t^^ ^

:>^

w'^:.

If

\^:

H.V
-3 . oi-T^via^i'^t.
il

^P
ELEMENTARY THEORY OF STRUCTURES
 McGraw-Hill Civil Engineering Series

Harmer E. Davis, Consulting Editor

Babbitt • Engineering in Public Health

Babbitt and Dolaxd •
Water Supply Engineering
Benjamin •
Statically Indeterminate Structures
Davis, Troxell, and Wiskocil •
The Testing and Inspection of

Engineering Materials
Dunham Foundations of Structures
•

Dunham The Theory and Practice of Reinforced Concrete

•

Dunham and Young Contracts, Specifications, and Law for

•
Engineers
Gaylord and Gaylord Structural Design •

Hennes and Ekse Fundamentals of Transportation Engineering

•

Henry Design and Construction of Engineering Foundations

•

Krynine and Judd Principles of Engineering Geology and Geotechnics

•

LiNSLEY and Franzini Elements of Hydraulic Engineering '

LiNSLEY, KoHLER, AND Paulhus Applied Hydrology *

LiNSLEY, KoHLER, AND Paulhus *

Hydrology for Engineers

Matson, Smith, and Hurd •

Traffic Engineering

Mead, Mead, and Akerman •

Contracts, Specifications, and
Engineering Relations
NoRRis, Hansen, Holley, Biggs, Namyet, and Minami •
Structural
Design for Dynamic Loads
Peurifoy •
Construction Planning, Equipment, and Methods
Peurifoy •
Estimating Construction Costs
Troxell and Davis •
Composition and Properties of Concrete
Tschebotarioff •
Soil Mechanics, Foundations, and Earth Structures

Urquhart, O'Rourke, and Winter Design of Concrete Structures •

Wang and Eckel Elementary Theory of Structures

•
Elementary Theory of Structures

CHU-KIA WANG, Ph.D.

Professor of Architectural Engineering
University of Illinois

CLARENCE LEWIS ECKEL, C.E,

Professor of Civil Engineering

Dean of the College of Engineering
University of Colorado

McGRAW-HILL BOOK COMPANY, INC.

New York Toronto London

1957
 ELEMENTARY THEORY OF STRUCTURES
Copyright © 1957 by the McGraw-Hill Book Company, Inc. Printed in the
United States of America. All rights reserved. This book, or parts thereof,
may not be reproduced in any form without permission of the publishers.
Library of Congress Catalog Card Number 56-11058

III

THE MAPLE PRESS COMPANY, YORK, PA.

 PREFACE

This text designed to present the essential principles of structural

is

analysis in a first course for architectural and civil engineering students.

The analysis of statically determinate structures is based on the laws of
statics, while that of statically indeterminate structures depends on both
the principles of statics and the geometric conditions of the deformed
structure. These principles are relatively simple; nevertheless experi-
ence shows that, in order to acquire proficiency and facility, a student
must expect to work a considerable number of problems involving the
appropriate conditions of statics and geometry.
Incidental to the presentation of basic principles in this text, special
emphasis has been given to illustrative examples. It is hoped that this
feature will relieve the teacher of undue blackboard routine and thereby
permit time for lively and fruitful class discussion.
Statically determinate structures are discussed in Chaps. 1 through

10. Chapters 11 through 15 are devoted to an introduction to the analy-

sis of statically indeterminate structures. With the exception of Chaps.
6, 7, and 10, which deal with applications to the analysis of structures
such as roof trusses, building bents, and bridge trusses, the basic concern
of this text is the use of general principles and methods of structural
analysis.
In schools where " unified '^ courses in structural analysis and design
are offered, this text may be used as a principal source book for the
"analysis" portion of '^ design" assignments.

Teachers who prefer to give a ''unified" treatment of statically deter-

minate and statically indeterminate structures will find that Chaps. 3
and 11, Chaps. 4 and 12, and Chaps. 5 and 13 may be conveniently used
in pairs.
Chapters 8 and 9, which deal with influence diagrams and criteria for
moving loads, are general in nature. The topics discussed in these
chapters are essential for an understanding of the structural analysis of
bridge trusses or other structures carrying moving loads. Although
students in architecture or architectural engineering are likely to be pri-
marily interested in building structures, they will find these chapters,
VI PREFACE
and perhaps Chap. 10, of value in adding to their over-all understanding
of the procedures of structural analysis.
The methods and moment distribution are treated
of slope deflection
separately in Chaps. 14 and 15. Again a choice is permitted in that the
slope-deflection method, and then the moment-distribution method, may
be studied; or both methods may be discussed in relation to a given
problem at the same time. If the latter choice is made, Chaps. 14 and
15 may be used together.
Although great care has been taken in checking calculations and manu-
script, the authors will appreciate notices of errors and suggestions for
improvement in future editions.
The authors wish to thank Mrs. C. K. Wang for her valuable assist-
ance, especially in typing the final manuscript of this text.

C. K. Wang
C. L. Eckel
 CONTENTS

Preface v

Chapter 1. General Introduction 1

1-1. Theory of Structures Defined. 1-2. Layout and Classification of Struc-

tures. 1-3. Loads on Structures. 1-4. Methods of Analysis.

Chapter 2. Equilibrium of Coplanar-force Systems 4

2-1. The Free Body. Equilibrium of Coplanar-concurrent-force Sys-

2-2.
tems. Equilibrium of Coplanar-parallel-force Systems. 2-4. Equilib-
2-3.
rium of General Coplanar-force Systems. 2-5. Reactions on a Three-hinged
Arch.

Chapter 3. Shears and Bending Moments in Beams 26

3-1. Definition of Shears and Bending Moments. 3-2. Relationship between
Load, Shear, and Bending Moment. 3-3. Shear and Bending-moment Equa-
tions. 3-4. Shear and Bending-moment Diagrams. 3-5. Bending-moment
Diagram by the Graphic Method.

Chapter 4. Analysis of Statically Determinate Rigid Frames and

Composite Structures 48
4-1. Analysis of Determinate Rigid Frames.
Statically 4-2. Analysis of
Statically Determinate Composite Structures.

Chapter 5. Stresses in Trusses 59

5-1. Stress Analysis of Trusses. 5-2. Method of Joints. 5-3. Method of
Sections. 5-4. Method of Moments and Shears. 5-5. The Graphic Method.

Chapter 6. Analysis of Roof Trusses 77

6-1. General Description. 6-2. Dead, Snow, and Wind Loads. 6-3. Wind
Loads as Recommended in the 1940 ASCE Final Report. 6-4. Combina-
tions of Loads.

Chapter 7. Analysis of Building Bents 97

7-1. General Description. 7-2. Methods of Analysis.

Chapter 8. Influence Diagrams 106

8-1. General Introduction. 8-2. Definition. 8-3. Influence Diagrams for
Reactions on a Beam. 8-4. Influence Diagram for Shear in a Beam. 8-5.
Influence Diagram for Bending Moment in a Beam. 8-6. Influence Diagram

vii
Vm CONTENTS
as a Deflection Diagram. 8-7. Influence Diagrams for Simple Trusses. 8-8.
InfluenceDiagram between Panel Points of a Truss. 8-9. Influence Dia-
grams for Reactions on a Truss. 8-10. Influence Diagram for Shear in a
Parallel-chord Truss. 8-11. Influence Diagram for Bending Moment at a
Panel Point in the Loaded Chord of a Truss. 8-12. Influence Diagram for
Pier or Floor-beam Reaction. 8-13. Influence Diagram for Bending
Moment at a Panel Point in the Unloaded Chord of a Truss. 8-14. Influ-
ence Diagram for Stress in a Web Member of a Truss with Inclined Chords.

Chapter 9. Criteria for Maxima: Moving Loads 138

9-1. General Introduction. 9-2. Maximum Reactions and Shears in Simple
Beams: Uniform Loads. 9-3. Maximum Reactions and Shears in Simple
Beams: Concentrated Loads. 9-4. ]\Iaximum Bending Moment at a Point in
a Simple Beam: Uniform Load. 9-5. Maximum Bending Moment at a Point
in a Simple Beam: Concentrated Loads. 9-6. Absolute Maximum Bending
Moment in a Simple Beam: Concentrated Loads. 9-7. Maximum Reactions
on Trusses. 9-8. Maximum Shear in a Panel of a Parallel-chord Truss. 9-9.
Maximum Bending Moment at a Panel Point in the Loaded Chord of a Truss.
9-10. Maximum Bending Moment at a Panel Point in the Unloaded Chord
of a Truss. 9-11. Maximum Stress in a Web Member of a Truss with
Inclined Chords.

Chapter 10. Analysis of Highway and Railw^ay Bridges 177

10-1. General Description. 10-2. Dead Load. 10-3. Live Load on Highway
Bridges. 10-4. Live Load on Railway Bridges. 10-5. Impact. 10-6. Use
of Counters in Trusses. 10-7. Analysis of Bridge Portals.

Chapter 11. Analysis of Statically Indeterminate Beams 230

11-1. Statically Determinate vs. Indeterminate Beams. 11-2.
Statically
Deflections and Slopes in Statically Determinate Beams. 1 1-3. The Moment-

area Method. 11-4. The Conjugate-beam Method. 11-5. The Unit-load

Method. 11-6. Law of Reciprocal Deflections. 11-7. Statically Inde-
terminate Beams with One Redundant. 11-8. Statically Indeterminate
Beams with Two Redundants. 11-9. Influence Diagrams for Statically
Indeterminate Beams.

Chapter 12. Analysis of Statically Indeterminate Rigid Frames . . 267

12-1. Statically Determinate vs. Indeterminate Rigid Frames.

Statically
12-2. Deflections of Statically Determinate Rigid Frames: the Moment-area
Method. 12-3. Deflections of Statically Determinate Rigid Frames: the
Unit-load Method. 12-4. Analysis of Statically Indeterminate Rigid
Frames by the Method of Consistent Deformation. 12-5. Influence Dia-
grams for Statically Indeterminate Rigid Frames.

Chapter 13. Analysis of Statically Indeterminate Trusses .... 295

13-1. Statically Determinate vs. Statically Indeterminate Trusses. 13-2.

Deflections of Statically Determinate Trusses: the Unit-load Method. 13-3.
Deflections of Statically Determinate Trusses: the Graphical Method. 13-4.

Analysis of Statically Indeterminate Trusses by the Method of Consistent

Deformation. 13-5. Influence Diagrams for Statically Indeterminate
Trusses.
 CONTENTS IX

Chapter 14. The Slope-deflection Method 318

14-1.General Description of the Slope-deflection Method. 14-2. Derivation
of the Slope-deflection Equations. 14-3. Application of the Slope-deflection
Method to the Analysis of Statically Indeterminate Beams. 14-4. Applica-
tion of the Slope-deflection Method to the Analysis of Statically Indetermi-
nate Rigid Frames. Case 1. Without Joint Movements. 14-5. Application
of the Slope-deflection Method to the Analysis of Statically Indeterminate
Rigid Frames. Case 2. With Joint Movements.

Chapter 15. The Moment-distribution Method 345

15-1.General Description of the Moment-distribution Method. 15-2. Appli-
cation of the Moment-distribution Method to the Analysis of Statically Inde-
terminate Beams. 15-3. Check on Moment Distribution. 15-4. Stiffness
Factor at the Near End of a Member When the Far End Is Hinged. 15-5.
Application of the Moment-distribution Method to the Analysis of Statically
Indeterminate Rigid Frames. Case 1. Without Joint Movements. 15-6.
Application of the Moment-distribution Method to the Analysis of Statically
Indeterminate Rigid Frames. Case 2. With Joint Movements.

Answers to Problems 377

Index 385
 CHAPTER 1

GENERAL INTRODUCTION

1-1. Theory of Structures Defined. Engineers design structures such

as bridges, buildings, ships, machine parts, as well as various kinds of
equipment and other structural installations. Incident to design, the
engineer must first determine the layout of the structure, its shape, and
its constituent members. Then he must estimate or otherwise determine
the loads which the structure is to carry. The theory of structures deals
with the principles and methods by which the direct stress, the shear and
bending moment, and the deflection at any section of each constituent
member in the structure may be calculated. The next phase of the
design is to proportion the members in accordance with the allowable
working stresses of the material and other requirements for the proper

A typical roof truss

Fig. 1-1

functioning of the structure. This work is generally within the scope of

texts on strength of materials or structural design and will not be dis-
cussed in this text. It may be well to point out that the process of design
may have to be repeated a number of times before a satisfactory final
design can be found.
Consider, for example, the design of a typical roof truss such as is

shown in Fig. 1-1. The process of design involves four stages: (1) a lay-
out of the truss is assumed; (2) the loading, which may consist of dead
load (weight of the roofing material and the truss itself), snow load, wind
load, or other loading, is estimated; (3) the direct stresses in the members
of the truss are found;and (4) the sizes of the members are determined in
accordance with the design specifications. This text will concern itself
primarily with the third stage, but with occasional reference to the second
1
J ELEMENTARY THEORY OF STRUCTURES
stage. The eventual reconciliation between the first and the fourth
stages is largely a matter of experience.
1-2. Layout and Classification of Structures. The laj^out of any struc-
ture depends largely on the function of the structure, the loading con-
ditions, and the properties of the material to be used. Except in routine
situations, the determination of the layout of any structure requires
knowledge, judgment, and experience. Usually after two or more lay-
outs for the same structure are carried through the initial design stage, a
comparison is made to determine the preferred design. Sometimes the
preliminary layout has to be modified to meet unanticipated conditions
encountered in the later stages of design.
Basicalh^ most structures may be classified as beams, rigid frames, or
trusses or combinations of these elements. A beam is a structural mem-
ber subjected to transverse loads onl3\ It is completeh^ analj^zed when
the shear and bending-moment values have been found. A rigid frame
is a structure composed of members connected by rigid joints (welded
joints, for instance). A frame is completelj^ analyzed when the
rigid
variations in direct stress, shear, and bending moment along the lengths
of all members have been found. A truss is a structure in which all
members are usually assumed to be connected by frictionless hinges. A
truss is completeh^ analyzed when the direct stresses in all members have
been determined. There are also structural members or machine parts
which may be subjected to the action of direct stress, shear and bending
moment, and t"v\'isting moment.
1-3. Loads on Structures. Generally, the loads on structures consist
of dead load, live load, and the dynamic or impact effects of the live load.
Dead load includes the weight of the structure itself; live load is the
loading to be carried by the structure; and impact is the dynamic effect
of the application of the live load. Thus, in building design, the weight
of the flooring, beams, girders, and columns makes up the dead load;
while the weight of movable partitions, furniture, etc., the snow load, and
the wind load are considered as live load. Often the live load comes on a
structure rather suddenly or as a mo^-ing or rolling load, as, for example,
when a train passes over a bridge. In this case the live load is increased
by an estimated percentage to include its dynamic effect. This increase
is called the impact load.
obvious that most of the dead load, except such items as the roofing
It is
on roof trusses, ceiling plaster under floors, and handrails on bridges,
cannot be determined until the members have actually been designed;
therefore, dead load has to be first assumed and then checked after the
sizes of the members have been determined. Except for unusual struc-
tures the dead-load stress normally constitutes only a relatively small
 GENEIL\L INTRODUCTION 3

percentage of the total stress in a member; so that in routine designs a

modification of the first design is seldom necessary.
In its passage across the structure, the position which the live load
assumes in order to cause a maximum direct stress, shear, or bending
moment at a particular section in a member is of great importance and
will receive comprehensive treatment in this text.
1-4. Methods of Analysis. In Art. 1-1 it was stated that the theory
of structures deals with the principles and methods by which the direct
stress, shear, and bending moment at any section of the member may be
found under given conditions of loading. Because the forces acting on a
structural member may usually be assumed to lie in the same plane and
are in equilibrium, fundamental structural analysis involves the use of
the three equations of equilibrium for a general coplanar-force system;
viz., 2Fx = 0, ZFy = 0, and Z3/ = 0. These three equations, together
with a good working knowledge of simple arithmetic, algebra, geometry,
trigonometry, and some calculus, are the necessary prerequisites for
studying the elementary theory of structures.
 CHAPTER 2

EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS

2-1. The Free Body. No matter how complicated a structure may be

it may be assumed to be cut into various members, parts, or sections, each
of which is under the action of a system of coplanar forces. Any one
member, from the whole structure, is called
part, or section, thus set free
a free body. A drawn and complete with the magni-
free body, clearly
tudes and directions (both known and unknown) of all the forces acting
on it, is called a free-body diagram. The facility and ease with which the
free-body diagrams are chosen and drawn are the key to the subject of
structural analysis.
The free body, being at rest within the structure, must be in equi-
librium under the action of all the coplanar forces acting on it. If the
magnitude, or the direction, or both, of some of these forces are unknown,
they can be found by the principles of statics, which are the three equa-
tions of equilibrium 2Fx = 0, 2Fy = 0, and 'EM = 0. In this chapter,
the methods of solving for these unknown magnitudes or directions will
be explained.
2-2. Equilibrium of Coplanar-concurrent-force Systems. If the free
body happens to be a point (a pin, for instance), the forces acting on it are
concurrent. The resultant of a coplanar-concurrent-force system must
be a single force, the x component of which is XF^ of the component forces
and the y component is XFy. Thus the two equations XFx = and
EFy = are necessary and sufficient to ensure that the resultant is zero
or that the coplanar-concurrent-force system is in equilibrium. These
two conditions for equilibrium permit the calculation of two unknowns,
which may be the magnitudes of two forces with known directions, or
the magnitude of one force with known direction and the direction of
another force with known magnitude.
It should be noted that the x and y directions are purely arbitrary;
thus, in applying the equation XFx = 0, any direction may be considered
as the X axis. Also, as long as the concurrent forces are in equilibrium
and have no resultant, the sum of the moments of the component forces
about any point in the plane must be zero. Thus in cases where they
may be more conveniently applied, the moment equations EM a = and
4
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS O

23/^ = may be substituted for either or both of the resolution equa-

tions ^F, = and i:Fy = 0.
In the graphic method of finding the resultant of a coplanar-concurrent-
force system, a zigzag line is drawn connecting successively the component
vectors taken in any convenient order; the resultant is then given by the
vector extending from the starting point of the first component vector
to the end point of the last component. Should the resultant be zero,
the end point of the last component force must coincide with the starting
point of the first component. Thus the graphic condition for the equi-
librium of a coplanar-concurrent-force system is that the force polygon
must close.
For example, if the four coplanar, concurrent forces ah, he, cd, and de
as sho^\^l in the space diagram of Fig. 2-la are in equilibrium, the points
A and E in the force polygon ABODE of Fig. 2-16 must coincide. Note
that the forces in Fig. 2-la may be designated in an irregular order,
although they are normally named in alphabetical order around point

(a) Space diagram

Fig. 2-1

in either the clock^^^se or counterclockwise direction. It should also be

noted that the position and direction, but not the magnitude, of the forces
must be plotted accurately in the space diagram; while both magnitude
and direction of the forces, and not the position, are represented in the
force polj'gon.
In the examples which follow, both the algebraic and graphic solutions
are given.
Example 2-1. If the four coplanar, concurrent forces Fi, Fo, Fz, and 7^4
shown in Fig. 2-2a are in equilibrium, find the magnitude and direction
of F4 which is arbitrarily assumed to act in the direction shown.
ALGEBRAIC SOLUTION. From ZF^ = 0,

50 cos 15° + 100 cos 45° - 80 sin 30° -h (F,),=

{Fa). = -(50)(0.966) - (100)(0.707) -h (80)(0.500) = -79.0 1b

Therefore (Fa), acts to the left as shown in Fig. 2-26 and not to the
right as assumed in Fig. 2-2a.
 ELEMENTARY THEORY OF STRUCTURES
A -149^ lb

Ji-lOOIb

-F3-80Ib

From ZFy = 0.

50 sin 15° - 100 cos 45° - 80 cos 30° -h (FA)y =

(F4), = -(o0){0.2o9) + (100)(0.707) + (80)(0.866) = +127.0 lb

(F4)y acts upward as assumed in Fig. 2-2a and as shown in Fig. 2-26.
Referring to Fig. 2-26.

'9.0
= =
tan <p
Y^ = 0.622 <t>
31.9^

(OaJz = 90° + 31.9° = 121.9°

127.01 127.0
F4 = = 149.6 lb
cos <p 0.849

CHECK. By 2(Fx), = (refer to Fig. 2-2c),

-hoO cos 0° - 149.6 cos 83.1° -f 100 cos 60° - 80 cos 45° «
+ (50;(lj - (149.6j (0.291 + (100;(0.o00j - (80; (0.707; «
j

-0.1 =

GRAPHIC SOLUTION. In Fig. 2-3. the forces AB. BC, and CD are
drawn in .succe.s.sion to scale and in direction as given. The magnitude
and direction of F4 are then given by the vector DA.
B

Scale.
1 in. - 100 lb
Ans.
/4- 150 lb

(a) Space diagram (6) Force polygon

Fig. 2-3
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 7

Example 2-2. If the four coplanar, concurrent forces Fi, F2, F3, and F4
shown in Fig. 2-4a are in equilibrium, find the magnitudes of F3 and F4
acting in the directions shown.
ALGEBRAIC SOLUTION. Since these forces are concurrent, this problem
may by resolving the forces into components
be solved parallel to the
X and y axes and applying the conditions XFx = and "ZFy = 0; how-
ever, this requires the use of simultaneous equations. A simpler solu-
tion involving the solution of two separate equations is obtained by
(1) finding F3 from SMz) = 0, (2) finding 7^4 from SMc = 0, and (3) check-
ing by 2Fx = and 2/^y = 0. It is usually easier to find the moment of

Fi y- 74.3 lb

y Jy
y
- 29.75 lb
0, A
\
\ u Psx \ \t
-\ 1^ / ^ (^ t
if '

(' 7i
\ /
\ K'J

— - --
? y^ '
^ <?
^^
> ^—^
X

r ^4
^A ../£>

B ^
"i-'i F2x-48lh '

'4x
Iclivi 3ior1-]Lft
i

(a)
F2y 36 lb (b )

Fig. 2-4

forces such as these about a point D by use of the principle of moments,

which states that the moment of a force about a point is equal to the
sum of the moments of its two rectangular components obtained by
resolution at any convenient point on the line of action of the force.
Thus, from XMd = (Fig. 2-46),

-(29.75) (7) - (74.3) (4) - (48) (1) ++ 7(^3). - 10(^3). =

(36) (10)
7(Fs), - = +193.45
10(Fz)y
But (Fa), - HiFz).
so (F3)x(7 - 10 X ^) = +193.45
(Fz), = -35.2 lb {Fz)y = -44.0 lb F3 = -56.3 lb

Note that circles are drawn around the arrowheads of {Fz)^, (Fsjy, and F3
in Fig. 2-46 to indicate that they are actually in directions opposite to
those assumed in Fig. 2-46. From l^Mc = 0,

(29.75) (0) + (74.3) (6) - (48) (8) + (36) (0) + 7(7^4): 10(^4). =
7(F4)x - 10(^4). = -61.8
But (F4), = HiF,).
{F,).{7 - 10 X Vs) = -61.8
(F,y. -16.85 1b (^4); 5.62 lb Fi = -17.76 1b
 ELEMENTARY THEORY OF STRUCTURES
CHECK
:f. = +29.75 - 48 + 35.2 - 16.85 «
+0.10 =
2F. +74.3 - 36 - 44.0 + 5.62 ^
-0.08 =

GRAPHIC SOLUTION. The space diagram and the force polygon are
shown in Fig. 2-5. AB = Fi and BC = Fo are first plotted; D is the
point of intersection of two lines drawn through C and A^ parallel,

Scale.
1 in. = 80 lb
Ans.
F3=CZ)=56.31b
F4-DA-17.81b

1 division = 1 ft

(a) Space diagram (6) Force polygon

Fig. 2-5

and F4. For equilibrium the force polygon ABC DA

respectively, to F3
must and the true magnitudes and directions of F3 and F4 are equal
close
to CD and DA acting as shown by the arrows in Fig. 2-56. The circles
drawn around the arrowheads of cd and da in the space diagram indicate
that initially the directions of these forces were incorrectly assumed.

F4 = 601b
F2 = 801b

Fi = 100 lb
Fi = 100 lb

rF4 = 60 1b xi
(2d solution)

(a) (6)

Fig. 2-6

Example 2-3. The four coplanar, concurrent forces Fi, F2, F3, and F4
shown in Fig. 2-6a are in equilibrium. Find the magnitude of Fz acting
as shown, and the direction of Fa.
ALGEBRAIC SOLUTION. The Xi^i axis is drawn perpendicular to the
line of action of F3 in Fig. 2-66. There are two positions of F^ = 60 lb,
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 9

each at an angle from the XiO^i axis, which will satisfy the condition

2(FJi = 0. FromS(FJi = 0,

-80 cos 30° + 100 cos 75° + 60 cos =

60 cos </> = (80) (0.866) - (100) (0.259) = 43.38
= ±43.7°

FIRST SOLUTION [(64)^ = 358.7°]. From ^F^, = 0,

+ 100 cos 30° - 80 sin 15° - F^ cos 45° + 60 cos 1.3° =

F3Cos45° = +(100)(0.866) - (80)(0.259) + (60)(1.000) = +125.9
^3 = 178.1 lb
CHECK. By XFy = 0,

+ 100 sin 30° + 80 cos 15° - 178.1 sin 45° - 60 sin 1.3° ^
+ (100) (0.500) + (80) (0.966) - (178.1) (0.707) - (60) (0.023) «
=
SECOND SOLUTION [(64) ^c = 271.3°]. FroiH 2Fa; = 0,

+ 100 cos 30° - 80 sin 15° - Fs cos 45° + 60 sin 1.3° =

F3Cos45° = +(100) (0.866) - (80) (0.259) + (60) (0.023) = +67.3
^3 = 95.2 lb
CHECK. By 2Fy = 0,

+ 100 sin 30° + 80 cos 15° - 95.2 sin 45° - 60 cos 1.3° «
+ (100) (0.500) + (80) (0.966) - (95.2) (0.707) - (60) (1.000) «
=
GRAPHIC SOLUTION. In the force polygon of Fig. 2-76, AB and BC
are first plotted. With A as center and 60 lb as radius, an arc isdrawn
i^4 = rfa = 601b
F2=6c=801b ' c
w 1
f
Fi = c6=100 1b /
/ \\
\ K'>r^^°
\^^v\ hx/^
A Scale.

/I
1 in. = 120 lb
45°\/|^ \l
e/ ^^
Ans.
Ist solution

/^y ^\ ^^ F3-CZ>i = 1781b

Fz^cd dXt--^^ »4x-358°

^
V / -^ 2d solution
F3-CZ)2 = 95 1b

(a) Space diagram (6) Force polygon

Fig. 2- 7

to intersect a line through C and parallel to cd. The points of inter-

section are Di and D2. As shown in Fig. 2-76, the values of 7^3 are
CD2 \Fz = 95 lb and (^4)^ = 272°] and CD^ [F, = 178 lb and {d,), =
10 ELEMENTARY THEORY OF STRUCTURES
358°]. Vectors D2A and DiA = A study of Fig. 2-76 shows
60 lb.

that there may be no solution, one solution, or two solutions depending

on whether ^D is shorter than, equal to, or longer than the perpendicu-

lar AE.
2-3. Equilibrium of Coplanar-parallel-force Systems. The resultant
of a coplanar-parallel-force system must be either a single force or a
couple. The resultant cannot be a single force if the summation of the
forces is zero and the resultant cannot be a couple if the summation of
the moments of the forces about any point in the plane of the forces is
zero. Thus the two equations SF = and SM = are necessary and
sufficient to ensure that the resultant is zero or the coplanar-parallel-force
system is in equilibrium. Two unknowns
can be found from these two
conditions of equilibrium. These unknowns may be the magnitude
(including sense or direction) and position of the same force, the magni-
tudes of two forces (usually called the reactions), or the magnitude of one
force with known position and the position of another force with known
magnitude.
In the algebraic solution, one of the two equations used must be a
moment equation, while the other may be either a resolution or a moment
equation.

(a) Space diagram (6) Force polygon

Fig. 2-8

The graphic conditions for equilibrium of a coplanar-parallel-force

system will now be established. In the space diagram of Fig. 2-Sa are
shown the positions of four coplanar-parallel forces ah, he, cd, and de. If
these four forces are in equilibrium, the line-force polygon ABODE of
Fig. 2-86 must close, or A and E must coincide to ensure that ZF = 0.

A point 0, called the pole, is arbitrarily chosen in Fig. 2-86 and rays OA
(or OE), OB, 00, and OD are drawn. The force AB is the vector sum of
AO and OB; BO, of BO and 00; OD, of 00 and OD; and DE, of DO and
OE. Or,
AB = AO-^ OB
BO = BO-^ 00
CD = 00-^ OD
DE = DO-^ OE
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 11

Adding and noting that OB -f> BO = 0, etc.,

AB -\^ BC -^ CD -^ DE = AO -^ OE
Thus the four original forces are now replaced by two forces AO and OE,
the resultant of which cannot be a single force because AO and OE are
equal and opposite. To ensure that the resultant is not a couple the
lines of action of ^0 and OE must be collinear. If in the space diagram
of Fig. 2-Sa, the strings oa, oh, oc, od, and oe are drawn in succession and
respectively parallel to the rays in the force polygon, oa and oe must be
collinear. The polygon 1-2-3-4-1 in Fig. 2-Sa is usually called the string
polygon, or the equilibrium polygon. Essentially the forces ao and oe
replace the given force.system and the fact that ao and oe are two collinear,
equal, and opposite forces ensures that the given force system is in equi-
librium. Thus the two graphic conditions for equilibrium of a coplanar-
parallel-forcesystem are: (1) the force polygon must close, and (2) the
string polygon must close. Note that, in the actual solution of a problem,
the force de would have been called da at the outset.

a a a
8

^ 3'^ 4' ^ X
f 1 3' (*; 6.4'

A B c

(a) (6)

Fig. 2-9

Example 2-4. If the four coplanar, parallel forces Fi, F2, Fz, and F4
shoAvn in Fig. 2-9a are in equilibrium, find the magnitude and position of F^.
ALGEBRAIC SOLUTION. From 2F = (Fig. 2-9a),

80 + 20 200 + F,
=
Fi = -100 lb or 100 lb downward

Because a negative sign for Fa is obtained, a circle is drawn around the

arrowhead of F^ in Fig. 2-9a to indicate that Fi is actually 100 lb down-
ward, as shown in Fig. 2-96. From XMc = (Fig. 2-9a),

(80) (7) + (20) (4) = (100) (x)

X = 6.4 ft

CHECK. By ^Ma = 0,

(20) (3) + (100) (13.4) « (200) (7)

1,400 = 1,400

GRAPHIC SOLUTION. In the space diagram of Fig. 2-lOa, the known

forces Fi, F2,and F3 are called ab, be, and cd. The unknown force 7^4 is
called da. The force polygon ABCD of Fig. 2-106 is drawn; F^ = DA
is scaled to be 100 lb downward. The pole is arbitrarily chosen and
12 ELEMENTARY THEORY OF STRUCTURES
rays OA, OB, etc., are drawn. The string oa is drawn parallel to the ray
OA, intersecting ah at any point such as 1. From 6 to 6 draw the string
ob or 1-2. From c to c draw the string oc or 2-3. Through point 3 draw
the string od which intersects the string oa at point 4. Point 4 is on the
line of action of da, which is scaled at x = 6.4 ft from cd.
It is to be noted that, in Fig. 2-106, AO and OB are components of
force AB and may be assumed to act through any point such as 1 on the

-801b ^2 = 201b F3 = 200 lb F,

3
oc^
2
d\a
^ \orf

1
^^ ,,__^oa.
\ 1

a h h c c -^4
3' 4' X Scale.
>|
1 in. = 200 lb
N \
Scale. Ans.
1 in. = 10 ft F4°DA= 100 lb
*=6.4ft
(a) Space diagram (6) Force polygon

Fig. 2-10

line of action of ah in Fig. 2-lOa. In the same manner ho and oc are

components of he and must pass through point 2. Likewise co and od
are components of cd and must pass through point 3. Components
along 1-2 and 2-3 cancel; thus do and oa are components of F4 and their
intersection must be on the line of action of Fa.
The three known forces have been named in the order ah, he, and cd
from the left toward the right. Actually any force can be called ah, and
the other two he and ed. For instance, Fz may be called ah; Fi, he; and
F2, cd. The same position for F4 will
Pl-lOOlb P2 = 801b Pz = 40 lb
be found by following the procedure
3'

! -M '
I
AB
described above.
Example 2-5. Three loads Pi, P2,
20'
and Fz act on the beam as shown in
v> 4..
Fig. 2-11. Find the reactions Va
Fig. 2-11
and Vb.
ALGEBRAIC SOLUTION. From I^Ma = 0,
(100) (4) + (80) (12) -h (40) (23) = 207^
2,280= 20F^
Vb = 114 lb
From XMb = 0,

20V A + (40) (3) = (100) (16) + (80) (8)

20VA = 2,120
Va = 106 lb
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 13

CHECK. By SF = 0,

100 + 80 + 40 106 + 114

220 220

GRAPHIC SOLUTION. The three known forces Pi, P2, and P3 are
designated as ah,be, and cd in the space diagram of Fig. 2-12a and plotted

as AB, BC, and CD in the force polygon of Fig. 2-126. The problem is
to locate the point E which will divide the load line ^D in segments
proportional to the reactions Va and Vb- The pole is arbitrarily

located and rays OA, OB, etc., are drawn. The unknown reactions in

Scale.
1 in. = 200 lb

Ans.
Va=EA= 106 lb
Vo = D£=114lb

la) Space diagram (6) Force polygon

Fig. 2-12

Fig. 2-12a are called de and ea. The strings oa, oh, etc., are drawn in
succession, respectively parallel to the corresponding rays. It is impor-
tant to note that the string oa, or 1-2, is drawn from a to a; oh, or 2-3,

from h toh; oc, or 3-4, from c to c; and od, or 4-5, from d to d. By joining
the points and 5, the string oe
1 from e to 6 is obtained. The ray OE
is then drawn parallel to oe. By scaling, Va = EA = 106 lb and
y^ = DE = 114 1b.

It is to ah, he, and ed may be assigned

be noted that the designations
to the three known any order and similarly de and ea to the two
forces in
unknown reactions Va and Vb. The reader will get good practice in
repeating this solution by designating P2 as ah. Pi as he, P3 as ed, Va as de,
and Vb as ea. In any event the unknowns must have a common letter,
which is 6 in this case.
Example 2-6. The five coplanar, parallel forces shown in Fig. 2-1 3a
are in equilibrium. Find the magnitude of F^ and the position of Fb.
ALGEBRAIC SOLUTION. From 2F = (Fig. 2- 13a),

50 + 80 + F4 = 60 +
40
F4 = -30 lb or 30 lb downward
14 ELEMENTARY THEORY OF STRUCTURES
(D
0.8'

4' 3' 5' \ 3.2'

B c
Si

SI ^ i i

(a) it)

Fig. 2-13

From 2Ma = (Fig. 2-13a),

(40) (4) + (60) (12) = (80) (9) + (50) (a: + 12)

50x = -440
X = -8.8 ft or 8.8 ft to the right of B
CHECK. By XMc = (Fig. 2-136),

(60) (3) + (50) (5.8) - (40) (5) + (30) (9)

470 = 470

GRAPHIC SOLUTION. The graphic solution is shown in Fig. 2-14. This

follows the preceding analysis and requires no detailed explanation. It
will be noted that strings od and oe intersect at point 5, thereby determin-
ing a point on the line of action of Fh or de.

b b c c a
E
oby^ v^C C

^^ oa ^ %^ A

3' 5'

r
M
1
1
D

B
Scale.
1 in.

Arts.
- 100 lb

F4-EA- 30 lb
X- 0.8ft

Scale.
1 in. -10 ft

(a) Space diagram (6) Force polygon

Fig. 2-14

2-4. Equilibrium of General Coplanar-force Systems. The resultant

of a general coplanar (noncurrent and nonparallel) force system must be
either a single force or a couple. The resultant cannot be a single force
if XFx and both zero and the resultant cannot be a
'LFy of the forces are
couple if XM of the forces about any point in the plane of the forces is
zero. Thus the three equations 2i^x = 0, SF^ = 0, and SM = are
necessary an dsufficient to ensure that the resultant is zero, or that the
general coplanar-force system is in equilibrium. From these three con-
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 15

ditions for equilibrium, three unknowns can be found. The more usual
combinations of unknowns are: (1) the magnitude, direction, and posi-
tion of the same force; (2) the magnitude of one force and both the
magnitude and direction of another; and (3) three magnitudes.
Although it is obvious that not more than two of the equations used
may be resolution equations, the resolution and moment equations may
be used freely so that the unknowns can be solved as directly as possible
without resorting to a system of simultaneous equations.
The two graphic conditions for equilibrium of a general coplanar-force
system are: (1) the force polygon must close, and (2) the string polygon
must close. Take, for instance, five coplanar forces Fi through F5, the
positions of which are given by ab, he, etc., in the space diagram of Fig.
2-15a and the magnitudes and directions are given by AB, BC, etc., in
the force polygon of Fig. 2-156. If the five forces are in equilibrium, the

{a) Space diagram (6) Force polygon

Fig. 2-15

firstrequirement is that points A and F must coincide in the force poly-

gon. This is equivalent to the two algebraic conditions 2/^^ = and
2Fj/ = 0. In case the five forces are known to be in equilibrium initially,
the force ef or EF would have been called ea or EA, thus eliminating
the letters / and F. Next, any arbitrary point 0, the pole, is chosen, and
the rays OA, OB, etc., are drawn. By beginning at any point 1 on ab, the
and 4-5 are drawn in succession, respectively parallel
strings 1-2, 2-3, 3-4,
to OB, OC, OD, and OE. (Note the rule, ''from 6 to 6 draw ob parallel
to OB, etc.") For equilibrium, the line joining points 1 and 5 must be
parallel to OA The reason is that each of the five given forces
or OF.
is replaced by two by ao and ob, for example), but the four
forces {ab
intervening pairs (such as ob and bo) balance themselves. The two
remaining forces ao and of, which are equivalent to the sum of Fi to F^,
inclusive, must therefore be equal, opposite, and coUinear (see string
1-5 in Fig. 2-15a).
Example 2-7. The three forces Fi, F2, and F^ shown in Fig. 2-16a and
another force F4 are in equilibrium. Find the magnitude, direction, and
position of F^.
16 ELEMENTARY THEORY OF STRUCTURES
Fi - 300 lb

^X<30°
(2,

(-3.0)

1 unit- 1ft
^^jSo.- 0.98' X
[^
379_8
160

^2= 150 lb

F3 - 200 lb

(a)
(6) (c)

Fig. 2-16

ALGEBRAIC SOLUTION. From SF^ = (Fig. 2-16a),

300 cos 30° + 200(%) + {F,), =

(F4)x = -379.8
From XFy = 0,

300 sin 30° - 150 - 200(^^) + {F,)y =

(F,)y = +160
SMo of Fi, F2, andFa = -(259.8) (3) + (150) (2) -\- (150) (3) + (120) (2)
+ (160)(1)
= —370.6 or 370.6 counterclockwise
Thus Mo of F4 = 370.6 ft-lb clockwise
Absolute value of x intercept of F^
Mo of Fi
{Fi\
370.6
2.32 ft
160

Absolute value of y intercept of Fi

Mo of Fi
(Fi).
370.6
0.98 ft
379.8
Referring to Fig. 2-166 and c,

OA = 2.32 ft OB = 0.98 ft

^^" * - - «-421 cf>

= 22.8° ^ 157.2°
sfo"^

'^4
_ 379.8
= = 412 lb
or» oo n noo

Thus Fi = 412 lb at 6^ = 157.2° passing through the point (-2.32,0)

or (0,-0.98).
CHECK. A check can be made by verifying that XM of the four forces
about each of the three points (2,3), ( — 3,0), or ( — 1,-2) is zero.
GRAPHIC SOLUTION. In the force polygon of Fig. 2-176, the forces
are laid out and the closing force Fi is determined. Next a pole such as
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 17

Fi = Q6 = 3001b

Scale.
1 in. = 400 lb
F3 = cc/=200lb Arts.
F4 = 412 1b
^4x = 157»
OA = 2.32 ft

(a) ffpace diagram (6) Force polygon

Fig. 2-17

is and rays OA, OB, OC, and OD are drawn. In the space
selected
diagram drawn parallel to the ray OA,
of Fig. 2-1 7a, the string oa is
intersecting ab at some convenient
point 1. Then strings oh (1-2), oc
200 1b
390 1b

(2-3),and od are drawn in succes-

Strings oa and od intersect
sion.
at point 4, which is on the line of
'^'
- 3'

action of da (Fi). 20'

Example 2-8. The beam shown

Ha ^By
in Fig. 2-18a has a roller support
ia)
at A and a hinge support at B.
390 1b
Find the magnitude of Ra and the 200 1b

magnitude and direction of Rb due

to the three loads shown. ^
^\ tL.
ALGEBRAIC SOLUTION. The hori- fT
zontal and vertical components of
the hinge reaction at B will be des-
(h)
ignated as {Rb)x and {Rb)^
Fig. 2-18
From SF:, = (Fig. 2-18a),

-f(200)(%) - (390) (K3) - (Rb). =

(Rb). = -30 1b
From 2Mb = 0,

20Ra = (200)(^^)(17) + (390)(i?l3)(10) + (80)(6)

Ra = 340 lb

From XMa = 0,

20{Rb\ = (200)(^^)(3) + (390)(iK3)(10) + (80)(14)

{RB)y = 260 lb
18 ELEMENTARY THEORY OF STRUCTURES
CHECK. By 2F^ = 0,

(200)^^) + (390)(i%3) + 80 - 340 + 260

600 = 600
Referring to Fig. 2-186,

tan 2^260 = 0.1154 = 6.6 d, = 83.4°

260 260
Rb = 262 1b
cos 6.6° 0.9934

It ^Adll be noted that a hinge reaction can be defined either b}^ its hori-
zontal and vertical components or by its magnitude and direction.
Because of its convenience, the former is recommended.

^J^^^^^f^ 1

4
A
3' \ 7 / 4'
-
''
\\
'1

a c a >^ hjc c d

200 1b 390 1b 80 lb Scale.

lin. = 400 1b"
Scale. Ans.
1 in.= 12'
/?^=£A = 340 1b
i?fi=I>f:=262 1b
^x=83''
(a) Space diagram (6) Force polygon

Fig. 2-19

GRAPHIC SOLUTION. In the force polygon of Fig. 2-196 the known

AB, BC, and CD.
forces are first plotted as provided Because a roller is
at A Ra must be vertical and the unknown
, point E must lie on a vertical
line through A. In the space diagram of Fig. 2-19a the string od (B-l)
must first be dra^^^l through B, the only known point on the Hne of action
of de (Rb). The strings oc (1-2), oh (2-3), and oa (3-4) are then drawn.
The closing line BA must be the string oe. From the pole a line is drawn
parallel to oe, intersecting the vertical line through A at E. Thus
Rb = DE arnd Ra = EA.
Example 2-9. The five coplanar forces acting on the 4- by 8-ft board
(Fig. 2-20a) are in equilibrium. Find the magnitudes of Fz, Fa, Fb. and
ALGEBRAIC SOLUTION. In Order to avoid the solution of simultaneous
equations, it will be desirable to find one of the unknowns bj^ taking
moments about the point of intersection of the lines of action of the other
two unknown forces. Thus from XMa = (Fig. 2-20a),

(200) (2) - (100) (2) (y5Fz){5) + {HF,){1) =

Fz = 58.8 lb (Fa): = 35.3 lb {Fz\ 47.0 lb
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 19
F2 - 200 lb

B 1
1 1 1 1 1

-100 1b 1 1 1 1 1
1

- -1- -
- - -r-T - T
1

T
> 1 1 1
,

._+._J__,__l__4-J^
1
/I ' ' '

1
'

1
1

1
1

1
I

1
^ 1

>r 1 division = 1 ft

A-lOOlb^ ::'!. V- F5= 151.3 lb

--Fox =135.3 lb
1 1 1 I I . I
;^
__L_J__J__l_ _l_^^-"_^_ _

F3x=35.31b
1 division = 1 ft

F3y=47.01b F4 = 220.6 lb

F3 = 58.81b
(b)

Fig. 2-20
From 2F:, = 0,

+ 100 + 35.3 + (/^5)x =

(F5). = -135.3 lb {F,)y = -67.6 lb F, = -151.3 lb

From ZFy = 0,

200 + 47.0 + F4 - 67.6

F4 = 220.6 lb

CHECK. The best check can be made by using as many of the previ-
ously unknown forces as possible in an independent equation. Thus from
XMb = (see Fig. 2-205),

+ (100)(1) - (200) (4) + (35.3) (4) + (47.0) (1) + (220.6) (6) - (67.6) (8)
- (135.3)(2) «
+ 1,611.8 - 1,611.4 =
GRAPHIC SOLUTION. The resultaut of the forces Fa and Fs must pass
through point A and will be designated Ra^. The four forces Fi, F2, F3,
and Rab are in equilibrium. The magnitude of F3 and both the magnitude
and direction of Rab can be found graphically as shown in Fig. 2-21.
Ra5 niay then be resolved into two components parallel to Fa and F5,
20 ELEMENTARY THEORY OF STRUCTURES
^2= 200 lb

b c
"
1 1 1 1 1 1

a Fi = 100 Ib^ 1 1 1 1
1

-r-t-T--t-
1

6 2^^-^^ja'

o6\
1

- r/^^l-

V;i_l__[r^r|r-
- h- -H
1 1

^
^3'
1/

/•
'^ 1 1

=
/oc
1
1/

\
1

F^
\ 1

Ra-o
Fz = 59 lb
1 division 1 ft
F4 = 221 lb
Scale. F5= 1511b
1 in. = 5 ft

(a) Space diagram (6) Force polygon

Fig. 2-21

The complete solution is shown in Fig. 2-21 and requires no detailed

explanation.
2-5. Reactions on a Three-hinged Arch. A common problem in struc-
tural analysis is to find the hinge reactions on a three-hinged arch by
algebraic or graphic methods. A three-hinged arch is a structure com-
posed of two curved segments, joined together by an internal hinge and
supported at two external hinges as shown in Fig. 2-22a. It is required
to find the horizontal and vertical components of the hinge reactions at
A, B, and C in this structure.

ALGEBRAIC METHOD. When the supports A and C are at the same

level, Va and Vc can be solved from XMc = and 2M^ = 0, respec-
tively, by taking the whole arch as a free body. Ha or He may be found
by taking moments about B, using either ^J5 or BC as free body.
If the supports A and C are not on the same level, it is convenient to

find Hb and Vb first. SM^ = by taking the left segment AB as free

body, and XMc = by taking the right segment BC as free body, will
furnish the two simultaneous equations from which Hb and Vb can be
solved.
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 21

GRAPHIC METHOD. The graphic method of finding the reactions on a

three-hinged arch requires that a funicular (or string) polygon be passed
through three given points. The general method will first be described
and then proof will be given for the construction. The three-hinged arch
in Fig. 2-23a is loaded as shown. The loads on the arch are plotted in
the force polygon of Fig. 2-236 as ABODE. It is required to find the
location of the pole such that the funicular (equilibrium) polygon
X-1-2-3-4-Z in Fig. 2-23a (as will be found later) will pass through the
three given points X, Y, and Z, which, in the present case, are the hinges
of the arch. A random pole 0' is arbitrarily chosen and the rays O'A,
O'B, etc., are drawn. Lines mn and pq are drawn through and Y and X
parallel to AC; lines st and uv are drawn through Y and Z and parallel
to CE. Starting at any point X' on mn, a trial funicular polygon
X'-V-2'-3'-4:'-Z' is drawn, with the strings respectively parallel to the

(a) Space diagram (6) Force polygon

Fig. 2-23

rays radiating from the trial pole 0'. The string 2^-3' (o'c) intersects pq
and St at Y' and Y'\ Join X'Y' and Y''Z\ Draw O'F parallel to X'Y'
and OV parallel to Y''Z\ Point 0, the pole being sought, is the point
of intersection of a line drawn parallel to XF at F and another line
drawn parallel to YZ at G. The rays OA, OB, etc., are then drawn.
The final funicular polygon X-1-2-3-4-Z will pass through the three given
points X, Y, and Z.
It will be shown that EO and OA are the external hinge reactions at
Z and X; and that CO and 00 are the internal hinge reaction acting on
the left and right segments, respectively.
PROOF. Suppose that the loads on the left segment are supported
by two through the points
parallel reactions X
and Y. These reactions
are FA and CF. Point F has been found by drawing O'F parallel to the
closing line X'Y'. Since point F must always take the same position
regardless of the location of the pole, the required pole must he on a line
22 ELEMENTARY THEORY OF STRUCTURES
drsLvm through F and parallel to XY. Any funicular polygon, with a
pole on FO, that passes through X, will also pass through Y. The same
reasoning applies to the right segment, and any string polygon with a
pole on GO will pass throughY and Z. Thus the required pole is the
point of intersection of FO and GO. The fact that OA and CO must
be
the hinge reactions on the left segment of the arch can be shown as
follows
AB = A0-{^ OB
BC = BO-^ OC
AB-^BC = AO-^OC
Since AO and OC two forces equivalent to all the loads acting
are the
on the left segment, OA
and CO (the opposites of ^0 and OC) must be
the supporting reactions. Similarly it can be shown that EO and OC
are the hinge reactions on the right segment of the arch.
40 k 50 k

20 k

7/9-42 3/^1 k

Ho=523/j^k

Vo=7lViik K = 58Vll k
(a)

40 k 20 k 30 k 50 k

20 k

T
H5=723/j^k

V5=llViik
H5=723/iik

00
Fig. 2-24

The polygon is sometimes known as the pressure line

final funicular
in the arch because each string (such as oh in Fig. 2-23a) is the line of
action of the resultant of all the forces (including the external hinge
reaction) on either side of this string.
Example 2-10. Find the hinge reactions at Lo, Ub, and Lg due to the
loads acting on the three-hinged arch shown in Fig. 2-24a.
ALGEBRAIC SOLUTION. From 2(moments about Lo) = by taking
the left segment as a free body (Fig. 2-246),

30^6 + 5OF5 = (20) (30) + (40) (10) + (20) (30)

3^6 + 5F5 = 160 (a)
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 23

or, from S (moments about L9) = by taking the right segment as a

free body (Fig. 2-24c),

2O//5 - 4OF5 = (30) (30) + (50)(«)(10) + (50)(%)(20)

2i/5 - 4F5 = 190 (b)
Solving (a) and (b),
H, = 72^1 kips
Vs = -llHi kips

From 2Fx = (Fig. 2-246),

i^o = 52^1 kips

From ZFy = (Fig. 2-246),

Vo = 71^1 kips

From SFx = (Fig. 2-24c),

H, = 42^1 kips

From 2Fy = (Fig. 2-24c)

79 = 58Ki kips

CHECK. A check can be made wdth SF^ = 0, XFy = 0, and 2 (moments

about Lo) = by taking the whole arch (Fig. 2-24a) as the free body.
A B

20 k

Scale.
1 in. = 1001c

(a) Space diagram (6) Force polygon

Fig. 2-25

GRAPHIC SOLUTION. The graphic solution is shown in Fig. 2-25. DO

and OA are the hinge reactions acting on the left segment; FO and OD
are the hinge reactions acting on the right segment.

PROBLEMS
2-1. By both the algebraic and graphic methods find the magnitude and direction
of Fa if the four coplanar, concurrent forces Fi, F^, Fz, and Ft are in equilibrium.
2-2. By both the algebraic and graphic methods find the magnitudes of F^ and F4
24 ELEMENTARY THEORY OF STRUCTURES

F2-601b A- 85 lb

__Fi = 601boL''j^J 0\ Fi" 120 lb

0|\ Fi = 1001b X

f2-75 1b
F3 = 90 lb
F2 = 801b
Fig. P2-1 Fig. P2-2 Fig. P2-3

acting in the directions shown if the four coplanar, concurrent forces Fi, F2, F3, and
F4 are in equilibrium.
2-3. By both the algebraic and graphic methods find the magnitude of F3 and
the direction of F4 if the four coplanar, concurrent forces Fi, F2, F3, and Fi are in
equilibrium.
F2-I001b F4

I
°' 1^1
Fl-2001b f3»601b

Fig. P2-4

2-4. By both the algebraic and graphic methods find the magnitude and position of
F4 if the four coplanar, parallel forces Fi, F2, Fz, and F^ are in equilibrium.

120 1b 601b 901b 15 k 5k

'
^ ^ ^
I j}
j^^^ J^^-h wmw I ; ;i;; i; ill

4 24' 4 1 18' t
4 f
^
Fig. P2-5 Fig. P2-6

2-5 and 2-6. By both the algebraic and graphic methods find the reactions Va and
Vb on the beam subjected to the loads as shown.

Fx " 100 lb

15 k
(-2,2)
F2 = 80 lb

\
60'
^(3,1)
.0 1
6k

25'
(-5,-2)

F3 - 120
/ lb
30 ^ "24 k
6 at 20' = 120'

Fig. P2-7 Fig. P2-8

2-7. By both the algebraic and graphic methods find the magnitude, direction, and
position of F^ if the four coplanar forces Fi, F2, F3, and F4 are in equilibrium.
 EQUILIBRIUM OF COPLANAR-FORCE SYSTEMS 25

By both the algebraic and graphic methods find the magnitude of Ra at the
2-8.
rollersupport and the magnitude and direction of Rb at the hinge support of the truss
subjected to loads as shown.

^5

\ /
^3
I

_L
Fi = 390 lb'
f-l-t-h

1 division = ifi T
F2 = 200 lb

Fig. P2-9

2-9. By both the algebraic and graphic methods find the magnitudes of Fz, F4, and
Fs if the five coplanar forces Fi to F^, inclusive, are in equilibrium.

24 k 18 k
1

12 k

8' 8'
6 k 3 k

8'
6 k
....^^^^^
r=:^
• '1* '

J
^ B-^ '

12' y 25'

?A
48'
^'t^
at 10' =
^ ^ "

Fig. P2-10 Fig. P2-11

2-10 and 2-11. By both the algebraic and graphic methods find the horizontal and
vertical components of each of the three hinge reactions.
 CHAPTER 3

SHEARS AND BENDING MOMENTS IN BEAMS

3-1. Definition of Shears and Bending Moments. A beam has been

defined as a structural member subjected to transverse loads only. Take,
for example, the beam sho^\Ti
in Fig. 3-1 a. The reactions Ra and Rb
can be found from the principles of equilibrium of a coplanar-parallel-
force sj^stem. The problem now is to investigate the internal forces
(shears and bending moments) within the beam. Suppose that the beam
is cut into the two segments by a right section at a distance x from A,

I i
^-^
i ; ; i i i i i
Pi

i 1 i
k.'ft
i i ; i i i
P.

It:
(«)
4
Pi

i' 'i X^V;, M, ^ k

M f
\ k \ \ k
I

r-A-c c I I I I I I I I w mz
T T^\---i
L-x
«Ar~r ib) (d)
\Rb
ZF
ZM W. M ZM Note.
V^ and M^ are shown in
their positive directions.

Vx ZF
(c) (e)

Fig. 3-1

and each segment is shown in Fig. 3-16 and d.

treated as a free body as
Assume the action of the segment on the right segment to be a shear-
left

ing force Vx upward and a bending moment Mx clockwise as shown in

Fig. 3-lc?. Consistent ^^ith the above assumption, the action of the right
segment on the left segment must be a shearing force Vx do^^'nward and
a bending moment Mx counterclock^^^ise, as sho\\Ti in Fig. 3-16. In this
text, these directions will hereafter be considered as those of positive
26
 SHEARS AND BENDING MOMENTS IN BEAMS 27

shear and positive bending moment. In other words, positive shear tends
to rotate an element of the beam dx in length in the clockwise direction,
and moment tends to compress the upper fibers of the
positive bending
element. (Check these definitions on both Fig. 3-lc and e.) Vx and Mx
as above defined are known as the shear and bending moment on a sec-
tion at a distance x from A .

By considering either the left or the right segment as the free body,
Vx and Mx can be found from the two equations of equilibrium S/^ =
and SM = 0. The values as determined by either method should agree
not only in magnitude but also in sign. Thus, from Fig. 3-16,

ZF = 0: Vx = Ra - Pi- wx (3-1)

SM = 0: Mx = Rax - Pi{x - a) - ^^^ (3-2)

or, from Fig. S-ld,

XF = 0: Vx = P2 + P, + w(L - X -h d) - Rb (3-3)
SM = 0: Mx = Rb{L - x) - P2(L -x - c) - P,(L - x^d)
-

w(L - ^ + dr (3^)

Thus, shear at a section is equal to the summation of all the upward

forces minus the downward forces to the left of the section, or it is equal
to the summation of all downward forces minus the upward forces on the
right of the section. Bending moment at a section is equal to the summa-
tion of all clockwise moments minus the counterclockwise moments of all
forces to the left of the section about the sec- i i * ^ * * i i 4 ; i-T-i
tion, or it is equal to the summation of all
counterclockwise moments minus the clock- t t
wise moments of all forces to the right of the
section about the section. (The moments
of upward forces, regardless of which side of
the section the forces are on, are always pos-
itive according to the above definitions.) Positive shear
If the definitions of shear and bending
moment as described in the preceding para-
graph are strictly followed, not only the nu-
merical magnitude, but also the sign, of the
"""^
Negative shear
shear or the bending moment at any section
Fig. 3-2
in a beam can be determined. Positive
shear and positive bending moment can be defined in another way on a
more physical basis. Shear at a section is said to be positive if, by imag-
ining the beam to be cut into two parts at the section, the left segment
tends to slide upward relative to the right segment (see Fig. 3-2) Bend- .

ing moment at a section is positive if the beam bends with the concave
28 ELEMENTARY THEORY OF STRUCTURES
side on the top (see Fig. 3-3). The follo^^ing examples will further em-
phasize the signs of shears and bending moments in beams.
Example 3-1. Determine the magnitudes and signs of the shear and
bending moment in the beam shown in Fig. 3-4 at a section whose dis-

tance from A is (a) 2 ft, (b) 13 ft, and (c) 22 ft.

16 k 30 k 10 k

t
.

Z I .
I

>^ i i
I
I I w ; i
I
^ ;
3k/ft
4 I i
I

4 I ;
I
'r-
20'
U-
*
Positive bending i

*\*
Negative bending
> i?^= 50.1k %= 83.9 k
Fig. 3-3 Fig. 3-4

SOLUTION. From 2J/5 = 0,

20Ra + (10) (4) = (78) (7) + (30) (8) + (16) (16)

Ra = 50.1 kips
From 1:Ma = 0,

20Rb = (78) (13) + (16) (4) + (30) (12) + (10) (24)

Rb = 83.9 kips
CHECK. By 2F = 0,

50.1 + 83.9 « 78 4- 16 + 30 + 10
134 = 134

(a) Fx and M^, when x = 2 it from A, are

Fx(left) = 50.1 - (3) (2) = +44.1 kips
Fx(right) = 16 + 30 + 10 + (3)(24) - 83.9 = +44.1 kips

M.(left) = (50.1) (2) - ^^^

2
= +94.2 kip-ft

- (3)(24)2
M.(right) = (83.9) (18) (16) (2) - (30) (10) - (10) (22)

= +94.2 kip-ft

(6) Vx and M^, when x = 13 ft from A, are

Fx(left) = 50.1 - (3)(13) - 16 - 30 = -34.9 kips

Fx(right) = 10 + (3)(13) -83.9 = -34.9 kips

ilf.(left) = (50.1)(13) - (16)(9) - (30)(1) - i^KM! = +223.8 kip-ft

iV/.(right) = (83.9) (7) - (10) (11) - ^^^1^ = +223.8 kip-ft

 SHEARS AND BENDING MOMENTS IN BEAMS 29

(c) Vx and Mx, when x = 22 it from A, are

F.(left) = 50.1 + 83.9 - 16 - 30 - (3) (22) = +22.0 kips
F.(right) = 10 + (3) (4) = +22.0 kips
(3) (22)^
Mx(left) = (50.1)(22) + (83.9)(2) - (16)(18) - (30)(10)
= -44.0kip-ft

M.(right) = -(10) (2) - (3) (4)^

-44.0kip-ft

Example 3-2. Determine the magnitudes and signs of the shear and
bending moment in the beam shown in Fig. 3-5 at sections (a) 7 ft and
(6) 17 ft from A.
SOLUTION. From XF = 0,

F^ ="6 + 9 + 24(0.8) = 34.2 kips

From XMb = 0,

Mb = (6) (20) + (9) (10) + H (0.8) (24) 2 = 440.4 kip-ft

3k/ft
6k 9k
4' 10' 10' 1.75 k/ft
I ^
^
^' 0.8k/ft i ^ -\ f
aE i I i i ^ i I i i i i >: ^ o
- 'p i
V^= 34.2 k Rb" I2k
Fig. 3-5 Fig. 3-6

CHECK. By 2AfA = 0,

(6) (4) + (9) (14) + H (0.8) (24) + 2 440.4 (34.2) (24)

820.8 = 820.8

(a) Vx and Mx, when a; = 7 ft from A, are

= -6 - (0.8) (7) = -11.6 kips
Fx(left)
= +9 + (0.8) (17) - 34.2 = -11.6 kips
Fx(right)
M.(left) = -(6) (3) - M (0.8) (7) = -37.6 kip-ft 2

Mx(right) = +(34.2) (17) - 440.4 - (9) (7) - K(0.8)(17)2

= -37.6 kip-ft
(b) Vx and Mx, when x = 17 ft from A, are

7x(left) = -6 - 9 - (0.8) (17) = -28.6 kips

7x(right) = -34.2 + (0.8) (7) = -28.6 kips
M,(left) = -(6)(13) - (9)(3) - M(0.8)(17)2 = -220.6 kip-ft
M,(right) = -440.4 + (34.2)(7) - M(0.8)(7)2 = -220.6 kip-ft
Example 3-3. Determine the magnitude and sign of the shear and
bending moment in the beam shown in Fig. 3-6 at a section 10 ft to the
right of A.
30 ELEMENTARY THEORY OF STRUCTURES
SOLUTION
Total load on the beam = 3^ (3) (24) = 36 kips

The center of gravity of the total load is at the third point of the span;
thus,
Ra = 24 kips Rb = 12 kips

Vx and Mx, when a: = 10 ft from A, are

F.(left) = 24 - K(3 + 1.75) (10) = +0.25 kip

y,(right) = -12 + 3^(1.75) (14) = -hO.25 kip
ilf.(left) = (24)(10) - K(3)(10)(2%) - K(1.75)(10)(i%)
= +110.83 kip-ft
Mx(right) = (12)(14) - M(1.75)(14)(i^) = +110.83 kip-ft
3-2. Relationship between Load, Shear, and Bending Moment. The
relationship between load, shear, and bending moment in a beam can be
expressed in the following two propositions:
Proposition 1 The rate of decrease of shear with respect to x, on any
.

section at a distance x from the left end of the beam, is equal to the
intensity of load at the section.
Proposition 2. The rate of increase of bending moment with respect
to x, on any section at a distance x from the left end of the beam, is equal
to the shear at the section.

A
i V.^dV,

dx

(6)

Fig. 3-7

PROOF. Let Vx and Mx be the shear and bending moment on a sec-

tion at a distance x from A of the beam shown in Fig. 3-7a. Likewise
Vx + dVx and Mx + dMx are theshear and bending moment on a section
at a distance x dx from A.
-\- The free-body diagram of the segment
of the beam which is dx in length is shown in Fig. 3-76. Note that the
load {Wx in intensity), shears, and bending moments are all shown in their
positive directions.
From XFy = (Fig. 3-76),

{Vx + dVx) - Wx dx
dVx = —Wx dx
dVx — w. (3-5)
dx
 SHEARS AND BENDING MOMENTS IN BEAMS 31

From -ZMb = (Fig. 3-76),

dx
V, rfx + Mx = Mx + dM, + w, dx

The term }/2'^x{dxy approaches zero as a limit when compared with the
other terms and can therefore be ignored. Simplifying,

dM. dx
dM,
(3-6)
dx
Equations (3-5) and symbolic expressions of Propositions 1 and
(3-6) are
2 stated above. These relations will be further exemplified in the
subsequent articles.
3-3. Shear and Bending-moment Equations. The shear and bending
moment Vx and ilf x on a section at a distance x from any arbitrary
reference point on the beam can be
16 k
expressed in terms of x. Such ex-
pressions are called shear and bend-
i i i U i ^ I i I i i 'i i i i i I i i

ing-moment equations. Any set of t C ^^,

D B EF
shear and bending-moment equa- ^
7?^= 50.1k Rb= 83.9k
tions, however, is applicable only (a)

to a definite segment of the beam. 16 k 30 k lOk

4' 2'

This segment is usually between two

I

k/ft|
3 k/ft 3 k/ft
till
I

I i I I TT-l
adjacent concentrated loads or re-
20'
B E F
actions. The reference point, from
t^
which X measured, may be either
is 50.1k 83.9 k
(6)
inside or outside of the segment. 16 k 30 k 10k
Example 3-4 Determine the shear
.

and bending-moment equations for

3 k/ft
I '
I I tf—t I
t
;' i i
3 k/ft
I I I
f
n f

segment CB of the beam shown in

'''}' B E F
20-

Fig. 3-8a using A, B, and C as the

A
50.1k 83.9 k
(c)
reference points, respectively.
16 k 30 k 10 k
SOLUTION. From XMb = and 4' 12'

I^Ma =
be 50.1 kips
0,
and
Ra and
83.9
Rb are
kips,
found
respec-
to
i
3 k/ft
i t 3 k/ft
I

D
I t

B E
aH
tively (see also Example 3-1). t
16-x ?
(a) Vx and Mx in segment CD in
50.1k 83.9 k
{d)

terms of x measured from A (Fig. Fig. 3-8

3-86). Identical expressions for Vx
and ilf X should be obtained whether the left or the right side of the beam
is considered. Although one seems to duplicate the other, both solutions
will be shown for purpose of clarification.

Left Vx = 50.1 - 16 - 3a; = 34.1 - Sx

Mx = 50.1a; - 16(a; - 4) - %x' = 64 + 34.1a; - 1.5a;2
32 ELEMENTARY THEORY OF STRUCTURES
Right Fx= 30 10 -h 3(26 - x)
+ - 83.9 = 34.1 - 3a:
M, = 83.9(20 - x) - 30(12 - x) - 10(24 - x) - %(26 - xY
= 64 + 34.1a; - l.Sa;^

Propositions 1 and 2 in Art. 3-2 can be illustrated by noting that the

derivative of the expression for Vx = 34.1 — 3a; with respect to a; is —3
kips per ft, or —Wx, and the derivative of

Mx = 64 + 34.1a; - l.Sa;^

is dMx/dx = 7^ = 34.1 - 3a;.

(b) Vx and Mx in segment CD in terms of x measured from B (Fig. 3-8c)

Left Vx = 50.1 - 16 - 3(20 - = 3a; - 25.9 a;)

Mx = 50.1(20 - x) - 16(16 - x) - %(20 - xY

= 146 + 25.9a; - l.Sa;^
Right Fx = 30 + 10 + 3(a; + 6) - 83.9 = 3a; - 25.9
Mx = 83.9a; - 30(a; - 8) - 10(a; + 4) - %(a; + 6)^
= 146 + 25.9a; - 1.5a;2
Note that, when x is increasing from right toward left, Eqs. (3-5) and
(3-6) in Art. 3-2 become

S = +-' (3-7)

and ^=-V^ (3-8)

In proving Eqs. (3-7) and (3-8), as shown in Fig. 3-76, it is necessary to

write Vx + dVx and Mx + dMx on the left face and Vx and Mx on the
right face of the element.
Thus, in the present case, the derivative of Fi = 3a; — 25.9 is +3, or
+Wx; the derivative of Mx = 146 -f- 25.9a; - 1.5a;2is25.9 - 3a;, or -Vx.
(c) Vx and Mx in CD in terms of x measured from C (Fig. 3-Sd) .

Left Vx = 50.1 - 16 - 3(a; -f- 4) = 22.1 - 3a;

Mx = 50.1(a; + 4) - 16a; - K(a; H- 4)^

= 176.4 -h 22.1a; - 1.5a;2

Right Fx = 30 +
10 H- 3(22 x) - 83.9 = 22.1 - 3a;
-
Mx = 83.9(16 - - 30(8 - x) - 10(20 - x) - M(22 -
x) x)^
= 176.4 -f- 22.1a; - 1.5a;2

Example 3-5. Determine the shear and bending-moment equations

for segment CD of the beam shown in Fig. 3-9a using A, B, and C as the
reference points, respectively.
SOLUTION. From 2F = and 2Mb = 0, Vb and Mb are found to be
34.2 kips and 440.4 kip-ft, respectively (see also Example 3-2).
(a) Vx and Mx in segment CD in terms of x measured from A (Fig. 3-96)
Left Vx = -6 - 0.8a;
Mx = -6(a; - 4) - M(0.8)a;2 = 24 - 6a; - 0.4x2
 SHEARS AND BENDING MOMENTS IN BEAMS 33

Right 7. = +9 + (0.8) (24 - x) - 34.2 = -6 - 0.8a;

Mx = -440.4 + 34.2(24 - x) - 9(14 x) - M(0.8)
(24 - xy = 24: - Qx - OAx''

6k 9k 6k 9k
4' 10' 10'
I I 10'
'"'
1 1
I 0-8k/ft I \
I 0.8k/ft f
24' / «3 ;c 24 -X /*^ "^

V^ = 34.2 k 34.2 k
(a) (&)

6k 9k
6k 9k
10'
10'

I 0.8k/ft
0.8 k/ft r 'I 0.8 k/ft f
\
aM
I

Mt CJ.
c ^^t
W^T
^ * i \ t i

20
i

-X
i i

24-

34 2k 34.2 k
(c) (d)

Fig. 3-9

(6) Fx and Mx in segment CD in terms of x measured from B (Fig. 3-9c)

Left F^ = -6 - 0.8(24 - x) = -25.2 +0.8x

Mx = -6(20 - x) - M (0.8) (24 - x)^
= -350.4 + 25.2X - 0.4a;2
Right 7, = +9 + 0.8a; - 34.2 = -25.2 + 0.8a:
Mx = -440.4 + 34.2a; - 9(a: - 10) - H(0.8)a;2
= -350.4 + 25.2a; - 0.4a;2
(c) Vx and Mx in segment CD in terms of x measured from C (Fig. 3-9d) .

Left Vx= -6 - 0.8(a; + 4) = -9.2 - 0.8a;

Mx = -6a; - M(0.8)(a; + 4)^ = -6.4 - 9.2a; - 0.4a;2
Right Vx = +9 + 0.8(20 - x) - 34.2 = -9.2 - 0.8a;
Mx = -440.4 + 34.2(20 - x) - 9(10 - x) - )^(0.8)(20 - x)^
= -6.4 - 9.2a; - 0.4a;2
a k/ft a k/ft

(3-V8)k/ft

i?4=24k i?B=12k /?4=24k i?3=12k

(a) (6)

Fig. 3-10

Example Determine the shear and bending-moment equations

3-6.
for the beam shown in Fig. 3-10 using A and B as the reference points,
respectively.
34 ELEMENTARY THEORY OF STRUCTURES
SOLUTION, (a) Vx and Mx in the beam in terms of x measured from A
(Fig. 3-lOa). The intensity of load Wx at a section distant x from A is

Wx =
3(24
24
-
— x)
= ^
^
..
- H^ ,.
l^PS per ft

Using the left free body,

Vx= 24.- H(3)a: - K(3 - 3^x)x = 24 - 3a: + Viex'

Mx = 24a; - K(3x)(%a:) - i^a:(3 - >ia:)(Ma:)
= 24a; - %a;2 + Ms^'
Using the right free body,

Vx= -12 + M(24 - - ^x) = 24 - 3a; + Ke^'

a;)(3

M. = 12(24 -X) - M(24 - x){3 - Vsx) ^^^^

= 24a; - %a;2 + i^g^'

Note that the relations dVx/dx = —Wx and dMx/dx = Vx are satisfied
by the above expressions for Wx, Vx, and Mx.
(h) Vx and Mx in the beam in terms of x measured from B (Fig. 3-106).

The intensity of load Wx at a section distant x from B is

3a; 1 ,

'^x = -^ = ^x .

kips per ft
.^

Using the left free body,

7. = 24 - H(3)(24 -x) - M(J^x)(24 - x)

= -12 + Mex'^
Mx = 24(24 -x) - M(3)(24 - a;)[%(24 - x)]
- M(M^)(24 - a;)[>^(24 - x)]
= 12a; - }i8x'

Using the right free body,

F. = -12 + M^(i^x) = -12 + Hex'

Mx = 12a; - M^(i^a;)(>^a;) = 12x - }48x'

In this case, x is increasing from right toward left; so the above expres-
sions for Wx, Vx, and Mx should satisfy the relations dVx/dx = -{-Wx and
dMx/dx = - Vx.
3-4. Shear and Bending-moment Diagrams. Shear and bending-
moment diagrams are the graphs showing the variations along the length
of the beam in the values of the shear and bending moment due to a fixed
loading condition. The shear and bending-moment curves or diagrams
can be plotted from the shear and bending-moment equations. Since
any set of shear and bending-moment equations is valid only mthin a
 SHEARS AND BENDING MOMENTS IN BEAMS 35

definite range of values of x, they can be used only to plot the shear and
bending-moment curves segment of the beam to which the equa-
for the
tions apply. If the beam is subjected to concentrated loads only, the

shear curves are horizontal and straight and the bending-moment curves
are sloping and linear; but if the beam is subjected to both concentrated
and uniformly distributed loads, the shear curves are linear and the
bending-moment curves are parabolic.
Shear and bending-moment diagrams can usually be drawn without
actually deriving the shear and bending-moment equations. Since the
shear at any section is equal to the summation of all upward forces minus
the downward forces to the left of the sec-
tion, the shear ordinate can be obtained by
summing all the forces to the left of the sec-
tion. Thus the procedure is that of ''step-
ping up and down with the forces from the
left of the beam."
end to the right end In-
asmuch summation of all forces act-
as the
ing on the beam must equal zero, the shear
diagram must close.
The bending-moment ordinate at any sec-
tion may be found by summing up the areas
of the shear diagram to the left of the sec-
Fig. 3-11
tion. This is true because it can be shown
that the change of moment between any two sections is equal to the area
of the shear diagram between these two sections.
Thus, referring to Fig. 3-11 and Eq. (3-6),

dM.
= V.
dx
from which dMx = Vx dx
Integrating between sections A and B,

dx

or Mb- Ma= r^
J ''a
Vx dx (3-9)

With the known value of the bending moment at the left end of the
beam, and by use of Eq. (3-9), the bending moment at all critical sections
of the beam can be computed and plotted to obtain the bending-moment
diagram. A check is always available because the bending moment at the
left end plus the total area of the shear diagram is equal to the bending

moment at the right end, or in other words, the bending-moment diagram

must close. It will be noted that the bending moments and shear areas
36 ELEMENTARY THEORY OF STRUCTURES
all carry signs, i.e., the bending moment at a section is positive if the
upper part of the section is in compression, and the shear area is positive
if it is above the horizontal reference line.

Since the derivative of the bending moment at a section with respect to

X is equal to the shear at the same section (or the slope of the tangent
to the bending-moment curve at any point is equal to the ordinate to
it is obvious that the bending moment
the shear curve at the same point),
is maximum where the shear equals zero. Such a section is
at a section
sometimes called the dangerous section, but probably the term ''section
16k 30k 10 k
4' 2'
I

^
3 k/ft I i I

^ ^ ^ * ^ * i i ^ i i i ^ i i i i I ;
1;^ ^

t 20'

i?^= 50.1k
A
i?B= 83.9 k
(a)

+ 50.1
^--^^38.1
(+176.40)- + 28 /^««°°'
^0.63' Bz I^-V +16

i(4 8L40)"^^«^ fe(-0.60) ^ ^2 +6^

C 7.37'
-1.9
^ 1
B e\f
(-351.20) ^(+6.00)
'J
D2
-31
5^ Bi
-55 .9
(6) Shear diagram

1 y^ ^^"^ >+
80
257.20

+ 176.40^<
'-3 ^\ 1-1.76'

\^
3E •1^
P B E F
A
C M D
m=6.23'
\Vb3 /^fi"•"
f>n

-94.00
(c) Bending moment diagram
Fig 3-12

ofmaximum bending moment" is more significant. The tangent to the

bending-moment curve at the section of maximum bending moment is
horizontal.
At a section where a concentrated load acts, there is an abrupt drop in
shear and thus an abrupt change in the slope of the tangent to the bend-
ing-moment curve. The bending-moment curve is therefore discontinu-
ous at such a section, or the tangent to the left branch of the bending-
moment curve is not collinear with that to the right branch.
Example 3-7. Draw shear and bending-moment diagrams for the
beam shown in Fig. 3-l2a.
 SHEARS AND BENDING MOMENTS IN BEAMS 37

SOLUTION. From XMb = and ^Ma = 0, Ra and Rb are found to be

50.1 and 83.9 kips, respectively (see Example 3-1).
The shear diagram is plotted as AA1C1C2D1D2B1B2E1E2F, as shown in
Fig. 3-125. AAi = 50.1 kips upward. AiCi is a straight line which
slopes down at the rate of 3 kips per ft. Thus CCi = 50.1 — 3(4) =
38.1 kips. C1C2 = load at C = 16 kips. Cd = 38.1 - 16 = 22.1 kips.
DDi = 22.1 - 3(8) = -1.9 kips. DD2 = -1.9 - 30 = -31.9 kips.
BBi = -31.9 - 3(8) = -55.9 kips. BB2 = -55.9 + 83.9 = +28 kips.
EEi = +28 - 3(4) = 16 kips. EE2 = 16 - 10 = 6 kips. The shear
at F is found to be 6 - 2(3) = 0.
The shear diagram shows that there are two values of shear at C;
+38.1 and +22.1 kips. In fact, the +38.1 kips is the value
viz., of the
shear at a section which is at an infinitesimal distance to the left of C
where the load of 16 -kips acts, and +22.1 kips is the value of shear
at a section which is at an infinitesimal distance to the right of C.
The location of point M, the section of maximum bending moment,
may be found as follows:

CM = ^ = ^^ = 7.37 ft

MD = ^' = i-9 = 0.63 ft

Wx 3
and CM + MD = 7.37 + 0.63 = 8 ft (check)

The shear areas are computed and entered in the shear diagram. Thus

AAiCiC = +M(50.1 + = +176.40 kip-ft

38.1) (4)
CC2M = +K(22.1)(7.37) = +81.40 kip-ft
MDDi = -3^ (1.9) (0.63) = -0.60 kip-ft
DD2B1B = -M(31.9 + 55.9)(8) = -351.20 kip-ft
BB2E1E = +M(28 + 16) (4) = +88.00 kip-ft
E2EF = +M(6)(2) = +6.00 kip-ft
The total area of the shear diagram is found to be zero. This verifies
the statement that the '^bending moment at the left end plus the total
shear area is equal to the bending moment at the right end.'' Note
that in this problem the bending moment is zero at both ends.
The bending-moment diagram is plotted as AC^MzDiB^EzF in Fig.
3-12c.
Ma =
Mc = Ma + = + 176.40 = +176.40
176.40
Mm = Mc + = 176.40 + 81.40 = +257.80
81.40
Md = 257.80 - 0.60 = +257.20
Mb = 257.20 - 351.20 = -94.00
Me = -94.00 + 88.00 = -6.00
Mf = -6.00 + 6.00 = {check)
38 ELEMENTARY THEORY OF STRUCTURES
Upon determination of these controlling values a smooth curve may be
drawn through the plotted points A, C3, M^, D3, Bz, Ez, and F. The
way in which the curvature occurs can best be visualized by inspecting
the shear diagram. For instance, the shear curve AiCi is linearly down-
ward, thus indicating that the slope to the moment curve ACz is con-
stantly decreasing from A to Cz. On the moment curve ACz^ the slope
at A is 50.1 kips, the slope at Cz is 38.1 kips, and the slope at other

intermediate points equal to the value of the shear ordinate at that

is

point. From this consideration, it is apparent that the moment curve

ACz is it is convex at the top and concave at the bottom.
parabolic and
has been stated that the slope of the tangent to the bending-moment
It
curve at any point is equal to the ordinate to the shear curve at the same
point. At point Cz on the bending-moment curve of Fig. 3-12c, the
CzA is +38.1, and that to curve C3M3 is
slope of the tangent to curve
+22.1; +38.1 and +22.1 are, respectively, the shear at a section just to
the left or right of C where a concentrated load of 16 kips acts.
The point at which bending moment is zero is called the point of
inflection. The distances DP = m or BP = n in Fig. 3-12c locate the
point of inflection P. The distances in and n, the sum of which must be
8 ft for a check, may be found as follows

Mp — Md = shear area between D and P

- 257.20 = -i.im(31.9 + 31.9 + 3m)
1.5??i2 + 31.9m = 257.20
?n2+ 21.27/n + 10.64^ = 171.47 + 10.64^
m + 10.64 = + V284.68 = ± 16.87
m = 16.87 - 10.64 = 6.23 ft
Mb — Mp = shear area between P and B
-94.00 - = -Hn(55.9 - 3/i + 55.9)
1.5n^ - 5o.9n = -94.00
71^ - 37.27/1 + 18.642^-62.67 + 18.64^
n - 18.64 = ± \/284.78 = ± 16.88
n = -16.88 + 18.64 = 1.76 ft
m+ ?i = 6.23 + 1.76 = 7.99 = 8 ft {check)

Example 3-8. Draw shear and bending-moment diagrams for the

beam shown in Fig. 3-1 3a.
SOLUTION. Considering the whole cantilever beam as a free body, from
XFy = (Fig. 3-1 3a),

F^ = 8 + 15 + 4 + 6 = 33 kips upward

From Sil/,! = 0,

Ma = (8)(9) + (15)(19) + (4)(24) + (6)(30)

= 633 kip-ft counterclockwise
 SHEARS AND BENDING MOMENTS IN BEAMS 39

CHECK. By ZMf = 0,

(33) (32) = 633 + (6) (2) + (4) (8) + (15) (13) + (8) (23)
1,056 = 1,056

The shear diagram by summing the forces to the

in Fig. 3-136 is plotted
left of the section concerned.Note that the moment Ma = 633 kip-ft
does not affect the shear diagram. The parts of the shear diagram are
computed to be +297, +250, +50, and +36 kip-ft, respectively. The
total shear area is +297 + 250 + 50 + 36 = +633 kip-ft, which checks
the relation Me - Ma = - (-633) = +633 kip-ft. In other words,
the increase in moment from A to
E is equal to the shear area between 15k
5'
4k
^
6k

A and E. Note that Ma is -633 <n I = 1 " i

'•
i 1
kip-ft because at section A the up- EF\

per fibers are in tension. 1- 33 k

32'

In plotting the moment diagram, (a)

+ 33 33

Ma is -633 kip-ft, 25 25

Mb = -633 + 297 = -336 kip-ft, ( + 297) I

(+250) 10 -10
Mc = -336 + 250 = -86 kip-ft, 1 +6 +6

il/^ = -86 + 50 = -36 kip-ft, B CD

!(+50) (+36)
EF
and Me = -36 + 36 = (check). (b) Shear diagram

Note that the bending-moment dia-

gram is bounded by straight lines
because the shear, or the slope of the
bending-moment curve, is constant
633
between concentrated loads. The (c) Bending moment diagram
positive slope of the moment curve Fig. 3-13
between A and B is constant at the
rate of 33 kip-ft per ft, or 33 kips, which is the constant value of shear
between A and B.
Example 3-9. Draw shear and bending-moment diagrams for the
beam shown in Fig. 3-14a.
SOLUTION. From XF = and SM^ = 0, Vb and Mb are found to
be 34.2 kips and 440.4 kip-ft, respectively (see Example 3-2).
The shear diagram in Fig. 3-146 is plotted by summing the forces
from the left toward the right. Since the summation of forces between
A and B must be zero, the shear diagram must close. The sum of the
shear areas is —6.4 — 132 — 302 = —440.4 kip-ft; thus the decrease in
bending moment from A to 5 is 440.4 kip-ft. The bending-moment
curve is plotted as shown in Fig. 3-14c. Note particularly the breaks in
slope at sections C and D.
40 ELEMENTARY THEORY OF STRUCTURES
Example 3-10. Draw shear and bending-moment diagrams for the
beam shown in Fig. 3-1 5a.
SOLUTION. From 2Mb = and ^Ma = 0, Ra and Rb are found to be
24 and 12 kips, respectively (see Example 3-3).
A freehand sketch of the shear diagram can be drawn as in Fig. 3-156.
First, the shear ordinates at A and B must be +24 and —12. The
slope of the second-degree curve at A is equal to the load intensity at A,
or —3 kips per ft; while the slope at B is zero. The location of the sec-
tion of zero shear may be determined by writing the expression for shear

6k 9k
10' 10'
3k/ft
0.8k/ft
1.732 k/ft
I I ; ; i~r

V^ = 34.2 k

i?A=24k i?B= 12 k
(a)

-12
(6) Shear diagram

+ 110.86

- 440.4

(c) Bending moment diagram (C) Bending moment diagram

Fig. 3-14 Fig. 3-15

at point M in terms of its distance Xm from B, equating this shear to

zero, and then solving for Xm. Thus,

oXm
Vm = 12 y2Xr
24
= SVS = 13.86 ft

In this particular case, the bending moment at section can be more M

conveniently found by using the left or right free bodies rather than the
shear areas. Considering as a free body, AM
Bending moment at M = (24)(10.14) - M(3)(10.14)(%)(10.14)
- >^(1.732)(10.14)(M)(10.14)
= 110.86 kip-ft
 SHEARS AND BENDING MOMENTS IN BEAMS 41

Considering BM as a free body,

Bending moment at ilf = (12)(13.86) - K(1.732)(13.86)(M)(13.86)
= 110.87 kip-ft (check)

A sketch of the bending-moment diagram is shown in Fig. 3-1 5c.

It is interesting to note that, if the total load of 36 kips is uniformly

distributed on a 24-ft span, the maximum bending moment at the mid-

span is K(36)(24) = 108 kip-ft.
3-5. Bending-moment Diagram by the Graphic Method. At times
when the span is long and the loading is complicated, it may be more
convenient to draw the bending-moment diagram by the graphic method,
which will now be explained.
PROOF. Let it be required to find the moment of the coplanar, parallel
forces ah{AB), hc{BC), cd{CD), and de{DE) shown in Fig. 3-16, about

A .

vo6
B' C U
5
P

e'
B X^
C
c ~H — -^^^
i
D
1

a' E Pole distance in kips

b b c c d d
\o a
(a) Space diagrami (6) Force polygon

Fig. 3-l()

point P. A pole is so chosen that its distance from the line ABODE is
a convenient number, such as 10 kips, 50 kips, etc. Note that the pole
distance H, although called a ''distance," is expressed in terms of units of
force, such as pounds or kips. The strings oa, oh, oc, od, and oe are then
drawn in succession. The prolongations of strings oa and oe intersect a
vertical line through P at points a' and e^ The distance a'e' is known
as the vertical intercept V. It will be shown that the moment of forces
AB, BC, CD, and DE, about P is equal to the product of H and F, or

Ml HV (3-10)

Note that H is a, force usually expressed in kips, and F is a distance

commonly measured in feet.
By definition (Fig. 3-16),

Mp = {AB){AT) + {BC){B'P) + {CD){C'P) + {DE){D'P) (3-11)

42 ELEMENTARY THEORY OF STRUCTURES
From similar triangles l-a'-h' and OAB,

-jTp = -fj- (ratio of base to altitude)

or {AB)(AT) = {aV)H (3-12a)

from
Like^N-ise, similar triangles 2-b'-c' and OBC, S-c'-d' and OCD, and
4-d'-e' and ODE,
=
{BC){B'P) {h'c')H (3-12b)
=
{CD){C'P) {c'd')H (3-12c)
{DE){D'P) = {d'e')H (3-12d)

Substituting (3-12) in (3-11),

Mp = {a'h' -h h'c' + c'd' -h c^'e')i^ = (a'e')^ = HV

which is Eq. (3-10).
Thus the momentof a group of vertical forces about any point is
equal to the product of the pole distance and the vertical intercept
between the points of intersection of the first and last strings on a vertical
Une through the moment center.

a b b c c d d e

P
W r

A' B- C 1}

r_of_ 6
1
V od^
a / 2 oc c' '
.
f e
'
6'

a'

(a) Space diagram (6) Force polygon

Fig. 3-17

Xow beam F'E' (Fig. 3-17) carrying loads AB,

consider the simple
BC, CD, and DE. The force polygon and the string polygon are drawn
as shown. It will be shown that to some scale the string polygon 1-2-3-4-
5-6-1 is the bending-moment diagram for the given loads. The bending
moment at P is equal to HV, wherein H is the conveniently chosen pole
distance and V is the vertical intercept within the string polygon under
the point P. By definition of bending moment,

Mp = {FA){F'P) - {AB){A'P) - {BC){BT) (3-13)

From similar triangles 1-f-a' and OF A,

fa' ^ FA
F'P H
or {FA){F'P) = iS'a'W (3-14a)
 SHEARS AND BENDING MOMENTS IN BEAMS 43

Likewise, from similar triangles 2-a'h' and OAB, and Z-b'-c' and OBC,

oTb; AB or {AB){AT) = {a'b')H (3-14b)

A'P H
BC =
or {BC){B'P) {h'c')H (3-14c)
B'P H
Substituting (3-14) in (3-13),

Mp = {fa' - a'h' - h'c')H = U'c')H = HV

Thus it is seen that the funicular, equilibrium, or string polygon repre-
sents to some scale the bending-moment diagram.
It is to be noted that, if a horizontal base line for the bending-moment
diagram is desired, it will be necessary to redraw the string polygon by
relocating the pole on the horizontal line through point F in Fig. 3-176.
Also note that the l^ending-moment diagram shown in Fig. 3-1 7a will be
inverted if the pole is chosen to the left of the load line AE,
8k 12k 6k 15k
b b d d
3' 4'

20'

h Scale. 1"
(a)
= 12'
Space diagram
i

(c) String polygon with pole at O, or bending

moment diagram at 1" = 240 k-ft .

id) String polygon with pole at (7, or bending Scale. l"-20k

moment diagram at 1" = 240 kft (6) Force polygon

Fig. 3-18

Example 3-11. Construct the bending-moment diagram for the beam

shown in Fig. 3- 18a by the graphic method.
SOLUTION. As shown in the space diagram of Fig. 3-18a, the loads
are called ah, he, cd, and de. Likewise the reactions are designated ef and
fa. The force polygon ABODE (Fig. 3-186) is drawn mth a convenient
pole distance of 20 kips. The string polygon 1-2-3-4-5-6-1 is constructed
as shown in Fig. 3-18c and then the closing line 1-6 or of is drawn. The
point F is located on the force polygon by drawing OF parallel to of.
44 ELEMENTARY THEORY OF STRUCTURES

In fact, EF and FA are, respectively, the reactions at the right and left
supports of the beam. Figure 3-18c is the required bending-moment
diagram. Bending-moment ordinates may be measured using a scale of
1 in. = 240 kip-ft, in which 240 kip-ft is the product of the pole distance

(20 kips) and the scale of the space diagram (1 in. = 12 ft). For instance

16 k 30 k 10 k

4' 2f

3k/ft
>> i i i i i i i i i M i i i ^ 1 M3CFM rw
20'
E F

(a)

16 k 30 k 10k
6k 6k 6k
6.k 6k 6k 6k 6k 6k 6k 6k 6k 6k I 6k
alb b\c d\e e\f f\g g\h i\j jU k\l l\m in n\o\p\q

Scale. 1" = 10'

'"
?
(6; Space diagram

oihOUoij

OjO
e.g. Af£,= //V£,= (50k)(5r) = 255kfl
'di String polygon with pole at Oj

02/f2£^

e.g. MifHVp^{50k){5.V)-255kfi 03m

Scale. l" = 40k ie) String polygon with f)oles at O2 and O3, or bending
(c) Force polygon moment diagram at 1" = 500 k-ft

Fig. 3-19

the bending moment at the 12-kip load measured to be 108 kip-ft, or

is

the product of V = 5.4 ft and H = 20 kips. If the bending-moment

diagram \^'ith a horizontal base Une is required, it is necessary to redraw

the string polygon as shown in Fig. 3-18c?, for which the pole 0' {O'F is

horizontal) has been used.

Example 3-12. Construct the bending-moment diagram for the beam
shown in Fig. 3-1 9a by the graphic method.
 SHEARS AND BENDING MOMENTS IN BEAMS 45

SOLUTION. The uniformly distributed load is replaced by a series of

evenly spaced concentrated loads. In this case, 13 loads of 6 kips each
will be used as shown in Fig. 3-196. (The larger the number of equivalent
concentrated loads, the more accurate will be the results.) The loads
are designated from a to ^ in the space diagram and the reactions are
called qr and ra. The force polygon AQ (Fig. 3-19c) is drawn and a pole
is chosen such that the pole distance i7 is a convenient number, which
is 50 kips in this case. The string polygon of Fig. 3-19d is then drawn
and the closing line OiV is found. The point R is located on the force
polygon by drawing OiR parallel to Oir. In fact, QR and RA are, respec-
tively, the reactions at the right and left supports of the beam. Figure
3-19d is the required bending-moment diagram, and the bending moment

8k 15k 4k 6k
9' 10' 5' 6'

T'Tl
I I I
'"
I

32'

Scale. 1"<=20'

(o)

e d d c c [^ ^1 a
H=40k

a e

oa

<ti
Ur^
CO yU
s y Scale. l" = 40k

s=
/
/ oe
(b) Foirce polygon

\ V ^
Scale.
„ =
1" (40) (20) =800k-ft

(c) Bending moment diagram

Fig. 3-20

\ at any point is V in feet and the pole

the product of the vertical intercept
distance H in For instance, the bending moment at D is equal to
kips.
HVd = (50 kips) (5.1 ft) = 255 kip-ft. Note that in this diagram OiV
and OiQ are the base lines from which the vertical intercepts are measured
either upward or downward, a fact which determines whether the bending
moment is positive or negative (this is always true when the pole is on the
left side of the force polygon). If a bending-moment diagram with
it is necessary to use the poles
horizontal base lines (Fig. 3-19e) is desired,
O2 and O3 {O2R and O3Q are horizontal) for the strings 02a to 02m and
Ozp to 03m, respectively (note that 02m and 03m meet at the right support
in Fig. 3-19e). Figure 3-19d or e is the bending-moment diagram with a
scale of 1 in. = 500 kip-ft, wherein 500 is the product of the pole distance
(50 kips) and the scale of the space diagram (1 in. = 10 ft).
46 ELEMENTARY THEORY OF STRUCTURES
It is to be noted that the bending-moment diagram thus plotted is

theoretically exact for the loading of Fig. 3-196, but only approximate
for the loading of Fig. 3-19a.
Example 3-13. Construct the bending-moment diagram for the beam
shown in Fig. 3-20a by the graphic method.
SOLUTION. Starting at the free end (Fig. 3-20c), the loads on the
cantilever beam are designated as ab, The force polygon be, ed, and de.
ABODE in Fig. 3-206 is drawn with pole
40 kips horizontally at // =
to the right of AE. This is done for convenience so that the string
polygon will look like the familiar bending-moment diagram. Had
the loads been designated ab, be, ed, and de starting from the left, the
pole would be located horizontally to the left of the low^er end of the
force polygon ABODE. The string polygon is drawn as shown in
Fig. 3-20c. This is the bending-moment diagram required with a
scale of 1 in. equal to H times the scale of the space diagram, or

1 in. = (40 kips) (20 ft) = 800 kip-ft

With this scale, the moment at the fixed end is measured to be 635 kip-ft,
which checks with the result of Example 3-8.

PROBLEMS
3-1. Determine the magnitudes and signs of the shear and bending moment in the
beam shown at a section whose distance from A is (a) 2 ft, (6) 10 ft, and (c) 15 ft.
as
Check each result by using both the left and right sides of the section as free bodies.

10 k 4k
4' 8' 6'

/ 3k/ft
/
i 1 1 . i 1 i i i 1 1

^ /
^ 18' ^ 1

'1
/

'Fig. P3-1

3-2.Determine the magnitudes and signs of the shear and bending moment in the
beam shown at a section whose distance from A is (a) 5 ft, (6) 10 ft, and (c) 20 ft.
as
Check each result by using both the left and right sides of the section as free bodies.

8k 30 k 20 k

3 k/ft

^^ i i
I

i i i i i W i

25'

Fig. P3-2

3-3. Determine the magnitude and sign of the shear and bending moment in the
beam as shown at a section 8 ft to the right of A.
 SHEARS AND BENDING MOMENTS IN BEAMS 47
2.4 k/ft

0.6k/ft

sA-A

Fig. P3-3

3-4. Determine the shear and bending-moment equations for segment CD of the
beam A and B as the reference points, respectively. Check each
of Prob. 3-1 using
equation by using both the left and right sides of the section as free bodies.
3-5. Determine the shear and bending-moment equations for segment DE of the
beam of Prob. 3-2 using A, B, and C as the reference points, respectively. Check
each equation by using both the left and right sides of the section as free bodies.
3-6. Determine the shear and bending-moment equations for the beam of Prob.
3-3 using A and B as the reference points, respectively.

5k 8k 3k
A WK/fl
wlc/ft
'
L 'A \ \ \ r I i-'l^ 'i

T Fig. P3-9
18'

Fig. P3-7 Fig. P3-8

u;k/ft
t^
^ \
V
Fig. P3-10 Fig. P3-11

8k 10k 14k ^12k^

P P IP F T ^

1/3 1 L/3 J L/3 ^L/4|L/4|L/4|L/4^

•Pi^
L '^
V A
Fig. P3-12 Fig. P3-13 Fig. P3-14

3-7 to 3-14. Draw shear and bending-moment diagrams for the

beams shown.
3-15. Draw beam of Prob. 3-1.
shear and bending-moment diagrams for the
3-16. Draw shear and bending-moment diagrams for the beam of Prob. 3-2.
3-17. Draw shear and bending-moment diagrams for the beam of Prob. 3-3.
3-18. Construct the bending-moment diagram for the beam of Prob. 3-9 by the
graphic method.
3-19. Construct the bending-moment diagram for the beam of Prob. 3-1 by the
graphic method.
3-20. Construct the bending-moment diagram for the beam of Prob. 3-2 by the
graphic method.
 CHAPTER 4

ANALYSIS OF STATICALLY DETERMINATE

RIGID FRAMES AND COMPOSITE STRUCTURES

4-1. Analysis of Statically Determinate Rigid Frames.^ A rigid frame

is a structure composed of members which are connected by relatively
stiff or rigid joints. In steel structures welded or riveted joints may be
designed as rigid joints, although beam-to-column connections are often
designed as semirigid or flexible, whether they are welded or riveted.
In reinforced-concrete structures monolithic construction is used, and
the concrete in the members meeting at a joint can be poured usually
at one time, thus forming a rigid joint. A rigid joint may exert a restrain-
ing moment at the end of a member, as distinguished from a pin-con-
nected joint which offers no such resistance. Two quadrangular frames

3k

3k

(c) Stable structure (6) Unstable structure

are shown in Fig. 4-1. One of these has rigid connections (Fig. 4-la)
at joints B and C, and the other has pinned connections (Fig. 4-16) at
these two joints. It is obvious that the structure shown in Fig. 4-la is
stable, while that in Fig. 4-16 is unstable. If the principles of static
equihbrium are applied to the whole frame ABCD, the external reaction
components at A and D for the two frames are identical; however, for
equilibrium ofmember AB, the joint B must exert a thrust of 5 kips
downward and a restraining moment of 6 kip-ft counterclockwise on
member BA (see Fig. 4-2a). The restraining moment of 6 kip-ft counter-
clockwise on member BA and of 6 kip-ft clockwise on member BC can
^ It
is more common for rigid frames to be statically indeterminate. For the differ-
ence between statically determinate and indeterminate rigid frames, see Chap. 13.
48
 DETERMINATE RIGID FRAMES AND COMPOSITE STRUCTURES 49

be developed by the rigid joint at B in Fig. 4-la, but not by the pin-
connected joint in Fig. 4-16. The free-body diagrams of joint B and
member BC are shown in Fig. 4-26 and c. Joint C must be regarded
as a rigid joint, although it happens in this case that there is no transverse
load acting on member CD and consequently no moment is required of
joint C. The free-body diagrams of joint C and of member CD are
shown in Fig. 4-2d and e. Thus one way of finding out whether a struc-
ture is stable or not is to draw the free-body diagrams of each member or
joint and see that the moment-resisting joints are rigid.
In the analysis of rigid frames, it is first necessary to find the external
reaction components. Free-body diagrams of all members are then
drawn, and from these the variation in direct stress, shear, and bending
moment in each individual member may be readily computed. For the
beginner it is advisable to sketch separate free-body diagrams of the rigid
joints themselves. It should be noted that the forces and moments (or
couples) acting on each free-body diagram, whether it is that of a joint, a

8k
5k 3k
5k 5k 3k 3k
6k-ft
5,
B
B
3k (c)

5k 3k
3k A (b) (d) D
Free body diagrams
5k 3k
(a) (e)

Fig. 4-2

member, or the whole frame, must fundamental equa-

satisfy the three
tions of equilibrium of a general coplanar-force system; i.e., SF^ = 0,
SFy = 0, and XM = 0. It should also be noted that the analysis is
self-checking, because by progressive use of the law that actions and
reactions (forces and moments) are equal and opposite, one completely
defined free body can be obtained in the end, and the equilibrium of this
last free body ensures the correctness of the solution.
It must be emphasized that only statically determinate structures with
rigid joints are discussed in this chapter. The definition of statically
indeterminate rigid frames and their methods of analysis will be con-
sidered in the latter part of this text.
Example 4-1. Draw free-body, shear, and bending-moment diagrams
for allmembers and joints of the rigid frame shown in Fig. 4-3a.
SOLUTION. The three unknown reaction components at the fixed sup-
port are Ma, Ha, and Va assumed to act as shown. By applying the
equations SF;, = 0, 2Fy = 0, and XMa = to the whole frame as a free
body, Ma, Ha, and Va are found to be 66 kip-ft counterclockwise, 5 kips
50 ELEMENTARY THEORY OF STRUCTURES
to the right, and 24 kips upward,
respectively. The free-body, shear, and
moment diagrams members AB, BC, and CD are drawn in succession
of
as shown in Fig. 4-36 to d. The reader is advised to sketch these diagrams
independently and check the results shown in Fig. 4-3. Note that plus
or minus signs as used in connection with the shear diagrams follow the
convention of Art. 3-1 vertical members are treated as beams by looking
;

from the right side of the page. It is not necessary to indicate moment
signs because all bending-moment diagrams are plotted on the compres-
sion side of the member, which means that the bending moment at any
one section is of such a direction that it causes compression on the side
of the member where it is plotted.

3k/ft
cc
5 B„ | 4 4 ; I ^ TT-lC 5

lir +12 -12^20

'+24 O'
+ 24

ppmnir™
116

116

\o,

(d) (e) (0
Note.
Bending moments are plotted
on the compression side.

Fig. 4-3

It should be remembered that the change in the bending moment

between any two sections is equal to the area of the shear diagram between
those two sections. For example, consider member BC. The bending
moment of 116 kip-ft is first plotted on the lower side of the bending-
moment diagram at end B (Fig. 4-3c). By adding the shear area of
+ 1^(24) (8) = 4-96 kip-ft between B and C to the moment at B, the
moment curve rises from 116 kip-ft on the lower side at B to 20 kip-ft
on the lower side at C, which checks with the known value of the end
moment at C on the free-body diagram of member BC. The free-body
diagrams of joints B and C are shown in Fig. 4-3e and /. As shown in
Fig. 4-3g the bending-moment diagram is plotted on the compression
side throughout the whole frame.
 DETERMINATE RIGID FRAMES AND COMPOSITE STRUCTURES 51

Example 4-2. Draw bending-moment diagrams

free-body, shear, and
for all members and frame shown in Fig. 4-4a.
joints of the rigid
SOLUTION. By applying the equations 2Fx = 0, S/^y = 0, and XMd =
to the whole frame as a free body, Md, Hd, and Vd are found to be
51 kip-ft clockwise, 5 kips to the left, and 24 kips upward, respectively.
The free-body, shear, and bending-moment diagrams of members AB^
BC, and CD are drawn in succession as shown in Fig. 4-46 to d. Par-
ticular attention should be called to the manipulation of the forces and
end moments acting on the member BC. It is convenient to first find

3k/ft

5 ^/^ \ \ { } ^

Fig. 4-4

the horizontal and vertical forces acting at B and C by the equations of

statics, and then replace these forces by two forces parallel and perpen-
dicular to the axis of the member as shown in Fig. 4-4c. The total
vertical uniform load on the member is (3) (4) = 12 kips, of which the
component in direction of BC is (12) (^5) = 7.2 kips and the component
perpendicular to BC is (12) (^^) = 9.6 kips. Thus the uniform load
perpendicular to BC is 9.6/5 = 1.92 kips per ft and the axial load along
BC is 7.2/5 = 1.44 kips per ft. The total axial stress in the member thus
increases from a compression of 11.2 kips at B to 11.2 -h 7.2 = 18.4 kips
at C. The shear and bending-moment diagrams for member BC are
52 ELEMENTARY THEORY OF STRUCTURES
then constructed in the usual manner. The free-body diagrams of
joints Band C are shown separately in Fig. 4-4e and /. The complete
bending-moment diagram, plotted on the compression side of the frame,
is shown in Fig. 4-4g.

Example 4-3. Draw free-body, shear, and bending-moment diagrams

for all members and joints of the rigid frame shown in Fig. 4-5a.

p'*

{h) Moment diagram,

plotted on the
compression side

15.4
(g)

Fig. 4-5

SOLUTION. Taking the whole frame as a free body (Fig. 4-5a),

SM^ = 0: (18) (3) + (20) (4) + (5) (12) = (10) (4) + IOVd
Vd = 15.4 kips upward
2F. = 0: 18 = 10 + Ha
Ha = S kips to the left
SF, = 0: Va + 15.4 = 20 + 5
Va = 9.6 kips upward
CHECK. By XMd = 0,

(9.6) (10) + (8) (2) -{- (18) (1) -h (5) (2) - (20) (6) + (10) (2)
140 = 140
 DETERMINATE RIGID FRAMES AND COMPOSITE STRUCTURES 53

The free-body, shear, and bending-moment diagrams for members AB,

BC, CE, and CD are shown in Fig. 4-56 to e. The free-body diagrams of
joints B and C are shown in Fig. 4-5/ and g. The free-body diagram of
joint C at which three members are joined should be particularly noted.

3k/ft 3k/ft

c
WUM

H^=8k

^^ ^^^/^^
192
8 B^ /^.
s/ r (s)

so^' M

m ^;,.
(A) f
34

Fig. 4-6

The bending-moment diagram for the whole frame is plotted on the

compression side as shown in Fig. 4-5/i.
Example 4-4. Draw free-body, shear, and bending-moment diagrams
members and joints of the rigid frame shown in Fig. 4-6a.
for all
54 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. The solution of this problem is fully shown in Fig. 4-6.
The advised to sketch these diagrams independently on a
reader is

separate piece of paper and check the results shown in Fig. 4-6. The
free-body diagram of joint B, on which an external horizontal load of
8 kips acts, should be particularly noted.
4-2. Analysis of Statically Determinate Composite Structures.^ There
are structures in which some members are primarily subjected to direct
stresses and others to bending stresses. Such structures may be called
'^beam trusses" or ''truss beams'' but are commonly known as composite
structures. In analyzing composite structures, it is important to recog-
nize which members are two-force members and which are three-force
members. A two-force member is one which is pin-connected at both
ends and is not subjected to any load between the end joints. A member
that does not satisfy these requirements is a three-force member. Thus,
when a member is treated as a two-force member, its own weight must
not be considered in the analysis (unless the member is in the vertical
position, in which case the member is subjected to a variable axial stress
only) For example, a typical free-body diagram of a two-force member
.

is shown in Fig. 4-7. Here member AB is pin-connected at both ends

and is subjected only to forces Fi, F2, and Fz at A and forces Fa, F5, and F^
at B. Naturally the six forces Fi through F^ acting on AB must be in
equilibrium. Let Ri replace Fi, F2, and Fz and R2 replace F4, F5, and F^;
or Ri is the resultant of Fi, F2, and F3
and R2 that of F4, F5, and Fe. Then
Ri and R2 must be in equilibrium.
Inasmuch as two forces in equihbrium
must be collinear, equal, and opposite,
the member AB must be subjected to
either direct axial tension, as is shown
in Fig. 4-7, or direct axial compression.
Thus members which satisfy the defi-
nition of two-force members as stated
above can always be considered to be
under the action of two equal tensile or compressive forces at the ends,
hence the name two-force members.
Once the two-force members are differentiated from the three-force
members in a composite structure, free-body diagrams can be sketched.
The forces and moments acting on each free-body diagram may be deter-
mined by using freely the three equations of statics and the fact that
actions and reactions are always equal and opposite.
^ For treatment of statically indeterminate composite structures, see C. K. Wang,
"Statically Indeterminate Structures," chap. 12, McGraw-Hill Book Company
Inc., New York, 1953.
 DETERMINATE RIGID FRAMES AND COMPOSITE STRUCTURES 55

Example 4-5. Analyze completely the composite structure shown in

Fig. 4-8a.
SOLUTION. The A and C are shown
external reaction components at
Ha, Va, He,
in Fig. 4-8a as and Vc- The free-body diagrams of members
BC and AB are shown in Fig. 4-86 and c. Member BC is recognized as a
two-force member. Obviously the resultant of He and Vc must be T

1^ = 1401b

//^= 336 lb Jt

r=3641b 7^3,= 1401b

:^ //. = 3361b
^

V^ = 1001b 2401b V;i

r
= 1001b
2401b
7; = 336 lb

(a)
+ 100 + 100
r= 364 lb (+700)
"^^ >f^C=140rb
-<—^^LC
/^= 3361b
( - 700)
- 140 - 140
700

j5 7:^=336 lb

ry=1401b| ^^^
T=3641b
(&) (0
Fig. 4-8

in the direction of BC. Thus the ratio of Vc to He must be %2- The

computations are as follows:

Fig. 4-8a: XMa = = 5He

(240) (7) He = 336 1b
Vc = ^/. i2He = ^2(336) = 140 1b
2F. = Ha^ He = 336 lb
2F„ = Va = 240 - Ve = 240 - 140 = 100 lb
Fig. 4-86: 2F. = T, = He = 336 lb
2F„ = T^ = Ve = 140 lb

Fig. 4-8c. The forces acting on this free-body diagram have been
determined above. As a check, these forces should be tested for equi-
librium before the shear and bending-moment diagrams are drawn.
Example 4-6. Analyze completely the composite structure shown in
Fig. 4-9a.
SOLUTION. In this structure ^C is a two-force member and AB and
BC are three-force members. By treating the whole structure as a free
body (Fig. 4-9a), Ha, Va, and Vb are found to be 0, 10, and 6 kips, respec-
56 ELEMENTARY THEORY OF STRUCTURES
C 4k

//i=lV3k H^-iy^k

V,
FT
=9k 12k V2=3k
(6)

H2=l^^k

V3-3k
H3=lV3ki
h^>^
//2=iV3ir~{ V2-lk
V,=9k V2 = 3k
I
V^ = 10k

(0 f^)

tively. The free-body diagrams of the three members and the three
joints are shown in Fig. 4-96 to g. The order of analysis follows:

1. Fig. 4-9&: 7i = 9 kips 72 = 3 kips

2. Fig. 4-9e: AC = 1% kips Zfi = IM kips
3. Fig. 4-96: H2 = IM kips
4. Fig. 4-9/: Hz = IM kips 73 = 3 kips
5. Fig. 4-9c^: H, = \\i kips 74 = 1 kip
6. Fig. 4-9d: Check by 2M =
7. Fig. 4-9^: Check by 2Fx = and ^Fy =
must be noted that the reactions Va and Vb are external forces
It
acting on the joints A and B. The shear and bending-moment diagrams
for members AB and BC are not shown but may be supplied by the reader.
Note that it is necessary to replace the horizontal and vertical forces
on member BC by forces parallel and perpendicular to the axis of the
member before shear and bending-moment diagrams can be drawn. The
reader is again advised to repeat the analysis and to note that the solution
is entirely self-checking.
Example 4-7. Analyze completely the composite structure shown in
Fig. 4-lOa.
 DETERMINATE RIGID FRAMES AND COMPOSITE STRUCTURES 57

7k

\0__ 6k ^B C 6k
6k

2k

jv^-7k li)-8kT
1
7k
(&)

7k 8k
8k-ft

\i01c

7k
\10k
id) \ m
Fig. 4-10

SOLUTION. CD is recognized to be the only two-force member. (Thus

the ratio of Vd to Hd must be % or ^ in order that their resultant may-
be in the direction Of CD.) The remainder of the solution is shown in
Fig. 4-10 and mil not be explained in detail. It will be interesting to
solve a problem similar to this one, but with the addition of a load in
some direction on member CD (see Prob. 4-11).

PROBLEMS
15 k
15 k
3' y '±
i
I

4k/ft

(;
^
(N
It /
4'
I;
B /
3' 5'

l&
16'

0'

A
--'
^Jw-
Fig. P4-2 Fig. P4-3

3k/ft
rrr
B
i i 4 1 i W li
4' c
4'

8'
*- 9k
D
r— 8'

E
15'

Fig. P4-4 Fig. P4-5

58 ELEMENTARY THEORY OF STRUCTURES
3k/ft
i i i i i i i i
1 i 1
C

6'
"^\ 5'

10 k
By^ D — -

lO* y ly

12'
^ 8' 12'
-

A
Fig. P4-6 Fig. P4-7

4-1 to 4-7. Draw free-body, shear, and bending-moment diagrams for all members
and joints of the rigid frames shown.

12k

' '
I
I I
9001b 9001b

Fig. P4-8 Fig. P4-9

15 k 10 k

12k-

Fig. P4-10 Fig. P4-11

4-8 to 4-11. Analyze completely the composite structures shown.

 CHAPTER 5

STRESSES IN TRUSSES

5-1. Stress Analysis of Trusses. A truss is a structure composed of

individual members joined together so as to form a series of triangles.
The joints are usually assumed to act as frictionless hinges or pins, but
they are often riveted or otherwise connected so that some end restraint
is developed. Generally all the loads on a truss act or are assumed to
act only at the joints. Thus it follows that all members of a truss are
assumed to be two-force members which are subjected only to direct
axial stress (tension or compression). A truss is completely analyzed
when the direct axial stresses {kind and amount) in all members have
been determined.
Either algebraic or graphic methods may be used in the stress analysis
of trusses. The algebraic method consists essentially of isolating certain
parts of the truss by cutting through some members and treating the
stresses in these members as external forces acting on the free body.
If the force system acting on the free body is concurrent, as in a joint,
two unknown stresses can be found. If the force system acting on the
free body is nonconcurrent, three unknown stresses can be found. Such
a free body is usually obtained by cutting a section through three members
and treating either side of the truss as a free body. The former pro-
cedure is commonly known as the method of joints and the latter as the
method of sections. One variation of the method of sections is called
the method of moments and shears. This procedure will be illustrated
in Art. 5-4. The graphic method of stress analysis is equivalent to the
algebraic method of joints. As a matter of convenience, the force
polygons for the concurrent-force system acting on each joint are com-
pounded to form a combined stress diagram, and from this, all unknown
stresses may be measured or scaled. As will be explained later, the
"closing'' of the combined stress diagram provides a valuable check in
the graphic analysis.
The three methods of stress analysis of trusses, viz., the algebraic
methods of joints and of sections (or of moments and shears) and the
graphic method, will be treated in the following articles.
59
60 ELEMENTARY THEORY OF STRUCTURES
5-2. Method of Joints. The external reaction components are first

determined by taking the whole truss as a free body. The two equations
of equilibrium are then applied to the free-body diagrams of all joints in
succession, so that not more than two unknown stresses are involved at
each joint. It is advisable to start first with the joint at the left and
proceed joint by joint to the middle of the structure and then start with
the joint at the right and work back to the middle, thereby obtaining
three checks at the junction. These three checks ensure the correctness
of the solution. It is also advisable to indicate on the truss diagram
not only the total stress in each member, but also the horizontal and
vertical components of this stress, so that the equilibrium of any one
joint can be checked at a glance.
To explain why there are always three independent checks in the
method of joints, the relation between the number of joints and number
of members in a truss will be developed. Since a truss is an assemblage
of triangles, it takes three members and three joints to form the first
triangle, and each additional triangle requires two additional members
but only one additional joint. Let m and j, respectively, equal the
total number of members and joints in a truss; then (m — 3) and
{j — 3) are the number of additional members and joints beyond the
first triangle. From the previous discussion, the number of additional
members is always twice the number of additional joints, or

(m - 3) = 2(i - 3)
Simplifying, m= 2j — 3 (5-1)

The free-body diagram of each joint furnishes two conditions of statics,

so there are altogether 2j conditions, but only m = 2j — 3 unknown

stresses the three extra conditions will therefore be available for checking
;

the solution.
Example 5-1. Using the method of joints, determine the kind and
amount of stress in each member of the truss shown in Fig. 5-1 a.

SOLUTION. applying the equations of statics to Fig. 5-la, Vq and

By
Fa are found to be 19.6 and 22.4 kips, respectively. The free-body
diagrams of joints Lo, Ui, Li, L3, L2, and U2 are drawn in succession in
Fig. 5-16 to g. Two unknown bar stresses can be solved at each joint;
thus,

1. Joint Lo (Fig. 5-16): LoUi = 24.5 kips compression

LoLi = 14.7 kips tension
2. Joint Ui (Fig. 5-lc): U1U2 = 14.7 kips compression
UiLi = 1.6 kips tension
3. Joint Li (Fig. 5-ld): L1U2 = 2.7 kips compression
L1L2 = 16.8 kips tension
 STRESSES IN TRUSSES 61

When the middle of the truss has been reached, it is usually preferable to
start with the right end and work back to the middle where three checks
may be obtained.

4. Joint Ls (Fig. 5-le): L2L3 = 16.8 kips tension

UiLz = 28.0 kips compression
5. Joint L2 (Fig. 5-1/) U2L2 = 24.0 kips tension
2F^ = {check)
6. Joint U2 (Fig. 5-1^) : SFx = [check)
i:Fy = (check)

it is helpful to show on all free-

In applying the resolution equations,
body diagrams the horizontal and vertical projections of the resultant

22.4 k

1.6
24

1.6
14.7 16.8 (check) 16.8
-<
16.8

22.4
(d) (e)

14.7 Uo

/2T ^ 16!8>
24
2f:( = (check)
IFy - (check)

fe)

Fig. 5-1

stress in each member. In Fig. shown the answer diagram on

5-l/i is
which the stresses with their horizontal and vertical components in all
members are indicated. Note that a minus sign indicates a compressive
stress and a positive sign indicates tension. The arrows in Fig. 5-lh
represent the direction of the action of the member on the joint, not the
action of the joint on the member. Take, for example, member LqUi,
62 ELEMENTARY THEORY OF STRUCTURES
The member pushing on the joints Lo and Ui] thus the joints are push-
is

ing on the ends of the member. This indicates a compressive stress.

Usually the experienced computer can omit diagrams such as Fig. 5-16 to
g and work only with a diagram like Fig. 5-l/i.
Example 5-2. Using the method of joints, determine the kind and
amount of stress in each member of the truss shown in Fig. 5-2a.
SOLUTION. The solution by the method of joints is shown in Fig. 5-26.
The reactions at Lo and Li are 2,000 lb each. First take joint Lo as a
free body. To satisfy SFy = at this joint, LqUi must push on the
joint with an effective vertical component of 1,500 lb. By proportion,
the corresponding horizontal component is 3,000 lb and the total stress

1000 lb

H +3000 ^
Li +2000 "£2" +2000 X3 +3000 ' 4 ^

2000 2000

Fig. 5-2

in member LqUi is 3,360 lb (compression). To satisfy the condition

that ZFx = at Lo, member LoLi must
on the joint with 3,000 lb
pull
(tension). Because only two unknowns can be involved at each joint,
consideration next given to joint Ui (there are three unknowns at
is

Li), which is shown as a free body in Fig. 5-2c. To satisfy 2Fn = 0,

1,000 cos 6 = UiLi cos d, and UiLi = 1,000 lb compression. By resolv-
ing forces along the member it is seen that stress in U1U2 is equal to that
in LoL^'i (compression). Joints Li, L2, and U2 are taken as free bodies
in order and the condition 2Fy = for joint U2 is used as a check.
Because symmetry, this is the only independent check available.
of
5-3. Method of Sections. The method of sections involves the passing
of a section through three members and applying the three equations of
statics to the free body on either side of the section. Usually the force
 STRESSES IX TRUSSES 63

system acting on such a free body is nonconcurrent and nonparallel;

thus the unknown stresses in the three cut members can be found. Take,
for example, the truss shown in Fig. o-3a. The stresses in members
U2UZ, UiLz, and L2L3 can be found by passing a section through these
three members and treating either side (Fig. 5-36 or c) as a free body. In
Fig. 5-36 and c, the stresses in members U2LZ and L2L3 are assumed to be
tensile and that in U^Uz compressive. Positive results from subsequent
computations will confirm the assumption. In case of a negative result,
a circle should be drawn around the original arrowhead. This will call

1
\
V2

7
/ \ \7>
<. 1^5

Li L, \ ] \h \ •>
\
"•
\

t
(a)
1

Fig. 5-3

attention to the fact that the assumed direction of the stress in this
member was incorrect. Note that, in Fig. 5-36 or c, any one unkno\\ii
stress may be found independently from a moment equation about the
point of intersection of the two other unknown stresses.
The method of sections is particularly useful when only the stresses
in some members are desired, can be used for solving stresses in all
or it

members. In this event, the equilibrium at each joint can be used to

check the correctness of the solution. The method of joints is con-
venient when at least one of thetwo unknowns at each joint is either
horizontal or vertical in direction. When both unknowns are inclined in
direction, it may be more convenient to apply the moment, instead of
resolution equations to the free body of the joint. This procedure \Aill
be illustrated in Example 5-5,
64 ELEMENTARY THEORY OF STRUCTURES
Example 5-3. Using the method of sections, determine the kind and
amount of stress in members U1U2, L1U2, and L1L2 of the truss shown in
Fig. 5-4a.
SOLUTION. By taking the whole truss as a free body, Vo and Vz are
found to be 19.6 and 22.4 kips, respectively. Next, a section is passed
through members U1U2, L1U2, and L1L2 as shown in Fig. 5-46. Direc-
tions of the unknown forces or stresses are assumed as indicated on the
cut members. Although it is not necessary to anticipate correctly the

18k 18 k

22.4 k

Fig. 5-4

kind of stress, it is advisable to try to avoid negative results by visualiz-

ing the truss action and making the best prediction possible (a negative
sign means that the calculated force or stress acts opposite to the assumed
direction). Taking the left side as the free body,

bending moment at L]
U,U2 =
12
(19.6) (9)
= 14.7 kips compression
12
bending moment at U2
L1L2 =
12
(19.6) (25) - (18) (16)
= 16.8 kips tension
12
ZFy = 0: + 19.6 - 18 - (LiC/2)(%) =
L1U2 = 2.7 kips compression

Taking the right side as the free body,

bending moment at Li
U1U2 =
12
(22.4) (25) - (24) (16)
12
14.7 kips compression (check)
bending moment at U2
LiL:
12
(22.4) (9)
= 16.8 kips tension (check)
12
 STRESSES IN TRUSSES 65

SF, = 0: +22.4 - 24 + (Li?72)(^^) =

L1U2 = 2.7 kips compression (check)

Example 5-4. Using the method of sections, determine the kind and
amount of stress in members U1U2, L1U2, and L1L2 of the truss shown in
Fig. 5-5a.
1000 lb

20001b

20001b

SOLUTION. By passing a plane through members U1U2, L1U2, and

L1L2 of the truss in Fig. 5-5a, the free-body diagrams of the left and
right sides are obtained as shown in Fig. 5-56 and c. Taking the left
segment (Fig. 5-56) as the free body,

2M about U2 = 0: (2,000 - =
500)(16) (1,000)(8) + (LiL2)(8)
=
L1L2 2,000 lb tension
XM about Li = 0: (2,000 - 500)(8) = (^iC72)//(4)
(UiU2)h = 3,000 lb
(UiU2)v = 1,500 lb
U1U2 = 3,360 lb compression

Note that the stress in U\U2 has been resolved at Ui into its horizontal
and vertical components and that the moment of these two components
about Li is equal to the moment of their resultant about Li.
66 ELEMENTARY THEORY OF STRUCTURES
Silf about Lo = 0: (1,000)(8) = (L,U2)v{S)
{LiU2)v = 1,000 lb
{LiU2)h = 1,0001b
L1U2 = 1,410 lb tension

24k (^0^

'^"^ \)-—^(,AC)^^
/ >.

6'

/
/ ^A>
^ J^
/ ,,^— WC)„
/M-i
6'
\\ ^y
//^=8k
1
«
T
•1
t
1

^ = 6k Vg=18k 6k
(a) (6)

24 k

(5cv[v7^(^^v /|^^''

t t \
18k 6k 18 k
(c)
(rf)

Fig. 5-6

Taking the right segment (Fig. 5-5c) as the free body,

2M about V2 = 0: (2,000 - 500)(16) = (1,000)(8) + (LiL2)(8)

L1L2 = 2,000 lb tension (check)
2ikf about Li = 0:
(2,000 - 500) (24) = (1,000) (16) + (1,000) (8) + {UiU2)hW
{UiU2)h = 3,000 1b
(UiU2)v = 1,500 lb
U1U2 = 3,360 lb compression (check)
1:M about Lo = 0:
(2,000 - 500)(32) = (1,000)(24) + (1,000)(16) + (LiU2)v(S)
(LiU2)v = 1,000 lb
(LiU2)h = 1,000 lb
L1U2 = 1,410 lb tension (check)

Example 5-5. Determine the kind and amount of stress in each

member of the truss shown in Fig. 5-6a. Use either the method of
joints or the method of sections.
 STRESSES IN TRUSSES 67

SOLUTION. The external reaction components at A and B must first

be determined. From 2Fx = on the whole free body, Ha must be
equal to 8 kips to the left. The 24-kip vertical load will be reacted by
12 kips each at A and B, and the 8-kip horizontal load will produce a
clockwise moment of 96 kip-ft about A or B, which can be balanced by a
counterclockwise reaction couple of 96/16 = 6 kips upward at B and
6 kips downward at A. Thus,

Va = 12 - ^^^^ = 12 - 6 = 6 kips upward

7^ = 12 + = 12 + 6 = 18 kips upward
^^^Y^
The free-body diagrams of joints A and B are shown in Fig. 5-66 and c.
Here the resolution equations of equilibrium are not readily applicable,
but the moment equations can be used advantageously. Taking joint A
(Fig. 5-66) as the free body,

XMc = 0: + (8) (12) = (AD)h{Q)

(6) (8)
(A D)h = 24: kips
AAD)v = (AD)HiH) = 18 kips
AD = {AD)h(H) = 30 kips tension
XMn = 0: (6)(8) + (8)(6) + (AC)^(6) =
{AC)h = — 16 kips [draw circles around arrowheads of {AC)h, {AC)vj
and AC]
(AC)v = (AC)i/(%) = 24 kips (numerically only)

AC = {AC)h{ —^— ) = 28.8 kips compression

Taking joint B (Fig. 5-6c) as the free body,

SMc = 0: (18)(8) = (BZ))^(6)

(BD)h = 24 kips
{BD)v = (24) (M) = 18 kips
BD = (24) (^i) = 30 kips tension
^Md = 0: (18) (8) = (BC)h(6)
{BC)h = 24 kips
(BC)v = (24)(%) = 36 kips

BC = (36)
m-
——
I
j
= 43.2 kips compression

As soon as they are calculated, the values of the horizontal and vertical
components as well as the resultant stress in each member are entered in
the answer diagram shown in Fig. 5-6d. By inspection of the equilibrium
of joint C, the stress in CD is found to be 36 kips tension. The equi-
librium of all joints can be readily checked on the answer diagram.
68 ELEMENTARY THEORY OF STRUCTURES
6-4. Method of Moments and Shears. The method of moments and
shears may be considered as merely a variation of the method of sections.
A horizontal truss subjected to vertical loads resembles a beam in its
structural action. The top chord of the truss takes compression and the
bottom chord takes tension. In a parallel chord truss the web member
(vertical or diagonal) resists the shearing force at the section. Consider,
for example, the truss shown in Fig. 5-7. The stress in member UaUb,
by reason of the method of sections, is equal to the bending moment at
La divided by the height of the truss. Similarly, the stress in member
L4L5 is equal to the bending moment at L'5 divided by the height of the
truss. Note that the bending moment at any joint of a truss may be
computed by using either the left or the right free body. It is also obvi-
ous that, in this simply supported truss, the entire top chord is in com-
pression and the entire bottom chord in tension. The stress in the
diagonal LiUf,, by observing ZFy = in the method of sections, is equal
to the shear at the section multi-
plied by the secant of the angle
which the member makes with the
vertical. If the shear at the sec-
tion is positive, the stress in L^Ub is
compressive, and if the shear is neg-
Fig. 5-7
ative, the stress is tensile. The
same procedure applies to vertical members, except the stress in a verti-
cal member of a horizontal-chord truss is numerically equal to the shear
at the section, and the kind of stress may be determined by inspection.
If the upper chord or the lower chord (or sometimes both) cut by a sec-
tion is inclined in direction, the vertical component of the stress in the
web member is equal to the shear at the section modified by the vertical
component of the stress in the inclined chord member.
The conception of having the chord members resist the bending
moment and the web members take the shearing force at a section may be
conveniently applied to finding the stresses in all members of a truss.
After the stresses have thus been found, each joint should be checked for
equilibrium; i.e., the forces acting on the joint must satisfy the two
conditions ZF^ = and ZFy = 0. If this check is applied at every joint,
the correctness of the solution is assured.
Example 5-6. Determine the kind and amount of stress in each mem-
ber of the truss shown in Fig. 5-8 by the method of moments and shears.
SOLUTION. Let
d = panel length
h = height of truss
s = length of diagonal
6 = angle between diagonal and vertical
 STRESSES IX TRUSSES 69

Then tan d = sec d =

It is noted that the top chord compression and the bottom chord
is in
is in tension; so the kind of stress in chord members is no longer in ques-
tion. The amount of the chord stresses mav be found as shown below.

bending moment atTi

LoL/i = J
L1L2 = J
= ShzPd = 6^or
oi p
, ~
.,,,
tan
n
if

n n
bending moment at L2 or 17%
l\U2 = I2L:
{sy2P){2d) - p{d)
QP tan d

bending moment at L3 or t/j

U.Vz = L3I4
(31 2 P) (3d) - P(d-{-2d)
lyzP tan e

bending moment at L4
Uzl\
h
(3HF)(4rf) - Pid -\-2d-\- 3d)
SP tan d

By the method of shears, the stresses in the diagonals Lolu ^ 1L2,

U2LZ, and Ud^i are found to be Sy^P sec d, 2^2^ sec 6, UoP sec d, and

31^ p a^p
Fig. 5-8

y^P sec 6] those in the verticals U2L2 and U^Lz are I'^oP and ^o^- respec-
tively. The kind of stress in these web members is determined by inspec-
tion. The stress in the vertical hanger L'lLi is -\-P, and that in the center
vertical UXa is zero. Since the truss is symmetrically loaded, the
stresses in the members on the right side of the center line are the same
as those in the corresponding members on the left segment of the truss.
In this solution it \\i\\ be noted that the stresses in the chord members
are all expressed in terms of P tan d; in the diagonals, P sec 6; and in
the verticals, P. The coefl&cients themselves are sometimes kno^^^l as
70 ELEMENTARY THEORY OF STRUCTURES
index stresses, and the method here treated may be called the method of
index stresses.
Again, the correctness of the solution may be checked by observing
that the two resolution equations are satisfied at each joint.

Fig. 5-9

Example 5-7. Determine the kind and amount of stress in each

member of the truss shown in Fig. 5-9 by the method of moments and
shears.
SOLUTION. The }^P and %P. The stresses
reactions are noted to be
in the diagonals are first found by the method of shears. Note that
only the coefficients of P sec 6 are entered on the diagonals of Fig. 5-9.
The stresses in the chord members may be found by taking moments.
For instance, the compressive stress in U2Ui is equal to the bending
moment at L2 divided by /i; or

U^Uz = (M^IM = y^p tan d (compression)

Likewise, the tensile stress in L4L5 is equal to the bending moment at Us

divided by /i; or

= (^^^)(^^) = 2P
L4L6 .

tan d (tension)

In Fig. 5-9, only the coefficients of P tan 6are written on the chord
members. The equilibrium at each joint is then reviewed to check the
solution completely.
Example 5-8. Determine the kind and amount of stress in each mem-
ber of the truss shown in Fig. 5-10 by the method of moments and shears.
SOLUTION. Finding the reactions,

2M, = 0: 70 = mjL^ f + 10(3)

= kips

ZMo = 0: F, = 10(4) + 12(5) + 9(6) ^ ^2 kips

CHECK. By 2n = 0, 9 -h 22 = 10 -M2 + 9
31 = 31
 STRESSES IN TRUSSES 71

The stresses in the lower chord members and the horizontal components
of the stresses in the upper chord members are determined by taking
moments. Thus,

^ , ^ ^
= bending moment at 1 (9) (24) ^ .

LoLi = L1L2 ^—fta = r^A =

,

9 kips
24 24
= bending m^^nenta^ ^ (9)(2K_24)
L^3 = {U.U.), ^ ^3 ^^ ^^^^

bending moment at 3 ^ (9) (3) (24)

LaL4 = {U2U^)h = ~ = 18 kips
36 36
bending moment at 4 _ (9) (4) (24)
~ = 24 kips
36 36

= bending moment at 5 _ [(22) (2) - (9)(1)]24

LJLi = (UbU6)H
32 32
= 26.25 kips
bending moment at 6 (22) (24)
LsL, J J
L/qL/j = ^:^ — PT"^
— ZZ Kips
.

24 24

The vertical components upper chord members

of the stresses in the
are found from the horizontal components computed above and the
slopes of the upper chords.

10 k
7 at 24' = 168'

Vo-9k V7-22k
Fig. 5-10

The stresses in the vertical members exclusive of UiLi, UaL^, and L^e^e
and the vertical components of the stresses in the diagonals are deter-
mined by the method of shears. When a section is cut through an
inclined upper chord, however, the vertical component of the stress in
this upper chord must be considered in applying the equation i:Fy =
to the free body on either side of the section. For instance, considering
the free body to the right of section A A (Fig. 5-10) and remembering
that the vertical component of the stress in UaUb is 4 kips,
'

{LdJ^)v + 22 - 9 - 12 - 4 =
(L4L^5)f = 3 kips
72 ELEMENTARY THEORY OF STRUCTURES
After the vertical components of the stresses in the diagonals are com-
puted, the horizontal components can easily be found by simple propor-
tion using the slopes of the diagonals. The stresses in UiLi, U4L4, and
UeLe are found by considering XFy = at joints Li, Ua, and Le.
Equilibrium at each joint can now be reviewed and the total stresses
in the inclined members can be found from their horizontal and vertical
components. The complete answer diagram is shown in Fig. 5-10.
6-5. The Graphic Method. Two steps are involved in the graphic
method of stress analysis of trusses: (1) the external reaction components
must be determined by either the algebraic or graphic methods previously
discussed in Chap. 2, and (2) the internal stresses in all members can be
obtained from the stress diagram, which is the superposition of all the
individual force polygons for the concurrent-force systems acting on each
joint.
In order that the second phase of this work can be systematically per-
formed, a definite scheme of notation and procedure has been suggested,
as shown in Fig. 5-1 la. The external reaction components may be
determined by one of the methods discussed in Chap. 2. Thus the
balanced external-force system acting on the truss includes Pi, P2, Va, P3,
Pi, Vo, and Ho, named in the clockwise order around the truss. Each
space between two adjacent external forces is labeled with a capital
letter in consecutive order, by starting with the letter A and then pro-
ceeding in the clockwise direction. Thus the force P4 can be called AB,
the force V^ called CD; etc. (see Fig. 5-1 la). Each triangle is labeled
with a numeral by starting with the number 1 and proceeding from
left to right. The two numbers, or one letter and one number, on oppo-
site sides of each member are used to represent the magnitude and direction
of the stress in the member. By referring to joint Lo, and reading around
the joint in the clockwise direction, the stress in member LqUi will be
called A-l, not 1-A but when referring to joint Ui, the stress in the same
;

member LqUi should be called 1-^, not ^-1. It will be found that this
distinction is of great importance in determining the kind of stress in a
member.
The procedure by which the stress diagram in Fig. 5-116 is drawn will
now be given. The force polygon of the external-force system is first
plotted as ABCDEFGA. The points 1, 2, 3, 4, 5, and 6 are then to be
located. Point 1 is at the intersection of a line through A parallel to
member A-l (or LqUi) and a line through F parallel to member 1-F (or
LoLi). With the locations of points 1 and E known in Fig. 5-116, point 2
is the intersection of two lines through 1 and E, respectively parallel to
members 1-2 (or LiUi) and 2-E (or LiLz). In this case it may be advis-
able to locate the points 1, 2, and 3 first and then points 6, 5, and 4.

The fact that Une 3-4 should be vertical (or parallel to member L2U2)
 STRESSES IN TRUSSES 73

serves as a check on the accuracy of the graphic solution. By proceeding

from each end and closing up near the middle, the cumulative effect of
graphic errors can be largely eliminated.
The amount of the stress in member 2-3 (or U1L2) is given by the
magnitude of the vector 2-3 in the stress diagram. The kind of stress

(f) Jomt U2

(c) Joint Lq
Fig. 5-11

can be determined from the following consideration. The stress in

member U1L2 is called 2-3 when read clockwise around joint L2. The
direction of 2-3 in Fig. 5-116 is downward to the right, which, when

referred to joint L2, means that the member exerts compression on the
joint. The stress in member U1L2 is therefore compressive. The same
conclusion can be reached if the stress in member U1L2 is called 3-2
when referred to joint Ui. The direction of 3-2 in Fig. 5-116 is upward
74 ELEMENTARY THEORY OF STRUCTURES
to the left, which, with reference to joint Ui, means compression on the
joint. Thus the amount and kind of the stress in each member can be
readily found from the stress diagram.
In fact, the stress diagram
merely the superposition of the individual
is

force polygons for the concurrent-force systems at each joint. By con-

sidering joint Lo as a free body, the unknown stresses in members LoUi
and LoLi can be scaled from the force polygon shown in Fig. 5-1 Ic. The
force polygons for joints Li, Ui, and Uz are shown in Fig. 5-1 Ic? to /.
A comparison of the force polygons shown in Fig. 5-1 Ic to / and others
which may be similarly drawn for the remaining joints with the combined
stress diagram will demonstrate the statement made at the beginning

18k B
'Ui -14.7 Uo

Vo = 19.6 k 24 k v;3 = 22.4 k 2

(a)

Scale. 1" = 20'

Scale. l" = 30k Scale. l" = 12k
(b) Space diagram (c) Force polygon (d) Stress diagram

Fig. 5-12

of this paragraph and explain the reasoning incident to the combined

stress diagram.
Example 5-9. Determine the kind and amount of stress in each mem-
ber of the truss shown in Fig. 5-12a by the graphic method.
SOLUTION. The reactions are determined by the graphic method.
The space diagram and the force polygon are shown in Fig. 5-126 and c.
It should be noted that the known forces are designated ab and he and
the unknown reactions are indicated by cd and da. By scaling the
force polygon in Fig. 5-12c, it is found that Vz = CD = 22.4 kips
and Vo = BA = 19.6 kips.
The previously mentioned notation for the graphic solution of the
internal stresses is now inserted in Fig. 5-12a; A, B, C, and around D
the truss and 1, 2, 3, and 4 in the internal triangles. The stress diagram
shown in Fig. 5-12rf is drawn in accordance with this notation. The stress
 STRESSES IN TRUSSES 75

in each member is scaled in Fig. 5-1 2d and then written on the proper
member in Fig. 5-12a.
Example 5-10. Determine the kind and amount of stress in each
member of the truss shown in Fig. 5- 13a by the graphic method.
1000 lb

10001b ^'h

^'^'
5001b

Vn=20001b V4= 20001b

(a)

Scale. 1" = 2000 lb

(6) Stress diagram

k SOLUTION.
Fig. 5-13

Because of symmetry, the reactions are 2,000 lb each.

The stress diagram is shown in Fig. 5-136, and the scaled stresses are
written on the truss members in Fig. 5-13a. It may be interesting to
note that, in the case of this simple and symmetrical stress diagram, it is
relatively simple to compute the stresses from the geometry of the stress
diagram.

^^ PROBLEMS
5-1 to 5-4. Determine the kind and amount of stress in each member of the truss
shown by the method of joints.
6-5. Using the method of sections, determine the kind and amount of stress in
members UzUz, L2C/3, and L2L3 of the truss in Prob. 5-1.
5-6. Using the method of sections, determine the kind and amount of stress in
members U2UZ, U2L3, and L2L3 of the truss in Prob. 5-2.
76 ELEMENTARY THEORY OF STRUCTURES
6k 12k 16 k 2C k 8k 1000 lb

Uo
Ih U^ ^3 ^ 1000 lb

/
/ \\ / / \ ^
500
10001b ^2

lb
^
hL \ \^2/ \
^3
4 at 15' = 60'

Fig. P5-1 Fig. P5-2

5001b
10001b

t/2 f/3 10001b dI ri

^--^ ^4 \
1000 lb c\^ ^

'-ISk
\\ ^2 L3 i
t\
^4 5
§

tAlv
Yh

-^-i
F G
5 at 15' = 75' 20'

Fig. P5-3 Fig. P5-4

5-7. Using the method of sections, determine the kind and amount of stress in
members U1U2, U2L2, and L2L3 of the truss in Prob. 5-3.
5-8. Using the method of sections, determine the kind and amount of stress in
members BC, CF, and FG of the truss in Prob. 5-4,
5-9. Determine the kind and amount of stress in each member of the truss in
Prob. 5-3 by the method of moments and shears.

f/i U2 U3 t/4 t/5

U2 U3

^^
A\N/ /\. Li ] £2 L3
1
..6,'
^ 3
1 P P
L=10c/

Fig. P5-10 Fig. P5-11

5-10 and 5-11. Determine the kind and amount of stress in each member of the
truss shown by the method moments and shears.
of
5-12 to 5-15. Determine the kind and amount of stress in each member of the
trusses shown in Probs. 5-1 to 5-4, inclusive, by the graphic method.
 CHAPTER 6

ANALYSIS OF ROOF TRUSSES

6-1. General Description. In building design wherein an open, unob-

structed space with a width of more than 40 or 50 ft is to be provided,
the roof is commonly supported by roof trusses spaced from about 15 to
25 ft apart. These trusses may rest on columns or on masonry walls
along the sides of the building. A roof truss attached to its supporting
columns is commonly called a bent, the analysis of which will be treated
in Chap. 7. The discussion in the present chapter will be limited to roof
trusses supported on masonry walls. If the span of the roof truss is

Purlins

Rafters

k Fig. 6-1

small, say less than 40 or 50 ft, the truss may usually be anchored to the
wall at both ends, or the anchor bolts at one end may pass through
slotted holes in the bearing plate to make some provision for expansion
or contractiondue to temperature changes. For longer spans, trusses
should be hinged at one end and supported on rollers or rockers at the
other end.
A sketch showing the typical roof construction is shown in Fig. 6-1.
Here the roof trusses are supported on continuous masonry walls. The
distance between adjacent roof trusses is called the hay. The purlins
are longitudinal beams which rest on the top chord, and preferably at the
joints of the truss. Unless the purlins are placed at the joints, the top
77
78 ELEMENTARY THEORY OF STRUCTURES
chord will be subjected to combined bending and direct stresses. The
roof covering (with or without sheathing) may rest directly on the purlins,
or on rafters which are in turn supported by the purlins. A typical
interior roof truss receives purlin loads from both sides; so it supports
roof loads on the equivalent of one whole bay length. The span of the
roof truss is the horizontal distance between the supports; the rise is the
vertical distance from the ridge to the eaves; the pitch is the ratio of the
rise to the span. Although the walls and the purlins assist in maintain-
ing longitudinal stability, additional bracing is usually necessary. The

mm^ ^^^ fa) Howe truss

(e) Fan or cambered
fan truss

D
(/) Saw-tooth truss

(6) Pratt truss

(c) Warren truss

(g) Scissors truss

(d) Fink or cambered (h) Three-hinged arch

Fink truss
Fig. 6-2

longitudinal or diagonal bracing members which run from truss to truss

may be in the plane bottom chord, or both; thus the
of the top chord, the
names top-chord bracing or bottom-chord bracing. Although approximate
stress calculations may be made, actual design of the bracing system is
largely a matter of experience and judgment.
The selection of the type of the roof truss to be used generally depends
on the length of span, the amount of loading, and the kind of materials
to be used. Eight common types of roof trusses are shown in Fig. 6-2a
to h. The vertical members of the Howe truss or the diagonal members
 ANALYSIS OF ROOF TRUSSES 79

normally in tension and may economically be steel

of the Pratt truss are
rods if is built of wood.
the rest of the truss The top chords of Howe,
Pratt, and Warren trusses may be horizontal (or with sufficient slope to
provide drainage) or with pronounced slopes as indicated. The Fink
and Fan trusses, with or without camber in the lower chord, are usually
built of steel. The sawtooth truss is suitable for mill buildings where
light from windows facing to the north is desired. The scissors truss is
often used in church structures. Three-hinged arches, with or without
tie rods, are often used in buildings with long spans, such as armories or
gymnasiums.
6-2. Dead, Snow, and Wind Loads. A truss carries its own weight,
the weight of the bracing and ceiling and other suspended loads, and the
loads from the purlins. One of the requirements for true truss action
is that the loads be applied at the joints only. Sometimes purlins are
placed on the top chord between panel points. In this case the loads are
distributed to the adjacent joints during the truss analysis, but both the
direct stresses as determined from the truss analysis and the bending
stresses due to the intermediate purlin loads between panel points must
be considered in the design of the top chords. The weight of the truss
and the bracing system may be assumed to be divided among the joints
on the top chord, while ceiling and suspended loads are assumed to be
carried by the appropriate lower chord joints.
The loads on roof trusses generally consist of dead, snow, and wind
loads. The dead load includes (1) the weight of roof covering (with
or without sheathing), rafters if any, and purlins; (2) the weight of the
bracing system (3) the weight of the truss itself and (4) ceiling and other
;
;

suspended loads. If desired, (1) and (4) may be ascertained before the
beginning of the truss analysis; items (2) and (3) must be first assumed
and then reviewed after design calculations have been made. Fortu-
nately items (2) and (3) are usually a small part of the total load; so
even a rather large error in their assumed values may have a relatively
insignificant effect on the resultant maximum stresses. Roof coverings
are commonly corrugated steel, asphalt or asbestos, various types of
shingles, tiles, slates, or thin concrete slabs, and tar and gravel. Sheath-
ing or rafters may or maynot be used, depending on whether the roof-
ing material is self-supporting. The weight of item (1) above can easily
vary from 5 to 25 The weight of a
lb per square foot of roof surface.
plastered ceiling may
be 8 to 10 lb per square foot of horizontal surface.
The weight of the bracing system may vary from 3^ to 1^ lb per square
foot of roof surface.
The weight of the roof truss is usually estimated by use of an appropri-
ate empirical formula. Two of these are given here; others are available
in handbooks.
80 ELEMENTARY THEORY OF STRUCTURES
For wooden roof trusses,

w = 0.5 + 0.07dL (H. S. Jacoby) (6-1)

For steel roof trusses, - ^'^ '^/Iris'^ p 't t^X/''^ /«>-=/

w = 0A + 0.04L (C. E. Fowler) (6-2)

In the above lormulas, w is the weight of truss in pounds per square foot
of horizontal surface, and L is the span in feet. It should be noted that
any empirical formula should be used with discretion, or with adapting
modifications. Some preliminary estimates are often advisable when
empirical formulas are used.
The snow load which may come to the roof depends on the climate
of the locality and on the pitch of the roof truss. The density of dry
snow may be taken at 8 lb per cu ft, and that of wet snow at 12 lb per
cu ft. Snow tends to pile up on flat roof surface. The snow load per
square foot of inclined roof surface may be assumed as (1 — 0/60) times
the estimated load per square foot of flat surface, wherein Q is the angle
in degrees between the inclined roof surface and the horizontal. If
appropriate for the climate, the snow load may be assumed to be 15, 20,
or 25 lb per square foot of roof surface.
The wind load on a roof surface depends on the pitch of the roof truss
and on the velocity of mnd, which in turn is a function of the height
of the building. The wind pressure p in pounds per square foot of vertical
surface due to a wind velocity Y in miles per hour is usually assumed to
vary from p = 0.003 F^ to p = 0.0047^. Thus a provision for wind
pressures of 20, 25, or 30 lb per square foot of vertical surface may provide
for wind velocities of 75, 85, or 95 mph. The wind pressure normal to an
inclined roof surface is usually found by use of the Duchemin empirical
formula (1829),
2 sin fr.ON
P" = P 1 + sin^ e
^^-^)

in which pn is the normal pressure on an inclined roof surface at an angle B

with the horizontal, and p is the assumed pressure on the vertical surface.
Recent investigations have shown that wind may not only exert
pressure on the windward side; it may actually exert suction on the
leeward side. not likely to affect the design of the members
Suction is

of the roof truss supported on masonry walls, but its effect is of enough
importance to point up the necessity for sufficient anchorage to prevent
lifting of the truss. Also, if the windows on the windward or leeward
side are open or broken, pressure or suction may come to the inside of
the roof. Although the use of the Duchemin formula as discussed above
is generally considered to be conservative, these newer conceptions are

noteworthy.
 ANALYSIS OF ROOF TRUSSES 81

6-3. Wind Loads as Recommended in the 1940 ASCE Final Report.

Sub-committee 31 of the American Society of Civil Engineers made some
definite recommendations in regard to wind forces in its 1940 final report.^
Although the report prescribes wind forces for both plane and round roof
surfaces, only those for plane surfaces have been abstracted from the
above-mentioned source as follows
1. A uniformly distributed force of 20 psf for the first 300 ft above

ground level, increased above this level by 2.5 psf for each additional

p=+9
+9
/
CO

2
•s^
a y
10 20 m 40 50 60 10 20 30 40 50
"^ ^ Roof slope/a, degrees Roof slope a, degrees

i^ r,
/ 7
*
CO
-9
i p=-9
-12 -J" -12^
P= -li^

(a) External force on windward side (6) External force on leeward side

p= +12

+ 4.5

10 20 30 40

C Estimated percentage
-4.5 •> n of wall opening

"'>
P--9
(c) Internal force on both sides of roof

Fig. 6-3

100 ft of height, is recommended as a standard wind load for the United

States and Canada.
2. For plane surfaces inclined to the wind and not more than 300 ft

above the ground, the external wind force may be pressure or suction,
depending on the exposure and the slope. For a windward slope inclined
at not more than 20° to the horizontal, a suction of 12 psf is recommended;
for slopes between 20 and 30°, a suction uniformly diminishing from 12 psf
to (p = 1.20a: — 36); and for slopes between 30 and 60°, a pressure
increasing uniformly from to 9 psf (p = 0.30q: — 9). On the lee-
1 Final Report of Sub-committee 31, Trans. ASCE, vol. 105, pp. 1713-1720, 1940.
82 ELEMENTARY THEORY OF STRUCTURES
ward slope, for all inclinations in excess of zero, a suction of 9 psf is
recommended.
3. For a flat roof a normal external suction of not less than 12 psf

should be considered as applied to the entire roof surface.

4. For buildings that are normally airtight an internal pressure or

suction of 4.5 psf should be considered as acting normal to the walls

and the roof. For buildings ^^'ith 30 per cent or more of the wall surfaces
open, or subject to being open, an internal pressure of 12 psf, or an

(a) Duchemin's, wind from left (6) Duchemin's, wind from right

(c) ASCE, extemiil force plus internal id) ASCE. external force plus internal
pressure, wind from left pressure, wind from right

(e) ASCE, external force plus internal (f) ASCE, external force plus internal
suction, wind firom left suction,wind from right

Fig. 6-4

internal suction of 9 psf. is recommended; for buildings with wall open-

ings varying from an internal pressure
to 30 per cent of the wall space,
varjang uniformly from 4.5 to 12 psf (p 4.5 + 0.25/1, n = percentage =
of opening), or an internal suction varying uniformly from 4.5 to 9 psf
(p = +
0.15m) is recommended.
4.5
5. design wind force applied to any surface of a building is to be
The
a combination of the afore-mentioned appropriate external and internal
wind forces.
6. When ^s^ind surfaces are more than 300 ft above the ground, the
external and internal wind forces should be scaled up in the proportion
 ANALYSIS OF ROOF TRUSSES 83

that the prescribed wind force on plane surfaces normal to the wind
at the level under consideration bears to 20 psf
The external and internal wind forces on inclined-plane surfaces at
not more than 300 ft above the ground, as described in items 2, 3, and
4 above, may be indicated graphically as in Fig. 6-3. The determination
of wind loads on roof trusses on the basis of the 1940 ASCE final report is

illustrated in the following examples.

Example Compare the wind loads on the roof truss shown in
6-1.
Fig. 6-4a (a) by the use of the Duchemin formula and (6) in accordance
with the 1940 ASCE recommendation. The bay distance between
trusses is 15 ft and the normal pressure in the direction of the wind is
20 psf. Assume 20 per cent wall opening.
SOLUTION
Length of AB (Fig. 6-4a) = V^' + 6^ = 7.21 ft
Tributary area of roof surface per panel = (15) (7.21)
= 108.2 sq ft

(a) Duchemin formula

^^^ 2 sin e
^^^ 2(2/^A3)^ ^
^^ ^^
^ + sm2 (2/ Vl3)'
1 d 1 -I-

Wind panel load = (16.97) (108.2) = 1,836 or 1,840 lb

(h) ASCE recommendation

a = = 33.7°
arctan %
External force, windward side = 0.30(33.7) — 9
= 1.11 psf (pressure)
External force, leeward side = 9 psf (suction)
Internal force, both sides = 4.5 -|- 0.25(20)
= 9.5 psf (pressure)
Internal force, both sides = 4.5 + 0.15(20)
= 7.5 psf (suction)
Thus, taking the inward panel load as positive and the outward as
negative, for the case of external force plus internal pressure,

Wind panel load, windward side = ( + 1.11 - 9.5)(108.2)

= -908 or -910 1b
Wind panel load, leeward side = (-9 - 9.5) (108.2)
= -2,002 or -2,000 1b
Considering external force plus internal suction,

Wind panel load, windward side = ( + 1.11 + 7.5)(108.2)

= +932 or +930 lb
Wind panel load, leeward side = (-9 + 7.5)(108.2)
= -162 or -160 1b
84 ELEMENTARY THEORY OF STRUCTURES
The wind panel loads as computed above are summarized in Fig. 6-4.
It is apparent that the use of the Duchemin formula is conservative as
far as maximum combined stresses (with the same sign as the dead-load
stress in the member) are concerned. But the ASCE loading of Fig. 6-4c
or d should be carefully considered because it may cause a stress of the
opposite sign numerically larger than the dead-load stress so that the
member must be designed to take both tension and compression. Also
the required anchorage of the truss must be investigated. The loading
shown in Fig. 6-4e and /, being less than that shown in Fig. 6-4a and h,

may not have any significance. The reader is advised to familiarize himself
with the ASCE recommendation, especially for use in unusual situations.
Example 6-2. Compare the wind loads on the roof truss shown in Fig.
6-5a (a) by the use of the Duchemin formula and (h) in accordance with
the 1940 ASCE recommendation. The bay distance between trusses is
18 ft and the normal pressure in the direction of the wind is 20 psf.
Assume more than 30 per cent wall opening.
SOLUTION
Length of AB (Fig. 6-5a) = V^-^' + 3.752 _ g 335 f^

Tributary area of roof surface per panel = (18) (8.385)

= 150.93 sq ft

(a) Duchemin formula

Pn
^"
= V
^
-^—2sin^
^

+ sm2 ^^r-k = ^^
20
2(l/-v/5)
, = ,.f., .
14.91 psf
1 d 1 + ( 1/V5)2
Wind panel load = (14.91) (150.93) = 2,250 lb

(6) ASCE recommendation

a = arctan 3^ = 26.57°
External force, windward side = 36 — 1.20(26.57)
= 4.12 psf (suction)
External force, leeward side = 9 psf (suction)
Internal force, both sides = 12 psf (pressure)
Internal force, both sides = 9 psf (suction)

Taking the inward panel load as positive and the outward as negative,
for the case of external force plus internal pressure,

Wind panel load, windward side = (-4.12 - 12)(150.93)

= -2,433 or -2,430 1b
Wind panel load, leeward side = (-9 - 12) (150.93) = -3,170 lb

Considering external force plus internal suction.

Wind panel load, windward side = (-4.12 + 9)(150.93)
= -773 or +770 lb
Wind panel load, leeward side = (-9 + 9) (150.93) =
 ANALYSIS OF ROOF TRUSSES 85

The wind panel computed above are summarized in Fig. 6-5.

loads as
conserva-
Again it is apparent that the use of the Duchemin formula is
due dead, snow,
tive with regard to the maximum combined stresses
to
per cent
and wind loads. But, because of the possibility of more than 30
wall opening, the internal pressure, together with external suction,
6-5c and d. In some
yields large outward panel loads as shown in Fig.
members these loads may produce stresses numerically larger than the

1*S)t)V.
.-1

A
60'
*

(a) Duchemin's, wind from left (6) Duchemin's, wind from right

^o,0vTN^%,?7;^^
,^x^^\^^X?^^^'^o

(c) ASCE, extemsJ force plus internal (d) ASCE, external force plus internal
pressure, wind from left pressure, wind from right

(e) ASCE, external force plus internal (/; ASCE, external force plus internal
suction, wind from left suction,wind from right

Fig. 6-5

dead-load stresses. Inasmuch as these dead and wind stresses are of the
opposite sign, there will be stress reversals for which provision must be

made.
6-4. Combinations of Loads. The function of stress analysis is to
provide the designer with the most probable maximum or minimum
(reversal) stresses to which any truss member may be subjected;
conse-

quently consideration must be given to the combinations of dead, snow,

and wind loads which are to be accommodated in the design. The usual

combinations are: (1) dead plus full snow on both sides, (2) dead plus

wind on either side, (3) dead plus half snow on both sides plus wind
from either side, (4) dead plus full snow on the leeward side and wind
on the windward side, (5) dead plus ice (which may be 5 to 10 lb
per
86 ELEMENTARY THEORY OF STRUCTURES
square foot of roof surface) on both sides plus wind on either side, and
(6) dead plus ice on both sides plus full snow on the leeward side plus wind
on the windward side. Much depends on the judgment of the designer
as to which load combinations should be used.
Because maximum winds may come only occasionally and are usually
of short duration, most specifications allow a 333^ per cent increase in
the unit working stress in cases where wind effect is included. In such
a case, any of the load combinations (2) to (6), if selected, will not be
controlling unless the resulting combined stress exceeds % times that
indicated in condition (1). Or stated differently, load condition (1)
will control the design if the resulting stress exceeds ^ of that from any
other load combination in whichmnd load is included.
Ordinarily a consideration of load combinations (1), (2), and (3)
willprovide adequate design data. If there may be a reversal of stress
in any one member, it must be caused by wind from the opposite side.
The minimum stress, or maximum stress opposite in sign to that of dead

Fig. 6-6

load, if any, must be due to load combination (2), and not load combina-
tion (3), because the inclusion of half snow on both combina-
sides in load
tion (3) will only nullify some of the reverse stress due to wind. It has
been found, however, that with the exception of cases involving small
dead load and large wind load, there are very few cases of stress reversal
in roof trusses supported on masonry walls. In the usual cases then, it
seems desirable to devise some sort of '^ equivalent" vertical loading to
simulate the combined effect of snow and wind on the basis of normal
working stress. It is logical to conclude that this '^equivalent" vertical
loading should be either that of full snow only or half snow plus a certain
fraction of the wind pressure normal to the roof surface. Here again,
only experience and judgment can help to decide what to use as '^ equiv-
alent" loading for snow and wind.
The following two examples will serve to illustrate the general pro-
cedure of stress analysis of roof trusses. The reader will note that the
assumed data are purely arbitrary.
Example 6-3. On the basis of normal working stress for vertical load-
ing determine the maximum and minimum (if opposite in sign) stresses
 ANALYSIS OF ROOF TRUSSES 87

in all members of the roof truss shown in Fig. 6-6. The data for the
analysis are:

Bay length = 15 ft Span = 36 ft Rise = 12 ft

The dead load consists of

Weight of roofing, rafters, and purlins

= 16 lb per square foot of roof surface
Weight system = 1 lb per square foot of roof surface
of bracing
Weightof truss = 0.5 + 0.075L
= 3.2 lb per square foot of horizontal surface
Weight of ceiling = 10 lb per square foot of horizontal surface

The snow load is 10 lb per square foot of roof surface.

The wind load is 30 lb per square foot of vertical surface.

Duchemin formula,IfVn — ^v ~^

1
—
-\-
i

^^tt.
sin^ 6

The truss is anchored at both ends; wind reactions to be assumed

parallel; 333^ per cent increase in working stress allowed when wind
effect is included.
Load combinations are

(1) Dead + full snow on both sides

(2) Dead + wind from either side
(3) Dead + half snow on both sides + wind from either side

SOLUTION, (a) Dead-load stresses.

Length of AB (Fig. 6-6) = V4' + 6^ = 7.21 ft

Tributary area of roof surface per (15) (7.21) = 108.2 sq ft

panel =
Tributary area of horizontal surface per panel = (15) (6) = 90 sq ft
Panel load on top chord = (16 + 1) (108.2) (3.2) (90) = 2,130 lb +
Panel load on bottom chord = (10) (90) = 900 lb

The dead panel loads as computed above are shown in Fig. 6-7a.
The weight of truss and bracing has been assumed to act at the panel
points on the top chord. Some designers assume this load to be equally
divided between the top and bottom panel points, but this is generally
an unnecessary refinement. The stresses in all members of the truss may
be determined by either the algebraic or the graphic method. The
graphic solution has always been considered as standard procedure, but
it is often found that the algebraic method actually takes much less time,

especially when a computing machine is available. It should again be

noted that each method is self-checking. Because both the algebraic
88 ELEMENTARY THEORY OF STRUCTURES
21301b

D 21301b ^ 1 =
1 .jy"^^

\^
2 S/E \ ^
N^
>1 xc* \ ^ ^K
9001b 9001b 9001b 9001b 9001b
i
90901b 90901b

(a) Dead panel loads (6) Dead load stresses

Fig. 6--

and the graphic methods have been treated at length in Chap. 5, they
will not be shoTSTi here. The dead-load stresses are shown in Fig. 6-76.
(6) Snow-load stresses.

Panel load on top chord = (10) (108.2) = 1,080 lb

The snow panel loads are shown in Fig. 6-8a and the snow-load stresses
in allmembers of the truss are shown in Fig. 6-86.
10801b
o
10801b D 10801b S D
o '/
c/g ^\£
^
'^ \\
S
x^
•
!N

1 B/° \'^.+
T/fCv^
^^
/^ \ /^\s^
32401b 32401b
(a) Snow panel loads <6) Snow load stresses

Fig. 6-8

(c) Wind-load stresses. From the Duchemin formula,

2 sin d 2(2^'Vi3)
Pn = p = 30
1 -h sin- 1 + (2/Vl3)2
= 25.45 lb per square foot of roof surface
Panel load = (25.45) (108.2) = 2,750 lb

Since the truss is anchored at both ends and \s'ind reactions are assumed
to be parallel to the loads, only the stresses due to wind from the left
need to be determined because the stresses due to ^^'ind from the right
will be opposite-handed to those due to -wind from left. The stresses
due to wind from left are shown in Fig. 6-96.
{d) Combination of stresses. In comparing Figs. 6-76, 6-86, and 6-96,
it is seen that wind from either side will not cause stress in any member

opposite in sign to that of dead or snow loads. Inasmuch as there can

be no reversal of stress in any member, load condition (2) (dead plus
 ANALYSIS OF ROOF TRUSSES 89

wind from either need not be considered at all. The maximum

side)
stress in a member onthe basis of normal working stress for vertical
loading is, then, the larger of the two values: (1) dead plus full snow on
both sides, or (2) three-fourths of dead, plus half snow on both sides,

Wind (from left) panel loads Wind load (from left) stresses

Fig. 6-9

plus wind from either side. Table 6-1 shows the combination of stresses.
In examining this table, it is found that the %(!> + S/2 -}- Wl or Wr)
condition controls the maximum stress in members Be, Cc, and Cd only.

Table 6-1. Combination of Stresses

Mem-
ber
D S Wl Wr D+ S ..f + Max
stress
Wl or Wr Wl or Wr^

AB -13,660 -4,870 -5,840 -4,470 -18,530 -21,935 -16,450 -18,530

BC -10,920 -3,890 -4,700 -4,470 -14,810 -17,565 -13,170 -14,810
CD - 8,190 -2,920 -3,550 -4,470 -11,110 -14,120 -10,590 -11,110
Ab + 11,360 +4,050 + 7,020 + 2,070 + 15,410 + 20,405 + 15,300 + 15,410
be + 11,360 + 4,050 +7,020 + 2,070 + 15,410 +20,405 + 15,300 + 15,410
cd + 9,090 + 3,240 +4,540 +2,070 + 12,330 + 15,250 + 11,440 + 12,330
Bb + 900 + 900 + 900
Be - 2,730 - 970 -2,980 - 3,700 - 6,195 - 4,650 - 4,650
Cc + 2,420 + 540 + 1,650 + 2,960 + 4,340 + 3,260 + 3,260
Cd - 3,790 -1,350 -4,130 - 5,140 - 8,595 - 6,450 - 6,450
Dd + 6,960 + 2,160 + 3,300 + 3,300 + 9,120 + 11,340 + 8.510 + 9,120

Example 6-4. On the basis of normal working stress for vertical load-
ing determine the maximum and minimum (if opposite in sign) stresses
in all members of the Fink roof truss shown in Fig. 6-10. The data
for the analysis are:

Span = 60 ft Rise = 15 ft Bay length = 18 ft

The dead load consists of

Weight of roofing, rafters, and purlins

= 14 lb per square foot of roof surface
Weight of bracing system = 1 lb per square foot of roof surface
90 ELEMENTARY THEORY OF STRUCTURES
Weight of truss = 0.4 + 0.04L
= 2.8 lb per square foot of horizontal surface
No ceiling or other suspended loads

The snow load is 15 lb per square foot of roof surface.

The wind load is 25 lb per square foot of vertical surface.

Duchemin formula, Pn = p
1 + sin2 d

The truss is hinged at the left end and supported on rollers at the right
end; 33}-^ per cent increase in working stress allowed when wind effect
is included.

Load combinations are

(1) dead + fullsnow on both sides

(2) dead + wind from either side
(3) dead + half snow on both sides + wind from either side

Equivalent loading for snow and wind: 20 lb per square foot of roof
surface on the basis of normal working stress for vertical loading. (Com-
pare maximum stresses due to load combinations stated above with

Fig. 6-10

stresses due to dead load plus equivalent loading for snow and wind
condition.)
SOLUTION, (a) Dead-load stresses

Length of AB (Fig. 6-10) = Vil.bY + (3.75)2 = g 335 ft

Tributary area of roof surface per panel = (18) (8.385) = 150.93 sq ft

Tributary area of horizontal surface per panel = (18) (7.5) = 135 sq ft
Panel load on top chord = (14 + 1) (150.93) + (2.8) (135) = 2,640 lb

The dead panel loads and stresses are shown in Fig. 6-11. The stresses
are obtained by multiplying those in Fig. 6-12 by 2.64, because the
dead-load panel loads are 2.64 times the 1,000-lb panel loads indicated in
Fig. 6-12. Although a graphic solution might have been used, the
stressesshown in Fig. 6-12 are the results of an algebraic solution.
In the algebraic solution of the stresses shown in Fig. 6-12, it should be
noted that the slope of the top chord is 1 on 2, the slope of members BK,
CL, and DM
is 2 on 1, the slope of members CK, EM, and is 4 on 3, ML
 ANALYSIS OF ROOF TRUSSES 91
26401b

26401b -ooJU 26401b

26401b^r>t^ /^S^ X^^J:^ 26401b

2640 lb ^qC G 26401b

13201b -B' [i 13201b

" +1S.480 K +15.840 L

10,5601b 10,5601b

Dead panel loads and stresses

Fig. 6-11

1000

4000 4000
Fig. 6-12

and member CM is horizontal. The slope of EML is 4 on 3 because

Length of AC = 15 = y VS
^Yj f

Length of AL = AC = 18.75 ft

tan angle ELR

ER 15 ^

LR 30 - 18.75

The methods of joints or sections may be used interchangeably. The

stresses may be solved in the following order:
1. Joint A members AB and
: AK
2. Section 1-1: member CK
3. Joint K\ members BK and KL
4. Joint B member BC (one check available)
:

5. Section 2-2: member BE, ME, and LP

6. Joint Dmembers CD and DM
: (first use 2il/c = to find the ver-
tical component of DM)
7. Joint M\ members CM and LM
8. Joint C: member CL (one check available)
9. Joint E: check SF^ =
92 ELEMENTARY THEORY OF STRUCTURES
The computer is advised to indicate the horizontal and vertical com-
ponents of the stress in every member as shown in Fig. 6-12. A final
review to see that the conditions ZFx = and 2Fy = are satisfied at every
joint checks the correctness of the solution.
If used for the stresses in the truss sho\Mi in
the graphic solution is

Fig. 6-12, some be encountered after the points 1, 2, and 3

difficulty will
(Fig. 6-13a) are located on the stress diagram (Fig. 6-13c). Apparently
there are three unknown stresses at either joints C or L, but the difficulty
can be surmounted by use of the modified truss shown in Fig. 6-136.
From a comparison of the original Fink truss in Fig. 6- 13a with the modi-
fied Fink truss of Fig. 6-136, it is obvious that the stress in member E'-6

2
t" r
}
^ \
G'
c \^J ^F }

IT

B C J^^^h^ J'

A' R k?^V/"V"\y^13^
K L \ P Q \

' 2
L

(a) Fink truss ''original)

ib) Fink truss (with substitute members) (c) Stress diagram

Fig. 6-13

has been unchanged by the alteration (in the algebraic method, stress in
member £"-6 can be found by cutting section 2-2 and taking moments
about L). Thus points 1, 2, 3, and 6 (or 13, 12, 11, and 8) must assume
the same positions in the stress diagrams for the trusses shown in either
Fig. 6-13a or 6. The graphic solution, then, consists of (1) determining
points 1, 2, and 3 with reference to Fig. 6-13a or 6; (2) determining points
4' and 6 with reference to Fig. 6-136 only; and (3) with point 6 known,
determining 5, 4, and then 7 with reference to Fig. 6-13a. This pro-
cedure applies equally well to the right half of the truss.
(6) Snow-load stresses.

Panel load on top chord = (15) (150.93) = 2,260 lb

The snow panel loads and stresses due to these loads are shown in
 ANALYSIS OF ROOF TRUSSES 93
22601b

22601b . ir 22601b

22601b

+ 15,820 K +13,560 L

90401b
90401b
Snow panel loads and stresses

Fig. 6-14

Fig. 6-14. The stresses are obtained by multiplying those in Fig. 6-12
by 2.26.
(c) Wind-load stresses. From the Duchemin formula,
2 sin d 2(1/V5)
P 25
1 + sin2 d 1 _!_ ( l/V5)2
= 18.63 lb per square foot of roof surface
Panel load - (18.63) (150.93) = 2,810 lb

The due to wind from left and right are shown in Fig. 6-15a
stresses
and h.These stresses may be obtained by multiplying those in Fig.
6-16a and d by 2.81. It should be noted that the hinged support is at
the left end and the roller support is at the right end.

50301b

69101b 31401b
(a) Stresses due to wind from left

50301b
'f +1260 K +

31401b 69101b
(6) Stresses due to wind from right

Fig. 6-15
94 ELEMENTARY THEORY OF STRUCTURES

1118 (d) - (c) reversed 2460

Fig. 6-16

Although the graphic method could have been readily used, the stresses
shown in Fig. 6-16a were calculated by the algebraic method. In this
connection, it should be noted that the members shown by dotted lines
in Fig. 6-1 6a are not stressed because of wind from left.
Because there
are no loads acting at joints F, G, H, N, P, or Q, the triangle LEJ satisfies
It is also apparent that the stresses in
the requirements for truss action.
FN and HQ must be zero to satisfy

2F(in direction perpendicular to top chord) =

at joints F and H, Then stresses in NG and GQ must be zero to satisfy
 ANALYSIS OF ROOF TRUSSES 95

SF(in directions perpendicular to ENP or PQJ) = 0, respectively, at

joints A^ and Q. Finally, similar considerations at joints G and P show
that the stresses in GP and ENP must be In the graphic solution
zero.
points 7 to 13, inclusive, will coincide on the stress diagram.
By superimposing Fig. on Fig. 6-16a, Fig. 6-16c is obtained.
6-16?>

The stresses due to wind from left when the right support is hinged are
I shown in Fig. 6-1 6c, and the stresses due to wind from right when the
left support is hinged are shown in Fig. 6-16d, which is Fig. 6-16c reversed.

{d) Stresses due to dead load plus equivalent loading for snow and wind.

Dead-load panel load

2,640 lb =
Panel load due to equivalent loading for snow and wind
= (20) (150.93) = 3,020 1b
Total panel load due to dead and equivalent loads
= 2,640 -f 3,020 = 5,660 lb

The stresses due to dead load plus equivalent loading for snow and
wind are shown in Fig. 6-17. These stresses are obtained by multiply-
ing those in Fig. 6-12 by 5.66.

56601b

56601b 1t7 56601b

5660 lb n1 :^ /£> \^ i J^ 5660 lb

22,6401b 22,6401b

Stresses due to dead and equivalent superimposed loads

Fig. 6-17

(e) Combination of stresses. A

comparison of Fig. 6-11 with Fig. 6-15
shows that wind from either side will not cause stress in any member
opposite in sign to that of dead load. Load combination (2), therefore,
is not critical. Only load combinat'ons (1) and (3) need be considered.
In Table 6-2, the maximum stress in each member, which is the larger
oi D -{- S and ^i{D + S/2 + W), is compared with the stress due to
D -\- E, or dead plus equivalent loads. The D -\- E stresses appear
much larger than the maximum stresses as determined from the specified
load combinations. However, the equivalent loading for both snow and
wind, 20 psf in this case, is only 5 psf more than the snow load of 15 psf.
Thus the effect of wind on design is rather small. This is to be expected
because of the allowance of 333^ per cent increase in unit working stresses
when wind effect is included.
96 ELEMENTARY THEORY OF STRUCTURES

Table 6-2. Combination of Stresses

Mem- D-^ S ..f. Max

D S Wl* Wr* D +£
ber Wl stress
or Wr Wl or Wr^

AB -20.660 -17.690 -12.640 -12,640 -38,350 -42,150 -31,610 -38,350 -44.300

BC -19.480 -16.680 -12.640 -12,640 -36,160 -40.460 -30,340 -36,160 -41.770
CD -18.300-15.670 -12,640 -12,640 -33.970 -38.780 -29,080 -33,970 -39.240
DE -17.120-14.650 -12,640 -12,640 -31,770 -37,080 -27,810 -31.770 -36,700
AK + 18.480+15.820 + 15,710 + 10,680 + 34.300 + 42.100 + 31,580 + 34.300 + 39,620
KL + 15.840| + 13.560 + 12.570 + 7,540 + 29.400 + 35.190 + 26,390 + 29.400 +33.960
LP + 10,560 + 9.040 + 6.280 + 1,260 + 19,600 + 21,360 + 16,020 + 19.600 +22.640
BK-DM - 2.360 - 2.020 - 2.810 - 2.810 - 4,380 - 6,180 - 4,640 - 4.640- 5.060
CK-CM + 2.640 + 2.260 + 3.140 + 3,140 + 4.910 + 6,910 + 5,180 + 5.180,+ 5.660
CL - 4,720 - 4,040 - 5.620 - 5,620 - 8.760 -12,300 - 9,270 - 9,270-10,130
EM + 7.920 + 6.780 + 9.430 + 9,430 + 14.700 + 20,740 + 15,560 + 15,560 +16.980
LM + 5.280 + 4,520 + 6.280 + 6.280 + 9.800 + 13,820 + 10,360 + 10.360 +11.320

• The stresses recorded are the larger ones on either side of the center line since the truss will
symmetrically fabricated.

PROBLEMS
6-1. Rework Example 6-3 if the height of the truss is changed from 12 to 9 ft.
6-2. Rework Example 6-4 if the height of the truss is changed from 15 to 20 ft.
 CHAPTER 7

ANALYSIS OF BUILDING BENTS

7-1. General Description. In Chap. 6 it was stated that roof trusses

for buildings may rest on masonry walls or on columns along the sides
of the building. A roof truss attached to its supporting columns is
commonly called a bent. Two bents are shown in Fig. 7-1, one without
and the other with knee braces. Before the action of the knee brace
can be explained, however, it should be mentioned that the vertical faces
of such buildings are usually covered with siding supported on girts,
which are attached to the columns. The wind pressure on the vertical
face of the building is therefore carried by the girts to the columns, just

as the purlins bring their loads to the roof truss. There will be, then,
horizontal forces acting on the Avindward column. Because it is assumed
that the roof truss is pin-connected to the columns, the bent shown in
Fig. 7-la is unstable when wind is acting as indicated. However, if the
knee braces are connected to the inner faces of the columns, as shown in
Fig. 7-16, the bent becomes stable, because the structure cannot collapse
until the knee braces fail. The knee braces are therefore most essential
in obtaining lateral stability when the structure resists inclined or hori-
zontal loading. A bent as shown in Fig. 7-16 is often called a knee-braced
bent.
Depending on the structural details used, the columns of a bent may
be considered to be hinged, partially fixed, or fixed at the base. The
stress analysis of a bent with hinged column bases will be considered first.
For any kind of loading, there will be two unknown (horizontal and
97
98 ELEMENTARY THEORY OF STRUCTURES
vertical) reaction components at each hinge, or a total of four unknowns.
When the whole benttaken as a free body, there are, however, only
is

three independent equations of statics, and the structure is statically

indeterminate. (Any structure whose reactions cannot be determined
by the equations of statics alone is statically indeterminate. The treat-
ment of statically indeterminate structures will be discussed in the latter
part of this book.) By making a reasonable assumption about the values
of the horizontal reactions, an ''approximate" solution may be obtained
by statics. When the bent is subjected to vertical loads only, it is
usually assumed that the horizontal reactions at the hinges are zero; if
inclined loads are present, the horizontal reactions are assumed to be
equal. Once the reactions have been obtained, the analysis is straight-
forward. However, it should be noted that the two columns are three-
force members, while all other members are two-force members. As

far as a knee-braced bent is concerned, this statement is of utmost

importance and needs to be kept constantly in mind.
When the lower ends of the columns of a bent are fixed or partially
fixed, there will be three reaction components (a horizontal force, a
vertical force, and a resisting moment) at each support, or a total of six
unknowns. Three more conditions must therefore be provided or
assumed in addition to the three equations of statics, ^he first as sump-
tion, which is the same as that for a bent with hinged supports, is that
the horizontal reactions are zero for vertical loadings only and are equal
for inclined loadings. The other two assumptions involve the arbitrary
location of the points of inflection in the two columns, as shown in Fig.
7-2a. If the column bases are rigidly fixed, the points of inflection may
be assumed at /i/2 from the base wherein h is the distance from the foot
of the knee brace to the base of the column. If the column bases are
partially fixed (a condition comparable with the ordinary details of
structural-steel columns), the points of inflection may be assumed at h/3
from the base. Because the points of inflection are equivalent to hinges,
the analysis of the portion of the bent above the points of inflection will be
 ANALYSIS OF BUILDING BENTS 99

identical with that of a bent with hinged supports. The deformation

of the columns with fixed or partially fixed bases under the action of
wind pressure is shown in Fig. 7-26.
7-2. Methods of Analysis. In the analysis of a building bent, either
methods may be used. A bent may be subjected to
algebraic or graphic
dead, snow, and wind loads as well as crane or other special loadings.
Under dead and snow loads, the horizontal reactions at the bases of the
columns are assumed to be zero. Thus the columns are subjected to
direct axial compression only, the stresses in the knee braces are zero,
and the stresses in the members of the truss proper are identical with
those of a truss supported on masonry walls. Under wind loads, the
columns are three-force members subjected to shear and bending moment,
as well as direct stress. The knee braces and all other truss members
are two-force members subjected to direct tension or compression. If
the column bases are hinged and the horizontal reactions are assumed
to be equal, all external reaction components may be found by taking
the entire bent as a free body. If the column bases are partially or fully
fixed, the horizontal and vertical forces at the assumed points of inflec-
tion are first found by taking the portion of bent above the points of
inflection as a free body. If each column is taken as a free body, the
stress in the knee brace and the horizontal and vertical reactions of the
column at the end joint of the truss may be found. The shears and
bending moments, as well as direct stresses in both columns, can then be
determined. Finally, the stresses in all members of the truss may be
found by either the algebraic or the graphic methods.
Example 7-1. On the basis of normal working stress for vertical load-
ing, determine the maximum and minimum (if opposite in sign) stresses
in all members of the knee-braced bent shown in Fig. 7-3. The data
for the analysis are:
Bay length = 18 ft
The dead load consists of

Weight of roofing, rafters, and purlins

= 14 lb per square foot of roof surface
Weight of bracing system = 1 lb per square foot of roof surface
Weight of truss = 0.4 + 0.04L
= 2.8 lb per square foot of horizontal surface
No ceiling or other suspended loads.
The snow load is 15 lb per square foot of roof surface.
The wind load is 25 lb per square foot of vertical surface.

Duchemin formula, Pn = v
^
-^

1
— —sm^
;

-j-
r-x—
d

Assume points of inflection to be at 7.5 ft above base of columns.

100 ELEMENTARY THEORY OF STRUCTURES
Structural details at column bases are such that the points of inflec-
tion may be taken at one-third of the distance from the column base to
the foot of the knee brace. A 331:3 per cent increase in allowable working
stress is permitted when wind effect is included.
Load combinations are

(1) dead + full snow on both sides

(2) dead + wind from either side
(3) dead + half snow on both sides + ^^nd from either side

Equivalent loading for snow and wind: 20 lb per square foot of roof
surface on the basis of normal working stress for vertical loading. (Com-
pare maximum stresses due to load combinations stated above with
stresses due to dead load plus the equivalent loading for snow and wind
condition.)

Fig. 7-3

SOLUTION. The dimensions of the Fink roof truss in Example 6-4

and also the vertical loadings on it are identical with those of the Fink
truss of the knee-braced bent in this problem. When only vertical loads
act on the bent, the horizontal reactions are assumed to be zero; thus the
columns take direct stresses only and the dead-load stresses in the knee
braces are zero. The stresses in the other members of the bent due to
vertical loads are therefore identical with those of the truss supported on
walls. Thus the values shown in columns (2), (3), and (10) of Table 7-1
are taken from Table 6-2 of Example 6-4.
The direct stresses, shears, and bending moments in the columns, and
the direct stresses in all other members of the bent due to wind loads \vill
now be determined. The inclined panel loads on the truss due to wind
from the left are taken from Example 6-4 and are shown in Fig. 7-4. If
the girts are spaced at 7.5 ft apart, the horizontal panel load on the
column due to a wind pressure of 25 psf will be

(Bay length)(girt spacing) (wind pressure) = (18) (7.5) (25) = 3,375 1b

The horizontal panel loads due to wind are also shown in Fig. 7-4.
 ANALYSIS OF BUILDING BENTS 101

The complete stress analysis of the knee-braced bent due to wind loads
is shown diagrammatically in Fig. 7-6. A brief description of the order
of calculations follows.
1. Fig. 7-4: Determination of horizontal reactions. The wind load
of 1,688 lb at the column base goes directly into the support; no transfer
of any portion of it to the other support is made.

Total horizontal force (exclusive of wind load at column base)

= 1,688 + (3) (3,375) + (4)(2,810)(l/->/5) = 16,840 lb
Therefore

Horizontal reaction at the left support = 1,688 + 3^(16,840)

= (1,688 + 8,420) lb
Horizontal reaction at the right support = 3^-2(16,840) = 8,420 lb

k-* ^ H2 = 8420

Fig. 7-4

2. Fig. 7-6a: Portion of bent above the points of inflection as free body.
From 2(ilf about the left point of inflection) = 0,

72 = 6,925 lb

From S(il/ about the right point of inflection) = 0,

Fi = 3,128 lb
Check by i:Fy = 0.

3. Fig. 7-66 and c: Lower portions of columns as free bodies.

From 2i¥ = (Fig. 7-66),

ilf 1 = 63,150 ft-lb

From Sikf = (Fig. 7-6c),

M2 = 63,150 ft-lb
102 ELEMENTARY THEORY OF STRUCTURES

V3-7136 V4 - 13,283

1688 '^=^8403 323,3

6416
3375 4008 ao^iosK ^4= 16,840

3375

3375 8420 8420

3128
r 6925
id) (e)

3128 6925

I
t8420 1 8420

1688 I
Hi = 8420+1688 H2 = 8420
I

to Af, = 63.150 |Oa/2 = 63,150

Vi = 3128 V2 = 6925
(6) (c)

71 36 13,283
in
1688' r8403
1

3375
5010 --
+
1-A
3375 in

3375 i
CO
+
1688 +
^

63.15oC 1688+8420
3128
(/)

^ 7/4 = 16.840

13,283
 ANALYSIS OF BUILDING BENTS 103

4. Fig. 7-Qd: Upper portion of left column as free body.

From XM about top = 0, H component of knee brace = 5,010 lb

Then V component of knee brace = 4,008 lb
and Stress in knee brace = 6,416 lb
From XH = 0, Hz = 8,403 lb
From SF = 0, Vz = 7,136 lb
Check by 2(il/ about the point of inflection) = 0.

5. Fig. 7-6e: Upper portion of right column as free body.

From 2(il/ about top) =0, H component of knee brace = 25,260 lb

Then V component of knee brace = 20,208 lb
and Stress in knee brace = 32,348 lb
From 2i7 = 0, H^ = 16,840 lb
From 27 = 0, V,= 13,283 lb
Check by 2(3/ about point of inflection) = 0.

Fig. 7-6/ and g\ Free-body, shear, and bending-moment diagrams

6.

of both columns. The sign convention for the shear and bending-
moment diagrams is that of treating the columns as beams when viewed
from the right side.
7. The methods of joints and sec-
Fig. 7-6/i: Stress analysis of truss.
tions are freely used. have been determined, it will be
After all stresses

advisable to check each joint by using the equations 2Fx = and

2Fj/ = 0. A graphic solution for this part of the problem may be used
to advantage.
Table 7-1 shows the various combinations of stresses. Values in col-
umns 4 and 5 are taken from Fig. 7-6. Note that the Wr (wind from
the right) stresses in the left half of the bent are equal to the Wl (wind
from the left) stresses in the right half of the bent. The maximum
stresses in column la are all of the same sign as those of dead-load stresses;
naturally the D -j- S/2 +W condition, not the D -\- W condition, con-
trols. In column 76, only minimum stresses with signs opposite to those
of maximum stresses are listed; these minimum stresses are, of course,
due to the D -{• W condition without the inclusion of S/2. The D -\- S
condition is entered in column 6. Maximum values shown in column 9
are the larger of the values listed in columns 6 and 8a. The minimum
values in column 9 are identical with those in column 86.
A comparison columns 9 and 10 indicates that the use of equivalent
of
vertical loading to replace both snow and wind effects is unsafe in the
case of building bents. Column 9 shows that many members are sub-
jected to stress reversals. Thej^ should be designed accordingly. As
indicated in column 10, the use of an ''equivalent" loading does not
reveal the true nature of these reversals of stress.
104 ELEMENTARY THEORY OF STRUCTURES
o o c o c ^ c c o o o o" -^-^
Ot^'^O(NO'^OOc0cC00(N o c
&i3

+ O cct^C^t^OC5:00;^0-^C5cc
Tf^ClcCOrOC^iOiOiOOo"^'
0;c
'^*•'<f^;cc^:^c(N
•s-^ ^^^ (m'(N
c^c^
Q IIII4- + II +I++I++
OQ c g^ :8 o o o o o o
-^ ^: — c^ -^
s t^ o O^ os^ O »0 lO X o ;

a Tf CO »0 -"T (N
OQ 1—1

+ 4- : 1 1 + 1 ++
1
1

•-< ^
8J
S8^^888§^iSi§§8g
ro—'C;t^rC'^o;cci^c;'<r<N(N-o-c
oo-occ — c^TTXioot^c^-^Cic
6 'q' ^5rc:cr:rC(M-^ „^^(M^_
Tj-cr.

1 1 I+4-+I++I4-+I ~ 1

-ooooc
1 1

c
oo
00 t^
• • -o
w3
• -o
t^ ^Tf r? oi TT j^
o
•

X o
•

's 'o o t^
:
^^ c; • cr. lO lit
3
o --r r: irt TT (N
o ;^
+ ++ — .2

oooooooooooooooc
: 1 1 : 1 1

IA>
*j
ore — o TT TT Ci TT
re Lt
'o
cr.
—
c: i>—X -r
X— — ^ <N c^
t^ ->c
cr. •>:? cr. cr.
-.c
-.:5
li::
2
— 0;S'^TrxiCTrXir:oi>(N'^;cc c
5 ;^ '

rcrc(NC^<N— <
^^^c^^^ o
>
m 1 1 1 I+ + 4-I+H-I++I 1 1

_ 3:

1
H c
O
•.q
OO
^ c^
• • -o
a O
?3
•OO'-OO
X — O <N <N
C5 t^ rc TT t^
"i ^ ft:
9. <N in;

CO c^"^ :s Tj- i> c: <M 1

b s + ++ 1 1 + ++ is
^
o 1 1

S=^
/
z ^
(>» '-M
*^
Q Zi K^^
^ S
E^
•<
2
X
4- ^
^
|||^|2|||||||||| >> c,Q o
^\oi (NO^rec:'^t^«o^OTfrc::;(M(N(M
— (N — r?(N(M
1 t> TTrr:rC(N(N ^ o ^m5 ^ fc
o
O
5 + o
1 1 1 I+ + +I++I++I 1 1
m ^\s t<
— Q
— C
«
-\ \
-c ri-
.
f,-
1

t>-
^
^M^^^
lO-^t-t-OOOXOO-^OO
--m^ "-m^

ce — r;t-^t'T:^^:c:c^. t-t-x
S«^ Sw^ iw^ -^^^ >^,^ '^00' NwK >«|0r -^^^ \^^
oc
^.^

;rc
s^
E
o
H + ^ xocc — TrcJCJ'^TTTfX'^c: c:c-
'
S
1.2 roccrororc(N— -- ^ —
n
<
.'^

IIII+ + +I++I4-+ II
tx

E-
SS2ESSS^g^5S9?§9
t^t^'-'-'t>;x?5
?:i5ii^:(^^cr 04
c
a;
:i>.
^^
jo
. X
H- ++ +II1 1 4-III + I s
ioiSr^r^?cc:o — SS2?:?.
—>^c:->:?r^ut(NX'- ?5^?
— xcr. r^'^ — —
-J ^^ c
^ Tf COCO00X(M-^l>(NXrCl> — xcct^c: X
77 I++I I+4-I+ + +I
1

Q,^,,-^^^,^,^-,^^^-,^^^,—^,-^,-«,-^^
1
o
"^
cioot^oo52^c522^x?5"'^Tr
-sCO-^c:;xicooc<i(NOt^iC cc Jo
^^^
c
V2
s
7777+7+I++I++ _ X
OOOOOOOOOOOOOOOC 1 1

OXOC^XTrcr;-0^^(N(NX o
^^ Orrrc — '<rXiOccocct^c:(N -o-c
^Oir: ^
"3
Cj
£i o" Ci' x' t>-'' x" O o' CsT (N (N T*'' t-' iC C C .2
7777777i++i4-H- 77
o E
Si

o
^-v
^^ cQoqg^e^^^^'^^^^^^
^CQOQ^^^i<Cj;j^;ii.^S^-^ft; •
s 2:
 ANALYSIS OF BUILDING BENTS 105

PROBLEM

Fig. P7-1

7-1. Determine the maximum and minimum (if opposite in sign) stresses in all
members of theknee-braced bent as shown on the basis of normal working stress for
vertical loading. Assume points of inflection to be at 10 ft from base of columns.
Other required data for the analysis are the same as given in Example 7-1.
 CHAPTER 8

INFLUENCE DIAGRAMS

8-1. General Introduction. In the foregoing chapters involving the

analysis of beams and been assumed in all cases that the
trusses, it has
was known. Shear and bending-moment
position of the applied loads
diagrams have been determined for beams under one given condition of
loading. Methods for determining the stresses in the members of a truss
under one given condition of loading have also been studied. Many-
structures, however, are subjected to the frequent passage in either direc-
tion of moving uniform or concentrated loads. A crane-runAvay girder
in an industrial building is subjected to moving wheel loads. The beams
which support conveyor systems transporting raw materials or finished
products in factories are subjected to systems of moving loads. Girders
or trusses in bridges carrying highway traffic are subjected to the moving
wheel loads of heavy trucks or some equivalent loading system which
simulates the streams of cars or trucks passing over the structure. Gird-
ers or trusses in bridges carrying railway traffic are subjected to the
wheel loads of heavy locomotives followed by uniform train loads of
indefinite length. These examples illustrate some types of moving loads
for which structures must be designed.
In designing a beam or truss to carrymoving loads, the position of
these loads on the structure must be somehow determined so that the
shear or bending moment at a section, or the stress in a member of the
truss, will be the maximum which may ever happen. In this connection
it must be emphasized that the position of loads which may cause maxi-
mum shear at a section will not necessarily cause maximum bending
moment at this same section, or a condition of loading which causes
maximum bending moment at one section may not cause maximum
bending moment at some other section. When a maximum is being
sought, whether be shear, bending moment, or stress in a member, the
it

first is to determine the critical position of the moving

consideration
loads. A study of influence diagrams will provide an understanding and,
in many cases, the best solution for this problem.
8-2. Definition. Preliminary to the study of the effect of a system of
moving loads which may consist of both concentrated and uniform loads,
106
 INFLUENCE DIAGRAMS 107

it will be desirable to consider first just one moving concentrated load.

For instance, the effect of a single moving concentrated load on the shear
at section C of simple beam AB (Fig. 8-la) is to be found. As a matter
of convenience, the magnitude of the moving concentrated load is
assumed to be unity. When the unit load
is on segment CB at a distance Xb from B

(Fig. 8-16), the left and right reactions are

Ra = Xb/L and Rb = {L — Xb)/L and the
(a)
shear at C is, by considering tIC as a free
body, Xb

Xb
Vc = Ra= -\-
^ L-^b
fn Rb \

or, by considering BC as a free body,

(6)

- Rb= +1 L Xb Xb
-{-1 = +
k=^ Rb- it
The shear at C, Vc = -\-Xb/L, due to a (c)

unit load at a distance of Xb from B, is

plotted directly under the position of the

unit load, as shown in Fig. S-ld. When
the unit load is at B, Vc = 0] when the
unit load is at an infinitesimal distance to
the right of C, Vc =
-\-h/L. These values
are plotted at B
and C and the points Bi
and C3 are connected by a straight line
(Fig. S-ld). When the unit load is on segment ^C at a distance Xa from
A (Fig. 8-lc), the left and right reactions are Ra = {L — Xa)/L and
Rb = Xa/L. Considering AC as a free body, the shear at C is

Vc = Ra-1= ^^^" - 1 = -
If

Or, with BC as a free body.

Vc= -Rb = -^
When the unit load is at A, Xa = 0, Vc = when the unit load is at an
0;
infinitesimal distance to the left of C, Xa = Vc = —a/L. These
a and
values are plotted at A and C and the straight line A1C2 is drawn as
shown in Fig. 8-lc?. The enclosed diagram A1C2C3B1 of Fig. 8-lc? is

called the influencediagram for shear at C. Thus the shear at C due to

the unit load at any position on the span is equal to the ordinate on the
influence diagram directly under the load.
108 ELEMENTARY THEORY OF STRUCTURES
From the preceding discussion, it is seen that an influence diagram is

constructed to show the variation in the effect of a single moving unit

load on some function at any section of a structure (such as shear at C
in Fig. 8-la). Influence diagrams may be constructed for a function such
as reaction, shear, bending moment, deflection, or stress in a member.
The ordinate to the influence diagram shows the desired function due to a
single unit load at the position of the ordinate. Usually influence dia-
grams are constructed by first calculating ordinates which represent the
value of the desired function due to a unit load as it moves across the
span. These ordinates are calculated and plotted under the load at each
critical position of the load, and then lines connecting the extremities of
successive ordinates thus calculated
enclose the influence diagram for
the given function.
It is well to note that a shear or
bending-moment diagram shows
the effect of stationary loads on
shears or bending moments at all
sections in a structure, while the
shear or bending-moment influence
diagram shows the effect of one
ib) Influence diagram for left reaction
moving unit load on the shear or
bending moment at the section for
which the influence diagram has
been constructed.
8-3. Influence Diagrams for Re-

(c) Influence diagram for right reaction actions on a Beam. Let it be re-
Fig. 8-2 quired to construct the influence
diagrams for the left and right re-

actions of the simple beam AJ5 as shown in Fig. 8-2a. A unit load is

placed at a distance x from the left support. By taking moments about

point B, the left reaction is found to be Ra = {L - x)/L, and by taking
moments about point A, the right reaction is Rb = x/L.
In the influence diagram for the left reaction, the value of

-
Ra = y = L X

is plotted directly under the unit load, which is now at x from point A.
Since i/ is a linear function of x and its expression is appUcable between
a; = and x = L, it will only be necessary to compute the values of y
at X = and x = L. The influence diagram for Ra is then obtained by
drawing a straight line connecting the upper extremities of the ordinates
at points A and B. Thus,
 INFLUENCE DIAGRAMS 109

When X = 0, y = —— = —— =
j j
1

When X = Lj y = —— = —— =
=: =:

ju Li

The influence diagram for the left reaction is plotted as ^^11^251 as in

Fig. 8-26. Similarly the influence diagram for the right reaction is

A1B1B2 as shown in Fig. 8-2c. At A the ordinate isy = x/L = 0/L =

and at B the ordinate is y = x/L = L/L = 1.
For concentrated loads Pi and P2 (not shown) at points 1 and 2 in
Fig. 8-2a, the left reaction is

Ra = Piyi + P22/2

This may be explained by the fact that, since yi is the left reaction due
to a unit load at point 1, the left reaction due to Pi at point 1 is Piyi.

Similarly, the left reaction due to P2 at point 2 is P22/2. The left reaction
due to Pi and P2 is thus Ra = Pit/i + P22/2.
The above influence diagrams may also be used to calculate reactions
due to uniform loading. Suppose a uniform load of intensity w lb per
lin ft (not shown) is applied between points 1 and 2 on the beam shown

in Fig. 8-2a. The reaction at A, due to a load w dx is dRA = yw dx, in

which y = {L — x)/L. Thus,

R. I yw dx = w y dx

It is noted that / y dx is the area of the influence diagram between

ordinates t/i and z/2. Therefore the left reaction may be obtained if the
area of the influence diagram covered by the uniform load is multiplied
by the intensity of the uniform loading.
Example 8-1. Given a beam 25 ft long which overhangs the left sup-
port by 5 ft as shown in Fig. 8-3a. Construct the influence diagrams for
Ra and Rb- Compute the maximum upward and downward (if any)
reactions due to (a) a moving uniform Hve load of 400 lb per lin ft and
(b) two concentrated loads of 10 kips each at 4 ft apart.

SOLUTION. First consider the influence diagram for Ra. When the
unit load is at a distance x from Bj

/?A = + ^ for ^ a; ^ 25

At point C: a; = 25 and Ra = +^^0 = +1.25

At point A: X = 20 and Ra = ^'^%o = +1.00
At point B: x = and Ra = +%o =
no ELEMENTARY THEORY OF STRUCTURES
The influence diagram for the left reaction is plotted as C1C2B1 as
shown in Fig. 8-36.
Next consider the influence diagram for Rb- When the unit load is

between points A and B at a distance x from A ,

x
Ri^ " for ^ .r ^ 20
'
20
At point .4: x = and Rb = +920 =
At point B: x = 20 and R. + 2^0
2Q = +1.00

WTien the unit load is between points C and A at a distance x from A,

— X
Ri = forO ^ ^• < 5
20
At point A: x = and Rb= --920 =
At point C: x = 5 and Rb= --^20 = 0.25

The influence diagram for the right reaction is C1C2A1B2B1 as shown

m Fig. 8-3c.
To ascertain the eflfect of mo\ang uniform or concentrated loads on the
reactions, the loading positions for maxima must be determined. For

(c) Influence diagram for Rb

Fig. 8-3

simple systems of loadings as given in this example, such positions can

be easily found by inspection of the influence diagrams. These positions
are summarized Inasmuch as the entire influence diagram
in Fig. 8-4.
for the left reaction above the base line, there is no possibility of a
is

downward left reaction at any time. Thus, in Fig. 8-46, no loads are
sho^vn. For a maximum upward reaction at the left support due to
 INFLUENCE DIAGRAMS 111

uniform load, the uniform load must cover the entire length. This
reaction may be found by multiplying the intensity of the load and the
entire area of the influence diagram. The two concentrated loads will
produce a maximum left reaction when the loads are placed over the
highest possible influence ordinates. These positions are shown in Fig.
8-4a. At the right support, however, the uniform load should cover
only the 20-ft segment between the supports for maximum upward reac-
tion and the o-ft overhang for maximum downward reaction. The two
concentrated loads are placed so that the positive or negative ordinates
are, numerically and respectively, the largest possible (see Fig. 8-4c and d).

lOk 10k
0-4k/ft 1.
4' I

i » I

'
;
^
I
A
,
t
,^
I
^
t
*
I
^
;
^
4
^

20'
i
^
I
^
t
^
i
^
I
^
U
t tb
o
I
/-.t__i__
c
20'
TlB

(aj For maximum upward reaction at A

-aB
(6) For maximum downward reaction at A
10k „ 10k
o.4k;ft 4

i^W I W I I I I M iB C
^
I ,

i±i.
Cc) For maximum upward reaction at B
10k ., 10k
0.4 k/ft
C ^"^~^^ Z.B tzl1^
Ct j,B

(d) For maximum downward reaction at B

Fig. 8-4

The actual values of the maximums may be computed by considering

the loadings show^n in Fig. 8-4, or by use of the influence diagrams, viz.,
R = w (area of influence diagram covered) or R = ^Py. The beginner
is advised to find all required maxima by both methods. For instance,
the maximum downward reaction at the right support (Fig. S-M) due
to the moving uniform load is

(0 4) (5)^
Rb = /9W9n\ ^ ^-^^ ^P downward

By the use of the influence diagram,

R^ = (0.4) ^-Mi^ = 0.25 kip downward (check)

Due to the two moving concentrated loads,

P =
Rb —
-(10)(1)
+ —
(10)(5)
-^-^ = 3^1-
kips downward
, .

or Rb = (10)[0.25 + 0.25(>^)] = 3 kips downward (check)

112 ELEMENTARY THEORY OF STRUCTURES
The required maxima are summarized in the follo-sving table. The
reader may check each value by both methods as illustrated above.

Max Ra, kips Max Rb, kips

Loading
(+) Upward ( — Downward (+) Upward
) ( — Downward
)

Uniform load -f- 6.25 + 4 -0.25

Concentrated loads. +23 + 18 -3

8-4. Influence Diagram for Shear in a Beam. The influence diagram

for shear at section C in the beam AB shown in Fig. S-oa will be con-
Unity
structed. For a load of unity at dis-
tance X from A, the reactions are
Ra = {L - x)/L and Rb = x/L. If
X < a, the shear at C is i^^ — 1 ; or

L -
(a)

^2r
The shear influence ordinate at C is

— a/L. It is also noted that, for this

condition of loading, the shear at C is
equal numerically to the right reac-
tion. Consequently, the portion of
the right-reaction influence diagram
(6) Influence diagram for shear at C
Fig. 8-5
between A and C may be used as a
shear influence diagram for C; how-
ever, to conform to usual sign conventions, this line is plotted as Aid
with negative ordinates. li x > a, the shear at C is the left reaction and

L - X
F(

Thus, for the portion of the beam between C and B, the left-reaction
influence diagram is also the shear influence diagram. At C the shear
influence ordinate is (L —
a)/L as shown in Fig. 8-56.
The diagram A1C2CZB1 is the shear influence diagram for section C.
It will be noted that the ordinate C1C3 is the positive shear at C when the
unit load is applied at an infinitesimal distance to the right of C and the
ordinate C1C2 is the negative shear at C when the unit load is appUed
at an infinitesimal distance to the left of C.

Example 8-2. A beam 50 ft long rests on its left support and extends
10 ft beyond the right support as shown in Fig. 8-6a. Construct the
influence diagram for shear at a section midway between the supports.
 INFLUENCE DIAGRAMS 113

Compute the numerical maximum shear at C due to (a) a moving uni-

form live load of 400 lb per lin ft and (6) two concentrated loads of
10 kips each at 4 ft apart.
SOLUTION. For loads between A and C, the shear at C is the left
reaction minus the load, and for loads to the right of C, the shear at C
is the left reaction.
B
20'
C
^
40' 10'

(a)

+ 0.50

(6) Influence diagram for shear at C

Fig. 8-6

When the unit load is at a distance x from A and x < 20,

Vc-Ra-1- -^q- - 1 - -
40
When X > 20,
40 - a:
Vc = R.
40
When X > 40,

- a: - 40
Vc = Ra-^
40
Thus 7c = at x = 0; Vc = -0.50 at x = 20; Vc = +0.50 at a; = 20;
Fc = at a: = 40; and Vc = —0.25 at a; = 50. When values of
Vc are plotted at points A, C, B, and D, the shear influence diagram
AiC2CzBiD2Di of Fig. 8-66 is obtained. It is noted that loads on seg-
ments AC and BD cause negative shears at C while positive shears are
produced by loads on segment CB.
If it is assumed that the moving uniform load may be broken into
segments of any length, numerically the maximum shear is the negative
shear at C when the uniform load covers portions AC and BD of the
span, as shown in Fig. 8-7a. This shear is (Fig. 8-7a)

Vc ^ Ra - (0.4) (20) = 5.5 - 8 = -2.5 kips

From the influence diagram,

Vc = 0.4 (area of AidCs + area of B^D.D^)

= 0.4 [-(H) (0.50) (20) - H(0.25)(10)]
= —2.5 kips (check)
114 ELEMENTARY THEORY OF STRUCTURES

I I 4
0.4k/ft
I I

20'
W I I

20' B
i
'
0.4k/ft
I I

10'
4 W,
40'
> i?n=6.5k
(a)

10k 10k
16'

A
B"
40'
^A=9k i2B=llk
ib)

Fig. 8-7

The two concentrated loads should be placed as shown in Fig. 8-76 to

cause maximum shear at C.

Vc = Ra = 9 kips

Or, from the influence diagram,

Vc = ^Py = 10(0.50) H- 10(0.50) (i%o) = 9 kips {check)

8-5. Influence Diagram for Bending Moment in a Beam. The influ-

ence diagram for bending moment at section C in the beam AB oi Fig.
8-8a is shown in Fig. 8-86. For a load of unity at distance of x from A
in Fig. 8-8a, the reactions are Ra = {L — x)/L and Rb = x/L. If

X < a, the bending moment at C, by considering AC as the free body, is

Mc = Racl — l{a — x) = (a) l{a — x)

= > -a)

The same expression can be obtained by considering CB as the free body.

Thus,

Mi Rb(L a)=l{L a)

This is the equation of line A1C2. Note that Mc = when the unit
load is at X = 0, and Mc = {a/L) (L — a) when the unit load is at a: = a.

li X > a, by considering AC as the free body.

Mi RAa = r (^ x)

or, by considering CB as the free body,

Mc = Rb{L - a) - 1{x - a)

{L — a) — l{x — a) = J {L — x)
 INFLUENCE DIAGRAMS 115

This the equation of line C2B1. Note that Mc = {a/L){L — a) when

is

X = and Mc =
a, when x = L. It will be observed that A1C2 and
B1C2 have a common ordinate (a/L)(L — a) at C.
It happens that the influence diagram for bending moment at C in
simple beam ^J5 is identical with the bending moment diagram for a
unit load at C. The interpretation P= unity
of these two diagrams, however, is

entirely different ; the influence ordi-

nate always gives the bending mo- L
ment at C as the unit load moves
across the span, while the bending-
moment diagram shows the bending
moments at various sections due to a
fixed unit load at C. Thus, for con-
venience, the influence diagram for
bending moment at C may be con-
structed by placing a load of unity
at C (the critical section) and then
(6) Influence diagram for bending moment at C
dra^ving the bending-moment dia-
Fig. 8-8
gram for this load.
Example 8-3. A beam 33 ft long rests on the left support and over-
hangs the right support 6 ft, as shown in Fig. 8-9a. Construct the
bending-moment influence diagram for a section from the left
(7 at 9 ft
support. Compute the maximum positive and negative bending moment

c B
9' 6'
IS'

27'

(a)

(6) Influence diagram for bending moment at C

Fig. 8-9

at C due to (a) a moving uniform live load of 400 lb per lin ft and (h)
two concentrated loads of 10 kips each at 4 ft apart.
SOLUTION. This influence diagram may be constructed by placing
unity at critical points A, C, B, and D and, in each instance, calculating
116 ELEMENTARY THEORY OF STRUCTURES
the bending moment at C. The value of each bending moment will be
plotted under the load as shown in Fig. 8-96. Thus:

For unity at A : Ale = and ordinate ^ = i

For unity at C: Ale = +6 ft and ordinate C1C2 = +6 ft

For unity at B: Ale = and ordinate Bi =
For unity at D: Ale = — 2 ft and ordinate D1D2 = —2 ft
It is to be noted that, in the case of influence diagrams for reaction
or shear, the influence ordinate is the ratio of the reaction or shear to
the moving load and therefore is merely an abstract number without an
attached dimensional unit. The bending-moment influence ordinate,
however, is the ratio of the bending moment at the section to the moving
load; or dimensionally speaking, this ratio is FL/F = L. Thus the
ordinates C1C2 and D1D2 in Fig. 8-9 are -|-6 ft and —2 ft, respectively.

10k 10k
9' 4' 14'
0.4 k/ft

=±i
I I

A ^ ^ ^ ^ ^ i ^ i ^ i -~r-r
B B'
27'
27' 6'

k 1123/27 k 84/27 k
5.4 .4 k
(a) For maximum positive bending moment at C
10k 101

0.4 k/ft

-a
2/3k

(6) For
B

maximum
r
22/3k
T 226/2^

negative bending
k
moment at C
2226/27 k

Fig. 8-10

The positions which the moving uniform or concentrated loads must

take to cause positive or negative bending moments at C may be ascer-
tained by inspection of the influence diagram A1C2B1D2D1. These posi-
tions areshown in Fig. 8-10. Note that, for maximum positive bending
moment at C, one of the two equal concentrated loads is placed at C
and the other on the long segment CB. Certainly any movement of
these two loads toward the right will give a smaller positive bending
moment at C as they are descending down the slope C2B1 of the influence
diagram. Any movement toward the left will also give a smaller positive
bending moment at C because the left load comes down on a steeper slope
C2A1, while the right load goes up on the flatter slope B1C2', or, the loss
because of the left load is more than the gain from the right load. Thus
any further movement of the two concentrated loads in either direction
from the position shown in Fig. 8-10 will cause a decrease in the positive
bending moment at C. Consequently the critical position for these
loads has been determined. Had the system of concentrated loads been
more complicated (more loads at varied spacings), it would have been
 INFLUENCE DIAGRAMS 117

difficult to determine by simple inspection the position which these loads

should take on the span to cause a maximum positive bending moment
at C. Problems of this nature will be rigorously treated in the next
chapter.
The actual values of the maximum positive or negative bending
moment at C may be found by applying statics to the free-body diagrams
of Fig. 8-10 or by the use of the influence diagram. Computations by
both methods are shown below.

For the uniform load,

Max +il/c = (5.4) (9) - M(0.4)(9)2 = 32.4 kip-ft

or = (0.4) (area of A1C2B1) = (0.4) (3^) (6) (27) = 32.4 kip-ft
Max -Mc = -(0.267) (9) = -2.4 kip-ft
or = (0.4) (area of B1D1D2) = (0.4) [-3^(6) (2)] = -2.4 kip-ft

For the two concentrated loads,

Max +Mc = {^^%7){9) = ^2% = 106.67 kip-ft

or = 10(6) -f 10(6) (i^ls) = 60 + 46.67 = 106.67 kip-ft
Max -Mc = -(^%7)(9) = -^% = -26.67 kip-ft
or = -(10) (2) - 10(2) (%) = -20 - 6.67 = -26.67 kip-ft
8-6. Influence Diagram as a Deflection Diagram. Influence diagrams
for reaction, shear, or bending moment as described in the preceding
articlesmay be determined on the basis of the conception that influence
diagrams are deflection diagrams.
1. To obtain the influence diagram for a reaction, remove the support

giving resistance for this reaction and introduce a unit displacement in

the direction of the reaction. The area enclosed between the original
and the final positions of the beam is the required influence diagram.
Thus, for the simple beam AB shown in Fig. 8-1 la, the influence dia-
grams for Ra and Rb are given by the enclosed area between the original
position (1) and the final position (2), as shown in Fig. 8-116 and c.

Similarly, for the overhanging beam AB shown in Fig. 8-12a, the influence
diagrams for Ra and Rb are shown in Fig. 8-126 and c. For the canti-
lever beam AB shown in Fig. 8-1 3a, the influence diagram for Ra is
obtained by sliding the fixed support vertically upward a unit displace-
ment as shown in Fig. 8-136.
To
2. obtain the influence diagram for shear at a section, cut the
beam at the section and lift the cut end at the right a unit displacement
relative to the cut end at the left, without introducing relative rotation
at the section. Thus in Figs. S-lldand S-12d, A1C2 and C^Bi are parallel
in order that there is no relative rotation at C. Similarly, the influence
diagram for Vc in the cantilever beam AB is obtained by cutting the
118 ELEMENTARY THEORY OF STRUCTURES
beam at C and lifting the cut end at the right a unit distance, as shown
in Fig. 8-13c. In this case, AiCi must remain horizontal; therefore C2-B2
must also be horizontal in order that there is no relative rotation at C.
3. To obtain the influence diagram for bending moment at a section,
insert a hinge at the section so that there no moment resistance at the
is

section and introduce a unit relative rotation at the section. Thus, in

Fig. 8-1 le, if there were no relative rotation at C2, A1C2 would have gone
straight to B2. C2B2 rotates around C2 for 1 radian to position C2B1.

(b) Influence diagram for /?^

Do
(c) Influence diagram for Rq

(1) fii

(c) Influence diagram for Rg

(d) Influence diagram for Vq

C2

(e) Influence diagram for Mq (e) Influence diagram for Mq

Fig. 8-11 Fig. 8-12

displacements are small, CiBi = C2B1, and B1B2 = 1 radian

If all vertical
times CiBi. Thus C1C2 = {B,B2/L){AiCi) = ab/L. It should be noted
that, although 1 radian is defined to be about 57.3° in trigonometry, it
must be regarded as a very small unit of measure so that the length B1B2
may be equal to that of an arc with radius equal to C2B1 or CiBi. Note
also the influence diagrams for Mc in Fig. 8-12e and for Ma and Mc in
Fig. S-13d and e.

This method of obtaining influence diagrams by use of deflection

diagrams has been demonstrated in a few cases, and the influence dia-
 INFLUENCE DIAGRAMS 119

grams thus found are seen to be identical with those determined analyt-
ically. A formal proof of this principle now seems to be in order. If it
can be proved that the ordinate y in Fig. 8-14c is equal to Mc in Fig.
8-146, then by definition A1C2B1 must be the influence diagram for Mc-
The underlying principle for making this proof is this If the force system :

acting on a structure is in equilibrium, and if this structure undergoes a

change of shape or position without resulting internal stresses or strains,
, ^ ^ the total external work done by the
elements (forces and moments) in
the balanced force system must be
(a) zero, because, if there is any exter-
nal work done, by the law of con-

^i (1) P= unity

(6) Influence diagram for i?^

DB
r i^ B2

1^ (1)
(a)

(c) Influence diagram for V^ P= unity

M,
0)
'9$ _L_
f Rb\
(6) Statics

(d) Influence diagram for Af^^

Cl (1) Bi

Ci (1)

(e) Influence diagram for Mq (c) Geometry

Fig. 8-13 Fig. 8-14

servation of energy, such work must be stored in the beam as internal

elastic energy. Now, when the balanced force system {Ra, Rb, P, Vc, Vc,
Mc, Mc) through the change in shape from position
in Fig. 8-146 goes 1

to position 2 as shown in Fig. 8- 14c, the total external work done is W

W = Ra{0) + 7^;,(0) - Py + Vc{C,C2) - Vc(C,C,) + Mc(d) + Mc{<t>)
= -Py + Mc{d ^ 0) -Py + Mc(l radian) •= -(!)(//) + McW
=
Thus
y = Mc
Note that the work done by the downward force P in going through
the upward displacement y is negative. Similar procedures can be used
120 ELEMENTARY THEORY OF STRUCTURES
to prove that influence diagrams for reaction and shear are deflection
diagrams as described above in items (1) and (2).
The relationship between influence diagrams and deflection diagrams
is important and can be used as an auxiUary method either to get a

preUminary sketch of or to make a final visual check on any influence

diagram.
8-7. Influence Diagrams for Simple Trusses. Usually two trusses
are used to carry moving loads, with one truss on each side of the traffic.
In Fig. 8-15 is shown a through truss bridge composed of two six-panel
Pratt trusses joined together by floor beams and other bracing in the
transverse direction (some not shown). The
stringers are simple beams
supported on the beams and with spans equal to the panel length
floor
of the truss. The moving wheel loads are carried by the bridge floor
to the stringers, which are supported by the floor beams. The floor beams
carry the stringer reactions to the panel points of the trusses. For

Stringers

End floor beam

Floor beams
Fig. 8-15

instance, an axle with two wheel loads of P each, acting on the stringers in
the third panel as shown in Fig. 8-15, causes two loads of Ph/d each on
floor beam LiL'o and two loads of Pa/d each on floor beam L3L3. As
shown almost invariably placed symmetrically
in Fig. 8-15, stringers are
on the floorbeams; consequently, in this case loads of Ph/d are trans-
ferred to joints L2 and Lg. Likewise loads of Pa/d are transferred to
joints L3 and L3.
This truss bridge is supported at the four corners as shown in Fig. 8-15
and requires the use of end floor beams to support the exterior ends of the
stringers in the end panels of the bridge. Sometimes a bridge is designed
so that both the trusses and the exterior ends of the stringers in the end
panels rest directly on the abutments or piers. If such is the case, end
floor beams are not required.
When structures, particularly railway and highway bridges, carry
moving loads which may occupy any position on the span, it is necessary
to determine the position of the load system causing maximum tensile
and/or compressive stress in any one member of the truss. It is to be
 INFLUENCE DIAGRAMS 121

noted that the position which causes maximum stress in one member
may not produce maximum stress in any other member. Obviously, the
loading condition which causes maximum tensile stress in one member
will not produce maximum compressive stress in this same member.
Influence diagrams provide a convenient method for developing criteria
which may be used to determine these critical loading conditions. Once
the critical condition of loading has been determined, the methods
explained in Chap. 5 or the influence
diagram itself may be used to calculate
Pa
the maximum stress in any member.
Stringer
8-8. Influence Diagram between
Panel Points of a Truss. It will be
proved that the influence diagram be-
tween panel points of a truss must be
bounded by a straight line. Let Fig.
8-166 represent the influence diagram
for the stress in some member of the
bridge truss shown in Fig. 8-1 6a.
Assume that a unit load at panel point
2 causes a stress equal to 2/2 in this
(b) Influence diagram for stress in a member
member and a unit load at panel point
Fig. 8-16
3 causes a stress equal to 2/3. If two
w^heel loads of P
each act on symmetrical stringers in the third panel at
distances a and h from the adjacent panel points, components Pb/d and
Pa/d, respectively, will be transmitted to panel points 2 and 3 on each
truss. Therefore, the stress y in the member is

Ph Ph P(d - h)
XPy (?/2) + -r (2/3)
d
2/2 + yz

= p Vz + (2/2 - yz)
d

If a straight line is drawn connecting the upper ends of the ordinates

2/2 and 2/3, the expression within the brackets is seen to be the value of the
ordinate under P This demonstration indi-
in the influence diagram.
cates that the influencediagram between panel points of a truss is always
bounded by straight lines. This statement should be constantly kept
in mind when influence diagrams for trusses are being determined because,
once the influence ordinates at all the panel points have been calculated,
the influence diagram is constructed by drawing straight lines through
the extremities of these ordinates.
8-9. Influence Diagrams for Reactions on a Truss. The influence
diagrams for reactions on a truss with end floor beams are different from
those for a truss without end floor beams. When end floor beams are
122 ELEMENTARY THEORY OF STRUCTURES
used, the loads on the floor system are transmitted to the abutments
all

or piers through the end pedestals. Thus, if end floor beams are used
in the bridge shown in Fig. 8- 17a, the reaction at the end pedestals is
unity when a placed on symmetrical stringers at their
pair of unit loads is

junction with the end floor beam. When the unit load moves to the
panel points b, c, d, e, f, or g, the end reaction decreases linearly from
unity to zero. Thus, as shown in
Fig. 8-176, the influence diagram for
the left reaction of a truss with end
1,^ floor beams is the same as the reac-
tion influence diagram for a simple
beam.
When end floor beams are not used,
the exterior ends of the end stringers
rest directly on the abutments or
piers. For the truss shown in Fig.
(6) Influence diagram for left reaction 8-1 7a, when the moving unit load is
(with end floor beams)
to the right of panel point h, the left
reaction is the same whether or not
end floor beams are used. Thus the
portion of the influence diagram be-
h ci di gj f^
tween panel points b and g in Fig.
(c) Influence diagram for left reaction 8- 17c is identical with that drawn for
(without end floor beams)
this segment of the truss in Fig. 8-176.
Fig. 8-17
When the unit load is on the end
panel, however, say at a distance x {x < p) from a, the left reaction of
a truss is equal to that of a beam equal in length to that of the truss
minus the stringer reaction at a, thus

L — X (p — x) _ X
Ri ~
p p
The left reaction may also be found by distributing the panel-point
load at b to the left and right ends of the truss. The panel-point load at
b is x/p, and the reaction at the left end is

X (L - p)
Ri =
p L P L
When X = 0, the unit load acts at the end of the stringer and goes directly
to the abutment or pier; therefore, its effect on the reaction at the left

pedestal of the truss is zero. Thus the influence diagram for the left

reaction of a truss without end floor beams is constructed as shown in

Fig. 8-17c. It will be noted that the triangle 010262 is an influence
diagram for the left reaction of a simple span p.
 INFLUENCE DIAGRAMS 123

Example 8-4. The exterior ends of the end stringers of the bridge
shown in Fig. 8-18a rest directly on abutments. Draw the influence
diagram for the left reaction of one truss. Compute the maximum left

reaction due to (a) a moving uniform load of 4 kips per lin ft on the bridge
and (5) two moving axle loads of 60
kips each, at 8 ft on centers.
SOLUTION.
for the left reaction is
The influence diagram
found to be as
/ / /\
h 1 2 3
6 at 30' = 180'
4 5 '

show^n in Fig. 8-186. When the unit

1

load on the stringer is at panel point

0, all of it goes directly into the abut-
ment and the left reaction on the truss
is zero. The left reaction becomes
+%, +%, +%, +%, +M, and zero (6) Influence diagram for jRq

as the unit load moves from panel

points 1 to 6. Inasmuch as it has
been proved that the influence curve
betw^een panel points is always linear,
the correctness of the influence dia- 60 k

gram shown in Fig. 8-185 is assured. i?0=150k i26=150k

(c)
To cause a maximum left reaction
on the truss, the moving uniform load
of 4 kips per lin ft should cover the
entire 180 ft of the bridge, or 2 kips
per lin ft to each truss. The value
of this reaction may be computed by
(d)
several different methods. One pro- Fig. 8-18
cedure is to multiply the area of the
influence diagram by the intensity of the uniform loading, or in this case
by 2 kips per lin ft. Thus

^0 = (2)[3^(%)(180)] = (2)(75) = 150 kips

A second method is to subtract the exterior reaction on the end stringer

from the left reaction of an equivalent simple beam 180 ft long. Thus
^0 = M(2)(180) - M(2)(30) = 150 kips

A third procedure is to determine the left reaction of the truss from the
actual panel-point loads on the truss as shown in Fig. 8-18c; thus

Ri >^(60)(5 -h 4 4- 3 + 2 + 1) = 150 kips

From an inspection of the influence diagram, it is seen that the two

moving concentrated loads should take positions in the second panel as
shown in Fig. 8- 18c?.
124 ELEMENTARY THEORY OF STRUCTURES

(a) By the influence-diagram method,

R, = 30^1 + S0y2 = 30(%) + 30(J^)(i4^^5o) = 48.67 kips

(h) By the simple-beam method (truss replaced by equivalent simple

beam),

30(142)+30(150)^^gg,
no
jgQ

(c) By the panel-loads method (the

loads are replaced by panel concentra-
tions as sho\s'n in Fig. 8-18c?),

= 8(4) + 52(5)
48.67 kips
Ro
6

Although it may seem longer than

the other procedures, the last method, (6) Influence diagram for shear in panel 2-3

in which the actual panel loads are

determined, gives the best picture of
how the loads on the floor system are
transferred to the panel points of the
truss
C D

f ^1 ^2 ^2 I

^
(b) Influence diagram for shear in panel cd

Fig. 8-19 Fig. 8-20

8-10. Influence Diagram for Shear in a Parallel-chord Truss. The

influence diagram for shear in the panel cd of a parallel-chord truss ag
(Fig. 8-19a) is shown in Fig. 8-196. For loads at panel points to the

left of panelthe shear in the panel is equal to the right reaction;

cd,

therefore, the portion of the right-reaction influence

diagram aiC2, plotted
as shown, may be used for shear in panel cd. The ordinate C1C2 is -m/n.
 INFLUENCE DIAGRAMS 125

For loads at panel points to the right of panel cd, the shear in the panel
is equal to the left reaction; therefore, the portion of the left-reaction
influence diagram gid^, plotted as shown, may be used for shear in panel
cd. The ordinate rfic?2 is -\-{n — m — l)/n. Joining C2 and d^ b}' a
straight line will give the complete influence diagram for the shear in
panel cd as a ic 2c? 2^1.
Because CiCoi and didoi are similar triangles, the distances Cii and idi
in Fig. 8-196 may be found by simple proportion. Thus

= —- C1C2
=
m/n m
Cii Cidi p (m/n) —m— —
C1C2 + ,
,
dido
,
-\- [(n l)/n] n 1

and

_ , did2 _ {n — m — l)/n _ n — m — \
'^' ~ ""'^^
C1C2 + did2
~ ^ {m '/?) -f [in - m- l)/n] ~ n - 1 ^

At this time it aaiII be interesting to note that the distances igi and iai
are, respectiveh^ n times the distances idi and ici. The proof follows:

iQi [in — m (n - 1)] p + {n - m- l)p — 1 \n - 1) +

1) 1
= n
id, [{n - m - 1) ^{ti - l)]p l/(/i - 1)

iai ^ [m {n - l)]p + mp ^ l:\?i - 1) + 1 _

and "
ici [m{n-l)]p 1(^-1) """
The point i in the influence diagram for shear in panel cd is sometimes
called the ''load divide" because it is seen that, if a unit load is placed at i,

the shear in the panel is zero. In other words, the point i divides the
span aiQi in the same ratio as it does the panel length Cidi so that the
left reaction at a is equal to the panel concentration at
c. It is to be
noted that loads to the right of point i produce positive shear (tension in
member Cd) in the panel, while loads to the left of i produce negative
shear (compression in member Cd) in the panel. Inasmuch as the stress
in diagonal Cd equals the shear in the panel cd times sec 6, influence
diagram aiCod^gi may be used as a stress influence diagram for member
Cd.
It ^^•ill be noted that the ordinates to diagram aiCid-iQi ma}' be con-
structed by placing unity first at c and then at d. With unit}' at c, the
shear in panel cd is —m/n, which is plotted as CiCo. For unity at d,
the shear in the panel is +(/! — m — 1) >i, which is the value of ordinate
dido.
Example 8-5. Draw the influence diagram for the shear in panel 2-3
of the parallel-chord truss shown in Fig. 8-20. Compute the maximum
positive and negative shears in panel 2-3 due to (a) a moving uniform
load of 2 kips per lin ft on each truss and (6) two moving concentrated
loads of 30 kips each at 8 ft apart on each truss.
126 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. The influence diagram for the shear in panel 2-3 as shown
in Fig. 8-206 can be obtained by placing the unit load at panel points
0, 1, 2, . 6 in succession and computing the shear in the third panel.
. . ,

The distances 2-i and 2-3 are found by dividing the panel length 2-3, or
25 ft, into two parts in the ratio of % to %, or 2 to 3; thus,

(2-^) =10 ft
2-' =(25)

3-- = (25)
(2-1:3)
= 15 ft

The critical positions of the uniform load to cause maximum positive

or negative shears in panel 2-3 are shown in Fig. 8-20c and d. These
values can be found by multiphdng the area of the influence diagram
covered by the uniform load by the intensity of loading. Thus

Max -h72-3 = +(2)(M)(%)(90) = -f-4o kips

Max -72-3 = -(2)(M)(%)(60) = -20 kips

These shear values can also be found from the free-body diagrams of the
truss as shown in Fig. 8-20c and d] thus

Max -I-F2-3 = Ro - P2
2(90)2 2(15)2
= 54 — 9 = 4-45 kips
(2) (150) 2(25)

Note that Ro is equal to the moment of P2 to P5, inclusive about the

right support divided by the span of the truss. Also the moment of
P2 to P5 about the right support is equal to that of the 90 ft of uniform
load about the right support. P2 in Fig. 8-20c is the portion of the
uniform load on panel 2-3 transferred to panel point 2. Similarly,

2(60)2 2(10)'
Max -72-3 = -(Pe - P3) = -
[2^ - ^^J = -(24 - 4)
= -20 kips

The critical positions for the two moving concentrated loads on the
span to cause maximum positive or negative shears in panel 2-3 are
shown in Fig. 8-20e and /. From the influence-diagram method,

Max +72.3 = ^Py = (30)(%) + 30(%)(6Jf5) = -f-28.4 kips

Max -72-3 = ^Py = -(30)(%) - 30{H)i^Ho) = -18.4 kips
From the panel-loads method (Fig. 8-20e and/),

Max +72-3 = Po =
2P4 + 3P3
o
6
2(9.6) + 3(50.4)
+ 28.4 kips
 INFLUENCE DIAGRAMS 127

Max — 72-3 = -R& = -

Pi + 2P2
6
9.6 + 2(50.4)
= -18.4 kips

8-11. Influence Bending Moment at a Panel Point in the

Diagram for
Loaded Chord of a Truss. The influence diagram for bending moment
at panel point c of the truss in Fig. 8-2 la is shown in Fig. 8-216. This
influence diagram is exactly the same as the influence diagram for bending
moment at point c in a simple beam ag. For unity on the segment ac
at a distance x from a, the bending
moment at c is

S2X
Mc = R2S2

For unity on the segment eg, the

B C D E F '

/
/\ / V / \\/ \ \ \
(6) Influence diagram for bending moment at L2
b c d e f \
^
S2=(n-m)p 1

L'^ np
,

,2 k/ft
i i A i ?7ri"~rKT TPTT

«•= ^" «< 8k

'^"^ I

120k t? 120k
(c)

30 k 0,30

01 Cl g^
f 50 k 10 k ^
(b) Influence diagram for bending moment at c id)

Fig. 8-21 Fig. 8-22

bending moment at c is

= Si(L — x)
Mc = RiSi

By substituting = Si a; in either expression for Mc it is seen that the

common ordinate C1C2 is S1S2/L.
It just happens, as noted before in Art. 8-5, that the influence diagram
aiC2gi may be obtained by constructing the bending-moment diagram
for point c in beam ag. For unity at c, the left reaction is S2/L and the
bending moment at c is S1S2/L.
Example 8-6. Draw the influence diagram for bending moment at L2
ID the loaded chord of the truss shown in Fig. 8-22a. Compute the
128 ELEMENTARY THEORY OF STRUCTURES
maximum bending moment at L2 due to (a) a moving uniform load of
2 kips per on each truss, and (6) two moving concentrated loads of
lin ft

30 kips each at 8 ft on centers, on each truss.

SOLUTION. The influence diagram for the bending moment at L2 is
shown in Fig. 8-226. The ordinate under L2 is found by placing the
unit load at L2 and computing the bending moment at L2', thus

Influence ordinate at L2 = /?o(48) = (%)(48) = 32 ft

For maximum bending moment at L2 it is seen that the moving uni-

form load covers the entire span as shown in Fig. 8-22c. The two moving
concentrated loads should be placed as shown in Fig. 8-22c^ to give the
maximum bending moment at L2. As in previous illustrative problems,
the value of the bending moment may be computed by the use of the
influence-diagram, the equivalent simple-beam, or the panel-loads
method.
Thus, by the influence-diagram method, for the uniform load,

Max M at L2 = (2) (M) (32) (144) = 4,608 kip-ft

and for the concentrated loads,

Max M atL2 = SPi/ = (30) (32) + 30(32) (s^e) = 1,840 kip-ft

By the simple-beam method, for the uniform load,

Max M at L2 = ^^1^ (48) - ^^ = 4,608 kip-ft

and for the concentrated loads,

Max MatL, = A8R0 = (48) ^^Ml^+^^M = i,840 kip-ft

By the panel-loads method, for the uniform load,

Max M at L2 = [120(2) - 48(1)] (24) = 4,608 kip-ft

and for the concentrated loads.

Max M at L2 = 48/^o = 48(3.^(10) + M(50)] = 1,840 kip-ft

It will be noted that the maximum compressive stress in members

U1U2 orU2Uz due to any moving load system is equal to the
of this truss
maximum bending moment at L2 divided by the height of the truss.
8-12. Influence Diagram for Pier or Floor-beam Reaction. The com-
mon support for the two simple structures shoAvn in Fig. 8-23a must take
the right reaction of span Li and the left reaction of span L2. The reac-
tion influence diagrams are constructed as shown in Fig. 8-236. It will
 INFLUENCE DIAGRAMS 129

be noted that this diagram is similar to the bending-moment influence

diagram for point e in a simple beam with span Li + L2 as shown in Fig.
8-23c. The ordinates in Fig. 8-23c when multiplied by (Li + L'^IL\Li
become numerically equal to those in Fig. 8-236. Thus, to cause a
maximum pier reaction at e, the critical position for a system of moving
loads is the same as that for maximum bending moment at point e in a
simple beam with span (Li + L2). In fact, it is seen that the value of
this reaction is equal to (Li + L2)IL\Li times the value of the bending
moment at e.

In Fig. 8-24a, the floor beam

must support the right
at panel point 5
reaction from the stringer in panel ah and the left reaction from the
stringer in panel he. The reaction influence diagram for both stringers
as shown in Fig. 246 is therefore the influence diagram for the floor-beam

(6) Influence diagram for pier reaction at e (6) Influence diagram for floor beam reaction
or stress in member Bh
«2

(c) Influence diagram for bending moment at e (c) Influence diagram for bending moment at
in simple beam of span = L^ + L2 center of simple beam of span = 2p

Fig. 8-23 Fig. 8-24

reaction, and, in this case, also for the stress in member Bh. This
diagram is similar to the bending-moment influence diagram for the
mid-point of a simple beam with a span equal to two times the panel
length, or 2p, as shown in Fig. 8-24c. The ordinates in Fig. 8-24c are
p/2 times those in Fig. 8-246. Thus the maximum floor-beam reaction
or the maximum member Bh due to any system of moving loads
stress in
is2/p times the maximum bending moment at the mid-point of a simple
beam with span 2p.
On occasions the conversion of one influence diagram to another as
illustrated above is a useful device.
8-13. Influence Diagram Bending Moment at a Panel Point in the
for
Unloaded Chord of a Truss. influence diagram for bending moment
The
at panel point C in the unloaded chord of the truss shown in Fig. 8-25a is
drawn in Fig. 8-256.
As has been previously shown, for a load of unity at a distance x
130 ELEMENTARY THEORY OF STRUCTURES
from a on segment ac, the bending moment at C is

X
Mc = R2S2 = S2
J
and for unity on segment dg, the bending moment at C is

nr =
Mc n
RiSi =
{L
^
—
J
—X)
- Si

Thus, if the unit load is outside the panel cd, the bending moment at C
the same as though C were on the loaded chord. In Fig. 8-256, aiCzgi

ABC
is

1 «2

D E

V viA vA\/\/\
F

y
/W\
b c p d e f
'
L
(a)

01 ci Ci di 8\

(6) Influence diagram for bending moment at C

Fig. 8-25

is the influence diagram for bending moment at C if C were on the loaded

chord. The ordinates C1C2, did^, and CiCs are
ac
C1C2 = R2S2 = J-
S2

77 n dg
did2 = iiiSi = -j- Si

S1S2
and C1C3 =

In Art. 8-8 it has been shown that the influence diagram between adjacent
panel points of a truss is composed of straight-line segments; consequently
diagram aiC2d2gi is the influence diagram for bending moment at panel
point C. It is to be noted that, had point C been vertically above point c,
the influence diagrams for bending moment at C or c would be identical.
 INFLUENCE DIAGRAMS 131

In an actual problem, it is suggested that ordinates C1C2, did2, and C1C3

as shown in Fig. 8-256 be all computed and their correctness be checked
by verifying that aiCid and gid2Cz are straight lines. The usable part
of the influence is, of course, only aic^digi.
diagram
Example 8-7. the influence diagram for bending moment at
Draw
panel point Uz in the unloaded chord of the Warren truss shown in
Fig. 8-26a. Compute the maximum bending moment at Uz due to (a) a
moving uniform load of 2 kips per Hn ft on each truss, and (h) two moving
concentrated loads of 30 kips each at 8 ft on centers on each truss.

Ui U2 U3 Wt U5 Ue

/\/ \/\/'\/ \A
LiH L2 I3 L, L6
16'

6 at 24' - 144'

b n c

(b) Influence diagram for bending moment at C/3

/\
*f/i i\^ I A/\At /\i\i
i/i
/\2k/ft/\
\A i\i i/i i\i
7\
y'l i\^

48k Aa\,
48k 48k
/Ifilr
48k
^fiL- 48k
4ftL- I

120 k 120 k
(c)

312/3 k 28V3k

Fig. 8-26

SOLUTION. The influence diagram for bending moment at C/s is shown

in Fig. 8-266. The ordinate he is the bending moment at Vz due to unity
at L2; thus
he = 7^6(84) = 3^(84) = 28 ft

Similarly, eg is the bending moment at Uz due to unity at Lz, or

eg = Roi^O) = 1^(60) = 30 ft

The ordinate at point / in Fig. 8-266 is the bending moment at a point

60 ft from the left end of a 144-ft simple beam, or

Ordinate at point/ = ^^^ = 35 ft

Verification that aef and fgd are straight lines may now be made.
132 ELEMENTARY THEORY OF STRUCTURES
From an inspection of the influence diagram, it is seen that for maxi-
mum bending moment at Uz the critical loading conditions for the moving
uniform load and the system of concentrated loads, respectively, are as
shown in Fig. 8-26c and d. The maximum bending moment at L\ may
be conveniently found by the influence-diagram method or the panel-
loads (Fig. 8-26c and d) method. Due to the uniform load,
Max M at U3 = (2) (area of aegda)
= (2)[H(28)(48) + K(28 + 30)(24) + M(30)(72)]
= 4,896 kip-ft
or Max M at U, = (120) (60) - 48(12 + 36) = 4,896 kip-ft

Due to the two concentrated loads.

Max M at Uz = XPy = (50) (30) + 10(28) = 1,780 kip-ft

or = (30) (30) + 30[30 - %4(2)] = 1,780 kip-ft
or
Max M at Uz = i?o(60) - 10(12) = (31%) (60) - 120 = 1,780 kip-ft

It is to be noted that, for any live-load system, the maximum tensile

stress in member L2L3 is equal to the maximum bending moment at Uz

divided by the height of the truss.

8-14. Influence Diagram for Stress in a Web Member of a Truss with
Inclined Chords. A truss with inclined chords is shown in Fig. 8-27a.
Point is the intersection of the chords BC and he. The stress in web
member Be equals the moment about of the forces on either side of a
section through members BC, Be, and he, divided by the arm t.

For unity on segment ah, at a distance x from a, the compressive stress

in Be is conveniently found by taking moments about with the right
side of the section as a free body; thus

Be= ^e-f-L
-R,-^ =
xe +L
-L-^~
If the length ah is substituted for x in the above equation, ordinate 6162
in the influence diagram for stress in member Be is found to be

, , ah e -{- L
^'^^=
-L—r
If 62^1 is prolonged to 02 directly under 0, ordinate O1O2 is found to be

e e -\- L
O1O2 = T
L t

For unity at a distance x from a and on segment eg, the tensile stress
in Be may be found by taking moments about with the left side of the
 INFLUENCE DIAGRAMS 133

section as a free body; thus

e L-xe
Bc= +R,-^= +-^-^
If the length ac is substituted for x in the above equation, ordinate C1C2

is found to be •

cge
C1C2 = + L t

Now points 02, C2, and gi can be proved to lie in the same straight line by
showing that
C1C2 ^ CiQi
O1O2 OiQi

Because the influence diagrams between adjacent panel points must be

bounded by straight hues, it is seen that a 16202^1 is the influence diagram

(6) Influence diagram for stress in member Be

Fig. 8-27

for the stress in member Be. In a problem like this, the critical ordinates
6162 and C1C2 are first calculated by placing unity at panel points h and c,
and a verification of their correctness can be made by showing that 62^1
and giC2, when both prolonged, intersect under the moment center 0.
Point i may be located by dividing the panel length he in the ratio of bib
and C1C2.
Example 8-8. Draw the influence diagram for stress in member U1L2
of the truss shown in Fig. 8-28a. Compute the maximum tensile and
compressive stresses in member U1L2 due to (a) a moving uniform load of
2 kips per lin ft on each truss, and (6) two moving concentrated loads of
30 kips each at 8 ft on centers on each truss.
134 ELEMENTARY THEORY OF STRUCTURES
A
V-l ^^3 U^

o-

^( + 20/i8secei)

"^
9.23'

c(- sec^l) 14.77'

5/i8

(6) Influence diagram for stress in member V^L2

iJo- 85.21k I
Pi -9.09 k

2 k /ft

7?fi= 7.67 k

i?0=38V3k

«6=8V3k

Fig. 8-28

SOLUTION. The influence diagram for the stress in member U1L2 is

shown in Fig. 8-286. The ordinate he is the compressive stress in member
C/1L2found by placing unity at joint Li and taking moments about point
with the right side of section A A as a free body.

„ 10 panels . . .
he = -R6'ir'-—r' (sec di)
6 panels
= -3^(1%) sec 01= -%8sec0i

The minus sign for he means compressive stress. The ordinate cf is the
tensile stress in member UiU found by placing unity at joint L2 and
 rsTLrxxcE diagrams 135

taking moments about point O with the left side of section -4.-1 as a free
bodT.

= +%(^ sec ^, = +5^8 aec Bi

The positive sign for cf means tensile stre^. The calculated values of
ordinates he and be verified by showing that point g. the inter-
cf should
section of the prolongations of ea and df. is directly under point 0.
The distances hi and ci in the influence diagram may be found by divid-
ing the panel length be in the ratio of [>e to rf or 5 to 8. Thus

"=<24)^g = 9^3ft

"= (24)
5^ = 14.77 ft

The critical conditions for the moving uniform load, to cause maximum
tensile and c(Mnpressive stresses in member UiLj, are shown in Fig. 8-28c
and dy respectively. Thus
Max tension in L'lLi = +(2) (area of ifd)
= -f (2)(K)(9f 8 sec ^,)(110.77)
= -S-49.23 sec Oi kips
Max compression in I'lL* = — (2) (area of aei)
= -(2)(M)(^8sec«i)(33^)
= -9.23 sec ^ikips
Or. from Fig. 8-28c and d.

^
= — 4 - Pi(5)
AT
Max tension m
* r- r
- •

L iLj ^ ^-^ sec ^i

b
4 So .21) - 5(9.09)

= —49.23 sec ^i kips

-
AT
Max compression mlIT
iL* = ^•(10) ^^
—^
D
Pi(6)
^^^ sec ^i

_ (7.67)(10^ - (3.oo)(6) „.
— ^ sec oi
,
o
= -9.23 sec ^1 kips

The positions which the two moving concentrated loads should take
for maximum tensile and compressive stresses in member UiLj are shown
in Fig. 8-28^ and /. Thus
Max tension in liU = ^Py = -30(S{s sec ^i)(l -h s^ge)
= —25.55 sec ^i kips
Max compression in =
UiU ZPy = -30(^8 sec ^,)(1 -h ^^4)
= - 13.88 sec Oi kips
136 ELEMENTARY THEORY OF STRUCTURES
Or. from Fig. 8-28e and /,

Max tension in U1L2 = H ^— sec ^i

= + ^^^^^(^^ sec ^1 = +25.55 sec Bi kips

Re{lO)
Max compression in f/iLj sec 01

(8M)(iQ)
sec 01
6
-13.88 sec 01 kips

PROBLEMS
8-1. Given a simple beam 30 ft long, construct influence diagrams and compute the
maximum values due to a moving uniform load of 1 kip per ft and a movable concen-
trated load of 30 kips for (a) the left reaction, (b) the shear and bending moment at a
section 10 ft from the left end, and (c) the shear and bending moment at the mid-point
of the beam.
8-2. Given a simple beam 40 ft long, construct influence diagrams and compute the
maximum values due to a moving uniform load of 0.8 kip per ft for (a) the left reaction
and shear at sections 5 ft, 10 ft, 15 ft, and 20 ft from the left end, {b) the bending
moment at sections 5 ft, 10 ft, 15 ft, and 20 ft from the left end.
8-3. A cantilever beam 15 ft long is fixed at the right end. Construct shear and
bending-moment influence diagrams for sections 5 ft, 10 ft, and 15 ft from the free
end. Calculate maximum shears and bending moments at these sections in this beam
due to a moving uniform load of 0.6 kip per ft and a movable concentrated load of
10 kips.
8-4. A beam 40 ft long is supported at the left end and 30 ft from the left end.
Construct influence diagrams and compute the maximum values due to a dead load of
0.5 kip per ft, a moving live load of 0.8 kip per ft, and a movable concentrated load of
12 kips for (a) the reactions, (b) the shear at sections 10 ft and 20 ft from the left end,
(c) the bending moment at sections 10 ft, 20 ft, and 30 ft from the left end.

8-5. A beam 65 ft long is supported at 10 ft from the left end and at 15 ft from the
right end. Construct influence diagrams and compute the maximum values due to a
dead load of 0.8 kip per ft, a moving live load of 1.2 kips per ft, and two movable
concentrated loads of 10 kips each spaced 10 ft apart, for (a) the reactions, (6) the
shear at the left support, at 10 ft and 20 ft from the left support, and at the right
support, (c) the bending moment at the left support, at 10 ft and 20 ft from the left
support, and at the right support.

Fig. P8-6

8-6. Given the Pratt truss as shown, construct influence diagrams and compute the
maximum and minimum (if any) values due to a moving uniform load of 1.2 kips per
 INFLUENCE DIAGRAMS 137

ftand a movable concentrated load of 20 kips for (a) the shear in panels LoLi, L1L2,
and L2L3; (b) the bending moment at panel points Li, L2, and L3; (c) the stress in
member UiLi.
Given the Warren truss as shown, construct influence diagrams and compute
8-7.
the maximum and minimum (if any) values due to a moving uniform load of 0.64 kip
per ft and a movable concentrated load of 18 kips for (a) the shear in panels LqLi,
LiL-2, and L^Lz] {b) the bending moment at panel points Li and L2; (c) the bending
moment at panel points Ui, U2, and U3.

V 2^ —u, U,
J6
,

Ui .

Lo/ / \\ \ X 7pi \"° ^

Li L2 L3 L4 L^ Le \L 7
7 at 20' = 140'

Fig. P8-7 Fig. P8-8

8-8. Giveiy the Parker truss asshown, construct influence diagrams and compute
the maximum and minimum any) values due to a dead load of 0.5 kip per ft, a
(if

mo\-ing live load of 1.5 kips per ft, and a movable concentrated load of 20 kips for
(a) the stress in members ^1^2 and L2L3, (5) the stress in member UiLi, (c) the stress
in member U2L2 and U2L3.
 CHAPTER 9

CRITERIA FOR MAXIMA: MOVING LOADS

9-1. General Introduction. Beams and trusses are often designed to

carry a system of moving loads. Theoretically these loads may occupy
an number
infinite of positions on the structure, but obviously, if adequate
provision is made for the maximum condition of loading, the designer
need not be concerned about the other and lesser conditions of loading
on the structure. Thus when a system of moving loads such as a uni-
form load or a series of concentrated loads, or both, passes over a beam
or a bridge floor supported by trusses, the effect on the shear or bending
moment at a section of the beam, or the stress in a member of the truss
is different for various positions of the loading. Before its maximum
effect on a specific function such as, for instance, the bending moment at
a section, can be calculated, the critical position of the load system on the
structure must be determined. In general these critical positions are
different for maximum effects on different functions.
In the preceding chapter, influence diagrams for various functions in
beams and trusses have been studied. As illustrated in Chap. 8, the
critical position of a uniform load or one or two concentrated loads can
usually be determined by simple inspection of the appropriate influence
diagrams. For more complicated conditions of loading, however, it is
difficult to find the critical positions by ordinary inspection or by other
simple cut-and-try methods. In this chapter, definite criteria for deter-
mining the positions of loading necessary to cause maximum effects will
be developed.
9-2. Maximum Reactions and Shears in Simple Beams: Uniform
Loads. A glance at the reaction influence diagrams in Fig. 8-2 indicates
at once that a uniformly distributed load produces a maximum left or
right reaction when the beam is fully loaded, i.e., when the uniformly
distributed load covers the full length of the beam.
The maximum positive and negative shears at some section C, at
distance x from A in the simple beam AB (Fig. 9-la), due to the passage
of a uniformly distributed load, be determined. The influence
will
diagram for shear at section C is constructed as shown in Fig. 9-16.
From this influence diagram it is seen that segment CB should be covered
138
 CRITERIA FOR MAXIMA! MOVING LOADS 139

with uniform load to cause the maximum positive shear at C, and seg-
ment AC should be covered to cause maximum negative shear at C.
Thus, from Fig. 9-lc,

(L - xy
Max positive shear at C + Ri = +v 2L
and, from Fig. 9- Id,

Max negative shear at C = — R2 = —

~2L

In the course of design of a beam or girder to carry a uniform live load,

the variation in the maximum positive or negative shears throughout the
length of the beam may be required.
For instance, in a built-up steel girder,
the rivet pitch at a section depends on
the maximum combined (dead plus live
plus impact) shear at the section. If the
beam in Fig. 9-2a is subjected to a uni-
form live load of p per linear foot, the

Influence diagram for shear at C L

(6)

(a)

<^^ t i I I ; i i i ,
pa-x)2
L-x ^h
=
p(L-x)2 I
i?i R2
2L *
(b)
(0

* i i I i <^

L-x

^^-JT
id)

Fig. 9-1

maximum positive and negative shear curves due to live load are shown,
respectively, in Fig. 9-26 and c. Both curves are parabolic.
Usually only the maximum numerical shear, regardless of sign, is
needed in the design. Thus, for the beam in Fig. 9-2a, the maximum
numerical value of the shear in the left half of the beam is the sum of the
dead-load positive shear plus the maximum positive shear due to live
load and impact; the maximum numerical value of the shear in the right
half of the beam is the sum of the dead-load negative shear plus the
maximum negative shear due to live load and impact. Of course, the
maximum value of the shear at a section in the left half of the beam is

numerically equal to that at a similar section in the right half of the beam.
140 ELEMENTARY THEORY OF STRUCTURES
Example 9-1. Determine the maximum combined shears at 6-ft inter-
vals in a 48-ft simple beam subjected to a dead load of 500 lb per lin ft,

a live load of 1,200 lb per lin ft, and an impact equal to 20 per cent of the
live load.
SOLUTION. The complete solution is shown diagrammatically in Fig.
9-3. The shear diagram due to dead load is shown in Fig. 9-36. Note
that the end shear due to dead load
"^ ——— S 2 ? ^ ^ '
'

^ is M (0.5) (48) = 12 kips. The curve

48'
for maximum positive shear due to
(a) live load is shown in Fig. 9-3c. This
curve is drawn after live-load shears
have been calculated at intervals
along the beam. For instance, the
value of the live-load shear at sec-
(6) Dead load shear
tion D (36 ft from the right end) is
28.80

A TT (1.2)(36)2
jr
Vd due ,
to LL = + ^a
-

g)
= -h 16.20 kips
(c) Live load positive shear
As shown in Fig. 9-Sd, the maximum
positive shear values due to impact
5.76
4.41 3.24 9 9R 1.44
are 20 per cent of those in Fig. 9-3c.
(d) Impact positive shear The combined shear values in the
the beam are the sums of
left half of
46.56
those shown in Fig. 9-36, c, and d,
and are shown in Fig. 9-3e. Note
that the shear values shown in Fig.
9-3e are to be used in the design of
the beam. This curve may be called
the
'
'
design shear curve. '
Also note
that, in this particular case, the de-
sign shear curve almost linear. To
is
(e) Combined shear diagram
save time, some designers simply
Fig. 9-3
compute the shear ordinates at the
ends and center of the beam (46.56 and 8.64 in Fig. 9-3e) and connect the
upper extremities of these ordinates by straight lines. When this is done,
the slight errors at the intervening sections are on the side of safety.
9-3. Reactions and Shears in Simple Beams Concentrated
Maximum :

Loads. A beam AB carrying a moving system of five concen-

simple
trated loads Pi, P2, Ps at spacings 61, 62, 63, 64 is shown in Fig. 9-4.
. . . ,

These five loads may take any position on the span. It is required to
determine the critical loading position to cause the maximum numerical
shear at section C. By inspection of the influence diagram for shear at
 CRITERIA FOR MAXIMA: MOVING LOADS 141

section C of Fig. 9-4rf, it is obvious that the maximum positive shear at C

will be numerically larger than the maximum negative shear at C; this
will always be true as long as section C is to the left of the mid-span.
If the loads shown in Fig. 9-4a come upon the span from the right
toward the left, certainly most of these loads should climb up the influence
line BiCz as high as possible, in order to give the maximum positive
shear at C. It is to be expected that the position of the loads shown in
Fig. 9-46 (Pi at C) may produce the maximum positive shear at C.
With Pi at C, the shear at C is equal to i^i. By bringing P2 to C (Fig.
9-4c), the shear at C is equal to Pi — Pi. The loading condition with
P2 at C, however, may or may not cause a larger shear at C than the
loading condition with Pi at C
This depends on whether the gain in the
value of Pi is greater than the de-
crease in shear due to Pi.
Let G = Pi + P2 + P3 + P4 + P5. 1 ^
''
1 -
Vn
The left reaction in Fig. 9-4c is larger (a)

than that in Fig. 9-46 by the amount ^^2

^1 ^3 ^^4 ^5
Ghi/L because, in finding the value of
Pi by taking moments about P, the
moment arm of all the loads is in-
Rl
creased by 61 and the increase in re- (6)

action due to the forward movement

of 61 2P(6i)/L or
is Gh^/L. Thus in ?2 P3 Pi h
determining whether Pi at C or P2
at C will give the larger shear at C it

isonly necessary to compare the gain

(Gbi/L) with the loss (Pi). If the
gain is smaller than the loss, then Pi
at C causes maximum shear at C.
If the gain is larger than the loss,
then P2 at C causes a larger shear at
(d) Influence diagram for shear at section C
C than does Pi at C. If the latter Fig. 9-4
happens, it will then be necessary to
find out whether P3 will have to be moved forward to C. Theoreti-
cally this comparison between two successive loads should be repeated
until the forward load at the section controls, but very rarely will it be
necessary to bring up the third or fourth load.
If the traffic shown in Fig. 9-5a moves on the span from the left toward
the right, one of the loading conditions shown in Fig. 9-56 or c may cause
the larger positive shear at C. The loading conditions with Ps at C
(Fig. 9-56) and then P4 at C (Fig. 9-5c) should first be compared. If P5
at C controls, stop with P5 at C. If P4 at C controls, then compare the
effect of P4 at C with that of P3 at C.
142 ELEMENTARY THEORY OF STRUCTURES
Referring to Fig. 9-4, suppose that it is now required to determine
whether Pi at A or P2 at A will cause the larger left reaction. By bringing
F2 forward to A, the gain in the reaction is (P2 + P3 + Pa + P-^hi/L
exclusive of Pih/L, because Pi moves off the span. The loss in the
left reaction is, of course, Pi. Let G' be the sum of all loads on the
span exclusive of the forward load.
^^^"^ ^^^ (G'h^/L) larger
h^fj,2hZ±J '^ ^^^"^ is
I i I r i than the loss (Pi), P2 at A will give
^°^
the larger Pi; otherwise Pi at A mil
cause a larger reaction. If the for-
P. P4 P3 P2 Pi mer is true, the conditions with P2 at
^ l^^i ^1^2,L h I

^ A and then with P3 at A should be

f c i compared. The gain now becomes
l^
'
G'62/L in which G' = P3 + P4 + P5
(6) and the loss is P2.
The reader mil probably be wise to
fesj bj check through the example below and
*
I &4

—
I

^ -^ —^62 1

^
I

A then read the preceding discussion

I again.
^^
Example9-2. Determine the max-
PiG 9_5 imum left reaction and the maximum
numerical shear at section C in the
simple beam AB due to the passage of the six wheel loads shown in Fig.
9-6a in either direction.
SOLUTION. The case when the traffic comes upon the span from the
right toward the left will be considered first.

For maximum reaction at A,

Comparing Pi at A and P2 at ^,

Gain = ^ = -^^^ = 8-4 kips Loss = Pi = 6 kips

Comparing P2 at A and P3 at A,

^ G'b (54)(o) .... p = ^ .

Gam =
.

= — pQ— = ^-^ kips J

Loss = P2 9 kips
,

-J-

Thus, as shown in Fig. 9-66, the reaction Pa is a maximum with Po at ^.

12(38 -h 44 -h 49) -h 18(55) ., .

Max ^
^^ ^ .

Ra = 9 -\ ^^ ^
wK—^ - = oL/ ,
kips
oU
For maximum shear at C,

Comparing Pi at C and P2 at C,

Gain = ^ = ^^^ = 9.2 kips Loss = Pi = 6 kips

 CRITERIA FOR MAXIMA*. MOVING LOADS 143

6k 9k 18k 12k 12k 12k

01. @ (3) (p ®
( ®
5',[. 6' 5'.
.|.
I
(a)

9k 18k 12k 12k 12k

T^B
38'

60'

(6)

6k 9k 18k 12k 12k 12k

A^ (1) ® (3) (4) © (6)

12' .. C 18'

60'

(c)

12k 12k 12k 18k 9k 6k

(?) Q @ ®
f
(d)

12k 12k 12k 18k 9k 6k

aB
.30'

60'

(e)

12k 12k 12k 18k 9k 6k

Aa
60'
.

| .t .1 .

if)

Fig. 9-6

Comparing Po at C and P3 at C,

^ G6 (69)(5) . .. D = n •

= P2
.

Gain = ^r =
. ,

^ = o./o kips Loss

-r 1
9 kips

Thus, as shown in Fig. 9-6c, the maximum shear at C occurs when P2

is at C.

Max Vc = Ra -
_ 12(18
"60
= 29.5 kips
Pi
H- 24 + 29) + 18(35) + 9(40) + 6(48) __

When the traffic comes upon the span from the left toward the right,

the six concentrated loads will move on the span in the order shown in
Fig. 9-6d.
144 ELEMENTARY THEORY OF STRUCTURES
For maximum reaction Sit A,

Comparing Pq at A and P5 at A,

Gain = -j- = = 5.7 kips Loss = Pg = 12 kips

^J
Thus, as shown in Fig. 9-6e, the reaction Ra is a maximum with Pq at
A,

Max Ra = 12 -{-
^(30) + 9(38) + 18(43) + 12(49 + 54)
60
= 54.2 kips

For maximum shear at C,

Comparing Pq at C and P5 at C,

Gain = ^ = ^^^^
L/ bU
= 6.9 kips Loss = Pg = 12 kips

Thus, as shown in Fig. 9-6/, Pe at C causes the maximum shear at C.

Max Vc = Ra
^ 6(10) + 9(18) + 18(23) + 12(29 + 34 + 40)
60
= 31.2 kips

From the above computations it is seen that the larger maximum left

reaction and maximum shear at C are both caused by traffic moving from
the left toward the right. Thus, the final results are

Max Ra = 54.2 kips

Max Fc = 31.2 kips

9-4. Maximum Bending Moment Beam Uni-

at a Point in a Simple :

form Load. It is obvious that the maximum

bending moment at any
section in a simple beam due to a moving uniform live load occurs when
the beam is fully loaded. This is evident from a consideration of the
influence diagram for bending moment shown in Fig. 8-8. The beam
must be fully covered by the moving uniform load to produce maximum
bending moment at any point between the supports.
9-5. Maximum Bending Moment at a Point in a Simple Beam Con- :

centrated Loads. Let it be required to determine the position which a

series of concentrated loads must take on the simple beam AB (Fig. 9-7 a)
so that the bending moment at C will be a maximum.
Before a direct attack is made on this problem, some preliminary con-
siderations will be discussed. First, if the line segment in Fig. 9-Sb is a
portion of an influence diagram, the value of the function when the
 CRITERIA FOR MAXIMA! MOVING LOADS 145

structure is loaded with Pi, P2, P3, P4 as shown in Fig. 9-8a is SPt/ or
(Pi2/i + P22/2 4- PzUi -\- PiVi)' If Vg is the influence ordinate under the
resultant G = 2P of these forces, it can be shown that

GijG = PiVi + P2?/2 + P32/3 + PiVi

PROOF. The proof follows:

Gog = Pidi + P2a2 + Psaz + P^ai (principle of moments)

but ao = —
m ai = —
m a2 = —
m etc. (by geometry. Fig. 9-86)

Sub.M..«ng, O
(I)
= P,
(S) + P.
(S) + r.
(S) +
P.
(S)
or (t2/g = Pii/i + P22/2 + P32/3 + P42/4

The influence diagram for bending moment at (7 in a simple beam AB

is shown in Fig. 9-76. The ordinate at C2 is ab/L. The slope of hne
A1C2 is 6/L (vertical) on 1 (horizontal) and the slope of line Bid is a/L
on 1.
Pi P2 P3 P4

A(i)
T
—— (2)

a *Gi
1
(J) C (4) (^ (6)

feW
(7) (8) B

I
"
°1
^2
1
G

'1- - OG
L ^

(o) a,

»
'2

y V,

^
>ll ^ U-Ax
a_
^1

Cl
L
>2

_6
^2

^A*
T^T^^^^^^

^1

(6) Influence diagram for bending moment at C

Fig. 9-7

When a series of concentrated loads takes a position as shown in

Fig. 9-7aon the span, wdth Gi to the left of C and G2 to the right of C,
the bending moment at C is

Ale = GiVi + G22/2

If these loads move a small distance Aa; to the left, the bending moment
at C is

M'c = G,y[ + G,y',

As a result of this small movement of Ax toward the left, the increase in
146 VMJBMESfWJkKT YHBOKT €tr BMMMJMUmmXB

Bjr dKridEoig ewsrf tank in the abore equalioo hjr Ar^ Hie nle of
oT the bentfing momail ml C with respedt to Hie mawaaeBi Ar m

)
— ^J"! -h W^l') T Gri =" G Y — Gt

. ^ tht pordon of the load

r.iie beadins ni<jcneat at C

jm/L)} —

^n^'tf CTnni^Mi

dM ^
Gas
1S = T-^
In odcnlii^ ilt lias bem dianimrtntiid tdhat, wlien dMfdK = (V or

dXj, tfe conditian for m^aiwuH m or miwimiam is oblnnaL A

eondiiian enslB wlwn dM/ldx paaagji IdiEoa^ mo from pontire to
tins; TiiL„ idwn the knd at C is eameoAEBKA at an infinitcgii
to Hk n^ of C, d^if/<& = C^/2.) — Gi k pondwe, but wlwn tins load
 CRITERIA FOR MAXIMA! MOVING LOADS U7
I
is considered at an infinitesimal distance to the left of C,

dM Ga ,,

becomes neg:\ t vei

Summarizing, the criterion for maximum bending moment at C

distant a and b. respectively, from the left and right ends of a simple
beam AB with span equal to L. is that the critical load should be placed
at C so that the value of Ga/L lies between the two values of Gi, one
et 9k 18kl2k 12k 12k

.^•.|.5-.|.6-.[.5-4V[

(a)

As ^fl
:o' 20' 20' 20'

80*

(*)

9k ISkl2kl2k 12k
sf Sf ^f ^f

Cc^
r
9k iski:kr:k r:k
.0

t
(d)

6k 9k lSkl2k 12k I2k

g C^ L4^ ^5^ ^

Fio. 9-9

including and the other not including the load at C {G is the total load
on the span and Gi is the load on segment AC),
Sometimes several ditTerent loads at C will satisfy the criterion. In
this event, the bending moment at C must be calculated for each condi-
tion of loiiding that satisfies the criterion. A comparison of calculated
results will determine the greatest bending moment.
Example 9-3. Determine the maximum bending moments at sec-
tions C, D. and E of the simple beam AB (Fig. 9-96) due to the passage
of the series of concentrated loads as shown in Fig. 9-9a.
SOLUTION. Traffic may come on the span from either direction but,
if maximum bending moments at corresponding sections on either side
148 ELEMENTARY THEORY OF STRUCTURES
of the center line are computed, it is necessary to assume traffic in one
direction only (usually right to left). A Uttle reflection will show that the
maximum bending moment at C due to loads from the left is identical
with that at E due to loads from the right, and the maximum bending
moment at E due to loads from the left is identical Avith that at C due to
loads from the right.
In view of the above discussion, only traffic from the right toward
the left will be considered in the subsequent computations.

Maximum bending moment at C. From the following tabulation it is

seen that P3 at C satisfies the criterion for maximum bending moment
at C. With Pz at C, the value of Gi (load on AC) varies from
Pi -f- P2 = 15 kips

to Pi + P2 + P3 = 33 kips; the value of Ga/L = G^(20)/80 = yiG

(G = load on span) is 6% = 17.25. The criterion is satisfied because
17.25 between 15 and 33. With P2 at C, Gi varies from 6 to 15 kips,
lies

both of which are smaller than Ga/L = 17.25 kips, thus indicating
that more loads should be brought onto AC, With P4 at C, the value
of Gi varies from 33 to 45 kips. Both these values are greater than
Ga/L = 17.25 kips, and indicate that too many loads are now on AC.
It will always be advisable to make sure that the loads immediately
before and after the one (or sometimes more than one) which satisfies
the criterion do not satisfy the criterion.

Load at C G, HG G, Yes or no

P2 6 }i{69) = 17.25 15 No
Pz 15 17.25 33 Yes
P4 33 17.25 45 No

With P3 at C, the maximum bending moment at C is (Fig. 9-9c)

Max Mc =
12(43 + 49 + 54) + 18(60) + 9(65) + 6(73)
^^q)
80
- [9(5) + 6(13)]
3,855
- 123 = 840.75 kip-ft

Maximum bending moment at D

Load at D G, y2G G, Yes or no

Pz 15 3^(69) =34.5 33 No
P4 33 34.5 45 Yes
Ps 45 34.5 57 No
 CRITERIA FOR MAXIMA! MOVING LOADS 149

With Pi at D, the maximum bending moment at D is (Fig. 9-9d)

Max Md =
1^(^Q + 35 + 40) + 18(46) + 9(51) + 6(59)
^^^^
80
- [(18)(6) +9(11) +6(19)]
2,889
321 = 1,123.5 kip-ft

Maximum bending moment at E

Load at E Gi HG G, Yes or no

Pa 33 Him = 51 75 45 No
Pb 45 51 75 57 Yes
Pe 57 51 75 69 No

With Pb Sit E, the maximum bending moment at E is (Fig. 9-96)

Max Me
12(14 + 20 + 25) + 18(31) + 9(36) + 6(44)
(60)
80
- [12(5) + 18(11) + 9(16) + 6(24)]
= ^(1,854) - 546 = 844.5 kip-ft

Thus the maximum bending moment due to the passage of this series
of concentrated loads in either direction is 844.5 kip-ft at C or E and
1,123.5 kip-ft at D.
Example 9-4. Determine the maximum bending moment at section C
of the simple beam AB (Fig. 9-106) due to the passage of a uniform load
which is 12 ft in length as shown in Fig. 9-lOa.

^k/ft

L 12' J
(a)

x= 4' ,

3k/ft
I

t i I

C|
I

4 I WW Q-
I

10' 20'

30'

(6)

Fig. 9-10

SOLUTION. If the uniform load is of indefinite length, the whole

span must be loaded to produce maximum bending moment at C, or,
for that matter, at any other section in the beam. In this problem,
however, the uniform load has a definite length of 12 ft. The criterion
dM/dx = (Ga/L) — Gi = will be used to determine the position of
this uniform load on the span so that the bending moment at C may be
150 ELEMENTARY THEORY OF STRUCTURES
maximum. Let x be the length of uniform load on segment AC, then
Gi = 3x kins
^r. kips, G = l^fi kins and
36 kips, 1Q.<^ =
n/T, = i%o
anH a/L 1,

Gi = 360^) - 3a: = a: = 4 ft
L
Thus the uniform load should take the position shown in Fig. 9-106.

Max iMc = Ra(10) - ^ = ^^^^ (10) - 24 = 192 kip-ft

9-6. Absolute Maximum Bending Moment in a Simple Beam: Con-

centrated Loads. Absolute maximum bending moment in a beam is
defined as the largest bending moment which may ever occur in the beam
due to the passage of a series of loads. The significance of this moment is
apparent, as the strength of the beam must be proportioned accordingly.
be noted that the location of the section at which the absolute
It is to
maximum bending moment occurs is not yet known. Inasmuch as the

(Fd ® ® ® ® ® ® *s
a\-^
"^ c
L-c-. = ^
L

Fig. 9-11

bending-moment diagram of a beam subjected to concentrated loads

only is composed of broken straight-line segments, the absolute maximum
bending moment must occur under a load. If the maximum bending
moment under each load is first determined, then the largest of these
maxima will be the absolute maximum bending moment in the beam.
During the passage of a system of loads across a beam as shown in
Fig. 9-11, the bending moment under each load varies with every position
of the load. In the present instance, it is required to find the position of
P3 for the maximum bending moment that occurs under P3. Let
G = the sum of the loads which remain on the beam, then

(L X)
Ra = G

and the bending moment under P3 is

M = Rax (Pia + P26)

c — x)
X - (Pia + Pib)
 CRITERIA FOR MAXIMA: MOVING LOADS 151

Differentiating to determine the condition for maximum bending moment,

dM = G .J — ^ — s
=0 ri
-r- y (L 2X C)
dx L
and X = — ^

In the above equation x = {L — c)/2 is the distance from A to P3; the

distance from G to B (Fig. 9-11) is (L — c — x) which is also (L — c)/2.
Thus bending moment at P 3 to be a maximum, P3 and G must be
for the
equidistant from the left and right ends of the beam. In other words,
the maximum bending moment under any concentrated load occurs
when this concentrated load and the center of gravity of all loads on the
span are at equal distances from the center line of the beam.
It should be noted that, during the passage of a system of concentrated
loads, a maximum bending moment occurs under each load when the
above condition is met. Several calculations may be necessary to deter-
mine the greatest of these maximum bending moments. Usually, how-
ever, it will be found that the absolute maximum bending moment, or the
greatest of these maximum bending moments, occurs under the load
which is nearest to the center of gravity of the system.
If two unequal loads at a fixed distance apart move across a beam, the
maximum bending moment occurs under the heavier load.
If two equal loads of P at a distance of a apart move across a simple
beam, maximum bending moment occurs under either load when it is at a
distance x = L/2 — a/4 from either end of the beam. The maximum
bending moment can be found to be

M -U^-iJ
If the spacing of the loads is maximum bending
greater than 0.586L,
moment occurs under one load when it is at mid-span. This maximum
spacing may be obtained by equating the single-load bending moment
to the maximum value for two equal loads. Thus

PL
4 -U'-ti
Solving, a = 0.586L

The maximum bending moment caused by three equal loads, spaced

a distance a apart, occurs at the middle and is M = (P/4)(3L — 4a).
If a is greater than 0.450L, the maximum bending moment occurs with
two loads on the beam.
152 ELEMENTARY THEORY OF STRUCTURES
Example 9-5. Determine the absolute maximum bending moment in
the simple beam AB (Fig. 9-126) due to the passage of the series of
concentrated loads shown in Fig. 9- 12a.
SOLUTION. Let X be the distance between Pe and the center of gravity
G of the six loads. Taking moments about Pe,
69j = 12(6 + 11) -f 18(17) + 9(22) + 6(30)
X = 12.870 ft

Thus wheel Pa is nearest to G and the distance between Pi and G is

1.870 ft. Pi and G are placed at equal distances from the center or the

6k 9k 18k il2kl2k 12k

0) (2) C5) lO) CO (6)

_1.870'

6' + 5' 6'

(a)

Aa 0) C2) ® i® © (6)

39.065' ! 39.065' 1

> 80'

(b)

Fig. 9-12

ends of the beam as shown in Fig. 9-126. With the loads in this position,
the bending moment under Pi is

Max M at Pi = 7^^(40.935) - (moments of Pi, P2, and P3 about Pi)

69(40.935)
(40.935) - [18(6) + 9(11) + 6(19)]
80
= 1,445.3 - 321 1,124.3 kip-ft

This moment may also be found by using the right free body; thus

Max M at P4 = Pb(39.065) - (moments of P5 and Pe about Pi)

69(39.065)
(39.065) - 12(5 + 11)
80
= 1,316.2 - 192 = 1,124.2 kip-ft (check)

In Example was found that these same loads produced a maxi-

9-3, it
mum bending moment of 1,123.5 kip-ft at the middle of an 80-ft beam
with Pi at the middle. With P4 placed at 0.935 ft from the center, the
bending moment at P4 becomes a trifle larger, or 1,124.3 kip-ft. Usually
the absolute maximum bending moment in a beam occurs under the load,
w^hich, if placed at the mid-point of the beam, will cause the maximum
bending moment at the mid-section. It is seen that the difference
 CRITERIA FOR MAXIMA! MOVING LOADS 153

between the absolute maximum bending moment and the maximum

bending moment at the center is, in this case, very small. This difference,
however, may be quite large in cases when there are fewer than three or
four concentrated loads in the system, or when the spacings between
loads are relatively large in comparison with the span.
9-7. Maximum Reactions on Trusses. As explained in Art. 8-9, the
influence diagram for the left reaction of a simple truss with end floor
beams, wherein all loads on the structure are transferred to the abut-
ments through the pedestals, is the same as that for a simple beam and is
shown in Fig. 9-136. The criteria previously explained in Art. 9-3 may
be used, but in general, the max-
imum reaction will occur when the
span is loaded as much as pos-
sible and heavy loads are near the
support.
The influence diagram shown in
Fig. 9- 13c is for a truss with end
panel stringers resting directly on
the abutment. This influence dia-
gram is similar to that for bending (6) Influence diagram for the left reaction
moment at 6 of a simple beam with (with end floor beam)

span= ag. The criterion for max- ^ b

imum is

or (?i =
V (c) Influence diagram for the left reaction
(without end floor beam)
This means that the maximum end Fig. 9-13
reaction on a truss having stringers
which on the abutment occurs when the average load on the end
rest
panel equals the average load on the entire span.
Example 9-6. Compute the maximum left reaction for the simple
truss of Fig. 9-146 due to the passage of the system of loads as shown in
Fig. 9-14a, (a) when end floor beams are used, (6) when the exterior ends
of the end stringers rest directly on the abutments.
SOLUTION. When end floor beams are used, the influence diagram for
i^o of the truss is identical with that of the left reaction of a 108-ft simple
beam (Fig. 9-14c). Thus the maximum simple-beam reaction is also the
maximum reaction on the truss. For the maximum simple-beam
reaction.

Comparing Pi at Lo and P2 at Lo,

Gain = G^ ^ 139.6(5) "^

6(2.5)
6.60 kips
L 108 108
Loss = Pi = 5 kips
154 ELEMENTARY THEORY OF STRUCTURES
Comparing P2 at Lo and P3 at Lo,

^ . G'6 135.6(5) ,
6(2.5) . .^ , .

Loss = P2 = 10 kips

From the above calculations, it is seen that, when P2 is at Lo, reaction

Po is a maximum. Note that, in the comparison of Pi and P2 at Lo,

lOk
5k / / \ 10k 1.2k/ft
CD(2)(3)(4)C5) ^S
Indefinite length

(a) Loading, per stringer

Ui V2 t/3

(c) Influence diagram (or Rq

(with end floor beams)

{d) Influence diagram for Rq

(no end floor beams)

Fig. 9-14

the load on the span exclusive of Pi when Pi is at Lo is equal to

40 + 1.2(83) = 139.6 kips

In the forward movement of 5 ft, this load of 139.6 kips climbs up

the influence Une for a full distance of 5 ft; thus the gain in reaction is
(139.6) (5)/108 = 6.46 kips, but, in the meantime 5 ft more of the uni-
form load, or (1.2) (5) = 6 kips come onto the span. The center of
gravity of this 5-ft uniform load climbs up the span only 2.5 ft; thus the
gain in reaction is only 6(2.5)/108 = 0.14 kip. The total gain is there-
fore 6.46 4- 0.14 = 6.60 kips. With P2 at Lo,

Max R M (1.2) (88)^ -f- 10(93 + 98 + 103 + 108) ^ ^^ ^4 kips

108

J
 CRITERIA FOR MAXIMA: MOVING LOADS 155

When the exterior ends of the end stringers rest directly on the abut-
ments, the influence diagram for Ro of the truss is shown in Fig. 9-14d.
This influence diagram is similar to that for the bending moment at the
one-sixth point of a simple beam. The criterion is therefore Gi = G/6.

G
Load at Li G^i G, Yes or no
6

P2 5 12% = 21.5 15 No
Pz 15 13% = 22.5 25 Yes
P4 25 14% = 23.5 35 No

With Pz at Li, the reaction Ro on the truss may be found by subtract-

ing the reaction at the exterior end of the end stringer from the simple-
beam reaction at Lo.

3^(1.2)(7.5;)2 + 10(80 + 85 + 90 + 95) + 5(100) 10(5) + 5(10)

Ro
108 18
7,375 100
68.29 - 5.55 = 62.74 kips
108 18

9-8. Maximum Shear in a Panel of a Parallel-chord Truss The influ-

ence diagram for shear in panel cd
of the parallel-chord truss ag of Fig.
/
9-1 5a

L. The
is

slope of gid2
shown
is

slope
in Fig. 9-156.
aia^ over aiQi, or
of 020^2 is (ciC2
The
1

+ didi)
on / b
mp
\H/r\ J
c
p
d '-
1

over p, or [m/n -{- {n —m— V)/n\ L ='np

over p, or (n — 1) on L.
Now it is desired to determine the
condition of loading which causes
maximum positive shear in panel cd.
Let G2 equal the resultant of the
loads on panel cd (although G2 may
fall on either segment Cii or idi of
the influence diagram, in most cases
it will be within segment id^). Let
Gz equal the resultant of the loads
on segment dg. Generally there
will be no load on segment ac be-
cause such loading will cause neg-
(c) Influence diagram for shear in panel de
ative shear in panel cd (see next
Fig. 9-15
paragraph). As shown by the or-
dinates in Fig. 9-155, the positive shear in panel cd is

Vcd = G2y2 + Gzyz

156 ELEMENTARY THEORY OF STRUCTURES
If the system of loads moves a small distance Ax toward the left, the
positive shear in panel cd becomes

y'cd = G,y', + G,y',

The change or increase in shear is

=
AF — V
* cd
— V cd^
'
== G,{y 3-2.',) - G^ (y^ -yd
but
- = i Ax and -y'.= n
y'z !/s 2/2
z'-
Thus
AF^' Ax - gAf-'-
or, by letting G == G2 + G,,

AV _G - G2 G2{n - 1) . _ G -GiU
-
_ 1 fo ,
Ax L L np p \n /

A maximum is attained if AF/Ao; changes from positive through zero to

negative. This be possible if a load
will is placed at panel point d, so
that, if it is not considered in G2, (G/n) — Gi is positive, and if it is,

{G/n) — G2 is negative. The criterion for maximum positive shear in

panel cd is therefore

or, in other words, the value of G/n falls within the two values of G2.
In fact, for the usual case this criterion may be developed by referring
to Art. 8-10. In this article itwas proved that, in the triangle ig\d2 in
Fig. 9-156, the ratio of idi to igi is l/n. Inasmuch as the portion of the
influence diagram which should be loaded for positive shear in panel cd
is similar to the influence diagram for bending moment at point di in a
simple beam of span equal to igi, the criterion becomes

igi n

in is the load on idi and G is the load on both idi and digi.
which G2 It
will be noted that this derivation by similarity excludes the possibility
of any load on segment Cii while the preceding derivation does not
dictate this requirement. As pointed out before, except in unusual cases
there is generally no load on segment Cii. In any event, the criterion
G2 = G/n is applicable whenever G2 is the load within panel cd and G is

the total load on the span.

1
 CRITERIA FOR MAXIMA: MOVING LOADS 157

It may be pointed out that on occasion the criterion Gi = G/n can

be satisfied only when some loads extend into segment ac. In such a
case, G is the sum of Gi, G2, and G^, where Gi is the load on segment ac,
G2 is the load within panel cd, and G^ is the load on segment dg. The
derivation follows the usual pattern, except that in Fig. 9-1 5a and b, there
should be added a load Gi on segment ac and ordinates 1/1 and y[ on portion
aiC2 of the influence diagram.
For maximum negative shear in panel cd, portion aiic2 of the influence
diagram in Fig. 9-156 should be loaded when the traffic comes from the
left toward the right. However, a comparison of the influence diagrams
for shear in panels cd and de as shown in Fig. 9-156 and c will show that
triangle aiiC2 of Fig. 9-156 is identical with triangle igie2 of Fig. 9-1 5c
except they are on opposite sides of the base line. Therefore the maxi-
mum negative shear in any panel of a truss is numerically equal to the
maximum an opposite-handed panel on the other side
positive shear in
of the center line of the truss. Once the maximum positive shear in
every panel of a truss between the left and right supports has been found,
the maximum positive and negative shears in every panel are known.
In Fig. 9-1 5a, the maximum tensile and compressive stresses in diagonal
Cd in the parallel-chord truss ag equal the maximum positive and nega-
tive shears, respectively, in the panel Cd multiplied by the secant of the
angle the diagonal makes with the vertical.
Example 9-7. Compute the maximum positive and negative shears
in panels 0-1, 1-2, and 2-3 of the truss of Fig. 9-166 due to the passage of
the system of loads shown in Fig. 9-1 6a.
SOLUTION. The maximum positive shears in all panels due to the given
loading ^vill first be computed.

Maximum positive shear in panel 0-1

G
Load at Li G2 G2 Yes or no
6

P2 5 ^^% =21.5 15 No
Pz 15 13^^ = 22.5 25 Yes
Pi 25 141.^ = 23.5 35 No

With Pz at Li, the shear in panel 0-1 may be found by subtracting

the panel load at Lo from the left reaction on the truss (see Fig. 9-16c).

y "' ^ K(1.2)(75)^ 4- 10(80 + 85 -h 90 + 95) + 5(100) 10(5) -h 5(10)

108 18
= 68.29 - 5.55 = 62.74 kips
158 ELEMEXTABT THEOKT OF STRUCTURES
10 k

Indgfinite Wagtfa

{$) For nuuL po«. thear in panel 4-5

Fig. 9-16
 l»

f|
1

hmAtiU * 1

ft

ft
1
^
Willi ^% %t Lt^ th^ dh«wr in pMi^ \^ mai^ |i« f^wMl l^t^ salMtf^M^dniil

V
II

hmiMl» «^ ^ Y«fewiift

'^ • «^«-»>.a * X<?»

i

ft 3> ^-M..S 1« T«K

ft i* ¥>>** » Xm
ft % -^-»«..
1
^

WittK P» nl t%

l;s

-IM
Wilk Pi nl L^

r^« — i> ^
Sa^ - 3k3S » ^ M kn»
160 ELEMENTARY THEORY OF STRUCTURES
Thus the maximum positive shear of 24.99 kips occurs in panel 2-3 when
P2 is at L3.

Maximum positive shear in panel 3-4

G
Load at L4 G, G, Yes or no
6

Pi 5 No

P2 5 15 Yes

Pz 15 ^^ = - 25 No

With P2 at L4, the shear in panel 3-4 may

be found by subtracting
the panel load at L; from the left reaction on the truss (see Fig. 9-16/).

73-4 =
i^(1.2)(16)2 -K 10(21 + 26 + 31 -f 36) -h 5(41) _ 5(5)
108 18
= 13.88 - 1.39 = 12.49 kips

Maximum positive shear in panel 4-5

G
Load at L5 G, G2 Yes or no
6

Pi 3^ = 5.83 5 No
P2 5 4^^ = 7.50 15 Yes

Pz 15 1^^ = 8.10 25 No

With P2 at Ls, the shear in panel 4-5 is, from Fig. 9-16gr,

V,
10(3 + 8 + 13 + 18) + 5(23) 5(5)
4.95 - 1.39 = 3.56 kips
108 18

The shear in panel 5-6 can only be negative; therefore the maximum
positive shear in this panel due to the given system of loads is zero.
Thus the maximum positive shear in panel 0-1 is 62.74 kips, or the
maximum compression in member LqU\ is 62.74 sec 6 = 78.42 kips.
The maximum negative shear in panel 0-1 is zero; therefore tension
never occurs in diagonal LqUi.
The maximum positive shear in panel 1-2 is 42.04 kips; so the maximum
tension in member U1L2 is 42.04 sec = 52.55 kips. The maximum
negative shear in panel 1-2 is 3.56 kips; so the maximum compression
in member U1L2 is 3.56 sec 6 = 4.45 kips.
The maximum positive shear in panel 2-3 is 24.99 kips; so the maximum
compression in member U2L2 is 24.99 kips and the maximum tension in
 CRITERIA FOR MAXIMA*. MOVING LOADS 161

diagonal C/2L3 is 24.99 sec d =

The maximum negative
31.24 kips.
shear in panel 2-3 is 12.49 kips; so the tension in member maximum
U2L2 is 12.49 kips, and the maximum compression in member U2L3 is
12.49 sec d = 15.61 kips.
9-9. Maximum Bending Moment at a Panel Point in the Loaded Chord
of a Truss. The influence diagram for bending moment at c of the
parallel-chord truss ag (Fig. 9-17a) is shown in Fig. 9-176. Let Gi equal
the resultant of the loads to the left of panel point cand G2 the resultant
of the loads on segment eg. Inasmuch as this influence diagram is

/^ \ K 6 c
^
d e
S2«(n-m)p
f

L'-np
'1

(a)

«i «i 81
(6) Influence diagram for bending moment at c

Fig. 9-17

identical with that for bending moment at c of a simple beam ag, the
criterion for maximum is

Gsi
= G^
L
in which G = Gi -\- G2. It is thus seen that the maximum bending
moment at a panel point in the loaded chord of a truss occurs when the
average load to the left of the panel point equals the average load on the
entire span.
Example 9-8. Compute the maximum
bending moments at panel
points Li, L2, and L3 of the truss shown in Fig. 9-186 due to the passage
of thesystem of loads shown in Fig. 9-18a.
SOLUTION. From the discussion in Art. 9-5 be recalled that the it will

maximum bending moment at a point in the a beam due to left half of

traffic from the left is equal to that at a corresponding point in the right
half due to traffic from the right. For instance, in the present problem,
the maximum bending moment at panel point L2 due to traffic from the
left is equal to that at L4 due to traffic from the right. Thus, in order
to find the maximum
bending moments at points Li, L2, and L3 due to
only be necessary to find the maximum
traffic in either direction, it will
bending moments at all five panel points due to traffic from the right only.
162 ELEMENTARY THEORY OF STRUCTURES
,10 k
5k //\\ i.2k/ft

Indefinite length
^

(a) Loading, per stringer

^1 Ui f/3 U\

(c) For max. M at Lj

67'

( T) (?) @ (5) V////////////////////////////^^

(d) For max..Af at L^

54'

(D (2) ® (4) (6) f^^^^^yy^yyy^^y^yyy^^^^yyy^^yyyyy.

1-3

(e) For max, Af at L3

I
x-34.5'

Ci)@®(4)(5) yyyyy^yyyyy^^yyy. -^y^yyyyyyy'^^^'^^'^^y^^

La h
if) For max. M at L4 or Lg
Fig. 9-18

Maximum bending moment at L]

Load at Li G, (?i Yes or no

6

P2 5 12^^ =21.5 15 No
P3 15 13^ =22.5 25 Yes
P4 25 i4>^ = 23.5 35 No

With PaatLi (Fig. 9-18c),

M at Li = H(1.2)(75)2 + 10(80 + 85 + 90 + 95) + 5(100)

6
- [10(5) + 5(10)]
7,375
100 = 1,229.2 - 100 = 1,129.2 kip-ft
 CRITERIA FOR MAXIMA*. MOVING LOADS 163

Maximum bending moment at L2

G
Load at L2 G, C?i Yes or no
3

119 4
P* 25
- 39 8 35 No
3
125 4
P, 35 41 8 45 Yes
3

WithPsatLs (Fig. 9-lSd),

ilf at L2 = H(1.2)(67)2 + 10(72 + 77 + 82 + 87) + 5(92)

- [10(5 + 10 + 15) + 5(20)]

6,333.4
- 400 = 1,711.1 kip-ft

Maximum
bending moment at Lz. Let x equal the length of uniform
load which passes to the left of panel point L3 (Fig. 9-18e). Then
Gi = 45 + 1.2x and G = 45 + 1.2(x + 54). Equating Gi = }4G,

45 + 1.2a; = 3^[45 + 1.2(a: + 54)]

Solving for x,

X = 16.5 ft

For this loading condition (Fig. 9-18e),

M,^ at ^ ^ = H(1.2)(70.5)2 + 10(75.5 + 80.5 + 85.5 + 90.5) + 5(95.5)

jL3 ^

- [K(1.2)(16.5)2 + 10(21.5 + 26.5 + 31.5 + 36.5) + 5(41.5)]

fi 77Q f\^
= ^^ '
- 1,530.85 = 1,859.0 kip-ft
z

Maximum bending moment at L^. Let x equal the length of uniform

load which passes to the left of panel point L4 (Fig. 9-18/). Then
Gi = 45 + 1.2a; and (? = 45 + 1.2(a; + 36). Equating Gi = JiG,

45 + 1.2a; = %[45 + 1.2(a; + 36)]

Solving for x,

X = 34.5 ft

This condition of loading is shown in Fig. 9-18/. A comparison of

Fig. 9-18/ and Fig. 9-186 \vill show that the two conditions of load-
ing are identical. This must be true because the average load on the
span, [(1.2) (70.5) + 45]/108 = 1.2 kips per Un ft, is equal to the average
164 ELEMENTARY THEORY OF STRUCTURES
load on either L0L3 or LqLa- Thus Fig. 9-18e or/ must also be the critical
loading position for maximum bending moment at L5.

Max M at L4 = %(6,779.65) - [3^(1.2) (34.5)2

+ 10(39.5 + 44.5 + 49.5 + 54.5) + 5(59'.5)]
= 4,519.8 - 2,891.7 = 1,628.1 kip-ft

Maximum bending moment at L5. Referring to either Fig. 9-18e or /,

Max M atLs = %(6,779.65) - [K(1.2)(52.5)2

+ 10(57.5 +
62.5 -h 67.5 + 72.5) + 5(77.5)]
= 5,649.7 - 4,641.2 = 1,008.5 kip-ft

From the above computations it is seen that the maximum bending

moments at L4 and L5 are somewhat smaller, respectively, than those at
L2 and Li. This understandable because the average load per linear
is

foot in the range of the concentrated loadsis larger than the intensity

of the subsequent uniform load.

The maximum bending moment at Li, which in this case is also the
maximum bending moment at Ui, is the larger of 1,129.2 or 1,008.5 kip-ft.
The maximum tension in members LoLi or L1L2 is therefore

1,129.2 ,^^^,.
'

^ = 47.05 kips

The maximum bending moment at L2 or U2 is the larger of 1,711.1 or

1,628.1 kip-ft. The maximum compression in member U1U2 or the
maximum tension in member L2L3 is therefore 1,711.1/24 = 71.30 kips.
The maximum bending moment at L3 is 1,859.0 kip-ft. The maximum
compression in member [/2^3 is therefore 1,859.0/24 = 77.42 kips.
9-10. Maximum Bending Moment at a Panel Point in the Unloaded
Chord of a Truss.Let it be required to derive the criterion for maximum
bending moment at panel point C in the unloaded chord of the truss
shown in Fig. 9-19a. In this derivation, the panel point C need not be
at halfway between panel points c and d, but it is at a horizontal distance
h to the right of panel point c. The influence diagram for bending
moment at C is shown in Fig. 9-196. The ordinate C1C2 is the bending
moment at C with unity at c; thus

C1C2 = 112S2 = — S2
n

The ordinate did2 is the bending moment at C with unity at d; thus

jj = HiSi
D = n
— m— 1
didi Si
n
 CRITERIA FOR maxima: moving loads 1 165
'2

b
A B , .
C D E F

/\/i\\/\\/\/t\/\
h c d e f
;^s
mp P (n- m- \)p
L= np

-^ -Ax ^1 -*| ^Ax di ^ ^^x

(6) Influence diagram for bending moment at C
Fig. 9-19

The slope of aid is C1C2 divided by aiCi, or

oi
Slope £
01 aiC2 = —=
^1^2
aiCi
(m/n)s2
^
mp
=
S2
T"
L
The slope of g\d2 is did^ divided by gidi, or

- m -
Slope
,
01 gid2 = didi
— r- = —
[{n
7
l)/n\si
rx
Si
= t

The slope of C2C?2 is {did^ — C1C2) divided by Cidi, or

. ,
= did2 — C1C2 n — m — 1 Si m S2
Slope 01 C2a2 3-
Cirfi n p n p

(siu — Siin — Si — S2m)

~L
1
= 7- [siU — Sim — Si — (np — Si)m]

= [n(si - mp) - si] = 2^

{nh ~ ^1) = ^
~
J- ;^

Now if Gi, G2, and G3 are, respectively, the resultant loads on segments
166 ELEMENTARY THEORY OF STRUCTURES
ac, cd, and dg of the truss shown in Fig. 9-19a, the bending moment at C is
Mc = Gzijs + G22/2 + Giiji
If the system of loads moves a small distance Ax toward the left, the
bending moment at C becomes

M'c = G,y', + G2y'2 + Giy[

The change in Mc due to this movement of Ax is

Mc = M'c - Mc = Gz{y', - y) - ^2(2/2 - 2/2)

- G,{y^ - y[)

- -
but 2/3 2/
= ^ Aa: 2/2 2/; = Q - ^j Ax y,-y[ = ?l Ax

Substituting and solving for AM/Ax,

AM
Ax

(c_ffi_o,,|j_o.(-»-|;)-e.
^ Si +
_
— p
Cr
Si
7 tri
^2
—^ 6
Cr2 -
L Lf p
GSi
L ((..+0.^)

As explained before, the loading condition for maximum bending moment

at C is attained when {Gsi/L) — [Gi -{- Gi{h/p)] changes from positive
to negative. This will occur when a load is at either panel point c or d.
When a load placed at panel point c, AM/ Ax may be positive if the
is

load is included in G2, but AM

/Ax is negative if this load is included in Gi.
The criterion may also be satisfied by placing a load at panel point dj
the condition being that, if this load is not included in Gi, AM / Ax is

positive, and if it is included in G2, AM / Ax becomes negative. Usually,

then, there are two loading conditions (one with a load at c, the other
with a load at d) which satisfy the criterion; the larger of the two results
thus obtained is the required maximum.
In most cases the panel point in the unloaded chord is halfway between
the adjacent panel points in the loaded chord, thus making 6/p equal to
Yi. The criterion for maximum bending moment then becomes

If the panel point in the unloaded chord is directly above a panel point

in the loaded chord, the maximum bending moment at the former is

identical with that at the latter.

\
 CRITERIA FOR MAXIMA: MOVING LOADS 167

Example 9-9. Compute the maximum bending moment at panel points

Ui, U2, and Uz of the truss of Fig. 9-206 due to the passage of the system
of loads shown in Fig. 9-20a.
SOLUTION. The maximum bending moment at each of the five panel
points in the unloaded chord due to traffic from the right will be deter-
mined. Then the larger of the maximum bending moments at Ui or Us
willbe the maximum at either Ui or U^, and the larger of those at U2 or
Ui will be the maximum at either U2 or U^.
10 k
5k / / W 1.2k/ft

5' 5' 5' 5' 5 Indefinite length

{a) Loading, per stringer

(72 f/3 V,

86'

0( 2 ) ( 3)

f
(4) (5)

(c)
x^yy//////,^

For max. bending moments at

', 'A X- ^ < v//,

U-^ and V^
y/, y////,
1
x=10.5' 72'

ffl@@(^o ^VTT? ^/////////////y/2 /'/'^^^//A

" tl:
f 3.5'^^
(d) For max. bending moments at U^, U4, and f/5

Fig. 9-20

Maximum bending moment at U\. The influence diagram for bend-

ing moment at (7i is moment at Li. Both
similar to that for bending
are in the form of a triangle with a peak ordinate of 9.6 ft at Li in the
first case and of 19.2 ft in the latter. Therefore the maximum bending
moment at Ui is one-half of that at Li with a load at Li. The criterion
becomes Gi on LoLi equals G/5.

Load at Li G on LqLi G/b G on LoLi Yes or no

^^2 2^28.44
Pz 15 25 No

P. 25 1^^-2=29.64 35 Yes

Ps 35 1^^-2=30.84 45 No
168 ELEMENTARY THEORY OF STRUCTURES
With P4 at Li (Fig. 9-20c),

Max M at Ui
M(1.2)(86)2 + 10(91 + 96 + 101 + 106) + 5(111)

- 10(5 + 10) - 5(15)

= ^ (^^^^ - 225 j = 3^(1,786.52 - 225) = 780.8 kip-ft

Maximum bending moment at U2

Load at Li G, + G2/2 HoG G, + G2/2 Yes or no

Ps 15 + ^^2^ = 35.4 Mo (142. 2) =42.66 25 + '\' = 40A No

Pa 25+^^2^ = 43.4 Ko(148.2) = 44.46 Yes

P5 35+^22^-51.4 Ko(154.2) =46.26 45 + 2^2^ = 56.4 No

Note that, in theabove test for criterion, Gi is the load on panel LoLi;
G2, on panel and
L1L2', G, the total load on the entire span. Inasmuch as
the criterion is satisfied by placing P4 at Li and the uniform load already
extends to the left of panel point L2, there is no possibility that any con-
centrated load placed at L2 will satisfy the criterion.
The bending moment at U2 for any known loading condition may be
found by averaging the bending moments at Li and L2 or by taking
moments directly about point 172-
3^(1.2)(86)2 + 10(91 + 96 + 101 + 106) + 5(111)
M at Li
- [10(5 + 10) + 5(15)]
8,932.6
225 = 1,561.5 kip-ft

M at L2 (8,932.6) (2)

- [3^(1.2)(14)2 + 10(19 + 24 + 29 + 34) + 5(39)]

= 3,573.0 - 1,372.6 = 2,200.4 kip-ft
M at [72 = J^(M at Li + M at L2) = ^(1,561.5 + 2,200.4)
= 1,881.0 kip-ft
Taking moments directly about U2,

M at U. = Ko(8.932.6)
- H(1.2)(14)^ + 10(19 +j4) ^^^^
[

+ 10(17 + 22) + 5(27)

= 2,679.8 - 798.8 = 1,881.0 kip-ft

i
 CRITERIA FOR MAXIMA I MOVING LOADS 169

It should be remembered that the above loads are actually applied only
at the panel points of the truss.
Maximum bending moment at U^. Let x equal the length of the uni-
form load which passes to the left of L2 (Fig. 9-20d). Then,

Gi = 45 + 1.2a: G2 = 28.8 G = 45 -f 1.2(a: + 72)

Remembering that Gi + 3^(r2 = 3^(t,

(45 + 1.2a;) + K(28.8) = ^[45 + 1.2(a; + 72)]

Solving, X = 10.5 ft

For this condition of loading (Fig. 9-20d),

Max M SLtU^ = K[M(1.2)(82.5)2 + 10(87.5 + 92.5 + 97.5 + 102.5)

+5(107.5)] - [(14.4)(12) + (12.6)(17.25)
+ 10(27.5 + 32.5 + 37.5 + 42.5) + 5(47.5)]

= ^^:^^±^ _ 2,027.6 = 2,183.0 kip-ft

Maximum bending moment at Ua. It is apparent that the condition

of loading shown in Fig. 9-20d will satisfy the criterion Gi ^2^2 = l/i^G. +
A check, however, will be made.

Gi = 45 + 1.2(34.5) = 86.4 G2 = 28.8 G = 45 + 1.2(82.5) = 144

84.6 + 1^(28.8) = Ko(144)
100.8 = 100.8
Max Mat U, = Ko(8,421.25) - [(14.4)(12) + (1.2) (34.5) (29.25)
+ 10(51.5 + 56.5 + 61.5 + 66.5) + 5(71.5)]
= 5,894.9 - 4,101.2 = 1,793.7 kip-ft

Maximum bending moment at U^. The maximum bending moment

at U5 is one-half of that at L4. The criterion becomes Gi on LqLa equals
40/5. For the condition of loading shown in Fig. 9-20d,

Gi on L0L4 = 45 + 1.2(58.5) = 115.2 G = 144

Thus, Gi = HG
Max M Sit U5 = H(M Sit L,)
= Mf^(8,421.25) - [M(1.2)(58.5)2
+ 10(63.5 + 68.5 73.5+ + 78.5) + 5(83.5)]}
= H(6,737.0 - 5,310.8) = 713.1 kip-ft

Because traffic moves in either direction, the maximum bending

moment at Ui is therefore 780.8 kip-ft (the larger of the computed
values 780.8 or 713.1), that at U2 is 1,881.0 kip-ft (the larger of 1,881.0 or
1,793.7), and that at Uz is 2,183.0 kip-ft.
9-11. Maximum Stress in a Web Member of a Truss with Inclined
Chords. The influence diagram for the stress in member Be of the truss
(
170 ELEMENTARY THEORY OF STRUCTURES
with inclined chords (Fig. 9-2 la) is shown in Fig. 9-216. Due to a
moving load system the stress in member Be is tensile when portion igi
of the span is loaded, and it is compressive when portion iai is loaded.
Because triangle ic2gi is similar to the influence diagram for bending
moment at section Ci of a simple beam with span equal to igi, the criterion

O'--

o,^

(b) Influence diagram for stress in member Be

Fig. 9-21

formaximum tension in the member, due to a series of concentrated loads

coming on the span from the right, becomes

(j2 (on ici or biCi) = G ^

Likewise the criterion for maximum compression in the member, due
to a series of concentrated loads coming on the span from the left, is

lb I
G2 (on ihi or c 161) = G
iai

In a numerical problem it will be best to compute first the ordinates

6162 and C1C2 in the influence diagram by placing unity at panel points
h and c, respectively. The influence diagram thus obtained, such as
aib2C2gi in Fig. 9-216, should be checked in that the prolongations of
6201 and giC2 should intersect at point Oi directly under the moment
center 0. The distances bii and ici are then computed by dividing the
panel length into segments proportional to the ratio of 6162 to C1C2.
 CRITERIA FOR MAXIMA! MOVING LOADS 171

Finally, the criteria may be numerically established by using the relation-

ships described in the preceding paragraph.
For completeness in treatment, the criterion for maximum stress in a
web member of a truss with inclined chords will be derived for the general
case. Referring to Fig. 9-216,

— m —
C\Ci
n
= Ka —+ m-fl
e
,

r^
^
sec 6
^
= n
n
1

e +m+
1
e
— -
l
sec d

6 + m
— —np-^
e -{-

_—+p^
71 ^
O1O2 = Kg r— sec 6
^ = r sec 6
+m -

+ m-j-l
j

_ ,
6
,

l ne ,

Slope of giC2 = —- = 77 , r-T\ sec d

Slope of ai02 = —r- =

aibi
-n
L{e
—+ m +r—
i

rr sec 6
1)

6162 + C1C2 mn -\- en — e .

cji
Slope ofc
7
O2C2 = = T7 sec
r
hiCi L{e + m +r^r^
\

1)

Let G2 equal the total load on panel he and G^ the total load on the seg-
ment eg. The tensile stress in member Be is

T = G^y, + G22/2

If the system of loads moves a small distance Ax to the left, the tensile
stress in Be becomes
r = G^yi + G2yi
The increase in tension is

AT = r -T = G,(y', - 2/3) - G2{y2 - yi)

^ e sec Q .
ri (^^ + en — e) sec ^ ^

L(e +m+ 1) + m + 1)
L{e

from which -— =
Ax L(e
77 —+ m +—^;

1)
[Gse — (j2(mn + en — e)]
sec ^
j(^ _ ^^sj^ _ ^^^^^ + en - e)]
L(e +m+ 1)

_ n(m -\- e) sec 6 ^G e _ fy\

L(e +m+ 1) \n e +m /

Therefore, to obtain tjie maximum tension in member Be, the load

placed at panel point c must yield a value of (G/n)[e/{e m)] which is +
between the two values of G2, when the load assumed to be off or on
is

the panel he. Note that G2 is the total load on panel he and G is the total
load on the span.
Similarly, the criterion for maximum compression in member Be will
be derived by letting G2 equal the total load on panel eh and Gz equal the
172 ELEMENTARY THEORY OF STRUCTURES
total load on segment ab. If the system of loads moves a small distance
Ax to the right, the increase in compression in member Be is

AC = (raCslope of aib2)Ax — G2(slope of 62C2)Aa:

^ + n) sec ^
(e
^ ^ {mn -\- en — e) sec 6 ,

L{e + m + 1) L{e +m+ 1)

from which -— =
Ax ^—
L{e
.

-j-
—
m+—1)
[G^ie + n) - G^imn + ^ en - e)]
^-'

[{G - G2){e + n) - (?2(mn + en - e)]

+ m + 1)
L(6
_ n{e m 1)
-\-
-i- sec 6 /G e -\- n ^\
~ L(e + m + +m+ ~ 7
1) V^ e 1

_ n sec $ /G e -\- n ^\
Zr~ \n e +m+ 1
~ 7

Thus, to obtain the maximum compression in member Be, a load must be

placed at panel point h so that the value of {G/n)[{e + n)/{e +m+ 1)]
lies between the two values of G2. As in the preceding case, G is the
total load on the span and the two values of G2 are the load in panel he
without and then with the load at point h.
Although general expressions for the criteria have been derived, a
direct procedure based on the use of the influence diagram is usually
preferable.
Example 9-10. Compute the maximum tension and compression in
member U1L2 of the truss in Fig. 9-226 due to the passage of the system
of loads shown in Fig. 9-22a.
SOLUTION. The moment center (Fig. 9-226) is found to be at four
panels to the left of Lq. The influence diagram for the stress in member
U1L2 is shown in Fig. 9-22c. The ordinate at L2 is the stress in U1L2 due
to unity at L2; or
4
Ordinate at L2 = Ro ,
4
—+r o2 sec d

= {%){%) sec e = fls sec d tension

The ordinate at Li is the stress in U1L2 due to unity at Li; or

Ordinate at Li =
4 + 6
7^6 . .
^ sec 6

= {H)(^%) sec 6 = ^is sec 6 compression

The correctness of the influence diagram may be verified by extending

the two line segments to an intersection under and with an ordinate of
^/'is sec 6 Sit this point.
 CRITERIA FOR MAXIMA'. MOVING LOADS 173

10 k
5k //\\ 1.2k/ft

(c) Influence diagram for stress in member UiL2

42.48k l-39k
(d) For max. tension in U1L2

^-^^^
47.59k
(e) For max. tension in U1L2

if)
n C4^@a)^

For max, compression

\L2
1.39k
in U1L2
4.95 k

Fig. 9-22

Maximum tension in U1L2. In Fig. 9-22c, the ratio of l-i to i-2 is 5

to 8; so i-2 = Hs oi the panel length. The criterion for maximum

tension in U1L2 is

8
t-6 9^
8^
13
£3+4
=(-
8 + 52
Ig
15
174 ELEMENTARY THEORY OF STRUCTURES
This same result may be obtained by substituting n = 6, e = 4, and
m = 1 in (?2 = {G/n)[e/{e + m)].

Load at Li G^ Vi^G G, Yes or no

Pi ^5(101.4) = 13.52 5 No
P2 5 ^5(107. 4) = 14.32 15 Yes
Pz 15 ^5(113.4) = 15.12 25 Yes
P4 25 K5(119.4) = 15.92 35 No

It is noted that both conditions P2 at L2, and P3 at L2 satisfy the

criterion. The larger stress in U1L2 as given by either condition will be
the maximum tension in member U1L2. With P2 at L2 (Fig. 9-22(i), the
left reaction is

R, = M(l-2)(52)^ + 10(57 + 62 + 67 + 72) + 5(77) ^ ^^.48 kips

108

The panel load at Li is

(5) (5)
PatLi 1.39 kips
18

Taking moments about 0,

UJ.2 = (^^•^^)W^ (1.39)(5)

^^^ ^ ^ 27.16 sec d kips ten§ion

With P3

=
at L2 (Fig. 9-226),

H(1.2)(57)^ + 10(62 + ^ + 72+77) +5(82) ^

I
ii;„ ^^_^g ^.^^

P at L.= ^0(^\+/(^°^=
18
5.55 kips

Taking moments about 0,

47.59(4) - 5.55(5)
^, ,
U 1L2 = n sec 6 = 27.10 sec 6 kips tension

Thus the maximum tension in U1L2 is 27.16 sec 6 kips.

Maximum compression in U1L2. In Fig. 9-22c, the ratio of l-i to i-2
is 5 to 8; so l-i = ^{3 of the panel length. The criterion for maximum
compression in U1L2 is

G2
Hs G
5
^±^
^i^ = ^5 13 + 1 5 + 13 18

The same result may be obtained by substituting n = 6, 6 = 4, and

m= 1 in G2 = {G/n)l{e + n)/{e +m+ 1)]
 CRITERIA FOR MAXIMA: MOVING LOADS 175

Load at Li G, Vi^G G, Yes or no

Pi ^8(35) = 9.72 5 No
P2 5 ^8(45) = 12.5 15 Yes
P3 15 ^8(48. 6) = 13.5 25 No

With the loads moving on the span from the left and P2 at Li (Fig.

9-22f)
10(3 + 8 + 13 + 18) + 5(23) ^
Ri ^^^^ ^^^
108
5(5)
Panel load at L2 = 1.39 kips
18
Taking moments about ;

4.95(10) - 1.39(6)
UiL', sec d = 6.86 sec 6 kips compression
6

Thus the maximum compression in U1L2 is 6.86 sec 6 kips.

PROBLEMS
9-1. determine the maximum combined shears at 10-ft intervals in a 60-ft simple
Kfe«w3»^ubjected to a dead load of 0.8 kip per lin ft, a live load of 1.6 kips per lin ft,

and an impact equal to 20 per cent of the live load.

9-2. A simple beam 40 ft long carries moving loads of 5 kips, 10 kips, and 10 kips
spaced 5 ^t apart. Calculate the maximum left reaction, and the maximum shear and
bending moment at a section 10 ft from the left end.
9-3. A beam 50 ft long is supported at 10 ft from the left end and at the right end.
It carries moving loads of 5 kips, 10 kips, and 10 kips spaced 5 ft apart. Calculate
the maximum left reaction, and the maximum bending moment at the left support.
9-4. A simple beam 15 ft long carries two moving concentrated loads of 12 kips
each spaced 10 ft apart. Calculate the maximum shear and the absolute maximum
bending moment in the beam due to these loads.
9-5. A simple beam 20 ft long carries two moving concentrated loads of 10 kips
each spaced 4 ft apart. Calculate the maximum shear and the absolute maximum
bending moment in the beam due to these loads.
9-6. A simple beam 28 ft long carries two moving concentrated loads of 10 kips
and 5 kips spaced 9 ft on centers. Calculate the maximum shear and the absolute
maximum bending moment in the beam due to these loads.
9-7. A simple beam 24 ft long carries moving loads of 10 kips, 20 kips, and 15 kips
spaced 4 ft apart. Find the maximum shear and the absolute maximum bending
moment caused by this system of loading.
9-8. A simple beam 24 ft long carries a system of loads spaced 5 ft on centers. The
loads are 20 kips each. Calculate the maximum end shear and the maximum shear
at a section 4 ft from the left end. Also calculate the absolute maximum bending
moment.
9-9. A simple beam 30 ft long carries a moving uniform load of 2 kips per ft and two
movmg concentrated loads of 8 kips each spaced 10 ft on centers. Calculate the
maximum shear and the absolute maximum bending moment in the beam due to
these loads.
176 ELEMENTARY THEORY OF STRUCTURES
9-10. A simple beam 48 ft long carries a system of moving loads as shown. Calcu-
late (a) the maximum and the maximum shear at sections 12 ft and 24 ft
left reaction
from the left end, (6) the maximum bending moment at sections 12 ft and 24 ft from
the left end, and (c) the absolute maximum bending moment in the beam.

10 k 20k 20k 20k 20k 20k 20k

lOk lOk 10k 10k
5k
5' 5' 5' 5' 5' Indefinite length

+ W w 1.2k/ft
1
10' 5' 5' 10' 5' 5'

Fig. P9-10 Fig. P9-11

9-11. A simple beam 60 ft long carries a system of moving loads as shown. Calcu-
late (a) the maximum left reaction and the maximum shear at sections 15 ft and 30 ft
from the leftthe maximum bending moment at sections 15 ft and 30 ft from
end, (&)
the left the absolute maximum bending moment in the beam.
end, and (e)

9-12. A simple beam 50 ft long carries a system of moving loads as shown. Calcu-
late (a) the maximum bending moment at sections 20 ft and 25 ft from the left end,
and (6) the absolute maximum bending moment in the beam.

20 k 30k 30k 30k 30k

10' 5' 5' 5'

Fig. P9-12 Fig. P9-13

9-13. Given the Pratt truss as shown, calculate the maximum and minimum (if
any) values due to the system of moving wheel loads as given in Prob. 9-10 for (a) the
shear in panels LoLi, L1L2, and L2L3; (6) the bending moment at panel points Li, L2,
and Lz; (c) the stress in member UiLi.

Ui U2 Ue

L2 ^3 L4 Ls
6 at 20' = 120'

Fig. P9-14

9-14. Given the Warren truss as shown, calculate the maximum bending moments
at panel points Ui, Ui, and Uz due to the system of moving wheel loads as given in
Prob. 9-10.
C/3 Ui
U2^^^fK
Ux

Fig. P9-15

9-15. Given the Parker truss as shown, calculate the maximum and minimum (if
any) values due to the system of moving wheel loads as given in Prob. 9-10 for (a) the
stress in members UiUi. and L2Z/3, (6) the stress in member C/iLi, (c) the stress in
members JJiL^ and UzLz.
 CHAPTER 10

ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES

10-1. General Description. Although the floor slab in highway-

bridges is usually a reinforced-concrete slab, superstructures of bridges
which carry highway or railway traffic are often built of steel. In recent
years there has been a marked increase in the number of rigid-frame high-
way bridges built of reinforced concrete. The discussion in the present
chapter will be limited to steel-girder or truss bridges carrying highway or
railway traffic.

Bridge superstructures may make use of floor beams, but when the
span is short and loading is light, a highway-bridge floor slab may rest on
several joists or stringers running in the direction of the traffic and sup-
ported directly on the end piers. In the case of deck railway bridges,
the ties supporting the rails commonly rest on the top of two parallel
built-up plate girders. In the analysis of the highway stringers or rail-

way plate girders as described above, it is necessary to determine the

and bending moment along the span due to dead load,
variation in the shear
and impact.
live load,
For longer spans and heavier loadings, or for other reasons, it may be
uneconomical or infeasible to support the stringers or plate girders directly
on the end piers. The span is then divided into several panels of equal
length. The two or more stringers supporting the highway-bridge slab or
the two stringers supporting the open or solid deck of a single-track railway
bridge, now are simple beams with spans equal to the panel length and
supported on floor beams in the transverse direction. The floor beams
are in turn supported at the panel points of the two main girders on each
side of the traffic. In the analysis of the main girders, the object is to
determine the maximum combined shears in the panels and maximum
combined bending moments at the panel points. When trusses are used
instead of plate girders, the object will then be to determine the maximum
and minimum combined stresses in all members of either truss.
A deck-truss bridge is one in which the floor beams as described in the
preceding paragraph are connected to the panel points at the top chords
of the main trusses; a pony-truss bridge is one in which the floor beams
are connected to the panel points at the lower chords of the two main
177
178 ELEMENTARY THEORY OF STRUCTURES
trusses, which are so low that overhead lateral bracing becomes imprac-
ticable; a through-truss bridge is one in which the floor beams are con-
nected to the panel points at the lower chords of the main trusses, with
complete upper and lower lateral, portal, and sway bracing. A typical
four-panel through-truss bridge with stringers, end floor beams, floor
beams, main trusses, upper lateral bracing, lower lateral bracing, portal
bracing,and sway bracing is shown in Fig. 10-1.
The more commonly used types of trusses are shown in Fig. 10-2.
The diagonals of the Pratt truss serve mainly in tension; while those of
the Howe truss are usually in compression. The Howe truss is generally
built of timber, with vertical members of steel.The Parker truss is
really a curved-chord Pratt truss; the decreasing height of the truss
toward the ends tends to equalize the required chord areas. The Balti-
more and Pennsylvania trusses are, respectively, horizontal-chord and

Upper laterEil bracing

Sway bracing

Portal bracing

Stringers

Lower lateral bracing

Elnd floor beam
Main truss

Floor beam
Fig. 10-1

curved-chord Pratt trusses with subdivided panels, with the main diag-
onals running across two subpanels. Subverticals, subties, or substruts
are shown in Fig. 10-2e and /; the subties or substruts may be used in
either the Baltimore or the Pennsylvania truss. The K truss, like the
Baltimore and Pennsylvania trusses, may be used to provide an appropri-
ate panel length (20 to 25 ft) and a height consistent with the length of
span (one-sixth to one-eighth of the span).
10-2. Dead Load. The dead load carried by any structure is the
weight of the structure itself. The dead load on a bridge includes (1)
the weight of the floor system, and (2) the weight of the main girders or
trusses together with the bracing system. The floor system is usually
designed first; and its weight is therefore known prior to the analysis
of the main girders or trusses. The weight of the main girders or trusses
together with lateral bracing, however, must be assumed in the analysis
and then reviewed after the design has been made. A fairly good esti-
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 179

mate of the weight of the structure can usually be made by comparison

with existing bridges or by use of an appropriate empirical formula.
An ordinary 6-in. reinforced-concrete slab in the floor system of a high-
way bridge may weigh 75 psf while the stringers and floor beams may
weigh from 12 to 20 psf of roadway surface. The weight of an open-floor
railway track, including rails, ties, and fastenings, may be approximately
500 lb per lin ft.

End post Hip vertical

(a) Pratt Truss (b) Howe Truss

(c) Warren Truss (d) Parker Truss

Subtle

(e) Baltimore Truss Subvertical

Substrut

Subvertical

(g) K-Truss

Fig. 10-2

For estimating the weight of highway girders or trusses, the following

formula, which is a variation of the old Johnson, Bryan, Turneaure
formula, 1is suggested here. Again, any weight formula must be used
with discretion; the actual weight of the structure as designed should be
compared with the assumed weight used in the analysis. It may then be
necessary to revise the design.

Ly/p L{h - 16)

w
20 10
+ 50 (10-1)

^Hool and Kinne, "Steel and Timbor Structures," 2d ed., p. 359, McGraw-Hill
Book Company, Inc., New York, 1942.
180 ELEMENTARY THEORY OF STRUCTURES
where w = weight, lb perft, of each girder or truss, including floor beams,
but not stringers and slab
L = length of span, ft

b = width of roadway, ft

p = live load per lin ft of each girder or truss

For two-lane bridges, use
p = 600 lb per ft for HIO loading
p = 900 lb per ft for H 15 loading
p = 1,200 lb per ft for H20 loading
Note: See Art. 10-3 for definition of H loadings.
The weight of single-track railway bridges, for spans up to about 300 ft,

may be estimated by use of the following formulas, where ^

w = weight, lb per ft, of bridge (both girders or both trusses) includ-

ing stringers and floor beams
L = length of span, ft

Deck plate girders: w = /c(12.5L + 100) (10-2)

Through plate girders: w = k{14:L -\- 450) (10-3)
Riveted or pin-connected trusses:

w = k{SL + 700) (10-4)

In Eqs. (10-2) and (10-3),

k = 0.90 for E40 loading

k = 1.00 for E50 loading
k = 1.10 for E60 loading
In Eq. (10-4), k = 0.875 for E40 loading
k = 1.00 for E50 loading
k = 1.125 for E60 loading

Note: See Art. 10-4 for definition of E loadings.

At best the preceding formulas for estimating dead loads are rough
approximations of the weight of various types of bridge structures. They
should not be considered as entirely reliable for design. The type of
truss, the amount of live load and impact, and especially the allowable
working stresses to be used in the design of the structure are all factors
influencing the weight of the structure. Perhaps the best estimate of
dead load can be obtained by comparison with similar existing structures.
Example 10-1. Estimate the dead load on a 120-ft highway girder
bridge with floor beams spaced 20 ft apart. The width of roadway is
20 ft and live load is H20 loading. Draw shear and bending-moment
diagrams due to the estimated dead load for the main girder.
1 Ibid., p. 290.
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 181

SOLUTION. The dead load per foot of girder is estimated as follows

Assume a 6-in. floor slab which weighs 75 psf

(^) (75) (width of roadway) = (K)(75)(20) = 750 lb per ft

Assume weight of stringers at 6 psf

(3^) (6) (width of roadway) = (3^) (6) (20) = 60 lb per ft

12 k 24 k 24 k 24 k 24 k 24 k 12 k (if end floor beam is used)

t I \ I I i_

6 at 20' = 120'

(a) Panel loads «

+ 60k
1 +36k
(+ 12001,,,,, J + 12k
/ 1 • •

(+240) -12k L 1

-36k 1

-60k
(6) Shear diagram

2160 k-ft
1920 k-ft

1200 k-ft

(c) Bending moment diagram

Fig. 10-3

Estimate weight of main girder including floor beams (Eq. 10-1)

LVp L{h- 16)

"^
20 ^ 10 ^^^
_ 120 Vr^, ,

'
120(20 - 16)
+ 50 = 306 lb per ft
20 10

Total dead load per ft of girder

750 + 60 + 306 = 1,116, or say, 1,200 lb per ft

In practice the slab, stringers, and floor beams are designed before-
hand; thus their weight (with allowance for sidewalks, curbs, handrails,
etc.) can be determined prior to the analysis of the girder.

Dead panel load = (1.2 kips per ft) (panel length in ft)
= (1.2) (20) = 24 kips

The panel loads and shear and bending-moment diagrams due to

dead load for the main girder are shown in Fig. 10-3.
182 ELEMENTARY THEORY OF STRUCTURES
Example 10-2. Estimate the dead load on a single-track, 150-ft
through-truss railway bridge with floor beams spaced 25 ft apart. The
truss is a six-panel Parker truss as shown in Fig. 10-4. Assume Cooper's
E60 Determine the dead-load stresses in all members.
loading.
SOLUTION. Assuming the use of Eq. (10-4), the weight of the trusses,
bracing systems, stringers, and floor beams per foot of bridge is

w = kiSL + 700) - 1.125(8 X 150 + 700)

= 2,137.5 lb per ft of bridge or 1,068.8 lb per ft of truss

Because the weight of the stringers and floor beams comes to the
bottom chord, a common procedure is to assume that one-third of the
weight estimated above be applied at the top chord and two-thirds at
the bottom chord. The weight of track, 500 lb per foot of bridge or
250 lb per foot of truss, also acts on the bottom chord. Thus

Top panel load = (M) (1,068.8) (25) = 8,907 lb = 8.91 kips

Bottom panel load = [(%) (1,068.8) + 250](25)
= 24,063 lb = 24.06 kips
Total panel load (top and bottom) = (1,068.8 + 250) (25)
= (1,318.8)(25)
= 32,970 lb
= 32.97 kips

It is usuallymore convenient to first compute the dead-load stresses

in all members by assuming that the total panel loads are all applied at
the bottom chord, and then modify
Us U^
C/2
Ui
.

the stresses in the vertical members

by adding a compressive stress equal
Lo
to the amount of the top panel load.
L2 L3 La An inspection of Fig. 10-5 will show
i/4

6 at 25' = 150'
how this can be done.
Fig. 10-4
The dead panel loads and the cor-
all members
responding stresses in
are shown in Fig. 10-5a, which shows the summation of the stresses
shown in Fig. 10-56 and c.
The stresses shown in Fig. 10-56 are usually found by the algebraic
method. The suggested procedure is (1) to determine the amount and
kind of stress in each member (also the horizontal and vertical components
of the stress if the member is inclined) independently by the method of
sections, and (2) to review the equilibrium of each joint and see that the
two equations of equilibrium XFx = and XFy = are satisfied at each
joint. In this way, the correctness of the solution can be demonstrated.
The lengths of all inclined members are computed and entered on

Fig. 10-6. The inclined top chord member [/i(72 is extended to intersect
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 183

8.91 8.91 8.91

- -109.09 ^f -109.09 ..

+ 73.59 +73.59 +96.97 1 +96.97 +73.59 +73.59

24.06 24.06 24.06 ?4.06 24.06

82.425 82.425
(a)

82.425 82.425

8.91
+
8.91 8.91
8.91 8.91
-- ..,0
[^-3— \s /
/\ 05
OS
/\
X\ \\ // //
00
(30 en
00
\f

/
00
1 1

» ' '
.0 1

1
8.91 8.91 8.91 8.91 8.91

(c)

(a)-(6) + (c)
Fig. 10-5

2 3 4
1 1 \ I

A-^

2.5 P 2.5 P
P= 32.97 k
rf = 25'

Fig. 10-6
184 ELEMENTARY THEORY OF STRUCTURES
the bottom chord at point A. The distance ALq is found to be ^}yid in
which d is the panel length. From similar triangles U1U2M2 and A UiLi,
ALi LiU: ALi 28
or
^'
U1M2 M2U2 d 6
Solving, AU = 2%d = ^%d
and ALo = ALi- d = ^Vsd - d = ^d
The computations for the stresses in all members of the truss follow.

Section 1-1 (Fig. 10-7)

2M about Ui = 0:

(2.5P)(25)
LqLi 73.59 kips tension
28
2F. = 0:

(LoUi)v = 2.5P = 82.42 kips

{LoUi)h = (82.42) (2%8) = 73.59 kips
/ 37.54 \
LoUi = (82.42) ^ 110.51 kips compression
V 28 y
Joint Li

^Fy = 0: LiUi = 32.97 kips tension

SF. = 0: L1L2 = LqLi = 73.59 kips tension

iUiU2)v

Ar-' •L
11/3^ (UlL2)v

2.5 P 2.5 P
Fig. 10-7 Fig. 10-8

Section 2-2 (Fig. 10-8)

ZM about L2 = 0:

{2.5P){2d) - Pd 4.Pd
{UiU2)h
34 34
(4) (32.97) (25)
= 96.97 kips
34
{UiU2)v = (96.97) (^^5) = 23.27 kips

U1U2 = (96.97) (^^) = 99.72 kips compression

 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 185

XM about A = 0:

= (2.5P)(ih'^) - Pi^Vsd)
(UiL2)v

13.5
P = 26.18 kips
17
(UiL,)„ = (26.18)(2%8) = 23.38 kips

f/iLs = (26.18) (^1^) = 35.10 kips tension

Section 3-3 (Fig. 10-9)

2il/ about A = 0:
(2.5P){^Hd) - P{iysd) - Pi^Vzd)
UU
2L/2

2.3 P 2.5 P
Fig. 10-9 Fig. 10-10

Section 4-4 (Fig. 10-10)

XM about L3 = 0:
(2.5P)(3d) - {P){2d) - Pd
U2U-
34
(4.5) (32.97) (25)
= 109.09 kips compression
34
XM about U2 = 0:

— {2.5P){2d) - Pd
L^Lz
34
(4) (32.97) (25)
96.97 kips tension
34
SK =
{U2Lz)v = 2.5P - P - P = (0.5) (32.97) = 16.48 kips
(U2Lz)n = (16.48) (2^^4) = 12.12 kips

U2L, = (16.48) = 20.46 kips tension

(^H^)
Joint Uz
XFy = 0: UzL, =
186 ELEMENTARY THEORY OF STRUCTURES
The with their horizontal and vertical components in the case
stresses,
of inclined members, are entered on
Fig. 10-56. The equilibrium of each
joint can then be checked.
Example 10-3. Estimate the dead load on a single-track 240-ft
through-truss railway bridge with floor beams spaced 20 ft apart. The
truss is of the Baltimore type shown in Fig. 10-11. The live load is
Cooper's E60 loading. Determine the dead-load stresses in all members.
U4 Ug Vs

My / M^ ^ /\ ^9 48'

Lo Li
K L2 L3 L4 L5 Lq
A A A L-j Lg L(j Ljo I'll L12

12 at 20' = 240'

Fig. 10-11

SOLUTION.Assuming the use of Eq. (10-4), the weight of the trusses,

bracing systems, stringers, and floor beams per foot of bridge is

w = k{SL + 700) = 1.125(8 X 240 + 700)

= 2,947.5 lb per foot of bridge or 1,473.8 lb per foot of
truss

Although it may
be more accurate to assume that one-third and two-
thirds of this weight may be distributed to the top and bottom chords,
respectively, the fact that the live-load and impact stresses (especially
in long spans) are relatively much larger than the dead-load stresses
probably makes this refinement unnecessary. In this problem all dead
load will be assumed to act on the lower chord. By assuming the weight
of track to be 250 lb per foot of truss, the total panel load will be

(1,473.8 + 250) (20) = 34,476 lb or 34.48 kips

The dead-load stresses shown in Fig. 10-12 have been found by the
algebraic method. As in the preceding problem, it is advisable to deter-
mine the stress in each member independently, and then check the
The procedure is outlined below,
equilibrium of each joint by inspection.
but detailed computations are not shown.
1. Members Liilf 1, L3M3, and L^M^. For example, from 2^^ = at
joint Li,
LiMi = 34.48 kips tension

2. Members L2M1, MzUa, and M^Uq. For example, from Silf about
Lo = with joint Mi as the free body (Fig. 10- 13a),

(L2Mi)v = MP
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 187

5.5 P
P= 34.48 k sin ^ = 0.6402 tan e = 0.8333
(f=20' cos d = 0.7682 seed =1.3017

Fig. 10-12

and from SAf about C72 = with joint Mz as the free body (Fig. 10-136),

{MzU,)v = 3^P
3. Members LqMi, M1U2, U2MZ, M3L4, UMh, Mf,L&. By cutting
sections 1-1 through 6-6, respectively (Fig. 10-12), the vertical component
of the stress in each of these members can be found from ZF^ = 0.

£/2 U2 iM3U4)H
%^
!1^4

/
A(Af3t/4)ir

(^2Mi),

VL4
(^2A^i)h

(o) (6)

Fig. 10-13

4. Members LoLi = L1L2, L3L3 = L3L4, and LJj^ = L5Z/6. The stresses
in these members can be found from 2M about Mi = 0, section 2-2;
SM about U2 = 0, section 3-3; and 2M about Ui = 0, section 5-5.
5. Members U2UA dnd UaUq. The stresses in these members can be
found from Silf about La = 0, section 3-3; and 2A/ about Le = 0, section
5-5. It should be noted that, in taking moments about L4, section 3-3, the
panel load at joint Lz is outside of the free body (Fig. 10-14). This also
happens with XM about Le = 0, section 5-5.
188 ELEMENTARY THEORY OF STRUCTURES
6. Members U2L2, UJLa, and UeLe. The stresses in these members
can be found by using 2Fy = at joints L2, f/4, and U&.
10-3. Live Load on Highway Bridges. The Uve load to be used in the
design of highway bridges is given in ''Standard Specifications for High-
way Bridges," 6th edition, 1953, pubhshed by the American Association
of State Highway Officials. As noted in Art. 3.2.5 of these Specifications,
there are five typical loadings for highway bridges:
1. H20-44 standard truck or lane loading
2. H15-44 standard truck or lane loading
3. HI 0-44 standard truck or lane loading
4. H20-S16-44 standard truck or lane loading
5. H15-S 12-44 standard truck or lane loading
These standard truck or lane loadings are shown diagrammatically in
Fig. 10-15.
The standard truck and lane loadings under the same designation are
approximately equivalent loadings; however, the one which causes the

5.5 p
Fig. 10-14

larger stresses should always be used. The HI 0-44 and HI 5-44 loadings
are 50 and 75 per cent, respectively, of the H20-44 loading. In the case
of the H20-44 loading, the number 20 after H indicates the gross weight
in tons of the standard truck. Twenty per cent of this weight, 4 tons or
8 kips, assumed to be on the front axle; and 80 per cent, 16 tons or
is

32 kips, is assumed on the rear axle. The front and rear axles are 14 ft
apart. The manner in which the total weight of the standard truck and
load is assumed to be distributed to the four wheels is shown in Fig. 10-16.
The number 44 after each loading designation refers to the 1944 edition
of the specifications. The H15-S 12-44 loading is 75 per cent of the
H20-S16-44 loading. The number after H indicates the gross weight
in tons of the tractor truck, which is identical with the standard truck
of the same weight; and the number after S indicates the gross weight
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 189
32 k
8k
XI
14'
_ , ,\ 18 k for moment
Cone. ioadL„, . ,

/26 k for shear

I
(o) H2044 Standard truck loading
i Uniform load = 0.64 k/ft

24 k
6k

14'
O (f) H20-44 or H20-S16-44 Lane loading

(6) H15-44 Standard truck loading

13.5 k for moment

16 k Cone, load
4k 19.5 k for shear

14'
O 1 Uniform load =0. 48 k/ft

(g) H15-44 or H15S12-44 Lane loading

(c) HlO-44 Standard truck loading.

32 k 32 k
8k

14'
O 14' to 30'
O 9 k for moment
Cone, load
13 k for shear
Whatever is most critical
(d) H20-S16-44 Standard truck loading

24 k 24 k
i Uniform load = 0.32 k/ft

6k

14'
O 14' to 30'
a {h) HlO-44 Lane loading

Whatever is most critical

(c) H15-S12.44 Standard truck loading

Fig. 10-15

hW Q
14'

Width of each rear tire equals

O.IW OAW 1 inch per ton of total weight
loaded truck

s s:

W= total weight of truck and load

Fig. 10-16

in tons of the single axle of the semitrailer. The variable spacing of

14 to 30 intended to approximate closely the tractor trailers now in
ft is

common use and to provide a more satisfactory loading for continuous

spans. The lane loading consists of a uniform load of indefinite length
and a floating concentrated load (or two concentrated loads for negative
190 ELEMENTARY THEORY OF STRUCTURES
moment in continuous spans) placed on the span so as to produce maxi-
mum stress. It is to be noted that the same lane loading is equivalent to
both the H and the H-S truck loadings with the same weight of truck.
In the preceding paragraph it has been stated that the truck and lane
loadings are equivalent, but whichever gives the larger stress is to be used.
Generally the lane loading controls for long spans and the truck loading
controls for short spans. For simple beams, the dividing point has been
found to be 56 ft for maximum moment and 33 ft for maximum end shear
in the case of H loadings, and 140 ft for maximum moment and 120 ft
for maximum end shear in the case of H-S loadings (see Appendix A
of the Specifications).
As stipulated in Arts. 3.2.6 and 3.2.9 of the Specifications, the number
of lanes of live load which should be attributed to each girder or truss of
a highway bridge depends on the width of roadway. In the examples

32k 8k 32k 8k

30' 60'

1
(a) Truck loading

26 k Jk

W 0.64 k/ft 0.64 k/ft

T
'
I

30'

^ ;

(6)
I I 4

Lane loading
; I ; I

60'
I I I ; i ;

Fig. 10-17

which follow, be assumed that only one lane of live load goes to
it will

each girder or truss under consideration. In case the number of lanes

of live load on the girder or truss is different from one, the necessary
modification can be easily made.
Example 10-4. maximum end shear, (b) the maxi-
Compute (a) the
mum shear at the quarter point, maximum bending moment at the
(c) the
quarter point, (d) the maximum bending moment at the center, and (e)
the absolute maximum bending moment in the span due to one lane of
H20-44 loading on two simple beams. One has a span of 30 ft and the
other a span of 60 ft.

SOLUTION

(a) Maximum end shear (Fig. 10-17)

30-ft span:

Truck loading, F. = 32 + (8)(i%o) = 36.27 kips {controls)

Lane loading, Fe = 26 -|-
(H) (0.64) (30) = 35.6 kips
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 191

60-ft span:

Truck loading, 7« = 32 + (8)(4%o) = 38.13 kips

Lane loading, Fe = 26 + (3^) (0.64) (60) = 45.2 kips (controls)

32 k 8k 32k 8k
15' 14' 31' .,

i 1 1

60'
30' J
1

(a) Truck loading

26 k 26 k
7.5' 15'
0.64 k/ft -J 0.64 k/ft

i i ^ i i r 1 1 i I 1 1 ] 1 i 1 i
1
30' 60' t
1

(b) Lane loading

Fig. 10-18

(6) Maximum shear at the quarter point (Fig. 10-18)

30-ft span:

'22
Truck loading, F^ = (32)
w) + <«i
(m) 26.27 kips (controls)

Lane loading, V^ = (26) (^i) + = 24.9 kips

^'^"(^jlof^'
60-ft span:

Truck loading, F„ = (32)(4%o) + (SX^Ho) = 28.13 kips

Lane loading, Vy, = (26)(%) + ^'^;owi^?^' = 30.3 kips (controls)

(2) (60)

32k 8k 32 k
15' 14' 31'
'
"•] "• ,

!
1
1. .

Y'' i 1 -1

30' 60'
1
1
•|

(a) Truck loading

18 k 18 k
15'
0.64 k/ft
I t I I i 1^^ i i 1 i 1 i i i 1 i 1 1 i i*"

30' 60'

(6) Lane loading

Fig. 10-19

(c) Maximum bending moment at the quarter point (Fig. 10-19). Al-
though the criteria in Chap. 9 may be apphed, the condition of loading
producing maximum bending moment at the quarter point or mid-span
can be readily visuaUzed if the influence diagrams are sketched for these
sections.
192 ELEMENTARY THEORY OF STRUCTURES
30-ft span:
(8)(8-5) +^(32)(22.5)
Truck loading, My, = ^ j^^ ^.^^^ {controls)

Lane loading, M„ = ^i^^fMl + MD (7.5)(22.5)

= 155.25 kip-ft

It is to be noted that the bending moment at a section C, at distances

of a and 6, respectively, from the left and right supports of a simple
beam, which is loaded with a uniform load of w per linear foot over the
entire span L, is equal to Mc = {wL/2)a — {wa^/2) or Mc = wah/2.

60-ft span:

Truck loading, M^ = (18)(31) + (32)(45) ^ ^^2 kip-ft (con^roZs)

Lane loadmg, M, = ^A^KiS) + (0|4) ^^^^^^^^

= 418.5 kip-ft
32 k 8k 32 k 8k

t=±
30' 16'

30' 60'

(a) Truck loading

18k 18k
^^' 30'
.
• J
*| U.64k/ft
I.
'
I
0.64 k/ft
I

i i 1 1 i i i
t I ^ I 4 I I f I ^ I I I i ^ 31
30' 60'

(6) Lane loading

Fig. 10-20

(d) Maximum bending moment at the center (Fig. 10-20)

30-ft span:
^(l) +^(32)(15) ^
Truck loading, Mi_ = ^44 kip-ft (controls)

Lane 1
loading,
J- 7,^
3ft = -—j—
(18) (30)-
+,
(0.64) (30)2
^
^
—- = 207 .

kip-ft

60-ft span:

Truck loadmg, Mt = ffld^) + (32)(30) ^ ^^^ .j^.^,

Lane loading, Mt = ^« + »^' = 558 kip-ft (con^.o^.)

(e) Absolute maximum bending moment in the span (Fig. 10-21).

The absolute maximum bending moment in the span due to the lane
loading is the same as the maximum bending moment at the center of the
span.
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 193

Truck loading:
(40) (13.6)'
30-ft span, M max =
30
= 246.6 kip-ft (controls)

(40) (28.6)^
60-ft span, max M = = 545.3 kip-ft
60

The results of the above computations are summarized in the following

table. Note that generally the truck loading controls in the 30-ft span
and, with the exception of maximum bending moment at the quarter

32 k 8k
13.6' Vf\ 11.2'

13.6'
i=L
30' 60'

Truck loading

Fig. 10-21

point of the 60-ft span, the lane loading controls in the 60-ft span. For
H loadings, the lane loading causes larger end shear for spans over 33 ft
and larger bending moment for spans over 56 ft.

Span 30 ft 60 ft

H20-44 loading Truck Lane Truck Lane

Max end shear, kips 36.27 35.6 38.13 45.2
Max shear at quarter point, kips .... 26.27 24.9 28.13 30.3
Max moment at quarter point, kip-ft 197 155.25 422 418.5
Max moment at center, kip-ft 244 207 544 558
Absolute max moment, kip-ft 246.6 207 545.3 558

Example 10-5. maximum end shear, (h) the maxi-

Compute (a) the
mum the maximum bending moment at
shear at the quarter point, (c)

the quarter point, (d) the maximum bending moment at the center, and
(e) the absolute maximum bending moment in the span due to one lane of

H20-S 16-44 loading on two simple beams. One has a span of 100 ft
and the other a span of 160 ft.

SOLUTION

(a) Maximum end shear (Fig. 10-22)

100-ft span:

Truck loading, F. = 32 -f (32)(«^{oo) (8)(^%oo) +

= 65.28 kips (controls)
Lane loading, F^ = 26 + (3^) (0.64) (100) = 58 kips
194 ELEMENTARY THEORY OF STRUCTURES
32k 32k 8k 32k 32k 8k

a
14'

100'
72'

,
^44
160'
132'

(a) Truck loading

26 k 26 k

0.64 k/ft 0.64 k/ft

i i i I ; i ; i I 1 i \ i \ i 1 1 1 i 1 i I i \ i

100' 160' t
5 1

(6) Lane loading

Fig. 10-22

160-ft span:

Truck loading, Fe = 32 + (32)(i4%6o) + (S){^^Heo)

= 67.8 kips
Lane loading, 7e = 26 + (K) (0.64) (160) = 77.2 kips {controls)

32k 32k 8k 32k 32k 8k

25'

¥H 1 14' I 14'

100'
47'

-H4-
160'
92'

(a) Truck loading

26 k 26 k
25' 40'
c
1
0.64 k/ft J 0.64 k/ft
1 i ; 1 i 4 1 i J

100' 1 160^
1^ 4
(6) Lane loading

Fig. 10-23

(b) Maximum shear at the quarter point (Fig. 10-23)

100-ft span:

Truck loading, Vy, = (32)(7^ioo) (32)(6Koo) + + (8)(4Koo)

= 47.28 kips (controls)
(0.64) (75)2
Lane loading, Fi/, = (26)(%) + (2) (100)
37.5 kips

160-ft span:

Truck loading, Vy, = (32)(i2%6o) + (32) (i^^leo) + (8)(^?l6o)

= 49.8 kips (controls)
Lane loading Vy, = (26)(%) + = 48.3 kips
^^f^l[^^^'

(c) Maximum bending moment at the quarter point (Fig. 10-24). Both
truck loading positions, as shown in Fig. 10-24a and b, satisfy the crite-
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 195

rion for with G/^ = '^% = 18 in either case and

maximum moment,
(?i = 8 to 40 for truck heading to the left and Gi
= to 32 for truck
heading to the right. To determine which truck loading position gives
the larger moment, it is only necessary to compare the ordinates 2/1 and

8k 32k 32k 8k 32k 32k

26' |14'|14'| 106'
'^\^'^'{ '' J
-
v v v
100' 160' 1

(a) Truck loading (heading left)

32k 32k 8k 32k 32k 8k

40' 92'
'•

^T[
100' 160'

(6) Truck loading (heading right)

30

(c) Influence diagram for bending moment at the quarter point

18k 18k
[
, 25' . I .
40' .
'
I
*I

\ \
I

\ \
0.64 k/ft
\ \ \ \ \ \ \
r
t
I i i
|

I I i
0.64 k/ft
ii i i ; i w i

100' 160'
J
(d) Lane loading

Fig. 10-24

2/2 as. shown in the influence diagrams of Fig. 10-24c. Thus it is seen
that the second condition (Fig. 10-246) is critical.

100-ft span:

Truck loading, My^ ^

(8) (47) + (32) (61) + (32) (75)
4
= 1,182 kip-ft (controls)

Lane loading, My, = MM + (^ 4 '

2
(25)(75)

= 937.5 kip-ft

160-ft span:

Truck loading, ikf 1,4 = (8) (92) + (32) (106) + (32) (120)

= 1,992 kip-ft

Lane loading, My, = ilM?0) ^ (0|4) ^^^^^^^O)

= 2,076 kip-ft (controls)

196 ELEMENTARY THEORY OF STRUCTURES
8k 32k 32k 8k 32k 32k
36' 66' |l4'|l4'| 66'

1 1 I 1 V V '1

100' 160'

(a) Truck loading

50' 18k 50' 80' 18 k 80'

0.64 k/ft 0.64 k/ft 1

'
\ i i I i i \ i \ i \ I i '
1 i I i ; 1 i

100' 100'

{b) Lane loading

Fig. 10-25

(d) Maximum bending moment at the center (Fig. 10-25)

100-ft span:

Truck loading, Mt = (32)(36) + (32)(50) + (8)(64) _ ^^^^^^^

= 1,520 kip-ft (controls)

Lane 1
loading,
^. .r
ilf
(18)(100)
t = ^—^ + ,
^
(0.64)(100)2
^^ = ,or;ni- U
1,250 kip-ft

160-ft span

Truck loading, Mt = (32) (66) + 32(80) + (8) (94) _ ^^^^^^^

= 2,600 kip-ft
_ (18) (160) (0.64) (160)^
^- ^(18)(160) +
Lane 1
loading, Ml
TiT
= -^
4
,

"^ -^

= 2,768 kip-ft (controls)

8k 32W 32k 8k 32k 32k

63.667'

47.667' 77.667'

100' 160'

Truck loading

Fig. 10-26

(e) Absolute maximum bending moment in the span (Fig. 10-26).

The absolute maximum bending moment in the span due to the lane
loading is the same as the maximum bending moment at the center.
Truck loading

100-ft span, max M = ^^^^^^^^^^ - (8)(14)

= 1,523.9 kip-ft (controls)

160-ft span, max M = (72)(77^667)^ _ ^^^^^^^

= 2,602.5 kip-ft
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 197

The above computations are summarized in the following

results of the
table. Note that generally the truck loading controls in the 100-ft span
and, with the exception of maximum shear at the quarter point of the
160-ft span, the lane loading controls in the 160-ft span. For H-S load-
ings, the lane loading causes larger end shear for spans over 120 ft and
larger bending moment for spans over 140 ft.

Span 100 ft 160 ft

H20-S16-44 loading Truck Lane Truck Lane

Max end shear, kips 65.28 58 67.8 77.2
Max shear at quarter point, kips 47.28 37.5 49.8 48.3
Max moment at quarter point, kip-f t 1,182 937.5 1,992 2,076
Max moment at center, kip-f t 1,520 ,250 2,600 2,768
Absolute max moment, kip-f t 1,523.9 ,250 2,602.5 2,768

Example 10-6. A 120-ft highway girder bridge has floor beams spaced
20 ft on centers as shown in Fig. 10-27. Compute the maximum shears
in panels 0-1, 1-2, and 2-3 and the maximum bending moments at points
1, 2, and 3, due to one lane of H20-
S 16-44 loading per girder. i
1 > ^ 1 1
a

SOLUTION. Inasmuch as there are L 6 at 20' = 120'

only three concentrated loads in the

Fig. 10-27
H-S truck loading, and a uniform
load with one concentrated load in the lane loading, the position of loads
for maximum moments can be readily determined by
shears or bending
The values of maximum shears or
inspection of the influence diagrams.
bending moments may be computed by the use of either the influence
diagrams or the free-body diagrams of the girder.

Maximum shear in panel 0-1 (Fig. 10-28)

Truck loading:

Fo-i = (32)(%) + (32)(^)(8Jfoo) + (8)(%)(^Koo)

= 54.4 kips (controls)

Lane loading:

Vo.i = (26)(%) + (0.64) (M)(%) (120) = 53.67 kips

Maximum shear in panel 1-2 (Fig. 10-29)

Truck loading:

7i.2 = (32)(%) + (32)(%)(6%o) + (8)(%)(5%o)

= 42.4 kips (controls)
198 €xr

ask sk sk

II
(W^'Hxfc
35-1 2fk ST

*
1

37^1 kqiB

: = 32)00 + (32)0O(^«) + (8)ai)PJi#)

= 30.4 Vjps (eoRlrob)

Lane loadnig:
F« = (26)^^ - "-^4^(^)00^2) = 24.52 kips

/~ "

--^^
-

•
t
... t .:. t ? t
"'.'
*''
L
R"
1

1
-•

« *
A A

,^i;-irndki«d
«|-ik«kii
Lfk
— "
.' '- i '

I
i
(ir 1.

Fi^s ::-5
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 199

Maximum bending moment at point 1 (Fig. 10-31)

Truck loading:

^ ^ (8)(72) + (32)(86) + (32)(100) ^ ^^ ^^^^ ^^^^^

6

Lane loading:

Ml = ^^^^i^^^
6
+ 3^(O.W)(20)(100) = 940 kip-ft

Maximum bending moment at point 2 (Fig. 10-32)

Truck loading:
3/. = Hm{o2) + (32)(66) + (32)(80)](2)
= 1,696 kip-ft {controls)
Lane loading

M. = (MO) + (OM (^)(80) = i,5(M kijHft

(») Treekloadn!^
IBk

I I
(cl L«

Fig. 10-32 Fig. 10«

A/aximum bending moment at point 3 (Fig. 10-33)

Truck loading:

^^ ^ (32)(46) + (32)(60) -h (8)(74) _ ^^^^^^^

= 1,880 kip-ft (controls)

200 ELEMENTARY THEORY OF STRUCTURES
Lane loading

Example 10-7. Compute the maximum and minimum stresses in all

members of the sLx-panel through-truss highway bridge shown in Fig.
10-34 due to the passage of one lane

u 3- U3 f^4
, of H20-i4 loading per truss.

r /
/]
1
K \ / 7\*;i
\
L:
\

\/
Li
/
/
La ^5 \L6
clined
soLUTiox. The lengths of all
members and the point of
tersection of the prolongation of the
in-
in-

25' = 50'
'
6 a-. I
inclined upper chord with the lower

Fig. 10-34 chord should first be determined as

shown in Fig. 10-35.
The maximum stresses in the chord members are easily found after
the maximum bending moments at panel points 1, 2, and 3 have been
determined. The minimum chord stresses are, of course, zero.

Fig. 10-35

Maximum bending moment at panel point 1 (Fig. 10-36)

3/i = (8) (111) + (32) (125)

= 814.7 kip-ft
6

^ (18)(125) ^ (064) =
or ^^^ (25) (125) 1,375 kip-ft {controls)
D 2
1 375
:Max stress in LoLi and LiLo = -^^^^ = 49.1 kips tension
28

32k 8k 32k 8k
25'

4
IH'I

1
1

6 at 25' -ISC
111' ,

1
50*

S 6 at 25' = 150'
86'

(a) Truck loading (a) Truck loading

18k 18k
^'
0.64 k/ft 0.64 k/ft
!

\ i i \ I i I i i i * * i i i

6 at 25'- 150' 6 at 25' = ISC

(6) Lane loading (6) Lane loading

Fig. 10-36 Fig. 10-37

 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 201

Maximum bending moment at panel point 2 (Fig. 10-37)

(8)(86) +^(32)(100)
M. = ^ ^^96 kip-ft

or M2 = ^^^^'^J-^^ + ^^^ (50) (100) = 2,200 kip-ft (controls)

2 200
Max stress in L2L3 = —57— =64.7 kips tension

TT TT 2,200 /25.7l\ ^^ ^ .
TVT
Max X
stress in
'

U1U2 = '

i ^r ) = ,

66.5 kips compression

I2k 8k
32k 8I1
75' 1
14' I
61'

3
6 at 25' = 150'

(a) Truck loading

18 k
75'
. I
•j 0.64 k/ft
* * * * * ^ i 'I- ^ i i i * i i
3
6 at 25' = 150'

(6) Lane loading

Fig. 10-38

Maximum bending moment at panel point 3 (Fig. 10-38)

or M 3 = M^) + (0:^iKM! = 2,475 kip-ft (controls)

2 475
Max stress in U2UZ = -^ol~
~ ^^.8 kips compression

The maximum and minimum stresses in the web members can best
be determined by use of the influence diagrams.

Maximum stress in LqUi (Fig. 10-39)

Max stress in Lof/i = 1.117[(32)(1) + (8)(11H25)]

= 43.7 kips compression
or Max stress in LoUi = (26)(1.117) + (0.64) (3^^) (1.1 17) (150)
= 82.7 kips compression (controls)

Maximum stress in UiLi (Fig. 10-40)

Max stress in UiL, = (32) (1) + (8) (1^5)

= 35.5 kips tension
or Max stress in UiL, = (26) (1) -\- (0.64)(>^)(1)(50)
= 42.0 kips tension (controls)
 ELEMENTARY THEORY OF STRUCTURES

5/g sec ^=1.117 12

(a) Influence diagram for stress in
6

(a) Influence diagram for stress in LqUi UiLi

32k 8k
25
32k 8k

150'
111'
25'

^ 150'
:i
(6) Truck loading (6) Truck loading

26 k 26 k
25' ,
25' J 25' I

0.64 k/ft
.
i i 1 1 \ i 1 ; 1 1 1 1 1 1 J

150' 150'
1

Cc) Lane loading (c) Lane loading

Fig. 10-39 Fig. 10-40

Member U1L2 The influence diagram for the stress in member U1L2
is shown in Fig. 10-41.

Max stress in U.L^ = (32) (0.578) + (8)(0.578)(8Koo)

= 22.5 kips tension
or Max stress in U1L2 = (26) (0.578) + (0.64) (i^) (0.578) (115.07)
= 33.5 kips tension (controls)
Min stress in U1L2 = (32)(0.381) + (8)(0.381)(iK5)
= 13.5 kips compression
or Min stress in U,L2 = (26) (0.381) + (0.64) (M) (0.381) (34.93)
= 14.2 kips compression (controls)

+ 44/102 sec 0= +0.578

+ 58/,,- +0.569

-29/jQ2 sec^= -0.381 -33/102 =-0.324

Influence diagram for stress in U1L2 Influence diagram for stress in C72l'2

Fig. 10-41 Fig. 10-42

Member U2L2. The influence diagram for the stress in member U2L2
is shown in Fig. 10-42.

Max stress in U2L2 = (32) (0.569) + (8)(0.569)(3^^o)

= 21.5 kips tension
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 203

or Max stress in U2L2 (26) (0.569) + (0.64) (^2.:) (0.569) (65.93)

= 26.8 kips tension (controls)
Min stress in U2L2 = (32) (0.324) + (8) (0.324) (^Hs)
= 12.5 kips compression
or Min stress in U2L: (26) (0.324) + (0.64) (M) (0.324) (84.07)
17.1 kips compression (controls)

Usually the kind of live-load stress having the same sign as the dead-
load stressis called the maximum live-load stress. In this case (member
U2L2), the dead-load stress is tensile because the area of the influence
diagram above the base line is larger
4- V2 sec = + 0.624
than that below the base line.

Member UiLz. The influence dia-

gram for the stress in member U2LZ is
shown in Fig. 10-43.
60'

\,
I
1

\ 2 ' 90' .1
'1

r
-V3sec0= -0.413
Influence diagram for stress in U2L2 Max. and min. LL stresses

Fig. 10-43 Fig. 10-44

Max stress in C/2L3 = (32)(0.624) + (8) (0.624) (63.^5)

= 24.0 kips tension
or Max stress in U2Li = (26) (0.624) + (0.624) (H) (0.64) (90)
= 34.2 kips tension (controls)
Min stress in t/aLs = (32) (0.413) + (8)(0.413)(3^^o)
= 15.6 kips compression
or Min stress in [/2L3 = (26) (0.413) +
(0.64) (>^) (0.413) (60)
= 18.7 kips compression (controls)

The maximum and minimum stresses in all members of the truss due
to the passage of one lane of H20-44 loading per truss are entered on the
left and right sides, respectively, of the truss diagram in Fig. 10-44.

10-4. Live Load on Railway Bridges. The live load to be used in the
design of railway bridges is given in the ''Specifications for Steel Railway
Bridges," 1952 edition, of the American Railway Engineering Associa-
tion. It is stated in Art. 203 of this specification that the recommended
live load for each track is the Cooper E72 load as shown in Fig. 10-45,
but the engineer may specify the live load to be used, such load to be
proportional to the recommended load, with the same axle spacing.
Naturally the live load to be used in the design of a railway bridge
should be determined from the weights of the heaviest locomotives and
204 ELEMENTARY THEORY OF STRUCTURES
train loads may be expected to pass over the bridge during its
which
Ufetime. It would, however, be rather tedious and probably unwar-
ranted to compare the effects of the numerous types of locomotive load-
ings, each with different axle loads at different spacings. In 1894,
Theodore Cooper suggested the use of Cooper E40 load, a standard
which was supposed to be the equivalent of the various types of locomo-
tive loadings. The letter E means engine; 40 is the weight of the driver
axle in kips; and the train load is 4 kips per lin ft. The present Cooper
E72 load is '^^io times the original Cooper E40 load, with the same axle
spacings.
In 1923, D. B. Steinman proposed the M60 loading, which is approxi-
mately equivalent to Cooper E75 for short spans and to Cooper EGO for
long spans. A summary of Steinman's extensive studies may be found
in Locomotive Loadings for Railway Bridges, vol. 86 of the Transactions
of the American Society of Civil Engineers.
The discussion in this text will not deal with the choice of loading in a
particular situation. The E loading will be used to illustrate the typical

36k
zo
//W
72k 72k 72k 72k
0000 nn //\\
nn/n onoo oooo
46.8 k
36k 72k72k72k72k
46.8 k
.^k/ft
p.....;;.^'/?......

5' 5' 5 6' 5' 8' 5' 5' 5 6' 5' 5'

Cooper E12 load, per track

Fig. 10-45

method of computing maximum shears, bending moments, and stresses

due to locomotive and train loading.
Before numerical examples for finding the maximum shears and bend-
ing moments in a girder, or the maximum and minimum stresses in
members of a truss due to Cooper's loadings are shown, it will be advis-
able to look at the content of Fig. 10-46. The wheels of the first loco-
motive are numbered from 1 to 9, and those of the second locomotive
from 10 to 18. The axle loads and their spacings are respectively shown
above and below the wheel designations. The train load follows the
wheel concentrations and is a uniform load of indefinite length. As
typical for the other wheel concentrations, the five values listed in the
column under wheel 13 are noted to be 74, 763.2, 727.2, 27,604.8, and
24,940.8. The distance from wheel 1 to wheel 13 is 74 ft; the summation
of axle loads 1 through 13 is 763.2 kips; the summation of axle loads 2
through 13 is 727.2 kips; the summation of moments of axle loads 1
through 13 about wheel 13 is 27,604.8 kip-ft; and the summation of
moments of axle loads 2 through 13 about wheel 13 is 24,940.8 kip-ft.
The reader is advised to compute and construct Fig. 10-46 independently
on a separate sheet of paper. This table will be of use in solving the
problems which follow.
ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 205

S i
1 i i s

s© s § i
i i

1© .
S
s i i 2
o
s g 1
i© § S

1© s i s
1
i

1 s
s© ??
i i
s
ifi 00 CO

s© .
? s g
1 1
X

g© .
s
£ §
*
?•

g© s
3 i 1 1
2

8© s S 2 i
1
\
? 2 CO -.'-'

1© >
,?
i 1
-T

1© ?
i i
i
o i

s ^ s
i© i 1

i® ^ o
n 1
s

s 1
P© i i 1

s© .
2 S
2
§ s

g
?id .
= 5 g i

E2© - 'S 7 =

s§0 o s o

\
^ ^
^ qT ^ qT
C qT c
c c
a 5 c _^ C
13
o
g
H Q,
w :5
w 5
w
206 ELEMENTARY THEORY OF STRUCTURES
Another statement which may become quite useful later in moment
computations will be made and proved here. The moment of forces
Pi to Pn, inclusive, about point B (Fig. 10-47) is equal to the moment of
forces Pi to Pn, inclusive, about point A plus the product of the sum of

Po p-^

03

a-Z

ai

Fig. 10-47

Pi to Pn, inclusive, and the distance AB. Expressed in a formula, the

above statement becomes

Mb = Ma-\- [2(Pi to Pn, inclusive)] (^5) (10-5)

Proof

Ma = Piai + P2a2 + Pa^a + • ' '

+ Pnttn
Mb = Pi{a, + AB) + P2(a2 + AB) + P,{a, + .45) + • •
•

-f- Pn{an + AB)

= Pitti + P2a2 + Psas + • • •
+ Pnan
+ (P1 + P2 + P3+ • • • ^-Pn){AB)
= Ma + [2(Pi to Pn, inclusive)] (AB)

Thus, in Fig. 10-46,

S[M of 1 to (12 or 13) about 13] = 2(lf of 1 to 12 about 12)

-f [SP(1 to 12)](5)
= 24,148.8 + (691.2)(5)
= 27,604.8 kip-ft
or
S[ikf of 2 to (12 or 13) about 13] = S(ilf of 2 to 12 about 12)
4- [2P(2 to 12)](5)
= 21,664.8 -f (655.2)(o)
= 24,940.8 kip-ft

way of computing the values shown

This procedure suggests a convenient
in the lasttwo lines of Fig. 10-46. Each subsequent value may be
computed from the preceding one along the horizontal line.
Example 10-8. Two 72-ft plate girders support a single-track railway
bridge. Compute (a) the maximum end shear, (6) the maximum shear
at the quarter point, (c) the maximum bending moment at the quarter
point, and (d) the maximum bending moment at the center due to the
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 207

Cooper E72 load. In each case, find the equivalent uniform load which
will cause the same maximum effect.

SOLUTION
Maximum end shear (Fig. 10-48). Compare Pi at A with P2
(a)
at A. When Pi is at A, G on span (not including Pi) = 655.2 kips
and Gh/L = (655.2) (8)/72 = 72.8 kips > Pi. When P2 is at A, G on
span (including P2) = 799.2 kips, and

Gb ^ (799.2) (^)
^ .8 kips > Pi.
L 72

Thus, by moving P2 forward to A, the gain in the end reaction is between

72.8 and 88.8 kips, while the loss is equal to Pi = 36 kips. The gain is
greater than the loss; therefore P2 at A will cause a larger end shear or
reaction than Pi at A.
Compare P2 at A with P3 at A. When P2 is at A, G on span (not
including P2) ^ 727.2 kips and Gb/L = (727.2) (5)/72 = 50.5 kips < P2.
When P3 is at A, G on span (including P3) = 727.2 kips, and

Gh
50.5 kips < P2.
L
Thus P2 at A will cause the maximum end shear.

72'
i c
1

S"*'
h
-t

72'

Fig. 10-48 Fig. 10-49

Since the single track is supported by two girders, only one-half of

the load per track goes to each girder. Referring to Fig. 10-48,

Max R
576.8 + (799.2) (1)
204 kips
-(r-^ 72

Let w = equivalent uniform load per track

Max Ra = m^w){72) = 204
w = 11.33 kips per ft

(b) Maximum shear at the quarter point (Fig. 10-49). Compare Pi

at C with P2 at C. When Pi is at C, G on span = 511.2 kips, and
Gb/L = (511.2)(8)/72 = 56.8 kips > Pi. When P2 is at C,

G on span = 547.2 kips

and Gb/L = (547.2) (8) /72 = 60.8 kips > Pi.

208 ELEMEXTART THEORY OF STRUCTUKEB
Compare Pj at C with P, at C. When P, is at C,

G on span = 547.2 kips

and Gb/L = (547.2) (5) /72 = 38.0 kips < P^. When P, is at C,

G on span = 619.2 kips

and Gh L = (619.2)(5)/72 = 43.0 kips < Pi. Thus P 2 at C will cause

the maximum shear at C. Referring to Fig. 10-49,

l6.67.5.2+^(547.2>(6)
^^ ^.^ ^
Q [
_ 3^ j ^ ^^0.6 Wps

Let w = equivalent uniform load per track

Max T c - " ^^"^

-(2y(72r
ifr = 11.91 kips per ft

!_L J.
(a)

iirt^H-^
,rj 72k/ft

Fig. 10-50

(c) Maximum bending moment at the quarter point ^Tig. 10-50^

Wheel
G Yes or no
at C G: G^
4

Pz 36 547 2 4= 136.8 108 X

Pi 108 - 4 = 154.8 180 >/

P. 180 . 4 = 172. 8\ 216 ^

- 4 = 163. 8^
Pii 82.8 -- 4 = 155 7 154.8 X
Pit 108 - 4 = 153 180 ^
Pi» 180 64S,4 = 162 216 X
612/4 = 153
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 209

In this case both P3 at C and P12 at C satisfy the criterion for maximum
bending moment at C so calculations must be made for both conditions
of loading.

With P, at C (Fig. 10-50a),

Mc
21,052.8 + (619.2)(3)
828 ]-• 449.8 kip-ft
With P12 at C (Fig. 10-506),

_
Mi ~
1 12,585.6 + (511.2)(19) + (K)(7.2)(14)2
828
2
= 2,461.5 kip-ft (controls)

Note that the moment of wheels 10

H
to 18 about 18 is identical with the
moment of wheels 1 to 9 about 9, '1
which is found to be 12,585.6 kip-ft 36'

72'
in Fig. 10-46. Similarly the moment
ofwheels 10 to 12 about 12 is the (a)

same as the moment of wheels 1 to 3

about 3.
13 ^".7.2
Thus P12C gives a larger bend-
at k/ft

ing moment at C than does P3. rt1 >'

(ft
D 36'
1
Let w = equivalent uniform load
7 1

per track 1

(b)

M(M^)(18)(54) = 2,461.5 Fig. 10-51

w = 10.13 kips per ft

(d) Maximum bending moment at the center (Fig. 10-51)

Wheel at D G,
G Gi Yes or no
2

P4 180 511.2/2 = 255.6 252 X

Ps 252 547.2/2 =273.6 324 V
Pe 324 619.2/2 =309.6 370.8 X
P12 248.4 651.6/2 =352.8 320.4 X
PlZ 273.6 612/2 = 306 345.6 V
Pu 345.6 / 648/2 = 324 \ 370.8 X
^601. 2/2 = 300. 6

In this case, both P5 and Pu satisfy the criterion for maximum bending
moment at D] consequently calculations must be made for both con-
ditions of loading.

ill
210 ELEMENTARY THEORY OF STRUCTURES
With Pb at D (Fig. 10-51a),

M. = 16.075.2 + (547.2)(3) _ ^^gg = 3,085.2 kip-ft

\ [ 2

With Pi3 at D (Fig. 10-516),

Moment of Pi to Pis about B = 58,910.4 + (1,022.4)(1)

= 59,932.8 kip-ft
Moment of Pi to P? about B = 7,758 -h (417.6) (73)
= 38,242.8 kip-ft
Moment of Pg to Pis about B = 59,932.8 -
38,242.8
= 21,690 kip-ft
Moment of Pi to P13 about D = 27,604.8 kip-ft
Moment of Pi to Pi about D = 7,758 + (417.6) (37)
= 23,209.2 kip-ft
Moment of Pg to P13 about D = 27,604.8 - 23,209.2
= 4,395.6 kip-ft

In some more elaborate moment tables^ other than the one shown in
Fig. 10-46, the moment of any group of axle loads about the first and
last axle load in this group may be read off directly.

M. = 21,690 + (H)(7.2)(1)2
4,395.6 = 3,225.6 kip-ft
I

It is seen that P13 causes a greater bending moment at D than does P5.
Let w = equivalent uniform load per track

Hmw)(72y = 3,225.6
w= 9.96 kips per ft

The now be examined.

significance of the equivalent uniform load will
In the Cooper E72 uniform load which follows the two loco-
load, the
motives is 7.2 kips per Un ft; the average load per Unear foot under each
locomotive is 511.2/48 = 10.65 kips per ft, while the average load under
the driver axles \s'^% = 14.4 kips per lin ft. Thus the equivalent uni-
form load must, in the first place, be larger than 7.20 kips per ft; and it
increases as the loaded length decreases. Generally, it is larger for shear
than for bending moment.
Attention of interested readers should be called to the fact that tables
or curves giving equivalent uniform loads for Cooper's loading are avail-
able.- Once the equivalent uniform load for a special situation is known,
1 Hool and Kinne, "Stresses in Framed Structures," 2d ed., p. 106, McGraw-Hill
Book Company, Inc., New York, 1942.
2 Ibid., p. 134; Trans. ASCE, vol. 86, p. 610, 1923.
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 211

the desired maximum shear, bending moment, or stress in a member may

be readily computed.
Example 10-9. The floor beams of a 120-ft single-track railway girder
bridge are spaced 20 ft shown in Fig. 10-52. Compute
center to center as
the maximum and 2-3 and the maximum bending
shears in panels 0-1, 1-2,
moments at points 1, 2, and 3 due to the Cooper E72 load. In each case,
find the equivalent uniform load which will give the same maximum effect.

6 at 20' = 120'

63:
Influence diagram for shear in panel 0-1

FiG. 10-52 Fig. 10-53

SOLUTION

(a) Maximum shear in panel 0-1 (Fig. 10-53)

G
Wheel at 1 G: G: Yes or no
(3

Pt 36 1.022.4 6 = 170.4 lOS X

Px 108 1.051.2/6 = 175.2 180 >/

P* ISO 1.0S7.2/6 = 181.2 252 V

P. 216 1.0S7.2'6 = 181.2 2SS X

With P3 at 1,

^ 1
["
58.910.4 + (1.0 2 2.4)(4) -f 3.6(4)- _ 828]
"'
2 L 120 20 J
242.04 kips (controls)

With P4 at 1,

^1 [
58,910.4 + (1.022.4)(9) -h 3.6(9)- _ 1,728
"•'
2 L 120 20
= 241.82 kips

Let w = equivalent uniform load per track

(Mt^)(M)(?^)(120) = 242.04
w = 9.68 kips per ft
212 ELEMENTARY THEORY OF STRUCTURES

(6) Maximum shear in panel 1-2 (Fig. 10-54)

G
Wheel at 2 G, G, \ es or no
6

Pi 36 /S35.2/6 = 139. 2\ 108 X

V 882/6 = 147 /
P3 108 / 822, 6 = 147 \ 180 V
V928.8/6 = 154.8/
P. 180 928.8/6 = 154.8 252 X

=
43,347.6 _ 828
= 159.92 kips
Fi.2
120 20

Let w = equivalent uniform load per track

(H«')(K)(M)(96) = 109.92
w = 9.99 kips per ft

at 20' = 120'
6 at 20' - 120*
\ f

Influence diagram for shear in panel 2-3 Influence diagram for shear in panel 1-2

Fig. 10-54 Fig. 10-55

(c) Maximuryi shear in panel 2-3 (Fig. 10-55)

G ^
WTieel at 3 G2 ^' Yes or no
6 1

Pi 547.2 6 = 91.2 36 X
P. 36 619.2/6 = 103.2 108 V
Pz 108 691.2/6 = 115.2 180
P, ISO 763.2/6 =127.2 252 X

With Po at 3,

"21.052.8 -h (619.2)(4) 288"

90.84 kips
[ 120 20 J

With Pz at 3,

-
1
[
24,148.8 + (691.2) (4) _ 828
2.3 ^ ,

J20
12 20"
2L
91.44 kips (controls)
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 213

Let w = equivalent uniform load per track

(M«')(3^)(%)(72) = 91.44
w = 10.16 kips per ft

(d) Maximum bending moment at point 1

G
Wheel at 1 G, G, Yes or no
6

I P2 36 1,022.4/6 = 170.4 108 X

Pz 108 1,051.2/6 = 175.2 180
Pa 180 1,087.2/6 = 181.2 252
Ps 216 1,087.2/6 = 181.2 288 X

With Pz at 1,

Ml
58,910.4 + (1,022.4)(4) + 3.6(4)'
- 828
6
= 4,840.8 kip-ft (controls)

With P4 at 1

M _ 1 [58,910.4 + (1,022.4)(9) + 3.6(9)2 -

' - 2 6
1,728
L
= 4,836.3 kip-ft

^Tieel at 5 Gi HG (?i Yes or no

Pl3 691.2 (^)(928.8) = 774.0 763.2

Pu 763.2 [(5^) (928. 8) = 774 835.2
L(^i)(975.6) = 813 oJ
Pl5 835.2 (^) (1,022. 4) = 852.0 882
P16 882 (5^)(1,051.2) = 876.0 928.8

With Pi4 at 5,

(48,920.4) (5)
Ms = 31,420.81 = 4,673.1 kip-ft
2

With Pi5 at 5,

Ms,
53,798.4 + a,022.4)(4) ^^^ _ J ^ ^^^.^ ^ kip-ft
g^^g^^
Let w = equivalent uniform load per track

M(M^)(20)(100) = 4,840.8
w = 9.68 kips per ft
214 ELEMENTARY THEORY OF STRUCTURES
(e) Maximum bending moment at point 2

G
Wheel at 2 G, Gr Yes or no
3

252 975.6/3 = 325.2 324 X

Pe 324 1,044/3 = 348 370.8
Pt 370.8 1,080/3 = 360 417.6 X

With Pe at 2,

58,910.4 + (1,022.4)(3) + 3.6(3)'

- 5,904

= 7,383 kip-ft

Wheel at 4 Gi VsG Gi Yes or no

Pn 547.2 r (^i)(975.6) =650.4] 619.2 X

L(^^)(l,022.4) = 681. 6J
Pl2 619.2 (2^) (1,022. 4) =681.6 619.2 V
Pl3 691.2 (2^)(1,058.4) = 705.6 763.2 V
Pu 763.2 (3i)(l,094.4) = 729.6 835.2 X

With Pi2 at 4,

(58,910.4) (2)
Ma = 24,148.8 = 7,562.4 kip-ft

With Pi3 at 4,

Af 4 =
1 58,910.4 + (1,022.4)(5) + 3.6(5)2
(2) - 27,604.8

= 7,568.4 kip-ft (controls)

Let w = equivalent uniform load per track

3^(3^Ti;)(40)(80) = 7,568.4
w = 9.46 kips per ft

(/) Maximum bending moment at point 3

G Yes or no M3, kip-ft

Wheel at 3 G, Gi
2

Ps 417.6 975.6/2 =487.8 464.4 X

P9 464.4 1,022.4/2 = 511.2 511.2 8,179.2
Pio 511.2 1,072.8/2 = 536.4 547.2 V 8,223.2
Pn 511.2 1,094.4/2 = 547.2 583.2 V 8,273.7
Pl2 511.2 1,058.4/2 = 529.2 583.2 V 8,224.2
Pl3 511.2 1,022.4/2 = 511.2 583.2 8,129.7
Pm 511.2 986.4/2 =483.2 583.2 X
 ANALYSIS OF HIGHWAY AXD RAILWAY BRIDGES 215

Let w = equivalent uniform load per track

ys(y2w){120) 8,273.7
9.19 kips per ft

Example 10-10. Compute the maximum and minimum stresses in

members dLi and U1L2 of the six-panel through-truss single track

Fig. 10-56

railway bridge shown in Fig. 10-56 due to the Cooper E72 load. In each
case, find the equivalent uniform load which will cause the same maximum
or minimum effect.

SOLUTION, Maximum and minimum

(a) stresses in member UiLi
(Fig. 10-57). The maximum floor-
beam reaction at Li will be the
maximum in hanger UiLi.
stress
As shown Chap. 8, a comparison
in
of the influence diagram for stress in
member UiLi with that for bending (a) Influence diagram for stress in UiLi

moment at the center of a simple

beam with span equal to two times
the panel length (Fig. 10-57) indi-
cates that the maximum stress in
UiLi is equal to 2/d times the max- (6) Influence diagram for bendingmoment
at center of a simple beam with span = 2d
imum bending moment at the mid-
Fig. 10-57
dle of the simple beam.
In the present case, the maximum bending moment at the middle of a
50-ft beam will be computed first.

G
Wheel at center Gi G Yes or no
2

P, 108 417.6/2 = 208.8 180 X

P4 180 /417.6/2 = 208. 8\ 252 V
\464.4/2 = 232.2/
P. 252 464.4/2 =232.2 324 X
511.2^2 = 255.6
216 ELEMENTARY THEORY OF STRUCTURES

M at middle of 50-ft beam = 1 10,2G3.6

- 1,728

= 1,701.8 kip-ft
Max stress in C/iLi = (1,701.9) (^^5)
= 136.2 kips tension

Let w = equivalent uniform load per track

mw){25) = 136.2
w = 10.90 kips per ft

Min stress in UiLi =

+ '*Vl02 6ec^- +0.578

- 29/102 sec 9 = - 0.381

Influence diagram for stress in U1L2

Fig. 10-58

(6) Maximum and minimum stresses in member U1L2 (Fig. 10-58).

With traffic coming on the span from the right, for maximum tensile stress
in member ?7iL2, a load should be placed at L2.

Wheel at L2 G2
i^^ = o-« G2 Yes or no

P2 36 (0.131) (1,022. 4) = 133.9 108 X

Pt 108 (0.131)(1,051.2) = 137.7 180
Pi 180 (0.131)(1,087.2) = 142.4 252 X

With P3 at L2,

Ro =
58,910.4 + (1,022.4)(4) + 3.6(4)'
^ 150
= 210.19 kips
Panel load at Li = Vzi^^Hs) = 16.56 kips
(210.19)(iM) - (16.56)(i%) ^ 37.54
Max stress in U1L2 = X
(1%) 28
= 164.1 kips tension

Let w = equivalent uniform load per track

(M^)(>i) (0.578) (115.07) = 164.1

w = 9.87 kips per ft
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 217

With traffic coming on the span from the left, for maximum com-
pressive stress in member U1L2, a load should be placed at Li.

Wheel at L, G2 3^«4g = 0.28.G G2 Yes or no

Pi (0.284) (324) = 92.0 36 X

P2 36 (0.284) (370. 8) =105.3 108 V
Pz 108 (0.284) (417. 6) = 118.6 180 V
P< 180 "(0.284) (417. 6) = 118.0] 252 X
_ (0.284) (464. 4) = 131. 9 J

With P2 at Li,
"

Ri
1 5,904 + (370.8) (1)
20.92 kips
2 L 150
Panel load at L2 = m^^%5) = 5.76 kips
= (20.92) (13^ + 6)
- 5.76(1^ + 2) ,, 37.54
Min stress in U1L2 X
i^Vs + 2) 28
40.1 kips compression
With P3 at Li,

27.25 kips

Min stress in U1L2 = 16.56( 11^^ + 2) ^ 27.54

i^Vs + 2) 28
40.1 kips compression

Let w = equivalent uniform load per track

(>^«')(M) (0.381) (34.93) 40.1

w 12.05 kips per ft

10-5. Impact. When highway or railway traffic passes over a bridge,

the structure is subjected to not only the static but also the dynamic
effect of the moving loads. The static effect has been treated in Arts.
10-3 and The dynamic effect, usually called impact, is ordinarily
10-4.
considered to be proportional to the static effect. Thus the shears, bend-
ing moments, or stresses due to impact are found by multiplying those
due to live load by a fraction, known as the impact factor.
The impact factor is usually expressed in terms of the loaded length
of the bridge; the shorter this length, the larger the value of the factor.
It should not be hard to surmise that the impact factor for railway
bridges is much larger than that for highway bridges.

In the 1953 ''Standard Specifications for Highway Bridges" of the

American Association of State Highway Officials, the impact formula is
218 ELEMENTARY THEORY OF STRUCTURES
given in Art. 3.2.12 (c) as

' = LTT25 (!«-«)

in which / = impact fraction (maximum 30 per cent)

L = length, ft, span which is loaded to
of the portion of the
produce the maximum stress in the member
For computing truck-load moments, L is the length of span; except for
cantilever arms, L is the distance from the moment center to the far
end of truck. For shear due to truck loads, L is the length of the loaded
portion of the span from the point under consideration to the reaction;
except for cantilever arms, / = 30 per cent is used.
In the 1952 ''Specifications for Steel Railway Bridges" of the American
Railway Engineering Association, the impact formula is given in Art.
206 as
(a) The rolling effect: Vertical forces due to the rolling of the train
from side to side, acting downward on one rail and upward on the other,
the forces on each rail being equal to 10 per cent of the axle loads.
(6) The direct vertical effect: Downward forces, distributed equally
to the two rails and acting normal to the top-of-rail plane, due, in the case
of steam locomotives, to hammer blow, track irregularities, speed effect,
and car impact, and equaling the following percentage of the axle loads:
(1) For beam spans, stringers, girders, floor beams, posts of deck
truss spans carrying load from floor beam only, and floor beam hangers.

For L less than 100 ft: / = 60 -

^ (10-7)

For L 100 ft or more: / =

i^\q + 10 (1^-8)

(2) For truss spans: / = ^ \_ + 15 (10-9)

or due, in the case of rolling equipment without hammer blow (diesels,

electric locomotives, tenders alone, etc.), to track irregularities, speed

effect, and car impact, and equaling the following percentage of axle

loads
or 2

= -
For L less than 80 ft: 7
^^
40 (10-10)

For L 80 ft or more: I = ^^^ + 1^ (10-11)

where L = length, ft, center to center of supports for stringers, transverse

floorbeams without stringers, longitudinal girders and trusses (main
members) or L = length, ft, of the longer adjacent supported stringer,
longitudinal beam, girder or truss for impact in floor beams, floor beam
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 219

hangers, subdiagonals of trusses, transverse girders, supports for longi-

tudinal and transverse girders and viaduct columns.
Example 10-1 1. maximum and minimum impact stresses
Compute the
in all members of the six-panel through-truss highway bridge shown in
Fig. 10-59 due to the passage of one lane of H20-44 loading i>er truss.

Fig. 10-59

soLUTiox.The maximum and minimum live-load stresses have been

previously computed in Example 10-7. The impact formula

50
L + 125

is used to calculate the impact stresses as shown in Table 10-1.

T-AJBLE 10-1

Max stress Min stress

Mem-
ber LL Loaded Impact Impact LL Loaded Impact Impact
stress length L factor / stress stress length L factor / stress

+49.1 150 0.182 + 8.9

L,Lr +49.1 150 0.182 + 8.9
+64.7 150 0.182 +11.8
-66.5 150 0.182 -12.1
-72.8 150 0.182 -13.2
-82.7 150 0.182 -15.0
+42.0 50 0.286 + 12.0
+33.5 115.07 0.208 + 7.0 -14.2 34.93 0.300* -4.3
+26.8 65.93 262 + 7 -17.1 84.07 0.239 -4 1

+34.2 90 0.232 + 7.9 -18.7 60 0.270 -5.0

U^L^

* Max 30 per cent.

Example 10-12. Determine the impact factor for the live-load shears
and bending moments in the railway girder described in Example 10-9.
soLUTiox. In accordance with the 1952 AREA specifications, regard-
less of the loaded length of live load, the same impact factor will be used
for shears and moments at all points of the girder.
220 ELEMENTARY THEORY OF STRUCTURES
The distance between center lines of rails will be assumed to be 5 ft
and that of girders 8 ft.

The Impact due to rolling effect is shown in Fig. 10-OOa. It is seen

that this may increase the live-load stress in one girder by 12.5 per cent
and, at the same time, decrease the live-load stress in the other girder
by a like amount. If steam locomotives are assumed to run on this
bridge, the direct vertical effect (Fig. 10-606) can be considered by using
Eq. (10-8). Thus

J ( 1,800 , . A , / 1,800 ,
, A
^ = \r^^o + ^V P'' ''""^ ^ Vr2o^io +
^V P''
"^""^

= (22.5 -h 10) per cent = 32.5 per cent

The total impact factor including both the rolling and direct vertical
effects is (12.5 + 32.5) per cent = 45 per cent.
20% of load 20% of load 32.5% of load 32.5% of load
per rail per rail per rail per rail

I 5' c.c. rails I I ^ 5' cc. rails ^ I

^ ^

L
I 8' c.c. girders
1 I
t
I
^
8' c.c.
.

girders 1J
12.5% 12.5% 32.5% 32.5X

(a) Rolling effect (6) Direct vertical effect

Fig. 10-60

Use of Counters in Trusses. The maximum or minimum com-

10-6.
bined stress in any member of a bridge truss is the sum of stresses due
to dead load, live load, and impact. If the sign of the maximum com-
bined stress is opposite to that of the minimum combined stress, the
member is subjected to stress reversal. Members subjected to stress
reversal are usually designed, depending on the governing specifications,
to accommodate by 50 per cent of the smaller.
either stress, increased
Also, regardless of the compressive stress a member has to
amount of
take, the limiting value of the slenderness ratio must not be exceeded
(the maximum value of the slenderness ratio is usually limited to 120
for main compression members). In case of main members subjected to
a small reversal in compression, this requirement is quite severe.
The diagonals in bridge trusses, especially those in the panels near the
middle of the span, are usually subjected to stress reversals. Designers
sometimes provide two diagonals in such a panel so that either takes
tension only. When two diagonals are provided in the same panel, the
diagonal in which the dead-load stress is tensile is called the main diag-
onal, and the other diagonal, in which the dead-load stress is compressive,
is called the counter. Thus in Fig. 10-61, members C/3L4 and LdJh are
main diagonals, and members Lzlj\ and U^L^, if used, are counters.
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 221

t/i U2 f/3 U4 Us Ue U^
A /\
/\ V

// \\ \\ /
/ /
/ \ \
""

'

Li Lo L3 1,4 L5 Lg L7 jLg

8 at 20' = 160'

Dead load: 1200 lb/ft of truss; V3 at top, % at bottom

Live load: 3000 lb/ft of truss
Impact factor: 45X for all members
Fig. 10-61

Counters LzV and V ^^ \v411 be needed if the minimum combined stress

I in
^

members ^^^4 and LJJ found on the assumption that no counters

c^,

are used, is compressive, and if compression is not allowed in any diagonal.

It will be appropriate to compare the maximum and minimum stresses in
all members of a bridge
° truss when - ^6 ,tan ^9 - 7.5 tan 5 - 8 tan 5
counters are not used with those
of the same truss when counters are
used. In the following example,
for simplicity of illustration, a Pratt
truss subjected to a moving uni-
form live load will be considered.
Complications^ will necessarily arise
in other more elaborate forms of
when the live load
trusses, especially
involves a system of moving con-
centrated loads; however, a simple
example is used to afford an initial
understanding of the effect of the
use of counters in bridge trusses.
Example 10-13. Compute the
1 1 1
1 \

maximum and minimum combined -102.8 .

-128.6 -137.1
stresses in all members of the bridge
truss shown in Fig. 10-61, (a) when
counters are not used, and
counters are used.
(6) when 4/ ~
\' \
SOLUTION. Maximum and mini-
/
1 + 60.0 , .
+ 60.0 , . +102.8 , . +128.6
mum combined stresses in all mem- 16 . 16 16 16
84
(c)
bers will first be computed when no
Fig. 10-62
counters are used.
The dead-load stresses can be found conveniently by the method
outUned in Fig. 10-62 where stresses in all members due to unit panel

' Consult section on bridge trusses, Hool and Kinne, "Stresses in Framed Struc-
tures," 2d ed., McGraw-Hill Book Company, Inc., New York, 1942.
222 ELEMENTARY THEORY OF STRUCTURES
loads at the lower chord are shown in Fig. 10-62a. In this problem,

Dead-load panel load = (1.2) (20)

= 24 kips (8 kips at top and 16 kips at bottom)
24 tan = (24)(29^8) = 17.14 kips

24 sec 6 = (24) (y^?9^±^\ = 29.49 kips

The dead-load stresses due to panel loads of 24 kips each at the lower
chord as shown in Fig. 10-626 are found by multiplying the stresses in
Fig. 10-62a by 24. If one-third of the dead load is assumed to go to the

top chord, then a compression of 8 kips is added to the stress in each

vertical member of Fig. 10-626 to give the corrected dead-load stresses as
shown in Fig. 10-62c.

I
7i V2 Us Ua Us Ue U^
(-)Area= 5.71 sec 5

/
/\ \ \ / / / \ \ \ \/ / / \
§5
5.71'

Lo ^1 ^-2 1^3 L, Ls 1:'6 J^7 ^8

8 at 20' = 160' 2/8 sec e

(d) Influence diagram for stress in U2L3

(-)Area=1.43 sec
+ 6/«secd (+)Area= 12.86 +%

Vasec
(6) Influence diagram for stress in U1L2 Influence diagram for stress in U3L2
(e)

(+)Area=5. 71
\+2; )Area= 12.86 sec d

^-"'^ —^'^ \ l_1 /I OQ» ;.57'

'^
(-) Area = 35^72^ 11.43' \
5.71'

\ (/)
-3/8 seed (

Influence diagram for stress in

+ )Area- 22.86
UsLt
sec ff

(c) Influence diagram for stress in L^ L2

Fig. 10-63

The maximum live-load stresses in all chord members as well as those

in the end posts can be found by multiplying the dead-load stresses in
these members by the ratio of live load to dead load, or 2.5 in this case.
The minimum live-load stresses in these members are, of course, zero.
The maximum live-load stress in the hip vertical (UiLi) is equal to the
live-load panel load, or (3) (20) =
60 kips tension; the is zero. minimum
The maximum and minimum live-load stress in the web members
UiL2y U2L2, U2L3, UJjz, and UzL4 can best be found by use of the influ-
ence diagrams. Thus, if each area as indicated in Fig. 10-63 is multiplied
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 223

by the intensity of live load, or 3 kips per ft, the appropriate maximum
or minimum live-load stress will be obtained.
The reader is now advised to check the table of stresses (Table 10-2).
From Table 10-2 it is noted that, if members
stress reversal in all diagonal
is to be avoided, counters must be provided in panels
L^Ui and UJLi^
LJji and L4L5. When counters are used, the lines preceded by a single
asterisk in Table 10-2 are to be replaced by those preceded by double
asterisks.
Table 10-2

Live-load stress Impact stress Combined stress

Dead-load
Member stress
Max Min Max Min Max Min

C/1C/2 -102.8 -257.1 -115.7 -475.6 -102.8

C/2C/3 -128.6 -321.4 -144.6 -594.6 -128.6
UzU, -137.1 -342.8 -154.3 -634.2 -137.1
LoLi -h 60.0 -hl50.0 + 67.5 +277.5 + 60.0
L1L2 + 60.0 -1-150.0 + 67.5 +277.5 + 60.0
Ld^z -hl02.8 +257.1 + 115.7 +475.6 + 102.8
L3L4 + 128.6 +321.4 + 144.6 +594.6 + 128.6
LoC/i -103.2 -258.1 -116.1 -477.4 -103.2
u,u + 16.0 + 60.0 + 27.0 + 103.0 + 16.0
UiL2 -h 73.7 + 189.6 - 5.3 + 85.3 - 2.4 +348.6 + 66.0
U2L2 - 44.0 -107.2 + 17.1 - 48.2 + 7.7 -199.4 - 19.2
UiLz -h 44.2 + 131.7 -21.1 + 59.3 - 9.5 +235.2 + 13.6
*UzLz - 20.0 - 68.6 +38.6 - 30.9 + 17.4 -119.5 + 36.0
*UzL, -h 14.7 + 84.3 -47.4 + 37.9 -21.3 + 136.9 - 54.0
*U,L, - 8.0 - 8.0 - 8.0
**Ud.z - 20.0 - 68.6 - 30.9 -119.5 - 8.0
- 8.0
**UzL, + 14.7 + 84.3 + 37.9 + 136.9
**UJL, -h 4.0 - 38.6 - 17.4 - 52.0 - 8.0
- 8.0
**UU, - 14.7 + 47.4 + 21.3 + 54.0

* Counters are not in action.

** Counters are in action.

L3C/4 and UJjf, provided, the maximum combined

Even with counters
main diagonals V^L^ or LJJf, remains the same ( + 136.9 kips)
stress in the
as when counters are not used, and the minimum combined stress is zero,
since the diagonals cannot carry compression. For instance, zero com-
bined stress occurs in member UzL^. when the live load covers a short
distance at the end of the span so that the compression caused by the
left
Uve load plus impact just balances the tension due to dead load. When
the chord members UzL\ and L3L4 are parallel, as in this problem, the
maximum combined stress (tension) in the counters L^Ua or [/4L5 is
224 ELEMENTARY THEORY OF STRUCTURES
equal to the minimum combined stress (compression) in the main diag-
onals U3Li or L4L\. The minimum combined stress in the counters is,
of course, zero.
The effect of the counters on the stresses in other members will be now
investigated. The loading diagram when counter L^Ui is in action is
shown in Fig. 10-64, but it should be noted that the live load does not
actually come to the lower chord as uniform load. The moment center
for UsUi is now at L3 and that for L3L4, at t/4. However, the combined
stresses in the chord members due to this partial live load will not be
criticalw^hen compared with those due to full hve load, when the main
diagonals are in action. Thus the use of counters will not affect the chord
stresses at all. The maximum combined stress in U^Lz is —119.5 kips
and occurs when main diagonal U^Li is in action; and, as can be observed
from Fig. 10-64, when the counter is in action, the minimum combined
stress is —8 kips. When the main diagonals are in action, the stress in
UJji is —8 kips, and when counter L^Ui is in action (Fig. 10-64), the
8k 8k 8k 8k 8k 8k 8k
\
U2 Us J
if/6 if/7

\
f

/Live load\ /

^0,
A 1 i 1 i IN i i A 7V\ \ Ls
\l2 1^5
16 k 1( 5k le)k 16 k 1(5k 16k 16k
8 at 20 = 160'
'

Loading diagram— counter L3C/4 inaction

Fig. 10-64

dead-load stress is +4 kips, the live-load stress is —38.6 kips with

impact of —17.4 kips; so the combined stress is —52.0 kips.
10-7. Analysis of Bridge Portals. A sketch showing the typical con-
struction of a single-track through-truss railway bridge is shown in
Fig. 10-65. The upper and lower lateral trusses serve not only to tie
the two main trusses together as a matter of practical necessity for
lateral stability, but they are also subjected to lateral forces. These
lateral forces include wind pressure on the vertical surface of the structure
and the live load (the train, for instance), and the sway or vibratory
forces in the lateral direction during the passage of the live load. The
amount of these wind or lateral forces to be considered in the design is
usually given in the governing specifications, such as the AREA specifica-
tions for railway bridges or the AASHO specifications for highway
bridges. Generally, when the stresses due to lateral loads are combined
with those due to vertical loads (dead, live, and impact) the allowable
design stresses are increased by a percentage, which is also dictated by the

appropriate specifications.
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 225

The lateral forces are usually treated as moving concentrated loads

on the upper or lower lateral trusses. The cross diagonals may be
assumed to take tension only, or if appropriately designed each diagonal
may be assumed to take half the shear in the panel with one diagonal in
compression and the other in tension. The lateral forces will also cause
stresses in some members of the main truss, especially in the lower
chord. The complete analysis of bridge trusses under the action of
lateral forces will not be treated here; interested readers are referred to
Hool and Kinne's ''Stresses in Framed Structures," 2d ed., p. 334,
McGraw-Hill Book Company, Inc., New York, 1942.
Wind forces acting on the vertical surface of the structure are usually
assumed to be equally divided between the upper and lower lateral
Main truss

Lower lateral truss

Fig. 10-65

systems. The loads on the lower lateral truss go directly into the end
supports, while those on the upper lateral truss are carried by the portal
frames to the supports at either end of the bridge. The stress analysis
of the portal frames (such as the one shown in Fig. 10-65) will be con-
sidered in the subsequent discussion.
The analysis of a bridge portal is very similar to that of a building
bent. The end posts of the bridge truss are analogous to the columns
of the bent. Take, for example, a typical portal frame shown in Fig.
10-66. The load P is one-half of the total wind or other lateral forces
assumed to act on the top chord of the through truss. The lower ends
of the end posts are assumed to be either fixed or partially fixed because
of the large compressive stresses in the end posts due to vertical loads.
In the case of the combination of large wind pressure with small vertical
loads (such as a wind of 50 psf on an unloaded highway bridge), the end
226 ELEMENTARY THEORY OF STRUCTURES
p p

\/\^ c

+
1'2
«
p P
- „^ Points of P P
inflection -^ > -c
rt "^ 2 2
s
u
, P(a + 6}
^2- «

1 7^
^1 ^2

(a) (6) (c)

Fig. 10-66

posts may be considered to be partially fixed. The usual assumptions

used in the analysis are, then, that the horizontal reactions are equal
and that the points of inflection are at one-third to one-half of the clear
length of the end post from the end joint of the main truss. Thus, in
Fig. 10-66a,

}1\ = H2 ~ p
2
Pc
In Fig. 10-666: Ml = M
In Fig. 10-66c: Vi = V2
P{a + h)

In Fig. 10-66c, the end posts are three-force members, but all others
are two-force members.

Rigid joints
^KXX3 zs
A"^
(a) Girder portal (6) Lattice portal (c) A-frame portal (d) Portal with (e) Portal for wide
diagonal bracing roadway

Fig. 10-67

10 k

Assumed
12k

^ Assumed
y^

Assumed
As
10k

^
Assumed
10k

K 18' 18' 20'

li 16'

(a) (ft) (c) (d)

Fig. 10-68
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 227

10 10
,90
10k B C 10 5 B Cj 5
c
~~B
1^90
10 10

m
1 1

-10 -10
5 '-'^

- +
18' o.
10 90
(a)

12kB +3C 12^ ;^19.5 7.5

M
15 6 «-

--
13.5
o
+
6!

13.5
3
M 18' ^60
J?'' 12
I (6)

10 kB_ 16.25+ 6.25C 10, 16.25 ^ 6.25

—I*^
I So

/^ r9

11.25
I'
11.25
9
J=
I

s
'n"
o
I.
A 20'
^50 50

(0

11.25
I

.'

10k B -16.25 C 10^ 16.25 <^^ 6.25

B kll.25
I
— in J]' I
in

11.25 + 11.25

iH ^50
^9 D
''^50
5_ ^n
11.25 11.25
(d)
16

lOkB-20 -10 -no IQ^y^ 15 -«--

c
p
B
12
^ +10 -10 Oj 12}
-> in
10 To*' +
+

N 9 T-

in
..M +u
40' 10 60
D
C
(e)

Fig. 10-69
228 ELEMENTARY THEORY OF STRUCTURES
Some common types of portal frames are shown in Fig. 10-67.
In the
wherein two short diagonals are present in
lattice portal of Fig. 10-G76
each panel, the vertical shear may be equally divided between the
diagonals, each taking an equal amount of tension or compression. In
the portal with diagonal bracing (Fig. 10-67o?), it will be economical
in the design to assume that the diagonals take tension only. It is to be
noted that the transverse load P may act in either direction, to the left
or right.
Example 10-14. Analyze each of the portal frames shown in Fig. 10-68.
Direct stresses, shears, and bending moments in the end posts and
direct stresses in all other members are required.
SOLUTION. The analysis of these portal frames involves little that is
new; so complete details will not be given. Nevertheless, the reader
will benefit by a review of the principles of statics by analyzing these
frames independently and checking the results with those given in
Fig. 10-69.

PROBLEMS
10-1. Estimate the dead load on a single-track 160-ft through-truss railway bridge
with floor beams spaced 20 ft apart. The truss is an eight-panel Parker truss as
shown. Assume Cooper's E60 loading. Determine the dead-load stresses in all
members.

Fig. PlO-1

10-2. Compute (a) the maximum end shear, (6) the maximum shear at the quarter

point, (c) maximum bending moment at the quarter point, {d) the maximum
the
bending moment at the center, and (e) the absolute maximum bending moment in the
span due to one lane of H20-44 loading on two simple beams. One has a span of
40 ft and the other a span of 80 ft.
10-3. Compute (a) the maximum end shear, {h) the maximum shear at the quarter
point, (c) the maximum bending moment at the quarter point, {d) the maximum
bending moment at the center, and (e) the absolute maximum bending moment
in the span due to one lane of H20-S 16-44 loading on two simple beams. One has a
span of 80 ft and the other a span of 180 ft.
12 3 4 5

5 at 25' = 125'
F
Fig. PlO-4

10-4. A 125-ft highway girder bridge has floor beams spaced 25 ft on centers as
shown. Compute the maximum shears in panels 0-1, 1-2, and 2-3 and the maximum
bending moments at points 1 and 2 due to one lane of H20-S1G-44 loading.
 ANALYSIS OF HIGHWAY AND RAILWAY BRIDGES 229

lO-Bj Compute the maximum and minimum stresses in all members of the eight-
pan<5rthrough-truss highway bridge due to the passage of one lane of H20-44 loading
per truss. For dimensions of the truss, see sketch for Prob. 10-1.
Two 60-ft plate girders support a single-track railway bridge. Compute (a)
10-6.
maximum end shear, (b) the maximum shear at the quarter point, (c) the maxi-
the
mum bending moment at the quarter point, and (d) the maximum bending moment
at the center due to the Cooper E72 load. In each case, find the equivalent uniform
load which will cause the same maximum effect.
10-7. The floor beams of a 1 60-ft single-track railway girder bridge are spaced 20 ft
on centers as shown. Compute the maximum shears in panels 0-1, 1-2, 2-3, and 3-4

8 at 20' = 160'

Fig. PlO-7

and the maximum bending moments at points 1, 2, 3, and 4 due to the Cooper E72
load. In each case, find the equivalent uniform load which will give the same maxi-
mum effect.
10-8. Compute, the maximum and minimum stresses in members UiLi, U2L2, and
(72^8 of the eight-panel through-truss single-track railway bridge as shown in Prob.
10-1 due to the Cooper E72 load. In each case, find the equivalent uniform load
which will cause the same maximum or minimum effect.

L/i f/2 Us C/4 Us

/\ \ / / \^
/ \ \/ / \
L, 1'^2 ^3 ^4 ^^5
"S

^6
6 at 20' = 120'

Fig. PlO-9

10-9. The bridge truss shown is subjected to a dead load of 1,200 lb per foot of
truss (one-third at top; two-thirds at bottom) and a live load of 3,000 lb per foot of
truss with 45 per cent impact. Compute the maximum and minimum combined
stresses in all members {a) when counters are not used, and {h) when counters are
used.

10 k 10 k lOk lOk
v /y\s-7
XX
"*"

^l
> <
- - / S

«3
CM ^ r

CO h.

20'
JL J — 20'
..

(b) (c)

Fig. PlO-10

10-10. Analyze each of the portal frames as shown. Required are the direct
stresses, shears, and bending moments in the end posts and the direct stresses in all
other members.
 11

ANALYSIS OF STATICAIXY DO^TERMINATi: BE.AVCS

r I r

A <r
! f ^
3^
J
t/rifcT
! !
-
zi —T^
—
r t <l^
(a' Simpie beam 'ft

Ffift. Ut-I. Statically

ODe cr '
'be
rwu!ti._ _-- ^ -aid to be frtaritkalBy

two uQikiiioiirr.
-
^iiimhwiI lijr Ifce farars cf skaftks aBoBBCL The

Fig. 11-1 are ^^^.^.^^ n^r.f^^

If a bcaun reafts en ^ as dMrana ia ¥^ 11-2K, tibe Ibce

Fj^ F^ an e i&Ert drtri BMBMwl bcfiore ribear and
nMT"* - .-, r\o-^T»T-r/»iffjBii
^HBDOe Q^BBS OBb^
irifi|pl(|My

two rndrpra ifcmt i eqiti :iiaDS <ai this bean canni; be

bf the I^w tinaircfiQffe a beana mttiiie €» Hfcree

aaidto ^^^ - --iifee to tibe fint degree btcaML

it ladcB one eond. ^ statiei to gme the three
 ANALYSIS OF STATICALLY IND ETERMINATE BEA:MS 231

condition, which cannot be is furnished bv the

an equation of statics,
geometry of the elastic cun.'e of the beam.
This elastic curve is sketched
in dotted form in Fig. 1 1-2a. A simple beam supported at A and C and
subjected to the known forces Pi and Pi, and the unknown force F^, is
shown in Fig. 1 1-26. If the physical properties of the beam and the force*
in Fig. 11-26 are identical with those in Fig. ll-2a, the elastic curves must
also be identical.The value of Va in Fig. 1 1-26 may be determined from
when the simple beam AC is subjected to the loads Pi
the condition that,
and Pi and an unknown force Va, the vertical deflection of point B is equal

Pi p. P. A ^1 A Pi P.

A h \
c A 1. i C A i. I c A - a c

TT
\
JS^ T^^~-JS^ -1"- "^"^^X-' -JT^ -^^-^. ---'X--^.

\ t t \ \ t I t t
^A ^ ^A % Va %
(6) (d)

Fig. 11-2

to zero. downward deflection Aa (Fig. 11-36) at B due

In other words, the
to Pi and Pi must be equal to the upward deflection Va^a (Fig. II-3c) at
B due to Fb. wherein 5 5 is the upward deflection at B due to a unit
upward load acting at point B of simple beam AC. Thus Va^a = Aa
and Vb is equal to As 5^. Once Va in Fig. ll-3a or Fig. 11-26 has thus
been determined from a condition of geometry. Vj_ and Vc may be found
from the two laws of statics. A beam, then, may be statically inde-
terminate, but staticallv and geometricallv. no beam is indeterminate.

p. p, p, p.

|C, -— _c
^
A \ B C A \ 3 A,.---.r

^^' t
" 1 i i
t
% Vb Vc I
^3

(«) (8) (c)

Fig. 11-3

By inspection of Fig. Il-2c, it is seen that Vc may first be determined

by the geometrical condition that the vertical deflection of point C of
the overhanging beam AB, subjected Pi and Pi and an to the loads
unknown force Vc, is equal to zero. Likewise, as shown in Fig. ll-2c?y
Vj_ may first be determined independently by applying this suae cnuB^
tion of geometry at A. The unknown reaction, which is s^eted to be
first found by the condition of geometry, is said to be redundant. If
the beam has only one redundant, it is statically indeterminate to the
first degree. This is true even though any one of the three unknown
reactionsmay be chosen as the redundant and must consequently be
determined by a condition of geometry.
232 ELEMENTARY THEORY OF STRUCTURES
Additional statically indeterminate beams are shown in Fig. 11-4.
The beams of Fig. ll-4a and h are statically indeterminate to the first
degree; those of Fig. ll-4c and d, to the second degree; and those of
Fig. ll-4e and /, to the third degree. The degree of indeterminacy is

equal to the number of redundants, the numerical values of which can

be found only from the conditions of the geometry of the elastic curve.
For example, Vb and Vc in Fig. llAd, when chosen to be the redundants,
can be found from the two conditions that the deflections at B and C
must both be zero when the five forces Vb, Vc, Pi, P2, and P3 act on the
cantilever beam fixed at A. If, however, Ma and Va are chosen to be

the redundants, they can be found from the conditions that the slope
of the elastic curve at A is zero and the deflection at A is also zero if
the forces Va, Pi, P2, Pz, and the couple Ma act on the overhanging
beam BC.

Pi P2 Pi Pi P3 Pi P2 Pi Pa

A \ B i ^ A i B j C \ D A \ B \ C \ D \ E

t ! tilt ! t I t t f

(o) (c) (e)

*1 ^2 1 2 P^ Pi Pi P^ Pa

^A ^A \ B \ ^^A j^A \ B I C \ ^^ Aa\ B \ C \ B \

^ iji
t
^B
^ v^
t
^B
t
Vc
^^ t
Vb
tVc t
v^
(*) id) (/)

Fig. 11-4, Statically Indeterminate Beams.

Before going into the analysis of statically indeterminate beams by the

method of consistent deformation as described above, it will be necessary
to discuss methods of finding the vertical deflections or the slopes at various
points on the elastic curve of a statically determinate beam. For instance,
in Fig. and c, A^ and 5^ are merely deflections in a statically deter-
1 1-36
minate simple beam.
11-2. Deflections and Slopes in Statically Determinate Beams. There
are generally three methods by which the geometry of the elastic curve
(deflections and slopes) of statically determinate beams may be com-
puted; viz., the double-integration method, the moment-area method,
and the unit-load method. The double-integration method is usually
treated in the texts on strength of materials. Since this method is not
as convenient in its application as the other two, it will not be treated
here. The moment-area method, with the conjugate-beam method as
its special case, is perhaps the most powerful method of all. The unit-
load method is classic in nature and is especially useful as a spot check
 ANALYSIS OF STATICALLY IXDETERMIXATE BEAMS 233

on the results found by the moment-area method. Actually the method

to be used is a matter of individual preference.
11-3. The Moment-area Method. A simple beam AB subjected to
two concentrated loads is shown in Fig. ll-oa. The bending-moment
diagram for this loading is shown in Fig. 11-56. Points 1 and 2 at a
distance dx apart on the neutral axis of the unloaded beam will deflect
vertically downward to positions 1' and 2' when the beam is loaded. An
enlarged view of the deformed beam between the sections V and 2' is

(6) Bending moment diagram (c)

Fig. 11-5

shown in Fig. ll-5c. be shown that the angle dd between the

It will
V and 2' is equal to the area of the
tangents to the elastic curve at points
moment diagram between sections 1 and 2 divided by EI; or

dd
M dx (11-1)
EI
Because points 1 and 2 are at an infinitesimal distance dx apart, the
bending moment may be assumed to be constant on this short segment
and the elastic curve r-2' becomes a circular arc. In Fig. ll-5c, 2'-5'
is drawn parallel to I'-S'. Angle 5'-2'-4' is equal to the central angle at
0, which in turn is equal to the angle between the tangents at V and 2'.
Thus
arc 4'-5' arc 4'-5'
dd = angle 5'-2'-4' = ">'.:
(11-2)

But arc 4'-5' is

the total elongation of the lower extreme fiber, the orig-
inal length of which was 3'-5' or V-2'. The unit tensile stress at the
lower extreme fiber, from the flexure formula, is Mco/I. By Hooke's law,

arc V-o' = Mco/I (3'-5') (11-3)

ET"^'
234 ELEMENTARY THEORY OF STRUCTURES
Substituting (11-3) in (11-2)

de
arc 4'-5' Mc2 dx M dx
C2 EI
which is Eq. (11-1). It is to be
noted that Eq. (11-1) is general
in nature; points T and 2' at a
distance dx apart may be on the
elasticcurve of any portion of any
kind of a beam.
Let A'B' be the elastic curve of
an originally straight portion AB
in a beam of indefinite length (Fig.
ll-6a). The moment diagram for
(by Jl/-diagram
the segment AB is shown in Fig.
11-66 and the M/EI diagram in
Fig. ll-6c. Unless I is constant
throughout AB, the M/EI diagram
will not be similar to the M
diagram.
(c) £y -diagram
The angle between the tangent A'Bi
at A and the tangent at B' will be
Fig. 11-6
B'
and the vertical dis-
called [B]a'^' ,

tance B'Bx of point B' from the tangent at A' will be called [Ab' from
tangent at A']. It can be proved that

Wa'^' = ^FTF diagram between A and B (11-4)

EI
and
[Ab' from tangent at A']

= moment
M
of ^rr diagram between A and B about B (11-5)

Equations (11-4) and (11-5) are the two moment-area theorems.

Moment-area Theorem 1. The angle in radians or change in slope
between the tangents at any two points on a continuous elastic curve is
equal to the area of the M/EI diagram between these two points.
Moment-area Theorem 2. The deflection of a second point on a con-
tinuous elastic curve, measured in a direction perpendicular to the
original straight axis of the member, from the tangent at a first point
on the elastic curve, is equal to the moment of the M/EI diagram between
these two points about the second point.
Equations (11-4) and (11-5) will now be proved. Take points 1'
and 2' at dx apart on the elastic curve A'B' shown in Fig. ll-6a. Draw
tangents at 1' and 2' and prolong them to intercept a distance rfA on
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 235

B'Bi. Let dd equal the angle between the tangents at 1' and 2' and x
the distance from 1' or 2' to B. In Fig. ll-6a, it should be noted that
the vertical dimensions are very small when compared with the hori-
zontal distance and that the length of any curve or inclined distance
may be considered to be equal to its horizontal projection. By applying
Eq. (11-1) to Fig. ll-Qahc, it is seen that

Integrating both sides between the limits A and B,

rB CB
M
J A JA El^^
or [Q]a'^' =
M diagram between A and B
^FTf (11-4)
Ell

Also, from Fig. ll-6a,

d^ = xde

Integrating both sides between the limits A and B

rB
/ d^ = I
rB
xde =
rB
X
M dx
J A. J A. J A. EI
or

[Ab' from tangent at A']

= moment of ^^^
M diagram between A and B about B (11-5)

The application of the two moment-area theorems to finding slopes

and deflections of the elastic curves of statically determinate beams will
be illustrated by the following examples.
Example 11-1. By the moment-area method find the slopes of the
elastic curve at A and B and the vertical deflection at the center of the
beam shown in Fig. ll-7a.
SOLUTION. Since / is constant, the M /EI diagram is similar to the
M diagram and is as shown in Fig. 11-76. Applying the first moment-
area theorem,

= 2 wL'^ L wL^
Ba = Bb = [C]c'^ = area Ai
3 8^;/ 2 2^EI
Thus =
wU
Ba ^ /wx clockwise
24:EI

Bb = i^TTFTf counterclockwise
24:EI
236 ELEMENTARY THEORY OF STRUCTURES
Applying the second moment-area theorem,

Ac BiB = Ab from tangent at C = moment of area Ai about B

wL'
2iEI
A r = downward
16 384^;/

Example 11-2. By the moment-area method find the slope and deflec-
tion at B of the cantilever beam shown in Fig. ll-8a.
SOLUTION. The tangent at A is observed to be horizontal. Thus

Ob = Wa^' = area Ai L = rr^pTP clockwise

2EI " QEI
3
Ab BB' = Ab' from tangent at A = moment of area A about B i

wU 3 . wL^
downward
QEI4: SEI

Example 11-3. By the moment-area method find the slope and deflec-
tion under the load in the beam shown in Fig. ll-9a.

Ac Ib\J>>^^
Si
"'•?
L ' ~f
A4 * _* * *

Constant /
(a) Constant I

SEI

3/4 L

ill Ei
VieL
2EI
(6) ^-diagram (6) ^-diagram

Fig. 11-7 Fig. 11-8

SOLUTION. The bending-moment diagram for the beam in Fig. ll-9a

is shown in Fig. 11-96. In applying the moment-area theorems, the
M/EI diagram, not the M diagram, must be used; however, if / is

constant throughout the member, the M/EI diagram is similar to the

M diagram. In such a case, convenient to work with the properties
it is

of the M diagram and divide the results by EI to determine the correct

slope or deflection. If / is not constant, it is advisable to choose the
smallest value of / as a standard unit {h) and then express the other
/'s in terms of this unit (n/c). Thus the ordinates of the diagram M
are divided by the values of n to give the modified M diagram. In
slope and deflection calculations, the properties of the modified M dia-
gram can be used, except that the results should be divided by Eh.

1
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 237

In this problem, Ic = 200 in.^; n = 1 for portions AC and EB, and

n = 1.5 for portion CE. The modified M diagram is shown in Fig. ll-9c.
The location of the horizontal tangent is unknown in this beam, but
the deflections at A and B are known to be zero.

BBi = Ab from tangent at A

(moment of areas 1 to 6, inclusive, about B)
EL
1

EL [(180) (20) + (60) (17) + (90) (16) + (270) (12) + (108) (9)

+ (108) (4)]
kip-cu ft
= 10,704
EIc

dA
^ BBi ^
~ AB ~
10,704
24
= 446 \,j
hiic
— clockwise
In tracing the elastic curve from A to D' the tangent rotates progres-
,

I sively in the counterclockwise direction an amount equal to the sum

16k

60
1j5-/=200in.4
60

/=300in.4
12'

24'

VJi=10k Vn=6k
£= 30,000 k/in.2
(a)

90

(6) M-diagram {d) Elastic curve

Fig. 11-9

of the areas 1, 2, and 3, divided by EL, or

1 kip-sq ft

EL
(area of 1, 2, and 3) = 180 + 60 + 90 = 330
EL
Since there exists a clockwise rotation of 446 kip-sq it/ EL at A, and the
tangent rotates 330 kip-sq it/EL counterclockwise from A toward D',
the slope of the tangent at D' is

= ^^P-^q^^
Od 446 330 lie clockwise
EL
(116)(144)
= 2.784 X 10~^ radians clockwise
(30,000) (200)
238 ELEMENTARY THEORY OF STRUCTURES

^n = DD' = DDi - D,D' = (AD){dA) - [^D' from tangent at A]

= {AD){eA) - and 3 about D)
^Y (ii^onient of 1, 2,

= (9)(446_) - [(180)(5) + (60)(2) + (90)(1)]

= 2,904 T^y — downward

(2,904) (1,728) n QQr •
^ a
= = ^'^^^ ^^- downward
(30,000) (200)

For purpose of demonstration, Od and Ad will be found by tracing the

elastic curve from end B toward D.

AAi = A^ from tangent at B

= ^^^ (moment of areas 1 to 6, inclusive, about A)
tjlc

= ;^^[(180)(4) + (60)(7) + (90)(8) + (270)(12) + (108)(15)

+ (108) (20)]
kip-cu ft
8,880
EIc
. AA^ 8,880 ^_kip-sqft
— — counterclockwise
=
•

6b = -^ ^ = 370 ^Y^
, i .

An
. r,
Z-±
,

tj 1 c

In tracing the elastic curve from B to D' the tangent rotates progres-
,

sively in the clockwise direction for an amount equal to the sum of the
areas of 6, 5, and 4, divided by EIc, or

^ (area of 6, 5, and 4) = 108 + 108 + 270 = 486 ^^^1^,^ ^^

Since there exists a counterclockwise rotation of 370 kip-sq it/ EIc at B,

and the tangent turns 486 kip-sq it/EIc clockwise from B toward D'
the slope of the tangent at D' is

Bd = 486 - 370 = 116 —w^

tjlc
clockwise {check)

Ad = DD' = DD2 - D2D' = (BD){dB) - [Ad' from tangent at B]

= {BD){6b) — ^T (moment of 6, 5, and 4 about D)

= (15)(370) - [(108)(11) + (108)(6) -h (270)(3)]

= 2,904 ^^j
tjlc
— downward {check)

Example 11-4. By method find the slope and deflec-

the moment-area
tion at the free end of the overhanging beam shown in Fig. 11-lOa.
SOLUTION. The shear and bending-moment diagrams are shown in
Fig. 11-106 and c. Since / is constant for the entire length of the beam,
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 239

the M diagram need not be modified. For convenience the diagram M

in Fig. 11-lOc willbe decomposed into the elements shown in Fig. ll-lOrf.
The free-body diagram for BC (Fig. 11-lOc^) indicates that this segment
is subjected to a uniform load and an end moment, and for each of these

the M diagram is drawn separately as parts 1 and 2.

8.k

3k/ft
\ \ \ \ \ \ \ \ \ \

16'

35 k Constant / = 288 in .< 21 k

£= 30,000 k/in.2 21k
(a)

+ 27

{d) M-diagram in parts

(6) Shear diagram

73.5

(c) 3f-diagram (g) Elastic curve

Fig. 11-10

BBi = Afi from tangent at C

= -^Tj [(moment of 1) — (moment of 2)]
til
kip-cu ft
-

111
%(96)(16)(8) M(48)(16)(i^ 6,144
Er~
Be
BBj^
BC
6,144
16
= 384„^ . kip-sq
\^^
EI
— counterclockwise
ft . i i
•

OO 1

= Jj -El (^""^^ ^^
^EI ^''''^^ ^^ ^ EI ^^^^ ~ ^'^-^ + ^^^^

— 256 ^^
EI
— counterclockwise = 256 —^~r^— clockwise
EI
256 area 3
Ba (256 144) = 112^^-i-^lockwise
EI EI EI
(112)(144)
= 1.867 X 10~^ radians clockwise
(30,000K288)
(256)(6) H(48)(6)(4)
A. = AA AA,- A,A' =
EI EI
= ^. _ kip-cu
960
j^j
^— upward
ft , (960) (1,728)
(30,000) (288)
= 0.192 in. upward
240 ELEMENTARY THEORY OF STRUCTURES
11-4.The Conjugate-beam Method. A simple beam with some load-
ing shown in Fig. 11-lla. The conjugate beam for this situation, as
is

shown in Fig. 11-116, is this same simple beam loaded with the M/EI dia-
gram for the actual loading on the beam as shown in Fig. 11-lla. It will
be shown that the -positive shear at section C (Fy of the conjugate beam is
equal to the clockwise rotation at C {6c) of the real beam, and the positive
bending moment at C (M^) of the conjugate beam is equal to the downward
deflection at C (Ac) of the real beam.

Applying the moment-area theorems,

6c = Oa

_ moment
- [^ area on
of M/EI
AC') =
1
=
area on
s-r-^

AB
-
- ([yj area on AC ]
about B / M
(
^

J
.
^^ area on AC
and
Ac = CC = CCi - CiC = OAiAC)

-(moment of ^ area on AC about C

moment of M/EI area on AB about B
(AC)

— moment ( of yj area on AC about C 1

The above two expressions for 6c and Ac are actually the shear and
bending moment at section C of the conjugate beam. Thus

6c = V'c (11-6)
Ac = M'c (11-7)
also 6a = Y'a (11-8)
6b = V'b (11-9)

Although A and B in the above derivation are the two end points of a
simple beam, they could have been any two points with zero deflection
on a continuous elastic curve. In fact, in the general case, the conjugate-
beam method may be applied between any two points with known deflec-
tions. In this event the rotationmeasured from the direction of the
is

straight line joining the two points on the elastic curve and the deflection
perpendicular to the original straight axis of the member is measured
from this same straight line.
Example 11-5. By the conjugate-beam method find the slope and
deflection under the load in the beam shown in Fig. ll-12a.
SOLUTION. The M
diagram and the modified diagram on the basis M
of Ic = 200 in.^ are shown in Fig. 11-126 and c. Applying the conjugate-
beam method to the elastic curve AB of Fig. ll-12c,
 ANALYSIS OF STATICALLY INDETERMINATE BE.\MS 241
16 k

A JST ^ -2^B
/- 200 m.^ /=jo; '.n.-* / = 2rOin.-« 1
6' 12' e'

1
•

^i ^i
ib, Cczy^gi-.e beaza (c; Modified M-diagrzm and elastic cunre

Fig. 11-11 Fig. 11-12

(108)(4) + (108>(9^ 4- (270H12) -^ f90)(16) + (60) (17)

+ (180) (20)
^1 = 24i:/c
kip-sq ft
= 446
EI.
+ (90) (8) -f (270) (12) + (108) (15) + (108) (20) ~
n=
., (180) (4) -f (60) (7)
24£'/c

= 370^^^
''' ^ ^^
= Vn = VI- ^.^ = (446 - 330)
en
EL ^^
^^
= 116 ^V7
Eh
= 2.784 X 10-' radians clockwise

^D = 3/i> = 9F1 - ^^ [(90) (1) + (60) (2) + (180) (5)]

(4,014 - 1,110)
EI.

= 2,904 ^^7 ^^
= 0S36 in. downward
242 ELEMENTARY THEORY OF STRUCTURES
The above values of On and Ad (or F^ and M'^) could have been found
by using DB, instead of AD, as the free body.
Example 11-6. By the moment-area and/or conjugate-beam method
find the slope and deflection at the free end of the overhanging beam
shown in Fig. ll-13a.
SOLUTION. The shear and bending-moment diagrams are shown in
Fig. 11-136 and c. Since I is constant, the M diagram need not be
modified for slope and deflection computations. The conjugate-beam

8k
H 3k/ft

XV ' ^ ^

16'
^ ^ ^ ^
^x^
f
35 k 21k
Constant /- 288 in.^
£= 30,000 k/in.2
(a)
+ 27

-21 384

(6) Shear diagram (d) 3f-diagram in parts

73.5

(e) Elastic curve

(c) JVf-diagram

Fig. 11-13

method will be applied to the elastic curve BC. The elastic curve BC
shown in Fig. ll-13e is considered to be the composition of the two elastic

curves shown in Fig. ll-13d. Thus

Obi ~ 0b2 V — VB2

^ [^(1,024) - %(384)] = 256 ^^g ^^
clockwise

Applying the first moment-area theorem between B and A',

Oa = Bb
di - ^ {M area on AB) = ^ (256 - 144)

112 „j — clockwise
1.867 X 10"^ radians clockwise
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 243

M(48)(6)(4) 1,536 - 576

A^ = AA' = AAi - AiA' = QOb - ^
EI EI
^^P-'^ ^^
= 960 upward
EI
= 0.192 in. upward

A comparison Examples 11-5 and 11-6 with Examples 11-3 and 11-4
of
indicates that the two solutions for each problem are really identical.
The conception of the conjugate beam, when applied between two points
of zero (or known) deflections, seems to give the required results more
directly than the moment-area method; however, the latter deals phys-
ically with the elastic curve at almost every step, and this is sometimes an
advantage.
11-5. The Unit-load Method. Let it be required to find the deflection
Ac or the slope dc (Figs. 11-146and 11-156) at point C in a simple beam
AB carrying two concentrated loads Pi and P2 as shown. In either

J ^ P2 ^1 1 P2

1 '2 2
+ -

4^^L^
firirVr~^'
+
4" ^-V" ^U'
dx dl dx + dL
"vh&p^
hc-^ ^C dx+dl + dL
62+^2

(a) (c)

Fig. 11-14

Unit couple Unit couple

____?L, ^.h-.c t2. _J_. = _4.LlP u+S^ 'U+S

2_
^

¥i dx+dl
B A
dx + dL
1/ -i^. T^Yyb
82+^2
dx + dl + dL
fa) (c)

Fig. 11-15

Fig. 11-146 or 11-156, Pi and P2 cause deflections Ai and A2 at points 1 and

2, and an internal tensile stress *S in a typical fiber with a cross-sectional
area dA and of an original length dx, but which is elongated by an amount
dL. When Pi and P2 are gradually applied to the beam, by the law of
conservation of energy, the total external work done on the beam must
be equal to the total internal elastic energy stored in the beam, or

MPiAi H- M^2A2 = ^VzS dL (11-10)

A unit load or a unit couple applied at C will cause deflections 5i and 62

at points 1 and 2 and a deflection be or rotation 0c at point C as shown in

244 ELEMENTARY THEORY OF STRUCTURES
Figs. ll-14a or 11-1 5a. A typical fiber with an original length dx and a
cross-sectional area dA will be subjected to an internal tensile stress u
and an elongation of dl. Again, with gradual application of either the
unit load or unit couple, the external work equals the internal energy.
Accordingly,

H(l)(5c) = i:y2udl (ll-lla)

or 3^(l)(<^c) = i:y2udl (11-116)

Now if the unit load or the unit couple is first applied to the unloaded
beam and the loads Pi and P^ are then added, the resulting condition
will be as shown in Figs. ll-14c or ll-15c. The external work done on
the beam due to the gradual application of the unit load or the unit
couple is 3^(l)(5c) or H(l)(0c), w^hile the internal energy is 'Zyu dl.

Because Pi and P2 go through the displace-

ments Ai and A2 when the forces increase
gradually from zero to Pi and P2, the external
work done by Pi and P2 is J^PiAi + J^P2A2.
The external work done by the unit load or
the unit couple, which is already on the beam,
in going through the additional displacement
Elongation
Ac or rotation dc is (l)(Ac) or (1)(M- The
Fig. 11-16 work therefore
total external is

3^(1) (5c) + M^iAi + M^2A2 + (l)(Ac)

in the case of Fig. ll-14c, and H(l)(0c) + M^iAi + M^2A2 + (1)(M in

the case of Fig. ll-15c. Bythe same reasoning, the total internal energy
is X^u dl + xyS dL + ^u dL. (In fact, the internal energy in a typical
fiber can be represented by the area of the force-displacement diagram as
show^n in Fig. 11-16.) Equating the total external work to the total
internal energy,

M(i)(5c) + H^iAi + M^2A2 + (i)(Ac) = i:}iudi + i^ysdL

-\-XudL (ll-12a)
or

H(i)(<Ac) + H^iAi + MP2A2 + Wide) = ^yiudi + ^y2SdL

+ XudL (ll-12b)

Subtracting Eqs. (11-10) and (U-ll) from Eq. (11-12),

(l)(Ac) = ^udL (ll-13a)

or (l)((9c) = ^udL (ll-13b)

It must be pointed out that the value of u in Eq. (ll-13a) is different

from the value of u in Eq. (ll-13b); one is the stress in a typical fiber
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 245

due to the unit load, and the other is the stress in a typical fiber due to the
unit couple.
Let M equal the bending moment in the beam due to Pi and P2 and m
that due to the unit load or the unit couple. Then, in a typical fiber,

u = ^ dA
and dL = —v- ^ dx
Substituting the above expressions for u and dL in Eq. (11-13),

(l)(A.)=f c^A^d.

aKec) =
f"^ (ii-i4b)

Note again that the m in Eq. (ll-14a) is the bending moment due to
the unit load and the m in Eq. (ll-14b) is due to the unit couple.
The application of Eqs. (ll-14a) and /\

(ll-14b) to the calculation of the deflec- w 1 * 1 4 t i i i i i ~^~j n

tionsand slopes (or rotations) in statically |
l j
determinate beams will be illustrated by constant/
the following examples. (°^

Example 11-7. Using the unit-load (a^ ^

method, calculate the slopes of the elastic \* ^
curve at A and B and the vertical deflec- f 7-
^
(6) m for ^^
tion at the center of the beam shown m
•
1 r ^ ^ i

l i l
'^ '^
Fig. 1 l-17a. ^ h 4- H „

SOLUTION ^
t- \
V2 1/2
(c)mforAc
Mmdx over AB
,^ Fig. 11-17
EI
^ - -
= /
\{wLI2)x
^ (^a:V2)][l
^
^^—^
(1/L)x] dx
—^ ——^— With origm at A
or Oa = —^ -
p \{wLl2)x
-
(i/;xV2)][(l/L)a:]
^
_; ^^^^ ^ ^
-*
—
dx .^. .
^
with origin at, B
.

Jo rji

clockwise
24£'/

A positive result indicates that the direction of Oa is the same as that

of the unit couple, clockwise in this case.
246 ELEMENTARY THEORY OF STRUCTURES

Ac = /
f Mm dx over AC
Ar^ X

-\-
[
\
-^^ d^
over
r^n
BC
^j ^
L L
[ H{wL/2)x - wxy2WAx)dx
{ (^ [{wL/2)x-{wxy2WAx)dx
'^
Jo EI Jo EI
origin at A origin at B
'
[iwL/2)x - iwx^/2)](}4x) dx bwL* ,
= 384B7 downward
L Wl
A positive result indicates that the direction of Ac is the same as that
of the unit load, downward in this case.
Example 11-8. Use the unit-load method to calculate the slope and
deflection at B of the cantilever beam shown
in Fig. ll-18a. , f-i-x
SOLUTION H
Constant /
Mm dx over AB^ . (a)

EI
'
(— V2)(-l)
EI
dx _ wU
6EI
^^^^^^.^^
^\ 0^
(6) m for 6q
origin at B

^B - j
f Mm
-^^dx .^
over AB I

^ 1

^ r i-wx^m-x) dx ^ wL^ downward '^'""'-^^

Jo EI 8EI Fig. 11-18
origin at B
Example 11-9. By the unit-load method calculate the slope and deflec-
tion under the load in the beam shown in Fig. ll-19a.
SOLUTION. At this time it may be advisable to examine the dimen-
sional units in Eqs. (ll-14a) and (ll-14b). In a numerical problem,
when the dimensional units are like those in Fig. 11-19, the dimensional
equation of (l)(Az>) = f Mm dx/ EI is

^
(ksi)(m.^)

The dimensional equation of (l)(^z)) = fMm dx/EI is

/.. lUN/ A- ^
=
(kip-ft)(ft-lb)(ft) ^
X
...
(ft-lb) (radian)
^ ^^ n --w-—4Y^
(ksi)(m.^)
l-^^

Referring to Fig. 11-19,

0. ^ ^^-^^ on AC + on CD + f''^f\nDE
f f'±^-^
+ JM^ on EB

1
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 247

16 k

Vd E B 8k
U ^ — H 3k/ft
1.5 /c Ic
jt f
^ t * i i ^ * * i r
12'

24'
6'

35k
P 16'

?
21k
V^ = 10k V„ = 6k
/ = 200 in.4
Constant /= 288 in.'*

£= 30,000 k/in.2 £= 30,000 k/in.2

(a) (a)

1 ft-lb
1 ft-lb-

— B
r

i
„<5^

I
1/24 lb
B

1/24 lb
A

X1/16 lb

(6) m for 9/1

1/16 lb

(6) m for Bd

lib

A B A

t
t t
13/8 lb 3/8 lb
5/8 lb 3/8 lb
(c) m for A© (c) m for Ayi

Fig. 11-19 Fig. 11-20

EU, = r (lOx) (_ ^ x) dx + i-5 /J (lOx) (- ij x) rfx

origin at A origin at A

+ i^f (6x)(2^x).x + /;(6x)(A.),.

B B
origin at origin at

= 116kip-sqft

(3(

A,=
= ,^^^onAC+ [M^onCD^ [ 'J^
EI
on DE
/ EI EI
,

/''
,
Mmdx ^o

E/.A. =
/J
(lOx)
(I
origin at
x) dx
A
+ i-^
£ (lOx)
(I
origin at
x) dx
A

origin at B origin at B
= 2,904 kip-cu ft

Example 11-10. Using the unit-load method calculate the slope and
deflection at the free end of the overhanging beam shown in Fig. ll-20a.
248 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. Referring to Fig. 11-20,

Mm dx Mm dx on BC
d.
/= EI on AB 4-
/ EI
16

EIOa = (~Sx)(^l) dx -}- (2lx-Mx'){Hex)dx

origin at A origin at C
= 112 kip-sq ft

= (112)(144)
Ba = 1.867 X 10~^ radians clockwise
(30,000) (288)
Mm dx on AB Mm dx on BC
Aa -{ EI -h
/ri6 EI
EIAa = (-Sx)(-x)dx 4-
/
{2\x - y2x')(-ysx) dx
origin at A J origin at C
= —960 kip-cu ft downward or 960 kip-cu ft upward
(960) (1,728)
Aa = = 0.192 in. upward
(30,000) (288)

11-6. Law of Reciprocal Deflections. As applied to beams the law of

reciprocal deflections has three different versions. These are:
1. If 8ab is the deflection at A due to a unit load at B, and dsA is the
deflection at B due to a unit load at A, then Bab = ^ba-
Unit couple Unit couple

B A B A
5b a] ^.''"^
.^'i
5ba

mA " bending moment due to niA = bending moment due to TTiA = bending moment due to
unit load at A unit couple at A unit couple at A

Unit couple

A R

---I'^AB
*^AB
mg - bending moment due to mg - bending moment due to mg - bending moment due to
unit load at B unit load at B unit couple at B

(a) (b) (c)

Fig. 11-21

2. If <t>AB is A due to a unit load at B and 5ba

the rotation at is the
deflection at due to a unit couple at A, then (t)AB = ^ba-
B
3. If 0AB is the rotation at A due to a unit couple at B and (t>BA is the
rotation at B due to a unit couple at A, then <f>AB = <t>BA-
The proofs of these three statements can be very simple. Referring
to Fig. ll-21a.
Let mA = bending moment due to unit load at A
ms = bending moment due to unit load at B
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 249

Then = f Mm dx 1 msmA dx
^AB
EI - j EI
-J
and ^BA =
f Mm dx 1 mAmB dx
-J EI -J EI
Therefore ^AB = ^BA
Referring to Fig. 11-216,

f Aim dx f mAms dx
8ba = ~
J EI J El
f Aim dx f msmA dx
<f>AB = '

J EI J EI
Therefore <f>AB
= ^BA
Referring to Fig. 11-21C,

= f Aim dx f mAms dx
(i>BA -
j EI J EI
= f Mm dx f msmA dx
<t>AB
EI ~ EI
-J J
Therefore <i>AB
= 4>BA

The second statement is sometimes puzzling because one wonders how

an angle in radians can be equal to a deflection in, for instance, feet.

In fact, for <I>ab = ^ba to hold true, if the unit load at B is 1 kip, the unit
couple at A must be 1 kip-ft.
Example 11-11. Referring to Fig. 11-22, verify all reciprocal relations
after determining the deflection and rotation at B due to a 1-kip load or a
1-kip-ft couple at A, and the deflection and rotation at A due to a 1-kip
load or a 1-kip-ft couple at B by the conjugate-beam method.
SOLUTION. Applying the conjugate-beam method to Fig. ll-22a,

EldBA = M's = (16) (3) - M(l)(3)(l) = 46.5 kip-cu ft

^- =
EI<f>BA =
mrnm
= -
7^
=
16 M(l)(3)
''''
=
"" ''~'
''

14.5 kip-sq ft
'"^""^^^

(14 5) (144)
^^^ ^ ^ ^^^^ ^ ^^"^ radians counterclockwise
(30 000)7288)

From Fig. 11-226,

Eld^,^ = AI'^ = (2)(3) - y2(H){S)(l) = 5.75 kip-cu ft

EI<t>'BA = F^ = 2 - y2{H){3) = 1.75 kip-sq ft

<f>BA = "" ^^^ ^ ^^~^ radians counterclockwise

(30 000) (288)
250 ELEMENTARY THEORY OF STRUCTURES
Ik Ik

B I L
'i'BAU
18'

1/6 k 5/6k
2/3 1/3
£= 30,000 k/in.2 /=288in.4
4k-ft

2.5k-ft

20k-ft2 16k-ft2 8.75k-ft2

(a)

Ik-ft

'BA
\B
"^BA-

1/18 k Vi8k Vl8k

+ 2/3k-ft
+ l/6k-ft

Vsk-ft -i/eTfTj
1 k-ff^ 2 k-ft2 2.75 k-ft2 3.25k.ft2

ib) (d)

Fig. 11-22

From Fig. 11 -22c,

EUab = M'a = (8.75) (6) - H(l)(6)(2) = 46.5 kip-sq ft

8ab = 0.775 X 10-3 ft downward

EIct>AB = V'a = 8.75 - 3^(1) (6) = 5.75 kip-sq ft
(j)AB = ^^% X 10~^ radians clockwise
From Fig. ll-22c?,

EId[AB M[A - (2.75)(6) - M(M)(6)(2) = 14.5kip-cuft

^'ab = "^^^^ X 10-6 ft downward
EH'ab = V'a 2.75 - >^(>^)(6) = 1.75 kip-sq
= ft

0^^ = 1'^% X 10-6 radians clockwise

Thus the four reciprocal relations which are now verified are

hBA = ^AB ^BA = ^AB <t>BA = ^AB ^ud ^5^ = 4>a 3

11-7. statically Indeterminate Beams with One Redundant. The

analysis of statically indeterminate beams with one redundant by the
method of consistent deformation will be illustrated by the following two
examples.
Example Analyze the statically indeterminate beam shown in
11-12.
Fig. ll-23aby the method of consistent deformation. Draw the shear
and bending-moment diagrams and sketch the elastic curve.
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 251

4 k /ft

18'
A!, \
c ^^ -"% »N=iw5c
t
Constant / Ki = 39 k VB=57k
(a) (d) Basic determinate beam
under action of redundant (/) Free-body diagram

4k/ft

(6) Basic determinate beam

under the applied load
33
(g) Shear diagram

64.125

(e) iW-diagram

'1152 k-ft
126
(c) Af-diagram of (6)
(h) Bending moment diagram

Fig. 11-23

FIRST SOLUTION. If Vb is chosen to be the redundant, the support at

B is assumed to be removed and the basic determinate beam becomes a
cantilever beam fixed at A only. The given beam of Fig. ll-23a is the
sum of those in Fig. 11-236 and d. Equating A^ to Vb^b,

F. = ^
Ob

From Fig. ll-23c,

EIAb = moment of area A1A2B1B2 about B

= moment of area A1A2C1 about B
— moment of area B1B2C1 about B
= (M)(l,152)(24)(12) - M(V2)(6)(-1.5)
= 110,592 216 = 110,808 kip-cu ft
+
From Fig. ll-23e,

EHb = moment of area A1A2B1 about B

= 3^(18) (18) (12) = 1,944 kip-cu ft

^, EIAb 110,808 .- , . .
Thus Vb = ^^. = 1. = 57 kips upward
EUb .
1,944

By statics (Fig. 11-23/),

Fa = 96 — 57 = 39 kips upward
Ma = (96) (12) - (57) (18) = 126 kip-ft counterclockwise

The shear and bending-moment diagrams are shown in Fig. 11-23^

and h, and the elastic curve is sketched in Fig. 11-23/.
 VUEMEaSTAKW TBEJOm €fF SOBSXJWVIfEES

SDCO^CD saumioa:. If If^ is Miwniigd to be the redandbnt^ die

in fig. ll-2la beeomes the ann of Uie two curednngpng
in ng. ll-2lfr and ^ From fi&. ll-2lfr and c;

= 972 - 216 ^ 756 kip^ fit

From Fig- ll-24f

^/i^ji = Fi = 6 kqHHi fit

?i-ff

-^ -, r?.

5i«:afci

By arftatics (Kg- 11-2^.

126 + 96(6)^ ^,^

ExamtpU 11-13- Analj^ze Ithc fitalirally indetenninafte beam XL

Fi& ll'25a bf «lie mediod of eonaitenlt drfonnalian. Dnw llie

and benfing-mnnenlt dbgraniffi and dkeltcfa Idle dbstic carve.
wussrmxxums. H F* k dioaai as the redondanl^ the beam of
the two simple beams
g^
Fig- ll-25« may be eoDfflidered to be eqpivallent to
ll-25«, Ai.
diavninFig.ll'^&andd. For iiaro defltttion at il in Fig.
in F«. 11-256 mart be equal to Vm^m in fi& ll'25d;
thn
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 253

I R = -;r-

Afi is conveniently calculated by the conjugate-beam method. Portion 4

of the bending-moment diagram in Fig. ll-25r is equal to the bending-
moment diagram of a simple beam with span of 12 ft and subjected to a

15 k 15k
1
4' I 4' ,-T ^ *
}' • • • i^>^ ft ,'-' ]Vph^ |- • • '! 15k ft

L ?^ J_ r2[ J I ---'
t t t
4.35 k 21,75 k 6.90k
ConstAnl /
(d) Free-body diagram
(f)

15 k + 11.10

*' *'
\
'
\ 1.5k ft
I

—r-:2^

(-15.8?) -6.90
(b)
-10.65
(£) Shear diagram

79.2
e9.6
^v
(2)^ vT"
M
^ ' : 6'
V;-25.6
1 !

25.2
1 27S 4/15 S>^475 2 : :i6

\;i-5S3.2 22/3' 11/3' V<^-544.S (h) Bending moment diagram

(c) M-d\&grim of (6)

Fig. 1-125

P, Py P^ P,

U_i ' '{-"!» A. ^^^^

v'n \:

(«) (6)

Fig. 11-26

uniform load of 1.5 kips per ft. This can be proved by comparing the
bending moments at point C of the beams shown in Fig. ll-26a and h,

{Mc in Fig. ll-26a) - (J/c in Fig. 11-26?))

In the above equation, it is seen that the difference in the bending

momenta at C in Fig. ll-2Ga and 6, which is the vertical intercept within,

for instance, area of portion 4 in Fig. ll-25c, is equal to the bending

254 ELEMENTARY THEORY OF STRUCTURES
moment in a simple beam with span equal to h and subjected to a uniform
load. From Fig. ll-25c,
EISb = M's = (583.2)(8) - (158.4)(M) - (278.4)(4)
= 3,340.8 kip-cu ft

In Fig. ll-25c?, 6b is used in the sense that it is the upward deflection

at B due to a 1-kip upward load at B however, this is numerically equal
;

to the downward deflection at B due to a 1-kip downward load at B, as

shown in Fig. ll-25e. Referring to Fig. ll-25e,

EI 5b = M'b = (25.6) (8) - (19.2) (2?^) = 153.6 kip-cu ft

rj.. j^ EI\b 3,340.8 .^, -_ , . ,

thus B = t^tT~ = 'T^^o~F~ ^ 21. /o kips upward
^^-E~n
\
hi Ob loo.b

By statics (Fig. 11-25/),

Va = 4.35 kips upward

Vc = 6.90 kips upward
The shear and bending-moment diagrams of the given beam are shown
in Fig. 11-25^ and h, and the elastic curve is sketched on Fig. 11-25/.
SECOND SOLUTION. When T"^ is chosen as the redundant, the given
beam of Fig. ll-27a is the equivalent of the two overhanging beams
shown in Fig. 11-276 and d.

From Fig. ll-27c,

EIOb = 2^(area 2) -H(area 1)

= 2^(360) - M(216) = 132 kip-sq ft

EIAa = EldBiS) + moment of area 3 about .4

= (132) (8) 4- (120) (62^) = 1,856 kip-cu ft

From Fig. ll-27e,

EIOb = %(area 4) = %(48) = 32 kip-sq ft

EI8a = SEIOb + moment of area 5 about A
= 8(32) + 32(5M) = ^ kip-cu ft

Thus F^ = 1^ = j^ = 4.35 kips upward

By statics,
Vb = 21.75 kips upward
Vc = 6.90 kips upward
11-8. Statically Indeterminate with Two Redundants. The
Beams
analysis of statically indeterminate beams with two redundants by the
method of consistent deformation will be illustrated by the following
two examples.

i
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 255

15k

'
T i I i i i I Ar r^C
i?
12'

Constant /
(a) (d)

15 k

1.5k/ft
^^j^^i-.l
Jr
i^ I I ^A

Ik
(b)

(e)

15 k

^ 1.5 k/a

! f t
60 4.35 k 21.75 k 6.90 k

(c) Af-diagram of (6),in parts f/j Free-body diagram

Fig. 11-27

Example Analyze the statically indeterminate beam shown in

11-14.
Fig. ll-28aby the method of consistent deformation. Draw the shear
and bending-moment diagrams and sketch the elastic curve.
SOLUTION. The beam in Fig. ll-28a is statically indeterminate to the
second degree, because it has four unknown
Ma, Va, reaction components
Vb, and Vc, while statics provides only two independent equations of
equilibrium for a coplanar-parallel-force system. When Vb and Vc are
chosen as the redundants, the given beam in Fig. ll-28a becomes the
composite of the three cantilever beams shown in Fig. ll-2Sbcd. The
notation 8pq will be used to designate the deflection at P due to a unit
load at Q. Thus the two equations for consistent deformation are

Ab
and Vb^cb -} Vc^cc Ac
Values of As, Ac, ^bb, ^bc = ^cb and bcc will be found by the moment-
area method. Referring to Fig. 11-286,

EIAb = moment of areas 1, 2, 3, and 4 about B

= (624)(6%) + (360)(5M) + (360)(2%) + (216)(li^)
= 7,328 kip-cu ft
EI^c = moment of areas 1, 2, 3, 4, and 5 about C
= (624)(18%) + (360)(17i^) + (360)(14%) + (216)(13H)
+ >i(108)(12)(9)
= 29,936 kip-cu ft
256 ELEMENTARY THEORY OF STRUCTURES
15 k
4'
I
4' ik

Al
|

* i
1.5k/ft
i i n M
12*
^BC

Constant /
(a)
i
15k

. , 1.5k/ft
' ' 15 k
1 ---V--
-^ ' i' 4'

Ab

6k 19 11/12 k 7Vi2k
(g)

+ 1011/12
+ 6 4.72'
—^^
'
,
"I
1

"-^
71/12
(/i) Shear diagram

13 16.72

yc^Bc
11
23
C (ij Bending moment diagram
30 27
Vc
(d)

Ik
hB c
'U-
JSCB F^^
-® 23
8
(c) 0) Af-diagram in parts

Fig. 11-28

Referring to Fig. ll-28e,

ElbsB = moment of area 6 about B

= (32) (53^^) = 170?^ kip-cuft
ElhcB = moment of area 6 about C
= (32)(17M) = 55-4% kip-cu f
Referring to Fig. 11-28/,

Eldsc = moment of areas 7 and 8 about B

= (80) (oM) + (•i8)(2%) = 554% kip-cu ft

Elbcc = moment of areas 7, 8, and 9 about C

= (80)(17>^) + (48)(14?^) + (72)(8) = 2,6662^ kip-cu ft

Note that 8cb and dsc have been computed independently, but, by the
law of reciprocal deflections, they are naturally equal.
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 257

The two equations for consistent deformation become, numerically,

1^4 ^^ + 8^0 ^^ ^ 29,936

Solving,
Vb = 19^ H2 kips upward
Vc = 7^2 kips upward
By statics,

Va = S3 - Vb - Vc = Q kips upward
Ma = (15) (4) + (18) (14) - SVb - 20 Fc
= 11 kip-ft counterclockwise

The free-body, shear, and bending-moment diagrams for the given

beam are shown in Fig. ll-2Sghi and the elastic curve is sketched in
Fig. 11-28^.
A check on the correctness of the above solution could be made by
choosing, say, Va and Ma as the redundants. It is preferable, however,
to check the consistency of the elastic curve by use of the bending-moment
diagram in Fig. The bending-moment diagram in Fig.
ll-28z. ll-28i
may be resolved into parts as shown in Fig. ll-28j. Applying the conju-
gate-beam method to span AB,

EIBa = M(area 10) - %(area 12) - M(area 13)

I = 60 - M(44) - 3^^(92) =
EIBb = H(area 10) - M(area 12) - M(area 13)
= 60 - H(44) - %(92) = -16 or 16 kip-sq ft clockwise

Applying the conjugate-beam method to span BC,

EIBb = M(area 11) - %(area 14) = H(216) - %(138)

= 16 kip-sq ft clockwise

The fact that ^a = and Bb in span AB i^ equal to Bb in span BC is

assurance that two conditions of geometry have been satisfied in this

beam which is statically indeterminate to the second degree. Thus the
correctness of the solution is ensured.
Example Analyze the statically indeterminate beam shown in
11-15.
Fig. ll-29aby the method of consistent deformation. Draw the shear
and bending-moment diagrams and sketch the elastic curve.
SOLUTION. With four unknown reaction components and only two
equations of statics, the given beam is statically indeterminate to the
second degree. If Ma and Mb are chosen as the redundants, the given
beam with fixed ends becomes the composite of the three simple beams
258 ELEMENTARY THEORY OF STRUCTURES

Mb
^-4f T^
Constant /
(a)

±
-f^^-^^t _
_ _ ^P\^^^^

JJmh-^
V^-^(3a-f6) ^B-^(36 + a)
(O
+ ^(3a.6)

(36+0)
(f) Shear diagram
2Pa263

P"
Pa6^U^ ^^^
fg^ Bending moment diagram

Fig. 11-29

shown in Fig. ll-296cc?. The two conditions for consistent deformation

are
^A = (f>Al -{- <i>A2

0B = (t>Bl + 052

Applying the conjugate-beam method to Fig. 11-296,

Pah
eioa = y'A =
l 6L
(a + 26)

Pab^ / _^ 6\ Pa'h (2 \ Pah

eibb = y'B = T
2L
,

6L
(2a + h)

From Fig. ll-29c,

EHai = V'ai = MaL

3

EIcl>Bl = Tbi = MaL

From Fig. ll-29c?,
MbL
EI(f>A2 = Va2
6
MbL
EI(t> B2 v\52
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 259

Substituting,
MaL MbL Pah
(a + 26)
6L
MaL MbL Pah
(2a + h)
6 "^ 3 6L
Solving,
Pab^ Pha^
Ma = and Mb L2

The free-body, shear, and bending-moment diagrams, and the elastic

curve are shown in Fig. ll-29e/gf.

11-9. Influence Diagrams of Statically Indeterminate Beams. Let it
be required to construct the influence diagram for the reaction at B of
the continuous beam ABC shown in Fig. ll-30a. To determine the

Ik Ik

1 Tmr ^ ». ^
^c
T hpB
/A'}/ I
5bb

Va
(d) Elastic curve for unit load
(a)
Ik at B of simple beam AC

. i s
^««^ >wt
5bp

(b)

(e) Influence diagram for reaction

^BB at B oi continuous beam ABC
%r
Ik
(c)

Fig. 11-30

influence diagram, it will be necessary to compute the values of Vb for

various positions of the unit load on the beam. It would appear, then,
that each ordinate of the influence diagram must be evaluated by a
statically indeterminate analysis. For instance, for a unit load at point P
in Fig. ll-30a, Vbequal to 8bp in Fig. 11-306 divided by Bbb in Fig.
is

ll-30c. In the influence diagram for Vb (Fig. ll-30e), this value of

Vb = ^Bp/^BB is the ordinate at P. However, by the law of reciprocal
deflections, hBP equal to bps] thus the influence ordinate Vb at P is
is

equal to ^pb/^bb, in which the numerator is the deflection at P due to a

unit load at B and the denominator is a constant. The elastic curve
for a unit load at B as shown in Fig. ll-30d gives the values of 8pb along
the beam. The influence diagram for Vb is obtained by dividing all

ordinates in the elastic curve by Sbb-

260 ELEMENTARY THEORY OF STRUCTURES
This discussion again demonstrates the fact that influence diagrams
are deflection diagrams. Thus, in this case, the influence diagram for
Vb may be obtained by introducing a unit deflection at point B in the
simple beam AC.
If it is required to construct the influence diagram for the reactions at
B and C of the continuous beam ABCD shown in Fig. 11-3 la, it wiU be
necessary to compute Vb and Vc for a unit load at any position such as
P. Applying the two conditions
deformation to Fig.
for consistent
50'
n-Slbcd, 4
Vb^BB + Vc^BC — ^BP ^
Constant /
Vb^cb ~h Vc^cc = ^cp (a)

Ik

A i B C D
r^ P
lb) Influence diagram for Rq
(a) 1.000

Ik

hp k:p ,^'^

(b)

^B^B
A Vi ^y~ Vb^cb D
(d) Influence diagram for Rq

Vb + 6.11
(c)

Vc^BC:. ^yc}cc ^D (e) Influence diagram for Mj

-2.08
(d)
Ik -G.OO
(f) Influence diagram for ^3
A P B + 8.40
^Q^. t-^'^
5qb
mB ^CB I

(e)

Fig. 11-31 Fig. 11-32

All the 6 quantities in the above equations may be taken from the
elastic curve for a unit load at 7^ on a simple beam AD, because, by use
of symmetry and the law of reciprocal deflections,

^BC = ^CB
^CC — ^BB
5np = ^PB
^CP = ^BQ = ^QB
 .

ANALYSIS OF STATICALLY INDETERMINATE BEAMS 261

Note that points P and Q are symmetrical with respect to the center
line of the beam. Thus, for a symmetrical three-span continuous beam,
a single elastic curve as shown in Fig. ll-31e will supply all the necessary
data for computing the values of Ik
Vb and Vc for various positions of
the unit load. 1 2 3 ' 15 6 7 8

Example 11-16. Construct the ^''^'

\ ,^23 533
.
043
533 \
23.75V.^ ^-^ -'20,250
influence diagrams for Rq, Rz, Rs, J, J,
o ^i''3'8,250 6£/ J^

Ml, Ms, and il/5 in the continuous d 43,750 " -

-''51.750 6£/ °
6£/ 56,250 58.500
6£/
beam shown in Fig. ll-32a. 6£/ 6£7
(a)
SOLUTION. Although any one of
the three reactions can be made the 18.75

redundant, for convenience Rz will

be chosen as the redundant. In
Table 11-1 are shown the values of
Rz, Ro, Rs, Ml, Mz, and M5 due to
a unit load at the successive points.
The influence diagrams shown in
Fig. 11-32 are plotted by using the
values of the influence ordinates in
Table 11-1.
To illustrate the procedure for
making these calculations, the com-
putations for the values of Rz, Ro, Rs, Mi, Mz, and M5 due to a unit load
at point 4 (Fig. ll-33c) will be shown. ^7543 in Fig. ll-33a is equal to
the bending moment at point 4 in the conjugate beam shown in Fig.
11-336.

Eldiz = (343.75)(40) - M(15)(40)(4%) = 9,750 kip-cu ft

Table 11-1. Influence Table

Load at R. ^8 Ml il/3 M,
033

+ 1.000
1 +0.422 +0.611 -0.033 +6.11 -1.66 -1.00
2 +0.778 +0.264 -0.042 +2.64 -2.08 -1.25
3 + 1.000
4 + 1.040 -0.150 +0.110 -1.50 -4.50 +3.30
5 +0.920 -0.200 +0.280 -2.00 -6.00 +8.40
6 +0.680 -0.175 +0.495 -1.75 -5.25 +4.85
7 +0.360 -0.100 +0.740 -1.00 -3.00 +2.20
8 + 1.000
262 ELEMENTARY THEORY OF STRUCTURES
01 23456789 10

30' 40' 30'

Rar 4- X,
R3 Ri
Constant I I
(a)

1.00

(6) Influence diagram for Rq

1.00

(c) Influence diagram for i?3

+ 5.98

(d) Influence diagram for Mi

-2.59 -3.33
(e) Influence diagram for M3
+ 6.67

(f) Influence diagram for M^

Fig. 11-34

Likewise, ^7533 is equal to the bending moment at point 3.

EI^z^ = (406.25) (30) - 3^(18.75) (30) (10) = 9,375 kip-cu ft

9,750
Thus R. = ^^ = ^4^ = 1.040 kips upward
9,375

By statics (Fig. ll-33c)

/2o = 0.150 kip downward

Rs = 0.110 kip upward

From the free-body diagram shown as Fig. ll-33c,

Ml = -lORo = -1.50kip-ft
M, = -SORo = -4.50kip-ft
M, = +30/^8 = +3.30 kip-ft
Other values in the influence table are similarly computed. The
required influence diagrams are plotted in Fig. 11-32.
 ANALYSIS OF STATICALLY INDETERMINATE BEAMS 263

Example 11-17. Construct the influence diagrams for Rq, Rz, ikfi, il/3,

and M5 in the continuous beam shown in Fig. ll-34a.

SOLUTION
Table 11-2. Influence Table

Load at 7^0 Rz Ml Ma M5

+ 1.000
1 +0.598 +0.469 +5.98 -2.08 -0.74
2 +0.247 +0.836 +2.47 -2.59 -0.93
3 + 1.000
4 -0.108 +0.896 -1.08 -3.25 +2.50
5 -0.111 +0.611 -1.11 -3.33 +6.67
6 -0.058 +0.271 -0.58 -1.75 +2.50
7
8 +0.025 -0.108 +0.25 +0.74 -0.93
9 +0.020 -0.086 +0.20 +0.59 -0.74
10

The values in Table 11-2 are computed, and the required influence
diagrams are drawn as shown in Fig. 11-34.

Ik

012 ,,3 4 8 10

523
573 ^^-"27.000
533 §43 553 563
-''52,200
65,800'

««'2°%^9;000-99W^°'°°^''''''
(a)

455

3 I 7 10

^ 4

t \
8k 0.896 k 0.271k 0.058 k
(c)

Fig. 11-35

For purpose of illustration, computations for values of i?o, Rz, Mi, Mz,
and Ms in the continuous beam due to a unit load at point 4 will be
shown. In this analysis, Rz and /?? are chosen as the redundants. With
264 ELEMENTARY THEORY OF STRUCTURES
a unit load at point 4, the two conditions for consistent deformation are

^3533 + -^-537 = 5u

All 5 quantities may be taken from the elastic curve of Fig. ll-35a.
They have been computed from the conjugate beam shown in Fig. 11-356.
Thus
^7577 = Eld,^ = M'^ = (595) (30) - H(21)(30)(10)
= 88,200 kip-cu ft

E[bz-j = Elb,^ = M', = (455)(30) - M(9)(30)(10)

= 73,800 kip-cu ft

^7534 = 7:7543 = M\ = (455) (60) - H(18)(60)(20)

= 99,000 kip-cu ft

Elh, = EUn = EI8,z = M', = (455) (40) - M(12)(40)(4%)

= 90,000 kip-cu ft

Substituting and solving,

Rz = 0.8958 kip upward

Ri = 0.2708 kip upward
By statics (Fig. ll-35c),

Ro = 0.1083 kip downward

RiQ = 0.0583 kip downward
From Fig. ll-35c,

Ml = -lORo = -1.083 kip-ft

M, = -SORo = -3.249 kip-ft
M, = -50/^10 + 20R7 = +2.501 kip-ft

The influence diagrams as shown in Fig. 11-34 will clearly indicate

the spans which should be loaded with uniform live load for maximum
effect. For instance, for maximum negative moment at point 3, the
two adjacent spans should be loaded but the third span should remain
unloaded.

PROBLEMS
11-1. By the moment-area method calculate the slopes of the elastic curve at
A, B, and C and the vertical deflection at C of the simple beam.
11-2. By the moment-area method find the slope and deflection at the free end B
of the cantilever beam AB.
11-3. By the moment-area method find the slope and deflection at the free end A
of the cantilever beam AB.
11-4. By the moment-area method find the slopes at ends A and B and the vertical
deflection at the center of the simple beam.
 H

ANALYSIS OF STATICALLY INDETERMINATE BEAMS 265

12 k

1^ I n P

E= 30,000 k/m.2
20'

-
—
/=250in.4
^^ ,

Constant /
L
—JB
J

Fig. Pll-1 Fig. pi 1-2

I I I I L 1 i i i i
£
L/l_ L/2 ^L/4 . L/2 L/4'^
'
'^ L ,

Constant / Constant I

Fig. Pll-3 Fig. pi 1-4

11-5. By the moment-area method find the slopes at ends A and B and the vertical
deflection at the center of the simple beam.

12 k
10'

W i i t I i WW t p
-*L 12! ^0.6 k/ft
nn.
2/e Ic
20'
L/4 L/2 L/4

E = 30,000 k/in.2 /= 200 in.4

Fig. pi 1-5 Fig. pi 1-6

11-6. By the moment-area method calculate the slope and deflection at the free
eno^ of the overhanging beam ABC.
11-7. Solve Prob. 11-1 by the conjugate-beam method. Determine also the loca-
tion, and amount of the maximum deflection.
11-8. Solve Prob. 11-4 by the conjugate-beam method.
11-9. Solve Prob. 11-5 by the conjugate-beam method.
11-10. Solve Prob. 11-6 by the moment-area and/or conjugate-beam method.

0.6 k/ft

20'

£=30,000 k/in.2 /= 200 in."*

Fig. Pll-11

11-1/i. By the moment-area and/or conjugate-beam method calculate the slope and
defection at the free end C of the overhanging beam ABC.
11-12. Solve Prob. 11-1 by the unit-load method.
11-13. Solve Prob. 11-2 by the unit-load method.
11-14. Solve Prob. 11-3 by the unit-load method.
11-15. Solve Prob. 11-4 by the unit-load method.
11-16. Solve Prob. 11-5 by the unit-load method.
11-17. Solve Prob. 11-6 by the unit-load method.
11-18. Solve Prob. 11-11 by the unit-load method.
11-19 to 11-24. Analyze the statically indeterminate beams shown by the method
of consistent deformation. Draw the shear and bending-moment diagrams and sketch
the elastic curve.
266 ELEMENTARY THEORY OF STRUCTURES

i 1

.
I i

Constant /
i i i i i
?B
I. —
Constant /
, j,
''
^

Fig. pi 1-19 Fig. pi 1-20

1
'
I
M
I f t

'
*******
L
V

cB

Constant / Constant 7

Fig. Pll-21 Fig. pi 1-22

k \ k k \

7^777 fTTyf hi
rC
18'
*' - I
I

Constant I Constant I

Fig. Pll-23 Fig. Pll-24:

012345678
A '

40'
'
^
"^ '

40'
^

Constant 7
Fig. pi 1-25

11-25. Construct the influence diagrams for 72o, ^4, M2, and M4 in the continuous

beam shown.

^
01 23 '
' '
4
^
567 89
^
'
10 11 12

40' 40' 40'

'X-
Constant 7

Fig. pi 1-26

11-26. Construct the influence diagrams for 7?o, ^4, M2, M4, and Me in the con-

tinuous beam shown.

 CHAPTER 12

ANALYSIS OF STATICALLY INDETERMINATE

RIGID FRAMES

12-1. StaticallyDeterminate vs. Statically Indeterminate Rigid

Frames. A frame has been defined as a structure in which the
rigid
members are joined together by rigid connections, such as welded steel
joints or monolithic connections in reinforced concrete. In contradis-
tinction to a hinged joint, members are not free to rotate around a
rigid joint when subjected to geometric deformation, or more specifically,
the angle between the tangents to the elastic curves of any two adjacent
members must remain constant. When deformations of rigid frames are
evaluated quantitatively, this requirement is of paramount importance.

Pi

B 1

P2

"^^^-^ Ha
Vb M^G"

Ca) (6)

Fig. 12-1. Statically Determinate Rigid Frames.

Rigid frames may be statically determinate or statically indeterminate;

in the case of single-story rigid frames, the degree of indeterminacy is

equal to the number of unknown reaction components in excess of the

three unknow^ns which are dependent on the three conditions of equi-
librium for a general coplanar-force system. For instance, the two rigid
frames shown in Fig. 12-1 are statically determinate because in either
frame there are only three unknown reaction components. Ha, Va, and
Vd in Fig. 12-la and Ma, Ha, and Va in Fig. 12-16. However, the rigid
frames shown in Fig. 12-2 are statically indeterminate, the degree of
267
268 ELEMENTARY THEORY OF STRUCTURES
indeterminacy being equal to the number of unknown reaction com-
ponents minus 3. Thus the rigid frame of Fig. 12-2a is statically inde-
terminate to the first degree; of Fig. 12-26, to the second degree; of
Fig. 12-2c, to the third degree; and of Fig. l2-2d, to the fifth degree.
As beams, the basic method of
in the case of statically indeterminate
analyzing statically indeterminate rigid frames is the method of consistent
deformation. Other more convenient methods, such as the slope-deflec-
tion or the moment-distribution methods, may be used to analyze both
statically indeterminate beams and rigid frames. The two last-men-
tioned methods will be treated in Chaps. 14 and 15. In the method of
consistent deformation, a basic determinate structure is first derived
from the original indeterminate structure by removing the redundant
reaction components and treating them as unknown forces or moments
acting on the basic structure. Then these unknown redundant forces

C B

Hr H,
H^

J Vd Mp Vf

(o) (b) (c) (d)

Fig. 12-2. Statically Indeterminate Rigid Frames.

or moments are determined from an equal number of geometrical condi-

tions relating to the deformed structure. These usually require zero
deflection or rotation at the location of each redundant. Numerical
applications of this method will be shown in Art. 12-4.
In order that statically indeterminate rigid frames may be analyzed
by the method of consistent deformation, it is first necessary to discuss
the methods of finding deflections and rotations in a deformed statically
determinate rigid frame. The two methods to be treated here are the
unit-load method and the moment-area (including the conjugate-beam)
method.
12-2. Deflections of Statically Determinate Rigid Frames : the
Moment-area Method. In the preceding chapter the moment-area
(including conjugate-beam) method was used to determine the deflec-
tions and slopes at various points in statically determinate beams. The
moment-area or conjugate-beam theorems may be applied to each straight
member in a rigid frame by treating the member as a beam or a member
subjected to bending only. Actually the members in a rigid frame are
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 269

subjected to combined bending and direct stress and must be treated as

such in the design of the cross sections. In general the axial deformation
in the member due to the direct stress is almost always so small in com-
parison with the transverse deflection due to bending that direct stress
deformation may be neglected when considering the geometric distortion
of the whole rigid frame. In accordance with this fundamental assump-
tion, the lengths of allmembers in a rigid frame are assumed to remain
unchanged during Thus the moment-area method may be
distortion.
used to determine the transverse deflections of any member in a rigid
frame.
Since any application of the moment-area or the conjugate-beam
theorems involves the computation for some geometric property of the
elastic curve, it is most desirable to indicate the result thus obtained
on a sketch of the elastic curve. To determine deflections or rotations
in statically determinate rigid frames, the following steps are necessary:
1. Compute reaction components. Draw a free body and the bending-
moment diagram for each member.
2. Draw the Mdiagram on the rigid frame. A modified AI diagram
is required if the members have different moments of inertia.

3. By visualization determine the most probable new locations of the

joints after deformation. Sketch the elastic curve.

4. Apply the moment-area or conjugate-beam theorems to evaluate

any desired quantity on the elastic curve.

The above procedure will be used in the solutions of the following
examples.
Example 12-1. By the moment-area method determine the rotation,
horizontal deflection, and vertical deflection of each joint in the rigid
frame shown in Fig. 12-3a.
SOLUTION. The reaction components Ha, Va, and Vd are first com-
puted by applying the equations of statics to Fig. 12-3a.

2Ma = 0: (10) (12) + (72) (12) = 247z>

Fz) = 41 kips
SF, = 0: Ha = 10 kips
^Fy = 0: 7^ = 72 - 41 = 31 kips

CHECK. By 1:Md = 0,

(31) (24) + (10) (12) = (72) (12)

864 = 864

The free-body, shear, and bending-moment diagrams of all members

areshown in Fig. 12-Sbcd. When all free bodies are in equilibrium, and
270 ELEMENTARY THEORY OF STRUCTURES
all shear and bending-moment diagrams close, the correctness of the
analysis indicated in Fig. l2-3bcd is certain.
The bending-moment diagram for each member in Fig. 12-Sbcd is

next drawn on a sketch of the rigid frame, as shown in Fig. 12-3e. If

members vary in cross section, the modified diagram of M Fig. 12-3/
may be obtained by dividing the ordinates in the M diagram of Fig. 12-3e
31 41
3k/ft 120
3k/ft
n t * \ i i * *
C -Q-f-T-T
«D
4/c
10k — _^ 3/c ca
^
10
(+1601/6) 41
3/c _^^J
O] p^^ (-2801/6)
W
l=10k,
^ A
24'
A
31
120
2801/6 -41
41
(d)

^ = 31k Vx,-41k (6) (c)

(o)

®V_ii
2801/6

1) 93Vi8
^^TTTTTmTTT]TmTTTr>>^^

(e) M-diagram (/) Mod. Af-diagram

20,736
17,616 ^^^^^ . 17,616
1bi cI

(ff) Elastic curve

Fig. 12-3

by the number of /c's in the respective members. The properties of the

modified M
diagram on member AB may be easily found as it is, or it
may be decomposed into triangles 1 and 2 -|- 3 as shown in Fig. 12-3/.
By referring to Fig. 12-36 it is seen that actually member AB may be
considered to be a simple span subjected first to a 10-kip load without
end moments and then to an end moment of 120 kip-ft. Triangle 1 in
Fig. 12-3/ is the modified M
diagram due to the upper end moment only,
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 271

while the triangle 2 + 3 is that for the 10-kip load only. Similarly
triangle 4 is the modified M diagram on member BC due to the left end
moment only and parabola 5 is that due to the uniform load on a simple
beam BC. Obviously it will be more convenient to use the properties
of triangle 4 and parabola 5 than to use those of the combined modified
M diagram as shown on member BC of Fig. 12-3/.
A sketch of the deformed structure is shown in Fig. 12-3^. It is
noted that A and A' must coincide because of the hinge at A. D moves
out horizontally to D'. Now, since the lengths of members AB and
CD assumed not to change (even though there are direct stresses in
are
them), the deflected points B' and C
must remain on the same elevation
as B and C. ^Moreover, BB' and CC must be equal in order that the
length of member BC remains constant. The elastic curve B'C must be
concave on the top because the bending moment on member BC causes
compression throughout the entire length. Similarh', elastic curve A'B'
is concave to the left and CD' is straight. Because the joints at B and C
are assumed to be rigid, the tangents at these joints both rotate so that
the angles at B' and C
are still right angles.
Upon completion of the above-described preliminaries, the actual com-
putations for the rotations and deflections now become relatively simple.
Applying the conjugate-beam method to member BC,

EIcBb = M(area 5) + %(area 4)

= >^(%)(54)(24) -f %(>^)(30)(24) = 672 kip-sq ft

6b = 672 —^^p^— clockwise

EIcBc = K(area 5) -f M(area 4)
= H(864) + M(360) = 552 kip-sq ft

dc = 552 — — :p~ counterclockwise

Applying the moment-area theorems to member AB,

EIcBa = EISb + (area 1 + area 2 + area 3)

= 672 -h 360 H- 40 -f- 80 = 1,152 kip-sq ft

Oa = 1,152^^ clockwise
BB' = BBi - B,B'

= 18^.4 — (moment of areas and 3 about B)

YY 1, 2,

= ^^[(18)(1,152) - (360)(6) - (40)(4) - (80)(10)]

kip-cu ft
= 17,616--^^—
272 ELEMENTARY THEORY OF STRUCTURES
or BB' = AA. = AAi -f AiA2
= yj^ (moment of areas 1, 2, and 3 about A) -\- ISOb

= ^^ [(360)(12) + (80)(8) H- (40)(14) + 18(672)]

kip-cu ft
= 17,616
EIc
kip-cu ft
CC = BB' = 17,616
EIc
Referring to CD',

Qd = Be = 552 —^-4^^ — counterclockwise

DD' = DDi + DiD' = CC + Udc

1 kip-cu ft
[17,616 12(552)] = 24,240
EIc
-f-
EL
24' '^' 1/4
18
*0^ 1 B C 1
c
4/c
18 oi
1/4 1/4

37,

Ik
t 1/4
HA=lk
1/4
VA=V4k
(6) Free-body diagrams
(a)

1836

(c) A/ diagram (d) Mod. JV/-diagram

Fig. 12-4

Example 12-2. By the moment-area method determine the rotation,

horizontal deflection, and vertical deflection of each joint in the rigid
frame shown in Fig. 12-4a.
SOLUTION. The reaction components due to the 1-kip horizontal load
at D are found by applying the equations of statics to Fig. 12-4a. Thus

Ha = 1 kip to the left

Va = yi kip downward
and Vd = /4: kip upward
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 273

The free-body diagrams of the individual members of the frame are

shown in Fig. 12-46. The M and the modified M diagrams are shown in
Fig. 12-4c and d, respectively. The elastic curve A'B'C'D' is sketched
and shown in Fig. 12-4e.
Applying the conjugate-beam method to member BCj

EIcBb = %(area 2) + >i(area 3)

= %(54) + M(36) = 48 kip-sq ft clockwise
Elcdc = >^(area 2) + %(area 3)
= M(54) + M(36) = 42 kip-sq ft counterclockwise

Applying the moment-area theorems to member AB,

EIcdA = EIcdB + (area 1)

= 48 + 54 = 102 kip-sq ft clockwise
BB' = BBi - B^B'
= 18^^ — ^PTT (moment of area 1 about B)
hilc

EIc(BB') = 18(102) - (54) (6) = 1,512 kip-cu ft

or BB' = AA2 = AAi + A1A2

= y^^ (moment of area 1 about A) + ISOb

EIc{BB') = (54) (12) + 18(48) = 1,512 kip-cu ft

Applying the moment area theorems to member CD,

EIcBd = EIoBc + (area 4)

= 42 + 24 = 66 kip-sq ft counterclockwise
DD' = DD, + D1D2 + D2D'
= CC + + tjlc
wr (nioment of area 4 about D)
12dc

EIc{DD') = 1,512 + 12(42) + (24) (8) = 2,208 kip-cu ft

or DD' = CC2 = CC' + C'C2 = CC {C^C2 -C^C) -f-

= CC -h 12^£> — ^^rr (moment of area 4 about C)

EIc(DD') = 1,512 + 12(66) - (24) (4) = 2,208 kip-cu ft

Example 12-3. By the moment-area method determine the rotation,

horizontal deflection, and vertical deflection of each joint in the rigid
frame shown in Fig. 12-5a.
SOLUTION. By applying the three equations of statics to the free
body shown in Fig. 12-5a, the reaction components are found to be
Ha = 10 kips to the left, Va = 72 kips upward, and Ma = 984 kip-ft
counterclockwise. The M
and the modified diagrams are shown inM
Fig. 12-56 and c, respectively. By starting at the fixed end at A, the
274 ELEMENTARY THEORY OF STRUCTURES
curve A'B'C'D' may be sketched as shown in Fig. 12-5c?. The
elastic
moment-area theorems are then appUed successively to the individual
members of the frame.

= area 2 + area 3 = (288) (18) -j- >^(40)(12)

EIcBb
= 5,184 + 240 = 5,424 kip-sq ft clockwise
EIc{BB') = moment of areas 2 and 3 about B
= (5. 184) (9) + (240) (14) = 50,016 kip-cu ft
EIcBc = EIcBb + (area 1) = 5,424 + 1,728
= 7,152 kip-sq ft clockwise

CiC = CiCo -h CiC = 2A:dB -h Jt (moment of area 1 about C)

10 k

//.= 10k

M. = 9S4 k-ft (6) JVf-diagram

Vi = 72 k
(a)

50,016
(2)
B
216

®
AA'
n
(c) Mod. A/-diagram
35.808-

{d) Elastic curve

Fig. 12-5

EliCiC) = (24)(5,424) -f- (1,728)(18) = 161,280 kip-cu ft

kip-cu ft
DDi = CiC = 161,280
EIc

e^ = dc = 7,152 ^^Jfy^ — clockwise

D2D' = DiD' - D,D2 = I2dc - CCi

EIciDiD') = 12(7,152) - 50,016 = 35,808 kip-cu ft

Example 12-4. By the moment-area method determine the rotation,

horizontal deflection, and vertical deflection of each joint in the rigid
frame shown in Fig. 12-6a.
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 275

SOLUTION. The reaction components at A due to the 1 kip-ft moment

atD are found by the equations of statics to be FIa = 0, Va = 0, and
Ma = 1 kip-ft clockwise (Fig. 12-Ga). The diagram, the modified M
M diagram, and a sketch of the elastic curve are shown in Fig. 12-Qbcd.

EIcBb = area 1=6 kip-sq ft counterclockwise

EIc(BB') = moment of area 1 about B = (6) (9) = 54 kip-cu ft
Elcdc = EIcOb + area 2 = 6 + 6 = 12 kip-sq ft counterclockwise
CCi = BB'

CiC = C1C2 + C2C' = 24:03 + 7^ (moment of area 2 about C)

B C

Ha'O A
Afa-lk-ftC "
(b) M-diagram

^-0
(a)

1/4
1/4
f®
:> B C ^^-16

(D^

Z>L

5 A
(c) Mod. M-diagram {d) Elastic curve

Fig. 12-6

ElciCiC) = (24) (6) + (6) (12) = 216 kip-cu ft

EIcOd = Elcdc + area 3 = 12 + 4 = 16 kip-sq ft counterclockwise

D2D = CiC
D2D' = D,D, + DzD' - D1D2
1 54
= I2dc + ^PTT (moment of area 3 about D) — j^^
J^Ic tile
EhiD^D') = (12)(12) + (4)(6) - 54 = 114 kip-cu ft

Example 12-5. By the moment-area method determine the rotation,

horizontal deflection, and vertical deflection of each joint in the rigid
frame shown in Fig. 12-7a.
276 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. The reaction components at A due to the 1-kip horizontal
load at D are found by statics to he Ha = I kip to the left, Va = 0, and
Ma = 6 kip-ft counterclockwise (Fig. 12-7 a). The M diagram, the
modified M diagram, and the sketched elastic curve are shown in Fig.
12-7 bed. Note that B and B' happen to coincide in this particular
problem.

EIcOb = (area 2) - (area 1)

= 24 — 6 = 18 kip-sq ft counterclockwise
EIc{BB') = (moment of area 1 about B) — (moment of area 2 about B)
= (6) (16) - (24) (4) =
Elcdc = ELBb + (area 3)
= 18 + 72 = 90 kip-sq ft counterclockwise
CC = CCi + CiC = 2-idB + -Jt-
rjic
(moment of area 3 about C)

24'

47c

37,
•Ik

L,
'

ilfA-6k-ftCr (6) A7-diagram

(a)

w B-F

A-A'
N
(c) Mod. M-diagram (d) Elastic curve

Fig. 12-7

EIc{CC') = (24)(18) + (72)(12) = 1,296 kip-cu ft

EIcOd = EIcBc + (area 4) = 90 24 +
= 114 kip-sq ft counterclockwise

DiD' = Z)iZ)2 + D2D' = I2dc + -Ji-

tic
(moment of area 4 about D)

EIc{D,D') = (12)(90) + (24)(8) = 1,272 kip-cu ft

 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 277

Example 12-6. By the moment-area method determine the rotation,

horizontal deflection, and vertical deflection of each joint in the rigid
frame shown in Fig. 12-8a.

24'

4/c

3/c

Ik
M^-24k-ftC'w (6) Af-diagram
VA-lk
I (a)

$0= 216
"V
®

(c) Mod. M-diagram (d) Elastic curve

Fig. 12-8

SOLUTION. The reaction components at A due to the 1-kip vertical

load at D are found by statics to be Ha =
0, Va = I kip downward, and

Ma = 24 kip-ft clockwise (Fig. 12-8a). The diagram, the modified M

M diagram, and the sketched elastic curve are shown in Fig. 12-Sbcd,

EIcOb = area 1 = 144 kip-sq ft counterclockwise

EIcBB' = moment of area 1 about B
= (144) (9) = 1,296 kip-cu ft

EIcBc = EIcBb + (area 2) = 144 + 72 = 216 kip-sq ft

CiC = C1C2 4" C 2C

= 24^5 -f -^rr (moment of area 2 about C)

tjlc
EIc(CiC') = (24) (144) + (72) (16) = 4,608 kip-cu ft

Bd = Gc
D2D' = DiD' - D1D2 = I2dc - D1D2
EIciD^D') = (12) (216) - 1,296 = 1,296 kip-cu ft

12-3. Deflections of Statically Determinate Rigid Frames : the Unit-

load Method. The derivation and application of the unit-load method to
finding deflections and rotations in statically determinate beams were
278 ELEMENTARY THEORY OF STRUCTURES
described in Art. 11-5. The formulas previously derived are;

(l)(Ac) = Mm dx (ll-14a)
EI
and

(mec)
Mm dx (ll-14b)
-j: EI
Equations (ll-14a) and (ll-14b) may also be used to find the deflection
or rotation at any point in a statically determinate rigid frame provided
that only the effect of bending stress on the distortion of the frame is

considered. The expression on the right side of Eqs. (ll-14a) and

(ll-14b) must include the summation or integration through all the
members of the rigid frame. Since the product of M and m is involved
in these summations, it is important to use the same sign convention
for both M and m. Usually a bending moment which causes compression
in the outer fibers of the section is considered to be positive.

lOk-

H^=10k^*-«--^A
1/4

l^-31k V£,= 41k 1/4

(a) For values of M (b) For values of m

Fig. 12-9

The following examples will illustrate the application of the unit-load

method to the finding of deflections and slopes in statically determinate
rigid frames.
Example 12-7. By the unit-load method determine the horizontal
deflection at the roller support of the rigid frame shown in Fig. 12-9a.

SOLUTION

Segment of frame AE EB BC CD

Origin A E C D
Limits . . to 12 to 6 to 24 Oto 12
M \0x 120 41x - ^x2
m X X + 12 12 + Hx X
/ Sic 3/. 4/. Sic
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 279
r\2 re
EIcAh oi D = H / {10x){x)dx + Vs / (120) (x + I2)dx
/•24

+ (41x - %a:2)(12 + }4x)dx

~
12 16

=[ 9
+ 20x2 ^ 430^
_

123x2 3x' 41 3 _
+ 2 2 '48 + 128
L
1,920 + 3,600 + 18,720 = 24,240 kip-cu ft

The fact that a positive result is obtained indicates that the direction
of the horizontal deflection at D agrees with that of the unit load applied
at D as shown in Fig. 12-96 (toward the right in this case). Thus

Anoi D = 24,240 ^'^l^ ^^

to the right

Example 12-8. By the unit-load method determine the horizontal

deflection at the roller support of the rigid frame shown in Fig. 12-lOa.

24'

4/c

3/c

37.
Ik

Ik t t
1/4 k 1/4
{
1/4 k V4
(a) For values of M (b) For values of m
Fig. 12-10

SOLUTION

Segment of frame AB BC CD

Origin A B D
Limits Oto 18 to 24 to 12
M X 18 - Hx X
m X 18 - Hx X
/ sic 47c 3/c

EIcAh oi D = yi 1^'^ x'dx-\-}i

= [y9xV + [81x %x2 + yig2xV + -
P (18 - }4xydx -h M /o'^ x' dx
[}'^a:3]o^2

= 648 -f 1,368 + 192 = 2,208 kip-cu ft

280 ELEMENTARY THEORY OF STRUCTURES

Thus Anof D = 2,208 ^}2^^ to the right

Example 12-9. By the unit-load method determine the rotation,

horizontal deflection, and vertical deflection at the free end of the rigid
frame shown in Fig. 12-1 la.

lOk — ^

c
H^-lOk
Wyl-984k-ftG|,
V^-72k
(a) ForM (b) For m for $[)

Bi ,C

D ^1

D 24
G
(c) For m for ^ at D (d) For m for Ay at D
Fig. 12-11

SOLUTION

Segment of frame AE EB BC CD

Origin A E C D
Limits ... . Oto 12 to 6 Oto 24 to 12
M -984 + lOx -804 -^x^
7n for Od -1 -1 -1 -1
m for A// at /) -6 +x -6 + (x + 12) 4-12 +x
m for Av Sit D -24 -24 —X
/ 3/c 3/c 4/c 3/c

EUd = Vs
P (-984 + 10x){-l)dx + Vs j^ {-SQ>^){-l)dx
-hy4P{-Hx')(-l)dx
= [328x - Hx'W + [288x]o« + [Vhx'V
= 3,696 -i- 1,728 + 1,728 = 7,152 kip-sq ft

kip-sq ft
Od = 7,152 clockwise
EIc
 .

ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 281

EIc^H Sit D = H f^'^ (-984 -h 10x)(-6 + x)dx

+ Vs
f'
(-864)(x + 6)dx + H //' {-y2x'){-^l2)d:
10x3
1.968.r - 174x2 + 144x2 - l,728x]o^

+ l-r2X%h
= 480 - 15,552 - 20,736 = -35,808
kip-cu ft
Ah Sit D = 35,808
EL to the left

EIcAv at D = }4 r' (-984 + 10x)(-24)f/x + H /^^ (-864)(-24)c?x

-\-Hl''i-y2x'){-x)dx
= [7,872x - + [6,912x]o« + iy32xV-'
40x2]oi2
= 88,704 + 41,472 + 31,104 = 161,280
Av Sit D = 161,280 ^^^'^^ ^^ downward
Example 12-10. By the unit-load method determine the rotation,
horizontal deflection, and vertical deflection at the free end of the rigid
frame shown in Fig. 12- 12a.

SOLUTION

Segment of frame AB BC CD

Origin . . . A C D
Limits. . to 18 0to24 Oto 12
M +1 +1 +1
m for do. -hi + 1 -hi
m for A^ at D -6 -hx -hl2 +x
m for Ar at D +24 -f-x

/ 3/c 47c 3/c

= 16 kip-sq ft counterclockwise

EIJiH s.iD = yz
P (+l)(-6 + x)dx + Yi
P (+l)(+12)rfi

+ Ys
f^"
{+i){+x)dx
= 114 kip-cu ft to the right

EIcAy at D = 1^
P (+l)(+24)rfx + Yi //' ( + l)i+x)dx
+ M/o"( + l)(0)rfx
= 216 kipHcu ft upward
282 ELEMENTARY THEORY OF STRUCTURES
24'

4/<:

3/c

Ik-ft
"Ji

Ik-ft C iC
(a) ForAf (6) For m for 9/)

Br S. ^C

D -*-!

,G 24 Ci
(c) For m for A// at D (d) For m for Ay at D
Fig. 12-12

Example 12-11. By the unit-load method determine the rotation,

horizontal deflection, and vertical deflection at the free end of the rigid
frame shown in Fig. 12-13a.

SOLUTIOX

Segment of frame AB BC CD

Origin A C D
Limits Oto 18 Oto 24 to 12
M -6 +a: + 12 +J-
m for do +1 +1 + 1

m for Afr at Z> -6 +x + 12 +T

m for Ar at D +24 +x
/ 3/. 4/. 3/c

ELBo = Yi
P (-6 + x)( + l)rfz + MP i + 12){ + l)dx

= 114 kip-sq ft counterclockwise

EIJ,„ at D = M P (-6 + xy-dx + M P i+ny-dx

+H /„" (+xydx
= 1,272 kip-cu ft to the right
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 283
/* 18 r 24
£/.Ak at Z) = M /^ i-6 + x)(+2-i)dx + }i j^ i+l2)(+x)dx

+H /„" i+x){0)dx
= 1,296 kip-cu ft upward

24'

4/c

37c

3/c

C
6k-ft G C
(a) For M (6) For m for ^£,

Bf iC B| iC

,c 24C ,,

(c) For m for A// at D (d) For m for Ay at D

Fig. 12-13

Example 12-12. By the unit-load method determine the rotation,

horizontal deflection, and vertical deflection at the free end of the rigid
frame shown in Fig. 12-14a.

SOLUTION

Segment of frame AB BC CD

Origin A C D
Limits to 18 Oto 24 to 12
M +24 +x
m for Od +1 +1 +1
m for Afl^ at D -6 +2: + 12 +0:
m for Af at D +24 +x
/ 37. 37e

Ehdn = Yz
P (+24)(+l)dx +H P {+x)( + l)dx

+ yzl (0)( + l)dx

= 216 kip-sq ft counterclockwise
284 ELEMENTARY THEORY OF STRUCTURES

EI Ah at Z) = y^j^\ (+24)(-6 + x)dx + >i

J^''
(+x)( + i2)rfa;

+ M/,''(0)(+x)rfa:
= 1,296 kip-cu ft to the right

EIc^v at Z) = M /,'' {-^2^ydx + 3^ J^'" {-^xYdx + M /,'' (O)^cia:

= 4,608 kip-cu ft upward

24'

4/.

3/c
3/.

c
Ik 0-

24k-ftC ,C
Ik
(a) ForM (6) For m for ^^j

Bt ,C

=C 24^

(c) For m for A« at D (d) For m for Ay at D

Fig. 12-14

12-4. Analysis of Statically Indeterminate Rigid Frames by the Method

of Consistent Deformation. The analysis of statically indeterminate
rigid frames by the method of consistent deformation is quite similar to
that of statically indeterminate beams. indeterminacy The degree of
of the rigidframe is first observed. A basic determinate rigid frame is
then derived from the given indeterminate rigid frame by removing the
restraints (equal in number to the degree of indeterminacy) and treating
the redundant reaction components as loads acting on the basic structure.
The unknown redundant foroes are determined from an equal number of
conditions of consistent deformation, which usually require zero rotation
or zero deflection at the location of each redundant. Once the redundants
are known, the remaining reaction components may be determined by the
laws of statics.
Example 12-13. Analyze the rigid frame shown in Fig. 12- 15a by the
method of consistent deformation. Draw the free-body, shear, and
bending-moment diagrams for all members. Sketch the elastic curve
of the deformed structure.

i
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 285

SOLUTION. A single-span rigid frame with two hinged supports is

statically indeterminate to the first degree because there are four unknown
reaction components Ha, Va, Hd, and Vd as shown in Fig. 12- 15a, but
only three equations of statics. In this case the basic determinate struc-
ture may be obtained by replacing the hinged support at /) by a roller
support. Then the original structure in Fig. 12- 15a is the equivalent
of the two structures shown in Fig. 12-156 and c. In the present instance,
the deflection Ai? to the right and caused by the loading on the statically

3k/ft 3k/ft
I I t I i C^
C
10 k
10 k-

^+
(b) (c)

Fig. 12-15

determinate structure must be wholly counteracted by the redundant

reaction Hd w^hich must act to the left. To satisfy the condition of zero
horizontal deflection at D, Hd must be equal to Ad/^d- From the result
of either Example 12-1 or Example 12-7,

EIcAd = 24,240 kip-cu ft to the right

and from either Example 12-2 or Example 12-8,

EIJd = 2,208 kip-cu ft to the left

Therefore
EIcAd 24,240
Hd = 10.98 kips to the left
EIcdD 2,208

By statics (Fig. 12-16a),

Ha = 0.98 kip to the right

Va = 33.74 kips upward
Vd = 38.26 kips upward

The free-body, shear, and bending-moment diagrams for the three

members are shown in Fig. 12-I6bcd. The modified diagram and its M
parts are shown in Fig. 12-16e. The elastic curve is shown in Fig. 12-16/.
Numerical properties of the elastic curve may be computed as shown
below.
 286 ELEMENTARY THEORY OF STRUCTURES
=
EIcOb >^(area 4) - %(area 5) - M(area 6)
= 3^^(864) - %(233) - M(396) = 145 kip-sq ft clockwise
EIcSc = >2(area 4) - >^(area 5) - %(area G)
= 1^^(864) - M(233) - %(396)
= 90 kip-sq ft counterclockwise
EIcdA = 145 - (area 3) + (area 1) + (area 2)
= 145 - 233 + 40 + 80 = 32 kip-sq ft clockwise
Elcdn = (area 7) - 90 = 263 - 90 = 173 kip-sq ft clockwise
EIciBB') = (18) (32) -h (moment of area 3 about B)
— (moment of areas 1 and 2 about B)
= 576 -h 233(6) - (40) (4) - (80) (10)
= 1,014 kip-cu ft to the right
ElciCC) = (12) (173) - (moment of area 7 about C)
= 2,076 - (263) (4) = 1,024 kip-cu ft to the right

3k/ft
i t I I 4 i-r-r

lOk-

Hjr,= 10.98 k

24'

V^= 33.74 k \^ = 38.26 k

(a) ib) (c)

3k/ft
10.98
10.9^

77, 2\^
33.74
c|b:; .s
38.26
^ |1014| 2076
33.74 (+189.8) 7, 57^-438 1024 , 1052
12.75 I'lR C '

'1(7
'

n B
//X^fT=9o-
/^A = 32
^~"'
^
ty
1 On- 173

A-A'

77.6
(e) Mod, AZ-diagram if) Elastic curve

131.8
(d)

Fig. 12-16

The fact that all free-body diagrams (Fig. \2-\Qiabcd) are in equi-
librium shows that all equations of statics are satisfied, and the fact that
BB' and CC are found to be equal shows that the one important condition
of geometry has been satisfied. When all the conditions of statics and
of the geometry of deformation have been satisfied, the correctness of
the solution is ensured.
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 287

Example 12-14. Analyze the rigid frame shown in Fig. 12-17a by the
method of consistent deformation. Draw the free-body, shear, and
bending-moment diagrams of all members. Sketch the elastic curve of
the deformed structure.
SOLUTION. Because there are six unknown reaction components, and
only three equations of statics, a single-span rigid frame with two fixed
supports is statically indeterminate to the third degree. Since any three
of the six reaction components may be taken as the redundants, there
are several different ways of choosing the basic determinate structure.
If Md, Hd, and Vd are chosen the redundants, the basic determinate
structure is a cantilever structure with a fixed support at A and a free
end at D. The original statically indeterminate structure in Fig. 12-17a

3k/ft 3k/ft
^^D^<i>M

10k 10k + D'

Wd^vm

1 D'
M, (5
^nhuM H
(6) (c)

J^D^(i>H

+
Vd^w
D

(d) (c)

Fig. 12-17

then becomes the equivalent of the four determinate structures shown in

Fig. 12-I7bcde.
Let 6, Ah, and Av be the rotation, horizontal deflection, and vertical
deflection at D due to the applied loads on the basic determinate struc-
ture; 8^^f, dnM, and 8vm be those due to a 1 kip-ft counterclockwise
moment at D; d^n, 8hh, and 8vh be those due to a 1-kip horizontal load
acting to the right at D; and 5^^, 8hv, and dw be those due to a 1-kip
vertical load actingupward at D. The three conditions of consistent
deformation requiring zero rotation and zero horizontal and vertical
deflections at D are:

6d + Md8<i,m + Hd84>h + Vd8<j,v =

Ah + Md8hm + Hd^iih + Vd^hv =
A^ -|- Md8vm "h Hodvii -h Vd8vv =
By assigning a positive sign to counterclockwise rotation, horizontal
288 ELEMENTARY THEORY OF STRUCTURES
deflection to the right, and vertical deflection upward, the following
values have been taken from the results of Examples 3 to 6, or Examples 9
to 12:

=
EIcBd -7,152 kip-sq ft EIchM = +16 kip-sq ft
EIc^H = -35,808 kip-cu ft EIchHM = +114 kip-cu ft
EIcAv = -161,280 kip-cu ft EIcdvM = +216 kip-cu ft
EIcd^^H = 4-114 kip-sq ft Elcd^v = +216 kip-sq ft
EIcdHH = +1,272 kip-cu ft EIcdHv = +1,296 kip-cu ft
EIcSvH = +1,296 kip-cu ft Elcdvv = +4,608 kip-cu ft

3k/ft

4/c

lOk-
3/c

//o= 13.78 k
;/a= 3.78 k
O
Afo=55.4kft A
24'
A/^-24.8k-ftCr
\j4 = 35.72 k Vo= 36.28 k
(a)

3 k /ft
13.78
^'^'^ I I i 4 Iil3.78
XC
i i i

.2GfB
103.2 OP
35.72
C]QiiO.O
3G.28
35.72 (+212.6)

Py^ (-219.4)

36.28

(e) Mod. Afdiagram

103.2 110.0 B B'

id)

'Ob
/%
DD"

AA'
if) Elastic curve

Fig. 12-18

In the above tabulation, it should be noted that

8hv = 8vH

According to the law must hold

of reciprocal deflections, these relations
true. be noted that in the moment-area method each
It should also
value in the pair of 6's is independently determined; therefore, the fact

J
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 289

that they must be equal serves as an excellent check. In the unit-load

method, however, the expressions for both 6 values are identical, and
this does not afford an independent check.
Substituting the above deflection quantities in the equations of con-
sistent deformation,

-7,152 + 16Md + IUHd + 216Vd =

-35,808 + IUMd + l,272i/z, + l,296Fz) =
-161,280 + 2mMD + 1,29QHd -f 4,608 Fz) =
Solving,

Md = +55.4 kip-ft or 55.4 kip-ft counterclockwise

Hd = - 13.78 kips or 13.78 kips to the left
Vd = +36.28 kips or 36.28 kips upward

Applying the laws of statics to Fig. 12-18a,

Ha = 3.78 kips to the right

Va = 35.72 kips upward
Ma = 24.8 kip-ft clockwise

The free-body, shear, and bending-moment diagrams of the three

members of the frame are shown in Fig. 12-lSbcd. The modified M
diagram and its separate parts are shown in Fig. 12-18e. The elastic
curve is shown in Fig. 12-18/. Numerical properties of the elastic curve
may be computed as shown below.
Elcds = 3^ (area 5) - %(area 6) - 3^ (area 7)
= H(864) - M(309.6) - M(330.0)
= 116 kip-sq ft clockwise
or EIc6b = (area 4) — (area 3) — (areas 1 and 2)
= 309.6 - 74.7 - 40 - 80
= 115 kip-sq ft clockwise (check)
EIcBc = 3^^ (area 5) - 3^^ (area 6) - %(area 7)
= M(864) - M(309.6) - %(330.0)
= 109 kip-sq ft counterclockwise
or Elcdc = (area 8) — (area 9)
= 220.2 - 111.0 = 109 kip-sq ft counterclockwise (check)
EIc(BB') = (moment of area 4 about jB)
— (moment of areas 1, 2, and 3 about B)
= (309.6) (6) - (40) (4) - (80) (10) - (74.7) (12)
= 1 kip-cu ft to the right
EIc(CC') = (moment of area 9 about C) — (moment of area 8 about C)
= (111.0)(8) - (220.2)(4)
= 7 kip-cu ft to the right (check)
290 ELEMENTARY THEORY OF STRUCTURES
Inasmuch as this rigid frame with two fixed supports is statically
indeterminate to the third degree, three consistent checks must be
expected in the computed results of the elastic curve. These are:
(1) EIcBb = 116 kip-sq ft clockwise as computed for member BC should
be equal to EIcBb = 115 kip-sq ft clockwise as computed for member AB\
(2) Elcdc = 109 kip-sq ft counterclockwise as computed for member BC
should be equal to Elcdc = 109 kip-sq ft counterclockwise as computed
for member CD, and (3) EldBB') = 1 kip-cu ft to the right should be
equal to EIc(CC') = 7 kip-cu ft to the right. last equality seems
This
absurd but it is nevertheless within the permissible limit of the error
expected because, in this instance, either 1 or 7 is the difference in the
last significant number of values with three or four significant figures.

Ik

^^DD

H
(a) (C)

*-lk

(d)

Fig. 12-19

12-5. Influence Diagrams for Statically Indeterminate Rigid Frames.

Statically indeterminate rigid frames may carry moving loads, especially
on the horizontal member when this structure is used in bridges or to
carry crane loads in industrial buildings. In these cases it may be neces-

sary to construct influence diagrams for bending moments or other

functions. With the aid of the influence diagrams, the critical loading
conditions for maximum bending moments or other desired functions
can usually be determined by inspection or by some cut-and-try method.
The discussion in this article will be limited to influence diagrams for a
single-span rigid frame with two hinged supports. It will be seen that

the use of the law of reciprocal deflections greatly simplifies the work
involved in the computations for the ordinates in the influence diagram
for the horizontal reaction. For instance, let it be required to find the
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 291

horizontal reaction H due to a 1-kip vertical load at P in the rigid frame

with two hinged supports shown in Fig. 12-19a. Applying the method of
consistent deformation to Fig. 12-19a6c,

Odd
in which 5dp is the horizontal deflection at D due to a unit vertical load
at P and ^dd is the horizontal deflection at D due to a unit horizontal load
4 .5
r\

24'
o B B' c ^ C
4/c

ZIc /

/
r"
3/c

TTifr H
\ /
I
/

^st"^
/ Ik
U)
\ 1
1

ji/4k
1
Ik \ 1
\
^^-^\a-a'
H ""^
^'^^ ^°*^- ^-diagram and elastic curve
1/4 kI

4.5
3.9375

B' 12 15 18 21

V = 43'
(6) Influence diagram for H (e) Conjugate beam B'C

00
t^ ^N
-^
t^ / (O N. 00

(c)
•T
•^
<N

Influence diagram for bending

^
moment at midpoint of EC 0.11655 k

1
0.34586 k
0.11655 k

0.65414 k|
(/)

Fig. 12-20

at D. But, by the law of reciprocal deflections, 8dp in Fig. 12-196 is

equal to 8pd in Fig. 12-19c?. Substituting,

H (Fig. 12-19a)
^ ^ ^
= 'r g'g- ]l-]lil
12-19d)
8dd (Fig.
Thus the elastic curve B'C in Fig. 12-19d is geometrically similar to the
influence diagram for //. In fact, the influence ordinates for may be H
obtained by dividing the ordinates to the elastic curve B'C by the con-
stant 8dD' This method will be used in the following illustrative example.
292 ELEMENTARY THEORY OF STRUCTURES
Example 12-15. Construct the influence diagrams for the horizontal
reaction and for the bending moment at the mid-point of the horizontal
member in the rigid frame shown in Fig. 12-20a.
SOLUTION'. The required influence diagrams are shown in Fig. 12-206
and c. To illustrate the procedure, the values in the influence table
for a unit load at a point 9 ft from point B will be computed. Applying
the conjugate-beam method to the elastic curve B'C (Fig. 12-20e),

EIc5,D = (48)(9) - M(4.o)(9)(6) - M (3.9375) (9) (3)

= 257.34 kip-cu ft

Referring to Examples 12-2 or 12-8,

EIcBdd (Fig. 12-20<f) = 2,208 kip-cu ft

T.^LE 12-1. Influence Table

Distance kip-cu ft Moment at mid-

'^^
from B, ft EI. Odd point of BC, ft

3 124.0.3125 0.05617 4-0.657

6 209.25 0.09477 -1-1.578
9 257.34375 0.11655 +2.752
12 270 0.12228 -f4.166
15 248.90625 0.11273 4-2.809
18 195.75 0.08865 + 1.670
21 112.21875 0.05082 +0.738
24

Thus
H (due to 1-kip load at 9 ft from B) = ^^ = 0.1 1G6 kip

Referring to Fig. 12-20/,

M (at mid-point of BC) = (0.34586)(12) - (0.11655)(12)

= +2.752 kip-ft

PROBLEMS
12-1 to 12-3. By the moment-area method determine the rotation, horizontal
deflection, and vertical deflection of each joint in the rigid frame shown.
12-4, By the moment-area method determine the rotation, horizontal deflection,
and vertical deflection at the free end of the rigid frame shown due to a l-kipnft
moment acting counterclockwise at the free end.
12-5. By the moment-area method determine the rotation, horizontal deflection,
and vertical deflection at the free end of the rigid frame shown due to a 1-kip hori-
zontal load acting to the right at the free end.
 ANALYSIS OF STATICALLY INDETERMINATE RIGID FRAMES 293

12-6. By the momciit-arca method dctcrinine the rotation, horizontal deflection,

and vertical deflection at the free end of the rigid frame shown due to a 1-kip vertical
load acting upward at the free end.

18 k
15' 21' 36'

4lc 4lc

2/c 2/c
3/r 3/.

Ik W^
A*^

Fig. P12-1. 7 Fig. P12-2, 8

12-7 and 12-8. By the unit-load method determine the horizontal deflection at the
support of the rigid frame shown.
roller

18 k
15' 21' 36'

4/c
Mr

2/.
2/c
3/, 3/c
CO (M

D DTTfrr-
TTfrr

Fig. P12-3, 9 Fig. P12-4 to 6 and Pr2-10 to 12

12-9. By the unit-load method determine the rotation, horizontal deflection, and
vertical deflection at the free end of the rigid frame shown.
12-10 to 12-12. Solve Probs. 12-4 to 12-0 by the unit-load method.

18k
18 k

=4 4/c
21'
15'

4/r
21'

2/c
2/c
^Ic
3/c

'r
Fig. P12-13 Fig. P12-14

12-13 and 12-14. Analyze the rigid frame shown by the method of consistent
deformation. Draw the free-body, shear, and bending-moment diagrams for all
members. Sketch the elastic curve of the deformed structure.
294 ELEMENTARY THEORY OF STRUCTURES

B ^Ic C

3 2Ic

CM
3/c

^M
-,^
Fig. P12-15

12-16. Construct the influence diagrams for the horizontal reaction and for the
bending moment at the mid-point of the horizontal member in the rigid frame shown.
 CHAPTER 13

ANALYSIS OF STATICALLY INDETERMINATE TRUSSES

13-1. Statically Determinate vs. Statically Indeterminate Trusses. A

truss is a structure in which all members are usually considered to be
connected by smooth pins, and subjected to loads appUed only at the
joints. All members in a truss are thus two-force members. A truss is
completely analyzed when the kind and amount of direct stress in each
member are determined.
A truss is statically determinate if it can be completely analyzed by
the laws of statics alone. A truss may be statically indeterminate
because it has either external redundant reactions or internal redundant
members or both, the degree of indeterminacy being equal to the com-
bined number of redundant reactions and members. If a truss is sub-
jected to a general coplanar-force system, the number of external redun-
dants is equal to the total number of external reaction components minus
3. The number of internal redundant members present in a truss is
equal to the total number of members minus (2j — 3) in which j is the
total number of joints. This statement may
be proved by first establishing that there are
two conditions of statics for each joint as a
free body, or a total of 2j conditions to match
2j unknowns, three of which are the three
external reaction components, thus leaving
2j — 3 conditions of statics available for determining the stresses in the
internal members. Another method of arriving at the same conclusion
is to postulate that a truss is internally stable if it consists of a series of
triangles as shown in Fig. 13-1. The first triangle is made up of three
joints and three members; each successive triangle requires two additional
members but only one additional joint. Thus, if m is the number of mem-
bers in the truss and j is the number of joints, (w — 3) = 2(j — 3), or
m = 2j - 3.

The truss shown in Fig. 13-2a is statically determinate because it

just has three unknown external reaction components and

2j - 3 = 2(16) - 3 = 29 members
295
296 ELEMENTARY THEORY OF STRUCTURES
This is also obvious because the truss diagram is a compilation of simple
triangles. The truss shown in Fig. 13-26 is statically indeterminate to
the second degree because it has two external redundant reaction com-
ponents, although it has just the right number of internal members,
m = 2j - 3 = 2(28) - 3 = 53. The truss shown in Fig. 13-2c is stat-
ically indeterminate to the third degree because it has one external and
two internal redundants. The truss of Fig. 13-2c^ is statically inde-
terminate to the third degree because it has three internal redundants.

Fig. 13-2

The analysis of statically indeterminate trusses by the method of

consistent deformation requires first of all a knowledge of the methods

of determining deflections of statically determinate trusses. The unit-
load method and the graphical (WilHot-Mohr diagram) method will be
discussed in the next two articles.
13-2. Deflections of Statically Determinate Trusses: the Unit-load
Method. Incidental to developing the unit-load determining method of

the deflections of statically determinate beams, one formula takes the

form
(l)(Ac) = ^udL (ll-13a)

This formula may

be applied to the calculation of deflections of trusses.
In a truss there is a finite number of members in which u, the stress
due to the unit load, is constant for each member and dL is the change
in the length of the member due to the applied loads. Applying Hooke's
law,
SL
dL = (13-1)
AE
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 297

wherein *S is the stress in the member due to the applied loads, L and A
are, respectively, the length and cross-sectional area of the member, and
E is the modulus of elasticity of the material. Substituting (13-1) in
(ll-13a),
^ SuL
= 2tE (13-2)

Example Determine the horizontal and vertical deflections of

13-1.
shown in Fig. 13-3a by the unit-load method.
joint L2 of the truss
SOLUTION. The stress S in each truss member (kind and amount) is
computed as indicated in Fig. 13-36. The horizontal and vertical unit
loads are separately applied and the resulting u stresses calculated as

Numbers in ( ) are areas in sq in.

£- 30,000 k/in.2 S in kips
(a) (6)

Ui U2 ^l _o.9o U2 _o_9o Us _o_3o U^

for A// at L2 u for Ay at L2

(c) (d)

Fig. 13-3

shown in Fig. 13-3c and d, respectively. The algebraic methods of

joints or sections may be freely used; by showing the horizontal and
vertical components of the stresses in the inclined members, the equi-
librium of the concurrent-force system at each joint may be easily
checked by inspection.
The required computations are shown in Table 13-1. Thus

Ah of L2 + 63 X 10-3 in. or 0.063 in. to the right

and Av of Lo +244.7 X 10-3 in. or 0.2447 in. downward

It is to be noted that a positive result means that the actual direction

of the deflection or movement agrees with that of the assumed unit load.
Example 13-2. Determine the horizontal and vertical deflections of
joint L3 of the truss shown in Fig. 13-4a by the unit-load method.
298 ELEMENTARY THEORY OF STRUCTURES

Table 13-1

A, AT ^^ u for Ah at L2 Av at L2
Mem- u for Av
sq A// at = Sw AL, = 2u AL,
ber in. kips 10-3 in. at L2
in. L2 10-3 in. 10-3 in.

C/1C/2 180 4 -19.5 - 29.25 -0.90 + 26.325

U.Uz 180 4 -19.5 - 29.25 -0.90 + 26.325
UzU, 180 4 -16.5 - 24.75 -0.30 + 7.425
LqLi 180 4 +21 + 31.5 +1 +31.5 -fO.45 + 14.175
L,L, 180 4 +21 + 31.5 +1 +31.5 +0.45 + 14.175
L2LZ 180 4 + 18 + 27 +0.60 + 16.200
L3L4 180 4 + 18 + 27 +0.60 + 16.200
L4L5 180 4 + 16.5 + 24.75 +0.30 + 7.425
LoC/i 300 5 -35 - 70 -0.75 + 52.500
C/1L2 300 5 - 2.5 - 5 +0.75 - 3.750
L2C/3 300 5 + 2.5 + 5 +0.50 + 2.500
UzL, 300 5 - 2.5 - 5 -0.50 + 2.500
U,Ls 300 5 -27.5 - 55 -0.50 + 27.500
U,Lr 240 2 +30 + 120
U2L2 240 2
UzLz 240 2
UJ., 240 2 +22 + 88 +0.40 + 35.200
s +63 +244.70

SOLUTION. The values of S, u for A// at L3, and u for Af at L3, are
shown in Fig. 13-46cc?. Computations for A// and Af at joint L3 are
shown in Table 13-2. Thus

Ai/ of Lj -64.8 X 10-3 in. or 0.0648 in. to the left

Af of Lj +320.7 X 10-3 in. or 0.3207 in. downward

Table 13-2

AT ^^ u for Ah at Lz Af at Lz
Mem- = 2w u for Af
= 2u
sq Ah at AL, AL,
ber in. kips 10-3 in. at Lz
in. Lz 10-3 in. 10-3 in.

72 4 +36 +21.6 + 1.50 + 32.4

U,U2 72 4 + 18 + 10.8 +0.75 + 8.1
LoLi 72 4 -54 -32.4 + 1 -32.4 -2.25 + 72.9
L1L2 72 4 -36 -21.6 + 1 -21.6 -1.50 + 32.4
L2L3 72 4 -18 -10.8 + 1 -10.8 -0.75 + 8.1
f/oLi 120 5 +30 +24 + 1.25 + 30.0
C/1L2 120 5 +30 +24 + 1.25 + 30.0
C/2L3 120 5 +30 +24 + 1.25 + 30.0
C/iLi 96 2 -24 -38.4 -1 + 38.4
C/2L2 96 2 -24 -38.4 -1 + 38.4
2 -64.8 +320.7

L
 ^

ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 299

Vo (4) Ui (4) U2

24 k
Numbers in ( ) are areas in sq in.
£= 30,000 k/in2 S in kips
(a) (b)

— f/o f^i
2.25

1 2.25

u for Afj at L3 u for Ay at L3

Fig. 13-4

Example 13-3. Determine the relative movement in the direction

L1U2 between the joints Li and U2 of the truss shown in Fig. 13-5a by
the unit-load method.
SOLUTION. The separating relative movement between the joints Li
and U2, in the direction L1U2, is equal to the sum of the tw^o absolute
movements in the direction L1U2 of joints Li and U2 (the movement
upward to the right of joint U2 and the movement downward to the left

16 k

Numbers in ( ) are areas in sq in.

£= 30,000 k/in.2 S in kips Values of u
(a) (&) (c)

Fig. 13-5

of joint Li). Each of the two absolute movements may be found by

applying a single unit load at a time, but by applying a pair of unit loads
at Liand U2 as shown in Fig. 13-5c, the combined relative movement may
be found directly by Zi^(AL), in which u is the stress in any truss member
due to the pair of unit loads. As shown in Table 13-3, this relative
movement is +83.25 X IQ-^ in., or 0.08325 in. away from each other.
13-3. Deflections of Statically Determinate Trusses : the Graphical
Method. Since a truss is an assembly of triangles, the sides of which
300 ELEMENTARY THEORY OF STRUCTURES
Table 13-3

A, SL
A = 2u(AL),
Member sq u
10-3 in.
in. kips 10-3 in.
in.

UlU2 180 4 -12 -18 +0.6 -10.8

LqLi 180 4 + 10.5 + 15.75
LIL2 180 4 + 10.5 + 15.75 +0.6 + 9.45
L2LZ 180 4 + 12 + 18
uu. 300 5 -17.5 -35
C/1L2 300 5 + 2.5 + 5 -1 - 5
UzLz 300 5 -20 -40
U,L, 240 2 + 12 +48 +0.8 +38.4
U2L2 240 2 + 16 +64 +0.8 +51.2
2 +83.25

remain straight while undergoing changes in length, it can be surmised

that the shape of the deformed trussmay be graphically determined by
using the new lengths of members as the sides of the component triangles.
But the changed lengths of the members are only a little longer or shorter
than the original lengths, a fact which makes the deformed truss almost
coincide with the original truss when ordinary scales are used. This
difficulty may be avoided by using two different scales when plotting
the original lengths L and the changes in length AL.
Consider, for example, the Warren truss ABODE in Fig. 13-6a. Let
+7 units be the lengthening in member AB; +6, +3, and +9 units the
lengthenings in members BC, BD, and BE; —8, —5, and —4 units the
shortenings in members AC, CD, and DE. The shape of the deformed
truss may be determined by drawing the new triangles A'B'C, B'C'D',
and B'D'E' in succession.
Starting with triangle A'B'C assume
, that joint A' of the deformed
truss coincides with joint A of the original truss (Fig. 13-6a), and assume
also that the direction of A'B' coincides with that of AB. B' then, must
,

fall at 7 units to the right of B in the direction AB. Joint C the only ,

unknown point on triangle A'B'C may , be determined by the intersec-

tion of two A' and B' as centers and the lengths of A'C and
arcs, using
B'C as radii. This is performed by the following procedure: From C
measure CCi = 8 units toward A (or A') because member AC is short-
ened by 8 units. ACi is then the new length of member AC. Now,
instead of drawing the arc with A' as center and A'Ci as radius, which
can be done only if CCi is plotted on the same scale as that for AC, a.
perpendicular to AC is drawn at Ci. This is an approximation which is
permissible because the deformations are very small in comparison with
ANALYSIS OF STATICALLY INDETERMINATE
TRUSSES 301

(h)

Fig. 13-6
302 ELEMENTARY THEORY OF STRUCTURES
the original lengths of the members. This also permits CCi to be plotted
equal to 8 units and to a large scale. Now draw B'C2 parallel and equal
to BC and extend 6 units beyond C2 to C3. B'C^ is the new length of mem-
ber BC because B'C2 represents the original length and C2C3 = 6 units
is the increase in length. A perpendicular to CzB' at C3 and the one
previously drawn at Ci will intersect at C. Triangle A'B'C is now
completed.
of joint C may be scaled from C to
Although the displacement in C
Fig. 13-6a, be preferable to isolate and draw the displacement
it will
polygon CC2CzC'CiC separately as shown in Fig. 13-6c?. In this diagram
only AL is involved and therefore the scale may be made as large as
desirable.
Next, draw the triangle B'C'D' for which B' and
, C have already been
located. Draw C'Di parallel and equal to CD. From Di measure
D1D2 = member CD shortens
5 units to the left because the units.
C'D2 is the new member CD. Draw B'D^ parallel and equal
length of
to BD and from D^ measure D^D^ = 3 units which is the increase in
length of member BD. B'D^ is the new length of member BD. Per-
pendiculars to lines C'D2 and B'D^,, erected at points Z>2 and D^, respec-
tively, D' which completes the triangle B'C'D'. Note
intersect at ,

that the displacement polygon DD^D^D'D2DiD may be draw^n sep-

arately, as shown in Fig. 13-6e. The displacement of joint D is from
D to D'.
It is now necessary to locate joint E' in the triangle B'D'E' . Draw
D'Ei parallel and equal to DE. From Ei measure E1E2 = 4 units toward
jD' because this is D'E2 is the new
the shortening in the member.
length of member DE. Prolong B'E to E^, making B'E^ = BE, or
EE3 = 7 units. Because member BE lengthens, from Ez lay off

EsEi = 9 units
to the right. B'Ea is the new length of member BE. Perpendiculars to
lines D'E2 and B'E^, erected at points E2 and E4, respectively, intersect
at E', which is the new location of joint E. The displacement polygon
EEaE'E2EiE may also be drawn separately, as shown in Fig. 13-6/.
If the displacement polygons, shown separately in Fig. 13-66 to /, are

superimposed upon one another, and points A, B, C, D, and E are kept

coincident, the combined diagram of Fig. 13-Qg will be obtained. The
deflections of joints A, B, C, D, and E may be scaled in Fig. 13-6^' from
the common point A {B,C,D, or E) outward to points A' (A A' = 0),
B', C, D' and E\
, respectively. This common point A (or A') is called the
reference point. The combined diagram, shown as Fig. 13-6^, is called
the deformation diagram or the Williot diagram.
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 303

The preceding explanationof the Williot diagram has been given in

detail to ensure a basic understanding of the analysis. In practice,
however, this construction may be greatly simplified. After choosing
joint A as the reference point and member AB sls the reference member,
start with A' (Fig. 13-6^) and then, because the lengthening in AB is
7 units, locate point B' 7 units to the right of A. Because joint C
moves 8 units downward to the left relative to joint A, from A' lay off
A^Ci = 8 units downward to the left and parallel to the direction of
member AC. Since joint C moves upward to the left relative to joint
B, from B' measure B'Cz = 6 units upward to the left and parallel to the
direction of member BC. The two perpendiculars to the directions of
members AC and BC, erected at points d
and Ca, respectively, intersect
at C. From C draw C'D2 = 5 units horizontally to the left; and from
B' measure B^D^ = 3 units upward to the right in the direction of
member BD. The two perpendiculars drawn at D2 and 2)4 intersect
at D'. From D' lay off D'E2 = 4 units upward to the left in the direction
of member DE, and from B' draw B^E^ = 9 units horizontally to the right.
The two perpendiculars drawn at E2 and E4 intersect at E'. While
studying this paragraph the reader is advised to make an independent
sketch of the Williot diagram as shown in Fig. 13-6g.
The deformed truss now assumes the form A'B'C'D'E' as shown in
Fig. 13-6a, and the movement of each joint may be measured in Fig.
13-6^ from the reference point to the single prime point in question.
Obviously it is necessary to rotate the deformed truss A'B'C'D'E' of

Fig. 13-6a through a clockwise angle, in radians, of E'E^ divided by the

span of the truss so as to bring the point E' to the same level as the hinge
at Aand to coincide with point E^. The additional displacements of
all joints due to this rotation may be found as follows: In Fig. 13-6g', a
vertical line through the hinge A (also called A' or A") and a horizontal
line through E' are made to intersect at E" (see Fig. 13-6;); then the
vertical distance E" A" is the movement of joint E due to the above-
mentioned rotation. With A"E" as a base (Fig. 13-6/i), A"B"C"D"E"
is drawn similar to the original truss ABCDE.In this case, the rotation
diagram A"B"C"D"E" should be drawn on the left of the base A"E"
because the criterion is that A"B"C"D"E" can be rotated 90° to a
position parallel and similar to the original truss, ABCDE.
can be shown that the movements of joints B, C, D, E due to this
It
rotation may be scaled in Fig. 13-Qh from the double-prime point in
question to the hinge A (A' or A"). For instance, it is required to
prove that D"A" in Fig. 13-6/i equals Une DA in Fig. 13-6a times the
angle of rotation and that D"A" in Fig. 13-6/t is perpendicular to line
DA in Fig. 13-6a.
304 ELEMENTARY THEORY OF STRUCTURES
PROOF. Since triangle D" A"E" (Fig. 13-6/i) is similar to triangle DAE
(Fig. 13-6a), E"A" is perpendicular to EA, and
angle D"A"E" = angle DAE
D" A" is perpendicular to DA. Also,

D"A" E"A"
DA EA
E"A"
but angle of rotation
EA
D"A"
Therefore, angle of rotation
DA
or D"A" {DA) times the angle of rotation

If Fig. 13-6/i is superimposed on Fig. 13-6^, Fig. 13-6/ is obtained.

The total movement of each joint is the vector sum of the ''rotation
movement" from the double-prime point to the hinge A {A' or A") and
the ''deformation movement" from the reference point (in this case it
happens that the hinge is chosen as the reference point) to the single-
prime point. (The vector diagram for joint D is shown as Fig. 13-6z.)

Double prime point

(any joint)

Hinge
(singleprime \<^
or double prime; \^ ^ ^
Single prime point
(any joint)

Reference point
(single prime)

Fig. 13-7

More simply, the total movement may be measured directly from the
double-prime point to the single-prime point in Fig. 13-6;. Thus, in the
graphical solution for the magnitude and direction of the deflection of
each joint in a truss, a diagram similar to Fig. 13-6j is all that is necessary
and the deflection is always measured from a double-prime point to the
corresponding single-prime point. Figure 13-6^ is known as the Williot
diagram. Fig. 13-6/i as the Mohr diagram, and Fig. 13-6j as the Williot-
Mohr diagram.
If some joint other than the hinge had been chosen as the reference

point, then owing to deformation only the hinge would have moved
from the reference point to the single-prime point for the hinge. It is
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 305

necessary, therefore, in addition to the deformation and rotation effects,

to translate the deformed truss through the displacement from the single-
prime point for the hinge back to the reference point. Figure 13-7
shows the combination of all three vectors, but again, the total movement
may be measured directly from the double-prime point to the single-
prime point.
t/l (4)
t/2 t/i -12 .^2 ^1 -13 ^2

Numbers in ( ) are areas in sq in.

£- 30,000 k/in2 S in kips AI in 10-3 in.
(a) (&) (0

Uif

U{ 1^"-

15.75

Reference point. Lq Reference point. U\

Reference member LqL\ Reference member. U\L2
id)

Note. Deflection is measured from the double

prime point to the single prime point

Fig. 13-8

Example 13-4. Using the graphical method, determine the horizontal

and vertical deflections of all joints of the truss shown in Fig. 13-8a.
SOLUTION. The stresses in all members of the truss due to the applied
loading and the corresponding changes in the lengths of the members
are shown in Fig. 13-86 and c.
306 ELEMENTARY THEORY OF STRUCTURES
Two graphical solutions areshown in Fig. IS-Sd and e. It is to be
noted that, had the same scale been used in both solutions, that shown
in Fig. 13-8e would take much less space than does the solution in Fig.
IS-Sd. A good choice of reference point and the selection of a reference
member with comparatively little rotation will require a relatively small
rotation or correction diagram and therefore will yield a more compact
graphical solution than might otherwise be obtained. Thus, if the
assumed reference member or line does not rotate, the Mohr diagram
reduces to a point and vanishes, as illustrated in Example 13-5. Note
also that the rotation diagram should be drawn to the left or right of LJ'Lj',
depending on whether a 90° rotation will bring it to a position parallel
and similar to that of the original truss.
Example 13-5. Using the graphical method, determine the horizontal
and vertical deflections of all joints of the truss shown in Fig. 13-9a.
SOLUTION. The stresses in all members of the truss due to the applied
loading and the changes in the lengths of the members of the truss are
shown in Fig. 13-96 and c.
Two graphical solutions are shown in Fig. 13-9d and e. Note that,
in Fig. 13-9c, a fictitious member UqLq is added, and its deformation is
zero. In Fig. 13-9c?, the Mohr rotation diagram is a point diagram,
viz., all the double-prime points coincide. The rotation diagram in
Fig. 13-9e is constructed by first identifying Uq Avith Uq and Lq with LJ
and then making a similar truss diagram with UqLq as a base and at 90°
to the original horizontal position of the truss.
13-4. Analysis of Statically Indeterminate Trusses by the Method of
Consistent Deformation. A truss may be statically indeterminate
because of the presence of (1) external redundant reaction components,
(2) internal redundant members, or (3) a combination of external and
internal redundants. The analysis of statically indeterminate trusses
by the method of consistent deformation involves removing the external
redundant supports or cutting the internal redundant members and then
treating their actions as unknown forces on the basic determinate truss.
These unknown forces are then solved from an equal number of conditions
of the geometry of the deformed truss, requiring zero deflection in the
direction of the external redundant or a relative movement between the
cut ends of the internal redundant equal to the deformation of the
redundant member. Because of the limited scope of this text, the follow-
ing examples will deal only with trusses statically indeterminate to the
first degree. One of these has an external redundant reaction and the
other involves an internal redundant member.
Example 13-6. Analyze the statically indeterminate truss shown in
Fig. 13-1 Oa by the method of consistent deformation.
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 307

.^0 (4) A (4) ^2 54if^0 +36 ^1 +18 ^^2 1^^0+21.6^1+10.8^2

24 k
24 k
Numbers in ( ) are areas in sq in.
£= 30,000 k/in.2 5 in kips AL in 10-3 in.
(a) (c)
ib)

L'q-Uq-Lq

L'o-L'q

Reference point. Uq Reference point, t/j

Reference member. UoLq Reference member. U1L2
id) (e)

Note. Deflection is measured from the double

prime point to the single prime point

Fig. 13-9

SOLUTION. Any one of the three vertical reactions may be chosen as

the redundant force. If R2 is assumed to be redundant, the truss in
Fig. 13-lOa is the equivalent of the two trusses shown in Fig. 13-106 and c.
For zero deflection at L2,

02

in which A 2 is the downward

due to the applied loads in
deflection of L2
the basic determinate truss, and deflection at L2 due to a
82 is the upward
1-kip upward load at L2 in the basic determinate truss. However, as
shown in Fig. 13-11, 62 may also be the downward deflection at L2 due to
308 ELEMENTARY THEORY OF STRUCTURES
a 1-kip downward load at L2. Let S be the stress in any member of
the truss in Fig. 13-lOa, aS' the stress in the corresponding member of the
simple truss in Fig. 13-106, and u the stress in this same member in the
truss shown in Fig. 13-11. Then
Ul (4) Uz (4)

AE
A_2
R2 = S' - R2U
52

Numbers in ( ) are areas in sq in. The computations may be conven-

£= 30,000 k/in.2
(a)
shown in Table
iently tabulated as
13-4. The answer diagram is shown
in Fig. 13-12. As noted in Fig.
13-12, it is important to be sure that
the two resolution equations are
satisfied at every joint of the truss.
When a solution as shown by the
answer diagram of Fig. 13-12 has
been obtained, it is desirable to
make an independent check of the

Ui t/2 U3 t

\ \
Lo/ \ \ \
c
L3 La
"'"
t t
0.6 k 1 k 0.4 k

For values of u
Fig. 13-11

Ui +5.59 U2 +5.59 f^3 -8.14 U4

11.27 k
27.88 k

Answer diagram
Fig. 13-12

calculated stresses. One procedure is to assume either Ro or 7?5 as the

redundant and then make another independent solution of the problem.
It may be more convenient, however, to use a basic determinate struc-
ture different from the one used in the initial analysis and then check
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 309

Table 13-4

A, S'uL
Mem- ^, -S', -R2U, S = S' - Riu,
u AE' AE'
ber in. kips 10-3 in. 10-3 in. kips kips
in.

U,U2 180 4 -19.5 -0.90 + 26.325 + 1.215 +25.09 + 5.59

U2Uz 180 4 -19.5 -0.90 + 26.325 + 1.215 +25.09 + 5.59
UzU, 180 4 -16.5 -0.30 + 7.425 +0.135 + 8.36 - 8.14
LqLi 180 4 +21 +0.45 + 14.175 +0.30375 -12.55 + 8.45
uu 180 4 +21 +0.45 + 14.175 +0.30375 -12.55 + 8.45
L2L3 180 4 + 18 +0.60 + 16.200 +0.540 -16.73 + 1.27
L3L4 180 4 + 18 +0.60 + 16.200 +0.540 -16.73 + 1.27
L4L5 180 4 + 16.5 +0.30 + 7.425 +0.135 - 8.36 + 8.14
LoC/i 300 5 -35 -0.75 + 52.500 + 1.125 +20.91 -14.09
u,u 300 5 - 2.5 +0.75 - 3 750 . + 1.125 -20.91 -23.41
UUz 300 5 + 2.5 +0.50 + 2.500 +0.500 -13.94 -11.44
UzL, 300 5 - 2.5 -0.50 + 2.500 +0.500 + 13.94 + 11.44
UJLs 300 5 -27.5 -0.50 + 27.500 +0.500 + 13.94 -13.56
U,L, 240 2 +30 +30
UJL^ 240 2
UzLz 240 2
UJ., 240 2 +22 +0.40 + 35.200 +0.640 -11.15 + 10.85
Ro +28 +0 60 -16.73 + 11.27
R. -1 +27.88 +27.88
Rf> +22 +0.40 -11.15 + 10.85
2 +244.700 +8.7775
(= A2) (= 52)

a geometrical condition. For instance, a truss supported at Lo and Li

and subjected to the loads of 30, 20, and 10.85 kips at Lo, L2, and L5,
respectively, as shown in Fig. 13-13a may be used. Note that the load
at L5 is the previously calculated reaction but that it is now considered
^1 +5.59 ^2 +5.59 ^3 -8.14 Ua Ui +2.25 U2 +2.25 ^3 +0.75 ^^4

11^7 k

Fig. 13-13

to be an upward load acting at L5. The condition of consistent deforma-

tion is that the deflection of this truss at L5 should equal zero. Values
of u are shown in Fig. 13-136, and computations for the deflection at L5
are arranged in Table 13-5. Thus, when a solution satisfies the necessary
conditions of both statics and geometry, its correctness is ensured.
310 ELEMENTARY THEORY OF STRUCTURES

Table 13-5

A, AT ^^
s, u AL,
Member sq u
in. kips 10-3 in.
in. 10-3 in.

U1U2 180 4 + 5.59 + 8.39 +2.25 + 18.88

C/2C/3 180 4 + 5.59 + 8.39 +2.25 + 18.88
C/3C/4 180 4 - 8.14 -
12.21 +0.75 - 9.16
LqLi 180 4 + 8.45 +
12.67 -1.125 -14.25
L,L2 180 4 + 8.45 +
12.67 -1.125 -14.25
L2L3 180 4 + 1.27 +
1.90 -1.5 - 2.85
L3L4 180 4 + 1.27 +
1.90 -1.5 - 2.85
L4L5 180 4 + 8.14 - 12.21 -0.75 - 9.16
Lot/i 300 5 -14.09 - 28.18 + 1.875 -52.84
C/1L2 300 5 -23.41 - 46.82 -1.875 +87.79
L2C/3 300 5 -11.44 - 22.88 -1.25 +28.60
t/3L4 300 5 4-11.44 + 22.88 + 1.25 +28.60
C/4L5 300 5 -13.56 - 27.12 + 1.25 -33.90
C/iLi 240 2 +30 + 120
C/2L2 240 2
C/3L3 240 2
C/4L4 240 2 + 10.85 + 43.40 -1 -43.40
2 + . 09 «

Example 13-7. Using the method of consistent deformation, analyze

the statically indeterminate truss shown in Fig. 13-14a.
SOLUTION. The given truss has no external redundant, but internally
ithas one redundant member (m = 10, j = 6). Any one of the six
members around or within the middle panel may be cut without impair-

Numbers in ( ) are areas in sq in.

£- 30,000 k/in^
(a)

Normally, however, one of the two

ing the static stability of the truss.
diagonals is taken as the redundant because obviously there is one extra
or unnecessary diagonal in the middle panel. If member L1U2 is cut,

the given truss may be replaced by the two basic determinate trusses
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 311

shown in Fig. 13-146 and c. One of these carries the applied loads and
the other is subjected to the pair of forces X and X, the unknown tensile
stress in the redundant member. Let

A = relative movement together between Li and U2 due to the

applied loads
6 = relative movement together between Li and U2 due to a pair of
1-kip loads at joints Li and U2 and acting in the direction of
the cut member (Fig. 13-15)

Then the condition for consistent deformation becomes

XL
(A + X6) = -f
AE
or X = -
8 + (L/AE)
because actually joints Li and U2 should move apart to accommodate an
elongation of XL/ AE in the redundant member. Let S' be the stress

v-^ ^dJ Values of u

Fig. 13-15

in any member shown in Fig. 13-146, u be that due to a pair

of the truss
of 1-kip loads at joints Li and U2 as shown in Fig. 13-15, and S be the
resultant stress in the member, then

u'L
S = S' -huX
AE
Note that the above two summations do not include the redundant
member.
From Table 13-6, it is seen that

A = y -^ = -83.25 X 10-3 in.

and 5 - V ^'^ - -h8.20 X 10-3 in.

312 ELEMENTARY THEORY OF STRUCTURES

Table 13-6

^, u u
Mem- Tu S' S'uL u'^L S = SuL
sq (Fig. {u){X) (Fig.
ber in.
in.
kips
13-15)
AE AE ,S' +uX 13-17)
AE
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)

UxU^ 180 4 -12 -0.6 + 10.8 +0.54 -4.90 -16.90 +0.6 -15.21
LqLi 180 4 + 10.5 + 10.50
uu 180 4 + 10.5 -0.6 - 9.45 +0.54 -4.90 + 5.60 +0.6 + 5.04
uu 180 4 + 12 + 12.00
LoUl 300 5 -17.5 -17.50
u,u 300 5 + 2.5 + 1.0 + 5 +2.00 +8.16 + 10.66
L1U2 300 5 +8.16 + 8.16 -1 -16.32
u,u 300 5 -20 -20.00
C/iLi 240 2 + 12 -0.8 -38.4 +2.56 -6.53 + 5.47 +0.8 + 17.50
U2L2 240 2 + 16 -0.8 -51.2 +2.56 -6.53 - 9.47 +0.8 +30.31
s -83.25 +8.20 +21.32
(= A) ( = 5)

Therefore
(-83.25) 83.25
X =
300
+ 10.20
= +8.16 kips
(+8.20) +
5 X 30

With the stress in the redundant known, the total stress S in any member
is found from S = S' -\- uX, as shown in column 9 of Table 13-6. The
answer diagram is shown in Fig. 13-16.

^1 -16.90 U2

Answer diagram
Fig. 13-16

conditions of static equilibrium may be checked by making sure

The
that the two resolution equations are satisfied at every joint as shown in
Fig. 13-16. The geometrical condition is checked by calculating the
relative movement between joints Ui and L2 of the truss in Fig. 13- 17a,
which is the truss shown in Fig. 13-16 with member U1L2 cut and its

action replaced by a pair of forces of 10.66 kips each. As shown in Table

13-6 (columns 10 and 11), joints Ui and L2 move 21.32 X 10-^ in. apart,
which is just equal to the elongation caused by a tension of 10.66 kips
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 313

Vi -16.90 f:'2 Ui -kjjs U2

V^alues of u
ib)

Fig. 13-r

in member U1L2. The correctness of the solution is ensured because the

conditions of statics and geometry are now both satisfied.
13-5. Influence Diagrams for Statically Indeterminate Trusses. A
truss supported at more than two points is called a continuous truss and
is, of course, statically indeterminate. In the analysis of statically inde-
terminate trusses carrj'ing a system of mo\'ing loads, influence diagrams
for the reactions or stresses in the truss members must be constructed in
order to determine critical loading positions. The discussion in this
article will be limited to the determination of the influence diagram
for the intermediate reaction on a two-span continuous truss.

X = A

t Ik t
Rb
tRc Ik
(a) (4
Fig. 13-18

Letbe required to construct the influence diagram for Rb of the

it

truss shown
in Fig. 13-18a. In the development of this influence dia-
gram, it will be recalled that the value of Rb due to a 1-kip load at P
must be plotted directly under P, as shown in Fig. 13-20. According
to the method of consistent deformation (Fig. 13-18), the value of Rb due
to a 1-kip load at P is

'BB

By the law of reciprocal deflections.

Sbp =
Therefore Rb =

Values of Bpb and Sbb may be taken directly from Fig. 13-196, which
is the elastic cur^-e of the simple truss AC due to a unit load at B. Thus
the required influence diagram which is shown in Fig. 13-206 is similar
to the elastic curv^e of Fig. 13-196. The scale has been changed because
314 ELEMENTARY THEORY OF STRUCTURES
the ordinates to the elastic curve have been divided by the constant
dsB and these ratios are plotted in the reverse direction as the positive
ordinates of the influence diagram. Thus it is seen that the influence
diagram for Rb is actually the deflection diagram obtained by applying
a load at B of the simple truss ^C so that the deflection caused at B is

equal to unity.

(fc) Elastic curve

Fig. 13-19

Similarly, influence diagrams for Ra or Re may be obtained inde-

pendently by introducing a load at A or C of the overhanging truss BC
or AB so that the deflection at A or C is equal to unity.
Example 13-8. Construct the influence diagram for the reaction at L2
in the truss shown in Fig. 13-2 la.
SOLUTION. The required influence diagram is shown in Fig. 13-216.
To calculate the influence ordinates, the reaction at L2 is removed and a

(6) Influence diagram for Rb

Fig. 13-20

load of 1 kip is placed at L 2 as shown in Fig. 13-22a. Then by any con-

venient method, either algebraic or graphic, determine the vertical deflec-
tions of the lower-chord panel points of the truss L0L5 and plot the elastic
curve as shown in Fig. 13-226.
Inasmuch as the influence ordinate at L2 should be unity, divide the
ordinates to the deflection curve in Fig. 13-226 by 8.7775 and plot these

f
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 315
Ul (4) U2 (4) t/3 (4) U4

Numbers in ( ) are areas in sq in

£- 30,000 k/in2
(a)

(6) Influence diagram for R2 (6) Elastic curve in 10-3 in.

Fig. 13-21 Fig. 13-22

ratios above the base line as shown in Fig. 13-216. This is the required
influence diagram.

PROBLEMS
13-1. Determine the horizontal and vertical deflections of joint L3 of the truss
shown by the unit-load method.
13-2. Determine the horizontal and vertical deflections of joint L5 of the truss
shown by the unit-load method.
13-3. Determine the horizontal and vertical deflections of joints Lq and L2 of the
truss shown by the unit-load method.
13-4. Determine the relative movement in the direction U2LZ between the joints
U2 and L3 of the truss shown by the unit-load method.
13-5. Determine the relative movement in the direction L2UZ between the joints
Lz and U3 of the truss shown by the unit-load method.
13-6 to 13-8. Determine the horizontal and vertical deflections of all joints of the
truss shown by the graphical method.

Area of all chord members = 3 sq in.

Area of all web members = 2 sq in.
£- 30.000 k/in2
Fig. P13-1, 6

13-9. Determine the relative movement in the direction U2L3 between the joints
C/2and Lz of the truss shown by the graphical method.
13-10. Determine the relative movement in the direction LzUz between the joints
L2 and Uz of the truss shown by the graphical method.
316 ELEMENTARY THEORY OF STRUCTURES
Ui U2 U3 U4 Us

3 at 18' = 54' 2 at 18' = 36'

Area of all chord members = 3 sq in.

Area of all web members = 2 sq in.
£- 30,000 k/in.2
Fig. P13-2, 7

u2,

^0
Li L2
(4) \ '^ 'i^
4 at 6' = 24'

25 k

Numbers in ( ) are areas in sq in.

jE:= 30,000 k/in.2

Fig. P13-3, 8

CTi (4) C/2 (4) C/3 (4) t^4 I7l (4) 1/2 (4) f/3 (4) ^1

/(4) (4)\ (4)\ /(4) (4)\ ^

f Li L2 ^kL3 L4 L5
36' ,
24'
1

15 k
Numbers in ( ) are areas in sq in. Numbers in ( ) are areas in sq in.

£- 30,000 k/in.2 £- 30,000 k/in.2

Fig. P13-4, 9 Fig. P13-5, 10

t/l Ui t/3 t/4 [/g

/ \/\/^/\/\^
-^
iio ^' 12k^
20 k¥^
3 at 18' = 54' 2 at 18' = 36'
p •r
Area of all chord members = 3 sq in.
Area of all web members - 2 sq in.^
£= 30,000 k/in.2
Fig. P13-11, 12

13-11. Using the reaction at Lz as the redundant, analyze the statically indetermi-
nate truss shown by the method of consistent deformation.
13-12. Using the reaction at Lg as the redundant, analyze the statically indetermi-
nate truss shown by the method of consistent deformation.
 ANALYSIS OF STATICALLY INDETERMINATE TRUSSES 317

Ui (4) U2 (4) Us (4) U4

to

(4)\ /(4)\ /(4) (4)\

'^ L, L2 ^L, L, ^5
36' 1 24'
' "

1
15 k
Numbers in ( ) are areas in sq in.
£= 30,000 k/in 2
Fig. P13-13, 14

13-13. Using member U2L3 as the redundant, analyze the statically indeterminate
truss shown by the method of consistent deformation.
13-14. Using member L2U3 as the redundant, analyze the statically indeterminate
truss shown by the method of consistent deformation.

Area of all chord members = 3 sq in.

Area of all web members = 2 sq in.
£= 30,000 k/in.2
Fig. P13-15

13-15. Construct the influence diagrams for the reactions at Lq, L3, and L5 in the
truss shown.

Ui (4) U2 (4) Us (4) U4

f4
/(4) (4)\ /(4)\ /(4) (4)\
^ Li
36'
L2 ^kLs L4 L5
24'
'
I "

Numbers in ( ) are areas in sq in.

£:- 30,000 k/in.2

Fig. P13-16

13-16. Construct the influence diagrams for the stresses in members C/2^3 and
L2U3 in the truss shown.

i
 CHAPTER 14

THE SLOPE-DEFLECTION METHOD

14-1. General Description of the Slope -deflection Method. The

slope-deflection method may be used to analyze all types of statically
indeterminate beams or rigid frames. In this method all joints are con-
sidered rigid; i.e., the angles between members at the joints are considered
not to change in value as loads are applied. Thus the joints at the
interior supports of statically indeterminate beams can be considered 180°
rigid joints; and ordinarily the joints in rigid frames are 90° rigid joints.
When beams or rigid frames are deformed, the rigid joints are considered
to rotate only as a whole; in other terms, the angles between the tangents
to the various branches of the elastic curve meeting at a joint remain the
same as those in the original undeformed structure.
In the slope-deflection method the rotations of the joints are treated
as unknowns. It willbe shown later that for any one member bounded
by two joints the end moments can be expressed in terms of the end
rotations. But, to satisfy the condition of equilibrium, the sum of the
end moments which any joint exerts on the ends of members meeting
there must be zero, because the rigid joint in question is subjected to the
sum of these end moments (only reversed in direction). This equation
of equilibrium furnishes the necessary condition to cope with the unknown
and when these unknown joint rotations are found,
rotation of the joint,
the end moments can be computed from the slope-deflection equations
which will be derived in the next article.
The discussion in the preceding paragraph will be clarified by the
following example: suppose that it is required to analyze the rigid frame
loaded as shown in Fig. 14- la. This frame is statically indeterminate
to the sixth degree. The method of consistent deformation could be
used, but the amount work involved would make that method too
of
laborious. Because the frame is kept from horizontal movement by
its connection at A and vertical movement is prevented by the fixed

bases at D and E, and since axial deformation of the members is usually

neglected, all joints of this frame must remain in their original locations.
(The cases in which some joints may change position when the frame is
deformed will be taken up later.) Clockwise joint rotations, as shown in
318

-»
 fHE mX»FEr^EFLJECnOSi: MHHraOD 319

F%, I4rla, are conadoed to be pmiim* The free-bodjr cliagTaiEfis of all

membas aieshcmn in ¥igr 14>1^. At uay one end of each member^
tiiere aie three leaetion conqMrnents: direct poll or thmst^ end diear^ and
end mnment^ Tbe end moment wbidk acts at end A of member AB i»
denoted as JTi^; that at end B of member AB„ as JifA«, Cmmiewthdtwiae
end moments acting cm the members aie considered to be 'potiiiWj. po»-
thre end moments being diovn in Fig^ 14-lfr. It is posnible^ bf the use
of the dope-deflection equations to be derired in the next ardde^ to
e>iw gsM the end
; moments of each member in terms of the end rotations
and the loading which acts on the member. Unis the ci^ end

^ ^ B
M.
—h
^
- ^ ^ir

^^c
^G|| > ^c P^*^
\r)Msjj r^^'McE

m
J^3-fO*kr "^^^
|OiWfo«

-J^EC

m. <tM« pro^jTem rxi5.7 h^. -^^xp r»^ss5ed in terms of the twO' Tinkno'wn joirit

ro'naini«>CL§- cre<r-oodv dx^gFain^ of all jiOTiits- are ^Kqwh m Fig. 1-t-ic,

T'tif:

Of eoorse^ the action of the member on the joint consbts of a force in the
directian of the axis of the member^ a foioe pcipendicallar to this axis^ and
a moment^ each being the oppcnite of the action of the joint on the mem-
ber. In Fig, 14-lc:, on^ the moments are Aasmn, These moments are
drawn in their 'pomiiam direction^ which is cioobetsr. Fck- eqnilibfinm,
summation of all moments acting at each joint must be lero. Thus
Joint eondfition at Bz t+ Mmc + Mmi^ =
Joint condition at Ci Mem + McM =
The abore two equations are and sufficient to detemnne the
of $m and fc- All end then be foand bj
320 ELEMENTARY THEORY OP STRUCTURES
the known joint rotations in the slope-deflection equations. By the prin-
ciples of statics the direct stress and shear and bending-moment diagrams
for eachmember may be found.
It has been repeatedly pointed out that the analysis of statically
indeterminate structures must satisfy both statics and geometry. In the
slope-deflection method of analyzing rigid frames, the conditions required
of the geometry' of the deformed structure, which are those of the rigidity
of the joints, are satisfied at the outset by calhng the joint rotation one
single unknown at each joint. Thus the conditions of statics, requiring
that the sum of moments acting on each joint be zero, are used to solve
for the joint rotations.
14-2. Derivation of the Slope -deflection Equations. In the slope-
deflection equations, the end moments acting at the ends of a member are
expressed in terms of the end rotations and the loading on the member.

^b|<U
^
""bIMba MpAB MfBA
^'
^Al ^
0B1
0A2

Constant EI
(o) (6) (c) («ft

Fig. 14-2

Thus for span AB in Fig. 14-2a it is required to express Mab and

shown
Mba in terms of theend rotations 6 a and Ob and the apphed loading Pi
and P2. Note that the end moments are shown as counterclockwise
(positive) and the end rotations are shown as clockwise (positive). Now,
with the apphed loading on the member, the fixed end moments Mfab
and Mfba (both shown as counterclockwise) are required to hold the
tangents fixed at the ends (Fig. 14-26). The additional end moments 3/^
and Ms should be such as to cause rotations of Ba and Ob, respectively.
If Sai and $bi are the end rotations caused by M'a and 6a2 and dB2 by

M'b (Fig- 14-2c and d), the conditions required by geometry- are

6a = —Sai-\- 6a2
(14-1)
6b = -\-6b\ -~ 6b2
By superposition,
Mab = M,AB + M'a
(14-2)
Mba = MfBA + M'b

By the conjugate-beam method

noa,

M'aL
ui
3EI ^^' - 6ET
(14-3)
M'bL
U2
6EI ^^' - 3El
 THE SLOPE-DEFLECTION METHOD 321

Substituting Eqs. (14-3) in Eqs. (14-1),

= _ ^'aL ^ M',L
^^
(14-4)
M'^L M'bL
63= +
6^7 3EI
Solving Eqs. (14-4) for M'^ and M^,
9F T
(14-5)
M'b = +^(-26^ - Ba)

Substituting Eqs. (14-5) in Eqs. (14-2),

MxB = MpAB + ^
1FT
{-2eA - Bb)
(14-6)
Mba = Mfba -\-^{-2dB- Oa)

Equations (14-6) are the slope-deflection equations which express the

end moments in terms of the end rotations and the appHed loading.
Note again that counterclockwise moments acting at the ends of the
member (Mab, Mba, Mfab, and Mfba) are positive and clockwise rota-
tions {Ba and Bb) are positive.
The slope-deflection equations as shown in Eqs. (14-6) express the
end moments in terms end rotations and the applied loading.
of the
If, in addition to the applied loading, the end joints are subjected to

unequal movements in the direction perpendicular to the axis of the

member, additional fixed end moments M'pj^j^ and M'pj^j^ (Fig. 14-3c) are
induced to act on the member in order to keep the tangents at the ends
fixed. Then M^ and M's should be such as to cause rotations of Ba and
Bb, respectively. As in the preceding case, the conditions required of
geometry are
Ba = —Bai-{-Ba2
(14-1)
Bb = -\-Bbi — Bb2

By superposition,

Mab = Mfab + MJ,^^ + M'^

(14-7)
Mba = Mfba + M'fba + M'^
From Eqs. (14-5),

M'a = + ~
9FT
{-2eA - Bb)
(14-5)
M'b = +^ lET
{-203 - Ba)
 322 ELEMENTARY THEORY OF STRUCTURES
P^ p. The moment -area method will be
s used to determine Mf^B ^^^^ ^^'pba
G ^^d T (Fig. 14-4). Note that R, the an-
measured from the original direc-
y gle
Constant EI ^J: I

JQ t ion of member AB to the Une join-

(a)
^^fsA
ing the displaced joints, is positive
II
when clockwise. Note also that

^ = L
G
Bv the first moment-area theorem.

cr

Moment diagram
Fig. 14-4

Change of slope between tangents at A and B'

= area of
M
^r diagram between A and B' =
EI
or ^^FAB ~ -^^ FBA

Deflection of B' from tangent ^t A =

^J [''¥^]{h)
MfabL' = A
6EI

or -^^FAB — '^^FBA ~ (1^8)

]J. L

Substituting Eqs. (14-5) and (14-8) in Eqs. (14-7),

2EI
MaB = ^J^FAB -\- {-26 A - ds + 37?)
L (14-9)
2EI - Oa^
Mba = Mfba + {-2eB sR)

iii
 THE SLOPE-DEFLECnOX METHOD 323

Equations (14-9) are the general slope-deflection equations which express

the end moments in terms of the end rotations, the applied loading, and
the angle R between the line joining the deflected end joints and the
original direction of the member. If the angle R is zero, Eqs. (14-9)

reduce to Eqs. (14-6^.

14-3. Application of the Slope -deflection Method to the Analysis of
Statically Indeterminate Beams. The procedure for analj^zing statically

UllllUllrtilllllUr JMiiM I
Wn-TT
I
Fig. 14-5

indeterminate beams by the slope-deflection method is as follows (see

(Fig. 14-5):
1. Determine the fixed-end moments due to the applied loads at the

ends of each span, using the formulas shown in Fig. 14-6.

2. Determine the known value of R for each span (R = Oin most cases

unless there are imequal settlements at the supports).

.3. Express all end moments in terms of the fixed-end moments due
to the applied loads, the known values of R. and the unknown joint
rotations by using the slope-deflection equations.
4. Establish simultaneous equations with the rotations at the supports
as unknowns by apphing the conditions that the sum of the end moments

I
i I - ^ ' '

r «
u-

*
k
«
f:

B
t

r '
'4
Fig. 14-6

acting on the ends of the two members meeting at each support is zero
(except at the fixed end).
5. Solve for the rotations at all supports.
6. Substitute the rotations back into the slope-deflection equations,

and compute the end moments.

7. Determine all reactions, draw shear and bending-moment diagrams,

and sketch the elastic cur^-e.

Exumpk 14-1. .\nalyze the beam shown in Fig. 14-7a by the slope-
deflection method. Draw shear and bending-moment diagrams. Sketch
the elastic cur\'e.
324 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. In the slope-deflection equations

Mab = Mfab + ^{-26 a- Sb)

2FT
Mba = Mfba + --^{-2dB- Sa)

the coefficient 2EI/L is different for each span. If 2EI/L values

the
for all spans are made A^ times smaller, the only effect will be to make all
d values N times larger, while the products of the expressions 2EI/L and
(— 2^near " ^far) , or thc valucs of the end moments, remain unchanged.
If the absolute magnitudes of the B values are not of direct interest,
then, the relative values of 2EI /L may be used in the above equations.
I/L are
If the relative values of called the relative stiffness K, the slope-
deflection equations become

Mab = Mfab + KAB(-2dA - M (14-10)

Mba = Mfba + KAB{-2dB - 6 a)

4k/ft Relative stiffness. Since there is

M. Cl
******* ^ I ^ i
-^
only one span (or member) in this
18'
4- beam, the stiffness of the member
Rb
Constant / may be regarded as unity.
k/ft
126 (^- f t I I I I I 1=3 Fixed-end moments.

End reactions due 4(^18)2

to applied loads
End reactions due
+ 36 +36 + 24
Mfab "*"
12
+ 108kip-ft
to end moments + 3 - 3
4(18)2
Total end reaction + 39 +33 + 24
Mfba - 108 kip-ft
Rb= 57 k
12

Slope-deflection equations. By us-

ing the modified slope-deflection
equations (14-10) and noting that

-33 6 A is equal to zero, the following

Shear diagram expressions for the end moments
64.125 are obtained:

Mab = +108 + l{-2dA - Bb)

72
= +108 - Bb
126 Mba = -108 + 1{-2Bb - Ba)
M-diagram = - 108 - 2Bb

Joint condition
Elastic curve

Fig. 14-7 Mba = -72 kip-ft

Note:End moment acting on member BA at B is 72 kip-ft clockwise,

ov Mba = -72 kip-ft.
 THE SLOPE-DEFLECnOX METHOD 325

Therefore,
-108 - 2^B = -72
Ob =- -18
Substituting,

3/,5= ^lOS -63 = +108 - (-18) = +126kip-ft

Mba = -lOS - 2t?5 = -IDS - 2(-18) = -72 kip-ft {check)

It is to be noted that, in the expression $b = — 18, the minus sign indi-

cates that the rotation atB is couni4:rclocktDise, but the value 18 is only
value of the rotation being 18 {2EI L).
relative, the true
The end momenta determined above are shown to act at the ends of
span AB (Fig. 14-7). Note again that positive end moments are counter-
clockwise when acting on the member. Thus Mab is 126 kip-ft counter-
clockwise and Mba is 72 kip-ft clockwise. The end reactions are com-
puted as the summation of those due to the applied loads and those due
to the end moments, as shown in Fig. 14-7. The shear and bending-
moment diagrams and the elastic curv^e are also shown in Fig. 14-7.
Examph 14-2. Analyze the continuous beam shown in Fig. 14-S by
the sloi>e-deflection method. Draw shear and bending-moment dia-
grams. Sketch the elastic curve.
soLiTiox. The valuer of the relative stiffness and the fixed-end
momenta are computed and shown in Fig. 14-8. Extreme care must be
exercised in determining these values because the subsequent computa-
tion, even though its own correctness can be checked and thus assured,
depends nevertheless on these preliminary- quantities.

Relative siifn^ess

AB I 24
s

BC T2
^^^

Fixed-end momenta

,,
Mfab = n
,
(15H8)
o — ,-.,. .^
= +lo kip-ft

Mfba = —15 kip-ft

Mtbc = + ih^^^ = +18 kip-ft

^fFCB = - 18 kip-ft

Slope^^fleciion equations. The modified slope-deflection equations

Mab = MrAB + KABi-^BA - Sb)

Mba = ^^FBA ~^ A.4i?(. — 2t'js — ^^i)
326 ELEMENTARY THEORY OF STRUCTURES
15k
' 4'
\ ,

T" H 1.5k/ft
Y ; ; i I ; i ; i~r ^=^=^'^
T± 12'

Constant I

Relative stiffness //8X24-(D ^/l2X24-(2) 1

Fixed end moments + 15 -15 + 18 -18

15 k
1.5 k
J_ i>Q
f t i i t i 313
25.2
End reactions due
to applied loads
+ 7.50 + 7.50 + 9.00 + 9.00

End reactions due

to end moments
-3.15 + 3.15 + 2.10 -2.10

Total end reaction + 4.35 + 10.65 + 11.10 + 6.90

i?5- 21.75 k i^= 6.90 k
+ 11.10
.35 ^•^' ,
(+41.07?^^ 1

1 (+17.4) 1

(-42.6)
{-IbAiy -6.90

Shear diagram
15.87

Elastic cur\'e

Fig. 14-8
will be used.

Mab = +15 + 3(-2(9a - Ob) = +15 - 6^a - 3^b

3/^^ = _15 4- 3(-20B _ ^^) = -15 - 6(95 - '^Ba

M^c = +18 + 2(-2^B - dc) = +18 - \eB - 2ec

3/c5 = -18 + 2(-20c - 61b)
= -18 - 4:dc - 263

Joint conditions

Joint A: Mae = -6^.1 - 3(95 = -15 (a)

Joint B: Mba + 3/bc= -3<9.4 - IOOb - 2ec (b)

Joint C: McB = -2^B - -^dc = +18 (c)

Subtracting twice (b) from (a).

+ 17eB + 4flc = -9 (d)

Adding 8.5(c) to (d),

-30ec = +144
ec = -4.8 (e)
 THB SLOPE-DEFLECnOX METHOD 327

Substituting (e) in (d),

-\-l70B - 19.2 = -9
Ob = +0.6 (f)

Substituting (f) in (a),

-6^^ - l.S = -15
e. = -2.2 (g)

Computation of end moments

Mab = -15 - (oSa - SBb = +15 - 6(+2.2) - 3(+0.6) =

Mba = -15 - ^Ob - 3^^ = -15 - 6(+0.6) - 3(+2.2^
= -25.2kip-ft
.V^c = -18 - -l^B - 2ec = +18 - 4(+0.6) - 2(-4.S
= -25.2kip-ft
M^s = -IS - 4^c - 2^5 = -IS - 4(-4.8) - 2(+0.C -

Note that the three joint conditions Mab = 0, Mba t Mbc = 0, and
Mcb = are satisfied.
The computations for the reactions, the shear and bending-moment
diagrams, and the elastic cune are all shown in Fig. 14-8.
EXAMPLE 1+3. Analyze the continuous beam shown in Fig. 14-9 by
the slope-deflection method. Draw shear and bending-moment dia-
grams. Sketch the elastic curve.
soLUTiox. The values of the relative stiffness and the fixed-end
moments are computed and shown in Fig. 1+9.
Slope-deflection equations. In this problem, 6a is zero because the
beam is fixed at A.

Mab = -15 -r 3(-2d^ - Ob) = -15 - 3^3

Mba = -15 + sl-2eB - Ba) = -lo - D^s
Mbc = +18 + 2{-2dB - Oc) = +18 - 4^^ - 2^c
Mcb = -18 + 2(-2^c - Sb) = -IS - 4^c - 2^;,

Joint conditions

Joint B: Mba + Mbc = - lO^B --2(?c = (a)

Joint C: Mcb = - 2^^ - 4^c = +1S (b)

Subtracting twice (a> from (b),

+ 18^5 = -24
Ob = -1.333 (c)

Substituting (c) inL(b),

-2(+1.333) _ 4dc = +18

dc = -5.167 (d)
 T

328 ELEMENTARY THEORY OF STRUCTURES

Computation of end moments
= +15 - 3( + 1.333) = +11.00 kip-ft
^
Mj,B= +15 - ?>eB

Mba = -15 - 6<9b = -15 - 6( + 1.333) = -23.00 kip-ft

Mbc = +18 - ^Bb -2dc = -18 - 4( + 1.333) - 2(-5.167)
= +23.00 kip-ft

McB = -18 - 4^c - 2dB = -18 - 4(-5.167) - 2( + 1.333) =

15 k

A* 4" H 1.5k/ft
1 1
dt i
******* * r-i—T
12'

Constant /

Relative stiffness w<24-(3) //12X24-(D 1

Fixed end moments + 15 -15 + 18 -18

1

15 k
l.Sk/ft

iiC^
i ^*ll*llli
23
End reactions due
+
to applied loads
+ 7.5 + 7.5 9 +9
End reactions due
to end moments -1.5 + 1.5 + 1.917 -1.917

Total end reaction +6 +9 + 10.917 + 7.083

i2c= 7.083 k

7.083

Shear diagram
16.72

11

Elastic curve

Fig. 14-9

Note that the joint conditions Mba + Mbc = and Mcb = are
satisfied. The reactions, shear and bending-moment diagrams, and the
elasticcurve are shown in Fig. 14-9.
Example 14-4. Analyze the continuous beam shown in Fig. 14-10
owing to the effect of a ]4-m. settlement at support B by the slope-
deflection method. Calculate the reactions and draw the shear and
bending-moment diagrams. Sketch the elastic curve.
 THE SLOPE-DEFLECTION METHOD 329
A-

/-4/c ^-34
ec 40*

£- 30.000 k/iiL2 /c - 1000 in.4

Values of A in rad. + 0.69444x10-3 -1.04167x10-3

ValuMofi^imk-ft 27,778 31.250

&
55.412
^e-
52.962
1.3240
End reactiona due to end moments + 1.8062 — 1.8062 1.3240
i?4-+l.S062k i?B- -3.1302k i2c- +1.3240k
+ 1.8062

- 1.3240
Shear diagram

52.962

-yL--'^
Elastic curve

Fig. 14-10

SOLUTION. The general slope-deflection equations

2EI
Mab = Mfab + L {-2dA - Ob + SR)

2EI
Mba = Mfba + {-263 - Ba + 3i?)

will be used. In this example there are no fixed-end moments due to

applied loadings, but there are R values for spans AB and BC. Inas-
much as the absolute values of R in radians are known, the absolute
values of 2EI L in kip-feet must be used.

Span AB'. R = + -:^?P.— = -hO.69444 X IQ-' radians

/20 m.
2EI ^ (2) (30,000) (4,000)
27,778 kip-ft
L (U4)(60)
0.500 in.
Span BC: R = - -1.04167 X 10-^ radians
480 in.

2EI ^ (2) (30 ,000) (3,000)

= 31,250 kip-ft
L (144)(40)
330 ELEMENTARY THEORY OF STRUCTURES
Slope-defledion equations

Mab = + 27,778[-2^^ - ^b + 3(+0,69444 X \i^\

= -27,778^^ + 57.862
Mba = + 27,778[-2(9b - Ba -^ 3(+0.6d444 X 10-*)]
= -55,556^5 + 57,862
Mbc = + 31/250[-2^£ - ^c + 3(- 1,04167 X 10-")]
= -62,500^B - 31,250^c - 97.656
McB = + 31,250[-2^c - i9b + 3(- 1.04167 X 10-»)]
= -62,50a^c - 31,250^£ - 97.656

Joint conditions

Joint B:

Mba + Mbc = -118,056^5 - 31,250^c = +39.794 (a

Joint C: Mcb = -31.250^£ - 62,500^c = +97.656 (b)

Subtracting twice (a) from (bj.

+204.862^5 = +18,068
Bb = +0.088196 X 10-" radians (c)

Substituting (c) in (b),

-31,250(+0.0866196 X 10-^) - 62,500^c = +97,656

Be = -1.60659 X 10-2 radians (d)

Computation of end m-oments

Mab = -27,778(+0.088196 X 10-^) + 57.862 = +55.412 kipHft

Mba = -55,556(+0.088196 X lO"*) + 57.862 = +52,962 Mp-ft
M^c = -62,500(+0.088196 X 10-^) - 3 1,250(- 1.60659 X lO"*)
- <^7.656
= -52.962 kip-ft
Mcb = -62,500(- 1.60659 X 10-«) - 3 1,250 (+0.088196 X lO-^J
— 97,056
=
The reactions, shear and bending-moment diagrams, and the daffllic

cun^e are shown in Fig, 14-10.

A quick check of the above solution may be made by applyiiog the
moment-area method to the elastic curve Aown in Fig, 14-lOj

BB' = deflection of B' from the taaagent at A

= t(55.412)(6%)(40) - (52.962) (^9i)<20)]
(SO.O^kIoOO)
= 0.500 in. (check)
 THE SLOPE-DEFLECnOX METHOD 331

Deflection of C from the tangent at A

= [(55.412)(e^,)(80) - (52.962)(e%)(60)l
(30,0^k!,000)
- (30,oia)to00)
K52.962)(4%)(s^l
= {check)

14-4. Application of the Slope -deflection Method to the Analysis of

Statically Indeterminate Rigid Frames. Ca-^^A 1 Without Joint Move-
merits. The modified slope-deflection equations

Mab = Mfab + Kab(-29a - Os)

Mba = Mfba -h Kab^-^Bb - Ba)

may be used to analyze statically indeterminate rigid frames wherein all

joints remain fixed in location during deformation. Again the axial
deformation in the members due to direct stress is neglected in applying

i 1 -::: B ^ -5€J>5
i

Fig. 14-11 Fig. 14-12

the conditions of consistent deformation to the analysis; but direct

stresses, together with shears and bending moments, must be considered
in the design of sections. The conditions for consistent deformation are
those of the rigidity of joints, or the angle between any two tangents
to the elastic cur\'es meeting at one joint must remain the same as that
in the original undeformed structure. In the slope-deflection method
the rotation at each joint is considered as the unknown, while the con-
unknown is one of statics; i.e., the sum of the
dition corresponding to this
moments, as expressed by the slope-deflection equations, acting on the
joint is equal to zero. Thus there are always as many conditions of
statics as unknown rotations. After the latter are solved, aU end
moments may be found from the slope-deflection equations. With all
end moments known, the direct stresses, shears, and bending moments
in all members are foimd by apphing the principles of statics to the
indi\'idual members.
Example 14-5. Analyze the rigid frame shown in Fig. 14:-11 by the
slope-deflection method. Find the direct stresses, shears, and bending
moments in all members. Sketch the deformed structure.
332 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. Joints A, B, C, and D must all remain fixed in location;
thus R is zero for allmembers.
Relative stiffness and fixed-end moments (Fig. 14-12)

AB 9
t(»)
BC <») 12
a* = *)

BD (^" = t),eo, 10

3(20)2
Span AB: Mfab ^ ^ + 100 kip-ft
12
Mfba - 100 kip-ft

Span BC: M FBC +

3(15)2
12
= +56.25 kip-ft

Mfcb -56.25 kip-ft

Slope-deflection equations. Oa and do are known to be zero.

Mab = +100 + 9(-2^^ - Bb) = +100 - ^Bb

= +100 - 9(- 0.27902) = +102.51 kip-ft
Mba = -100 + 9(-2^B - Ba) = -100 - ISBb
= -100 - 18(- 0.27902) = -94.98 kip-ft
Mbc = +56.25 + 12(-26>B - Be) = +56.25 - 24^^ - l2Bc
= +56.25 - 24(-0.27902) - 12(-2.20425)
= +89.40 kip-ft
McB = -56.25 + 12{-2Bc - Bb) = -56.25 - 24^c - 12Bb
= -56.25 - 24(-2.20425) - 12(-0.27902) =
Mbd = + 10(-2^B - Bd) = -20(- 0.27902) = +5.58 kip-ft
Mdb = + 10(-2^i, - Bb) = -IOBb = -10(- 0.27902)
= +2.79 kip-ft
Joint conditions

Mba + Mbc + Mbd = -62(9,j - 12Bc = +43.75 (a)

McB = -12Bb - 24(9c = +56.25 (b)

Subtracting twice (a) from (b),

+ 112(95 = -31.25
Bb = -0.27902 (c)

Substituting (c) in (b),

-12(-0.27902) 24i9c = +56.25

Be = -2.20425 (d)
 THE SLOPE-DEFLECTIOX METHOD 333
58.084
3k ft ^ B 3k ft c <^5.58
0.698,
* I I i 1^
0.698
.dnnzjzj. X3_2 B
I

-r30 Cf^ 22.500 - 22.500*2) 0.698

102.5 ^ -
+
'+30.37
+ 30.376
0.37 - 0.376
+ 29.624
r^.9s 89.40^1+
% 5.960
28.460 +16.54C
5.960r^0

+ 28j460 (+135.00)
(* 153.79)
9.S75' 5.513',

I
0.69S

^2.79
b
10.125'
540 58.084
(- 146.26;
- 29.624
51.2S

102.51 94.98

3k/ft
,698 k I i i * * ^ *
I
i t i i ^~~) r

102..51k-ftG|'^
t
30.376 k 16.540 k

Check.
^.H- 0.698 k 1H=0
i\-=o
|02.79k-ft ZM=0
58.084 k
(a) Free-body diagram of whole frame

51.28 45.60

102.51

(b) Bending moment diagram

H B ^
Check,
By moment-area method.

5b=^8.37
(c) Elastic curve
^--66.1

Fig. 14-14

The free-body, shear, and bending-moment diagrams of the indi^'idllal

members are shown in Fig. 14-13. The free-bod\' diagram, bending-
moment diagram, and the elastic cun'e of the whole frame are shown in
Fig. 14-14.
Example 14-6. Analyze the rigid frame shown in Fig. 14-15 by the
slope-deflection method. Draw shear and bending-moment diagrams.
Sketch the deformed structure.
334 ELEMENTARY THEORY OF STRUCTURES
SOLUTION. Joints A and D are fixed. Joints B and C cannot move
in the vertical direction, but each may shift the same distance in the
horizontal direction. In the present example, however, on account
of the symmetry, both in the properties of the frame and in the applied

300
^ 300

D ®

A o Do
Relative stifftiess & FEM

Fig. 14-15 Fig. 14-16

loading, joints B and C will not have any horizontal displacement. Thus
R is equal to zero for all members.

Relative stiffness and fixed-end moments (Fig. 14-16)

AB, CD (2lc ^ Ic\

(30)
\ 20 10/

/5/c ^ Ic\
EC 6/
(30)
\30

Mfbc = + ^^ = +300 kip-ft

Mfcb = -300 kip-ft

Slope-deflection equations. 6a and dn are known to be zero. By

symmetry, dc = —Ob.
60 60
163.64 4k/ft .163.64
l,2-273
I ; I ; |

12^273 12.273
163.64 60 jl^ 163.64

12.273 12.273

to81.82 81.82
60 60

163.64 163.64

Fig. 14-17
 THE SLOPE-DEFLECTION METHOD 335

Mab = + 3(-2^^ - Ob) = -SOb = -3(+27.273) -81.82 kip-ft

Mba = + 3(-2dB - Ba) = -^Bb = -6(+27.273) -1G3.G4 kip-ft
Mbc = +300 + b{-2BB - Be) = +300 - 5(9^
= +300 - 5(+27.273) = +163.64 kip-ft
McB = -300 + 5(-2(9c - Bb) = -300 + 5Bb
= -300 + 5(+27.273) = -163.64 kip-ft
Mod = + 3(-2^c - Bd) = +6(9zj
= +6(+27.273) = +163.64 kip-ft
Mdc = + 3(-2^D - Be) = +SBb
= -l-3(+27.273) = +81.82 kip-ft
Joint condition

Joint B: Mba + Mbc =

-UBb + 300 =
e^ = -f27.273

The free-body, shear, and bending-moment diagrams of the individual

members are shown in Fig. 14-17. The free-body diagram, bending-
moment diagram, and the elastic curve of the whole frame are shown in
Fig. 14-18.
286.36

4k/ft
i I i

Check.
2H=0
2y=o
SM=0
12.273 k

Of
60 k
Od'O

(a) Free-body diagram of whole frame (6) Bending moment diagram (c) Elastic curve

Fig. 14-18

14-5. Application of the Slope -deflection Method to the Analysis of

Statically Indeterminate Rigid Frames. Case 2 With Joint Movements.
When loads are applied to statically indeterminate rigid frames, there
are cases in which j oints move unknown distances, although usually
some
in known Take, for instance, the rigid frame of Fig. 14-19a;
directions.
joints D, E, and F are fixed; but joints A, B, and C may all move equal
distances in the horizontal direction. This horizontal movement is
generally called sidesway. Assume that the amount of sidesway is A
to the right; then

Rad — Rbe = Yj- RcF = 77"

ih 112 lis
336 ELEMENTARY THEORY OF STRUCTURES
Thus the slope-deflection equations (14-9)

OW J
MAB = Mfab + -^ (-2(9a - Bb^ 3R)
OW J (14-9)
MBA = MfBA + -j^ {-203 - Oa-^ SR)

must be used for members AD, BE, and CF. It is necessary, then, to
seek another condition to cope with the unknown amount of sidesway A.

r)Mj,j, <^Mbe Io^c

Hf
T Jk
<

"Omfc
E He

Hd
o M, O.M,EB

Oi^^DA Omda
Vd
(a) (b) Free-body diagrams of AD, BE and CF
Fig. 14-19

By applying the equations of statics to the free bodies of members AD,

BE, and CF (Fig. 14-196),

rj Pihi
^^ = + ^7 +
Mad + Mda
,
.

—m—
Mbe + Meb
He =
H2
Hf =
McF + Mfc
H,

Applying the equation of statics 2i/ = to the whole frame shown in

Fig. 14-19a,

-{-Pi — Hd ~ He — Hf = (14-11)

Equation (14-11) is generally called the shear equation or the 6en^ equa-
tion. It furnishes the extra condition corresponding to the additional
unknown A.
The application of the slope-deflection method to the analysis of
single-span, one-story, statically indeterminate rigid frames in which
 THE SLOPE-DEFLECTION METHOD 337

some joints are displaced during deformation will be illustrated by the

following examples.
Example 14-7. Analyze the rigid frame shown in Fig. 14-20a by the
slope-deflection method. Draw shear and bending-moment diagrams.
Sketch the deformed structure.

3k/ft
I ; I ; ; ; i © -144
C B
B'
+ 144

RCD-
10 k
® ®
^AB

%
(6) Relative stiffness & FEM (c) Sidesway

Fig. 14-20

SOLUTION. Relative stiffness and fixed-end moments (Fig. 14-20&)

AB /3^_
Vi8
Mey (12)

BC /47c ^/c\ (12)

\24 6^

CD

(10)(12)(6)2
Span AB Mfab = + (18)2
= +13.33 kip-ft

M FBA MgM! = _26.67 kip-ft

3(24)2
Span BC: M FBC= +
12
= -1-144 kip-ft

Mfcb = — 144 kip-ft

Relative values of R (Fig. 14-20c)

Rab A (36) 2R

Rbc

RCD ^(36) ZR

If Rab = 2R, then Rcd = SR.

338 ELEMENTARY THEORY OF STRUCTURES
Slope-deflection equations. The slope-deflection equations

Mab = Mfab -\-^{-2dA - eB + 3R)

(14-9)
Mba = Mfba + -^ {-^Bb - Ba^ SR)

may be modified to take the following form:

Mab = Mfab + Kab{-2Ba - Bb + Rr.d

(14-12)
Mba = Mfba + Kab{-2Bb - Ba + Rr.x)

in which Kab is the relative stiffness and i^rei is the relative size of the
angle between the original axis and the straight line joining the deflected
ends. Note that 2EI /L is replaced by Kab and 3/^ is replaced by Rr^\.
These changes do not affect the values of the end moments at all. Equa-
tions (14-12) will be used in the present example.

Mab = +13.33 + 2{-2Ba - Bb ^ 2R)

= +13.33 - 4(9a - 2Bb + ^R
Mba = -26.67 + 2{-2Bb - Ba + 2R)
= -26.67 - ^Bb - 2Ba + 4J?
Mbc = +144 + 2{-2Bb - Be) = +144 - 4(9^ - 2Bc
McB = +144 + 2(-2(9c - Bb) = -144 - 46>c - 2Bb
Mod = + 3(-2(9c - 6*^ + SR) = -QBc - SBd + 9R
Mdc = + Si-2BD - 6*0 + ^R) = -QBd - SBc + QR
Joint conditions

Mab = -4(9a - 2Bb + 4.R = -13.33 (a)

Mba + Mbc = -2Ba - SBb - 2Bc + 4/? = -117.33 (b)

McB + Mod =0 -2Bb - lOBc - SBd -\- 9R = +144 (c)

Mdc = -SBc - QBd -\-9R = (d)

Shear condition (Fig. 14-21). Substituting

„ = 10 , Mab + Mba
liA -FT ~r
3 '
18

and -H^D = Mod + Mdc

12

in the shear equation +10 — Ha — Hd = 0,

in _ Mabj^MbA _ McD + Mdc _

+ 1U
I

-(^12
^
(
i
-u
-h
18 12
^
y
+360 - 120 - 2{Mab + Mba) - 3(Mcd + Mdc) =
12^A + 12Bb + 27^c + 27^D - 70i^ = -266.67 (e)
 ?

THE SLOPE-DEFLECnOX ICETHOD 339

Solving the five siiniiltaneous equations (a) to (e),

Oa = +5.369
$B = +24.174
Be = -15.152
$1, = +28.761
R= +14.123
In deriving the shear equation (e). Mab and Mdc could have been
called zero; the equation then obtained will be actually the combination
of Eqs. (a), (d), and (e) shown above.

3k"S ry^cD
t t 4 I B

10 k 10

OmJ
'A

Fig. 14-21
33L745
,- --,-5 ?k fi c
1 : ^f

t^isL-
lOk-
* 3o-"ir ^.^^

"\
11 .14- ^ \
/•
-^\-^ 355

iTja
13L74
Fig. 14-22

Computation of end moments

Mab = +13.33 - 4(+5.369) - 2^-24.174) + 4(+14.123) =

Mba = -26.67 - 4(+24.174) - 2(+5.369) + 4(+14.123)
= -77.61 kip-ft
Mbc = +144 - 4(+24.174) - 2(- 15.152) = +77.61 kip-ft
McB = -144 - 4(- 15.152) - 2(+24.174) = -131.74 kip-ft
McD = -61,-15.152) - 3(+28.761) + 9(+ 14.123)
= +131.74 kip-ft
Mdc = -6(+28.761) - 3(-l5.152) + 9(+14.123) =
340 ELEMENTARY THEORY OF STRUCTURES
The free-body, shear, and bending-moment diagrams of the individual
members are shown in Fig. 14-22. The free-body diagram, bending-
moment diagram, and elastic curve of the whole frame are shown in
Fig. 14-23.
112.17

3k/ft
i I I I I ^

10 k 77.61

131.74

10.978 k
v
0.979 k
38.255 k A
Check. Check. By moment-area method.
t
2//-0
33.745 k
SV-0 <t4- +32.21^ Bd' + 172.57 ^
2Af-0
Ob- + 145.04 ^
k-ft2
A- 1016.8^
^C- -90.91^

(o) Free-body diagram of whole frame (b) Bending moment diagram (c) Elastic cxirve

Fig. 14-23

Example 14-8. Analyze the rigid frame shown in Fig. 14-24a by the
slope-deflection method. Draw shear and bending-moment diagrams.
Sketch the deformed structure.

3k/ft
* i i i 4 i
+
144
C B 1b' y
B 4/c C 144
i£>

ZIc N
...
10 k
3/, (D
C5
D ^ —J-
rr

A
24'

(a) (6) Relative stiffness & FEM (c) Sidesway

Fig. 14-24

SOLUTION. Relative stiffness and fixed-end moments (Fig. 14-246)

AB 6* =*)<-) 2

BC = '-) 3
Ci' '/)

CD - <-' 3
e// ';)

Span AB: M^ab = + ^"^^1^,*^^' = +13.33 kip-ft

 :

THE SLOPE-DEFLECTION METHOD 341

10(6)(12)2
M FBA —
(18)^
-26.67 kip-ft

Span BC: Mfbc = + ^^' = +144 kip-ft

Mfcb = — 144 kip-ft

Relative values of R (Fig. 14-24c)

Rab (^^) 2R
rs

Rbc

RcD ^''^ Sit:

T2
If Rab = 2R, then Rcd = SR.
3k/ft
in + ^Mcn
i i i

10k lOk-

'Hr ,^_^ij Mcd+Mdc

^M:DC 'M:x>c
»}» < Ha

Fig. 14-25

Slope-deflection equations. The slope-deflection equations

MaB = MfaB + KAB(-2dA - eB-{- Rrel)

Mba = Mfba + KAB{-2eB - Ba^ i^rel)

will be used. 6A and ^d are known to be zero.

Mab = +13.33 + 2(-2^A - Ob + 2R) = +13.33 - 2(9^ + 4:R

Mba = -26.67 + 2(-2(9b - 0^ + 2i?) = -26.67 - ^Ob + 4/?
Mbc = +144 + 2(-2^^ - Sc) = +144 - Ub - 2dc
McB = -144 H- 2(-2^c - Bb) = -144 - 4(9c - 2(9^
McD = + 3(-2^c - Od + SR) = -Qdc + 9R
Mdc = + 3(-2(9d - ^c + 3i^) = -3(9c + 9R
Two joint conditions and one shear condition (Fig. 14-25)

Joint B: Mba + Mbc = -SOb - 200 + 4/^ = 117.33 (a)

Joint C: Mcb + McD = -2(9b - lOdc + 9/^ = +144 (b)

Shear

+ 10 H. Hd = 120J 270c + 707? = +266.67 (c)

342 ELEMENTARY THEORY OF STRUCTURES
Solving the three simultaneous equations (a) to (c),

03 = +19.260
Be = -18.154
R = +0.109
Computation of end moments
Mab = +13.33 - 2( + 19.260) + 4(+0.109) = -24.75 kip-ft
Mba = -26.67 - 4( + 19.260) +
-103.27 kip-ft 4( + 0.109) =
Mbc = +144 - 4( + 19.260) - 2(- 18.154) = +103.27 kip-ft
McB = -144 - 4(- 18.154) - 2( + 19.260) = -109.90 kip-ft
McD = -6(- 18.154) + 9(+0.109) = +109.90 kip-ft
Mdc = -3(- 18.154) + 9(+0.109) = +55.44 kip-ft
35.724 36.276
q103;27 3k/ft 109.90
13.779
i f I I 3=3=^4
103.2
q| + 36.000
0.276
+ 36.000 f-^
+
13.779
0.2761 109.90
++ + 35.724 + 36.276
10
+ 35.724
13.779 e

00 t>
'^55.44'
+ I
36.276

'024.75
35.724 36.276

103.27
109.90

Fig. 14-26

109.43

3k/ft
t t t I t I f Cllc

10k
13.779 k

055.44k-ft
3.779 k
36.276 k

024.75k-ft
35.724 k Check. Check. By moment-area method,
ZH-0
IV- ^B- +115.56^' ^-7-85^
SM-0
^C-- 108.93-^
(a) Free-body diagram (6) Bending moment diagram (c) Elastic curve

Fig. 14-27

The free-body, shear, and bending-moment diagrams of the individual

members are shown in Fig. 14-26. The free-body diagram, bending-
moment diagram, and elastic curve of the whole frame are shown in
Fig. 14-27.
 THE SLOPE-DEFLECTION METHOD 343
^
\' PROBLEMS
14-1 to 14-22. Analyze the statically indeterminate beams or rigid frames shown
by the slope-deflect ion method. Draw shear and bending-moment diagrams. Sketch
the deformed structure.

30 k 10k 30k
12' 1.6' J. 12^ .L 8'

>* * * i
2k/ft
^
±=1
i
,

* * * r.
j yi
1
l 4 \

^1
I

A \
2k/ft
\ \ \
i
\ t =r
20'
20'

Constant / Constant I

Fig. P14-1 Fig. P14-2

60 k 60k
12' 28' 12' I 28'
I

t=t
"^X
*
* *
3k/ft

37.
3/e B^ 2^c 4^
A-^ I
i-
r ^ I
3/^
3k/ft
I I w
D^f^
I I
2lc
X3c
'^
40' 20' 40' 20'

Fig. P14-3 Fig. P14-4

60k 60 k
12' 28' 28'
I .

Jit
I

3k/ft it^
\ 3k/ft
r^
I

Ai t IM I t t I \ \\ a;: *
5^^^=?*^ '/.•
B^ jt
2/c 3/c ' fi

20' 40' 20'

Fig. P14-5 Fig. P14-6

B B
A^
3/c 'fJl/2" 2/ 3/c ;^Jy2" 2/,
^
40'
^ 20' 40'

J5:= 3000 k/m.2 /c = 10,000 in.'* £ = 3000 k/in.2 /j. = 10,000 in.-*

Fig. P14-7 Fig. P14-8

B
ITe AJl/2" 2/e
rC
3/c
40' '^-^ 20' 40' 20'

£ - 3000 k/in.2 j^ =, iq.OOO in.4 E « 3000 k/in.2 j^ ^ 10,000 in.'*

Fig. P14-9 Fig. P14-10

60 k
12' I 28'

i 3k/ft

^yf 3Ic bA- 2/c 'Ac 3/c A^

40' 20' 4£ J

Fig. P14-11
60k
12' I
28'

\ I
i
i i i I I I
3k/ft
t » I I » » *
'^D

40' 20* 40*

Fig. P14-12
 —

344 ELEMENTARY THEORY OF STRUCTURES

3k/ft 3k/ft
. / ; 1 i t + ',

3Ic CV 3/c B 3/c C 3/c

21, 2/c "^

2V 20'
-
20'
a- 20- ,

Fig. P14-13 Fig. P14-14

18'
30 k
18'
18' r
1
.8-

2k/ft '

2k/ft
c ^ i i * C
B ; 1 1 4 B *

5/c 5/c

3/c 3/c 04 3/c 3/c CM

D
- ^L
7f It-
36'
36' ' J
"l

Fig. P14-15 Fig. P14-16

30k 30k
18' I, 18' ,^ 18' 1
^
18'

* *
T 2k/ft T
I I I V I
i
I
2 k/ft
^ ~t
i

5/c B 5/c

3Jc 3/c 3/c

E
h-
F7.

36' 36'

Fig. P14-17 Fig. P14-18

w i i ^
1.5k/ft
i i

5/c
i w i
;-

Dk
3/.
3L

36'

Fig. P14-19 Fig. P14-20

1.5k/ft 1.5k/ft

-9k

Fig. P14-21 Fig. P14-22

 CHAPTER 15

THE MOMENT-DISTRIBUTION METHOD

15-1. General Description of the Moment-distribution Method. The

moment-distribution method may be used to analyze all types of statically
indeterminate beams or rigid frames. Essentially it consists in solving
the simultaneous equations in the slope-deflection method by successive
approximations. In developing the method, it will be helpful to consider
the following problem: If a clockwise moment of Ma kip-ft is applied
at the simple support of a straight member of constant cross section
simply supported at one end and fixed
at the other end, find the rotation Oa
at the simple support and the moment
Mb Sit the fixed end (Fig. 15-1). The £f>^
method of consistent deformation will
be used. The condition of geometry
required is, in this case,

di or Obi = dB2 (15-1)

By the conjugate-beam method, Mmrni^m^

MaL

1
— MaL Bb2 — MbL (15-2) ib)

QEI SEI ^A2

Mb
Substituting Eq. (15-2) in Eq. (15-1), ^ MbL
2

MaL MbL M.
QEI SEI 'ama (15-3)
^""^^^^^i^miL^^
(c)

Also, Fig. 15-1

— —
MaL MbL MaL (y2MA)L
Ba &A1 0A2
SEI QEI SEI QEI
L
AEI
Ma (15-4)

Solving for Ma in Eq. (15-4),

4.EI
Ma = (15-5)

345
346 ELEMENTARY THEORY OF STRUCTURES
Thus, for a span AB simply supported at A and fixed at B, a
which is

clockwise rotation of Oa may

be effected by applying a clockwise moment
of Ma = {4:EI/L)dA at A, and this in turn induces a clockwise moment of
]\jg = 3^ Ma on the member at B, The expression 4:EI/L is usually
called the stiffness factor, which is defined as the moment required to be
applied at A to cause a rotation of 1 radian at A of a span AB simply
supported at A andB; the number +3^ is the carry-over factor,
fixed at
which is the ratio of the moment induced at B to the moment applied
at A. Note that the same sign convention is used in the moment-
distribution method as in the slope-deflection method.

24 k

A^ i ; i i i
3k/ft
w ; i i i 1 1 i i

ii
zd
3/. 27.

20' 20'

(a)

100
e ^ © £)60
100 60
I

L js^i?_r Q I

(b)
Locking moment Locking moment

^°°^^100
^C^^^ 60
CO-^
^ —^40" "locking moment"
(c)

Fig. 15-2

Consider the continuous beam ABC shown in Fig. 15-2a. If the

joints A, B, and C are to be restrained against rotation, the moments as
shown in Fig. 15-2 must be appHed, and these are, in fact, the fixed-end
moments on spans AB and BC. The restraining moments required to
hold the joints against rotation are (1) 100 kip-ft counterclockwise at A,
(2) 100 - 60 = 40 kip-ft clockwise at B, (3) 60 kip-ft clockwise at C.
These restraining moments are sometimes called the ''locking" moments
to 'Mock" the joints against rotation. Note that the joint B shown as a
free body in Fig. 15-2c is in equiUbrium under the action of the fixed-end
moments, which are opposite in direction to those acting on the members,
and the locking moment. The procedure may be described as follows:
First lock all three joints. Then release joint B only. Joint B, now
under the action of 40 kip-ft counterclockwise, will rotate a certain
amount counterclockwise direction, which will in turn induce coun-
in the
terclockwise moments at B to act on BA and BC in amounts proportional
 THE MOMENT-DISTRIBUTION METHOD 347

to the stiffness factors of each, with a sum

40 kip-ft. The relative stiff-
of

ness factors of BA and BC are 3 Thus 3/(3 + 2) = 0.600 times

and 2.

40 kip-ft, or 24 kip-ft counterclockwise, will act on BA, and

= 0.400 times 40 kip-ft

(2 + 3)

or 16 kip-ft counterclockwise, will act on BC. The numbers 0.600 and

0.400 are usually called the distribution factors. Now lock joint B in
its new position, and release joint C, which will rotate a certain amount
in the counterclockwise direction. This rotation must be such as to
induce a counterclockwise moment of 60 kip-ft to act on CB at C. Joint
A is a fixed support; so it need not be released at all. Thus the first
cycle of the moment distribution has been completed (see the adjoining
moment-distribution table). To summarize, all joints are first locked
by locking moments +100, —100, +60, —60 acting on all members,
then joints B
and C are released in succession, and the ''balancing'^
moments are 0, +24, +16, and +60 (joint A is a fixed support).
When a balancing moment of +24 kip-ft is placed at B of span AB,
one-half of this amount, or +12 kip-ft, is induced Sii A on AB. In the
Moment-distribution Table

Joint A B C

Member AB BA BC CB

Distribution factors . . . 0.600 0.400 1.000

Cycle 1 FEM -hlOO -100 + 60 -60

Balance + 24 + 16 +60

Cycle 2 Carry-over + 12 +30 + 8

Balance - 18 -12 - 8

Cycle 3 Carry-over - 9 - 4 - 6
Balance -h 2.4 + 1.6 + 6

Cycle 4 Carry-over + 1.2 + 3 + 0.8

Balance - 1.8 - 1.2 - 0.8

Cycle 5 Carry-over - 0.9 - 0.4 - 0.6

Balance -h 0.24 + 0.16 + 0.6

and so on, to any desired degree of accuracy

Total end moments + 103.3 - 93.16 +93.16

(5 cycles)
348 ELEMENTARY THEORY OF STRUCTURES
same manner, one-half of the -hl6 kip-ft, or +8 kip-ft, acts at C on CB,
and one-half of the -hOO kip-ft, or +30 kip-ft, acts at B on BC. These
moments -f-12, 0, +30, +8 are called the ''carry-over" moments. They
are kept out during the first balancing and now are considered as new
locking moments to lock the joints in position after the first rotations.
Then joints B and C are released for the second time, and the second
rotations at B and C induce balancing moments as shown in the second
cycle and carry-over moments as shown in the first line of the third cycle.
The same process is repeated for as many cycles as desired to bring the
balancing or carry-over moments to very small magnitudes. Thus any
degree of accuracy can be obtained, and the work required decreases as
the required accuracy decreases. The final, or total, end moments are
obtained by adding all numbers in the respective columns.
Thus the moment-distribution method consists in successively locking
and releasing the joints; the first locking moments are the fixed-end

A 4k/ft
* I I * I I I I I I i V
A; ---,.. r -^ 18'

<i.
Constant /
B (a)

Constant £7 .
^ _,,,
B^r+T2
6£/A ,6EIR
= MpBA -+-^
\f nf
MfaB ,

-^-T- (6) Relative stiffness & FEM

Fig. 15-3 Fig. 15-4

moments due to the applied loading; after the first balancing, the succes-
sive locking moments are the carry-over moments which are induced to
act at the other ends of the respective spans by the balancing moments.
The reader is advised to read this article again and again while working
through the rest of this chapter.
15-2. Application of the Moment-distribution Method to the Analysis
of Statically Indeterminate Beams. As described in general in the pre-
ceding article, the moment-distribution method may be used to analyze
statically indeterminate beams due to any applied loading. This method
may also be used to analyze statically indeterminate beams due to the
yielding of supports. The physical concept involved is that the joints
are first locked against rotation and then displaced to conform with the
amount of yielding; the locking moments acting on the ends of each
member will be the fixed-end moments as derived in Eq. (14-8), which
are repeated here and shown in Fig. 15-3. Then the joints are released
or balanced; the carry-overs become the next unbalances, the joints are
again balanced, and so on. In other words, the fixed-end moments due
to the movement of one end relative to the other in a direction perpen-
 THE MOMENT-DISTRIBUTION METHOD 349

dicular to the original direction of the member are treated in exactly

the same manner as those due to the applied loadings.
Example 15-1. Analyze the beam shown in Fig. 15-4a by the moment-
distribution method. Draw shear and bending-moment diagrams.
Sketch the elastic curve.
SOLUTION. Relative stiffness and fixed-end moments (Fig. 15-46).
Since there is only one span (or member) in this beam, the stiffness of
this member can be regarded as unity.

Mfab = + ^^' = +108 kip-ft

Mfba = - 108 kip-ft

When the overhang BC is treated as a cantilever fixed at B, the fixed-end

moment at B is +4(6) V2 = +72 kip-ft because it acts counterclockwise
at B on BC. However, BC has no stiffness and it should not be treated
as a member.
Table 15-1. Moment Distribution

Joint A B

Member AB BA BC

K 1 1

Cycle DF 1.000

1 FEM + 108 -108 +72

Balance + 36

2 Carry-over + 18
Balance

Total + 126 - 72 +72

Check:
Change + 18 + 36
-H (change) - 18 - 9
Sum + 27
^rei = sum/ — K . . .
- 27
Check

For explanation of check, see Art. 15-3.

In the moment-distribution table (see Table 15-1) the distribution

factor at end A of AB is shown by a dash because at the fixed end A
there is no need of releasing or balancing, while the distribution factor
at B of BA is equal to 1.000 because member BA is the only member with
350 ELEMENTARY THEORY OF STRUCTURES
stiffness entering jointB. In the body of the moment-distribution table,
a dash isused for zero. In the first cycle the ''unbalance" at joint B
is —108 + 72 = —36 kip-ft. All of the balancing moment, +36 kip-ft,
is placed under BA. The moment distribution comes to an automatic
stop at the end of the second cycle. The total moments are +126, —72,
and +72 as shown. Whenever a stage of moment distribution comes
The checking
to an automatic stop, the results are theoretically exact.
procedure will be explained in the next article.

The reactions, shear and bending-moment diagrams, and the elastic

curve are shown in Example 14-1 and will not be repeated here.

^J^
h
—— 15 k

=±-I bJl
i
i

+
1.5k/ft

v'
'
^^^c
^
Constant /
(o)

(3) -15 ^} -18

^ ^+ 15 Bt^ + IS ^C
(6) Relative stiffness & FEM
Fig. 15-5

Example Analyze the continuous beam shown in Fig. 15-5a by

15-2.
the moment-distribution method. Draw shear and bending-moment
diagrams. Sketch the elastic curve.
SOLUTION. Relative stiffness and fixed-end moments (Fig. 15-56)

AB i
(^^)

BC A (2^)

M,As= +^^ = +15 kip-ft

Mfba = — 15 kip-ft

Mfbc = + —{2"^' = +18 kip-ft

Mfcb = — 18 kip-ft

In the moment-distribution table (see Table 15-2) the distribution

factors are first computed. The DF (distribution factor) at joint A
(or C) is 1.000 on member AB (or CB) because there is only one member
entering the joint. The DF's at joint B at 3/(3 + 2) = 0.600 on member
=
BA and 2/(3 + 2) 0.400 on member BC. In cycle 1, the joints are
released and balanced. The ''unbalance" at A is +15.00; so thebalanc-
 THE MOMENT-DISTRIBUTION METHOD 351

ing moment is —15.00. The '^unbalance" at J5 is

-15.00 + 18.00 = +3.00

so the balancing moments are —(0.600) (3.00) = —1.80 and
-(0.400) (3.00) = -1.20 on BA and BC, respectively

The ''unbalance" at C is —18.00; so the balancing moment is +18.00.

The carry-overs as shown in the first line of the second cycle are +3^ times
the balancing moments placed at the far ends of the respective members
in the preceding Une. These carry-overs are the new unbalances, which
in turn are balanced in the same manner. The process is repeated, keep-
ing all figures to two decimal places. In this example it is observed
that the total end moments at all joints are not materially affected by
the third (and fourth) cycle of moment distribution. This suggests
that no further moment distribution is needed. By adding the moments

Table 15-2. Moment Distribution

Joint A B C

Member AB BA BC CB

K 3 3 2 2

Cycle DF 1.000 0.600 0.400 1.000

1 FEM + 15.00 -15.00 + 18.00 -18.00

Balance -15.00 - 1.80 - 1.20 + 18.00
2 Carry-over - 0.90 - 7.50 + 9.00 - 0.60
Balance + 0.90 - 0.90 - 0.60 + 0.60

3 Carry-over - 0.45 + 0.45 + 0.30 - 0.30

Balance + 0.45 - 0.45 - 0.30 + 0.30

4 Carry-over - 0.22 + 0.22 + 0.15 - 0.15

Balance + 0.22 - 0.22 - 0.15 + 0.15

Total -25.20 +25.20

Check:
Change -15.00 -10.20 + 7.20 + 18 00
-y2 (ch ange) + 5.10 + 7.50 - 9.00 - 3.60
Sum . . - 9.90 - 2.70 - 1.80 + 14 40
^rel = sum/ — K + 3.30 + 0.90 + 0.90 - 7.20
Ch eck

For explanation of check, see Art. 15-3.

352 ELEMENTARY THEORY OF STRUCTURES
in the respective columns the total end moments are obtained. These
total end moments check with those obtained by the slope-deflection
method in Example 14-2.
For reactions, shear and bending-moment diagrams, and the elastic
curve, see Example 14-2.
In carrjdng out the work in the moment-distribution table, it is advis-
able tofirst put down all the signs in any one line at the same time and

then the numerical values. This helps the computer to concentrate

on one operation at a time. Note also that at the time of each balancing
the sum of the balancing moments placed at the ends of members meeting
at one joint must be equal to the total unbalance at that joint. This
will ensure that the sum of the total end moments acting on all members
meeting at any one joint is zero.

15 k
4' I 4'
^
-*\ 1.5k/ft -

A -i .±^ ^ ^ ^ ^ ^

12'
^ ^ ^ ^ ^
Xc
Constant I
(a)

(3) -15 © 18
15

(6) Relative stiffness & FEM

Fig. 15-6

Example 15-3. Analyze the continuous beam shown in Fig. lo-6a

bj^ the moment-distribution method. Draw shear and bending-moment
diagrams. Sketch the elastic curve.
SOLUTION. Relative stiffness and fixed-end moments (Fig. 15-66)

AB (24)

BC (24)
12

M FAB _^15(8)
8
^ +15kip.ft
Mfba = —15 kip-ft
1.5(12)2
Mf BC ,

= +18 kip-ft
12
Mfcb = -18 kip-ft

In the moment-distribution table (see Table 15-3) the distribution

first computed.
factors are Since joint A is a fixed support, it can resist
any moment assigned to it and therefore need not be released. In such
 THE MOMENT-DISTRIBUTION METHOD 353

a case no distribution factor is applicable to A . The distribution factors

at joints B and C are determined as previously explained. Joint A
requires no balancing; so the balancing moments in all cycles are zero,
for which a dash is used in the table. The moment distribution
is carried

out as indicated. By keeping all figures to two decimal places, the table

Table 15-3. Moment Distribution

Joint A B C

Member AB BA BC CB

K 3 3 2 2

Cycle DF 0.600 0.400 1.000

1 FEM + 15.00 -15.00 + 18.00 -18.00

Balance - 1.80 - 1.20 + 18.00
2 Carry-over - 0.90 + 9.00 - 0.60
Balance - 5.40 - 3.60 + 0.60

3 Carry-over - 2.70 + 0.30 - 1.80

Balance - 0.18 - 0.12 + 1.80

4 Carry-over - 0.09 + 0.90 - 0.06

Balance - 0.54 - 0.36 + 0.06

5 Carry-over - 0.27 + 0.03 - 0.18

Balance - 0.02 - 0.01 + 0.18

6 Carry-over - 0.01 + 0.09

Balance - 0.05 - 0.04

7 Carry-over - 0.02 - 0.02

Balance + 0.02

8 Carry-over + 0.01
Balance - 0.01

Total + 11.01 -23.00 +23.00

Check:
Change - 3.99 - 8.00 + 5.00 + 18 00
-'A (ch ange) + 4.00 + 2.00 - 9.00 - 2.50
Sum . . + 0.01 - 6.00 - 4.00 + 15 50
d,ei = sum/-K + 2.00 + 2.00 - 7.75
Check Ch eck

For explanation of check, see Art. 15-3.

354 ELEMZXTART THEORY OF STRUCTURES
comes to an automatic stop at the end of the ei^th cycle. The total
end moments obtained by adding the moments in the respjective columns
check quite closely with those of Example 14-3 in which the slope-
deflection solution was used.
For reactions, shear and bending-moment diagrams, and the elastic
cur\-e, see Example 14-3.

^
A -2^^

7= 47.

/e- 1000 im.«

^^ — 5^<^

Example 15-4. Analyze the continuous beam shown in Fig. 15-7a

owing to the effect of a }4-ui. settlement at support B by the moment-
distribution method. Draw shear and bending-moment diagrams.
Sketch the elastic curve.
soLunox. Relaiite stiffness and fixed-end moments (Hg. 15-76)

^B I
I
\60 ^= fi (120)
15/
! 8

BC ^(120) I
9

QEIR
MfaB = y^FBA —
.^e<3«g^^(,o^)_,,^^^„
MrBc = ^IrcB = -i
j
—
^ _ 6^30.000)^3.000)
(-w) = -»^«"^'
The solution of this problem by the moment-distribution method is

shown in Table 15-4. For reactions, shear and bending-moment dia-

grams, and the elastic curve, see Elxample 14-4.
 THE MOMENT-DISTRIBUTION METHOD 355

Table 15-4. Moment Distribution

Joint A B C
1

Member AB BA BC CB

K 8 8 9 9

Cycle DF 0.4706 0.5294 1.0000

1 FEM -h57.86 -h57.86 -97.66 -97.66

Balance -1-18.73 -f21.07 -h97.66

2 Carry-over + 9.36 +48.83 + 10.54

Balance -22.98 -25.85 -10.54

3 Carry-over -11.49 - 5.27 -12.92

Balance + 2.48 + 2.79 + 12.92
4 Carry-over -f 1.24 + 6.46 + 1.40
Balance - 3.04 - 3.42 - 1.40

5 Carry-over - 1.52 - 0.70 - 1.71

Balance + 0.33 + 0.37 + 1.71

6 Carry-over -h 0.16 + 0.86 + 0.18

Balance - 0.40 - 0.46 - 0.18

7 Carry-over - 0.20 - 0.09 - 0.23

Balance + 0.04 + 0.05 + 0.23

8 Carry-over + 0.02 + 0.12 + 0.02

Balance - 0.06 - 0.06 - 0.02

9 Carry-over - 0.03 - 0.01 - 0.03

Balance + 0.01 + 0.03

10 Carry-over -h 0.02
Balance - 0.01 - 0.01

Total 4-55.40 -1-52.95 -52.95

Check:
Change - 2.46 - 4.91 -h44.71 +97.66
- yi (change) + 2.46 -f 1.23 -48.83 -22.36
Sum - 3.68 - 4.12 +75.30
Orel = sum/ — K -h 0.460 + 0.458 - 8.367
Check Ch 2Ck

For explanation of check, see Art. 15-3.

356 ELEMENTARY THEORY OF STRUCTURES
15-3. Check on Moment Distribution. The moment-distribution table
begins with relative stiffness and fixed-end moments (due to the apphed
loading and or settlement of supports) and concludes with the required
end moments. A check may easily be made to ensure that the correct
end moments have been obtained on the basis of the relative stiffness
and fixed-end moments at the beginning of the table. Of course, the first
ob\-ious check is to see whether or not the moments are balanced at
each interior joint and the moment is zero at each exterior simple (or
hinged) support; this check is on the conditions of statics. There is
also a check on the conditions of geometry. This may be made by find-
ing the absolute or relative values of the rotation at each joint.
The slope-deflection equations may be written as
2EI
^J^AB = ^ItFAB H
J
— — 2d A —
{ 6b)
^1^^)
2EI
3/sA = Mtfba H X— — 208 —
{ 6 a)

in which Mtfab and Mtfba represent the total fixed-end moments due
to the appHed loading and settlement of supports, or
^EIR
Mtfab — M fab H"
L
6EIR
Mtfba = Mfba "h

In Eqs. (15-6), the end moments are expressed in terms of the fixed-
end moments and the end rotations. Conversely, the end rotations
may be expressed in terms of the fixed-end moments and the final end
moments. Solving Eqs. (15-6) for 6 a and 63,

(Mab — Mtfab) — hzi-^^BA ~ Mtfba)

Ba
-ZEl/L
(15-7
{Mb A — Mtfba) — }^{Mab — Mtfab)
Be
-SEI/L
Equations (15-7) give the absolute values of the end rotations. If

only the relative values of Ba and 6b are desired, the expression ^^EI/L
in thedenominator can be replaced by iv^i. Thus.

{Mab — Mtfab) — }4{Mba - Mtfba)

(Ba) rel
(15-8)
{Mba — Mtfba) — }'2{Mab — Mtfab)
(^fl)rel — ^

In Eqs. (15-7) or (15-8), {Mab - Mtfab) and {Mba - Mtfba) are

the changes in moment from the "fixed-end moment" to the ''final end
moment." Thus the absolute or the relative value of the rotation
 THE MOMENT-DISTRIBUTION METHOD 357

at any one end of a member is equal to the change in the moment at the
near end minus one-half of the change in the moment at the far end and
then di\'ided by -3EI/L or -K^,. Or

Absolute value of ^^ear end

(change) ae.r end + (- K)(change)f>,ead n Q^

-
^^^'^^
-3EI/L
Relative value of dner^: end

(change) near end + "^( ) (change)far end

(15-10)
K.

Thus either the absolute or relative values of the rotation at the ends
of each member may be computed by the use of Eqs. (15-9) or (15-10).
A check on the condition of geometry is that the rotations at the ends of
all members meeting at one joint must be equal and that the rotation

at the fixed support must be zero. It is to be noted that, for the purpose
of this check, the use of the relative values of the rotations is recom-
mended. K occasions arise where the absolute values are desired, Eq.
(15-9) must be used.
For example, at the end of Table 15-3 for moment distribution in
Example 15-3, the check made by applying Eq. (15-10) is shown. The
relative values of the rotations at joints A, B, and C are 0, +2.00, +2.00,
and —7.75. The check on the two conditions of geometry are that the
rotation at A is zero and that the same value of the rotation at B is
obtained from spans BA and BC.
It must be again noted that the above check has nothing to do with
the correctness of the values of the relative stiffness and of the fixed-end
moments used at the beginning of the moment-distribution solution; i.e.,

if these are incorrect, the final answers wHl be correspondingly wrong

even though they meet the test of the check.
15-4. Stiffness Factor at the of a Member When the Far
Near End
End Is Hinged. The has been defined as the moment
stiffness factor
required to rotate the tangent to the elastic curve at the near end of a
member through 1 radian when the far end is fixed; for a member with
constant cross section, this stiffness factor is ^EI/L (Fig. 15-8a). Now,
if the far end is hinged instead of being fixed, the moment required to
rotate the tangent at the near end through 1 radian will be 3EI/L
instead of 4iEI/L (Fig. 15-86). This may be derived easily by the conju-
gate-beam method. From Fig. 15-85,

Ele^ = Rl = -^ or M, = ^-^e^=l(^^y_, (15-11)

Thus the stiffness factor at the near end when the far end is hinged is

3EI/L, or three-fourths of that when the far end is fixed. This stiffness
358 ELEMENTARY THEORY OF STRUCTURES
factor, 'iEI/L, will be called the modified stiffness factor in subsequent
discussions.
In Fig. 15-9 four members AE, BE, CE, and DE meet at a rigid joint
E. li A, B, and D are fixed and C is hinged, any unbalanced moment
at joint E will cause a certain amount of rotation at joint E, or the
unbalance on the ends of
will distribute itself into four parts to act
EA, EB, EC, and ED Kae, Kbe, H{Kce), and Kde.
in the ratio of
One-half of the balancing moments placed at E on members EA, EB, and
ED will then be carried to A, B, and Z>; but no carry-over to the hinge
4EI /^^
Ma
„
^ Mq" 1/2 Mj^

3EI
Ma'

Fig. 15-9

is necessary, because by modifying the stiffness factor of EC to three-

fourths of its usual value, provision is made for the fact that the moment
at C should always be zero. If, however, the "unbalance" at E is

distributed in the ratio of Kae, Kbe, Kce, and Kde, one-half of the balanc-
ing moment placed at E on EC must be carried over to C and joint C
must be balanced in every cycle of moment distribution. The latter
procedure has been followed in deaUng with the exterior simple support
examples of Art. 15-2. The alternate procedure involving modifica-
in all
tion of the stiffness of members with exterior simple or hinged supports
will now be shown.
 THE MOMENT-DISTRIBUTION METHOD 359

When a continuous beam or rigid frame has one or more exterior simple
or hinged supports, the moment-distribution procedure may be performed
by the regular method or the modified-stiffness method. In general, the
modified-stiffness method maintains accuracy to the same number of
significant figures with a shorter moment-distribution table. In apply-
ing the modified-stiffness method, the stiffness of the member with an
exterior simple or hinged support is changed to three-fourths of its usual

value. For such an exterior support, no distribution factor is shown

because only one release is needed when there is an initial fixed-end
moment. After the first balance, no further carry-overs are brought
to the simple support. To prevent a common mistake, it is suggested
that a vertical arrow be drawn immediately after the first cycle under
the exterior simple support in the moment-distribution table.
The distribution factors are determined from the modified stiffness of
members with hinged ends. Upon checking, however, the unmodified
values of the relative stiffness should be used in B.^x = sum/ — K.
Example 15-5. Using the modified-stiffness procedure, solve Example
15-2 by the moment-distribution method.

SOLUTION
T.A^LE 15-5. Moment Distribution

Joint A B c

Member AB BA BC CB

K 3 3 2 2

ModifierI Jf^ 2.25 2.25 1.5 1.5

Cycle DF 0.600 0.400

1 FEM + 15.00 -15.00 + 18.00 -18.00

Balance -15.00 - 1.80 - 1.20 + 18.00
2 Carry-over - 7.50 + 9.00
Balance , ,
- 0.90 - 0.60 [

Total -25.20 +25.20

Check:
Chan^re -15.00 -10.20 + 7.20 + 18 00
-M ( change) + 5.10 + 7.50 - 9.00 - 3.60
Sum. - q 90 - 2.70 - 1.80 + 14 40
^rci = sum/ — K . .
+ : 5.30 + 0.90 + 0.90 — r.20
Ch eck
360 ELEMENTARY THEORY OF STRUCTURES
Example 15-6. Using the modified-stiffness procedure, solve Example
15-3 by the moment-distribution method.

SOLUTION
Table 15-6. Moment Distribution

Joint A B c

Member AB BA BC CB

K 3 3 2 2

Modifiec \K 3 3 1.5 1.5

Cycle DF 0.667 0.333

1 FEM + 15.00 -15.00 + 18.00 -18.00

Balance - 2.00 - 1.00 + 18.00
2 Carry-over - 1.00 + 9.00
•

Balance - 6.00 - 3.00

3 Carry-over - 3.00
Balance ,

Total + 11.00 -23.00 +23.00

Check:
Change - 4.00 - 8.00 + 5.00 + 18.00
-}i (change) + 4.00 + 2.00 - 9.00 - 2.50
Sum - 6.00 - 4.00 + 15.50
^rei = sum/ — K. . . + 2.00 + 2.00 - 7.75
Check Ch ^ck

Example 15-7. Solve the moment distribution in Example 15-4 by the

modified-stiffness method.

3k/ft \ (^ -100 B ^ - 56.25 _

A", ; +100 o + 56.25 £

Do
Relative stiffness & FEM
Fig. 15-10 Fig. 15-11
 THE MOMENT-DISTRIBUTION METHOD 361

SOLUTION
Table 15-7. Moment Distribution

Joint A B c

Member. AB BA BC CB

K 8 8 9 9

Modified K 8 8 6.75 6.75

Cycle DF 0.5424 0.4576

1 FEM +57.86 +57.86 -97.66 -97.66

Balance +21.59 + 18.21 +97.66

2 Carry-over + 10.80 +48.83

Balance -26.48 -22.35

3 Carry-over -13.24
Balance ,

Total +55.42 +52.97 -52.97

Check:
Change - 2.44 - 4.89 +44.69 +97.66
-y2 (c hange) + 2.44 + 1.22 -48.83 -22.34
Sum . .
- 3.67 - 4.14 +75.32
Orel = sum/ — K. . . . + 0.459 + 0.460 - 8.369
Check Ch eck

15-5. Application of the Moment-distribution Method to the Analysis

of Statically Indeterminate Rigid Frames. Case 1 Without Joint Move-
ments. The application of the moment-distribution method to the
analysis of statically indeterminate rigid frames wherein no joint move-
ments or ''sidesway" is involved is very similar to that of beams as
discussed in the previous articles, except that in the case of rigid frames
there are frequently more than two members meeting in one joint. In
I such cases the unbalance at any joint is distributed to the ends of the
several members meeting at the joint in the ratio of their relative stiff-
nesses. There are a number of ways in which the work for the moment-
distribution procedure may be arranged, but a tabular form in which all
members meeting at the same joint are grouped together is used in this
text and is suggested as the most convenient form.
Example 15-8. Analyze the rigid frame shown in Fig. 15-10 by the
I moment-distribution method. Draw shear and bending-moment dia-
grams. Sketch the deformed structure.
 . .

362 ELEMENTARY THEORY OF STRUCTURES

SOLUTION. i2 = f or all members.

Relative stiffness and fixed-end moments (Fig. 15-11)

AB "»> 9
20

BC <™ 12
a^ = ^)

BD (f^-
'") 10
=
3(20)2
M FAB = + 12
= 4-lOOkip-ft

Mfba = -lOOkip-ft
3(15)2
Mfbc + = +56.25 kip-ft
12
Mfcb 56.25 kip-ft

Moment distribution (see Tables 15-8 and 15-9)

Table 15-8. Moment Distribution, Modified-stiffness Method

Joint A B C D

MembeiT AB BA BC BD CB DB
K 9 9 12 10 12 10

Modified K 9 9 9 10 9 10

Cycle DF 0.3214 0.3214 0.3572

1 FEM + 100.00 100.00 +56.25 -56.25

Balance + 14.06 + 14.06 + 15.63 +56.25

2 Carry-over + 7.03 +28.12 + 7.82

Balance — 9.04 - 9.04 -10.04

3 Carry-over - 4.52 -5.02

Balance

Total . + 102.51 - 94.98 +89.39 + 5.59 +2.80

Check:
Charige + 2.51 + 5.02 +33.14 + 5.59 +56.25 +2.80
-H (change) - 2.51 1.26 -28.12 -1.40 -16.57 -2.80
Sum + 3.76 + 5.02 + 4.19 +39.68
= sum/ — /C. 0.418 - 0.418 - 0.419 - 3 .307
Check Check Check
 .

THE MOMENT-DISTRIBUTION METHOD 363

Table 15-9. Moment Distribution, Regular Method

Joint A B C D

Member AB BA BC BD CB DB
K 9 9 12 10 12 10

Cycle DF 0.2903 0.3871 0.3226 1.000

1 FEM + 100.00 -100.00 +56.25 -56.25

Balance + 12.70 + 16.94 + 14.11 +56.25

2 Carry-over + 6.35 +28.12 + 8.47 +7.06

Balance - 8.16 -10.89 - 9.07 - 8.47

3 Carry-over - 4.08 - 4.24 - 5.44 -4.54

Balance + 1.23 + 1.64 + 1.37 + 5.44

4 Carry-over + 0.62 + 2.72 + 0.82 +0.68

Balance - 0.79 - 1.05 - 0.88 - 0.82

5 Carry-over - 0.40 - 0.41 - 0.52 -0.44

Balance + 0.12 + 0.16 + 0.13 + 0.52

6 Carry-over + 0.06 + 0.26 + 0.08 +0.06

Balance - 0.08 - 0.10 - 0.08 - 0.08

7 Carr3'-over - 0.04 - 0.04 - 0.05 -0.04

Balance + 0.01 + 0.02 + 0.01 + 0.05

8 Carry-over + 0.02 + 0.01

Balance - 0.01 - 0.01 - 0.01

Total + 102.51 - 94.97 +89.39 + 5.58 +2.78

Check:
Charige + 2.51 + 5.03 +33.14 + 5.58 +56.25 +2.78
-y2 (change) - 2.52 - 1.26 -28.12 - 1.39 -16.57 -2.79
Sum - 0.01 + 3.77 + 5.02 + 4.19 +39.68 -0.01
= sum/ — K. - 0.419 - 0.418 - 0.419 - 3.307
Check Check Check

For reactions, shear and bending-moment diagrams, and the elastic

curve, see Example 14-5.
Example 15-9. Analyze therigid frame shown in Fig. 15-12 by the
moment-distribution method. Draw shear and bending-moment dia-
grams. Sketch the deformed structure.

L
364 ELEMENTARY THEORY OF STRUCTURES
k/ft
B 300
300

Q a

Relative Btiffness & FEM

Fig. 15-12 Fig. 15-13

SOLUTION. R = OioT all members.

Relative stiffness and fixed-end moments (Fig. 15-13)

AB, CD (2L ^ Ic\

(30)
V20 10/

BC

4(30)2
M FBC + 12
+300 kip-ft

Mfcb = -300 kip-ft

Moment distribution (see Table 15-10)

For reactions, shear and bending-moment diagrams, and the elastic

curve, see Example 14-6.
15-6. Application of the Moment-distribution Method to the Analysis
Indeterminate Rigid Frames. Case 2 With Joint Movements.
of Statically
The procedure for applying the moment-distribution method to the
analysis of statically indeterminate rigid frames in which sidesway or
joint movements are involved consists in the following:
1. The joints are first held against sidesway. The fixed-end moments
caused by the apphed loadings are distributed, and a first set of balanced
end moments is obtained.
2. The unloaded frame is then assumed to have a certain amount of

sidesway which will cause a set of fixed-end moments. These fixed-end

moments are then distributed, and a second set of balanced end moments
is obtained.
3. resulting set of end moments may be obtained by adding the
The
and the product of a ratio and the second set, the ratio being
first set

determined by use of the shear condition, as will be explained.

 THE MOMENT-DISTRIBUTION METHOD 365

Table 15-10. Moment Distribution

Joint A B C D

Member AB BA BC CB CD DC

K 3 3 5 5 3 !
3

Cycle DF 0.375 0.625 1

0.625 375 !

1 FEM -300.0 -300.0

Balance -112.5 -187.5 1+187.5 -112.5

2 Carr>--over -56 2 ^93.8 - 93.8 -56 2

Balance - 35.2 - 58.6 [+ 58.6 + 35.2

3 Carn--over -17.6 - 29 3 - 29 3 -rl7.6

Balance - 11.0 - 18. 3 V 18 3 - 11.0

4 Carry-over - 5.5 ^ 9.2 9.2

-
— 5.5
Balance 3.4 5.8 + 5.8 3.4

5 Cany-over - 1.7 - 2.9 2.9 - 1.7

Balance 1.1 1-S — 1.8 - 1.1

6 Carry-over - 0.6 - OS' - •:> 9 - 0.6

Balance 0.3 ( — :3

7 Carr>--over - 0.1 — 0.3 - 0.3 - 0.1

Balance - 0.1 - 0.2 ;+ 0.2 - 0.1

8 Caro-over - 1 - 1

Balance - 1 — 1

Total -81.7 -163.6 —16:3 6 -163 6 -163 6 —81 7

Check: i
'

-81.7-163.6
i I i

Change i -136.4 '+136.4 '+163.6 +81.7

-H (change) i
+81.8 + 40.8 - 68 2 + 682 - 40.8 -81.8
Sum + 0.1 -122 8 -2(M 6 +204.6 +122.8 I
- 0.1
•„i =sum/-^....|
I

+ 40.93 - 40.92 - 40.92 - 40.93

Check Check Check Check

Take, for example, the rigid frame shown in Fig. 15-1 4a. It is required
to analyze this statically indeterminate rigid frame by the moment-
distribution method. The given frame shown in Fig. 15-14a is equivalent
to the sum of Fig. 15-146 and Fig. 15-14:C. In Fig. 15-146 the joints
B and C are held against sidesway by the fictitious support at C, the
horizontal reaction of which is denoted as //^. If the fictitious support
366 ELEMENTARY THEORY OF STRUCTURES
at C is removed, the force H'^ would act at joint C. In Fig. 15-1-k,
A" is the sides way caused by any arbitrary force H'^. If i/J; i^ equal to
kH'c, where k is the unknown ratio, the actual amount of the sidesway.
A' (Fig. 15-14a),must be equal to k\" Let Jr^g. M'^.k- ^^'bc^ ^^'cb^ ^^'cd- .

and M'dc be the balanced moments obtained by distributing the fixed-end

moments due to the appUed loading in which joints B and C are permitted
to rotate but not to move from their original positions (Fig. 15-146).

B ^i ^ B 1 4*—He- *«C B
A"
B c C c B" c

-(- A times

A JL^ A
D_ D D
\^^AB ---Hb \p\fXB
A
1
1
1

1
D D
(«) (6) (c)

Fig. 15-14

Let Mb^, M'bc ^^cb^ ^^cd^ and M'^c be the balanced moments
3/;;b,
obtained by distributing the fixed-end moments due to any assumed
amount A" of the horizontal movement of joint B or C. The shear
condition required of the frame shown in Fig. 15- 14a is

H., -h Ho Pi
Since
Tj
riA = Mab +
r
Mba ,

r and Hz> =
hi

the shear condition becomes

Pifl . McD + J/pc

= Pi (15-12)
Ai

Also, by superposition,

3/Ba = 3/^^ + k{M%j,)

Mbc = .V'bc + kiM'^c) (15-13)
.1/dc = Mdc + A(-U5;c)

By substituting Eqs. (15-13) in Eq. (15-12),

A.
+ h.

(15-U)

The unknown ratio A: can then be found by soUing Eq. (15-14). Once k
 THE MOMENT-DISTRIBUTIOX METHOD 367

is known, all end moments acting on the frame of Fig. 15-14a may be
found from Eqs. (15-13).
\\liere two or more unknown movements of sidesway are involved,
the resulting set of end moments may be expressed as the sum of (1)
the balanced end moments by distributing the fixed-end moments due
to the appUed loading, and (2) the products of an unknown ratio and the
balanced end moments found by distributing the fixed-end moments
due to a certain amount movement in sidesway, and (3) the
of the first
products of a second unknown and the balanced end moments due
ratio
to a certain amount of the second movement in sidesway, and so on. The
unknown ratios are determined from the shear conditions.
The procedure discussed above will be illustrated by the following
examples.
Example 15-10. Analyze the rigid frame shown in Fig. 15- 15a by the
moment-distribution method. Draw shear and bending-moment dia-
grams. Sketch the deformed structure.
SOLUTION. Relative stiffness

^'^ =
|(S t)(12)
Distribution of fixed-end moments due to the applied loading (see Fig.
15-156)

3f...«=+(l««^P-'= +13.33 kip-ft

(18)^

M FBA (10)(6)(12)^
(18)^
= -26.67 kip-ft

3(24)2
MpBC = + = -f 144 kip-ft
12
Mfcb = - 144 kip-ft

For distribution of these fixed-end moments, see Table 15-11.

Distribution of fixed-end moments due to sidesway (see Fig. 15- 15c)

Fixed-end Moments Due to Sidesway

Relative magnitudes

Mfab = Mfba -|-6E(3/c)A/(18)« + 1^24 X 1.296 +4 -h400

MwCD — y( FDC +6Er.3/c)A/(12)« + K44 X 1.206 +9 -1-900
368 ELEMENTARY THEORY OF STRUCTURES

-144 B '€
+ 144
H-
lOk

4=

(6) fOf due to applied loading (c) FEM due to sidesway

Fig. 15-15

Note that only the relative magnitudes of the fixed-end moments due
to an assumed amount of sidesway A are required.
For distribution of these fixed-end moments, see Table 15-12.
3k, ft
O-'^^BA
i i I i I I I ^ I I I ; c

'«
+
lOk 10-
^ c
^'
+ ac
ts

:s
Hd
+ D .^^
2i~ o,Mdc'O

O Ma r -

Fig. 15-16

Determination of ratio k. The shear condition (Fig. 15-16) is

Ha ^ Hd = 10

or

Simplifying, 20/^^ -h Mba) + S{Mcd + Mdc) = +240

Substituting

Mab = + A-(0) =
3/5, = -99.75 + A-( + 156.7) = -99.75 + 156.7A-

3/cz) = -h99.70 + A-(-h 227.0) = +99.70 + 227.0/:

3/^c = + A'(0) =
in the above equation and solving for the ratio A;,

2(-99.75 + 156.7A) + 3(+99.70 + 227.0A-) = +240

k = -h0.1412
 THE MOMENT-DISTRIBUTION METHOD 369

Table 15-11. Distribution of FEM Due to the Applied Loading

Joint A B C D

Member. AB BA BC CB CD DC
K 2 2 2 2 3 3

Modified K 1.5 1.5 2 2 2.25 2.25

Cycle DF 0.4286 0.5714 0.4706 0.5294

1 FEM + 13.33-26.67 + 144.00 -144.00

Balance -13.33 -50.29 - 67.04 + 67.77 +76.23

2 Carry-over - 6.67 + 33.88 - 33.52

Balance -11.66 - 15.55 + 15.77 + 17.75
3 Carry-over + 7.88 - 7.78
Balance - 3.38 - 4.50 + 3.66 + 4.12

4 Carry-over + 1.83 - 2.25

Balance - 0.78 - 1.05 + 1.06 + 1.19

5 Carry-over + 0.53 - 0.52

Balance - 0.23 - 0.30 + 0.24 + 0.28

6 Carry-over + 0.12 - 0.18

Balance - 0.05 - 0.07 + 0.08 + 0.10

7 Carry-over + 0.04 - 0.04

Balance - 0.02 - 0.02 + 0.02 + 0.02

8 Carry-over '

+ 0.01 - 0.01
Balance - 0.01 + 0.01

Total -99.75 + 99.75 - 99.70 +99.70

Check:
Change -13.33 -73.08 - 44.25 + 44.30 +99.70 t1
-M (c hange) +36.54 + 6.66 - 22.15 + 22.12 -49.85
Sum. +23 21 -66.42 - 66.40 + 66.42 +99.70 -49 S.^

^rel = sum/— i^. . . .

-1 1.60 +33.21 + 33.20 - 33.21 -33.23 + 1(3.62

CIleck Che ck
370 ELEMENTARY THEORY OF STRUCTURES

Table 15-12. Distribution of FEM Due to Sidesway

Joint . A B C D

Member AB BA BC CB CD DC
K 2 2 2 2 3 3

Modified K 1.5 1.5 2 2 2.25 2.25

Cycle DF 0.4286 0.5714 . 4706 0.5294

1 FEM +400.0 +400.0 +900.0 +900.0

Balance -400.0 -171.4 -228.6 -423.5 -476.5 -900.0

2 Carrj'-over -200.0 -211.8 -114.3 -450.0 1

Balance + 176.5 +235.3 +265.6 +298.7

3 Carry-over + 132.8 + 117.6

Balance - 56.9 - 75.9 - 55.3 - 62.3

4 Carry-over - 27.6 - 38.0

Balance + 11.8 + 15.8 + 17.9 + 20.1

5 Carry-over + 9.0 + 7.9

Balance - 3.8 - 5.2 - 3.7 - 4.2

6 Carry-over - 1.8 - 2.6

Balance + 0.8 + 1.0 + 1.2 + 1.4

7 Carry-over + 0.6 + 0.5

Balance - 0.3 - 0.3 - 0.2 - 0.3

8 Carry-over - 0.1 - 0.1

Balance + 0.1 + 0.1 ,

Total + 156.7 -156.7 -227.0 +227.0

Check:
Change -400.0 -243.3 -156.7 -227.0 -673.0 -900.0
- }i (change) -hl21.6 +200.0 +113.5 + 78.4 +450.0 +336.5
Sum -278.4 - 43.3 - 43.2 -148.6 -223.0 -563.5
»r*l = Bum/—K.. +139.2 +
21.6 + 21.6 + 74.3 + 74.3 + 187.8
Ch eck Ch E^ck
 THE MOMENT-DISTRIBUTION METHOD 371

Combination of the two sets of balanced moments (see Table 15-13)

Table 15-13. Combination of the Two Sets of Balanced Moments

Joint A B a D

Member AB BA BC CB CD DC
FE-M from Table 15-11.... + 13.33 - 26.67 + 144.00 -144.00

+0.1412 times FEM from

Table 15-12 . . . +56.48 + 56.48 + 127.08 + 127.08
Total FEM +69.81 + 29.81 + 144.00 -144.00 + 127.08 + 127.08
Balanced moments from
Table 15-11 - 99.75 + 99.75 - 99.70 + 99.70

+0.1412 times balanced

moments in Table 15-12 . + 22.13 - 22.13 - 32.05 + 32.05

Total balanced moments. . .

- 77.62 + 77.62 -131.75 + 131.75
Check:
Change -69.81 -107.43 - 66.38 + 12.25 + 4.67 -127 08
— l-o (change) +53.72 + 34.90 - 6.12 + 33.19 + 63.54 - 2.34
Sum -16.09 - 72.53 - 72.50 + 45.44 + 68.21 -129 42
^rei = sum/ —K + 8.04 + 36.26 + 36.25 - 22.72 - 22.74 + 43.14
Ch eck Ch eck

For reactions, shear and bending-moment diagrams, and the elastic

curve, see Example 14-10.

A
B
5 + 144
v>
CM
(^
-144
o §
B '

B'

§
Ci
^ \

1
+ 1
\
+ 4
lOk (^
(?:
« Do f D
CO '8
^
A +
7ii
A +

(a) (6) FEM due to applied loading (c) FEM due to sidesway
Fig. 15-17

Example 15-11. Analyze the rigid frame shown in Fig. 15-1 7a by

the moment-distribution method. Draw shear and bending-moment
diagrams. Sketch the deformed structure.
 —

372 ELEMENTARY THEORY OF STRUCTURES

SOLUTION. Relative stiffness

AB (12)
\18 6/

BC (12) 2
6/
I

\24

CD (12)
Vl2 4:J

Distribution of FEM due to the applied loading (see Fig. 15-176)

-V... = +«!«= (18)2

+13.33 kip-ft

(1Q)(6)(12)2
^^ FBA — ~ = -26.67 kip-ft
(18)2
3(24)2
^i FBC = + = +144 kip-ft
12
Mfcb - 144 kip-ft
MpcD Mfdc =
For distribution of these fixed-end moments, see Table 15-14.
Distribution of fixed-end moments due to sidesway (see Fig. 15-17c)

FixED-EXD Moments Due to Sidesway

Relative magnitudes

Mfab = ^Ifba -l-6£;(3/.)A/(18)2 +^^24 X 1,296 +4 +400

Mfcd = ^Ifdc +6£(3/.)A/(12)2 -hH44 X 1,296 +9 +900

Note that only the relative magnitudes of the fixed-end moments due
assumed amount of sidesway A are required.
to. an

For distribution of these fixed-end moments, see Table 15-15.

; i i i i
3 k/ft
i i
—— —
[
i
r
*
I
* T—i
"ri
O^BA n^cD
B

10k- 10'

Hr
p
OMdc

Om.^AB

Fig. 15-18
 THE MOMENT-DISTRIBUTIOX METHOD 373

Determination of ratio k. The shear condition (Fig. 15-18) is

Hd =

SimpUh-ing.
10
y+
2{Mab
—— ——
^ Mab
Ha-\-
-h
18
Mba
+
,

+ 3/^^) -h S{Mcd + 3/z>c) = +240

^fcD
10
+ Mpc
12 = in
^^

Substituting Mab = -25.12 + ^-(+342.1)

MsA = -103.58 + ^•(+284.2)
McD = +109.47 + ^(+394.7)
Mj,c = +54.73 + ^(+&47.4)
T KBLE 15-14. D ISTRIBUTION OF FEM Due TO THE Applied Loading

Joint A B
t

c 1 D

Member AB BA CB CD DC

K 2 2 2 ' 2 3 3

Cycle DF 0.5000 0.5000 '

0.4000 0.6000

1 FEM +13.33 - 26.67 + 144.00 ^-144.00

Balance - 58.66 - 58.67 — 57.60 + 86.40

2 Carry-over -29.33 - 28.80 - 29.33 +43.20

Balance - 14.40 - 14.40 - 11.73 - 17.60

3 Carr>--over - 7.20 - 5.86 - 7.20 + 8.80

Balance - 2.93 - 2.93 + 2.88 + 4.32

Carry-over - 1.46 -r 1.44 - 1.46 + 2.16

Balance 0.72 - 0.72 -r 0.58 - f 0.88

Cany--over - 0.36 - 0.29 - 0.36 + 0.44

Balance 0.15 - 0.14 ^ 0.14 - f 0.22

Cany-over - 0.08 1
+ 07 - 0.07 1+ 0.11
Balance - 0.04 - 03 + 0.03 0.04

Carry-over 0.02 02 j- 0.02 + 0.02

Balanced - 0.01 01 + 0.01 + 0.01

Total -25.12 -103.58 +103.58 109.4' 109.4: +54.73

Check:
Change -38.451- 76.91 - 40 42 + 34 53 + 109.47 +54.73
-1^ (change).... 1+38.46+ 19.22 - 17 26 j+ 20.21 - 27.36 -54.74
Sum + O.Oll- 57.69 - 57 68 + 54.74 + 82.11 - 0.01
?rel = sum/— /T. .1 + 28.84 + 28 84 '- 27.37 - 27.37
Check Check Check Check
374 ELEMENTARY THEORY OF STRUCTURES

Table 15-15. Distribution of FEM Due to Sidesway

Joint A B C D

Member AB BA BC CB CD DC
K. 2 2 2 2 3 3

Cycle DF 0.5000 0.5000 0.4000 0.6000

1 FEM +400.0 +400.0 +900.0 +900.0

Balance -200.0 -200.0 -360.0 -540.0

2 Carry-over -100.0 -180.0 -100.0 -270.0

Balance + 90.0 + 90.0 + 40.0 + 60.0

3 Carry-over + 45.0 + 20.0 + 45.0 + 30.0

Balance - 10.0 - 10.0 - 18.0 - 27.0

4 Carry-over - 5.0 - 9.0 - 5.0 - 13.5

Balance + 4.5 + 4.5 + 2.0 + 3.0

5 Carry-over + 2.2 + 1.0 + 2.2 + 1.5

Balance - 0.5 - 0.5 - 0.9 - 1.3

6 Carry-over - 0.2 - 0.4 - 0.2 - 0.6

Balance + 0.2 + 0.2 + 0.1 + 0.1

7 Carry-over + 0.1 + 0.1

Balance - 0.1

Total +342.1 +284.2 -284.2 -394.7 +394.7 +647.4

Check:
Change - 57.9 -115.8 -284.2 -394.7 -505.3 -252.6
-^ (change) + 57.9 + 29.0 + 197.4 + 142.1 + 126.3 +252.6
Sum - 86.8 - 86.8 -252.6 -379.0
Orel = sum/ — K. . + 43.4 + 43.4 + 126.3 + 126.3
Check Ch(3Ck Ch(3ck Check

in the above equation and solving for the ratio k,

2(-128.70 + 626.3A:) + 3( + 164.20 + 1,042.1A:) = +240

4,378.9/c = +4.8
k = +0.0011
 THE MOMENT-DISTRIBUTION METHOD 375

Combination of the two sets of balanced moments (see Table 15-16)

Table 15-16. Combination of the Two Sets of Balanced Moments

Joint A B r,
(_ D

Member AB BA BC CB CD DC
F^M from Table 15-14 + 13.33 - 26.67 + 144.00 + 144.00
+0.0011 times FEM from
Table 15-15 + 0.44 + 0.44 + 0.99 + 0.99

Total F^M + 13.77 - 26.23 + 144.00 -144.00 + 0.99 + 0.99

Balanced moments from

Table 15-11 -25.12 -103.58 + 103.58 -109.48 + 109.47 +54.73

+0.0011 times balanced

moments in Table 15-15. . + 0.38 + 0.31 - 0.31 - 0.43 + 0.43 + 0.71

Total balanced moments. . . . -24.74 -103.27 + 103.27 -109.90 + 109.90 +55.44

Check:
Change -38.51 - 77.04 - 40.73 + 34.10 + 108.91 +54.45
- }i (change) +38.52 + 19.26 - 17.05 + 20.36 - 27.22 -54.46
Sum + 0.01 - 57.78 - 57.78 + 54.46 + 81.69 - 0.01
dre\ = sum/ — K + 28.89 + 28.89 - 27.23 - 27.23
Check Ch Bck Ch eck Check

For reactions, shear and bending-moment diagrams, and the elastic

curve, see Example 14-11.

PROBLEMS
15-1 to 15-22. Analyze the statically indeterminate beams or rigid frames shown in
Probs. 14-1 to 14-22 by the moment-distribution method. Wherever applicable,
use both the regular and the modified-stiffness methods in moment distribution.
Draw shear and bending-moment diagrams. Sketch the deformed structure.
t
 ANSWERS TO PROBLEMS

2-1. Fi = 138.42 lb at (^4)x = 53°38'.

2-2. Fz = 24.62 lb; F4 = -104.82 lb.
2-3. (a)Fz = 54.58 (6) Fz = -38.68 lb, {d,}, = 243°16'.
lb, (^4)x = 176°44';

2-4. Fi = 160 lb at 0.5 ft to left of Fi.

2-5. Va = 105 1b; Vb = 165 1b.
2-6. Va = llVz lb; Fb = 50^ lb.
2-7. Fi = 202.2 lb at (04)x = 2°40' passing through (0,+3.99').
2-8. Ra = 24.68 kips; Rb =21.19 kips at ((9b)x = 136°22'.
2-9. Fz = 825 lb; F4 = 606.25 lb; F5 = -168.75 lb.
2-10. Ha = 12 kips to the right; Va = 16 kips upward; He = 12 kips to the left;
Fc = 11 kips upward.
2-11. Ha = 2S}i kips to the right; Va = 172^ kips upward; He = I7}i kips to
the left; Vc = 24>^ kips upward.
3-1. (a) V = +44 kips, M = -234 kip-ft; (5) V = +10 kips, M = -38 kip-ft;
(c) F = +4 kips, M = -12 kip-ft.

3-2. (a) F = -23 kips, M = -45.5 kip-ft; (6) F = +47.44 kips, 3/ = +58.32
kip-ft; (c) F = -6.56 kips, M = +178.72 kip-ft.
3_3. V = +1.68 kips; M = +74.88 kip-ft.
3-4. A as origin, F = +40 - 3a:, M = -288 + 40x - 1.5^2; B as origin, T^ =
-14 + 3x, M = -54 + 14a: - 1.5x2.
3-5. A as origin, F = -6.56, M = -6.56x + 309.92; B as origin, F = -6.56,
M = -6.56x + 264; C as origin, F = -6.56, M = +6.56x + 100.
3-6. A as origin, F = +18 - 2.4x + 0.045x2, M = +18x - 1.2x2 + 0.015x'
B as origin, F = -12 + 0.6x + 0.045x2, M = +12x - 0.3x2 - 0.015x3.
3-7. At support, V = -P, M = -PL.
3-8. At support, F = -\-wL, M = -MwL^
3-9. At support, F = +16 kips, M = -190 kip-ft.
3-10. At center, F = 0, M = -\-yswLK
3-11. At center, F = 0, M = -{-}iPL.
3-12. At one-third point, F = +P, M = +}iPL.
3-13. At center, F = ±>^P, M = +HP^.
3-14. Max M = +142.2 kip-ft under 1+kip load.
3-15. At support, F = +50 kips, M = -328 kip-ft.
3-16. Max M = +199.01 kip-ft at 16.187 ft from the right support.
3-17. Max M = +75.736 kip-ft at 9.028 ft from the left support.
3-18. See Prob. 3-9.
3-19. See Probs. 3-1 and 3-15.
3-20. See Probs. 3-2 and 3-16.
4-1. Ma = 0; Mb = 38.4 kip-ft (compression inside) ; Mc = 173.4 kip-ft (compres-
sion inside); Md = 19.8 kip-ft (compression inside).
4-2. Ma = Mc = Md = 0; Mb = 38.4 kip-ft (compression outside); M at
15-kip load = 62.55 kip-ft (compression outside).
377
378 ELEMENTARY THEORY OF STRUCTURES
4-3. Ma = Mb = 128 kip-ft (compression inside); Mc = 50 kip-ft (compression
inside).
4-4. Max M in BC = +115.2 kip-ft at 9. 033 ft from B.
4-5. Mba = 24 kip-ft (compression inside); Mbc = 373.5 kip-ft (compression
inside); Mbd = 349.5 kip-ft (compression inside).
4-6. Max M in CD = 150 kip-ft (compression outside) at 2.167 ft from C.
4-7. Mb = 90 kip-ft (compression outside); Mc = 54 kip-ft (compression outside).
4-8. T in rod = 1,500 lb.
4-9. Direct stress in AB = 2.67 kips tension; direct stress in BC = 4.80 kips
compression; direct stress in AD = 6.93 kips compression; direct stress in CD = 0.27
kip tension; shear in AD = -f-4.80 kips.
4-10. Mc = 41.143 kip-ft (compression inside); direct stress in CD = 6.343 kips
compression; shear in CD = 2.743 kips.
4-11. Ha = 0.025 kip to the right; Va = 8.50 kips upward; Hd = 8.625 kips to
the left; Vd = 16.50 kips upward; Mb = 29 kip-ft (compression inside).
5-1. UoUi = 0, LoLi = -1-16.5 kips, U1U2 = -24 kips, L1L2 = -|-16.5 kips,
U2UZ = -24 kips, L2L3 = +19.5 kips, UzUa = 0, L3L4 = +19.5 kips, UoLo = -6
kips, LoUi = -27.5 kips, UiLi = 0, U1L2 = +12.5 kips, U2L2 = -16 kips, Ljf/s =
+7.5 kips, UzLz = 0, UzLi = —32.5 kips, UaLa = —8 kips.
5-2. LoUi = -4,507 lb, LoLi = +3,750 lb, l7iL^ = -3,6061b, L1L2 = +3,750 lb,
U2UZ = -2,704 lb, L2L3 = +3,000 lb, L\Li = 0, U1L2 = -9011b, U2L2 = +500 lb,
U2LZ = -1,250 1b, UzLz = +2,000 lb.
5-3. LoUi = -40.72 kips, LoLi = +28.80 kips, U1U2 = -31.31 kips, L1L2 =
+28.80 kips, U2UZ = -19.80 kips, L2L3 = +29.70 kips, UzUa = -20.87 kips,
L3L4 = +13.20 kips, UJji = -18.66 kips, L4L5 = +13.20 kips, f/iLi = +18.00 kips,
U1L2 = +1.27 kips, U2L2 = +23.10 kips, UiLz = -16.50 kips, UzLz = +6.60 kips,
LzUi = +9.33 kips, 4L4 = 0. C'

5-4. AB = +1,118 lb, AF = -1,000 lb, BC = +1,565 lb, FG = -2,000 lb,

CD = +2,012 lb, GK = -4,000 lb, DE = +2,460 lb, CG = -1,789 lb, BF = DH =
-894 lb, GH = +2,000 lb, CF = CH = +1,000 lb, EH = +3,000 lb.
5-5. See Prob. 5-1.
5-6. See Prob. 5-2.
5-7. See Prob. 5-3.
5-8. See Prob. 5-4.
5-9. See Prob. 5-3.
5-10. U1U2 = -4Ptan U ,U = +P, ^2^/3 = -4.5P tan e, U2L2 = -0.5P,
d,

LoLi = +2.5P tan d, L1L2 = +2.5P tan d, Lol\ = -2.5P sec 6, L2L3 =
UzLz = 0,
+4P tan d, U1L2 = +1.5P sec 6, (/2L3 = +0.5P sec d.
5-11. U1U2 = -4P tan d, UUi = -2P sec d, UiUz = -6P tan d, UxU =
+2P sec e, LoLi = +2P tan d, L1U2 = -P sec d, LxL. = +5P tan e, UiLi =
+ P sec d, L.Lz = +6P tan d, LJ^ = 0.
5-12. See Prob. 5-1.
5-13. See Prob. 5-2.
5-14. See Prob. 5-3.
6-15. See Prob. 5-4.
6-1. Dead panel load = 2,000 lb on top chord and 900 lb on bottom chord. Snow
panel load = 1,000 lb. Wind panel load = 2,250 lb. Max caused by D -j- S:
AB = -21,800; BC = -17,440; CD = -13,080; Ab = +19,500; be = +19,500;
cd = +15,600; Bb = +900; Dd = +8,700. Max caused by ^(Z) + S/2 + Wl
or Wr): Be = -4,960; Ce = +2,900; Cd = -6,270.
6-2. Dead panel load = 2,810 lb. Snow panel load = 2,430 lb. Wind panel
load = 3,440 lb. (D + E) panel load = 6,060 lb. Max caused hy D -\- S: AB ^
 ANSWERS TO PROBLEMS 379

-33,060; BC = -30,150; CD = -27,200; DE = -24,350; = -|-27,510; AK

KL = +23,590; LP = +15,720. Max caused by ^{0 + 8/2 + Wl or Wr):
BK = DM= -5,090; CK = C\U = +4,590; CL = -10,190; £3/ = +13,770;
LM = +9,180. (D + L^) stresses: .45 = -38,240; BC = -34,880; CD = -31,520;
D£: = -28,160; AK = +31,820; A'L = +27,270; LP = +18,180; BK = CM =
-5,040; CK = CM
= +4,550; CL = -10,090; = +13,640; EM
= +9,090. LM
7-1. Dead panel load = 2,810 lb. Snow panel load = 2,430 lb. Wind panel
load = 3,440 lb on truss and 4,500 lb on column, {D E) panel load = 6,060 lb. +
Max caused by D + S: AB = -33,060; BC = -30,150; CD = -27,260; DE =
-24,350; AK
= +27,510; KL = +23,590; LP = +15,720. Max caused by
HiD 5/2 + Wl or Wr); BK =
+ DM
= -5,090; CK = +10,640; = +4,590; CM
CL = -13,550; EM
= +16,800; LM
= +12,210; RK = -35,800. Min caused
by HiD + TFl or Wr): AB = +15,170; 5C = +16,340; KL = -12,850; LP =
-2,860; CK = -24,720; CL = +11,090; = -8,420; EM
= -9,990; RK = LM
+8,240. For stresses due to (D + £") condition, see answers to Prob. 6-2.
8-1. (a) 45 kips; (6) +26.67 kips, -11.67 kips, 300 kip-ft; (c) ±18.75 kips,
337.5 kip-ft.
8-2. (a) 16 kips, +12.25 kips, 9 kips, 6.25 kips, 4 kips; (6) 70 kip-ft, 120 kip-ft.
150 kip-ft, 160 kip-ft.
8-3. (a) -13 kips, -16 kips, -19 kips; (b) -57.5 kip-ft, -130 kip-ft, -217.5
kip-ft.
8-4. (a) 30.67 kips, 50.67 kips; (b) +15 kips, -18 kips; (c) -11.67 kip-ft to
+201.67 kip-ft, -73.33 kip-ft to +193.33 kip-ft, -380 kip-ft.
8-6. (a) 82.75 kips, 99.625 kips; (b) —30 kips on and +57.75 kips on right of
left

left support, +34.25 kips, —18.125 kips, —62.125 kips on left and +50 kips on right

of right support; (c) -200 kip-ft; -86.25 kip-ft to +372.5 kip-ft, -102.5 kip-ft to
+485 kip-ft, -425 kip-ft.
8-6. (a) +76.67 kips, -5.73 to +51.73 kips, -16.27 to +31.6 kips; (b) +1,533
kip-ft, +2,453 kip-ft, +2,760 kip-ft; (c) +44 kips.
8-7. (a) +46.4 kips, -5.6 to +28.8 kips, ±15.2 kips; (6) +1,160 kip-ft, +1,740
kip-ft; (c) +580 kip-ft, +1,405 kip-ft, +1,650 kip-ft.

8-8. (a) 189 kips compression, 183 kips tension; {b) 60 kips tension; (c) 29.36 kips
compression to 33.93 kips tension, 36.21 kips compression to 31.38 kips tension.
9-1. 81.6 kips, +56 kips, +33.6 kips, +10.8 kips.
9-2. 22.5 kips, 16.25 kips, 162.5 kip-ft.
9-3. 28.75 kips, -150 kip-ft.
9-4. 16 kips, 45 kip-ft.
9-5. 18 kips, 81 kip-ft.
9-6. 13.39 kips, 83.70 kip-ft.
9-7. 38.33 kips, 220.1 kip-ft.
9-8. 58.33 kips, 41.67 kips, 300 kip-ft.
9-9. 43.33 kips; 306.2 kip-ft at 14.13 ft from the left support.
9-10. (a) 43.55 kips, 26.22 kips, 12.08 kips; (b) 358.6 kip-ft; 461.3 kip-ft; (c)
462.6 kip-ft with 15.14 ft ofuniform load on the span.
9-11. (a) 93.33 kips, 60.83 kips, 30 kips; (6) 975kip-ft, 1,250 kip-ft; (c)l,255.12kip-ft.
9-12. (a) 1,250 kip-ft, 1,300 kip-ft; (6) 1,300 kip-ft.
9-13. (a) +68.21 kips, -3.96 to +45.71 kips, -13.46 to +27.21 kips; (6) 1,364.17
kip-ft, 2,065.83 kip-ft, 2,269.38 kip-ft; (c) 34.25 kips.
9-14. 682.08 kip-ft, 1,713.75 kip-ft, 2,153.02 kip-ft.
9-15. (a) 105.42 kips compression, 102.25 kips tension; (6) 34.25 kips tension;
(r) 22.26 kips compression to 24.36 kips tension, 26.00 kips compression to 23.76 kips
tension.
380 ELEMENTARY THEORY OF STRUCTURES
10-1. Panel load = 7.43 kips at top and 19.85 kips at bottom; LoLi = L1L2 =
+95.48; L2L3 = +130.94; L3L4 = +146.14; Lof/i = -135.01; UiU^ = -135.00;
UiUz = -147.75; C/3C/4 = -155.88; U1L2 = +50.14; U^Lz = +24.34; UzL, =
+ 16.76; UiLi = +19.85; U2L2 = -15.61; U^Lz = +0.85; U^L^ = -7.43.
10-2.

40-ft span 80-ft span

H20-44 loading
Truck Lane Truck Lane

Max end shear, kips 37.2 38.8 38.6 51 6

Max V at }i point, kips 27.2 26.7 28.6 33.9
Max M at }i point, kip-ft 272 231 572 654
Max M at center, kip-ft 344 308 744 872
Absolute max M, kip-ft 345.96 308 744.98 872

10-3.

80-ft span 180-ft span

H20-S 16-44 loading
Truck Lane Truck Lane

Max end shear, kips . . . 63.6 51.6 68.27 83 6

Max V at 3^^ point, kips 45.6 33.9 50.27 51 9
Max M at }yi point, kip-ft 912 654 2,262 2,551.5
Max M at center, kip-ft 1,160 872 2,960 3,402
Absolute max M, kip-ft 1,164.9 872 2,962.2 3,402

10-4.

Shear, kips Moment, kip-ft

H20-S16-44
Panel 0-1 Panel 1-2 Panel 2-3 Point 1 Point 2

Truck 52.224 37.82^ 23.424 1,305.6 1,913.6

Lane 52.8 33.6, 18.4 1,160 1,740

10-5. LoLi= L1L2 = to +60.55; L2L3 = to +83.04; L3L4 = to +92.68:

LoUi = -95.52 to 0; UxU2 = -85.61 to 0; UiUz = -93.70 to 0; U3U, = -98.86
to 0; U1L2 = -13.54 to +43.50; C/2L3 = -20.57 to +33.33; U3L, = -22.09 to
+33.95; UiLi = to +38.8; U2L2 = -22.72 to +23.43; U^Lz = -18.31 to +28.35;
U,L, = 0.
10-6. (a) 175.68 kips, 11.712 kips per ft; (b) 108.18 kips, 12.821 kips per ft; (c)

1,809.45 kip-ft with wheel 12 at H point, 10.723 kips per ft; (d) 2,291.4 kip-ft, 10.184
kips per ft.

10-7. 7o-i = 325.75 kips, 9.307 kips per ft; 7i.2 = 246.55 kips, 9.588 kips per ft;

F2-3 = 176.35 kips, 9.876 kips per ft; F3.4 114.76 kips, 10.042 kips per ft; A/, =
=
6,515.1 kip-ft, 9.307 kips per ft; M2 = 10,740.3 kip-ft, 8.950 kips per ft; Mz = 13,255.9
kip-ft 8.837 kips per ft; M4 = 14,317.2 kip-ft, 8.948 kips per ft.
 ANSWERS TO PROBLEMS 381

10-8. UiLi = 118.0 kips tension (11.80 kips per ft), U2L2 = 101.3 kips compression
(10.00 kips per ft) or 79.3 kips tension (11.11 kips per ft), U2L3 = 149.0 kips tension
(9.93 kips per ft) or 68.0 kips compression (11.18 kips per ft).
10-9. With or without counters: TiTj = -76.8 to -355.2; IWz = -86.4 to
-399.6; LoLi = L1L2 = +48.0 to +222.0; L2L3 = +76.8 to -355.2; Lof/i = -76.8
to -355.4; UiLi = +16.0 to +103.0; L\L2 = +35.0 to +224.4. Without counters:
l\L2 = +14.8 to -98.3; U2L3 = -29.2 to +115.6; UJjz = -8.0. With counters:
UiLi = to +115.6; LiUz = to +29.2; U2L2 = -8.0 to -98.3; U3L3 = -8 to
-30.8.
10-10. (a) Moment = 120 kip-ft clockwise acting on ends of cross member;
(6) direct stress in upper horizontals = —10.75 kips and +0.75 kip, direct stress in
lower horizontals = +5.75 kips and —5.75 kips; (r) direct stress in the cross hori-
zontal member = — 15 kips and +5 kips; {d) direct stress in the cross horizontal
member = —18^^, —5, and -\-S}i kips.
11-1. Oa = 5.484 X 10-3 radian clockwise, 0^ = 4193 >< IQ-^ radian counterclock-
wise, dc = 2.580 X 10-3 radian clockwise, Ac = 0.3251 in. downward.
11-2. Ob = PL^/2EI clockwise; As = PL^SEI downward.
11-3. Ob = livLy-^SEI counterclockwise; Ab = 41irLV384£'/ downward.
11-4. Oa = llirLV384E/ clockwise; Ob = UwL^/SS^EI counterclockwise; A at
center = 19iyLV2,048E/ downward.
11-5. Oa = Ob = 7wLy2o6EIc, A at center = 93tt'LV12,288£/. downward.
11-6. dc = 261.6 kip-sqft/£/ counterclockwise; Ac = 1,052.8 kip-cu ft /^/upward.
11-7. See Prob. 11-1. A^ax = 0.3695 in. downward at 8.985 ft from left support.
11-8. See Prob. 11-4.
11-9. See Prob. 11-5.
11-10. See Prob. 11-6.
11-11. dc = 161.6 kip-sq ft /E/ counterclockwise; Ac = 652.8 kip-cu ft /E7 upward.
11-12. See Prob. 11-1.
11-13. See Prob. 11-2.
11-14. See Prob. 11-3.
11-15. See Prob. 11-4.
11-16. See Prob. 11-5.
11-17. See Prob. 11-6.
11-18. See Prob. 11-11.
11-19. Ra = %ivL upward; Rb = ^s^vL upward; Mb = /s«^'^^ clockwise.
11-20. Ra = 8.8487 kips upward; Rb = 19.5417 kips upward; Re = 1.9904 kips
downward.
11-21. Ra = Rc = ?i6-P upward; Rb = i^s-P upward.
11-22. Ma = Mb = icL^/l2: R^^ = Rb = wL/2.
11-23. Ra = 11.6558 kips upward; Rb = 15.1104 kips upward; Rc = 0.3662 kip
downward; Ma = 37.535 kip-ft counterclockwise.
11-24. Ra = ^HeP upward; Rb = ^^4 upward; Rc = 2^6 P upward; Mc =
^^8 PL clockwise.
11-25. Rii 0; 0.3672; 0.6875; 0.9141; 1.0000; 0.9141; 0.6875; 0.3672; 0. 3/o: 0;
-g3.828; +8.125; +3.361; 0; -1.641; -1.875; -1.172; 0.
11-26. Ri.O; +0.3906; +0.7250; +0.9469; +1.0000; +0.8531: +0.5750; +0.2594;
0; -0.1312; -0.1500; -0.0938; 0. M,: 0; -2.500; -4.000; -3.500; 0; -2.875;
-3.000; -1.625; 0; +0.875; +1.000; +0.625; 0.
12-1. Oa = Ob = 503.66 kip-sq ft/£:/c clockwise; dc = 593.89 kip-sq H/EIc
counterclockwise; do = 1,054.69 kip-sq ft/EIc counterclockwise; Ah oi A = 30,231
I kip-cu ft/£:/, to the left; Ah oi B or C = 21,165 kip-cu it/EIc to the left.
382 ELEMENTARY THEORY OF STRUCTURES
12-2. Oa = 171 kip-s(i it /Eh clockwise; Ob= 90 kip-sq it/ Eh clockwise; dc = 99
kip-8(i it/EIc counterclockwise; On = 195 kip-sq it /Eh counterclockwise; A// of
A = 6,504 kip-cu it /Eh to the left; Ah of B or C = 3,912 kip-cu it/Eh to the left.
12-3. Oa = 4,246.65 kip-sq it/Eh counterclockwise; Ah oi A = 36,004.5 kip-cu
it/Eh Av of A = 145,933.65 kip-cu it /Eh downward.
to the right;
12-4. Oa= 26 kip-sq it/Eh counterclockwise; Ah oi A = 291 kip-cu it /Eh to the
i-ight; Ay of A = 450 kip-cu it /Eh downward.
12-5. Oa = 291 kip-sq it/Eh counterclockwise; Ah of A = 4,560 kip-cu it /Eh to
the right; Av of A = 4,644 kip-cu it/ Eh downward.
12-6. Oa = 450 kip-sq it /Eh clockwise; A// of A = 4,644 kip-cu it/ Eh to the left;
Ar of A = 14,256 kip-cu it/Eh upward.
12-7. See Prob. 12-1.
12-8. See Prob. 12-2.
12-9. See Prob. 12-3.
12-10. See Prob. 12-4.
12-11. See Prob. 12-5.
12-12. See Prob. 12-6.
12-13. Ha = 4.648 kips to the right; Va = 12.125 kips upward; Hd = 2.552 kips
to the right; Vd = 5.875 kips upward; 6a = 291.2 kip-sq it/EIc counterclockwise;
6b = 85.3 kip-sq it /Eh clockwise; dc = 133.7 kip-sq it/ Eh counterclockwise;
dn = 148.3 kip-sq ft /^/c counterclockwise; Ah oi B = Ah oi C = 2,982 kip-cu it/EIc
to the left.
12-14. Ma = 37.722 kip-ft clockwise; i/A = 5.5730 kips to the right; Va = 10.8614
kips upward; Md = 2.230 kip-ft clockwise; Hd = 1.6270 kips to the right; Vn =
7.1386 kips upward; Ob = 111.91 kip-sq it/Eh clockwise; dc = 92.05 kip-sq it/Eh
counterclockwise; Ah oi B = Ah oi C = 347.0 kip-cu it /Eh to the left.

12-15. Horizontal reaction at 6-ft intervals: 0; 0.07034; 0.11439; 0.13076; 0.11808;

0.07495; 0. Moment at the mid-point: 0; 1.523; 3.598; 6.254; 3.520; 1.426; 0.
13-1. Ah = 0.19728 in. to the right; Ay = 0.91194 in. downward.
13-2. Ah = 0.0144 in. to the left; Ay = 0.3237 in. downward.
13-3. Ah of Lo = 0.144 in. to the right; Ay of Lo = 0.697 in. downward; A// of
L2 = 0.072 in. to the right; Ay of L2 = 0.102 in. downward.
13-4. A = 0.051 in. toward each other.
13-5. A = 0.051 in. away from each other.
13-6. See Prob. 13-1.
13-7. See Prob. 13-2.
13-8. See Prob. 13-3.
13-9. See Prob. 13-4.
13-10. See Prob. 13-5.
13-11. R3 = 911.94/37.71 = 24.183 kips.
13-12. R, = 323.70/104.75 = 3.090 kips.
13-13. Stress in U2L3 = 6.25 kips compression.
13-14. Stress in L2U3 = 6.25 kips tension.
13-15. Influence ordinates for Rz: 0; +0.4861; +0.8576; +1.0000; +0.5859; 0.

Influence ordinates for R,: 0; -0.0916; -0.1146; 0; +0.4484; -|- 1.0000.

13-16. Influence ordinates for U2L3 = 0; -H4', -K2;0; -^4; -^2- Influence
ordinates for L2C/3 = 0; +^4; +^2; 0; +^4; +^2-
14-1. Ra = 21.24 kips upward; Rb = 48.76 kips upward; Mb = 215.2 kip-ft
clockwise.
14-2. Rb = 50.44 kips upward; Re = 41.56 kips upward; Mc = 167.2 kip-ft
clockwise.
14-3. Mba = -594.34 kip-ft; dB = 1,481.2 kip-sq it/EI counterclockwise.
 ANSWERS TO PROBLEMS 383

14-4. Mab = +853.1 kip-ft; Mba = -350.G kip-ft; Ob = 668.7 kip-sq it/EI
counterclockwise.
14-6. Mba = -629.66 kip-ft; A/cb = +164.83 kip-ft; ^b = -1,324.2 kip-sq ft /JB//
counterclockwise.
Mab = +849.48 kip-ft; Mba = -357.83 kip-ft; Mcb = +28.92 kip-ft;
14-6.
Ob = —644.7 kip-sq it/EI counterclockwise.
14-7. Mba = +83.705 kip-ft; Oa = 1.9345 X IQ-^ radian clockwise; Ob = 0.7441
X 10~3 radian counterclockwise; dc = 2.7530 X 10"^ radian counterclockwise.
14-8. Mab = +105.794 kip-ft; Mba = +113.932 kip-ft; Ob = 0.2604 X 10-^
radian counterclockwise; dc = 2.9948 X 10~^ radian counterclockwise.
14-9. Mbc = -125.000 kip-ft; Mcb = -192.708 kip-ft; Oa = 2.375 X 10-^
radian clockwise; Bb = 1.625 X 10"^ radian counterclockwise.
14-10. Mab = +132.534 kip-ft; Mba = +167.411 kip-ft; Mcb = -213.914 kip-ft;
dB = 1.1161 X 10~3 radian counterclockwise.
14-11. Mbc = +531.50 kip-ft; Mcs = -293.25 kip-ft; Ob = 1,760.5 kip-sq ft /i&/c
counterclockwise; Be = 1,363.3 kip-sq it/ clockwise. Eh
14-12. Mab = +878.08 kip-ft; Mba = -300.64 kip-ft; Mcd = +199.84 kip-ft;
Mdc = -500.08 kip-ft; Ob = 835.2 kip-sq it /Eh counterclockwise; dc = 667.2
kip-sq it/Eh clockwise.
14-13. Va = 24.632 kips upward; He = 5.053 kips to the right; Vc = 4.263 kips
downward; Mc = 28.42 kip-ft counterclockwise; Hd = 5.053 kips to the left; Vd =
39.631 kips upward; Md = 25.26 kip-ft counterclockwise; Bb = 94.73 kip-sq it /Eh
counterclockwise.
14-14. Mab = Mbc = +100 kip-ft; Mba = Mcb = -100 kip-ft; Mbe = Meb =
0; Ob = dc= 0.
14-15. Ha = 8.4017 kips to
the right; Va = 51 kips upward; 6 a = 268.85 kip-
sq it/Eh counterclockwise; Ob = 537.70 kip-sq it /Eh clockwise.
14-16. Ha = 14.1025 kips to the right; Va = 51 kips upward; Ma = 112.82
kip-ft clockwise; Ob = 451.29 kip-sq it/Eh clockwise.
14-17. Hd = 5.8938 kips to the right; Vd = 42.2686 kips upward; He = 0;
Ve = 119.4628 kips upward; Od = 188.60 kip-sq it /Eh counterclockwise; Oa =
377.19 kip-sq it /Eh clockwise; ds = Oe = 0.
14-18. Hd = 10.3912 kips to the right; Vd = 43.3025 kips upward; = 83.13 Md
kip-ft clockwise; He = 0; Ve = 117.3950 kips upward; M^ = 0; Ba = 332.53
kip-sq it/Eh; ds = 0.
14-19. Ha = 8.860 kips to the right; Va = 29.523 kips upward; Hd = 0.140 kip
to the right; Vd = 24: All kips upward; dA = 360.4 kip-sq it /Eh counterclockwise;
Bb = 118.1 kip-sq it /Eh clockwise; dc = 227.1 kip-sq it /Eh counterclockwise;
dD = 144.5 kip-sq it /Eh counterclockwise; Ah oi B = 3,616 kip-cu it /Eh to the left.
14-20. Ha = 10.409 kips to the right; Fa = 27.428 kips upward; Ma = 66.13 kip-ft
clockwise; Hd = 1.409 kips to the left; Vd = 26.572 kips upward; ds = 165.3
kip-sq it/Eh clockwise; dc = 183.8 kip-sq it /Eh counterclockwise; dD = 47.5
kip-sq /Eh clockwise; Ah oi B = 199 kip-cu it /Eh to the left.
it
14-21. Ha = 7.858 kips to the right; Va = 28.836 kips upward; Hd = 1.142 kips
to the right; Vd = 25.164 kips upward; Md = 30.77 kip-ft clockwise; dA = 271.1
kip-sq it /Eh counterclockwise; dB = 153.3 kip-sq it /Eh. clockwise; dc = 232.6
kip-sq it /Eh counterclockwise; Ah oi B = 2,333 kip-cu it /Eh to the left.
14-22. Ha = 11.088 kips to the right; Fa = 27.511 kips upward; Ma = 74.16 kip-ft
clockwise; Hd = 2.088 kips to the left; Vd = 26.489 kips upward; Md = 15.07 kip-ft
counterclockwise; dB = 153.8 kip-sq it /Eh clockwise; Be = 175.8 kip-sq it /Eh
counterclockwise; Ah oi B = 412 kip-cu it /Eh to the left.
15-1 to 15-22. See answers for Probs. 14-1 to 14-22.
 INDEX

Absolute maximum bending moment, Conjugate-beam method, 240, 268

150 Consistent-deformation method, appli-
Arch, three-hinged, 20 cation of, to statically inde-
Areas, shear, 35 terminate beams, 250, 254
to statically indeterminate
Beams, deflections of, 232 rigid frames, 284
by conjugate-beam method, 240 to statically indeterminate
by moment-area method, 233 trusses, 306
by unit-load method, 243 Cooper's loading, 204
statically indeterminate, 230 Counters in trusses, 220
analysis of, by consistent-defor-
mation method, 250, 254
by moment-distribution Deflections, of beams {see Beams)
method, 348 reciprocal, law of, 248
by slope-deflection method, 323 of rigid frames, 268, 277
influence diagrams for, 259 of trusses, 296, 299
Bending moment, absolute maximum, Diagrams, bending-moment, 34, 41
150 influence {see Influence diagrams)
definition of, 26 Mohr, 304
diagrams, 34, 41 shear, 34
equations, 31 space, 5, 10, 15
influence diagram for, 114, 127, 129 stress, 72
maximum, due to moving loads, 144, Williot, 302
161, 164 Distribution factors, 347
section of, 36 Duchemin formula, 80
Bent, building, 97-105
Bridge portals, analysis of, 224
Bridge trusses, types of, 177 Equilibrium polygon, 11, 43
Bridges, dead load on, 178 Equivalent loading, on railway bridges,
highway, impact load on, 217 207, 210
live load on, 188 on roof trusses, 86
railway Railway bridges)
{see
Building bents, 97-105
Floor-beam reaction, influence dia-
Carry-over factor, 346 gram for, 128
Composite structures, statically Force polygon, 5, 10, 15
determinate, 54 Force systems, coplanar-concurrent-, 4
385
386 INDEX
Force systems, cophmai-parallel-, 10 Maximum reactions due to moving
general coplanar, 14 loads, 138, 140, 153
Free body, 4 Maximum shear due to moving loads,
Funicular polygon, 21, 43 138, 140, 155
Mohr diagram, 304
Moment-area method, 233, 268
Highway bridges, impact load on, 217 Moment distribution, check on, 356
live load on, 188 Moment-distribution method, 345
application of, to statically indeter-
minate beams, 348
Impact load, on highway bridges, 217 to statically indeterminate rigid
on railway bridges, 218 frames, 361, 364
Indeterminacy, degree of, 232, 268 sidesway in, 364
Index stresses, method of, 70 Moments, principle of, 7
Inflection, point of, 38, 98 Moments and shears, method of, 68
Influence diagrams, for bending
moment, 114, 127, 129
as deflection diagrams, 117, 260, 291, Point of inflection, 38, 98
314 Pole distance, 41
for pier or floor-beam reaction, 128 Polygon, equilibrium, 11, 43
for reactions, 108, 121 force, 5, 10, 15
for shear, 112, 124 funicular, 21, 43
for statically indeterminate beams, string, 11, 43
259 Pressure line, 22
for statically indeterminate rigid
frames, 290
for statically indeterminate trusses, Railway bridges, equivalent loading on,
313 207, 210
for stress in web members of a truss impact load on, 218
with inclined chords, 132 live load on, 203
Intercept, vertical, 41 Reactions, influence diagrams for, 108,

121
maximum, due to moving loads, 138,
Joints, method of, 60 140, 153
Reciprocal deflections, law of, 248
Reversal of stress, 103
Knee brace, 97
Rigid frames, deflections of, 268, 277
statically determinate, 48
statically indeterminate, 267
Live load, on highway bridges, 188
on railway bridges, 203 analysis of, by consistent-deforma-
tion method, 284
Loading (see Equivalent loading)
by moment-distribution
method, 361, 364
Maximum bending moment, absolute, by slope-deflection method, 331,
150 335
due to moving loads, 144, 161, 164 influence diagrams for, 290
section of, 36 Roof trusses, 77
 INDEX 387

Roof trusses, equivalent loading on, Stress in web members of truss with
86 inclined chords, influence dia-
snow and wind loads on. 79, 81 gram for, 132
types 78
of, maximum. 169
weight of. 80 Stress diagram, 72
String polygon, 11, 43

Sections, method of, 62

Shear, definition of, 26 Three-force member, 54, 98
influence diagram for. 112. 124 Three-hinged arch, 20
maximum, due to mo\'ing loads. 138. Truss with inclined chords, stress in
140, 155 web members of, influence dia-

Shear areas, 35 gram for, 132

Shear diagrams. 34 maximum, 169
Shear equations, 31 Trusses, bridge. 177

Sides way, in moment-distribution counters in, 220

method, 364 deflections of, 296, 299
in slope-deflection method, 335 roof (see Roof trusses)

Slope-deflection equations, 320 295

statically indeterminate,

Slope-deflection method, 318 analysis 306of.

application of, to statically indeter- influence diagrams for, 313

minate beams. 323 Two-force member, 54, 59, 98, 295
to statically indeterminate rigid
frames, 331. 335
Unit-load method, 243, 277, 296
sidesway in, 335
Space diagram. 5, 10, 15
Steinman. D. B., 204 Vertical intercept. 41
Stiffness factor, 346
modified, 358
Stress, reversal of, 104 Williot diagram, 302

k
•>-=Vi'^V '^^-^
•-?.\ •'.r^li
V ..1

- 'i; -' :
'\^V
f '
"'•
*• "^
< '\'"^f!!^^
^
'>X ^}p^
i^ j\^
^-,^

J; '"^j^^
^

, >'^i:^
1^
i< f< 'o"V o
>" •
ri
'Z*.^ '4:
^^ .^*
'- ^ ?i

''':^

sV«{,

;-'i