. If no previous defeaturing has been done, setting the tolerance will automatically merge all
gaps by the specified value.
Menu Paths
Main Menu>Preprocessor>Modeling>Topo Repair>Preferences
GAPPLOT, Lab
Plots all joined or disjoined lines (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label defining the type of plot:
OPEN
Plots disjoined edges. This is the default option.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
706
GAPOPT
CLOSED
Plots all edge lines that are currently "sewn" to adjacent areas.
ALL
Plots all line segments (both joined and disjoined). Disjoined lines plot in a different color.
Notes
Disjoined lines are those lines which have only one area attached to them. Typically, these occur around
gaps in the model and should be merged with adjacent disjoined lines using the GAPMERGE (topological
repair) or ARFILL (geometry repair) command.
This command is available for both topological and geometric repair for models imported from CAD files
through the Default IGES option. If topological repair is active, you can specify what is plotted through the
GAPOPT,OESELE command.
Menu Paths
Main Menu>Preprocessor>Modeling>Geom Repair>Plt Model Gaps>Closed Edges
Main Menu>Preprocessor>Modeling>Geom Repair>Plt Model Gaps>Open Edges
Main Menu>Preprocessor>Modeling>Geom Repair>Plt Model Gaps>Opn & Closed
Main Menu>Preprocessor>Modeling>Topo Repair>Plt Model Gaps>Closed Edges
Main Menu>Preprocessor>Modeling>Topo Repair>Plt Model Gaps>Open Edges
Main Menu>Preprocessor>Modeling>Topo Repair>Plt Model Gaps>Opn & Closed
GAUGE, Opt, FREQ
Gauges the problem domain for a magnetic edge-element formulation.
SOLUTION: Analysis Options
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS
Opt
Type of gauging to be performed:
ON
Perform tree gauging of the edge values (default).
OFF
Gauging is off. (You must specify custom gauging via APDL specifications.)
STAT
Gauging status (returns the current Opt and FREQ values)
FREQ
The following options are valid when Opt = ON:
0
Generate tree-gauging information once, at the first load step. Gauging data is retained for subsequent
load steps. (This behavior is the default.)
1
Repeat gauging for each load step. Rewrites the gauging information at each load step to accom-
modate changing boundary conditions on the AZ degree of freedom (for example, adding or deleting
AZ constraints via the D or CE commands).
707
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GAUGE
Notes
The GAUGE command controls the tree-gauging procedure required for electromagnetic analyses using an
edge-based magnetic formulation (elements SOLID117, SOLID236 and SOLID237).
Gauging occurs at the solver level for each solution (SOLVE). It sets additional zero constraints on the edge-
flux degrees of freedom AZ to produce a unique solution; the additional constraints are removed after
solution.
Use the FREQ option to specify how the command generates gauging information for multiple load steps.
Access the gauging information via the _TGAUGE component of gauged nodes. The program creates and
uses this component internally to remove and reapply the AZ constraints required by gauging. If FREQ =
0, the _TGAUGE component is created at the first load step and is used to reapply the tree gauge constraints
at subsequent load steps. If FREQ = 1, the tree-gauging information and the _TGAUGE component are
generated at every load step
If gauging is turned off (GAUGE,OFF), you must specify your own gauging at the APDL level.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Magnetics>Options Only>Gauging
Main Menu>Solution>Load Step Opts>Magnetics>Options Only>Gauging
/GCMD, WN, Lab1, Lab2, Lab3, Lab4, Lab5, Lab6, Lab7, Lab8, Lab9, Lab10, Lab11, Lab12
Controls the type of element or graph display used for the GPLOT command.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which this command applies (defaults to 1)
Lab1, Lab2, Lab3, . . . , Lab12
Command labels (for example, PLNSOL,S,X)
Notes
This command controls the type of element or graph display that appears when you issue the GPLOT
command when the /GTYPE,,(ELEM or GRPH) entity type is active. If you have multiple plotting windows
enabled, you can also use /GCMD to select one window when you wish to edit its contents.
For related information, see the descriptions of the GPLOT and /GTYPE commands in this manual.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Multi-Plot Contrls
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
708
/GCMD
/GCOLUMN, CURVE, STRING
Allows the user to apply a label to a specified curve.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
CURVE
Curve number on which label will be applied (integer value between 1 and 10).
STRING
Name or designation that will be applied to the curve (8 characters max).
Notes
This command is used for an array parameter plot (a plot created by the *VPLOT command). Normally the
label for curve 1 is "COL 1", the label for curve 2 is "COL 2" and so on; the column number is the field con-
taining the dependent variables for that particular curve. Issuing /GCOLUMN,CURVE, with no string value
specified resets the label to the original value.
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Curve
GENOPT
Specifies "General options" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>General Options
709
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GENOPT
GEOM, K2D, NDIV
Defines the geometry specifications for the radiation matrix calculation.
AUX12: Radiation Matrix Method
MP ME ST PR <> <> <> <> <> <> <> PP <> EME MFS
K2D
Dimensionality key:
0
3-D geometry
1
2-D geometry (plane or axisymmetric)
NDIV
Number of divisions in an axisymmetric model. Used only with K2D = 1. Defaults to 0 (2-D plane). The
2-D model is internally expanded to a 3-D model based on the number of divisions specified (6
NDIV
n
NSOL)
and some apply only to response surface sets (i.e., where NSOL+1 n NSOL+NRSS).
*GET,Par, PDS, n, Item1, IT1NUM, Item2, IT2NUM
Description IT1NUM Item1
Name of the analysis method used in the n-th solution set (MCS,
RSM). A character parameter of up to 8 characters is returned. This
applies only to solution sets (i.e., "n" ranges from 1 to NSOL).
METH
Number of simulation samples that are available for postprocessing
in the n-th result set. If the n-th result set is a solution set (i.e., 1
NSIM
n
NSOL)
and some apply only to response surface sets (i.e., where NSOL+1 n NSOL+NRSS).
*GET,Par, PDS, n, Item1, IT1NUM, Item2, IT2NUM
Description IT1NUM Item1
Item2 = RP; IT2NUM=j; The minimum sampled value of the j-th
defined random output parameter in the n-th result set.
MIN
Item2 = RV; IT2NUM=j; The minimum sampled value of the j-th
defined random input variable in the n-th result set.
MAX
Item2 = RP; IT2NUM=j; The minimum sampled value of the j-th
defined random output parameter in the n-th result set.
MAX
Item2 = VTYP; Type of the level values of the design-of-experiment
for a central composite design of the i-th defined random input
i CCDL
variable that has been used during the execution of the n-th solu-
tion set (PROB, PHYS). A character parameter of up to 8 characters
is returned. This applies only to solution sets (i.e., n ranges from 1
to NSOL).
Item2 = LOPT; Type of the level definition of the design-of-experi-
ment for a central composite design of the i-th defined random
i CCDL
input variable that has been used during the execution of the n-th
solution set (BND, ALL). A character parameter of up to 8 characters
is returned. This applies only to solution sets (i.e., n ranges from 1
to NSOL).
Item2 = LDEF; IT2NUM=j; Flag indicating if the j-th design-of-exper-
iment level for a central composite design of the i-th defined ran-
i CCDL
dom input variable has been specified for the n-th solution set.
(0=NO, 1=YES). This applies only to solution sets (i.e., n ranges from
1 to NSOL).
Item2 = LVAL; IT2NUM=j; Level value for the j-th design-of-experi-
ment level for a central composite design of the i-th defined ran-
i CCDL
dom input variable that has been used during the execution of the
n-th solution set. This applies only to solution sets (i.e., n ranges
from 1 to NSOL).
Item2 = VTYP; Type of the level values of the design-of-experiment
level for a Box-Behnken Matrix design of the i-th defined random
i BBML
input variable that has been used during the execution of the n-th
solution set. (PROB, PHYS). A character parameter of up to 8 char-
acters is returned. This applies only to solution sets (i.e., n ranges
from 1 to NSOL).
Item2 = LOPT; Type of the level definition of the design-of-experi-
ment level for a Box-Behnken Matrix design of the i-th defined
i BBML
random input variable that has been used during the execution of
the n-th solution set (BND, ALL). A character parameter of up to 8
763
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*GET
Entity = PDS, ENTNUM = n (n-th result set).
The numbering or ordering of the result sets is used as follows: If NSOL is the number of
solution sets and NRSS is the number response surface sets then the solution sets are indexed
from 1 to NSOL and the response surface sets are indexed from NSOL+1 to NSOL+NRSS.
Note that some options listed below apply only to solution sets (i.e., where 1
n
NSOL)
and some apply only to response surface sets (i.e., where NSOL+1 n NSOL+NRSS).
*GET,Par, PDS, n, Item1, IT1NUM, Item2, IT2NUM
Description IT1NUM Item1
characters is returned. This applies only to solution sets (i.e., n
ranges from 1 to NSOL).
Item2 = LDEF; IT2NUM=j; Flag indicating if the j-th design-of-exper-
iment level for a Box-Behnken Matrix design of the i-th defined
i BBML
random input variable specified for the n-th solution set. (0=NO,
1=YES).
Item2 = LVAL; IT2NUM=j; Level value for the j-th design-of-experi-
ment level for a Box-Behnken Matrix design of the i-th defined
i BBML
random input variable that has been used during the execution of
the n-th solution set. This applies only to solution sets (i.e., n ranges
from 1 to NSOL).
Item2 = XSOL; Index of the solution set the response surface set
identified by the n-th result set is associated with. This applies only
to response surface sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
RSST
Item2 = NFRP; Number of fitted random output parameters in the
response surface set identified by the n-th result set is associated
RSST
with. This applies only to response surface sets (i.e., n ranges from
NSOL+1 to NSOL+NRSS).
Item2 = XFRP; Index of the random output parameter that has been
fitted with the RSFITcommand to fit the i-th response surface in
i RSUR
the response surface set identified by the n-th result set. IT1NUM=i
ranges from 1 to NFRP (see Item1=RSST, Item2=NFRP). This applies
only to response surface sets (i.e., n ranges from NSOL+1 to
NSOL+NRSS).
Item2 = RMOD; Label for the response surface model that has been
used in the RSFITcommand to fit the i-th response surface in the
i RSUR
response surface set identified by the n-th result set. IT1NUM=i
ranges from 1 to NFRP (see Item1=RSST, Item2=NFRP). A character
parameter of up to 8 characters is returned. This applies only to
response surface sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = YTRT; Label of the type of transformation that has been
used in the RSFITcommand to fit the random output parameter
i RSUR
of i-th response surface in the response surface set identified by
the n-th result set. IT1NUM=i ranges from 1 to NFRP (see
Item1=RSST, Item2=NFRP). A character parameter of up to 8 char-
acters is returned. This applies only to response surface sets (i.e.,
n ranges from NSOL+1 to NSOL+NRSS).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
764
*GET
Entity = PDS, ENTNUM = n (n-th result set).
The numbering or ordering of the result sets is used as follows: If NSOL is the number of
solution sets and NRSS is the number response surface sets then the solution sets are indexed
from 1 to NSOL and the response surface sets are indexed from NSOL+1 to NSOL+NRSS.
Note that some options listed below apply only to solution sets (i.e., where 1
n
NSOL)
and some apply only to response surface sets (i.e., where NSOL+1 n NSOL+NRSS).
*GET,Par, PDS, n, Item1, IT1NUM, Item2, IT2NUM
Description IT1NUM Item1
Item2 = YTRV; Transformation value of the transformation that has
been used in the RSFITcommand to fit the random output para-
i RSUR
meter of i-th response surface in the response surface set identified
by the n-th result set. IT1NUM=i ranges from 1 to NFRP (see
Item1=RSST, Item2=NFRP). This applies only to response surface
sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = FILT; Label of the filtering type of the regression terms that
has been used in the RSFITcommand to fit the random output
i RSUR
parameter of i-th response surface in the response surface set
identified by the n-th result set. IT1NUM=i ranges from 1 to NFRP
(see Item1=RSST, Item2=NFRP). A character parameter of up to 8
characters is returned. This applies only to response surface sets
(i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = CONF; Confidence level value of the regression term filtering
that has been used in the RSFITcommand to fit the random output
i RSUR
parameter of i-th response surface in the response surface set
identified by the n-th result set. IT1NUM=i ranges from 1 to NFRP
(see Item1=RSST, Item2=NFRP). This applies only to response surface
sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = YBOX; Box-Cox transformation value lambda of the re-
sponse surface equation for the i-th fitted random output parameter
i RSEQ
in the n-th result set. IT1NUM=i ranges from 1 to NFRP (see
Item1=RSST, Item2=NFRP). This applies only to response surface
sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = NTRM; Number of regression terms of the response surface
equation for the i-th fitted random output parameter in the n-th
i RSEQ
result set. IT1NUM=i ranges from 1 to NFRP (see Item1=RSST,
Item2=NFRP). This applies only to response surface sets (i.e., n
ranges from NSOL+1 to NSOL+NRSS).
Item2 = TTYP; IT2NUM=j; Type of the j-th regression term of the
response surface equation for the i-th fitted random output para-
i RSEQ
meter in the n-th result set. IT1NUM=i ranges from 1 to NFRP (see
Item1=RSST, Item2=NFRP). IT2NUM=j ranges from 1 to NTRM (see
Item1=RSEQ, Item2=NTRM). This applies only to response surface
sets (i.e., n ranges from NSOL+1 to NSOL+NRSS). Possible return
values are:
1 = term is a constant (this term does not involve any random input
variables)
765
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*GET
Entity = PDS, ENTNUM = n (n-th result set).
The numbering or ordering of the result sets is used as follows: If NSOL is the number of
solution sets and NRSS is the number response surface sets then the solution sets are indexed
from 1 to NSOL and the response surface sets are indexed from NSOL+1 to NSOL+NRSS.
Note that some options listed below apply only to solution sets (i.e., where 1
n
NSOL)
and some apply only to response surface sets (i.e., where NSOL+1 n NSOL+NRSS).
*GET,Par, PDS, n, Item1, IT1NUM, Item2, IT2NUM
Description IT1NUM Item1
2 = term is a linear term (this term involves only one random input
variable)
3 = term is a purely quadratic term involving only one random input
variable (this term involves only one random input variable)
4 = term is a mixed quadratic term involving two random input
variables (this term involves two random input variables)
Item2 = XRV1; IT2NUM=j; Index of the first random input variable
involved in the j-th regression term of the response surface equation
i RSEQ
for the i-th fitted random output parameter in the n-th result set.
IT1NUM=i ranges from 1 to NFRP (see Item1=RSST, Item2=NFRP).
This applies only to response surface sets (i.e., n ranges from
NSOL+1 to NSOL+NRSS). An error appears if the term does not in-
volve a random input variable, i.e. if the term is a constant.
Item2 = XRV2; IT2NUM=j; Index of the second random input variable
involved in the j-th regression term of the response surface equation
i RSEQ
for the i-th fitted random output parameter in the n-th result set.
IT1NUM=i ranges from 1 to NFRP (see Item1=RSST, Item2=NFRP).
This applies only to response surface sets (i.e., n ranges from
NSOL+1 to NSOL+NRSS). An error appears if the term does not in-
volve a second random input variable, i.e. if the term is not a mixed
quadratic term.
Item2 = COEF; IT2NUM=j; Regression coefficient of the j-th regression
term of the response surface equation for the i-th fitted random
i RSEQ
output parameter in the n-th result set. IT1NUM=i ranges from 1
to NFRP (see Item1=RSST, Item2=NFRP). This applies only to re-
sponse surface sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = SLOP; IT2NUM=j; Scaling slope of the j-th random input
variable of the response surface equation for the i-th fitted random
i RSEQ
output parameter in the n-th result set. IT1NUM=i ranges from 1
to NFRP (see Item1=RSST, Item2=NFRP). This applies only to re-
sponse surface sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Item2 = ICPT; IT2NUM=j; Scaling intercept of the j-th random input
variable of the response surface equation for the i-th fitted random
i RSEQ
output parameter in the n-th result set. IT1NUM=i ranges from 1
to NFRP (see Item1=RSST, Item2=NFRP). This applies only to re-
sponse surface sets (i.e., n ranges from NSOL+1 to NSOL+NRSS).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
766
*GET
Menu Paths
Main Menu>Prob Design>Prob Method>Response Surface
Utility Menu>Parameters>Get Scalar Data
/GFILE, SIZE
Specifies the pixel resolution on Z-buffered graphics files.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SIZE
Pixel resolution. Defaults to a pixel resolution of 800. Valid values are from 256 to 2400.
Command Default
800 pixels
Notes
Defines the pixel resolution on subsequently written graphics files (Jobname.GRPH) for software Z-buffered
displays [/TYPE]. Lowering the pixel resolution produces a "fuzzier" image; increasing the resolution produces
a "sharper" image but takes a little longer.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Redirect Plots>To GRPH File
Utility Menu>PlotCtrls>Redirect Plots>To PSCR File
/GFORMAT, Ftype, NWIDTH, DSIGNF
Specifies the format for the graphical display of numbers.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Ftype
FORTRAN format types (G is the default if this field is left blank.)
G
Gxx.yy. xx and yy are described below.
F
Fxx.yy
E
Exx.yy
NWIDTH
Total width (12 maximum) of the field (the xx in Ftype). Defaults to 12.
767
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GFORMAT
DSIGNF
Number of digits after the decimal point (yy in F or E format) or number of significant digits in G format.
Range is 1 to xx-6 for Ftype = G or E; and 0 to xx-3 for Ftype = F. The default is a function of Ftype
and NWIDTH.
Notes
Lets you control the format of the graphical display of floating point numbers. Issue /GFORMAT,STAT to
display the current settings; issue /GFORMAT,DEFA to let ANSYS choose the format for the graphical display
of floating numbers.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Floating Point Format
/GLINE, WN, STYLE
Specifies the element outline style.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
STYLE
Outline key:
0
Solid element outlines (default)
1
Dashed element outlines
-1
No element outlines
Command Default
Solid element outlines
Notes
Determines the element outline style. Often used when node numbers are displayed to prevent element
lines from overwriting node numbers.
Unless you are using an OpenGL or Starbase driver, the dashed element outline option (/GLINE,WN,1) is not
available in the following situations:
Z-buffered displays (/TYPE,WN,6).
Capped Z-buffered displays (/TYPE,WN,7).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
768
/GLINE
Qslice Z-buffered displays (/TYPE,WN,8).
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Edge Options
/GMARKER, CURVE, KEY, INCR
Specifies the curve marking style.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
CURVE
Curve number markers will be applied on (integer value between 1 and 10).
KEY
Marker key:
0
No markers will be applied (default).
1
TRIANGLES will be applied.
2
SQUARES will be applied.
3
DIAMONDS will be applied.
4
CROSSES will be applied.
INCR
Determines the curve marking frequency. (a whole number value between 1 and 255). If INCR = 1,
markers are displayed at every data point on the curve. If INCR = 2 then markers are displayed at every
second data point. If INCR = 3 then they are displayed at every third data point.
Command Default
No markers will be applied.
Notes
The user-specified markers will not be drawn when the area under the curve is color-filled (/GROPT, FILL).
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Curve
769
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GMARKER
GMATRIX, SYMFAC, Condname, NUMCOND, --, Matrixname
Performs electric field solutions and calculates the self and mutual conductance between multiple con-
ductors.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
SYMFAC
Geometric symmetry factor. Conductance values are scaled by this factor which represents the fraction
of the total device modeled. Defaults to 1.
Condname
Alphanumeric prefix identifier used in defining named conductor components.
NUMCOND
Total number of components. If a ground is modeled, it is to be included as a component.
--
Unused field.
Matrixname
Array name for computed conductance matrix. Defaults to GMATRIX.
Notes
To invoke the GMATRIX macro, the exterior nodes of each conductor must be grouped into individual
components using the CM command. Each set of independent components is assigned a component name
with a common prefix followed by the conductor number. A conductor system with a ground must also include
the ground nodes as a component. The ground component is numbered last in the component name se-
quence.
A ground conductance matrix is a matrix relating current to a voltage vector. A ground matrix cannot be
applied to a circuit modeler such as SPICE. The lumped conductance matrix is a matrix formed by a combin-
ation of lumped "arrangements" of voltage differences between conductors. You can use the lumped con-
ductance terms in a circuit modeler such as SPICE to represent conductances between conductors.
You must enclose all name-strings in single quotes in the GMATRIX command line.
GMATRIX works with the following elements:
SOLID5 (KEYOPT(1) = 9)
SOLID98 (KEYOPT(1) = 9)
LINK68
PLANE230
SOLID231
SOLID232
This command is available from the menu path shown below only if existing results are available.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
770
GMATRIX
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Component Based>Self/Mutual conductance
GMFACE, Lab, N
Specifies the facet representation used to form solid models.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Valid Labels:
FINE
Value that determines how coarse the facets will be.
N
An integer value between one (small) and ten (large) that determines the tolerances that will be applied
to the creation of arcs and surfaces. Ten will create many facets, which may in turn cause ANSYS to run
very slowly. One will create fewer facets, which may in turn cause larger tolerance errors.
Menu Paths
Utility Menu>PlotCtrls>Style>Solid Model Facets
*GO, Base
Causes a specified line on the input file to be read next.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Base
"Go to" action:
:label --
A user-defined label (beginning with a colon (:), 8 characters maximum). The command reader will
skip (and wrap to the beginning of the file, if necessary) to the first line that begins with the
matching :label.
Caution
This label option may not be mixed with do-loop or if-then-else constructs.
STOP --
This action will cause an exit from the ANSYS program at this line.
771
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*GO
Command Default
Read lines sequentially.
Notes
Causes the next read to be from a specified line on the input file. Lines may be skipped or reread. The *GO
command will not be executed unless it is part of a macro, user file (processed by *USE), an alternate input
file (processed by /INPUT), or unless it is used in a batch-mode input stream. Jumping into, out of, or within
a do-loop or an if-then-else construct to a :label line is not allowed.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
/GO
Reactivates suppressed printout.
SESSION: List Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Reactivates printout suppressed with the /NOPR command without producing any output. The /GOPR
command has the same function except that it also produces a command response from the program.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Rigid Region
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
/GOLIST
Reactivates the suppressed data input listing.
SESSION: List Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Reactivates printout of the data input listing suppressed with /NOLIST.
This command is valid in any processor, but only within a batch run [/BATCH].
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
772
/GO
Menu Paths
This command cannot be accessed from a menu.
/GOPR
Reactivates suppressed printout.
SESSION: List Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Reactivates printout suppressed with the /NOPR command. The /GO command has the same function except
that it does not produce a command response from the program.
This command is valid in any processor.
Menu Paths
Main Menu>Solution>Time Controls>Time Step Prediction
GP, NODE1, NODE2, Lab, STIF, GAP, DAMP
Defines a gap condition for transient analyses.
SOLUTION: Gap Conditions
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE1
Node I of gap. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI).
NODE2
Node J of gap (must be different from NODE1). Non-grounded gap nodes must be defined as master
degrees of freedom or be unconstrained, active DOF in a full analysis type. Grounded gap nodes (those
not defined as MDOF) need not appear elsewhere in the model.
Lab
Direction of gap action in the nodal coordinate system (implied from the following force labels): FX, FY,
FZ, MX, MY, MZ.
STIF
Stiffness (Force/Length) of closed gap (may be positive or negative).
Note
High stiffness requires a small integration time step for numerical stability.
773
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GP
GAP
Initial size of gap. A zero (or positive) value assumes an initially open gap. A negative value defines an
interference condition. For a rotational gap, GAP should be in radians.
DAMP
Damping coefficient (Force*Time/Length) of closed gap using pseudo velocity (Newmark finite difference
expansion scheme).
Notes
Defines a gap condition for the reduced or mode superposition transient (ANTYPE,TRANS) analysis. If used
in SOLUTION, this command is valid only within the first load step.
Repeat GP command for additional gap conditions. Gaps are numbered sequentially as input.
Note
Gaps may be renumbered by the program during the solution (see output listing)
The reduced or mode superposition transient analysis (ANTYPE,TRANS and TRNOPT,REDUC or MSUP) does
not allow gap action with the standard ANSYS gap elements. Gap conditions, however, producing the same
effect, may be defined. The gap condition simulates the basic gap action of the COMBIN40 element. The
gap condition is treated as an explicit force (equal to the interference times contact stiffness) and affects
only the load vector calculation and not the reduced stiffness matrix. The interference is calculated from the
displacement extrapolated from the previous time points. A gap condition may be defined between a master
degree of freedom and ground or another master degree of freedom. When a non-reduced mode extraction
method is used, a master degree of freedom implies an unconstrained, active degree of freedom. Gap nodes
not defined as master degrees of freedom or attached to an element are assumed to be grounded. Grounded
gap nodes need not be given a spatial location nor do they need to be located on an element. Gap conditions
may be defined in parallel (across the same nodes), with varying gap and stiffness values, to simulate a
nonlinear (piecewise) force-deflection curve.
The gap direction is determined from the force label input on the GP command, i.e., FX defines a translational
gap acting in the UX nodal degree of freedom direction, and MZ defines a rotational gap acting in the
nodal ROTZ degree of freedom direction. The actual degree of freedom directions available for a particular
node depends upon the degrees of freedom associated with the element types [ET] at that node. For example,
degrees of freedom available with BEAM3 elements are UX, UY, and ROTZ only (so that only gap labels FX,
FY, and MZ are valid).
If the coordinate systems of the nodes connecting the gap are rotated relative to each other, the same degree
of freedom may be in different directions. The gap, however, assumes only a one-dimensional action. Nodes
I and J may be anywhere in space (preferably coincident). No moment effects are included due to noncoin-
cident nodes. That is, if the nodes are offset from the line of action, moment equilibrium may not be satisfied.
The contact stiffness value represents the stiffness of the closed gap. Stiffness values are related to the in-
tegration time step size and should be physically reasonable. High stiffness will require a small integration
time step, otherwise, due to the displacement extrapolation, the solution may go unstable. Negative stiffness
values may be used with gaps in parallel to produce a decreasing force-deflection curve.
The gap conditions, if any, should be defined in the first load step. Appearances in succeeding load steps
are ignored. The order of specifying the gap nodes is important, i.e., a gap condition connecting two nodes
will act differently depending upon which node is specified first on the GP command. For example, for Node
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
774
GP
1 at X = 0.0, Node 2 at X = 0.1, and the gap defined from Node 1 to 2, a displacement of Node 1 greater
than Node 2 will cause the gap to close. For the gap defined from Node 2 to 1, a displacement of Node 2
greater than Node 1 will cause the gap to close (like a hook action). In general, the gap closes whenever
the separation (defined as U
J
- U
I
+ GAP) is negative. U
J
is the displacement of node J, U
I
is the displacement
of node I, and GAP is the input gap value. The gap force output appears in the printout only for the time
steps for which the gap is closed. A negative spring force is always associated with a closed gap (even with
the hook option).
The nonlinear gap damping provided through the DAMP field runs faster than a full transient analysis using
a gap element (COMBIN40). Only ANTYPE = TRANS and TRNOPT = MSUP allow the nonlinear gap damping
action. Damping conditions are ignored for the reduced transient analysis method.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Dynamic Gap Cond>Define
Main Menu>Solution>Dynamic Gap Cond>Define
GPDELE, GAP1, GAP2, GINC
Deletes gap conditions.
SOLUTION: Gap Conditions
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
GAP1, GAP2, GINC
Delete gap conditions from GAP1 to GAP2 (defaults to GAP1) in steps of GINC (defaults to 1).
Notes
Deletes gap conditions defined with the GP command. Gap conditions following those deleted are automat-
ically compressed and renumbered. If used in SOLUTION, this command is valid only within the first load
step.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Dynamic Gap Cond>Delete
Main Menu>Solution>Dynamic Gap Cond>Delete
775
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GPDELE
GPLIST, GAP1, GAP2, GINC
Lists the gap conditions.
SOLUTION: Gap Conditions
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
GAP1, GAP2, GINC
List gap conditions from GAP1 to GAP2 (GAP2 defaults to GAP1) in steps of GINC (defaults to 1). If
GAP1 = ALL (default), GAP2 and GINC are ignored and all gap conditions are listed.
Notes
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Loads>Dynamic Gap Cond>List All
Main Menu>Preprocessor>Loads>Dynamic Gap Cond>List Specified
Main Menu>Solution>Dynamic Gap Cond>List All
Main Menu>Solution>Dynamic Gap Cond>List Specified
Utility Menu>List>Other>Gap Conditions
GPLOT
Controls general plotting.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This command displays all entity types as specified via the /GTYPE command. Only selected entities (NSEL,
ESEL, KSEL, LSEL, ASEL, VSEL) will be displayed. See the descriptions of the /GTYPE and /GCMD commands
for methods of setting the entity types displayed.
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Multi-Plots
/GRAPHICS, Key
Defines the type of graphics display.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Key
Graphics key:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
776
GPLIST
FULL
Display all model geometry and results.
POWER
Activate PowerGraphics (default when GUI is on).
Command Default
PowerGraphics ON (Key = POWER).
Notes
The /GRAPHICS command specifies the type of graphics display. Key = POWER activates the PowerGraphics
capability. PowerGraphics offers faster plotting than the Key = FULL option, and speeds up element, results,
area, line, and volume displays. PowerGraphics mode (the default) is automatically invoked when the GUI
is accessed. This action supersedes all prior macros or start up routines (start.ans, config.ans, etc.). Full
graphics mode can be accessed only by issuing /GRAPHICS, FULL after the GUI is active.
Results values (both printed and plotted) may differ between the Key = FULL and Key = POWER options
because each option specifies a different set of data for averaging and display. For Key = FULL, all element
and results values (interior and surface) are included. For Key = POWER, only element and results values
along the model exterior surface are processed.
Caution
If you have specified one facet per element edge for PowerGraphics displays (via the /EFACET command
or via choices from the General Postproc or Utility Menus), PowerGraphics does not plot midside nodes.
The /EFACET command is only applicable to element type displays. (See the descriptions of these commands
for more information.)
Maximum values shown in plots can differ from printed maximum values. This is due to different averaging
schemes used for plotted and printed maximum values.
PowerGraphics displays do not average at geometric discontinuities. The printouts in PowerGraphics will,
however, provide averaging information at geometric discontinuities if the models do not contain shell
elements. Carefully inspect the data you obtain at geometric discontinuities.
Note
In Full Graphics mode, it is possible to deselect an individual node, select all elements (including
the element that contains that node), and then perform postprocessing calculations on those
elements and have that unselected node not be considered in those calculations. However, if
PowerGraphics is active, postprocessing always displays based on selected elements.
PowerGraphics does not support membrane shell elements such as SHELL41, and these elements using the
membrane-stiffness-only option (KEYOPT1) = 1): SHELL181, SHELL208, and SHELL209.
Commands that are not supported by PowerGraphics are listed below. These commands are executed using
the Key = FULL option, regardless of whether PowerGraphics is activated. Only certain options for /CTYPE,
777
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GRAPHICS
/EDGE, /ESHAPE, *GET, /PNUM, /PSYMB, SHELL, and *VGET are not supported by PowerGraphics. (See the
descriptions of these commands for more information.)
/SSCALE PRETAB /PBF ESYS /CTYPE
/SHRINK PRSECT PLCRACK *GET DNSOL
TALLOW PRVECT PLETAB LAYER /EDGE
*VGET /PSYMB PLLS /NORMAL ERNORM
*VPUT SALLOW PLSECT NSEL ESORT
SHELL /PNUM NSORT
Menu Paths
Utility Menu>PlotCtrls>Style>Hidden-Line Options
/GRESUME, Fname, Ext, --
Sets graphics settings to the settings on a file.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to GSAV if Fname is blank.
--
Unused field.
Notes
Causes a file to be read to reset the graphics slash (/) commands as they were at the last /GSAVE command.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Restore Plot Ctrls
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
778
/GRESUME
/GRID, KEY
Selects the type of grid on graph displays.
GRAPHICS: Graphs
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Grid key:
0 (OFF)
No grid.
1 (ON)
Full grid (X and Y grid lines).
2 (X)
Partial grid (X grid lines only).
3 (Y)
Partial grid (Y grid lines only)
Command Default
No grid.
Notes
Selects the type of grid on graph displays. Graphs with multiple Y-axes can have multiple grids [/GRTYP].
The grid of the first curve is also used as the background grid (above and behind the curve). Grids for other
curves are limited to be under the curves. See also /GTHK and /GROPT for other grid options.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Grid
/GROPT, Lab, KEY
Sets various line graph display options.
GRAPHICS: Graphs
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Apply display style as selected from the following labels:
AXDV
Axis division (tick) marks (defaults to KEY = ON).
AXNM
Axis scale numbers (defaults to KEY = ON, which puts numbers at the back plane of the graph). If
KEY = FRONT, numbers are on the front plane of the graph.
779
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GROPT
AXNSC
Axis number size scale factor. Input the scale value for KEY (defaults to 1.0).
ASCAL
Automatic scaling of additional Y-axes for multi-curve [/GRTYP, 2 or 3] graphs (defaults to KEY =
ON). If KEY = OFF, use base Y-axis scaling (see the /YRANGE command).
LOGX
Log X scale (defaults to KEY = OFF (linear)).
LOGY
Log Y scale (applies only to the base Y axis) (defaults to KEY = OFF (linear)).
FILL
Color fill areas under curves (defaults to KEY = OFF).
CGRID
Superimpose background grid [/GRID] over areas under filled curves (defaults to KEY = OFF).
DIG1
Number of significant digits before decimal point for axis values. Input the value for KEY (defaults
to 4).
DIG2
Number of significant digits after decimal point for axis values. Input the value for KEY (defaults to
3).
VIEW
View key for graph displays (defaults to KEY = OFF, in which case the view is (0,0,1) for 2-D graph
displays or (1,2,3) for 3-D graph displays). If KEY = ON, the view settings for graph displays are the
same as the view settings for the model.
REVX
Plots the values on the X-axis in reverse order.
REVY
Plots the values on the Y-axis in reverse order.
DIVX
Determines the number of divisions (grid markers) that will be plotted on the X axis.
DIVY
Determines the number of divisions (grid markers) that will be plotted on the Y axis.
LTYP
Determines whether ANSYS generated (KEY = 1) or system derived (KEY = 0) fonts will be used for
the axis labels.
CURL
Determines the position of the curve labels. If (KEY = 1), the curve label will be plotted in the legend
column, and the label will be displayed in the same color as the curve. If (KEY = 0) the curve labels
will be plotted near the curve. (default).
XAXO
When you use this label, the subsequent KEY value will determine an offset amount from the default
(along the bottom) location for the X axis. If KEY = 1.0, a full offset occurs (the X axis is moved to
the top of the graph). If KEY = 0.5, the axis is offset to the midpoint of the graph, and if KEY = 0
the axis remains in the original position, along the bottom of the graph. For any offset, a grey copy
of the original axis (containing the axis numbering) remains at the original location.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
780
/GROPT
YAXO
When you use this label, the subsequent KEY value will determine an offset amount from the default
(along the left side of the graph) location for the Y axis. If KEY = 1.0, a full offset occurs (the Y axis
is moved to the right side of the graph). If KEY = 0.5, the axis is offset to the midpoint of the graph,
and if KEY = 0 the axis remains in the original position, along the left side of the graph. For any
offset, a gray copy of the original axis (containing the axis numbering) remains at the original location.
KEY
Option values:
OFF (0)
Do not apply selected style.
ON (1)
Apply selected style.
nnnn
If Lab is DIG1 or DIG2, input the number of digits.
nn
If Lab is AXNSC, input the scale factor.
FRONT
If Lab is AXNM, FRONT may also be input.
Ndiv
If Lab is DIVX or DIVY, determines the number of divisions (1-99) that will be applied to the axis.
Kfont
If Lab is LTYP, Kfont is ON (1) or OFF(0). ON will use ANSYS generated fonts for the axis labels, while
OFF will use SYSTEM (Windows, X-system, etc.) fonts. The default value is ON (ANSYS fonts).
Notes
Sets various line graph display options. Issue /GROPT,STAT to display the current settings. Issue /GROPT,DEFA
to reset the default specifications. ANSYS informs you that graph view manipulation is inactive unless you
have issued the /GROPT,VIEW,ON command. See the /AXLAB, /GRTYP, /GRID, and /GTHK commands for
other graph control options.
Automatic scaling using the /XRANGE and /YRANGE commands will often yield inappropriate range values
for logarithmic scales (/GROPT, LOGX or /GROPT, LOGY).
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Axes
Utility Menu>PlotCtrls>Style>Graphs>Modify Curve
Utility Menu>PlotCtrls>Style>Graphs>Modify Grid
781
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GROPT
GRP, SIGNIF, Label
Specifies the grouping mode combination method.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SIGNIF
Combine only those modes whose significance level exceeds the SIGNIF threshold. For single point,
multipoint, or DDAM response (SPOPT,SPRS, MPRS or DDAM), the significance level of a mode is defined
as the mode coefficient of the mode, divided by the maximum mode coefficient of all modes. Any mode
whose significance level is less than SIGNIF is considered insignificant and is not contributed to the
mode combinations. The higher the SIGNIF threshold, the fewer the number of modes combined.
SIGNIF defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0. (This mode combination
method is not valid for SPOPT,PSD.)
Label
Label identifying the combined mode solution output.
DISP
Displacement solution (default). Displacements, stresses, forces, etc., are available.
VELO
Velocity solution. Velocities, "stress velocities," "force velocities," etc., are available.
ACEL
Acceleration solution. Accelerations, "stress accelerations," "force accelerations," etc., are available.
Notes
The SIGNIF value set with this command (including the default value of 0.001) overrides the SIGNIF value
set with the MXPAND command.
This command is also valid for PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Mode Combine
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
782
GRP
/GRTYP, KAXIS
Selects single or multiple Y-axes graph displays.
GRAPHICS: Graphs
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KAXIS
Axis selection key:
0 or 1
Single Y-axis. Up to 10 curves scaled to a single Y-axis.
2
Additional Y-axes (one for each curve) (3 curves maximum). Allows better scaling of curves with
widely differing numbering ranges.
3
Same as 2 but with additional Y-axis and curves projected out of the plane (6 curves maximum). Allows
clearer display with an isometric view. The default view when KAXIS = 3 is View,1,1,2,3.
Command Default
Single Y-axis graph (except as noted for selection key 3).
Notes
The basic line graph has one or more curves plotted against the same Y and X axes. Multiple curve graphs
can also be plotted with individual Y axes and the same X axis. The Y axis of the first curve is referred to as
the base Y-axis and the Y axes of the other curves as additional Y axes. Curves are numbered sequentially
from 1 (the base curve) in the order in which they are displayed. See the /AXLAB, /GROPT, /GRID, and
/GTHK commands for other display options.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Axes
/GSAVE, Fname, Ext, --
Saves graphics settings to a file for later use.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
783
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GSAVE
Ext
Filename extension (8 character maximum).
The extension defaults to GSAV if Fname is blank.
--
Unused field.
Notes
This command does not save all graphics settings, but only those that may be reset by the /RESET command.
The database remains untouched. Use the /GRESUME command to read the file. Repeated use of the /GSAVE
command overwrites the previous data on the file. The following commands are saved by /GSAVE:
/VSCALE /PSYMB /GTHK /DSCALE /ANGLE
/VUP /RATIO /GTYPE /EDGE /AXLAB
/WINDOW /SHRINK /LIGHT /ESHAPE /CLABEL
/XRANGE /SSCALE /NORMAL /FOCUS /COLOR
/YRANGE /TRIAD /NUMBER /GCMD /CONTOUR
/TRLCY /PBC /GLINE /CPLANE
/TYPE /PLOPTS /GRID /CTYPE
/VCONE /PNUM /GROPT /CVAL
/VIEW /PSF /GRTYP /DIST
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Save Plot Ctrls
GSBDATA, LabZ, VALUEZ, LabX, VALUEX, LabY, VALUEY
Specifies the constraints or applies the load at the ending point for generalized plane strain option.
SOLUTION: FE Constraints
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
LabZ
Constraint or load at the ending point in the fiber Z direction.
F
Apply a force in the fiber direction (default).
LFIBER
Define a length change in the fiber direction.
VALUEZ
Value for LabZ. The default is zero.
LabX
Constraint or load on rotation about X.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
784
GSBDATA
MX
Supply a moment to cause the rotation of the ending plane about X (default).
ROTX
Define a rotation angle (in radians) of the ending plane about X.
VALUEX
Value for LabX. The default is zero.
LabY
Constraint or load on rotation about Y
MY
Supply a moment to cause the rotation of the ending plane about Y (default).
ROTY
Define a rotation angle (in radians) of the ending plane about Y.
VALUEY
Value for LabY. The default is zero.
Notes
All inputs are in the global Cartesian coordinate system. For more information about the generalized plane
strain feature, see Generalized Plane Strain Option of Current-Technology Solid Elements in the Element
Reference.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Gen Plane Strain
Main Menu>Solution>Define Loads>Apply>Structural>Gen Plane Strain
GSGDATA, LFIBER, XREF, YREF, ROTX0, ROTY0
Specifies the reference point and defines the geometry in the fiber direction for the generalized plane
strain element option.
PREP7: Meshing
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
LFIBER
Fiber length from the reference point. Defaults to 1.
XREF
X coordinate of the reference point. Defaults to zero.
YREF
Y coordinate of the reference point. Defaults to zero.
ROTX0
Rotation of the ending plane about X in radians Defaults to zero.
ROTY0
Rotation of the ending plane about Y in radians Defaults to zero.
785
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GSGDATA
Notes
The ending point is automatically determined from the starting (reference) point and the geometry inputs.
All inputs are in the global Cartesian coordinate system. For more information about the generalized plane
strain feature, see Generalized Plane Strain Option of Current-Technology Solid Elements in the Element
Reference.
Menu Paths
Main Menu>Preprocessor>Modeling>Genl plane strn
GSLIST, Lab
When using generalized plane strain, lists the input data or solutions.
SOLUTION: FE Constraints
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Lab
Specify the content to be listed.
GEOMETRY
List the data input using GSGDATA
BC
List the data input using GSBDATA.
REACTIONS
When the command is issued in POST1, list the reaction force at the ending point,
and the moment about X and Y if the corresponding constraints were applied.
RESULTS
When the command is issued in POST1, list the change of fiber length at the ending point during
deformation and the rotation of the ending plane about X and Y during deformation.
ALL
List all of the above (default).
Notes
This command can be used to list the initial position of the ending plane, the applied load or displacements
in the fiber direction, the resulting position of the ending plane after deformation, and the available reaction
forces and moments at the ending point.
All inputs and outputs are in the global Cartesian coordinate system. For more information about the gen-
eralized plane strain feature, see Generalized Plane Strain Option of Current-Technology Solid Elements in
the Element Reference.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Other>Genl Plane Strn
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
786
GSLIST
GSSOL, NVAR, Item, Comp, Name
Specifies which results to store from the results file when using generalized plane strain.
POST26: Set Up
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
NVAR
Arbitrary reference number or name assigned to this variable. Variable numbers can be 2 to NV (NUMVAR)
while the name can be an eight byte character string. Overwrites any existing results for this variable.
Item
Label identifying item to be stored.
LENGTH
Change of fiber length at the ending point.
ROT
Rotation of the ending plane during deformation.
F
Reaction force at the ending point in the fiber direction.
M
Reaction moment applied on the ending plane.
Comp
Component of the item, if Item = ROT or M.
X
The rotation angle or reaction moment of the ending plane about X.
Y
The rotation angle or reaction moment of the ending plane about Y.
Name
Thirty-two character name identifying the item on the printout and display. Defaults to the label formed
by concatenating the first four characters of the Item and Comp labels.
Notes
This command stores the results (new position of the ending plane after deformation) for generalized plane
strain. All outputs are in the global Cartesian coordinate system. For more information about the generalized
plane strain feature, see Generalized Plane Strain Option of Current-Technology Solid Elements in the Element
Reference.
Menu Paths
Main Menu>TimeHist Postpro>Variable Viewer
787
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
GSSOL
/GST, Lab, Lab2
Turns Graphical Solution Tracking (GST) on or off.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Lab
Determines whether the Graphical Solution Tracking feature is active. Specify ON to activate GST, or OFF
to deactivate the feature.
Lab2
Activates generation of interface and field convergence files (ANSYS MFX analyses only).
Notes
For interactive runs using the GUI [/MENU,ON] or graphics [/MENU,GRPH] mode, ANSYS directs GST graphics
to the screen. For interactive sessions not using GUI or graphics mode, or for batch sessions, GST graphics
are saved in the ANSYS graphics file Jobname.GST. The file Jobname.GST can be viewed with the DISPLAY
program. You must select All File Types to access it. For more information on the DISPLAY program see
Getting Started with the DISPLAY Program in the Basic Analysis Guide.
The GST feature is available only for nonlinear structural, thermal, electric, magnetic, fluid, or CFD simulations.
For more information about this feature and illustrations of the GST graphics for each analysis type, see the
ANSYS Analysis Guide for the appropriate discipline. See also the CNVTOL command description.
The GST feature is also available for a p-method electrostatic analysis. For interactive sessions with the p-
method electrostatic preference set, ANSYS directs the GST graphics to the screen.
When running an ANSYS MFX analysis, specify /GST,ON,ON to generate both the interface (Jobname.NLH)
and field convergence (Fieldname.GST) files for monitoring the analysis. This field is not available on the
GUI.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Output Ctrls>Grph Solu Track
Main Menu>Solution>Load Step Opts>Output Ctrls>Grph Solu Track
GSUM
Calculates and prints geometry items.
PREP7: Keypoints
PREP7: Lines
PREP7: Areas
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Calculates and prints geometry items (centroid location, moments of inertia, length, area, volume etc.) asso-
ciated with the selected keypoints, lines, areas, and volumes. Geometry items are reported in the global
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
788
/GST
Cartesian coordinate system. For volumes, a unit density is assumed unless the volumes have a material as-
sociation via the VATT command. For areas, a unit density (and thickness) is assumed unless the areas have
a material (and real constant) association via the AATT command. For lines and keypoints, a unit density is
assumed, irrespective of any material associations [LATT, KATT, MAT]. Items calculated by GSUM and later
retrieved by a *GET or *VGET commands are valid only if the model is not modified after the GSUM command
is issued. This command combines the functions of the KSUM, LSUM, ASUM, and VSUM commands.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Calc Geom Items>Of Geometry
/GTHK, Label, THICK
Sets line thicknesses for graph lines.
GRAPHICS: Graphs
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Apply thicknesses as selected from the following labels:
AXIS
Modify thickness of ordinate and abscissa axes on graph displays.
GRID
Modify thickness of grid lines on graph displays.
CURVE
Modify thickness of curve lines (when no area fill [/GROPT]).
THICK
Thickness ratio (whole numbers only, from -1 to 10):
1
Do not draw the curve, but show only the markers specified by /GMARKER.
0 or 1
Thin lines.
2
The default thickness.
3
1.5 times the default thickness.
etc.
(up to 10)
Notes
Sets line thicknesses for graph lines (in raster mode only). Use /GTHK,STAT to show settings.
This command is valid in any processor.
789
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GTHK
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Axes
Utility Menu>PlotCtrls>Style>Graphs>Modify Curve
Utility Menu>PlotCtrls>Style>Graphs>Modify Grid
/GTYPE, WN, LABEL, KEY
Controls the entities that the GPLOT command displays.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which this command applies (defaults to 1)
LABEL
This represents the type of entity to display:
NODE
Nodes
ELEM
Elements
KEYP
Keypoints
LINE
Lines
AREA
Areas
VOLU
Volumes
GRPH
Graph displays
KEY
Switch:
0
Turns the entity type off.
1
Turns the entity type on.
Notes
The /GTYPE command controls which entities the GPLOT command displays. NODE, ELEM, KEYP, LINE, AREA,
and VOLU are on by default. When ELEM is activated, you can control the type of element displayed via the
/GCMD command (which also controls the type of graph display). When the GRPH entity type is activated,
all other entity types are deactivated. Conversely, when any of the NODE, ELEM, KEYP, LINE, AREA, and VOLU
entity types are active, the GRPH entity type is deactivated.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
790
/GTYPE
The /GTYPE command gives you several options for multi-window layout:
One window
Two windows (left and right or top and bottom of the screen)
Three windows (two at the top and one at the bottom of the screen, or one top and two bottom windows
Four windows (two at the top and two at the bottom)
Once you choose a window layout, you can choose one of the following: multiple plots, replotting, or no
redisplay.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Multi-Plot Contrls
791
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/GTYPE
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
792
H Commands
HARFRQ, FREQB, FREQE
Defines the frequency range in the harmonic response analysis.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
FREQB
Frequency (Hz) at the beginning of the FREQB to FREQE range (if FREQE > FREQB). If FREQE is blank,
the solution is done only at frequency FREQB.
FREQE
Frequency at end of this range. Solutions are done at an interval of (FREQE-FREQB)/NSBSTP, ending
at FREQE. No solution is done at the beginning of the frequency range. NSBSTP is input on the NSUBST
command. See EXPSOL for expansion pass solutions.
Notes
Defines the frequency range for loads in the harmonic response analysis (ANTYPE,HARMIC).
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Freq and Substps
Main Menu>Solution>Load Step Opts>Time/Frequenc>Freq and Substps
/HBC, WN, Key
Determines how boundary condition symbols are displayed in a display window.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window reference number. This number can be any window numbered 1 to 5, or ALL (for all active
windows). Defaults to 1
Key
Key to enable/disable hidden surface boundary condition symbol display for 2-D graphics devices and
to request improved pressure contour display for 2-D and 3-D devices: Key = ON, YES or 1 will show
your BC symbols on the hidden surfaces and use an improved pressure contour display. Key = OFF, NO
or 0 (default) will hide the symbols .
793
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Command Default
/HBC,WN,Off
For 2-D graphics devices (X11, win32, PNG, et al.), boundary condition symbols are NOT drawn to the hidden
surface display. For both 2-D and 3-D graphics devices, no enhanced rendering of pressure contours occurs.
Menu Paths
Utility Menu>PlotCtrls>Style>Hidden-Line Options
HBMAT, Fname, Ext, --, Form, Matrx, Rhs, Mapping
Writes an assembled global matrix in Harwell-Boeing format.
AUX2: Binary Files
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Defaults to the current Jobname if left blank.
Ext
Filename extension (8 character maximum).
Defaults to .MATRIX if left blank.
--
Unused field.
Form
Specifies format of output matrix file:
ASCII
Write output matrix file in ASCII form.
BIN
Write output matrix file in binary form.
Matrx
Specify which matrix to write to the output matrix file:
STIFF
Write stiffness matrix to output matrix file. Valid for all types of analyses that write a .FULL file.
MASS
Write mass matrix to output matrix file. Valid for buckling, substructure, and modal analyses. If .FULL
file was generated in a buckling analysis, then this label will write stress stiffening matrix to output
matrix file.
DAMP
Write damping matrix to output matrix file. Only valid for damped modal analyses.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
794
HBMAT
Rhs
Specifies whether to write the right-hand side vector to output matrix file:
YES
Write right-hand side vector to output matrix file
NO
Do not write right-hand side vector to output matrix file
Mapping
Specifies whether to write the mapping file. This file is always named Fname.MAPPING.
YES
Write the mapping file
NO
Do not write the mapping file (default)
Command Default
By default, assuming a proper filename and extension have been entered, writes the stiffness matrix and
right-hand side vector to jobname.MATRIX in the current working directory in ASCII format.
Notes
This command is used to copy a matrix from the assembled global matrix file (.FULL file) or from the su-
perelement matrix file (.SUB file) as specified on the FILE command and write it in Harwell-Boeing format
to a new file, named jobname.MATRIX. The Harwell-Boeing format is widely used by other applications
that deal with matrices.
The assembled global matrix file is created during solution depending on the analysis type, equation solver,
and other solution options. By default, the assembled global matrix file is never deleted at the end of solution.
For most analysis types, the Sparse direct solver, the ICCG solver, and the AMG solver (when available) will
write a .FULL file. With the exception of the reduced (MODOPT,REDUC) and Variational Technology
(MODOPT,VT) options, all mode extraction methods used for buckling and modal analyses will write a
properly formatted .FULL file to be used with the HBMAT command.
When dumping the stiffness matrix for transient and harmonic analyses, be aware that the element mass
matrix values (and possibly element damping matrix values) are incorporated into the globally assembled
stiffness matrix. Thus, the globally assembled stiffness matrix represents more than the stiffness of the
model for these analysis types. Please refer to the Theory Reference for the Mechanical APDL and Mechanical
Applications for more details.
When dumping a .FULL file, the rows and columns corresponding to specified constraints (e.g., D commands)
are eliminated from the system of equations and therefore are not written to the .MATRIX file. Also, rows
and columns corresponding to eliminated (slave) degrees of freedom from coupling and/or constraint
equations (e.g., CE, CP commands) are also eliminated from the system of equations and are not written to
the .MATRIX file. The DOFs that are eliminated from any coupling and/or constraint equations are determined
internally by the solution code and may not match what you specified via the CE/CP (or similar) commands.
When dumping a .SUB file, the full nxn matrix will be written to the .MATRIX file for either symmetric or
unsymmetric matrices, regardless of whether any of the matrix coefficients are zero-valued. When dumping
a .FULL file, only the lower triangular part of the matrix will be written to the .MATRIX file if the matrix
is symmetric; the full matrix is written if the matrix is unsymmetric. Only matrix coefficients that are greater
than zero will be written.
795
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HBMAT
The Harwell-Boeing format is column-oriented. That is, non-zero matrix values are stored with their corres-
ponding row indices in a sequence of columns. However, since the ANSYS matrix files are stored by row
and not column, when the HBMAT command is used with a non-symmetric matrix, the transpose of the
matrix is, in fact, written.
The WRFULL command, in conjunction with the SOLVE command, can be used to generate the assembled
global matrix file and eliminate the equation solution process and results output process.
The mapping file can be used to map the matrix equation numbers found in the .MATRIX file directly to
the corresponding node numbers and degrees of freedom.
When dumping a CMS .SUB file, the last rows/columns of the matrix are non-physical degrees of freedom
added internally by the CMS process and cannot be mapped directly to a node number or particular degree
of freedom.
Menu Paths
Utility Menu>File>List>Binary Files
Utility Menu>List>Files>Binary Files
/HEADER, Header, Stitle, Idstmp, Notes, Colhed, Minmax
Sets page and table heading print controls.
POST1: Listing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Header
ANSYS page header (system, date, time, version, copyright, title, etc.):
ON
Turns this item on (default for batch mode; not available for interactive mode).
OFF
Turns this item off.
(blank)
Retains the previous setting.
Stitle
Subtitles (see /STITLE command): ON, OFF, or (blank) (see above).
Idstmp
Load step information (step number, substep number, time value): ON, OFF, or (blank) (see above).
Notes
Information relative to particular table listings: ON, OFF, or (blank) (see above).
Colhed
Column header labels of table listings (currently only for single column tables): ON, OFF, or (blank) (see
above).
Minmax
Minimum/maximum information or totals after table listings: ON, OFF, or (blank) (see above).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
796
/HEADER
Command Default
All specifications are on (batch mode); Header = OFF, all other specifications are on (interactive mode)
Notes
Sets specifications on or off for page and table heading print controls associated with the POST1 PRNSOL,
PRESOL, PRETAB, PRRSOL, and PRPATH commands.
Note
If the printout caused a top-of-form (page eject to top of next page), the top-of-form is also
suppressed with the printout. Issue /HEADER,STAT to display the current settings. Issue /HEAD-
ER,DEFA to reset the default specifications.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
HELP, Name
Displays help information on ANSYS commands and element types.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Command name or element type. Examples are: HELP,MP or HELP,SOLID45 (or HELP,45). For a list of
elements of a particular type, enter HELP,BEAM, HELP,SOLID, HELP,HYPER, etc.
Notes
If Name uniquely matches a command or element name, the description for that command or element will
be displayed in the Help Window. For command help, you must type the complete command name (including
the * or /). The help system cannot find partial matches. If * is used at the beginning of the string, it will be
interpreted as an ANSYS * command.
For help on topics that are not ANSYS commands or element types (for example, help for the word "mater-
ial"), use the index or full text search features of the ANSYS online help system.
The HELP command is valid only in GUI mode. To obtain help when not in GUI mode, you can either activate
the GUI by typing /MENU,ON, or you can activate the help system directly by issuing /UI,HELP.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
797
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HELP
HELPDISP, Commandname
Displays help information on DISPLAY program commands.
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Commandname
Any DISPLAY command. If blank, a list of DISPLAY commands is produced.
Menu Paths
It is part of the DISPLAY command.
HEMIOPT, HRES
Specifies options for Hemicube view factor calculation.
SOLUTION: Radiosity
AUX12: Radiosity Solver
MP ME <> PR PRN <> <> <> <> <> <> PP <> EME MFS
HRES
Hemicube resolution. Increase value to increase the accuracy of the view factor calculation. Defaults to
10.
Menu Paths
Main Menu>Preprocessor>Radiation Opts>View Factor
Main Menu>Radiation Opt>Radiosity Meth>View Factor
Main Menu>Solution>Radiation Opts>View Factor
HFADP, Lab
Turns a high-frequency adaptive error calculation on or off.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
Adaptive error label:
ON
Turns on adaptive error calculation.
OFF
Turns off adaptive error calculation.
Notes
Adapative error calculation must be turned on to automatically refine the mesh with the HFEREFINE macro.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
798
HELPDISP
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
HFANG, Lab, PHI1, PHI2, THETA1, THETA2
Defines or displays spatial angles of a spherical radiation surface for antenna parameter calculations.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
Spatial angle label.
ANGLE
Define spatial angles (default).
STATE
Display spatial angles. PHI1, PHI2, THETA1, and THETA2 are ignored.
PHI1, PHI2
Starting and ending angles (degrees) in the spherical coordinate system. Defaults to 0.
THETA1, THETA2
Starting and ending angles (degrees) in the spherical coordinate system. Defaults to 0.
Notes
Defines or displays spatial angles of a spherical radiation surface. See Spherical Coordinates in the High-
Frequency Electromagnetic Analysis Guide. Use this command only with PLFAR,Lab = ANT or PRFAR,Lab =
ANT.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>List Results>Field Extension>Direct Gain
Main Menu>General Postproc>List Results>Field Extension>Efficiency
Main Menu>General Postproc>List Results>Field Extension>Power Gain
Main Menu>General Postproc>List Results>Field Extension>Rad Power
Main Menu>General Postproc>Plot Results>Field Extension>Direct Gain
799
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFANG
HFARRAY, NUMX, NUMY, PX, PY, SKEW, PHASEX, PHASEY
Defines phased array antenna characteristics.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
NUMX
Number of array units in X-direction (defaults to 1).
NUMY
Number of array units in Y-direction (defaults to 1).
PX
Spatial periodicity in X-direction in meters.
PY
Spatial periodicity in Y-direction in meters.
SKEW
Skew angle for triangular periodicity from X-axis to Y-axis in degrees (defaults to 90).
PHASEX
Initial phase angle difference between array units in X-direction in degrees (defaults to 0).
PHASEY
Initial phase angle difference between array units in Y-direction in degrees (defaults to 0).
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>List Results>Field Extension>Direct Gain
Main Menu>General Postproc>List Results>Field Extension>Far Field
Main Menu>General Postproc>List Results>Field Extension>Pattern
Main Menu>General Postproc>Plot Results>Field Extension>Direct Gain
Main Menu>General Postproc>Plot Results>Field Extension>Far Field
Main Menu>General Postproc>Plot Results>Field Extension>Pattern
HFDEEM, Filename, Snp, PORTNUM1, L1, Filename1, Ext1, PORTNUM2, L2, Filename2, Ext2
Calibrates S-parameter phase shift.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Filename
A Touchstone file containing the S-parameters of a n-port system.
Snp
Extension of the n-port Touchstone file.
PORTNUM1
Port number of excitation port. Defaults to 1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
800
HFARRAY
L1
Distance from the S-parameter extraction plane to the reference plane at PORTNUM1 in meters. Defaults
to 0.
Filename1
A two-column Touchstone format file containing frequencies and propagating constants.
Ext1
Extension of the Filename1 file.
PORTNUM2
Port number of output port. Defaults to 1.
L2
Distance from the S-parameter extraction plane to the reference plane at PORTNUM2 in meters. Defaults
to 0.
Filename2
A two-column Touchstone format file containing frequencies and propagating constants.
Ext2
Extension of the Filename2 file.
Notes
The first line of the Touchstone format files Filename1 and Filename2 is a general description of the output
data and it starts with an exclamation point (!). The second line starts with a # and it is followed by a frequency
unit of GHz, MHz, KHz, or Hz (for example, #GHz). The first column of the two-column format file is the fre-
quency. The second column is the propagating constant corresponding to the working frequency.
Menu Paths
This command cannot be accessed from a menu.
HFEIGOPT, Lab, Val1
Specifies high frequency electromagnetic modal analysis options.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
High frequency modal analysis type:
CAVITY
Perform a 3-D eigenvalue analysis (default for HF119 and HF120).
CUTOFF
Perform a 2-D cutoff frequency analysis (default for HF118).
GAMMA
Perform a 2-D propagating constant analysis.
Val1
Valid for the GAMMA option only. If Lab = GAMMA, Val1 is the frequency in Hz.
801
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFEIGOPT
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
HFEREFINE, FACTOR, NUMLIST
Automatically refines high-frequency tetrahedral elements (HF119) or lists high-frequency brick elements
(HF120) with the largest error.
PREP7: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FACTOR
Percentage applied in adaptive mesh refinement scheme. This percentage of the elements with local
error greater that the average error will be refined. FACTOR should be in the range of 0 to 100 percent
and it defaults to 10 percent. Used only with HF119 elements.
NUMLIST
Number of brick elements to be listed. Defaults to 25. Used only with HF120 elements.
Notes
For HF119 elements, HFEREFINE automatically refines the mesh so that the measured discretization error
will decrease. Execution deletes all boundary conditions and excitation sources on the finite element model.
For HF120 elements, HFEREFINE provides a list of elements with the largest error.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>HF Refine
HFMODPRT, FREQ
Calculates electromagnetic field distribution for a modal port.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FREQ
Working frequency (Hz). No default.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
802
HFEREFINE
Notes
This command automatically generates a port electromagnetic field by solving the 2-D eigenvalue problem
and stores the solution as 3-D harmonic excitation and matching condition.
This command must be issued in SOLUTION when a modal port is defined and the SPSWP command is not
used.
Menu Paths
This command cannot be accessed from a menu.
HFPA, Lab, Local, VAL1, VAL2
Specifies a radiation scan angle for a phased array antenna analysis.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
Enter SCAN in this field.
Local
Local coordinate system number (defaults to 0).
VAL1
Angle from x-axis towards y-axis, , in degrees (defaults to 0).
VAL2
Angle from +z-axis towards -z-axis, , in degrees (defaults to 0).
Notes
See Spherical Coordinates in the High-Frequency Electromagnetic Analysis Guide for an illustration of the co-
ordinate system.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
803
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFPA
HFPCSWP, FREQB, FREQE, FREQINC, NUMMODE, IMPDMULT, ZOPT
Calculates propagating constants and characteristic impedance of a transmission line or waveguide over
a frequency range.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FREQB
Beginning frequency of the analysis (in Hertz).
FREQE
Ending frequency of the analysis (in Hertz). If blank, FREQE is set to FREQB and one solution is performed.
(This is recommended for initially verifying the model setup.)
FREQINC
Frequency increment between solutions. If FREQINC is blank, solutions are performed at FREQB and
FREQE only (default).
NUMMODE
Number of required modes. Defaults to 1.
IMPDMULT
Impedance multiplier (defaults to 1). For a half symmetry model, use 2.
ZOPT
Impedance calculation inputs:
0
Power and current (PI) definition (default).
1
Power and voltage (PV) definition.
2
Voltage and current (VI) definition.
Notes
HFPCSWP calculates propagating constants and characteristic impedance for a transmission line or waveguide
over a frequency range from FREQB to FREQE, in increments of FREQINC. The output data is written to
the file HFPCSWP.OUT. Use the PLTLINE command to plot the results.
HFPCSWP can only be used with HF118 elements.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Solve>Electromagnet>HF Emag>2D Freq Sweep
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
804
HFPCSWP
HFPOWER, ARG1, ARG2
Calculates power terms of a multi-port network.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
ARG1
Excitation source port number.
ARG2
Matched output port number.
Notes
HFPOWER calculates the input power, reflected power, return loss, and power reflection coefficient for a
driven port. If a matching output port is defined, it also calculates transmitted power, insertion loss and the
power transmission coefficient. For lossy materials and conducting surfaces, HFPOWER also calculates the
time-averaged dissipated power. The elements must be selected to calculate the dissipated power.
If ARG1 and ARG2 are both blank, no ports are specified and only the dissipated power is calculated.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Element Based>Power
HFPORT, Portnum, Porttype, Local, Opt1, Opt2, VAL1, VAL2, VAL3, VAL4, VAL5, VAL6, VAL7, VAL8, VAL9, VAL10,
VAL11
Specifies input data for waveguide, plane wave, lumped gap, modal, and modal lumped gap ports.
SOLUTION: Misc Loads
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Portnum
Port number. You specify the integer number for exterior and interior ports by the SF and BF family of
commands, respectively. The number must be between 1 and 50. If Portnum = STATUS, provides the
status of port option settings.
Porttype
Port type:
COAX
Cylindrical coaxial waveguide.
RECT
Rectangular waveguide.
CIRC
Circular waveguide.
PARA
Parallel plate waveguide.
805
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFPORT
PLAN
Incident plane wave.
LGAP
Lumped gap port.
MODAL
Modal port.
MGAP
Modal lumped gap port.
Local
A previously defined local Cartesian coordinate system number (>10) or 0 (global Cartesian) used to
specify the geometric properties of the waveguide. Defaults to 0. The local Z-direction must be the dir-
ection of wave propagation. The origin of the local coordinate system must be centered about the face
of the waveguide port without considering symmetry. A local Cartesian coordinate system number is
not necessary for Porttype = MODAL and MGAP.
Opt1
For Porttype = COAX, RECT, CIRC, PARA, and LGAP, Opt1 defines the mode type:
TEM
Transverse electromagnetic wave. See note below.
TEmn
Transverse electric wave. See note below.
TMmn
Transverse magnetic wave. See note below.
Note
For Porttype = COAX, only the TEM mode is available. For Porttype = LGAP, Opt1 is not
used.
For Porttype = PLAN, Opt1 defines the extracted harmonic component:
ALL
No specified extraction (default).
TE
Transverse electric wave.
TM
Transverse magnetic wave.
For Porttype = MODAL, Opt1 is the required number of modes. Defaults to 1.
For Porttype = MGAP, Opt1 is equal to 1 and defaults to 1.
Opt2
EXT
Exterior port.
INT
Interior port.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
806
HFPORT
For Porttype = PLAN, LGAP, and MGAP, only Opt2 = INT is available.
VAL1, VAL2, VAL3, . . . , VAL11
If Porttype = COAX:
VAL1
Inner radius of the coaxial waveguide.
VAL2
Outer radius of the coaxial waveguide.
VAL3
Zero-to-peak amplitude of the voltage between the inner and the outer conductors. If blank, the
port will create a matching impedance.
VAL4
Phase angle of the applied voltage (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the applied voltage input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7-VAL11
Not used.
If Porttype = RECT:
VAL1
Width of the rectangular waveguide.
VAL2
Height of the rectangular waveguide.
VAL3
Zero-to-peak amplitude of the electric field component E
z
for a TM wave or the magnetic field
component H
z
for a TE wave. If blank, the port will appear as a matching impedance.
VAL4
Phase angle of the applied field (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the field component input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7-VAL11
Not used.
If Porttype = CIRC
VAL1
Radius of the circular waveguide.
VAL2
Not used.
VAL3
Zero-to-peak amplitude of the electric field component E
z
for a TM wave or the magnetic field
component H
z
for a TE wave. If blank, the port will appear as a matching impedance.
807
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFPORT
VAL4
Phase angle of the applied field (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the field component input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7-VAL11
Not used.
If Porttype = PARA:
VAL1
Width of the parallel plate waveguide (defaults to 1).
VAL2
Separation between the two plates.
VAL3
Zero-to-peak amplitude of the electric field component E
y
for a TEM wave, electric field component
E
z
for a TM wave, or the magnetic field component H
z
for a TE wave. If blank, the port will appear
as a matching impedance.
VAL4
Phase angle of the applied field (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the field component input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7-VAL11
Not used.
If Porttype = PLAN:
VAL1
X Component of incident planar wave (V/m) in local coordinate system. Defaults to 0.
VAL2
Y Component of incident planar wave (V/m) in local coordinate system. Defaults to 0.
VAL3
Z Component of incident planar wave (V/m) in local coordinate system. Defaults to 0.
VAL4
Angle between incident wave vector and X-axis ().
VAL5
Angle between incident wave vector and Z-axis ().
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7
Extracted wave number in X direction (m in TEmn or TMmn). Defaults to 0.
VAL8
Extracted wave number in Y direction (n in TEmn or TMmn). Defaults to 0.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
808
HFPORT
VAL9
Spatial periodicity in X direction in meters. It must be defined if the m value specified by VAL7 is
not equal to zero. Defaults to 0.
VAL10
Spatial periodicity in Y direction in meters. It must be defined if the n value specified by VAL8 is not
equal to zero. Defaults to 0.
VAL11
Skew angle for triangular periodicity from X-axis to Y-axis in degrees. Defaults to 90.
See Spherical Coordinates in the High-Frequency Electromagnetic Analysis Guide for an illustration of the
coordinate system.
Note
VAL1, VAL2, and VAL3 = 0 for Porttype = PLAN and Opt2 = INT.
If Porttype = LGAP:
VAL1
Width of the lumped gap port.
VAL2
Separation between the central conductor and ground.
VAL3
Zero-to-peak amplitude of the voltage between the central conductor and ground. Defaults to 0.
VAL4
Phase angle of the applied voltage (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the field component input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7
Not used.
VAL8
Characteristic impendance of lumped gap port. Defaults to 50 ohm.
VAL9-VAL11
Not used.
If Porttype = MODAL:
VAL1
Impedance multiplier (defaults to 1). For a half symmetry model, use 2.
VAL2
Impedance calculation inputs:
PI
Power and current (PI) definition (default).
809
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFPORT
PV
Power and voltage (PV) definition.
VI
Voltage and current (VC) definition.
VAL3
Zero-to-peak amplitude of the electric field. Defaults to 0.
VAL4
Phase angle of the applied electric field (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the field component input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7-VAL11
Not used.
If Porttype = MGAP:
VAL1
Not used.
VAL2
Not used.
VAL3
Zero-to-peak amplitude of the voltage between the central conductor and ground. Defaults to 0.
VAL4
Phase angle of the applied voltage (in degrees). Defaults to 0 degrees.
VAL5
Input power (time-average). If power is input, it overrides the field component input.
VAL6
Distance from S-parameter extraction plane to reference plane. Defaults to 0.
VAL7
Not used.
VAL8
Characteristic impendance of lumped gap port. Defaults to 50 ohm.
VAL9-VAL11
Not used.
Notes
The origin of the local coordinate system must be at the center of the waveguide structure. For a rectangular
waveguide, the X and Y axes of the local coordinate system must be parallel to the width and height of the
waveguide, respectively. For a parallel plate waveguide or a lumped gap port, the X and Y axes of the local
coordinate system must be parallel to the width and separation, respectively.
The following apply to the TEmn and TMmn mode types:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
810
HFPORT
For a rectangular waveguide, the suffix m and n mean the variation of the field along the wide side
and narrow side of the waveguide, respectively.
For a circular waveguide, the suffix m and n mean the variation of the field along the angular and radial
directions, respectively.
For a parallel plate waveguide, the suffix m is 0 and the suffix n means the variation of the field between
the plates.
For a coaxial waveguide, only the TEM mode is available.
For a lumped gap port, only the TEM mode or blank is available.
For a lumped gap port, a uniformly distributed electromagnetic field is assumed (that is, E
y
and H
x
components
exist in the local coordinate system).
For a modal lumped gap port, a voltage path connecting the central conductor and the ground must be
defined using the LPRT command.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior
Port>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior
Port>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Interior
Port>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Interior
Port>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Modify Port
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Port Status
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior Port>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior Port>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Interior Port>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Interior Port>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Modify Port
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Port Status
HFSCAT, Lab
Specifies a high-frequency scattering analysis.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
Label identifying scattering analysis options:
OFF
Do not perform a scattering analysis.
811
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HFSCAT
SCAT
Perform a scattering analysis and store the scattering field (default).
TOTAL
Perform a scattering analysis and store the total field.
Notes
Specifies a high frequency scattering analysis and the type of electromagnetic field output. HFSCAT,SCAT
provides a scattering field output, E
sc
, which is required for the calculation of Radar Cross Section (RCS).
HFSCAT,TOTAL provides a sum of the incident and scattering fields, E
total
= E
inc
+E
sc
.
Use the PLWAVE command to specify the incident field, E
inc
.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
HFSYM, KCN, Xkey, Ykey, Zkey
Indicates the presence of symmetry planes for the computation of high-frequency electromagnetic fields
in the near and far field domains (beyond the finite element region).
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
KCN
Coordinate system reference number. KCN may be 0 (Cartesian), or any previously defined local Cartesian
coordinate system number (>10). Defaults to 0.
Xkey
Key for electromagnetic field boundary condition, as prescribed for the solution, corresponding to the
x = constant plane:
None
No electric wall or magnetic wall boundary conditions (default).
PEC
Perfect electric conductor (electric wall boundary conditions).
PMC
Perfect magnetic conductor (magnetic wall boundary conditions).
Ykey
Key for electromagnetic field boundary condition, as prescribed for the solution, corresponding to the
y = constant plane:
None
No electric wall or magnetic wall boundary conditions (default).
PEC
Perfect electric conductor (electric wall boundary conditions).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
812
HFSYM
PMC
Perfect magnetic conductor (magnetic wall boundary conditions).
Zkey
Key for electromagnetic field boundary condition, as prescribed for the solution, corresponding to the
z = constant plane:
None
No electric wall or magnetic wall boundary conditions (default).
PEC
Perfect electric conductor (electric wall boundary conditions).
PMC
Perfect magnetic conductor (magnetic wall boundary conditions).
Notes
HFSYM uses the image principle to indicate symmetry planes (x, y, or z = constant plane) for high-frequency
electromagnetic field computations outside the modeled domain. A perfect magnetic conductor (PMC)
boundary condition must be indicated even though it occurs as a natural boundary condition.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>List Results>Field Extension>Direct Gain
Main Menu>General Postproc>List Results>Field Extension>Efficiency
Main Menu>General Postproc>List Results>Field Extension>Far Field
Main Menu>General Postproc>List Results>Field Extension>Near Field
Main Menu>General Postproc>List Results>Field Extension>Pattern
Main Menu>General Postproc>List Results>Field Extension>Power Gain
Main Menu>General Postproc>List Results>Field Extension>Rad Power
Main Menu>General Postproc>List Results>Field Extension>RCS
Main Menu>General Postproc>List Results>Field Extension>RCS Normalized
Main Menu>General Postproc>Path Operations>Map onto Path>HF Near Field
Main Menu>General Postproc>Plot Results>Field Extension>Direct Gain
Main Menu>General Postproc>Plot Results>Field Extension>Far Field
Main Menu>General Postproc>Plot Results>Field Extension>Pattern
Main Menu>General Postproc>Plot Results>Field Extension>RCS
Main Menu>General Postproc>Plot Results>Field Extension>RCS Normalized
HMAGSOLV, FREQ, NRAMP, CNVA, CNVV, CNVC, CNVE, NEQIT
Specifies 2-D or axisymmetric harmonic magnetic solution options and initiates the solution.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
FREQ
Analysis frequency (Hz).
813
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HMAGSOLV
NRAMP
Number of ramped substeps for the first load step of a nonlinear 2-D harmonic electromagnetic solution.
Defaults to 3. If NRAMP = -1, ignore the ramped load step entirely.
CNVA
Convergence tolerance on the program calculated reference value for the magnetic vector potential
degree of freedom. Defaults to 0.001.
CNVV
Convergence tolerance on the program calculated reference value for the time-integrated electric potential
VOLT. Defaults to 0.001.
CNVC
Convergence tolerance on the program calculated reference value for the current degree of freedom
CURR. Defaults to 0.001.
CNVE
Convergence tolerance on the program calculated reference value for the voltage drop degree of freedom
EMF. Defaults to 0.001.
NEQIT
Maximum number of equilibrium iterations per load step. Defaults to 50.
Notes
HMAGSOLV invokes an ANSYS macro which specifies harmonic electromagnetic solution options and initiates
the solution. The macro is applicable to any ANSYS 2-D or axisymmetric linear or nonlinear harmonic analysis.
Results are only stored for the final converged solution. (In POST1, issue *SET,LIST to identify the load step
of solution results.) The macro internally determines if a nonlinear analysis is required based on magnetic
material properties defined in the database.
The macro performs a two-load-step solution sequence. The first load step ramps the applied loads over a
prescribed number of substeps (NRAMP), and the second load step calculates the converged solution. For
linear problems, only a single load step solution is performed. The ramped load step can be bypassed by
setting NRAMP to -1.
A 3-D harmonic electromagnetic analysis is available for linear solutions only and does not require this
solution macro.
The following analysis options and nonlinear options are controlled by this macro: KBC, NEQIT, NSUBST,
CNVTOL, OUTRES.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Solve>Electromagnet>Harmonic Analys>Opt&Solv
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
814
HMAGSOLV
HPGL, Kywrd, Opt1, Opt2
Specifies various HP options.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
If Kywrd = MODEL, command format is HPGL,MODEL,Pmod.
Pmod
Valid plotter model: 7475A (default), 7550A, 7580B, 7585B, 7586B, COLORPRO, DRAFTPRO, or DRAFTMAS-
TER.
If Kywrd = PAPER, command format is HPGL,PAPER,Size,Orien.
Size
Valid paper size: A (default), B, C, D, E, A4, A3, A2, A1, A0, CARCH, DARCH, or EARCH.
Orien
Orientation: HORIZONTAL (default) or VERTICAL.
If Kywrd = COLOR, command format is HPGL,COLOR,KEY.
KEY
Pen choice:
0
Single pen
1
Multiple pens
If Kywrd = DIRECT, command format is HPGL,DIRECT,Port.
Used to direct plotter commands to a port. Use HPGL,FILE to redirect output from a port back to a file. This
option for Kywrd is valid in the DISPLAY program only.
Port
Port name for direct connection.
Notes
This command is available in both the ANSYS and DISPLAY programs. It is valid for Hewlett Packard Graphics
Language (HPGL) format files selected in the ANSYS program with /SHOW,HPGL (or HPGL2), or with
/SHOWDISP,HPGL (or HPGL2) in the DISPLAY program.
An output file is generated for each plot. The ANSYS file is named JobnameNN.HPGL. In the DISPLAY pro-
gram, this file is named HPGLnn. This file remains open for a subsequent /NOERASE plot, and will be incom-
plete until the program is closed (/EXIT), or until the next file is opened by the next /ERASE plot request.
Menu Paths
Utility Menu>PlotCtrls>Redirect Plots>To HPGL File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL2 File
815
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HPGL
HPTCREATE, TYPE, ENTITY, NHP, LABEL, VAL1, VAL2, VAL3
Defines a hard point.
PREP7: Hard Points
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
TYPE
Type of entity on which the hard point will be created.
LINE
Hard point will be created on a line.
AREA
Hard point will be created within an area (not on the boundaries).
ENTITY
Number of the line or area on which the hard point will be created.
NHP
Number assigned to the hard point. Defaults to the lowest available hard point number.
LABEL
If LABEL = COORD, VAL1, VAL2, and VAL3 are the respective global X, Y, and Z coordinates. If LABEL
= RATIO, VAL1 is the parameter value (this is available only for lines). Valid parameter values are between
0 and 1. VAL2 and VAL3 are ignored.
VAL1
If LABEL = RATIO, ratio value for line. If LABEL = COORD, global X coordinate value.
VAL2
If LABEL = COORD, global Y coordinate value.
VAL3
If LABEL = COORD, global Z coordinate value.
Notes
The ability to enter a parameter value provides a simple way of positioning hard points on lines. For example,
to place a hard point halfway along a line, one can simply specify a VAL1 value of 0.5.
For models imported through the DEFAULT IGES filter, you can place hard points on models only by specifying
coordinates (you can't place a hard point using interactive picking).
If you issue any commands that update the geometry of an entity, such as Boolean or simplification com-
mands, any hard points associated with that entity are deleted. Therefore, you should add any hard points
after completing the solid model. If you delete an entity that has associated hard points, those hard points
are either
Deleted along with the entity (if the hard point is not associated with any other entities).
Detached from the deleted entity (if the hard point is associated with additional entities).
When archiving your model (CDWRITE), hardpoint information cannot be written to the IGES file. The Job-
name.cdb file can be written with the CDWRITE,DB option.
Hard points are only applicable for area and volume meshing, not for beams.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
816
HPTCREATE
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Hard Points>Hard PT on area>Hard PT by coordinates
Main Menu>Preprocessor>Modeling>Create>Hard Points>Hard PT on line>Hard PT by coordinates
Main Menu>Preprocessor>Modeling>Create>Hard Points>Hard PT on line>Hard PT by ratio
Main Menu>Preprocessor>Modeling>Create>Keypoints>Hard PT on area>Hard PT by coordinates
Main Menu>Preprocessor>Modeling>Create>Keypoints>Hard PT on area>Hard PT by picking
Main Menu>Preprocessor>Modeling>Create>Keypoints>Hard PT on line>Hard PT by coordinates
Main Menu>Preprocessor>Modeling>Create>Keypoints>Hard PT on line>Hard PT by picking
Main Menu>Preprocessor>Modeling>Create>Keypoints>Hard PT on line>Hard PT by ratio
HPTDELETE, NP1, NP2, NINC
Deletes selected hardpoints.
PREP7: Hard Points
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Delete the pattern of hard points beginning with NP1 to NP2 in steps of NINC (defaults to 1). If NP1 =
ALL, NP2 and NINC are ignored and the pattern is all selected hard points [KSEL]. If NP1 = P, graphical
picking is enabled and all remaining command fields are ignored.
Notes
Deletes all attributes attached to the designated hard points as well as the hard points themselves. If any
entity is attached to a designated hard point, the command detaches the hard point from that entity (the
program will alert you that this will occur).
Menu Paths
Main Menu>Preprocessor>Modeling>Delete>Hard Points
HRCPLX, LOADSTEP, SUBSTEP, OMEGAT, 1STLCASE, 2NDLCASE
Computes and stores in the database the time-harmonic solution at a prescribed phase angle.
POST1: Set Up
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
LOADSTEP
Load step number of the data set to be read (defaults to 1).
SUBSTEP
Substep number within LOADSTEP.
OMEGAT
Angle in degrees ( (angle) times T (time)).
If 360, the amplitude is supplied.
817
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HRCPLX
All others supply results at that angle. For example, if the angle is set to 0.0, the real part of the
solution is supplied. If the angle is set to -90 the imaginary part of the solution is supplied.
1STLCASE
First load case number (defaults to 1).
2NDLCASE
Second load case number (defaults to 2).
Notes
HRCPLX invokes a macro which combines the real and imaginary parts of the solution. If the angle is specified,
it produces the following:
{ } { } cos { } sin R R R
R I
Where:
R
R
and R
I
are, respectively, the real and imaginary parts of the results quantity (e.g. the nodal displacements,
the reaction forces, ...).
is the angle (OMEGAT).
1STLCASE and 2NDLCASE are internally generated load cases. You may want to specify these to avoid
overwriting an existing load case number 1 or 2.
Not all results computed by this command are valid. See Summable, Non-Summable and Constant Data in
the Basic Analysis Guide for more information.
For postprocessing amplitudes, the only appropriate coordinate system is the solution coordinate system
(RSYS ,SOLU). When displaying the displacement amplitudes, use a contour display (PLNSOL command).
Because a deformed shape display (PLDISP command) could lead to a non-physical shape, the displacement
scaling is off by default (/DSCALE,,OFF).
For postprocessing cylindrical geometry, it is suggested that you rotate the element coordinate systems into
the appropriate cylindrical system (EMODIF,,ESYS) before running the solution and then view the results
in this system (RSYS,SOLU) in POST1.
Since HRCPLX performs load case combinations, it alters most of the data in the database. In particular, it
alters applied loads such as forces and imposed displacements. To restore the original loads in the database
for a subsequent analysis, reissue the SET command in POST1 to retrieve the real and imaginary set data.
To animate the solution over one period, use the ANHARM command.
See the Theory Reference for the Mechanical APDL and Mechanical Applications for more information on har-
monic analysis equations and their relationship to real and imaginary data sets.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
818
HRCPLX
HREXP, ANGLE
Specifies the phase angle for the harmonic analysis expansion pass.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
ANGLE
Phase angle (degrees) for expansion pass. If ALL (default), use both 0.0 (real) and 90.0 (imaginary) phase
angles.
Notes
Specifies the phase angle where the expansion pass will be done for a harmonic reduced or harmonic mode
superposition expansion pass.
For a specific angle, the following real solution is stored in the results (*.rst) file:
{ } { cos( )}
max
u u
i i
Where:
i is the degree of freedom number.
u
i
max
is the amplitude of the i th degree of freedom solution
i
is the phase shift angle of the i th degree of freedom solution
is the supplied phase shift angle (ANGLE)
If ANGLE is ALL, both the real and imaginary parts of the solution are stored in the results file.
For more details about the solution equations, see Harmonic Response Analyses in the Theory Reference for
the Mechanical APDL and Mechanical Applications.
This command is ignored if the HROPT command has been issued with Method = VT or Method = VTRU.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>By Load Step
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>By Time/Freq
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>Range of Solu's
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>By Load Step
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>By Time/Freq
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>Range of Solu's
819
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HREXP
HROPT, Method, MAXMODE, MINMODE, MCout, Damp
Specifies harmonic analysis options.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Product Restrictions
Method
Solution method for the harmonic analysis:
FULL
Full method (default).
REDUC
Reduced method.
MSUP
Mode superposition method.
VT
Variational Technology method (based on FULL harmonic algorithm).
VTPA
Variational Technology perfect absorber method (based on FULL harmonic algorithm)..
VTRU
Variational Technology reuse method (based on FULL harmonic algorithm)..
Note
To use any Variational Technology method, you must have an HPC license.
The Variational Technology perfect absorber method (applicable to elements HF119 and
HF120 only) (Method = VTPA) is about 20% faster but slightly less accurate than the Variational
Technology method (Method = VT).
The Variational Technology reuse method (Method = VTRU) simply does the stress pass of a
previous run using the Variational Technology method (Method = VT) using the .RSX file.
It can be used to refine the frequency range and sample density without needing to redo the
entire analysis.
MAXMODE
Largest mode number to be used to calculate the response (for Method = MSUP). Defaults to the highest
mode calculated in the preceding modal analysis. This option does not apply for Method = VT, VTPA,
or VTRU.
MINMODE
Smallest mode number to be used (for Method = MSUP). This option does not apply for Method = VT,
VTPA, or VTRU. Defaults to 1.
MCout
Modal coordinates output key (valid only for the mode superposition method MSUP):
NO
No output of modal coordinates (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
820
HROPT
YES
Output modal coordinates to the text file jobname.MCF.
Note
MCout provides a singular (one-time) output of the modal coordinates. It will not provide
the information for LSWRITE/LSSOLVE operations.
Damp
Damping mode for frequency-dependent material properties (valid only for the Variational Technology
Method VT).
Hysteretic
Not proportional to the frequency.
Viscous
Proportional to the frequency (default).
Notes
Specifies the method of solution for a harmonic analysis (ANTYPE,HARMIC). If used in SOLUTION, this com-
mand is valid only within the first load step. See the product restrictions indicated below.
To include residual vectors in your mode-superposition harmonic analysis, specify RESVEC,ON.
This command is also valid in PREP7.
If Method = FULL, this command is valid in Distributed ANSYS.
Product Restrictions
The VT, VTPA, and VTRU options are only available with a HPC license.
Available Products Command
Option
Method
MP ME ST <> <> <> <> <> EM EH <> PP <> EME MFS FULL
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS REDUC
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS MSUP
Menu Paths
Main Menu>DesignXplorer>Solution>Solve
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
821
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
HROPT
HROUT, Reimky, Clust, Mcont
Specifies the harmonic analysis output options.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Reimky
Real/Imaginary print key:
ON
Print complex displacements as real and imaginary components (default).
OFF
Print complex displacements as amplitude and phase angle (degrees).
Clust
Cluster option (for HROPT,MSUP):
OFF
Uniform spacing of frequency solutions (default).
ON
Cluster frequency solutions about natural frequencies.
Mcont
Mode contributions key (for HROPT,MSUP):
OFF
No print of mode contributions at each frequency (default).
ON
Print mode contributions at each frequency.
Notes
Specifies the harmonic analysis (ANTYPE,HARMIC) output options. If used in SOLUTION, this command is
valid only within the first load step. OUTPR,NSOL must be specified to print mode contributions at each
frequency.
This command is ignored if the HROPT command has been issued with Method = VT, VTPA, or VTRU. Dis-
placements are not available at expanded frequencies with these solution methods.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
822
HROUT
I Commands
IC, NODE, Lab, VALUE, VALUE2, NEND, NINC
Specifies initial conditions at nodes.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
NODE
Node at which initial condition is to be specified. If ALL, apply to all selected nodes (NSEL). If NODE =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may be substituted for NODE.
Lab
Degree-of-freedom label for which the initial condition is to be specified. If ALL, use all appropriate labels.
Structural labels: UX, UY, or UZ (displacements or linear velocities); ROTX, ROTY, or ROTZ (rotations
or angular velocities). Thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP (temperature).
FLOTRAN fluid labels: PRES (pressure); VX, VY, or VZ (velocities); ENKE or ENDS (turbulent kinetic
energy or turbulent energy dissipation); SP01 through SP06 (multiple species mass fractions) or their
user-defined names (MSSPEC). Electric label: VOLT (voltage).
Magnetic labels: MAG (scalar magnetic potential); AX, AY, or AZ (vector magnetic potentials).
VALUE
Initial value of the degree of freedom (first-order value). Defaults to the program default for that degree
of freedom (0.0 for structural analysis, TUNIF for thermal analysis, etc.). Values are in the nodal coordinate
system and in radians for rotational degrees of freedom.
VALUE2
Second-order degree of freedom value, mainly used to specify initial structural velocity. Defaults to the
program default for that degree of freedom (0.0 for structural analysis). Values are in the nodal coordinate
system and in radians/time for rotational degrees of freedom.
NEND, NINC
Specifies the same initial condition values at the range of nodes from NODE to NEND (defaults to NODE),
in steps of NINC (defaults to 1).
Notes
The IC command specifies initial conditions, which are the initial values of the specified degrees of freedom.
It is valid only for a static analysis and full method transient analysis (TIMINT,ON and TRNOPT,FULL). For
the transient, the initial value is specified at the beginning of the first load step, that is, at time = 0.0.
Initial conditions should always be step applied (KBC,1) and not ramped. (In a transient analysis when
SOLCONTROL is ON, KBC,1 is the default as long as TIMINT is also on.)
If constraints (D, DSYM, etc.) and initial conditions are applied at the same node, the constraint specification
overrides. Exercise caution when specifying constraints. The degree-of-freedom values start from zero, or
823
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
the first value given in the table when table name is specified. To match the nonzero initial condition value
with the initial value for degree-of-freedom constraint, use a table for the degree-of-freedom constraint.
For thermal analyses, any TUNIF specification should be applied before the IC command; otherwise, the
TUNIF specification is ignored. If the IC command is input before any TUNIF specification, use the ICDELE
command and then reissue any TUNIF specification and then follow with the IC command.
Be sure to define consistent initial conditions. For example, if you define an initial velocity at a single degree
of freedom, the initial velocity at every other degree of freedom will be 0.0, potentially leading to conflicting
initial conditions. In most cases, you should define initial conditions at every unconstrained degree of freedom
in your model.
After a solution has been performed, the specified initial conditions are overwritten by the actual solution
and are no longer available. You must respecify them if you want to perform a subsequent analysis. You
may want to keep a database file saved prior to the first solution for subsequent reuse.
If you use the CDWRITE command to archive your model, first-order values (initial displacements, temperat-
ures, etc.) specified via the IC command are not written to the archive file; however, second-order (structural
velocity) terms are written.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Initial Condit'n>Define
Main Menu>Solution>Define Loads>Apply>Initial Condit'n>Define
ICDELE
Deletes initial conditions at nodes.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
Deletes all initial conditions previously specified with the IC command at all nodes.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>Initial Condit'n
Main Menu>Solution>Define Loads>Delete>Initial Condit'n
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
824
ICDELE
ICE, ELEM, Lab, VALUE
Specifies initial conditions on elements.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element to which initial condition applies. If All, apply initial condition to all selected elements [ ESEL].
If ELEM = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI). A component name may be substituted for ELEM.
Lab
Valid initial conditions label. FLOTRAN fluid labels: VFRC (volume fraction).
VALUE
Initial value for the volume fraction.
Notes
The ICE command specifies initial conditions on selected elements.
Once a solution has been performed, the specified initial conditions will be overwritten by the actual solution,
and are no longer available. You must respecify them if you want to perform a re-analysis.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Volume Fract>Init Loads>On
Elements
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Volume Fract>Init Loads>On Elements
ICEDELE, ELEM, Lab
Deletes initial conditions on elements.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element at which initial conditions are to be deleted. If ALL, initial conditions at all selected elements
[ESEL] are deleted. If ELEM = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may be substituted for ELEM.
Lab
Valid initial conditions label. FLOTRAN fluid labels: VFRC (volume fraction).
Notes
Deletes all initial conditions previously specified with the ICE command at all elements.
This command is also valid in PREP7.
825
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ICEDELE
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Volume Fract>Init Loads>On
Elements
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Volume Fract>Init Loads>On Elements
ICELIST, ELEM, Lab
Lists initial conditions on elements.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
List initial condition for elements for ELEM (defaults to ALL). If ELEM = All, initial conditions for all selected
elements [ESEL} are listed. If ELEM = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in GUI). A component name may be substituted for ELEM.
Lab
Valid initial conditions label. FLOTRAN fluid labels: VFRC (volume fraction).
Notes
Lists the initial on elements specified by the ICE command. Listing applies to all the selected elements [ESEL].
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Elem Init Condit'n>On Picked Elemts
ICLIST, NODE1, NODE2, NINC, Lab
Lists the initial conditions.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
NODE1, NODE2, NINC
List initial conditions for nodes NODE1 to NODE2 (defaults to NODE1) in steps of NINC (defaults to 1).
If NODE1 = ALL (default), NODE2 and NINC are ignored and initial conditions for all selected nodes
[NSEL] are listed. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may be substituted for NODE1 (NODE2 and NINC are ignored).
Lab
Velocity key:
DISP
Specification is for first order degree of freedom value (displacements, temperature, etc.) (default).
VELO
Specification is for second order degree of freedom value (velocities).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
826
ICELIST
Notes
Lists the initial conditions specified by the IC command. Listing applies to all the selected nodes [NSEL] and
DOF labels. ICLIST is not the same as the DLIST command. All the initial conditions including the default
conditions are listed for the selected nodes.
Initial conditions for a FLOTRAN analysis are primary (first order) degrees of freedom and are thus listed with
the DISP key.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Initial Condit'n>List Picked
Main Menu>Solution>Define Loads>Apply>Initial Condit'n>List Picked
Utility Menu>List>Loads>Initial Conditions>On Picked Nodes
/ICLWID, FACTOR
Scales the line width of circuit builder icons.
GRAPHICS: Scaling
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
FACTOR
Multiplication factor applied to the default line width (defaults to 1). The minimum is 1 and the maximum
is 6.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Scale Icon
/ICSCALE, WN, FACTOR
Scales the icon size for elements supported in the circuit builder.
GRAPHICS: Scaling
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
WN
Window number (or ALL) to which command applies (defaults to 1).
FACTOR
Factor applied to the default icon size (defaults to 1).
827
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/ICSCALE
Notes
Scaling the icon size can provide better visualization of the circuit components when using the Circuit
Builder (an interactive builder available in the ANSYS GUI).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Scale Icon
ICVFRC, Geom, VAL1, VAL2, VAL3, VAL4
Sets the initial volume fraction field for a geometry.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
Geom
Geometry:
CIRC
Circle.
ELPT
Ellipse.
VAL1, VAL2
Location of the center. VAL1 and VAL2 are the x and y coordinates, respectively.
VAL3, VAL4
If Geom = CIRC, VAL3 is the radius of the circle and VAL4 is not used. If Geom = ELPT, VAL3 and VAL4
are the x and y semiaxes of the ellipse, respectively.
Notes
The ICVFRC command sets the initial volume fraction field for a geometry. The initial VFRC is set to one for
elements completely within the geometry. The initial VFRC is set equal to the fraction of element area
within the geometry for elements intersected by the boundary of the geometry.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Volume Fract>Init Loads>By
Geom>Circle
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Volume Fract>Init Loads>By
Geom>Elliptic
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Volume Fract>Init Loads>By Geom>Circle
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Volume Fract>Init Loads>By Geom>Elliptic
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
828
ICVFRC
*IF, VAL1, Oper1, VAL2, Base1, VAL3, Oper2, VAL4, Base2
Conditionally causes commands to be read.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
VAL1
First numerical value (or parameter which evaluates to a numerical value) in the conditional comparison
operation. VAL1, VAL2, VAL3, and VAL4 can also be character strings (enclosed in quotes) or parameters
for Oper = EQ and NE only.
Oper1
Operation label. A tolerance of 1.0E-10 is used for comparisons between real numbers:
EQ --
Equal (for VAL1 = VAL2).
NE --
Not equal (for VAL1 VAL2).
LT --
Less than (for VAL1 < VAL2).
GT --
Greater than (for VAL1 > VAL2).
LE --
Less than or equal (for VAL1
VAL2).
GE --
Greater than or equal (for VAL1
VAL2).
ABLT --
Absolute values of VAL1 and VAL2 before < operation.
ABGT --
Absolute values of VAL1 and VAL2 before > operation.
VAL2
Second numerical value (or parameter which evaluates to a numerical value) in the conditional compar-
ison operation.
Base1
Action based on the logical expression (Oper1) being true. If false, continue reading at the next line.
This is conditional, except for the IF-THEN-ELSE constructs described below; any of the following constructs
(through Base1 = THEN) cause all subsequent fields to be ignored:
:label --
A user-defined label (beginning with a colon (:), 8 characters maximum). The command reader will
skip (and wrap to the beginning of the file, if necessary) to the first line that begins with the
matching :label.
829
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*IF
Caution
This label option may not be mixed with do-loop or if-then-else constructs.
STOP --
This action will cause an exit from the ANSYS program at this line, unless running in interactive mode.
In interactive mode, the program will not stop.
EXIT --
Exit the current do-loop [*EXIT].
CYCLE --
Skip to the end of the current do-loop [*CYCLE].
THEN --
Make this *IF an if-then-else construct (see below).
The following optional values determine the connection between the two logical clauses Oper1 and
Oper2
AND --
True if both clauses (Oper1 and Oper2) are true.
OR --
True if either clause is true.
XOR --
True if either (but not both) clause is true.
VAL3
Third numerical value (or parameter which evaluates to a numerical value).
Oper2
Operation label. This will have the same labels as Oper1, except it uses Val3 and Val4. A tolerance
of 1.0E-10 is used for comparisons between real numbers.
VAL4
Fourth numerical value (or parameter value which evaluates to a numerical value).
Base2
Action based on the logical expression (Oper1 and Oper2) being true. They will be the same values as
Base1, except as noted.
Command Default
Read commands sequentially.
Notes
Conditionally causes commands to be read from a specific block or at a specific location. Twenty levels of
nested *IF blocks are allowed. Jumping to a :label line is not allowed with keyboard entry. Jumping into,
out of, or within a do-loop or an if-then-else construct to a :label line is not allowed. Using *IF interactively
or from the command line prevents rereading the file to find a label. To do so, use batch mode or /INPUT.
The following is an example of an if-then-else construct:
*IF,VAL1,Oper,VAL2,THEN
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
830
*IF
----
*ELSEIF,VAL1,Oper,VAL2
----
*ELSEIF,VAL1,Oper,VAL2
----
*ELSE
----
*ENDIF
where "----" represents a block of any number of commands. Any number of *ELSEIF clauses (or none) may
be included (in the location shown). One *ELSE clause (at most) may be included (in the location shown).
The *IF command is executed by evaluating its logical expression. If it is true, the block of commands fol-
lowing it is executed. The construct is considered to be complete and the command following the *ENDIF
is executed next. If the logical expression is false, the next *ELSEIF command (if any) following the block is
executed. The execution logic is the same as for *IF. The effect is that the logical expressions in the *IF and
the *ELSEIF commands are sequentially tested until one is found to be true. Then the block of commands
immediately following the expression is executed, which completes the execution of the if-then-else construct.
If all *IF and *ELSEIF expressions are false, the block following the *ELSE command is executed, if there is
one. Only one block of commands (at most) is executed within the if-then-else construct. If a batch input
stream hits an end-of-file during a false *IF condition, the ANSYS run will not terminate normally. You will
need to terminate it externally (use either the UNIX kill function or the Windows task manager). The *IF,
*ELSEIF , *ELSE, and *ENDIF commands for each if-then-else construct must all be read from the same file
(or keyboard).
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
IGESIN, Fname, Ext, --
Transfers IGES data from a file into ANSYS.
AUX15: IGES
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
831
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
IGESIN
The extension defaults to CAD if Fname is blank.
--
Unused field.
Notes
Reads a file containing IGES data and transfers it into the ANSYS database. The file transferred is the IGES
Version 5.1, ASCII format file. IGES (Initial Graphics Exchange Specification) is a neutral format developed by
the U.S. Dept. of Commerce, National Institute of Standards and Technology. There is no output transfer file
written since the transferred data is read directly into the ANSYS database.
You can import multiple files into a single database, but you must use the same import option (set with the
IOPTN command) for each file.
The IOPTN command sets the parameters for reading the file (refer to this command for the options and
their support for IGES entities). The two main parameters are FACETED and SMOOTH. You cannot change
the IOPTN settings once a file has been imported.
When a file is read using the FACETED option, the data is stored in an defeaturing database. Once the
transferred data is stored, a special set of geometry repair and enhancement tools, designed specifically for
use with files imported from CAD systems, are available. Files read through the SMOOTH option use the
standard database.
Menu Paths
Utility Menu>File>Import
IGESOUT, Fname, Ext, --, ATT
Writes solid model data to a file in IGES Version 5.1 format.
PREP7: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to IGES if Fname is blank.
--
Unused field.
ATT
Attribute key:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
832
IGESOUT
0
Do not write assigned numbers and attributes of the solid model entities to the IGES file (default).
1
Write assigned numbers and attributes of solid model entities (keypoints, lines, areas, volumes) to
the IGES file. Attributes include MAT, TYPE, REAL, and ESYS specifications as well as associated solid
model loads and meshing (keypoint element size, number of line divisions and spacing ratio) spe-
cifications.
Notes
Causes the selected solid model data to be written to a coded file in the IGES Version 5.1 format. Previous
data on this file, if any, are overwritten. Keypoints that are not attached to any line are written to the output
file as IGES entity 116 (Point). Lines that are not attached to any area are written to the output file as either
IGES Entity 100 (Circular Arc), 110 (Line), or 126 (Rational B-Spline Curve) depending upon whether the ANSYS
entity was defined as an arc, straight line, or spline. Areas are written to the output file as IGES Entity 144
(Trimmed Parametric Surface). Volumes are written to the output file as IGES entity 186 (Manifold Solid B-
Rep Object). Solid model entities to be written must have all corresponding lower level entities selected
(use ALLSEL,BELOW,ALL) before issuing command. Concatenated lines and areas are not written to the IGES
file; however, the entities that make up these concatenated entities are written.
Caution
Section properties assigned to areas, lines and other solid model entities will not be maintained
when the model is exported from ANSYS using IGESOUT.
If you issue the IGESOUT command after generating a beam mesh with orientation nodes, the
orientation keypoints that were specified for the line [LATT] are no longer associated with the
line and are not written out to the IGES file. The line does not recognize that orientation keypoints
were ever assigned to it, and the orientation keypoints do not "know" that they are orientation
keypoints. Thus the IGESOUT command does not support (for beam meshing) any line operation
that relies on solid model associativity. For example, meshing the areas adjacent to the meshed
line, plotting the line that contains the orientation nodes, or clearing the mesh from the line that
contains orientation nodes may not work as expected. See Meshing Your Solid Model in the
Modeling and Meshing Guide for more information about beam meshing.
You cannot access this command for models that have been imported from IGES using the FACETED trans-
lator (see the IOPTN command).
Menu Paths
Utility Menu>File>Export
833
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
IGESOUT
/IMAGE, Label, Fname, Ext, --
Allows graphics data to be captured and saved.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Label specifying the operation to be performed:
CAPTURE
Capture the image from the graphics window to a new window.
RESTORE
Restore the image from a file to a new window.
SAVE
Save the contents of the graphic window to a file.
DELETE
Delete the window that contains the file.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
If no extension is specified, bmp will be used on Windows systems, and img will be used on UNIX systems.
--
Unused field.
Menu Paths
This command cannot be accessed from a menu.
IMAGIN, IR, IA, --, --, Name, --, --, FACTA
Forms an imaginary variable from a complex variable.
POST26: Operations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA
Reference number of the variable to be operated on.
--, --
Unused fields.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
834
/IMAGE
Name
Thirty-two character name for identifying the variable on the printout and displays. Embedded blanks
are compressed upon output.
--, --
Unused fields.
FACTA
Scaling factor (positive or negative) applied to variable IA (defaults to 1.0).
Notes
This command forms a new variable from a complex variable by storing the imaginary part as the real part.
The imaginary part can then be used in other operations. Used only with harmonic analyses (ANTYPE,HARMIC).
Complex variables are stored in two-column arrays with the real component stored in the first column and
the imaginary component stored in the second column. This command extracts the value stored in the
second column (i.e., imaginary component). However, with harmonic analyses, all variables are stored in
two-column arrays as complex variables. If the variable is not complex, then the same value is stored in both
columns. This command will extract the variable in the second column of the array, even if this variable is
not the imaginary component of a complex variable.
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Imaginary Part
IMESH, LAKY, NSLA, NTLA, KCN, DX, DY, DZ, TOL
Generates nodes and interface elements along lines or areas.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LAKY
Copies mesh according to the following:
LINE or 1
Copies line mesh (default).
AREA or 2
Copies area mesh.
NSLA
Number that identifies the source line or area. This is the line or area whose mesh will provide the pattern
for the interface elements. ANSYS copies the pattern of the line or area elements through the area or
volume to create the mesh of area or volume interface elements.
NTLA
Number that identifies the target line or area. This is the line or area that is opposite the source line or
area specified by NSLA. Add NTLA to obtain the copied mesh from the source line or area.
KCN
Number that identifies the particular ANSYS coordinate system.
835
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
IMESH
DX, DY, DZ
Incremental translation of node coordinates in the active coordinate system (DR, D, DZ for cylindrical,
and DR, D, D for spherical or toroidal). The source line or area coordinates + DX, DY, DZ = the target
line or area coordinates. If left blank, ANSYS automatically estimates the incremental translation.
TOL
Tolerance for verifying topology and geometry. By default, ANSYS automatically calculates the tolerance
based on associated geometries.
Notes
Generates nodes and interface elements along lines or areas. The IMESH command requires that the target
line or area exactly match the source line or area. Also, both target and source lines or areas must be in the
same area or volume. The area or volume containing the source line or area must be meshed before executing
IMESH, while the area or volume containing the target line or area must be meshed after executing IMESH.
For three dimensional problems where LAKY = AREA, ANSYS fills the interface layer according to the following
table:
ANSYS fills the interface layer with: If source mesh consists of:
Hexahedral elements Quadrilateral elements
Degenerated wedge elements Triangle elements
Combination hexahedral and degenerated wedge
elements
Combination quadrilateral and triangle
elements
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Interface Mesh>2D Interface
Main Menu>Preprocessor>Meshing>Mesh>Interface Mesh>3D Interface
IMMED, KEY
Allows immediate display of a model as it is generated.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Immediate mode key:
0
Display only upon request, i.e., no immediate display (default with the GUI off).
1
Display immediately as model is generated (default with the GUI on).
Command Default
As described above.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
836
IMMED
Notes
Allows immediate display of a model (as it is generated) without a screen erase or a display request. Available
only during an interactive session at a graphics display terminal. A valid graphics device name must first be
specified on the /SHOW command.
The IMMED command allows you to control whether or not the model is displayed immediately as it is
generated in an interactive session. By default in the GUI, your model will immediately be displayed in the
Graphics Window as you create new entities (such as areas, keypoints, nodes, elements, local coordinate
systems, boundary conditions, etc.). This is called immediate mode graphics. Also note that symbols (such
as boundary conditions, local coordinate system triads, etc.) are shown immediately and will be present on
subsequent displays unless you "turn off" the appropriate symbol using the GUI plot controls function or
the appropriate graphics specification command.
An immediate image will also be automatically scaled to fit nicely within the Graphics Window -- a feature
called automatic scaling. The new scaling is usually apparent on the automatic replot associated with imme-
diate mode. To suppress automatic replot, issue /UIS,REPLOT,0. (With automatic replot suppressed, the im-
mediate image may not always be automatically scaled correctly.)
Note
An immediate display in progress should not be aborted with the usual system "break" feature
(or else the ANSYS session itself will be aborted). When you run the ANSYS program interactively
without using the GUI, immediate mode is off by default.
This command is valid only in PREP7.
Menu Paths
Utility Menu>PlotCtrls>Erase Options>Immediate Display
IMPD, Vpath, Ipath, Vsymm, Isymm
Calculates the impedance of a conductor at a reference plane.
POST1: Magnetics Calculations
MP ME <> <> <> <> <> <> <> EH <> PP <> EME MFS
Vpath
Path name for a predefined path [PATH command] for calculating the EMF (voltage drop) from the
conductor to a reference point. The path should start at the outer conductor wall and end at a reference
voltage point.
Ipath
Path name for a predefined path [PATH command] for calculating the MMF (current) in a conductor.
The path should traverse a closed contour surrounding the conductor, and you should define the path
in a counterclockwise direction.
Vsymm
Symmetry factor applied to the calculated EMF (voltage drop). The EMF (voltage drop) from the conductor
to the reference point is multiplied by Vsymm.
837
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
IMPD
Isymm
Symmetry factor applied to the calculated current. The calculated current is multiplied by Isymm.
Notes
Used in a harmonic high-frequency electromagnetic analysis, IMPD calculates the impedance of a conductor
at a reference plane from the EMF (voltage) and MMF (current) at the reference plane. The EMF (voltage
drop) is calculated by a line integral from the input path name (specified by the Vpath argument) that extends
from the conductor to a reference point. The MMF (current) is calculated by a closed path around the con-
ductor from the input path name (specified with Ipath). In cases having modeled symmetry, you can
multiply the voltage drop or current by symmetry factors (Vsymm and Isymm respectively).
This command macro returns the scalar parameters Zre and Zim, representing the real and imaginary com-
ponents of the impedance.
See magnetic macros for further details.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>Impedance
INISTATE, Action, Val1, Val2, Val3, Val4, Val5, Val6, Val7, Val8, Val9
Defines initial state data and parameters.
PREP7: Data Tables
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Action
Specifies action for defining or manipulating initial state data:
SET
Use Action = SET to designate initial state coordinate system, data type and material type paramet-
ers. See "Command Specification for Action = SET" (p. 839).
DEFINE
Use Action = DEFINE to specify the actual state values, and the corresponding element, integration
point, or layer information. See "Command Specifications for Action = DEFINE" (p. 839).
WRITE
Use Action = WRITE to write the initial state values to a file when the SOLVE command is issued.
See "Command Specifications for Action = WRITE" (p. 840).
READ
Use Action = READ to read the initial state values from a file. See "Command Specifications for Ac-
tion = READ" (p. 840).
LIST
Use Action = LIST to read out the initial state data. See "Command Specifications for Action =
LIST" (p. 841).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
838
INISTATE
DELETE
Use Action = DELE to delete initial state data from a selected set of elements. See "Command
Specifications for Action = DELETE" (p. 841)
Notes
For more information about initial state, and to view the list of elements supported, see "Initial State" in the
Basic Analysis Guide.
Command Specification for Action = SET
INISTATE, SET, Val1, Val2
Val2 = Val1 =
Coordinate system. Val2 will be an integer corresponding to the coordinate
system:
CSYS
-2 = Element Coordinate System
-1 = Material Coordinate System
0 = Global Cartesian Coordinate System
0 - 10 = Any ANSYS defined coordinate system.
>= 11 = Any user defined coordinate system ID.
Data type. Val2 is the type of data that will be set on the subsequent
INISTATE,DEFINE command:
DTYP
STRE = Stress data (default)
EPEL = Strain data
EPPL = Plastic strain data
Material type. Val2 is the material ID. Using Val2 = -1 disables material-
based initial state and enables integration-point-based initial state data.
MAT
Notes
Action = SET specifies and modifies the environment into which you will define the initial state data (via
a subsequent INISTATE,DEFINE command). Otherwise, subsequent INISTATE,DEFINE data is input as initial
state data in the global Cartesian coordinate system.
Command Specifications for Action = DEFINE
INISTATE, DEFINE, ELID, Eint, Klayer, Parmint, Cxx, Cyy, Czz, Cxy,
Cyz, Cxz
ELID --
Element ID number. If (blank), defaults to current element selection.
Eint --
Gauss integration point (defaults to ALL). Not valid for material-based initial state data.
Klayer --
Layer number (for layered solid/shell elements) or cell number for beam elements. Blank for nonlayered,
non-beam, elements, and not valid (ignored) for material-based initial state data.
839
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
INISTATE
ParmInt --
Section integration point within a layer, or cell-integration point for beams (typically four integration
points). The default value is ALL. Not valid for material-based initial state data.
Cxx, Cyy, Czz, Cxy, Cyz, Cxz --
Stress (S), strain (EPEL), or plastic strain (EPPL) values.
Notes
You can issue the INISTATE command repeatedly to define multiple sets of initial state values. Initial state
data can be specified according to elements, layers or integration points.
When the initial state parameters are being defined based on the material, (INISTATE,SET,MAT,MATID), the
ELID value designates the element number and all subsequent values are ignored.
Command Specifications for Action = WRITE
INISTATE, WRITE, FLAG, , , , CSID, Dtype
FLAG --
Set this value to 1 to generate the initial state file, or 0 to disable initial state file generation.
CSID --
Determines the coordinate system for the initial state:
0 (default)
Write in global Cartesian coordinate system for solid elements.
-1 (or MAT)
Write in material coordinate system
-2 (or ELEM)
Write in element coordinate system for link, beam, and layered elements.
Dtype --
Sets the data type to be written in the IST file:
S
Output stresses.
EPEL
Output elastic strain.
EPPL
Output plastic strain.
Notes
Default is 0 for solid elements and -2 for link, beam, and shell elements.
Command Specifications for Action = READ
INISTATE, READ, Fname, Ext, Path
Read initial state data from a standalone initial state file of the specified name (Fname) and file name extension
(Ext), located in the specified path (Path). The initial state file must be in a comma-delimited ASCII file
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
840
INISTATE
format, consisting of individual rows for each stress/strain item, with each row consisting of columns separated
by commas.
Notes
Use the READ option to apply complex sets of initial state data to various elements, cells, layers, sections
and integration points. See "Initial State" in the Basic Analysis Guide for additional information.
Command Specifications for Action = LIST
INISTATE, LIST, ELID
Lists initial state data for elements with ID = ELID. If ELID is blank, all initial state data for all selected ele-
ments are listed.
Command Specifications for Action = DELETE
INISTATE, DELE, ELID
Deletes initial state data for elements with ID = ELID. If ELID is blank, all initial state data for all selected
elements are deleted.
Menu Paths
This command cannot be accessed from a menu.
/INPUT, Fname, Ext, --, LINE, LOG
Switches the input file for the commands that follow.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to the current Jobname if Ext is specified.
Ext
Filename extension (8 character maximum).
--
Unused field.
LINE
A value indicating either a line number in the file or a user-defined label in the file from which to begin
reading the input file.
(blank), 0, or 1
Begins reading from the top of the file (default).
841
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/INPUT
LINE_NUMBER
Begins reading from the specified line number in the file.
:label
Begins reading from the first line beginning with the matching user-defined label :label (beginning
with a colon (:), 8 characters maximum).
LOG
Indicates whether secondary input from this command should be recorded in the command log
(File.LOG) and the database log:
0
Record only the /INPUT command on the log (default).
1
Record commands in the specified secondary file as they are executed.
Notes
Switches the input file for the next commands. Commands are read from this file until an end-of-file or an-
other file switching directive is read. An end-of-file occurs after the last record of the file or when a /EOF
command is read. An automatic switch back one level (to the previous file) occurs when an end-of-file is
encountered. Twenty levels of nested file switching are allowed. Note that files including *DO, *USE, *ULIB,
and the "Unknown Command" Macro have less nesting available because each of these operations also uses
a level of file switching. For an interactive run, a /INPUT,TERM switches to the terminal for the next input.
A /EOF read from the terminal then switches back to the previous file. A /INPUT (with a blank second field)
switches back to the primary input file.
Setting LOG = 1 on /INPUT causes all commands read from the specified file to be recorded in the command
log (File.LOG) and the internal database command log [LGWRITE]. This option is recommended if the
log file will be used later (e.g., as batch input or as an analysis file for design optimization). The LOG = 1
option is only valid when the /INPUT occurs in the primary input file. Using LOG = 1 on a nested /INPUT
or on a /INPUT within a do-loop will have no effect (i.e., commands in the secondary input file are not
written to the command log).
This command is valid in any processor.
Menu Paths
Utility Menu>File>Read Input from
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
842
/INPUT
/INQUIRE, StrArray, FUNC
Returns system information to a parameter.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
StrArray
Name of the "string array" parameter that will hold the returned values. String array parameters are
similar to character arrays, but each array element can be as long as 128 characters. If the string para-
meter does not exist, it will be created.
FUNC
Specifies the type of system information returned:
LOGIN --
Returns the pathname of the login directory on UNIX systems or the pathname of the default directory
(including drive letter) on Windows systems.
DOCU --
Returns the pathname of the ANSYS docu directory.
APDL --
Returns the pathname of the ANSYS APDL directory.
PROG --
Returns the pathname of the ANSYS executable directory.
AUTH --
Returns the pathname of the directory in which the license file resides.
USER --
Returns the name of the user currently logged-in.
DIRECTORY --
Returns the pathname of the current directory.
JOBNAME --
Returns the current Jobname. The value of Jobname can be up to 250 characters in length.
RSTDIR --
Returns rst directory (FILE command).
RSTFILE --
Returns rst file name (FILE command).
RSTEXT --
Returns rst file extension (FILE command).
PSEARCH --
Returns path used for "unknown command" macro (/PSEARCH command).
Returning the Value of an Environment Variable to a Parameter
If FUNC=ENV, the command format is /INQUIRE,StrArray,ENV,ENVNAME,Substring. In this instance,
ENV specifies that the command should return the value of an environment variable. The following defines
the remaining fields:
843
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/INQUIRE
ENVNAME
Specifies the name of the environment variable.
Substring
If Substring = 1, the first substring (up to the first colon (:)) is returned. If Substring = 2, the second
substring is returned, etc. For Windows platforms, the separating character is semicolon (;). If this argument
is either blank or 0, the entire value of the environment variable is returned.
Returning the Value of a Title to a Parameter
If FUNC = TITLE, the command format is /INQUIRE,StrArray,TITLE,Title_num. In this context, the value
of Title_num can be blank or 1 through 5. If the value is 1 or blank, the title is returned. If the value is 2
through 5, a corresponding subtitle is returned (2 denoting the first subtitle, and so on).
Returning Information About a File to a Parameter
The /INQUIRE command can also return information about specified files within the file system. For these
capabilities, the format is /INQUIRE,Parameter,FUNC,Fname, Ext, --. The following defines the fields:
Parameter
Name of the parameter that will hold the returned values.
FUNC
Specifies the type of file information returned:
EXIST --
Returns a 1 if the specified file exists, and 0 if it does not.
DATE --
Returns the date stamp of the specified file in the format yyyymmdd.hhmmss.
SIZE --
Returns the size of the specified file in MB.
WRITE --
Returns the status of the write attribute. A 0 denotes no write permission while a 1 denotes write
permission.
READ --
Returns the status of the read attribute. A 0 denotes no read permission while a 1 denotes read
permission.
EXEC --
Returns the status of the execute attribute (this has meaning only in UNIX). A 0 denotes no execute
permission while a 1 denotes execute permission.
LINES --
Returns the number of lines in an ASCII file.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
844
/INQUIRE
--
Unused field.
Notes
The /INQUIRE command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
INRES, Item1, Item2, Item3, Item4, Item5, Item6, Item7, Item8
Identifies the data to be retrieved from the results file.
POST1: Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item1, Item2, Item3, . . . , Item8
Data to be read into the database from the results file. May consist of any of the following labels:
ALL
All solution items (default).
BASIC
NSOL, RSOL, NLOAD, STRS, FGRAD, and FFLUX items.
NSOL
Nodal DOF solution.
RSOL
Nodal reaction loads.
ESOL
Element solution items (includes all of the following):
NLOAD
Element nodal loads.
STRS
Element nodal stresses.
EPEL
Element elastic strains.
EPTH
Element thermal, initial, and swelling strains.
EPPL
Element plastic strains.
EPCR
Element creep strains.
FGRAD
Element nodal gradients.
845
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
INRES
FFLUX
Element nodal fluxes.
MISC
Element miscellaneous data (SMISC and NMISC).
Notes
Identifies the type of data to be retrieved from the results file for placement into the database through
commands such as SET, SUBSET, and APPEND. INRES is a companion command to the OUTRES command
controlling data written to the database and the results file. Since the INRES command can only flag data
that has already been written to the results file, care should be taken when using the OUTRES command
to include all data you wish to retrieve for postprocessing later on.
Menu Paths
Main Menu>General Postproc>Data & File Opts
INRTIA
Specifies "Inertial loads" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Loads>Inertia Loads
Utility Menu>List>Status>Solution>Inertia Loads
INT1, IR, IY, IX, --, Name, --, --, FACTA, FACTB, CONST
Integrates a variable.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result. Table values represent integrated sum of IY to current table position of IX.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
846
INRTIA
IY, IX
Integrate variable IY with respect to IX.
--
Unused field.
Name
Thirty-two character name for identifying the variable on the printout and displays. Embedded blanks
are compressed upon output.
--, --
Unused fields.
FACTA, FACTB
Scaling factors (positive or negative) applied to the corresponding variables (default to 1.0).
CONST
Initial value.
Notes
Integrates variables according to the operation:
IR = (FACTA x IY) d(FACTB x IX) + CONST
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Integrate
INTSRF, Lab
Integrates nodal results on an exterior surface.
POST1: Special Purpose
MP ME <> <> <> <> <> FL <> <> DY PP <> EME MFS
Lab
Label indicating degree of freedom to be integrated:
PRES
Pressure.
TAUW
Wall shear stress.
FLOW
Both pressure and wall shear stress.
Notes
Integrates nodal results on a surface. Use node selection (such as the EXT option of the NSEL command) to
indicate the surface(s) of element faces to be used in the integration. A surface can be "created" by unselecting
elements (such as unselecting non-fluid elements that are adjacent to fluid elements for the postprocessing
of fluid flow result). Element faces attached to the selected nodes will be automatically determined. All nodes
847
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
INTSRF
on a face must be selected for the face to be used. The integration results will cancel for nodes on common
faces of adjacent selected elements.
Integration results are in the active coordinate system (see the RSYS command). The type of results coordinate
system must match the type used in the analysis. However, you may translate and rotate forces and moments
as needed. Use the *GET command (Utility Menu> Parameters> Get Scalar Data) to retrieve the results.
Menu Paths
Main Menu>General Postproc>Nodal Calcs>Surface Integral
IOPTN, Lab, VAL1
Controls options relating to importing a model.
AUX15: IGES
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label identifying the import option. The meaning of VAL1 will vary depending on Lab.
STAT
List overall status of import facilities, including current option values. VAL1 is ignored.
DEFA
Set default values for all import options. VAL1is ignored.
MERG
Entity merge option. VAL1 can be:
YES
Automatic merging is performed (default).
NO
No merging of entities.
SOLID
Solid option. VAL1 can be:
YES
Solid is created automatically (default).
NO
No solid created.
GTOLER
Entity merging tolerance. Valid arguments for VAL1 depend on value of IGES. If IGES = FACETED,
then GTOLER,VAL1 is a numeric value used as a multiplying factor to change the maximum model
dimension. If IGES = SMOOTH, then GTOLER,VAL1 can be:
DEFA
Use system defaults (default).
FILE
Use tolerance from the imported file.
n
A user-specified tolerance value.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
848
IOPTN
IGES
IGES import option. VAL1 can be:
STAT
List status of IGES related options in the output window.
SMOOTH (or RV52)
Use more robust IGES revision 5.2 import function (default).
FACETED (or RV53)
Use defeaturing database.
SMALL
Small areas option. VAL1 can be:
YES
Small areas are deleted (default).
NO
Small areas are retained.
VAL1
Additional input value as described under each option for Lab.
Command Default
Merging will be performed during the IGES transfer with no global solid model tolerance (GTOLER) used.
Notes
Controls various options during a model file transfer (e.g., IGESIN command). A global solid model tolerance
(GTOLER) can be specified.
IOPTN,SMALL,YES (default value) will delete small areas and can cause geometrical inconsistencies that
could cause the import process to abort. Retaining the small areas will increase processor time and memory
usage.
The two IGES translation options, SMOOTH and FACETED, determine which graphics database will be used
when IGES files are read into ANSYS. If the SMOOTH option is chosen (default), the data is stored in the
standard ANSYS graphics database. If the FACETED option is used, the defeaturing database is used and a
special set of geometry repair and enhancement tools, designed specifically for use with files imported from
CAD systems, are made available.
The SMOOTH option is capable of reading in any rational B-spline curve entity (type 126), or rational B-spline
surface entity (type 128) with a degree less than or equal to 20. Attempts to read in B-spline curve or surface
entities of degree higher than 20 may result in error messages.
The FACETED option translates all IGES topological and geometric entities. The option ignores data such as
dimensions, text, annotation entities, structure entities, and any IGES entities that the filter doesn't recognize.
The FACETED option is not recommended for large, complex geometries.
If you issue the /CLEAR command, the IOPTN settings will return to their default.
If IOPTN, MERG, YES is used, merging of coincident geometry items is performed automatically when the
IGESIN command is issued. (In other words, an internal NUMMRG,KP command is issued.) The model will
be merged with the consideration tolerance (TOLER on NUMMRG) set equal to 0.75 * the shortest distance
849
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
IOPTN
between the endpoints of any active line. See the NUMMRG command for more information on these toler-
ances. In most cases, this default merging is appropriate. The IOPTN command should be used when you
want to turn off merging operations, override the default merging and specify a global solid model tolerance
value (GTOLER), or turn off the automatic creation of solids (SOLID). If used, the IOPTN command should
be issued before the IGESIN command. You cannot change these options once your model has been imported
or created. If you must change these options, clear the database using the /CLEAR command (or exit and
restart ANSYS), set the correct options, and reimport or recreate the model.
This command is valid in any processor.
Menu Paths
Utility Menu>File>Import
IRLF, KEY
Specifies that inertia relief calculations are to be performed.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS
KEY
Calculation key:
0
No inertia relief calculations.
1
Counterbalance loads with inertia relief forces.
-1
Precalculate masses for summary printout only (no inertia relief ).
Command Default
No inertia relief calculations.
Notes
The IRLF command specifies that the program is to calculate accelerations to counterbalance the applied
loads (inertia relief ). Displacement constraints on the structure should be only those necessary to prevent
rigid-body motions (3 are needed for a 2-D structure and 6 for a 3-D structure). The sum of the reaction
forces at the constraint points will be zero. Accelerations are calculated from the element mass matrices and
the applied forces. Data needed to calculate the mass (such as density) must be input. Both translational
and rotational accelerations may be calculated.
This option applies only to the static (ANTYPE,STATIC) analysis. Nonlinearities, elements that operate in the
nodal coordinate system, and axisymmetric or generalized plane strain elements are not allowed. Models
with both 2-D and 3-D element types are not recommended. Loads may be input as usual. Displacements
and stresses are calculated as usual. Use IRLIST to print inertia relief calculation results. The mass and moment
of inertia summary printed before the solution is accurate (because of the additional pre-calculations required
for inertia relief ). See Inertia Relief in the Theory Reference for the Mechanical APDL and Mechanical Applications
for calculation details. See also the Structural Analysis Guide for procedural details.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
850
IRLF
If the inertia relief calculation is to be performed in the second or later load step, you must specify EMAT-
WRITE,YES in the initial load step for the element matrices needed to perform the calculations to be available.
When a superelement (MATRIX50) is present in the model, any DOF constraints that you need to apply (D)
on a degree of freedom (DOF) belonging to the superelement must be applied in the use pass of the MAT-
RIX50 element (not in the generation pass). The command has no effect in the generation pass of a substruc-
ture. In the expansion pass, precalculation of masses for summary printout (IRLF,-1) occurs only on elements
that are part of the substructure.
This command is also valid in PREP7.
Distributed ANSYS Restriction A KEY value of 1 is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Inertia Relief
Main Menu>Preprocessor>Loads>Load Step Opts>Output Ctrls>Incl Mass Summry
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Inertia Relief
Main Menu>Solution>Load Step Opts>Output Ctrls>Incl Mass Summry
IRLIST
Prints inertia relief summary table.
POST1: Listing
MP ME ST PR PRN DS DSS FL <> <> <> PP <> EME MFS
Notes
Prints the inertia relief summary data, including the mass summary table, the total load summary table, and
the inertia relief summary table resulting from the inertia relief calculations. These calculations are performed
in the solution phase [SOLVE or PSOLVE] as specified by the IRLF command.
Inertia relief output is stored in the database rather than in the results file (Jobname.RST). When you issue
IRLIST, ANSYS pulls the information from the database, which contains the inertia relief output from the
most recent solution [SOLVE or PSOLVE].
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
851
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
IRLIST
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
852
J Commands
JPEG, Kywrd, OPT
Provides JPEG file export for ANSYS displays.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Kywrd
Specifies various JPEG file export options.
QUAL
If Kywrd = QUAL, then OPT is an integer value defining the JPEG quality index on an arbitrary scale
ranging from 1 to 100. The default value is 75.
ORIENT
If Kywrd = ORIENT, then OPT will determine the orientation of the entire plot. OPT can be either
Horizontal (default) or Vertical.
COLOR
If Kywrd = COLOR, then OPT will determine the color depth of the saved file. OPT can be 0, 1, or
2, corresponding to Black and White, Grayscale, and Color (default), respectively.
TMOD
If Kywrd = TMOD, then OPT will determine the text method. OPT can be either 1 or 0, corresponding
to bitmap text (default) or line stroke text, respectively.
DEFAULT
If Kywrd = DEFAULT, then all of the default values, for all of the Kywrd parameters listed above, are
active.
OPT
OPT can have the following names or values, depending on the value for Kywrd (see above).
1 to 100
If Kywrd = QUAL, a value between 1 and 100 will determine the quality index of the JPEG file.
Horizontal, Vertical
If Kywrd = ORIENT, the terms Horizontal or Vertical determine the orientation of the plot.
0,1,2
If Kywrd = COLOR, the numbers 0, 1, and 2 correspond to Black and White, Grayscale and Color,
respectively.
1,0
If Kywrd = TMOD, the values 1 and 0 determine whether bitmap (1) or stroke text (0) fonts will be
used
Menu Paths
Utility Menu>PlotCtrls>HardCopy>ToFile
853
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
JSOL, NVAR, ELEM, ITEM, COMP, Name
Specifies result items to be stored for the joint element.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVAR
Arbitrary reference number or name assigned to this variable. Variable numbers can be 2 to NV (NUMVAR)
while the name can be an eight-byte character string. Overwrites any existing results for this variable.
ELEM
Element number for which to store results.
Item
Label identifying the item. Valid item labels are shown in Table 224: JSOL - Valid Item and Component
Labels (p. 854) below.
Comp
Component of the Item (if required). Valid component labels are shown in Table 224: JSOL - Valid Item
and Component Labels (p. 854) below.
Name
Thirty-two character name identifying the item on printouts and displays. Defaults to a label formed by
concatenating the first four characters of the Item and Comp labels.
Notes
This command is valid for the MPC184 joint elements. The values stored are for the free or unconstrained
degrees of freedom of a joint element. Relative reaction forces and moments are available only if stiffness,
damping, or friction is associated with the joint element.
Table 224 JSOL - Valid Item and Component Labels
Description Comp Item
x, y, or z relative displacement. X, Y, Z U
x, y, or z relative rotation. X, Y, Z ROT
x, y, or z relative linear velocity. X, Y, Z VEL
x, y, or z relative angular velocity. X, Y, Z OMG
x, y, or z relative linear acceleration. X, Y, Z ACC
x, y, or z relative angular acceleration. X, Y, Z DMG
Relative reaction forces in the local x, y, or z direction. X, Y, Z RF
Relative reaction moments in the local x, y, or z direction. X, Y, Z RM
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
854
JSOL
K Commands
K, NPT, X, Y, Z
Defines a keypoint.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPT
Reference number for keypoint. If zero, the lowest available number is assigned [NUMSTR].
X, Y, Z
Keypoint location in the active coordinate system (may be R, , Z or R, , ). If X = P, graphical picking
is enabled and all other fields (including NPT) are ignored (valid only in the GUI).
Notes
Defines a keypoint in the active coordinate system [CSYS] for line, area, and volume descriptions. A previously
defined keypoint of the same number will be redefined. Keypoints may be redefined only if it is not yet at-
tached to a line or is not yet meshed. Solid modeling in a toroidal system is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Keypoints>In Active CS
Main Menu>Preprocessor>Modeling>Create>Keypoints>On Working Plane
KATT, MAT, REAL, TYPE, ESYS
Associates attributes with the selected, unmeshed keypoints.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT, REAL, TYPE, ESYS
Material number, real constant set number, type number, and coordinate system number to be associated
with selected, unmeshed keypoints.
Notes
Keypoints subsequently generated from the keypoints will also have these attributes. These element attributes
will be used when the keypoints are meshed. If a keypoint does not have attributes associated with it (by
this command) at the time it is meshed, the attributes are obtained from the then current MAT, REAL,TYPE,
and ESYS command settings. Reissue the KATT command (before keypoints are meshed) to change the
attributes. A zero (or blank) argument removes the corresponding association.
855
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
If any of the arguments MAT, REAL, TYPE, or ESYS are defined as -1, then that value will be left unchanged
in the selected set.
In some cases, ANSYS can proceed with a keypoint meshing operation even when no logical element type
has been assigned via KATT,,,TYPE or TYPE. For more information, see the discussion on setting element
attributes in Meshing Your Solid Model in the Modeling and Meshing Guide.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>All Keypoints
Main Menu>Preprocessor>Meshing>Mesh Attributes>Picked KPs
KBC, KEY
Specifies stepped or ramped loading within a load step.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
KEY
Ramping key:
0
Loads are linearly interpolated (ramped) for each substep from the values of the previous load step
to the values of this load step.
1
Loads are step changed (stepped) at the first substep of this load step to the values of this load step
(i.e., the same values are used for all substeps). Useful for rate-dependent behavior (e.g., creep, vis-
coplasticity, etc.) or transient load steps only.
Command Default
When SOLCONTROL is ON, ANSYS performs ramped loading if ANTYPE,STATIC, or if ANTYPE,TRANS and
TIMINT,OFF. It performs stepped loading if ANTYPE,TRANS and TIMINT,ON. (In a transient analysis, TIMINT,ON
is the default.)
When SOLCONTROL is OFF, ramped loading for all types of transient or nonlinear analysis.
Notes
Specifies whether loads applied to intermediate substeps within the load step are to be stepped or ramped.
Used only if DTIME on the DELTIM command is less than the time span or, conversely, if NSBSTP on the
NSUBST command is greater than one. Flags (FSI, MXWF, MVDI, etc.) are always stepped.
For ramped loading (KBC,0), when a load is applied for the first time, it is interpolated from zero to the value
of the current load step, and not from the initial condition or value of the DOF from the previous load step.
Tabular boundary conditions do not support ramping and instead apply their full value regardless of the
KBC setting. Cyclic symmetry analyses applies only stepped loading, regardless of the KBC setting.
Irrespective of the KBC setting, loads are usually step-removed. See Stepping or Ramping Loads in the Basic
Analysis Guide for more information.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
856
KBC
It sometimes is difficult to obtain successful convergence with stepped loading in a nonlinear transient
problem. If divergence is encountered, check if stepped loading was used by default, then determine if it is
appropriate for the analysis.
If you run an analysis with SOLCONTROL,ON, but do not issue the KBC command, ANSYS will choose
whether or not to use stepped or ramped loads. The program-chosen option will be recorded on the load
step files as KBC,-1.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Transient
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Freq and Substps
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time and Substps
Main Menu>Solution>Analysis Type>Sol'n Controls>Transient
Main Menu>Solution>Load Step Opts>Time/Frequenc>Freq and Substps
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time and Substps
KBETW, KP1, KP2, KPNEW, Type, VALUE
Creates a keypoint between two existing keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KP1
First keypoint. If KP1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI).
KP2
Second keypoint.
KPNEW
Number assigned to the new keypoint. Defaults to the lowest available keypoint number.
Type
Type of input for VALUE.
RATIO
Value is the ratio of the distances between keypoints as follows: (KP1-KPNEW)/(KP1-KP2).
DIST
Value is the absolute distance between KP1 and KPNEW (valid only if current coordinate system is
Cartesian).
VALUE
Location of new keypoint, as defined by Type (defaults to 0.5). If VALUE is a ratio (Type = RATIO) and
is less than 0 or greater than 1, the keypoint is created on the extended line. Similarly, if VALUE is a
distance (Type = DIST) and is less than 0 or greater than the distance between KP1 and KP2, the key-
point is created on the extended line.
857
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KBETW
Notes
Placement of the new keypoint depends on the currently active coordinate system [CSYS]. If the coordinate
system is Cartesian, the keypoint will lie on a straight line between KP1 and KP2. If the system is not Cartesian
(e.g., cylindrical, spherical, etc.), the keypoint will be located as if on a line (which may not be straight) created
in the current coordinate system between KP1 and KP2. Note that solid modeling in a toroidal coordinate
system is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Keypoints>KP between KPs
KCALC, KPLAN, MAT, KCSYM, KLOCPR
Calculates stress intensity factors in fracture mechanics analyses.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
KPLAN
Key to indicate stress state for calculation of stress intensity factors:
0
Plane strain and axisymmetric condition (default).
1
Plane stress condition.
MAT
Material number used in the extrapolation (defaults to 1).
KCSYM
Symmetry key:
0 or 1
Half-crack model with symmetry boundary conditions [DSYM] in the crack-tip coordinate system. K
II
= K
III
= 0. Three nodes are required on the path.
2
Like 1 except with antisymmetric boundary conditions (K
I
= 0).
3
Full-crack model (both faces). Five nodes are required on the path (one at the tip and two on each
face).
KLOCPR
Local displacements print key:
0
Do not print local crack-tip displacements.
1
Print local displacements used in the extrapolation technique.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
858
KCALC
Notes
Calculates the stress intensity factors (K
I
, K
II
, and K
III
) associated with homogeneous isotropic linear elastic
fracture mechanics. A displacement extrapolation method is used in the calculation (see POST1 - Crack
Analysis in the Theory Reference for the Mechanical APDL and Mechanical Applications). This method assumes
that the displacement calculations are for the plane strain state. If the displacement calculations are performed
using a plane stress formulation, the calculation of the stress intensity factors can be converted to the plane
strain state by using KPLAN = 1. ANSYS Uses minor Poisson's ratio (MP,NUXY) for the stress intensity factor
calculation, therefore the material's Poisson's ratio must be defined using MP,NUXY command. The PATH
and PPATH commands must be used to define a path with the crack face nodes (NODE1 at the crack tip,
NODE2 and NODE3 on one face, NODE4 and NODE5 on the other (optional) face). A crack-tip coordinate
system, having x parallel to the crack face (and perpendicular to the crack front) and y perpendicular to the
crack face, must be the active RSYS and CSYS before KCALC is issued.
Menu Paths
Main Menu>General Postproc>Nodal Calcs>Stress Int Factr
KCENTER, Type, VAL1, VAL2, VAL3, VAL4, KPNEW
Creates a keypoint at the center of a circular arc defined by three locations.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Type of entity used to define the circular arc. The meaning of VAL1 through VAL4 will vary depending
on Type. If Type = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI).
KP
Arc is defined by keypoints.
LINE
Arc is defined by locations on a line.
VAL1, VAL2, VAL3, VAL4
Values used to specify three locations on the arc (see table below).
KPNEW
Number assigned to new keypoint. Defaults to the lowest available keypoint number.
Definitions:
If Type = KP, inputs VAL1 through VAL4 are defined as follows:
VAL1
First keypoint.
VAL2
Second keypoint.
VAL3
Third keypoint.
859
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KCENTER
VAL4
Arc radius. If VAL4 = 0 or blank (default), the arc is defined by the three keypoints specified as VAL1,
VAL2, and VAL3 and arc radius is not used. If VAL4 is nonzero, VAL1, VAL2, and VAL4 are used to
calculate the center point, and VAL3 is used to locate the center point as follows:
VAL4 > 0
Center point and VAL3 are on the same side of the line between the first two keypoints.
VAL4 < 0
Center point and VAL3 are on opposite sides of the line between the first two keypoints.
If Type = LINE, inputs VAL1 through VAL4 are defined as follows:
VAL1
Line number.
VAL2
Line ratio (0 to 1) indicating the first location (defaults to 0).
VAL3
Line ratio (0 to 1) indicating the second location (defaults to 0.5).
VAL4
Line ratio (0 to 1) indicating the third location (defaults to 1).
Notes
KCENTER should be used in the Cartesian coordinate system (CSYS,0) only. This command provides three
methods to define a keypoint at the center of three locations. As shown below, the center point can be
calculated based on a) three keypoints, b) three keypoints and a radius, or c) three locations on a line. Note
that for method c, if a circular line is specified by VAL1, VAL2 through VAL4 are not needed.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
860
KCENTER
VAL1
imaginary
circular arc
imaginary
circular arc
imaginary
circular arc
VAL1
VAL1
VAL1 (circular line)
VAL1 (arbitrary line)
VAL2
VAL2
VAL2
VAL2
VAL3
VAL3
VAL3
VAL3
VAL4
(+ radius)
VAL4
(- radius)
VAL4
KPNEW
KPNEW
KPNEW
KPNEW
KPNEW
(a) Three keypoints
(b) Three keypoints and a radius
(c) Locations on line
imaginary
circular arc
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Keypoints>KP at center>3 keypoints
Main Menu>Preprocessor>Modeling>Create>Keypoints>KP at center>3 KPs and radius
Main Menu>Preprocessor>Modeling>Create>Keypoints>KP at center>Location on line
KCLEAR, NP1, NP2, NINC
Deletes nodes and point elements associated with selected keypoints.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Delete mesh for keypoints NP1 to NP2 (defaults to NP1) in steps of NINC (defaults to 1). If NP1 = ALL,
NP2 and NINC are ignored and the mesh for all selected keypoints [KSEL] is deleted. If NP1 = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NP1.
861
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KCLEAR
Notes
Deletes all nodes and point elements associated with selected keypoints (regardless of whether the nodes
or elements are selected). Nodes associated with non-point elements will not be deleted. Attributes assigned
as a result of KATT are maintained. In the program's response to the command, if a keypoint is tallied as
"cleared," it means either its node or element reference was deleted.
Menu Paths
Main Menu>Preprocessor>Meshing>Clear>Keypoints
KDELE, NP1, NP2, NINC
Deletes unmeshed keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Delete keypoints from NP1 to NP2 (defaults to NP1) in steps of NINC (defaults to 1). If NP1 = ALL, NP2
and NINC are ignored and all selected keypoints [KSEL] are deleted. If NP1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI). A component name may
also be substituted for NP1 (NP2 and NINC are ignored).
Notes
Deletes selected keypoints. A keypoint attached to a line cannot be deleted unless the line is first deleted.
Menu Paths
Main Menu>Preprocessor>Modeling>Delete>Keypoints
Main Menu>Preprocessor>Modeling>Topo Repair>Delete>Keypoints
KDIST, KP1, KP2
Calculates and lists the distance between two keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KP1
First keypoint in distance calculation. If KP1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
KP2
Second keypoint in distance calculation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
862
KDELE
Notes
KDIST lists the distance between keypoints KP1 and KP2, as well as the current coordinate system offsets
from KP1 to KP2, where the X, Y, and Z locations of KP1 are subtracted from the X, Y, and Z locations of
KP2 (respectively) to determine the offsets. KDIST is valid in any coordinate system except toroidal [CSYS,3].
KDIST returns a variable, called "_RETURN," which contains the distance value. You can use this value for
various purposes; for example, to set the default number of line divisions to be generated along region
boundary lines [ESIZE,_RETURN]. In interactive mode, you can access this command by using the Model
Query Picker (Utility Menu> List> Picked Entities), where you can also access automatic annotation functions,
and display the value on your model.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Check Geom>KP distances
KEEP, Key
Stores POST26 definitions and data during active session.
POST26: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Key
State or value
On or 1
Allows you to exit and reenter /POST26 without losing your current time history variable information.
Keeps a cache of the /POST26 variable information including the active file name (FILE), variable
definitions (NSOL, ESOL, GAPF, RFORCE, SOLU, and EDREAD) and stored variable data in memory
for the current ANSYS session.
Off or 0
/POST26 variable information is deleted when you exit /POST26.
Command Default
ON - Hold time history information in memory. You can, for example, move back and forth between /POST1
and /POST26 without redefining and storing the time history variables each time you enter /POST26.
Notes
Your variable information is saved in memory only for the current active ANSYS session. It is deleted when
you exit ANSYS. This information is also deleted when you issue /CLEAR, RESUME, SOLVE, or RESET.
When you reenter /POST26 all time history variable data is available for use. When you issue STORE,NEW,
variable definitions created by math operations such as ADD or PROD will not be restored. However, variables
defined with NSOL, ESOL, GAPF, RFORCE, SOLU, and EDREAD will be restored. Only the last active results
file name is kept in memory (FILE).
863
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KEEP
Commands such as LAYERP26, SHELL, and FORCE that specify the location or a component of data to be
stored will retain the setting at the time of exiting /POST26 .
Menu Paths
Main Menu>TimeHist Postpro>Settings>Data
KESIZE, NPT, SIZE, FACT1, FACT2
Specifies the edge lengths of the elements nearest a keypoint.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPT
Number of the keypoint whose lines will be adjusted. If ALL, use all selected keypoints [KSEL]. If NPT =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
SIZE
Size of elements along lines nearest keypoint NPT (overrides any other specified size). If SIZE is zero
(or blank), use FACT1 or FACT2.
FACT1
Scale factor applied to a previously defined SIZE. Not used if SIZE is input.
FACT2
Scale factor applied to the minimum element division at keypoint NPT for any attached line. This feature
is useful with adaptive mesh refinement. Not used if SIZE or FACT1 is input.
Notes
Affects only the line divisions adjacent to the keypoint on lines not previously assigned divisions by other
line commands [LESIZE, etc.]. The remaining line divisions are determined from the division nearest the
keypoint at the other end of the line (specified by another KESIZE command or the ESIZE command). Divisions
are transferred to the lines during the mesh operation. If smart element sizing is being used [SMRTSIZE],
KESIZE specifications may be overridden (i.e., a smaller element size may be used) to accommodate curvature
and small features.
This command is valid in any processor. The command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Keypoints>All KPs
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Keypoints>Clr Size
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Keypoints>Picked KPs
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
864
KESIZE
KEYOPT, ITYPE, KNUM, VALUE
Sets element key options.
PREP7: Element Type
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITYPE
Element type number as defined on the ET command.
KNUM
Number of the KEYOPT to be defined (KEYOPT(KNUM)).
VALUE
Value of this KEYOPT.
Notes
Alternative to inputting KEYOPT values on ET command. Must be used if KEYOPT(7) or greater values are
to be input. ITYPE must first be defined with the ET command.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Flow Environment>FLOTRAN Coor Sys
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Flow Environment>FLOTRAN Coor Sys
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
KEYPTS
Specifies "Keypoints" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
This command cannot be accessed from a menu.
865
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KEYPTS
KEYW, Keyword, KEY
Sets a keyword used by the GUI for context filtering (GUI).
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Keyword
A keyword which, when set to either true or false, changes the behavior of the GUI.
KEY
Keyword switch:
0
Sets the keyword to "false."
1
Sets the keyword to "true."
Notes
Defines a keyword used by the GUI for context filtering. This is a command generated by the GUI and may
appear in the log file (Jobname.LOG) if the GUI is used. This command is usually not typed in directly in
an ANSYS session.
This command is valid in any processor.
Menu Paths
Main Menu>Preferences
KFILL, NP1, NP2, NFILL, NSTRT, NINC, SPACE
Generates keypoints between two keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2
Beginning and ending keypoints for fill-in. NP1 defaults to next to last keypoint specified, NP2 defaults
to last keypoint specified. If NP1 = P, graphical picking is enabled and all remaining command fields are
ignored (valid only in the GUI).
NFILL
Fill NFILL keypoints between NP1 and NP2 (defaults to |NP2-NP1|-1). NFILL must be positive.
NSTRT
Keypoint number assigned to first filled-in keypoint (defaults to NP1 + NINC).
NINC
Add this increment to each of the remaining filled-in keypoint numbers (may be positive or negative).
Defaults to (NP2-NP1)/(NFILL + 1), i.e., linear interpolation.
SPACE
Spacing ratio. Ratio of last division size to first division size. If > 1.0, divisions increase. If < 1.0, divisions
decrease. Ratio defaults to 1.0 (uniform spacing).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
866
KEYW
Notes
Generates keypoints (in the active coordinate system) between two existing keypoints. The two keypoints
may have been defined in any coordinate system. However, solid modeling in a toroidal coordinate system
is not recommended. Any number of keypoints may be filled in and any keypoint numbering sequence may
be assigned.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Keypoints>Fill between KPs
KGEN, ITIME, NP1, NP2, NINC, DX, DY, DZ, KINC, NOELEM, IMOVE
Generates additional keypoints from a pattern of keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME
Do this generation operation a total of ITIME times, incrementing all keypoints in the given pattern
automatically (or by KINC) each time after the first. ITIME must be more than 1 for generation to occur.
NP1, NP2, NINC
Generate keypoints from the pattern of keypoints beginning with NP1 to NP2 (defaults to NP1) in steps
of NINC (defaults to 1). If NP1 = ALL, NP2 and NINC are ignored and the pattern is all selected keypoints
[KSEL]. If NP1 is negative, NP2 and NINC are ignored and the last |NP1| keypoints (in sequence from
the highest keypoint number) are used as the pattern to be repeated. If NP1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI). A component name may
also be substituted for NP1 (NP2 and NINC are ignored).
DX, DY, DZ
Keypoint location increments in the active coordinate system (DR, D, DZ for cylindrical, DR, D, D for
spherical).
KINC
Keypoint increment between generated sets. If zero, the lowest available keypoint numbers are assigned
[NUMSTR].
NOELEM
Specifies if elements and nodes are also to be generated:
0
Generate nodes and point elements associated with the original keypoints, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether keypoints will be moved or newly defined:
0
Generate additional keypoints as requested with the ITIME argument.
867
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KGEN
1
Move original keypoints to new position retaining the same keypoint numbers (ITIME, KINC, and
NOELEM are ignored). Valid only if the old keypoints are no longer needed at their original positions.
Corresponding meshed items are also moved if not needed at their original position.
Notes
Generates additional keypoints (and corresponding mesh) from a given keypoint pattern. The MAT, TYPE,
REAL, and ESYS attributes are based upon the keypoints in the pattern and not upon the current settings.
Generation is done in the active coordinate system. Keypoints in the pattern may have been defined in any
coordinate system. However, solid modeling in a toroidal coordinate system is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Keypoints
KL, NL1, RATIO, NK1
Generates a keypoint at a specified location on an existing line.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the line. If negative, the direction of line (as interpreted for RATIO) is reversed. If NL1 = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
RATIO
Ratio of line length to locate keypoint. Must be between 0.0 and 1.0. Defaults to 0.5 (divide the line in
half ).
NK1
Number to be assigned to keypoint generated at division location (defaults to lowest available keypoint
number [NUMSTR]).
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Keypoints>On Line
Main Menu>Preprocessor>Modeling>Create>Keypoints>On Line w/Ratio
KLIST, NP1, NP2, NINC, Lab
Lists the defined keypoints or hard points.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
List keypoints from NP1 to NP2 (defaults to NP1) in steps of NINC (defaults to 1). If NP1 = ALL (default),
NP2 and NINC are ignored and all selected keypoints [KSEL] are listed. If NP1 = P, graphical picking is
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
868
KL
enabled and all remaining command fields are ignored (valid only in the GUI). A component name may
also be substituted for NP1 (NP2 and NINC are ignored).
Lab
Coordinate listing key:
(blank)
List all keypoint information.
COORD
Suppress all but the keypoint coordinates (shown to a higher degree of accuracy than when displayed
with all information).
HPT
List only hard point information.
Notes
Lists keypoints in the active display coordinate system [DSYS]. An attribute (TYPE, MAT, REAL, or ESYS) listed
as a zero is unassigned; one listed as a positive value indicates that the attribute was assigned with the
KATT command (and will not be reset to zero if the mesh is cleared); one listed as a negative value indicates
that the attribute was assigned using the attribute pointer [TYPE, MAT, REAL, or ESYS] that was active
during meshing (and will be reset to zero if the mesh is cleared).
This command is valid in any processor.
Menu Paths
Utility Menu>List>Keypoints>Coordinates +Attributes
Utility Menu>List>Keypoints>Coordinates only
Utility Menu>List>Keypoints>Hard Points
KMESH, NP1, NP2, NINC
Generates nodes and point elements at keypoints.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Mesh keypoints from NP1 to NP2 (defaults to NP1) in steps of NINC (defaults to 1). If NP1 = ALL, NP2
and NINC are ignored and all selected keypoints [KSEL] are meshed. If NP1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI). A component name may
also be substituted for NP1.
Notes
Missing nodes required for the generated elements are created and assigned the lowest available numbers.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Keypoints
869
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KMESH
KMODIF, NPT, X, Y, Z
Modifies an existing keypoint.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPT
Modify coordinates of this keypoint. If NPT = ALL, modify coordinates of all selected keypoints [KSEL].
If NPT = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the
GUI). A component name may also be substituted for NPT.
X, Y, Z
Replace the previous coordinate values assigned to this keypoint with these corresponding coordinate
values. Values are interpreted according to the active coordinate system (R, , Z for cylindrical, R, , for
spherical). If X = P, graphical picking is used to locate keypoint and Y and Z are ignored. A blank retains
the previous value. You cannot specify Y = P.
Notes
Lines, areas, and volumes attached to the modified keypoint (if any) must all be selected and will be redefined
using the active coordinate system. However, solid modeling in a toroidal coordinate system is not recom-
mended.
Caution
Redefined entities may be removed from any defined components and assemblies. Nodes and
elements will be automatically cleared from any redefined keypoints, lines, areas, or volumes.
The KMODIF command moves keypoints for geometry modification without validating underlying entities.
To merge keypoints and update higher order entities, issue the NUMMRG command instead.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Keypoints>Set of KPs
Main Menu>Preprocessor>Modeling>Move / Modify>Keypoints>Single KP
KMOVE, NPT, KC1, X1, Y1, Z1, KC2, X2, Y2, Z2
Calculates and moves a keypoint to an intersection.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPT
Move this keypoint. If NPT = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NPT.
KC1
First coordinate system number. Defaults to 0 (global Cartesian).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
870
KMODIF
X1, Y1, Z1
Input one or two values defining the location of the keypoint in this coordinate system. Input "U" for
unknown value(s) to be calculated and input "E" to use an existing coordinate value. Fields are R1, 1,
Z1 for cylindrical, or R1, 1, 1 for spherical.
KC2
Second coordinate system number.
X2, Y2, Z2
Input two or one value(s) defining the location of the keypoint in this coordinate system. Input "U" for
unknown value(s) to be calculated and input "E" to use an existing coordinate value. Arguments are R2,
2, Z2 for cylindrical, or R2, 2, 2 for spherical.
Notes
Calculates and moves a keypoint to an intersection location. The keypoint must have been previously defined
(at an approximate location) or left undefined (in which case it is internally defined at the SOURCE location).
The actual location is calculated from the intersection of three surfaces (implied from three coordinate
constants in two different coordinate systems). Note that solid modeling in a toroidal coordinate system is
not recommended. See the MOVE command for surface and intersection details. The three (of six) constants
easiest to define should be used. The program will calculate the remaining three coordinate constants. All
arguments, except KC1, must be input. Use the repeat command [*REPEAT] after the KMOVE command to
move a series of keypoints, if desired.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Keypoints>To Intersect
KNODE, NPT, NODE
Defines a keypoint at an existing node location.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPT
Arbitrary reference number for keypoint. If zero, the lowest available number is assigned [NUMSTR].
NODE
Node number defining global X, Y, Z keypoint location. If NODE = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI). A component name may also be substituted
for NODE.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Keypoints>On Node
871
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KNODE
KPLOT, NP1, NP2, NINC, Lab
Displays the selected keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Display keypoints from NP1 to NP2 (defaults to NP1) in steps of NINC (defaults to 1). If NP1 = ALL (de-
fault), NP2 and NINC are ignored and all selected keypoints [KSEL] are displayed.
Lab
Determines what keypoints are plotted (one of the following):
(blank)
Plots all keypoints.
HPT
Plots only those keypoints that are hard points.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Keypoints>Hardpoints
Utility Menu>Plot>Keypoints>Keypoints
Utility Menu>Plot>Specified Entities>Keypoints
KPSCALE, NP1, NP2, NINC, RX, RY, RZ, KINC, NOELEM, IMOVE
Generates a scaled set of (meshed) keypoints from a pattern of keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Set of keypoints (NP1 to NP2 in steps of NINC) that defines the pattern to be scaled. NP2 defaults to
NP1, NINC defaults to 1. If NP1 = ALL, NP2 and NINC are ignored and the pattern is defined by all se-
lected keypoints. If NP1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NP1 (NP2 and NINC are ignored).
RX, RY, RZ
Scale factors to be applied to the X, Y, Z keypoint coordinates in the active coordinate system (RR, R,
RZ for cylindrical; RR, R, R for spherical). The R and R scale factors are interpreted as angular offsets.
For example, if CSYS = 1, an RX, RY, RZ input of (1.5,10,3) would scale the specified keypoints 1.5 times
in the radial and 3 times in the Z direction, while adding an offset of 10 degrees to the keypoints.) Zero,
blank, or negative scale factor values are assumed to be 1.0. Zero or blank angular offsets have no effect.
KINC
Increment to be applied to the keypoint numbers for generated set. If zero, the lowest available keypoint
numbers will be assigned [NUMSTR].
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
872
KPLOT
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Nodes and point elements associated with the original keypoints will be generated (scaled) if they
exist.
1
Nodes and point elements will not be generated.
IMOVE
Specifies whether keypoints will be moved or newly defined:
0
Additional keypoints will be generated.
1
Original keypoints will be moved to new position (KINC and NOELEM are ignored). Use only if the
old keypoints are no longer needed at their original positions. Corresponding meshed items are also
moved if not needed at their original position.
Notes
Generates a scaled set of keypoints (and corresponding mesh) from a pattern of keypoints. The MAT, TYPE,
REAL, and ESYS attributes are based on the keypoints in the pattern and not the current settings. Scaling is
done in the active coordinate system. Keypoints in the pattern could have been generated in any coordinate
system. However, solid modeling in a toroidal coordinate system is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Scale>Keypoints
KREFINE, NP1, NP2, NINC, LEVEL, DEPTH, POST, RETAIN
Refines the mesh around specified keypoints.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NP1, NP2, NINC
Keypoints (NP1 to NP2 in increments of NINC) around which the mesh is to be refined. NP2 defaults
to NP1, and NINC defaults to 1. If NP1 = ALL, NP2 and NINC are ignored and all selected keypoints are
used for refinement. If NP1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NP1 (NP2 and NINC are ignored).
LEVEL
Amount of refinement to be done. Specify the value of LEVEL as an integer from 1 to 5, where a value
of 1 provides minimal refinement, and a value of 5 provides maximum refinement (defaults to 1).
DEPTH
Depth of mesh refinement in terms of the number of elements outward from the indicated keypoints
(defaults to 1).
POST
Type of postprocessing to be done after element splitting, in order to improve element quality:
873
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KREFINE
OFF
No postprocessing will be done.
SMOOTH
Smoothing will be done. Node locations may change.
CLEAN
Smoothing and cleanup will be done. Existing elements may be deleted, and node locations may
change (default).
RETAIN
Flag indicating whether quadrilateral elements must be retained in the refinement of an all-quadrilateral
mesh. (The ANSYS program ignores the RETAIN argument when you are refining anything other than
a quadrilateral mesh.)
ON
The final mesh will be composed entirely of quadrilateral elements, regardless of the element quality
(default).
OFF
The final mesh may include some triangular elements in order to maintain element quality and
provide transitioning.
Notes
KREFINE performs local mesh refinement around the specified keypoints. By default, the indicated elements
are split to create new elements with 1/2 the edge length of the original elements (LEVEL = 1).
KREFINE refines all area elements and tetrahedral volume elements that are adjacent to the specified key-
points. Any volume elements that are adjacent to the specified keypoints, but are not tetrahedra (for example,
hexahedra, wedges, and pyramids), are not refined.
You cannot use mesh refinement on a solid model that contains initial conditions at nodes [IC], coupled
nodes [CP family of commands], constraint equations [CE family of commands], or boundary conditions or
loads applied directly to any of its nodes or elements. This applies to nodes and elements anywhere in the
model, not just in the region where you want to request mesh refinement. See Revising Your Model in the
Modeling and Meshing Guide for additional restrictions on mesh refinement.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Keypoints
KSCALE, KINC, NP1, NP2, NINC, RX, RY, RZ
Generates a scaled pattern of keypoints from a given keypoint pattern.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KINC
Do this scaling operation one time, incrementing all keypoints in the given pattern by KINC. If KINC =
0, keypoints will be redefined at the scaled locations.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
874
KSCALE
NP1, NP2, NINC
Scale keypoints from pattern beginning with NP1 to NP2 (defaults to NP1) in steps of NINC (defaults
to 1). If NP1 = ALL, NP2 and NINC are ignored and pattern is all selected keypoints [KSEL]. If NP1 = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NP1 (NP2 and NINC are ignored).
RX, RY, RZ
Scale factor ratios. Scaling is relative to the origin of the active coordinate system (RR, R, RZ for cylindrical,
RR, R, R for spherical). If > 1.0, pattern is enlarged. If < 1.0, pattern is reduced. Ratios each default to
1.0.
Notes
Generates a scaled pattern of keypoints from a given keypoint pattern. Scaling is done in the active coordinate
system (see analogous node scaling [NSCALE]). Solid modeling in a toroidal coordinate system is not recom-
mended.
Menu Paths
This command cannot be accessed from a menu.
KSCON, NPT, DELR, KCTIP, NTHET, RRAT
Specifies a keypoint about which an area mesh will be skewed.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPT
Keypoint number at concentration. If NPT = ALL, use all selected keypoints. If remaining fields are blank,
remove concentration from this keypoint (if unmeshed). If NPT = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI). A component name may also be substituted
for NPT.
DELR
Radius of first row of elements about keypoint.
KCTIP
Crack tip singularity key:
0
Do not skew midside nodes, if any, within the element.
1
Skew midside nodes of the first row of elements to the 1/4 point for crack tip singularity.
NTHET
Number of elements in circumferential direction (defaults to approximately one per 45 (or one per 30,
if KCTIP = 1)).
RRAT
Ratio of 2nd row element size to DELR (defaults to 0.75, or 0.5 if KCTIP = 1).
875
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KSCON
Notes
Defines a concentration keypoint about which an area mesh will be skewed. Useful for modeling stress
concentrations and crack tips. During meshing, elements are initially generated circumferentially about, and
radially away, from the keypoint. Lines attached to the keypoint are given appropriate divisions and spacing
ratios. Only one concentration keypoint per unmeshed area is allowed. Use KSCON,STAT to list current status
of concentration keypoints. The KSCON command does not support 3-D modeling.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>Concentrat KPs>Create
Main Menu>Preprocessor>Meshing>Size Cntrls>Concentrat KPs>List
KSEL, Type, Item, Comp, VMIN, VMAX, VINC, KABS
Selects a subset of keypoints or hard points.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Restore the full set.
NONE
Unselect the full set.
INVE
Invert the current set (selected becomes unselected and vice versa).
STAT
Display the current select status.
The following fields are used only with Type = S, R, A, or U:
Item
Label identifying data. Valid item labels are shown in the table below. Some items also require a com-
ponent label. If Item = PICK (or simply P), graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI). Defaults to KP.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
876
KSEL
Comp
Component of the item (if required). Valid component labels are shown in the table below.
VMIN
Minimum value of item range. Ranges are keypoint numbers, coordinate values, attribute numbers, etc.,
as appropriate for the item. A component name (as specified on the CM command) may also be substi-
tuted for VMIN (VMAX and VINC are ignored). If Item = MAT, TYPE, REAL, or ESYS and if VMIN is positive,
the absolute value of Item is compared against the range for selection; if VMIN is negative, the signed
value of Item is compared. See the KLIST command for a discussion of signed attributes.
VMAX
Maximum value of item range. VMAX defaults to VMIN.
VINC
Value increment within range. Used only with integer ranges (such as for keypoint numbers). Defaults
to 1. VINC cannot be negative.
KABS
Absolute value key:
0
Check sign of value during selection.
1
Use absolute value during selection (sign ignored).
Command Default
All keypoints are selected.
Notes
Selects a subset of keypoints or hard points. For example, to select a new set of keypoints based on keypoint
numbers 1 through 7, use KSEL,S,KP,,1,7. The selected subset is used when the ALL label is entered (or implied)
on other commands, such as KLIST,ALL. Only data identified by keypoint number are selected. Data are
flagged as selected and unselected; no data are actually deleted from the database.
This command is valid in any processor.
For selections based on non-integer numbers (coordinates, results, etc.), items that are within the range
VMIN -Toler and VMAX + Toler are selected. The default tolerance Toler is based on the relative values
of VMIN and VMAX as follows:
If VMIN = VMAX, Toler = 0.005 x VMIN.
If VMIN = VMAX = 0.0, Toler = 1.0E-6.
If VMAX VMIN, Toler = 1.0E-8 x (VMAX - VMIN).
877
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KSEL
Use the SELTOL command to override this default and specify Toler explicitly.
Table 225 KSEL - Valid Item and Component Labels
Valid Item and Component Labels KSEL, Type, Item, Comp, VMIN, VMAX, VINC,
KABS
Description Comp Item
Keypoint number. KP
Keypoint numbers on exterior of selected lines (ignore remaining
fields).
EXT
Hard point number. HPT
X, Y, or Z location in the active coordinate system. X,Y,Z LOC
Material number associated with the keypoint. MAT
Element type number associated with the keypoint. TYPE
Real constant set number associated with the keypoint. REAL
Element coordinate system associated with the keypoint. ESYS
Menu Paths
Utility Menu>Select>Entities
KSLL, Type
Selects those keypoints contained in the selected lines.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of keypoint select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
878
KSLL
KSLN, Type
Selects those keypoints associated with the selected nodes.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of keypoint select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Notes
Valid only if the nodes were generated by a meshing operation [KMESH, LMESH, AMESH, VMESH] on a
solid model that contains the associated keypoints.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
KSUM
Calculates and prints geometry statistics of the selected keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Calculates and prints geometry statistics (centroid location, moments of inertia, etc.) associated with the
selected keypoints. Geometry items are reported in the global Cartesian coordinate system. A unit density
is assumed, irrespective of any material associations [KATT, MAT]. Items calculated by KSUM and later re-
trieved by a *GET or *VGET command are valid only if the model is not modified after the KSUM command
is issued.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Calc Geom Items>Of Keypoints
879
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KSUM
KSYMM, Ncomp, NP1, NP2, NINC, KINC, NOELEM, IMOVE
Generates a reflected set of keypoints.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Ncomp
Symmetry key:
X
X (or R) symmetry (default).
Y
Y (or ) symmetry.
Z
Z (or ) symmetry.
NP1, NP2, NINC
Reflect keypoints from pattern beginning with NP1 to NP2 (defaults to NP1) in steps of NINC (defaults
to 1). If NP1 = ALL, NP2 and NINC are ignored and pattern is all selected keypoints [KSEL]. If Ncomp =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NP1 (NP2 and NINC are ignored).
KINC
Keypoint increment between sets. If zero, the lowest available keypoint numbers are assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Generate nodes and point elements associated with the original keypoints, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether keypoints will be moved or newly defined:
0
Generate additional keypoints.
1
Move original keypoints to new position retaining the same keypoint numbers (KINC and NOELEM
are ignored). Valid only if the old keypoints are no longer needed at their original positions. Corres-
ponding meshed items are also moved if not needed at their original position.
Notes
Generates a reflected set of keypoints (and corresponding mesh) from a given keypoint pattern by a symmetry
reflection (see analogous node symmetry command, NSYM). The MAT, TYPE, REAL, and ESYS attributes are
based upon the keypoints in the pattern and not upon the current settings. Reflection is done in the active
coordinate system by changing a particular coordinate sign. Keypoints in the pattern may have been generated
in any coordinate system. However, solid modeling in a toroidal coordinate system is not recommended.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
880
KSYMM
Menu Paths
Main Menu>Preprocessor>Modeling>Reflect>Keypoints
KTRAN, KCNTO, NP1, NP2, NINC, KINC, NOELEM, IMOVE
Transfers a pattern of keypoints to another coordinate system.
PREP7: Keypoints
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCNTO
Reference number of coordinate system where the pattern is to be transferred. Transfer occurs from the
active coordinate system.
NP1, NP2, NINC
Transfer keypoints from pattern beginning with NP1 to NP2 (defaults to NP1) in steps of NINC (defaults
to 1). If NP1 = ALL, NP2 and NINC are ignored and pattern is all selected keypoints [KSEL]. If NP1 = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NP1 (NP2 and NINC are ignored).
KINC
Keypoint increment between sets. If zero, the lowest available keypoint numbers are assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Generate nodes and point elements associated with the original keypoints, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether keypoints will be moved or newly defined:
0
Generate additional keypoints.
1
Move original keypoints to new position retaining the same keypoint numbers (KINC and NOELEM
are ignored). Valid only if the old keypoints are no longer needed at their original positions. Corres-
ponding meshed items are also moved if not needed at their original position.
Notes
Transfers a pattern of keypoints (and corresponding mesh) from one coordinate system to another (see
analogous node transfer command, TRANSFER). The MAT, TYPE, REAL, and ESYS attributes are based upon
the keypoints in the pattern and not upon the current settings. Coordinate systems may be translated and
rotated relative to each other. Initial pattern may be generated in any coordinate system. Coordinate values
are interpreted in the active coordinate system and are transferred directly. Solid modeling in a toroidal
coordinate system is not recommended.
881
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KTRAN
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Transfer Coord>Keypoints
KUSE, KEY
Specifies whether or not to reuse the factorized matrix.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
KEY
Reuse key:
0
Program decides whether or not to reuse the previous factorized stiffness matrix.
1
Force the previous factorized stiffness matrix to be reused. Used mainly in a restart. Forcing reuse
of the matrix is a nonstandard use of the program, and should be done with caution. For instance,
using this option and changing the number of elements, or the number or type of degrees of freedom,
may cause an abort.
-1
All element matrices are reformed and are used to reform a new factorized stiffness matrix.
Command Default
Program makes decision.
Notes
Overrides the program logic to determine whether or not to reuse the previous factorized stiffness matrix
for each substep of this load step. Applies only to static or full transient analyses and to full harmonic analyses
if the frequency is not changed for continuing loadsteps. For full harmonic analyses, only KEY = 1 or KEY
= 0 is valid.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Reuse LN22 Matrix
Main Menu>Solution>Load Step Opts>Other>Reuse LN22 Matrix
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
882
KUSE
KWPAVE, P1, P2, P3, P4, P5, P6, P7, P8, P9
Moves the working plane origin to the average location of keypoints.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1, P2, P3, . . . , P9
Keypoints used in calculation of the average. At least one must be defined. If P1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI).
Notes
Moves the origin of the working plane to the average of the specified keypoints. Averaging is based on the
active coordinate system.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Offset WP to>Keypoints
KWPLAN, WN, KORIG, KXAX, KPLAN
Defines the working plane using three keypoints.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number whose viewing direction will be modified to be normal to the working plane (defaults
to 1). If WN is a negative value, the viewing direction will not be modified. If fewer than three points are
used, the viewing direction of window WN will be used instead to define the normal to the working
plane.
KORIG
Keypoint number defining the origin of the working plane coordinate system. If KORIG = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI).
KXAX
Keypoint number defining the x-axis orientation (defaults to the x-axis being parallel to the global X-
axis; or if the normal to the working plane is parallel to the global X-axis, then defaults to being parallel
to the global Y-axis).
KPLAN
Keypoint number defining the working plane (the normal defaults to the present display view [/VIEW]
of window WN).
Notes
Defines a working plane to assist in picking operations using three keypoints as an alternate to the WPLANE
command. The three keypoints also define the working plane coordinate system. A minimum of one keypoint
883
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
KWPLAN
(at the working plane origin) is required. Immediate mode may also be active. See WPSTYL command to
set the style of working plane display.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Align WP with>Keypoints
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
884
KWPLAN
L Commands
L, P1, P2, NDIV, SPACE, XV1, YV1, ZV1, XV2, YV2, ZV2
Defines a line between two keypoints.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1
Keypoint at the beginning of line. If P1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
P2
Keypoint at the end of line.
NDIV
Number of element divisions within this line. Normally this field is not used; specifying divisions with
LESIZE, etc. is recommended.
SPACE
Spacing ratio. Normally this field is not used, as specifying spacing ratios with the LESIZE command is
recommended. If positive, SPACE is the nominal ratio of the last division size (at P2) to the first division
size (at P1). If the ratio is greater than 1, the division sizes increase from P1 to P2, and if less than 1,
they decrease. If SPACE is negative, then |SPACE| is the nominal ratio of the center division size to those
at the ends.
The following fields are used only if specified end slopes on the line are desired, otherwise zero curvature
end slopes will be automatically calculated to produce a line which is "straight" in the active coordinate
system. To specify end slopes, use the following fields to define a "slope vector" (one for each end of the
line, if desired) that has its tail at the origin and its head at the point XV, YV, ZV in the active coordinate
system [CSYS]. The corresponding end slope of the line will then be parallel to this "slope vector."
XV1, YV1, ZV1
Location (in the active coordinate system) of the head of the "slope vector" corresponding to the slope
at the P1 end of the line. The tail of the vector is at the origin of the coordinate system.
XV2, YV2, ZV2
Location of the head of the "slope vector" corresponding to the slope at the P2 end of the line.
Notes
Defines a line between two keypoints from P1 to P2. The line shape may be generated as "straight" (in the
active coordinate system) or curved. The line shape is invariant with coordinate system after it is generated.
Note that solid modeling in a toroidal coordinate system is not recommended. A curved line is limited to
180. Lines may be redefined only if not yet attached to an area.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>In Active Coord
885
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
L2ANG, NL1, NL2, ANG1, ANG2, PHIT1, PHIT2
Generates a line at an angle with two existing lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the first line to be hit (touched by the end of the new line). If negative, assume P1 (see below)
is the second keypoint of the line instead of the first. If NL1 = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI).
NL2
Number of the second line to be hit. If negative, assume P3 is the second keypoint of the line instead
of the first.
ANG1
Angle of intersection (usually zero or 180) of generated line with tangent to first line.
ANG2
Angle of intersection (usually zero or 180) of generated line with tangent to second line.
PHIT1
Number to be assigned to keypoint generated at hit location on first line (defaults to lowest available
keypoint number [NUMSTR]).
PHIT2
Number to be assigned to keypoint generated at hit location on second line (defaults to lowest available
keypoint number [NUMSTR]).
Notes
Generates a straight line (PHIT1-PHIT2) at an angle (ANG1) with an existing line NL1 (P1-P2) and which
is also at an angle (ANG2) with another existing line NL2 (P3-P4). If the angles are zero the generated line
is tangent to the two lines. The PHIT1 and PHIT2 locations on the lines are automatically calculated. Line
P1-P2 becomes P1-PHIT1, P3-P4 becomes P3-PHIT2, and new lines PHIT1-P2, PHIT2-P4, and
PHIT1-PHIT2 are generated. Line divisions are set to zero (use LESIZE, etc. to modify).
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Angle to 2 Lines
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Norm to 2 Lines
L2TAN, NL1, NL2
Generates a line tangent to two lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the first line generated line is tangent to. If negative, assume P1 (see below) is the second
keypoint of the line instead of the first. If NL1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
886
L2ANG
NL2
Number of the second line generated line is tangent to. If negative, assume P3 is the second keypoint
of the line instead of the first.
Notes
Generates a line (P2-P3) tangent at point P2 to line NL1 (P1-P2) and tangent at point P3 to line NL2
(P3-P4).
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Tan to 2 Lines
LANG, NL1, P3, ANG, PHIT, LOCAT
Generates a straight line at an angle with a line.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the line to be hit (touched by the end of the new line). If negative, assume P1 (see below)
is the second keypoint of the line instead of the first. If NL1 = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI).
P3
Keypoint at which generated line must end.
ANG
Angle of intersection of generated line PHIT-P3 with tangent to line P1-P2 at PHIT. If 0 (default), the
generated line is tangent to NL1 toward end P1; if 90, the generated line is perpendicular to NL1. If 180,
the generated line is tangent to NL1 toward end P2. ANG can be any value, but is adjusted to the corres-
ponding acute angle with respect to LOCAT. See "Notes" (p. 887) for a discussion of accuracy.
PHIT
Number to be assigned to keypoint generated at hit location (defaults to lowest available keypoint
number [NUMSTR]).
LOCAT
Approximate location of PHIT in terms of the ratio of the distance along the line (NL1) to the length
of the line. LOCAT can range from 0 to 1. If LOCAT is blank, the point will be located with less speed
and accuracy, and an arbitrary location may result.
Notes
Generates a straight line (PHIT-P3) at an angle (ANG) with a line NL1 (P1-P2). The location of PHIT on the
line is automatically calculated. Line P1-P2 becomes P1-PHIT and new lines PHIT-P2 and PHIT-P3 are
generated. Line divisions are set to zero (use LESIZE, etc. to modify).
PHIT is positioned closest to LOCAT for the given angle, ANG. To ensure better performance, it is recom-
mended that LOCAT be input, even if it is 0.
887
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LANG
The program uses an iterative procedure to position PHIT. The procedure is not exact, with the result that
the actual value of ANG will sometimes differ slightly from the specified value.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>At angle to line
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Normal to Line
LARC, P1, P2, PC, RAD
Defines a circular arc.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1
Keypoint at one end of circular arc line. If P1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
P2
Keypoint at other end of circular arc line.
PC
Keypoint defining plane of arc and center of curvature side (with positive radius). Must not lie along the
straight line from P1 to P2. PC need not be at the center of curvature.
RAD
Radius of curvature of the arc. If negative, assume center of curvature side is opposite to that defined
by PC. If RAD is blank, RAD will be calculated from a curve fit through P1, PC, and P2.
Notes
Defines a circular arc line from P1 to P2. The line shape is generated as circular, regardless of the active
coordinate system. The line shape is invariant with coordinate system after it is generated.
When dealing with a large radius arc (1e3), or if the location of the arc you create is far away from the origin
of your coordinate system, anomalies may occur. You can prevent this by creating the arc at a smaller scale,
and then scaling the model back to full size (LSSCALE).
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Arcs>By End KPs & Rad
Main Menu>Preprocessor>Modeling>Create>Lines>Arcs>Through 3 KPs
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
888
LARC
/LARC, XCENTR, YCENTR, XLRAD, ANGLE1, ANGLE2
Creates annotation arcs (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XCENTR
Arc X center location (-1.0 < X < 1.0).
YCENTR
Arc Y center location (-1.0 < Y < 1.0).
XLRAD
Arc radius length.
ANGLE1
Starting angle of arc.
ANGLE2
Ending angle of arc. The arc is drawn counterclockwise from the starting angle, ANGLE1, to the ending
angle, ANGLE2.
Notes
Defines annotation arcs to be written directly onto the display at a specified location. This is a command
generated by the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation
is used. This command is not intended to be typed in directly in an ANSYS session (although it can be included
in an input file for batch input or for use with the /INPUT command).
All arcs are shown on subsequent displays unless the annotation is turned off or deleted. Use the /LSPEC
command to set the attributes of the arc.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
LAREA, P1, P2, NAREA
Generates the shortest line between two keypoints on an area.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1
First keypoint of line to be generated. If P1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
P2
Second keypoint of line to be generated.
NAREA
Area containing P1 and P2, or area to which generated line is to be parallel.
889
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LAREA
Notes
Generates the shortest line between two keypoints, P1 and P2, both of which lie on an area. The generated
line will also lie on the area. P1 and P2 may also be equidistant (in global Cartesian space) from the area
(and on the same side of the area), in which case a line parallel to the area is generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Overlaid on Area
LARGE, IR, IA, IB, IC, Name, --, --, FACTA, FACTB, FACTC
Finds the largest (the envelope) of three variables.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA, IB, IC
Reference numbers of the three variables to be operated on. If only two, leave IC blank. If only one,
leave IB blank also.
Name
Thirty-two character name for identifying the variable on the printout and displays. Embedded blanks
are compressed upon output.
--, --
Unused fields.
FACTA, FACTB, FACTC
Scaling factors (positive or negative) applied to the corresponding variables (default to 1.0).
Notes
Creates a new variable by finding the largest of up to three variables according to the operation:
IR = Largest of (FACTA x IA, FACTB x IB, FACTC x IC)
The comparison is done at each time location, so that the new variable is the "envelope" of the three existing
variables.
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Find Maximum
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
890
LARGE
LATT, MAT, REAL, TYPE, --, KB, KE, SECNUM
Associates element attributes with the selected, unmeshed lines.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT, REAL, TYPE
Material number, real constant set number, and type number to be associated with selected, unmeshed
lines.
--
Unused field.
KB, KE
Beginning and ending orientation keypoints to be associated with selected, unmeshed lines. ANSYS uses
the location of these keypoints to determine how to orient beam cross sections during beam meshing.
Beam elements may be created along a line with a constant orientation by specifying only one orientation
keypoint (KB), or a pre-twisted beam may be created by selecting different orientation keypoints at each
end of the line (KB and KE). (For a line bounded by two keypoints (KP1 and KP2), the orientation vector
at the beginning of the line extends from KP1 to KB, and the orientation vector at the end of the line
extends from KP2 to KE. The orientation vectors are used to compute the orientation nodes of the ele-
ments.)
SECNUM
Section identifier to be associated with selected, unmeshed lines. For details, see the description of the
SECTYPE and SECNUM commands.
Notes
The element attributes specified by the LATT command will be used when the lines are meshed.
Lines subsequently generated from the lines will also have the attributes specified by MAT, REAL, TYPE,
and SECNUM. If a line does not have these attributes associated with it (by this command) at the time it is
meshed, the attributes are obtained from the then current MAT, REAL, TYPE, and SECNUM command settings.
In contrast, the values specified by KB and KE apply only to the selected lines; that is, lines subsequently
generated from these lines will not share these attributes. Similarly, if a line does not have KB and KE attrib-
utes associated with it via the LATT command at the time it is meshed, ANSYS cannot obtain the attributes
from elsewhere. See the discussion on beam meshing in Meshing Your Solid Model in the Modeling and
Meshing Guide for more information.
Reissue the LATT command (before lines are meshed) to change the attributes. A zero (or blank) argument
removes the corresponding association. If any of the arguments are defined as -1, then that value will be
left unchanged in the selected set..
In some cases, ANSYS can proceed with a line meshing operation even when no logical element type has
been assigned via LATT,,,TYPE or TYPE. See Meshing Your Solid Model in the Modeling and Meshing Guide
for more information about setting element attributes.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>All Lines
Main Menu>Preprocessor>Meshing>Mesh Attributes>Picked Lines
891
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LATT
LAYER, NUM
Specifies the element layer for which data are to be processed.
POST1: Controls
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
NUM
Layer-processing mode:
N
The layer number to process. The default value is 0.
FCMAX
Processes the layer with the largest failure criteria.
Command Default
The default is N = 0, meaning that the entire element is considered to be the default layer. Accordingly,
the results data are from the bottom of the bottom layer and the top of the top layer.
Notes
Specifies the element layer for which results data are to be listed, plotted, or otherwise processed.
Applies to stress and strain data for layered elements SHELL163, SHELL181, SHELL281, ELBOW290, SOLID185,
SOLID186, SOLSH190, SHELL208, SHELL209, REINF264, and REINF265; heat flux and heat gradient for SHELL131
and SHELL132.
The SHELL command may then be used (with shell elements) to specify a location (TOP, MID, BOT) within
the layer for output. (The SHELL command does not apply to thermal shell elements SHELL131 and SHELL132.)
Transverse shear stresses for MID are linearly averaged from TOP and BOT, and do not reflect a parabolic
distribution. Setting KEYOPT(8) = 2 for SHELL181, SHELL281, SHELL208, SHELL209, and ELBOW290 writes the
mid-surface values directly to the results file and yields more accurate values than linear averaging.
Because energy is a per-element quantity, you cannot use this command for energy output.
When using the LAYER command with SHELL181, SOLID185, SOLID186, SOLSH190, SHELL208, SHELL209,
SHELL281, and ELBOW290, KEYOPT(8) must be set to 1 (or 2 for SHELL181, SHELL281, ELBOW290, SHELL208,
and SHELL209) in order to store results for all layers.
When NUM = FCMAX, you must provide the failure criterion input. If using FC input, ANSYS processes all
structural elements.
When NUM = FCMAX, you must provide the failure criterion input. For more information, see the document-
ation for the FC command.
Use this command with RSYS,LSYS to display results in the layer coordinate system for a particular layer.
For the ANSYS LS-DYNA product, this command works differently than described above. For SHELL163, you
must first use EDINT during the solution phase to define the integration points for which you want output
data. Be aware that the output location for SHELL163 data is always at the integration point, so "top" and
"bottom" refer to the top or bottom integration point, not necessarily the top or bottom surface. For more
information, see the ANSYS LS-DYNA User's Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
892
LAYER
Menu Paths
Main Menu>General Postproc>Options for Outp
Utility Menu>List>Results>Options
LAYERP26, NUM
Specifies the element layer for which data are to be stored.
POST26: Controls
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
NUM
Layer-processing mode:
N
The layer number to process. The default value is 1.
Command Default
The default is N = 1, meaning that results data are from the first layer. If KEYOPT(8) = 0, first layer actually
means the bottom of the bottom layer and the top of the top layer.
Notes
Defines the element layer for which results data are to be stored for postprocessing. Applies to stress and
strain data for layered elements BEAM161, SHELL163, SHELL181, SOLID185, SOLID186, SOLSH190, SHELL208,
SHELL209, SHELL281, REINF265, and ELBOW290.
The SHELL command can be used (for shell elements) to specify a location (TOP, MID, BOT) within the layer
for selection on the ESOL command. Transverse shear stresses for MID are linearly averaged from TOP and
BOT, and do not reflect a parabolic distribution. Setting KEYOPT(8) = 2 for SHELL181, SHELL208, SHELL209,
SHELL281, and ELBOW290 writes the mid-surface values directly to the results file and yields more accurate
values than linear averaging.
That this command cannot be used for energy output, as energy is a per-element quantity.
When using the LAYERP26 command with SHELL181, SOLID185, SOLID186, SOLSH190, SHELL208, or SHELL209,
KEYOPT(8) must be set to 1 (or 2 for SHELL181, SHELL208, SHELL209, SHELL281, and ELBOW290) in order to
store results for all layers.
For the ANSYS LS-DYNA product, this command works differently than described above. For SHELL163 and
BEAM161, you must first use EDINT during the solution phase to define the integration points for which
you want output data. Be aware that the output location for SHELL163 data is always at the integration
point, so "top" and "bottom" refer to the top or bottom integration point, not necessarily the top or bottom
surface. For more information, see the ANSYS LS-DYNA User's Guide.
In POST26, the ESOL data stored is based on the active LAYERP26 specification at the time the data is
stored. To store data at various specifications (for example, layers 2 and 5), issue a STORE command before
each new specification.
893
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LAYERP26
Menu Paths
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
LAYLIST, IEL, LAYR1, LAYR2, Mplab1, Mplab2
Lists real constants material properties for layered elements.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
IEL
Element number to be listed. If ALL, list all selected elements [ESEL] of the appropriate type. If blank
and the current element type is a layered element type, list data from the current real constant table in
the layered format.
LAYR1, LAYR2
Range of layer numbers to be listed. If LAYR1 is greater than LAYR2, a reverse order list is produced.
LAYR1 defaults to 1. LAYR2 defaults to LAYR1 if LAYR1 is input or to the number of layers if LAYR1
is not input.
Mplab1, Mplab2
Material property labels (e.g., EX) to be listed along with the layer real constants.
Notes
Lists real constants and any two material properties for layered shell and solid elements.
If matrix input is selected (KEYOPT(2) = 2 or 3), LAYR1, LAYR2, Mplab1, and Mplab2 are not used.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Elements>Layered Elements
Utility Menu>List>Properties>Layer Data
LAYPLOT, IEL, LAYR1, LAYR2
Displays the layer stacking sequence for layered elements.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
IEL
Element number for the display. If blank and the current element type is a layered element type, display
data from the current real constant table.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
894
LAYLIST
LAYR1, LAYR2
Range of layer numbers to be displayed. If LAYR1 is greater than LAYR2, a reversed order display is
produced. Up to 20 layers may be displayed at a time. LAYR1 defaults to 1. LAYR2 defaults to LAYR1
if LAYR1 is input or to the number of layers (or to 19+LAYR1, if smaller) if LAYR1 is not input.
Notes
Displays the layer stacking sequence as defined in the real constant table for layered shell and solid elements
in a form where the layers are visible (like a sheared deck of cards). The element x-axis is shown as 0.0 degrees.
Layers are cross-hatched and color coded for clarity. The hatch lines indicate the layer angle (real constant
THETA) and the color coding is for material identification (real constant MAT). The actual orientation of a
specific layer in three dimensional space can be seen using /PSYMB, LAYR. Layer thickness can be displayed
using the /ESHAPE and EPLOT commands.
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Layered Elements
LCABS, LCNO, KABS
Specifies absolute values for load case operations.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LCNO
Load case pointer number. If ALL, apply to all selected load cases [LCSEL].
KABS
Absolute value key:
0
Use algebraic values of load case LCNO in operations.
1
Use absolute values of load case LCNO in operations.
Command Default
Use algebraic values.
Notes
Causes absolute values to be used in the load case operations [LCASE or LCOPER]. Absolute values are
taken prior to assigning a load case factor [LCFACT] and are applied only to defined load cases [LCDEF].
Menu Paths
Main Menu>General Postproc>Load Case>Calc Options>Absolut Value
895
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LCABS
LCASE, LCNO
Reads a load case into the database.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LCNO
Load case pointer number [LCDEF,STAT]. Defaults to 1.
Command Default
Load case 1.
Notes
Reads a load case into the database. Load cases are created as described on the LCDEF or LCWRITE com-
mands. The results portion of the database and the applied forces and displacements are cleared before
reading the data in. Absolute values [LCABS] and scale factors [LCFACT] can be applied during the read
operation.
Menu Paths
Main Menu>General Postproc>Load Case>Read Load Case
LCCALC
Specifies "Load case settings" as the subsequent status topic.
POST1: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
This command is also valid for rezoning.
Menu Paths
Utility Menu>List>Status>General Postproc>Load Case Calcs
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
896
LCASE
LCCAT, NL1, NL2
Concatenates multiple lines into one line for mapped meshing.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2
Lines to be concatenated. If NL1 = ALL, NL2 is ignored and all selected lines [LSEL] are concatenated.
If NL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the
GUI). A component name may also be substituted for NL1 (NL2 is ignored).
Notes
Concatenates multiple, adjacent lines (the input lines) into one line (the output line) in preparation for
mapped meshing. An area that contains too many lines for mapped meshing can still be mapped meshed
if some of the lines in that area are first concatenated (see Meshing Your Solid Model in the Modeling and
Meshing Guide for details on mapped meshing restrictions).
LCCAT is meant to be used solely for meshing and cannot be used for any other purposes. Specifically, (a)
the output line and any areas that have the output line on their line list [ALIST] cannot be used as input to
any other solid modeling operation (not even another LCCAT command); and (b) the output line cannot
accept solid model boundary conditions [DL, SFL].
The output line will take on the element divisions of the input lines and will not accept element divisions
that are directly assigned [LESIZE]. The output line from the LCCAT operation will be coincident with the
input lines and the input lines will be retained. Consider the LCOMB command instead of LCCAT if you wish
to delete the input lines and if the lines to be combined have similar slopes at the common keypoint(s).
When an LCCAT command is issued, area line lists [ALIST] that contain all of the input lines will be updated
so that the area line lists refer to the output line instead of the input lines. Deletion of the output line [LDELE]
effectively reverses the LCCAT operation and restores area line lists to their original condition.
You can use the LSEL command to select lines that were created by concatenation, and then follow it with
an LDELE,ALL command to delete them. Also see Meshing Your Solid Model in the Modeling and Meshing
Guide for a discussion on how to easily select and delete concatenated lines in one step.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Areas>Mapped>Concatenate>Lines
Main Menu>Preprocessor>Meshing>Mesh>Volumes>Mapped>Concatenate>Lines
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Geometry Controls>Concatenate
Lines
897
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LCCAT
LCDEF, LCNO, LSTEP, SBSTEP, KIMG
Creates a load case from a set of results on a results file.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LCNO
Arbitrary pointer number (1-99) to be assigned to the load case specified by LSTEP, SBSTEP and by
the FILE command. Defaults to 1 + previous value.
LSTEP
Load step number to be defined as the load case. Defaults to one.
SBSTEP
Substep number. Defaults to the last substep of the load step.
KIMG
Used only with results from complex analyses:
0
Use real part of complex solution
1
Use imaginary part.
Notes
Creates a load case by establishing a pointer to a set of results on a results file (written during the ANSYS
solution phase). This pointer (LCNO) can then be used on the LCASE or LCOPER commands to read the
load case data into the database.
Issue LCDEF,ERASE to delete all load case pointers (and all load case files, if any). Issue LCDEF,LCNO,ERASE
to delete only the specific load case pointer LCNO (and its file, if any). With the ERASE options, all pointers
are deleted; however only files with the default extension [LCWRITE] are deleted. Issue LCDEF,STAT for
status of all selected load cases [LCSEL], or LCDEF,STAT,ALL for status of all load cases. The STAT command
may be used to list all load cases. See also LCFILE to establish a pointer to a set of results on a load case
file (written by LCWRITE). Harmonic element data read from a result file load case is stored at the zero-degree
position.
Menu Paths
Main Menu>General Postproc>Load Case>Create Load Case
Main Menu>General Postproc>Load Case>Erase Load Case
Main Menu>General Postproc>Load Case>List Load Cases
LCFACT, LCNO, FACT
Defines scale factors for load case operations.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
LCNO
Load case pointer number. If ALL, apply to all selected load cases [LCSEL].
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
898
LCDEF
FACT
Scale factor applied to load case LCNO. Blank defaults to 1.0.
Command Default
All factors are 1.0.
Notes
Defines scale factors to be used in the load case operations [LCASE or LCOPER]. Scale factors are applied
after an absolute value operation [LCABS] and are applied only to defined load cases [LCDEF].
Menu Paths
Main Menu>General Postproc>Load Case>Calc Options>Scale Factor
LCFILE, LCNO, Fname, Ext, --
Creates a load case from an existing load case file.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
LCNO
Arbitrary (1-99) pointer number assigned to this load case.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to the LCNO value preceded by an" L" (for values 1099) or by an" L0" (for values
19).
--
Unused field.
Notes
Creates a load case by establishing a pointer to an existing load case file [LCWRITE]. This pointer (LCNO)
can then be used on the LCASE or LCOPER commands to read the load case data into the database. This
command is typically used to reestablish load case pointers in a new ANSYS session (pointers are not saved
on the database file), or when more than one pointer to a single load case is desired. See the LCDEF command
for status and erase operations. See also LCDEF to establish a pointer to a set of results on a results file
(written during the ANSYS solution phase).
899
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LCFILE
Menu Paths
Main Menu>General Postproc>Load Case>Create Load Case
LCLEAR, NL1, NL2, NINC
Deletes nodes and line elements associated with selected lines.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
Delete mesh for lines NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to 1). If NL1 = ALL, NL2
and NINC are ignored and the mesh for all selected lines [LSEL] is deleted. If NL1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for NL1 (NL2 and NINC are ignored).
Notes
Deletes all nodes and line elements associated with selected lines (regardless of whether the nodes or ele-
ments are selected). Nodes shared by adjacent meshed lines and nodes associated with non-line elements
will not be deleted. Attributes assigned as a result of LATT are maintained. In the program's response to
the command, if a line or keypoint is tallied as "cleared," it means either its node or element reference was
deleted.
Menu Paths
Main Menu>Preprocessor>Meshing>Clear>Lines
LCOMB, NL1, NL2, KEEP
Combines adjacent lines into one line.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the first line to be combined. If NL1 = ALL, NL2 is ignored and all selected lines [LSEL] are
combined. If NL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NL1 (NL2 is ignored).
NL2
Number of the second line to be combined.
KEEP
Specifies whether to keep the input entities:
0
Delete lines NL1 and NL2 and their common keypoint. Keypoints will not be deleted if they are
meshed or if they are attached to other lines. Lines will not be deleted if they are attached to different
areas.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
900
LCLEAR
1
Keep NL1, NL2, and their common keypoint. (The common keypoint will not be attached to the
output line.)
Notes
Combines adjacent lines into one line (the output line). This operation will effectively "undo" the LDIV oper-
ation. Line divisions are set to zero (use LESIZE, etc. to modify). Lines attached to the same area(s) can also
be combined. See also the LCCAT command for line concatenation capability.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Add>Lines
LCOPER, Oper, LCASE1, Oper2, LCASE2
Performs load case operations.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Oper
Valid operations are:
ZERO
Zero results portion of database (LCASE1 ignored).
SQUA
Square database values (LCASE1 ignored).
SQRT
Square root of database (absolute) values (LCASE1 ignored).
LPRIN
Recalculate line element principal stresses (LCASE1 ignored). Stresses are as shown for the NMISC
items of the ETABLE command for the specific line element type.
ADD
Add LCASE1 to database values.
SUB
Subtract LCASE1 from database values.
SRSS
Square root of the sum of the squares of database and LCASE1.
MIN
Compare and save in database the algebraic minimum of database and LCASE1.
MAX
Compare and save in database the algebraic maximum of database and LCASE1.
ABMN
Compare and save in database the absolute minimum of database and LCASE1 (based on magnitudes,
then apply the corresponding sign).
901
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LCOPER
ABMX
Compare and save in database the absolute maximum of database and LCASE1 (based on magnitudes,
then apply the corresponding sign).
LCASE1
First load case in the operation (if any). See LCNO of the LCDEF command. If ALL, repeat operations using
all selected load cases [LCSEL].
Oper2
Valid operations are:
MULT
Multiplication: LCASE1*LCASE2
LCASE2
Second load case. Used only with Oper2 operations.
Notes
LCOPER operates on the database and one or two load cases according to:
Database = Database Oper (LCASE1 Oper2 LCASE2)
where operations Oper and Oper2 are as described above. Absolute values and scale factors may be applied
to the load cases before the operations [LCABS, LCFACT]. If LCASE1 is not specified, only operation Oper
is performed on the current database. If LCASE2 is specified, operation Oper2 will be performed before
operation Oper. If LCASE2 is not specified, operation Oper2 is ignored. Solution items not contained
[OUTRES] in either the database or the applicable load cases will result in a null item during a load case
operation. Harmonic element data read from a result file load case are processed at zero degrees. All load
case combinations are performed in the solution coordinate system, and the data resulting from load case
combinations are stored in the solution coordinate system. The resultant data are then transformed to the
active results coordinate system [RSYS] when listed or displayed.
Menu Paths
Main Menu>General Postproc>Load Case>Add
Main Menu>General Postproc>Load Case>Line Elem Stress
Main Menu>General Postproc>Load Case>Min & Max
Main Menu>General Postproc>Load Case>Square
Main Menu>General Postproc>Load Case>Square Root
Main Menu>General Postproc>Load Case>SRSS
Main Menu>General Postproc>Load Case>Subtract
LCSEL, Type, LCMIN, LCMAX, LCINC
Selects a subset of load cases.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
902
LCSEL
S
Select a new set.
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Restore the full set.
NONE
Unselect the full set.
INVE
Invert the current set (selected becomes unselected and vice versa).
STAT
Display the current select status.
LCMIN
Minimum value of load case pointer range.
LCMAX
Maximum value of load case pointer range. LCMAX defaults to LCMIN.
LCINC
Value increment within range. Defaults to 1. LCINC cannot be negative.
Command Default
All load cases are selected.
Notes
Selects a subset of load cases for other operations. For example, to select a new set of load cases based on
load cases 1 through 7, use LCSEL,S,1,7. The subset is used when the ALL label is entered (or implied) on
other commands, such as LCFACT, LCABS, LCOPER, etc. Load cases are flagged as selected and unselected;
no load case pointers [LCDEF, LCWRITE, LCFILE] are actually deleted from the database.
Menu Paths
Main Menu>General Postproc>Load Case>Calc Options>Sele Ld Cases
903
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LCSEL
LCSL, NL1, NL2, NL3, NL4, NL5, NL6, NL7, NL8, NL9
Divides intersecting lines at their point(s) of intersection.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NL3, . . . , NL9
Numbers of lines to be intersected. If NL1 = ALL, NL2 to NL9 are ignored and the intersection of all
selected lines is found. If NL1 = P, use graphical picking to specify lines (NL2 to NL9 are ignored).
Notes
Divides intersecting (classifies) lines at their point(s) of intersection. The original lines (and their corresponding
keypoint(s)) will be deleted by default. See the BOPTN command for the options available to Boolean oper-
ations. Element attributes and solid model boundary conditions assigned to the original entities will not be
transferred to the new entities generated.
Menu Paths
This command cannot be accessed from a menu.
LCSUM, Lab
Specifies whether to process non-summable items in load case operations.
POST1: Results
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lab
Combination option
(blank)
Only combine summable items [default].
ALL
Combine all items including non summable items.
Notes
Allows non-summable items (e.g. plastic strains) to be included in load combinations. Issue LCSUM,ALL before
the first load case operation (LCXX command). May also be used to include nonsummable items in the ap-
pending of a results file (RAPPND command).
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
904
LCSL
LCWRITE, LCNO, Fname, Ext, --
Creates a load case by writing results to a load case file.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
LCNO
Arbitrary pointer number (1-99) to be assigned to this load case.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to the LCNO value preceded by an "L" (for values 1099) or by an "L0" (for values
19).
--
Unused field.
Notes
Creates a load case by writing the results data in the database to a load case file. The database remains
unchanged by this operation. A pointer is also established to the written set of results on the load case file.
This pointer (LCNO) can then be used on the LCASE or LCOPER commands to read the load case data into
the database. By default, only summable results data (such as displacements, stresses, elastic strains) and
constant results data (such as volume) are written to the load case file unless requested (LCSUM command).
Non-summable results data (such as plastic strains, strain energy), boundary conditions, and nodal loads are
not written to the load case file. The load case file may be named by default or by a user name. Rewriting
to the same file overwrites the previous data. See the LCDEF command for status and erase operations.
Menu Paths
Main Menu>General Postproc>Load Case>Write Load Case
LCZERO
Zeroes the results portion of the database.
POST1: Load Case Calculations
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Notes
Often used before the LCOPER command. Same as LCOPER,ZERO.
905
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LCZERO
Menu Paths
Main Menu>General Postproc>Load Case>Zero Load Case
LDELE, NL1, NL2, NINC, KSWP
Deletes unmeshed lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
Delete lines from NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to 1). If NL1 = ALL, NL2 and
NINC are ignored and all selected lines [LSEL] are deleted. If NL1 = P, graphical picking is enabled and
all remaining command fields are ignored (valid only in the GUI). A component name may also be sub-
stituted for NL1 (NL2 and NINC are ignored).
KSWP
Specifies whether keypoints are also to be deleted:
0
Delete lines only.
1
Delete lines, as well as keypoints attached to lines but not attached to other lines.
Notes
A line attached to an area cannot be deleted unless the area is first deleted.
Menu Paths
Main Menu>Preprocessor>Meshing>Concatenate>Del Concats>Lines
Main Menu>Preprocessor>Meshing>Mesh>Areas>Mapped>Del Concats>Lines
Main Menu>Preprocessor>Meshing>Mesh>Volumes>Mapped>Del Concats>Lines
Main Menu>Preprocessor>Modeling>Delete>Del Concats>Lines
Main Menu>Preprocessor>Modeling>Delete>Line and Below
Main Menu>Preprocessor>Modeling>Delete>Lines Only
Main Menu>Preprocessor>Modeling>Topo Repair>Delete>Line and Below
Main Menu>Preprocessor>Modeling>Topo Repair>Delete>Lines Only
LDIV, NL1, RATIO, PDIV, NDIV, KEEP
Divides a single line into two or more lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the line to be divided. If negative, assume P1 (see below) is the second keypoint of the line
instead of the first for RATIO. If ALL, divide all selected lines [LSEL]. If NL1 = P, graphical picking is en-
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
906
LDELE
abled and all remaining command fields are ignored (valid only in the GUI). A component name may
also be substituted for NL1.
RATIO
Ratio of line length P1-PDIV to line length P1-P2. Must be between 0.0 and 1.0. Input ignored if NDIV
> 2.
PDIV
Number to be assigned to keypoint generated at division location (defaults to lowest available keypoint
number [NUMSTR]). Input ignored if NL1 = ALL or NDIV > 2. If PDIV already exists and lies on line NL1,
divide line at PDIV (RATIO must also be 0.0). If PDIV already exists and does not lie on line NL1, PDIV
is projected and moved to the nearest point on line NL1 (if possible). PDIV cannot be attached to an-
other line, area, or volume.
NDIV
The number of new lines to be generated from old line (defaults to 2).
KEEP
Specifies whether to keep the input entities:
0
Modify old line to use new keypoints and slopes.
1
Do not modify old line. New lines will overlay old line and have unique keypoints.
Notes
Divides a single line NL1 (defined from keypoint P1 to keypoint P2) into two or more lines. Line NL1 becomes
the new line beginning with keypoint P1 and new lines are generated ending at keypoint P2. If the line is
attached to an area, the area will also be updated. Line divisions are set to zero (use LESIZE, etc. to modify).
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Line into 2 Ln's
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Line into N Ln's
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Lines w/ Options
LDRAG, NK1, NK2, NK3, NK4, NK5, NK6, NL1, NL2, NL3, NL4, NL5, NL6
Generates lines by sweeping a keypoint pattern along path.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NK1, NK2, NK3, . . . , NK6
List of keypoints in the pattern to be dragged (6 maximum if using keyboard entry). If NK1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). If NK1 = ALL,
all selected keypoints (except those that define the drag path) will be swept along the path. A component
name may also be substituted for NK1.
NL1, NL2, NL3, . . . , NL6
List of lines defining the path along which the pattern is to be dragged (6 maximum if using keyboard
entry). Must be a continuous set of lines.
907
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LDRAG
Notes
Generates lines (and their corresponding keypoints) by sweeping a given keypoint pattern along a charac-
teristic drag path. If the drag path consists of multiple lines, the drag direction is determined by the sequence
in which the path lines are input (NL1, NL2, etc.). If the drag path is a single line (NL1), the drag direction
is from the keypoint on the drag line that is closest to the first keypoint of the given pattern to the other
end of the drag line.
The magnitude of the vector between the keypoints of the given pattern and the first path keypoint remains
constant for all generated keypoint patterns and the path keypoints. The direction of the vector relative to
the path slope also remains constant so that patterns may be swept around curves. Keypoint and line
numbers are automatically assigned (beginning with the lowest available values [NUMSTR]). For best results,
the entities to be dragged should be orthogonal to the start of the drag path. Drag operations that produce
an error message may create some of the desired entities prior to terminating.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extrude>Keypoints>Along Lines
LDREAD, Lab, LSTEP, SBSTEP, TIME, KIMG, Fname, Ext, --
Reads results from the results file and applies them as loads.
SOLUTION: FE Constraints
SOLUTION: FE Body Loads
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Valid load label:
TEMP
Temperatures from a thermal analysis are applied as body force nodal loads (BF) in a structural
analysis, an explicit dynamic analysis, or other type of analysis.
If the thermal analysis uses SHELL131 or SHELL132, temperatures are applied as body force element
loads (BFE). In most cases, only the top and bottom temperatures from SHELL131 and SHELL132 are
used by the structural shell elements; any interior temperatures are ignored.
All temperatures are used for SHELL181 using section input, and SHELL281 using section input; for
these elements, therefore, the number of temperature points at a node generated in the thermal
model must match the number of temperature points at a node needed by the structural model.
When using SHELL131 or SHELL132 information for the LDREAD operation, all element types should
specify the same set of thermal degrees of freedom.
Can also be temperatures applied as nodal loads or initial conditions (see KIMG description).
FORC
Forces from an electromagnetic analysis are applied as force loads (F) in a structural analysis or a
FLOTRAN analysis. LDREAD,FORC reads coupling forces. When transferring static magnetic forces
from a SOLID117 model, ensure that all nodes and elements of the bodies intended to receive the
forces are selected prior to transfer. See the discussion on force computation in the Low-Frequency
Electromagnetic Analysis Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
908
LDREAD
For a full harmonic magnetic analysis, FORC represents the time-averaged force (use in conjunction
with KIMG = 2). Values are in the nodal coordinate system for the force loads (F).
HGEN
Heat generations from a magnetic analysis are applied as body force loads (BFE) in a thermal or a
FLOTRAN analysis. For a full harmonic analysis, HGEN represents the time-averaged heat generation
load (use in conjunction with KIMG = 2).
HFLU
Heat fluxes from a FLOTRAN analysis are applied as surface loads in a thermal analysis.
EHFLU
Surface losses from a high frequency electromagnetic analysis are applied as time-average surface
heat flux in a thermal analysis.
JS
Source current density from a current-conduction analysis are applied as body force loads (BFE).
Values are in the global Cartesian coordinate system.
EF
Electric field element centroid values from an electrostatic analysis are applied as body force loads
(BFE) in a magnetic analysis. Values are in the global Cartesian coordinate system.
PRES
Pressures from a FLOTRAN analysis are applied as surface loads (SFE) in a structural analysis. For
shell elements, use the KIMG parameter to establish the face on which to apply the pressure.
REAC
Reaction loads from any analysis are applied as force loads (F) in any analysis. Values are in the
nodal coordinate system.
HFLM
FLOTRAN predicted film coefficient (and associated bulk temperature) are applied as surface loads
(film coefficient and bulk temperature) in a thermal analysis (SFE). FLOTRAN film coefficients may
be either positive or negative, depending on the direction of heat transfer, but ANSYS thermal analyses
require positive film coefficients. If the FLOTRAN film coefficient is negative, LDREAD adjusts the
bulk temperature to be equal to twice the wall temperature minus the bulk temperature. The adjusted
bulk temperature and the now positive film coefficient make the amount and direction of heat
transfer consistent with the FLOTRAN analysis.
H
Nodal magnetic field values from magneto static analysis are applied as nodal loads (BF) in a high-
frequency electromagnetic analysis. Values are in the global Cartesian coordinate system.
FLRZ
Lorentz forces (F = J * B). Used only with SOLID117.
LSTEP
Load step number of the data set to be read. Defaults to 1. If LAST, ignore SBSTEP and TIME and read
the last data set.
SBSTEP
Substep number (within LSTEP). If zero (or blank), LSTEP represents the last substep of the load step.
TIME
Time-point identifying the data set to be read. Used only if both LSTEP and SBSTEP are zero (or blank).
If TIME is between two solution time points on the results file, a linear interpolation is done between
the two data sets. If TIME is beyond the last time point on the file, use the last time point.
909
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LDREAD
KIMG
When used with results from harmonic analyses (ANTYPE,HARMIC) KIMG establishes which set of data
to read:
0
Read the real part of the solution. Valid also for Lab = EHFLU to read in time-average heat flux.
1
Read the imaginary part of the solution.
2
Calculate and read the time-average part. Meaningful for Lab = HGEN or FORC.
When used with the PRES label, KIMG represents the shell element face on which to apply the pressure:
1
Apply pressure to face 1
2
Apply pressure to face 2
When used with the TEMP label, KIMG indicates how temperatures are to be applied. In an explicit dy-
namic analysis, KIMG = 0 is the only valid option for applying temperature loads.
0
Apply temperatures as body loads
1
Apply temperatures as nodal loads
2
Apply temperatures as initial conditions.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to RST if Fname is blank.
--
Unused field.
Notes
Reads results data from the results file and applies them as loads. You also can use LDREAD to apply results
from an analysis defined with one physics environment as loads on a second analysis using a different
physics environment. The values are applied as loads for field coupling effects (such as output temperatures
from a thermal analysis as input to a structural analysis). Nodal loads are applied only to selected nodes.
Element loads are applied only to selected elements. Element surface loads are applied only to selected
elements where all face nodes for that surface are selected. Additionally, to assure proper distribution of
the surface loads, select only the nodes on the element face where the surface load is to be applied. Scaling
and accumulation specifications are applied as the loads are read (BFCUM for body force loads (heat gener-
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
910
LDREAD
ation loads are not accumulated), SFCUM for surface loads, and FCUM for force loads). These commands
do not work for tabular boundary conditions or temperature loads applied to an explicit dynamic analysis
via LDREAD. Use the appropriate list command to list the results (BFLIST or BFELIST for body force loads,
SFELIST for surface loads, and FLIST for force loads). Values may be redefined after being read by issuing
LDREAD again with a different load step and substep, or time value.
When you use LDREAD in an explicit dynamic analysis to read in temperatures, you cannot use the ED-
LOAD,,TEMP command to apply temperature loading. Furthermore, any temperature loading defined by
LDREAD cannot be listed or deleted by the EDLOAD command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Temperature>From
Therm Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Heat Generat>From Mag
Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Forces>Forces>From Mag Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Initial Condit'n>Temp from ANSYS
Main Menu>Preprocessor>Loads>Define Loads>Apply>Initial Condit'n>Temp from Fluid
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>Temperature>From
Therm Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Excitation>AppCurrDens>From
Elec An
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>Electric Field>From Elec
An
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>From Mag Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Force/Moment>From Reactions
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>From Fluid Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Temperature>From Therm Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Convection>From Fluid Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flux>From EMAG Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flux>From Fluid Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Generat>From Mag Analy
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>From ANSYS
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>From Flotran
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Temp From ANSYS
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Temperature>From Therm Analy
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Heat Generat>From Mag Analy
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Forces>Forces>From Mag Analy
Main Menu>Solution>Define Loads>Apply>Initial Condit'n>Temp from ANSYS
Main Menu>Solution>Define Loads>Apply>Initial Condit'n>Temp from Fluid
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>Temperature>From Therm Analy
Main Menu>Solution>Define Loads>Apply>Magnetic>Excitation>AppCurrDens>From Elec An
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>Electric Field>From Elec An
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>From Mag Analy
Main Menu>Solution>Define Loads>Apply>Structural>Force/Moment>From Reactions
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>From Fluid Analy
Main Menu>Solution>Define Loads>Apply>Structural>Temperature>From Therm Analy
Main Menu>Solution>Define Loads>Apply>Thermal>Convection>From Fluid Analy
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flux>From EMAG Analy
911
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LDREAD
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flux>From Fluid Analy
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Generat>From Mag Analy
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>From ANSYS
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>From Flotran
Main Menu>Solution>Loading Options>Temp From ANSYS
LESIZE, NL1, SIZE, ANGSIZ, NDIV, SPACE, KFORC, LAYER1, LAYER2, KYNDIV
Specifies the divisions and spacing ratio on unmeshed lines.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the line to be modified. If ALL, modify all selected lines [LSEL]. If NL1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI). A component name
may also be substituted for NL1.
SIZE
If NDIV is blank, SIZE is the division (element edge) length. The number of divisions is automatically
calculated from the line length (rounded upward to next integer). If SIZE is zero (or blank), use ANGSIZ
or NDIV.
ANGSIZ
The division arc (in degrees) spanned by the element edge (except for straight lines, which always result
in one division). The number of divisions is automatically calculated from the line length (rounded upward
to next integer).
NDIV
If positive, NDIV is the number of element divisions per line. If -1 (and KFORC = 1), NDIV is assumed
to be zero element divisions per line. TARGE169 with a rigid specification ignores NDIV and will always
mesh with one element division.
SPACE
Spacing ratio. If positive, nominal ratio of last division size to first division size (if > 1.0, sizes increase, if
< 1.0, sizes decrease). If negative, |SPACE| is nominal ratio of center division(s) size to end divisions size.
Ratio defaults to 1.0 (uniform spacing). For layer-meshing, a value of 1.0 normally is used. If SPACE =
FREE, ratio is determined by other considerations
KFORC
KFORC 0-3 are used only with NL1 = ALL. Specifies which selected lines are to be modified.
0
Modify only selected lines having undefined (zero) divisions.
1
Modify all selected lines.
2
Modify only selected lines having fewer divisions (including zero) than specified with this command.
3
Modify only selected lines having more divisions than specified with this command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
912
LESIZE
4
Modify only nonzero settings for SIZE, ANGSIZ, NDIV, SPACE, LAYER1, and LAYER2. If KFORC =
4, blank or 0 settings remain unchanged.
LAYER1
Layer-meshing control parameter. Distance which defines the thickness of the inner mesh layer, LAYER1.
Elements in this layer are uniformly-sized with edge lengths equal to the specified element size for the
line (either through SIZE or line-length/NDIV). A positive value for LAYER1 is interpreted as an absolute
length, while a negative value in interpreted as a multiplier on the specified element size for the line.
As a general rule, the resulting thickness of the inner mesh layer should be greater than or equal to the
specified element size for the line. If LAYER1 = OFF, layer-meshing control settings are cleared for the
selected lines. The default value is 0.0
LAYER2
Layer-meshing control parameter. Distance which defines the thickness of the outer mesh layer, LAYER2.
Elements in this layer transition in size from those in LAYER1 to the global element size. A positive value
of LAYER2 is interpreted as an absolute length, while a negative value is interpreted as a mesh transition
factor. A value of LAYER2 = -2 would indicate that elements should approximately double in size as the
mesh progresses normal to LAYER1. The default value is 0.0.
KYNDIV
0, No, and Off means that SmartSizing cannot override specified divisions and spacing ratios. Mapped
mesh fails if divisions do not match. This defines the specification as "hard".
1, Yes, and On means that SmartSizing can override specified divisions and spacing ratios for curvature
or proximity. Mapped meshing can override divisions to obtain required matching divisions. This defines
the specification as" soft".
Notes
Defines the number of divisions and the spacing ratio on selected lines. Lines with previously specified divi-
sions may also be changed.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Layers>Clr Layers
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Layers>Picked Lines
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Lines>All Lines
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Lines>Clr Size
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Lines>Copy Divs
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Lines>Flip Bias
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Lines>Picked Lines
913
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LESIZE
LEXTND, NL1, NK1, DIST, KEEP
Extends a line at one end by using its slope.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the line to be extended. If NL1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
NK1
Number of keypoint at the end of line NL1 to be extended.
DIST
Distance that the line will be extended.
KEEP
Specifies whether to keep the input entities:
0
Modify old line to use new keypoints and slopes.
1
Do not modify old line. New line will overlay old line and have unique keypoints.
Notes
Extends a line at one end by using its slope. Lines may be redefined only if not yet attached to an area. Line
divisions are set to zero (use LESIZE, etc. to modify). Note that solid modeling in a toroidal coordinate system
is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extend Line
LFILLT, NL1, NL2, RAD, PCENT
Generates a fillet line between two intersecting lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the first intersecting line. If NL1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
NL2
Number of the second intersecting line.
RAD
Radius of fillet to be generated. Radius should be less than the lengths of the two lines specified with
NL1 and NL2.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
914
LEXTND
PCENT
Number to be assigned to generated keypoint at fillet arc center. If zero (or blank), no keypoint is gen-
erated.
Notes
Generates a fillet line between two intersecting lines NL1 (P1-PINT) and NL2 (P2-PINT). Three keypoints
may be generated, two at the fillet tangent points (PTAN1 and PTAN2) and one (optional) at the fillet arc
center (PCENT). Line P1-PINT becomes P1-PTAN1, P2-PINT becomes P2-PTAN2, and new arc line
PTAN1-PTAN2 is generated. Generated keypoint and line numbers are automatically assigned (beginning
with the lowest available values [NUMSTR]). Line divisions are set to zero (use LESIZE, etc. to modify).
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Line Fillet
LFSURF, SLINE, TLINE
Generates surface elements overlaid on the edge of existing solid elements and assigns the extra node
as the closest fluid element node.
PREP7: Elements
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
SLINE
Component name for the surface lines of the meshed solid areas.
TLINE
Component name for the target lines meshed with fluid elements.
Notes
This command macro is used to generate surface effect elements overlaid on the surface of existing plane
elements and, based on proximity, to determine and assign the extra node for each surface element. The
underlying areas of the solid region and the fluid lines must be meshed prior to calling this command macro.
The active element type must be SURF151 with appropriate settings for KEYOPT(4), KEYOPT(5), KEYOPT(6),
and KEYOPT(8).
The surface lines of the solid and the target lines of the fluid are grouped into components and named using
the CM command. The names must be enclosed in single quotes (e.g., 'SLINE') when the LFSURF command
is manually typed in.
When using the GUI method, node and element components are created through the picking dialog boxes
associated with this command.
The macro is applicable for the SURF151 and FLUID116 element types.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Attch to Flu-
id>Area to Fluid
915
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LFSURF
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Attch to Fluid>Line
to Fluid
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Attch to Flu-
id>Node to Fluid
LGEN, ITIME, NL1, NL2, NINC, DX, DY, DZ, KINC, NOELEM, IMOVE
Generates additional lines from a pattern of lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME
Do this generation operation a total of ITIMEs, incrementing all keypoints in the given pattern automat-
ically (or by KINC) each time after the first. ITIME must be > 1 for generation to occur.
NL1, NL2, NINC
Generate lines from pattern beginning with NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to
1). If NL1 = ALL, NL2 and NINC are ignored and pattern is all selected lines [LSEL]. If NL1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NL1 (NL2 and NINC are ignored).
DX, DY, DZ
Keypoint location increments in the active coordinate system (--, D, DZ for cylindrical, --, D, -- for
spherical).
KINC
Keypoint increment between generated sets. If zero, the lowest available keypoint numbers are assigned
[NUMSTR].
NOELEM
Specifies if elements and nodes are also to be generated:
0
Generate nodes and elements associated with the original lines, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether to redefine the existing lines:
0
Generate additional lines as requested with the ITIME argument.
1
Move original lines to new position retaining the same keypoint numbers (ITIME, KINC, and NOELM
are ignored). Valid only if the old lines are no longer needed at their original positions. Corresponding
meshed items are also moved if not needed at their original position.
Notes
Generates additional lines (and their corresponding keypoints and mesh) from a given line pattern. The MAT,
TYPE, REAL, and ESYS attributes are based upon the lines in the pattern and not upon the current settings.
End slopes of the generated lines remain the same (in the active coordinate system) as those of the given
pattern. For example, radial slopes remain radial, etc. Generations which produce lines of a size or shape
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
916
LGEN
different from the pattern (i.e., radial generations in cylindrical systems, radial and phi generations in
spherical systems, and theta generations in elliptical systems) are not allowed. Note that solid modeling in
a toroidal coordinate system is not recommended. New line numbers are automatically assigned (beginning
with the lowest available values [NUMSTR]).
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Lines
Main Menu>Preprocessor>Modeling>Move / Modify>Lines
LGLUE, NL1, NL2, NL3, NL4, NL5, NL6, NL7, NL8, NL9
Generates new lines by "gluing" lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NL3, . . . , NL9
Numbers of the lines to be glued. If NL1 = ALL, all selected lines will be glued (NL2 to NL9 will be ig-
nored). If NL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NL1.
Notes
Use of the LGLUE command generates new lines by "gluing" input lines. The glue operation redefines the
input lines so that they share keypoints at their common ends. The new lines encompass the same geometry
as the original lines. This operation is only valid if the intersections of the input lines are keypoints at the
ends of those lines. See the Modeling and Meshing Guide for an illustration. See the BOPTN command for
an explanation of the options available to Boolean operations. Element attributes and solid model boundary
conditions assigned to the original entities will not be transferred to the new entities generated.
The LGLUE command results in the merging of keypoints at the common end of the lines. The keypoints
of the lower numbered line will be kept. This means one must be aware of line numbering when multiple
LGLUE commands are applied to avoid any ungluing of geometry.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Glue>Lines
LGWRITE, Fname, Ext, --, Kedit
Writes the database command log to a file.
SESSION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
917
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LGWRITE
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to LGW if Fname and Ext are blank.
--
Unused field.
Kedit
Flag to suppress nonessential commands:
NONE
Do not suppress any commands (default).
COMMENT
Write nonessential commands as comments (starting with !).
REMOVE
Do not write nonessential commands or comments.
Notes
Writes the database command log to a named file. The database command log contains all commands that
were used to create the current database. These commands are recorded in the database as they are issued,
and saved in the database file (File.DB) whenever the database is saved. The LGWRITE command extracts
these commands from the database and writes them to a file. Nonessential commands (for listing, graphics
displays, help, etc.) can be excluded from the file by using the Kedit field. The file resulting from LGWRITE
can be used as command input to the program. This command is most useful if the session log file
(File.LOG), which is normally saved during an interactive session, has been lost or corrupted. LGWRITE
cannot be used after design optimization looping because the database is cleared [/CLEAR] automatically
at each loop; use File.LOG instead.
This command is valid in any processor.
Menu Paths
Utility Menu>File>Write DB Log File
/LIGHT, WN, NUM, INT, XV, YV, ZV, REFL
Specifies the light direction for the display window.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
NUM
Ambient or directional light key:
0
Ambient light (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
918
/LIGHT
1
Directional light.
INT
Light intensity factor (defaults to 0.3 for ambient, 1.0 for directional). This option is valid only for 3-D
devices).
XV, YV, ZV
Light direction (valid only for NUM = 1). The directional light source is parallel to the line from point XV,
YV, ZV to the origin, in the global Cartesian system origin. Defaults to the viewing direction [/VIEW].
REFL
Light reflectance factor (valid only for NUM = 1 and 3-D devices).
Command Default
Use ambient light.
Notes
Defines the light direction for the window. Use this command only with 3-D graphics devices or 2-D devices
when Z-buffering is used [/TYPE,,(6 or 7)]. The ambient light has no direction, only an intensity. You can
position the directional light source by defining a point (in the global Cartesian coordinate system) repres-
enting a point along the light directional line. This point, and the global Cartesian coordinate system origin,
define the line along which the light is positioned looking toward the origin. You can use any point along
the light line; for example, both (1.,1.,1.) and (2.,2.,2.) give the same light effect. For 3-D graphics devices
only, the directional light source also has intensity and reflectance factors.
By choosing the highest intensity ambient light for 3-D graphics devices (via the command /LIGHT,WN,0,1),
you can nullify color shading and other effects of directional lighting.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Light Source
LINA, NL, NA
Finds the intersection of a line with an area.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL
Number of line to be intersected. If NL = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
NA
Number of area to be intersected.
919
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LINA
Notes
Finds the intersection of a line with an area. New lines will be generated where the lines intersect the areas.
If the regions of intersection are only points, new keypoints will be generated instead. See the Modeling and
Meshing Guide for an illustration. See the BOPTN command for the options available to Boolean operations.
Element attributes and solid model boundary conditions assigned to the original entities will not be transferred
to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Intersect>Line with Area
LINE
Specifies "Lines" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
This command cannot be accessed from a menu.
/LINE, X1, Y1, X2, Y2
Creates annotation lines (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
X1
Line X starting location (-1.0 < X < 2.0).
Y1
Line Y starting location (-1.0 < Y < 1.0).
X2
Line X ending location (-1.0 < X < 2.0).
Y2
Line Y ending location (-1.0 < Y < 1.0).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
920
LINE
Notes
Defines annotation lines to be written directly onto the display at a specified location. This is a command
generated by the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation
is used. This command is not intended to be typed in directly in an ANSYS session (although it can be included
in an input file for batch input or for use with the /INPUT command).
All lines are shown on subsequent displays unless the annotation is turned off or deleted. Use the /LSPEC
command to set the attributes of the line.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
LINES, N
Specifies the length of a printed page.
POST26: Listing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
N
Number of lines per page (defaults to 20). (Minimum allowed = 11).
Command Default
20 lines per page.
Notes
Specifies the length of a printed page (for use in reports, etc.).
Menu Paths
Main Menu>TimeHist Postpro>Settings>List
LINL, NL1, NL2, NL3, NL4, NL5, NL6, NL7, NL8, NL9
Finds the common intersection of lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NL3, . . . , NL9
Numbers of lines to be intersected. If NL1 = ALL, find the intersection of all selected lines and NL2 to
NL9 are ignored. If NL1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NL1.
921
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LINL
Notes
Finds the common (not pairwise) intersection of lines. The common intersection is defined as the regions
shared (in common) by all lines listed on this command. New lines will be generated where the original lines
intersect. If the regions of intersection are only points, new keypoints will be generated instead. See the
Modeling and Meshing Guide for an illustration. See the BOPTN command for the options available to Boolean
operations. Element attributes and solid model boundary conditions assigned to the original entities will
not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Intersect>Common>Lines
LINP, NL1, NL2, NL3, NL4, NL5, NL6, NL7, NL8, NL9
Finds the pairwise intersection of lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NL3, . . . , NL9
Numbers of lines to be intersected pairwise. If NL1 = ALL, find the pairwise intersection of all selected
lines and NL2 to NL9 are ignored. If NL1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI). A component name may be substituted for NL1.
Notes
Finds the pairwise intersection of lines. The pairwise intersection is defined as any and all regions shared by
at least two lines listed on this command. New lines will be generated where the original lines intersect
pairwise. If the regions of pairwise intersection are only points, new keypoints will be generated. See the
Modeling and Meshing Guide for an illustration. See the BOPTN command for the options available to Boolean
operations. Element attributes and solid model boundary conditions assigned to the original entities will
not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Intersect>Pairwise>Lines
LINV, NL, NV
Finds the intersection of a line with a volume.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL
Number of line to be intersected. If NL = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
NV
Number of volume to be intersected.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
922
LINP
Notes
Finds the intersection of a line with a volume. New lines will be generated where the lines intersect the
volumes. If the regions of intersection are only points, new keypoints will be generated instead. See the
Modeling and Meshing Guide for an illustration. See the BOPTN command for the options available to Boolean
operations. Element attributes and solid model boundary conditions assigned to the original entities will
not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Intersect>Line with Volume
LIST, LEVEL
Lists out the sets in the results file.
AUX3: Results Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This command lists the results set number, the load step, substep, and time step for each set. It also shows
all sets marked for deletion.
Menu Paths
This command cannot be accessed from a menu.
*LIST, Fname, Ext, --
Displays the contents of an external, coded file.
SESSION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
Notes
Displays the contents of an external, coded file. The file to be listed cannot be in use (open) at the time
(except for the error file, File.ERR, which may be displayed with *LIST,ERR).
923
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*LIST
Use caution when you are listing active ANSYS files via the List> Files> Other and File> List> Other menu
paths. File I/O buffer and system configurations can result in incomplete listings unless the files are closed.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
LLIST, NL1, NL2, NINC, Lab
Lists the defined lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
List lines from NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to 1). If NL1 = ALL (default), NL2
and NINC are ignored and all selected lines [LSEL] are listed. If NL1 = P, graphical picking is enabled
and all remaining command fields are ignored (valid only in the GUI). A component name may also be
substituted for NL1 (NL2 and NINC are ignored).
Lab
Determines what type of listing is used (one of the following):
(blank)
Prints information about all lines in the specified range.
RADIUS
Prints the radius of certain circular arcs, along with the keypoint numbers of each line. Straight lines,
non-circular curves, and circular arcs not internally identified as arcs (which depends upon how each
arc is created) will print a radius value of zero.
LAYER
Prints layer-mesh control specifications.
HPT
Prints information about only those lines that contain hard points. HPT is not supported in the GUI.
ORIENT
Prints a list of lines, and identifies any orientation keypoints and any cross section IDs that are asso-
ciated with the lines. Used for beam meshing with defined orientation nodes and cross sections.
Notes
There are 2 listings for the number of element divisions and the spacing ratio. The first listing shows assign-
ments from LESIZE only, followed by the hard key (KYNDIV). See LESIZE for more information. The second
listing shows NDIV and SPACE for the existing mesh, if one exists. Whether this existing mesh and the mesh
generated by LESIZE match at any given point depends upon meshing options and the sequence of
meshing operations.
A "-1" in the "nodes" column indicates that the line has been meshed but that there are no interior nodes.
An attribute (TYPE, MAT, REAL, or ESYS) listed as a zero is unassigned; one listed as a positive value indicates
that the attribute was assigned with the LATT command (and will not be reset to zero if the mesh is cleared);
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
924
LLIST
one listed as a negative value indicates that the attribute was assigned using the attribute pointer [TYPE,
MAT, REAL, or ESYS] that was active during meshing (and will be reset to zero if the mesh is cleared).
This command is valid in any processor.
Menu Paths
Utility Menu>List>Lines
LMATRIX, SYMFAC, Coilname, Curname, Indname
Calculates an inductance matrix and the total flux linkage for an N-winding coil system.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
SYMFAC
Geometric symmetry factor. Inductance terms are scaled by this factor which represents the fraction of
the total device modeled. Default is 1.
Coilname
Alphanumeric prefix identifier for coil label used in defining named element coil components. Default
is 'coil.'
Curname
Name of a predefined parameter array containing the nominal coil currents of the system. The array
must be defined (see *DIM command) prior to calling the LMATRIX macro. Default is 'cur.'
Indname
Name of the array parameter to be created by LMATRIX containing the calculated inductance matrix
and the flux linkage in each coil. A text file of the same name with an extension .TXT is created con-
taining the matrix data. Default is 'lmatrix.'
Notes
LMATRIX calculates the differential inductance matrix for an N-winding system where N is the number of
coils in the system, and calculates the total flux linkage in each coil. LMATRIX may only be executed after
the solution of a problem with nominal currents applied to the coils at a desired operating point. The array
Indname has N rows and N+1 columns. The N x N block is the differential inductance matrix; the N+1th
column contains the total flux linkage, with the ith row corresponding to the ith coil. See the Theory Reference
for the Mechanical APDL and Mechanical Applications for more details.
To invoke the LMATRIX macro, for the classical formulations, the elements for each coil must be grouped
into a component using the CM command. Each set of independent coil elements is assigned a component
name with the prefix Coilname followed by the coil number. For the solenoidal formulations, you must
make the exciting node with a F,AMPS load a node component using the CM command. The classical and
solenoidal formulations cannot be mixed.
To invoke the LMATRIX macro, the vector array parameter Curname with dimension N must be defined
and named using the *DIM command. You must set each vector array entry equal to the nominal current
per turn in the corresponding coil at the operating point. Zero current values must be approximated by a
negligibly small applied current.
925
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LMATRIX
Do not apply (or remove) inhomogeneous loads before using the LMATRIX command. Inhomogeneous
loads are those created by:
Degree of freedom commands (D, DA, etc.) specifying nonzero degrees of freedom values on nodes or
solid model entities
Any CE command with a nonzero constant term
Do not put any loads (for example, current) on elements not contained in the element component.
Operating solutions must be obtained through static analysis before calling LMATRIX. All name-strings must
be enclosed in single quotes in the LMATRIX command line. The geometric symmetry factor, Symfac,
represents the fraction of the device modeled, disregarding any current source primitives.
LMATRIX works only with magnetic field elements: PLANE53, SOLID96, and SOLID97, and with SOL-
ID117/SOURC36 solenoidal formulations. You can use interface element INTER115 to connect SOLID96 and
SOLID97 regions.
For more information, see LMATRIX in the Low-Frequency Electromagnetic Analysis Guide.
See the Theory Reference for the Mechanical APDL and Mechanical Applications and Electric and Magnetic
Macros in the Low-Frequency Electromagnetic Analysis Guide for details.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Solve>Electromagnet>Static Analysis>Induct Matrix
LMESH, NL1, NL2, NINC
Generates nodes and line elements along lines.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
Mesh lines from NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to 1). If NL1 = ALL, NL2 and
NINC are ignored and all selected lines [LSEL] are meshed. If NL1 = P, graphical picking is enabled and
all remaining command fields are ignored (valid only in the GUI). A component name may also be sub-
stituted for NL1 (NL2 and NINC are ignored).
Notes
Generates nodes and line elements along lines. Missing nodes required for the generated elements are created
and assigned the lowest available numbers.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Lines
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
926
LMESH
LNCOLLAPSE, LINE, KEYPOINT
Collapse a line segment to a keypoint (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LINE
The ID of the line segment to be collapsed.
KEYPOINT
The ID of the existing keypoint (one of the keypoints connected to the line segment) that will remain
after the collapse operation.
Notes
This command is available only for repairing the geometry of models imported from CAD systems (Default
IGES option).
This command does not check for any entity penetration resulting from the collapse.
Menu Paths
Main Menu>Preprocessor>Modeling>Simplify>Toolkit>Collapse Lines
LNDETACH, LINE1, LINE2, LNINC
Detaches lines from neighboring geometric entity (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LINE1, LINE2, LNINC
Detach lines from LINE1 to LINE2 (defaults to LINE1) in steps of LNINC (defaults to 1).
Notes
Use this command to detach non-manifold lines from their neighboring geometric entities. This command
is available only for repairing the geometry of models imported from CAD systems.
Menu Paths
Main Menu>Preprocessor>Modeling>Geom Repair>Detach Lines
927
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LNDETACH
LNFILL, KP1, KP2
Creates a straight line between two keypoints (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KP1, KP2
Creates a straight line from keypoint defined by KP1 to keypoint defined by KP2. If KP1 = P; graphical
picking is enabled and the remaining command field is ignored.
Notes
This command is available only for repairing the geometry of models imported from CAD systems (Default
IGES option).
Menu Paths
Main Menu>Preprocessor>Modeling>Geom Repair>Fill Lines
LNMERGE, LN1, LN2, LN3, LN4, LN4, LN6, LN7, LN8, LN9, LN10
Merges two or more connected line segments (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LN1, LN2, LN3, . . . , LN10
List of lines that define the set of line segments to merge into one or more line segments. The command
merges all the lines from the selected set that are connected to only one other line. If LN1 = P, graphical
picking is enabled and all remaining arguments are ignored (valid only in the GUI). If LN1 = ALL, all se-
lected lines will be merged and all remaining arguments are ignored. The connected lines are automat-
ically grouped together to form a set of merged lines.
Notes
Use the LNMERGE command to merge disproportionately small line segments into a single line segment.
This command is available only for repairing the geometry of models imported from CAD systems (Default
IGES option). Only those line segments specified in the line list that are connected together will be joined
by the merge operation. The command considers two lines to be connected if those two lines (and only
those two lines) connect together at a single keypoint.
While the LNMERGE command allows you to join multiple lines in one merge operation, the preferred
method is to merge pairs of line segments to successively build a single line segment. Using this technique
can prevent problems during meshing.
The LNMERGE command does not check the angle of inclination between adjacent lines during the merge
operation.
Menu Paths
Main Menu>Preprocessor>Modeling>Simplify>Toolkit>Merge Lines
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
928
LNFILL
LNSPLIT, LINE, PARAM
Splits a line segment into two line segments (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LINE
ID identifying the line to be split.
PARAM
Arc length parameter specifying the location of the keypoint which will join the two new line segments
(the "split point"). Valid arguments are values between 0 and 1, measured from the line's starting keypoint.
Negative values between 0 and -1 are measured from the line's ending keypoint.
Notes
This command is available only for repairing the geometry of models imported from CAD systems (Default
IGES option).
Menu Paths
Main Menu>Preprocessor>Modeling>Simplify>Toolkit>Split Lines
LNSRCH, Key
Activates a line search to be used with Newton-Raphson.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Key
Line search key:
OFF
Do not use a line search.
ON
Use a line search. Note, adaptive descent is suppressed when LNSRCH is on unless explicitly requested
on the NROPT command. Having line search on and adaptive descent on at the same time is not recom-
mended.
AUTO
ANSYS automatically switches line searching ON and OFF between substeps of a load step as needed.
This option is recommended.
Command Default
Line search off, unless contact elements are present.
Notes
The default values given for this command assume SOLCONTROL,ON (the default). See the description of
SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and SOLCONTROL,OFF.
929
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LNSRCH
Activates a line search to be used with the Newton-Raphson method [NROPT]. Line search is an alternative
to adaptive descent (see Line Search in the Theory Reference for the Mechanical APDL and Mechanical Applic-
ations).
LNSRCH,AUTO can be very efficient for problems in which LNSRCH is needed at only certain substeps.
You cannot use line search [LNSRCH], automatic time stepping [AUTOTS], or the DOF solution predictor
[PRED] with the arc-length method [ARCLEN, ARCTRM]. If you activate the arc-length method after you set
LNSRCH, AUTOTS, or PRED, a warning message appears. If you choose to proceed with the arc-length
method activation, ANSYS disables your line search, automatic time stepping, and DOF predictor settings.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Line Search
Main Menu>Solution>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Solution>Load Step Opts>Nonlinear>Line Search
LOCAL, KCN, KCS, XC, YC, ZC, THXY, THYZ, THZX, PAR1, PAR2
Defines a local coordinate system by a location and orientation.
DATABASE: Coordinate System
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCN
Arbitrary reference number assigned to this coordinate system. Must be greater than 10. A coordinate
system previously defined with this number will be redefined.
KCS
Coordinate system type:
0 or CART
Cartesian
1 or CYLIN
Cylindrical (circular or elliptical)
2 or SPHE
Spherical (or spheroidal)
3 or TORO
Toroidal
XC, YC, ZC
Location (in the global Cartesian coordinate system) of the origin of the new coordinate system.
THXY
First rotation about local Z (positive X toward Y).
THYZ
Second rotation about local X (positive Y toward Z).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
930
LOCAL
THZX
Third rotation about local Y (positive Z toward X).
PAR1
Used for elliptical, spheroidal, or toroidal systems. If KCS = 1 or 2, PAR1 is the ratio of the ellipse Y-axis
radius to X-axis radius (defaults to 1.0 (circle)). If KCS = 3, PAR1 is the major radius of the torus.
PAR2
Used for spheroidal systems. If KCS = 2, PAR2 = ratio of ellipse Z-axis radius to X-axis radius (defaults
to 1.0 (circle)).
Notes
Defines a local coordinate system by origin location and orientation angles. The local coordinate system is
parallel to the global Cartesian system unless rotated. Rotation angles are in degrees and redefine any pre-
vious rotation angles. See the CLOCAL, CS, CSWPLA, and CSKP commands for alternate definitions. This
local system becomes the active coordinate system [CSYS]. Local coordinate systems may be displayed with
the /PSYMB command.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Local Coordinate Systems>Create Local CS>At Specified Loc
LOVLAP, NL1, NL2, NL3, NL4, NL5, NL6, NL7, NL8, NL9
Overlaps lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NL3, . . . , NL9
Numbers of lines to be overlapped. If NL1 = ALL, NL2 to NL9 are ignored and all selected lines are
overlapped. If NL1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NL1.
Notes
Overlaps lines. Generates new lines which encompass the geometry of all the input lines. The new lines are
defined by the regions of intersection of the input lines, and by the complementary (non-intersecting) regions.
See the Modeling and Meshing Guide for an illustration. This operation is only valid when the region of inter-
section is a line. See the BOPTN command for an explanation of the options available to Boolean operations.
Element attributes and solid model boundary conditions assigned to the original entities will not be transferred
to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Overlap>Lines
931
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LOVLAP
LPLOT, NL1, NL2, NINC
Displays the selected lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
Display lines from NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to 1). If NL1 = ALL (default),
NL2 and NINC are ignored and display all selected lines [LSEL].
Notes
Mesh divisions on plotted lines are controlled by the LDIV option of the /PSYMB command.
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Lines
Utility Menu>Plot>Specified Entities>Lines
LPRT, PORTNUM, LNUM, Ltype, CS, X1, Y1, Z1, X2, Y2, Z2
Defines impedance and calibration lines for excitation eigenfield.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
PORTNUM
Port number. If PORTNUM = LIST, lists lines. If PORTNUM = DELE, deletes lines.
LNUM
Line number. A maximum of 20 lines can be defined for each port. The line number should be between
1 and 1000.
Ltype
Type of line:
IMPD
Impedance line.
CIMP
Impedance and calibration line.
CS
Local coordinate system number at port.
X1, Y1, Z1
The local coordinates of the starting point.
X2, Y2, Z2
The local coordinates of the ending point.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
932
LPLOT
Notes
The direction of a line is from starting point to ending point.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
LPTN, NL1, NL2, NL3, NL4, NL5, NL6, NL7, NL8, NL9
Partitions lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NL3, . . . , NL9
Numbers of lines to be operated on. If NL1 = ALL, NL2 to NL9 are ignored all selected lines are used.
If NL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the
GUI). A component name may be substituted for NL1.
Notes
Partitions lines. Generates new lines which encompass the geometry of all the input lines. The new lines are
defined by both the regions of intersection of the input lines and the complementary (non-intersecting)
regions. See the Modeling and Meshing Guide for an illustration. See the BOPTN command for an explanation
of the options available to Boolean operations. Element attributes and solid model boundary conditions
assigned to the original entities will not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Partition>Lines
LREFINE, NL1, NL2, NINC, LEVEL, DEPTH, POST, RETAIN
Refines the mesh around specified lines.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
Lines (NL1 to NL2 in increments of NINC) around which the mesh is to be refined. NL2 defaults to NL1,
and NINC defaults to 1. If NL1 = ALL, NL2 and NINC are ignored and all selected lines are used for re-
finement. If NL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NL1 (NL2 and NINC are ignored).
LEVEL
Amount of refinement to be done. Specify the value of LEVEL as an integer from 1 to 5, where a value
of 1 provides minimal refinement, and a value of 5 provides maximum refinement (defaults to 1).
933
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LREFINE
DEPTH
Depth of mesh refinement in terms of the number of elements outward from the indicated lines (defaults
to 1).
POST
Type of postprocessing to be done after element splitting, in order to improve element quality:
OFF
No postprocessing will be done.
SMOOTH
Smoothing will be done. Node locations may change.
CLEAN
Smoothing and cleanup will be done. Existing elements may be deleted, and node locations may
change (default).
RETAIN
Flag indicating whether quadrilateral elements must be retained in the refinement of an all-quadrilateral
mesh. (The ANSYS program ignores the RETAIN argument when you are refining anything other than
a quadrilateral mesh.)
ON
The final mesh will be composed entirely of quadrilateral elements, regardless of the element quality
(default).
OFF
The final mesh may include some triangular elements in order to maintain element quality and
provide transitioning.
Notes
LREFINE performs local mesh refinement around the specified lines. By default, the indicated elements are
split to create new elements with 1/2 the edge length of the original elements (LEVEL = 1).
LREFINE refines all area elements and tetrahedral volume elements that are adjacent to the specified lines.
Any volume elements that are adjacent to the specified lines, but are not tetrahedra (for example, hexahedra,
wedges, and pyramids), are not refined.
You cannot use mesh refinement on a solid model that contains initial conditions at nodes [IC], coupled
nodes [CP family of commands], constraint equations [CE family of commands], or boundary conditions or
loads applied directly to any of its nodes or elements. This applies to nodes and elements anywhere in the
model, not just in the region where you want to request mesh refinement. For additional restrictions on
mesh refinement, see Revising Your Model in the Modeling and Meshing Guide.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Lines
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
934
LREFINE
LREVERSE, LNUM, NOEFLIP
Reverses the normal of a line, regardless of its connectivity or mesh status.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LNUM
Line number of the line whose normal direction is to be reversed. If LNUM = ALL, the normals of all se-
lected lines will be reversed. If LNUM = P, graphical picking is enabled. A component name may also be
substituted for LNUM.
NOEFLIP
Indicates whether you want to change the normal direction of the existing elements on the reversed
line(s) so that they are consistent with each line's new normal direction.
0
Make the normal direction of existing elements on the reversed line(s) consistent with each line's
new normal direction (default).
1
Do not change the normal direction of existing elements on the reversed line(s).
Notes
You cannot use the LREVERSE command to change the normal direction of any element that has a body
or surface load. We recommend that you apply all of your loads only after ensuring that the element normal
directions are acceptable.
Real constants (such as nonuniform shell thickness and tapered beam constants) may be invalidated by an
element reversal.
For more information, see Revising Your Model in the Modeling and Meshing Guide.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Reverse Normals>of Lines
LROTAT, NK1, NK2, NK3, NK4, NK5, NK6, PAX1, PAX2, ARC, NSEG
Generates circular lines by rotating a keypoint pattern about an axis.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NK1, NK2, NK3, . . . , NK6
List of keypoints in the pattern to be rotated (6 maximum if using keyboard entry). If NK1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). If NK1 = ALL,
all selected keypoints (except PAX1 and PAX2) will define the pattern to be rotated. A component name
may also be substituted for NK1.
PAX1, PAX2
Keypoints defining the axis about which the keypoint pattern is to be rotated.
935
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LROTAT
ARC
Arc length (in degrees). Positive follows right-hand rule about PAX1-PAX2 vector. Defaults to 360.
NSEG
Number of lines (8 maximum) around circumference. Defaults to minimum required for 90 (maximum)
arcs, i.e., 4 for 360, 3 for 270, etc.
Notes
Generates circular lines (and their corresponding keypoints) by rotating a keypoint pattern about an axis.
Keypoint patterns are generated at regular angular locations (based on a maximum spacing of 90). Line
patterns are generated at the keypoint patterns. Keypoint and line numbers are automatically assigned
(beginning with the lowest available values [NUMSTR]).
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extrude>Keypoints>About Axis
LSBA, NL, NA, SEPO, KEEPL, KEEPA
Subtracts areas from lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL
Line (or lines, if picking is used) to be subtracted from. If ALL, use all selected lines. If NL = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NL.
NA
Area (or areas, if picking is used) to be subtracted. If ALL, use all selected areas. A component name may
also be substituted for NA.
SEPO
Behavior if the intersection of the lines and the areas is a keypoint or keypoints:
(blank)
The resulting lines will share keypoint(s) where they touch.
SEPO
The resulting lines will have separate, but coincident keypoint(s) where they touch.
KEEPL
Specifies whether NL lines are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NL lines after LSBA operation (override BOPTN command settings).
KEEP
Keep NL lines after LSBA operation (override BOPTN command settings).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
936
LSBA
KEEPA
Specifies whether NA areas are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete areas after LSBA operation (override BOPTN command settings).
KEEP
Keep areas after LSBA operation (override BOPTN command settings).
Notes
Generates new lines by subtracting the regions common to both NL lines and NA areas (the intersection)
from the NL lines. The intersection can be a line(s) or keypoint(s). If the intersection is a keypoint and SEPO
is blank, the NL line is divided at the keypoint and the resulting lines will be connected, sharing a common
keypoint where they touch. If SEPO is set to SEPO, NL is divided into two unconnected lines with separate
keypoints where they touch. See the Modeling and Meshing Guide for an illustration. See the BOPTN command
for an explanation of the options available to Boolean operations. Element attributes and solid model
boundary conditions assigned to the original entities will not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Line by Area
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>With Options>Line by Area
LSBL, NL1, NL2, SEPO, KEEP1, KEEP2
Subtracts lines from lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Line (or lines, if picking is used) to be subtracted from. If ALL, use all selected lines. Lines specified in
this argument are not available for use in the NL2 argument. If P, graphical picking is enabled (valid
only in the GUI) and all remaining fields are ignored. A component name may also be substituted for
NL1.
NL2
Line (or lines, if picking is used) to subtract. If ALL, use all selected lines (except those included in the
NL1 argument). A component name may also be substituted for NL2.
SEPO
Behavior if the intersection of the NL1 lines and the NL2 lines is a keypoint or keypoints:
(blank)
The resulting lines will share keypoint(s) where they touch.
SEPO
The resulting lines will have separate, but coincident keypoint(s) where they touch.
KEEP1
Specifies whether NL1 lines are to be deleted:
937
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSBL
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NL1 lines after LSBL operation (override BOPTN command settings).
KEEP
Keep NL1 lines after LSBL operation (override BOPTN command settings).
KEEP2
Specifies whether NL2 lines are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NL2 lines after LSBL operation (override BOPTN command settings).
KEEP
Keep NL2 lines after LSBL operation (override BOPTN command settings).
Notes
Generates new lines by subtracting the regions common to both NL1 and NL2 lines (the intersection) from
the NL1 lines. The intersection can be a line(s) or point(s). If the intersection is a point and SEPO is blank,
the NL1 line is divided at the point and the resulting lines will be connected, sharing a common keypoint
where they touch. If SEPO is set to SEPO, NL1 is divided into two unconnected lines with separate keypoints
where they touch. See the Modeling and Meshing Guide for an illustration. See the BOPTN command for an
explanation of the options available to Boolean operations. Element attributes and solid model boundary
conditions assigned to the original entities will not be transferred to the new entities generated. LSBL,ALL,ALL
will have no effect since all the lines (in NL1) will be unavailable as NL2 lines.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Line by Line
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>With Options>Line by Line
Main Menu>Preprocessor>Modeling>Operate>Booleans>Subtract>Lines
Main Menu>Preprocessor>Modeling>Operate>Booleans>Subtract>With Options>Lines
LSBV, NL, NV, SEPO, KEEPL, KEEPV
Subtracts volumes from lines.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL
Line (or lines, if picking is used) to be subtracted from. If ALL, use all selected lines. If NL = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NL.
NV
Volume (or volumes, if picking is used) to be subtracted. If ALL, use all selected volumes. A component
name may also be substituted for NV.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
938
LSBV
SEPO
Behavior if the intersection of the NL lines and the NV volumes is a keypoint or keypoints:
(blank)
The resulting lines will share keypoint(s) where they touch.
SEPO
The resulting lines will have separate, but coincident keypoint(s) where they touch.
KEEPL
Specifies whether NL lines are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NL lines after LSBV operation (override BOPTN command settings).
KEEP
Keep NL lines after LSBV operation (override BOPTN command settings).
KEEPV
Specifies whether NV volumes are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NV volumes after LSBV operation (override BOPTN command settings).
KEEP
Keep NV volumes after LSBV operation (override BOPTN command settings).
Notes
Generates new lines by subtracting the regions common to both NL lines and NV volumes (the intersection)
from the NL lines. The intersection can be a line(s) or point(s). If the intersection is a point and SEPO is
blank, the NL1 line is divided at the point and the resulting lines will be connected, sharing a common
keypoint where they touch. If SEPO is set to SEPO, NL1 is divided into two unconnected lines with separate
keypoints where they touch. See the Modeling and Meshing Guide for an illustration. See the BOPTN command
for an explanation of the options available to Boolean operations. Element attributes and solid model
boundary conditions assigned to the original entities will not be transferred to the new entities generated.
LSBL,ALL,ALL will have no effect since all the lines (in NL1) will be unavailable as NL2 lines.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Line by Volume
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>With Options>Line by Volume
Main Menu>Preprocessor>Modeling>Operate>Divide>Line by Volu
939
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSBV
LSBW, NL, SEPO, KEEP
Subtracts the intersection of the working plane from lines (divides lines).
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL
Line (or lines, if picking is used) to be subtracted from. If NL = ALL, use all selected lines. If NL = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be input for NL.
SEPO
Behavior of the created boundary.
(blank)
The resulting lines will share keypoint(s) where they touch.
SEPO
The resulting lines will have separate, but coincident keypoint(s).
KEEP
Specifies whether NL lines are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NL lines after LSBW operation (override BOPTN command settings).
KEEP
Keep NL lines after LSBW operation (override BOPTN command settings).
Notes
Generates new lines by subtracting the intersection of the working plane from the NL lines. The intersection
will be a keypoint(s). The working plane must not be in the same plane as the NL line(s). If SEPO is blank,
the NL line is divided and the resulting lines will be connected, sharing a common keypoint where they
touch. If SEPO is set to SEPO, NL is divided into two unconnected lines with separate keypoints. See the
Modeling and Meshing Guide for an illustration. See the BOPTN command for an explanation of the options
available to Boolean operations. Element attributes and solid model boundary conditions assigned to the
original entities will not be transferred to the new entities generated. Areas that completely contain the input
lines will be updated if the lines are divided by this operation.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Line by WrkPlane
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>With Options>Line by WrkPlane
Main Menu>Preprocessor>Modeling>Operate>Divide>Line by WrkPlane
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
940
LSBW
LSCLEAR, Lab
Clears loads and load step options from the database.
SOLUTION: Load Step Operations
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Lab
Label identifying the data to be cleared:
SOLID
Delete only solid model loads.
FE
Delete only finite element loads.
INER
Delete only inertia loads (ACEL, etc.).
LFACT
Initialize only load factors (on DCUM, FCUM, SFCUM, etc.).
LSOPT
Initialize only load step options.
ALL
Delete all loads and initialize all load step options and load factors.
Notes
Loads are deleted, and load step options are initialized to their default values.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All F.E. Loads
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Inertia Lds
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Loads & Opts
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All SolidMod Lds
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Reset Factors
Main Menu>Preprocessor>Loads>Load Step Opts>Reset Options
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>All
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Areas
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Elements
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Keypoints
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Lines
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Nodes
Main Menu>Solution>Define Loads>Delete>All Load Data>All F.E. Loads
Main Menu>Solution>Define Loads>Delete>All Load Data>All Inertia Lds
Main Menu>Solution>Define Loads>Delete>All Load Data>All Loads & Opts
Main Menu>Solution>Define Loads>Delete>All Load Data>All SolidMod Lds
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Reset Factors
Main Menu>Solution>Load Step Opts>Reset Options
941
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSCLEAR
LSDELE, LSMIN, LSMAX, LSINC
Deletes load step files.
SOLUTION: Load Step Operations
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
LSMIN, LSMAX, LSINC
Range of load step files to be deleted, from LSMIN to LSMAX in steps of LSINC. LSMAX defaults to
LSMIN, and LSINC defaults to 1. If LSMIN = ALL, all load step files are deleted (and LSMAX and LSINC
are ignored). The load step files are assumed to be named Jobname.Sn, where n is a number assigned
by the LSWRITE command (01--09,10,11, etc.). On systems with a 3-character limit on the extension, the
"S" is dropped for numbers > 99.
Notes
Deletes load step files in the current directory (written by the LSWRITE command).
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Delete LS Files
Main Menu>Solution>Define Loads>Operate>Delete LS Files
LSEL, Type, Item, Comp, VMIN, VMAX, VINC, KSWP
Selects a subset of lines.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Restore the full set.
NONE
Unselect the full set.
INVE
Invert the current set (selected becomes unselected and vice versa).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
942
LSDELE
STAT
Display the current select status.
The following fields are used only with Type = S, R, A, or U:
Item
Label identifying data. Valid item labels are shown in the table below. Some items also require a com-
ponent label. If Item = PICK (or simply P), graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI). Defaults to LINE.
Comp
Component of the item (if required). Valid component labels are shown in the table below.
VMIN
Minimum value of item range. Ranges are line numbers, coordinate values, attribute numbers, etc., as
appropriate for the item. If VMIN = 0.0, a tolerance of 1.0E-6 is used, or 0.005 x VMIN if VMIN = VMAX.
A component name (as specified on the CM command) may also be substituted for VMIN (VMAX and
VINC are ignored). If Item = MAT, TYPE, REAL, ESYS, or NDIV and if VMIN is positive, the absolute value
of Item is compared against the range for selection; if VMIN is negative, the signed value of Item is
compared. See the LLIST command for a discussion of signed attributes.
VMAX
Maximum value of item range. VMAX defaults to VMIN.
VINC
Value increment within range. Used only with integer ranges (such as for line numbers). Defaults to 1.
VINC cannot be negative.
KSWP
Specifies whether only lines are to be selected:
0
Select lines only.
1
Select lines, as well as keypoints, nodes, and elements associated with selected lines. Valid only with
Type = S.
Command Default
All lines are selected.
Notes
Selects lines based on values of a labeled item and component. For example, to select a new set of lines
based on line numbers 1 through 7, use LSEL,S,LINE,,1,7. The subset is used when the ALL label is entered
(or implied) on other commands, such as LLIST,ALL. Only data identified by line number are selected. Data
are flagged as selected and unselected; no data are actually deleted from the database.
If Item = LCCA, the command selects only those lines that were created by concatenation. The KSWP field
is processed, but the Comp, VMIN, VMAX, and VINC fields are ignored.
If Item = HPT, the command selects only those lines that contain hard points.
Item = RADIUS is only valid for lines that are circular arcs.
943
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSEL
LSEL is valid in any processor.
For selections based on non-integer numbers (coordinates, results, etc.), items that are within the range
VMIN -Toler and VMAX +Toler are selected. The default tolerance Toler is based on the relative values
of VMIN and VMAX as follows:
If VMIN = VMAX, Toler = 0.005 x VMIN.
If VMIN = VMAX = 0.0, Toler = 1.0E-6.
If VMAX VMIN, Toler = 1.0E-8 x (VMAX - VMIN).
Use the SELTOL command to override this default and specify Toler explicitly.
Table 226 LSEL - Valid Item and Component Labels
Valid Item and Component Labels LSEL, Type, Item, Comp, VMIN, VMAX, VINC,
KSWP
Description Comp Item
Line number. LINE
Line numbers on exterior of selected area (ignore remaining fields). EXT
X, Y, or Z center location in the active coordinate system. X,Y,Z LOC
Unit vector component of outward tangent at beginning of line. X,Y,Z TAN1
Unit vector component of outward tangent at end of line. X,Y,Z TAN2
Number of divisions within the line. NDIV
Spacing ratio of line divisions. SPACE
Material number associated with the line. MAT
Element type number associated with the line. TYPE
Real constant set number associated with the line. REAL
Element coordinate system associated with the line. ESYS
Cross section ID number. [SECNUM] SEC
Length of the line. LENGTH
Radius of the line. RADIUS
Line number (selects only lines with associated hard points). HPT
Concatenated lines (selects only lines that were created by concat-
enation [LCCAT]).
LCCA
Menu Paths
Utility Menu>Select>Entities
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
944
LSEL
LSLA, Type
Selects those lines contained in the selected areas.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of line select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
LSLK, Type, LSKEY
Selects those lines containing the selected keypoints.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of line select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
LSKEY
Specifies whether all contained line keypoints must be selected [KSEL]:
0
Select line if any of its keypoints are in the selected keypoint set.
945
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSLK
1
Select line only if all of its keypoints are in the selected keypoint set.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
LSOPER
Specifies "Load step operations" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Load Step Operations
/LSPEC, LCOLOR, LINSTL, XLNWID
Specifies annotation line attributes (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LCOLOR
Line color:
0
Black
1
Red-Magenta
2
Magenta
3
Blue-Magenta
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
946
LSOPER
4
Blue
5
Cyan-Blue
6
Cyan
7
Green-Cyan
8
Green
9
Yellow-Green
10
Yellow
11
Orange
12
Red
13
Dark Gray
14
Light Gray
15
White
LINSTL
Line style:
0
Solid line.
1
Dashed line.
XLNWID
Line width multiplier (1.0 to 20.0). Defaults to 1.0.
Notes
Specifies annotation line attributes to control certain characteristics of the lines created via the /LINE, /LARC,
/LSYMBOL, /POLYGON, /PMORE, /PCIRCLE, and /PWEDGE commands. This is a command generated by
the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation is used. This
command is not intended to be typed in directly in an ANSYS session (although it can be included in an
input file for batch input or for use with the /INPUT command).
This command is valid in any processor.
947
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/LSPEC
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
LSREAD, LSNUM
Reads load and load step option data into the database.
SOLUTION: Load Step Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LSNUM
Identification number of the load step file to be read. Defaults to 1 + highest number read in the current
session. Issue LSREAD,STAT to list the current value of LSNUM. Issue LSREAD,INIT to reset LSNUM to 1.
The load step files are assumed to be named Jobname.Sn, where n is a number assigned by the
LSWRITE command (01--09,10,11, etc.). On systems with a 3-character limit on the extension, the "S" is
dropped for LSNUM > 99.
Notes
Reads load and load step option data from the load step file into the database. LSREAD will not clear the
database of all current loads. However, if a load is respecified with LSREAD, then it will overwrite the existing
load. See the LSWRITE command to write load step files, and the LSDELE command to delete load step
files. LSREAD removes any existing SFGRAD specification.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Read LS File
Main Menu>Solution>Load Step Opts>Read LS File
LSSCALE, NL1, NL2, NINC, RX, RY, RZ, KINC, NOELEM, IMOVE
Generates a scaled set of lines from a pattern of lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1, NL2, NINC
Set of lines (NL1 to NL2 in steps of NINC) that defines the pattern to be scaled. NL2 defaults to NL1,
NINC defaults to 1. If NL1 = ALL, NL2 and NINC are ignored and the pattern is defined by all selected
lines. If NL1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only
in the GUI). A component name may also be substituted for NL1 (NL2 and NINC are ignored).
RX, RY, RZ
Scale factors to be applied to the X, Y, Z keypoint coordinates in active coordinate system (RR, R, RZ for
cylindrical; RR, R, R for spherical). Note that the R and R scale factors are interpreted as angular
offsets. For example, for CSYS,1, RR, R, RZ input of (1.5,10,3) would scale the specified keypoints 1.5
times in the radial and 3 times in the Z direction, while adding an offset of 10 degrees to the keypoints.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
948
LSREAD
Zero, blank, or negative scale factor values are assumed to be 1.0. Zero or blank angular offsets have no
effect.
KINC
Increment to be applied to keypoint numbers for generated set. If zero, the lowest available keypoint
numbers will be assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Nodes and line elements associated with the original lines will be generated (scaled) if they exist.
1
Nodes and line elements will not be generated.
IMOVE
Specifies whether lines will be moved or newly defined:
0
Additional lines will be generated.
1
Original lines will be moved to new position (KINC and NOELEM are ignored). Use only if the old
lines are no longer needed at their original positions. Corresponding meshed items are also moved
if not needed at their original position.
Notes
Generates a scaled set of lines (and their corresponding keypoints and mesh) from a pattern of lines. The
MAT, TYPE, REAL, and ESYS attributes are based on the lines in the pattern and not the current settings.
Scaling is done in the active coordinate system. Lines in the pattern could have been generated in any co-
ordinate system.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Scale>Lines
LSSOLVE, LSMIN, LSMAX, LSINC
Reads and solves multiple load steps.
SOLUTION: Load Step Operations
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
LSMIN, LSMAX, LSINC
Range of load step files to be read and solved, from LSMIN to LSMAX in steps of LSINC. LSMAX defaults
to LSMIN, and LSINC defaults to 1. If LSMIN is blank, a brief command description is displayed. The
load step files are assumed to be named Jobname.Sn, where n is a number assigned by the LSWRITE
command (01--09,10,11, etc.). On systems with a 3-character limit on the extension, the "S" is dropped
for numbers > 99.
949
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSSOLVE
Notes
LSSOLVE invokes an ANSYS macro to read and solve multiple load steps. The macro loops through a series
of load step files written by the LSWRITE command. The macro file called by LSSOLVE is called LS-
SOLVE.MAC.
LSSOLVE cannot be used with the birth-death option.
LSSOLVE is not supported for cyclic symmetry analyses.
LSSOLVE does not support restarts.
LSSOLVE does not apply to FLOTRAN.
Menu Paths
Main Menu>Solution>Solve>From LS Files
LSTR, P1, P2
Defines a straight line irrespective of the active coordinate system.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1
Keypoint at the beginning of line. If P1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI)
P2
Keypoint at the end of line.
Notes
Defines a straight line from P1 to P2 using the global Cartesian coordinate system. The active coordinate
system will be ignored. The line shape is invariant with the coordinate system after it is generated. Lines
may be redefined only if not yet attached to an area.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Straight Line
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
950
LSTR
LSUM
Calculates and prints geometry statistics of the selected lines.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Calculates and prints geometry statistics (length, centroid, moments of inertia, etc.) associated with the se-
lected lines. Geometry items are reported in the global Cartesian coordinate system. A unit density is assumed,
irrespective of any material associations [LATT, MAT]. Items calculated by LSUM and later retrieved by a
*GET or *VGET command are valid only if the model is not modified after the LSUM command is issued.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Calc Geom Items>Of Lines
LSWRITE, LSNUM
Writes load and load step option data to a file.
SOLUTION: Load Step Operations
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
LSNUM
Number to be assigned to the load step file name for identification purposes. Defaults to 1 + highest
LSNUM used in the current session. Issue LSWRITE,STAT to list the current value of LSNUM. Issue
LSWRITE,INIT to reset to 1. The load step file will be named Jobname.Sn, where n is the specified
LSNUM value (preceded by "0" for values 1-9). On systems with a 3-character limit on the file name ex-
tension, the "S" is dropped for LSNUM > 99.
Command Default
The default behavior of LSWRITE is dependent on the SOLCONTROL command. When solution control is
on (default behavior), the LSWRITE command does NOT write the default values for commands affected by
solution control. When solution control is off (SOLCONTROL,OFF), LSWRITE does write the default values
for commands affected by solution control. See the SOLCONTROL documentation for a list of those com-
mands.
Notes
Writes all load and load step option data for the selected model to a load step file for later use. LSWRITE
does not capture changes made to real constants (R), material properties (MP), couplings (CP), or constraint
equations (CE). Solid model loads will not be saved if the model is not meshed. Solid model loads, if any,
are transferred to the finite element model. Issue LSCLEAR,FE to delete finite element loads. One file is
written for each load step. Use the LSREAD command to read a single load step file, and the LSDELE com-
mand to delete load step files. Use the LSSOLVE command to read and solve the load steps sequentially.
LSWRITE does not support the following commands: DJ, FJ, GSBDATA, GSGDATA, ESTIF, EKILL, EALIVE,
MPCHG, and OUTRES. These commands will not be written to the load step file.
951
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSWRITE
LSWRITE cannot be used with the birth-death option.
LSWRITE does not apply to FLOTRAN.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Write LS File
Main Menu>Solution>Load Step Opts>Write LS File
/LSYMBOL, X, Y, SYMANG, SYMTYP, SYMSIZ, KEYBMP
Creates annotation symbols (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
X
X location for symbol (-1.0 < X < 2.0).
Y
Y location for symbol (-1.0 < Y < 1.0).
SYMANG
Symbol orientation angle.
SYMTYP
Symbol type:
1
Arrow.
2
Tee.
3
Circle.
4
Triangle.
5
Star.
SYMSIZ
Symbol size multiplier (0.1 to 20.0). Defaults to 1.0.
KEYBMP
If KEYBMP = 1, the annotation is a bitmap. SYMTYP will then be a number from 1-99, indicating the
bitmap type (see notes), and X and Y will define the lower left corner of the bitmap. The SYMANG,
SYMSIZarguments are ignored. If KEYBMP = 0, or blank, then the argument definitions above apply.
Notes
Defines annotation symbols to be written directly onto the display at a specified location. This is a command
generated by the GUI and will appear in the log file (Jobname.LOG) if annotation is used. This command
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
952
/LSYMBOL
is not intended to be typed in directly in an ANSYS session (although it can be included in an input file for
batch input or for use with the /INPUT command).
All symbols are shown on subsequent displays unless the annotation is turned off or deleted. Use the /LSPEC
command to set the attributes of the symbol.
The KEYBMP argument reads the symtype argument to determine which bitmap to insert. This bitmap is
defined by an integer between 1 and 99. Numbers 1 through 40 correspond to the standard texture values
found in the /TXTRE command, while numbers 51 through 99 correspond to user supplied bitmaps, as
defined using the Filename option of the /TXTRE command. Numbers 51 through 57 are predefined (the
logos, clamps and arrows available from the GUI) but can be overridden. Numbers 41 through 50 are reserved.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
LSYMM, Ncomp, NL1, NL2, NINC, KINC, NOELEM, IMOVE
Generates lines from a line pattern by symmetry reflection.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Ncomp
Symmetry key:
X
X symmetry (default).
Y
Y symmetry.
Z
Z symmetry.
NL1, NL2, NINC
Reflect lines from pattern beginning with NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to 1).
If NL1 = ALL, NL2 and NINC are ignored and pattern is all selected lines [LSEL]. If NL1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NL1 (NL2 and NINC are ignored).
KINC
Keypoint increment between sets. If zero, the lowest available keypoint numbers are assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Generate nodes and elements associated with the original lines, if they exist.
1
Do not generate nodes and elements.
953
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LSYMM
IMOVE
Specifies whether areas will be moved or newly defined:
0
Generate additional lines.
1
Move original lines to new position retaining the same keypoint numbers (KINC and NOELEM are
ignored). Valid only if the old lines are no longer needed at their original positions. Corresponding
meshed items are also moved if not needed at their original position.
Notes
Generates a reflected set of lines (and their corresponding keypoints and mesh) from a given line pattern
by a symmetry reflection (see analogous node symmetry command, NSYM). The MAT, TYPE, REAL, and ESYS
attributes are based upon the lines in the pattern and not upon the current settings. Reflection is done in
the active coordinate system by changing a particular coordinate sign. The active coordinate system must
be Cartesian. Lines in the pattern may have been generated in any coordinate system. However, solid
modeling in a toroidal coordinate system is not recommended. Lines are generated as described in the
LGEN command.
See the ESYM command for additional information about symmetry elements.
Menu Paths
Main Menu>Preprocessor>Modeling>Reflect>Lines
LTAN, NL1, P3, XV3, YV3, ZV3
Generates a line at the end of, and tangent to, an existing line.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NL1
Number of the line the generated line is tangent to. If negative, assume P1 (see below), instead of P2,
is the second keypoint of line NL1. If NL1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
P3
Keypoint at which generated line must end.
The following fields are used only if a specified slope on the end of the new line is desired, otherwise a zero
curvature end slope will be automatically calculated. To specify the end slope, use the following fields to
define a "slope vector" that has its tail at the origin and its head at the point XV, YV, ZV in the active co-
ordinate system [CSYS]. The corresponding end slope of the line will then be parallel to this "slope vector."
XV3, YV3, ZV3
Location (in the active coordinate system) of the head of the "slope vector" corresponding to the slope
at the P3 end of the line. The tail of the vector is at the coordinate system origin.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
954
LTAN
Notes
Generates a line (P2-P3) tangent at end point (P2) of line NL1 (P1-P2).
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Lines>Tangent to Line
LTRAN, KCNTO, NL1, NL2, NINC, KINC, NOELEM, IMOVE
Transfers a pattern of lines to another coordinate system.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCNTO
Reference number of coordinate system where the pattern is to be transferred. Transfer occurs from the
active coordinate system. The coordinate system type and parameters of KCNTO must be the same as
the active system.
NL1, NL2, NINC
Transfer lines from pattern beginning with NL1 to NL2 (defaults to NL1) in steps of NINC (defaults to
1). If NL1 = ALL, NL2 and NINC are ignored and pattern is all selected lines [LSEL]. If NL1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NL1 (NL2 and NINC are ignored).
KINC
Keypoint increment between sets. If zero, the lowest available keypoint numbers are assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Generate nodes and elements associated with the original lines, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether lines will be moved or newly defined:
0
Generate additional lines.
1
Move original lines to new position retaining the same keypoint numbers (KINC and NOELM are ig-
nored). Valid only if the old lines are no longer needed at their original positions. Corresponding
meshed items are also moved if not needed at their original position.
Notes
Transfers a pattern of lines (and their corresponding keypoints and mesh) from one coordinate system to
another (see analogous node transfer command, TRANSFER). The MAT, TYPE, REAL, and ESYS attributes are
based upon the lines in the pattern and not upon the current settings. Coordinate systems may be translated
and rotated relative to each other. Initial pattern may be generated in any coordinate system. However,
955
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LTRAN
solid modeling in a toroidal coordinate system is not recommended. Coordinate and slope values are inter-
preted in the active coordinate system and are transferred directly. Lines are generated as described in the
LGEN command.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Transfer Coord>Lines
LUMPM, Key
Specifies a lumped mass matrix formulation.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Key
Formulation key:
OFF
Use the element-dependent default mass matrix formulation (default).
ON
Use a lumped mass approximation.
Command Default
Use the default element mass matrix.
Notes
This command is also valid in PREP7. If used in SOLUTION, this command is valid only within the first load
step.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>New Analysis
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>New Analysis
LVSCALE, FACT, LDSTEP
Scales the load vector for mode superposition analyses.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
FACT
Scale factor applied to both the real and imaginary (if they exist) components of the load vector. Defaults
to 0.0.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
956
LUMPM
LDSTEP
Specifies the load step number (from the modal analysis) to be used in MSUP transient or harmonic
analyses. Defaults to 1. The maximum value is 60.
Command Default
No load vector is applied.
Notes
Specifies the scale factor for the load vector that was created in a modal (ANTYPE,MODAL) analysis. Applies
only to the mode superposition transient (ANTYPE,TRANS), mode superposition harmonic response (AN-
TYPE,HARMIC), and PSD spectrum (pressure PSD) analyses.
The LVSCALE command supports tabular boundary conditions (%TABNAME_X%) for FACT input values only
as a function of time in the mode superposition transient (ANTYPE,TRANS) or as a function of frequency in
mode superposition harmonic response (ANTYPE,HARMIC).
MPC contact generates constraint equations that can include constant terms (included on the right-hand
side of the system equation). The LVSCALE command scales the constant terms.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Load Vector>For Mode Super
Main Menu>Preprocessor>Loads>Define Loads>Apply>Load Vector>For PSD
Main Menu>Preprocessor>Loads>Define Loads>Delete>Load Vector>For Mode Super
Main Menu>Preprocessor>Loads>Define Loads>Delete>Load Vector>For PSD
Main Menu>Solution>Define Loads>Apply>Load Vector>For Mode Super
Main Menu>Solution>Define Loads>Apply>Load Vector>For PSD
Main Menu>Solution>Define Loads>Delete>Load Vector>For Mode Super
Main Menu>Solution>Define Loads>Delete>Load Vector>For PSD
LWPLAN, WN, NL1, RATIO
Defines the working plane normal to a location on a line.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number whose viewing direction will be modified to be normal to the working plane (defaults
to 1). If WN is a negative value, the viewing direction will not be modified.
NL1
Number of line to be used. If NL1 = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI).
957
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
LWPLAN
RATIO
Location on NL1, specified as a ratio of the line length. Must be between 0.0 and 1.0. If RATIO = P, use
graphical picking to specify location on the line.
Command Default
Working plane is parallel to the Global X-Y plane at Z = 0.0.
Notes
Defines a working plane (to assist in picking operations) normal to a location on a line. See WPSTYL command
to set the style of working plane display.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Align WP with>Plane Normal to Line
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
958
LWPLAN
M Commands
M, NODE, Lab1, NEND, NINC, Lab2, Lab3, Lab4, Lab5, Lab6
Defines master degrees of freedom for reduced and superelement generation analyses.
SOLUTION: Master DOF
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE
Node number at which master degree of freedom is defined. If ALL, define master degrees of freedom
at all selected nodes [NSEL]. If NODE = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI). A component name may also be substituted for NODE.
Lab1
Valid degree of freedom label. If ALL, use all appropriate labels. Structural labels: UX, UY, or UZ (displace-
ments); ROTX, ROTY, or ROTZ (rotations). Thermal labels: TEMP, TBOT, TE2, TE3, . . ., TTOP (temperature).
Electric labels: VOLT (voltage).
NEND, NINC
Define all nodes from NODE to NEND (defaults to NODE) in steps of NINC (defaults to 1) as master degrees
of freedom in the specified direction.
Lab2, Lab3, Lab4, . . . , Lab6
Additional master degree of freedom labels. The nodes defined are associated with each label specified.
Notes
Defines master degrees of freedom (MDOF) for reduced (dynamic and superelement generation) analyses.
If defined for other analyses, MDOF are ignored. If used in SOLUTION, this command is valid only within the
first load step. Scalar MDOF (VOLT) are not permitted in structural analyses if mass or damping matrices are
reduced. Reduced analyses normally apply only to the UX, UY, UZ, ROTX, ROTY, and ROTZ labels.
Repeat M command for additional master degrees of freedom. The limit for the number of master nodes
used is determined by the maximum system memory available.
The reduced modal (ANTYPE,MODAL), reduced transient (ANTYPE,TRANS), reduced harmonic response
(ANTYPE,HARMIC), and the substructure (ANTYPE,SUBSTR) analyses utilize the matrix condensation technique
to reduce the structure matrices to those characterized by a set of master degrees of freedom.
Master degrees of freedom are identified by a list of nodes and their nodal directions. The actual degree of
freedom directions available for a particular node depends upon the degrees of freedom associated with
element types [ET] at that node. For example, degrees of freedom available with BEAM3 elements are UX,
UY, and ROTZ only. There must be some mass (or stress stiffening in the case of the buckling analysis) asso-
ciated with each master degree of freedom (except for the VOLT label). The mass may be due either to the
distributed mass of the element or due to discrete lumped masses at the node. If a master degree of freedom
is specified at a constrained point, it is ignored. If a master degree of freedom is specified at a coupled node,
it should be specified at the prime node of the coupled set. Master degrees of freedom can also be generated
automatically (during solution) by issuing the TOTAL command in PREP7 or SOLUTION.
959
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Transient displacements and forces, used to apply motion to a structure in the reduced transient or reduced
harmonic response analysis, must be applied at a master degree of freedom. Substructure analysis connection
points must be defined as master degrees of freedom.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Master DOFs>User Selected>Define
Main Menu>Preprocessor>Modeling>CMS>CMS Interface>Define
Main Menu>Solution>Master DOFs>User Selected>Define
MADAPT, ERRTARGT, NADAPT, NMAX, KPLT, Ksmooth, KLST, KCD, DEVICE
Adaptively meshes and solves an edge-based model.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
ERRTARGT
Target percentage for Zienkiewitz Zhu magnetic flux error (defaults to 5).
NADAPT
Maximum number of adaptive steps (defaults to 5).
NMAX
Maximum number of elements at which the iterations may continue (defaults to 50,000). Limits the
number of elements that can be chosen for refinement.
KPLT
Plotting options:
0
No plot (default)
1
Elements and H
2
BERR error estimates
3
BDSG, BEPC error estimates
4
Adaptive details
If KPLT > 0, then MADAPT disables PowerGraphics for the adaptive solution.
Ksmooth
Smoothing options for refinement
0
No postprocessing will be done (default).
1
Smoothing will be done. Node locations may change.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
960
MADAPT
2
Smoothing and cleanup will be done. Existing elements may be deleted, and node locations may
change.
KLST
Listing options
0
No printing (default)
1
Final report
2
Report at each iteration step
3
Report Zienkiewitz Zhu magnetic errors BERR and BSGD
4
Regular details
5
Full details
6
Refine details
7
Track
KCD
Allows you to issue a CDWRITE or CDREAD at every iteration.
0
Do not issue CDWRITE or CDREAD (default).
1
Issue CDWRITE at every iteration (to save every mesh variation). This option issues CDWRITE,geom,
writing the information to jobnameN.cdb.
2
Issue CDREAD at every iteration (to read every mesh variation). Reads from an existing job-
nameN.cdb.
DEVICE
Defines the output device for plotting.
0
Screen only (default)
1
JPEG frames. Each frame is written to a file (jobnameN.jpg by default). See /SHOW.
Notes
MADAPT invokes a predefined ANSYS macro for adaptive meshing and solution of edge-based magnetic
analyses. The macro causes repeated runs of the PREP7, SOLUTION, and POST1 phases of the ANSYS program
with mesh density refinements based upon the percentage error in energy norm.
961
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MADAPT
The MADAPT command macro requires a second, user-defined macro, which must be named madaptld.mac
and must reside in the same directory where ANSYS is being run. This madaptld macro must contain loads
and boundary conditions, based on permanent geometry or solid model features (such as sides or vertices).
Loads specified in the madaptld macro cannot be based on node numbers because the node numbers
will change throughout the refinement process. This secondary macro is required because the MADAPT
macro process must delete all loads and boundary conditions at every refinement step.
MADAPT refines tetrahedral volume elements based on error. Hexahedra, wedges, and pyramids are not
refined (see NREFINE).
This command is also valid at the Begin level.
Menu Paths
This command cannot be accessed from a menu.
MAGOPT, Value
Specifies options for a 3-D magnetostatic field analysis.
SOLUTION: Load Step Options
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS
Value
Option key:
0
Calculate a complete H field solution in the entire domain using a single (reduced) potential.
Caution
When used in problems with both current sources and iron regions, errors may result due
to numerical cancellation.
1
Calculate and store a preliminary H field in "iron" regions (
r
1). Requires flux-parallel boundary
conditions to be specified on exterior iron boundaries. Used in conjunction with subsequent solutions
with VALUE = 2 followed by VALUE = 3. Applicable to multiply-connected iron domain problems.
2
Calculate and store a preliminary H field in "air" regions (
r
= 1). The air-iron interface is appropriately
treated internally by the program. Used in conjunction with a subsequent solution with VALUE = 3.
Applicable to singly-connected iron domain problems (with subsequent solution with VALUE = 3)
or to multiply-connected iron domain problems (when preceded by a solution with VALUE = 1 and
followed by a solution with VALUE = 3).
3
Use the previously stored H field solution(s) and calculate the complete H field.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
962
MAGOPT
Notes
Specifies the solution sequence options for a 3-D magnetostatic field analysis using a scalar potential (MAG).
The solution sequence is determined by the nature of the problem.
You cannot use constraint equations with Value = 1.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Magnetics>Options Only>DSP Method
Main Menu>Preprocessor>Loads>Load Step Opts>Magnetics>Options Only>GSP Method
Main Menu>Preprocessor>Loads>Load Step Opts>Magnetics>Options Only>RSP Method
Main Menu>Solution>Load Step Opts>Magnetics>Options Only>DSP Method
Main Menu>Solution>Load Step Opts>Magnetics>Options Only>GSP Method
Main Menu>Solution>Load Step Opts>Magnetics>Options Only>RSP Method
MAGSOLV, OPT, NRAMP, CNVCSG, CNVFLUX, NEQIT, BIOT,CNVTOL
Specifies magnetic solution options and initiates the solution.
SOLUTION: Load Step Options
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS
OPT
Static magnetic solution option:
0
Vector potential (MVP) or edge formulation (default).
1
Combined vector potential and reduced scalar potential (MVP-RSP).
2
Reduced scalar potential (RSP).
3
Difference scalar potential (DSP).
4
General scalar potential (GSP).
NRAMP
Number of ramped substeps for the first load step of a nonlinear MVP or MVP-RSP solution. Defaults to
3. If NRAMP = -1, ignore the ramped load step entirely.NRAMP is ignored for linear magnetostatics.
CNVCSG
Tolerance value on the program-calculated reference value for the magnetic current-segment convergence.
Used for the MVP, the MVP-RSP, and the edge formulation solution options (OPT = 0 and 1). Defaults
to 0.001.
CNVFLUX
Tolerance value on the program-calculated reference value for the magnetic flux convergence. Used for
all scalar potential solution options (OPT = 2, 3, 4). Defaults to 0.001.
963
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MAGSOLV
NEQIT
Maximum number of equilibrium iterations per load step. Defaults to 25.
BIOT
Option to force execution of a Biot-Savart integral solution [BIOT,NEW] for the scalar potential options.
Required if multiple load steps are being performed with different current source primitives (SOURC36
elements).
0
Do not force execution of Biot-Savart calculation (default); Biot-Savart is automatically calculated
only for the first solution.
1
Force execution of Biot-Savart calculation.
CNVTOL
Sets the convergence tolerance for AMPS reaction. Defaults to 1e-3. Applicable only when KEYOPT(1) =
5 on SOLID117.
Notes
MAGSOLV invokes an ANSYS macro which specifies magnetic solution options and initiates the solution.
The macro is applicable to any ANSYS magnetostatic analysis using the magnetic vector potential (MVP),
reduced scalar potential (RSP), difference scalar potential (DSP), general scalar potential (GSP), or combined
MVP-RSP formulation options. Results are only stored for the final converged solution. (In POST1, issue
*SET,LIST to identify the load step of solution results.) The macro internally determines if a nonlinear analysis
is required based on magnetic material properties.
If you use the BIOT option and issue SAVE after solution or postprocessing, the Biot-Savart calculations are
saved to the database, but will be overwritten upon normal exit from the program. To save this data after
issuing SAVE, use the /EXIT,NOSAVE command. You can also issue the /EXIT,SOLU command to exit ANSYS
and save all solution data, including the Biot-Savart calculations, in the database. Otherwise, when you issue
RESUME, the Biot-Savart calculation will be lost (resulting in a zero solution).
The MVP, MVP-RSP, and edge formulation options perform a two-load-step solution sequence. The first load
step ramps the applied loads over a prescribed number of substeps (NRAMP), and the second load step
calculates the converged solution. For linear problems, only a single load step solution is performed. The
ramped load step can be bypassed by setting NRAMP to -1.
The RSP option solves in a single load step using the adaptive descent procedure. The DSP option uses two
load steps, and the GSP solution uses three load steps.
The following analysis options and nonlinear options are controlled by this macro: KBC, NEQIT, NSUBST,
CNVTOL, NROPT, MAGOPT, and OUTRES.
You cannot use constraint equations with OPT = 4.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Solve>Electromagnet>Static Analysis>Opt&Solv
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
964
MAGSOLV
/MAIL, --, Address, Fname, Ext
Mails file to the specifed address.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
--
Unused field.
Address
Email address (up to 64 characters) of the intended recipient of the file.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
Notes
Issue the /MAIL command to alert someone when a long-running job has completed, as shown in this ex-
ample:
...
SOLVE
/MAIL,,yourname@yourdomain.com,jobdone,txt
If you are running ANSYS in a Microsoft Windows environment, you must configure BLAT (a freeware program
provided with ANSYS). Before issuing the /MAIL command, enter this command from the C:\Program
Files\Ansys Inc\V120\Ansys\Bin\Intel directory:
If you are running ANSYS in a Windows x64 environment, enter this command from the C:\Program
Files\Ansys Inc\V120\Ansys\Bin\Winx64 directory:
blat -install smtphost.bar.com you@bar.com
For more information about the BLAT program, see the BLAT_README.TXT file in the same directory.
Because Windows cannot operate on a busy file, you cannot mail standard ANSYS output. Instead, redirect
the output to a file and then mail that file, as shown in this example:
...
/PREP7
/OUT,scratch.out
:
:
FINISH
/OUT
/MAIL,,YourName@YourDomain.com,scratch.out
FINISH
Menu Paths
This command cannot be accessed from a menu.
965
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/MAIL
MAPSOLVE, MAXSBSTEP
Maps solved node and element solutions from an original mesh to a new mesh.
SOLUTION: Rezoning
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
MAXSBSTEP
The maximum number of substeps for rebalancing the residuals. The default value is 5.
Notes
Used during the rezoning process, the MAPSOLVE command maps solved node and element solutions from
the original mesh to the new mesh and achieves equilibrium based on the new mesh.
Additional substeps are necessary to reduce the residuals to zero.
During the rebalancing stage, the external loads and time remain unchanged.
The MAPSOLVE command is valid for manual rezoning (REZONE,MANUAL) only.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Manual Rezoning>Map Results
MASTER
Specifies "Master DOF" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Master DOF
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
966
MAPSOLVE
MAT, MAT
Sets the element material attribute pointer.
PREP7: Meshing
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT
Assign this material number to subsequently defined elements (defaults to 1).
Notes
Identifies the material number to be assigned to subsequently defined elements. This number refers to the
material number (MAT) defined with the material properties [MP]. Material numbers may be displayed
[/PNUM].
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>Default Attribs
Main Menu>Preprocessor>Modeling>Create>Elements>Elem Attributes
MATER
Specifies "Material properties" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Materials
MCHECK, Lab
Checks mesh connectivity.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Operation:
967
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MCHECK
ESEL
Unselects the valid elements.
Notes
Wherever two area or volume elements share a common face, MCHECK verifies that the way the elements
are connected to the face is consistent with their relative normals or integrated volumes. (This may detect
folds or otherwise overlapping elements.)
MCHECK verifies that the element exterior faces form simply-connected closed surfaces. (This may detect
unintended cracks in a mesh.)
MCHECK warns if the number of element facets in a 2-D loop or 3-D shell is not greater than a computed
limit. This limit is the smaller of either three times the number of faces on one element, or one-tenth the
total number of element faces in the model. (This may detect holes in the middle of a mesh.)
The MCHECK command will perform a number of validity checks on the selected elements, including:
1. Normal check: Wherever two area elements share a common edge, MCHECK verifies that the ordering
of the nodes on each element is consistent with their relative normals.
2. Volume check: Wherever two volume elements share a common face, MCHECK verifies that the sign
of the integrated volume of each element is consistent.
3. Closed surface check: MCHECK verifies that the element exterior faces form simply-connected closed
surfaces (this may detect unintended cracks in a mesh).
4. Check for holes in the mesh: MCHECK warns if the number of element faces surrounding an interior
void in the mesh is small enough to suggest one or more accidentally omitted elements, rather than
a deliberately formed hole. For this test, the number of faces around the void is compared to the
smaller of a) three times the number of faces on one element, or b) one-tenth the total number of
element faces in the model.
Menu Paths
Main Menu>Preprocessor>Meshing>Check Mesh>Connectivity>Ck Connectvty
Main Menu>Preprocessor>Meshing>Check Mesh>Connectivity>Sel Bad Connt
MDAMP, STLOC, V1, V2, V3, V4, V5, V6
Defines the damping ratios as a function of mode.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
STLOC
Starting location in table for entering data. For example, if STLOC = 1, data input in the V1 field applies
to the first constant in the table. If STLOC = 7, data input in the V1 field applies to the seventh constant
in the table, etc. Defaults to the last location filled + 1.
V1, V2, V3, . . . , V6
Data assigned to six locations starting with STLOC. If a value is already in this location, it will be redefined.
Blank values for V2 to V6 leave the corresponding previous value unchanged.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
968
MDAMP
Notes
Defines the damping ratios as a function of mode. Table position corresponds to mode number. These ratios
are added to the DMPRAT value, if defined. Use STAT command to list current values. Applies to the mode
superposition harmonic response (ANTYPE,HARMIC), the mode superposition linear transient dynamic
(ANTYPE,TRANS), and the spectrum (ANTYPE,SPECTR) analyses. Repeat MDAMP command for additional
constants (10000 maximum).
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Damping
Main Menu>Solution>Load Step Opts>Time/Frequenc>Damping
MDELE, NODE, Lab1, NEND, NINC, Lab2, Lab3, Lab4, Lab5, Lab6
Deletes master degrees of freedom.
SOLUTION: Master DOF
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE, Lab1, NEND, NINC
Delete master degrees of freedom in the Lab1 direction [M] from NODE to NEND (defaults to NODE) in
steps of NINC (defaults to 1). If NODE = ALL, NEND and NINC are ignored and masters for all selected
nodes [NSEL] are deleted. If Lab1 = ALL, all label directions will be deleted. If NODE = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NODE.
Lab2, Lab3, Lab4, . . . , Lab6
Delete masters in these additional directions.
Notes
Deletes master degrees of freedom. If used in SOLUTION, this command is valid only within the first load
step.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Master DOFs>User Selected>Delete
Main Menu>Solution>Master DOFs>User Selected>Delete
969
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MDELE
MDPLOT, Function, Dmpname, Scale
Plots frequency-dependent modal damping coefficients calculated by DMPEXT.
SOLUTION: Dynamic Options
MP ME <> <> <> <> <> <> <> <> <> PP <> EME MFS
Function
Function to display.
d_coeff
Damping coefficient
s_coeff
Squeeze coefficient
d_ratio
Damping ratio
s_ratio
Squeeze stiffness ratio
Dmpname
Array parameter name where damping information is stored. Defaults to d_damp.
Scale
Indicates whether to perform a linear or a double logarithmic plot.
LIN
Perform a linear plot. Default
LOG
Perform a double logarithmic plot.
Notes
See Thin Film Analysis for more information on thin film analyses.
Menu Paths
Main Menu>General Postproc>Plot Results>ThinFilm>Plot Dmp Parm
MEMM, Lab, Kywrd
Allows the current session to keep allocated memory
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
When Lab = KEEP, the memory manager's ability to acquire and keep memory is controlled by Kywrd
Kywrd
Turns the memory keep mode on or off
ON
Keep any memory allocated during the analysis.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
970
MDPLOT
OFF
Use memory dynamically and free it up to other users after use (default).
Command Default
Kywrd = OFF.
Notes
You can use the MEMM command to ensure that memory intensive operations will always have the same
memory available when the operations occur intermittently. Normally, if a large amount of memory is allocated
for a specific operation, it will be returned to the system once the operation is finished. This option always
maintains the highest level used during the analysis until the analysis is finished.
The MEMM command does not affect the value you specify with the -m switch. When you allocate memory
with the -m switch, that amount will always be available. However, if dynamic memory allocation in excess
of the-m value occurs, you can use the MEMM command to ensure that amount is retained until the end
of your analysis.
Menu Paths
This command cannot be accessed from a menu.
/MENU, Key
Activates the Graphical User Interface (GUI).
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Key
Activation key:
ON
Activates the menu system (device dependent).
GRPH
Enters non-GUI graphics mode. This option is intended for use on graphics devices that do not
support the full Motif-based GUI.
Command Default
GUI is on if entering the ANSYS program through the launcher. GUI is off if entering using the ANSYS execution
command.
Notes
Activates the Graphical User Interface (GUI).
971
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/MENU
Caution
if you include the /MENU,ON command in your start120.ans, it should be the last command in
the file. Any commands after /MENU,ON may be ignored. (It is not necessary to include the /SHOW
and /MENU,ON commands in start120.ans if you will be using the launcher to enter the ANSYS
program.)
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
MESHING
Specifies "Meshing" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Meshing
MFANALYSIS, Key
Turns an ANSYS Multi-field solver analysis on or off.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Key
Multifield analysis key:
ON
Activates an ANSYS Multi-field solver analysis.
OFF
Deactivates an ANSYS Multi-field solver analysis (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
972
MESHING
Notes
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>Select method
Main Menu>Solution>Multi-field Set Up>Select method
MFBUCKET, Key, Value
Turns a bucket search on or off.
SOLUTION: Multi-field Interface Mapping
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Key
Bucket search key:
ON
Activates a bucket search (default).
OFF
Deactivates a bucket search. A global search is then activated.
Value
Scaling factor (%) used to determine the number of buckets for a bucket search. Defaults to 50%.
Notes
A bucket search will more efficiently compute the mapping of surface and volumetric interpolation data
across field interfaces (flagged by the FSIN label using SF, SFA, SFE, or SFL or the FVIN label using BFE).
The number of buckets used to partition a flagged interface is equal to the scaling factor (%) times the total
number of interface elements. For example, for the default scaling factor of 50% and a 10,000 element inter-
face, 5,000 buckets are used.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
973
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFBUCKET
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Setup>Global
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Mapping
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Setup>Global
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Mapping
MFCALC, FNUMB, FREQ
Specifies a calculation frequency for a field in an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Time Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB
Field number set by the MFELEM command.
FREQ
Perform calculation every Nth ANSYS Multi-field solver time step. Defaults to 1 for every time step.
Notes
This command only applies to a harmonic analysis of the specified field. It is useful when a field contributes
negligible field interaction within a single ANSYS Multi-field solver time step.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Frequency
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Frequency
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
974
MFCALC
MFCI, VAL1, VAL2
Sets the control parameters used by the conservative (CPP) interpolation scheme.
SOLUTION: Multi-field Interface Mapping
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
VAL1
Controls the pixel resolution. The higher the resolution, the more accurate and more expensive the
conservative (CPP) interpolation will be. Valid values are 10 to 256; defaults to 100.
VAL2
The separation factor to handle any gap between the two surfaces. It is a relative measure of the gap,
normalized by the averaged element face sizes from both sides of the interface. Defaults to 0.1.
Notes
In a conservative (CPP) interpolation scheme as specified on the MFLCOMM command, each element face
is first divided into n number of faces, where n is the number of nodes on the face. The three-dimensional
faces are then converted onto a two-dimensional polygon made up of rows and columns of dots called
pixels. By default, these pixels have a resolution of 100 x 100; use VAL1 to increase the resolution and improve
the accuracy of the algorithm. See Load Interpolation in the Coupled-Field Analysis Guide for more information
on interpolation schemes and adjusting the pixel resolution for the conservative interpolation scheme.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Mapping
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Mapping
MFCLEAR, Option, Value
Deletes ANSYS Multi-field solver analysis settings.
SOLUTION: Multi-field Global Controls
MP <> <> <> <> <> <> <> EM <> <> PP <> <> <>
Product Restrictions
Option
SOLU
Resets all ANSYS solution commands except KBC to their default states. This option clears analysis
options when setting up different fields for an ANSYS Multi-field solver analysis.
FIELD
Deletes all ANSYS Multi-field solver specifications for the specified field number.
SINT
Deletes all ANSYS Multi-field solver specifications for the specified surface interface number.
VINT
Deletes all ANSYS Multi-field solver specifications for the volumetric interface number.
975
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFCLEAR
ORD
Deletes the analysis order specified by the MFORDER command.
EXT
Deletes external fields specified by the MFEXTER command
MFLC
Deletes load transfers specified by the MFLCOMM command
Value
Use only for Option = FIELD, SINT, or VINT.
Option = FIELD deletes all ANSYS Multi-field solver specifications for the field number Value. Value
defaults to 0 (no deletions). A Value of -1 deletes the ANSYS Multi-field solver settings for all fields.
Option = SINT deletes surface interface specifications for the surface interface number Value. Value
defaults to 0 (no deletions). A Value of -1 deletes all surface interface specifications. This command
deletes the ANSYS Multi-field solver commands, not the boundary conditions themselves.
Option = VINT deletes all volumetric interface specifications for the volume interface number Value.
Value defaults to 0 (no deletions). A Value of -1 deletes all volumetric interface specifications. This
command deletes the ANSYS Multi-field solver commands, not the boundary conditions themselves.
Notes
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Clear
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Clear
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Clear
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Clear
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
MFCMMAND, FNUMB, Fname, Ext
Captures field solution options in a command file.
SOLUTION: Multi-field Definitions
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB
Field number specified by the MFELEM command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
976
MFCMMAND
Fname
Command file name specified for the field number. Defaults to field "FNUMB".
Ext
Extension for Fname. Defaults to .cmd.
Notes
All relevant solution option commands for the specified field are written to a file with the extension .cmd.
Refer to the commands in the following tables in the Command Reference: Analysis Options, Nonlinear Options,
Dynamic Options, and Load Step Options.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Capture
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Capture
MFCONV, Lab, TOLER, --, MINREF
Sets convergence values for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Convergence Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Lab
Valid labels:
Force labels: FX, FY, FZ
Displacement labels: UX, UY, UZ
Temperature label: TEMP
Heat flux label: HFLU (for surface load transfer)
Heat generation label: HGEN (for volume load transfer)
If Lab = ALL, MFCONV applies the convergence value to all nine variables.
TOLER
Convergence tolerance about program calculated reference value (the L2 norm of the new load in a
multi-field analysis). Defaults to 0.01 (1%) for all labels. Must be less than 1.0.
--
Unused field.
MINREF
The minimum value allowed for the program calculated reference value. If negative, no minimum is
enforced. Defaults to 1.0e-6 for all labels. Not available in the GUI. MINREF corresponds to ||
new
|| as
defined in Set up Stagger Solution in the Coupled-Field Analysis Guide.
977
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFCONV
Notes
MFCONV sets convergence values for variables at the ANSYS Multi-field solver interface.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Stagger>Convergence
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Convergence
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Stagger>Convergence
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Convergence
MFDTIME, DTIME, DTMIN, DTMAX, Carry
Sets time step sizes for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Time Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
DTIME
Multi-field time step size. If automatic time stepping is being used [see Notes below], DTIME is the
starting time step.
DTMIN
Minimum time step. Defaults to DTIME.
DTMAX
Maximum time step. Defaults to DTIME.
Carry
Time step carryover key.
OFF
Use DTIME as the starting time step for the next restart run (default).
ON
Use the final time step from the previous run as the starting time step for the next restart run.
Notes
This command specifies time step sizes for an ANSYS Multi-field solver analysis. If either DTMIN or DTMAX
is not equal to DTIME, auto time-stepping is turned on for the multi-field loop. ANSYS will automatically
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
978
MFDTIME
adjust the time step size for the next multi-field step between DTMIN and DTMAX, based on the status of
the current convergence, the number of target stagger iterations (specified by MFITER), and the actual
number of iterations needed to reach convergence at the current time step.
If auto time-stepping is turned off, the time step size must be evenly divisible into the end time (specified
by MFTIME) minus the start time (0 for a new analysis or a restart time specified by MFRSTART).
You can use a smaller time step within each ANSYS field analysis. This is called subcycling. Use the DELTIM
and AUTOTS commands to subcycle a structural, thermal, or electromagnetic analysis. Use the FLDATA4
command to subcycle a fluid analysis.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Time Ctrl
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Time Ctrl
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Time Ctrl
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Time Ctrl
MFELEM, FNUMB, ITYPE1, ITYPE2, ITYPE3, ITYPE4, ITYPE5, ITYPE6, ITYPE7, ITYPE8, ITYPE9, ITYPE10
Defines a field by grouping element types.
SOLUTION: Multi-field Definitions
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB
Field number for a group of element types.
ITYPE1, ITYPE2, ITYPE3, . . . , ITYPE10
Element types defined by the ET command.
Notes
You can define up to ten element types per field.
Define only element types that contain elements in the field. Do not include MESH200 because it is a mesh-
only element that does not contribute to the solution.
This command is also valid in PREP7.
979
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFELEM
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Define>Define
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Define>Define
MFEM, FNUMB, ITYPE1, ITYPE2, ITYPE3, ITYPE4, ITYPE5, ITYPE6, ITYPE7, ITYPE8, ITYPE9, ITYPE10
Add more element types to a previously defined field number.
SOLUTION: Multi-field Definitions
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB
Existing field number defined by the MFELEM command.
ITYPE1, ITYPE2, ITYPE3, . . . , ITYPE10
Element types defined by the ET command.
Notes
You can add up to ten element types per MFEM command. This command should not be used after an initial
solution.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Define>Add elems
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Define>Add elems
MFEXTER, FNUMB1, FNUMB2, FNUMB3, FNUMB4, FNUMB5, FNUMB6, FNUMB7, FNUMB8, FNUMB9, FNUMB10,
FNUMB11, FNUMB12, FNUMB13, FNUMB14, FNUMB15, FNUMB16, FNUMB17, FNUMB18, FNUMB19, FNUMB20
Defines external fields for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Definitions
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB1, FNUMB2, FNUMB3, . . . , FNUMB20
External field numbers defined by the MFELEM command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
980
MFEM
Notes
This command specifies external field numbers to be used for load transfer in an ANSYS Multi-field solver
analysis. Use the MFIMPORT command to import the external fields.
Use the MFELEM command to specify external field numbers. Use the MFORDER command to specify the
solution order for the external fields.
You can define a maximum of 20 fields.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Setup>External
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Setup>External
MFFNAME, FNUMB, Fname
Specifies a file name for a field in an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Definitions
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB
Field number specified by the MFELEM command.
Fname
File name. Defaults to field "FNUMB".
Notes
All files created for the field will have this file name with the appropriate extensions.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Define>Define
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Define>Define
981
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFFNAME
MFFR, Fname, Lab, RFINI, RFMIN, RFMAX
Setup Multi-Field relaxation factors for field solutions.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Fname
Field name (MFX) or number (MFS). Must be the ANSYS field (cannot be a FLOTRAN or CFX field).
Lab
Label name. Valid values are DISP and TEMP.
RFINI
Initial relaxation factor. Defaults to 0.75.
RFMIN
Minimum relaxation factor. Defaults to RFINI.
RFMAX
Maximum relaxation factor. Defaults to RFINI.
Notes
Use this command to relax the field solutions in fluid-solid interaction analyses and thermal-thermal analyses
for a better convergence rate in coupled problems, especially cases that need dynamic relaxation. The ANSYS
field that has the MFFR command applied will do only one nonlinear stagger iteration within each multi-
field stagger; the convergence of the ANSYS field solver will be satisfied through multiple multi-field staggers.
Note that the CFX field solver can have multiple iterations within the field solver; see the CFX documentation
for more details. ANSYS will not terminate the nonlinear field solution until the ANSYS field solver converges
or reaches the maximum number of multi-field staggers as specified on MFITER.
The interface load relaxation (MFRELAX) will be automatically turned off for the corresponding surface loads
that have MFFR applied. The automatic change of the relaxation factor for accelerating the nonlinear con-
vergence of the coupled field solution is based on Aitken's acceleration scheme.
This command is valid only with coupled problems involving surface load transfer only. No subcycling is al-
lowed for the field solver if using this command.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
982
MFFR
MFIMPORT, FNUMB, Option, Fname, Ext
Imports a new field into a current ANSYS Multi-field solver analysis.
PREP7: Database
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB
Field number specified by the MFELEM command.
Option
Selects data to read.
DB
Reads a CDB file. The CDB file name and extension are specified by Fname and Ext.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to CDB if Fname is blank.
Notes
The field to be imported should be written to a CDB file (CDWRITE command). This file is read into the
database, offsetting all existing element type numbers, node numbers, etc. in order to accommodate the
imported field. (See the NUMOFF command for information on offset capabilities.) It then updates all of the
previously issued MFxx commands to the new element type numbers. A new field is created using the specified
field number, which must not currently exist. If there are no ANSYS Multi-field solver command files written
for the existing fields in the database, one will be written for each field with the default name (see the
MFCMMAND command). A MFCMMAND will be issued for the imported field as well.
Repeat the MFIMPORT command to import additional fields. If you are importing a FLOTRAN fluid field
along with other fields, you must import the FLOTRAN fluid field last to ensure the FLOTRAN elements have
material type 1. FLOTRAN requires its elements to have material type 1.
If the fluid field is imported last and MP commands were used to define the FLOTRAN material properties,
the non-FLOTRAN elements will have material numbers greater than 1. However, if MP commands were not
used to define the FLOTRAN material properties, some non-FLOTRAN elements may also have material
number 1. This material numbering anomaly can occur, because if a MP command is not present in the
fluid fields CDB file, then MFIMPORT will not offset the material numbers when importing the fluid field.
Thus, it is possible for FLOTRAN and non-FLOTRAN elements to share material number 1. The solution will
not be affected, because the MFS internally manages the material numbers to ensure that the correct ma-
terial properties are used with each field. However, the ability to select elements based on their material
properties may be compromised. If unique material numbers are desired, NUMOFF,MAT,1 can be explicitly
included in the fluid field's CDB file to force the material numbers for the non-FLOTRAN elements to be
offset.
983
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFIMPORT
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Import
MFINTER, Option
Specifies the interface load transfer interpolation option for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Option
Interface load transfer option:
CONS
Conservative formulation for load transfer.
NONC
Nonconservative formulation for load transfer (default).
Notes
This command only applies to the interpolation method for forces, heat flux, and heat generation. Displace-
ment and temperature transfers are always nonconservative.
For more information on conservative and nonconservative load transfer, see Load Transfer in the Coupled-
Field Analysis Guide.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Setup>Global
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Setup>Global
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
984
MFINTER
MFITER, MAXITER, MINITER, TARGET
Sets the number of stagger iterations for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Convergence Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
MAXITER
Maximum number of iterations. Defaults to 10.
MINITER
Minimum number of iterations. Defaults to 1.
TARGET
Target number of iterations. Defaults to 5.
Notes
The number of stagger iterations applies to each time step in an ANSYS Multi-field solver analysis. MINITER
and TARGET are valid only when multi-field auto time stepping is on (MFDTIME).
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Stagger>Iterations
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Iterations
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Stagger>Iterations
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Iterations
MFLCOMM, Type, Fname1, Intname1, Label1, Fname2, Intname2, Label2, Option
Defines a load transfer for code coupling analyses.
SOLUTION: Multi-field Load Transfer
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Type
Set to SURF for a surface load transfer. Only surface load transfers are available for MFX.
Fname1
Sets the field solver name for the server (sending) code with a case-sensitive character string of up to
80 characters.
985
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFLCOMM
Intname1
Sets the interface name or number for the field solver of the server code. ANSYS interfaces are numbered
and are defined by the SF family of commands (SF, SFA, or SFE) with the FSIN surface load label. CFX
interfaces use names, which are set in CFX-Pre.
Label1
Sets the surface load label for the field solver of the server code with a character string of up to 80
characters. ANSYS uses a combination of the label and option to determine what data is transferred
(e.g., heat flows and not fluxes are sent with the label/option pair HFLU/CPP). ANSYS cannot serve total
force or total force density to CFX for either formulation. CFX will send the data requested by the label
regardless of the option. CFX labels that have more than one word must be enclosed in single quotes.
Note that this field is case-sensitive; i.e., FORC will work, but forc will not.
Definition Option ANSYS La-
bel
CFX Label
Sets the surface load label to force. CPP FORC 'Total Force'
Sets the surface load label to the total
force per unit area associated with a mesh
node.
NONC FORC 'Total Force
Density'
Sets the surface load label to displace-
ment.
NONC, CPP DISP 'Total Mesh
Displace-
ment'
Sets the surface load label to temperature. NONC, CPP TEMP Temperature
Sets the surface load label to total wall
heat flow.
CPP HFLU 'Wall Heat
Flow'
Sets the surface load label to heat flux. NONC HFLU 'Wall Heat
Flux'
Fname2
Sets the field solver name for the client (receiving) code with a character string of up to 80 characters.
Intname2
Sets the interface name or number for the field solver of the client code with a character string of up
to 80 characters. ANSYS interfaces are numbered and are defined by the SF family of commands (SF,
SFA, or SFE) with the FSIN surface load label. CFX interfaces use names, which are set in CFX-Pre.
Label2
Sets the surface load label for the field solver of the client code with a character string of up to 80
characters. ANSYS uses a combination of the label and option to determine what data is transferred
(e.g., heat flows and not fluxes are sent with the label-option pair HFLU/CPP). CFX will send the data
requested by the label regardless of the option. CFX labels that have more than one word must be en-
closed in single quotes. Note that this field is case-sensitive; i.e., FORC will work, but forc will not.
Option
NONC
Profile preserving: Sets the interface load transfer to the nonconservative formulation (default for
displacement and temperature). In the nonconservative formulation, the force density (or heat flux)
is transferred across the interface, preserving the density profile between the two fields.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
986
MFLCOMM
CPP
Conservative: Uses a local conservative formulation while preserving the density profile (default for
total force and wall heat flow). In the conservative formulation, total force (or heat flow) must be
transferred across the interface from the CFX field solver to the ANSYS field solver.
Notes
ANSYS input should always be in consistent units for its model.
ANSYS uses a combination of the label and option to determine what data to transfer. CFX will send exactly
the data requested by the label, regardless of the option. However, for the NONC option, the CFX label must
be Total Force Density or Wall Heat Flux and for the CPP option, the CFX label must be Total Force or Wall
Heat Flow.
For more information on profile preserving and conservative load transfer, see Load Interpolation in the
Coupled-Field Analysis Guide. Mapping Diagnostics are also available; however, if the improperly-mapped
nodes are based on the CFX mesh, you should ignore the ANSYS-generated components because the CFX
nodes are not present in the ANSYS database.
If you are working interactively, you can choose two pre-defined combinations, Mechanical or Thermal, or
you can choose a Custom option. If you choose the Mechanical load type, then the Total Force Density
and Total Mesh Displacement data (corresponding to the ANSYS FORC and DISP labels, respectively) is
transferred. If you choose the Thermal load type, then the Temperature and Wall Heat Flux data (correspond-
ing to the ANSYS TEMP and HFLU labels, respectively) is transferred. If you choose Custom, you can select
any valid combination of label and option as described above.
The ANSYS Multi-field solver solver does not allow you to switch the load transfer direction for the same
load quantity across the same interfaces for a restart run. For example, if Field1 sends temperature to and
receives heat flow from Field2 across Interface 1 in a previous solution, then you cannot make Field1 send
heat flow to and receive temperatures from Field2 across the same interface in a restart run, even if you
cleared the corresponding load transfer command.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Load Transfer
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Load Transfer
987
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFLCOMM
MFLIST, Option, Value
Lists the settings for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Option
ALL
Lists all ANSYS Multi-field solver analysis options.
SOLU
Lists all solution-related ANSYS Multi-field solver options.
FIELD
Lists all ANSYS Multi-field solver options related to the specified field number.
SINT
Lists all surface interface information for the specified surface interface number.
VINT
Lists all volumetric interface information for the specified volumetric interface number.
Value
Use only for Option = FIELD, SINT, or VINT.
Option = FIELD lists all ANSYS Multi-field solver options for the field number Value. Value defaults to
0, which lists information for all fields.
Option = SINT lists all surface interface information for the interface number Value. Value defaults to
0, which lists information for all surface interfaces.
Option = VINT lists all volumetric interface information for interface number Value. Value defaults to
0, which lists information for all volumetric interfaces.
Notes
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Status
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Status
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Status
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
988
MFLIST
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Status
MFMAP, Lab1, Lab2, Filename, Opt
Calculates, saves, resumes, or deletes mapping data in an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Interface Mapping
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Lab1
Operation label:
CALC
Calculate mapping data and keep it in memory (default).
SAVE
Calculate mapping data, keep it in memory, and save it to a file. (If MFMAP,CALC or MFMAP,RESU
have been issued, just save it to a file.)
RESU
Resume the mapping from a file and keep it in memory.
DELE
Free the mapping memory.
Lab2
Applicable mapping label:
ALL
Surface and volumetric mapping.
SURF
Surface mapping only.
VOLU
Volumetric mapping only.
Filename
The file name for a mapping data file (filename.sur for surface mapping and filename.vol for
volumetric mapping). Defaults to Jobname. Applies to the commands MFMAP,SAVE and MFMAP,RESU
only.
Opt
File format:
BINA
Binary file (default).
ASCI
ASCII file.
Notes
This command calculates, saves, resumes, or deletes mapping data. It defaults to calculating the mapping
data. If MFMAP has not been previously issued, the mapping data will be automatically calculated during
the solution process. On the other hand, the ANSYS Multi-field solver solver will use previously created
mapping data. Resumed mapping files must have load transfer specifications that are consistent with those
of the current MFSURFACE and MFVOLUME commands and the ANSYS database.
989
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFMAP
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Interface>Mapping
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Interface>Mapping
MFORDER, FNUMB1, FNUMB2, FNUMB3, FNUMB4, FNUMB5, FNUMB6, FNUMB7, FNUMB8, FNUMB9, FNUMB10,
FNUMB11, FNUMB12, FNUMB13, FNUMB14, FNUMB15, FNUMB16, FNUMB17, FNUMB18, FNUMB19, FNUMB20
Specifies field solution order for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
FNUMB1, FNUMB2, FNUMB3, . . . , FNUMB20
Field numbers defined by the MFELEM command .
Notes
You can define up to twenty fields in an ANSYS Multi-field solver analysis.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Setup>Order
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Setup>Order
MFOUTPUT, FREQ
Specifies results file output frequency for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Time Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
FREQ
N
Write solution every Nth (and the last) time step. Defaults to 1, for every time step.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
990
MFORDER
-N
Writes up to -N equally spaced results (for multifield auto time stepping).
NONE
Suppresses writing of results for all multifield time steps.
ALL
Writes results for every multifield time step (default).
LAST
Writes results for only the last multifield time step.
%array%
Where %array% is the name of an n X 1 X 1 dimensional array parameter defining n key times, the
data for the specified solution results item is written at those key times. Key times in the array
parameter must appear in ascending order. Value must be greater than or equal to the ending time
values for the load step.
For restart runs (see MFRSTART command), either change the parameter values to fall between the
beginning and ending time values of the load step, or erase the current settings and reissue the
command with a new array parameter.
For more information about defining array parameters, see the *DIM command documentation.
Notes
A MFOUTPUT setting overrides any other output frequency setting (OUTRES). To select the solution items,
use the OUTRES command.
For the case of Freq = -n and Freq = %array%, the results at the time points which first time reaches or exceeds
the targeting ouptupt time points will be written.
FLOTRAN ignores a MFOUTPUT setting. You should write out every time step in a FLOTRAN analysis.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Frequency
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Frequency
991
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFOUTPUT
*MFOURI, Oper, COEFF, MODE, ISYM, THETA, CURVE
Calculates the coefficients for, or evaluates, a Fourier series.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Oper
Type of Fourier operation:
FIT --
Calculate Fourier coefficients COEFF from MODE, ISYM, THETA, and CURVE.
EVAL --
Evaluate the Fourier curve CURVE from COEFF, MODE, ISYM andTHETA
COEFF
Name of the array parameter vector containing the Fourier coefficients (calculated if Oper = FIT, required
as input if Oper = EVAL). See *SET for name restrictions.
MODE
Name of the array parameter vector containing the mode numbers of the desired Fourier terms.
ISYM
Name of the array parameter vector containing the symmetry key for the corresponding Fourier terms.
The vector should contain keys for each term as follows:
0 or 1 --
Symmetric (cosine) term
-1 --
Antisymmetric (sine) term.
THETA, CURVE
Names of the array parameter vectors containing the theta vs. curve description, respectively. Theta
values should be input in degrees. If Oper = FIT, one curve value should be supplied with each theta
value. If Oper = EVAL, one curve value will be calculated for each theta value.
Notes
Calculates the coefficients of a Fourier series for a given curve, or evaluates the Fourier curve from the given
(or previously calculated) coefficients. The lengths of the COEFF, MODE, and ISYM vectors must be the
same--typically two times the number of modes desired, since two terms (sine and cosine) are generally re-
quired for each mode. The lengths of the CURVE and THETA vectors should be the same or the smaller of
the two will be used. There should be a sufficient number of points to adequately define the curve--at least
two times the number of coefficients. A starting array element number (1) must be defined for each array
parameter vector. The vector specifications *VLEN, *VCOL, *VABS, *VFACT, and *VCUM do not apply to
this command. Array elements should not be skipped with the *VMASK and the NINC value of the *VLEN
specifications. The vector being calculated (COEFF if Oper is FIT, or CURVE if Oper is EVAL) must exist as
a dimensioned array [*DIM].
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
992
*MFOURI
Menu Paths
Utility Menu>Parameters>Array Operations>Matrix Fourier
MFPSIMUL, gname, Fname1, Fname2
Sets up a field solver group to simultaneously process with code coupling analyses.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
gname
Sets the group name with a character string of up to 80 characters.
Fname1, Fname2
Sets the field solver 1 and field solver 2 names, which are processed simultaneously, with a character
string of up to 80 characters.
Notes
This command is used to define a group of simultaneously-processed field solvers in an MFX analysis. For
example, to define group g1 with field solvers ansys-code and cfx-code, enter MFPS,g1,ansys-
code,cfx-code.
To indicate groups of sequentially-processed field solvers for your MFX analysis, create two groups (g1 and
g2).
A field solver refers to a specific instance of an ANSYS or CFX solver execution that is defined by the respective
input file(s) referenced when starting the solver (through the launcher or from the command line). The field
solver names that are referenced in several MFX commands must be consistent with the names that will be
used when starting the coupled simulation.
Note
When running MFX from the launcher, you must use ANSYS and CFX (uppercase) as the field
solver names (MFPSIMUL) in your input file.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
993
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFPSIMUL
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
MFRC, FREQ, MAXFILES
Controls file writing for multiframe restarts for the ANSYS Multi-field solver.
SOLUTION: Multi-field Time Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
FREQ
Frequency at which the .Xnnn files are written.
0
Do not write any .Xnnn files for this simulation
LAST
Write the .Xnnn files for the last multifield time step (default).
N
If N is a positive number, write the .Xnnn file every Nth multifield time step.
If N is a negative number write N equally spaced .Xnnn file for the entire simulation time. If auto
time stepping is on, write the file at the multifield time step which first exceeds one of the equally
spaced time points.
MAXFILES
Maximum number of .Xnnn files to save for a multifield analysis.
0
Do not overwrite existing .Xnnn files (default). The total maximum number of .Xnnn files for one
run is 999. If this number is reached before the analysis is complete, the analysis will continue, but
will no longer write .Xnnn files.
N
The maximum number of .Xnnn file to keep for this multifield simulation. When this number of
.Rnnn have been written, ANSYS will overwrite the first .Xnnn file of the subsequent multifield
time step.
Notes
This command sets up the restart parameters for a multiframe restart, allowing you to restart an analysis
from any multifield time step for which there is a .Rnnn file. You can perform a multiframe restart only for
nonlinear static and full transient structural, thermal and thermal- structural (see RESCONTROL for details).
If you have many multifield time steps, and are writing .Xnnn files frequently, use MAXFILES to limit the
number of .Xnnn files saved, since these files can fill up your disk quickly.
For a CFX analysis, you must also configure the MFOUTPUT and MFRSTART settings to ensure consistent
time points for post processing and restart simulation.
For MFX simulation, the RESCONTROL command will be ignored.
Default Behavior
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
994
MFRC
In nonlinear static and full transient analyses, the default behavior is multiframe restart. (MFRC, LAST). By
default, the .Rnnn file will be written at the last multifield time step . A ..Rnnn file and corresponding load
set of .ldhifiles is also written at the multifield time step prior to the abort point. of the run if an abort
file was used or if the job terminated because of a failure to converge or some other solution error. No in-
formation at the aborted multifield time step will be saved in either the .Rnnn file or the .ldhi file.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
To be provided at time of release.
MFRELAX, Lab, VALUE, Option
Sets relaxation values for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Convergence Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Lab
Valid labels:
Force label: FORC
Displacement label: DISP
Temperature label: TEMP
Heat flux label: HFLU
Heat generation label: HGEN
If Lab = ALL, MFRELAX applies the relaxation value to all variables.
VALUE
Relaxation value. Defaults to 0.75 for all labels.
Option
Valid options are:
RELX
Uses relaxation method for load transfer (default).
LINT
Uses a linear interpolation for loaf transfer.
Notes
MFRELAX sets relaxation values for the load transfer variables at a surface or volume interface. Option =
RELX will usually give you a more stable and smooth load transfer and is suitable for strongly coupled
995
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFRELAX
problems (such as FSI problems). Option = LINT is suitable for weakly coupled problems because it will
transfer the full load in fewer stagger iterations.
See the MFFR and MFITER commands for more information on relaxation in the ANSYS Multi-field solver.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Stagger>Relaxation
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Relaxation
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Stagger>Relaxation
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Relaxation
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
MFRSTART, TIME, Type
Specifies restart status for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Time Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
TIME
Restart time
0
New analysis (Default)
-1
Restart from the last result set from a previous run. For a singleframe restart, the ANSYS field solver
checks the ANSYS database (.DB) file to find the latest time; for a multiframe restart, ANSYS checks
the last result file (.Rnnn) file to find the latest time.
n
Specify any positive number for the actual time point from which the ANSYS Multi-field solver will
restart. ANSYS checks the availability of the result set and database file.
Type
Restart type
SING
Use singleframe restart for the ANSYS field solver (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
996
MFRSTART
MULT
Use multiframe restart for the ANSYS field solver. Valid only for MFX.
Notes
For a singleframe restart, you must resume the current database using RESUME before issuing the MFRSTART.
For a multiframe restart, the appropriate .Rnnn files must exist in the working directory. The MULT option
is valid only for MFX analyses. You cannot use MFRSTART,-1,MULT to restart from a diverged or crashed
run. A converged time point must be provided by MFRSTART,TIME,MULT.
For MFX analyses, ANSYS always passes an actual time value to CFX (zero for a new analysis or a positive
value for a restart run) and CFX verifies the consistency with the initial results file. For more details about
ANSYS restart capabilities, please see Restarting an Analysis in the Basic Analysis Guide.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Time Ctrl
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Time Ctrl
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Time Ctrl
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Time Ctrl
MFSORDER, gname1, gname2
Sets up the solution sequence of simultaneous field solver groups for code coupling analyses.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
gname1, gname2
Specifies the group name for groups defined by the MFPSIMUL command with a character string of up
to 80 characters.
Notes
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
997
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFSORDER
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Solution Ctrl
MFSURFACE, INUMB, FNUMB1, Label, FNUMB2
Defines a surface load transfer for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Load Transfer
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
INUMB
Interface number for load transfer. The interface number corresponds to the interface number specified
by the surface flag FSIN (SFxxcommands).
FNUMB1
Field number of sending field.
Label
Valid surface load labels:
Force label: FORC
Displacement label: DISP
Temperature label: TEMP
Heat flux label: HFLU
FNUMB2
Field number for receiving field.
Notes
This command is also valid in PREP7.
The ANSYS Multi-field solver solver does not allow you to switch the load transfer direction for the same
load quantity across the same interfaces for a restart run. For example, if Field1 sends temperature to and
receives heat flow from Field2 across Interface 1 in a previous solution, then you cannot make Field1 send
heat flow to and receive temperatures from Field2 across the same interface in a restart run, even if you
cleared the corresponding load transfer command.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Interface>Surface
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
998
MFSURFACE
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Interface>Surface
MFTIME, TIME
Sets end time for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Time Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
TIME
End time of an ANSYS Multi-field solver analysis. Defaults to 1.
Notes
A MFTIME setting overrides any other end time setting (TIME or FLDATA4).
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Time Ctrl
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Time Ctrl
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Time Ctrl
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Time Ctrl
MFTOL, Key, Value, Toler
Turns normal distance checking on for surface mapping in an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Interface Mapping
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Key
Normal distance key
ON
Activates normal distance checking.
OFF
Deactivates normal distance checking (default).
999
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MFTOL
Value
The normal distance tolerance for surface mapping. Defaults to 1.0e-6. If Toler = REL, Value is dimen-
sionless. If Toler = ABS, Value has the dimensions of length.
Toler
Tolerance definition key
REL
Activates relative gap tolerance, which is independent of units (default).
ABS
Activates absolute gap tolerance.
Notes
For a dissimilar mesh interface, the nodes of one mesh are mapped to the local coordinates of an element
in the other mesh. When normal distance checking is activated, the mapping tool checks the normal distance
from the node to the nearest element. The node is considered improperly mapped if the normal distance
exceeds the tolerance value. The mapping tool creates a component to graphically display the improperly
mapped nodes. See Mapping Diagnostics in the Coupled-Field Analysis Guide for more information.
When using relative gap tolerance (Toler = REL), the normal distance tolerance is derived from the product
of the relative tolerance Value and the largest dimension of the Cartesian bounding box for a specific in-
terface. Therefore, each interface will have a different normal distance tolerance , even though MFTOL is a
global command.
This command is also valid in PREP7.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Product Restrictions
ANSYS Mechanical is available only for an MFX analysis. ANSYS Multiphysics is valid for both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Setup>Global
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Mapping
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Setup>Global
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Advanced Set Up>Mapping
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1000
MFTOL
*MFUN, ParR, Func, Par1
Copies or transposes an array parameter matrix.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting array parameter matrix. See *SET for name restrictions.
Func
Copy or transpose function:
COPY --
Par1 is copied to ParR
TRAN --
Par1 is transposed to ParR. Rows (m) and columns (n) of Par1 matrix are transposed to resulting
ParR matrix of shape (n,m).
Par1
Array parameter matrix input to the operation.
Notes
Operates on one input array parameter matrix and produces one output array parameter matrix according
to:
ParR = f(Par1)
where the function (f ) is either a copy or transpose, as described above.
Functions are based on the standard FORTRAN definitions where possible. ParR may be the same as Par1.
Starting array element numbers must be defined for each array parameter matrix if it does not start at the
first location. For example, *MFUN,A(1,5),COPY,B(2,3) copies matrix B (starting at element (2,3)) to matrix A
(starting at element (1,5)). The diagonal corner elements for each submatrix must be defined: the upper left
corner by the array starting element (on this command), the lower right corner by the current values from
the *VCOL and *VLEN commands. The default values are the (1,1) element and the last element in the
matrix. No operations progress across matrix planes (in the 3rd dimension). Absolute values and scale factors
may be applied to all parameters [*VABS, *VFACT]. Results may be cumulative [*VCUM]. Array elements
should not be skipped with the *VMASK and the NINC value of the *VLEN specifications. The number of
rows [*VLEN] applies to the Par1 array. See the *VOPER command for details.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Matrix Functions
1001
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*MFUN
MFVOLUME, INUMB, FNUMB1, Label, FNUMB2
Defines a volume load transfer for an ANSYS Multi-field solver analysis.
SOLUTION: Multi-field Load Transfer
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
INUMB
Interface number for load transfer. The interface number corresponds to the interface number specified
by the volume flag FVIN (BFE command).
FNUMB1
Field number of sending field.
Label
Valid volume load labels:
Force label: FORC
Displacement label: DISP
Temperature label: TEMP
Heat generation: HGEN
FNUMB2
Field number for receiving field.
Notes
This command is also valid in PREP7.
The ANSYS Multi-field solver solver does not allow you to switch the load transfer direction for the same
load quantity across the same interfaces for a restart run. For example, if Field1 sends temperature to and
receives heat flow from Field2 across Interface 1 in a previous solution, then you cannot make Field1 send
heat flow to and receive temperatures from Field2 across the same interface in a restart run, even if you
cleared the corresponding load transfer command.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFS-Single Code>Interface>Volume
Main Menu>Solution>Multi-field Set Up>MFS-Single Code>Interface>Volume
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1002
MFVOLUME
MFWRITE, Fname, Ext
Writes an ANSYS master input file for MFX multiple code coupling.
SOLUTION: Multi-field Global Controls
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Fname defaults to Jobname.
Ext
Filename extension (8 character maximum).
Ext defaults to dat.
Notes
When working interactively, you need to issue this command as the last step in your setup process. This
command will write out the input file that you will then use to submit the MFX analysis. This file will include
the /SOLU, SOLVE, and FINISH commands.
See Multi-field Commands in the Coupled-Field Analysis Guide for a list of all ANSYS Multi-field solver commands
and their availability for MFS and MFX analyses.
Product Restrictions
ANSYS Mechanical allows MFX analyses but not MFS analyses. ANSYS Multiphysics allows both MFS and MFX
analyses.
Menu Paths
Main Menu>Preprocessor>Multi-field Set Up>MFX-ANSYS/CFX>Write input
Main Menu>Solution>Multi-field Set Up>MFX-ANSYS/CFX>Write input
MGEN, ITIME, INC, NODE1, NODE2, NINC
Generates additional MDOF from a previously defined set.
SOLUTION: Master DOF
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
ITIME, INC
Do this generation operation a total of ITIMEs, incrementing all nodes in the set by INC each time
after the first. ITIME must be > 1 for generation to occur. All previously defined master degree of
freedom directions are included in the set. A component name may also be substituted for ITIME.
1003
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MGEN
NODE1, NODE2, NINC
Generate master degrees of freedom from set beginning with NODE1 to NODE2 (defaults to NODE1) in
steps of NINC (defaults to 1). If NODE1 = ALL, NODE2 and NINC are ignored and set is all selected nodes
[NSEL]. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI).
Notes
Generates additional master degrees of freedom from a previously defined set. If used in SOLUTION, this
command is valid only within the first load step.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Master DOFs>User Selected>Copy
Main Menu>Solution>Master DOFs>User Selected>Copy
MIDTOL, KEY, TOLERB, RESFQ
Sets midstep residual criterion values for structural transient analyses.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
KEY
Midstep residual criterion activation key.
ON or 1
Activate midstep residual criterion in a structural transient analysis (default).
OFF or 0
Deactivate midstep residual criterion in a structural transient analysis.
STAT
List the current midstep residual criterion setting.
TOLERB
Midstep residual tolerance or reference value for bisection. Defaults to 100 times the TOLER setting of
the CNVTOL command.
If TOLERB > 0, it is used as a tolerance about the typical force and/or moment to compare midstep re-
sidual force and/or moment for convergence.
If TOLERB < 0, it is used as a reference force value against which the midstep residual force is compared
for convergence. The reference force value is used to compute a reference moment value for midstep
residual moment comparison.
If midstep residual force and/or moment has not converged and AUTOTS,ON is used, then TOLERB is
also used to predict time step size for bisection.
RESFQ
Key to use response frequency computation along with midstep residual criterion for automatic time
stepping (AUTOTS,ON).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1004
MIDTOL
OFF or 0
Do not calculate response frequency and do not consider it in the automatic time stepping (default).
ON or 1
Calculate response frequency and consider it in the automatic time stepping.
Command Default
For transient structural analysis, the out-of-balance residual is not checked at the midstep.
Notes
When TOLERB is input as a tolerance value (TOLERB > 0), the typical force and/or moment from the regular
time step is used in the midstep residual force and/or moment comparison.
In a structural transient analysis, the suggested tolerance range of TOLERB (TOLERB > 0) is as follows:
TOLERB = 1 to 10 times the TOLER setting of the CNVTOL command for high accuracy solution.
TOLERB = 10 to 100 times the TOLER setting of the CNVTOL command for medium accuracy solution.
TOLERB = more than 100 times the TOLER setting of the CNVTOL command for low accuracy solution.
If the structural transient analysis is elastic and linear, and the load is constant or changes slowly, use a
smaller value of TOLERB to achieve an accurate solution. If the analysis involves large amounts of energy
dissipation, such as elastic-plastic material, TOLERB can be larger. If the analysis includes contact or rapidly
varying loads, a smaller value of TOLERB should be used if high frequency response is important; otherwise,
a larger value of TOLERB may be used to enable faster convergence with larger time step sizes.
For more information on how the midstep criterion is used by the program, see Midstep Residual for Struc-
tural Dynamic Analysis in the Theory Reference for the Mechanical APDL and Mechanical Applications.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Transient
Main Menu>Solution>Analysis Type>Sol'n Controls>Transient
MITER, NEL1, NEL2, RAD, NDIV, ESTRT, EINC
Defines a mitered bend in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NEL1, NEL2
Element numbers of the two intersecting straight pipes. Defaults to the last two straight pipe elements
nearest the intersection of the last two runs.
RAD
Bend radius. If LR, use long radius standard (1.5 x OD) (default). If SR, use short radius standard (1.0 x
OD).
NDIV
Number of divisions (elements) along bend (defaults to 2). A node is generated at the end of each division.
1005
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MITER
ESTRT
Number to be assigned to first element of bend (defaults to MAXEL + 1).
EINC
Element number increment (defaults to 1).
Notes
Defines a mitered bend of piecewise straight pipe elements (PIPE16) in place of the intersection of two
previously defined straight pipe elements [RUN]. This command is similar to the BEND command except
that straight pipe elements are used to form the bend instead of curved (elbow) elements.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Define Pipes>Miter
/MKDIR, Dir
Creates a directory.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Dir
The directory to create (248 characters maximum on Unix and Linux; 233 on Windows). If no path is
provided, it will be created in the current working directory. Must be a valid name (and path) for the
system you are working on.
Notes
Creates a directory on the computer ANSYS is currently running on.
Menu Paths
This command cannot be accessed from a menu.
MLIST, NODE1, NODE2, NINC
Lists the MDOF of freedom.
SOLUTION: Master DOF
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE1, NODE2, NINC
List master degrees of freedom from NODE1 to NODE2 (defaults toNODE1) in steps of NINC (defaults
to 1). If NODE1 = ALL (default), NODE2 and NINC are ignored and masters for all selected nodes [NSEL]
are listed. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NODE1 (NODE2 and NINC are
ignored).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1006
/MKDIR
Notes
Lists the master degrees of freedom. Master degrees of freedom generated from the TOTAL command
cannot be listed until after the first load step.
Menu Paths
Main Menu>Preprocessor>Loads>Master DOFs>User Selected>List All
Main Menu>Preprocessor>Loads>Master DOFs>User Selected>List Picked
Main Menu>Solution>Master DOFs>User Selected>List All
Main Menu>Solution>Master DOFs>User Selected>List Picked
Utility Menu>List>Other>Master DOF>At All Nodes
Utility Menu>List>Other>Master DOF>At Picked Nodes
MMASS, Option, ZPA
Specifies the missing mass response calculation.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Option
Flag to activate or deactivate missing mass response calculation.
1 (ON or YES)
Activate.
0 (OFF or NO)
Deactivate (this is the default value)
ZPA
Zero Period Acceleration Value. If a scale factor FACT is defined in the SVTYP command, it applies to
this value.
Notes
For more information on spectrum analysis combination methods, see Combination of Modes
Missing mass calculation is valid only for single point excitation response spectrum analysis (SPOPT, SPRS)
and for multiple point response spectrum analysis (SPOPT, MPRS) performed with base excitation using
acceleration response spectrum loading. Missing mass is supported in a spectrum analysis only when the
preceding modal analysis is performed with Block Lanczos, PCG Lanczos or Supernode eigensolver (Method
=LANB, LANPCG, or SNODE on the MODOPT command).
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Menu paths generated with the final build.
1007
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MMASS
MMF
Calculates the magnetomotive force along a path.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
MMF invokes an ANSYS macro which calculates the magnetomotive force (mmf ) along a predefined path
[PATH]. It is valid for both 2-D and 3-D magnetic field analyses. The calculated mmf value is stored in the
parameter MMF.
A closed path [PATH], passing through the magnetic circuit for which mmf is to be calculated, must be
defined before this command is issued. A counterclockwise ordering of points on the PPATH command will
yield the correct sign on the mmf. The mmf is based on Ampere's Law. The macro makes use of calculated
values of field intensity (H), and uses path operations for the calculations. All path items are cleared upon
completion. The MMF macro sets the "ACCURATE" mapping method and "MAT" discontinuity option of the
PMAP command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>MMF
MODCONT, MLSkey
Create multiple load vectors in a modal analysis.
SOLUTION: Dynamic Options
MP ME ST PR PRN DS DSS <> <> EH <> PP <> EME MFS
MLSkey
Multiple Load Step Key:
OFF
Perform the modal analysis for each load step. (default)
ON
Perform the modal analysis only for the first load step; form the load vector for each load step and
write it to the *.mode file.
Notes
Use the LVSCALE command to apply the desired load in a mode superposition transient or harmonic ana-
lysis.
Specifies additional modal analysis (ANTYPE,MODAL) options.
Multiple load steps in MODCONT are not supported for the PSOLVE command.
This command applies only to the Block Lanczos, PCG Lanczos, and Supernode methods.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1008
MMF
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
To be provided at time of release.
MODE, MODE, ISYM
Specifies the harmonic loading term for this load step.
SOLUTION: Load Step Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
MODE
Number of harmonic waves around circumference for this harmonic loading term (defaults to 0).
ISYM
Symmetry condition for this harmonic loading term (not used when MODE = 0):
1
Symmetric (UX, UY, ROTZ, TEMP use cosine terms; UZ uses sine term) (default).
-1
Antisymmetric (UX, UY, ROTZ, temp use sine terms; UZ uses cosine term).
Command Default
MODE = 0, ISYM = 1.
Notes
Used with axisymmetric elements having nonaxisymmetric loading capability (e.g., PLANE25, SHELL61, FLUID81,
etc.). For analysis types ANTYPE,MODAL, HARMIC, TRANS, and SUBSTR, the term must be defined in the first
load step and may not be changed in succeeding load steps.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>For Harmonic Ele
Main Menu>Solution>Load Step Opts>Other>For Harmonic Ele
1009
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MODE
MODIFY, SET, LSTEP, ITER, CUMIT, TIME, Ktitle
Changes the listed values of the data in a set.
AUX3: Results Files
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
SET
Set of data in results file to be modified.
LSTEP
The new load step number.
ITER
The new load substep number.
CUMIT
The new cumulative iteration.
TIME
The new time/frequency value.
Ktitle
Indicates if the set title should be modified.
0
Keep the original title.
1
Change the title to the title specified with the most current /TITLE command.
Notes
Use this command to change the listed values in a data set in a results file. Using this command does not
change any actual model data; it affects only the values listed in the results file.
For example, if you start with the following results file:
SET TIME/FREQ LOAD STEP SUBSTEP CUMULATIVE
1 1.0000 1 1 1
first load set
2 2.0000 2 1 2
second load set
3 3.0000 3 1 3
third load set
4 4.0000 4 1 4
fourth load set
and you then issue the following commands:
/title, modified title for set number 3
modify,3,5,2,5,4.5,1
The modified results file would look like this:
SET TIME/FREQ LOAD STEP SUBSTEP CUMULATIVE
1 1.0000 1 1 1
first load set
2 2.0000 2 1 2
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1010
MODIFY
second load set
3 4.5000 5 2 5
modified title for set number 3
4 4.0000 4 1 4
fourth load set
Menu Paths
Main Menu>Preprocessor>Loads>Other>For Harmonic Ele
Main Menu>Solution>Other>For Harmonic Ele
MODMSH, Lab
Controls the relationship of the solid model and the FE model.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Relationship key:
STAT
Gives status of command (default). This applies only to the CHECK option (no status is provided for
the DETACH option).
NOCHECK
Deactivates the checking of the solid model and the finite element model. Allows elements and
nodes generated with the mesh commands to be modified directly (EMODIF, NMODIF, EDELE,
NDELE, etc.). Also deactivates solid model hierarchical checking so that areas attached to volumes
may be deleted etc.
CHECK
Reactivates future checking of the solid model.
DETACH
Releases all associativity between the current solid model and finite element model. ANSYS deletes
any element attributes that were assigned to the affected solid model entities through default attrib-
utes (that is, through the TYPE, REAL, MAT, SECNUM, and ESYS command settings and a subsequent
meshing operation). However, attributes that were assigned directly to the solid model entities (via
the KATT, LATT, AATT, and VATT commands) are not deleted.
Notes
Affects the relationship of the solid model (keypoints, lines, areas, volumes) and the finite element model
(nodes, elements, and boundary conditions).
Specify Lab = NOCHECK carefully. By deactivating checking, the solid model database can be corrupted by
subsequent operations.
After specifying Lab = DETACH, it is no longer possible to select or define finite element model items in
terms of the detached solid model or to clear the mesh.
1011
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MODMSH
Menu Paths
Main Menu>Preprocessor>Checking Ctrls>Model Checking
MODOPT, Method, NMODE, FREQB, FREQE, Cpxmod/PRMODE, Nrmkey, --, BlockSize
Specifies modal analysis options.
SOLUTION: Dynamic Options
MP ME ST PR PRN DS DSS <> <> EH <> PP <> EME MFS
Product Restrictions
Method
Mode extraction method to be used for the modal analysis.
LANB
Block Lanczos
LANPCG
PCG Lanczos
SNODE
Supernode modal solver.
REDUC
Householder (reduced)
UNSYM
Unsymmetric matrix. This option cannot be followed by a subsequent spectrum analysis.
DAMP
Damped system. This option cannot be followed by a subsequent spectrum analysis.
QRDAMP
Damped system using QR algorithm. This option cannot be followed by a subsequent spectrum
analysis.
VT
Solve with the Variational Technology method of ANSYS DesignXplorer.
NMODE
The number of modes to extract. The value can depend on the value supplied for Method. Defaults to
the number of master DOFs when Method = REDUC. For the other methods, NMODE has no default and
must be specified. If Method = LANB, LANPCG, or SNODE, the number of modes that can be extracted
can equal the DOFs in the model after the application of all boundary conditions.
Recommendation:
When Method = REDUC, NMODE should be less than half of the number of master DOFs.
When Method = LANPCG, NMODE should be less than 100 to be computationally efficient.
When Method = SNODE, NMODE should be greater than 100 for 2-D plane or 3-D shell/beam models
and greater than 250 for 3-D solid elements to be computationally efficient.
FREQB
The beginning, or lower end, of the frequency range of interest.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1012
MODOPT
For Method = LANB, UNSYM, DAMP, and QRDAMP, FREQB also represents the first shift point for the
eigenvalue iterations. If values for UNSYM or DAMP are zero or blank, the default value is -1.0. For the
other methods, the default is internally computed. Eigenvalue extraction is most accurate near the shift
point; multiple shift points are used internally in the LANB and QRDAMP methods. For LANB, LANPCG,
UNSYM, DAMP, and QRDAMP methods with a positive FREQB, eigenvalues are output beginning at the
shift point and increase in magnitude. For UNSYM and DAMP methods with a negative FREQB value,
eigenvalues are output beginning at zero magnitude and increase.
Choosing higher FREQB values with the LANPCG and SNODE methods may lead to inefficient solution
times because these methods will find all eigenvalues between zero and FREQB before finding the re-
quested modes between FREQB and FREQE.
FREQE
The ending, or upper end, of the frequency range of interest (in Hz). The default is 1e8 for Method =
LANB, LANPCG, and QRDAMP. The default for Method = SNODE is described below. The default for all
other methods is to calculate all modes, regardless of their maximum frequency.
The default is 100 Hz for Method = SNODE. To maintain solution efficiency, you should not set the
FREQE value too high; for example, not higher than 5000 Hz for an industrial problem. The higher the
FREQE value used for the SNODE method, the more solution time it will take and the more eigenvalues
it could produce. For example, if FREQE is set to 1e8, it will cause the underlying supernodal structures
to find all the possible eigenvalues of each group of supernodes; hence, it will take an excessive amount
of solution time.
Cpxmod/PRMODE
CPXMOD (Valid only when Method = QRDAMP).
ON
Calculate complex eigenmode shapes.
OFF
Do not calculate complex eigenmode shapes. This is the default.
PRMODE
The number of reduced modes to print. Valid only when Method = REDUC.
Nrmkey
Mode shape normalization key:
OFF
Normalize the mode shapes to the mass matrix (default).
ON
Normalize the mode shapes to unity instead of to the mass matrix. If a subsequent spectrum or
mode superposition analysis is planned, the mode shapes should be normalized to the mass matrix
(Nrmkey = OFF).
--
Unused field.
BlockSize
The block vector size to be used with the Block Lanczos eigensolver (used only when Method = LANB).
BlockSize must be an integer value between 0 and 16. When BlockSize = zero or blank, the code
decides the block size internally (normally, a value of 8 is used). Typically, higher BlockSize values
are more efficient under each of the following conditions:
1013
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MODOPT
When running in out-of-core mode and there is not enough physical memory to buffer all of the
files written by the Block Lanczos eigensolver (and thus, the time spent doing I/O is considerable).
Many modes are requested (>100).
Higher-order solid elements dominate the model.
The memory usage only slightly increases as BlockSize is increased. It is recommended that you use
a value divisible by 4 (4, 8, 12, or 16).
Notes
Specifies modal analysis (ANTYPE,MODAL) options. Additional options used only for the Supernode (SNODE)
eigensolver are specified by the SNOPTION command. If Method = LANPCG, ANSYS automatically switches
to the PCG solver internally for this modal analysis. You can further control the efficiency of the PCG solver
with the PCGOPT and EQSLV commands. Method = LANPCG is available in both shared-memory ANSYS
and Distributed ANSYS.
The Block Lanczos method is strongly recommended for high-frequency magnetic eigenvalue problems. The
initial frequency guess is not critical. The ratio of FREQE to FREQB can be up to 1e6. The PCG Lanczos
method is not supported for high-frequency magnetic eigenvalue problems.
For models that involve a non-symmetric element stiffness matrix, as in the case of a contact element with
frictional contact, the QR damp eigensolver (MODOPT, QRDAMP) extracts modes in the modal subspace
formed by the eigenmodes from the symmetrized eigenproblem. The QR damp eigensolver symmetrizes
the element stiffness matrix on the first pass of the eigensolution, and in the second pass, eigenmodes are
extracted in the modal subspace of the first eigensolution pass. For such non-symmetric eigenproblems,
you should verify the eigenvalue and eigenmode results using the non-symmetric matrix eigensolver
(MODOPT, UNSYM ).
This command is also valid in PREP7.
Product Restrictions
Available Products Command
Option
Method
MP ME ST PR PRN DS DSS <> <> EH <> PP <> EME MFS LANB
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS LANPCG
MP ME ST PR PRN <> <> <> <> EH <> PP <> EME MFS SNODE
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS REDUC
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS UNSYM
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS DAMP
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS QRDAMP
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <> VT
Menu Paths
Main Menu>DesignXplorer>Solution>Solve
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1014
MODOPT
MONITOR, VAR, Node, Lab
Controls contents of three variable fields in nonlinear solution monitor file.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
VAR
One of three variable field numbers in the monitor file whose contents can be specified by the Lab field.
Valid arguments are integers 1, 2, or 3. See Notes section for default values.
Node
The node number for which information is monitored in the specified VAR field. In the GUI, if Node =
P, graphical picking is enabled. If blank, the monitor file lists the maximum value of the specified
quantity (Lab field) for the entire structure.
Lab
The solution quantity to be monitored in the specified VAR field. Valid labels for solution quantities are
UX, UY, and UZ (displacements); ROTX, ROTY, and ROTZ (rotations); and TEMP (temperature). Valid labels
for reaction force are FX, FY, and FZ (structural force) and MX, MY, and MZ (structural moment). Valid
label for heat flow rate is HEAT. For defaults see the Notes section.
Notes
This command is only active when SOLCONTROL,ON.
The monitor file is an ASCII file which is automatically created and saved when SOLCONTROL is active (ON).
The monitor file always has an extension of .mntr, and takes its file name from the specified Jobname. If
no Jobname is specified, the file name defaults to file.
You must issue this command once for each solution quantity you want to monitor at a specified node at
each load step. You cannot monitor a reaction force during a linear analysis. The variable field contents can
be redefined at each load step by reissuing the command. The monitored quantities are appended to the
file for each load step.
The following example shows the format of a monitor file. Note that the file only records the solution substep
history when a substep is convergent.
SOLUTION HISTORY INFORMATION FOR JOB: file.mntr
LOAD SUB- NO. NO. TOTL INCREMENT TOTAL VARIAB 1 VARIAB 2 VARIAB 3
STEP STEP ATTMP ITER ITER TIME/LFACT TIME/LFACT MONITOR MONITOR MONITOR
UZ MZ MxPl
1 1 1 3 3 0.25000 0.25000 1.4145 0.19076E-06 0.78886E-30
1 2 1 2 5 0.25000 0.50000 2.8283 0.92989E-06 0.78886E-30
1 3 1 2 7 0.37500 0.87500 4.9467 0.33342E-05 0.78886E-30
1 4 1 2 9 0.12500 1.0000 5.6519 0.16826E-05 0.78886E-30
2 1 1 6 15 0.20000E-03 1.0002 4.2198 515.23 0.78886E-30
2 2 2 6 26 0.10000E-03 1.0003 4.4849 593.03 0.78886E-30
2 3 1 3 29 0.10000E-03 1.0004 4.7531 611.45 0.78886E-30
2 4 1 3 32 0.15000E-03 1.0006 5.0696 621.83 0.78886E-30
2 5 1 4 36 0.22500E-03 1.0008 5.4428 628.42 0.78886E-30
2 6 1 4 40 0.33750E-03 1.0011 5.8928 632.78 0.78886E-30
2 7 1 5 45 0.50625E-03 1.0016 6.4454 635.62 0.78886E-30
2 8 1 7 52 0.75938E-03 1.0024 7.1375 637.22 0.78886E-30
2 9 1 5 57 0.75938E-03 1.0031 7.7422 637.66 0.78886E-30
2 10 1 6 63 0.11391E-02 1.0043 8.5588 637.42 0.78886E-30
2 11 2 3 72 0.76887E-03 1.0050 9.0721 636.96 0.78886E-30
1015
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MONITOR
2 12 1 3 75 0.76887E-03 1.0058 9.5648 636.35 0.78886E-30
2 13 1 3 78 0.11533E-02 1.0070 10.277 635.25 0.78886E-30
2 14 1 4 82 0.17300E-02 1.0087 11.306 633.37 0.78886E-30
2 15 1 6 88 0.25949E-02 1.0113 12.802 630.21 0.78886E-30
2 16 1 5 93 0.25949E-02 1.0139 14.273 626.81 0.78886E-30
2 17 1 7 100 0.38924E-02 1.0178 16.477 621.42 0.78886E-30
2 18 1 6 106 0.38924E-02 1.0217 18.704 615.77 0.78886E-30
2 19 2 4 116 0.26274E-02 1.0243 20.229 611.83 0.78886E-30
2 20 1 4 120 0.26274E-02 1.0269 21.777 607.80 0.78886E-30
The following details the contents of the various fields in the monitor file:
LOAD STEP
The current load step number.
SUBSTEP
The current substep (time step) number.
NO. ATTEMPT
The number of attempts made in solving the current substep. This number is equal to the number of
failed attempts (bisections) plus one (the successful attempt).
NO. ITER
The number of iterations used by the last successful attempt.
TOTL. ITER
Total cumulative number of iterations (including each iteration used by a bisection).
INCREMENT
TIME/LFACT
Time or load factor increments for the current substep.
TOTAL TIME/LFACT
Total time (or load factor) for the last successful attempt in the current substep.
VARIAB 1
Variable field 1. In this example, the field is reporting the UZ value. By default, this field lists the CPU
time used up to (but not including) the current substep.
VARIAB 2
Variable field 2. In this example, the field is reporting the MZ value. By default, this field lists the maximum
displacement in the entire structure.
VARIAB 3
Variable field 3. By default (and in the example), this field reports the maximum equivalent plastic strain
in the entire structure.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Monitor
Main Menu>Solution>Load Step Opts>Nonlinear>Monitor
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1016
MONITOR
*MOPER, ParR, Par1, Oper, Par2, Par3, kDim, --, kOut, LIMIT
Performs matrix operations on array parameter matrices.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting array parameter matrix. See *SET for name restrictions.
Par1
First array parameter matrix input to the operation. For Oper = MAP, this is an N x 3 array of coordinate
locations at which to interpolate. ParR will then be an N(out) x M array containing the interpolated
values.
Oper
Matrix operations:
INVERT --
(*MOPER, ParR, Par1, INVERT)
Square matrix invert: Inverts the n x n matrix in Par1 into ParR. The matrix must be well conditioned.
Warning
Non-independent or ill-conditioned equations can cause erroneous results.
MULT --
(*MOPER, ParR, Par1, MULT, Par2)
Matrix multiply: Multiplies Par1 by Par2. The number of rows of Par2 must equal the number of
columns of Par1 for the operation. If Par2 is input with a number of rows greater than the number
of columns of Par1, matrices are still multiplied. However, the operation only uses a number of
rows of Par2 equal to the number of columns of Par1.
COVAR --
(*MOPER, ParR, Par1, COVAR, Par2)
Covariance: The measure of association between two columns of the input matrix (Par1). Par1, of
size m runs (rows) by n data (columns) is first processed to produce a row vector containing the
mean of each column which is transposed to a column vector (Par2) of n array elements. The Par1
and Par2 operation then produces a resulting n x n matrix (ParR) of covariances (with the variances
as the diagonal terms).
CORR --
(*MOPER, ParR, Par1, CORR, Par2)
Correlation: The correlation coefficient between two variables. The input matrix (Par1), of size m
runs (rows) by n data (columns), is first processed to produce a row vector containing the mean of
each column which is then transposed to a column vector (Par2) of n array elements. The Par1
and Par2 operation then produces a resulting n x n matrix (ParR) of correlation coefficients (with
a value of 1.0 for the diagonal terms).
1017
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*MOPER
SOLV --
(*MOPER, ParR, Par1, SOLV, Par2)
Solution of simultaneous equations: Solves the set of n equations of n terms of the form a
n1
x
1
+
a
n2
x
2
+
...
+ a
nn
x
n
= b
n
where Par1 contains the matrix of a-coefficients, Par2 the vector(s) of b-
values, and ParR the vector(s) of x-results. Par1 must be a square matrix. The equations must be
linear, independent, and well conditioned.
Warning
Non-independent or ill-conditioned equations can cause erroneous results.
SORT --
(*MOPER, ParR, Par1, SORT, Par2)
Matrix sort: Sorts matrix Par1 according to sort vector Par2 and places the result in Par1. Rows
of Par1 are moved to the corresponding positions indicated by the values of Par2. Par2 may be
a column of Par1 (in which case it will also be reordered). ParR is the vector of initial row positions.
Sorting Par1 according to ParR should reproduce the initial ordering.
NNEAR --
(*MOPER, ParR, Par1, NNEAR, Toler)
Nearest Node: Quickly determine all the nodes within a specified tolerance of a given array.
ParR is a vector of the nearest selected nodes, or 0 if no nodes are nearer than Toler. Par1 is the
n x 3 array of coordinate locations.
ENEAR --
(*MOPER, ParR, Par1, ENEAR, Toler)
Nearest Element: Quickly determine the elements with centroids that are within a specified tolerance
of the points in a given array.
ParR is a vector of the nearest selected elements, or 0 if no element centroids are nearer than Toler.
Par1 is the n x 3 array of coordinate locations.
MAP --
(*MOPER, ParR, Par1, MAP, Par2, Par3, kDim, kOut, LIMIT)
Maps the results from another program onto your ANSYS finite element model. For example, you
can map pressures from a CFD analysis onto your model for a structural analysis.
When you map results, the subsequent Par2 and Par3 arguments define your input values and
their locations, and the arguments that follow determine the search area and interpolation schemes
(see below).
For Oper = MAP, output points are incorrect if they are not within the boundaries (area or volume)
set via the specified input points. Also, calculations for out-of-bound points require much more
processing time than do points that are within bounds.
When mapping results from one analysis to another (Oper = MAP), Par1 will be your final N(out)
x 3 array of points. Par2 will be an N(in) x M array that corresponds to the points in Par3. For each
point in the destination mesh, all possible triangles in the source mesh are searched to find the best
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1018
*MOPER
triangle containing each point. It then does a linear interpolation inside this triangle. You should
carefully specify your interpolation method and search criteria in order to provide faster and more
accurate results (see LIMIT, below). Results mapping (Oper = MAP) is available from the command
line only.
Par2
Second array parameter matrix input to the operation. For the COVAR and CORR operations, this para-
meter must exist as a dimensioned array vector without specified values since its values (means) will be
calculated as part of the operations. For MAP, this will be an {N(in) x M} array of values to be interpolated,
where N(in) is the number of points to interpolate from, and M is the number of values at each point.
For the ENEAR and NNEAR operations, this parameter specifies the tolerance for the search.
Par3
Third array parameter, used for Oper = MAP. This is an N x 3 array of coordinate locations corresponding
to the values in Par2.
kDim
Interpolation criteria; used for Oper = MAP:
If kDim = 2 or 0, two dimensional interpolation is applied (interpolate on a surface).
If kDim = 3, three dimensional interpolation is applied (interpolate on a volume).
--
Unused field
kOut
Outside region results; used for Oper = MAP
If kOut = 0, use the value(s) of the nearest region point for points outside of the region.
If kOut = 1, set results extrapolated outside of the region to zero.
LIMIT
Number of nearby nodes considered for interpolation (Oper = MAP). Minimum = 5, default = 20. Lower
values will reduce processing time, however, some distorted or irregular meshes will require a higher
LIMIT value to encounter three nodes for triangulation.
Notes
Each starting array element number must be defined for each array parameter matrix if it does not start at
the first location. For example, *MOPER,A(2,3),B(1,4),MULT,C(1,5) multiplies submatrix B (starting at element
(1,4)) by submatrix C (starting at element (1,5)) and puts the result in matrix A (starting at element (2,3)).
The diagonal corner elements for each submatrix must be defined: the upper left corner by the array starting
element (on this command), the lower right corner by the current values from the *VCOL and *VLEN com-
mands. The default values are the (1,1) element and the last element in the matrix. No operations progress
across matrix planes (in the 3rd dimension). Absolute values and scale factors may be applied to all parameters
[*VABS, *VFACT]. Results may be cumulative [*VCUM]. Array elements should not be skipped with the
*VMASK and the NINC value of the *VLEN specifications. See the *VOPER command for details.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Matrix Operations
1019
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*MOPER
MOPT, Lab, Value
Specifies meshing options.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Meshing option to be specified (dictates the meaning of Value):
AORDER
Mesh by ascending area size order. Set Value to ON to mesh smaller areas first. Using this results
in finer meshes in critical areas for volume meshes; this can be used for cases where SMRTSIZE does
not mesh as needed. Default is OFF.
EXPND
Area mesh expansion (or contraction) option. (This option is the same as SMRTSIZE,,,EXPND.) This
option is used to size internal elements in an area based on the size of the elements on the area's
boundaries. Value is the expansion (or contraction) factor. For example, issuing MOPT,EXPND,2
before meshing an area will allow a mesh with elements that are approximately twice as large in
the interior of an area as they are on the boundary. If Value is less than 1, a mesh with smaller
elements on the interior of the area will be allowed. Value for this option should be greater than
0.5 but less than 4. Value defaults to 1, which does not allow expansion or contraction of internal
element sizes (except when using AESIZE sizing). If Value = 0, the default value of 1 will be used.
The actual size of the internal elements will also depend on the TRANS option, or upon AESIZE or
ESIZE sizing, if used.
TETEXPND
Tet-mesh expansion (or contraction) option. This option is used to size internal elements in a volume
based on the size of the elements on the volume's boundaries. Value is the expansion (or contraction)
factor. For example, issuing MOPT,TETEXPND,2 before meshing a volume will allow a mesh with
elements that are approximately twice as large in the interior of the volume as they are on the
boundary. If Value is less than 1, a mesh with smaller elements on the interior of the volume will
be allowed. Value for this option should be greater than 0.1 but less than 3. Value defaults to 1,
which does not allow expansion or contraction of internal element sizes. If Value = 0, the default
value of 1 will be used. If Value is greater than 2, mesher robustness may be affected. The TETEXPND
option is supported for both the VMESH and FVMESH commands. Tet-mesh expansion is the only
mesh control supported by FVMESH.
TRANS
Mesh transition option. (This option is the same as SMRTSIZE,,,,TRANS.) This option is used to control
how rapidly elements are permitted to change in size from the boundary to the interior of an area.
Value is the transitioning factor. Value defaults to 2.0, which permits elements to approximately
double in size as they approach the interior of the area. (If Value = 0, the default value of 2 will be
used.) Value must be greater than 1 and, for best results, should be less than 4. The actual size of
the internal elements will also depend on the EXPND option, or upon AESIZE or ESIZE sizing, if used.
AMESH
Triangle surface meshing option. Valid inputs for Value are:
DEFAULT
Let ANSYS choose which triangle mesher to use. In most cases, ANSYS will choose the main triangle
mesher, which is the Riemann space mesher. If the chosen mesher fails for any reason, ANSYS
invokes the alternate mesher and retries the meshing operation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1020
MOPT
MAIN
ANSYS uses the main triangle mesher (Riemann space mesher), and it does not invoke an alternate
mesher if the main mesher fails. The Riemann space mesher is well suited for most surfaces.
ALTERNATE
ANSYS uses the first alternate triangle mesher (3-D tri-mesher), and it does not invoke another
mesher if this mesher fails. This option is not recommended due to speed considerations. However,
for surfaces with degeneracies in parametric space, this mesher often provides the best results.
ALT2
ANSYS uses the second alternate triangle mesher (2-D parametric space mesher), and it does not
invoke another mesher if this mesher fails. This option is not recommended for use on surfaces
with degeneracies (spheres, cones, etc.) or poorly parameterized surfaces because poor meshes
may result.
QMESH
Quadrilateral surface meshing option. (Quadrilateral surface meshes will differ based on which triangle
surface mesher is selected. This is true because all free quadrilateral meshing algorithms use a triangle
mesh as a starting point.) Valid inputs for Value are:
DEFAULT
Let ANSYS choose which quadrilateral mesher to use. In most cases, ANSYS will choose the main
quadrilateral mesher, which is the Q-Morph (quad-morphing) mesher. For very coarse meshes,
ANSYS may choose the alternate quadrilateral mesher instead. In most cases, the Q-Morph
mesher results in higher quality elements. If either mesher fails for any reason, ANSYS invokes
the other mesher and retries the meshing operation. (Default.)
MAIN
ANSYS uses the main quadrilateral mesher (Q-Morph mesher), and it does not invoke the alternate
mesher if the main mesher fails.
ALTERNATE
ANSYS uses the alternate quadrilateral mesher, and it does not invoke the Q-Morph mesher if
the alternate mesher fails. To use the alternate quadrilateral mesher, you must also select
MOPT,AMESH,ALTERNATE or MOPT,AMESH,ALT2.
VMESH
Tetrahedral element meshing option. Valid inputs for Value are:
DEFAULT
Let ANSYS choose which tetrahedra mesher to use. ANSYS always uses the alternate tetrahedra
mesher when meshing with p-elements. Otherwise, it usually uses the main tetrahedra mesher.
MAIN
Use the main tetrahedra mesher (Delaunay technique mesher). (GHS3D meshing technology by
P. L. George, INRIA, France.) For most models, this mesher is significantly faster than the alternate
mesher.
ALTERNATE
Use the alternate tetrahedra mesher (advancing front mesher). This mesher is the ANSYS Revision
5.2 tetrahedra mesher. It does not support the generation of a tetrahedral volume mesh from
facets [FVMESH]. If this mesher is selected and you issue the FVMESH command, ANSYS uses
the main tetrahedra mesher to create the mesh from facets and issues a warning message to
notify you.
SPLIT
Quad splitting option for non-mapped meshing. If Value = 1, ON, or ERR, quadrilateral elements in
violation of shape error limits are split into triangles (default). If Value = 2 or WARN, quadrilateral
1021
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MOPT
elements in violation of either shape error or warning limits are split into triangles. If Value = OFF,
splitting does not occur, regardless of element quality.
LSMO
Line smoothing option. Value can be ON or OFF. If Value = ON, smoothing of nodes on area
boundaries is performed during smoothing step of meshing. During smoothing, node locations are
adjusted to achieve a better mesh. If Value = OFF (default), no smoothing takes place at area
boundaries.
CLEAR
This option affects the element and node numbering after clearing a mesh. If Value = ON (default),
the starting node and element numbers will be the lowest available number after the nodes and
elements are cleared. If Value = OFF, the starting node and element numbers are not reset after
the clear operation (which was the default behavior for ANSYS versions prior to Release 5.3).
PYRA
Transitional pyramid elements option. Value can be ON or OFF. If Value = ON (default), ANSYS
automatically creates transitional pyramid elements, when possible. Pyramids may be created at the
interface of tetrahedral and hexahedral elements, or directly from quadrilateral elements. For pyramids
to be created, you must also issue the command MSHAPE,1,3D (degenerate 3-D elements). If Value
= OFF, ANSYS does not create transitional pyramid elements.
TIMP
Identifies the level of tetrahedra improvement to be performed when the next free volume meshing
operation is initiated [VMESH, FVMESH]. (For levels 2-5, improvement occurs primarily through the
use of face swapping and node smoothing techniques.) Valid inputs for Value are:
0
Turn off tetrahedra improvement. Although this value can lead to faster tetrahedral mesh creation,
it is not recommended because it often leads to poorly shaped elements and mesh failures.
1
Do the minimal amount of tetrahedra improvement. (Default.) This option is supported by the
main tetrahedra mesher only [MOPT,VMESH,MAIN]. If the alternate tetrahedra mesher
[MOPT,VMESH,ALTERNATE] is invoked with this setting, ANSYS automatically performs tetrahedra
improvement at level 3 instead [MOPT,TIMP,3].
2
Perform the least amount of swapping/smoothing. No improvement occurs if all tetrahedral
elements are within acceptable limits.
3
Perform an intermediate amount of swapping/smoothing. Some improvement is always done.
4
Perform the greatest amount of swapping/smoothing. Meshing takes longer with this level of
improvement, but usually results in a better mesh.
5
Perform the greatest amount of swapping/smoothing, plus additional improvement techniques.
This level of improvement usually produces results that are similar to those at level 4, except for
very poor meshes.
6
For linear tetrahedral meshes, this value provides the same level of improvement as MOPT,TIMP,5.
For quadratic tetrahedral meshes, this value provides an additional pass of cleanup. This value
is supported for both the main [MOPT,VMESH,MAIN] and alternate [MOPT,VMESH,ALTERNATE]
tetrahedra meshers.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1022
MOPT
STAT
Display status of MOPT settings. Value is ignored.
DEFA
Set all MOPT options to default values. Value is ignored.
Value
Value, as described for each different Lab above.
Notes
See the Modeling and Meshing Guide for more information on the MOPT command and its options.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesher Opts
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Global>Area Cntrls
Main Menu>Preprocessor>Meshing>Size Cntrls>ManualSize>Global>Volu Cntrls
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Global Meshing
Options
Utility Menu>List>Status>Preprocessor>Solid Model
MORPH, Option, --, Remeshopt, ElemSet, ARMAX, VOCH, ARCH, STEP, TIME
Specifies morphing and remeshing controls.
PREP7: Morphing
MP ME <> <> <> <> <> <> <> <> <> PP <> EME MFS
Option
OFF
Turns off morphing for field elements (default).
ON
Turns on morphing for field elements.
--
Unused field
Remeshopt
OFF
Do not remesh (default).
ON
Remesh when element qualities fall below values specified by ARMAX, VOCH, or ARCH as explained
below. Valid only when Option is ON.
ElemSet
ALL
Remesh all selected elements if the quality of the worst defined element falls below any quality re-
quirement (default when Remeshopt = ON).
1023
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MORPH
CompName
Specify a component name, up to 8 characters. All elements included in this component name are
remeshed if the quality of the worst element falls below any quality requirement.
ARMAX
The maximum allowable element generalized aspect ratio. Defaults to 5. See Element Qualities in the
Fluids Analysis Guide for the maximum generalized aspect ratio.
VOCH
The maximum allowable change of element size (area or volume). Defaults to 3. See Element Qualities
in the Fluids Analysis Guide for maximum allowable change of element size.
ARCH
The maximum allowable element aspect ratio change. Defaults to 3. See Element Qualities in the Fluids
Analysis Guide for maximum allowable element aspect ratio change.
STEP
The frequency of element quality checking, based on time steps. A quality check takes place at the intervals
defined by STEP. Defaults to 1 (quality check at every step).
TIME
A quality check takes place at the time point specified. Defaults to -1 (a quality check at every time
point).
Notes
MORPH is applicable to any non-structural field analysis (not including fluid elements). It activates displace-
ment degrees of freedom for non-structural elements so that boundary conditions may be placed on the
field mesh to constrain the movement of the non-structural mesh during morphing. It morphs the non-
structural mesh using displacements transferred at the surface interface between the structural field and
the non-structural field. The displacements of non-structural elements are mesh displacements to avoid
mesh distortion, but have no physical meaning except at the interface. MORPH does not support surface,
link, or shell elements, or any element shape other than triangles, quads, tets, and bricks. Morphed fields
must be in the global Cartesian system (CSYS = 0).
After each remesh, new databases and results files are written with the extensions .rth0n and .db0n,
where n is the remesh file number (FieldName.rth01, FieldName.rth02, ... and FieldName.db01,
FieldName.db02, etc.). The original database file is FieldName.dbo. The FieldName.db01, Field-
Name.db02, etc. files have elements that are detached from the solid model.
Remeshing has the following restrictions:
Valid only for the electrostatic elements (PLANE121, SOLID122, and SOLID123)
Limited to triangle (2-D) and tetrahedral (3-D) options of these elements
Valid only for the MFS solver
No body loads allowed in the interior nodes of the remeshing domain
Nodes on the boundary cannot be remeshed; remeshing will not work if morphing failed on the surface
nodes
Not suitable for extreme area or volume changes
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1024
MORPH
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Element Morphing
Main Menu>Solution>Load Step Opts>Other>Element Morphing
MOVE, NODE, KC1, X1, Y1, Z1, KC2, X2, Y2, Z2
Calculates and moves a node to an intersection.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Move this node. If NODE = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NODE.
KC1
First coordinate system number. Defaults to 0 (global Cartesian).
X1, Y1, Z1
Input one or two values defining the location of the node in this coordinate system. Input "U" for unknown
value(s) to be calculated and input "E" to use an existing coordinate value. Fields are R1, 1, Z1 for cyl-
indrical, or R1, 1, 1 for spherical or toroidal.
KC2
Second coordinate system number.
X2, Y2, Z2
Input two or one value(s) defining the location of the node in this coordinate system. Input "U" for un-
known value(s) to be calculated and input "E" to use an existing coordinate value. Fields are R2, 2, Z2
for cylindrical, or R2, 2, 2 for spherical or toroidal.
Notes
Calculates and moves a node to an intersection location. The node may have been previously defined (at
an approximate location) or left undefined (in which case it is internally defined at the SOURCE location).
The actual location is calculated from the intersection of three surfaces (implied from three coordinate
constants in two different coordinate systems). The three (of six) constants easiest to define should be used.
The program will calculate the remaining three coordinate constants. All arguments, except KC1, must be
input. Use the repeat command [*REPEAT] after the MOVE command to define a line of intersection by re-
peating the move operation on all nodes of the line.
Surfaces of constant value are implied by some commands by specifying a single coordinate value. Implied
surfaces are used with various commands [MOVE, KMOVE, NSEL, etc.]. Three surfaces are available with
each of the four coordinate system types. Values or X, Y, or Z may be constant for the Cartesian coordinate
system; values of R, , or Z for the cylindrical system; and values of R, , for the spherical and toroidal systems.
For example, an X value of 3 represents the Y-Z plane (or surface) at X=3. In addition, the parameters for
the cylindrical and spherical coordinate systems may be adjusted [CS, LOCAL] to form elliptical surfaces. For
surfaces in elliptical coordinate systems, a surface of "constant" radius is defined by the radius value at the
X-axis. Surfaces of constant value may be located in local coordinate systems [LOCAL, CLOCAL, CS, or CSKP]
to allow for any orientation.
1025
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MOVE
The intersection calculation is based on an iterative procedure (250 iterations maximum) and a tolerance of
1.0E-4. The approximate location of a node should be sufficient to determine a unique intersection if more
than one intersection point is possible. Tangent "intersections" should be avoided. If an intersection is not
found, the node is placed at the last iteration location.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Nodes>To Intersect
MP, Lab, MAT, C0, C1, C2, C3, C4
Defines a linear material property as a constant or a function of temperature.
PREP7: Materials
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
Product Restrictions
Lab
Valid material property label. Applicable labels are listed under "Material Properties" in the input table
for each element type in the Element Reference. See Linear Material Properties of the Element Reference
for more complete property label definitions:
EX
Elastic moduli (also EY, EZ).
ALPX
Secant coefficients of thermal expansion (also ALPY, ALPZ).
CTEX
Instantaneous coefficients of thermal expansion (also CTEY, CTEZ).
THSX
Thermal strain (also THSY, THSZ).
REFT
Reference temperature. Must be defined as a constant; C1 through C4 are ignored.
PRXY
Major Poisson's ratios (also PRYZ, PRXZ).
NUXY
Minor Poisson's ratios (also NUYZ, NUXZ).
GXY
Shear moduli (also GYZ, GXZ).
DAMP
K matrix multiplier for damping.
Note
If used in an explicit dynamic analysis, the value corresponds to the percentage of
damping in the high frequency domain. For example, 0.1 roughly corresponds to 10%
damping in the high frequency domain.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1026
MP
DMPR
Constant material damping coefficient.
MU
Coefficient of friction.
DENS
Mass density.
C
Specific heat.
ENTH
Enthalpy.
KXX
Thermal conductivities (also KYY, KZZ).
HF
Convection or film coefficient.
EMIS
Emissivity.
QRATE
Heat generation rate.
VISC
Viscosity.
SONC
Sonic velocity.
RSVX
Electrical resistivities (also RSVY, RSVZ).
PERX
Electric relative permittivities (also PERY, PERZ).
Note
If you enter permittivity values less than 1 for SOLID5, PLANE13, or SOLID98, the program
interprets the values as absolute permittivity. Values input for PLANE223, SOLID226, or
SOLID227 are always interpreted as relative permittivity.
MURX
Magnetic relative permeabilities (also MURY, MURZ).
MGXX
Magnetic coercive forces (also MGYY, MGZZ).
LSSM
Magnetic loss tangent.
LSST
Electric loss tangent.
SBKX
Seebeck coefficients (also SBKY, SBKZ).
1027
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MP
MAT
Material reference number to be associated with the elements (defaults to the current MAT setting
[MAT]).
C0
Material property value, or if a property-versus-temperature polynomial is being defined, the constant
term in the polynomial. C0 can also be a table name (%tabname%); if C0 is a table name, C1 through
C4 are ignored.
C1, C2, C3, C4
Coefficients of the linear, quadratic, cubic, and quartic terms, respectively, in the property-versus-temper-
ature polynomial. Leave blank (or set to zero) for a constant material property.
Notes
MP defines a linear material property as a constant or in terms of a fourth order polynomial as a function
of temperature. (See the TB command for nonlinear material property input.) Linear material properties
typically require a single substep for solution, whereas nonlinear material properties require multiple substeps;
see Linear Material Properties in the Element Reference for details.
If the constants C1 - C4 are input, the polynomial
Property = C0 + C1(T) + C2(T)
2
+ C3(T)
3
+ C4(T)
4
is evaluated at discrete temperature points with linear interpolation between points (that is, a piecewise
linear representation) and a constant-valued extrapolation beyond the extreme points. First-order properties
use two discrete points (9999). The MPTEMP or MPTGEN commands must be used for second and higher
order properties to define appropriate temperature steps. To ensure that the number of temperatures defined
via the MPTEMP and MPTGEN commands is minimally sufficient for a reasonable representation of the
curve, ANSYS generates an error message if the number is less than N, and a warning message if the number
is less than 2N. The value N represents the highest coefficient used; for example, if C3 is nonzero and C4 is
zero, a cubic curve is being used which is defined using 4 coefficients so that N = 4.
A polynomial input is not valid in an explicit dynamic analysis or for Lab = DAMP. C1, C2, C3, and C4 are
ignored.
The use of tabular material properties (C0 = %tabname%) is available only for FLUID116. Fluid properties
can be evaluated as a function of pressure, temperature, velocity, time, and location (independent variables).
Use the *DIM command to create the table of property values as a function of the independent variables.
Then refer to this table via the MP command for the property. If using temperature or pressure independent
variables, you need to activate the appropriate pressure or temperature degrees of freedom on the element.
Tabular material properties are calculated before the first iteration (that is, using initial values [IC]). For more
information about using table-type array parameters, see the discussion on applying tabular boundary
conditions in the Basic Analysis Guide.
When defining a reference temperature (MP,REFT), you can convert temperature-dependent secant coefficients
of thermal expansion (SCTE) data from the definition temperature to the reference temperature. To do so,
issue the MPAMOD command.
This command is also valid in SOLUTION.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1028
MP
Product Restrictions
Available Products Command Option
Lab
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS ALPX
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS ALPY
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS ALPZ
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS C
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS CTEX
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS CTEY
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS CTEZ
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS DAMP
MP ME ST PR PRN DS DSS <> <> EH DY PP <> EME MFS DENS
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS DMPR
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS EMIS
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS ENTH
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS EX
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS EY
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS EZ
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS GXY
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS GXZ
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS GYZ
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS HF
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS KXX
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS KYY
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS KZZ
MP <> <> <> <> <> <> <> EM EH <> PP <> EME <> LSSM
MP ME <> <> <> <> <> <> EM EH <> PP <> EME MFS LSST
MP ME <> <> <> <> <> <> EM <> DY PP <> EME MFS MGXX
MP ME <> <> <> <> <> <> EM <> DY PP <> EME MFS MGYY
MP ME <> <> <> <> <> <> EM <> DY PP <> EME MFS MGZZ
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS MU
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS MURX
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS MURY
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS MURZ
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS NUXY
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS NUXZ
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS NUYZ
MP ME <> <> <> <> <> FL EM EH DY PP <> EME MFS PERX
1029
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MP
MP ME <> <> <> <> <> FL EM EH DY PP <> EME MFS PERY
MP ME <> <> <> <> <> FL EM EH DY PP <> EME MFS PERZ
MP ME ST PR PRN DS DSS FL <> <> DY PP <> EME MFS PRXY
MP ME ST PR PRN DS DSS FL <> <> DY PP <> EME MFS PRXZ
MP ME ST PR PRN DS DSS FL <> <> DY PP <> EME MFS PRYZ
MP ME ST PR PRN DS DSS <> EM <> DY PP <> EME MFS QRAT
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS REFT
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS RSVX
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS RSVY
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS RSVZ
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS SBKX
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS SBKY
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS SBKZ
MP ME <> <> <> <> <> FL <> <> DY PP <> EME MFS SONC
MP ME ST <> <> <> <> FL <> <> DY PP <> EME MFS THSX
MP ME ST <> <> <> <> FL <> <> DY PP <> EME MFS THSY
MP ME ST <> <> <> <> FL <> <> DY PP <> EME MFS THSZ
MP ME ST <> <> <> <> FL <> <> DY PP <> EME MFS VISC
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
MPAMOD, MAT, DEFTEMP
Modifies temperature-dependent secant coefficients of thermal expansion.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
MAT
Material number for which the secant coefficients of thermal expansion (SCTE's) are to be modified.
Defaults to 1.
DEFTEMP
Definition temperature at which the existing SCTE-versus-temperature tables were defined. Defaults to
zero.
Notes
This command converts temperature-dependent SCTE data (properties ALPX, ALPY, ALPZ) from the definition
temperature (DEFTEMP) to the reference temperature defined by MP,REFT or TREF. If both the MP,REFT
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1030
MPAMOD
and TREF commands have been issued, the reference temperature defined by the MP,REFT command will
be used.
This command does not apply to the instantaneous coefficients of thermal expansion (properties CTEX, CTEY,
CTEZ) or to the thermal strains (properties THSX, THSY, THSZ).
See Linear Material Properties of the Element Reference and the Theory Reference for the Mechanical APDL and
Mechanical Applications for more details.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Convert ALPx
Main Menu>Preprocessor>Material Props>Convert ALPx
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Convert ALPx
MPCHG, MAT, ELEM
Changes the material number attribute of an element.
PREP7: Materials
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT
Assign this material number to the element. Material numbers are defined with the material property
commands [MP].
ELEM
Element for material change. If ALL, change materials for all selected elements [ESEL].
Notes
Changes the material number of the specified element. Between load steps in SOLUTION, material properties
cannot be changed from linear to nonlinear, or from one nonlinear option to another.
If you change from one MKIN model to another MKIN model, the different MKIN models need to have the
same number of data points. This requirement also applies if you change from one KINH model to another
KINH model, or from one CHABOCHE model to another CHABOCHE model.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Change Mat Num
Main Menu>Preprocessor>Material Props>Change Mat Num
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Change Mat Num
1031
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPCHG
MPCOPY, --, MATF, MATT
Copies linear material model data from one material reference number to another.
PREP7: Materials
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
--
Unused field
MATF
Material reference number from where material property data will be copied.
MATT
Material reference number to where material property data will be copied.
Notes
The MPCOPY command copies linear material properties only, which are all properties defined through the
MP command. If you copy a model that includes both linear and yield behavior constants (for example, a
BKIN model), the MPCOPY and TBCOPY, ALL commands are used together to copy the entire model. All
input data associated with the model is copied, that is, all data defined through the MP and TB commands.
Also, if you copy a material model using the Material Model Interface (Edit> Copy), both the commands
MPCOPY and TBCOPY, ALL are issued, regardless of whether the model includes linear constants only, or
if it includes a combination of linear and yield behavior constants.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
MPDATA, Lab, MAT, STLOC, C1, C2, C3, C4, C5, C6
Defines property data to be associated with the temperature table.
PREP7: Materials
MP ME <> <> <> <> <> <> EM EH <> PP <> EME MFS
Product Restrictions
Lab
Valid property label. Applicable labels are listed under "Material Properties" in the input table for each
element type in the Element Reference. See Linear Material Properties of the Element Reference for more
complete property label definitions:
EX
Elastic moduli (also EY, EZ).
ALPX
Secant coefficients of thermal expansion (also ALPY, ALPZ). (See also MPAMOD command for adjust-
ment to reference temperature).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1032
MPCOPY
CTEX
Instantaneous coefficients of thermal expansion (also CTEY, CTEZ).
THSX
Thermal strain (also THSY, THSZ).
REFT
Reference temperature (may not be temperature dependent).
PRXY
Major Poisson's ratios (also PRYZ, PRXZ).
NUXY
Minor Poisson's ratios (also NUYZ, NUXZ).
GXY
Shear moduli (also GYZ, GXZ).
DAMP
K matrix multiplier for damping.
DMPR
Constant material damping coefficient.
MU
Coefficient of friction.
DENS
Mass density.
C
Specific heat.
ENTH
Enthalpy.
KXX
Thermal conductivities (also KYY, KZZ).
HF
Convection or film coefficient.
EMIS
Emissivity.
QRATE
Heat generation rate.
VISC
Viscosity.
SONC
Sonic velocity.
RSVX
Electrical resistivities (also RSVY, RSVZ).
PERX
Electric relative permittivities (also PERY, PERZ).
MURX
Magnetic relative permeabilities (also MURY, MURZ).
1033
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPDATA
MGXX
Magnetic coercive forces (also MGYY, MGZZ).
LSST
Dielectric loss tangent.
SBKX
Seebeck coefficients (also SBKY, SBKZ).
The MPDATA command may also be used to enter temperature dependent properties for fluids in a
CFD analysis with FLOTRAN via FLUID141 and FLUID142. Valid MPDATA labels for a CFD analysis in a
non-solid region are:
DENS
Density of fluid. This is the same as the label used to specify mass density with the FLDATA command.
C
Specific heat of fluid. This is equivalent to the SPHT label used to specify conductivity with the
FLDATA command.
KXXX
Thermal conductivity of fluid.
VISC
Viscosity of fluid. This is the same as the label used to specify kinematic velocity with the FLDATA
command.
MAT
Material reference number to be associated with the elements (defaults to 1 if you specify zero or no
material number).
STLOC
Starting location in table for generating data. For example, if STLOC = 1, data input in the C1 field is
the first constant in the table. If STLOC = 7, data input in the C1 field is the seventh constant in the
table, etc. Defaults to the last location filled + 1.
C1, C2, C3, . . . , C6
Property data values assigned to six locations starting with STLOC. If a value is already in this location,
it is redefined. A blank (or zero) value for C1 resets the previous value in STLOC to zero. A value of zero
can only be assigned by C1. Blank (or zero) values for C2 to C6 leave the corresponding previous values
unchanged.
Notes
Defines a table of property data to be associated with the temperature table. Repeat MPDATA command
for additional values (100 maximum). Temperatures must be defined first [MPTEMP]. Also stores assembled
property function table (temperature and data) in virtual space.
This command is also valid in SOLUTION.
Product Restrictions
In ANSYS without Emag enabled, the MUR_ and MG__ properties are not allowed. In ANSYS Professional,
all structural and thermal properties are allowed except DAMP and MU. In ANSYS Emag, only the RSV_, PER_,
MUR_, and MG__ properties are allowed. Only products that include ANSYS Emag can use the LSST property.
The SBK_ property is only available in ANSYS Multiphysics, ANSYS PrepPost, and ANSYS ED.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1034
MPDATA
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
MPDELE, Lab, MAT1, MAT2, INC,--, LCHK
Deletes linear material properties.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Material property label (see MP command for valid labels). If ALL, delete properties for all applicable labels.
MAT1, MAT2, INC
Delete materials from MAT1 to MAT2 (defaults to MAT1) in steps of INC (defaults to 1). If MAT1 = ALL,
MAT2 and INC are ignored and the properties for all materials are deleted.
--
Unused field.
LCHK
Specifies the level of element-associativity checking:
NOCHECK
No element-associativity check occurs. This option is the default.
WARN
When a section, material, or real constant is associated with an element, ANSYS issues a message
warning that the necessary entity has been deleted.
CHECK
The command terminates, and no section, material, or real constant is deleted if it is associated with
an element.
Notes
This command is also valid in SOLUTION.
The LCHK argument is valid only when Lab = ALL.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
1035
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPDELE
MPDRES, LabF, MATF, LabT, MATT
Reassembles existing material data with the temperature table.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabF
Material property label associated with MATF.
MATF
Material reference number of property to restore from virtual space.
LabT
Material property label associated with MATT (defaults to label associated with MATF).
MATT
Material reference number assigned to generated property (defaults to MATF).
Notes
Restores into the database (from virtual space) a data table previously defined [MP] for a particular property,
assembles data with current database temperature table, and stores back in virtual space as a new property.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Other>Change Mat Props>Modify Temps
Main Menu>Preprocessor>Material Props>Modify Temps
Main Menu>Solution>Other>Change Mat Props>Modify Temps
/MPLIB, R-W_opt, PATH
Sets the default material library read and write paths.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
R-W_opt
Determines what path is being set. Possible values are:
READ
Set the read path.
WRITE
Set the write path.
STAT
Report what read and write paths are currently in use.
PATH
The directory path to be used for material library files.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1036
MPDRES
Notes
The /MPLIB command sets two path strings used in conjunction with the material library feature and the
MPREAD and MPWRITE commands.
For MPREAD, when you use the LIB option and no directory path is given in the file name, the command
searches for the file in these locations: the current working directory, the user's home directory, the user-
specified material library directory (as defined by the /MPLIB,READ,PATH command), and /ansys_dir/mat-
lib.
For MPWRITE, when you use the LIB option and the directory portion of the specification for the material
library file is blank, the command writes the material library file to the directory specified by the
/MPLIB,WRITE,PATH command (if that path has been set). If the path has not been set, the default is to
write the file to the current working directory.
The Material Library files supplied with the distribution disks are meant for demonstration purposes only.
These files are not intended for use in customer applications.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Lib
Path Status
Main Menu>Preprocessor>Material Props>Material Library>Lib Path Status
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Lib Path Status
MPLIST, MAT1, MAT2, INC, Lab, TEVL
Lists linear material properties.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT1, MAT2, INC
List materials from MAT1 to MAT2 (defaults to MAT1) in steps of INC (defaults to 1). If MAT1= ALL (default),
MAT2 and INC are ignored and properties for all material numbers are listed.
Lab
Material property label (see the MP command for labels). If ALL (or blank), list properties for all labels.
If EVLT, list properties for all labels evaluated at TEVL.
TEVL
Evaluation temperature for Lab = EVLT listing (defaults to BFUNIF).
Notes
For Lab = EVLT, when the property is from tables, the MPPLOT command will not be valid because the
property could be a function of more than temperature.
This command is valid in any processor.
1037
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPLIST
Menu Paths
Utility Menu>List>Properties>All Materials
Utility Menu>List>Properties>All Matls, All Temps
Utility Menu>List>Properties>All Matls, Specified Temp
Utility Menu>List>Properties>Specified Matl, All Temps
MPPLOT, Lab, MAT, TMIN, TMAX, PMIN, PMAX
Plots linear material properties as a function of temperature.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Linear material property label (EX, EY, etc.) [MP].
MAT
Material reference number. Defaults to 1.
TMIN
Minimum abscissa value to be displayed.
TMAX
Maximum abscissa value.
PMIN
Minimum property (ordinate) value to be displayed.
PMAX
Maximum property value.
Notes
When the property is from tables, the MPPLOT command will not be valid because the property could be
a function of more than temperature.
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Materials
MPREAD, Fname, Ext, --, LIB
Reads a file containing material properties.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including directory). If you do not specify the
LIB option, the default directory is the current working directory. If you specify the LIB option, the
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1038
MPPLOT
default is the following search path: the current working directory, the user's home directory, MPLIB_DIR
(as specified by the /MPLIB,READ,PATH command) and /ansys_dir/matlib (as defined by installation).
If you use the default for your directory, you can use all 248 characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
If you omit the default extension is MP. extension is units_MPL, where units is the system of units currently
in use. (See the description of the /UNITS command.) For example, if /UNITS is set to SI, the extension
defaults to SI_MPL.
--
Unused field.
LIB
Reads material library files previously written with the MPWRITE command. (See the description of the
LIB option for the MPWRITE command.) The only allowed value for LIB is LIB.
The LIB field indicates that the specified file was written by MPWRITE using the LIB option, and that the
file is consistent with the material library file format. When the MPREAD command executes, the ANSYS
program reads material properties defined in the specified file into the current ANSYS working database.
The currently selected material, as defined by the MAT command (MAT,MAT), determines the material
number used when reading the material properties. The LIB option for MPREAD and MPWRITE supports
storing and retrieving both linear and nonlinear properties.
Notes
Material properties written to a file without the LIB option do not support nonlinear properties. Also, prop-
erties written to a file without the LIB option are restored in the same material number as originally defined.
To avoid errors, use MPREAD with the LIB option only when reading files written using MPWRITE with the
LIB option.
If you omit the LIB option for MPREAD, this command supports only linear properties.
Material numbers are hardcoded. If you write a material file without specifying the LIB option, then read
that file in using the MPREAD command with the LIB option, the ANSYS program will not write the file to
a new material number. Instead, it will write the file to the "old" material number (the number specified on
the MPWRITE command that created the file.)
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Export
Library
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Import
Library
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Select
Units
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Read from File
Main Menu>Preprocessor>Material Props>Material Library>Export Library
Main Menu>Preprocessor>Material Props>Material Library>Import Library
1039
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPREAD
Main Menu>Preprocessor>Material Props>Material Library>Select Units
Main Menu>Preprocessor>Material Props>Read from File
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Export Library
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Import Library
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Select Units
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Read from File
MPRINT, KEY
Specifies that radiation matrices are to be printed.
AUX12: Radiation Matrix Method
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
KEY
Print key:
0
Do not print matrices.
1
Print matrices.
Command Default
Matrices are not printed.
Notes
Specifies that the element and node radiation matrices are to be printed when the WRITE command is issued.
If KEY = 1, form factor information for each element will also be printed.
Menu Paths
Main Menu>Radiation Opt>Matrix Method>Write Matrix
MPTEMP, STLOC, T1, T2, T3, T4, T5, T6
Defines a temperature table for material properties.
PREP7: Materials
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
STLOC
Starting location in table for entering temperatures. For example, if STLOC = 1, data input in the T1
field applies to the first constant in the table. If STLOC = 7, data input in the T1 field applies to the
seventh constant in the table, etc. Defaults to the last location filled + 1.
T1, T2, T3, . . . , T6
Temperatures assigned to six locations starting with STLOC. If a value is already in this location, it will
be redefined. A blank (or zero) value for T1 resets the previous value in STLOC to zero. A value of zero
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1040
MPRINT
can only be assigned by T1. Blank (or zero) values for T2 to T6 leave the corresponding previous values
unchanged.
Command Default
No temperature table defined (i.e., properties must be defined as a constant or linear function of temperature
with the MP command).
Notes
Defines a temperature table to be associated with the property data table [MPDATA]. These temperatures
are also used for polynomial property evaluation, if defined [MP]. Temperatures must be defined in non-
descending order. Issue MATER $ STAT to list the current temperature table. Repeat MPTEMP command
for additional temperatures (100 maximum). If all arguments are blank, the temperature table is erased.
For clear definition, the temperature range you define with the MPTEMP command should include the entire
range you'll use in subsequently defined materials. To assist the user in this, the first (and only the first) ex-
cursion out of the temperature range defined by the MPTEMP commands is flagged with a warning message.
Similarly, the reference temperature (TREF or MP,reft commands) should also fall in this same temperature
range. If not and MP,alpx was used, a note will be output. If not, and MP,ctex or MP,thsx was used, an error
message will be output.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
MPTGEN, STLOC, NUM, TSTRT, TINC
Adds temperatures to the temperature table by generation.
PREP7: Materials
MP ME ST PR PRN <> <> FL <> <> DY PP <> EME MFS
STLOC
Starting location in table for generating temperatures. Defaults to last location filled + 1.
NUM
Number of temperatures to be generated (1-100).
TSTRT
Temperature assigned to STLOC location.
TINC
Increment previous temperature by TINC and assign to next location until all NUM locations are filled.
Notes
Adds temperatures to the temperature table by generation. May be used in combination (or in place of ) the
MPTEMP command.
1041
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPTGEN
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Other>Change Mat Props>Generate Temp
Main Menu>Preprocessor>Material Props>Generate Temp
Main Menu>Solution>Other>Change Mat Props>Generate Temp
MPTRES, Lab, MAT
Restores a temperature table previously defined.
PREP7: Materials
MP ME ST PR PRN <> <> FL <> <> DY PP <> EME MFS
Lab
Material property label [MP].
MAT
Material reference number.
Notes
Restores into the database (from virtual space) a temperature table previously defined [MP] for a particular
property. The existing temperature table in the database is erased before this operation.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Other>Change Mat Props>Restore Temps
Main Menu>Preprocessor>Material Props>Restore Temps
Main Menu>Solution>Other>Change Mat Props>Restore Temps
MPWRITE, Fname, Ext, --, LIB, MAT
Writes linear material properties in the database to a file (if the LIB option is not specified) or writes both
linear and nonlinear material properties (if LIB is specified) from the database to a file.
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including directory). If you do not specify the
LIB option, the default directory is the current working directory. If you specify LIB and you have
specified a material library directory (via the /MPLIB command), that directory is the default. Otherwise,
the default is the current working directory. If you use the default for your directory, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1042
MPTRES
Ext
Filename extension (8 character maximum).
If you omit the LIB option, the default extension is MP. If you specify the LIB option, the default exten-
sion is units_MPL, where units is the system of units currently in use. (See the description of the /UNITS
command.) For example, if /UNITS is set to BIN, the extension defaults to BIN_MPL.
--
Unused field.
LIB
The only value allowed for this field is the string "LIB."
The LIB option indicates that you wish to have properties associated with the material (MAT) written to
the specified material library file using the material library file format. The material library file format is
ASCII-text-based ANSYS command input. Certain commands associated with this format have been
modified to interpret the string "_MATL" to mean the currently selected material. This feature makes the
material library file independent of the material number in effect when the file was written; this enables
you to restore the properties into the ANSYS database using the material number of your choice. The
LIB option also enables you to save both linear and nonlinear properties. If you omit the LIB option, you
can save linear properties only.
MAT
Specifies the material to be written to the named material library file. There is no default; you must either
specify a material or omit the MAT argument. Even if you specify a MAT value, the ANSYS program ignores
it if the LIB argument is not specified.
Notes
Writes linear material properties currently in the database to a file. The file is rewound before and after
writing.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Export
Library
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Import
Library
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Library>Select
Units
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Write to File
Main Menu>Preprocessor>Material Props>Material Library>Export Library
Main Menu>Preprocessor>Material Props>Material Library>Import Library
Main Menu>Preprocessor>Material Props>Material Library>Select Units
Main Menu>Preprocessor>Material Props>Write to File
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Export Library
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Import Library
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Library>Select Units
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Write to File
1043
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MPWRITE
/MREP, NAME, ARG1, ARG2, ARG3, . . . , ARG4, ARG5, ARG6, ARG7, ARG8, ARG9, ARG10, ARG11, ARG12, ARG13,
ARG14, ARG15, ARG16, ARG17, ARG18
Enables you to reissue the graphics command macro "name" during a replot or zoom operation.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NAME
The name identifying the macro file or macro block on a macro library file. The name can contain up to
eight characters maximum and must begin with a letter.
ARG1, ARG2, ARG3, . . . , ARG18
Values to be passed into the file or block.
Notes
This command reissues the graphics command macro "name" during a replot operation [/REPLOT] or a
zoom [/ZOOM] operation. The ANSYS program passes the command macro arguments to the replot and
zoom feature for use by the graphics macro. You should place the s-MREP command at the end of the
graphics command macro, following the last graphics command within the macro, to enable the replot or
zoom feature.
Menu Paths
This command cannot be accessed from a menu.
MSADV, SPNUM, MTHA
Specifies the approach to discretize the advection term in a species transport equation.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
MTHA
Choice of approach to discretize the advection term:
MSU
Monotone Streamline Upwind (MSU) approach.
SUPG
Streamline Upwind / Petrov-Galerkin (SUPG) approach (default).
COLG
Collocated Galerkin (COLG) approach.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis. See Advection Term
in the Theory Reference for the Mechanical APDL and Mechanical Applications for more information on the
advection term.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1044
/MREP
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSAVE, Key
Sets the solver memory saving option. This option only applies to the PCG solver.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Key
Activation key:
0 or OFF
Use global assembly for the stiffness matrix (and mass matrix, when using MODOPT,LANPCG) of the
entire model.
1 or ON
Use an element-by-element approach when possible to save memory during the solution. In this
case, the global stiffness (and mass) matrix is not assembled; element stiffness (mass) is regenerated
during PCG or MODOPT,LANPCG iterations.
Notes
MSAVE,ON only applies to and is the default for parts of the model using the following element types with
linear material properties that meet the conditions listed below.
SOLID92
SOLID95
SOLID186 (Structural Solid only)
SOLID187
The following conditions must also be true:
The PCG solver has been specified.
The analysis is a static analysis. Topological optimization is not supported.
Small strains are assumed (NLGEOM,OFF).
No prestress effects (PSTRES) are included.
No stress stiffening effects (SSTIF) are included when small strains are assumed (NLGEOM,OFF).
All nodes on the supported element types must be defined (i.e., the midside nodes cannot be removed
using the EMID command).
For elements with thermally dependent material properties, MSAVE,ON applies only to elements with
uniform temperatures prescribed.
The default element coordinate system must be used.
1045
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSAVE
If you manually force MSAVE,ON by including it in the input file, the model can include the following addi-
tional conditions:
The analysis can be a modal analysis using the PCG Lanczos method (MODOPT,LANPCG).
Large deflection effects (NLGEOM,ON) are included.
SOLID45 (brick shapes and KEYOPT(2) = 0 only) elements can be included.
SOLID185 (brick shapes and KEYOPT(2) = 3 only) elements can be included.
All other element types or other parts of the model that don't meet the above criteria will be solved using
global assembly (MSAVE,OFF). This command can result in memory savings of up to 70% over the global
assembly approach for the part of the model that meets the criteria. Depending on the processor speed
and the computer manufacturer, the solution time may increase or decrease when this feature is used.
This memory saving feature runs in parallel when multiple processors are used with the /CONFIG command
or with Distributed ANSYS. The gain in performance with using multiple processors with this feature turned
on should be similar to the default case when this feature is turned off. Performance also improves when
using the 2 x 2 x 2 reduced integration option for SOLID95 elements or the uniform reduced integration
option for SOLID186 elements.
This command does not support the layered option of the SOLID185 and SOLID186 elements.
When using MSAVE,ON with the PCGOPT command, note the following restrictions: .
For static and modal analyses, MSAVE,ON is not valid when using a Lev_Diff value of 5 on the
PCGOPT command; Lev_Diff will automatically be reset to 2.
For modal analyses, MSAVE,ON is not valid with the StrmCk option of the PCGOPT command; Strmck
will be set to OFF.
When using MSAVE,ON for modal analyses, no .FULL file will be created. The .FULL file may be necessary
for subsequent analyses (e.g., harmonic, transient mode superposition, or spectrum analyses). To generate
the .FULL file, rerun the modal analysis using the WRFULL command.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
MSCAP, SPNUM, Capkey, UPPER, LOWER
Activates and controls mass fraction capping for a species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
Capkey
Key to activate mass fraction capping:
OFF
Capping not enforced (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1046
MSCAP
ON
Capping will be enforced.
UPPER, LOWER
Upper and lower bounds on mass fraction if capping is activated. Default to 1.0 and 0.0 respectively.
Command Default
No mass fraction capping.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSDATA, ALGEB, UGAS
Defines multiple species data applicable to all species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
ALGEB
The algebraic species number whose mass fraction is calculated by subtracting the sum of the mass
fractions of all other species from 1.0. This ensures that the sum of the mass fractions of all the species
is 1.0. Defaults to 2.
UGAS
The universal gas constant. Defaults to 8314.3 (SI units).
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
1047
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSDATA
*MSG, Lab, VAL1, VAL2, VAL3, VAL4, VAL5, VAL6, VAL7, VAL8
Writes an output message via the ANSYS message subroutine.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Lab
Label for output and termination control:
INFO --
Writes the message with no heading (default).
NOTE --
Writes the message with a "NOTE" heading.
WARN --
Writes the message with a "WARNING" heading. Also writes the message to the errors file, Job-
name.ERR.
ERROR --
Writes the message with a "ERROR" heading and causes run termination (if batch) at earliest "clean
exit" point. Also writes the message to the errors file, Jobname.ERR.
FATAL --
Writes the message with a "FATAL ERROR" heading and causes run termination immediately. Also
writes the message to the errors file, Jobname.ERR.
UI --
Writes the message with a "NOTE" heading and displays it in the message dialog box. This option is
most useful in GUI mode.
VAL1, VAL2, VAL3, . . . , VAL8
Numeric or alphanumeric character values to be included in message. Values may be the results of
parameter evaluations. All numeric values are assumed to be double precision. The FORTRAN nearest
integer (NINT) function is used to form integers for the %I specifier.
Notes
Allows writing an output message via the ANSYS message subroutine. Also allows run termination control.
This command is used only when contained in a prepared file read into the ANSYS program (i.e., *USE,/INPUT,
etc.). A message format must immediately follow the *MSG command (on a separate line, without parentheses,
as described below).
The message format may be up to 80 characters long, consisting of text strings and predefined "data
descriptors" between the strings where numeric or alphanumeric character data are to be inserted. The
normal descriptors are %I for integer data, %G for double precision data, %C for alphanumeric character
data, and %/ for a line break. The corresponding FORTRAN data descriptors are I9, 1PG16.9 and A8, respectively.
Each descriptor must be preceded by a blank. There must be one data descriptor for each specified value
(8 maximum) in the order of the specified values.
Enhanced descriptions may also be used:
w is field width %w.pE
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1048
*MSG
p is precision %w.pG
%w.pF
a single percent sign %%
character string %wC; %wS
left justify string %-wC; %-wS
w blank characters %wX
integer format %wI
pad integer with leading zeros rather than blanks %0wI
w is field width; p is number of characters filled %0w.pI
Do not begin *MSG format lines with *IF, *ELSE , *ELSEIF , or *ENDIF . If the last nonblank character of the
message format is an ampersand (&), a second line will also be read as a continuation of the format. Up to
nine continuations (ten total lines) may be read. If normal descriptions are used, then consecutive blanks
are condensed into one blank upon output, and a period is appended. Up to ten lines of output of 72
characters each may be produced (using the %/ descriptor). Two examples follow.
Here is an example of the *MSG command and a format to print a message with two integer values and
one real value:
*MSG, INFO, 'Inner',25,1.2,148
Radius ( %C) = %I, Thick = %G, Length = %I
The output line is:
Radius (Inner) = 25, Thick = 1.2, Length = 148.
Here is an example illustrating multiline displays in GUI message windows:
*MSG,UI,Vcoilrms,THTAv,Icoilrms,THTAi,Papprnt,Pelec,PF,indctnc
Coil RMS voltage, RMS current, apparent pwr, actual pwr, pwr factor: %/&
Vcoil = %G V (electrical angle = %G DEG) %/&
Icoil = %G A (electrical angle = %G DEG) %/&
APPARENT POWER = %G W %/&
ACTUAL POWER = %G W %/&
Power factor: %G %/&
Inductance = %G %/&
VALUES ARE FOR ENTIRE COIL (NOT JUST THE MODELED SECTOR)
Note
The /UIS,MSGPOP command controls which messages are displayed in the message dialog box
when the GUI is active. All messages produced by the *MSG command are subject to the /UIS
specification, with one exception, If Lab = UI, the message will be displayed in the dialog box
regardless of the /UIS specification.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
1049
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*MSG
MSHAPE, KEY, Dimension
For elements that support multiple shapes, specifies the element shape to be used for meshing.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Key indicating the element shape to be used:
0
Mesh with quadrilateral-shaped elements when Dimension = 2-D mesh with hexahedral-shaped
elements when Dimension = 3-D.
1
Mesh with triangle-shaped elements when Dimension = 2-D mesh with tetrahedral-shaped elements
when Dimension = 3-D.
Dimension
Specifies the dimension of the model to be meshed:
2D
2-D model (area mesh).
3D
3-D model (volume mesh).
Command Default
Since specification of element shape [MSHAPE] and meshing type [MSHKEY] are so closely related, the
element shape that ANSYS meshes with depends on the combination of the values that are set for the two
commands. The table below explains what happens when you fail to specify values for these settings.
How it affects the mesh... Your action...
ANSYS uses quadrilateral-shaped or hexahed-
ral-shaped elements to mesh the model, de-
You issue the MSHAPE command with no
arguments.
pending on whether you are meshing an area
or a volume.
ANSYS uses the default shape of the element
to mesh the model. It uses the type of mesh-
ing that you specified.
You do not specify an element shape, but you
do specify the type of meshing to be used
[MSHKEY].
ANSYS uses the default shape of the element
to mesh the model. It uses whichever type of
meshing is the default for that shape.
You specify neither an element shape nor the
type of meshing to be used.
Notes
If no value is specified for Dimension the value of KEY determines the element shape that will be used
for both 2-D and 3-D meshing. In other words, if you specify MSHAPE,0, quadrilateral-shaped and hexahedral-
shaped elements will be used. If you specify MSHAPE,1, triangle-shaped and tetrahedral-shaped elements
will be used.
The MSHAPE, MSHKEY, and MSHMID commands replace the functionality that was provided by the ESHAPE
command in ANSYS 5.3 and earlier releases.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1050
MSHAPE
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Volumes>Mapped>4 to 6 sided
Main Menu>Preprocessor>Meshing>Mesher Opts
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Global Meshing
Options
MSHCOPY, KEYLA, LAPTRN, LACOPY, KCN, DX, DY, DZ, TOL, LOW, HIGH
Simplifies the generation of meshes that have matching node element patterns on two different line
groups (in 2-D) or area groups (3-D).
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEYLA
Copy line mesh (default) if LINE, 0 or 1. Copy area mesh if AREA, or 2.
LAPTRN
Meshed line/area to be copied, or a component name containing a list. If LAPTRN = P, graphical picking
is enabled (valid only in the GUI).
LACOPY
Unmeshed line/area to get copied mesh, or a component name containing a list. If LACOPY = P,
graphical picking is enabled (valid only in the GUI).
KCN
In coordinate system KCN, LAPTRN + DX DY DZ = LACOPY.
DX, DY, DZ
Node location increments in the active coordinate system (DR, D, DZ for cylindrical, DR, D, D for
spherical or toroidal).
TOL
Tolerance. Defaults to 1.e--4.
LOW
Name of low node component to be defined (optional).
HIGH
Name of high node component to be defined (optional).
Notes
Matching meshes are used for rotational (cyclic) symmetry, or for contact analysis using coupling or node-
to-node gap elements. See Using CPCYC and MSHCOPY Commands in the Modeling and Meshing Guide for
more information.
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Area Mesh
Main Menu>Preprocessor>Modeling>Copy>Line Mesh
1051
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSHCOPY
MSHKEY, KEY
Specifies whether free meshing or mapped meshing should be used to mesh a model.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Key indicating the type of meshing to be used:
0
Use free meshing (the default).
1
Use mapped meshing.
2
Use mapped meshing if possible; otherwise, use free meshing. If you specify MSHKEY,2, SmartSizing
will be inactive even while free meshing non-map-meshable areas.
Command Default
As stated above, free meshing is the default. However, since the MSHKEY and MSHAPE settings are closely
related, you should refer to the table in the MSHAPE command description for more information about
defaults.
Notes
The MSHKEY, MSHAPE, and MSHMID commands replace the functionality that was provided by the ESHAPE
command in ANSYS 5.3 and earlier releases.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Areas>Mapped>3 or 4 sided
Main Menu>Preprocessor>Meshing>Mesh>Areas>Target Surf
Main Menu>Preprocessor>Meshing>Mesh>Volumes>Mapped>4 to 6 sided
Main Menu>Preprocessor>Meshing>Mesher Opts
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Global Meshing
Options
MSHMID, KEY
Specifies placement of midside nodes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Key indicating placement of midside nodes:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1052
MSHKEY
0
Midside nodes (if any) of elements on a region boundary follow the curvature of the boundary line
or area (the default).
1
Place midside nodes of all elements so that element edges are straight. Allows coarse mesh along
curves.
2
Do not create midside nodes (elements will have removed midside nodes).
Notes
The MSHMID, MSHAPE, and MSHKEY commands replace the functionality that was provided by the ESHAPE
command in ANSYS 5.3 and earlier releases.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesher Opts
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Global Meshing
Options
MSHPATTERN, KEY
Specifies pattern to be used for mapped triangle meshing.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Key indicating triangle pattern to be used (the figures below illustrate the pattern that will be used for
each value of KEY):
0
Let ANSYS choose the pattern (the default). ANSYS maximizes the minimum angle of the triangular-
shaped elements that are created.
1
Unidirectional split at node I.
2
Unidirectional split at node J.
1053
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSHPATTERN
KEY = 0
KEY = 1 KEY = 2
Notes
"Mapped triangle meshing" refers to the ANSYS program's ability to take a map-meshable area and mesh it
with triangular elements, based on the value of MSHPATTERN,KEY. This type of meshing is particularly
useful for analyses that involve the meshing of rigid contact elements.
The MSHPATTERN command is valid only when you have specified that ANSYS use triangle-shaped elements
[MSHAPE,1,2D] (or you are meshing with an element that supports only triangles), and you have also specified
mapped meshing [MSHKEY,1] to mesh an area.
For details about mapped meshing with triangles, see the Modeling and Meshing Guide.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesher Opts
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Global Meshing
Options
MSMASS, SPNUM, Value
Specifies the mass type for a transient species analysis.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified. If ALL, all numbers are used.
Value
Mass type for transient species analysis:
LUMP
Lumped mass matrix (default).
CONS
Consistent mass matrix.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1054
MSMASS
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSMETH, SPNUM, KEY
Specifies the method of solution of the species transport equations.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
KEY
Key defining the method of solution for the specified species number:
0
No solution of equations for species SPNUM
1
Tri-Diagonal Matrix Algorithm (TDMA).
2
Conjugate residual method.
3
Preconditioned Conjugate Residual method.
4
Preconditioned Generalized Minimum Residual (PGMR) solution method (default).
5
Sparse Direct method.
6
Preconditioned BiCGStab method (PBCGM).
Command Default
The Preconditioned Generalized Minimum Residual (PGMR) method is used for all species.
Notes
The Tri-Diagonal Matrix Algorithm (TDMA) method is a special version of the standard Gauss-Seidel iterative
method for the solution of sets of algebraic equations. The number of iterations (sweeps) to be performed
is specified with the MSSOLU command. No convergence criterion is required for the TDMA method.
Three methods are semi-direct solution methods based on search directions. They are conjugate direction
iterative techniques which develop a solution as a linear combination of search directions. The Conjugate
Residual method requires the least memory, but stalls when solving ill-conditioned problems. The Precondi-
tioned Conjugate Residual method requires much more memory but performs better for ill-conditioned
matrix problems. The PGMR method is memory-intensive; by necessity, it incorporates a tight convergence
criterion. The Preconditioned BiCGStab method (PBCGM) requires less memory than the PGMR method and
1055
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSMETH
sometimes performs better than the PGMR method. The number of search vectors and the convergence
criterion are specified with the MSSOLU command.
The Sparse Direct method is based on Gaussian elimination to factorize the matrix. This method is memory
intensive and creates temporary files on the hard disk. It is robust and can be used for symmetric as well as
non-symmetric equation systems.
See the Fluids Analysis Guide for additional information on the FLOTRAN Solvers.
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSMIR, SPNUM, Value
Sets modified inertial relaxation factors for multiple species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
Value
Modified inertial relaxation factor. Value defaults to 0 (modified inertial relaxation off ).
Notes
Value must be a positive real number. A Value between 0.1 and 1.0 is recommended. A larger Value
provides a more robust scheme, but it may yield a slower convergence.
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1056
MSMIR
MSNOMF, SPNUM, FRACTION
Specifies the initial value of nominal mass fraction for a species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
FRACTION
The initial mass fraction of the entire problem domain for this species. Defaults to 0.0. The sum of the
mass fractions for all species should equal 1.0.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis and is required if the
CMIX option has been activated for a property [FLDATA7,PROT,Label,CMIX].
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSPROP, SPNUM, Label, Type, NOMINAL, COF1, COF2, COF3
Defines the fluid properties of a species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
Label
Label identifying the property being defined:
DENS
Density.
VISC
Viscosity.
COND
Thermal conductivity.
MDIF
Mass diffusion coefficient.
SPHT
Specific heat.
Type
Type of property:
1057
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSPROP
CONSTANT
Constant property (default). The property does not vary with temperature.
LIQUID
Liquid property. Density varies according to a second order polynomial relationship, and all other
properties follow Sutherland's law for liquids.
GAS
Gas property. Density varies according to the ideal gas law, and all other properties follow Sutherland's
law for gases.
NOMINAL
Nominal value of the property being defined. For CONSTANT fluid types, the property remains at this
value and does not vary. For GAS and LIQUID property types, this is the value of the property correspond-
ing to the temperature defined by COF1.
COF1
Temperature corresponding to the NOMINAL value of the property (for GAS and LIQUID property types
only; see Notes section). Not required for Label = SPHT.
COF2, COF3
Second and third coefficients for temperature variation of the property. Not required for Label = SPHT.
Notes
If the property type is CONSTANT, the equation used is as follows:
Lab = NOMINAL
If the property type is LIQUID, a second order polynomial relationship is used for density, and Sutherland's
law for liquids is used for the other properties:
DENS = NOMINAL + COF2*(T-COF1) + COF3*(T-COF1)
2
Property = NOMINAL * EXP[COF2*(1/T-1/COF1) + COF3*(1/T-1COF1)
2
]
where T is the temperature of the node where the property is being calculated.
If the property type is GAS, the ideal gas law is used for density, and Sutherland's law for gases is used for
other properties:
DENS = NOMINAL * (P/COF2 / (T/COF1)
Property = NOMINAL * (T/COF1)
1.5
* (COF1 + COF2)/(T + COF2)]
where P and T are the pressure and temperature of the node where the property is being calculated.
Specific heat is always a CONSTANT. Also, property types (Type) such as TABLE, USER, POWL, BIN, etc. are
not available for individual species. They are valid only for the bulk fluid.
If a property type is entered incorrectly (e.g. a misspelling), a CONSTANT property type is assumed and the
property is assigned the NOMINAL value.
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1058
MSPROP
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSQUAD, QDIF, QSRC
Specifies the quadrature order for multiple species elements.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
QDIF
Quadrature order for diffusion term integration:
0
One-point quadrature (default).
1
Same as 0, except a distributed value of temperature is used to evaluate temperature-dependent
properties.
2
Two-point quadrature (default for axisymmetric models).
QSRC
Quadrature order for source term integration:
0
One-point quadrature (default).
1
Same as 0, except a distributed value of temperature is used to evaluate temperature-dependent
properties.
2
Two-point quadrature (default for axisymmetric models).
Command Default
As described above.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
1059
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSQUAD
MSRELAX, SPNUM, CONC, MDIF, EMDI, STAB
Specifies relaxation factors for a multiple species transport analysis.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
CONC
Species concentration relaxation factor. Defaults to 0.5.
MDIF
Mass diffusion coefficient relaxation factor. Defaults to 0.5.
EMDI
Effective mass diffusion coefficient relaxation factor (used for turbulent flow). Defaults to 0.5.
STAB
Inertial relaxation factor for solution of the transport equation. Defaults to 1.0 x 10
+20
.
Command Default
As described above for each relaxation factor.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSSOLU, SPNUM, NSWEEP, MAXI, NSRCH, CONV, DELMAX
Specifies solution options for multiple species transport.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
NSWEEP
Number of Tri-Diagonal Matrix Algorithm (TDMA) sweeps. Valid only for the TDMA method [MSMETH].
Defaults to 100.
MAXI
Maximum number of iterations allowed for the semi-direct methods (conjugate residual and precondi-
tioned conjugate residual methods, chosen with the MSMETH command). Defaults to 1000.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1060
MSRELAX
NSRCH
Number of search vectors used for the semi-direct methods. Defaults to 2. If you are using the Precondi-
tioned BiCGStab Method (PBCGM) solver, the number of search directions is 1 to 8, with 2 as the default.
New search vectors are made orthogonal to NSRCH previous vectors in the solution of the unsymmetric
matrix systems.
CONV
Convergence criterion for the semi-direct methods. It represents the factor by which the inner product
of the residual vector is reduced during the solution of the equations at any global iteration. Defaults
to 1.0 x 10
-12
. If the convergence criterion has not been achieved, the algebraic solver issues a warning
message, and the execution of FLOTRAN continues normally.
DELMAX
Minimum normalized rate of change which will permit the semi-direct solution methods to continue.
Used to terminate the semi-direct solvers in the event that stall occurs. Defaults to 1.0 x 10
-9
. If the
methods stall, the solver increments the solution only a very small amount despite the fact that the
correct solution has been not been achieved (or perhaps even approached). The maximum nodal difference
between the solutions, normalized to the value of the variable, is compared to DELMAX, and the solution
is terminated if the value is less than DELMAX. Termination of the algebraic solver due to the small rate
of change is considered a normal function, and no warning message is printed. Execution of FLOTRAN
continues normally.
Command Default
As described above.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSSPEC, SPNUM, Name, MOLWT, SCHMIDT
Specifies the name, molecular weight, and Schmidt number of a species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
Name
Name to be assigned to the species, up to 4 characters long. Defaults to SP01 for species 1, SP02 for
species 2, ..., SP06 for species 6. This name can be used in place of the species number when specifying
mass fraction boundary conditions and in postprocessing.
1061
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSSPEC
Note
The GUI always shows the default names, not the user-defined names.) Name should not be
the same as an existing degree of freedom label.
MOLWT
Molecular weight for the species. Required only for gases (determined by the property type on MSPROP
command). Defaults to 29.0.
SCHMIDT
Schmidt number (diffusion term divisor) for the species. Required only for gases (determined by the
property type on MSPROP command). Defaults to 1.0.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
/MSTART, Label, KEY
Controls the initial GUI components.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Label identifying the GUI component:
ZOOM
Pan, Zoom, Rotate dialog box, off by default.
WORK
Offset Working Plane dialog box, off by default.
WPSET
Working Plane Settings dialog box, off by default.
ABBR
Edit Toolbar/Abbreviations dialog box, off by default.
PARM
Scalar Parameters dialog box, off by default.
SELE
Select Entities dialog box, off by default.
ANNO
Annotation dialog box, off by default.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1062
/MSTART
HARD
Hard Copy dialog box, off by default.
UTIL
Turns on the pre-ANSYS 6.1 (UIDL) GUI, off by default.
KEY
Switch value:
OFF or 0
Component does not appear when GUI is initialized.
ON or 1
Component appears when GUI is initialized.
Command Default
Same as Label defaults.
Notes
Controls which components appear when the Graphical User Interface (GUI) is initially brought up. This
command is valid only before the GUI is brought up [/MENU,ON] and is intended to be used in the start120.ans
file. It only affects how the GUI is initialized; you can always bring up or close any component once you are
in the GUI.
This command is valid only at the Begin Level.
Menu Paths
This command cannot be accessed from a menu.
MSTERM, SPNUM, STER, TTER
Sets the convergence monitors for species.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
STER
Termination criteria for steady-state analysis. Defaults to 1 x 10
-8
.
TTER
Termination criteria for transient analysis. Defaults to 1 x 10
-6
.
Notes
Repeat command to set each species number as required.
All specified criteria must be met before the case is terminated.
1063
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MSTERM
If a termination criterion for a specific species number is set negative, the termination check is ignored for
that particular species.
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
MSVARY, SPNUM, Lab, Key
Allows species properties to vary between global iterations.
PREP7: FLOTRAN Multiple Species
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
SPNUM
Species number, from 1 to 6. Must be specified.
Lab
Label identifying the species property:
DENS
Density.
VISC
Viscosity.
COND
Thermal conductivity.
MDIF
Mass diffusion coefficient.
Key
Key to allow property variation between global iterations:
OFF
Variation not allowed (default).
ON
Variation allowed.
Command Default
No property is allowed to vary between global iterations within a load step.
Notes
This command is valid for the multiple species transport option in a FLOTRAN analysis.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1064
MSVARY
Menu Paths
Main Menu>Preprocessor>FLOTRAN Set Up>Multiple Species
Main Menu>Solution>FLOTRAN Set Up>Multiple Species
*MWRITE, ParR, Fname, Ext, --, Label, n1, n2, n3
Writes a matrix to a file in a formatted sequence.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the array parameter. See *SET for name restrictions.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
If the file name fields are left blank, the default file is the current output file.
Ext
Filename extension (8 character maximum).
--
Unused field.
Label
Can use a value of IJK, IKJ, JIK, JKI, KIJ, KJI, or blank (JIK).
n1, n2, n3
Write as (((ParR(i,j,k), k = 1,n1), i = 1, n2), j = 1, n3) for Label = KIJ. n1, n2, and n3 default to the
corresponding dimensions of the array parameter ParR.
Notes
Writes a matrix or vector to a specified file in a formatted sequence. You can also use the *VWRITE command
to write data to a specified file. Both commands contain format descriptors on the line immediately following
the command. The format descriptors can be in either Fortran or C format.
Fortran format descriptors are enclosed in parentheses. They must immediately follow the *MWRITE command
on a separate line of the same input file. The word FORMAT should not be included. The format must specify
the number of fields to be written per line, the field width, the placement of the decimal point, etc. There
should be one field descriptor for each data item written. The write operation uses the available system
FORTRAN FORMAT conventions (see your system FORTRAN manual). Any standard FORTRAN real format
(such as (4F6.0), (E10.3,2X,D8.2), etc.) and character format (A) may be used. Integer (I) and list-directed (*)
descriptors may not be used. Text may be included in the format as a quoted string. The FORTRAN descriptor
must be enclosed in parentheses and the format must not exceed 80 characters (including parentheses).
The C format descriptors are used if the first character of the format descriptor line is not a left parenthesis.
C format descriptors may be up to 80 characters long, consisting of text strings and predefined "data
1065
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*MWRITE
descriptors" between the strings where numeric or alphanumeric character data are to be inserted. The
normal descriptors are %I for integer data, %G for double precision data, %C for alphanumeric character
data, and %/ for a line break. Each descriptor must be preceded by a blank. There must be one data descriptor
for each specified value in the order of the specified values. The enhanced formats described in *MSG may
also be used.
The starting array element number must be defined. Looping continues in the directions indicated by the
Label argument. The number of loops and loop skipping may also be controlled with the *VLEN and *VMASK
commands, which work in the n2 direction (by row on the output file), and by the *VCOL command, which
works in the n1 direction (by column in the output file). The vector specifications *VABS and *VFACT apply
to this command, while *VCUM does not apply to this command. See the *VOPER command for details. If
you are in the GUI, the *MWRITE command must be contained in an externally prepared file and read into
ANSYS (i.e., *USE, /INPUT, etc.).
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Parameters>Write to File
MXPAND, NMODE, FREQB, FREQE, Elcalc, SIGNIF, MSUPkey
Specifies the number of modes to expand and write for a modal or buckling analysis.
SOLUTION: Dynamic Options
SOLUTION: Nonlinear Options
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS
NMODE
Number of modes to expand and write. If blank, expand and write all modes within the frequency range
specified. If -1, do not expand and do not write modes to the results file during the analysis.
FREQB
Beginning, or lower end, of frequency range of interest. If FREQB and FREQE are both blank, expand
and write the number of modes specified without regard to the frequency range. Defaults to the entire
range.
FREQE
Ending, or upper end, of frequency range of interest.
Elcalc
Element calculation key:
NO
Do not calculate element results and reaction forces (default).
YES
Calculate element results and reaction forces, as well as the nodal degree of freedom solution.
SIGNIF
Expand only those modes whose significance level exceeds the SIGNIF threshold. The significance level
of a mode is defined as the mode coefficient of the mode, divided by the maximum mode coefficient
of all modes. Any mode whose significance level is less than SIGNIF is considered insignificant and is
not expanded. The higher the SIGNIF threshold, the fewer the number of modes expanded. SIGNIF
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1066
MXPAND
defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0. SIGNIF value is only used for single-
point or DDAM response (SPOPT,SPRS or DDAM) analyses.
MSUPkey
Element result superposition key:
NO
Do not write element results to the mode file Jobname.MODE.
YES
Write element result to the mode file for use in the expansion pass of a subsequent mode superpos-
ition, PSD, transient, or harmonic analysis (default if Elcalc = YES).
Command Default
Mode shapes (displacements) are expanded and output to the results file (so that if you need to inspect the
mode shapes, it is not necessary to run the analysis again). Element stresses are not expanded.
Notes
Specifies the number of modes to expand and write over a frequency range for a modal (ANTYPE,MODAL)
or buckling (ANTYPE,BUCKLE) analysis. For reduced analyses, an expansion is required. If used in SOLUTION,
this command is valid only within the first load step.
There is no limit on the number of expanded modes (NMODE). However, there is a limit on the maximum
number of modes used via the *GET,,MODE command, mode combinations, and the MDAMP command.
With MSUPkey = YES, the computed element results (Elcalc = YES) are written to Jobname.MODE for
subsequent use in downstream mode superposition analyses, including harmonic, transient and PSD analyses.
This reduces computation time significantly for the combination or expansion passes.
In a Distributed ANSYS analysis, you must issue MXPAND to specify the number of modes to expand when
computing the modes and mode shapes. In a Distributed ANSYS run, MXPAND cannot be issued in an ex-
pansion pass (EXPASS).
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>Expand Modes
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>Expand Modes
1067
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
MXPAND
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1068
N Commands
N, NODE, X, Y, Z, THXY, THYZ, THZX
Defines a node.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Node number to be assigned. A previously defined node of the same number will be redefined. Defaults
to the maximum node number used +1.
X, Y, Z
Node location in the active coordinate system (R, , Z for cylindrical, R, , for spherical or toroidal). If
X = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
THXY
First rotation about nodal Z (positive X toward Y).
THYZ
Second rotation about nodal X (positive Y toward Z).
THZX
Third rotation about nodal Y (positive Z toward X).
Notes
Defines a node in the active coordinate system [CSYS]. The nodal coordinate system is parallel to the global
Cartesian system unless rotated. Rotation angles are in degrees and redefine any previous rotation angles.
See the NMODIF, NANG, NROTAT, and NORA commands for other rotation options.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>In Active CS
Main Menu>Preprocessor>Modeling>Create>Nodes>On Working Plane
NANG, NODE, X1, X2, X3, Y1, Y2, Y3, Z1, Z2, Z3
Rotates a nodal coordinate system by direction cosines.
PREP7: Nodes
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
NODE
Rotate coordinate system of this node.
X1, X2, X3
Global X, Y, Z components of a unit vector in new nodal X direction.
1069
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Y1, Y2, Y3
Global X, Y, Z components of a unit vector in new nodal Y direction.
Z1, Z2, Z3
Global X, Y, Z components of a unit vector in new nodal Z direction.
Notes
Rotates a nodal coordinate system to the orientation specified by the X, Y and Z direction cosines. Existing
rotation specifications on the node are redefined. If only two of the three unit vectors are specified, the third
is defined according to the right hand rule. It is the responsibility of the user to ensure that input direction
cosines are orthogonal in a right-handed system.
See the NMODIF, NROTAT, and NORA commands for other rotation options.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>RotateNode>By Vectors
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>By Vectors
NAXIS, Action, Val
Generates nodes for general axisymmetric element sections.
PREP7: Nodes
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Action
Specifies one of the following command behaviors:
GEN
Generates nodes around the axis of an axisymmetric section. This action is the default behavior.
CLEAR
Clears all nodes around the axis of an axisymmetric section.
EFACET
Specifies the number of facets between nodal planes to display using PowerGraphics and the
/ESHAPE,1 command.
Val
Tolerance value or number of facets:
TOLER
When Action = GEN, the tolerance to use for merging the generated nodes around the axis.
NUM
When Action = EFACET, the number of facets per element edge for element plots:
AUTO
Use program-chosen facets per edge (default).
1
Use 1 facet per edge (default for elements with 9, 10, 11, or 12 node planes). Shows nodal and
integration planes only.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1070
NAXIS
2
Use 2 facets per edge (default for elements with 5, 6, 7, or 8 node planes, and maximum for
elements with 9, 10, 11, or 12 node planes).
3
Use 3 facets per edge (default for elements with 3 or 4 node planes, and maximum for elements
with 6, 7, or 8 node planes).
4
Use 4 facets per edge (maximum for elements with 5 node planes).
5
Use 5 facets per edge (maximum for elements with 4 node planes).
6
Use 6 facets per edge (maximum for elements with 3 node planes).
Notes
The NAXIS command generates or clears the nodes for general axisymmetric element sections. The command
applies to elements SOLID272 and SOLID273.
The generate option (Action = GEN) operates automatically on any current-technology axisymmetric element.
Any nodes within the tolerance value (TOLER) of the axis are merged into a single node. The default tolerance
is 1.0e-4.
If you want to change the number of nodes, use the clear option (Action = CLEAR) before regenerating
the nodes.
To cause the 3-D element plot to appear more like the actual 3-D model, use NAXIS,EFACET,NUM, where
NUM > 1. In this case, the coordinate system specified for displaying element and nodal results (RSYS) must
be either global (RSYS,0) or solution (RSYS,SOLU); otherwise, ANSYS resets NUM to 1.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>About AXIS Section
NCNV, KSTOP, DLIM, ITLIM, ETLIM, CPLIM
Sets the key to terminate an analysis.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
KSTOP
Program behavior upon nonconvergence:
0
Do not terminate the analysis if the solution fails to converge.
1
Terminate the analysis and the program execution if the solution fails to converge (default).
2
Terminate the analysis, but not the program execution, if the solution fails to converge.
1071
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NCNV
DLIM
Terminates program execution if the largest nodal DOF solution value (displacement, temperature, etc.)
exceeds this limit. Defaults to 1.0E6 for all DOF except MAG and A and FLOTRAN DOF. Defaults to 1.0E10
for MAG and A. Defaults to 1.0E20 for the FLOTRAN VX, VY, VZ, PRES, and TEMP DOF. Does not apply to
the FLOTRAN ENKE and ENDS DOF.
ITLIM
Terminates program execution if the cumulative iteration number exceeds this limit (defaults to infinity).
ETLIM
Terminates program execution if the elapsed time (seconds) exceeds this limit (defaults to infinity).
CPLIM
Terminates program execution if the CPU time (seconds) exceeds this limit (defaults to infinity).
Command Default
As described for each argument above.
Notes
Sets the key to terminate an analysis if not converged, or if any of the following limits are exceeded for
nonlinear and full transient analyses: DOF (displacement), cumulative iteration, elapsed time, or CPU time
limit. Applies only to static and transient analyses (ANTYPE,STATIC and ANTYPE,TRANS), and FLOTRAN
analyses. Time limit checks are made at the end of each equilibrium iteration. This command is ignored
during an optimization run.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Advanced NL
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Criteria to Stop
Main Menu>Solution>Analysis Type>Sol'n Controls>Advanced NL
Main Menu>Solution>Load Step Opts>Nonlinear>Criteria to Stop
NDELE, NODE1, NODE2, NINC
Deletes nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1, NODE2, NINC
Delete nodes from NODE1 to NODE2 (defaults to NODE1) in steps of NINC (defaults to 1). If NODE1 =
ALL, NODE2 and NINC are ignored and all selected nodes [NSEL] are deleted. If NODE1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NODE1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1072
NDELE
Notes
Deletes selected nodes that are not connected to elements. Nodes may also be redefined instead of deleted,
if desired. Boundary conditions (displacements, forces, etc.) as well as any coupling or constraint equations
containing the deleted nodes are also deleted.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Main Menu>Preprocessor>Modeling>Delete>Nodes
NDIST, ND1, ND2
Calculates and lists the distance between two nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ND1
First node in distance calculation. If ND1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
ND2
Second node in distance calculation.
Notes
NDIST lists the distance between nodes ND1 and ND2, as well as the current coordinate system offsets from
ND1 to ND2, where the X, Y, and Z locations of ND1 are subtracted from the X, Y, and Z locations of ND2
(respectively) to determine the offsets. NDIST is valid in any coordinate system except toroidal [CSYS,3].
NDIST returns a variable, called "_RETURN," which contains the distance value. You can use this value for
various purposes, such as the calculation of distributed loads. In interactive mode, you can access this
command by using the Model Query Picker (Utility Menu> List> Picked Entities), where you can also access
automatic annotation functions and display the value on your model.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Check Geom>ND distances
1073
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NDIST
NDSURF, Snode, Telem, DIMN
Generates surface elements overlaid on the edge of existing elements and assigns the extra node as the
closest fluid element node.
PREP7: Elements
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
Snode
Component name for the surface nodes of the solid elements.
Telem
Component name for the target fluid elements.
DIMN
Model dimensionality:
2
2-D model.
3
3-D model.
Notes
This command macro is used to generate surface effect elements (SURF151 or SURF152) overlaid on the
surface of existing plane or solid elements and, based on proximity, to determine and assign the extra node
(FLUID116) for each surface element. The active element type must be SURF151 or SURF152 with appropriate
settings for KEYOPT(4), KEYOPT(5), KEYOPT(6), and KEYOPT(8).
The surface nodes of the plane or solid elements must be grouped into a node component and the fluid
elements must be grouped into an element component and named using the CM command. The names
must be enclosed in single quotes (e.g., 'NOD') when the NDSURF command is manually typed in.
When using the GUI method, node and element components are created through the picking dialog boxes
associated with this command.
The macro is applicable for the SURF151, SURF152, and FLUID116 element types.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Attch to Flu-
id>Area to Fluid
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Attch to Fluid>Line
to Fluid
Main Menu>Preprocessor>Modeling>Create>Elements>Surf / Contact>Surf Effect>Attch to Flu-
id>Node to Fluid
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1074
NDSURF
NEQIT, NEQIT, FORCEkey
Specifies the maximum number of equilibrium iterations for nonlinear analyses.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> EM <> DY PP <> EME MFS
NEQIT
Maximum number of equilibrium iterations allowed each substep.
FORCEkey
One iteration forcing key (used only when NEQIT = 1 and SOLCONTROL,ON):
FORCE
Forces one iteration per substep.
(blank)
SOLCONTROL controls the minimum number of iterations (internally set to two iterations per substep
if SOLCONTROL,ON).
Note
Using one iteration per substep may result in unconverged solutions for nonlinear analysis,
and the program may not indicate divergence in this case. This option is intended primarily
for use by the ANSYS Workbench interface. Keep in mind that forcing one iteration per substep
is only recommended under very specific conditions; for example, nonlinearity in bonded
penalty type contact. Under these conditions the solution typically converges in one iteration.
Command Default
When SOLCONTROL,ON, NEQIT is set between 15 and 26 depending on the physics of the problem. When
SOLCONTROL,OFF, NEQIT defaults to 25 for all cases.
Notes
See the description of SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and
SOLCONTROL,OFF.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Equilibrium Iter
Main Menu>Solution>Analysis Type>Sol'n Controls>Nonlinear
Main Menu>Solution>Load Step Opts>Nonlinear>Equilibrium Iter
1075
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NEQIT
/NERR, NMERR, NMABT, --, IFKEY, NUM
Limits the number of warning and error messages displayed.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NMERR
Maximum number of warning and error messages displayed per command. Defaults to 5 for interactive
runs with the GUI turned on, 20 for interactive runs with the GUI turned off, 200 for batch runs. If NMERR
is negative, the absolute value of NMERR is used as the maximum number of warning and error messages
written to the error file (file.ERR) per command, as well as the maximum number of messages displayed
per command.
NMABT
Maximum number of warning and error messages allowed per command before run aborts (must be
greater than zero). Maximum value is 99,999,999. Defaults to 10,000.
--
Unused field.
IFKEY
Specifies whether or not to abort if an error occurs during a /INPUT operation:
0 or OFF
Do not abort. This option is the default.
1 or ON
Abort.
NUM
The number of invalid command warnings before a stop warning will be issued:
0
Disables the stop warning/error function.
n
An integer value representing the number of warnings that will be encountered before prompting
the user to stop (default = 5). The first error encountered will ALWAYS result in a prompt.
Note
Invalid command warnings and error tracking are mutually exclusive.
Command Default
As described above.
Notes
Limits the number of warning and error messages displayed for any one command in an interactive run.
Warning and error messages continue to be written to Jobname.ERR regardless of these limits (unless
NMERR is negative).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1076
/NERR
Issue this command with NUM = n to specify the number of "invalid command" warnings to be encountered
before the user is prompted to stop. You can then continue or abort the run. If you choose to abort the run,
the log file can be saved so that any of the processing up to that point can be appended to an input that
rectifies the condition. A batch run always aborts on the first error. Issue /NERR,STAT to list current settings.
Issue /NERR,DEFA to reset values to initial defaults.
An IFKEY value of 1 or ON causes the ANSYS program to abort immediately upon encountering an error
during a file /INPUT operation. However, use of this option may cause the following conditions to occur:
The /INPUT command may abort if issued for a log file (jobname.log).
Some macros may abort.
A CAD connection may fail after reading only a small portion of a CAD model.
The command is valid in any processor.
Menu Paths
Utility Menu>MenuCtrls>Message Controls
NFORCE, ITEM
Sums the nodal forces and moments of elements attached to nodes.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
ITEM
Specifies the selected set of nodes for summing forces and moments for contact elements.
(blank)
Sums the nodal forces of elements for all selected nodes and excludes contact elements (elements
169-177).
CONT
Sums the nodal forces of elements for contact nodes only.
BOTH
Sums the nodal forces of elements for all selected nodes, including contact elements.
Notes
Sums and prints, in each component direction for each selected node, the nodal force and moment contri-
butions of the selected elements attached to the node. If all elements are selected, the sums are usually
zero except where constraints or loads are applied. The nodal forces and moments may be displayed
[/PBC,FORC and /PBC,MOME]. Use PRESOL to print nodal forces and moments on an element-by-element
basis. You can use the FORCE command to specify which component (static, damping, inertia, or total) of
the nodal load is to be used. Nodal forces associated with surface loads are not included.
This vector sum is printed in the global Cartesian system. Moment summations are about the global origin
unless another point is specified with the SPOINT command. The summations for each node are printed in
the global Cartesian system unless transformed [RSYS]. This command is generally not applicable to
1077
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NFORCE
axisymmetric models because moment information from the NFORCE command is not correct for axisym-
metric elements.
Selecting a subset of elements [ESEL] and then issuing this command will give the forces and moments re-
quired to maintain equilibrium of that set of elements. The effects of nodal coupling and constraint equations
are ignored. The option ITEM = CONT provides the forces and moments for the contact elements (CONTA171,
CONTA172, CONTA173, CONTA174, CONTA175, CONTA176, and CONTA177). Setting ITEM = BOTH provides
the forces and moments for all selected nodes, including contact elements.
This command also includes the FSUM command function which vectorially sums and prints, in each com-
ponent direction for the total selected node set, the nodal force and moment contributions of the selected
elements attached to the selected node set.
Menu Paths
Main Menu>General Postproc>Nodal Calcs>Sum @ Each Node
NGEN, ITIME, INC, NODE1, NODE2, NINC, DX, DY, DZ, SPACE
Generates additional nodes from a pattern of nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME, INC
Do this generation operation a total of ITIME times, incrementing all nodes in the given pattern by
INC each time after the first. ITIME must be > 1 for generation to occur.
NODE1, NODE2, NINC
Generate nodes from the pattern of nodes beginning with NODE1 to NODE2 (defaults to NODE1) in
steps of NINC (defaults to 1). If NODE1 = ALL, NODE2 and NINC are ignored and the pattern is all selected
nodes [NSEL]. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NODE1 (NODE2 and NINC are
ignored).
DX, DY, DZ
Node location increments in the active coordinate system (DR, D, DZ for cylindrical, DR, D, D for
spherical or toroidal).
SPACE
Spacing ratio. Ratio of last division size to first division size. If > 1.0, divisions increase. If < 1.0, divisions
decrease. Ratio defaults to 1.0 (uniform spacing).
Note
The average spacing ratio remains 1.0, such that the location of the last generated set will be
the same regardless of SPACE. SPACE only serves to skew the position of the nodes between
the pattern set and the last set.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1078
NGEN
Notes
Generates additional nodes from a given node pattern. Generation is done in the active coordinate system.
Nodes in the pattern may have been generated in any coordinate system.
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Nodes>Copy
NKPT, NODE, NPT
Defines a node at an existing keypoint location.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Arbitrary reference number for node. If zero or blank, defaults to the highest node number +1 [NUMSTR].
NPT
Keypoint number defining global X, Y, Z location. If NPT = All, then a node will be placed at each selected
keypoint. If NPT = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NPT.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>On Keypoint
NLDIAG, Label, Key
Sets nonlinear diagnostics functionality.
SOLUTION: Nonlinear Options
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
Label
Specifies a diagnostic function:
NRRE
Store the Newton-Raphson residuals information.
EFLG
Identify or display elements or nodes that violate the criteria.
MAXF
Set a maximum number of files (MaxFile) to create. Information is written to Jobname.ndxxx
or Jobname.nrxxx, where xxx iterates from 001 through MaxFile. When the maximum number
of files is reached, the counter resets to 001 and earlier files are overwritten.
CONT
Write contact information to a single Jobname.cnd text file during solution.
1079
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NLDIAG
Key
Sets the characteristics of the diagnostic function:
OFF, or 0
Suppresses writing of diagnostic information. This value is the default.
ON, or 1
Writes diagnostic information to the Jobname.ndxxx, Jobname.nrxxx, or Jobname.cnd file.
(If Label = CONT, this option is the same as the SUBS option described below.)
ITER
Writes contact diagnostic information at each iteration. Valid only when Label = CONT.
SUBS
Writes contact diagnostic information at each substep. Valid only when Label = CONT.
LSTP
Writes contact diagnostic information at each load step. Valid only when Label = CONT.
STAT
Lists information about the diagnostic files in the current working directory.
DEL
Deletes all diagnostic files in the current working directory.
MaxFile
Sets the maximum number of files to create (Jobname.ndxxx or Jobname.nrxxx). Valid values
are 1 through 999. The default is 4. (If you change the default MaxFile value, that value is retained
until you change it again.) Valid only when Label = MAXF.
Command Default
No nonlinear diagnostic files are written.
Notes
The NLDIAG command is a nonlinear diagnostics tool valid for nonlinear structural analyses. It is a debugging
tool for use when you must restart after an unconverged solution. The command creates Jobname.ndxxx,
Jobname.nrxxx, or Jobname.cnd files in the working directory to store the information you specify.
Newton-Raphson residual Issue the NLDIAG,NRRE,ON command to create Jobname.nrxxx files which
store the relevant Newton-Raphson residual information for the last MaxFile equilibrium iterations. Issue
a NLDPOST,NRRE,STAT command to list the load step, substep, time, and equilibrium iteration corresponding
to each of the Jobname.nrxxx files in the working directory, then issue a PLNSOL,NRRES,,,,FileID
command to point to the file from which you want to create a contour plot of your Newton-Raphson residuals.
For NLDIAG,NRRE,ON, any Jobname.nrxxx files in the current (working) directory are overwritten when
you issue a new SOLVE command (or restart).
Element components that violate criteria Issue a NLDIAG,EFLG,ON command to create Jobname.ndxxx
files which store IDs for elements violating the following criteria:
Too large a distortion
Elements contain nodes that have near zero pivots for nonlinear analyses
Too large a plastic/creep strain increment (CUTCONTROL)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1080
NLDIAG
Elements for which mixed u-P constraints are not satisfied (mixed U-P option of 18x solid elements
only)
For NLDIAG,EFLG,ON, all Jobname.ndxxx diagnostic files in the current (working) directory are deleted
when you issue a new SOLVE command (or restart).
In the solution processor (/SOLU), use the STAT option to list the active status of this command. In the
postprocessor (/POST1), issue a NLDPOST,EFLG,STAT command to list the load step, substep, time, and
equilibrium iteration corresponding to each of the Jobname.ndxxx files in the working directory, then issue
a NLDPOST,EFLG,CM,FileID command to create element components that violate the criteria.
Contact diagnostic Issue the NLDIAG,CONT,ON command to create a Jobname.cnd file which stores
contact information for all defined contact pairs at all substeps. Alternatively, you may issue one of the fol-
lowing commands to store contact information at a specific frequency:
NLDIAG,CONT,ITER to write at each iteration
NLDIAG,CONT,SUBS to write at each substep (default)
NLDIAG,CONT,LSTP to write at each load step
File Jobname.cnd is written during solution and lists, on a pair-base, the following contact information:
Number of contact elements in contact
[1]
Number of contact elements in "sticking" contact status
Maximum chattering level
Maximum contact penetration/Minimum gap
[2]
Maximum closed gap
Maximum normal contact stiffness
Minimum normal contact stiffness
Maximum resulting pinball
Maximum elastic slip distance
Maximum tangential contact stiffness
Minimum tangential contact stiffness
Maximum sliding distance
Maximum contact pressure
Maximum friction stress
Average contact depth
Maximum closed penetration
Number of contact points having too much penetration
1. If >0, this quantity indicates the number of contact elements in contact. Other values are interpreted
as follows:
0 indicates that the contact pair is in near-field contact status.
-1 indicates that the contact pair is in far-field contact status.
-2 indicates that the contact pair is inactive (symmetric to asymmetric contact)
1081
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NLDIAG
2. A positive value indicates penetration and a negative value indicates a gap. If the contact pair has a
far-field contact status, penetration and gap are not available and the value stored is the current pinball
radius.
In the solution processor (/SOLU), use the NLDIAG,CONT,STAT command to list the active status of the
contact information. If you subsequently issue a new SOLVE command (or restart), the Jobname.cnd file
in the current (working) directory is not deleted; information is appended to it. Delete the existing file
(NLDIAG,CONT,DEL command) if you do not want to retain diagnostic information from previous solutions.
For more information, see nonlinear diagnostics.
Distributed ANSYS supports only the contact diagnostic option (NLDIAG,CONT).
Menu Paths
Main Menu>Solution>Diagnostics>Nonlinear Diagnostics
NLDPOST, Label, Key, FileID, Prefix
Gets element component information from nonlinear diagnostic files.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Label
Specifies the type of command operation:
EFLG
Element flag for nonlinear diagnostics.
NRRE
Newton-Raphson residuals.
Key
Specifies the command action:
STAT
List information about the diagnostic files (Jobname.ndxxx or Jobname.nrxxx) in the current
directory.
For Label = EFLG, the listing gives a summary that associates the loadstep, substep, time, equilib-
rium iteration number, cumulative iteration number, and the number of elements that fail each cri-
teria with a specific file ID (Jobname.ndxxx). Use the list to create element components (via the
CM option) based on the cumulative iteration number.
For Label = NRRE, the listing provides a summary that associates the loadstep, substep, time,
equilibrium iteration number, and cumulative iteration number with a specific file ID (Job-
name.nrxxx). Use the list to identify the respective file ID for creating Newton-Raphson residual
contour plots (PLNSOL,NRRE,,FileID).
DEL
Delete Jobname.ndxxx or Jobname.nrxxx files in the working directory, if any exist.
CM
Create components for elements that violate criteria. This value is valid only when Label = EFLG.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1082
NLDPOST
FileID
Valid only when Label = EFLG and Key = CM, this value specifies file IDs:
IDnum
The file ID number. Creates the element components from the diagnostic files corresponding to the
specified file ID number in the working directory.
ALL
Creates element components from all available diagnostic files residing in the working directory.
This value is the default if you do not specify an IDnum value.
Prefix
Sets the prefix name for components. Specify up to 21 alphanumeric characters.
Command Default
The NLDPOST command returns no nonlinear diagnostic information.
Notes
Based on the nonlinear diagnostic results (created via the NLDIAG,EFLG command), the NLDPOST command
creates element components with predefined names.
The following table lists the diagnostic criteria and component names (with specified prefix and without).
Here xxx corresponds to the file ID (FileID) of Jobname.ndxxx or Jobname.nrxxx.
Without prefix specified: If prefix is specified: Elements with:
ndxxx_eppl prefix_xxx_eppl Too large a plastic strain increment
ndxxx_epcr prefix_xxx_epcr Too large a creep strain increment
ndxxx_mxup prefix_xxx_mxup Unsatisfied mixed u-P constraints
ndxxx_hdst prefix_xxx_hdst Too much distortion
ndxxx_pivt prefix_xxx_pivt Negative/small pivots
For more information, see Performing Nonlinear Diagnostics.
Menu Paths
Main Menu>General Postproc>Nonlinear Diagnostics
NLGEOM, Key
Includes large-deflection effects in a static or full transient analysis.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Key
Large-deflection key:
1083
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NLGEOM
OFF
Ignores large-deflection effects (that is, a small-deflection analysis is specified). This option is the
default.
ON
Includes large-deflection (large rotation) effects or large strain effects, according to the element type.
Command Default
Large-deflection effects are ignored.
Notes
Large-deflection effects are categorized as either large deflection (or large rotation) or large strain, depending
on the element type. These are listed (if available) under Special Features in the input data table for each
element in the Element Reference. If used during the solution (/SOLU), this command is valid only within the
first load step.
In a large-deflection analysis, pressure loads behave differently than other load types. For more information,
see Load Direction in a Large-Deflection Analysis.
The gyroscopic matrix (that occurs due to rotational angular velocity) does not support large-deflection effects.
The theoretical formulations for the gyroscopic matrix support small deflection (linear formulation) only.
This command is also valid in PREP7.
Product Restrictions
In ANSYS Professional NLT, large deflection effects should not be turned on if 2-D solid (PLANEn) or 3-D
solid (SOLIDn) elements are defined. ANSYS Professional NLS supports NLGEOM,ON for plane and solid ele-
ments.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Basic
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Sol'n Controls>Basic
NLHIST, Key, Name, Item, Comp, NODE, ELEM, SHELL, LAYER
Specify result items to track during solution.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Key
Specifies the command operation:
NSOL
Nodal solution data.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1084
NLHIST
ESOL
Element nodal data.
PAIR
Contact data (pair-based).
STAT
Displays a list of items to track.
OFF or 0
Deactivates tracking of all variables. This value is the default.
ON or 1
Activates tracking of all variables. Tracking also activates whenever any specification changes.
DEL
Removes the specified variable from the set of result items to track. If Name = ALL (default), all spe-
cifications are removed.
Name
The 32-character user-specified name.
Item, Comp
Predetermined output item and component label for valid elements. See the Element Reference for more
information.
NODE
Valid node number (if Key = NSOL or ESOL), or valid real constant set number (if Key = PAIR) identifying
a contact pair. Required input for NSOL, ESOL, and PAIR items.
ELEM
Valid element number for element results. Used for ESOL items. If ELEM is specified, then a node number
that belongs to the element must also be specified in the NODE field.
SHELL
Valid labels are TOP, MID or BOT. This field can specify the location on shell elements for which to retrieve
data. Used only for element nodal data (ESOL).
LAYER
Layer number (for layered elements only). Used only for element nodal data (ESOL).
Notes
The NLHIST command is a nonlinear diagnostics tool allowing you to monitor diagnostics results of interest
in real time during solution.
You can track a maximum of 50 variables during solution. The specified result quantities are written to the
file Jobname.nlh. Nodal results and contact results are written for every converged substep (irrespective
of the OUTRES command setting) while element results are written only at time points specified via the
OUTRES command. For time points where element results data is not available, a very small number is
written instead. If the conditions for contact to be established are not satisfied, 0.0 will be written for contact
results.
Results tracking is available only for a nonlinear structural analysis (static or transient), a nonlinear steady-
state thermal analysis, or a transient thermal analysis (linear or nonlinear). All results are tracked in the
Solution Coordinate System (that is, nodal results are in the nodal coordinate system and element results
are in the element coordinate system). Results tracking is not available for FLOTRAN analyses.
1085
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NLHIST
The Jobname.nlh file is an ASCII file that lists each time point at which a converged solution occurs along
with the values of the relevant result quantities.
The GUI option Solution> Results tracking provides an interface to define the result items to be tracked.
The GUI also allows you to graph one or more variables against time or against other variables during
solution. You can use the interface to graph or list variables from any .nlh file generated by the ANSYS
program.
You can also track results during batch runs. Either access the ANSYS Launcher and select File Tracking
from the Tools menu, or type nlhist120 at the command line. Use the supplied file browser to navigate
to your Jobname.nlh file, and click on it to invoke the tracking utilty. You can use this utilty to read the
file at any time, even after the solution is complete (the data in the file must be formatted correctly).
Specifications that you set via the NLHIST command are not saved in the database (.db file).
Table 227 NLHIST - Valid NSOL Item and Component Labels
Description Comp Item
X, Y, or Z structural displacement. X, Y, Z U
X, Y, or Z structural rotation. X, Y, Z ROT
X, Y, or Z structural reaction force. X, Y, Z F
X, Y, or Z structural reaction moment. X, Y, Z M
Temperature. - TEMP [1]
Maximum or minimum temperature in the model. MAX, MIN TEMP
Reaction heat flow. - HEAT [2]
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels TBOT, TE2, TE3, . . ., TTOP
instead of TEMP.
2. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels HBOT, HE2, HE3, . . .,
HTOP instead of HEAT.
Table 228 NLHIST - Valid ESOL Item and Component Labels
Description Comp Item
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity. INT "
Equivalent stress. EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity. INT "
Elastic equivalent strain. EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity. INT "
Plastic equivalent strain. EQV "
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1086
NLHIST
Description Comp Item
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity. INT "
Creep equivalent strain. EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity. INT "
Thermal equivalent strain. EQV "
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Accumulated equivalent creep strain. CREQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
ETABLE items are not supported for ESOL items.
PAIR solution quantities are output on a "per contact pair" basis. As a consequence, the corresponding values
listed in the Jobname.nlh file represent a minimum or a maximum over the associated contact pair, as
detailed in the table below.
Table 229 NLHIST - Valid PAIR Item and Component Labels
Description Comp Item
If >0, number of contact elements in contact. Other values
are interpreted as follows:
ELCN CONT
0 indicates the contact pair is in near-field contact status.
-1 indicates the contact pair is in far-field contact status.
-2 indicates that the contact pair is inactive (symmetric to
asymmetric contact).
Number of contact elements in sticking contact status. ELST "
Maximum chattering level. CNOS "
Maximum penetration (or minimum gap). [1] PENE "
Maximum closed (geometrical) gap. CLGP "
Maximum total sliding distance. SLID "
Maximum elastic slip distance. ESLI "
Maximum normal contact stiffness. KNMX "
Maximum tangential contact stiffness. KTMX "
1087
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NLHIST
Description Comp Item
Minimum normal contact stiffness. KNMN "
Minimum tangential contact stiffness. KTMN "
Maximum pinball radius. PINB "
Maximum contact pressure. PRES "
Maximum frictional stress. SFRI "
Average contact depth. CNDP "
Maximum closed (geometrical) penetration. CLPE "
Number of contact points having too much penetration. LGPE "
1. For PENE, a positive value indicates a penetration, and a negative value indicates a gap. If the contact
pair has a far-field contact status, penetration and gap are not available, and the value stored for PENE
is the current pinball radius.
Distributed ANSYS supports only the tracking of pair-based contact data (NLHIST,PAIR) and the applicable
supporting options (STAT / ON / OFF / DEL).
Menu Paths
Main Menu>Solution>Results Tracking
NLIST, NODE1, NODE2, NINC, Lcoord, SORT1, SORT2, SORT3, KINTERNAL
Lists nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1, NODE2, NINC
List nodes from NODE1 to NODE2 (defaults to NODE1) in steps of NINC (defaults to 1). If NODE1 = ALL
(default), NODE2 and NINC are ignored and all selected nodes [NSEL] are listed. If NODE1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NODE1 (NODE2 and NINC are ignored).
Lcoord
Coordinate listing key:
(blank)
List all nodal information
COORD
Suppress all but the XYZ coordinates (shown to a higher degree of accuracy than when displayed
with all information).
SORT1
First item on which to sort. Valid item names are NODE, X, Y, Z, THXY, THYZ, THXZ
SORT2, SORT3
Second and third items on which to sort. Valid item names are the same as for SORT1.
KINTERNAL
Internal nodes listing key:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1088
NLIST
(blank)
List only external nodes.
INTERNAL
List all nodes, including internal nodes.
Notes
Lists nodes in the active display coordinate system [DSYS]. Nodal coordinate rotation angles are also listed
(relative to the global Cartesian coordinate system).
Node listing can be in a sorted order (ascending). SORT2, for example, will be carried out on nodes having
equal values of SORT1.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>List Results>Sorted Listing>Sort Nodes
Utility Menu>List>Nodes
NLOG, IR, IA, --, --, Name, --, --, FACTA, FACTB
Forms the natural log of a variable.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA
Reference number of the variable to be operated on.
--, --
Unused fields.
Name
Thirty-two character name identifying the variable on printouts and displays. Embedded blanks are
compressed for output.
--, --
Unused fields.
FACTA
Scaling factor applied to variable IA (defaults to 1.0).
FACTB
Scaling factor (positive or negative) applied to the operation (defaults to 1.0).
Notes
Forms the natural log of a variable according to the operation:
1089
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NLOG
IR = FACTB*LN(FACTA x IA)
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Natural Log
NLOPT
Specifies "Nonlinear analysis options" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM <> DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Nonlinear Options
NMODIF, NODE, X, Y, Z, THXY, THYZ, THZX
Modifies an existing node.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE
Modify coordinates of this node. If ALL, modify coordinates of all selected nodes [NSEL]. If NODE = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NODE.
X, Y, Z
Replace the previous coordinate values assigned to this node with these corresponding coordinate values.
Values are interpreted in the active coordinate system (R, , Z for cylindrical; R, , for spherical or tor-
oidal). Leaving any of these fields blank retains the previous value(s).
THXY
First rotation of nodal coordinate system about nodal Z (positive X toward Y). Leaving this field blank
retains the previous value.
THYZ
Second rotation of nodal coordinate system about nodal X (positive Y toward Z). Leaving this field blank
retains the previous value.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1090
NLOPT
THZX
Third rotation of nodal coordinate system about nodal Y (positive Z toward X). Leaving this field blank
retains the previous value.
Notes
Modifies an existing node. Nodal coordinate system rotation angles are in degrees and redefine any existing
rotation angles. Nodes can also be redefined with the N command.
See the NROTAT, NANG, and NORA commands for other rotation options.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>RotateNode>By Angles
Main Menu>Preprocessor>Modeling>Move / Modify>Nodes>Set of Nodes
Main Menu>Preprocessor>Modeling>Move / Modify>Nodes>Single Node
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>By Angles
NOCOLOR, KEY
Removes color from graphics displays.
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Color key:
0
Color the displays.
1
Do not color the displays.
2
Do not shade the displays.
Command Default
Color graphics displays (device dependent).
Menu Paths
It is part of the DISPLAY program.
1091
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NOCOLOR
NODES
Specifies "Nodes" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Nodes
/NOERASE
Prevents the screen erase between displays.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Command Default
Issue the command with no arguments to prevent automatic screen erase before creating the next display.
Notes
Preventing the normal screen erase between requested displays allows you to overlay multiple views.
Clearing the screen with the ERASE command (Utility Menu> PlotCtrls> Erase Options> Erase screen)
active simply clears the display area. Subsequent replots will provide the cumulative plots previously generated
by the /NOERASE command.
For 3-D devices, you can issue /DV3D,DELS to suppress repeated screen overlays and generate clear contour
plots.
Use the /ERASE command to reactivate automatic screen erase.
For 3-D devices (/SHOW,3D), the model in all active windows will be the same, even if you issue a different
display command (NPLOT, EPLOT, etc.) for each active window. Use the Multi-Plot command (GPLOT) to
display different entities, in different windows, on 3-D devices.
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1092
NODES
Menu Paths
Utility Menu>PlotCtrls>Erase Options>Erase between Plots
/NOLIST
Suppresses the data input listing.
SESSION: List Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Command Default
Issue the command with no arguments to suppress the data input listing.
Notes
Printout is suppressed until a /GOLIST command is read or the end of the listing is encountered.
This command is valid in any processor, but only within a batch run [/BATCH].
Menu Paths
This command cannot be accessed from a menu.
NOOFFSET, Label
Prevents the CDREAD command from offsetting specified data items
PREP7: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Specifies items not to be offset.
NODE
Node numbers
ELEM
Element numbers
KP
Keypoint numbers
LINE
Line numbers
AREA
Area numbers
VOLU
Volume numbers
1093
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NOOFFSET
MAT
Material numbers
TYPE
Element type numbers
REAL
Real constant numbers
CSYS
Coordinate system numbers
SECN
Section numbers
CP
Coupled set numbers
CE
Constraint equation numbers
CLEAR
All items will be offset
STATUS
Shows which items are specified notto be offset.
Notes
The NOOFFSET command specifies data items not to be offset by a set of data read from a CDREAD com-
mand.
Menu Paths
This command cannot be accessed from a menu.
NOORDER, Lab
Re-establishes the original element ordering.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Lab
Turns element reordering on or off.
ON (or blank)
Re-establishes original element ordering (default).
OFF
Original ordering is not used and program establishes its own ordering at the beginning of the
solution phase.
Notes
If Lab = ON, the original element ordering is re-established and no automatic reordering occurs at the be-
ginning of the solution phase. Use Lab = OFF only to remove the effect of a previous NOORDER command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1094
NOORDER
This command affects only those elements that were defined up to the point that this command is issued.
See the WSORT and WAVES commands for reordering.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Reset Elem Order
/NOPR
Suppresses the expanded interpreted input data listing.
SESSION: List Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Command Default
Issuing this command with no arguments suppresses the interpreted data input print out.
Notes
Suppresses printout of interpreted input data, including information labeled as "Notes." When this printout
is not suppressed, the data input to the analysis is echoed to the output file in an expanded format. Printout
is suppressed until a /GOPR or /GO command is read.
Use of /NOPR is not recommended when the graphical user interface (GUI) is active. The GUI sometimes issues
"hidden" /NOPR and /GOPR command sequences, which will countermand user-issued /NOPR commands,
thus making the use of /NOPR in the GUI environment unpredictable.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Rigid Region
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Smooth Data
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Smooth Data
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Smooth Data
Main Menu>Solution>Loading Options>Smooth Data
Main Menu>Solution>Time Controls>Time Step Prediction
Main Menu>TimeHist Postpro>Smooth Data
1095
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/NOPR
NORA, AREA, NDIR
Rotates nodal coordinate systems to surface normal
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
AREA
The area number containing the nodes to be rotated to their normals. If ALL, applies to all selected areas
(see the ASEL command). If AREA = P, graphical picking is enabled.
NDIR
Direction of the normal. If NDIR = -1, the nodal coordinate system is rotated in the opposite direction
of the surface normal. The default is the same direction as the surface normal.
Notes
The NORA command rotates the X-axis of the nodal coordinate system to the surface normal. The rotated
nodal coordinate systems may be displayed through the /PSYMB command. In case multiple areas are se-
lected, there could be conflicts at the boundaries. If a node belongs to two areas that have a different normal,
its nodal coordinate system will be rotated to the area normal with the lowest number. You can use the
AREVERSE and ANORM commands to rotate the surface normals in the appropriate direction. Keep the
following in mind when using the NORA command:
If the nodal coordinate system is parallel to the global Cartesian system, it is not displayed through the
/PSYMB command.
Previously specified rotation on the selected nodes are overridden.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>On Areas
NORL, LINE, AREA, NDIR
Rotates nodal coordinate systems perpendicular to line normal
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LINE
Line number containing the nodes to be rotated. If ALL, applies to all selected lines (see the LSEL com-
mand). If LINE = P, graphical picking is enabled.
AREA
The area number containing the selected lines. The normal of the line(s) selected is supposed to lie on
this area. Defaults to the lowest numbered selected area containing the line number.
NDIR
Direction of the normal. If NDIR = -1, the nodal coordinate system is rotated in the opposite direction
of the line normal. The default is the same direction as the surface normal.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1096
NORA
Notes
The NORL command rotates the X-axis of the nodal coordinate perpendicular to the line normal. The rotated
nodal coordinate systems may be displayed through the /PSYMB command. In case multiple lines are selected,
there could be conflicts at the boundaries. If a node belongs to two lines that have a different normal, its
nodal coordinate system will be rotated to the line normal with the lowest number. Keep the following in
mind when using the NORL command:
If the nodal coordinate system is parallel to the global Cartesian system, it is not displayed through the
/PSYMB command.
Previously specified rotation on the selected nodes are overridden.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>On Lines
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>with Area
/NORMAL, WN, KEY
Allows displaying area elements by top or bottom faces.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
KEY
Display key:
0
No face distinction.
1
Show only area elements having their positive normals directed toward the viewing point.
-1
Show only area elements having their positive normals directed away from the viewing point.
Command Default
Do not distinguish between top face and bottom face elements.
Notes
/NORMAL allows you to select area elements and area plots by the top or bottom faces. It is useful for
checking the normal directions on shell elements. The positive normal (element Z direction) is defined by
the right-hand rule following the node I, J, K, L input direction. This command is available only with raster
or hidden-line displays, for WIN32 or X11 2-D displays only.
This command is valid in any processor.
1097
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/NORMAL
Menu Paths
Utility Menu>PlotCtrls>Style>Shell Normals
NPLOT, KNUM
Displays nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KNUM
Node number key:
0
No node numbers on display.
1
Include node numbers on display. See also /PNUM command.
Notes
Produces a node display. Only selected nodes [NSEL] are displayed. Elements need not be defined. See the
DSYS command for display coordinate system.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>On Areas
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>On Lines
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>with Area
Utility Menu>Plot>Nodes
NPRINT, N
Defines which time points stored are to be listed.
POST26: Listing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
N
List data associated with every N time (or frequency) point(s), beginning with the first point stored (defaults
to 1).
Command Default
List all stored points.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1098
NPLOT
Notes
Defines which time (or frequency) points within the range stored are to be listed.
Menu Paths
Main Menu>TimeHist Postpro>Settings>List
NREAD, Fname, Ext, --
Reads nodes from a file.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to NODE if Fname is blank.
--
Unused field.
Notes
The read operation is not necessary in a standard ANSYS run but is provided as a convenience to users
wanting to read a coded node file, such as from another mesh generator or from a CAD/CAM program. Data
should be formatted as produced with the NWRITE command. Only nodes that are within the node range
specified with the NRRANG command are read from the file. Duplicate nodes already in the database will
be overwritten. The file is rewound before and after reading. Reading continues until the end of the file.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>Read Node File
NREFINE, NN1, NN2, NINC, LEVEL, DEPTH, POST, RETAIN
Refines the mesh around specified nodes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NN1, NN2, NINC
Nodes (NN1 to NN2 in increments of NINC) around which the mesh is to be refined. NN2 defaults to
NN1, and NINC defaults to 1. If NN1 = ALL, NN2 and NINC are ignored and all selected nodes are used
1099
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NREFINE
for refinement. If NN1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NN1 (NN2 and NINC are ignored).
LEVEL
Amount of refinement to be done. Specify the value of LEVEL as an integer from 1 to 5, where a value
of 1 provides minimal refinement, and a value of 5 provides maximum refinement (defaults to 1).
DEPTH
Depth of mesh refinement in terms of number of elements outward from the indicated nodes (defaults
to 1).
POST
Type of postprocessing to be done after element splitting, in order to improve element quality:
OFF
No postprocessing will be done.
SMOOTH
Smoothing will be done. Node locations may change.
CLEAN
Smoothing and cleanup will be done. Existing elements may be deleted, and node locations may
change (default).
RETAIN
Flag indicating whether quadrilateral elements must be retained in the refinement of an all-quadrilateral
mesh. (The ANSYS program ignores the RETAIN argument when you are refining anything other than
a quadrilateral mesh.)
ON
The final mesh will be composed entirely of quadrilateral elements, regardless of the element quality
(default).
OFF
The final mesh may include some triangular elements in order to maintain element quality and
provide transitioning.
Notes
NREFINE performs local mesh refinement around the specified nodes. By default, the indicated elements
are split to create new elements with 1/2 the edge length of the original elements (LEVEL = 1).
NREFINE refines all area elements and tetrahedral volume elements that are adjacent to the specified nodes.
Any volume elements that are adjacent to the specified nodes, but are not tetrahedra (for example, hexahedra,
wedges, and pyramids), are not refined.
You cannot use mesh refinement on a solid model that contains initial conditions at nodes [IC], coupled
nodes [CP family of commands], constraint equations [CE family of commands], or boundary conditions or
loads applied directly to any of its nodes or elements. This applies to nodes and elements anywhere in the
model, not just in the region where you want to request mesh refinement. For additional restrictions on
mesh refinement, see Revising Your Model in the Modeling and Meshing Guide.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Refine At>Nodes
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1100
NREFINE
NRLSUM, SIGNIF, Label
Specifies the Naval Research Laboratory (NRL) sum mode combination method.
SOLUTION: Spectrum Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
SIGNIF
Combine only those modes whose significance level exceeds the SIGNIF threshold. For single point,
multipoint, or DDAM response (SPOPT,SPRS, MPRS or DDAM), the significance level of a mode is defined
as the mode coefficient of the mode, divided by the maximum mode coefficient of all modes. Any mode
whose significance level is less than SIGNIF is considered insignificant and is not contributed to the
mode combinations. The higher the SIGNIF threshold, the fewer the number of modes combined.
SIGNIF defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0. (This mode combination
method is not valid for SPOPT,PSD.)
Label
Label identifying the combined mode solution output.
DISP
Displacement solution (default). Displacements, stresses, forces, etc., are available.
VELO
Velocity solution. Velocities, "stress velocities," "force velocities," etc., are available.
ACEL
Acceleration solution. Accelerations, "stress accelerations," "force accelerations," etc., are available.
Notes
This command is also valid in PREP7. This mode combination method is usually used for SPOPT,DDAM.
Product Restrictions
NRLSUM is not allowed in ANSYS Professional.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Mode Combine
1101
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NRLSUM
NROPT, Option, --, Adptky
Specifies the Newton-Raphson options in a static or full transient analysis.
SOLUTION: Nonlinear Options
MP ME ST <> PRN <> <> <> EM <> <> PP <> EME MFS
Option
Option key:
AUTO
Let the program choose the option (default).
FULL
Use full Newton-Raphson.
MODI
Use modified Newton-Raphson.
INIT
Use the previously computed matrix (initial-stiffness).
UNSYM
Use full Newton-Raphson with unsymmetric matrices of elements where the unsymmetric option
exists (see Newton-Raphson Option in the Structural Analysis Guide for more information).
--
Unused field.
Adptky
Adaptive descent key:
ON
Use adaptive descent (default if frictional contact exists; if one of the elements CONTAC12 or CON-
TAC52 is present; or if plasticity exists and one of the elements PIPE20, BEAM23, BEAM24, or PIPE60
is present). Explicit ON is valid only if Option = FULL.
OFF
Do not use adaptive descent (default in all other cases).
Notes
The default values given for this command assume SOLCONTROL,ON (the default). See the description of
SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and SOLCONTROL,OFF.
Indicates the Newton-Raphson option used to solve the nonlinear equations in a static or full transient
analysis. If used in SOLUTION, this command is valid only within the first load step.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1102
NROPT
NROTAT, NODE1, NODE2, NINC
Rotates nodal coordinate systems into the active system.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
NODE1, NODE2, NINC
Rotate nodes from NODE1 to NODE2 (defaults to NODE1) in steps of NINC (defaults to 1). If NODE1 =
ALL, NODE2 and NINC are ignored and all selected nodes [NSEL] are rotated. If NODE1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NODE1 (NODE2 and NINC are ignored).
Command Default
No automatic nodal rotation.
Notes
Rotates nodal coordinate systems into the active coordinate system. Nodal coordinate systems may be
automatically rotated into the active (global or local) coordinate system as follows: Rotations in Cartesian
systems will have nodal x directions rotated parallel to the Cartesian X direction. Rotations in cylindrical,
spherical or toroidal systems will have the nodal x directions rotated parallel to the R direction. Nodes at
(or near) a zero radius location should not be rotated. Nodal coordinate directions may be displayed [/PSYMB].
Nodal forces and constraints will also appear rotated when displayed if the nodal coordinate system is rotated.
For FLOTRAN analyses, nodal coordinate systems should only be rotated parallel to the global Cartesian
system.
ANSYS LS-DYNA (explicit dynamics) does not support the NROTAT command. If you have rotated nodes in
the implicit phase of an implicit-to-explicit sequential solution, you must rotate the nodes back to the
global Cartesian direction before switching from implicit to explicit elements (ETCHG,ITE). Use the EDNROT
command in the explicit run to maintain the same displacement constraints as were used on rotated nodes
in the implicit run.
Note
When the nodal coordinate systems are defined, they remain parallel to the global Cartesian
system unless subsequently rotated.
Previously specified rotations on the specified nodes are overridden.
See the NMODIF, NANG, and NORA commands for other rotation options.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>RotateNode>To Active CS
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Active CS
1103
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NROTAT
NRRANG, NMIN, NMAX, NINC
Specifies the range of nodes to be read from the node file.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NMIN, NMAX, NINC
Node range is defined from NMIN (defaults to 1) to NMAX (defaults to 99999999) in steps of NINC (defaults
to 1).
Notes
Defines the range of nodes to be read [NREAD] from the node file. Also implies an element range since only
elements fully attached to these nodes will be read from the element file.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>Read Node File
NSCALE, INC, NODE1, NODE2, NINC, RX, RY, RZ
Generates a scaled set of nodes from a pattern of nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
INC
Do this scaling operation one time, incrementing all nodes in the given pattern by INC. If INC = 0,
nodes will be redefined at the scaled locations.
NODE1, NODE2, NINC
Scale nodes from pattern of nodes beginning with NODE1 to NODE2 (defaults to NODE1) in steps of
NINC (defaults to 1). If NODE1 = ALL, NODE2 and NINC are ignored and pattern is all selected nodes
[NSEL]. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NODE1 (NODE2 and NINC are ignored).
RX, RY, RZ
Scale factor ratios. Scaling is relative to the origin of the active coordinate system (RR, R, RZ for cylindrical,
RR, R, R for spherical or toroidal). If absolute value of ratio > 1.0, pattern is enlarged. If < 1.0, pattern
is reduced. Ratios default to 1.0 (each).
Notes
Generates a scaled pattern of nodes from a given node pattern. Scaling is done in the active coordinate
system. Nodes in the pattern may have been generated in any coordinate system.
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Nodes>Scale & Copy
Main Menu>Preprocessor>Modeling>Move / Modify>Nodes>Scale & Move
Main Menu>Preprocessor>Modeling>Operate>Scale>Nodes>Scale & Copy
Main Menu>Preprocessor>Modeling>Operate>Scale>Nodes>Scale & Move
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1104
NRRANG
NSEL, Type, Item, Comp, VMIN, VMAX, VINC, KABS
Selects a subset of nodes.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Restore the full set.
NONE
Unselect the full set.
INVE
Invert the current set (selected becomes unselected and vice versa).
STAT
Display the current select status.
The following fields are used only with Type = S, R, A, or U:
Item
Label identifying data. Valid item labels are shown in the table below. Some items also require a com-
ponent label. If Item = PICK (or simply P), graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI). Defaults to NODE.
Comp
Component of the item (if required). Valid component labels are shown in the table below.
VMIN
Minimum value of item range. Ranges are node numbers, set numbers, coordinate values, load values,
or result values as appropriate for the item. A component name (as specified on the CM command) may
also be substituted for VMIN (VMAX and VINC are ignored).
VMAX
Maximum value of item range. VMAX defaults to VMIN for input values. For result values, VMAX defaults
to infinity if VMIN is positive, or to zero if VMIN is negative.
VINC
Value increment within range. Used only with integer ranges (such as for node and set numbers). Defaults
to 1. VINC cannot be negative.
KABS
Absolute value key:
1105
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSEL
0
Check sign of value during selection.
1
Use absolute value during selection (sign ignored).
Command Default
All nodes are selected.
Notes
Selects a subset of nodes. For example, to select a new set of nodes based on node numbers 1 through 7,
use NSEL,S,NODE,,1,7. The subset is used when the ALL label is entered (or implied) on other commands,
such as NLIST,ALL. Only data identified by node number are selected. Data are flagged as selected and un-
selected; no data are actually deleted from the database.
When selecting nodes by results, the full graphics value is used, regardless of whether PowerGraphics is on.
Solution result data consists of two types, 1) nodal degree of freedom--results initially calculated at the
nodes (such as displacement, temperature, pressure, etc.), and 2) element--results initially calculated elsewhere
(such as at an element integration point or thickness location) and then recalculated at the nodes (such as
stresses, strains, etc.). Various element results also depend upon the recalculation method and the selected
results location [AVPRIN, RSYS, FORCE, LAYER and SHELL].
You must have all the nodes (corner and midside nodes) on the external face of the element selected to
use Item = EXT.
This command is valid in any processor.
For Selects based on non-integer numbers (coordinates, results, etc.), items that are within the range
VMIN-Toler and VMAX+Toler are selected. The default tolerance Toler is based on the relative values
of VMIN and VMAX as follows:
If VMIN = VMAX, Toler = 0.005 x VMIN.
If VMIN = VMAX = 0.0, Toler = 1.0E-6.
If VMAX VMIN, Toler = 1.0E-8 x (VMAX-VMIN).
Use the SELTOL command to override this default and specify Toler explicitly.
Table 230 NSEL - Valid Item and Component Labels
Valid Item and Component Labels NSEL, Type,Item,Comp,VMIN,VMAX,VINC,KABS
Valid item and component labels for input values are:
Description Comp Item
Node number. NODE
Nodes on exterior of selected elements (ignore remaining
fields).
EXT
X, Y, or Z location in the active coordinate system. X, Y, Z LOC
THXY, THYZ, or THZX rotation angle. XY, YZ, ZX ANG
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1106
NSEL
Valid Item and Component Labels NSEL, Type,Item,Comp,VMIN,VMAX,VINC,KABS
Valid item and component labels for input values are:
Description Comp Item
Master node number. M
Coupled set number. CP
Constraint equation set number. CE
Any of X, Y, or Z structural displacements. Amplitude only,
if complex.
U D
X, Y, or Z structural displacement. Amplitude only, if
complex.
UX, UY, UZ "
Any of X, Y, or Z structural rotations. Amplitude only, if
complex.
ROT "
X, Y, or Z structural rotation. Amplitude only, if complex. ROTX, ROTY,
ROTZ
"
Temperature. TEMP, TBOT, TE2,
TE3, . . ., TTOP
"
Pressure. PRES "
Electric potential. VOLT "
Magnetic scalar potential. MAG "
Any of X, Y, or Z fluid velocities. V "
X, Y, or Z fluid velocity. VX, VY, VZ "
Any of X, Y, or Z magnetic vector potentials. Amplitude
only, if complex.
A "
X, Y, or Z magnetic vector potential. Amplitude only, if
complex.
AX, AY, AZ "
Current. CURR "
Electromotive force drop. EMF "
Turbulent kinetic energy or energy dissipation (FLOTRAN). ENKE, ENDS "
Any of X, Y, or Z structural forces. Amplitude only, if
complex.
F F
X, Y, or Z structural force. Amplitude only, if complex. FX, FY, FZ "
Any of X, Y, or Z structural moments. Amplitude only, if
complex
M "
X, Y, or Z structural moment. Amplitude only, if complex. MX, MY, MZ "
Heat flow. HEAT, HBOT, HE2,
HE3, . . ., HTOP
"
Fluid flow. FLOW "
Current flow. AMPS "
Magnetic flux. FLUX "
Any of X, Y, or Z magnetic current segment components.
Amplitude only, if complex.
CSG "
1107
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSEL
Valid Item and Component Labels NSEL, Type,Item,Comp,VMIN,VMAX,VINC,KABS
Valid item and component labels for input values are:
Description Comp Item
X, Y, or Z magnetic current segment component. Amp-
litude only, if complex.
CSGX, CSGY, "
if complex. CSGZY,
Electric charge. CHRG "
Electric charge density. CHRGD "
Nodal temperature. TEMP BF
Nodal fluence. FLUE "
Nodal heat generation rate. HGEN "
Any of X, Y, or Z current densities. Amplitude only, if
complex.
JS "
X, Y, or Z current density. Amplitude only, if complex. JSX, JSY, JSZ "
Magnetic virtual displacements flag. MVDI "
Table 231 NSEL - Valid Item and Component Labels for Nodal DOF Result Values
Description Comp Item
X, Y, or Z structural displacement or vector sum. X, Y, Z, SUM U
X, Y, or Z structural rotation or vector sum. X, Y, Z, SUM ROT
Temperature. TEMP
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity or vector sum. X, Y, Z, SUM V
X, Y, or Z magnetic vector potential or vector sum. X, Y, Z, SUM A
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy (FLOTRAN). ENKE
Turbulent energy dissipation (FLOTRAN). ENDS
Table 232 NSEL - Valid Item and Component Labels for Element Result Values
Description Comp Item
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity or equivalent stress. INT, EQV "
Component total strain (EPEL + EPPL + EPCR). X, Y, Z, XY, YZ, XZ EPTO
Principal total strain. 1,2,3 "
Total strain intensity or total equivalent strain. INT, EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1108
NSEL
Description Comp Item
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity or elastic equivalent strain. INT, EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1,2,3 "
Plastic strain intensity or plastic equivalent strain. INT, EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1,2,3 "
Creep strain intensity or creep equivalent strain. INT, EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity or thermal equivalent strain. INT, EQV "
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Contact status. STAT1 CONT
Contact penetration. PENE "
Contact pressure. PRES "
Contact friction stress. SFRIC "
Contact total stress (pressure plus friction). STOT "
Contact sliding distance. SLIDE "
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
Component pressure gradient or vector sum. X, Y, Z, SUM PG
Component electric field or vector sum. X, Y, Z, SUM EF
Component electric flux density or vector sum. X, Y, Z, SUM D
Component magnetic field intensity or vector sum. X, Y, Z, SUM H
Component magnetic flux density or vector sum. X, Y, Z, SUM B
Component electromagnetic forces or vector sum. X, Y, Z, SUM FMAG
Densities used for topological optimization. This applies
to nodes attached to the following elements: PLANE82,
SOLID92, SOLID95.
TOPO
1109
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSEL
1. For more information on the meaning of contact status and its possible values, see Reviewing Results
in POST1 in the Contact Technology Guide.
Table 233 NSEL - Valid Item and Component Labels for FLOTRAN Nodal Result Values
Description Comp Item
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Turbulent viscosity coefficient. CMUV
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Delete Elements
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
Utility Menu>Select>Entities
NSLA, Type, NKEY
Selects those nodes associated with the selected areas.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of node select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1110
NSLA
NKEY
Specifies whether only interior area nodes are to be selected:
0
Select only nodes interior to selected areas.
1
Select all nodes (interior to area, interior to lines, and at keypoints) associated with the selected
areas.
Notes
Valid only if the nodes were generated by an area meshing operation [AMESH, VMESH] on a solid model
that contains the selected areas.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
NSLE, Type, NodeType, Num
Selects those nodes attached to the selected elements.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of node select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
NodeType
Label identifying type of nodes to consider when selecting:
ALL
Select all nodes of the selected elements (default).
ACTIVE
Select only the active nodes. An active node is a node that contributes DOFs to the model.
INACTIVE
Select only inactive nodes (such as orientation or radiation).
CORNER
Select only corner nodes.
1111
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSLE
MID
Select only midside nodes.
POS
Select nodes in position Num.
FACE
Select nodes on face Num.
Num
Position or face number for Nodes = POS or FACE.
Notes
NSLE selects NodeType nodes attached to the currently-selected set of elements. Only nodes on elements
in the currently-selected element set can be selected.
Note
When using degenerate hexahedral elements, NSLE, U,CORNER and NSLE,S,MID will not select
the same set of nodes because some nodes appear as both corner and midside nodes.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>CMS>CMS Superelements>By Picking
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
Utility Menu>Select>Entities
NSLK, Type
Selects those nodes associated with the selected keypoints.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of node select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1112
NSLK
Notes
Valid only if the nodes were generated by a keypoint meshing operation [KMESH, LMESH, AMESH, VMESH]
on a solid model that contains the selected keypoints.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
NSLL, Type, NKEY
Selects those nodes associated with the selected lines.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of node select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
NKEY
Specifies whether only interior line nodes are to be selected:
0
Select only nodes interior to selected lines.
1
Select all nodes (interior to line and at keypoints) associated with the selected lines.
Notes
Valid only if the nodes were generated by a line meshing operation [LMESH, AMESH, VMESH] on a solid
model that contains the associated lines.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
1113
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSLL
NSLV, Type, NKEY
Selects those nodes associated with the selected volumes.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of node select:
S
Select a new set (default).
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
NKEY
Specifies whether only interior volume nodes are to be selected:
0
Select only nodes interior to selected volumes.
1
Select all nodes (interior to volume, interior to areas, interior to lines, and at keypoints) associated
with the selected volumes.
Notes
Valid only if the nodes were generated by a volume meshing operation [VMESH] on a solid model that
contains the selected volumes.
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
NSMOOTH, NPASS
Smooths selected nodes among selected elements.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NPASS
Number of smoothing passes. Defaults to 3.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1114
NSLV
Notes
Repositions each selected node at the average position of its immediate neighbors on the selected elements.
The node positions converge after some number of smoothing passes. For some initial conditions, NPASS
may need to be much larger than 3. If the boundary of a mesh is to be undisturbed (usually desirable), the
boundary nodes should be unselected before issuing NSMOOTH.
Menu Paths
This command cannot be accessed from a menu.
NSOL, NVAR, NODE, Item, Comp, Name, SECTOR
Specifies nodal data to be stored from the results file.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVAR
Arbitrary reference number or name assigned to this variable. Variable numbers can be 2 to NV (NUMVAR)
while the name can be an eight byte character string. Overwrites any existing results for this variable.
NODE
Node for which data are to be stored.
Item
Label identifying the item. Valid item labels are shown in the table below. Some items also require a
component label.
Comp
Component of the item (if required). Valid component labels are shown in the table below.
Name
Thirty-two character name identifying the item on printouts and displays. Defaults to a label formed by
concatenating the first four characters of the Item and Comp labels.
SECTOR
For a cyclic symmetry solution, the sector number for which the results from NODE are to be stored.
Notes
Stores nodal degree of freedom and solution results in a variable. For more information, see Data Interpreted
in the Nodal Coordinate System in the Modeling and Meshing Guide.
Table 234 NSOL - Valid Item and Component Labels
Valid Item and Component Labels NSOL, NVAR,NODE,Item,Comp,Name
Valid item and component labels for nodal degree of freedom results are:
Description Comp Item
X, Y, or Z structural displacement. X, Y, Z U
X, Y, or Z structural rotation. X, Y, Z ROT
Temperature. TEMP[1]
1115
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSOL
Valid Item and Component Labels NSOL, NVAR,NODE,Item,Comp,Name
Valid item and component labels for nodal degree of freedom results are:
Description Comp Item
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity in a fluid analysis. X, Y, Z V
X, Y, or Z magnetic vector potential in an electromagnetic
analysis.
X, Y, Z A
X, Y, or Z velocity in a structural transient dynamic analysis
(ANTYPE,TRANS).
X, Y, Z VEL
X, Y, or Z acceleration in a structural transient dynamic
analysis (ANTYPE,TRANS).
X, Y, Z ACC
X, Y, or Z rotational velocity in a structural transient dy-
namic analysis (ANTYPE,TRANS).
X, Y, Z OMG
X, Y, or Z rotational acceleration in a structural transient
dynamic analysis (ANTYPE,TRANS).
X, Y, Z DMG
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy. ENKE
Turbulent energy dissipation. ENDS
Table 235 NSOL - Valid Item and Component Labels for FLOTRAN Nodal Results
Description Comp Item
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Turbulent viscosity coefficient. CMUV
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1116
NSOL
Description Comp Item
Radiation heat flux. RDFL
Table 236 NSOL - Valid Item and Component Labels for ANSYS LS-DYNA Nodal Results
Description Comp Item
X, Y, or Z nodal displacements. X, Y, Z U
X, Y, or Z nodal rotations. X, Y, Z ROT
X, Y, or Z nodal velocities. X, Y, Z V
X, Y, or Z nodal accelerations. X, Y, Z A
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels TBOT, TE2, TE3, . . ., TTOP
instead of TEMP.
Menu Paths
Main Menu>Drop Test>Time History>Graph Variables
Main Menu>Drop Test>Time History>List Variables
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
NSORT, Item, Comp, ORDER, KABS, NUMB, SEL
Sorts nodal data.
POST1: Results
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Label identifying the item to be sorted on. Valid item labels are shown in the table below. Some items
also require a component label.
Comp
Component of the item (if required). Valid component labels are shown in the table below.
ORDER
Order of sort operation:
0
Sort into descending order.
1
Sort into ascending order.
KABS
Absolute value key:
0
Sort according to real value.
1
Sort according to absolute value.
1117
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSORT
NUMB
Number of nodal data records to be sorted in ascending or descending order (ORDER) before sort is
stopped (remainder will be in unsorted sequence) (defaults to all nodes).
SEL
Allows selection of nodes in the sorted field.
(blank)
No selection (default).
SELECT
Select the nodes in the sorted list.
Command Default
Use ascending node number order.
Notes
Values are in the active coordinate system [CSYS for input data or RSYS for results data]. Various element
results also depend upon the recalculation method and the selected results location [AVPRIN, RSYS, SHELL,
ESEL, and NSEL]. If simultaneous load cases are stored, the last sorted sequence formed from any load case
applies to all load cases. Use NUSORT to restore the original order. This command is not valid with Power-
Graphics.
Table 237 NSORT - Valid Item and Component Labels
Valid Item and Component Labels NSORT, Item,Comp,ORDER,KABS,NUMB,SEL
Valid item and component labels for input values are:
Description Comp Item
X, Y, or Z location. X, Y, Z LOC
THXY, THYZ, or THZX rotation angle. XY, YZ, ZX ANG
Table 238 NSORT - Valid Item and Component Labels for Nodal DOF Result Values
Description Comp Item
X, Y, or Z structural displacement or vector sum. X, Y, Z, SUM U
X, Y, or Z structural rotation or vector sum. X, Y, Z, SUM ROT
Temperature (includes TEMP, TBOT, TE2, TE3, . . ., TTOP
values).
TEMP
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity or vector sum. X, Y, Z, SUM V
X, Y, or Z magnetic vector potential or vector sum. X, Y, Z, SUM A
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy (FLOTRAN). ENKE
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1118
NSORT
Description Comp Item
Turbulent energy dissipation (FLOTRAN). ENDS
Table 239 NSORT - Valid Item and Component Labels for Element Result Values
Description Comp Item
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2,3 "
Stress intensity or equivalent stress. INT, EQV "
Component total strain (EPEL + EPPL + EPCR). X, Y, Z, XY, YZ, XZ EPTO
Principal total strain. 1, 2, 3 "
Total strain intensity or total equivalent strain. INT, EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity or elastic equivalent strain. INT, EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity or plastic equivalent strain. INT, EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity or creep equivalent strain. INT, EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity or thermal equivalent strain. INT, EQV "
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Contact status. STAT1 CONT
Contact penetration. PENE "
Contact pressure. PRES "
Contact friction stress. SFRIC "
Contact total stress (pressure plus friction). STOT "
Contact sliding distance. SLIDE "
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
Component pressure gradient or vector sum. X, Y, Z, SUM PG
1119
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSORT
Description Comp Item
Component electric field or vector sum. X, Y, Z, SUM EF
Component electric flux density or vector sum. X, Y, Z, SUM D
Component magnetic field intensity or vector sum. X, Y, Z, SUM H
Component magnetic flux density or vector sum. X, Y, Z, SUM B
Component electromagnetic forces or vector sum. X, Y, Z, SUM FMAG
Densities used for topological optimization. This applies
to nodes attached to the following types of elements:
PLANE82, SOLID92, SOLID95, and PLANE183.
TOPO
1. For more information on the meaning of contact status and its possible values, see Reviewing Results
in POST1 in the Contact Technology Guide.
Table 240 NSORT - Valid Item and Component Labels for FLOTRAN Nodal Result Values
Description Comp Item
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
Menu Paths
Main Menu>General Postproc>List Results>Sorted Listing>Sort Nodes
Utility Menu>Parameters>Get Scalar Data
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1120
NSORT
NSTORE, TINC
Defines which time points are to be stored.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
TINC
Store data associated with every TINC time (or frequency) point(s), within the previously defined range
of TMIN to TMAX [TIMERANGE]. (Defaults to 1)
Command Default
Store every point.
Notes
Defines which time (or frequency) points within the range are to be stored.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Data
NSUBST, NSBSTP, NSBMX, NSBMN, Carry
Specifies the number of substeps to be taken this load step.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
NSBSTP
Number of substeps to be used for this load step (i.e., the time step size or frequency increment). If
automatic time stepping is used [AUTOTS], NSBSTP defines the size of the first substep. If SOLCON-
TROL,ON and contact elements TARGE169, TARGE170, CONTA171, CONTA172, CONTA173, CONTA174,
CONTA175, CONTA176, or CONTA177 are used, defaults to 1 or 20 substeps, depending on the physics
of the problem. If SOLCONTROL,ON and none of these contact elements are used, defaults to 1 substep.
If SOLCONTROL,OFF, defaults to the previously specified value (or 1, if there is no previously specified
value).
NSBMX
Maximum number of substeps to be taken (i.e., the minimum time step size) if automatic time stepping
is used. If SOLCONTROL,ON, ANSYS determines the default depending on the physics of the problem.
If SOLCONTROL,OFF, defaults to the previously specified value (or NSBSTP, if there is no previously
specified value).
NSBMN
Minimum number of substeps to be taken (i.e., the maximum time step size) if automatic time stepping
is used. If SOLCONTROL,ON, ANSYS determines the default depending on the physics of the problem.
If SOLCONTROL,OFF, defaults to the previously specified value (or 1, if there is no previously specified
value).
Carry
Time step carryover key (ANSYS determines the default depending on the physics of the problem):
1121
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSUBST
OFF
Use NSBSTP to define time step at start of each load step.
ON
Use final time step from previous load step as the starting time step (if automatic time stepping is
used).
If SOLCONTROL,ON, ANSYS determines the default depending on the physics of the problem. If SOL-
CONTROL,OFF, defaults to OFF.
Notes
See DELTIM for an alternative input. It is recommended that all fields of this command be specified for
solution efficiency and robustness.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Basic
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Freq and Substps
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time and Substps
Main Menu>Solution>Analysis Type>Sol'n Controls>Basic
Main Menu>Solution>Load Step Opts>Time/Frequenc>Freq and Substps
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time and Substps
NSVR, ITYPE, NSTV
Defines the number of variables for user-programmable element options.
PREP7: Element Type
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
ITYPE
Element type number as defined on the ET command.
NSTV
Number of extra state variables to save (must be no more than 840).
Command Default
No extra variables are saved.
Notes
Defines the number of extra variables that need to be saved for user-programmable (system-dependent)
element options, e.g., material laws through user subroutine USERPL. ITYPE must first be defined with the
ET command.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1122
NSVR
NSYM, Ncomp, INC, NODE1, NODE2, NINC
Generates a reflected set of nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Ncomp
Symmetry key:
X
X (or R) symmetry (default).
Y
Y (or ) symmetry.
Z
Z (or ) symmetry.
INC
Increment all nodes in the given pattern by INC to form the reflected node pattern.
NODE1, NODE2, NINC
Reflect nodes from pattern beginning with NODE1 to NODE2 (defaults to NODE1) in steps of NINC (de-
faults to 1). If NODE1 = ALL, NODE2 and NINC are ignored and pattern is all selected nodes [NSEL]. If
NODE1 = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI). A component name may also be substituted for NODE1 (NODE2 and NINC are ignored).
Notes
Generates nodes from a given node pattern by a symmetry reflection. Reflection is done in the active co-
ordinate system by changing a particular coordinate sign. Nodes in the pattern may have been generated
in any coordinate system. Nodal rotation angles are not reflected.
Symmetry reflection may be used with any node pattern, in any coordinate system, as many times as desired.
Reflection is accomplished by a coordinate sign change (in the active coordinate system). For example, an
X-reflection in a Cartesian coordinate system generates additional nodes from a given pattern, with a node
increment added to each node number, and an X coordinate sign change. An R-reflection in a cylindrical
coordinate system gives a reflected "radial" location by changing the "equivalent" Cartesian (i.e., the Cartesian
system with the same origin as the active cylindrical system) X and Y coordinate signs. An R-reflection in a
spherical coordinate system gives a reflected "radial" location by changing the equivalent Cartesian X, Y,
and Z coordinate location signs. Nodal coordinate system rotation angles are not reflected.
Menu Paths
Main Menu>Preprocessor>Modeling>Reflect>Nodes
1123
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NSYM
/NUMBER, NKEY
Specifies whether numbers, colors, or both are used for displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NKEY
Numbering style:
0
Color (terminal dependent) the numbered items and show numbers.
1
Color the numbered items. Do not show the numbers.
2
Show the numbers. Do not color the items.
-1
Do not color the items or show the numbers. For contour plots, the resulting display will vary (see
below).
Notes
Specifies whether numbers, colors, or both are used for numbering displays [/PNUM] of nodes, elements,
keypoints, lines, areas, and volumes.
Shading is also available for terminals configured with more than 4 color planes [/SHOW]. Color automatically
appears for certain items and may be manually controlled (off or on) for other items.
When you suppress color (NKEY = -1) your contour plots will produce different results, depending on your
graphics equipment. For non-3-D devices (X11, Win32, etc.) your contour plot will be white (no color). For
3-D devices, such as OpenGL, the resulting display will be in color.
The following items are automatically given discrete colors: Boundary condition symbols [/PBC], curves on
graph displays, and distorted geometry on postprocessing displays. Contour lines in postprocessing displays
are automatically colored based upon a continuous, rather than a discrete, spectrum so that red is associated
with the highest contour value. On terminals with raster capability [/SHOW], the area between contour lines
is filled with the color of the higher contour.
Explicit entity colors or the discrete color mapping may be changed with the /COLOR command.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Numbering
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1124
/NUMBER
NUMCMP, Label
Compresses the numbering of defined items.
PREP7: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Items to be compressed:
NODE
Node numbers
ELEM
Element numbers
KP
Keypoint numbers
LINE
Line numbers
AREA
Area numbers
VOLU
Volume numbers
MAT
Material numbers
TYPE
Element type numbers
REAL
Real constant numbers
CP
Coupled set numbers
SECN
Section numbers
CE
Constraint equation numbers
ALL
All item numbers
Notes
The NUMCMP command effectively compresses out unused item numbers by renumbering all the items,
beginning with one and continuing throughout the model. The renumbering order follows the initial item
numbering order (that is, compression lowers the maximum number by "sliding" numbers down to take
advantage of unused or skipped numbers). All defined items are renumbered, regardless of whether or not
they are actually used or selected. Applicable related items are also checked for renumbering as described
for the merge operation [NUMMRG].
Compressing material numbers [NUMCMP,ALL or NUMCMP,MAT] does not update the material number
referenced by either of the following:
1125
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NUMCMP
A temperature-dependent convection or surface-to-surface radiation load [SF, SFE, SFL, SFA]
Real constants for multi-material elements (such as PIPE17 and SOLID65)
Compression is usually not required unless memory space is limited and there are large gaps in the numbering
sequence.
You cannot access this command for models that have been imported from IGES files using the FACETED
translator (see the IOPTN command).
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Compress Numbers
NUMEXP, NUM, BEGRNG, ENDRNG, Elcalc
Specifies solutions to be expanded from reduced analyses.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NUM
The number of solutions to expand. This value is required.
Num
Number of solutions to expand.
ALL
Expand all substeps between BEGRNG and ENDRNG (provided that ENDRNG > 0). If BEGRNG and
ENDRNG have no specified values, this option expands all substeps of all load steps.
BEGRNG, ENDRNG
Beginning and ending time (or frequency) range for expanded solutions. The default is 0 for both values.
Elcalc
The element-calculation key:
YES
Calculate element results, nodal loads, and reaction loads. This value is the default.
NO
Do not calculate these items.
Command Defaults
Issuing this command with no arguments is invalid. You must specify the number of solutions, or all solutions,
to expand (NUM). The default value for both the beginning (BEGRNG) and ending (ENDRNG) time or frequency
is 0. The default behavior of the command is to calculate element results, nodal loads, and reaction loads
(Elcalc = YES).
Notes
Specifies a range of solutions to be expanded from analyses that use reduced or mode superposition
methods (ANTYPE,HARMIC, TRANS, or SUBSTR).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1126
NUMEXP
For ANTYPE,TRANS, NUM, evenly spaced solutions are expanded between time BEGRNG and time ENDRNG.
For ANTYPE,HARMIC, NUM, evenly spaced solutions are expanded between frequency BEGRNG and frequency
ENDRNG.
The first expansion in all cases is done at the first point beyond BEGRNG (that is, at BEGRNG + (ENDRNG -
BEGRNG) / NUM)).
For a single expansion of a solution, or for multiple expansions when the solutions are not evenly spaced
(such as in a mode superposition harmonic analysis with the cluster option), ANSYS recommends issuing
one or more EXPSOL commands.
The NUMEXP command is invalid in these cases:
In a substructing analysis (ANTYPE,SUBST) when a factorized matrix file (the .LN22 file generated by
the sparse solver) does not exist, causing ANSYS to employ the full-resolve method.
If the full-resolve option is selected using the SEOPT command.
In both situations, use the EXPSOL command to perform a single expansion for each solution desired.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>Range of Solu's
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>Range of Solu's
NUMMRG, Label, TOLER, GTOLER, Action, Switch
Merges coincident or equivalently defined items.
PREP7: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Items to be merged:
NODE
Nodes
ELEM
Elements
KP
Keypoints (will also merge lines, areas, and volumes)
MAT
Materials
TYPE
Element types
REAL
Real constants
1127
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NUMMRG
CP
Coupled sets
CE
Constraint equations
ALL
All items
TOLER
Range of coincidence. For Label = NODE and KP, defaults to 1.0E-4 (based on maximum Cartesian co-
ordinate difference between nodes or keypoints). For Label = MAT, REAL, and CE, defaults to 1.0E-7
(based on difference of the values normalized by the values). Only items within range are merged. (For
keypoints attached to lines, further restrictions apply. See the GTOLER field and Merging Solid Model
Entities below.)
GTOLER
Global solid model tolerance -- used only when merging keypoints attached to lines. If specified,
GTOLER will override the internal relative solid model tolerance. See Merging Solid Model Entities below.
Action
Specifies whether to merge or select coincident items.
SELE
Select coincident items but do not merge. Action = SELE is only valid for Label = NODE.
(Blank)
Merge the coincident items (default).
Switch
Specifies whether the lowest or highest numbered coincident item is retained after the merging operation.
This option does not apply to keypoints; i.e., for Label = KP, the lowest numbered keypoint is retained
regardless of the Switch setting.
LOW
Retain the lowest numbered coincident item after the merging operation (default).
HIGH
Retain the highest numbered coincident item after the merging operation.
Notes
After issuing the command, the area and volume sizes (ASUM and VSUM) may give slightly different results.
In order to obtain the same results as before, use /FACET, /NORMAL, and ASUM / VSUM.
The merge operation is useful for tying separate, but coincident, parts of a model together. If not all items
are to be checked for merging, use the select commands (NSEL, ESEL, etc.) to select items. Only selected
items are included in the merge operation for nodes, keypoints, and elements.
By default, the merge operation retains the lowest numbered coincident item. Higher numbered coincident
items are deleted. Set Switch to HIGH to retain the highest numbered coincident item after the merging
operation. Applicable related items are also checked for deleted item numbers and if found, are replaced
with the retained item number. For example, if nodes are merged, element connectivities (except superele-
ments), mesh item range associativity, coupled degrees of freedom, constraint equations, master degrees
of freedom, gap conditions, degree of freedom constraints, nodal force loads, nodal surface loads, and
nodal body force loads are checked. Merging material numbers [NUMMRG,ALL or NUMMRG,MAT] does not
update the material number referenced:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1128
NUMMRG
By temperature-dependent film coefficients as part of convection load or a temperature-dependent
emissivity as part of a surface-to-surface radiation load [SF, SFE, SFL, SFA]
By real constants for multi-material elements (such as PIPE17 and SOLID65)
If a unique load is defined among merged nodes, the value is kept and applied to the retained node. If loads
are not unique (not recommended), only the value on the lowest node (or highest if Switch = HIGH) will
be kept, except for "force" loads for which the values will be summed if they are not defined using tabular
boundary conditions.
Note
The unused nodes (not recommended) in elements, couplings, constraint equations, etc. may
become active after the merge operation.
The Action field provides the option of visualizing the coincident items before the merging
operation.
Caution
When merging entities in a model that has already been meshed, the order in which you issue
multiple NUMMRG commands is significant. If you want to merge two adjacent meshed regions
that have coincident nodes and keypoints, always merge nodes [NUMMRG,NODE] before merging
keypoints [NUMMRG,KP]. Merging keypoints before nodes can result in some of the nodes be-
coming "orphaned"; that is, the nodes lose their association with the solid model. Orphaned nodes
can cause certain operations (such as boundary condition transfers, surface load transfers, and
so on) to fail. However, using NUMMRG should be avoided if at all possible, as the procedure
outlined above may even cause meshing failure, especially after multiple merging and meshing
operations.
After a NUMMRG,NODE, is issued, some nodes may be attached to more than one solid entity. As a result,
subsequent attempts to transfer solid model loads to the elements may not be successful. Issue NUMMRG,KP
to correct this problem. Do NOT issue VCLEAR before issuing NUMMRG,KP.
For NUMMRG,ELEM, elements must be identical in all aspects, including the direction of the element co-
ordinate system.
For certain solid and shell elements (181, 185, 190, etc) ANSYS will interpret coincident faces as internal and
eliminate them. To prevent this from occurring, shrink the entities by a very small factor to delineate coin-
cident items (/SHRINK, 0.0001) and no internal nodes, lines, areas or elements will be eliminated.
When working with solid models, you may have better success with the gluing operations (AGLUE, LGLUE,
VGLUE). Please read the following information when attempting to merge solid model entities.
Gluing Operations vs. Merging Operations
Adjacent, touching regions can be joined by gluing them (AGLUE, LGLUE, VGLUE) or by merging coincident
keypoints (NUMMRG,KP, which also causes merging of identical lines, areas, and volumes). In many situations,
either approach will work just fine. Some factors, however, may lead to a preference for one method over
the other.
Geometric Configuration
1129
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NUMMRG
Gluing is possible regardless of the initial alignment or offset of the input entities. Keypoint merging is
possible only if each keypoint on one side of the face to be joined is matched by a coincident keypoint on
the other side. This is commonly the case after a symmetry reflection (ARSYM or VSYMM) or a copy (AGEN
or VGEN), especially for a model built entirely in ANSYS rather than imported from a CAD system. When the
geometry is extremely precise, and the configuration is correct for keypoint merging, NUMMRG is more ef-
ficient and robust than AGLUE or VGLUE.
Model Accuracy
Gluing, like all ANSYS boolean operations, requires that the input entities meet the current boolean tolerance
(BTOL). Otherwise, AGLUE or VGLUE may fail. In such cases, relaxing the tolerance may allow the glue to
complete. An advantage of gluing is that it is unlikely to degrade the accuracy of a geometric model. Key-
point merging can operate on almost any combination of entities (although you may have to override the
default tolerances on NUMMRG). However, it can also introduce or increase accuracy flaws, making later
boolean operations less likely to succeed. If the input tolerances are too large, NUMMRG can collapse out
small lines, areas, or volumes you intended to keep, possibly rendering the model unusable.
Mesh Status
Gluing, like all ANSYS boolean operations, requires that the input entities be unmeshed. Keypoint merging
is effective for meshed models under the right conditions. More information on keypoint merging follows.
Merging Solid Model Entities:
When merging solid model entities (Label = KP or ALL), keypoint locations are used as the basis for merging.
Once keypoints are merged, any higher order solid model entities (lines, areas, and volumes), regardless of
their select status or attachment to the merged keypoints, are considered for merging.
Keypoints that are attached to lines will be merged only if:
X, Y, and Z are each less than TOLER
where,
X is the X component of the distance between keypoints,
Y is the Y component of the distance between keypoints, and
Z is the Z component of the distance between keypoints;
and
X Y Z
2 2 2
+ +
is less than 1E-5 times the length of the longest line attached to those keypoints
(internal relative solid model tolerance), or
X Y Z
2 2 2
+ +
is less than GTOLER (global solid model
tolerance) if specified.
The TOLER field is a consideration tolerance. If a keypoint is within TOLER of another keypoint, then those
two keypoints are candidates to be merged. If, when "moving" the higher numbered keypoint, the distance
exceeds the internal relative solid model tolerance, or the global solid model tolerance (GTOLER) if specified,
the keypoints will not be merged. Lines, areas, and volumes are considered for merging in a similar manner.
The internal relative solid model tolerance should be overridden by the global solid model tolerance
(GTOLER) only when absolutely necessary. GTOLER is an absolute tolerance; if specified, relative lengths of
lines in the model will no longer be considered in the merge operation. If GTOLER is too large, you can
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1130
NUMMRG
"merge-out" portions of your model accidently, effectively defeaturing the model. If GTOLER is used, it is
recommended that NUMMRG be preceded by saving the database (since undesired merges of solid model
entities could occur).
Note
Use of the NUMMRG command does not cause changes to a model's geometry; only the topology
is affected.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Merge Nodes
Main Menu>Preprocessor>Modeling>Delete>Pre-tens Elemnts
Main Menu>Preprocessor>Numbering Ctrls>Merge Items
NUMOFF, Label, VALUE
Adds a number offset to defined items.
PREP7: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Apply offset number to one of the following sets of items:
NODE
Nodes
ELEM
Elements
KP
Keypoints
LINE
Lines
AREA
Areas
VOLU
Volumes
MAT
Materials
TYPE
Element types
REAL
Real constants
CP
Coupled sets
1131
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NUMOFF
SECN
Section numbers
CE
Constraint equations
CSYS
Coordinate systems
VALUE
Offset number value (cannot be negative).
Notes
Useful for offsetting current model data to prevent overlap if another model is read in. CDWRITE automat-
ically writes the appropriate NUMOFF commands followed by the model data to File.CDB. Therefore,
when the file is read, any model already existing in the database is offset before the model data on the file
is read.
Offsetting material numbers with this command [NUMOFF,MAT] does not update the material number ref-
erenced by either of the following:
A temperature-dependent convection or surface-to-surface radiation load [SF, SFE, SFL, SFA]
Real constants for multi-material elements (such as PIPE17, and SOLID65).
Therefore, a mismatch may exist between the material definitions and the material numbers referenced.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Add Num Offset
NUMSTR, Label, VALUE
Establishes starting numbers for automatically numbered items.
PREP7: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Apply starting number to one of the following sets of items:
NODE
Node numbers. Value defaults (and is continually reset) to 1 + maximum node number in model.
Cannot be reset lower.
ELEM
Element numbers. Value defaults (and is continually reset) to 1 + maximum element number in
model. Cannot be reset lower.
KP
Keypoint numbers. Value defaults to 1. Only undefined numbers are used. Existing keypoints are
not overwritten.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1132
NUMSTR
LINE
Line numbers. Value defaults to 1. Only undefined numbers are used. Existing lines are not over-
written.
AREA
Area numbers. Value defaults to 1. Only undefined numbers are used. Existing areas are not over-
written.
VOLU
Volume numbers. Value defaults to 1. Only undefined numbers are used. Existing volumes are not
overwritten.
DEFA
Default. Returns all starting numbers to their default values.
VALUE
Starting number value.
Notes
Establishes starting numbers for various items that may have numbers automatically assigned (such as element
numbers with the EGEN command, and node and solid model entity numbers with the mesh [AMESH,
VMESH, etc.] commands). Use NUMSTR,STAT to display settings. Use NUMSTR,DEFA to reset all specifications
back to defaults. Defaults may be lowered by deleting and compressing items (i.e., NDELE and NUMCMP,NODE
for nodes, etc.).
Note
A mesh clear operation (VCLEAR, ACLEAR, LCLEAR, and KCLEAR) automatically sets starting
node and element numbers to the highest unused numbers. If a specific starting node or element
number is desired, issue NUMSTR after the clear operation.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Reset Start Num
Main Menu>Preprocessor>Numbering Ctrls>Set Start Number
Main Menu>Preprocessor>Numbering Ctrls>Start Num Status
NUMVAR, NV
Specifies the number of variables allowed in POST26.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV
Allow storage for NV variables. 200 maximum are allowed. Defaults to 10 (except for an explicit dynamics
analysis, which defaults to 30). TIME (variable 1) should also be included in this number.
Command Default
Allow storage for 10 variables (30 variables for an explicit dynamics analysis).
1133
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NUMVAR
Notes
Specifies the number of variables allowed for data read from the results file and for data resulting from an
operation (if any). For efficiency, NV should not be larger than necessary. NV cannot be changed after data
storage begins.
Menu Paths
Main Menu>TimeHist Postpro>Settings>File
NUSORT
Restores original order for nodal data.
POST1: Results
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This command restores the nodal data to its original order (sorted in ascending node number sequence)
after an NSORT command. Changing the selected nodal set [NSEL] also restores the original nodal order.
Menu Paths
Main Menu>General Postproc>List Results>Sorted Listing>Unsort Nodes
NWPAVE, N1, N2, N3, N4, N5, N6, N7, N8, N9
Moves the working plane origin to the average location of nodes.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
N1, N2, N3, . . . , N9
Nodes used in calculation of the average. At least one must be defined. If N1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI).
Notes
Averaging is based on the active coordinate system.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Offset WP to>Nodes
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1134
NUSORT
NWPLAN, WN, NORIG, NXAX, NPLAN
Defines the working plane using three nodes.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number whose viewing direction will be modified to be normal to the working plane (defaults
to 1). If WN is a negative value, the viewing direction will not be modified. If fewer than three points are
used, the viewing direction of window WN will be used instead to define the normal to the working
plane.
NORIG
Node number defining the origin of the working plane coordinate system. If NORIG = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI).
NXAX
Node number defining the x-axis orientation (defaults to the x-axis being parallel to the global X-axis;
or if the normal to the working plane is parallel to the global X-axis, then defaults to being parallel to
the global Y-axis).
NPLAN
Node number defining the working plane (the normal defaults to the present display view [/VIEW] of
window WN).
Notes
Defines a working plane to assist in picking operations using three nodes as an alternate to the WPLANE
command. The three nodes also define the working plane coordinate system. A minimum of one node (at
the working plane origin) is required. Immediate mode may also be active. See the WPSTYL command to
set the style of the working plane display.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>Surface Operations>Create Surface>Sphere>At Node
Utility Menu>WorkPlane>Align WP with>Nodes
NWRITE, Fname, Ext, --, KAPPND
Writes nodes to a file.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
1135
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
NWRITE
Ext
Filename extension (8 character maximum).
The extension defaults to NODE if Fname is blank.
--
Unused field.
KAPPND
Append key:
0
Rewind file before the write operation.
1
Append data to the end of the existing file.
Notes
Writes selected nodes [NSEL] to a file. The write operation is not necessary in a standard ANSYS run but is
provided as a convenience to users wanting a coded node file. Data are written in a coded format. The
format used is (I8, 6G20.13) to write out NODE,X,Y,Z,THXY,THYZ,THZX. If the last number is zero (i.e., THZX
= 0), or the last set of numbers are zero, they are not written but are left blank. Therefore, you must use a
formatted read to process this file. Coordinate values are in the global Cartesian system.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Nodes>Write Node File
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1136
NWRITE
O Commands
OCDATA, VAL1, VAL2, VAL3, VAL4, VAL5, VAL6
Defines an ocean environment using non-table data.
PREP7: Ocean
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1, VAL2, VAL3, . . . , VAL6
Values describing the basic ocean environment or a wave condition.
Notes
The OCDATA specifies non-table data that defines the ocean environment, such as the ocean section number,
the depth of the ocean to the mud line, the ratio of added mass over added mass for a circular cross section,
or the wave theory to apply. The terms VAL1, VAL2, etc. are specialized according to the input set required
for the given ocean environment.
ANSYS interprets the data input via the OCDATA command within the context of the most recently issued
OCTYPE command.
You must input values in the order indicated:
Ocean Data Type: BASIC (OCTYPE,,BASIC)
VAL1
DEPTH -- The depth of the ocean to the mud line. This value is required.
VAL2
MATOC -- The material number of the ocean. This value is required.
VAL3
KFLOOD -- The inside-outside-fluid interaction key, as follows:
0 -- The pressures inside and outside of the pipe are independent. This is the default behavior.
1 -- The pressure inside of the pipe is set to equal the pressure outside of the pipe.
VAL4
Ci -- The ratio of added mass of the external fluid over added mass for a circular cross section. This
value defaults to 1.0 if Ci = 0.0 or no value is specified. If this value should be 0.0, enter a negative
number.
VAL5
Cb -- The ratio of buoyancy force used over buoyancy force based on outside diameter and water
density. This value defaults to 1.0 if Cb = 0.0 or no value is specified. If this value should be 0.0, enter a
negative number.
Ocean Data Type: WAVE (OCTYPE,,WAVE)
1137
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VAL1
KWAVE -- The wave theory to apply to the current profile, as follows:
0 -- Small Amplitude Airy Wave Theory without modifications. This value is the default.
1 -- Small Amplitude Wave Theory with empirical modification of depth decay function.
2 -- Stokes Fifth Order Wave Theory.
3 -- Stream Function Wave Theory.
10 -- API for computing particle velocities and accelerations due to waves and current. (For more
information, see the description of the userPartVelAcc subroutine in the Programmer's Manual
for ANSYS.)
VAL2
THETA -- The wave direction .
VAL3
WAVELOC -- The wave location, as follows:
0 -- Waves act on elements at their actual locations. This is the default behavior.
1 -- Elements are assumed to be at wave peak.
2 -- Upward vertical wave velocity acts on elements.
3 -- Downward vertical wave velocity acts on elements.
4 -- Elements are assumed to be at wave trough.
VAL4
KCRC -- The wave-current interaction key.
Adjustments to the current profile are available via the KCRC constant of the water motion table (illus-
trated in Figure 4: Velocity Profiles for Wave-Current Interaction(p. 1139)). Typically, these options are used
only when the wave amplitude is large relative to the water depth, such that significant wave-current
interaction exists. Valid values are:
0 -- Use the current profile (as input) for wave locations below the mean water level, and the top
current profile value for wave locations above the mean water level. This is the default behavior.
1 -- "Stretch" (or compress) the current profile to the top of the wave.
2 -- Same as KCRC = 1, but also adjusts the current profile horizontally such that total flow continuity
is maintained with the input profile. All wave directions (j) must be identical.
VAL5
KMF -- The MacCamy-Fuchs adjustment key, typically used only for larger-diameter pipes in relatively
shallow water. Valid values are:
0 -- Do not apply the adjustment. This is the default behavior.
1 -- Apply the adjustment.
VAL6
PRKEY -- The wavelength wave printout key:
0 -- No extra printout. This value is the default.
1 -- Include the extra printout.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1138
OCDATA
Figure 4: Velocity Profiles for Wave-Current Interaction
Mean Water
Surface
Mud Line
Constant (KCRC = 0)
Stretch (KCRC = 1)
Continuity (KCRC = 2)
Horizontal arrows represent input velocities
Water Surface
Menu Paths
Main Menu>Preprocessor>Sections>Pipe>Ocean>Add/Edit
OCDELETE, OCFIRST, OCLAST, OCINC, --, LCHK
Deletes ocean information from the ANSYS database.
PREP7: Ocean
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
OCFIRST
First ocean ID to be deleted; defaults to first available ocean in the database.
OCLAST
Last ocean ID to be deleted; defaults to last available ocean in the database.
OCINC
Increment of the ocean ID; defaults to 1.
--
Unused field.
LCHK
Specifies the level of element-associativity checking:
NOCHECK
No element-associativity check occurs. This option is the default.
WARN
When an ocean ID is associated with an element, ANSYS issues a message warning that the necessary
entity has been deleted.
1139
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OCDELETE
CHECK
The command terminates, and no ocean ID is deleted if it is associated with an element.
Notes
Deletes one or more specified ocean IDs and their associated data from the ANSYS database.
Menu Paths
Main Menu>Preprocessor>Sections>Pipe>Ocean>Delete
OCLIST, SFIRST, SLAST, SINC, Details, Type
Summarizes all currently defined ocean environments.
PREP7: Ocean
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SFIRST
First ocean environment to be summarized. Defaults to first available ocean ID in the database.
SLAST
Last ocean environment to be summarized. Defaults to last available ocean ID in the database.
SINC
Increment of the ocean ID. Defaults to 1.
Details
Determines the content of the summarized ocean information:
BRIEF
Lists only the ocean IDs requested. This option is the default.
GROUP
If a basic ocean environment calls for current and/or waves, this option lists them too.
Notes
The OCLIST command summarizes the ocean properties for all defined ocean environments in the current
session of ANSYS.
Menu Paths
To be provided at time of release.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1140
OCLIST
OCTABLE, VAL1, VAL2, VAL3, VAL4, VAL5, VAL6
Defines an ocean environment using table data.
PREP7: Ocean
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
VAL1, VAL2, VAL3, . . . , VAL6
Values describing the basic ocean environment, a current condition, or a wave condition.
Notes
The OCTABLE specifies table data that defines the ocean environment. The terms VAL1, VAL2, etc. are
specialized according to the input set required for the given ocean environment.
ANSYS interprets the data input via the OCTABLE command within the context of the most recently issued
OCTYPE command.
There is no limit to the number of data input. You can issue this command as often as necessary. You must
input values in the order indicated:
Ocean Data Type: BASIC (OCTYPE,,BASIC)
Data to provide in the value fields:
RE, CDy, CDz, CT, CMy, CMz
where
RE = Reynolds number for coefficients for this command. Input these values in ascending order from
one command to the next.
CDy = Drag coefficient in the element y direction (normal).
CDz = Drag coefficient in the element z direction (normal). This value defaults to CDy.
CT = Drag coefficient in the element x direction (tangential).
CMy = Coefficient of inertia in the element y direction.
CMz = Coefficent of inertia in the element z direction.
Ocean Data Type: CURRENT (OCTYPE,,CURR)
Data to provide in the value fields:
Z, W, Th, T
where
Z = Coordinate location of the drift current. Input these values in descending order from one command
to the next. (The first Z value must be zero and the last one must be -Depth.) If the current is constant,
only one OCTABLE command is necessary and this value is not required.
W = Velocity of the drift current.
Th = Angle of the drift current.
T = Temperature at this location.
Ocean Data Type: WAVE (OCTYPE,,WAVE)
Data to provide in the value fields:
A, Per, Ps, Lw
where
A = Peak-to-trough amplitude
Per = Wave period.
1141
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OCTABLE
Ps = Phase shift.
Lw = Wavelength. This value is optional.
When using Stokes Fifth Order Wave Theory (KWAVE = 2 on the OCDATA command), only one OCTABLE
command is necessary. Wavelength (Lw) is not input.
When using Stream Function Wave Theory (KWAVE = 3 on the OCDATA command), X(j) / (H * T * G) is
defined in A(j), where j is the number of the OCTABLE command. Also, DPT / LO = Per(2), L / LO = Per(3),
H / DPT = Per(4), and / G * H * T) = Per(5). Phase shift is input in Ps(1).
Menu Paths
Main Menu>Preprocessor>Sections>Pipe>Ocean>Add/Edit
OCTYPE, IDKEY, DataType, Name, OCIDC, OCIDW
Specifies the type of ocean environment data to follow.
PREP7: Ocean
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
IDKEY
The ocean environment, current, or wave ID number:
OCID
The ocean ID. Specify this value when DataType = BASIC.
OCIDC
The ocean current ID. Specify this value when DataType = CURR.
OCIDW
The ocean wave ID. Specify this value when DataType = WAVE.
DataType
The type of ocean data to be input following this command:
BASIC
Defines the basic ocean environment. The BASIC argument is required for any ocean loading.
CURR
Defines ocean drift current. The CURR argument is optional and uses the ocean environment ID
defined via the SOCEAN command.
WAVE
Defines an ocean wave state. The WAVE argument is optional and uses the ocean environment ID
defined via the SOCEAN command.
Specify basic, current, or wave input via the OCDATA and OCTABLE commands. For an
example input fragment showing how to use the ocean environment data types, see the
Notes section.
Name
An eight-character name for the ocean environment. An ocean name can consist of letters and numbers,
but cannot contain punctuation, special characters, or spaces.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1142
OCTYPE
OCIDC
The ocean current ID. Specify this value only when DataType = BASIC.
OCIDW
The ocean wave ID. Specify this value only when DataType = BASIC.
Notes
The OCTYPE command specifies the type of ocean environment data to follow (basic, current, or wave)
along with the appropriate ID number. Issue this command before inputting your ocean environment data
(OCDATA and OCTABLE).
The ocean environment ID number OCID must be identical to the OCID value specified via the SOCEAN
command. (SOCEAN associates the ocean ID with the most recently defined element section.)
Ocean loading applies only to pipe elements PIPE288 and PIPE289.
The ocean ID number (OCID), ocean current ID number (OCIDC), or ocean wave ID number (OCIDW) must
differ from each other and from any element section ID number.
The following example shows how you can use the basic (DataType = BASIC), current (DataType = CURR),
and wave (DataType = WAVE) ocean data types within the context of a simple input file fragment:
Do=1.5 ! outside diameter
th=0.1 ! wall thickness
height=10 ! wave height
CS=2 ! surface current speed
Depth=100 ! water depth
matwat=2 ! material number id of the ocean
secpipe= 1 ! section number of the pipe
idwat=2 ! id number of the ocean
idcur=3 ! id number of the current loading
idwav=4 ! id number of the wave loading
!
sectype,secpipe,pipe,,pipetest
secdata,Do,th,16 ! 16 cells around circumference
socean,idwat ! must be present for every section loaded by ocean
!
octype,idwat,basic,idcur,idwav
ocdata,Depth,matwat,0,-1,-1 ! suppress added mass and buoyancy
octable,,.5,.5,,2 ! CD = .5, CM = 2
octype,idcur,curr
octable, 0.0,CS ! input free surface current speed
octable,-Depth,0.00 ! input ocean floor current speed of 0.0
!
octype,idwav,wave
ocdata,2 ! request Stokes wave theory
octable,height,8 ! wave period of 8 seconds
slist,all ! lists pipe section AND
! mentions ocean loading (socean,idwat)
oclist,all ! lists details of ocean loading
To associate your ocean environment with an element section, issue the SOCEAN command.
Menu Paths
Main Menu>Preprocessor>Sections>Pipe>Ocean>Add/Edit
1143
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OCTYPE
OMEGA, OMEGX, OMEGY, OMEGZ, KSPIN
Specifies the rotational velocity of the structure.
SOLUTION: Inertia
MP ME ST PR PRN DS DSS FL <> <> DY PP <> EME MFS
OMEGX, OMEGY, OMEGZ
Rotational velocity of the structure about the global Cartesian X, Y, and Z axes.
KSPIN
The spin-softening key:
0 or OFF
No modification of the stiffness matrix of the element component due to rotation. This value is the
default except in a dynamic analysis when the Coriolis effect is taken into account in the rotating
reference frame (CORIOLIS,ON); in this case, ANSYS applies spin-softening by default.
1 or ON
Decrease radial stiffness of the element component due to rotation (that is, include spin-softening
effects).
Notes
This command specifies the rotational velocity of the structure about each of the global Cartesian axes (right-
hand rule). Rotational velocities may be defined in these analysis types:
Static (ANTYPE,STATIC)
Harmonic (ANTYPE,HARMIC) -- Full or mode superposition
Transient (ANTYPE,TRANS) -- Full or mode superposition
Substructuring (ANTYPE,SUBSTR)
The OMEGA command supports tabular boundary conditions (%TABNAME_X%, %TABNAME_Y%, and
%TABNAME_Z%) for OMEGA_X, OMEGA_Y, and OMEGA_Z input values (*DIM) for full transient and harmonic
analyses.
Rotational velocities are combined with the element mass matrices to form a body force load vector term.
Units are radians/time. Related commands are ACEL, CGLOC, CGOMGA, DCGOMG, and DOMEGA.
The KSPIN option allows you to adjust the stiffness of a rotating body to account for dynamic mass effects.
The adjustment is called spin-softening and applies to a modal (ANTYPE,MODAL), harmonic (ANTYPE,HARM),
or transient (ANTYPE,TRANS) analysis only. By default, ANSYS does not include the spin-softening adjustment
in the analysis except for dynamic analyses where you have applied the Coriolis effect (CORIOLIS,ON) in the
rotating reference frame. To account for this exception, if the default is requested, the KSPIN value appears
as -1 in all saved files containing loads (such as .CDB, .Snn, and .LDHI). The adjustment approximates the
effects of geometry changes caused by large-deflection circumferential motion in a small-deflection analysis.
The KSPIN option is not intended for a large-deflection static analysis; in such a case, use the NLGEOM,ON
command to account for this effect. The command is usually used in conjunction with prestressing (PSTRES).
If you have applied the Coriolis effect (CORIOLIS) using a stationary reference frame, the OMEGA command
takes the gyroscopic damping matrix into account for the elements listed in the Stationary Reference
Frameheading in the notes section of the CORIOLIS command. The element axis must pass through the
global Cartesian origin. ANSYS verifies that the rotation vector axis is parallel to the axis of the element; if
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1144
OMEGA
not, the gyroscopic effect is not applied. After issuing the OMEGA command when the Coriolis or gyroscopic
effect is present, a subsequently issued CMOMEGA command has no effect.
In a modal harmonic or transient analysis, you must apply the load in the modal portion of the analysis.
ANSYS calculates a load vector and writes it to the mode shape file, which you can apply via the LVSCALE
command.
The overall stiffness matrix should not be reused (KUSE) if OMEGA command parameters change between
substeps and the spin-softening key (KSPIN) is active. See Reuse of Matrices or Structure Matrices in the
Theory Reference for the Mechanical APDL and Mechanical Applications for more information.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>Global
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Inertia>Angular Veloc>Global
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>Global
Main Menu>Solution>Define Loads>Delete>Structural>Inertia>Angular Veloc>Global
OPADD, NRES, NUM1, NUM2, C1, C2
Forms a set of optimization parameters by adding two sets.
OPTIMIZATION: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NRES
Number assigned to results set. If same as existing set, the existing values will be overwritten by these
results (defaults to next available set number).
NUM1
Number of first design set to be used in operation.
NUM2
Number of second design set to be used in operation (may be blank).
C1
Scale factor applied to NUM1 (defaults to 1.0).
C2
Scale factor applied to NUM2 (defaults to 1.0).
Notes
Forms a result set of parameters by adding two existing design sets according to the operation:
NRES = (C1 x NUM1) + (C2 x NUM2)
OPADD can also be used to scale results for a single set. If no set number is specified for the results set
(NRES), it defaults to the next available set number. Because no compression of set numbers is done, the
next available set number will be either 1 + the highest existing set number, or 1 + the highest set number
used (if sets with higher numbers previously existed but have been removed).
1145
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPADD
Menu Paths
Main Menu>Design Opt>Design Sets>Combine
OPANL, Fname, Ext, --
Defines the analysis file to be used for optimization looping.
OPTIMIZATION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
Command Default
If interactive, no default; if batch, use the batch input stream (File.BAT).
Notes
The optimization looping file must be specified for an optimization which is performed interactively. The
file must exist at the time OPANL is issued. In this file, where the /PREP7 and /OPTcommands occur, they
must be the first nonblank characters on the line (i.e., do not use the $ delimiter on any /PREP7 or /OPT
command lines).
Menu Paths
Main Menu>Design Opt>Analysis File>Assign
OPCLR
Clears the optimization database.
OPTIMIZATION: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Clears the optimization database. All settings are reset to their default values, and all design sets are deleted.
This command is useful between multiple (independent) optimization analyses in the same batch run or
interactive session.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1146
OPANL
Menu Paths
Main Menu>Design Opt>Opt Database>Clear & Reset
OPDATA, Fname, Ext, --
Identifies the file where optimization data is to be saved.
OPTIMIZATION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to OPT if Fname is blank.
--
Unused field.
Command Default
Optimization data is saved on File.OPT.
Notes
Identifies the file where optimization data is to be automatically saved during looping. Automatic saves occur
at the end of each loop or iteration, and upon finishing out of the optimizer [FINISH]. (To save the data
elsewhere in the optimizer, use OPSAVE.) Optimization data written to the file include DVs, SVs, saved design
sets, analysis file name, etc. The data can be restored via the OPRESU command.
Menu Paths
Main Menu>Design Opt>Controls
OPDEL, NSET1, NSET2
Deletes optimization design sets.
OPTIMIZATION: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET1
First design set in range to be deleted. Defaults to 0. If NSET1 is blank and NSET2 is
NSET2,
where N is an existing set number) are permanently removed from the optimization database. The original
set numbers are retained for remaining design sets. Note that no compression of set numbers is done. Thus,
if the highest set is deleted, the next available set number will be the previously existing highest set number
+ 1. (A maximum of 130 design sets can be stored in the optimization database.)
Menu Paths
Main Menu>Design Opt>Design Sets>Select/Delete
OPEQN, KFOBJ, KFSV, KWGHT, KOPPR, INOPT
Controls curve fitting for the subproblem approximation method.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KFOBJ
Objective function curve fit:
0
Quadratic plus cross-term curve fit for objective function (default).
1
Linear curve fit.
2
Quadratic curve fit.
3
Quadratic plus cross-term fit (same as 0).
KFSV
State variables curve fit:
0
Quadratic curve fit for state variables (default).
1
Linear curve fit.
2
Quadratic curve fit (same as 0).
3
Quadratic plus cross-term curve fit.
KWGHT
Weighting factors:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1148
OPEQN
0
Weights applied to design sets based on triple products of distance in design space, objective
function values, and feasibility/infeasibility (KWGHT = 2, 3, and 4, below) (default).
1
All weights set to unity.
2
Weights based on distance in design space.
3
Weights based on objective function values.
4
Weights based on feasibility/infeasibility.
KOPPR
Optimization printout:
0
No printout of approximation data (default).
N
Printout every nth optimization loop.
INOPT
Approximation reformulation:
0
Approximations are updated every loop (default).
N
Approximations are completely reformulated every nth loop.
Notes
Controls the curve fitting for the optimization approximations used with the subproblem approximation
method [OPTYPE,SUBP]. See Function Approximations in the Theory Reference for the Mechanical APDL and
Mechanical Applications for details.
Menu Paths
Main Menu>Design Opt>Method/Tool
OPERATE
Specifies "Operation data" as the subsequent status topic.
POST26: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
1149
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPERATE
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
This command cannot be accessed from a menu.
OPEXE
Initiates optimization looping.
OPTIMIZATION: Run
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Initiates optimization looping using commands condensed from the analysis file [OPANL]. The type of op-
timization to be performed is specified by the OPTYPE command. Upon issuing this command, optimization
looping begins and the design variables are updated as necessary based on the type of optimization chosen.
Control remains with the ANSYS program until convergence or termination occurs, at which time the com-
mands occurring after OPEXE are executed. This command is not allowed within a do-loop [*DO].
Menu Paths
Main Menu>Design Opt>Run
OPFACT, Type
Defines the type of factorial evaluation to be performed.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Type of factorial evaluation. Number of iterations is determined by the number of design variables (n).
FULL
Perform a full factorial analysis (default); 2**n iterations are required. n must be
7.
1/2
Perform a 1/2 fractional factorial analysis. (1/2)*2**n iterations are required. n must be
8.
1/4
Perform a 1/4 fractional factorial analysis. (1/4)*2**n iterations are required. n must be
9.
1/8
Perform a 1/8 fractional factorial analysis. (1/8)*2**n iterations are required. n must be
10.
1/16
Perform a 1/16 fractional factorial analysis. (1/16)*2**n iterations are required. n must be
10.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1150
OPEXE
1/32
Perform a 1/32 fractional factorial analysis. (1/32)*2**n iterations are required. n must be
10.
1/64
Perform a 1/64 fractional factorial analysis. (1/64)*2**n iterations are required. n must be
10.
Command Default
Perform a full factorial analysis.
Notes
This command is valid for the factorial evaluation method of optimization [OPTYPE,FACT]. The factorial
evaluation is defined as either full or fractional. The number of iterations performed is dependent on the
number of design variables that have been defined (n).
Menu Paths
Main Menu>Design Opt>Method/Tool
OPFRST, NITR, SIZE, DELTA
Defines specifications for the first order optimization method.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NITR
Maximum number of optimization iterations for the next execution [OPEXE]. Defaults to previously
specified value, if any, otherwise defaults to 10.
SIZE
Limit (in percent) that is applied to the size of each line search step. For example, setting SIZE = 10
will limit design variable changes to 10 percent of the maximum range of design space (defined by
feasible design variable limits specified on OPVAR commands) at each iteration. Defaults to previously
specified value, if any; otherwise, defaults to 100 (percent).
DELTA
The forward difference (in percent) applied to the design variable range that is used to compute the
gradient. For example, the shift in a DV is DELTA*(MAX - MIN)/100, where MIN and MAX are specified
on the OPVAR command. Defaults to previously specified value, if any; otherwise, defaults to 0.2 (percent).
Command Default
NITR = 10, SIZE = 100, DELTA = 0.2.
Notes
This command is valid for the first order method of optimization [OPTYPE,FIRST].
1151
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPFRST
Menu Paths
Main Menu>Design Opt>Method/Tool
OPGRAD, Dset, DELTA
Specifies which design set will be used for gradient evaluation.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Dset
BEST
Use the best design set as the evaluation point (default).
LAST
Use the last design set as the evaluation point.
n
Use design set number n as the evaluation point.
DELTA
The forward difference applied to the design variable range that is used to compute the gradient. For
example, the shift in a design variable (DV) is defined as DELTA*(MAX-MIN)/100, with MIN and MAX as
specified on the OPVAR command. Defaults to 0.5.
Command Default
As described for each argument above.
Notes
This command is valid for the gradient evaluation method of optimization [OPTYPE,GRAD]. OPGRAD specifies
which point (design set) in design space will be used to evaluate the gradient and what the forward difference
will be.
Menu Paths
Main Menu>Design Opt>Method/Tool
OPKEEP, Key
Specifies whether to save the best-set results and database file.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Key
Save key:
OFF
Do not save the results and database files for the best design set (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1152
OPGRAD
ON
Save results and database files (File.BRST and File.BDB) for the best design set during optim-
ization looping.
Command Default
Do not save the database and results for the best design set.
Notes
Specifies whether or not the results file and database file corresponding to the best design set (based on
all existing sets) should be saved during optimization. The saved files will be named File.BRST (or
File.BRTH for thermal analysis, File.BRMG for magnetic analysis, File.BRFL for a FLOTRAN CFD ana-
lysis) and File.BDB. If Key = ON, both files will be saved at the end of each optimization loop which results
in a new best design set. File.RST (results) and File.DB (database) will be saved as usual for the last
design set, regardless of the Key setting. If multiple analyses are done within an optimization loop, then
only the last analysis is considered for the Key = ON option. For example, if a thermal analysis is followed
by a stress analysis, the files saved for the best set will be File.BRST and File.BDB, which will both
correspond to the structural analysis.
Menu Paths
Main Menu>Design Opt>Controls
OPLFA, Name, Effect, EMIN, EMAX
Displays the results of a factorial evaluation.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name. The parameter must have been previously defined as the objective function or a state
variable [OPVAR].
Effect
Display effects for parameter Name.
MAIN
Display main effects (default).
2FAC
Display two-factor interactions.
3FAC
Display three-factor interactions.
EMIN
Minimum value of effect to be displayed. Default - computed minimum.
EMAX
Maximum value of effect to be displayed. Default - computed maximum.
1153
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPLFA
Notes
Displays the results of a factorial analysis in bar chart form. Only the 10 largest interaction values will be
displayed in the specified range (EMIN to EMAX). This command is only valid after a factorial evaluation
[OPTYPE,FACT].
Menu Paths
Main Menu>Design Opt>Design Sets>Tool Results>Graph>Factorial
OPLGR, Pname, Dvnam1, Dvnam2, Dvnam3, Dvnam4, Dvnam5, Dvnam6
Graphs the results of a gradient evaluation.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Pname
Response parameter name. The parameter must have been previously defined as the objective function
or a state variable [OPVAR].
Dvnam1, Dvnam2, Dvnam3, . . . , Dvnam6
Design variable names.
Notes
This command is only valid after a gradient evaluation [OPTYPE,GRAD]. Pname is graphed with respect to
a plus or minus 1% change in design variable (Dvnam1, ... Dvnam6).
Menu Paths
Main Menu>Design Opt>Design Sets>Tool Results>Graph>Gradient
OPLIST, SET1, SET2, LKEY
Displays the parameters for design sets.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SET1, SET2
Display values of all analysis parameters from SET1 (defaults to the highest existing set) to SET2 (defaults
to SET1). If SET1 = ALL, display for all sets.
LKEY
Listing key:
0
List only scalar parameters related to optimization.
1
List all analysis scalar parameters.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1154
OPLGR
Notes
Displays the values of the parameters for specified design sets. Note that a *STATUS command may also
be issued to display various optimization data.
Menu Paths
Main Menu>Design Opt>Design Sets>List
OPLOOP, Read, Dvar, Parms
Specifies controls for optimization looping.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Read
Indicates where to begin reading the analysis file during optimization looping. Note that the /PREP7
command (and the /OPT command) must occur as the first nonblank characters on a line in the analysis
file (i.e., do not use the $ delimiter). Defaults to previously specified label, if any; otherwise, defaults to
TOP:
TOP
Read from the first line.
PREP
Read from the first occurrence of /PREP7.
Dvar
Indicates how to treat parameters which have been designated as design variables (DVs) during optim-
ization looping. Defaults to previously specified label, if any; otherwise, defaults to IGNORE:
IGNORE
Do not process DV parameter assignments in the analysis file during looping.
PROCESS
Process DV parameter assignments in the analysis file during looping.
Parms
Indicates which types of parameters to save during optimization looping. Defaults to previously specified
label, if any; otherwise, defaults to SCALAR:
SCALAR
Save scalar parameters only.
ALL
Save all parameters, both scalar and array.
Command Default
Read = TOP, Dvar = IGNORE, and Parms = SCALAR.
Menu Paths
Main Menu>Design Opt>Controls
1155
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPLOOP
OPLSW, Pname, Dvnam1, Dvnam2, Dvnam3, Dvnam4, Dvnam5, Dvnam6
Graphs the results of a global sweep generation.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Pname
Response parameter name. The parameter must have been previously defined as the objective function
or a state variable (OPVAR).
Dvnam1, Dvnam2, Dvnam3, . . . , Dvnam6
Design variable names.
Notes
This command is only valid after a global sweep generation [OPTYPE,SWEEP]. Pname is graphed with respect
to the specified design variables (Dvnam1, ... Dvnam6). The design variables are normalized (0 to 1) along
the X-axis.
Menu Paths
Main Menu>Design Opt>Design Sets>Tool Results>Graph>Sweeps
OPMAKE
Creates a design set using active scalar parameter values.
OPTIMIZATION: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
The next available design set number is assigned to the new design set. Because no compression of set
numbers is done, the next available set number will be either 1 + the highest existing set number, or 1 +
the highest set number used (if sets with higher numbers previously existed but have been removed). Multiple
design sets can be created by repeated use of PARRES (to read parameters from a file) and OPMAKE.
Menu Paths
Main Menu>Design Opt>Design Sets>Create
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1156
OPLSW
OPNCONTROL, Lab, VALUE, NUMSTEP
Sets decision parameter for automatically increasing the time step interval.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Lab
DOF
DOF label used to base a decision for increasing the time step (substep) interval in a nonlinear or
transient analysis. The only DOF label currently supported is TEMP.
OPENUPFACTOR
Factor for increasing the time step interval. Use Lab = OPENUPFACTOR (when AUTOTS,ON is invoked)
and specify a VALUE greater than 1.0. Default for OPENUPFACTOR is 1.5. The upper limit for OPEN-
UPFACTOR is 10.0. However, we recommend using a value not greater than 3.0.
VALUE, NUMSTEP
Two values used in algorithm for deciding if the time step interval can be increased. The time step interval
is increased if the maximum absolute value of the incremental solution at the specified DOF label is less
than VALUE for the number of contiguous time steps specified by NUMSTEP. The default for VALUE is
0.1 and the default for NUMSTEP is 3. NUMSTEP is used only when Lab = DOF.
Notes
This command is available only for nonlinear or transient analysis, and only when SOLCONTROL,ON.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Open Control
Main Menu>Solution>Load Step Opts>Nonlinear>Open Control
OPPRNT, Key
Activates detailed optimization summary printout.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Key
Summary key:
OFF
Do not print details (default).
ON
Print the detailed summary.
FULL
Same as Key = ON, but includes a full list of all design sets.
Notes
Activates detailed optimization summary printout to the primary output during optimization analysis.
1157
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPPRNT
Menu Paths
Main Menu>Design Opt>Controls
OPRAND, NITR, NFEAS
Defines the number of iterations for a random optimization.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NITR
Maximum number of optimization iterations for the next execution [OPEXE]. Defaults to previously
specified value, if any; otherwise, defaults to 1.
NFEAS
Desired total number of feasible design sets (including existing feasible sets). Once NFEAS feasible sets
are achieved, random iterations are terminated. If 0 is input, NITR random iterations will be performed
independent of the number of feasible designs that are obtained. Defaults to previously specified value,
if any; otherwise, defaults to 0.
Command Default
Perform 1 iteration.
Notes
This command is valid for the random design generation method of optimization [OPTYPE,RAND].
Menu Paths
Main Menu>Design Opt>Method/Tool
OPRESU, Fname, Ext, --
Reads optimization data into the optimization database.
OPTIMIZATION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to OPT if Fname is blank.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1158
OPRAND
--
Unused field.
Notes
Reads optimization data from the specified file into the optimization database. Optimization data must have
been previously saved to this file [OPSAVE], or automatically written to this file in an earlier optimization
analysis. All existing parameters will be deleted and replaced by those stored on the resumed file. (To save
existing parameters, use PARSAV before OPRESU.)
Menu Paths
Main Menu>Design Opt>Opt Database>Resume
OPRFA, Name
Prints the results of a factorial evaluation.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name. The parameter must have been previously defined as the objective function or a state
variable [OPVAR]. If Name = ALL, print results for all optimization response variables (default).
Notes
This command is only valid after a factorial evaluation [OPTYPE,FACT].
Menu Paths
Main Menu>Design Opt>Design Sets>Tool Results>Print
OPRGR, Name
Prints the results of a gradient evaluation.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name. The parameter must have been previously defined as the objective function or a state
variable [OPVAR]. If Name = ALL, print results for all optimization response variables (default).
Notes
This command is only valid after a gradient evaluation [OPTYPE,GRAD].
1159
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPRGR
Menu Paths
Main Menu>Design Opt>Design Sets>Tool Results>Print
OPRSW, Name
Prints the results of a global sweep generation.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name. The parameter must have been previously defined as the objective function or a state
variable (OPVAR). If Name = ALL, print results for all optimization response variables (default).
Notes
This command is only valid after a global sweep generation [OPTYPE,SWEEP].
Menu Paths
Main Menu>Design Opt>Design Sets>Tool Results>Print
OPSAVE, Fname, Ext, --
Writes all optimization data to a file.
OPTIMIZATION: Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to OPT if Fname is blank.
--
Unused field.
Notes
Writes all optimization data to File.OPT or a named file. Saved data includes optimization data only (DVs,
SVs, saved design sets, analysis file name, etc.). The data can be restored [OPRESU] for an optimization restart.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1160
OPRSW
Menu Paths
Main Menu>Design Opt>Opt Database>Save
OPSEL, NSEL
Selects design sets for subsequent optimization looping.
OPTIMIZATION: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSEL
Number of best design sets to be selected. If NSEL = -1, select all feasible designs. If NSEL is positive
and no objective function is defined, the following occurs: all infeasible design sets are removed; next
the lowest number (oldest) feasible design sets are removed until NSEL sets are left (i.e., the latest
feasible design sets are kept).
Notes
A number of best design sets or all feasible design sets may be specified. ("Best" design sets are determined
by the objective function value for each feasible design.) All design sets not selected are permanently removed
from the optimization database. The original set numbers are retained for remaining design sets. Note that
no compression of set numbers is done. Thus, if the highest set is removed, the next available set will still
be the previously used highest set number + 1. (A maximum of 130 design sets can be stored in the optim-
ization database.)
Menu Paths
Main Menu>Design Opt>Design Sets>Select/Delete
OPSUBP, NITR, NINFS
Defines number of iterations for subproblem approximation method.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NITR
Maximum number of optimization iterations for the next execution [OPEXE]. Defaults to previously
specified value, if any; otherwise, defaults to 30.
NINFS
Maximum number of consecutive infeasible design sets that are allowed. Once NINFS consecutive in-
feasible sets are encountered, iterations are terminated. Defaults to previously specified value, if any;
otherwise, defaults to 7.
Notes
This command is valid for the subproblem approximation method of optimization [OPTYPE,SUBP].
1161
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPSUBP
Menu Paths
Main Menu>Design Opt>Method/Tool
OPSWEEP, Dset, NSPS
Specifies the reference point and number of evaluation points for a sweep generation.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Dset
BEST
Use the best design set as the reference point (default).
LAST
Use the last design set as the reference point.
N
Use design set number N as the reference point.
NSPS
Number of evaluation points used for each design variable sweep. Two to ten evaluation points are al-
lowed. Defaults to 2 (i.e., the minimum and maximum design variable values).
Command Default
As described by the arguments above.
Notes
This command is valid for the sweep evaluation method of optimization [OPTYPE,SWEEP].
Menu Paths
Main Menu>Design Opt>Method/Tool
/OPT
Enters the design optimizer.
SESSION: Processor Entry
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Enters the ANSYS optimizer for design optimization by iterative analyses. Upon issuing this command, if
there are no existing design sets, all currently defined numerical scalar parameters are established as design
set number 1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1162
OPSWEEP
This command is valid only at the Begin Level.
Menu Paths
Main Menu>Design Opt
OPTYPE, Mname
Specifies the optimization method to be used.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Product Restrictions
Mname
Name of the optimization method to be used:
SUBP
Subproblem approximation method. Specifications for this method are defined on the OPSUBP
command.
FIRST
First order optimization method. Specifications for this method are defined on the OPFRST command.
RAND
Perform random iterations. Specifications for this method are defined on the OPRAND command.
RUN
Perform a single optimization run with current parameter values for the defined design variables.
FACT
Perform factored design iterations. Specifications for this method are defined on the OPFACT com-
mand.
GRAD
Compute a gradient (design sensitivity) at a point in design space. Specifications for this method
are defined on the OPGRAD command.
SWEEP
Perform global sweeps through global design space starting from a single design set. Specifications
for this method are defined on the OPSWEEP command.
USER
User-supplied external optimization. See the Advanced Analysis Techniques Guide for user optimization
information (USEROP routine). Parameters used for this method are defined on the OPUSER command.
Command Default
Perform a single optimization run (Mname = RUN) with current parameter values for the defined design
variables.
Notes
Specifies the optimization method to be used upon execution [OPEXE].
1163
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPTYPE
Product Restrictions
Available Products Command Op-
tion Mname
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS SUBP
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS FIRST
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS OPRAND
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS RUN
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS FACT
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS GRAD
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS SWEEP
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS USER
Menu Paths
Main Menu>Design Opt>Method/Tool
OPUSER, NITR, VAL1, VAL2, VAL3, VAL4, VAL5, VAL6, VAL7, VAL8
Defines specifications for user-supplied external optimization.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
NITR
Maximum number of optimization iterations for the next execution [OPEXE]. Defaults to previously
specified value, if any; otherwise, defaults to 1.
VAL1, VAL2, VAL3, . . . , VAL8
Numerical values (double precision) for user-supplied optimizer. Values may be the results of parameter
evaluations. Values are written to File.OPT after the parameter set data. VAL1--VAL8 default to pre-
viously specified values, if any; otherwise, default to 0.
Notes
Defines the number of iterations and the parameter values for user-supplied external optimization [OP-
TYPE,USER]. Values are passed to the user routine (USEROP) through the optimization common block.
Menu Paths
Main Menu>Design Opt>Method/Tool
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1164
OPUSER
OPVAR, Name, Type, MIN, MAX, TOLER
Specifies the parameters to be treated as optimization variables.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name (must be a scalar parameter).
Type
Optimization variable type:
DV
Design variable. MAX must also be specified.
SV
State variable, also referred to as a constrained variable. (Usually defined via a *GET command.) MIN,
or MAX, or both must be specified.
OBJ
Objective function (variable to be minimized). Only one objective function is allowed. MIN and MAX
are not used.
DEL
Deletes this optimization variable (does not delete the parameter). This option only has meaning if
the parameter Name was previously defined as an optimization variable (DV, SV, or OBJ). The para-
meter retains the value assigned during the last loop. MIN and MAX are not used.
MIN
Minimum value for this parameter. For Type = DV, MIN must be greater than 0.0 and defaults to
0.001*(MAX). For Type = SV, there will be no lower limit if MIN is left blank, but if MIN = 0.0, the lower
limit is zero. MIN is ignored for Type = OBJ. MIN is also ignored for Type = DEL (except for the special
case when Name = (blank) and Type = DEL, as described next).
When the GUI is on, if a Delete operation (in a Design Variables or State Variables dialog box) writes
this command to a log file (Jobname.LOG or Jobname.LGW), you will observe that Name is blank,
Type = DEL, and MIN is an integer number. In this case, the GUI has assigned a value of MIN that cor-
responds to the location of a chosen variable name in the dialog box's list. It is not intended that you
type in such a location value for MIN in an ANSYS session. However, a file that contains a GUI-generated
OPVAR command of this form can be used for batch input or for use with the /INPUT command.
MAX
Maximum value for this parameter. Required for Type = DV and ignored for Type = OBJ or DEL. For
Type = SV, there will be no upper limit if MAX is left blank, but if MAX = 0.0, the upper limit is zero.
TOLER
For Type = DV and OBJ, acceptable change between loops for convergence; defaults to 0.01*(current
value) for OBJ and 0.01*(MAX-MIN) for DV. For Type = SV, defines feasibility tolerance. For double-sided
SV limits (MAX and MIN specified), the default is 0.01*(MAX-MIN). For a single-sided SV limit (MAX or MIN
specified), the default is 0.01*(specified limit) or, if the absolute value of the limit is < 1, the default is
0.01*(current SV value). TOLER is ignored for Type = DEL.
Command Default
No optimization parameters defined.
1165
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OPVAR
Notes
Specifies the parameters to be treated as optimization variables. Variable constraints (MIN and MAX) and
tolerances (TOLER) are also specified on this command. If the specified parameter (Name) does not exist at
the time OPVAR is issued, the OPVAR command defines that parameter and assigns a value of zero to it.
Up to 60 DV and 100 SV parameters are allowed.
Menu Paths
Main Menu>Design Opt>Design Variables
Main Menu>Design Opt>Objective
Main Menu>Design Opt>State Variables
OUTOPT
Specifies "Output options" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Output Options
OUTPR, Item, FREQ, Cname
Controls the solution printout.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Item for print control:
BASIC
Basic quantities (nodal DOF solution, nodal reaction loads, and element solution) (default).
NSOL
Nodal DOF solution.
RSOL
Nodal reaction loads.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1166
OUTOPT
ESOL
Element solution.
NLOAD
Element nodal loads. When nonlinear stabilization is active, the stabilization force/moments are also
printed.
SFOR
Stabilization force/moment at the applicable nodes (valid only when nonlinear stabilization is active).
VENG
Element energies. When nonlinear stabilization is active, the energy dissipation due to stabilization
is also printed.
V
Nodal velocity (applicable to structural transient analysis only (ANTYPE,TRANS)).
A
Nodal acceleration (applicable to structural transient analysis only (ANTYPE,TRANS)).
ALL
All of the above solution items.
FREQ
Print solution for this item every FREQ th (and the last) substep of each load step. If -n, print up to n
equally spaced solutions (for automatic time stepping). If NONE, suppress all printout for this item for
this load step. If ALL, print solution for this item for every substep. If LAST, print solution for this item
only for the last substep of each load step. For a modal analysis, use NONE or ALL.
Cname
Name of the component, created with the CM command, defining the selected set of nodes or elements
for which this specification is active. If blank, the set is all entities.
Note
The component named must be of the same type as the item, i.e. nodal or element. A com-
ponent name is not allowed with the BASIC or ALL labels.
Command Default
No printout.
Notes
Controls the solution items to be printed, the frequency with which they are printed (in static, transient, or
full harmonic analyses), and the set of nodes or elements to which this specification applies (in static, tran-
sient, or full harmonic analyses). An item is associated with either a node (NSOL, RFORCE, V, and A items)
or an element (all of the remaining items). The specifications are processed in the order that they are input.
Up to 50 specifications (OUTPR and OUTRES commands combined) may be defined. Use OUTPR,STAT to
list the current specifications and use OUTPR,ERASE to erase all the current specifications.
As described above, OUTPR writes some or all items (depending on analysis type) for all elements. To restrict
the solution printout, use OUTPR to selectively suppress (FREQ = NONE) the writing of solution data, or
first suppress the writing of all solution data (OUTPR,ALL,NONE) and then selectively turn on the writing of
solution data with subsequent OUTPR commands.
1167
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OUTPR
If the generalized plane strain feature is active and OUTPR is issued, the change of fiber length at the ending
point during deformation and the rotation of the ending plane about X and Y during deformation will be
printed if any displacement at the nodes is printed. The reaction forces at the ending point will be printed
if any reaction force at the nodes is printed.
Nodal reaction loads (Item = RSOL) are processed according to the specifications listed for the PRRSOL
command.
Result printouts for interactive sessions are suppressed for models with more than 10 elements.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Output Ctrls>Solu Printout
Main Menu>Solution>Load Step Opts>Output Ctrls>Solu Printout
/OUTPUT, Fname, Ext, --, Loc
Redirects text output to a file or to the screen.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
Filename and directory path (248 character maximum, including directory) to which text output will be
redirected (defaults to Jobname if Ext is specified). For interactive runs, Fname = TERM (or blank) re-
directs output to the screen. For batch runs, Fname = blank (with all remaining command arguments
blank) redirects output to the default system output file.
Ext
Filename extension (8 character maximum).
--
Unused field.
Loc
Location within a file to which output will be written:
(blank)
Output is written starting at the top of the file (default).
APPEND
Output is appended to the existing file.
Command Default
Text output is written to the screen for interactive runs and to the system output file for batch runs (see the
Basic Analysis Guide).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1168
/OUTPUT
Notes
Text output includes responses to every command and GUI function, notes, warnings, errors, and other in-
formational messages. Upon execution of /OUTPUT,Fname, Ext, ..., all subsequent text output is redirected
to the file Fname.Ext. To redirect output back to the default location, issue /OUTPUT (no arguments).
Note
When using the GUI, output from list operations [NLIST, DLIST, etc.] is always sent to a list window
regardless of the /OUTPUT setting. The output can then be saved on a file or copied to the
/OUTPUT location using the File menu in the list window.
This command is valid in any processor.
Menu Paths
Utility Menu>File>Switch Output to>File
Utility Menu>File>Switch Output to>Output Window
OUTRES, Item, Freq, Cname
Controls the solution data written to the database.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Item
Results item for database and file write control:
ALL
All solution items except SVAR and LOCI. This value is the default.
CINT
J-integral results.
ERASE
Resets OUTRES specifications to their default values.
STAT
Lists the current OUTRES specifications.
BASIC
Write only NSOL, RSOL, NLOAD, STRS, FGRAD, and FFLUX records to the results file and database.
NSOL
Nodal DOF solution.
RSOL
Nodal reaction loads.
V
Nodal velocity (applicable to structural full transient analysis only (ANTYPE,TRANS)).
A
Nodal acceleration (applicable to structural full transient analysis only (ANTYPE,TRANS)).
1169
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OUTRES
ESOL
Element solution (includes all items following):
NLOAD
Element nodal, input constraint, and force loads (also used with the /POST1 commands PRRFOR,
NFORCE, and FSUM to calculate reaction loads).
STRS
Element nodal stresses.
EPEL
Element elastic strains.
EPTH
Element thermal, initial, and swelling strains.
EPPL
Element plastic strains.
EPCR
Element creep strains.
FGRAD
Element nodal gradients.
FFLUX
Element nodal fluxes.
LOCI
Integration point locations.
SVAR
State variables (used only by USERMAT).
MISC
Element miscellaneous data (SMISC and NMISC items of the ETABLE command).
Freq
Specifies how often (that is, at which substeps) to write the specified solution results item. The following
values are valid:
Description Value
Writes the specified results item every nth (and the last) substep of each
load step.
n
Writes up to n equally spaced solutions (for automatic loading). -n
Suppresses writing of the specified results item for all substeps. NONE
Writes the solution of the specified solution results item for every substep.
This value is the default for a harmonic analysis (ANTYPE,HARMIC) and
for any expansion pass (EXPASS,ON).
ALL
Writes the specified solution results item only for the last substep of each
load step.This value is the default for a static (ANTYPE,STATIC) or transient
(ANTYPE,TRANS) analysis.
LAST
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1170
OUTRES
Description Value
Where array is the name of an n x 1 x 1 dimensional array parameter
defining n key times, the data for the specified solution results item is
written at those key times.
Key times in the array parameter must appear in ascending order.
Values must be greater than or equal to the beginning values of the
%ar-
ray%
load step, and less than or equal to the ending time values of the load
step.
For multiple-load-step problems, either change the parameter values
to fall between the beginning and ending time values of the load step
or erase the current settings and reissue the command with a new array
parameter.
For more information about defining array parameters, see the *DIM
command documentation.
Cname
The name of the component, created with the CM command, defining the selected set of elements or
nodes for which this specification is active. If blank, the set is all entities. A component name is not allowed
with the ALL, BASIC, or RSOL items.
Command Default
The OUTRES command writes the specified solution results item for every substep. The exceptions to the
default behavior are as follows:
For static (ANTYPE,STATIC) or transient (ANTYPE,TRANS) analyses, the default is to write the specified
solution results item for the last substep of each load step.
For a harmonic analysis (ANTYPE,HARMIC) and any expansion pass, the default is to write the specified
solution results item for every substep.
For reduced or mode superposition transients, the default is to write the reduced displacements file for
every 4th substep (as well as the last substep); however, if gap conditions exist, the default is to write
the specified solution results item of every substep.
Notes
The OUTRES command allows you to specify the following:
The solution item (Item) to write to the database (and to the reduced displacement and results files)
The frequency (Freq) at which the solution item is written (applicable to static, transient, or full har-
monic analyses)
The set of elements or nodes (Cname) to which your specification applies.
The command generates a specification for controlling data storage for each substep, activating storage of
the specified results item for the specified substeps of the solution and suppressing storage of that item for
all other substeps.
1171
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OUTRES
You can issue multiple OUTRES commands in an analysis. After the initial command creating the storage
specification, subsequent OUTRES commands modify the specification set for each substep. The command
processes your specifications at each substep in the order in which you input them. If you specify a given
solution item twice, output is based upon the last specification. Therefore, issue multiple OUTRES commands
carefully and in the proper sequence.
Besides OUTRES, another output-control command named OUTPR exists which controls solution printout.
You can issue up to 50 output-control commands (either OUTRES or OUTPR, or some combination of both)
in an analysis.
Note
Issuing OUTRES,ERASE erases the existing output specifications and resets the counted number
of OUTRES commands to zero. Issuing OUTPR,ERASE affects the OUTPR command in the same
way.
A given OUTRES command has no effect on results items not specified. For example, an OUTRES,ESOL,LAST
command does not affect NSOL data; that is, it neither activates nor suppresses NSOL data storage in any
substep.
Caution
In the results-item hierarchy, certain items are subsets of other items. For example, element
solution (ESOL) data is a subset of all (ALL) solution data. Therefore, an OUTRES,ALL command
can affect ESOL data. Likewise, an OUTRES command that controls ESOL data can affect a portion
of ALL data. The example OUTRES commands illustrate the interrelationships between results
items and the necessity for employing the OUTRES command thoughtfully.
To suppress ALL data at every substep, issue an OUTRES,ALL,NONE command. (An OUTRES,ERASE command
does not suppress ALL data at every substep.)
The NSOL, RSOL, V, and A solution items are associated with nodes. All remaining solution items are associated
with elements.
The boundary conditions (constraints and force loads) are written to the results file only if either nodal or
reaction loads (NLOAD or RSOL items) are also written.
When specifying a Freq value, observe the following:
For a modal analysis, the only valid values are NONE or ALL.
If you issue multiple OUTRES commands during an analysis, you cannot specify a key time array para-
meter (%array%) in a given OUTRES command and then specify a different Freq option in a sub-
sequent OUTRES command.
For a coupled-field analysis using the ANSYS Multi-field Solver, Freq cannot be NONE for Item = NSOL or
ESOL because interface loads have to be retrieved from the database.
The OUTRES command is also valid in /PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1172
OUTRES
Example
When issuing an OUTRES command, think of a matrix in which you set switches on and off. When a switch
is on, a results item is stored for the specified substep. When a switch is off, a results item is suppressed for
a specified substep.
Assuming a static (ANTYPE,STATIC) analysis, this example shows how the matrix looks after issuing each
OUTRES command in this six-substep solution.
NSUBST,6
OUTRES,ERASE
OUTRES,NSOL,2
OUTRES,ALL,3
OUTRES,ESOL,4
SOLVE
To simplify the example, only a subset of the available solution items appears in the matrix.
OUTRES,ERASE -- After issuing this command, the default output specifications are in effect, as shown:
Results Item Specification
Substep
ALL
BASIC
ESOL RSOL NSOL
EPTH EPEL FGRAD STRS NLOAD
off off off off off off off 1
off off off off off off off 2
off off off off off off off 3
off off off off off off off 4
off off off off off off off 5
ON ON ON ON ON ON ON 6
OUTRES,NSOL,2 -- This command modifies the initial specifications so that NSOL is turned on for substeps
2, 4 and 6, and turned off for substeps 1, 3 and 5, as shown:
Results Item Specification
Substep
ALL
BASIC
ESOL RSOL NSOL
EPTH EPEL FGRAD STRS NLOAD
off off off off off off off 1
off off off off off off ON 2
off off off off off off off 3
off off off off off off ON 4
off off off off off off off 5
ON ON ON ON ON ON ON 6
1173
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OUTRES
OUTRES,ALL,3 -- This command further modifies the specifications so that ALL is turned on for substeps 3
and 6, and turned off for substeps 1, 2, 4 and 5, as shown:
Results Item Specification
Substep
ALL
BASIC
ESOL RSOL NSOL
EPTH EPEL FGRAD STRS NLOAD
off off off off off off off 1
off off off off off off off 2
ON ON ON ON ON ON ON 3
off off off off off off off 4
off off off off off off off 5
ON ON ON ON ON ON ON 6
OUTRES,ESOL,4 -- This command once again modifies the specifications so that ESOL is turned on for the
fourth and last substeps, and turned off for substeps 1, 2, 3 and 5, as shown:
Results Item Specification
Substep
ALL
BASIC
ESOL RSOL NSOL
EPTH EPEL FGRAD STRS NLOAD
off off off off off off off 1
off off off off off off off 2
off off off off off ON ON 3
ON ON ON ON ON off off 4
off off off off off off off 5
ON ON ON ON ON ON ON 6
SOLVE
When obtaining the solution, results data are stored as follows:
Results Items Stored Substep
No data 1
No data 2
NSOL and RSOL data 3
ESOL data 4
No data 5
ALL data 6
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1174
OUTRES
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Basic
Main Menu>Preprocessor>Loads>Load Step Opts>Output Ctrls>DB/Results File
Main Menu>Solution>Analysis Type>Sol'n Controls>Basic
Main Menu>Solution>Load Step Opts>Output Ctrls>DB/Results File
1175
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
OUTRES
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1176
P Commands
PADELE, DELOPT
Deletes a defined path.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
DELOPT
Path delete option (one of the following):
ALL
Delete all defined paths.
NAME
Delete a specific path from the list of path definitions. (Substitute the actual path name for NAME.)
Command Default
Deletes the currently active path.
Notes
Paths are identified by individual path names. To review the current list of path names, issue the command
PATH,STATUS.
This command is valid in the general postprocessor.
Menu Paths
Main Menu>General Postproc>Path Operations>Archive Path>Retrieve>Path from array
Main Menu>General Postproc>Path Operations>Delete Path>All Paths
Main Menu>General Postproc>Path Operations>Delete Path>By Name
Main Menu>Preprocessor>Path Operations>Delete Path>All Paths
Main Menu>Preprocessor>Path Operations>Delete Path>By Name
/PAGE, ILINE, ICHAR, BLINE, BCHAR
Defines the printout and screen page size.
POST1: Listing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ILINE
Number of lines (11 minimum) per "page" or screen. Defaults to 24. Applies to interactive non-GUI to
the screen output only.
1177
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ICHAR
Number of characters (41 to 132) per line before wraparound. Defaults to 80. Applies to interactive non-
GUI to the screen output only.
BLINE
Number of lines (11 minimum) per page. Defaults to 56. Applies to batch mode [/BATCH], diverted
[/OUTPUT], or interactive GUI [/MENU] output.
BCHAR
Number of characters (41 to 240 (system dependent)) per line before wraparound. Defaults to 132. Applies
to batch mode [/BATCH], diverted [/OUTPUT], or interactive GUI [/MENU] output.
Command Default
As defined by the items above.
Notes
Defines the printout page size for batch runs and the screen page size for interactive runs. Applies to the
POST1 PRNSOL, PRESOL, PRETAB, PRRSOL, and PRPATH commands. See the /HEADER command for ad-
ditional controls (page ejects, headers, etc.) that affect the amount of printout. A blank (or out-of-range)
value retains the previous setting. Issue /PAGE,STAT to display the current settings. Issue /PAGE,DEFA to
reset the default specifications.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
PAGET, PARRAY, POPT
Writes current path information into an array variable.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
PARRAY
The name of the array parameter that the ANSYS program creates to store the path information. If the
array parameter already exists, it will be replaced with the current path information.
POPT
Determines how data will be stored in the parameter specified with PARRAY:
POINTS
Store the path points, the nodes (if any), and coordinate system. (For information on defining paths
and path points, see the descriptions of the PATH and PPATH commands.)
TABLE
Store the path data items. (See the PDEF command description for path data items.)
LABEL
Stores path data labels.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1178
PAGET
Notes
Use the PAGET command together with the PAPUT command to store and retrieve path data in array
variables for archiving purposes. When retrieving path information, restore the path points (POINTS option)
first, then the path data (TABLE option), and then the path labels (LABEL option).
Menu Paths
Main Menu>General Postproc>Path Operations>Archive Path>Store>Path in array
PAPUT, PARRAY, POPT
Retrieves path information from an array variable.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
PARRAY
Name of the array variable containing the path information.
POPT
Specifies which path data to retrieve:
POINTS
Retrieve path point information (specified with the PPATH command and stored with the PA-
GET,POINTS command). The path data name will be assigned to the path points.
TABLE
Retrieve path data items (defined via the PDEF command and stored with the PAGET,,TABLE com-
mand).
LABEL
Retrieve path labels stored with the PAGET,,LABEL command.
Notes
When retrieving path information, restore path points (POINTS option) first, then the path data (TABLE option),
and then the path labels (LABEL option).
Menu Paths
Main Menu>General Postproc>Path Operations>Archive Path>Retrieve>Path from array
PARESU, Lab, Fname, Ext, --
Restores previously saved paths from a file.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Read operation:
1179
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PARESU
ALL
Read all paths from the selected file (default).
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to PATH if Fname is blank.
--
Unused field.
Notes
This command removes all paths from virtual memory and then reads path data from a file written with the
PASAVE command. All paths on the file will be restored. All paths currently in memory will be deleted.
Menu Paths
Main Menu>General Postproc>Path Operations>Archive Path>Retrieve>Paths from file
Main Menu>Preprocessor>Path Operations>Retrieve>Paths from file
PARTSEL, Type, PMIN, PMAX, PINC
Selects a subset of parts in an explicit dynamic analysis.
DATABASE: Selecting
<> <> <> <> <> <> <> <> <> <> DY PP <> <> <>
Type
Label identifying type of select. Because PARTSEL is a command macro, the label must be enclosed in
single quotes.
'S'
Select a new set (default).
'R'
Reselect a set from the current set.
'A'
Additionally select a set and extend the current set.
'U'
Unselect a set from the current set.
'ALL'
Select all parts.
'NONE'
Unselect all parts.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1180
PARTSEL
'INVE'
Invert the current selected set.
The following fields are used only with Type = 'S', 'R', 'A', or 'U':
PMIN
Minimum part number in the range to be selected.
PMAX
Maximum part number in the range to be selected (defaults to PMIN).
PINC
Part number increment in the range to be selected (default =1).
Notes
PARTSEL invokes an ANSYS macro that selects parts in an explicit dynamic analysis. When PARTSEL is ex-
ecuted, an element component is automatically created for each existing part. For example, the elements
that make up PART 1 are grouped into the element component _PART1. Each time the PARTSEL command
is executed, components for unselected parts will be unselected. To plot selected parts, choose Utility
Menu> Plot> Parts in the GUI or issue the command PARTSEL,'PLOT'.
After selecting parts, if you change the selected set of nodes or elements and then plot parts, the nodes
and elements associated with the previously selected parts (from the last PARTSEL command) will become
the currently selected set.
Note
A more efficient way to select and plot parts is to use the ESEL (with ITEM = PART) and EPLOT
commands. We recommend using ESEL instead of PARTSEL since PARTSEL will be phased out
in a future release. Note that the menu path mentioned above for plotting parts does not work
with the ESEL command; use Utility Menu> Plot> Elements instead.
In an explicit dynamic small restart analysis (EDSTART,2), PARTSEL can be used to unselect a part during
the solution even if it is referenced in some way (such as in a contact definition). (Note that ESEL cannot
be used for this purpose.) However, in a new analysis or a full restart analysis (EDSTART,3), all parts that are
used in some type of definition must be selected at the time of solution.
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Utility Menu>Select>Select Parts
1181
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PARTSEL
PARRES, Lab, Fname, Ext, --
Reads parameters from a file.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Lab
Read operation:
NEW --
Replace current parameter set with these parameters (default).
CHANGE --
Extend current parameter set with these parameters, replacing any that already exist.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to PARM if Fname is blank.
--
Unused field.
Notes
Reads parameters from a coded file. The parameter file may have been written with the PARSAV command.
The parameters read may replace or change the current parameter set.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Restore Parameters
PARSAV, Lab, Fname, Ext, --
Writes parameters to a file.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Lab
Write operation:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1182
PARRES
SCALAR --
Write only scalar parameters (default).
ALL --
Write scalar and array parameters. Parameters may be numeric or alphanumeric.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to PARM if Fname is blank.
--
Unused field.
Notes
Writes the current parameters to a coded file. Previous parameters on this file, if any, will be overwritten.
The parameter file may be read with the PARRES command.
PARSAV/PARRES operations truncate some long decimal strings, and can cause differing values in your
solution data when other operations are performed. A good practice is to limit the number of decimal places
you will use before and after these operations.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Save Parameters
PASAVE, Lab, Fname, Ext, --
Saves selected paths to an external file.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Write operation:
S
Saves only selected paths.
ALL
Saves all paths (default).
Pname
Saves the named path (from the PSEL command).
1183
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PASAVE
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to PATH if Fname is blank.
--
Unused field.
Notes
Saves the paths selected with the PSEL command to an external file (Jobname.path by default). Previous
paths on this file, if any, will be overwritten. The path file may be read with the PARESU command.
This command is valid in POST1.
Menu Paths
Main Menu>General Postproc>Path Operations>Archive Path>Store>Paths in file
Main Menu>Preprocessor>Path Operations>Store>Paths in file
PATH, NAME, nPts, nSets, nDiv
Defines a path name and establishes parameters for the path.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NAME
Name for this path (eight characters maximum. If nPts is blank, set the current path to the path with
this name. If nPts is greater than zero, create a path of this name. If a path with this name already exists,
replace it with a new path. If the NAME value is STATUS, display the status for path settings.
nPts
The number of points used to define this path. The minimum number is two, and the maximum is 1000.
Default is 2.
nSets
The number of sets of data which you can map to this path. You must specify at least four: X, Y, Z, and
S. Default is 30.
nDiv
The number of divisions between adjacent points. Default is 20. There is no maximum number of divisions.
Notes
The PATH command is used to define parameters for establishing a path. The path geometry is created by
the PPATH command. Multiple paths may be defined and named; however, only one path may be active
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1184
PATH
for data interpolation [PDEF] and data operations [PCALC, etc.]. Path geometry points and data are stored
in memory while in POST1. If you leave POST1, the path information is erased. Path geometry and data may
be saved in a file by archiving the data using the PASAVE command. Path information may be restored by
retrieving the data using the PARESU command.
For overlapping nodes, the lowest numbered node is assigned to the path.
The number of divisions defined using nDiv does NOT affect the number of divisions used by PLSECT and
PRSECT.
For information on displaying paths you have defined, see the Basic Analysis Guide.
Menu Paths
Main Menu>General Postproc>List Results>Path Items
Main Menu>General Postproc>Path Operations>Archive Path>Retrieve>Path from array
Main Menu>General Postproc>Path Operations>Archive Path>Retrieve>Paths from file
Main Menu>General Postproc>Path Operations>Define Path>By Location
Main Menu>General Postproc>Path Operations>Define Path>By Nodes
Main Menu>General Postproc>Path Operations>Define Path>On Working Plane
Main Menu>General Postproc>Path Operations>Define Path>Path Status>Defined Paths
Main Menu>General Postproc>Path Operations>Delete Path>All Paths
Main Menu>General Postproc>Path Operations>Delete Path>By Name
Main Menu>General Postproc>Path Operations>Recall Path
Main Menu>Preprocessor>Path Operations>Define Path>By Location
Main Menu>Preprocessor>Path Operations>Define Path>By Nodes
Main Menu>Preprocessor>Path Operations>Define Path>On Working Plane
Main Menu>Preprocessor>Path Operations>Define Path>Path Status>Defined Paths
Main Menu>Preprocessor>Path Operations>Delete Path>All Paths
Main Menu>Preprocessor>Path Operations>Delete Path>By Name
Main Menu>Preprocessor>Path Operations>Recall Path
Main Menu>Preprocessor>Path Operations>Retrieve>Paths from file
Utility Menu>List>Status>General Postproc>Path Operations
PAUSE
Temporarily releases the current product license.
SESSION: Run Controls
MP ME ST PR PRN DS DSS FL EM EH DY PP VT EME MFS
Notes
The PAUSE command temporarily releases (or pauses) the current product license so that another application
can use it.
This application consumes a license as soon as you launch it, and retains that license until it is finished. If
you launch the product interactively, the license is retained until you either close the application or issue a
PAUSE command via the command line.
1185
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PAUSE
No other operation (other than SAVE or /EXIT) is possible in the current application while use of the product
license is paused.
When the second application has finished and releases the license, issue an UNPAUSE command via the
command line to restore use of the license to the current application.
For more information, see the ANSYS, Inc. Licensing Guide.
Menu Paths
This command cannot be accessed from a menu.
/PBC, Item, --, KEY, MIN, MAX, ABS
Shows boundary condition (BC) symbols and values on displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Label identifying the item:
U
Applied translational constraints (UX, UY, UZ).
ROT
Applied rotational constraints (ROTX, ROTY, ROTZ).
TEMP
Applied temperatures (TEMP, TBOT, TE2, TE3, . . ., TTOP).
PRES
Applied fluid pressures.
V
Applied flow velocities (VX, VY, VZ).
SP0n
Mass fraction of species n, where n = 1 to 6 (FLOTRAN). If a species is given a user-defined name
[MSSPEC], use that name instead of SP0n.
ENKE
Turbulent kinetic energy (FLOTRAN).
ENDS
Turbulent energy dissipation (FLOTRAN).
VOLT
Applied voltages.
MAG
Applied scalar magnetic potentials.
A
Applied vector magnetic potentials.
CHRG
Applied electric charge.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1186
/PBC
F or FORC
Applied structural forces (FX, FY, FZ).
M or MOME
Applied structural moments (MX, MY, MZ).
HEAT
Applied heat flows (HEAT, HBOT, HE2, HE3, . . ., HTOP).
FLOW
Applied fluid flow.
AMPS
Applied current flow.
FLUX
Applied magnetic flux.
CSG
Applied magnetic current segments.
MAST
Master degrees of freedom.
CP
Coupled nodes.
CE
Nodes in constraint equations.
NFOR
POST1 nodal forces.
NMOM
POST1 nodal moments
RFOR
POST1 reaction forces.
RMOM
POST1 reaction moments (MX, MY, MZ).
PATH
Path geometry (undistorted) associated with the PATH command after a PDEF or PVECT command
has been issued.
ACEL
Global acceleration (ACELX, ACELY, ACELZ vector).
OMEG
Global angular velocity (OMEGX, OMEGY, OMEGZ vector) and acceleration (DOMEGX, DOMEGY,
DOMEGZ vector).
WELD
Applied spotwelds (ANSYS LS-DYNA).
ALL
Represents all appropriate labels.
--
Unused field.
1187
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PBC
KEY
Symbol key:
0
Do not show symbol.
1
Show symbol.
2
Plot value next to symbol.
MIN
Minimum value in a range of values plotted on screen.
MAX
Maximum value in a range of values plotted on screen.
ABS
Absolute number. If KEY = 2 and ABS = 0, a number falling between the MIN and MAX is displayed. If
ABS is not specified, it defaults to 0. If KEY = 2 and ABS = 1, an absolute value falling between the MIN
and MAX is displayed. ABS = 1 lets you eliminate the display of numbers whose absolute values are less
than a desired tolerance. For example, if ABS = 1, MIN = 10 and MAX = 1e8, values such as .83646 and
-5.59737 are not displayed.
Command Default
No symbols displayed.
Notes
The /PBC command adds degree of freedom constraint, force load, and other symbols to displays.
Symbols are applied to the selected nodes only. All arrow and arrowhead symbols are oriented in the nodal
coordinate system and lie in two perpendicular planes. Force arrows are scaled proportional to their mag-
nitude. (If KEY = 1, use /VSCALE to change arrow length.) For scalar quantities, the specific component
direction (i.e., x, y, or z) of the symbol has no meaning, but the positive or negative sense (e.g., positive or
negative x) represents a positive or negative scalar value, respectively.
The effects of the /PBC command are not cumulative (that is, the command does not modify an existing
setting from a previously issued /PBC command). If you issue multiple /PBC commands during an analysis,
only the setting specified by the most recent /PBC command applies.
Use /PSTATUS or /PBC,STAT to display settings. Use /PBC,DEFA to reset all specifications back to default.
See the /PSF and /PBF commands for other display symbols.
In a cyclic symmetry analysis, the /PBC command is deactivated when cyclic expansion is active (/CYCEX-
PAND,,ON). To view boundary conditions on the basic sector, deactivate cyclic expansion (/CYCEXPAND,,OFF)
and issue this command: /PBC,ALL,,1
Issuing the command /PBC,PATH, ,1 displays all defined paths.
The /PBC command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1188
/PBC
Menu Paths
Main Menu>General Postproc>Path Operations>Define Path>On Working Plane
Main Menu>General Postproc>Path Operations>Plot Paths
Main Menu>Preprocessor>Path Operations>Define Path>On Working Plane
Main Menu>Preprocessor>Path Operations>Plot Paths
Utility Menu>PlotCtrls>Symbols
/PBF, Item, --, KEY
Shows magnitude of body force loads on displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Label identifying the item:
TEMP
Applied temperatures.
FLUE
Applied fluences.
HGEN
Applied heat generation rates.
JS
Applied current density magnitude.
JSX
X-component of current density.
JSY
Y-component of current density.
JSZ
Z-component of current density.
PHASE
Phase angle of applied load.
MVDI
Applied magnetic virtual displacements flag.
CHRGD
Applied electric charge density.
VLTG
Applied voltage drop.
FORC
Applied force density (FLOTRAN only).
--
Unused field.
KEY
Symbol key:
1189
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PBF
0
Do not show body force load contours.
1
Show body force load contours.
2
Show current density as a vector (not a contour).
Command Default
No body force load contours displayed.
Notes
Shows body force loads as contours on displays for the selected elements.
The effects of the /PBF command are not cumulative (that is, the command does not modify an existing
setting from a previously issued /PBF command). If you issue multiple /PBF commands during an analysis,
only the setting specified by the most recent /PBF command applies.
Use /PSTATUS or /PBF,STAT to display settings. Use /PBF,DEFA to reset all specifications back to default.
See also the /PSF and /PBC command for other display contours.
Portions of this command are not supported by PowerGraphics [/GRAPHICS,POWER].
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Symbols
PCALC, Oper, LabR, Lab1, Lab2, FACT1, FACT2, CONST
Forms additional labeled path items by operating on existing path items.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Oper
Type of operation to be performed. See "Notes" (p. 1191) below for specific descriptions of each operation:
ADD
Adds two existing path items.
MULT
Multiplies two existing path items.
DIV
Divides two existing path items (a divide by zero results in a value of zero).
EXP
Exponentiates and adds existing path items.
DERI
Finds a derivative.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1190
PCALC
INTG
Finds an integral.
SIN
Sine.
COS
Cosine.
ASIN
Arcsine.
ACOS
Arccosine.
LOG
Natural log.
LabR
Label assigned to the resulting path item.
Lab1
First labeled path item in operation.
Lab2
Second labeled path item in operation. Lab2 must not be blank for the MULT, DIV, DERI, and INTG op-
erations.
FACT1
Factor applied to Lab1. A (blank) or '0' entry defaults to 1.0.
FACT2
Factor applied to Lab2. A (blank) or '0' entry defaults to 1.0.
CONST
Constant value (defaults to 0.0).
Notes
If Oper = ADD, the command format is:
PCALC,ADD,LabR,Lab1,Lab2,FACT1,FACT2,CONST
This operation adds two existing path items according to the operation:
LabR = (FACT1 x Lab1) + (FACT2 x Lab2) + CONST
It may be used to scale the results for a single path item.
If Oper = MULT, the command format is:
PCALC,MULT,LabR,Lab1,Lab2,FACT1
Lab2 must not be blank. This operation multiplies two existing path items according to the operation:
LabR = Lab1 x Lab2 x FACT1
If Oper = DIV, the command format is:
1191
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PCALC
PCALC,DIV,LabR,Lab1,Lab2,FACT1
Lab2 must not be blank. This operation divides two existing path items according to the operation:
LabR = (Lab1/Lab2) x FACT1
If Oper = EXP, the command format is:
PCALC,EXP,LabR,Lab1,Lab2,FACT1,FACT2
This operation exponentiates and adds existing path items according to the operation:
LabR = (|Lab1|
FACT1
) + (|Lab2|
FACT2
|)
If Oper = DERI, the command format is:
PCALC,DERI,LabR,Lab1,Lab2,FACT1
Lab2 must not be blank. This operation finds a derivative according to the operation:
LabR = FACT1 x d(Lab1)/d(Lab2)
If Oper = INTG, the command format is:
PCALC,INTG,LabR,Lab1,Lab2,FACT1
Lab2 must not be blank. This operation finds an integral according to the operation:
LabR FACT Lab d Lab
s
1 1 2 ( )
Use S for Lab2 to integrate Lab1 with respect to the path length. S, the distance along the path, is auto-
matically calculated by the program when a path item is created with the PDEF command.
If Oper = SIN, COS, ASIN, ACOS, or LOG, the command format is:
PCALC,Oper,LabR,Lab1,,FACT1,CONST
where the function (SIN, COS, ASIN, ACOS or LOG) is substituted for Oper and Lab2 is blank.
The operation finds the resulting path item according to one of the following formulas:
LabR = FACT2 x sin(FACT1 x Lab1) + CONST
LabR = FACT2 x cos(FACT1 x Lab1) + CONST
LabR = FACT2 x sin
-1
(FACT1 x Lab1) + CONST
LabR = FACT2 x cos
-1
(FACT1 x Lab1) + CONST
LabR = FACT2 x log(FACT1 x Lab1) + CONST
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1192
PCALC
Menu Paths
Main Menu>General Postproc>Path Operations>Add
Main Menu>General Postproc>Path Operations>ArcCosine
Main Menu>General Postproc>Path Operations>ArcSine
Main Menu>General Postproc>Path Operations>Cosine
Main Menu>General Postproc>Path Operations>Differentiate
Main Menu>General Postproc>Path Operations>Divide
Main Menu>General Postproc>Path Operations>Exponentiate
Main Menu>General Postproc>Path Operations>Integrate
Main Menu>General Postproc>Path Operations>Multiply
Main Menu>General Postproc>Path Operations>Natural Log
Main Menu>General Postproc>Path Operations>Sine
PCGOPT, Lev_Diff , --, ReduceIO, StrmCk, Wrtfull, Memory
Controls PCG solver options.
SOLUTION: Analysis Options
MP ME ST PR PRN DS DSS <> EM <> <> PP <> EME MFS
Lev_Diff
Indicates the level of difficulty of the analysis. Valid settings are AUTO or 0 (default), 1, 2, 3, 4, or 5. This
option applies to both the PCG solver when used in static and full transient analyses and to the PCG
Lanczos method in modal analyses. Use AUTO to let ANSYS automatically choose the proper level of
difficulty for the model. Lower values (1 or 2) generally provide the best performance for well-conditioned
problems. Values of 3 or 4 generally provide the best performance for ill-conditioned problems; however,
higher values may increase the solution time for well-conditioned problems. Higher level-of-difficulty
values typically require more memory. Using the highest value of 5 essentially performs a factorization
of the global matrix (similar to the sparse solver) and may require a very large amount of memory. If
necessary, use Memory to reduce the memory usage when using Lev_Diff = 5. Lev_Diff = 5 is
generally recommended for small- to medium-sized problems when using the PCG Lanczos mode extrac-
tion method.
For example, models containing elongated elements (i.e., elements with high aspect ratios) and models
containing contact elements can lead to ill-conditioned problems. To determine if your problem is ill-
conditioned, view the Jobname.PCS file to see the number of PCG iterations needed to reach a con-
verged solution. Generally, static or full transient solutions that require more than 1500 iterations are
considered to be ill-conditioned for the PCG solver.
--
Unused field.
ReduceIO
Controls whether the PCG solver will attempt to reduce I/O performed during equation solution:
AUTO
Automatically chooses whether to reduce I/O or not (default).
YES
Reduces I/O performed during equation solution in order to reduce total solver time.
NO
Does NOT reduce I/O performed during equation solution.
1193
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PCGOPT
This option applies to both the PCG solver when used in static and full transient analyses and to the
PCG Lanczos method in modal analyses.
StrmCk
Controls whether or not a Sturm sequence check is performed:
OFF
Does NOT perform Sturm sequence check (default).
ON
Performs Sturm sequence check
This option applies only when using the PCG Lanczos method in modal analyses. When using this option,
a factorization must be performed and will require a very large amount of memory for extra computations.
This option is generally recommended for small- to medium-sized problems. If the Sturm sequence check
takes a large amount of computing time, use the Jobname.ABT file to abort the Sturm check, or press
the STOP button if in interactive mode.
Wrtfull
Controls whether or not the .FULL file is written.
ON
Write .FULL file (default)
OFF
Do not write .FULL file.
This option applies only when using the PCG Lanczos method in modal analyses because the .FULL
file is never written when using the PCG solver in static or full transient analyses. If using MSAVE,ON
and conditions for the MSAVE command are met, a complete .FULL file is never written, regardless of
this option.
This option is useful in Distributed ANSYS because assembling the global stiffness and mass matrices
on the master machine before writing the .FULL file can take a considerable amount of memory. By
setting Wrtfull = OFF, this assembly process is skipped on the master machine, decreasing the amount
of memory required to compute the modes and mode shapes. Wrtfull = OFF does not affect the
results for the modes and mode shapes. However, without a .FULL file, the participation factor table
computations do not occur.
To generate the .FULL file, such as for a harmonic, transient mode superposition, or spectrum analysis,
rerun the modal analysis with Wrtfull = ON, or use the WRFULL command.
Memory
Controls whether to run using in-core or out-of-core mode when using Lev_Diff = 5.
AUTO
Automatically chooses which mode to use. (default).
INCORE
Run using in-core mode.
OOC
Run using out-of-core mode.
This option applies to both the PCG solver when used in static and full transient analyses and to the
PCG Lanczos method in modal analyses. However, this option only applies when using Lev_Diff =
5 and does not apply to Distributed ANSYS. Forcing the INCORE option will require a large amount of
memory and is only recommended for small-to-medium sized problems. Using the OOC option will make
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1194
PCGOPT
the PCG solver use the least amount of memory (relative only to Lev_Diff = 5) at the cost of addi-
tional I/O, which will likely decrease performance as compared to an in-core run.
Notes
ReduceIO works independently of the MSAVE command in the PCG solver. Setting ReduceIO to YES can
significantly increase the memory usage in the PCG solver.
To minimize the memory used by the PCG solver with respect to the Lev_Diff option only, set
Lev_Diff = 1 if you do not have sufficient memory to run the PCG solver with Lev_Diff = AUTO.
The MSAVE,ON command is not valid when using Lev_Diff = 5. In this case, the Lev_Diff value will
automatically be reset to 2. The MSAVE,ON command is also not valid with the StrmCk option. In this
case, StrmCk will be set to OFF.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
PCIRC, RAD1, RAD2, THETA1, THETA2
Creates a circular area centered about the working plane origin.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RAD1, RAD2
Inner and outer radii (either order) of the circle. A value of either zero or blank for either RAD1 or RAD2,
or the same value for both RAD1 and RAD2, defines a solid circle.
THETA1, THETA2
Starting and ending angles (either order) of the circular area. Used for creating a circular sector. The
sector begins at the algebraically smaller angle, extends in a positive angular direction, and ends at the
larger angle. The starting angle defaults to 0.0 and the ending angle defaults to 360.0. See the Modeling
and Meshing Guide for an illustration.
Notes
Defines a solid circular area or circular sector centered about the working plane origin. For a solid circle of
360, the area will be defined with four keypoints and four lines. See the CYL4 and CYL5 commands for al-
ternate ways to create circles.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Areas>Circle>By Dimensions
1195
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PCIRC
/PCIRCLE, XCENTR, YCENTR, XLRAD
Creates an annotation circle (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XCENTR
Circle X center location (-1.0 < X < 2.0).
YCENTR
Circle Y center location (-1.0 < Y < 1.0).
XLRAD
Circle radius length.
Notes
Creates an annotation circle to be written directly onto the display at a specified location. This is a command
generated by the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation
is used. This command is not intended to be typed in directly in an ANSYS session (although it can be included
in an input file for batch input or for use with the /INPUT command).
All circles are shown on subsequent displays unless the annotation is turned off or deleted. Use the /LSPEC
and the /PSPEC command to set the attributes of the circle.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
PCONV, TOLER, Item, Comp, NODE, Surf
Sets convergence values for p-method solutions.
SOLUTION: p-Method
MP ME ST PR PRN <> <> FL EM <> <> PP <> EME MFS
TOLER
Tolerance (as a percentage) used for convergence checking. The analysis is considered to be converged
if the values of all specified items change by less than TOLER at the end of a p-loop. Defaults to 5. If
TOLER = STAT, list the currently specified p-convergence criteria. If TOLER = ERASE, erase all p-conver-
gence specifications. If TOLER = DELE, delete only the tolerance for the specification denoted by the
Item, Comp, NODE.
Item
Convergence based on (Valid component labels are described in Table 241: PCONV - Valid Item and
Component Labels (p. 1198) below):
SE
Strain energy or stored electrostatic energy (default). Does not require a component label.
S
Stress.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1196
/PCIRCLE
EPEL
Strain.
U
Translational displacement.
ROT
Structural rotation. (Valid only for SHELL150.)
VOLT
Electric potential.
EF
Electric field strength.
D
Electric flux density.
EFORC
Global electrostatic (Maxwell Stress Tensor) force.
Comp
Component of the Item. (Not required for Item = SE, VOLT, or EFORC). If Item = S or EPEL, Comp defaults
to EQV; if Item = U, ROT, EF, or D, Comp defaults to SUM. Valid component labels for each Item are
described in Table 241: PCONV - Valid Item and Component Labels (p. 1198) below.
NODE
Node number where convergence checking is to be performed. Valid only for Item = S, EPEL, U, ROT,
VOLT, EF, or D. For valid Item labels, a node number must be entered in this field. If NODE = P, graph-
ical picking is enabled (valid only in the GUI). A component name may be substituted for NODE.
Surf
Surface of a p-element shell structure where the convergence checking for NODE will be performed.
Valid only for Item = S or EPEL.
TOP
Top surface (default).
MID
Middle surface.
BOT
Bottom surface.
Command Default
Convergence is checked between successive p-loops by comparing the change in strain energy (SE) to
TOLER.
Notes
When you use the menu paths to delete p-method convergence criteria, the ANSYS program writes the
command PCONV,DELE,n to its log file.
The p-level (polynomial level, i.e., the order or level n of a polynomial, where 2
8) is increased for
each loop until the values of the specified items (Item,Comp) change less than the specified tolerance
(TOLER) between loops, or until the maximum p-level has been reached [PPRANGE]. Convergence checking
1197
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PCONV
is performed only on included elements [PINCLUDE]. Up to 10 separate specifications may be activated at
one time. Valid Item and Comp labels are shown next.
Table 241 PCONV - Valid Item and Component Labels
Description Comp Item
Strain energy or stored electrostatic energy SE
Component stress X, Y, Z, XY, YZ, XZ S
Principal stress 1, 2, 3 "
Stress intensity INT "
Equivalent stress EQV "
Component elastic strain X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain 1, 2, 3 "
Elastic strain intensity INT "
Elastic equivalent strain EQV "
X, Y, or Z structural displacement or vector sum X, Y, Z, SUM U
X, Y, or Z structural rotation or vector sum (valid only for
SHELL150)
X, Y, Z, SUM ROT
Electric potential VOLT
Electric field strength X, Y, Z, SUM EF
Electric flux density X, Y, Z, SUM D
Global electrostatic (Maxwell Stress Tensor) force X, Y, Z, SUM EFORC
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Convergence Crit
Main Menu>Solution>Load Step Opts>p-Method>Convergence Crit
/PCOPY, KEY
Automatically generates hard copies for HP UNIX work stations.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Copy key:
0
No specification setting for automatic hard copy of display.
1
Set specification for automatic hard copy after each display.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1198
/PCOPY
NOW
(Action) Produce hard copy of current display (KEY is not reset to 1).
Command Default
No automatic hard copy of display.
Notes
Sets automatic hard copy specification. This command is available only on HP work stations, and only during
interactive runs with the /SHOW specification active (for terminals with hard copy capability).
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
PCORRO, CTK
Specifies the allowable exterior corrosion thickness for a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
CTK
Allowable corrosion thickness.
Notes
Specifies the allowable exterior corrosion thickness for a piping run. See the PREP7 RUN command.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Specifications
PCROSS, LabXR, LabYR, LabZR, LabX1, LabY1, LabZ1, LabX2, LabY2, LabZ2
Calculates the cross product of two path vectors along the current path.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabXR
Label assigned to X-component of resultant vector.
LabYR
Label assigned to Y-component of resultant vector.
LabZR
Label assigned to Z-component of resultant vector.
1199
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PCROSS
LabX1
X-component of first vector label (labeled path item).
LabY1
Y-component of first vector label.
LabZ1
Z-component of first vector label.
LabX2
X-component of second vector label (labeled path item).
LabY2
Y-component of second vector label.
LabZ2
Z-component of second vector label.
Menu Paths
Main Menu>General Postproc>Path Operations>Cross Product
PDANL, Fname, Ext, --
Defines the analysis file to be used for probabilistic looping.
PROBABILISTIC: Deterministic Model
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
Command Default
If interactive, no default; if batch, use the batch input stream (File.BAT).
Notes
The analysis file containing the deterministic, parameterized model must be specified if the probabilistic
analysis is performed interactively. The file must exist at the time the PDANL command is issued. In this file,
where /PREP7 and /PDS occur, they must be the first nonblank characters on the line (do not use the $
delimiter on any /PREP7 and /PDS command lines).
By default the analysis files specified with PDANL are executed from the top. All definitions of random input
variables (for example, APDL parameters defined as random input variables with the PDVAR command, using
*SET or Pname = ... ) are ignored in the analysis file. The PDS takes control of setting the values of random
input variable values for each loop.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1200
PDANL
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Prob Design>Analysis File>Assign
PDCDF, Rlab, Name, Type, CONF, NMAX
Plots the cumulative distribution function.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo
Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface
Analyses).
Name
Parameter name. The parameter must have been previously defined as a random input variable or a
random output parameter with the PDVAR command.
Type
Type of cumulative distribution curve to be displayed.
EMP
Show an empirical cumulative distribution curve.
GAUS
Show a cumulative distribution curve in a normal distribution plot. A random variable based on a
normal or Gaussian distribution appears as a straight line in this type of plot.
LOGN
Show a cumulative distribution curve in a log-normal plot. A random variable based on a log-normal
distribution appears as a straight line in this type of plot.
WEIB
Show a cumulative distribution curve in a Weibull distribution plot. A random variable based on a
Weibull distribution appears as a straight line in this type of plot.
CONF
Confidence level. The confidence level is used to plot confidence bounds around the cumulative distri-
bution function. The value for the confidence level must be between 0.0 and 1.0 and it defaults to 0.95
(95%). Plotting of the confidence bound is suppressed for CONF
ij
= 1 for i = j
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1202
PDCFLD
ij
= 0 for i j
Here,
ij
is the correlation coefficient between the i-th and j-th selected FE entity (node or element
centroid).
LEXP
Calculate the correlation coefficient according to a linear-exponential decay function.
ij
i j
L
D x x
C
_
,
exp
({ }, { })
Here, D({x
i
} , {x
j
}) is the domain distance between {x
i
}, {x
j
}, and {x
i
} and {x
j
} are the coordinate vectors
of the i-th and j-th selected FE entity (node or element centroid), and C
L
is the correlation length of
the random field as specified in the CLENGTH field.
QEXP
The correlation coefficient is calculated according to a quadratic- exponential decay function.
ij
i j
L
D x x
C
_
,
_
,
exp
({ }, { })
2
Here, D({x
i
} , {x
j
}) is the domain distance between {x
i
}, {x
j
}, and {x
i
} and {x
j
} are the coordinate vectors
of the i-th and j-th selected FE entity (node or element centroid), and C
L
is the correlation length of
the random field as specified in the CLENGTH field.
DIST
Calculate only D({x
i
} , {x
j
}). D({x
i
} , {x
j
}) is the domain distance between {x
i
}, {x
j
} , and {x
i
} and {x
j
} are
the coordinate vectors of the i-th and j-th selected FE entity (node or element centroid). The CLENGTH
argument is ignored for this option.
CLENGTH
Correlation length of the correlation field. The correlation length is a characteristic length that influences
how strongly two elements of a random field are correlated with each other. The larger the value of
CLENGTH, the stronger the correlation between the random field elements. CLENGTH is required for
Ctype = LEXP and Ctype = QEXP; it must be a nonzero, positive number.
Notes
Calculates a correlation field for a probabilistic analysis involving a random field. Random fields are random
effects with a spatial distribution; the value of a random field not only varies from simulation to simulation
at any given location, but also from location to location. The correlation field describes the correlation
coefficient between two different spatial locations. Random fields can be either based on element properties
(typically material) or nodal properties (typically surface shape defined by nodal coordinates). Hence, random
fields are either associated with the selected nodes or the selected elements. If a random field is associated
with elements, then the correlation coefficients of the random field are calculated based on the distance of
the element centroids.
For more information, see Probabilistic Design in the Advanced Analysis Techniques Guide.
Note that for correlation fields, the domain distance D({x
i
} , {x
j
}) is not the spatial distance |{x
i
} - {x
j
}|, but
the length of a path between {x
i
} and {x
j
} that always remains inside the finite element domain. However,
1203
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDCFLD
exceptions are possible in extreme meshing cases. For elements that share at least one node, the PDCFLD
evaluates the distance by directly connecting the element centroids with a straight line. If these neighboring
elements form a sharp inward corner then it is possible that the domain distance path lies partly outside
the finite element domain, as illustrated below.
After the correlation coefficients have been calculated and stored in the ANSYS parameter ParR, then use
the PDCORR command to define the correlations between the elements of the random field.
Menu Paths
Main Menu>Prob Design>Prob Definitns>Correl Field
PDCLR, Type
Clears the probabilistic design database.
PROBABILISTIC: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Specifies the part of the probabilistic database to be cleared.
ALL
Clear the entire probabilistic database. Both the preprocessing and postprocessing parts are cleared.
POST
Clear only the postprocessing part of the probabilistic database. This is necessary if you want to
modify the deterministic model (the analysis file) or probabilistic model (random variables, correlations
between random variables, or the random output parameter) after a probabilistic analysis has already
been performed.
Notes
Clears the probabilistic design system (PDS) database. The settings are reset to their default values and the
memory is cleared. Remember that the result files containing the results of the probabilistic loops are never
deleted unless you intentionally delete them. We recommend that you use this command before switching
to a new probabilistic analysis using a different probabilistic model (a different analysis loop file or determ-
inistic model), or changing random input variables, their correlations, or random output parameters. Clearing
the probabilistic database is not necessary if the probabilistic model remains the same and will be analyzed
with a different probabilistic method or different method options. See the PDEXE command for restrictions.
Before issuing the PDCLR command, you should save the probabilistic database using the PDSAVE command.
Menu Paths
Main Menu>Prob Design>Prob Database>Clear & Reset
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1204
PDCLR
PDCMAT, Rlab, Matrix, Name1, Name2, Corr, SLEVEL, Popt
Prints the correlation coefficient matrix.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo
Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface
Analyses).
Matrix
Keyword for the type of correlation coefficient matrix.
IO
Matrix of correlation coefficients between random input variables and output parameters.
II
Matrix of correlation coefficients between random input variables and other random input variables
OO
Matrix of correlation coefficients between random output parameters and other random output
parameters.
S
Correlation coefficient between a single random parameter (input or output) and another random
parameter (input or output). The probabilistic design parameters must be specified in Name1 and
Name2 for this option.
Name1, Name2
Parameter names. The parameters must have been previously defined as a random input variable or a
random output parameter with the PDVAR command. These parameters are used for Matrix = S only
and are ignored for the other Matrix keywords.
Corr
Keyword for the type of correlation coefficients to be used for the output.
RANK
Spearman rank-order correlation coefficient (default).
LIN
Pearson linear correlation coefficient.
SLEVEL
Significance level. The value for the significance level must be between 0.0 and 1.0. The default value is
0.025 (2.5%).
Popt
Specifies whether the probabilities should be printed with the correlation coefficients.
0
Print only the correlation coefficients.
1
Print both the correlation coefficients and the probabilities (default).
1205
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDCMAT
Command Default
Rlab as described above, Matrix = IO, Corr = RANK, SLEVEL = 0.025, Popt = 1
Notes
Prints the correlation coefficient matrix.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
For all correlation coefficients the probabilistic design tool evaluates the probability that the correlation
coefficient can be neglected. The evaluation of this probability is based on statistical test theory. The larger
this probability is the likelier it is that the correlation coefficient does not really reflect an observable statist-
ical interdependence between the parameters involved. If this probability exceeds the significance level as
specified by the SLEVEL parameter, the correlation coefficient should be regarded as negligible or insigni-
ficant. The higher the significance level SLEVEL, the more correlation coefficients are considered significant.
Using the Popt parameter you can also get a list of the probabilities and review them as to how far they
exceed the significance level or how far they stay below it.
The PDCMAT command cannot be used to postprocess the results in a solution set that is based on Response
Surface Methods, only Monte Carlo Simulations.
Menu Paths
Main Menu>Prob Design>Prob Results>Trends>Correl Matrix
PDCORR, Name1, Name2, CORR
Specifies the correlation between two random input variables.
PROBABILISTIC: Preprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name1
Parameter name. The parameter must have been previously defined as a random input variable with
the PDVAR command.
Name2
Parameter name. The parameter must have been previously defined as a random input variable with
the PDVAR command. Must be different from Name1.
CORR
Specification of the correlation:
Value
Sets the correlation coefficient between Name1 and Name2 to the specified value. If this correlation
coefficient was already defined it will be changed to this new value. The correlation coefficient must
be between -1.0 and +1.0.
DEL
Delete the previously defined correlation between Name1 and Name2.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1206
PDCORR
Command Default
No correlations defined between random input variables.
Notes
Specifies the correlations between two random input variables. The PDS tool generates correlated samples
if correlations exist. This applies to both probabilistic methods (Monte Carlo Simulation and Response Surface
Methods).
If there are correlated random input variables, the sampled locations of the random input variables reflect
the correlation as illustrated below for the case of two standard normal distributed variables X
1
and X
2
. The
illustrations show no correlation (correlation coefficient 0.0), a relatively moderate negative correlation
(correlation coefficient -0.6), and a relative strong positive correlation (correlation coefficient 0.9).
Menu Paths
Main Menu>Prob Design>Prob Definitns>Correlation
PDDMCS, NSIM, --, Astop, ACCMEAN, ACCSTDEV, CHECK, Seed
Specifies options for Monte Carlo Simulations using direct sampling.
PROBABILISTIC: Methods
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSIM
Number of simulation loops of the analysis.
--
Unused field.
Astop
Autostop option label.
AUTO
Enable Autostop. When Autostop is used, the PDS feature continues the simulation loops until the
convergence criteria for the mean value and the standard deviation have been met or until the
number of simulations NSIM are complete, whichever comes first. The convergence criteria (mean
value and standard deviations of all random output parameters) are specified by the ACCMEAN and
ACCSTDEV parameters. The criteria are met if the mean value and the standard deviations converge
within the accuracy specified in the ACCMEAN and ACCSTDEV options. The convergence check is
done every i-th loop, where i is specified in the CHECK parameter.
ALL
Disable Autostop option. All Monte Carlo Simulations as specified by NSIM are performed (default).
1207
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDDMCS
ACCMEAN
Accuracy of the mean values of all random output parameters that must be met for the Autostop option.
Default is 0.01 (1%). ACCMEAN is ignored for Astop = ALL. The convergence of the mean values is met
if for all random output parameters y the equation holds:
y i y i CHECK
y i
ACCMEAN CHECK CHECK
( ) ( )
( )
, ,
with i 2 3
where the value of CHECK is specified in the CHECK option.
ACCSTDEV
Accuracy of the standard deviations of all random output parameters that must be met for the Autostop
option. The default is 0.02 (2%). ACCSTDEV is ignored for Astop = ALL. The convergence for the
standard deviations is met if for all random output parameters y the equation holds:
y y
y
i i CHECK
i
ACCSTDEV CHECK CHECK
( ) ( )
( )
, ,
with i 2 3
where the value of CHECK is specified in the CHECK option.
CHECK
Sets how often conditions for convergence are checked for Autostop. The PDS checks if the convergence
criteria are met every i-th loop, where i is given by the CHECK parameter. The default value is 10. It is
not recommended to use CHECK = 1, because it could cause Autostop to terminate the simulations
prematurely. The mean values and standard deviation might not show large differences between con-
secutive simulation loops but might still have a visible "global" trend if viewed over several simulations.
This behavior indicates that convergence has not really been achieved. If you set CHECK = 1 (or another
small number), then Autostop is not able to detect such a global trend. CHECK is ignored for Astop =
ALL.
Seed
Seed value label. Random number generators require a seed value that is used to calculate the next
random number. After each random number generation finishes, the seed value is updated and is used
again to calculate the next random number. ANSYS initializes the seed value with the system time when
the ANSYS session started.
CONT
Continues updating using the derived seed value (default).
TIME
Initializes the seed value with the system time. You can use this if you want the seed value set to a
specific value for one analysis and then you want to continue with a "random" seed in the next
analysis. It is not recommended to "randomize" the seed value with the Seed = TIME option for
multiple analyses. If the Monte Carlo simulations requested with this command will be appended to
previously existing simulations, then the Seed option is ignored and Seed = CONT is used.
INIT
Initializes the seed value using 123457 (a typical recommendation). This option leads to identical
random numbers for all random input variables when the exact analysis will be repeated, making it
useful for benchmarking and validation purposes (where identical random numbers are desired). If
the Monte Carlo simulations requested with this command will be appended to previously existing
simulations, then the Seed option is ignored and Seed = CONT is used.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1208
PDDMCS
Value
Uses the specified (positive) value for the initialization of the seed value. This option has the same
effect as Seed = INIT, except you can chose an arbitrary (positive) number for the initialization. If
the Monte Carlo simulations requested with this command will be appended to previously existing
simulations, then the Seed option is ignored and Seed = CONT is used.
Command Default
NSIM = 30, Astop = AUTO, ACCMEAN = 0.01, ACCSTDEV = 0.02, CHECK = 10, Seed = CONT.
Notes
Defines the number of simulations, the specification of the Autostop option, how frequently the Autostop
option is checked and the seed value for random number generation. If the Autostop option is enabled,
then the number of simulations given for NSIM can be regarded as the maximum number of simulations
to be performed. The Autostop option may terminate the simulations before NSIM simulations are done if
the convergence criteria are met.
Menu Paths
Main Menu>Prob Design>Prob Method>Monte Carlo Sims
PDDOEL, Name, Method, Vtype, Lopt, VAL1, VAL2, VAL3, VAL4, VAL5
Defines design of experiment levels for an individual random input variable.
PROBABILISTIC: Preprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name. The parameter name must have been previously defined as a random input variable
using the PDVAR command.
Method
Specifies the response surface method for which the levels of the design of experiment are to be defined.
This field must not be left blank.
CCD
Use the Central Composite Design method. The design experiment levels of a central composite
design are defined in the fields VAL1 to VAL5.
BBM
Use the Box-Behnken Matrix method. The design experiment levels of a Box-Behnken Matrix design
are defined in the fields VAL1 to VAL3. The fields VAL4 and VAL5 are ignored
Vtype
Specifies the type of the values of the design of experiment levels.
PROB
The design of experiment levels are specified in terms of probabilities. This is the default.
PHYS
The design of experiment levels are specified in terms of physical values.
1209
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDDOEL
Lopt
Specifies the type of the design of experiment levels, indicating if they are defined by lower and upper
bound only (default) or all specified by the user.
BND
You specify the lower and upper bounds for the design of experiment levels. The values for interme-
diate levels are calculated automatically at run time (default). The lower and upper levels of the
design of experiment itself can be specified either in terms of probabilities or in terms of physical
values, depending on the Vtype field.
For Lopt = BND and Method = CCD only the entries VAL1 and VAL5 are processed and they rep-
resent the lower and upper bound values of a central composite design. The intermediate levels
VAL2 to VAL4 are evaluated automatically. For Lopt = BND and Method = BBM only the entries
VAL1 and VAL3 are processed and they represent the lower and upper bound values of a Box-
Behnken Matrix design respectively. The intermediate level VAL2 is evaluated automatically.
NOTE: The intermediate levels between the lower and upper bounds are calculated so they are at
equal intervals along the curve (interpolated linearly in the physical space whether the curve is
symmetrical or not) regardless of whether the lower and upper bounds are specified as probabilities
(Vtype = PROB) or as physical values (Vtype = PHYS).
ALL
You explicitly specify all necessary design of experiment levels. The design of experiment levels can
be in terms of probabilities or in terms of physical values, depending on the Vtype field.
VAL1, VAL2, VAL3, . . . , VAL5
Values for the levels of the design of experiment for the random input variable Name. Must be specified
in ascending order. All probabilities must be between 0.0 and 1.0.
For response surface methods based on a Central Composite Design (Method = CCD) VAL1 through
VAL5 correspond to the five design of experiment levels of this method.
For response surface methods based on a Box-Behnken Matrix design (Method = BBM) VAL1 through
VAL3 correspond to the three design of experiment levels of this method. The entries for VAL4 and
VAL5 are ignored for Method = BBM.
Command Default
For Meth = CCD, VAL1 = 0.005, VAL2 = 0.05, VAL3 = 0.5, VAL4 = 0.95, VAL5 = 0.995.
For Meth = BBM, VAL1 = 0.01, VAL2 = 0.5, VAL3 = 0.99.
Notes
If Vtype = PHYS, you must enter values for VAL1 through VAL3 or VAL5 (depending on the Method and
Lopt option you choose). If Vtype = PROB and you do not enter values, they default to the values shown
below.
For Method = CCD:
Lopt = ALL Lopt = BND
VAL1 through VAL5 must not be blank. VAL1 and VAL5 must not be blank. Vtype = PHYS
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1210
PDDOEL
VAL1 = 0.005, VAL2 = 0.05, VAL3 = 0.50, VAL4 =
0.95, VAL5 = 0.995
VAL1 = 0.005, VAL5 = 0.995 Vtype = PROB
For Method = BBM:
Lopt = ALL Lopt = BND
VAL1 through VAL3 must not be blank. VAL1 and VAL3 must not be blank. Vtype = Phys
VAL1 = 0.01, VAL2 = 0.50, VAL3 = 0.99 VAL1 = 0.01, VAL3 = 0.99 Vtype = Prob
See Probabilistic Design in the Advanced Analysis Techniques Guide for more information on the PDS methods.
Menu Paths
Main Menu>Prob Design>Prob Method>Response Surface
PDEF, Lab, Item, Comp, Avglab
Interpolates an item onto a path.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Label assigned to the resulting path item (8 characters maximum). This item may be used as input for
other path operations.
Item
Label identifying the item for interpolation. Valid item labels are shown in Table 242: PDEF - Valid Item
and Component Labels (p. 1212) below. Some items also require a component label.
Comp
Component of the item (if required). Valid component labels are shown in Table 242: PDEF - Valid Item
and Component Labels (p. 1212) below.
Avglab
Option to average across element boundaries:
AVG
Average element results across elements (default).
NOAV
Do not average element results across elements. If the parameter DISCON = MAT on the PMAP
command, this option is automatically invoked.
Notes
Defines and interpolates a labeled path item along a predefined path (PATH). Path item results are in the
global Cartesian coordinate directions unless transformed (RSYS). A path item must be defined before it
can be used with other path operations. Additional path items may be defined from the PVECT, PCALC,
PDOT, and PCROSS commands. Path items may be listed (PRPATH) or displayed (PLPATH, PLPAGM). A
maximum number of path items permitted is established by the nSets argument specified with the PATH
command.
1211
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDEF
When you create the first path item (PDEF or PVECT), the program automatically interpolates four path
items which are used to describe the geometry of the path. These predefined items are the position of the
interpolated path points (labels XG, YG, and ZG) in global Cartesian coordinates, and the path length (label
S). For alternate methods of mapping the path geometry (to include, for example, material discontinuity)
see the PMAP command. These items may also be listed or displayed with the PRPATH, PLPATH, and
PLPAGM commands.
If specifying that load case operations act on principal/equivalent stresses (SUMTYPE,PRIN), derived quant-
ities (principal and equivalent stresses/strains) will be zero for path plots. A typical use for such a case involves
mode combinations in a response spectrum analysis.
The number of interpolation points on the path is defined by the nDiv argument on the PATH command.
See Mapping Nodal and Element Data onto the Path in the Theory Reference for the Mechanical APDL and
Mechanical Applications for details. Use PDEF,STAT to list the path item labels. Use PDEF,CLEAR to erase all
labeled path items, except the path geometry items (XG, YG, ZG, S).
Table 242 PDEF - Valid Item and Component Labels
Description Comp Item
Valid item and component labels for nodal degree of freedom results are:
X, Y, or Z structural displacement or vector sum. X, Y, Z, SUM U
X, Y, or Z structural rotation or vector sum. X, Y, Z, SUM ROT
Temperature. TEMP[1]
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity or vector sum. X, Y, Z, SUM V
X, Y, or Z magnetic vector potential or vector sum. X, Y, Z, SUM A
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy (FLOTRAN). ENKE
Turbulent energy dissipation (FLOTRAN). ENDS
Valid item and component labels for element results are:
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity or Equivalent stress. INT, EQV "
Component total strain (EPEL + EPPL + EPCR). X, Y, Z, XY, YZ, XZ EPTO
Principal total strain. 1, 2, 3 "
Total strain intensity or total equivalent strain. INT, EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity or elastic equivalent strain. INT, EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1212
PDEF
Description Comp Item
Plastic strain intensity or plastic equivalent strain. INT, EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity or creep equivalent strain. INT, EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity or thermal equivalent strain. INT, EQV "
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
For contact results PowerGraphics is applicable for 3-D models only.
Contact status. STAT2 CONT
Contact penetration. PENE "
Contact pressure. PRES "
Contact friction stress. SFRIC "
Contact total stress (pressure plus friction). STOT "
Contact sliding distance. SLIDE "
Contact gap distance. GAP "
Total heat flux at contact surface. FLUX "
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
Component pressure gradient or vector sum. X, Y, Z, SUM PG
Component electric field or vector sum. X, Y, Z, SUM EF
Component electric flux density or vector sum. X, Y, Z, SUM D
Component magnetic field intensity or vector sum. X, Y, Z, SUM H
Component magnetic flux density or vector sum. X, Y, Z, SUM B
Component electromagnetic force or vector sum. X, Y, Z, SUM FMAG
Any user-defined element table label (see ETABLE com-
mand).
Lab ETAB
Applied and calculated temperatures along a defined
path.
TEMP BFE
Valid item labels for FLOTRAN nodal results are:
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
1213
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDEF
Description Comp Item
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Turbulent viscosity coefficient. CMUV
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
Specific heat. SPHT
Radiation heat flux. RDFL
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels TBOT, TE2, TE3, ..., TTOP
instead of TEMP.
2. For more information on the meaning of contact status and its possible values, see Reviewing Results
in POST1 in the Contact Technology Guide.
Menu Paths
Main Menu>General Postproc>Path Operations>Clear Path Items
Main Menu>General Postproc>Path Operations>Define Path>Path Status>Current Path
Main Menu>General Postproc>Path Operations>Map onto Path
Main Menu>General Postproc>Path Operations>Map onto Path>FE Results
PDEXE, Slab, MRUN, NFAIL, FOPT, Fname
Executes the probabilistic analysis.
PROBABILISTIC: Run Probabilistic Analysis
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Slab
Solution label. This is the name given to the files containing the results of different probabilistic analyses
performed with a particular probabilistic model. Different solutions using different probabilistic methods
or method options are stored in separate solution sets identified with this solution label. If the solution
label is blank, ANSYS automatically assigns Slab = "SOLUnn", where nn is the current two-digit number
of the probabilistic analysis (PDS allows a maximum of 10 solution sets identified by 10 solution labels);
for example, the first analysis is SOLU01, the second is SOLU02, and so on up to SOLU10). This field
cannot contain blanks. Maximum length of this field is 32 characters; if the field contains more than 32
characters, it will be truncated to 32 characters.
MRUN
Keyword for the processing of the simulation loops
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1214
PDEXE
SER
Executes the simulation loops of the probabilistic analysis using a single machine; serial processing
(default).
PAR
Executes the simulation loops of the probabilistic analysis in parallel using the ANSYS parallel-pro-
cessing tool. See Probabilistic Design in the Advanced Analysis Techniques Guide.
NFAIL
This parameter sets the number of loops that are allowed to fail before the execution of the probabilistic
analysis is terminated. For example, if geometric parameters are used as random variables then for some
simulations (loops) the loop will fail because meshing was not successful. This might happen infrequently
which might be acceptable. However, if the probabilistic problem is incorrectly defined then loops could
fail frequently or even in every loop. You can define an upper limit for the number of failed loops that
will be tolerated before the probabilistic analysis will be terminated. If you use NFAIL = 0 (default) then
failed loops are not tolerated and the probabilistic analysis will terminate the first time a loop fails.
NFAIL is used only if MRUN = PAR.
FOPT
Keyword for the file option. This determines if a copy of the sample file should be saved or not.
DEL
The sample file will be deleted (default).
COPY
A copy of the sample file will be saved as the file name specified with the Fname option.
Fname
Name of the copy of the sample file. The directory for storing the sample file is always the current
working directory. The file extension is always .csmp. This option is ignored for FOPT = DEL.
Command Default
Slab = SOLUnn (see above) , MRUN = SER, NFAIL = 0, FOPT = DEL.
Notes
Runs the probabilistic analysis. The results of the probabilistic analysis are identified with the solution label
specified here. This solution label is used for the postprocessing of the results. The maximum number of
probabilistic analyses that can be executed, stored, and post-processed based on the same deterministic
and probabilistic model in one session with the PDS is 10. However, it is not counted as a separate probab-
ilistic analysis if the same solution label is used multiple times in order to append to or overwrite previous
results.
The probabilistic results are stored in result files in ASCII format.
If you specify a solution label that has already been used, the results will be appended at the end of the
results file.
After the PDEXE command has been issued, you can no longer change the probabilistic model using prob-
abilistic preprocessing commands; doing so would invalidate the probabilistic analysis results and potentially
corrupt the probabilistic database. Also, you are not allowed to change the probabilistic analysis file using
a PDANL command after the first PDEXE has been executed.
1215
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDEXE
If you need to change the probabilistic model (analysis file, random input variables, etc.) after the PDEXE
command has been issued, you should use the PDCLR, POST command to clear the probabilistic results.
We recommend that you first save the results using the PDSAVE command.
By default, ANSYS creates a sample file (called Jobname.samp) containing all values of all random input
parameters; unless you specify otherwise, this file is deleted after the analysis is finished. With the FOPT
and Fname options you can determine if and where a copy of that file should be stored for later use in an-
other analysis.
Menu Paths
Main Menu>Prob Design>Run>Exec Serial >Run Serial
PDHIST, Rlab, Name, NCL, Type
Plots the frequency histogram.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo
Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface
Analyses).
Name
Parameter name. The parameter must have been previously defined as a random input variable or a
random output parameter with the PDVAR command.
NCL
Number of classes for the histogram plot. This is the number of bars shown in the histogram. NCL must
be a positive number. If this field is left blank, ANSYS calculates an appropriate number of classes based
on the sample size. ANSYS divides the range between the smallest and largest sample value into NCL
classes of equal width and determines the histogram frequencies by counting the number of hits that
fall in the classes.
Type
Type of histogram.
ABS
Absolute frequency histogram. This is the actual number of hits in each class.
REL
Relative frequency histogram (default). This is the number of hits in the individual classes divided
by the total number of samples.
NORM
Normalized frequency histogram. This is the number of hits in the individual classes divided by the
total number of samples and divided by the width of the class. This normalization makes the histogram
comparable to the probability density function.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1216
PDHIST
Command Default
Rlab as described above, NCL as described above, Type = REL.
Notes
Plots the frequency histogram.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
The PDHIST command cannot be used to postprocess the results in a solution set that is based on Response
Surface Methods, only Monte Carlo Simulations.
Menu Paths
Main Menu>Prob Design>Prob Results>Statistics>Histogram
PDINQR, Rpar, Name, Type, VAL
Evaluates statistical characteristics of a random input variable.
PROBABILISTIC: Preprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rpar
Name of scalar ANSYS parameter into which the characteristic value is stored.
Name
Parameter name. The parameter must have been previously defined as a random input variable using
the PDVAR command.
Type
Keyword for the type of information you need to retrieve.
MEAN
Mean value of the random input variable specified in Name.
STDV
Standard deviation of the random input variable
PDF
Probability density function of the random input variable specified in Name at a given value specified
by VAL.
CDF
Cumulative distribution function of the random input variable specified in Name at a given value
specified by VAL.
INV
Inverse cumulative distribution function of the random input variable specified in Name at a given
probability specified by VAL.
1217
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDINQR
VAL
Value needed for Type = PDF, CDF, or INV. For Type = PDF and Type = CDF, this is the value of the
random input variable at which the probabilistic density or cumulative distribution function should be
evaluated. For Type = INV, VAL indicates the probability at which you want the inverse cumulative
distribution function evaluated.
Notes
Evaluates statistical characteristics of a random input variable. The result is stored in the ANSYS parameter
Rpar.
Menu Paths
Main Menu>Prob Design>Prob Definitns>Inquire
PDLHS, NSIM, NREP, ISopt, --, Astop, ACCMEAN, ACCSTDV, CHECK, Seed
Specifies options for Monte Carlo Simulations using Latin-Hypercube sampling.
PROBABILISTIC: Methods
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSIM
Number of simulation loops per repetition cycle.
NREP
Number of repetition cycles of the analysis.
ISopt
Latin-Hypercube sampling divides the domain of each random input variable into intervals of equal
probability. The interval sampling option ISopt determines where the samples are located within each
interval.
RAND
Picks a random location within the interval (default).
MEAN
Picks the mean value location within the interval.
MEDI
Picks the median value location within the interval.
--
Unused field.
Astop
Autostop option label.
AUTO
Enable Autostop. When Autostop is used, the PDS feature continues the simulation loops until the
convergence criteria for the mean value and the standard deviation have been met or until the
number of simulations NSIM are complete, whichever comes first. The convergence criteria (mean
value and standard deviations of all random output parameters) are specified by the ACCMEAN and
ACCSTDEV parameters. The criteria are met if the mean value and the standard deviations converge
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1218
PDLHS
within the accuracy specified in the ACCMEAN and ACCSTDEV options. The convergence check is
done every i-th loop, where i is specified in the CHECK parameter.
ALL
Disable Autostop option. All Monte Carlo Simulations as specified by NSIM and NREP are performed
(default).
ACCMEAN
Accuracy of the mean values of all random output parameters that must be met to activate Autostop.
Default is 0.01 (1%). ACCMEAN is ignored for Astop = ALL. The convergence for the mean values is met
if for all random output parameters y the following equation is true:
y i y i CHECK
y i
ACCMEAN CHECK CHECK
( ) ( )
( )
, ,
with i 2 3
where the value of CHECK is given by the CHECK option.
ACCSTDEV
Accuracy of the standard deviations of all random output parameters that must be met to activate
Autostop. The default is 0.02 (2%). ACCSTDEV is ignored for Astop = ALL. The convergence for the
standard deviations is met if for all random output parameters y the following equation is true:
y y
y
i i CHECK
i
ACCSTDEV CHECK CHECK
( ) ( )
( )
, ,
with i 2 3
where the value of CHECK is given by the CHECK option.
CHECK
Sets how often conditions for convergence are checked for Autostop. The PDS feature checks if the
convergence criteria are met every i-th loop, where i is given by the CHECK parameter. The default
value is 10. It not recommended to use CHECK = 1, because it could cause Autostop to terminate the
simulations prematurely. The mean values and standard deviation might not show large differences
between all simulation loops but might still have a visible "global" trend if viewed over several simulations.
This behavior indicates that convergence has not really been achieved. If you set CHECK = 1, then
Autostop is not able to detect such a global trend. CHECK is ignored for Astop = ALL.
Seed
Seed value label. Random number generators require a seed value that is used to calculate the next
random number. After each random number generation finishes, the seed value is updated and is used
again to calculate the next random number. ANSYS initializes the seed value with the system time when
the ANSYS session started.
CONT
Continues updating using the derived seed value (default).
TIME
Initializes the seed value with the system time. You can use this if you want the seed value set to a
specific value for one analysis and then you want to continue with a "random" seed in the next
analysis. It is not recommended to "randomize" the seed value with the Seed = TIME option for
multiple analyses. If the Monte Carlo simulations requested with this command will be appended to
previously existing simulations, then the Seed option is ignored and Seed = CONT is used.
1219
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDLHS
INIT
Initializes the seed value using 123457 (a typical recommendation). This option leads to identical
random numbers for all random input variables when the exact analysis will be repeated, making it
useful for benchmarking and validation purposes (where identical random numbers are desired). If
the Monte Carlo simulations requested with this command will be appended to previously existing
simulations, then the Seed option is ignored and Seed = CONT is used.
Value
Uses the specified (positive) value for the initialization of the seed value. This option has the same
effect as Seed = INIT, except you can chose an arbitrary (positive) number for the initialization. If
the Monte Carlo simulations requested with this command will be appended to previously existing
simulations, then the Seed option is ignored and Seed = CONT is used.
Command Default
NSIM = 30, NREP = 1, ISopt = RAND, Astop = ALL, ACCMEAN = 0.01, ACCSTDEV = 0.02, CHECK = 10,
Seed = CONT.
Notes
Defines the number of simulations per repetition cycle, number of repetition cycles, specification of the
Autostop option, checking frequency for the Autostop option, and the seed value for random number gen-
eration.
For Latin-Hypercube sampling, it is advantageous to divide the total number of requested simulations into
a few repetitions. This adds more randomness to the sampling process. If NTOT is the total number of sim-
ulations, then as a rough rule of thumb NTOT should be NREP =
NTOT
4
repetitions. The number obtained
with this rule of thumb must be adjusted such that NTOT = NREP*NSIM. For example if NTOT = 1000 then
NREP =
1000
4
= 10, so the 1000 simulations can be done in 100 simulations with 10 repetitions. If for ex-
ample NTOT = 100 then NREP =
100
4
= 3.16, which means that the 100 simulations could be broken up
into either 2*50 or 4*25 simulations.
If Autostop is enabled then the maximum number of simulations to be performed is given by NSIM*NREP.
The Autostop option will terminate the simulations before the NSIM*NREP simulations are done if the
convergence criteria are met.
Menu Paths
Main Menu>Prob Design>Prob Method>Monte Carlo Sims
PDMETH, Method, Samp
Specifies the probabilistic analysis method.
PROBABILISTIC: Methods
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Method
Label for the probabilistic analysis method.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1220
PDMETH
MCS
Monte Carlo Simulation
RSM
Response Surface Method
Samp
Label for the sampling techniques. The sampling technique determines the values of the random input
variables during the simulation loops.
DIR
Direct or Crude Monte Carlo Sampling. This technique randomly samples the random input variables
according to their distribution functions without "memory" of previous simulations.
The parameters for a Monte Carlo Simulation using direct sampling are specified with the PDDMCS
command.
LHS
Latin Hypercube Sampling (default). Valid only for Method = MCS. For this sampling technique the
random input variables are sampled randomly according to their distribution functions, efficiently
stratifying the samples into layers and avoiding the re-use of those layers. The sampling process has
a "memory" of previous simulations, which prevents accumulation of clusters of samples. In addition,
this sampling strategy forces the extreme ends of a distribution function to participate in the sampling.
This generally leads to smoother distribution functions of the sampled set.
The parameters for a Monte Carlo Simulation using Latin-Hypercube sampling are specified with the
PDLHS command.
USER
User specified sampling. Valid only for Method = MCS. In this case you provide a file containing the
sampling "points" (values) of all random input variables for all simulation loops. These samples are
simply executed and it is your responsibility to specify the samples correctly. The probabilistic design
system can perform only limited checks on the samples you provide. ANSYS allows Monte Carlo
specific postprocessing operations on the results generated with user-specified samples. The para-
meters for the user-supplied sampling technique are specified with the PDUSER command.
CCD
Central Composite Design. Valid only for Method = RSM. A central composite design is composed
of a center point, axis points, and corner points, called factorial points. Using large numbers of random
input variables produces prohibitively large numbers of factorial points; therefore, ANSYS automat-
ically reduces the number of factorial points by switching to a fractional plan for the factorial part
of the design. See the PDDOEL command for more information.
Note
This option is only valid for 2 to 20 random input variables. You will receive an error if
you have specified fewer than 2 or more than 20 random input variables.
BBM
Box-Behnken Matrix Design. Valid only for Method = RSM. A Box-Behnken Matrix design is composed
of a center point plus the points at the middle of the edges of the hypercube in the space of random
input variables. A Box-Behnken design might be advantageous if the corner points of the hypercube
represent very extreme conditions that are undesirable and therefore should not be used for the
sampling. See the PDDOEL command for more information.
1221
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDMETH
Note
This option is only valid for 3 to 12 random input variables. You will receive an error if
you have specified fewer than 3 or more than 12 random input variables.
USER
User specified sampling. In this case you provide a file containing the sampling "points" (values) of
all random input variables for all simulation loops. These samples are simply executed and it is your
responsibility to specify the samples correctly. The PDS can perform only limited checks on the
samples you provide, if user-supplied sampling technique are specified with the PDUSER command.
Command Default
Method = MCS, Samp = LHS.
Notes
Specifies the probabilistic analysis method and the sampling technique used for the individual probabilistic
analysis method.
Menu Paths
Main Menu>Prob Design>Prob Method>Monte Carlo Sims
Main Menu>Prob Design>Prob Method>Response Surface
PDOT, LabR, LabX1, LabY1, LabZ1, LabX2, LabY2, LabZ2
Calculates the dot product of two path vectors along the current path.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabR
Label assigned to dot product result.
LabX1
X-component of first vector label (labeled path item).
LabY1
Y-component of first vector label (labeled path item).
LabZ1
Z-component of first vector label (labeled path item).
LabX2
X-component of second vector label (labeled path item).
LabY2
Y-component of second vector label (labeled path item).
LabZ2
Z-component of second vector label (labeled path item).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1222
PDOT
Menu Paths
Main Menu>General Postproc>Path Operations>Dot Product
PDPINV, Rlab, Name, PROB, --, CONF
Prints the result of the inversion of a probability.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo
Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface
Analyses).
Name
Parameter name. The parameter must have been previously defined as a random input variable or a
random output parameter with the PDVAR command.
PROB
Target probability for which the random parameter value should be determined.
--
Unused field.
CONF
Confidence level. The confidence level is used to print the confidence bounds on the random parameter
value. The value for the confidence level must be between 0.0 and 1.0 and it defaults to 0.95 (95%).
Printing of confidence bound is suppressed for CONF
0.5. This parameter is ignored for response
surface methods results postprocessing.
Command Default
Rlab as described above, CONF = 0.95.
Notes
Prints the value for the random parameter Name at which the probability that there are simulation values
lower than that value is equal to PROB. This corresponds to an inversion of the cumulative distribution
function (see PDCDF command) at a given probability. In this sense the PDPINV is doing the opposite of
the PDPROB command. The PDPROB command evaluates a probability for a given random parameter value
and the PDPINV command evaluates the random parameter value that corresponds to a given probability.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
The confidence level is a probability expressing the confidence that the value for the requested result is in
fact between the confidence bounds. The larger the confidence level, the wider the confidence bounds.
Printing the confidence bounds only makes sense for postprocessing Monte Carlo simulation results, where
the confidence bounds represent the accuracy of the results. With increasing sample sizes, the width of the
1223
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDPINV
confidence bounds gets smaller for the same confidence level. For response surface analysis methods, the
number of simulations done on the response surface is usually very large; therefore, the accuracy of the
results is determined by the response surface fit and not by the confidence level.
The PDPINV command cannot be used to postprocess the results in a solution set that is based on Response
Surface Methods, only Monte Carlo Simulations.
Menu Paths
Main Menu>Prob Design>Prob Results>Statistics>Inverse Prob
PDPLOT, Name, PLOW, PUP
Plots the distribution curves of a defined random input variable.
PROBABILISTIC: Preprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name. The parameter name must have been previously defined as a random input variable
using the PDVAR command.
PLOW
Lower probability level used to determine the lower boundary of the curve. This probability must be
between 0.0 and 1.0 and it defaults to 0.0025. This parameter is used to determine the lower plotting
range (boundary) in case the random input variable does not have a minimum value (such as Gauss).
PUP
Upper probability level used to determine the upper boundary of the curve. This probability must be
between 0.0 and 1.0 and it defaults to 0.9975.
Command Default
PLOW = 0.0025 (applied if distribution type has no minimum), PUP = 0.9975 (applied if distribution type has
no maximum).
Notes
Plots the distribution of a defined random input variable. The PDPLOT command generates a probability
density function plot as well as a cumulative distribution function plot of the random variable. The probab-
ilities PLOW and PUP are used to determine the plot range of the random input variable values. To do this,
the probabilities are converted into random input variable values using the inverse cumulative distribution
function of the random input variable as shown in the following illustration.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1224
PDPLOT
Using the probabilities ensures that the boundaries are always feasible and meaningful for the random input
variable regardless of its distribution type.
If PLOW is left blank, then a minimum value of the distribution is used for plotting, provided it exists (for
example, uniform distribution). If the distribution type has no minimum value (for example, a Gaussian dis-
tribution) then the default value is used to determine the lower plotting range value. The same applies for
the maximum value, if PUP is left blank.
Menu Paths
Main Menu>Prob Design>Prob Definitns>Plot
PDPROB, Rlab, Name, Relation, LIMIT, --, CONF
Prints a probability result.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo
Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface
Analyses).
Name
Parameter name. The parameter must have been previously defined as a random input variable or a
random output parameter using the PDVAR command.
Relation
Relation label for the relation between probabilistic design parameter Name and the limit value LIMIT:
LT
Less than (default).
GT
Greater than.
LIMIT
Limit value.
--
Unused field.
CONF
Confidence level. The confidence level is used to print the confidence bounds on the probability. The
value for the confidence level must be between 0.0 and 1.0 and it defaults to 0.95 (95%). Printing of
confidence bound is suppressed for CONF
0.5. This parameter is ignored for response surface methods
results postprocessing.
Command Default
Rlab as described above, Relation = LT, CONF = 0.95.
1225
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDPROB
Notes
Prints the probability that the probabilistic design input or output variable denoted with Name is smaller
or larger than a certain limit value.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
Use the relation label Relation to specify if you want to print a traditional probability value (LT) or the
exceedence probability (GT). The LIMIT directly specifies at which value of Name (the design parameter)
the probability should be evaluated. If LIMIT is between two sample values of Name the resulting probab-
ility is linearly interpolated between the sample values. If LIMIT is smaller than all sample values of Name
the probability is 0.0 for Relation = LT and 1.0 for Relation = GT. If LIMIT is greater than all sample
values for Name the probability is 1.0 for Relation = LT and 0.0 for Relation = GT.
The confidence level is a probability expressing the confidence that the value for the requested probability
is in fact between the confidence bounds. The larger the confidence level, the wider the confidence bounds.
Printing the confidence bounds only makes sense for postprocessing Monte Carlo simulation results, where
the confidence bounds represent the accuracy of the results. With increasing sample sizes, the width of the
confidence bounds gets smaller for the same confidence level. For response surface analysis methods, the
number of simulations done on the response surface is usually very large; therefore, the accuracy of the
results is determined by the response surface fit and not by the confidence level.
The PDPROB command cannot be used to postprocess the results in a solution set that is based on Response
Surface Methods, only Monte Carlo Simulations.
Menu Paths
Main Menu>Prob Design>Prob Results>Statistics>Probabilities
PDRAG, PX1, PY1, PZ1, H1, PX2, PY2, PZ2, H2, Kcord
Defines the external fluid drag loading for a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
PX1, PY1, PZ1
External fluid drag pressure (global Cartesian components) at height H1.
H1
Height (along Kcord coordinate) for first drag pressure.
PX2, PY2, PZ2
External fluid drag pressure (global Cartesian components) at height H2.
H2
Height (along Kcord coordinate) for second drag pressure.
Kcord
Coordinate direction for height value (in the global Cartesian coordinate system):
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1226
PDRAG
X
X coordinate.
Y
Y coordinate (default).
Z
Z coordinate.
Notes
Defines the external fluid drag loading (pressure) as a function of height for a piping run. See the PREP7
RUN command. The element drag pressure is determined from the centroid height and linear interpolation.
Pressures are assigned to the elements as they are generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Loads
PDRESU, Fname, Ext, --
Reads the probabilistic model data and loads it into the database.
PROBABILISTIC: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to pds if Fname is blank; otherwise, no default.
--
Unused field.
Command Default
Fname = jobname, Ext = pds, the default directory is the current working directory.
Notes
Reads the probabilistic model data from the specified file and loads it into the database. Probabilistic analyses
results are not stored in the database with the PDRESU command, rather they reside in separate results
files. Analyses results are loaded automatically (one-by-one and on demand) when a probabilistic postpro-
cessing command is issued.
1227
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDRESU
Menu Paths
Main Menu>Prob Design>Prob Database>Resume
PDROPT, RVAR, CORR, STAT, SHIS, HIST, CDF, SENS, CMAT, CONF
Specifies the options for an HTML report.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RVAR
Specifies in which form to show the definitions of random variables in the report.
0
Using tables (including name and distribution parameter) and figures (including a probability density
function plot and a cumulative distribution plot) (default).
1
Using tables only.
2
Using figures only.
3
None.
CORR
Specifies if a table describing the correlation between random variables should be included in the report.
0
Yes, include this table (default).
1
No, do not include this table.
STAT
Specifies which statistics to include in the report. In general, statistics are provided in a tabular form.
0
Statistics of the random output parameters only (default).
1
Statistics of the random input variables only.
2
Statistics of both the random input variables and the random output parameters.
3
None.
SHIS
Specifies which sample history plots to include in the report. This option applies to the random output
parameters only.
0
None (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1228
PDROPT
1
Mean value and standard deviation as a sample plot.
2
Mean value, standard deviation and sample values as a sample plot.
3
All types of sample plots - mean value, standard deviation, minimum value, maximum values, and
the sample values.
HIST
Specifies which histogram plots to include in the report.
0
Histogram of the random output parameters only (default).
1
Histogram of the random input variables only.
2
Histogram of both the random input variables and the random output parameters.
3
None.
CDF
Specifies which cumulative distribution function (CDF) plots to include in the report.
0
CDF of the random output parameters only (default).
1
CDF of the random input variables only.
2
CDF of both the random input variables and the random output parameters.
3
None.
SENS
Specifies which sensitivity plots to include in the report.
0
Plots the sensitivities for all random output parameters based on Spearman-rank-order correlation
coefficient (default).
1
Plots the sensitivities for all random output parameters based on linear (Pearson) correlation coefficient.
2
Plots the sensitivities according to option SENS=1 and SENS=2.
3
None.
CMAT
Specifies which correlation matrices to include in the report.
0
Correlation matrix between random output parameters and random output parameters only (default).
1229
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDROPT
1
Correlation matrix between random input variables and random output parameters only.
2
Correlation matrix between random input variables and random input variables only.
3
Correlation matrices according to option CMAT=0 and CMAT=1.
4
Correlation matrices according to option CMAT=0 and CMAT=2.
5
Correlation matrices according to option CMAT=1 and CMAT=2.
6
Correlation matrices according to option CMAT=0, CMAT=1, and CMAT=2.
7
None.
CONF
Confidence level. The confidence level is used to plot confidence bounds for the history value. The value
for the confidence level must be between 0.0 and 1.0 and it defaults to 0.95 (95%). Confidence bound(s)
plotting is suppressed for CONF
0.5. This option is ignored if the report does not include plots for
which confidence bounds are applicable.
Command Default
RVAR = 0, CORR = 0, STAT = 0, SHIS = 0, HIST = 0, CDF = 0, SENS = 0, CMAT = 0, CONF = 0.95.
Notes
Specifies the options for an HTML report. An HTML report includes a description of the deterministic model,
the probabilistic model, the probabilistic methods used for the analyses, and the results obtained from the
analyses. The deterministic model is documented in the report by including a link to the analysis file (see
PDANL command). In addition, an element plot of the component is shown, if available, based on the current
view settings. The command ALLSEL is issued automatically prior to the respective plot command.
Menu Paths
Main Menu>Prob Design>Prob Results>Report>Report Options
/PDS
Enters the probabilistic design system.
PROBABILISTIC: Auxiliary
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Enters the Probabilistic Design System (PDS). This command is valid only at the Begin Level.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1230
/PDS
Menu Paths
Main Menu>Prob Design
PDSAVE, Fname, Ext, --
Writes the probabilistic model data to a file.
PROBABILISTIC: Database
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to pds if Fname is blank; otherwise, no default.
--
Unused field.
Command Default
Fname = Jobname, Ext = pds, the default directory is the current working directory.
Notes
Writes the probabilistic model data to a file. Saved data include probabilistic data only; the results of the
probabilistic analyses are not stored in the file (rather, these are stored in separate result files).
Menu Paths
Main Menu>Prob Design>Prob Database>Save
PDSCAT, Rlab, Name1, Name2, Type, ORDER, NMAX
Plots a scatter graph.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command, or the response surface set label defined in an
RSFIT command.
1231
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDSCAT
Name1, Name2
Parameter names. The parameters must have been previously defined as a random input variable or a
random output parameter using the PDVAR command. The parameter data for Name1 is shown on the
X-axis and the parameter data for Name2 is shown on the Y-axis in the plot.
Type
Keyword for the type of trendline curve.
POLY
Polynomial trendline (default).
NONE
A trendline is not plotted.
ORDER
Order of the polynomial trendline. This parameter is used only for Type = POLY. ORDER must be a
positive number. There is no maximum for ORDER provided there are enough data points to evaluate
a polynomial of the requested order. Default is 1.
NMAX
Maximum number of points plotted in the scatter plot. If there are more sample data, then only the first
NMAX points are plotted. The default value is 10,000.
Command Default
Rlab as described above, Type = POLY, ORDER = 1, NMAX = 10,000
Notes
Plots a scatter graph with or without a trendline. The scatter plot shows the simulated points for two random
parameters. Random input variables and random output parameters are valid for both X- and Y-axis. The
mean value of both parameters are marked with separate green lines. The point where the green lines cross
marks the statistical center of gravity of the cloud of all simulated data points of the two parameters.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
If the parameter data for Name1 includes negative values, fitting a logarithmic trendline is not possible and
the logarithmic trendline plot is suppressed if requested. The same applies for an exponential trendline if
the data for the Name2 includes negative values.
Because of the amount of data involved, the scatter plot is limited to NMAX points. If shown, the trendline
is evaluated only on the NMAX points shown. However, the statistical information shown in the plot, such
as the mean value lines for both parameters and the correlation coefficients listed in the legend are based
on the full set of samples. If required, you can increase NMAX to plot more points, but this will affect the
time needed to process the PDSCAT command. If NMAX is less than the total amount of simulated points,
which is typically possible for Response Surface Methods, you will see an appropriate warning in the plot
legend.
Menu Paths
Main Menu>Prob Design>Prob Results>Trends>Scatter Plot
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1232
PDSCAT
PDSENS, Rlab, Name, Chart, Type, SLEVEL
Plots the probabilistic sensitivities.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command (if you are directly postprocessing Monte Carlo
Simulation results), or the response surface set label defined in an RSFIT command (for Response Surface
Analyses). The PDSENS command cannot be used to postprocess the results in a solution set that is
based on Response Surface Methods, only Monte Carlo Simulations.
Name
Parameter name. The parameter must have been previously defined as a random output parameter using
the PDVAR command.
Chart
Keyword for the type of chart to be plotted.
BAR
Bar chart of the absolute sensitivities.
PIE
Pie chart of relative and normalized sensitivities.
BOTH
Both pie and bar charts plotted side by side (default).
Type
Keyword for the type of correlation coefficients used to evaluate the sensitivities.
RANK
Spearman rank-order correlation coefficient (default).
LIN
Pearson linear correlation coefficient.
SLEVEL
Significance level. The value for the significance level must be between 0.0 and 1.0 and it defaults to
0.025 (2.5%).
Command Default
Rlab as described above, Chart = BOTH, Corr = RANK, SLEVEL = 0.025
Notes
Plots the probabilistic sensitivities.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
Evaluation of the probabilistic sensitivities is based on the correlation coefficients between all random input
variables and the random output parameter specified by Name. You can chose which correlation coefficient
should be used for that evaluation using the Corr option. For all sensitivity values, the probabilistic design
1233
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDSENS
tool evaluates the probability that the sensitivity can be neglected, based on statistical test theory. If this
probability exceeds the significance level as specified by the SLEVEL parameter, the sensitivity value should
be regarded as negligible or insignificant. The higher the significance level (SLEVEL) the more sensitivities
are considered as significant. The sensitivity plot includes the significant sensitivities only and lists the insig-
nificant ones separately.
Menu Paths
Main Menu>Prob Design>Prob Results>Trends>Sensitivities
PDSHIS, Rlab, Name, Type, CONF
Plots the sample history values.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Rlab
Result set label. Identifies the result set to be used for postprocessing. A result set label can be the
solution set label you defined in a PDEXE command, or the response surface set label defined in an
RSFIT command.
Name
Parameter name. The parameter must have been previously defined as a random input variable or a
random output parameter using the PDVAR command.
Type
Keyword to identify which type of data is to be plotted:
SAMP
Sampled values (default).
MEAN
Mean values for results based on Monte Carlo simulation methods only.
STDV
Standard deviations for results based on Monte Carlo simulation methods only.
MIN
Minimum values for results based on Monte Carlo simulation methods only.
MAX
Maximum values for results based on Monte Carlo simulation methods only.
MEAN, STDV, MIN, and MAX are only valid for Monte Carlo methods.
CONF
Confidence level. The confidence level is used to plot confidence bounds for the history value. The value
for the confidence level must be between 0.0 and 1.0 and it defaults to 0.95 (95%). Confidence bound(s)
plotting is suppressed for CONF
0.5. This option is ignored for Type = SAMP (no confidence bounds
are plotted).
Command Default
Rlab as described above, Type = SAMP, CONF = 0.95
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1234
PDSHIS
Notes
Plots the sample history values as a function of the number of simulation loops.
If Rlab is left blank, then the result set label is inherited from the last PDEXE command (Slab), RSFIT
command (RSlab), or the most recently used PDS postprocessing command where a result set label was
explicitly specified.
The confidence level is a probability expressing the confidence that the value for the requested probability
is in fact between the confidence bounds. The larger the confidence level, the wider the confidence bounds.
For Type = MEAN and Type = STDV, lower and upper confidence curves are plotted. For Type = MEAN,
the mean value curve starts at the first simulation and the confidence bounds start with simulation number
2. For Type = MIN only the upper confidence bound is shown (the interpretation is that by a certain prob-
ability the true minimum value is below this curve). This probability (or confidence) is set using CONF. Likewise,
for Type = MAX, only the lower confidence bound is shown. For all Type options, confidence curves are
plotted starting with the simulation at which enough data is available to calculate the bounds. However,
for scaling reasons, no confidence bounds are plotted for simulation numbers 1 to 5 even if they might be
available.
Menu Paths
Main Menu>Prob Design>Prob Results>Statistics>Sampl History
PDUSER, Fname, Ext, --
Specifies options for user-specified sampling methods.
PROBABILISTIC: Methods
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
Notes
If user-specified sampling methods are requested with the PDMETH, MCS, USER command or the PDMETH,
RSM, USER command, then you need to specify which file contains the sample data. For more information
on the format of this file, see Probabilistic Design in the Advanced Analysis Techniques Guide.
Menu Paths
Main Menu>Prob Design>Prob Method>Monte Carlo Sims
Main Menu>Prob Design>Prob Method>Response Surface
1235
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDUSER
PDVAR, Name, Type, PAR1, PAR2, PAR3, PAR4
Specifies the parameters to be treated as probabilistic design variables.
PROBABILISTIC: Preprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Name
Parameter name (must be a scalar ANSYS parameter). The parameter must have been previously defined
as a random input variable or a random output parameter with the PDVAR command. See the *SET
command for restrictions about ANSYS parameters.
Type
Probabilistic design variable type. This is the statistical distribution type. For more information on each
of these types, see Probabilistic Design in the Advanced Analysis Techniques Guide.
BETA
Beta distributed random variable.
PAR1 = Shape parameter. Defaults to 2.0.
PAR2 = Shape parameter. Defaults to 2.0.
PAR3 = Lower minimum value. Defaults to 0.0.
PAR4 = Upper maximum value. Defaults to 1.0.
EXPO
Exponential distributed random variable.
PAR1 = Decay parameter . Must be larger then 0.0 and defaults to 1.0.
PAR2 = Shift or minimum value. Defaults to 0.0.
PAR3, PAR4 are ignored.
GAMA
Gamma distributed random variable.
PAR1 = Decay parameter . Must be larger then 0.0 and defaults to 1.0.
PAR2 = Exponential parameter k. Must be larger then 0.0 and defaults to 1.0.
PAR3, PAR4 are ignored. Exponential distributed random variable.
GAUS
Gaussian (Normal) distributed random variable.
PAR1 = Mean value. Defaults to 0.0.
PAR2 = Standard deviation. Must be larger then 0.0 and defaults to 1.0.
PAR3, PAR4 are ignored.
LOG1
Lognormal distributed random variable specified directly with the statistical parameters mean value
and standard deviation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1236
PDVAR
PAR1 = Mean value. Must be larger then 0.0 and defaults to 1.0.
PAR2 = Standard deviation. Must be larger then 0.0 and defaults to 1.0.
PAR3, PAR4 are ignored.
LOG2
Lognormal distributed random variable specified with the statistical parameters mean value and
standard deviation of the logarithm of the random values.
PAR1 and PAR2 must also be defined. PAR1 = Mean value of the logarithm of the data. Defaults to
0.0.
PAR2 = Standard deviation of the logarithm of the data. Must be larger then 0.0 and defaults to 1.0.
PAR3, PAR4 are ignored.
UNIF
Uniform distributed random variable. Note that PAR1 must be less than PAR2.
PAR1 = Minimum value. Defaults to 0.0.
PAR2 = Maximum value. Defaults to 1.0.
PAR3, PAR4 are ignored.
TGAU
Truncated Gaussian distributed random variable. Note that PAR3 must be less than PAR4.
PAR1 = Mean value of the untruncated Gaussian distribution. Defaults to 0.0.
PAR2 = Standard deviation of the untruncated Gaussian distribution. Must be larger then 0.0 and
defaults to 1.0.
PAR3 = Minimum value and lower truncation boundary. Defaults to -3.0.
PAR4 = Maximum value and upper truncation boundary. Defaults to +3.0.
TRIA
Triangular distributed random variable. Note that PAR1 must be less than PAR2 which must be less
than PAR3.
PAR1 = Minimum value. Defaults to 0.0.
PAR2 = Most Likely Value (MLV). Defaults to 0.5.
PAR3 = Maximum value. Defaults to 1.0.
PAR4 is ignored.
WEIB
Weibull (Type III smallest) distributed random variable. Note that PAR2 must be greater than PAR3.
If PAR3 = 0.0, the random distribution is equivalent to a two-parameter Weibull distribution.
PAR1 = Weibull exponent. Must be larger then 0.0 and defaults to 1.0.
PAR2 = Characteristic value. Must be larger then 0.0 and defaults to 1.0.
1237
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PDVAR
PAR3 = Shift or minimum value. Defaults to 0.0.
PAR4 is ignored.
RESP
Random output or response parameter.PAR1 to PAR4 are not used.
DEL
Deletes this probabilistic design variable (does not delete the ANSYS parameter). This option is only
valid if the parameter Name was previously defined as a probabilistic design variable (using Type
= BETA, ..., WEIB or Type = RESP). The parameter retains the value assigned during the last probab-
ilistic design loop.PAR1 to PAR4 are not used.
PAR1, PAR2, PAR3, PAR4
Parameters of the distribution function. The parameters must be specified according to the requirements
of the individual distribution types described above.
Notes
Specifies the parameters to be treated as probabilistic design variables. A random input variable is specified
by the name of the ANSYS parameter, the type of the distribution function (Type) and its distribution
parameters (PAR1, ..., PAR4). A random output parameter is specified by the name of the ANSYS parameter
and the type identifying it as a random output parameter (Type = RESP).
Menu Paths
Main Menu>Prob Design>Random Input
PDWRITE, File, Fnam, Lnam
Generates an HTML report for the probabilistic analyses.
PROBABILISTIC: Postprocessing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
File
File name and directory path (248 characters maximum, including directory) from which to read the report.
If you do not specify a directory path, it will default to your working directory and you can use all 248
characters for the file name.
Fnam
First name of the author of the report (32 characters maximum). This first name must not include blanks.
Lnam
Last name of the author of the report (32 characters maximum). This last name must not include blanks.
Notes
Generates an HTML report for the probabilistic analysis. An HTML report includes a description of the de-
terministic model, the probabilistic model, the probabilistic methods used for the analyses and the results
obtained from the analyses.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1238
PDWRITE
Menu Paths
Main Menu>Prob Design>Prob Results>Report>Generate Report
PEMOPTS, TOLER, Method
Defines percentage tolerance and error estimation method for electrostatic p-Method solution.
SOLUTION: Controls
MP ME <> <> <> <> <> <> EM <> <> PP <> EME MFS
TOLER
Percentage tolerance used to determine which elements to fix at the current p-level. If an element's error
energy is less that this tolerance, its p-level will be fixed at the current level. Defaults to 5%.
Method
Specifies the error estimation method.
ZZ
Zienkiewicz-Zhu error estimator. Applicable to problems with a single dielectric material (default).
DUAL
Dual field method. Applicable to problems with single or multiple dielectric materials.
Notes
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Solution Options
Main Menu>Solution>Load Step Opts>p-Method>Solution Options
PERBC2D, LOC1, LOC2, LOCTOL, R1, R2, TOLR, OPT, PLNOPT
Generates periodic constraints for 2-D planar magnetic field analyses.
PREP7: Special Purpose
MP <> <> <> <> <> <> <> EM EH <> PP <> EME <>
LOC1
Constant coordinate location of the first plane of nodes. For PLNOPT = 1 or 2, the constant coordinate
location is the global Cartesian coordinate system [CSYS,0] location in the X or Y direction respectively.
For PLNOPT = 0, the location is the angle in the global cylindrical coordinate system [CSYS,1].
LOC2
Constant coordinate location of the second plane of nodes. For PLNOPT = 1 or 2, the constant coordinate
location is the global Cartesian coordinate system [CSYS,0] location in the X or Y direction respectively.
For PLNOPT = 0, the location is the angle (in degrees) in the global cylindrical coordinate system [CSYS,1].
1239
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PERBC2D
LOCTOL
Tolerance on the constant coordinate location for node selection. Defaults to .00001 for PLNOPT = 1
or 2 and .001 degrees for PLNOPT = 0.
R1
Minimum coordinate location along the second plane of nodes. For PLNOPT = 1 or 2, the coordinate
location is the global Cartesian coordinate system location in the Y or X direction respectively. For
PLNOPT = 0, the coordinate location is the radial coordinate value in the global cylindrical coordinate
system. Periodic conditions are not applied to nodes at this location.
R2
Maximum coordinate location along the second plane of nodes. For PLNOPT = 1 or 2, the coordinate
location is the global Cartesian coordinate system location in the Y or X direction respectively. For
PLNOPT = 0, the coordinate location is the radial coordinate value in the global cylindrical coordinate
system. Periodic conditions are not applied to nodes at this location.
TOLR
Tolerance dimension on node selection along the plane of nodes. Defaults to .00001.
OPT
Periodic option:
0
Odd symmetry (default). Apply constraint equations such that AZ(i) = -AZ(j).
1
Even symmetry. Apply node coupling such that AZ(i) = AZ(j).
PLNOPT
Symmetry plane option:
0
Planes of constant angle in the global cylindrical coordinate system [CSYS,1].
1
Planes parallel to the global Cartesian X axis [CSYS,0].
2
Planes parallel to the global Cartesian Y axis [CSYS,0].
Notes
PERBC2D invokes an ANSYS macro which generates periodic boundary condition constraints for 2-D planar
magnetic field analysis. The macro is restricted to node pairs sharing common coordinate values along
symmetry planes separated by a constant coordinate value. Planes (or lines) must lie at either constant
angles (PLNOPT = 0), constant X values (PLNOPT = 1), or constant Y values (PLNOPT = 2). PERBC2D applies
constraint equations (OPT = 0, odd symmetry) or node coupling (OPT = 1, even symmetry) to each node
pair sharing a common coordinate value along the symmetry planes. By default, periodic conditions are not
applied at the first and last node pairs on the symmetry planes unless the input location values, R1 and R2,
are adjusted to be less than or greater than the actual node coordinate values. Nodes are selected for ap-
plication of the constraints using the NSEL command with tolerances on the constant coordinate location
(LOCTOL) and the coordinate location along the plane (RTOL).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1240
PERBC2D
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>VectorPot>Periodic
BCs
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>VectorPot>Periodic BCs
PERI, DX, DY, DZ
Specifies periodic boundary conditions in an incompressible flow analysis.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
DX
Offset in the X direction of the second boundary from the first.
DY
Offset in the Y direction of the second boundary from the first.
DZ
Offset in the Z direction of the second boundary from the first.
Notes
Specified offsets must be consistent with the flow field coordinate system, as dictated by the FLUID141 and
FLUID142 KEYOPT(3) setting.
The meshes at the two boundaries must be identical.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
PEXCLUDE, ELEM
Specifies elements to be excluded from p-level escalations.
SOLUTION: p-Method
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
ELEM
Element number of element to be excluded. If ELEM = ALL, exclude all elements. If ELEM = STAT, give
status of excluded elements. If ELEM = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI). A component name may be substituted for ELEM.
Command Default
No elements are excluded.
1241
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PEXCLUDE
Notes
The elements chosen by this command will be excluded from convergence calculations. As a result, these
elements will not have their p-levels increased throughout the solution iterations.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Stored Energy>Exclude Elems
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Strain Energy>Exclude Elems
Main Menu>Solution>Load Step Opts>p-Method>Stored Energy>Exclude Elems
Main Menu>Solution>Load Step Opts>p-Method>Strain Energy>Exclude Elems
PFACT, TBLNO, Excit, Parcor
Calculates participation factors for the PSD or multi-point response spectrum table.
SOLUTION: Spectrum Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
TBLNO
Input PSD (Power Spectral Density) table number for which participation factors are to be calculated.
Excit
Label defining the location of excitation:
BASE
Base excitation (default).
NODE
Nodal excitation.
Parcor
Label defining excitation type (applies only to SPOPT,PSD analysis). Used only when partially correlated
excitation is due to wave propagation or spatial correlation. Defaults to partially correlated excitation as
defined by COVAL and QDVAL commands.
WAVE
Excitation defined by PSDWAV command.
SPAT
Excitation defined by PSDSPL command.
Notes
Calculates the participation factors for a particular PSD or multi-point response spectrum table defined with
the PSDVAL command. The Jobname.DB file must contain modal solution data in order for this command
to calculate the participation factor. There must be a PFACT command for each excitation spectrum. You
are limited to 100 excitations.
This command is also valid in PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1242
PFACT
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>MultiPt>Calculate PF
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>PSD>Calculate PF
Main Menu>Solution>Load Step Opts>Spectrum>MultiPt>Calculate PF
Main Menu>Solution>Load Step Opts>Spectrum>PSD>Calculate PF
PFLUID, DENS
Defines the contained fluid density for a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
DENS
Density of the contained fluid.
Notes
See the PREP7 RUN command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Specifications
PGAP, NLOC, K, DX, DY, DZ, GAP, ELEM
Defines a spring-gap constraint in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Node where gap is to be placed. Defaults to current run starting point.
K
Spring constant value (must be positive).
DX, DY, DZ
Increment (in terms of the active coordinate system components) to determine gap ground point. Element
length must not be zero. Constraints are automatically generated at the ground point.
GAP
Gap size (defaults to the element length).
ELEM
Element number to be assigned to gap (defaults to MAXEL + 1).
1243
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PGAP
Notes
Defines a spring-gap constraint (gap element, CONTAC52) at a given location in a piping run. Gives spring
constraint resistance after a specified gap is closed. See the RUN command.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Spring-Gap Supp
PGRAPH, Option, Fname, Fext, --
Specifies the location from which graphics data will be retrieved for viewing.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Option
Switch that controls the PGR file write operations :
Off
Use the data currently in the data base for display (default).
On
Use the PowerGraphics data in the PGR file for display.
Fname
File name and directory path (248 characters maximum, including directory) from which to read the PGR
file. If you do not specify a directory path, it will default to your working directory and you can use all
248 characters for the file name.
The file name defaults to Jobname, or File if no jobname is specified.
Fext
Filename extension (8 character maximum).
--
Unused field.
Notes
Menu Paths
Main Menu>Solution>Output Controls>PGR File
Main Menu>General Postproc>Write PGR File
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1244
PGRAPH
PGRSET, Lstep, SBSTEP, --, KIMG, TIME, --, NSET
Defines the data set to be read from the PGR file.
POST1: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lstep
Load step number of the data set to be read (defaults to 1):
N
Read load step N.
FIRST
Store the first data set on the PGR file (SBSTEP and TIME are ignored).
LAST
Store the last data set on the PGR file. (SBSTEP and TIME are ignored).
NEXT
Store the next data set on the PGR file (SBSTEP and TIME are ignored). If at the last data set, the
first data set will be read as the next.
LIST
Scan the PGR file and list a summary of each load step. (FACT, KIMG, TIME and ANGLE are ignored.)
SBSTEP
Substep number (within Lstep). If Lstep = LIST, SBSTEP = 0 or 1 lists the basic step information,
whereas SBSTEP = 2 also lists the load step title, and labels imaginary data sets if they exist. Default
maximum is 1000. When the number of substeps exceeds this limit, you need to issue SET,Lstep,LAST
to bring in the 1000th load step. Use /CONFIG to increase the limit.
--
Unused field.
KIMG
Used only with results from complex analyses.
0
Store real part of complex solution.
1
Store imaginary part.
TIME
Time-point identifying the data set to be read. For the harmonic response analyses, time corresponds
to the frequency. If both Lstep and SBSTEP are zero (or blank), read data set at time = TIME. If TIME
is beyond the last time point on the file, the last time point will be used. If TIME doe not match a time
value on the PGR file, the nearest time point in the data set will be used.
--
Unused field.
NSET
Data set number of the data set to be read. If a positive value for NSET is entered, Lstep, SBSTEP,
KIMG, and TIME are ignored. Available set numbers can be determined by PGRSET,LIST.
1245
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PGRSET
Notes
Defines the data set to be read from the PGR file into the database. Various operations may also be performed
during the read operation.
Menu Paths
Main Menu>Solution>Output Ctrls>PGR File
Main Menu>General Postproc>Write PGR File
PGSAVE, Fname, Fext, --, DataType, InteriorKey, Append
Creates a PowerGraphics (PGR) file from results data.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname, or File if no jobname is specified.
Fext
Filename extension (8 character maximum).
--
Unused field.
DataType
Data type to create. This specification applies to discontinuous results data such as stress, strain, and
field data. .
0
Save as nodal-averaged data (default). Used by the PLNSOL command.
1
Reserved.
2
Save both nodal averaged and unaveraged data. Used by the PLESOL and PLNSOL commands.
InteriorKey
Key that controls whether or not internal model data is saved to the PGR file. Internal data is required
in order to use graphical slicing, capping, vector displays, or Isosurface displays for data display. Interior
data is also required for the AVRES,,FULL option, where surface results take into consideration the interior
element contributions.
0
Save exterior surface data only. (default)
1
Save exterior and interior data.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1246
PGSAVE
Append
Append or overwrite data to the specified.
0
Append the data to the existing file (default). Note that the geometry must not have been changed
between the write operations.
1
Overwrite the data file.
Notes
The user may elect to write results to the PGR file in a specified order. The specifications (including /ESHAPE
parameters) must be defined before saving the file. The GUI will allow the user to write the current set, or
all sets from the results file.
For contact analysis, only flex-flex analysis is supported.
Menu Paths
Main Menu>Solution>Output Ctrls>PGR File
Main Menu>General Postproc>Write PGR File
PGSELE, Type, Item, --, VMIN, VMAX, VINC
Select a subset of elements for display with the PGR viewer.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying type of select
S
Select a new set (default)
R
Reselect a set from the current set
A
Additionally select a set, extending the current set.
U
Unselect a set from the current set.
ALL
Restore the full set. Subsequent command arguments are ignored.
INVE
Invert the current set (change selected to unselected; change unselected to selected). Subsequent
command arguments are ignored.
STAT
Display the current select status. Subsequent command arguments are ignored.
Item
Label identifying the data. Valid labels are:
1247
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PGSELE
If Type = STAT and Item = (blank) or ELEM, the element selection information is provided.
Type - Element type number
Mat - Material number
Real - Real constant number
Sec - Section identification number
Comp or CM - Component name(s). (see notes, below)
Pick - Pick entities via a picking window. (see notes, below)
(Defaults to Type).
--
Unused field.
VMIN
Minimum value of Item range. Ranges are attribute numbers.
VMAX
Maximum value for Item range. VMAX defaults to VMIN for input values.
VINC
Value increment within Item range. Used only with integer ranges (such as for attribute numbers). De-
faults to 1. VINC cannot be negative.
Command Default
All elements are selected.
Notes
When Item = Pick, interactive picking is enabled. The displayed facets of the elements you pick will be
flagged according to how they are resolved in PowerGraphics. Exterior elements will be flagged on each
facet surface, while elements with no external (surface) component will be flagged at the centroid of that
element.
When Item = Comp, the subsequent values for VMIN, VMAX, and VINC, are replaced by valid component
names (32 character max). You can name up to 16 components in this fashion. All named components must
contain elements.
This command is valid in POST1
Menu Paths
Main Menu>General Postproc>Results Viewer
PGWRITE, Label, Fname, Fext, --, DataType, InteriorKey, Append
Writes selected solution data to the PGR file for faster post processing access.
SOLUTION: Misc Loads
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Switch that controls the PGR file write operations :
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1248
PGWRITE
Off
Do not write the PGR file during the solution.
On
Write the PGR file during solution (default).
STAT
Display the current status of the PGR file settings.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname, or File if no jobname is specified.
Fext
Filename extension (8 character maximum).
--
Unused field.
DataType
Data type to create. This specification applies to discontinuous results data such as stress, strain, and
field data. .
0
Save as nodal-averaged data (default). Used by the PLNSOL command.
1
Reserved.
2
Save both nodal averaged and unaveraged data. Used by the PLESOL and PLNSOL commands.
InteriorKey
Key that controls whether or not internal model data is saved to the PGR file. Internal data is required
in order to use graphical slicing, capping, vector displays, or Isosurface displays for data display. Interior
data is also required for the AVRES,,FULL option, where surface results take into consideration the interior
element contributions.
0
Save exterior surface data only. (default)
1
Save exterior and interior data.
Append
Append or overwrite data to the specified file. .
0
Append the data to the existing file (default). Note that the geometry must not have been changed
between the write operations.
1
Overwrite the data file.
1249
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PGWRITE
Notes
In interactive mode, you cannot read the PGR file while PGWRITE is active. If you solve your model and
want to plot the model while still in SOLUTION, you must first close the PGR file by issuing PGWRITE,Off.
This occurs automatically when you exit SOLUTION and enter POST1.
When you wish to view your stress results in another coordinate system, you must generate your PGR file
from the results file, in POST1, in that coordinate system.
Menu Paths
Main Menu>Solution>Output Ctrls>PGR File
Main Menu>General Postproc>Write PGR File
PHYSICS, Option, Title, Fname, Ext, --
Writes, reads, or lists all element information
PREP7: Special Purpose
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Option
Specifies what to do with element information:
WRITE
Write all appropriate element types, key options, real constants, material properties, solution analysis
options, load step options, constraint equations, coupled nodes, defined components, and GUI
preference settings to the file specified with the Fname and Ext arguments.
READ
Deletes all solution information (material properties, solution options, load step options, constraint
equations, coupled nodes, results, and GUI preference settings) then reads all the information listed
above into the ANSYS database from the location specified by the Fname and Ext arguments.
LIST
Lists currently defined physics files and their titles.
DELETE
Deletes a specified physics file and its title from the database.
CLEAR
Deletes all material properties, solution options, load step options, constraint equations, coupled
nodes, results, and GUI preference settings from the database. Does NOT clear the active physics file
title from the database.
STATUS
Displays information about all active elements and settings.
Title
A user-defined title that quickly identifies a set of physics settings. For example, you might use "Fluid,"
"Structural," or "Magnetic" as titles. A title can contain up to 64 characters. It can be entered in lower or
upper case. Lower case is internally converted to upper case within the program.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1250
PHYSICS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname. Previous data on this file, if any, are overwritten.
Ext
Filename extension (8 character maximum).
The extension defaults to PHn if Fname is blank, where n is a number between one and nine, depending
on how many times you have issued the PHYSICS command. (You can have up to nine separate physics
files.) If you issue the command more than nine times, the ANSYS program will require you to delete an
existing file.
--
Unused field.
Notes
Use the PHYSICS command when you are performing a multiphysics analysis that involves two different
disciplines (for example, CFD and structural analysis) and you cannot solve both analyses simultaneously.
Once you have set up physics environments for both analyses, you can use the PHYSICS,READ command
to change between the defined physics environments. For more information about doing multiphysics
analyses, see Sequential Coupled-Field Analysis in the Coupled-Field Analysis Guide.
The PHYSICS command outputs all solution information, including analysis options, to the Jobname.PHn
file described above. Although it also outputs components, the ANSYS program does not list entities (nodes,
elements, lines, etc.).
PHYSICS,WRITE will overwrite existing physics files with the same title (even if the name is different). In
other words, if the directory has a physics file with the same title as the active physics file title, but a different
name, the PHYSICS,WRITE command will overwrite the existing physics file and use the existing filename,
not the filename specified on the PHYSICS,WRITE command.
Menu Paths
Main Menu>Preprocessor>Physics>Environment>Clear
Main Menu>Preprocessor>Physics>Environment>Delete
Main Menu>Preprocessor>Physics>Environment>List
Main Menu>Preprocessor>Physics>Environment>Read
Main Menu>Preprocessor>Physics>Environment>Status
Main Menu>Preprocessor>Physics>Environment>Write
Main Menu>Solution>Physics>Environment>Clear
Main Menu>Solution>Physics>Environment>Delete
Main Menu>Solution>Physics>Environment>List
Main Menu>Solution>Physics>Environment>Read
Main Menu>Solution>Physics>Environment>Status
Main Menu>Solution>Physics>Environment>Write
1251
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PHYSICS
/PICE, Item, --, KEY
Shows initial conditions on elements as contours on displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Item
Label identifying the item:
VFRC
Volume fraction.
--
Unused field.
KEY
Symbol key:
0
Do not show initial condition contours.
1
Show initial condition contours.
Command Default
No initial condition contours displayed.
Notes
Shows initial conditions as contours on displays for the selected elements. Use /PSTATUS or /PICE,STAT to
display settings. Use /PICE,DEFA to reset all specifications back to default.
/PICE is overridden if /PBF, /PSF, and /PBC are all on.
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Utility Menu>PlotCtrls>Symbols
PINCLUDE, ELEM
Specifies elements to be included in p-level escalations.
SOLUTION: p-Method
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
ELEM
Element number of element to be included. If ELEM = ALL, include all elements. If ELEM = STAT, give
status of included elements. If ELEM = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI). A component name may be substituted for ELEM.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1252
/PICE
Command Default
All elements are included.
Notes
The elements chosen by this command may have their p-levels increased throughout the solution iterations.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Stored Energy>Include Elems
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Strain Energy>Include Elems
Main Menu>Solution>Load Step Opts>p-Method>Stored Energy>Include Elems
Main Menu>Solution>Load Step Opts>p-Method>Strain Energy>Include Elems
PINSUL, DENS, ITK
Defines the external insulation constants in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
DENS
Insulation density.
ITK
Insulation thickness.
Command Default
No insulation.
Notes
Defines the external insulation constants in a piping run. See the RUN command.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Specifications
1253
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PINSUL
PIPE
Specifies "Pipe modeling" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Piping Module
PIVCHECK, KEY, PRNTCNTRL
Prevents a batch mode, linear static analysis from stopping when a negative or zero equation solver pivot
value is encountered.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
KEY
Determines whether to stop or continue an analysis when a negative or zero equation solver pivot value
is encountered:
ON
Default value. ANSYS checks for negative or zero pivot values on linear static analyses performed
with the sparse and PCG solvers. When one is encountered, an error is issued, stopping the job. A
negative pivot value may be valid for some multiphysics analyses (e.g. Electromagnetic or Thermal);
this key has no effect in these cases.
OFF
Pivots will not be checked. This key continues the analysis in spite of a zero or negative pivot value.
The program will proceed until other error checking routines are encountered.
PRNTCNTRL
Provides print options. Print output with these options will be sent to the default output file, not to the
files created by the nonlinear diagnostic tools (NLDIAG).
ONCE or blank
Default value. Print maximum and minimum pivot information once per loadstep if the SPARSE
solver (the default solver) is used.
EVERY
Print maximum and minimum pivot information every Newton-Raphson iteration or at every iteration
within a substep of every loadstep. This option is provided for nonlinear diagnostics.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1254
PIPE
Command Default
ON, stop the analysis and report an error status to the user.
Notes
This command is valid only for linear static analyses. In a nonlinear analysis, a negative pivot may be valid.
Normally, rigid body motions in a nonlinear analysis will be trapped by error routines checking infinitely
large displacements (DOF limit exceeded) or nonconvergence status. An under-constrained model may avoid
the pivot check, but fail with a DOF limit exceeded error.
This command is applicable only to batch mode (not in the GUI). While in an ANSYS interactive session, it
is generally desirable to remain in the session, even though a pivot error has occurred. Warning messages
will be displayed on the screen, allowing the user to take corrective action.
Finally, machine precision will determine if a warning or an error is issued for the small pivot case. In all
cases there will be some notification that the model needs to be checked closely for an accurate solution.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
PLCAMP, Option, SLOPE, UNIT, FREQB, Cname, STABVAL
Plots Campbell diagram data for applications involving rotating structure dynamics.
POST1: Special Purpose
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Option
Flag to activate or deactivate sorting of forward or backward whirl frequencies:
0 (OFF or NO)
No sorting.
1 (ON or YES)
Sort. This value is the default.
SLOPE
The slope of the line to be plotted. This value must be greater than 0. The line represents the number
of excitations per revolution of the rotor. For example, SLOPE = 1 represents one excitation per revolution,
usually resulting from unbalance. The default is no line.
UNIT
Specifies the unit of measurement for rotational angular velocities:
RDS
Rotational angular velocities in radians per second (rad/s). This value is the default.
RPM
Rotational angular velocities in revolutions per minute (RPMs).
1255
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLCAMP
FREQB
The beginning, or lower end, of the frequency range of interest. The default is zero.
Cname
The rotating component name.
STABVAL
Flag to plot the stability values:
0 (OFF or NO)
Plot the frequencies (the imaginary parts of the eigenvalues in Hz). This value is the default.
1 (ON or YES)
Plot the stability values (the real parts of the eigenvalues in Hz).
2
Plot the logarithmic decrements.
For more information about complex eigenvalues and corresponding logarithmic decrements, see
Damped Method in the Theory Reference for the Mechanical APDL and Mechanical Applications.
Notes
To take the gyroscopic effect into account when plotting a Campbell diagram, first issue the CORIOLIS
command in the SOLUTION module.
The PLCAMP command is valid only for modal analyses (QR damped [MODOPT,QRDAMP] or damped
[MODOPT,DAMP] methods only). The command works over two or more load step results generated with
an ascending order of rotational velocity (OMEGA or CMOMEGA).
Complex eigenvalues are necessary (MODOPT,QRDAMP,,,,Cpxmod = ON), and you must specify the number
of modes to expand (MXPAND) at each load step.
In some cases where modes are not in the same order from one load step to the other, sorting (Option =
1) the frequencies can help to obtain a correct plot. Sorting is based on the comparison between complex
mode shapes calculated at two successive load steps.
At each load step, ANSYS compares the mode shape to the loads at other loads steps to determine whirl
direction at the load step. If applicable, a label appears (in the plot legend) representing each whirl mode
(BW for backward whirl and FW for forward whirl).
ANSYS considers the SLOPE value only if the frequencies are plotted (STABVAL = OFF).
At each load step, ANSYS checks for instability (based on the sign of the real part of the eigenvalue). The
labels stable or unstable appear in the plot legend for each frequency curve.
The rotational velocities of a named component (Cname) are displayed on the X-axis.
To plot the Campbell diagram for a prestressed structure, first issue a CAMPBELL,ON command in the static
portion of the analysis.
In general, ANSYS, recommends plotting a Campbell diagram only when your analysis is performed in a
stationary reference frame (CORIOLIS,,,,RefFrame = ON).
For a usage example of the PLCAMP command, see Sample Campbell Diagram Analysis.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1256
PLCAMP
Menu Paths
Main Menu>General Postproc>Rotor Dynamics>Plot Campbell
PLCINT, ACTION, ID, Node, Cont, Dtype
Plots the contour integral (CINT) result data.
POST1: Results
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Action
PATH
Plots CINT quantities according to path number (default).
FRONT
Plots CINT quantities distribution along the crack front.
ID
Crack ID number.
Node
Crack tip node number (default = ALL).
Use only for ACTION = PATH. Plots CINT contour for an individual crack tip node.
Cont
Contour number (Default = ALL).
Use only for ACTION = FRONT. Plots CINT distribution along the crack for a given path.
Dtype
Data type to output:
JINT
J-integral (default)
IIN1
Interaction integral 1
IIN2
Interaction integral 2
IIN3
Interaction integral 3
K1
Mode 1 stress-intensity factor
K2
Mode 2 stress-intensity factor
K3
Mode 3 stress-intensity factor
1257
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLCINT
Menu Paths
This command cannot be accessed from a menu.
PLCONV, Item, Comp, NODE, Surf
Plots the convergence curve for specified items from a p-method solution.
POST1: Results
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Item
Label identifying the convergence criterion to be plotted. (Valid component labels are described in
Table 243: PLCONV - Valid Item and Component Labels (p. 1259).)
ALL
Simultaneously plot the convergence curve for all specified Item criteria versus p-level (default).
Does not require a component label.
SE
Plot strain energy curve or stored electrostatic energy curve, versus p-level. Does not require a
component label.
S
Plot stress curve versus p-level. Valid component labels are described in Table 243: PLCONV - Valid
Item and Component Labels (p. 1259) below.
EPEL
Plot strain curve versus p-level. Valid component labels are described in Table 243: PLCONV - Valid
Item and Component Labels (p. 1259) below.
U
Plot translational displacement curve versus p-level. Valid component labels are described in
Table 243: PLCONV - Valid Item and Component Labels (p. 1259) below.
ROT
Plot structural rotation versus p-level (valid only for SHELL150).
VOLT
Plot electric potential curve versus p-level. Does not require a component label.
EF
Plot electric field strength curve versus p-level. Valid component labels are described in
Table 243: PLCONV - Valid Item and Component Labels (p. 1259) below.
D
Plot electric flux density curve versus p-level. Valid component labels are described in
Table 243: PLCONV - Valid Item and Component Labels (p. 1259) below.
EFORC
Plot global electrostatic (Maxwell Stress Tensor) force curve versus p-level. Valid component labels
are described in Table 243: PLCONV - Valid Item and Component Labels (p. 1259) below.
Comp
Component of the Item. Not required for Item = ALL, SE, VOLT, or EFORC. Valid Comp values (and as-
sociated Item values) are in Table 243: PLCONV - Valid Item and Component Labels (p. 1259) below. (Defaults
to ALL: plot the convergence curve for all Item and Comp criteria simultaneously.)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1258
PLCONV
NODE
Node number where convergence checking was performed. Valid only for Item = S, EPEL, U, ROT, VOLT,
EF, or D. Defaults to all nodes having Item and Comp previously specified as convergence criteria
[PCONV].
Surf
Surface of a p-element shell structure where the convergence curve for NODE will be plotted. Valid only
for Item = S or EPEL. Defaults to all surfaces.
TOP
Top surface (default).
MID
Middle surface
BOT
Bottom surface.
Command Default
Display all specified criteria simultaneously.
Notes
The previously specified convergence criterion [PCONV] is graphed versus the characteristic p-level. The
characteristic p-level is the polynomial level range from the minimum defined [PPRANGE] to the maximum
p-level reached by any element in the model. The Item value corresponds to the convergence criterion
indicated by PCONV. All Item values previously designated by the PCONV command will be plotted.
Table 243 PLCONV - Valid Item and Component Labels
Description Comp Item
All convergence criteria (default) ALL ALL
Strain energy or stored electrostatic energy SE
Component stress X, Y, Z, XY, YZ, XZ S
Principal stress 1, 2, 3 "
Stress intensity INT "
Equivalent stress EQV "
Component elastic strain X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain 1, 2, 3 "
Elastic strain intensity INT "
Elastic equivalent strain EQV "
X, Y, or Z structural displacement or vector sum X, Y, Z, SUM U
All applicable displacement items ALL "
X, Y, or Z structural rotation or vector sum (valid only for
SHELL150)
X, Y, Z, SUM ROT
Electric potential VOLT
Electric field strength X, Y, Z, SUM EF
Electric flux density X, Y, Z, SUM D
1259
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLCONV
Description Comp Item
Global electrostatic (Maxwell Stress Tensor) force X, Y, Z, SUM EFORC
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>p-Method>p-Convergence
PLCPLX, KEY
Specifies the part of a complex variable to display.
POST26: Display
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
KEY
Complex variable part:
0
Amplitude.
1
Phase angle.
2
Real part.
3
Imaginary part.
Notes
Used only with harmonic analyses (ANTYPE,HARMIC).
All results data are stored in the form of real and imaginary components and converted to amplitude and/or
phase angle as specified via the PLCPLX command. The conversion is not valid for derived results (such as
principal stress/strain, equivalent stress/strain and USUM).
Menu Paths
Main Menu>TimeHist Postpro>Settings>Graph
PLCRACK, LOC, NUM
Displays cracking and crushing locations in SOLID65 elements.
POST1: Special Purpose
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
LOC
Location at which symbols are to be displayed:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1260
PLCPLX
0
Plot symbols at integration points (default).
1
Plot symbol at element centroids (averaged).
NUM
Crack to be displayed:
0
Plot all cracks (default).
1
Plot only the first crack.
2
Plot only the second crack.
3
Plot only the third crack.
Notes
PLCRACK displays circles at locations of cracking or crushing in concrete elements. Cracking is shown with
a circle outline in the plane of the crack, and crushing is shown with an octahedron outline. If the crack has
opened and then closed, the circle outline will have an X through it. Each integration point can crack in up
to three different planes. The first crack at an integration point is shown with a red circle outline, the second
crack with a green outline, and the third crack with a blue outline.
Symbols shown at the element centroid (LOC = 1) are based on the status of all of the element's integration
points. If any integration point in the element has crushed, the crushed (octahedron) symbol is shown at
the centroid. If any integration point has cracked or cracked and closed, the cracked symbol is shown at the
element centroid. If at least five integration points have cracked and closed, the cracked and closed symbol
is shown at the element centroid. Finally, if more than one integration point has cracked, the circle outline
at the element centroid shows the average orientation of all cracked planes for that element.
Portions of this command are not supported by PowerGraphics [/GRAPHICS,POWER].
Menu Paths
Main Menu>General Postproc>Plot Results>ConcPlot>Crack/Crush
PLDISP, KUND
Displays the displaced structure.
POST1: Results
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS
KUND
Undisplaced shape key:
0
Display only displaced structure.
1261
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLDISP
1
Overlay displaced display with similar undisplaced display (appearance is system-dependent).
2
Same as 1 except overlay with undisplaced edge display (appearance is system-dependent).
Notes
Displays the displaced structure for the selected elements.
For information on true scale plots, refer to the description of the /DSCALE command [/DSCALE,,1.0].
Menu Paths
Main Menu>General Postproc>Plot Results>Deformed Shape
Utility Menu>Plot>Results>Deformed Shape
Utility Menu>PlotCtrls>Animate>Deformed Shape
PLESOL, Item, Comp, KUND, Fact
Displays the solution results as discontinuous element contours.
POST1: Results
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Label identifying the item. Valid item labels are shown in Table 244: PLESOL - Valid Item and Component
Labels (p. 1263) below. Some items also require a component label.
Comp
Component of the item (if required). Valid component labels are shown in Table 244: PLESOL - Valid Item
and Component Labels (p. 1263) below.
KUND
Undisplaced shape key:
0
Do not overlay undeformed structure display
1
Overlay displaced contour plot with undeformed display (appearance is system-dependent)
2
Overlay displaced contour plot with undeformed edge display (appearance is system-dependent)
Fact
Scale factor for 2-D display of contact items (defaults to 1). A negative scaling factor may be used to
invert the display.
Notes
Displays the solution results as element contours discontinuous across element boundaries for the selected
elements. For example, PLESOL,S,X displays the X component of stress S (that is, the SX stress component).
Various element results depend on the calculation method and the selected results location (AVPRIN, RSYS,
and ESEL). Contours are determined by linear interpolation within each element, unaffected by the surround-
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1262
PLESOL
ing elements (i.e., no nodal averaging is performed). The discontinuity between contours of adjacent elements
is an indication of the gradient across elements. Component results are displayed in the active results co-
ordinate system [RSYS] (default is the global Cartesian). See the ETABLE and PLETAB commands for displaying
items not available through this command (such as line element results).
For PowerGraphics displays [/GRAPHICS,POWER], results are plotted only for the model exterior surface. The
items marked with [1 (p. 1267)] in Table 244: PLESOL - Valid Item and Component Labels (p. 1263) are not supported
by PowerGraphics.
Table 244 PLESOL - Valid Item and Component Labels
Description Comp Item
Valid item and component labels for element results are:
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity. INT "
Equivalent stress. EQV "
Component elastic strain. X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity. INT "
Elastic equivalent strain. EQV "
Component plastic strain. X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity. INT "
Plastic equivalent strain. EQV "
Component creep strain. X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity. INT "
Creep equivalent strain. EQV "
Component thermal strain. X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity. INT "
Thermal equivalent strain. EQV "
Swelling strain. EPSW
Component total mechanical strain (EPEL + EPPL + EPCR). X, Y, Z, XY, YZ, XZ EPTO
Principal total mechanical strain. 1, 2, 3 "
Total mechanical strain intensity. INT "
Total mechanical equivalent strain. EQV "
Total mechanical and thermal strain (EPEL + EPPL + EPCR
+ EPTH).
X, Y, Z, XY, YZ, XZ EPTT
Principal total mechanical and thermal strain. 1, 2, 3 "
Total mechanical and thermal strain intensity. INT "
1263
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLESOL
Description Comp Item
Total mechanical and thermal equivalent strain. EQV "
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Accumulated equivalent creep strain. CREQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Elastic strain energy density. ELASTIC SEND
Plastic strain energy density. PLASTIC "
Creep strain energy density. CREEP "
Damage variable DMG CDM
Maximum previous strain energy for virgin material LM "
Maximum of all failure criteria defined at the current loca-
tion. [1][4]
MAX FAIL
Maximum Strain Failure Criterion. [1][4] EMAX "
Maximum Stress Failure Criterion. [1][4] SMAX "
Tsai-Wu Strength Index Failure Criterion. [1][4] TWSI "
Inverse of Tsai-Wu Strength Ratio Index Failure Criterion.
[1][4]
TWSR "
User-defined failure criteria. [1][4][5] USR1, USR2, ..., USR9 "
Layer number where the maximum of all failure criteria
over the entire element occurs. [1][4]
LAY FCMX
Number of the maximum-failure criterion over the entire
element: 1 - EMAX 2 - SMAX 3 - TWSI 4 - TWSR 5~13 -
USR1~USR9
[1 (p. 1267)][4 (p. 1267)]
FC "
Value of the maximum failure criterion over the entire ele-
ment. [1][4]
VAL "
State variable. [1] 1, 2, 3, ... N SVAR
Gasket component stress. X, XY, XZ GKS
Gasket component total closure. X, XY, XZ GKD
Gasket component total inelastic closure. X, XY, XZ GKDI
Gasket component thermal closure. X, XY, XZ GKTH
Interface traction (stress) X, XY, XZ SS
Interface separation X, XY, XZ SD
Contact status[3]: STAT[2] CONT
3-closed and sticking
2-closed and sliding
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1264
PLESOL
Description Comp Item
1-open but near contact
0-open and not near contact
Contact penetration PENE "
Contact pressure PRES "
Contact friction stress SFRIC "
Contact total stress (pressure plus friction) STOT "
Contact sliding distance SLIDE "
Contact gap distance GAP "
Total heat flux at contact surface FLUX "
Total number of contact status changes during substep. CNOS
Fluid penetration pressure FPRS
Component thermal gradient or vector sum. X, Y, Z, SUM TG
Component thermal flux or vector sum. X, Y, Z, SUM TF
Component pressure gradient or vector sum. X, Y, Z, SUM PG
Component electric field or vector sum. X, Y, Z, SUM EF
Component electric flux density or vector sum. X, Y, Z, SUM D
Component magnetic field intensity or vector sum. X, Y, Z, SUM H
Component magnetic flux density or vector sum. X, Y, Z, SUM B
Component electromagnetic force or vector sum [1]. X, Y, Z, SUM FMAG
Pointing vector component or sum [1]. X, Y, Z, SUM P
Structural error energy [1]. SERR[6]
Absolute value of the maximum variation of any nodal
stress component [1].
SDSG[6]
Thermal error energy [1]. TERR[6]
Absolute value of the maximum variation of any nodal
thermal gradient component [1].
TDSG[6]
X, Y, or Z structural force [1]. X, Y, Z F
X, Y, or Z structural moment [1]. X, Y, Z M
Heat flow [1]. HEAT
Fluid flow [1]. FLOW
Current flow [1]. Use FORCE for type. AMPS
Charge [1]. Use FORCE for type. CHRG
Magnetic flux [1]. FLUX
X, Y, or Z magnetic current segment component [1]. X, Y, Z CSG
"Stiffness" energy or thermal heat dissipation. Same as
TENE[1].
SENE
Elemental energy dissipation due to stabilization [1]. STEN
Thermal heat dissipation or "stiffness" energy. Same as SENE
[1].
TENE
1265
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLESOL
Description Comp Item
Kinetic energy [1]. KENE
Element Joule heat generation [1]. JHEAT
Source current density for low-frequency magnetic analyses.
Total current density (sum of conduction and displacement
X, Y, Z, SUM JS
current densities) in low frequency electric analyses. Com-
ponents (X, Y, Z) and vector sum (SUM). [1].
Total measureable current density in low-frequency electro-
magnetic analyses. (Conduction current density in a low-
X, Y, Z, SUM JT
frequency electric analysis.) Components (X, Y, Z) and vector
sum (SUM). [1].
Conduction current density for elements that support con-
duction current calculation. Components (X, Y, Z) and vector
sum (SUM). [1].
X, Y, Z, SUM JC
Magnetic Reynolds number [1]. MRE
Volume of volume element [1]. VOLU
Centroid X, Y, or Z location (based on shape function) in
the active coordinate system [1].
X, Y, Z CENT
Body temperatures (calculated from applied temperatures)
as used in solution (area and volume elements only).
TEMP BFE
Element summable miscellaneous data value at sequence
number snum (shown in the Output Data section of each
element description found in the Element Reference).
snum SMISC
Element non-summable miscellaneous data value at se-
quence number snum (shown in the Output Data section
of each element description found in the Element Reference).
snum NMISC
Densities used for topological optimization. This applies to
the following types of elements: PLANE82, SOLID92, SOL-
ID95.
TOPO
Material cap plasticity model only: Cohesion; hydrostatic
compaction yielding stress; I1 at the transition point at
C0,X0,K0,ZONE,
DPLS,VPLS
CAP
which the shear and compaction envelopes intersect; zone
= 0: elastic state, zone = 1: compaction zone, zone = 2:
shear zone, zone = 3: expansion zone; effective deviatoric
plastic strain; volume plastic strain.
Material EDP creep model only (not including the cap
model): Equivalent creep stress; equivalent creep strain.
CSIG,CSTR EDPC
Fictive temperature. TEMP FICT
Components of Biots effective stress. X,Y,Z,XY,YZ,ZX ESIG
Principal stresses of Biots effective stress. 1, 2, 3 "
Stress intensity of Biots effective stress. INT "
Equivalent stress of Biots effective stress. EQV "
Total porosity (Gurson material model). TPOR DPAR
Porosity due to void growth. GPOR "
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1266
PLESOL
Description Comp Item
Porosity due to void nucleation. NPOR "
Fluid flow flux in poromechanics. X,Y,Z FFLX
1. Not supported by PowerGraphics
2. Status notation is for all contact elements except CONTAC12 and CONTAC52. If you are using these
elements, check the output listing in the corresponding element documentation.
3. For MPC-based contact definitions, the value of STAT can be negative. This indicates that one or more
contact constraints were intentionally removed to prevent over-constraint. STAT = -3 is used for MPC
bonded contact; STAT = -2 is used for MPC no-separation contact.
4. Works only if FC command information is provided.
5. Works only if user failure criteria routine is provided.
6. Some element- and material-type limitations apply. For more information, see the documentation for
the PRERR command.
Menu Paths
Main Menu>General Postproc>Plot Results>Contour Plot>Element Solu
Utility Menu>Plot>Results>Contour Plot>Elem Solution
PLETAB, Itlab, Avglab
Displays element table items.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Itlab
User-defined label, as specified with the ETABLE command, of item to be displayed.
Avglab
Averaging operation:
NOAV
Do not average element items at common nodes (default).
AVG
Average the element items at common nodes.
Notes
Displays items stored in the table defined with the ETABLE command for the selected elements. For display
purposes, items are assumed to be constant over the element and assigned to each of its nodes. Contour
display lines (lines of constant value) are determined by linear interpolation within each element from the
nodal values. These nodal values have the option of being averaged (values are averaged at a node
whenever two or more elements connect to the same node) or not averaged (discontinuous). The discon-
tinuity between contour lines of adjacent elements is an indication of the gradient across elements.
Portions of this command are not supported by PowerGraphics [/GRAPHICS,POWER].
1267
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLETAB
Menu Paths
Main Menu>General Postproc>Element Table>Plot Elem Table
Main Menu>General Postproc>Plot Results>Contour Plot>Elem Table
Utility Menu>Plot>Results>Contour Plot>Elem Table Data
PLFAR, Lab, Opt, PHI1, PHI2, NPH1, THETA1, THETA2, NTHETA, VAL1, VAL2
Plots electric far fields and far field parameters.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
Plot Parameter:
EF
Electric Field
RCS
Radar cross section
RCSN
Normalized radar cross section
ANT
Antenna parameters
Opt
As shown below, data entered in the Opt field will vary, depending on the print parameter.
Lab
As shown below, data entered in the Lab field will vary, depending on the print option (Opt).
Table 245 Valid Lab Data Labels
Opt Lab
SUM the maximum total rE-field (default) EF
X the maximum rE-field in x-direction
Y the maximum rE-field in y-direction
Z the maximum rE-field in z-direction
PHI the maximum rE-field in -direction
THETA the maximum rE-field in -direction
R the maximum rE-field in r-direction
LHCP the maximum left-hand circularly polarized
component
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1268
PLFAR
Opt Lab
RHCP the maximum right-hand circularly polar-
ized component
ARCP the circular polarization axial ratio
LW3X the maximum dominant component for
an x-polarized aperture with Ludwigs third defini-
tion of cross definition
LW3Y the maximum dominant component for
an y-polarized aperture with Ludwigs third defini-
tion of cross definition
ARCP the linear polarization axial ratio
TOTAL Radar echo area (default)
PP - polarization
RCS and RCSN
PT - polarization
TP - polarization
TT - polarization
HH H-H (horizontal horizontal) polarization
HV H-V (Horizontal Vertical) polarization
VH V-H (Vertical Horizontal) polarization
VV V-V (Vertical Vertical) polarization
TE2D 2-D TE incident plane wave
TM2D 2-D TM incident plane wave
EPCT radiation pattern (default)
EPPL radiation pattern in polar coordinate system
plot
ANT
UMAX the maximum radiation intensity
DGCT directivity in Cartesian coordinate system
plot
DGPL directivity in polar coordinate system plot
DGMX peak directivity
PGCT gain in Cartesian coordinate system plot
PGPL gain in polar coordinate system plot
1269
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLFAR
Opt Lab
PGMX peak gain
RGCT realized gain in Cartesian coordinate system
plot
RGPL realized gain in polar coordinate system plot
RGMX peak realized gain
EFF efficiency
PRAD radiated power
The following PHI1, PHI2, NPHI, THETA1, THETA2, and NTHETA arguments are used only with Lab = EF,
RCS, RCSN and ANT and Opt = EPCT, EPPL, DGCT, DGPL, PGCT, PGPL, RGCT and RGPL.
PHI1, PHI2
Starting and ending angles (degrees) in the spherical coordinate system. Defaults to 0.
NPHI
Number of divisions between the starting and ending angles for data computations. Defaults to 0.
THETA1, THETA2
Starting and ending angles (degrees) in the spherical coordinate system. Defaults to 0.
NTHETA
Number of divisions between the starting and ending angles for data computations. Defaults to 0.
VAL1
Used only with Lab = ANT and Opt = PGCT, PGPL, PGMX, RGCT, RGPL, RGMX, EFF or Lab = RCS or RCSN
and Opt = TE2D, TM2D. For Lab = ANT and Opt = PGCT, PGMX, RGCT, RGMX, EFF, VAL1 = the incident
power, Pinc, in watts. For Lab = RCS or RCSN and Opt = TE2D or TM2D, VAL1 = thickness of the 3-D
model in the z direction, defaults to 1.
VAL2
Used only with Lab = ANT and Opt = PGCT, PGPL, PGMX, EFF. VAL2 = the |s11| of the lossless port with
a single propagation mode to calculate the power delivered to the radiating antenna structure, i.e.
P P s
del inc
( | | ) 1
11
2
. Defaults to 0.
Notes
PLFAR plots electric far fields and far field parameters as determined by the equivalent source principle.
Use this command to plot electric far field, radar cross section, antenna parameters.
See Spherical Coordinates in Hi-Frequency Electromagnetic Analysis Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1270
PLFAR
PLFSS, Filename, Ext, Opt, Lab1, Lab2
Plots reflection and transmission parameters of a frequency selective surface solution.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Filename
Name of the results file generated by the SPFSS command.
Ext
Filename extension (8 character maximum).
Opt
Format of plotted parameter:
MAG
Magnitude (default)
ANG
Angle
DB
Magnitude in dB units
Lab1, Lab2
Labels for reflection and transmission parameters:
REFL
Reflection coefficient.
TRAN
Transmission coefficient.
PREFL
Power reflection coefficient.
PTRAN
Power transmission coefficient.
IL
Insertion loss (dB only).
RL
Return loss (dB only).
Notes
Plots reflection and transmission parameters results of the SPFSS command.
If Lab2 is the same as Lab1 or blank, only one plot is generated.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
1271
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLFSS
PLF2D, NCONT, OLAY, ANUM, WIN
Generates a contour line plot of equipotentials.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
NCONT
Number of contour lines to display. Issue in multiples of 9 (i.e., 9, 18, 27, etc.). Default is 27 contour lines.
OLAY
Overlay:
0
Overlay edge outlines by material number.
1
Overlay edge outlines by real constant number.
ANUM
Highest material or real constant attribute number. Command will cycle through ANUM element display
overlays. Defaults to 10.
WIN
Window number to which command applies. Defaults to 1.
Notes
PLF2D invokes an ANSYS macro which plots equipotentials of the degree of freedom AZ. These equipotential
lines are parallel to flux lines and thus give a good representation of flux patterns. In the axisymmetric case,
the display is actually r*AZ where "r" is the node radius. The macro overlays (OLAY) edge outlines by mater-
ial number or real constant number (ANUM) and allows user control over the number of contour lines to
display (NCONT).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>Contour Plot>2D Flux Lines
Utility Menu>Plot>Results>Flux Lines
PLLS, LabI, LabJ, Fact, KUND
Displays element table items as contoured areas along elements.
POST1: Element Table
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
LabI
Label of element table item [ETABLE] for node I magnitude.
LabJ
Label of element table item for node J magnitude.
Fact
Scale factor for display (defaults to 1). A negative scaling factor may be used to invert the display.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1272
PLF2D
KUND
Undisplaced shape key:
0
Display selected items on undeformed shape.
1
Display selected items on deformed shape.
Notes
Displays selected items (e.g., shears and moments) as a contoured area (trapezoid) display along line elements
and 2-D axisymmetric shell elements (e.g., shear and moment diagrams). Three sides of the trapezoid are
formed by the element (one side) and lines at nodes I and J of length proportional to the item magnitude
and displayed normal to the element and the viewing direction (the two parallel sides).
Portions of this command are not supported by PowerGraphics [/GRAPHICS,POWER].
Menu Paths
Main Menu>General Postproc>Plot Results>Contour Plot>Line Elem Res
PLNEAR, Lab, Opt, KCN, VAL1, VAL2, VAL3, VAL4, VAL5, VAL6, VAL7
Plots the electric field in the near zone exterior to the equivalent source surface.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Lab
Print the maximum E-file:
SPHERE
on the spherical structure
PATH
along the path
Opt
SUM the maximum total E-field (default)
X the maximum E-field in x-direction
Y the maximum E-field in y-direction
Z the maximum E-field in z-direction
PHI the maximum E-field in -direction
THETA the maximum E-field in -direction
R the maximum E-field in r-direction
LHCP the maximum left-hand circularly polarized component
RHCP the maximum right-hand circularly polarized component
ARCP the circular polarization axial ratio
LW3X the maximum dominant component for an x-polarized aperture with Ludwigs third definition
of cross definition
LW3Y the maximum dominant component for an y-polarized aperture with Ludwigs third definition
of cross definition
1273
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLNEAR
ARCP the linear polarization axial ratio
KCN
KCN is the coordinate system reference number. It may be 0 (Cartesian) or any previously defined local
coordinate system number (>10). Defaults to 0.
VAL1, VAL2, VAL3, . . . , VAL7
For LAB = SPHERE:
VAL1
the radius of spherical surface in spherical coordinate system.
VAL2
starting angle (degree) in the spherical coordinate system. Defaults to 0.
VAL3
ending angle (degree) in the spherical coordinate system. Defaults to 0.
VAL4
Number of divisions between the starting and ending angles for data computations. Defaults to
0.
VAL5
starting angle (degree) in the spherical coordinate system. Defaults to 0.
VAL6
ending angle (degree) in the spherical coordinate system. Defaults to 0.
VAL7
Number of divisions between the starting and ending angles for data computations. Defaults to
0.
For Lab = PATH, PLNEAR computes the electric field for the path data points for the path currently
defined by the PATH and PPATH commands.
Notes
PLNEAR uses the equivalent source principle to calculate the electric field in the near zone exterior to the
equivalent source surface (flagged with the Maxwell surface flag in the preprocessor) for one of the following
locations:
A spherical surface in the KAN coordinate system
A path defined by the PATH and PPATH commands
To plot the electromagnetic field results for a path, use the PLPAGM or PLPATH commands.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1274
PLNEAR
PLNSOL, Item, Comp, KUND, Fact, FileID
Displays results as continuous contours.
POST1: Results
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Label identifying the item. Valid item labels are shown in Table 246: PLNSOL - Valid Item and Component
Labels (p. 1275) below. Some items also require a component label.
Comp
Component of the item (if required). Valid component labels are shown in Table 246: PLNSOL - Valid Item
and Component Labels (p. 1275) below.
KUND
Undisplaced shape key:
0
Do not overlay undeformed structure display
1
Overlay displaced contour plot with undeformed display (appearance is system-dependent)
2
Overlay displaced contour plot with undeformed edge display (appearance is system-dependent)
Fact
Scale factor for 2-D display for contact items. Default value is 1. A negative scaling factor may be used
to invert the display.
FileID
The file index number (obtained via the NLDIAG,NRRE,ON command). Valid only for Item = NRRE.
Notes
Displays the solution results as continuous contours across element boundaries for the selected nodes and
elements. For example, PLNSOL,S,X displays the X component of stress S (that is, the SX stress component).
Various element results depend upon the recalculation method and the selected results location [AVPRIN,
RSYS, LAYER, SHELL, and NSEL]. Contours are determined by linear interpolation within each element from
the nodal values, which are averaged at a node whenever two or more elements connect to the same node
(except for FMAG, which is summed at the node).
For PowerGraphics displays [/GRAPHICS,POWER], results are plotted only for the model exterior surface. The
items marked with [2 (p. 1279)] are not supported by PowerGraphics. To plot midside nodes, you must first
issue /EFACET,2.
Table 246 PLNSOL - Valid Item and Component Labels
Description Comp Item
Valid item and component labels for nodal degree of freedom results are:
X, Y, or Z structural displacement or vector sum X, Y, Z, SUM U
X, Y, or Z structural rotation or vector sum X, Y, Z, SUM ROT
Temperature TEMP[1]
Pressure PRES
1275
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLNSOL
Description Comp Item
Electric potential VOLT
Magnetic scalar potential MAG
X, Y, Z, SUM V X, Y, or Z fluid velocity or vector sum in a fluid analysis, or
X, Y, or Z velocity or vector sum in an ANSYS LS-DYNA
analysis.
X, Y, Z, SUM A X, Y, or Z magnetic vector potential or vector sum in an
electromagnetic analysis, or
X, Y, or Z acceleration or vector sum in an ANSYS LS-DYNA
analysis.
X, Y, or Z velocity or vector sum in a structural transient
dynamic analysis (ANTYPE,TRANS).
X, Y, Z, SUM VEL
X, Y, or Z acceleration or vector sum in a structural transient
dynamic analysis (ANTYPE,TRANS).
X, Y, Z, SUM ACC
X, Y, or Z rotational velocity or vector sum in a structural
transient dynamic analysis (ANTYPE,TRANS).
X, Y, Z, SUM OMG
X, Y, or Z rotational acceleration or vector sum in a structur-
al transient dynamic analysis (ANTYPE,TRANS).
X, Y, Z, SUM DMG
Turbulent kinetic energy (FLOTRAN) ENKE
Turbulent energy dissipation (FLOTRAN) ENDS
Mass fraction of species n, where n = 1 to 6 (FLOTRAN). If
a species is given a user-defined name [MSSPEC], use that
name instead of n.
SP0n
Warping WARP
Plot the Newton-Raphson residuals from the file you ob-
tained via the NLDIAG,NRRE,ON command. The FNRM and
FX, FY, FZ, FNRM, MX,
MY, MZ, MNRM
NRRE
MNRM labels are computed as the square root of the sum
of the squares of the residual component forces or mo-
ments (FX,FY,FZ, MX, MY, MZ). [6]
When KUND = 0, use the absolute value of the residual
from the files (default).
Valid item and component labels for element results are:
Component stress X, Y, Z, XY, YZ, XZ S
Principal stress 1, 2, 3 "
Stress intensity INT "
Equivalent stress EQV "
Component elastic strain X, Y, Z, XY, YZ, XZ EPEL
Principal elastic strain 1, 2, 3 "
Elastic strain intensity INT "
Elastic equivalent strain EQV "
Component thermal strain X, Y, Z, XY, YZ, XZ EPTH
Principal thermal strain 1, 2, 3 "
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1276
PLNSOL
Description Comp Item
Thermal strain intensity INT "
Thermal equivalent strain EQV "
Component plastic strain X, Y, Z, XY, YZ, XZ EPPL
Principal plastic strain 1, 2, 3 "
Plastic strain intensity INT "
Plastic equivalent strain EQV "
Component creep strain X, Y, Z, XY, YZ, XZ EPCR
Principal creep strain 1, 2, 3 "
Creep strain intensity INT "
Creep equivalent strain EQV "
Swelling strain EPSW
Component total mechanical strain (EPEL + EPPL + EPCR) X, Y, Z, XY, YZ, XZ EPTO
Principal total mechanical strain 1, 2, 3 "
Total mechanical strain intensity INT "
Total mechanical equivalent strain EQV "
Component total mechanical and thermal strain (EPEL +
EPPL + EPCR + EPTH)
X, Y, Z, XY, YZ, XZ EPTT
Principal total mechanical and thermal strain 1, 2, 3 "
Total mechanical and thermal strain intensity INT "
Total mechanical and thermal equivalent strain EQV "
Components of Biots effective stress. X,Y,Z,XY,YZ,ZX ESIG
Principal stresses of Biots effective stress. 1, 2, 3 "
Stress intensity of Biots effective stress. INT "
Equivalent stress of Biots effective stress. EQV "
Total porosity (Gurson material model). TPOR DPAR
Porosity due to void growth. GPOR "
Porosity due to void nucleation. NPOR "
Equivalent stress (from stress-strain curve) SEPL NL
Stress state ratio SRAT "
Hydrostatic pressure HPRES "
Accumulated equivalent plastic strain EPEQ "
Accumulated equivalent creep strain CREQ "
Plastic state variable PSV "
Plastic work/volume PLWK "
Elastic strain energy density ELASTIC SEND
Plastic strain energy density PLASTIC "
Creep strain energy density CREEP "
Damage variable DMG CDM
1277
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLNSOL
Description Comp Item
Maximum previous strain energy for virgin material LM "
Maximum of all failure criterion defined for this node [2][7] MAX FAIL
Maximum Strain Failure Criterion [2][7] EMAX "
Maximum Stress Failure Criterion [2][7] SMAX "
Tsai-Wu Strength Index Failure Criterion [2][7] TWSI "
Inverse of Tsai-Wu Strength Ratio Index Failure Criterion
[2][7]
TWSR "
User-defined failure criteria [2][7][8] USR1, USR2, ..., USR9 "
State variable [2] 1, 2, 3, ... N SVAR
Gasket component stress X, XY, XZ GKS
Gasket component total closure X, XY, XZ GKD
Gasket component total inelastic closure X, XY, XZ GKDI
Gasket component thermal closure X, XY, XZ GKTH
Interface traction (stress) X, XY, XZ SS
Interface separation X, XY, XZ SD
Fictive temperature TEMP FICT
For contact results PowerGraphics is applicable for 3-D models only.
Contact status[5]: STAT[4] CONT[3]
3-closed and sticking
2-closed and sliding
1-open but near contact
0-open and not near contact
Contact penetration PENE "
Contact pressure PRES "
Contact friction stress SFRIC "
Contact total stress (pressure plus friction) STOT "
Contact sliding distance SLIDE "
Contact gap distance GAP "
Total heat flux at contact surface FLUX "
Total number of contact status changes during substep. CNOS "
Fluid penetration pressure FPRS "
Component thermal gradient or vector sum X, Y, Z, SUM TG
Component thermal flux or vector sum X, Y, Z, SUM TF
Component pressure gradient or vector sum X, Y, Z, SUM PG
Component electric field or vector sum X, Y, Z, SUM EF
Component electric flux density or vector sum X, Y, Z, SUM D
Component magnetic field intensity or vector sum X, Y, Z, SUM H
Component magnetic flux density or vector sum X, Y, Z, SUM B
Component electromagnetic force or vector sum [2]. X, Y, Z, SUM FMAG
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1278
PLNSOL
Description Comp Item
Conduction current density for elements that support con-
duction current calculation. Components (X, Y, Z) and vector
sum (SUM). [2].
X, Y, Z, SUM JC
Body temperatures (calculated from applied temperatures)
as used in solution (area and volume elements only).
TEMP BFE
Densities used for topological optimization. This applies to
nodes attached to the following types of elements:
PLANE82, SOLID92, SOLID95.
TOPO
Valid item labels for FLOTRAN nodal results are:
Total temperature TTOT
Heat flux HFLU
Heat transfer (film) coefficient HFLM
Fluid laminar conductivity COND
Pressure coefficient PCOE
Total (stagnation) pressure PTOT
Mach number MACH
Stream function (2-D applications only) STRM
Fluid density DENS
Fluid laminar viscosity VISC
Specific heat [2] SPHT
Fluid effective viscosity EVIS
Turbulent viscosity coefficient CMUV
Fluid effective conductivity ECON
Y+, a turbulent law of the wall parameter YPLU
Shear stress at the wall TAUW
Surface tension coefficient SFTS
Laminar mass diffusion coefficient for species n, where n
= 1 to 6
LMDn
Effective mass diffusion coefficient for species n, where n
= 1 to 6
EMDn
Radiation heat flux [2] RDFL
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels TBOT, TE2, TE3, . . ., TTOP
instead of TEMP to view the individual temperature degree of freedom. When other thermal elements
are included in the model, they should be unselected to avoid plotting undefined information. To view
all temperatures in the same plot, set /ESHAPE,1 and /GRAPHICS,POWER and issue PLNSOL,TEMP.
2. Not supported by PowerGraphics
3. For the CONT items for elements CONTA171 through CONTA177, the reported data is averaged across
the element. To obtain a more meaningful STAT value, use the PLESOL command.
4. Status notation is for all contact elements except CONTAC12 and CONTAC52. If you are using these
elements, check the output listing in the corresponding element documentation.
1279
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLNSOL
5. For MPC-based contact definitions, the value of STAT can be negative. This indicates that one or more
contact constraints were intentionally removed to prevent over-constraint. STAT = -3 is used for MPC
bonded contact; STAT = -2 is used for MPC no-separation contact.
6. When plotting Newton-Raphson residual items (Item = NRRE) from a file on the deformed geometry,
the displacements are based on the current set of results in the database. These displacements may
not correspond to the loadstep and substep in the .nrxxxxx file. For more information about
.nrxxxxx files and nonlinear diagnostics postprocessing, see the description of the NLDPOST com-
mand and Performing Nonlinear Diagnostics.
7. Works only if FC command information is provided.
8. Works only if user failure criteria routine is provided.
Menu Paths
Main Menu>Drop Test>Animate Results
Main Menu>General Postproc>Plot Results>Contour Plot>Nodal Solu
Utility Menu>Plot>Results>Contour Plot>Nodal Solution
Utility Menu>PlotCtrls>Animate>Animate Over Results
Utility Menu>PlotCtrls>Animate>Animate Over Time
Utility Menu>PlotCtrls>Animate>Deformed Results
Utility Menu>PlotCtrls>Animate>Dynamic Results
Utility Menu>PlotCtrls>Animate>Isosurfaces
Utility Menu>PlotCtrls>Animate>Mode Shape
Utility Menu>PlotCtrls>Animate>Q-Slice Contours
Utility Menu>PlotCtrls>Animate>Time-harmonic
/PLOPTS, Label, KEY
Controls graphics options on subsequent displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Apply display items as selected from the following labels:
INFO
Controls the display of the legend (ON or OFF) and allows the choice of preset or Multi-legend
placement. Control is provided by the KEY values. (Defaults to KEY=3 when the GUI is on. Defaults
to KEY= 2 otherwise.)
LEG1
Header portion of legend column (defaults to ON).
LEG2
View portion of legend column (defaults to ON (except off with contour displays)).
LEG3
View the contour section of the legend column (defaults to ON).
FRAME
Frame border lines around windows (defaults to ON).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1280
/PLOPTS
TITLE
Title (bottom left text) (defaults to ON).
MINM
Min-Max symbols on contour displays (defaults to ON).
LOGO
ANSYS logo (defaults to OFF (displayed as text at top of legend column)). If KEY = ON, the text is
removed from legend column but the logo symbol is displayed in whichever active window is either
in the uppermost right corner and on top, or if there is no window in that location, then in the
window to the furthest right of the screen. Version information remains in the legend column.
WINS
Controls whether graphics windows automatically stretch or shrink to adjust to screen size as the
legend column is turned off or on [/PLOPTS,INFO] (defaults to ON). If WINS is on and the legend
column is changed from off to on, all windows are shrunk regardless of what their correct size is.
WP
Working plane (defaults to OFF). The working plane is drawn as part of the display (not just an
overlaid image as in WPSTYL). This option is best used in combination with a hidden-line technique
[/TYPE].
DATE
Controls the display of the date and time in your legend. Subsequent KEY values control the display
as follows:
Off or 0 - No date or time displays are included in your legend.
1 - Only the date is shown.
2 (default) - both the date and time are shown.
FILE
Controls the display of the ANSYS jobname in your legend. Subsequent KEY values control the display
as follows:
Off or 0 (default) - The ANSYS jobname is NOT included in your legend.
On or 1 - The ANSYS jobname is included in your legend.
KEY
Switch:
OFF or 0
Do not apply this display item. For Label = DATE, no time or date are displayed.
ON or 1
Apply this display item. For Label = DATE, show only the date.
AUTO or 2
For Label = INFO, initiate Auto-legend mode. If the display has contours, the legend is ON; if the
display has no contours, the legend is OFF. For Label = DATE, display both the date and time.
3
For Label = INFO , switch to Multi-legend mode. See the /UDOC command for the available legend
configurations.
Command Default
See individual label defaults.
1281
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PLOPTS
The Multi-legend mode (/PLOPTS,INFO,3) is the default for contour legend displays.
Notes
Use /PLOPTS,STAT to display settings. Use /PLOPTS,DEFA to reset all specifications back to their defaults.
When you perform multiple results displays, contours on the legend column may be truncated. To avoid
this, specify /PLOPTS,LEG1,0.
The Multi-legend mode provides a number of legend data item priority and placement options. These options
are accessed via the GUI at Utility Menu> PlotCtrls> Window Controls> Window Options. The /UDOC
command provides command line options for this capability.
This command is valid in any processor.
This command is not available for Academic Research or Teaching level products
Menu Paths
Utility Menu>PlotCtrls>Window Controls>Reset Window Options
Utility Menu>PlotCtrls>Window Controls>Window Options
PLORB
Displays the orbital motion of a rotating structure
POST1: Results
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
When a structure is rotating and the Coriolis or gyroscopic effect is taken into account (CORIOLIS), nodes
lying on the rotation axis generally exhibit an elliptical orbital motion. The PLORB command displays the
orbit of each rotating node as well as the deformed shape at time t = 0 (the real part of the solution).
To print the characteristics of the orbital path traversed by each node, issue the PRORB command.
The PLORB command is valid for the following line elements: BEAM4, PIPE16, BEAM188, BEAM189, PIPE288,
and PIPE289.
Your model must also involve a rotational velocity (OMEGA or CMOMEGA) with Coriolis enabled in a sta-
tionary reference frame (CORIOLIS,,,,RefFrame = ON).
A SET command should be issued after PLORB to ensure proper output for subsequent post processing
commands.
The coordinate system for displaying nodal results must be global Cartesian (RSYS,KCN = 0).
Menu Paths
Main Menu>General Postproc>Rotor Dynamics>Plot orbit
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1282
PLORB
PLOT, NSTRT, NEND, NINC
Forms a display.
DISPLAY: Action
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSTRT, NEND, NINC
Display plots sequentially from number NSTRT to NEND in steps of NINC. NSTRT defaults to the next
plot. NEND defaults to NSTRT. NINC defaults to 1. If NSTRT = ALL, display all plots from the beginning
of the file. If NEND = ALL, display to the end of the file.
Notes
Output will be to the terminal or to a file, depending on the driver [/SHOWDISP]. The INTERLEAF and DUMP
drivers produce an output file for each plot named INTLnn and DUMPnn, with nn sequentially ranging from
00 to 99. A blank line after the PLOT command causes the next plot to be formed.
Menu Paths
It is part of the DISPLAY command.
PLOTTING
Specifies "Plotting settings" as the subsequent status topic.
POST26: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>TimeHist Postproc>Plot
1283
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLOTTING
PLPAGM, Item, Gscale, Nopt
Displays path items along the path geometry.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
The path data item to be displayed on the currently active path (defined by the PATH command). Valid
path items are those defined with the PDEF or PLNEAR commands.
Gscale
Scale factor for the offset from the path for the path data item displays. Defaults to 1.0.
Nopt
Determines how data is displayed:
(blank)
Do not display nodes, and scale the display based on the currently selected node set (default).
NODE
Display path item data along with the currently selected set of nodes. The display geometry is scaled
to the selected node set.
Notes
You can use the Gscale argument to scale the contour display offset from the path for clarity. You need
to type all six characters to issue this command.
Menu Paths
Main Menu>General Postproc>Path Operations>Plot Path Item>On Geometry
Main Menu>General Postproc>Plot Results>Plot Path Item>On Geometry
PLPATH, Lab1, Lab2, Lab3, Lab4, Lab5, Lab6
Displays path items on a graph.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab1, Lab2, Lab3, . . . , Lab6
Labels identifying the path items to be displayed. Up to six items may be drawn per frame. Predefined
path geometry items XG, YG, ZG, and S [PDEF] may also be displayed.
Notes
The path must have been defined by the PATH and PPATH commands. Path items and their labels must
have been defined with the PDEF, PVECT, PCALC, PDOT, PCROSS, or PLNEAR commands. Path items may
also be printed with the PRPATH command. Graph scaling may be controlled with the /XRANGE, /YRANGE,
and PRANGE commands. You need to type all six characters to issue this command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1284
PLPAGM
Menu Paths
Main Menu>General Postproc>Path Operations>Plot Path Item>On Graph
Main Menu>General Postproc>Plot Results>Plot Path Item>On Graph
Utility Menu>Plot>Results>Path Plot
PLSCH, Fname, Ext, Lab, Port
Converts and plots scattering, admittance, or impedance parameters on a Smith chart.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Fname
File name and directory path of a Touchstone file (248 characters maximum, including the characters
needed for the directory path). A Touchstone file may be created in ANSYS by performing a frequency
sweep using the SPSWP command macro, or may it be supplied from another source. An unspecified
directory path defaults to the working directory; in this case, you can use all 248 characters for the file
name.
Ext
Filename extension (8 character maximum).
Lab
Output parameter:
S
Scattering parameter (default)
Y
Admittance parameter
Z
Impedance parameter
Port
Port number. Defaults to 1.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>Smith Chart
1285
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLSCH
PLSECT, Item, Comp, RHO, KBR
Displays membrane and membrane-plus-bending linearized stresses.
POST1: Path Operations
MP ME ST PR PRN DS DSS <> <> <> DY PP <> EME MFS
Item
Label identifying the item to be processed. Valid item labels are shown in Table 247: PLSECT - Valid Item
and Component Labels (p. 1286) below. Items also require a component label.
Comp
Component of the item. Valid component labels are shown in Table 247: PLSECT - Valid Item and Component
Labels (p. 1286) below.
RHO
In-plane (X-Y) average radius of curvature of the inside and outside surfaces of an axisymmetric section.
If zero (or blank), a plane or 3-D structure is assumed. If nonzero, an axisymmetric structure is assumed.
Use a very large number (or -1) for an axisymmetric straight section.
KBR
Through-thickness bending stresses key for an axisymmetric analysis (RHO 0):
0
Include the thickness-direction bending stresses.
1
Ignore the thickness-direction bending stresses.
2
Include the thickness-direction bending stress using the same formula as the Y (axial direction )
bending stress. Also use the same formula for the shear stress.
Notes
Calculates and displays the membrane and membrane-plus-bending linearized stresses (as described for the
PRSECT command) along a path section [PATH] as a graph. The path section is defined by two points specified
with the PPATH command. For linearized stress calculations, the path must be defined with nodes. The path
must be entirely within the selected elements (that is, there must not be any element gaps along the path).
The total stress (equivalent to the PLPATH display) is also displayed. This command always uses 48 divisions
along the path, regardless of the number of divisions defined by PATH.
Portions of this command are not supported by PowerGraphics [/GRAPHICS,POWER].
Table 247 PLSECT - Valid Item and Component Labels
Description Comp Item
Valid item and component labels for element results are:
Component stress. X, Y, Z, XY, YZ, XZ S
Principal stress. 1, 2, 3 "
Stress intensity or equivalent stress. INT, EQV "
Menu Paths
Main Menu>General Postproc>Path Operations>Linearized Strs
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1286
PLSECT
Main Menu>General Postproc>Plot Results>Plot Path Item>Lineariz Strs
PLSYZ, Fname, Ext, Lab, Opt, VAL_I1, VAL_J1, VAL_I2, VAL_J2, VAL_I3, VAL_J3, VAL_I4, VAL_J4
Converts and plots network parameters versus frequency or plots losses versus frequency.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Fname
File name and directory path of a Touchstone file (248 characters maximum, including the characters
needed for the directory path). A Touchstone file may be created in ANSYS by performing a frequency
sweep using the SPSWP command macro, or may it be supplied from another source. An unspecified
directory path defaults to the working directory; in this case, you can use all 248 characters for the file
name. Fname is only used for network parameters (Lab = S, Y, or Z).
Ext
Extension of Touchstone file (.snp where n is the number of ports). Ext is only used for network
parameters (Lab = S, Y, or Z).
Lab
S
Scattering parameter (default)
Y
Admittance parameter
Z
Impedance parameter
REFL
Reflection loss
INSL
Insertion loss
ISOL
Isolation loss
VSWR
Voltage standing wave ratio
Opt
For Lab = S, Y, or Z, format of Touchstone file:
MAG
Magnitude (default)
ANG
Angle
DB
Magnitude in dB units. Only available for scattering parameters (Lab = S).
REAL
Real part of the parameter
1287
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLSYZ
IMAG
Imaginary part of the parameter
For Lab = REFL, INSL, or ISOL, DB (magnitude in dB units) is the only available option. For Lab = VSWR,
MAG (mangitude) is the only available option.
VAL_I1
For Lab = S, Y, or Z, index I of first parameter plotted. For example, SIJ. Defaults to 1.
For Lab = REFL, INSL, ISOL, or VSWR, output port number.
VAL_J1
For Lab = S, Y, or Z, index J of first parameter plotted. For example, SIJ. Defaults to 1.
For Lab = REFL, INSL, ISOL, or VSWR, input port number.
VAL_I2, VAL_J2, VAL_I3, VAL_J3, VAL_I4, VAL_J4
For Lab = S, Y, or Z, index I and J of second, third, and fourth parameters plotted. For example, SIJ. No
default.
For Lab = REFL, INSL, ISOL, or VSWR, VAL_I2 through VAL_J4 are not used.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>S,Y,Z Parameters
PLTD, Fname, Ext, Lab, Opt, Vkey, NFFT, TSTART, TRISE, PORTI1, PORTJ1, PORTI2, PORTJ2, PORTI3, PORTJ3,
PORTI4, PORTJ4
Displays TDR/TDT waveforms, an impedance profile, or a total waveform.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Fname
A Touchstone file containing the S-parameters of a n-port system.
Ext
Extension of the n-port Touchstone file.
Lab
Display type:
SPAR
Display TDR and TDT waveforms.
IMPD
Display impedance profile.
TDWF
Display total waveform.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1288
PLTD
Opt
Excitation source:
STEP
Step-like source (default).
IMPU
Gaussian impulse.
Vkey
Control key:
Off
No display of the source waveform (default).
On
Display the source waveform.
NFFT
Number of interpolation points for Fast Fourier Transformation. It must be an integer power of 2. NFTT
must be greater than or equal to 256 and less than or equal to 4096. Defaults to 512.
TSTART
Source beginning time in seconds. Defaults to 0.
TRISE
Source rise time. For a step-like source, it is the time from 10% of the rise to 90% of the rise. For a
Gaussian impulse, it is the time from 50% of the rise to 50% of the rise. The rise time must be greater
than or equal to 5 time steps. If the TRISE input is less than 5 time steps, it is changed to 5 time steps.
Defaults to 5 time steps.
PORTI1
Driven port number. Defaults to 1.
PORTJ1
Output port number. Defaults to 1.
PORTI2
Driven port number. No default.
PORTJ2
Output port number. No default.
PORTI3
Driven port number. No default.
PORTJ3
Output port number. No default.
PORTI4
Driven port number. No default.
PORTJ4
Output port number. No default.
Notes
The time step is given by dt = 1/(2fmax) where fmax is the maximum frequency.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1289
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLTD
Menu Paths
This command cannot be accessed from a menu.
PLTIME, TMIN, TMAX
Defines the time range for which data are to be displayed.
POST26: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
TMIN
Minimum time (defaults to the first point stored).
TMAX
Maximum time (defaults to the last point stored).
Command Default
Use the previously defined range [TIMERANGE].
Notes
Defines the time (or frequency) range (within the range stored) for which data are to be displayed. Time is
always displayed in the Z-axis direction for 3-D graph displays. If XVAR = 1, time is also displayed in the X-
axis direction and this control also sets the abscissa scale range.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Graph
PLTLINE, Filename, Ext, Lab, Modenum
Plots port transmission line data generated by the HFPCSWP or SPSWP macros.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Filename
A file name containing the transmission line results.
Ext
Filename extension (8 character maximum).
Lab
Data type:
BETA
Propagating constant.
ZC
Characteristic impedance.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1290
PLTIME
EREFF
Effective dielectric constant.
Modenam
ALL
Plot data for all modes (default).
N
Plot data for the Nth mode.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
PLTRAC, Analopt, Item, Comp, TRPNum, Name, MXLOOP, TOLER, OPTION, ESCL, MSCL
Displays a particle flow or charged particle trace on an element display.
POST1: Trace Points
MP <> <> <> <> <> <> FL EM <> <> PP <> <> <>
Analopt
Analysis option
FLUID
Particle trace in fluid flow (default)
ELEC
Particle trace in electric field
MAGN
Particle trace in magnetic field
EMAG
Particle trace in presence of both electric and magnetic fields
Item
Label identifying the item to be contoured. Valid item labels are shown in Table 248: PLTRAC - Valid Item
and Component Labels (p. 1293) below. Some items also require a component label. If Item is blank, display
only the path trajectory.
Comp
Component of the item (if required). Valid component labels are shown in Table 248: PLTRAC - Valid Item
and Component Labels (p. 1293) below.
TRPNum
Trace point number for storing trajectory data for use with PATH logic. Defaults to 0 (no trajectory path
data is stored for further processing with PATH logic).
1291
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLTRAC
Name
Name of prefix of array variable. Defaults to TRAC. NamePOIN stores trajectory path points for trace point
number TRPNum. If Analopt = ELEC, MAGN, or EMAG, two additional array parameters, NameDATA
and NameLABL, store trajectory path data and labels for the same TRPNum.
MXLOOP
Maximum number of loops traced by a particle. Defaults to 25 for Opt = FLUID; otherwise, defaults to
1000.
TOLER
Length tolerance used for particle trajectory geometry calculation. Valid only for Analopt = ELEC,
MAGN, or EMAG. If particle trace appears to terminate inside an element, adjusting the length tolerance
may be necessary. Defaults to 1.0 x 10
-8
.
OPTION
Flow trace option:
0
Use the undeformed mesh for computing the flow trace.
1
Use the deformed mesh for computing the flow trace.
ESCL
Electric field scale factor. Setting this scale factor affects only the tracing, not the field solution results.
A negative factor corresponds to the opposite vector direction. Valid only for Analopt = ELEC or EMAG.
Defaults to 1.
MSCL
Magnetic field scale factor. Setting this scale factor affects only the tracing, not the field solution results.
A negative factor corresponds to the opposite vector direction. Valid only for Analopt = MAGN or
EMAG. Defaults to 1.
Notes
For a specified item, the variation of the item is displayed along the particle trace as a color-contoured ribbon.
The TRPOIN command must be used to define a point on the trajectory path. Multiple traces may be displayed
simultaneously by defining multiple trace points. Issue the TRPLIS command to list the current tracing
points. Issue the TRPDEL command to delete tracing points defined earlier. Use the PAPUT command with
the POIN option to retrieve the particle trajectory points as path points.
The model must be 3-D for the ELEC, MAGN, and EMAG analysis options.
Three array parameters are created at the time of the particle trace: TRACPOIN, TRACDATA and TRACLABL.
These array parameters can be used to put the particle velocity and the elapsed time into path form. The
procedure to put the arrays into a path named PATHNAME is as follows:
*get,npts,PARM,TRACPOIN,DIM,x
PATH,PATHNAME,npts,9,1
PAPUT,TRACPOIN,POINTS
PAPUT,TRACDATA,TABLES
PAPUT,TRACLABL,LABELS
PRPATH,S,T_TRACE,VX_TRACE,VY_TRACE,VZ_TRACE,VS_TRACE
Not used if Analopt = FLUID. If working in the GUI, use the "All information" option to retrieve information
from all three arrays at once.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1292
PLTRAC
If OPTION is set to 1, the deformed mesh is based on the displacement degrees of freedom UX, UY, and
UZ, which must be available in the load step.
Table 248 PLTRAC - Valid Item and Component Labels
Description Comp Item
Valid item and component labels for nodal degree of freedom results are:
Temperature. TEMP
Pressure. PRES
X, Y, or Z fluid velocity or vector sum. X, Y, Z, SUM V
Turbulent kinetic energy. ENKE
Turbulent energy dissipation. ENDS
Valid item and component labels for FLOTRAN nodal results are:
Total temperature. TTOT
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Specific heat. SPHT
Fluid effective viscosity. EVIS
Turbulent viscosity coefficient. CMUV
Fluid effective conductivity. ECON
Valid item labels for Analopt = ELEC nodal results are:
Electric potential. VOLT
Valid item labels for Analopt = MAGN or EMAG nodal results are:
Color contour displayed. None
See the Basic Analysis Guide for more information on particle flow and charged particle traces. See Animation
in the Basic Analysis Guide for information on particle trace animation.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>Particle Trace
Main Menu>General Postproc>Plot Results>Plot Flow Tra
Utility Menu>Plot>Results>Flow Trace
Utility Menu>PlotCtrls>Animate>Particle Flow
1293
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLTRAC
PLVAR, NVAR1, NVAR2, NVAR3, NVAR4, NVAR5, NVAR6, NVAR7, NVAR8, NVAR9, NVAR10
Displays up to ten variables in the form of a graph.
POST26: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVAR1, NVAR2, NVAR3, . . . , NVAR10
Variables to be displayed, defined either by the reference number or a unique thirty-two character name.
If duplicate names are used the command will plot the data for the lowest-numbered variable with that
name.
Notes
Variables are displayed vs. variable N on the XVAR command. The string value will be a predefined, unique
name. For complex variables, the amplitude is displayed by default [PLCPLX]. Each PLVAR command produces
a new frame. See the /GRTYP command for displaying multiple variables in a single frame with separate Y-
axes.
Menu Paths
Main Menu>TimeHist Postpro>Graph Variables
PLVAROPT, Lab1, Lab2, Lab3, Lab4, Lab5, Lab6, Lab7, Lab8, Lab9, Lab10
Displays up to ten parameters in the form of a graph.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab1, Lab2, Lab3, . . . , Lab10
Names of the parameters to be displayed on this frame.
Command Default
No display.
Notes
Displays up to ten parameters in the form of a graph. Parameters are displayed vs. Lab on the XVAROPT
command (defaults to set number) in the order corresponding to an ascending order of the XVAROPT
parameter. See the /GRTYP command for displaying multiple parameters in a single frame with separate Y-
axes. Each PLVAROPT command produces a new frame.
Menu Paths
Main Menu>Design Opt>Design Sets>Graphs/Tables
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1294
PLVAR
PLVECT, Item, Lab2, Lab3, LabP, Mode, Loc, Edge, KUND
Displays results as vectors.
POST1: Results
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Item
Predefined vector item (from Table 249: PLVECT - Valid Item Labels (p. 1296) below) or a label identifying
the i-component of a user-defined vector.
Lab2
Label identifying the j-component of a user-defined vector. In most cases, this value must be blank if
Item is selected from Table 249: PLVECT - Valid Item Labels (p. 1296). Individual principal stresses (Item =
S) or principal strains (Item = EPxx) may be plotted by specifying the value as 1, 2, or 3.
Lab3
Label identifying the k-component of a user-defined vector. Must be blank if Item is selected from list
below or for 2-D user defined vector.
LabP
Label assigned to resultant vector for display labeling (defaults to Item).
Mode
Vector or raster mode override key:
(blank)
Use the setting of KEY on the /DEVICE command.
RAST
Use raster mode for PLVECT displays.
VECT
Use vector mode for PLVECT displays.
Loc
Vector location for display of field element results:
ELEM
Display at element centroid (default).
NODE
Display at element nodes.
Nodal results quantities will only be displayed at nodes, not at element centroids.
Edge
Edge display override key:
(blank)
Use the setting of Key on the /EDGE command.
OFF
Deactivate the edge display.
ON
Activate the edge display.
KUND
Undisplaced shape key:
1295
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLVECT
0
Display vectors on undeformed mesh or geometry.
1
Display vectors on deformed mesh or geometry.
Notes
Displays various solution results as vectors (arrows) for the selected nodes and/or elements (elements must
contain at least three nodes that are not colinear). For example, PLVECT,U displays the displacement vector
for all selected nodes. For section displays [/TYPE], the vectors are shown only on the section face (i.e., cutting
plane). The PLVECT display of principal strains and stresses (Item = S, EPTO, EPEL, EPPL, EPCR, or EPTH) on
a "cut" of the model (/TYPE,,1 ,5,7,8, or 9) is not supported. The resulting plot displays the vectors on all
selected elements, not on just the sliced surface. See the /VSCALE command to scale vector lengths. Vector
magnitudes may be shown as a contour display with the PLNSOL command. Various results also depend
upon the recalculation method and the selected results location [LAYER, SHELL, and NSEL].
Items may be selected from a set of recognized vector labels (Item) or a vector may be defined from up
to three scalar labels (Item,Lab2,Lab3). Scalar labels may be user-defined with the ETABLE command. The
vectors appear on an element display as arrows showing the relative magnitude of the vector and its direction.
The predefined items will be shown either at the node or at the element centroid, depending on what item
is being displayed and depending on the Loc setting. User defined ETABLE items will be shown at the
element centroid, regardless of the Loc setting. Stress vectors appear as arrows at the element centroid,
with the arrowheads pointing away from each other for tension and toward each other for compression.
For PowerGraphics, vector arrow displays are generated in Global Cartesian (RSYS = 0). All subsequent displays
will revert to your original coordinate system.
When vector mode is active (Mode = VECT), use the Z-buffered display type [/TYPE,,6] to maximize speed
of PLVECT plots (other hidden display types may make plotting slow). For PowerGraphics [/GRAPHICS,POWER],
the items marked with [1 (p. 1297)] are not supported by PowerGraphics.
It is possible to plot principal stresses (Item = S) or principal strains (Item = EPxx) individually. To do so,
specify a Lab2 value of 1, 2, or 3. For example, the following are valid commands:
PLVECT,S,1,,,VECT,ELEM,ON,0
PLVECT,EPEL,3,,,VECT,NODE,ON,0
Table 249 PLVECT - Valid Item Labels
Description Item
Valid item labels for nodal degree of freedom vector results are:
Structural displacement vector. U
Structural rotation vector. ROT
Velocity vector. V
Magnetic vector potential vector. A
Fluid flux in poromechanics. FFLX
Valid item labels for structural element results are:
Principal stresses[1]. S
Principal total strain (EPEL + EPPL + EPCR)[1]. EPTO
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1296
PLVECT
Description Item
Principal elastic strains[1]. EPEL
Principal plastic strains[1]. EPPL
Principal creep strains[1]. EPCR
Principal thermal strains[1]. EPTH
Valid item labels for field element results are:
Thermal gradient vector. TG
Thermal flux vector. TF
Pressure gradient vector. PG
Electric field vector. EF
Electric flux density vector. D
Magnetic field intensity vector. If Lab2 is blank, then Item
is interpreted as one of the predefined labels. Otherwise,
H
Item is interpreted as a user-defined ET label and ANSYS
will request a nonblank Lab2/Lab3 according to the di-
mension of the problem.
Magnetic flux density vector. B
Electromagnetic force vector. FMAG
Poynting vector. P
Source current density vector for low-frequency magnetic
analyses. Total current density vector (sum of conduction
JS
and displacement current densities) in low frequency
electric analyses.
Total measureable current density vector in low-frequency
electromagnetic analyses. (Conduction current density
vector in a low-frequency electric analysis.)
JT
Conduction current density vector for elements that sup-
port conduction current calculation.
JC
1. Not supported by PowerGraphics
Menu Paths
Main Menu>General Postproc>Plot Results>Vector Plot>Predefined
Main Menu>General Postproc>Plot Results>Vector Plot>User-defined
Utility Menu>Plot>Results>Vector Plot
Utility Menu>PlotCtrls>Animate>Q-Slice Vectors
1297
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PLVECT
PLVFRC, CONT
Displays volume fractions in a volume of fluid (VOF) analysis.
PREP7: FLOTRAN Miscellaneous
MP <> <> <> <> <> <> FL <> <> <> PP <> <> <>
CONT
Contour setting:
0
Smooth plot with contour levels at 0.5 and 1.0 (default).
1
Smooth plot with current settings of contour levels.
Notes
The PLVFRC command macro places the volume fraction results into an element table and then plots the
element table data (i.e., ETABLE,VFRC,NMISC,1 and then PLETAB,VFRC,1). If CONT = 0 (default), the command
macro issues the /CVAL,ALL,0.5,1.0 command before the ETABLE command. This means PLETAB uses contour
levels of 0.5 and 1.0 to indicate the free surface and fluid regions. If CONT = 1, PLETAB uses the current
contour settings.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Drop Test>Animate Results
Main Menu>General Postproc>Plot Results>Contour Plot>Volume Fraction
Utility Menu>PlotCtrls>Animate>Animate Over Results
Utility Menu>PlotCtrls>Animate>Animate Over Time
Utility Menu>PlotCtrls>Animate>Time-harmonic
Utility Menu>PlotCtrls>Multi-Plot Contrls
PLWAVE, Ex, Ey, Ez, AngX, AngZ
Specifies a free-space time-harmonic incident plane electromagnetic wave.
SOLUTION: Misc Loads
MP <> <> <> <> <> <> <> <> <> <> PP <> <> <>
Ex
Electric field amplitude in x direction.
Ey
Electric field amplitude in y direction.
Ez
Electric field amplitude in z direction.
AngX
Angle between incident wave vector and X-axis (Phi).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1298
PLVFRC
AngZ
Angle between incident wave vector and Z-axis (Theta).
Notes
Defines an incident plane wave for the entire solution domain. See Spherical Coordinates in the High-Frequency
Electromagnetic Analysis Guide.
This command cannot be used for the analysis of periodic structures (see the HFPORT command).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>Planewav0>Define
Wave
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>Planewav0>Wave Status
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>Planewav0>Define Wave
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>Planewav0>Wave Status
/PMACRO
Specifies that macro contents be written to the session log file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This command forces the contents of a macro or other input file to be written to Jobname.LOG. It is valid
only within a macro or input file, and should be placed at the top of the file. /PMACRO should be included
in any macro or input file that calls GUI functions.
Menu Paths
This command cannot be accessed from a menu.
PMAP, FORM, DISCON
Creates mapping of the path geometry by defining path interpolation division points.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
FORM
Defines the mapping method:
UNIFORM
Maps uniform divisions (specified on the nDiv argument of the PATH command) between specified
points. This is the default.
1299
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PMAP
ACCURATE
Map geometry using a small division at the beginning and end of each segment. This gives you ac-
curate derivatives, integrals, tangents, and normals for curves which do not have continuous slopes
at the specified points. To create nonuniform divisions, the nDiv argument of the PATH command
must be greater than 2.
DISCON
Sets mapping for discontinuities in the field. The divisions are modified to put a point just before and
just after the discontinuity. The valid label is MAT, for a material discontinuity. No discontinuity is the
default. Discontinuity mapping involves the NOAV option on the PDEF command.
Menu Paths
Main Menu>General Postproc>Path Operations>Define Path>Path Options
Main Menu>Preprocessor>Path Operations>Define Path>Path Options
PMETH
Specifies "p-Method" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
This command is also valid in SOLUTION.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Utility Menu>List>Status>p-Method
/PMETH, Key, OPTION
Activates the p-method solution options in the Graphical User Interface (GUI).
SOLUTION: p-Method
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Key
ON
Activates the p-method solution options.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1300
PMETH
OFF
Deactivates the p-method solution options (default).
STAT
Lists the current setting of Key.
OPTION
Activates discipline option if Key = ON.
0
Activates the structural p-method solution options (default).
1
Activates the electrostatic p-method solution options.
Command Default
The p-method solution options are hidden (not shown) in the GUI.
Notes
When this option is activated, only the options applicable to a p-method solution will be displayed in the
GUI. This command is not required outside of the GUI.
Menu Paths
Main Menu>Preferences
PMGTRAN, Fname, FREQ, Fcnam1, Fcnam2, Pcnam1, Pcnam2, Ecnam1, Ccnam1
Summarizes electromagnetic results from a transient analysis.
POST26: Special Purpose
MP <> <> <> <> <> <> <> EM <> <> PP <> <> <>
Fname
File name (8 characters maximum) to which tabular data and plot files will be written. Must be enclosed
in single quotes when the command is manually typed in. Defaults to MG_TRNS. The data file extension
is .OUT and the plot file extension is .PLT.
FREQ
Frequency of solution output. Defaults to 1. Every FREQth solution on the results file is output.
Fcnam1, Fcnam2
Names of element components for force calculation. Must be enclosed in single quotes when the com-
mand is manually typed in.
Pcnam1, Pcnam2
Names of element components for power loss calculation. Must be enclosed in single quotes when the
command is manually typed in.
Ecnam1, Ccnam1
Names of element components for energy and total current calculations, respectively. Must be enclosed
in single quotes when the command is manually typed in.
1301
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PMGTRAN
Notes
PMGTRAN invokes an ANSYS macro which calculates and summarizes electromagnetic results from a tran-
sient analysis. The results are summarized by element components and listed on the screen as well as written
to a file (Fname.OUT). Also, graph plots of results as a function of time are created and written to a file
(Fname.PLT) for use in the DISPLAY program.
Two components may be selected for the summary of electromagnetic forces (see FMAGSUM), two for
power loss, and one each for stored energy (see SENERGY) and total current (see CURR2D). See the referenced
commands for other restrictions.
PMGTRAN is restricted to MKSA units.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>TimeHist Postpro>Elec&Mag>Magnetics
PMLOPT, ESYS, Lab, Xminus, Xplus, Yminus, Yplus, Zminus, Zplus
Defines perfectly matched layers (PMLs) for a high-frequency analysis.
PREP7: Perfectly Matched Layers
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
ESYS
Element coordinate system number. ESYS may be 0 (global Cartesian) or any previously defined local
Cartesian coordinate system number (>10). Defaults to 0.
Lab
Label defining the number of dimensions:
ONE
A one-dimensional PML region.
THREE
A three-dimensional PML region (default).
Xminus
For harmonic analysis, normal reflection coefficient in negative X direction of ESYS. Defaults to 1.E3
(equivalent to -60 dB).
Xplus
For harmonic analysis, normal reflection coefficient in positive X direction of ESYS. Defaults to 1.E3
(equivalent to -60 dB).
Yminus
For harmonic analysis, normal reflection coefficient in negative Y direction of ESYS. Defaults to 1.E3
(equivalent to -60 dB).
Yplus
For harmonic analysis, normal reflection coefficient in positive Y direction of ESYS. Defaults to 1.E3
(equivalent to -60 dB).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1302
PMLOPT
Zminus
For harmonic analysis, normal reflection coefficient in negative Z direction of ESYS. Defaults to 1.E3
(equivalent to -60 dB).
Zplus
For harmonic analysis, normal reflection coefficient in positive Z direction of ESYS. Defaults to 1.E3
(equivalent to -60 dB).
Notes
PMLOPT defines perfectly matched layers (PML) for a high-frequency analysis. Each PML region must have
a uniquely defined element coordinate system. Normal reflection coefficient values for a harmonic analysis
must be less than 1.
Issue PMLOPT,STAT to list the current normal reflection coefficient or attenuation factor settings for a PML
region. Issue PMLOPT,CLEAR to clear all normal reflection coefficient settings and restore them to the defaults.
Issue PMLOPT,ESYS,CLEAR to clear all normal reflection coefficient settings for this element coordinate system
and restore them to the defaults.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>PMLOpt>Clear
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>PMLOpt>Define
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>PMLOpt>Status
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>PMLOpt>All
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>PMLOpt>On ESYS
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>PMLOpt>Clear
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>PMLOpt>Define
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>PMLOpt>Status
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>PMLOpt>All
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>PMLOpt>On ESYS
PMLSIZE, FREQB, FREQE, DMIN, DMAX, THICK, ANGLE
Determines number of PML layers.
PREP7: Perfectly Matched Layers
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FREQB
Minimum operating frequency
FREQE
Maximum operating frequency
DMIN
Minimum distance from radiation source to PML interface.
DMAX
Maximum distance from radiation source to PML interface.
THICK
Thickness of PML region. Defaults to 0.
1303
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PMLSIZE
ANGLE
Incident angle of wave to the PML interface. Defaults to 0.
Notes
PMLSIZE determines the number of PML layers for acceptable numerical accuracy.
PMLSIZE must be issued before any meshing commands. If the thickness of the PML region is known, it
determines an element edge length (h) and issues ESIZE,h. If the thickness of the PML region is unknown,
it determines the number of layers (n) and issues ESIZE,,n.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>PML
PMOPTS, TOLER
Defines percentage tolerance for a p-Method solution.
SOLUTION: p-Method
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
TOLER
Percentage tolerance used to determine which elements to fix at the current p-level. If an element's
strain energy error is less than this tolerance its p-level will be fixed at the current level. Defaults to 5.
Command Default
Tolerance of 5% will be used.
Notes
Defines the p-level tolerance for p-method solutions (TOLER). The default value is satisfactory for most
solutions.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>p-Method>Solution Options
Main Menu>Solution>Load Step Opts>p-Method>Solution Options
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1304
PMOPTS
/PMORE, --, X5, Y5, X6, Y6, X7, Y7, X8, Y8
Creates an annotation polygon (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
--
Unused field.
X5
X location for vertex 5 of polygon (-1.0 < X < 2.0).
Y5
Y location for vertex 5 of polygon (-1.0 < Y < 1.0).
X6
X location for vertex 6 of polygon (-1.0 < X < 2.0).
Y6
Y location for vertex 6 of polygon (-1.0 < Y < 1.0).
X7
X location for vertex 7 of polygon (-1.0 < X < 2.0).
Y7
Y location for vertex 7 of polygon (-1.0 < Y < 1.0).
X8
X location for vertex 8 of polygon (-1.0 < X < 2.0).
Y8
Y location for vertex 8 of polygon (-1.0 < Y < 1.0).
Notes
Defines the 5th through 8th vertices of an annotation polygon [/POLYGON]. This is a command generated
by the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation is used.
This command is not intended to be typed in directly in an ANSYS session (although it can be included in
an input file for batch input or for use with the /INPUT command).
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
PNGR, Kywrd, OPT, VAL
Provides PNG file export for ANSYS displays.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Kywrd
Specifies various PNG file export options.
1305
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PNGR
COMP
If Kywrd = COMP, then OPT is either ON or OFF (blank is interpreted as OFF). This option allows
you to turn PNG file compression ON or OFF. If OPT = ON, then The VAL field is read to determine
the degree of compression. See the VALUE argument for acceptable compression values.
ORIENT
If Kywrd = ORIENT, then OPT will determine the orientation of the entire plot. OPT can be either
Horizontal (default) or Vertical.
COLOR
If Kywrd = COLOR, then OPT will determine the color depth of the saved file. OPT can be 0, 1, or
2, corresponding to Black and White, Grayscale, and Color (default), respectively.
TMOD
If Kywrd = TMOD, then OPT will determine the text method. OPT can be either 1 or 0, corresponding
to bitmap text (default) or line stroke text, respectively.
DEFAULT
If Kywrd = DEFAULT, then all of the default values, for all of the Kywrd parameters listed above, are
active.
STAT
Shows the current status of PNG file export.
OPT
OPT can have the following names or values, depending on the value for Kywrd (see above).
ON, OFF
If Kywrd = COMP, the values On and Off control the use of compression. The degree of compression
is determined by VAL
Horizontal, Vertical
If Kywrd = ORIENT, the terms Horizontal or Vertical determine the orientation of the plot.
0, 1, 2
If Kywrd = COLOR, the numbers 0, 1, and 2 correspond to Black and White, Grayscale and Color,
respectively.
1, 0
If Kywrd = TMOD, the values 1 and 0 determine whether bitmap (1) or stroke text (0) fonts will be
used
VAL
VAL is active only when Kywrd = COMP, and determines the degree of compression applied to the ex-
ported file (see above).
1
Apply the default, optimum value for compression. This value represents the best combination of
speed and compression. It varies according to the release level of the ZLIB compression package.
1-9
Use this value to specify a specific compression level. 1 is the lowest compression level (fastest) and
9 is the highest compression level (slowest).
Menu Paths
Utility Menu>PlotCtrls>Redirect Plots
Utility Menu>PlotCtrls>Hard Copy
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1306
PNGR
/PNUM, Label, KEY
Controls entity numbering/coloring on plots.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Type of numbering/coloring:
NODE
Node numbers on node and element plots.
ELEM
Element numbers and colors on element plots.
DOMAIN
Domains set via DECOMP.
SEC
Section numbers and colors on element and solid model plots (see "Notes" (p. 1308)).
MAT
Material set numbers and colors on element and solid model plots (see "Notes" (p. 1308)).
TYPE
Element type reference numbers and colors on element and solid model plots (see "Notes" (p. 1308)).
REAL
Real constant set numbers and colors on element and solid model plots (see "Notes" (p. 1308)).
ESYS
Element coordinate system numbers on element and solid model plots (see "Notes" (p. 1308)).
PART
Element part numbers and colors on element plots (applicable to ANSYS LS-DYNA only).
LOC
Location numbers/colors of the element in matrix assembly order on element plots.
Note
LOC and ELEM numbers will be the same unless the model has been reordered.
KP
Keypoint numbers on solid model plots.
LINE
Line numbers on solid model plots (both numbers and colors on line plots).
AREA
Area numbers on solid model plots (both numbers and colors on area plots).
VOLU
Volume numbers on solid model plots (both numbers and colors on volume plots).
SVAL
Stress (or any contour) values on postprocessing plots, and surface load values and colors on model
plots (when surface load symbols are on [/PSF]). For tabular boundary conditions, the table-evaluated
1307
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PNUM
values will be displayed on node, element, or contour displays in POST1 when load symbols (/PBF,
/PSF, /PBC) are on and TABNAM is OFF.
TABNAM
Table names for tabular boundary conditions. If this label is turned on, the table name appears next
to the appropriate symbol, arrow, face outline, or contour as dictated by the /PSF, /PBC, and /PBF
commands.
STAT
Shows current settings for /PNUM.
DEFA
Resets all /PNUM specifications back to default.
KEY
Switch:
0
Turns OFF numbers/colors for specified label.
1
Turns ON numbers/colors for specified label.
Notes
This command specifies entity numbering and coloring for subsequent plots.
The MAT, TYPE, REAL, and ESYS labels activate both the numbering and coloring of the corresponding attrib-
utes for EPLOT, KPLOT, LPLOT, APLOT, and VPLOT. The ELEM, MAT, TYPE, REAL, ESYS, PART (ANSYS LS-
DYNA only), and LOC labels are mutually exclusive, i.e., only one can be specified at a time. Also, turning on
a LINE, AREA, or VOLU label will turn off the MAT, TYPE, REAL, and PART labels.
PowerGraphics [/GRAPHICS,POWER] displays for/PNUM can be problematic. /PNUM,ELEM will display errat-
ically depending on other display command specifications, while /PNUM,LOC and /PNUM,ESYS are not
supported.
Element and volume numbers are not visible for 3-D elements and volumes when Z-buffering is turned on
(/TYPE,,[6,7, or 8]).
/PNUM,DOMAIN,ON or OFF plots the subdomains used to create the CPU domains for Distributed ANSYS
following a solution or substructures following auto substructure generation.
Use /PSTATUS or /PNUM,STAT to show settings. Use /PNUM,DEFA to reset all specifications back to default.
Use the /NUMBER command to control whether numbers and colors are displayed together.
This command is valid in any processor
Menu Paths
Utility Menu>PlotCtrls>Numbering
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1308
/PNUM
POINT
Specifies "Point flow tracing settings" as the subsequent status topic.
POST1: Status
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>General Postproc>Trace Points
POLY
Creates a polygonal area based on working plane coordinate pairs.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Defines a polygonal area on the working plane. The area will be defined with NPT keypoints and NPT lines,
where NPT (must be at least 3) is the number of coordinate pairs defined with the PTXY command. See the
RPOLY and RPR4 commands for other ways to create polygons.
Menu Paths
This command cannot be accessed from a menu.
/POLYGON, NVERT, X1, Y1, X2, Y2, X3, Y3, X4, Y4
Creates annotation polygons (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVERT
Number of vertices of polygon (3
NVERT
PCOLOR
15):
1373
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PSPEC
0
Black.
1
Red-Magenta.
2
Magenta.
3
Blue-Magenta.
4
Blue.
5
Cyan-Blue.
6
Cyan.
7
Green-Cyan.
8
Green.
9
Yellow-Green.
10
Yellow.
11
Orange.
12
Red.
13
Dark Gray.
14
Light Gray.
15
White.
KFILL
Polygon fill key:
0
Hollow polygon.
1
Filled polygon.
KBORDR
Polygon border key:
0
No border.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1374
/PSPEC
1
Border.
Notes
Creates annotation polygon attributes to control certain characteristics of the polygons created via the
/POLYGON, /PMORE, /PCIRCLE and /PWEDGE commands. This is a command generated by the Graphical
User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation is used. This command is
not intended to be typed in directly in an ANSYS session (although it can be included in an input file for
batch input or for use with the /INPUT command).
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
PSPRNG, NLOC, TYPE, K, DX, DY, DZ, ELEM
Defines a spring constraint in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Node where spring is to be placed. Defaults to current run starting point.
TYPE
Type of spring:
TRAN
Translational (default).
ROT
Rotational.
K
Spring constant value (must be positive).
DX, DY, DZ
Increment (in terms of the active coordinate system components) to determine spring ground point.
Spring length must not be zero. Constraints are automatically generated at the ground point.
ELEM
Element number to be assigned to spring (defaults to the previous maximum element number (MAXEL
+ 1)).
Notes
Defines a spring constraint (spring element COMBIN14) at a given location in a piping run. See the PREP7
RUN command.
1375
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PSPRNG
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Spring Support
/PSTATUS, WN
Displays the global or window display specifications.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number for status (defaults to global specifications).
Notes
Displays the current global or window display specifications. Global display specifications are common to
all windows (e.g. /SHOW, etc.). Window display specifications are specific to one window (e.g. /VIEW, /TYPE,
etc.).
This command is valid in any processor.
Menu Paths
Utility Menu>List>Status>Graphics>General
Utility Menu>List>Status>Graphics>Window 1
Utility Menu>List>Status>Graphics>Window 2
Utility Menu>List>Status>Graphics>Window 3
Utility Menu>List>Status>Graphics>Window 4
Utility Menu>List>Status>Graphics>Window 5
PSTRES, Key
Specifies whether prestress effects are calculated or included.
SOLUTION: Nonlinear Options
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS
Key
Prestress key:
OFF
Do not calculate (or include) prestress effects (default).
ON
Calculate (or include) prestress effects.
Notes
Specifies whether or not prestress effects are to be calculated or included. Prestress effects are calculated
in a static or transient analysis for inclusion in a buckling, modal, harmonic (Method = FULL or REDUC),
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1376
/PSTATUS
transient (Method = REDUC), or substructure generation analysis. If used in SOLUTION, this command is
valid only within the first load step.
If the prestress effects are to be calculated in a nonlinear static or transient analysis (for a prestressed modal
analysis of a large-deflection solution), you can issue a SSTIF,ON command (rather than a PSTRES,ON com-
mand) in the static analysis.
If you apply thermal body forces during a static analysis to calculate prestress effects, do not delete the
forces during any subsequent full harmonic response analyses. If you delete the thermal body forces, the
thermal prestress effects will not be included in the harmonic response analysis. Temperature loads used to
define the thermal prestress will also be used in the full harmonic response analysis as sinusoidally time-
varying temperature loads.
A prestress effect applied with non-follower loads resists rigid body rotation of the model. For example, an
unsupported beam with axial tensile forces applied to both ends will have two nonzero rotational rigid body
modes.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Basic
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Sol'n Controls>Basic
/PSYMB, Label, KEY
Shows various symbols on displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Show symbols as selected from the following labels:
CS
Local coordinate systems.
NDIR
Nodal coordinate systems (on rotated nodes only).
ESYS
Element coordinate systems (element displays only).
LDIR
Line directions (line displays only).
LDIV
Controls the display of element divisions on lines.
ADIR
Area direction symbol (for keypoint, line, area and volume plots).
1377
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PSYMB
LAYR
Layer orientations (relative to the projected element x-axis) or fiber orientations in smeared reinforcing
elements. Used only within an element display. Use KEY to specify the layer number.
PCON
Convergence criterion symbols on nodes (for a p-method analysis).
ECON
Element mesh symbols on keypoints and lines.
DOT
Larger symbols displayed for node and keypoint locations. When Label = DOT, KEY = 1 by default.
XNOD
Extra node of surface or circuit elements.
FBCS
Force boundary condition scaling. Subsequent KEY value determines whether or not to scale the
applied and derived forces/moments to the same maximum value.
DEFA
Resets the symbol keys so that ANSYS displays none of the symbols controlled by the /PSYMB
command. The value of the KEY field is ignored.
STAT
Prints the status of the settings of the symbol keys controlled by the /PSYMB command. The KEY
field is ignored.
KEY
Symbol key:
-1
Effective only if Label = LAYR and solid shape element display (/ESHAPE) is active. Orientation of
all layers appears with the solid shape element display.
0
No symbol (default). If Label = LDIV, then KEY= 0 indicates that the displayed element divisions
will correspond to the existing mesh (the word MESHED or EXISTING can also be substituted). Also,
for Label = LDIV, if you execute any meshing command (such as AMESH or VMESH), KEY is set to
0 (MESHED) automatically. If Label = FBCS, then KEY= 0 indicates that boundary condition scaling
will not be common. The applied and derived forces/moments will be scaled to their respective
maximum values.
1
Include symbol. If Label = LDIV, then KEY = 1 indicates that the displayed line divisions will corres-
pond to the value assigned by LESIZE (the word LESIZE can also be substituted). Also, for Label =
LDIV, if you execute the LESIZE command, KEY is set to 1 (LESIZE) automatically. If Label = FBCS,
then KEY= 1 indicates that boundary condition scaling will be common. The applied and derived
forces/moments will be scaled to the same maximum value.
N
If Label = LAYR, then N is equal to the layer number. If Label = DOT, then N can be equal to
0,1,.....15, indicating the dot size. If Label = LDIV, then KEY = -1, indicates that no element divisions
will be displayed (the word OFF can also be substituted).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1378
/PSYMB
Notes
Includes various symbols on the display. Triads are right-handed with x displayed as the longest leg. Where
color is displayed, x is white, y is green, and z is blue. For beams, x is always along the length of the element.
For lines, an arrow represents the direction of a line, from the beginning keypoint to the end keypoint. See
/PLOPTS command for additional display options. Use /PSTATUS or /PSYMB,STAT to display settings. Use
/PSYMB,DEFA to reset all specifications back to their defaults. The command /PSYMB,ECON,1 causes the
symbol "M" to be displayed on keypoints and lines associated with meshed entities. When you issue the
command /PSYMB,DOT,1, a larger symbol is displayed for each node and keypoint location.
PowerGraphics (/GRAPHICS,POWER) does not support /PSYMB,ESYS and /PSYMB,LAYR.
If KEY = N and PowerGraphics is off, the centroid of the surface elements is connected to the extra node
using a gray line. However, if PowerGraphics is on, the color of the line connecting the centroid to the extra
node is the same as that for the elements themselves (as determined by /PNUM).
When Label = LAYR, the layer systems can be visualized with all current-technology layered elements and
the smeared reinforcing element REINF265.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>On Areas
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>On Lines
Main Menu>Preprocessor>Modeling>Move / Modify>RotateNode>To Surf Norm>with Area
Main Menu>Preprocessor>Trefftz Domain>TZ Symmetry
Utility Menu>PlotCtrls>Symbols
PTEMP, TOUT, TIN
Defines the pipe wall temperatures in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
TOUT
Outer pipe wall temperature. If NONE, reset temperature specification to none (BFUNIF will be assigned).
TIN
Inner pipe wall temperature (defaults to TOUT).
Command Default
Assign uniform temperature BFUNIF to elements.
Notes
Defines the pipe wall temperatures in a piping run. These temperatures are assigned to the elements as
they are generated. See the PREP7 RUN command.
1379
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PTEMP
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Loads
PTR, LOC, BASE
Dumps the record of a binary file.
AUX2: Binary Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LOC, BASE
Dump the file record starting at pointer LOC. BASE is the base pointer, and would be used if LOC is a
relative pointer.
Notes
Dumps the record of the file named on the AUX2 FILEAUX2 command according the format specified on
the FORM command.
Menu Paths
This command cannot be accessed from a menu.
PTXY, X1, Y1, X2, Y2, X3, Y3, X4, Y4
Defines coordinate pairs for use in polygons and prisms.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
X1, Y1, X2, Y2, X3, Y3, X4, Y4
X and Y coordinate pairs on the working plane.
Notes
Defines coordinate pairs for use in polygons and prisms [POLY, RPRISM]. The coordinates must be in the
Cartesian coordinate system. The coordinate pairs must be input in a continuous order. PTXY may be repeated
(up to 100 pairs) until the required pairs have been defined. The pairs will be saved until either the POLY
or PRISM command is entered. Use PTXY,STAT to list the saved coordinate pairs. Use PTXY,DELE to delete
all the saved coordinate pairs. See the RPOLY, RPRISM, and RPR4 commands for other ways to create
polygons and prisms.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1380
PTR
PUNIT, KOPT
Selects the system of length units to be used in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
KOPT
Units key:
0
Input units are consistent (no conversions are done) (default).
FTIN or 1
English units (feet A, inch B, fraction of inch C/D). Use A+B+C/D format for PDRAG, BRANCH, RUN,
BEND, MITER, REDUCE, VALVE, BELLOW, FLANGE, PSPRNG, and PGAP commands. Precede by "-
'' sign for negative coordinates. (Example: 5+6+7/16 for 5 ft. 6-7/16 in., +3 for 3 in., -0+3 for -3 in.,
+0+9/16 for 9/16 in.).
The two signs should not be consecutive. A, B, C, and D must be integers. Use B+C/D format for
PSPEC, PINSUL, and PCORRO commands. (Example: 2 for 2 in., 3+1/2 for 3-1/2 in., +3/8 for 3/8 in.)
METRIC or 2
Metric units (meter A, centimeter B, fraction of cm C/D). Use as explained for English units. (Example:
5+6+7/10 for 5 m 6-7/10 cm with PDRAG command.)
Command Default
Input units are consistent (no conversions are done).
Notes
Selects the system of length units to be used for the piping commands. Mixed length units require a + sign
to delimit (or position) the units in the system and are converted to the smallest unit of the system (inches
or centimeters) upon input.
Note
This conversion is local only to pure length units of the piping commands listed. Other units and
units for other commands must be input to be consistent with the smallest length unit of the
system used.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Specifications
1381
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
PUNIT
PVECT, Oper, LabXR, LabYR, LabZR
Interpolates a set of items onto a path.
POST1: Path Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Oper
Valid operations for geometry operations along a path are:
NORM
Defines a unit normal vector at each interpolation point in the direction of the cross product of the
tangent to the path and the active Z axis. Resulting vector components are in the active coordinate
system (which must be Cartesian).
TANG
Defines a unit vector tangent to the path at each interpolation point. Vector components are in the
active coordinate system (which must be Cartesian).
RADI
Defines the position vector of each interpolation point of the path from the center of the active co-
ordinate system (which must be Cartesian).
LabXR
Label (8 characters maximum) assigned to X-component of the resulting vector.
LabYR
Label (8 characters maximum) assigned to Y-component of the resulting vector.
LabZR
Label (8 characters maximum) assigned to Z-component of the resulting vector.
Notes
Defines and interpolates a set of labeled path items along predefined path [PATH] and performs various
geometric operations on these path items. A path item must be defined before it can be used with other
path operations. Additional path items may be defined with the PDEF, PCALC, PDOT, and PCROSS commands.
Path items may be listed or displayed with the PLPATH, PLPAGM and PRPATH commands. Path geometry
items (XG, YG, ZG, S) are automatically interpolated (in the active CSYS) if not done so previously with the
PDEF command.
Menu Paths
Main Menu>General Postproc>Path Operations>Unit Vector
/PWEDGE, XCENTR, YCENTR, XLRAD, ANGLE1, ANGLE2
Creates an annotation wedge (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XCENTR
Wedge X center location (-1.0 < X < 2.0).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1382
PVECT
YCENTR
Wedge Y center location (-1.0 < Y < 1.0).
XLRAD
Wedge radius length.
ANGLE1
Starting angle of wedge.
ANGLE2
Ending angle of wedge. The wedge is drawn counterclockwise from the starting angle, ANGLE1, to the
ending angle, ANGLE2.
Notes
Creates an annotation wedge to be written directly onto the display at a specified location. This is a command
generated by the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation
is used. This command is not intended to be typed in directly in an ANSYS session (although it can be included
in an input file for batch input or for use with the /INPUT command).
All wedges are shown on subsequent displays unless the annotation is turned off or deleted. Use the /LSPEC
and the /PSPEC command to set the attributes of the wedge.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
1383
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/PWEDGE
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1384
Q Commands
QDVAL, TBLNO1, TBLNO2, SV1, SV2, SV3, SV4, SV5, SV6, SV7
Defines PSD quadspectral values.
SOLUTION: Spectrum Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
TBLNO1
First input PSD table number associated with this spectrum.
TBLNO2
Second input PSD table number associated with this spectrum.
SV1, SV2, SV3, . . . , SV7
PSD quadspectral values corresponding to the frequency points [PSDFRQ].
Notes
Defines PSD quadspectral values to be associated with the previously defined frequency points. Repeat
QDVAL command with the same table number for additional points. Unlike autospectra [PSDVAL], the
quadspectra can be positive or negative. The quadspectral curve segment where there is a sign change is
interpolated linearly (the rest of the curve segments use log-log interpolation). For better accuracy, choose
as small a curve segment as possible wherever a sign change occurs.
Two table numbers are required since values are off-diagonal terms. This command is valid for SPOPT,PSD
only.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>PSD>Correlation>Quadspectral
Main Menu>Solution>Load Step Opts>Spectrum>PSD>Correlation>Quadspectral
1385
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
QFACT
Calculates the quality factor for high-frequency electromagnetic resonators.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> <> <> <> PP <> <> <>
Notes
The QFACT command macro calculates the quality factor for high-frequency electromagnetic resonators. It
returns the quality factor as a scalar parameter, QFACT. To compute the quality factor, the macro uses the
stored energy, surface losses, and dielectric losses.
See magnetic macros for further details.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Cavity>Q-factor
QSOPT, Opt
Specifies quasi static radiation options.
SOLUTION: Radiosity
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
Opt
Quasi static option:
OFF
Do not run transient radiation problem to steady-state (default).
ON
Run transient radiation problem to steady-state.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Quasi-Static
Main Menu>Solution>Load Step Opts>Time/Frequenc>Quasi-Static
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1386
QFACT
QUAD, NODE1, NINTR, NODE2, NFILL, NSTRT, NINC, PKFAC
Generates a quadratic line of nodes from three nodes.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NODE1
Begin fill-in from this node location. If NODE1 = P, graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI).
NINTR
Intermediate or guiding node. Quadratic curve will pass through this location. NINTR may have any
node number and any location. If the quadratic line also generates a node with number NINTR, the
generated location overrides the previous NINTR location.
NODE2
End quadratic fill-in at this node location.
NFILL
Fill-in NFILL nodes between NODE1 and NODE2 (defaults to |NODE2-NODE1|-1). NFILL must be positive.
NSTRT
Node number assigned to first filled-in node (defaults to NODE1 + NINC).
NINC
Add this increment to each of the remaining filled-in node numbers (may be positive or negative). Defaults
to (NODE2-NODE1)/(NFILL + 1), i.e., linear interpolation.
PKFAC
Peak location factor. If PKFAC=0.5, the peak of the quadratic shape occurs at the NINTR location. If 0.0
< PKFAC < 0.5, the peak occurs to the NODE2 side of the NINTR location. If 0.5 < PKFAC < 1.0, the
peak occurs to the NODE1 side of the NINTR location. Defaults to 0.5.
Notes
Generates a quadratic line of nodes (in the active coordinate system) from three nodes. The three nodes
determine the plane of the curve and may have been defined in any coordinate system. Any number of
nodes may be filled-in and any node number sequence may be assigned.
The quadratic line feature uses three nodes (NODE1,NINTR,NODE2) to determine the plane of the curve.
The curve passes through the three points, beginning from NODE1, through the intermediate (or guiding)
point NINTR, and toward NODE2.
Generated nodes are also quadratically spaced. If the guiding node number is within the set being generated,
it will be relocated according to the quadratic spacing.
The peak location factor is used to determine how the quadratic fits through the three points. Various
nodal progressions can be obtained by different combinations of PKFAC and the guiding node location. If
the guiding node is at mid-length between NODE1 and NODE2, 0.293
n
= natural frequency of the system,
k m /
n
= ratio of viscous damping to critical damping, c/c
cr
X
o
= ground displacement
The solution of this equation for the maximum response,
X
max
, at various frequencies results in the spectral
response curve. See the Theory Reference for the Mechanical APDL and Mechanical Applications for calculation
details.
Calculations are based on a numerical integration scheme with the displacement time-history data from the
file as the input ground-forcing function. The integration time step (argument DTIME on the RESP command)
and the damping coefficient (argument RATIO) are constant over the frequency range. The number of cal-
culations done per displacement spectral response curve is the product of the number of input solution
points (TMAX-TMIN)DTIME and the number of oscillating systems (frequencies located in variable LFTAB).
Input solution points requested (by DTIME and the frequency range) at a time not corresponding to an
actual displacement solution time on the file are read from the next available time. The user has the option
of calculating either a displacement, velocity, or acceleration spectral response.
Menu Paths
Main Menu>TimeHist Postpro>Generate Spectrm
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1412
RESP
RESUME, Fname, Ext, --, NOPAR, KNOPLOT
Resumes the database from the database file.
DATABASE: Set Up
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname.
Ext
Filename extension (8 character maximum).
The extension defaults to DB if Fname is blank.
--
Unused field.
NOPAR
Parameter resume key:
0
All data in the database, including the scalar parameters, are replaced with the data saved on File.DB
(default).
1
All data in the database, except the scalar parameters, are replaced with the data saved on File.DB.
Note
This option should not be used if array parameters are defined, since existing array para-
meters might be redefined with arbitrary values. See PARSAV and PARRES for a more
general method of preventing the replacement of both scalar and array parameters.)
KNOPLOT
If equal to 1, will suppress automatic plot. Otherwise, if the GUI is on and this RESUME command was
not read from a file, the selected elements from Fname are plotted. (If there are no selected elements,
selected nodes are plotted. If no nodes, volumes; if no volumes, areas; if no areas, lines; if no lines, key-
points. If there are no selected keypoints, the screen is erased.)
Notes
The RESUME command resumes a database file into the ANSYS program. The command causes the database
file (File.DB) to be read, thereby resetting the database (including any geometry settings) either a) as it
was at the last SAVE command, or b) as it was saved with the last /EXIT command, whichever was last. For
multiple load step analyses (because only the data for one load step at a time may reside in the database),
the load step data restored to the database will correspond to the load step data written when the save
occurred.
1413
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RESUME
If the database file was saved in another ANSYS product, it may contain element type and KEYOPT specific-
ations which are invalid in the "resuming" product. Immediately after the database resume is completed,
you should redefine these invalid element types and KEYOPT settings to valid ones (ET, KEYOPT).
RESUME checks your database for mixed mode (both SMOOTH and FACETED) geometry. ANSYS notifies you
if both types of geometry are found in the database. You can continue if you get this message; however,
you may encounter problems during Boolean operations. If you do not want to continue the analysis with
this database, you must either recreate or reimport the geometry.
This command is valid in any processor. If used in SOLUTION, this command is valid only within the first
load step.
Menu Paths
Utility Menu>File>Resume from
Utility Menu>File>Resume Jobname.db
RESVEC, Key
Calculates or includes residual vectors.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Key
Residual vector key:
OFF
Do not calculate or include residual vectors. This option is the default.
ON
Calculate or include residual vectors.
Command Default
No residual vectors are calculated or included in the analysis.
Notes
In a modal analysis, the RESVEC command calculates residual vectors. In a mode-superposition transient
dynamic, mode-superposition harmonic, PSD or spectrum analysis, the command includes residual vectors.
If rigid-body modes exist, pseudo-constraints are required for the calculation. Issue the D,,,SUPPORT command
to specify only the minimum number of pseudo-constraints necessary to prevent rigid-body motion.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1414
RESVEC
RESWRITE, Fname
Appends results data from the database to a results file.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name extension varies as follows:
.RST for structural, fluid, or coupled-field analyses
.RTH for thermal or electrical analyses
.RMG for magnetic analyses
.RFL for FLOTRAN analyses
Notes
The RESWRITE command appends to the specified file a data set by writing the results data currently in the
database. If the file does not exist, it will be created including the geometry records. The current load step,
substep, and time value are maintained. All data (summable and nonsummable) are written.
The command is primarily intended for use in a top-down substructuring analysis, where the full model is
resumed and the results data read from the use pass results file (SET), and subsequently from all substructure
expansion pass results files (APPEND). The full set of data in memory can then be written out via the
RESWRITE command to create a complete results file (as though you had run a nonsubstructured analysis).
Menu Paths
This command cannot be accessed from a menu.
*RETURN, Level
Returns input stream to a higher level.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Level
Number of levels to move up from the current level.
Negative --
Move relative to current level. For example: *Return,-2 will go up two levels from the current level.
Positive --
Move to absolute level. For example: *Return,2 will go to level 2.
Level 0 is the primary input file.
1415
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*RETURN
Notes
This command is used to jump to the macro call sequence, ending the current macro file, and returning to
the line after the calling line in the previous file. Unlike the *GO command, this command may be used inside
*IF or *DO constructs.
Menu Paths
This command cannot be accessed from a menu.
REXPORT, Target, --, --, LSTEP, SBSTEP, Fname, Ext, --
Exports displacements from an implicit run to ANSYS LS-DYNA.
SOLUTION: Explicit Dynamics
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
Target
The type of analysis run to which displacements are exported.
OFF
Ignore initial displacements.
DYNA
Get initial displacements from an earlier implicit (ANSYS) run and export to an explicit ANSYS LS-
DYNA run (Default).
--, --
Unused fields.
LSTEP
Load step number of data to be exported. Defaults to the last load step.
SBSTEP
Substep number of data to be exported. Defaults to the last substep.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name does not have a default; you must specify a name. It CANNOT be the current Jobname.
Ext
Filename extension (8 character maximum).
The extension must be an RST extension (default). Currently, only structural results are allowed.
--
Unused field.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1416
REXPORT
Notes
This command exports the displacements, rotations, and temperatures calculated in an ANSYS implicit ana-
lysis into the `drelax' file, which is subsequently read in by ANSYS LS-DYNA when a dynamic relaxation or
stress initialization is conducted [EDDRELAX].
This command is not written to the Jobname.CDB file when the CDWRITE command is issued.
Menu Paths
Main Menu>Preprocessor>LS-DYNA Options>Constraints>Read Disp
Main Menu>Solution>Constraints>Read Disp
REZONE, Option, LDSTEP, SBSTEP
Initiates the rezoning process, sets rezoning options, and rebuilds the database.
SOLUTION: Rezoning
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Option
The rezoning method to employ:
MANUAL
Manual rezoning. This method is the default.
LDSTEP
The load step number at which rezoning should occur. The default value is the highest load step number
found in the Jobname.Rnnn files (for the current jobname and in the current directory).
SBSTEP
The substep number of the specified load step (LDSTEP) at which rezoning should occur. The default
value is the highest substep number found in the specified load step in the Jobname.Rnnn files (for
the current jobname and in the current directory).
Notes
The REZONE command rebuilds the database (.db file) based on the specified load step and substep in-
formation, and updates nodes to their deformed position for remeshing.
Before issuing this command, clear the databse via the /CLEAR command.
Currently, ANSYS supports only manual rezoning (Option = MANUAL).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Manual Rezoning>Start
1417
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
REZONE
RFILSZ
Estimates file sizes.
RUNSTATS: Run Statistics Estimator
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Notes
Gives file size estimates for File.ESAV, File.EMAT, File.EROT, File.LN22, File.FULL, File.RST,
File.RTH, and File.RMG. File sizes are given in megabytes (MB). These file estimates are for the solution
of the model that currently resides in the database. This command will cause reordering [WAVES] if reordering
was not already done on the model. See File Management and Files in the Basic Analysis Guide for file de-
scriptions.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Run-Time Stats>Individual Stats
RFORCE, NVAR, NODE, Item, Comp, Name
Specifies the total reaction force data to be stored.
POST26: Set Up
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
NVAR
Arbitrary reference number assigned to this variable (2 to NV [NUMVAR]). Overwrites any existing results
for this variable.
NODE
Node for which data are to be stored. If NODE = P, graphical picking is enabled (valid only in the GUI).
Item
Label identifying the item. Valid item labels are shown in the table below. Some items also require a
component label.
Comp
Component of the item (if required). Valid component labels are shown in the table below.
Name
Thirty-two character name identifying the item on printouts and displays. Defaults to an eight character
label formed by concatenating the first four characters of the Item and Comp labels.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1418
RFILSZ
Notes
Defines the total reaction force data (static, damping, and inertial components) to be stored from single
pass (ANTYPE,STATIC or TRANS) solutions or from an expansion pass of reduced two-pass (ANTYPE,HARMIC
or TRANS) solutions.
Table 255 RFORCE - Valid Item and Component Labels
Valid item and component labels for node results are:
Description Comp Item
X, Y, or Z structural force. X,Y,Z F
X, Y, or Z structural moment. X,Y,Z M
Heat flow. HEAT[1]
Fluid flow. FLOW
Current flow. AMPS
Magnetic flux. FLUX
X, Y, or Z magnetic current segment component. X,Y,Z CSG
Voltage drop VLTG
Current CURT
Charge CHRG
1. For SHELL131 and SHELL132 elements with KEYOPT(3) = 0 or 1, use the labels HBOT, HE2, HE3, . . .,
HTOP instead of HEAT.
Menu Paths
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
/RGB, Kywrd, PRED, PGRN, PBLU, N1, N2, NINC, NCNTR
Specifies the RGB color values for indices and contours.
POST26: Set Up
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Kywrd
Determines how RGB modifications will be applied.
INDEX
Specifies that subsequent color values apply to ANSYS color indices (0-15).
CNTR
Specifies that subsequent color values apply to contours (1-128). Applies to C-option devices only
(i.e. X11C or Win32C).
PRED
Intensity of the color red, expressed as a percentage.
1419
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/RGB
PGRN
Intensity of the color green, expressed as a percentage.
PBLU
Intensity of the color blue, expressed as a percentage.
N1
First index (0-15), or contour (1-128) to which the designated RGB values apply.
N2
Final index (0-15), or contour (1-128) to which the designated RGB values apply.
NINC
The step increment between the values N1 and N2 determining which contours or indices will be con-
trolled by the specified RGB values.
NCNTR
The new maximum number of contours (1-128).
Notes
Issuing the /CMAP command (with no filename) will restore the default color settings.
Menu Paths
Utility Menu>PlotCtrls>Redirect Plots>To GRPH File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL2 File
Utility Menu>PlotCtrls>Redirect Plots>To PSCR File
RIGID, Dof1, Dof2, Dof3, Dof4, Dof5, Dof6
Specifies known rigid body modes (if any) of the model.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Dof1, Dof2, Dof3, . . . , Dof6
Up to six global Cartesian directions of the rigid modes. For a completely free 2-D model, use ALL or
UX, UY, ROTZ. For a completely free 3-D model, use ALL or UX, UY, UZ, ROTX, ROTY, ROTZ. For a con-
strained model, use UX, UY, UZ, ROTX, ROTY, or ROTZ, as appropriate, to specify each and every uncon-
strained direction which exists in the model (not specifying every direction may cause difficulties in ex-
tracting the modes).
Notes
Specifies known rigid body modes (if any) of the model. This command applies only to a component mode
synthesis (CMS) analysis (see the CMSOPT command). Any rigid body modes specified must be permitted
by the applied displacement constraints (i.e., do not specify a rigid body mode in a constrained direction).
Reissue the command to redefine the specification. If used in SOLUTION, this command is valid only within
the first load step.
This command is also valid in PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1420
RIGID
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
RIGRESP, Option, Method, Val1, Val2
Specifies the rigid response calculation.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Option
Flag to activate or deactivate the rigid response calculation:
1 (ON or YES)
Activate.
2 (OFF or NO)
Deactivate. This value is the default.
Method
Method used to calculate the rigid response:
GUPTA
Gupta method.
LINDLEY
Lindley-Yow method.
Val1
If Method = GUPTA, Val1 represents the frequency F1 in Hertz.
If Method = LINDLEY, Val1 is the Zero Period Acceleration (ZPA). If a scale factor is defined (FACT in
the SVTYP command), it is used to scale this value
Val2
If Method = GUPTA, Val2 represents the frequency F2 in Hertz.
Notes
For more information on the rigid response calculation, see Rigid Responses in the Theory Reference for the
Mechanical APDL and Mechanical Applications
This rigid response calculation is only valid for single point response spectrum analysis (SPOPT, SPRS) and
multiple point response spectrum analysis (SPOPT, MPRS) with combination methods (SRSS), complete
quadratic (CQC) or Rosenblueth (ROSE)
This command is also valid in PREP7.
1421
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RIGRESP
Product Restrictions
Only Sptype = SPRS is allowed in ANSYS Professional.
Menu Paths
To be supplied.
RIMPORT, Source, Type, Loc, LSTEP, SBSTEP, Fname, Ext, --, SPSCALE, MSCALE
Imports initial stresses from an explicit dynamics run into ANSYS.
SOLUTION: FE Body Loads
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Source
The type of analysis run from which stresses are imported.
OFF
Ignore initial stresses.
DYNA
Get initial stresses from an earlier explicit (ANSYS LS-DYNA) run (default).
Type
Type of data imported. Note that this is an ANSYS-defined field; the only valid value is STRESS.
Loc
Location where the data is imported. Note that this is an ANSYS-defined field; the only valid value is
ELEM (data imported at the element integration points).
LSTEP
Load step number of data to be imported. Defaults to the last load step.
SBSTEP
Substep number of data to be imported. Defaults to the last substep.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name does not have a default; you must specify a name. It CANNOT be the current Jobname.
Ext
Filename extension (8 character maximum).
The extension must be an RST extension (default).
--
Unused field.
SPSCALE
Stabilization factor. This factor is used in a springback analysis to scale (up or down) the initial stiffness
of the applied spring. No default; input a value only if you want to activate stabilization. If SPSCALE is
blank, stabilization is not activated.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1422
RIMPORT
MSCALE
Acceptable stabilization stiffness (defaults to 1.0 X 10
--4
). In a springback analysis, iterations will stop
when the applied spring stiffness comes down to this value. MSCALE is not used if SPSCALE is blank.
Notes
This command imports initial stress information into ANSYS from an earlier explicit (ANSYS LS-DYNA) run.
The stress state from SHELL163 and SOLID164 elements in the explicit analysis is imported to the correspond-
ing SHELL181 and SOLID185 implicit elements. For the shell elements, the current shell element thickness
is also imported. This command is valid only before the first SOLVE command of the implicit analysis (which
comes after the explicit analysis) and is ignored if issued after subsequent SOLVE commands (that is, stresses
will not be re-imported).
RIMPORT is typically used to perform springback analysis of sheet metal forming. We recommend that you
use SHELL163 elements in the explicit analysis with 3 to 5 integration points through the thickness. This
ensures that the through-thickness stress distribution is transferred accurately to the SHELL181 elements. If
more than 5 integration points are used, ANSYS imports resultants (forces and moments) to the SHELL181
elements. This implies that linearization of the through-thickness stress distribution is assumed in SHELL181
elements. If SHELL163 uses full integration in the shell plane, stress and thickness data are averaged and
then transferred. For the solid elements, the stress at the SOLID164 element centroid is transferred to the
SOLID185 element centroid. If SOLID164 has full integration, the stress is averaged and then transferred.
When the SPSCALE argument is specified, artificial springs with exponentially decaying stiffness (as a
function of iterations) are applied. This technique is recommended only for those cases in which there are
severe convergence difficulties. In general, you should first attempt a springback analysis without using the
stabilization factors SPSCALE and MSCALE. (For more information on springback stabilization, see the ANSYS
LS-DYNA User's Guide.)
This command is not written to the Jobname.CDB file when the CDWRITE command is issued. Further,
the RIMPORT information is not saved to the database; therefore, the RIMPORT command must be reissued
if the database is resumed.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Other>Import Stress
Main Menu>Solution>Define Loads>Apply>Structural>Other>Import Stress
RITER, NITER
Supplies an estimate of the number of iterations for time estimates.
RUNSTATS: Run Statistics Estimator
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
NITER
Estimated number of iterations (or load steps in a linear, static analysis) for the analysis (defaults to 1).
1423
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RITER
Command Default
One iteration.
Notes
This estimate will be used by the program to calculate estimated run times [RTIMST].
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Run-Time Stats>Iter Setting
RLIST, NSET1, NSET2, NINC
Lists the real constant sets.
PREP7: Real Constants
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET1, NSET2, NINC
List real constant sets from NSET1 to NSET2 (defaults to NSET1) in steps of NINC (defaults to 1). If
NSET1 = ALL (default), ignore NSET2 and NINC and list all real constant sets [R].
Notes
The real constant sets listed contain only those values specifically set by the user. Default values for real
constants set automatically within the various elements are not listed.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Properties>All Real Constants
Utility Menu>List>Properties>Specified Real Constants
RMALIST
Lists all defined master nodes for a ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1424
RLIST
Menu Paths
Main Menu>ROM Tool>Setup>Master Nodes>List
RMANL, Fname, Ext, --, Dimn, Oper
Assigns model database, dimensionality, and operating direction for the ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Fname
Database file name and directory path (248 characters maximum, including directory). The file name
defaults to Jobname.
Ext
File extension (8 character maximum). The extension defaults to db.
--
Unused field.
Dimn
Model dimensionality:
2
2-D models
3
3-D Models
Oper
Primary operating direction:
X
direction
Y
direction
Z
direction
Notes
Required Inputs:
Model database containing a structural physics file, title "STRU".
Model database containing an electrostatic physics file, title "ELEC".
Model database containing an area or volume component of the electrostatic domain to be morphed,
title "AIR".
Model database containing a node component of the neutral plane nodes, named "NEUN".
Model database containing node components of conductors, named "CONDi", where "i" is the conductor
number.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1425
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMANL
Menu Paths
Main Menu>ROM Tool>Setup>Model Features
RMASTER, Node, Lab
Defines master nodes for the ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Node
Node number at which master degree of freedom is defined If Node = P, graphical picking is enabled
and all remaining command fields are ignored (valid only in the GUI).
Lab
Valid labels are "ADD" (default) and "DEL".
Notes
Defines master nodes for the ROM. Master nodes are used to track the total displacement of a structure in
the operating direction [RMANL]. They may be used as attachment points for 1-D structural elements during
a ROM use pass via the UX degree of freedom.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Setup>Master Nodes>Define
Main Menu>ROM Tool>Setup>Master Nodes>Delete
RMCAP, RefName, C1, C2
Defines lumped capacitance pairs between conductors C1 and C2 for a ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
RefName
Reference name for capacitance pair definition.
C1
First conductor (between 1 and 5).
C2
Second conductor (between 1 and 5).
Notes
For a capacitance definition between conductor C1 and C2, node components COND%C1% and COND%C2%
(see CM command) must be present containing the conductor nodes. If C1 and C2 are blank, the capacitance
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1426
RMASTER
definition with RefName will be deleted. (For example, if C1 = 1, and C2 = 2, then node components COND1
and COND2 must be defined).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Setup>Capacitances>Define>All Capacitances
Main Menu>ROM Tool>Setup>Capacitances>Define>Single Capacitance
Main Menu>ROM Tool>Setup>Capacitances>Delete
RMCLIST
Lists all lumped capacitance pairs defined.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Setup>Capacitances>Define>List
Main Menu>ROM Tool>Setup>Capacitances>List
/RMDIR, Dir
Removes (deletes) a directory.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Dir
The directory to remove. If no path is provided, it will be assumed to be in the current working directory.
All files in the directory are also removed.
Notes
Removes a directory on the computer ANSYS is currently running on. No warning or prompt is given, so use
with extreme caution.
Menu Paths
This command cannot be accessed from a menu.
1427
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/RMDIR
RMEMRY
Prints memory statistics for the current model.
RUNSTATS: Run Statistics Estimator
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Notes
Memory statistics include work space usage, database size, binary input/output buffers, and available ANSYS
scratch space. Also the maximum model size that will fit in the available ANSYS scratch space is displayed.
The memory statistics are displayed in units of kilobytes (KB) or megabytes (MB).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Run-Time Stats>Individual Stats
RMFLVEC
Writes eigenvectors of fluid nodes to a file for use in damping parameter extraction.
POST1: Special Purpose
MP ME <> <> <> <> <> <> <> <> <> PP <> EME MFS
Notes
RMFLVEC extracts the modal information from the modal results file for all nodes specified in a node com-
ponent called 'FLUN'. This component should include all nodes which are located at the fluid-structural in-
terface. Mode shapes, element normal orientation, and a scaling factor are computed and stored in a file
Jobname.EFL. For damping parameter extraction, use the DMPEXT command macro. See "Thin Film
Analysis" for more information on thin film analyses.
FLUID136 and FLUID138 are used to model the fluid interface. Both the structural and fluid element types
must be active. The fluid interface nodes must be grouped into a component 'FLUN'. A results file of the
last modal analysis must be available.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>ThinFilm>Extract Eigv
Main Menu>Solution>ThinFilm>DampExtract>Eigenfrequency
Main Menu>Solution>ThinFilm>DampExtract>Frequency Range
Main Menu>Solution>ThinFilm>RayleighDamp
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1428
RMEMRY
RMLVSCALE, Nload, Fact1, Fact2, Fact3, Fact4, Fact5
Defines element load vector scaling for a ROM use pass.
REDUCED ORDER MODELING: Use Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Nload
Total number of load cases to be considered within a ROM use pass. If Nload = "DELETE", all defined
load vectors are deleted.
Fact1, Fact2, Fact3, . . . , Fact5
Scale factors applied to load vectors (maximum 5). Defaults to 0.
Notes
Specifies the element load scale factor applied to a ROM analysis use pass. Element load vectors are extracted
from a Static Analysis using the RMNDISP command. Up to 5 element load vectors may be scaled and applied
to a ROM use pass.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Load Vector>For ROM
Main Menu>Preprocessor>Loads>Define Loads>Delete>Load Vector>For ROM
Main Menu>Solution>Define Loads>Apply>Load Vector>For ROM
Main Menu>Solution>Define Loads>Delete>Load Vector>For ROM
RMMLIST
Lists all mode specifications for the ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Mode Selection>List
1429
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMMLIST
RMMRANGE, Mode, Key, Min, Max, Nstep, Damp, Scale
Defines and edits various modal parameters for the ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Mode
Mode number. Must be lower or equal to the number of modes extracted via the RMNEVEC command.
Key
Mode classification key. Valid keys are:
DOMINANT
Dominant mode
RELEVANT
Relevant mode
UNUSED
Unused mode. Do not consider mode in ROM.
Min
Lower bound for fit range of mode.
Max
Upper bound for fit range of mode.
Nstep
Number of equidistant steps in fit range of mode.
Damp
Modal damping factor. Defaults to 0.0.
Scale
Modal scaling factor.
Notes
When selected manually (RMMSELECT), modes must be classified as dominant, relevant, or unused. Dominant
modes (Key = DOMINANT) are basis functions with large amplitudes. Relevant modes (Key = RELEVANT)
are influenced by the dominant modes but do not cause interactions among themselves due to the small
amplitude. This assumption leads to essential speed up of the sample point generator (see RMSMPLE).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Mode Selection>Edit
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1430
RMMRANGE
RMMSELECT, Nmode, Method, Dmin, Dmax
Selects modes for the ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Nmode
Total number of modes to be selected
Method
Method for mode selection. Valid methods are:
TMOD
Automated selection using a test load. TMOD must be enclosed in single quotes.
NMOD
First Nmode eigenmodes. NMOD must be enclosed in single quotes.
Dmin
Lower bound for total deflection range.
Dmax
Upper bound for total deflection range.
Notes
Select pertinent modes for use in a ROM. Pertinent mode selection may be enhanced by using the deflection
state of the structure representative of the operating nature of the device (Method = TMOD). A static ana-
lysis with an applied Test Load may be used. The test load displacements must be extracted at the neutral
plane of the device (if the device is stress-stiffened), or at any plane of the device (non-stress-stiffened). A
node component "NEUN" must be defined for the plane of nodes, and the displacements extracted using
the RMNDISP command prior to issuing this command. If Method = NMOD, use the first Nmode eigenmodes
to select the pertinent modes for the ROM tool. Only those modes are selected that act in the operating
direction of the structure [RMANL].
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Required Input Files
jobname.evx, jobname.evy, jobname.evz, jobname.evn, jobname.evl
Optional Input File
Test load and element load neutral plane displacement files: jobname.tld, jobname.eld
Menu Paths
Main Menu>ROM Tool>Mode Selection>Select
1431
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMMSELECT
RMNDISP, LoadT, Loc
Extracts neutral plane displacements from a test load or element load solution for the ROM method.
REDUCED ORDER MODELING: Preparation
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
LoadT
Load type. Load type must be an alphanumeric string enclosed in single quotes. Valid load types are
'TLOAD' for the test load and 'ELOAD' for the element load.
Loc
Determines whether file will be overwritten or appended. Valid labels are 'WRITE' or 'APPEND'. Defaults
to 'WRITE' for test load.
Notes
This command extracts the displacements at a neutral plane of a model. If LoadT = 'TLOAD', extract displace-
ments for a test load on a structure that represents the expected deflection state. A test load is used to assist
in the automatic mode selection for the ROM mode characterization. If LoadT = 'ELOAD', extract the neutral
plane displacements for an element load that will be used in the use pass of a ROM analysis. Typical element
loads are gravity, and pressure loading. The element loads may be scaled [RMLVSCALE] during the use pass.
The command requires a node component named "NEUN" to be defined. These nodes represent the nodes
at the neutral plane of a structure (in the case of a stress-stiffened structure), or at any plane in the structure
(non stress-stiffened case).
For LoadT = 'TLOAD', node displacements are written to the file jobname.tld. For LoadT = 'ELOAD',
node displacements are written to the file jobname.eld. Up to 5 element load cases may be written to
the file jobname.eld.
This command is only valid in POST1.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Output Files
For a test load jobname.tld
For an element load jobname.eld
Menu Paths
Main Menu>General Postproc>ROM Operations>Extract NP DISP
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1432
RMNDISP
RMNEVEC
Extracts neutral plane eigenvectors from a modal analysis for the ROM method.
REDUCED ORDER MODELING: Preparation
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Notes
This command extracts the eigenvectors at a neutral plane of a model from a modal analysis. The modal
analysis must have expanded modes [MXPAND] in order to process the data. Only the first 9 modes are
considered. The command requires a node component named "NEUN" to be defined. These nodes represent
the nodes at the neutral plane of a structure (in the case of a stress-stiffened structure), or at any plane in
the structure (non stress-stiffened case).
This command is only valid in POST1.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Output Files
jobname.evx, jobname.evy, jobname.evz, jobname.evn, jobname.evl
Menu Paths
Main Menu>General Postproc>ROM Operations>Extract NP Eigv
RMODIF, NSET, STLOC, V1, V2, V3, V4, V5, V6
Modifies real constant sets.
PREP7: Real Constants
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSET
Existing set to be modified.
STLOC
Starting location in table for modifying data. For example, if STLOC = 1, data input in the V1 field is the
first constant in the set. If STLOC = 7, data input in the V1 field is the seventh constant in the set, etc.
Must be greater than zero.
V1
New value assigned to constant in location STLOC. If zero (or blank), a zero value will be assigned.
V2, V3, V4, . . . , V6
New values assigned to constants in the next five locations. If blank, the value remains unchanged.
Notes
Allows modifying (or adding) real constants to an existing set [R] at any location.
This command is also valid in SOLUTION.
1433
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMODIF
Menu Paths
This command cannot be accessed from a menu.
RMORE, R7, R8, R9, R10, R11, R12
Adds real constants to a set.
PREP7: Real Constants
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
R7, R8, R9, . . . , R12
Add real constants 7 to 12 (numerical values or table names) to the most recently defined set.
Notes
Adds six more real constants to the most recently defined set. Repeat the RMORE command for constants
13 to 18, again for 19-24, etc.
If using table inputs (SURF151, SURF152, FLUID116, CONTA171, CONTA172, CONTA173, CONTA174, and
CONTA175 only), enclose the table name in % signs (e.g., %tabname%).
When copying real constants to new sets, ANSYS recommends that you use the command input. If you do
use the GUI, restrict the real constant copy to only the first six real constants (real constants seven and
greater will be incorrect for both the master and copy set).
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Spring>Nonlin Rotary
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Mechanical>Spring>Nonlin Trans
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>Transducer>ElecMech
Main Menu>Preprocessor>Modeling>Create>Circuit>Edit Real Cnst
RMPORDER, Ord1, Ord2, Ord3, Ord4, Ord5, Ord6, Ord7, Ord8, Ord9
Defines polynomial orders for ROM functions.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Ord1, Ord2, Ord3, . . . , Ord9
Polynomial orders for modes. Ordi specifies the polynomial order for modei. Modes are ordered as ex-
tracted from a modal analysis using the RMNEVEC command. Defaults to 0 if mode i is unused; default
to nstep(i) -1 for dominant or relevant modes, where nstep(i) is the number of equidistant steps in fit
range of mode i. nstep(i) is automatically set by RMMSELECT or modified by the RMMRANGE command.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1434
RMORE
Menu Paths
Main Menu>ROM Tool>Resp Surface>Poly Order
RMRESUME, Fname, Ext, --
Resumes ROM data from a file.
REDUCED ORDER MODELING: Set Up
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Fname
Name and directory path of the ROM database file (248 character maximum). Default to Jobname.
Ext
Extension of the ROM database file. Default to .rom.
--
Unused field.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>ExpansionPass
Main Menu>Preprocessor>Loads>Analysis Type>New Analysis
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>ROM>Database
Main Menu>ROM Tool>Rom Database>Resume
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>ExpansionPass
Main Menu>Solution>Analysis Type>New Analysis
RMRGENERATE
Performs fitting procedure for all ROM functions to generate response surfaces.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Notes
The fitting procedure uses modal analysis data and function data generated using the RMSMPLE command
and specifications set forth in the RMROPTIONS command. The files jobname_ijk.pcs (modes i, j, k)
will be generated containing the coefficients of the response surfaces. These files are needed for the ROM
Use Pass along with a ROM data base file [RMSAVE].
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1435
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMRGENERATE
Input Files
Strain energy and capacitance data file jobname_ijk.dec
Output Files
Response surface coefficients jobname_ijk.pcs (modes i, j, k)
Menu Paths
Main Menu>ROM Tool>Resp Surface>Fit Functions
RMROPTIONS, RefName, Type, Invert
Defines options for ROM response surface fitting.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
RefName
Reference name of ROM function to be fitted. Valid reference names are "SENE" for the strain energy of
the structural domain and any capacitance reference name previously defined by means of RMCAP
command for the electrostatic domain.
Type
Type of fitting function to be applied for regression analysis. Valid types are:
LAGRANGE
Lagrange type (default)
REDLAGRANGE
Reduced Lagrange type
PASCAL
Pascal type
REDPASCAL
Reduced Pascal type
Invert
Flag to specify whether data should be inverted prior to fitting.
0
Do not invert data (default for SENE)
1
Invert data input on Refname field (recommended for capacitance functions).
Notes
The objective of response surface fit is to compute an analytical expression for the strain energy and the
capacitance as functions of modal amplitudes.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1436
RMROPTIONS
Menu Paths
Main Menu>ROM Tool>Resp Surface>Options
RMRPLOT, RefName, Type, Mode1, Mode2
Plots response surface of ROM function or its derivatives with respect to the dominant mode(s).
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
RefName
Reference name of ROM function. Valid reference names are "SENE" for the strain energy of the mechan-
ical domain and any capacitance definition, previously defined by means of the RMCAP command, for
the electrostatic domain.
Type
Type of data to be plotted. Valid types are:
FUNC
Response surface (default)
FIRST
First derivative of response surface with respect to Mode1.
SECOND
Second derivative of response surface with respect to Mode1 and Mode2.
Mode1
First mode number (used for Type = "FIRST" and Type = "SECOND" only).
Mode2
Second mode number (used for Type = "SECOND" only).
Notes
The objective of response surface fit is to compute an analytical expression for the strain energy and the
capacitance as functions of modal amplitudes. This command assumes that the coefficient files job-
nam_ijk.pcs are available [RMRGENERATE]. Visualization of the response surface will help to evaluate
the validity of the function fit.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Resp Surface>Plot
1437
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMRPLOT
RMRSTATUS, RefName
Prints status of response surface for ROM function.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
RefName
Reference name of ROM function. Valid reference names are "SENE" for the strain energy of the mechan-
ical domain and any capacitance reference names [RMCAP], for the electrostatic domain.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Resp Surface>Status
RMSAVE, Fname, Ext, --
Saves ROM data to file.
REDUCED ORDER MODELING: Set Up
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Fname
Name and directory path of the ROM database file. Default to Jobname.
Ext
Extension of the ROM database file. Default to .rom.
--
Unused field.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>ROM Tool>Rom Database>Save
RMSMPLE, Nlgeom, Cap, Seqslv, Eeqslv
Runs finite element solutions and obtains sample points for the ROM method.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Nlgeom
Specify whether a large or small deflection analysis is to be performed for the mechanical domain:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1438
RMRSTATUS
OFF (or 0)
Perform small deflection analysis (default).
ON (or 1)
Perform large deflection analysis.
Cap
Capacitance calculation method.
CHARGE
Compute capacitance based on the charge voltage relationship (default).
CMATRIX
Employ CMATRIX macro to calculate capacitance.
Seqslv
Solver for structural analysis:
SPARSE
Sparse direct equation solver (default).
PCG
Pre-conditioned Conjugate Gradient iterative equation solver.
Eeqslv
Solver for electrostatic analysis:
SPARSE
Sparse direct equation solver (default).
JCG
Jacobi Conjugate Gradient iterative equation solver.
ICCG
Incomplete Cholesky Conjugate Gradient iterative equation solver.
Notes
This command prepares and runs multiple finite element solutions on the Structural domain and the Elec-
trostatic domain of a model to collect sample points of data for ROM response curve fitting. The command
requires a model database [RMANL] and two Physics Files (Structural domain, titled "STRU" and an Electro-
static domain, titled "ELEC"; see PHYSICS command). Also required is a complete ROM database generated
from the ROM Tools. The Cap = CHARGE method is preferred when capacitance to "infinity" is not required.
Capacitance conductor pairs are defined by the RMCAP command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Required Input
Model Database filename.db
ROM Database jobname.rom, jobname.evx, jobname.evy, jobname.evz
Output Files
Strain energy and capacitance data files jobname_ijk.dec (mode i, j, k).
1439
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RMSMPLE
Menu Paths
Main Menu>ROM Tool>Sample Pt Gen>Compute Points
RMUSE, Option, Usefil
Activates ROM use pass for ROM elements.
REDUCED ORDER MODELING: Use Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Option
Type of data to be plotted. Valid types are:
1 or "ON"
Activates ROM use pass.
0 or "OFF"
Deactivates ROM use pass.
Usefil
Name of the reduced displacement file (.rdsp) created by the ROM Use Pass (required field only for
the Expansion Pass).
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>ExpansionPass
Main Menu>Preprocessor>Loads>Analysis Type>New Analysis
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>ExpansionPass
Main Menu>Solution>Analysis Type>New Analysis
RMXPORT
Exports ROM model to external VHDL-AMS simulator.
REDUCED ORDER MODELING: Generation Pass
MP <> <> <> <> <> <> <> <> <> <> PP <> EME <>
Notes
Use this command to generate all files necessary to run the ROM analysis in an external VHDL-AMS Simulator.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1440
RMUSE
Output Files
VHDL files: Initial.vhd, S_ams_ijk.vhd, Cxxx_ams_ijk.vhd, transducer.vhd.
Menu Paths
Main Menu>ROM Tool>Export>VHDL-AMS
ROCK, CGX, CGY, CGZ, OMX, OMY, OMZ
Specifies a rocking response spectrum.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
CGX, CGY, CGZ
Global Cartesian X, Y, and Z location of center of rotation about which rocking occurs.
OMX, OMY, OMZ
Global Cartesian angular velocity components associated with the rocking.
Notes
Specifies a rocking response spectrum effect in the spectrum (ANTYPE,SPECTR) analysis.
The excitation direction with rocking included is not normalized to one; rather, it scales the spectrum. For
more information, see Participation Factors and Mode Coefficients.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Settings
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Settings
ROSE, SIGNIF, Label, TD
Specifies the Rosenblueth mode combination method.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SIGNIF
Combine only those modes whose significance level exceeds the SIGNIF threshold. For single point,
multipoint, or DDAM response (SPOPT, SPRS, MPRS, or DDAM), the significance level of a mode is defined
as the mode coefficient of the mode, divided by the maximum mode coefficient of all modes. Any mode
whose significance level is less than SIGNIF is considered insignificant and does not contribute to the
mode combinations. The higher the SIGNIF threshold, the fewer the number of modes combined. SIGNIF
defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0.
1441
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
ROSE
Label
Label identifying the combined mode solution output.
DISP
Displacement solution (default). Displacements, stresses, forces, etc., are available.
VELO
Velocity solution. Velocities, "stress velocities," "force velocities," etc., are available.
ACEL
Acceleration solution. Accelerations, "stress accelerations," "force accelerations," etc. are available.
TD
Time duration for earthquake or shock spectrum. TD defaults to 10.
Notes
For more information on spectrum analysis combination methods, see Combination of Modes
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Menu paths generated with the final build.
RPOLY, NSIDES, LSIDE, MAJRAD, MINRAD
Creates a regular polygonal area centered about the working plane origin.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSIDES
Number of sides in the regular polygon. Must be greater than 2.
LSIDE
Length of each side of the regular polygon.
MAJRAD
Radius of the major (or circumscribed) circle of the polygon. Not used if LSIDE is input.
MINRAD
Radius of the minor (or inscribed) circle of the polygon. Not used if LSIDE or MAJRAD is input.
Notes
Defines a regular polygonal area on the working plane. The polygon will be centered about the working
plane origin, with the first keypoint defined at = 0. The area will be defined with NSIDES keypoints and
NSIDES lines. See the RPR4 and POLY commands for other ways to create polygons.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>By Circumscr Rad
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1442
RPOLY
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>By Inscribed Rad
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>By Side Length
RPR4, NSIDES, XCENTER, YCENTER, RADIUS, THETA, DEPTH
Creates a regular polygonal area or prism volume anywhere on the working plane.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NSIDES
The number of sides in the polygon or prism face. Must be greater than 2.
XCENTER, YCENTER
Working plane X and Y coordinates of the center of the polygon or prism face.
RADIUS
Distance (major radius) from the center to a vertex of the polygon or prism face (where the first keypoint
is defined).
THETA
Angle (in degrees) from the working plane X-axis to the vertex of the polygon or prism face where the
first keypoint is defined. Used to orient the polygon or prism face. Defaults to zero.
DEPTH
The perpendicular distance (either positive or negative based on the working plane Z direction) from
the working plane representing the depth of the prism. If DEPTH = 0 (default), a polygonal area is created
on the working plane.
Notes
Defines a regular polygonal area anywhere on the working plane or prism volume with one face anywhere
on the working plane. The top and bottom faces of the prism are polygonal areas. See the RPOLY, POLY,
RPRISM, and PRISM commands for other ways to create polygons and prisms.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>Hexagon
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>Octagon
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>Pentagon
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>Septagon
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>Square
Main Menu>Preprocessor>Modeling>Create>Areas>Polygon>Triangle
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>Hexagonal
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>Octagonal
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>Pentagonal
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>Septagonal
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>Square
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>Triangular
1443
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RPR4
RPRISM, Z1, Z2, NSIDES, LSIDE, MAJRAD, MINRAD
Creates a regular prism volume centered about the working plane origin.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Z1, Z2
Working plane Z coordinates of the prism.
NSIDES
Number of sides in the polygon defining the top and bottom faces of the prism. Must be greater than
2.
LSIDE
Length of each side of the polygon defining the top and bottom faces of the prism.
MAJRAD
Radius of the major (or circumscribed) circle of the polygon defining the top and bottom faces of the
prism. Not used if LSIDE is input.
MINRAD
Radius of the minor (or inscribed circle) of the polygon defining the top and bottom faces of the prism.
Not used if LSIDE or MAJRAD is input.
Notes
Defines a regular prism volume centered about the working plane origin. The prism must have a spatial
volume greater than zero. (i.e., this volume primitive command cannot be used to create a degenerate
volume as a means of creating an area.) The top and bottom faces are polygonal areas that are parallel to
the working plane but neither face need be coplanar with (i.e., "on") the working plane. The first keypoint
defined for each face is at = 0. See the RPR4 and PRISM commands for other ways to create prisms.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>By Circumscr Rad
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>By Inscribed Rad
Main Menu>Preprocessor>Modeling>Create>Volumes>Prism>By Side Length
RPSD, IR, IA, IB, ITYPE, DATUM, Name
Computes response power spectral density (PSD).
POST26: Special Purpose
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previous variable, the previous variable will be overwritten with this result.
IA, IB
Reference numbers of the two variables to be operated on. If only one, leave IB blank.
ITYPE
Defines the type of response PSD to be calculated:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1444
RPRISM
0,1
Displacement (default).
2
Velocity.
3
Acceleration.
DATUM
Defines the reference with respect to which response PSD is to be calculated:
1
Absolute value.
2
Relative to base (default).
Name
Thirty-two character name identifying variable on listings and displays. Embedded blanks are compressed
for output.
Notes
This command computes response power spectral density (PSD) for the variables referenced by the reference
numbers IA and IB. The variable referred by IR will contain the response PSD. You must issue the STORE,PSD
command first; File.PSD must be available for the calculations to occur.
The response power spectral density you calculate will include all modes, including those corresponding to
a significance level smaller than the threshold ( PSDCOM, SIGNIF). See POST26 - Response Power Spectral
Density in the Theory Reference for the Mechanical APDL and Mechanical Applications for more information
on these equations.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>TimeHist Postpro>Calc Resp PSD
RSFIT, RSlab, Slab, Name, Rmod, Ytrans, Yval, Xfilt, CONF
Fit a response surface for an output parameter in a solution set.
PROBABILISTIC: Response Surfaces
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSlab
Response surface set label. Identifies the response surface results for later postprocessing. This label can
be used as Rlab for postprocessing response surface results, and for evaluating the fit results with
RSPLOT and RSPRNT. You must have separate identification labels for the solution sets and for the re-
sponse surface sets, because you can perform a response surface fit based on Monte Carlo results. For
this case, the original Monte Carlo results are identified by the solution set label provided in the PDEXE
command and the results generated with the fitted response surfaces are identified by the response
1445
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RSFIT
surface set label (RSlab) provided here. RSlab must not contain blanks. Maximum length of this field
is 16 characters; if this field contains more than 16 characters, it will be truncated to 16 characters.
Slab
Solution set label that identifies the probabilistic analysis containing the results to be fitted. This solution
set label is defined with the PDEXE command.
Name
Parameter name. The parameter must have been previously defined as a random output parameter with
the PDVAR command. The result values of this parameter (stored in the solution set identified by Slab)
are fitted as a function of all random input variables.
Rmod
Regression model to use for the evaluation of the response surface.
LIN
Regression model with linear terms only.
QUAD
Regression model with linear and pure quadratic terms (no cross-terms).
QUAX
Regression model with linear and all quadratic terms (including cross-terms). (Default.)
Ytrans
Option for the transformation type applied to the output parameter identified with Name.
NONE
The values of the output parameter Name are not transformed (default).
EXP
The values of the output parameter Name are transformed according to Y* = exp(Y). The transformed
values Y* are used for the fitting process.
LOGA
The values of the output parameter Name are transformed according to Y* = log
a
(Y), where the basis
a is specified by Yval. The transformed values Y* are used for the fitting process.
LOGN
The values of the output parameter Name are transformed according to Y* = In(Y). The transformed
values Y* are used for the fitting process.
LOG10
The values of the output parameter Name are transformed according to Y* = log
10
(Y). The transformed
values Y* are used for the fitting process.
SQRT
The values of the output parameter Name are transformed according to Y* =
y
. The transformed
values Y* are used for the fitting process.
POW
The values of the output parameter Name are transformed according to Y* = Y
a
, where the exponent
a is specified by Yval. The transformed values Y* are used for the fitting process.
BOX
The values of the output parameter Name are transformed according to the Box-Cox transformation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1446
RSFIT
Y* =
Y i-1
i
for
i
0
Y* = ln(Y) for
i
= 0
The parameter
i
is automatically determined within the interval [-2, 2]. The step length for this in-
terval search is given in Yval. The transformed values Y* are used for the fitting process.
Yval
Value needed for Ytrans = LOGA, Ytrans = POW, and Ytrans = BOX. Ignored for all other options
of Ytrans. For Ytrans = LOGA and Ytrans = POW, the default value is Yval = 1.0. For Ytrans =
BOX, the default value is Yval = 0.01.
Xfilt
Option that specifies if irrelevant terms of the regression model should be filtered out.
NONE
Regression term filtering is not done. A full regression model as specified with the Rmod option is
evaluated.
FSR
Use the "forward-stepwise-regression" to automatically filter out individual and irrelevant terms of
the regression model. If irrelevant terms are sorted out this leads to a smaller number of coefficients
to be determined in the regression analysis, which then leads to increased accuracy of the coefficients
of the remaining terms in the regression model (default).
CONF
Confidence level that is used to filter the terms of the regression model if the Xfilt = FSR option has
been selected. The value of CONF must be between 0.0 and 1.0 (default is 0.95). The higher this value,
the more terms will be filtered out. Consequently, higher CONF values lead to a fewer terms in the re-
gression model. Likewise, lower CONF values lead to more terms being included in the regression
model.
Command Default
Rmod = QUAX, Ytrans = NONE, Xfilt = FSR, CONF = 0.95, Yval as explained above.
Notes
Evaluates a response surface for a random output parameter Name. The result values for the parameter
Name as stored in the solution set Slab are fitted as a function of all random input parameters. There can
be only one response surface set for each solution set. A response surface set can include the fitted response
surfaces of one or more output parameters.
Note
If the RSFIT command is used for an existing response surface set for which Monte Carlo simula-
tions have already been generated using the RSSIMS command then these Monte Carlo samples
are deleted by the RSFIT command. In this case they need to be generated again using the
RSSIMS command. It is necessary to delete the Monte Carlo simulations in order to make sure
that the samples, their statistics and correlations are consistent with the response surfaces.
1447
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RSFIT
Menu Paths
Main Menu>Prob Design>Response Surf>Fit Resp Surf
RSOPT, Opt, Filename, Ext, Dir
Creates or loads the radiosity mapping data file for SURF251 or SURF252 element types.
SOLUTION: Radiosity
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Opt
File option:
SAVE
Write the radiosity mapping data to a file. (Default)
LOAD
Read in the specified mapping data file.
Fname
File name for radiosity mapping data file. Defaults to Jobname.
Ext
Filename extension for radiosity mapping data file (default = .rsm).
Dir
Directory path for radiosity mapping data file. If you do not specify a directory path, it will default to
your working directory.
Notes
Use this command to manually create or load a radiosity mapping data file. This command is useful if you
want to create the mapping data file without issuing SAVE or CDWRITE, or if you want to specify that the
file be located in a directory other than your working directory. Also use this command to manually load an
existing mapping data file during a restart.
Menu Paths
This command cannot be accessed from a menu.
RSPEED, MIPS, SMFLOP, VMFLOP
Supplies system performance information for use in time estimates.
RUNSTATS: Run Statistics Estimator
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
MIPS
MIPS rating of computer (defaults to 4). This value is ignored if SMFLOP is specified.
SMFLOP
Scalar MFLOPS rating of computer (defaults to MIPS/4).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1448
RSOPT
VMFLOP
Vector MFLOPS rating of computer (Defaults to MIPS/2).
Notes
Supplies system performance information to the program for its use in estimating run times [RTIMST].
Normally this command is invoked through the SETSPEED macro as executed by the ANSYS installation
process.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Run-Time Stats>System Settings
RSPLIT, Option, Label, Name1, Name2, Name3, Name4, Name5, Name6, Name7, Name8, Name9, Name10,
Name11, Name12, Name13, Name14, Name15, Name16
Creates one or more results file(s) from the current results file based on subsets of elements.
POST1: Special Purpose
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Option
Specify which results to include for the subset of elements.
ALL
Write all nodal and element results based on the subset of elements (default).
EXT
Write only the nodal and element results that are on the exterior surface of the element subset. The
results data will be averaged as in PowerGraphics (see AVRES) when this results file is brought into
POST1. Only valid for solid elements.
Label
Define where the element subset is coming from.
ALL
Use all selected element components (CMSEL) (default).
ESEL
Use the currently selected (ESEL) set of elements. Name1 defines the results file name.
LIST
Use Name1 to Name16 to list the element component and/or assembly names (that contain element
components).
Name1, Name2, Name3, . . . , Name16
Up to 16 element component and/or assembly names (that contain element components).
Command Default
Write all data available for the element subset.
1449
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RSPLIT
Notes
Results files will be named based on the element component or assembly name, e.g., Cname.rst, except
for the ESEL option, for which you must specify the results file name (no extension) using the Name1 field.
Note that the .rst filename will be written in all uppercase letters (CNAME.rst) (unless using the ESEL
option); when you read the file, you must specify the filename using all uppercase letters (i.e., file,CNAME).
You may repeat the RSPLIT command as often as needed. All results sets on the results file are processed.
Use /AUX3 to produce a results file with just a subset of the results sets.
Use INRES to limit the results data written to the results files.
The subset geometry is also written so that no database file is required to postprocess the subset results
files. You must not resume any database when postprocessing one of these results files. The input results
file must have geometry written to it (i.e., do not use /CONFIG,NORSTGM,1).
Applied forces and reaction forces are not apportioned if their nodes are shared by multiple element subsets.
Their full values are written to each results file.
Each results file renumbers its nodes and elements starting with 1.
This feature is useful when working with large models. For more information on the advantages and uses
of the RSPLIT command, see Splitting Large Results Files in the Basic Analysis Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
RSPLOT, RSlab, YName, X1Name, X2Name, Type, NPTS, PLOW, PUP
Plot a response surface.
PROBABILISTIC: Response Surfaces
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSlab
Response Surface set label. Identifies the response surfaces generated by the RSFIT command.
YName
Parameter name. The parameter must have been previously defined as a random output parameter with
the PDVAR command.
X1Name
Parameter name. The parameter must have been previously defined as a random input variable with
the PDVAR command.
X2Name
Parameter name. The parameter must have been previously defined as a random input variable with
the PDVAR command. X2Name must be different than X1Name.
Type
Type of the response surface visualization.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1450
RSPLOT
2D
2-D contour plot.
3D
3-D surface plot.
NPTS
Number of grid points for both the X1-axis and the X2-axis. The grid points are used for the evaluation
of the response surface. The number must be between 1 and 500. Defaults to 20. If NPTS = 0 or greater
than 500, then a value of 20 is used.
PLOW
Lower probability level used to determine the lower boundary (plotting range) of the curve in case the
random input variable does not have a minimum value (such as Gauss). This probability must be between
0.0 and 1.0. Defaults to 0.0025.
PUP
Upper probability level used to determine the upper boundary of the curve. This probability must be
between 0.0 and 1.0. Defaults to 0.9975.
Command Default
NPTS = 20, PLOW = 0.0025 (applied if distribution type has no minimum), PUP = 0.9975 (applied if distribution
type has no maximum)
Notes
Plots the response surface of an output parameter YName as a function of two input parameters X1Name
and X2Name.
If PLOW is left blank, then a minimum value of the distribution is used for plotting, provided it exists (for
example, uniform distribution). If the distribution type has no minimum value (for example, Gaussian distri-
bution), then the default value is used to determine the lower plotting range value. The same is true for the
maximum value if PUP is left blank.
In addition to the response surface, the sampling points that are fitted by the response surface are also
plotted by this command. However, sampling points falling outside of the plotting range defined by the
PLOW and PUP fields will not be shown in the plot.
Menu Paths
Main Menu>Prob Design>Response Surf>Plt Resp Surf
RSPRNT, RSlab, YName, Xout
Print a response surface.
PROBABILISTIC: Response Surfaces
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSlab
Response Surface set label. Identifies the response surfaces generated by the RSFIT command.
1451
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RSPRNT
YName
Parameter name. The parameter must have been previously defined as a random output parameter with
the PDVAR command. Identifies the parameter for which a response surface has been generated by the
RSFIT command.
Xout
An option if an extended print-out of more feedback about goodness-of-fit and the details of the regres-
sion analysis of the response surface is requested.
No
Use the standard print-out (default).
Yes
Use the extended print-out.
Notes
Prints the results and details of a response surface analysis generated by the RSFIT command. For the specified
output parameter Yname, the fitting details such as the individual terms of the response surface model and
their corresponding coefficients are listed. The command also produces a comparison of the original values
of Yname used for the fitting process and the approximate values derived from the fitting, and some
goodness of fit measures.
If Xout = Yes, then more information about the regression analysis of the response surface will be printed.
For example, the confidence intervals on the regression coefficients and the correlation between the regression
coefficients among others.
Menu Paths
Main Menu>Prob Design>Response Surf>Prn Resp Surf
RSSIMS, RSlab, NSIM, Seed
Performs Monte Carlo simulations on response surface(s).
PROBABILISTIC: Response Surfaces
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSlab
Response Surface set label. Identifies the response surfaces generated by the RSFIT command.
NSIM
Number of simulation loops on the response surfaces that will be generated for all random output
parameters. If the RSSIMS command is issued multiple times using the same response surface set label
the NSIM Monte Carlo simulations is appended to previous ones. The default value for NSIM is 10,000.
Seed
Seed value label. Random number generators require a seed value that is used to calculate the next
random number. After each random number generation finishes, the seed value is updated and is used
again to calculate the next random number. By default ANSYS initializes the seed value with the system
time (one time only) when the ANSYS session started.
CONT
Continues updating using the derived seed value (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1452
RSSIMS
TIME
Initializes the seed value with the system time. You can use this if you want the seed value set to a
specific value for one analysis and then you want to continue with a "random" seed in the next
analysis. It is not recommended to "randomize" the seed value with the Seed = TIME option for
multiple analyses. If the Monte Carlo simulations requested with this command will be appended to
previously existing simulations, then the Seed option is ignored and Seed = CONT is used.
INIT
Initializes the seed value using 123457. This value is a typical recommendation used very often in
literature. This option leads to identical random numbers for all random input variables when the
exact analysis will be repeated, making it useful for benchmarking and validation purposes (where
identical random numbers are desired). If the Monte Carlo simulations requested with this command
will be appended to previously existing simulations, then the Seed option is ignored and Seed =
CONT is used.
Value
Uses the specified (positive) value for the initialization of the seed value. This option has the same
effect as Seed = INIT, except you can chose an arbitrary (positive) number for the initialization. If
the Monte Carlo simulations requested with this command will be appended to previously existing
simulations, then the Seed option is ignored and Seed = CONT is used.
Command Default
NSIM = 10,000, Seed = CONT
Notes
Generate the Monte Carlo simulations on the response surfaces that are included in a response surface set.
Simulations are evaluated only for the output parameters that have been fitted in a response surface set
using the RSFIT command.
If the RSSIMS command is issued multiple times using the same response surface label the probabilistic
design system appends the samples generated here to the previous ones. This way you can start with a
moderate NSIM number and add more samples if the probabilistic results are not accurate enough.
Menu Paths
Main Menu>Prob Design>Response Surf>RS Simulation
RSTAT
Prints the FE model statistics of the model.
RUNSTATS: Run Statistics Estimator
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
Prints the finite element model statistics of the model currently in the database. The maximum node and
element number used are displayed, as well as the number of nodes and elements selected.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1453
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
RSTAT
Menu Paths
Main Menu>Run-Time Stats>Individual Stats
RSTMAC, File1, Lstep1, Sbstep1, File2, Lstep2, Sbstep2, TolerN, MacLim, Cname, KeyPrint
Calculates modal assurance criterion (MAC) and matches nodal solutions from two results files.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
File1
Name (32 characters maximum) corresponding to first Jobname (.db and .rst files).
Lstep1
Load step number of the results to be read in File1.
N
Reads load step N. Defaults to 1.
Sbstep1
Substep number of the results to be read in File1.
N
Reads substep N.
All
Reads all substeps. This value is the default.
File2
Name (32 characters maximum) corresponding to second Jobname (.rst files).
Lstep2
Load step number of the results to be read in File2.
N
Reads load step N. Defaults to 1.
Sbstep2
Substep number of the results to be read in File2.
N
Reads substep N.
All
Reads all substeps. This value is the default.
TolerN
Tolerance for nodes matching, where TolerN > 0. The default value is 0.01 (1%).
MacLim
Smallest acceptable MAC value. Must be
0 and
8, default = 8
Ro = Outer radius of the tube
N = Number of cells along the circumference; N
8, default = 8
T = Number of divisions through the radius;
default = 2
This subtype is similar to type PIPE. However, ele-
ments using PIPE account for internal or external
pressures, whereas elements using CTUBE do not.
Type: BEAM, Subtype: I
W1
W2
t1
t2
t3
z
y
W3
Type: BEAM, Subtype: CHAN
W1
W2
W3
t1
t2
t3
z
y
Data to provide in the value fields:
Data to provide in the value fields:
W1, W2, W3, t1, t2, t3
W1, W2, W3, t1, t2, t3
where
where
W1, W2 = Lengths of the flanges
W1, W2 = Width of the top and bottom flanges
W3 = Overall depth
W3 = Overall depth
t1, t2 = Flange thicknesses
t1, t2 = Flange thicknesses
t3 = Web thicknesses
t3 = Web thicknesses
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1474
SECDATA
Type: BEAM, Subtype: L Type: BEAM, Subtype: Z
W1
W2
t1
t2
z
y
W1
W2
W3
t1
t2
t3
z
y
Data to provide in the value fields:
W1, W2, t1, t2
where
W1, W2 = Leg lengths
t1, t2 = Leg thicknesses
Data to provide in the value fields:
If W2 is a negative value, the section will be flipped.
W1, W2, W3, t1, t2, t3
where
W1, W2 = Flange lengths
W3 = Overall depth
t1, t2 = Flange thicknesses
t3 = Stem thicknesses
Type: BEAM, Subtype: HATS
W1 W2
W3
W4
t3
t4
t1
t2
t5
z
y
Type: BEAM, Subtype: T
W1
W2
t1
t2
y
z
Data to provide in the value fields:
W1, W2, t1, t2
Data to provide in the value fields:
where
W1, W2, W3, W4, t1, t2, t3, t4, t5
W1 = Flange width
where
W2 = Overall depth
W1, W2 = Width of the brim
t1 = Flange thicknesses
W3 = Width of the top of the hat
t2 = Stem thicknesses
W4 = Overall depth
1475
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
If W2 is a negative value, the section will be flipped. t1, t2 = Thickness of the brim
t3 = Thickness of the top of the hat
t4, t5 = Web thicknesses
Type: BEAM, Subtype: ASEC Type: BEAM, Subtype: HREC
W1
W2
t1
t4
y
z
t2
t3
Arbitrary: User-supplied integrated section
properties instead of basic geometry data.
Data to provide in the value fields:
A, Iyy, Iyz, Izz, Iw, J, CGy, CGz, SHy, SHz
where
A = Area of section
Iyy = Moment of inertia about the y axis
Iyz = Product of inertia
Izz = Moment of inertia about the z axis
Iw = Warping constant
J = Torsional constant
CGy = Y coordinate of centroid
Data to provide in the value fields:
CGz = Z coordinate of centroid
W1, W2, t1, t2, t3, t4
SHy = Y coordinate of shear center
where
SHz = Z coordinate of shear center
W1 = Outer width of the box
SECPLOT cannot display an ASEC plot.
W2 = Outer height of the box
t1, t2, t3, t4 = Wall thickness
Type: BEAM, Subtype: MESH
User-defined mesh.
Data required is created by the SECWRITE
command and is read into ANSYS by the
SECREAD command.
When a user mesh is input via SECREAD, ANSYS
calculates the area, second moments of inertia,
centroid, torsion constant, warping rigidity, and
shear center.
If you redefine a material for a composite cross-
section after creating the section, you must re-
issue the SECREAD command.
Pipes
Type: PIPE
Pipe sections are referenced by the PIPE288, PIPE289, and ELBOW290 elements.
Data to provide in the value fields:
D
o
, T
w
, N
c
, S
s
, N
t
, M
int
, M
ins
, T
ins
where
D
o
= Outside diameter of pipe. Does not include T
ins
.
T
w
= Wall thickness.
N
c
= Number of cells along the circumference. This value must be 8 or greater. The default value is 8.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1476
SECDATA
S
s
= Section number of the shell representing the pipe wall. Valid with ELBOW290 only.
N
t
= Number of cells through the pipe wall. Valid values are 1, 3, 5, 7, and 9. The default value is 1. Cells
are graded such that they are thinner on the inner and outer surfaces. Valid with PIPE288 and PIPE289
only.
M
int
= Material number of fluid inside of the pipe. The default value is 0 (no fluid). Valid with PIPE288
and PIPE289 only.
M
ins
= Material number of material external to the pipe (such as insulation, biofouling, or armoring). The
default value is 0 (no external material). Valid with PIPE288 and PIPE289 only.
T
ins
= Thickness of material external to the pipe, such as insulation. The default value is 0 (no external
material). Valid with PIPE288 and PIPE289 only.
External material (M
ins
) adds mass and increases hydraulic diameter, but does not add to stiffness. (M
ins
is
valid for PIPE288 and PIPE289 only.)
Internal fluid
Pipe wall
T
w
Insulation
Radial cell divisions
(N
T
= 3 shown)
D
o
T
ins
Circumferential cell divisions
(N
c
= 8 shown)
External fluid
Taper
Type: TAPER
Taper sections are referenced by BEAM188, and BEAM189 elements. After specifying the tapered section
type ( SECTYPE,,TAPER), issue separate SECDATA commands to define each end of the tapered beam.
1477
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
Data to provide in the value fields:
Sec_IDn, XLOC, YLOC, ZLOC
where
Sec_IDn = Previously defined beam section at ends 1 (I) and 2 (J).
XLOC, YLOC, ZLOC = The location of Sec_IDn in the global Cartesian coordinate system.
General Axisymmetric
Type: AXIS
General axisymmetric sections are referenced by the SOLID272 and SOLID273 elements. Use this command
to locate the axisymmetric axis.
Data to provide in the value fields:
Pattern 1 (two points):
1, x1, y1, z1, x2, y2, z2
Pattern 2 (coordinate system number plus axis [1 = x, 2 = y, 3 = z] ):
2, csys, axis
Pattern 3 (origin plus direction):
3, xO, yO, zO, xdir, ydir, zdir
where xdir, ydir, and zdir are direction cosines.
For general axisymmetric (AXIS) sections using Pattern 2, csys must be a Cartesian coordinate system.
Shells
Type: SHELL
Shell sections are referenced by the SHELL131, SHELL132, SHELL181, SOLID185 Layered Solid, SOLID186
Layered Solid, SOLSH190, SHELL208, SHELL209, and SHELL281 elements.
Data to provide in the value fields:
TK, MAT, THETA, NUMPT, LayerName
where
TK = Thickness of shell layer. Zero thickness (not valid for SHELL131 and SHELL132) indicates a dropped
layer. The sum of all layer thicknesses must be greater than zero. The total thickness can be tapered via
the SECFUNCTION command.
MAT = Material ID for layer (any current-technology material model is available for SHELL181, SOLID185
Layered Solid, SOLID186 Layered Solid, SOLSH190, SHELL208, and SHELL209, including UserMat). MAT is
required for a composite (multi-layered) laminate. For a homogeneous (single-layered) shell, the default
is the MAT command setting. Use the TREF and/or the MP, REFT commands to address multiple reference
temperatures.
THETA = Angle (in degrees) of layer element coordinate system with respect to element coordinate
system (ESYS).
NUMPT = Number of integration points in layer. The GUI permits 1, 3, 5, 7, or 9 points (default = 3).
However a higher odd number may be specified in the command. The integration rule used is Simpson's
Rule. (NUMPT is not used by SHELL131 and SHELL132.)
LayerName = The layer name (up to 72 characters) in an analysis using FiberSIM data (SECTYPE,,SHELL,
FIBERSIM). The layer name is case-sensitive and must match the ply name in the FiberSIM .xml file. (In
a non-FiberSIM analysis, this value serves only as a comment in the input.)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1478
SECDATA
In an analysis using FiberSIM data:
The THETA value is ignored,
The order of the SECDATA commands does not matter as the layer order comes from the .xml file.
Therefore, repeated values of LayerName should not be used.
Every LayerName used on the .xml file should be on a separate SECDATA command.
Issue separate SECDATA commands to define each of the layers in the section.
Pretension
Type: PRETENSION
Pretension sections are referenced by the PRETS179 element.
Data to provide in the value fields:
node, nx, ny, nz
where
node = Pretension node number.
nx = Orientation in global Cartesian x direction.
ny = Orientation in global Cartesian y direction.
nz = Orientation in global Cartesian z direction.
The following usage is typical:
SECTYPE, 1, PRETENSION
SECDATA, 13184, 0.000, 0.000, 1.000
SECMODIF, 1, NAME, example
SLOAD, 1, PL01, TINY, FORC, 100.00 , 1, 2
The PRETENSION section options of SECTYPE and SECDATA are documented mainly to aid in the under-
standing of data written by CDWRITE. ANSYS recommends that you generate pretension sections using
PSMESH.
Joints
Type: JOINT
Joint sections are referenced by MPC184 joint elements.
Data to provide in the value fields:
length1, length2, length3, angle1, angle2, angle3
where
length1-3 = Reference lengths used in the constitutive calculations.
angle1-3 = Reference angles used in the constitutive calculations.
The following table shows the lengths and angles to be specified for different kinds of joints.
1479
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
Reference lengths and angles Subtype Joint
angle1 for x-axis revolute joint
angle3 for z-axis revolute joint
REVO Revolute
angle1 and angle3 UNIV Universal
length1 SLOT Slot
length2 and length3, PINP Point-in-plane
length1 PRIS RevoluteTransla-
tional
length1 and angle1 for x-axis cylindrical joint
length3 and angle3 for z-axis cylindrical joint
CYLI Cylindrical
length2, length3, and angle1 for x-axis planar joint
length1, length2, and angle3 for z-axis planar joint
PLAN Planar
(not used) WELD Weld
(not used) ORIE Orient
(not used) SPHE Spherical
length1, length2, length3, angle1, angle2, angle3 -- De-
pends upon which unconstrained relative degrees of
freedom are available.
GENE General
length3 and angle3 SCRE Screw
The reference length and angle specifications correspond to the free relative degrees of freedom in a joint
element for which constitutive calculations are performed. These values are used when stiffness and/or
damping are specified for the joint elements.
If the reference lengths and angles are not specified, they are calculated from the default or starting config-
uration for the element.
See MPC184 or the individual joint element descriptions for more information on joint element constitutive
calculations.
Reinforcing
Type: REINF
Reinforcing sections are referenced by REINF264 and REINF265 elements.
- - - - - - - - - - - - - - - - - - - - -
Type: REINF, Subtype: DISCRETE
This subtype defines discrete reinforcing fibers with arbitrary orientations. Issue separate SECDATA commands
to define each reinforcing fiber.
Data to provide in the value fields:
where
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1480
SECDATA
MAT, A, PATT, V1, V2, V3, V4, V5
MAT = Material ID for the fiber. (Refer to the REINF264 documentation for available material models.)
The material ID is required for all fibers, and no default for this value is available.
A = Cross-section area of the reinforcing fiber.
PATT = Input pattern code indicating how the location of this fiber is defined with respect to the base
element. Available input patterns are LAYn, EDGo, and BEAM.
V1, V2, V3, V4, V5 = Values to define the location of the reinforcing fiber (depending on the
PATT pattern code used), as shown:
PATT : LAYn
Description: The discrete reinforcing fiber is placed in the middle of a layer in a layered base element. The
orientation of the fiber within the layer is adjustable via offsets with respect to a specified element edge.
Required input:
V1 (or n) -- The number of the layer in the base element on which to apply the reinforcing fiber. The
default value is 1.
V2 (or e) -- The number to indicate the element edge to which the offsets are measured. The default
value is 1.
V3 and V4 (or Y1 and Y2) -- The normalized distances from the fiber to the two ends of the specified
element edge. Valid values for Y1 and Y2 are 0.0 through 1.0. The default value of Y1 is 0.5. The default
value of Y2 is Y1.
When applied to 8-node or 20-node layered solid elements:
Base layer n
I
J
K
L
M
P
O
N
(Y
1
= 1)
(Y
2
= 1)
(Y
1
= 0)
Y
2
(Y
2
= 0)
Y
1
II
JJ
Reinforcing
e = 1
1481
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
Base layer n
I
J
K
L
M
P
O
N
(Y
1
= 1)
(Y
2
= 1)
(Y
1
= 0)
Y
2
(Y
2
= 0)
Y
1
II
JJ
e = 2
R
e
i
n
f
o
r
c
i
n
g
When applied to 4-node or 8-node layered shell elements:
Base layer n
I
J
K
L
(Y
1
= 1)
(Y
2
= 1)
(Y
1
= 0)
Y
2
(Y
2
= 0)
Y
1
II
JJ
Reinforcing
e = 1
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1482
SECDATA
Base layer n
I
J
K
L
(Y
1
= 1)
(Y
2
= 1)
(Y
1
= 0)
Y
2
(Y
2
= 0)
Y
1
II
JJ
e = 2
R
e
i
n
f
o
r
c
i
n
g
PATT : EDGo
Description: The orientation of the discrete reinforcing fiber is similar to one of the specified element edges.
The fiber orientation can be further adjusted via offsets with respect to the specified element edge.
Required input:
V1 (or e) -- The number to indicate the element edge to which the offsets are measured. The default
value is 1.
V2 and V3 (or Y1 and Z1) -- The normalized distances from the fiber to the first end of the specified
element edge. Valid values for Y1 and Z1 are 0.0 through 1.0. The default value for Y1 and Z1 is 0.5.
V4 and V5 (or Y2 and Z2) The normalized distances from the fiber to the second end of the specified
element edge. Value values for Y2 and Z2 are 0.0 through 1.0. The default value for Y2 is Y1, and the
default value for Z2 is Z1.
If the base element is a beam or link, ANSYS ignores values V2 through V5 and instead places
the reinforcing in the center of the beam or link.
When applied to 8-node or 20-node solid elements:
1483
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
I
J
K
L
M
P
O
N
II
e = 1
Y
1
Z
1
(Z
1
= 1)
(Y
1
= 0)
(Z
1
= 0)
Y
2
Z
2
JJ
(Y
2
= 0)
(Z
2
= 0)
(Y
2
= 1)
(Z
2
= 1)
e = 2
I
J
K
L
M
P
O
N
II
Y
1
Z
1
(Z
1
= 1)
(Y
1
= 0)
(Z
1
= 0)
Y
2
Z
2
JJ
(Y
2
= 0)
(Z
2
= 0)
(Y
2
= 1)
(Z
2
= 1)
(Y
1
= 1)
R
e
i
n
f
o
r
c
i
n
g
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1484
SECDATA
e = 3
I
J
K
L
M
P
O
N
II
Y
1
(Z
1
= 1)
(Y
1
= 0)
(Z
1
= 0)
JJ
(Y
2
= 0)
(Z
2
= 0)
(Y
2
= 1)
(Z
2
= 1)
(Y
1
= 1)
Y
2
Z
2
Z
1
When applied to tetrahedral elements:
L
K
J
I
(Z
1
= 1)
(Z
2
= 1)
(Y
1
= 1)
(Y
2
= 1)
(Y
2
= 0)
(Z
2
= 0)
(Y
1
= 0)
(Z
1
= 0)
Y
1
Z
1
Y
2
Z
2
II
JJ
e = 1
1485
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
L
K
J
I
(Z
1
= 1)
(Z
2
= 1)
(Y
2
= 1)
(Y
1
= 1)
(Y
2
= 0)
(Z
2
= 0)
(Y
1
= 0)
Z
1
Y
2
Z
2
II
JJ
Y
1
e = 2
L
K
J
I
(Z
2
= 1)
(Y
2
= 1)
(Y
1
= 1)
(Y
2
= 0)
(Z
2
= 0)
(Y
1
= 0)
(Z
1
= 0)
Y
2
Z
2
II
JJ
Y
1
Z
1
e = 3
When applied to 3-D shell elements:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1486
SECDATA
I
J
K
L (Y
1
= 1)
(Y
2
= 1)
(Y
1
= 0)
Y
2
(Y
2
= 0)
Y
1
II
JJ
Reinforcing
(Z
1
= 0)
(Z
1
= 1)
Z
1
Z
2
= Z
1
e = 1
I
J
K
L
(Y
1
= 1)
(Y
2
= 1)
(Y
1
= 0)
Y
2
(Y
2
= 0)
Y
1
II
JJ
R
e
i
n
f
o
r
c
i
n
g
Z
1
Z
2
= Z
1
Z
1
= 0
e = 2
When applied to beam or link elements:
1487
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
e = 1
II
JJ
I
J
PATT : BEAM
Description: Use this specialized input pattern for defining reinforcing in regular constant and tapered
beams.
Required input:
V1 and V2 (or Y1 and Z1) -- Y and Z offsets with respect to the section origin in the first beam section
referred by the base beam element. The default value for Y1 and Z1 is 0.0.
V3 and V4 (or Y2 and Z2) -- Y and Z offsets with respect to the section origin in the second beam section
referred by the base beam element. The default value for Y2 is Y1, and the default value for Z2 is Z1.
(Because V3 and V4 values apply only to tapered beams, ANSYS ignores them if the base beam has a
constant section.)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1488
SECDATA
II
JJ
I
J
y
z
y
z
(y
2
, z
2
)
(y
1
, z
1
)
Base beam
Beam section #2
Beam section #1
R
e
i
n
f
o
r
c
i
n
g
- - - - - - - - - - - - - - - - - - - - -
Type: REINF, Subtype: SMEAR
This subtype defines reinforcing fibers in homogenous layered forms. Issue separate SECDATA commands
to define each of the layers of reinforcing fibers with uniform cross-section area and spacing.
Data to provide in the value fields:
MAT, A, S, KCN, THETA, PATT, V1, V2
where
MAT = Material ID for layer. (Refer to the REINF265 documentation for available material models.) The
material ID is required for all layers, and no default for this value is available.
A = Cross-section area of a single reinforcing fiber.
S = Distance between two adjacent reinforcing fibers.
KCN = Local coordinate system reference number for this layer. (See the LOCAL command documentation
for more information about local coordinate systems.) If this value is not provided, the layer adopts a
default layer reference system. (See the REINF265 documentation for information about default layer
systems.)
1489
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
THETA = Angle (in degrees) of the final layer coordinate system with respect to the default layer system
or the layer system specified in the KCN field.
PATT = Input pattern code indicating how the location of this layer is defined with respect to the base
element. Available input patterns are LAYn and ELEf.
V1, V2 = Values to define the location of the reinforcing layer, as shown:
PATT : LAYn
Description: The smeared reinforcing layer is placed in the middle of a layer in a layered base element.
Required input: V1 (or n) -- The number of the layer in the base element on which to apply the reinforcing
layer. The default value is 1.
When applied to layered solid elements:
Base layer n
I
II
J
K
KK
L
LL
JJ
M
P
O
N
Reinforcement
When applied to layered shells:
I
II
J
K
KK
L
LL
JJ
Base layer n
Reinforcement
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1490
SECDATA
PATT : ELEf
Description: The smeared reinforcing layer is oriented parallel to one of three adjacent element faces.
Required input:
V1 (or f) -- The number to indicate the base element face. The default value is 1.
V2 (or d) -- The normalized distance from the layer to the specified base element face. Valid values for
d are 0.0 through 1.0. The default value is 0.5.
When applied to 8-node or 20-node solid elements:
I (d = 0)
II
J (d = 1)
K
KK
L
LL
JJ
M
P
O
d
N
f = 2
I
II
J (d = 0)
K
KK
L
LL
JJ
M
P
O
d
N (d = 1)
f = 1
I
II
J (d = 0)
K (d = 1)
KK
L
LL
JJ
M
P
O
N
f = 3
d
When applied to tetrahedral elements:
1491
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECDATA
d
JJ
KK, LL
K
J (d = 1)
I
II
L (d = 0)
f = 2
d
JJ
KK, LL
K
J (d = 0)
I
II
L (d = 1)
f = 1
d JJ
KK, LL
K (d = 1)
J (d = 0)
I
II
L
f = 3
When applied to 3-D shell elements:
L
K
J
I
LL
JJ
KK
II
d = 0
d = 1
d
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1492
SECDATA
Menu Paths
Main Menu>Preprocessor>Sections>Axis>Add
Main Menu>Preprocessor>Sections>Axis>Edit
Main Menu>Preprocessor>Sections>Beam>Common Sections
Main Menu>Preprocessor>Sections>Beam>Taper Sections>By Picked Nodes
Main Menu>Preprocessor>Sections>Beam>Taper Sections>By XYZ Location
Main Menu>Preprocessor>Sections>Joints>Add / Edit
Main Menu>Preprocessor>Sections>Pipe>Add
Main Menu>Preprocessor>Sections>Pipe>Edit
Main Menu>Preprocessor>Sections>Reinforcing>Add / Edit
Main Menu>Preprocessor>Sections>Shell>Lay-up>Add / Edit
SECFUNCTION, TABLE, KCN
Specifies shell section thickness as a tabular function.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
TABLE
Table name or array parameter reference for specifying thickness.
KCN
Local coordinate system reference number for this tabular function evaluation.
Notes
The SECFUNCTION command is associated with the section most recently defined via the SECTYPE command.
A table (TABLE) can define tabular thickness as a function of coordinates. Alternately, you can use an array
parameter (indexed by node number) that expresses the function to be mapped. (For example, func (17)
should be the desired shell thickness at node 17.) To specify a table, enclose the table or array name in
percent signs (%) (SECFUNCTION,%tablename%). Use the *DIM command to define a table.
The table or array defines the total shell thickness at any point in space. In multilayered sections, the total
thickness and each layer thickness are scaled accordingly.
The Function Tool is a convenient way to define your thickness tables. For more information, see "Using the
Function Tool" in the Basic Analysis Guide.
If you do not specify a local coordinate system (KCN), ANSYS interprets the table in global XYZ coordinates.
When KCN = NODE, ANSYS interprets TABLE as an array parameter (indexed by node number) that expresses
the function to be mapped. For information about local coordinate systems, see the documentation for the
LOCAL command.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Lay-up>Add / Edit
1493
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECFUNCTION
SECJOINT, Kywrd, Val1, Val2, Val3, Val4, Val5, Val6
Defines local coordinate systems at joint element nodes and other data for joint elements.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Kywrd
Keyword that indicates the type of joint element data being defined.
LSYS or blank
Define local coordinate systems at the nodes that form the MPC184 joint element.
RDOF
Define the relative degrees of freedom to be fixed for an MPC184-General joint element.
PITC
Define the pitch of an MPC184-Screw joint element.
FRIC
Define the geometric quantities required for Coulomb frictional behavior in the MPC184-Revolute
or MPC184-Translational joint element.
Val1, Val2, Val3, Val4, Val5, Val6
The meaning of Val1 through Val6 changes, depending on the value of Kywrd.
If Kywrd = LSYS (or blank), Val1 and Val2 are Identifiers of the local coordinate systems at nodes I
and J, respectively, of the joint element. Val3 through Val6 are not used.
If Kywrd = RDOF, Val1 through Val6 are the relative degrees of freedom to be fixed for a general
joint element. Input 1 for UX, 2 for UY, 3 for UZ, 4 for ROTX, 5 for ROTY, and 6 for ROTZ. You may input
the DOFs in any order.
If Kywrd = PITC, Val1 is the pitch of the screw joint element; pitch is defined as the ratio of relative
axial displacement (length units) to relative rotation (in radians). Val2 through Val6 are not used.
If Kywrd = FRIC, Val1 through Val3 are defined as follows.
For Revolute Joint:
Val1 = outer radius
Val2 = inner radius
Val3 = effective length
For Translational Joint:
Val1 = effective length
Val2 = effective radius
Notes
Use this command to define additional section data for MPC184 joint elements. To overwrite the current
values, issue another SECJOINT command with the same Kywrd value. The data input on this command is
interpreted based on the most recently issued SECTYPE command.
Menu Paths
Main Menu>Preprocessor>Sections>Joints>Add / Edit
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1494
SECJOINT
/SECLIB, Option, Path
Sets the default section library path for the SECREAD command.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Option
READ
Sets the read path (default).
STATUS
Reports the current section library path setting to the Jobname.LOG file.
Path
Defines the directory path from which to read section library files.
Notes
When the SECREAD command is issued without a directory path, the command searches for a section library
in the following order:
The user's home directory
The current working directory
The path specified by the /SECLIB command
Menu Paths
Main Menu>Preprocessor>Sections>Section Library>Library Path
SECLOCK, dof, MINVALUE, MAXVALUE, dof, MINVALUE, MAXVALUE, dof, MINVALUE, MAXVALUE
Specifies locks on the components of relative motion in a joint element.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
dof
Local degree of freedom to be locked.
MINVALUE
Low end of the range of allowed movement for the specified DOF.
MAXVALUE
High end of the range of allowed movement for the specified DOF.
Notes
Specify up to three DOFs to be locked. Locks are activated when the limit values are reached, and further
motion in that DOF is frozen. If necessary, you may repeat the command.
1495
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECLOCK
Menu Paths
Main Menu>Preprocessor>Sections>Joints>Add / Edit
SECMODIF, SECID, Kywrd
Modifies a pretension section
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
If Kywrd = NORM, command format is SECMODIF,SECID, NORM, NX, NY, NZ, KCN
SECID
Unique section number. This number must already be assigned to a section.
NORM
Keyword specifying that the command will modify the pretension section normal direction.
NX, NY, NZ
Specifies the individual normal components to modify.
KCN
Coordinate system number. This can be either 0 (Global Cartesian), 1 (Global Cylindrical) 2 (Global
Spherical), 4 (Working Plane), 5 (Global Y Axis Cylindrical) or an arbitrary reference number assigned to
a coordinate system.
If Kywrd = NAME, the command format is SECMODIF,SECID, NAME, Name
SECID
Unique section number. This number must already be assigned to a section.
NAME
Action key that instructs the command to change the name of the specified pretension section.
Name
The new name to be assigned to the pretension section.
Notes
The SECMODIF command either modifies the normal for a specified pretension section, or changes the
name of the specified pretension surface.
Menu Paths
Main Menu>Preprocessor>Sections>Pretension>Modify Name
Main Menu>Preprocessor>Sections>Pretension>Modify Normal
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1496
SECMODIF
SECNUM, SECID
Sets the element section attribute pointer.
PREP7: Cross Sections
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
SECID
Defines the section ID number to be assigned to the subsequently-defined elements by the LMESH, E,
or EN commands. Defaults to 1. See SECTYPE for more information about the section ID number.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>Default Attribs
Main Menu>Preprocessor>Modeling>Create>Elements>Elem Attributes
SECOFFSET, Location, OFFSET1, OFFSET2, CG-Y, CG-Z, SH-Y, SH-Z
Defines the section offset for cross sections.
PREP7: Cross Sections
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
Location, OFFSET1, OFFSET2, CG-Y, CG-Z, SH-Y, SH-Z
The location of the nodes in the section. All are dependent on the type. See the "Notes" (p. 1497) section
of this command description for details about these values for the various section types.
Notes
The SECOFFSET command is divided into three types: Beams, Pipes, and Shells.
The offsets defined by the SECOFFSET command are associated with the section most recently defined using
the SECTYPE command. Not all SECOFFSET location values are valid for each subtype.
For the thermal shell elements, SHELL131 and SHELL132, the node offset specified by SECOFFSET is used
in thermal contact analyses. Otherwise, the SECOFFSET command has no effect on the solution for these
elements and is used only for visualization purposes.
Beams
Type: BEAM
Argument data to provide:
Location, OFFSETY, OFFSETZ, CG-Y, CG-Z, SH-Y, SH-Z
Location
CENT --
Beam node will be offset to centroid (default).
SHRC --
Beam node will be offset to shear center.
1497
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECOFFSET
ORIGIN --
Beam node will be offset to origin of the cross section.
USER --
Beam node will be offset to the location specified by the OFFSETY and OFFSETZ arguments.
OFFSETY, OFFSETZ
Values that locate the node with respect to the default origin of the cross section when the Location
argument is set to USER. Valid only when USER is set.
The following figure illustrates the offsets for a channel cross section, and shows the relative locations
of SHRC and CENT.
Figure 5: Offsets for a CHAN Section Subtype
OFFSET-Y OFFSET-Z
SHRC CENT
USER
ORIGIN
z
y
CG-Y, CG-Z, SH-Y, SH-Z
Override the program-calculated centroid and shear centroid locations.
Note
This option should only be used by advanced users modeling composite cross sections.
Pipes
Type: PIPE
Argument data to provide:
OFFSETY, OFFSETZ
OFFSETY, OFFSETZ
Values that locate the node with respect to the center of the pipe.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1498
SECOFFSET
Shells
Type: SHELL
Argument data to provide:
Location, OFFSET
Location
TOP --
Shell node will be offset to top of the section.
MID --
Shell node will be offset to midplane of the section (default).
BOT --
Shell node will be offset to bottom of the section.
USER --
Shell node will be offset to the location specified by the OFFSET argument.
OFFSET
Value that locates the node with respect to the default origin (midplane) of the section when the Loc-
ation argument is set to USER. Valid only when USER is set.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>Common Sections
Main Menu>Preprocessor>Sections>Beam>Custom Sections>Read Sect Mesh
Main Menu>Preprocessor>Sections>Pipe>Add
Main Menu>Preprocessor>Sections>Pipe>Edit
Main Menu>Preprocessor>Sections>Shell>Lay-up>Add / Edit
SECPLOT, SECID, VAL1, VAL2, VAL3
Plots the geometry of a beam, pipe, shell, or reinforcing section to scale.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SECID
The section ID number (as defined via the SECTYPE command).
VAL1, VAL2, VAL3
Values that control the information to be plotted. See the "Notes" (p. 1499) section of this command de-
scription for more information. For clarity, the labels VAL1, VAL2, and VAL3 are renamed according to
the section type.
Notes
The SECPLOT command is valid only for "Beams and Pipes" (p. 1500), "Shells" (p. 1501), and "Reinforcings" (p. 1502).
Custom sections created using FiberSIM data do not support SECPLOT.
1499
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECPLOT
SECPLOT cannot display the plot of an ASEC (arbitrary section) subtype.
Beams and Pipes
Plots the geometry of the beam or pipe section to scale depicting the centroid, shear center, and origin.
SECPLOT also lists various section properties such as Iyy, Iyz, and Izz.
Data to be supplied in the value fields:
MESHKEY
Beam or pipe section mesh display options:
0 --
Does not display beam pipe section mesh.
1 --
Display beam or pipe section mesh.
2 --
Display the section mesh with node numbers.
3 --
Display the section mesh with cell numbers.
4 --
Display the section mesh with material numbers and colors.
5 --
Display the section mesh with material colors only.
Options 2 through 5 do not depict centroid and shear center, nor do they list section properties.
Following is a sample section plot for the beam section type:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1500
SECPLOT
Shells
Plots the layer arrangement of the shell section showing the layer material and orientation.
Data to be supplied in the value fields:
LAYR1, LAYR2
The range of layer numbers to be displayed. If LAYR1 is greater than LAYR2, a reversed order display
is produced. Up to 20 layers may be displayed at the same time. LAYR1 defaults to 1. LAYR2 defaults
to LAYR1 if LAYR1 is input or to the number of layers (or to 19+LAYR1, if smaller) if LAYR1 is not input.
Following is a sample section plot for the shell section type:
1501
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECPLOT
Reinforcings
Plots the arrangement of a reinforcing section within the base element.
Data to be supplied in the value fields:
REINF1, REINF2, OVERLAY
REINF1, REINF2 -- The numerical range of reinforcings to be displayed. The default REINF1 value
is 1. The default REINF2 value is the number of reinforcings.
OVERLAY -- The section ID of the base element within which to display the reinforcing section. The
section appears translucent and the reinforcing section is solid. Valid values are:
SOLID -- Display a translucent solid block over the reinforcing section
SECID -- A number corresponding to a specific section ID of the base element.
If no OVERLAY value is specified, ANSYS displays the reinforcing section only.
Following is a sample section plot for the reinforcing section type:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1502
SECPLOT
For more information about reinforcing, see the documentation for the SECDATA command, and the REINF264
and REINF265 elements.
Menu Paths
Main Menu>Preprocessor>Sections>Beam>Plot Section
Main Menu>Preprocessor>Sections>Pipe>Plot Section
Main Menu>Preprocessor>Sections>Reinforcing>Plot Section
Main Menu>Preprocessor>Sections>Shell>Lay-up>Plot Section
SECREAD, Fname, Ext, --, Option
Reads a custom section library or a user-defined section mesh into ANSYS.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Fname
Section library file name and directory path containing the section library file (248 characters maximum,
including directory). If you do not specify a directory path, it will default to your working directory and
you can use all 248 characters for the file name.
When the SECREAD command is given without a directory path, the command searches for a section
library in the following order:
The user's home directory
The current working directory
The path specified by the /SECLIB command
The file name defaults to Jobname if Fname is left blank.
Ext
Filename extension (8 character maximum).
The extension defaults to SECT if Ext is left blank.
1503
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECREAD
--
Unused field.
Option
LIBRARY
Reads in a library of sections and their associated section data values; the default. A section library
may be created by editing the section-defining portions of the Jobname.LOG file and saving it with
a .SECT suffix.
MESH
Reads in a user mesh section file containing the cell connectivity, cell flags, and nodal coordinates
for the current beam section of subtype MESH as defined by SECTYPE. See the "Notes" (p. 1504) section
of this command description for details about user mesh section files. SECWRITE builds mesh files
based on 2-D models you create.
Notes
The SECREAD command operates on the section specified via the most recently issued SECTYPE command.
Issue a separate SECREAD command for each section ID that you want to read into ANSYS.
For shell sections incorporating FiberSIM data, the .xml file containing the FiberSIM data must exist in its
original format. ANSYS cannot use the data if you have altered the file in any way (for example, by opening
the file in an XML editor, making changes, and then saving it again).
Sample User Section Cell Mesh File
Here are excerpts from a sample user section mesh file for a section with 75 nodes, 13 cells, and 9 nodes
per cell for a two-hole box section. Illustrations of the two-hole box section and the cell mesh for it appear
later in this command description.
75 13
First Line:
1 3 25 23 2 14 24 12 13 1
3 5 27 25 4 16 26 14 15 1
Cells Section:
5 7 29 27 6 18 28 16 17 1
7 9 31 29 8 20 30 18 19 1
9 11 33 31 10 22 32 20 21 1
23 25 45 43 24 36 44 34 35 1
...
49 51 73 71 50 62 72 60 61 1
51 53 75 73 52 64 74 62 63 1
0 0.0 0.0
0 0.025 0.0
0 0.05 0.0
0 5.0175 0.0
...
0 19.98 10.00
0 20.00 10.00
Nodes Sec-
tion:
The mesh file is divided into three sections: the First Line, the Cells Section, and the Nodes Section. Here
are brief descriptions of the contents of each.
First Line: The First Line defines the number of nodes and the number of cells for the mesh.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1504
SECREAD
Cells Section: The Cells Section contains as many lines as there are cells. In this example, there are thirteen
cells, so there are thirteen lines in this section. In each line, the number 1 that follows the cell connectivity
information is the material number.
Cell nodal connectivity must be given in a counterclockwise direction, with the center node being the ninth
node. For details, see Figure 7: Cell Mesh for the Two-hole Box Section(p. 1506).
Nodes Section: The Nodes Section contains as many lines as there are nodes. In this example, there are 75
nodes, so there are a total of 75 lines in this section. Each node line contains the node's boundary flag, the
Y coordinate of the node, and the Z coordinate of the node. Currently, all node boundary flags appear as
0s in a cell mesh file (as illustrated in Figure 6: Two-hole Box Section(p. 1505)). Since all node boundary flags
are 0, SECREAD ignores them when it reads a cell mesh file into ANSYS.
There cannot be any gaps in the node numbering of a cell mesh. The nodes in a cell mesh must be numbered
consecutively, with the first node having a node number of 1, and the last node having a node number that
is equal to the maximum number of nodes in the cell mesh.
Figure 6: Two-hole Box Section
0.05"
0.03"
0.03" 0.04"
10" 10" 0.03"
1505
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECREAD
Figure 7: Cell Mesh for the Two-hole Box Section
65 66 67
54
43
68 71 69 70 72 73 74 75
64
53
23
12
1 3 2 4 5 6 7 8 9 10 11
22
33
34 42
A nine node cell
Menu Paths
Main Menu>Preprocessor>Sections>Beam>Custom Sections>Read Sect Mesh
Main Menu>Preprocessor>Sections>Section Library>Import Library
SECSTOP, dof, MINVALUE, MAXVALUE, dof, MINVALUE, MAXVALUE, dof, MINVALUE, MAXVALUE
Specifies stops on the components of relative motion in a joint element.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
dof
Local degree of freedom to be stopped.
MINVALUE
Low end of the range of allowed movement for the specified DOF.
MAXVALUE
High end of the range of allowed movement for the specified DOF.
Notes
Stops restrict motion in a DOF; motion beyond the MINVALUE or MAXVALUE is prevented (motion away
from a limit is allowed). You can specify up to three stops. If necessary, you can repeat the command.
Menu Paths
Main Menu>Preprocessor>Sections>Joints>Add / Edit
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1506
SECSTOP
SECTYPE, SECID, Type, Subtype, Name, REFINEKEY
Associates section type information with a section ID number.
PREP7: Cross Sections
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
SECID
Section identification number.
Type
BEAM
Defines a beam section.
PIPE
Defines a pipe section.
AXIS
Define the axis for a general axisymmetric section.
TAPER
Defines a tapered beam section. The sections at the end points must be topologically identical.
SHELL
Defines a shell section.
PRETENSION
Defines a pretension section.
JOINT
Defines a joint section.
REINF
Defines a reinforcing section.
GENB
Defines a nonlinear general (temperature-dependent) beam section.
GENS
Defines a preintegrated general (temperature-dependent) shell section.
Subtype
When Type = BEAM, the possible beam sections that can be defined for Subtype are:
Rectangle RECT
Quadrilateral QUAD
Circular solid CSOLID
Circular tube CTUBE
Channel CHAN
I-shaped section I
Z-shaped section Z
L-shaped section L
T-shaped section T
Hat-shaped section HATS
Hollow rectangle or box HREC
1507
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECTYPE
Arbitrary section -- integrated cross-section inertia properties
supplied by user
ASEC
User-defined mesh -- see the SECREAD command for more
information about this data
MESH
The following figure shows the shape of each cross section subtype:
RECT QUAD CSOLID CTUBE CHAN I
Z L T HATS HREC ASEC
When Type = GENB, the possible nonlinear general beam sections that can be defined for Subtype
are:
The generalized-stress/generalized-strain relationship is
elastic (linear or nonlinear).
ELASTIC
The generalized-stress/generalized-strain relationship is
elasto-plastic (and allows for permanent deformation).
PLASTIC
When Type = SHELL, an optional, special-purpose Subtype is available:
Specifies that layer data from the FiberSIM product will be
used to generate the section.
FIBERSIM
When Type = JOINT, the possible joint sections that can be defined for Subtype are:
Universal joint UNIV
Revolute joint REVO
3-D Slot joint SLOT
Point-in-plane joint PINP
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1508
SECTYPE
Translational joint PRIS
Cylindrical joint CYLI
Planar joint PLAN
Weld joint WELD
Orient joint ORIE
Spherical joint SPHE
General joint GENE
Screw joint SCRE
Bushing joint BUSH
When Type = REINF, the possible reinforcing sections that can be defined for Subtype are:
Discrete reinforcing. The reinforcing fibers are arbitrarily ori-
ented and modeled individually.
DISC
Smeared reinforcing. The reinforcing fibers are homogenous
and defined as a membrane.
SMEAR
Name
An eight-character name for the section. Name can be a string such as "W36X210" or "HP13X73" for
beam sections. Section name can consist of letters and numbers, but cannot contain punctuation, special
characters, or spaces.
REFINEKEY
Sets mesh refinement level for thin-walled beam sections. Valid values are 0 (the default - no mesh re-
finement) through 5 (high level of mesh refinement). This value has meaning only when Type = BEAM.
Notes
SECTYPE sets the section ID number, section type, and subtype for a section. If the section ID number is
not specified, ANSYS increments the highest section ID number currently defined in the database by one.
A previously-defined section with the same identification number will be redefined. The geometry data de-
scribing this section type is defined by a subsequent SECDATA command. Define the offsets by a subsequent
SECOFFSET command. The SLIST command lists the section properties, and the SECPLOT command displays
the section to scale. The SECNUM command assigns the section ID number to a subsequently-defined beam
element.
For a beam section (Type = BEAM), a subsequent SECDATA command builds a numeric model using a nine-
node cell for determining the properties (Ixx, Iyy, etc.) of the section and for the solution to the Poisson's
equation for torsional behavior. See Beam Analysis and Cross Sections in the Structural Analysis Guide for
examples using the section commands.
For a tapered beam section (Type = TAPER), two subsequent SECDATA commands are required (one for
each end section). Section ends must be topologically identical (same Subtype, number of cells and mater-
ial IDs).
1509
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SECTYPE
For a nonlinear general beam section (Type = GENB), the Subtype and REFINEKEY options do not apply.
Subsequent commands are necessary to define the section: BSAX, BSM1, BSM2, BSTQ, BSS1, BSS2, BSMD,
and BSTE are available. ANSYS ignores all other section commands for this section type.
For a preintegrated general shell section (Type = GENS), the Subtype and REFINEKEY options do not
apply. Subsequent commands are necessary to define the section: SSPA, SSPB, SSPD, SSPE, SSMT, SSBT,
and SSPM are available. ANSYS ignores all other section commands for this section type.
The PRETENSION section options of SECTYPE and SECDATA are documented primarily to aid your under-
standing of the data written by the CDWRITE command. ANSYS recommends that you generate pretension
sections via PSMESH.
To display elements with shapes determined from the section definition, issue the /ESHAPE command.
Menu Paths
Main Menu>Preprocessor>Sections>Axis>Add
Main Menu>Preprocessor>Sections>Axis>Edit
Main Menu>Preprocessor>Sections>Beam>Common Sections
Main Menu>Preprocessor>Sections>Beam>Custom Sections>Read Sect Mesh
Main Menu>Preprocessor>Sections>Beam>NL Generalized
Main Menu>Preprocessor>Sections>Beam>Taper Sections>By Picked Nodes
Main Menu>Preprocessor>Sections>Beam>Taper Sections>By XYZ Location
Main Menu>Preprocessor>Sections>Joints>Add / Edit
Main Menu>Preprocessor>Sections>Pipe>Add
Main Menu>Preprocessor>Sections>Pipe>Edit
Main Menu>Preprocessor>Sections>Reinforcing>Add / Edit
Main Menu>Preprocessor>Sections>Shell>Lay-up>Add / Edit
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
SECWRITE, Fname, Ext, --, ELEM_TYPE
Creates an ASCII file containing user mesh section information.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to Jobname if Fname is left blank.
Ext
Filename extension (8 character maximum).
The extension defaults to SECT if Ext is left blank.
--
Unused field.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1510
SECWRITE
ELEM_TYPE
Element type attribute pointer (ET) for the elements that are part of the section. See SECREAD for a
detailed description.
Notes
Before creating a user mesh file, you must create a model using ANSYS 2-D meshing capabilities. Use PLANE82
or MESH200 with KEYOPT(1) = 7 (quadrilateral with 8 nodes option) to model the cells. SECWRITE creates
an ASCII file that contains information about the nodes and cells that describe a beam section. For detailed
information on how to create a user mesh file, see Creating Custom Cross Sections with a User-defined Mesh
in the Structural Analysis Guide.
Menu Paths
Main Menu>Preprocessor>Sections>Custom Sectns>Write From Areas
SED, SEDX, SEDY, SEDZ
Defines the excitation direction for a single-point response spectrum.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SEDX, SEDY, SEDZ
Global Cartesian coordinates of a point that defines a line (through the origin) corresponding to the
excitation direction. For example: 0.0, 1.0, 0.0 defines global Y as the spectrum direction. Spectrum values
are not scaled with this input.
Notes
The excitation direction without rocking (ROCK) is normalized to one so that the SEDX, SEDY and SEDZ
values do not scale the spectrum.
The excitation direction with rocking is not normalized. The SEDX, SEDY and SEDZ values must be consistent
with the OMX, OMY and OMZ values on the ROCK command. The calculated direction then scales the spectrum.
For more information, see Participation Factors and Mode Coefficients.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>DDAM Options
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Settings
Main Menu>Solution>Load Step Opts>Spectrum>DDAM Options
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Settings
1511
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SED
SEDLIST, Sename, KOPT
Lists the DOF solution of a superelement after the use pass.
PREP7: Superelements
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Sename
Name of the superelement on File.DSUB to be listed. If a number, it is the element number of the
superelement as used in the use pass. If ALL, list results for all superelements.
KOPT
List key:
0
List summary data only.
1
List full contents. Be aware that the listing may be extensive.
Notes
Lists the degree of freedom solution of a superelement after the substructure use pass. Results may be listed
for any superelement on File.DSUB.
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>List Results>Superelem DOF
Utility Menu>List>Results>Superelem DOF Solu
SEEXP, Sename, Usefil, Imagky, Expopt
Specifies options for the substructure expansion pass.
SOLUTION: Analysis Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Sename
The name (case-sensitive) of the superelement matrix file created by the substructure generation pass
(Sename.SUB). Defaults to the initial jobname File. If a number, it is the element number of the su-
perelement as used in the use pass.
Usefil
The name of the file containing the superelement degree-of-freedom (DOF) solution created by the
substructure use pass (Usefil.DSUB).
Imagky
Key to specify use of the imaginary component of the DOF solution. Applicable only if the use pass is a
harmonic (ANTYPE,HARMIC) analysis:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1512
SEDLIST
OFF
Use real component of DOF solution (default).
ON
Use imaginary component of DOF solution.
Note
If all solutions are to be expanded (NUMEXP,ALL), Imagky is ignored and both the real and
imaginary solutions are expanded.
Expopt
Key to specify whether the superelement (ANTYPE,SUBSTR) expansion pass (EXPASS,ON) should transform
the geometry:
OFF
Do not transform node or element locations (default).
ON
Transform node or element locations in the FE geometry record of the .rst results file.
Notes
Specifies options for the expansion pass of the substructure analysis (ANTYPE,SUBSTR). If used in SOLUTION,
this command is valid only within the first load step.
If you specify geometry transformation (Expopt = ON), you must retrieve the transformation matrix (if it
exists) from the specified .SUB file. The command updates the nodal X, Y, and Z coordinates to represent
the transformed node locations. The Expopt option is useful when you want to expand superelements
created from other superelements (via SETRAN or SESYMM commands). For more information, see Supere-
lement Expansion in Transformed Locations and Plotting or Printing Mode Shapes.
This command is also valid in /PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>ExpansionPass>Single Expand>Expand Superelem
Main Menu>Solution>Load Step Opts>ExpansionPass>Single Expand>Expand Superelem
/SEG, Label, Aviname, DELAY
Allows graphics data to be stored in the local terminal memory.
GRAPHICS: Set Up
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Storage key:
1513
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/SEG
SINGL
Store subsequent display in a single segment (overwrites last storage).
MULTI
Store subsequent displays in unique segments [ANIM].
DELET
Delete all currently stored segments.
OFF
Stop storing display data in segments.
STAT
Display segment status.
PC
This option only applies to PC versions of ANSYS and only when animating via the AVI movie player
(i.e., /DEVICE,ANIM,2). This command appends frames to the File.AVI, so that the animation goes
in both directions (i.e., forward--backward--forward). You must have a current animation file to use
this option.
Aviname
Name of the animation file that will be created when each frame is saved. The .AVI extension is applied
automatically. Defaults to Jobname.AVI if no filename is specified.
DELAY
Delay factor between each frame, in seconds. Defaults to 0.015 seconds if no value is specified.
Command Default
No segment storage.
Notes
Allows graphics data to be stored in the terminal local memory (device-dependent). Storage occurs concur-
rently with the display.
Although the information from your graphics window is stored as an individual segment, you cannot plot
directly (GPLOT) from the segment memory.
For the DISPLAY program, the Aviname and DELAY fields are ignored.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Redirect Plots>Delete Segments
Utility Menu>PlotCtrls>Redirect Plots>Segment Status
Utility Menu>PlotCtrls>Redirect Plots>To Segment Memory
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1514
/SEG
SEGEN, Mode, nSuper, mDof, stopStage
Automatically generate superelements.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Mode
Specify action to take (must be specified as one of the following):
AUTO
Turn on feature.
OFF
Turn off feature.
nSuper
Number of superelements to create. The minimum number of superelements is 2, and the maximum
number of superelements is 999. Note that the number of requested superelements may not be the
same as the number of defined superelements (see "Notes" (p. 1515) for more details).
mDof
Specifies whether to use the master DOF defined by the user.
YES
Use master DOF defined by the user with the M command.
NO
Use the master DOF defined by the automatic generation process. Be aware that this option can
generate a large number of master DOFs (see "Notes" (p. 1515) for more details).
stopStage
Specifies when to stop the automatic superelement generation process.
PREVIEW
Preview the superelements only; stop after creating the domains which will become the superelements,
and after creating master DOF on the interfaces between each domain.
GEN
Create (generate) the superelements.
Command Default
No command default. If the command is issued, a Mode must be specified. By default, if the AUTO mode is
chosen, the SEGEN command automatically creates 10 superelements and defines master DOFs at all DOFs
on the interfaces between superelements.
Notes
This command can be used to quickly generate a set of superelements. Each superelement is created in a
separate file (jobnameXXX.sub, where XXX is a positive number from 1 to 999).
Due to the heuristics in the automatic domain decomposer, which is used to define the domains that will
become superelements, the number of defined superelements may exceed the number of requested super-
elements. Use the mDof and stopStage options to determine exactly how many superelements will be
created, the interfaces between each superelement, and where master DOF will be defined. With the
/PNUM,DOMAIN command, you can graphically (visually) preview the elements in each superelement. Then,
1515
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SEGEN
if required, you can add additional master DOF to (or remove from) the boundaries of the superelements.
Use the SEGEN command again with stopStage = GEN to actually create the superelements.
ANSYS automatically defines master DOF at each of the following: all interface DOF between superelements,
all DOF attached to contact elements (TARGE169 to CONTA177), and all DOF associated with nodes having
a point load defined. Note that for regular superelements, all interface DOFs must be defined as master
DOFs for the correct solution to be obtained. However, for CMS superelements, some of the interface DOFs
can be removed without a significant loss of accuracy.
For the case when mDof = YES, you should select the preview option first (stopStage = PREVIEW) to
verify exactly how many superelements will be created and where the superelement boundaries are located.
If more superelements will be created than were requested, you should define master DOF on the interface(s)
between all superelements.
This command is valid only for substructuring analyses (ANTYPE,SUBSTR). Use SEOPT to specify any options
for all of the superelements (e.g., which matrices to reduce), and possibly CMSOPT for any CMS substructuring
analysis. Note that the created superelements will follow the current /FILNAME instead of SENAME from
SEOPT. Also, only one load vector will be written to each .SUB file. Multiple load steps are not supported
with the automatic superelement generation process.
During the actual creation of the superelements, the output is redirected to jobname.autoTemp.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
SELIST, Sename, KOPT, KINT
Lists the contents of a superelement matrix file.
PREP7: Superelements
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Sename
The name (case-sensitive) of the superelement matrix file created by the substructure generation pass
(Sename.SUB). Defaults to the current Jobname. If a number, it is the element number of the supere-
lement as used in the use pass.
KOPT
List key:
0
List summary data only.
1
List contents, except load vectors and matrices.
2
List contents, except matrices.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1516
SELIST
3
List full contents. Be aware that the listing may be extensive.
KINT
Integer printout format key:
OFF
Default.
ON
Long format for large integers.
Notes
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Utility Menu>List>Other>Superelem Data
SELM
Specifies "Superelements" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Utility Menu>List>Status>Preprocessor>Superelements
1517
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SELM
SELTOL, Toler
Sets the tolerance for subsequent select operations.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Toler
Tolerance value. If blank, restores the default tolerance logic.
Notes
For selects based on non-integer numbers (e.g. coordinates, results, etc.), items within the range VMIN -
Toler and VMAX + Toler are selected, where VMIN and VMAX are the range values input on the xSEL
commands (ASEL, ESEL, KSEL, LSEL, NSEL, and VSEL).
The default tolerance logic is based on the relative values of VMIN and VMAX as follows:
If VMIN = VMAX, Toler = 0.005 x VMIN.
If VMIN = VMAX = 0.0, Toler = 1.0E-6.
If VMAX VMIN, Toler = 1.0E-8 x (VMAX-VMIN).
This command is typically used when VMAX-VMIN is very large so that the computed default tolerance is
therefore large and the xSEL commands selects more than what is desired.
Toler remains active until respecified by a subsequent SELTOL command. A SELTOL < blank > resets back
to the default Toler logic.
Menu Paths
This command cannot be accessed from a menu.
SENERGY, OPT, ANTYPE
Determines the stored magnetic energy or co-energy.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM EH <> PP <> EME <>
OPT
Item to be calculated:
0
Stored magnetic energy.
1
Stored magnetic co-energy.
ANTYPE
Analysis type:
0
Static or transient.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1518
SELTOL
1
Harmonic.
Notes
SENERGY invokes an ANSYS macro which calculates the stored magnetic energy or co-energy for all selected
elements. (For a harmonic analysis, the macro calculates a time-averaged (rms) stored energy.) A summary
table listing the energy by material number is produced. The energy density is also calculated and stored
on a per-element basis in the element table [ETABLE] with the label MG_ENG (energy density) or MG_COENG
(co-energy density). The macro erases all other items in the element table [ETABLE] and only retains the
energy density or co-energy density. Use the PLETAB and PRETAB commands to plot and list the energy
density. The macro is valid for static and low-frequency magnetic field formulations. The macro will not
calculate stored energy and co-energy for the following cases:
Orthotropic nonlinear permanent magnets.
Orthotropic nonlinear permeable materials.
Temperature dependent materials.
SENERGY is restricted to MKSA units.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Element Based>Co-Energy
Main Menu>General Postproc>Elec&Mag Calc>Element Based>Energy
SEOPT, Sename, SEMATR, SEPR, SESST, EXPMTH
Specifies substructure analysis options.
SOLUTION: Analysis Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Sename
The name (case-sensitive) assigned to the superelement matrix file. The matrix file will be named Se-
name.SUB. This field defaults to Fname on the /FILNAME command.
SEMATR
Matrix generation key:
1
Generate stiffness (or conductivity) matrix (default).
2
Generate stiffness and mass (or conductivity and specific heat) matrices.
3
Generate stiffness, mass and damping matrices.
SEPR
Print key:
1519
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SEOPT
0
Do not print superelement matrices or load vectors.
1
Print both load vectors and superelement matrices.
2
Print load vectors but not matrices.
SESST
Stress stiffening key:
0
Do not save space for stress stiffening in a later run.
1
Save space for the stress stiffening matrix (calculated in a subsequent generation run after the ex-
pansion pass).
EXPMTH
Expansion method for expansion pass:
BACKSUB
Save necessary factorized matrix files for backsubstitution during subsequent expansion passes (de-
fault). This normally results in a large usage of disk space
RESOLVE
Do not save factorized matrix files. Global stiffness matrix will be reformed during expansion pass.
This option provides an effective way to save disk space usage. This option cannot be used if the
use pass uses large deflections (NLGEOM,ON).
Notes
Specifies substructure analysis options (ANTYPE,SUBSTR). If used in SOLUTION, this command is valid only
within the first load step.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
SESYMM, Sename, Ncomp, INC, File, Ext, --
Performs a symmetry operation on a superelement within the use pass.
PREP7: Superelements
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Sename
The name (case-sensitive) of the superelement matrix file created by the substructure generation pass
(Sename.SUB). Defaults to the current Jobname. If a number, it is the element number of a previously
defined superelement in the current use pass.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1520
SESYMM
Ncomp
Symmetry key:
X
X symmetry (default).
Y
Y symmetry.
Z
Z symmetry.
INC
Increment all nodes in the superelement by INC.
File
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
This field must be input.
Ext
Filename extension (8 character maximum).
The extension defaults to SUB.
--
Unused field.
Notes
Performs a symmetry operation on a superelement within the substructure use pass by reversing the sign
of component Ncomp in the global Cartesian coordinate system. The node numbers are incremented by
INC. The new superelement is written to File.SUB in the current directory (by default). All master node
nodal coordinate systems must be global Cartesian (no rotated nodes allowed).
The maximum number of transformations for a given superelement is five (including SETRAN, SESYMM,
and the large rotation transformation if NLGEOM is ON in the use pass).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Superelem>By Reflection
1521
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SESYMM
*SET, Par, VALUE, VAL2, VAL3, VAL4, VAL5, VAL6, VAL7, VAL8, VAL9, VAL10
Assigns values to user-named parameters.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
An alphanumeric name used to identify this parameter. Par may be up to 32 characters, beginning with
a letter and containing only letters, numbers, and underscores. Examples: ABC A3X TOP_END. ANSYS
command names, function names, label names, component and assembly names, etc., should not be
used. Parameter names beginning with an underscore (e.g., _LOOP) are reserved for use by ANSYS and
should be avoided. Parameter names ending in an underscore are not listed by the *STATUS command.
Array parameter names must be followed by a subscript, and the entire expression must be 32 characters
or less. Examples: A(1,1) NEW_VAL(3,2,5) RESULT(1000). There is no character parameter substitution for
the Par field. Table parameters that are used in command fields (where constant values are normally
given) are limited to 32 characters.
VALUE
Numerical value or alphanumeric character string (up to 32 characters enclosed in single quotes) to be
assigned to this parameter. Examples: A(1,3)=7.4 B='ABC3'. May also be a parameter or a parametric ex-
pression. Examples: C=A(1,3) A(2,2)=(C+4)/2. If blank, delete this parameter. Example: A= deletes para-
meter A.
VAL2, VAL3, VAL4, VAL5, VAL6, VAL7, VAL8, VAL9, VAL10
If Par is an array parameter, values VAL2 through VAL10 (up to the last nonblank value) are sequentially
assigned to the succeeding array elements of the column. Example: *SET,A(1,4),10,11 assigns A(1,4)=10,
A(2,4)=11. *SET,B(2,3),'file10','file11' assigns B(2,3)='file10', B(3,3)='file11'.
Notes
Assigns values to user-named parameters that may be substituted later in the run. The equivalent (and re-
commended) format is
Par = VALUE,VAL2,VAL3, . . . , VAL10
which may be used in place of *SET,Par, ... for convenience.
This command is valid in any processor.
Parameter Definitions
Parameters (numeric or character) may be scalars (single valued) or arrays (multiple valued in one, two, or
three dimensions). An unlimited number of parameter names may be defined in any ANSYS run. For very
large numbers of parameters, it is most efficient to define them in alphabetical order.
Parameter values may be redefined at any time. Array parameters may also be assigned values within a do-
loop [*DO] for convenience. Internally programmed do-loop commands are also available with the *VXX
commands (*VFILL). Parameter values (except for parameters ending in an underscore) may be listed with
the *STATUS command, displayed with the *VPLOT command (numeric parameters only), and modified
with the *VEDIT command (numeric parameters only).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1522
*SET
Older ANSYS-supplied macro files may use parameter names that do not begin with an underscore. Using
these macros embedded in your own macros may cause conflicts if the same parameter names are used.
Parameters can also be resolved in comments created by the /COM command (see /COM for complete
documentation). A parameter can be deleted by redefining it with a blank VALUE. If the parameter is an
array, the entire array is deleted. Parameters may also be defined by a response to a query with the *ASK
command or from an "ANSYS-supplied" value with the *GET command.
Array Parameters
Array parameters must be dimensioned [*DIM] before being assigned values unless they are the result of
an array operation or defined using the implied loop convention. Scalar parameters that are not defined are
initialized to a "near" zero value. Numeric array parameters are initialized to zero when dimensioned, and
character array parameters are initialized to blank. An existing array parameter must be deleted before it
can be redimensioned. Array parameter names must be followed by a subscript list (enclosed in parentheses)
identifying the element of the array. The subscript list may have one, two, or three values (separated by
commas). Typical array parameter elements are A(1,1), NEW_VAL(3,2,5), RESULT(1000). Subscripts for defining
an array element must be integers (or parameter expressions that evaluate to integers). Non-integer values
are rounded to the nearest integer value. All array parameters are stored as 3-D arrays with the unspecified
dimensions set to 1. For example, the 4th array element of a 1-dimensional array, A(4), is stored as array
element A(4,1,1). Arrays are patterned after standard FORTRAN conventions.
Numerical Parameter Substitution
If the parameter name Par is input in a numeric argument of a command, the numeric value of the para-
meter (as assigned with *SET, *GET, =, etc.) is substituted into the command at that point. Substitution occurs
only if the parameter name is used between blanks, commas, parentheses, or arithmetic operators (or any
combination) in a numeric argument. Substitution can be prevented by enclosing the parameter name Par
within single quotes ( ' ), if the parameter is alone in the argument; if the parameter is part of an arithmetic
expression, the entire expression must be enclosed within single quotes to prevent substitution. In either
case the character string will be used instead of the numeric value (and the string will be taken as 0.0 if it
is in a numeric argument).
A forced substitution is available in the text fields of the /TITLE, /STITLE, /TLABEL, /AN3D, /SYP (ARG1--ARG8),
and *ABBR commands by enclosing the parameter within percent (%) signs. Also, parameter substitution
may be forced within the file name or extension fields of commands having these fields by enclosing the
parameter within percent (%) signs. Array parameters [*DIM] must include a subscript (within parentheses)
to identify the array element whose value is to be substituted, such as A(1,3). Out-of-range subscripts result
in an error message. Non-integer subscripts are allowed when identifying a TABLE array element for substi-
tution. A proportional linear interpolation of values among the nearest array elements is performed before
substitution. Interpolation is done in all three dimensions.
Note
Interpolation is based upon the assigned index numbers which must be defined when the table
is filled [*DIM].
Character Parameter Substitution
Most alphanumeric arguments permit the use of character parameter substitution. When the parameter
name Par input, the alphanumeric value of the parameter is substituted into the command at that point.
1523
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*SET
Substitution can be suppressed by enclosing the parameter name within single quotes ( ' ). Forced substitution
is available in some fields by enclosing the parameter name within percent (%) signs. Valid forced substitution
fields include command name fields, Fname (filename) or Ext (extension) arguments, *ABBR command
(Abbr arguments), /TITLE and /STITLE commands (Title argument) and /TLABEL command (Text argu-
ment). Character parameter substitution is also available in the *ASK, /AN3D, *CFWRITE, *IF, *ELSEIF, *MSG,
*SET, *USE, *VREAD, and *VWRITE commands. Character array parameters must include a subscript (within
parentheses) to identify the array element whose value is to be substituted.
Parameter Expressions
If a parameter operation expression is input in a numeric argument, the numeric value of the expression is
substituted into the command at that point. Allowable operation expressions are of the form
E1oE2oE3 ...oE10
where E1, E2, etc. are expressions connected by operators (o). The allowable operations (o) are
+ - * / ** < >
For example, A+B**C/D*E is a valid operation expression. The * represents multiplication and the ** represents
exponentiation.
Note
Exponentiation of a negative number (without parentheses) to an integer power follows standard
FORTRAN hierarchy conventions; that is, the positive number is exponentiated and then the sign
is attached. Thus, -4**2 is evaluated as -16. If parentheses are applied, such as (-4)**2, the result
is 16.
A parameter is evaluated as a number within parentheses before exponentiation. Exponentiation of a neg-
ative number to a non-integer power is performed by exponentiating the positive number and prepending
the minus sign, for example, -4**2.3 is -(4**2.3). The < and > operators allow conditional substitution. For
example, E1<E2 substitutes the value of E1 if the comparison is true or the value of E2 if the comparison is
false.
Spaces should not be used around operation symbols since * (a space and a star) makes the rest of the
line a comment. Operation symbols (or symbols and signs) may not be immediately adjacent to each other.
Parentheses may be used to separate symbols and signs, to determine a hierarchy of operations, or for
clarity. For example, A**(-B) must be used instead of A**-B. Numbers ending with +0nn or -0nn are assumed
to be of exponential form (as written on files by some computer systems) so that 123-002 is 123E-2 while
123-2 is 121. This form of exponential data should not be input directly. The default hierarchy follows the
standard FORTRAN conventions, namely:
operations in parentheses (innermost first)
then exponentiation (right to left)
then multiplication or division (left to right)
then unary association (such as +A or -A)
then addition or subtraction (left to right)
then logical evaluations (left to right).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1524
*SET
Expressions (E) may be a constant, a parameter, a function, or another operation expression (of the form
E1oE2oE3 ...oE10). Functions are of the form FTN(A) where the argument (A) may itself be of the form
E1oE2oE3 ...oE10. Operations are recursive to a level of four deep (three levels of internally nested parentheses).
Iterative floating point parameter arithmetic should not be used for high precision input because of the
accumulated numerical round off-error. Up to 10 expressions are accepted within a set of parenthesis.
Valid functions (which are based on standard FORTRAN functions where possible) are:
Sine SIN(X)
Cosine COS(X)
Tangent TAN(X)
Arcsine ASIN(X)
Arccosine ACOS(X)
Arctangent ATAN(X)
Arctangent (Y/X) with the sign of each component considered ATAN2(Y,X)
Hyperbolic sine SINH(X)
Hyperbolic cosine COSH(X)
Hyperbolic tangent TANH(X)
Square root SQRT(X)
Absolute value ABS(X)
Absolute value of X with sign of Y. Y=0 results in positive sign SIGN(X,Y)
Nearest integer NINT(X)
Remainder of X/Y Y=0 returns zero (0) MOD(X,Y)
Exponential EXP(X)
Natural log LOG(X)
Common log LOG10(X)
Random number, where X is the lower bound, and Y is the upper bound RAND(X,Y)
Random sample of Gaussian distributions, where X is the mean, and Y is
the standard deviation
GDIS(X,Y)
Lowercase equivalent of character parameter CPARM LWCASE(CPARM)
Uppercase equivalent of character parameter CPARM UPCASE(CPARM)
Numeric value of character parameter CPARM (If CPARM is a numeric
parameter, returns 0.0)
VALCHR(CPARM)
Character value of numerical parameter PARM. For ABS(PARM) < 10,
character value format is F8.5; for 10
PARM < 100,000,000, format is also F8.0. Otherwise result is 0.0 and is not
a character value.
Set the n2 bit in value b1 (bits are numbered from 0 to 31) IBSET(b1,n2)
Clear the n2 bit in value b1 IBCLR(b1,n2)
Test the n2 bit in value b1 (return true (1.0) if bit is set) BTEST(b1,n2)
Bitwise AND of value b1 and b2 BITAND(b1,b2)
1525
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*SET
Bitwise OR of value b1 and b2 BITOR(b1,b2)
Bitwise XOR of value b1 and b2 BITXOR(b1,b2)
Set the b2 bits in b1 BITSET(b1,b2)
Clear the b2 bits in b1 BITCLEAR(b1,b2)
Function arguments (X,Y, etc.) must be enclosed within parentheses and may be numeric values, parameters,
or expressions. Input arguments for angular functions must evaluate to radians by default. Output from an-
gular functions are also in radians by default. See the *AFUN command to use degrees instead of radians
for the angular functions. See the *VFUN command for applying these parameter functions to a sequence
of array elements. Additional functions, called "get functions" are described with the *GET command. For
SINH, COSH, TANH, and EXP, if X > 80, the value returned is zero.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Section
Main Menu>Solution>Define Loads>Delete>Structural>Section
Utility Menu>Parameters>Scalar Parameters
SET, Lstep, Sbstep, Fact, KIMG, TIME, ANGLE, NSET, ORDER
Defines the data set to be read from the results file.
POST1: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lstep
Load step number of the data set to be read (defaults to 1):
N
Read load step N.
FIRST
Read the first data set (Sbstep and TIME are ignored).
LAST
Read the last data set (Sbstep and TIME are ignored).
NEXT
Read the next data set (Sbstep and TIME are ignored). If at the last data set, the first data set will
be read as the next.
PREVIOUS
Read the previous data set (Sbstep and TIME are ignored). If at the first data set, the last data set
will be read as the previous.
NEAR
Read the data set nearest to TIME (Sbstep is ignored). If TIME is blank, read the first data set.
LIST
Scan the results file and list a summary of each load step. (KIMG, TIME, ANGLE, and NSET are ig-
nored.)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1526
SET
Sbstep
Substep number (within Lstep). Defaults to the last substep of the load step (except in a buckling or
modal analysis). For a buckling (ANTYPE,BUCKLE) or modal (ANTYPE,MODAL) analysis, Sbstep corres-
ponds to the mode number. Specify Sbstep = LAST to store the last substep for the specified load step
(that is, issue a SET,Lstep,LAST command).
If Lstep = LIST, Sbstep = 0 or 1 lists the basic step information. Sbstep = 2 also lists the basic step
information, but includes the load step title, and labels imaginary data sets if they exist.
Fact
Scale factor applied to data read from the file. If zero (or blank), a value of 1.0 is used. A nonzero factor
excludes non-summable items. Harmonic velocities or accelerations may be calculated from the displace-
ment results from a Modal (ANTYPE,MODAL) or Harmonic Response (ANTYPE,HARMIC) analyses. If Fact
= VELO, the harmonic velocities (v) are calculated from the displacements (d) at a particular frequency
(f ) according to the relationship v = 2 fd. Similarly, if Fact = ACEL, the harmonic accelerations (a) are
calculated as a = (2 f )
2
d.
If Lstep = LIST in an analysis employing rezoning, Fact can restrict the output to a file type such as
.rst or .rsnn. For example, to see only the portion of the data set associated with the .rs01 file,
issue a SET,LIST,,RS01 command.
KIMG
Used only with complex results (harmonic and complex modal analyses).
0 or REAL
Store the real part of complex solution (default).
1, 2 or IMAG
Store the imaginary part of a complex solution.
3 or AMPL
Store the amplitude
4 or PHAS
Store the phase angle. The angle value, expressed in degrees, will be between -180 and +180.
TIME
Time-point identifying the data set to be read. For the harmonic response analyses, time corresponds
to the frequency. For the buckling analysis, time corresponds to the load factor. Used only in the following
cases: If Lstep = NEAR, read the data set nearest to TIME. If both Lstep and Sbstep are zero (or
blank), read data set at time = TIME. Do not use TIME to identify the data set to be read if you used
the arc-length method (ARCLEN) in your solution. If TIME is between two solution time points on the
results file, a linear interpolation is done between the two data sets (except in rezoning). Solution items
not written to the results file (OUTRES) for either data set will result in a null item after data set interpol-
ation. If TIME is beyond the last time point on the file, the last time point will be used.
ANGLE
Circumferential location (0.0 to 360). Defines the circumferential location for the harmonic calculations
used when reading from the results file. The harmonic factor (based on the circumferential angle) is
applied to the harmonic elements (PLANE25, PLANE75, PLANE78, FLUID81, PLANE83, and SHELL61) of
the load case. See the Theory Reference for the Mechanical APDL and Mechanical Applications for details.
1527
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SET
Note
The factored values of applied constraints and loads will overwrite any values existing in the
database. If ANGLE = NONE, all harmonic factors are set to 1 and postprocessing will yield
the solution output. When using ANGLE = NONE with MODE > 0, the combined stresses and
strains are not valid. The default value of ANGLE is 0.0, but if the SET command is not used,
the effective default is NONE.
NSET
Data set number of the data set to be read. If a positive value for NSET is entered, Lstep, Sbstep,
KIMG, and TIME are ignored. Available set numbers can be determined by SET,LIST.
ORDER
Sorts the harmonic index results in ascending order of eigenfrequencies or buckling load multipliers.
This value applies to cyclic symmetry buckling and modal analyses only, and is valid only when Lstep
= FIRST, LAST, NEXT, PREVIOUS, NEAR or LIST.
Notes
Defines the data set to be read from the results file into the database. Various operations may also be per-
formed during the read operation. The database must have the model geometry available (or use the RESUME
command before the SET command to restore the geometry from Jobname.DB). Values for applied con-
straints [D] and loads [F] in the database will be replaced by their corresponding values on the results file,
if available. (See the description of the OUTRES command.) In a single load step analysis, these values are
usually the same, except for results from harmonic elements. (See the description of the ANGLE value above.)
In an interactive run, the sorted list (ORDER option) is also available for results-set reading via a GUI pick
option.
During rezoning, Lstep values NEXT, PREVIOUS, and NEAR are not available.
In ANSYS, you can postprocess results without issuing a SET command if the solution results were saved to
the database file (Jobname.DB). Distributed ANSYS, however, can only postprocess using the results file
(Jobname.RST) and cannot use the Jobname.DB file since no solution results are written to the database.
Therefore, you must issue a SET command before postprocessing in Distributed ANSYS.
When postprocessing amplitudes or phases (KIMG = AMPL or PHAS):
The only appropriate coordinate system is the solution coordinate system (RSYS ,SOLU). When displaying
the displacement amplitudes, use a contour display (PLNSOL command). Because a deformed shape
display (PLDISP command) could lead to a non-physical shape, the displacement scaling is off by default
(/DSCALE,,OFF).
The conversion is not valid for derived results such as principal stress/strain, equivalent stress/strain,
and USUM.
Cyclic symmetry results are not supported.
Menu Paths
Main Menu>General Postproc>List Results>Detailed Summary
Main Menu>General Postproc>List Results>Detailed Summary (Freq ordered)
Main Menu>General Postproc>Read Results>By Load Step
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1528
SET
Main Menu>General Postproc>Read Results>By Pick
Main Menu>General Postproc>Read Results>By Pick (Freq ordered)
Main Menu>General Postproc>Read Results>First Set
Main Menu>General Postproc>Read Results>Last Set
Main Menu>General Postproc>Read Results>Next Set
Main Menu>General Postproc>Read Results>Previous Set
Main Menu>General Postproc>Results Summary
Main Menu>General Postproc>Results Summary (Freq ordered)
Utility Menu>List>Results>Load Step Summary
SETFGAP, GAP, ROPT, --, PAMB, ACF1, ACF2, PREF, MFP
Updates or defines the real constant table for squeeze film elements.
PREP7: Real Constants
MP ME <> <> <> <> <> <> <> <> <> PP <> EME MFS
GAP
Gap separation.
ROPT
Real constant set option.
0
Creates separate real constant sets for each selected element with the specified real constant values
(default).
1
Updates existing real constant sets. The gap separation is updated from displacement results in the
database. Other real constants are updated as specified in the command input parameters.
--
Unused field
PAMB
Ambient pressure.
ACF1, ACF2
Accommodation factor 1 and 2.
PREF
Reference pressure for mean free path.
MFP
Mean free path.
Notes
This command is used for large signal cases to update the gap separation real constant on a per-element
basis. Issue this command prior to solution using the default ROPT value to initialize real constant sets for
every fluid element. After a solution, you can re-issue the command to update the real constant set for a
subsequent analysis. See "Thin Film Analysis" for more information on thin film analyses.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1529
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SETFGAP
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Real Constants>ThinFilm
Main Menu>Preprocessor>Real Constants>ThinFilm
Main Menu>Solution>Load Step Opts>Other>Real Constants>ThinFilm
SETRAN, Sename, KCNTO, INC, File, Ext, --, DX, DY, DZ, NOROT
Creates a superelement from an existing superelement.
PREP7: Superelements
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Sename
The name (case-sensitive) of the file containing the original superelement matrix created by the generation
pass (Sename.SUB). The default is the current Jobname. If Sename is a number, it is the element
number of a previously defined superelement in the current use pass.
KCNTO
The reference number of the coordinate system to where the superelement is to be transferred. The
default is the global Cartesian system. Transfer occurs from the active coordinate system.
INC
The node offset. The default is zero. All new element node numbers are offset from those on the original
by INC.
File
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
This field requires input.
Ext
Filename extension (8 character maximum).
The extension defaults to SUB.
--
Unused field.
DX, DY, DZ
Node location increments in the global Cartesian coordinate system. Defaults to zero.
NOROT
Node rotation key:
0
The nodal coordinate systems of the transferred superelement rotate into the KCNTO system. (That
is, the nodal coordinate systems rotate with the superelement.) The superelement matrices remain
unchanged. This value is the default.
1
The nodal coordinate systems do not rotate. (That is, they remain fixed in their original global orient-
ation.) The superelement matrices and load vectors are modified if any rotations occur.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1530
SETRAN
Notes
The SETRAN command creates a superelement from an existing superelement and writes the new element
to a file. You can then issue an SE command to read the new element (during the use pass).
You can create a superelement from an original by:
Transferring the original's geometry from the active coordinate system into another coordinate system
(KCNTO)
Offsetting its geometry in the global Cartesian coordinate system (DX, DY, and DZ )
Offsetting its node numbers (INC).
A combination of methods is valid. If you specify both the geometry transfer and the geometry offset, the
transfer occurs first.
If you specify rotation of the transferred superelement's nodal coordinate systems into the KCNTO system
(NOROT = 0), the rotated nodes cannot be coupled via the CP command; in this case, issue the CE command
instead. If you specify no rotation of the nodal coordinate systems (NOROT = 1) for models with displacement
degrees of freedom, and KCNTO is not the active system, the superelement Sename must have six MDOF
at each node that has MDOF; therefore, only elements with all six structural DOFs are valid in such cases.
There is no limit to the number of copies that can be made of a superelement, provided the copies are all
generated from the same original superelement. However, nested copies are limited to five. In other words,
the total number of different Sename usages on the SETRAN and SESYMM commands is limited to five.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Superelem>By CS Transfer
Main Menu>Preprocessor>Modeling>Create>Elements>Superelem>By Geom Offset
SEXP, LabR, Lab1, Lab2, EXP1, EXP2
Forms an element table item by exponentiating and multiplying.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabR
Label assigned to results. If same as existing label, the existing values will be overwritten by these results.
Lab1
First labeled result item in operation.
Lab2
Second labeled result item in operation (may be blank).
EXP1
Exponent applied to Lab1.
EXP2
Exponent applied to Lab2.
1531
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SEXP
Notes
Forms a labeled result item (see ETABLE command) for the selected elements by exponentiating and mul-
tiplying two existing labeled result items according to the operation:
LabR = (|Lab1|
EXP1
) x (|Lab2|
EXP2
)
Roots, reciprocals, and divides may also be done with this command.
Menu Paths
Main Menu>General Postproc>Element Table>Exponentiate
SF, Nlist, Lab, VALUE, VALUE2
Specifies surface loads on nodes.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Nlist
Nodes defining the surface upon which the load is to be applied. Use the label ALL or P, or a component
name. If ALL, all selected nodes [NSEL] are used (default). If P, graphical picking is enabled and all re-
maining command fields are ignored (valid only in the GUI).
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each element
type in the Element Reference.
Label Description Surface Load Label Discipline
pressure PRES Structural
frequency (harmonic analyses
only)
FREQ
convection CONV[1] Thermal
heat flux HFLUX[1]
radiation RAD
surface-to-surface radiation RDSF
fluid-structure interaction flag FSI[2] Acoustic fluid
impedance boundary flag IMPD
constant total pressure PTOT
Maxwell force flag MXWF Magnetic
magnetic circuit interface MCI
surface charge density CHRGS Electric
Maxwell force flag MXWF
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1532
SF
Label Description Surface Load Label Discipline
number 1-50 for a waveguide
exterior port
PORT High-frequency
electromagnet-
ic
surface shielding properties SHLD
Maxwell surface flag of equival-
ent source surface
MXWF
surface impedance IMPD
field-surface interface number FSIN[3] Field-surface
interface
Fluid flow flux FFLX Poromechanics
1. Thermal labels CONV and HFLUX are mutually exclusive.
2. For an acoustic analysis, apply the fluid-structure interaction flag (Label = FSI) to only the FLUID129
or FLUID130 elements.
3. For a fluid-solid interaction analysis, apply the field-surface interface flag (Label = FSIN) twice: once
for the fluid side (FLUID141 or FLUID142 elements) and once for the solid side.
VALUE
Surface load value or table name reference for specifying tabular boundary conditions.
If Lab = PRES, VALUE is the real component of the pressure.
If Lab = CONV, VALUE is typically the film coefficient and VALUE2 (below) is typically the bulk temper-
ature. If VALUE = -N, the film coefficient may be a function of temperature and is determined from the
HF property table for material N [MP]. The temperature used to evaluate the film coefficient is usually
the average between the bulk and wall temperatures, but may be user-defined for some elements.
If Lab = MCI, VALUE indicates current direction (-1; current flow into the element face (IN), +1; current
flow out of the element face (OUT). If Lab = RAD, VALUE is surface emissivity.
If Lab = PORT, VALUE is a port number representing a waveguide exterior port. The port number must
be an integer between 1 and 50.
If Lab = SHLD, VALUE is surface conductivity.
If Lab = IMPD, VALUE is resistance in ohms/square.
If Lab = RDSF, VALUE is the emissivity value; the following conditions apply: If VALUE is between 0 and
1, apply a single value to the surface. If VALUE= -N, the emissivity may be a function of the temperature,
and is determined from the EMISS property table for material N (MP). The material N does not need to
correlate with the underlying solid thermal elements.
If Lab = FSIN in a Multi-field solver (single or multiple code coupling) analysis, VALUE is the surface in-
terface number. If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VALUE is not used unless the
ANSYS analysis is performed using the Multi-field solver.
VALUE2
Second surface load value (if any).
1533
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SF
If Lab = PRES, VALUE2 is the imaginary component of the pressure. Imaginary pressures can only be
used by SURF153 and SURF154 and can only be used for a full harmonic response analysis (HROPT,FULL),
or by a mode superposition harmonic response analysis (HROPT,MSUP) if the mode extraction method
is Block Lanczos (MODOPT,LANB), PCG Lanczos (MODOPT,LANPCG), or Supernode (MODOPT,SNODE).
If Lab = CONV, VALUE2 is typically the bulk temperature.
If Lab = RAD, VALUE2 is the ambient temperature.
If Lab = SHLD, VALUE2 is relative permeability and defaults to 1.0.
If Lab = IMPD, VALUE2 is reactance in ohms/square.
If Lab = RDSF, VALUE2 is the enclosure number. Radiation will occur between surfaces flagged with
the same enclosure numbers. If the enclosure is open, radiation will also occur to ambient. If VALUE2
is negative radiation direction is reversed and will occur inside the element for the flagged radiation
surfaces. Negative value of enclosure number is applicable for FLUID141 and FLUID142 elements, to
model radiation occurring between surfaces inside the fluid domain.
If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VALUE2 is the surface interface number (not
available from within the GUI).
If Lab = PORT, VALUE2 is not used.
Notes
Individual nodes may not be entered for this command. The node list is to identify a surface and the Nlist
field must contain a sufficient number of nodes to define an element surface. The loads are internally stored
on element faces defined by the specified nodes. All nodes on an element face must be specified for the
face to be used, and the element must be selected. If all nodes defining a face are shared by an adjacent
face of another selected element, the face is not free and will not have a load applied. If more than one
element can share the same nodes (for example, a surface element attached to a solid element), select the
desired element type before issuing the SF command. The SF command applies only to area and volume
elements. For shell elements, if the specified nodes include face one (which is usually the bottom face) along
with other faces (such as edges), only face one is used. Where faces cannot be uniquely determined from
the nodes, or where the face does not fully describe the load application, use the SFE command. A load key
of 1 (which is typically the first loading condition on the first face) is used if the face determination is not
unique. A uniform load value is applied over the element face.
See the SFBEAM command for applying surface loads to beam elements. See the SFGRAD command for
an alternate tapered load capability. See the SFFUN command for applying loads from a node vs. value
function. Also see the SFE command for applying tapered loads on individual element faces. Use the SFDELE
command to delete loads applied with this command. Use the SFCUM command to accumulate (add) surface
loads applied with SF.
Tabular boundary conditions (VALUE = %tabname% and/or VALUE2 = %tabname%) are available for the
following surface load labels (Lab) only: PRES (real and/or imaginary components), CONV (film coefficient
and/or bulk temperature) or HFLUX, and RAD (surface emissivity and ambient temperature). Use the *DIM
command to define a table.
This command is also valid in PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1534
SF
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>AppImped_E>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>AppShield>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior
Port>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>AppInfinite>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>AppMaxwell>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Field Surface Intr>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Field Surface>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Impedance>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Flag>AppInfinite>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Flag>AppMCI>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppMaxwell>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Node Components
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Convection>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flux>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Radiation>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Surface Rad>On Nodes
Main Menu>Preprocessor>Trefftz Domain>AppInfinite>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>AppImped_E>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>AppShield>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior Port>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Flag>AppInfinite>On Nodes
Main Menu>Solution>Define Loads>Apply>Electric>Flag>AppMaxwell>On Nodes
Main Menu>Solution>Define Loads>Apply>Field Surface Intr>On Nodes
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Field Surface>On Nodes
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Impedance>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Flag>AppInfinite>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Flag>AppMCI>On Nodes
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppMaxwell>On Nodes
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Node Components
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Nodes
Main Menu>Solution>Define Loads>Apply>Thermal>Convection>On Nodes
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flux>On Nodes
Main Menu>Solution>Define Loads>Apply>Thermal>Radiation>On Nodes
Main Menu>Solution>Define Loads>Apply>Thermal>Surface Rad>On Nodes
1535
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SF
SFA, AREA, LKEY, Lab, VALUE, VALUE2
Specifies surface loads on the selected areas.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
AREA
Area to which surface load applies. If ALL, apply load to all selected areas [ASEL]. If AREA = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
may be substituted for AREA.
LKEY
Load key associated with surface load (defaults to 1). Load keys (1,2,3, etc.) are listed under "Surface
Loads" in the input data table for each element type in the Element Reference. LKEY is ignored if the
area is the face of a volume region meshed with volume elements.
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each area type
in the Element Reference.
Label Description
Body Load
Label
Discipline
pressure PRES Structural
convection CONV[1] Thermal
heat flux HFLUX[1]
radiation RAD
surface-to-surface radiation RDSF
fluid-structure interaction
flag
FSI[2] Fluid
impedance boundary flag IMPD
Maxwell force flag MXWF Magnetic
magnetic circuit interface MCI
surface charge density CHRGS Electric
Maxwell force flag MXWF
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
number 1-50 for a waveguide
exterior port
PORT High-frequency
electromagnet-
ic
surface shielding properties SHLD
surface impedance IMPD
field-surface interface num-
ber
FSIN[3] Field-surface
interface
1. Thermal labels CONV and HFLUX are mutually exclusive.
2. For an acoustic analysis, apply the fluid-structure interaction flag (Label = FSI) to only the FLUID129
or FLUID130 elements.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1536
SFA
3. For a fluid-solid interaction analysis, apply the field-surface interface flag (Label = FSIN) twice: once
for the fluid side (FLUID141 or FLUID142 elements) and once for the solid side.
VALUE
Surface load value or table name reference for specifying tabular boundary conditions.
If Lab = CONV, VALUE is typically the film coefficient and VALUE2 (below) is typically the bulk temper-
ature. If Lab = CONV and VALUE = -N, the film coefficient may be a function of temperature and is de-
termined from the HF property table for material N [MP]. The temperature used to evaluate the film
coefficient is usually the average between the bulk and wall temperatures, but may be user-defined for
some elements.
If Lab = MCI, VALUE indicates current direction (-1; current flow into the element face (IN), +1; current
flow out of the element face (OUT).
If Lab = RAD, VALUE is the surface emissivity.
If Lab = PORT, VALUE is a port number representing a waveguide exterior port. The port number must
be an integer between 1 and 50.
If Lab = SHLD, VALUE is surface conductivity.
If Lab = IMPD, VALUE is resistance in ohms/square.
If Lab = RDSF, VALUE is the emissivity value; the following conditions apply: If VALUE is between 0 and
1, apply a single value to the surface. If VALUE = -N, the emissivity may be a function of the temperature,
and is determined from the EMISS property table for material N (MP). The material N does not need to
correlate with the underlying solid thermal elements.
If Lab = FSIN in a Multi-field solver (single or multiple code coupling) analysis, VALUE is the surface in-
terface number and LKEY is ignored. If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VALUE is
not used unless the ANSYS analysis is performed using the Multi-field solver.
VALUE2
Second surface load value (if any).
If Lab = CONV, VALUE2 is typically the bulk temperature.
If Lab = RADVALUE2 is ambient temperature.
If Lab = SHLD, VALUE2 is relative permeability (defaults to 1.0).
If Lab = IMPD, VALUE2 is reactance in ohms/square.
If Lab = RDSF, VALUE2 is the enclosure number. Radiation will occur between surfaces flagged with
the same enclosure numbers. If the enclosure is open, radiation will also occur to ambient. If VALUE2
is negative radiation direction is reversed and will occur inside the element for the flagged radiation
surfaces. Negative value of enclosure number is applicable for FLUID141 and FLUID142 elements, to
model radiation occurring between surfaces inside the fluid domain.
If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VALUE2 is the surface interface number (not
available from within the GUI).
1537
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFA
Notes
Surface loads may be transferred from areas to elements with the SFTRAN or SBCTRAN commands. See
the SFGRAD command for an alternate tapered load capability.
Tabular boundary conditions (VALUE = %tabname% and/or VALUE2 = %tabname%) are available for the
following surface load labels (Lab) only: PRES (real and/or imaginary components), CONV (film coefficient
and/or bulk temperature) or HFLUX, and RAD (surface emissivity and ambient temperature). Use the *DIM
command to define a table.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>AppImped_E>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>AppShield>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior
Port>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>AppInfinite>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>AppMaxwell>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Field Surface Intr>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Field Surface>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Impedance>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Flag>AppInfinite>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Flag>AppMCI>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppMaxwell>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Convection>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flux>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Radiation>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Surface Rad>On Areas
Main Menu>Preprocessor>Trefftz Domain>AppInfinite>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>AppImped_E>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>AppShield>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>EMPorts>Exterior Port>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Flag>AppInfinite>On Areas
Main Menu>Solution>Define Loads>Apply>Electric>Flag>AppMaxwell>On Areas
Main Menu>Solution>Define Loads>Apply>Field Surface Intr>On Areas
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Field Surface>On Areas
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Impedance>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Flag>AppInfinite>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Flag>AppMCI>On Areas
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppMaxwell>On Areas
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Areas
Main Menu>Solution>Define Loads>Apply>Thermal>Convection>On Areas
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flux>On Areas
Main Menu>Solution>Define Loads>Apply>Thermal>Radiation>On Areas
Main Menu>Solution>Define Loads>Apply>Thermal>Surface Rad>On Areas
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1538
SFA
SFACT, TYPE
Allows safety factor or margin of safety calculations to be made.
POST1: Element Table
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
TYPE
Type of calculation:
0
No nodal safety factor or margin of safety calculations.
1
Calculate and store safety factors in place of nodal stresses.
2
Calculate and store margins of safety in place of nodal stresses.
Command Default
No nodal safety factor or margin of safety calculations.
Notes
Allows safety factor (SF) or margin of safety (MS) calculations to be made for the average nodal stresses ac-
cording to:
SF = SALLOW/|Stress|
MS = (SALLOW/|Stress|) -- 1.0
Calculations are done during the display, select, or sort operation (in the active coordinate system [RSYS])
with results stored in place of the nodal stresses. Use the PRNSOL or PLNSOL command to display the results.
Note
The results are meaningful only for the stress (SIG1, SIGE, etc.) upon which SALLOW is based.
Nodal temperatures used are those automatically stored for the node. Related commands are
SFCALC, SALLOW, TALLOW.
Menu Paths
Main Menu>General Postproc>Safety Factor>Restore NodeStrs
Main Menu>General Postproc>Safety Factor>SF for Node Strs
1539
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFACT
SFADELE, AREA, LKEY, Lab
Deletes surface loads from areas.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
AREA
Area to which surface load deletion applies. If ALL, delete load from all selected areas [ASEL]. If AREA
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may be substituted for AREA.
LKEY
Load key associated with surface load (defaults to 1). See the SFA command for details.
Lab
Valid surface load label. If ALL, use all appropriate labels. See the SFA command for labels.
Notes
Deletes surface loads (and all corresponding finite element loads) from selected areas.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Surface Ld>On All Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>AppImped_E>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>AppShield>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>DelExtPort>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppInfinite>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppMaxwell>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Field Surface Intr>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Field Surface>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Impedance>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppMCI>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Pressure>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Ambient Rad>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Convection>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Heat Flux>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Radiation>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Surface Rad>On Areas
Main Menu>Solution>Define Loads>Delete>All Load Data>All Surface Ld>On All Areas
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>AppImped_E>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>AppShield>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>DelExtPort>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppInfinite>On Areas
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppMaxwell>On Areas
Main Menu>Solution>Define Loads>Delete>Field Surface Intr>On Areas
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1540
SFADELE
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Field Surface>On Areas
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Impedance>On Areas
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Areas
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppMCI>On Areas
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Areas
Main Menu>Solution>Define Loads>Delete>Structural>Pressure>On Areas
Main Menu>Solution>Define Loads>Delete>Thermal>Ambient Rad>On Areas
Main Menu>Solution>Define Loads>Delete>Thermal>Convection>On Areas
Main Menu>Solution>Define Loads>Delete>Thermal>Heat Flux>On Areas
Main Menu>Solution>Define Loads>Delete>Thermal>Radiation>On Areas
Main Menu>Solution>Define Loads>Delete>Thermal>Surface Rad>On Areas
SFALIST, AREA, Lab
Lists the surface loads for the specified area.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
AREA
Area at which surface load is to be listed. If ALL (or blank), list for all selected areas [ASEL]. If AREA = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may be substituted for AREA.
Lab
Valid surface load label. If ALL (or blank), use all appropriate labels. See the SFA command for labels.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Surface Loads>On All Areas
Utility Menu>List>Loads>Surface Loads>On Picked Areas
SFBEAM, ELEM, LKEY, Lab, VALI, VALJ, VAL2I, VAL2J, IOFFST, JOFFST, LENRAT
Specifies surface loads on beam and pipe elements.
SOLUTION: FE Surface Loads
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
ELEM
Element to which surface load is applied. If ALL, apply load to all selected beam elements (ESEL). If ELEM
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may be substituted in ELEM.
1541
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFBEAM
LKEY
Load key associated with surface load (defaults to 1). Load keys (1, 2, 3, etc.) are listed under "Surface
Loads" in the input table for each element type in the Element Reference. For beam and some pipe ele-
ments, the load key defines the load orientation.
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each element
type in the Element Reference. Structural labels: PRES (pressure).
VALI, VALJ
Surface load values at nodes I and J. If VALJ is blank, it defaults to VALI. If VALJ is zero, a zero is used.
VAL2I, VAL2J
Second surface load values at nodes I and J. Currently not used.
IOFFST, JOFFST
Offset distance from node I (toward node J) where VALI is applied, and offset distance from node J
(toward node I) where VALJ is applied, respectively.
LENRAT
Offset distance flag:
0
Offset is in terms of length units (default).
1
Offset is in terms of a length ratio (0.0 to 1.0).
Notes
Specifies surface loads on the selected beam elements. Distributed loads are applied on a force-per-length
basis (that is, the width of the underlying element is not considered). To list and delete surface loads applied
with this command, use the SFELIST and SFEDELE commands, respectively.
If no offset values (IOFFSET and JOFFSET) are specified, the load is applied over the full element length.
Values may also be input as length fractions, depending on the LENRAT setting. For example, assuming a
line length of 5.0, an IOFFST of 2.0 with LENRAT = 0 or an IOFFST of 0.4 with LENRAT = 1 represent the
same point. If JOFFST = -1, VALI is assumed to be a point load at the location specified via IOFFST, and
VALJ is ignored. (IOFFSET cannot be equal to -1.) The offset values are stepped even if you issue a KBC,0
command.
Offsets are only available for element types BEAM3, BEAM4, BEAM23, BEAM24, BEAM44, and BEAM54. Offsets
are also available for BEAM188 and PIPE288 if using the cubic shape function (KEYOPT(3) = 3) for those element
types.
Only constant pressures are supported for element types PIPE16, PIPE17, PIPE18, PIPE20, PIPE59, and PIPE60.
Tapered pressures are not recognized for these elements.
To accumulate (add) surface loads applied with this command, use the SFCUM,,ADD command. Use the
same offset values used on the previous SFBEAM command (for a given element face); otherwise, the loads
do not accumulate. If no offsets are specified, the command applies the previous offset values.
This command is also valid in PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1542
SFBEAM
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Beams
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Beams
SFCALC, LabR, LabS, LabT, TYPE
Calculates the safety factor or margin of safety.
POST1: Element Table
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
LabR
Label assigned to results. If same as existing label, the existing values will be overwritten by these results.
LabS
Labeled result item corresponding to the element stress.
LabT
Labeled result item corresponding to the element temperature.
TYPE
Type of calculation:
0 or 1
Use safety factor (SF) calculation.
2
Use margin of safety (MS) calculation.
3
Use 1/SF calculation.
Notes
Calculates safety factor (SF) or margin of safety (MS) as described for the SFACT command for any labeled
result item (see ETABLE command) for the selected elements. Use the PRETAB or PLETAB command to
display results. Allowable element stress is determined from the SALLOW-TALLOW table [SALLOW, TALLOW].
Menu Paths
Main Menu>General Postproc>Safety Factor>SF for ElemTable
SFCUM, Lab, Oper, FACT, FACT2
Specifies that surface loads are to be accumulated.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Lab
Valid surface load label. If ALL, use all appropriate labels.
1543
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFCUM
Label Description
Body
Load La-
bel
Discipline
pressure PRES Structural
convection CONV[1] Thermal
heat flux HFLUX[1]
load vector number SELV Substructure
surface charge density CHRGS Electric
Maxwell force flag MXWF
Maxwell force flag MXWF Magnetic
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
1. Thermal labels CONV and HFLUX are mutually exclusive.
Oper
Accumulation key:
REPL
Subsequent values replace the previous values (default).
ADD
Subsequent values are added to the previous values.
IGNO
Subsequent values are ignored.
FACT
Scale factor for the first surface load value. A (blank) or '0' entry defaults to 1.0.
FACT2
Scale factor for the second surface load value. A (blank) or '0' entry defaults to 1.0.
Command Default
Replace previous values.
Notes
Allows repeated surface loads (pressure, convection, etc.) to be replaced, added, or ignored. Surface loads
are applied with the SF, SFE, and SFBEAM commands. Issue the SFELIST command to list the surface loads.
The operations occur when the next surface load specifications are defined. For example, issuing the SF
command with a pressure value of 25 after a previous SF command with a pressure value of 20 causes the
current value of that pressure to be 45 with the add operation, 25 with the replace operation, or 20 with
the ignore operation. All new pressures applied with SF after the ignore operation will be ignored, even if
no current pressure exists on that surface.
Scale factors are also available to multiply the next value before the add or replace operation. A scale factor
of 2.0 with the previous "add" example results in a pressure of 70. Scale factors are applied even if no previous
values exist. Issue SFCUM,STAT to show the current label, operation, and scale factors. Solid model boundary
conditions are not affected by this command, but boundary conditions on the FE model are affected.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1544
SFCUM
Note
The FE boundary conditions may still be overwritten by existing solid model boundary conditions
if a subsequent boundary condition transfer occurs.
SFCUM does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Surface Loads
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Surface Loads
SFDELE, Nlist, Lab
Deletes surface loads.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Nlist
Label defining where to find the list of nodes:
ALL
Use all selected nodes [NSEL]. If P, use graphical picking in GUI. A component label may be substituted
for Nlist.
Lab
Valid surface load label. If ALL, use all appropriate labels. See the SF command for labels.
Notes
Deletes surface loads as applied with the SF command. Loads are deleted only for the specified nodes on
external faces of selected area and volume elements. For shell elements, if the specified nodes include face
one (which is usually the bottom face) along with other faces (such as edges), only the loads on face one
will be deleted. The element faces are determined from the list of selected nodes as described for the SF
command. See the SFEDELE command for deleting loads explicitly by element faces.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>AppImped_E>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>AppShield>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>DelExtPort>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppInfinite>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppMaxwell>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Field Surface Intr>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Field Surface>On Nodes
1545
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFDELE
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Impedance>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppMCI>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Pressure>On Node Components
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Pressure>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Ambient Rad>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Convection>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Heat Flux>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Radiation>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Surface Rad>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>AppImped_E>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>AppShield>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>DelExtPort>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppInfinite>On Nodes
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppMaxwell>On Nodes
Main Menu>Solution>Define Loads>Delete>Field Surface Intr>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Field Surface>On Nodes
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Impedance>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppMCI>On Nodes
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Nodes
Main Menu>Solution>Define Loads>Delete>Structural>Pressure>On Node Components
Main Menu>Solution>Define Loads>Delete>Structural>Pressure>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Ambient Rad>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Convection>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Heat Flux>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Radiation>On Nodes
Main Menu>Solution>Define Loads>Delete>Thermal>Surface Rad>On Nodes
SFE, ELEM, LKEY, Lab, KVAL, VAL1, VAL2, VAL3, VAL4
Specifies surface loads on elements.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element to which surface load applies. If ALL, apply load to all selected elements [ESEL]. If ELEM = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may be substituted for ELEM.
LKEY
Load key or face number associated with surface load (defaults to 1). Load keys (1,2,3, etc.) are listed
under "Surface Loads" in the input data table for each element type in the Element Reference.
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each element
type in the Element Reference.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1546
SFE
Label Description
Surface-
Load Label
Discipline
pressure PRES Structural
frequency (harmonic ana-
lyses only)
FREQ
convection CONV[1] Thermal
heat flux HFLUX[1]
radiation RAD
surface-to-surface radiation RDSF
fluid-structure interaction
flag
FSI[2] Fluid
volume fraction for VOF
method
VFRC
impedance boundary flag IMPD
Maxwell force flag MXWF Magnetic
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
load vector number SELV Substructure
surface charge density CHRGS Electric
Maxwell force flag MXWF
waveguide port number PORT High-frequency
electromagnet-
ic
surface shielding properties SHLD
field-surface interface num-
ber
FSIN[3] Field-surface
interface
Fluid flow flux FFLX Poromechanics
1. Thermal labels CONV and HFLUX are mutually exclusive.
2. For an acoustic analysis, apply the fluid-structure interaction flag (Label = FSI) to only the FLUID129
or FLUID130 elements.
3. For a fluid-solid interaction analysis, apply the field-surface interface flag (Label = FSIN) twice: once
for the fluid side (FLUID141 or FLUID142 elements) and once for the solid side.
KVAL
Value key. If Lab = PRES:
0 or 1
VAL1 through VAL4 are used as real components of pressures.
2
VAL1 through VAL4 are used as imaginary components of pressures.
Value key. If Lab = CONV:
0 or 1
VAL1 through VAL4 are used as the film coefficients.
1547
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFE
2
VAL1 through VAL4 are the bulk temperatures.
Value key. If Lab = SHLD:
0 or 1
VAL1 through VAL4 are used as the electrical conductivities.
2
VAL1 through VAL4 are used as the relative permeabilities.
Value key. If Lab = RAD:
0 or 1
VAL1 through VAL4 are used as the emissivities.
2
VAL1 through VAL4 are ambient temperatures.
Value key. If Lab = RDSF:
0 or 1
VAL1 is the emissivity value between 0 and 1.
2
VAL1 is the enclosure number.
Value key. If Lab = SELV:
0 or 1
VAL1 is the multiplier on real load vector LKEY.
2
VAL1 is the multiplier on imaginary load vector LKEY.
Value key. If Lab = VFRC:
0 or 1
VAL1 is the boundary load value. VAL1 defaults to 1.
2
VAL1 is the wetting status. VAL1 defaults to 0 or non-wetting.
If only one set of data is supplied (either emissivities or temperatures when Lab = RAD; or either film
coefficients or temperatures when Lab = CONV; or either conductivity or relative permeability when
Lab = SHLD), the other set of data defaults to previously specified values (or zero if not previously spe-
cified).
VAL1
First surface load value (typically at the first node of the face) or the name of a table for specifying tab-
ular boundary conditions. Face nodes are listed in the order given for "Surface Loads" in the input data
table for each element type in the Element Reference. For example, for SOLID45, the item 1-JILK associates
LKEY = 1 (face 1) with nodes J, I, L, and K. Surface load value VAL1 then applies to node J of face 1. To
specify a table, enclose the table name in percent signs (%), e.g., %tabname%. Use the *DIM command
to define a table. Only one table can be specified, and it must be specified in the VAL1 position; tables
specified in the VAL2, VAL3, or VAL4 positions will be ignored. VAL2 applies to node I, etc.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1548
SFE
If Lab = PRES and KVAL = 2, VAL1 is the imaginary pressure, which is used only by SURF153 or SURF154
in full harmonic response analyses (HROPT,FULL), or by a mode superposition harmonic response ana-
lysis (HROPT,MSUP) if the mode extraction method is Block Lanczos (MODOPT,LANB), PCG Lanczos
(MODOPT,LANPCG), or Supernode (MODOPT,SNODE).
If Lab = CONV, KVAL = 0 or 1, and VAL1 = -N, the film coefficient is assumed to be a function of tem-
perature and is determined from the HF property table for material N [MP]. The temperature used to
evaluate the film coefficient is usually the average between the bulk and wall temperatures, but may
be user defined for some elements.
If Lab = PORT, VAL1 is a port number representing a waveguide port. The port number must be an integer
between 1 and 6.
If Lab = RDSF, KVAL = 0 or 1, and VAL1 = -N, the emissivity is assumed to be a function of the temper-
ature, and is determined from the EMISS property table for material N (MP). The material N does not
need to correlate with the underlying solid thermal elements. If Lab = RDSF, KVAL = 2, and VAL1 is
negative, radiation direction is reversed and will occur inside the element for the flagged radiation surfaces.
Negative value of enclosure number is applicable for FLUID141 and FLUID142 elements, to model radiation
occurring between surfaces inside the fluid domain.
If Lab = VFRC, VAL1 is the boundary value or the wetting status, as defined by KVAL above. If Lab =
VFRC and KVAL = 2, a VAL1 setting of 1 indicates a wetted boundary. At a wetted boundary, the fluid
upstream keeps the associated elements full.
If Lab = FSIN in a Multi-field solver (single or multiple code coupling) analysis, VAL1 is the surface interface
number. KVAL is not used. If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VAL1 is not used
unless the ANSYS analysis is performed using the Multi-field solver.
VAL2, VAL3, VAL4
Surface load values at the 2nd, 3rd, and 4th nodes (if any) of the face. If all three values are blank, all
three default to VAL1, giving a constant load. Zero or other blank values are used as zero. If a table is
specified in VAL1, then any tables specified in VAL2, VAL3, or VAL4 will be ignored.
If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VAL2 is the surface interface number (not
available from within the GUI). VAL3 and VAL4 are not used.
Notes
Specifies surface loads on selected elements.
Caution
You cannot use the SFE command with the INFIN110 or INFIN111 elements without prior know-
ledge of element face orientation, i.e., you must know which face is the exterior in order to flag
it. Also, the surface effect elements, SURF153 and SURF154, use LKEY to allow the use of tangential
and other loads (see SURF153and SURF154 of the Element Reference for more details).
Tapered loads may be applied over the faces of most elements. Exceptions include elements PIPE16, PIPE17,
PIPE18, PIPE20, PIPE59, and PIPE60, where tapered pressures are not recognized. Only constant pressures
are supported for these elements.
1549
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFE
For beam elements allowing lateral surface loads that may be offset from the nodes, use the SFBEAM
command to specify the loads and offsets. See the SF command for an alternate surface load definition
capability based upon node numbers. See the SFGRAD command for an alternate tapered load capability.
Use the SFCUM command to accumulate (add) surface loads applied with SFE.
The SFE command can also be used to define fluid pressure penetration loads (Lab = PRES) at a contact
interface. For this type of load, LKEY = 1 is used to specify the pressure values and LKEY = 2 is used to
specify starting points and penetrating points. See Applying Fluid Penetration Pressure in the Contact Tech-
nology Guide for details on how to apply this type of load.
You can specify a table name only when using structural (PRES) and thermal (CONV (film coefficient and/or
bulk temperature), HFLUX), boundary value and wetting status (VFRC), and surface emissivity and ambient
temperature (RAD) surface load labels. The Volume of Fluid method (Lab = VFRC) is applicable only for
FLUID141.
When a tabular function load is applied to an element, the load will not vary according to the positioning
of the element in space.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Ele-
ments
Main Menu>Preprocessor>Loads>Define Loads>Apply>Field Surface Intr>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Impedance>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Apply>Load Vector>For Superelement
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Element Components
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Convection>On Elements>Tapered
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Convection>On Elements>Uniform
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flux>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Radiation>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Surface Rad>On Elements
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Elements
Main Menu>Solution>Define Loads>Apply>Field Surface Intr>On Elements
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Impedance>On Elements
Main Menu>Solution>Define Loads>Apply>Load Vector>For Superelement
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Element Components
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Elements
Main Menu>Solution>Define Loads>Apply>Thermal>Convection>On Elements>Tapered
Main Menu>Solution>Define Loads>Apply>Thermal>Convection>On Elements>Uniform
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flux>On Elements
Main Menu>Solution>Define Loads>Apply>Thermal>Radiation>On Elements
Main Menu>Solution>Define Loads>Apply>Thermal>Surface Rad>On Elements
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1550
SFE
SFEDELE, ELEM, LKEY, Lab
Deletes surface loads from elements.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element to which surface load deletion applies. If ALL, delete load from all selected elements [ESEL]. If
ELEM = P, graphical picking is enabled and all remaining command fields are ignored (valid only in the
GUI). A component name may be substituted for ELEM.
LKEY
Load key associated with surface load (defaults to 1). If ALL, delete surface loads for all load keys.
Lab
Valid surface load label. If ALL, use all appropriate labels. See the SFE command for labels.
Notes
Deletes surface loads from selected elements. See the SFDELE command for an alternate surface load deletion
capability based upon selected nodes.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Surface Ld>On All Elems
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Boundary>AppImped_E>On Ele-
ments
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Ele-
ments
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppInfinite>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppMaxwell>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Field Surface Intr>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Impedance>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Volume Fract>Bound Loads>On
Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Load Vector>For Superelement
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppMCI>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Pressure>On Element Compon-
ents
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Pressure>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Ambient Rad>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Convection>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Heat Flux>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Radiation>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Surface Rad>On Elements
Main Menu>Solution>Define Loads>Delete>All Load Data>All Surface Ld>On All Elems
Main Menu>Solution>Define Loads>Delete>Electric>Boundary>AppImped_E>On Elements
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Elements
1551
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFEDELE
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppInfinite>On Elements
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppMaxwell>On Elements
Main Menu>Solution>Define Loads>Delete>Field Surface Intr>On Elements
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Impedance>On Elements
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Volume Fract>Bound Loads>On Elements
Main Menu>Solution>Define Loads>Delete>Load Vector>For Superelement
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Elements
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppMCI>On Elements
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Elements
Main Menu>Solution>Define Loads>Delete>Structural>Pressure>On Element Components
Main Menu>Solution>Define Loads>Delete>Structural>Pressure>On Elements
Main Menu>Solution>Define Loads>Delete>Thermal>Ambient Rad>On Elements
Main Menu>Solution>Define Loads>Delete>Thermal>Convection>On Elements
Main Menu>Solution>Define Loads>Delete>Thermal>Heat Flux>On Elements
Main Menu>Solution>Define Loads>Delete>Thermal>Radiation>On Elements
Main Menu>Solution>Define Loads>Delete>Thermal>Surface Rad>On Elements
SFELIST, ELEM, Lab
Lists the surface loads for elements.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
ELEM
Element at which surface load is to be listed. If ALL (or blank), list loads for all selected elements [ESEL].
If ELEM = P, graphical picking is enabled and all remaining command fields are ignored (valid only in
the GUI). A component name may be substituted for ELEM.
Lab
Valid surface load label. If ALL (or blank), use all appropriate labels.
Label Description
Body Load
Label
Discipline
pressure PRES Structural
convection CONV[1] Thermal
heat flux HFLUX[1]
radiation RAD
surface-to-surface radiation RDSF
fluid-structure interaction
flag
FSI[2] Fluid
volume fraction for VOF
method
VFRC
impedance boundary flag IMPD
Maxwell force flag MXWF Magnetic
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
load vector number SELV Substructure
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1552
SFELIST
Label Description
Body Load
Label
Discipline
surface charge density CHRGS Electric
Maxwell force flag MXWF
waveguide port number PORT High-frequency
electromagnet-
ic
surface shielding properties SHLD
1. Thermal labels CONV and HFLUX are mutually exclusive.
2. If Lab = FSI, only the fluid elements must be selected for the flag to be applied.
Notes
The surface loads listed correspond to the current database values. The database is not updated for surface
loads in POST1. Surface loads specified in tabular form, however, do list their values corresponding to the
current results set in POST1.
For SURF151 or SURF152 elements with an extra node for radiation and/or convection calculations (KEYOPT(5)
= 1), the bulk temperature listed is the temperature of the extra node. If the thermal solution does not
converge, the extra node temperature is not available for listing.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Surface Loads>On All Elements
Utility Menu>List>Loads>Surface Loads>On Picked Elems
SFFUN, Lab, Par, Par2
Specifies a varying surface load.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each element
type in the Element Reference. Issue SFFUN,STATUS to list current command settings.
Label Description
Body Load
Label
Discipline
pressure PRES Structural
convection CONV[1] Thermal
heat flux HFLUX[1]
surface charge density CHRGS Electric
1. Thermal labels CONV and HFLUX are mutually exclusive.
1553
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFFUN
Par
Parameter containing list of surface load values. If Lab = CONV, values are typically the film coefficients
and Par2 values (below) are typically the bulk temperatures.
Par2
Parameter containing list of second surface load values (if any). If Lab = CONV, the Par2 values are
typically the bulk temperatures. Par2 is not used for other surface load labels.
Notes
Specifies a surface load "function" to be used when the SF or SFE command is issued. The function is supplied
through an array parameter vector which contains nodal surface load values. Node numbers are implied
from the sequential location in the array parameter. For example, a value in location 11 applies to node 11.
The element faces are determined from the implied list of nodes when the SF or SFE command is issued.
Zero values should be supplied for nodes that have no load. A tapered load value may be applied over the
element face. These loads are in addition to any loads that are also specified with the SF or SFE commands.
Issue SFFUN (with blank remaining fields) to remove this specification. Issue SFFUN,STATUS to list current
settings.
Starting array element numbers must be defined for each array parameter vector. For example, SF-
FUN,CONV,A(1,1),A(1,2) reads the first and second columns of array A (starting with the first array element
of each column) and associates the values with the nodes. Operations continue on successive column array
elements until the end of the column. Another example to show the order of the commands:
*dim,nodepres,array,2
nodepres(1)=11,12
/prep7
et,1,42
n,1
n,2,1
n,3,1,1
n,4,,1
e,1,2,3,4
sfe,1,1,pres,1,3
sfelist ! expected answer: 3 at both nodes 1 and 2
sfedel,all,pres,all
sffun,pres, nodepres(1)
sfe,1,1,pres,1,5
sfelist ! expected answer: 5+11=16 at node 1, 5+12=17 at node 2
fini
SFFUN does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>For Surface Ld>Node Function
Main Menu>Solution>Define Loads>Settings>For Surface Ld>Node Function
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1554
SFFUN
SFGRAD, Lab, SLKCN, Sldir, SLZER, SLOPE
Specifies a gradient (slope) for surface loads.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each element
type in the Element Reference.
Label Description
Body Load
Label
Discipline
pressure (real pressure only) PRES Structural
convection (bulk temperat-
ures only)
CONV[1] Thermal
heat flux HFLUX[1]
surface charge density CHRGS Electric
1. Thermal labels CONV and HFLUX are mutually exclusive.
SLKCN
Reference number of slope coordinate system (used with Sldir and SLZER to determine COORD).
Defaults to 0 (the global Cartesian coordinate system).
Sldir
Slope direction in coordinate system SLKCN:
X
Slope is along X direction (default). Interpreted as R direction for non-Cartesian coordinate systems.
Y
Slope is along Y direction. Interpreted as direction for non-Cartesian coordinate systems.
Z
Slope is along Z direction. Interpreted as direction for spherical or toroidal coordinate systems.
SLZER
Coordinate location (degrees for angular input) where slope contribution is zero (CVALUE = VALUE). Allows
the slope contribution to be shifted along the slope direction. For angular input, SLZER should be
between 180 if the singularity [CSCIR] is at 180 and should be between 0 and 360 if the singularity
is at 0.
SLOPE
Slope value (load per unit length or per degree).
Notes
Specifies a gradient (slope) for surface loads. All surface loads issued with the SF, SFE, SFL, or SFA commands
while this specification is active will have this gradient applied (for complex pressures, only the real component
will be affected; for convections, only the bulk temperature will be affected). The load value, CVALUE, calcu-
lated at each node is:
CVALUE = VALUE + (SLOPE X (COORD-SLZER))
1555
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFGRAD
where VALUE is the load value specified on the subsequent SF, SFE, SFL, or SFA commands and COORD is
the coordinate value (in the Sldir direction of coordinate system SLKCN) of the node. Only one SFGRAD
specification may be active at a time (repeated use of this command replaces the previous specification with
the new specification). Issue SFGRAD (with blank fields) to remove the specification. Issue SFGRAD,STAT to
show the current command status. The SFGRAD specification (if active) is removed when the LSREAD (if
any) command is issued.
For elements PIPE16, PIPE17, PIPE18, PIPE20, and PIPE60, tapered pressures are not recognized. Only constant
pressures are supported for these elements.
SFGRAD does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>For Surface Ld>Gradient
Main Menu>Solution>Define Loads>Settings>For Surface Ld>Gradient
SFL, LINE, Lab, VALI, VALJ, VAL2I, VAL2J
Specifies surface loads on lines of an area.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LINE
Line to which surface load applies. If ALL, apply load to all selected lines [LSEL]. If LINE = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may be substituted for LINE.
Lab
Valid surface load label. Load labels are listed under "Surface Loads" in the input table for each element
type in the Element Reference.
Label Description
Body Load
Label
Discipline
pressure PRES Structural
convection CONV[1] Thermal
heat flux HFLUX[1]
radiation RAD
surface-to-surface radiation RDSF
fluid-structure interaction
flag
FSI[2] Fluid
volume fraction for VOF
method
VFRC
impedance boundary flag IMPD
Maxwell force flag MXWF Magnetic
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1556
SFL
Label Description
Body Load
Label
Discipline
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
surface charge density CHRGS Electric
Maxwell electrostatic force
flag
MXWF
field-surface interface num-
ber
FSIN[3] Field-surface
interface
1. Thermal labels CONV and HFLUX are mutually exclusive.
2. For an acoustic analysis, apply the fluid-structure interaction flag (Label = FSI) to only the FLUID129
or FLUID130 elements.
3. For a fluid-solid interaction analysis, apply the field-surface interface flag (Label = FSIN) twice: once
for the fluid side (FLUID141 or FLUID142 elements) and once for the solid side.
VALI, VALJ
Surface load values at the first keypoint (VALI) and at the second keypoint (VALJ) of the line, or table
name for specifying tabular boundary conditions. If VALJ is blank, it defaults to VALI. If VALJ is zero,
a zero is used. If Lab = CONV, VALI and VALJ are the film coefficients and VAL2I and VAL2J are the
bulk temperatures. To specify a table, enclose the table name in percent signs (%), e.g., %tabname%.
Use the *DIM command to define a table. If Lab = CONV and VALI = -N, the film coefficient may be a
function of temperature and is determined from the HF property table for material N [MP]. If Lab = RAD,
VALI and VALJ values are surface emissivities and VAL2I and VAL2J are ambient temperatures. The
temperature used to evaluate the film coefficient is usually the average between the bulk and wall
temperatures, but may be user defined for some elements. If Lab = RDSF, VALI is the emissivity value;
the following condition apply: If VALI = -N, the emissivity may be a function of the temperature and
is determined from the EMISS property table for material N [MP]. If Lab = VFRC, VALI is the boundary
value (defaults to 1). If Lab = FSIN in a Multi-field solver (single or multiple code coupling) analysis,
VALI is the surface interface number. If Lab = FSIN in a unidirectional ANSYS to CFX analysis, VALJ is
the surface interface number (not available from within the GUI) and VALI is not used unless the ANSYS
analysis is performed using the Multi-field solver.
VAL2I, VAL2J
Second surface load values (if any). If Lab = CONV, VAL2I and VAL2J are the bulk temperatures. If Lab
= RAD, VAL2I and VAL2J are the ambient temperatures. If Lab = RDSF, VAL2I is the enclosure number.
Radiation will occur between surfaces flagged with the same enclosure numbers. If the enclosure is open,
radiation will occur to the ambient. Negative value of enclosure number is applicable for FLUID141 and
FLUID142 elements, to model radiation occurring between surfaces inside the fluid domain. VAL2I and
VAL2J are not used for other surface load labels. If VAL2J is blank, it defaults to VAL2I. If VAL2J is
zero, a zero is used. To specify a table (Lab = CONV), enclose the table name in percent signs (%), e.g.,
%tabname%. Use the *DIM command to define a table. If Lab = VFRC, VAL2I is the wetting status
value (defaults to 0 or non-wetting). If Lab = VFRC, a VAL2I setting of 1 indicates a wetted boundary.
At a wetted boundary, the fluid upstream keeps the associated elements full.
Notes
Specifies surface loads on the selected lines of area regions. The lines represent either the edges of area
elements or axisymmetric shell elements themselves. Surface loads may be transferred from lines to elements
1557
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFL
with the SFTRAN or SBCTRAN commands. See the SFE command for a description of surface loads. Loads
input on this command may be tapered. See the SFGRAD command for an alternate tapered load capability.
You can specify a table name only when using structural (PRES) and thermal (CONV (film coefficient and/or
bulk temperature), HFLUX), boundary value and wetting status (VFRC), and surface emissivity and ambient
temperature (RAD) surface load labels. VALJ and VAL2J are ignored for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>AppInfinite>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Flag>AppMaxwell>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Field Surface Intr>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Field Surface>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/ANSYS>Impedance>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Flag>AppInfinite>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Other>AppMaxwell>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pressure>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Convection>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Heat Flux>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Radiation>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Surface Rad>On Lines
Main Menu>Solution>Define Loads>Apply>Electric>Excitation>AppSurfChar>On Lines
Main Menu>Solution>Define Loads>Apply>Electric>Flag>AppInfinite>On Lines
Main Menu>Solution>Define Loads>Apply>Electric>Flag>AppMaxwell>On Lines
Main Menu>Solution>Define Loads>Apply>Field Surface Intr>On Lines
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Field Surface>On Lines
Main Menu>Solution>Define Loads>Apply>Fluid/ANSYS>Impedance>On Lines
Main Menu>Solution>Define Loads>Apply>Magnetic>Flag>AppInfinite>On Lines
Main Menu>Solution>Define Loads>Apply>Magnetic>Other>AppMaxwell>On Lines
Main Menu>Solution>Define Loads>Apply>Structural>Pressure>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Convection>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Heat Flux>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Radiation>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Surface Rad>On Lines
SFLDELE, LINE, Lab
Deletes surface loads from lines.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LINE
Line to which surface load deletion applies. If ALL, delete load from all selected lines [LSEL]. If LINE =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may be substituted for LINE.
Lab
Valid surface load label. If ALL, use all appropriate labels. See the SFL command for labels.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1558
SFLDELE
Notes
Deletes surface loads (and all corresponding finite element loads) from selected lines.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Surface Ld>On All Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppInfinite>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Electric>Flag>AppMaxwell>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Field Surface Intr>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Field Surface>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/ANSYS>Impedance>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Fluid/CFD>Volume Fract>Bound Loads>On
Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Structural>Pressure>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Ambient Rad>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Convection>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Heat Flux>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Radiation>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Delete>Thermal>Surface Rad>On Lines
Main Menu>Solution>Define Loads>Delete>All Load Data>All Surface Ld>On All Lines
Main Menu>Solution>Define Loads>Delete>Electric>Excitation>AppSurfChar>On Lines
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppInfinite>On Lines
Main Menu>Solution>Define Loads>Delete>Electric>Flag>AppMaxwell>On Lines
Main Menu>Solution>Define Loads>Delete>Field Surface Intr>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Field Surface>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/ANSYS>Impedance>On Lines
Main Menu>Solution>Define Loads>Delete>Fluid/CFD>Volume Fract>Bound Loads>On Lines
Main Menu>Solution>Define Loads>Delete>Magnetic>Flag>AppInfinite>On Lines
Main Menu>Solution>Define Loads>Delete>Magnetic>Other>AppMaxwell>On Lines
Main Menu>Solution>Define Loads>Delete>Structural>Pressure>On Lines
Main Menu>Solution>Define Loads>Delete>Thermal>Ambient Rad>On Lines
Main Menu>Solution>Define Loads>Delete>Thermal>Convection>On Lines
Main Menu>Solution>Define Loads>Delete>Thermal>Heat Flux>On Lines
Main Menu>Solution>Define Loads>Delete>Thermal>Radiation>On Lines
Main Menu>Solution>Define Loads>Delete>Thermal>Surface Rad>On Lines
SFLEX, FFAX, FFBY, FFBZ, FFTO, FFTSY, FFTSZ
Sets flexibility factors for the currently defined pipe element section.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
FFAX
Factor to increase axial flexibility. The default value is 1.0.
1559
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFLEX
FFBY
Factor to increase bending flexibility about element y axis (bending in the element x-z plane). The default
value is 1.0.
FFBZ
Factor to increase bending flexibility about element z axis (bending in the element x-y plane). The default
value is FFBY.
FFTO
Factor to increase torsional flexibility. The default value is 1.0.
FFTSY
Factor to increase transverse shear flexibility in the element x-z plane. The default value is 1.0.
FFTSZ
Factor to increase transverse shear flexibility in the element x-y plane. The default value is FFTSY.
Notes
The SFLEX command sets section-flexibility factors for sections used by current-technology pipe elements.
To increase stiffness, use a flexibility factor of less than 1.0.
The FFBY and FFTSY arguments affect motion in the element x-z plane, and the FFBZ and FFTSZ arguments
affect motion in the element x-y plane. For stout pipe structures with low slenderness ratios, set both FFBY
and FFTSY--and/or both FFBZ and FFTSZ (the related bending and transverse shear factors)--to the same
value to obtain the expected flexibility effect.
When issued, the SFLEX command applies to the pipe section most recently defined via the SECTYPE
command.
SFLEX is valid only for linear material properties and small strain analyses. The command does not support
offsets, temperature loading, or initial state loading.
Menu Paths
Main Menu>Preprocessor>Sections>Pipe>Flexibility Factors
SFLIST, NODE, Lab
Lists surface loads.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
NODE
Node at which surface load is to be listed. If ALL (or blank), list for all selected nodes [NSEL]. If NODE =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may be substituted for NODE.
Lab
Valid surface load label. If ALL (or blank), use all appropriate labels.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1560
SFLIST
Label Description
Body
Load La-
bel
Discipline
pressure PRES Structural
convection CONV Thermal
heat flux HFLUX
radiation RAD
surface-to-surface radiation RDSF
fluid-structure interaction
flag
FSI[1] Fluid
impedance boundary flag IMPD
Maxwell force flag MXWF Magnetic
magnetic circuit interface MCI
Exterior surface flag for IN-
FIN110 and INFIN111
INF Infinite ele-
ment
surface charge density CHRGS Electric
Maxwell force flag MXWF
waveguide port number PORT High-frequency
electromagnet-
ic
surface shielding properties SHLD
1. If Lab = FSI, only the fluid elements must be selected for the flag to be applied.
Notes
Lists the surface loads as applied with the SF command. Loads are listed only for the specified nodes on
external faces of selected area and volume elements. Use SFELIST for line elements. The surface loads listed
correspond to the current database values. The database is not updated for surface loads in POST1. Surface
loads specified in tabular form, however, do list their values corresponding to the current results set in
POST1.
For SURF151 or SURF152 elements with an extra node for radiation and/or convection calculations (KEYOPT(5)
= 1), the bulk temperature listed is the temperature of the extra node. If the thermal solution does not
converge, the extra node temperature is not available for listing.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Surface Loads>On All Nodes
Utility Menu>List>Loads>Surface Loads>On Picked Nodes
1561
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFLIST
SFLLIST, LINE, Lab
Lists the surface loads for lines.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
LINE
Line at which surface load is to be listed. If ALL (or blank), list for all selected lines [LSEL]. If LINE = P,
graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may be substituted for LINE.
Lab
Valid surface load label. If ALL (or blank), use all appropriate labels. See the SFL command for labels.
Notes
Lists the surface loads for the specified line.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Loads>Surface Loads>On All Lines
Utility Menu>List>Loads>Surface Loads>On Picked Lines
SFSCALE, Lab, FACT, FACT2
Scales surface loads on elements.
SOLUTION: FE Surface Loads
MP ME ST PR PRN <> <> FL <> <> <> PP <> EME MFS
Lab
Valid surface load label. If ALL, use all appropriate labels.
Label Description
Body Load
Label
Discipline
pressure PRES Structural
convection CONV[1] Thermal
heat flux HFLUX[1]
load vector number SELV Substructure
surface charge density CHRGS Electric
1. Thermal labels CONV and HFLUX are mutually exclusive.
FACT
Scale factor for the first surface load value. Zero (or blank) defaults to 1.0. Use a small number for a zero
scale factor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1562
SFLLIST
FACT2
Scale factor for the second surface load value. Zero (or blank) defaults to 1.0. Use a small number for a
zero scale factor.
Notes
Scales surface loads (pressure, convection, etc.) in the database on the selected elements. Surface loads are
applied with the SF, SFE, or SFBEAM commands. Issue the SFELIST command to list the surface loads. Solid
model boundary conditions are not scaled by this command, but boundary conditions on the FE model are
scaled.
Note
Such scaled FE boundary conditions may still be overwritten by unscaled solid model boundary
conditions if a subsequent boundary condition transfer occurs.
SFSCALE does not work for tabular boundary conditions.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Scale FE Loads>Surface Loads
Main Menu>Solution>Define Loads>Operate>Scale FE Loads>Surface Loads
SFTRAN
Transfer the solid model surface loads to the finite element model.
SOLUTION: Solid Surface Loads
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Notes
Surface loads are transferred only from selected lines and areas to all selected elements. The SFTRAN oper-
ation is also done if the SBCTRAN command is issued or automatically done upon initiation of the solution
calculations [SOLVE].
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Operate>Transfer to FE>Surface Loads
Main Menu>Solution>Define Loads>Operate>Transfer to FE>Surface Loads
1563
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SFTRAN
/SHADE, WN, Type
Defines the type of surface shading used with Z-buffering.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
Type
Shading type:
FACET or 0
Facet shading (one color per area face) (default).
GOURAUD or 1
Gouraud smooth shading (smooth variation of color based on interpolated vertex colors).
PHONG or 2
Phong smooth shading (smooth variation of color based on interpolated vertex normals).
Command Default
Facet shading.
Notes
Defines the type of surface shading used on area, volume, and PowerGraphics [/GRAPHICS,POWER] displays
when software Z-buffering is enabled [/TYPE]. This command is only functional for 2-D display devices.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Hidden-Line Options
SHELL, Loc
Selects a shell element or shell layer location for results output.
POST1: Controls
POST26: Controls
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Loc
Location within shell element (or layer) to obtain stress results:
TOP
Top of shell element (or layer) (default).
MID
Middle of shell element (or layer). The default method averages the TOP and BOT values to obtain
a mid value. Setting KEYOPT(8) = 2 for SHELL181, SHELL208, SHELL209, and ELBOW290, or KEYOPT(11)
= 2 for SHELL63, uses MID results obtained directly from the results file.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1564
/SHADE
BOT
Bottom of shell element (or layer).
Command Default
Shell element (or layer) top location.
Notes
Selects the location within a shell element (or a shell layer) for results output (nodal stresses, strains, etc.).
Applies to POST1 selects, sorts, and output [NSEL, NSORT, PRNSOL, PLNSOL, PRPATH, PLPATH, etc.], and
is used for storage with the POST26 ESOL command. For example, SHELL,TOP causes item S of the POST1
PRNSOL command or the POST26 ESOL command to be the stresses at the top of the shell elements. For
layered shell elements, use the LAYER (POST1) or LAYERP26 (POST26) command to select the layer. The
SHELL command does not apply to the layered thermal shell elements, SHELL131 and SHELL132. For
PowerGraphics [/GRAPHICS,POWER], the SHELL,MID command affects both the printed output and the
displayed results, while the SHELL (TOP or BOT) command prints and displays both the top and bottom
layers simultaneously.
In POST26, the ESOL data stored is based on the active SHELL specification at the time the data is stored.
To store data at various specifications (for example, stresses at the top and bottom locations), issue a STORE
command before each new specification.
Menu Paths
Main Menu>General Postproc>Options for Outp
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
Utility Menu>List>Results>Options
/SHOW, Fname, Ext, VECT, NCPL
Specifies the device and other parameters for graphics displays.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
Device name, filename, or keyword, as listed below:
<device name>
Any valid graphics display device name (e.g., X11, 3-D etc.). Defaults to X11 for most systems. See
Getting Started with Graphics in the Basic Analysis Guide for details. A device name must be defined
before activating the Graphical User Interface (GUI). Once the GUI is activated, the device name
cannot be changed for that ANSYS session, except for switching between X11 and X11C.
<filename>
Name of graphics file to which graphics displays are to be diverted (248 characters maximum). Should
not be the same as a valid device name or any other Fname option. Plots are written to the file
Filename.Ext (or just Filename.Ext if Ext is left blank) in the working directory. This file can
be appended only during the same session; reissuing the same filename in a new session will overwrite
existing filenames from previous sessions. Although multiple filenames can be used within one session,
1565
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/SHOW
only the last file created or accessed will be appended. Issuing /SHOW, CLOSE, or starting a new
session will prevent access to any previously created files.
TERM
Graphics displays are switched back to the last-specified device name.
CLOSE
This option purges the graphics file buffer. The CLOSE option should be issued any time you are
changing graphics devices or file output types during an ANSYS session. Graphics displays are
switched back to the last-specified device name, and any open graphics files are closed. The CLOSE
option is similar to the TERM option, however, with the CLOSE option, another process, such as the
DISPLAY program, can access the data in the graphics file. The CLOSE option causes graphics file
buffers to be flushed to the graphics file.
FILE
Graphics displays are switched back to the last-specified file name.
OFF
Graphics display requests are ignored.
(blank)
If blank in interactive mode, graphics will be displayed on screen as requested by display commands
(no file written); If blank in batch mode, graphics data will be written to Jobname.GRPH.
PSCR
Creates PostScript graphic files that are named Jobnamennn.eps, where nnn is a numeric value
that is incremented by one as each additional file is created; that is, Jobname000.eps, Job-
name001.eps, Jobname002.eps, and so on . (See the PSCR command for options.) Ignores the
Ext and NCPL fields.
HPGL
Creates Hewlett-Packard Graphics Language files that are named Jobnamennn.hpgl, where nnn
is a numeric value that is incremented by one as each additional file is created; that is, Job-
name000.hpgl, Jobname001.hpgl, Jobname002.hpgl, and so on. (See the HPGL command
for options.) Ignores the Ext and NCPL fields.
HPGL2
Creates Hewlett-Packard Graphics Language files that are named Jobnamennn.hpgl, where nn is
a numeric value that is incremented by one as each additional file is created; that is, Jobname000.hp-
gl, Jobname001.hpgl, Jobname02.hpgl, and so on. The HPGL2 files have enhanced color. (See
the HPGL command for options.) Ignores the Ext field.
JPEG
Creates JPEG files that are named Jobnamennn.jpg, where nnn is a numeric value that is incre-
mented by one as each additional file is created; that is, Jobname000.jpg, Jobname001.jpg,
Jobname002.jpg, and so on. Ignores the Ext field.
TIFF
Creates tagged image format files that are named Jobnamennn.tif, where nnn is a numeric value
that is incremented by one as each additional file is created; that is, Jobname000.tif, Job-
name001.tif, Jobname002.tif, and so on. This value for the Fname argument ignores the Ext
field. (See the TIFF command for options.)
PNG
Creates PNG (Portable Network Graphics) files that are named Jobnamennn.png, where nnn is a
numeric value that is incremented by one as each additional file is created; that is, Jobname000.png,
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1566
/SHOW
Jobname001.png, Jobname002.png, and so on. This value for the Fname argument ignores the
Ext field. (See the PNGR command for options.)
VRML
Creates Virtual Reality Meta Language files named Jobname000.wrl that can be displayed on 3-
D Internet web browsers. Ignores the Ext and NCPL fields.
Ext
Filename extension (8 character maximum).
VECT
Specifies raster or vector display mode. This affects area, volume, and element displays, as well as geo-
metric results displays such as contour plots. See the /DEVICE command for an alternate way to toggle
between raster and vector mode. Changing VECT also resets the /TYPE command to its default.
0
Raster display (color filled entities; default)
1
Vector display (outlined entities; i.e., "wireframe")
NCPL
Sets the number of color planes (4 to 8). Default is device-dependent. NCPL is not supported by all
graphics devices.
Command Default
For interactive runs, display is shown on the screen; for batch runs, display commands are ignored (graphics
file not written).
Notes
Specifies the device to be used for graphics displays, and specifies other graphics display parameters. Display
may be shown at the time of generation (for interactive runs at a graphics display terminal) or diverted to
a file for later processing with the DISPLAY program. Issue /PSTATUS for display status.
Batch runs do not have access to the fonts available on your system. The Courier and Helvetica font files
used for JPEG, PNG and TIFF batch output are copyrighted by Adobe Systems Inc. and Digital Equipment
Corp. Permission to use these trademarks is hereby granted only in association with the images described
above. Batch run JPEG output is produced at the default quality index value of 75, unless specified otherwise.
Interactive displays will default to eight color planes (NCPL = 8) for most monitors, while graph file output
will default to eight color planes for VRML output, and four color planes for PSCR, HPGL, HPGL2, JPEG, PNG,
TIFF and FILE33.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Device Options
Utility Menu>PlotCtrls>Redirect Plots>To GRPH File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL File
Utility Menu>PlotCtrls>Redirect Plots>To HPGL2 File
Utility Menu>PlotCtrls>Redirect Plots>To PSCR File
Utility Menu>PlotCtrls>Redirect Plots>To Screen
1567
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/SHOW
/SHOWDISP, Dname, --, --, NCPL
Defines the display driver name.
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Dname
Valid driver name (see Getting Started with Graphics in the Basic Analysis Guide for details):
<device name>
Any linked terminal driver (such as X11, TEKTRONIX, etc.)
HPGL
Hewlett-Packard Graphics Language
HPGL2
Hewlett-Packard Graphics Language with enhanced color. (See the HPGL command for options.)
Ignores the NCPL field.
INTERLEAF
Interleaf ASCII Format, OPS Version 5.0
POSTSCRIPT
PostScript, Version 1.0 Minimally Conforming
DUMP
ASCII Text Dump
--, --
Unused fields.
NCPL
Number of color planes (4 to 8). Default is device-dependent.
Menu Paths
It is part of the DISPLAY command.
SHPP, Lab, VALUE1, VALUE2
Controls element shape checking.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Shape checking option. (When Lab = WARN, STATUS, SUMMARY, or DEFAULT, the remaining arguments
are ignored.)
ON
Activates element shape checking. New elements, regardless of how they are created, are tested
against existing warning and error limits. (The existing limits may be the default limits, or previously
modified limits.) Elements that violate error limits produce error messages and either (a) cause a
meshing failure, or (b) for element creation or storage other than AMESH or VMESH, are not stored.
Elements that violate warning limits produce warning messages. If shape checking was previously
turned off [SHPP,OFF] and you turn it on, existing elements are marked as untested; use the CHECK
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1568
/SHOWDISP
command to retest them. With this option, you may also specify a value for VALUE1 to turn individual
shape tests on. If you do not specify a value for VALUE1, all shape tests are turned on.
WARN
Activates element shape checking; however, in contrast to SHPP,ON, elements that violate error
limits do not cause either a meshing or element storage failure. Instead, they produce warning
messages to notify you that error limits have been violated. This option does not alter current shape
parameter limits. Since the default shape parameter error limits are set to allow almost any usable
element, the elements this option allows, which would otherwise be forbidden, are likely to be very
poorly shaped.
OFF
Deactivates element shape checking. This setting does not alter current shape parameter limits. Use
of this option is risky, since poorly shaped elements can lead to analysis results that are less accurate
than would otherwise be expected for a given mesh density. With this option, you may also specify
a value for VALUE1 to turn individual shape tests off. If you do not specify a value for VALUE1, all
element shape tests are turned off.
SILENT
Determines whether element shape checking runs in silent mode. In silent mode, ANSYS checks
elements without issuing warnings, with the exception of the generic warnings that it issues at solution.
With this option, you must also specify a value for VALUE1 (During the execution of certain com-
mands, ANSYS automatically runs element shape checking in silent mode, then internally summarizes
the shape test results for all of the new or modified elements. ANSYS does this when it executes any
of the following commands: AGEN, AMESH, AREFINE, ARSYM, ATRAN, CDREAD, EGEN, ENGEN,
ENSYM, EREAD, EREFINE, ESYM, ET, FVMESH, KREFINE, LREFINE, NREFINE, TIMP, VEXT, VGEN,
VIMP, VMESH, VOFFST, VROTAT, VSWEEP, VSYMM, and VTRAN.)
STATUS
Lists the shape parameter limits currently in effect, along with status information about element
shape checking (for example, whether any individual shape tests are off, whether any of the shape
parameter limits have been modified, and so on).
SUMMARY
Lists a summary of element shape test results for all selected elements.
DEFAULT
Resets element shape parameter limits to their default values. Also, if any individual tests were turned
off, turns them back on. (The SHPP,DEFAULT command may be useful if any parameter limits were
previously altered by using the MODIFY option.)
OBJECT
Determines whether element shape test results data is stored in memory. When this option is turned
on, an "object" is created for storing test results in memory. When this option is turned off, no object
is created and no data is stored; thus, any operation that requires shape parameters for an existing
element (such as use of the CHECK command) causes the shape parameters to be recomputed.
(Note the distinction between storing the data in memory and storing it in the database; regardless
of whether this option is turned on or off, no element shape test results data will be stored in the
database. The element shape parameter object is deleted automatically before any solution.) This
setting is independent of shape checking status, with one exception--if shape checking is turned off
[SHPP,OFF], the object is not created. Keep in mind that recomputing shape parameters is more
computationally expensive than retrieving them from the object. With this option, you must also
specify a value for the VALUE1 argument; the VALUE2 argument is ignored.
1569
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SHPP
LSTET
Determines, for Jacobian ratio tests, whether sampling is done at integration points (DesignSpace
product method), or at corner nodes. When this option is turned on, sampling is done at integration
points, and the default limits for h-element Jacobian ratios are a warning tolerance of 10 and an error
tolerance of 40. When this option is turned off, sampling is done at corner nodes, and the corres-
ponding default limits are a warning tolerance of 30 and an error tolerance of 1000. Sampling at the
integration points (option on) results in a lower Jacobian ratio, but that ratio is also subjected to a
more restrictive error limit. Some elements that have passed the integration point sampling criterion,
have failed the corner mode sampling criterion. Because of this, use integration point sampling only
for simple linear analyses. For other types of analyses (e.g., nonlinear, electromagnetic), use sampling
at corner nodes, which is the more conservative approach. With this option, you must also specify
a value for the VALUE1 argument; the VALUE2 argument is ignored.
MODIFY
Indicates that you want to respecify a shape parameter limit. With this option, you must also specify
values for the VALUE1 and VALUE2 arguments.
FLAT
Determines the warning and error limits used to test elements that may exhibit nonzero/nonconstant
Z coordinates. With this option, you must also specify values for the VALUE1 and/or VALUE2 argu-
ments.
VALUE1
Valid for the ON, OFF, SILENT, OBJECT, LSTET, MODIFY, and FLAT options only. When Lab = ON or OFF,
use VALUE1 to individually control (that is, turn off or turn on) specific element shape tests. Thus,
VALUE1 can be ANGD (SHELL28 corner angle deviation tests), ASPECT (aspect ratio tests), PARAL (deviation
from parallelism of opposite edges tests), MAXANG (maximum corner angle tests), JACRAT (Jacobian
ratio tests), WARP (warping factor tests), or ALL (all tests). When Lab = SILENT, VALUE1 can be ON (to
turn silent mode on) or OFF (to turn silent mode off ). When Lab = OBJECT, VALUE1 can be either 1,
YES, or ON to turn on storage of element shape test data (the default); or it can be 0, NO, or OFF to turn
off storage of element shape test data (delete the data and recompute as necessary). When Lab = LSTET,
VALUE1 can be either 1, YES, or ON to choose Jacobian sampling at integration points; or it can be 0,
NO, or OFF to choose Jacobian sampling at nodes (the default). When Lab = MODIFY, VALUE1 is the
numeric location (within the shape parameter limit array) of the shape parameter limit to be modified.
Locations are identified in the element shape checking status listing [SHPP,STATUS]. For more information,
see the examples in the Notes section. When Lab = FLAT, VALUE1 is the warning limit for XY element
constant Z sets performed at CHECK or SOLVE. The default is 1.0e-8.
VALUE2
Valid for the MODIFY and FLAT options only. When Lab = MODIFY, specifies the new limit for the shape
parameter that is in the location indicated by the VALUE1 argument. See the examples in the Notes
section. When Lab = FLAT, VALUE2 is the error limit. The default is 1.0e-2.
Command Default
All shape checking tests are on [SHPP,ON,ALL] with default shape parameter limits. Silent mode is off.
Memory object storage of element shape parameters is on.
Notes
The following examples illustrate how to use the SHPP,MODIFY,VALUE1,VALUE2 command to respecify
shape parameter limits. Assume that you issued the SHPP,STATUS command, and you received the output
below:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1570
SHPP
ASPECT RATIO (EXCEPT FLOTRAN OR EMAG)
QUAD OR TRIANGLE ELEMENT OR FACE
WARNING TOLERANCE ( 1) = 20.00000
ERROR TOLERANCE ( 2) = 1000000.
.
.
.
MAXIMUM CORNER ANGLE IN DEGREES (EXCEPT FLOTRAN OR EMAG)
TRIANGLE ELEMENT OR FACE
WARNING TOLERANCE (15) = 165.0000
ERROR TOLERANCE (16) = 179.9000
Notice that in the sample output, the warning tolerance for aspect ratios is set to 20. Now assume that you
want to "loosen" this shape parameter limit so that it is less restrictive. To allow elements with aspect ratios
of up to 500 without causing warning messages, you would issue this command:
SHPP,MODIFY,1,500
Also notice that each shape parameter's numeric location within the shape parameter limit array appears
in the sample output within parentheses. For example, the numeric location of the aspect ratio shape
parameter (for warning tolerance) is 1, which is why "1" is specified for the VALUE1 argument in the example
command above.
Now notice that the sample output indicates that any triangle element with an internal angle that is greater
than 179.9 degrees will produce an error message. Suppose that you want to "tighten" this shape parameter
limit, so that it is more restrictive. To cause any triangle or tetrahedron with an internal angle greater than
170 degrees to produce an error message, you would issue this command:
SHPP,MODIFY,16,170
The existence of badly shaped elements in a model may lead to certain computational errors that can cause
your system to abort during ANSYS solution. Therefore, you run the risk of a system abort during solution
any time that you turn element shape checking off entirely, run shape checking in warning-only mode, turn
off individual shape checks, or loosen shape parameter limits.
Changing any shape parameter limit marks all existing elements as untested; use the CHECK command to
retest them.
Since the shape parameter limit array was completely reorganized at ANSYS 5.4, you should revise any input
files created prior to 5.4 that contain limit changes so that they reflect the reorganized data structure.
For more information about element shape checking, see Meshing Your Solid Model in the Modeling and
Meshing Guide.
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Checking Ctrls>Shape Checking
Main Menu>Preprocessor>Checking Ctrls>Toggle Checks
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Element Shape
Checking>Shape Checking
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Mesh Controls>Element Shape
Checking>Toggle Checks
1571
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SHPP
/SHRINK, RATIO
Shrinks elements, lines, areas, and volumes for display clarity.
GRAPHICS: Scaling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RATIO
Shrinkage ratio (input as a decimal (0.0 to 0.5)). Defaults to 0.0 (no shrinkage). Values greater than 0.5
default to 0.1 (10% shrinkage).
Command Default
Full size entities.
Notes
Shrinks the elements, lines, areas, and volumes so that adjacent entities are separated for clarity. This command
is not valid with p-elements. Portions of this command are not supported by PowerGraphics [/GRAPH-
ICS,POWER].
If only the common lines of non-coplanar faces are drawn (as per the /EDGE command), then this command
is ignored.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Size and Shape
SHSD, RID, Action
Creates or deletes shell-solid interface to be used in shell-to-solid assemblies.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
RID
The real constant set ID that identifies the contact pair on which a shell-to-solid assembly is defined. If
ALL, all selected contact pairs will be considered for assembly.
Action
CREATE
(default) Builds new shell and contact elements to be used in shell-solid assemblies. New elements
are stored as internally-created components.
DELETE
Deletes the nodes and elements created during a previous execution of SHSD,RID,CREATE for the
real constant set identified by RID.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1572
/SHRINK
Notes
The SHSD command creates a shell-solid interface to be used in shell-to-solid assemblies, or deletes a pre-
viously-created shell-solid interface. Virtual shell elements and additional CONTA175 elements are created
at the contact pair identified by RID when Action = CREATE. Set Action = DELETE to remove the generated
nodes and elements at the contact pair identified by RID.
The SHSD command is active only when the following element KEYOPTs of associated CONTA175 and
TARGE170 element types are predefined:
Detail KEYOPT Element
MPC algorithm KEYOPT(2) = 2
CONTA175 bonded contact KEYOPT(12) = 5, 6
contact normal perpendicular to target surface KEYOPT(4) = 0
types of constraints: (1 = solid-solid, 2 = shell-shell) KEYOPT(5) = 1, 2 TARGE170
The method used to build shell and contact elements depends on the KEYOPT(5) setting of the target element
type associated with the real constant set identified with the RID argument. If KEYOPT(5) = 1 (solid-solid
constraint), the virtual shell elements are built perpendicularly to the pre-existing shell elements attached
to the contact elements. They geometrically follow the contact interface edge and are built on both sides
of this interface in such a way that each new shell element (SHELL181) has two nodes that belong to the
associated pre-existing shell element in the shell edge. (See Figure 8: Virtual Shell Elements Following the
Contact Interface Edge (p. 1573).) The width of the new shell elements is half the thickness of the pre-existing
shell element. The CONTA175 elements are then created at each node of the virtual shell elements where
no CONTA175 element exists. The new contact elements are identified by the same contact pair ID as the
pre-existing contact elements.
Figure 8: Virtual Shell Elements Following the Contact Interface Edge
Shell elements
(thickness t)
Contact nodes
Solid elements
Target elements
(top only)
New "virtual"
shell elements
gap
New CONTA175
elements
t
If KEYOPT(5) = 2 (shell-shell constraint), the virtual shell elements (SHELL181: low order; SHELL281: high
order) overlap the existing high or low order target elements identified with the RID argument, and share
their nodes. Only those target elements close enough to the contact interface (identified using the PINB real
constant) are overlapped. ANSYS uses the FTOLN real constant (defaults to half the shell element thickness)
to define an influence distance. The associated virtual shell elements are created only for those target elements
1573
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SHSD
which lie partially inside the influence distance region (see Figure 9: Virtual Shell Elements Overlapping Target
Elements (p. 1574)).
Figure 9: Virtual Shell Elements Overlapping Target Elements
Shell Elements
(thickness t)
Contact nodes
Solid elements
Target elements
(top only)
New "virtual"
shell elements
FTOLN (influence
distance)
For the bonded always option (KEYOPT(12) = 5), any contact node inside the pinball region (gap < PINB) is
included in the KEYOPT(5) = 2 process. A relatively small PINB value may be used to prevent false contact.
PINB defaults to 25% of the contact depth for small deformation analyses.
For the bonded initial option (KEYOPT(12) = 6), only those contact nodes which initially lie inside the adjust-
ment zone (gap < ICONT) are always included in the KEYOPT(5) = 2 process. ICONT defaults to 5% of the
contact depth.
For both processes, the new nodes and elements are stored in internally-named components. The internal
naming convention is based on the real constant set ID specified by RID, as illustrated in the following
table.
SHSD_ND_RID Nodes
SHSD_CN_RID Contact Elements
SHSD_SH_RID Shell Elements
Issuing SHSD,RID,DELETE deletes components based on their generated names. Only components whose
names match the internal naming convention will be deleted.
Caution
Do not rename or manually delete generated components. Use the SHSD command to delete
generated components.
Renaming or manually deleting generated components will cause these components to be ignored
when SHSD,RID,DELETE is executed and when ANSYS searches for these components to verify
if SHSD,RID,CREATE can be safely executed. Manually renaming or deleting generated components
and reissuing SHSD,RID,CREATE may result in erroneous generation of virtual shell or contact
elements.
See Modeling a Shell-Solid Assembly in the Contact Technology Guide for more information.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1574
SHSD
Menu Paths
Main Menu>Preprocessor>Coupling / Ceqn>Shell/Solid Interface
SLIST, SFIRST, SLAST, SINC, Details, Type
Summarizes the section properties for all defined sections in the current session of ANSYS.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SFIRST
First section ID to be summarized. Defaults to first available section in the database.
SLAST
Last section ID to be summarized. Defaults to last available section in the database.
SINC
Increment of the section ID; defaults to 1.
Details
Determines the content of the summarized information for beams and shells.
BRIEF
For beams, lists only the section integrated properties (such as Area, Iyy, and Iyz). This option is the
default.
FULL
For beams, lists the section integrated properties, as well as the section nodal coordinates, section
cell connectivity information, and section cell integration point coordinates. For shells, the section
stiffness (membrane, bending, membrane-bending coupling and transverse shear) are printed.
The shell section stiffness listed considers elastic behavior of materials at reference temperature only.
The elements that use the section data may alter the transverse shear stiffness based on slenderness
considerations (in addition to the shear correction factors shown).
Section stiffness terms listed via the FULL option do not include section offsets. The ANSYS program
considers section offsets during the solution phase of the analysis.
44
You can use section commands with BEAM44 elements instead of defining real constants to describe
the elements. Issuing SLIST,,,,44 lists the section integrated properties, as well as the real constant
commands that are equivalent to your section command specifications. A section made of multiple
materials cannot be modeled using BEAM44.
This option applies only to beam sections.
GROUP
If a section calls other sections, this option lists those sections too.
Type
The section type. Valid arguments are ALL (the default) and the types available on the SECTYPE command.
1575
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SLIST
Notes
For sections integrating FiberSIM data (SECTYPE,,SHELL,FIBERSIM), the SLIST command output displays the
relevant layer information.
Because the actual section can vary from element to element, the information provided by the Details
= FULL option is not available when using the FiberSIM-ANSYS interface.
By default, the command lists information concerning all sections; however, you can limit the output to only
beam or pretension sections via the Type key.
Sample Output
Following is sample output from the SLIST,,,,BRIEF command for a rectangular beam section subtype (SEC-
TYPE,,BEAM,RECT):
LIST SECTION ID SETS 1 TO 1 BY 1
SECTION ID NUMBER: 1
BEAM SECTION TYPE: Rectangle
BEAM SECTION NAME IS:
BEAM SECTION DATA SUMMARY:
Area = 6.0000
Iyy = 4.5000
Iyz = 0.11281E-15
Izz = 2.0000
Warping Constant = 0.23299
Torsion Constant = 4.7330
Center of Gravity Y =-0.30973E-16
Center of Gravity Z = 0.15376E-15
Shear Center Y =-0.22957E-13
Shear Center Z = 0.31281E-13
Beam Section is offset to CENTROID of cross section
Following is sample output from the SLIST command for a FiberSIM subtype (SECTYPE,,SHELL,FIBERSIM):
LIST SECTION ID SETS 1 TO 4 BY 1
INPUT SECTION ID NUMBER 4
INPUT SECTION IS SHELL SECTION TYPE USING FIBERSIM DATA
INPUT SHELL SECTION NAME lamin1
Thickness MatID Num Intg. Pts Layer ID
0.2000 1 0 STEEL
1.0000 2 0 ALUM1
0.2000 3 0 ALUM2
This table is a summary of the SECDATA command input and describes the individual
plies/layers and does not normally represent the actual layup used. Layup data is
available in FiberSIM.
Shell Section is offset to BOTTOM surface of Shell
Section Solution Controls
Tolerance in Thickness Direction = 0.15500
Angle Tolerance = 15.000
Added Mass Per Unit Area = 5.00000E-02
Elastic Foundation Stiffness = 100.00
Parsing Debug = 1
Computed Real Constants = 1
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1576
SLIST
Menu Paths
Main Menu>Preprocessor>Sections>List Sections
SLOAD, SECID, PLNLAB, KINIT, KFD, FDVALUE, LSLOAD, LSLOCK
Load a pretension section.
PREP7: Cross Sections
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
SECID
Unique section number. The number must already be assigned to a pretension section.
PLNLAB
Label representing the pretension load sequence number in the format "PLnn" where nn is an integer
from 1 through 99 (for example, PL01 through PL99).
Specify a value of DELETE to delete all loads on the specified pretension section (SECID). In this case,
the command ignores any other argument values.
KINIT
Initial action key for pretension load PL01. (This field is omitted for PL02 and up.) Three scenarios are
possible:
LOCK
Constrains (connects) the cutting plane on the pretension section. This value is the default.
SLID
Unconstrains (disconnects) the cutting plane on the pretension section.
TINY
Applies a very small pretension load (0.1% of FDVALUE) before the desired load is established. The
small load prevents convergence problems which can occur when the desired load is not established
in the first load step. This value is valid only if KFD = FORC.
KFD
Force/Displacement key. Specifies whether FDVALUE is a force or a displacement:
FORC
Apply a force on the specified pretension section. This value is the default.
DISP
Apply a displacement (adjustment) on the specified pretension section.
FDVALUE
Pretension load value. If KFD = FORC, this value is a pretension force. If KFD = DISP, this value is a pre-
tension displacement (adjustment).
LSLOAD
Load step in which to apply the FDVALUE.
LSLOCK
The load step in which the displacement value resulting from the pretension force is locked. This value
is valid only if KFD = FORC.
1577
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SLOAD
Command Default
The default pretension load value FDVALUE is zero (no load). A positive value puts the pretension elements
in tension.
No default exists for the LSLOAD applied load step value. You must specify the load step in which to apply
the FDVALUE.
No default exists for the LSLOCK locked load step value. You must specify the load step in which to lock
the FDVALUE.
Notes
The SLOAD command applies pretension loads to specified pretension sections (created via the PSMESH
command). A pretension load is ramp-applied (KBC = 0) if it is a force (KFD = FORC), and step-applied (KBC
= 1) if it is a displacement (KFD = DISP).
You can lock the load value at a specified load step. When locked, the load changes from a force to a dis-
placement, and ANSYS applies the load as a constant displacement in all future load steps. Locking is useful
when applying additional loadings. The additional loadings alter the effect of the initial load value, but because
locking transforms the load into a displacement, it preserves the initial load's effect.
Example 1 Applying a Load
The following command shows how to establish loads on a pretension section:
SLOAD,1,PL01,TINY,FORC,5000,2,3
In this example, the load is applied to pretension section 1, and the sequence begins with the initial action
key, KINIT, set to TINY. A small stabilization load (5 = 0.10% of 5000) is applied in the first load step, as
the actual pretension force is not applied until the second load step. The next four fields set the actual load:
the KFD value FORC specifies the type of load, FDVALUE defines the pretension load value (5000), LSLOAD
specifies the load step in which the force is applied (2), and the LSLOCK field specifies the load step in
which the force is locked (3). Additional sets of four fields can be used to define additional loads.
Example 2 Editing an Existing Load
You can use the SLOAD command to edit (overwrite) existing loads on a pretension section. This example
changes the load on pretension section 1 (set above) to 6000:
SLOAD,1,PL01,,,6000,2,3
Unspecified values (blank fields), as shown in this example, remain unchanged from prior settings. If no
prior specifications exist, then default values (KINIT = LOCK and KFD = FORC) apply.
Example 3 Deleting All Loads
The command can also delete all loads on a specified pretension section, as shown here:
SLOAD,1,DELETE
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1578
SLOAD
Example 4 Locking a Pretension Element
For a prestressed modal analysis, this command locks the pretension element:
SLOAD,1,PL01,LOCK,DISP,0,1,2
Multiple Loadings
The SLOAD command allows you to apply multiple loadings. You can add up to 15 loadings (PL01 through
PL15), or delete loadings, for any given pretension section(s).
Example 5 Applying Multiple Loadings
The following SLOAD commands, issued in the order shown, establish a pretension load sequence in pre-
tension section 2 with a force of 25 in load step (LS) 2, locked in LS 3-6, a force of 50 in LS 7, locked in LS
8-11, a force of 75 in LS 12, locked in LS 13 and beyond:
SLOAD,2,PL01,LOCK,FORC,25,2,3
SLOAD,2,PL02,,FORC,50,7,8
SLOAD,2,PL03,,FORC,75,12,13
At the same time, you can issue SLOAD commands to apply loads on other pretension sections. For example,
in addition to the commands listed above, you could issue the following command to apply a load on pre-
tension section 3:
SLOAD,3,PL01,LOCK,FORC,25,3,4
Using the Pretension Section Loads GUI
Any addition or deletion of a loading applies to the selected sections only. ANSYS does not apply or delete
a load until you click on the Apply or OK button.
After you have successfully solved for a specified LSLOAD (GUI field Apply at LS) and eventually LSLOCK
(GUI field Lock at LS) value, you cannot modify that loading's settings during subsequent steps of the ana-
lysis. Similarly, you cannot delete loadings that you have already partially or completely solved.
You can select more than one pretension section at a time in order to specify identical loadings on them.
Before you completely solve a given loading, any combination of pretension sections is valid. The following
limitations apply:
After you have completely solved one or more loadings, ANSYS allows multiple selection of only those
pretension sections having
-- the same number of defined loadings, and
-- the identical loading number from the most recent completely solved loading.
A multiple selection meeting the necessary criteria retains the settings that are identical for all selected
pretension sections and leaves all other fields blank.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Pretnsn Sectn
1579
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SLOAD
Main Menu>Preprocessor>Loads>Define Loads>Delete>All Load Data>All Section Lds
Main Menu>Solution>Define Loads>Apply>Structural>Pretnsn Sectn
Main Menu>Solution>Define Loads>Delete>All Load Data>All Section Lds
SLPPLOT, Prefer, VALUE
Displays line loops smaller than a specified size (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Prefer
Preference for line loop display. If Prefer = FACTOR, the command displays all line loops whose radius
is smaller than the radius of the model (taken between the two most distant keypoints) times VALUE.
This is the default preference. If Prefer = RADIUS, the command displays all line loops whose radius
is smaller than that specified by VALUE.
VALUE
Numeric argument for Prefer.
Notes
Use this command to locate and display disproportionately small line loops when repairing the geometry
of models imported from CAD files. Line loops matching the criteria specified in Prefer and VALUE both
display in a different color and include their IDs. This command is available only for models imported through
the Default IGES option.
Menu Paths
Main Menu>Preprocessor>Modeling>Simplify>Detect/Display>Small Loops
SLSPLOT, Prefer, VALUE
Displays line segments smaller than a specified size (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Prefer
Preference for line segment display. If Prefer = FACTOR, the command displays all line segments whose
length is smaller than the average length of lines within the model times VALUE. This is the default
preference. If Prefer = LENGTH, the command displays all line lengths smaller than that specified by
VALUE.
VALUE
Numeric argument for Prefer.
Notes
Use this command to locate and display disproportionately small line segments when repairing the geometry
of models imported from CAD files. Line lengths matching the criteria specified in Prefer and VALUE both
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1580
SLPPLOT
display in a different color and include their IDs. This command is available only for models imported through
the Default IGES option.
Menu Paths
Main Menu>Preprocessor>Modeling>Simplify>Detect/Display>Small Lines
SMALL, IR, IA, IB, IC, Name, --, --, FACTA, FACTB, FACTC
Finds the smallest of three variables.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA, IB, IC
Reference numbers of the three variables to be operated on. If only two, leave IC blank. If only one,
leave IB blank also.
Name
Thirty-two character name identifying the variable on printouts and displays. Embedded blanks are
compressed for output.
--, --
Unused fields.
FACTA, FACTB, FACTC
Scaling factors (positive or negative) applied to the corresponding variables (defaults to 1.0).
Notes
Finds the smallest of three variables according to the operation:
IR = smallest of (FACTA x IA, FACTB x IB, FACTC x IC)
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Find Minimum
SMAX, LabR, Lab1, Lab2, FACT1, FACT2
Forms an element table item from the maximum of two other items.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabR
Label assigned to results. If same as existing label, the existing values will be overwritten by these results.
1581
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SMAX
Lab1
First labeled result item in operation.
Lab2
Second labeled result item in operation (may be blank).
FACT1
Scale factor applied to Lab1. A (blank) or '0' entry defaults to 1.0.
FACT2
Scale factor applied to Lab2. A (blank) or '0' entry defaults to 1.0.
Notes
Forms a labeled result item (see ETABLE command) for the selected elements by comparing two existing
labeled result items according to the operation:
LabR = (FACT1 x Lab1) cmx (FACT2 x Lab2)
where "cmx" means "compare and save maximum." If absolute values are requested [SABS,1], the absolute
values of Lab1 and Lab2 are used.
Menu Paths
Main Menu>General Postproc>Element Table>Find Maximum
/SMBC, Mode
Controls the display of solid model boundary condition symbols and labels.
DATABASE: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Mode
CENT
Solid model boundary condition symbols and labels appear at the centroid of the solid model entity
(default).
TESS
Solid model boundary condition symbols and labels appear inside each constituent element of the
tesselation.
Notes
Mode = CENT is designed to reduce the clutter of boundary condition symbols in solid model plots. For ex-
ample, if you have assigned normal pressure loads to an area, you may choose to display the pressures as
arrows with the /PSF command using /PSF,PRES,NORM,2. When Mode = CENT, the pressure arrow is displayed
at the centroid of the area. When Mode = TESS, a pressure arrow is displayed at the centroid of each polygon
of the area's tesselation.
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1582
/SMBC
Menu Paths
This command cannot be accessed from a menu.
SMBODY
Specifies "Body loads on the solid model" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Body Loads
SMCONS
Specifies "Constraints on the solid model" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>DOF Constraints
1583
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SMCONS
SMFOR
Specifies "Forces on the solid model" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Forces
SMIN, LabR, Lab1, Lab2, FACT1, FACT2
Forms an element table item from the minimum of two other items.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabR
Label assigned to results. If same as existing label, the existing values will be overwritten by these results.
Lab1
First labeled result item in operation.
Lab2
Second labeled result item in operation (may be blank).
FACT1
Scale factor applied to Lab1. A (blank) or '0' entry defaults to 1.0.
FACT2
Scale factor applied to Lab2. A (blank) or '0' entry defaults to 1.0.
Notes
Forms a labeled result item (see ETABLE command) for the selected elements by comparing two existing
labeled result items according to the operation:
LabR = (FACT1 x Lab1) cmn (FACT2 x Lab2)
where "cmn" means "compare and save minimum." If absolute values are requested [SABS,1], the absolute
values of Lab1 and Lab2 are used.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1584
SMFOR
Menu Paths
Main Menu>General Postproc>Element Table>Find Minimum
SMOOTH, Vect1, Vect2, DATAP, FITPT, Vect3, Vect4, DISP
Allows smoothing of noisy data and provides a graphical representation of the data.
POST26: Special Purpose
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Vect1
Name of the first vector that contains the noisy data set (i.e., independent variable). You must create
and fill this vector before issuing SMOOTH.
Vect2
Name of the second vector that contains the dependent set of data. Must be the same length as the
first vector. You must create and fill this vector before issuing SMOOTH.
DATAP
Number of data points to be fitted, starting from the beginning of the vector. If left blank, the entire
vector will be fitted. The maximum number of data points is 100,000 (or greater, depending on the
memory of the computer).
FITPT
Order of the fitting curve that will be used as a smooth representation of the data. This number should
be less than or equal to the number of the data points. Default (blank) is one-half the number of data
points. Maximum number of smoothed data fitting order is the number of data points up to 50. Depending
on this number, the smoothed curve will be one of the following:
1
Curve is the absolute average of all of the data points.
2
Curve is the least square average of all of the data points.
3 or more
Curve is a polynomial of the order (n-1), where n is the number of data fitting order points.
Vect3
Name of the vector that contains the smoothed data of the independent variable. This vector should
have a length equal to or greater than the number of smoothed data points. In batch (command) mode,
you must create this vector before issuing the SMOOTH command. In interactive mode, the GUI auto-
matically creates this vector (if it does not exist). If you do not specify a vector name, the GUI will name
the vector smth_ind.
Vect4
Name of the vector that contains the smoothed data of the dependent variable. This vector must be the
same length as Vect3. In batch (command) mode, you must create this vector before issuing the
SMOOTH command. In interactive mode, the GUI automatically creates this vector (if it does not exist).
If you do not specify a vector name, the GUI will name the vector smth_dep.
DISP
Specifies how you want to display data. No default; you must specify an option.
1585
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SMOOTH
1
Unsmoothed data only
2
Smoothed data only
3
Both smoothed and unsmoothed data
Notes
You can control the attributes of the graph using standard ANSYS controls (/GRID, /GTHK, /COLOR, etc.). If
working interactively, these controls appear in this dialog box for convenience, as well as in their standard
dialog boxes. You must always create Vect1 and Vect2 (using *DIM) and fill these vectors before
smoothing the data. If you're working interactively, ANSYS automatically creates Vect3 and Vect4, but if
you're working in batch (command) mode, you must create Vect3 and Vect4 (using *DIM) before issuing
SMOOTH. Vect3 and Vect4 are then filled automatically by ANSYS. In addition, ANSYS creates an additional
TABLE type array that contains the smoothed array and the unsmoothed data to allow for plotting later with
*VPLOT. Column 1 in this table corresponds to Vect1, column 2 to Vect2, and column 3 to Vect4. This
array is named Vect3_SMOOTH, up to a limit of 32 characters. For example, if the array name is X1, the
table name is X1_SMOOTH.
This command is also valid in PREP7 and SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>Replace vs Add>Smooth Data
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Smooth Data
Main Menu>Solution>Define Loads>Settings>Replace vs Add>Smooth Data
Main Menu>Solution>Loading Options>Smooth Data
Main Menu>TimeHist Postpro>Smooth Data
SMRTSIZE, SIZLVL, FAC, EXPND, TRANS, ANGL, ANGH, GRATIO, SMHLC, SMANC, MXITR, SPRX
Specifies meshing parameters for automatic (smart) element sizing.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SIZLVL
Overall element size level for meshing. The level value controls the fineness of the mesh. (Any input in
this field causes remaining arguments to be ignored.) Valid inputs are:
n
Activate SmartSizing and set the size level to n. Must be an integer value from 1 (fine mesh) to 10
(coarse mesh). Remaining arguments are ignored, and argument values are set as shown in
Table 256: SMRTSIZE - Argument Values for h-elements and p-elements (p. 1588).
STAT
List current SMRTSIZE settings.
DEFA
Set all SMRTSIZE settings to default values (as shown in Table 256: SMRTSIZE - Argument Values for
h-elements and p-elements (p. 1588) for size level 6).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1586
SMRTSIZE
OFF
Deactivate SmartSizing. Current settings of DESIZE will be used. To reactivate SmartSizing, issue
SMRTSIZE,n.
FAC
Scaling factor applied to the computed default mesh sizing. Defaults to 1 for h-elements (size level 6),
which is medium. Values from 0.2 to 5.0 are allowed.
EXPND
Mesh expansion (or contraction) factor. (This factor is the same as MOPT,EXPND,Value.) EXPND is used
to size internal elements in an area based on the size of the elements on the area's boundaries. For ex-
ample, issuing SMRTSIZE,,,2 before meshing an area will allow a mesh with elements that are approxim-
ately twice as large in the interior of an area as they are on the boundary. If EXPND is less than 1, a mesh
with smaller elements on the interior of the area will be allowed. EXPND should be greater than 0.5 but
less than 4. EXPND defaults to 1 for h-elements (size level 6), which does not allow expansion or contrac-
tion of internal element sizes (except when using AESIZE element sizing). (If EXPND is set to zero, the
default value of 1 will be used.) The actual size of the internal elements will also depend on the TRANS
option or upon AESIZE or ESIZE sizing, if used.
TRANS
Mesh transition factor. (This factor is the same as MOPT,TRANS,Value.) TRANS is used to control how
rapidly elements are permitted to change in size from the boundary to the interior of an area. TRANS
defaults to 2.0 for h-elements (size level 6), which permits elements to approximately double in size from
one element to the next as they approach the interior of the area. (If TRANS is set to zero, the default
value will be used.) TRANS must be greater than 1 and, for best results, should be less than 4. The actual
size of the internal elements will also depend on the EXPND option or upon AESIZE or ESIZE sizing, if
used.
ANGL
Maximum spanned angle per lower-order element for curved lines. Defaults to 22.5 degrees per element
(size level 6). This angle limit may be exceeded if the mesher encounters a small feature (hole, fillet, etc.).
(This value is NOT the same as that set by DESIZE,,,,ANGL.) This option does not apply to p-element
meshes.
ANGH
Maximum spanned angle per higher-order element for curved lines. Defaults to 30 degrees per element
(size level 6). This angle limit may be exceeded if the mesher encounters a small feature (hole, fillet, etc.).
(This value is NOT the same as that set by DESIZE,,,,,ANGH.)
GRATIO
Allowable growth ratio used for proximity checking. Defaults to 1.5 for h-elements (size level 6). Values
from 1.2 to 5.0 are allowed; however, values from 1.5 to 2.0 are recommended.
SMHLC
Small hole coarsening key, can be ON (default for size level 6) or OFF. If ON, this feature suppresses
curvature refinement that would result in very small element edges (i.e., refinement around small features).
SMANC
Small angle coarsening key, can be ON (default for all levels) or OFF. If ON, this feature restricts proximity
refinement in areas where it is ill-advised (that is, in tight corners on areas, especially those that approach
0 degrees).
MXITR
Maximum number of sizing iterations (defaults to 4 for all levels).
1587
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SMRTSIZE
SPRX
Surface proximity refinement key, can be off (SPRX = 0, which is the default for all levels) or on via two
different values (SPRX = 1 or SPRX = 2). If SPRX = 1, surface proximity refinement is performed and
any shell elements that need to be modified are modified. If SPRX=2, surface proximity refinement is
performed but no shell elements are altered.
Command Default
SmartSizing is off and DESIZE is used for automatic element sizing.
Notes
If a valid level number (1 (fine) to 10 (coarse)) is input on SIZLVL, inputs for remaining arguments are ignored,
and the argument values are set as shown in Table 256: SMRTSIZE - Argument Values for h-elements and p-
elements (p. 1588).
Note
The settings are different for h-elements and p-elements (indicated by the "h" and "p" headings
in the table). The first column contains SIZLV data, ranging from 10 (coarse) to 1 (fine); 6 is the
default.
Table 256 SMRTSIZE - Argument Values for h-elements and p-elements
SPRX MXITR SMANC SMHLC GRA-
TIO
ANGH ANGL TRANS EXPND FAC
p h p h p h p h p h p h p h p h p h p h
off off 4 4* on on on on 2.0 2.0 30.0 45.0* - 45.0 2.0 2.0* 2.5 2.0 5.0 5.0 10
off off 4 4* on on on on 1.9 1.9 30.0 45.0* - 36.0 2.0 2.0* 2.35 1.75 4.0 3.0 9
off off 4 4* on on on on 1.8 1.8 30.0 45.0* - 30.0 2.0 2.0* 2.25 1.5 3.0 1.875 8
off off 4 4* on on on on 1.7 1.7 30.0 36.0* - 26.0 2.0 2.0* 1.7 1.0 2.5 1.5 7
off off 4 4* on on on on 1.7 1.5* 30.0 30.0* - 22.5 2.0 2.0* 1.5 1.0* 1.875 1.0* 6*
off off 4 4* on on on on 1.6 1.5 30.0 27.0 - 18.0 2.0 2.0* 1.25 1.0* 1.5 0.65 5
off off 4 4* on on on off 1.5 1.5 30.0 22.0 - 15.0 2.0 2.0* 1.0 1.0* 1.0 0.4 4
off off 4 4* on on off off 1.5 1.5 22.0 18.0 - 12.0 2.0 2.0* 1.0 1.0* 0.8 0.3 3
off off 4 4* on on off off 1.5 1.5 18.0 15.0 - 10.0 2.0 2.0* 1.0 1.0* 0.6 0.25 2
off off 4 4* on on off off 1.4 1.4 15.0 15.0 - 7.5 2.0 2.0* 1.0 1.0* 0.4 0.2 1
*
Default
Where appropriate, SmartSizing will start with AESIZE settings. Elsewhere, it will start with any defined ES-
IZE,SIZE setting. It will locally override AESIZE or ESIZE for proximity and curvature. SmartSizing ignores
any ESIZE,,NDIV setting.
LESIZE line division and spacing specifications will be honored by SmartSizing, unless you give permission
for SmartSizing to override them (for proximity or curvature) by setting KYNDIV to 1. Lines not having an
LESIZE specification are meshed as well as they can be.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1588
SMRTSIZE
This command is also valid for rezoning.
Menu Paths
Main Menu>Preprocessor>Meshing>Size Cntrls>SmartSize>Adv Opts
Main Menu>Preprocessor>Meshing>Size Cntrls>SmartSize>Basic
Main Menu>Preprocessor>Meshing>Size Cntrls>SmartSize>Status
SMSURF
Specifies "Surface loads on the solid model" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>Surface Loads
SMULT, LabR, Lab1, Lab2, FACT1, FACT2
Forms an element table item by multiplying two other items.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabR
Label assigned to results. If same as existing label, the existing values will be overwritten by these results.
Lab1
First labeled result item in operation.
Lab2
Second labeled result item in operation (may be blank).
FACT1
Scale factor applied to Lab1. A (blank) or '0' entry defaults to 1.0.
FACT2
Scale factor applied to Lab2. A (blank) or '0' entry defaults to 1.0.
1589
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SMULT
Notes
Forms a labeled result item (see ETABLE command) for the selected elements by multiplying two existing
labeled result items according to the operation:
LabR = (FACT1 x Lab1) x (FACT2 x Lab2)
May also be used to scale results for a single labeled result item. If absolute values are requested [SABS,1],
the absolute values of Lab1 and Lab2 are used.
Menu Paths
Main Menu>General Postproc>Element Table>Multiply
SNOPTION, RangeFact, BlockSize, RobustLev, Compute, --, Solve_Info
Specifies Supernode (SNODE) eigensolver options.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RangeFact
Factor used to control the range of eigenvalues computed for each supernode. The value of RangeFact
must be a number between 1.0 and 5.0. By default the RangeFact value is set to 2.0, which means
that all eigenvalues between 0 and 2*FREQE are computed for each supernode (where FREQE is the
upper end of the frequency range of interest as specified on the MODOPT command). As the RangeFact
value increases, the eigensolution for the SNODE solver becomes more accurate and the computational
time increases.
BlockSize
BlockSize to be used when computing the final eigenvectors. The value of Blocksize must be
either MAX or a number between 1 and NMODE, where NMODE is the number of modes to be computed
as set on the MODOPT command. Input a value of MAX to force the algorithm to allocate enough
memory to hold all of the final eigenvectors in memory and, therefore, only read through the file con-
taining the supernode eigenvectors once. Note that this setting is ONLY recommended when there is
sufficient physical memory on the machine to safely hold all of the final eigenvectors in memory.
RobustLev
Parameter used to control the robustness of the SNODE eigensolver. The value of RobustLev must be
a number between 0 and 10. Lower values of RobustLev allow the eigensolver to run in the most ef-
ficient manner for optimal performance. Higher values of RobustLev often slow down the performance
of the eigensolver, but can increase the robustness; this may be desirable if a problem is detected with
the eigensolver or its eigensolution.
Compute
Key to control which computations are performed by the Supernode eigensolver:
EVALUE
The eigensolver computes only the eigenvalues.
EVECTOR
The eigensolver computes only the eigenvectors (must be preceded by a modal analysis where the
eigenvalues were computed using the Supernode eigensolver).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1590
SNOPTION
BOTH
The eigensolver computes both the eigenvalues and eigenvectors in the same pass (default).
--
Unused field
Solve_Info
Solver output option:
OFF
Turns off additional output printing from the Supernode eigensolver (default).
PERFORMANCE
Turns on additional output printing from the Supernode eigensolver, including a performance summary
and a summary of file I/O for the Supernode eigensolver. Information on memory usage during as-
sembly of the global matrices (that is, creation of the Jobname.FULL file) is also printed with this
option.
Command Default
RangeFact = 2.0. BlockSize is set to min(NMODE,40), where NMODE is the number of modes to be
computed as set on the MODOPT command. RobustLev = 0. Compute = BOTH. Additional output is not
printed (Solve_Info = OFF).
Notes
This command specifies options for the Supernode (SNODE) eigensolver.
Setting RangeFact to a value greater than 2.0 will improve the accuracy of the computed eigenvalues and
eigenvectors, but will often increase the computing time of the SNODE eigensolver. Conversely, setting
RangeFact to a value less than 2.0 will deteriorate the accuracy of the computed eigenvalues and eigen-
vectors, but will often speedup the computing time of the SNODE eigensolver. The default value of 2.0 has
been set as a good blend of accuracy and performance.
The SNODE eigensolver reads the eigenvectors and related information for each supernode from a file and
uses that information to compute the final eigenvectors. For each eigenvalue/eigenvector requested by the
user, the program must do one pass through the entire file that contains the supernode eigenvectors. By
choosing a BlockSize value greater than 1, the program can compute BlockSize number of final eigen-
vectors for each pass through the file. Therefore, smaller values of BlockSize result in more I/O, and larger
values of BlockSize result in less I/O. Larger values of BlockSize also result in significant additional
memory usage, as BlockSize number of final eigenvectors must be stored in memory. The default
Blocksize of min(NMODE,40) is normally a good choice to balance memory and I/O usage.
The RobustLev field should only be used when a problem is detected with the accuracy of the final solution
or if the Supernode eigensolver fails while computing the eigenvalues/eigenvectors. Setting RobustLev
to a value greater than 0 will cause the performance of the eigensolver to deteriorate. If the performance
deteriorates too much or if the eigensolver continues to fail when setting the RobustLev field to higher
values, then switching to another eigensolver such as Block Lanczos or PCG Lanczos is recommended.
Setting Compute = EVALUE causes the Supernode eigensolver to compute only the requested eigenvalues.
During this process a Jobname.SNODE file is written; however, a Jobname.MODE file is not written. Thus,
errors will likely occur in any downstream computations that require the .MODE file (for example, participation
factor computations, mode superpostion transient/harmonic analysis, PSD analysis). Setting Compute =
EVECTOR causes the Supernode eigensolver to compute only the corresponding eigenvectors. The Job-
1591
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SNOPTION
name.SNODE file and the associated Jobname.FULL file are required when requesting these eigenvectors.
In other words, the eigenvalues must have already been computed for this model before computing the
eigenvectors. This field can be useful in order to separate the two steps (computing eigenvalues and com-
puting eigenvectors).
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
SOCEAN, OCID
Specifies an ocean environment and associates it with an element section.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
OCID
The ocean environment ID number.
Notes
The SOCEAN command specifies an ocean environment and associates the specified ocean ID (OCID) to
the element section most recently defined via the SECTYPE command. (See Table 2.56: Ocean(p. 39) for the
commands used to define your ocean environment.)
Ocean loading applies to the PIPE288 and PIPE289 elements only.
Menu Paths
Main Menu>Preprocessor>Sections>Pipe>Add
Main Menu>Preprocessor>Sections>Pipe>Edit
SOLCONTROL, Key1, Key2, Key3, Vtol
Specifies whether to use optimized nonlinear solution defaults and some enhanced internal solution al-
gorithms.
SOLUTION: Analysis Options
MP ME ST PR PRN DS DSS <> EM EH <> PP <> EME MFS
Key1
Optimized defaults activation key:
ON or 1
Activates optimized defaults for a set of commands applicable to nonlinear solutions. This is the
default. The majority of solution command defaults are listed in the "Notes" (p. 1593) section below.
See also the description of individual solution commands for default values.
OFF or 0
Restores defaults to pre-ANSYS 5.4 values (see the Default States table below). Internal solution al-
gorithms work as for pre-ANSYS 5.4.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1592
SOCEAN
Key2
Check contact state key. This key is operable only when the optimized defaults are active (Key1 = ON)
and a contact or nonlinear status element is present in the model. When check contact state is active,
ANSYS will base the time step size on the specifications of KEYOPT(7) for all contact elements. KEYOPT(7)
for contact elements can be used to ensure that the time step interval accounts for changes in the
contact status. Also, when Key2 = ON, ANSYS ensures the time step is small enough to account for
changes in nonlinear element status (applies to LINK10, COMBIN7, COMBIN14, COMBIN39, and COMBIN40
elements). Valid arguments for the key are:
ON or 1
Activate time step predictions based on specifications of element KEYOPT(7) or the nonlinear status
of the element (applies to LINK10, COMBIN7, COMBIN14, COMBIN39, and COMBIN40 elements).
OFF or 0
Time step predictions not based on contact status or nonlinear element status (default).
Key3
Pressure load stiffness key. In general, use the default setting. Use a non-default setting only if you en-
counter convergence difficulties. Pressure load stiffness is automatically included when using eigenvalue
buckling analyses (ANTYPE,BUCKLE), equivalent to Key3 = INCP. For all other types of analyses, valid
arguments for Key3 are:
NOPL
Do not include pressure load stiffness for any elements.
no entry (default)
Include pressure load stiffness for elements SURF153, SURF154, SHELL181, PLANE182, PLANE183,
SOLID185, SOLID186, SOLID187, SOLSH190, BEAM188, BEAM189, FOLLW201, SHELL208, SHELL209,
SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290.
Do not include pressure load stiffness for elements PLANE42, SOLID45, SOLID65, PLANE82, SOLID92,
and SOLID95.
INCP
Include pressure load stiffness for elements PLANE42, SOLID45, SOLID65, PLANE82, SOLID92, SOLID95,
SURF153, SURF154, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188,
BEAM189, SOLSH190, FOLLW201, SHELL208, SHELL209, SOLID272, SOLID273, SHELL281, SOLID285,
PIPE288, PIPE289, and ELBOW290.
Vtol
Tolerance for volumetric compatibility check in current-technology plane and solid elements (such as
PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, SOLSH190, SOLID272, and SOLID273) when mixed
u-P formulation is employed. When displacement convergence is not checked, the default value is 1.0
x 10
-5
; when displacement convergence is checked, the default value is 1.0 x 10
-3
. Vtol can range from
0.0 to 1.0, but recommended values are between 1.0 x 10
-5
and 1.0 x 10
-2
. For more details, see Volumetric
Constraint Equations in u-P Formulations in the Theory Reference for the Mechanical APDL and Mechanical
Applications.
Notes
The SOLCONTROL command is designed to provide reliable and efficient default solution settings for single-
field full structural nonlinear or full transient analysis, or single-field thermal analysis. (It is not applicable for
reduced transient analysis.) The optimized default settings and advanced internal solution algorithms can
be used to solve the majority of structural/thermal, nonlinear/transient problems with the least amount of
1593
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SOLCONTROL
user intervention. The SOLCONTROL command is ON by default. In most cases, to successfully solve a
problem the user only needs to:
Choose NLGEOM,ON for large displacement/strain analysis.
Choose NROPT,UNSYM to access unsymmetric pressure load stiffness, material stiffness, friction behavior,
etc.
Provide the initial step size through the NSUBST or DELTIM command.
To achieve this, the SOLCONTROL command chooses better default settings for a number of commands
within ANSYS and uses optimized internal solution algorithms.
THOPT settings take precedence over SOLCONTROL settings.
Commands that ANSYS issues are written to files (such as log and load step files) as COMMAND,-1 (for ex-
ample, KBC,-1).
Note
The state of the SOLCONTROL command is not written by the CDWRITE or LSWRITE commands
(so that the CDWRITE file does not rigidly define an individual solution command). Also note
that switching SOLCONTROL ON and OFF between load steps is not recommended.
The SOLCONTROL command also serves as a "reset" command; when you issue SOLCONTROL, all the
control commands set earlier in the interactive or input session are reset to their original default values.
Text database files (.CDB files created by CDWRITE) and load step files (.S01, .S02, .Snn, etc. files created
by LSWRITE) should be handled with care when SOLCONTROL,ON (default). Files of these types sometimes
contain control commands that you did not issue intentionally. These extra commands can overwrite the
default settings specified by SOLCONTROL. To avoid overwriting the SOLCONTROL settings when you are
using .CDB files, use the following procedure:
1. Read the .CDB files into ANSYS.
2. Enter the solution processor [/SOLU].
3. Issue SOLCONTROL,ON.
4. Issue the desired control commands to overwrite the SOLCONTROL defaults as needed.
To use .Snn files properly, you should preview and edit them. Delete the unwanted solution commands
before you execute the LSSOLVE command.
The following table lists the nonlinear solution parameters and algorithm differences when the command
is active and inactive.
Table 257 SOLCONTROL - Default States Table
SOLCONTROL OFF SOLCONTROL ON Argu-
ment
Command
General Options
Previously specified value. If
no specified value, defaults
to 1.
1 load step if contact elements
TARGE169, TARGE170, CONTA171,
CONTA172, CONTA173, CONTA174,
NSBSTP NSUBST
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1594
SOLCONTROL
SOLCONTROL OFF SOLCONTROL ON Argu-
ment
Command
CONTA175, CONTA176, CONTA177
are not present; if these elements are
present, 1 or 20 substeps, depending
on the physics of the program
Previously specified value. If
no specified value, defaults
to NSBSTP.
Determined by ANSYS NSBMX
Previously specified value. If
no specified value, defaults
to 1.
Determined by ANSYS NSBMN
OFF Determined by ANSYS Carry
Previously specified value, if
any.
1 time span of the load step if con-
tact elements TARGE169, TARGE170,
DTIME DELTIM
CONTA171, CONTA172, CONTA173,
CONTA174, CONTA175, CONTA176,
CONTA177 are not present; if these
elements are present, 1 or 1/20 of
the time span of the load step, de-
pending on the physics of the pro-
gram
Previously specified value. If
no specified value, defaults
to DTIME
Determined by ANSYS DTMIN
Previously specified value. If
no specified value, defaults
to time span of load step.
Determined by ANSYS DTMAX
OFF Determined by ANSYS Carry
0 for all types of transient or
nonlinear analysis
0 (ramped) for static nonlinear,
structural and thermal (steady state)
KEY KBC
analyses, as well as for transient
analyses when TIMINT,OFF. 1
(stepped) for transient structural and
thermal analyses when TIMINT,ON.
(TIMINT,ON is the default for transi-
ent analyses.)
OFF Chosen by program Key AUTOTS
Uses sparse solver. If PCG
solver is chosen, sets multi-
Uses sparse solver. If PCG solver is
chosen, sets multiplier to 2.0 for
Newton-Raphson iteration.
EQSLV
plier to 1.0 for Newton-
Raphson iteration.
Write all the default values
for solution control com-
mands.
Does not write default values for
most of the relevant solution control
commands or options listed in this
table.
CDWRITE and
LSWRITE
Not available Active MONITOR
1595
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SOLCONTROL
SOLCONTROL OFF SOLCONTROL ON Argu-
ment
Command
Nonlinear Options
Force or moment conver-
gence tolerance = 0.1% Dis-
Force or moment convergence toler-
ance = 0.5% Displacement tolerance
= 5%
TOLER CNVTOL
placement tolerance not
checked.
For force or moment, 1.0 for
heat flow, 1.0E-6 otherwise,
0
0.01 for force or moment; for heat
flow and others the same as SOL-
CONTROL ,OFF
MINREF
25 Between 15 and 26, depending on
the physics of the problem.
NEQIT NEQIT
Use ARCLEN as in Release
5.3.
A more aggressive scheme to open-
up time step is used. A more stable
ARCLEN algorithm is used.
ARCLEN
OFF On, unless SOLID65 is present. Sskey PRED
OFF Automatically turned ON when con-
tact elements present.
Key LNSRCH
5% 15% PLSLIMIT CUTCONTROL
20 13 NPOINT
Not Available .01 TEMP OPNCONTROL
Not Available 3 NUM-
STEP
OFF ON for geometrically nonlinear ana-
lysis (NLGEOM, ON).
Key SSTIF
Automatically toggled on
and off depending on
OFF, except: when frictional contact
exists; when elements CONTAC12 or
ADPTKY NROPT
whether plasticity or friction-
al contact exists or not.
CONTAC52 are present; or when
plasticity exists and one of the ele-
ments PIPE20, BEAM23, BEAM24, or
PIPE60 is present.
.5 1.0 THETA TINTP
.2 0.0 TOL
Element Options
Time prediction depending
on KEYOPT(7).
Time prediction independent of
KEYOPT(7) (default) except when re-
quested.
CONTAC12, CONTAC52
Same Adaptive descent ON when friction
is present.
CONTAC12, CONTAC52
Same Adaptive descent ON when plasticity
is present.
PIPE20, BEAM23, BEAM24,
PIPE60
KEYOPT(2) = 0 Consistent tangent KEYOPT(2) = 1
when NLGEOM,ON.
BEAM4, SHELL63
Algorithm Behavior
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1596
SOLCONTROL
SOLCONTROL OFF SOLCONTROL ON Argu-
ment
Command
Not available Active Deformed element shape
(Jacobi) check used as criter-
ia for early bisection
0.0 1.0 for thermal analysis. Euler backward theta (for
first order equations)
Write all the default values
for solution controls com-
mands.
Does not write default values for any
of the relevant commands or options
listed in this table.
Log file
When zero CONVTOL, MIN-
REF value is used.
Automatically corrected when values
become too small by using a reac-
Moment reference values
tion force times an element charac-
teristic length.
Check on nonconvergent
patterns not implemented.
Check on nonconvergent patterns.
Time step is opened up less aggress-
Automatic time step
scheme
Time step is opened up ively; the increase factor (used in
more aggressively; the in- calculating the degree in which the
crease factor = 2.0 (in most time step is opened) = 1.5 (in most
cases). No physics depend-
ency involved.
cases). The calculation also takes into
account the physics of the problem.
Not available SOLCONTROL,ON or OFF Reset of all solution control
defaults in one command.
When force norm is smaller
than 1, the MINREF value is
used as the REF value.
When force norm is smaller than 1,
the calculated force value is still used
as the REF value. If the calculated
Nonlinear convergence cri-
terion
force value is approaching machine
zero, the MINREF value is used as
REF.
Detailed message printed
for each iteration.
Simplified message, not printed in
some cases.
Warning message printed
when negative diagonal in
matrix is discovered.
Use Jobname.ABT file to
control normal abort.
Available in GUI. Jobname.ABT file
can also be used.
Stop button
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Nonlinear>Mixed U-P Toler
Main Menu>Preprocessor>Loads>Load Step Opts>Solution Ctrl
Main Menu>Solution>Load Step Opts>Nonlinear>Mixed U-P Toler
Main Menu>Solution>Load Step Opts>Solution Ctrl
1597
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SOLCONTROL
SOLU, NVAR, Item, Comp, Name
Specifies solution summary data per substep to be stored.
POST26: Set Up
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
NVAR
Arbitrary reference number assigned to this variable (2 to NV [NUMVAR]).
Item
Label identifying the item. Valid item labels are shown in the table below. Some items may also require
a component label.
Comp
Component of the item (if required). Valid component labels are shown in the table below. None are
currently required.
Name
Thirty-two character name identifying the item on printouts and displays. Defaults to an eight character
label formed by concatenating the first four characters of the Item and Comp labels.
Notes
See also the PRITER command of POST1 to display some of these items directly. Valid for a static or full
transient analysis. All other analyses have zeros for the data. Valid item and component labels for solution
summary values are:
Valid Item Labels
Description Item
Total arc-length load factor (ratio of the load at equilibrium to the total applied
load)
ALLF
Arc-length load factor increment (change in ALLF) ALDLF
Normalized arc-length radius ARCL
Convergence indicator. CNVG
Maximum creep ratio. CRPRAT
Current segment convergence value. CSCV
Current convergence value. CUCV
Displacement convergence value. DICV
Descent parameter. DSPRM
Time step size. DTIME
Number of equilibrium iterations. EQIT
Fluid flow convergence value. FFCV
Force convergence value. FOCV
Heat flow convergence value. HFCV
Cumulative number of iterations. NCMIT
Cumulative number of load steps NCMLS
Cumulative number of substeps NCMSS
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1598
SOLU
Valid Item Labels
Description Item
Magnetic flux convergence value. MFCV
Moment convergence value. MOCV
Maximum degree of freedom value. MXDVL
Pressure convergence value. PRCV
Maximum plastic strain increment. PSINC
Response frequency for 2nd order systems. RESFRQ
Response eigenvalue for 1st order systems. RESEIG
Rotation convergence value. ROCV
Scalar magnetic potential convergence value. SMCV
Temperature convergence value. TECV
Velocity convergence value. VECV
Voltage convergence value. VOCV
Vector magnetic potential convergence value. VMCV
Menu Paths
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
/SOLU
Enters the solution processor.
SESSION: Processor Entry
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This command is valid only at the Begin Level.
Menu Paths
Main Menu>Solution
1599
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/SOLU
SOLUOPT
Specifies "Solution options" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Solution>General
SOLVE
Starts a solution.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Starts the solution of one load step of a solution sequence based on the current analysis type and option
settings.
Menu Paths
Main Menu>DesignXplorer>Solution>Solve
Main Menu>Drop Test>Solve
Main Menu>Solution>Run FLOTRAN
Main Menu>Solution>Solve
Main Menu>Solution>Solve>Current LS
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1600
SOLUOPT
SORT
Specifies "Sort settings" as the subsequent status topic.
POST1: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>General Postproc>Sort Module
SOURCE, X, Y, Z
Defines a default location for undefined nodes or keypoints.
PREP7: Keypoints
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
X, Y, Z
Global Cartesian coordinates for source nodes or keypoints (defaults to the origin).
Command Default
Global Cartesian origin.
Notes
Defines a global Cartesian location for undefined nodes or keypoints moved during intersection calculations
[MOVE or KMOVE].
Menu Paths
This command cannot be accessed from a menu.
1601
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SOURCE
SPACE, NODE
Defines a space node for radiation using the Radiation Matrix method.
AUX12: Radiation Matrix Method
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE
Node defined to be the space node.
Command Default
No space node (no radiation to space).
Notes
A space node is required in an open system to account for radiation losses.
If using SPACE with the ANSYS Multi-field solver (MFS), you must capture this command in the command
file using MFCMMAND. This step is necessary because at the end of each field computation, this command
is unset.
Menu Paths
Main Menu>Radiation Opt>Matrix Method>Other Settings
SPADP, FREQ, NUMADP, RMSSP, FACTB, FACTE, SLVOPT, SLVACC
Automatically refines a HF119 tetrahedral element mesh based on S-parameter convergence.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> <> <> <> <>
Freq
Working frequency (Hz).
NUMADP
Maximum number of refinement iterations (0 to 10). Defaults to 0.
RMSSP
Root mean square error criteria for |S|. Defaults to 0.01.
FACTB
Beginning refinement percentage (0 to 100). Defaults to 10.
FACTE
Ending refinement percentage (0 to 100). Defaults to 10.
SLVOPT
Solver option:
0
SPARSE
1
QMR (symmetric matrices)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1602
SPACE
2
ICCG (unsymmetric matrices)
SLVACC
Convergence criteria for an iterative solver. Defaults to 1.0E-6.
Notes
The refinement factor varies linearly with the number of refinement iterations.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
SPARM, PORTI, PORTJ, PAROPT
Calculates parameters between ports of a network system.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
PORTI
Port number of the excited port with a excitation mode. (See the description of the HFPORT command.)
PORTJ
Port number of the output port. This could be used for a multiport system. All ports but PORTI must
be matched.
PAROPT
Output parameter option:
0
S-parameters (default).
1
Reflection coefficient and voltage standing wave ratio.
Notes
SPARM calculates parameters for multiport or single port (PORTI = PORTJ) systems.
If PAROPT = 0, the SPARM command macro returns the following complex S parameters: SII, dBSII, SIIPHR,
SJI, dBSJI, and SJIPHR, where "I" is the port number for the excited port, "J" is the output port number and
PHR is the phase angle.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Port>S-Parameters
1603
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPARM
SPCNOD, ENCL, NODE
Defines a space node for radiation using the Radiosity method.
SOLUTION: Radiosity
AUX12: Radiosity Solver
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
ENCL
Radiating surface enclosure number. Defaults to 1. If ENCL = STAT, the command lists all enclosure space
nodes. If ENCL = DELE, the command deletes all enclosure space nodes.
NODE
Node defined to be the space node.
Notes
For open systems, an enclosure may radiate to a space node (NODE).
Open systems may be characterized by one or more enclosures (ENCL). Each enclosure may radiate to a
different space node (NODE).
For a space node that is not part of the finite element model, specify the temperature using the D command.
For the first load step, the space node temperature ramps from the uniform temperature specified by the
TUNIF command to the temperature specified by the D command. For subsequent load steps, it ramps from
the previous value of the space node temperature. For intermediate load steps, use the SPCNOD,DELETE
command and specify the space node temperature again to ramp from the uniform temperature.
For a space node that is part of the finite element model, the temperature is that calculated during the finite
element solution.
Menu Paths
Main Menu>Preprocessor>Radiation Opts>Solution Opt
Main Menu>Radiation Opt>Radiosity Meth>Solution Opt
Main Menu>Solution>Radiation Opts>Solution Opt
SPCTEMP, ENCL, TEMP
Defines a free-space ambient temperature for radiation using the Radiosity method.
SOLUTION: Radiosity
AUX12: Radiosity Solver
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
ENCL
Radiating surface enclosure number. Defaults to 1. If ENCL = STAT, the command lists all enclosure space
temperatures. If ENCL = DELE, the command deletes all enclosure space temperatures.
TEMP
Temperature of free-space in the reference temperature system. The temperature will be offset by the
value specified in the TOFFST command for internal calculations.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1604
SPCNOD
Notes
For open systems, an enclosure may radiate to the free-space ambient temperature (TEMP).
Open systems may be characterized by one or more enclosures (ENCL). Each enclosure may radiate to a
different free-space ambient temperature (TEMP).
For the first load step, the space temperature ramps from the uniform temperature specified by the TUNIF
command to the temperature specified by the SPCTEMP command. For subsequent load steps, it ramps
from the previous value of the space temperature. For intermediate load steps, use the SPCTEMP,DELETE
command and specify the space temperature again to ramp from the uniform temperature.
If using SPCTEMP with the ANSYS Multi-field solver (MFS), you must capture this command in the command
file using MFCMMAND. This step is necessary because at the end of each field computation, this command
is unset.
Menu Paths
Main Menu>Preprocessor>Radiation Opts>Solution Opt
Main Menu>Radiation Opt>Radiosity Meth>Solution Opt
Main Menu>Solution>Radiation Opts>Solution Opt
SPEC
Specifies "Miscellaneous specifications" as the subsequent status topic.
POST1: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>General Postproc>Output Options
1605
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPEC
SPFSS, FREQB, FREQE, FREQINC, PHIB, PHIE, PHIINC, THETAB, THETAE, THETAINC, SWPOPT, EFACC, OUTPUT,
FILEOPT
Performs frequency sweep or angle sweep calculations to determine reflection and transmission parameters
of a frequency selective surface.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FREQB
Frequency (Hz) at the beginning of the FREQB to FREQE range. If FREQE is blank, the solution is done
only at frequency FREQB.
FREQE
Frequency at end of this range.
FREQINC
Frequency increment. The number of solutions performed is [(FREQE - FREQB) / FREQINC] + 1. Solutions
are always performed at FREQB and FREQE.
PHIB
Angle (Degree) of spherical coordinates at the beginning of the PHIB to PHIE range. If PHIE is blank,
the solution is done only at angle PHIB.
PHIE
Angle at end of this range.
PHIINC
angle increment. The number of solutions performed is [(PHIE - PHIB) / PHIINC] + 1. Solutions are
always performed at PHIB and PHIE.
THETAB
Angle (Degree) of spherical coordinates at the beginning of the THETAB to THETAE range. If THETAE
is blank, the solution is done only at angle.
THETAE
Angle at end of this range.
THETAINC
angle increment. The number of solutions performed is [(THETAE - THETAB) / THETAINC] + 1. Solutions
are always performed at THETAB and THETAE.
SWPOPT
Solution Method (See HROPT command):
0
Variational Technology (default). See notes below.
1
Full method.
2
Variational Technology using a perfect absorber. See notes below.
EFACC
Electric field accuracy for fast exploration method. Defaults to 0.01.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1606
SPFSS
OUTPUT
0
Create minimal size results file required to compute S-parameters for the last port excitation case,
for all frequencies (default).
1
Create complete results file for the last port excitation case, for all frequencies.
2
Create complete results file for the last port excitation case, for the last frequency only. Valid only
for full method.
FILEOPT
0
Output S-parameter magnitude and phase angle in Touchstone file.
1
Output S-parameter magnitude in dB and phase angle in Touchstone file.
Notes
Only one parameter among frequency, angle, and angle can be chosen to sweep each time. An output
file filename.fnp (n = 1 or 2) is generated after the sweep solution.
To use any Variational Technology method, you must have the separately licensed VT Accelerator. The
Variational Technology Perfect Absorber method (Method = VTPA) is about 20% faster but slightly less ac-
curate than the Variational Technology method (Method = VT).
The Variational Technology methods (SWPOPT = 0 or 2) are only available for frequency sweep calculations.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
SPH4, XCENTER, YCENTER, RAD1, RAD2
Creates a spherical volume anywhere on the working plane.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XCENTER, YCENTER
Working plane X and Y coordinates of the center of the sphere.
RAD1, RAD2
Inner and outer radii (either order) of the sphere. A value of zero or blank for either RAD1 or RAD2
defines a solid sphere.
1607
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPH4
Notes
Defines either a solid or hollow spherical volume anywhere on the working plane. The sphere must have a
spatial volume greater than zero. (i.e., this volume primitive command cannot be used to create a degenerate
volume as a means of creating an area.) A sphere of 360 will be defined with two areas, each consisting of
a hemisphere. See the SPHERE and SPH5 commands for other ways to create spheres.
When working with a model imported from an IGES file (DEFAULT import option), you can create only solid
spheres. If you enter a value for both RAD1 and RAD2 the command is ignored.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Primitives>Solid Sphere
Main Menu>Preprocessor>Modeling>Create>Volumes>Sphere>Hollow Sphere
Main Menu>Preprocessor>Modeling>Create>Volumes>Sphere>Solid Sphere
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Create>Volume>Sphere>Solid Sphere
SPH5, XEDGE1, YEDGE1, XEDGE2, YEDGE2
Creates a spherical volume by diameter end points.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XEDGE1, YEDGE1
Working plane X and Y coordinates of one edge of the sphere.
XEDGE2, YEDGE2
Working plane X and Y coordinates of the other edge of the sphere.
Notes
Defines a solid spherical volume anywhere on the working plane by specifying diameter end points. The
sphere must have a spatial volume greater than zero. (i.e., this volume primitive command cannot be used
to create a degenerate volume as a means of creating an area.) A sphere of 360 will be defined with two
areas, each consisting of a hemisphere. See the SPHERE and SPH4 commands for other ways to create
spheres.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Sphere>By End Points
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Create>Volume>Sphere>By End Points
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1608
SPH5
SPHERE, RAD1, RAD2, THETA1, THETA2
Creates a spherical volume centered about the working plane origin.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RAD1, RAD2
Inner and outer radii (either order) of the sphere. A value of zero or blank for either RAD1 or RAD2
defines a solid sphere.
THETA1, THETA2
Starting and ending angles (either order) of the sphere. Used for creating a spherical sector. The sector
begins at the algebraically smaller angle, extends in a positive angular direction, and ends at the larger
angle. The starting angle defaults to 0.0 and the ending angle defaults to 360.0. See the Modeling and
Meshing Guide for an illustration.
Notes
Defines either a solid or hollow sphere or spherical sector centered about the working plane origin. The
sphere must have a spatial volume greater than zero. (i.e., this volume primitive command cannot be used
to create a degenerate volume as a means of creating an area.) Inaccuracies can develop when the size of
the object you create is much smaller than the relative coordinate system values (ratios near to or greater
than 1000). If you require an exceptionally small sphere, create a larger object, and scale it down to the ap-
propriate size.
For a solid sphere of 360, you define it with two areas, each consisting of a hemisphere. See the SPH4 and
SPH5 commands for the other ways to create spheres.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Sphere>By Dimensions
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Create>Volume>Sphere>By Dimensions
SPICE, Fname, Ext, RMSERR, Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8, Z9, Z10, Z11, Z12, Z13, Z14, Z15, Z16
Generates a SPICE subcircuit model using S-parameters from a Touchstone file.
POST1: Special Purpose
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path) of the Touchstone input file with s-parameters. An unspecified directory path defaults to the
working directory; in this case, you can use all 248 characters for the file name.
Ext
Extension of Touchstone file (.snp where n is the number of ports).
RMSERR
The convergence criteria (root-mean-square error) for the SPICE subcircuit model. Default determined
by ANSYS.
1609
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPICE
Z1, Z2, Z3, . . . , Z16
Reference impedance for ports 1, 2, 3, 16. Default to 50 ohms.
Notes
SPICE allows a maximum of 16 reference impedances.
The frequency range is determined by the frequency range of the Touchstone file.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
SPLINE, P1, P2, P3, P4, P5, P6, XV1, YV1, ZV1, XV6, YV6, ZV6
Generates a segmented spline through a series of keypoints.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1, P2, P3, . . . , P6
Keypoints through which the spline is fit. At least two must be defined. If P1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI).
The following fields are used only if specified end slopes on the line are desired, otherwise zero curvature
end slopes will be automatically calculated to produce a line which is "straight" in the active coordinate
system. To specify end slopes, use the following fields to define a "slope vector" (one for each end of the
line, if desired) that has its tail at the origin and its head at the point XVn,YVn, ZVn in the active coordinate
system [CSYS]. The corresponding end slope of the line will then be parallel to this "slope vector."
XV1, YV1, ZV1
Location (in the active coordinate system) of the head of the "slope vector" corresponding to the slope
at the P1 end of the spline. The tail of the vector is at the origin of the coordinate system.
XV6, YV6, ZV6
Location of the head of the "slope vector" corresponding to the slope at the P6 (or the last keypoint if
fewer than six specified) end of the spline.
Notes
The output from this command is a series of connected lines (one line between each pair of keypoints) that
together form a spline. Note that solid modeling in a toroidal coordinate system is not recommended.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Lines>Splines>Segmented Spline
Main Menu>Preprocessor>Modeling>Create>Lines>Splines>With Options>Segmented Spline
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1610
SPLINE
SPLOT, NA1, NA2, NINC, MESH
Displays the selected areas and a faceted view of their underlying surfaces
PREP7: Areas
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NA1
Starting area for display of areas and underlying surfaces. If NA1 = ALL (default), NA2 and NINC are ig-
nored and all selected areas are displayed (ASEL command).
NA2
Last area to be displayed.
NINC
Numeric value setting steps between NA1 and NA2 for display. Default value is (1).
MESH
Specifies a rectangular mesh density used to display the underlying surface (default 4, i.e. 4 x 4).
Notes
This command is valid in any processor. The plot output displays the external and internal trim curves and
underlying surface. You cannot obtain a faceted view of your surface areas when you are using the /EXPAND
command to create larger graphics displays.
Use APLOT for trimmed surface display.
Menu Paths
This command cannot be accessed from a menu.
SPOINT, NODE, X, Y, Z
Defines a point for moment summations.
POST1: Special Purpose
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NODE
Node number of the desired point. If zero, use X,Y,Z to describe point.
X, Y, Z
Global Cartesian coordinates of the desired summation point. Used if NODE is 0. Defaults to (0,0,0).
Command Default
No point is defined by default; you must either specify a node or coordinates.
Notes
Defines a point (any point other than the origin) about which the tabular moment summations are computed
[NFORCE, FSUM]. If force summations are desired in other than the global Cartesian directions, a node
1611
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPOINT
number must be specified on the NODE field, and the desired coordinate system must be activated with
RSYS.
Menu Paths
Main Menu>General Postproc>Nodal Calcs>Summation Pt>At Node
Main Menu>General Postproc>Nodal Calcs>Summation Pt>At XYZ Loc
SPOPT, Sptype, NMODE, Elcalc
Selects the spectrum type and other spectrum options.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Product Restrictions
Sptype
Spectrum type:
SPRS
Single point excitation response spectrum (default). See also the SVTYP command.
MPRS
Multiple point excitation response spectrum.
DDAM
Dynamic design analysis method.
PSD
Power spectral density.
NMODE
Use the first NMODE modes from the modal analysis. Defaults to all extracted modes, as specified by the
MODOPT and BUCOPT commands. NMODE cannot be larger than 10000.
Elcalc
Element calculation key (for Sptype = PSD only):
NO
Do not include stress responses in the calculations (default).
YES
Include stress responses in the calculations.
Notes
Valid only for a spectrum analysis (ANTYPE,SPECTR). This operation must be preceded by a modal solution
(ANTYPE,MODAL) with the appropriate files available. If used in SOLUTION, this command is valid only
within the first load step.
This command is also valid in PREP7.
Product Restrictions
Only Sptype = SPRS is allowed in ANSYS Professional.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1612
SPOPT
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
SPREAD, VALUE
Turns on a dashed tolerance curve for the subsequent curve plots.
POST26: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
VALUE
Amount of tolerance. For example, 0.1 is 10%.
Command Default
No tolerance curve.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Graph
SPSCAN, FREQ, LOCAL, PHIB, PHIE, PHIINC, THETAB, THETAE, THETAINC, FILEOPT, SLVOPT, SLVACC, SYMOPT
Performs a harmonic analysis of a unit cell over a range of angles and extracts the S-parameter.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FREQ
Working frequency in Hz.
LOCAL
Local coordinate system number (defaults to 0).
PHIB
Beginning angle from x-axis towards y-axis, , in degrees (defaults to 0).
PHIE
Ending angle from x-axis towards y-axis, , in degrees (defaults to 0).
PHIINC
Increment of in degrees (defaults to 0).
THETAB
Beginning angle from +z-axis towards -z-axis, , in degrees (defaults to 0).
THETAE
Ending angle from +z-axis towards -z-axis, , in degrees (defaults to 0).
THETAINC
Increment of in degrees (defaults to 0).
1613
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPSCAN
FILEOPT
0
Output S-parameter magnitude and phase angle to Jobname.scan.
1
Output S-parameter magnitude in dB and phase angle to Jobname.scan.
SLVOPT
Solver option:
0
SPARSE
1
QMR (symmetric matrices)
2
ICCG (unsymmetric matrices)
SLVACC
Convergence criteria for an iterative solver. Defaults to 1.0E-6.
SYMOPT
Key for S-parameter output format:
0
The S-parameter matrix is symmetrical (default). The following are examples.
For a 2 port system that has only port 1 excited, the assumption is that S
12
= S
21
and S
22
= S
11
.
S
S S
S S
[ ]
1
]
1
11 12
21 22
For a 3 port system that has only port 1 excited, the assumption is that S
12
= S
21
and S
13
= S
31
.
S
S S S
S
S
[ ]
1
]
1
1
1
11 12 13
21 0 0
31 0 0
1
The S-parameter matrix is unsymmetrical. The following are examples.
For a 2 port system that has only port 1 excited, the assumption is that S
12
and S
22
are zero.
S
S
S
[ ]
1
]
1
11 0
21 0
For a 3 port system that has only port 1 excited, the assumption is that S
12
, S
22
, S
32
, S
13
, S
23
, and S
33
are all zero.
S
S
S
S
[ ]
1
]
1
1
1
11 0 0
21 0 0
31 0 0
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1614
SPSCAN
Notes
See the figure Spherical Coordinates in the High-Frequency Electromagnetic Analysis Guide for an illustration
of the coordinate system.
Menu Paths
Main Menu>Solution>Solve>S-Par Angle Sweep
SPSWP, FREQB, FREQE, FREQINC, SWPOPT, EFACC, OUTPUT, FILEOPT, SLVOPT, SLVACC, SYMOPT
Computes S-parameters over a frequency range and writes them to a file.
SOLUTION: Analysis Options
MP <> <> <> <> <> <> <> <> EH <> PP <> <> <>
FREQB
Frequency (Hz) at the beginning of the FREQB to FREQE range. If FREQE is blank, the solution is done
only at frequency FREQB.
FREQE
Frequency at end of this range.
FREQINC
Frequency increment. The number of solutions performed is [(FREQE - FREQB)/FREQINC] + 1. Solutions
are always performed at FREQB and FREQE.
SWPOPT
Solution Method (See HROPT command):
0
Variational Technology (default).
1
Full method.
2
Variational Technology using a perfect absorber.
Note
To use any Variational Technology method, you must have the separately licensed VT Accelerator.
The Variational Technology Perfect Absorber method (Method = VTPA) is about 20% faster but
slightly less accurate than the Variational Technology method (Method = VT).
EFACC
Electric field accuracy for fast exploration method. Defaults to 0.01.
OUTPUT
0
Create minimal size results file required to compute S-parameters for the last port excitation case
(see "Notes" (p. 1617)), for all frequencies. (default).
1615
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPSWP
1
Create complete results file for the last port excitation case (see "Notes" (p. 1617)), for all frequencies.
2
Create complete results file for the last port excitation case (see "Notes" (p. 1617)), for the last frequency
only, valid only for full method.
FILEOPT
0
Output S-parameter magnitude and phase angle in Touchstone file.
1
Output S-parameter magnitude in dB and phase angle in Touchstone file.
2
Output S-parameter real and imaginary parts in Touchstone file.
SLVOPT
0
SPARSE
1
QMR (symmetric matrices).
2
ICCG (unsymmetric matrices).
SLVACC
Convergence criteria for an iterative solver. Defaults to 1.0E-6.
SYMOPT
Key for S-parameter output format:
0
The S-parameter matrix is symmetrical (default). The following are examples.
For a 2 port system that has only port 1 excited, the assumption is that S
12
= S
21
and S
22
= S
11
.
S
S S
S S
[ ]
1
]
1
11 12
21 22
For a 3 port system that has only port 1 excited, the assumption is that S
12
= S
21
and S
13
= S
31
.
S
S S S
S
S
[ ]
1
]
1
1
1
11 12 13
21 0 0
31 0 0
1
The S-parameter matrix is unsymmetrical. The following are examples.
For a 2 port system that has only port 1 excited, the assumption is that S
12
and S
22
are zero.
S
S
S
[ ]
1
]
1
11 0
21 0
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1616
SPSWP
For a 3 port system that has only port 1 excited, the assumption is that S
12
, S
22
, S
32
, S
13
, S
23
, and S
33
are all zero.
S
S
S
S
[ ]
1
]
1
1
1
11 0 0
21 0 0
31 0 0
Notes
This command computes S-parameters by sequencing through a series of harmonic solutions with different
port excitations over the desired frequency range. To use this function, port flag boundary conditions must
be previously set (see SF, SFA for exterior waveguide ports, or BFA, BFL or BF for interior waveguide or
transmission line ports). Ports should be numbered sequentially from "1" with no gaps in the numbering.
Resulting S-parameters are written to a file jobname.snp, where n is the number of ports. The file is
written in Touchstone format.
S-parameters require a sequence of solutions whereby for each solution, one port is "excited" and the other
ports are "matched". Each solution represents one column of a S-parameter matrix (i.e., if Port 1 is excited
for a three-port system, the resulting column represents the S11, S21 and S31 S-parameters). A full S-para-
meter matrix for a n-port system at one frequency requires n solutions alternating "excited" and "matched"
port boundary conditions at each port. The SPSWP command will solve a column of the S-parameter matrix
for each port that has a defined excitation. If all ports have a defined excitation, then the full S-parameter
matrix will be computed.
To prepare the ports for SPSWP, each port must be defined (flagged) as a port and it must have a port ex-
citation defined. For waveguide ports, excitation is defined using the HFPORT command. For transmission
line ports, excitation is defined using the BFA, BFL, or BF commands using the JS, H, or EF load options.
Transmission line excitation must be defined on the port surface itself (nodes, lines, or areas).
Menu Paths
Main Menu>Solution>Solve>S-Par Sweep
SPTOPT
Specifies "Spectrum analysis options" as the subsequent status topic.
SOLUTION: Status
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
1617
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SPTOPT
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Show Status
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Show Status
Utility Menu>List>Status>Solution>Spectrum Options
SQRT, IR, IA, --, --, Name, --, --, FACTA
Forms the square root of a variable.
POST26: Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Arbitrary reference number assigned to the resulting variable (2 to NV [NUMVAR]). If this number is the
same as for a previously defined variable, the previously defined variable will be overwritten with this
result.
IA
Reference number of the variable to be operated on.
--, --
Unused fields.
Name
Thirty-two character name identifying the variable on printouts and displays. Embedded blanks are
compressed for output.
--, --
Unused fields.
FACTA
Scaling factor (positive or negative) applied to variable IA (defaults to 1.0).
Notes
Forms the square root of a variable according to the operation:
IR FACTA IA
Menu Paths
Main Menu>TimeHist Postpro>Math Operations>Square Root
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1618
SQRT
*SREAD, StrArray, Fname, Ext, --, nChar, nSkip, nRead
Reads a file into a string array parameter.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
StrArray
Name of the "string array" parameter which will hold the read file. String array parameters are similar to
character arrays, but each array element can be as long as 128 characters. If the string parameter does
not exist, it will be created. The array will be created as: *DIM,StrArray,STRING,nChar,nRead
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
nChar
Number of characters per line to read (default is length of the longest line in the file).
nSkip
Number of lines to skip at the start of the file (default is 0).
nRead
Number of lines to read from the file (default is the entire file).
Notes
The *SREAD command reads from a file into a string array parameter. The file must be an ASCII text file.
Menu Paths
This command cannot be accessed from a menu.
SRSS, SIGNIF, Label
Specifies the square root of sum of squares mode combination method.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
SIGNIF
Combine only those modes whose significance level exceeds the SIGNIF threshold. For single point,
multipoint, or DDAM response (SPOPT,SPRS, MPRS or DDAM), the significance level of a mode is defined
as the mode coefficient of the mode, divided by the maximum mode coefficient of all modes. Any mode
whose significance level is less than SIGNIF is considered insignificant and is not contributed to the
mode combinations. The higher the SIGNIF threshold, the fewer the number of modes combined.
1619
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SRSS
SIGNIF defaults to 0.001. If SIGNIF is specified as 0.0, it is taken as 0.0. (This mode combination
method is not valid for SPOPT,PSD.)
Label
Label identifying the combined mode solution output.
DISP
Displacement solution (default). Displacements, stresses, forces, etc., are available.
VELO
Velocity solution. Velocities, "stress velocities," "force velocities," etc., are available.
ACEL
Acceleration solution. Accelerations, "stress accelerations," "force accelerations," etc., are available.
Notes
This command is also valid for PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>MultiPt>Mode Combine
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Mode Combine
SSBT, B
T
11
, B
T
22
, B
T
12
, T
Specifies preintegrated bending thermal effects for shell sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
B
T
11
, B
T
22
, B
T
12
Bending thermal effects component [B
T
].
T
Temperature.
Notes
The behavior of shell elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
A B
B D
T T
M
B
S E
T
I
T
T
'
1
]
1
1
'
'
( - )
[ ] [ ]{{ }
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1620
SSBT
The SSBT command, one of several preintegrated shell section commands, specifies the bending thermal
effects quantity (submatrix [B
T
] data) for a preintegrated shell section. The section data defined is associated
with the section most recently defined (via the SECTYPE command).
The [B
T
] quantity represents bending stress resultants caused by a unit raise in temperature on a fully con-
strained model. For a layered composite shell, it is usually necessary to specify both the [B
T
] and [M
T
]
quantities (by issuing the SSBT and SSMT commands, respectively).
Unspecified values default to zero.
Related commands are SSPA, SSPB, SSPD, SSPE, SSMT, and SSPM.
If you are using the SHELL181 or SHELL281 element's Membrane option (KEYOPT(1) = 1), it is not necessary
to issue this command.
For complete information, see Using Preintegrated General Shell Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
/SSCALE, WN, SMULT
Sets the contour multiplier for topographic displays.
GRAPHICS: Scaling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
SMULT
Contour multiplier that factors in results based on the product of the multiplier and the result being
plotted. Defaults to 0.0 (no topographic effects).
Command Default
No topographic contour effects.
Notes
Use this command to scale values to the geometry when the contours are shown elevated. For section displays
[/TYPE], the elevation is performed perpendicular to the section face.
Nonzero contour multipliers factoring in large results (stresses or displacements) can produce very large
distortion, causing images to disappear. To bring a distorted image back into view, reduce the contour
multiplier value.
Portions of this command are not supported by PowerGraphics [/GRAPHICS,POWER].
1621
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/SSCALE
Menu Paths
Utility Menu>PlotCtrls>Style>Contours>Contour Style
SSLN, FACT, SIZE
Selects and displays small lines in the model.
PREP7: Lines
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
FACT
Factor used to determine small lines. FACT times the average line length in the model is used as the
line length limit below which lines will be selected.
SIZE
Line length limit for line selection. Lines that have a length less than or equal to SIZE will be selected.
Used only if FACT is blank.
Notes
SSLN invokes a predefined ANSYS macro for selecting small lines in a model. Lines that are smaller than or
equal to the specified limit (FACT or SIZE) are selected and line numbers are displayed. This command
macro is useful for detecting very small lines in a model that may cause problems (i.e., poorly shaped elements
or a meshing failure) during meshing. All lines that are not "small" will be unselected and can be reselected
with the LSEL command.
Menu Paths
Main Menu>Preprocessor>Modeling>Check Geom>Sel Small Lines
SSMT, M
T
11
, M
T
22
, M
T
12
, T
Specifies preintegrated membrane thermal effects for shell sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
M
T
11
, M
T
22
, M
T
12
Membrane thermal effects component [M
T
].
T
Temperature.
Notes
The behavior of shell elements is governed by the generalized-stress/generalized-strain relationship of the
form:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1622
SSLN
N
M
A B
B D
T T
M
B
S E
T
I
T
T
'
1
]
1
1
'
'
( - )
[ ] [ ]{{ }
The SSMT command, one of several preintegrated shell section commands, specifies the membrane thermal
effects quantity (submatrix [M
T
] data) for a preintegrated shell section. The section data defined is associated
with the section most recently defined (via the SECTYPE command).
The [M
T
] quantity represents membrane stress resultants caused by a unit raise in temperature on a fully
constrained model. For a layered composite shell, it is usually necessary to specify both the [M
T
] and [B
T
]
quantities (by issuing the SSMT and SSBT commands, respectively).
Unspecified values default to zero.
Related commands are SSPA, SSPB, SSPD, SSPE, SSBT, and SSPM.
For complete information, see Using Preintegrated General Shell Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
SSPA, A
11
, A
21
, A
31
, A
22
, A
32
, A
33
, T
Specifies a preintegrated membrane stiffness for shell sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
A
11
, A
21
, A
31
, A
22
, A
32
, A
33
Membrane stiffness component (symmetric lower part of submatrix [A]).
T
Temperature.
Notes
The behavior of shell elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
A B
B D
T T
M
B
S E
T
I
T
T
'
1
]
1
1
'
'
( - )
[ ] [ ]{{ }
The SSPA command, one of several preintegrated shell section commands, specifies the membrane stiffness
quantity (submatrix [A]) for a preintegrated shell section. The section data defined is associated with the
section most recently defined (via the SECTYPE command).
Unspecified values default to zero.
1623
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SSPA
Related commands are SSPB, SSPD, SSPE, SSMT, SSBT, and SSPM.
For complete information, see Using Preintegrated General Shell Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
SSPB, B
11
, B
21
, B
31
, B
22
, B
32
, B
33
, T
Specifies a preintegrated coupling stiffness for shell sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
B
11
, B
21
, B
31
, B
22
, B
32
, B
33
Coupling stiffness component (symmetric lower part of submatrix [B]).
T
Temperature.
Notes
The behavior of shell elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
A B
B D
T T
M
B
S E
T
I
T
T
'
1
]
1
1
'
'
( - )
[ ] [ ]{{ }
The SSPB command, one of several preintegrated shell section commands, specifies the coupling stiffness
quantity (submatrix [B] data) for a preintegrated shell section. The section data defined is associated with
the section most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are SSPA, SSPD, SSPE, SSMT, SSBT, and SSPM.
If you are using the SHELL181 or SHELL281 element's Membrane option (KEYOPT(1) = 1), it is not necessary
to issue this command.
For complete information, see Using Preintegrated General Shell Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1624
SSPB
SSPD, D
11
, D
21
, D
31
, D
22
, D
32
, D
33
, T
Specifies a preintegrated bending stiffness for shell sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
D
11
, D
21
, D
31
, D
22
, D
32
, D
33
Bending stiffness component (symmetric lower part of submatrix [D]).
T
Temperature.
Notes
The behavior of shell elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
A B
B D
T T
M
B
S E
T
I
T
T
'
1
]
1
1
'
'
( - )
[ ] [ ]{{ }
The SSPD command, one of several preintegrated shell section commands, specifies the bending stiffness
quantity (submatrix [D] data) for a preintegrated shell section. The section data defined is associated with
the section most recently defined (via the SECTYPE command).
Unspecified commands default to zero.
Related commands are SSPA, SSPB, SSPE, SSMT, SSBT, and SSPM.
If you are using the SHELL181 or SHELL281 element's Membrane option (KEYOPT(1) = 1), it is not necessary
to issue this command.
For complete information, see Using Preintegrated General Shell Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
SSPE, E
11
, E
21
, E
22
, T
Specifies a preintegrated transverse shear stiffness for shell sections.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
E
11
, E
21
, E
22
Transverse shear stiffness component (symmetric lower part of submatrix [E]).
T
Temperature.
1625
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SSPE
Notes
The behavior of shell elements is governed by the generalized-stress/generalized-strain relationship of the
form:
N
M
A B
B D
T T
M
B
S E
T
I
T
T
'
1
]
1
1
'
'
( - )
[ ] [ ]{{ }
The SSPE command, one of several preintegrated shell section commands, specifies the transverse shear
stiffness quantity (submatrix [E] data) for a preintegrated shell section. The section data defined is associated
with the section most recently defined (via the SECTYPE command).
Unspecified values default to zero.
Related commands are SSPA, SSPB, SSPD, SSMT, SSBT, and SSPM.
If you are using the SHELL181 or SHELL281 element's Membrane option (KEYOPT(1) = 1), it is not necessary
to issue this command.
For complete information, see Using Preintegrated General Shell Sections.
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
SSPM, DENS, T
Specifies mass density for a preintegrated shell section.
PREP7: Cross Sections
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
DENS
Mass density.
T
Temperature.
Notes
The SSPM command, one of several preintegrated shell section commands, specifies the mass density (as-
suming a unit thickness) for a preintegrated shell section. The value specified is associated with the section
most recently defined (via the SECTYPE command).
Related commands are SSPA, SSPB, SSPD, SSPE, SSMT, and SSBT.
For complete information, see Using Preintegrated General Shell Sections.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1626
SSPM
Menu Paths
Main Menu>Preprocessor>Sections>Shell>Pre-integrated
SSTIF, Key
Activates stress stiffness effects in a nonlinear analysis.
SOLUTION: Nonlinear Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Key
Stress stiffening key:
OFF
No stress stiffening is included (default unless NLGEOM,ON).
ON
Stress stiffening is included (default if NLGEOM,ON).
Command Default
SSTIF will be turned on if NLGEOM,ON; otherwise it will be turned off.
Notes
Activates stress stiffness effects in a nonlinear analysis (ANTYPE,STATIC or TRANS). (The PSTRES command
also controls the generation of the stress stiffness matrix and therefore should not be used in conjunction
with SSTIF.) If used within the solution processor, this command is valid only within the first load step.
When SOLCONTROL and NLGEOM are ON, SSTIF defaults to ON. This normally forms all of the consistent
tangent matrix. However, for some special nonlinear cases, this can lead to divergence caused by some
elements which do not provide a complete consistent tangent (notably, elements outside the 18x family).
In such a case, ANSYS recommends issuing an SSTIF,OFF command to achieve convergence. For current-
technology elements, setting SSTIF,OFF when NLGEOM is ON has no effect (because stress stiffness effects
are always included).
The default values given for this command assume SOLCONTROL,ON (the default). See the description of
SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and SOLCONTROL,OFF.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
1627
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SSTIF
SSUM
Calculates and prints the sum of element table items.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Calculates and prints the tabular sum of each existing labeled result item [ETABLE] for the selected elements.
If absolute values are requested [SABS,1], absolute values are used.
Menu Paths
Main Menu>General Postproc>Element Table>Sum of Each Item
STABILIZE, Key, Method, VALUE, SubStpOpt, FORCELIMIT
Activates stabilization for all elements that support nonlinear stabilization.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> FL EM EH <> PP <> EME MFS
Key
Key for controlling nonlinear stabilization:
OFF
Deactivate stabilization. This value is the default.
CONSTANT
Activate stabilization. The energy-dissipation ratio or damping factor remains constant during the
load step.
REDUCE
Activate stabilization. The energy-dissipation ratio or damping factor is reduced linearly to zero at
the end of the load step from the specified or calculated value.
Method
The stabilization-control method:
ENERGY
Use the energy-dissipation ratio as the control. This value is the default when Key OFF.
DAMPING
Use the damping factor as the control.
VALUE
The energy-dissipation ratio (Method = ENERGY) or damping factor (Method = DAMPING). This value
must be greater than 0 when Method = ENERGY or Method = DAMPING. When Method = ENERGY,
this value is usually a number between 0 and 1.
SubStpOpt
Option for the first substep of the load step:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1628
SSUM
NO
Stabilization is not activated for the first substep even when it does not converge after the minimal
allowed time increment is reached. This value is the default when Key OFF.
MINTIME
Stabilization is activated for the first substep if it still does not converge after the minimal allowed
time increment is reached.
ANYTIME
Stabilization is activated for the first substep. Use this option if stabilization was active for the previous
load step via Key = CONSTANT.
FORCELIMIT
The stabilization force limit coefficient, such that 0 < FORCELIMIT < 1. The default value is 0.2. To omit
a stabilization force check, set this value to 0.
Command Default
Once issued, the stabilization effects of the command remain until you issue either a STABILIZE command
(with no arguments) or a STABILIZE,OFF command. If you issue the command with no arguments, the effect
is to deactivate stabilization.
Notes
Once issued, a STABILIZE command remains in effect until you reissue the command.
For the energy dissipation ratio, specify VALUE = 1.0e-4 if you have no prior experience with the current
model; if convergence problems are still an issue, increase the value gradually. The damping factor is mesh-
, material-, and time-step-dependent; an initial reference value from the previous run (such as a run with
the energy-dissipation ratio as input) should suggest itself.
Exercise caution when specifying SubStpOpt = MINTIME or ANYTIME for the first load step; ANSYS recom-
mends this option only for experienced users. If stabilization was active for the previous load step via Key
= CONSTANT and convergence is an issue for the first substep, specify SubStpOpt = ANYTIME.
When the L2-norm of the stabilization force (CSRSS value) exceeds the L2-norm of the internal force multiplied
by the stabilization force coefficient, ANSYS issues a message displaying both the stabilization force norm
and the internal force norm. The FORCELIMIT argument allows you to change the default stabilization
force coefficient (normally 20 percent).
This command stabilizes the degrees of freedom for current-technology elements only. Other elements can
be included in the FE model, but their degrees of freedom are not stabilized.
For more information about nonlinear stabilization, see Unstable Structures in the Structural Analysis Guide.
For additional tips that can help you to achieve a stable final model, see Keep It Simple in the Structural
Analysis Guide.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Advanced NL
Main Menu>Solution>Analysis Type>Sol'n Controls>Advanced NL
1629
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
STABILIZE
STAOPT, Method
Specifies static analysis options.
SOLUTION: Analysis Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Method
Solution method for the static analysis:
DEFA
Regular ANSYS solve (default).
VT
Solve with Variational Technology.
Notes
Specifies the method of solution for a static analysis (ANTYPE,STATIC). If used in SOLUTION, this command
is valid only within the first load step.
The VT option using an ANSYS Mechanical HPC license is valid for either thermal or structural nonlinear
analyses, where it attempts to reduce the total number of iterations. You can also use the VT option with a
DX license for ANSYS DesignXplorer design parameter studies.
After a Design Exploration solution (SOLVE command), the Variational Technology model is "frozen", i.e.,
you cannot add or delete input variables.
This command is also valid in PREP7.
Product Restrictions
The VT option is only available with the VT product or a HPC or DX license.
Menu Paths
Main Menu>DesignXplorer>Solution>Solve
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
STAT
Displays the status of database settings.
DATABASE: Set Up
DISPLAY: Action
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
In the DISPLAY program, STAT will show the current status of the program settings.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1630
STAOPT
In the ANSYS program, STAT is a command generated by the GUI and will appear in the log file (Job-
name.LOG) if status is requested for some items under Utility Menu> List> Status. Generally, STAT will
be preceded by one of the commands listed below, which specifies the particular topic that status was re-
quested for.
If entered directly into the program, the STAT command should be immediately preceded by the desired
topic command listed below. In processors other than those listed below (e.g., AUX12), no topic command
should proceed STAT.
This command is valid in any processor.
PREP7 topic commands (and their corresponding topics) are:
Topic Topic Command
Element types ETYPE
Real constants RCON
Material properties MATER
Data table properties TBLE
Solid model primitives PRIM
Keypoints KEYPTS
Lines LINE
Areas AREAS
Volumes VOLUMES
Solid model information GEOMETRY
Meshing MESHING
Booleans BOOL
Nodes NODES
Elements ELEM
Superelements SELM
Pipe modeling PIPE
Node digitizing DIGIT
Node coupling COUPLE
Constraint equations CEQN
Model reordering REORDER
SOLUTION topic commands (and their corresponding topics) are:
Topic Topic Command
Analysis types ATYPE
Master DOF MASTER
Reduced transient gap conditions GAP
Element birth and death (deactivation) DEACT
Load step operations LSOPER
Constraints on nodes FECONS
1631
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
STAT
Topic Topic Command
Forces on nodes FEFOR
Surface loads on elements FESURF
Body loads on elements FEBODY
Constraints on the solid model SMCONS
Forces on the solid model SMFOR
Surface loads on the solid model SMSURF
Body loads on the solid model SMBODY
Inertial loads INRTIA
General options GENOPT
Dynamic analysis options DYNOPT
Nonlinear analysis options NLOPT
Output options OUTOPT
Biot-Savart options BIOOPT
Spectrum analysis options SPTOPT
Solution options SOLUOPT
FLOTRAN data settings FLOTRAN
POST1 topic commands (and their corresponding topics) are:
Topic Topic Command
Data definition settings DEFINE
Sort settings SORT
Print settings PRINT
Display settings DISPLAY
Calculation settings CALC
Path data settings PATH
Load case settings LCCALC
Directly defined data status DATADEF
Fatigue data status FATIGUE
Point flow tracing settings POINT
Miscellaneous specifications SPEC
POST26 topic commands (and their corresponding topics) are:
Topic Topic Command
Data definition settings DEFINE
Operation data OPERATE
Print settings PRINT
Plotting settings PLOTTING
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1632
STAT
Menu Paths
Main Menu>Prob Design>Prob Database>Status
Main Menu>Solution>Solve>Current LS
*STATUS, Par, IMIN, IMAX, JMIN, JMAX, KMIN, KMAX, LMIN, LMAX, MMIN, MMAX, KPRI
Lists the current parameters and abbreviations.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
Specifies the parameter or sets of parameters listed. For array parameters, use IMIN, IMAX, etc. to specify
ranges. Use *DIM to define array parameters. Use *VEDIT to review array parameters interactively. Use
*VWRITE to print array values in a formatted output. If Par is blank, list all scalar parameter values, array
parameter dimensions, and abbreviations. If ARGX, list the active set of local macro parameters (ARG1
to ARG9 and AR10 to AR99) [*USE].
The following are possible values for Par
ALL or blank --
Lists all parameters (except local macro parameters and those with names beginning or ending with
an underbar) and toolbar abbreviations.
_PRM --
Lists only parameters with names beginning with an underbar (_). These are ANSYS internal paramet-
ers.
PRM_ --
Lists only parameters with names ending with an underbar (_). A good APDL programming convention
is to ensure that all parameters created by your system programmer are named with a trailing un-
derbar.
ABBR --
Lists all toolbar abbreviations.
PARM --
Lists all parameters (except local macro parameters and those with names beginning or ending with
an underbar).
PARNAME --
Lists only the parameter specified. PARNAME cannot be a local macro parameter name.
ARGX --
Lists all local macro parameter values (ARG1- AR99) that are non-zero or non-blank.
IMIN, IMAX, JMIN, JMAX, KMIN, KMAX, LMIN, LMAX, MMIN, MMAX
Range of array elements to display (in terms of the dimensions (row, column, plane, book, and shelf ).
Minimum values default to 1. Maximum values default to the maximum dimension values. Zero may be
input for IMIN, JMIN, and KMIN to display the index numbers. See *TAXIS command to list index
numbers of 4- and 5-D tables.
1633
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*STATUS
KPRI
Use this field to list your primary variable labels (X, Y, Z, TIME, etc.).
1
List the labels (default). YES, Y, or ON are also valid entries.
0
Do not list the labels. NO, N, or OFF are also valid entries.
Notes
You cannot obtain the value for a single local parameter (e.g., *STATUS,ARG2). You can only request all
local parameters simultaneously using *STATUS,ARGX.
This command is valid in any processor.
Menu Paths
Utility Menu>List>Other>Named Parameter
Utility Menu>List>Other>Parameters
Utility Menu>List>Status>Parameters>All Parameters
Utility Menu>List>Status>Parameters>Named Parameters
/STATUS, Lab
Lists the status of items for the run.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Items to list status for:
ALL
List all below (default).
TITLE
List only titles, Jobname, and revision number.
UNITS
List only units.
MEM
List only memory data statistics.
DB
List only database statistics
CONFIG
List only configuration parameters.
GLOBAL
Provides a global status summary.
SOLU
Provides a solution status summary.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1634
/STATUS
PROD
Provides a product summary.
Notes
Displays various items active for the run (such as the ANSYS revision number, Jobname, titles, units, config-
uration parameters, database statistics, etc.).
This command is valid in any processor.
Menu Paths
Utility Menu>List>Status>Global Status
STEF, VALUE
Specifies Stefan-Boltzmann radiation constant.
AUX12: General Radiation
SOLUTION: Radiosity
MP ME ST PR <> <> <> <> <> <> <> PP <> EME MFS
VALUE
Stefan-Boltzmann constant (defaults to 0.119E-10 Btu/hr/in
2
/ R
4
).
Command Default
0.119E-10 Btu/hr/in
2
/ R
4
.
Notes
You can use this command in the general preprocessor (PREP7) and in the Solution processor to specify the
Stefan-Boltzmann constant for a FLOTRAN analysis using radiation surface boundary conditions.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Displacement>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Displacement>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Displacement>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Displacement>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Forces>Body Forces>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Turbulence>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Turbulence>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Turbulence>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Velocity>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Velocity>On Keypoints
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Velocity>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Velocity>On Nodes
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Volume Fract>Bound Loads>On
Elements
1635
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
STEF
Main Menu>Preprocessor>Loads>Define Loads>Apply>Fluid/CFD>Volume Fract>Bound Loads>On
Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Ambient Rad>On Areas
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Ambient Rad>On Elements
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Ambient Rad>On Lines
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Ambient Rad>On Nodes
Main Menu>Preprocessor>Radiation Opts>Solution Opt
Main Menu>Radiation Opt>Matrix Method>Other Settings
Main Menu>Radiation Opt>Radiosity Meth>Solution Opt
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Displacement>On Areas
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Displacement>On Keypoints
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Displacement>On Lines
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Displacement>On Nodes
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Forces>Body Forces>On Nodes
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Turbulence>On Areas
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Turbulence>On Lines
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Turbulence>On Nodes
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Velocity>On Areas
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Velocity>On Keypoints
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Velocity>On Lines
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Velocity>On Nodes
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Volume Fract>Bound Loads>On Elements
Main Menu>Solution>Define Loads>Apply>Fluid/CFD>Volume Fract>Bound Loads>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Ambient Rad>On Areas
Main Menu>Solution>Define Loads>Apply>Thermal>Ambient Rad>On Elements
Main Menu>Solution>Define Loads>Apply>Thermal>Ambient Rad>On Lines
Main Menu>Solution>Define Loads>Apply>Thermal>Ambient Rad>On Nodes
Main Menu>Solution>Radiation Opts>Solution Opt
/STITLE, NLINE, Title
Defines subtitles.
DATABASE: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NLINE
Subtitle line number (1 to 4). Defaults to 1.
Title
Input up to 70 alphanumeric characters. Parameter substitution may be forced within the title by enclosing
the parameter name or parametric expression within percent (%) signs. If Title is blank, this subtitle
is deleted.
Notes
Subtitles (4 maximum) are displayed in the output along with the main title [/TITLE]. Subtitles do not appear
in GUI windows or in ANSYS plot displays. The first subtitle is also written to various ANSYS files along with
the main title. Previous subtitles may be overwritten or deleted. Issue /STATUS to display titles.
This command is valid in any processor.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1636
/STITLE
Menu Paths
This command cannot be accessed from a menu.
STORE, Lab, NPTS
Stores data in the database for the defined variables.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Valid labels:
MERGE
Merge data from results file for the time points in memory with the existing data using current
specifications (default).
NEW
Store a new set of data, replacing any previously stored data with current result file specifications
and deleting any previously-calculated (OPER) variables. Variables defined using the ANSOL command
are also deleted.
APPEN
Append data from results file to the existing data.
ALLOC
Allocate (and zero) space for NPTS data points.
PSD
Create a new set of frequency points for PSD calculations (replacing any previously stored data and
erasing any previously calculated data).
NPTS
The number of time points (or frequency points) for storage (used only with Lab = ALLOC or PSD). The
value may be input when using POST26 with data supplied from other than a results file. This value is
automatically determined from the results file data with the NEW, APPEN, and MERGE options. For the
PSD option, NPTS determines the resolution of the frequency vector (valid numbers are between 1 and
10, defaults to 5).
Command Default
Merge newly-defined variables with previously stored variables for the time points stored in memory using
the current specifications. If STORE is preceded by TIMERANGE or NSTORE, the default is STORE,NEW.
Notes
This command stores data from the results file in the database for the defined variables [NSOL, ESOL, SOLU,
JSOL] per specification [FORCE, LAYERP26, SHELL]. See the Basic Analysis Guide for more information.
The STORE,PSD command will create a new frequency vector (variable 1) for response PSD calculations
[RPSD]. This command should first be issued before defining variables [NSOL, ESOL, RFORCE] for which
response PSD's are to be calculated.
1637
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
STORE
Menu Paths
Main Menu>TimeHist Postpro>Store Data
SUBSET, Lstep, SBSTEP, FACT, KIMG, TIME, ANGLE, NSET
Reads results for the selected portions of the model.
POST1: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lstep
Load step number of the data set to be read (defaults to 1):
N
Read load step N.
FIRST
Read the first data set (SBSTEP and TIME are ignored).
LAST
Read the last data set (SBSTEP and TIME are ignored).
NEXT
Read the next data set (SBSTEP and TIME are ignored). If at the last data set, the first data set will
be read as the next.
NEAR
Read the data set nearest to TIME (SBSTEP is ignored). If TIME is blank, read the first data set.
LIST
Scan the results file and list a summary of each load step. (FACT, KIMG, TIME and ANGLE are ig-
nored.)
SBSTEP
Substep number (within Lstep). For the buckling (ANTYPE,BUCKLE) analysis or the modal (ANTYPE,MOD-
AL) analysis, the substep corresponds to the mode number. Defaults to last substep of load step (except
for ANTYPE,BUCKLE or MODAL). If Lstep = LIST, SBSTEP = 0 or 1 lists the basic step information,
whereas SBSTEP = 2 also lists the load step title, and labels imaginary data sets if they exist.
FACT
Scale factor applied to data read from the file. If zero (or blank), a value of 1.0 is used. Harmonic velocities
or accelerations may be calculated from the displacement results from a modal (ANTYPE,MODAL) or
harmonic response (ANTYPE,HARMIC) analyses. If FACT = VELO, the harmonic velocities (v) are calculated
from the displacements (d) at a particular frequency (f ) according to the relationship v = 2 fd. Similarly,
if FACT = ACEL, the harmonic accelerations (a) are calculated as a = (2 f )
2
d.
KIMG
Used only with results from complex analyses:
0
Store real part of complex solution
1
Store imaginary part.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1638
SUBSET
TIME
Time-point identifying the data set to be read. For the harmonic responses analyses, time corresponds
to the frequency. For the buckling analysis, time corresponds to the load factor. Used only in the following
cases: If Lstep is NEAR, read the data set nearest to TIME. If both Lstep and SBSTEP are zero (or
blank), read data set at time = TIME. If TIME is between two solution time points on the results file, a
linear interpolation is done between the two data sets. Solution items not written to the results file
[OUTRES] for either data set will result in a null item after data set interpolation. If TIME is beyond the
last time point on the file, use the last time point.
ANGLE
Circumferential location (0.0 to 360). Defines the circumferential location for the harmonic calculations
used when reading from the results file. The harmonic factor (based on the circumferential angle) is
applied to the harmonic elements (PLANE25, PLANE75, PLANE78, FLUID81, PLANE83, and SHELL61) of
the load case. See the Theory Reference for the Mechanical APDL and Mechanical Applications for details.
Note that factored values of applied constraints and loads will overwrite any values existing in the
database.
NSET
Data set number of the data set to be read. If a positive value for NSET is entered, Lstep, SBSTEP,
KIMG, and TIME are ignored. Available set numbers can be determined by *SET,LIST.
Notes
Reads a data set from the results file into the database for the selected portions of the model only. Data
that has not been specified for retrieval from the results file by the INRES command will be listed as having
a zero value. Each time that the SUBSET command is issued, the data currently in the database will be
overwritten with a new set of data. Various operations may also be performed during the read operation.
The database must have the model geometry available (or used the RESUME command before the SUBSET
command to restore the geometry from File.DB).
Menu Paths
Main Menu>General Postproc>Read Results>By Load Step
Main Menu>General Postproc>Read Results>By Set Number
Main Menu>General Postproc>Read Results>By Time/Freq
SUCALC, RSetName, lab1, Oper, lab2, fact1, fact2, const
Create new result data by operating on two existing result data sets on a given surface.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSetName
Eight character name for new result data.
lab1
First result data upon which to operate.
Oper
Mathematical operation to perform.
1639
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SUCALC
ADD
(lab1 + lab2 + const)
SUB
(lab1 - lab2 + const)
MULT
(lab1 * lab2 + const)
DIV
(lab1 / lab2 + const)
EXP
(lab1 ^ fact1 + lab2 ^ fact2 + const)
COS
(cos (lab1) + const)
SIN
(sin (lab1) + const)
ACOS
(acos (lab1) + const)
ASIN
(asin (lab1) + const)
ATAN
(atan (lab1) + const)
ATA2
(atan2 (lab1 / lab2) + const)
LOG
(log (lab1) + const)
ABS
(abs (lab1) + const)
ZERO
(0 + const)
lab2
Second result data upon which to operate.
fact1
First scaling factor (for EXP option only).
fact2
Second scaling factor (for EXP option only).
const
Constant added to the values in the resulting data.
Menu Paths
Main Menu>General Postproc>Surface Operations>Math Operations>Absolute
Main Menu>General Postproc>Surface Operations>Math Operations>Add
Main Menu>General Postproc>Surface Operations>Math Operations>ArcCosine
Main Menu>General Postproc>Surface Operations>Math Operations>ArcSine
Main Menu>General Postproc>Surface Operations>Math Operations>ArcTangent
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1640
SUCALC
Main Menu>General Postproc>Surface Operations>Math Operations>ArcTangent2
Main Menu>General Postproc>Surface Operations>Math Operations>Cosine
Main Menu>General Postproc>Surface Operations>Math Operations>Divide
Main Menu>General Postproc>Surface Operations>Math Operations>Exponentiate
Main Menu>General Postproc>Surface Operations>Math Operations>Initialize
Main Menu>General Postproc>Surface Operations>Math Operations>Multiply
Main Menu>General Postproc>Surface Operations>Math Operations>Natural Log
Main Menu>General Postproc>Surface Operations>Math Operations>Sine
Main Menu>General Postproc>Surface Operations>Math Operations>Subtract
SUCR, SurfName, SurfType, nRefine, Radius, blank, blank, TolOut
Create a surface.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SurfName
Eight character surface name.
SurfType
Surface type.
CPLANE
Surface is defined by the cutting plane in window one (controlled by the working plane (/CPLANE,1),
NOT the view settings (/CPLANE,0)).
SPHERE
Surface is defined by a spherical surface centered about the working plane origin.
INFC
Surface is defined by a cylindrical surface centered about the working plane origin and extending
indefinitely in the positive and negative Z directions.
nRefine
Refinement level.
For SurfType = CPLANE
The refinement level of the surface "mesh". This will be an integer between 0 and 3 (default = 0).
See Notes below.
For SurfType = SPHERE
The number of divisions along a 90 arc (minimum = 9). The default is 9.
For SurfType = INFC
The number of divisions along a 90 arc (minimum = 9). The default is 9.
Radius
Appropriate radius value (for INFC or SPHERE).
TolOut
Tolerance value for inclusion of element facets within a prescribed volume. (for INFC)
Notes
This command creates a new surface and stores the following data for that surface:
1641
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SUCR
GCX, GCY, GCZ - global Cartesian coordinates at each point on the surface.
NORMX, NORMY, NORMZ - components of the unit normal at each point on the surface.
DA - the contributory area of each point.
For SurfType = CPLANE, nRefine refers to the number of points that define the surface. An nRefine
value of zero is used for points where the element face intersects the cutting plane.
If SurfType = CPLANE and nRefine = 0, the points reside at the section cuts where the element intersects
the cutting plane. Increasing nRefine from 0 to 1 will subdivide each surface facet into 4 subfacets, and
increase the number of points at which results can be interpolated.
For SurfType = CPLANE , the setting from the /EFACET command will affect the creation of surface facets
and the quality of the fit of the surface in the model. SUCR employs geometry data from PowerGraphics to
aid in determining where the surface intersects the model. If /EFACET,1 is in effect when the SUCR command
is issued, then the curvature of high order elements (that is, elements with midside nodes) will be ignored.
If your model contains high order elements, you can see a better fit for your surface if /EFACET,2 is in effect
when the SUCR command is issued. Currently, the SUCR command interprets /EFACET,4 to mean /EFACET,2.
For SurfType = INFC, a default tolerance of 0.01 will be applied to include the vertices of any facets that
fall out of the cylinder definition. this tolerance increases the facet size by 1% to check for inclusion. Excluding
facets under such a small tolerance may yield unacceptable (aesthetically) results. Increasing the tolerance
by a larger amount (0.1 or 10%) will sometimes yield smother edges along the surface you create.
Menu Paths
Main Menu>General Postproc>Surface Operations>Create Surface>Inf. Cylinder
Main Menu>General Postproc>Surface Operations>Create Surface>On Cuttng Plane
Main Menu>General Postproc>Surface Operations>Create Surface>Sphere>At Node
Main Menu>General Postproc>Surface Operations>Create Surface>Sphere>By Dimensions
SUDEL, SurfName
Delete geometry information as well as any mapped results for specified surface.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SurfName
Eight character surface name.
SurfName = ALL will delete all surface geometry and result infromation.
Menu Paths
Main Menu>General Postproc>Surface Operations>Delete Surfaces
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1642
SUDEL
SUEVAL, Parm, lab1, Oper
Perform operations on a mapped item and store result in a scalar parameter.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Parm
APDL parameter name.
lab1
Eight character set name for the first set used in calculation.
Oper
Operation to perform:
SUM
Sum of lab1 result values.
INTG
Integral of lab1 over surface.
AVG
Area-weighted average of a result item [(lab1*DA) / (DA)]
Notes
The result of this operation is a scalar APDL parameter value. If multiple surfaces are selected when this
command is issued, then the operation is carried out on each surface individually and the parameter reper-
esents the culmulative value of the operation on all selected surfaces.
Menu Paths
Main Menu>General Postproc>Surface Operations>Math Operations>Average Result
Main Menu>General Postproc>Surface Operations>Math Operations>Integrate Results
Main Menu>General Postproc>Surface Operations>Math Operations>Sum of Results
SUGET, SurfName, RSetName, Parm, Geom
Moves surface geometry and mapped results to an array parameter.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SurfName
Eight character surface name.
RSetName
Eight character result name.
Parm
APDL array parameter name (up to 32 characters).
Geom
Switch controlling how data is written.
1643
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SUGET
ON (or 1 or YES)
Writes geometry data and interpolated results information to the parameter.
OFF (or 0 or NO)
Writes only interpolated results information to the parameter. (Default)
Notes
For Geom = OFF (or 0 or NO), only results information is written to this parameter.
For Geom = ON (or 1 or YES), both geometry data and results information are written to this parameter.
Geometry data includes 7 data items: (GCX, GCY, GCZ, NORMX, NORMY, NORMZ, and DA). Results information
is then written to the 8th column of the parameter. SetNames of GCX, GCY, GCZ, NORMX, NORMY, NORMZ,
and DA are predefined and computed when SUCR is issued.
Menu Paths
Main Menu>General Postproc>Surface Operations>Results to Array
SUMAP, RSetName, Item, Comp
Map results onto selected surface(s).
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSetName
Eight character name for the result being mapped.
Item
Label identifying the item.
Valid item labels are defined in PLNSOL. Some items also require a component label.
If Item = CLEAR, the specified result set will be deleted from all selected surfaces
Comp
Component label of item (if required).
Notes
Results are mapped in current RSYS. This command interpolates and stores results data on to each of the
selected surfaces. SUMAP, ALL, CLEAR will delete all results sets from all selected surfaces.
Menu Paths
Main Menu>General Postproc>Surface Operations>Clear Results
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1644
SUMAP
SUMTYPE, Label
Sets the type of summation to be used in the following load case operations.
POST1: Results
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Summation type
COMP
Combine element component stresses only. Stresses such as average nodal stresses, principal stresses,
equivalent stresses, and stress intensities are derived from the combined element component stresses.
Default.
PRIN
Combine principal stress, equivalent stress, and stress intensity directly as stored on the results file.
Component stresses are not available with this option.
Notes
Issue SUMTYPE,PRIN when you want to have a load case operation (LCOPER) act on the principal / equivalent
stresses instead of the component stresses. Also issue SUMTYPE,PRIN when you want to read in load cases
(LCASE). Note that the SUMTYPE setting is not maintained between /POST1 sessions.
SUMTYPE,PRIN also causes principal nodal values to be the average of the contibuting principal element
nodal values (see AVPRIN,1).
BEAM188 and BEAM189 elements compute principal stress, equivalent stress, and stress intensity values on
request instead of storing them on the results file; SUMTYPE,PRIN does not apply for these elements.
Menu Paths
Main Menu>General Postproc>Load Case>Calc Options>Stress Options
SUPL, SurfName, RSetName, KWIRE
Plot result data on all selected surfaces or on a specified surface.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SurfName
Eight character surface name. ALL will plot all selected surfaces.
RSetName
Eight character result name.
KWIRE
Plot in context of model.
0
Plot results without the outline of selected elements.
1645
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SUPL
1
Plot results with the outline of selected elements.
Notes
If RSetName is left blank, then the surface geometry will be plotted. If the Setname portion of the argument
is a vector prefix (i.e. if result sets of name SetNameX, SetNameY and SetNameZ exist), ANSYS will plot these
vectors on the surface as arrows. For example, SUPL, ALL, NORM will plot the surface normals as vectors on
all selected surfaces, since NORMX NORMY and NORMZ are pre-defined geometry items.
Menu Paths
Main Menu>General Postproc>Surface Operations>Plot Results
Main Menu>General Postproc>Surface Operations>Plot Vectors
SUPR, SurfName, RSetName
Print global status, geometry information and/or result information.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SurfName
Eight character surface name. If SurfName = ALL, repeat printout for all selected surfaces.
RSetName
Eight character result set name.
Notes
When no arguments are specified, SUPR generates a global status summary of all defined surfaces. If only
SurfName is specified, the geometry information for that surface is printed. If both SurfName and RSet-
Name are specified, the value of the results set at each point, in addition to the geometry information, is
printed.
Menu Paths
Main Menu>General Postproc>Surface Operations>Print Results
Main Menu>General Postproc>Surface Operations>Status>Global
SURESU, --, Fname, Fext, Fdir
Read a set of surface definitions and result items from a file and make them the current set.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
--
Unused field.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1646
SUPR
Fname
Eight character name.
Fext
Extension name.
Fdir
Optional path specification.
Notes
Reading (and therefore resuming) surface and result definitions from a file overwritea any existing surface
definitions.
Reading surfaces back into the postprocessor (/POST1) does not insure that the surfaces (and their results)
are appropriate for the model currently residing in /POST1.
Menu Paths
Main Menu>General Postproc>Surface Operations>Resume Surfaces
SUSAVE, Lab, Fname, Fext, Fdir
Saves surface definitions to a file.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Eight-character surface name.
If Lab = ALL (default), then all surfaces are saved to the file.
If Lab = S, only currently selected surfaces are saved to the file.
Fname
File name and directory path (248 character maximum, including directory). If you do not specify a dir-
ectory path, the default is your working directory and you can use all 248 characters for the file name.
The file name defaults to the jobname.
Fext
File name extension (eight-character maximum). The extension defaults to surf.
Fdir
Optional path specification.
Notes
The SUSAVE command saves surface definitions (geometry information)--and any result items mapped onto
the surfaces--to a file.
Issuing the SUSAVE command has no effect on the database. The database remains unchanged.
Subsequent executions of the SUSAVE command overwrite previous data in the file.
1647
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SUSAVE
To read the contents of the file created via the SUSAVE command, issue the SURESU command.
Menu Paths
Main Menu>General Postproc>Surface Operations>Save Surfaces
SUSEL, Type, Name1, Name2, Name3, Name4, Name5, Name6, Name7, Name8
Selects a subset of surfaces
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of select:
S
Selects a new set (default).
R
Reselects a set from the current set.
A
Additionally selects a set and extends the current set.
U
Unselects a set from the current set.
ALL
Also selects all surfaces.
NONE
Unselects all surfaces.
Name1, Name2, Name3, . . . , Name8
Eight character surface names
Notes
The selected set of surfaces is used in the following operations: SUMAP, SUDEL, SUCALC, SUEVAL, and
SUVECT.
Menu Paths
Main Menu>General Postproc>Surface Operations>Select Surfaces
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1648
SUSEL
SUVECT, RSetName, lab1, Oper, lab2, Offset
Create new result data by operating on two existing result vectors on a given surface.
POST1: Surface Operations
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RSetName
Eight character name of the result data output. There will be one or three RSetName values depending
on the operation specified in Oper.
lab1
Eight character name of the mapped data that forms vector 1. Specified sets must exist on all selected
surfaces for this operation to take place. The names NORM and GC will be reserved for normals and for
global (x, y, z).
Oper
DOT
Computes dot product between lab1 and lab2 vectors. The result is a scalar parameter (RSetName)
and each value within the set can be modified (incremented) via Offset.
CROSS
Computes cross product between lab1 and lab2 vectors. Each X, Y, Z value in the result can be
modified (incremented) via Offset.
SMULT
Scales (lab1x, lab1y, lab1z) vector by scalar lab2. Each X,Y,Z value in the result can be modified
(incremented) via Offset.
lab2
Eight character name of the mapped data that forms vector 2. Sets with names Lab2X, Lab2Y, and Lab2Z
must exist on all selected surfaces for operation to take place. For Oper = SMULT a scalar value or an-
other predefined scalar item (e.g., DA) can be supplied.
Offset
An offset value to be applied to the resultant RSetName. One value is specified for Oper = DOT, and
three values are specified for Oper = SMULT.
Menu Paths
Main Menu>General Postproc>Surface Operations>Math Operations>Cross Product
Main Menu>General Postproc>Surface Operations>Math Operations>Dot Product
Main Menu>General Postproc>Surface Operations>Math Operations>Scale Result
1649
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SUVECT
SV, DAMP, SV1, SV2, SV3, SV4, SV5, SV6, SV7, SV8, SV9
Defines spectrum values to be associated with frequency points.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
DAMP
Damping ratio for this response spectrum curve. If the same as a previously defined curve, the SV values
are added to the previous curve. Up to four different curves may be defined, each with a different
damping ratio. Damping values must be input in ascending order.
SV1, SV2, SV3, . . . , SV9
Spectrum values corresponding to the frequency points [FREQ]. Values are interpreted as defined with
the SVTYP command. SV values should not be zero. Values required outside the frequency range use
the extreme input values.
Notes
Defines the spectrum values to be associated with the previously defined frequency points [FREQ]. Applies
only to the single-point response spectrum. Damping has no effect on the frequency solution. Damping
values are used only to identify SV curves for the mode combinations calculation. Only the curve with the
lowest damping value is used in the initial mode coefficient calculation. Use STAT command to list current
spectrum curve values.
Repeat SV command for additional SV points (100 maximum per DAMP curve). SV values are added to the
DAMP curve after the last nonzero SV value.
The interpolation method between response spectrum points and curves is specified using KeyInterp in
the SVTYP command. It is logarithmic by default.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Spectr Values
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Spectr Values
SVPLOT, OptionScale, damp1, damp2, damp3, damp4
Displays input spectrum curves.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
OptionScale
Flag to activate or deactivate input spectrum value scaling:
OFF
Do not scale the input spectrum values with scale factor FACT (SVTYP command). This is the default
value.
ON
Scale the input spectrum values with scale factor FACT (SVTYP command)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1650
SV
damp1
Damping ratio corresponding to DAMP (SV command) defining the first spectrum curve.
damp2
Damping ratio corresponding to DAMP (SV command) defining the second spectrum curve.
damp3
Damping ratio corresponding to DAMP (SV command) defining the third spectrum curve.
damp4
Damping ratio corresponding to DAMP (SV command) defining the fourth spectrum curve.
Notes
You can display up to four input spectrum tables (SV and FREQ commands) with log X scale. If no damping
ratio is specified, all spectrum tables are displayed.
This command is valid in any processor.
Menu Paths
To be supplied.
SVTYP, KSV, FACT, KeyInterp
Defines the type of single-point response spectrum.
SOLUTION: Spectrum Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
KSV
Response spectrum type:
0
Seismic velocity response spectrum loading (SV values interpreted as velocities with units of
length/time).
1
Force response spectrum loading (SV values interpreted as force amplitude multipliers).
2
Seismic acceleration response spectrum loading (SV values interpreted as accelerations with units
of length/time
2
).
3
Seismic displacement response spectrum loading (SV values interpreted as displacements with units
of length).
4
PSD loading (SV values interpreted as acceleration
2
/(cycles/time), such as (in/sec
2
)
2
/Hz (not g
2
/Hz)).
(Not recommended)
FACT
Scale factor applied to spectrum values (defaults to 1.0). Values are scaled when the solution is initiated
[SOLVE]. Database values remain the same.
1651
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SVTYP
KeyInterp
Key to activate or deactivate the linear interpolation between input response spectrum points and input
response spectrum curves:
0 (OFF or NO)
Deactivate linear and use logarithmic interpolation. This value is the default.
1 (ON or YES)
Activate linear interpolation.
Command Default
Seismic velocity response spectrum.
Notes
Defines the type of single-point response spectrum [SPOPT]. The seismic excitation direction is defined with
the SED command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>SinglePt>Settings
Main Menu>Solution>Load Step Opts>Spectrum>SinglePt>Settings
SWADD, Ecomp, SHRD, NCM1, NCM2, NCM3, NCM4, NCM5, NCM6, NCM7, NCM8, NCM9
Adds more surfaces to an existing spot weld set.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Ecomp
Name of an existing spot weld set that was previously defined using SWGEN.
SHRD
Search radius. Defaults to 4 times the spot weld radius defined for the spot weld set (SWRD on SWGEN).
NCM1, NCM2, NCM3, . . . , NCM9
Surfaces to be added to the spot weld set. Each surface can be input as a predefined node component
or a meshed area number.
Notes
This command adds surfaces to an existing spot weld set defined by the SWGEN command. You can add
additional surfaces by repeating the SWADD command. However, the maximum number of allowable surfaces
(including the 2 surfaces used for the original set defined by SWGEN) for each spot weld set is 11. See
Adding Surfaces to a Basic Set for more information.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>SpotWeld>Add More Surfaces>By Areas
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1652
SWADD
SWDEL, Ecomp
Deletes spot weld sets.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Ecomp
Name of an existing spot weld set that was previously defined using SWGEN. If Ecomp = ALL (default)
all spot welds are deleted.
Notes
This command deletes spot weld sets previously defined by the SWGEN command.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>SpotWeld>Delete
SWGEN, Ecomp, SWRD, NCM1, NCM2, SND1, SND2, SHRD, DIRX, DIRY, DIRZ, ITTY, ICTY
Creates a new spot weld set.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Ecomp
Name to identify the new spot weld. This name will be used for the element component containing the
new contact, target, and beam elements generated for the spot weld set.
SWRD
Spot weld radius.
NCM1
Name of a component containing nodes on the first spot weld surface, or a meshed area number for
the surface.
NCM2
Name of a component containing nodes on the second spot weld surface, or a meshed area number
for the surface.
SND1
Node number of the first spot weld node corresponding to the first surface (NCM1). This node can be
on or close to the first surface.
SND2
Node number of the second spot weld node corresponding to the second surface (NCM2). This node
can be on or close to the second surface. ANSYS will create the node if it is not specified.
SHRD
Search radius. Defaults to 4 times the spot weld radius SWRD.
DIRX, DIRY, DIRZ
Spot weld projection direction in terms of normal X, Y, and Z components.
1653
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SWGEN
ITTY
Target element type ID.
ICTY
Contact element type ID.
Notes
This command creates a new spot weld set. You can add more surfaces to the set using SWADD after the
initial SWGEN command. However, the maximum number of allowable surfaces (including the 2 surfaces
used for the original set) for each spot weld set is 11.
Ecomp, SWRD, NCM1, NCM2, and SND1 must be specified. SND2, SHRD, DIRX, DIRY, DIRZ, ITTY, ICTY
are optional inputs. If the second spot weld node (SND2) is specified, that node is used to determine the
spot weld projection direction, andDIRX, DIRY and DIRZ are ignored.
If ITTY is specified, the following corresponding target element key option must be set: KEYOPT(5) = 4. If
ICTY is specified, the following corresponding contact element key options must be set: KEYOPT(2) = 2,
KEYOPT(12) = 5.
Use the SWLIST and SWDEL commands to list or delete spot welds. See Creating a Basic Spot Weld Set with
SWGEN for detailed information on defining spot welds.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>SpotWeld>Add More Surfaces>By Node
Component
Main Menu>Preprocessor>Modeling>Create>Elements>SpotWeld>Create New Set>By Area
Main Menu>Preprocessor>Modeling>Create>Elements>SpotWeld>Create New Set>By Node Com-
ponent
SWLIST, Ecomp
Lists spot weld sets.
PREP7: Elements
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Ecomp
Name of an existing spot weld set that was previously defined using SWGEN. If Ecomp = ALL (default),
all spot weld sets are listed.
Notes
This command lists spot weld node, beam, and contact pair information for all defined spot weld sets, or
for the specified set. To ensure that all defined spotwelds are listed, issue CMSEL,ALL (to select all components)
before issuing the SWLIST command.
When SWLIST is issued in POST1, the beam forces and moments are output. For the case of a deformable
spot weld, the stresses are also output in the beam local coordinate system.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1654
SWLIST
Menu Paths
Main Menu>General Postproc>List Results>SpotWeld Solution
Main Menu>Preprocessor>Modeling>Create>Elements>SpotWeld>List
Utility Menu>List>Results>SpotWeld Solution
SYNCHRO, RATIO, Cname
Specifies whether the excitation frequency is synchronous or asynchronous with the rotational velocity
of a structure.
SOLUTION: Inertia
MP ME ST <> <> <> <> <> EM <> <> PP <> EME MFS
RATIO
The ratio between the frequency of excitation and the frequency of the rotational velocity of the structure.
This value must be greater than 0. The default is an unbalance excitation (RATIO = 1.0).
Cname
The name of the rotating component on which to apply the harmonic excitation.
Notes
The SYNCHRO command specifies whether the excitation frequency is synchronous or asynchronous with
the rotational velocity of a structure in a harmonic analysis. Use the command to take into account rotating
harmonic forces on rotating structures.
ANSYS calculates the rotational velocity of the structure from the excitation frequency f, defined (via the
HARFRQ command) as = 2f / RATIO. The rotational velocity is applied along the direction cosines of
the rotation axis (specified via an OMEGA or CMOMEGA command).
Specifying any value for RATIO causes a general rotational force excitation and not an unbalance force. To
define an unbalance excitation force (F =
2
* Unb), RATIO should be left blank (the nodal unbalance Unb
is specified via the F command).
The SYNCHRO command is valid only for a full-solution harmonic analysis (HROPT,Method = FULL) involving
a rotating structure (OMEGA or CMOMEGA) with Coriolis enabled in a stationary reference frame (CORIOL-
IS,,,,RefFrame = ON).
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Inertia>Angular Veloc>Synchron-
ous
Main Menu>Solution>Define Loads>Apply>Structural>Inertia>Angular Veloc>Synchronous
1655
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
SYNCHRO
/SYP, String, ARG1, ARG2, ARG3, ARG4, ARG5, ARG6, ARG7, ARG8
Passes a command string and arguments to the operating system.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
String
Command string (cannot include commas). See also the /SYS command.
ARG1, ARG2, ARG3, . . . , ARG8
Arguments to be appended to the command string, separated by blanks, commas, or other delimiter
characters (see the Operations Guide). The arguments may be numbers, parameters, or parametric expres-
sions.
Notes
Passes a command string to the operating system for execution, along with arguments to be appended to
the command string. See the Operations Guide for details. ANSYS may not be aware of your specific user
environment. In particular, this command may not recognize UNIX aliases, depending on the hardware
platform and user environment.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
/SYS, String
Passes a command string to the operating system.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
String
Command string, up to 75 characters (including blanks, commas, etc.). The specified string is passed
verbatim to the operating system, i.e., no parameter substitution is performed.
Notes
Passes a command string to the operating system for execution (see the Operations Guide). Typical strings
are system commands such as list, copy, rename, etc. Control returns to the ANSYS program after the system
procedure is completed. ANSYS may not be aware of your specific user environment. In particular, this
command may not recognize UNIX aliases, depending on the hardware platform and user environment.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1656
/SYP
T Commands
TALLOW, TEMP1, TEMP2, TEMP3, TEMP4, TEMP5, TEMP6
Defines the temperature table for safety factor calculations.
POST1: Element Table
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS
TEMP1, TEMP2, TEMP3, . . . , TEMP6
Input up to six temperatures covering the range of nodal temperatures. Temperatures must be input in
ascending order.
Notes
Defines the temperature table for safety factor calculations [SFACT, SALLOW]. Use STAT command to list
current temperature table. Repeat TALLOW command to zero table and redefine points (6 maximum).
Safety factor calculations are not supported by PowerGraphics. Both the SALLOW and TALLOW commands
must be used with the Full Model Graphics display method active.
Menu Paths
Main Menu>General Postproc>Safety Factor>Allowable Strs>Reset Temps
Main Menu>General Postproc>Safety Factor>Allowable Strs>Temp-depend
*TAXIS, ParmLoc, nAxis, Val1, Val2, Val3, Val4, Val5, Val6, Val7, Val8, Val9, Val10
Defines table index numbers.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParmLoc
Name and starting location in the table array parameter for indexing. Indexing occurs along the axis
defined with nAxis.
nAxis
Axis along which indexing occurs. Valid labels are:
1 --
Corresponds to Row. Default.
2 --
Corresponds to Column.
1657
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
3 --
Corresponds to Plane.
4 --
Corresponds to Book.
5 --
Corresponds to Shelf.
ALL --
Lists all index numbers. Valid only if Val1 = LIST.
Val1, Val2, Val3, . . . , Val10
Values of the index numbers for the axis nAxis, starting from the table array parameter location
ParmLoc. You can define up to ten values.
To list the index values specified with nAxis, issue Val1 = LIST. If Val1 = LIST, Val2 - Val10 are ig-
nored.
Notes
*TAXIS is a convenient method to define table index values. These values reside in the zero column, row,
etc. Instead of filling values in these zero location spots, use the *TAXIS command. For example,
*TAXIS,longtable(1,4,1,1),2,1.0,2.2,3.5,4.7,5.9
would fill index values 1.0, 2.2, 3.5, 4.7, and 5.9 in nAxis 2 (column location), starting at location 4.
To list index numbers, issue *TAXIS,ParmLoc, nAxis, LIST, where nAxis = 1 through 5 or ALL.
Menu Paths
This command cannot be accessed from a menu.
TB, Lab, MAT, NTEMP, NPTS, TBOPT, EOSOPT, FuncName
Activates a data table for nonlinear material properties or special element input.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Product Restrictions
Lab
Type of data table:
AHYPER
Anisotropic Hyperelasticity model (SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
SHELL208, SHELL209, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290.
See AHYPER Specifications (p. 1677) for more information.
ANEL
Anisotropic elastic matrix (SOLID5, PLANE13, SOLID98, SHELL181, PLANE182, PLANE183, SOLID185,
SOLID186, SOLID187, SOLSH190, SHELL208, SHELL209, SOLID272, SOLID273, SHELL281, SOLID285,
PIPE288, PIPE289, and ELBOW290, and explicit dynamic elements SOLID164 and SOLID168). Also
PLANE223, SOLID226, SOLID227. See ANEL Specifications (p. 1674) for more information.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1658
TB
ANISO
Anisotropic plasticity (LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, SOLID95, and PLANE183). See ANISO Specifications (p. 1668) for more in-
formation.
BB
Bergstrom-Boyce (PLANE182, PLANE183, SHELL181, SOLID185, SOLID186, SOLID187, SOLSH190,
SHELL208, SHELL209, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290).
See Bergstrom-Boyce Specifications (p. 1678)for more information.
BH
Magnetic field data (SOLID5, PLANE13, PLANE53, SOLID62, SOLID96, SOLID97, SOLID98, SOLID117,
SOLID236, SOLID237). See BH Specifications (p. 1682) for more information.
BISO
Bilinear isotropic hardening using von Mises or Hill plasticity.
For von Mises plasticity: LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186,
SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF264, REINF265,
SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290, and explicit dynamic
elements PLANE162, SHELL163, SOLID164, and SOLID168.
For Hill plasticity: PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182,
PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209,
REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290.
See BISO Specifications (p. 1667) for more information.
BKIN
Bilinear kinematic hardening using von Mises or Hill plasticity
For von Mises plasticity: LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, SOLID185, SOLID186, SOLID187, BEAM188,
BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SHELL281, PIPE288, PIPE289, EL-
BOW290, PLANE182, and PLANE183, and explicit dynamic elements LINK160, BEAM161, PLANE162,
SHELL163, SOLID164, and SOLID168.
For Hill plasticity: PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182,
PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209,
REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290.
See BKIN Specifications (p. 1665) for more information.
CAST
Cast iron material model (PLANE182 (not applicable for plane stress), PLANE183 (not applicable for
plane stress), SOLID185, SOLID186, SOLID187, SOLSH190, SOLID272, SOLID273, SOLID285, PIPE288,
and PIPE289). See CAST Specifications (p. 1669) for more information.
CHABOCHE
Chaboche nonlinear kinematic hardening using von Mises or Hill plasticity (PLANE42, SOLID45,
PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273,
SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290). See CHABOCHE Specifications (p. 1666) for
more information.
1659
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
CDM
Mullins effect for Isotropic Hyperelasticity models (SHELL181, PLANE182, PLANE183, SOLID185, SOL-
ID186, SOLID187, SOLSH190, SHELL208, SHELL209, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288,
PIPE289, and ELBOW290). See Mullins Effect Specifications (p. 1681) for more information.
CNDE
Anisotropic electric current conductivity ( HF119, HF120). See CNDE Specifications (p. 1682) for more
information.
CNDM
Anisotropic magnetic current conductivity ( HF119, HF120). See CNDM Specifications (p. 1683) for more
information.
COMP
Composite material models (explicit dynamic elements PLANE162, SHELL163, SOLID164, and SOLID168).
See COMP Specifications (p. 1686) for more information.
CONCR
Concrete element data (SOLID65) or concrete damage model (explicit dynamic elements SOLID164
and SOLID168). See CONCR Specifications (p. 1669) for more information.
CREEP
Viscoplasticity/creep. ANSYS can model pure creep, creep with isotropic hardening plasticity, or creep
with kinematic hardening plasticity using both von Mises or Hill potentials. See Material Model
Combinations in the Element Reference for further information on combining models.
For explicit creep with von Mises potential: LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45,
PIPE60, SOLID62, SOLID65, PLANE82, SOLID92, and SOLID95.
For implicit creep with von Mises or Hill potential: PLANE42, SOLID45, PLANE82, SOLID92, SOLID95,
LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189,
SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273, SOLID285, SHELL281,
PIPE288, PIPE289, and ELBOW290.
See CREEP Specifications (p. 1676) for more information.
CZM
Cohesive zone material model (contact elements CONTA171, CONTA172, CONTA173, CONTA174,
CONTA175, CONTA176, CONTA177, and interface elements INTER202, INTER203, INTER204, INTER205
) . See CZM Specifications (p. 1689) for more information.
DISCRETE
Explicit spring-damper (discrete) material models (COMBI165). See DISCRETE Specifications (p. 1686) for
more information.
DP
Drucker-Prager plasticity (LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, PLANE83, SOLID92, and SOLID95). See DP Specifications (p. 1670) for more inform-
ation.
DPER
Anisotropic electric permittivity ( HF118, HF119, HF120, PLANE223, SOLID226, SOLID227). See DPER
Specifications (p. 1683) for more information.
EDP
Extended Drucker-Prager model for granular materials such as rock, concrete, soil, ceramics and
other pressure dependent materials. (PLANE182(not applicable for plane stress), PLANE183 (not ap-
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1660
TB
plicable for plane stress), SOLID185, SOLID186, SOLID187, SOLSH190, SOLID272, SOLID273, SOLID285,
PIPE288, and PIPE289). See EDP Specifications (p. 1670) for more information.
ELASTIC
Elastic material property (LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, CPT212, CPT213, CPT215, CPT216, CPT217,
PLANE223, SOLID226, SOLID227, REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285,
and ELBOW290). Elastic properties can be defined as frequency or temperature dependent properties
for use in full harmonic analyses (use TBFIELD). See ELASTIC Specifications (p. 1675) for more information.
EOS
Equation of state (explicit dynamic elements only). See EOS Specifications (p. 1687) for more information.
EVISC
Viscoelastic element data (via explicit dynamic elements BEAM161, PLANE162, SOLID164, and SOL-
ID168). See EVISC Specifications (p. 1687) for more information.
FCON
Fluid conductance data (FLUID116). See FCON Specifications (p. 1687) for more information.
FOAM
Foam material models (explicit dynamic elements PLANE162, SOLID164, and SOLID168). See FOAM
Specifications (p. 1688) for more information.
FRIC
Coefficient of friction based on Coulomb's Law (CONTAC12, CONTAC52, CONTA171, CONTA172,
CONTA173, CONTA174, CONTA175, CONTA176, CONTA177, and CONTA178) or user defined friction
properties (CONTA171, CONTA172, CONTA173, CONTA174, CONTA175, CONTA176, CONTA177, and
CONTA178). See FRIC Specifications (p. 1690) for more information.
GASKET
Gasket material model (INTER192, INTER193, INTER194, and INTER195). See GASKET Specifications (p. 1691)
for more information.
GCAP
Geological Cap material model (explicit dynamic elements SOLID164 and SOLID168). See GCAP Spe-
cifications (p. 1688) for more information.
GURSON
Gurson pressure dependent plasticity model for porous metals (PLANE182 (not applicable for plane
stress), PLANE183 (not applicable for plane stress), SOLID185, SOLID186, SOLID187, SOLID272, SOL-
ID273, and SOLID285). See GURSON Specifications (p. 1671) for more information.
HFFDLD
Frequency-dependent lossy dielectric (HF119 and HF120). See HFFDLD Specifications (p. 1683) for more
information.
HFLM
Film coefficient data (FLUID116). See HFLM Specifications (p. 1683) for more information.
HILL
Hill anisotropy when combined with other material options simulates plasticity, viscoplasticity, and
creep -- all with the Hill potential. See Material Model Combinations in the Element Reference for
further information on combining models (PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180,
SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190,
SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288,
PIPE289, and ELBOW290). See HILL Specifications (p. 1669) for more information.
1661
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
HONEY
Honeycomb material models (explicit dynamic elements PLANE162, SOLID164, and SOLID168). See
HONEY Specifications (p. 1688) for more information.
HYPER
Hyperelasticity models [Mooney-Rivlin, Ogden, Neo-Hookean, Polynomial form, Arruda-Boyce, Gent,
Yeoh, Blatz-Ko, Ogden foam, and user-defined] (SHELL181, PIPE288, PIPE289, ELBOW290, PLANE182,
PLANE183, SOLID185, SOLID186, SOLID187, SOLSH190, SHELL208, SHELL209, SOLID272, SOLID273,
SHELL281, and SOLID285). See HYPER Specifications (p. 1678) for more information.
JOIN
Linear and nonlinear elastic stiffness, linear and nonlinear damping, and frictional behavior (MPC184).
See JOIN Specifications (p. 1671) for more information.
KINH
Multilinear kinematic hardening using von Mises or Hill plasticity.
For von Mises plasticity: LINK1, LINK8, PLANE13, PIPE20, BEAM23, PLANE42, SOLID45, PIPE60, SOLID62,
PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SHELL281, PIPE288,
PIPE289, and ELBOW290.
For Hill plasticity: PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182,
PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209,
REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, and ELBOW290. KINH is the same
as MKIN with TBOPT = 2, but with fewer restrictions on the number of points per curve and the
number of temperatures. See KINH Specifications (p. 1666) for more information.
LSEM
Anisotropic electric and magnetic loss tangents (HF119 and HF120). See LSEM Specifications (p. 1684)
for more information.
MELAS
Multilinear elasticity (LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, and SOLID95 ). See MELAS Specifications (p. 1675) for more information.
MISO
Multilinear isotropic hardening using von Mises or Hill plasticity.
For von Mises plasticity: LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, SOLID95, LINK180, SHELL181,PLANE182, PLANE183, SOLID185, SOLID186,
SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272,
SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290.
For Hill plasticity: PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182,
PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209,
REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, and ELBOW290.
See MISO Specifications (p. 1667) for more information.
MKIN
Multilinear kinematic hardening using von Mises or Hill plasticity.
For von Mises plasticity: LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186,
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1662
TB
SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272,
SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290.
For Hill plasticity: PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180, SHELL181, PLANE182,
PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209,
REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, and ELBOW290.
See MKIN Specifications (p. 1667) for more information.
MOONEY
Mooney-Rivlin hyperelastic element data (explicit dynamic elements PLANE162, SHELL163, SOLID164,
and SOLID168). See MOONEY Specifications (p. 1680) for more information.
MUR
Anisotropic relative permeability ( HF118, HF119, HF120). See MUR Specifications (p. 1684) for more
information.
NLISO
Voce isotropic hardening law or power law for modeling nonlinear isotropic hardening using von
Mises or Hill plasticity (LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273,
SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290). See NLISO Specifications (p. 1668) for more
information.
PIEZ
Piezoelectric matrix (SOLID5, PLANE13, SOLID98). Also PLANE223, SOLID226, SOLID227. See PIEZ
Specifications (p. 1684) for more information.
PLASTIC
Nonlinear plasticity with stress-vs.-plastic strain data (LINK180, SHELL181, PIPE288, PIPE289, ELBOW290,
PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208,
SHELL209, REINF264, REINF265, SOLID272, SOLID273, SHELL281, and SOLID285). See PLASTIC Specific-
ations (p. 1668) for more information.
PLAW
Plasticity laws (explicit dynamic elements LINK160, BEAM161, PLANE162, SHELL163, SOLID164, and
SOLID168). See PLAW Specifications (p. 1689) for more information.
PM
Porous media material model (coupled elements CPT212, CPT213, CPT215, CPT216, and CPT217).
See PM Specifications (p. 1685) for more information.
PRONY
Prony series constants for viscoelastic materials (LINK180, SHELL181, PIPE288, PIPE289, ELBOW290,
PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208,
SHELL209, REINF264, REINF265, SOLID272, SOLID273, SHELL281, and SOLID285). See PRONY Specific-
ations (p. 1681) for more information.
PZRS
Piezoresistive materials (PLANE223, SOLID226, SOLID227). See PZRS Specifications (p. 1685) for more
information.
RATE
Rate-dependent plasticity (viscoplasticity) when combined with the BISO, MISO, NLISO or PLASTIC
material options, or rate-dependent anisotropic plasticity (anisotropic viscoplasticity) when combined
with the HILL and BISO, MISO, NLISO or PLASTIC material options (PLANE42, SOLID45, PLANE82,
SOLID92, SOLID95, LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
1663
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273,
SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290). For more information about combining
material models, see Material Model Combinations in the Element Reference.
The Chaboche option includes an explicit function for directly defining static yield stresses of mater-
ials.
An Anand unified plasticity model is also available, requiring no combination with other material
models and supporting the following elements: PLANE182 and PLANE183 (neither applicable for
plane stress), SOLID185, SOLID186, SOLID187, SOLSH190, SOLID272, SOLID273, SOLID285, PIPE288,
and PIPE289.
See RATE Specifications (p. 1677) for more information.
SDAMP
Material structural damping coefficient (PLANE182, PLANE183, SOLID185, SOLID186, SOLID187,
PLANE223, SOLID226, SOLID227, SOLID272, SOLID273, and SOLID285). Damping coefficients can be
defined as frequency or temperature dependent properties for use in full harmonic analyses (use
TBFIELD). SDAMP specifies damping in terms of the loss factor, which is equal to 2x the damping
ratio. See SDAMP Specifications (p. 1675) for more information.
SHIFT
Shift function for viscoelastic materials (LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186,
SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, PIPE288, PIPE289, ELBOW290, RE-
INF264, REINF265, SOLID272, SOLID273, SHELL281, and SOLID285). See SHIFT Specifications (p. 1681)
for more information.
SMA
Shape memory alloy for simulation of hysteresis superelastic behavior with no performance deform-
ation. Plane stress is not supported. See Shape Memory Alloys in the Element Reference, and Shape
Memory Alloy in the Theory Reference for the Mechanical APDL and Mechanical Applications for further
information. The TB,SMA option is valid for PLANE182 and PLANE183 (with plane strain or axisym-
metric stress states), and for SOLID185, SOLID186, SOLID187, SOLSH190, SOLID272, SOLID273, and
SOLID285. See SMA Specifications (p. 1673) for more information.
STATE
User-defined state variables, used with TB,USER or TB,CREEP [when the USER CREEP option is chosen
(TBOPT = 100)]. When used with TB,USER, TB,STATE is only used with the USERMAT subroutine (not
USERPL). When used with TB,CREEP, TB,STATE is only used with the USERCREEP subroutine. See
STATE Specifications (p. 1691) for more information.
SWELL
Swelling constants (LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62,
SOLID65, PLANE82, SOLID92, and SOLID95). See SWELL Specifications (p. 1677) for more information.
UNIAXIAL
Uniaxial stress-strain relation associated with the Cast Iron material model (PLANE182 (not applicable
for plane stress), PLANE183 (not applicable for plane stress), SOLID185, SOLID186, SOLID187, SOLSH190,
SOLID272, SOLID273, and SOLID285). See UNIAXIAL Specifications (p. 1674) for more information.
USER
User-defined material model; general purpose except for incompressible material models (LINK180,
SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190,
SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288,
PIPE289, and ELBOW290).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1664
TB
Also, user-defined plasticity or viscoplasticity (LINK1, LINK8, PIPE20, BEAM23, BEAM24, PLANE42,
SOLID45, PIPE60, SOLID62, SOLID65, PLANE82, SOLID92, SOLID95, and PLANE183).
See USER Specifications (p. 1692) for more information.
WATER
Water motion table data (PIPE59). See WATER Specifications (p. 1691) for more information.
MAT
Material reference number (defaults to 1; maximum equals 100,000).
NTEMP
The number of temperatures for which data will be provided (if applicable). Specify temperatures via
the TBTEMP command.
NPTS
For most labels where NPTS is defined, the number of data points to be specified for a given temperature.
Define data points via the TBDATA or TBPT commands.
EOSOPT
Indicates which equation of state model will be used. Used only for explicit dynamics, and only when
Lab = EOS.
1
Linear polynomial equation of state
2
Gruneisen equation of state
3
Tabulated equation of state
FuncName
The name of the function to be used (entered as %tabname%, where tabname is the name of the
table created by the Function Tool). Valid only when Lab = JOIN (joint element material) and nonlinear
stiffness or damping are specified on the TBOPT field (see JOIN Specifications (p. 1671)). The function must
be previously defined using the Function Tool. To learn more about how to create a function, see "Using
the Function Tool" in the Basic Analysis Guide.
---
Data Table Specifications
Following is a listing of all valid TB command labels (Lab values). For each material type, the data table in-
cludes requirements for the NTEMP, NPTS, and TBOPT options, along with links to more detailed document-
ation if needed.
Kinematic Hardening Tables
BKIN Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 6, Maximum = 6
NPTS:
Not used.
1665
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
TBOPT:
Stress-strain options (not used in an explicit dynamics analysis).
0 --
No stress relaxation with temperature increase (this is not recommended for nonisothermal problems).
1 --
Rice's hardening rule, which takes into account stress relaxation with increasing temperature (default).
References:
Bilinear Kinematic Hardening Constants (TB,BKIN) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
Nonlinear Inelastic Models in the ANSYS LS-DYNA User's Guide.
CHABOCHE Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum value of NTEMP is such
that NTEMP x (1 + 2NPTS) = 1000
NPTS:
Number of kinematic models to be superposed. Default = 1, Maximum value of NPTS is such that NTEMP
x (1 + 2NPTS) = 1000
TBOPT:
Not used.
References:
Nonlinear Kinematic Hardening Constants (TB,CHABOCHE) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
KINH Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 40
NPTS:
Number of data points to be specified for a given temperature. Default = 20, Maximum = 20
TBOPT:
Use 0 or leave blank to define stress -vs- total strain curve.
Use 4 or enter PLASTIC to define stress -vs- plastic strain curve. This option supports only elements
LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186, SOLID187, BEAM188, BEAM189, SOLSH190,
SHELL208, SHELL209, REINF264, REINF265, SOLID272, SOLID273, SOLID285, SHELL281, PIPE288, PIPE289,
and ELBOW290, .
References:
Multilinear Kinematic Hardening Constants (TB,KINH or TB,MKIN) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1666
TB
MKIN Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 5, Maximum = 5
NPTS:
Not used.
TBOPT:
Stress-strain options.
0 --
No stress relaxation with temperature increase (this is not recommended for nonisothermal problems);
also produces thermal ratcheting (default).
1 --
Recalculate total plastic strain using new weight factors of the subvolume.
2 --
Scale layer plastic strains to keep total plastic strain constant; agrees with Rice's model (TB, BKIN
with TBOPT = 1). Produces stable stress-strain cycles.
References:
Multilinear Kinematic Hardening Constants (TB,KINH or TB,MKIN) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
Isotropic Hardening Tables
BISO Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 6, Maximum = 6
NPTS:
Not used.
TBOPT:
Not used.
References:
Bilinear Isotropic Hardening Constants (TB,BISO) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
Nonlinear Inelastic Models in the ANSYS LS-DYNA User's Guide.
MISO Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 20
NPTS:
Number of data points to be specified for a given temperature. Default = 20, Maximum = 100
TBOPT:
Not used.
1667
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
References:
Multilinear Isotropic Hardening Constants (TB,MISO) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
NLISO Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 20
NPTS:
Number of data points to be specified for a given temperature. Default = 4, Maximum = 4
TBOPT:
Isotropic hardening options.
VOCE --
Voce hardening law. (default).
POWER --
Power hardening law..
References:
Nonlinear Isotropic Hardening Constants (TB,NLISO) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
PLASTIC Specifications
NTEMP:
The number of temperature points (default = 1). You can specify up to 20.
NPTS:
The number of stress versus plastic strain data points (default = 20). You can specify up to 100.
TBOPT:
Type of plastic hardening:
MISO --
Multilinear isotropic hardening plasticity.
KINH --
Multilinear kinematic hardening plasticity.
References:
Plasticity in the Structural Analysis Guide.
Multilinear Kinematic Hardening in the Element Reference.
Multilinear Isotropic Hardening in the Element Reference.
Anisotropic Plasticity Tables
ANISO Specifications
NTEMP:
Not used.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1668
TB
NPTS:
Not used.
TBOPT:
Not used.
References:
Anisotropic Constants (TB,ANISO) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
HILL Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 40
NPTS:
Not used.
TBOPT:
Not used.
References:
Hill's Anisotropy Constants (TB,HILL) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
Nonmetal and Other Plasticity Tables
CAST Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1; Max = 10.
NPTS:
Not used.
TBOPT:
Defines hardening type.
ISOTROPIC --
Specifies cast iron plasticity with isotropic hardening.
References:
Cast Iron Plasticity Material Constants (TB,CAST) in the Element Reference.
CONCR Specifications
NTEMP:
Number of temperatures for which data will be provided (used only if TBOPT = 0 or 1). Default = 6,
Maximum = 6
NPTS:
Not used.
TBOPT:
Concrete material options.
1669
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
0 or 1 --
General ANSYS concrete option for element SOLID65.
2 --
Concrete damage model for explicit dynamic elements SOLID164 and SOLID168.
References:
SOLID65 in the Element Reference.
Concrete Damage Model in the ANSYS LS-DYNA User's Guide.
DP Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
Drucker-Prager Constants (TB,DP) in the Element Reference.
Plastic Material Options in the Structural Analysis Guide.
EDP Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1.
NPTS:
Number of data points to be specified for a given temperature.
TBOPT:
EDP material options.
LYFUN --
LInear yield function.
PYFUN --
Power law yield function.
HYFUN --
Hyperbolic yield function.
LFPOT --
Linear flow potential function.
PFPOT --
Power law flow potential function.
HFPOT --
Hyperbolic flow potential function.
CYFUN --
CAP yield function.
CFPOT --
CAP flow potential function.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1670
TB
References:
See Extended Drucker Prager in the Element Reference.
Also see Plastic Material Options in the Structural Analysis Guide.
GURSON Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1.
NPTS:
Number of data points to be specified for a given temperature.
TBOPT:
GURSON material options.
BASE --
Basic model without nucleation or coalescence. (default)
SNNU --
Strain controlled nucleation.
SSNU --
Stress controlled nucleation.
COAL --
Coalescence
References:
See Gurson's Model in the Element Reference.
Also see Plastic Material Options in the Structural Analysis Guide.
JOIN Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1.
NPTS:
Number of data points to be specified for a given temperature. NPTS is ignored if TBOPT = STIF or
DAMP.
If Coulomb friction is specified, NPTS is used only for TBOPT = MUS1, MUS4, and MUS6.
TBOPT:
Joint element material options.
Linear stiffness behavior:
STIF --
Linear stiffness.
Nonlinear stiffness behavior:
JNSA --
Nonlinear stiffness behavior in all available components of relative motion for the joint element.
JNS1 --
Nonlinear stiffness behavior in local UX direction only.
1671
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
JNS2 --
Nonlinear stiffness behavior in local UY direction only.
JNS3 --
Nonlinear stiffness behavior in local UZ direction only.
JNS4 --
Nonlinear stiffness behavior in local ROTX direction only.
JNS5 --
Nonlinear stiffness behavior in local ROTY direction only.
JNS6 --
Nonlinear stiffness behavior in local ROTZ direction only.
Linear damping behavior:
DAMP --
Linear damping.
Nonlinear damping behavior:
JNDA --
Nonlinear damping behavior in all available components of relative motion for the joint element.
JND1 --
Nonlinear damping behavior in local UX direction only.
JND2 --
Nonlinear damping behavior in local UY direction only.
JND3 --
Nonlinear damping behavior in local UZ direction only.
JND4 --
Nonlinear damping behavior in local ROTX direction only.
JND5 --
Nonlinear damping behavior in local ROTY direction only.
JND6 --
Nonlinear damping behavior in local ROTZ direction only.
Friction Behavior:
Coulomb friction coefficient -
The values can be specified using either TBDATA (NPTS = 0) or TBPT (NPTS is nonzero).
MUS1 --
Coulomb friction coefficient (stiction) in local UX direction only.
MUS4 --
Coulomb friction coefficient (stiction) in local ROTX direction only.
MUS6 --
Coulomb friction coefficient (stiction) in local ROTZ direction only.
Coulomb friction coefficient - Exponential Law -
Use TBDATA to specify
s
,
d
, and c for the exponential law.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1672
TB
EXP1 --
Exponential law for friction in local UX direction only.
EXP4 --
Exponential law for friction in local ROTX direction only.
EXP6 --
Exponential law for friction in local ROTZ direction only.
Elastic slip:
SL1 --
Elastic slip in local UX direction only.
SL4 --
Elastic slip in local ROTX direction only.
SL6 --
Elastic slip in local ROTZ direction only.
TMX1 --
Critical force in local UX direction only.
TMX4 --
Critical moment in local ROTX direction only.
TMX6 --
Critical moment in local ROTZ direction only.
Stick-stiffness:
SK1 --
Stick-stiffness in local UX direction only.
SK4 --
Stick-stiffness in local ROTX direction only.
SK6 --
Stick-stiffness in local ROTZ direction only.
Interference fit force/moment:
FI1 --
Interference fit force in local UX direction only.
FI4 --
Interference fit moment in local ROTX direction only.
FI6 --
Interference fit moment in local ROTZ direction only.
References:
MPC184 Joint Material Models (TB,JOIN) in the Element Reference.
SMA Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, maximum = 40.
1673
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
NPTS:
Not used.
TBOPT:
Not used
References:
Shape Memory Alloy Material Model (TB,SMA) in the Element Reference.
Shape Memory Alloys in the Structural Analysis Guide, and Shape Memory Alloy in the Theory Reference
for the Mechanical APDL and Mechanical Applications.
UNIAXIAL Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1; Max = 10.
NPTS:
Number of data points to be specified for a given temperature. Default = 20; Max = 20.
TBOPT:
Defines stress-strain relationship for cast iron plasticity.
TENSION --
Defines stress-strain relation in tension
COMPRESSION --
Defines stress-strain relation in compression.
References:
Cast Iron Plasticity Material Constants (TB,CAST) in the Element Reference.
Elasticity Tables
ANEL Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 6, Maximum = 6. NTEMP is not used
for explicit dynamic elements.
NPTS:
Not used.
TBOPT:
Anisotropic elastic matrix options.
0 --
Elasticity matrix used as supplied (input in stiffness form).
1 --
Elasticity matrix inverted before use (input in flexibility form). This option is not valid for explicit
dynamic elements.
References:
Anisotropic Elastic Material Model (TB,ANEL) in the Element Reference.
Linear Elastic Models in the ANSYS LS-DYNA User's Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1674
TB
ELASTIC Specifications
NTEMP:
Not used.
NPTS:
Number of properties to be defined for the material option.
2 --
Isotropic property (EX, NUXY). This is the default.
9 --
Orthotropic properties with minor Poisson's ratio (EX, EY, EZ, GXY, GYZ, GXZ, NUXY, NUYZ, NUXZ).
NOTE: All 9 constants must be defined for orthotropic material, even for the 2-D case.
21 --
Anisotropic property with stiffness form of input (D11, D21, D31, D41, D51, D61, D22, D32, D42, D52,
D62, D33, D43, ..... D66).
TBOPT:
Elasticity options:
USER --
User-defined linear elastic properties. For more information, see the documentation for the
user_tbelastic subroutine in the Guide to ANSYS User Programmable Features, available on the
ANSYS product distribution media.
References:
See the TBFIELD command for more information on defining temperature and/or frequency dependent
properties.
Full Harmonic Response Analysis in the Structural Analysis Guide.
MELAS Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 20
NPTS:
Number of data points to be specified for a given temperature. Default = 20, Maximum = 100
TBOPT:
Not used.
References:
Multilinear Elastic Constants (TB,MELAS) in the Element Reference.
Multilinear Elasticity in the Structural Analysis Guide.
SDAMP Specifications
NTEMP:
Not used.
NPTS:
Number of properties to be defined for the material option.
1675
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
1 --
Structural damping coefficient. This is the default.
TBOPT:
Not Used.
References:
See the TBFIELD command for more information on defining temperature and/or frequency dependent
properties.
Full Harmonic Response Analysis in the Structural Analysis Guide.
Rate-Dependent Plasticity Tables
CREEP Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum value of NTEMP is such
that NTEMP x NPTS = 1000 for implicit creep and 250 for explicit creep.
NPTS:
Number of data points to be specified for a given temperature. Default = 12 for implicit creep and 72
for explicit creep, Maximum value of NPTS is such that NTEMP x NPTS = 1000 for implicit creep and
250 for explicit creep.
TBOPT:
Creep model options.
0 --
(or Blank) Explicit creep option. Creep model is defined by constants C
6
, C
12
, and C
66
, through TBDATA.
See Primary Explicit Creep Equation for C6 = 0 through Irradiation Induced Explicit Creep Equation
for C66 = 5 in the Element Reference for the associated equations. (Applicable to LINK1, LINK8, PIPE20,
BEAM23, BEAM24, PLANE42, SOLID45, PIPE60, SOLID62, SOLID65, PLANE82, SOLID92, and SOLID95.
C
6
= 100 defines the USER CREEP option for explicit creep. You must define the creep law using the
subroutine USERCR.F. See the Guide to ANSYS User Programmable Features for more information.
1 through 13 --
Implicit creep option. See Table 2.4: "Implicit Creep Equations" in the Element Reference for a list of
available equations. Use TBTEMP and TBDATA to define temperature-dependent constants. (Applic-
able to PLANE42, SOLID45, PLANE82, SOLID92, SOLID95, LINK180 , SHELL181, PLANE182, PLANE183,
SOLID185, SOLID186 , SOLID187 , BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264,
REINF265, SOLID272, SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290).
100 --
USER CREEP option (applicable only to LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186,
SOLID187, BEAM188, BEAM189, SOLSH190, SHELL208, SHELL209, REINF264, REINF265, SOLID272,
SOLID273, SHELL281, SOLID285, PIPE288, PIPE289, and ELBOW290). You must define the creep law
using the subroutine USERCREEP.F. See the Guide to ANSYS User Programmable Features for more
information. Use TBTEMP and TBDATA to define temperature-dependent constants. For implicit
creep, use with TB,STATE for defining the number of state variables.
References:
Creep Equations (TB,CREEP) in the Element Reference.
Creep in the Structural Analysis Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1676
TB
RATE Specifications
NTEMP:
The number of temperatures for which data will be provided. The default is 1. The maximum NTEMP
value is such that NTEMP x NPTS = 1000.
NPTS:
The number of data points to be specified for a given temperature. The default is 2. The maximum NPTS
value is such that NTEMP x NPTS = 1000.
TBOPT:
Rate-dependent viscoplasticity options.
PERZYNA --
Perzyna option (default)
PEIRCE --
Peirce option
CHABOCHE --
Chaboche option
ANAND --
Anand option
References:
Rate-Dependent Plastic (Viscoplastic) Material Models (TB,RATE) in the Element Reference.
Viscoplasticity in the Structural Analysis Guide.
Rate-Dependent Plasticity in the Theory Reference for the Mechanical APDL and Mechanical Applications.
SWELL Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
Swelling Equation Constants (TB,SWELL) in the Element Reference.
Swelling in the Structural Analysis Guide.
Hyperelasticity Tables
AHYPER Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1.
NPTS:
Number of data points to be specified for a given temperature.
1677
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
TBOPT:
Anisotropic hyperelastic material options.
POLY --
Anisotropic potential.
AVEC --
Define the A vector.
BVEC --
Define the B vector.
PVOL --
Volumetric potential.
References:
Anisotropic Hyperelasticity in the Element Reference.
Anisotropic Hyperelasticity in the Theory Reference for the Mechanical APDL and Mechanical Applications
Bergstrom-Boyce Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1. Maximum must be a value such
that (NTEMP x NPTS) <= 1000.
NPTS:
Number of material constants. If TBOPT = ISO, then NPTS = 7. If TBOPT = 1, then NPTS = 1.
TBOPT:
Isochoric or volumetric strain-energy function:
ISO --
Define material constants for isochoric strain energy.
PVOL --
Define material constants for volumetric strain energy.
References:
Bergstrom-Boyce in the Theory Reference for the Mechanical APDL and Mechanical Applications.
Bergstrom-Boyce Material Constants (TB,BB) in the Element Reference.
Bergstrom-Boyce Hyperviscoelastic Material Model in the Structural Analysis Guide.
HYPER Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum value of NTEMP is such
that NTEMP x NPTS = 1000
NPTS:
Number of data points to be specified for a given temperature, except for TBOPT = MOONEY, where
NPTS is the number of parameters in the Mooney-Rivlin model (2 [default], 3, 5, or 9).
TBOPT:
Hyperelastic material options.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1678
TB
MOONEY --
Mooney-Rivlin model (default). You can choose a 2 parameter Mooney-Rivlin model with NPTS = 2
(default), or a 3, 5, or 9 parameter model by setting NPTS equal to one of these values.
References:
Mooney-Rivlin Hyperelastic Material Constants (TB,HYPER,,,,MOONEY) in the Element Reference.
Mooney-Rivlin Hyperelastic Option in the Structural Analysis Guide.
OGDEN --
Ogden model. For NPTS, the default = 1 and the maximum is such that NTEMP x NPTS x 3 = 1000.
References:
Ogden Hyperelastic Material Constants (TB,HYPER,,,,OGDEN) in the Element Reference.
Ogden Hyperelastic Option in the Structural Analysis Guide.
NEO --
Neo-Hookean model. For NPTS, the default = 2 and the maximum = 2.
References:
Neo-Hookean Hyperelastic Material Constants (TB,HYPER,,,,NEO) in the Element Reference.
Neo-Hookean Hyperelastic Option in the Structural Analysis Guide.
POLY --
Polynomial form model. For NPTS, the default = 1 and the maximum is such that NTEMP x NPTS =
1000.
References:
Polynomial Form Hyperelastic Material Constants (TB,HYPER,,,,POLY) in the Element Reference.
Polynomial Form Hyperelastic Option in the Structural Analysis Guide.
BOYCE--
Arruda-Boyce model. For NPTS, the default = 3 and the maximum = 3.
References:
Arruda-Boyce Hyperelastic Material Constants (TB,HYPER,,,,BOYCE) in the Element Reference.
Arruda-Boyce Hyperelastic Option in the Structural Analysis Guide.
GENT --
Gent model. For NPTS, the default = 3 and the maximum = 3.
References:
Gent Hyperelastic Material Constants (TB,HYPER,,,,GENT) in the Element Reference.
Gent Hyperelastic Option in the Structural Analysis Guide.
YEOH --
Yeoh model. For NPTS, the default = 1 and the maximum is such that NTEMP x NPTS x 2 = 1000.
1679
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
References:
Yeoh Hyperelastic Material Constants (TB,HYPER,,,,YEOH) in the Element Reference.
Yeoh Hyperelastic Option in the Structural Analysis Guide.
BLATZ --
Blatz-Ko model. For NPTS, the default = 1 and the maximum = 1.
References:
Blatz-Ko Foam Hyperelastic Material Constants (TB,HYPER,,,,BLATZ) in the Element Reference.
Blatz-Ko Hyperelastic Option in the Structural Analysis Guide.
FOAM --
Hyperfoam (Ogden) model. For NPTS, the default = 1 and the maximum is such that NTEMP x NPTS
x 3 = 1000.
References:
Ogden Compressible Foam Hyperelastic Material Constants (TB,HYPER,,,,FOAM) in the Element Reference.
Ogden Compressible Foam Hyperelastic Option in the Structural Analysis Guide.
USER --
User-defined hyperelastic model. See the ANSYS Guide to User Programmable Features for details.
References:
User-Defined Hyperelastic Material Constants (TB,HYPER,,,,USER) in the Element Reference.
User-Defined Hyperelastic Option in the Structural Analysis Guide.
MOONEY Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 6, Maximum = 6.
NPTS:
(Not used for explicit dynamic elements.)
TBOPT:
Mooney-Rivlin material option, applicable to explicit dynamic elements PLANE162, SHELL163, SOLID164,
and SOLID168.
0 --
Direct input of hyperelastic material constants (default).
1 --
Reserved for future use.
2 --
Material constants to be calculated by the LS-DYNA program from experimental data. This option is
only valid for explicit dynamic elements.
References:
Nonlinear Elastic Models in the ANSYS LS-DYNA User's Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1680
TB
Mullins Effect Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1. Maximum must be a value such
that (NTEMP x NPTS) <= 1000.
NPTS:
Number of data points to be specified for a given temperature.
TBOPT:
Mullins effect option:
PSE2 --
Pseudo-elastic model with modified Ogden-Roxburgh damage function. Requires NPTS = 3.
References:
Mullins Effect in the Theory Reference for the Mechanical APDL and Mechanical Applications.
Mullins Effect Constants (TB,CDM) in the Element Reference.
Mullins Effect Material Model in the Structural Analysis Guide.
Viscoelasticity Tables
PRONY Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1; Max = 100.
NPTS:
Number of pairs of Prony series. Default = 1 pair; Max = 100 pairs.
The total number of data points allowed is 1000.
TBOPT:
Defines the relaxation behavior for viscoelasticity.
1 --
(or SHEAR) relaxation behavior of the shear response.
2 --
(or BULK) relaxation behavior of the volumetric response.
References:
Viscoelastic Material Model in the Element Reference.
SHIFT Specifications
NTEMP:
Allows one temperature for which data will be provided.
NPTS:
Number of material constants to be entered as determined by the shift function specified by TBOPT.
3 --
for TBOPT = 1 or WLF
1681
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
2 --
for TBOPT = 2 or TN
n
f
--
for TBOPT = 3 or FICT, where n
f
is the number of partial fictive temperatures
TBOPT:
Defines the shift function
1 or WLF --
William-Landel-Ferry shift function
2 or TN --
Tool-Narayanaswamy shift function
3 or FICT --
Tool-Narayanaswamy with fictive temperature shift function
100 --
(or USER) User-defined shift function.
References:
Viscoelastic Material Model in the Element Reference.
Multiphysics Tables
BH Specifications
NTEMP:
Not used.
NPTS:
Number of data points to be specified. Default = 20, Maximum = 500
TBOPT:
Not used.
References:
Magnetic Materials in the Element Reference.
Additional Guidelines for Defining Regional Material Properties and Real Constants in the Low-Frequency
Electromagnetic Analysis Guide.
CNDE Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
High-Frequency Electromagnetic Material Models in the Element Reference.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1682
TB
CNDM Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
High-Frequency Electromagnetic Material Models in the Element Reference.
DPER Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used for HF118, HF119, and HF120.
Permittivity matrix options for PLANE223, SOLID226, and SOLID227:
0 --
Permittivity matrix at constant strain [
S
] (used as supplied)
1 --
Permittivity matrix at constant stress [
T
] (converted to [
S
] form before use)
References:
Anisotropic Electric Permittivity Material Model (TB,DPER) in the Element Reference
HFFDLD Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
High-Frequency Electromagnetic Materials in the Element Reference.
Specifying Material Properties in the High-Frequency Electromagnetic Analysis Guide.
HFLM Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 20
NPTS:
Number of data points to be specified for a given temperature. Default = 1, Maximum = 100
1683
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
TBOPT:
Not used.
References:
FLUID116 in the Element Reference.
LSEM Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
High-Frequency Electromagnetic Materials in the Element Reference.
Specifying Material Properties in the High-Frequency Electromagnetic Analysis Guide.
MUR Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Anisotropic relative permeability options:
0 --
Input permeability matrix.
1 --
Generate permeability matrix for B-H nonlinear material with uniform dc internal magnetic field
2 --
Generate permeability matrix for B-H nonlinear material with non-uniform dc internal magnetic field
References:
High-Frequency Electromagnetic Material Models in the Element Reference.
PIEZ Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Piezoelectric matrix options.
0 --
Piezoelectric stress matrix [e] (used as supplied)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1684
TB
1 --
Piezoelectric strain matrix [d] (converted to [e] form before use)
References:
Piezoelectric Material Model (TB,DPER) in the Element Reference.
Piezoelectric Analysis in the Coupled-Field Analysis Guide.
PZRS Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Piezoresistive matrix options
0 --
Piezoresistive stress matrix (used as supplied)
1 --
Piezoresistive strain matrix (used as supplied)
References:
Piezoresistive Material Model (TB,PZRS) in the Element Reference.
Piezoresistive Analysis in the Coupled-Field Analysis Guide.
Porous Media Tables
PM Specifications
NTEMP:
The number of temperatures. Default = 1. Maximum must be a value such that (NTEMP x NPTS) <=
1000.
NPTS:
The number of material constants. Default = 4. Maximum must be a value such that (NTEMP x NPTS)
<= 1000.
TBOPT:
Porous media options:
PERM --
Permeability
BIOT --
Biot coefficient
References:
Porous Media Constants (TB,PM) in the Element Reference.
Pore-Fluid-Diffusion-Structural Analysis in the Coupled-Field Analysis Guide.
Porous Media Flow in the Theory Reference for the Mechanical APDL and Mechanical Applications
1685
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
Also see VM260 in the Verification Manual.
ANSYS LS-DYNA Tables
COMP Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
Composite Damage Model in the ANSYS LS-DYNA User's Guide.
DISCRETE Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Explicit spring-damper (discrete) material options.
0 --
Linear elastic spring (translational or rotational elastic spring) (default)
1 --
Linear viscous damper (linear translational or rotational damper)
2 --
Elastoplastic spring (elastoplastic translational or rotational spring with isotropic hardening)
3 --
Nonlinear elastic spring (nonlinear elastic translational or rotational spring with arbitrary force/dis-
placement response moment/rotation dependency)
4 --
Nonlinear viscous damper (nonlinear damping with arbitrary force/velocity response moment/rota-
tional velocity dependency)
5 --
General nonlinear spring (general nonlinear translational or rotational spring with arbitrary loading
and unloading definitions)
6 --
Maxwell viscoelastic spring (Maxwell viscoelastic translational or rotational spring)
7 --
Inelastic tension or compression-only spring (inelastic tension or compression only, translational or
rotational spring)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1686
TB
References:
Spring-Damper (Discrete) Models in the ANSYS LS-DYNA User's Guide.
EOS Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Equation of state (explicit dynamics elements only - no default, must specify).
1 --
Johnson-Cook material model - for strain, strain rate, and temperature dependent impact/forming
analyses.
2 --
Null material model - for allowing equation of state to be considered without computing deviatoric
stresses.
3 --
Zerilli-Armstrong material model - for metal forming processes in which the stress depends on strain,
strain rate, and temperature.
4 --
Bamman material model - for metal forming processes with strain rate and temperature dependent
plasticity. Does not require an additional equation of state (EOSOPT is not used).
5 --
Steinberg material model - for modeling high strain rate effects in solid elements with failure.
References:
Equation of State Models in the ANSYS LS-DYNA User's Guide.
EVISC Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
Nonlinear Elastic Models in the ANSYS LS-DYNA User's Guide.
FCON Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum = 20
NPTS:
Number of data points to be specified for a given temperature. Default = 1, Maximum = 100
TBOPT:
Not used.
1687
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
References:
FLUID116 in the Element Reference.
FOAM Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Foam material options for explicit dynamics elements (no default - must specify).
1 --
Rigid, closed cell, low density polyurethane foam material model.
2 --
Highly compressible urethane foam material model.
3 --
Energy absorbing foam material model.
4 --
Crushable foam material model.
References:
Foam Models in the ANSYS LS-DYNA User's Guide.
GCAP Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
Pressure Dependent Plasticity Models in the ANSYS LS-DYNA User's Guide.
HONEY Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
Foam Models in the ANSYS LS-DYNA User's Guide.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1688
TB
PLAW Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Plasticity options for explicit dynamics elements (no default - must specify).
1 --
Isotropic/kinematic hardening model.
2 --
Strain rate dependent plasticity model used for metal and plastic forming analyses.
3 --
Anisotropic plasticity model (Barlat and Lian).
4 --
Strain rate dependent plasticity model used for superplastic forming analyses.
5 --
Strain rate dependent isotropic plasticity model used for metal and plastic forming analyses.
6 --
Anisotropic plasticity model (Barlat, Lege, and Brem) used for forming processes.
7 --
Fully iterative anisotropic plasticity model for explicit shell elements only.
8 --
Piecewise linear plasticity model for explicit elements only.
9 --
Elastic-plastic hydrodynamic model for explicit elements only.
10 --
Transversely anisotropic FLD (flow limit diagram) model for explicit elements only.
11 --
Modified piecewise linear plasticity model for explicit shell elements only.
12 --
Elastic viscoplastic thermal model for explicit solid and shell elements only.
References:
Nonlinear Inelastic Models in the ANSYS LS-DYNA User's Guide.
Pressure Dependent Plasticity Models in the ANSYS LS-DYNA User's Guide.
Special Material Tables
CZM Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1.
1689
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
NPTS:
Number of data points to be specified for a given temperature.
TBOPT:
Cohesive zone material options.
EXPO --
Exponential material behavior (valid for interface elements only).
CBDD --
Bilinear material behavior with linear softening characterized by maximum traction and maximum
separation (valid for contact elements only).
CBDE --
Bilinear material behavior with linear softening characterized by maximum traction and critical energy
release rate (valid for contact elements only).
References:
Interface Delamination and Failure Simulation in the Structural Analysis Guide.
Cohesive Zone Material Constants (TB,CZM) in the Element Reference.
FRIC Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1; Max = 40.
NTEMP is not used for the following situations:
Isotropic or orthotropic friction defined in terms of field data (TBFIELD command)
User-defined friction (TBOPT = USER)
NPTS:
Number of data points to be specified for user-defined friction (TBOPT = USER). Not used for TBOPT =
ISO or TBOPT = ORTHO.
TBOPT:
Friction options:
ISO --
Isotropic friction (one coefficient of friction, MU). This option is valid for all 2-D and 3-D contact ele-
ments. (Default.)
ORTHO --
Orthotropic friction (two coefficients of friction, MU1 and MU2). This option is valid for CONTA173,
CONTA174, CONTA175, CONTA176, and CONTA177.
USER --
User defined friction. This option is valid for CONTA171, CONTA172, CONTA173, CONTA174, CONTA175,
CONTA176, CONTA177, and CONTA178 elements.
References:
Contact Friction in the Element Reference.
See also the TBFIELD command for more information on defining a coefficient of friction that is
dependent on temperature, time, normal pressure, sliding distance, or sliding relative velocity.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1690
TB
GASKET Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1.
NPTS:
Number of data points to be specified for a given temperature. Default = 5 for TBOPT = PARA. Default
= 1 for all other values of TBOPT.
TBOPT:
Gasket material options.
PARA --
Gasket material general parameters.
COMP --
Gasket material compression data.
LUNL --
Gasket linear unloading data.
NUNL --
Gasket nonlinear unloading data.
TSS --
Transverse shear data.
References:
Gasket Materials in the Element Reference.
Gasket Joints Simulation in the Structural Analysis Guide.
WATER Specifications
NTEMP:
Not used.
NPTS:
Not used.
TBOPT:
Not used.
References:
PIPE59 in the Element Reference.
User Tables
STATE Specifications
NTEMP:
Not used.
NPTS:
Number of state variables. Maximum = 1000
TBOPT:
Not used.
1691
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
References:
User-Defined Material Constants (TB,USER) and Implicit Creep Equations in the Element Reference.
User Defined Material and Implicit Creep Procedure in the Structural Analysis Guide.
USER Specifications
NTEMP:
Number of temperatures for which data will be provided. Default = 1, Maximum value of NTEMP is such
that NTEMP x NPTS = 1000
NPTS:
Number of data points to be specified for a given temperature. Default = 48, Maximum value of NPTS
is such that NTEMP x NPTS = 1000
TBOPT:
Not used.
References:
User-Defined Material Constants (TB,USER) in the Element Reference.
User-Defined Material Model in the Structural Analysis Guide.
Notes
TB activates a data table to be used with subsequent TBDATA or TBPT commands. The table space is ini-
tialized to zero values. Data from this table are used for certain nonlinear material descriptions as well as
for special input for some elements. See Material Data Tables (Implicit Analysis) in the Element Reference for
a description of table types (Lab) or the elements that require the table for special data. See Material Models
in the ANSYS LS-DYNA User's Guide for a description of data table input required for explicit dynamic mater-
ials. See the MP command for linear material property input.
The type of data table specified remains active until the TB command is reissued. More than one type of
data table may be defined for each material (for example, MISO and CREEP), except that only one type of
plasticity/elasticity may be used for each material.
This command is also valid in SOLUTION.
Product Restrictions
Available Products Command Option
Lab
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS AHYPER
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS ANAN
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS ANEL
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS ANIS
MP ME ST <> <> <> <> <> EM <> <> PP <> EME MFS BH
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS BISO
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS BKIN
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1692
TB
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS BOYC
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS CAST
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS CHAB
MP ME ST <> <> <> <> <> <> EH <> PP <> EME MFS CNDE
MP ME ST <> <> <> <> <> <> EH <> PP <> EME MFS CNDM
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS COMP
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS CONC
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS CREEP
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS CZM
<> <> <> <> <> <> <> <> <> <> DY PP <> <> <> DISC
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS DP
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS DPER
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS EDP
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS ELASTIC
<> <> <> <> <> <> <> <> <> <> DY PP <> <> <> EOS
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS EVIS
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS FCON
<> <> <> <> <> <> <> <> <> <> DY PP <> <> <> FOAM
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS FRIC
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS GASKET
<> <> <> <> <> <> <> <> <> <> DY PP <> <> <> GCAP
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS HFLM
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS HILL
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS HONEY
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS HYPER
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS HYPER - MOONEY
(NPTS = 2)
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS HYPER - OGDEN
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS HYPER - NEO
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS KINH
MP ME ST PR PRN <> <> <> <> <> DY PP <> EME MFS JOIN
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS MELA
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS MISO
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS MKIN
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS MOON
MP ME ST <> <> <> <> <> <> EH <> PP <> EME MFS MUR
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS NL
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS NLIS
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS PFLO
1693
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TB
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS PIEZ
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS PLASTIC
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS PLAW
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS PROONY
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS PZRS
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS RATE
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS SDAMP
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS SHIFT
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS SMA
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS STATE
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS SWELL
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS UNIAXIAL
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS USER
MP ME ST <> <> <> <> <> <> <> DY PP <> EME MFS WATER
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
TBCOPY, Lab, MATF, MATT
Copies a data table from one material to another.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lab
Data table label (see the TB command for valid labels, and see "Notes" (p. 1694) for Lab = ALL).
MATF
Material reference number where data table is to be copied from.
MATT
Material reference number where data table is to be copied to.
Notes
The TBCOPY command, with Lab = ALL, copies all of the nonlinear data defined by the TB command. If
you copy a model that includes both yield behavior constants and linear constants (for example, a BKIN
model), TBCOPY,ALL and MPCOPY are used together to copy the entire model. All input data associated
with the model is copied, that is, all data defined through the TB and MP commands.
Also, if you copy a material model using the Material Model Interface (Edit> Copy), both the commands
TBCOPY,ALL and MPCOPY are issued, regardless of whether the model includes linear constants only, or if
it includes a combination of linear and yield behavior constants.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1694
TBCOPY
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
TBDATA, STLOC, C1, C2, C3, C4, C5, C6
Defines data for the data table.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
STLOC
Starting location in table for entering data. For example, if STLOC = 1, data input in the C1 field applies
to the first table constant, C2 applies to the second table constant, etc. If STLOC=5, data input in the
C1 field applies to the fifth table constant, etc. Defaults to the last location filled + 1. The last location
is reset to 1 with each TB or TBTEMP command.
C1, C2, C3, . . . , C6
Data values assigned to six locations starting with STLOC. If a value is already in this location, it is re-
defined. A blank value leaves the existing value unchanged.
Notes
Defines data for the table specified on the last TB command at the temperature specified on the last TBTEMP
command (if applicable). The type of data table specified in the last TB command determines the number
of data values needed in TBDATA. Data values are linearly interpolated for temperatures that fall between
user defined TBTEMP values. See Material Data Tables (Implicit Analysis) of the Element Reference for the
number of data values required for different material behavior options.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
TBDELE, Lab, MAT1, MAT2, INC
Deletes previously defined data tables.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lab
Data table label (see the TB command for valid labels). If ALL, delete all data tables.
1695
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBDELE
MAT1, MAT2, INC
Delete tables for materials MAT1 to (MAT2 defaults to MAT1) in steps of INC (defaults to 1). If MAT1=
ALL, ignore MAT2 and INC and delete data tables for all materials.
Notes
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
TBFIELD, Type, Value
Defines values of field variables for the material data tables.
PREP7: Data Tables
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS
Type
Type of field variable:
FREQ
A frequency will be specified in Value
TEMP
A temperature will be specified in Value
TIME
A time will be specified in Value
NPRES
A normal pressure will be specified in Value
SLDA
A total sliding distance (algebraic) will be specified in Value
SLDI
A total sliding distance (absolute) will be specified in Value
SLRV
A sliding velocity will be specified in Value
Value
The field value to be referenced (use this command multiple times to enter values of different field
variables).
Notes
You use this command with TB,ELASTIC, or TB,SDAMP to define multiple field values for frequency and/or
temperature dependent material data tables. You can also use this command with TB,FRIC to define friction
that is dependent on temperature, time, normal pressure, sliding distance, and sliding velocity.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1696
TBFIELD
Note
When defining friction, keep in mind the following points:
The TEMP value specified on this command corresponds to the average temperature on the
contact surface for contact elements CONTA171, CONTA172, CONTA173, CONTA174, CON-
TA175, CONTA176, and CONTA177. For contact element CONTA178, the TEMP value corres-
ponds to the average temperature of the nodes.
The TIME value specified on this command corresponds to the analysis time specified on the
TIME command.
The algebraic sliding distance (SLDA) specified on this command is the total sliding distance
(the algebraic sum) as reported in the element output definitions table for the contact ele-
ments (for example, TASS and TASR output items for CONTA174).
The absolute sliding distance (SLDI) specified on this command is the total accumulated
sliding distance (the absolute sum) as reported in the element output definitions table for
the contact elements (for example, AASS and AASR output items for CONTA174).
When used with TB,FRIC, field variables defined by TBFIELD are only available for isotropic
friction (TBOPT = ISO) and orthotropic friction (TBOPT = ORTHO); they are not available for
user-defined friction (TBOPT = USER).
Enter the appropriate TB,ELASTIC, TB,SDAMP, or TB,FRIC option first to define your data tables as field
variable-dependent. Then issue the TBFIELD command to define your field values. Data values must be
defined in ascending order for all field quantities. If a field value is to be held constant, it should be defined
only once; subsequent definitions will be ignored. There is no limit on the number of values you can specify.
The specified field value remains active until the next TBFIELD command is input. Once you define the field
value(s), use TBDATA to define your data for the data tables.
See Understanding How ANSYS Interpolates Field Variables in the Element Reference for more information
on the interpolation scheme used for field-dependent material properties.
See Full Harmonic Response Analysis in the Structural Analysis Guide for more information on using TBFIELD
with TB,ELASTIC or TB,SDAMP.
See Contact Friction in the Element Reference for more information on using TBFIELD with TB,FRIC.
Menu Paths
This command cannot be accessed from a menu.
TBFT, Oper, ID, Option1, Option2, Option3, Option4, Option5, Option6, Option7
Performs material curve fitting operations.
PREP7: Materials
MP ME ST <> PRN <> <> <> <> <> DY PP <> EME MFS
Use material curve fitting to evaluate your experimental data and correlate it to the nonlinear material
models built into ANSYS. You apply coefficients to your curve data, determine a fit with existing model, and
1697
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBFT
write the data according to the table configurations outlined in the TB command. Curve fitting is available
for nonlinear models associated with the material behaviors listed below:
For hyperelastic material behavior, see "TBFT Specifications for Hyperelastic Models" (p. 1698), below.
For creep material behavior, see "TBFT Specifications for Creep Models" (p. 1701), below.
For viscoelastic material behavior, see "TBFT Specifications for Viscoelastic Models" (p. 1703), below.
TBFT Specifications for Hyperelastic Models
TBFT, Oper, ID, Option1, Option2, Option3, Option4, Option5, Option6, Option7
Oper
The specific curve fitting operation:
FADD
Define a constitutive model.
FDEL
Delete a constitutive model.
FSET
Write data related to a constitutive model to the database (same as TB command).
SET
Initialize coefficients of a constitutive model for nonlinear curve-fitting procedure.
CDEL
Deletes coefficients at current reference temperature. Applicable only for temperature dependent
coefficients.
SOLVE
Solve for coefficients.
FIX
Fix (hold constant) the coefficient you specify in Option4.
EADD
Add experimental data.
EDEL
Delete experimental data.
LIST
List all data associated with the material model represented by the material ID number.
PLOT
Plot curves.
ID
The material reference number (same as MAT argument used in the TB command). Valid entry is any
number greater than zero (default = 1) but less than 100,000.
Option1
For curve fit function operations (Oper = FADD, FDEL, FSET, SET, CDEL, SOLVE or FIX) this field specifies
the category (HYPER).
For adding, deleting, or plotting your experiment (Oper = EADD, EDEL, or PLOT), this field specifies the
experimental data type. Valid entries are: UNIA, BIAX, SHEA, SSHE,and VOLU (default = UNIA).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1698
TBFT
Option2
For curve fit function operations (Oper = FADD, FDEL, FSET, SET, CDEL, SOLVE, or FIX), this field specifies
constitutive model type. The valid entries are listed in Table 259: Hyperelastic Options (p. 1700) below.
When you need to specify a filename from which to get experimental data (Oper = EADD), place that
string here. Valid entry is any filename string. You can enter the entire path\filename.extension
string and leave the next two fields (Option3 and Option4) blank, or you can specify the name here,
the extension in the next field, and the path following.
For plotting (Oper = PLOT), enter HYPER.
Option3
For Oper = FADD, FDEL, FSET, CDEL, SET, SOLVE or FIX, some of the cases specified inOption2 will require
that the polynomial order be specified. The applicable values for the order specification are listed in
Table 259: Hyperelastic Options (p. 1700).
If a filename for experimental data is being specified in Option2 (Oper = EADD), this field will contain
the file extension.
For plotting operations (Oper = PLOT), use this field to specify the constitutive model name (see
Table 259: Hyperelastic Options (p. 1700) ).
Option4
When you are working on a specific coefficient (Oper = FIX), this field specifies the index of that coeffi-
cient. Valid entries vary from 1 to n, where n is the total number of coefficients (default = 1).
For Oper = SET, see Table 258: Set Operations (p. 1699), below.
If a filename for experimental data is being specified in Option2 (Oper = EADD), this field will contain
the directory/path specification.
For Oper = PLOT, this field specifies the order of the polynomial if applicable, as listed in Table 259: Hy-
perelastic Options (p. 1700) .
If Oper = SOLVE, this value specifies the curve-fitting procedure. Valid entries are 0 for unnormalized
least squares curve-fitting procedure, and 1 for normalized least squares curve-fitting procedure.
Option5
When you are working on a specific coefficient (Oper = FIX), this field specifies the index of that coeffi-
cient. Valid entries vary from 1 to N, where N is the total number of coefficients (default = 1)
For Oper = SET, see Table 258: Set Operations (p. 1699), below.
Table 258 Set Operations
Option5 Option4 Purpose
Value of that coefficient Index of coefficient Set the value of the coefficient.
1 for ON and 0 for OFF tdep Turn temperature dependency
ON/OFF*
Temperature value tref Set reference temperature
* - Temperature data should be supplied in the same units as the TREF command.
1699
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBFT
If Oper = SOLVE, use this field to specify the number of iterations to be used in the calculation of the
coefficients. Valid entry is any positive integer (default = 1000)
Option6
If Oper = SOLVE, specifies the allowed tolerance in residual change to stop an iteration. Valid entry is
0.0 to 1.0 (default = 0.0).
Option7
If Oper = SOLVE, specifies the allowed tolerance in coefficient change to stop an iteration. Valid entry
is 0 to 1 (default = 0).
Table 259 Hyperelastic Options
Option3 Option2 Option1
2, 3, 5, 9 MOON HYPER
1 to N POLY HYPER
1 to N OGDEN HYPER
1 to N YEOH HYPER
NA BOYC HYPER
NA GENT HYPER
NA NEO HYPER
NA BLAT HYPER
1 to N FOAM HYPER
Following table summarizes the format for hyperelastic operations via the TBFT command:
Hyperelastic Model Command Summary
Option7 Option6 Option5 Option4 Option3 Option2 Option1 ID Oper
Order Option HYPER ID FADD
Order Option HYPER ID FDEL
Order Option HYPER ID FSET
Value Index Order Option HYPER ID SET
CTOL RTOL Num Iter
Norm
Flag
Order Option HYPER ID SOLVE
Fix / Un-
fix Flag
Index Order Option HYPER ID FIX
Direct-
ory
Exten-
sion
File Exp Type ID EADD
Exp Type
/ Index
ID EDEL
Order Option HYPER
Exp Type
/ Index
ID PLOT
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1700
TBFT
TBFT Specifications for Creep Models
TBFT, Oper, ID, Option1, Option2, Option3, . . . , Option7
Oper
The specific curve fitting operation:
FADD
Define a constitutive model.
FDEL
Delete a constitutive model.
FSET
Write data related to a constitutive model to the database (same as TB command).
SET
Initialize coefficients of a constitutive model for nonlinear curve-fitting procedure.
CDEL
Deletes coefficients at current reference temperature. Applicable only for temperature dependent
coefficients.
SOLVE
Solve for coefficients.
FIX
Fix (hold constant) the coefficient you specify in Option4.
EADD
Add experimental data.
EDEL
Delete experimental data.
LIST
List all data associated with the material model represented by the material ID number.
PLOT
Plot curves.
ID
The material reference number (same as MAT argument used in the TB command). Valid entry is any
number greater than zero (default = 1) but less than 100,000.
Option1
For curve fit function operations (Oper = FADD, FDEL, FSET, SET, CDEL, SOLVE or FIX) this field specifies
the category (CREEP).
For adding, deleting, or plotting your experiment (Oper = EADD, EDEL, or PLOT), this field specifies the
experimental data type (CREEP).
Option2
For curve fit function operations (Oper = FADD, FDEL, FSET, SET, CDEL, SOLVE, or FIX), this field specifies
constitutive model type. The valid entries are listed in Table 261: Creep Options (p. 1703) below.
When you need to specify a filename from which to get experimental data (Oper = EADD), place that
string here. Valid entry is any filename string. You can enter the entire path\filename.extension
string and leave the next two fields (Option3 and Option4) blank, or you can specify the name here,
the extension in the next field, and the path following.
1701
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBFT
For plotting (Oper = PLOT), enter HYPER.
Option3
If a filename for experimental data is being specified in Option2 (Oper = EADD), this field will contain
the file extension.
For plotting operations (Oper = PLOT), use this field to specify the constitutive model name (see
Table 261: Creep Options (p. 1703) ).
Option4
When you are working on a specific coefficient (Oper = FIX), this field , specifies the index of that coef-
ficient. Valid entries vary from 1 to n, where n is the total number of coefficients (default = 1).
For Oper = SET, see Table 260: Set Operations (p. 1702), below.
If a filename for experimental data is being specified in Option2 (Oper = EADD), this field will contain
the directory/path specification.
If Oper = SOLVE, this value specifies the curve-fitting procedure. Valid entries are 0 for unnormalized
least squares curve-fitting procedure, and 1 for normalized least squares curve-fitting procedure.
Option5
If Oper = SOLVE, use this field to specify the number of iterations to be used in the calculation of the
coefficients. Valid entry is any positive integer (default = 1000)
If you specify a coefficient to be held constant (Oper = FIX), enter a 1 to FIX the coefficient, and a 0 to
allow it to vary (turn fixing OFF).
For Oper = SET, see Table 260: Set Operations (p. 1702), below.
Table 260 Set Operations
Option5 Option4 Purpose
Value of coefficient Index of coefficient Set the value of the coefficient
1 for ON and 0 for OFF tdep Turn temperature dependency
ON/OFF*
Temperature value tref Set reference temperature
* - When tdep is OFF, the Arrhenius term (e.g. C4) in the strain hardening creep equation is calculated.
Because TBTEMP is not included in the curve fitting regimen, the experimental data (in absolute temper-
ature) must be supplied using the /TEMP command.
* - When tdep is set ON, the Arrhenius term is set to 0 and the constants are separately computed for
each temperature. The curve fitting regimen uses TBTEMP. In this case, you must supply the experimental
data using the same units as the TREF command.
Option6
If Oper = SOLVE, specifies the allowed tolerance in residual change to stop an iteration. Valid entry is
0.0 to 1.0 (default = 0.0).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1702
TBFT
Option7
If Oper = SOLVE, specifies the allowed tolerance in coefficient change to stop an iteration. Valid entry
is 0 to 1 (default = 0).
Table 261 Creep Options
Option Name Category
NA SHAR CREEP
NA THAR CREEP
NA GEXP CREEP
NA GGRA CREEP
NA GBLA CREEP
NA MTHA CREEP
NA MSHA CREEP
NA GGAR CREEP
NA EXPO CREEP
NA NORT CREEP
NA PSTH CREEP
NA PSRP CREEP
NA GTHA CREEP
Following table summarizes the format for creep operations via the TBFT command:
Creep Model Command Summary
Option7 Option6 Option5 Option4 Option3 Option2 Option1 ID Oper
NA Option CREEP ID FADD
NA Option CREEP ID FDEL
NA Option CREEP ID FSET
Value Index NA Option CREEP ID SET
CTOL RTOL Num Iter
Norm
Flag
NA Option CREEP ID SOLVE
Fix / Un-
fix Flag
Index NA Option CREEP ID FIX
Direct-
ory
Exten-
sion
File Exp Type ID EADD
Exp Type
/ Index
ID EDEL
Option CREEP
Exp Type
/ Index
ID PLOT
TBFT Specifications for Viscoelastic Models
TBFT, Oper, ID, Option1, Option2, Option3, . . . , Option7
1703
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBFT
Oper
The specific curve fitting operation:
FCASE
Define a case/constitutive model for viscoelasticity (prony).
FADD
Define a constitutive model.
FDEL
Delete a constitutive model.
FSET
Write data related to a constitutive model to the database (same as TB command).
SET
Initialize coefficients of a constitutive model for nonlinear curve-fitting procedure.
CDEL
Deletes coefficients at current reference temperature. Applicable only for temperature dependent
coefficients.
SOLVE
Solve for coefficients.
FIX
Fix (hold constant) the coefficient you specify in Option4.
EADD
Add experimental data.
EDEL
Delete experimental data.
LIST
List all data associated with the material model represented by the material ID number.
PLOT
Plot curves.
ID
The material reference number (same as MAT argument used in the TB command). Valid entry is any
number greater than zero (default = 1) but less than 100,000.
Option1
This option is set to VISCO for adding coefficients (Oper = FADD). The FADD operation is used to specify
the shear order, the bulk order or the shift option.
This option (Oper = CASE) is set to NEW or FINI. The FADD commands are always enclosed by TBFT,FCASE,
ID, NEW and TBFT,FCASE, ID, FINI. See "Material Curve Fitting" in the Structural Analysis Guide for more
information.
For other curve fit function operations (Oper = FDEL, FSET, SET, CDEL, SOLVE or FIX) this field is set to
CASE.
For adding, deleting or plotting your experiment (Oper = EADD, EDEL, or PLOT), this field specifies the
experiment type. Valid entries are: SDEC (Shear Modulus vs. Time) or BDEC (Bulk Modulus vs. Time).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1704
TBFT
Option2
For defining your viscoelastic case (Oper = FCASE), you specify PVHE.
For curve fit function operations (Oper = FDEL, FSET, SET, CDEL, SOLVE, or FIX), this field specifies the
case name.
To create a new case (Oper = FADD), the valid options are listed in Table 263: Viscoelastic Options (p. 1706)
below.
When you need to specify a filename from which to get experimental data (Oper = EADD), place that
string here. Valid entry is any filename string. You can enter the entire path\filename.extension
string and leave the next two fields (Option3 and Option4) blank, or you can specify the name here,
the extension in the next field, and the path following.
For plotting (Oper = PLOT), enter the CASE name.
Option3
For Oper = FCASE, this field specifies the CASE name.
For Oper = FADD, this field specifies the order
If a filename for experimental data is being specified in Option2 (Oper = EADD), this field will contain
the file extension.
For plotting operations (Oper = PLOT), use this field to specify the CASE name.
Option4
When you are working on a specific coefficient (Oper = FIX), this field specifies the index of that coeffi-
cient. Valid entries vary from 1 to n, where n is the total number of coefficients (default = 1).
For Oper = SET, see Table 262: Set Operations (p. 1705), below.
You can also specify TREF to indicate the reference temperature, or COMP for a partial/complete solution
(only for bulk, only for shear, or all coefficients).
If a filename for experimental data is being specified in Option2 (Oper = EADD), this field will contain
the directory/path specification.
If Oper = SOLVE, this value specifies the curve-fitting procedure. Valid entries are 0 for unnormalized
least squares curve-fitting procedure, and 1 for normalized least squares curve-fitting procedure.
Option5
For Oper = SET, see Table 262: Set Operations (p. 1705), below.
Table 262 Set Operations
Option5 Option4 Purpose
Value of coefficient Index of coefficient Set the value of the coefficient
1 for ON and 0 for OFF tdep Turn temperature dependency
ON/OFF
Temperature value tref Set reference temperature
pshea, pbulk, or pvhe comp Set current component to solve for
1705
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBFT
If Oper = SOLVE, use this field to specify the number of iterations to be used in the calculation of the
coefficients. Valid entry is any positive integer (default = 1000).
If you are specifying a coefficient to be held constant (Oper = FIX), a 1 FIXES the specified coefficient,
and a 0 allows it to vary (turns fixing OFF).
Option6
If Oper = SOLVE, specifies the allowed tolerance in residual change to stop an iteration. Valid entry is
0.0 to 1.0 (default = 0.0).
Option7
If Oper = SOLVE, specifies the allowed tolerance in coefficient change to stop an iteration. Valid entry
is 0 to 1 (default = 0).
Table 263 Viscoelastic Options
Option Name Category
1 to N PSHEAR VISCO
1 to N PBULK VISCO
TN, WLF, NONE SHIFT VISCO
Following table summarizes the format for viscoelastic operations via the TBFT command:
Viscoelastic Models
Option7 Option6 Option5 Option4 Option3 Option2 Option1 ID Oper
Case
Name
PVHE NEW ID FCASE
FINI ID FCASE
Order Option VISCO ID FADD
NA
Case
Name
CASE ID FDEL
NA
Case
Name
CASE ID FSET
Value Index NA
Case
Name
CASE ID SET
CTOL RTOL Num Iter
Norm
Flag
NA
Case
Name
CASE ID SOLVE
Fix / Un-
fix Flag
Index NA
Case
Name
CASE ID FIX
Direct-
ory
Exten-
sion
File Exp Type ID EADD
Exp Type
/ Index
ID EDEL
NA
Case
Name
CASE
Exp Type
/ Index
ID PLOT
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1706
TBFT
Notes
This command provides tools for comparing experimental material data to the calculated data ANSYS supplies
for various non linear material options. Based on curve fitting plot comparisons and error norms, you choose
the model to use in solution according to the best fit. All of the capabilities of the TBFT-series of commands
are accessible interactively via the standard material GUI. See "Material Curve Fitting" in the Structural Ana-
lysis Guide for more information.
You can display material model data associated with both the TB command and the TBFT,FSET command
by issuing TBLIST,ALL,ALL.
Material model data associated with the latest TB or TBFT,FSET command overwrites previous data.
You can display material model data associated with both the TB command and the TBFT,FSET command
by issuing TBLIST,ALL,ALL.
The uniaxial, biaxial, and shear experimental data use engineering stress. The volumetric data uses true stress.
See the Element Reference for details on experimental data for creep and viscoelasticity.
Menu Paths
Main Menu> Preprocessor> Material Props> Material Models
TBLE
Specifies "Data table properties" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and will appear
in the log file (Jobname.LOG) if status is requested for some items under Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
Utility Menu>List>Status>Preprocessor>Data Tables
1707
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBLE
TBLIST, Lab, MAT
Lists the data tables.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lab
Data table label (see the TB command for labels). Defaults to the active table. If ALL, list data for all labels.
MAT
Material number to be listed (defaults to the active material). If ALL, list data tables for all materials.
Notes
This command is a utility command, valid anywhere.
Menu Paths
Utility Menu>List>Properties>Data Tables
TBMODIF, ROW, COL, VALUE
Modifies data for the data table (GUI).
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
ROW, COL
The row and column numbers of the table entry to be modified.
VALUE
The new value to be used in the ROW, COL location.
Notes
Modifies data for the table specified on the last TB command. For temperature-dependent data, the temper-
ature specified on the last TBTEMP command is used. This is a command generated by the Graphical User
Interface (GUI). It will appear in the log file (Jobname.LOG) if a TB data table is graphically edited in
spreadsheet fashion. This command is not intended to be typed in directly in an ANSYS session (although
it can be included in an input file for batch input or for use with the /INPUT command).
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1708
TBLIST
TBPLOT, Lab, MAT, TBOPT, TEMP, SEGN
Displays the data table.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH DY PP <> EME MFS
Lab
Data table label. Valid labels are: MKIN, KINH, MELAS, MISO, BKIN, BISO, BH, GASKET, and JOIN. Defaults
to the active table label. For B-H data, also valid are: NB to display NU-B
2
, MH to display MU vs. H, and
SBH, SNB, SMH to display the slopes of the corresponding data.
MAT
Material number to be displayed (defaults to the active material).
TBOPT
Gasket material or joint element material option to be plotted.
The following gasket material options are valid only when Lab = GASKET:
ALL
Plots all gasket data.
COMP
Plots gasket compression data only.
LUNL
Plots gasket linear unloading data with compression curve.
NUNL
Plots gasket nonlinear unloading data only.
The following joint element material options are valid only when Lab = JOIN:
JNSA
Plots nonlinear stiffness data that is applicable to all relevant directions.
JNSn
Plots only the specified nonlinear stiffness data. The "n" can be 1, 4, or 6. For example, JNS4 plots
only the nonlinear stiffness data specified in the local direction 4 (ROTX).
JNDA
Plots nonlinear damping data that is applicable to all relevant directions.
JNDn
Plots only the specified nonlinear damping data. The "n" can be 1, 4, or 6. For example, JND4 plots
only the nonlinear damping data specified in the local direction 4 (ROTX).
JNFA
Plots nonlinear hysteretic friction data that is applicable to all relevant directions.
JNFn
Plots only the specified nonlinear hysteretic friction data. The "n" can be 1, 4, or 6. For example, JNF4
plots only the nonlinear hysteretic friction data specified in local direction 4 (ROTX).
TEMP
Specific temperature at which gasket data or joint element material data will be plotted (used only when
Lab = GASKET or JOIN). Use TEMP = ALL to plot gasket data or joint element material data at all tem-
peratures.
1709
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TBPLOT
SEGN
Segment number of plotted curve (valid only when Lab = GASKET):
NO
Segment number is not added to plotted curve (default).
YES
Segment number is added to plotted curve. This option is ignored if the number of data points in
a curve exceeds 20.
Notes
Only data for stress-strain, B-H, gasket curves, or joint element nonlinear material model curves may be
displayed. TBOPT and TEMP are used only when Lab = GASKET or JOIN. SEGN is used only when Lab =
GASKET.
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Data Tables
TBPT, Oper, X, Y
Defines a point on a nonlinear data curve.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Oper
Operation to perform:
DEFI
Defines a new data point (default). The point is inserted into the table in ascending order of X. If a
point already exists with the same X value, it is replaced.
DELE
Deletes an existing point. The X value must match the X value of the point to be deleted (Y is ignored).
X
The X value of the point (for example, strain, H, etc.).
Y
The corresponding Y value of the point (for example, stress, B, etc.).
Notes
TBPT defines a point on a nonlinear data curve (such as a stress-strain curve, B-H curve, etc.) at the temper-
ature specified on the last TBTEMP command. The meaning of the X and Y values will depend on the type
of data table specified on the last TB command (MISO, BH, etc.).
This command is also valid in SOLUTION.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1710
TBPT
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
TBTEMP, TEMP, KMOD
Defines a temperature for the data table.
PREP7: Data Tables
MP ME ST PR PRN <> <> <> EM <> DY PP <> EME MFS
TEMP
Temperature value (defaults to 0.0 if KMOD is blank).
KMOD
If blank, TEMP defines a new temperature. If an integer, 1 to NTEMP (from the TB command), modify
that previously defined temperature to the TEMP value, unless TEMP is blank, then that previously
defined temperature is reactivated. Use TBLIST to list temperatures and data. The next TBDATA or TBPT
commands also add or change the data at this temperature. If KMOD = CRIT (and TEMP is blank), the
next TBDATA values are failure-criteria keys. If KMOD = STRAIN (and TEMP is blank), the next TBDATA
values are strains as described for the MKIN property option (see Material Data Tables (Implicit Analysis)
of the Element Reference).
Notes
Defines a temperature to be associated with the data on the subsequent TBPT or TBDATA commands. The
temperature remains active until the next TBTEMP command is input. Data values must be defined with
the temperatures in ascending order. Temperatures previously associated with a data table may also be
modified.
This command is also valid in SOLUTION.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
TCHG, ENAME1, ENAME2, ETYPE2
Converts 20-node degenerate tetrahedral elements to their 10-node non-degenerate counterparts.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ENAME1
Name (or the number) of the 20-node tetrahedron element that you want to convert. This argument is
required.
1711
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TCHG
ENAME2
Name (or the number) of the 10-node tetrahedron element to which you want to convert the ENAME
elements. This argument is required.
ETYPE2
Element TYPE reference number for ENAME2. If ETYPE2 is 0 or is not specified, ANSYS chooses the
element TYPE reference number for ENAME2. See the "Notes" (p. 1712) section for details. This argument
is optional.
Notes
The TCHG command allows you to specify conversion of any selected 20-node brick that is degenerated
into a tetrahedron to a 10-node tetrahedron.
The TCHG command is useful when used in with the MOPT,PYRA command. Twenty-node pyramid shaped
elements may be used in the same volume with 10-node tetrahedra.
Performing a conversion is likely to create circumstances in which more than one element type is defined
for a single volume.
If specified, ETYPE2 will usually be the same as the local element TYPE number (ET,ITYPE) that was assigned
to ENAME2 with the ET command. You can specify a unique number for ETYPE2 if you prefer. Although
ETYPE2 is optional, it may be useful when two or more ITYPEs have been assigned to the same element
(for example, if two SOLID92 elements have been established in the element attribute tables for the current
model, use the ETYPE2 argument to distinguish between them). If ETYPE2 is nonzero and it has not already
been assigned to an element via ET, ANSYS assigns the ETYPE2 value to ENAME2 as its element TYPE ref-
erence number.
If ETYPE2 is 0 or is not specified, ANSYS determines the element TYPE reference number for ENAME2 in
one of these ways:
If ETYPE2 is 0 or is not specified, and ENAME2 does not appear in the element attribute tables, ANSYS
uses the next available (unused) location in the element attribute tables to determine the element TYPE
reference number for ENAME2.
If ETYPE2 is 0 or is not specified, and ENAME2 appears in the element attribute tables, ANSYS uses
ENAME2 's existing element TYPE reference number for ETYPE2 . (If there is more than one occurrence
of ENAME2 in the element attribute tables (each with its own TYPE reference number), ANSYS uses the
first ENAME2 reference number for ETYPE2 .)
See Meshing Your Solid Model of the Modeling and Meshing Guide for detailed information about converting
degenerate tetrahedral elements.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Change Tets
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1712
TCHG
/TEE, Label, Fname, Ext, --
Writes a list of commands to a specified file at the same time that the commands are being executed.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Label
Indicates how ANSYS is to interpret this /TEE command:
NEW --
Signals the beginning of the command text that is to be written to Fname. If Fname already exists,
specifying NEW causes the contents of Fname to be overwritten.
APPEND --
Indicates that you want to append to Fname the command text that follows.
END --
Signals the end of the command text that is to be written to or appended to Fname.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
If you plan to execute the file as if it were an ANSYS command, use the extension .mac.
--
Unused field.
Notes
You can use the /TEE command to record a macro to a specified file at the same time that the macro is
being executed. It is similar to the UNIX tee command.
For more information about the /TEE command, see the "Introducing APDL" of the ANSYS Parametric Design
Language Guide.
The following example illustrates the use of the /TEE command. If you issue these commands:
/tee,new,myfile,mac
et,1,42,0,0,1
ex,1,3e7
/tee,end
/tee,append,myfile,mac
n,1,8
n,5,11
fill
ngen,5,5,1,5,1,0,1
/tee,end
the content of myfile.mac is:
et,1,42,0,0,1
ex,1,3e7
1713
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/TEE
n,1,8
n,5,11
fill
ngen,5,5,1,5,1,0,1
This command is valid in any processor, but only during an interactive run.
Menu Paths
This command cannot be accessed from a menu.
TEE, NCENT, TYPE, ELEM, EINC, L1, L2, L3
Defines a tee in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NCENT
Node where three straight pipes intersect forming a tee (or "Y"). Defaults to last starting branch node
[BRANCH].
TYPE
Type of tee:
WT
Welding tee (default).
r = (D
0
- t
w
) / 2
h = 4.4 t
w
/ r
SIF = 0.9 / (h
2/3
)
If (SIF < 1) SIF = 1
UFT
Unreinforced fabricated tee.
r = (D
0
- t
w
) / 2
h = t
w
/ r
SIF = 0.9 / (h
2/3
)
If (SIF < 1) SIF = 1
ELEM
Element number to be assigned to first tee leg (defaults to the previous maximum element number
(MAXEL) + 1).
EINC
Element number increment (defaults to 1).
L1, L2, L3
Tee leg lengths (corresponding in order of increasing straight pipe element numbers). Must be less than
the straight pipe length. Defaults to 2 x OD of straight pipe (for each leg).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1714
TEE
Notes
Defines a tee in place of the tee intersection of three previously defined straight pipe elements. See the
PREP7 RUN command. The new tee is also composed of three PIPE16 straight pipe elements, but of the leg
lengths specified and with the appropriate tee factors calculated. Three new nodes are generated at the
ends of the tee. The original three straight pipes are automatically "shortened" to meet the ends of the tee.
The tee specifications and loadings are taken from the corresponding three straight pipes.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Define Pipes>Pipe Tee
TERM, Kywrd, Opt1, Opt2, Opt3
Specifies various terminal driver options.
DISPLAY: Driver Options
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
If Kywrd = COPY, command format is TERM,COPY,NCOPY
NCOPY
Activate hard copy device for NCOPY (0,1,2, etc.) copies.
If Kywrd = LOOP, command format is TERM,LOOP,NLOOP,PAUSE. Used only with PLOT,ALL
NLOOP
Loop NLOOP times back to beginning of file when end of file is reached.
PAUSE
Pause PAUSE seconds between plots.
If Kywrd = NOPROM, command format is TERM,NOPROM,KEY
KEY
Prompt key:
0
Display prompt line for prompt.
1
Use terminal bell for prompt.
Notes
Used only with terminal driver names on /SHOWDISP command.
This command is also valid in PREP7.
Menu Paths
This command cannot be accessed from a menu.
1715
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TERM
THOPT, Refopt, REFORMTOL, NTABPOINTS, TEMPMIN, TEMPMAX
Nonlinear transient thermal solution option.
SOLUTION: Analysis Options
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
Refopt
Matrix reform option.
FULL
Use the full Newton-Raphson solution option (default).
QUASI
Use a selective reform solution option based on REFORMTOL.
LINEAR
Use a linear solution option with no matrix reform.
REFORMTOL
Property change tolerance for Matrix Reformation (.05 default). The thermal matrices are reformed if the
maximum material property change in an element (from the previous reform time) is greater than the
reform tolerance.
NTABPOINTS
Number of points in Fast Material Table (64 default).
TEMPMIN
Minimum temperature for Fast Material Table. Defaults to the minimum temperature defined by the
MPTEMP command for any material property defined.
TEMPMAX
Maximum temperature for Fast Material Table. Defaults to the maximum temperature defined by the
MPTEMP command for any material property defined.
Notes
This solution logic is only supported by the ICCG and the JCG solvers (EQSLV command). You must issue
the THOPT command before the EQSLV command to use the QUASI option.
For Refopt = QUASI, results from a restart may be different than results from a single run because the
stiffness matrices are always recreated in a restart run, but may or may not be in a single run (depending
on the behavior resulting from the REFORMTOL setting). Additionally, results may differ between two single
runs as well, if the matrices are reformed as a result of the REFORMTOL setting.
Because the matrices are reformed based on material properties, only Refopt = FULL should be used for
the radiosity solver method.
THOPT settings take precedence over SOLCONTROL settings.
ANTYPE,,RESTART (including multiframe restart) is not supported by THOPT, LINEAR option.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>Analysis Options
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1716
THOPT
TIFF, Kywrd, OPT
Provides TIFF file Export for ANSYS Displays.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Kywrd
Specifies various TIFF file export options.
COMP
If Kywrd = COMP, then OPT controls data compression for the output file. If COMP = 0, then com-
pression is off. If COMP = 1 (default), then compression is on.
ORIENT
If Kywrd = ORIENT, then OPT will determine the orientation of the entire plot. OPT can be either
Horizontal (default) or Vertical.
COLOR
If Kywrd = COLOR, then OPT will determine the color attribute of the saved file. OPT can be 0, 1,
or 2, corresponding to Black and White, Grayscale, and Color (default), respectively.
TMOD
If Kywrd = TMOD, then OPT will determine the text method. OPT can be either 1 or 0, corresponding
to bitmap text (default) or line stroke text, respectively.
DEFAULT
If Kywrd = DEFAULT, then all of the default values, for all of the Kywrd parameters listed above, are
active.
OPT
OPT can have the following names or values, depending on the value for Kywrd (see above).
1 or 0
If Kywrd = COMP, a value or 1 (on) or 0 (off ) will control compression for the TIFF file.
Horizontal, Vertical
If Kywrd = ORIENT, the terms Horizontal or Vertical determine the orientation of the plot.
0, 1, 2
If Kywrd = COLOR, the numbers 0, 1, and 2 correspond to Black and White , Grayscale and Color,
respectively.
1, 0
If Kywrd = TMOD, the values 1 and 0 determine whether bitmap (1) or stroke text (0) fonts will be
used
Menu Paths
Utility Menu>PlotCtrls>HardCopy>ToFile
1717
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TIFF
TIME, TIME
Sets the time for a load step.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM <> DY PP <> EME MFS
TIME
Time at the end of the load step.
Command Default
Previous TIME + 1.0 (at each load step), corresponding to the load step number.
Notes
Associates the boundary conditions at the end of the load step with a particular TIME value.
TIME must be a positive, nonzero, monotonically increasing quantity that "tracks" the input history. Units
of time should be consistent with those used elsewhere (for properties, creep equations, etc.).
Typically, for the first load step TIME defaults to 1. However, for the first load step of a reduced transient
analysis (ANTYPE,TRANS and TRNOPT,REDUC) or mode superposition transient analysis (ANTYPE,TRANS
and TRNOPT,MSUP), the TIME command is ignored and a static solution is performed at TIME = 0.
For a full transient analyses, the command's default behavior does not apply. You must specify a time for
each load step and it must be greater than the time at the end of the prior load step.
TIME does not apply to modal (ANTYPE,MODAL), harmonic response (ANTYPE,HARMIC), or substructure
(ANTYPE,SUBSTR) analyses.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Basic
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time and Substps
Main Menu>Solution>Analysis Type>Sol'n Controls>Basic
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time and Substps
Main Menu>Solution>Time Controls>Solution Time
TIMERANGE, TMIN, TMAX
Specifies the time range for which data are to be stored.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
TMIN
Minimum time (defaults to first time (or frequency) point on the file).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1718
TIME
TMAX
Maximum time (defaults to last time (or frequency) point on the file).
Command Default
Include all time (or frequency) points in the range.
Notes
Defines the time (or frequency) range for which data are to be read from the file and stored in memory. Use
the NSTORE command to define the time increment.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Data
TIMINT, Key, Lab
Turns on transient effects.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Key
Transient effects key:
OFF
No transient effects (static or steady-state).
ON
Include transient (mass or inertia) effects.
Lab
Degree of freedom label:
ALL
Apply this key to all appropriate labels (default).
STRUC
Apply this key to structural DOFs.
THERM
Apply this key to thermal DOFs.
ELECT
Apply this key to electric DOFs.
MAG
Apply this key to magnetic DOFs.
FLUID
Apply this key to fluid DOFs.
Command Default
Include transient effects (ON) if ANTYPE,TRANS, exclude transient effects (OFF) if ANTYPE,STATIC.
1719
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TIMINT
Notes
Indicates whether this load step in a full transient analysis should use time integration, that is, whether it
includes transient effects (e.g. structural inertia, thermal capacitance) or whether it is a static (steady-state)
load step for the indicated DOFs. Transient initial conditions are introduced at the load step having Key =
ON. Initial conditions are then determined from the previous two substeps. Zero initial velocity and acceler-
ation are assumed if no previous substeps exist. See the Structural Analysis Guide, the Thermal Analysis Guide,
and the Low-Frequency Electromagnetic Analysis Guide for details.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Transient
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time Integration>Amplitude
Decay
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time Integration>Newmark
Parameters
Main Menu>Solution>Analysis Type>Sol'n Controls>Transient
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time Integration>Amplitude Decay
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time Integration>Newmark Parameters
TIMP, ELEM, CHGBND, IMPLEVEL
Improves the quality of tetrahedral elements that are not associated with a volume.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ELEM
Identifies the tetrahedral elements to be improved. Valid values are ALL and P. If ELEM = ALL (default),
improve all selected tetrahedral elements. If ELEM = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI).
CHGBND
Specifies whether to allow boundary modification. Boundary modification includes such things as changes
in the connectivity of the element faces on the boundary and the addition of boundary nodes. (Also see
the Notes section below for important usage information for CHGBND.)
0
Do not allow boundary modification.
1
Allow boundary modification (default).
IMPLEVEL
Identifies the level of improvement to be performed on the elements. (Improvement occurs primarily
through the use of face swapping and node smoothing techniques.)
0
Perform the least amount of swapping/smoothing.
1
Perform an intermediate amount of swapping/smoothing.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1720
TIMP
2
Perform the greatest amount of swapping/smoothing.
3
Perform the greatest amount of swapping/smoothing, plus additional improvement techniques
(default).
Notes
The TIMP command enables you to improve a given tetrahedral mesh by reducing the number of poorly-
shaped tetrahedral elements (in particular, the number of sliver tetrahedral elements)--as well as the overall
number of elements--in the mesh. It also improves the overall quality of the mesh.
TIMP is particularly useful for an imported tetrahedral mesh for which no geometry information is attached.
Regardless of the value of the CHGBND argument, boundary mid-nodes can be moved as long as you are
not using p-method analysis. When CHGBND = 0 and you are using p-method analysis, boundary mid-nodes
cannot be moved. (ANSYS issues an error message if it would be necessary to move boundary mid-nodes
in order to generate valid quadratic elements.)
When loads or constraints have been placed on boundary nodes or mid-nodes, and boundary mid-nodes
are later moved, ANSYS issues a warning message to let you know that it will not update the loads or con-
straints.
No boundary modification is performed if shell or beam elements are present in the mesh, even when
CHGBND = 1.
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Improve Tets>Detached Elems
TINTP, GAMMA, ALPHA, DELTA, THETA, OSLM, TOL, --, --, AVSMOOTH, ALPHAF, ALPHAM
Defines transient integration parameters.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
GAMMA
Amplitude decay factor for 2nd order transient integration, e.g., structural dynamics (used only if ALPHA,
DELTA, ALPHAF, and ALPHAM are blank). Defaults to 0.005.
ALPHA
2nd order transient integration parameter (used only if GAMMA is blank). Defaults to 0.2525.
DELTA
2nd order transient integration parameter (used only if GAMMA is blank). Defaults to 0.5050.
THETA
1st order transient (e.g., thermal transient) integration parameter. Defaults to 0.5 if SOLCONTROL is OFF.
Defaults to 1.0 if SOLCONTROL is ON.
1721
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TINTP
OSLM
Specifies the oscillation limit criterion for automatic time stepping of 1st order transients (e.g., thermal
transients). Defaults to 0.5 with a tolerance of TOL.
TOL
Tolerance applied to OSLM. Defaults to 0.0.
--, --
Unused fields.
AVSMOOTH
Smooth flag option:
0
Include smoothing of initial velocity (1st order system) or initial acceleration (2nd order system)
(default).
1
Do not include smoothing.
ALPHAF
Interpolation factor in HHT algorithm for force and damping terms (used only if GAMMA is blank). Defaults
to 0.005.
ALPHAM
Interpolation factor in HHT algorithm for inertial term (used only if GAMMA is blank). Defaults to 0.0.
Notes
Used to define the transient integration parameters. For more information on transient integration parameters,
refer to the Theory Reference for the Mechanical APDL and Mechanical Applications.
For structural transient analyses, you may choose between the Newmark and HHT time integration methods
(see the TRNOPT command). In this case, if GAMMA is input and the integration parameters ALPHA, DELTA,
ALPHAF, and ALPHAM are left blank, the program will calculate the integration parameters. Alternatively,
you can input these integration parameters directly on this command. However, for the unconditional sta-
bility and second order accuracy of the time integration, these parameters should satisfy a specific relationship,
as described in Description of Structural and Other Second Order Systems of the Theory Reference for the
Mechanical APDL and Mechanical Applications.
In a transient piezoelectric analysis, required input for this command is ALPHA = 0.25, DELTA = 0.5, and
THETA = 0.5. For a coupled electromagnetic-circuit transient analysis, use THETA = 1.0, the default value,
to specify the backward Euler method.
The default values given for this command assume SOLCONTROL,ON (the default). If your analysis is using
SOLCONTROL,OFF, you may need to adjust some defaults to prevent the solution from diverging. See the
description of SOLCONTROL for a complete listing of the defaults set by SOLCONTROL,ON and SOLCON-
TROL,OFF.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Transient
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time Integration>Amplitude
Decay
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1722
TINTP
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time Integration>Newmark
Parameters
Main Menu>Solution>Analysis Type>Sol'n Controls>Transient
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time Integration>Amplitude Decay
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time Integration>Newmark Parameters
/TITLE, Title
Defines a main title.
DATABASE: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Title
Input up to 72 alphanumeric characters. Parameter substitution may be forced within the title by enclosing
the parameter name or parametric expression within percent (%) signs.
Notes
The title is carried through the printout and written on various files. The title written to a file is the title
defined at that time. Special characters may be used within the title text. Subtitles may also be defined
[/STITLE].
This command is valid in any processor.
Menu Paths
Utility Menu>File>Change Title
/TLABEL, XLOC, YLOC, Text
Creates annotation text (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XLOC
Text X starting location (-1.0 < X < 1.6).
YLOC
Text Y starting location (-1.0 < Y < 1.0).
Text
Text string (60 characters maximum). Parameter substitution may be forced within the text by enclosing
the parameter name or parametric expression within percent (%) signs.
Notes
Defines annotation text to be written directly onto the display at a specified location. This is a command
generated by the Graphical User Interface (GUI) and will appear in the log file (Jobname.LOG) if annotation
is used. This command is not intended to be typed in directly in an ANSYS session (although it can be included
in an input file for batch input or for use with the /INPUT command).
1723
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/TLABEL
All text is shown on subsequent displays unless the annotation is turned off or deleted. Use the /TSPEC
command to set the attributes of the text.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
TOCOMP, Refname, Type, NUMLC, LCARR
Defines single or multiple compliance as the topological optimization function.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Refname
Reference name (8 character string). TOCOMP,Refname with the other fields blank deletes the defined
compliance function.
Type
Sets the type of compliance function. Valid types are:
SINGLE
Sets the topological optimization function for Refname to single compliance. NumLC is the load
case number.
MULTIPLE
Sets the topological optimization function to weighted summation of individual compliances. NumLC
is the number of load cases considered
NUMLC
For Type = Single, NumLC is the load case identifier. For Type = Multiple, NumLC is the total number
of load cases considered.
LCARR
Used when Type = Multiple. The array (of dimension NumLC) defining the load case weights. Array
name must be enclosed in % signs: %WEIGHT%.
Notes
The function defined using TOCOMP is used as an objective or constraint for the topological optimization,
as defined further in TOVAR. When used as the objective, the optimization distributes the material so that
the compliance measure is minimized (stiffness is maximized).
Menu Paths
Main Menu>Topological Opt>Set Up>Advanced Opt>Topo Function
Main Menu>Topological Opt>Set Up>Basic Opt
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1724
TOCOMP
TODEF, ACCUR
Defines parameters for and initializes topological optimization.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
ACCUR
Accuracy used for termination and convergence checking. Default = 0.0005.
Notes
Initializes the topological optimization problem and defines the accuracy for the solution. Issue this command
before TOLOOP.
Menu Paths
Main Menu>Topological Opt>Run
TOEXE
Executes one topological optimization iteration.
OPTIMIZATION: Run
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
Runs one topological optimization iteration, leading to the prediction of a new shape, defined by means of
element densities. Before issuing TOEXE, you must perform a static or modal analysis solution depending
on the type of objective and constraints you defined. TOEXE performs a convergence test based on relative
objective and element density change with an accuracy as specified by TODEF.
Menu Paths
This command cannot be accessed from a menu.
TOFFST, VALUE
Specifies the temperature offset from absolute zero to zero.
SOLUTION: Analysis Options
AUX12: General Radiation
MP ME ST PR PRN DS DSS FL <> <> <> PP <> EME MFS
VALUE
Degrees between absolute zero and zero of temperature system used (should be positive).
1725
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TOFFST
Notes
Specifies the difference (in degrees) between absolute zero and the zero of the temperature system used.
Absolute temperature values are required in evaluating certain expressions, such as for creep, swelling, radi-
ation heat transfer, MASS71, etc. (The offset temperature is not used in evaluating emissivity.) Examples are
460 for the Fahrenheit system and 273 for the Celsius system. The offset temperature is internally included
in the element calculations and does not affect the temperature input or output. If used in SOLUTION, this
command is valid only within the first load step.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Temperature Units
Main Menu>Preprocessor>Material Props>Temperature Units
Main Menu>Preprocessor>Radiation Opts>Solution Opt
Main Menu>Radiation Opt>Radiosity Meth>Solution Opt
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Temperature Units
Main Menu>Solution>Radiation Opts>Solution Opt
TOFREQ, Refname, Type, Nfreq, Frqarr, Targval
Defines single or mean frequency formulation as the topological optimization function.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Refname
Reference name (8 character string). TOFREQ,Refname with other fields blank deletes the defined
compliance function.
Type
Defines type of frequency topological optimization function. Valid types are:
SINGLE
Defines single frequency as the topological optimization function.
WEIGHTED
Defines a weighted summation of individual frequencies as the topological optimization function.
RECIPROCAL
Uses a reciprocal formulation as the topological optimization function.
EUCLIDEAN
Uses a Euclidean Norm formulation as the topological optimization function.
Nfreq
For Type = SINGLE, Nfreq is the frequency identifier. For Type = WEIGHTED, RECIPROCAL, or EUCLIDEAN,
Nfreq is the total number of frequencies considered.
Frqarr
For Type = WEIGHTED or RECIPROCAL, Frqarr is the array of dimension Nfreq that defines the weights
for mean frequency formulation. For Type = EUCLIDEAN, Frqarr is the array (of dimension Nfreq)
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1726
TOFREQ
defining the target values for mean frequency formulation. The array name must be enclosed in % signs:
%ARRAY%.
Targval
For Type = RECIPROCAL, Targval specifies the target value for mean frequency formulation.
Targval Frqarr Nfreq Type
N/A N/A Frequency identifier SINGLE
N/A Defines weights Total number of frequencies WEIGHTED
Target value Defines weights Total number of frequencies RECIPROCAL
N/A Defines target values Total number of frequencies EUCLIDEAN
Notes
Single, when the frequency function is set as the objective (TOVAR), distributes material so that the single
frequency specified is maximized.
Weighted (available only as an objective function) distributes material so that the specified mean frequency
formulation is maximized.
Reciprocal (available only as an objective function) is a mean frequency formulation that defines a smoother
function than Single. It is best used when two modes whose eigenfrequencies occur in the given mean
formulation exchange their orders during optimization. The eigenfrequency that is closest to the target fre-
quency (set with Targval) experiences the largest increase and is maximized.
Euclidean Norm Formulation (available only as an objective function) is used to shift single or multiple ei-
genfrequencies up. The mean eigenfrequency is utilized to smooth the optimization process. This must be
used as the objective where the optimization procedure distributes material so that the mean frequency is
minimized.
If the optimization problem does not converge, try specifying a different type of frequency formulation.
Each formulation uses different methods to calculate the frequency optimization, so depending on your
specific problem, one might work better than another. See the Theory Reference for the Mechanical APDL and
Mechanical Applications for more information on these formulations.
Menu Paths
Main Menu>Topological Opt>Set Up>Advanced Opt>Topo Function
TOGRAPH, Type, Refname
Plots iteration solution of topological optimization.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Type
OBJ
Plot objective history versus iteration number.
1727
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TOGRAPH
CON
Plot constraint history versus iteration number.
Refname
Used if multiple constraints were defined. Defaults to the first constraint name.
Notes
Plots all iterations of the topological optimization solution, using either objective or constraint history.
Menu Paths
Main Menu>Topological Opt>Graph History
TOLIST
Lists all topological optimization functions currently defined.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
Provides a list of all defined topological functions, in the order they were defined.
Menu Paths
Main Menu>Topological Opt>Set Up>Advanced Opt>List Functions
TOLOOP, NITER, PLOT
Execute several topological optimization iterations.
OPTIMIZATION: Run
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NITER
Number of iterations to be performed (maximum of 100). Default is 1.
PLOT
Sets display of topological densities for each iteration:
0
Do not display the results of each iteration.
1
Display the results of each iteration.
Command Default
TOLOOP,1,0.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1728
TOLIST
Notes
Invokes a macro to solve, postprocess, and plot each iteration. For compliance cases, you must write at least
one load step (using LSWRITE) before issuing this command. LSSOLVE is used for static analyses if there
are multiple load steps. The Block Lanczos eigensolver is used for modal analyses. PLNSOL,TOPO and TOEXE
are used for each iteration. The macro terminates when either the number of iterations or the convergence
criteria is met, as specified in TODEF,ACCUR.
Menu Paths
Main Menu>Topological Opt>Run
*TOPER, ParR, Par1, Oper, Par2, FACT1, FACT2, CON1
Operates on table parameters.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
Name of the resulting table parameter. The command will create a table array parameter with this name.
Any existing parameter with this name will be overwritten.
Par1
Name of the first table parameter.
Oper
The operation to be performed: ADD. The operation is: ParR(i,j,k) = FACT1*Par1(i,j,k) + FACT2 *Par2(i,j,k)
+CON1
Par2
Name of the second table parameter.
FACT1
The first table parameter multiplying constant. Defaults to 1.
FACT2
The second table parameter multiplying constant. Defaults to 1.
CON1
The constant increment for offset. Defaults to 0.
Notes
*TOPER operates on table parameters according to: ParR(i,j,k) = FACT1*Par1(i,j,k) + FACT2 *Par2(i,j,k) +CON1
Par1 and Par2 must have the same dimensions and the same variable names corresponding to those dimen-
sions. Par1 and Par2 must also have identical index values for rows, columns, etc.
If you want a local coordinate system for the resulting array, you must dimension it as such using the *DIM
command before issuing *TOPER.
1729
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*TOPER
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Table Operations
TOPLOT, AVRG
Plot current topological density distribution.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
AVRG
Sets TOPLOT to show smoothed nodal solution or non-smoothed element solution.
0
Plot smoothed nodal solution. Default.
1
Plot non-smoothed element solution.
Notes
Invokes a macro to postprocess the current element pseudo densities for topological optimization. TOPLOT,0
uses PLNSOL,TOPO, and TOPLOT,1 uses PLESOL,TOPO to plot the current density distribution predicted by
the topological optimization.
Menu Paths
Main Menu>Topological Opt>Plot Dens Unavg
Main Menu>Topological Opt>Plot Densities
TOPRINT, Type, Refname
Print iteration solution history of topological optimization.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Type
OBJ
Print objective history versus iteration number.
CON
Print constraint history versus iteration number.
Refname
Used if multiple constraints were defined. Defaults to the first constraint name.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1730
TOPLOT
Notes
Prints one iteration of the topological optimization solution, using either objective or constraint history.
Menu Paths
Main Menu>Topological Opt>Print History
TORQ2D
Calculates torque on a body in a magnetic field.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
TORQ2D invokes an ANSYS macro which calculates mechanical torque on a body in a magnetic field. The
body must be completely surrounded by air (symmetry permitted), and a closed path [PATH] passing through
the air elements surrounding the body must be available. A counterclockwise ordering of nodes on the
PPATH command will give the correct sign on the torque result. The macro is valid for 2-D planar analysis.
The calculated torque is stored in the parameter TORQUE. A node plot showing the path is produced in in-
teractive mode. The torque is calculated using a Maxwell stress tensor approach. Path operations are used
for the calculation, and all path items are cleared upon completion. See the TORQC2D command for torque
calculation based on a circular path.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>Torque
TORQC2D, RAD, NUMN, LCSYS
Calculates torque on a body in a magnetic field based on a circular path.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
RAD
Radius of the circular path. The nodes for the path are created at this radius.
NUMN
Number of nodes to be created for the circular path. The greater the number of nodes, the higher the
accuracy of the torque evaluation. Defaults to 18.
LCSYS
(Optional) Local coordinate system number to be used for defining the circular arc of nodes and the
path. Defaults to 99. (If a local system numbered 99 already exists, it will be overwritten by this default.)
1731
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TORQC2D
Notes
TORQC2D invokes an ANSYS macro which calculates the mechanical torque on a body using a circular path.
It is used for a circular or cylindrical body such as a rotor in an electric machine. The body must be centered
about the global origin and must be surrounded by air elements. The air elements surrounding the path at
radius RAD must be selected, and elements with a high-permeability material should be unselected prior to
using the macro. The macro is valid for 2-D planar analyses only. For a harmonic analysis, the macro calculates
the time-average torque. Radial symmetry models are allowed, i.e., the model need not be a full 360 model.
The calculated torque is stored in the parameter TORQUE. If the model is not a full 360 model, TORQUE
should be multiplied by the appropriate factor (such as 4.0 for a 90 sector) to obtain the total torque. A
node plot showing the path is produced in interactive mode.
The torque is calculated via a circular path integral of the Maxwell stress tensor. The circular path and the
nodes for the path are created by the macro at the specified radius RAD. Path operations are used for the
calculation, and all path items are cleared upon completion. See the TORQ2D command for torque calculation
based on an arbitrary, non-circular path.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Path Based>Circular Torq
TORQSUM, Cnam1, Cnam2, Cnam3, Cnam4, Cnam5, Cnam6, Cnam7, Cnam8, Cnam9
Summarizes electromagnetic torque calculations on element components.
POST1: Magnetics Calculations
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Cnam1, Cnam2, Cnam3, . . . , Cnam9
Names of existing element components for which Maxwell or virtual work boundary conditions were
applied in the preprocessor. Must be enclosed in single quotes (e.g., 'CNAM1') when the command typed
in the command input box.
Notes
TORQSUM invokes an ANSYS macro that summarizes the Maxwell and virtual work torque values. The element
components must have had appropriate Maxwell or virtual work boundary conditions established in the
preprocessor prior to solution in order to retrieve torques (see the FMAGBC command). The torque values
are stored on a per-element basis for the adjacent air layer elements surrounding the components and are
retrieved and summed by the macro. For a harmonic analysis, the calculated torque represents a time-average
value.
TORQSUM is valid only for 2-D planar analysis.
Menu Paths
Main Menu>General Postproc>Elec&Mag Calc>Component Based>Torque
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1732
TORQSUM
TORUS, RAD1, RAD2, RAD3, THETA1, THETA2
Creates a toroidal volume.
PREP7: Primitives
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
RAD1, RAD2, RAD3
Three values that define the radii of the torus. You can specify the radii in any order. The smallest of the
values is the inner minor radius, the intermediate value is the outer minor radius, and the largest value
is the major radius. (There is one exception regarding the order of the radii values--if you want to create
a solid torus, specify zero or blank for the inner minor radius, in which case the zero or blank must occupy
either the RAD1 or RAD2 position.) At least two of the values that you specify must be positive values;
they will be used to define the outer minor radius and the major radius. See the diagram in the Notes
section for a view of a toroidal sector showing all radii.
THETA1, THETA2
Starting and ending angles (either order) of the torus. Used for creating a toroidal sector. The sector
begins at the algebraically smaller angle, extends in a positive angular direction, and ends at the larger
angle. The starting angle defaults to 0 and the ending angle defaults to 360.
Notes
Defines a toroidal volume centered about the working plane origin. A solid torus of 360 will be defined
with four areas, each area spanning 180 around the major and minor circumference.
To create the toroidal sector shown below, the command TORUS,5,1,2,0,180 was issued. Since "1" was the
smallest radii value specified, it defined the inner minor radius; since "2" was the intermediate radii value
specified, it defined the outer minor radius; and since "5" was the largest radii value specified, it defined the
major radius. The values "0" and "180" defined the starting and ending angles of the torus.
Inner
minor
radius
Outer
minor
radius
Major
radius
1733
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TORUS
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Torus
TOSTAT
Displays topological optimization status and results information.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Notes
Displays information for defined functions from TOVAR, TODEF, and TOTYPE, and results information such
as final volume, number of iterations, and convergence status.
Menu Paths
Main Menu>Topological Opt>Set Up>Advanced Opt>Status
Main Menu>Topological Opt>Status
TOTAL, NTOT, NRMDF
Specifies automatic MDOF generation.
SOLUTION: Master DOF
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NTOT
Total number of master degrees of freedom to be used in the analysis, including specified (NS, see below)
master degrees of freedom. NTOT must be greater than NS if any automatic generation is to be done.
NRMDF
Rotational masters key:
0
Include all degrees of freedom in automatic master selection.
1
Exclude rotational degrees of freedom (and VOLT degrees of freedom in a piezoelectric analysis)
from automatic selection.
Command Default
Do not use any automatically generated MDOF.
Notes
Specifies automatic master degree of freedom (MDOF) generation. The limit on the number of MDOF is
equal to the maximum in-memory wavefront size (see the Basic Analysis Guide). If NS is defined as the
number of master degrees of freedom specified with the M or MGEN command, NTOT-NS additional master
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1734
TOSTAT
degrees of freedom will be automatically generated during the solution phase if TOTAL is used. NS may be
zero, i.e., all master degrees or freedom can be automatically generated. After the solution phase, generated
masters become specified masters (NS = NTOT) so that they may be listed, displayed, modified, etc. The
TOTAL command is ignored in subsequent solutions unless masters are deleted, such that NS<NTOT. If used
in SOLUTION, this command is valid only within the first load step.
During the matrix triangulation (wavefront) operation, the first NTOT degrees of freedom are temporarily
identified as masters and then are replaced as degrees of freedom with lower K/M ratios are found. Degrees
of freedom matching the user specified set (if any) are permanently identified. The wavefront builds to NTOT
and will have a minimum (and final) value of NTOT. The final set of automatic masters identified will be
those corresponding to the lowest modes of the structure.
Constrained degrees of freedom are excluded from the automatic master selection. Constraints may be
defined to prevent undesirable modes from being present (thus preventing the corresponding MDOF from
being selected). For example, if symmetry constraints are imposed, degrees of freedom producing only
symmetric modes will be selected. In-plane rotational degrees of freedom for shell elements lying in a
global plane are automatically excluded. All rotational degrees of freedom can be excluded during the
automatic selection if desired.
If automatic master selection is used in the reduced linear transient dynamic (ANTYPE,TRANS) analysis or
the reduced harmonic response (ANTYPE,HARMIC) analysis, be sure to force the selection [M] of any degrees
of freedom having nonzero displacement or force inputs. If automatic master selection is used in the super-
element generation pass (ANTYPE,SUBSTR), be sure to force the selection of connection points to nonsuper-
elements.
Automatically selected masters are shown in the solution listing (and not in preprocessing listings) as follows:
in the reduced eigenvector solution for modal (ANTYPE,MODAL).
in the reduced displacement solution for harmonic response (ANTYPE,HARMIC).
in the reduced displacement solution for linear transient dynamic (ANTYPE,TRANS).
in the matrix or load vector printout for substructures (ANTYPE,SUBSTR).
In the substructure generation pass (ANTYPE,SUBSTR), a mass matrix must be available if the TOTAL option
is to be used.
The TOTAL command is used in conjunction with a specific solver developed in an earlier version of the
ANSYS program. ANSYS automatically chooses this solver when the TOTAL command is executed. Other
solvers are not available with the TOTAL command.
We recommend that you define as many MDOF as you can based on your knowledge of the dynamic char-
acteristics of the structure and let the program choose a few additional masters with the TOTAL command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Master DOFs>Program Selected
Main Menu>Solution>Master DOFs>Program Selected
1735
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TOTAL
TOTYPE, Type
Specifies solution method for topological optimization.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Type
Choose the solution method to use:
OC
Use Optimality Criteria (OC) approach. (Default)
SCP
Use Sequential Convex Programming (SCP) approach.
Notes
The OC approach is applicable to problems with only volume as the constraint (or volume as the only
constraint). The SCP approach is applicable to all valid combinations of objectives and constraints. See
TOVAR for a description of valid combinations.
Menu Paths
Main Menu>Topological Opt>Run
TOVAR, Refname, Type, LOWER, UPPER, Boundtype
Specifies the objective and constraints for the topological optimization problem.
OPTIMIZATION: Specifications
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Refname
Reference name (8 character string), previously defined with TOCOMP and TOFREQ, or VOLUME (reserved
name; default).
Type
Valid types for this command are:
OBJ
Specifies the objective for the topological optimization problem. The Refname(s) specified must
already be defined (TOCOMP,Refname, or TOFREQ) before defining your constraint(s). LOWER,
UPPER, and Boundtype are not used if the specified Refname is the objective. To specify total
volume as the objective function, use VOLUME as the Refname. For single or multiple compliance,
see TOCOMP. For frequency formulations (single, weighted mean, reciprocal mean, or Euclidean
normal), see TOFREQ.
CON
Specifies the constraint for the topological optimization problem. Requires a previously defined ob-
jective function (TOVAR,Refname,OBJ). LOWER, UPPER, and Boundtype must be specified as listed
below.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1736
TOTYPE
DEL
Deletes the previously defined topological optimization objective or constraint named in Refname.
LOWER, UPPER, and Boundtype are not used.
LOWER
Lower bound for the constraint (Type = CON).
UPPER
Upper bound for the constraint (Type = CON). Default is no defined upper bound.
Boundtype
For Type = CON, specifies whether the specified bounds are actual values, or indicate percentages.
PERCENT
Indicates that the values specified in LOWER and UPPER should be treated as percentages of the
original value. The original value is 100%; only values less than 100 are valid. Depending on the type
of constraint, this value indicates a decrease or increase of the initial values. For example, specify 50
to reduce the volume by 50%, or specify 30 to increase the compliance by 30%.
ACTUAL
Indicates that the values specified in LOWER and UPPER should be treated as actual values.
Command Default
TOVAR,VOLUME,OBJ
Notes
You must set the objective first, then set the constraint(s).
If a single or multiple compliance function (see TOCOMP) is specified as objective function, only the VOLUME
function is allowed as a constraint.
If a single, weighted mean, reciprocal mean, or euclidean norm frequency (see TOFREQ) objective is selected,
only the VOLUME function is allowed as a constraint.
If the VOLUME function is selected as the objective function, then only a single or multiple compliance
constraint (see TOCOMP) is allowed as constraint function. In this case, a multiple single compliance constraint
definition is also possible. LOWER is not applicable if Type = OBJ or DEL, and is not used if Boundtype =
PERCENT.
UPPER is not applicable if Type = OBJ or DEL. If Boundtype = PERCENT, and RefName = VOLUME, UPPER
specifies the percentage of volume reduction. If Boundtype = PERCENT, and RefName is a single or multiple
compliance function, UPPER specifies the percentage of compliance increase.
If Boundtype = PERCENT, the valid range for UPPER is [5%, 95%].
Menu Paths
Main Menu>Topological Opt>Set Up>Advanced Opt>Topo Objective
Main Menu>Topological Opt>Set Up>Advanced Opt>TopoConstraint>By Percentage
Main Menu>Topological Opt>Set Up>Advanced Opt>TopoConstraint>By Value
Main Menu>Topological Opt>Set Up>Advanced Opt>TopoConstraint>Delete
Main Menu>Topological Opt>Set Up>Basic Opt
1737
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TOVAR
TRANS, Fname, Ext, --
Reformats File.GRPH for improved performance with plotters.
DISPLAY: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name defaults to TRAN33.
Ext
Filename extension (8 character maximum).
--
Unused field.
Notes
Reformats current Fname.GRPH data (based on color) for improved performance with pen plotters.
Menu Paths
It is part of the DISPLAY command.
TRANSFER, KCNTO, INC, NODE1, NODE2, NINC
Transfers a pattern of nodes to another coordinate system.
PREP7: Nodes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCNTO
Reference number of coordinate system where the pattern is to be transferred. Transfer occurs from the
active coordinate system.
INC
Increment all nodes in the given pattern by INC to form the transferred node pattern.
NODE1, NODE2, NINC
Transfer nodes from pattern beginning with NODE1 to NODE2 (defaults to NODE1) in steps of NINC
(defaults to 1). If NODE1 = ALL, NODE2 and NINC are ignored and the pattern is all selected nodes
[NSEL]. If NODE1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component may be substituted for NODE1 (NODE2 and NINC are ignored).
Notes
Transfers a pattern of nodes from one coordinate system to another. Coordinate systems may be translated
and rotated relative to each other. Initial pattern may be generated in any coordinate system. Coordinate
values are interpreted in the active coordinate system and are transferred directly.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1738
TRANS
A model generated in one coordinate system may be transferred to another coordinate system. The user
may define several coordinate systems (translated and rotated from each other), generate a model in one
coordinate system, and then repeatedly transfer the model to other coordinate systems. The model may be
generated in any type of coordinate system (Cartesian, cylindrical, etc.) and transferred to any other type of
coordinate system. Coordinate values (X, Y, Z, or R, , Z, or etc.) of the model being transferred are interpreted
in the active coordinate system type, regardless of how they were generated. Values are transferred directly
and are interpreted according to the type of coordinate system being transferred to. For example, transferring
from a Cartesian coordinate system to a cylindrical coordinate system (not recommended) would cause X
= 2.0 and Y = 3.0 values to be directly interpreted as R = 2.0 and = 3.0 values, respectively.
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Transfer Coord>Nodes
*TREAD, Par, Fname, Ext, --, NSKIP
Reads data from an external file into a table array parameter.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
Table array parameter name as defined by the *DIM command.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
File name has no default.
Ext
Filename extension (8 character maximum).
Extension has no default.
--
Unused field.
NSKIP
Number of comment lines at the beginning of the file being read that will be skipped during the reading.
Default = 0.
Notes
Use this command to read in a table of data from an external file into an ANSYS table array parameter. The
external file may be created using a text editor or by an external application or program. The external file
must be in tab-delimited, blank-delimited, or comma-delimited format to be used by *TREAD. The ANSYS
TABLE type array parameter must be defined before you can read in an external file. See *DIM for more in-
formation.
1739
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*TREAD
This command is not applicable to 4- or 5-D tables.
Menu Paths
Utility Menu>Parameters>Array Parameters>Read from File
TREF, TREF
Defines the reference temperature for the thermal strain calculations.
SOLUTION: Load Step Options
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
TREF
Reference temperature for thermal expansion.
Note
If the uniform temperature (TUNIF) is not specified, it is also set to this value.
Command Default
Reference temperature is 0.0 degrees.
Notes
Defines the reference temperature for the thermal strain calculations in structural analyses and explicit dy-
namic analyses. Thermal strains are given by *(T-TREF), where is the coefficient of thermal expansion
(for more on this see the Theory Reference for the Mechanical APDL and Mechanical Applications). Input the
strain via ALPX, ALPY, ALPZ (the secant or mean coefficient value), or CTEX, CTEY, CTEZ (the instantaneous
coefficient value), or the thermal strain value (THSX, THSY, THSZ). T is the element temperature. If is tem-
perature-dependent, TREF should be in the range of temperatures you define using the MPTEMP command.
Reference temperatures may also be input per material by specifying a value on the MP material property
command:
MP,REFT,MAT,C0.
Only a constant (non-temperature-dependent) value is valid. The value input on the TREF command applies
to all materials not having a specified material property definition.
To convert temperature-dependent secant coefficients of thermal expansion (SCTE) data (properties ALPX,
ALPY, ALPZ) from the definition temperature to the reference temperature defined via a TREF (or MP,REFT)
command, issue the MPAMOD command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Settings>Reference Temp
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Reference Temp
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1740
TREF
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Reference Temp
Main Menu>Solution>Define Loads>Settings>Reference Temp
Main Menu>Solution>Load Step Opts>Other>Reference Temp
Main Menu>Solution>Loading Options>Reference Temp
/TRIAD, Lab
Shows the global XYZ coordinate triad on displays.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Display triad as follows:
ORIG
Display triad at global origin (default).
OFF
Turn off triad display.
LBOT
Display triad in lower left screen corner.
RBOT
Display triad in lower right screen corner.
LTOP
Display triad in upper left screen corner.
RTOP
Display triad in upper right screen corner.
Notes
For efficiency, ANSYS 3-D graphics logic maintains a single data structure (segment), which includes the triad
as a 3-D data object. If a 3-D device is involved (/SHOW,3D), and the ANSYS graphics are not being displayed
as multi-plots, then the triad location is determined by the view settings for Window #1. A request for triad
display anywhere except for the origin could yield an improper display in windows 2 through 5. The program
displays the same segment in all windows. The view settings of each window constitute the only difference
in the display in the active windows.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Window Controls>Reset Window Options
Utility Menu>PlotCtrls>Window Controls>Window Options
1741
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/TRIAD
/TRLCY, Lab, TLEVEL, N1, N2, NINC
Specifies the level of translucency.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Apply translucency level to the items specified by the following labels:
ELEM
Elements. Use N1, N2, NINC fields for element numbers.
AREA
Solid model areas. Use N1, N2, NINC fields for area numbers.
VOLU
Solid model volumes. Use N1, N2, NINC fields for volume numbers.
ISURF
Isosurfaces (surfaces of constant stress, etc., value). Translucency varies with result value, to a max-
imum of the specified translucency level.
CM
Component group. Use N1 for component name, ignore N2 and NINC.
CURVE
Filled areas under curves of line graphs. Use N1, N2, NINC fields for curve numbers.
ZCAP
If /TYPE,WN,ZCAP is the current display type, then /TRLCY,ZCAP,TLEVEL will display the model in
window WN with the portion of the model in front of the section plane displayed at the translucency
level TLEVEL.
ON, OFF
Sets the specified translucency display on or off. All other fields are ignored.
TLEVEL
Translucency level: 0.0 (opaque) to 1.0 (transparent).
N1, N2, NINC
Used only with labels as noted above. Apply translucency level to Lab items numbered N1 to N2 (defaults
to N1) in steps of NINC (defaults to 1). If N1 is blank or ALL, apply specified translucency level to entire
selected range. If Lab is CM, use component name for N1 and ignore N2 and NINC. A value of N1 = P
allows you to graphically pick elements, areas, and volumes. You can then assign translucency levels to
the entities via the picker. The Lab and TLEVEL fields are ignored when translucency is applied by
picking.
Command Default
Zero translucency (opaque) level.
Notes
Specifies the level of translucency for various items. Issue /TRLCY,DEFA to reset the default (0) translucency
levels. This command is valid only on selected 2-D and 3-D graphics devices; see in the Basic Analysis Guide
for more information on applying translucency.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1742
/TRLCY
For 2-D devices, ANSYS displays only the visible faces of the items being displayed. The information behind
the facing planes is not displayed. Issuing the /SHRINK command will force the hardware to display inform-
ation behind the translucent items.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Translucency
TRNOPT, Method, MAXMODE, Dmpkey, MINMODE, MCout, TINTOPT
Specifies transient analysis options.
SOLUTION: Dynamic Options
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Product Restrictions
Method
Solution method for the transient analysis:
FULL
Full method (default).
REDUC
Reduced method.
MSUP
Mode superposition method.
VT
Variational Technology method.
MAXMODE
Largest mode number to be used to calculate the response (for Method = MSUP). Defaults to the highest
mode calculated in the preceding modal analysis.
Dmpkey
Damping option (for Method = REDUC):
DAMP
Include the effects of damping if present (default).
NODAMP
Ignore the effects of damping, even if present.
MINMODE
Smallest mode number to be used (for Method = MSUP). Defaults to 1.
MCout
Modal coordinates output key (valid only for the mode superposition method MSUP):
NO
No output of modal coordinates (default).
YES
Output modal coordinates to the text file jobname.MCF.
1743
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TRNOPT
TINTOPT
Time integration method for the transient analysis:
NMK or 0
Newmark algorithm (default).
HHT or 1
HHT algorithm (valid only for the full transient method).
Notes
Specifies transient analysis (ANTYPE,TRANS) options. If used in SOLUTION, this command is valid only within
the first load step. Use the TINTP command to set transient integration parameters.
The VT option using an ANSYS Mechanical HPC license is valid for either thermal or structural analyses, where
it attempts to reduce the total number of iterations. It is a FULL transient solution.
To include residual vectors in your mode-superposition transient analysis (Method=MSUP), specify RESVEC,ON.
This command is also valid in PREP7.
Product Restrictions
The VT option is only available with a HPC license.
Additional product restrictions for the TRNOPT command are shown in the table below.
Available Products Command
Option
Method
MP ME ST <> <> <> <> <> EM <> <> PP <> EME MFS FULL
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS REDUC
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS MSUP
Menu Paths
Main Menu>Preprocessor>Loads>Analysis Type>Analysis Options
Main Menu>Preprocessor>Loads>Analysis Type>New Analysis
Main Menu>Preprocessor>Loads>Analysis Type>Sol'n Controls>Transient
Main Menu>Solution>Analysis Type>Analysis Options
Main Menu>Solution>Analysis Type>New Analysis
Main Menu>Solution>Analysis Type>Sol'n Controls>Transient
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1744
TRNOPT
TRPDEL, NTRP1, NTRP2, TRPINC
Deletes particle flow or charged particle trace points.
POST1: Trace Points
MP <> <> <> <> <> <> FL EM <> <> PP <> EME <>
NTRP1, NTRP2, TRPINC
Delete points from NTRP1 to NTRP2 (defaults to NTRP1) in steps of TRPINC (defaults to 1). If NTRP1
= ALL, NTRP2 and TRPINC are ignored and all trace points are deleted. If NTRP1 = P, graphical picking
is enabled and all remaining command fields are ignored (valid only in the GUI).
Notes
Deletes particle flow or charged particle trace points defined with the TRPOIN command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>Dele Trace Pt
TRPLIS, NTRP1, NTRP2, TRPINC, Opt
Lists the particle flow or charged particle trace points.
POST1: Trace Points
MP <> <> <> <> <> <> FL EM <> <> PP <> EME <>
NTRP1, NTRP2, TRPINC
List points from NTRP1 to NTRP2 (defaults to NTRP1) in steps of TRPINC (defaults to 1). If NTRP1 =
ALL, NTRP2 and TRPINC are ignored and all trace points are listed. If NTRP1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI).
Opt
Opt = LOC lists the trace point number location (X, Y, Z). Default.
Opt = PART lists the trace point number particle settings (velocity, charge, mass).
Notes
Lists the particle flow or charged particle trace points in the active display coordinate system [DSYS]. Trace
points are defined with the TRPOIN command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>List Trace Pt
1745
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TRPLIS
TRPOIN, X, Y, Z, VX, VY, VZ, CHRG, MASS
Defines a point through which a particle flow or charged particle trace will travel.
POST1: Trace Points
MP <> <> <> <> <> <> FL EM <> <> PP <> EME <>
X, Y, Z
Coordinate location of the trace point (in the active coordinate system). If X = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI).
VX, VY, VZ
Particle velocities in the X, Y and Z directions (in the active coordinate system).
CHRG
Particle charge.
MASS
Particle mass.
Notes
Defines a point through which a particle flow or charged particle trace [PLTRAC] will travel. Multiple points
(50 maximum) may be defined which will result in multiple flow traces. Use TRPLIS to list the currently
defined trace points and TRPDEL to delete trace points.
The VX, VY, VZ, CHRG, and MASS arguments only apply to charged particles.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>Defi Trace Pt
TRTIME, TIME, SPACING, OFFSET, SIZE, LENGTH
Defines the options used for the PLTRAC (particle flow or charged particle trace) command.
POST1: Animation
MP ME ST PR PRN <> <> FL EM <> <> PP <> EME MFS
TIME
Total Trace Time (seconds) (defaults to 0, which is the full flow trace).
SPACING
Particle spacing in seconds (defaults to 0).
OFFSET
Particle offset in seconds (defaults to 0). It is used internally in the ANFLOW macro to produce an anim-
ation of particle flow in a flowing fluid or charged particle motion in an electric or magnetic field.
SIZE
Particle size (defaults to 0, which is a line).
LENGTH
Particle length fraction (defaults to .1).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1746
TRPOIN
Command Default
Full particle flow or charged particle trace.
Notes
The TRTIME command varies the type of PLTRAC display produced. Particle flow or charged particle traces
follow a particle's path in the forward and backward direction of travel. The DOF selected determines the
color of the particle trace. SPACING defines the particle spacing in seconds from adjacent particles in the
stream line. OFFSET defines the offset in seconds from the spacing set by the SPACING argument.
LENGTH defines the particle length fraction. The default value (.1), means the particle occupies 10% of the
flow region, and the other 90% is a color-coded line.
SIZE sets the radius of the particle. Use SPACING, OFFSET and LENGTH only when SIZE is nonzero (i.e.,
the particle is bigger than the line).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>General Postproc>Plot Results>Time Interval
TSHAP, Shape
Defines simple 2-D and 3-D geometric surfaces for target segment elements.
PREP7: Elements
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Shape
Specifies the geometric shapes for target segment elements TARGE169 and TARGE170.
LINE
Straight line (2-D, 3-D) (Default for 2-D)
PARA
Parabola (2-D, 3-D)
ARC
Clockwise arc (2-D)
CARC
Counterclockwise arc (2-D)
CIRC
Complete circle (2-D)
TRIA
Three-node triangle (3-D) (Default for 3-D)
TRI6
Six-node triangle (3-D)
QUAD
Four-node quadrilateral (3-D)
1747
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TSHAP
QUA8
Eight-node quadrilateral (3-D)
CYLI
Cylinder (3-D)
CONE
Cone (3-D)
SPHE
Sphere (3-D)
PILO
Pilot node (2-D, 3-D)
POINT
Point (rigid surface node) (2-D, 3-D)
Notes
Use this command to specify the target segment shapes for the rigid target surface associated with surface-
to-surface contact (TARGE169, CONTA171, CONTA172 (2-D) and TARGE170, CONTA173, CONTA174 (3-D)), 3-
D beam-to-beam contact (TARGE170 and CONTA176), and 3-D line-to-surface contact (TARGE170 and CON-
TA177). Once you issue TSHAP, all subsequent target elements generated via the direct element generation
technique will have the same shape, until you issue TSHAP again with a different Shape value.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Elements>Elem Attributes
/TSPEC, TCOLOR, TSIZE, TXTHIC, PANGLE, IANGLE
Creates annotation text attributes (GUI).
GRAPHICS: Annotation
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
TCOLOR
Text color (0
TCOLOR
15):
0
Black.
1
Red-Magenta.
2
Magenta.
3
Blue-Magenta.
4
Blue.
5
Cyan-Blue.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1748
/TSPEC
6
Cyan.
7
Green-Cyan.
8
Green.
9
Yellow-Green.
10
Yellow.
11
Orange.
12
Red.
13
Dark Gray.
14
Light Gray.
15
White.
TSIZE
Text size factor.
TXTHIC
Text thickness key:
1
normal.
2
twice as thick.
3
three times as thick.
4
four times as thick.
PANGLE
Text path angle (0.0 < angle < 360.0).
IANGLE
Text italic angle (0.0 < angle < 45.0).
Notes
Defines annotation text attributes to control certain characteristics of the text created via the /TLABEL
command. This is a command generated by the Graphical User Interface (GUI) and will appear in the log file
(Jobname.LOG) if annotation is used. This command is not intended to be typed in directly in an ANSYS
session (although it can be included in an input file for batch input or for use with the /INPUT command).
1749
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/TSPEC
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Annotation>Create Annotation
TSRES, Array
Defines an array of key times at which the time-stepping strategy changes.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> FL EM <> DY PP <> EME MFS
Array
Identifies an Nx1x1 array parameter containing the key times at which the heat transfer time-stepping
strategy changes (the time step is reset to the initial time step based on DELTIM or NSUBST settings).
The array name must be enclosed by % signs (e.g., %array%). See *DIM for more information on array
parameters.
Notes
Time values in the array parameter must be in ascending order and must not exceed the time at the end
of the load step as defined on the TIME command. The time increment between time points in the array
list must be larger than the initial time step defined on the DELTIM or NSUBST command. Time values must
also fall between the beginning and ending time values of the load step. For multiple load step problems,
you must either change the parameter values to fall between the beginning and ending time values of the
load step or reissue the command with a new array parameter. To clear the array parameter specification,
issue TSRES,ERASE. Results can be output at the requested time points if the array or time values in the array
are also specified in the OUTRES command using FREQ=%array%. Use this command to reset the time-
stepping strategy within a load step. You may need to reset the time-stepping strategy when using tabular
time-varying boundary conditions.
See Steady-State Thermal Analysis of the Thermal Analysis Guide for more information on applying boundary
conditions via tabular input. See Transient Thermal Analysis of the Thermal Analysis Guide for more information
on defining the key time array.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Preprocessor>Loads>Load Step Opts>Time/Frequenc>Time and Substps
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time - Time Step
Main Menu>Solution>Load Step Opts>Time/Frequenc>Time and Substps
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1750
TSRES
TUNIF, TEMP
Assigns a uniform temperature to all nodes.
SOLUTION: FE Body Loads
MP ME ST PR PRN DS DSS FL EM EH DY PP <> EME MFS
TEMP
Uniform temperature assigned to the nodes. If a TEMP value is not specified, the uniform temperature
is set to zero.
Command Default
Set the uniform temperature to the reference temperature (defined via the TREF command).
Notes
TUNIF is a convenient form of the more general BFUNIF command.
In a transient or nonlinear thermal analysis, the uniform temperature is used during the first iteration of a
solution as follows:
as the starting nodal temperature (except where temperatures are explicitly specified [D, DK]),
to evaluate temperature-dependent material properties.
In a structural analysis or an explicit dynamic analysis, the uniform temperature is used as the default tem-
perature for thermal strain calculations and material property evaluation (except where body load temper-
atures are specified (BF, BFE, BFK, LDREAD). In other scalar field analyses, the uniform temperature is used
for material property evaluation.
Because TUNIF (or BFUNIF,TEMP) is step-applied in the first iteration, issue a BF,ALL,TEMP,Value command
to ramp on a uniform temperature load.
When the TUNIF command is used in an explicit dynamic analysis, you cannot apply temperature loading
via the EDLOAD,,TEMP command. Furthermore, temperature loading defined by TUNIF cannot be listed or
deleted by the EDLOAD command.
The command default sets the uniform temperature to the reference temperature defined via the TREF
command only (and not the MP,REFT command).
If using the command default to set the uniform temperature (to the reference temperature set via TREF),
you can convert temperature-dependent secant coefficients of thermal expansion (SCTEs) from the definition
temperature to the uniform temperature. To do so, issue the MPAMOD command.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Define Loads>Apply>Electric>Boundary>Temperature>Uniform
Temp
Main Menu>Preprocessor>Loads>Define Loads>Apply>Magnetic>Boundary>Temperature>Uniform
Temp
Main Menu>Preprocessor>Loads>Define Loads>Apply>Structural>Temperature>Uniform Temp
1751
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TUNIF
Main Menu>Preprocessor>Loads>Define Loads>Apply>Thermal>Temperature>Uniform Temp
Main Menu>Preprocessor>Loads>Define Loads>Settings>Uniform Temp
Main Menu>Preprocessor>LS-DYNA Options>Loading Options>Uniform Temp
Main Menu>Solution>Define Loads>Apply>Electric>Boundary>Temperature>Uniform Temp
Main Menu>Solution>Define Loads>Apply>Magnetic>Boundary>Temperature>Uniform Temp
Main Menu>Solution>Define Loads>Apply>Structural>Temperature>Uniform Temp
Main Menu>Solution>Define Loads>Apply>Thermal>Temperature>Uniform Temp
Main Menu>Solution>Define Loads>Settings>Uniform Temp
Main Menu>Solution>Loading Options>Uniform Temp
TVAR, KEY
Changes time to the cumulative iteration number.
POST26: Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KEY
Time key:
0
Time is used for the variable TIME.
1
NCUMIT is used for the variable TIME.
Command Default
TIME is the variable TIME.
Notes
Changes the meaning of the time variable to the cumulative iteration number (NCUMIT) variable. Data can
be read from the file, printed, and displayed as a function of NCUMIT rather than time. All POST26 descriptions
applying to TIME then apply to NCUMIT.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Data
/TXTRE, Lab, NUM, N1, N2, NINC
Controls application of texture to selected items.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
You can apply texture according to the following labels:
ELEM
Apply texture to elements N1 through N2 in steps of NINC.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1752
TVAR
AREA
Apply texture to areas N1 through N2 in steps of NINC.
VOLU
Apply texture to volumes N1 through N2 in steps of NINC.
CM
Apply texture to the component named in N1. N2 and NINC are ignored.
ON, OFF
Sets the specified texture display on or off. All other fields are ignored.
File
If Lab = File, the command format is /TXTRE, File, Key_Index, Fname, Fext, --, Format (This
variant of the command is applicable to 2-D drivers).
Key_Index
The texture index associated with the file. If the number fifty-one (51) is used, the imported bitmap
will be used as the window's logo.
Fname
File name and directory path (248 characters maximum, including the characters needed for the
directory path). An unspecified directory path defaults to the working directory; in this case, you
can use all 248 characters for the file name.
Fext
Filename extension (8 character maximum).
--
Unused field.
Format
The file format. If Format = 0, the file is a pixmap (UNIX) or Bitmap (PC). The file cannot contain
a compressed image, and the PC file must be 8 or 24 bit BI_RGB format. If Format = 1 or JPEG,
then the file is in JPEG (Joint Photographic Experts Group) format. If Format = 2 or PNG, then
the file is in PNG (Portable Network Graphics) format.
NUM
Select the texture index number from the following list:
0
No Texturing
1
Aluminum
2
Aluminum, Brushed
3
Steel With Bumps
4
Steel, Embossed
5
Iron
6
Steel, Pattern
1753
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/TXTRE
7
Steel, Riveted
8
Steel, Scratched
9
Tin
10
Metal
11
Steel, Etched
12
Metal, Hot
13
Iron, Grainy
14
Metal, Rusty
15
Brick
16
Block
17
Wood
18
Wood, Light
19
Wood, Walnut
20
Plastic, Hard Blue
21
Plastic, Light Blue
22
Plastic, Hard Red
31
Gold
32
Brass
33
Silver
34
Plastic, Black
35
Plastic, Ivory
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1754
/TXTRE
36
Plastic, Blue
37
Plastic, Red
38
Plastic, Yellow
39
Plastic, Green
40
Plastic, Brown
N1, N2, NINC
Apply texture to Lab items numbered N1 through N2 in steps of NINC (defaults to 1). If Lab = CM,
then N1 is used to for the component name and N2 and NINC are ignored. If Lab = ELEM, AREA, or
VOLU and N1 = blank or ALL, then the specified texture will be applied to all entities of type Lab. If N1
= P, then graphical picking is enabled.
Command Default
No texture (/TXTRE,DEFA)
Notes
This command is available for 3-D Open GL devices. 2-D devices are supported only for the Lab = File
variation of the command, allowing imported bitmaps to be used for texturing and annotation. Textures
can affect the speed of many of your display operations. You can increase the speed by temporarily turning
the textures off (Utility Menu> PlotCtrls> Style> Texturing(3D)> Display Texturing). This menu selection
toggles your textures on and off. When textures are toggled off, all of the texture information is retained
and reapplied when texturing is toggled back on.
For some displays, the texture will appear distorted because of a technique used to enhance 3-D displays
(/DV3D,TRIS,1). Disabling this function (/DV3D,TRIS,0) will improve the quality of some texture displays.
Disabling the TRIS option of the /DV3D command will slow down 3-D displays significantly. Be sure to reapply
the TRIS option after you obtain a satisfactory output.
Specifying /TXTRE,DEFA removes all texturing.
Menu Paths
Utility Menu>PlotCtrls>Style>Texturing(3D)
/TYPE, WN, Type
Defines the type of display.
GRAPHICS: Style
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
1755
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/TYPE
Type
Display type. Defaults to ZBUF for raster mode displays or BASIC for vector mode displays:
BASIC or 0
Basic display (no hidden or section operations).
SECT or 1
Section display (plane view). Use the /CPLANE command to define the cutting plane.
HIDC or 2
Centroid hidden display (based on item centroid sort).
HIDD or 3
Face hidden display (based on face centroid sort).
HIDP or 4
Precise hidden display (like HIDD but with more precise checking). Because all facets are sorted, this
mode can be extremely slow, especially for large models.
CAP or 5
Capped hidden display (same as combined SECT and HIDD with model in front of section plane re-
moved).
ZBUF or 6
Z-buffered display (like HIDD but using software Z-buffering).
ZCAP or 7
Capped Z-buffered display (same as combined SECT and ZBUF with model in front of section plane
removed).
ZQSL or 8
QSLICE Z-buffered display (same as SECT but the edge lines of the remaining 3-D model are shown).
HQSL or 9
QSLICE precise hidden display (like ZQSL but using precise hidden).
Command Default
ZBUF for raster mode displays; BASIC for vector mode displays.
Notes
Defines the type of display, such as section display or hidden-line display. Use the /DEVICE command to
specify either raster or vector mode.
The SECT, CAP, ZCAP, ZQSL, and HQSL options produce section displays. The section or "cutting" plane is
specified on the /CPLANE command as either normal to the viewing vector at the focus point (default), or
as the working plane.
When you use PowerGraphics, the section display options (Section, Slice, and Capped) use different averaging
techniques for the interior and exterior results. Because of the different averaging schemes, anomalies may
appear at the transition areas. In many cases, the automatically computed MIN and MAX values will differ
from the full range of interior values. You can lessen the effect of these anomalies by issuing AVRES,,FULL
(Main Menu> General Post Proc> Options for Outp). This command sets your legend's automatic contour
interval range according to the minimum and maximum results found throughout the entire model.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1756
/TYPE
With PowerGraphics active (/GRAPHICS,POWER), the averaging scheme for surface data with interior element
data included (AVRES,,FULL) and multiple facets per edge (/EFACET,2 or /EFACET,4) will yield differing
minimum and maximum contour values depending on the Z-Buffering options (/TYPE,,6 or /TYPE,,7). When
the Section data is not included in the averaging schemes (/TYPE,,7), the resulting absolute value for the
midside node is significantly smaller.
The HIDC, HIDD, HIDP, ZBUF, ZQSL, and HQSL options produce displays with "hidden" lines removed. Hidden
lines are lines obscured from view by another element, area, etc. The choice of non-Z-buffered hidden-line
procedure types is available only for raster mode [/DEVICE] displays. For vector mode displays, all non-Z-
buffered "hidden-line" options use the same procedure (which is slightly different from the raster procedures).
Both geometry and postprocessing displays may be of the hidden-line type. Interior stress contour lines
within solid elements can also be removed as hidden lines, leaving only the stress contour lines and element
outlines on the visible surfaces. Midside nodes of elements are ignored on postprocessing displays. Overlap-
ping elements will not be displayed.
The ZBUF, ZCAP, and ZQSL options use a specific hidden-line technique called software Z-buffering. This
technique allows a more accurate display of overlapping surfaces (common when using Boolean operations
or /ESHAPE on element displays), and allows smooth shaded displays on all interactive graphics displays.
Z-buffered displays can be performed faster than HIDP and CAP type displays for large models. See also the
/LIGHT, /SHADE, and /GFILE commands for additional options when Z-buffering is used.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Hidden-Line Options
TYPE, ITYPE
Sets the element type attribute pointer.
PREP7: Meshing
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITYPE
Assign this type number to the elements (defaults to 1).
Command Default
ITYPE = 1.
Notes
Activates an element type number to be assigned to subsequently defined elements. This number refers to
the element type number (ITYPE) defined with the ET command. Type numbers may be displayed [/PNUM].
In some cases, ANSYS can proceed with a meshing operation even when no logical element type has been
assigned via TYPE or XATT,,,TYPE. For more information, see the discussion on setting element attributes
in Meshing Your Solid Model in the Modeling and Meshing Guide.
1757
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TYPE
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>Default Attribs
Main Menu>Preprocessor>Modeling>Create>Elements>Elem Attributes
Main Menu>Preprocessor>Modeling>Operate>Extrude>Elem Ext Opts
TZAMESH, Tvolu, SIZE, NDIV
Meshes the areas of a volume to create Trefftz nodes.
PREP7: Trefftz Domain
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Tvolu
Component name for the Trefftz volume. The component name must be enclosed in single quotes in
the TZAMESH command line.
SIZE
Edge length between Trefftz nodes on surface boundaries (i.e. lines). The areas of the component volume
will be meshed with this specification to create the Trefftz nodes. If SIZE is zero (or blank), use NDIV.
NDIV
Number of element divisions along the surface boundary lines. Defaults to 2.
Notes
The command macro is used to create Trefftz nodes by meshing the surface areas of the selected volumes.
The nodes of the meshed surface areas are grouped into the node component TZ_NOD. There is no solid
model associativity with the Trefftz volume and the Trefftz nodes. The Trefftz nodes (component TZ_NOD)
are used to create the Trefftz substructure (TZEGEN command).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Trefftz Domain>Mesh TZ Geometry
TZDELE
Deletes the Trefftz superelement, associated constraint equations and all supporting Trefftz files.
PREP7: Trefftz Domain
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
Deletes the Trefftz superelement, associated constraint equations and all supporting Trefftz files previously
generated with the TZEGEN command.
You should delete the Trefftz superelement if you are going to create new Trefftz nodes and generate a
new Trefftz superelement.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1758
TZAMESH
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Trefftz Domain>Superelement>Delete TZ
TZEGEN
Generates a Trefftz domain substructure and defines a Trefftz superelement for use in electrostatic ana-
lysis.
PREP7: Trefftz Domain
MP <> <> <> <> <> <> <> EM <> <> PP <> EME <>
Notes
The command generates a Trefftz domain substructure from the Trefftz node component (TZ_NOD) and the
flagged infinite surfaces of the exterior finite element domain [SF command with the INF surface load label].
The Trefftz nodes should be uniformly scattered between the modeled components and the exterior of the
finite element mesh. The nodes must not be attached to any underlying electrostatic finite elements. The
nodes may be created automatically from a solid model volume using the TZEGEN command macro.
The substructure named Jobname.sub is created and automatically brought into the problem as a super-
element (MATRIX50) using the next available element number. Also created is a set of constraint equations
relating the Trefftz nodes to the surface nodes of the flagged exterior finite element domain. The TZEGEN
command creates temporary files during the substructure generation. These include the following:
Jobname.TZN - Trefftz nodes
Jobname.TZE - Trefftz surface facets on the FE boundary
Jobname.TZX - Surface nodes of the FE boundary
Jobname.TZM - Trefftz material
The TZEGEN macro is only valid for 3-D electrostatics analysis. In addition, the bounding surface of the finite
element domain must contain no symmetry planes.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Trefftz Domain>Superelement>Generate TZ
1759
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
TZEGEN
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1760
U Commands
/UCMD, Cmd, SRNUM
Assigns a user-defined command name.
APDL: Abbreviations
MP ME ST <> <> <> <> <> EM EH <> PP <> EME MFS
Argument Descriptions
Cmd
User-defined command name. Only the first four characters are significant. Must not conflict with any
ANSYS command name or any user "unknown command" macro name.
SRNUM
User subroutine number (1 to 10) programmed for this command. For example, the command
/UCMD,MYCMD,3 will execute subroutine USER03 whenever the command MYCMD is entered. Use a
blank command name to disassociate SRNUM from its command. For example, /UCMD,,3 removes MYCMD
as a command.
Notes
Assigns a user-defined command name to a user-programmable (system-dependent) subroutine. This feature
allows user-defined commands to be programmed into the ANSYS program. Once programmed, this command
can be input to the program like other commands, and can also be included in the ANSYS start-up file. See
*ULIB for another way of defining user commands.
Up to 10 subroutines are available for user-defined commands (USER01 to USER10). Users must have system
permission, system access, and knowledge to write, compile, and link the appropriate subprocessors into
the ANSYS program at the site where it is to be run. All routines should be written in FORTRAN. For more
information on FORTRAN compilers please refer to either the ANSYS, Inc. Windows Installation Guide or the
ANSYS, Inc. UNIX/Linux Installation Guide for details specific to your platform or operating system. The USER01
routine is commented and should be listed from the distribution media (system dependent) for more details.
Issue /UCMD,STAT to list all user-defined command names. Since a user-programmed command is a non-
standard use of the program, the verification of any ANSYS run incorporating these commands is entirely
up to the user. In any contact with ANSYS customer support regarding the performance of a custom version
of the ANSYS program, you should explicitly state that a user programmable feature has been used. See the
Advanced Analysis Techniques Guide for a general description of user-programmable features and Guide to
ANSYS User Programmable Features for a detailed description of these features.
This command is valid only at the Begin Level.
Menu Paths
This command cannot be accessed from a menu.
1761
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/UDOC, WIND, Class, Key,
Determines position and content for the multi-legend options.
GRAPHICS: Labeling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WIND
The window number to which the command applies. (defaults to 1)
Class
The type (and relative importance) of legend item being displayed:
CNTR
Contour legend. This legend item is controlled separately from the other legend items (see note
below).
DATE
The items in the DATE class include the date and time, or the ANSYS graphical logo (/PLOPTS,LOGO,1).
This item is shown by default in all plots.
GWIN
The items in the GWIN class include the entity acronyms that appear in the legend of a multiplot of
entities (Nodes, Elements, Keypoints, Lines, Areas, Volumes). GWIN items are shown by default for
all GPLOT displays.
TYPE
Items in the TYPE class include the plot type (e.g. ELEMENTS, MATERIALS, NODAL SOLUTIONS, etc.).
TYPE items are shown by default in all plots.
TYP2
Items in the TYP2 class include supplementary type information, such as DMAX and SMAX for nodal
solutions. TYP2 items are shown by default in all plots.
INUM
Items in the INUM class include the number labels generated by the /PNUM command. This class is
displayed by default in all plots that contain /PNUM information.
BCDC
The items in the BCDC class include labels created by the /PBC command. This class is shown by
default in all plots which contain /PBC information.
VECT
Items in the VECT class include labels created by the PLVECT command. This class is shown by default
for all PLVECT plots.
SURF
The items in the SURF class include labels from the /PSF legend. This class is shown by default on
all plots of surface boundary conditions.
BODY
Items from the BODY class include labels from the /PBF legend. This class is shown by default in all
plots of body forces.
PSTA
Items from the PSTA class include stress scaling statistics, such as the /SSCALE setting. This class is
not shown as the default for any type of plot, and must be specifically referenced to display the in-
cluded data.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1762
/UDOC
VIEW
The items in the VIEW class include view statistics. This class is not shown as the default for any type
of plot, and must be specifically referenced to display the included data.
MISC
The items in the MISC class include supplementary labels like /EXPANDED and Stress Section Cross
Section. This class is not shown as the default for any type of plot, and must be specifically referenced
to display the included data.
KEY
Switch:
Do not display the legend. OFF or 0 --
If the value for Class is CNTR, these are the four acceptable values
for the contour legend position.
LEFT, RIGHT, TOP or BOTTOM --
If any value other than CNTR is used for Class, these are the two
acceptable values for the text data.
LEFT or RIGHT --
Notes
The legend classes conform to the controls specified in the window options panel (PlotCtrls> Window
Controls> Window Options). In many instances, the legend controls specified with the /PLOPTS command
will take precedence and override /UDOC specifications. For instance:
/PLOPTS,LEG1,OFF will disable the TYPE, TYP2, INUM, and MISC classes, regardless of the /UDOC settings.
/PLOPTS,LEG2,OFF will disable the VIEW class, regardless of the /UDOC settings.
/PLOPTS,LEG3,OFF will disable the PSTA class, regardless of the /UDOC settings.
All items in a class are listed with the same X coordinate (except for contours). The contents of the text
classes are dumped onto the display window from top to bottom, in order of class importance.
The font specification for text items that are included in the user-specified legends are controlled with the
/DEVICE command (PlotCtrls> Font Controls> Anno/Graph Font).
The floating point values for the data presented in the legend(s) are controlled by the /GFORMAT command.
Menu Paths
Utility Menu>PlotCtrls>Style>MultiLegend Options>Contour Legend
Utility Menu>PlotCtrls>Style>MultiLegend Options>Text Legend
/UI, Func, Type, Format, Screen, Color, Krev, Orient, Compress, Quality
Activates specified GUI dialog boxes.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Func
Label identifying the dialog box to be activated:
1763
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/UI
HELP
Activates the online help system. Valid only in non-UI graphics mode (/MENU,GRPH).
VIEW
Activates the Pan, Zoom, Rotate dialog box
WPSE
Activates the Working Plane Settings dialog box.
WPVI
Activates the Offset Working Plane dialog box.
RESULT
Activates the Query Picking Menu for reviewing results.
QUERY
Activates the Query Picked Entities (preprocess) dialog box.
COPY
Activates the Hard Copy dialog box.
ANNO
Activates the 2D Annotation dialog box.
AN3D
Activates the 3D Annotation dialog box.
SELECT
Activates the Select Entities dialog box.
NSEL
Activates a picking menu to select nodes.
ESEL
Activates a picking menu to select elements.
KSEL
Activates a picking menu to select keypoints.
LSEL
Activates a picking menu to select lines.
ASEL
Activates a picking menu to select areas.
VSEL
Activates a picking menu to select volumes.
REFRESH
Refreshes the graphics window (non-UI mode only).
COLL
Controls the collapse of the ANSYS Main Menu when a FINISH command is issued. See Type below
for a discussion of the arguments.
Type
Label identifying the type of select operation. Valid only for the following Func labels; NSEL, ESEL, KSEL,
LSEL, ASEL, and VSEL:
S
Select a new set.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1764
/UI
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Label identifying the type of results data to be queried. Valid only for Func = RESULT:
NODE
Nodal solution data (h-elements only).
ELEMENT
Element solution data.
GRID
Subgrid solution data (p-elements only).
Label specifying the behavior of the ANSYS Main Menu after a FINISH command is issued. Note that
this does not affect user interaction with the Main Menu. Valid only for Func = COLL:
YES, 1 or blank
Allows the Main Menu to collapse after FINISH command.
NO or 0
Prevents Main Menu collapse after FINISH command.
If Func = COPY, and Type = SAVE, command format is /UI,Func,Type,Format,Screen,Color,Krev,Ori-
ent,Compress,Quality. The remaining fields (after Type) identify the portion of the screen, the type of
file, and the method of display that the portion of the screen is to be saved as.
Format
PSCR
Encapsulated Postscript File Format.
TIFF
Tagged Image File Format.
EPSI
Encapsulated Postscript with TIFF preview.
BMP
(PC only) Bitmap (Windows) file format.
WMF
(PC only) Windows Metafile format.
EMF
(PC only) Enhanced Metafile format.
JPEG
JPEG (Joint Photographic Experts Group) file format.
Screen
FULL
Saves the entire screen in the specified format.
1765
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/UI
GRAPH
Saves only the ANSYS Graphic window.
Color
MONO
A two color (black and white) file is saved.
GRAY
The specified file format is saved in gray scale.
COLOR
The file is saved at the specified color depth.
Krev
NORM
Saves file as shown on the screen.
REVERSE
Saves file with the background color reversed.
Orient
LANDSCAPE
Saves file in landscape mode.
PORTRAIT
Saves file in portrait mode.
Compress
YES
Compresses TIFF files and EPS files with TIFF preview (default).
NO
Saves files with no compression.
Quality
1,2,,,100
JPEG quality index, with 100 being the maximum quality level.
Notes
Allows you to activate specified GUI dialog boxes directly in either GUI or non-GUI mode.
The /UI command itself is valid in any processor, however certain dialog boxes are accessible only in a par-
ticular processor (e.g., /UI,RESULT,... is valid only in the General Postprocessor).
ANSYS JPEG software is based in part on the work of the Independent JPEG Group, Copyright 1998, Thomas
G. Lane.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>Manual Rezoning>Create Remesh Zone(s)>Create Rezone Area
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1766
/UI
UIMP, MAT, Lab1, Lab2, Lab3, VAL1, VAL2, VAL3
Defines constant material properties (GUI).
PREP7: Materials
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT
Material number.
Lab1, Lab2, Lab3
Material property labels (see the MP command for valid labels).
VAL1, VAL2, VAL3
Values corresponding to three labels.
Notes
Defines constant material properties. This is a command generated by the Graphical User Interface (GUI)
and will appear in the log file (Jobname.LOG) if material properties are specified using the Material Prop-
erties dialog box. This command is not intended to be typed in directly in an ANSYS session (although it
can be included in an input file for batch input or for use with the /INPUT command).
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Change Mat Props>Material Models
Main Menu>Preprocessor>Material Props>Material Models
Main Menu>Solution>Load Step Opts>Other>Change Mat Props>Material Models
/UIS, Label, VALUE
Controls the GUI behavior.
SESSION: Run Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Behavior control key:
BORD
Controls the functionality of the mouse buttons for dynamic viewing mode only. When Label =
BORD, the three values that follow control the functionality of the LEFT, MIDDLE and RIGHT buttons,
respectively (see below).
MSGPOP
Controls which messages from the ANSYS error message subroutine are displayed in a message
dialog box.
REPLOT
Controls whether or not an automatic replot occurs after functions affecting the model are executed.
ABORT
Controls whether or not ANSYS displays dialog boxes to show the status of an operation in progress
and to cancel that operation.
1767
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/UIS
DYNA
Controls whether the dynamic mode preview is a bounding box or the edge outline of the model.
This label only applies to 2-D display devices (i.e., /SHOW,XII or /SHOW,WIN32). This "model edge
outline" mode is only supported in PowerGraphics [/GRAPHICS,POWER] and is intended for element,
line, results, area, or volume displays.
PICK
Controls how graphical entities are highlighted from within the ANSYS Select menu.
POWER
Controls whether or not PowerGraphics is active when the GUI is initiated. The ANSYS program default
status is PowerGraphics ON; this command is used (placed in the start.ans file) when full
graphics is desired on start up.
DPRO
Controls whether or not the ANSYS input window displays a dynamic prompt. The dynamic prompt
shows the correct command syntax for the command, as you are entering it.
UNDO
Controls whether or not the session editor includes nonessential commands or comments in the file
it creates. You can use this option to include comments and other materials in the session editor
file.
LEGE
Controls whether or not the multi-legend is activated when you start the GUI. The multi-legend allows
you specify the location of your legend items in each of the five graphics windows. You can place
this option in your start.ans file and have the GUI start with the legend items in a pre-specified
location.
PBAK
Controls whether or not the background shading is activated when you start the GUI. Background
shading. You can place this option in your start.ans file and control whether or not background
shading is activated.
ZPIC
Controls the sorting order for entities that are coincident (directly in front of or behind each other)
to a picked spot on your model. When you pick a spot on your model that could indicate two or
more entities, a message warns you of this condition, and a list of the coincident entities can be
generated. The VALUE term (below) will determine the sort order.
HPOP
Controls the prioritization of your GUI windows when the contents are ported to a plot or print file
(/UI,COPY,SAVE). OpenGL (3D) graphics devices require that the ANSYS Graphics Screen contents be
set in front of all overlying windows in order to port them to a printer or a file. This operation can
sometimes conflict with /NOERASE settings. See the VALUE term (below) to determine the available
control options.
VALUE
Values controlling behavior if Label = BORD:
(These values control the operation according to syntax : /UIS,BORD,LEFT,MIDDLE,RIGHT)
1
PAN, controls dynamic translations.
2
ZOOM, controls zoom, and dynamic rotation about the view vector.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1768
/UIS
3
ROTATE, controls dynamic rotation about the screen X and Y axes.
Note
You can designate any value for any button, or designate the same value for all three
buttons. If no value is specified, default is LEFT = PAN, MIDDLE = ZOOM and RIGHT =
ROTATE.
Values controlling behavior if Label = MSGPOP:
0
All messages displayed.
1
Only notes, warnings, and errors displayed.
2
Only warnings and errors displayed (default).
3
Only errors displayed.
Values controlling behavior if Label = REPLOT:
0
No automatic replot.
1
Automatic replot (default).
Values controlling behavior if Label = ABORT:
ON
Display status and cancellation dialog boxes (default).
OFF
Do not display status and cancellation dialog boxes.
1
Same as ON.
0
Same as OFF.
Values controlling behavior if Label = DYNA:
0
Use model edge outline when possible (default).
1
Use bounding box preview.
Values controlling behavior if Label = PICK:
0
Picked keypoints and nodes are enclosed by a square. Picked lines are overlaid by a thicker line.
Picked areas, volumes, and elements (non-point/non-line) are redrawn with highlighting colors.
1769
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/UIS
However, if the pick is a box, circle, or polygon pick, the highlighting for all entitles consists only of
a square placed around the entity's centroid.
1
Picked entities are not highlighted.
2
5.1 highlighting (that is, no XOR).
3
Picked entities are highlighted as in VALUE = 0, except that, for a box, circle, or polygon pick, the
picked areas, volumes, and elements (non-point/non-line) are redrawn with highlighting colors. This
technique is slower than the VALUE = 0 technique.
Values controlling behavior if Label = POWER:
0
Start GUI in Full Graphics mode.
1
Start GUI in PowerGraphics mode (default).
Values controlling behavior if Label = DPRO:
0 or OFF
Do not display the dynamic prompt.
1 or ON
Display the dynamic prompt (default).
Values controlling behavior if Label = UNDO:
0 or None
Do not suppress any commands (default).
1 or Comment
Write the nonessential commands to the session editor file as comments (with a ! at the beginning).
2 or Remove
Do not write nonessential commands or comments.
Values controlling behavior if Label = LEGE:
0 or OFF
Start GUI with the enhanced legend off (default).
1 or ON
Start GUI with the enhanced legend capability activated.
Values controlling behavior if Label = PBAK:
0 or OFF
Start the GUI with the no background shading (default).
1 or ON
Start the GUI with background shading activated.
Values controlling behavior if Label = HPOP:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1770
/UIS
0 or OFF
No rewrite operations are performed to compensate for items that obscure or overlay the graphics
window (default).
1 or ON
The Graphics screen contents are replotted to ensure that they are situated in front of all other
windows. If /NOERASE is detected, this operation is suppressed.
Notes
Controls certain features of the Graphical User Interface (GUI), including whether the ANSYS program displays
dialog boxes to show you the status of an operation (such as meshing or solution) in progress and to enable
you to cancel that operation. Issue /UIS,STAT for current status. Issue /UIS,DEFA to reset default values for
all labels. Issue /UIS,Label,STAT and /UIS,Label,DEFA for status and to reset a specific Label item.
A /UIS,HPOP,1 command employs a fast redraw method which does not allow entering the legend logic for
a /PLOPTS,INFO,1 or /PLOPTS,INFO,2 command. However, the legend is redrawn for /PLOPTS,INFO,3 because
that command also allows a fast redraw.
This command is valid in any processor.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Utility Menu>MenuCtrls>Message Controls
Utility Menu>PlotCtrls>Device Options
*ULIB, Fname, Ext, --
Identifies a macro library file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
Ext
Filename extension (8 character maximum).
--
Unused field.
Command Default
No macro library file.
1771
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*ULIB
Notes
Identifies a macro library file for the *USE command. A library of macros allows blocks of often used ANSYS
commands to be stacked and executed from a single file. The macro blocks must be enclosed within block
identifier and terminator lines as shown in the example below. If you want to add comment lines to a macro
block, you may place them anywhere within the macro block. (This includes placing them directly on the
lines where the macro block identifier and the macro block terminator appear, as shown in the example.)
Do not place comment lines (or any other lines) outside of a macro block.
ABC! Any valid alphanumeric name (8 characters maximum)
! identifying this data block
---! ANSYS data input commands
---
---
/EOF! Terminator for this data block
XYZ! Identify another data block (if desired)
---! ANSYS data input commands
---
---
/EOF! Terminator for this data block
(etc.)
The name of the macro library file is identified for reading on the *ULIB command. The name of the macro
block is identified on the *USE command. The commands within the macro block are copied to a temporary
file (of the macro block name) during the *USE operation and executed as if a macro file of that name had
been created by the user. The temporary file is deleted after it has been used. Macro block names should
be acceptable filenames (system dependent) and should not match user created macro file names, since
the user macro file will be used first (if it exists) before the library file is searched. Macro blocks may be
stacked in any order. Branching [*GO or *IF] external to the macro block is not allowed.
This command is valid in any processor.
Menu Paths
Utility Menu>Macro>Execute Data Block
UNDELETE, Option, Nstart, Nend
Removes results sets from the group of sets selected for editing.
AUX3: Results Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Option
Specifies which sets are to be removed from the selected sets.
SET
Specifies one or more particular sets in the results file that are to be removed from the group of sets
selected for deletion.
ALL
Removes all selected sets that are currently selected for deletion.
Nstart
The first set to be removed from the set selected for deletion.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1772
UNDELETE
Nend
The final set to be removed from the set selected for deletion. This field is used only if operating on
more than one sequential set.
Notes
Use this command if you have previously marked a set for deletion (with the DELETE command) and now
wish to keep that set instead of deleting it.
Menu Paths
This command cannot be accessed from a menu.
UNDO, Kywrd
Allows the user to modify or save commands issued since the last RESUME or SAVE command.
DATABASE: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Kywrd
NEW
Create an editable GUI window that allows the user to alter the commands issued since the most
recent SAVE or RESUME operations (GUI only).
Notes
The UNDO command brings up the session editor, a text window that displays all of the program operations
since the last SAVE or RESUME command. You can modify command parameters, delete whole sections of
text and even save a portion of the command string to a separate file. The file is named jobname000.cmds,
with each subsequent save operation incrementing the filename by one digit.
Note
The session editor file can be changed only by the session editor. If you rename your database
file outside of ANSYS and then resume that database, the session editor will display the old file-
name.
For more information on the session editor, see Using the Session Editor in the Operations Guide.
Menu Paths
Main Menu>Preprocessor>Session Editor
1773
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
UNDO
/UNITS, Label, LENFACT, MASSFACT, TIMEFACT, TEMPFACT, TOFFSET, CHARGEFACT, FORCEFACT, HEATFACT
Annotates the database with the system of units used.
DATABASE: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Label
Label to denote the system of units used in this job:
USER
User-defined system (default).
SI
International system (m, kg, s, K).
MKS
MKS system (m, kg, s, C).
uMKS
MKS system (m, kg, s, C).
CGS
CGS system (cm, g, s, C).
MPA
MPA system (mm, Mg, s, C).
BFT
U. S. Customary system using feet (ft, slug, s, F).
BIN
U. S. Customary system using inches (in, lbf*s
2
/in, s, F).
If Label = USER, the remaining fields on this command may be used to enter conversion factors that are
appropriate for the user-defined system of units.
LENFACT
Conversion factor to meter (m). Default = 1.
MASSFACT
Conversion factor to kilogram (kg). Default = 1.
TIMEFACT
Conversion factor to second (s). Default = 1.
TEMPFACT
Conversion factor to Kelvin (K). Default = 1.
TOFFSET
Temperature offset from absolute zero in degrees Kelvin. Default = 0.
CHARGEFACT
Conversion factor to Coulomb. Default = 1.
FORCEFACT
Conversion factor to Newton (N). Default = 1.
HEATFACT
Conversion factor to Joule (J). Default = 1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1774
/UNITS
Command Default
User-defined units.
Notes
Allows the user to set a marker in the database indicating the system of units used. The setting may be re-
viewed with the /STATUS command at the Begin level. The units label and conversion factors on this com-
mand are for user convenience only and have no effect on the analysis or data. That is, /UNITS will not
convert database items from one system to another (e.g., from U. S. Customary to SI, etc.). The units setting
will be written to the file of IGES data [IGESOUT or CDWRITE], which can then be read by many programs
that read IGES files. The user must still use consistent units for the results to be valid.
If you choose the MKS system of units, the EPZRO option for the EMUNIT command is set to 8.85 e-12 F/m.
(EPZRO specifies alternate free-space permittivity.)
For micro-electromechanical systems (MEMS), where dimensions are on the order of microns, see the con-
version factors in System of Units in the Coupled-Field Analysis Guide.
If you use the ANSYS ADAMS Interface to export model information to the ADAMS program, the /UNITS
command is required to ensure the correct transfer of data between ANSYS and ADAMS. You may choose
a predefined unit system label (Label = SI, CGS, etc.) or you can select the user-defined system option
(Label = USER) and input the appropriate conversion factors (LENFACT, MASSFACT, TIMEFACT, and
FORCEFACT). The conversion factors will be written to the ADAMS input file Jobname.MNF in order to
correctly generate the load. For more information, see Export to ADAMS in the Advanced Analysis Techniques
Guide.
All differences between the base solution units used by the ANSYS and CFX solvers will be noted in the
ANSYS output file. Unit conversions are automatically applied to all loads transferred unless Label = USER.
Unit conversions are not applied to any of the loads transferred between the ANSYS and CFX solvers if they
use a user-defined unit system.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
UNPAUSE
Restores use of a temporarily released product license.
SESSION: Run Controls
MP ME ST PR PRN DS DSS FL EM EH DY PP VT EME MFS
Notes
The UNPAUSE command restores use of a temporarily released (paused) product license. The command is
valid only after a previously issued PAUSE command.
1775
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
UNPAUSE
When use of the product license is paused via the PAUSE command, no other operation (other than SAVE
or /EXIT) is possible until you issue the UNPAUSE command.
For more information, see the documentation for the PAUSE command and the ANSYS, Inc. Licensing Guide.
Menu Paths
This command cannot be accessed from a menu.
UPCOORD, FACTOR, Key
Modifies the coordinates of the active set of nodes, based on the current displacements.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
FACTOR
Scale factor for displacements being added to nodal coordinates. If FACTOR = 1.0, the full displacement
value will be added to each node, 0.5, half the displacement value will be added, etc. If FACTOR = -1,
the full displacement value will be subtracted from each node, etc.
Key
Key for zeroing displacements in the database:
OFF
Do not zero the displacements (default).
ON
Zero the displacements.
Notes
The UPCOORD command uses displacements stored in the ANSYS database, and not those contained
within the results file, Jobname.RST. Nodal coordinates are updated each time the command is issued.
After updating, both the nodal displacements and rotations are set to zero if Key = ON.
For structural solutions with an updated mesh, unless the coefficient matrix is otherwise reformed (e.g., a
new analysis or NLGEOM,ON) it should first be reformed by issuing a KUSE,-1 command.
For a multiphysics simulation where a CFD or electromagnetic field is being coupled to a structure undergoing
large displacements, all (or a portion) of the surrounding field mesh may take part in the structural solution
to "move" with the displacing structure. You can use the UPCOORD command with a suitable FACTOR to
update the coordinates of the nodes using the newly computed displacements. The mesh will now conform
with the displaced structure for subsequent field solutions. However, the mesh should always be restored
to its original location by using an UPCOORD,FACTOR command before performing any subsequent struc-
tural solutions. This is true for both repeated linear solutions, and for nonlinear restarts. (All saved displace-
ments are relative to the original mesh location.)
This command is not intended to replace either the large displacement or birth and death logic.
This command is also valid in PREP7.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1776
UPCOORD
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>Updt Node Coord
Main Menu>Solution>Load Step Opts>Other>Updt Node Coord
UPGEOM, FACTOR, LSTEP, SBSTEP, Fname, Ext, --
Adds displacements from a previous analysis and updates the geometry of the finite element model to
the deformed configuration.
PREP7: Elements
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
FACTOR
Multiplier for displacements being added to coordinates. The value 1.0 will add the full value of the
displacements to the geometry of the finite element model. Defaults to 1.0.
LSTEP
Load step number of data to be imported. Defaults to the last load step.
SBSTEP
Substep number of data to be imported. Defaults to the last substep.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The field must be input (no default).
Ext
Filename extension (8 character maximum).
The extension must be an RST extension.
--
Unused field.
Notes
This command updates the geometry of the finite element model according to the displacement results of
the previous analysis and creates a revised geometry at the deformed configuration. This command works
on all nodes (default) or on a selected set of nodes. If this command is issued repeatedly, it creates a revised
geometry of the finite element model in a cumulative fashion, i.e., it adds displacement results on the pre-
viously generated deformed geometry. The solid model geometry is not updated by this command.
Menu Paths
Main Menu>Preprocessor>Modeling>Update Geom
1777
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
UPGEOM
*USE, Name, ARG1, ARG2, ARG3, ARG4, ARG5, ARG6, ARG7, ARG8, ARG9, AR10, AR11, AR12, AR13, AR14, AG15,
AR16, AR17, AR18
Executes a macro file.
APDL: Macro Files
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Name
Name (32 characters maximum, beginning with a letter) identifying the macro file or a macro block on
a macro library file.
ARG1, ARG2, ARG3, . . . , AR18
Values passed into the file or block where the parameters ARG1 through ARG9 and AR10 through AR18
are referenced. Values may be numbers, alphanumeric character strings (up to 8 characters enclosed in
single quotes), parameters (numeric or character) or parametric expressions. See below for additional
details.
Notes
Causes execution of a macro file called Name, or, if not found, a macro block "Name" on the macro library
file [*ULIB]. Argument values (numeric or character) are passed into the file or block and substituted for
local parameters ARG1, ARG2, ..., AR18. The file Name may also be executed as an "unknown command" (i.e.,
without the *USE command name) as described below.
A macro is a sequence of ANSYS commands (as many as needed) recorded in a file or in a macro block in
a library file (specified with the *ULIB command). The file or block is typically executed with the *USE
command. In addition to command, numerical and alphanumeric data, the macro may include parameters
which will be assigned numerical or alphanumerical character values when the macro is used. Use of the
macro may be repeated (within a do-loop, for example) with the parameters incremented. A macro is defined
within a run by "enclosing" a sequence of data input commands between a *CREATE and a *END command.
The data input commands are passive (not executed) while being written to the macro file. The macro file
(without *CREATE and *END ) can also be created external to ANSYS.
Up to 99 specially named scalar parameters called ARG1 to AR99 are locally available to each macro. Note
that the prefix for the first 9 parameters is "ARG," while the prefix for the last 90 is "AR." A local parameter
is one which is not affected by, nor does it affect, other parameters, even those of the same name, which
are used outside of the macro. The only way a local parameter can affect, or be affected by, parameters
outside the macro is if values are passed out of, or into, the macro by an argument list. Parameters ARG1
through AR18 can have their values (numeric or character) passed via the argument list on the *USE command
(ARG1 through AR19 can be passed as arguments on the "unknown command" macro). Parameters AR19
through AR99 (AR20 through AR99 in the "unknown command" macro) are available solely for use within
the macro; they cannot be passed via an argument list. Local parameters are available to do-loops and to
/INPUT files processed within the macro. In addition to an ARG1--AR99 set for each macro, another ARG1--
AR99 set is available external to all macros, local to "non-macro" space.
A macro is exited after its last line is executed. Macros may be nested (such as a *USE or an "unknown
command" within a macro). Each nested macro has its own set of 99 local parameters. Only one set of local
parameters can be active at a time and that is the set corresponding to the macro currently being executed
or to the set external to all macros (if any). When a nested macro completes execution, the previous set of
local parameters once again becomes available. Use *STATUS,ARGX to view current macro parameter values.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1778
*USE
An alternate way of executing a macro file is via the "unknown command" route. If a command unknown
to the ANSYS program is entered, a search for a file of that name (plus a .MAC suffix) is made. If the file
exists, it is executed, if not, the "unknown command" message is output. Thus, users can write their own
commands in terms of other ANSYS commands. The procedure is similar to issuing the *USE command with
the unknown command in the Name field. For example, the command CMD,10,20,30 is internally similar to
*USE,CMD,10,20,30. The macro file named CMD.MAC will be executed with the three parameters. The *USE
macro description also applies to the "unknown command" macro, except that various directories are searched
and a suffix (.MAC) is assumed. Also, a macro library file is not searched.
A three-level directory search for the "unknown command" macro file may be available (see the Operations
Guide). The search order may be: 1) a high-level system directory, 2) the login directory, and 3) the local
(working) directory. Use the /PSEARCH command to change the directory search path. For an "unknown
command" CMD, the first file named CMD.MAC found to exist in the search order will be executed. The
command may be input as upper or lower case, however, it is converted to upper case before the file name
search occurs. On systems that uniquely support both upper and lower case file names, the file with the
matching lower case name will be used if it exists, otherwise, the file with the matching upper case name
will be used. All macro files placed in the apdl directory must be upper case.
Note, since undocumented commands exist in the ANSYS program, the user should issue the command in-
tended for the macro file name to be sure the "unknown command" message is output in the processor
where it's to be used. If the macro is to be used in other processors, the other processors must also be
checked.
This command is valid in any processor.
Menu Paths
Utility Menu>Macro>Execute Data Block
/USER, WN
Conveniently resets /FOCUS and /DIST to USER.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
Notes
Conveniently resets scale parameters to USER on the /FOCUS and /DIST commands. Scale parameters will
be internally respecified to those used for the last display. Convenient when the last scale parameters were
automatically calculated. User specified parameters hold until changed or removed [/AUTO]. Parameters
may be reset on the individual commands after this command has been issued.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>View Settings>Automatic Fit Mode
1779
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/USER
USRCAL, Rnam1, Rnam2, Rnam3, Rnam4, Rnam5, Rnam6, Rnam7, Rnam8, Rnam9
Allows user-solution subroutines to be activated or deactivated.
SOLUTION: Load Step Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
Rnam1, Rnam2, Rnam3, . . . , Rnam9
User-defined solution subroutine names to be activated. Up to nine may be defined on one command
or multiple commands may be used. If Rnam1 = ALL, activate all valid user subroutines. If Rnam1 =
NONE, deactivate all valid user subroutines. All characters are required:
USREFL
Allows user defined scalar field (body force) loads.
USERCV
Allows user defined convection (surface) loads.
USERPR
Allows user defined pressure (surface) loads.
USERFX
Allows user-defined heat flux (surface) loads.
USERCH
Allows user-defined charge density (surface) loads.
USERFD
Computes the complex load vector for the frequency domain logic (PIPE59).
USEROU
Allows user supplied element output.
USERMC
Allows user control of the hygrothermal growth).
USOLBEG
Allows user access before each solution.
ULDBEG
Allows user access before each load step.
USSBEG
Allows user access before each substep.
UITBEG
Allows user access before each equilibrium iteration.
UITFIN
Allows user access after each equilibrium iteration.
USSFIN
Allows user access after each substep.
ULDFIN
Allows user access after each load step.
USOLFIN
Allows user access after each solution.
UANBEG
Allows user access at start of ANSYS run.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1780
USRCAL
UANFIN
Allows user access at end of ANSYS run.
UELMATX
Allows user access to element matrices and load vectors.
USWVVEAC
Allows user access to hydrodynamic loads on applicable line elements.
Command Default
No user-solution subroutines are active (even if linked into the program).
Notes
Allows certain user-solution subroutines to be activated or deactivated (system-dependent). This command
only affects the subroutines named. Other user subroutines (such as user elements, user creep, etc.) have
their own activation controls described with the feature.
The routines are commented and should be listed after performing a custom installation from the distribution
media for more details. See also the Advanced Analysis Techniques Guide for a general description of user-
programmable features.
Users must have system permission, system access, and knowledge to write, compile, and link the appropriate
subroutines into the ANSYS program at the site where it is to be run. All routines should be written in
FORTRAN. (For more information on FORTRAN compilers please refer to either the ANSYS, Inc. Windows In-
stallation Guide or the ANSYS, Inc. UNIX/Linux Installation Guide for details specific to your platform or oper-
ating system.) Issue USRCAL,STAT to list the status of these user subroutines. Since a user-programmed
subroutine is a nonstandard use of the program, the verification of any ANSYS run incorporating these
commands is entirely up to the user. In any contact with ANSYS customer support regarding the performance
of a custom version of the ANSYS program, you should explicitly state that a user programmable feature
has been used.
This command is also valid in PREP7.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Other>User Routines
Main Menu>Solution>Load Step Opts>Other>User Routines
USRDOF, Action, DOF1, DOF2, DOF3, DOF4, DOF5, DOF6, DOF7, DOF8, DOF9, DOF10
Specifies the degrees of freedom for the user-defined element USER300.
PREP7: Elements
MP ME ST <> <> <> <> FL EM EH <> PP <> EME MFS
Action
One of the following command operations:
1781
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
USRDOF
DEFINE
Specify the degrees of freedom (DOFs). This value is the default.
LIST
List all previously specified DOFs.
DELETE
Delete all previously specified DOFs.
DOF1, DOF2, DOF3, . . . , DOF10
The list of DOFs.
Notes
The USRDOF command specifies the degrees of freedom for the user-defined element USER300.
Although you can intersperse other commands as necessary for your analysis, issue the USRDOF command
as part of the following general sequence of commands:
1. Issue the ET command for element USER300, followed by the related TYPE command.
2. Issue both the USRELEM and USRDOF commands (in either order).
3. Define your element using USER300.
The DOF list (DOF1 through DOF10) can consist of up to 10 DOFs. Use any valid and appropriate DOF (such
as UX, UY, UZ, ROTX, ROTY, ROTZ, AX, AY, AZ, VX, VY, VZ, PRES, WARP, TEMP, VOLT, MAG, ENKE, ENDS, EMF,
and CURR).
You can specify a maximum of 10 DOFs per USRDOF command. To define additional DOFs, issue the command
again.
The maximum number of DOFs for a user-defined element--the number of nodes times the number of DOFs
per node--cannot exceed 480.
To learn more about user-defined elements, see Creating a New Element in the Programmer's Manual for
ANSYS.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
USRELEM, NNODES, NDIM, KeyShape, NREAL, NSAVEVARS, NRSLTVAR, KEYANSMAT, NINTPNTS, KESTRESS,
KEYSYM
Specifies the characteristics of the user-defined element USER300.
PREP7: Elements
MP ME ST <> <> <> <> FL EM EH <> PP <> EME MFS
NNODES
The number of nodes.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1782
USRELEM
NDIM
The number of dimensions (of nodal coordinates). Valid values are 2 or 3.
KeyShape
One of the following element shape options:
ANYSHAPE
Any shape (that is, no specified shape). This value is the default. (The ANSYS MeshTool is unavailable.)
POINT
Point.
LINE
Straight line.
TRIAN
Triangle.
QUAD
Quadrilateral. This shape can be degenerated to a triangle.
TET
Tetrahedron.
BRICK
Brick. This shape can be degenerated to a wedge, pyramid, or tetrahedron.
NREAL
The number of real constants.
NSAVEVARS
The number of saved variables.
NRSLTVAR
The number of variables saved in results files.
KEYANSMAT
Key for element formulation control:
0
Create your own material codes within the element formulation. In this case, the real constants are
available to input material properties. You can also input linear material properties via MP and MP-
DATA commands .
1
Use ANSYS standard material routines or the USERMAT subroutine to form structural material data.
ANSYS material properties must be input in the standard way (as you would for non-user-defined
elements). This value is invalid when KeyShape = ANYSHAPE.
NINTPNTS
The maximum number of integration points (used when KEYANSMAT = 1).
KESTRESS
Key for the element stress state (used when KEYANSMAT = 1):
0
Plane stress elements.
1
Axisymmetric elements.
1783
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
USRELEM
2
Plane strain elements.
3
3-D solid elements.
4
3-D solid-shell elements.
5
Generalized plane strain elements.
6
Beam elements.
7
Link/truss elements.
8
3-D shell elements.
9
Axisymmetric shell elements.
KEYSYM
Key for specifying whether element stiffness matrices are symmetric or unsymmetric:
0
Symmetric.
1
Unsymmetric.
Notes
The USRELEM command specifies the characteristics of the user-defined element USER300.
Although you can intersperse other commands as necessary for your analysis, issue the USRELEM command
as part of the following general sequence of commands:
1. Issue the ET command for element USER300, followed by the related TYPE command.
2. Issue both the USRELEM and USRDOF commands (in either order).
3. Define your element using USER300.
The number of real constants (NREAL) can refer to geometry quantities, material quantities, or any parameters
for element formulation.
ANSYS saves variables in the .esav file to preserve element data when you specify a positive NSAVEVARS
value. When KEYANSMAT = 0, all variables of both material and kinematic formulation are saved. When
KEYANSMAT = 1, only the variables for kinematic formulation (such as deformation gradient tensor) are
saved; in this case, the material routine saves all necessary material data automatically.
Element data saved in results files (NRSLTVAR) are accessible only as nonsummable miscellaneous data.
ANSYS saves stress and total strain data for structural elements in the .rst file automatically (as it does for
equivalent variables such as thermal gradient and thermal flux in thermal elements); therefore, NRSLTVAR
does not need to include stress and total strain data.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1784
USRELEM
To learn more about creating user-defined elements, see Creating a New Element in the Programmer's
Manual for ANSYS.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
1785
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
USRELEM
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1786
V Commands
V, P1, P2, P3, P4, P5, P6, P7, P8
Defines a volume through keypoints.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
P1
Keypoint defining starting corner of volume. If P1 = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI).
P2
Keypoint defining second corner of volume.
P3
Keypoint defining third corner of volume.
P4
Keypoint defining fourth corner of volume.
P5
Keypoint defining fifth corner of volume.
P6
Keypoint defining sixth corner of volume.
P7
Keypoint defining seventh corner of volume.
P8
Keypoint defining eighth corner of volume.
Notes
Defines a volume (and its corresponding lines and areas) through eight (or fewer) existing keypoints. Keypoints
must be input in a continuous order. The order of the keypoints should be around the bottom and then the
top. Missing lines are generated "straight" in the active coordinate system and assigned the lowest available
numbers [NUMSTR]. Missing areas are generated and assigned the lowest available numbers.
Note
Solid modeling in a toroidal coordinate system is not recommended.
Certain faces may be condensed to a line or point by repeating keypoints. For example, use
V,P1,P2,P3,P3,P5,P6,P7,P7 for a triangular prism or V,P1,P2,P3,P3,P5,P5,P5,P5 for a tetrahedron. See
SOLID45 Geometry in the Element Reference for analogous element sketches of the valid condensed face
patterns.)
1787
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Using keypoints to produce partial sections in CSYS = 2 can generate anomalies; check the resulting volumes
carefully.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volumes>Arbitrary>Through KPs
V2DOPT, GEOM, NDIV, HIDOPT, NZONE
Specifies 2-D/axisymmetric view factor calculation options.
SOLUTION: Radiosity
AUX12: Radiosity Solver
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
GEOM
Choice of geometry:
0
Planar (default).
1
Axisymmetric
NDIV
Number of divisions for axisymmetric geometry. Defaults to 20.
HIDOPT
Viewing option:
0
Hidden (default).
1
Non-hidden
NZONE
Number of zones for view factor calculation. Defaults to 200.
Command Default
V2DOPT allows you to select options for 2-D view factor calculation. The geometry type can be set to either
2-D plane or axisymmetric (defaults to plane). You can also define the number of divisions (defaults to 20)
for an axisymmetric geometry. This command also allows you to select either hidden or non-hidden viewing
option (defaults to hidden) and the number of zones for view factor calculation (defaults to 200).
Menu Paths
Main Menu>Preprocessor>Radiation Opts>View Factor
Main Menu>Radiation Opt>Radiosity Meth>View Factor
Main Menu>Solution>Radiation Opts>View Factor
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1788
V2DOPT
VA, A1, A2, A3, A4, A5, A6, A7, A8, A9, A10
Generates a volume bounded by existing areas.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
A1, A2, A3, . . . , A10
List of areas defining volume. The minimum number of areas is 4. If A1 = ALL, use all selected [ASEL]
areas and ignore A2 to A10. If A1 = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI). A component name may also be substituted for A1.
Notes
This command conveniently allows generating volumes from regions having more than eight keypoints
(which is not allowed with the V command). Areas may be input in any order. The exterior surface of a VA
volume must be continuous, but holes may pass completely through it.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Volume by Areas
Main Menu>Preprocessor>Modeling>Create>Volumes>Arbitrary>By Areas
Main Menu>Preprocessor>Modeling>Geom Repair>Create Volume
*VABS, KABSR, KABS1, KABS2, KABS3
Applies the absolute value function to array parameters.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
KABSR
Absolute value of results parameter:
0 --
Do not take absolute value of results parameter (ParR).
1 --
Take absolute value.
KABS1
Absolute value of first parameter:
0 --
Do not take absolute value of first parameter (Par1 or ParI).
1 --
Take absolute value.
KABS2
Absolute value of second parameter:
1789
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VABS
0 --
Do not take absolute value of second parameter (Par2 or ParJ).
1 --
Take absolute value.
KABS3
Absolute value of third parameter:
0 --
Do not take absolute value of third parameter (Par3 or ParK).
1 --
Take absolute value.
Command Default
Do not use absolute values.
Notes
Applies an absolute value to parameters used in certain *VXX and *MXX operations. Typical absolute value
applications are of the form:
ParR = |f(|Par1|)|
or
ParR = |(|Par1| o |Par2|)|
The absolute values are applied to each input parameter value before the operation and to the result value
after the operation. Absolute values are applied before the scale factors so that negative scale factors may
be used. The absolute value settings are reset to the default (no absolute value) after each *VXX or *MXX
operation. Use *VSTAT to list settings.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Operation Settings
VADD, NV1, NV2, NV3, NV4, NV5, NV6, NV7, NV8, NV9
Adds separate volumes to create a single volume.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NV3, . . . , NV9
Numbers of volumes to be added. If NV1 = ALL, add all selected volumes and ignore NV2 to NV9. If NV1
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may also be substituted for NV1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1790
VADD
Notes
Adds separate volumes to create a single volume. The original volumes (and their corresponding areas, lines
and keypoints) will be deleted by default [BOPTN]. See the BOPTN command for the options available to
Boolean operations. Element attributes and solid model boundary conditions assigned to the original entities
will not be transferred to the new entities generated. Concatenated entities are not valid with this command.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Add>Volumes
Main Menu>Preprocessor>Modeling>Operate>Booleans>Add>Volumes
VALVE, NLOC, LENG, MASS, SIF, FLEX, ARINS, ELEM
Defines a valve in a piping run.
PREP7: Piping
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NLOC
Node where valve is to be placed (as described below). Defaults to current run starting point.
LENG
Length of valve (defaults to larger pipe OD).
MASS
Dry mass (weight/gravity) of valve without insulation (defaults to equivalent straight pipe mass). Note,
acceleration [ACEL] must be nonzero for weight to be calculated.
SIF
Stress intensification factor (defaults to 1.0).
FLEX
Bending flexibility factor (defaults to 0.5).
ARINS
Insulation surface area (defaults to equivalent straight pipe insulation area). Units (length
2
) must be
consistent with the smallest unit of the system used (not mixed) regardless of the PUNIT option.
ELEM
Element number to be assigned to valve (defaults to the previous maximum element number (MAXEL)
+ 1).
Notes
Defines a valve (straight pipe element (PIPE16) with adjusted specifications and loadings) at a given location
in a piping run. See the PREP7 RUN command. The location may be 1) between two adjacent colinear straight
pipes, 2) between an adjacent straight pipe and a different piping component, or 3) at the end of a straight
pipe.
For Case 1, two new nodes are generated at the ends of the valve. The two straight pipes are automatically
"shortened" to meet the ends of the valve. The valve specifications and loadings are taken from the corres-
ponding two straight pipes.
1791
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VALVE
For Case 2, one new node is generated at one end of the valve. The straight pipe is automatically "shortened"
to meet this end of the valve. The other end of the valve meets the other piping component. The valve
specifications and loadings are taken from the straight pipe.
For Case 3, one new node is generated at the free end of the valve. The other end of the valve meets the
straight pipe. The valve specifications and loadings are taken from the straight pipe.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Piping Models>Define Pipes>Valve
VARDEL, NVAR
Deletes a variable (GUI).
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NVAR
The reference number of the variable to be deleted. NVAR is as defined by NSOL, ESOL, etc.
Notes
Deletes a POST26 solution results variable. This is a command generated by the Graphical User Interface
(GUI). It will appear in the log file (Jobname.LOG) if a POST26 variable is deleted from the "Defined Time-
History Variables" dialog box. This command is not intended to be typed in directly in an ANSYS session
(although it can be included in an input file for batch input or for use with the /INPUT command).
Menu Paths
Main Menu>TimeHist Postpro>Define Variables
Main Menu>TimeHist Postpro>Elec&Mag>Circuit>Define Variables
VARNAM, IR, Name
Names (or renames) a variable.
POST26: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
IR
Reference number of the variable (2 to NV [NUMVAR]).
Name
Thirty-two character name for identifying variable on printouts and displays. Embedded blanks are
compressed for output.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Graph
Main Menu>TimeHist Postpro>Settings>List
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1792
VARDEL
VATT, MAT, REAL, TYPE, ESYS, SECNUM
Associates element attributes with the selected, unmeshed volumes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
MAT, REAL, TYPE, ESYS, SECNUM
Material number, real constant set number, type number, coordinate system number, and section number
to be associated with selected, unmeshed volumes.
Notes
These element attributes will be used when the volumes are meshed. If a volume does not have attributes
associated with it (by this command) at the time it is meshed, the attributes are obtained from the then
current MAT, REAL, TYPE, ESYS, and SECNUM command settings. Reissue the VATT command (before
volumes are meshed) to change the attributes. A zero (or blank) argument removes the corresponding asso-
ciation.
If any of the arguments MAT, REAL, TYPE, ESYS or SECNUM are defined as -1, then that value will be left
unchanged in the selected set.
In some cases, ANSYS can proceed with a volume meshing operation even when no logical element type
has been assigned via VATT,,,TYPE or TYPE. For more information, see the discussion on setting element
attributes in Meshing Your Solid Model of the Modeling and Meshing Guide.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>All Volumes
Main Menu>Preprocessor>Meshing>Mesh Attributes>Picked Volumes
VCLEAR, NV1, NV2, NINC
Deletes nodes and volume elements associated with selected volumes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
Delete mesh for volumes NV1 to NV2 (defaults to NV1) in steps of NINC (defaults to 1). If NV1 = ALL,
NV2 and NINC are ignored and mesh for all selected volumes [VSEL] is deleted. If NV1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). A component
name may also be substituted for NV1 (NV2 and NINC are ignored).
Notes
Deletes all nodes and volume elements associated with selected volumes (regardless of whether the nodes
or elements are selected). Nodes shared by adjacent meshed volumes and nodes associated with non-volume
elements will not be deleted. Attributes assigned as a result of VATT are maintained. In the program's response
to the command, if a volume, area, line, or keypoint is tallied as "cleared," it means either its node or element
reference was deleted.
1793
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VCLEAR
Menu Paths
Main Menu>Preprocessor>Meshing>Clear>Volumes
*VCOL, NCOL1, NCOL2
Specifies the number of columns in matrix operations.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
NCOL1
Number of columns to be used for Par1 with *MXX operations. Defaults to whatever is needed to fill
the result array.
NCOL2
Number of columns to be used for Par2 with *MXX operations. Defaults to whatever is needed to fill
the result array.
Command Default
Fill all locations of the result array from the specified starting location.
Notes
Specifies the number of columns to be used in array parameter matrix operations. The size of the submatrix
used is determined from the upper left starting array element (defined on the operation command) to the
lower right array element (defined by the number of columns on this command and the number of rows
on the *VLEN command).
The default NCOL is calculated from the maximum number of columns of the result array (the *DIM column
dimension) minus the starting location + 1. For example, *DIM,R,,1,10 and a starting location of R(1,7) gives
a default of 4 columns ( starting with R(1,7), R(1,8), R(1,9), and R(1,10)). Repeat operations automatically ter-
minate at the last column of the result array. Existing values in the rows and columns of the results matrix
remain unchanged where not overwritten by the requested input or operation values.
The column control settings are reset to the defaults after each *MXX operation. Use *VSTAT to list settings.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Operation Settings
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1794
*VCOL
/VCONE, WN, PHI
Defines the view cone angle for perspective displays.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
PHI
View cone angle (0.0 to 85.) to define perspective. Use PHI = 45.0 for typical perspective. Increase
angle for more perspective, decrease angle for less. If the distance [/DIST] is not specified, it will be
automatically calculated to give full window magnification. If the distance is also specified, PHI controls
both the perspective and the magnification. The larger the angle, the more the perspective and the less
the magnification. Defaults to 0.0 (no perspective).
Command Default
0.0 degrees (no perspective (parallel projection)).
Notes
Perspective shows the true depth of the object in the display. A variable magnification results since the back
plane of the object is further from the observer than the front plane. The largest magnification occurs at
the front plane. With perspective, the magnification factor (MAGF) is not only a function of the distance
from the object, but also the window shape and the perspective (or view cone) angle as follows:
MAGF
d TAN
/
( )
2
where
, for square windows, is the largest in-plane vertical or horizontal dimension, d is the distance from
the observer to the plane of
(usually the front plane of the object), and is the view cone angle (defined
with the /VCONE command). The bigger the cone angle, the more the perspective. The magnification factor
proportionally decreases with increasing . The distance can be defined with the /DIST or the /FOCUS
command. Note, the distance input on the /DIST command is equal to d only if the focus point is located
on the plane of
. It is recommended that if a general perspective is desired (i.e., not any specific cone
angle), use = 45.0 (since TAN(45.0) = 1.0) and let the d value be automatically calculated for full window
magnification.
Note that any number of /DIST, /FOCUS, and /VCONE combinations can be used to produce the same
magnification. Distances less than the object depth will produce views from within the object.
A magnification factor of 1.0 just fills the window. If the automatic scaling option is used [/AUTO], the
magnification factor is fixed at 0.91 (to allow a 10% margin around the object) and d is automatically calculated
for the given /VCONE and /FOCUS values. Any value of between 0.0 and 85.0 (usually 45.0) may be used
to activate the perspective. Views from inside the object are not possible when d is automatically calculated
(use manual scaling [/USER] along with /DIST specification).
This command is valid in any processor.
1795
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/VCONE
Menu Paths
Utility Menu>PlotCtrls>View Settings>Perspective View
VCROSS, LabXR, LabYR, LabZR, LabX1, LabY1, LabZ1, LabX2, LabY2, LabZ2
Forms element table items from the cross product of two vectors.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabXR, LabYR, LabZR
Label assigned to X, Y, and Z-component of resultant vector.
LabX1, LabY1, LabZ1
X, Y, and Z-component of first vector label.
LabX2, LabY2, LabZ2
X, Y, and Z-component of second vector label.
Notes
Forms labeled result items for the selected element from the cross product of two vectors:
{LabXR, LabYR, LabZR} = {LabX1, LabY1, LabZ1} X {LabX2, LabY2, LabZ2}
Data must be in a consistent coordinate system. Labels are those associated with the ETABLE command.
Menu Paths
Main Menu>General Postproc>Element Table>Cross Product
*VCUM, KEY
Allows array parameter results to add to existing results.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
KEY
Accumulation key:
0 --
Overwrite results.
1 --
Add results to the current value of the results parameter.
Command Default
Overwrite results.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1796
VCROSS
Notes
Allows results from certain *VXX and *MXX operations to overwrite or add to existing results. The cumulative
operation is of the form:
ParR = ParR + ParR(Previous)
The cumulative setting is reset to the default (overwrite) after each *VXX or *MXX operation. Use *VSTAT
to list settings.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Operation Settings
VCVFILL, A1, A2, A3, A4, A5, A6, A7, A9, A9, A10
Fills cavities and bosses in volumes (for models imported from CAD files).
PREP7: CAD Repair
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
A1, A2, A3, . . . , A10
List of areas that define the volume. If A1 = P, graphical picking is enabled and all remaining arguments
are ignored (valid only in the GUI). If A1 = ALL, all selected areas will be used and all remaining arguments
are ignored.
Notes
Use this command to fill in holes and cavities or eliminate protrusions (bosses) in models imported from
CAD files. In essence, this command removes features projecting above or below a plane, such as bosses
and cavities. This command is available only for models imported through the Default IGES option.
You must select all areas related to the feature before it can be eliminated, and the order in which those
areas are selected can be important. If the boss or cavity is an isolated entity, the area selection order does
not matter. However, if the boss or cavity is attached to an area, that area must be the last area selected in
the list. If the attached area is not the last area in the selection list the command may distort the geometry
of the model.
Menu Paths
Main Menu>Preprocessor>Modeling>Simplify>Toolkit>Fill Cavity
Main Menu>Preprocessor>Modeling>Simplify>Toolkit>Remove Boss
1797
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VCVFILL
VDDAM, VF, VA, VB, VC
Specifies the velocity spectrum computation constants for the analysis of shock resistance of shipboard
structures.
SOLUTION: Spectrum Options
MP ME ST <> <> <> <> <> <> <> <> PP <> EME MFS
VF
Direction-dependent velocity coefficient for elastic or elastic-plastic analysis option (Default = 0).
VA, VB, VC
Coefficients for the DDAM velocity spectrum equations. See the Theory Reference for the Mechanical APDL
and Mechanical Applications. Default for these coefficients is zero.
Notes
This command specifies velocity coefficients to analyze shock resistance of shipboard equipment. These
coefficients are used to compute mode coefficients according to the equations given in the Theory Reference
for the Mechanical APDL and Mechanical Applications. The form of these equations is based on the Naval NRL
Dynamic Design Analysis Method. This command, along with the ADDAM and SED commands, is used with
the spectrum (ANTYPE,SPECTR) analysis as a special purpose alternative to the SV, FREQ, and SVTYP com-
mands. The mass and length units of the model must be in pounds and inches, respectively.
This command is also valid in PREP7.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Spectrum>DDAM Options
Main Menu>Solution>Load Step Opts>Spectrum>DDAM Options
VDELE, NV1, NV2, NINC, KSWP
Deletes unmeshed volumes.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
Delete volumes from NV1 to NV2 (defaults to NV1) in steps of NINC (defaults to 1). If NV1 = ALL, NV2
and NINC are ignored and all selected volumes [VSEL] are deleted. If NV1 = P, graphical picking is enabled
and all remaining command fields are ignored (valid only in the GUI). A component name may also be
substituted for NV1 (NV2 and NINC are ignored).
KSWP
Specifies whether keypoints, lines, and areas are also deleted:
0
Delete volumes only (default).
1
Delete volumes, as well as keypoints, lines, and areas attached to the specified volumes but not
shared by other volumes.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1798
VDDAM
Menu Paths
Main Menu>Preprocessor>Modeling>Delete>Volume and Below
Main Menu>Preprocessor>Modeling>Delete>Volumes Only
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Delete>Volume and Below
Main Menu>Preprocessor>Trefftz Domain>TZ Geometry>Delete>Volumes Only
VDGL, NV1, NV2, NINC
Lists keypoints of a volume that lie on a parametric degeneracy.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
List keypoints that lie on a parametric degeneracy on volumes from NV1 to NV2 (defaults to NV1) in
steps of NINC (defaults to 1). If NV1 = ALL (default), NV2 and NINC will be ignored and keypoints on
all selected volumes [VSEL] will be listed. If NV1 = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI). You may also substitute a component name for NV1
(ignore NV2 and NINC).
Notes
See the Modeling and Meshing Guide for details about parametric degeneracies.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Check Geom>Show Degeneracy>List Degen Volus
Main Menu>Preprocessor>Modeling>Operate>Booleans>Show Degeneracy>List Degen Volus
VDOT, LabR, LabX1, LabY1, LabZ1, LabX2, LabY2, LabZ2
Forms an element table item from the dot product of two vectors.
POST1: Element Table
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LabR
Label assigned to dot product result.
LabX1, LabY1, LabZ1
X, Y, and Z-component of first vector label.
LabX2, LabY2, LabZ2
X, Y, and Z-component of second vector label.
Notes
Forms labeled result items for the selected element from the dot product of two vectors:
1799
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VDOT
LabR = {LabX1, LabY1, LabZ1} {LabX2, LabY2, LabZ2}
Data must be in a consistent coordinate system. Labels are those associated with the ETABLE command.
Menu Paths
Main Menu>General Postproc>Element Table>Dot Product
VDRAG, NA1, NA2, NA3, NA4, NA5, NA6, NLP1, NLP2, NLP3, NLP4, NLP5, NLP6
Generates volumes by dragging an area pattern along a path.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NA1, NA2, NA3, . . . , NA6
List of areas in the pattern to be dragged (6 maximum if using keyboard entry). If NA1 = P, graphical
picking is enabled and all remaining command fields are ignored (valid only in the GUI). If NA1 = ALL,
all selected areas will be swept along the path. A component name may also be substituted for NA1.
NLP1, NLP2, NLP3, . . . , NLP6
List of lines defining the path along which the pattern is to be dragged (6 maximum if using keyboard
entry). Must be a continuous set of lines. To be continuous, adjacent lines must share the connecting
keypoint (the end keypoint of one line must also be first keypoint of the next line).
Notes
Generates volumes (and their corresponding keypoints, lines, and areas) by sweeping a given area pattern
along a characteristic drag path. If the drag path consists of multiple lines, the drag direction is determined
by the sequence in which the path lines are input (NLP1, NLP2, etc.). If the drag path is a single line (NLP1),
the drag direction is from the keypoint on the drag line that is closest to the first keypoint of the given area
pattern to the other end of the drag line.
The magnitude of the vector between the keypoints of the given pattern and the first path keypoint remains
constant for all generated keypoint patterns and the path keypoints. The direction of the vector relative to
the path slope also remains constant so that patterns may be swept around curves. Lines are generated
with the same shapes as the given pattern and the path lines.
Keypoint, line, area, and volume numbers are automatically assigned (beginning with the lowest available
values [NUMSTR]). Adjacent lines use a common keypoint, adjacent areas use a common line, and adjacent
volumes use a common area. For best results, the entities to be dragged should be orthogonal to the start
of the drag path. Drag operations that produce an error message may create some of the desired entities
prior to terminating.
If element attributes have been associated with the input area via the AATT command, the opposite area
generated by the VDRAG operation will also have those attributes (i.e., the element attributes from the input
area are copied to the opposite area). Note that only the area opposite the input area will have the same
attributes as the input area; the areas adjacent to the input area will not.
If the input areas are meshed or belong to a meshed volume, the area(s) can be extruded to a 3-D mesh.
Note that the NDIV argument of the ESIZE command should be set before extruding the meshed areas.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1800
VDRAG
Alternatively, mesh divisions can be specified directly on the drag line(s) (LESIZE). See the Modeling and
Meshing Guide for more information.
You can use the VDRAG command to generate 3-D interface element meshes for elements INTER194 and
INTER195. When generating interface element meshes using VDRAG, you must specify the line divisions to
generate one interface element directly on the drag line using the LESIZE command. The source area to be
extruded becomes the bottom surface of the interface element. Interface elements must be extruded in
what will become the element's local x direction, that is, bottom to top.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extrude>Areas>Along Lines
*VEDIT, Par
Allows numerical array parameters to be graphically edited.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
Name of the array parameter to be edited.
Notes
Invokes a graphical editing system that displays array parameter values in matrix form, and allows the use
of the mouse to edit individual values. The starting array subscripts must be defined, such as *VEDIT,A(4,6,1),
to indicate the section of the array to be edited. The array section starts at the specified array element and
continues to the maximum extent of the array parameter. Row and column index values may be set or
changed in any plane, and those values will be applied to all planes. The menu system must be on [/MENU]
when this command is issued. Graphical editing is not available for character array parameters. The *VEDIT
command can not be used in a macro or other secondary input file.
This command is not applicable to 4- or 5-D arrays.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Parameters>Define/Edit
1801
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VEDIT
VEORIENT, VNUM, Option, VALUE1, VALUE2
Specifies brick element orientation for volume mapped (hexahedron) meshing.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
VNUM
Number identifying volume for which elements are to be oriented (no default).
Option
Option for defining element orientation:
KP
Orientation is determined by two keypoints on the volume. Input the keypoint numbers (KZ1 and
KZ2) in fields VALUE1 and VALUE2, respectively. The element z-axis points from KZ1 toward KZ2.
Element x and y directions point away from KZ1 along edges of the volume to make a right-hand
triad. (The element x- and y-axes are uniquely determined by this specification.)
LINE
Orientation is determined by one of the lines defining the volume. Input the line number in field
VALUE1. The element z direction follows the direction of the line. Input a negative value if the desired
z direction is opposite to the direction of the specified line. (The element x- and y-axes are uniquely
determined by this specification.) (VALUE2 is not used.)
AREA
Orientation is determined by one of the areas defining the volume. The area represents the desired
element top surface. Input the area number as VALUE1. The shortest line in the volume connected
to the area will be used to specify the element z direction. (If more than one shortest line exists, the
lowest numbered of those is used.) Element x and y directions are not uniquely specified by this
option. (VALUE2 is not used.)
THIN
Align the element z normal to the thinnest dimension of the volume. The shortest line in the volume
is used to specify the element z direction. (If more than one shortest line exists, the lowest numbered
of those is used.) Element x and y directions are not uniquely specified by this option. (VALUE1 and
VALUE2 are not used.)
DELE
Delete the previously defined volume orientation for the specified volume (VNUM). (VALUE1 and
VALUE2 are not used.)
VALUE1, VALUE2
Parameters required for the element z-axis direction specification. The meaning of VALUE1 and VALUE2
will depend on the chosen Option. See the description of Option above for details.
Command Default
Elements are not oriented in any specific manner.
Notes
Use VEORIENT before the VMESH command to specify the desired orientation of brick elements in a mapped
mesh. VEORIENT has no effect on tetrahedron meshes, extruded meshes (VROTAT, VDRAG, VEXT, etc.), or
swept meshes (VSWEEP).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1802
VEORIENT
Proper brick orientation is essential for certain element types such as SOLID185 Layered Solid, SOLID186
Layered Solid, and SOLSH190. In such cases, use VEORIENT or EORIENT to achieve the desired orientation.
For other brick element types, you may need to specify element orientation to control orthotropic material
property directions without concern for the element connectivity. For those cases, the ESYS command is
the preferred method of specifying the material property directions.
For Option = LINE, AREA, and THIN, the orientation will be internally converted to an equivalent Option
= KP specification (KP,KZ1,KZ2). Use the VLIST command to view the element orientations (in terms of KZ1
and KZ2) associated with each volume.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh Attributes>Volume Brick Orient>Delete Specification
Main Menu>Preprocessor>Meshing>Mesh Attributes>Volume Brick Orient>Z Along Line
Main Menu>Preprocessor>Meshing>Mesh Attributes>Volume Brick Orient>Z by 2 Keypoints
Main Menu>Preprocessor>Meshing>Mesh Attributes>Volume Brick Orient>Z in Thin Direction
Main Menu>Preprocessor>Meshing>Mesh Attributes>Volume Brick Orient>Z Normal to Area
VEXT, NA1, NA2, NINC, DX, DY, DZ, RX, RY, RZ
Generates additional volumes by extruding areas.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NA1, NA2, NINC
Set of areas (NA1 to NA2 in steps of NINC) that defines the pattern to be extruded. NA2 defaults to
NA1, NINC defaults to 1. If NA1 = ALL, NA2 and NINC are ignored and the pattern is defined by all se-
lected areas. If NA1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NA1 (NA2 and NINC are ignored).
DX, DY, DZ
Increments to be applied to the X, Y, and Z keypoint coordinates in the active coordinate system (DR,
D, DZ for cylindrical; DR, D, D for spherical).
RX, RY, RZ
Scale factors to be applied to the X, Y, and Z keypoint coordinates in the active coordinate system (RR,
R, RZ for cylindrical; RR, R, R for spherical). Note that the R and R scale factors are interpreted as
angular offsets. For example, if CSYS = 1, RX, RY, RZ input of (1.5,10,3) would scale the specified key-
points 1.5 times in the radial and 3 times in the Z direction, while adding an offset of 10 degrees to the
keypoints. Zero, blank, or negative scale factor values are assumed to be 1.0. Zero or blank angular offsets
have no effect.
Notes
Generates additional volumes (and their corresponding keypoints, lines, and areas) by extruding and scaling
a pattern of areas in the active coordinate system.
If element attributes have been associated with the input area via the AATT command, the opposite area
generated by the VEXT operation will also have those attributes (i.e., the element attributes from the input
area are copied to the opposite area). Note that only the area opposite the input area will have the same
attributes as the input area; the areas adjacent to the input area will not.
1803
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VEXT
If the areas are meshed or belong to meshed volumes, a 3-D mesh can be extruded with this command.
Note that the NDIV argument on the ESIZE command should be set before extruding the meshed areas.
Scaling of the input areas, if specified, is performed first, followed by the extrusion.
In a non-Cartesian coordinate system, the VEXT command locates the end face of the volume based on the
active coordinate system. However, the extrusion is made along a straight line between the end faces. Note
that solid modeling in a toroidal coordinate system is not recommended.
Caution
Use of the VEXT command can produce unexpected results when operating in a non-Cartesian
coordinate system. For a detailed description of the possible problems that may occur, see Solid
Modeling in the Modeling and Meshing Guide.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extrude>Areas>By XYZ Offset
*VFACT, FACTR, FACT1, FACT2, FACT3
Applies a scale factor to array parameters.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
FACTR
Scale factor applied to results (ParR) parameter. Defaults to 1.0.
FACT1
Scale factor applied to first parameter (Par1 or ParI). Defaults to 1.0.
FACT2
Scale factor applied to second parameter (Par2 or ParJ). Defaults to 1.0.
FACT3
Scale factor applied to third parameter (Par3 or ParK). Defaults to 1.0.
Command Default
Use 1.0 for all scale factors.
Notes
Applies a scale factor to parameters used in certain *VXX and *MXX operations. Typical scale factor applic-
ations are of the form:
ParR = FACTR*f(FACT1*Par1)
or
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1804
*VFACT
ParR = FACTR*((FACT1*Par1) o (FACT2*Par2))
The factors are applied to each input parameter value before the operation and to the result value after the
operation. The scale factor settings are reset to the default (1.0) after each *VXX or *MXX operation. Use
*VSTAT to list settings.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Operation Settings
*VFILL, ParR, Func, CON1, CON2, CON3, CON4, CON5, CON6, CON7, CON8, CON9, CON10
Fills an array parameter.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting numeric array parameter vector. See *SET for name restrictions.
Func
Fill function:
DATA --
Assign specified values CON1, CON2, etc. to successive array elements. Up to 10 assignments may
be made at a time. Any CON values after a blank CON value are ignored.
RAMP --
Assign ramp function values: CON1+((n-1)*CON2) , where n is the loop number [*VLEN]. To specify
a constant function (no ramp), set CON2 to zero.
RAND --
Assign random number values based on a uniform distribution: RAND(CON1,CON2), where CON1 is
the lower bound (defaults to 0.0) and CON2 is the upper bound (defaults to 1.0).
GDIS --
Assign random sample of Gaussian distributions: GDIS(CON1,CON2), where CON1 is the mean (defaults
to 0.0) and CON2 is the standard deviation (defaults to 1.0).
TRIA --
Assigns random number values based on a triangular distribution: TRIA(CON1,CON2,CON3), where
CON1 is the lower bound (defaults to 0.0), CON2 is the location of the peak value
(CON1
CON2
CON3, CON1 if 0
CON1, or CON3 if
CON3
0), and CON3 is the upper bound (defaults to 1.0 + CON1 if CON1
0 or 0.0 if CON1
0).
BETA --
Assigns random number values based on a beta distribution: BETA(CON1,CON2,CON3,CON4), where
CON1 is the lower bound (defaults to 0.0), CON2 is the upper bound (defaults to 1.0+CON1 if CON1
0
or 0.0 if CON1
0), and CON3 and CON4 are the alpha and beta parameters, respectively, of the beta
function. Alpha and beta must both be positive; they default to 1.0.
1805
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VFILL
GAMM --
Assigns random number values based on a gamma distribution: GAMM(CON1,CON2,CON3), where
CON1 is the lower bound (defaults to 0.0), and CON2 and CON3 are the alpha and beta parameters.
respectively, of the gamma function. Alpha and beta must both be positive; they default to 1.0.
CON1, CON2, CON3, . . . , CON10
Constants used with above functions.
Notes
Operates on input data and produces one output array parameter vector according to:
ParR = f(CON1, CON2, ...)
where the functions (f ) are described above. Operations use successive array elements [*VLEN, *VMASK]
with the default being all successive elements. For example, *VFILL,A,RAMP,1,10 assigns A(1) = 1.0, A(2) =
11.0, A(3) = 21.0, etc. *VFILL,B(5,1),DATA,1.5,3.0 assigns B(5,1) = 1.5 and B(6,1) = 3.0. Absolute values and
scale factors may be applied to the result parameter [*VABS, *VFACT]. Results may be cumulative [*VCUM].
See the *VOPER command for details.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Parameters>Fill
VFOPT, Opt, Filename, Ext, Dir, Format
Specifies options for view factor file.
SOLUTION: Radiosity
AUX12: Radiosity Solver
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
Opt
View factor option:
NEW
Compute view factors and write them to a file.
OFF
Do not recompute view factors it they already exist in the database, otherwise compute them (default).
READ
Read view factors from a binary file. For subsequent SOLVE commands, switch to the default option
(OFF).
NONE
Do not write view factors to a file.
Fname
File name for view factor matrix. Defaults to Jobname.
Ext
Filename extension for view factor matrix (default = vf ).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1806
VFOPT
Dir
Directory path for view factor matrix. If you do not specify a directory path, it will default to your working
directory.
Format
File format for view factor file. Defaults to a Binary format.
BINA
Binary file (default).
ASCI
ASCII file.
Notes
This command allows you to deactivate the view factor computation (Opt = OFF) if the view factors already
exist in the database. OFF is the default upon encountering the second and subsequent SOLVE commands
in the solution processor. When using the READ option, only a previously-calculated view factor binary file
is supported.
Menu Paths
Main Menu>Preprocessor>Radiation Opts>View Factor
Main Menu>Radiation Opt>Radiosity Meth>View Factor
Main Menu>Solution>Radiation Opts>View Factor
VFQUERY, SRCELEM, TARELEM
Queries and prints element Hemicube view factors and average view factor.
AUX12: Radiosity Solver
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
SRCELEM
Elements representing the source radiating surfaces used to query the view factor at the target element(s).
If SRCELEM = P, graphical picking is enabled (valid only in the GUI). If SRCELEM = ALL, all selected ele-
ments will have their view factors queried. A component name may also be substituted for SRCELEM.
Selected elements must be flagged for surface to surface radiation in order to query view factors (SF,
SFA, or SFE with Lab = RDSF). The view factors must have been previously computed.
TARELEM
Element for view factor query. If TARELEM = P, graphical picking is enabled (valid only in the GUI). If
TARELEM = ALL, all selected elements will have their view factors queried. A component name may also
be substituted for TARELEM. Selected elements must be flagged for surface to surface radiation in order
to query view factors (SF, SFA, or SFE with Lab = RDSF). The view factors must have been previously
computed.
Notes
View factors for each target element will be printed.
An average view factor for all target elements will be computed. (Use *GET to retrieve the average value).
1807
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VFQUERY
When resuming a database, issue the command VFOPT,READ before issuing the VFQUERY command.
Menu Paths
Main Menu>Radiation Opt>Radiosity Meth>Query
VFSM, Action, ENCL, Scale
Scales view factor matrix to yield row sum values equal to one.
AUX12: Radiosity Solver
MP ME <> PR <> <> <> <> <> <> <> PP <> EME MFS
Action
Action to be performed:
Define
Define a view factor summation (default)
Clear
Resets the scaling method to 0 for all enclosures. All subsequent arguments are ignored.
Status
Outputs scaling method for each enclosure in the model.
ENCL
Enclosure number for the view factor scaling.
Scale
Scaling factor key:
0
The view factor matrix is not changed and no scaling is applied (default).
1
Use scaling.
Notes
For a perfect enclosure the view factor row sum should be 1.0. Computational errors may yield a row sum
< 1.0, requiring a space mode temperature specification for the enclosure, and yielding an unrealistic tem-
perature distribution. Use the VFSM command to correct this problem.
Menu Paths
To be provided at time of release.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1808
VFSM
*VFUN, ParR, Func, Par1, CON1, CON2, CON3
Performs a function on a single array parameter.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting numeric array parameter vector. See *SET for name restrictions.
Func
Function to be performed:
ACOS --
Arccosine: ACOS(Par1).
ASIN --
Arcsine: ASIN(Par1).
ASORT --
Par1 is sorted in ascending order. *VCOL, *VMASK, *VCUM, and *VLEN,,NINC do not apply.
*VLEN,NROW does apply.
ATAN --
Arctangent: ATAN(Par1).
COMP --
Compress: Selectively compresses data set. "True" (*VMASK) values of Par1 (or row positions to be
considered according to the NINC value on the *VLEN command) are written in compressed form
to ParR, starting at the specified position.
COPY --
Copy: Par1 copied to ParR.
COS --
Cosine: COS(Par1).
COSH --
Hyperbolic cosine: COSH(Par1).
DIRCOS --
Direction cosines of the principal stresses (nX9). Par1 contains the nX6 component stresses for the
n locations of the calculations.
DSORT --
Par1 is sorted in descending order. *VCOL, *VMASK, *VCUM, and *VLEN,,NINC do not apply.
*VLEN,NROW does apply.
EULER --
Euler angles of the principal stresses (nX3). Par1 contains the nX6 component stresses for the n
locations of the calculations.
EXP --
Exponential: EXP(Par1).
1809
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VFUN
EXPA --
Expand: Reverse of the COMP function. All elements of Par1 (starting at the position specified) are
written in expanded form to corresponding "true" (*VMASK) positions (or row positions to be con-
sidered according to the NINC value on the *VLEN command) of ParR.
LOG --
Natural logarithm: LOG(Par1).
LOG10 --
Common logarithm: LOG10(Par1).
NINT --
Nearest integer: 2.783 becomes 3.0, -1.75 becomes -2.0.
NOT --
Logical complement: values
0.0 (false) become 1.0 (true). Values > 0.0 (true) become 0.0 (false).
PRIN --
Principal stresses (nX5). Par1 contains the nX6 component stresses for the n locations of the calcu-
lations.
PWR --
Power function: Par1**CON1. Exponentiation of any negative number in the vector Par1 to a non-
integer power is performed by exponentiating the positive number and prepending the minus sign.
For example, -4**2.3 is -(4**2.3).
SIN --
Sine: SIN(Par1).
SINH --
Hyperbolic sine: SINH(Par1).
SQRT --
Square root: SQRT(Par1).
TAN --
Tangent: TAN(Par1).
TANH --
Hyperbolic tangent: TANH(Par1).
TANG --
Tangent to a path at a point: the slope at a point is determined by linear interpolation half way
between the previous and next points. Points are assumed to be in the global Cartesian coordinate
system. Path points are specified in array Par1 (having 3 consecutive columns of data, with the
columns containing the x, y, and z coordinate locations, respectively, of the points). Only the starting
row index and the column index for the x coordinates are specified, such as A(1,1). The y and z co-
ordinates of the vector are assumed to begin in the corresponding next columns, such as A(1,2) and
A(1,3). The tangent result, ParR, must also have 3 consecutive columns of data and will contain the
tangent direction vector (normalized to 1.0); such as 1,0,0 for an x-direction vector.
NORM --
Normal to a path and an input vector at a point: determined from the cross-product of the calculated
tangent vector (see TANG) and the input direction vector (with the i, j, and k components input as
CON1, CON2, and CON3). Points are assumed to be in the global Cartesian coordinate system. Path
points are specified in array Par1 (having 3 consecutive columns of data, with the columns containing
the x, y, and z coordinate locations, respectively, of the points). Only the starting row index and the
column index for the x coordinates are specified, such as A(1,1). The y and z coordinates of the vector
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1810
*VFUN
are assumed to begin in the corresponding next columns, such as A(1,2) and A(1,3). The normal
result, ParR, must also have 3 consecutive columns of data and will contain the normal direction
vector (normalized to 1.0); such as 1,0,0 for an x-direction vector.
LOCAL --
Transforms global Cartesian coordinates of a point to the coordinates of a specified system: points
to be transformed are specified in array Par1 (having 3 consecutive columns of data, with the
columns containing the x, y, and z global Cartesian coordinate locations, respectively, of the points).
Only the starting row index and the column index for the x coordinates are specified, such as A(1,1).
The y and z coordinates of the vector are assumed to begin in the corresponding next columns, such
as A(1,2) and A(1,3). Results are transformed to coordinate system CON1 (which may be any valid
coordinate system number, such as 1,2,11,12, etc.). The transformed result, ParR, must also have 3
consecutive columns of data and will contain the corresponding transformed coordinate locations.
GLOBAL --
Transforms specified coordinates of a point to global Cartesian coordinates: points to be transformed
are specified in array Par1 (having 3 consecutive columns of data, with the columns containing the
local coordinate locations (x, y, z or r, , z or etc.) of the points). Only the starting row index and the
column index for the x coordinates are specified, such as A(1,1). The y and z coordinates (or and
z, or etc.) of the vector are assumed to begin in the corresponding next columns, such as A(1,2) and
A(1,3). Local coordinate locations are assumed to be in coordinate system CON1 (which may be any
valid coordinate system number, such as 1,2,11,12, etc.). The transformed result, ParR, must also
have 3 consecutive columns of data, with the columns containing the global Cartesian x, y, and z
coordinate locations, respectively.
Par1
Array parameter vector in the operation.
CON1, CON2, CON3
Constants (used only with the PWR, NORM, LOCAL, and GLOBAL functions).
Notes
Operates on one input array parameter vector and produces one output array parameter vector according
to:
ParR = f(Par1)
where the functions (f ) are described below. Functions are based on the standard FORTRAN definitions where
possible. Out-of-range function results (or results with exponents whose magnitudes are approximately
greater than 32 or less than -32) produce a zero value. Input and output for angular functions may be radians
(default) or degrees [*AFUN]. ParR may be the same as Par1. Starting array element numbers must be
defined for each array parameter vector if it does not start at the first location. For example, *VFUN,A,SQRT,B(5)
takes the square root of the fifth element of B and stores the result in the first element of A. Operations
continue on successive array elements [*VLEN, *VMASK] with the default being all successive elements.
Absolute values and scale factors may be applied to all parameters [*VABS, *VFACT]. Results may be cumu-
lative [*VCUM]. Skipping array elements via *VMASK or *VLEN for the TANG and NORM functions skips
only the writing of the results (skipped array element data are used in all calculations). See the *VOPER
command for details.
This command is valid in any processor.
1811
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VFUN
Menu Paths
Utility Menu>Parameters>Array Operations>Vector Functions
VGEN, ITIME, NV1, NV2, NINC, DX, DY, DZ, KINC, NOELEM, IMOVE
Generates additional volumes from a pattern of volumes.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
ITIME
Do this generation operation a total of ITIMEs, incrementing all keypoints in the given pattern automat-
ically (or by KINC) each time after the first. ITIME must be > 1 for generation to occur.
NV1, NV2, NINC
Generate volumes from pattern beginning with NV1 to NV2 (defaults to NV1) in steps of NINC (defaults
to 1). If NV1 = ALL, NV2 and NINC are ignored and the pattern is all selected volumes [VSEL]. If NV1
= P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI).
A component name may also be substituted for NV1 (NV2 and NINC are ignored).
DX, DY, DZ
Keypoint location increments in the active coordinate system (--, D, DZ for cylindrical, --, D, -- for
spherical).
KINC
Keypoint increment between generated sets. If zero, the lowest available keypoint numbers are assigned
[NUMSTR].
NOELEM
Specifies if elements and nodes are also to be generated:
0
Generate nodes and elements associated with the original volumes, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether to redefine the existing volumes:
0
Generate additional volumes as requested with the ITIME argument.
1
Move original volumes to new position retaining the same keypoint line, and area numbers (ITIME,
KINC, and NOELEM are ignored). Corresponding meshed items are also moved if not needed at their
original position.
Notes
Generates additional volumes (and their corresponding keypoints, lines, areas and mesh) from a given volume
pattern. The MAT, TYPE, REAL, and ESYS attributes are based upon the volumes in the pattern and not upon
the current settings of the pointers. End slopes of the generated lines remain the same (in the active coordin-
ate system) as those of the given pattern. For example, radial slopes remain radial, etc. Generations which
produce volumes of a size or shape different from the pattern (i.e., radial generations in cylindrical systems,
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1812
VGEN
radial and phi generations in spherical systems, and theta generations in elliptical systems) are not allowed.
Note that solid modeling in a toroidal coordinate system is not recommended. Volume, area, and line
numbers are automatically assigned (beginning with the lowest available values [NUMSTR]).
Menu Paths
Main Menu>Preprocessor>Modeling>Copy>Volumes
Main Menu>Preprocessor>Modeling>Move / Modify>Volumes
*VGET, ParR, Entity, ENTNUM, Item1, IT1NUM, Item2, IT2NUM, KLOOP
Retrieves values and stores them into an array parameter.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting vector array parameter. See *SET for name restrictions.
Entity
Entity keyword. Valid keywords are NODE, ELEM, KP, LINE, AREA, VOLU, etc. as shown for Entity = in
the table below.
ENTNUM
The number of the entity (as shown for ENTNUM = in the table below).
Item1
The name of a particular item for the given entity. Valid items are as shown in the Item1 columns of
the table below.
IT1NUM
The number (or label) for the specified Item1 (if any). Valid IT1NUM values are as shown in the IT1NUM
columns of the table below. Some Item1 labels do not require an IT1NUM value.
Item2, IT2NUM
A second set of item labels and numbers to further qualify the item for which data is to be retrieved.
Most items do not require this level of information.
KLOOP
Field to be looped on:
0 or 2 --
Loop on the ENTNUM field (default).
3 --
Loop on the Item1 field.
4 --
Loop on the IT1NUM field. Successive items are as shown with IT1NUM.
5 --
Loop on the Item2 field.
6 --
Loop on the IT2NUM field. Successive items are as shown with IT2NUM.
1813
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VGET
Notes
Retrieves values for specified items and stores the values in an output vector of a user-named array parameter
according to:
ParR = f(Entity, ENTNUM, Item1, IT1NUM, Item2, IT2NUM)
where (f ) is the *GET function; Entity, Item1, and Item2 are keywords; and ENTNUM, IT1NUM, and
IT2NUM are numbers or labels corresponding to the keywords. Looping continues over successive entity
numbers (ENTNUM) for the KLOOP default. For example, *VGET,A(1),ELEM,5,CENT,X returns the centroid x-
location of element 5 and stores the result in the first location of A. Retrieving continues with element 6, 7,
8, etc., regardless of whether the element exists or is selected, until successive array locations are filled. Use
*VLEN or *VMASK to skip locations. Absolute values and scale factors may be applied to the result parameter
[*VABS, *VFACT]. Results may be cumulative [*VCUM]. See the *VOPER command for general details. Results
can be put back into an analysis by writing a file of the desired input commands with the *VWRITE command.
See also the *VPUT command.
Both *GET and *VGET retrieve information from the active data stored in memory. The database is often
the source, and sometimes the information is retrieved from common memory blocks that ANSYS uses to
manipulate information. Although POST1 and POST26 operations use a *.rst file, GET data is accessed
from the database or from the common blocks. Get operations do not access the *.rst file directly.
The *VGET command retrieves both the unprocessed real and the imaginary parts (original and duplicate
sector nodes and elements) of a cyclic symmetry solution.
This command is valid in any processor.
Table 264 *VGET - PREP7 Items
PREP7 Items
Entity = NODE, ENTNUM = n (node number)
Description IT1NUM Item1
X, Y, or Z location in the active coordinate system. X, Y, Z LOC
THXY, THYZ, THZX rotation angle. XY, YZ, ZX ANG
Select status of node n (-1 - unselected, 0 - undefined, 1 - selec-
ted).
NSEL
Returns the list of selected nodes (ENTNUM is ignored). NLIST
Entity = ELEM, ENTNUM = n (element number)
Description IT1NUM Item1
Node number at position 1,2,--20 of element n. 1,2,--20 NODE
Centroid X, Y, or Z location (based on shape function) in the
active coordinate system.
X, Y, Z CENT
Number of element adjacent to face 1, 2, -- 6. 1, 2, -- 6 ADJ
Number assigned to attribute name = MAT, TYPE, REAL, ESYS,
ENAM, or SECN).
name ATTR
Characteristic element geometry. Length of line element (straight
line between ends), area of area element, or volume of volume
GEOM
element. Issuing *VGET for an element returns a signed value.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1814
*VGET
PREP7 Items
To always get a positive value, issue *VABS,1 just prior to issuing
*VGET,par(n),ELEM,x,GEOM.
Select status of element n (-1 - unselected, 0 - undefined, 1 -
selected).
ESEL
Element shape test result for selected element n, where
Test=ANGD (SHELL28 corner angle deviation), ASPE (aspect
Test SHPAR
ratio), JACR (Jacobian ratio), MAXA (maximum corner angle),
PARA (deviation from parallelism of opposite edges), or WARP
(warping factor).
Returns the list of selected elements (ENTNUM is ignored). ELIST
Entity = KP, ENTNUM = n (keypoint number)
Description IT1NUM Item1
X, Y, or Z location in the active coordinate system. X, Y, Z LOC
Number assigned to attribute (name = MAT, TYPE, REAL, ESYS,
NODE or ELEM).
name ATTR
Divisions (element size setting) from KESIZE command. DIV
Select status of keypoint n (-1 - unselected, 0 - undefined, 1 -
selected).
KSEL
Returns the list of selected keypoints (ENTNUM is ignored). KLIST
Entity = LINE, ENTNUM = n (line number)
Description IT1NUM Item1
Keypoint number at position 1 or 2. 1,2 KP
Number assigned to attribute (name = MAT, TYPE, REAL, ESYS,
NNOD, NELM, or NDIV). NNOD = number of nodes, NELM =
number of elements, NDIV = number of divisions.
name ATTR
Length. LENG
Select status of line n (-1 - unselected, 0 - undefined, 1 - selec-
ted).
LSEL
Returns the list of selected lines (ENTNUM is ignored). LLIST
Entity = AREA, ENTNUM = n (area number)
Description IT1NUM Item1
Loop number. Must be input if LINE number is to be retrieved. 1,2,-- I LOOP
Description IT2NUM Item2
Line number at position 1, 2, --- p. 1, 2, -- p LINE
Number assigned to attribute (name = MAT, TYPE, REAL, ESYS,
SECN, NNOD, or NELM). NNOD = number of nodes, NELM =
number of elements.
name ATTR
Area (after last ASUM). AREA
Select status of area n (-1 - unselected, 0 - undefined, 1 - selec-
ted).
ASEL
1815
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VGET
PREP7 Items
Returns the list of selected areas (ENTNUM is ignored). ALIST
Entity = VOLU, ENTNUM = n (volume number)
Description IT1NUM Item1
Shell number. Must be input if AREA number is to be retrieved. 1,2,--l SHELL
Description IT2NUM Item2
Area number at position 1, 2, --- p. 1, 2, -- p AREA
Number assigned to attribute (name = MAT, TYPE, REAL, ESYS,
NNOD, or NELM). NNOD = number of nodes, NELM = number
of elements.
name ATTR
Volume (after last VSUM). VOLU
Select status of volume n (-1 - unselected, 0 - undefined, 1 -
selected).
VSEL
Returns the list of selected volumes (ENTNUM is ignored). VLIST
Entity = CDSY, ENTNUM = n (coordinate system number)
Description IT1NUM Item1
X, Y, or Z origin location (global Cartesian coordinate). X, Y, Z LOC
THXY, THYZ, or THZX rotation angle () relative to the global
Cartesian coordinate system.
XY, YZ, ZX ANG
Number assigned to attribute (name = KCS, KTHET, KPHI, PAR1,
or PAR2). A -1.0 is returned for KCS if coordinate system is un-
defined).
name ATTR
Entity = RCON, ENTNUM = n (real constant set number)
Description IT1NUM Item1
Real constant value for constant 1,2,--- m. 1,2,--m CONST
Entity = TLAB, ENTNUM = n (Tlab is the data table label: BKIN, MKIN, MISO, etc. as
described on the TB command. n is the material number.)
Description IT1NUM Item1
Temperature value (if any) at which to retrieve table data. val TEMP
Description IT2NUM Item2
Constant number whose value is to be retrieved (see Data Tables
- Implicit Analysis in the Element Reference). For constants input
num CONST
as X, Y points, the constant numbers are consecutive with the
X constants being the odd numbers, beginning with one.
Table 265 *VGET - POST1 Items
Entity = NODE, ENTNUM = n (node number)
Vector items are in the active results coordinate system unless otherwise specified.
Description IT1NUM Item1
Valid labels for nodal degree of freedom results are:
X, Y, or Z structural displacement. X, Y, Z U
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1816
*VGET
Entity = NODE, ENTNUM = n (node number)
Vector items are in the active results coordinate system unless otherwise specified.
Description IT1NUM Item1
X, Y, or Z structural rotation. X, Y, Z ROT
Temperature. For SHELL131 and SHELL132 elements with
KEYOPT(3) = 0 or 1, use TBOT, TE2, TE3, . . ., TTOP instead
TEMP
of TEMP. Alternative get functions: TEMP(N), TBOT(N),
TE2(N), etc.
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity. X, Y, Z V
X, Y, or Z magnetic vector potential. X, Y, Z A
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy (FLOTRAN). ENKE
Turbulent energy dissipation (FLOTRAN). ENDS
Valid labels for element nodal results are:
Description IT1NUM Item1
Component stress. X, Y, Z, XY, YZ,
XZ
S
Principal stress. 1,2,3 "
Stress intensity or equivalent stress. INT, EQV "
Component total strain (EPEL + EPPL + EPCR). X, Y, Z, XY, YZ,
XZ
EPTO
Principal total strain. 1, 2, 3 "
Total strain intensity or total equivalent strain. INT, EQV "
Component elastic strain. X, Y, Z, XY, YZ,
XZ
EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity or elastic equivalent strain. INT, EQV "
Component plastic strain. X, Y, Z, XY, YZ,
XZ
EPPL
Principal plastic strain. 1, 2, 3 "
Plastic strain intensity or plastic equivalent strain. INT, EQV "
Component creep strain. X, Y, Z, XY, YZ,
XZ
EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity or creep equivalent strain. INT, EQV "
Component thermal strain. X, Y, Z, XY, YZ,
XZ
EPTH
1817
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VGET
Entity = NODE, ENTNUM = n (node number)
Vector items are in the active results coordinate system unless otherwise specified.
Description IT1NUM Item1
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity or thermal equivalent strain. INT, EQV "
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Component magnetic field intensity from current sources
(in the global Cartesian coordinate system).
X, Y, Z HS
Body temperatures (calculated from applied temperatures)
as used in solution.
TEMP BFE
Component thermal gradient and sum. X, Y, Z, SUM TG
Component thermal flux and sum. X, Y, Z, SUM TF
Component pressure gradient and sum. X, Y, Z, SUM PG
Component electric field and sum. X, Y, Z, SUM EF
Component electric flux density and sum. X, Y, Z, SUM D
Component magnetic field intensity and sum. X, Y, Z, SUM H
Component magnetic flux density and sum. X, Y, Z, SUM B
Component electromagnetic force and sum. X, Y, Z, SUM FMAG
Note
Element nodal results are the average nodal value of the selected elements.
Valid labels for FLOTRAN nodal results are:
Description IT1NUM Item1
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1818
*VGET
Entity = NODE, ENTNUM = n (node number)
Vector items are in the active results coordinate system unless otherwise specified.
Description IT1NUM Item1
Fluid effective viscosity. EVIS
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
Entity = ELEM, ENTNUM = n (element number)
Valid labels for element results are:
Any user-defined element table label (see ETABLE com-
mand).
Lab ETAB
Menu Paths
Utility Menu>Parameters>Get Array Data
VGET, Par, IR, TSTRT, KCPLX
Moves a variable into an array parameter vector.
POST26: Special Purpose
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Par
Array parameter vector in the operation.
IR
Reference number of the variable (1 to NV [NUMVAR]).
TSTRT
Time (or frequency) corresponding to start of IR data. If between values, the nearer value is used.
KCPLX
Complex number key:
0
Use the real part of the IR data.
1
Use the imaginary part of the IR data.
Notes
Moves a variable into an array parameter vector. The starting array element number must be defined. For
example, VGET,A(1),2 moves variable 2 (starting at time 0.0) to array parameter A. Looping continues from
array element A(1) with the index number incremented by one until the variable is filled. The number of
loops may be controlled with the *VLEN command (except that loop skipping (NINC) is not allowed). For
multi-dimensioned array parameters, only the first (row) subscript is incremented.
1819
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VGET
Menu Paths
Main Menu>TimeHist Postpro>Table Operations>Variable to Par
VGLUE, NV1, NV2, NV3, NV4, NV5, NV6, NV7, NV8, NV9
Generates new volumes by "gluing" volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NV3, . . . , NV9
Numbers of the volumes to be glued. If NV1 = ALL, all selected volumes will be glued (NV2 to NV9 will
be ignored). If NV1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NV1.
Notes
Use of the VGLUE command generates new volumes by "gluing" input volumes. The glue operation redefines
the input volumes so that they share areas along their common boundaries. The new volumes encompass
the same geometry as the original volumes. This operation is only valid if the intersections of the input
volumes are areas along the boundaries of those volumes. See the Modeling and Meshing Guide for an illus-
tration. See the BOPTN command for an explanation of the options available to Boolean operations. Element
attributes and solid model boundary conditions assigned to the original entities will not be transferred to
the new entities generated.
The VGLUE command results in the merging of areas, lines, and keypoints at the common volume bound-
aries. The areas, lines, and keypoints of the lower numbered volume will be kept. This means one must be
aware of volume numbering when multiple VGLUE commands are applied to avoid any "ungluing" of geo-
metry.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Glue>Volumes
/VIEW, WN, XV, YV, ZV
Defines the viewing direction for the display.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
XV, YV, ZV
The object is viewed along the line from point XV,YV,ZV (in the global coordinate system) to the global
coordinate system origin. For section displays, the cutting plane is assumed to be perpendicular to this
line. If XV = WP, modify view to be normal to the currently defined working plane. Defaults to (0,0,1).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1820
VGLUE
Command Default
0,0,1 view. The default reference orientation is X-axis horizontal to the right, Y-axis vertical upward, and Z-
axis out from the screen (normal). See /VUP command to change reference orientation.
Notes
The view line is always normal to the screen. The view is selected by defining a point (in the global Cartesian
coordinate system) representing a point along the viewing line. This point, and the global Cartesian coordinate
system origin, define the line along which the object is viewed while looking toward the origin. Any point
along the view line may be used, i.e., (1,1,1) and (2,2,2) give the same view. The display orientation may be
changed as desired [/ANGLE]. The display coordinate system type may be changed (from Cartesian to cyl-
indrical, spherical, toroidal, etc.) with the DSYS command.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>Path Operations>Define Path>On Working Plane
Main Menu>Preprocessor>Path Operations>Define Path>On Working Plane
Utility Menu>PlotCtrls>Pan, Zoom, Rotate
Utility Menu>PlotCtrls>View Settings>Viewing Direction
VIMP, VOL, CHGBND, IMPLEVEL
Improves the quality of the tetrahedral elements in the selected volume(s).
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
VOL
Number of the volume containing the tetrahedral elements to be improved. If VOL = ALL (default), improve
the tetrahedral elements in all selected volumes. If VOL = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI). A component name may also be substituted for VOL.
CHGBND
Specifies whether to allow boundary modification. Boundary modification includes such things as changes
in the connectivity of the element faces on the boundary and the addition of boundary nodes. (Also see
"Notes" (p. 1822) below for important usage information for CHGBND.)
0
Do not allow boundary modification.
1
Allow boundary modification (default).
IMPLEVEL
Identifies the level of improvement to be performed on the elements. (Improvement occurs primarily
through the use of face swapping and node smoothing techniques.)
0
Perform the least amount of swapping/smoothing.
1821
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VIMP
1
Perform an intermediate amount of swapping/smoothing.
2
Perform the greatest amount of swapping/smoothing.
3
Perform the greatest amount of swapping/smoothing, plus additional improvement techniques
(default).
Notes
VIMP is useful for further improving a volume mesh created in ANSYS [VMESH], especially quadratic tetra-
hedral element meshes.
The VIMP command enables you to improve a given tetrahedral mesh by reducing the number of poorly-
shaped tetrahedral elements (in particular, the number of sliver tetrahedral elements)--as well as the overall
number of elements--in the mesh. It also improves the overall quality of the mesh.
Regardless of the value of the CHGBND argument, boundary mid-nodes can be moved as long as you are
not using p-method analysis. When CHGBND = 0 and you are using p-method analysis, boundary mid-nodes
cannot be moved. (ANSYS issues an error message if it would be necessary to move boundary mid-nodes
in order to generate valid quadratic elements.)
When loads or constraints have been placed on boundary nodes or mid-nodes, and boundary mid-nodes
are later moved, ANSYS issues a warning message to let you know that it will not update the loads or con-
straints.
Even when CHGBND = 1, no boundary modification is performed on areas and lines that are not modifiable
(for example, areas that are adjacent to other volumes or that contain shell elements, or lines that are not
incident on modifiable areas, contain beam elements, or have line divisions specified for them [LESIZE]).
Menu Paths
Main Menu>Preprocessor>Meshing>Modify Mesh>Improve Tets>Volumes
VINP, NV1, NV2, NV3, NV4, NV5, NV6, NV7, NV8, NV9
Finds the pairwise intersection of volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NV3, . . . , NV9
Numbers of volumes to be intersected pairwise. If NV1 = ALL, NV2 to NV9 are ignored and the pairwise
intersection of all selected volumes is found. If NV1 = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI). A component name may also be substituted for NV1.
Notes
Finds the pairwise intersection of volumes. The pairwise intersection is defined as all regions shared by any
two or more volumes listed on this command. New volumes will be generated where the original volumes
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1822
VINP
intersect pairwise. If the regions of pairwise intersection are only areas, new areas will be generated. See
the Modeling and Meshing Guide for an illustration. See the BOPTN command for an explanation of the options
available to Boolean operations. Element attributes and solid model boundary conditions assigned to the
original entities will not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Intersect>Pairwise>Volumes
VINV, NV1, NV2, NV3, NV4, NV5, NV6, NV7, NV8, NV9
Finds the intersection of volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NV3, . . . , NV9
Numbers of volumes to be intersected. If NV1 = ALL, NV2 to NV9 are ignored, and the intersection of
all selected volumes is found. If NV1 = P, graphical picking is enabled and all remaining command fields
are ignored (valid only in the GUI). A component name may also be substituted for NV1.
Notes
Finds the common (not pairwise) intersection of volumes. The common intersection is defined as the regions
shared (in common) by all volumes listed on this command. New volumes will be generated where the ori-
ginal volumes intersect. If the regions of intersection are only areas, new areas will be generated instead.
See the Modeling and Meshing Guide for an illustration. See the BOPTN command for an explanation of the
options available to Boolean operations. Element attributes and solid model boundary conditions assigned
to the original entities will not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Intersect>Common>Volumes
*VITRP, ParR, ParT, ParI, ParJ, ParK
Forms an array parameter by interpolation of a table.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting array parameter. See *SET for name restrictions.
ParT
The name of the TABLE array parameter. The parameter must exist as a dimensioned array of type TABLE
[*DIM].
1823
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VITRP
ParI
Array parameter vector of I (row) index values for interpolation in ParT.
ParJ
Array parameter vector of J (column) index values for interpolation in ParT (which must be at least 2-
D).
ParK
Array parameter vector of K (depth) index values for interpolation in ParT (which must be 3-D).
Notes
Forms an array parameter (of type ARRAY) by interpolating values of an array parameter (of type TABLE) at
specified table index locations according to:
ParR = f(ParT, Parl, ParJ, ParK)
where ParT is the type TABLE array parameter, and ParI, ParJ, ParK are the type ARRAY array parameter
vectors of index values for interpolation in ParT. See the *DIM command for TABLE and ARRAY declaration
types. Linear interpolation is used. The starting array element number for the TABLE array (ParT) is not used
(but a value must be input). Starting array element numbers must be defined for each array parameter
vector if it does not start at the first location. For example, *VITRP,R(5),TAB(1,1),X(2),Y(4) uses the second
element of X and the fourth element of Y as index values (row and column) for a 2-D interpolation in TAB
and stores the result in the fifth element of R. Operations continue on successive array elements [*VLEN,
*VMASK] with the default being all successive elements. Absolute values and scale factors may be applied
to the result parameter [*VABS, *VFACT]. Results may be cumulative [*VCUM]. See the *VOPER command
for details.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Vector Interpolate
*VLEN, NROW, NINC
Specifies the number of rows to be used in array parameter operations.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
NROW
Number of rows to be used with the *VXX or *MXX operations. Defaults to the number of rows needed
to fill the result array.
NINC
Perform the operation on every NINC row (defaults to 1).
Command Default
Fill all locations of the result array from the specified starting location.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1824
*VLEN
Notes
Specifies the number of rows to be used in array parameter operations. The size of the submatrix used is
determined from the upper left starting array element (defined on the operation command) to the lower
right array element (defined by the number of rows on this command and the number of columns on the
*VCOL command). NINC allows skipping row operations for some operation commands. Skipped rows are
included in the row count. The starting row number must be defined on the operation command for each
parameter read and for the result written.
The default NROW is calculated from the maximum number of rows of the result array (the *DIM row dimen-
sion) minus the starting location + 1. For example, *DIM,R,,10 and a starting location of R(7) gives a default
of 4 loops (filling R(7), R(8), R(9), and R(10)). Repeat operations automatically terminate at the last row of the
result array. Existing values in the rows and columns of the results matrix remain unchanged where not
overwritten by the requested input or operation values.
The stride (NINC) allows operations to be performed at regular intervals. It has no effect on the total number
of row operations. Skipped operations retain the previous result. For example, *DIM,R,,6, with a starting
location of R(1), NROW = 10, and NINC = 2 calculates values for locations R(1), R(3), and R(5) and retains
values for locations R(2), R(4), and R(6). A more general skip control may be done by masking [*VMASK].
The row control settings are reset to the defaults after each *VXX or *MXX operation. Use *VSTAT to list
settings.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Operation Settings
VLIST, NV1, NV2, NINC
Lists the defined volumes.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
List volumes from NV1 to NV2 (defaults to NV1) in steps of NINC (defaults to 1). If NV1 = ALL (default),
NV2 and NINC are ignored and all selected volumes [VSEL] are listed. If NV1 = P, graphical picking is
enabled and all remaining command fields are ignored (valid only in the GUI). A component name may
also be substituted for NV1 (NV2 and NINC are ignored).
Notes
An attribute (TYPE, MAT, REAL, or ESYS) listed as a zero is unassigned; one listed as a positive value indicates
that the attribute was assigned with the VATT command (and will not be reset to zero if the mesh is cleared);
one listed as a negative value indicates that the attribute was assigned using the attribute pointer [TYPE,
MAT, REAL, or ESYS] that was active during meshing (and will be reset to zero if the mesh is cleared). A "-1"
in the "nodes" column indicates that the volume has been meshed but there are no interior nodes. The
volume size is listed only if a VSUM command has been performed on the volume. Volume orientation at-
tributes (KZ1 and KZ2) are listed only if a VEORIENT command was previously used to define an orientation
for the volume.
1825
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VLIST
This command is valid in any processor.
Menu Paths
Utility Menu>List>Volumes
VLSCALE, NV1, NV2, NINC, RX, RY, RZ, KINC, NOELEM, IMOVE
Generates a scaled set of volumes from a pattern of volumes.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
Set of volumes (NV1 to NV2 in steps of NINC) that defines the pattern to be scaled. NV2 defaults to
NV1, NINC defaults to 1. If NV1 = ALL, NV2 and NINC are ignored and the pattern is defined by all se-
lected volumes. If NV1 = P, graphical picking is enabled and all remaining command fields are ignored
(valid only in the GUI). A component name may also be substituted for NV1 (NV2 and NINC are ignored).
RX, RY, RZ
Scale factors to be applied to the X, Y, and Z keypoint coordinates in active coordinate system (RR, R,
RZ for cylindrical; RR, R, R for spherical). Note that the R and R scale factors are interpreted as angular
offsets. For example, if CSYS = 1, RX, RY, RZ input of (1.5,10,3) would scale the specified keypoints 1.5
times in the radial and 3 times in the Z direction, while adding an offset of 10 degrees to the keypoints.
Zero, blank, or negative scale factor values are assumed to be 1.0. Zero or blank angular offsets have no
effect.
KINC
Increment to be applied to keypoint numbers for generated set. If zero, the lowest available keypoint
numbers will be assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Nodes and elements associated with the original volumes will be generated (scaled) if they exist.
1
Nodes and elements will not be generated.
IMOVE
Specifies whether volumes will be moved or newly defined:
0
Additional volumes will be generated.
1
Original volumes will be moved to new position (KINC and NOELEM are ignored). Use only if the
old volumes are no longer needed at their original positions. Corresponding meshed items are also
moved if not needed at their original position.
Notes
Generates a scaled set of volumes (and their corresponding keypoints, lines, areas, and mesh) from a pattern
of volumes. The MAT, TYPE, REAL, and ESYS attributes are based on the volumes in the pattern and not the
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1826
VLSCALE
current settings. Scaling is done in the active coordinate system. Volumes in the pattern could have been
generated in any coordinate system. However, solid modeling in a toroidal coordinate system is not recom-
mended.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Scale>Volumes
*VMASK, Par
Specifies an array parameter as a masking vector.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par
Name of the mask parameter. The starting subscript must also be specified.
Command Default
No mask parameter specified (use true for all operations).
Notes
Specifies the name of the parameter whose values are to be checked for each resulting row operation. The
mask vector usually contains only 0 (for false) and 1 (for true) values. For each row operation the corresponding
mask vector value is checked. A true value allows the operation to be done. A false value skips the operation
(and retains the previous results). A mask vector can be created from direct input, such as M(1) = 1,0,0,1,1,0,1;
or from the DATA function of the *VFILL command. The NOT function of the *VFUN command can be used
to reverse the logical sense of the mask vector. The logical compare operations (LT, LE, EQ, NE, GE, and GT)
of the *VOPER command also produce a mask vector by operating on two other vectors. Any numeric
vector can be used as a mask vector since the actual interpretation assumes values less than 0.0 are 0.0
(false) and values greater than 0.0 are 1.0 (true). If the mask vector is not specified (or has fewer values than
the result vector), true (1.0) values are assumed for the unspecified values. Another skip control may be input
with NINC on the *VLEN command. If both are present, operations occur only when both are true. The
mask setting is reset to the default (no mask) after each *VXX or *MXX operation. Use *VSTAT to list settings.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Operation Settings
1827
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VMASK
VMESH, NV1, NV2, NINC
Generates nodes and volume elements within volumes.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
Mesh volumes from NV1 to NV2 (defaults to NV1) in steps of NINC (defaults to 1). If NV1 = ALL, NV2
and NINC are ignored and all selected volumes [VSEL] are meshed. If NV1 = P, graphical picking is enabled
and all remaining command fields are ignored (valid only in the GUI). A component name may also be
substituted for NV1 (NV2 and NINC are ignored).
Notes
Missing nodes required for the generated elements are created and assigned the lowest available numbers
[NUMSTR]. During a batch run and if elements already exist, a mesh abort will write an alternative database
file (File.DBE) for possible recovery.
Tetrahedral mesh expansion [MOPT,TETEXPND,Value] is supported for both the VMESH and FVMESH
commands.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Volumes>Free
Main Menu>Preprocessor>Meshing>Mesh>Volumes>Mapped>4 to 6 sided
VOFFST, NAREA, DIST, KINC
Generates a volume, offset from a given area.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NAREA
Area from which generated volume is to be offset. If NAREA = P, graphical picking is enabled and all
remaining command fields are ignored (valid only in the GUI).
DIST
Distance normal to given area at which keypoints for generated volume are to be located. Positive normal
is determined from the right-hand rule keypoint order.
KINC
Increment to be applied to the keypoint numbers between sets. If zero, keypoint numbers will be auto-
matically assigned beginning with the lowest available value [NUMSTR].
Notes
Generates a volume (and its corresponding keypoints, lines, and areas) by offsetting from an area. The direction
of the offset varies with the given area normal. End slopes of the generated lines remain the same as those
of the given pattern.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1828
VMESH
If element attributes have been associated with the input area via the AATT command, the opposite area
generated by the VOFFST operation will also have those attributes (i.e., the element attributes from the input
area are copied to the opposite area). Note that only the area opposite the input area will have the same
attributes as the input area; the areas adjacent to the input area will not.
If the areas are meshed or belong to meshed volumes, a 3-D mesh can be extruded with this command.
Note that the NDIV argument on the ESIZE command should be set before extruding the meshed areas.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extrude>Areas>Along Normal
VOLUMES
Specifies "Volumes" as the subsequent status topic.
PREP7: Status
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
This is a status [STAT] topic command. Status topic commands are generated by the GUI and appear in the
log file (Jobname.LOG) if status is requested for some items by choosing Utility Menu> List> Status. This
command will be immediately followed by a STAT command, which will report the status for the specified
topic.
If entered directly into the program, the STAT command should immediately follow this command.
Menu Paths
This command cannot be accessed from a menu.
*VOPER, ParR, Par1, Oper, Par2, CON1, CON2
Operates on two array parameters.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting array parameter vector. See *SET for name restrictions.
Par1
First array parameter vector in the operation. May also be a scalar parameter or a literal constant.
Oper
Operations:
ADD--
Addition: Par1+Par2.
1829
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VOPER
SUB --
Subtraction: Par1-Par2.
MULT --
Multiplication: Par1*Par2.
DIV --
Division: Par1/Par2 (a divide by zero results in a value of zero).
MIN --
Minimum: minimum of Par1 and Par2.
MAX --
Maximum: maximum of Par1 and Par2.
LT --
Less than comparison: Par1<Par2 gives 1.0 if true, 0.0 if false.
LE --
Less than or equal comparison: Par1
Par2 gives 1.0 if true, 0.0 if false.
EQ --
Equal comparison: Par1 = Par2 gives 1.0 if true, 0.0 if false.
NE --
Not equal comparison: Par1 Par2 gives 1.0 if true, 0.0 if false.
GE --
Greater than or equal comparison: Par1
Par2 gives 1.0 if true, 0.0 if false.
GT --
Greater than comparison: Par1>Par2 gives 1.0 if true, 0.0 if false.
DER1 --
First derivative: d(Par1)/d(Par2). The derivative at a point is determined over points half way
between the previous and next points (by linear interpolation). Par1 must be a function (a unique
Par1 value for each Par2 value) and Par2 must be in ascending order.
DER2 --
Second derivative: d
2
(Par1)/d(Par2)
2
. See also DER1.
INT1 --
Single integral: Par1 d(Par2), where CON1 is the integration constant. The integral at a point is
determined by using the single integration procedure described in the Theory Reference for the
Mechanical APDL and Mechanical Applications.
INT2 --
Double integral: Par1 d(Par2), where CON1 is the integration constant of the first integral and
CON2 is the integration constant of the second integral. If Par1 contains acceleration data, CON1
is the initial velocity and CON2 is the initial displacement. See also INT1.
DOT --
Dot product: Par1 . Par2. Par1 and Par2 must each have three consecutive columns of data,
with the columns containing the i, j, and k vector components, respectively. Only the starting row
index and the column index for the i components are specified for Par1 and Par2, such as A(1,1).
The j and k components of the vector are assumed to begin in the corresponding next columns,
such as A(1,2) and A(1,3).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1830
*VOPER
CROSS --
Cross product: Par1 x Par2. Par1, Par2, and ParR must each have 3 components, respectively.
Only the starting row index and the column index for the i components are specified for Par1,
Par2, and ParR, such as A(1,1). The j and k components of the vector are assumed to begin in the
corresponding next columns, such as A(1,2) and A(1,3).
GATH --
Gather: For a vector of position numbers, Par2, copy the value of Par1 at each position number
to ParR. Example: for Par1 = 10,20,30,40 and Par2 = 2,4,1; ParR = 20,40,10.
SCAT --
Scatter: Opposite of GATH operation. For a vector of position numbers, Par2, copy the value of
Par1 to that position number in ParR. Example: for Par1 = 10,20,30,40,50 and Par2 = 2,1,0,5,3;
ParR = 20,10,50,0,40.
ATN2 --
Arctangent: arctangent of Par1/Par2 with the sign of each component considered.
Par2
Second array parameter vector in the operation. May also be a scalar parameter or a literal constant.
CON1
First constant (used only with the INT1 and INT2 operations).
CON2
Second constant (used only with the INT2 operation).
Notes
Operates on two input array parameter vectors and produces one output array parameter vector according
to:
ParR = Par1 o Par2
where the operations (o) are described below. ParR may be the same as Par1 or Par2. Absolute values
and scale factors may be applied to all parameters [*VABS, *VFACT]. Results may be cumulative [*VCUM].
Starting array element numbers must be defined for each array parameter vector if it does not start at the
first location, such as *VOPER,A,B(5),ADD,C(3) which adds the third element of C to the fifth element of B
and stores the result in the first element of A. Operations continue on successive array elements [*VLEN,
*VMASK] with the default being all successive elements. Skipping array elements via *VMASK or *VLEN for
the DER_ and INT_ functions skips only the writing of the results (skipped array element data are used in
all calculations).
Parameter functions and operations are available to operate on a scalar parameter or a single element of
an array parameter, such as SQRT(B) or SQRT(A(4)). See the *SET command for details. Operations on a se-
quence of array elements can be done by repeating the desired function or operation in a do-loop [*DO].
The vector operations within the ANSYS program (*VXX commands) are internally programmed do-loops
that conveniently perform the indicated operation over a sequence of array elements. If the array is multidi-
mensional, only the first subscript is incremented in the do-loop, that is, the operation repeats in column
vector fashion "down" the array. For example, for A(1,5), A(2,5), A(3,5), etc. The starting location of the row
index must be defined for each parameter read and for the result written.
The default number of loops is from the starting result location to the last result location and can be altered
with the *VLEN command. A logical mask vector may be defined to control at which locations the operations
are to be skipped [*VMASK]. The default is to skip no locations. Repeat operations automatically terminate
1831
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VOPER
at the last array element of the result array column if the number of loops is undefined or if it exceeds the
last result array element. Zeroes are used in operations for values read beyond the last array element of an
input array column. Existing values in the rows and columns of the results matrix remain unchanged where
not changed by the requested operation values. The result array column may be the same as the input array
column since results in progress are stored in a temporary array until being moved to the results array at
the end of the operation. Results may be overwritten or accumulated with the existing results [*VCUM]. The
default is to overwrite results. The absolute value may be used for each parameter read or written [*VABS].
A scale factor (defaulting to 1.0) is also applied to each parameter read and written [*VFACT].
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Vector Operations
VOVLAP, NV1, NV2, NV3, NV4, NV5, NV6, NV7, NV8, NV9
Overlaps volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NV3, . . . , NV9
Numbers of volumes to be operated on. If NV1 = ALL, NV2 to NV9 are ignored and all selected volumes
are used. If NV1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NV1.
Notes
Overlaps volumes. Generates new volumes which encompass the geometry of all the input volumes. The
new volumes are defined by the regions of intersection of the input volumes, and by the complementary
(non-intersecting) regions. See the Modeling and Meshing Guide for an illustration. This operation is only
valid when the region of intersection is a volume. See the BOPTN command for an explanation of the options
available to Boolean operations. Element attributes and solid model boundary conditions assigned to the
original entities will not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Overlap>Volumes
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1832
VOVLAP
*VPLOT, ParX, ParY, Y2, Y3, Y4, Y5, Y6, Y7, Y8
Graphs columns (vectors) of array parameters.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParX
Name of the array parameter whose column vector values will be the abscissa of the graph. If blank, row
subscript numbers are used instead. ParX is not sorted by the program.
ParY
Name of the array parameter whose column vector values will be graphed against the ParX values.
Y2, Y3, Y4, . . . , Y8
Additional column subscript of the ParY array parameter whose values are to be graphed against the
ParX values.
Notes
The column to be graphed and the starting row for each array parameter must be specified as subscripts.
Additional columns of the ParY array parameter may be graphed by specifying column numbers for Y2,
Y3, ...,Y8. For example, *VPLOT,TIME (4,6), DISP (8,1),2,3 specifies that the 1st, 2nd, and 3rd columns of array
parameter DISP (all starting at row 8) are to be graphed against the 6th column of array parameter TIME
(starting at row 4). The columns will be graphed from the starting row to their maximum extent. See the
*VLEN and *VMASK commands to limit or skip data to be graphed. The array parameters specified on the
*VPLOT command must be of the same type (type ARRAY or TABLE; [*DIM]. Arrays of type TABLE will be
graphed as continuous curves. Arrays of type ARRAY will be displayed in bar chart fashion.
The normal curve labeling scheme for *VPLOT is to label curve 1 COL 1, curve 2 COL 2 and so on. You
can use the /GCOLUMN command to apply user-specifed labels (8 characters maximum) to your curves. See
Modifying Curve Labels in the ANSYS Parametric Design Language Guide for more information on using
/GCOLUMN.
When a graph plot reaches minimum or maximum y-axis limits, ANSYS indicates the condition by clipping
the graph. The clip appears as a horizontal magenta line. ANSYS calculates y-axis limits automatically; however,
you can modify the (YMIN and YMAX) limits via the /YRANGE command.
This command is valid in any processor.
Menu Paths
Utility Menu>Plot>Array Parameters
1833
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VPLOT
VPLOT, NV1, NV2, NINC, DEGEN, SCALE
Displays the selected volumes.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NINC
Display volumes from NV1 to NV2 (defaults to NV1) in steps of NINC (defaults to 1). If NV1 = ALL (default),
NV2 and NINC are ignored and all selected volumes [VSEL] are displayed.
DEGEN
Degeneracy marker:
(blank)
No degeneracy marker is used (default).
DEGE
A red star is placed on keypoints at degeneracies (see the Modeling and Meshing Guide). Not available
if /FACET,WIRE is set.
SCALE
Scale factor for the size of the degeneracy-marker star. The scale is the size in window space (-1 to 1 in
both directions) (defaults to .075).
Notes
Displays selected volumes. (Only volumes having areas within the selected area set [ASEL] will be plotted.)
With PowerGraphics on [/GRAPHICS,POWER], VPLOT will display only the currently selected areas. This
command is also a utility command, valid anywhere. The degree of tessellation used to plot the volumes is
set through the /FACET command.
Menu Paths
Main Menu>Preprocessor>Modeling>Check Geom>Show Degeneracy>Plot Degen Volus
Main Menu>Preprocessor>Modeling>Operate>Booleans>Show Degeneracy>Plot Degen Volus
Utility Menu>Plot>Specified Entities>Volumes
Utility Menu>Plot>Volumes
VPTN, NV1, NV2, NV3, NV4, NV5, NV6, NV7, NV8, NV9
Partitions volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1, NV2, NV3, . . . , NV9
Numbers of volumes to be operated on. If NV1 = ALL, NV2 to NV9 are ignored and all selected volumes
are used. If NV1 = P, graphical picking is enabled and all remaining command fields are ignored (valid
only in the GUI). A component name may also be substituted for NV1.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1834
VPLOT
Notes
Partitions volumes. Generates new volumes which encompass the geometry of all the input volumes. The
new volumes are defined by the regions of intersection of the input volumes, and by the complementary
(non-intersecting) regions. See the Modeling and Meshing Guide for an illustration. See the BOPTN command
for an explanation of the options available to Boolean operations. Element attributes and solid model
boundary conditions assigned to the original entities will not be transferred to the new entities generated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Partition>Volumes
*VPUT, ParR, Entity, ENTNUM, Item1, IT1NUM, Item2, IT2NUM, KLOOP
Restores array parameter values into the ANSYS database.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the input vector array parameter. See *SET for name restrictions. The parameter must exist
as a dimensioned array [*DIM] with data input.
Entity
Entity keyword. Valid keywords are shown for Entity = in the table below.
ENTNUM
The number of the entity (as shown for ENTNUM= in the table below).
Item1
The name of a particular item for the given entity. Valid items are as shown in the Item1 columns of
the table below.
IT1NUM
The number (or label) for the specified Item1 (if any). Valid IT1NUM values are as shown in the IT1NUM
columns of the table below. Some Item1 labels do not require an IT1NUM value.
Item2, IT2NUM
A second set of item labels and numbers to further qualify the item for which data is to be stored. Most
items do not require this level of information.
KLOOP
Field to be looped on:
0 or 2 --
Loop on the ENTNUM field (default).
3 --
Loop on the Item1 field.
4 --
Loop on the IT1NUM field. Successive items are as shown with IT1NUM.
1835
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VPUT
5 --
Loop on the Item2 field.
6 --
Loop on the IT2NUM field. Successive items are as shown with IT2NUM.
Notes
The *VPUT command is not supported for PowerGraphics displays. Inconsistent results may be obtained if
this command is not used in /GRAPHICS, FULL.
Plot and print operations entered via the GUI (Utility Menu> Pltcrtls, Utility Menu> Plot) incorporate the
AVPRIN command. This means that the principal and equivalent values are recalculated. If you use *VPUT
to put data back into the database, issue the plot commands from the command line to preserve your data.
This operation is basically the inverse of the *VGET operation. Vector items are put directly (without any
coordinate system transformation) into the ANSYS database. Items can only replace existing items of the
database and not create new items. Degree of freedom results that are replaced in the database are available
for all subsequent postprocessing operations. Other results are changed temporarily and are available mainly
for the immediately following print and display operations. The vector specification *VCUM does not apply
to this command. The valid labels for the location fields (Entity, ENTNUM, Item1, and IT1NUM) are listed
below. Item2 and IT2NUM are not currently used. Not all items from the *VGET list are allowed on *VPUT
since putting values into some locations could cause the database to be inconsistent.
This command is valid in any processor.
Table 266 *VPUT - POST1 Items
Entity = NODE, ENTNUM = n (node number)
Description IT1NUM Item1
Valid labels for nodal degree of freedom results are:
X, Y, or Z structural displacement. X, Y, Z U
X, Y, or Z structural rotation. X, Y, Z ROT
Temperature. For SHELL131 and SHELL132 elements with
KEYOPT(3) = 0 or 1, use TBOT, TE2, TE3, . . ., TTOP instead
TEMP
of TEMP. Alternative get functions: TEMP(N), TBOT(N),
TE2(N), etc.
Pressure. PRES
Electric potential. VOLT
Magnetic scalar potential. MAG
X, Y, or Z fluid velocity. X, Y, Z V
X, Y, or Z magnetic vector potential. X, Y, Z A
Current. CURR
Electromotive force drop. EMF
Turbulent kinetic energy (FLOTRAN). ENKE
Turbulent energy dissipation (FLOTRAN). ENDS
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1836
*VPUT
Entity = NODE, ENTNUM = n (node number)
Description IT1NUM Item1
Valid labels for element nodal results are:
Description IT1NUM Item1
Component stress. X, Y, Z, XY, YZ,
XZ
S
Principal stress. 1, 2, 3 "
Stress intensity or equivalent stress. INT, EQV "
Component total strain (EPEL + EPPL + EPCR). X, Y, Z, XY, YZ,
XZ
EPTO
Principal total strain. 1,2,3 "
Total strain intensity or total equivalent strain. INT, EQV "
Component elastic strain. X, Y, Z, XY, YZ,
XZ
EPEL
Principal elastic strain. 1, 2, 3 "
Elastic strain intensity or elastic equivalent strain. INT, EQV "
Component plastic strain. X, Y, Z, XY, YZ,
XZ
EPPL
Principal plastic strain. 1,2,3 "
Plastic strain intensity or plastic equivalent strain. INT, EQV "
Component creep strain. X, Y, Z, XY, YZ,
XZ
EPCR
Principal creep strain. 1, 2, 3 "
Creep strain intensity or creep equivalent strain. INT, EQV "
Component thermal strain. X, Y, Z, XY, YZ,
XZ
EPTH
Principal thermal strain. 1, 2, 3 "
Thermal strain intensity or thermal equivalent strain. INT, EQV "
Swelling strain. EPSW
Equivalent stress (from stress-strain curve). SEPL NL
Stress state ratio. SRAT "
Hydrostatic pressure. HPRES "
Accumulated equivalent plastic strain. EPEQ "
Plastic state variable. PSV "
Plastic work/volume. PLWK "
Component thermal gradient. X, Y, Z TG
Component thermal flux. X, Y, Z TF
Component pressure gradient. X, Y, Z PG
Component electric field. X, Y, Z EF
Component electric flux density. X, Y, Z D
Component magnetic field intensity. X, Y, Z H
1837
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VPUT
Entity = NODE, ENTNUM = n (node number)
Description IT1NUM Item1
Component magnetic flux density. X, Y, Z B
Component electromagnetic force. X, Y, Z FMAG
Valid labels for FLOTRAN nodal results are:
Description IT1NUM Item1
Total temperature. TTOT
Heat flux. HFLU
Heat transfer (film) coefficient. HFLM
Fluid laminar conductivity. COND
Pressure coefficient. PCOE
Total (stagnation) pressure. PTOT
Mach number. MACH
Stream function. (2-D applications only.) STRM
Fluid density. DENS
Fluid laminar viscosity. VISC
Fluid effective viscosity. EVIS
Fluid effective conductivity. ECON
Y+, a turbulent law of the wall parameter. YPLU
Shear stress at the wall. TAUW
Entity = ELEM, ENTNUM = n (element number)
Valid labels for element results are:
Any user-defined element table label (see ETABLE com-
mand).
Lab ETAB
Menu Paths
Utility Menu>Parameters>Array Operations>Put Array Data
VPUT, Par, IR, TSTRT, KCPLX, Name
Moves an array parameter vector into a variable.
POST26: Special Purpose
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Par
Array parameter vector in the operation.
IR
Arbitrary reference number assigned to this variable (1 to NV [NUMVAR]). Overwrites any existing results
for this variable.
TSTRT
Time (or frequency) corresponding to start of IR data. If between values, the nearer value is used.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1838
VPUT
KCPLX
Complex number key:
0
Use the real part of the IR data.
1
Use the imaginary part of the IR data.
Name
Thirty-two character name identifying the item on printouts and displays. Defaults to the label formed
by concatenating VPUT with the reference number IR.
Notes
At least one variable should be defined (NSOL, ESOL, RFORCE, etc.) before using this command. The starting
array element number must be defined. For example, VPUT,A(1),2 moves array parameter A to variable 2
starting at time 0.0. Looping continues from array element A(1) with the index number incremented by one
until the variable is filled. Unfilled variable locations are assigned a zero value. The number of loops may be
controlled with the *VLEN command (except that loop skipping (NINC) is not allowed). For multi-dimensioned
array parameters, only the first (row) subscript is incremented.
Menu Paths
Main Menu>TimeHist Postpro>Table Operations>Parameter to Var
*VREAD, ParR, Fname, Ext, --, Label, n1, n2, n3, NSKIP
Reads data and produces an array parameter vector or matrix.
APDL: Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting array parameter vector. See *SET for name restrictions. The parameter must
exist as a dimensioned array [*DIM]. String arrays are limited to a maximum of 8 characters.
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
If the Fname field is left blank, reading continues from the current input device, such as the terminal.
Ext
Filename extension (8 character maximum).
--
Unused field.
Label
Can take a value of IJK, IKJ, JIK, JKI, KIJ, KJI, or blank (IJK).
1839
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VREAD
n1, n2, n3
Read as (((ParR (i,j,k), k = 1,n1), i = 1, n2), j = 1, n3) for Label = KIJ. n2 and n3 default to 1.
NSKIP
Number of lines at the beginning of the file being read that will be skipped during the reading. Default
= 0.
Notes
Reads data from a file and fills in an array parameter vector or matrix. Data are read from a formatted file
or, if the menu is off [/MENU,OFF] and Fname is blank, from the next input lines. The format of the data to
be read must be input immediately following the *VREAD command. The format specifies the number of
fields to be read per record, the field width, and the placement of the decimal point (if none specified in
the value). The read operation follows the available FORTRAN FORMAT conventions of the system (see your
system FORTRAN manual). Any standard FORTRAN real format (such as (4F6.0), (E10.3,2X,D8.2), etc.) or alpha-
numeric format (A) may be used. Alphanumeric strings are limited to a maximum of 8 characters for any
field (A8). For storage of string arrays greater than 8 characters, the *SREAD command can be used. Integer
(I) and list-directed (*) descriptors may not be used. The parentheses must be included in the format and the
format must not exceed 80 characters (including parentheses). The input line length is limited to 128 char-
acters.
A starting array element number must be defined for the result array parameter vector (numeric or character).
For example, entering these two lines:
*VREAD,A(1),ARRAYVAL
(2F6.0)
will read two values from each line of file ARRAYVAL and assign the values to A(1), A(2), A(3), etc. Reading
continues until successive row elements [*VLEN, *VMASK, *DIM] are filled.
For an array parameter matrix, a starting array element row and column number must be defined. For example,
entering these two lines:
VREAD,A(1,1),ARRAYVAL,,,IJK,10,2
(2F6.0)
will read two values from each line of file ARRAYVAL and assign the values to A(1,1), A(2,1), A(3,1), etc.
Reading continues until n1 (10) successive row elements are filled. Once the maximum row number is
reached, subsequent data will be read into the next column (e.g., A(1,2), A(2,2), A(3,2), etc.)
For numerical parameters, absolute values and scale factors may be applied to the result parameter [*VABS,
*VFACT]. Results may be cumulative [*VCUM]. See the *VOPER command for details. If you are in the GUI
the *VREAD command must be contained in an externally prepared file read into the ANSYS program (i.e.,
*USE, /INPUT, etc.).
This command is not applicable to 4- or 5-D arrays.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Parameters>Read from File
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1840
*VREAD
VROTAT, NA1, NA2, NA3, NA4, NA5, NA6, PAX1, PAX2, ARC, NSEG
Generates cylindrical volumes by rotating an area pattern about an axis.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NA1, NA2, NA3, . . . , NA6
List of areas in the pattern to be rotated (6 maximum if using keyboard entry). Areas must lie to one
side of, and in the plane of, the axis of rotation. If NA1 = P, graphical picking is enabled and all remaining
command fields are ignored (valid only in the GUI). If NA1 = ALL, all selected areas will define the pattern
to be rotated. A component name may also be substituted for NA1.
PAX1, PAX2
Keypoints defining the axis about which the area pattern is to be rotated.
ARC
Arc length (in degrees). Positive follows right-hand rule about PAX1-PAX2 vector. Defaults to 360.
NSEG
Number of volumes (8 maximum) around circumference. Defaults to minimum required for 90 (maximum)
arcs, i.e., 4 for 360, 3 for 270, etc.
Notes
Generates cylindrical volumes (and their corresponding keypoints, lines, and areas) by rotating an area pattern
(and its associated line and keypoint patterns) about an axis. Keypoint patterns are generated at regular
angular locations (based on a maximum spacing of 90). Line patterns are generated at the keypoint patterns.
Arc lines are also generated to connect the keypoints circumferentially. Keypoint, line, area, and volume
numbers are automatically assigned (beginning with the lowest available values). Adjacent lines use a
common keypoint, adjacent areas use a common line, and adjacent volumes use a common area.
To generate a single volume with an arc greater than 180, NSEG must be greater than or equal to 2.
If element attributes have been associated with the input area via the AATT command, the opposite area
generated by the VROTAT operation will also have those attributes (i.e., the element attributes from the
input area are copied to the opposite area). Note that only the area opposite the input area will have the
same attributes as the input area; the areas adjacent to the input area will not.
If the given areas are meshed or belong to meshed volumes, the 2-D mesh can be rotated (extruded) to a
3-D mesh. See the Modeling and Meshing Guide for more information. Note that the NDIV argument on the
ESIZE command should be set before extruding the meshed areas.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Extrude>Areas>About Axis
1841
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VROTAT
VSBA, NV, NA, SEPO, KEEPV, KEEPA
Subtracts areas from volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV
Volume (or volumes, if picking is used) to be subtracted from. If ALL, use all selected volumes. If P,
graphical picking is enabled (valid only in the GUI) and remaining fields are ignored. A component name
may also be substituted for NV.
NA
Area (or areas, if picking is used) to subtract. If ALL, use all selected areas. A component name may also
be substituted for NA.
SEPO
Behavior of the touching boundary:
(blank)
The resulting volumes will share area(s) where they touch.
SEPO
The resulting volumes will have separate, but coincident area(s) where they touch.
KEEPV
Specifies whether NV volumes are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NV volumes after VSBA operation (override BOPTN command settings).
KEEP
Keep NV volumes after VSBA operation (override BOPTN command settings).
KEEPA
Specifies whether NA areas are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NA areas after VSBA operation (override BOPTN command settings).
KEEP
Keep NA areas after VSBA operation (override BOPTN command settings).
Notes
Generates new volumes by subtracting the regions common to both the volumes and areas (the intersection)
from the NV volumes. The intersection will be an area(s). If SEPO is blank, the volume is divided at the area
and the resulting volumes will be connected, sharing a common area where they touch. If SEPO is set to
SEPO, the volume is divided into two unconnected volumes with separate areas where they touch. See the
Modeling and Meshing Guide for an illustration. See the BOPTN command for an explanation of the options
available to Boolean operations. Element attributes and solid model boundary conditions assigned to the
original entities will not be transferred to the new entities generated.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1842
VSBA
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Volume by Area
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>With Options>Volume by Area
VSBV, NV1, NV2, SEPO, KEEP1, KEEP2
Subtracts volumes from volumes.
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV1
Volume (or volumes, if picking is used) to be subtracted from. If ALL, use all selected volumes. Volumes
specified in set NV2 are removed from set NV1. If P, graphical picking is enabled (valid only in the GUI)
and remaining fields are ignored. A component name may also be substituted for NV1.
NV2
Volume (or volumes, if picking is used) to subtract. If ALL, use all selected volumes (except those included
in the NV1 argument). A component name may also be substituted for NV2.
SEPO
Behavior if the intersection of the NV1 volumes and the NV2 volumes is an area or areas:
(blank)
The resulting volumes will share area(s) where they touch.
SEPO
The resulting volumes will have separate, but coincident area(s) where they touch.
KEEP1
Specifies whether NV1 volumes are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NV1 volumes after VSBV operation (override BOPTN command settings).
KEEP
Keep NV1 volumes after VSBV operation (override BOPTN command settings).
KEEP2
Specifies whether NV2 volumes are to be deleted:
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NV2 volumes after VSBV operation (override BOPTN command settings).
KEEP
Keep NV2 volumes after VSBV operation (override BOPTN command settings).
1843
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VSBV
Notes
Generates new volumes by subtracting the regions common to both NV1 and NV2 volumes (the intersection)
from the NV1 volumes. The intersection can be an volume(s) or area(s). If the intersection is an area and
SEPO is blank, the NV1 volume is divided at the area and the resulting volumes will be connected, sharing
a common area where they touch. If SEPO is set to SEPO, NV1 is divided into two unconnected volumes
with separate areas where they touch. See the Modeling and Meshing Guide for an illustration. See the BOPTN
command for an explanation of the options available to Boolean operations. Element attributes and solid
model boundary conditions assigned to the original entities will not be transferred to the new entities
generated. VSBV,ALL,ALL will have no effect because all the volumes in set NV1will have been moved to
set NV2.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Subtract>Volumes
Main Menu>Preprocessor>Modeling>Operate>Booleans>Subtract>With Options>Volumes
Main Menu>Preprocessor>Modeling>Operate>Subract>Volumes
VSBW, NV, SEPO, KEEP
Subtracts intersection of the working plane from volumes (divides volumes).
PREP7: Booleans
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
NV
Volume (or volumes, if picking is used) to be subtracted from. If NV = ALL, use all selected volumes. If
NV = P, graphical picking is enabled (valid only in the GUI). A component name may also be input for
NV.
SEPO
Behavior of the created boundary.
(blank)
The resulting volumes will share area(s) where they touch.
SEPO
The resulting volumes will have separate, but coincident area(s).
KEEP
Specifies whether NV volumes are to be deleted.
(blank)
Use the setting of KEEP on the BOPTN command.
DELETE
Delete NV volumes after VSBW operation (override BOPTN command settings).
KEEP
Keep NV volumes after VSBW operation (override BOPTN command settings).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1844
VSBW
Notes
Generates new volumes by subtracting the intersection of the working plane from the NV volumes. The in-
tersection will be an area(s). If SEPO is blank, the volume is divided at the area and the resulting volumes
will be connected, sharing a common area where they touch. If SEPO is set to SEPO, the volume is divided
into two unconnected volumes with separate areas. The SEPO option may cause unintended consequences
if any keypoints exist along the cut plane. See the Modeling and Meshing Guide for an illustration. See the
BOPTN command for an explanation of the options available to Boolean operations. Element attributes and
solid model boundary conditions assigned to the original entities will not be transferred to the new entities
generated.
Issuing the VSBW command under certain conditions may generate a topological degeneracy error. Do not
issue the command if:
A sphere or cylinder has been scaled. (A cylinder must be scaled unevenly in the XY plane.)
A sphere or cylinder has not been scaled but the work plane has been rotated.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>Volu by WrkPlane
Main Menu>Preprocessor>Modeling>Operate>Booleans>Divide>With Options>Volu by WrkPlane
Main Menu>Preprocessor>Modeling>Operate>Divide>Volu by WrkPlane
/VSCALE, WN, VRATIO, KEY
Scales the length of displayed vectors.
GRAPHICS: Scaling
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
VRATIO
Ratio value applied to the automatically calculated scale factor (defaults to 1.0, i.e., use scale factor as
automatically calculated).
KEY
Relative scaling key:
0
Use relative length scaling among vectors based on magnitudes.
1
Use uniform length scaling for all vector lengths.
Command Default
Automatic vector scaling.
1845
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/VSCALE
Notes
Allows scaling of the vector length displayed with the PLVECT command of POST1 and the /PBC and /PSF
commands. Also allows the scaling of the element (i.e., /PSYMB,ESYS) and the nodal (i.e., /PSYMB,NDIR)
coordinate system symbols.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>Plot Results>Vector Plot>Predefined
Utility Menu>Plot>Results>Vector Plot
Utility Menu>PlotCtrls>Multi-Plot Contrls
Utility Menu>PlotCtrls>Style>Vector Arrow Scaling
*VSCFUN, ParR, Func, Par1
Determines properties of an array parameter.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
ParR
The name of the resulting scalar parameter. See *SET for name restrictions.
Func
Functions:
MAX --
Maximum: the maximum Par1 array element value.
MIN --
Minimum: the minimum Par1 array element value.
LMAX --
Index location of the maximum Par1 array element value. Array Par1 is searched starting from its
specified index.
LMIN --
Index location of the minimum Par1 array element value. Array Par1 is searched starting from its
specified index.
FIRST --
Index location of the first nonzero value in array Par1. Array Par1 is searched starting from its
specified index.
LAST --
Index location of the last nonzero value in array Par1. Array Par1 is searched starting from its
specified index.
SUM --
Sum: Par1 (the summation of the Par1 array element values).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1846
*VSCFUN
MEDI --
Median: value of Par1 at which there are an equal number of values above and below.
MEAN --
Mean: ( Par1)/NUM, where NUM is the number of summed values.
VARI --
Variance: ( ((Par1-MEAN)**2))/NUM.
STDV --
Standard deviation: square root of VARI.
RMS --
Root-mean-square: square root of ( (Par1**2))/NUM.
NUM --
Number: the number of summed values (masked values are not counted).
Par1
Array parameter vector in the operation.
Notes
Operates on one input array parameter vector and produces one output scalar parameter according to:
ParR = f(Par1)
where the functions (f ) are described below. The starting array element number must be defined for the
array parameter vector. For example, *VSCFUN,MU,MEAN,A(1) finds the mean of the A vector values, starting
from the first value and stores the result as parameter MU. Operations use successive array elements [*VLEN,
*VMASK] with the default being all successive array elements. Absolute values and scale factors may be
applied to all parameters [*VABS, *VFACT]. Results may be cumulative [*VCUM]. See the *VOPER command
for details.
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Operations>Vector-Scalar Func
VSEL, Type, Item, Comp, VMIN, VMAX, VINC, KSWP
Selects a subset of volumes.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of volume select:
S
Select a new set (default).
R
Reselect a set from the current set.
1847
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VSEL
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
ALL
Restore the full set.
NONE
Unselect the full set.
INVE
Invert the current set (selected becomes unselected and vice versa).
STAT
Display the current select status.
The following fields are used only with Type = S, R, A, or U:
Item
Label identifying data. Valid item labels are shown in the table below. Some items also require a com-
ponent label. If Item = PICK (or simply P), graphical picking is enabled and all remaining command
fields are ignored (valid only in the GUI). Defaults to VOLU.
Comp
Component of the item (if required). Valid component labels are shown in the table below.
VMIN
Minimum value of item range. Ranges are volume numbers, coordinate values, attribute numbers, etc.,
as appropriate for the item. A component name (as specified on the CM command) may also be substi-
tuted for VMIN (VMAX and VINC are ignored). If Item = MAT, TYPE, REAL, or ESYS and if VMIN is pos-
itive, the absolute value of Item is compared against the range for selection; if VMIN is negative, the
signed value of Item is compared. See the VLIST command for a discussion of signed attributes.
VMAX
Maximum value of item range. VMAX defaults to VMIN.
VINC
Value increment within range. Used only with integer ranges (such as for volume numbers). Defaults to
1. VINC cannot be negative.
KSWP
Specifies whether only volumes are to be selected:
0
Select volumes only.
1
Select volumes, as well as keypoints, lines, areas, nodes, and elements associated with selected
volumes. Valid only with Type = S.
Command Default
All volumes are selected.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1848
VSEL
Notes
Selects volumes based on values of a labeled item and component. For example, to select a new set of
volumes based on volume numbers 1 through 7, use VSEL,S,VOLU,,1,7. The subset is used when the ALL
label is entered (or implied) on other commands, such as VLIST,ALL. Only data identified by volume number
are selected. Data are flagged as selected and unselected; no data are actually deleted from the database.
This command is valid in any processor.
For Selects based on non-integer numbers (coordinates, results, etc.), items that are within the range
VMIN-Toler and VMAX+Toler are selected. The default tolerance Toler is based on the relative values
of VMIN and VMAX as follows:
If VMIN = VMAX, Toler = 0.005 x VMIN.
If VMIN = VMAX = 0.0, Toler = 1.0E-6.
If VMAX VMIN, Toler = 1.0E-8 x (VMAX-VMIN).
Use the SELTOL command to override this default and specify Toler explicitly.
Table 267 VSEL - Valid Item and Component Labels
VSEL Type, Item, Comp, VMIN, VMAX, VINC, KABS
Description Comp Item
Volume number. VOLU
X, Y, or Z center (picking "hot spot" location in the active co-
ordinate system).
X, Y, Z LOC
Material number associated with the volume. MAT
Element type number associated with the volume. TYPE
Real constant set number associated with the volume. REAL
Element coordinate system associated with the volume. ESYS
Menu Paths
Utility Menu>Select>Entities
VSLA, Type, VLKEY
Selects those volumes containing the selected areas.
DATABASE: Selecting
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Type
Label identifying the type of volume select:
S
Select a new set (default).
R
Reselect a set from the current set.
1849
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VSLA
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
VLKEY
Specifies whether all contained volume areas must be selected [ASEL]:
0
Select volume if any of its areas are in the selected area set.
1
Select volume only if all of its areas are in the selected area set.
Notes
This command is valid in any processor.
Menu Paths
Utility Menu>Select>Entities
*VSTAT
Lists the current specifications for the array parameters.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Notes
Lists the current specifications for the *VABS, *VCOL, *VCUM, *VFACT, *VLEN, and *VMASK commands.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
VSUM, LAB
Calculates and prints geometry statistics of the selected volumes.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LAB
Controls the degree of tessellation used in the calculation of area properties. If LAB = DEFAULT, area
calculations will use the degree of tessellation set through the /FACET command. If LAB = FINE, area
calculations are based on a finer tessellation.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1850
*VSTAT
Notes
Calculates and prints geometry statistics (volume, centroid location, moments of inertia, etc.) associated with
the selected volumes. Geometry items are reported in the global Cartesian coordinate system. A unit density
is assumed unless the volumes have a material association via the VATT command. Items calculated by
VSUM and later retrieved by a *GET or *VGET command are valid only if the model is not modified after
the VSUM command is issued.
Setting a finer degree of tessellation will provide area calculations with greater accuracy, especially for thin,
hollow models. However, using a finer degree of tessellation requires longer processing.
For very thin volumes, such that the ratio of the minimum to the maximum dimension is less than 0.01, the
VSUM command can provide erroneous volume information. To ensure that such calculations are accurate,
make certain that you subdivide such volumes so that the ratio of the minimum to the maximum is at least
0.05.
Menu Paths
Main Menu>Preprocessor>Modeling>Operate>Calc Geom Items>Of Volumes
VSWEEP, VNUM, SRCA, TRGA, LSMO
Fills an existing unmeshed volume with elements by sweeping the mesh from an adjacent area through
the volume.
PREP7: Meshing
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
VNUM
Number identifying the volume that is to be meshed by VSWEEP. If VNUM = P, graphical picking is enabled,
you will be prompted to choose the volume or volumes based on the setting of EXTOPT,VSWE,AUTO.
This argument is required.
ALL is a valid input value that when selected sends all the selected volumes to the sweeper. If VNUM
= ALL, each volume that can be swept will be and those not able to be swept will be unmeshed or
meshed with tets depending upon the setting of EXTOPT,VSWE,TETS.
A component name is a valid input value. All volumes that are part of the specified component will be
sent to the sweeper.
SRCA
Number identifying the source area. This is the area whose mesh will provide the pattern for the volume
elements. (If you do not mesh the source area prior to volume sweeping, ANSYS meshes it internally
when you initiate volume sweeping.) ANSYS sweeps the pattern of the area elements through the volume
to create the mesh of volume elements. You cannot substitute a component name for SRCA.
This argument is optional. If VNUM = ALL or is a component containing more than one volume, SRCA is
ignored. If SRCA is not provided or if it is ignored, VSWEEP attempts to automatically determine which
area should be the target area.
1851
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VSWEEP
TRGA
Number identifying the target area. This is the area that is opposite the source area specified by SRCA.
You cannot substitute a component name for TRGA.
This argument is optional. If VNUM = ALL or component containing more than one volume, TRGA is ig-
nored. If TRGA is not provided or if it is ignored, VSWEEP attempts to automatically determine which
area should be the target area.
LSMO
Value specifying whether ANSYS should perform line smoothing during volume sweeping. (The value
of this argument controls line smoothing for the VSWEEP command only; it has no effect on the setting
of the MOPT command's LSMO option.) This argument is optional.
0
Do not perform line smoothing. This is the default.
1
Always perform line smoothing. This setting is not recommended for large models due to speed
considerations.
Notes
If the source mesh consists of quadrilateral elements, ANSYS fills the volume with hexahedral elements. If
the source mesh consists of triangles, ANSYS fills the volume with wedges. If the source mesh consists of a
combination of quadrilaterals and triangles, ANSYS fills the volume with a combination of hexahedral and
wedge elements.
In the past, you may have used the VROTAT, VEXT, VOFFST, and/or VDRAG commands to extrude a meshed
area into a meshed volume. However, those commands create the volume and the volume mesh simultan-
eously. In contrast, the VSWEEP command is intended for use in an existing unmeshed volume. This makes
VSWEEP particularly useful when you have imported a solid model that was created in another program,
and you want to mesh it in ANSYS.
For related information, see the description of the EXTOPT command (although EXTOPT sets volume
sweeping options, it does not affect element spacing). Also see the detailed discussion of volume sweeping
in Meshing Your Solid Model of the Modeling and Meshing Guide.
Menu Paths
Main Menu>Preprocessor>Meshing>Mesh>Volume Sweep>Sweep
VSYMM, Ncomp, NV1, NV2, NINC, KINC, NOELEM, IMOVE
Generates volumes from a volume pattern by symmetry reflection.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Ncomp
Symmetry key:
X
X symmetry (default).
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1852
VSYMM
Y
Y symmetry.
Z
Z symmetry.
NV1, NV2, NINC
Reflect volumes from pattern beginning with NV1 to NV2 (defaults to NV1) in steps of NINC (defaults
to 1). If NV1 = ALL, NV2 and NINC are ignored and the pattern is all selected volumes [VSEL]. If NV1 =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NV1 (NV2 and NINC are ignored).
KINC
Keypoint increment between sets. If zero, the lowest available keypoint numbers are assigned [NUMSTR].
NOELEM
Specifies whether nodes and elements are also to be generated:
0
Generate nodes and elements associated with the original volumes, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether volumes will be moved or newly defined:
0
Generate additional volumes.
1
Move original volumes to new position retaining the same keypoint numbers (KINC and NOELEM
are ignored). Corresponding meshed items are also moved if not needed at their original position.
Notes
Generates a reflected set of volumes (and their corresponding keypoints, lines, areas and mesh) from a given
volume pattern by a symmetry reflection (see analogous node symmetry command, NSYM). The MAT, TYPE,
REAL, and ESYS attributes are based upon the volumes in the pattern and not upon the current settings.
Reflection is done in the active coordinate system by changing a particular coordinate sign. The active co-
ordinate system must be a Cartesian system. Volumes in the pattern may have been generated in any co-
ordinate system. However, solid modeling in a toroidal coordinate system is not recommended. Volumes
are generated as described in the VGEN command.
See the ESYM command for additional information about symmetry elements.
Menu Paths
Main Menu>Preprocessor>Modeling>Reflect>Volumes
1853
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VSYMM
/VT
Enters the Variational Technology preprocessor.
SESSION: Processor Entry
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Notes
Enters the Variational Technology processor.
Menu Paths
Main Menu>DesignXplorer VT>Solution>Solve
VTCLR, Type
Clears the Variational Technology database.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Type
Specifies the part of the Variational Technology database to be cleared. Valid labels are:
ALL
Clears the entire Variational Technology database. Both the preprocessing part (settings for input
variables, result parameters, etc.) and results part (results for the derivatives) are cleared.
RSLT
Clear only the results part of the Variational Technology database. This is necessary if you want to
modify the Variational Technology model (for example, adding a new input variable and so on) after
a solution has already been performed.
Notes
Clear the database of the Variational Technology module. The settings are reset to their default values and
the memory is cleared. Remember that the result files containing the results of Variational Technology (see
VTRFIL command) are never deleted unless you intentionally delete them. We recommend that you use
this command before switching to a new solution, which, for example, includes different input variables or
different element or node components for input variables. See the various Variational Technology commands
for changes that are possible after the solution has been performed.
If the entire database is cleared with VTCLR,ALL then also all definitions for input variables, result parameters,
solution method, result file storage and so on are either deleted and/or reset to their default values.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Other>Clear Database
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1854
/VT
VTDISC, Name, ElComp
Defines an element component as a discrete input variable for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name for the input variable. This is a string of up to 250 characters used for postprocessing purposes.
Blanks are allowed in the string.
ElComp
The name of either of an element component or the number of a single element.
Command Default
None
Notes
Defines a component containing a group of elements as a discrete input variable of the DesignXplorer. In
a discrete optimization the DesignXplorer will evaluate the dependency of the result parameter(s) (see the
VTRSLT command) as a function of the discrete case of either including the group of elements in the finite-
element model or excluding the group of elements from the model.
The ElComp field defines a group of elements that are considered as either being included or excluded
from the finite-element model during a discrete optimization.
If a solution has already been performed with Variational Technology using STAOPT,VT and the SOLVE
command, then the Variational Technology database is locked. In this case no input variables can be added
to the model. You cannot change the specific definitions of an existing input variable. If you want to modify
the Variational Technology model by adding a discrete variable, you first have to unlock the model with the
VTCLR command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Discrete Variable
VTEVAL, --, Mode
Triggers evaluation of generated results based on input variables specified via the VTVMOD command.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
--
Unused field.
Mode
The mode number for normal modes. By default, driven by the application.
1855
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTEVAL
Command Default
Defaults as outlined above.
Notes
Used in conjunction with the VTVMOD command, VTEVAL evaluates the designated variables and loads
the results into the ANSYS database. You can employ the ANSYS postprocessing tools to view the results,
or perform an optimization or PDS loop.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu> DesignXplorer VT> Postprocessing> Evaluate Results
VTFREQ, Name, MIN, MAX, INC, RedOpt
Defines the frequency as input variable for the harmonic sweep functionality of VT Accelerator.
VARIATIONAL TECHNOLOGY: DesignXplorer
VARIATIONAL TECHNOLOGY: VT Harmonic Sweep
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name of the variable. This is a string of up to 250 characters used for postprocessing purposes. Blanks
can be included in the string.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
INC
Number of increments within the range of the input variable as defined by the MIN and MAX fields.
RedOpt
Reduction option that specifies what to do if the DesignXplorer needs to reduce the variation range
from MIN to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Command Default
There are no default values for the command fields.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1856
VTFREQ
Notes
Defines the frequency as an input variable. Variational Technology will evaluate a frequency sweep from the
specified minimum to the maximum frequency values given in the MIN and MAX fields. The frequency sweep
will include as many frequency increments as specified in the INC field.
If a solution has already been performed with Variational Technology using STAOPT,VT and the SOLVE
command, then the Variational Technology database is locked. In this case no input variables can be added
to the model. You cannot change the specific definitions of an existing input variable. If you want to modify
the Variational Technology model by adding a discrete variable, you first have to unlock the model with the
VTCLR command.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Frequency
VTGEOM, Name, MIN, MAX, RedOpt, Order
Defines a geometry parameter created with ANSYS Mesh Morpher as a DesignXplorer input variable.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name of an input variable. This is a string of up to 256 characters used for post-processing purposes. It
can have blanks included in the string.
MIN
Minimum value of the input variable. By default, driven by the application.
MAX
Maximum value of the input variable. By default, driven by the application.
RedOpt
Reduction option that specifies what to do if the DesignXplorer needs to reduce the variation range
from MIN to MAX. Valid labels are;
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Order
Deviation Order. By default, driven by the application.
Command Default
Defaults as outlined above.
1857
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTGEOM
Notes
If you do not use ANSYS Mesh Morpher, you can ignore the VTGEOM command.
The VTGEOM command allows you to define a geometry parameter created with ANSYS Mesh Morpher as
an input variable of the DesignXplorer.
The filename of the ANSYS Mesh Morpher database has to be defined using the VTRFIL command.
DesignXplorer uses the same RSX file to store its results. If the RSX file does not exist or does not contain
the expected geometry parameters and associated morphed mesh, then the VTGEOM command is ignored.
In the general case, the fields MIN and MAX should not be used because a range of variation is already
defined in ANSYS Mesh Morpher. Nethertheless, the MIN and MAX fields allow to define a smaller range of
variation to compute the VT results. Althought is it not recommended, it is also possible to define a larger
range of variation. In this case, a warning message is always generated.
The MIN and MAX field must always be specified or left blank together. If only one field is specified, then
the VTGEOM command is ignored.
If the Order is not specified, it will be increased until the accuracy requirements of the VTRSLT command
are satisfied. When you set a value for Order, no further adjustments for accuracy will be performed. Order
should only be used when an input parameter has an influence with a limited order.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
VTIN, Name, MIN, MAX, RedOpt, Lab, Comp, --, VarType, Order
Defines an inertial load as an input variable for DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name for the input variable. This is a string of up to 250 characters used for postprocessing purposes.
Blanks can be included in the string.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
RedOpt
Reduction option that specifies what to do if DesignXplorer needs to reduce the variation range from
MIN to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1858
VTIN
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Lab
Valid inertial load label. This label must be defined for the inertial load using the ANSYS command of
the same name prior to issuing this command. Applicable labels are:
ACEL
Linear acceleration.
CGOMGA
Rotational velocity of the global origin.
DCGOMG
Rotational acceleration of the global origin.
DOMEGA
Rotational acceleration of the structure.
OMEGA
Rotational velocity.
--
Unused field.
Comp
X, Y,Z or ALL
If Comp =ALL, the VarType must be FACT.
VarType
Label indicating how the variability of vectors should be handled, when COMP = ALL is used. Valid labels
are:
FACT
The variability of vector is described by a scaling factor that is applied to all 3 vectors components.
This means that the MIN field is the value of the minimum factor to be applied and the MAX field is
the maximum factor to be applied.
ADD
The variability of vector components is described by an add-on that is applied to vector components.
This means that the MIN field is the minimum value that is added to the vector components and
the MAX field contains the maximum value that is added to the vector components. As such the
values specified for MIN and MAX have the same unit as the vector itself.
VAL
The variability of the vectors is described by the variability of the vector components itself. This is
only applicable if ALL has not been used. This means that the MIN field is the minimum value that
the vector component itself can take and the MAX field contains the maximum value, which the
vector component can take. This is the default.
Order
Derivation order. By default, driven by the application.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1859
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTIN
Menu Paths
This command cannot be accessed from a menu.
VTMETH, SoluType, ApprType, ModeTrack
Defines the solution options for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
SoluType
Specifies the type of solution that needs to be evaluated during the solution run of the DesignXplorer.
Valid labels are:
INDP
Builds the solution approximation assuming that the input variables are independent. Therefore,
mixed input variables interactions will be neglected.
FULL
Builds the full solution approximation, addressing the interactions between the input variables (if
necessary). This option is the default.
ApprType
Specifies the type of approximation function that is used to build the DesignXplorer. Valid labels are:
ROMS
Uses a Reduced Order Model Sweep approximation. This method is the more accurate, it supports
multiple input variables per element, continuous as well as discrete input variables. However it is
limited to linear static analysis (structural and thermal), and cannot accept more than 10 continuous
input variables or 20 discrete input variables. Continuous and Discrete input variables cannot be
mixed.
PADE
Uses a series expansion based on Pade approximations using rational functions. This method is more
accurate than TAYLOR, however it is limited to a single input variable per element and does not
support geometry input variables.
TAYL
Uses a Taylor series expansion as the approximation function. This method supports multiple input
variables per element, i.e. Young's Modulus and thickness, geometry input variables, linear static and
normal modes analysis. However, Discrete input variables is not supported.
AUTO
This option is the default and allows DesignXplorer VT to determine automatically the most appro-
priate method to use. This is the recommended option.
ModeTrack
Specifies whether or not mode tracking should be in a modal solution of the DesignXplorer. Valid labels
are:
YES
Mode Tracking will be used in a modal solution. This option is the default.
NO
Mode Tracking will not be used in a modal solution.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1860
VTMETH
Command Default
Defaults as outlined above.
Notes
Determines which approximation function of the result parameters with respect to the input variables shall
be calculated during the DesignXplorer solution run.
The default ApprType=AUTO is recommended. It selects the ROMS method each time it is possible, or the
most appropriate method in case ROMS does not apply to the model.
If Variational Technology is not used in connection with a modal analysis, then the ModeTrack option is ig-
nored.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Solution>Solution Method
VTMP, Name, MIN, MAX, RedOpt, Lab, MAT, ElComp, VarType, Order
Defines a material property as an input variable for DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name for the input variable. This is a string of up to 250 characters used for postprocessing purposes.
Blanks can be included in the string.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
RedOpt
Reduction option that specifies what to do if DesignXplorer needs to reduce the variation range from
MIN to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Lab
Valid material property label. This label must be defined for the material number MAT using the MP or
MPDATA command prior to issuing this command. Applicable labels are:
EX
Elastic moduli (also EY, EZ).
1861
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTMP
NUXY
Minor Poisson's ratios (also NUYZ, NUXZ).
GXY
Shear moduli (also GYZ, GXZ).
DENS
Material density
KXX
Conductivity (only in thermal analysis), also KYY, KZZ
ALPX
Thermal Expansion, also ALPY, ALPZ
MAT
Material identification number
ElComp
Name of an element component or number of a single individual element. If ElComp = ALL, then all
elements are considered for the VTMP command.
VarType
Label indicating how the variability of temperature dependent material properties should be handled.
This is particularly useful to describe how temperature dependent material properties are handled. At
this release, only one label (the default) is valid:
FACT
The variability of the material property is described by a scale factor that is applied to the material
property values. In the case of temperature dependent material properties, a scale factor will be
applied to all defined material property values. This means that the MIN field is the value of the
minimum factor to be applied and the MAX field is the maximum factor to be applied. As such the
values specified for MIN and MAX are unitless quantities. VarType = FACT is the default for temper-
ature dependent material properties.
ADD
The variability of the material property is described by an add-on that is applied to the material
property values. In the case of a temperature dependent material property, the defined property
values are shifted all at the same time. This means that the MIN field is the minimum value that is
added to the material property values and the MAX field contains the maximum value that is added
to the material property values. Hence, the values specified for MIN and MAX have the same unit as
the material property itself.
VAL
The variability of the material property is described by the variability of the material property value
itself. This is only applicable if the material property is not temperature dependent. This means that
the MIN field is the minimum value that the material property itself can take and the MAX field
contains the maximum value which the material property values can take. This is the default if the
material properties are not temperature dependent.
Order
Derivation order. By default, driven by the application.
Command Default
As described above.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1862
VTMP
Notes
Defines a material property of a single element or a set of elements as an input variable of the DesignXplorer.
The DesignXplorer will evaluate the derivatives of the result parameter(s) (see VTRSLT command) with respect
to the material property defined here. The derivatives are used to determine an approximation function for
the result parameter(s) as a function of the material property in the range specified by the MIN and MAX
fields.
The type of the material property is defined by the Lab field, which must not be left blank. The material
property must exist or be specified using the MP or MPDATA command prior to issuing the VTMP command.
Using either the MAT field or the ElComp field a set of element is selected for which the material property
is varied. The user must define one of them and leave the respective other one blank. All selected elements
must be associated with the same material number. This is of course always given in the MAT field. If selected
by the ElComp field, an error will be issued if an element is found that is associated with a material number
that is different from the one associated with any other element in the element component.
If the material properties are temperature dependent, then the temperature dependency is defined by either
several data points at different temperatures (MP) or by coefficients defining the temperature dependent
polynomial (MPDATA). To describe a variation of the material property with one single parameter is only
possible by using a scalar addition to the temperature dependency curve (VarType = ADD) or by using a
scalar multiplication factor for the entire temperature dependency curve (VarType = FACT). Varying the
material property value itself using VarType = VAL is not possible, if the material property is temperature
dependent.
An error message is generated and the VTMP command is ignored when the current value is outside the
variation range.
If the Order is not specified, it will be increased until the accuracy requirements of the VTRSLT command
are satisfied. When you set a value for Order, no further adjustments for accuracy will be performed. Order
should only be used when an input parameter has an influence with a limited order.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Material Property
VTOP, Oper, Name, Value
Defines options value for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Oper
Label for the type of operation that is performed. Valid labels are:
SET
Add an option (or if it already exists, modify its value).
DELETE
Delete an option.
1863
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTOP
RESET
Reset an option to its default value. If the option name is not specified, this applies to all options.
Name
Name of the option to define or modify. If this field is left blank, the command applies to all options.
Valid labels are listed in the table below.
Value
If Oper=SET, this is the value assigned to the option. Valid labels and values for each option are listed
in the table below.
Notes
Table of valid Name and Value labels.
Value Description Value Option Description Name
Normal (default) PUBLIC Controls the level of details of the out-
put messages.
VERBOSITY
Detailed messages DETAILED
Maximum verbosity VERBOSE
Interpolation deactivated (default) 0 Activates an additional interpolation
phase at the end of the ROMS Variation-
OPT_FAST_Per-
formInterpol
Interpolation activated 1
al solve in order to provide fast post-
processing evaluations.
Out-of-core is deactivated (default). 0 Activates/deactivates the out-of-core
post-processing mode to evaluate
OUTOFCORE
Out-of-core is activated. 1
large results databases using less
memory. Valid for TAYLOR and ROMS
approximation types.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
VTPOST
Launches the DesignXplorer postprocessing application.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Notes
The postprocessing application provides a viewer for Variational Technology results files. You can only launch
this application if the ANSYS GUI is active; that is, it cannot be launched from batch mode.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1864
VTPOST
Menu Paths
Main Menu>DesignXplorer VT>Postprocessing>SolutionViewer
VTRAN, KCNTO, NV1, NV2, NINC, KINC, NOELEM, IMOVE
Transfers a pattern of volumes to another coordinate system.
PREP7: Volumes
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
KCNTO
Reference number of coordinate system where the pattern is to be transferred. Transfer occurs from the
active coordinate system. The coordinate system type and parameters of KCNTO must be the same as
the active system.
NV1, NV2, NINC
Transfer volumes from pattern beginning with NV1 to NV2 (defaults to NV1) in steps of NINC (defaults
to 1). If NV1 = ALL, NV2 and NINC are ignored and the pattern is all selected volumes [VSEL]. If NV1 =
P, graphical picking is enabled and all remaining command fields are ignored (valid only in the GUI). A
component name may also be substituted for NV1 (NV2 and NINC are ignored).
KINC
Keypoint increment between sets. If zero, the lowest available keypoint numbers are assigned [NUMSTR].
NOELEM
Specifies whether elements and nodes are also to be generated:
0
Generate nodes and elements associated with the original volumes, if they exist.
1
Do not generate nodes and elements.
IMOVE
Specifies whether to redefine the existing volumes:
0
Generate additional volumes.
1
Move original volumes to new position retaining the same keypoint numbers (KINC and NOELEM
are ignored). Corresponding meshed items are also moved if not needed at their original position.
Notes
Transfers a pattern of volumes (and their corresponding keypoints, lines, areas and mesh) from one coordinate
system to another (see analogous node transfer command, TRANSFER). The MAT, TYPE, REAL, and ESYS at-
tributes are based upon the volumes in the pattern and not upon the current settings. Coordinate systems
may be translated and rotated relative to each other. Initial pattern may be generated in any coordinate
system. However, solid modeling in a toroidal coordinate system is not recommended. Coordinate and slope
values are interpreted in the active coordinate system and are transferred directly. Volumes are generated
as described in the VGEN command.
1865
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTRAN
Menu Paths
Main Menu>Preprocessor>Modeling>Move / Modify>Transfer Coord>Volumes
VTREAL, Name, MIN, MAX, RedOpt, Lab, NSET, ElComp, VarType, Order
Defines a real constant property as an input variable for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name of the input variable. This is a string of up to 250 characters used for postprocessing purposes.
Blanks are allowed in the string.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
RedOpt
Reduction option that specifies what to do if the DesignXplorer needs to reduce the variation range
from MIN to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Lab
Valid real constant label. This label must refer to a valid property of the real constant set as defined in
the R command. Applicable labels are:
TK
Shell thickness. This applies only to shell elements of type SHELL181, where the element thickness
has been defined with the R command.
STIF
Spring stiffness. This applies only to combination elements of type COMBIN14, where the spring
stiffness has been defined with the R command.
MASS
Lumped mass. This applies only to structural mass elements of type MASS21 with KEYOPT(3) = 2,
where the lumped mass has been defined with the R command.
NSET
Real constant set identification number as defined by the R command.
ElComp
Name of an element component or number of a single individual element. If ElComp = ALL, then all
elements are considered for the VTREAL command.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1866
VTREAL
VarType
Label indicating how the variability of a spatially dependent element thickness should be handled. This
label is ignored if the shell element thickness is constant for all elements selected by either the NSET
or the ElComp field. Valid labels are:
FACT
The variability of the section property is described by a scale factor that is applied to the section
property values. In case the section properties are different from element to element, a scale factor
will be applied to all section property values. This means that the MIN field is the value of the min-
imum factor to be applied and the MAX field is the maximum factor to be applied. As such the values
specified for MIN and MAX are unitless quantities. VarType = FACT is the default if the section
property is different from element to element.
ADD
The variability of the section property is described by an add-on that is applied to the section
property values. In case the section properties are different from element to element, then the section
property values are shifted at the same time by this add-on. This means that the MIN field is the
minimum value that is added to the section property values and the MAX field contains the maximum
value that is added to the property values. Hence, the values specified for MIN and MAX have the
physical unit of a length.
VAL
The variability of the section property is described by the modification of the section property value
itself. This is only applicable if the section property value is identical and constant for all selected
elements. This means that the MIN field is the minimum value that the section property value itself
can take and the MAX field contains the maximum value, which the section property value can take.
This is the default if the section property is identical and constant from element to element.
Order
Derivation order. By default, driven by the application.
Notes
Defines the real constant property of a single element or a set of elements as an input variable of the
DesignXplorer. The real constant property must be specified with the R command prior to the VTREAL
command. The DesignXplorer will evaluate the derivatives of the result parameter(s) (see VTRSLT command)
with respect to the section property defined here. The derivatives are used to determine an approximation
function for the result parameter(s) as a function of the section property in the range specified by the MIN
and MAX fields.
Using either the NSET field or the ElComp field a set of elements is selected for which the real constant
property is varied. The user must define one of these two fields and leave the respective other one blank.
If the real constant property is different from element to element, then each element has different values
for the real constant properties at the individual nodes. To describe a variation of the real constant property
with one single parameter is only possible by using a scalar addition to all real constant property values
(VarType = ADD) or by using a scalar multiplication factor for all real constant property values (VarType
= FACT). Varying the real constant property by its value itself using VarType = VAL is only possible if the
element thickness is constant and identical for all elements.
An error message is generated and the VTREAL command is ignored when the current value is outside the
variation range.
1867
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTREAL
If the Order is not specified, it will be increased until the accuracy requirements of the VTRSLT command
are satisfied. When you set a value for Order, no further adjustments for accuracy will be performed. Order
should only be used when an input parameter has an influence with a limited order.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Real Constant
VTRFIL, Fname, Ext, Dir
Specifies the file to which DesignXplorer results are written.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Fname
File name (32 characters maximum). Defaults to Jobname.
Ext
File name extension (8 characters maximum). Defaults to rsx if Fname is blank; otherwise, no default.
Dir
Directory name (64 characters maximum). Defaults to the current working directory.
Command Default
Fname = Jobname, Ext = rsx, the default directory is the current working directory.
Notes
Specifies which file is used to store the results of DesignXplorer. The file contains the description of the
higher order derivatives of the result parameters (selected by the VTRSLT command) with respect to the
input variables (selected by commands VTFREQ, VTMP, VTSEC, ).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Solution>Results File
VTRSLT, Name, Entity, Type, Comp, ACC, CompName
Defines a result quantity for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name of the result parameter. This is a string of up to 250 characters used for postprocessing purposes.
Blanks can be included.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1868
VTRFIL
Entity
Entity keyword. Valid keywords are listed in the tables below.
Type
Type of result parameter. Valid labels are listed in the table below depending on the Entity field.
Comp
Component of the result parameter type. Valid labels are listed in the table below depending on the
Type field.
ACC
Required accuracy for the requested result parameter. Default values depend on the individual result
parameter. The default value is 0.02 (2%).
CompName
Name of an element or node component, depending on the Entity field. If CompName = ALL then
all entities as specified in the Entity field are considered for the VTRSLT command. CompName = ALL
is the default.
Command Default
ACC = 2%, CompName = ALL.
Notes
Defines the result quantity, for which an approximation function with respect to the input variables will be
calculated.
In static analysis, displacement and temperature do not need to be specified, as they are automatically
computed for all nodes of the model and for all components.
The accuracy provided with the ACC field should not be confused with the error due to meshing of the finite-
element model. The accuracy is the maximum difference you can expect between an evaluation of the results
of DesignXplorer at any point of the input space of the defined input variables and a rerun of a full ANSYS
model at the same location of the input space.
The accuracy for all result parameters will be used to derive a global accuracy measure using the minimum
value. This means that if there are several result parameters defined with the VTRSLT command and some
have a requested accuracy of ACC = 0.05 and one has a requested accuracy of ACC = 0.01, then an accuracy
of 0.01 (1%) will be adopted for all result parameters.
If the derivation order has been specified for one parameter, then the accuracy is not guaranteed for any
variation involving this parameter.
For a normal modes analysis the result type MASS cannot be selected.
Table 268 Table of Valid Labels for the Type and Comp Fields for Entity = ELEM
Description Comp Type
Structural mass based on the elements. MASS
Component nodal stress of the element(s) (not averaged at nodes). ALL S
Component nodal temperature gradient of the element(s) (not
averaged at nodes).
ALL TG
1869
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTRSLT
Description Comp Type
Component nodal heat flux of the element(s) (not averaged at
nodes).
ALL TF
Note
For Entity = ELEM an existing component name including a set of elements must be specified
in the CompName field.
Table 269 Table of valid labels for the Type and Comp fields for Entity = Node
Description Comp Type
Displacement ALL U
Temperature ALL TEMP
Nodal reaction forces in the nodal coordinate system FX, FY, FZ,
MX, MY, MZ,
ALL
RF
Note
For Entity = NODE an existing component name including a set of nodes must be specified in
the CompName field.
Table 270 Table of valid labels for the Type and Comp fields for Entity = Mode
Description Comp Type
All mode shapes and the corresponding eigenfrequencies that are
evaluated as part of the solution.
MODE
All eigenfrequencies that are evaluated as part of the solution.
Compared to the Type = MODE option, the Type = FREQ option
FREQ
does not provide the solutions for the mode shapes. The option
will save memory compared to the Type = MODE option.
Note
For Entity = MODE, all modes are always taken into consideration, that is, the only valid user
input is CompName = ALL.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Result Quantity
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1870
VTRSLT
VTSEC, Name, MIN, MAX, RedOpt, Lab, SECID, LAYERID, ElComp, VarType, Order
Defines a section property as an input variable for DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name for the input variable. This is a string of up to 250 characters used for postprocessing purposes.
Blanks can be included.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
RedOpt
Reduction option that specifies what to do if DesignXplorer must reduce the variation range from MIN
to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Lab
Valid section property label. This label must refer to a valid property of the section as defined by the
SECTYPE and SECDATA commands. The applicable label is:
TK
Shell thickness. This applies only if the section property has been defined with Type = SHELL of
the SECTYPE command.
A
Area of section.
Iyy
Moment of inertia about the y axis.
Iyz
Product of inertia.
Izz
Moment of interia about the z axis.
Iw
Wrapping constant.
J
Torsional constant.
SECID
Section identification number as defined by the SECTYPE command and further specified by the SECDATA
command.
LAYERID
Layer identification number as defined by the SECDATA command.
1871
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTSEC
ElComp
Name of an element component or number of a single individual element. If ElComp = ALL, then all
elements are considered for the VTSEC command.
VarType
Label indicating how the variability of a spatially dependent element thickness should be handled. This
label is ignored if the shell element thickness is constant for all elements selected by either the SECID
or the ElComp field. Valid labels are:
FACT
The variability of the section property is described by a scale factor that is applied to the section
property values. In case the section properties are different from element to element, a scale factor
will be applied to all section property values. This means that the MIN field is the value of the min-
imum factor to be applied and the MAX field is the maximum factor to be applied. As such the values
specified for MIN and MAX are unitless quantities. VarType = FACT is the default if the section
property is different from element to element.
ADD
The variability of the section property is described by an add-on that is applied to the section
property values. In case the section properties are different from element to element, then the section
property values are shifted at the same time by this add-on. This means that the MIN field is the
minimum value that is added to the section property values and the MAX field contains the maximum
value that is added to the section property values. Hence, the values specified for MIN and MAX have
the physical unit of a length.
VAL
The variability of the section property is described by the modification of the section property value
itself. This is only applicable if the section property value is identical and constant for all selected
elements. This means that the MIN field is the minimum value that the section property value itself
can take and the MAX field contains the maximum value, which the section property value can take.
This is the default if the section property is identical and constant from element to element.
Order
Derivation order. By default, driven by the application.
Notes
Defines the section property of a single element or a set of elements as an input variable of DesignXplorer.
The section property must be specified with the SECTYPE and SECDATA commands prior to the VTSEC
command. The DesignXplorer will evaluate the derivatives of the result parameter(s) (see VTRSLT command)
with respect to the section property defined here. The derivatives are used to determine an approximation
function for the result parameter(s) as a function of the section property in the range specified by the MIN
and MAX fields.
Using either the SECID field or the ElComp field a set of elements is selected for which the section property
is varied. The user must define one of these two fields and leave the respective other one blank.
If the section property is different from element to element, then each element has different values for the
section property at the individual nodes. To describe a variation of the section property with one single
parameter is only possible by using a scalar addition to all section property values (VarType = ADD) or by
using a scalar multiplication factor for all section property values (VarType = FACT). Varying the section
property by its value itself using VarType = VAL is only possible, if the section property is constant and
identical for all elements.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1872
VTSEC
An error message is generated and the VTSEC command is ignored when the current value is outside the
variation range.
If the Order is not specified, it will be increased until the accuracy requirements of the VTRSLT (p. 1868)
command are satisfied. When you set a value for Order, no further adjustments for accuracy will be performed.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Section Property
VTSFE, Name, MIN, MAX, RedOpt, Lab, LKEY, ElComp, VarType, Order
Defines a surface load as an input variable for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name of the input variable. This is a string of up to 256 characters used for postprocessing purposes.
Blanks are allowed in the string.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
RedOpt
Reduction option that specifies what to do if the DesignXplorer needs to reduce the variation range
from MIN to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
Lab
Valid surface load label. This label must be defined using the SFE or SF command prior to issuing this
command. Applicable labels are:
PRES
Pressure, Force
HFILM
Convection surface loads (film coefficient)
TBULK
Convection surface loads (bulk temperature)
HFLUX
Heat flux
1873
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTSFE
LKEY
Load key associated with surface load (not applicable when Lab is different from PRES). Valid load keys
are: 1,2, 3, 4, or 5 for element SURF154 and 1,2, 3, or 4 for element SURF153.
ElComp
Name of an element component or number of a single individual element. If ElComp = ALL, then all
elements are considered for the VTSFE command.
VarType
Label indicating how the variability of surface load should be handled. Valid labels are:
FACT
The variability of the surface load is described by a scale factor that is applied to the load values.
This means that the MIN field is the value of the minimum factor to be applied and the MAX field is
the maximum factor to be applied. As such the values specified for MIN and MAX are unitless
quantities. VarType = FACT is the default for surface load parameters.
ADD
The variability of the surface load is described by an add-on that is applied to the load values. This
means that the MIN field is the minimum value that is added to the load values and the MAX field
contains the maximum value that is added to the surface load values. Hence, the values specified
for MIN and MAX have the same unit as the surface load itself.
Order
Derivation order. By default, driven by the application.
Notes
Defines the surface load of a single element or a set of elements as an input variable of the DesignXplorer.
The DesignXplorer will evaluate the derivatives of the result parameter(s) (see VTRSLT command) with respect
to the surface load defined here. The derivatives are used to determine an approximation function for the
result parameter(s) as a function of the surface load in the range specified by the MIN and MAX fields.
The type of the surface load is defined by the Lab field, which must not be left blank. The surface load must
exist or be specified using the SFE command prior to issuing the VTSFE command.
Using the ElComp field a set of elements is selected for which the surface load is varied.
If the Order is not specified, it will be increased until the accuracy requirements of the VTRSLT command
are satisfied. When you set a value for Order, no further adjustments for accuracy will be performed. Order
should only be used when an input parameter has an influence with a limited order.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1874
VTSFE
VTSL, Type, Varname
Selects a subset of elements associated with an VT input variable.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Type
Label identifying the type of select:
S
Select a new set. This is the default.
R
Reselect a set from the current set.
A
Additionally select a set and extend the current set.
U
Unselect a set from the current set.
Varname
Name of an VT input variable, defined with the VTMP, VTREAL, VTSEC or VTDISC commands.
Command Default
There are no default values for the command fields.
Notes
This command does not apply to output variables or VTGEOM input variables.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu> DesignXplorer VT> Setup> Frequency
VTSTAT, EntyLis
Print the status of the DesignXplorer definitions and settings into a separate window.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
EntyLis
Option that specifies whether the nodes or element lists should be printed for those Variational Technology
commands where nodes or element lists have been specified. Valid labels are:
NO
Do not print out the list of nodes or elements (default).
YES
Print out the list of nodes or elements.
1875
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTSTAT
Notes
Prints the status the user definitions and settings of the DesignXplorer to a separate window.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Other>Status
VTTEMP,Name, MIN, MAX, RedOpt, --, VarType, Order
Defines the temperature as input variable for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name for the input variable. This is a string of up to 250 characters used for postprocessing purposes.
Blanks are allowed in the string.
MIN
Minimum value of the input variable.
MAX
Maximum value of the input variable.
RedOpt
Reduction option that specifies what to do if the DesignXplorer needs to reduce the variation range
from MIN to MAX. Valid labels are:
CONT
Continue with the calculation of the derivatives. This is the default.
STOP
Stop the calculation of the derivatives. This option is necessary if a reduction of the variation range
is not acceptable.
--
Unused field.
VarType
Label indicating how the variability of temperature should be handled. Valid labels are:
FACT
The variability of the temperature is described by an scaling factor that is applied to the temperature.
This means that the MIN field is the value of the minimum factor to be applied and the MAX field is
the maximum factor to be applied. Hence, the values specified for MIN and MAX are unitless quant-
ities. VarType = FACT is the default for temperature parameters.
ADD
The variability of the temperature is described by an add-on that is applied to the temperature values.
This means that the MIN field is the minimum value that is added to the temperature values and
the MAX field contains the maximum value that is added to the temperature values. Hence, the values
specified for MIN and MAX have the same unit as the temperature itself.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1876
VTTEMP
Order
Derivation Order. By default, driven by the application.
Command Default
Defaults as outlined above.
Notes
Defines the temperature as an input variable.
If the Order is not specified, it will be increased until the accuracy requirements of the VTRSLT command
are satisfied. When you set a value for Order, no further adjustments for accuracy will be performed.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Temperature Load
VTVMOD, Name, Oper, Value
Modifies the status or current value of an input variable for the DesignXplorer.
VARIATIONAL TECHNOLOGY: DesignXplorer
<> <> <> <> <> <> <> <> <> <> <> <> VT <> <>
Name
Name for an input or results variable of the DesignXplorer. This is a string of up to 250 characters used
for postprocessing purposes. Blanks can be used. An input variable with this name must be previously
defined using one of the following commands: VTFREQ, VTMP, VTRSLT,
Oper
Label for the type of operation that is performed on the input variable Name. Valid labels are:
ACT
Activates an input variable in the DesignXplorer. This operation is ignored, if the variable specified
is already activated.
DEACT
Deactivates an input variable of the DesignXplorer. This operation is ignored, if the variable specified
is not activated.
DEL
Deletes an input or results variable of the DesignXplorer.
SET
Sets the value of an input variable of the DesignXplorer.
Value
Is the value of the input variable for the VTEVAL command.
1877
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
VTVMOD
Notes
Modifies the status of an input variable for the DesignXplorer that have been previously defined, by activating,
deactivating or deleting it.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>DesignXplorer VT>Setup>Modify
VTYPE, NOHID, NZONE
Specifies the viewing procedure used to determine the form factors for the Radiation Matrix method.
AUX12: Radiation Matrix Method
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
NOHID
Type of viewing procedure:
0
Hidden procedure.
1
Non-hidden (faster, but less general) procedure.
NZONE
Number of sampling zones for the hidden procedure (100 maximum for 3-D, 1000 maximum for 2-D).
Defaults to 20 for 3-D, 200 for 2-D. Number of points is 2*NZONE for 2-D and 2*NZONE*(NZONE+1) for
3-D.
Command Default
Hidden procedure using 20 zones for 3-D, 200 zones for 2-D.
Menu Paths
Main Menu>Radiation Opt>Matrix Method>Write Matrix
/VUP, WN, Label
Specifies the global Cartesian coordinate system reference orientation.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number (or ALL) to which command applies (defaults to 1).
Label
Orientation:
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1878
VTYPE
Y
Y vertical upward, X horizontal to the right, Z out from the screen (default).
-Y
Y vertical downward, X horizontal to the left, Z out from the screen.
X
X vertical upward, Y horizontal to the left, Z out from the screen.
-X
X vertical downward, Y horizontal to the right, Z out from the screen.
Z
Z vertical upward, Y horizontal to the right, X out from the screen. With this choice, you should use
a view other than the /VIEW default of (0,0,1).
-Z
Z vertical downward, Y horizontal to the left, X out from the screen. With this choice, you should
use a view other than the /VIEW default of (0,0,1).
Command Default
Y vertical upward, X horizontal to the right, Z out from the screen.
Notes
Specifies the global Cartesian coordinate system reference orientation. The /VIEW and /ANGLE commands
may be used to reorient the view and are relative to this reference orientation. All coordinate systems are
right-handed.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>View Settings>Viewing Direction
*VWRITE, Par1, Par2, Par3, Par4, Par5, Par6, Par7, Par8, Par9, Par10, Par11, Par12, Par13, Par14, Par15, Par16,
Par17, Par18, Par19
Writes data to a file in a formatted sequence.
APDL: Array Parameters
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
Par1, Par2, Par3, . . . , Par19
You can write up to 19 parameters (or constants) at a time. Any Par values after a blank Par value are
ignored. If you leave them all blank, one line will be written (to write a title or a blank line). If you input
the keyword SEQU, a sequence of numbers (starting from 1) will be written for that item.
1879
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
*VWRITE
Notes
You use *VWRITE to write data to a file in a formatted sequence. Data items (Par1, Par2, etc.) may be array
parameters, scalar parameters, character parameters (scalar or array), or constants. You must evaluate expres-
sions and functions in the data item fields before using the *VWRITE command, since initially they will be
evaluated to a constant and remain constant throughout the operation. Unless a file is defined with the
*CFOPEN command, data is written to the standard output file. Data written to the standard output file
may be diverted to a different file by first switching the current output file with the /OUTPUT command.
You can also use the *MWRITE command to write data to a specified file. Both commands contain format
descriptors on the line immediately following the command. The format descriptors can be in either Fortran
or C format.
You must enclose Fortran format descriptors in parentheses. They must immediately follow the *VWRITE
command on a separate line of the same input file. Do not include the word FORMAT. The format must
specify the number of fields to be written per line, the field width, the placement of the decimal point, etc.
You should use one field descriptor for each data item written. The write operation uses your system's
available FORTRAN FORMAT conventions (see your system FORTRAN manual). You can use any standard
FORTRAN real format (such as (4F6.0), (E10.3,2X,D8.2), etc.) and alphanumeric format (A). Alphanumeric
strings are limited to a maximum of 8 characters for any field (A8) using the Fortran format. Use the C
format for string arrays larger than 8 characters. Integer (I) and list-directed (*) descriptors may not be used.
You can include text in the format as a quoted string. The parentheses must be included in the format and
the format must not exceed 80 characters (including parentheses). The output line length is limited to 128
characters.
The C format descriptors are used if the first line of the format descriptor is not a left parenthesis. C
format descriptors are up to 80 characters long, consisting of text strings and predefined "data descriptors"
between the strings where numeric or alphanumeric character data will be inserted. The normal descriptors
are %I for integer data, %G for double precision data, %C for alphanumeric character data, and %/ for a line
break. Each descriptor must be preceded by a blank. There must be one data descriptor for each specified
value (8 maximum) in the order of the specified values. The enhanced formats described in *MSG may also
be used.
For array parameter items, you must define the starting array element number. Looping continues (incre-
menting the vector index number of each array parameter by one) each time you output a line, until the
maximum array vector element is written. For example, *VWRITE,A(1) followed by (F6.0) will write one value
per output line, i.e., A(1), A(2), A(3), A(4), etc. You write constants and scalar parameters with the same values
for each loop. You can also control the number of loops and loop skipping with the *VLEN and *VMASK
commands. The vector specifications *VABS, *VFACT, and *VCUM do not apply to this command. If looping
continues beyond the supplied data array's length, zeros will be output for numeric array parameters and
blanks for character array parameters. For multi-dimensioned array parameters, only the first (row) subscript
is incremented. See the *VOPER command for details. If you are in the GUI, the *VWRITE command must
be contained in an externally prepared file and read into ANSYS (i.e., *USE, /INPUT, etc.).
This command is valid in any processor.
Menu Paths
Utility Menu>Parameters>Array Parameters>Write to File
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1880
*VWRITE
W Commands
/WAIT, DTIME
Causes a delay before the reading of the next command.
APDL: Process Controls
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Argument Descriptions
DTIME
Time delay (in seconds). Maximum time delay is 59 seconds.
Notes
The command following the /WAIT will not be processed until the specified wait time increment has elapsed.
Useful when reading from a prepared input file to cause a pause, for example, after a display command so
that the display can be reviewed for a period of time. Another "wait" feature is available via the *ASK com-
mand.
This command is valid in any processor.
Menu Paths
This command cannot be accessed from a menu.
WAVES, Wopt, OLDMAX, OLDRMS
Initiates reordering.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Wopt
Option for comparison:
MAX
Use maximum wavefront value for comparison (default).
RMS
Use RMS wavefront value for comparison.
OLDMAX, OLDRMS
Maximum and RMS wavefront values of model to be used in place of the old values. OLDRMS defaults
to OLDMAX (and vice versa). If neither is specified, each defaults to its calculated old value.
1881
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Notes
Initiates the element reordering based upon the previously defined starting wave lists (if any). Reordering
affects only the element order for the solution phase and not the element numbers (input referring to element
numbers, such as element pressures, is unaffected by reordering).
Note
The new order is retained only if new the new maximum or RMS wavefront values are lower than
the old values, as described below. See the WSORT command for another reordering procedure.
The resulting element ordering can be shown by listing the wavefront history [WFRONT,1] or by
displaying elements with their element location numbers [/PNUM].
The WAVES reordering procedure is based upon a list of nodes defining where the element reordering is
to start (either input on the WSTART command, or automatically determined if not input). If a list is input,
additional starting wave lists may be defined to allow the user to block or guide the wave. An input starting
list usually consists of one node for a line element model, a line of nodes for an area element model, or a
plane of nodes for a volume element model. Elements are ordered from the first starting wave list in a dir-
ection that attempts to minimize the wavefront. Nodes are deleted and added to the total wave list set as
reordering progresses through the model. Elements attached to nodes on succeeding starting lists, if any,
are ignored at this time, thereby temporarily blocking that path. Whenever no more elements can be re-
ordered, the procedure is repeated starting from the nodes defined on the second starting list (if any). The
same node number should not appear on more than one list. If no starting list is input, a starting node is
automatically selected from an element weighting procedure (see Element Reordering in the Theory Reference
for the Mechanical APDL and Mechanical Applications).
Reordering may be made from the previously reordered model or from the initial model (by issuing the
SAVE command before reordering and then restoring that model back to PREP7 with the RESUME command
after reordering). The WFRONT command is useful for calculating and printing the current wavefront statistics
at any time without causing any reordering of the elements. If a model is to be used for both a thermal and
a structural analysis, the reordered statistics should be based upon the structural model (because of the
higher number of degrees of freedom per node).
The reordering procedure treats separate portions of the model (i.e., not connected by elements) as discon-
tinuous. Reordering automatically continues across a discontinuity as best as possible.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Reorder by List
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1882
WAVES
WERASE
Erases all reordering wave lists.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Notes
The REORDER then STAT commands will display the current wave lists.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Erase Wave List
WFRONT, KPRNT, KCALC
Estimates wavefront statistics.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
KPRNT
Wavefront history print key:
0
Print current wavefront statistics.
1
Print current wavefront statistics but also print wavefront history (wavefront at each element). Elements
are listed in the reordered sequence.
KCALC
Calculation options:
0
Wavefront estimate assumes maximum model DOF set at each node and does not include the effects
of master degrees of freedom and specified displacement constraints.
1
Wavefront estimate uses the actual DOF set at each node and does not include the effects of master
degrees of freedom and specified displacement constraints. More time consuming than estimated
wavefront. KPRNT = 1 is not available with this option.
Notes
Estimates wavefront statistics of the model as currently ordered.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1883
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
WFRONT
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Est. Wavefront
/WINDOW, WN, XMIN, XMAX, YMIN, YMAX, NCOPY
Defines the window size on the screen.
GRAPHICS: Set Up
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window reference number (1 to 5). Defaults to 1. This number, or ALL (for all active windows), may be
used on other commands.
XMIN, XMAX, YMIN, YMAX
Screen coordinates defining window size. Screen coordinates are measured as -1.0 to 1.67 with the origin
at the screen center. For example, (-1,1.67,-1,1) is full screen, (-1,0,-1,0) is the left bottom quadrant. If
XMIN = OFF, deactivate this previously defined window; if ON, reactivate this previously defined window.
If FULL, LEFT, RIGH, TOP, BOT, LTOP, LBOT, RTOP, RBOT, form full, half, or quarter window. If SQUA, form
largest square window within the current graphics area. If DELE, delete this window (cannot be reactivated
with ON).
NCOPY
Copies the current specifications from window NCOPY (1 to 5) to this window. If NCOPY = 0 (or blank),
no specifications are copied.
Command Default
One window at full screen.
Notes
Defines the window size on the screen. Windows may occupy a separate section of the screen or they may
overlap. Requested displays are formed in all windows according to the selected window specifications.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Window Controls>Copy Window Specs
Utility Menu>PlotCtrls>Window Controls>Delete Window
Utility Menu>PlotCtrls>Window Controls>Window Layout
Utility Menu>PlotCtrls>Window Controls>Window On or Off
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1884
/WINDOW
WMID, Key
Specifies reordering options for the WAVES command.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Key
Determines whether midside nodes are considered when reordering.
NO
Do not consider midside nodes when reordering (default).
YES
Consider midside nodes when reordering. This option is useful for models where line elements are
only attached to midside nodes of solid elements.
Notes
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
This command cannot be accessed from a menu.
WMORE, NODE1, NODE2, NINC, ITIME, INC
Adds more nodes to the starting wave list.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
NODE1, NODE2, NINC
Add another node set to the previous starting wave list. Set is NODE1 to NODE2 (defaults to NODE1) in
steps of NINC (defaults to 1). If NODE1 is negative, delete (instead of add) this node set from previous
starting wave list.
ITIME, INC
Add other node sets to the same starting wave list by repeating the previous node set with NODE1 and
NODE2 incremented by INC (defaults to 1) each time after the first. ITIME is the total number of sets
(defaults to 1) defined with this command.
Notes
Adds more nodes to (or modifies) the previous starting wave list (if any) [WSTART]. Repeat WMORE command
to add more nodes to the previous starting wave list. Up to 10,000 nodes may be defined (total, for all
starting waves).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Extend Wave List
1885
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
WMORE
WPAVE, X1, Y1, Z1, X2, Y2, Z2, X3, Y3, Z3
Moves the working plane origin to the average of specified points.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
X1, Y1, Z1
Coordinates (in the active coordinate system) of the first point. If X1 = P, graphical picking is enabled
and all remaining command fields are ignored (valid only in the GUI).
X2, Y2, Z2
Coordinates (in the active coordinate system) of the second point.
X3, Y3, Z3
Coordinates (in the active coordinate system) of the third point.
Notes
Moves the origin of the working plane to the average of the specified points. A point is considered specified
only if at least one of its coordinates is non-blank, and at least one point (1, 2, or 3) must be specified. Blank
coordinates of a specified point are assumed to be zero. Averaging is based on the active coordinate system.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Offset WP to>Global Origin
Utility Menu>WorkPlane>Offset WP to>Origin of Active CS
Utility Menu>WorkPlane>Offset WP to>XYZ Locations
WPCSYS, WN, KCN
Defines the working plane location based on a coordinate system.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number whose viewing direction will be modified to be normal to the working plane (defaults
to 1). If WN is a negative value, the viewing direction will not be modified.
KCN
Coordinate system number. KCN may be 0,1,2 or any previously defined local coordinate system number
(defaults to the active system).
Notes
Defines a working plane location and orientation based on an existing coordinate system. If a Cartesian
system is used as the basis (KCN) for the working plane, the working plane will also be Cartesian, in the X-
Y plane of the base system. If a cylindrical, spherical, or toroidal base system is used, the working plane will
be a polar system in the R- plane of the base system.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1886
WPAVE
If working plane tracking has been activated (CSYS,WP or CSYS,4), the updated active coordinate system
will be of a similar type, except that a toroidal system will be updated to a cylindrical system. See the Mod-
eling and Meshing Guide for more information on working plane tracking.
This command is valid in any processor.
Menu Paths
Main Menu>General Postproc>Surface Operations>Create Surface>Sphere>At Node
Utility Menu>WorkPlane>Align WP with>Active Coord Sys
Utility Menu>WorkPlane>Align WP with>Global Cartesian
Utility Menu>WorkPlane>Align WP with>Specified Coord Sys
WPLANE, WN, XORIG, YORIG, ZORIG, XXAX, YXAX, ZXAX, XPLAN, YPLAN, ZPLAN
Defines a working plane to assist in picking operations.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number whose viewing direction will be modified to be normal to the working plane (defaults
to 1). If WN is a negative value, the viewing direction will not be modified. If fewer than three points are
used, the viewing direction of window WN will be used instead to define the normal to the working
plane.
XORIG, YORIG, ZORIG
Global Cartesian coordinates of the origin of the working plane coordinate system.
XXAX, YXAX, ZXAX
Global Cartesian coordinates of a point defining the x-axis orientation. The x-axis aligns with the projection
of the line from this orientation point to the origin.
XPLAN, YPLAN, ZPLAN
Global Cartesian coordinates of the third point defining the working plane. This point will also define
the location of the positive XY-sector of the working plane coordinate system.
Command Default
Working plane parallel to the global X-Y plane at Z = 0.0.
Notes
Defines a working plane to assist in picking operations using the coordinates of three noncolinear points.
The three points also define the working plane coordinate system. A minimum of one point (the working
plane origin) is required. Immediate mode may also be active. See WPSTYL command to set the style of
working plane display.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Align WP with>XYZ Locations
1887
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
WPLANE
WPOFFS, XOFF, YOFF, ZOFF
Offsets the working plane.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XOFF, YOFF, ZOFF
Offset increments defined in the working plane coordinate system. If only ZOFF is used, the working
plane will be redefined parallel to the present plane and offset by ZOFF.
Notes
Changes the origin of the working plane by translating the working plane along its coordinate system axes.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Offset WP by Increments
WPROTA, THXY, THYZ, THZX
Rotates the working plane.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
THXY
First rotation about the working plane Z axis (positive X toward Y).
THYZ
Second rotation about working plane X axis (positive Y toward Z).
THZX
Third rotation about working plane Y axis (positive Z toward X).
Notes
The specified angles (in degrees) are relative to the orientation of the working plane.
This command is valid in any processor.
Menu Paths
Utility Menu>WorkPlane>Offset WP by Increments
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1888
WPOFFS
WPSTYL, SNAP, GRSPAC, GRMIN, GRMAX, WPTOL, WPCTYP, GRTYPE, WPVIS, SNAPANG
Controls the display and style of the working plane.
DATABASE: Working Plane
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
SNAP
Snap increment for a locational pick (1E-6 minimum). If -1, turn off snap capability. For example, a picked
location of 1.2456 with a snap of 0.1 gives 1.2, with 0.01 gives 1.25, with 0.001 gives 1.246, and with
0.025 gives 1.250 (defaults to 0.05).
GRSPAC
Graphical spacing between grid points. For graphical representation only and not related to snap points
(defaults to 0.1).
GRMIN, GRMAX
Defines the size of a square grid (if WPCTYP = 0) to be displayed over a portion of the working plane.
The opposite corners of the grid will be located at grid points nearest the working plane coordinates of
(GRMIN,GRMIN) and (GRMAX,GRMAX). If a polar system (WPCTYP = 1), GRMAX is the outside radius of
grid and GRMIN is ignored. If GRMIN = GRMAX, no grid will be displayed (defaults to -1.0 and 1.0 for
GRMIN and GRMAX respectively).
WPTOL
The tolerance that an entity's location can deviate from the specified working plane, while still being
considered on the plane. Used only for locational picking of vertices for polygons and prisms (defaults
to 0.003).
WPCTYP
Working plane coordinate system type:
0
Cartesian (default). If working plane tracking is on [CSYS,4], the updated active coordinate system
will also be Cartesian.
1
Polar. If working plane tracking is on, the updated active coordinate system will be cylindrical.
2
Polar. If working plane tracking is on, the updated active coordinate system will be spherical.
GRTYPE
Grid type:
0
Grid and WP triad.
1
Grid only.
2
WP triad only (default).
WPVIS
Grid visibility:
0
Do not show GRTYPE entities (grid and/or triad) (default).
1889
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
WPSTYL
1
Show GRTYPE entities. Cartesian working planes will be displayed with a Cartesian grid, polar with
a polar grid.
SNAPANG
Snap angle (0--180) in degrees. Used only if WPCTYP = 1 or 2. Defaults to 5 degrees.
Command Default
Snap capability on with an increment of .05, spacing between grid points of 0.1, Cartesian coordinate system,
WP triad displayed, and a tolerance of .003.
Notes
Use WPSTYL,DEFA to reset the working plane to its default location and style. Use WPSTYL,STAT to list the
status of the working plane. Blank fields will keep present settings.
It is possible to specify SNAP and WPTOL values that will cause conflicts during picking operations. Check
your values carefully, and if problems are noted, revert to the default values.
WPSTYL with no arguments will toggle the grid on and off. The working plane can be displayed in the non-
GUI interactive mode only after issuing a /PLOPTS,WP,1 command. See the Modeling and Meshing Guide for
more information on working plane tracking. See /PLOPTS command for control of hidden line working
plane.
This command is valid in any processor.
Menu Paths
Main Menu>Preprocessor>Modeling>Create>Circuit>Builder>ROM>ElecStruc
Main Menu>Preprocessor>Modeling>Create>Circuit>Set Grid
Utility Menu>List>Status>Working Plane
Utility Menu>WorkPlane>Display Working Plane
Utility Menu>WorkPlane>Offset WP by Increments
Utility Menu>WorkPlane>Show WP Status
Utility Menu>WorkPlane>WP settings
WRFULL, Ldstep
Stops solution after assembling global matrices.
SOLUTION: Load Step Options
MP ME ST PR PRN <> <> <> EM EH <> PP <> EME MFS
Ldstep
Specify action to take:
OFF or 0
Turn off feature (default)
N
Turn on feature and set it to stop after assembling the global matrices and writing the .FULL file
for load step N.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1890
WRFULL
Command Default
By default the WRFULL command is turned OFF.
Notes
This command is used in conjunction with the SOLVE command to generate the assembled matrix file
(.FULL file) only. The element matrices are assembled into the relevant global matrices for the particular
analysis being performed and the .FULL file is written. Equation solution and the output of data to the
results file are skipped. To dump the matrices written on the .FULL file into Harwell-Boeing format, use
the HBMAT command in /AUX2. To copy the matrices to a postscript format that can be viewed graphically,
use the PSMAT command.
To use the LSSOLVE macro with this command, you may need to modify the LSSOLVE macro to properly
stop at the load step of interest.
This command only valid for linear static, full harmonic, and full transient analyses when the sparse direct
solver is selected. This command is also valid for buckling or modal analyses with any mode extraction
method. This command is not valid for nonlinear analyses or analyses containing p-elements.
In general, the assembled matrix file .FULL contains stiffness, mass, and damping matrices. However, the
availability of the matrices depends on the analysis type chosen when the file is written.
Menu Paths
Main Menu>Preprocessor>Loads>Load Step Opts>Stop Solution
Main Menu>Solution>Load Step Opts>Stop Solution
WRITE, Fname
Writes the radiation matrix file.
AUX12: Radiation Matrix Method
MP ME ST PR PRN <> <> <> <> <> <> PP <> EME MFS
Fname
File name and directory path (248 characters maximum, including the characters needed for the directory
path). An unspecified directory path defaults to the working directory; in this case, you can use all 248
characters for the file name.
The file name Defaults to Jobname.
Notes
Writes radiation matrix file (File.SUB) for input to the substructure thermal "use" pass. Subsequent WRITE
operations to the same file overwrite the file.
Menu Paths
Main Menu>Radiation Opt>Matrix Method>Write Matrix
1891
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
WRITE
WSORT, Lab, KORD, --, Wopt, OLDMAX, OLDRMS
Initiates element reordering based upon a geometric sort.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
Lab
Coordinate (in the active system) along which element centroid locations are sorted. Valid labels are: X,
Y, Z, ALL. If ALL (default), all three directions will be used, and the order corresponding to the lowest
MAX or RMS wavefront value will be retained.
KORD
Sort order:
0
Sort according to ascending coordinate values.
1
Sort according to descending coordinate values.
--
Unused field.
Wopt
Option for comparison:
MAX
Use maximum wavefront value for comparison (default).
RMS
Use RMS wavefront value.
OLDMAX, OLDRMS
MAX and RMS wavefront values of model to be used in place of the old values. OLDRMS defaults to
OLDMAX (and vice versa). If neither is specified, each defaults to its calculated old value.
Notes
Initiates element reordering based upon a geometric sort of the element centroid locations. Wave lists, if
any [WSTART], are ignored. Reordering affects only the element order for the solution phase and not the
element numbers (input referring to element numbers, such as element pressures, is unaffected by reordering).
Note
The new order is retained only if new the new maximum or RMS wavefront values are lower than
the old values, as described below. See the WAVES command for another reordering procedure
and for more details on reordering. The resulting element ordering can be shown by listing the
wavefront history [WFRONT,1] or by displaying elements with their element location numbers
[/PNUM].
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Reorder by XYZ
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1892
WSORT
WSPRINGS
Creates weak springs on corner nodes of a bounding box of the currently selected elements.
SOLUTION: Misc Loads
MP ME ST PR PRN DS DSS <> <> <> <> PP <> EME MFS
Notes
WSPRINGS invokes a predefined ANSYS macro that is used during the import of loads from the ADAMS
program into the ANSYS program. WSPRINGS creates weak springs on the corner nodes of the bounding
box of the currently selected elements. The six nodes of the bounding box are attached to ground using
COMBIN14 elements. The stiffness is chosen as a small number and can be changed by changing the real
constants of the COMBIN14 elements. This command works only for models that have a geometric extension
in two or three dimensions. One dimensional problems (pure beam in one axis) are not supported.
For more information on how WSPRINGS is used during the transfer of loads from the ADAMS program to
ANSYS, see Import Loads into ANSYS in the Advanced Analysis Techniques Guide.
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
Menu Paths
Main Menu>Solution>ADAMS Connection>Import fr ADAMS
WSTART, NODE1, NODE2, NINC, ITIME, INC
Defines a starting wave list.
PREP7: Element Reordering
MP ME ST PR PRN <> <> <> EM <> <> PP <> EME MFS
NODE1, NODE2, NINC
Define a set of nodes in the starting wave list from NODE1 to NODE2 (defaults to NODE1) in steps of
NINC (defaults to 1). If NODE1 = ALL, ignore remaining fields and use all selected nodes [NSEL].
ITIME, INC
Add more node sets to the same starting wave list by repeating the previous node set with NODE1 and
NODE2 incremented by INC (defaults to 1) each time after the first. ITIME is the total number of sets
(defaults to 1) defined with this command.
Command Default
If no starting waves, a starting node will be automatically defined when the WAVES command is issued.
Notes
Defines a starting wave list (optional) for reordering with the WAVES command. Repeat WSTART command
to define other starting wave lists (20 maximum).
Distributed ANSYS Restriction This command is not supported in Distributed ANSYS.
1893
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
WSTART
Menu Paths
Main Menu>Preprocessor>Numbering Ctrls>Element Reorder>Define Wave List
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1894
WSTART
X Commands
/XFRM, LAB, X1, Y1, Z1, X2, Y2, Z2
Controls the centroid or the axis of dynamic rotation.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
LAB
The location or entity (centroid) used to define the center or axis of rotation.
NODE
If NODE is chosen for the center of rotation, the node number will be X1. If the rotation is to be
about an axis, then X1 and Y1 define the two nodes between which a line is drawn to determine
the axis. The remaining arguments are ignored.
ELEMENT
If ELEMENT is chosen for the center of rotation, the element number will be X1. If the rotation is to
be about an axis, then X1 and Y1 define the two elements between which a line is drawn to determine
the axis. The remaining arguments are ignored.
KP
If KP is chosen for the center of rotation, the keypoint number will be X1. If the rotation is to be
about an axis, then X1 and Y1 define the two keypoints between which a line is drawn to determine
the axis. The remaining arguments are ignored.
LINE
If LINE is chosen for the center of rotation, the line number will be X1. If the rotation is to be about
an axis, then X1 and Y1 define the two lines between which a line is drawn to determine the axis.
The remaining arguments are ignored.
AREA
If AREA is chosen for the center of rotation, the area number will be X1. If the rotation is to be about
an axis, then X1 and Y1 define the two areas between which a line is drawn to determine the axis.
The remaining arguments are ignored.
VOLUME
If VOLUME is chosen for the center of rotation, the volume number will be X1. If the rotation is to
be about an axis, then X1 and Y1 define the two volumes between which a line is drawn to determine
the axis. The remaining arguments are ignored.
XYZ
If XYZ is chosen for the center of rotation, the location of that center is determined by the coordinates
X1, Y1, Z1. If values are specified for X2, Y2, Z2, then the axis of rotation will be about the line
between those two points.
OFF
If LAB = OFF, DEFAULT, FOCUS or if no value is specified, then the center of rotation is set at the
FOCUS point, as defined by the /FOCUS command.
1895
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
X1
The entity number or X coordinate for the center of rotation.
Y1
The entity number or Y coordinate for the center of rotation.
Z1
The Z coordinate for the center of rotation.
X2
The X coordinate for the axis of rotation.
Y2
The Y coordinate for the axis of rotation.
Z2
The Z coordinate for the axis of rotation.
Command Default
Issuing /XFRM, with no LAB defined sets the center of rotation at the focal point specified by the /FOCUS
command.
Notes
The /XFRM command is active only when the cumulative rotation key is specified ON for the /ANGLE com-
mand (KINCR = 1). This command affects dynamic manipulations only.
For center rotation, the middle mouse button will rotate the model about the screen Z axis and the right
mouse button will rotate the model about the screen X and Y axis.
For rotation about an axis, the middle mouse button will rotate the model about the defined axis of rotation
and the right mouse button will be deactivated.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>ViewSettings>Rotational Center>ByPick
/XRANGE, XMIN, XMAX
Specifies a linear abscissa (X) scale range.
GRAPHICS: Graphs
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
XMIN
Minimum abscissa scale value.
XMAX
Maximum abscissa scale value.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1896
/XRANGE
Command Default
Automatically select X-range scale to include all data being displayed.
Notes
Specifies a linear abscissa (X) scale range for the line graph display. Use /XRANGE,DEFAULT to return to
automatic scaling.
Automatic scaling will often yield inappropriate range values for logarithmic scales (/GROPT, LOGX).
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Axes
XVAR, N
Specifies the X variable to be displayed.
POST26: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
N
X variable number:
0 or 1
Display PLVAR values vs. time (or frequency).
n
Display PLVAR values vs. variable n (2 to NV [NUMVAR]).
1
Interchange time and PLVAR variable numbers with time as the curve parameter. PLVAR variable
numbers are displayed uniformly spaced along X-axis from position 1 to 10.
Command Default
Use time or frequency.
Notes
Defines the X variable (displayed along the abscissa) against which the Y variable(s) [PLVAR] are to be dis-
played.
Menu Paths
Main Menu>TimeHist Postpro>Settings>Graph
1897
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
XVAR
XVAROPT, Lab
Specifies the parameter to be used as the X-axis variable.
OPTIMIZATION: Display
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
Lab
Parameter to be used as the X variable. Defaults to set numbers.
Command Default
Use design set numbers as the X variable.
Notes
Specifies the parameter to be used as the X-axis variable for graphical displays produced by the PLVAROPT
command, and for the first column of the tabular listings produced by the PRVAROPT command. Design
sets are automatically sorted in a sequence corresponding to an ascending order of the specified parameter.
Menu Paths
Main Menu>Design Opt>Design Sets>Graphs/Tables
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1898
XVAROPT
Y Commands
/YRANGE, YMIN, YMAX, NUM
Specifies a linear ordinate (Y) scale range.
GRAPHICS: Graphs
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
YMIN
Minimum ordinate scale value.
YMAX
Maximum ordinate scale value.
NUM
Y-axis number to which range applies (defaults to 1). Valid numbers are 1 to 3 for /GRTYP,2 and 1 to 6
for /GRTYP,3. If ALL, apply to all Y-axes.
Command Default
Automatically select Y-range scale to include all data being displayed.
Notes
Specifies a linear ordinate (Y) scale range for the line graph display. Use /YRANGE,DEFAULT to return to
automatic scaling. For multiple Y-axes graphs [/GRTYP], see /GROPT, ASCAL to automatically scale the addi-
tional Y-axes.
Automatic scaling will often yield inappropriate range values for logarithmic scales (/GROPT, LOGY).
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Style>Graphs>Modify Axes
1899
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1900
Z Commands
/ZOOM, WN, Lab, X1, Y1, X2, Y2
Zooms a region of a display window.
GRAPHICS: Views
MP ME ST PR PRN <> <> FL EM EH DY PP <> EME MFS
WN
Window number to which command applies (defaults to 1).
Lab
Label to define the desired type of zoom:
OFF
Turns zoom off (refits image of entire model to the window).
BACK
Goes back to previous zoom setting (five successive back ups, maximum).
SCRN
Interprets X1,Y1 as the screen coordinates of the center of a square zoom region; X2,Y2 as the screen
coordinates of a point on one side of that square.
RECT
Interprets X1,Y1 and X2,Y2 as the screen coordinates of two opposite corners of a rectangular zoom
region.
Notes
Zooms (centers and magnifies) the specified region of a display window. /ZOOM will operate on a display
that has been formed by an explicit graphics action command (APLOT, EPLOT, etc.). /ZOOM has no effect
on an "immediate" graphics display. When /ZOOM is executed, the display is automatically replotted such
that the specified zoom region is centered and sized to fill the window.
Auto resizing is disabled when you issue the /ZOOM command. To restore auto resizing, issue the /AUTO
command, or select FIT from the Pan, Zoom, Rotate box.
This command is valid in any processor.
Menu Paths
Utility Menu>PlotCtrls>Pan, Zoom, Rotate
1901
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1902
Index
Symbols
*ABBR command, 93
*AFUN command, 107
*ASK command, 169
*CFCLOS command, 254
*CFOPEN command, 255
*CFWRITE command, 256
*CREATE command, 316
*CYCLE command, 331
*DEL command, 364
*DIM command, 379
*DO command, 401
*DOWHILE command, 405
*ELSE command, 508
*ELSEIF command, 509
*END command, 519
*ENDDO command, 519
*ENDIF command, 520
*EXIT command, 570
*GET command, 711
*GO command, 771
*IF command, 829
*LIST command, 923
*MFOURI command, 992
*MFUN command, 1001
*MOPER command, 1017
*MSG command, 1048
*MWRITE command, 1065
*REPEAT command, 1406
*RETURN command, 1415
*SET command, 1522
*SREAD command, 1619
*STATUS command, 1633
*TAXIS command, 1657
*TOPER command, 1729
*TREAD command, 1739
*ULIB command, 1771
*USE command, 1778
*VABS command, 1789
*VCOL command, 1794
*VCUM command, 1796
*VEDIT command, 1801
*VFACT command, 1804
*VFILL command, 1805
*VFUN command, 1809
*VGET command, 1813
*VITRP command, 1823
*VLEN command, 1824
*VMASK command, 1827
*VOPER command, 1829
*VPLOT command, 1833
*VPUT command, 1835
*VREAD command, 1839
*VSCFUN command, 1846
*VSTAT command, 1850
*VWRITE command, 1879
/AN3D command, 116
/ANFILE command, 123
/ANGLE command, 125
/ANNOT command, 129
/ANUM command, 146
/ASSIGN command, 172
/AUTO command, 176
/AUX12 command, 179
/AUX15 command, 179
/AUX2 command, 178
/AUX3 command, 178
/AXLAB command, 182
/BATCH command, 183
/CFORMAT command, 256
/CLABEL command, 264
/CLEAR command, 265
/CLOG command, 268
/CMAP command, 271
/COLOR command, 294
/COM command, 297
/CONFIG command, 299
/CONTOUR command, 302
/COPY command, 303
/CPLANE command, 311
/CTYPE command, 323
/CVAL command, 326
/CWD command, 328
/CYCEXPAND command, 328
/DELETE command, 365
/DEVDISP command, 373
/DEVICE command, 375
/DFLAB command, 377
/DIRECTORY command, 382
/DIST command, 383
/DSCALE command, 406
/DV3D command, 415
/EDGE command, 455
/EFACET command, 501
/EOF command, 525
/ERASE command, 532
/ESHAPE command, 541
/EXIT command, 570
/EXPAND command, 573
/FACET command, 584
/FDELE command, 590
1903
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
/FILNAME command, 598
/FOCUS command, 685
/FORMAT command, 688
/GCMD command, 708
/GCOLUMN command, 709
/GFILE command, 767
/GFORMAT command, 767
/GLINE command, 768
/GMARKER command, 769
/GO command, 772
/GOLIST command, 772
/GOPR command, 773
/GRAPHICS command, 776
/GRESUME command, 778
/GRID command, 779
/GROPT command, 779
/GRTYP command, 783
/GSAVE command, 783
/GST command, 788
/GTHK command, 789
/GTYPE command, 790
/HBC command, 793
/HEADER command, 796
/ICLWID command, 827
/ICSCALE command, 827
/IMAGE command, 834
/INPUT command, 841
/INQUIRE command, 843
/LARC command, 889
/LIGHT command, 918
/LINE command, 920
/LSPEC command, 946
/LSYMBOL command, 952
/MAIL command, 965
/MENU command, 971
/MKDIR command, 1006
/MPLIB command, 1036
/MREP command, 1044
/MSTART command, 1062
/NERR command, 1076
/NOERASE command, 1092
/NOLIST command, 1093
/NOPR command, 1095
/NORMAL command, 1097
/NUMBER command, 1124
/OPT command, 1162
/OUTPUT command, 1168
/PAGE command, 1177
/PBC command, 1186
/PBF command, 1189
/PCIRCLE command, 1196
/PCOPY command, 1198
/PDS command, 1230
/PICE command, 1252
/PLOPTS command, 1280
/PMACRO command, 1299
/PMETH command, 1300
/PMORE command, 1305
/PNUM command, 1307
/POLYGON command, 1309
/POST1 command, 1311
/POST26 command, 1311
/PREP7 command, 1322
/PSEARCH command, 1361
/PSF command, 1363
/PSPEC command, 1373
/PSTATUS command, 1376
/PSYMB command, 1377
/PWEDGE command, 1382
/QUIT command, 1388
/RATIO command, 1396
/RENAME command, 1404
/REPLOT command, 1406
/RESET command, 1410
/RGB command, 1419
/RMDIR command, 1427
/RUNST command, 1464
/SECLIB command, 1495
/SEG command, 1513
/SHADE command, 1564
/SHOW command, 1565
/SHOWDISP command, 1568
/SHRINK command, 1572
/SMBC command, 1582
/SOLU command, 1599
/SSCALE command, 1621
/STATUS command, 1634
/STITLE command, 1636
/SYP command, 1656
/SYS command, 1656
/TEE command, 1713
/TITLE command, 1723
/TLABEL command, 1723
/TRIAD command, 1741
/TRLCY command, 1742
/TSPEC command, 1748
/TXTRE command, 1752
/TYPE command, 1755
/UCMD command, 1761
/UDOC command, 1762
/UI command, 1763
/UIS command, 1767
/UNITS command, 1774
/USER command, 1779
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1904
Index
/VCONE command, 1795
/VIEW command, 1820
/VSCALE command, 1845
/VT command, 1854
/VUP command, 1878
/WAIT command, 1881
/WINDOW command, 1884
/XFRM command, 1895
/XRANGE command, 1896
/YRANGE command, 1899
/ZOOM command, 1901
~CAT5IN command, 83
~CATIAIN command, 84
~PARAIN command, 85
~PROEIN command, 86
~SATIN command, 87
~UGIN command, 88
A
A command, 91
AADD command, 91
AATT command, 92
ABBRES command, 94
ABBSAV command, 95
ABEXTRACT command, 93
ABS command, 96
ACCAT command, 97
ACEL command, 97
ACLEAR command, 99
ADAMS command, 99
ADAPT command, 100
ADD command, 102
ADDAM command, 102
ADELE command, 103
ADGL command, 104
ADRAG command, 104
AESIZE command, 105
AFILLT command, 106
AFLIST command, 106
AFSURF command, 107
AGEN command, 108
AGLUE command, 109
AINA command, 110
AINP command, 110
AINV command, 111
AL command, 111
ALIST command, 112
ALLSEL command, 113
ALPFILL command, 114
ALPHAD command, 114
AMAP command, 115
AMESH command, 115
ANCNTR command, 118
ANCUT command, 119
ANCYC command, 120
ANDATA command, 121
ANDSCL command, 122
ANDYNA command, 122
ANFLOW command, 124
ANHARM command, 125
ANIM command, 126
ANISOS command, 127
ANMODE command, 127
ANMRES command, 128
ANORM command, 131
ANSOL command, 132
ANSTOAQWA command, 134
ANSTOASAS command, 137
ANTIME command, 142
ANTYPE command, 143
AOFFST command, 147
AOVLAP command, 148
APLOT command, 148
APPEND command, 149
APTN command, 151
ARCLEN command, 151
ARCOLLAPSE command, 152
ARCTRM command, 153
ARDETACH command, 154
AREAS command, 154
AREFINE command, 155
AREMESH command, 156
AREVERSE command, 157
ARFILL command, 157
ARMERGE command, 158
AROTAT command, 158
ARSCALE command, 159
ARSPLIT command, 160
ARSYM command, 161
ASBA command, 162
ASBL command, 163
ASBV command, 164
ASBW command, 166
ASEL command, 167
ASKIN command, 170
ASLL command, 171
ASLV command, 171
ASUB command, 173
ASUM command, 173
ATAN command, 174
ATRAN command, 175
ATYPE command, 176
AUTOTS command, 177
AVPRIN command, 179
1905
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
AVRES command, 181
B
BCSOPTION command, 184
BELLOW command, 186
BEND command, 186
BETAD command, 187
BF command, 187
BFA command, 191
BFADELE command, 193
BFALIST command, 194
BFCUM command, 195
BFDELE command, 196
BFE command, 198
BFECUM command, 200
BFEDELE command, 202
BFELIST command, 203
BFESCAL command, 203
BFINT command, 205
BFK command, 206
BFKDELE command, 208
BFKLIST command, 209
BFL command, 210
BFLDELE command, 212
BFLIST command, 213
BFLLIST command, 213
BFSCALE command, 214
BFTRAN command, 215
BFUNIF command, 215
BFV command, 217
BFVDELE command, 218
BFVLIST command, 219
BIOOPT command, 220
BIOT command, 220
BLC4 command, 221
BLC5 command, 222
BLOCK command, 222
BOOL command, 223
BOPTN command, 223
BRANCH command, 225
BSAX command, 226
BSM1 command, 227
BSM2 command, 228
BSMD command, 227
BSPLIN command, 229
BSS1 command, 230
BSS2 command, 231
BSTE command, 232
BSTQ command, 232
BTOL command, 233
BUCOPT command, 234
C
C*** command, 235
CALC command, 235
CAMPBELL command, 236
CBDOF command, 237
CDOPT command, 239
CDREAD command, 239
CDWRITE command, 241
CE command, 244
CECHECK command, 246
CECMOD command, 246
CECYC command, 247
CEDELE command, 248
CEINTF command, 248
CELIST command, 249
CENTER command, 250
CEQN command, 251
CERIG command, 251
CESGEN command, 253
CFACT command, 253
CGLOC command, 257
CGOMGA command, 258
CHECK command, 259
CHKMSH command, 260
CINT command, 260
CIRCLE command, 263
CISOL command, 264
CLOCAL command, 266
CLOG command, 267
CLRMSHLN command, 268
CM command, 269
CMACEL command, 270
CMATRIX command, 272
CMDELE command, 273
CMDOMEGA command, 274
CMEDIT command, 275
CMGRP command, 276
CMLIST command, 277
CMMOD command, 278
CMOMEGA command, 278
CMPLOT command, 281
CMROTATE command, 282
CMSEL command, 283
CMSFILE command, 285
CMSOPT command, 286
CMWRITE command, 288
CNCHECK command, 288
CNVTOL command, 292
COMPRESS command, 298
CON4 command, 298
CONE command, 299
CONJUG command, 301
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1906
Index
CORIOLIS command, 304
COUPLE command, 306
COVAL command, 306
CP command, 307
CPCYC command, 309
CPDELE command, 310
CPINTF command, 310
CPLGEN command, 311
CPLIST command, 312
CPMERGE command, 313
CPNGEN command, 314
CPSGEN command, 315
CQC command, 315
CRPLIM command, 317
CS command, 317
CSCIR command, 318
CSDELE command, 319
CSKP command, 320
CSLIST command, 321
CSWPLA command, 321
CSYS command, 322
CURR2D command, 324
CUTCONTROL command, 325
CVAR command, 327
CYCLIC command, 331
CYCOPT command, 333
CYCPHASE command, 337
CYL4 command, 339
CYL5 command, 340
CYLIND command, 341
CZDEL command, 341
CZMESH command, 342
D
D command, 345
DA command, 348
DADELE command, 351
DALIST command, 353
DAMORPH command, 354
DATA command, 355
DATADEF command, 356
DCGOMG command, 356
DCUM command, 357
DCVSWP command, 358
DDELE command, 359
DDOPTION command, 361
DEACT command, 362
DECOMP command, 363
DEFINE command, 363
DELETE command, 365
DELTIM command, 366
DEMORPH command, 367
DERIV command, 368
DESIZE command, 369
DESOL command, 370
DETAB command, 372
DIG command, 378
DIGIT command, 378
DISPLAY command, 382
DJ command, 384
DJDELE command, 385
DJLIST command, 387
DK command, 387
DKDELE command, 389
DKLIST command, 390
DL command, 390
DLDELE command, 393
DLIST command, 395
DLLIST command, 396
DMOVE command, 396
DMPEXT command, 397
DMPRAT command, 398
DNSOL command, 398
DOF command, 402
DOFSEL command, 403
DOMEGA command, 404
DSCALE command, 406
DSET command, 407
DSPOPTION command, 408
DSUM command, 410
DSURF command, 411
DSYM command, 411
DSYS command, 413
DTRAN command, 414
DUMP command, 414
DVMORPH command, 417
DYNOPT command, 418
E
E command, 421
EALIVE command, 421
EDADAPT command, 422
EDALE command, 423
EDASMP command, 424
EDBOUND command, 425
EDBVIS command, 428
EDBX command, 427
EDCADAPT command, 428
EDCGEN command, 430
EDCLIST command, 435
EDCMORE command, 435
EDCNSTR command, 436
EDCONTACT command, 439
EDCPU command, 441
1907
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
EDCRB command, 441
EDCSC command, 442
EDCTS command, 443
EDCURVE command, 444
EDDAMP command, 446
EDDBL command, 448
EDDC command, 449
EDDRELAX command, 450
EDDUMP command, 451
EDELE command, 452
EDENERGY command, 452
EDFPLOT command, 453
EDGCALE command, 454
EDHGLS command, 456
EDHIST command, 456
EDHTIME command, 457
EDINT command, 458
EDIPART command, 459
EDIS command, 461
EDLCS command, 462
EDLOAD command, 463
EDMP command, 468
EDNB command, 471
EDNDTSD command, 472
EDNROT command, 473
EDOPT command, 474
EDOUT command, 475
EDPART command, 477
EDPC command, 479
EDPL command, 479
EDPVEL command, 480
EDRC command, 482
EDRD command, 484
EDREAD command, 485
EDRI command, 487
EDRST command, 488
EDRUN command, 488
EDSHELL command, 489
EDSOLV command, 491
EDSP command, 491
EDSTART command, 492
EDTERM command, 494
EDTP command, 495
EDVEL command, 496
EDWELD command, 499
EDWRITE command, 500
EGEN command, 503
EINTF command, 504
EKILL command, 505
ELBOW command, 505
ELEM command, 507
ELIST command, 507
EMAGERR command, 510
EMATWRITE command, 511
EMF command, 511
EMFT command, 512
EMID command, 512
EMIS command, 513
EMODIF command, 514
EMORE command, 515
EMSYM command, 515
EMTGEN command, 516
EMUNIT command, 517
EN command, 518
ENDRELEASE command, 520
ENERSOL command, 521
ENGEN command, 522
ENORM command, 523
ENSYM command, 524
EORIENT command, 525
EPLOT command, 527
EQSLV command, 527
ERASE command, 532
EREAD command, 533
EREFINE command, 533
EREINF command, 535
ERESX command, 535
ERNORM command, 536
ERRANG command, 537
ESCHECK command, 537
ESEL command, 538
ESIZE command, 543
ESLA command, 544
ESLL command, 544
ESLN command, 545
ESLV command, 546
ESOL command, 547
ESORT command, 550
ESSOLV command, 551
ESTIF command, 553
ESURF command, 554
ESYM command, 556
ESYS command, 557
ET command, 557
ETABLE command, 558
ETCHG command, 565
ETCONTROL command, 567
ETDELE command, 568
ETLIST command, 568
ETYPE command, 568
EUSORT command, 569
EWRITE command, 569
EXP command, 571
EXPAND command, 572
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1908
Index
EXPASS command, 575
EXPROFILE command, 576
EXPSOL command, 577
EXTOPT command, 578
EXTREM command, 580
EXUNIT command, 580
F
F command, 583
FATIGUE command, 585
FC command, 585
FCCHECK command, 587
FCDELE command, 587
FCLIST command, 588
FCUM command, 588
FDELE command, 589
FE command, 591
FEBODY command, 592
FECONS command, 592
FEFOR command, 593
FELIST command, 593
FESURF command, 593
FILE command, 594
FILEAUX2 command, 595
FILEAUX3 command, 595
FILEDISP command, 596
FILL command, 596
FILLDATA command, 597
FINISH command, 598
FITEM command, 599
FJ command, 600
FJDELE command, 601
FJLIST command, 601
FK command, 602
FKDELE command, 603
FKLIST command, 604
FL command, 605
FLANGE command, 605
FLDATA command, 606
FLDATA1 command, 609
FLDATA10 command, 629
FLDATA11 command, 630
FLDATA12 command, 631
FLDATA13 command, 632
FLDATA14 command, 633
FLDATA15 command, 634
FLDATA16 command, 634
FLDATA17 command, 635
FLDATA18 command, 636
FLDATA19 command, 638
FLDATA2 command, 611
FLDATA20 command, 639
FLDATA20A command, 641
FLDATA20B command, 642
FLDATA21 command, 643
FLDATA22 command, 644
FLDATA23 command, 646
FLDATA24 command, 647
FLDATA24A command, 650
FLDATA24B command, 652
FLDATA24C command, 653
FLDATA24D command, 654
FLDATA24E command, 655
FLDATA24F command, 656
FLDATA24G command, 657
FLDATA24H command, 658
FLDATA25 command, 659
FLDATA26 command, 660
FLDATA27 command, 661
FLDATA28 command, 663
FLDATA29 command, 664
FLDATA3 command, 612
FLDATA30 command, 666
FLDATA31 command, 667
FLDATA32 command, 668
FLDATA33 command, 671
FLDATA34 command, 672
FLDATA35 command, 672
FLDATA36 command, 673
FLDATA37 command, 674
FLDATA38 command, 675
FLDATA39 command, 676
FLDATA4 command, 613
FLDATA40 command, 678
FLDATA4A command, 616
FLDATA5 command, 618
FLDATA6 command, 621
FLDATA7 command, 622
FLDATA8 command, 626
FLDATA9 command, 628
FLIST command, 679
FLLIST command, 679
FLOCHECK command, 680
FLOTRAN command, 681
FLREAD command, 681
FLST command, 682
FLUXV command, 683
FMAGBC command, 683
FMAGSUM command, 684
FOR2D command, 686
FORCE command, 686
FORM command, 687
FP command, 689
FPLIST command, 690
1909
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
FREQ command, 690
FRQSCL command, 691
FS command, 692
FSCALE command, 693
FSDELE command, 693
FSLIST command, 694
FSNODE command, 694
FSPLOT command, 695
FSSECT command, 696
FSSPARM command, 697
FSUM command, 697
FTCALC command, 698
FTRAN command, 699
FTSIZE command, 699
FTWRITE command, 700
FVMESH command, 700
G
GAP command, 703
GAPF command, 703
GAPFINISH command, 704
GAPLIST command, 704
GAPMERGE command, 705
GAPOPT command, 706
GAPPLOT command, 706
GAUGE command, 707
GENOPT command, 709
GEOM command, 710
GEOMETRY command, 710
GMATRIX command, 770
GMFACE command, 771
GP command, 773
GPDELE command, 775
GPLIST command, 776
GPLOT command, 776
GRP command, 782
GSBDATA command, 784
GSGDATA command, 785
GSLIST command, 786
GSSOL command, 787
GSUM command, 788
H
HARFRQ command, 793
HBMAT command, 794
HELP command, 797
HELPDISP command, 798
HEMIOPT command, 798
HFADP command, 798
HFANG command, 799
HFARRAY command, 800
HFDEEM command, 800
HFEIGOPT command, 801
HFEREFINE command, 802
HFMODPRT command, 802
HFPA command, 803
HFPCSWP command, 804
HFPORT command, 805
HFPOWER command, 805
HFSCAT command, 811
HFSYM command, 812
HMAGSOLV command, 813
HPGL command, 815
HPTCREATE command, 816
HPTDELETE command, 817
HRCPLX command, 817
HREXP command, 819
HROPT command, 820
HROUT command, 822
I
IC command, 823
ICDELE command, 824
ICE command, 825
ICEDELE command, 825
ICELIST command, 826
ICLIST command, 826
ICVFRC command, 828
IGESIN command, 831
IGESOUT command, 832
IMAGIN command, 834
IMESH command, 835
IMMED command, 836
IMPD command, 837
INISTATE command, 838
INRES command, 845
INRTIA command, 846
INT1 command, 846
INTSRF command, 847
IOPTN command, 848
IRLF command, 850
IRLIST command, 851
J
JPEG command, 853
JSOL command, 854
K
K command, 855
KATT command, 855
KBC command, 856
KBETW command, 857
KCALC command, 858
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1910
Index
KCENTER command, 859
KCLEAR command, 861
KDELE command, 862
KDIST command, 862
KEEP command, 863
KESIZE command, 864
KEYOPT command, 865
KEYPTS command, 865
KEYW command, 866
KFILL command, 866
KGEN command, 867
KL command, 868
KLIST command, 868
KMESH command, 869
KMODIF command, 870
KMOVE command, 870
KNODE command, 871
KPLOT command, 872
KPSCALE command, 872
KREFINE command, 873
KSCALE command, 874
KSCON command, 875
KSEL command, 876
KSLL command, 878
KSLN command, 879
KSUM command, 879
KSYMM command, 880
KTRAN command, 881
KUSE command, 882
KWPAVE command, 883
KWPLAN command, 883
L
L command, 885
L2ANG command, 886
L2TAN command, 886
LANG command, 887
LARC command, 888
LAREA command, 889
LARGE command, 890
LATT command, 891
LAYER command, 892
LAYERP26 command, 893
LAYLIST command, 894
LAYPLOT command, 894
LCABS command, 895
LCASE command, 896
LCCALC command, 896
LCCAT command, 897
LCDEF command, 898
LCFACT command, 898
LCFILE command, 899
LCLEAR command, 900
LCOMB command, 900
LCOPER command, 901
LCSEL command, 902
LCSL command, 904
LCSUM command, 904
LCWRITE command, 905
LCZERO command, 905
LDELE command, 906
LDIV command, 906
LDRAG command, 907
LDREAD command, 908
LESIZE command, 912
LEXTND command, 914
LFILLT command, 914
LFSURF command, 915
LGEN command, 916
LGLUE command, 917
LGWRITE command, 917
LINA command, 919
LINE command, 920
LINES command, 921
LINL command, 921
LINP command, 922
LINV command, 922
LIST command, 923
LLIST command, 924
LMATRIX command, 925
LMESH command, 926
LNCOLLAPSE command, 927
LNDETACH command, 927
LNFILL command, 928
LNMERGE command, 928
LNSPLIT command, 929
LNSRCH command, 929
LOCAL command, 930
LOVLAP command, 931
LPLOT command, 932
LPRT command, 932
LPTN command, 933
LREFINE command, 933
LREVERSE command, 935
LROTAT command, 935
LSBA command, 936
LSBL command, 937
LSBV command, 938
LSBW command, 940
LSCLEAR command, 941
LSDELE command, 942
LSEL command, 942
LSLA command, 945
LSLK command, 945
1911
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
LSOPER command, 946
LSREAD command, 948
LSSCALE command, 948
LSSOLVE command, 949
LSTR command, 950
LSUM command, 951
LSWRITE command, 951
LSYMM command, 953
LTAN command, 954
LTRAN command, 955
LUMPM command, 956
LVSCALE command, 956
LWPLAN command, 957
M
M command, 959
MADAPT command, 960
MAGOPT command, 962
MAGSOLV command, 963
MAPSOLVE command, 966
MASTER command, 966
MAT command, 967
MATER command, 967
MCHECK command, 967
MDAMP command, 968
MDELE command, 969
MDPLOT command, 970
MEMM command, 970
MESHING command, 972
MFANALYSIS command, 972
MFBUCKET command, 973
MFCALC command, 974
MFCI command, 975
MFCLEAR command, 975
MFCMMAND command, 976
MFCONV command, 977
MFDTIME command, 978
MFELEM command, 979
MFEM command, 980
MFEXTER command, 980
MFFNAME command, 981
MFFR command, 982
MFIMPORT command, 983
MFINTER command, 984
MFITER command, 985
MFLC command, 985
MFLIST command, 988
MFMAP command, 989
MFORDER command, 990
MFOUTPUT command, 990
MFPS command, 993
MFRC command, 994
MFRELAX command, 995
MFRSTART command, 996
MFSO command, 997
MFSURFACE command, 998
MFTIME command, 999
MFTOL command, 999
MFVOLUME command, 1002
MFWR command, 1003
MGEN command, 1003
MIDTOL command, 1004
MITER command, 1005
MLIST command, 1006
MMASS command, 1007
MMF command, 1008
MODCONT command, 1008
MODE command, 1009
MODIFY command, 1010
MODMSH command, 1011
MODOPT command, 1012
MONITOR command, 1015
MOPT command, 1020
MORPH command, 1023
MOVE command, 1025
MP command, 1026
MPAMOD command, 1030
MPCHG command, 1031
MPCOPY command, 1032
MPDATA command, 1032
MPDELE command, 1035
MPDRES command, 1036
MPLIST command, 1037
MPPLOT command, 1038
MPREAD command, 1038
MPRINT command, 1040
MPTEMP command, 1040
MPTGEN command, 1041
MPTRES command, 1042
MPWRITE command, 1042
MSADV command, 1044
MSAVE command, 1045
MSCAP command, 1046
MSDATA command, 1047
MSHAPE command, 1050
MSHCOPY command, 1051
MSHKEY command, 1052
MSHMID command, 1052
MSHPATTERN command, 1053
MSMASS command, 1054
MSMETH command, 1055
MSMIR command, 1056
MSNOMF command, 1057
MSPROP command, 1057
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1912
Index
MSQUAD command, 1059
MSRELAX command, 1060
MSSOLU command, 1060
MSSPEC command, 1061
MSTERM command, 1063
MSVARY command, 1064
MXPAND command, 1066
N
N command, 1069
NANG command, 1069
NAXIS command, 1070
NCNV command, 1071
NDELE command, 1072
NDIST command, 1073
NDSURF command, 1074
NEQIT command, 1075
NFORCE command, 1077
NGEN command, 1078
NKPT command, 1079
NLDIAG command, 1079
NLDPOST command, 1082
NLGEOM command, 1083
NLHIST command, 1084
NLIST command, 1088
NLOG command, 1089
NLOPT command, 1090
NMODIF command, 1090
NOCOLOR command, 1091
NODES command, 1092
NOOFFSET command, 1093
NOORDER command, 1094
NORA command, 1096
NORL command, 1096
NPLOT command, 1098
NPRINT command, 1098
NREAD command, 1099
NREFINE command, 1099
NRLSUM command, 1101
NROPT command, 1102
NROTAT command, 1103
NRRANG command, 1104
NSCALE command, 1104
NSEL command, 1105
NSLA command, 1110
NSLE command, 1111
NSLK command, 1112
NSLL command, 1113
NSLV command, 1114
NSMOOTH command, 1114
NSOL command, 1115
NSORT command, 1117
NSTORE command, 1121
NSUBST command, 1121
NSVR command, 1122
NSYM command, 1123
NUMCMP command, 1125
NUMEXP command, 1126
NUMMRG command, 1127
NUMOFF command, 1131
NUMSTR command, 1132
NUMVAR command, 1133
NUSORT command, 1134
NWPAVE command, 1134
NWPLAN command, 1135
NWRITE command, 1135
O
OCDATA command, 1137
OCDELETE command, 1139
OCLIST command, 1140
OCTABLE command, 1141
OCTYPE command, 1142
OMEGA command, 1144
OPADD command, 1145
OPANL command, 1146
OPCLR command, 1146
OPDATA command, 1147
OPDEL command, 1147
OPEQN command, 1148
OPERATE command, 1149
OPEXE command, 1150
OPFACT command, 1150
OPFRST command, 1151
OPGRAD command, 1152
OPKEEP command, 1152
OPLFA command, 1153
OPLGR command, 1154
OPLIST command, 1154
OPLOOP command, 1155
OPLSW command, 1156
OPMAKE command, 1156
OPNCONTROL command, 1157
OPPRNT command, 1157
OPRAND command, 1158
OPRESU command, 1158
OPRFA command, 1159
OPRGR command, 1159
OPRSW command, 1160
OPSAVE command, 1160
OPSEL command, 1161
OPSUBP command, 1161
OPSWEEP command, 1162
OPTYPE command, 1163
1913
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
OPUSER command, 1164
OPVAR command, 1165
OUTOPT command, 1166
OUTPR command, 1166
OUTRES command, 1169
P
PADELE command, 1177
PAGET command, 1178
PAPUT command, 1179
PARESU command, 1179
PARRES command, 1182
PARSAV command, 1182
PARTSEL command, 1180
PASAVE command, 1183
PATH command, 1184
PAUSE command, 1185
PCALC command, 1190
PCGOPT command, 1193
PCIRC command, 1195
PCONV command, 1196
PCORRO command, 1199
PCROSS command, 1199
PDANL command, 1200
PDCDF command, 1201
PDCFLD command, 1202
PDCLR command, 1204
PDCMAT command, 1205
PDCORR command, 1206
PDDMCS command, 1207
PDDOEL command, 1209
PDEF command, 1211
PDEXE command, 1214
PDHIST command, 1216
PDINQR command, 1217
PDLHS command, 1218
PDMETH command, 1220
PDOT command, 1222
PDPINV command, 1223
PDPLOT command, 1224
PDPROB command, 1225
PDRAG command, 1226
PDRESU command, 1227
PDROPT command, 1228
PDSAVE command, 1231
PDSCAT command, 1231
PDSENS command, 1233
PDSHIS command, 1234
PDUSER command, 1235
PDVAR command, 1236
PDWRITE command, 1238
PEMOPTS command, 1239
PERBC2D command, 1239
PERI command, 1241
PEXCLUDE command, 1241
PFACT command, 1242
PFLUID command, 1243
PGAP command, 1243
PGRAPH command, 1244
PGRSET command, 1245
PGSAVE command, 1246
PGSELE command, 1247
PGWRITE command, 1248
PHYSICS command, 1250
PINCLUDE command, 1252
PINSUL command, 1253
PIPE command, 1254
PIVCHECK command, 1254
PLCAMP command, 1255
PLCINT command, 1257
PLCONV command, 1258
PLCPLX command, 1260
PLCRACK command, 1260
PLDISP command, 1261
PLESOL command, 1262
PLETAB command, 1267
PLF2D command, 1272
PLFAR command, 1268
PLFSS command, 1271
PLLS command, 1272
PLNEAR command, 1273
PLNSOL command, 1275
PLORB command, 1282
PLOT command, 1283
PLOTTING command, 1283
PLPAGM command, 1284
PLPATH command, 1284
PLSCH command, 1285
PLSECT command, 1286
PLSYZ command, 1287
PLTD command, 1288
PLTIME command, 1290
PLTLINE command, 1290
PLTRAC command, 1291
PLVAR command, 1294
PLVAROPT command, 1294
PLVECT command, 1295
PLVFRC command, 1298
PLWAVE command, 1298
PMAP command, 1299
PMETH command, 1300
PMGTRAN command, 1301
PMLOPT command, 1302
PMLSIZE command, 1303
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1914
Index
PMOPTS command, 1304
PNGR command, 1305
POINT command, 1309
POLY command, 1309
POPT command, 1310
POUTRES command, 1312
POWERH command, 1313
PPATH command, 1314
PPLOT command, 1315
PPRANGE command, 1315
PPRES command, 1316
PRANGE command, 1316
PRCAMP command, 1317
PRCINT command, 1318
PRCONV command, 1319
PRCPLX command, 1320
PRECISION command, 1320
PRED command, 1321
PRENERGY command, 1322
PRERR command, 1323
PRESOL command, 1324
PRETAB command, 1328
PRFAR command, 1329
PRI2 command, 1331
PRIM command, 1332
PRINT command, 1332
PRISM command, 1333
PRITER command, 1333
PRJSOL command, 1333
PRNEAR command, 1334
PRNLD command, 1336
PRNSOL command, 1337
PROD command, 1342
PRORB command, 1343
PRPATH command, 1343
PRRFOR command, 1344
PRRSOL command, 1345
PRSECT command, 1345
PRSYZ command, 1347
PRTIME command, 1348
PRVAR command, 1348
PRVAROPT command, 1349
PRVECT command, 1349
PSCONTROL command, 1351
PSCR command, 1352
PSDCOM command, 1355
PSDFRQ command, 1356
PSDGRAPH command, 1357
PSDRES command, 1357
PSDSPL command, 1358
PSDUNIT command, 1359
PSDVAL command, 1360
PSDWAV command, 1361
PSEL command, 1362
PSMAT command, 1365
PSMESH command, 1367
PSOLVE command, 1371
PSPEC command, 1373
PSPRNG command, 1375
PSTRES command, 1376
PTEMP command, 1379
PTR command, 1380
PTXY command, 1380
PUNIT command, 1381
PVECT command, 1382
Q
QDVAL command, 1385
QFACT command, 1386
QSOPT command, 1386
QUAD command, 1387
QUOT command, 1388
R
R command, 1391
RACE command, 1392
RADOPT command, 1393
RALL command, 1394
RAPPND command, 1394
RATE command, 1395
RBE3 command, 1397
RCON command, 1398
RDEC command, 1398
RDELE command, 1399
REAL command, 1400
REALVAR command, 1401
RECTNG command, 1401
REDUCE command, 1402
reinforcing, 535
REMESH command, 1402
REORDER command, 1405
RESCONTROL command, 1408
RESET command, 1410
RESP command, 1411
RESUME command, 1413
RESVEC command, 1414
RESWRITE command, 1415
REXPORT command, 1416
REZONE command, 1417
RFILSZ command, 1418
RFORCE command, 1418
RIGID command, 1420
RIGRESP command, 1421
RIMPORT command, 1422
1915
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
RITER command, 1423
RLIST command, 1424
RMALIST command, 1424
RMANL command, 1425
RMASTER command, 1426
RMCAP command, 1426
RMCLIST command, 1427
RMEMRY command, 1428
RMFLVEC command, 1428
RMLVSCALE command, 1429
RMMLIST command, 1429
RMMRANGE command, 1430
RMMSELECT command, 1431
RMNDISP command, 1432
RMNEVEC command, 1433
RMODIF command, 1433
RMORE command, 1434
RMPORDER command, 1434
RMRESUME command, 1435
RMRGENERATE command, 1435
RMROPTIONS command, 1436
RMRPLOT command, 1437
RMRSTATUS command, 1438
RMSAVE command, 1438
RMSMPLE command, 1438
RMUSE command, 1440
RMXPORT command, 1440
ROCK command, 1441
ROSE command, 1441
RPOLY command, 1442
RPR4 command, 1443
RPRISM command, 1444
RPSD command, 1444
RSFIT command, 1445
RSOPT command, 1448
RSPEED command, 1448
RSPLIT command, 1449
RSPLOT command, 1450
RSPRNT command, 1451
RSSIMS command, 1452
RSTAT command, 1453
RSTMAC command, 1454
RSTOFF command, 1455
RSURF command, 1456
RSYMM command, 1458
RSYS command, 1459
RTHICK command, 1462
RTIMST command, 1462
RUN command, 1463
RWFRNT command, 1464
S
SABS command, 1465
SADD command, 1465
SALLOW command, 1466
SARPLOT command, 1467
SAVE command, 1467
SBCLIST command, 1468
SBCTRAN command, 1469
SDELETE command, 1469
SE command, 1470
SECCONTROLS command, 1471
SECDATA command, 1473
SECFUNCTION command, 1493
SECJOINT command, 1494
SECLOCK command, 1495
SECMODIF command, 1496
SECNUM command, 1497
SECOFFSET command, 1497
SECPLOT command, 1499
SECREAD command, 1503
SECSTOP command, 1506
SECTYPE command, 1507
SECWRITE command, 1510
SED command, 1511
SEDLIST command, 1512
SEEXP command, 1512
SEGEN command, 1515
SELIST command, 1516
SELM command, 1517
SELTOL command, 1518
SENERGY command, 1518
SEOPT command, 1519
SESYMM command, 1520
SET command, 1526
SETFGAP command, 1529
SETRAN command, 1530
SEXP command, 1531
SF command, 1532
SFA command, 1536
SFACT command, 1539
SFADELE command, 1540
SFALIST command, 1541
SFBEAM command, 1541
SFCALC command, 1543
SFCUM command, 1543
SFDELE command, 1545
SFE command, 1546
SFEDELE command, 1551
SFELIST command, 1552
SFFUN command, 1553
SFGRAD command, 1555
SFL command, 1556
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1916
Index
SFLDELE command, 1558
SFLEX command, 1559
SFLIST command, 1560
SFLLIST command, 1562
SFSCALE command, 1562
SFTRAN command, 1563
SHELL command, 1564
SHPP command, 1568
SHSD command, 1572
SLIST command, 1575
SLOAD command, 1577
SLPPLOT command, 1580
SLSPLOT command, 1580
SMALL command, 1581
SMAX command, 1581
SMBODY command, 1583
SMCONS command, 1583
SMFOR command, 1584
SMIN command, 1584
SMOOTH command, 1585
SMRTSIZE command, 1586
SMSURF command, 1589
SMULT command, 1589
SNOPTION command, 1590
SOCEAN command, 1592
SOLCONTROL command, 1592
SOLU command, 1598
SOLUOPT command, 1600
SOLVE command, 1600
SORT command, 1601
SOURCE command, 1601
SPACE command, 1602
SPADP command, 1602
SPARM command, 1603
SPCNOD command, 1604
SPCTEMP command, 1604
SPEC command, 1605
SPFSS command, 1606
SPH4 command, 1607
SPH5 command, 1608
SPHERE command, 1609
SPICE command, 1609
SPLINE command, 1610
SPLOT command, 1611
SPOINT command, 1611
SPOPT command, 1612
SPREAD command, 1613
SPSCAN command, 1613
SPSWP command, 1615
SPTOPT command, 1617
SQRT command, 1618
SRSS command, 1619
SSBT command, 1620
SSLN command, 1622
SSMT command, 1622
SSPA command, 1623
SSPB command, 1624
SSPD command, 1625
SSPE command, 1625
SSPM command, 1626
SSTIF command, 1627
SSUM command, 1628
STABILIZE command, 1628
STAOPT command, 1630
STAT command, 1630
STEF command, 1635
STORE command, 1637
SUBSET command, 1638
SUCALC command, 1639
SUCR command, 1641
SUDEL command, 1642
SUEVAL command, 1643
SUGET command, 1643
SUMAP command, 1644
SUMTYPE command, 1645
SUPL command, 1645
SUPR command, 1646
SURESU command, 1646
SUSAVE command, 1647
SUSEL command, 1648
SUVECT command, 1649
SV command, 1650
SVPLOT command, 1650
SVTYP command, 1651
SWADD command, 1652
SWDEL command, 1653
SWGEN command, 1653
SWLIST command, 1654
SYNCHRO command, 1655
T
TALLOW command, 1657
TB command, 1658
TBCOPY command, 1694
TBDATA command, 1695
TBDELE command, 1695
TBFIELD command, 1696
TBFT command, 1697
TBLE command, 1707
TBLIST command, 1708
TBMODIF command, 1708
TBPLOT command, 1709
TBPT command, 1710
TBTEMP command, 1711
1917
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
TCHG command, 1711
TEE command, 1714
TERM command, 1715
THOPT command, 1716
TIFF command, 1717
TIME command, 1718
TIMERANGE command, 1718
TIMINT command, 1719
TIMP command, 1720
TINTP command, 1721
TOCOMP command, 1724
TODEF command, 1725
TOEXE command, 1725
TOFFST command, 1725
TOFREQ command, 1726
TOGRAPH command, 1727
TOLIST command, 1728
TOLOOP command, 1728
TOPLOT command, 1730
TOPRINT command, 1730
TORQ2D command, 1731
TORQC2D command, 1731
TORQSUM command, 1732
TORUS command, 1733
TOSTAT command, 1734
TOTAL command, 1734
TOTYPE command, 1736
TOVAR command, 1736
TRANS command, 1738
TRANSFER command, 1738
TREF command, 1740
TRNOPT command, 1743
TRPDEL command, 1745
TRPLIS command, 1745
TRPOIN command, 1746
TRTIME command, 1746
TSHAP command, 1747
TSRES command, 1750
TUNIF command, 1751
TVAR command, 1752
TYPE command, 1757
TZAMESH command, 1758
TZDELE command, 1758
TZEGEN command, 1759
U
UIMP command, 1767
UNDELETE command, 1772
UNDO command, 1773
UNPAUSE command, 1775
UPCOORD command, 1776
UPGEOM command, 1777
USRCAL command, 1780
USRDOF command, 1781
USRELEM command, 1782
V
V command, 1787
V2DOPT command, 1788
VA command, 1789
VADD command, 1790
VALVE command, 1791
VARDEL command, 1792
VARNAM command, 1792
VATT command, 1793
VCLEAR command, 1793
VCROSS command, 1796
VCVFILL command, 1797
VDDAM command, 1798
VDELE command, 1798
VDGL command, 1799
VDOT command, 1799
VDRAG command, 1800
VEORIENT command, 1802
VEXT command, 1803
VFOPT command, 1806
VFQUERY command, 1807
VFSM command, 1808
VGEN command, 1812
VGET command, 1819
VGLUE command, 1820
VIMP command, 1821
VINP command, 1822
VINV command, 1823
VLIST command, 1825
VLSCALE command, 1826
VMESH command, 1828
VOFFST command, 1828
VOLUMES command, 1829
VOVLAP command, 1832
VPLOT command, 1834
VPTN command, 1834
VPUT command, 1838
VROTAT command, 1841
VSBA command, 1842
VSBV command, 1843
VSBW command, 1844
VSEL command, 1847
VSLA command, 1849
VSUM command, 1850
VSWEEP command, 1851
VSYMM command, 1852
VTCLR command, 1854
VTDISC command, 1855
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1918
Index
VTEVAL, 1855
VTFREQ command, 1856
VTGEOM, 1857
VTIN command, 1858
VTMETH command, 1860
VTMP command, 1861
VTOP command, 1863
VTPOST command, 1864
VTRAN command, 1865
VTREAL command, 1866
VTRFIL command, 1868
VTRSLT command, 1868
VTSEC, 1871
VTSFE command, 1873
VTSL command, 1875
VTSTAT command, 1875
VTTEMP, 1876
VTVMOD command, 1877
VTYPE command, 1878
W
WAVES command, 1881
WERASE command, 1883
WFRONT command, 1883
WMID command, 1885
WMORE command, 1885
WPAVE command, 1886
WPCSYS command, 1886
WPLANE command, 1887
WPOFFS command, 1888
WPROTA command, 1888
WPSTYL command, 1889
WRFULL command, 1890
WRITE command, 1891
WSORT command, 1892
WSPRINGS command, 1893
WSTART command, 1893
X
XVAR command, 1897
XVAROPT command, 1898
1919
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
Index
Release 12.0 - 2009 SAS IP, Inc.All rights reserved.- Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.
1920