Finite Element Method

MSM 1333

Chapter 3.2
spring & truss

Prepared by:

Dr. Yeak Su Hoe

C22-432, U.T.M. Skudai
s.h.yeak@utm.my
012-7116604

Feb 2012
1
 Spring theory

F1x F2x F3x

x

k1 k2

Problem
Analyze the behavior of the system composed of the two springs loaded
by external forces as shown above

Given
F1x , F2x ,F3x are external loads. Positive directions of the forces are
along the positive x-axis
k1 and k2 are the stiffnesses of the two springs

2
 Spring theory
F1x F2x F3x
x

k1 k2
Solution
Step 1: In order to analyze the system we break it up into smaller parts,
i.e., “elements” connected to each other through “nodes”
F1x k1 F2x k2 F3x
x
1 2 3
Element 1 Element 2
d1x d2x d3x
Node 1
Unknowns: nodal displacements d1x, d2x, d3x,

3
 Spring theory
F1x k1 F2x k2 F3x
x
1 2 3
Element 1 Element 2
d1x d2x d3x
Node 1
Solution
Step 2: Analyze the behavior of a single element (spring)

© 2002 Brooks/Cole Publishing / Thomson Learning™

Two nodes: 1, 2
Nodal displacements: d̂ 1x d̂ 2x
Nodal forces: f̂ 1x f̂ 2x
Spring constant: k
4
 Spring theory

© 2002 Brooks/Cole Publishing / Thomson Learning™

Local (x̂ , ŷ , ẑ) and global (x,y,z) coordinate systems

5
 Spring theory
Behavior of a linear spring (recap)

F x
k k
1 F
d k
d

F = Force in the spring

d = deflection of the spring
k = “stiffness” of the spring
Hooke’s Law
F = kd
6
 Spring theory

f̂1x f̂ 2x
© 2002 Brooks/Cole Publishing / Thomson Learning™

Hooke’s law for our spring element

f̂ 2x = k (d̂ 2x − d̂ 1x ) Eq (1)
Force equilibrium for our spring element (recap free body diagrams)
f̂1x + f̂ 2x = 0
Eq (2)
⇒ f̂1x = − f̂ 2x = − k (d̂ 2x − d̂ 1x )
Collect Eq (1) and (2) in matrix form
 f̂1x   k - k   d̂ 1x 
f̂ = k̂ d̂  = - k k   
 f̂ 2x  142 4  d̂ 2x 
Element force Element nodal 123 3 12 3
Element displacement f̂ k̂ d̂
vector
stiffness vector
matrix 7
 Spring theory
Note T
1. The element stiffness matrix is “symmetric”, i.e. k̂ = k̂
2. The element stiffness matrix is singular, i.e.,

2 2
det ( k̂ ) = k −k = 0
The consequence is that the matrix is NOT invertible. It is not
possible to invert it to obtain the displacements. Why?
The spring is not constrained in space and hence it can attain multiple
positions in space for the same nodal forces e.g.,
 f̂1x  2 - 2  1  - 2 
 = - 2   = 
 f̂ 2x   2  2  2 
 f̂1x  2 - 2   3  - 2 
 = - 2   = 
 f̂ 2x   2  4  2 
8
 Spring theory
Solution
Step 3: Now that we have been able to describe the behavior of each
spring element, lets try to obtain the behavior of the original structure
by assembly

Split the original structure into component elements

Element 2
Element 1
1 k1 2 2 k2 3

(1) (2)
f̂ 1x(1) d̂ 1x
(1) f̂ 2x(1) d̂ 2x f̂ (2)
d̂ (2) f̂ 2x(2) d̂ 2x
1x 1x

 f̂1x(1)   k 1 - k 1   d̂ 1x
(1) 
  f̂1x(2)   k 2 - k 2   d̂ 1x
(2) 

 (1)  = - k   (1)   (2)  = - k   (2) 
 f̂ 2x  142 1 k 1   d̂ 2x   f̂ 2x  1 42 k 2   d̂ 2x 
123 43 123 123 2
43 123
(1 ) (1 ) (1 )
k̂ (2) (2) (2)
f̂ d̂ f̂ k̂ d̂
Eq (3) Eq (4)
9
 Spring theory
To assemble these two results into a single description of the response of
the entire structure we need to link between the local and global variables.

Question 1: How do we relate the local (element) displacements back to

the global (structure) displacements?

F1x k1 F2x k2 F3x

x
1 2 3
Element 1 Element 2
d1x d2x d3x
Node 1
(1)
d̂ 1x = d 1x
(1) (2)
Eq (5)
d̂ 2x = d̂ 1x = d 2x
d̂ (2)
2x = d 3x

10
 Spring theory
Hence, equations (3) and (4) may be rewritten as

 f̂1x(1)   k1 - k 1   d 1x   f̂1x(2)   k2 - k 2   d 2x 
 (1)  = - k     (2)  = - k   
 f̂ 2x   1 k 1   d 2x   f̂ 2x   2 k 2   d 3x 
Or, we may expand the matrices and vectors to obtain
 f̂1x(1)   k 1 − k 1 0   d 1x   0  0 0 0   d 1x 
 ˆ (1)  - k  d   f̂ (2)  0 k  d 
 f̂ 2x  =  1x  = − k 2   2x 
 1 k 1 0   2x   f̂ (2) 
 2
0   0 0 0  1d2 
3x  12 2x   0 - k 2 k 2  1d2 
3x 
123 1 442 4 43 3 3 1 442 4 43 3
(1 ) e d (2)e (2)e d
(1 ) e
k̂ f̂ k̂
f̂
Eq (6) Eq (7)
(1) e
k̂ Expanded element stiffness matrix of element 1 (local)
(1) e
f̂ Expanded nodal force vector for element 1 (local)
d Nodal load vector for the entire structure (global)
11
Question 2: How do we relate the local (element) nodal forces back to
the global (structure) forces? Draw 5 FBDs
F1x k1 F2x k2 F3x
x
1 A B 2 C D3
d1x d2x d3x
2 3

F1x f̂1x(1) f̂ 2x(1) f̂1x(2) f̂ 2x(2) F3x

F2x
© 2002 Brooks/Cole Publishing / Thomson Learning™

At node 1 : F1x - f̂1x(1) = 0

At node 2 : F2x - f̂ 2x(1) − f̂1x(2) = 0
At node 3 : F3x - f̂ 2x(2) = 0

12
 Spring theory
In vector form, the nodal force vector (global)
 F1x   f̂1x 
(1)

   (1) (2) 
F =  F2x  =  f̂ 2x + f̂1x 
 F   f̂ (2) 
 3x   2x 
Recall that the expanded element force vectors were
 f̂1x(1)   0 
(1) e  (1)  (2)e  f̂ (2) 
f̂ =  f̂ 2x  and f̂ =  1x 
0   f̂ (2) 
   2x 
Hence, the global force vector is simply the sum of the expand element
nodal force vectors F  1x
  (1) e (2)e
F =  F2x  = f̂ + f̂
F 
 3x 
13
 Spring theory
But we know the expressions for the expanded local force vectors from
Eqs (6) and (7)
(1) e (1)e (2)e (2)e
f̂ = k̂ d and f̂ = k̂ d
Hence

d =  k̂ + k̂  d

(1) e (2)e (1)e (2)e (1)e (2)e
F = f̂ + f̂ = k̂ d + k̂
 
F=Kd

F = Global nodal force vector

d = Global nodal displaceme nt vector
K = Global stiffness matrix
= sum of expanded element stiffness matrices
14
 Spring theory
For our original structure with two springs, the global stiffness matrix is

 k 1 − k 1 0  0 0 0 
K =  - k 1 k 1 0  +  0 k 2 − k 2 
 0 0 0   0 - k 2 k 2 
1 442 4 43 1 442 4 43
(1 ) e (2)e
k̂ k̂

 k1 − k1 0 
=  - k 1 k1 + k 2 − k 2 
 0 - k2 k 2 

NOTE
1. The global stiffness matrix is symmetric
2. The global stiffness matrix is singular
15
 Spring theory
The system equations F = K d imply

 F1x   k1 − k1 0   d 1x 
  - k  
 F2x  =  1 k1 + k 2 − k 2  d 2x 
F   0 - k2 k 2  d 
 3x   3x 

F1x = k 1d 1x − k 1d 2x
⇒ F2x = -k 1d 1x + ( k 1 + k 2 ) d 2x − k 2 d 3x
F3x = -k 2 d 2x + k 2 d 3x

These are the 3 equilibrium equations at the 3 nodes.

16
 Spring theory
F1x k1 F2x k2 F3x
x
1 A B 2 C D3
d1x d2x d3x
2 3

F1x f̂1x(1) f̂ 2x(1) f̂1x(2) f̂ 2x(2) F3x

F2x
© 2002 Brooks/Cole Publishing / Thomson Learning™

F1x = k 1 (d 1x − d 2x ) = f̂1x(1)
At node 1 : F1x - f̂1x(1) = 0
(1) (2)
F2x = -k 1d 1x + ( k 1 + k 2 ) d 2x − k 2 d 3x
At node 2 : F2x - f̂ − f̂ =0
2x 1x
= − k 1 (d 1x − d 2x ) + k 2 (d 2x − d 3x )
At node 3 : F3x - f̂ 2x(2) = 0
= f̂ 2x(1) + f̂1x(2)
F3x = -k 2 (d 2x − d 3x ) = f̂ 2x(2)

17
 Spring theory

Notice that the sum of the forces equal zero, i.e., the structure is in static
equilibrium.

F1x + F2x+ F3x =0

Given the nodal forces, can we solve for the displacements?

To obtain unique values of the displacements, at least one of the nodal

displacements must be specified.

18
 Spring theory
Direct assembly of the global stiffness matrix

Global
F1x k1 F2x k2 F3x
x
1 2 3
Element 1 Element 2
d1x d2x d3x

Local
Element 2
Element 1
1 k1 2 2 k2 3

(1) (2)
f̂ 1x(1) d̂ 1x
(1) f̂ 2x(1) d̂ 2x f̂ (2)
d̂ (2) f̂ 2x(2) d̂ 2x
1x 1x

19
 Spring theory

Node element connectivity chart : Specifies the global node number

corresponding to the local (element) node numbers

ELEMENT Node 1 Node 2 Local node number

1 1 2 Global node number

2 2 3

20
 Spring theory
Stiffness matrix of element 1 Stiffness matrix of element 2

d1x d2x d2x d3x

(1)  k1 - k 1  d1x (2)  k2 - k 2  d2x
k̂ = k̂ = 
- k 1 k 1  d2x - k 2 k 2  d3x

Global stiffness matrix

d1x d2x d3x
 k1 - k1 0  d1x

K = - k 1 k1 + k 2 
- k 2  d2x
 0 - k2 k 2  d3x
Examples: Problems 2.1 and 2.3 of Logan
21
Trusses: Engineering structures that are composed only of
two-force members. e.g., bridges, roof supports
Actual trusses: Airy structures composed of slender members
(I-beams, channels, angles, bars etc) joined together at their
ends by welding, riveted connections or large bolts and pins

A typical truss structure

Gusset plate

22
Ideal trusses:
Assumptions
• Ideal truss members are connected only at their ends.
• Ideal truss members are connected by frictionless pins (no
moments)
• The truss structure is loaded only at the pins
• Weights of the members are neglected

A typical truss structure

Frictionless pin 23
 Trusses

These assumptions allow us to idealize each truss member

as a two-force member (members loaded only at their
extremities by equal opposite and collinear forces)

member in
compression

member in
tension

Connecting pin
24
 FEM analysis scheme
Step 1: Divide the truss into bar/truss elements connected to each
other through special points (“nodes”)

Step 2: Describe the behavior of each bar element (i.e. derive its
stiffness matrix and load vector in local AND global coordinate
system)

Step 3: Describe the behavior of the entire truss by putting together the
behavior of each of the bar elements (by assembling their stiffness
matrices and load vectors)

Step 4: Apply appropriate boundary conditions and solve

25
 Trusses

Stiffness matrix of bar element

E, A

© 2002 Brooks/Cole Publishing / Thomson Learning™

L: Length of bar
A: Cross sectional area of bar
E: Elastic (Young’s) modulus of bar
û (x̂ ) :displacement of bar as a function of local coordinate bar x̂
The strain in the bar at x̂ d û
ε( x̂ ) =
d x̂
The stress in the bar (Hooke’s law)
σ (x̂ ) = E ε( x̂ )
26
 Trusses
d̂ 2x
Tension in the bar
T( x̂ ) = EAε  x̂  x̂
x̂ û (x̂ ) =  1 −  d̂ 1x + d̂ 2x
x̂ d̂ 1x  L L
L
Assume that the displacement û (x̂is) varying linearly along the bar
 x̂  x̂
û (x̂ ) =  1 −  d̂ 1x + d̂ 2x
 L L
d û d̂ 2x − d̂ 1x
Then, strain is constant along the bar: ε= =
d x̂ L
Stress is also constant along the bar:
E
(
σ = Eε = d̂ 2x − d̂ 1x
L
)
Tension is constant along the bar: T = EAε =
EA
L
(
d̂ 2x − d̂ 1x )
{
k

The bar is acting like a spring with stiffness EA

k=
L
27
 Trusses
Springs theory
E, A

© 2002 Brooks/Cole Publishing / Thomson Learning™ Element /local nodal

d̂ = displacement vector
Two nodes: 1, 2
Nodal displacements: d d̂ 1x d
d̂ 2x
Nodal forces: f̂ 1x f̂ 2x
Spring constant: k = EA
L
Element stiffness matrix in local coordinates
 f̂1x   k - k   d̂ 1x 
f̂ = k̂ d̂  = - k k   
 f̂ 2x  142 4  d̂ 2x 
Element force Element nodal 123 3 12 3
Element displacement f̂ k̂ d̂
vector
stiffness vector
matrix 28
 Trusses
2 bars cases
E1, A1
E2, A2

L2
L1

This is equivalent to the following system of springs

E 1A 1 E 2A 2
k1 = k2 =
L1 L2
x
Element 1 2 Element 2 3
1
d1x d2x d3x

PROBLEM
29
Problem 1: Find the stresses in the two-bar assembly loaded as shown
below
E, 2A
E, A
1 P
2 3
L
L
Solution: This is equivalent to the following system of springs
2EA EA
k1 = k2 =
L L x
Element 1 2 Element 2 3
1
d1x d2x d3x

We will first compute the displacement at node 2 and then the stresses
within each element

30
 Trusses
The global set of equations can be generated using the technique
developed in the springs theory
 k1 − k1 0   d 1x   F1x 
−k     
 1 k1 + k 2 −k2  d 2 x  =  F2 x 
 0 −k2 k 2   d 3 x  F 
 3x 
here d1 x = d 3 x = 0 and F2 x = P
Hence, the above set of equations may be explicitly written as
− k1d 2 x = F1 x (1)
( k1 + k 2 ) d 2 x = P (2)
− k 2 d 2 x = F3 x (3)
P PL
From equation (2) d2x = =
k1 + k 2 3 EA
31
 Trusses
To calculate the stresses:
For element #1 first compute the element strain
(1) d 2 x − d1 x d 2 x P
ε = = =
L L 3 EA
and then the stress as
(1) (1) P
σ = Eε = (element in tension)
3A
Similarly, in element # 2
(2)d3x − d 2 x d2x P
ε = =− =−
L L 3 EA
(2) (2) P
σ = Eε = − (element in compression)
3A

32
 Trusses

© 2002 Brooks/Cole Publishing / Thomson Learning™

Inter-element continuity of a two-bar structure

33
 Trusses

Bars in a truss have various orientations

member in
compression

member in
tension

Connecting pin
34
 Trusses d 2y , f 2y
x̂
dˆ 2y , fˆ2y = 0 d̂ 2x , f̂ 2x
y ŷ
d 2x , f 2x
d 1y , f1y
dˆ 1y , fˆ1y = 0 θ
d̂ 1x , f̂ 1x d 1x , f 1x

x
At node 1: At node 2:
d̂ 1y d̂ 2y

d 1y d 2y
θ d̂ 1x θ d̂ 2x
d 1x d 2x f̂ 2y = 0
f̂1y = 0

f1y f 2y
θ f̂1x θ f̂ 2x
f1x f 2x
35
 Trusses
In the global coordinate system, the vector of nodal displacements
and loads
 d 1x   f1x  Element /local nodal
d̂ = displacement vector
d  f 
   
d =  1y ; f =  1y  Global nodal
d 2x f 2x d=
    displacement
 d 2y   f 2y  vector

Our objective is to obtain a relation of the form

f = k d
4×1 4× 4 4×1

Where k is the 4x4 element stiffness matrix in global coordinate system

36
 Trusses
The key is to look at the local coordinates
x̂
y ŷ dˆ 2y , fˆ2y = 0
d̂ 2x , f̂ 2x
 f̂1x  k - k   d̂ 1x 
 = - k   
dˆ 1y , fˆ1y = 0
θ  f̂ 2x   k   d̂ 2x 
d̂ 1x , f̂ 1x
EA
x k=
L
Rewrite as  f̂1x  k 0 -k 0   d̂ 1x 
  0   
 f̂1y   0 0 0   d̂ 1y  f̂ = k̂ d̂
 =  
 f̂ 2x  - k 0 k 
0  d̂ 2x 
 f̂    
 2y  0 0 0 0  d̂
 2y 

37
 Trusses

NOTES

1. Assume that there is no stiffness in the local ^y direction.

2. If you consider the displacement at a point along the local x direction as

a vector, then the components of that vector along the global x and y
directions are the global x and y displacements.

3. The expanded stiffness matrix in the local coordinates is symmetric and

singular.

38
 Trusses
NOTES
5. In local coordinates we have f̂ = k̂ d̂
4×1 4× 4 4×1

But or goal is to obtain the following relationship

f = k d
4×1 4× 4 4×1

Hence, need a relationship between d and d

d̂
and between f̂ and f d̂ 1y
 d̂ 1x  Need to understand how
 d 1x   
d 1y
the components of a
d  θ d̂ 1x
 1y   d̂ 1y  vector change with
d =   d̂ =   d 1x
d̂ 2y coordinate
d 2x
   d̂ 2x  transformation
 d 2y   d̂  d 2y
 2y  θ d̂ 2x
d 2x
39
 Trusses
Transformation of a vector in two dimensions

y vx v y cos θ Angle θ is measured

ŷ x̂
v̂ x positive in the
v̂ y
θ counter clockwise
v vy
v x sin θ direction from the +x
axis)
θ
v y sin θ x
v x cos θ

The vector v has components (vx, vy) in the global coordinate system and
^ , ^v ) in the local coordinate system. From geometry
(vx y

v̂ x = v x cos θ + v y sin θ
v̂ y = − v x sin θ + v y cos θ
40
 Trusses

In matrix form

 v̂ x   cos θ sin θ   v x 
 =  − sin θ  
 v̂ y   cos θ   v y 
Or Direction cosines
 v̂ x   l m vx  l = cos θ
 = −m    where
 v̂ y   l  vy  m = sin θ

Transformation matrix for a single vector in 2D

*  l m *
v̂ = T v
T = relates
− m l 
 v̂ x   v x  are components of the same
where v̂ =   and v =  
 v̂ y   v y  vector in local and global
coordinates, respectively.
41
Relationship between d̂ and d for the truss element
d̂ 1y
At node 1 d̂ 1x  *  d 1x 
 =T   d 1y
 d̂ 1y   d 1y  θ d̂ 1x
d 1x d̂ 2y
At node 2 d̂ 2x  *  d 2x 
 =T   d 2y
 d̂ 2y   d 2y  θ d̂ 2x
Putting these together d 2x
d̂ = T d

 d̂ 1x   l m 0 0   d 1x 
  − m l  d 
 d̂ 1y  T * 0
 =  0 0   1y  T = *
 0 0 l m   d 2x 
4× 4
0 T 
 d̂ 2x 
 d̂    d 
104 4 0 − m l   2y 
12 2y 
3 42 4 4 43 123
T d
d̂ 42
 Relationship between f̂ and f for the truss element
At node 1 f̂1x  f̂1y
*  f 1x 
 =T  
 f̂1y   f1y  f1y
θ f̂1x
f̂ 2y
At node 2 f̂ 2x  *  f 2x 
f1x
 =T  
 f̂ 2y   f 2y 
f 2y
θ f̂ 2x
Putting these together f̂ = T f f 2x
T * 0
 f̂1x  T = *
 l m 0 0   f1x  4× 4
0 T 
  − m l  f 
 f̂1y   0 0   1y  transformation matrix T
 =
 f̂ 2x   0 0 l m   f 2x  is orthogonal,

 f̂    f  T
−1
=T
T

104 4 0 − m l   2y 
12 2y 
3 42 4 4 43 123
T f
f̂ Use the property that l2+m2=1 43
 Trusses
Putting all the pieces together

x̂ f̂ = T f
y ŷ d̂ 2y , f̂ 2y
d̂ 2x , f̂ 2x d̂ = T d
θ f̂ = k̂ d̂
d̂ 1y , f̂1y
d̂ 1x , f̂ 1x
⇒ T f = k̂ˆ T d
x
(
⇒ f = T k̂ T d
1
424 3
−1
)
The desired relationship is f = k d k
4×1 4× 4 4×1

T is the element stiffness matrix in the

Where k = T k̂ T
4× 4 4 × 4 4× 4 4× 4 global coordinate system
44
 Trusses

 l m 0 0  k 0 -k 0
− m 0  0
l 0 0 0 0 
T= k̂ = 
 0 0 l m  - k 0 k 0
   
 0 0 −m l  0 0 0 0

 l2 lm − l2 − lm 
 2 
T EA  lm m2 − lm −m 
k = T k̂ T =
L  − l2 − lm l2 lm 
 
 − lm − m2 
2
lm m

45
 Trusses
Computation of the direction cosines

2 (x2,y2)
x 2 − x1 L
l = cos θ =
L
y 2 − y1 θ
m = sin θ =
L 1 (x ,y )
1 1

What happens if I reverse the node numbers?

x − x2 1 (x1,y1)
l ' = cos θ = 1 = −l L
L
y − y2
m ' = sin θ = 1 = −m θ
L 2 (x ,y )
2 2
Question: Does the stiffness matrix change?
46
 Trusses
Example Bar element for stiffness matrix evaluation
© 2002 Brooks/Cole Publishing / Thomson Learning™ E = 30 × 10 6 psi
A = 2 in 2
L = 60 in
θ = 30 o 3
l = cos 30 =
2
1
m = sin 30 =
 3 3 3 3  2
 − − 
 4 4 4 4 
 3 1
−
3
−
1 
k=
(
30 × 10 6 (2 )  4 ) 4 4 4  lb
60  3 3 3 3  in
 − − 
 4 4 4 4 
 3 1 3 1 
 − − 
 4 4 4 4 

47
 Trusses
Computation of element strains
© 2002 Brooks/Cole Publishing / Thomson Learning™

Recall that the element strain is

 d̂ 1x 
 
d̂ 2x − d̂ 1x 1  d̂ 1y 
ε= = [− 1 0 1 0 ] 
L L d̂ 2x 
d̂ 
 2y 
d û d̂ 2x − d̂ 1x 1
ε= = = [− 1 0 1 0 ] d̂
d x̂ L L
E
σ = Eε = d̂ 2x − d̂ 1x
L
( ) 1
= [− 1 0 1 0 ] T d
L
T = EAε =
EA
L
(
d̂ 2x − d̂ 1x )
{
k
EA
k=
L 48
 Trusses

 l m 0 0 
− m l 0 0 
1
ε = [− 1 0 1 0 ]  d
L  0 0 l m 
 
 0 0 −m l 
1
= [− l − m l m ]d
L
 d 1x 
d 
1  1y 
= [− l − m l m ]
L d 2x 
 
d 2y 

49
 Trusses
Computation of element stresses stress and tension

Recall that the element stress is

E
L
( )E
σ = Eε = d̂ 2x − d̂ 1x = [− l
L
−m l m ]d

Recall that the element tension is

d û d̂ 2x − d̂ 1x
ε= =
EA d x̂ L
T = EAε = [ −l −m l m ]d
L E
(
σ = Eε = d̂ 2x − d̂ 1x
L
)
T = EAε =
EA
L
(
d̂ 2x − d̂ 1x )
{
k
EA
k=
L 50
 Trusses
Steps in solving a problem

Step 1: Write down the node-element connectivity table linking

local and global nodes; also form the table of direction cosines
(l, m)
Step 2: Write down the stiffness matrix of each element in
global coordinate system with global numbering

Step 3: Assemble the element stiffness matrices to form the

global stiffness matrix for the entire structure using the
node element connectivity table

Step 4: Incorporate appropriate boundary conditions

Step 5: Solve resulting set of reduced equations for the unknown

displacements
Step 6: Compute the unknown nodal forces
51
 Trusses
Node element connectivity table

ELEMENT Node 1 Node 2

1 1 2
2 2 3

3 3 1

1 2 (x2,y2)
L
El 1 60 El 3
θ
2 60 60 3 1 (x ,y )
1 1
El 2
52
Stiffness matrix of element 1 Stiffness matrix of element 2
d1x d1y d2x d2y d2x d2y d3x d3y
  d1x   d2x
   
k
(1)
=  d1y k
(2)
=  d2y
  d2x   d3x
   
  d2y   d3y

Stiffness matrix of element 3

There are 4 degrees of
d3x d3y d1x d1y
freedom (dof) per element
  d3x (2 per node)
 
k
(3)
=  d3y
  d1x
 
  d1y
53
 (1)
k
Global stiffness matrix
d1x d1y d2x d2y d3x d3y
  d1x
  (2)
  d1y k
  d2x
K=  d2y
 
  d3x
( 3)
k
  d3y
  6× 6

How do you incorporate boundary conditions?

54
 Trusses
Example 2 The length of bars 12 and 23 are equal (L)
y E: Young’s modulus
3 A: Cross sectional area of each bar
El#2 P2
Solve for
P1 (1) d2x and d2y
El#1 (2) Stresses in each bar
2
45o
x
1 Solution

Step 1: Node element connectivity table

ELEMENT Node 1 Node 2

1 1 2
2 2 3

55
 Trusses
3
Table of nodal coordinates El#2 P2
Node x y P1
1 0 0 El#1 2
2 Lcos45 Lsin45 45o
x
1
3 0 2Lsin45 ELEMENT Node 1 Node 2

1 1 2

Table of direction cosines 2 2 3

ELEMENT Length x2 − x1 y 2 − y1
l= m=
length length
1 L cos45 sin45
2 L -cos45 sin45

56
 Trusses
Step 2: Stiffness matrix of each element in global coordinates with
global numbering d û d̂ − d̂ 2x 1x
ε= =
Stiffness matrix of element 1 d x̂ L

 l 2
lm −l 2
− lm  σ = Eε =
E
L
(
d̂ 2x − d̂ 1x )
 2 
(1) EA  lm m2 − lm −m  T = EAε =
EA
(
d̂ 2x − d̂ 1x )
k =
L  −l 2 lm  {L
− lm l2 k
  k=
EA
 − lm −m2 lm 2
m  L

d1x d1y d2x d2y

1 1 −1 −1  d1x
1 1 − 1 − 1 
EA   d1y
=
2L  − 1 − 1 1 1  d2x
 
 −1 −1 1 1  d2y
57
 Trusses
Stiffness matrix of element 2
d2x d2y d3x d3y
 1 − 1 − 1 1  d2x
 −1 1 1 − 1 
(2) EA   d2y
k =
2L  − 1 1 1 − 1  d3x
  d
 1 − 1 − 1 1  3y
ELEMENT Node 1 Node 2

1 1 2
2 2 3

58
 Trusses
Step 3: Assemble the global stiffness matrix
1 1 −1 −1 0 0
1 1 − 1 − 1 0 0 
  The final set of equations is
EA  − 1 − 1 2 0 −1 1 
K=   Kd = F
2L  − 1 − 1 0 2 1 − 1
0 0 −1 1 1 − 1
 
(1)
k
0 0 1 −1 −1 1 
d1x d1y d2x d2y d3x d3y
  d1x
 d1 x   F1x    d (2)
d  F    1y k
 1y   1y    d2x
d 2 x   P1  K=
d =  , F =    d2y
d 2 y   P2   
  d3x
(3)
d3x   F3 x  k
      d3y
 3 y 
d  F3 y    6× 6 59
 Trusses
3
El#2 P2
Step 4: Incorporate boundary conditions
P1
 0 
 0   d1 x   F1x  El#1 2
  d  F 
 1y   1y  45o
d2x  d 2 x   P1  x
d =   d =   , F =   1
d 2 y   P2 
d2y   d3x   F3 x 
 0   
 3 y 
d
 
 F3 y 
 
 0 
Hence reduced set of equations to solve for unknown displacements at
node 2

EA 2 0   d 2x   P1 
Kd = F 2 L  0  
2 d2y 
 =  
 P2 

60
 Trusses
d û d̂ 2x − d̂ 1x
ε= =
Step 5: Solve for unknown displacements d x̂ L
 P1 L  E
(
σ = Eε = d̂ 2x − d̂ 1x )
d 2x   E A  L
  =   EA
( )
d
 2y P
 2 L T = EAε =
L
d̂ 2x − d̂ 1x
{
 E A  EA
k

k=
L
Step 6: Obtain stresses in the elements
σ = Eε =
E
L
( )E
d̂ 2x − d̂ 1x = [− l
L
−m l m ]d
0

For element #1:  d1 x  0

d 
(1) E 1 1 1 1   1y 
σ = − −   
L 2 2 2 2   2x 
d
 d 2 y 
E P + P2
= (d 2 x + d 2 y ) = 1
2L A 2 61
 Trusses

For element #2: d2x 

 
(2) E 1 1 1 1  d2 y  0
σ =  − −   
L 2 2 2 2   3x 
d
0
 d 3 y 
E P1 − P2
= (d 2 x − d 2 y ) =
2L A 2

σ = Eε =
E
L
( ) E
d̂ 2x − d̂ 1x = [− l
L
−m l m ]d

62
 Multi-point constraints

 d1 x   F1x 
d  F 
  1 y  1y 
d 2 x   P 
d =  , F =  
d 2 y   F2 y 
d 3x   F3 x 
   
 3 y 
d  F3 y 

© 2002 Brooks/Cole Publishing / Thomson Learning™

Figure 3-19 Plane truss with inclined boundary conditions

at node 3

63
 Trusses
Problem 3: For the plane truss
P=1000 kN,
y L=length of elements 1 and 2 = 1m
P El#2 3 E=210 GPa
A = 6×10-4m2 for elements 1 and 2
2 = 6 2 ×10-4 m2 for element 3
El#1
El#3
Determine the unknown displacements and
45o reaction forces.
x
1
Solution
Step 1: Node element connectivity table
ELEMENT Node 1 Node 2
1 1 2
2 2 3
3 1 3
64
 Trusses
y
Table of nodal coordinates
P El#2 3
Node x y
1 0 0 2
El#1
2 0 L El#3
45o
3 L L 1 x

Table of direction cosines

x 2 − x1
ELEMENT Length x2 − x1 y − y1 l = cos θ =
l= m= 2 L
length length y 2 − y1
1 L 0 1 m = sin θ =
L
2 L 1 0

3 L 2 1/ 2 1/ 2
65
 Trusses
Step 2: Stiffness matrix of each element in global coordinates with
global numbering
Stiffness matrix of element 1
 l2 lm −l 2 − lm 
 2 
(1) EA  lm m2 − lm −m 
k =
L  −l 2 − lm l2 lm 
 
 − lm −m2 lm 2
m 

d1x d1y d2x d2y

0 0 0 0  d1x
-4 
9
(210 × 10 )(6 × 10 )  0 1 0 − 1  d1y
=
1 0 0 0 0  d2x
 
0 −1 0 1  d2y
66
 Trusses
Stiffness matrix of element 2 d2x d2y d3x d3y
1 0 −1 0  d2x
-4 
(2)
9
(210 × 10 )(6 × 10 )  0 0 0 0  d2y
k =
1  −1 0 1 0  d3x
 
0 0 0 0  d3y
Stiffness matrix of element 3
d1x d1y d3x d3y
 0.5 0.5 − 0.5 − 0.5  d1x
 − 0.5 
(3) (210 × 10 9 )(6 2 × 10 -4 )  0.5 0.5 − 0.5 d1y
k =
2  − 0.5 − 0.5 0.5 0.5  d3x
 
 − 0.5 − 0.5 0.5 0.5  d3y

67
 Trusses
Step 3: Assemble the global stiffness matrix

 0.5 0.5 0 0 − 0.5 − 0.5 

 0.5 1.5 0 − 1 − 0.5 − 0.5 

5
 0 0 1 0 −1 0  N/m
K = 1260 × 10  
 0 −1 0 1 0 0 
 − 0.5 − 0.5 −1 0 1.5 0.5 
 
 − 0.5 − 0.5 0 0 0.5 0.5 

The final set of equations is Kd = F Eq(1)

68
 Trusses
Step 4: Incorporate boundary conditions $
y
x$
 0  y
 0  3
  P El#2
 d 2 x 
d =   2
 0  El#1
 d 3x  El#3  d1 x   F1x 
  d 
 
F 
 1y 
 d 3 y  45o 1 y

d 2 x   P 
x d =  , F =  
1
d 2 y   F2 y 
d 3x   F3 x 
   
 3 y 
d  F3 y 
Also, $d3y = 0 in the local coordinate system of element 3

How do I convert this to a boundary condition in the global (x,y)

coordinates?

69
 Trusses
$
y
x$
 F1 x  y
 F 1 y  3
 P El#2
 P 
F =   2
 F2 y  El#1
 F3 x  El#3  d1 x   F1x 
  d 
 
F 
 1y 
 F 3 y  45o 1 y

d 2 x   P 
x d =  , F =  
1
d 2 y   F2 y 
d 3x   F3 x 
   
 3 y 
d  F3 y 
Also, Fˆ3 x = 0 in the local coordinate system of element 3

How do I convert this to a boundary condition in the global (x,y)

coordinates?

70
 Trusses
Using coordinate transformations
 d 3 x 
$  l m   d 3x  1
  =     l = m =
 d$ 3 y  −m l  d3y  2

 1 1   1 
 d$ 3 x   2 2   d 3x   2
( 3x
d + d 3 y )

⇒   =    =  
 d$ 3 y  − 1 1  d3y   1
   2
( d 3 y − d 3 x ) 
2 2 

$d3y = 0 (Multi-point constraint)

1
⇒ d$ 3 y =
2
(d 3 y − d 3x )= 0

⇒ d 3y − d 3x = 0 Eq (2)

71
 Trusses
Similarly for the forces at node 3
 Fˆ3 x   l m  F3 x  1
 ˆ =   , l = m =
 F3 y  − m l   F3 y  2
 1 1   1 
 Fˆ3 x   2   F  

(F3x + F )
3y 

2 3 x 2
 ˆ = 1 1   F3 y   1
= 
 F3 y  −  (− F3 x + F3 y )
 2 2   2 

Fˆ3 x = 0
1
ˆ
F3 x = (F3 y + F3 x ) = 0
2 Eq (3)
F3 y + F3 x = 0
72
 Trusses
Therefore we need to solve the following equations simultaneously
 F1 x 
Kd = F Eq(1)  F 1 y 

Eq(2)  P 
d 3y − d 3x = 0 F =  
 F2 y 
F3 y + F3 x = 0 Eq(3)
 F3 x 
 
 F 3 y 
Incorporate boundary conditions and reduce Eq(1) to
 0 
 1 −1 0  d2x   P   0 
     
1 2 6 0 × 1 0 5  − 1 1 .5 0 . 5   d 3x  =  F3 x   d 2 x 
d  F  d =  
 0 0 .5 0 . 5   3y   3y   0 
 d 3x 
 0.5 0.5 0 0 − 0.5 − 0.5   
 0.5 1.5 0 − 1 − 0.5 − 0.5   d 3 y 

 0 0 1 0 −1 0 
K = 1260 × 10 5  
 0 −1 0 1 0 0 
 − 0.5 − 0.5 −1 0 1.5 0.5 
 
 − 0.5 − 0.5 0 0 0.5 0.5  73
 Trusses
Write these equations out explicitly

1260 × 105 (d 2x − d 3x ) = P Eq(4)

1 2 6 0 × 1 0 5 ( − d 2 x + 1 . 5 d 3 x + 0 . 5 d 3 y ) = F 3 x Eq(5)
1 2 6 0 × 1 0 5 ( 0 .5 d 3 x + 0 .5 d 3 y ) = F3 y Eq(6)

Add Eq (5) and (6)

1 2 6 0 × 1 0 5 ( − d 2 x + 2 d 3 x + d 3 y ) = F 3 x + F 3 y = 0 using Eq(3)

⇒ 1260 × 105 (− d 2x + 3d 3x ) = 0 using Eq(2)

⇒ d 2 x = 3 d 3 x Eq(7)
⇒ 1 2 6 0 × 1 0 5 (3 d 3 x − d 3 x ) = P
Plug this into Eq(4)
⇒ 2520 × 105 d 3x = 106
P=1000 kN
74
 Trusses
⇒ d 3 x = 0 .0 0 3 9 6 8 m
d 2x = 3 d 3 x = 0 .0 1 1 9 m

Compute the reaction forces

 F1 x   0 − 0 .5 − 0 .5 
F   0
 1 y   − 0 .5 − 0 . 5  d2x 
 
 2y
F = 1 2 6 0 × 1 0 5
 0 0 0   d 3x 
F    d 
 3x  −1 1 .5 0 .5   3y 
 F 3 y   0 0 .5 0 . 5 
−500
−500
 
=  0  kN
−500
 
 5 0 0  75
 Trusses
Physical significance of the stiffness matrix

In general, we will have a stiffness matrix of the form

 k 11 k 12 k 13 
K =  k 21 k 22 k 23 
 k 31 k 32 k 33 

And the finite element force-displacement relation

 k 11 k 12 k 13   d 1   F1 
k  d  =  F 
 21 k 22 k 23   2   2
 k 31 k 32 k 33   d 3   F3 

76
 Trusses
Physical significance of the stiffness matrix

The first equation is

Force equilibrium
k 11 d 1 + k 12 d 2 + k 13 d 3 = F1 equation at node 1

Columns of the global stiffness matrix

What if d1=1, d2=0, d3=0 ?

While d.o.f 2 and 3 are held fixed
F1 = k 11 Force along d.o.f 1 due to unit displacement at d.o.f 1
F2 = k 21 Force along d.o.f 2 due to unit displacement at d.o.f 1
F3 = k 31 Force along d.o.f 3 due to unit displacement at d.o.f 1

Similarly we obtain the physical significance of the other entries

of the global stiffness matrix
77
 Trusses

In general

k ij = keeping
Force at d.o.f ‘i’ due to unit displacement at d.o.f ‘j’
all the other d.o.fs fixed

78
 Trusses
Example The length of bars 12 and 23 are equal (L)
y E: Young’s modulus
3 A: Cross sectional area of each bar
El#2 P2
Solve for d2x and d2y using the “physical
P1 interpretation” approach
El#1 2
45o
x
1 Solution

Notice that the final set of equations will be of the form

 k11 k12   d 2 x   P1 
k   = 
 21 k 22   d 2 y   P2 
Where k11, k12, k21 and k22 will be determined using the “physical
interpretation” approach
79
To obtain the first column  k11  d2x = 1
1   apply
y δ 2 = 1.cos(45) =
2  k 21  y d2 y = 0
3 F2y=k21
F2y=k21
F2x=k11 T2
El#2
2 F2x=k11
2’
El#1 T1 2
x 1 x
1 δ1 = 1.cos(45) =
2
d2x=1

Force equilibrium Force-deformation relations

EA
∑F x = k11 − T1 cos(45) − T2 cos(45) = 0 T1 =
L
δ1
∑F y = k21 − T1 sin(45) + T2 sin(45) = 0 EA
T2 = δ2
L
80
 Trusses
Combining force equilibrium and force-deformation relations
( T1 + T2 ) EA
k11 = = (δ1 + δ 2 )
2 2L
(T1 − T2 ) EA
k21 = = (δ1 − δ 2 )
2 2L
Now use the geometric (compatibility) conditions (see figure)
1
δ1 = 1.cos(45) =
2
1
δ 2 = 1.cos(45) =
2
Finally
EA EA 2 EA
k11 = (δ1 + δ 2 ) = ( )=
2L 2L 2 L
EA
k21 = (δ1 − δ 2 ) = 0
2L
81
To obtain the second column  k12  d = 0
  apply 2 x

y  k 22 y d2 y = 1
3 1
δ 2 = −1.cos(45) = − F2y=k22
2
2’ T2
El#2 d2y=1
2 F2x=k12
El#1 T1 2
x 1 x
1 δ1 = 1.cos(45) =
2

Force equilibrium Force-deformation relations

EA
∑F x = k12 − T1 cos(45) − T2 cos(45) = 0 T1 =
L
δ1
∑F y = k22 − T1 sin(45) + T2 sin(45) = 0 EA
T2 = δ2
L
82
 Trusses
Combining force equilibrium and force-deformation relations
( T1 + T2 ) EA
k12 = = (δ1 + δ 2 )
2 2L
(T1 − T2 ) EA
k22 = = (δ1 − δ 2 )
2 2L
Now use the geometric (compatibility) conditions (see figure)
1
δ1 = 1.cos(45) =
2
1
δ 2 = −1.cos(45) = − This negative is due to compression
2
Finally
EA
k12 = (δ1 + δ 2 ) = 0
2L
EA EA 2 EA
k22 = (δ1 − δ 2 ) = ( )=
2L 2L 2 L
83
© 2002 Brooks/Cole Publishing / Thomson Learning™ 3D Truss (space truss)

84
In local coordinate system f̂ = k̂ d̂
 f̂ 1x   d̂ 1x 
   k 0 0 −k 0 0  
 f̂ 1y   0 0 0 0 0 0   d̂ 1y 
    
 f̂ 1z   0 0 0 0 0 0   d̂ 1z 
 =  
 f̂ 2x   − k 0 0 k 0 0   d̂ 2x 
 f̂   0 0 0 0 0 0   d̂ 
 2y     2y 
 f̂   0 0 0 0 0 0  
d̂ 
 2z   2z 

85
 The transformation matrix for a single vector in 3D
*
d̂ = T d

 l1 m1 n1  l1, m1 and n1 are the direction cosines of x ^

T = l 2 n 2 
*
m2
l1 = cos θ x
 l 3 m3 n 3  m1 = cos θ y

© 2002 Brooks/Cole Publishing / Thomson Learning™

n1 = cos θ z

86
Transformation matrix T relating the local and global displacement and
load vectors of the truss element

d̂ = T d
T * 0
T = *
f̂ = T f
6× 6
0 T 

Element stiffness matrix in global coordinates

T
k =T k̂ T
6× 6 6× 6 6 × 6 6 × 6

87
  l1 2 l1 m1 l1 n1 − l1
2
− l1m1 − l1 n1 
 2 2 
 l1m1 m1 m1n1 − l1 m1 − m1 − m1 n1 
EA  l1 n1 − n1 
2 2
T m1 n1 n1 l1 n1 m1 n1
k = T k̂ T =  
L  − l1 2 − l1 m1 − l1 n1 l1
2
l1 m1 l1 n1 
− l m − m1
2
− m1 n1 l1 m1 m1
2
m1 n1 
 1 1 2 2

 − l1 n1 − m1n1 − n1 l1 n1 m1 n1 n1 

Notice that the direction cosines of only the local ^x axis enter the k
matrix

88