Introduction to

Structural Analysis
 Introduction to
Structural Analysis

Debabrata Podder and Santanu Chatterjee

First edition published 2022
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press

2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

© 2022 Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, LLC

Reasonable efforts have been made to publish reliable data and information, but the author and pub-
lisher cannot assume responsibility for the validity of all materials or the consequences of their use.
The authors and publishers have attempted to trace the copyright holders of all material reproduced in
this publication and apologize to copyright holders if permission to publish in this form has not been
obtained. If any copyright material has not been acknowledged please write and let us know so we may
rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or here-
after invented, including photocopying, microfilming, and recording, or in any information storage or
retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.com
or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-
750-8400. For works that are not available on CCC please contact [email protected]

Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used
only for identification and explanation without intent to infringe.

ISBN: 978-0-367-53272-7 (hbk)

ISBN: 978-0-367-53273-4 (pbk)
ISBN: 978-1-003-08122-7 (ebk)

DOI: 10.1201/9781003081227

Typeset in Times
by KnowledgeWorks Global Ltd.
Contents
Preface.................................................................................................................... xiii
Authors...................................................................................................................... xv

PART I  I ntroduction to Structural Analysis,

Loads, Material, and Section Properties

Chapter 1 Introduction to Structural Analysis....................................................... 3

1.1 Introduction................................................................................ 3
1.2 Historical Background................................................................ 3
1.3 Importance of Structural Analysis.............................................8

PART II  Analysis of Statically Determinate Structures

Chapter 2 Types of Structures and Loads............................................................ 13

2.1 Introduction.............................................................................. 13
2.2 Structural Classifications.......................................................... 13
2.2.1 Tension Structures....................................................... 13
2.2.2 Compression Structures.............................................. 14
2.2.3 Trusses......................................................................... 15
2.2.4 Shear Structures.......................................................... 16
2.2.5 Bending Structures...................................................... 16
2.3 Structural Systems for Transmitting Loads and Load Path....... 17
2.4 Dead Loads............................................................................... 19
2.5 Live Loads................................................................................20
2.6 Wind Loads..............................................................................20
2.7 Snow Loads.............................................................................. 21
2.8 Earthquake Loads..................................................................... 22
2.9 Hydrostatic and Soil Pressure...................................................24
2.10 Thermal and Other Effects.......................................................25
2.11 Load Combinations..................................................................25
2.12 Analytical Model......................................................................25

Chapter 3 Material and Section Properties.......................................................... 29

3.1 Introduction.............................................................................. 29
3.2 Simple Stress-Strain Relationship............................................ 29
3.3 Young’s Modulus or Modulus of Elasticity.............................. 30
v
vi Contents

3.4 Secant Modulus........................................................................ 31

3.5 Tangent Modulus...................................................................... 31
3.6 Shear Modulus or Modulus of Rigidity.................................... 32
3.7 Yield Strength........................................................................... 32
3.8 Ultimate Strength..................................................................... 33
3.9 Modulus of Rupture in Bending............................................... 33
3.10 Modulus of Rupture in Torsion................................................. 33
3.11 Poisson’s Ratio.......................................................................... 33
3.12 Coefficient of Thermal Expansion........................................... 33
3.13 Elastic Assumptions.................................................................34
3.14 Sturctural Nonlinearity............................................................34
3.15 Cross-Sectional Area................................................................ 35
3.16 Center of Gravity and Centroid................................................ 35
3.17 Elastic Neutral Axis................................................................. 36
3.18 Second Moment of Area and Radius of Gyration.................... 36
3.19 Elastic Section Modulus........................................................... 38

Chapter 4 Basic Concepts of Generalized Coordinates, Lagrangian,

and Hamiltonian Mechanics............................................................... 39
4.1 Introduction and Concept of Generalized Coordinates............ 39
4.2 Concept of Configuration Space and Phase Space...................40
4.3 Introduction to Lagrangian and Hamiltonian Formulation
of Mechanics............................................................................ 41
4.3.1 Hamilton’s Equation of Motion................................... 45
4.4 Concept of Symmetry and Conservation Laws........................46

Chapter 5 Equilibrium and Support Reactions.................................................... 47

5.1 Introduction.............................................................................. 47
5.2 Equilibrium of Structures......................................................... 47
5.2.1 Global and Local Equilibrium of Structures............... 48
5.3 Free Body Diagrams................................................................. 50
5.4 Sign Convention....................................................................... 51
5.5 External and Internal Forces.................................................... 53
5.6 Types of Supports for Structures.............................................. 58
5.7 Release of Internal Reactions or Member Forces..................... 58
5.7.1 Releasing Bending Moment........................................ 58
5.7.2 Releasing Shear Force................................................. 61
5.7.3 Releasing Axial Force................................................. 61
5.7.4 Releasing Axial Force and Bending Moment............. 61
5.7.5 Releasing Shear Force and Bending Moment............. 61

Chapter 6 Indeterminacy and Stability of Structure............................................ 63

6.1 Introduction.............................................................................. 63
6.2 Structural Indeterminacy......................................................... 63
Contents vii

6.3 Static Indeterminacy and Stability........................................... 63

6.3.1 Static Indeterminacy of Rigid Structures....................64
6.3.1.1 Shortcut Method for Determining
Internal Static Indeterminacy of Rigid
Structures..................................................... 68
6.3.1.2 Degree of Statical Indeterminacy
When Load is Applied to the
Structure...................................................... 70
6.3.2 Static Indeterminacy of Pin-Jointed
Structures.................................................................... 72
6.3.2.1 Shortcut Method for Determining
Internal Static Indeterminacy of Pin-
Jointed Structures........................................ 73
6.3.3 External and Internal Stability of
Structures.................................................................... 74
6.3.3.1 External Stability of Structures................... 74
6.3.3.2 Internal Stability of Structures.................... 75
6.4 Kinematic Indeterminacy of Structures................................... 77
6.4.1 Kinematic Indeterminacy of Pin-Jointed
Structures or Truss...................................................... 78
6.4.2 Kinematic Indeterminacy of Rigid-Jointed
Structures.................................................................... 79
6.4.3 Summary of All Formulations for Static and
Kinematic Indeterminacy............................................ 83
6.5 Principle of Superposition........................................................ 83

Chapter 7 Plane Trusses and Space Trusses........................................................ 85

7.1 Introduction.............................................................................. 85
7.2 Common Types of Trusses....................................................... 85
7.3 Classification of Coplanar Trusses........................................... 85
7.4 Assumptions on Analysis of Trusses........................................ 87
7.5 Arrangement of Members of Coplanar Trusses – Internal
Stability.................................................................................... 89
7.6 Static Determinacy, Indeterminacy, and Instability of
Coplanar Trusses and Their Solution Methods........................90
7.6.1 Method of Joints..........................................................92
7.6.2 Method of Sections...................................................... 95
7.6.3 Method of Tension Coefficients..................................97
7.6.4 Graphical Method of Truss Analysis........................ 101
7.6.4.1 Bow’s Notation.......................................... 101
7.6.5 Henneberg’s Method of Solution for Complex
Trusses....................................................................... 107
7.7 Compound Trusses................................................................. 111
7.8 Space Trusses......................................................................... 112
7.9 Zero-Force Members of Trusses............................................ 118
viii Contents

Chapter 8 Beams and Frames, Shear, and Bending Moments........................... 121

8.1 Introduction............................................................................ 121
8.2 Axial Force, Shear, and Bending Moments........................... 121
8.3 Shear and Bending Moment Diagrams for a Beam................ 122
8.4 Qualitative Discussion on the Deflected Shape of
Beams..................................................................................... 128
8.5 Relationships between Loads, Shear, and Bending
Moments................................................................................. 130
8.6 Shear and Bending Moment Diagram of Frames................... 136

Chapter 9 Deflections of Beams by Geometric Methods.................................. 143

9.1 Introduction............................................................................ 143
9.2 Deflected Shapes and Elastic Curve....................................... 143
9.3 Double Integration Method.................................................... 146
9.4 Moment-Area Method............................................................ 147
9.5 Conjugate Beam Method........................................................ 150
9.6 Macaulay’s Method................................................................ 154

Chapter 10 Energy Principles and Deflection of Beam....................................... 157

10.1 Introduction............................................................................ 157
10.2 Strain Energy and Pure Bending............................................ 157
10.3 Principle of Virtual Work....................................................... 162
10.3.1 Principle of Virtual Displacements for Rigid
Bodies........................................................................ 163
10.3.2 Principle of Virtual Forces for Deformable
Bodies........................................................................ 166
10.4 Deflection of Trusses by Virtual Work Method..................... 167
10.5 Deflection of Beams by Virtual Work Method...................... 170
10.6 Deflection of Frames by Virtual Work Method..................... 171
10.7 Castigliano’s Theorem............................................................ 173
10.8 Maxwell-Betti Law of Reciprocal Deflections....................... 177

Chapter 11 Rolling Loads and Influence Lines and Their Applications............. 181
11.1 Introduction............................................................................ 181
11.2 Influence Lines for Beams and Frames by Equilibrium
Method.................................................................................... 181
11.3 Qualitative Influence Lines and Müller-Breslau’s
Principle.................................................................................. 188
11.3.1 Müller-Breslau Principle........................................... 188
11.4 Influence Lines for Floor Girders........................................... 190
11.5 Influence Lines for Trusses.................................................... 194
11.6 Maximum Influence at a Point Due to a Series of
Concentrated Loads................................................................ 197
Contents ix

11.7 Maximum Influence at a Point Due to a Uniformly

Distributed Live Load............................................................204
11.8 Absolute Maximum Shear and Moment................................205
11.8.1 Absolute Maximum Bending Moment for
Series of Concentrated Loads...................................208
11.9 Influence Lines for Deflections.............................................. 214

Chapter 12 Cables, Arches, and Suspension Bridges.......................................... 217

12.1 Introduction............................................................................ 217
12.2 Cables..................................................................................... 217
12.3 Cables Subjected to Concentrated Loads............................... 217
12.4 Cable Subjected to a Uniformly Distributed Load................. 221
12.5 Arches..................................................................................... 226
12.6 Three-Hinged Arches............................................................. 227
12.7 Three-Hinged Stiffening Girders........................................... 235

Chapter 13 Analysis of Symmetric Structures.................................................... 243

13.1 Introduction............................................................................ 243
13.2 Symmetric and Antisymmetric Components of
Loadings................................................................................. 243
13.3 Symmetric and Antisymmetric Components of
Loadings.................................................................................244
13.4 Behavior of Symmetric Structures under Symmetric
and Antisymmetric Loadings.................................................246

PART III  A
 nalysis of Statically
Indeterminate Structures

Chapter 14 Introduction to Statically Indeterminate Structures.......................... 251

14.1 Introduction............................................................................ 251
14.2 Advantages of Indeterminate Structure................................. 252
14.3 Disadvantages of Indeterminate Structure............................. 253

Chapter 15 Approximate Analysis of Statically Indeterminate

Structures.......................................................................................... 255
15.1 Introduction............................................................................ 255
15.2 Assumptions for Approximate Analysis................................ 255
15.2.1 Assumptions about the Location of Points of
Inflection................................................................... 256
15.2.2 Assumptions about the Distribution of Forces
and Reactions............................................................ 258
x Contents

15.3 Vertical Loads on Building Frames....................................... 258

15.4 Lateral Loads on Building Frames: Portal Method................ 259
15.5 Lateral Loads on Building Frames: Cantilever Method......... 265

Chapter 16 Method of Consistent Deformations................................................. 271

16.1 Introduction Force Method of Analysis: General
Procedure................................................................................ 271
16.2 Structures with a Single Degree of Indeterminacy................ 271
16.3 Method of Least Work............................................................ 273
16.4 Structures with Multiple Degree of Indeterminacy............... 276
16.4.1 Shear and Bending Moment Diagrams of
Three-Span Continuous Beams................................ 279
16.5 Support Settlements, Temperature Changes, and
Fabrication Errors................................................................... 279
16.5.1 Temperature Changes and Fabrication Errors........... 282

Chapter 17 Influence Lines for Statically Indeterminate Structures................... 285

17.1 Introduction of Influence Lines for Statically
Indeterminate Structures........................................................ 285
17.2 Influence Lines for Beams...................................................... 285
17.3 Influence Lines for Trusses.................................................... 292
17.4 Qualitative Influence Lines by the Müller-Breslau’s
Principle and Influence Line for Frames................................ 294
17.5 Alternate Approach for Finding Influence Line Diagrams
for Indeterminate Beams........................................................ 298

Chapter 18 Slope Deflection Method................................................................... 305

18.1 Introduction............................................................................ 305
18.2 Slope Deflection Equations and Analysis of Continuous
Beams..................................................................................... 305
18.3 Members with Far End Hinged.............................................. 315
18.4 Analysis of Frames without Any Sidesway............................ 318
18.5 Analysis of Frames with Sidesway......................................... 323

Chapter 19 Moment Distribution Method............................................................ 327

19.1 Introduction............................................................................ 327
19.2 General Principles and Definitions......................................... 327
19.2.1 Sign Convention........................................................ 327
19.2.2 Fixed-End Moments (FEMs)..................................... 327
19.2.3 Member Stiffness Factors......................................... 328
19.2.4 Joint Stiffness Factor................................................. 329
19.2.5 Distribution Factor (DF)............................................ 331
Contents xi

19.2.6 Member Relative Stiffness Factor............................. 332

19.2.7 Carry over Factor...................................................... 333
19.3 Basic Concept of Moment Distribution Method.................... 333
19.4 Stiffness Factor Modifications................................................ 335
19.4.1 Member Pin Supported at Far End............................ 335
19.4.2 Symmetric Beam and Loading................................. 336
19.4.3 Symmetric Beam with Unsymmetric Loading......... 337
19.5 Analysis of Continuous Beams.............................................. 339
19.6 Analysis of Frames without Sidesway.................................... 341
19.7 Analysis of Frames with Sidesway.........................................344
19.7.1 Multistorey Frames................................................... 350

Chapter 20 Kani’s Method or Rotation Contribution Method............................. 353

20.1 Introduction............................................................................ 353
20.2 Basic Concept......................................................................... 353
20.2.1 Members without Relative Lateral
Displacement............................................................. 353
20.2.2 Members with Relative Lateral Displacement.......... 362
20.3 Analysis of Frames with Sidesway with Vertical
Loadings................................................................................. 366
20.4 Analysis of Frames with Sidesway with Vertical
Loading and Horizontal Loading at Nodal Points................. 378
20.5 Analysis of Frames with Columns with Unequal Height....... 387

Chapter 21 Column Analogy Method.................................................................. 399

21.1 Introduction............................................................................ 399
21.2 Basic Concept......................................................................... 399
21.3 Development of the Column Analogy Method......................402
21.4 Stiffness and Carry over Factors Determined by
Method of Column Analogy...................................................406
21.5 Fixed End Moments due to Support Settlement.....................407
21.6 Analysis of Portal Frames......................................................409

Chapter 22 Beams and Frames Having Nonprismatic Members......................... 413

22.1 Introduction............................................................................ 413
22.2 Deflections and Loading Properties of Nonprismatic
Members................................................................................. 413
22.3 Moment Distribution for Structures Having Nonprismatic
Members................................................................................. 414
22.3.1 Beam Pin Supported at Far End................................ 415
22.3.2 Symmetric Beam and Loading................................. 416
22.3.3 Symmetric Beam with Antisymmetric Loading....... 416
22.3.4 Support Settlement.................................................... 416
xii Contents

22.4 Slope Deflection Equation for Structures Having

Nonprismatic Members.......................................................... 417
22.4.1 Loads......................................................................... 417
22.4.2 Relative Joint Translation.......................................... 417
22.4.3 Rotation at A.............................................................. 418
22.4.4 Rotation at B.............................................................. 418

Chapter 23 Introduction to Matrix Structural Analysis....................................... 419

23.1 Introduction............................................................................ 419
23.2 Analytical Model.................................................................... 419
23.3 Member Stiffness Relations in Local Coordinates
for 2D Truss............................................................................ 420
23.4 Coordinate Transformation for 2D Truss............................... 422
23.5 Displacement Transformation Matrix for 2D Truss............... 423
23.6 Force Transformation Matrix................................................. 424
23.7 Member Global Stiffness Matrix for 2D Truss...................... 425
23.8 Application of Stiffness Method for Truss Analysis.............. 428
23.9 Application of Stiffness Method for Space Truss Analysis... 430
23.10 Application of Stiffness Method for Beam Analysis............. 430
23.11 Beam Structure Complete Global Stiffness Matrix............... 432
23.12 Application of Stiffness Method for Frame Analysis............. 434

Chapter 24 Introduction to Plastic Analysis of Structure.................................... 437

24.1 Introduction............................................................................ 437
24.2 Stress-Strain Curve of a Ductile Material.............................. 437
24.3 Plastic Moment....................................................................... 438
24.4 Methods of Analysis...............................................................440
24.4.1 The Maximum Principle, Static Theorem, Lower
Bound Theorem, or Safe Theorem............................444
24.4.2 The Minimum Principle, Kinematc Theorem,
Upper Bound Theorem, or Unsafe Theorem.............444
24.4.3 The Uniqueness Theorem......................................... 445
24.5 Static Method for Determining Collapse Load...................... 445
24.6 Kinematic Method for Determining Collapse Load..............449
24.7 Plastic Analysis of Portal Frames........................................... 452

Appendix A............................................................................................................ 457

Appendix B............................................................................................................ 459
Appendix C............................................................................................................465
Appendix D............................................................................................................ 467
Bibliography.......................................................................................................... 489
Index....................................................................................................................... 491
Preface
Structural analysis, as a subject, is a very important and conceptual subject in its
own right among all other subjects that constitute the field of civil engineering.
Starting from ancient times to present day, this subject has flourished depending
upon the need and complexity of the structures coming from the advancement of
human civilization. Structures are mostly classified into two categories, determi-
nate and indeterminate structures. Determinate structures are the preliminary basic
structures that need to be understood well first. Basic principles like elastic limit,
Hook’s law, the principle of superposition, and equilibrium conditions need to be
studied very meticulously before jumping into more complex indeterminate struc-
tures. Structures that we encounter in everyday life are mostly indeterminate struc-
tures. However, the analysis process and basic compatibility conditions like force
and moment equilibrium equations and superposition principle within elastic limit
remain the same as those for determinate structures. That is why we have organized
this book into two parts. In the first part, students will get acquainted with the
basic important theories and analysis procedures for determinate structures. Having
understood the basic principles, in the second part of this book, the indeterminate
structural analysis will be presented, and various methods of solving the problems
are described in detail.
This book is primarily aimed at students taking degree courses in civil engi-
neering. However, students enrolled in master’s degree programs in structural engi-
neering may use this book to brush up on concepts that were taught during their
undergrad study.
No matter how much theory we study, it is very important to go through as many
examples as possible to get hands-on experience toward solving the problems with
theoretical background that can only be understood by going through the chapters.
That is why we have also included solved examples on various problems in different
chapters as well as in the Appendix section, so that students can practice on their
own by studying the text first and then understanding the examples.
This book is mostly self-contained. Preliminary ideas about elastic properties and
some ideas of basic calculus and partial derivatives that are mostly taught at the first
year or second year of degree courses in universities and colleges are assumed to be
known. Apart from that, this book can be studied without taking references to other
books or materials. However, for more solved examples and more advanced ideas on
these topics, students are advised to go through any book provided in the bibliogra-
phy section as they wish.
The authors will be highly open to suggestions regarding further development of
this book and inclusion of any specific topics that may enhance the applicability to
a broader audience.
We are sincerely thankful to all our colleagues, friends, and family members for
standing by our side to complete this book and convert this hardworking effort into
a memorable journey.

xiii
xiv Preface

We thank you all for choosing this book for your course work, and we wish all
the best to everyone on their future endeavors and successful careers in this exciting
field.

Debabrata Podder
Santanu Chatterjee
December 2021
Authors
Dr. Debabrata Podder is currently working as an Assistant Professor (CE) at
National Institute of Technology Meghalaya, India. He has completed his engineering
education from Jadavpur University and Indian Institute of Technology Kharagpur.
After a brief stint as a civil engineer at Shapoorji Pallonji & Co. Ltd., Mumbai,
he took to academics. He has published the book, Residual Stresses, Distortions
and Their Mitigation for Fusion Welding (ISBN: 978-3-659-94213-6) and papers in
several peer-reviewed journals. His research interests include finite element model-
ling and simulation of engineering structures, theoretical and computational solid
mechanics, structural analysis and design, and welding-induced deformations and
residual stresses.

Mr. Santanu Chatterjee has a BE from Jadavpur University, Kolkata in construc-

tion engineering. He has more than 14 years’ research and industrial experience in
the field of civil/structural design engineering. He has also been involved in civil
design of several solar power plants. Mr. Chatterjee also takes interest in concepts
of string theory and quantum mechanics. Recently, he got his paper selected for
the International Conference on Mathematical Modeling in Physical Sciences, and
another of his papers on solar engineering was awarded as Best Scientific Paper
at the National Institute of Technology Durgapur and the International Conference
on Renewable Energy (ICCARE-2019). Another paper on pile design and its rel-
evance (with IS-2911 part 1 section 2 for solar projects) was published at the 37th
European Photovoltaic Solar Energy Conference and Exhibition (EUPVSEC 2020).
In addition, a paper on quantization of classical string was also published in the
International Journal of Physics (IOP) (vol. 1391, 2020).

xv
Part I
Introduction to Structural
Analysis, Loads, Material,
and Section Properties
 1 Introduction to
Structural Analysis

1.1 INTRODUCTION
This chapter provides a general introduction to the subject of structural analysis
and development of this subject from ancient ages to modern days with the help of
various engineers, scientists, and philosophers. Some historical structures with their
distinct characteristic features are also discussed in this chapter. One should have
some knowledge of the historical backgrounds and time-to-time development of this
rich and important subject. This chapter will increase interest and zeal in learning
this subject by heart and setting deeper insight that may open up new avenues of the
analysis process to enhance and enrich this subject to the next level.

1.2  HISTORICAL BACKGROUND

Through the contribution from various civilizations, structural analysis as we know
it today evolved over several thousand years. From living under different natural
habitats like tree shelters and caves, humans started colonizing beside various rivers
throughout the world. There they started making their shelters using stones, clay,
bricks, and numerous cementitious materials by organizing them into different geo-
metrical shapes as per their needs. Initially, they adopted these shapes based on vari-
ous thumb rules that evolved from their past experiences. They slowly modified their
building types/constructions according to their diverse socio­economic, cultural,
religious, and security needs. Among these countless ancient structures, the ‘Great
Bath’ of Mohenjo-daro (Figure 1.1) is believed to be the world’s oldest public pool,
built during the Indus Valley Civilization (3300–1300 BCE). The pool measures
approximately 12-m long and 7-m wide, with a maximum depth of 2.4 m with two
wide staircases lead down into the pool. The pool was watertight, covered by finely
fitted bricks laid on edge with gypsum plaster. Most historians agree that, this would
have been used for special religious occasions where its water was believed to purify
and renew the well-being of the bathers.
Ancient Egyptian (3150–323 BCE) builders are mainly known for their
astounding pyramid-building capabilities. They also knew the techniques of
post and lintel constructions (e.g., Karnak Temple, 2000–1700 BCE). Pharaoh
Djoser’s official Imhotep, a famous architect, and scholar, designed Egypt’s first
Step Pyramid, a pharaoh’s tomb at Saqqara that looks like a stairway to heaven
in about 2600 BCE. Imhotep is referred to as the world’s first structural engineer.
Unbaked mud-brick and stone were two principal building materials in ancient
Egypt.

DOI: 10.1201/9781003081227-2 3
4 Introduction to Structural Analysis

FIGURE 1.1  The ‘Great Bath’ of Mohenjo-daro, 3300–1300 BCE.

Source: https://www.harappa.com/slide/great-bath-mohenjo-daro-0.

For determining sizes of structural members, the Egyptians, the Indus Valley
Civilians, and other ancient builders surely had some kinds of empirical rules
drawn from their previous experiences because, regarding the development of
any theory of structural analysis, there is, however, no evidence found from their
civilizations.
The ancient Greeks (1200–323 BCE) are famously known for their Post and
Lintel-type constructions. Their magnificent Doric, Ionic, and Corinthian order tem-
ples were of such types. The example par excellence is undoubtedly the Parthenon
(Figure 1.2) of Athens, built in the mid-5th century BCE. The temple was built to
house the gigantic statue of Athena and advertise to the world the glory of Athens,
which still stands majestically on the city’s acropolis.
Although the ancient Greeks were blessed with many famous mathematicians,
physicists, inventors, philosophers, scientists, and they also built some magnificent
structures, their contributions to structural theory and analysis were few and far
between. Pythagoras of Samos (about 582–500 BCE) is famous for the right-angle
theorem that bears his name. However, a Babylonian clay tablet confirms that this
theorem was known by the Sumerians in about 2000 BCE. Archimedes of Syracuse
(287–212 BCE) developed some fundamental principles of statics and introduced the
term center of gravity.
The ancient Romans (753 BCE–476 CE) were also excellent builders. Alongside
various Post and Lintel constructions, they mastered the art of building arches,
vaults, and domes, which helped them to cover wider space more easily. Their mas-
tery was further enhanced by the development of concrete (opus caementicium),
which was typically made from a mixture of lime mortar, water, sand, pozzolana,
Introduction to Structural Analysis 5

FIGURE 1.2  Parthenon of Athens, 447–432 BCE.

Source: http://employees.oneonta.edu/farberas/arth/arth200/politics/parthenon.html.

tuff, travertine, brick, and rubbles. Some of the unusual additives were also mixed
with this concrete, such as horsehair, which made the concrete less prone to crack-
ing; animal blood that increased the resistance to frost damage. By implementing
their methods, Romans built various temples, basilicas, pantheons, theatres, amphi-
theaters, public baths, triumphal arches, bridges, aqueducts, roads, lighthouses, etc.
But like Greeks, they too had very less knowledge of structural analysis and made
even less scientific progress in structural theory. They built their majestic structures
from an artistic point of view based on various empirical rules gained from their past
experiences. If those rules got clicked, the structures would have been survived, or
else it got collapsed.
Karl-Eugen Kurrer in his book, ‘The History of the Theory of Structures:
Searching for Equilibrium’, divided the evolution history of structural analysis
into some particular periods and broke those down further into phases as shown in
Figure 1.3.
The preparatory period (1575–1825) of the development and evolution of struc-
tural analysis stretches over around 250 years and is characterized by the direct
application of mathematics and mechanics of that time to simple load-bearing ele-
ments in structures. During this time, buildings and structures were designed mostly
based on empirical knowledge and theory. The theory was evident primarily in the
form of geometrical design and dimensioning rules. This period mainly focused on
the formulation of beam theory. The orientation phase (1575–1700) is characterized
by the sciences (mathematics and mechanics) of this new age discovering the build-
ing industry. In the middle of this phase (1638), the final book of Galileo, ‘Discorsi e
Dimostrazioni Matematiche, intorno a due nuove scienze (Dialogue Concerning
Two New Sciences)’, was published, which is a scientific testament covering much of
6 Introduction to Structural Analysis

FIGURE 1.3  Evaluation periods of structural analysis.

his work in Physics over the preceding 30 years. Galileo’s Dialogue contributed ele-
ments of strength of materials in the form of first beam theory to the menu, though it
was erroneous. Robert Hooke (1635–1703) took the next step and discovered the law
of elasticity in 1660, which later becomes as Hooke’s law. As for the first time, the
differential and integral calculus appeared around 1700; they found their place as a
tool in the applications of astronomy, theoretical mechanics, geodesy, and construc-
tion for very obvious reasons and thus started the application phase (1700–1775).
Mathematicians and natural scientists such as Leibniz (1646–1716), Bernoulli, and
Leonhard Euler (1707–1783) made progress on the beam theory and the theory of
elastic curve. In the first half of the 18th century, engineering schools were devel-
oped for the very first time in France where a scientific self-conception based on the
Introduction to Structural Analysis 7

applications of differential and integral calculus was established. Bernard Forest de

Bélidor (1697–1761) in his book (1729), which was based on the calculus, dealt with
earth pressure, arches, and beams in great detail. Mathematics became a useful tool
for budding civil engineers in the application phase. At the beginning of the initial
phase (1775–1825), Coulomb (1736–1806) in his paper (1776) first applied differen-
tial and integral calculus to beam, arch, and earth pressure theories in a coherent
form and thus provided a method for knowledge of structural analysis. In this phase,
the statics of solid bodies was the only indirectly applied mathematics unlike the
application phase and calculus, which became an integral part of higher technical
education.
In 1826, Louis Henri Navier (1785–1836) published his work where he discussed
the practical bending theory and thus initiated the discipline formation period
(1825–1900). Karl Culmann (1821–1881) expanded his trussed framework theory
(1851) to form the graphical statics (1864/1866). His work was an attempt to give the
structural analysis a mathematical legitimacy through projective geometry. James
Clerk Maxwell (1831–1879), Emil Winkler (1835–1888), Otto Mohr (1835–1918),
Alberto Castigliano (1847–1884), Heinrich Müller-Breslau (1851–1925), and Viktor
Lvovich Kirpichev (1845–1913) consequentially created the linear elastic theory
of trusses. Müller-Breslau with his force method – a general method for calculat-
ing statically indeterminate trusses – rounded off the discipline formation period.
Navier’s practical bending theory formed the basic foundation of structural analysis
in the constitution phase (1825–1850). Due to his work, civil and structural engi-
neers now no longer have to rely solely on their experience-based constructional
knowledge rather they can create and optimize structural models through an itera-
tive design process. Structural analysis got established in the establishment phase
(1850–1875) in continental Europe as iron bridges became common after 1850.
Culmann’s trussed framework theory and graphical statics became the incarna-
tion of iron bridge-building in search of economic use of materials. In the classical
phase (1875–1900), Müller-Breslau developed (1886) general theory of linear elastic
trusses based on the principle of virtual forces as Culmann’s graphical statics was
less suitable for analyzing statically indeterminate systems.
Structural analysis experienced a significant expansion of its scientific area of
study in the consolidation period (1900–1950). Around 1915, the growth in rein-
forced concrete in the construction industry led to the development of the theory
of framed structures and 20 years later to the development of the theory of plates
and shells. When the existing methods of analysis reached their limits during the
skyscraper boom (the 1920s), Hardy Cross (1885–1959) provided (1930) a com-
paratively easier iterative method, known as the moment distribution method, that
could solve the internal forces with a high degree of static indeterminacy quickly.
The accumulation phase (1900–1925) consisted of introducing statically indeter-
minate primary systems instead of statically determinate basic systems and thus
more attention being paid to the deformations of statically indeterminate systems.
Consideration of the secondary stresses in trussed frameworks and analyzing load-
bearing systems of rigid frames became important issues in this phase. Due to the
8 Introduction to Structural Analysis

increase in the use of reinforced concrete, plate and shell structures became an
area of study in the middle of the accumulation phase. The coherent and consistent
arrangement of structural analysis arose out of the principle of virtual displace-
ments at the end of this phase. The contents of structural analysis became tested
and consolidated from the inputs/challenges of the multiple disciplines such as rein-
forced concrete construction, mechanical and plant engineering, crane-building,
and, finally, aircraft engineering. In the invention phase (1925–1950), structural
analysis was characterized by several new developments, such as the theory of
plates and shell structures, development of displacement method alongside the force
method, inclusion of nonlinear phenomena (second-order theory, plasticity), and
formation of numerical methods.
The aircraft industry also reached their limits due to the continuous demands
of rationalizing the calculations of airplane structures in the integration period
(from 1950 to date). To make the airplanes lightweight and stable under the action
of dynamic loads, engineers were lacking some reliable numerical tools where
the whole body can be subdivided into some finite number of elements, consider-
ing them individually in the mechanical sense and then again putting them back
together choosing the right boundary conditions. What is exactly the creator of the
finite element method – Turner, Clough, Martin, and Topp – did in 1956. In the
innovation phase (1950–1975), modern structural mechanics emerged into a theo-
retical level and practical level automation of the structural calculations was initi-
ated. Various numerical methods mostly the finite element method gained more
and more popularity in this phase. In 1960, Ray William Clough (1920–2016)
gave this name, and in 1967, Olgierd Cecil Zienkiewicz (1921–2009) and Yau
Kai Cheung outlined it in a monograph for the first time. In the diffusion phase
(from 1975 to date), the introduction of desktop computers, computer networks,
and lastly the Internet revolutionized computer-assisted structural calculations into
everyday reality.

1.3  IMPORTANCE OF STRUCTURAL ANALYSIS

Structural engineering is the method of mathematical analysis and art of scheduling,
designing, and constructing safe and economical structures without jeopardizing the
overall integrity and which will serve their intended purposes in their anticipated
lifetime. Structural analysis is the first major step of any structural engineering proj-
ect, its function being the prediction of the performance of the planned structure.
Without proper analysis, the critical force, and moments and the corresponding criti-
cal stresses will not be generated. Without the critical loads and their effect on the
structure, a structural engineer cannot properly fix the size and types of structural
members or elements so that it would not collapse under any circumstances in its
entire lifetime. Most of the structures are designed to have a life span of 50 years on
an average. However, as per international codes and standards design life span may
Introduction to Structural Analysis 9

FIGURE 1.4  Different phases of a structural engineering project.

be long enough than this, and to meet higher design life, necessary coefficients for
load increment and material degradation need to be considered in the design steps.
However, for all structural engineers, necessary analysis steps need to be mastered
well before carrying out any real-life projects. Without complete understanding the
nature of critical loads and their combinations, a structure cannot be declared safe
for human use. The various phases of a structural engineering project are shown in
Figure 1.4.
Part II
Analysis of Statically
Determinate Structures
 2 Types of Structures
and Loads

2.1 INTRODUCTION
This chapter provides a general introduction of different types of loads and their
combinations in structural analysis. We introduce five common types of structures:
tension structures, compression structures, trusses, shear structures, and bending
structures. Finally, we consider the development of the simplified models of real
structures for the purpose of analysis. At the very least, loads and determination
of load intensity play an important role in structural analysis. Without a good idea
of the loads acting on the structure, we will not be able to analyze and design to
satisfy the criticality criteria in structural design, leading to improper member selec-
tion, which will ultimately jeopardize structural integrity and quality. Thereby stu-
dents are encouraged to go through this chapter without skipping to the next sections
where the main topics and different analysis procedures are introduced.

2.2  STRUCTURAL CLASSIFICATIONS

It is to be understood that the most crucial decision made by a structural engineer in
implementing an engineering project is the choice of the kind of structure to be used
for supporting or transferring loads to the foundation systems. Most commonly used
structures can be classified into five basic types, depending on the primary loads and
stresses that may develop in their members under active design actions. However,
it should be realized that any two or more of such basic structural types described
in this section below may be combined in a single structure, such as a building or
a bridge, to meet the structure’s functional requirements. Depending on the joint
type, the structures can also be classified into two broad categories, i.e., pin jointed
structures or trusses and rigid jointed structures or frames. In the case of pin jointed
structures, as members are connected through frictionless pins, reacting bending
moment gets released or does not accumulate at the joints, but it is not so in case of
rigid jointed structures. Selecting a proper type of structure or combination of sev-
eral fundamental units to create the final complex structure profoundly impacts the
overall safety and serviceability.

2.2.1  Tension Structures

Tension structures undergo purely tensile force and subsequently uniform tensile
stress across their cross sections under the action of external loads. As tensile stress
is distributed uniformly over the cross-sectional areas of tension members, the mate-
rial of such a structure is utilized in the most effective manner. Tension structures
DOI: 10.1201/9781003081227-4 13
14 Introduction to Structural Analysis

FIGURE 2.1  Tension structure – (a) Cable subjected to point load and (b) uniformly dis-
tributed load (udl).

composed of flexible steel cables are most generally preferred than other steel sec-
tions to support bridges and large-span roofs and domes. Due to their flexibility,
cables have negligible bending stresses and thus subject to only tensile forces. Under
imposed loads, a cable assumes a shape (the most common shape of laded cable
is known as catenary) that enables it to support the load by tensile forces alone. In
other words, the shape of a cable transforms/changes according to the loads (mag-
nitude and direction) acting on it. For example, the shapes that a single cable may
attain under the application of two different concentrated loads (P1 and P2) are pro-
vided in Figure 2.1 (a) for the ease of understanding, whereas the cable transforms
its shape under the action of uniformly distributed load (w) all over it as shown in
Figure 2.1 (b).
Besides cable structures, other types of tension structures used most frequently
elsewhere are vertical rods as hangers (to support balconies or tanks or any overhead
structures) and membrane structures like tents and roofs of large-span domes.

2.2.2 Compression Structures
Compression structures mainly develop compressive stresses under the action of
externally imposed loads. The most common examples of compression structures
are columns (Figure 2.2) and arches. Columns are straight structural elements
subjected to axially compressive loads. Although in most of the cases, there are
moments as well as axial compressive loads that act on the column, which produces
uniaxial or biaxial bending stresses in the column. For an axially loaded column in
which the load path has no eccentricity with respect to the central axis of column,
generates compressive stress only. In the case of columns acted upon by compres-
sive as well as bending stresses, such structural elements are called as beam-column
elements. Normally beam-column terminology is used for steel structures. For RCC
structures, columns are designed including uniaxial or biaxial stresses if applicable,
so that failure modes do not occur during their entire life span.
An arch is a curved structure that looks like an inverted cable. These kinds of
structures generally support roofs or long-span bridges. Arches develop mainly
compressive forces under the action of external loads. They are designed in such a
Types of Structures and Loads 15

FIGURE 2.2  Column subjected to (a) axial compressive load only, (b) axial compression
with uniaxial bending, and (c) axial compression with biaxial bending.

way so that they develop compression forces under the action of main design forces.
As Arches are rigid structures, unlike cables, they produce secondary bending and
shear stresses if the loading condition changes. If these secondary effects are signifi-
cant, they should also be considered during the design phase.
The compression structures are susceptible to buckling failure or instability.
So, they should be adequately designed considering the bracing arrangements if
necessary.

2.2.3  Trusses
Ideally, trusses are composed of straight members connected at their ends by fric-
tionless hinged connections to form a stable configuration. As the external loads are
applied only to the joints of a truss, its members either get elongated or shortened.
That means for an ideal truss, its members are either subjected to pure tension or
pure compression. In the actual scenario, the truss members are connected through
gusset plates using bolting or welding, which produces rigid joints instead of fric-
tionless hinges. These rigid joints generate secondary bending stresses when the
external load is applied or due to the self-weight of individual members. In most of
the cases, these secondary bending stresses are negligible, and the assumption of
frictionless hinges yields satisfactory results.
As the trusses are light weight and have high strength, they are the most com-
monly used types of structures supporting roofs of buildings. Please refer Chapter 7
for details of trusses and their analysis procedure.
16 Introduction to Structural Analysis

FIGURE 2.3  Multistory framed building with shear walls.

2.2.4 Shear Structures
Shear structures are the structural elements that develop mainly in-plane shear, with
relatively small bending stresses under the application of external loads. Due to
the external loads, shear is developed between two adjacent sections of a structure,
which tries to slide the sections opposite each other. Usually, in concrete structures,
shear walls are being designed and constructed in high-rise structures (as shown in
Figure 2.3) to absorb any sudden horizontal shear load due to earthquake or wind.
For shearing-resistant structural design, engineers need to take special design cau-
tion as per guidelines presented in various international codes and standards to
ensure structural integrity and safety even under severe accidental loads.

2.2.5  Bending Structures

Bending structures are structural elements subject to transverse loads and develop
bending stresses as a result. As discussed in compression members, sometimes the
bending moment is created due to eccentricity on the applied external load, even in
columns. In such cases, the column needs to be designed against both vertical com-
pression and uniaxial or biaxial bending effects. Typically, structures subject to pure
bending are called beam elements. Beams axis is oriented along its length direction,
and its cross-section is perpendicular to it. The beam is a structural element whose
one dimension is much larger than its other two dimensions. In RCC and steel, beams
Types of Structures and Loads 17

subject to pure bending stress need to be taken care of by proper design calculation
and selecting an adequate section to withstand the effect under bending stress. Beams
subject to pure bending only are called Euler-Bernoulli beams and beams subject to
bending as well as shear are called shear deformable beams or Timoshenko beams. If
the beam’s wavelength is longer than six times its height, shear deformation and rota-
tional inertia do not play any significant role. This type of beam can be treated as an
Euler-Bernoulli beam; otherwise, they are treated as Timoshenko beams. Both the
beams have different stress-strain characteristics. As per various international codes
and standards, analysis procedures and guidelines must be followed meticulously to
ensure structural integrity even under adverse loading conditions.

2.3 STRUCTURAL SYSTEMS FOR TRANSMITTING

LOADS AND LOAD PATH
In most buildings, bridges, walkover platforms, and other civil engineering facilities,
two or more of the previous section’s primary structural types are combined to form
a structural system that can transfer the imposed loads to the ground through the
foundation system. Such structures are also called framing systems or frameworks,
and the components or elements of such an assembly are called structural members.
The buildings and bridges are analyzed and designed as per various international
codes and standards to withstand loads in both the vertical and horizontal directions
or any other general direction. Loads acting in any generalized direction can be
transformed into vertical and horizontal components. These components are applied
separately to the structural system during the analysis process to calculate effec-
tive net stress developed in various structural elements comprising the system. The
vertical loads, generated mainly to the occupancy, self-weight, dust, and snow or
rain, are commonly referred to as gravity loads or dead loads. The horizontal loads,
imposed mainly due to wind and earthquakes, are called lateral loads. The term load
path is considered to describe how a load acting on the structure as a whole is trans-
mitted, through various structural elements of the structural system, to the ground.
Depending on the type of load to be transferred, there are two primary load paths:
gravity load path and lateral load path.
The gravity load path of a single-story building is mainly from slab to beam,
beam to column, column to foundation system, and finally from foundation system
to the soil. Any vertical distributed area load, such as snow, dust, or rain, acting on
the roof slab is first transferred to the secondary beams as a distributed line load
(force per unit length of the member). As primary beams support the secondary
beams, the secondary beam reactions become concentrated forces on the supporting
points of the primary beams. Similarly, the primary beams supported by columns
transmit the load, via their support reactions, to the columns as axial compressive
forces and/or moments. The columns transmit the load to the foundation system
underneath it, which finally disperses the load to the ground soil.
The load transfer mechanism described above is yet much more straightforward
than it happens. Actual load transfer from slab to primary and/or secondary beams
is determined effectively by applying yield line theory. According to this theory,
18 Introduction to Structural Analysis

FIGURE 2.4  Slab to beam (primary) load transfer mechanism.

the loads transferred from slab to surrounding beams depend on many factors like
the slab panel position (i.e., interior or exterior panel), edge fixity conditions of the
slab, surrounding beams arrangement, and dimensions. Let us consider a typical roof
and beam arrangement as shown in Figure 2.4.
The above triangular and trapezoidal distributed load will act on the correspond-
ing primary beams, as shown in the figure. Ultimately, these loads will be transferred
to columns as axial compressive loads and/or moments from the primary beams con-
nected with the columns as per the tributary area of that particular column. This is
the primary load path for gravity load transfer in all civil engineering structures.
The tributary area, as shown in Figure 2.5, is related to the load path and is used to
determine the amount of loads that beams, girders, columns, and walls carry. For
the lateral load path, horizontal loads coming from wind or seismic force are passed
to the periphery beams of the supporting slabs through shear action to the columns
and progressively to the foundation systems. Horizontal loads can also be trans-
ferred through lateral load resisting elements such as shear walls to the foundation
directly. In general, horizontal actions produce more complex load distribution pat-
terns, and the analysis procedure is also very much complicated compared to gravity
loads. Hence, more detailed analysis as per codal provisions and dynamic theories
must be studied for detailed horizontal load analysis and its effect on the structures.
Types of Structures and Loads 19

FIGURE 2.5  Tributary area for (a) beam ‘AB’ and (b) column ‘C’.

Modern-day computer programs are primarily employed for such analysis. However,
for smaller structural elements, different analysis procedures for horizontal loads are
required to be studied for a basic understanding of the process.

2.4  DEAD LOADS

Dead loads are gravity loads of constant magnitudes and fixed positions that act
permanently on the structure. Such loads consist of the structural system’s weights
and all other material and equipment permanently attached to the structural system.
For example, the dead loads for a building structure include the weights of frames,
framing and bracing systems, floors, roofs, ceilings, walls, stairways, heating and
air-conditioning systems, plumbing, electrical systems, and other permanent mis-
cellaneous items. The weight of the structure and all various internal equipment
are not known in advance of design and is usually assumed based on experience.
After the structure has been designed and the member sizes ascertained, the actual
weight is computed by using the member sizes and the unit weights of materials. The
actual weight is then compared to the assumed weight, and the design is reviewed
and revised accordingly if required. The unit weights of some common construction
materials are given in Table 2.1. The weights of permanent service equipment, such

TABLE 2.1
Unit Weights of Various Common Structural
Materials
Material Unit Weights (KN/m3)
Concrete 25
Steel 78.5
Wood 6.3
Aluminum 25.9
Brick 18.8
Cement plaster 2.4
20 Introduction to Structural Analysis

as heating and air-conditioning systems, are usually obtained from the suppliers of
respective items and from manual books of the same.

2.5  LIVE LOADS

Live loads are the transient loads, i.e., loads having variable magnitudes with respect
to time and/or locations caused by the use and occupancy of the building or other
structure. More specifically, the term live loads is applied to address all loads and
actions on the structure which do not include construction or environmental loads
such as snow load, wind load, rain load, earthquake load, floor load, or dead load.
However, since the probabilities of occurrence for environmental loads are different
from those due to the use of structures, the current codes use the term live loads to
refer only to those variable loads caused by the use of the structure. Any such variable
loads are referred to as live loads i.e., the weight of people, furniture, appliances, cars
and other vehicles, and equipment. The intensity of live loads that need to be included
during structural analysis and design are generally specified in various international
building codes and standards. As the point of application of live loads may change,
each member of the structure must be designed for the load location that will produce
the maximum most severe stress conditions in members. Different members of a
structure may reach their maximum allowable stress intensities at different positions
for the given applied load. The method for calculating a live load location at which a
particular response characteristic, such as stress resultant, or a deflection, of a struc-
ture is maximum (or minimum) is discussed in influence lines in Chapters 11 and 17.

2.6  WIND LOADS

Wind loads are produced when a structure is subjected to wind flow toward and
around the structure. The intensity of wind loads that may act on a structure depends
on the geographical position of the structure, wind obstructions in its vicinity, such
as nearby houses, forestry, or any kind of wind obstructions present all around the
structure, the plan and elevational geometry of the structure, and the natural vibra-
tional characteristics of the structure itself. Distance from the seashore is also a
major factor for the intensity of wind load that will act on the structure. Various
international codes provide necessary guidelines for wind analysis that need to be
incorporated during the design analysis stage, but most of the codes apply the same
fundamental relationship between the wind speed V and the dynamic pressure p
induced on a flat surface normal to the wind flow. This can be formulated by apply-
ing Bernoulli’s principle and is given by the following equation:
1
p= mV 2
2
in which m is the mass density of air. Using the unit weight of air as 12.02 N/m3 for
the standard atmosphere (at sea level, with a temperature of 15°C ) and expressing the
wind speed V in m/s, the dynamic pressure p in N/m2 or pascal is given by:
1  12.02  2
p= ×  V =  0.613V
2
2  9.81 
Types of Structures and Loads 21

However, IS:875 (Part 3):2015 uses the following equation for net wind pressure due
to design wind speed:

pz = 0.6 Vz 2

where:
pZ = wind pressure at height z, in N/m2
Vz = design wind speed at height z, in m/s
Vz is determined from the following equation:

Vz = Vb k1 k2 k3 k4  

where:
k1 = probability factor (risk coefficient)
k2 = terrain roughness and height factor
k3 = topography factor
k4 = importance factor for the cyclonic region
Vb = basic wind speed at the site location

For all the above-mentioned factors, IS:875 (Part 3): 2015 provides detailed clauses
and tables with values that need to be incorporated depending on the site location.
On the other hand, Uniform Building Code, i.e., UBC 1997 (Chap 16, Div. III,
Sec. 1620) code uses the following formula for wind load analysis:

P = CeCq qs I w

where:
P = design wind pressure
Ce = combined height, exposure, and gust factor coefficient
Cq = pressure coefficient for the structure or portion of the structure under
consideration
qs = wind stagnation pressure at the standard height of 33 ft (10,000 mm)
I w = importance factor

To determine the above values, UBC 1997 provided various design tables and guide-
lines to pick appropriate values as per site condition and build structural details.

2.7  SNOW LOADS

If the site location is such that there is snowfall during every winter season, snow
load needs to be considered while designing and analyzing structures. The design
snow load intensity on a structure is based on the ground snow thickness as per its
geographical location. Various international codes and standards provide a neces-
sary guideline for the snow load analysis process. In particular, ASCE 7 provides
a clear guideline for calculating snow load intensity at any elevated structural level
based on the ground peak snow thickness and meteorological data at the site location.
22 Introduction to Structural Analysis

At first, the ground snow load was required to be calculated. Then the design snow
load intensity at the roof level or at an elevated level of the structure is calculated
by using factors as the structure’s exposure to wind and its thermal, geometric, and
functional properties and characteristics. As per ASCE 7, the following equation is
used for calculating snow load intensity at the flat roof level of a building,

p f = 0.7 CeCt I s pg

where:
p f = snow load on flat roofs (“flat” = roof slope ≤ 5°), in kN/m2
pg = ground snow load intensity, in kN/m2
Ce = exposure factor
Ct = thermal factor
I s = importance factor

The design snow load for a sloped roof can be determined by multiplying the cor-
responding flat-roof snow load by a roof slope factor Cs :

ps = Cs p f

There are several clauses and guidelines provided in ASCE 7 to determine the snow
load intensity at the roof level using the above formulas.

2.8  EARTHQUAKE LOADS

An earthquake is a sudden vibration and movement of a portion of the earth’s sur-
face. Although the ground surface vibrates in both horizontal and vertical directions
during the quake, the magnitude of the vertical component of ground vibration is
usually very small and does not have an effect at a considerable scale on most struc-
tures. The horizontal component of ground vibration is the severe one that induces
large dynamic stresses in the structural element that causes damage and collapse
of a particular structure. Due to severity, earthquake force must be considered in
designs of structures located in high earthquake prone locations. During an earth-
quake, the foundation of the structure vibrates with the ground, following Newton’s
law of inertia. On the other hand, the above-ground portion of the structure (gener-
ally called superstructure), because of the inertia of mass, always tries to resist the
motion, thereby causing the structure and all its elements or members to vibrate in
the horizontal direction as well. These vibrations induce horizontal shear forces in
the structural elements. To predict the resulting stresses accurately in the structural
element due to earthquake, dynamic analysis, considering the mass (generally called
seismic mass of the structure) and stiffness characteristics of the structure, must be
applied. However, for small-to-medium height buildings, most international codes
provide a guideline of what is popularly known as equivalent static forces analy-
sis or equivalent static method. The equivalent static method starts from the base
shear force that will be applied due to the earthquake at the foundation level of the
Types of Structures and Loads 23

structure. Then this foundation shear force is required to be distributed along with
various levels of the building. International codes provide all necessary equations
and tables of empirical values to distribute the horizontal base shear force to various
parts of the building. ASCE 7 provides quick guidelines for equivalent static method
using the following equation for base shear force analysis.

V =  C S W

where:
W = seismic mass of building comprising total dead load and a prescribed per-
centage of live loads
CS = seismic response coefficient. This coefficient is a function of several other
coefficients as stated below:

S DS
CS =  
R /I e
where:
S DS = design spectral response acceleration in the short period range
R = response modification coefficient
I e = importance factor

However, IS:1893 (Part 1):2016 provides the following equation for calculating base
shear, VB  for a structure.

 Z   Sa 
 2   g 
VB = AhW = ×W
 R
I

where:
Ah = design horizontal earthquake acceleration coefficient for a structure
W = seismic weight of building
Z = seismic zone factor
I = importance factor
g = design acceleration coefficient for different soil types, normalized with peak
Sa

ground acceleration, corresponding to natural period T of the structure (con-

sidering soil-structure interaction if required)
R = response reduction factor

IS:1893 (Part 1):2016 provides detailed tables of empirical values for the coefficients
mentioned above based on seismic zones and structure material type. After carry-
ing out the base shear value, it needs to be distributed among various floor levels of
the building proportionate to that particular floor’s seismic mass to the total seismic
mass of the structure.
24 Introduction to Structural Analysis

2.9  HYDROSTATIC AND SOIL PRESSURE

Structures constructed to retain water, such as dams and tanks, as well as struc-
tures in coastal areas partially or fully submerged in water, need to be analyzed and
designed to resist water pressure. This water is the most stagnant type and does not
create any dynamic effects on the structure. Hence hydrostatics is the key principle
while designing these types of structures. Hydrostatic or water pressure acts normal
to the structure’s submerged surface as per Pascal’s law, with its magnitude varying
linearly with height, as shown in Figure 2.6. Thus, the hydrostatic pressure on the
structure at a depth h, from the top surface, and the resultant thrust exerted on the
structure can be written by the following equations, respectively.

p =  γ gh
 

1
Pw = γ gH 2
2

where:
p = hydrostatic pressure on the structure at a depth h
γ = density of the water or liquid retained by the structure
g =  acceleration due to gravity at the location
h = depth at which pressure need to be measured
Pw = resultant thrust on the structure
H = total depth of water

Underground structures like basement walls and floors, and retaining walls need
to be designed to resist soil pressure. Generally, for designing purposes, active soil
pressure gives the most conservative result. The vertical soil pressure is given by
the same equation as that for hydrostatic pressure with change in γ only, represent-
ing the unit weight of the soil or the earth. The lateral soil pressure is calculated
based on specific parameters known as active and passive earth pressure coefficients.

FIGURE 2.6  Hydrostatic pressure on water retaining structures.

Types of Structures and Loads 25

These coefficients depend on the type of soil, and one needs to multiply the vertical
pressure with these coefficients only to get the horizontal soil pressure acting on the
structure. Sometimes, part of the structure remains on the earth, and partly it is sub-
merged, or sometimes the earth is completely submerged in the water itself. In those
cases, both water and soil pressures need to be applied to the structure separately.
Due to water and earth, the net destabilizing force needs to be added together to
proportionate and design sections for the structures correctly.

2.10  THERMAL AND OTHER EFFECTS

Statically indeterminate structures are prone to be affected due to unusual stresses
induced from temperature changes, shrinkage of material, fabrication errors, and
uneven settlements of supports. These effects significantly impact the increase in
overall stress conditions in the structural elements, ultimately causing the unusual
collapse of a structure partly or wholly. International codes and standards provide
important guidelines toward these forces and their inclusion into design calculations.
In most cases, one or more expansion joints are provided if the structure’s length
exceeds 45 m to cater to thermal expansion effects due to a temperature change.
Fabrication errors induce inappropriate induction of moments and faulty joint func-
tioning, which will cause untimely failure of the structure.

2.11  LOAD COMBINATIONS

All the load cases discussed above are first applied to the structure, and support reac-
tions, member forces, and member stresses are preferably carried out. After that, all
loads need to be combined systematically in such a way to produce the most severe
load cases, which the structure needs to withstand during its entire life span. This
process of combining all loads to produce a resultant effect on the structural ele-
ments and check overall stability and integrity is called load combinations. Various
factors, such as an increase in magnitude and probability of occurrence, need to be
incorporated while combining the loads. Hence during load combinations, various
factors are multiplied with the individual load cases called safety factors. But one
needs to follow specific rules as per the design code being selected for the intended
job. Hence, we provide a few most commonly used design load combinations as per
different international codes and standards for reference in Table 2.2.

2.12  ANALYTICAL MODEL

An analytical model is a simplified pictorial or graphical representation of the origi-
nal structure for the purpose of analysis. The purpose of the model is to simplify
the design and analysis of a complicated structure. The analytical model depicts or
represents, as accurately as possible, the natural characteristics of the structure of
interest to the designer and engineers while removing much of the details about the
members, connections, and so on, that is believed to have little effect on the desired
overall characteristics of the original structure under the application of prescribed
loads. Formation of the analytical model is one of the most crucial steps of the design
 26
TABLE 2.2
Load Combinations used in Different Countries

American Australian Mexican New Zealand Canadian European Indian

ACI 318-11 AS 3600-2009 RCDF 2004 NZS 3101-06 CSA A23.3-04 EURO 2-2004 IS 456:2000
1.4DL 1.35DL 1.4( DL ± LL ) 1.35DL 1.4DL - 1.5 ( DL + LL )

0.9DL ± WL 0.9DL ± WL 0.9DL ± WL 0.9DL ± WL 1.25 DL + 1.5 LL ± 0.4WL DL ± 1.5WL 1.5( DL ± WL )

1.2 DL + LL + 0.5 Lr ± WL 1.2 DL + 0.4 LL ± WL 1.1DL + 1.1LL ± WL 1.2 DL + 0.4 LL ± WL 1.25 DL + 1.5S ± 0.5 LL 1.35 DL ± 1.5WL 0.9 DL ± 1.5WL

1.2 DL + 1.6 LL ± 0.5WL DL + 0.4 LL ± EQ 0.9DL ± EQ DL + 0.4LL ± EQ 1.25 DL + 0.5 LL ± 1.4WL 1.35 DL + 1.5 LL ± 0.9WL 1.2( DL + LL ± WL )

Introduction to Structural Analysis

0.9DL ± EQ 1.2 DL + 0.6 LL 1.1DL + 1.1LL ± EQ 1.2DL + 0.6LL DL + 0.5 LL ± EQ ± S 1.35 DL + 1.5WL ± 0.9 LL 1.5( DL ± EQ)

1.2 DL + 1.6 S ± 0.5WL 1.2DL + 1.5LL - 1.2DL + 1.5LL - DL ± EQ 0.9 DL ± 1.5EQ

1.2DL + LL + 0.5S ± WL - - - - DL + 0.3 LL ± EQ 1.2(DL + LL ± EQ)
1.2 DL + LL + 0.2 S ± EQ - - - - - -

Note: DL: Dead load; LL: Live load; EQ: Earthquake load; Lr : Roof live load; WL: Wind load; S: Snow load.
Types of Structures and Loads 27

process; it demands experience and knowledge of design practices and thorough

knowledge and understanding of the behavior of the original structures. Based on
the structural modeling, we do mathematical modeling, enabling us to analyze the
structure most effectively to get the most severe stress concentrations and its loca-
tion in the structure. Based on this critical load, designers can analyze and design
the structure on such a scale to resist structural failure and ensure structural integrity
throughout the entire life span of the structure. In chapters following this, students
will apply this method to the detailed analysis of several determinate and indeter-
minate structures by line diagrams/analytical models acted upon with prescribed
imposed loads.
 3 Material and Section
Properties

3.1 INTRODUCTION
This chapter provides a general overview of different types of material and their
properties. In this chapter, we will briefly introduce several important definitions
and terminologies related to material properties, which directly impact determining
bending stress and member deflection parameters. Important parameters like differ-
ent elastic modulus and stress-strain relationships will be discussed in this chapter.
Students are encouraged to brush up on their previous knowledge on strength of
materials by going through this chapter in detail.

3.2  SIMPLE STRESS-STRAIN RELATIONSHIP

As per Hook’s law, within the elastic limit, stress (σ ) is proportional to strain (ε ) .
This is also called the principle of elasticity:

σ ∝  ε

In the form of equality, we need a proportionality constant ( K ) known as the elastic

constant of material or coefficient of elasticity or modulus of elasticity or elastic
modulus of the material.

σ = Kε

Depending upon different types of elasticity tests, the proportionality constant also
named accordingly. For linear elongation of materials, the proportionality constant
is known as Young’s modulus, for shear resistance test, it is called shear coefficient
of the materials, etc. Modulus of elasticity plays an important role in its behavior
under different stress and loading conditions. A standard stress-strain experiment
on a ductile material produces a stress-strain curve as shown in Figure 3.1, from
which we can study the material properties under various loading conditions. In this
graph, the portion generated from origin and moves ahead in a straight liner path
is called the elastic range of material or the proportional limit. Within this range,
when the load is removed, the structure regains its original shape and size, and we
say the material was within the elastic range. Different material has a diverse elastic
range. After this proportional limit, nonlinearity arrives in the stress-strain curve
associated with stress-induced plastic flow in the material. By studying this graph,
we can make an excellent analytical comparison on the limit up to which the materi-
als can retain their original shape and size. By adequately checking these test curves,
we can select the right kind of material for our intended work.
DOI: 10.1201/9781003081227-5 29
30 Introduction to Structural Analysis

FIGURE 3.1  Typical stress-strain curve for ductile material.

The engineering measures of stress and strain denoted as σ e and ε e , respectively,

are determined from the measured load and deflection using the original cross-
sectional area of the specimen A0  and original length L0 as:
P δ
σe = , εe =
A0 L0
If the stress σ e is plotted against ε e , an engineering stress-strain curve is obtained.
The engineering stress-strain curve must be interpreted with caution beyond the
elastic limit because the specimen’s dimension experiences substantial change from
its original values. Using actual or true stress instead of engineering stress can pro-
vide more direct measures of the material’s response in this plastic flow range.

3.3  YOUNG’S MODULUS OR MODULUS OF ELASTICITY

As stated in the previous section, Young’s modulus ( E ) is the proportionality con-
stant of the material under linear deformation. It is the slope of the stress-strain curve
in the elastic limit. So it is defined as linear stress versus linear stain, and it can be
written as:
P /A
E = 
∆l /l
where, P is the tension or compression force applied on the specimen within the
elastic limit, A is the cross-sectional area of the specimen, ∆l is the change in length
due to linear elongation or contraction, and l is the actual or original length prior to
start of test.
Material and Section Properties 31

3.4  SECANT MODULUS

Secant modulus is the slope of a line drawn from the origin of the stress-strain dia-
gram and intersecting the curve at the point of interest. It is denoted by Es . This is
also known as the static modulus of elasticity. In Figure 3.2, the secant modulus has
been shown in different stress levels. Secant modulus can also be shown as a percent-
age of Young’s modulus (e.g., 0.6E or 0.8E). It is used to describe the stiffness (i.e.,
the load required to produce unit deformation) of the material in its inelastic zone of
the stress-strain curve.

3.5  TANGENT MODULUS

The slope of the tangent at any point in the stress-strain curve is known as the tan-
gent modulus of elasticity ( ET ). Based on the tangent locations, it can have different
values. For example, a tangent modulus becomes equal to Young’s modulus of elas-
ticity if drawn within the stress-strain curve’s linear range. Outside the linear region,
it is always less than Young’s modulus. In Figure 3.2, the tangent modulus is drawn at
half of the ultimate stress level. It is mainly used to describe the stiffness of the mate-
rial in the nonlinear range. Structural materials that do not have any well-defined
yield point like concrete, tangent modulus provide a better opportunity to study the
stress-strain characteristics and their elasticity range.

FIGURE 3.2  Different modulus of elasticity.

32 Introduction to Structural Analysis

3.6  SHEAR MODULUS OR MODULUS OF RIGIDITY

Within elastic limit, the ratio of shear stress to shear strain is known as modulus of
rigidity or shear modulus G. It is used to explain how a material resists a transverse
deformation in its elastic limit.

3.7  YIELD STRENGTH

At the yield point in the stress-strain graph, materials begin to have permanent (unre-
coverable) deformation. Some materials have well-defined yield points, whereas
some do not have. In the absence of a distinct yield point, a 0.2% offset is considered
to get an approximate yield point as shown in Figure 3.3 (a). The stress and strain
corresponding to the yield point are called yield stress and yield strain, respectively.
The yield point is crucial in steel structural design and analysis. At this stress, the
material specimen continues to elongate without an increase in stress. Sometimes, at
this point, a sudden decrease in stress is also observed. Thus, we get two yield points,
the upper yield point and the lower yield point. The corresponding stresses are called
upper yield stress and lower yield stress, respectively. The lower yield point is gener-
ally considered as the true characteristic yield point of a material.
The upper and lower yield points are shown for low carbon steel in Figure 3.3 (b).
The stress corresponding to the lower yield point is the accurate yield stress of the
material. The yield strength is defined in this case as the average stress at the lower
yield point.

FIGURE 3.3  (a) Yield strength using offset method and (b) upper and lower yield points for
low carbon steel.
Material and Section Properties 33

3.8  ULTIMATE STRENGTH

The maximum tensile load, compressive load, or shear load a material capable of car-
rying divided by the original cross-sectional area is called ultimate tensile strength,
ultimate compressive strength, or ultimate shear strength of the specimen.
In Figure 3.3 (b), the ultimate tensile strength of the material is pointed out.
Beyond the point of ultimate tensile strength, the material appears to strain soften
in the engineering stress-strain graph so that each increment of additional strain
requires smaller stress and necking also starts after this point.

3.9  MODULUS OF RUPTURE IN BENDING

A specimen will fail (rupture) at the point of maximum stress where it exceeds the
ultimate strength of the material. The modulus of rupture in bending, also known
as flexural strength or transverse rupture strength, or bend strength, gives the maxi-
mum load-carrying capacity right before the material breaks or yields. The modulus
of rupture is force per unit area, which is generally calculated by dividing the ulti-
mate bending moment a section can carry just before failure by the section modulus
of interest.

3.10  MODULUS OF RUPTURE IN TORSION

The modulus of rupture in torsion is the same as that for the modulus of rupture in
bending. In this case, the only difference is that, instead of bending moment, tor-
sional resistance needs to be considered, and it needs to be divided with torsional
section modulus.

3.11  POISSON’S RATIO

Poisson’s ratio is the ratio between two strains. It is defined as the ratio of transverse
strain to the corresponding axial strain on a material stressed along one axis. When
a bar is elongated, its cross-sectional area gets reduced. The change in diameter to
the original diameter is called the transverse or lateral strain. On the other hand, the
linear change in length to its original length is called the axial or longitudinal strain.

3.12  COEFFICIENT OF THERMAL EXPANSION

The coefficient of thermal extension is defined as the change in the original length
of the specimen due to a per degree Celsius rise in temperature. Mathematically it
can be expressed as:

∆L
α=
L0 × ∆T

where ∆L is the change in length, α is the thermal expansion coefficient, L0 is the

original length, and ∆T is the temperature change. For different materials, the
34 Introduction to Structural Analysis

coefficient of thermal expansion assumes different values. Specialist literature needs

to be studied to determine the coefficient of thermal expansion for the chosen mate-
rial. However, the abovementioned coefficient is also called the coefficient of lin-
ear thermal expansion. There are other two thermal expansion coefficients, surface
expansion (β ) and volume expansion (γ ). The former is defined as the change in area
due to a per degree change in temperature of the specimen, and the latter is defined
as the change in volume due to a per degree change in temperature. There exists a
relationship between these three coefficients for the same material specimen that is
given next:

β γ
α= =
2 3

3.13  ELASTIC ASSUMPTIONS

The deformation is said to be elastic if the external force produces a deformation
that disappears with the removal of the load. Thus, the elastic behavior implies the
absence of any permanent deformation.
The assumptions of linear elasticity are as follows:

1. The body is continuous, which means the whole-body volume is considered

to be filled with continuous matter without any void.
2. The body is perfectly elastic, which means the body should wholly obey
Hooke’s law of elasticity.
3. The body is homogeneous, which means the elastic properties are the same
throughout the body.
4. The body is isotropic, which means that the elastic properties of the body
remain the same in all directions.
5. Displacements and strains components are small.

3.14  STURCTURAL NONLINEARITY

Although our preliminary discussion based on elasticity grounds seems perfect, due
to some structural failure mechanism, we came to know that this phenomenon arises
due to the nonlinear properties inherent to the parent material used as structural
elements. Also, these failure modes cannot be analyzed or quantifiable based on
the elastic failure modes of the structures that we are accustomed to. Some impor-
tant engineering phenomena as listed next can only be accessed based on nonlinear
analysis:

1. Collapse or buckling of structures due to sudden overloads.

2. Progressive damage behavior due to long-lasting severe loads.
3. For certain structures (e.g., cables), nonlinear phenomena need to be
included in the analysis, even for service load calculations.
Material and Section Properties 35

Real-world behaviors are primarily nonlinear that cannot be obtained through

simplified linear models. In recent years, thus, the need for nonlinear analysis is
mainly due to the following reasons:

1. To use optimized structures.

2. To use new materials.
3. To address safety-related issues of structures more rigorously.

The sources of nonlinearity are due to multiple system properties, for example, mate-
rials, geometry, nonlinear loading, and constraints. Here are some examples:

1. Geometric nonlinearity: In this case, a structural component experiences

large deformation, which causes it to experience nonlinear behavior. A typi-
cal example is a fishing rod.
2. Material nonlinearity: Here, the component goes beyond the yield limit. As
a result, the stress-strain relationship becomes nonlinear, and the material
deforms permanently.
3. Contact: When two components come into contact, they can experience an
abrupt change in stiffness, resulting in localized material deformation at
the region.

To perform nonlinear analysis, we essentially need to adopt the following philosophy:

1. Stay with relatively small and reliable analytical models.

2. Perform a linear analysis first.
3. Refine the model by introducing nonlinearities as desired at various stages.

Interested students are encouraged to refer to various textbooks and references on

this subject to gain knowledge. This textbook is based on elastic analysis of the struc-
ture, and all stress and strain levels of the structure will be such that the material will
always be within the elastic limit.

3.15  CROSS-SECTIONAL AREA

A cross-sectional area represents an intersection of an object by a plane along or
across its axis. We usually take transverse cross sections while determining the
stress calculations. Cross-sectional areas depend on the geometry of the structural
element. A rectangular beam will have a rectangle cross section, whereas a circular
column will have a cross section like a circle in transverse direction.

3.16  CENTER OF GRAVITY AND CENTROID

The center of gravity (CG) is the point on an object through which its entire weight
acts in the direction of gravity, while the centroid is the geometric center of the
object. For isotropic and homogeneous objects with a regular geometric shape, the
36 Introduction to Structural Analysis

CG of weight and geometric center (i.e., centroid) coincide. For anisotropic objects,
the geometric center and weight CG do not coincide.

3.17  ELASTIC NEUTRAL AXIS

Let us first visualize the bending of a beam under any flexural loading (a load is called
flexural load when it is acting in a perpendicular direction to the longitudinal axis of
the beam, i.e., if the loading causes bending of a structure). So, under this loading, the
beam will deflect in the downward direction and become slightly curved, as shown
in Figure 3.4. The top fiber of the beam element will be in a state of compression,
whereas the bottom fiber of the beam will be in a state of tension. Thus, between these
two extreme fibers, there will be one location where the fibers are unaffected due to
this bending. That particular beam level where the net stress concentration is zero is
called the neutral plane, and the imaginary line drawn along this plane is called the
neutral axis of the beam. As we are dealing with structural bending within the elastic
limit, this neutral axis is sometimes also called the elastic neutral axis of the member
under study. For more detail on this, refer to Chapter 5 of this book.

FIGURE 3.4  Elastic deformation of beam under flexural loading and neutral axis.

3.18  SECOND MOMENT OF AREA AND RADIUS OF GYRATION

The second moment of area or moment of inertia is a term used to describe the
capacity of a cross section to resist rotation. It is always considered with respect
to a reference axis such as X–X or Y–Y, as shown in Figure 3.5. The reference axis
is usually the centroidal axis. It is a mathematical property of a section concerned
Material and Section Properties 37

FIGURE 3.5  Second moment of area or moment of inertia.

with a surface area and its distribution about the reference axis (axis of interest).
Mathematically, it is written as:

∫
I x − x =   y 2 dA
A

∫
I y − y = x 2 dA
A

where y is the distance from the X-axis to area dA and x is the distance from the
Y-axis to area dA.
For regular geometric shaped members, like a rectangle with width b and depth
d , the moment of inertia of the section is:

bd 3
Ix−x = 
12
The radius of Gyration or Gyradius is the several related measures of the size of an
object, a surface, or an ensemble of points. It is calculated as the root mean square
distance of the objects’ parts from either its CG or an axis.
In structural engineering, the two-dimensional radius of gyration is used to
describe the distribution of cross-sectional area in a beam around its centroidal axis.
It is the distance at which the entire area must be assumed to be concentrated so
that the product of the area and the square of this distance will equal the moment of
inertia of the actual area about the given axis. Here the radius of gyration is given by
the following formula:

Ix
kx =
A

where I x is the moment of inertia about the X-axis, and A is the area. If no axis is
specified, centroid axis is assumed. Here the radius of gyration is used to compare
38 Introduction to Structural Analysis

how various structural shapes will behave under compression along an axis. It is
used to predict buckling in a compression member or beam. It is used to describe the
cross-sectional area distribution in a column around its centroidal axis. If more area
is distributed further from the axis, it will offer greater resistance to buckling. The
most efficient column section to resist buckling is a circular pipe because it has its
area distributed as far away as possible from the centroid.

3.19  ELASTIC SECTION MODULUS

Elastic section modulus (S) is the ratio of moment of inertia of the cross section (I) to
the distance from the neutral axis to the most extreme fiber (y). So, it can be written
as S = I/y. For the rectangular section, the elastic section modulus of the section is:

I bd 2
Z =  = 
b /2 6

Section modulus is a geometric property of a given cross section and is used to

design beams or flexural members.
 4 Basic Concepts of
Generalized Coordinates,
Lagrangian, and
Hamiltonian Mechanics

4.1 INTRODUCTION AND CONCEPT OF

GENERALIZED COORDINATES
This chapter is independent and mostly has no direct connection with all other chap-
ters in this book. Variational principle as a subject is very interesting and has a wide
range of applicability in almost all fields of science and engineering. Hence, a short
introduction to this subject with few examples and some important theorems related
to this beautiful analytical technique has been provided here. Interested students
may find this chapter as a steppingstone toward more advanced mechanics and struc-
tural analysis books. Moreover, this technique has a wide range of applicability in
structural dynamics.
In this chapter, we will gain a brief insight into the vast area of mathematical
field called generalized coordinates and its subsequent development and applica-
tion in formulating Lagrangian and Hamiltonian mechanics. We are more famil-
iar with the Cartesian coordinates where three orthogonal axes (namely, x, y, z)
play the role of reference axes, and all other geometrical properties of particle or
objects are expressed with respect to these axes. Although Cartesian coordinates
or more specifically coordinate geometry helps us to formulate algebraic equa-
tions related to the orthogonal coordinates, sometimes it is found that all problems
of mechanics are not so simple or appropriate enough to express the geometri-
cal properties via only Cartesian coordinates. Also, it is found that a particular
mechanical problem, which has very complex structure in Cartesian coordinates to
express its equation of motion, becomes much easier if we use some other param-
eters that act as coordinates related to certain mechanical properties (like position
and momenta of the particle), the analysis of the system to understand its behavior
becomes much easier. There is also another advantage of using generalized coor-
dinates. Equations expressed in terms of generalized coordinates are applicable to
any coordinate system; one needs only to put appropriate values at the appropriate
place in the expression that is made up of generalized coordinates. In the following
section, we will develop and discuss some examples to see this powerful method
in action.

DOI: 10.1201/9781003081227-6 39
40 Introduction to Structural Analysis

4.2  CONCEPT OF CONFIGURATION SPACE AND PHASE SPACE

Configuration space is a set of geometric coordinates that define the configuration of
a system. If the system consists of a single-point particle, then configuration space
for that particle would be the three coordinates ( x , y, z ) that define its location in
three-dimensional space. Time is also taken as a parameter whenever we are dealing
with a dynamical system. So, for a static system of a single point, three coordinates
form the configuration space, whereas for dynamical system of a single point, three
Cartesian coordinates and a time parameter form the configuration space of the sys-
tem. So, in mathematical language, we can write:

For static system of single particle configuration space = ( x , y, z )

For dynamical system of single particle configuration space = ( x , y, z , t )

For a rigid body or a system consisting of many particles, the configuration space
consists of n-tuples forming the configuration space under study. So, for rigid body
or many particle systems, the configuration space can be expressed mathematically
as follows:

For static system of many particles = ( x n , yn , zn )

For dynamic system with many particle = ( x n , yn , zn , t )

So basically, configuration space is what we are most familiar with from our basic
introduction to coordinate geometry in secondary standard mathematics. However,
as it is evident from the mathematical forms, configuration space does not pro-
vide enough information about the nature of motion, deflection, and rotation of the
objects under study. To understand the nature of the object under application of force
or moment, we need to extend the configuration space to include more information
about the behavior of the system under application of forces and moments. When we
include some parameter in the configuration space to express the nature of motion of
the particle/particles under study, then we form phase space. A phase space of parti-
cles or particle is the space consisting of Cartesian coordinate as well as momentum
of the particle. So, the form of phase space coordinate is something like this:

Phase space for any dynamic system = ( x n , yn , zn , pxn , pyn , pzn , t )

For static system, there will be no momentum. Hence, phase space and configuration
space become same for static systems. Although we are writing x, y, z, it needs to
be understood that any set of parameters that define configuration of the system are
equally valid to be included as coordinates in phase or configuration space.
In the context of generalized coordinates, the corresponding momentum in such
cases is called generalized momentum. Generalized momentum might have differ-
ent form than mass times velocity, which we are most familiar with from our basic
concept of mechanics. The actual form and equation expressing the configuration
of a system is dependent upon the coordinate system that we are using to express
it. An example will be most welcome here to establish the concept of generalized
Generalized Coordinates, Lagrangian, and Hamiltonian Mechanics 41

FIGURE 4.1  Boyle’s law of gas – volume vs. pressure curve.

coordinates. From secondary school days, we have studied Boyle’s law that states
that when the temperature remains constant, the pressure and volume of a gas are
inversely proportional. Therefore, pressure times volume is constant, and what we
get from here is the rectangular hyperbola with the mutual orthogonal axes as pres-
sure and volume parameters of gas. P versus V diagram for gas obeying Boyle’s law
is shown in Figure 4.1 for better understanding.
In this plane, the coordinate of any point does not represent its physical position
in two-dimensional Cartesian plane, whereas in this plane, every point represents
the state of gas having this much pressure for this much volume and vice versa.
So, distance between two points in this plane has got no meaning in contrast to
usual distance between two points in Cartesian plane. However, this plane is most
appropriate for gases to express their state and nature of the change of state due
to the change in the pressure or volume parameters. So, we see that, it is not suit-
able to express the configuration of a system with usual Cartesian coordinates all
time. However, from this example, we can state that for the problem at hand, our
generalized coordinates for a gaseous system is the pressure and volume in P–V
plane. So, as stated earlier, generalized coordinates may not be coordinates at all
that we are familiar with from the concepts of coordinate geometry. The power of
generalized coordinates helps us to use parameter as coordinate of a system, which
is suitable to express its configuration under phase space or configuration space.
Phase space plays an important role toward formation of two most powerful tools
to analyze mechanical systems under different situation. So, in the next section, we
will study the formation of Lagrangian and Hamiltonian mechanics by the help of
phase space of the system.

4.3 INTRODUCTION TO LAGRANGIAN AND HAMILTONIAN

FORMULATION OF MECHANICS
From our earlier discussion, we can take the generalized coordinates for a particle
in phase space as (q, q, t) and try to form the equation of motion with this. The
Lagrangian of any system is defined as the difference between the kinetic and poten-
tial energy of the system under observation. Now, in this section, we will express the
42 Introduction to Structural Analysis

generalized coordinate by letter q and generalized velocity by q. Time parameter is

kept in its usual form t. So, Lagrangian of a system is the function:
Lagrangian of a system = L (q, q , t )
And the Euler-Lagrange equation of motion is defined as:

d  ∂L  ∂L
−  =0
dt  ∂q  ∂q

Typically, this equation is derived from the laws of calculus of variation. Euler’s
theorem on calculation of minima for a function of several variables has this form.
Interested readers may refer to any book mentioned in the reference for the same.
Above equation is not only applicable for mechanical systems but for all class of
problems where we are dealing with functions of several variables. More specifi-
cally, Lagrangian defined above is a functional whose arguments are functions. To
see how powerful the above equation is, let us consider the following problem of
basic Euclidian geometry and prove the assertion that the shortest distance between
any two points in plane is the straight line joining the two points.

Example 4.1: Prove that the shortest distance between two points

in a plane (two-dimensional for sake of ease of calculation) is the straight
line joining the two points.

SOLUTION:  We know from basic coordinate geometry that distance between

any two points ( x1, y 2 ) and ( x1, y 2 ) in two-dimensional Cartesian plane is given by:

s 2 =  ( x1 − x2 ) +  ( y1 −   y 2 )
2 2

When the two points are sufficiently close to each other, we can express the
above equation in differential form as follows:

ds 2 =  ( dx ) +  ( dy )
2 2

or,
2
 dy 
ds = dx 1+  
 dx 

Now, from the above Euler-Lagrange equation, we can choose the parameters of
Lagrangian (functional) as follows:

dy d d
L = ds; q = y , q = = y ;  =
dx dt dx
In this coordinate system, since the Lagrangian is not a direct function of y, we
have,

∂L
=0
∂y
Generalized Coordinates, Lagrangian, and Hamiltonian Mechanics 43

and,

∂y y
=
∂y 1+ y 2

Inserting all the above terms in the Euler-Lagrange equation, we get,

d  y 
    = 0
dx  1+ y 
2

which implies that,

 y 
  = constant = c
 1+ y 2 

So, this solution is only valid if and only if, y = a

where ‘a’ is another constant related to earlier constant by the following:

c
a=
1− c 2
So, from y = a, we get dy   = adx, which upon integration yields,

∫ dy = a∫ dx
or,

y   = ax + b

which is the equation of straight line.

So, from the above example, we can understand how powerful the Euler-
Lagrange equation is, and it is applicable for any problem involved finding out the
extremum of functionals irrespective of the dynamical systems.

Example 4.2:  Using Euler-Lagrange equation solve the equation of elastic

line and provide physical explanation of the solution.

SOLUTION:  Here we want to apply the Euler-Lagrange equation to a well-known

equation of an elastic line. From the concept of strength of materials, we know that
the elastic line equation for beams is in the following form:

d 2y M
=−
dx 2 EI

where y is the transverse direction in which deflection takes place, x is the axis
of beam, and M is the bending moment acting on the beam due to any external
transverse loading. EI is the product of elastic modulus of the beam and moment
of inertia of the beam cross section, respectively. We want to investigate this
44 Introduction to Structural Analysis

equation and our goal is to find the condition for maximum bending moment. So,
let us define the Lagrangian of the system under study as:

d 2y M
L= +
dx 2 EI

If the beam is subjected to a uniformly distributed force of intensity w, then bend-

ing moment from any section from distance x from its left support is given by:

wl wx 2
M= x−
2 2

So, with this expression inserted in the above equation, we get,

d 2y 1  wl wx 2 
L= +   x−  
dx 2
EI  2 2 

So, Lagrangian is function of both y  and  x. Hence, we need to solve this in

two ways. First, we will set:

d 2y M
L= −
dx 2 EI
with,

∂L
=0
∂y

And, Euler-Lagrange equation yields,

d  d 2y M 
  − =0
dx  dx 2 EI 

or, expanding the terms,

d 3y d  M 
−    = 0
dx 3 dx  EI 

or,

d 3y d  M  V 
=    =   (4.1)
dx 3 dx  EI   EI 

where V is the shear force acting in the transverse direction of the beam. In sec-
ond case with the substitution of the moment expression, we will get,

d  d 2y 1  wl wx 2  
  2 −  x− =0
dx  dx EI  2 2  
Generalized Coordinates, Lagrangian, and Hamiltonian Mechanics 45

or, completing the differentiation we get,

d 3y 1  wl wx   
−     −   = 0
dx 3 EI  2 2 

or,

d 3y 1  wl wx   
=     −  (4.2)
dx 3 EI  2 2 

Comparing equations (4.1) and (4.2), we get,

 wl wx   
V =   − 
 2 2 

which is precisely the expression for shear force induced in the beam due to
external uniformly applied load w on the beam in the transverse direction. Here
is the beauty of this method. We have neither analyzed the beam by drawing
cross section at a distance x from left support, nor have we drawn any induced
force acting at that section. But the equation helps us to derive the exact expres-
sion of the shear force at a distance x from the support. Same is applicable for
moment also. We can start by assuming that the induced shear force at a dis-
tance x is V , and from that, we will get the expression of moment in this section.
So, in summary, we can analyze the internal forces of beams without diving
into tidier details of shear force and bending moment diagrams, by applying
this method.

4.3.1 Hamilton’s Equation of Motion

From the above discussion on Lagrangian system, it is found that Lagrange’s equa-
tion of motion is second-order differential equation that imposes a tedious solving
process upon integration. So, to avoid difficulty, we can try to formulate mechani-
cal laws in first-order differential equations. That is what precisely we get using
Hamilton’s equation of motion. Hamilton’s equation of motion in generalized coor-
dinate is expressed as follows:
N

H =  ∑p q − L ( q, p, t )
i =1
i i

∂H
p i = −
∂qi

∂H
qi =  
 
∂ pi

In elementary terms, Hamiltonian function H is the sum of kinetic and potential

energy of a system or particle under study. So, the form of equation as written
above clearly indicates that we are having equations of motion involving first-order
46 Introduction to Structural Analysis

derivatives only. Although nothing new has been added in the overall mechanical
laws, Hamiltonian provides a more excellent way to solve equations of motion and a
profound understanding of the system’s dynamics that help build a more elegant way
of solving the problem.
We will not dive into much finer details with Hamilton’s equation of motion since
they are more relevant for understanding the dynamical system. Still, familiarity
with these laws helps one study a different advanced topic in mechanics. Interested
readers may consult the textbooks mentioned in the reference for a deeper under-
standing of these elegant laws.

4.4  CONCEPT OF SYMMETRY AND CONSERVATION LAWS

To this end, we will conclude the basic concept of conservation laws and their rela-
tion to symmetry of the system under study. As we have seen from Lagrange’s equa-
tion of motion, when Lagrangian is independent of q, then the equation reduces to,

d  ∂L 
=0
dt  ∂q 

or,

d
( p) = 0
dt

where the momentum of the particle is related to Lagrangian by the following

equation:

 ∂L 
p=
 ∂q 

So, the equation simply implies that when the Lagrangian is not function of one of
the coordinates then the corresponding momentum related to that coordinate axis
becomes conserved or constant of motion. This coordinate q is called ignorable or
cyclic coordinates. To this point, we quote few lines related to this phenomenon from
Goldstein’s book of classical mechanics:

If the generalized coordinate corresponding to a displacement is cyclic, it means that

translation of the system, as if rigid, has no effect on the problem. In other words, if
the system is invariant under translation along a given direction, the corresponding
linear momentum is conserved.

Thus, momentum conservation is directly related to the translational symmetry

of the system. The same is applicable for angular momentum conservation also.
Interested readers may refer to Goldstein’s book on classical mechanics Chapter 2 to
have a detailed understanding of the same.
 5 Equilibrium and
Support Reactions

5.1 INTRODUCTION
In this chapter, we will study various conditions regarding the equilibrium of struc-
tures. To establish the equations for the equilibrium of structures, we will also
investigate various types of supports and end conditions based on which the over-
all structural stability can be ensured. Equilibrium of various internal and external
forces and moments acting on the structures will also be explored.

5.2  EQUILIBRIUM OF STRUCTURES

Equilibrium of a structure or a particle is a state in which there is no net resul-
tant force acting on the structure. Hence, under the condition of static equilibrium,
there will be no motion due to the application of external forces and moments.
Several forces from various directions as well as moments can act on a structure. In
equilibrium there will be no net resultant force acting on the structure, effectively
imposing structural static condition. Although we are mostly dealing with statics
here, hence, topics like dynamic equilibrium or D’Alembert’s force law will not be
referred. So, our subject matter here is the static equilibrium of structure and its
various conditions. Following Newton’s law, when there is no resultant force acting
on the structure, then the acceleration of the structure will be a null vector or zero.
Following the same logic, the equilibrium condition in terms of vector equation can
be written as:
  
F = ma = 0

Expanding the vector equation along three mutually perpendicular axes ( x ,  y,  z ) ,
we can write:
   
Fx + Fy + Fz = 0

Here all the above vectors are mutually independent; hence, we can write:
 
Fx = 0
 
Fy = 0
 
Fz = 0

DOI: 10.1201/9781003081227-7 47
48 Introduction to Structural Analysis

So, we get the first condition of equilibrium. It simply says when the total force
acting on a system or a structure along three mutually perpendicular axes is indi-
vidually equal to the null vector or zero, the structure or system is said to be in the
state of equilibrium. Each component of force along three axes individually needs
to be zero; otherwise, along with the nonzero component, there will be acceleration
following Newton’s second law of motion and thereby disturbing the equilibrium of
the structure or the system. The above equation is also true for moment vectors. So,
under equilibrium condition, the net moment acting on a structure at any fixed point
of reference will be zero or null vector. Mathematically we can write this as:
   
Mx + M y + Mz = 0

So, the equilibrium condition tells us the following for moments as well:
 
Mx = 0
 
My = 0
 
Mz = 0

In summary, we can reinstate the equilibrium condition in a much more compact

form:
   
∑ F = 0; ∑ M =  0
i i

where i varies from 1 to 3, representing three mutually perpendicular axes under

consideration and it is very important to note that moment summation has been taken
with respect to a particularly chosen point. Moment due to all acting forces about
this chosen point need to be considered in the above equation and it should not be
such that moment about various points can make a body remain under equilibrium
condition. In the next section, we will investigate two important types of equilibrium
conditions, global and local equilibrium of structure and its elements. Expanding the
vector equation in terms of components, we can write the equilibrium conditions as:
 
Fx iˆ + Fy ˆj + Fz kˆ = 0 and, M x iˆ + M y ˆj + M z kˆ = 0

where iˆ,  ˆj , and k̂ are the unit vectors along three mutually orthogonal axes ( x ,  y,  z ),
respectively. Refer to Figure 5.1 for force vectors along three mutually orthogonal
axes for the ease of understanding.

5.2.1 Global and Local Equilibrium of Structures

We can apply the above equations for equilibrium on a structure as a whole or its
elements individually. The former case is called global equilibrium and the latter
case is known as the local equilibrium. This can be better understood with the fol-
lowing example. Let us consider a simple two-dimensional (2D) truss system with an
Equilibrium and Support Reactions 49

FIGURE 5.1  Force vector in a three-dimensional ( x ,  y,  z ) plane.

external load P acting on the structure as shown in Figure 5.2; though, we will learn
about truss in Chapter 7 in detail.
Now to calculate the support reactions at A and B due to the externally applied
load, we need to consider the equilibrium of the structure with the external load along
with the support reactions. The force diagram for the truss is shown in Figure 5.3
(for a list of support reactions depending on its different types refer to Section 5.6).
From the force diagram shown below, we can calculate the support reactions by
the application of equilibrium equations along two axes (since this is a 2D truss).
To determine the support reactions, we need to consider the total structure and its
geometry without paying attention to its internal members and their orientations. So,
to get support reactions we have to apply the concept of global equilibrium condition

FIGURE 5.2  Two-dimensional truss with external loading.

50 Introduction to Structural Analysis

FIGURE 5.3  Two-dimensional truss with external loading and support reactions – global
equilibrium.

FIGURE 5.4  Two-dimensional truss members with axial force – local equilibrium.

of the complete structure as a whole. After finding the support reactions, we need
to calculate the forces acting on each member of the truss. Now equilibrium condi-
tions need to be applied to each member to find out the forces acting on them. This
is the essence of the local equilibrium conditions. Thus, from the above discussion,
it is found that the global equilibrium condition is applied to the whole structure,
whereas local equilibrium condition is applied to each member comprising the struc-
ture separately to complete analyses. Individual truss members with the axial force
acting on them are shown in Figure 5.4 according to the present loading and support
condition for ease of understanding. After getting the values of support reactions
using global equilibrium conditions and thereafter by the application of equilibrium
equations for each member, we can easily calculate these axial forces. We will apply
these concepts in Chapter 7 while analyzing 2D and 3D trusses.
To understand the forces acting on a system and its resultant and subsequently
equilibrium condition, we need to carefully draw the free body diagrams, indicating
all acting forces on the system or structure. In the next section, we will investigate
the free body diagram of structures and will analyze the equilibrium conditions in
more detail.

5.3  FREE BODY DIAGRAMS

Free body diagrams are the most essential part of structural analysis. As stated
earlier, proper free body diagram with all acting forces at the proper location
Equilibrium and Support Reactions 51

allows us to analyze the structure effectively and correctly. As stated in the previ-
ous section, while analyzing the 2D truss, we have applied the concept of global
and local equilibrium conditions. Now, before we apply the equilibrium equations,
we must draw a force diagram or free body diagram of the truss by removing all
physical supports and showing support reactions at proper locations where they
are acting in the actual structure. So, the free body diagram enables us to draw
the force diagram as depicted in Figure 5.3 for a typical 2D truss. Without proper
free body diagram, we will not be able to analyze and determine the unknown
forces (like support reactions and member forces) acting on the structures. Also,
improper schematic of the free body diagram will lead to erroneous structural
analysis with faulty data.

5.4  SIGN CONVENTION

We will direct our attention toward the sign conventions that need to be consid-
ered while analyzing structures under equilibrium conditions. For support reactions
under global equilibrium, we usually consider the vertical reaction force as positive,
when it is directed toward the positive direction of y axis. Similarly, horizontal sup-
port reaction is taken as positive when it is directed toward the positive direction of
x axis or along positive direction of z axis. Moments, on the other hand, are taken
positive when they are directed in an anticlockwise direction. So, at the initial stage
of analysis, we consider all unknown forces and moments positive and continue
our analysis in this fashion. At the end, the algebraic signs of the calculated val-
ues indicate the proper directions of the support reactions and moments. A simple
example is appreciated here to understand the concept. Let us consider a simple
beam element under an axial pull P in the positive direction of x axis as shown in
Figure 5.5 (a).
Under this loading, the free body diagram under global equilibrium condition
may be drawn as shown in Figure 5.5 (b).
So as per the equilibrium condition along x axis, we can write:

Rx + P = 0

or,

Rx =   − P

Since there is no external force acting on the beam in y direction, hence, there will be
no support reaction in this direction. Applying equilibrium equation in y direction,
we get simply Ry = 0.
From the above solution, we get the magnitude of horizontal support reaction as
P and the algebraic sign in front of it indicates that the direction of the reaction force
is contrary to what we assumed prior to the analysis. So, with proper sign convention
and correct free body diagram, we will get the actual direction of reaction forces
after carrying out the analysis using principles of global equilibrium conditions.
52 Introduction to Structural Analysis

FIGURE 5.5  (a) Example problem of simple beam with horizontal load and (b) free body
diagram of simple beam with axial load.

But one point is needed to take care of very seriously. We cannot change or mix
our sign conventions while solving a problem. Whatever positive direction we have
assumed initially, it should remain the same while solving the problem until its
completion.
Now we will direct our attention toward the sign conventions for internal forces of
a structure under local equilibrium conditions. So, our aim is to understand the origin
of shear force and bending moments due to external loading acting on the structure.
In the case of a beam under transverse loading, if we take any section at a certain
distance from support then there will be internal forces and moments acting on that
section, which try to balance the effects produced by transverse external loading.
The resisting force induced at such arbitrary section is called shear force and the
resisting moment generated at the same section is called the bending moment. The
most commonly adopted positive directions of bending moment and shear force are
shown in Figure 5.6 for ease of understanding.
While analyzing any structure to determine the shear force and bending moments,
we will always use the diagram below or choosing the positive directions for shear
force and moment. In the following section, we will use these sign conventions all
along and we will get hands-on experience on its application.
Equilibrium and Support Reactions 53

FIGURE 5.6  Simply supported beam under transverse point load.

5.5  EXTERNAL AND INTERNAL FORCES

In this section, we will study different types of external and internal forces act-
ing on structural elements. By external force, we generally indicate the externally
applied loads acting on the structure as a whole or on a member. For example, in
the previous case of a 2D truss, we have seen that a point load is acting at node C
of the truss as a whole, and due to that load, we can calculate the support reactions
and member forces. So, the point load in this example is the external force and
forces generated on individual members due to this external force can be taken as
the internal forces.
But in any case, a structure needs to be in a state of equilibrium globally or
locally, as explained in the previous section. Normally for civil engineering struc-
tures, the external forces arises due to wind load, seismic load, impact load, self-
weight of the members comprising the structures, etc. Due to these loads, the internal
forces will be induced inside the structural members to maintain the equilibrium
condition without jeopardizing structural integrity. These induced forces are called
as internal reactions or member forces of the structures. To better understand the
effect of internal and external forces acting on structural elements, let us consider an
example. Let a simply supported beam be subjected to a uniformly distributed load
54 Introduction to Structural Analysis

FIGURE 5.7  (a) Simply supported beam under uniformly distributed load throughout its
length and (b) free body diagram of the beam up to section x–x from support A under flexural
loading.

w as shown in Figure 5.7 (a). Under the application of this load, let us consider a sec-
tion at a distance x from the left support ‘A’.
Now, as the beam is subjected to a uniformly distributed load, the free body dia-
gram of the beam up to section X − X will be as depicted in Figure 5.7 (b).
So, we have drawn one vertical force Vx and moment M x in the section shown
in Figure 5.7 (b). These two quantities are new, and where do they come from? To
understand this, let us first visualize the bending of a beam under any kind of flexural
loading (a load is called flexural load when it is acting in a perpendicular direction to
the longitudinal axis of a member, i.e., if the loading causes bending of a structure).
So, under this loading, the beam will be deflected in the downward direction and
will become slightly curved as shown in Figure 5.8 (a). The top fiber of the beam
element will be in a state of compression, whereas the bottom fiber of the beam will
be in a state of tension. If we consider a small element of beam under the externally
applied load P as shown in Figure 5.8 (b), then that particular section tries to move
the element toward its direction compared to the other parts of the beam, which are
connected with supports. Let us consider a section at a distance x from the left sup-
port with separated elements to understand the concept.
However, the actual movement of beam element under section p − q and m − n is
resisted by the other elements of the beam at its immediate vicinity. Now, in any case,
the section should remain under equilibrium condition. Hence, to ensure that, there
must be some forces that are needed to be developed at sections p − q and m − n,
which will counteract the external loading P applied on the element. This force that
Equilibrium and Support Reactions 55

FIGURE 5.8  (a) Deflection diagram of beam under flexural loading and (b) tendency of a
small element to move toward the applied loading on the beam.

is developed at these sections to counteract external load is called shear force. Now
let us draw the force diagram to justify the concept and form the local equilibrium
conditions, relating external load and shear force.
The positive direction of shear forces is maintained as per the sign convention
discussed in the previous section.
Applying local equilibrium condition for section x for 0 ≤ x ≤ l /2, we can write:

Vx − RA = 0

or,

Vx = RA

The free body diagram is shown in Figure 5.9 (a).

For the range l /2 ≤ x ≤ l , we need to consider Figure 5.9 (b) for the free body dia-
gram to form the force equilibrium equation:

Vx − RA + P = 0

or,

Vx = RA − P
56 Introduction to Structural Analysis

FIGURE 5.9  (a) Free body diagram of beam section with induced shear force at an arbi-
trary section 0 ≤ x ≤ l /2, (b) free body diagram of beam section with induced shear force for
l /2 ≤ x ≤ l , (c) free body diagram of beam for bending moment in 0 ≤ x ≤ 1/2, and (d) free
body diagram of beam for bending moment in l /2 ≤ x ≤ l.

Since all the terms in the right-hand side of the above equation is known, hence, from
this expression, we can calculate the value of shear force within the range l /2 ≤ x ≤ l.
The free body diagram is shown in Figure 5.9 (b).
An interesting point is that shear force changes its sign at the point x = l /2, which
is left as an exercise for the readers.
Now as discussed above for shear force, we can understand the formation of bend-
ing moments at the arbitrary section x from left support. To understand this concept,
let us consider the Figure 5.9 (c) for the free body diagram with bending moments
drawn at its positive orientation.
Now from local equilibrium condition, we can write for the range 0 ≤ x ≤ l /2:

M x −   RA x = 0

or,

M x =   RA x

Since support reactions are already calculated from global equilibrium conditions at
the beginning of the analysis, hence from the above expression, we can calculate the
bending moment by varying x from 0 to l /2.
Equilibrium and Support Reactions 57

To get the bending moment for the range l /2 ≤ x ≤ l , we have to draw another sec-
tion by increasing the range of x. We can do that by taking the section beyond the
point l /2 and drawing the free body diagram as shown in Figure 5.9 (d). So, from this
figure, we get after applying the local equilibrium condition, for the range l /2 ≤ x ≤ l:

l
M x − RA x + P  x −  = 0
 2
or,
l
M x = RA x − P  x − 
 2

So, from the above expression, we get the bending moments for rest of the portion
of the beam under the applied loading. From these two equations, it can be seen that
bending moment value is the same at a distance l /2 from the support, which is left
as an exercise for the readers. It is to be noted that, though the shear force and bend-
ing moment are shown separately in Fig. 5.9 for the sake of understanding, they get
induced simultaneously at a particular section.
The above concept of analysis can be applied to any type of lodging for beams by
carefully drawing free body diagrams with forces and proper sign conventions. We
will always get the above sets of algebraic equations, and solving the same, we not
only get the magnitudes but also the direction of the shear and bending moments. A
thorough understanding of the above method of analysis is the main building block
to cover more advanced topics on this subject.
Now we have understood how the internal forces/internal reactions/member forces
generate due to the application of external loads. The internal reactions or member
forces can be of various types like axial force, bending moment, shear force, and
torsional moment depending upon the type of members and nature of external loads
in the structures. For example, if it is a member of a pin jointed structure or truss, the
member force or the internal reaction will be generated solely as an axial force as
these types of structures are loaded only at the joints. The member forces generated
in a truss member is shown in Figure 5.10. In this figure, Fx and Fy are the external

FIGURE 5.10  Member forces in pin jointed structures.

58 Introduction to Structural Analysis

FIGURE 5.11  Member forces in rigid jointed structures for two dimensions: (a) for beams
(b) for frames.

reaction forces at the pin joints. In case of rigid jointed structures, the generated
member forces at a particular section (Fx, Fy, Mz) are shown in Figure 5.11 (a) and
(b) for two dimensions. As the whole if the structure is in equilibrium, equal and
opposite internal reactions or member forces generates at each cut section, as shown
in Figure 5.11 (b).

5.6  TYPES OF SUPPORTS FOR STRUCTURES

Depending on the situation, a structure might have different support conditions. For
example, beams and columns in a building frame may be fixed at the support or
maybe simply supported. For civil engineering structures, the most common sup-
port types are listed in Table 5.1 with the respective support reaction forces. These
support reaction forces are basically external reactions of the structures. A thor-
ough understanding of these supports and support reactions is essential for analyz-
ing different types of structures and their behavior under the application of external
loading. Students are advised to go through and understand this table for all future
analysis of structures that will be followed.

5.7  RELEASE OF INTERNAL REACTIONS OR MEMBER FORCES

Depending on the situation, it becomes necessary to remove or release any internal
reaction forces or member forces such as axial force (AF), shear force (SF), bending
moment (BM), or torsional moment (TM) from a specified location of a structural
member. Some special types of joints are introduced in those particular locations of
a structural member to release these internal reactions, as discussed below:

5.7.1 Releasing Bending Moment

To release the bending moment in a specified location, internal hinge is provided.
The mechanism of introducing an internal hinge is shown in Figure 5.12 (a). By intro-
ducing an internal hinge, one can reduce bending moment at a particular section, or
Equilibrium and Support Reactions 59

TABLE 5.1
Different Types of Supports and Support Reactions
External Support
Type of Symbolic Reactions in 2D and
Support Representation Reactions in 2D 3D
Fixed In 2D
Three
(Rx , Ry , M z )
In 3D
Six
(Rx , Ry , Rz , M x , M y , M z)
Hinged In 2D
Two
(Rx , Ry )
In 3D
Three
(Rx , Ry , Rz)
Roller In 2D
One
(Ry  or  R )
In 3D
One
(Ry  or  R )
Rocker In 2D
One
(Ry  or  R )
In 3D
One (Ry   or  R)

Link In 2D
One
(R)
In 3D
One (R)
Horizontal In 2D
guided roller Two
(Ry , M z )
In 3D
Two
(Ry , M )
Vertical guided In 2D
roller Two
(Ry , M z )
In 3D
Two
(R, M )
60 Introduction to Structural Analysis

FIGURE 5.12  Release of internal reactions introducing special type of joints.

one can break the length of span of a very long beam so that the overall deflection
and bending moment can be reduced. Suppose rr is the number of released reac-
tion for a particular special type of joint. So, for the first case of joint, i.e., the case
shown in Figure 5.12 (a), only bending moment is released. Hence, rr = 1. Now for 3D
case, if internal hinge is provided, the number of released reaction would be 3, i.e.,
Rx ≠ 0,  Ry ≠ 0,  Rz ≠ 0,  M x = 0,  M y = 0,  M z = 0. So if we can formulate the number
of released reactions (rr ) in two and three dimensions, it will look like the following
two equations, respectively.

rr = ∑( m − 1); in 2D
*

rr = 3 ∑( m − 1); in 3D
*

where m*= number of members connected to the internal hinge.

Example 5.1:  Find the number of released reactions for the 2D framed
structure as shown in Figure 5.13.

SOLUTION:  Internal hinge is provided in joint D, H, and J in the above frame. At

joint D, three members (DA, DE, and DG) are connected, at joint H four members
(HE, HG, HI, and HK) are connected, and at joint J two members (JG and JK) are
connected.
( )
No. of released reactions are, rr = ∑ m* − 1 = (3 − 1) + (4 − 1) + (2 − 1) = 6 .
Equilibrium and Support Reactions 61

FIGURE 5.13  Example problem on released reaction.

5.7.2 Releasing Shear Force

Vertical guided roller is introduced to release the internal shear force in a beam as
shown in Figure 5.12 (b). Here bending moment and axial force remain as they are.
As only the shear force becomes zero, here the number of released reaction, rr is one.

5.7.3 Releasing Axial Force

Horizontal guided roller is introduced to release the internal axial force in a beam as
shown in Figure 5.12 (c). Here bending moment and shear force remain as they are.
As only the axial force gets released, here the number of released reaction, rr is one.

5.7.4 Releasing Axial Force and Bending Moment

Horizontal roller is introduced to release the internal axial force and bending moment
in a beam as shown in Figure 5.12 (d). Here the shear force remains as they are as
the beam cannot move in vertical direction along position C. As the axial force and
bending moment get released, here the number of released reaction, rr is two.

5.7.5 Releasing Shear Force and Bending Moment

Vertical roller is introduced to release the internal shear force and bending moment
in a beam as shown in Figure 5.12 (e). Here axial force remains as they are as the
beam cannot move in horizontal direction at location C. As the shear force and bend-
ing moment get released, here the number of released reaction, rr is two.
 6 Indeterminacy and
Stability of Structure

6.1 INTRODUCTION
Indeterminacy and structural stability are discussed in detail with lots of solved
example problems. Indeterminacy has been divided into two parts: static and kine-
matic indeterminacy. The principle of superposition has been discussed at the end.
As can be seen by going through this book, structural analysis procedures vary com-
pletely depending on the degree of indeterminacy. This chapter is the cornerstone of
structural analysis and a must-read for all whoever wants to master the theory and
analysis of this subject.

6.2  STRUCTURAL INDETERMINACY

First, let us understand the indeterminacy or degree of indeterminacy. For that, let us
consider two sets of algebraic equations as follow:

Set − 1 Set − 2
7 x + 5 y = 10 and 7 x + 5 y + 8 z = 10
10 x − y = 20 10 x − y − 4 z = 20

The first set of equations is solvable because the number of unknowns and equations
are the same. From this set, the value of x and y comes out to be 1.93 and −0.702,
respectively. As we can determine the values of unknowns from this set, this is a
determinate set. But for the second set, the number of unknowns and equations is
not the same. For this reason, the values of the unknowns cannot be determined
from this set. This set is thus called an indeterminate set. The same concept is appli-
cable in the structural analysis also. Here the unknown components are the forces
in a given structure or member and available independent degrees of freedom of a
particular structure. So, depending on the type of unknowns, the degree of struc-
tural indeterminacy can be classified into two types: (a) static indeterminacy (Ds )
and (b)  kinematic indeterminacy (Dk ). The static indeterminacy is related to the
unknown forces in a given structure, and kinematic indeterminacy is associated with
the available independent unknown degrees of freedom of a particular structure.

6.3  STATIC INDETERMINACY AND STABILITY

When we observe most of the engineering structures around us, we may notice that
multiple vertical members support one horizontal member. These different structural
parts form a complex structural system separate from the simply supported beams and
DOI: 10.1201/9781003081227-8 63
64 Introduction to Structural Analysis

structural elements. Statically indeterminate structures or redundant structures are those

structures that cannot be analyzed with the help of equilibrium equations of statics only.
Some additional compatibility equations are required to solve them entirely. There are a
total of three static equations of equilibrium in two dimensions, i.e., ∑ Fx = 0, ∑ Fy = 0,
and ∑ M z = 0 and in three dimensions, there are six, i.e., ∑ Fx = 0, ∑ Fy = 0, ∑ Fz = 0,
∑ M x = 0, ∑ M y = 0, and ∑ M z = 0. Based on the type of reaction forces of a structure,
static indeterminacy can be divided into two types: (a) external static indeterminacy
(Dse ) and (b) internal static indeterminacy (Dsi ). The external static indeterminacy is
related to the unknown external support reactions, and the internal static indeterminacy
is about unknown internal reactions or member forces. We can write:
Ds = Dse + Dsi
or,

Ds = ( re − no. of  equilibrium solutions )

+ {3m − ( 3 j + rr − no. of  equilibrium solutions )}
or,

Ds = ( re − 3) + {3m − ( 3 j + rr − 3)}; in two − dimensional ( 2D )

Similarly
Ds = ( re − 6 ) + {6m − ( 6 j + rr − 6 )}; in three − dimensional ( 3D )

6.3.1 Static Indeterminacy of Rigid Structures

The structures, the members of which are connected to rigid or moment-resisting
joints, are called rigid structures. For these types of structures, external loads can
be applied both on the members and joints of the structure. In each member, there
will be three internal reaction forces or member forces (Fx ,  Fy , and  M z) for a 2D
case, but for three dimensions, in each member, there will be six internal member
forces (Fx ,  Fy ,  Fz ,  M x ,  M y  and  M z). These internal member forces are considered
as unknowns. The other set of unknowns arrives from the external support reactions.
The known quantities are the equations of equilibrium considered at each joint and
the number of released reactions, as discussed in Chapter 5. Now the static indeter-
minacy in two dimensions can be written as:
Ds = No.  of   unknown − No.  of   known
or,
Ds = ( re + 3m ) − (3 j + rr ); in two dimensions.
and,
Ds = ( re + 6m ) − (6 j + rr ); in three dimensions
where re is the no. of external support reactions, m is the no. of members, j is the no.
of joints, and rr is the no. of released reactions in the structure.
Indeterminacy and Stability of Structure 65

FIGURE 6.1  Example problem on the degree of static indeterminacy.

Example 6.1:  Find the static indeterminacy (Ds ) of the frame as shown
in Figure 6.1 for the given support conditions.

SOLUTION:  Ds = No. of unknowns − No. of knowns

Ds = ( re + 3m) − ( 3 j + rr )

Finding no. of external reactions (re ):

At support A = 3 (fixed support)
At support B = 1 (roller support)
At support C = 2 (hinged support)
At support D = 2 (hinged support)
At support E = 2 (hinged support)
At support F = 2 (horizontally guided roller)
At support G = 2 (vertically guided roller)
Total no. of external reactions (re ) = 3 + 1+ 2 + 2 + 2 + 2 + 2 = 14
No. of members (m) in the rigid frame = 13
Total no. of unknowns, (re + 3m) = (14 + 3 × 13) = 53
Total no. of knowns, (3 j + rr ) = (3 × 14 + 0) = 42
Therefore, the degree of static indeterminacy for this 2D rigid frame,
Ds = 53 − 42 = 11. That means, to analyze this frame fully, extra 11 equations are
required.

Example 6.2: Find the static indeterminacy (Ds ) of the 3D frame as

shown in Figure 6.2 for the given support conditions.

SOLUTION:  Total no. of unknowns, (re + 6m) = (6 + 1+ 3 + 1+ 3 + 1) + (6 × 13) = 93

Total no. of knowns, (6 j + rr ) = (6 × 12) + 0 = 72
Therefore, the degree of static indeterminacy for this 3D rigid frame,
Ds = 93 − 72 = 21. That means, to analyze this frame fully, extra 21 equations are
required.
66 Introduction to Structural Analysis

FIGURE 6.2  Example problem on the degree of static indeterminacy.

Example 6.3: Find the static indeterminacy (Ds ) of the 2D frame as

shown in Figure 6.3 for the given support conditions and internal
hinges.

SOLUTION:  Total no. of unknowns, ( re + 3m) = (3 + 2 + 2 + 2) + (3 × 14) = 51

Internal hinge is provided at joint I, F, and K.
No. of released reactions (rr ) due to the internal hinges = ∑(m * −1) = (2 − 1) +
(4 − 1) + (3 − 1) = 6; where m * = number of members connected to the internal
hinge.
Total no. of knowns, (3 j + rr ) = (3 × 12) + 6 = 42
Therefore, the degree of static indeterminacy for this 2D rigid frame,
Ds = 51− 42 = 9. That means, to analyze this frame fully, extra 9 equations are
required.

FIGURE 6.3  Example problem on the degree of static indeterminacy.

Indeterminacy and Stability of Structure 67

FIGURE 6.4  Example problem on the degree of static indeterminacy.

Example 6.4: Find the static indeterminacy (Ds ) of the 2D frame as

shown in Figure 6.4 for the given support conditions.

SOLUTION:  Total no. of unknowns, (re + 3m) = (3 + 2 + 2 + 3) + (3 × 20) = 70

Total no. of knowns, (3 j + rr ) = (3 × 18) + 7 = 61
Therefore, the degree of static indeterminacy for this 2D rigid frame,
Ds = 70 − 61 = 9. That means, to analyze this frame fully, extra 9 equations are
required.

Example 6.5:  Find the static indeterminacy (Ds ) of the 2D frame as shown
in Figure 6.5 for the given support conditions.

FIGURE 6.5  Example problem on the degree of static indeterminacy.

68 Introduction to Structural Analysis

SOLUTION:  Total no. of unknowns, (re + 3m) = (3 + 2 + 2 + 1) + (3 × 15) = 53

Total no. of knowns, (3 j + rr ) = (3 × 13) + 0 = 39
Therefore, the degree of static indeterminacy for this 2D rigid frame,
Ds = 53 − 39 = 14. That means, to analyze this frame fully, extra 14 equations are
required.

Example 6.6:  Find the static indeterminacy (Ds ) of the frame as shown
in Figure 6.6 for the given support conditions and internal hinges.

SOLUTION:  Total no. of unknowns, (re + 6m) = (6 + 3 + 3 + 1+ 1+ 1) + (6 × 13) = 93

Internal hinge is provided at joint B, J, and in-between member DH.
No. of released reactions (rr ) due to the internal hinges = 3 ∑ ( m * −1) =
{ }
3 × ( 3 − 1) + ( 3 − 1) + ( 2 − 1) = 15; where m * is the number of members con-
nected to the internal hinge.
Total no. of knowns, (6 j + rr ) = (6 × 12) + 15 = 87
Therefore, the degree of static indeterminacy for this 2D rigid frame,
Ds = 93 − 87 = 6. That means, to analyze this frame fully, extra 6 equations are
required.

FIGURE 6.6  Example problem on the degree of static indeterminacy.

6.3.1.1 Shortcut Method for Determining Internal Static

Indeterminacy of Rigid Structures
We have already found the values of external and internal static indeterminacies 2D
and 3D rigid frames as follows:

Dse = ( re − 3); in 2D

Dse = ( re − 6 ); in 3D
Indeterminacy and Stability of Structure 69

FIGURE 6.7  Closed-loop formations in rigid frames.

and,

Dsi = {3m − ( 3 j + rr − 3)}; in 2D

Dsi = {6m − ( 6 j + rr − 6 )}; in 3D

From the abovementioned equations, we can see that finding external static indeter-
minacy is quite simple and straightforward. But finding the internal static indeter-
minacy is quite hectic. Chances of committing mistakes are quite high if sufficient
attention is not paid. For this reason, an alternative shortcut method is developed
to find the internal static indeterminacy of rigid frames. By this method, one can
determine the internal static indeterminacy simply looking at the given frame.
The internal static indeterminacy can be obtained quite easily by the following
equations:

Dsi = 3c − rr ; for 2D frames

Dsi = 6c − rr ; for 3D frames

where c is the number of closed loops formed in the given frame. To understand
the concept clearly, let us consider the rigid frame as given in Figure 6.7. Here one
closed loop is formed by the members BC, CD, DE, and EB. So, internal static inde-
terminacy according to the abovementioned equation, Dsi = 3 × 1 − 0 = 3. Readers
are encouraged to cross check the solution of the example problems 6.2–6.6 by this
shortcut method again. Closed loop exists only in frames, not in beams or trusses.

Example 6.7:  Find the static indeterminacy (Ds ) of the frame as shown
in Figure 6.1 for the given support conditions and internal hinges by the
technique of closed-loop formation.

SOLUTION:  Dse = 14 − 3 = 11; Dsi = 3c − rr = (3 × 0) − 0 = 0 ; as there is no closed-

loop formation (c = 0) in this problem. Therefore, Ds = 11+ 0 = 11. As we can see,
it matches the earlier solution.
70 Introduction to Structural Analysis

Example 6.8:  Find the static indeterminacy (Ds ) of the frames as shown
in Figure 6.8 for the given support conditions and internal hinges by the
technique of closed-loop formation.

SOLUTION:  Case: a

Dse = ( re − 3) =  3 − 3  =  0; Dsi = 3c − rr =  3 × 0 − 1  = −1

Ds = Dse + Dsi =  0 − 1  = −1
 

The frame is statically determinate but unstable. When Ds is negative. The given
structure becomes unstable. We will learn about unstable structures in this chap-
ter shortly.
Case: b

Dse = (re − 3) =  4 − 3  =  1; Dsi = 3c − rr =  3 × 2 − 1 

=  5 (here two closed − loop forms).

Ds = Dse + Dsi =  1  +  5  =  6

This frame is statically indeterminate to the 6th degree.

FIGURE 6.8  Example problem on the degree of static indeterminacy by closed-loop forma-
tions technique.

6.3.1.2 Degree of Statical Indeterminacy When Load is Applied

to the Structure
External loads are imposed on structures. Depending on the loading magnitude and
direction, the reaction forces are generated accordingly. In Figure 6.9, the nature
of support reactions is shown depending on the external load for a fixed beam. In
the first case, i.e., Figure 6.9 (a), we need to consider three equilibrium conditions
(∑ Fx = 0, ∑ Fy = 0, and ∑ M z = 0) as the inclined load is applied that produces two
components during load application. But for the second case, i.e., Figure 6.9 (b), as
only a vertical load is applied, there is no need to consider ∑ Fx = 0. As a result, for
Indeterminacy and Stability of Structure 71

FIGURE 6.9  Nature of support reaction depending on external loads.

the second case, the effective number of equilibrium equations reduces down to two
(∑ Fy = 0 and ∑ M z = 0 ).

Example 6.9: Find the static indeterminacy (Ds ) of the fixed beam as

shown in Figure 6.10 for the given load conditions.

SOLUTION: 

Dse = (re − 2) = (5 − 2) = 3

Dsi = {3m − (3 j + rr − 3)} = {3 × 2 − (3 × 3 + 0 − 3)} =  0

 

Ds =  3  +  0  =  3
 

In this problem, two support reactions were considered at each fixed end, and two
equations of equilibrium were considered.

FIGURE 6.10  Example problem on finding static indeterminacy when external loads are
applied.
72 Introduction to Structural Analysis

6.3.2 Static Indeterminacy of Pin-Jointed Structures

Pin-jointed structures or trusses are constructed by joining multiple members by
frictionless pin joints. Here the structural loads are only allowed to apply to the joints
of the trusses. As the members are connected through pin joints, no reactive moment
is allowed to accumulate in these structures, i.e., the bending moments at joints
turn out to be zero. That is why for pin-jointed structures, the moment equilibrium
becomes unnecessary, and thus we need to consider the remaining two equilibrium
conditions in each joint for the 2D case. For the 3D case, we need to consider three
equilibrium conditions per joint.
Moreover, as the external loads are only allowed to apply to the joints, only axial
force generates as the member force. We will study trusses in more detail in Chapter 7.
However, in this chapter, we will focus on its degree of static indeterminacy only.
So, for the pin-jointed structures, the member forces (m) and external support
reactions (re ) are considered as unknowns, and the number of equilibrium equations
at each joint (j) is the knowns.
Therefore,

Ds = No.  of   unknown − No.  of   known

Ds = ( m + re ) − 2 j; for 2D

Ds = ( m + re ) − 3 j; for 3D

Again, we can write, Ds = Dse + Dsi ; where external static indeterminacy,

Dse = ( re − 3); for 2D

Dse = ( re − 6 ); for 3D

and internal static indeterminacy,

Dsi = m − (2 j − 3); for 2D

Dsi = m − (3 j − 6); for 3D

If we add these Dse and Dsi , we will obtain the value of the Ds given above for 2D
and 3D cases.

Example 6.10:  Find the degree of static indeterminacy for the following
pin-jointed structure as shown in Figure 6.11.

SOLUTION: 

Ds = ( m + re ) − 2 j =  ( 9  +  3) − 2 × 6  =  0

The degree of static indeterminacy of this pin-jointed structure is 0 or the structure

is statically determinate.
Indeterminacy and Stability of Structure 73

FIGURE 6.11  Example problem on the degree of static indeterminacy of pin-jointed

structures.

6.3.2.1 Shortcut Method for Determining Internal Static

Indeterminacy of Pin-Jointed Structures
Like the shortcut method to determine the internal static indeterminacy of the rigid
structures, we can find the same here also. Only visual inspection is required to
implement this method. The internal static indeterminacy of a pin-jointed truss
depends on the triangle formation in its geometry. For example, consider Figure 6.12,
where the degree of internal indeterminacy is found based on its geometric shape,
i.e., triangle formations. Consider Figure 6.12 (a); here, the entire pin-jointed truss
consisting of two nonoverlapping triangles. We cannot make any more nonoverlap-
ping triangles from this geometry. So, in this case, the internal static indeterminacy
is zero. In Figure 6.12 (b), we can observe one overlapping triangle is there. We
can eliminate this triangle just by removing one member from this truss. So, this

FIGURE 6.12  Finding internal static indeterminacy of pin-jointed structures by shortcut

method.
74 Introduction to Structural Analysis

structure is internally indeterminate to degree one. In Figure 6.12 (c), there is no

triangle formed. One additional member is required to make two nonoverlapping
triangles from this geometry. So this structure has internal static indeterminacy of
degree −1. That means this geometry is internally unstable, or it will form a mecha-
nism. The truss shown in Figure 6.12 (d) is also internally statically determinate.

Example 6.11:  Find the degree of static indeterminacy for the following
pin-jointed structure as shown in Figure 6.13.

SOLUTION:  Dse = ( re − 3) = 3 − 3 = 0; Dsi = −2; here there is a deficiency of two

members as shown in dotted lines in Figure 6.13.
Ds = 0 − 2 = −2; this structure is statically unstable. If we observe the pin-jointed
structure carefully, we can also find that it will form mechanism.

FIGURE 6.13  Example problem on degree of static indeterminacy for pin-jointed structure.

6.3.3 External and Internal Stability of Structures

A structure should be stable enough to serve its intended purpose i.e., it should not
undergo rigid body motion under the influence of external loads, or it should not form
any mechanism. Stability can be of two types: (a) external stability and (b) internal
stability. External stability is related to the external support conditions and internal
stability is related to the shape of the structure.

6.3.3.1  External Stability of Structures

Let us take the example of a simply supported beam. For this beam, the total unknown
support reactions or force of constraints are three. Since this is a 2D structure, we
have three equations of equilibrium at our disposal. Namely, ∑ Fx = 0, ∑ Fy = 0,
∑ M z = 0 . So, the difference between unknown reactions and available equations is
3 − 3 = 0 . Hence, the structure is statically determinate.
In the case of a propped cantilever, it has three unknown support reactions
at the fixed and one unknown support reaction at the pinned roller end. Hence,
the total unknown support reaction is 3 + 1 = 4 and available statical equilibrium
Indeterminacy and Stability of Structure 75

equation is 3. Hence, the difference between them is 4 − 3 = 1. This means that

the structure is statically indeterminate and there will be an additional equation
that needs to be formed to completely analyze the structure. Indeterminacy related
to support reactions only is called external indeterminacy. The examples we have
provided above are all related to external indeterminacy. If external indeterminacy
is positive, the structure is externally stable and if it is negative, the structure is
externally unstable. Let us consider a beam supported at both ends by rollers. In
this case, the only constraint force or unknown support reactions are vertical reac-
tions at support. Now the available statical equilibrium, equations are three for
2D structures. Hence, the degree of indeterminacy in this case is 2 − 3 =   −1  < 0.
Since the structure is supported on two rollers, hence it will slide along horizontal
direction freely and is unstable. Hence, as we have stated, in the negative degree of
indeterminacy structure becomes unstable. From a practical point of view, exter-
nally unstable structures need to be avoided. Hence, knowledge of external deter-
minacy and its relation to the stability of the structure is very much important for
all. Structural engineers and others too should have a clear idea about these salient
features of structural analysis prior to start a career in structural engineering. For
external stability of trusses, all support reaction components should not be paral-
lel to each other, and all support reactions should not be concurrent, i.e., pass-
ing through the same point. See Figure 6.14 for stable and unstable trusses as an
example of external instability.
In Figure 6.14 (a), the truss is supported on the roller at both ends; hence, only
unknown support reactions are vertical ones as shown. As both these reactions are
parallel to each other, hence, the structure is unstable. For the truss in Figure 6.14 (b),
the support reactions are shown with arrows and not all of them are parallel to each
other. Hence, the truss is a stable one.

FIGURE 6.14  Example of (a) externally unstable truss (b) externally stable truss.

6.3.3.2  Internal Stability of Structures

A structure is internally stable, if it maintains its shape and remains a rigid body
even after detaching from its supports. Conversely, a structure is internally unstable
or nonrigid, if it undergoes large displacements under small disturbances when not
76 Introduction to Structural Analysis

FIGURE 6.15  Examples of internally stable structures.

supported externally. In the second case, the structure forms mechanism under the
influence of small disturbance. Some examples of internally stable and unstable
structures are shown in Figures 6.15 and 6.16, respectively.
Note that in Figure 6.15, each of the structures maintains its original shape even
after removing from their supports and moves as a rigid body. But the structures
in Figure 6.16 are consisting of two rigid parts ‘AB’ and ‘BC’ that are connected
by a hinge at location ‘B’. If removed from the supports, these two rigid parts will
start rotating relative to each other about the hinge ‘B’ under the influence of a
small disturbance and thus forms mechanism. The second set of structures are
internally unstable or nonrigid in nature. In reality, every engineering structure
undergoes deformation when external loads are applied. But for rigid structures,
these deformations are so small that those can be neglected. Rigid structures
offer greater resistance against their shape changes whereas the nonrigid struc-
tures offer negligible resistance and deform easily when the external supports are
removed. They can even collapse under their own weight when in unsupported
condition.

FIGURE 6.16  Examples of internally unstable structures.

Indeterminacy and Stability of Structure 77

6.4  KINEMATIC INDETERMINACY OF STRUCTURES

When loads are applied, a structure undergoes deformation. The kinematic indeter-
minacy or degrees of freedom of a structure, in general, are defined as the indepen-
dent joint displacements (translations or rotations) necessary to specify the deformed
shape of the structure when subjected to an arbitrary loading. In another way, we
can say kinematic indeterminacy is the total number of unrestrained displacement
(translations or rotations) components at the joint of that structure. If the displace-
ment component of the joint cannot be determined by the compatibility (related
to the shape of the structure) equation, then the structure is called kinematically
indeterminate.
In Figure 6.17, the deformed and undeformed shapes of a pin-jointed structure are
shown in two dimensions under the influence of external loads. The joint A is hinged
supported and joint B is on roller. Joint A cannot move in any direction and joint B
can only slide along X-direction. The other joints (C and D) of the truss are free and
can move in both directions. After deformation took place, suppose joint B moves
d1 distance along X-direction. Likewise, joint C moves d 2, and d3 distance along
X- and Y-directions, respectively, and joint D moves d 4 , and d5 distance along X- and

FIGURE 6.17  Degrees of freedom of a 2D pin-jointed structure.

78 Introduction to Structural Analysis

TABLE 6.1
Types of Degrees of Freedom as Per Joint Condition
Types of Joint No. of Degrees of Freedom Types of Degrees of Freedom
2D truss joint 2 ∆x , ∆y
3D truss joint 3 ∆x , ∆y, ∆z
2D beam/frame joint 3 ∆x , ∆y,θ z
3D beam/frame joint 6 ∆x , ∆y, ∆z ,θ x ,θ y ,θ z

Y-directions, respectively. So, as per the definition above, the degrees of freedom of
this pin-jointed structure can be written as follows:

 d1 
 
 d2 
 
d= d3 
 
 d4 
 d5 
 

In the abovementioned example, we can observe that this pin-jointed structure has
a total of five unrestrained displacement components at the joints. The two supports
provided the restraints at A and B. If no supports were provided, then the number of
joint displacements for this unsupported structure would be eight (two displacements
per joint). But we have found the number of degrees of freedom was five. So, we can
understand, a total of three displacement components were stopped by the supports
provided. Now, we can quickly formulate the number of degrees of freedom of a
structure as next:

   Number   of   joint   
 Number   of   degrees   Number   of   joint     
   displacements   by   displacements   
 of   freedom   or   degrees   of   =  −  
 kinematic   indeterminacy   the   unsupported     restrained   by 
   structure   supports 
   

Depending on the joint types, the number of degrees of freedom per joint can be
shown in Table 6.1.

6.4.1  Kinematic Indeterminacy of Pin-Jointed Structures or Truss

We have seen from Table 6.1 that a joint of a 2D truss has two degrees of freedom.
So, in unsupported conditions, the total degrees of freedom of a 2D truss, as shown
in Figure 6.18 (a), would be 2 × 3 = 6 . But this truss is not externally stable as it will
undergo a rigid body motion under the influence of external loads. To prevent this,
we need to restrain the truss in some joints with supports, as shown in Figure 6.18 (b).
Indeterminacy and Stability of Structure 79

FIGURE 6.18  (a) Unsupported degrees of freedom and (b) supported degrees of freedom
and support reactions of a 2D truss.

These supports will stop the movement of those joints in some specified directions
and thereby generate the reaction forces.
So as per the earlier discussion, we can also find the number of degrees of free-
dom or degree of kinematic indeterminacy ( Dk ) as:

Dk = 2 j − re ; for 2D

Dk = 3 j − re ; for 3D

where j is the no. of joints of the pin-jointed structure, re is the no. of restrained/
support reactions.

6.4.2  Kinematic Indeterminacy of Rigid-Jointed Structures

Beams and frames are so-called rigid-jointed structures. In beam and frame mem-
bers, the external load can be applied anywhere, and the joint rigidity can be of vari-
ous types. Members can undergo bending due to the presence of transverse loads.
That is why, additionally, reacting moments can be generated at the restrained joints
depending on the support conditions and applied loads.
We have seen from Table 6.1 that a joint of a 2D beam/frame has three degrees
of freedom. So, in unsupported conditions, the total degrees of freedom of a 2D
beam/frame, as shown in Figure 6.19 (a), would be 3 × 2 = 6 . But this beam/frame
is not externally stable as it will undergo a rigid body motion under the influence
of external loads. To prevent this, we need to restrain the member with joint sup-
ports, as shown in Figure 6.19 (b). These supports will stop the movement of those
joints in some specified directions and thereby generate the reaction forces in those
directions.
80 Introduction to Structural Analysis

FIGURE 6.19  (a) Unsupported degrees of freedom and (b) supported degrees of freedom
and support reactions of a 2D rigid-jointed member.

So as per the earlier discussion, we can also find the number of degrees of free-
dom or degree of kinematic indeterminacy ( Dk ) for a rigid-jointed structure as:

Dk = 3 j − re ; for 2D

Dk = 6 j − re ; for 3D

where j is the no. of joints of the pin-jointed structure, re is the no. of restrained/
support reactions.
The abovementioned two equations are true when the members are extensible or
compressible axially. But if the members are inextensible or axially rigid, the equa-
tions can be rewritten as:

Dk = 3 j − re − m; for 2D

Dk = 6 j − re − m; for 3D

where m is the no. of inextensible members.

This can be understood by looking at Figure 6.20. In Figure 6.20 (a), we can see
an extensible member whereas in Figure 6.20 (b), we can see an axially rigid mem-
ber under the external load P. In the first case, the beam will undergo a horizontal
deformation of ∆x 2 at joint 2, whereas in the second case it will not.
When the members of a frame are inextensible or rigid, then rigid body displace-
ment should be considered combinedly for multiple members. Consider the follow-
ing example problem to clarify this.
If there are released reactions, then the number of degrees of freedom will be
increased as per the release arrangements. For example, if internal hinge is provided,
Indeterminacy and Stability of Structure 81

FIGURE 6.20  Degrees of freedom of (a) an extensible (b) inextensible rigid-jointed mem-
ber in 2D.

the bending moment gets released over there. So, in this case, the number of released
reaction (rr ) would be 1. Now, finally, we can add the influence of the released reac-
tions in the abovementioned formulation and get the final form:

Dk = 3 j − re − m + rr ; for 2D

Dk = 6 j − re − m + rr ; for 3D

Note: If you want to calculate the degrees of kinematic indeterminacy (Dk ) using the
above formula, you must consider the location of the released reactions as a separate
joint. Then only the correct answer will come.

Example 6.12: Find the degree of kinematic indeterminacy for the

following frame as shown in Figure 6.21(a). Take all the members
to be axially rigid.

SOLUTION: If the frame members were extensible/compressible, then the

degrees of freedom would be as per Table 6.2, i.e., total 6 degrees of freedom is
possible if the members are extensible/compressible.
As per the equation given previously, the degree of kinematic indeterminacy of
this frame (for members are axially rigid).

Dk = 3 j − re − m

Dk = 3 × 3 − 3 − 2 = 4
82 Introduction to Structural Analysis

TABLE 6.2
Degrees of Freedom If the Members Are
Extensible/Compressible
Joint-B Joint-C
∇x B ∇xC
∇yB ∇yC
θ ZB θ ZC

FIGURE 6.21  Example problem on degree of kinematic indeterminacy for axially rigid
frame.

Here we can see the degrees of freedom gets reduced by 2 when the members
are axially rigid. This is because, ∇yB will not be there anymore as the column AB
cannot get shortened or elongated and for beam BC instead of two axial deforma-
tions (i.e., ∇xB and ∇xC ) at its ends, the whole assembly will move horizontally to
the ∇x amount as shown in Figure 6.21 (b).

Example 6.13:  Find the degree of kinematic indeterminacy for the frame
shown in Figure 6.1.

SOLUTION:

Dk = 3 j − re − m + rr

Dk = 3 × 14 − 14 − 0 + 0 = 28

As nothing has been mentioned in the problem regarding the axial stiffness of the
members, we will consider the members as extensible/compressible.
Therefore, the degree of kinematic indeterminacy = 28.
Indeterminacy and Stability of Structure 83

Example 6.14:  Find the degree of kinematic indeterminacy for the frame
shown in Figure 6.3.

SOLUTION: 

Dk = 3 j − re − m + rr

rr = ∑ (m − 1) = (4 − 1) + (3 − 1) + ( 2 − 1) = 6
*

Therefore, Dk = 3 × 12 − 9 − 0 + 6 = 33
As nothing has been mentioned in the problem regarding the axial stiffness of
the members, we will consider the members as extensible/compressible.
Therefore, the degree of kinematic indeterminacy = 33.

Example 6.15:  Find the degree of kinematic indeterminacy for the frame
shown in Figure 6.4.

SOLUTION: 

Dk = 3 j − re − m + rr = 3 × 18 − 10 − 0 + 7 = 51
As nothing has been mentioned in the problem regarding the axial stiffness of the
members, we will consider the members as extensible/compressible.
Therefore, the degree of kinematic indeterminacy = 33.
Note that, the location of the released reactions was considered as separate joints.

6.4.3 Summary of All Formulations for Static

and Kinematic Indeterminacy

Finally, we are putting all the formulations for static and kinematic indeterminacy
shown in this chapter in a tabular form (Table 6.3) as shown next. All the notations
mentioned in the equations are already explained in this chapter.

TABLE 6.3
Summary of All Formulations for Static and Kinematic Indeterminacy
Static Indeterminacy (Ds )
Kinematic
Type of Structure Dse Dsi Indeterminacy (Dk )
Truss 2D re − 3 m − ( 2 j − 3) ( 2 j − re )
3D re − 6 m − (3 j − 6) (3 j − re )
Beams/Frames 2D re − 3 ( 3c − rr ) or {3m − ( 3 j + rr − 3)} 3 j − re − m + rr
for beams, c = 0
3D re − 6 { }
(6c − rr ) or 6m − ( 6 j + rr − 6 ) 6 j − re − m + rr
for beams, c = 0

6.5  PRINCIPLE OF SUPERPOSITION

The principle of superposition is the most important axiom or method or statement/
concept for theory of structural analysis. It may be introduced as follows: the total
84 Introduction to Structural Analysis

FIGURE 6.22  Principle of superposition.

displacement or internal stress, at a point/location in a structure acted upon by sev-

eral external loadings, can be calculated by adding together the displacements or
internal stress generated by each of the external loads acting separately at the same
location as that of original structure. For this statement or axiom to be valid, it is
important to understand that a linear relationship exists among the loads, stresses,
and displacements. Thus, to apply the principle of superposition, two points need to
be satisfied by the structure as described next:

1. Material, which the structural member is made up of, should behave in a

linear-elastic manner, so that Hooke’s law is valid, and therefore the load
will be proportional to displacement. In other words, material should be
elastic in nature as per the stress-strain curve explained at the beginning of
this section.
2. The geometry and shape of the structural elements must not undergo sig-
nificant change when the loads are applied, i.e., plane section should remain
plane before and after loading. Large displacements will completely change
the position and orientation of the loads.

Throughout this book, these two requirements will be found to be satisfied in all
cases. Hence, we may not declare it at all places wherever we will apply this prin-
ciple. Only at the end of this book, in the ‘Plastic Analysis’ section, we will introduce
another concept that will not follow this principle and a separate plastic analysis
principle will be introduced to that section under that particular different concept. In
Figure 6.22, the principle of superposition has been explained.
 7 Plane Trusses and
Space Trusses

7.1 INTRODUCTION
In this chapter, we will study the detailed analysis procedures for plane and space
trusses. Truss is an important type of structural element that has multiple applications
for constructing large load-carrying structures like bridges, roofs, and buildings. A
sound understanding of the truss system and its analysis procedure is a must-have for
all working engineers and researchers. So, students are advised to go through this
chapter in detail to gain a good understanding of various analysis procedures for the
plane as well as space trusses. We will explore different methods for analyzing the
trusses, and by analysis, we actually mean to be able to calculate all external and
internal unknown forces acting on the truss. Analysis is the first step before carry-
ing out detailed design of structural members. So, proper analysis of truss with all
forces acting internally and externally will help us to do the design work efficiently
and correctly.

7.2  COMMON TYPES OF TRUSSES

A framework consisting of members joined at their end by flexible connections to
form a rigid structure is called a truss. Structural elements, such as I-beams, angles,
channels, and square hollow sections connected at their ends by welding, rivets, or
bolts, form a complete truss for a particular application. When the truss members
are all lying in the same plane, it is called a plane truss. The combination of two or
more plane trusses in different planes is said to form space trusses. There are many
types of plane trusses used in bridges, roofs, and highway over bridges. The most
commonly used types of plane trusses are shown in Figure 7.1 for reference.

7.3  CLASSIFICATION OF COPLANAR TRUSSES

The basic element of any truss is the triangular arrangement of elements as shown
in Figure 7.2 (a). Three bars connected at the ends with pins, rivets, bolts, or weld-
ing, form a rigid or internally stable triangular element. The triangular element is
important to notice since by connecting in this fashion, the element becomes rigid,
and there will be no relative movement among the members. This is not applicable
for four or more members connected at ends to form a quadrilateral or polygonal
element as shown in Figure 7.2 (b). Such an element is free to move at the joints, and
it will form a mechanism. To make it stable, we can connect any one of its diago-
nals and, thus, forming two triangular elements. So, for trusses, triangular elements
always play a key role in structural stability.
DOI: 10.1201/9781003081227-9 85
86 Introduction to Structural Analysis

FIGURE 7.1  Different types of trusses.

Toward this end, we can say that trusses can be of three types such as simple
truss, compound truss, and complex truss. A simple truss is the one made up of a sin-
gle triangle formed by three connected members and, thereafter, enlarging its basic
truss element depending on the span required as shown in Figure 7.3. However, the
final form and overall dimension of the truss will be calculated based on the actual
site requirements. The basic truss unit contains three members. Now if we want a
simple truss for a longer span, every time we need to add two new members with

FIGURE 7.2  Fundamental triangular element for plane truss.

Plane Trusses and Space Trusses 87

FIGURE 7.3  Every time addition of two new members and a new joint in simple truss
extension.

FIGURE 7.4  (a) compound truss and (b) complex truss.

a new joint from the basic truss unit. If we formulate the number of members (m)
present for a simple truss, it will take this form: m = 3 + 2 × ( j − 3) = 2 j − 3; j being
the no. of joints.
A compound truss can be formed by combining two or more triangular simple
trusses together to form a single rigid body. See Figure 7.4 (a) for better understanding
of compound truss. Here two simple trusses ABC and EFG are connected by three
members BE, CF, and BF. Note that these three members are nonparallel and noncon-
current to each other, which makes the whole truss system internally stable or rigid.
A complex truss is the one that cannot be classified as simple or compound. A
typical complex truss is shown in Figure 7.4 (b) for reference.

7.4  ASSUMPTIONS ON ANALYSIS OF TRUSSES

Elements of trusses are all axially loaded members. By this, what we mean, for
trusses, there will be no induced moment due to external loading on the members
forming the truss. So, the members are said to be in a state of tension or compression
only. Even under certain situations, there may be some members in which no force is
being induced due to external loading. Calculation of forces of various members of
the truss is carried out by applying the local equilibrium conditions to those mem-
bers. And the support reactions acting on the overall truss will be calculated based
on global equilibrium conditions. There are several methods for analyzing plane
88 Introduction to Structural Analysis

truss that will be developed in the subsequent sections of this chapter. The important
assumptions that help us to analyze trusses are as follows:

1. External loads act on joints only. If there is any load acting in between
nodes, then we need to distribute it equally among two adjacent joints.
2. All members are subjected to axial loads only. There will be no moment
acting on the members due to any kind of loading on the truss.
3. The connections between different members are perfectly hinged/pinned
through frictionless connections.
4. Self-weight of the truss is either ignored or assumed to be equally distrib-
uted among its various members.
5. Even if members are connected by welding, bolt, etc., a nominal moment
that can be generated due to imperfect fixity (other than frictionless hinge)
is ignored.
6. No matter how many members are connected by gusset plates at a node,
if axis of all members is passing through the same point at the joint, then
the members are assumed to be hinged/pin connected. Hence, no moment
will come into play at this joint. For better understanding of this particular
point, please see Figure 7.5. As can be seen, the axis of all members is
passing through the same point at the support, and hence, it is a pin/hinged
connection without any moment induced at the members.
7. The vector representing a force acting on a joint or a section is drawn on
the same side of the joint or section. Since truss members are a two-force
system (tension or compression-only), before analysis, we assume the mem-
ber is in tension and draw a force vector along the axis of the member at
the same joint where the member is located. After carrying out the detailed
analysis, we get the algebraic sign and value for the acting force on that par-
ticular member, and then we get the perfect knowledge about the tension/
compression state of the member.
8. Force on the member acting towards the joint or section is taken as com-
pression, whereas force away from the joint or section is taken as tension.
See the Figure 7.6 for better understanding.

We have presented the assumptions of truss analysis in the above paragraph without
any example. As we will go through the examples, we will get a clear idea how these

FIGURE 7.5  A typical truss member connection using gusset plates.

Plane Trusses and Space Trusses 89

FIGURE 7.6  Tension/compression convention for truss members.

assumptions help us analyze any particular truss correctly and efficiently. Provision
for expansion and contraction arising from temperature changes and for deformation/
deflections caused by the applied loads is usually made at one of the supports for big
trusses. Roller, rocker supports, or some kind of slip joint are provided to manage
such effects.

7.5 ARRANGEMENT OF MEMBERS OF COPLANAR

TRUSSES – INTERNAL STABILITY
We have already explained the different types of trusses used to model our actual
site requirement. Coplanar trusses are the simplest models that provide us easy
analysis and design procedures. Moreover, noncoplanar three-dimensional (3D)
trusses are made by combining several coplanar trusses with tie members at
the top and bottom of the structure. Various members of coplanar trusses are
arranged in a way so that from any side, each is seen to be composed of triangu-
lar shapes. The three-member triangular shapes provide the best to form a rigid
structure to construct the overall truss. Thus, for stability, we need to connect
various triangular elements of the truss in a fashion so that there are no noncon-
nected nodes available in the truss. If there is any nonconnected node, the force
distribution among various members will be uneven, and ultimately it will cause
the structure to collapse. However, it may be understood that we may mistakenly
provide more internal members than that are required for the overall structural
stability of the truss. In such a case, when there are more members than needed
for structural stability, they become statically indeterminate truss. We cannot
analyze such type of truss completely by applying only equilibrium equations. In
the following sections, we will discuss the conditions and equation to determine
whether any given truss is statically determinate or indeterminate. The term rigid
is applied in the sense of being non-collapsible and in the sense that deformation/
deflections of the members due to induced internal forces and strains are minus-
cule to be neglected.
90 Introduction to Structural Analysis

7.6 STATIC DETERMINACY, INDETERMINACY,

AND INSTABILITY OF COPLANAR TRUSSES
AND THEIR SOLUTION METHODS
Continuing from the previous discussion, let us concentrate on determining
whether a truss is statically determinate or indeterminate. We should remember
that this is the most crucial step before diving into analyzing the truss completely.
For truss, we have two kinds of determinacy – internal and external. External
determinacy is related to the nature and number of supports provided to the truss
as a whole. Say, for example, if we provide two pin/hinge supports in a truss, it
will be statically indeterminate because we can not calculate the horizontal sup-
port reactions that will be acting on the two pin supports from the equation of
global equilibrium. And the degree of indeterminacy will be one, which is the dif-
ference between the total number of unknown support reactions and the number
of equilibrium equations available. From this example, we can set the condition
or criteria to check the external indeterminacy of any given truss, which is as
follows:

Dse = ( re − 3); for two-dimensional  ( 2D )  truss

Dse = ( re − 6 ); for 3D truss

where Dse is the degree of external indeterminacy, re is the number of unknown

support reactions in a particular direction, number of global equilibrium equations
available = 3 or 6 for 2D and 3D, trusses, respectively, in general.
For one-pin and one-roller supports for a truss in 2D, we have total unknown
support reactions equal to 3 (two at pin and one at roller) depending on the direc-
tion of the external loading. If only vertical loading is applied on the truss, no
horizontal support reaction will be generated, and, in that case, horizontal equilib-
rium equation is also not required. If inclined load is applied, the unknown support
reactions will be 3, and the available equations for the global equilibrium is also
equal to 3, namely,

∑ F = 0, ∑ F = 0, and ∑ M
x y z = 0; ( for 2 − D trusses )

And, hence, the degree of indeterminacy is = 3 − 3 = 0, i.e., the truss is statically

determinate externally, and we can determine the unknown support reactions com-
pletely by applying equations of global equilibrium only.

∑ F = 0 and ∑ M
y z =0

At this stage, if someone feels to brush up equilibrium conditions and equations for
structures, it is highly recommended to go through Chapter 6 of global and local
equilibrium conditions as we will use those concepts all through this book for ana-
lyzing structures.
Plane Trusses and Space Trusses 91

Having gained sufficient knowledge about the external degree of indeterminacy,

we will now discuss the internal degree of indeterminacy for a 2D truss. Let us
think of a truss made up of m number of members connected by j number of joints
and suppose the total number of support reactions are r (mostly this will be three
for plane trusses supported at one end by pin and at the other end by roller). We
already know that for m number of members and r number of support reactions, total
unknown forces are = m   +   r . Since there are j number of joints, total number of
equilibrium equations available will be 2 j. Hence, for a statically determinate stable
truss, the following necessary condition should be met:

m   +  r = 2 j

For instance, if,

m   +  r   > 2 j

And if, for simplicity, we take r = 3, then we have more members than that are
required for the overall stability of the truss. Hence, at this situation, it is called
a truss having redundant members inside it. And hence, equations of statics are
not sufficient enough to calculate all the unknown member forces. Thus, the truss
becomes statically indeterminate internally.
Another instance is the following,

m   +  r   < 2 j

And as per previous discussion, let us take r = 3. So, this indicates that the truss is
having fewer members than are required for the overall stability. Thus, the truss is
not a stable one, and it will collapse eventually or form a mechanism. So, this condi-
tion refers to unstable trusses. Hence, following this logic, we can determine at the
very beginning, before doing any analysis, whether it is statically determinate or not.
If it is statically indeterminate, we will need additional equations other than equilib-
rium conditions to determine all unknown member forces.
For space trusses or 3D trusses, we can develop the same type of equations or con-
ditions for checking their degree of indeterminacy. For space trusses or 3D trusses
having m number of members, j number of joints, and r number of support reactions,
we have the following relationship:

m   +   r = 3 j; for determinate truss.

For trusses having redundant members, the above relationship will become:

m   +  r > 3 j

And for unstable trusses, the equation will be:

m   +  r < 3 j
92 Introduction to Structural Analysis

The degree of static indeterminacy as discussed in Chapter 6 is defined by the fol-

lowing expression:

Ds = m + re − 2 j   ( for plane truss )

Ds = m + re − 3 j   ( for 3 − D / space truss )

For statically determinate trusses, Ds = 0, which is self-evident from the above

discussions.
So, before we proceed to analyze a truss, we should first check whether it is stati-
cally determinate or not. That being said, we have arrived in a position to learn vari-
ous methods of truss analysis. We will study the following methods for truss analysis:

1. Method of joints
2. Method of sections
3. Method of tension coefficients
4. Graphical method of truss analysis
5. Henneberg’s method of solution for complex trusses

7.6.1 Method of Joints
Among the various methods listed above, the method of joint is the simplest and
most straightforward method for truss analysis. To apply this method, we need to
pay attention only to the equations of global and local equilibrium, and no special
consideration needs to be taken as such. In this method, we need to work joint by
joint of the truss. We will always start from a joint with no more than two unknown
forces (which must not be colinear). So, the appropriate choice is to start from the
nodes connected with the supports, because at the start before doing anything, we
calculate the support reaction acting on the supports due external loading on the
truss. Let us proceed with this method with a simple example to better understand
its underlying concepts.
In Figure 7.7, a simple truss with an external point load P acting on node C is
shown. All the members of the truss are of same length l. Hence, the angle between
members will be 60°. Now before doing any force calculations, first, let us check
whether this truss is statically determinate or not. For this plane truss we have:

m = 7
r=3
 
j=5
 

So, the degree of static indeterminacy, Ds = m + re − 2 j = 10 − 10 = 0

Thus, the given truss is statically determinate.
Next, we need to calculate the support reactions acting on the supports due to
external loading. As explained in the earlier chapter of equilibrium conditions, we
need to consider the truss with external loading and support reactions and apply
global equilibrium conditions to calculate the support reactions as per Figure 7.8.
Plane Trusses and Space Trusses 93

FIGURE 7.7  Plane truss for analysis example.

Now applying global equilibrium condition in x direction, namely, ∑ Fx = 0 yields,

HA = 0

And taking moment of all forces about support B, we get,

RA × 2l − P × (2l − l cos 60°) = 0

or,
3P
RA =
8
Now, for global equilibrium in y direction we have, ∑ Fy = 0, which means,

RA + RB = P

or,
5P
RB = P − RA =
8

FIGURE 7.8  Global equilibrium condition for determining unknown support reactions.
94 Introduction to Structural Analysis

FIGURE 7.9  Method of joint and free body diagram.

So, we get the unknown support reactions with their proper direction from the above
expressions. Now we have arrived at the point from where we can start analyzing the
truss internally to calculate its member forces by the method of joints.
Let us begin from joint A as shown in Figure 7.9, since already we have a known
support reaction i.e., the number of unknown forces is two at this joint, and draw the
free body diagram with all members replaced by tension forces (assumed) as shown
below:
Now we are dealing with local equilibrium conditions to calculate member forces
induced on members due to external loading. So, from the above free body diagram,
applying equilibrium condition in y direction, we get,

∑F = 0x

or,
FAC  Sin 60 + RA = 0
or,

RA 3P
FAC = − =−
sin 60 2√3
So, we get the member force FAC with its appropriate direction. The minus sign
indicates that the member is under compression. Now applying local equilibrium
condition in x direction, we get,

∑F = 0x

or,
FAE + FAC cos 60 = 0
or,
3P
FAE = − FAC cos 60 =
4 √3
Plane Trusses and Space Trusses 95

So, we get the member force on the member AE, and the positive sign indicates that
the said member is under tension as was assumed.
We may now proceed sequentially to other joints and draw appropriate free body
diagrams to calculate the member forces in those members. Complete analysis of
this truss applying this concept is left as an exercise to the reader.
Thus, we get a complete idea about the method of joints and how this method
helps us to determine the unknown member forces in various members of a given
truss due to external loading. This method is very simple and elegant. However, for
large trusses, this method becomes very tedious and time consuming. For a large
truss, if we want to calculate member force of any particular member, then this
method doesn’t give immediate result. To be able to do that, we need a more sophis-
ticated method that we are about to learn in the next section.

7.6.2 Method of Sections
This method is elegant in a sense that we can calculate the member force on any
member directly from this without carrying out detailed step-by-step analysis. We
can immediately calculate the member force induced on any member due to external
loading by taking a suitable section passing through that member. Sections drawn
through truss should be such that it passes from bottom to top or top to bottom of the
entire truss. Also, it should be borne in mind that no chosen section should contain
more than three members. With this simple logic, let us calculate the member force
on the member DE of the previous example directly after calculating the degree of
indeterminacy and support reactions. So, the section is as shown in Figure 7.10.
Once section is drawn, we need to replace the members by member forces through
which the section passes. Hence, the free body diagram of the truss after taking sec-
tion can be as shown in Figure 7.11.
Now, for local equilibrium condition in y −direction, we have ∑ Fy = 0
or,

FDE sin 60 − RB = 0

FIGURE 7.10  Method of section and placement of imaginary section line through truss.
96 Introduction to Structural Analysis

FIGURE 7.11  Free body diagram with member forces through as per the section line.

or,
5P
FDE =  
4 √3
So, member DE is under tension, and the member force acting on it is 5P /4 √ 3, and
our requirement is satisfied. We are able to calculate the member force directly by
taking the appropriate section through that member and then applying local equilib-
rium conditions. Also, from the above free body diagram, if we take moment about
joint B of all forces, then we should get the member force at D. A complete truss
analysis by this method is left as an exercise for the readers. Also, it is also interest-
ing to check whether analyses by these two different methods yield the same result.
Hint to calculate member force DC from above free body diagram:
Taking moment about joint B yields:

FDC × l sin 60 + FDE sin 60 × l cos 60 + FDE cos 60 × l sin 60 = 0

Since we have already calculated FDE , from the above expression, we can calculate
the unknown member force FDC.
In drawing the above free body diagram, we have not considered the left section
of the truss. If we draw the left section of the truss, then the following free body
diagram as shown in Figure 7.12, appears.

FIGURE 7.12  Left section free body diagram of the truss.

Plane Trusses and Space Trusses 97

From the above diagram, by applying local equilibrium condition in y direction,

we get:

∑F = 0
y

or,

FDE sin 60 + RA − P = 0

or,

P RA 5P
FED =   − = 
sin 60 sin 60 4 √ 3

Exactly same as that obtained from the right section, we have considered in the first
analysis.
Hence, by taking left and right sections and sketching the appropriate free body
diagrams, we can easily get the unknown member forces by simple application of
local equilibrium equations. Applying this concept for the whole truss to determine
all unknown member forces is left as an exercise for the readers.

7.6.3 Method of Tension Coefficients

The method of tension coefficient was developed by R.V. Southwell in 1920 and is
applicable for both plane and space trusses. This method is very elegant in the sense
that it can be directly converted into a computer program for quicker analysis. To
understand this method, let us take a single member AB of a plane truss in ( x , y )
plane and also let us write down the coordinates of the end nodes for the same.
The coordinate of node A = ( x a , ya ), and coordinate of node B = ( x b , yb ) as per
Figure 7.13. As per the concept, the member force ( Tab ) is to be considered as tensile

FIGURE 7.13  Member coordinates for tension coefficient method.

98 Introduction to Structural Analysis

for this member AB. The components of the member force ( Tab ) along the coordinate
axis can be written as:
Tx = Tab cos θ , and Tx = Tab sin θ ; where θ is the inclination of the member to the
x-axis. Here cos θ , sin θ are the direction cosines of the force vector and can be
expressed as follows:

x b − x a x ab y − ya yab
cos θ = = , and sin θ = b = ; where lab = x ab
2
+ yab
2
lab lab lab lab
is the length of the member AB.

Now, we can replace cos θ and sin θ in the expressions for Tx and Ty and rewrite the
equations as:

x ab y
Tx = Tab × = tab × x ab , and Ty = Tab × ab = tab × yab
lab lab

Here the parameters, Tl is the force per unit length of a member and is known as
tension coefficient of the member.
So, if somebody wants to calculate the x component of tensile force acting on this
member, then he/she needs to multiply the tension coefficient tab with the difference
in x coordinate of the two nodes as explained above. That means, the x component
of force on member AB =   tab × ( x b − x a ), and similarly y component of force acting
on this member will be as tab × ( yb − ya ). The tension coefficients are assumed to be
positive initially. But if it turns out to be negative after calculation, the force in that
member is taken as compressive. Now consider a truss joint j where three members
are connected as shown in Figure 7.14.
In Figure 7.14, the joint i of a truss is connected to other truss joints l, m, and k
with three members and a concentrated load P is acting on it. Now, applying the
above concept of tension coefficient, we can formulate the equilibrium condition of
joint j as below:

∑F = 0 = t
x jk × x jk + t jl × x jl + t jm × x jm + Px = 0

∑F = 0 = t
y jk × y jk + t jl × y jl + t jm × y jm + Py = 0

FIGURE 7.14  Joint j is connecting three members and load P is applied on it.
Plane Trusses and Space Trusses 99

FIGURE 7.15  Example on tension coefficient method.

In compact form, the above two equations can be written as:

∑t jk × x jk + Px = 0

∑t jk × y jk + Py = 0

Example 7.1: Establish the equilibrium equations along both the

coordinate axis for the truss joint A shown in Figure 7.15 by tension
coefficient method.

SOLUTION: In this figure, all the coordinates of the two connected members
are shown, and there is an external point load P acting on y direction at joint A.
Following the above rules for tension coefficients, we can write the equilibrium
equations as:

∑F = 0 x

t AB × ( x2 −   x1) +  t AC × ( x3 −   x1) = 0; since there is no external load acting on x

direction.

∑F = 0 y

t AB × ( y 2 −   y1) +  t AC × ( y3 −   y1) − P = 0; since P is acting toward the joint, it is com-

pressive in nature.
All the coordinates and external loads are known to us. Hence, we can solve
these simultaneous equations to get the unknown tension coefficients.
To summarize this method of analysis, we can follow the below procedure:

a. Mark the positive and negative directions of ( x , y ) for plane trusses and
( x , y , z ) for space trusses.
b. Start with the assumption that all members are in tension, i.e., member
forces are away from the joint at which all are connected.
c. Then write down the equilibrium equations at each joint as per the pro-
cedure just shown for a typical truss with external loading.
100 Introduction to Structural Analysis

d. Solve the equations for unknown tension coefficients.

e. Finally calculate the member forces by the relationship TAB =  t AB   LAB

This method is to some extent linked with the method of joints since we need to
move from joint to joint for calculating the unknown tension coefficients. Let us
work out few member forces by this method for the same truss we have consid-
ered in our previous section. All the coordinates are presented as per the geom-
etry of the truss. Origin has been assumed to be situated at joint A.
Let us begin our work from node A. At this node, there are two members AC
and AE. And there is a support reaction acting on node A as found from our previ-
ous analysis. So, in terms of tension coefficient in x direction, we can write:

l 
t AE × ( l − 0 ) +  t AC ×  − 0 = 0
2 
or,

t AE = −t AC / 2

And for the y direction, we can write,

t AE × ( 0 − 0 ) +  t AC × ( )
3 / 2l − 0 − P +  RA = 0

or,

5P
t AC =  
4 √ 3l
So, from previous expression we get,

5P
t AE = −
8 √ 3l

Hence, member force on AC, TAC = t AC × LAC  

FIGURE 7.16  Complete truss with the coordinates for tension coefficient.
Plane Trusses and Space Trusses 101

or,

5P 5P
TAC   =   × l = 
4 √ 3l 4√3

and,

5P 5P
TAE = − ×l =−
8 3l 8√3

So, we get the magnitude as well as direction of forces of the member (compression/
tension) from the above method. This can be checked with respect to the earlier
methods whether the direction and magnitude of force of the members are same.
As can be understood, this method mostly relies on the method of joints, since
we need to move joint by joint to get the tension coefficients and corresponding
forces acting on the members. Thus, for quick and elegant approach for calculat-
ing any arbitrary member forces, the method of section is the most effective. For
complete truss analysis using computer programs, the method of tension coeffi-
cients is most suitable. We can build an algorithm to incorporate large trusses by
demanding very few inputs to analyze it completely once the program analysis is
complete. Complete analysis of this truss using tension coefficients for all mem-
bers is left as an exercise for the reader.

7.6.4 Graphical Method of Truss Analysis

Now, we will learn how to analyze a truss by graphical method. In this method, no
real calculation is being made. Only a suitable choice of scales for drawing member
forces yields the desired result of unknown member forces. To be able to do that, we
need to learn a few basic theorems and notations for preparing ourselves for graphi-
cal analysis. This method was largely developed by several analysts in the early
19th century (the most famous among them were Maxwell and Betti). Although no
equations are required to be dealt with in this method, one must have a good under-
standing of the force vector, triangle rule of vector addition, parallelogram rule of
vector addition, etc. To this end, we need to understand Bow’s notation for describing
the force vector by the space enclosed by them. It will be clear from the following
diagrams.

7.6.4.1  Bow’s Notation

In this notation, all force vectors are represented by the space included by them. Let
us draw some coplanar force vector with spaces marked as shown in Figure 7.17.
In the given diagram, we have drawn three concurrent forces with directions indi-
cated by arrows. These force vectors are not named as per usual convention for vec-
tors. Instead, we have marked the spaces surrounding the vectors by letters A, B, C,
etc. So, starting from the left, the first force vector will be denoted as AB, second
one will be BC, and the third will be denoted as CA. So, for triangle rule of vector
addition, if we draw the triangle with small letters a, b, c, etc. to indicate the force
vectors, then it will be something as shown in Figure 7.18.
102 Introduction to Structural Analysis

FIGURE 7.17  Bow’s notation for graphical analysis.

FIGURE 7.18  Force triangle following Bow’s notation.

So, from the triangle of forces, it is closed under vector addition. It means that
the forces are not having net resultant force. Hence, the concurrent force system is in
a state of equilibrium. The force triangle is drawn with a suitable scale to represent
the forces. Also, once we start with any one of the vectors ab, bc, or ca, the triangle
of forces automatically falls in place following the rule for vector addition. Bow’s
notation helps us to neatly draw the force system replicating the actual forces acting
on any structure or members like this way. For nonconcurrent forces, we indicate the
forces in the same way of surrounding spaces as we did for concurrent forces. See
Figure 7.19 for a system of nonconcurrent forces and their nomenclature for better
understanding.

FIGURE 7.19  Several nonconcurrent forces.

Plane Trusses and Space Trusses 103

So, for the above nonconcurrent forces, the force vectors are AB, BC, CD, and
DE as per Bow’s notation. These forces are not in equilibrium. So, the force polygon
will be open-ended, and the final vector joining the first node to the last node of the
force polygon will be the resultant force for all these forces. To establish this fact
graphically, we will now draw some particular force diagrams with suitable scales
for calculating the resultant force and its location/position. Let us take an arbitrary
point O and draw the forces AB, BC, CD, and DE tail to tail as with proper direction
as per Figure 7.19.
Above diagram is known as the polar diagram. The pole is the point ‘o’. All forces
are drawn in a suitable scale say 1 mm = 1 kN from tail to tail as shown above. From
the individual small triangles, we can see that all the internal forces except force oa
and oe remain left (which are indicated by the dotted arrows). So, the resultant for all
given nonconcurrent forces is the force ae, connecting vector from point a to point e.
So, from this diagram, we can get the value of the resultant force by measuring the
length of arrow ae and applying the chosen scale factor to calculate how much kN
will be resultant force.

FIGURE 7.20  Polar diagram.

104 Introduction to Structural Analysis

FIGURE 7.21  Funicular polygon for system of nonconcurrent forces.

To find the location of the resultant force, we need to draw the diagram as shown
in Figure 7.21, where line 1 is drawn at a suitable distance from force AB, parallel
to oa line in polar diagram. Similarly, all other lines are drawn parallel to sides of
ob, oc, and od as per polar diagram. As can be seen from the polar diagram, the
resultant force is represented by the line ae, and as per triangle rule for vector addi-
tion, we get ae = ao + oe. So, we extend the line 1 and line 4 in the above diagram to
intersect at the point r as shown in Figure 7.21. Now we can draw the line through
point r parallel to the line ae of the polar diagram. Hence, by this method, we get
the location of the resultant force for nonconcurrent forces and it is near to the
force AB. The above shape of the line 1, 2, 3, 4, etc. looks very similar to a loaded
string or rope. If we hang several loads from a tied rope, then it will take a shape
something like the above diagram lines 1–2–3–4. This diagram is known as funicu-
lar polygon. A funicular polygon may be thought of a possible configuration of
equilibrium of a rope or string, suspended from its ends, and is loaded to the given
system of coplanar forces. It needs to be understood that a string takes the profile
of a straight-line segments when it is subjected to a series of concentrated forces. In
case of distributed loads, string assumes a smooth curve profile. The same concept
can be applied to general class of loading to determine location and magnitude of
resultant force. However, we will focus on the methods of truss analysis using the
graphical method just explained. Even this method can be applied to determine
center of gravity for plane areas with irregular geometry. Interested reader may
refer to some excellent textbook mentioned at the end of the chapter reference for
the same.
Plane Trusses and Space Trusses 105

FIGURE 7.22  Truss for graphical analysis.

Let us consider the same simple truss with point load at its top left joint as shown
below. But in case of graphical analysis, we need to apply Bow’s notation for forces
acting on the system. For that, let us define the spaces surrounding the truss as shown
in Figure 7.22.
So, as per Bow’s notation, the support reactions will be AC and CB, whereas the
unknown member forces will be A1, 12, 23, 2B, 1C, 3C, 3B, etc.
As we already know, to analyze any truss, we need to first find out support reac-
tions at the supports. For determining support reactions, we also know that whole
truss is visualized as a single rigid beam like structure with loads at the distance
from supports as shown in the drawing. Hence, in graphical analysis, we need to
draw a funicular polygon as shown in Figure 7.23, representing the truss as a rigid

FIGURE 7.23  Funicular polygon and polar diagram for truss example.
106 Introduction to Structural Analysis

beam with point load on top of it. So, the funicular polygon and polar diagram will
be something like as presented below. Polar diagram has been drawn by choosing
a suitable scale say 1 mm = 1 kN and then the lines oa and ob are drawn. It is to be
understood that first line of funicular polygon is to be drawn from any point as shown
below, and it is parallel to the line oa and meeting the projected line of external
applied force at some point R as shown in Figure 7.23.
Now from R, a second line is drawn parallel to line ob of polar diagram up to the
left end projection line of beam. This line is named as line 2 in the funicular polygon.
Now the line joining the two ends of line 1 and line 2 is known as the closing line
of the funicular polygon. This line is the one that helps us to determine the support
reactions from polar diagram. Once the closing line is drawn, a parallel line from o
to the closing line is drawn which intersect the ab line at some point c. Hence, ac
segment will be the left support reaction, and cb will be the right support reaction
that can be directly measured from the polar diagram. It is also to be noted that this
method of determination of support reaction is applicable for any determinate beams
with point loadings. Once the support reactions are known then we can proceed to
analyze the truss by simply force triangle and force polygon method remembering
the fact that we should start from joint where at least one force is known to us. Also,
it is to be noted that scale chosen during drawing the polar diagram should remain
same for the entire problem we are dealing with. Scale factor once chosen cannot be
changed in the middle of calculations. Hence, for the joint AC1A, we have the follow-
ing force triangle as shown in Figure 7.24.
Since all forces are in equilibrium, force triangle will be closed under vector addi-
tion. From the above force triangle, ac, is already known and has been drawn with
the chosen scale factor. Hence, all other forces fall neatly following the vector rule
of addition. To draw the forces c1 and a1, we can draw two lines parallel to the truss
members from point c and point a, and the intersection of these two lies will be the
point 1. Once the triangle is drawn, we can measure the length and multiply it with
the chosen scale factor to obtain the force values in the member. Also, from the force
triangle, we immediately get the state of the member. For example, from force trian-
gle, we get that member C1 is under tension while member A1 is under compression.
Now we can proceed to joint AB21 where three members are connected. Of
these three member forces, we already calculated the force on the member A1 in

FIGURE 7.24  Joint AC1A force triangle.

Plane Trusses and Space Trusses 107

FIGURE 7.25  Force polygon for joint AB21.

the previous analysis. So, the force polygon for this joint will be something like as
shown in Figure 7.25.
First of all, we need to draw the force ab which is the known external point load
acting on the truss as per the chosen scale. Then line a1 is drawn as this force is
already known from previous analysis, and this line is parallel to the member A1.
Once these two lines are drawn next, we need to draw two lines, namely, from
point b, b2 parallel to member B2 and from point 1, 12 parallel to member 12. These
two lines will intersect at the point 2. All forces will be directed as shown in the
force polygon, since this should also be closed under vector addition rule under equi-
librium conditions. As all known forces are drawn as per the chosen force scale, we
can measure the length of lines 12 and b2 to get the force values in the member. Also,
from force polygon, it is immediately clear that member B2 is under compression,
member 12 in under compression.
All other joints and force polygons can be drawn to scale as per above method to
complete the truss analysis as a whole. Once calculated, member forces from graphi-
cal analysis can be compared to that obtained from other earlier defined methods to
compare the error in graphical analysis. Graphical analysis provides a nice tool to
analyze trusses with excellent accuracy compared to the other theoretical analysis.
Professor Maxwell made outstanding contributions toward the graphical analysis
of structures. For detailed discussion and understanding on the same, students are
encouraged to consult the books mentioned in the reference.

7.6.5 Henneberg’s Method of Solution for Complex Trusses

The member force analysis for a complex truss can be determined using the method
of joints; however, the analysis procedure will need to write two equilibrium equa-
tions for each of the j joints of the truss and then solving the complete set of 2j equa-
tions simultaneously. This approach may be tedious for hand calculations, especially
in the case of large complex trusses. Therefore, a more direct qualitative method
for analyzing a complex truss, referred to as the method of substitute members or
Henneberg’s method, will be presented here in a stepwise fashion.

a. Reduction to a stable simple truss

At first, we need to determine the reactions at the supports and begin by imag-
ining how to analyze the truss by the method of joints, i.e., progressing from
108 Introduction to Structural Analysis

FIGURE 7.26  Henneberg’s method of bar replacement by axial force.

joint to joint and solving for each member force. If a joint is reached where
there are three unknown forces, we need to remove one of the members at the
joint and replace it by an imaginary member elsewhere in the truss. By doing
this, reconstruct the truss to be a stable simple truss.
b. External loading on simple truss
Load the simple truss with the actual loading P, then determine the force on
each member i. But first of all, we need to find out the reactions. Then one
could start at any joint say A to determine the forces in connected member at
that node, then progressing successively to other joints where not more than
three members meet.
c. Remove External loading on simple truss.
Just imagine a simple truss without the external load P. Now let us place
equal but opposite collinear unit loads on the truss at the two joints from
which the member was removed as shown in Figure 7.26. If these forces
develop a force on the ith truss member, then by proportion an unknown force
x in the removed member would exert a force on the ith member. We replace
one bar by its axial force X. Because the structure becomes unstable, we add
one extra bar (at another place) to ensure mechanical stability. The axial
force on the extra bar should be zero. Using this condition, we can calculate
the force on the removed bar.

Let us introduce two more terms to define force system in this method. First one
is N e ( P ) that represents force due to external loading with proper support reac-
tions. Second one is N e ( X = 1) is the force caused by self-equilibrium forces X with
no support reactions. And we will use superposition principle as per the following
scheme to obtain the result:

N e = N e ( P ) + N e ( X ) = N e ( P ) + XN e ( X = 1) = 0

From which we get,

Ne ( P )
X = − 
N e ( X = 1)

where, N e ( X = 1) is the force due to unit load X = 1 in equilibrium condition.

To understand this method of analysis, refer to the following example.
Plane Trusses and Space Trusses 109

Example 7.2:  Determine the Axial Force on the Indicated Member of

the Following Truss Using Henneberg’s Method of Analysis.

SOLUTION:  The given truss is complex in nature. We cannot analyze this truss
by the method of joint or the method of sections without involving large number
of equilibrium equations. Hence, Henneberg’s method is most appropriate for this
type of complex truss analysis.
First of all, we determine support reaction for tis truss by applying global equa-
tions of equilibrium.

40.3 + 70.3 40.1 − 70.3

VA = = 82.5 kN, RB = 40.1− 70.3 =   −42.5 kN, HA = 70 kN
4 4

Next, we replace the indicated bar with self-equilibrium force X as shown in

Figure 7.27 (c). Since we have removed a bar from the original truss, we have

  

     

FIGURE 7.27  (a) Henneberg’s method example problem. (b) Henneberg’s method example
problem. (c) Henneberg’s method example problem. (d) Henneberg’s method – Basic member
forces calculation. (e) Henneberg’s method – basic member forces calculation. (f) Henneberg’s
method – self equilibrium force calculation.
110 Introduction to Structural Analysis

provided an additional new bar connecting the far ends of the truss as shown in
figure to maintain stability and the overall integrity of the structure.
Now let us analyze the truss for basic load in the members connected at the
nodes of the target member by the method of joints.
From Figure 7.27 (d), we get,

∑F = 0
y

N2 = −N3

∑F X =0

2 2
− N2 + N3 − 70 = 0
2 2

or,

N3 = 49.5 kN

From the left-hand side force diagram, we get,

∑F X =0

N4 = N5

∑F = 0
y

2 2
40 + N4 + N5 = 0
2 2
40
N5 = − = −28.28 kN
2

Let us denote the unknown member force on the indicated member as N1. As per
Henneberg’s method, we need to carry out two separate analysis to determine
N B1, which is normal member force coming from basic force analysis, and N11,
which is generated from self-equilibrium force analysis with unit loads as will be
indicated in the force diagram. At first, we will determine N B1 and then N11.

∑M c =0

2 2
N5 × 2 + N3 × 4 + N1B × 2 = 0
2 2

or,

N1B = −7.5 kN
Plane Trusses and Space Trusses 111

Now we will carry out analysis for self-equilibrium forces.

∑F = 0
y

N2 = −N3

∑F X =0

2 2
− N2 + N3 − 1 = 0
2 2
N3 = 0.7071 kN

∑F = 0
y

Again, we get from left side force diagram,

N4 = −N5

∑F X =0

2 2
1− N4 + N5 = 0
2 2
or,

N5 =   −0.7071 kN

∑M c =0

2 2
N5 × 2 + N3 × 4 + N11 × 2 = 0
2 2
N11 = −0.5 kN

We have finally,
N B1 −7.5
N1 =   −   =  = 15 kN
N11 −0.5

So, by applying Henneberg’s analysis method, we can easily determine force on

any member without dealing with a large number of equilibrium equations, which
may result if we adopt the method of joints or the method of sections.

7.7  COMPOUND TRUSSES

In the beginning of this chapter, it was shown that compound trusses are formed by
joining two or more simple trusses together either by bars or by joints. Occasionally
112 Introduction to Structural Analysis

this type of truss is best computed by applying both the method of joints and the
method of sections. Hence, the same procedure for simple truss can be considered
for compound truss analysis.

7.8  SPACE TRUSSES

A space truss is the 3D version of plane trusses. An ideal space truss is formed by
joining of rigid members connected by ball and socket joints. As discussed earlier
for a plane truss, a three-member structure is connected to form a triangle, which
is the basic non-collapsible structure for truss. Similarly, for space truss, six rigid
members are connected together by ball and socket joint in the form of tetrahedron,
providing the basic non-collapsible space truss unit. Thus, refer to Figure 7.28 for the
basic tetrahedron formed by six members as an example of a simple space truss unit.
In the above tetrahedron, members AB and BC are joined together by ball and
socket joint at the node B. To make this triangular shape stable and rigid, we need
to connect a third member AC at the base. Similarly, other faces of the tetrahedron
are formed. Hence, after connecting all the members, we have something as shown
in Figure 7.29.
From the above geometrical object, it can be clearly seen that the unit is quite
stable even without any supports. However, we need to provide supports as per
the site condition and functional requirement of the truss. Thus, by this logic,
we can form the basic unit of simple 3D truss tetrahedrons. Similarly, we can
form large space truss by combining several unit tetrahedral trusses and join-
ing the members accordingly. Same was found applicable for the case of simple
trusses where we can join desired number of triangular units to form large simple
truss. For analysis of space trusses, it is much more convenient to use the vector
form of equations of equilibrium than the scalar form. So, in 3D space, we can
put the coordinates of each joint by choosing suitable ( x , y, z ) axes and denot-
ing member forces in terms of component of forces and unit vectors along three
axes. More clarity will be obtained once we analyze a simple space truss by vec-
tor method. The method of joints discussed earlier for plane truss can be made

FIGURE 7.28  Fundamental building element (tetrahedron) for space truss.

Plane Trusses and Space Trusses 113

FIGURE 7.29  A typical space truss fundamental tetrahedron.

directly applicable for space trusses by adopting simple vector equilibrium equa-
tion as follows for each joint:
 
∑ F = 0

where, 0 is the null vector. We always start from a joint where at least one force is
known to us and not more than three unknown forces are present. Similarly, the
method of section developed in earlier chapter for plane trusses may be extended
to analyze space trusses also. For that method, we need to apply force vector and
moment vector equilibrium equations simultaneously to get the unknown forces.
The vector equilibrium equations for the method of section for space truss will be
something as the following:
 
∑ F = 0
and,
 
∑ M = 0
If the unknown support reactions are six, then according to degree of indeterminacy
formula, for statically determinate space trusses, we have the following equality in
terms of member numbers and joint numbers.

m + 6 = 3j

Also, it needs to be noted that above moment and force vector equations, there will
be total six number of equations. Hence, during drawing a section for space truss,
we need to be careful that the section does not pass through more than six members.
114 Introduction to Structural Analysis

In general, the statical degree determinacy and stability of space truss can be
denoted as:

b + r   < 3 j  → statically indeterminate −  unsatble truss

b + r   =  3 j  → startically determinate  − check stabilty
b + r   > 3 j  → statically indeterminate  − check stability

where:
b = number of members of the space truss;
r = number of reactions;
j = number of joints.

Example 7.3:  Determine the reactions at the supports and the force in
each member of the space truss as shown in Figure 7.30.

SOLUTION:  Static Determinacy: The truss contains nine members and five joints
and is supported by six reactions. As, m + r = 3 j and the reactions and the truss
members are properly arranged, it is statically determinate.
Zero-force members: At joint D, three members, AD, CD, and DE, are con-
nected. Of these members, AD and CD lie in the same (xz) plane, whereas DE
does not. Since no external loads or reactions are applied at the joint, member DE
is a zero-force member.

FDE = 0

DE is a zero-force member. The remaining two members, DA, and DC are not
colinear. So, they must be zero-force members.

FAD = 0

FDC = 0

Support reactions:

+ ∑F Z =0

BZ + 70 = 0

BZ = 70 kN 

Taking moment about the Y-axis,

+ ∑M Y =0

BX × 3 + 70 × 6 − 70 × 3 = 0

BX = 70 kN ←
Plane Trusses and Space Trusses 115

FIGURE 7.30  (a) Example problem of a 3D space truss. (b) Example problem of a 3D space
truss: Joint equilibriums.

→+ ∑F = 0
x

−70 + CX = 0

C X = 70 kN →

Taking moment about the X-axis,

+ ∑M X =0
116 Introduction to Structural Analysis

− AY × 3 − BY × 3 + 115 × 1.5 + 70 × 6 = 0

AY + BY = 197.5 

↑+ ∑F = 0
y

AY + BY + CY − 115 = 0

197.5 + CY − 115 = 0

CY = 82.5  kN  ↓

Taking moment about the Z-axis,

+ ∑M Z =0

BY × 6 − 82.5 × 6 − 115 × 3 = 0

BY = 140  kN ↑

AY = 57.5  kN ↑

The projections of the truss members in the X, Y, and Z directions, as obtained

from Fig. 7.30 (b) and their lengths computed from these projections are tabulated
in the following.
Joint A: See Fig. 7.30 (b) (i)

↑+ ∑F = 0
y

Y 
57.5 +  AE  FAE = 0
 LAE 

in which the second term on the left-hand side represents the Y component of
FAE. Substituting the values of Y and L for member AE from the Table 7.1, we write,

 6 
57.5 +  F =0
 6.87  AE

FAE = 65.84  kN  (C )

+ ∑F Z =0

 3   1.5 
− F + × 65.84 = 0
 6.71 AC  6.87 

FAC = 32.15  kN  (T )

→+ ∑F = 0
x
Plane Trusses and Space Trusses 117

TABLE 7.1
Member Lengths of the 3D Space Truss
Projection
Member X (m) Y (m) Z (m) Length (m)
AB 6 0 0 6
BC 0 0 3 3
CD 6 0 0 6
AD 0 0 3 3
AC 6 0 3 6.71
AE 3 6 1.5 6.87
BE 3 6 1.5 6.87
CE 3 6 1.5 6.87
DE 3 6 1.5 6.87

 6   3 
FAB +  × 32.15 −  × 65.84 = 0
 6.71  6.87 

FAB = 0

Joint B: See Fig. 7.30 (b) (ii)

→+ ∑F = 0x

 3 
− F − 70 = 0
 6.87  BE

FBE = 160.3  kN  (C )

+ ∑F Z =0

 1.5 
−70 − FBC +  × 160.3 = 0
 6.87 

FBC = 35  kN  (C )

As all the unknown forces at joint B have been determined, we will use the remain-
ing equilibrium equation to check our computations:

↑+ ∑F = 0
y

 6 
140 −  × 160.3 = 0
 6.87 
118 Introduction to Structural Analysis

Hence, it is fine.
Joint C: See Fig. 7.30 (b) (iii)

↑+ ∑F = 0
y

 6 
−82.5 +  F =0
 6.87  CE

FCE = 94.46  kN  (T )

7.9  ZERO-FORCE MEMBERS OF TRUSSES

Some members do not carry any force in a truss system, or the force magnitude
in those members is zero. These types of members are called zero-force members.
These types of members are added to increase the stability or rigidity of the truss sys-
tem and to provide support during different loading conditions. The method of joints
will be greatly simplified, if one can identify zero-force members at the beginning.
By inspecting the truss joints properly, we can identify the zero-force members
in the following ways:

1. If two noncollinear members are connected at a joint, and no external force

or support reactions are acting at that joint, then those two members are
the zero-force members. In Figure 7.31 (a) of the truss system, members
AB, AD, CE, EF are the zero-force members. We can easily proof this by
considering equations of equilibrium at joint A and E. In Figure 7.31 (b),
we can see the zero-force members have been removed before beginning
the analysis.
2. If three members form a truss joint, of which two are collinear, and there
are no external loads, or support reactions acting on the joint, then the
third noncollinear member is the zero-force member. In Figure 7.32 (b)
of the truss system, members BE and CE are the zero-force members.

FIGURE 7.31  (Example I): Truss system (a) with zero-force members (b) without zero-
force members.
Plane Trusses and Space Trusses 119

FIGURE 7.32  (Example II): Truss system (a) with zero-force members (b) without zero-
force members.

In Figure 7.32 (b), we can see the zero-force members have been removed
before beginning the analysis.
3. If all the members and external forces at a joint lie in the same plane but
one member at that joint is out of plane, then force in the out of plane mem-
ber is zero. In Figure 7.33, members AB, AF, AD, and force P are in the
same plane but member AC is out of plane. This member AC is a zero-force
member.

FIGURE 7.33  (Example I): Zero-force members for 3D space truss.

120 Introduction to Structural Analysis

FIGURE 7.34  (Example II): Zero-force members for 3D space truss.

4. If three members at a joint do not lie in the same plane and there is no
external load or support reaction at that joint, then the force in the three
members are zero-force members. In Figure 7.34, members AB, AC, and
AD are zero-force members.
 8 Beams and Frames, Shear,
and Bending Moments

8.1 INTRODUCTION
Apart from trusses discussed in the previous chapter, members of which are always
subjected to axial forces only, the members of rigid frames and beams may be sub-
jected to shear forces and bending moments (as discussed in Chapter 5) as well as
axial forces under the action of external loads. The analysis and determination of
these internal forces and moments are necessary for the design of such structures.
The aim of this chapter is to gain a deeper understanding of the analysis of inter-
nal forces and bending moments that develop in beams, and the members of plane
frames, under the action of coplanar systems of external forces and moments. We
start by defining the three types of internally induced forces and moments – axial
forces, shear forces, and bending moments – that will act on any arbitrary cross
sections of beams and the members of plane frames. Then we will discuss the con-
struction of shear and bending moment diagrams by the method of sections. We
will also discuss qualitatively the deflected shapes of beams and the relationships
between loads, shears, and bending moments. Also, we will develop the methods for
constructing the shear and bending moment diagrams using these relationships and
equations. Finally, we will introduce the concept of classification of plane frames as
statically determinate, indeterminate, and unstable, and last but not least, analysis of
statically determinate plane frames.

8.2  AXIAL FORCE, SHEAR, AND BENDING MOMENTS

Internal forces were defined in Chapter 5 as the forces and moments exerted on a
portion or section of a structure by the rest of the structure and its loading conditions.
Consider, for example, the simply supported beam shown in Figure 8.1 (a). Same
kind of figure was used in Chapter 5 while discussing the bending moment and
shear force.
As the applied force is inclined, hence, there will be horizontal force of the same,
which will act along the axis of the beam. Hence, the horizontal support reaction at
B will balance the same satisfying the equilibrium condition along x axis ( ∑ Fx = 0 )
of the beam. This implies H B = Px . Similarly, the other internal shear force and
bending moment at the arbitrary section x from the left support can be calculated by
applying the global equilibrium equations for force and moments.
It is to be noted that without these internal forces and moments (as drawn in
the free-body diagram above), section x will not be in a state of equilibrium. Also,
under a general coplanar system of external loads and reactions, two perpendicular
force components and a moment or couple are necessary at any arbitrary section to
DOI: 10.1201/9781003081227-10 121
122 Introduction to Structural Analysis

FIGURE 8.1  (a) External loading applied on a beam, (b) Its free body diagram showing
internal resistive forces at a section along with the applied force and support reactions.

maintain that section of the beam in equilibrium. The two internal forces are in the
direction of, and perpendicular to, the centroidal axis of the beam at section x, as
shown in Figure 8.1 (b). The internal force T in the direction of the centroidal axis
of the beam is called the axial force, and the internal force Vx in the direction per-
pendicular to the centroidal axis is referred to as the shear force. The moment M x of
the internal couple is termed the bending moment at the section x. To calculate the
axial force acting along the axis of the beam, we need to focus on local equilibrium
conditions for the sections of beams under consideration. From the local equilib-
rium condition, we can write from ∑ Fx = 0, T + H B = 0 , or T = − H B . Since H B has
already been calculated from the global equilibrium condition, we now have the
complete information of the magnitude and direction of the axial force acting in
the beam at the centroidal axis. Similarly, applying the local equilibrium equations
for the left section of the free-body diagram, we can write, RA − Vx = 0, or Vx = RA ,
and  + ∑ M x − x = 0 implies M x − RA × x = 0, which gives M x = RA × x for the range
0  ≤ x   ≤ l /2 (same was derived in detail in Chapter 5 also). So, we have formed equa-
tions for calculating the shear force and bending moments induced in a beam under
external loading. These equations help us to form shear and bending moment dia-
grams for the beams, which will be explained in detail in the next section.

8.3  SHEAR AND BENDING MOMENT DIAGRAMS FOR A BEAM

Shear and bending moment diagrams indicate the graphical variation of values and
direction of these quantities along the length of the member. These diagrams can be
constructed by taking sections indicated in the previous section and in Chapter 5.
Multiple sections are chosen from one end of the member to the other (usually from
Beams and Frames, Shear, and Bending Moments 123

left to right), considering a successive change in external loading along the length to
determine the equations expressing the shear and bending moment in terms of the
distance of the section from the starting point usually chosen as the origin. Shear
and bending moment values determined from these equations are then drawn as
ordinates against the position with respect to the chosen member end as abscissa
to obtain the desired shear and bending moment diagrams. The said procedure is
established by an example that follows next.

Example 8.1: Determine and draw the shear and bending moment

diagram for the following beam with respect to the given external
loading condition.

SOLUTION:  From global equilibrium condition using moment equilibrium equa-

tion, ∑ M = 0, we can write the same by taking moment of all forces about right
support point (see the free-body diagram Figure 8.3 of entire beam with equiva-
lent force).
So, from the above free-body diagram, we can write using the said moment
equilibrium condition with respect to right support B:

l
RA × l − wl × =0
2

or,

l 1 wl
RA   = wl ×   × =
2 l 2

Now applying global force equilibrium condition along y direction, ∑ Fy = 0

RA + RB = wl

or,

wl
RB = wl −  RA =
2

Hence, the two unknown support reactions are calculated first. Then we take any
arbitrary section at a distance x from left support as shown in Figure 8.2 and we

FIGURE 8.2  Simply supported beam under uniformly distributed load w throughout its
length.
124 Introduction to Structural Analysis

FIGURE 8.3  Free-body diagram of beam for global equilibrium condition with equivalent
concentrated load wl at midspan.

can draw the free-body diagram indicating all acting forces in the beam as shown
in Figure 8.4.
Now applying local equilibrium condition for force in y direction, we get,

RA − wx − Vx = 0 for 0 ≤ x ≤ l

or, Vx = RA − wx

So, shear force is a function of x, when x = 0, we have Vx = RA − 0 = RA = wl / 2,

and for x = l / 2, we have Vx = 0. This expression for shear force is valid for the entire
span of the beam as the nature of loading is same throughout the span. Now, if
x = l, from the abovementioned expression, we get Vx = RA − wl = wl / 2 − wl = − wl2 .
Since shear force is linear function of x, hence, it will be a straight line and the
values of shear force will vary from wl / 2 to 0 from left support to midspan and
from 0 to −wl / 2 from the midspan up to the right support. The shear force diagram
is shown in Figure 8.5.
Once the shear force diagram is drawn, we can now focus on drawing the
bending moment diagram of the beam. To do that, let us use the free-body diagram
of Figure 8.4 and take the moment of all forces to the left of point o as shown
in the same. Hence, we can write following local moment equilibrium condition,
 + ∑ M = 0,
x
Mx + wx   × − RA x = 0
2

for the range of 0 ≤ x ≤ l, because nature of loading is same throughout the span.

FIGURE 8.4  Free-body diagram of section x from left support of beam.

Beams and Frames, Shear, and Bending Moments 125

FIGURE 8.5  Shear force diagram for the entire span of the beam.

FIGURE 8.6  Bending moment diagram for the entire span of the beam.

On solving, we get,

x
Mx = −wx × + RA x
2

So, from the abovementioned equation for x = 0, we get Mx = −wx × ( x / 2) +

RA x = 0 . So, bending moment is zero at the support. And for x = l / 2, we get
Mx = −wl / 2 × (l /4) + RAl / 2 = wl 2 /8, which is positive. Now, if x = l , we get
Mx = −wl × (l / 2) + RAl = 0 .
So, the algebraic equation for bending moment at any arbitrary section x is
parabolic in nature with the maximum value at the apex of the parabola as drawn
in Figure 8.6.

Example 8.2: Find the shear force and bending moment for the
following cantilever beam shown in Figure 8.7.

SOLUTION:  The equivalent loading diagram is shown in Figure 8.7 (b).

Consider entire beam as free body,

+ ∑M B =0
126 Introduction to Structural Analysis

FIGURE 8.7  Problem on shear force and bending moment for a cantilever beam.

−MB + 65 × 1.8 − 39 + 227.5 × 4.45 = 0

MB = 1090.375 kN − m

↑+ ∑F = 0
y

−65 × 3.5 − 65 + VB = 0

VB = 292.5 kN

1. From A to C (see Figure 8.8 (a))

↑+ ∑F = 0
y

−65x − V = 0

V = −65x  kN

+ ∑M 1−1 =0
Beams and Frames, Shear, and Bending Moments 127

FIGURE 8.8  Problem on shear force and bending moment for a cantilever beam: SF and
BM at different sections.

x
65x +M =0
2

M = −32.5x 2  Kn − m

This V and M is valid in the region 0 < x < 3.5 m interval.

2. From C to D (see Figure 8.8 (b))

↑+ ∑F = 0
y

−227.5 − V = 0

V = −227.5 kN

+ ∑M 2− 2 =0

227.5 ( x − 1.75) + M = 0

M = 398.125 − 227.5x

This V and M is valid in the region 3.5 < x < 4.4 m interval.
3. From D to B (see Figure 8.8 (c))

↑+ ∑F = 0
y
128 Introduction to Structural Analysis

−227.5 − 65 − V = 0

V = −292.5 kN

+ ∑M 3− 3 =0

227.5 ( x − 1.75) + 65 ( x − 4.4 ) − 39 + M = 0

M = 723.125 − 292.5x  kN − m

This V and M is valid in the region 4.4 < x < 6.2 m interval. The shear
force and bending moment diagrams are shown in Figure 8.8 (d) and (e),
for various sections of the beam.

So, following the abovementioned procedure shown in the example problems,

and varying the range of the section x along the beam, bending moment and shear
force for any general class of loading can be drawn. The values, including the sign
in front of them, will dictate the positive and negative directions of shear forces
and moments and should be drawn accordingly in the diagram.

8.4 QUALITATIVE DISCUSSION ON THE

DEFLECTED SHAPE OF BEAMS
The qualitative deflected shape of a structure is simply a rough sketch of the neutral
surface of the structure, in the deformed or bend position, under the action of applied
external loading conditions. This shape is also called the elastic line for the struc-
ture. These diagrams, which can be drawn without any knowledge of the values of
deflections/bending, provide valuable information about the behavior of structures
and are very useful in computing the numerical values of actual deflections/bending.
Detailed theoretical analysis or calculating the deflection of the structure under
external loading condition will be developed in the next chapter.
According to the sign convention described in Section 5.1, a positive bending
moment or sagging moment bends a beam concave downward (or toward the nega-
tive y direction as shown in Figure 8.6), whereas a negative bending moment bends
a beam concave upward (or toward the positive y direction). Thus, the sign of the
curvature at any point along the axis of a beam can be obtained from the bending
moment diagram itself. Using the signs of bending moments, a qualitative deflected
shape (elastic curve) of the beam, which is consistent with its support conditions, can
be easily sketched (see Figure 8.9). For example, deflection line diagram of the beam,
which we have incorporated in Example 8.1, is shown in Figure 8.8.
Consider another example of a simply supported beam with a cantilever overhung
near support B as shown in Figure 8.10. Point load is acting on the beam as shown
in Figure 8.10.
It is to be noted that a qualitative deflected shape is approximate, as it is drawn
based on the signs of curvatures; the values of deflections along the axis of the beam
Beams and Frames, Shear, and Bending Moments 129

FIGURE 8.9  Deflection line diagram for the simply supported beam with uniformly dis-
tributed loading.

are not known. In any case as stated earlier, the deflected shape of beam always fol-
lows the nature of bending moment diagram. Also, in the case of bending moment
diagrams, there may be points of discontinuity, but for the deflected shape, and
there will not be any point of discontinuity along its entire range. Deflection line, or
more specifically, elastic line, is always a continuous function and does not have any
discontinuity.

FIGURE 8.10  Deflection line diagram and bending moment diagram for the simply sup-
ported beam with a cantilever overhung.
130 Introduction to Structural Analysis

8.5 RELATIONSHIPS BETWEEN LOADS, SHEAR,

AND BENDING MOMENTS
The formation of shear and bending moment diagrams can be analyzed by using
the basic differential relationships that relate the loads, the shears, and the bending
moments. To derive these relationships, consider a beam subjected to an arbitrary
loading, as shown in Figure 8.11. As discussed in the previous sections, we are con-
sidering a simply supported beam with uniformly distributed load all over its entire
span. But it is to be noted that the relationships that will be formed ultimately are
applicable to all types of beams with any general class of loading.
Let us take a small segment of beam of length dx. The free-body diagram of this
beam segment with all internal and external forces acting on it will be something
like as shown in Figure 8.12.
Now taking moment of all forces about the left portion of beam, we get from
moment equilibrium equation:

dx
(Vx + dVx ) dx + ( M x + dM x ) − M x − wdx =0
2
Since the last quantity is very small (square of differential quantities), hence, can be
neglected. So, the modified differential equation takes the following form:

(Vx + dVx ) dx − ( M x + dM x ) + M x = 0

FIGURE 8.11  Simply supported beam under uniformly distributed load w throughout its
length.

FIGURE 8.12  Free-body diagram of beam segment.

Beams and Frames, Shear, and Bending Moments 131

or,

Vx dx + dVx dx − ( M x + dM x ) − M x = 0

Since the second term from left is the product of two differential quantities, we can
neglect that to get:

(Vx ) dx − ( dM x ) = 0
Implies,
dM x
Vx =
dx
Hence, we get the differential equation relating to the bending moment at any section
to the shear force at the same section of the beam element. This differential equation
also tells us that the slope of bending moment line is equal to the shear force at that
point of the beam element. This equation also indicates that at the point of maxi-
mum bending moment, the shear force will be zero. So, at that point, the shear force
line will cross the beam axis and will direct toward the other portion with respect
to the beam axis. Refer to the shear force and bending moment diagram drawn in
Figures 8.5 and 8.6, respectively. For that simply supported beam with uniformly
distributed load, bending moment is maximum at the midspan of the beam, and
hence, the shear force value at the same point is zero and the shear force line crosses
the axis of the beam to progress in the upper half portion.
Integrating the abovementioned equation between two boundary points say x A to
x B, we have:
xB

∫
M x =   Vx dx
xA

which means that bending moment of two points x A and x B is equal to the area under
the curve under the same boundary points of shear force diagram.
Now, considering the local force equilibrium equation along y direction, ∑ Fy = 0,
we get:

Vx − (Vx + dVx ) − wdx = 0

or,

dVx
=w
dx

So, we get the relationship between applied load intensity and the shear force at any
arbitrary section x of the beam element. Mathematically, it indicates that the slope
of the shear force line at any point of beam is the intensity of loading applied on the
beam at that point.
132 Introduction to Structural Analysis

In the case of concentrated loads, applying the local equilibrium condition in y

direction, we can write:

P + (Vx + dVx ) − Vx = 0

or,

dVx = − P

This equation indicates that the change in shear force at the point of application of
point load is equal to the magnitude of the point load. Note that all quantities are
related to each other through the derived differential equations and hence, an abrupt
change in one value leads to the abrupt changes in other quantities also.
It is also to be mentioned here that the relationship between bending moment and
shear force was earlier derived in Chapter 2 applying the concept of Euler Lagrange’s
equation. Same needs to be checked against the above-derived relationships for bet-
ter in-depth understanding.

Example 8.3:  Find the shear force and bending moment for the following
simply supported beam shown in Figure 8.13.

SOLUTION:  Finding reaction forces:

+ ∑M A =0

or,

−50 × 3 − 160 × 6 − 80 × 9 + RB × 12 = 0

∴ RB = 152.5 kN

RA = ( 50 + 160 + 80 ) − 152.5 = 137.5

FIGURE 8.13  Problem on a shear force and bending moment diagram for a simply sup-
ported beam.
Beams and Frames, Shear, and Bending Moments 133

Shear force diagram:

Point A: Since a positive (upward) concentrated force of 137.5 kN magnitude
acts at point A, the shear diagram increases abruptly from 0 to +137.5 kN at this
point.
Point C:
The shear just to the left of point C,
SB ,L = S A,R + area under load diagram between just to the right of A to just to
the left of C.

∴ SB ,L = 137.5 + 0 = 137.5 kN

and,

SB ,R = 137.5 kN − 50 kN = 87.5 kN

Point D:
SD ,L = SC ,R + area under the load diagram between just to the right of C to just
to the left of D.

∴ SD ,L = 87.5 + 0 = 87.5 kN

and,

SD ,R = 87.5 kN − 160 kN = −72.5 kN

Point E:
SE ,L = SD ,R + area under the load diagram between just to the right of D to just
to the left of E.

∴ SD ,L = −72.5 + 0 = −72.5 kN

and,

SE ,R = −72.5 kN − 80 kN = −152.5 kN

Point B:

SB ,L = −152.5 + 0 = −152.5 kN

SB ,R = −152.5 kN + 152.5 kN = 0

The shear force diagram is shown in Figure 8.14.

Bending moment diagram:
Point A: No couple is applied at end A. So, M A = 0
Point C: MC = M A +   area under the shear force diagram between A and
C. = 0 + 412.5 = 412.5 kN m.
134 Introduction to Structural Analysis

FIGURE 8.14  Shear force diagram for the example problem.

FIGURE 8.15  Bending moment diagram for the example problem.

Point D: MD = MC + area under the shear force diagram between C and D. =

412.5 + 262.5 = 675 kN m.
Point E: ME = MD + area under the shear force diagram between D and E. =
675 − 217.5 = 457.5 kN m.
Point B: MB = ME + area under the shear force diagram between D and E. =
457.5 − 457.5 = 0 kN m.
The bending moment diagram is shown in Figure 8.15.

Example 8.4:  Find the shear force and bending moment for the following
simply supported beam shown in Figure 8.16.

SOLUTION:  Finding reaction forces:

→+ ∑F = 0
x

Ax + 42 = 0

or,

Ax = −42 kN(←)

+ ∑M D =0
Beams and Frames, Shear, and Bending Moments 135

FIGURE 8.16  Problem on a shear force and bending moment diagram for a simply sup-
ported beam.

or,

− Ay × 37 + (15 × 20 × 27) + 145 + 56 × 5 = 0

∴ Ay = 230.4 kN(↑)

↑+ ∑F = 0
y

FIGURE 8.17  (a) Shear force and (b) bending moment diagram for a simply supported
beam.
136 Introduction to Structural Analysis

230.4 − (15 × 20) − 56 + Dy = 0

∴ Dy = 125.6 kN(↑)

Shear force diagram:

Point A: S A,R = 230.4 kN
Point B: SB ,R = 230.4 − (15 × 20 ) =   −69.6 kN
Point C: S C ,L = −69.6 + 0 = −69.6 kN
SC ,R =   −69.6 − 56 = −125.6 kN
Point D: SD ,L = −125.6 + 0 = −125.6 kN
SD ,R = −125.6 + 125.6 = 0 kN
The shear force diagram is shown in Figure 8.17 (a).
Bending moment diagram:
Point A: No couple is applied at end A. So, M A = 0
Point E: ME = 0 + 1769.47 = 1769.47 kN m (here slope of the bending moment
diagram is zero)
Point B: MB ,L = 1769.47 − 161.47 = 1608 kN m
MB ,R = 1608 − 145 = 1463 kN m
Point C: MC = 1463 − 835.2 ≈ 628 kN m
Point D: MD = 628 − 628 = 0 kN m
The bending moment diagram is shown in Figure 8.17 (b).

8.6  SHEAR AND BENDING MOMENT DIAGRAM OF FRAMES

In contrast to truss, frames are truss-like structures, which carry axial force as well
as moments. Hence, for frames, both force equilibrium and moment equilibrium
equations will be found useful to calculate the bending moments as well as axial
and shear forces. Like beams, we need to draw the shear and the bending moment
diagrams for each member as well. So, the connection among members of a frame is
primarily rigid, which induces joint moments as well.
As a whole (external and internal), a frame is considered to be statically determi-
nate if all forces, moments of each member of the frame can be determined by the
application of force and moment equilibrium equations alone. No additional equa-
tions are required for a complete analysis of the statically determinate frames. In
Chapter 6, the methods of determining the static indeterminacy of rigid frames were
shown in a very easy and elegant way.
Shear and bending moment diagrams of a frame are formed based on the same
principle as we have learned for beam elements. Since each frame element is capable
of carrying bending moment and shear force, hence, for all members shear and bend-
ing moment diagrams need to be drawn as part of complete frame analysis. To be
able to understand the process, let us analyze completely the following frame shown
in Figure 8.18. Method of analysis presented here can be applied to any structurally
determinate frame systems.
Let us determine the static determinacy of the above frame. For the given frame,
m = 3,  j = 4,  re = 3,   rr = 0.
Thus, the degree of static indeterminacy, Ds = ( re + 3m ) − ( 3 j + rr ) = ( 3 + 3 × 3) −
( 3 × 4 + 0 ) =  0. Hence, the frame is statically determinate. Once we get that we must
Beams and Frames, Shear, and Bending Moments 137

FIGURE 8.18  Example problem of a plane frame with external loading.

calculate the support reactions for the external loading applied on the structure. To
be able to do that, we need to apply the global equilibrium condition, considering the
frame as a whole.
Taking moment of all forces with respect to support A, we get:

5 × 30 × 30/2 −   Dy × 30 = 0

or,

Dy = 75 kN

Now, applying global equilibrium equation in y direction, we can write:

Dy +   Ay = 30 × 5

or,

Ay = 150 − 75 = 75 kN

Also applying global equilibrium condition in x direction, we get:

Ax = 0

Hence, we have calculated all support unknown support reactions of the frame by the
application of global equilibrium conditions/equations.
Now, let us draw the BMD and SFD of each member by the same principle as that
we have learned for beam elements. Let us begin the bending moment for member
AB. Since there is no external load acting on the same, hence, there will be no bend-
ing moment or shear force at the span of the beam.
For member BC, proceeding from joint B, the moment at joint B should be zero as
there is no moment coming from member AB at joint B. At midspan of member BC,
the bending moment will be w   l 2 /8, which is equal to 562.5 kNm. And at joint C,
138 Introduction to Structural Analysis

FIGURE 8.19  (a) Bending moment and (b) shear force diagrams of the example frame.

there will be no moment as for the span CD, there is no load acting on the frame and
hence, there will be no moment. So, combining all the facts, we have the following
bending moment diagram for the given frame.
Since there are no loads acting on the member AB and CD, the shear force dia-
gram for the same has not been drawn. For any general class of loading, BMD and
SFD can be drawn in the same way sketched earlier for the entire frame.

Example 8.5:  Find the shear force and bending moment for the following
frame shown in Figure 8.20.

SOLUTION:  Let us determine the static determinacy of the above frame. For the
given frame, m = 3,  j = 4,  re = 3,  rr = 0. Thus, the degree of static indeterminacy,
Ds = ( re + 3m) − ( 3 j + rr ) = ( 3 + 3 × 3) − ( 3 × 4 + 0 ) = 0 . Hence, the frame is statically
determinate.
Finding reaction forces:  + ∑ M A = 0

FIGURE 8.20  Example problem for shear force and bending moment of a 2D frame.
Beams and Frames, Shear, and Bending Moments 139

or,

−6 × 5 × 2.5 − 5 × 3 + RBy × 5 = 0

∴ RBy = 18 kN

↑+ ∑F = 0
y

or, RAy + RBy = 6 × 5

∴ RAy = 12 kN

→+ ∑F = 0
x

RAx = 5 kN

The entire frame is in static equilibrium. So, its every part is in equilibrium also.
First consider the column AC to be in equilibrium as shown in Figure 8.21.

+ ∑M A =0

FIGURE 8.21  Equilibrium of column AC.

140 Introduction to Structural Analysis

or,
−5 × 3 + MC = 0

∴ MC = 15 kN − m

↑+ ∑F = 0
y

∴ RCy = 12 kN

Now consider the beam CD to be in equilibrium as shown in Figure 8.22.

↑+ ∑F = 0
y

or,

RCy + RDy = 6 × 5

∴ RDy = 30 − 12 = 18 kN

+ ∑M C =0

or,

−MC − 6 × 5 × 2.5 + RDy × 5 + MD = 0

FIGURE 8.22  Equilibrium of column BD.

Beams and Frames, Shear, and Bending Moments 141

FIGURE 8.23  (a) Shear force diagram and (b) bending moment diagram.

or,

−15 − 75 + 90 + MD = 0

∴ MD = 0

Consider the equilibrium of column BD as shown in Figure 8.22.

↑+ ∑F = 0
y

RDy = 18 kN

The shear force and bending moment diagram is shown in Figure 8.23.
 9 Deflections of Beams
by Geometric Methods

9.1 INTRODUCTION
While acted upon by an external loading system, structures undergo deflection in
the same direction as that of the applied loading. In case external loading acts on
the structure in different directions, the resultant direction of the external loading
system dictates the deflection of the structure in that direction. Deflection of struc-
ture is a major part of serviceability requirements related to the design of struc-
ture. Structures that may be safe under the loads may not be safe under the case of
deflection. That is why all structures are being designed to satisfy both strength and
serviceability requirements simultaneously and independently. Once a structure is
found to be safe, satisfying both these conditions, it is regarded as a safe structure,
and design analysis can be concluded at that point. In this chapter, we will study
the deflection calculation for beams under different types of loading, and we will
also learn geometric methods to determine the deflection due to the application of
general class of loads acting on the structures externally. The methods of calcu-
lating structural deflection are classified into two parts, energy methods and geo-
metric method. As the chapter name suggests, we will learn the geometric method
of calculating structural deflection and energy method we will learn in a separate
chapter. The deformations we will deal with in this chapter are the linear elastic
deformations.

9.2  DEFLECTED SHAPES AND ELASTIC CURVE

Before we start developing the deflected shape geometric ideas, students are advised
to go through Section 5.4 of the equilibrium condition chapter. In that section, we
have already explained in detail that, under the application of transverse loading,
beam members undergo deflection in the direction of the applied force. Due to such
bending, the upper fibers of the beam element remain under compression, and the
bottom fibers of the beam remain under tension. To understand the concept, see
Figure 9.1 that is being reproduced here from Chapter 5.
Under these circumstances, we should be able to calculate the numerical value
of deflection by solving certain equations. To develop the equation of deflection of
a beam element under any general transverse loading, we need to look deeply into
the bending feature and its relationship with the radius of curvature of bending. See
Figure 9.2 for a typical beam element under any general transverse loading.
We have drawn the deflection line of the beam for any general class of transverse
loading, and we have marked two close section lines at, m and m′ which are ds apart

DOI: 10.1201/9781003081227-11 143

144 Introduction to Structural Analysis

FIGURE 9.1  Deflection diagram of beam under flexural loading.

on the deflection line, to have a closer look into the deflected shape of the beam. Let
us consider a small beam element ds between the two section lines. Connecting the
radius of the curved line from common center point O, let us denote the small angle
between the two radius lines as dθ . In Figure 9.2, the tangent at m makes an angle
θ with x axis. The geometric relationship between radius, angle of rotation of radius
vector, and arc length assumes the following form:

ds = rdθ

So,

1 dθ
= 
r ds

FIGURE 9.2  Differential beam element with radius of curvature for transversely loaded
beams.
Deflections of Beams by Geometric Methods 145

The modulus sign indicates we are only concerned about the numerical value of the
radius of curvature. As the angle θ decreases if the point m move moves along the
curve from A to B in Figure 9.2, hence, the actual sign will be as follows:
1 dθ
= −
r ds

In the case of pure bending, the relationship between moment and radius of curva-
ture is as follows:
1 M
= 
r EI Z

where, I Z is the moment of inertia about the z axis of the beam element (transverse
to the x direction of the beam). For all practical purposes, since the arc length will
be very small, we may write the following approximations without causing any geo-
metrical change on the curvature of the beam,

ds ≈ dx

and,
dy
θ   ≈ tan θ =
dx

So, substituting the values of ds and θ , we can write,

1 d2y
= − 2
r dx

And equating this with the moment relationship, we get,

d2y M
=−
dx 2 EI Z

Thus, the above differential equation is the required relationship between deflection
line and the bending moment of a beam element we sought for. It should be noted
that the sign in this equation depends on the direction of the coordinate axis. Solving
this differential equation with proper initial and boundary conditions, we can deter-
mine the deflection line of a beam element under any type of transverse loading
system. Let us go through a simple example of a simply supported beam with point
loading at the middle span of the beam. We want to calculate the deflection at middle
point of the beam where the load is being applied. From the qualitative discussion,
it is clear that deflection becomes maximum just under the load with maximum
intensity. Hence, we can assume that for the simply supported beam, let us find the
maximum deflection from the differential equation itself. This method of solving the
differential equation twice to get the deflection is known as the double integration
method, which will be introduced in the following section.
146 Introduction to Structural Analysis

9.3  DOUBLE INTEGRATION METHOD

From Figure 9.3, and earlier discussion, the bending moment at any section x from
the left support A is given by,
P
M = RA   x = x
2
So, the value of bending moment at the midspan of the beam (i.e., at x = l /2) is:

P Pl
M = RA   x = x=
2 4
Inserting the general expression for bending moment in the differential equation we
have,

d2y M P
=− =− x
dx 2 EI Z 2 EI Z
or, integrating the above w.r.t x we have,

dy P
= − x 2 +   c1
dx 4 EI Z
where, c1 is the integration constant. Integrating the above differential equation once
again we get,
P
y = − x 3 +   c1 x +   c2
12 EI Z
Now, we know that for any beams, at support, deflection will be zero. So, at
x = 0, y = 0 . Putting this boundary condition in the above expression, we get,

c2 = 0

So, we are left with the following expression for deflection line with one unknown
parameter, c1. To determine this parameter, we recall our earlier discussion on the
location of maximum value of deflection. The maximum deflection should occur at
the same point where the applied load has the maximum intensity. Hence, at x = l /2,

FIGURE 9.3  Simply supported beam with point load.

Deflections of Beams by Geometric Methods 147

the deflection should be maximum. So, at this point, the slope of the deflection line
should be zero, i.e., at this point dy /dx = 0.
Applying the mentioned boundary condition, we get,
2
P l
0=−   + c1
4 EI Z 2
or,
Pl 2
c1 =
16 EI Z

With this, the final form of deflection line relationship becomes,

P Pl 2
y=− x3 + x
12 EI Z 16  EI Z

Now, putting x = l /2 to get the desired deflection value at the middle span of beam,

Pl 3
y=
48  EI Z

Thus, by solving the differential equation with appropriate boundary conditions,

we can determine the actual deflection at any point of a transversely loaded beam.
Double integration method also provides the essential tool for solving the differential
equation by numerical integration method to calculate the deflection with a good
degree of accuracy.

9.4  MOMENT-AREA METHOD

Expanding the idea of double integration, we will learn a very simple yet powerful
tool in calculating the deflection of beam. This method is known as moment area
method. This method was developed by Charles E. Greene in 1873. To understand
this method, let us once again write down the general differential equation for deflec-
tion of beams:
d2y M
=−
dx 2 EI Z
Integrating the above general equation once between two arbitrary points A and B
on the beam, we get:
B
dy M

dx
= θ BA = θ B − θ A = −
EI Z
dx
∫
A

This integral gives the change in slope between two arbitrary points A and B on
the beam. Again, it should be noted that the sign in this equation depends on the
direction of the coordinate axis. For example, if we take y-axis positive upwards,
148 Introduction to Structural Analysis

θ ≈ − dx
dy
. Now we will obtain plus instead of minus. θ A and θ B are the slopes of the
elastic curve at point A and B, respectively, with respect to the undeformed axis of
the beam. θ BA denotes the angle between the tangents to the elastic curve at A and
B and ∫ BA EIMZ dx represents the area under the EI
M
diagram between points A and B.
This concept is known as the first moment area theorem. Continuing further, if we
integrate it once more, we have the following relationship:
B
M
yBA = −
∫ EI
A
Z
xdx

This relationship is the desired relationship that we are looking for. Here, yBA repre-
sents the tangential deviation of point B from the tangent at A, which is the deflection
of point B in the direction perpendicular to the undeformed axis of the beam from
the tangent at point A, and ∫ BA EIMZ x dx represents the moment of the area under the EI
M

diagram between points A and B about point B. This concept is known as the second
moment area theorem. Unlike the double integration method, moment area theorem
totally depends on the shape and type of bending moment diagram. So, to apply this
theorem, one needs to be very careful about drawing the bending moment diagram
accurately. These two concepts are explained through Figure 9.4.

FIGURE 9.4  Deflection of a beam: moment area method.

Deflections of Beams by Geometric Methods 149

FIGURE 9.5  Deflected shape of the beam and its bending moment diagram.

To understand this method, let us consider the same problem of simply supported
beam with point load acting at its midpoint as shown in Figure 9.5.
The deflected shape and the bending moment diagram for this beam is shown as
in Figure 9.5.
Now, to calculate deflection at midpoint of the beam, we need to draw two tan-
gents at point A and B of the deflection diagram as shown in Figure 9.5 (a). Therefore,
we can write θ A = yBA
L in which θ A is assumed to be small, so that tanθ A ≈ θ A . The
tangential deviation yBA can be expressed as:

yBA = moment of area of the M

EI diagram between A and B about B

1   1 L PL  1 L    1 L PL  L 2 L    PL3
i.e., yBA =   2 × 2 × 4 ×  3 × 2   +  2 × 2 × 4 ×  2 + 3 2    = 16 EI
EI    

PL3 PL2
Therefore, θ A = yBA
L = 16 EI × 1
L = 16 EI
150 Introduction to Structural Analysis

Again, from the second moment area theorem we can get the tangential deviation
yCA , as:

1  PL2 1 L  PL3
yCA = × × =
EI  16 3 2  96 EI

If the deflection is small, we can write ( yC + yCA ) = L2 × θ A,

where yC is the deflection of the beam under the load, and yCA represents the tangen-
tial deviation of point C from the tangent at A. Now, if we put the values of yCA and,
θ A in the above expression, we get the value of yC , which is the desired deflection
amount at the midpoint of the beam under this loading condition.
2 3 3
Therefore, yC = L2 × θ A − yCA = L2 × 16PLEI − 96PLEI = 48PLEI
This is in exact agreement with the earlier found value of deflection calculated using
double integration method.
Students are encouraged to apply this concept for several different kinds of beams
with different kinds of loading. This method gives a very quick assessment of the
deflection value and is purely geometrical in nature. However, this method has cer-
tain limitations. In the case of complex loading and irregular geometry of bending
moment diagram, it is difficult to determine the distance of CG points from points
of concern. Thus, without correct bending moment and CG distance, this method
does not provide the correct result. However, for simple loading and symmetry, this
method gives a very quick assessment of the deflection values. It is to be noted, for
any general loading system, irrespective of symmetry, double integration method
provided the most general way to get the desired result. Double integration method
does not require any kind of diagram to complete the analysis. Only deflection line
equation and few boundary conditions as per the nature of loading and support con-
ditions are sufficient to determine the exact value of deflection at any point on the
beam element. For ease of reference, few geometric shapes with area and distance of
CG are shown in Figure 9.6, which may be found useful while applying the moment
area method for different types of loads.

9.5  CONJUGATE BEAM METHOD

Continuing the discussion further on the deflection topic, let us learn another simple
yet powerful geometric method known as conjugate beam method. We begin this
from the same equation as that has been used for deflection line. This method was
proposed by Prof. Otto Mohr in 1868 and later on developed by Prof. Muller-Breslau
in 1885. In this method, sketching of elastic curve is not required. The name conju-
gate beam refers to an imaginary beam of same length as that of the original beam.
The width of the beam is assumed to be of the length EI. And the beam is loaded
with the bending moment diagram of the original beam. The slope and deflection are
then calculated by the following two theorems known as Mohr’s Theorem.
The shear force at any section of the conjugate beam is equal to the slope of the
elastic curve at the corresponding section of the original beam.
The bending moment at any section of the conjugate beam is equal to the deflec-
tion of the elastic curve at the corresponding section of the actual beam.
Deflections of Beams by Geometric Methods 151

FIGURE 9.6  Some common geometrical figures with CG distance and area.

To simplify the formation of conjugate beam, we simply load it with M/EI and
then apply Mohr’s theorems as stated above. Apart from the above theorems, there
are a few key steps one needs to follow to apply the theorem correctly. See Table 9.1
for such points that need to be followed while constructing conjugate beams.
To better understand the process, we take the following example of a cantilever
beam loaded with a point load at its free end as shown in Figure 9.7. We will apply
the concept of conjugate beam to calculate the slope and deflection at the free end of
the beam by conjugate beam method.
First, let us draw the bending moment diagram of the above cantilever beam. It is
as shown in Figure 9.8.
The bending moment at the free end of the beam is zero and it varies linearly up to
support. At the support, the bending moment is negative with value Pl. Now to draw
the conjugate beam, we first draw the axis line of the beam as the same length as
that of the original beam. Then, we draw the bending moment divided EI as applied
loading on the same beam. Following the above table, the fixed end of cantilever
beam will be replaced by free end and the free end of original beam will be replaced
152 Introduction to Structural Analysis

TABLE 9.1
Supports for Conjugate Beam
Actual Beam Conjugate Beam Remarks
Fixed end supports Free end supports Slope and deflection at the fixed end of real beam
is zero. So, the shear force and bending moment
values of conjugate beam is zero at the two
supports.
Free end Fixed end Slope and deflection at the free end of original
beam is not zero. So, the shear force and
bending moments at these locations will also not
be zero.
Simply supported or Simply supported Slope at the supports for original beam exists but
roller supported end end deflection is zero at both supports. So, in
conjugate beam, shear force exists at the
supports but bending moment should be zero at
both supports.

FIGURE 9.7  Cantilever beam with point load at its free end.

FIGURE 9.8  Bending moment diagram.

by fixed end in the conjugate beam. Thus, following these steps, the conjugate beam
with all modified support conditions and loading will look something like as shown
in Figure 9.9.
Now, we are left with a simple task. To calculate the deflection at free end of
original beam, we simply need to calculate the bending moment at that point of the
conjugate beam. From Figure 9.9, the bending moment at the support of the conju-
gate beam is given by:
1 Pl 2l Pl 3
yB = Bending moment at support  B =   × l × =
2 EI 3 3EI
Deflections of Beams by Geometric Methods 153

FIGURE 9.9  Conjugate beam of cantilever beam.

So, it is fairly simple process, and we get the correct expression for calculating deflec-
tion at the free end of the cantilever beam. Similarly, for slope of elastic line at this
end of real beam, we need to calculate the shear force at the same point of the conju-
gate beam. Hence, slope will be equal to shear force at support B of conjugate beam,
which is denoted as ½ Pl /EI ×   l = Pl 2 /2 EI . These values can be crosschecked by
the earlier discussed methods by calculating deflection and slope at the same point
of the beam. Students are also encouraged to calculate the deflection of a simply
supported beam with point load acting at the midspan of the beam. While apply-
ing conjugate beam method, students are encouraged to take care of the necessary
changes that are required for supports.
We will conclude this section by giving another example of a cantilever beam
applied by uniformly distributed load to its entire length as shown in Figure 9.10.
As we have done in our previous example, first we need to draw the actual
bending moment diagram and then this diagram, divided by EI, will be applied
as a load on the conjugate beam. Also, the free end of the real beam will be
changed to fixed end in conjugate beam and the fixed end of the beam will be
free end for conjugate beams. These diagrams are shown in Figure 9.11 for ease
of understanding.
From the above diagram, we will now calculate the bending moment at support
B of the conjugate beam. This will be wl 2 /2 EI × l /3 × ( l − 3l /8 ) = 5wl 4 /48 EI . So,
this is the desired deflection at the free end B due to this loading.
Again, students are encouraged to check this value by calculating deflec-
tion, applying other alternative methods introduced in the earlier sections of this
chapter.

FIGURE 9.10  Cantilever beam with UDL.

154 Introduction to Structural Analysis

FIGURE 9.11  Conjugate beam with loading.

9.6  MACAULAY’S METHOD

As we have seen, the double integration method is very tedious and involves consid-
ering several correct boundary conditions depending on the geometry of the deflec-
tion line. Macaulay’s method provides a simple method of integration in such a way
that the constants of integration are valid for all sections of the beam; even though
the law of bending moment varies from section to section. The following rules need
to be followed while applying Macaulay’s method of deflection calculation:

1. Origin is always to be considered at the left support of beam.

2. Positive and negative moments are considered anticlockwise and clockwise
direction, respectively.

With the above mentioned points, let us calculate the deflection of the simply sup-
ported beam under the application of point load at its midspan. As stated, we con-
sider that origin of the beam is at the left support of the beam. Now the bending
moment at any section x from the left support is given by M   =   Px /2. In this case,
the range of x varies from 0 to l /2; for a section more than l /2 from left support,
the bending moment will be M = Px /2 − P ( x − l /2 ). Refer Figure 9.12 for ease of
understanding.

FIGURE 9.12  Simply supported beam with point load at midspan.

Deflections of Beams by Geometric Methods 155

So, for the entire span of the beam, the bending moment equation can be written
as follows:

M = Px /2 − P ( x − l /2 )

When we want to calculate the bending moment within the range 0 ≤ x ≤ l /2, then,
we will only use the first half of the equation. When we want to calculate bending
moment within the range l /2 ≤ x ≤ l, we will use the complete equation as written
above. Inserting the above expression of bending moment in elastic line equation we
have:

d2y Px
EI 2
=− + P ( x − l /2 )
dx 2

Integrating the above equation, we get,

2
dy P P l
EI = − x 2 + C1 +  x − 
dx 4 2  2

It is to be noted that ( x − l /2 ) has been integrated as a whole and not part wise.
Integrating once again we get:

3
P 3 P l
EI y = − x + C1 x + C2 +  x − 
12 6  2

It may again be noted that the expression ( x − 2l ) has been integrated as a whole
2

and not part wise. We know that when x = 0,  y = 0 . Substituting these values in the
above equation we get,

C2 = 0

To get the above value of C2, we need to keep in mind that for 0 ≤ x ≤ l /2 , we need
to use the first half of the equation. We also know that at x = l ,  y = 0. So, for this
range we need to use the entire expression as written earlier for deflection. Putting
this value, we get,

3
P 3 P l
0=− × l + C1 × l +  0 − 
12 6  2

or,

Pl 2
C1   =
16
156 Introduction to Structural Analysis

Now, substituting this value in the differential equation, we get,

3
P 3 Pl 2 P l
EI y = − x + x+  x − 
12 16 6 2

Now, for x = l /2 , we need to use the two parts of the expression, to the left of the
doted section location line. So, deflection at midspan will be:

P 3 Pl 2 Pl 3
EI y = − x + x=
12 16 48 EI

This is in exact deflection value that was derived earlier by other methods.
 10 Energy Principles and
Deflection of Beam

10.1 INTRODUCTION
In the previous chapter, we have learned several geometric methods to determine
the deflection of beams under the application of different kinds of loading. All
these methods primarily rely on the bending moment diagrams and their geometric
shapes. Energy principles, on the other hand, are geometry independent. In energy
methods, we need to construct the equation based on energy conservation principle
or work energy theorems and on scalar relations. It means that we need not pay any
attention to the nature of bending moment diagrams and several other details related
to that bending moment diagrams. More will be understood in detail once we go
through this chapter in detail and practice the necessary mechanisms to form energy
equations.

10.2  STRAIN ENERGY AND PURE BENDING

Let us consider a straight prismatic simply supported beam AB of length l made of a
homogeneous material with two transverse external loads of magnitude P   acting on
it, as shown in Figure 10.1.
From the Bending Moment Diagram(BMD) and Shear Force Diagram (SFD), we
find that between C and D, there is no shear force acting in the beam and the moment
at this portion of the beam is uniform with value M = Pa. This condition is known
as the pure bending condition of a beam. Due to pure bending, the portion CD of the
beam will take the form of a circular arc. In this deformed state, each cross-section
of the beam, initially plane, is assumed to remain plane after bending. This beam
types is called Euler beams, although this condition is not valid for large bending or
impure bending state. Beams in which the plane sections before and after bending
do not remain plane are called shear deformable beams or Timoshenko beams. Our
principal focus will be on the Euler beams, and we will develop more details and
deeper insight into these types of beams and their deflections.
As a result of the deformation, as shown in Figure 10.2, fibers on the convex side
of the beam will be elongated slightly, whereas, on the concave side, the fibers will
be shortened. Somewhere between the top and bottom fiber, there will be a layer of
fiber that does not undergo any kind of change in length. This layer is known as the
neutral layer or neutral axis of the beam.
From the above diagram, the plane of two cross sections mn and pq intersect at a
common point O. Let us denote the angle between these two planes by dθ , and noting
that, dθ =  dx /r where 1/r is the curvature of the neutral axis. Now, at this situation,
we draw through point b on the neutral axis, a line p′q′ parallel to mn and indicating
DOI: 10.1201/9781003081227-12 157
158 Introduction to Structural Analysis

FIGURE 10.1  (a) Beam with transverse load, (b) BMD, and (c) SFD.

FIGURE 10.2  Deformed shape and neutral axis.

Energy Principles and Deflection of Beam 159

the original orientation of pq before bending. From this figure, we find that segment
cd ′ of any fiber at the distance y from neutral axis elongates by an amount d ′d = ydθ .
Since its original length is cd ′ = dx, the corresponding strain is:

  ydθ y
  =  =
dx r

If a fiber on the concave side of the neutral axis is considered, the distance y will
be negative, and the strain is also negative. Thus, all fibers on the concave side of
neutral axis will be in a state of compression. And following similar logic, we can
say, all fibers on the convex side of neutral axis will be under tension. Since we are
working well within the elastic limit, stress will be proportional to strain. Thus,

E
σ x   =   E   = y
r

This indicates fiber stress is linearly dependent on distance y from neutral axis. This
is only applicable when we are dealing with a case of pure bending and well within
the elastic limit of the beam material where Hook’s law is valid. We can locate the
neutral axis position on the cross-section by satisfying the condition that the resul-
tant force produced by the stress distribution over the cross-sectional area must be
equal to zero.
Let dA be a small elemental area of the cross section at a distance y from the
neutral axis (see Figure 10.3). Then force acting on this differential element is σ x dA .
Using the earlier derived relationship between stress and strain, we can write,

E
σ x dA = ydA
r

Since there cannot be any resultant force acting on the cross-section as the beam is
under static equilibrium, the integral of the above differential relationship over the
entire area must be zero. So,

∑F = 0 x

FIGURE 10.3  Stress distribution and location of neutral axis.

160 Introduction to Structural Analysis

or,
E

∫ r ydA = 0
From this, we can say that,

∫ y dA = 0
Ayc = 0

where A is the total area of the beam cross section and yc is the distance of neutral
axis from the CG of the beam. And since A  ≠ 0, hence, we conclude that, yc = 0. This
means that neutral axis is passing through the centroid of the beam. Total moment
(bending moment) acting in the section due to external loading is given by,

E
M=
∫ yσ x dA =
r ∫ y dA
2

The integral in the left-hand side is the second moment of area or moment of inertia
of the beam section. Hence, we form an important relationship related to the bending
moment developed inside the beam and its moment of inertia and radius of curvature
of its bending, which is given by the following relationship:

E
M= I
r
or,
1 M
=
r EI

Using the differential relationship of elastic line derivatives and radius of curvature,
we can write the following equation that we have already studied in the previous
chapter:
d2y M
=
dx 2 EI

The famous equation of elastic line we have again derived. Thus, the equation of
elastic line is a corollary of state of pure bending. Once in the state of pure bending,
the strain energy stored in the beam, per unit length (or per unit volume for three-
dimensional object), can be expressed as:
1
Strain energy (U ) =  ( stress  ×  strain )
2

or, within elastic limit, applying Hook’s law, we can write,

1 stress 2
U=  
2 E
Energy Principles and Deflection of Beam 161

From earlier discussion, we have related stress developed in an element with that of
the bending moment developed inside it. Suppose the length of the neutral axis, CD
is l, then we have (refer Figure 10.2),

rθ = l

or,

l Ml
θ= =
r EI

So, for the case of pure bending, the net work done in bending the beam by an angle
θ is given by,

1
U= Mθ
2

Using the above two equations we get,

1 1  Ml 
U= Mθ = M  
2 2  EI 

or,

1 1 M 2l
U= Mθ =  
2 2 EI

From the above equation, we can determine the strain energy due to pure bending.
If instead of total length, we investigate the strain energy element wise, i.e., by tak-
ing small length of beam element dx, we have the following integral relating strain
energy and bending moment developed inside the beam,
l
M 2 dx
U=
∫
0
2 EI

or, replacing the above by equations of elastic line, we get,

l
M 2 dx
U=
∫
0
2 EI

For linear elongation of prismatic element of length l, subjected to an axial load P,

having cross sectional area A, and volume Al, the strain energy stored in the beam
due to the elongation is given by:

1 1 ∆l 1 P Pl P 2l
U =  ( stress  ×strain ) × volume = × σ × × ( Al ) = × × × ( Al ) =
2 2 l 2 A AEl 2 EA
162 Introduction to Structural Analysis

The above relationship has been derived from Hook’s law of elasticity which is as
follows:

Stress
=E
Strain
or,
P /A
=E
∆l /l
From the above, we have,

Pl
∆l = 
AE
By inserting this expression in the strain energy relationship, we get the earlier
derived result. For a curious-minded reader, here is a hint on how to calculate deflec-
tion or elongation of member using the strain energy expression. If we differentiate
partially the strain energy expression for prismatic member of length l, we have the
following expression:
∂U Pl
=   =   ∆l
∂l AE
which is the elongation length we have found out earlier by the application of Hook’s
Law. Thus, we get an important relationship between strain energy stored in the
beam and deflection of the beam; partial derivative of the strain energy stored in a
beam with respect to the length variable of the beam gives the deflection of beam in
the same direction of the applied force or moment induced by the applied external
loads. We will develop this concept further in subsequent sections of this chapter.

10.3  PRINCIPLE OF VIRTUAL WORK

Let us first discuss the concept of work. In mechanics, a force or a couple does work
when it displaces a body dx or rotates a body dθ amount, respectively, in the same
direction as the force or couple. The amount of external work done is a scalar quan-
tity and can be expressed as, dUe = Fdx , in case of applied force; dUe = Mdθ , in case
of applied moment. If the total displacement is ∆, the work becomes,
∆

∫
Ue = Fdx
0

If the total angular displacement is θ radian, the work becomes,

θ

∫
Ue = Mdθ
0
Energy Principles and Deflection of Beam 163

Suppose the force gradually increases from zero to a final value F = P. In this pro-
cess, the displacement reaches its limiting value ∆. If the material behaves in a lin-
ear elastic manner, the force will be directly proportional to the displacement, i.e.,
F = P∆ x . If we replace this in the earlier integral expression, we obtain: Ue = 12 P∆ .
Similarly, for the case of angular rotation, we can write, Ue = 12 Mθ . If the P or M
remains constant during the displacement or rotation, the amount of work is given by
Ue = P∆ or Ue = Mθ , respectively.
Now, let us discuss about the virtual work. The principle of virtual work was first
introduced by John Bernoulli in 1717. This principle does not have any physical sig-
nificance, but it provides a powerful analytical tool for many problems in structural
mechanics. The word virtual means imaginary or not real. We already understood
that work is a dot product of force and displacement. The ‘work’ will become virtual,
if one of the product elements is imaginary. Both cannot be imaginary simultane-
ously as we need to analyze and correlate an actual structure. So, considering the
product elements to be virtual, we can develop two virtual work principles, namely,
the principle of virtual displacements for rigid bodies and the principle of virtual
forces for deformable bodies.

10.3.1 Principle of Virtual Displacements for Rigid Bodies

The principle states that, if a rigid body is in equilibrium under a system of forces
and if it is subjected to any small virtual rigid body displacement, the virtual work
done by the external forces is zero. This principle can be easily proved with the help
of a simple illustration as shown in Figure 10.4.
The free body diagram of the simply supported beam is shown in Figure 10.4 (b).
Now, suppose this beam is given an arbitrary small virtual rigid body displace-
ment from its initial equilibrium position ACB to A′C ′B ′. The total virtual rigid
body displacement can be decomposed into translation ∆ vx and ∆ vy in the x and y
direction, respectively, and a rotation θ v about A. When the beam undergoes this

FIGURE 10.4  Virtual displacement of a rigid body.

164 Introduction to Structural Analysis

virtual displacement from ACB to A′C ′B ′, the forces acting on it perform virtual
work as:

Wve = Wvx + Wvy + Wvr

where Wvx and Wvy are the virtual work done during translation in x and y direction,
respectively, and Wvr is the virtual work done during rotation. The virtual work done
during translation in x and y can be written as follows, respectively:

Wvx = Ax ∆ vx − Fx ∆ vx = ( Ax − Fx ) ∆ vx = (∑ F ) ∆ x vx

Wvy = Ay ∆ vy − Fy ∆ vy + By ∆ vy = ( Ay − Fy + By ) ∆ vy = (∑ F ) ∆ y vy

And the virtual work done by all the forces during the small virtual rotation θ v can
be expressed as:

Wvr = − Fy ( aθ v ) + By ( Lθ v ) = ( − aFy + LBy )θ v = ( ∑ M )θ A v

Now, we can write the total virtual work as:

Wve = ( ∑ F ) ∆ + ( ∑ F ) ∆ + ( ∑ M )θ
x vx y vy A v

As the beam is in equilibrium under the action of all the forces, we can write:
∑ Fx = 0, ∑ Fy = 0 and ∑ M A = 0. Therefore, Wve = 0. Hence, it proves the prin-
ciple. To this end, we will study Example 10.1 to understand applying the principle
of virtual displacement for rigid bodies in solving engineering problems.

Example 10.1:  Each rod in the following frame is uniform of mass m and
rigid. There is an acting force P at joint C in the total frame system as
shown in Figure 10.5. Calculate the angle θ between the two members
under equilibrium condition.

FIGURE 10.5  Example problem on principle of virtual displacement for rigid bodies.
Energy Principles and Deflection of Beam 165

FIGURE 10.6  Free body diagram of the frame under equilibrium.

SOLUTION:  We will solve this problem using principle of virtual displacement

for rigid bodies, which states that under any virtual displacement, total work done
by external forces is zero. Let us draw the free body diagram under equilibrium
condition with the following configuration as shown in Figure 10.6.
From Figure 10.6, under equilibrium condition, the following geometric rela-
tionships can be made,

θ
x = 2l  sin
2
so,

θ
δ x = l  cos δθ
2

similarly,

l θ
h =   cos
2 2

or,

l θ
δ h = −  sin δθ
4 2

Now, for the entire system, applying principle of virtual work for acting forces we
get:

P  δ x + 2mg  δ h = 0
166 Introduction to Structural Analysis

Substituting the virtual displacements with known parameters already derived,

we get,

θ l θ
Pl  cos  δθ − 2mg    sin  δθ = 0
2 4 2

From which, cancelling the common terms we get,

θ 2P
tan   =  
2 mg

or,

2P
θ = 2 tan−1
mg

10.3.2 Principle of Virtual Forces for Deformable Bodies

The principle states that, if a deformable structure is in equilibrium under a virtual
system of forces and couples and if it is subjected to any small real deformation con-
sistent with the support and continuity conditions of the structure, then the virtual
external work done by the virtual external forces and couples acting through the
real external displacements and rotations is equal to the virtual internal work done
by the virtual internal forces and couples acting through the real internal displace-
ments and rotations. This principle can be easily proved with the help of a simple
illustration as shown in Figure 10.7.
Suppose a truss is in equilibrium under the action of a virtual external force Pv .
Consider the free body diagram of joint C as shown in Figure 10.5 (b). As the whole
structure is in equilibrium, the joint C will be also in equilibrium. Since the problem

FIGURE 10.7  (a) Virtual forces acting on a deformable body, (b) equilibrium of joint C.
Energy Principles and Deflection of Beam 167

is two-dimensional, we can write the equations of equilibrium in x and y directions

as below:

∑ F = 0;  − F
x vCA Sinθ 2 + FvCB Sinθ1 = 0

∑ F = 0;   P − F
y v vCA Cosθ 2 − FvCBCosθ1 = 0

FvCB, and FvCA represent the virtual internal forces in members BC and AC, respec-
tively, and angle θ1, θ 2 are the angles of inclination of these members with respect
to the vertical axis. Now, let us assume that the joint C of the truss is given a small
real displacement ∆ along the positive direction of the y axis. This small deformation
∆, is consistent with the support conditions, i.e., the supports provided at joints A
and B are not displaced. In this truss, only joint C can move due to the application
of external virtual force Pv . So, the total virtual work for this truss can be written as:

Wv = Pv ∆ − FvCB ( ∆Cosθ1 ) − FvCA ( ∆Cosθ 2 )

or,

Wv = ( Pv − FvCBCosθ1 − FvCACosθ 2 ) ∆ = 0

or,

Pv ∆ = ( FvCBCosθ1 + FvCACosθ 2 ) ∆

or we can write, Wve = Wvi ; where Wve is the virtual external work done by the virtual
external force, Pv acting through the real displacement ∆, and Wvi is the virtual internal
work done by the virtual internal forces acting through real internal displacements.

10.4  DEFLECTION OF TRUSSES BY VIRTUAL WORK METHOD

Virtual work method can be applied to trusses to determine deflection of any member
under the application of the external loading. To understand the process, let us consider
the determinate truss as shown in Figure 10.8, with the external loads P1 and P2 as
applied at joint C and calculate vertical deflection at node B of the truss by this method.
Since above truss is statically determinate, hence, all member forces due to the
applied loading can be calculated by any of the methods we have learnt in the truss
analysis chapter. Let us denote the force in any member, say AC, by FAC . Also let us
take this force as tensile in nature. Hence, the elongation of the member under this
axial load will be:

FAC l
δ AC = 
AE

where A and E is the cross-sectional area and elastic modulus of the member under
study, respectively. Now, to determine the vertical deflection at point B, we apply a
unit load in the direction of our interest as shown in Figure 10.9.
168 Introduction to Structural Analysis

FIGURE 10.8  Finding deflection of truss by virtual work method.

Under the application of this unit load in the direction of vertical deflection at point
B (denoted by ∆), the net external virtual work done Wve by the unit load is given by,

Wve =   ∆ × 1

On the other hand, the total virtual internal work done Wvi for all members of the
truss is given by the following summation,

Wvi =  ∑F δ
vj j

FIGURE 10.9  Finding deflection of truss by virtual work method: unit load application.
Energy Principles and Deflection of Beam 169

where it is to be understood that the summation is over all members of the truss
and Fvj is the member force in jth member due to the unit load at point B. Since total
energy is conserved, hence, net external virtual work done due to unit load should
be same as that of the total internal work done by the truss elements. Hence, we may
write the following equation:

∆ × 1 =  ∑F δ vj j

or,

∑F
Fj L j
∆= vj
AE
where Fj is the actual member force in jth member of the truss under the application
of original loading.
Since all the terms in the right side of above equation is known, we can deter-
mine the vertical deflection at point B due to the applied force on the truss from the
above equation. We can solve this equation in an excellent tabular form, as shown in
Table 10.1, to calculate the virtual internal work done by all the truss members and,
after that, can obtain the required deflection values.
Now we can equate the virtual external work to the virtual internal work and can
solve for the desired deflection, ∆ easily.
i.e.,

∑
Fj L j
1× ∆ = Fvj
AE
Hence, the deflection at node B will be,
FAB L AB F L F L F L F L
∆ =  FvAB + FvBC BC BC + FvAC AC AC + FvCD CD CD + FvBD BD BD
AE AE AE AE AE

TABLE 10.1
Internal Work Done by Various Members of the Truss
Virtual Member

∑F
Real Member Force Due to Fj Lj
vj
Member Length (L) Force (Fj ) Unit Load (Fvj ) AE

FAB L AB
AB L AB FAB FvAB FvAB
AE
FBC LBC
BC LBC FBC FvBC FvBC
AE
FAC L AC
AC L AC FAC FvAC FvAC
AE
FCD LCD
CD LCD FCD FvCD FvCD
AE
FBD LBD
BD LBD FBD FvBD FvBD
AE
170 Introduction to Structural Analysis

10.5  DEFLECTION OF BEAMS BY VIRTUAL WORK METHOD

For beams, deflection due to application of external loading can be calculated using
principle of virtual work with unit load method as we have done for trusses in the
previous section. To understand the method, let us draw an example of calculat-
ing deflection at the midpoint C of a simply supported beam with load as shown in
Figure 10.10.
Now, from elastic line equation, we know that the slope of elastic line is related to
the internal bending moment of the beam by the following relationship:

dθ M
= 
dx EI

where M is the bending moment at a section x, from the left support point A. So,
we have,

Mdx
dθ =
EI

At this point, let the bending moment at the same point from left support A, due to
unit point load, be given by M v. So, the virtual internal work done due to the small
angular virtual displacement dθ , is given by,

Mdx
M v dθ = M v
EI

FIGURE 10.10  (a) Beam with original load, elastic line and (b) applied unit load at the
desired point.
Energy Principles and Deflection of Beam 171

However, the external work done by the unit load due to actual displacement ∆ C is
given by,

1 × ∆C

Following principle of virtual forces for deformable bodies, we have the following
relationship:
l
Mdx

∫
1 × ∆C = M v
0
EI

Since all the terms are known in the right-hand side of the above expression, hence,
evaluating the integral we can determine the actual deflection at the required point
on the beam.
We can summarize the deflection calculation by this method as follows:

1. Determine the actual bending moment at the required point on the beam
due to the actual applied load.
2. Apply unit load at the required point on the beam and calculate bending
moment due to unit load at the same point of the beam. Unit load direction
will be in the same as that for desired deflection direction of the beam.
3. Evaluate the integral as deduced above for the entire length of beam.
4. Value of the integral will be the actual desired deflection at the required
point of the beam.

10.6  DEFLECTION OF FRAMES BY VIRTUAL WORK METHOD

To determine the deflection, ∆, and rotation, θ , at a certain point of a frame, we need
to apply virtual unit load, or unit moment at that point. After that we can easily find
out the virtual external work done just by multiplying the unit load or unit moment
with the deflection, ∆ and rotation, θ of that point, respectively. As the portions of
the frame may undergo axial deformation as well as bending deformation, the total
virtual internal work done is the summation of virtual internal work due to bending
and due to the axial deformation. Therefore, the total internal virtual work (Wvi ) for
a frame can be written by the following expression:

Wvi = ∑ F  FL
v
AE  ∑ ∫ EI
+ M M
dx v

where Fv, F are the axial forces generated in the frame members due to virtual and
real forces, respectively, and M v, M are the bending moments generated due to vir-
tual and real loads, respectively. Now, if we equate this total internal virtual work
to the virtual external work, we can get the desired deflection or rotation values of a
particular point, respectively, as shown in the following:

1( ∆ ) = ∑ F  FL
v
AE  ∑ ∫ EI
+ M M
dx v
172 Introduction to Structural Analysis

or,

1(θ ) = ∑ F  FL
v
AE  ∑ ∫ EI
+ M M
dx v

The axial deformations of the frame members are much smaller than the bending
deformations and are generally neglected for common engineering materials. That is
why the first part of the right-hand side of the previous expressions is not considered
in the calculations, and we can obtain the deflection or rotation values of a particular
point, respectively as:

1( ∆ ) = ∑ ∫ MEIM dx
v

or,

1(θ ) = ∑ ∫ MEIM dx
v

Let us consider a frame as shown in Figure 10.11, with a system of external load-
ing acting on it. Here we will neglect the axial deformations of the frame members.
Suppose here that our objective is to find out the horizontal deflection at point B
due to application of these loads. To be able to do that, first we need to calculate the
actual bending moments and support reactions. Since we are dealing with statically
determinate frames, hence, this should not be a major issue.

FIGURE 10.11  (a) Frame with actual loading and (b) unit load applied at the desired loca-
tion and direction of deflection.
Energy Principles and Deflection of Beam 173

Once we complete the analysis of the frame with external actual loading, we apply
a unit load at the point B of the frame in the desired direction of deflection and again
calculate the support reactions and bending moments in all members of the frame.
Suppose we denote the actual bending moment in each member of the frame as follows:

• Actual bending moment due to external loading in member AB = M AB

• Actual bending moment due to external loading in member BC = M BC
• Actual bending moment due to external loading in member CD = MCD

Also, let us denote the bending moment in each member of the frame due to unit load
at point B as follows:

• Virtual bending moment due to point load at node B in member AB = M ′AB

• Virtual bending moment due to point load at node B in member BC = M ′BC
• Virtual bending moment due to point load at node B in member CD = MCD′

Hence, if we denote the actual horizontal deflection at node B by ∆ B, from principle

of virtual forces for deformable bodies we can write,
B C D
M dx M BC dx MCD dx

∫
1 × ∆ B = M ′AB AB + M BC
A
EI
′
EI ∫
B
+ MCD
′
EI ∫
C

or,
B C D
M dx M BC dx MCD dx

∫
∆ B = M ′AB AB + M BC
A
EI
′
∫
EI
B
+ MCD
′
EI ∫
C

Since all the terms in the right-hand side of the above relationship are known, hence,
evaluating the integral within the limit of the member lengths, we can evaluate the
actual deflection at node B due to the system of actual loads acting on the frame.
Thus, for frames, we can apply the same principles as that for the beams to evaluate
deflection at the desired direction and desired location.

10.7  CASTIGLIANO’S THEOREM

We already have seen an interesting connection between strain energy and deflection
of structural elements. As pointed out toward the end of Section 10.2, partial derivative
of internal strain energy with respect to load is equal to the deflection of structural ele-
ment toward the same force at its point of application. Although we have seen a simple
case for point loads, this statement is true for any kind of loading. Only thing we need
to be careful about is the correct expression for internal strain energy. At this point, we
would introduce Castigliano’s 2nd theorem, which is something like following:
Castigliano’s 2nd theorem: For linearly elastic structures, the first partial deriva-
tive of the strain energy of the structure with respect to any particular force or couple
gives the displacement or rotation of the point of application of that force or moment
in the direction of its line of action.
174 Introduction to Structural Analysis

Mathematically we can state the theorem as:

∂U ∂U
= ∆ i or = θ i; in which U is the strain energy; ∆ i is the deflection of the
∂ Pi ∂ Mi
point of application of the force Pi in the direction of Pi ; and θ i is the rotation
of the point of application of the couple M i in the direction of M i .

Another version of above theorem stipulates that partial derivative of strain energy
with respect to deflection provides the force acting in the direction of the deflection
at the same point where deflection has been measured. It is somewhat reciprocal to
the above theorem. People often consider this theorem as the 1st theorem and the
above theorem as the 2nd one. However, from application point of view, the measure-
ment of deflection is the major objective and that is the one we will dive deeper into.
To understand 2nd theorem and its importance, let us consider the example of a
cantilever beam with point load at its end as shown in Figure 10.12, and let us calcu-
late the deflection of the beam under the action of point load at its free end.
First, the bending moment at any section x from left support point which is:

M x =   − Px

The internal strain energy due to bending of the beam will be,
l
M x 2 dx
U = 
∫
0
2 EI

or, putting the value of M x into the above equation we get,

l
( − Px )2 dx
U = 
∫
0
2 EI

Now, to determine the deflection under point load at the free end of cantilever, we
take partial derivative of the above expression with respect to P and arrive at:
l
∂U ∂ ( − Px )2 dx
=   
∂P ∂P 2 EI ∫
0

FIGURE 10.12  Cantilever beam with point load at its end and deflection line.
Energy Principles and Deflection of Beam 175

or,
l
∂U Px 2 dx
∆=
∂P
=
∫ 0
EI

Evaluating the integral, we get,

Pl 3
∆ = 
3EI
Suppose we want to calculate the deflection of the same cantilever beam with uni-
formly distributed load instead of point load at its free end. In such a situation, we
have to apply a fictitious point load R at the free end and then we will carry out the
strain energy with the original UDL load, including the fictitious load R.
So, the bending moment at any section x from the left support will be given by,

wx 2
M x =   − Rx −
2
So, strain energy due to bending will be:

( − Rx − ( wx /2)) dx
l l 2
2
M x 2 dx
U=
∫
0
2 EI
=
∫
0
2 EI

Now the deflection will be the following partial derivative,

( − Rx − ( wx /2)) dx
l 2
2
∂U ∂
∆= =
∂ R ∂P ∫
0
2 EI

or,
l
( (
2 Rx + wx 2 /2 xdx ))
∆=
∫
0
2 EI

Now, as R was a fictitious force, hence, equating R = 0 in the above equation and
completing the integral we get,

wl 4
∆ = 
8 EI
In addition to the above, if we want to calculate the slope of deflection line at the
free end, we can do so by taking partial derivative of the strain energy with respect
to moment. Considering the same example of cantilever beam in Figure 10.13, we
want to calculate the slope of the deflection line at the free end of the cantilever. To
do so, since there is no actual external moment acting at the free end, we will apply a
176 Introduction to Structural Analysis

FIGURE 10.13  Cantilever beam with point load and fictitious moment at free end.

fictitious moment M at the free end and calculate the strain energy under the applica-
tion of both point load and fictitious moment M . So, the cantilever beam with ficti-
tious moment applied at its free end will look something as shown in Figure 10.13.
So, bending moment at any section x from support is given by:

M x =   − Px + M

Strain energy due to bending will be,

l l
M x 2 dx ( − Px + M )2 dx
U = 
∫2 EI
0
= 
∫
2 EI
0

Evaluating the integral completely we get,

P 2 l 3 MPl 2 M 2 l
U =  + + 
6 EI 2 EI 2 EI

So, the slope of deflection line at free end is given by,

∂U Pl 2 Ml
θ =  = +
∂ M 2 EI EI

Since, M is the fictitious moment, hence, putting M = 0 in the above expression we get,

Pl 2
θ = 
2 EI

Methods presented above for cantilever beam can be applied and extended to any
type of statically determinate beams and frames to calculate deflection at any point of
structure due to external loading conditions. Students should learn and understand the
method of applying this theorem to any statically determinate problems. This method
is completely independent of the geometry and shape of the bending moment and
shear force diagrams, and thus, it can be also named a geometry-independent method.
Castigliano’s theorem is somewhat generalized in nature. In a single equation,
it combines all kinds of forces (point loads, moment, torsion etc.) to its related con-
jugate displacements. If we take force as moment, the corresponding deflection or
Energy Principles and Deflection of Beam 177

displacement will be angular displacement. If we take force as point loads, deflection

or displacement will be linear displacement. Since the equation is related to energy,
which is scalar in nature, its application is more generalized than vector equations
like Newton’s Laws of motion. That is why we introduced the concept of generalized
coordinates and generalized forces in Chapter 2 of this book. Familiarity with the
generalized coordinates and force helps us to unify the concepts of mechanics in an
elegant way.
Also, compared to geometric methods, energy principles provide a much easier
way to determine the deflection of complex structures without much trouble. In geo-
metric methods, students need to draw bending moment and shear force diagram
accurately and then apply the methods to determine the deflections. In contrary,
energy does not require any such drawings to be made a priori. One needs to calcu-
late the strain energy of individual members as per the acting loads and then simply
apply principle of conservation of energy to determine the unknown deflections. In
any case, it is up to the analyst which method he or she adopts to analyze the structure.

10.8  MAXWELL-BETTI LAW OF RECIPROCAL DEFLECTIONS

Maxwell’s law is a special case of more general Betti’s law. Initially, in 1864, Maxwell
developed it. After that, in 1872, Betti presented a generalized form of it. Both laws
apply to any type of structure, whether beam, truss, or frame. These ideas will be
developed by considering a simple beam, as shown in Figure 10.14.
Suppose that this beam is subjected to two separate and independent sets of
forces, the system of forces P and the system of forces Q. The P system develops
the internal bending moment M P in the beam, while the Q system develops the
internal bending moment MQ . Let us imagine two situations. First, suppose that
the beam is at rest under the action of the P system of forces, and then we further

FIGURE 10.14  Derivation of Betti’s reciprocal law: (a) P system of forces and (b) Q system
of forces.
178 Introduction to Structural Analysis

deform the beam by applying the Q system of forces. As a second situation, sup-
pose that the reverse is true. i.e., the Q system is acting on the beam, and then we
further deform the beam by the P system of forces. In both situations, we may apply
the law of virtual work and thereby come to a very useful conclusion known as
Betti’s law, which states that: In any linear elastic structure in which the supports
are unyielding and the temperature is constant, the external virtual work done by a
system of forces P during the deformation caused by a system of forces Q is equal
to the external virtual work done by the Q system during the deformation caused
by the P system of forces. From Figure 10.14, the virtual external work done (Wve )
can be expressed as:

Wve = P1∆ Q1 + P2 ∆ Q 2 + .. + Pn ∆ Qn

or,
n

Wve = ∑P ∆
i =1
i Qi

The virtual internal work done in the beam can be expressed as:
L
M P MQ
Wvi =
∫
0
EI
dx

Now applying the principle of virtual forces of deformable bodies, we can write,

n L

∑P ∆ = ∫
M P MQ
i Qi dx
i =1
EI
0

Next, we assume that the beam is subjected to a Q set of forces and the deflections
caused by the P set of forces. By equating the external virtual work to the internal
virtual work, we can write:

m L

∑
MQ M P

j =1
Q j ∆ Pj =
∫0
EI
dx

The right-hand side of both the above expressions is the same. Thus, we can write:

∑in=1 Pi ∆ Qi = ∑ mj=1 Q j ∆ Pj; this is the mathematical statement of Betti’s law. Betti’s law
is a very useful principle and sometimes called generalized Maxwell’s law.

Maxwell’s law of reciprocal deflection states that: In any linear elastic structure in
which the supports are unyielding and the temperature is constant, the deflection
at a point i due to a unit load applied at a point j is equal to the deflection at j due
to a unit load at i. In this, the terms deflection and load are given in general sense,
Energy Principles and Deflection of Beam 179

FIGURE 10.15  Development of Maxwell’s reciprocal law: (a) P system of forces and
(b) Q system of forces.

to include rotation, and couple, respectively. Maxwell’s theorem can be realized by

observing Figure 10.15, as shown below.
As mentioned earlier, Maxwell’s law can be considered as a special case of Betti’s
law. Applying the Betti’s law, we can easily show that: fij = f ji, for the case shown in
Figure 10.15, and this is the mathematical statement of Maxwell’s theorem. In this
figure, fij represents the deflection at i due to the unit load at j, and f ji represents the
deflection at j due to the unit load at i. These deflections per unit load are referred to
as flexibility coefficients.
We will find the application of this theory in several chapters that follow. Even
in indeterminate structural analysis, we will frequently apply this principle to form
additional equations to eliminate the unknown forces and moments.
 11 Rolling Loads and
Influence Lines and
Their Applications

11.1 INTRODUCTION
In previous chapters, we have considered loads acting on the structure, which is
static in nature. It was assumed that the point of application of the load would remain
the same throughout the entire life span of the structural element. But in real-life
problems, loads shift their position in a structure. Thus, the location of maximum
internal forces and deflections due to the applied loads will also vary depending on
the position of the load at that instant. In this chapter, we will develop some methods
to analyze the structure under moving loads. For moving loads, placing unit load at
different beam locations and drawing subsequent bending and shear force diagrams
will be very useful. These diagrams will be used, at a later stage, on other structures
like frames to analyze internal forces and moments induced due to moving load
conditions. After completing this chapter, students are expected to acquire detailed
knowledge of the analysis process when a structure is subjected to moving loads.

11.2 INFLUENCE LINES FOR BEAMS AND FRAMES

BY EQUILIBRIUM METHOD
Consider a simply supported beam with unit point load at a distance x from the left
support as shown in Figure 11.1. At this location of the load, the support reactions
at the two ends of the beam can be calculated using global moment equilibrium con-
ditions (recall from Chapter 2, to determine unknown support reactions, we have
applied global equilibrium conditions on the overall structure, and to determine inter-
nal forces such as bending moment and shear forces, we applied local equilibrium
conditions).
Taking moment of all forces about point B, we get,

∑M B =0

(l − x )
RA = 1 ×
l
or,

x
RA = 1 −  
l
DOI: 10.1201/9781003081227-13 181
182 Introduction to Structural Analysis

FIGURE 11.1  Simply supported beam with unit point load at a distance x from left
support A.

So, RB will be,

x x
RB = 1 −  1 −    =  
 l  l

We developed two equations for support reactions, for the arbitrary location of exter-
nal load, and both the equations are linear functions of x. So, by varying x, we can
determine the support reactions for any location of the point load acting on the beam.
It is interesting to put x = 0 in the abovementioned equation. In that case, RA = 1
and RB = 0 . This result is quite reasonable as, in this case, point load is sitting just
at the left support point, and hence, total downward force will be balanced by the
support A, and thus support reaction at B will be zero. Similarly, by setting x = l in
the abovementioned equation, we find that, RA = 0,  RB = 1 and the reason is quite
self-explanatory. So, the influence line for support reactions for a simply supported
beam with point load can be drawn as Figure 11.2.
Main advantage of drawing influence line is that we can determine the support
reactions for any load other than unit one from the above triangles by applying prin-
ciple of similar triangles and multiplying the result by the actual load applied on the
structure. To understand this concept, let us calculate the support reaction at A when

FIGURE 11.2  Influence line diagram for support reactions of a simply supported beams.
Rolling Loads and Influence Lines 183

FIGURE 11.3  Influence line diagram for support reaction of example problem.

a load of 5 kN is acting at a distance 2 m from left support. Total length of beam is

5 m. We take help of the first influence line to determine support reaction RA.
From Figure 11.3, considering similar triangle ABC and BED, we have,

BE BA
=
ED AC
or,
BE × AC 3 × 1
ED = = = 0.6
BA 5
Since the actual load acting on the structure is 5 kN, the actual support reaction at
support A will be 0.6 × 5 = 3 kN.
Thus, the beauty of the influence line diagram can be well appreciated from the
abovementioned example. In similar way, one can find out the support reaction at B
due to any position of load on the beam using the relevant influence line diagram for
the same.
Having equipped with the concept of influence line diagrams for support reac-
tions, we can form the influence line diagrams for bending moment and shear force
also. Using the same beam as shown in Figure 11.1, let us derive the expression for
change in bending moment diagram at a distance l1 from the left support if a unit load
moves across the beam from left support to right support. Let the point of concern
be C as shown in Figure 11.4.

FIGURE 11.4  Influence line for bending moment at section C.

184 Introduction to Structural Analysis

FIGURE 11.5  Free-body diagram when unit load is at 0 ≤ x ≤ l1.

Now, for the range 0 ≤ x ≤ l1 (as shown in Figure 11.5):

x
M x =  1 −  l1 − 1 × ( l1 − x )
 l

( )
For, x =  l1, M x = 1 −   ll1 l1 ,  and for x = 0,  M x = 0. So, when unit load is placed at
x = l1, bending moment at section C will be (1 − ( l1 /l )) l1 and when unit load is placed
at left support point A, the bending moment at section C will be zero.
Now, for the range l1 ≤ x ≤ l (as shown in Figure 11.6).

M x = RB  ( l − l1 ) − 1 × ( x − l1 )

(
or, M x = xl ( l − l1 ) − 1 × ( x − l1 ) = x 1 − l1
l )
So, for x =  l1, we get,

M x =  1 − 1  l1
l

 l

And for x = l ,  M x = 0. So, when unit load is placed at right support point B, the bend-
ing moment at section C will be zero and when unit load is placed at section C, the
bending moment at the same location will be (1 − ll1 ) l1, which is in perfect agreement
with the earlier calculation.
So, the influence line diagram for the moment at point C will be linear in nature,
and its values are zero at both supports. It is customary to note the difference between
the bending moment diagram and the influence line diagram for bending moment.
In the bending moment diagram, we draw the bending moment of a beam at various
points along its span due to a fixed external load acting on it. On the other hand, in

FIGURE 11.6  Free-body diagram when unit load is at l1 ≤ x ≤ l.

Rolling Loads and Influence Lines 185

FIGURE 11.7  Influence line diagram of bending moment at point C.

the influence line for bending moment, we draw change in bending moment at any
particular section on a beam due to the movement of a unit concentrated load over its
span. The influence line for bending moment at point C due to unit point load moving
across the beam is shown in Figure 11.7.
For shear force at point C, let us write the first equation of shear force correspond-
ing to Figure 11.5.
For the range 0 ≤ x ≤ l1, we can find out the shear force by using the free-body
diagram and applying local force equilibrium equation.

x x
RA − 1 − VC = 0; VC = 1 − −1 = −
l l

So, at x = 0 , means when unit load is placed at left support point A, then shear force
at C, VC = 0 . Also, for x = l1, the shear force at C will be VC = −(l1 /l ).
For the other portion of the beam, when the unit load is beyond point C, i.e.,
l1 ≤ x ≤ l, then we can use the right section about C (refer to Figure 11.6 for proper
unit load location).
Constructing the shear force equation by using the free-body diagram shown in
Figure 11.6, and applying local force equilibrium equation,

x
VC − 1 + RB = 0; VC = 1 −
l

So, for x = l1, we have VC = 1 − ( l1 /l ) and for x = l, we have VC = 0 . And also, it is to

be noted that all the equations for shear forces are linear in x. So, influence line
for shear force at C due to varying position of unit load can be drawn as shown in
Figure 11.8.
As the unit load moves from A to C, the shear at C decreases linearly until it
becomes −(l1 /l ) , when the unit load reaches just to the left of C. As the unit load
crosses point C, the shear at C increases abruptly to 1 − ( l1 /l ). It then decreases
linearly as the unit load moves toward B until it becomes zero when the unit load
reaches the right support B.
For frames, similar mathematical equations can be formed depending on the
geometry of the frame and the chosen point of reference for which the influence
186 Introduction to Structural Analysis

FIGURE 11.8  Influence line diagram of shear force at point C.

lines need to be drawn. To understand this concept, let us take an example problem
for a frame as shown in Figure 11.9.
Now, we want to draw the influence line for vertical reaction and support moment
at point A of the above frame. To be able to do this, let us first draw the free-body
diagram as shown in Figure 11.10 of the above frame with the applied unit load.
Now for support reactions, applying global force equilibrium condition, we get,

RA − 1 = 0
or,
RA = 1

And, applying global moment equilibrium conditions about point A, we get,

M A − 1 × ( x − 6) = 0

FIGURE 11.9  Example problem for frame structure.

Rolling Loads and Influence Lines 187

FIGURE 11.10  Free-body diagram of frame.

or,

M A = 1 × ( x − 6)

So, for x = 0 , M A = −6, and when x = 6,  M A = 0. Also, for x = 18, M A = 12

So, with the above-calculated values and since all the equations are linear func-
tions of x, the influence line diagrams for support reaction and support moment at
point A can be shown in Figure 11.11 (a) and (b), respectively.

FIGURE 11.11  Influence line diagram for the example frame.

188 Introduction to Structural Analysis

So, by following this method, we can draw influence line diagrams of any response
function (reaction, shear, bending moment etc.) for any frame for any reference point
selected on the frame. Having gained enough knowledge about the construction of
influence lines, now, we will learn another important topic known as Müller-Breslau
principle for drawing qualitative construction of influence line diagrams.

11.3 QUALITATIVE INFLUENCE LINES

AND MÜLLER-BRESLAU’S PRINCIPLE
For quantitative construction of influence lines, the method based on equilibrium
equations is found to be very useful. In contrast to quantitative analysis, and in real
life problems, we need to understand the nature and behavior of influence lines with-
out calculating in detail the numerical values. It is the shape and nature of influ-
ence line that are of utmost important for analyzing any structure qualitatively and
instantly. Müller-Breslau principle provides the best tool for qualitative analysis
of structures to construct influence lines for shear, support reactions, and bending
moment. However, this principle cannot be used to construct deflection influence
line diagrams as will be apparent when we will learn in detail about this principle.

11.3.1 Müller-Breslau Principle
The influence line equations for force or moment can be derived using the deflected
shape of the released structure. The released structure is obtained by removing the
restraint related to the point for which the influence lines are to be drawn. Then, we
need to apply unit displacement or rotation at that location and in the direction of
the force or moment at the reference point of the structure so that only the force or
moment at the reference point and the unit load do the external work.
To understand the actual working procedure, as per the above principle, we need
to go through an example. Let us consider the same simply supported beam as was
shown in Figure 11.4 and try to form the influence line equations for bending moment
at point C of that beam by applying this principle. As our objective is to find the
equation for bending moment influence line at point C, we remove the restraint at C
by introducing a hinge connection at that point (remember in hinge there will be no
moment, and thus, we can eliminate the bending moment at C by this method). The
beam with a new hinge connection at point C will look something like as shown in
Figure 11.12.

FIGURE 11.12  Released structure of simply supported beam with hinge at point C.
Rolling Loads and Influence Lines 189

FIGURE 11.13  Released structure moment and virtual displacement diagrams.

As we have provided a hinge at point C, the beams AC and CB are now free to
rotate about this point. To keep the released beam in equilibrium, we apply a moment
at C, which is shown in Figure 11.13.
Then, we apply a virtual rotation at point C by unit value, and due to this rota-
tion, AC is rotated by say θ1 and BC is rotated by say θ 2, so that, θ1 +  θ 2 = 1. Refer to
Figure 11.13 for the displaced portion of the beams.
Now, applying the principle of virtual work, we get,

W = MCθ1 + MCθ 2 − 1 × y = 0

which implies,

MC (θ1 + θ 2 ) = 1 × y

or,

MC = y

which indicates that the deflected shape of the beam is equal to the influence line dia-
gram for the beam, which was then the claim of Müller-Breslau principle. Now from
the radius and angular displacement relationship, so, the ordinate, ∆ is given by,

∆   = l1θ1 = ( l − l1 )θ 2

or,

l1 × θ1 = ( l − l1 ) × θ 2

So,

( l − l1 )
θ1 = × θ2
l1
190 Introduction to Structural Analysis

Now, using the relationship, θ1 +  θ 2 = 1, we get,

l1
θ2 =
l

So, we get,

( l − l1 ) l1 l
∆= = l1  1 − 1 
l  l

This is in exact agreement with earlier derived expression for bending moment influ-
ence line equation for the portion CB of the beam by applying equilibrium methods.
By following the same steps, we can form the other influence line equations for reac-
tion and shear forces as well.
Although Müller-Breslau principle can determine the influence lines quantitatively,
but in real-life problems, this principle is mostly used to determine the influence lines
qualitatively. Qualitative nature of the influence lines helps us to analyze nature and
qualitative failure mode analysis for large structures and individual structural ele-
ments as well. For detailed analysis purpose, one may adopt Müller-Breslau principle
to obtain the qualitative influence line diagrams and then apply equilibrium methods
to obtain the complete influence line diagrams with numerical values.

11.4  INFLUENCE LINES FOR FLOOR GIRDERS

In the previous section, influence lines for beams and frames for reactions, shear, and
bending moments have been provided. We have developed the equations for influ-
ence lines for the case when unit load is directly moving on the same system. In most
bridges and building, there are main structural elements on which direct application
of loading is not possible. For example, let us consider the case of bridge girders. In
most of the structural bridge girders, deck is supported on smaller steel members
known as the stringers. Stringers are then supported on structural beams that are also
supported on the main beams known as girders. Thus, for girders, no matter what
type of loading has been applied on the deck level, it will be always transferred as
point loads. For structural analysis and design, it is customary to determine the actual
shear and bending moment values acting on the girder due to load applied on the deck.
To illustrate the necessary steps to draw influence lines for bridge girders, we
assume that the girders are supported at the ends as a simply supported beam. Refer
Figure 11.4 (a) for better understanding of the stringer/floor beam/girder system.
Influence line for support reactions will be same as that for the simply supported beam
we have already discussed in the first section of this chapter and can be derived as next.

+ ∑M F =0

i.e.,

− RA × l + 1 × ( l − x ) = 0
Rolling Loads and Influence Lines 191

or,
x
RA = 1 −
l

+ ∑M A =0

i.e.,

RF × l − 1 × x = 0

or,
x
RF =
l

Hence, influence lines are drawn without much ado for the support reactions as
shown in Figure 11.14 (b) and (c).
Next, let us consider shear force at a point G in the panel BC as shown in
Figure 11.14 (d). When the unit load is located to the left of point B, then shear force
at any point within the panel BC will be:

x l
VBC + RF = 0, VBC = −  for 0 ≤ x ≤
l 5

When unit load is placed within the panel BC, then force at support B (as shown in
Figure 11.15) on the girder needs to be incorporated in the equation of the shear force
for panel BC. So, in this case the shear force equation will be:

x 5x 4x l 2l
VBC = RA − RB =  1 −  −  2 −  = −1 + ;  ≤ x ≤
 l  l  l 5 5

Similarly, when the unit load is located to the right of the panel point C, the shear at
any point within the panel BC is:

x 2l
VBC = RA = 1 − ; ≤ x ≤ l
l 5

Thus, from these equations, it is clear that influence line for shear force does not actu-
ally depend on the location of unit load on the panel. Thus, shear force remains con-
stant for the entire panel of girder. Hence for girders, shear force is drawn for panels
and not point to point of the entire length and for that reason, it is called as panel shear.
Now, let us find out the influence line for the bending moment at point G, located
at panel BC. When the unit load is to the left of point B, bending moment at point G
can be written as:

x l
M G = RF ( l − a ) =  ( l − a ) for 0 ≤ x ≤
l 5
192 Introduction to Structural Analysis

FIGURE 11.14  Influence line diagram for panel BC of a bridge girder.

When unit load is located to the right of panel point C, then bending moment at G
is given by:

x 2l
M G = RA × a =  1 −  a  for ≤ x ≤ l
 l 5

When unit load is placed in between panel points B and C, then bending moment at
F is given by:

l l 2l
M G = RA × a − RB  a −   for ≤ x ≤
 5 5 5

Upon simplification, we get M G = ( 2l /5) − a − x (1 − 4a

l ) for 5l ≤ x ≤ 25l .
Rolling Loads and Influence Lines 193

FIGURE 11.15  Finding shear force for panel BC when the unit load is within this panel.

Thus, unlike shear, bending moment for panels varies along point to point on the
panel and is dependent on the location of the chosen point (point G in our case) on
the panel. Since all the equations are linear in x, the bending moment influence line
will also consist of straight-line segments as that found for shear force also. Using the
abovementioned equations, we can draw the influence line for bending moment for
panel BC as shown in Figure 11.16.
Following the same procedure, one can complete the influence lines for bending
moment diagrams for every panel of the girder and same has been left as an exercise.

FIGURE 11.16  Influence line diagram for bending moment at point G in panel BC of a
bridge girder.
194 Introduction to Structural Analysis

11.5  INFLUENCE LINES FOR TRUSSES

The flooring system used for bridge girders is also used for trusses. The floor or
deck slab is supported by stringer beams and stringer beams are supported on floor
beams. Floor beams are connected to the nodes of longitudinal trusses. Hence, no
matter the type of load moving on the deck, force transmitted to the truss at the
nodes are always point loads. A typical arrangement of longitudinal truss with deck-
ing arrangements is shown in Figure 11.17 for ease of understanding.
Like bridge girders, trusses are assumed to be supported freely at the ends. Hence,
influence lines for trusses will also consist of straight-line segment as it was derived
and drawn for bridge girders in the previous section.
To illustrate the mechanism of constructing influence lines for trusses, let us con-
sider the truss shown in Figure 11.17. On the floor slab, let us assume a unit load is
moving from left to right. Load from the floor will be transferred to the stringer
beams, from stringer beam to floor beams, and from floor beam to the nodal points
of the truss. Our target is to draw influence lines for vertical support reactions at
points A and D and to draw influence lines for member force in the member EB.
Influence lines for reactions are obtained by following the same procedure as that for

FIGURE 11.17  Typical deck/floor slab supporting arrangements for truss.

Rolling Loads and Influence Lines 195

FIGURE 11.18  Influence line diagrams for support reactions of the truss.

simply supported beams. Thus, applying the equations of equilibrium, the equations
of influence lines for the vertical reactions RA and RD can be obtained as shown in
Figure 11.18.

+ ∑M D =0

i.e.,

− RA × l + 1 × ( L − x ) = 0

or,

x
RA = 1 −
L

+ ∑M A =0
196 Introduction to Structural Analysis

i.e.,

RD × L − 1 × x = 0

or,

x
RD =
L

After completing the support reaction influence lines, now, we can move ahead to
determine the equations for influence line of member force for the member EB.
When the unit load is at the left of point B, then applying the method of sections as
shown in Figure 11.19, we can form the local equilibrium force equations as follows:
Applying local force equilibrium equation in y direction, we get,

L
RA − FEB  sin 60 − 1 = 0 for 0 ≤ x ≤
3

or,

2x L
FEB =   −  for 0 ≤ x ≤
L 3 3

FIGURE 11.19  Free-body diagram of truss members applying method of sections.

Rolling Loads and Influence Lines 197

FIGURE 11.20  Influence line diagram for member force FEB .

When unit load is placed beyond point B i.e., right of point B, then it is convenient
to work with the other half of the section. From that, we can form the equation of
equilibrium easily as shown next:

L
FEB  sin 60 + RD − 1 = 0 for  ≤x≤L
3

2  x L
or,  FEB =   1 −   for  ≤ x ≤ L
3  L  3

From the abovementioned equations, we can draw the influence lines for member
force FEB as shown in Figure 11.20.
The member force FEB was assumed to be tensile (Figure 11.19 (b)) while deriva-
tion of the influence line equations. So, a positive ordinate of the influence line indi-
cates that the unit load applied at that point causes a tensile force in the member EB
and vice versa. Thus, the influence line for FEB (Figure 11.20) indicates that member
EB will be in tension when unit load is located between B and D, whereas it will be
in compression when unit load is located between A and B. It should be noted that
when the unit load just crosses point B and moves toward the right, then the load in
member EB will become tensile from compressive.
By following the above footsteps, we can form the influence lines of all other
members of the truss.

11.6 MAXIMUM INFLUENCE AT A POINT DUE

TO A SERIES OF CONCENTRATED LOADS
First, we need to find out the influence lines of any response functions (reaction
force, axial force, shear force, bending moment etc.) for a particular point in a struc-
ture. Now, the maximum effect caused by a live concentrated force can be obtained
198 Introduction to Structural Analysis

by multiplying the peak ordinate of the influence line by the magnitude of the force.
So, for any arbitrary point load P, the following two important points need to be
understood clearly:

1. To obtain the influence lines for force P, we multiply the ordinates of the
unit load influence lines with the numerical value P.
2. To obtain maximum positive value of ordinate, we need to multiply the load
P with the maximum positive unit load influence line ordinate. Similarly, to
obtain the maximum negative influence line value, we need to multiply the
maximum negative value of the unit load influence line ordinate with the
numerical load value of P.

To understand this concept, let us consider the following beam with overhang por-
tion. We wish to calculate the maximum positive and negative bending moment val-
ues due to the moving load P. To be able to do that, we first draw the influence line for
bending moment at a chosen point (say point D) for the given beam applying moving
unit load. Upon constructing the unit load influence line for bending moment, we can
now conveniently decide the location of the actual load P to obtain maximum values
as illustrated graphically in Figure 11.21.
For most of the real-life bridge and flyover problems, during analysis, all traf-
fic loads are considered as a series of many concentrated loads placed at a fixed
distance from each other. Most of the time, these points are wheel locations of the
moving vehicles like trucks or big carriage vehicles. For these types of movements,
it is essential to determine the critical locations of these series point loads for which
maximum adverse shear or bending moment is generated on the bridge structures.
Influence lines provide the easiest and accurate way to determine the locations of
these series loads for which maximum adverse conditions are met. Design against
these adverse data provides a safe and reliable design of bridges and flyovers for the
entire design life span of these structures. In the following sections, we will learn
these techniques in detail and prepare ourselves to tackle such problems in real life.
Let us consider a series of moving point loads as shown in the following diagram.
This is the pictorial representation of a large truck or vehicle class AA as per IRC
norms. However, for our analysis purpose, we only need to know the center-to-center
distance of the wheels and load acting at those points. Suppose, we want to calculate
the maximum shear force at any specific point C on the span of the beam due to this
moving load. The process of determining the maximum shear due to the moving
load is calculated by trial-and-error method.
As shown in Figure 11.22, first we place the first wheel at the reference point C
on the beam. Since we already know the shear force influence line diagram for the
unit load, from similar triangles, we can determine the ordinates of the influence line
diagram under the load positions, and then we multiply the same with corresponding
load magnitudes to get the actual shear force that will be induced by the truck at the
position of interest, i.e., C. Hence, the shear force at C due to the first load position
will be:

VC = w1s1 + w2 s2 + w3s3 + w4 s4
Rolling Loads and Influence Lines 199

FIGURE 11.21  Maximum positive and negative bending moment locations for load P.

Now, let us move the vehicle toward left support A so that now the second wheel
rests at the reference point C. At this position, we can calculate the shear force at C
following the same logic explained previously. Similarly, we will place the other two
rear wheels, respectively, at the point C and calculate the shear forces.
Proceeding as per the earlier method, we calculate the shear force value for each
different position of the wheel loads on the reference point C of the beam. Once we
calculate all these values, we can determine the maximum value of the shear force
and wheel position for the same.
When many concentrated loads act on the span, the trial-and-error method of
finding maximum shear force at a point of interest becomes very tedious. Then by
200 Introduction to Structural Analysis

FIGURE 11.22  Different position of wheel load on the reference point C of the beam.
Rolling Loads and Influence Lines 201

finding the change in shear force, ∆V from case to case, it becomes easy. If each
computed ∆V is positive, the new position will yield a larger shear in the point of
interest than the previous one. Each movement is investigated until a negative change
in shear is found. When this occurs, the previous position of loads will give the criti-
cal value.
If the slope of the influence line is S, then we can write,

y2 − y1
S=
x 2 − x1

where y2 − y1 is the change in the ordinate of influence line, let say ∆V, and x 2 − x1
is the change in the x-coordinate. So, we can write, ∆V = S ( x 2 − x1 ). Therefore, the
change in shear, ∆V for a load P that moves form position x1 to x 2 over a beam can
be determined by:

∆V = PS ( x 2 − x1 )

Now, if the load moves past a point where there is discontinuity of jump in the influ-
ence line, then change in shear can be written as:

∆V = P ( y2 − y1 )

Similarly, for a horizontal movement ( x 2 − x1 ) of a concentrated load P, the change

in moment, ∆M, is equivalent to the magnitude of the load times the change in influ-
ence line ordinate under the load written as:

∆M = PS ( x 2 − x1 )

Like shear here also, as long as each computed ∆M is positive, the new position will
yield a larger bending moment in the point of interest than the previous one. Each
movement is investigated until a negative change in bending moment is found. When
this occurs, the previous position of loads will give the critical value.
At this end, we should appreciate the importance of drawing influence line for
unit load for internal force and moments, which enables us to analyze the same for
any kind of loading as per actual requirements. Also, it is to be noted that follow-
ing the same scheme, we can find out the position of the wheel load that will cause
maximum bending moment in the beam using the unit load bending moment influ-
ence line. We will show how to calculate maximum bending moment for a moving
uniformly distributed load (UDL) using influence line diagram in the next section.

Example 11.1:  Find the maximum influence of shear at a point C due to

series of concentrated loads as shown in Figure 11.23.

SOLUTION:  Method 1: Let us first solve the problem by trial-and-error method.

Three cases have been investigated as shown in Figure 11.24.
202 Introduction to Structural Analysis

FIGURE 11.23  Example problem on finding the location for maximum influence.

Case 1:
In Case 1, let us assume that the first wheel is located at point C.
The shear force at point C due to this case,

(VC )Case  1 = 8 (0.6) + 50 (0.5) + 30 (0.35) = 40.3 kN

Case 2:
In Case 2, let us assume that the second wheel is just to the right of point C. The
shear force at point C due to this case,

(VC )Case   2 = 8 ( −0.3) + 50 (0.6) + 30 (0.45) = 41.1 kN

Case 3:
In Case 3, let us assume that the third wheel is just to the right of point C. The shear
force at point C due to this case,

(VC )Case  3 = 8 ( −0.15) + 50 ( −0.25) + 30 (0.6) = 4.3 kN

Hence, Case 2 yields the largest value for VC and, therefore, represents critical
loading position.
Method 2: In this method, finding the change in shear force, ∆V from case to
case we will obtain the load position for maximum shear at point C.
First consider the loads of Case 1, moving 1 m to Case 2. When this occurs,
8-kN load jumps down (−1) and all loads move up the slope of the influence line.
This causes a change in shear force.

∆V1− 2 = 8 ( −1) + (8 + 50 + 30 )( 0.1)(1) = +0.8 kN

Since ∆V1− 2 is positive, Case 2 will yield a larger value of VC than Case 1. While
calculating ∆V1− 2, we should note that after jumping down (−1) in the influence
line diagram, the 8-kN load has moved forward and positioned 1 m ahead from
point C. In this region also (AC), the 8-KN load has climbed a positive slope of 0.1.
Now, investigating ∆V2− 3 which occurs when Case 2 moves to Case 3. Here
the loads of Case 2, moving 1.5 m to Case 3. We must account for the downward
(−1) jump of the 50-kN load, and all loads move up the slope of the influence line.

∆V2−3 = 50 ( −1) + (8 + 50 + 30 )( 0.1)(1.5) = −36.8 kN

Rolling Loads and Influence Lines 203

FIGURE 11.24  Investigations of various cases for obtaining the maximum influence at C.
204 Introduction to Structural Analysis

Since ∆V2− 3 is negative, Case 2 will yield a larger value of VC than Case 3. Hence
the critical load position is that of Case 2 as investigated previously by trial-and-
error method earlier.
Similarly, one can investigate the influence of bending moment at a particular
point after constructing the bending moment influence line for a point of interest,
as depicted in Figure 11.7.

11.7 MAXIMUM INFLUENCE AT A POINT DUE

TO A UNIFORMLY DISTRIBUTED LIVE LOAD
Let us consider a simply supported beam with a moving UDL of magnitude w acting
on it. At any instant, let the starting and the end point of the UDL be at a distance of
a and b, respectively, from the left support. We wish to calculate the bending moment
at point C of the beam, as shown in Figure 11.25. Influence line diagram for bending
moment at point C is also shown at the bottom of the main beam for the analysis
purpose.
For a small length dx at a distance x from support A, the equivalent point load
acting on the beam is wdx. Hence, total bending moment generated at point C, due
to the above differential load is ( wdx ) y, where y is the ordinate of the influence line
under the differential load. Now, we can find the total bending moment at point C
due to the whole UDL as given by the following integral:
b b

∫a
∫
MC =   wydx = w   ydx
a

FIGURE 11.25  Maximum influence at a point due to moving uniformly distributed load
(UDL).
Rolling Loads and Influence Lines 205

FIGURE 11.26  Arrangement of UDL for maximum positive and negative bending moment
at a point D.

The integral in the right-hand side in the abovementioned expression represents the
area under the influence line diagram between point a and b (shaded portion). This
integral also tells us that for maximum positive bending moment, the UDL needs to
be placed in that portion of the beam where the ordinates of the influence line dia-
gram are all positive. To determine the maximum negative bending moment, if there
was any overhang portion of the beam, then the UDL needs to be placed in that por-
tion where the ordinates of influence line diagram are all negative. The arrangement
for maximum negative and positive bending moment at a point of interest D on the
beam is shown in Figure 11.26.

11.8  ABSOLUTE MAXIMUM SHEAR AND MOMENT

So far, we have discussed maximum and minimum bending moment and shear force
values with respect to a reference point on the beam. Now, we want to develop a con-
cept to determine the absolute maximum value of the same quantities for the entire
beam for a given loading and its position. The location and magnitude of the absolute
maximum shear force or bending moment for a beam are challenging to formulate.
But we can easily find them constructing influence lines for shear force or bending
moment at chosen points along the entire length of the beam and then computing the
206 Introduction to Structural Analysis

maximum shear or moment for each point in the beam. These maximum values of
shear and moment, when plotted together with respect to their positions along the
beam, yield an ‘envelope of maximums’. From this envelope of maximums, we can
easily find out the absolute maximum value of shear or moment and its location.
As usual, we can start this analysis by taking a moving unit load on the beam. We
already knew that once the ordinates for the influence line diagrams are obtained,
we can determine the values for any magnitude of loads just by multiplying the ordi-
nates at that point of influence line diagrams. From our earlier discussion, for single
unit point load on the beam, if we superimpose both influence line diagrams for sup-
port reactions, then we can easily determine from this the envelope of the maximum
value of shear and its location.
We know for any arbitrary section at a distance a from left support on the simply
supported beam subjected to a point load P, the maximum positive and negative
shear, respectively, are:

a
Va = + P    1 − 
 l
Pa
Va =   −
l

These equations show that maximum positive and maximum negative shear at any sec-
tion a varies linearly with the distance from left support A. Suppose, we plot the values
of maximum positive and negative shear according to the abovementioned two equa-
tions simultaneously on the same beam axis along the entire length of the beam and join
all of them on both sides of the beam axis. In that case, an envelope will be obtained as
shown in Figure 11.27. From this envelope, we can determine the location and value of
absolute maximum shear force. It is at the support points where maximum positive or
negative shear force values are P, which is clearly seen from the envelope itself.
The envelope for maximum bending moment can be easily evaluated by con-
sidering the maximum bending moment at any section a from left support point A
expressed as,

a
M a = Pa    1 − 
 l

FIGURE 11.27  Envelope of maximum shear for single point load P.

Rolling Loads and Influence Lines 207

FIGURE 11.28  Envelope of maximum bending moment for single point load P.

Clearly, the bending moment also varies quadratically with respect to distance from
the left support point. Hence, unlike shear force, the envelope for bending moment
for the simply supported beam with single concentrated load will be parabolic in
nature as shown in Figure 11.28. And the absolute maximum bending moment from
the envelope itself can be found to be at midspan with value Pl/4.
Same can be obtained for UDLs also. For UDL, the maximum positive and nega-
tive shear at any section a from left support is given, respectively, by:

w
 ( l − a )
2
Va = +
2l
w
Va = −   ( a )2
2l
The abovementioned values can be obtained by placing the UDL load at the beam
segment with maximum positive or negative ordinate values of unit load influence
line for shear force and calculating the area under between the reference points.
As from the abovementioned expression, it is clear that the shear force varies qua-
dratically with respect to the distance of the load from the left support. Hence, the
envelope for maximum positive and negative shear force will be parabolic in nature,
which is shown in Figure 11.29.

FIGURE 11.29  Envelope of maximum shear for uniformly distributed load.

208 Introduction to Structural Analysis

FIGURE 11.30  Envelope of maximum bending moment for uniformly distributed load.

Similarly, the maximum positive bending moment values for UDL from a dis-
tance a from the left support will be:

wa
Ma =  ( l − a )
a

So, the bending moment equations also vary quadratically with respect to distance
from the left support. Hence, bending moment envelope will be parabolic in nature
and absolute maximum bending moment from the envelope itself can be found to
be at midspan with value wl 2 /8. The maximum bending moment envelope can be as
seen in Figure 11.30.

11.8.1 Absolute Maximum Bending Moment for Series

of Concentrated Loads

For a series of concentrated loads, influence line envelopes can be drawn along the
length of the member to determine the maximum response values. But to be able to
that, we have to place the loads at various locations along the length of the beam,
which will ultimately attract a large number of computational efforts. By inspection
also the critical position of loads and the associated absolute maximum moment can-
not be determined even for a simple beam. For this reason, these analyses are done
in computer with the help of various software. However, for our analysis purpose,
we will provide the method of obtaining the maximum response for simple beams.
For a simply supported beam, a series of concentrated loads P, 1 P2 , and P3 are applied
as shown in Figure 11.31. Since the moment diagram for a series of concentrated
loads consists of straight-line segments having peaks at each load, the absolute maxi-
mum moment will also occur under one of these loads. Let us assume the maximum
moment occurs under the load P2, and P2 is x distance apart from the beam’s cen-
terline. By positioning the load P2 from a fixed distance x from the centerline of the
beam, we can fix the location of other loads in the series as well. Now, to determine
the specific value of x, we need to find the resultant force of this load series, PR, and
its distance x measured from P2. After obtaining this, if we now sum up moments
with respect to support B, we can easily find out the reaction force Ay as follows:

+ ∑M B =0
Rolling Loads and Influence Lines 209

FIGURE 11.31  Condition for the absolute maximum moment in a simply supported beam
for series of concentrated loads.

or,
L
− ( Ay × L ) + PR  − ( x − x )  = 0
2 

Therefore,
1 L
Ay = ( PR )  − ( x − x )
L 2 

We have assumed the absolute maximum bending moment will occur under the load
P2. So first, we need to find the internal moment, M 2 generated under this load and
after that need to find the condition for, M 2 to be maximum. If the beam is sectioned
just to the left of P2 as shown in Figure 11.32, we can write the moment equilibrium
equation as under:
L
M 2 = Ay  − x  − P1d1
2 

FIGURE 11.32  Section of the beam just to the left of load P2.
210 Introduction to Structural Analysis

Now, replacing the value of Ay in the abovementioned expression, we obtain,

L
  M 2 = Ay  − x  − P1d1
2 

or,
1 L L
M2 = ( PR )  − ( x − x ) ×  − x  − P1d1
L 2  2

therefore,
PR L PR x PR x 2 PR xx
M2 = − − + − P1d1
4 2 L L

For M 2 to be maximum from this expression, we require,

dM 2
=0
dx
i.e.,
−2 PR x PR x x
+ = 0; ∴ x =
L L 2

This is a very important relationship and from this relationship, we can say that, for a sim-
ply supported beam with series of concentrated loads, maximum bending moment occurs
under a load, when midspan of the beam equally divides the distance between the said
load and the resultant of all loads acting on the beam. Thus, after noting the resultant of
the applied series loads, we can place each load about mid-span of the beam and calculate
bending moment, satisfying the abovementioned condition, to calculate the maximum
absolute bending moment acting on the beam. This method is widely adopted to calculate
the maximum bending moment for bridge girders during analysis and design works.

Example 11.2: Find the absolute maximum bending moment in the

simply supported beam for the series of as shown in Figure 11.23.

SOLUTION: First let us determine the magnitude and position of the resultant

force of the system as next,
PR = (8 + 50 + 30 ) = 84 kN

Suppose the resultant force (84 kN) is x distance apart from the 8-kN load. Now
we need to find x from the following equilibrium equation:

84 x = 50 (1) + 30 ( 2.5)
or,
x = 1.49 m

Now we can locate the resultant force as shown in Figure 11.33.

Rolling Loads and Influence Lines 211

FIGURE 11.33  Location and magnitude of resultant force for the load series.

Let us assume the absolute maximum bending moment occurs under the
50-kN load. The load and the resultant force are positioned equidistant from the
beam’s centerline as shown in Figure 11.34.
Now in this load position, let us sum up all the moments about point B,

+ ∑M B =0

− Ay (10 ) + 84 ( 4.755) = 0

or,
Ay = 39.94 kN

FIGURE 11.34  Position of load series loads assuming the absolute maximum moment
occurs under 50-kN load.
212 Introduction to Structural Analysis

FIGURE 11.35  Section of the beam just to the left of load 50 kN.

If the beam is sectioned just to the left of 50 kN as shown in Figure 11.35, we can
write the moment equilibrium equation as under:

−39.94 ( 4.755) + 8 (1) + Ms = 0

or,

Ms = 181.91 kN − m

Now let us check another possibility before concluding 181.91 kN m as the abso-
lute maximum bending moment. For that let us assume the absolute maximum
bending moment occurs under the 30-kN load. The load and the resultant force
are positioned equidistant from the beam’s centerline as shown in Figure 11.36.
Now in this load position, let us sum up all the moments about point B,

+ ∑M B =0

− Ay (10 ) + 84 ( 5.505) = 0

or,

Ay = 46.24 kN

If the beam is sectioned just to the left of 30 kN as shown in Figure 11.37, we can
write the moment equilibrium equation as under:

−46.24 ( 5.505) + 8 ( 2.5) + 50 (1.5) + Ms = 0

or,

Ms = 159.55 kN − m

Now, we can conclude that the absolute maximum bending moment will occur
under the 50-kN load and its value is 181.91 kN m.
Rolling Loads and Influence Lines 213

FIGURE 11.36  Position of load series loads assuming the absolute maximum moment
occurs under 30-kN load.

FIGURE 11.37  Section of the beam just to the left of load 30 kN.
214 Introduction to Structural Analysis

11.9  INFLUENCE LINES FOR DEFLECTIONS

Influence line for deflections provides the information of deflection due to a moving
concentrated unit load on a beam structure. To obtain influence line for deflection,
Maxwell’s reciprocal law of deflection is found to be most useful. Consider a simply
supported beam on which we want to determine the deflection influence line for the
reference point C on the beam as shown in Figure 11.38.
Suppose when unit load is placed at a distance x from left support (point X), the
deflection at point C due to this case is δ CX as shown in Figure 11.38 (a). Also, when

FIGURE 11.38  Influence line diagram for deflection.

Rolling Loads and Influence Lines 215

we place the unit load at point C itself, the vertical deflection at point X is given by
δ XC as shown in Figure 11.38 (b). Now, following Maxwell’s theorem, we can say
that:

δ XC = δ CX

Since the point X is arbitrary in nature, this relationship is applicable for any point
on the beam. From the abovementioned relationship, we can conclude that influence
line for deflection for any point on a beam can be determined by drawing the elas-
tic curve/deflection line for the same beam when unit load is applied at that point
itself. We have already learned how to draw deflection lines for any determinate
beams using different principles in earlier chapters. Hence, by placing unit load at
the required point on the beam and doing the same calculation for drawing elastic
curve will help us to draw the influence line diagram of deflection for the beam.
 12 Cables, Arches, and
Suspension Bridges

12.1 INTRODUCTION
In major structures, cables and arches are found to be the principal load-carrying ele-
ments, and in this chapter, we will explore some salient features of the same and their
structural analysis. The chapter starts with a general discussion of cables, leading
to the analysis of cables subjected to concentrated and uniformly distributed loads
(UDL). Since the most common type of arches is statically indeterminate, only the
special case of three-hinged arch will be discussed in this chapter. The knowledge
of this type of structural analysis will provide some insight into the core behavior of
all arched structures.

12.2 CABLES
Cables are used in structures to support and transfer loads from one member to
another. When cables are used to support suspension roofs, bridges, etc., they form
the main load-bearing element in the structure. During force analysis of such struc-
tures, the weight of the cable is mostly neglected; however, when cables are used
as guy wires for radio antennas, electrical transmission lines, etc., the cable weight
may become an important factor, and it has to be considered in the structural analy-
sis. Separate cases will be discussed in the sections as follows: a cable subjected to
concentrated loads and another one to a UDL. It is worthwhile to mention that these
loadings are coplanar with the cable, and corresponding requirements for equilib-
rium are formulated accordingly.
While deriving the equations and relations between the force in the cable and its
slope, it will be assumed that the cable is perfectly flexible and inextensible. Due to
flexibility, the cable does not induce shear force and bending moment, and thus, the
force acting in the cable will always be tangential to the cable at the same points
along its length. As it is inextensible, the cable has invariant length both before and
after the external load is applied. As a result, once the external load is applied, the
geometry of the cable remains fixed, and the cable or a segment of it can be treated
as a rigid body.

12.3  CABLES SUBJECTED TO CONCENTRATED LOADS

When a cable (neglecting self-weight) supports several concentrated loads, the cable
takes the form of different straight-line segments, each of which is acted upon by a
constant tensile force. Let us consider, for example, the cable shown in Figure 12.1.
Here, θ indicates the angle of the cable’s chord AB, and L is its span. If the distances
DOI: 10.1201/9781003081227-14 217
218 Introduction to Structural Analysis

FIGURE 12.1  Cable with concentrated loads.

L1 , L2 , L3 and the loads P1 , P2 are known, our goal is to calculate the nine unknowns
consisting of the tension in three segments of the cable, the four components of reac-
tion at support points A and B, and the vertical displacements yC , yD at the two points
C and D. For the calculation, we can write two equations of force equilibrium at
each of points A, B, C, and D. This results in a total of eight equations. To complete
the analysis and calculations, it will be necessary to know about the geometry of the
cable to obtain the necessary ninth equation.
For the sake of simplicity, if the cable’s total length L is specified, the Pythagorean
theorem can be used to relate each of the three segmental lengths, in terms of
L1 , L2 , L3 , yC , yD , and θ to the total length, L. Practically, this type of problem cannot
be solved by hand with ease. On the other hand, we can analyze the same problem by
specifying either of the vertical deflections yC , yD , instead of the total length of the
cable. Whatever way we may proceed, we can form the equilibrium equations and
complete the calculations to determine the unknown tension forces acting at differ-
ent segments of the loaded cable and the support reactions.
While carrying out equilibrium analysis for a problem like this one, the
unknown forces in the cable can also be determined by developing the equilibrium
equations for the entire cable or any portion. Example 12.1 provides these neces-
sary concepts.

Example 12.1:  Determine the tension in each segment of the following

cable subjected to the point loads as shown in Figure 12.2.

SOLUTION:  By the nature of this loading pattern, we can declare that there are
four unknown support reactions (HA ,VA , HB ,VB ) and three unknown tension forces
(TDB ,TCD ,TCA ) acting in the cable and a sag h. So, there is total eight unknowns.
These eight unknowns can be determined from eight available equilibrium equa-
( )
tions ∑ Fx = 0, ∑ Fy = 0 applied through A to D. Since the geometry is known in
detail from the given diagram, we can first determine the unknown tensile force in
the cable DB (Figure 12.3).
Cables, Arches, and Suspension Bridges 219

FIGURE 12.2  Example problem on cable under concentrated loads.

Taking moment about A,

 2.5   2.0 
TDB  1+ TDB  8 − 8 × 4 − 2 × 5.5 = 0
 3.2   3.2 

TDB = 7.44 kN

Now, let us consider the equilibrium of point D as shown in Figure 12.4.

→+ ∑F = 0
x

 2.5 
7.44 ×  − T cos θ CD = 0
 3.2  CD

FIGURE 12.3  Free-body diagram to determine the tension force TDB .

220 Introduction to Structural Analysis

FIGURE 12.4  Equilibrium of point D.

↑+ ∑F = 0
y

 2.0 
7.44 ×  − T sin θ CD − 2 = 0
 3.2  CD

Solving the abovementioned two equilibrium equations, we get θ CD = 24.51°,

TCD = 6.37 kN.
Now, let us consider the equilibrium of point C as shown in Figure 12.5.

→+ ∑F = 0
x

6.37 × cos 24.51° − TAC cos θ AC = 0

↑+ ∑F = 0
y

6.37 × sin 24.51° + TAC sin θ CA − 8 = 0

FIGURE 12.5  Equilibrium of point C.

Cables, Arches, and Suspension Bridges 221

Solving the abovementioned two equilibrium equations, we get, θ CA = 42.75°,

TAC = 7.89 kN.
From the geometry we get, h = 4 ( tan 42.75°) = 3.7 m

12.4  CABLE SUBJECTED TO A UNIFORMLY DISTRIBUTED LOAD

Cables provide a very efficient means of withstanding and transferring the dead
weight of girders or bridge decks having very long spans. A suspension bridge is
an important example in which the bridge deck is supported by the cable using a
series of equally spaced hangers. In Figure 12.6 (a), we have shown a cable under

FIGURE 12.6  (a) Cable with uniformly distributed load and (b) free-body diagram of the
small element ∆ x .
222 Introduction to Structural Analysis

the application of a UDL of intensity w. The cable will take a curvilinear shape as
shown in the figure. Exact nature of the curve will depend upon the load intensity
and support conditions, and based on the same, exact equation for the deflected shape
of the cable can be drawn. To analyze the problem, we will take an arbitrary small
element of loaded cable of length ∆s at a distance x from the left support as shown
in Figure 12.6. The projection of this small element ∆s, on the x-axis is suppose ∆x.
Free-body diagram of this small element is also shown in Figure 12.6 (b) for the
ease of understanding. Here the origin of the x , y coordinates has been chosen at
the lowest point of the cable as shown. The distributed force is represented as an
equivalent concentrated force of magnitude w ( ∆x ), and acting at a distance of ∆x/2
from point O.
Applying force equilibrium equations in horizontal direction, we get,

→+ ∑ F = 0 − T cos θ + (T + ∆T ) cos(θ + ∆θ ) = 0
x

Also, applying force equilibrium equation for vertical direction, we get,

↑+ ∑ F = 0 − T sin θ − w∆x + (T + ∆T )sin (θ + ∆θ ) = 0

y

Also, taking moment about point O of the free-body diagram we get,

+ ∑M o =0

∆x
( w∆x ) − T  cos θ   ∆y + T  sin θ   ∆x = 0
2

Dividing each of the abovementioned equations by ∆x and taking the limit ∆x → 0,

and hence, by ∆θ → 0 and ∆T → 0 , we obtain,

d ( T cos θ )
=0
dx

d ( T sin θ )
=w
dx

dy
= tan θ
dx

Integrating the first equation, where T = FH at x = 0 , we have,

T  cos θ = FH

which indicates horizontal component of tension force at any point along the cable
subjected to UDL remains constant.
Cables, Arches, and Suspension Bridges 223

Integrating the second equation, with the initial condition T sin θ = 0 at x = 0, we get,

T sin θ = wx

Dividing these last two equations, we get,

dy wx
tan θ = =
dx FH
Performing the integration with y = 0 at x = 0 yields,
w 2
y= x
2 FH
The preceding equation is the equation of parabola. The constant force FH can be
obtained by putting the boundary condition,

y = h, at  x = L

Putting this value in the parabolic equations, we get,

wL2
FH =
2h
Replacing the value of FH in the master equation, we finally get the equation of para-
bolic shape of the cable as follows:
h 2
y= x
L2

Also, from the equation, T  cos θ = FH , and T  sin θ = wx; maximum tension in the
cable occurs when θ is maximum, i.e., at x = L. So, from these equations, we get,

Tmax =   FH 2 + ( wL )2

or,
2
 wL2 
Tmax =    + ( wL )2
 2h 

i.e.,
2
L
Tmax = wL   + 1
 2h 

As stated earlier, we have neglected the self-weight of the cable during derivation of the
abovementioned relationship. In fact, when a cable is suspended and only acting load
on it is the self-weight, the cable assumes the shape of catenary. From the outcome of
the abovementioned analysis, it indicates that a cable will assume a parabolic shape,
224 Introduction to Structural Analysis

provided the dead load of the deck for a bridge will be uniformly distributed on the
horizontal projection length of the cable. Hence, if the bridge girder is supported by a
series of hangers, which are uniformly spaced and not far from each other, the load in
each hanger must be the same to ensure that the cable has a parabolic shape.
Taking help of this assumption, we can complete the structural analysis of the
girder or any other framework that is freely suspended by the cable. We will assume
that the girder is simply supported, and as a result, this will be a statically indetermi-
nate problem of degree one. Example 12.2 will elaborate on this concept for cable-
stayed bridge girder analysis.

Example 12.2: Determine the tension at points A, B, and C of the

following cable-stayed bridge girder as per Figure 12.7. Self-weight of
the girder is 70 kN/m.

SOLUTION:  The origin has been selected at point B, which is the lowest point
of the cable. Now, from our theoretical analysis, we can insert these geometrical
values to calculate the exact equation.

w 2
y= x
2FH

or,

70 2 35 2
y= x = x
2FH FH

FIGURE 12.7  Example problem of cable-supported bridge girder.

Cables, Arches, and Suspension Bridges 225

Assuming point C at a distance of x′ from point B,

35 2
10 = x′
FH

or,

FH = 3.5x ′ 2

Also, for point A,

35
 − ( 40 − x ′ ) 
2
15 =
FH 

or,

35
2  (
 − 40 − x ′ ) 
2
15 =
3.5x ′

Solving the quadratic equation, we get,

x′ = 17.98 m

So, once we get the value for x′ , we can substitute it to get the value of FH .
So,

FH = 3.5 × 17.982 = 1131.48 kN

Now, at point A, x = − ( 40 − 17.98) = −22.02 m

dy 35
tan θ A = = × 2 × ( −22.02) = −1.36
dx x =−22.02 1131.48

Hence, θ A = −53.67°
So,

FH 1131.48
TA = = = 1909.88 kN
cos θ A cos ( −53.67° )

Now, at point B, x = 0 m

dy 35
tan θ B = = × 2 × (0) = 0
dx x =0 1131.48

Hence, θ B = 0°
So,

FH 1131.48
TB = = = 1131.48 kN
cos θ B cos ( 0°)
226 Introduction to Structural Analysis

Now, at point C, x = 17.98 m

dy 35
tan θ C = = × 2 × (17.98) = 1.11
dx x =17.98 1131.48

Hence, θ C = 47.98°
So,
FH 1131.48
TC = = = 1690.32 kN
cos θ C cos ( 47.98° )

12.5 ARCHES
Arches can be constructed to control and reduce the bending moments in long-span
bridges, airplane hangars, etc. because one of the main distinguishing features of an
arch is the development of vertical reactions as well as horizontal thrusts at the sup-
ports, even in the absence of a horizontal load. Operationally, an arch act just like
an inverted cable, so it transmits its load mainly in compression. However, due to
its rigidity, it also has to resist bending and shear depending upon external loading
patterns and its geometric shape. Specifically, if the arch is parabolic in nature and
loaded by a uniform horizontally distributed vertical load, from our earlier knowledge
on the analysis of cables it follows that only compressive forces will be resisted by the
arch. Under such circumstances, the arch shape is called a funicular arch because no
bending or shear forces will be induced within the arch due to external loading.
A typical arch is shown in Figure 12.8, which specifies some of the names of vari-
ous parts of an arch used to define geometry.
Depending upon the requirements, different types of arches, as shown in
Figure 12.9, can be modeled to support different types of external loading. A fixed
arch (Figure 12.9 (a)) is often made from reinforced concrete. Although it may require
less material to construct than other types of arches, it must have a solid foundation

FIGURE 12.8  Different components of an arch.

Cables, Arches, and Suspension Bridges 227

FIGURE 12.9  Different types of arches: (a) fixed arch, (b) two-hinged arch, (c) three-hinged
arch, and (d) tied arch.

since it is indeterminate to the third degree, and additional stresses can be intro-
duced into the arch due to relative settlement of its supports. A two-hinged arch
(Figure 12.9 (b)) is generally made from metal or timber. It is indeterminate to the
first degree. Although it is not as rigid as a fixed arch, it is to some extent insensitive
to settlement. We could make this structure statically determinate by replacing one of
the hinges with a roller. But this will affect the capacity of the structure to resist bend-
ing along its span, and, as a result, it would serve as a curved beam and not as an arch.
A three-hinged arch (Figure 12.9 (c)), which is also made from metal or timber, is
statically determinate. It is not affected by settlement or temperature changes. Finally,
we can attach a tie rod (Figure 12.9 (d)) at the supports, so that the arch will behave
like a rigid body and thus we can avoid the need for larger foundation abutments. This
will also remain unaffected even under the relative settlement of supports.

12.6  THREE-HINGED ARCHES

We will now consider the analysis of a three-hinged arch such as the one shown in
Figure 12.10 (a) and (c), to provide some insight into how arches transmit loads. In
this case, the third hinge is located at the crown, C and the supports are located at
different heights as shown in the diagram. To determine the reactions at the sup-
ports, the arch is disassembled to construct the free-body diagram of each member
as shown in Figure 12.10 (b). It is clear from the free-body diagram that there are
six unknowns for which six equations of equilibrium are available. One method of
solving this problem is to apply the moment equilibrium equations about points A
and B. The simultaneous solution will yield the reactions at the internal hinge at C.
From the force equations of equilibrium and then the support reactions at A and B
are determined. Once obtained, the internal normal force, shear, and moment at any
228 Introduction to Structural Analysis

FIGURE 12.10  (a), (c) Three-hinged arch (source of ‘c’: https://structurae.net/en/

media/315533-via-guglielmo-marconi-cycle-bridge), and (b) and (d) free-body diagrams for
internal loads and external support reactions.

point along the arch can be found using the method of sections. Here, of course, the
section should be taken perpendicular to the axis of the arch at the point considered.
Here, for example, the free-body diagram of segment AD is shown in Figure 12.10 (d).
By applying moment equilibrium equations and remembering the fact that at
hinge locations, net moment will be zero, we can calculate the unknown forces
acting in the arch. Once the support reactions are obtained, we can determine the
bending moment and shear force at any typical section taken perpendicular to the
arch at that point. These concepts will be easier to understand once we go through
Examples 12.3 and 12.4 (Figures 12.11 and 12.15).

Example 12.3: Show that any point (like point D) of the uniformly

loaded parabolic arch as shown in Figure 12.11 will always be under
compression. The arch is hinged at locations A, B, and C.

SOLUTION: Since the UDL is acting on the horizontal bridge deck above the
arch, net point load acting on the arch at the crown is,

P = 3.5 × ( 20 + 10 + 10 ) = 140 kN

The free-body diagram will be like the one in Figure 12.12.

Taking moment about point A, we will get,

140 × 20 − Cy × 40 = 0
Cables, Arches, and Suspension Bridges 229

FIGURE 12.11  Three-hinged arch example problem-1.

or,
Cy = 70 kN ↑

Since at point B, there is a hinge support, the algebraic sum of moment at B will
be zero. The free-body diagram of the right half portion is shown in Figure 12.13.
Arch segment BC,
Taking moment about point B, we get,

+ ∑M B =0

or,
70 × 20 − C x × 15 − 70 × 10 = 0

C x = 46.67 kN ←

→+ ∑F = 0
x

FIGURE 12.12  Free-body diagram of the entire arch for support reactions.
230 Introduction to Structural Analysis

FIGURE 12.13  Free-body diagram of the right half portion of arch.

or,
Bx = 46.67 kN →

↑+ ∑F = 0
y

By + Cy = 70

or,
By = 0

Taking a section along the point D, from cable equation of parabolic shape, we get,

h 2
y= x
L2
where h = 15 m, L = 20 m. Hence, the elevation of point D will be,

15
y=− (10 2 ) = −3.75 m
20 2
And the slope of the tangent at point D is given by,

dy −15 × 2
tan θ D = = × 10 = −0.75
dx x =10 m 20 2

θ D = −36.87°

The free-body diagram at point D for calculating the forces is shown in Figure 12.14.
Thus, force and moment equilibrium equation will produce,

→+ ∑ F = 0 ⇒ 46.67 − N cos 36.87° −  V  sin 36.87° = 0

x D  D

↑ +∑ F = 0 ⇒ −35 + N sin36.87° − V cos36.87° = 0

y D  D

 +∑ M = 0 ⇒ M + 35 ( 5) − 46.67 ( 3.75) = 0

D D
Cables, Arches, and Suspension Bridges 231

FIGURE 12.14  Free-body diagram of section D.

Solving the abovementioned equations, we get,

ND = 58.41 kN

VD = 0

MD = 0

Please note that when arch is having a different shape and with unsymmetrical
loading, the bending moment and shear force values will be nonzero.

Example 12.4: A three-hinged parabolic arch is loaded as shown in

Figure 12.15. Calculate the support reactions and internal stresses for
the sections shown. The arch is hinged at locations A, B, and C.

SOLUTION:  The ordinate (y) at any point along a parabolic arch is given by:

y=
(
4yc Lx − x 2 )
2
L

FIGURE 12.15  Three-hinged Arch example problem 2.

232 Introduction to Structural Analysis

where yc is the height of the crown of the arch from the base = 15 m, L is the length
of the arch = 50 m, x is the horizontal coordinate of interest.
The coordinate is chosen at point A as shown in figure.
Hence,

y=
(
4 × 15 50 x − x 2 ) =  6 x −  3  2
   x
50 2 5 125 

Differentiating the abovementioned equation with respect to x.

dy  6  6 
= tan θ =   −  x
dx  5   125 

The values of y-ordinates, and tan θ for corresponding different x-coordinates of

interest are calculated as from the abovementioned equations and shown in
Table 12.1.

Support reactions

+ ∑M B =0

(− A y )
× 50 + (16 × 25 × 12.5) + ( 20 × 42) − (10 × 8.064 ) = 0

or,
Ay = 115.18 kN

+ ∑M A =0

(B y )
× 50 − (16 × 25 × 37.5) − ( 20 × 8) − (10 × 8.064 ) = 0

or,
By = 304.81 kN

Now, from span BC, we will find out the horizontal reaction Bx , considering the
moment equilibrium at crown C. In this portion about point C, the anticlockwise

TABLE 12.1
Values of Different y-Coordinates and Slopes for
Corresponding x-Coordinates of the Arch
Point x ( m) y ( m) tan θ
A  0 0 1.2
1  8 8.064 0.816
2 16 13.056 0.432
C 25 15 0
Cables, Arches, and Suspension Bridges 233

moment will sag the arch and clockwise moment will hog the arch. So, the anti-
clockwise moment is considered positive here.

+ ∑M R
C =0

(304.81× 25) − (Bx × 15) − (16 × 25 × 12.5) = 0

or, Bx = 174.68 kN ←

Now, from span AC, we will find out the horizontal reaction Ax , considering the
moment equilibrium at crown C. In this portion about point C, the anticlockwise
moment will hog the arch and clockwise moment will sag the arch. So, the clock-
wise moment is considered positive here.

+ ∑M L
C =0

(115.18 × 25) − ( Ax × 15) − ( 20 × 17) − 10 (15 − 8.064 ) = 0

or, Ax = 164.68 kN →

Internal stresses
Now we will determine the internal stresses (normal force, shear, and bending
moment) of the arch at the section of interests. For this, let us consider one free-
body diagram up to some arbitrary section ( xi , y i ) as shown in Figure 12.16.
Let us consider the moment equilibrium at that section,

+ ∑M = 0
or, M = Ay xi − Ax y i − P1x1

From the free-body diagram, we can obtain,

Ax + N  cos θ + Q  sin θ = 0

Ay − P1 + N  sin θ − Q  cos θ = 0

FIGURE 12.16  Free-body diagram of the arch up to some arbitrary section.

234 Introduction to Structural Analysis

Solving these two equations for different sections, we can get the values of N,
and Q. Let us find these internal forces for different sections of arch as follows:

Calculation for the left side of the hinge

For point A,

tan θ = 1.2, θ = 50.19°

M A = 0 (hinged support)

164.68 + N  cos 50.19° + Q  sin 50.19° = 0

115.18 + N  sin 50.19° − Q  cos 50.19° = 0

Solving this,

N = −193.91 kN, and  Q = −52.76 kN

For point 1,

tan θ = 0.816, θ = 39.21°

M1 = 115.18 × 8 − 164.68 × 8.064 = −406.54 kN − m

164.68 + 10 + N  cos 31.21° + Q  sin 31.21° = 0

115.18 − 20 + N sin31.21° − Q  cos 31.21° = 0

Solving this,

N = −198.71 kN, and Q = −9.11 kN

For point 2,

tan θ = 0.432, θ = 23.36°

M2 = 115.18 × 16 − 164.68 × 13.056 − 10 × (13.056 − 8.064 ) − 20 × 8

= −517.1 kN − m

164.68 + 10 + N  cos 23.36° + Q  sin 23.36° = 0

115.18 − 20 + N  sin 23.36° − Q  cos 23.36° = 0

Cables, Arches, and Suspension Bridges 235

Solving this,

N = −198.11 kN, and  Q = 18.12 kN

For point C,

tan θ = 0, θ = 0°

MC = 115.18 × 25 − 164.68 × 15 − 10 × (15 − 8.064 ) − 20 × 17 = 0 kN − m

164.68 + 10 + N   = 0

115.18 − 20 − Q = 0

Solving this,

N = −174.68 kN, and Q = 95.18 kN

12.7  THREE-HINGED STIFFENING GIRDERS

The curvature of the cable of an unstiffened bridge always changes as the live load
moves on the deck, because the cable of the suspension bridge is the main load-
bearing element. To avoid this, the bridge deck is stiffened by either providing two-
hinged or three-hinged stiffening girders. The stiffening girders are assumed to
transfer a uniform or equal load to each suspender, irrespective of the load posi-
tions on the bridge deck. When bridges are stiffened with three-hinged stiffening
girders, it is assumed that the cable retains its parabolic shape when subjected to
loads from moving traffic or from other sources and hence the load on the cable
will remain uniform throughout the entire length. When the bridge is subjected to
UDL, the total load is directly taken care of by the cable itself and does not affect
the stiffening girders. On the other hand, in the case of moving loads, the cable will
be assumed to carry uniform load, and, as a result, the stiffening girders will be
subjected to bending moments and shear forces as that happens for normal trans-
versely loaded beams.
Now, let us consider a suspension bridge with three-hinged stiffening girders as
shown in Figure 12.17. Let two external loads W1, and W2 act on the deck. Let we per
unit run UDL transferred to the cables through the suspenders.
Now, let us separately consider the equilibrium of cable and the stiffening girder
as shown in Figure 12.18.
For the cable part, we can easily prove that the vertical reactions, V = wel/2, and
the horizontal thrusts, H = wel2/8d, where d is the dip of the cable at center. The
maximum tension in the cable can be written as, Tmax = V 2 + H 2 .
236 Introduction to Structural Analysis

FIGURE 12.17  Three-hinged stiffening girder.

Now consider the girder which is subjected to two load systems (Figure 12.19),
such as part (a) the applied external load (Figure 12.19 (b)) system, W1, and part (b)
upward UDL of we per unit run from the suspenders (Figure 12.19 (c)).
Let us consider a section at a distance x form support A of the girder. The bend-
ing moment, M x at this section due to this two loading systems on the girder can be
expressed as,
wl x
M x = VA . x − W1 ( x − a )  −  e . x − we . x . 
 2 2 

or,
we. x
M x = M beam − [l − x ]
2

FIGURE 12.18  Equilibrium of (a) cable and (b) stiffening girder shown separately.
Cables, Arches, and Suspension Bridges 237

FIGURE 12.19  (a) Girder subjected to two load system, (b) the applied external load sys-
tem, W1, and (c) upward UDL of we per unit run from the suspenders.

Rewriting the abovementioned expression as,

we.l 2 4 dx
M x = M beam − ⋅ 2 [l − x ]
8d l
or,
M x = M beam − H . y

where M beam is the bending moment at the section due to girder load considering the
span as that of a simply supported girder.
Now, let us find the moment at hinge C using the abovementioned expression,
0 = MC − Hd   (as dip of the cable at center is d)
Therefore, horizontal thrust, H = MC /d
Similarly, the shear force at this section can be expressed as,
wl w
S x = [VA − W1 ] −  e − we x  = [VA − W1 ] − e ( l − x )
 2  2
238 Introduction to Structural Analysis

Rewriting the abovementioned expression as,

 w l 2 4d (l − 2 x ) 
S x = [VA − W1 ] −  e. × 
 8d l2 
or,

 w l 2 d  4 dx 
S x = [VA − W1 ] −  e. ×  2 [l − x ] 
 8 d dx l 
or,

 w l 2 dy 
S x = [VA − W1 ] −  e. × 
 8d dx 
or,

S x = [VA − W1 ] − [ H × tan θ ]

or,

S x = Sbeam − [ H × tan θ ]

where Sbeam is the shear force at the section due to girder load considering the span
as that of a simply supported girder.
The shear force and bending moment diagram as per the abovementioned expres-
sions of three-hinged girders are shown in Figure 12.20.

FIGURE 12.20  Three-hinged stiffening girder (a) bending moment and (b) shear force
diagram.
Cables, Arches, and Suspension Bridges 239

The bending moment diagram for the girder can be drawn by superposing the same
for a simply supported beam over the bending moment diagram due to Hy. The value
of H for a particular loading type is constant, and hence, the product Hy will be a
parabola. The diagram is obtained by taking the parabolic shape of the cables and mul-
tiplying its ordinates by H as shown in the bending moment diagram in Figure 12.20 (a).
Similarly, the shear force diagram is obtained by superposing the same of a simply
supported beam over the H tan θ diagram as shown in Figure 12.20 (b). The tan θ
varies linearly with x, and hence, the diagram will be straight line. As H is constant,
H tan θ will also vary linearly from H tan θ A at A to H tan θ B at B.

Example 12.5:  A suspension bridge of 110-m span has a three-hinged

stiffening girder supported by two cables having a central dip of 10 m
as shown in Figure 12.21. The roadway has a width of 5 m. The dead
load on the bridge is 10 kN/m2 while the live load is 20 kN/m2 that acts
right half of the span. Determine the shear force and bending moment
in the girder at 30 m from the right end. Find the maximum tension in
the cable for the current position of the live load.

FIGURE 12.21  Example problem on three-hinged stiffening girder.

240 Introduction to Structural Analysis

FIGURE 12.22  Finding reaction forces due to live load on the girder.

SOLUTION:  Dead load = 10 kN/m2 × 5 m = 50 kN/m, live load = 20 kN/m2 × 5 m =

100 kN/m. Dead load per girder = 50 kN/m/2 = 25 kN/m, live load per girder =
100 kN/m/2 = 50 kN/m,
Only considering the live load on each girder as shown in Figure 12.22, let us
find out the reaction forces.
Taking moment about B,

55
RA × 110 − 50 × 55 ×
2
or,

RA = 687.5 kN; RB = ( 50 × 55) − 687.5 = 2062.5 kN

Horizontal thrust,

MC 687.5 × 55
H= = = 3781.25 kN
d 10

Actual bending moment at 30 m from the right side due to the live
 50 × 30 2 
load = Mbeam − H.y =  2062.5 × 30 −  − 3781.25y
where,  2

4dx 4 × 10 × 30
y= (l − x) = × (110 − 30 ) = 7.93 m
l2 110 2

Now, putting the value of y in the abovementioned expression, we get the actual
bending moment at 30 m from support B, as 9389.69 kN m.
Likewise, we can calculate the actual shear force at 30 m from support B due
to the live load as,

S30 = Sbeam − H × tan θ  = ( 2062.5 − 50 × 30 ) − 3781.25 × tan θ

Cables, Arches, and Suspension Bridges 241

where,

dy 4d 4 × 10 × (110 − 2 × 30 )
tan θ = = 2 ( l − 2x ) = = 0.1653
dx l 110 2

Now, putting the value of tan θ in the abovementioned expression, we get the
actual shear at 30 m from support B, as −62.5 kN m.
Now to get the maximum tensile force in the cable, we need to consider both
the effects of dead loads and live loads. For this, we need to find the equivalent
UDL, w e acting in the suspenders due to dead load and live load.
wl 2 w × 110 2
We know horizontal thrust in cables under UDL, H = = = 151.25w.
8d 8 × 10
Again, we have obtained the horizontal thrust due to the live load as 3781.25 kN.
If we equate these two, we will get the value of w, which is basically the UDL due
to the live loads transferred from girder to cable through suspender.
Therefore,

w = 25 kN/m

so, the total UDL, w e , that will be transferred from girder to cable through sus-
pender is = UDL from dead load + UDL from live load = (25 + 25) = 50 kN/m.
Now from Figure 12.21 (a),

w e l 50 × 110
VA = VB = V = = = 2750 kN
2 2

w e l 2 50 × 110 2
HA = HB = H = = = 7562.5 kN
8d 8 × 10

Therefore, the maximum tension in the cable will be

Tmax = (2750 2 + 7562.52 ) = 8046.98 kN.
 13 Analysis of Symmetric
Structures

13.1 INTRODUCTION
In this chapter, we will introduce the concept of symmetric structures. Symmetry
may be related to loading conditions or geometry or both. Having learned this
method of analysis, one may acquire quick detection of structural methods to be
adopted for part of the structures and applying the results for the entire one to com-
plete the analysis. However, it is advisable to check the results of symmetric analysis
with complete analysis of the same structure ignoring the symmetry in case of any
confusion. In this way, one can check if there is any mismatch between the outcomes
of both the processes and detect any calculation or conceptual error thereof. In short,
structural geometry can be determined based on (a) symmetric supports, (b) sym-
metric loading, (c) symmetric geometry. If any one of these parameters is nonsym-
metric, the structure as a whole becomes antisymmetric.

13.2 SYMMETRIC AND ANTISYMMETRIC

COMPONENTS OF LOADINGS
In the case of structural analysis of any highly indeterminate structure, even for a
statically determinate structure, it can be simplified, provided the designer or analyst
can recognize those structures that are symmetric and support either symmetric or
antisymmetric loadings. In a general sense, a structure can be termed as being sym-
metric, provided that half of it develops the same internal loadings and deflections
as its mirror image reflected about its central axis. Generally, symmetry requires
the material properties, geometry, supports, and loading to be the identical on each
side of the structure. However, this does not always have to be the case as shown in
Figure 13.1. It is to be noted that for horizontal stability, a pin is required to sup-
port the truss horizontally. This truss is violating the symmetry condition due to its
supports, but as only vertical load is applied on this truss, the horizontal reaction
at the pin is zero, and so, this structure will deflect and produce the same internal
loading as its reflected counterpart. As a result, they can be classified as being sym-
metric. It is to be understood that this would not be the case for the frame shown
in Figure 13.2. If the fixed support at A is replaced by a pin or roller, the deflected
shape and internal forces would not be the same on its left and right sides, even if the
vertical loading is only applied.
As can be seen from Figure 13.2, the structure is perfectly symmetrical in this
situation. But as explained earlier, when we replace one fixed support at location A,
this change in support condition will lead to asymmetry in the structural analysis.

DOI: 10.1201/9781003081227-15 243

244 Introduction to Structural Analysis

FIGURE 13.1  Symmetrical structure – truss.

FIGURE 13.2  Frame structure with loading.

13.3 SYMMETRIC AND ANTISYMMETRIC

COMPONENTS OF LOADINGS
Sometimes, a symmetric structure supports an antisymmetric loading, (i.e., the load-
ing is considered to be antisymmetric with respect to an axis in its plane if the nega-
tive of the reflection of the loading about the axis is identical to the loading itself) such
as shown in Figures 13.3 and 13.4. Provided the structure is symmetric in geometry
and its loading is either symmetric or antisymmetric, the structural analysis will only
have to be carried out on half of the structure since the same (symmetric) or opposite
(antisymmetric) results will be generated on the other half of it. If a structure is geo-
metrically symmetric and its applied loading is unsymmetrical, then it is possible to
transform this loading into symmetric and antisymmetric components. To do this,
the loading is first divided in half, then it is reflected to the other side of the structure
Analysis of Symmetric Structures 245

FIGURE 13.3  Examples of antisymmetric loading.

and both symmetric and antisymmetric components are produced. For example, the
loading on the beam in Figure 13.5 is divided by two and reflected about the beam’s
axis of symmetry. From this, the symmetric and antisymmetric components of the
load are produced as shown in the same figure. When combined all together, these
components produce the original loading. A separate structural analysis can now
be performed using the symmetric and antisymmetric loading components and the
results superimposed to obtain the actual behavior of the structure.

FIGURE 13.4  Continuous beam with antisymmetric loading.

246 Introduction to Structural Analysis

FIGURE 13.5  Superposition of loading on continuous beam.

13.4 BEHAVIOR OF SYMMETRIC STRUCTURES UNDER

SYMMETRIC AND ANTISYMMETRIC LOADINGS
When a symmetric structure is subjected to a loading with respect to the structure’s
axis of symmetry, the response of the structure is also symmetric.
Displacement behavior along the axis of symmetry for symmetric loading results
neither in rotation (unless there is a hinge at such a point) nor any deflection perpen-
dicular to the axis of symmetry.
Force behavior along the axis of symmetry for symmetric loading results in zero
force along the axis of symmetry.
For a symmetric structure, which is subjected to a loading that is antisymmetric
with respect to the structure’s axis of symmetry, the response of the structure is also
antisymmetric. Displacement behavior along the axis of symmetry for antisymmetric
loading conditions results in no displacement along the axis of symmetry. Force behav-
ior along the axis of symmetry for antisymmetric loading results in zero force normal to
the axis of symmetry and zero bending moment. In Figure 13.6, the deflection pattern
of a symmetric frame is shown for both symmetric (Figure 13.6 (a)) and antisymmetric
(Figure 13.6 (c)) loading condition. It can be also noted that, considering the symmetric
and antisymmetric boundary conditions on the half frame along the location of axis of
axis of symmetry, we can easily determine the nature of response of the whole structure.
Analysis of Symmetric Structures 247

FIGURE 13.6  S
 ymmetric frame subjected to symmetric and antisymmetric loading and
respective boundary conditions for the half frame.

For general loading on a symmetric structure, the loading can be decomposed

into symmetric and antisymmetric components. Displacement and force boundary
conditions for symmetric and antisymmetric loadings along the axis of structural
symmetry apply. To obtain the total response, use superposition of the symmetric
and antisymmetric result.

Example 13.1:  Find the substructures for the analysis of symmetric and
antisymmetric responses for the statically indeterminate beam shown
in Figure 13.7.

SOLUTION:  The procedure for determining the substructures for analyzing sym-
metric and antisymmetric responses of the given indeterminate beam for this load-
ing condition is shown in Figure 13.8.
248 Introduction to Structural Analysis

FIGURE 13.7  Indeterminate beam for the example problem 13.1.

FIGURE 13.8  Finding substructures to analyze symmetric and antisymmetric response.

From this figure it is clear that if we add up the responses of symmetric loading
component as shown in Figure 13.8 (b) with the antisymmetric loading component
as shown in Figure 13.8 (c), we will get the original response of the given structure.
Putting the appropriate boundary conditions along the axis of symmetry on the
half structure, we will be able to find the response of the whole beam as well.
Part III
Analysis of Statically
Indeterminate Structures
 14 Introduction to Statically
Indeterminate Structures

14.1 INTRODUCTION
In Part III (Chapters 14–24) of this text, our principal attention will be on the analy-
sis of statically indeterminate structures. As discussed earlier, the support reactions
and internal forces (member force, shear force, bending moment) of statically deter-
minate structures can be determined from the equations of equilibrium (including
equations of conditions, if required). However, since indeterminate structures have
more supports and/or members (called redundant members) than required for static
stability and equilibrium, the equilibrium equations alone are insufficient to deter-
mine the unknown reactions and internal forces of such structures. There must be
some other relationships based on the geometry of deformation of structures or the
nature of constraints imposed on the structure. These additional relationships or
equations, which are termed as the compatibility conditions or equations, ensure that
the continuity of the displacements is maintained throughout the structure and that
the structure’s various parts remain connected together without any damage to the
stability of the structure. For example, at a rigid joint, all the members’ deflections
and rotations related to the joint must be the same.
Thus, analysis of an indeterminate structure requires, in addition to the dimensions
and geometric arrangement of members/elements of the structure, its cross-sectional
and material properties (such as cross-sectional areas, moments of inertia – both
geometric and mass moment of inertia, moduli of elasticity, etc.), which in turn, also
depend on the internal forces of the same. The design of an indeterminate structural
element is, therefore, carried out in an iterative manner, whereby the (relative) sizes
of each structural member is initially assumed and used to analyze the structure, and
the internal forces, thus, obtained are used to revise the member sizes and orientations.
If the revised member sizes are not same as to those initially assumed sections, then
the structure is reanalyzed using the latest member sizes. The iteration keeps continu-
ing till the member sizes based on the results of analysis are close to those assumed
for that analysis. Due to this iterative nature of structural analysis, computer-aided
programmers are called for to do the iterations correctly and within short time span.
Despite the difficulty in designing indeterminate structures, a great majority of struc-
tures being built in today’s modern world are statically indeterminate in nature; for
example, most modern reinforced concrete buildings are statically indeterminate, and all
the beams are supported at more than two supporting columns. These types of beams are
called continuous beams, for which we shall study the analysis procedure soon.
In this chapter, we will discuss some important advantages and disadvantages of
indeterminate structures as compared to determinate structures and will develop the
fundamental concepts for the analysis of indeterminate structures.

DOI: 10.1201/9781003081227-17 251

252 Introduction to Structural Analysis

14.2  ADVANTAGES OF INDETERMINATE STRUCTURE

The advantages of statically indeterminate structures over determinate structures
include the following:

1. Lesser Stresses and greater stiffness – The maximum stress intensity and
deflection in statically indeterminate structures is lower than those in similar
determinate structures. For example, let us consider the statically determinate
and indeterminate beams shown in Figure 14.1 (a) and (b), respectively.
Bending moment diagrams and deflections for the beams due to a uniformly
distributed load intensity, w, are also shown. The methods for analyzing
indeterminate structures will be introduced in subsequent chapters. It can
be understood from the figures that the maximum bending moment – and,
consequently, the maximum bending stress intensity as well as deflection – in
the indeterminate beam is much lesser than in the determinate beam. In other
way we can say, statically indeterminate structures generally have more stiff-
ness (i.e., smaller deflection), than those of similar determinate structures.
2. Redundancies – In statically indeterminate structures, if correctly designed,
acting loads are redistributed when certain structural elements become
overloaded or collapse in cases of overload due to earthquakes, wind, snow,
impact, and other events. In practical situations, indeterminate structures have
more members and/or supporting points than required for static stability. If a
part of such a structure fails, the entire structure will not collapse immediately,
and the loads will be redistributed to the adjacent members of the structure. In
transmission line towers, this is frequently seen where more members are con-
nected with the main leg elements to reduce the slenderness ratio of the main
members and, thus, ensure the overall structural stability even under worst

FIGURE 14.1  (a) Simply supported beam and (b) fixed beam bending moment and deflection
diagrams.
Introduction to Statically Indeterminate Structures 253

loading situations. These additional members are called redundant members

and play a crucial role in the overall structural stability. We will investigate the
effect of redundant members in other chapters of this section and establish the
analysis procedure for the same. Even in some redundant elements after car-
rying out analysis, it will be found that there is no force playing at all in those
elements. These elements are called zero force elements.

14.3  DISADVANTAGES OF INDETERMINATE STRUCTURE

The disadvantages of statically indeterminate structures over determinate structures
include the following:

1. Stresses due to support settlements – In many practical situations, supports

may get settled as shown in Figure 14.2, due to faulty design, weak soil
strata, or any unforeseen consequences due to excessive loading originated
from seismic or wind, or any other adverse conditions.
Under such circumstances, for determinate structures, no additional
stresses are developed due to such settlements. But in case of indetermi-
nate structures, support settlements induce additional internal force and
moments due to which structural elements undergoes higher stress concen-
trations than they are designed to withstand for. We will analyze the effects
of support settlements and additional force and moments coming from them
in subsequent chapters. But for now, it is to be understood that for inde-
terminate structures, support settlements cause adverse effect, which is a
major drawback of designing members as indeterminate in nature.
2. Stresses due to temperature change and fabrication errors – Change in
temperature and fabrication errors do not cause additional stresses in deter-
minate structures but may induce significant stresses in indeterminate ones
as shown in Figure 14.3, and 14.4, respectively.
Using the simple relationship for linear elongation/contraction of sol-
ids under change in temperature (α∆TL ), due to support fixity, axial loads
will come into play in the member or structural element. This will cause
additional direct stress generation in addition to the bending stress due to
external loading. This effect produces most adverse stress condition for fixed
beams. In such beams, the axial load due to temperature change will induce
axial compressive or tensile force, and thus, direct stress will be induced in

FIGURE 14.2  Effect of foundation settlement to determinate and indeterminate structures.

254 Introduction to Structural Analysis

FIGURE 14.3  Effect of temperature change to determinate and indeterminate structures.

FIGURE 14.4  Effect of fabrication errors to determinate and indeterminate structures.

the fixed beam along with the bending stress due to other external applied
loading. Thus, without proper care, the members will fail due to this com-
bined (direct as well as bending) excessive stress conditions. In case of fabri-
cation errors, if any member length accidentally becomes short or long, that
will be adjusted by the structural system by elongating or compressing the
particular member, respectively, for indeterminate structures. In contrast,
for determinate structures, joints will be shifted to new positions to accom-
modate the situation. Thus, material properties and their effect on change in
temperature, good fabrication practices need to be considered while analyz-
ing and designing indeterminate structures.

This chapter has provided a basic qualitative introduction to indeterminate struc-

tures and their advantages and disadvantages over determinate structures. We will
learn in detail in subsequent chapters about the analysis procedures for the indeter-
minate structures, which is the first step toward carrying out structural design engi-
neering work. We will also learn how to form additional equations depending upon
the geometry and constraint imposed by different supports.
 15 Approximate Analysis of
Statically Indeterminate
Structures

15.1 INTRODUCTION
The analysis of statically indeterminate structures applying the force and displace-
ment methods is considered as exact in the sense that the compatibility and equilib-
rium conditions of the structure are exactly satisfied in such an analysis. However,
the results of such an exact analysis represent the actual response to the extent that
the mathematical model of the structure reflects the actual structure under study.
Experimental analysis has established the fact, that the response of most common
types of structures under various acting loads can be correctly predicted by the force
and displacement methods, provided an accurate mathematical model of the struc-
ture is considered at the beginning of analysis.
An approximate method of analysis proves to be quite a convenient way to apply
in the preliminary planning phase of any project when several alternative design phi-
losophies of the structure are usually evaluated for optimized economic results. The
results of approximate analysis can also be used to assume the dimensions of vari-
ous structural elements needed to begin the exact analysis. The approximate dimen-
sions of various structural elements are then changed iteratively, using the results of
successive exact design and analyses, to achieve their final dimensions. Moreover,
approximate analysis is seldom used to roughly estimate the results of exact analysis
program, which due to its complexity and time-consuming calculations can be prone
to erroneous results. Finally, in recent times, there has been an increased interest
toward renovating and retrofitting older and heritage structures. So, a knowledge
and understanding of approximate methods used by the original designers is usually
helpful in a renovation of work progress.

15.2  ASSUMPTIONS FOR APPROXIMATE ANALYSIS

As discussed in the earlier chapter, statically indeterminate structures have more
support reactions and/or members/elements than are required for static equilibrium;
therefore, all the reactions and internal forces, including moments, of such struc-
tures cannot be determined from the equations of equilibrium alone. The additional
reactions and internal forces of an indeterminate structure are referred to as redun-
dants, and the number of redundants (i.e., the difference between the total number of
unknowns and the number of equilibrium equations available) is termed the degree
of indeterminacy of the structure. So, in order to calculate the reactions and internal
forces of an indeterminate structure, the equilibrium equations must be aided with
DOI: 10.1201/9781003081227-18 255
256 Introduction to Structural Analysis

additional equations, the number of which must be the same as that of the degree
of indeterminacy of the structure. In approximate analysis, these additional equa-
tions are formed by applying engineering judgment to make simplifying assump-
tions about the response of the structure under the action of external loads. The total
number of equations corresponding to assumptions must be equal to the degree of
indeterminacy of the structure. Each assumption leads to an independent relation-
ship between the unknown reactions and/or internal forces. The equations based on
the simplifying assumptions are then solved in line with the equilibrium equations of
the structure to determine the approximate values of its reactions and internal forces.
Two assumptions usually applied to carry out the approximate analysis are
detailed in the following sections.

15.2.1 Assumptions about the Location of Points of Inflection

As a starter, a qualitative deflected shape of the indeterminate structure is sketched
and applied to determine the location where the curvature of elastic line changes its
direction/sign or becomes zero. These points are known as the point of inflection.
At the inflection points, bending moments also become zero. Thus, to model zero
bending moment points, internal hinges are inserted in the indeterminate structure
at the assumed inflection points to obtain a general determinate structure. By insert-
ing internal hinge points, each hinge provided additional condition, thereby provid-
ing additional equations. Moreover, the inflection points should be placed in such
a way, so that the resulting determinate structure, thus formed, must be statically
and geometrically stable. The simplified determinate structure thus obtained is then
analyzed to determine the approximate values of the reactions and internal forces of
the original indeterminate structure.
To understand the above concept, let us consider an example of a statically indeter-
minate portal frame. See Figure 15.1 for the actual portal frame and load acting on it.
We start by drawing possible deflected shape of the frame due to the applied load-
ing. The deflection diagram has been shown in Figure 15.1 for reference. From this
diagram, it seems that the inflection point is situated near the center of the span CD.
Although to determine the exact location of inflection point, we need to carry out the
exact analysis. With this assumption, we insert a hinge at the middle of the span CD
and applying the concept of that moment at hinge or inflection point is zero; thus, we
get the following equations:

∑M B =0

Ay l − Ph = 0
or,
Ph
Ay =
l
∑F = 0
y

Ph
− + By = 0
l
Approximate Analysis of Statically Indeterminate Structures 257

FIGURE 15.1  (a) Indeterminate frame with external loading, (b) corresponding simplified
determinate structure, and (c) approximate bending moment diagram.

or,
Ph
By = 
l
Similarly, calculating bending moment at point B, we get from the left of section with
respect to point B:
l
M x BE = By − Bx h
2
Since, at point B, we have the point of inflection, hence,

M x BE = 0

Resulting,
l P
By = Bx h or, Bx =
2 2
Thus, progressing the abovementioned way, we can easily determine all other
unknown support reactions at the other support point A. So, approximate analysis is
found out to be handy while carrying out the structural analysis work for indetermi-
nate structures.
258 Introduction to Structural Analysis

15.2.2 Assumptions about the Distribution of Forces and Reactions

Approximate analysis of indeterminate structures can also be performed by assum-
ing that the distribution of forces among the members and support reactions of the
structures. The number of these assumptions adopted for the analysis of the structure
is equal to the degree of indeterminacy of the structure. Each assumption provides
an independent equation combining the unknown member forces and support reac-
tions. The equations are then solved simultaneously with the equilibrium equations
of the structure to calculate its approximate reactions and internal forces. For exam-
ple, the same problem in the previous example can be solved by assuming that hori-
zontal support reactions at two supports are equal. By assuming this, one can solve
and analyze the entire frame without any difficulty. The same is left as an exercise
for the readers.

15.3  VERTICAL LOADS ON BUILDING FRAMES

The most accepted procedure for approximate analysis of building frames subjected
to vertical loads (mainly due to gravity) required taking three assumptions about the
nature of each girder of the frame. To understand the concept, let us take a building
frame subjected to uniformly distributed loads w, as shown in Figure 15.2. From the
typical free body diagram of member DE and its deflected shape, the point of inflec-
tions is located near the column support points D and E. Since the member DE
is fixed and connected with the framing columns, hence, the column provides end

FIGURE 15.2  (a) Building frame subjected to vertical load, (b) typical girder, (c) simply
supported girder, and (d) ideally fixed girder.
Approximate Analysis of Statically Indeterminate Structures 259

restraint against rotation of the member due to applied loading. Actual location of the
inflection points depends upon the proper analysis of indeterminate structure, which
we will discuss later. However, for approximate analysis, we can at least get a feel of
the extreme conditions and its effect toward the deflection of the said member under
investigation. If the girder was simply supported, bending moment would be zero at
the support points and thus inflection points would lie at the support locations too.
But since this is not simply supported, hence for practical purposes, we can assume
that the point of inflection lies at 0.11L from each support. This is half of the actual
distance of inflection point as per exact analysis. However, for approximate analysis,
the assumed distance of 0.11L is proved to be very effective one.
The third assumption is a very simple one and it is known from the exact analysis
of structures. From exact analysis, for a building frame subjected to vertical loads,
member force (force acting along the axis of member) is very negligible and hence
can be neglected. We take the same assumption accordingly while analyzing the
same frame using approximate analysis procedure.

15.4  LATERAL LOADS ON BUILDING FRAMES: PORTAL METHOD

Response of any rectangular building frames is unique and different under lateral
(horizontal) loads than under vertical loads explained in the previous section. Hence,
completely different assumptions must be adopted in the approximate analysis for
lateral loads. Two methods are popularly applied for approximate analysis of rectan-
gular frames subjected to lateral loads. These are: (1) the portal method and (2) the
cantilever method. In this section, we will discuss the portal and cantilever methods.
In portal analysis, it is assumed that point of inflections lies at the midpoint of
each member of the portal frame under investigation. To understand this concept, let
us take the example of a simple portal frame that is indeterminate to third degree, as
shown in Figure 15.3.
As seen in the above diagram, there are three points of inflections in the frame
element. These three points of inflection can be modeled as internal hinge points at
the same location. By inserting the internal hinges at these members, the frame looks
similar to the one shown in Figure 15.4.

FIGURE 15.3  Building frame analysis by portal method.

260 Introduction to Structural Analysis

FIGURE 15.4  Building frame with internal hinges.

With the assumption that point of inflection lies at the center of each member, we
can solve the indeterminate structure by considering moment at the inflection point,
thereby eliminating the unknown support reactions one by one. Since all the internal
hinges act like virtual supports, hence, there will be induced support reactions due
to the external loading. We will show some steps of evaluating the unknown reaction
forces and others will be left as an exercise for the reader.
At each internal hinge points, we have marked unknown horizontal and verti-
cal reaction forces, as shown in Figure 15.4. To solve the above frame, we pass an
imaginary section through point E and G of the above frame and draw the free body
diagram of the sections with all forces acting at the respective cut locations (for recap
of Method of Sections refer Chapter 7).
From Figure 15.5, taking moment about G, we get:

h
P − Eyl = 0
2

or,
h
Ey = P
2l

FIGURE 15.5  Sectional force free body diagram of simplified determinate portal frame.
Approximate Analysis of Statically Indeterminate Structures 261

Similarly, by taking moment of all forces about point E, we get:

h
Gy = − P
2l
Now, the bending moment at the point of inflection F, we get:
h l
M F − Ex − Ey = 0
2 2
Since, bending moment at point F is zero, hence, we get:
h h l
Ex −P   =0
2 2l 2
or,
P
Ex =
2
Now, applying force equilibrium equation in x direction, we get:
P
Gx + −P=0
2
or,
P
Gx = 
2
Hence, by adopting a simple assumption about the location of inflection points, we
can determine the unknown support reactions by applying force and moment equi-
librium equations.
As we have just learned about the application of the portal method for statically
indeterminate single bay frame, the same understanding can be applied to multiple
bay portal frames of buildings to analyze the unknown forces and moments acting
in the frame due to eternal loading. To understand this, let us consider the following
indeterminate frame of a multi-bay building, as shown in Figure 15.6.

FIGURE 15.6  Multi-bay simplified building frame.

262 Introduction to Structural Analysis

FIGURE 15.7  Multi-bay building frame – shear force distribution.

Following the same procedure, we assume that the point of inflection exists at the
midpoint of each member of the above multi-bay frame. Now, before inserting the
internal hinges, the total number of unknown support reactions are = 3 × 4 and num-
ber of girders = 12. Now, with the introduction of internal hinges (ten in total), the
structure does not become statically determinate. Hence, with the internal hinges,
the net degree of indeterminacy becomes 12  −  10  =  2. Hence, additional two equa-
tions or conditions are required to make the structure statically determinate. Hence,
in addition to assumption on location of inflection point, we need another two condi-
tions to make this structure statically determinate, which will enable us to complete
the analysis.
As analyzed above, we have found under the application of horizontal loading,
the perimeter columns carry shear force equal to the half of the applied load-
ing. Since internal columns represent two legs of the two portal frames placed
side by side, hence, it may be assumed that the internal columns can carry twice
more load than the peripheral columns. Thus, this assumption provides additional
equations for each column in the portal frame. Hence, total available equations
become 1 × 3 = 3. However, we needed only two additional equations and we got
three in turn. However, since all these additional equations are derived based
on the same principle, hence, one extra equation than required does not pose
any problem for analysis. Hence, following this assumption, we get the following
Figure 15.7 related to the analysis program of multiple bays building frames by
portal frame method.
Although we have tried to provide the detailed approximate calculation tech-
niques by portal frame methods, we will provide Example 15.1 for better under-
standing of the concept below.

Example 15.1:  Calculate the member forces using portal frame method
of the building frame, as shown in Figure 15.8.

SOLUTION:  After inserting the hinges at midpoint of each member, we pass a

section a − a near the midportion of the frame and another section b − b at the
bottom of the total system. After that, we draw the free body diagram for the
Approximate Analysis of Statically Indeterminate Structures 263

FIGURE 15.8  Multi-bay building frame – example problem.

separated frames along with horizontal shears, as explained earlier. The free body
diagrams will be something like as shown in Figures 15.9 and 15.10.
From the free body diagrams as mentioned, applying force equilibrium equa-
tion in x direction:

∑F = 0
x

Thus,

10 − RC − 2RC − RC = 0

or,

RC = 2.5 kN

30 − SC − 2SC − SC = 0

or,

SC = 7.5 kN

Once we calculate the shear forces acting in the column, we are now able to ana-
lyze the forces in members, as will be described. Let us open a joint located at C
and draw free body diagram of the connected member at this node.

FIGURE 15.9  Free body diagram of portal frame of a–a section.

264 Introduction to Structural Analysis

FIGURE 15.10  Free body diagram of portal frame of b–b section.

Now, applying force equilibrium equation in y direction, we get:

VK +  VJ = 0

So,

VK = −VJ

Now, considering bending moment at point K of the member CK, we get:

8
MCK − 10 × =0
2
or,

MCK = 40 kNm

Now, since at joint C, total summation of moments will be zero to maintain equi-
librium condition, hence, we get:

MCJ =   −MCK =   −40 kNm

FIGURE 15.11  Free body diagram of joint C.

Approximate Analysis of Statically Indeterminate Structures 265

Once the nodal moments are calculated, we can take moment of all forces about
any points, J or K, which will help us to determine the other unknown forces act-
ing in the member. As for example, if we take moment of all forces about point J,
we get:

8 20
SC ×   −  VK × =0
2 2

Solving the above equation, we get:

7.5 × 4
VK = = 3 kN
10

Hence, for the node C and connected members in this node, we have completely
analyzed all unknown forces and moments for the same. Proceeding in this way,
students are encouraged to complete the analysis of this frame by the same fash-
ion. In the next section, we will introduce another effective approximate method
for analyzing multistoried building frames, which is popularly known as cantilever
method.

15.5 LATERAL LOADS ON BUILDING FRAMES:

CANTILEVER METHOD
This method was initially developed by A. C. Wilson in 1908 and is generally applied
mainly for the approximate analysis of tall building frames. Cantilever method
assumes that under the application of lateral loads, building frames act like canti-
lever element, as shown in Figure 15.12. From the strength of materials, we already
know that the axial stress on a cross section of a cantilever beam subjected to lateral/
transverse loads varies linearly with the distance from the centroidal axis (neutral
line), so that the longitudinal fibers of the beam on the concave side of the neutral line
are in the state of compression, whereas fibers on the convex side remain in tension.
As we determine the neutral line for the beams under the application of transverse
loading, for building, we determine the centroidal line accordingly. This centroidal
line behaves in the same way as the neutral line behaves for the building structure.
Framing columns are then considered as the different fibers of the beam, and axial
stress distribution, at the mid-height of column, is assumed to be linearly propor-
tional to the distance of the column from the centroidal axis of the building. If it
is further assumed that the cross-sectional areas of all the columns throughout the
building structure, the axial force in each column will be linearly proportional to the
distance of the column from the centroidal line of the building.
In addition to the abovementioned assumption, the cantilever method uses the
same assumption regarding the location of inflection points, the same as that in the
portal method. Thus, the assumptions of cantilever method can be stated as follows:

1. An inflection point is located at the midspan of each member of the frame.

2. In each story, the axial stress in columns is directly proportional to the dis-
tance of the column from the centroidal line.
266 Introduction to Structural Analysis

FIGURE 15.12  Tall building frame deflected shape under the application of horizontal load.

3. As this method also assumes the same location of inflection points as that
for the portal method, hence, at midspan of each member there will be
internal hinge indicating the location of the inflection point.

With this two assumptions and subsequent analysis procedure, we will show through
Example 15.2 that this method is very effective toward determining the approximate
force and moments in each and every framing elements of any tall building structure.

Example 15.2:  Calculate the member forces using portal frame method
of the building frame, as shown in the Figure 15.13.

SOLUTION: To solve this problem using cantilever method, we first insert the
internal hinge points at each member midspan location. Then we pass two imagi-
nary section lines along the hinge points to form the reduced or simplified frame.
In reduced frame thus obtained, we then draw the forces and moments as we
show in any free body diagram. The free body diagram of the frame is shown in
Figure 15.14.
Approximate Analysis of Statically Indeterminate Structures 267

FIGURE 15.13  Building frame for cantilever method.

Assuming all columns are of the same cross-sectional area = A, the centroidal
line location can be determined by taking area moments with respect to line CB:

x × ( 3A) = 20 × A + 35 × A

or,

x  =
( 20 × A + 35 × A) = 18.33 m
3A

Once the location of CG line is determined, we immediately know that the col-
umns to the left of CG line will be under tension, whereas the columns to the right

FIGURE 15.14  Free body diagram of reduced building frame.

268 Introduction to Structural Analysis

of CG line will be under compression. Hence, the force triangle will be as shown
Figure 15.14. Since the force in the columns is directly proportional to the distance
of the same from the centroidal axes, hence, from a similar triangle, we can write:

FCB F
=   FE
18.33 1.67
or,

FCB × 1.67
FFE = = 0.091 FCB
18.33
Similarly,

FIH F
=   CB
16.67 18.33
or,

FCB × 16.67
FIH =   = 0.91 FCB
18.33

Now calculating the moment at joint J, we get:

∑M = 0J

So,

6 × 10 − 20 × FFE − 35 × FIH = 0

or,

60 − 20 × 0.091 FCB − 35 × 0.91 FCB = 0

or,

FCB = 1.782 kN

So, from earlier derived relationships, we get:

FFE = 0.162 kN

FIH = 1.621 kN

Following the same procedure, we can determine the column forces of the lower
segment columns by drawing an imaginary line through the inflection points as
shown in Figure 15.12 and taking the drawing of the triangle of forces as per ten-
sion and compression states of the member. Column member forces for all other
columns are left as exercise for the interested readers.
Once all the column member forces are determined then we can proceed to
calculate the girder forces and moments by drawing suitable free body diagrams and
applying force or moment equilibrium equation or both suitably as per requirement.
Approximate Analysis of Statically Indeterminate Structures 269

FIGURE 15.15  Free body diagram of girder CF.

To understand the concept, let us calculate the member force and moments in the
girder CF of the above frame. The free body diagram of the girder with all forces
and moment is shown in Figure 15.15.
Taking moment about hinge J, we get:

MCB − 10 × 6 = 0

or,

MCB =  60 kNm

Since at joint C, sum of all moments should be zero to maintain equilibrium,

hence,

MCB +  MCF = 0

or,

MCF =   −60 kNm

Similarly, by taking moment of all forces about point C for the span CJ, we get:

VCB × 6 −  MCB = 0

or,

VCB = 10 kN

Following these procedures, we can carry out all the forces and moments in all
members of the frame using the cantilever method. All other girders are left as an
exercise for the readers. It is customary to mention here that by merely studying
theories and worked examples will not help one to master the underlying con-
cepts and problem-solving techniques. To be able to have a good hold on any
mathematical topic, it is of utmost importance to go through the exercises in detail
and carry out all necessary steps to obtain the result.
 16 Method of Consistent
Deformations

16.1 INTRODUCTION FORCE METHOD

OF ANALYSIS: GENERAL PROCEDURE
This chapter will discuss a general analysis procedure of the force (flexibility) method,
also called the method of consistent deformations, to analyze statically indeterminate
structures. James C. Maxwell introduced this method in 1864, and this involves
removing enough restraints and/or constraints from the indeterminate structure to
convert it into a statically determinate structure. This determinate structure, thus
obtained, must be statically (and geometrically) stable and is known as the primary
structure. The excess restraints (and/or constraints) taken away from the indeter-
minate structure to make it determinate primary structure are called redundant
restraints, and the reactions or internal forces related with these restraints are called
redundants. The redundants are then applied as unknown forces on the primary
structure, and their values are determined by solving the compatibility equations.
Compatibility equations are being solved based on the condition that the actual dis-
placement is a linear superposition of displacements of the primary structure with the
applied external load and load due to force of constraint. In this method of analysis,
the main unknown forces are the redundant reactions, and they must be evaluated
at first before evaluating other unknown parameters like deflections and rotations.
Since the forces are first determined, this method is known as the force method.
In this chapter, we will also learn how to apply Maxwell-Betti’s law to analyze
indeterminate structure to determine the unknown forces and moments.

16.2  STRUCTURES WITH A SINGLE DEGREE OF INDETERMINACY

To understand the concept of force method of analysis for single degree of indeter-
minacy, let us consider the following example of propped cantilever as shown in
Figure 16.1.
So, for the shown propped cantilever, we have three unknown support reactions
at A (one vertical, one horizontal, one moment) and another unknown support reac-
tion is at support B (since this is a roller support, we have only one unknown vertical
support reaction). So, the total unknown support reactions are four, and we have
three equilibrium equations, namely, ∑ Fx = 0,  ∑ Fy = 0,  ∑ M x = 0. So, degree of
indeterminacy = 4 − 3 = 1; hence, the system has one more reaction than required
for static stability. Now to analyze the structure, we need to remove one redundant
support reaction and make it to determinate primary structure. In primary structure,
there will be all external loading acting on the system.

DOI: 10.1201/9781003081227-19 271

272 Introduction to Structural Analysis

FIGURE 16.1  Propped cantilever.

So, for the applied loading, the bending moment at the support is M A = P ( l /2 ).
Now from the bending moment diagram, we can apply moment area theorem to cal-
culate the deflection at free end B, by taking moment of bending moment area about
support B and dividing the same by EI (flexural rigidity of the beam).

(1/2 ) × P ( l /2 ) × ( l /2 ) 5l 5Pl 3
δB = × =−
EI 6 48 EI

Once this deflection at free end has been calculated from the primary structure, we
are in a position to apply the unknown redundant reaction at the propped end to cal-
culate the deflection in the reverse direction. Here the downward deflection at B due
to the external loading is considered as negative.
Let the unknown support reaction at propped end is given by RB. Now due to this
support reaction, the deflection at end B in vertical upward direction will be:

(1/2 ) × RB l × l 2l RB l 3
δB = × =
EI 3 3EI

FIGURE 16.2  Primary structure and bending moment diagram due to external loading.
Method of Consistent Deformations 273

Since there is no actual deflection at the free end, the total algebraic sum of deflections
at the propped end should be zero (this is also known as principle of superposition).
Thus, we obtain the following equation:

RB l 3 5Pl 3
−   = 0
3EI 48 EI

or, solving the above equation we get,

RB l 3 5Pl 3
= 
3EI 48 EI
or,

5P
RB =   ↑
16

Hence, we have solved the indeterminate structure and calculated the redundant sup-
port reaction by the application of force method and in conjunction with principle of
superposition. As stated earlier, we have tackled the problem starting with redundant
force; hence, the name force method came into picture. This simple trick for calcu-
lating the redundant forces or moments by converting the indeterminate structure
into determinate primary structure can be applied in general type of indeterminate
structures as well for analysis.

16.3  METHOD OF LEAST WORK

In this portion, we consider an alternative procedure of the force method known as
the method of least work. For this analysis process, the compatibility conditions are
formed with the help of Castigliano’s second theorem instead by deflection super-
position method, as in the method we have studied previously. With this difference,
the two methods are very similar and demand essentially the same amount of com-
putational work. The method of least work generally proves to be easy for analyzing
composite structures that contain both axial force and flexural members (e.g., beams
supported by columns or cables). However, the method cannot be treated as general
as the method of consistent deformations in a sense that in its original process (as
presented here), this method cannot be applied for analyzing the impact of support
settlements, temperature changes, and fabrication errors.
To derive the method of least work, let us take a statically indeterminate beam
with unyielding supports subjected to an external loading P, as shown in Figure 16.3.

FIGURE 16.3  Indeterminate beam and its deflected shape.

274 Introduction to Structural Analysis

Suppose that we select the vertical reaction By at the support point B to be the redun-
dant reaction. By taking the redundant as an unknown load applied to the beam
along with the applied loading P, relation for the strain energy can be written as a
function of the known load P and the unknown redundant By as:

U =   f ( P , By )

According to Castigliano’s second theorem, the partial derivative of the strain energy
with respect force is equal to the deflection of the structure at the point where load
has been applied. Also, the direction of deflection will be consistent with the direc-
tion of the applied load. Hence, for deflection in the direction of redundant reaction,
the following partial differential equation forms:

∂U
=0
∂ By

Above equation indicates that the value of the redundant that satisfies equilibrium
and compatibility conditions, the strain energy of any elastic system attains maxi-
mum or minimum value. Since for linearly elastic bodies, there is no upper limit for
elastic strain energy, we conclude that, for true value of redundant, strain energy
attains minimum value. So, in short, the magnitude of redundant reaction in any
indeterminate structure must be such that the strain energy attains its minimum
value at that situation.
The method described above is suitable for structures with higher degrees of inde-
terminacy as well. In such case, there will be n number of partial differential equa-
tion combining the elastic stored energy and redundant forces. The number of partial
differential equation is same as that of the degrees of indeterminacy as shown below:

∂U
=0
∂ B1

∂U
=0
∂ B2







∂U
=0
∂ Bn

To understand the concept of energy principle, let us consider the previous section
example with the redundant force as Ry.
Method of Consistent Deformations 275

Under the application of load as was shown in Figure 16.2, the bending moment
at any section x from right support B is given by,

l
M x = Ry x − P  x − 
 2

So, total strain energy stored in the beam is given by,

l
M x2
U = 
∫
0
2 EI
dx

or, in full terms, the above relationship becomes:

2
  l 
 Ry x − P  x −
l 
2  

U = 
∫
0
2 EI
  dx

Now, from principle of least work, we know that the partial derivative of internal
strain energy with respect to redundant force will be zero. Hence, we get,

∂U
=0
∂ Ry

which implies,

M x ( ∂ M x / ∂ Ry )
l
∂U

∂ Ry
=
∫
0
EI
dx = 0

So,

∂M x
=
((
∂ Ry x − P ( x − ( l /2 )) / ∂ Ry ) )
∂ Ry EI

or,

∂M x x
=
∂ Ry EI

which implies,

∂U
l
M x ( ∂ M x / ∂ Ry )
l
( R x − P ( x − (l /2)))( x ) dx = 0
∫ ∫
y
= dx =
∂ Ry EI ( EI )( EI )
0 0
276 Introduction to Structural Analysis

or, performing the integration, we get,

l3  l3 l3 
Ry − P −  = 0
3  3 4

or, solving this equation, we get,

5P
Ry =  
16

It is same as we have derived in the previous section. So, method of virtual work pro-
vides another alternative way to effectively calculate the unknown support reactions
for indeterminate structures.

16.4 STRUCTURES WITH MULTIPLE DEGREE

OF INDETERMINACY
The method of consistent deformations introduced in the previous sections for ana-
lyzing structures with single degree of indeterminacy can easily be extended to the
analysis of structures with multiple degrees of indeterminacy. To understand the
concept, let us consider an example of four-span continuous beam subjected to a uni-
formly distributed load w, as shown in Figure 16.4 (a). As can be seen, there are six
unknown support reactions. Since this is a planar structure, the total available equi-
librium equations are three. Hence, the degree of indeterminacy will be = 6 − 3 = 3.
Following the logic of previous sections, we need to take any three unknown support
reaction as redundant. To begin with, let us consider the vertical reactions at support
B, C, and D to be the redundants. The roller supports at B, C, and D are then removed
from the given indeterminate beam to form statically determinate and stable primary
beam, as shown in Figure 16.4 (b). Once the primary beam is formed, we can now

FIGURE 16.4  (a) Indeterminate structure with multiple degrees of indeterminacy and
(b) corresponding primary structure.
Method of Consistent Deformations 277

FIGURE 16.5  Three-span continuous beam with redundant reactions.

determine the vertical deflection at the support s using any of the known methods.
Since there will be no deflection, we apply redundant forces as loads at these points
to calcite the deflection I opposite direction. The algebraic sum of deflection due to
these two cases should add up to zero.
To better understand the analysis procedure for the same, let us consider that the
uniformly distributed load applied on the three-span continuous beam as shown in
the following diagram is 2 kN/m. The span of each segment of the beam is 20 m.
Now we have here four supports with the support A as hinge, and all other sup-
ports are roller supports. Hence, total unknown is 2 + 1 + 1 + 1 = 5. Total available
equilibrium equations are 3 ( ∑ Fx = ∑ Fy = ∑ M z = 0 ). Hence, degree of indetermi-
nacy of the structure is 5 − 3 = 2.
Now, to solve this problem, we must take two support reactions as redundant.
Let us take RB and RC. In this situation, we remove roller supports at B and C to
make the determinate primary beam as shown in the above diagram. Next, we apply
unit load at the redundant location B and C separately and calculate the deflection
in the reverse order as shown in the above diagram. As can be seen from the unit
load diagrams, there are few new terms introduced, namely, f BB, f BC , fCC , and fCB.
These terms are known as flexibility coefficients. These coefficients are calculated
based on the deflection due to unit load applied at a certain point on the structure.
278 Introduction to Structural Analysis

f BB indicates the deflection at point B when unit load is applied there. First suffix
represents point of application of unit load and second suffix indicates the location at
which deflection has been considered. Thus, fCB is the deflection at point B when unit
load has been applied at point C. Thus, from the unit force diagrams, we can easily
determine the deflection in the direction of force by applying previously acquired
knowledge on any one of the deflection calculation methods.
Understanding the concept of flexibility coefficients is quite useful when we are
dealing with structures having multiple degrees of indeterminacy. Now, considering
the actual redundant force RB and RC, the total deflection at support B due to redun-
dant forces is f BB RB + f BC RC . So, the compatibility equation for deflection at support
B is given by,

∆ B − f BB RB − f BC RC = 0

Similarly, for support at C, we can have another compatibility equation as given

below:

∆ C − fCB RB − fCC RC = 0

Since all the parameters except the redundant forces are unknown, we can simulta-
neously solve these two equations to get two unknown support reactions.
Once redundant support reactions are obtained, we can apply principle of equi-
librium conditions alone to determine the bending moment and shear force diagram
for the entire beam. In Figure 16.6, the bending moment and shear force diagram is
shown. Though the detailed calculation to determine the support reactions and bend-
ing moment diagrams is left as an exercise for the readers.

FIGURE 16.6  (a) Bending moment and (b) shear force diagrams for three span continuous
beam.
Method of Consistent Deformations 279

16.4.1 Shear and Bending Moment Diagrams

of Three-Span Continuous Beams

The shear and bending moment diagrams of the previous example beam are shown
in Figure 16.6. The shapes of the shear and bending moment diagrams for continu-
ous beams can be drawn only after complete determination of redundant support
reactions as explained above. Once all unknown support reactions are known, we
can proceed from any one extreme support end, and proceed progressively toward
the other end as we have done for statically determinate beams. Once bending
moment equations are obtained, we vary the length parameter x along the beam axis
to determine magnitude of bending moment at different points on the axis of beam.
In general, the following observations will be formed once we complete the analysis
of the beam:

1. In case uniformly distributed loading, the bending moment diagrams in

general will be parabolic in nature with negative value at the interior sup-
ports and positive values near the mid span of the beam. At the extreme
supports, if they are hinged, then bending moment will be zero. Otherwise,
it will be generally negative.
2. In case of point loading, the bending moment diagrams will be linear seg-
ments with negative moment at the supports and positive moments at the
mid span of the beam.

However, actual values of the bending moments, of course, depend on the mag-
nitude of the loading as well as on the lengths and flexural rigidities of the spans
of the continuous beam. It is strongly suggested that students should carry out the
detailed analysis and match the magnitude of shear and bending moment as given in
Figure 16.6, along with proper algebraic signs.

16.5 SUPPORT SETTLEMENTS, TEMPERATURE CHANGES,

AND FABRICATION ERRORS
So far, we have discussed the analysis of structures with unsettled supports. As dis-
cussed in previous chapters, support settlements due to inadequate foundation sys-
tem and the like may result significant stresses in externally indeterminate structures
and must be considered in the designs. Support settlements, however, do not cause
any considerable effect on the stress conditions of structures that are internally inde-
terminate but externally determinate. This is majorly due to the fact that settlements
cause structures to move or rotate as a rigid body. The method of consistent deforma-
tions, as developed in the previous sections, can be conveniently modified to include
the effect of support settlements in the analysis.
Let us consider, as example, the same continuous beam subjected to a uniformly
distributed load w, as shown in Figure 16.7 (a). In this situation, let us also consider
that the supports B and C of the beam undergo small settlements ∆ B and ∆ C , respec-
tively, as shown in the same figure. To analyze the beam, like previous example, we
consider the vertical reactions RB and RC as the redundants. The supports B and C
280 Introduction to Structural Analysis

FIGURE 16.7  Three-span continuous beam with settlement of supports.

are then removed from the indeterminate beam to form the primary beam, which is
then applied separately to the external load w and the unit values of the redundants
RB and RC as shown in Figure 16.7 (b), (c), and (d), respectively. By realizing that the
deflections of the actual beam at supports B   and C are equal to the settlements ∆ B
and ∆ C , respectively, we obtain the compatibility equations as given below:

∆ BO + f BB RB + f BC RC = ∆ B

∆ CO + fCB RB + fCC RC = ∆ C
Method of Consistent Deformations 281

Since all the parameters except the redundant support reactions are unknown in
the above equations, we can easily solve them simultaneously to obtain the unknown
redundant forces and thereby complete the analysis of the indeterminate structure.
Although support settlements are usually defined with respect to the undeformed
position of the indeterminate structure, the magnitudes of such settlements to be
used in the compatibility equations must be calculated from the chord connecting the
deflected positions of the supports of the primary structure to the settled positions
of the redundant supports. Any such support settlement is positive if it has the same
direction as that assumed for the redundant. For our case, the beam of Figure 16.7,
since there is no settlement at supports A and D, the chord AD of the primary beam
remains same as that for the undeformed position of the indeterminate beam. Thus,
the settlements of supports B and C relative to the chord of the primary beam are
equal to the prescribed settlements ∆ B and ∆ C , respectively.
Now, let us consider a more severe case when all the supports of a beam undergo
settlement as shown in Figure 16.8. If we consider the reactions RB and RC as redun-
dants, then the displacements ∆ BF and ∆ CF to be the of supports B and C, respectively,
in respect to the chord of the primary beam should be applied in the compatibility
equations instead of the specified displacements ∆ B and ∆ C . This is because only the
displacements relative to the chord induce stresses in the beam. In other words, if all
the supports of the beam would have settled either by equal amounts or by amounts
so that the deformed positions of all of the supports would lie on a straight line, then
the beam would remain straight without bending, and no stresses would develop in
the beam. In this case, as no external load is present on the beam, the compatibility
equations can be written as:

f BB RB + f BC RC = ∆ BF

fCB RB + fCC RC = ∆ CF

FIGURE 16.8  Three-span continuous beam with settlement of all the supports.
282 Introduction to Structural Analysis

16.5.1  Temperature Changes and Fabrication Errors

Apart from support settlements, which affect only externally indeterminate struc-
tures, temperature changes and fabrication errors may affect the stress conditions
of both externally and internally indeterminate structures. The method to analyze
these structures subjected to temperature changes and/or fabrication errors is the
same as used previously for the case of external loads. The only difference is that
the primary structure is now applied with the prescribed temperature changes and/or
fabrication errors (instead of external loads) to evaluate its deflection at the locations
of redundants due to these cases. The redundants are then calculated by applying
the usual compatibility conditions that the deflections of the primary structure at
the locations of the redundants due to the combined effects of temperature changes
and/or fabrication errors and the redundants must equal the known deflections at
the corresponding locations on the actual indeterminate structure. The procedure is
explained by the Example 16.1.

Example 16.1:  Determine the reactions and the force in each member
of the truss shown in Figure 16.9 due to a temperature increase of 45°C
in member AB and a temperature drop of 20°C in member CD. Use the
method of consistent deformations. Cross-sectional area of each member
is 5000 mm2 and diagonal members are 3000 mm2, E = 200 GPa, and
coefficient of linear thermal expansion, α = 1.2 (10−5 )/°C.

SOLUTION: The degree of indeterminacy is m + r − 2 j = 6 + 3 − 2 × 4 = 1. Thus,

the truss is internally indeterminate to first degree. Hence, let us consider one
diagonal element, say AD, as the redundant member. Hence, the primary truss
is first obtained by removing the redundant member from the original structure.
Once it is removed, it will look similar to the one shown in Figure 16.10.

FIGURE 16.9  Original truss as per example problem statement.

Method of Consistent Deformations 283

FIGURE 16.10  Primary truss subjected to (a) temperature change and (b) unit tensile force
along direction of redundant AD.

Next, we apply prescribed temperature changes in the primary truss as shown

in Figure 16.10 (a), and separately a unit load applied in the direction of AD as
shown in Figure 16.10 (b). The virtual work done by any member due to change
in temperature will be:

∆ ADO +   fAD , ADFAD = 0

The above equation is the compatibility equation for the member AD.
where ∆ ADO is the relative displacement between joints A and D of the primary
structure due to temperature changes. fAD , AD is the flexibility coefficient denoting
the relative displacement between the same joints due to a unit value of the redun-
dant FAD . The virtual work expression for ∆ ADO can be given as:

∆ ADO   =   ∑α ( ∆T ) Lu AD

where, u AD is the force in the member due to unit load in the member AD as shown
in Figure 16.10 (b). Since there is no externally applied load on the truss, applying
global equilibrium equations for planner structure, we can determine the support
reactions of the truss, which will be all zeros. That is why in Figure 16.10 (b), we
have shown the support reactions with 0 values. Now, the value u AD is calculated
using standard procedure for truss analysis as we have learned in earlier chapters
on the same. As for example, let us calculate the member force AC due to unit ten-
sile force in member AD in the primary truss as shown in Figure 16.10 (b). Resolving
the forces and applying force equilibrium equation in y direction, we get,

FAC + 1× sin Ø = 0

or,

6
FAC =   − sin Ø =   − =   −0.6 kN
10
284 Introduction to Structural Analysis

Thus,

u AC =   −0.6 kN

Following the above procedure of method of joints (or any other methods as
one may wish), we can determine all the member forces from the primary truss in
Figure 16.10 (b).
The calculation process has been provided below in tabular form for ease of
understanding:

 kN 
u AD 
 kN  ( ∆T ) Lu AD 2
u AD L /A
Member L (m) A (sq. m) ∆T ( °C ) (kN/kN) (°C m) (1/m) F = u AD FAD (kN)
AB  8 0.005 45 −0.8 −288 1024 −32.067
CD  8 0.005 −20 −0.8 128 1024 −32.067
AC  6 0.005 0 −0.6 0  432 −24.05
BD  6 0.005 0 −0.6 0  432 −24.05
AD 10 0.003 0 1.0 0 3333.344  40.085
BC 10 0.003 0 1.0 0 3333.344  40.085
∑ −160 9578.688

Now, applying the above mentioned formula,

∆ ADO   =   ∑α ( ∆T ) Lu AD ( )
= 1.2  10 −5 ( −160 )(1.0 ) =   −0.00192 m =   −1.92 mm

(10 )m = 0.0479 mm/kN
−6

∑ u A L =   200 
2
1 9578.667
fAD , AD =   AD
= 47.893
E (10 ) 6
kN

∆ ADO
FAD = − = 40.084 kN ( tension)
fAD , AD

With the above obtained values, we can use the compatibility equation to obtain,

−1.92 + ( 0.0479) FAD = 0

or,

FAD = 40.084 kN (T )

Since the truss is statically determinate externally, the support reactions will be
zero due temperature changes. Also, member forces in other members can be
determined from the table and using the relationship, F = u ADFAD.
 17 Influence Lines for
Statically Indeterminate
Structures

17.1 INTRODUCTION OF INFLUENCE LINES FOR

STATICALLY INDETERMINATE STRUCTURES
In this chapter, we will discuss the methods for constructing influence lines for stati-
cally indeterminate structures. It may be recalled from Chapter 11 of Part II of this
book, that an influence line is a diagram of a response function of a structure as a
function of the position of a downward unit load moving across the structure. The
core procedure for constructing influence lines for indeterminate structures is very
much the same as that of determinate structures. The steps essentially demand cal-
culating the values of the response function of interest by varying positions of a unit
load on the structure and drawing the response function values as ordinates versus
the position of the unit load as abscissa to obtain the influence line.
As we have seen, the influence lines for forces and moments of determinate struc-
tures consist of straight-line segments. The influence lines are drawn in Chapter 11
by evaluating the ordinates for only a few locations of the unit load and connecting
them with straight lines. The influence lines for indeterminate structures, as we will
discover soon, are generally curved lines. Thus, the formation of influence lines for
indeterminate structures requires calculation of many more ordinates than necessary
in the case of determinate structures.
Although any of the methods of analysis of indeterminate structures presented
in Part III can be used for computing the ordinates of influence lines, we will use
the method of consistent deformations discussed in the previous chapter to learn the
method of drawing the influence lines for indeterminate structures.

17.2  INFLUENCE LINES FOR BEAMS

To begin our discussion on influence line diagrams, let us consider the continuous
beam with the support conditions as shown in Figure 17.1 (a).
As per Figure 17.1, the unit load is placed (for an instant) at a distance x from the
left support. Now our target is to draw the influence line for the support reaction at
B (i.e., for By). To be able to do that, we need to express By in terms of variable x.
Since the beam is indeterminate to the first degree, we need to remove one redundant
support reaction from the indeterminate beam to make the primary structure. So,
the roller support from B has been removed to convert it to primary structure. Once
the primary structure is formed, we apply separately the unit load at any arbitrary
DOI: 10.1201/9781003081227-20 285
286 Introduction to Structural Analysis

FIGURE 17.1  Finding influence line for reaction of an indeterminate beam.

Influence Lines for Statically Indeterminate Structures 287

point X at a distance x from the left support and redundant reaction By as shown in
Figure 17.1 (b) and (c) to form the compatibility equation. Since net displacement at
support B will be zero, hence, we get:

f BB ( By ) = 0
f BX + 

From which we get:

f BX  
By =   −
f BB

In which, flexibility coefficient, f BX gives the deflection at point B when a unit load is
placed at a distance x from left support. And  f BB represents the deflection at point B
due to a unit redundant force at B. We can apply the above equation easily to get the
influence line for the support reaction at B. To be able to do that, we can take help
from the Maxwell’s law of reciprocal deflection introduced in the earlier chapter.
According to this principle, the deflection at B due to a unit load at a distance x is the
same as deflection at x due to a unit load at point B. Thus, in summary, we have the
following relationship:

f BX = f XB

With this, the above equation changes to:

f XB  
By = −

f BB

This is the equation of influence line for reaction at B. Note that the deflections of
f XB and 
f BB are considered positive when in the upward direction. To understand the
concept, let us consider the following example problem.

Example 17.1:  Draw the influence lines for the reactions at the supports
A, B, and C of the indeterminate beam, and shear force and bending
moment at point E as shown in Figure 17.2.

SOLUTION:  This beam is indeterminate to the first degree. So first, the support at
point B is considered as redundant. Now, if we remove this redundant, we will get
the primary beam as shown in Figure 17.2 (b).
Influence line for redundant By
The value of redundant By for an arbitrary position X of the unit load can be
determined by solving the compatibility equation:

fBX + f ( )
BB By = 0 (17.1)

fBX
From which, By = − (17.2)
f
BB
288 Introduction to Structural Analysis

FIGURE 17.2  Example problem on finding the influence line diagram for reactions, shear,
and bending moment.

According to Maxwell’s law, fBX = fXB , we place the unit load at B on the primary
beam as shown in Figure 17.3 (a) and compute the deflections A through C by
using conjugate beam method. The conjugate beam is shown in Figure 17.3 (b),
from which we obtain the following:

fBA = fAB = 0

 9   1  1.5   3   24.75
fBD = fDB = −   × 3 −   × 3 ×  × =−
 EI   2  EI   3   EI

 9   1  3   6  36
fBB = −   × 6 −   × 6 ×   ×    = −
  EI   2   EI   3   EI
Influence Lines for Statically Indeterminate Structures 289

FIGURE 17.3  Conjugate beam for unit load at B.

 9   1  1.5   3   24.75
fBE = fEB = −   × 3 −   × 3 ×  × =−
 EI   2  EI   3   EI

fBC = fCB = 0

The negative signs indicate that these directions occur in the downward direc-
tions. The flexibility coefficient f
BB in equation (17.1) denotes the upward (positive)
deflection of primary beam at B due to the unit value of the redundant By as shown
in Figure 17.2 (c), whereas the deflection fBB represents the downward (negative)
deflection at B due to the external unit load at B.
36
Thus, we can write, f BB = − fBB = +
EI
The ordinates for the influence line for By can now be computed by applying
equation (17.2) successively for each position of unit load. For example, when the
unit load is located at A and C, the value of By is given by,

fBA f
By = − = − BA = 0

fBB f
BB

When the unit load is located at D, and E, the value of By is given by,

fBD f 24.75
By = − = − BE =   = 0.6875
f
BB

fBB 36
290 Introduction to Structural Analysis

FIGURE 17.4  Finding coordinate of influence line diagram for Ay , when the unit load is at
3 m from support A.

When the unit load is located at B, the value of By is given by,

fBB 36
By = − = =1
f
BB 36

Now that the By is known, the values of the ordinate of the influence lines for other
reactions can be obtained using equation of statics.
Influence line for Ay :
Let us consider the external unit load is located at D, which is 3 m from sup-
port A. In this position, the influence line ordinate for By is 0.6875. Now, from
Figure 17.4, using moment equilibrium about point C, we can easily find out the
ordinate of influence line for Ay as below,

+ ∑M C =0

− Ay (12) + 1( 9) − 0.6875 (6) = 0

or, Ay = 0.41
Now, we can vary the position of the unit load along the span of the beam and
can get the remaining ordinates of influence line diagram.
Similarly, we can find the influence line diagram for Cy as well using equation
of statics. The influence line diagrams for Ay , By , and  Cy are shown in Figure 17.5.
Influence lines for SE and ME
The ordinates of the influence lines for the shear and bending moment at E
can now be evaluated by placing the unit load successively at points A through
C on the indeterminate beam and by using the corresponding values of the
reactions computed previously. For example, as shown in Figure 17.6 (a), when
the unit load is located at point D, the values of the reactions are Ay = 0.41;
By = 0.69;  and  Cy = −0.095.
By considering the equilibrium of the free body of the portion of the beam to
the left of E, we obtain,

SE = 0.41 − 1+ 0.69 = 0.1

ME = 0.41( 9) − 1(6) + 0.69 ( 3) = −0.24

The values of remaining ordinates of the influence lines are computed in a similar
manner. Figure 17.6 shows the influence line diagrams for SE and ME .
Influence Lines for Statically Indeterminate Structures 291

FIGURE 17.5  Influence line diagrams for reaction forces.

FIGURE 17.6  Influence line diagrams for SE , and M E .

292 Introduction to Structural Analysis

17.3  INFLUENCE LINES FOR TRUSSES

Influence lines for trusses can be drawn with the help of flexibility coefficients as
that has been considered during the analysis of beams. To understand the concept,
let us proceed with Example 17.2.

Example 17.2:  Draw the influence line diagrams for member forces of
the truss member BC and CE as shown in Figure 17.7.

SOLUTION:  This truss is internally statically indeterminate to the degree one.

Now, let us consider the redundant member as the member CE. To determine
the influence line diagram for FCE , we place a unit load successively to the joints
B and C. For each position of the unit load, we apply the method of consistent
deformation to determine the value of member force CE, i.e., FCE . First, we remove
the member CE entirely from the original truss to convert it to determinate primary
truss. Under this situation, the primary member with unit load at joint B will look
something like the diagram as shown in Figure 17.8 (a).
By taking the overall force and moment equilibrium of the determinate
truss shown in Figure 17.8, we can determine the support reactions that are as
follows:

2
RA =
3

1
RB =
3

The above force values will be just interchanged when the unit load is shifted
from joint B to joint C due to symmetry. Under this situation, the force in the
member of the truss also needs to be calculated. Since the redundant has been
removed, it has become a statically determinate structure and hence, the mem-
ber forces can be calculated by applying any standard procedure we have
learned earlier. The member forces due to unit load at B and C are given in
Figure 17.8 (a) and (b), respectively for quick reference. Also, internal forces on
the member when a unit tensile load applied on the redundant member CE are
also provided in Figure 17.8 (c). Students need to check the correctness of these
values on their own.

FIGURE 17.7  Example problem for truss member force influence line.
Influence Lines for Statically Indeterminate Structures 293

FIGURE 17.8  Primary truss with unit load at B, C, and along the redundant member CE.

When the unit load is located at B, the compatibility equation can be written as

fCE ,B + fCE ,CE  FCE = 0

where fCE ,B denotes the relative displacement between joints C and E of the pri-
mary truss due to unit load at B and fCE ,CE relative displacement between the
same nodes due to a unit value of redundant force FCE . Once the compatibility
formed, we can develop the following table for ease of analysis using compat-
ibility method.

uB uCE L uC uCE L u 2CE L

Member L(mm) A( mm 2 ) uB uC uCE
A A A
AB 240 6 −0.888 −0.445 0 0 0 0
BC 240 6 −0.445 −0.888 −0.8 15.806 28.416 25.6
CD 240 6 −0.445 −0.888 0 0 0 0
EF 240 6 0.888 −0.445 −0.79 −28.061 −14.062 24.964
BE 180 4 −0.668 −0.334 −0.61 18.337 9.168 16.745
CF 180 4 0 −1.0 −0.61 0 27.45 16.745
AE 300 6 1.12 0.556 0 0 0 0
BF 300 4 −0.556 0.556 1 −41.7 41.7 75
CE 300 4 0 0 1 0 0 75
DF 300 6 0.556 1.12 0 0 0 0
∑ −35.618 120.796 234.054
294 Introduction to Structural Analysis

FIGURE 17.9  Influence line diagram for member force FCE .

Now, using virtual work method, we obtain:

fCE ,B =
1
E
  ∑ u uA L = − 35.618
B CE
E

∑ u A L = 234.054
2
1 CE
fCE ,CE =  
E E

By substituting these values in the above compatibility equation, we get:

fCE ,B
FCE =   − = 0.152 (T )
fCE ,CE

Similarly, when the unit load is placed at C., the compatibility equation may be
written as:
fCE ,C + fCE ,CEFCE = 0

Using the above table, we have:

fCE ,C =
1
E
  ∑ u uA L = 120.796
C CE
E

Substituting this value in the above equation we get:

fCE ,C 120.796
FCE = − =  − =   − 0.516 (C )
fCE ,CE 234.054

Thus, we determine the ordinate at B and C, respectively, from the above results.
The influence line diagram for FCE is shown in Figure 17.9.
Following this procedure, the influence lines for all other member forces can
be drawn.

17.4 QUALITATIVE INFLUENCE LINES BY THE MÜLLER-BRESLAU’S

PRINCIPLE AND INFLUENCE LINE FOR FRAMES
In many practical situations, such as when designing continuous beams or building
frames acted upon by uniformly distributed live loads, it is sufficient to draw only
the qualitative influence lines to predict where to place the live loads to produce the
maximum effect of the response functions. As we have already seen in statically
determinate structures, the Müller-Breslau’s principle offers a convenient method of
forming qualitative influence lines for indeterminate structures.
Influence Lines for Statically Indeterminate Structures 295

As we can remember from our determinate structural analysis, this principle

states that the influence line for a force (or moment) is given by the deflected shape
of the released structure obtained by removing the restraint corresponding to the
response function from the original structure and giving the released structure a
unit displacement (or rotation) at the location and in the direction of the response
function, so that only the response function and the unit load perform external work.
The method of drawing qualitative influence lines for indeterminate structures is
the same as that of determinate structures discussed in Chapter 11 of this book. The
procedure needs the following steps to be adopted for analysis:

1. Removing from the original structure the restraint corresponding to the

response function to obtain the released structure.
2. Imposing a small displacement (or rotation) to the released structure at the
location and in the positive direction of the response function.
3. Drawing a deflected shape of the released structure in line with its support
and continuity parameters.

The influence lines for indeterminate structures are generally found to be curved
lines. Once a qualitative influence line for force or bending moment has been con-
structed, it can be used to decide where to place the live loads to maximize the value
of the forces and/or moments to cause severity. As discussed in earlier sections, the
value of shear force or bending moment due to a uniformly distributed live load is
maximum positive (or negative) when the load is located in those portions of the
structure where the ordinates of the response function influence line are positive (or
negative). Because the influence line ordinates tend to diminish rapidly with distance
from the point of application of the unit load, live loads placed more than three span
lengths away from the location of the unit load generally have a negligible effect on
the value of the overall reaction or bending moments. With a known live load pattern,
an approximate analysis of the structure can be performed to determine the maxi-
mum value of the force or moments that we desire and are required for drawing the
influence lines. We shall follow the Example 17.3 that will demonstrate the proce-
dure for qualitative influence line construction using the abovementioned procedure.

Example 17.3:  Draw qualitative influence lines for the vertical reactions
at supports A and B for the continuous beam shown in Figure 17.10.
Also, show the arrangements of a uniformly distributed downward live
load w to cause the maximum positive reactions at supports A and B,
the maximum negative bending moment at B.

SOLUTION:  Influence line for RA

To obtain influence line for RA , following the Muller-Breslau principle, we first
remove the support at A (thus, we remove the redundant reaction at A), and then
apply a small displacement in the direction of the redundant force RA . In this situ-
ation, please keep in mind that the other supports to be kept as it is present in the
original structure. Thus, due to this small displacement at A, the overall deflected
shape of the beam will be the influence line for reaction at A. Pictorially, the
qualitative influence line for RA will be something like as shown in Figure 17.11 (a).
296 Introduction to Structural Analysis

FIGURE 17.10  Example problem for qualitative influence line diagram.

Now to maximize the effect due to live load w, we need to place the same over
the span AB of the beam where the ordinate of the influence line diagram is
positive.
Influence line for RB
Following the same procedure as that for RA , we can form the influence line for
RB by drawing the deflected shape of the beam, maintaining the constraints and
direction of redundant support reaction. The same has been drawn in Figure 17.11 (b)
for reference.
As can be seen for RB influence line, the total ordinate is positive near the span
and goes near the supports. Hence, for maximum response against the live load,
the load should be placed on the beam A to C over the whole span.
Influence line for MB
To determine the influence line for bending moment at B, we need to remove
the support at B and then replace the existing roller support by a hinge and
apply small angular displacement at the same location as shown in Figure 17.12.
Due  to this applied rotational displacement, the deflected shape of the entire
beam has also been drawn, and that is the influence line diagram for the bending
moment at B.
Thus, to cause maximum negative bending moment, we place the live load in
the span AB and BC, where the ordinate of the influence line diagram is negative.

FIGURE 17.11  Qualitative influence line for (a) RA and (b) RB diagrams as per example
problem.
Influence Lines for Statically Indeterminate Structures 297

FIGURE 17.12  Qualitative influence line diagrams for bending moment at B.

Once we have dealt with the continuous beams, let us focus on the formation of influ-
ence lines for statically determinate frames influence lines.
The Müller-Breslau principle provides a quick method for generating the gen-
eral shape of the influence line for building frames. Once the influence line shape
is determined, one can immediately decide on the location of the live loads to
form the greatest influence of the functions (reaction, shear, or moment) in the
frame.
The shape of the influence line for the positive moment at the center I of girder
FG of the building frame in Figure 17.13 is shown by the broken lines. Thus, uniform
loads would be placed only on girders AB, CD, and FG in order to create the largest
negative moment at I. With the frame in addition to imposed live load at the specified
locations in this manner, indeterminate analysis of the frame can then be applied to
calculate the critical moment at I.
Similarly, for the maximum positive moment at midspan of BC, we can place the
live load on EF, GH and BC to produce the maximum adverse effect due to live load
as that we have told just before for the maximum negative moment at the midspan
of FG.
So from the example, we find that the Müller-Breslau’s principle indeed provides
a very handy tool to determine the qualitative influence line diagram of the support
reactions and bending moment as required. Thus, by applying this principle, we can
easily determine the location and placement of live loads on any number of spans of
a continuous beam to obtain the maximum response.

FIGURE 17.13  Building frame qualitative influence lines.

298 Introduction to Structural Analysis

17.5 ALTERNATE APPROACH FOR FINDING INFLUENCE

LINE DIAGRAMS FOR INDETERMINATE BEAMS
This approach for finding the influence lines of indeterminate beams is based on a
mathematical model derived from the fundamental use of the flexibility method. The
mathematical model is based on describing the forces and deformations of the beam
as mathematical functions related by the consecutive integration process. This new
approach was reported by Dr. Moujalli Hourani [1] of Manhattan College in the pro-
ceeding of the American Society of Engineering Education Annual Conference and
Exposition in 2002. It was reported that using this new technique, students showed a
major improvement in their capabilities to solve problems of influence lines for indeter-
minate beams. They were capable of developing their computer programs using Excel/
Quattro and Mapple/MathCAD to solve problems of influence lines for multi-span
beams with various boundary conditions. This approach is not based on new theories
or principles but rather on a new methodology to solve a typical structural problem.
Along with the flexibility method, a mathematical model was used as summa-
rized below to evaluate the influence lines:

1. Determine the degree of static indeterminacy.

2. Choose the unknowns/redundants, name them X1 , X 2 ...........,  X n . The redun-
dant can be external reactions, internal forces or both. Make sure that the
structure remains stable.
3. Remove the dedundants, i.e., set the unknowns equal to zero. The structure
is now statically determinate, and it is called the released structure.
4. Remove all the loads and apply a unit load corresponding to X1, at the loca-
tion of X1.
5. Repeat the process for all the X’s.
6. Find the displacements corresponding to the unknowns in the released
structure.
7. Find the flexibility coefficients.
8. Use the principle of superposition and apply the compatibility conditions at
the locations of all the redundants.
9. Describe the load as a function YL in the X-Y coordinate system, as shown
in Figure 17.14. Since the influence line is based on applying a unit concen-
trated load, then, YL = 0 .
10. The shear is equal to YV = ∫ YL dx = C1
11. The bending moment YM = ∫ YV dx = C1 x + C2
12. The slope of the elastic curve is described by,

 1  Y dx =  1  ×  C x  + C x + C , assuming constant EI .
2
YS =
∫   M
EI
   1 
EI  2
2 3

13. The transverse deflection of the beam is described by,

1   x3  x2
YD =   ×  C1  + C2 + C3 x + C 4
 EI   6 2
Influence Lines for Statically Indeterminate Structures 299

FIGURE 17.14  Description of load function, YL .

The constants of integration can be determined by applying the geometric (slope,

deflection, and compatibility conditions) and the loading (shear, moment, and joint
equilibrium) boundary conditions. We will understand further about this concept
with the help of some example problems.

Example 17.4:  Draw the influence line for the vertical reaction at B, RB ,
for the beam shown in Figure 17.15. The beam is statically indeterminate
to the first degree. Choose as a redundant the vertical reaction at B,
By = X1.

SOLUTION:  Let us remove the redundant at B to get the primary structure. Now
the beam has become statically determinate as shown in Figure 17.15 (b). Now,
remove the applied load and apply a unit vertical load at B. This virtual beam
is shown in Figure 17.15 (c). Making use of the principle of superposition, the
relationship between these three figures is summarized by:

Figure 17.15 ( a )   =  Figure 17.15 (b )   +  Figure 17.15 ( c )   ×   X1

Applying the compatibility condition for the vertical deflection at B, gives:

−δ BD + F11× X1 = 0

Using Maxwell’s law of reciprocal deflection, we can set δ BD = δ DB, and the
reaction RB can be found from the above equation to be,

RB = X1 = δ DB /F11

As it can be seen from the above equation, the only figure needed is Figure 17.15 (c).
In the above equations, X1 = the vertical reaction at B due to a unit vertical load
at any point D along the beam; δ DB = the vertical displacement at D due to a unit
vertical load at B; F11 = the vertical displacement at B due to a unit vertical load at B.
The moment diagram of the beam in Figure 17.16 is drawn below:
Due to the discontinuity in the moment diagram, the equations for the two seg-
ments AB and BC are needed to describe the moment functions, x1 − y1 from A to B,
and x2 − y 2 from C to B.
300 Introduction to Structural Analysis

FIGURE 17.15  Finding influence line diagram of vertical reaction for two span indetermi-
nate beam.

From A to B (origin at A)

(YM )1 = − 
b
x
L

1   b  x 2 
(YS )1 =   × −  × + C3 
EI   L  2 

1   b  x 3 
(YD )1 =   × −  × + C3 x + C 4 
EI   L  6 

FIGURE 17.16  Moment diagram for the virtual beam.

Influence Lines for Statically Indeterminate Structures 301

From C to B (origin at C)

(YM )2 = 
a
x
L

1   a  x 2 
(YS )2 =  ×  × + K3 
EI   L  2 

1   a  x 3 
(YD )2 =  ×  × + K3 x + K 4 
EI   L  6 

Applying the boundary conditions to find the constants of integration.

At x1 = 0, (YD )1 = 0 , which implies, C4 = 0. At x2 = 0, (YD ) 2 = 0 , which implies,
K4 = 0 .
The compatibility condition for the slope at B, can be written as,
YS1at   x1 = a = YS 2at   x 2 = − b,, which gives,

a2 b2
−b + C3 = a + K3
2L 2L

Simplifying the above equation yields,

 ab  ab
C3 − K 3 =   × ( a + b ) = (17.3)
 2L  2

The compatibility condition for the vertical deflection at B, can be written as,
(YD )1  = (YD ) 2  , which gives,
at   x1 = a at   x2 =− b

− b × a3 ab3
+ C3a = − − K3b
6L 6L

Simplifying the above equation yields,

 ab 
C3a + K 3 b =   × ( a − b ) (17.4)
 6 

Solving equations (17.1) and (17.2) for C3 and K3, yield

 ab 
C3 =   × ( 2b + a )
 6L 

 ab 
K 3 = −   × ( b + 2a )
 6L 
302 Introduction to Structural Analysis

Having solved for all the constants of integration, the equations describing the
vertical displacement at any point along the beam due to unit vertical load at B
can be written as,

(YD )1 = 
1 
 ×  − bx + ab ( a + 2b ) x  for, 0 ≤ x ≤ a
3

6EIL  

(YD )2 = 
1 
 ×  ax − ab ( 2a + b ) x  for, − b ≤ x ≤ 0
3

6EIL  

Here, YD represents the vertical deflection at any point D due to a unit vertical load
at B, δ DB.
F11 = (YD )1 at x1 = a. So, we get, F11 = ( 6EIL
2
) a2b 2, and the vertical reaction at B, By
can be written as,

 1   − bx + xab ( a + 2b ) 
3

X1 = (YD )1 /F11 =   × for 0 ≤ x ≤ a

 6EIL   2  2 2
 
 6EIL  × a b 
 

TABLE 17.1
Summary of the Reaction Forces for the Given
Indeterminate Beam
Origin At Valid Range Equations of Reactions

A 0 ≤ x1 ≤ a Ay =
(
x 3 − x 3a 2 + 2ab + 2a 2 L )
2a 2 b

By =
(
− x 3 + x a 2 + 2ab )
2abL

Cy =
(
x x 2 − a2 )
2a 2 b
− ( x − L ) ( a /b ) + ab ( x − L )
3
a≤x≤L Ay =
2a 2 L

( x − L )( x 2 − 2 Lx + a 2 )
By =
2b 2 a
− x 3 + 3 Lx 2 + ax ( a − 4 L ) + a 2 L
Cy =
2b 2 L
− x 3 ( a /b ) + abx
C − b ≤ x2 ≤ 0 Ay =
2a 2 L

By =
(
x 3 − x b 2 + 2ab )
2b 2 a

Cy = −
(
x 3 − x 2ab + 3b 2 − 2b 2 L )
2a 2 L
Influence Lines for Statically Indeterminate Structures 303

Simplifying we get,

X1 = By =
(
− x 3 + x a2 + 2ab ) for 0 ≤ x ≤ a
2a2b

X1 = By =
(
x 3 + x b 2 + 2ab ) for − b ≤ x ≤ 0
2b 2a

Knowing, By , the other two reactions can be found by static equilibrium. The
equations of the reaction forces are summarized in Table 17.1.
These equations are used to determine the internal resisting forces at any point
in the beam. Readers are encouraged to plot the influence line diagrams of various
reaction forces using the above equations making one computer program.
 18 Slope Deflection Method

18.1 INTRODUCTION
In this chapter, we will discuss the whole idea for analyzing structures using the
slope deflection method. Once these concepts are presented and detailed steps
are developed, we will form the general slope deflection equations and then use
them to analyze statically indeterminate beams and frames. It is to be understood
that the slope deflection method is a displacement method or stiffness method,
unlike the force method or the method of consistent deformation discussed earlier.
Here the unknown displacements are found first, solving the structure’s equilibrium
equations.

18.2 SLOPE DEFLECTION EQUATIONS AND ANALYSIS

OF CONTINUOUS BEAMS
The method of consistent deformation studied in Chapter 16 is called a force method
of analysis because it demands developing equations that relate to the unknown
forces or moments in a structure. Unfortunately, its use is limited to systems that are
not highly indeterminate. This is because tedious work is required to set up the com-
patibility equations, and moreover, each equation written involves all the unknown
forces, making it difficult to solve the resulting set of equations unless computer
software is used. In contrast, the slope-deflection method, which is a displacement
method of analysis, is not as difficult to handle. As we shall understand soon, it
requires less time to write the necessary equations for the solution of a problem and
solve these equations for the unknown displacements or deformations and associated
internal forces and moments. Also, this process can be easily programmed on a com-
puter and applied to analyze a wide range of indeterminate structures.
The slope-deflection method was originally invented by Heinrich Manderla and
Otto Mohr for the purpose of studying secondary stresses in trusses. Later, in 1915,
G. A. Maney developed the modern version of this technique and applied this prin-
ciple to the analysis of indeterminate beams as well as framed structures.
The slope-deflection method is so named since it deals with the unknown slopes
and deflections to the applied load on a structure. To develop the general form of
the slope-deflection equations, we will consider the typical span AB of a continuous
prismatic beam as shown in Figure 18.1, which is subjected to arbitrary loading and
has a constant EI. We like to relate the beam’s internal end moments with its degrees
of freedom, namely, its angular and linear displacements, which could be caused by
a relative settlement between the supports, temperature change, due to the presence
of external force, etc. or due to any combination of the effects. Since we will be
developing a formula, the moments and angular displacements, cord rotations will
be considered positive when they act clockwise on the span, as shown in Figure 18.1.
Moreover, the linear displacement ∆ is considered positive, as shown in the figure,
DOI: 10.1201/9781003081227-21 305
306 Introduction to Structural Analysis

FIGURE 18.1  Two spans of a continuous beam with general loading with deflection diagrams.

since this displacement causes the centerline of the span and the span’s cord angle to
rotate clockwise. In Figure 18.1, the straight line joining the two nodes A to deflected
node B is the cord line of the beam, and the dotted line is the deflection curve of the
beam due to the external loading.
The slope-deflection equations can be formulated by using the principle of super-
position by taking separately the moments developed at each support due to each of
the displacements θ A, θ B, ∆ and then the external loads.
For angular displacement θ A only at A, let us consider the following beam with B
end fixed as shown in Figure 18.2.
Now, to determine the moment M AB to cause this angular displacement, we seek
help from the conjugate beam method. Since the deflection at the two ends of the real
beam is zero, corresponding moments at the end of the conjugate beam will also be
zero. As θ A is clockwise, the shear at end A′ of the conjugate beam acts downward.
Also, support conditions in real beams need to be changed while forming the con-
jugate beam. For students, it is advisable to stop here and go through Chapter 9 on
conjugate beam analysis for a quick recap. So, applying the appropriate loading and
support conditions, the bending moments at the end of the conjugate beam will be:

+ ∑M A′
1 M

2 EI 
L 1 M
= 0;         AB  L −    BA  L

3 2 EI 
2L
3
=0

+ ∑M B′
1 M

2 EI 
L 1 M
= 0;         BA  L −  AB  L

3 2 EI 
2L
3
+ θAL = 0

Solving these two simultaneous equations, we get,

4 EI
M AB = θA
L
2 EI
M BA = θA
L
Slope Deflection Method 307

FIGURE 18.2  Real beam and conjugate beam for finding angular displacement at A.

Similarly, for separate rotational displacement θ B, we can form the same type of
beam with A end fixed. Then applying the conjugate beam method, we can deter-
mine the following relationships between angular displacement and moment at the
supports:

4 EI
M BA = θB
L

2 EI
M AB = θB
L

Once the relationships are formed for angular displacements, we can form a similar
relationship for the linear displacement ∆ that occurred due to the settlement of the
support B as per Figure 18.1. For this analysis, we have both ends fixed beam with
settlement occurred at the fixed-end B as shown in Figure 18.3.
So, the cord of the beam rotates clockwise, but both ends do not rotate. This
results in equal but opposite support moments and shear force at both beams’ end,
as shown in Figure 18.3. To form the relationship, we again seek the help of the
308 Introduction to Structural Analysis

FIGURE 18.3  Fixed-end beam with support settlement at B.

conjugate beam theory, and in this case, both ends of the conjugate beam will be
the free end as per requirement. However, the displacement at B of the real beam
and the moment at the end of the conjugate beam at the same point should be the
same as ∆.

1 M  2  1 1 
 2   EI ( L )    3 L   −  2 ( L )  3 L   − ∆ = 0
   

6 EI
M AB = M BA = M = − ∆
L2

By our sign convention, this moment is indeed negative as for equilibrium, and it acts
in the anticlockwise direction.
It is to be noted that linear and angular displacements at the nodes occur due to the
loading in the span of the beam. Hence, we need to develop a method to transfer the
loading on the beam into equivalent force and moment at the nodes to be included in
the slope deflection equation. This is simply done by finding reaction moments that
Slope Deflection Method 309

each load develops at the nodes. So, the final moment that helps us transfer the span
loading of beams at the supports is the fixed-end moment at each support’s points or
nodes for the given class of loading on the span.
So, in addition to the above, we have the following two moments that need to be
included in the slope deflection equation:

M AB = ( FEM ) AB

M BA = ( FEM )BA

Let us find out the fixed-end moments for the beam shown in Figure 18.3 under uni-
formly distributed load (udl) throughout its span. The conjugate beam is shown in
Figure 18.3 (b). As both ends are fixed, the slope will be zero for both ends. So, from
the conjugate beam, we can write,

↑+ ∑F = 0
y

or,
2 wL2 1 M 
L× − 2   L
3 8 EI 
 2 EI  

Therefore,
wL2
M=
12

This is the amount of fixed end moment if udl is imposed throughout the span.

FIGURE 18.4  Fixed-end moment for a uniformly distributed load.

310 Introduction to Structural Analysis

Figures 18.5 and 18.6 show the fixed end moments for various loading and bound-
ary conditions of the beams.
Now we can combine all the terms, i.e., end moments due to each displacement
and loading, and can get the following slope deflection equation for the two nodes of
a continuous beam:

M AB =
2 EI  2θ + θ − 3 ∆  + ( FEM )
 L 
A B AB
L

M BA =
2 EI  2θ + θ − 3 ∆  + ( FEM )
 L 
B A BA
L
These equations are very neat and very elegant in solving. For each span of a continu-
ous beam, we write these equations, and then for internal nodes, we add the moments
at the two sides of the same node, which should add up to zero to maintain the
moment equilibrium at those interior nodes. From these, we solve the simultaneous

FIGURE 18.5  Fixed-end moments for various loading conditions (fixed-fixed condition).
Slope Deflection Method 311

FIGURE 18.6  Fixed-end moments for various loading conditions (fixed-hinged condition).

equations to determine the unknown parameters θ A and θ B. These necessary steps

will be clear once we will go through Example 18.1.

Example 18.1: Draw the Shear and Bending Moment diagram of the

continuous beam shown in Figure 18.7.

SOLUTION:  We start the analysis by calculating the fixed-end moments at each

node due to each span of the beam.
wl 2
(FEM) AB = − = −96 kNm
12
wl 2
(FEM)BA = = 96 kNm
12
3PL
(FEM)BC = − = −18 kNm
16
312 Introduction to Structural Analysis

FIGURE 18.7  Example problem.

The last node C of the continuous beam is hinged and (FEM)CB = 0. Also, since
node A is fixed, θ A = 0. As well as since the far-end C is pinned, the final moment
will also be zero at this location, i.e., MCB = 0.
So, with these values in hand, let us write the slope deflection equation as
follows:
2EI
M AB =
24
[ 2 × 0 + θ B − 3 × 0] − 96
or,

M AB = 0.0833EIθ B − 96

2EI
MBA =
L
[ 2θ B + 0 − 3 × 0 ] + 96
or,

MBA = 0.166EIθ B + 96

Now, for the span BC, we have,

2EI  ∆
MBC =
L  2θ B + θ C − 3 L  + (FEM)BC

or,

2EI
MBC =
L
[ 2θ B + θC − 3 × 0] − 18
or,

MBC = 0.50EIθ B + 0.25EIθ C − 18

Also, for the far-end C, we have:

2EI  ∆
MCB =
L  2θ C + θ B − 3 L  + (FEM)CB

or,
2EI
0=
L
[ 2θC + θ B − 3 × 0] + 0
Slope Deflection Method 313

From the above equation, we have the following relationship:

2θ C = −θ B

or,

θB
θC = −
2
So, substituting these values in other equations for θ B we get,

θB
MBC = 0.5EIθ B − 0.25EI × − 18
2
or,

MBC = 0.375EIθ B − 18

Now, since there are two moments at two sides of the same node B, these should
sum up to zero to maintain the moment equilibrium equation.
Thus, we have, MBA + MBC = 0 .
or,

0.166EIθ B + 96 + 0.375EIθ B − 18 = 0

or,

144.177
θB = −
EI

Substituting these values in the above three equations, we get,

144.177
M AB = − 0.0833EI − 96 =   −108 kN
EI
144.177
MBA = −0.166EI + 96 =  72.06 kN
EI
144.177
MBC = − 0.375EI − 18 = −72.06 kN
EI

Now, to determine the shear force, we need to draw the free-body diagrams as
shown in Figure 18.8.
So, from the above-left side free body diagram, we have,

VA × 24 + 72.06 − 108 − 48 × 12 = 0

or,

VA = 24.41 kN

VBA × 24 − 72.06 + 108 − 48 × 12 = 0

314 Introduction to Structural Analysis

FIGURE 18.8  Free body diagram of the beam span with end moments.

or,

VBA =   −22.5 kN

VBC × 8 − 12 × 4 − 72.06 = 0

VBC = 15.00 kN

Also,

VC + 15.00 − 12 = 0

or,

Vc = −3 kN

With the above values, the shear and bending moment diagram of the complete
beam can be drawn as shown in Figure 18.9.
Bending moment at the mid-span of span AB (M AB/ 2) can be calculated by tak-
ing a section at that point and taking a moment of all forces about the same of the
left section. So, we have,

M AB / 2 + 12 × 2 × 6 − 24.4 × 12 + 108 = 0

or,

M AB / 2 = 40.08 kNm

Similarly, the moment at mid-span of BC is:

MBC / 2 + 15 × 4 − 72.06 = 0

MBC / 2 =   −12.06 kNm

All above values are placed in the bending and shear force diagrams for ease of
understanding.
Slope Deflection Method 315

FIGURE 18.9  Shear force and bending moment diagram of the example problem.

18.3  MEMBERS WITH FAR END HINGED

The above formulations of slope deflection are derived based on the conditions where
the members are considered to be rigidly connected to joints at both sides so that the
member end rotations are equal to the rotations of the adjacent joints. But it may so
happen that the far end of a member is joined with a hinged connection. In those
cases, the end moment will become zero at those far ends, and the slope deflection
equations can also be modified accordingly. Let us assume that the end B is hinged
on a member AB. Then the slope deflection equations can be written as follows:

M AB =
2 EI  2θ + θ − 3 ∆  + ( FEM )
L  A B
L  AB

M BA =
2 EI  2θ + θ − 3 ∆  + ( FEM ) = 0
L  B A
L  BA
316 Introduction to Structural Analysis

Solving M BA = 0, we get that,

θA 3 ∆ L
θB = − − − ( FEM )BA
2 2 L 4 EI
Now, we can eliminate θ B and rewrite the slope deflection equation as:

3EI  ∆  FEM BA 
M AB =  θ A −  +  FEM AB −
L  L 2 
M BA = 0

Let us go through the following example to understand the concept.

Example 18.2: Figure 18.10 shows a continuous beam ABCD. Find

the moments at A, B, C, and D if the end A rotates by 0.004 radians
in the clockwise direction and the support B sinks by 6 mm. Take
E = 200 kN/mm2 and I = 9 × 107 mm4.

( )
SOLUTION:  For the data of this problem, EI = 200 × 9 × 107 / 106 = 18000 kNm 2
Span AB

2E ( 2I )  3∆  4EI  3× 6 
M AB =  2θ A + θ B − =  2 × 0.004 + θ B −  = 63 + 18000θ B
L  L  4  4000 

2E ( 2I )  3∆  4EI  3× 6 
MBA =  2θ B + θ A −  =  2θ B + 0.004 −  = −9 + 36000θ B
L L 4 4000 

Here the clockwise movement of the cord is taken as positive.

Span BC

2E ( 4I )  3∆  8EI  3× 6 
MBC =  2θ B + θ C + =  2θ B + θ C +  = 40.5 + 36,000θ B + 18,000θ C
L  L  8  8000 

2E ( 4I )  3∆  8EI  3× 6 
MCB =  2θ C + θ B + =  2θ C + θ B +  = 40.5 + 18,000θ B + 36,000θ C
L  L  8  8000 

Here the anticlockwise movement of the cord is taken as negative.

FIGURE 18.10  Problem on a continuous beam with support settlement.

Slope Deflection Method 317

Span CD

3EI  ∆  FEMDC 
MCD =  θ C −  +  FEMCD − 
L  L 2 

or,

(θC − 0 ) +  0 −  = 18,000θC

3EI 0
MCD =
3 2

Equilibrium condition at B,

MBA + MBC = 0

or,

−9 + 36,000θ B + 40.5 + 36,000θ B + 18,000θ C = 0

or,

72,000θ B + 18,000θ C = −31.5  (18.1)

Equilibrium condition at C,

MCB + MCD = 0

or,

40.5 + 18,000θ B + 36,000θ C + 18,000θ C = 0

or,

18,000θB + 54,000θ C = −40.5. (18.2)

Solving equations (18.1) and (18.2), we get,

θ B = −2.73 × 10 −4

θ C = − 6.59 × 10 −4

Substituting for θ B and θ C we get,

(
M AB = 63 + 18,000θ B = 63 + 18,000 −2.73 × 10 −4 = 58.1 kN )
( )
MBA = −9 + 36,000θ B = −9 + 36,000 −2.73 × 10 −4 = −18.82 kN

MBC = 40.5 + 36,000θ B + 18,000θ C

( ) (
= 40.5 + 36,000 −2.73 × 10 −4 + 18,000 −6.59 × 10 −4 = 18.81 kN )
318 Introduction to Structural Analysis

FIGURE 18.11  Bending moment diagram of the example problem.

MCB = 40.5 + 18,000θ B + 36,000θ C

( ) ( )
= 40.5 + 18,000 −2.73 × 10 −4 + 36,000 −6.59 × 10 −4 = 11.86 kN

( )
MCD = 18,000θ C = 18,000 −6.59 × 10 −4 = −11.86 kN

MDC = 0

The bending moment diagram is shown in Figure 18.11.

18.4  ANALYSIS OF FRAMES WITHOUT ANY SIDESWAY

A frame with the proper restrained condition will not sidesway toward left or right.
Moreover, even an improperly restrained frame may not move if it is symmetric in
terms of geometry and loading. The number of independent joint translations, i.e.,
sidesway degrees of freedom for an arbitrary plane frame subjected to general copla-
nar loading, can be expressed as:

ss = 2 j −  2 ( f + h ) + r + m 

where j is the number of joints; f is the number of fixed supports; h is the number of
hinged supports; r is the number of roller supports; and m is the number of inexten-
sible members. In the coplanar system, the independent joint translation can happen
either horizontally or vertically, i.e., in the above expression, 2j depicts the total num-
ber of translational degrees of freedom available. In the subtracted items, the fixed
and hinged support restrain two translations per node. The roller support restrains
one translation. Each inextensible member connecting two joints prevents one joint
translation in its axial direction.
For analysis of frames without sidesway, we will do that by solving a problem
discussed in Example 18.3.
Slope Deflection Method 319

FIGURE 18.12  Frame without sway problem.

Example 18.3:  Analyze the frame shown in Figure 18.12 and draw the
bending moment diagram.

SOLUTION:  In this example, the frame consists of four joints (j = 4), three
members (m = 3), two fixed supports (f = 2). So, the sidesway degrees of freedom,
ss = 2 × 4 −  2 ( 2 + 0 ) + 0 + 3 = 1. From this, we can understand that the frame can
undergo one independent joint translation. But further, if we observe, we can see
this frame is loaded symmetrically with respect to its axis of symmetry. So, this
frame is treated without sidesway condition.
In the above frame, all joints are fixed joints. In this situation, we first calculate
the fixed-end moments of each beam segment as follows:

(FEM) AB = 0 =  (FEM)CD

(FEM)BA = 0 = (FEM)DC

5wl 2
(FEM)BC = − =   −80 kNm
96

5wl 2
(FEM)CB = =  80 kNm
96

Note that, θ A and θ D = 0, since these two ends are fixed supported. Also, since there
will be no sidesway, ∆ in general, will be 0.
So,

M AB =
2 EI  2θ + θ − 3 ∆  + ( FEM )
 L 
A B AB
L
320 Introduction to Structural Analysis

or,
2 EI
M AB =   [ 2 × 0 +  θ B − 3 × 0 ] + 0
12
or,

M AB = 0.167 EIθ B

Similarly,

M BA =
2 EI  2θ + θ − 3 ∆  + ( FEM )
 L 
B A BA
L
or,
2 EI
M BA =   [ 2θ B +  0 − 3 × 0 ] +  ( FEM )BA + 0
12
or,

M BA = 0.334 EIθ B

Then,
2 EI  ∆
M BC = 2θ B +  θC − 3  + ( FEM )BC
L  L
or,

M BC =  0.5EIθ B + 0.25EIθC − 80

Similarly,

MCB =
2 EI  2θ + θ − 3 ∆  + ( FEM )
 L 
C B CB
L
or,

M CB =  0.5EIθC + 0.25EIθ B + 80

2 EI  ∆
MCD =
L  2θC + θ D − 3 L  + ( FEM )CD
or,
2 EI
MCD = [ 2θC + 0 − 3 × 0 ] + 0
8

MCD = 0.334 EIθC

2 EI  ∆
M DC =
L  2θ D +  θC − 3 L  + ( FEM )DC
Slope Deflection Method 321

or,

2 EI
M DC = [ 2 × 0 + θC − 3 × 0 ] + 0
8
or,

M DC = 0.167 EIθC

Now, applying the local moment equilibrium equation at joints B and C, respectively,
we get,

M BA + M BC = 0

MCB + M CD = 0

0.834 EIθ B + 0.25EIθC = 80

0.834 EIθC + 0.25EIθ B = −80

Solving these two simultaneous equations, we get,

137.27
θB =
EI

137.27
θC = −
EI

which shows that the members BA and CD are deflected by similar angle opposite in
direction without any sway. The qualitative deflected shape of the structure without
sway is shown in Figure 18.13.

FIGURE 18.13  Example problem of frame qualitative deflected shape without sidesway.
322 Introduction to Structural Analysis

Since the main unknown parameters are calculated, substituting these values in
the moment equation, we get,

M AB = 0.167 EIθ B = 22.9 kNm

M BA = 0.334 EIθ B = 45.76 kNm

M BC =  0.5EIθ B + 0.25EIθC − 80 = −45.76 kNm

MCB =  0.5EIθC + 0.25EIθ B + 80 = 40.75 kNm

MCD = 0.334 EIθC = −22.9 kNm

Using these end moments, support reaction of the frame can be obtained by using
force and moment equilibrium equations alone member-wise. Since this is a simple
work, it has been left as an exercise for the readers.
The frame’s shear force and bending moment diagram will be something like that
shown in Figure 18.14.

FIGURE 18.14  Shear force and bending moment diagram of example problem frame.
Slope Deflection Method 323

FIGURE 18.15  Frame with sidesway deflected shape.

In the next section, we will discuss the analysis procedure for frames with
sidesway.

18.5  ANALYSIS OF FRAMES WITH SIDESWAY

A frame will move or shift sideways when it or the loading acting on it is non-
symmetric. To understand this effect, consider the frame shown in Figure 18.15.
Here, the loading Q produces unequal moments at the joints B and C. These unequal
moments try to move joint B to the right, whereas try to displace joint C to the left.
Since M BC is larger than MCB, the net effect is a sidesway or sideway displacement
by ∆ of both joints B and C to the right, as shown in Figure 18.15. When dealing
with a slope-deflection equation to each column of this frame, we must consider the
column rotation as unknown in the equation. So, an extra equilibrium equation must
be included for the solution.
The techniques to solve this type of problem will be explained with the help of
Example 18.4.

Example 18.4:  Determine the moments at each joint of the frame shown
in Figure 18.16. EI is constant.

SOLUTION:  Since the frame is nonsymmetrical in nature, applied load at end B

will cause the frame to displace by ∆ , which remains the same for node C also.
Hence, both the members, AB and CD, are rotated with respect to their support
point by an angle, say α AB = ∆ /12 and α DC = ∆ /18, respectively, in the clockwise
direction. So, from this relationship, we can deduce that,

18
α AB = α DC
12
324 Introduction to Structural Analysis

FIGURE 18.16  Frame with sidesway example problem.

So, we need to incorporate it in the slope deflection equation while analyzing the
same.

2EI   18 
M AB =
12  2 ( 0 ) + θ B − 3  12 α DC   + 0 = EI ( 0.166θ B − 0.75α DC )
 

2EI   18 
MBA =
12  2θ B + 0 − 3  12 α DC   + 0 = EI ( 0.333θ B − 0.75α DC )
 

2EI
MBC =  2θ B + θ C − 3 ( 0 )  + 0 = EI ( 0.267θ B + 0.133θ C )
15 

2EI
MCB =  2θ C + θ B − 3 ( 0 )  + 0 = EI ( 0.267θ C + 0.133θ B )
15 

2EI
MCD =  2θ C + 0 − 3 (α DC )  + 0 = EI ( 0.222θ C −  0.333α DC )
15 

2EI
MDC =  2 ( 0 ) + θ C − 3 (α DC )  + 0 = EI ( 0.111θ C −  0.333α DC )
15 

So, in the above six equations, we have nine unknown forces. Another two equa-
tions will be related to moment equilibrium at the joints B and C, namely,

M AB + MBC = 0 (18.3)

MCB + MCD = 0 (18.4)

Since horizontal displacement is along the positive x-axis, we can have the follow-
ing force equilibrium condition for support reactions:

40 − HA − HD = 0
Slope Deflection Method 325

FIGURE 18.17  Column free body diagram for example problem.

The base support reactions or column shears can be related to the internal nodal
moments as per Figure 18.17.
So, from the above free body, we get,

M AB + MBA
HA = −
12

MDC + MCD
HD = −
18

So, from the horizontal equilibrium equation, we get,

M AB + MBA MDC + MCD

40 + + = 0. (18.5)
12 18

Now, substituting the earlier calculated moment expressions into equations

(18.3)–(18.5) and on simplification, we get,

0.6θ B + 0.133θ C − 0.75α DC = 0

0.133θ B + 0.489θ C − 0.333α DC = 0

480
0.6θ B + 0.222θ C − 1.944α DC = −
EI

Solving simultaneously, we get from the above equations,

EIθ B = 438.81

EIθ C = 136.18

EIα DC = 375.26
326 Introduction to Structural Analysis

FIGURE 18.18  Shear force and bending moment diagram of nonsymmetrical example frame.

Now, substituting back these above values in the original equations we get,

M AB = −208 kNm

MBA = −135 kNm

MBC = 135 kNm

MCB = 94.8 kNm

MCD = −94.8 kNm

MDC = 110 kNm

Students should pay utmost concentration to the formation of free body diagrams
for columns, which in turn provides us additional equations to solve the problem.
This procedure of column analysis is very important, and it will be frequently used
in subsequent chapters that follow. The shear force and bending moment diagram
for this problem is shown in Figure 18.18.
 19 Moment Distribution
Method

19.1 INTRODUCTION
The moment distribution method is another displacement method of analysis that
is a very elegant and convenient approach to apply once few elastic constants have
been determined. This is an iterative method in which formulating the equations for
unknowns is not even required. In this chapter, we will first state some of the impor-
tant definitions and points useful for understanding this analysis method. Then we
shall apply this method to solve problems related to statically indeterminate beams
and frames.

19.2  GENERAL PRINCIPLES AND DEFINITIONS

Prof. Hardy Cross invented the method of analyzing beams and frames using moment
distribution in 1930, in an era when computer facilities were not available. That is
why this method is also called the Hardy Cross method. By the time this method
was published, it had attracted the immediate attention of scholars, engineers, and
scientific professionals all over the world. Soon it has been recognized as one of the
most notable discoveries in structural analysis during the 20th century.
As will be explained in detail, moment distribution is a method of repetitive approx-
imations that may be carried out to any desired degree of accuracy. The method, at
first, assumes that each joint of a structure is fixed. Then, by unfixing and fixing each
joint in succession, the internal moments at the joints are ‘distributed’ and balanced
until the joints have rotated to their final or nearly ultimate positions. Readers will
find that this method of calculation is both repetitive and simple to apply. Before we
jump into the method directly, we will develop some key insights and definitions,
which will be found useful toward formulating the complete analysis procedure.

19.2.1 Sign Convention
We will consider the same sign convention as that has been used for the slope deflec-
tion equations: clockwise moments that act on the member will be taken as positive,
whereas counterclockwise moments will be taken as negative.

19.2.2 Fixed-End Moments (FEMs)

The moments at the ‘walls’ or perfectly fixed joints of a loaded member are called
fixed-end moments (FEMs). FEMs are already taught in the previous chapter, and
we need to remember these formulae for calculating numerical values of the moment
DOI: 10.1201/9781003081227-22 327
328 Introduction to Structural Analysis

based on the type of load applied to the structure. As for example, if a point load,
P, acts on the midspan of a beam of length l, the FEM will be ± Pl /8. In this way, if
the beam is subjected to a uniformly distributed load of intensity w, FEMs will be
± wl 2 /12. Different FEM due to the general class of loading is provided in Chapter 18,
in a tabular form.

19.2.3 Member Stiffness Factors

Consider a beam with one end pinned and another end fixed, as shown in Figure 19.1.
In this situation, a moment has been applied at the pinned end M AB. Due to this
moment, the pinned end will be rotated by an angle θ A as shown in Figure 19.1 (a).
Using slope deflection equation derived in the previous chapter, the moment can
be related to the rotation of the cord of the beam by the following equation: here,
θ B = ψ = FEM AB = 0.
2 EI
M AB = ( 2θ A )
l
i.e.,
4 EI
M AB = θA
l

The bending stiffness coefficient is defined as the moment that must be applied at the
end of a member to cause a unit angular displacement of that end, i.e.,
Setting, θ A = 1 radian we get,
4 EI
K=
l

FIGURE 19.1  Beam with end moments for stiffness coefficient calculation.
Moment Distribution Method 329

However, we will come up with situations where the far-end support may be pinned
or rocker supports in many cases. In that case, the stiffness factor will be changed,
and the fixed-beam stiffness factor formula cannot be used. Now, if the beam’s far
end is also hinged as shown in Figure 19.1 (b), we can write,
3EI
M AB = θA
l
In this case, the bending stiffness will be
3EI
K=
l
Here we can observe that the bending stiffness of the beam is reduced by 25% when
the fixed support at B is replaced by hinged support.

19.2.4  Joint Stiffness Factor

If several members are connected by a fixed joint at some location, following the
principle of superposition, the total stiffness factor at the joint will be the algebraic
sum of each and every member’s stiffness factors. This total sum of stiffness factors
indicates the moment required at a joint to rotate the joint by a unit angular displace-
ment. This is further explained in the following.
Suppose members AB, CB, DB, and EB are rigidly connected to joint B, as
shown in Figure 19.2. Now, a moment M is applied at joint B, causing it to rotate by
an angle θ . This applied moment, M, will now be resisted by all the four members
connected at joint B. If we consider the moment equilibrium of joint B as shown in
Figure 19.3, we can write,
M + M BA + M BC + M BD + M BE = 0

FIGURE 19.2  Joint stiffness and distribution factor.

330 Introduction to Structural Analysis

FIGURE 19.3  Free-body diagram of joint B and four members.

or,
M = − ( M BA + M BC + M BD + M BE )
As all the members are rigidly connected at joint B, the rotations of the ends B of
these members are the same as that of the joint. Now, the moments at B of the mem-
bers can be expressed in terms of joint rotation θ .

 4 EI1 
M BA =    θ   = K BAθ
 l1 

 4 EI 2 
M BC =    θ   = K BCθ
 l2 
Moment Distribution Method 331

 3EI 3 
M BD =    θ   = K BDθ
 l3 

 4 EI 4 
M BE =    θ   = K BEθ
 l4 

Now we can rewrite the earlier expression as,

 4 EI1 4 EI 2 3EI 3 4 EI 4 
M = − + + + θ
 l1 l2 l3 l4 

or,

M = ( K BA + K BC + K BD + K BE )θ

or,

M=− ( ∑ K )θB

Here ∑ K B is the joint stiffness factor or sum of the bending stiffness of all members
connected to joint B.

19.2.5 Distribution Factor (DF)

If a moment M is applied to a fixed connected joint, the connecting members will
each supply a portion of the resisting moment required to maintain the moment equi-
librium at that joint. That portion of the total resisting moment supplied by each
member is known as the distribution factor (DF). To calculate its value, let us con-
sider the joint shown in Figure 19.2. The rotational stiffness of a joint is defined as
the moment required to cause a unit rotation of the joint.
or,

M
θ
=− (∑ K ) B

So,

M
θ=−
∑ KB

After replacing the value θ in the earlier expressions of the member end moments,
we can write,

 K 
M BA = −  BA  M
 ∑ KB 
332 Introduction to Structural Analysis

 K 
M BC = −  BC  M
 ∑ KB 

 K 
M BD = −  BD  M
 ∑ KB 

 K 
M BA = −  BE  M
 ∑ KB 

From this, we can understand that the applied moment M is distributed to the four
members in proportion to their bending stiffness. The ratio ( K / ∑ K B ) is called the
DF for that member at end B. It represents the fraction of applied moment M that is
distributed to end B of a member.
If an applied moment M causes the joint to rotate an amount θ , each member j
rotates by this same amount. If the stiffness factor of jth member is K j, the moment
contributed by that member to maintain equilibrium at the joint will be M j = K jθ .
Since the total stiffness factor at the joint is given by K =   ∑ nJ =1 K j, hence,

Mi θKj Kj
DFi = = =
M θ ∑ j =1 K j ∑ K j
n

So, by taking all the individual member stiffness coefficients, we can use the above
formula to calculate the DF for each member by simply substituting the values in the
above equation.

19.2.6 Member Relative Stiffness Factor

For continuous beams, we frequently found that elastic constant E remains the
same throughout the entire beam. In this case, we can remove the common factor
4E, as in the DF equation, this common term gets canceled from numerator and
denominator. Hence, it is convenient just to determine the member’s relative stiff-
ness factor as:

Ij
Kj =
Lj  

when the far end of the beam is fixed and used for computation of DFs. If the far end
of a member is hinged, the relative stiffness factor will be,

3 Ij
Kj =  
4 Lj  
Moment Distribution Method 333

19.2.7 Carry over Factor

When we apply a moment M AB at end A, as shown in Figure 19.1 (a), a moment M BA
is developed at the fixed-end B. This developed moment at the fixed-end B due to the
applied moment at end A is called the carry-over moment.
We know that M AB = ( 4 EI /l )θ A and M BA = ( 2 EI /l )θ A. So, equating these two
relationships, we get:

1
M BA = M AB
2
or,

1
M′ = M
2
This indicates that moment M at pin end has been transferred to the wall or fixed
support as M/2. Hence, in the case of a beam with a far-end fixed, the carry-over
factor is +1/2. The plus sign indicates that both moments are acting in the same
direction (clockwise).

19.3  BASIC CONCEPT OF MOMENT DISTRIBUTION METHOD

To understand the basic concept of moment distribution, it is better to do it with the
help of an example. Refer to Figure 19.4 that indicates the typical arrangement of
a beam with loading conditions. We will apply the moment distribution method to
analyze the beam in a stepwise detailed manner, so that student gains a good basic
understanding of the approach.
The stiffness factor on either side of the node B is:

4 E ( 300 )
K BA = = 4 E ( 20 ) mm 4 /m
15

4 E ( 600 )
K BC = = 4 E ( 30 ) mm 4 /m
20

FIGURE 19.4  Example problem of moment distribution.

334 Introduction to Structural Analysis

So, the DFs will be:

4 E ( 20 )
DFBA = = 0.4
4 E ( 20 ) + 4 E ( 30 )

4 E ( 30 )
DFBC = = 0.6
4 E ( 20 ) + 4 E ( 30 )

At the wall or fixed support end, the stiffness is infinite. Hence, the DFs in this
respect will be:

4 E ( 20 )
DFAB = =0
∞ + 4 E ( 30 )

4 E ( 30 )
DFCB = =0
∞ + 4 E ( 30 )

Once the DFs are calculated, we will now calculate the FEMs, which is done as
follows:

ωl2
( FEM )BC = − = −6666.67 kNm
12

ωl2
( FEM )CB = = 6666.67 kNm
12

Now, to maintain equilibrium at joint B, we need to apply an equal but opposite

moment at the same joint B. Hence, the applied moment will be = +6666.67 kNm .
Now, this additional moment applied at B will be distributed in the span BA and BC
as per the DFs. So, BA will carry = +6666.67 kNm × 0.4 = +2666.67 kNm and BC
will carry +6666.67 kNm × 0.6 = +4000.00 kNm. These two moments are located at
joint B. Hence, as per the carry-over factor, the half of these moments will be carried
over to the farthest end of the beams. So, carried over moment at joint A will be =
1/2 × 2666.67 = 1333.34 kNm and at C will be = 1/2 × 4000 = 2000.00 kNm .
All the above methods are carried over by a tabular form, which helps us to keep
track of everything in a sequential manner. Table 19.1 is shown below for a ready
reference.
So, as we can see from Table 19.1, at joint B, the moment equilibrium is ensured
(since +2666.67 − 2666.67 = 0).
As long as we get this joint equilibrium ensured, we keep on this process, and
the iteration continues till, at each joint where a number of members meet, moment
equilibrium is satisfied. This tabular form is the elegance of this method that we have
told at the beginning of this chapter. The final bending moment diagram is left as an
exercise for the reader.
Moment Distribution Method 335

TABLE 19.1
Moment Distribution Process for this Problem
Fixed Fixed
Joint A B C
Member AB BA BC CB
DF 0 0.4 0.6 0
FEM −6666.67 6666.67
Balance B +2666.67 +4000.0
1 1
Carry over 1333.34 ← CO = + CO = + → 2000.0
2 2
Final moments +1333.34 +2666.67 −2666.67 +8666.67

19.4  STIFFNESS FACTOR MODIFICATIONS

We have already seen that for a beam with the far end fixed, the stiffness factor of
the beam is 4 EI /l and in section 19.2.3, we have seen if a beam’s far end is hinged,
the stiffness factor gets reduced by 25% and becomes 3EI /l. In this section, again,
we would like to develop the stiffness factor when the far end is hinged along with
few special cases by the conjugate beam method.

19.4.1 Member Pin Supported at Far End

Let us consider the following situation in which the rear end of the beam is pin sup-
ported, as shown in Figure 19.5.
Now taking a moment of all forces about support B, we get from conjugate beam:

1 M 2
VA × l −       l = 0
2  EI   3 
or,

Ml
VA = θ =
3EI
or,

3EI
M= θ
l

Thus, for a beam with far end pinned or rocker support, the stiffness factor will be:

3EI
K=
l

Also, the carry-over factor will be zero (0) since the pin end does not support any
moment. Hence, the moment will be zero at that support point.
336 Introduction to Structural Analysis

FIGURE 19.5  Pin or rocker supported far end and stiffness coefficient: (a) original beam
and (b) conjugate beam of real beam.

19.4.2 Symmetric Beam and Loading

If the beam is symmetric with respect to both loading and geometry, the bending
moment diagram of the beam will also be symmetric. As a result, modification of
stiffness factor can be modified based on the center span bending moment, as will
be clear from Figure 19.6.
Due to symmetry, the internal moment for joints B and C will be the same. Let
us consider this moment to be M, as shown in Figure 19.6 (b) for span BC conjugate
beam. We can write the following equation by taking a moment about joint C of the
conjugate beam forces:

M l
−VB′l + l    = 0
EI  2 
So,

Ml
VB′ = θ =
2 EI
Moment Distribution Method 337

FIGURE 19.6  Symmetric beam in geometry and loading.

or,

2 EI
M= θ
l

So, for symmetric loading, the stiffness factor will be:

2 EI
K=
l

Thus, moments for the half span of the beam can be distributed by using the above
stiffness of the factor measured at the center of the span of the beam.

19.4.3 Symmetric Beam with Unsymmetric Loading

In the case of a symmetric beam with antisymmetric loading, we can, like the previ-
ous case, analyze the beam concerning the center span of the beam, provided the
stiffness factor takes a different form than the usual one.
For the previous case, if the beam is loaded only for the portion AB and there is
no load in the BC span, the conjugate beam for the span AB of the beam will be as
shown in Figure 19.7.
As carried out in the earlier section, taking a moment of all forces about joint C
of the conjugate beam, as shown in Figure 19.7, we get:

1 M  l 5l 1 M l l
−VB′ l +         −         = 0
  
2 EI 2 6    
2 EI 2 6 
  
338 Introduction to Structural Analysis

FIGURE 19.7  Symmetric beam in geometry and antisymmetric loading.

or,

Ml
VB′ = θ =
6 EI

or,

6 EI
M= θ
l

So, the stiffness at the center of the span will be:

6 EI
K=
l
Moment Distribution Method 339

19.5  ANALYSIS OF CONTINUOUS BEAMS

Primary discussion on a continuous beam has been provided in Section 19.3. In
this section, we will make a detailed analysis and draw the shear force and bending
moment diagrams of a continuous beam as shown in Figure 19.8, following all the
previous concepts given in earlier sections.
At the very first, the DFs at each joint first need to be calculated:

4 EI 4 EI 4 EI
K AB = ,  K BC = ,  K CD = ,  K AB = K DC = ∞
12 12 8

So,

DFAB =  DFDC = 0

( 4 EI /12 ) 4 EI /12
DFBA =  DFBC =   = 0.5,  DFCB =   = 0.4,
4 EI /12 + 4 EI /12 4 EI /12 + 4 EI /8

4 EI /8
 DFCD = = 0.6
4 EI /12 + 4 EI /8

Next, we need to find out the FEMs of each span by considering imaginary clamps
at both ends of each span.

wl 2
FEM BC =   − =   −180 kNm, FEMCB = 180 kNm, 
12

Pl Pl
FEMCD =   − =   −200 kNm, FEM DC = = 200 kNm
8 8

Now, equipped with the above inputs, let us form the moment distribution table to
complete the analysis as done before.

FIGURE 19.8  A continuous beam arrangement and loading details.

340 Introduction to Structural Analysis

Joint A B C D
Member AB BA BC CB CD DC
DF 0 0.5 0.5 0.4 0.6 0
FEM 0 0 −180 180 −200 200
Balancing +180 (200 − 180) = +20
Moment
Dist. Mom. 0 180 × 0.5 = 90 180 × 0.5 = 90 20 × 0.4 = 8 20 × 0.6 =
12 0
Carry over 90/2 = 45 0 8/2 = 4 90/2 = 45 0  12/2 = 6
Mom.
Dist. Mom. 0 −4 × 0.5 = −2 −4 × 0.5 = −2 −45 × 0.4 = −45 × 0.6 = 0
−18 −27
Carry over −2/2 = −1 0 −18/2 = −9 −2/2 = −1 0 −27/2 =
Mom. −13.5
Dist. Mom. 0 9 × 0.5 = 4.5 9 × 0.5 = 4.5 1 × 0.4 = 0.4 1 × 0.6 = 0.6 0
Carry over 4.5/2 = 0 0.4/2 = 0.2 4.5/2 = 2.25 0 0.6/2 =
Mom. 2.25 0.3
Dist. Mom. 0 −0.2 × 0.5 = −0.2 × 0.5 = −2.25 × 0.4 = −2.25 × 0.6 0
−0.1 −0.1 −0.9 = −1.35
∑ M (FEM + 46.25 92.40 −92.40 215.75 −215.75 192.8
Dist. Mom +
carry over
Mom.)

So, the joint equilibrium of moments is ensured, as seen from the final sum-
mation row of the above table. We have provided the calculation stepwise and
detailedly so that readers gain sufficient confidence in solving other problems after
studying this. Detailed understanding of various steps involved is a must for all
students who want to gain complete control of this beautiful and elegant analysis
process. The overall steps involved in the moment distribution method is summa-
rized as below:

1. Calculate the DFs for members rigidly connected to the joints. The sum of
all the DFs at a joint must be equal to one.
2. Assuming all the free joints are clamped against rotation, calculate the
FEMs due to the external loads or support settlements (if any) for each
span. The clockwise moment is considered positive.
3. Calculate and distribute the unbalanced moment to the members connected
to a particular joint as per their DFs.
4. Carry over one-half of each distributed moment to the far end of the
member.
5. Repeat steps (3) and (4) until either all free joints are balanced or the unbal-
anced moments at these joints become negligibly small.
Moment Distribution Method 341

FIGURE 19.9  Example problem of (a) shear force, (b) bending moment, and (c) free-body
diagram of various beam spans.

6. Find out the final moments by algebraically summing up the fixed end,
distributed, and carry-over moments at each member end.
7. By considering the equilibrium of each span and joint, find out the end
shears and support reactions, respectively.
8. Draw the shear force and bending moment diagram.

The beam’s shear force and bending moment diagram is given in Figure 19.9 (a) and (b),
respectively. Upon drawing the free-body diagram of respective beam elements as
shown in Figure 19.9 (c), one can determine the shear force and support reactions by
applying equilibrium equations. The calculation part has been left as an exercise for
the students.

19.6  ANALYSIS OF FRAMES WITHOUT SIDESWAY

For frames without any sidesway, the moment distribution method can be conve-
niently applied as has been done for continuous beams. Following the same tabular
form, we will analyze a frame to determine the unknown moments at its joints.
Moreover, in special cases, the steps involved can be reduced by adopting modified
stiffness methods as and when required.
342 Introduction to Structural Analysis

Example 19.1: Determine the internal moments at the joints of the

below frame. All the supports are fixed – except E, D which are pinned,
and EI is cinstant for the entire frame.

SOLUTION: 

Joint A B C D E
Member AB BA BC CB CD CE DC EC
DF 0 0.545 0.455 0.330 0.298 0.372 1 1
FEM 0 0 −216 +216 0 0 0 0
Balancing 216 −216
Mom.
Dist. Mom 0 117.72 96.12 −71.28 −64.368 −80.352 0 0
Carry over 117.72/2 0 −71.28/2 = 96.12/2 = 0 0 0 0
Mom. = 58.86 −35.64 48.06
Dist. Mom 19.42 16.22 −15.86 −14.32 −17.88 0 0
Carry over 9.71 0 −15.86/2 = 16.22/2 = 0 0 0 0
Mom. −7.93 8.11
Dist. Mom 0 4.321 3.529 −2.676 −2.417 −3.017 0 0
Carry over 2.16 0 −1.338 1.765 0 0 0 0
Mom.
Dist. Mom 0 0.729 0.595 −0.582 −0.526 −0.6576 0 0
Carry over 0.365 0 −0.291 0.298 0 0 0 0
Mom.
Dist. Mom 71.095 142.19 142.19 183.835 −81.631 102.91 0 0

Since E and D are pinned, the stiffness factors for CD and CE can be computed by
K = 3EI /l , i.e., KCD = 3EI /15 and KCE = 3EI /12.

So, DFAB = 0,

4EI /15
DFBA = = 0.545
4EI /15 + 4EI /18
DFBC = 1− 0.545 = 0.455

4EI /18
DFCB =   = 0.330
4EI /18 + 3EI /15 + 3EI /12
3EI /15
DFCD =   = 0.298
4EI /18 + 3EI /15 + 3EI /12

DFCE = 1− 0.330 − 0.298 = 0.372

DFDC = 1

DFEC = 1

We used these values in the tabular form on the previous page to analyze the frame
using the moment distribution method. The shear force and bending moment dia-
gram of this frame is shown in Figure 19.11 (a) and (b), respectively.
Moment Distribution Method 343

FIGURE 19.10  Example problem of frame without sidesway.

FIGURE 19.11  (a) Shear force and (b) bending moment diagram of the frame without
sidesway.
344 Introduction to Structural Analysis

19.7  ANALYSIS OF FRAMES WITH SIDESWAY

In Chapter 18, on the slope deflection method, we have explained that frames
that are unsymmetrical in geometry or applied load or both are subject to sway.
A situation for a frame with unsymmetrical loading acting on it is given in
Figure 19.12. In this frame, the applied force tends to sway the frame to the
right by a deflection ∆ . Due to this deflection, there will be unequal moments
distributed in the frames. We can determine the final moments in the frame due
to sidesway by the application of the principle of superposition. First, we assume
that there is a virtual support at C, which makes the frame nonsway type. Then
we calculate the support reaction R at this support C by completing the moment
distribution of this nonsway frame. After that, we apply the same but opposite
restraining force R in the frame and calculate the internal moments once again.
One method for doing this last step is to assume a value for any one of the inter-
nal moments, say, M BA′ . Using moment distribution and statics, we can then deter-
mine the deflection ∆ ′ and R′ against the assumed value of M BA ′ . Since linear
deformation occurs within the elastic limit, actual force R will be proportional to
this. So that we can write,

′
M BA M BA
=
R R′

From which we get,

′ ( R /R ′ )
∴ M BA = M BA

To understand the abovementioned process in detail, we will go through a problem

discussed in Example 19.2, which will clear all our doubts and enable us to gain
confidence in the analysis procedure.

FIGURE 19.12  Frame with sidesway – preliminary analysis method.

Moment Distribution Method 345

Example 19.2: Determine the moments at each joint of the frame as

shown in Figure 19.13.

SOLUTION:  We first make the frame prevent from sidesway as was done in
Figure 19.13 (b), i.e., with virtual support at C.
Now, with the virtual supports, we first calculate the FEMs against the applied
loading:

4 2 (1)
FEMBC = −20 = −12.8 kNm
52

12 ( 4 )
FEMCB = 20 = 3.2 kNm
52

Then, the DFs are calculated:

4EI /5
DFAB = 0, DFBC = = 0.5, DFBA = 1−  DFBC = 0.5
4EI /5 + 4EI /5

4EI /5 4EI /5
 DFCB = = 0.5, DFCD = = 0.5, DFDC = 0
4EI /5 + 4EI /5 4EI /5 + 4EI /5

Now, we are in a position to solve the first stage of this problem by usual moment
distribution method as given in the tabular form:

Joint A B C D
Member AB BA BC CB CD DC
DF 0    0.5   0.5   0.5  0.5  0
FEM 0    0 −12.8   3.2  0  0
Balancing Mom. +12.8 −3.2
Dist. Mom.    6.4   6.4 −1.6 −1.6
Carry over Mom. 3.2    0   −0.8   3.2  0 −0.8
Dist. Mom. 0    0.40   0.40 −1.6 −1.6  0
Carry over Mom. 0.20    0   −0.80   0.20  0 −0.80
Dist. Mom. 0    0.40   0.40 −0.10 −0.10  0
Carry over Mom. 0.20    0   −0.05   0.20  0 −0.05
Dist. Mom. 0    0.025   0.025 −0.1 −0.1  0
∑ M (FEM + Dist. Mom. + 3.6    7.225   −7.225   3.4 −3.4 −1.65
carry over Mom.)

Once done with this, we need to draw free-body diagram of the columns to deter-
mine end shear force acting in the column (Figure 19.14).
346 Introduction to Structural Analysis

FIGURE 19.13  Example problem with sidesway and virtual support.

Taking moment about joint B in Column BA, we get:

RA ( 5) − 3.6 − 7.225 = 0

or,

RA = 2.165 kN

Similarly, by taking moment about joint C of column CD, we get:

RD ( 5) − 3.4 − 1.65 = 0

or,

RD = 1.01 kN

FIGURE 19.14  Column free-body diagram for force analysis.

Moment Distribution Method 347

So, from the overall frame, applying global equilibrium force equation in horizon-
tal direction, we get:

R − 2.165 + 1.01 = 0
or,

R = 1.155 kN
Now, we must apply an equal but opposite force of R on the frame at point C and
calculate the internal moments accordingly. This situation is shown in Figure 19.13 (c)
with the deflection diagram for better understanding of the reader.
In the above case, the joints B and C are assumed to be restrained against
rotation. Under this situation, the deflection ∆ induces moments in the columns,
which can be determined by using the concept of slope deflection equations.
Consider a member AB fixed at both ends and support B settles by an amount ∆.
Also let us consider that there is no load acting on the beam. So, applying slope
deflection equation we get:

2EI  ∆
M AB =
L  2θ A +  θ B − 3 L  +  (FEM) AB

Here, θ A =  θ B = 0, (FEM) AB = 0, and the cord members AB and DC rotate clock-

wise (ψ = ∆ /L = + ve). So, from the above equation we get:

6EI∆
M AB =   −
L2
Hence, for our columns DC and AB, we have induced moment due to the
deflection:

6EI∆
−
52
Also, since both ends, B and C of the frame in Figure 19.15, are happened to be
displaced by the same amount ∆′, and AB, DC have the same E, I, and length,
FEM in AB and DC will be the same. So, let us arbitrarily choose values of FEMs
as follows:

FEM AB =  FEMBA =  FEMCD =  FEMDC =   −100 kNm

FIGURE 19.15  Arbitrarily assumed fixed-end moments.

348 Introduction to Structural Analysis

Negative sign is due to anticlockwise FEM induced. With these values in hand, let
us carry out the moment distribution once again for the entire frame in the tabular
form as given below:

Joint A B C D
Member AB BA BC CB CD DC
DF 0 0.5 0.5 0.5 0.5 0
FEM −100 −100 0 0 −100 −100
Balancing Mom. 100 100
Dist. Mom 0 50 50 50 50 0
Carry over Mom. 25 0 25 25 0 25
Dist. Mom 0 −12.5 −12.5 −12.5 −12.5 0
Carry over Mom. −6.25 0 −6.25 −6.25 0 −6.25
Dist. Mom 0 3.125 3.125 3.125 3.125 0
Carry over Mom. 1.56 0 1.56 1.56 0 1.56
Dist. Mom 0 −0.78 −0.78 −0.78 −0.78 0
Carry over Mom. −0.39 0 −0.39 −0.39 0 −0.39
Dist. Mom 0 0.195 0.195 0.195 0.195 0
∑ M (FEM + Dist. Mom + −80.08 −59.96 59.96 59.96 −59.96 −80.08
carry over Mom.)

Once the table is ready, we can draw the same column free-body diagrams as
done in Figure 19.14 and analyze the case for column horizontal support reac-
tions. In this case, the support reactions in horizontal direction will be:

RA = 28.01 kN

RD = 28.01 kN

Thus, for the entire frame, applying the global equilibrium of force condition in
horizontal direction, we get:

R′ − 28.01− 28.01 = 0
or,

R′ = 56.02 kN
So, the end moments at the joint of this frame will be:

 1.155 
M AB = 3.6 + 
 56.02 
( −80.08) = 1.95 kNm

 1.155 
MBA = 7.225 + 
 56.02 
( −59.96) = 5.99 kNm

 1.155 
MBC = −7.225 + 
 56.02 
(59.96) = −5.99 kNm
Moment Distribution Method 349

 1.155 
MCB = 3.4 + 
 56.02 
(59.96) = 4.64 kNm

 1.155 
MCD = −3.4 + 
 56.02 
( −59.96) = −4.64 kNm

 1.155 
MDC = −1.65 + 
 56.02 
( −80.08) = −3.30 kNm

Now the analysis is complete. The final shear force and bending moment diagram
of this frame is shown in Figure 19.16.

FIGURE 19.16  (a) Shear force and (b) bending moment diagram of the given frame.
350 Introduction to Structural Analysis

19.7.1 Multistorey Frames
For multistorey frames, we analyze them in the same manner as that of a single-
storied frame explained in detail above. To elaborate the process, let us consider a
multistorey frame as shown in Figure 19.17. Under action of an unsymmetrical load-
ing, the frame will sway as shown in the same diagram. To do the analysis, we first
apply two supports at two suitable nodes of the frame to make it into nonsway-type
frame as shown in Figure 19.17 (b). After that, we carry out the moment distribution
of the frame with these virtual supports and calculate the virtual support reactions
R1 and R2 from the global equilibrium equations.

FIGURE 19.17  Multistorey frame with sway: (a) original frame, (b) frame with virtual sup-
ports to make nonsway-type (c), and (d) multistorey frame with support reactions for second
stage analysis.
Moment Distribution Method 351

Once the first part of analysis is complete, we apply the horizontal virtual dis-
placements one by one in two nodes keeping the other node as restrained and
carry out the moment distributions separately to calculate the virtual support
reactions R1′, R2′ , and R1′′ , R ′′2 , respectively. These displacements cause FEMs in
the frame and which can be assigned specific numerical values in the frame. By
distributing these FEMs and with the help of equations of equilibrium, we can find
out R1′, R2′ , and R1′′ , R ′′2 . See Figure 19.17 (c) and (d) for understanding the process.
Thus, at this stage, we have completed the second-stage analysis process of the
multistorey frame.
Since in the last two steps as per Figure 19.17 (c) and (d), we have to assume inter-
nal FEM values twice; hence, there must be some correction factors that need to be
considered while forming the equilibrium equations, which are given below:

R2 =   −C ′R2′ + C ′′R2′′

R1 = +C ′R1′ − C ′′R1′′

By solving the above two simultaneous equations, we will determine the values
of the correction factors C ′ and C′′. These correction factors are then multiplied
with the moments we get after completing analysis in stage one and two analysis.
The resultant moments are then found by adding these corrected moments to those
obtained for frame, shown in Figure 19.17 (b).
As this process of analysis for multistorey frame for large buildings will become
very much difficult to handle manually, computer programming or different struc-
tural analysis software is called for analysis of large multistorey frame. Also, mul-
tistorey frames involve several joint displacements independently, which is also
very difficult to analyze manually without some assumptions. The more assump-
tions we take, more erroneous the result becomes. Matrix analysis procedures
may also be adopted to solve this type of problems much efficiently than moment
distribution method. In the later chapter, we will provide introduction of matrix
structural analysis and provide some insight about the analysis process of multi-
storey frames.
20 Kani’s Method or Rotation
Contribution Method

20.1 INTRODUCTION
This method was introduced by Gasper Kani of Germany in 1947. This iterative
method is suitable for approximate analysis of multistory frames, beams, or other
statically indeterminate structures in their entirety, i.e., approximate values of the
internal forces for the whole frame, beams, or other indeterminate structures can be
determined by applying this method. This method is seldom used today due to the
advent of advanced structural analysis software programs. Still, for interested read-
ers of this subject, it is important to learn this method to have a quick analysis to get
approximate results of frames and beams, which can be supplied as input toward
preliminary design drawing preparations.

20.2  BASIC CONCEPT

In this section, we will discuss the basic concept of Kani’s method or rotation con-
tribution method. First, we will develop the idea without any lateral displacement at
the ends of a member. After that, we will formulate the same, considering the lateral
displacements at the ends.

20.2.1 Members without Relative Lateral Displacement

Let us consider AB, as shown in Figure 20.1 (a), one of the spans of a framed or a
continuous structure. The deformed shape of the member AB is shown in Figure 20.1 (b).
Let us assume that no lateral displacement occurs at the ends of member AB. Let the
end A rotates by an angle θ A under the influence of end moment M AB, and the end B
rotates by an angle θ B under the influence of end moment M BA. Let us assume that
clockwise end moments and clockwise rotations at the ends are positive. Now the
final end moments M AB and M BA can be split up into the components as shown in
Figure 20.1 (c)–(e).
First of all, the ends A and B of the member are considered as fixed. Corresponding
to this condition, the fixed end moments at A and B are determined as M AB and M BA,
respectively. Now, maintaining the fixity at end B, let the end A is rotated through an
angle θ A by applying a moment 2M AB ′ . Due to 2M AB
′ at end A, a moment of magnitude
′ in the same direction will be induced at end B. This moment is called as rotation
M AB
contribution of end A. Similarly at this stage, maintaining the fixity at end A, let the
end B is rotated through an angle θ B by applying a moment 2M BA ′ . Due to 2M BA
′ at

DOI: 10.1201/9781003081227-23 353

354 Introduction to Structural Analysis

FIGURE 20.1  Deformation components of a framed or continuous structure subjected to

external loading.
Kani’s Method or Rotation Contribution Method 355

FIGURE 20.2  A multistorey frame subjected to vertical loading only.

end B, a moment of magnitude M BA ′ in the same direction will be induced at end A.

This moment is called as rotation contribution of end B. So, the final moments M AB
and M BA can be expressed as follows:

M AB = M AB + 2 M AB
′ + M BA
′

M BA = M BA + 2 M BA
′ + M AB
′

The expression for the final moment at the near end of a member can be expressed
as the algebraic sum of:

1. The fixed end moment at the near end.

2. Twice the rotation contribution of the near end.
3. Rotation contribution of the far end.

Now let us consider a multistorey frame as shown in Figure 20.2. Suppose no lateral
joint displacement is occurring for any of the members of this frame. Let us consider
joint A, where HA, LA, DA, and BA members meet together. The end moments meet-
ing at this joint A can be written as:

M AB = M AB + 2 M AB
′ + M BA
′

M AL = M AL + 2 M AL
′ + M LA
′

M AH = M AH + 2 M AH
′ + M HA
′

M AD = M AD + 2 M AD
′ + M DA
′
356 Introduction to Structural Analysis

For the equilibrium of joint A, the sum of all the end moments at A must be zero.
i.e.,

∑M = 0 A

∑ M = ∑ M + 2∑ M ′ + ∑ M ′
Aj Aj Aj jA = 0 (20.1)

where ∑ M Aj is the algebraic sum of the fixed end moments at A for all members
meeting at A. ∑ M Aj′ is the algebraic sum of the rotation contributions at A for all
members meeting at A. ∑ M ′jA is the algebraic sum of the rotation contributions at the
far end joints with respect to joint A.
From equation (20.1), we can write,

∑M′ Aj
1
=    − 
 2 {∑ M + ∑ M ′ } (20.2)
Aj jA

We know for the member AB, with the end B fixed as shown in Figure 20.1 (d), the
moment required at A to rotate the joint A by θ A can be expressed as:
4 EI AB
′ =
2 M AB θ A = 4 EK ABθ A
L AB
where I AB /L AB = K AB ,  E = Young’s modulus.
i.e.,
′ = 2 EK ABθ A (20.3)
M AB

Now, as all the members meeting at joint A are rigidly connected, they will undergo
the same amount of rotation θ A. Assuming Young’s modulus are the same for all the
members, now we can write:

∑M′ Aj = 2 Eθ A ∑K Aj (20.4)

Now, dividing equation (20.3) by equation (20.4), we get,

′
M AB 2 EK ABθ A  
=
∑ M Aj
′ 2 Eθ A ∑ K Aj
or,

 K 
′ =  AB 
M AB
 ∑ K Aj 
∑ M ′ (20.5)
Aj

Now, from equations (20.2) and (20.5), we get,

M AB
 1 K AB 
′ =   −
 2 ∑ K Aj 
{∑ M + ∑ M ′ } (20.6)
Aj jA
Kani’s Method or Rotation Contribution Method 357

( )
The ratio − (1/2 )( K AB / ∑ K Aj ) is called the rotation factor for the member AB at joint A.
( )
Let us consider, µ AB = − (1/2 )( K AB / ∑ K Aj ) , we can rewrite equation (20.6) as,

′ =   µ AB
M AB {∑ M + ∑ M ′ } (20.7)
Aj jA

From equation (20.7), the summation of fixed end moments, ∑ M Aj , can be computed
easily and is a known quantity. The summation of far end rotation contributions, ∑ M ′jA ,
is also a known quantity, as in the first trial, we take the far end rotation moment as
zero to start with. Thus from equation (20.7), we can easily find the near-end rotation
contributions through some cycles until more accurate values are obtained. If the end
of a member is fixed, the rotation of that end is zero, and the rotation contribution of
that end is zero as well. If the end of a member is hinged or pinned, it is convenient to
consider the end as fixed and to take the relative stiffness as 43 LI .
We will give the details of this analysis procedure by carrying out the detailed
calculations of Example 20.1, which follow this section.

Example 20.1: Solve the continuous beam shown in Figure 20.3 by

rotation contribution method.

SOLUTION:  The fixed end moments are calculated as next, considering clockwise
moments as positive:

100 × 1× 22
M AB = − = −44.44 kNm
32

100 × 2 × 12
MBA = = 22.22 kNm
32

50 × 4 2
MBC = − = −66.67 kNm
12

50 × 4 2
MCB = = 66.67 kNm
12
60 × 4
MCD = − = −30 kNm
8
60 × 4
MDC = = 30 kNm
8

FIGURE 20.3  Example problem of a continuous beam by Kani’s method.

358 Introduction to Structural Analysis

TABLE 20.1
Calculating Rotation Factors for the Example Problem
Rotation Factor,
Relative Total Relative
 1 K 
Joint Members Stiffness, K Stiffness, ∑ K µ = −
 2 ∑ K 

BA 1.5I 1
−
B 3 I 4
BC 2I 1
−
4 4
CB 2I 1
−
C 4 3I 3
CD I 4 1
−
4 6

Next, we will evaluate the rotation factors at joints B and C using Table 20.1.
The sum of the fixed end moments at joint B,

∑M B = MBA + MBC

or, ∑M B = 22.22 − 66.67 = −44.45 kNm

The sum of the fixed end moments at joint C,

∑M C = MCB + MCD

or, ∑M C = 66.67 − 30 = 36.67 kNm

The scheme for proceeding with the method of rotation contribution is shown in
Figure 20.4, where the joints B and C are shown as two square boxes. The sum of
the fixed end moments at B and C is written in the smaller square boxes for B and
C, respectively. The rotation factors −1/4 for members BA and BC at B, and the
rotation factors −1/3 and −1/6 for members CB and CD at C are also shown in
the figure. The fixed end moments are entered above the horizontal line outside
the square boxes.

FIGURE 20.4  The scheme for proceeding with the method of rotation contribution.
Kani’s Method or Rotation Contribution Method 359

The rotation contribution can now be determined by a successive iteration

process as explained next:

Cycle 1
Consider joint B:
Applying equation (20.7) to this joint, we get,

′ =   µBA
MBA {∑ M + ∑ M′ }
B jB
j = A ,C

and
′ =   µBC
MBC {∑ M + ∑ M′ }
B jB
j = A ,C

where ∑ MB is the sum of fixed end moments at B = 22.22 − 66.67 = −44.45 kNm,
∑ M AB
′ = 0, since end A is fixed end, ∑ MCB
′ = 0, assumed to start with.
Substituting these values, we get,

 1
′ =   −  {−44.45 + 0 + 0} = 11.11 kNm
MBA
 4

 1
′ =  −  {−44.45 + 0 + 0} = 11.11 kNm
MBC
 4

Consider joint C:
′ and MCD
Rotation moments MCB ′ will be determined as follows:

′ =   µCB
MCB {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 11.11+ 0} = −15.93 kNm
MCB
 3

′ = µCD
MCD {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 11.11+ 0} = −7.96 kNm
MCD
 6

Cycle 2
Consider joint B:

′ =   µBA
MBA {∑ M + ∑ M′ }
B jB
j = A ,C

or,
 1
′ =  −  {−44.45 + 0 − 15.93} = 15.095 kNm
MBA
 4

′ =   µBC
MBC {∑ M + ∑ M′ }
B jB
j = A ,C
360 Introduction to Structural Analysis

or,

 1
′ =  −  {−44.45 + 0 − 15.93} = 15.095 kNm
MBC
 4
Consider joint C:

′ =   µCB
MCB {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 15.095 + 0} = −17.255 kNm
MCB
 3

′ = µCD
MCD {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 15.095 + 0} = −8.63 kNm
MCD
 6

Cycle 3
Consider joint B:

′ =   µBA
MBA {∑ M + ∑ M′ }
B jB
j = A ,C

or,
 1
′ =  −  {−44.45 + 0 − 17.255} = 15.43 kNm
MBA
 4

′ =   µBC
MBC {∑ M + ∑ M′ }
B jB
j = A ,C

or,
 1
′ =  −  {−44.45 + 0 − 17.255} = 15.43 kNm
MBC
 4

Consider joint C:

′ =   µCB
MCB {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 15.43 + 0} = −17.37 kNm
MCB
 3

′ = µCD
MCD {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 15.43 + 0} = −8.68 kNm
MCD
 6
Kani’s Method or Rotation Contribution Method 361

Cycle 4
Consider joint B:

′ =   µBA
MBA {∑ M + ∑ M′ }
B jB
j = A ,C

or,
 1
′ =  −  {−44.45 + 0 − 17.37} = 15.455 kNm
MBA
 4

′ =   µBC
MBC {∑ M + ∑ M′ }
B jB
j = A ,C

or,
 1
′ =  −  {−44.45 + 0 − 17.37} = 15.455 kNm
MBC
 4

Consider joint C:

′ =   µCB
MCB {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 15.455 + 0} = −17.375 kNm
MCB
 3

′ = µCD
MCD {∑ M + ∑ M′ }
C jC
j = B ,D

or,
 1
′ =  −  {36.67 + 15.455 + 0} = −8.6875 kNm
MCD
 6

The scheme for proceeding with the rotation contribution method after Cycle 4 is
shown in Figure 20.5.
After the fourth cycle, we may stop our iteration process as the difference
between two successive cycles, as can be seen from Figure 20.5, is becoming
negligible. Now we can proceed to calculate the end moments as follows with the
outcome of this fourth cycle.

FIGURE 20.5  The scheme for proceeding with the method of rotation contribution after
Cycle 4.
362 Introduction to Structural Analysis

FIGURE 20.6  (a) Determining the end moments by Kani’s method, (b) bending moment
diagram, and (c) shear force diagram.

For any span and from equation (20.1), we can write,

Final end moment at the near end = fixed end moment at the near end + twice
the rotation contribution of the near end + rotation contribution of the far end.
The calculation for the final end moment is shown in Figure 20.6 (a). The
corresponding bending moment and shear force diagram for this continuous
beam is shown in Figure 20.6 (b) and (c), respectively.

20.2.2 Members with Relative Lateral Displacement

In this section, we will reformulate the rotation contribution method considering
the members whose ends have undergone lateral displacement of ∆ amount as
shown in Figure 20.7.
From the slope deflection equations, we can calculate the end moments due to
this lateral displacement ∆ as follows:
2 EI
M AB =
L
( 2θ A + θ B − 3ψ AB ) + M AB
where θ A = θ B = 0 , as both ends are fixed; ψ AB = − ∆L , negative as the cord rotation
is anticlockwise; M AB = 0 , as no external load is present on the member AB. After
substituting all these values in the abovementioned equation, we can get,
2 EI  3∆  6 EI∆
M AB =  +  = + 2   = M BA
L  L L
Kani’s Method or Rotation Contribution Method 363

FIGURE 20.7  Member with relative lateral displacement ∆.

We have considered the clockwise moments as positive. And here also, we have
obtained the end moments as positive, so they should be clockwise in nature due to
anticlockwise cord rotation, as shown in the figure.
Let us rename, M AB = M BA = M AB
′′ = M BA
′′ = + 6 EI
L2
∆

When this kind of lateral displacements occur for a member, the final moments
at A and B are given by,

M AB = M AB + 2 M AB
′ + M BA
′ + M AB
′′

M BA = M BA + 2 M BA
′ + M AB
′ + M BA
′′

′′ = M BA
The quantity M AB ′′ is called the displacement contribution of the member AB.
The expression for the final moment at the near end of a member can be expressed
as the algebraic sum of:

1. The fixed end moment at the near end.

2. Twice the rotation contribution of the near end.
3. Rotation contribution of the far end.
4. The displacement contributions of the near end.

If several members meet at joint A, we can similarly derive the following expression
like as equation (20.7).

′ =   µ AB
M AB {∑ M + ∑ M ′ + ∑ M ′′ }
Aj jA Aj

Now, using the abovementioned equation, we can easily find out the rotation contri-
butions iteratively, including the effect of lateral displacement, and finally can obtain
the end moments.
If the lateral displacement ∆ is known, the displacement contribution, i.e., the
fixed end moment generated due to lateral displacement, can also be found. In that
364 Introduction to Structural Analysis

case, the net fixed end moment is calculated at each end. If ∆ is not known, then
additional equations are to be used to find out the results.
Let us go through Example 20.2 to understand the effect of lateral displacement.

Example 20.2: Find the member end moments of the three-span

continuous beam due to the support settlement of 3.5 mm at B, 9 mm
at C, and the loading condition shown in Figure 20.8 (a). For all members,
take I = 3.5 × 107 mm4 and E = 200 kN/mm2.

SOLUTION:  Let us consider clockwise moment and clockwise cord rotation as

positive.
Fixed end moments
Span AB
Resultant fixed end moment at support

60 × 62 6 × 200 × 3.5 × 107 × 3.5

A=− − = −180 − 4.083 = −184.083 kNm
12 62 × 109
Resultant fixed end moment at support B = +180 − 4.083 = 175.917 kNm
Span BC
Resultant fixed end moment at support

150 × 5 6 × 200 × 3.5 × 107 × ( 9 − 3.5)

B=− − = −93.75 − 9.24 =   −102.99 kNm
8 52 × 109
Resultant fixed end moment at support C = 93.75 − 9.24 = 84.51 kNm

FIGURE 20.8  Example problem on Kani’s method considering support settlement.

(a) Continuous beam and (b) cord rotations due to support settlements.
Kani’s Method or Rotation Contribution Method 365

TABLE 20.2
Calculating Rotation Factors for the Example Problem
Rotation Factor,
Relative Total Relative
 1 K 
Joint Members Stiffness, K Stiffness, ∑ K µ = −
 2 ∑ K 

BA I 5
−
6 11I 22
B
BC I 30 6
−
5 22
CB I 6
−
C 5 11I 22
CD I 30 5
−
6 22

Span CD
Resultant fixed end moment at support

80 × 6 6 × 200 × 3.5 × 107 × 9

C=− + = −60 + 10.5 = −49.5 kNm
8 62 × 109
Resultant fixed end moment at support D = +60 + 10.5 = 70.5 kNm
Next, we will evaluate the rotation factors at joints B and C using Table 20.2.
The scheme for proceeding with the rotation contribution method after Cycle 4
is shown in Figure 20.9.
The calculation for the final end moment is shown in Figure 20.10 (a). The cor-
responding bending moment and shear force diagram for this continuous beam
are shown in Figure 20.10 (b) and (c), respectively.

FIGURE 20.9  The scheme for proceeding with Kani’s method after Cycle 4.
366 Introduction to Structural Analysis

FIGURE 20.10  (a) Determining the end moments by Kani’s method for support settlement
problem, (b) bending moment diagram, and (c) shear force diagram.

20.3 ANALYSIS OF FRAMES WITH SIDESWAY

WITH VERTICAL LOADINGS
Let us consider the multistory frame shown in Figure 20.2 again and let HK represent
a vertical member of the frame. Let M HK and M KH be the end moments in the end H
and K, respectively, and the horizontal forces exerted by the frame on the member
HK at H and K be FH as shown in Figure 20.11 under equilibrium condition.
As the member HK is under equilibrium condition, we can write,

M HK + M KH + FH h = 0

FIGURE 20.11  Equilibrium of a vertical column of a multistory frame.

Kani’s Method or Rotation Contribution Method 367

or,
M HK + M KH
FH = −
h
FH represents the shear force at any section of the member HK. Let HK, AL, BM, IN,
JO be the vertical members of a story. Applying a similar concept, we can finally
find out the sum of shear forces in all the columns of a particular story as follows:

∑F H =−
∑ M HK + ∑ M KH
h
where ∑ M HK is the sum of the end moments at the upper ends of all the columns of
a particular story. ∑ M KH is the sum of the end moments at the lower ends of all the
columns of a particular story. ∑ FH is the story shear for a particular story. h is the
height of the columns of a particular story.
Obviously, the story shear for a particular story should be equal to the sum of all
horizontal external loads above that particular story. But in this case, the multistory
frame is only subjected to external vertical loads. So, for each story, the story shear
is equal to zero. So, we can write, ∑ M HK + ∑ M KH = 0 , for a particular story. From
the earlier discussions, we can write the general expressions for the end moments for
column HK as next,

M HK = M HK + 2 M HK
′ + M KH
′ + M HK
′′

M KH = M KH + 2 M KH
′ + M HK
′ + M KH
′′

The terminologies used in the abovementioned two equations are already explained.
Since the loading on the frame is vertical only, M HK = M KH = 0 . Now, if we add the
abovementioned two equations, we will get,

M HK + M KH = 3 M HK
′ + 3 M KH
′ + 2 M HK
′′
But,

∑M HK + ∑M KH =0

∴3 ∑M′ HK +3∑M′ KH +2 ∑ M ′′
HK =0

Therefore,

∑ M ′′HK
3
= − 
2 ∑M′ + ∑M′
HK KH
 (20.8)


The abovementioned equation establishes a relationship between the rotation contri-

butions and the displacement contributions. We already know for any member the
displacement contribution is,
6 EI∆
=
L2
368 Introduction to Structural Analysis

Now, the relative lateral displacement ∆ is the same for all the columns of a story.
Suppose the length, L and Young’s modulus, E is assumed to be the same for all the
columns. In that case, we can say the displacement contribution for a column of a
story is proportional to the moment of inertia of the section of the column.
i.e.,

′′ ∝ I
M HK

′′ ∝ K HK , if LHK = L , for
But we know the relative stiffness, K HK = I HK /LHK . So, M HK
all the columns of a story.
Therefore,

′′
M HK K HK
=
∑ M HK
′′ ∑ K HK

′′ =
M HK
K HK
∑ K HK ∑ M ′′ (20.9)
HK

From Eqs. (20.8) and (20.9), we get,

′′ =
M HK
K HK  3  
− 
∑ K HK  2   ∑M′ + ∑M′
HK KH
 (20.10)


The quantity ( K HK / ∑ K HK )( −3/2 ) is called the displacement factor for member HK.
In equation (20.10), ∑ M HK′ + ∑ M KH ′ represents the sum of rotation contributions
of the top and bottom ends of all the columns of a particular story. ∑ K HK represents
the sum of the relative stiffness of all the columns of a story concerned.
( )
Let us consider, γ HK = − 32 ∑KKHKHK , we can rewrite equation (20.10) as,

′′ =  γ HK 
M HK
 ∑M′ + ∑M′
HK KH
 (20.11)


Let us discuss Example 20.3 to understand the procedure just discussed.

Example 20.3: Find the member end moments for the portal frame
shown in Figure 20.12.

SOLUTION:  Fixed end moments

M AB = MBA = MCD = MDC = 0

190 × 4 × 82
MBC = − = −337.78 kNm
122

190 × 8 × 4 2
MCB = = 168.89 kNm
122
Kani’s Method or Rotation Contribution Method 369

FIGURE 20.12  Example problem on Kani’s method considering a portal frame under verti-
cal unsymmetrical load.

Rotation factors
Rotation factors are calculated using Table 20.3.
Displacement factors
Displacement factors are calculated using Table 20.4.
The sum of the fixed end moments at joint B,

∑M B = MBA + MBC

or,

∑M B = 0 − 337.78 = −337.78 kNm

TABLE 20.3
Calculating Rotation Factors for the Given Portal Frame
Rotation Factor,
Relative Total Relative
 1 K 
Joint Member Stiffness, K Stiffness, ∑ K µ = −
 2 ∑ K 
BA I 1
−
9 5I 5
B
BC 2I 18 3
−
12 10
CB 2I 3
−
C 12 5I 10
CD I 18 1
−
9 5
370 Introduction to Structural Analysis

TABLE 20.4
Calculating Displacement Factors for the Given Portal Frame
Displacement Factor,
Relative Total Relative
 3 K 
Vertical Member Stiffness, K Stiffness, ∑ K γ = −
 2 ∑ K 
AB I 3
−
9 2I 4
DC I 9 3
−
9 4

The sum of the fixed end moments at joint C,

∑M C =  MCB + MCD

or,
∑M C =  168.89 + 0 = 168.89 kNm

The scheme for proceeding with the method of rotation contribution is shown in
Figure 20.13, where the joints B and C are shown as two square boxes. The sum
of the fixed end moments at B and C is written in the smaller square boxes for B
and C, respectively. The rotation factors −1/5 for member BA and −3/10 for BC at
B and the rotation factors −1/5 for member CD and −3/10 for members CB at C are
also shown in the figure. The displacement factors are entered by the side of each
column. The fixed end moments are entered above the horizontal line outside the
square boxes.

FIGURE 20.13  The scheme for proceeding with Kani’s method.

Kani’s Method or Rotation Contribution Method 371

Iteration process: Cycle 1

Joint B
Summation of the fixed end moments = −337.78
Far end rotation contributions
At A =0
At C (assumed) =0
Summation of displacement contributions
BC =0
BA (assumed) =0
     
= −337.78
 1
′ =   −  ( −337.78)
MBA
 5

= 67.556 kNm
 3
′ =  −  ( −337.78)
MBC
 10 

= 101.334 kNm
Joint C
Summation of the fixed end moments = 168.89
Far end rotation contributions
At B = 101.334
At D =0
Summation of displacement contributions
CB =0
CD (assumed) =0
   
= 270.224
 3
′ =   −  ( 270.224 )
MCB
 10 

= −81.067 kNm
 1
′ =  −  ( 270.224 )
MCD
 5

= −54.045 kNm
Story one (there is only one story)
Rotation contribution at top of column BA = 67.556
Rotation contribution at top of column CD = −54.045
Rotation contribution at bottom of columns =0
   
= 13.511
 3
′′ =  −  (13.511)
M AB
 4

= −10.133
 3
′′ =  −  (13.511)
MCD
 4

= −10.133
372 Introduction to Structural Analysis

Iteration process: Cycle 2

Joint B
Summation of the fixed end moments = −337.78
Far end rotation contributions
At A =0
At C = −81.067
Summation of displacement contributions
BC =0
BA = −10.133
   
= −428.98
 1
′ =   −  ( −428.98)
MBA
 5

= 85.796 kNm
 3
′ =  −  ( −428.98)
MBC
 10 

= 128.694 kNm
Joint C
Summation of the fixed end moments = 168.89
Far end rotation contributions
At B = 128.694
At D =0
Summation of displacement
contributions
CB =0
CD = −10.133
   
= 287.451
 3
′ =   −  ( 287.451)
MCB
 10 

= −86.235 kNm
 1
′ =  −  ( 287.451)
MCD
 5

= −57.49 kNm
Story one (there is only one story)
Rotation contribution at top of column BA = 85.796
Rotation contribution at top of column CD = −57.49
Rotation contribution at bottom of columns =0
  
= 28.306
 3
′′ =  −  ( 28.306)
M AB
 4

= −21.229
 3
′′ =  −  ( 28.306)
MCD
 4

= −21.229
Kani’s Method or Rotation Contribution Method 373

Iteration process: Cycle 3

Joint B
Summation of the fixed end moments = −337.78
Far end rotation contributions
At A =0
At C = −86.235
Summation of displacement contributions
BC =0
BA = −21.229
   
= −445.244
 1
′ =   −  ( −445.244 )
MBA
 5

= 89.05 kNm
 3
′ =  −  ( −445.244 )
MBC
 10 

= 133.57 kNm
Joint C
Summation of the fixed end moments = 168.89
Far end rotation contributions
At B = 133.57
At D =0
Summation of displacement contributions
CB =0
CD = −21.229
  
= 281.231
 3
′ =   −  ( 281.231)
MCB
 10 

= −84.37 kNm
 1
′ =  −  ( 281.231)
MCD
 5

= −56.24 kNm
Story one (there is only one story)
Rotation contribution at top of column BA = 89.05
Rotation contribution at top of column CD = −56.24
Rotation contribution at bottom of columns =0
   
=32.81
 3
′′ =  −  ( 32.81)
M AB
 4

= −24.61
 3
′′ =  −  ( 32.81)
MCD
 4

= −24.61
374 Introduction to Structural Analysis

Iteration process: Cycle 4

Joint B
Summation of the fixed end moments = −337.78
Far end rotation contributions
At A =0
At C = −84.37
Summation of displacement contributions
BC =0
BA = −24.61
   
= −446.76
 1
′ =   −  ( −446.76)
MBA
 5

= 89.352 kNm
 3
′ =  −  ( −446.76)
MBC
 10 

= 134.03 kNm
Joint C
Summation of the fixed end moments = 168.89
Far end rotation contributions
At B = 134.03
At D =0
Summation of displacement contributions
CB =0
CD = −24.61
   
= 278.31
 3
′ =   −  ( 278.31)
MCB
 10 

= −83.49 kNm
 1
′ =  −  ( 278.31)
MCD
 5

= −55.66 kNm
Story one (there is only one story)
Rotation contribution at top of column BA = 89.352
Rotation contribution at top of column CD = −55.66
Rotation contribution at bottom of columns =0
   
= 33.692
 3
′′ =  −  ( 33.692)
M AB
 4

= −25.27
 3
′′ =  −  ( 33.692)
MCD
 4

= −25.27
Kani’s Method or Rotation Contribution Method 375

Iteration process: Cycle 5

Joint B
Summation of the fixed end moments = −337.78
Far end rotation contributions
At A =0
At C = −83.49
Summation of displacement contributions
BC =0
BA = −25.27
   
= −446.54
 1
′ =   −  ( −446.54 )
MBA
 5

= 89.308 kNm
 3
′ =  −  ( −446.54 )
MBC
 10 

= 133.96 kNm
Joint C
Summation of the fixed end moments = 168.89
Far end rotation contributions
At B = 133.96
At D =0
Summation of displacement contributions
CB =0
CD = −25.27
   
= 277.58
 3
′ =   −  ( 277.58)
MCB
 10 

= −83.274 kNm
 1
′ =  −  ( 277.58)
MCD
 5

= −55.52 kNm
Story one (there is only one story)
Rotation contribution at top of column BA = 89.308
Rotation contribution at top of column CD = −55.52
Rotation contribution at bottom of columns =0
   
= 33.788
 3
′′ =  −  ( 33.788)
M AB
 4

= −25.34
 3
′′ =  −  ( 33.788)
MCD
 4

= −25.34
The iteration process has been carried out up to the fifth cycle, as shown in
Figure 20.14 (a). After that, from the values of the fifth cycle, the final end moments
have been found, as shown in Figure 20.14 (b).
376 Introduction to Structural Analysis

FIGURE 20.14  Portal frame analysis by Kani’s method under vertical load. (a) The scheme
for proceeding with Kani’s method after Cycle 5 and (b) determining end moments by Kani’s
method.
Kani’s Method or Rotation Contribution Method 377

FIGURE 20.15  (a) Shear force diagram and (b) bending moment diagram of the portal
frame.

The shear force and bending moment diagrams have been given in
Figure 20.15 (a) and (b), respectively. Note that there is a slight deviation in the
values of bending moments as shown in the figure and with the values obtained
by Kani’s method after the fifth cycle. As we increase our number of cycles, we
will get closer to the values shown in the figure.
378 Introduction to Structural Analysis

20.4 ANALYSIS OF FRAMES WITH SIDESWAY WITH

VERTICAL LOADING AND HORIZONTAL
LOADING AT NODAL POINTS
Let us consider a multistoried frame subjected to horizontal loads applied to nodal
points along with the vertical loading, as shown in Figure 20.16. As was discussed
in Section 20.3, the sum of the shear forces in all the columns of a particular story
can be expressed as,

∑F H =−
∑ M HK + ∑ M KH
h

Let us consider fth story of a multistory frame. We can write, Sf = ∑ FH , where Sf is

the story shear of the fth story. Now we can rewrite the abovementioned expression as,

Sf = −
( ∑ M HK + ∑ M KH ) for the fth story
hf

where h f is the height of columns of the fth story, ∑ M HK is the sum of the end
moments at the upper ends of all the columns of the fth story, ∑ M KH is the sum of
the end moments at the lower ends of all the columns of the fth story.
Rearranging the abovementioned equation, we can write,

(∑ M + ∑ M ) = − S h
HK KH f f

FIGURE 20.16  Multistory frame subjected to horizontal loading at nodal points.

Kani’s Method or Rotation Contribution Method 379

The general expression of the end moments for a particular vertical member HK can
be written as:

M HK = M HK + 2 M HK
′ + M KH
′ + M HK
′′ = 2 M HK
′ + M KH
′ + M HK
′′

M KH = M KH + 2 M KH
′ + M HK
′ + M KH
′′ = 2 M KH
′ + M HK
′ + M KH
′′

Since M HK = M KH = 0 , because no external loads are acting in between the joints of

column HK.
Therefore,

M HK + M KH = 3 M HK
′ + 3 M KH
′ + 2 M HK
′′

i.e.,

∑M HK + ∑M KH =3 ∑M′ HK +3 ∑M′ KH ∑ M ′′
+2 HK = −S f hf

or,

∑ M ′′
2 HK = −S f hf − 3 ∑M′ HK −3 ∑M′ KH

or,

2 ∑ M ′′ HK
 S f hf
= −3 
 3
+ ∑M′ + ∑M′
HK KH



∴ ∑ M ′′ HK
3  S f hf
=− 
2 3
+ ∑M′ + ∑M′
HK KH

 (20.12)

For a particular story, the relative lateral displacement is the same for all the columns.
Let us assume the height of all the columns and Young’s modulus are also the same.
The displacement contribution for a column HK of the fth story,

6 EI∆ 6 E∆ I 6 E∆
′′ =
M HK = × = K HK
hf 2 hf hf hf

i.e.,
′′ ∝ K HK
M HK

So, we can say the displacement contribution of a particular column is proportional

to its relative stiffness.
Therefore,

′′
M HK K HK
=
∑ M HK
′′ ∑ K HK

′′ =
M HK
K HK
∑ K HK ∑ M ′′ HK
380 Introduction to Structural Analysis

Now, with the help of this expression, and using equation (20.12), we get,

′′ =
M HK
K HK
∑ K HK
 − 3   S f hf +
 
2  3 ∑M′ + ∑M′
HK KH

 (20.13)

M HK
 S f hf
′′ = γ HK 
 3
+ ∑M′ + ∑M′HK KH

 (20.14)

where γ HK = ( K HK / ∑ K HK )( −3/2 ) = displacement factor for HK, and the quantity

S f h f /3 is called the story moment. Let us discuss Example 20.4 to clarify the con-
cept just discussed.

Example 20.4: Analyze the portal frame under the influence of

horizontal load applied at joint B as shown in Figure 20.17. Assume all
the members have the same flexural rigidity.

SOLUTION:  Fixed end moments

M AB = MBA = MCD = MDC = MBC = MCB = 0

Rotation factors
Rotation factors are calculated using Table 20.5.
Storey shear

Sf = 15 kN

Storey moment

Sf hf 15 × 7
= = 35 kNm
3 3

FIGURE 20.17  Example problem on the portal frame under the influence of horizontal
nodal load.
Kani’s Method or Rotation Contribution Method 381

TABLE 20.5
Calculating Rotation Factors for the Given Portal Frame under
Horizontal Nodal Load
Rotation Factor,
Relative Total Relative
 1 K 
Joint Member Stiffness, K Stiffness, ∑ K µ = −
 2 ∑ K 
BA I 1
−
7 2I 4
B
BC I 7 1
−
7 4
CB I 1
−
C 7 2I 4
CD I 7 1
−
7 4

TABLE 20.6
Calculating Displacement Factors for the Given Portal Frame
Displacement Factor,
Vertical Relative Total Relative
 3 K 
Member Stiffness, K Stiffness, ∑ K γ = −
 2 ∑ K 
AB I 3
−
7 2I 4
DC I 7 3
−
7 4

Displacement factors
Displacement factors are calculated using Table 20.6.
Iteration: Cycle 1

Rotation Contributions
Joint B Joint C
Summation of the 0 Summation of the 0
fixed end moments fixed end moments
Far end rotation Far end rotation
contributions contributions
  At A ( M AB
′ ) 0   At B ( MBC′ ) 0 (assumed)
  At C ( MCB′ ) 0 (assumed)   At D ( MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC ( MBC
′′ ) 0   CB ( MCB
′′ ) 0
  BA ( MBA
′′ ) 0 (assumed)   CD ( MCD′′ ) 0 (assumed)
Total 0 Total 0
382 Introduction to Structural Analysis

Rotation Contributions
Joint B Joint C

′
MBA  1 ′
MCB  1
′ =  −  (0)
MBA ′ =  −  (0)
MCB
= µBA {∑ M B
 4
= 0 = MBC
′
= µCB {∑ M C
 4
= 0 = MCD
′
+ ∑ M′ jB + ∑ M ′jC

+ ∑ M ′′ }
Bj
+ ∑ M ′′ } Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of columns
′ )
  For AB (MBA 0
  For DC (MCD′ ) 0
Rotation contributions at the bottom of columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Sf hf ( ) 3
35
Total 35

MBA
Sh
′′ = γ BA  f f +
 3 ∑ M′ + ∑ M′
BA AB

 MBA
 3
′′ =  −  ( 35) = −26.25 = MCD
 4
′′

Iteration: Cycle 2

Rotation Contributions
Joint B Joint C
Summation of the fixed end 0 Summation of the 0
moments fixed end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB′ ) 0   At B (MBC ′ ) +6.56
  At C (MCB ′ ) 0   At D (MDC ′ ) 0
Displacement contributions Displacement
  BC (MBC′′ ) 0 contributions
  BA (M BA ′′ ) −26.25   CB (MCB′′ ) 0
  CD (MCD ′′ ) −26.25
Total −26.25 Total −19.69
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
 1
=  −  ( −26.25)
 4
= µCB {∑ M C
 1
=  −  ( −19.69 )
 4
+ ∑ M′ jB = +6.56 = MBC
′ + ∑ M′ jC = + 4.92 = MCD
′

+ ∑ M ′′ }
Bj + ∑ M ′′ } Cj
j = A ,C j = B ,D
Kani’s Method or Rotation Contribution Method 383

Displacement Contributions
Rotation contributions at the top of columns
′ )
  For AB (MBA +6.56
  For DC (MCD′ ) +4.92
Rotation contributions at the bottom of columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment ( )
Sf hf
3
35
Total 46.48

MBA
Sh
′′ = γ BA  f f +
 3 ∑ M′ + ∑
BA ′ 
M AB


MBA
 3
′′ =  −  ( 46.48 ) = −34.86 = MCD
 4
′′

Iteration: Cycle 3

Rotation Contributions
Joint B Joint C
Summation of the 0 Summation of the 0
fixed end moments fixed end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB′ ) 0   At B (MBC ′ ) +7.485
  At C (MCB ′ ) +4.92   At D (MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC (MBC′′ ) 0   CB (MCB′′ ) 0
  BA (MBA′′ ) −34.86   CD (MCD ′′ ) −34.86
Total −29.94 Total −27.375
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
 1
=  −  ( −29.94 )
 4
= µCB {∑ M C
 1
=  −  ( −27.375)
 4
+ ∑ M′ jB = +7.485 = MBC
′ + ∑ M′ jC = +6.84 = MCD
′

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of columns
′ )
  For AB (MBA +7.485
  For DC (MCD′ ) +6.84
Rotation contributions at the bottom of columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Sf hf ( ) 3
35
Total 49.33

MBA
Sh
′′ = γ BA  f f +
 3 ∑ M′ + ∑
BA ′ 
M AB


MBA
 3
′′ =  −  ( 49.33) = −36.99 = MCD
 4
′′
384 Introduction to Structural Analysis

Iteration: Cycle 4

Rotation Contributions
Joint B Joint C
Summation of the fixed 0 Summation of the fixed 0
end moments end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB ′ ) 0   At B (MBC ′ ) +7.54
  At C (MCB ′ ) +6.84   At D (MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC (MBC′′ ) 0   CB (MCB′′ ) 0
  BA (MBA ′′ ) −36.99   CD (MCD ′′ ) −36.99
Total −30.15 Total −29.45
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
 1
=  −  ( −30.15)
 4
= µCB {∑ M C
 1
=  −  ( −29.45)
 4
+ ∑ M′ jB = +7.54 = MBC
′ + ∑ M′ jC = +7.36 = MCD
′

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of columns
′ )
  For AB (MBA +7.54
  For DC (MCD′ ) +7.36
Rotation contributions at the bottom of columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Sf hf
( ) 3
35
Total 49.9

MBA
Sh
′′ = γ BA  f f +
 3 ∑ M′ + ∑
BA ′ 
M AB


MBA
 3
′′ =  −  ( 49.9 ) = −37.425 = MCD
 4
′′

Iteration: Cycle 5

Rotation Contributions
Joint B Joint C
Summation of the 0 Summation of the 0
fixed end moments fixed end moments
Far end rotation Far end rotation
contributions contributions
′ )
  At A (M AB 0 ′ )
  At B (MBC +7.52
′ )
  At C (MCB +7.36   At D (MDC′ ) 0
Kani’s Method or Rotation Contribution Method 385

Rotation Contributions
Joint B Joint C
Displacement Displacement
contributions contributions
′′ )
  BC (MBC 0   CB (MCB′′ ) 0
  BA (MBA′′ ) −37.425   CD (MCD ′′ ) −37.425
Total −30.1 Total −29.91
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
 1
=  −  ( −30.1)
 4
= µCB {∑ M C
 1
=  −  ( −29.91)
 4
+ ∑ M′ jB = +7.52 = MBC′ + ∑ M′ jC = +7.48 = MCD
′

+ ∑ M ′′ }Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of columns
′ )
  For AB (MBA +7.52
  For DC (MCD′ ) +7.48
Rotation contributions at the bottom of columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Sf hf ( ) 3
35
Total 50

MBA
Sh
′′ = γ BA  f f +
 3 ∑ M′ + ∑ M′
BA AB

 MBA
 3
′′ =  −  ( 50 ) = −37.5 = MCD
 4
′′

After performing the fifth cycle, it was found that the rotation and displacement
contribution values are getting converged, i.e., no more cycle is required. So, the
values obtained in the fifth cycle are used to calculate the final end moments. The
obtained values from all the iterations for rotation and displacement contribution
are given in Table 20.7.
The end moments have been calculated as shown in Figure 20.18.
The shear force and bending moment diagrams for the given portal frame are
shown in Figure 20.19 (a) and (b), respectively.

TABLE 20.7
Values of Rotation and Displacement Contributions for All the Iterations
Cycle ′
MBA ′
MBC ′
MCB ′
MCD ′′
MBA ′′
MCD
First cycle  0  0  0  0 −26.25 −26.25
Second cycle +6.56 +6.56 +4.92 +4.92 −34.86 −34.86
Third cycle +7.485 +7.485 +6.84 +6.84 −36.99 −36.99
Fourth cycle +7.54 +7.54 +7.36 +7.36 −37.425 −37.425
Fifth cycle +7.52 +7.52 +7.48 +7.48 −37.5 −37.5
386 Introduction to Structural Analysis

FIGURE 20.18  Determining end moments by Kani’s method.

FIGURE 20.19  (a) Shear force diagram and (b) bending moment diagram of the portal
frame. (Continued)
Kani’s Method or Rotation Contribution Method 387

FIGURE 20.19  (Continued)

20.5 ANALYSIS OF FRAMES WITH COLUMNS

WITH UNEQUAL HEIGHT
Let us consider the frame with unequal columns as shown in Figure 20.20 with a
horizontal load P acting at node A.
As the column is unsymmetrical both in geometry and loading conditions, it will
sway toward the right in the direction of the applied force. Due to equilibrium, base
shear forces drawn near each column’s supports will be developed, which will coun-
teract the effect of the horizontal external load P. Under this situation, the base shear
in each column can be written as:

( M AB + M BA )
H AB =  
h1

Likewise,

HCD =  
( MCD + M DC )
h2
388 Introduction to Structural Analysis

FIGURE 20.20  Frame with sidesway and unequal columns.

( M EF + M FE )
H EF =  
h3

HGH =  
( M GH + M HG )
h4

Following horizontal force equilibrium conditions, we can write:

P+ ∑H = 0
where ∑ H is the summation of all horizontal support reactions, i.e., total horizontal
shear in all columns of a particular story, as shown in Figure 20.20.
Thus,

P+ ∑ Mh = 0
where ∑ M is the sum of the end moments of all the individual columns in a
particular story as written above, and h is the corresponding length of those
columns.
Kani’s Method or Rotation Contribution Method 389

Now, if we take any reference height of a column, say hr, and which is the maxi-
mum height as well of all the columns, we can modify the abovementioned equations
as follows:

( M AB + M BA ) hr ( M AB + M BA )
H AB = × = C AB
hr h1 hr


HGH =
( M GH + M HG ) × hr = ( M GH + M HG ) C
GH
hr h4 hr

Now we can substitute these values back in the force equilibrium equations for mem-
ber AB to get:

M AB + M BA
P+ C AB = 0
hr

1
P+
hr
( M AB + 2 M AB′ + M BA′ + M AB′′ + M BA + 2 M BA′ + M AB′ + M BA′′ )C AB = 0
or,
1
P+ ( 2 M AB
′ + M BA
′ + M AB
′′ + 2 M BA
′ + M AB ′′ ) C AB = 0
′ + M BA
hr

Since M AB = M BA = 0 , as no external loads are applied in between the joints.

1
P+ ( 3M AB
′ + 3 M BA ′′ ) C AB = 0
′ + 2 M AB
hr

or,

3 Ph
′′ C AB = −    r − ( M AB
M AB ′ C AB ) 
′ C AB + M BA
2  3 

For all columns, we get:

∑ M ′′ CAB AB
3 Ph
= −    r −
2  3 ∑( M ′ C AB AB ′ C AB )  (20.15)
+ M BA


Now, moment generated at the ends of AB column due to sway of δ amount:

6 EI ABδ 6 Eδ I AB K
′′ =
M AB =   = C AB (20.16)
hAB 2 hAB hAB hAB
390 Introduction to Structural Analysis

where C = 6 Eδ . Furthermore, we can expand the abovementioned equation as

follows:

hr K h
′′
M AB = C × AB × r
hAB hAB hAB

or,

C h2
M ′′ABC AB = K AB   2r
hr hAB

or,

C
′′ C AB =
M AB 2
K ABC AB
hr

So, summing for all the columns, we get:

∑ M ′′ C AB AB =
C
hr ∑K AB
2
C AB

or,

∑ M AB
′′ C AB
C = hr (20.17)
∑ K ABC AB
2

Again writing equation (20.16),

K AB
′′ = C
M AB
hAB

     ∑ M AB
′′ C AB K AB
= hr ×
∑ K ABC AB
2
hAB

     h ∑ M AB
′′ C AB
= K AB r ×
hAB ∑ K ABC AB
2

′′ =
∴ M AB
K ABC AB
∑ K ABC AB
2
× ∑ M ′′ C AB AB (20.18)
Kani’s Method or Rotation Contribution Method 391

Substituting, the value of ∑ M AB

′′ C AB from equation (20.15) into equation (20.18),

′′ = −
∴ M AB
K ABC AB 3 Ph
×    r −
∑ K ABC AB 2  3
2 ∑( M ′ CAB AB ′ C AB ) 
+ M BA


Thus, the displacement coefficient is given by:

3 K C
γ AB = −   AB AB2
2 ∑ K ABC AB

And story moment is given by:

Phr

3

If there is no horizontal force acting on the frame system,

′′ = −
M AB
K ABC AB
∑ K ABC AB
2
3
×   
2  ∑( M ′ C
AB AB ′ C AB ) 
+ M BA


If heights of all columns are identical, then:

C AB = CCD = CEF = CGH = 1

In such case, the abovementioned equation becomes:

M AB
3 K  
′′ = −   AB
2 ∑ K AB ∑( M ′
AB ′ )
+ M BA

Let us discuss Example 20.5 to clarify the concept just discussed.

Example 20.5: Analyze the frame shown in Figure 20.21 using Kani’s

method.

SOLUTION:  Fixed end moments:

wl 2
MBC = − = −3 kNm
12

wl 2
MCB = = 3 kNm
12
392 Introduction to Structural Analysis

FIGURE 20.21  Example problem of frame system with loading.

Rotation contributions at different joints are as follows:

1 2I /6
µBA =   −   = −0.2
2 2I /6 + 3I /6

1 3I /6
µBC = −   = −0.3
2 2I /6 + 3I /6

1 3I /6
µCB =   −   =   −0.3
2 2I /6 + 3I /6

1 I /3
µCD =   −   =   −0.2
2 I /3 + 3I /6

Now, let us consider the height, hr = 6 m. Then, we get:

hr
C AB =   =1
hAB

hr
CCD = =2
hCD

Displacement factors are given by:

3 K C 3 2I /6 × 1
γ AB = −   AB AB2 = −   = −0.3
2 ∑ K ABC AB 2 2I /6 × 12 + I /3 × 0.52

3 I /3 × 0.5
γ CD = −   = −0.6
2 2I /6 × 12 + I /3 × 0.52
Kani’s Method or Rotation Contribution Method 393

Iteration: Cycle 1

Rotation Contributions
Joint B Joint C
Summation of the −3 Summation of the 3
fixed end moments fixed end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB ′ ) 0   At B (MBC ′ ) 0.9
  At C (MCB ′ ) 0 (assumed)   At D (MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC (MBC′′ ) 0   CB (MCB′′ ) 0
  BA (MBA ′′ ) 0 (assumed)   CD (MCD ′′ ) 0 (assumed)
Total −3 Total 3.9
′
MBA ′ = ( −0.2)( −3)
MBA ′
MCB ′ = ( −0.3)( 3.9)
MCB
= µBA {∑ M B
= 0.6 = µCB {∑ M C
= −1.17

+ ∑ M′ jB + ∑ M′ jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

′
MBC ′ = ( −0.3)( −3)
MBC ′
MCD ′ = ( −0.2)( 3.9)
MCD
= µBC {∑ M B
= 0.9 = µCD {∑ M C
= −0.78

+ ∑ M′ jB + ∑ M′ jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of columns
′ )
  For AB (MBA 0.6
  For DC (MCD′ ) −0.78
Rotation contributions at the bottom of
columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Phr ( ) 3
0
 Ph
′ = γ BA  r +
MBA ∑ (M′ BA

′ ) C AB 
+ M AB MBA {
′′ = ( −0.3)  ( 0.6 + 0 ) × 1 }
 3 
+ { }
( −0.78 + 0) × 2  = 0.288
 Ph
′′ = γ CD  r +
MCD ∑ (M′ CD

′ ) CCD 
+ MDC MCD {
′′ = ( −0.6)  ( 0.6 + 0 ) × 1 }
 3 
+ {( −0.78 + 0) × 2} = 0.576
394 Introduction to Structural Analysis

Iteration: Cycle 2

Rotation Contributions
Joint B Joint C
Summation of the fixed −3 Summation of the fixed 3
end moments end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB′ ) 0   At B (MBC ′ ) 1.165
  At C (MCB ′ ) −1.17   At D (MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC (MBC′′ ) 0   CB (MCB′′ ) 0
  BA (MBA′′ ) 0.288   CD (MCD ′′ ) 0.576
Total −3.882 Total 4.741
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
= ( −0.2)( −3.882)
= 0.7764
= µCB {∑ M C
= ( −0.3)( 4.741)
= −1.4222
+ ∑ M′ jB + ∑ M′ jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

′
MBC ′
MBC ′
MCD ′
MCD

= µBC {∑ M B
= ( −0.3)( −3.882)
= 1.165
= µCD {∑ M C
= ( −0.2)( 4.741)
= −0.9482
+ ∑ M ′jB + ∑ M ′jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of
columns
′ )
  For AB (MBA 0.7764
  For DC (MCD′ ) −0.9482
Rotation contributions at the bottom of
columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Phr
( ) 3
0
 Ph
′′ = γ BA  r +
MBA ∑ (M′ BA

′ ) C AB 
+ M AB MBA {
′′ = ( −0.3)  ( 0.7764 + 0 ) × 1 }
 3 
+ {( −0.9482 + 0) × 2} = 0.336
 Ph
′′ = γ CD  r +
MCD ∑ (M′CD

′ ) CCD 
+ MDC MCD {
′′ = ( −0.6)  ( 0.7764 + 0 ) × 1 }
 3 
+ {( −0.9482 + 0) × 2} = 0.672
Kani’s Method or Rotation Contribution Method 395

Iteration: Cycle 3

Rotation Contributions
Joint B Joint C
Summation of the −3 Summation of the 3
fixed end moments fixed end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB′ ) 0   At B (MBC ′ ) 1.226
  At C ((MCB ′ ) −1.4222   At D (MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC (MBC′′ ) 0   CB (MCB′′ ) 0
  BA ((MBA ′′ ) 0.336   CD (MCD ′′ ) 0.672
Total −4.0862 Total 4.898
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
= ( −0.2)( −4.0862)
= 0.81724
= µCB {∑ M C
= ( −0.3)( 4.898)
= −1.4694
+ ∑ M′ jB + ∑ M′ jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

′
MBC ′
MBC ′
MCD ′
MCD

= µBC {∑ M B
= ( −0.3)( −4.0862)
= 1.226
= µCD {∑ M C
= ( −0.2)( 4.898)
= −0.9796
+ ∑ M′ jB + ∑ M′ jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of
columns
′ )
  For AB (MBA 0.81724
  For DC (MCD′ ) −0.9796
Rotation contributions at the bottom of
columns
′ )
  For AB (M AB 0
  For DC (MDC′ ) 0
Story moment Phr ( ) 3
0
 Ph
′′ = γ BA  r +
MBA ∑ (M′ BA

′ ) C AB 
+ M AB MBA {
′′ = ( −0.3)  ( 0.81724 + 0 ) × 1 }
 3 
+ { }
( −0.9796 + 0) × 2  = 0.3426
 Ph
′′ = γ CD  r +
MCD ∑ (M′CD

′ ) CCD 
+ MDC MCD {
′′ = ( −0.6)  ( 0.81724 + 0 ) × 1 }
 3 
+ {( −0.9796 + 0) × 2} = 0.6852
396 Introduction to Structural Analysis

Iteration: Cycle 4

Rotation Contributions
Joint B Joint C
Summation of the −3 Summation of the 3
fixed end moments fixed end moments
Far end rotation Far end rotation
contributions contributions
  At A (M AB′ ) 0   At B (MBC′ ) 1.238
  At C (MCB ′ ) −1.4694   At D (MDC ′ ) 0
Displacement Displacement
contributions contributions
  BC (MBC′′ ) 0   CB (MCB′′ ) 0
  BA ((MBA ′′ ) 0.3426   CD ((MCD ′′ ) 0.6852
Total −4.127 Total 4.9232
′
MBA ′
MBA ′
MCB ′
MCB

= µBA {∑ M B
= ( −0.2)( −4.127)
= 0.8254
= µCB {∑ M C
= ( −0.3)( 4.9232)
= −1.477
+ ∑ M′ jB + ∑ M′ jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

′
MBC ′
MBC ′
MCD ′
MCD

= µBC {∑ M B
= ( −0.3)( −4.127)
= 1.238
= µCD {∑ M C
= ( −0.2)( 4.9232)
= −0.9846
+ ∑ M ′jB + ∑ M ′jC

+ ∑ M ′′ }
Bj + ∑ M ′′ }
Cj
j = A ,C j = B ,D

Displacement Contributions
Rotation contributions at the top of
columns
′ )
For AB (MBA 0.8254
For DC (MCD′ ) −0.9846
Rotation contributions at the bottom of
columns
′ )
For AB (M AB 0
For DC (MDC′ ) 0
Story moment Phr
( ) 3
0
Total 0.00
 Ph
′′ = γ BA  r +
MBA ∑ (MBA 
′ ) C AB 
′ + M AB MBA {
′′ = ( −0.3)  ( 0.8254 + 0 ) × 1 }
 3 
+ {( −0.9846 + 0) × 2} = 0.3431
 Ph
′′ = γ CD  r +
MCD ∑ (M′ CD

′ ) CCD 
+ MDC MCD {
′′ = ( −0.6)  ( 0.81724 + 0 ) × 1 }
 3 
+ {( −0.9796 + 0) × 2} = 0.6863
Kani’s Method or Rotation Contribution Method 397

TABLE 20.8
Values of Rotation and Displacement Contributions
for All the Iterations
Cycle ′
MBA MBC′ ′
MCB ′
MCD ′′
MBA ′′
MCD
First cycle 0.6 0.9 −1.17 −0.78 0.288 0.576
Second cycle 0.7764 1.165 −1.4222 −0.9482 0.336 0.672
Third cycle 0.81724 1.226 −1.4694 −0.9796 0.3426 0.6852
Fourth cycle 0.8254 1.238 −1.477 −0.9846 0.3431 0.6863

After performing the fourth cycle, it was found that the rotation and displacement
contribution values are getting converged, i.e., no more cycle is required. So, the
values obtained in the fourth cycle are used to calculate the final end moments.
The obtained values from all the iterations for rotation and displacement contribu-
tion are given in Table 20.8.
The end moments have been calculated as shown in Figure 20.22.
The shear force and bending moment diagrams are shown in Figure 20.23 (a)
and (b), respectively.

FIGURE 20.22  Determining end moments by Kani’s method.

398 Introduction to Structural Analysis

FIGURE 20.23  (a) Shear force diagram and (b) bending moment diagram of the portal
frame.
 21 Column Analogy Method

21.1 INTRODUCTION
The basic concept of the column analogy method circles around analyzing the stati-
cally indeterminate beams of single span, portal frames, closed box frames with
some assumptions related to the given properties of that structural element. We adopt
this method to determine the redundant force or moments of statically indeterminate
structural elements more easily than all other methods we have learned so far. This
method provides the exact value of the redundant force and moment in contrast to
other approximate methods like moment distribution. This method is based on the
similarity between the moments induced in a statically indeterminate structure, and
the stresses produced in an eccentrically loaded short column.

21.2  BASIC CONCEPT

We begin with the concept of stress for eccentrically loaded short columns. For such
columns, stress at any point is linearly proportional to the distance of the point from
the point of application of load (Figure 21.1).
As shown in Figure 21.1, stress ( f ) at any point A with coordinates ( x , y), can be
written as:

f = a + bx + cy

where a, b, c are constants.

So, for any small elemental area δ A near point A, the force acting in that area
will be:

δ P = fδ A

So, the total force acting on the column section can be expressed as:

∫
P =  δ P =
∫ fδ A = ∫ f dA
or,

P=
∫ ( a + bx + cy) dA

∫ ∫
P = a dA + b xdA + c ydA
∫
DOI: 10.1201/9781003081227-24 399
400 Introduction to Structural Analysis

FIGURE 21.1  Short column section with eccentric loading.

If the reference axis passes through the center of gravity of the section, the point O
coincides with center of gravity.

∫ xdA = 0 =   ∫ ydA
So, in that case, we are left with:

∫
P = a dA = aA

where A is the total area of cross section of the column.

So,
P
a=
A
The moment of the elementary force δ P about x axis is given by:

δ M x = yδ P

So, the total moment:

Mx =
∫ ydP = ∫ yf dA = ∫ y ( a + bx + cy) dA
or,

∫ ∫ ∫
M x = a ydA + b xydA + c y 2 dA = 0 + bI xy + cI x (21.1)

where I xy is the cross moment of inertia and I x is the moment of inertia about x axis.
Similarly, for moment about y axis, we get:

My =
∫ xdP = ∫ x f dA = ∫ x ( a + bx + cy) dA
Column Analogy Method 401

or,

M y = 0 + bI y + cI xy (21.2)

So, solving equations (21.1) and (21.2), we get:

M y I x − M x I xy
b = 
I y I x − I xy2
M x I y − M y I xy
c = 
I y I x − I xy2

Hence, by substituting these values in the original stress equation, we can determine
the full equation for stress at any point of the column. Moreover, it is customary to
note that if the reference axis is the principal axis, I xy = 0.

P My M
f= +  x+ x y
A Iy Ix

Now we set our attention toward the bending of a curved beam, CB as shown in
Figure 21.2.
In the following curved beam CB, the CB portion has been rotated through an
angle δθ at point C. Due to this rotation, point B has been shifted to new location B′.
Also, the point T assumes the new location T′. The angle TCT′ is δθ . Let δ V and δ H
are vertical and horizontal displacements of the point B′ with respect to B. Let angle
BTC is ϕ . So, angle B′ T ′ C will also be ϕ . Draw a line parallel to B′ T ′ at B. Let
the angle between this line and BT be δϕ . δϕ will be equal to δθ .
Since all the angles are infinitesimally small,
BB ′ = BC δθ
So,
δ V = BB ′ cos θ

FIGURE 21.2  Bending of a curved beam.

402 Introduction to Structural Analysis

or,
δ V = BC cos θ  δθ
Implies,
δ V = xδθ
Similarly,

δ H = BB ′ sin θ = BC  sin θ  δθ = yδθ

Considering the beam as a rigid body excepting a small length δ s at C, the change
in slope at C will be equal to Mδ s /EI , where M is the bending moment of the beam
at point C.
Mδ s
The change in slope at B will be .
EI
Thus, we get the following relationships:

Mδ s
δϕ =  
EI
Mxδ s
δV = 
EI
Myδ s
δH =
EI

Now we are ready to learn the column analogy method that will be developed in the
following section.

21.3  DEVELOPMENT OF THE COLUMN ANALOGY METHOD

Indeterminate structure consists of two types of bending moments, one moment (M i )
considering the structure as indeterminate one, and the other will be the moment
(M s ) in the primary structure by removing the redundancy from the structure. Hence,
the total moment can be written as:

M = M s + Mi

Let θ be the relative rotation of the ends of the structure, H be the relative horizontal
displacement of the ends, and V be the relative vertical displacement of the ends.

Mds
θ=
∫ EI
Myds
H=
∫ EI
Mxds
V=
∫ EI
Column Analogy Method 403

In case the relative rotation and displacements at the ends are zero:
Mds
θ =0=
∫ EI
or,
( M s + M i ) ds
θ =0=
∫ EI
So,
( M i ) ds = − ( M s ) ds

∫ EI ∫ EI
Similarly,
Myds ( M i y ) ds = − ( M s ) ds y
H=0=
∫ EI
⇒
∫ EI ∫ EI
Also,
Mxds ( M i x ) ds = − ( M s ) ds x
V =0=
∫ EI
⇒
∫ EI ∫ EI

Now, let us consider a short column of width 1/EI , and the load intensity is − M s .

M s ds
So, the total load, P = −
So,
∫ EI

M i ds

∫ EI
=P

or,
M i ds
dP =  
EI
or,
M i yds

∫ EI
=
∫ ydP
M i xds

∫ EI
=
∫ xdP
Now, comparing these values with the stress values calculated for column with
eccentric load, we get:

∫ f dA = P

∫ fydA = M = ∫ ydP x

∫ fxdA = M = ∫ xdP y
404 Introduction to Structural Analysis

Analogously, for the imagined column, ds /EI will be the elementary area, f will be
the stress at any point due to load intensity − M s or the total load − ∫ M s ds /EI .
So, we have got:

P M y I x − M x I xy M x I y − M y I xy
Mi = + ×x+ ×y
A I y I x − I xy
2
I y I x − I xy2

or,

P M y − M x ( I xy /I x ) M x − M y ( I xy /I y )
Mi = + ×x+ ×y
 (
A I y 1 − I xy /I x I y 
2
 ) I x  (
1 − I xy2 /I y I x 
 )
Putting,

I xy
M y′ = M y − M x
Ix

And,

I xy
M x′ = M x − M y
Iy

 I xy2 
I y′ =   I y  1 −
 I x I y 

 I xy2 
I x′ = I x  1 −
 I y I x 

Hence, the final bending moment equation becomes:

P M y′ M′
Mi = + ×x+ x ×y
A I y′ I x′

M s will be positive if it induces tension in the inside fibers and it will be negative
if it induces tension in the outer fibers. For positive M s , P will be tensile and nega-
tive, and for negative M s , P will be compressive and positive. M i is positive if f is
compressive. Also, for any structure, if the support is hinged, it does not offer any
resistance to rotation and can take any rotation. The flexural rigidity EI at a simply
supported end is zero, and hence, 1/EI is taken as infinite, and for a fixed support,
there will be no rotations and the flexural rigidity EI is infinite; hence, 1/EI is taken
as zero. The thickness of the load diagram is the same as the analogous column sec-
tion. So, if the beam is of uniform flexural rigidity, the thickness of the analogous
column and that of the load diagram on this column may be taken as unity, since the
quantity 1/EI cancels out.
Column Analogy Method 405

FIGURE 21.3  Example problem on column analogy method for fixed beam.

Example 21.1:  A fixed beam of span L carries a point load P eccentrically

on the span at a distance a from the left end and b from the right end
as shown in Figure 21.3(a). Find the fixed end moments at the ends of
the beam.

SOLUTION:  Let M AB and MBA be the fixed end moments at A and B, respectively,
and these are redundants. If we remove these two redundants, we will get the
basic determinate simply supported beam. For this simply supported beam, the Ms
diagram is shown in Figure 21.3 (b). The height of the Ms diagram is PabL .
The beam is having uniform flexural rigidity throughout its length. So, the thick-
ness of the Ms diagram and the thickness of the analogous column are taken as unity.

Total load on the analogous column = 12 L Pab

L × 1=
Pab
2

1
2 × a× Pab
L × 2a
3 + 21 × b × Pab
L × (a + b
3 )=
C.G. of the Ms diagram from A, = L+a
3
1
2 ×L× Pab
L

∴ Eccentricity of the load = ( L +3a ) − L

2 = 2a − L
6

Area of the analogous column section = L × 1 = L

1× L3 L3
Moment of inertia of the analogous column section = 12 = 12

Pab Pab
× 2a − L
Stress at any point of the column section, Mi = 2
± 2
L3
6
x
L 12
Pab( 2a − L )
or, Mi = Pab
2L ± L3
x
406 Introduction to Structural Analysis

Pab Pab ( 2a − L ) L Pab Pab 2

Stress at A, MiA = − 3
= 2 ( L − a) = 2
2L L 2 L L
Pab Pab ( 2a − L ) L Pab Pa2b
Stress at B, MiB = + 3
= 2 ( L + 2a − L ) = 2
2L L 2 2L L
Pab 2 Pab 2
Fixed end moment at A, M AB = MsA − MiA = 0 − 2
=− 2
L L
Pa2b Pa2b
Fixed end moment at B, MBA = MsB − MiB = 0 − 2
=− 2
L L

21.4 STIFFNESS AND CARRY OVER FACTORS DETERMINED

BY METHOD OF COLUMN ANALOGY
In simple words, stiffness is the value of moment to be applied at an end to cause
slope of one radian and carry over factor is the ratio of moment generated at the far
end due to the applied moment at the joint under investigation. Let us consider a fixed
beam of rigidity EI as shown in Figure 21.4. The loading on the analogous column
will be θ A at A and width of the column will be 1/EI as shown in Figure 21.4. For
the clockwise slope at A, the loading will be positive, and for clockwise slope at B,
loading will be negative.
So, for this analogous column, the area of cross section will be l × 1/EI . So,
we get:

P P×e× y
f =  + 
A Iy

FIGURE 21.4  Fixed beam and corresponding analogous column model.

Column Analogy Method 407

In the case of our analogous column model, we have P =  θ A ,  A = l /EI ,  e = l /2,

y = l /2,  I y = l 3 /12 EI .
So, substituting these values in the above equation, we get:

θA θ × ( l /2 ) × ( l /2 ) 4 EIθ A
f A = M AB = +  A 3 =
l /EI l /12 EI l

So,

4 EIθ A
M AB = M s + M i = 0 + f A = 0 +
l

Remembering the definition of stiffness, putting θ A = 1, we get:

4 EI
M AB =
l

This is in perfect agreement with the earlier derived stiffness factor for fixed beam
found in moment distribution chapter (Chapter 19).
Also, let us check the stress at support B:

P y
fB = −P×e×
A Iy

or,

θA 3EIθ A 2 EIθ A
fB = − =−
l /EI l l

So,

2 EIθ A 1
M BA = M s + M i = 0 + f B = − = − M AB
l 2

So, the carry over factor is 1/2 as we have already seen in the Chapter 19. So, all the
results by earlier methods as discussed can be derived by constructing an imaginary
column with geometric and loading properties as explained.

21.5  FIXED END MOMENTS DUE TO SUPPORT SETTLEMENT

Consider the same prismatic beam as drawn in Figure 21.5 with one exception. Let
the support at B settles by an amount ∆ with respect to support A. The slope of the
chord joining the supports A and B, at this settled condition, will be θ = ∆ /l .
408 Introduction to Structural Analysis

FIGURE 21.5  Fixed end moments due to settlement of supports.

So, following the earlier explained logic, the loading in the analogous column
model will be ∆ /l at B in downward direction and at A ∆ /l in the upward direction.
As usual, the width of this analogous column will be 1/EI . The net vertical load
acting on the column = P = 0.
So,
∆
My = ×l = ∆
l
P x ∆ × ( l /2 ) 6 EI∆
fB = + My × = 0 + 3 =+ 2
A Iy l /12 EI l
6 EI∆
M BA = 0 + f B =
l2
Similarly,
∆ × ( l /2 ) 6 EI∆
fA = 0 +  = 2
l 3 /12 EI l
So,
6 EI∆
M AB = 0 + f A =
l2
So, these are also in perfect agreement with the earlier derived result in Chapters 18
and 19.
Column Analogy Method 409

21.6  ANALYSIS OF PORTAL FRAMES

Continuing from the previous section, we will apply the method of column analogy to
analyze the portal frames. We will show this method through Example 21.2. Please
note this frame is not symmetrical due to loading and not due to its geometry.

Example 21.2:  Analyze the fixed portal frame shown in Figure 21.6.

SOLUTION:  Let us obtain a basic determinate structure as shown in Figure 21.7,

of the given portal frame with A as the roller end and D as the hinged end.

FIGURE 21.6  Portal frame example problem.

FIGURE 21.7  Determinate frame after removing the redundants.

410 Introduction to Structural Analysis

FIGURE 21.8  (a) M s diagram and (b) analogous column of the example problem.

Figure 21.8 (a) shows the Ms diagram, and Figure 21.8 (b) shows the analogous
column section. For AB the thickness of the section ( EI1 ) is taken equal to 2. For BC
and CD, the thickness of the section ( 21EI ) is taken equal to 1.
Load on the analogous column = Volume of the Ms diagram = 12 × 8 × 15 × 1 =
60 units.
This acts on BC at a distance 103 m from B.
Total area of the analogous column section, = ( 2 × 6) + (8 × 1) + (6 × 1) = 26 units
Let us take A as the origin. Let ( x , y ) be the coordinates of the centroid, with
respect to A.

x=
(8 × 4 ) + (6 × 8) = 3.08 m
26

y=
(12 × 3) + (8 × 6) + (6 × 3) = 3.92 m
26

2 × 63 1× 63
IBC = + = 216 units
3 3

I xx = IBC − A × 2.08 2 = 216 − 26 × (6 − 3.92) = 103.51 units

2

1× 83
I AB = + 6 × 82 = 554.67 units
3

Iyy = I AB − A × 3.08 2 = 554.67 − 26 × 3.082 = 308.02 units

I xy = A1x1y1 + A2x2y 2 + A3 x3y3

Column Analogy Method 411

or,

I xy = 12 × ( −3.08) × ( −0.92) + 8 × 0.92 × 2.08 + 6 × 4.92 × ( −0.92)

= 34.0 + 15.3 − 27.16 = 22.14 units

Mxx = P × ey = 60 × 2.08 = 124.8 units

 10 
Myy = P × ex = 60 ×  4 −  = 60 × 0.67 = 40 units
 3

Myy I xy −  MxxI xy ( 40 × 103.51) − (124.8 × 22.14 ) 1377.33

b =  = = = 0.044
Iyy I xx −  I xy 2 (308.02 × 103.51) − 22.14 2 31392.97

MxxIyy −  Myy I xy (124.8 × 308.02) − ( 40 × 27.16) 37354.5

c =  = = = 1.19
Iyy I xx −  I xy 2 (308.02 × 103.51) − 22.14 2 31392.97

Now, the stress at any point of the analogous column section,

P 60
Mi = + bx + cy = + 0.044 x + 1.19y = 2.31 + 0.044 x + 1.19y
A 26

Therefore,
Stress at A, (−3.08, −3.92)

MiA = 2.31+ 0.044 ( −3.08) + 1.19 ( −3.92) = −2.49 kNm

Stress at D, (4.92, −3.92)

MiD = 2.31+ 0.044 ( 4.92) + 1.19 ( −3.92) = −2.14 kNm

Stress at B, (−3.08, 2.08)

MiB = 2.31+ 0.044 ( −3.08) + 1.19 ( 2.08) = +4.65 kNm

Stress at C, (4.92, 2.08)

MiB = 2.31+ 0.044 ( 4.92) + 1.19 ( 2.08) = +5.0 kNm

Final Moments:

M A = MsA − MiA = 0 − ( −2.49) = +2.49 kNm

MD = MsD − MiD = 0 − ( −2.14 ) = +2.14 kNm

412 Introduction to Structural Analysis

FIGURE 21.9  Final bending moment diagram of the portal frame.

MB = MsB − MiB = 0 − ( 4.65) = −4.65 kNm

MC = MsC − MiC = 0 − ( 5.0 ) = −5.0 kNm

Figure 21.9 shows the bending moment diagram of the portal frame.
22 Beams and Frames
Having Nonprismatic
Members

22.1 INTRODUCTION
In this chapter, we will apply previously acquired knowledge of slope deflection and
moment distribution methods to analyze beams and frames composed of nonpris-
matic members. At first, how the necessary carry-over factors, stiffness factors, and
fixed end moments are obtained will be discussed in detail. Finally, the analysis of
statically indeterminate structures using the slope deflection and moment distribu-
tion methods will be discussed in detail.
Nonprismatic members are being used in recent large span structures like factory
shades and large ceremony hall canopies. Modern days pre-engineered buildings
are mostly used for storage sheds and various other industrial applications. In these
structures, nonprismatic sections are used instead of making large truss members for
holding the roof structure. Applications of these nonprismatic sections provide more
aesthetic views and open spaces for other internal work installations.

22.2 DEFLECTIONS AND LOADING PROPERTIES

OF NONPRISMATIC MEMBERS
As stated above, for economic design, girders used for long spans (in bridges or
industrial sheds) are designed to be nonprismatic, that is, to have a variable moment
of inertia. The most common form of nonprismatic member is the tapered section
that is having more depth near the support and lesser depth at the crown portion.
The depth of section is decided based on the detailed analysis and bending moment
and shear force acting on the member at critical sections. If we can express the
moment of inertia of these sections as function of x coordinates along the length of
the member then we can use virtual work or Castigliano’s methods to calculate the
deflections. For recollection of earlier learned equations, deflection is calculated by
applying the following:
l
∂M M
∆=
∫ ∂P   EI dx
0

In many cases due to irregular geometry, the exact solution of the above integral is
not possible. In those cases, various numerical methods and approximation tech-
niques may be adopted to calculate the approximate value of deflection.
DOI: 10.1201/9781003081227-25 413
414 Introduction to Structural Analysis

FIGURE 22.1  Stiffness coefficients and angular displacements.

To apply slope deflection or moment distribution methods, we need to quantify

the following structural properties:

1. Fixed end moments – Assuming that supports are fixed, fixed end moments
need to be calculated first for the given load applied on the same.
2. Stiffness factor (K) – The moment that is needed to be applied at the end of
a member to make unit rotation at the end is called stiffness factor.
3. Carry over factor – It gives the amount of moment transferred from pin
supported end to the fixed end of a structure.

There exists an important relationship between carry over factor and stiffness of
structural elements. To understand the relationship, see Figure 22.1.
From the above diagram, we can apply Maxwell-Betti reciprocal theorem that
stipulates that work done by loads in the first diagram with the displacement in the
second diagram should be equal to the work done by the second diagram forces with
the displacements in the first diagram. So, in short,
U AB = U BA
or,
K A ( 0 ) + C AB K A (1) = C BA K B (1) + K B ( 0 )
or,
C AB K A = C BA K B
Although the relationship has been formed quite comfortably, determining numeri-
cal values of the above factors often involves considerable labor and efforts. To over-
come these computational issues, design tables and graphs are often available in
many standard books (see ‘Bibliography’) for ready reference. Most commonly used
charts and tables are available in the Handbook of Frame Constants published by
Portland Cement Association.+

22.3 MOMENT DISTRIBUTION FOR STRUCTURES

HAVING NONPRISMATIC MEMBERS
After determining the fixed end moments, stiffness, and carry over factors for the
nonprismatic element of a structure, application of the moment distribution method
can be applied following the same procedure as outlined in Chapter 19. In this
Beams and Frames Having Nonprismatic Members 415

regard, remember that the distribution of moments may be shortened if a member

stiffness factor is modified due to conditions of end-span pin support and structure
symmetry or antisymmetry. Similar modifications are also required to be made to
nonprismatic members.

22.3.1  Beam Pin Supported at Far End

Let us consider the beam in Figure 22.2, which is pinned at its far end B. The abso-
lute stiffness factor is the moment applied at A such that it rotates the beam at A, this
can be calculated as follows. At first, let us assume that B is fixed end support, and a
moment is applied at support A, as shown in Figure 22.2 (b). The moment generated
at B is the carry over factor from A to B. Since B is not actually fixed, application of
the opposite moment to the beam, Figure 22.2 (c), will generate a moment at support A.
By superposition, the results of these two applications of moment produce the beam
loaded as shown in Figure 22.2 (a). So, the absolute stiffness factor of the beam at
support A is given by:
K A′ = K A (1 − C ABC BA )
In this case, K A is the absolute stiffness factor of the beam by assuming the far end
B of the beam is fixed. For the sake of confirmation, in the case of prismatic beams,
we know that K A = 4 EI /4 EI and C AB = C BA = 1/2. Substituting these values in the
above equation yields:
3EI
K A′ =
l
which is the well-known result from our earlier analysis of beams with far end pinned.

FIGURE 22.2  Principle of superposition for combined stiffness calculation.

416 Introduction to Structural Analysis

22.3.2 Symmetric Beam and loading

In this case, we need to determine K A′ that is needed to rotate the beam at support A,
θ A = 1 radian, while θ B = −1. In this case, we first assume that support B is fixed, and
in this situation, we apply a moment K A at support. Next, we apply negative moment
K B at support B by assuming support A is fixed. This induces a moment C BA K B at end
A. Now we apply superposition principle to get:

K A′ = K A − C BA K B = K A (1 − C AB )

In the case of prismatic beams, K A = 4 EI /l and C AB = 1/2, so that:

2 EI
K A′ =
l

which is the same as the earlier derived value.

22.3.3 Symmetric Beam with Antisymmetric Loading

Now we will derive the relationship for symmetric beams with antisymmetric load-
ing. To do this, we first take fixed support at B as fixed support and apply moment
K A at A. Similarly, applying moment K B at support B by keeping A as fixed will yield
another result. Combining these two, we get:

K A′ = K A + C BA K B = K A (1 + C AB )

Substituting the values for prismatic beams, we get K A = 4 EI /l and C AB = 1/2, that
gives:
6 EI
K A′ =
l
Hence, we have derived the earlier results once again by substituting the appropriate
values for different parameters for prismatic beams.

22.3.4 Support Settlement
Now we will deal with the case of support settlements. In the case of support set-
tlements, fixed end moments are developed at the joints, where the settlement has
occurred. This we have already seen in Chapter 18. Now, to derive the result, we first
take both supports of the beam as pin supported and apply a settlement at support B
of the beam by an amount ∆. So, the rotation at the supports will be θ A = θ B = ∆ /l .
Then, assuming B is fixed, we apply a moment M A′ = − K A ( ∆ /l ) to the support A.
Following this, we take A end as fixed and apply a moment M B′ = − K B ( ∆ /l ), so that
end rotates by an angle θ B = −∆ /l . Thus, fixed end moment at A is given by:
∆ C BA K B ∆
( FEM ) AB = − K A −
l l
Beams and Frames Having Nonprismatic Members 417

Applying the earlier relationship,

C AB K A = C BA K B

we get,
∆
( FEM ) AB = − K A (1 + C AB )
l
For prismatic member, we have:

4 EI 1
KA = , C AB =
l 2
Thus,
3EI∆
( FEM ) AB = −
l2
which is in exact agreement with the earlier derived values in slope deflection equa-
tion and moment distribution method chapters, i.e., Chapters 18 and 19, respectively.

22.4 SLOPE DEFLECTION EQUATION FOR STRUCTURES

HAVING NONPRISMATIC MEMBERS
We have already learned slope deflection equations for prismatic members in Chapter 18.
In this section, we will develop a generalized form of slope deflection equations so
that they can be applied on nonprismatic members. To be able to do this, we take help
of the results mentioned in the previous section and will formulate the equations in
the same manner as discussed in Chapter 18, that is, considering the impacts caused
by the imposed loads, relative joint displacement or support settlement, and each
joint rotation separately, and then superimposing the results thus obtained.

22.4.1 Loads
Loads are transformed into fixed end moments that are acting at the ends A and B
of the span. Our convention is that positive moments act clockwise and negative
moment acts anticlockwise in this chapter.

22.4.2 Relative Joint Translation

When a relative displacement (we commonly call this as support settlement) between
the joints occurs, the induced moments are determined from the following equations
developed in Section 22.3.4 at the two support ends A and B:

K ∆
FEM AB = −  A   (1 + C AB )
 l 
K ∆
FEM BA = −  B   (1 + C BA )
 l 
418 Introduction to Structural Analysis

22.4.3 Rotation at A
If chord at support point A rotates by θ A, the required moment in the span at A will be
K Aθ A. Also, this induces a moment of C AB K Aθ A = C BA K Bθ A at other support end B.

22.4.4 Rotation at B
If chord at support point B rotates by θ B, the required moment in the span at B will
be K Bθ B . Also, this induces a moment of C BA K Bθ B = C AB K Aθ B at other support end A.
So, the total moment produced due to this effect will give us the generalized slope
deflection equation that can be written down as follows:

∆
M AB = K A [θ A + C ABθ B − (1 + C AB ) + FEM AB
l 

∆
M BA = K B [θ B + C BAθ A − (1 + C BA ) + FEM BA
l 

By applying these equations for nonprismatic beams, we will get the final results in
a fashion already explained in Chapter 18, slope deflection equation for prismatic
beams.
23 Introduction to Matrix
Structural Analysis

23.1 INTRODUCTION
In this chapter, the principles of using the stiffness method for analyzing structures
will be explained. Although the procedure outlined here may seem very useful for
manual calculation, but this is a quite tedious method for large structures. However,
the repeated steps can be programmed in a computer as software package for solv-
ing large structures comprising of many structural elements. Modern computer pro-
grams systematically apply this procedure toward solving the structure with minor
input parameters. Few examples will be provided as we move on, which will help us
to develop the basics of matrix analysis techniques using stiffness matrix formula-
tion. Also, it is to be noted that bold-faced upper- and lowercase letters in this chapter
will represent matrix unless otherwise mentioned in the text.

23.2  ANALYTICAL MODEL

Matrix stiffness method requires following key steps to start the analysis process:

1. Subdividing the structure into a series of discrete members called elements.

2. Denoting their end points as nodes. For trusses, elements are represented
by each of the members that compose the truss, and the nodes represent the
joints.
3. The force-displacement equilibrium equations of each element are deter-
mined, and each member is combined in a way they are connected at the
respective nodes.
4. These relationships, expressed in the form of matrix for the entire structure,
are then grouped together into a single matrix known as the structure stiff-
ness matrix K.
5. Once a complete stiffness matrix is established, the unknown displacements
of the nodes can then be determined for any given loading on the structure.

The steps mentioned above are pretty tedious for manual calculations as for each ele-
ment in the structures, one stiffness matrix needs to be formed. Then this stiffness
matrix needs to be transformed into global stiffness matrix using transformation
matrix. Once the global stiffness matrix of each element is formed, all these matrices
need to be combined in a logical way, depending on their connectivity and arrange-
ment in the whole structure. Thus, the global complete stiffness matrix of the entire
structure is formed. For large structures having many elements, these steps are quite
tedious in nature for manual calculations. However, many computer programs are
DOI: 10.1201/9781003081227-26 419
420 Introduction to Structural Analysis

available these days to form the stiffness matrices and subsequently form the global
complete stiffness matrix. We will provide few examples with structures having a
small number of elements to provide the necessary feel toward the above mentioned
steps. Students using software in the near future will have more confidence toward
analyzing the output results by doing some elementary matrix operations.
We will explain each step mentioned above by taking a model two-dimensional
(2D) truss. At the end of the chapter, we will provide sufficient insight toward form-
ing stiffness matrices for more generalized frame elements.

23.3 MEMBER STIFFNESS RELATIONS IN LOCAL

COORDINATES FOR 2D TRUSS
We will establish the stiffness matrix for a single truss member here. Let us consider
a typical truss element as shown in Figure 23.1.
The axis passing through the member and aligned along the member is the local
x axis of the member denoted by x. Similarly, y axis is also shown. These axes
attached to the member or element is called local axis. The axial force acting at
the two ends of the member has also been shown by respective arrows. Now due to
the positive displacement d N at the near end node N of the member, an axial force
qN′ will be developed. The force-displacement equation for this member can be
written as:

AE
qN′ = dN
l

And at the far end, since it is pinned, maintaining force equilibrium, we will get:

AE
qF′ = − dN
l

Now, if the N node is pinned and F node is free, the above equations will be just
reversed, which can be written as:

AE
qN′′ = − dF
l

AE
qF′′ = dF
l
Thus, superimposing the above equations, for both type of displacement occurring
simultaneously, we will get:

AE AE
qN = dN − dF
l l

AE AE
qF = dF − dN
l l
Introduction to Matrix Structural Analysis 421

FIGURE 23.1  Typical two-dimensional truss element and local coordinate of the element.

Now, these two load displacement equations can be expressed in a one matrix equa-
tion as written below:

 qN  AE  1 −1   d N 
 =     
 qF  l  −1 1   d F 

or,

q = k ′d
422 Introduction to Structural Analysis

where,

AE   1 −1 
k′ =    
l  −1 1 

This is the stiffness matrix for the truss element NF in local coordinate system. The
four elements that comprise the matrix are known as the member stiffness influence
coefficients kij′. Physically, this means, force at joint i, when unit displacement is
imposed at joint j. For example, i = j = 1, then k11
′ is the force at joint 1, when far end
joint is held fixed, and displacement at the near joint is d N = 1. So,

AE
qN =   k11
′ =
l
Similarly, for i = 2,  j = 1, we will get:

AE
qF = k21
′ = −
l
These two are the first column, entered in the above stiffness matrix k ′.

23.4  COORDINATE TRANSFORMATION FOR 2D TRUSS

A truss is composed of many individual members or elements oriented in different
directions. In this section, we will learn the matrix method of transformation of
coordinates from local to global scale. This is of utmost importance since all mem-
ber local stiffness matrices need to be transformed into global stiffness matrix prior
to assembling the same to form the final single global stiffness matrix for the entire
structure.
To be able to do that, we need to consider the member orientation and cast on the
global coordinate system as shown in Figure 23.2. As already stated in the previ-
ous section, member local axis is the axis, which is passing through the member.
Depending upon the orientation of the member in the actual truss, there will be an
angle between local and global axes. The angle of inclination of the member with

FIGURE 23.2  Local and global coordinate system.

Introduction to Matrix Structural Analysis 423

respect to global axis needs to be determined first. The angles between the positive
global x, y axes and the positive local x ′ axis are defined as θ x and θ y. Let us denote
the global coordinates of the two ends of the member as ( x N , yN ) and ( x F , yF ) for N
and F, respectively.
Let us make two more simplifying assumptions related to cosine of two angles
θ x and θ y:

δ x = cos θ x

δ y = cos θ y

xF − x N xF − x N
δ x = cos θ x = =
l ( x F − x N )2 + ( yF − yN )2
yF − y N yF − y N
δ y = cos θ y = =
l ( x F − x N )2 + ( yF − yN )2
We will use the above relationships in the next section to develop the displacement
transformation matrix.

23.5 DISPLACEMENT TRANSFORMATION MATRIX

FOR 2D TRUSS
In global coordinate, the member displacement along local axes and corresponding
components in global axis need to be connected. Otherwise, there will be no way to
form the global matrix for the entire structure. We can form this relationship with the
help of the direction cosines δ x and δ y and noting the displacement components as
shown in Figure 23.3. Each joint of a truss element is having two degrees of freedom,
i.e., ∆ N x , ∆ N y at node N, and ∆ Fx , ∆ Fy at node F in the direction of global coordinates.
From components of vectors, we can write:

d N = ∆ N x cos θ x + ∆ N y cos θ y

FIGURE 23.3  Local and global displacement coefficients.

424 Introduction to Structural Analysis

Similarly, for the displacement at the other end F, we can write:

d F = ∆ Fx cos θ x + ∆ Fy cos θ y

In terms of δ x and δ y, we can rewrite the above two equations in the following matrix
form:

 ∆ Nx 
 
 dN   δ x δ y 0 0   ∆ Ny 
 =   
 d F   0 0 δ x δ y   ∆ Fx 

 ∆ Fy 
 

or in the compact matrix equation form, we can write:

d = T∆

where, T is the transformation matrix of displacement, which is known as displace-

ment transformation matrix.

 δx δy 0 0 
T= 
 0 0 δx δy 
 

23.6  FORCE TRANSFORMATION MATRIX

Next, we will investigate how the local and global force components can be related
to the transformation matrix T already developed in the previous section. To do the
same, refer to Figure 23.4 for global and local force components at the node N of the
truss member.

FN x = qN cos θ x = qN δ x

FN y = qN cos θ y = qN δ y

FIGURE 23.4  Local and global force coefficients.

Introduction to Matrix Structural Analysis 425

Similarly, for the other node F, the above two equations can be written as:

FFx = qF cos θ x = qF δ x

FFy = qF cos θ y = qF δ y

In matrix form, the above four equations can be combined in the following elegant
form:
 FN x   δ x 0 
   
 FN y   δ y 0   qN 
=   
 FFx   0 δ x   qF 
  
 FFy   0 δy 
   

or, it can be written in the compact matrix equation form as:

Q = TT q
where:

  δ x 0 
 
  δ y 0 
T =
T

 0  δ x 
 0  δ y 
 

is the transpose matrix of the original transformation matrix T. So, for force trans-
formation, we need to take transpose matrix of displacement transformation matrix.

23.7  MEMBER GLOBAL STIFFNESS MATRIX FOR 2D TRUSS

Now we have arrived at a point where we can form the member global stiffness
matrix. Please be cautious that this is not the complete global matrix of the entire
structure. We will form only a one-member global stiffness matrix that we were
discussing in all previous sections. If we substitute, d = T∆ into the equation q = k ′d,
we will get:

q = k ′ T∆

Now substituting this into equation Q = T T q we will get:

Q = T T k ′T∆

or,

Q = K∆
426 Introduction to Structural Analysis

where:

K = T T k ′ T

is the global stiffness matrix of the member NF as discussed in all the sections.
Substituting the values of T T,  k ′, and T , one can perform the matrix multiplication
to obtain:

Nx Ny Fx Fy
 δ 2
δ xδ y −δ −δ xδ y  N x
2
 x x

 δ xδ y δ y2 −δ xδ y −δ y2  N y
K = AE /l  
 −δ x2 −δ x δ y δ x2 δ x δ y  Fx
 
 −δ x δ y −δ y2 δ x δ y δ y2  Fy


Since 2D truss element has two degrees of freedom per node, this 4 × 4 symmetric
matrix is referenced with each global degree of freedom related to near end N fol-
lowed by far end F. All the global force and displacement components in the row
and the column are provided at the top and right side of the global stiffness matrix
for ease of understanding. Let us determine the structure stiffness matrix for a truss
given in Example 23.1 below.

Example 23.1: Determine the structure stiffness matrix for the two-

member truss as shown in Figure 23.5 (a). Consider AE as constant for
all members.

SOLUTION:  From the above configuration of the truss, it is clear that we will have
two unknown displacements at node 2 only since nodes 1, and 3 are constrained
in both the directions at supports. At node 2, we will have movement along x
and y axis globally. Locally, the movement will be the resultant of this global
displacement.

FIGURE 23.5  Two-member truss with node and member numbering.

Introduction to Matrix Structural Analysis 427

Since 2 is the near end and 3 is the far end of the truss, hence,

3− 0
δx = =1
3
0−0
δy = =0
3
Using the above global matrix and dividing each matrix element by the length of
the member (i.e., by 3.0 m), we will get,

1 2 3 4
 0.333 0 −0.333 0  1
 
0 0 0 0 2
K1 = AE  
 −0.333 0 0.333 0  3
 0 0 0 0  4
 

Now the above matrix is the global stiffness matrix for member 1. For member 1,
there are two nodes 2 and 3 of which 2 is free to move and 3 is constrained. Also,
since this is a 2D truss, hence, each node can have degrees of freedom. For node 2,
which is near end, is represented by 1 and 2, whereas for node 3, it is 3 and 4.
Writing displacement of DOFs at the top and side of the stiffness matrix helps us to
assemble the matrices easily to form the global stiffness matrix of the entire truss.
Now for next member 2, we have two nodes 2 and 1. Here also 2 is the near
node and 1 is the far end node. Also, for this member, we have again four DOFs at
two nodes 1, 2, 5, and 6 where 5 and 6 are the DOFs related to node 1.
So, for member 2, we have:

3− 0
δx = = 0.6
5
4−0
δy = = 0.8
5

1 2 5 6
 0.072 0.096 −0.072 −0.096  1
 
0.096 0.128 −0.096 −0.128 2
K 2 = AE  
 −0.072 −0.096 0.072 0.096  5
 −0.096 −0.128 0.096 0.128  6
 

Thus, the global stiffness matrix for member 2 becomes:

Since the structure consists of three nodes, so the total number of DOFs will
be 3 × 2 = 6. Hence, the final matrix will be 6 × 6 matrix. Now, two matrices are
added algebraically with entries in each individual matrix with zero in rows and
columns, as shown below:

K =  K 1 +  K 2
428 Introduction to Structural Analysis

or,

1 2 3 4 5 6
 0.333 0 −0.333 0 0 0  1
 
 0 0 0 0 0 0  2
 −0.333 0 0.333 0 0 0  3
K = AE  
0 0 0 0 0 0 4
 
 0 0 0 0 0 0  5
 0 0 0 0 0 0  6
 

1 2 3 4 5 6

 0.072 0 0.096 0 −0.072 −0.096  1
 
 0.096 0 0.128 0 −0.096 −0.128  2
 0 0 0 0 0 0  3
+ AE   4
0 0 0 0 0 0
 
 −0.072 0 −0.096 0 0.072 0.096  5
 −0.096
 0 −0.128 0 0.096 0.128  6

or,
 0.405 0.096 −0.033 0 −0.072 −0.096 
 
 0.096 0.128 0 0 −0.096 −0.128 
 −0.333 0 0.333 0 0 0 
K = AE  
0 0 0 0 0 0
 
 −0.072 −0.096 0 0 0.072 0.096 
 −0.096 −0.128 0 0 0.096 0.128 


When the computer is called for analysis, generally, the software and/or algorithm
of the analysis software starts with all zero elements in all cells of the 6 × 6 matrix.
As the individual structural element global stiffness matrices are formed, the same
are placed directly into their respective element positions in the global overall K
matrix, instead of formulating the individual element stiffness matrices, computing
and storing them, and finally assembling them.

23.8  APPLICATION OF STIFFNESS METHOD FOR TRUSS ANALYSIS

Once the structure global stiffness matrix is generated, the global force components
Q acting on the truss can then be related to its global displacements D using:
Q = KD
The above equation is known as the structure stiffness equation. We can partition
the above matrix equation to represent the known and unknown components of each
matrix. Refer to the following matrix partition into known and unknown compo-
nents as shown:
 Qk   K11 K12   Dk 
 =    
 Qu   K 21 K 22   Du 
Introduction to Matrix Structural Analysis 429

Here the subscripts k, u represent known and unknown components.

Expanding the above matrix equation further, we get:

Qk = K11 Du + K12 Dk

Qu = K 21 Du + K 22 Dk

In most of the cases, Dk = 0 since there are no movements at the support points. In
such cases, we have:

Qk =  K11 Du

Since the elements in matrix K11 represent the total resistance at a joint due to unit
displacement at that joint in either x or in y direction, the equation is the matrix
representation of the force equilibrium equation of the entire truss. Since the exter-
nal applied load are either all known or are zero, hence, inverting the above matrix
equation, we get:

Du = K11−1 Qk

So, once the above equation is solved for Du and noting that Dk = 0, from the second
equation to solve for unknown components of matrix Qu

Qu = K 21 Du

The member forces can be determined from the already developed equation in the
previous section, as reproduced below:

q = k ′TD

Expanding this equation yields:

 DNx 
 
 qN  AE  1 −1   δ x δ y 0 0   DNy 
 =         
 qF  l  −1 1   0 0 δ x δ y   DFx 
  
 DFy 


We know that qN = − qF following equilibrium condition, hence, only one force

among these two needs to be calculated. Say, for example,

 DNx 
 
AE  DNy 
qF =   −δ x −δ y δ x δ y   

l   DFx 
 
 DFy 

430 Introduction to Structural Analysis

23.9 APPLICATION OF STIFFNESS METHOD

FOR SPACE TRUSS ANALYSIS
The analysis for space truss also follows the same procedure as that of 2D truss. Only
in the case of coordinate displacement, there will be three direction cosines:

δ x = cos θ x
δ y = cos θ y
δ z = cos θ z
Due to this additional term, the transformation matrix becomes:

 δx δy δz 0 0 0 
T= 
 0 0 0 δx δy δz 
 
Substituting this in the equation, k =  T T k ′T , we get:

 δx 0 
 
 δy 0 
 
δz 0  AE  1 −1   δ x δ y δ z 0 0 0 
k=    
 0 δ x  l  −1 1   0 0 0 δ x δ y δ z 
   
 0 δy 
 0 δ z 
 
Completing the matrix computation, the stiffness matrix yields:

 δ x2 δ xδ y δ xδ z −δ x2 −δ x δ y −δ x δ z 
 
 δ xδ y δ y2 δ yδ z −δ x δ y −δ y2 −δ yδ z 
 
 δ zδ x δ zδ y δ z2 −δ z δ x −δ z δ y −δ z2 
k = AE /l  
 −δ 2
x −δ x δ y −δ z δ x δ x2 δ xδ y δ zδ x 
 
 −δ x δ y −δ y2 −δ z δ y δ xδ y δ y2 δ zδ y 
 −δ z δ x −δ z δ y −δ z2 δ zδ x δ zδ y δ z2 
 
Once this global matrix has been formed, we can proceed to analyze the truss as per
the procedures explained in Section 23.7 for 2D truss.

23.10 APPLICATION OF STIFFNESS METHOD

FOR BEAM ANALYSIS
In this section, we will develop stiffness matrix for a prismatic beam element.
Necessary formulas for finding the stiffness matrix have already been developed in
the slope deflection equation chapter. We will use those equations very frequently and
Introduction to Matrix Structural Analysis 431

FIGURE 23.6  Beam element with local coordinate system and node numbering.

recall them as and when required in this section. Students and readers are advised to
go through the slope deflection equation chapter in detail once again before progress-
ing far from this point to learn the basic tools for the stiffness coefficient calculation
for beams.
We have learnt from the slope deflection equation that if a support sinks by an
amount ∆, there will be forces and moments induced at the supports due to this
effect. We first define the local axes of beam elements as shown in Figure 23.6.
The positive direction of force and moments is also drawn in Figure 23.6 for
understanding the local coordinate and force momentum vector conventions. Let
the displacements at the nodes be denoted by δ y′ and rotation through an angle be
represented by δ z′ . Linear force and moment at supports are represented by Fy′ and
M z′ , respectively. Now, due to linear and angular displacement occurring indepen-
dently at two nodes, we will have the following force and moments induced at the
support:

6 EI
′ =
FNy δ Ny
′
l2

6 EI
FFy′ = δ Fy
′
l2

12 EI
′ =
M Ny δ Nz
′
l3

12 EI
′ =
M Fy δ Fz
′
l3

4 EI
′ =
RNz δ Nz
′
l

2 EI
′ =
RFz δ Fz
′
l
432 Introduction to Structural Analysis

The above load displacement relationship can be expressed in a matrix form as

follows:

 12 EI 6 EI 12EI 6EI 
 − 3 
l3 l2 l l2
 ′    

M Ny
  6EI 4 EI 6EI 2EI   δ Ny
′

− 2
 ′
FNy   l 2
l l l   δ Nz
′ 
 ′ =    
 M Fy  
12EI 6EI 12EI
− 3 − 2
6EI
− 2   δ Fy
′ 
 FFy′   l l l3 l δ
′ 
   6EI 2EI 6EI 4 EI
  Fz 
 − 2 
 l2 l l l 

The above matrix relationship can be expressed in a matrix equation form as follows:

Q = kδ

The symmetric matrix k is called the member stiffness matrix for beam elements.
Physically, all these components in the stiffness matrix represent the amount of force
required for a unit displacement in the given positive sense of the member. Also, it is
customary to note that the local and global axes for this beam element are the same
since all these coordinates are parallel to each other. So, the stiffness matrix for
beam element will retain its form in both coordinate systems.

23.11 BEAM STRUCTURE COMPLETE GLOBAL

STIFFNESS MATRIX
When all the beam member stiffness matrices for an entire structure have been
formed, one must combine them into the complete structure global stiffness matrix
K. This computation first relies on determining the location of each member in the
member stiffness matrix. In this case, the rows and columns of each k matrix (like
the above matrix for each member) are denoted by the two code numbers at the
near end of the member FNy , FNz followed by those at the far end FFy, FFz . So, when
combining the matrices, each element must be placed in the same location of the
complete global K matrix. Following this procedure, K matrix attains an order that
will be equal to the highest code number assigned to the beam element, since this
indicates the total number of degrees of freedom. Also, in the case of several mem-
bers connected to a node, their member stiffness coefficients will have the same
position in the global K matrix and so must be algebraically added to determine the
nodal stiffness influence coefficient for the entire structure. This step is important
since each coefficient represents the nodal resistance of the structure in a particular
sense when a unit displacement occurs either at the same or another node. We will
understand this concept of assembling each elemental matrix to form global stiffness
matrix through Example 23.2.
Introduction to Matrix Structural Analysis 433

Example 23.2:  Determine the reaction at the supports of the continuous

beam as shown in Figure 23.7 using the matrix method of analysis.

SOLUTION:  The beam has two elements or members and three nodes. Out
of these, 1–4 numbers are taken to indicate unconstrained DOFs, and 5 and
6 numbers are used to indicate constrained degrees of freedom.
The known load and displacement matrices are as follows:

 0 1
 
−5  2
Fk = 
 0 3
 0  4


0 5
δ = 
06

Each DOF is written at the side of each element of the above matrices for ease of
understanding.
Now member stiffness matrices will be prepared directly from the earlier
developed k matrix.
Now we will assemble these two matrices to form the global stiffness matrix of
the entire structure:

Q = Kδ

6 4 5 3
 1.5 1.5 −1.5 1.5  6
 
1.5 2 −1.5 1 4
k1 = EI  
 −1.5 −1.5 1.5 −1.5  5
 1.5 1 −1.5 2  3
 

FIGURE 23.7  Example beam problem with node number, beam number, DOFs.
434 Introduction to Structural Analysis

5 3 2 1
 1.5 1.5 −1.5 1.5  5
 
1.5 2 −1.5 1 3
k2 = EI  
 −1.5 −1.5 1.5 −1.5  2
 1.5 1 −1.5 2  1
 

 −1.5 1 
 0   2 0 1.5 0
 δ1 
   −1.5 1.5 −1.5 0 −1.5 0 
 0   1  δ2 
   −1.5 −1.5 0 −1.5 0  
0    δ3
 = 1 −1.5 4 1 0 1.5 

0
  0  δ4 
0 1 2 −1.5 1.5 
 δ5   0 
   1.5 −1.5 0 −1.5 3 −1.5  
 δ6   0  0 
0 1.5 1.5 −1.5 1.5 

Now, carrying out the multiplication, the equations that can be formed from the
above matrix are:
2δ 1 − 1.5δ 2 +  δ 3 = 0 = 0

5
−1.5δ 1 + 1.5δ 2 − 1.5δ 3 + 0 =   −
EI
δ 1 − 1.5δ 2 + 4δ 3 + δ 4 = 0

0 + 0 + δ 3 + 2δ 4 = 0
Upon solution, we get:
16.67
δ1 = −
EI
26.67
δ2 = −
EI
6.67
δ 3 =  −
EI
3.33
δ4 =
EI
With this calculated value, we can determine the unknown forces as:

Q5 = 10 kN
Q6 =   −5 kN

23.12 APPLICATION OF STIFFNESS METHOD

FOR FRAME ANALYSIS
Having gained a fair amount of knowledge of truss and beam elements, we can apply
these for the analysis of frames also. In case of frames, at each node, there are three
degrees of freedom available in 2-D since the node can displace linearly in x and y
Introduction to Matrix Structural Analysis 435

direction as well as there can be a rotation. So, for a frame element, there will be total
2 × 3 = 6 DOFs per element in two dimensions. Thus, for the stiffness, transforma-
tion, force, etc., matrices will be 6 × 6 matrix.
Also, unlike the beams, the frame elements can be oriented at different angles;
hence, we also need transformation matrix to transform from local to global coor-
dinate system. We will provide the required formulations in matrix forms in case of
frame elements for reference. These matrices can be formed by the same procedure
as that explained for truss and beams in previous sections of this chapter.
Typical stiffness matrix for a frame element in local coordinate system:

 AE /l 0 0 − AE /l 0 0 
 3 2 
 0 12 EI /l 6 EI /l 0 −12 EI /l 6 EI /l 2
3

 0 6 EI /l 2 4 EI /l 0 −6 EI /l 2 2 EI /l 
k′ =  
 − AE /l 0 0 AE /l 0 0 
 0 −12 EI /l 3 −6 EI /l 2 0 12 EI /l 3 −6 EI /l 2 
 2 
 0 6 EI /l 2 EI /l 0 −6 EI /l 2 4 EI /l 

To transform the above local stiffness matrix into global, we need the following
transformation matrix to operate:

 δx δy 0 0 0 0 
 
 −δ y δx 0 0 0 0 
 
0 0 1 0 0 0 
T  = 
 0 0 0 δx δy 0 
 
 0 0 0 −δ y δx 0 
 0 0 0 0 0 1 


Similarly, the force transformation matrix will be of the following form:

 δx −δ y 0 0 0 0 
 
 δy δx 0 0 0 0 
 
0 0 1 0 0 0 
T T  = 
 0 0 0 δx −δ y 0 
 
 0 0 0 δy δx 0 
 0 0 0 0 0 1 


Frame member global stiffness matrix can be formed by carrying out the following
operation:

K =  T T k ′T
436 Introduction to Structural Analysis

The global stiffness matrix for each frame element needs to be calculated by apply-
ing the above equation and the result is left as an exercise for readers. This is noth-
ing but carrying out the stepwise row by column matrix multiplication that we have
already done in previous sections.
After forming each member global stiffness matrix, we need to carry out the
assembly process as per the member orientation and member numbering of the entire
frame to form the overall global stiffness matrix. This method is quite tedious for
large frames, and that is why, computer programs are called for to carry out the same
automatically as per the algorithms set to do the operation. However, for small frame
elements, we can attempt to do it manually and compare the result from computer
analysis so that a concrete understanding of the underlying process remains at our
confidence.
 24 Introduction to Plastic
Analysis of Structure

24.1 INTRODUCTION
In this chapter, plastic analysis of structure is introduced. Plastic analysis has advan-
tages over elastic analysis in a way that members provide much more resistance
and practical bending features. Material can utilize its reserve strength that remains
unutilized in the elastic analysis if plastic analysis is adopted. We can define our limit
load more realistically in plastic analysis, and thus, our design will be more eco-
nomical. After completing this chapter, we will learn to calculate the plastic section
modulus of different sections. Different types of frames using this analysis method
will also be discussed in this chapter. A thorough understanding of this method of
analysis will provide readers a very solid foundation to work in design analysis field.

24.2  STRESS-STRAIN CURVE OF A DUCTILE MATERIAL

For a ductile material like steel, the stress strain diagram under tension looks similar
to the one shown in Figure 24.1.
Ideal ductile material is defined as the one that has a defined elastic range and
then plastic range follows. In the above stress-strain curve, from O to A, the line is
straight line, and in this portion the stress is proportional to strain. If you unload
the material from point A, it will reach its origin O. There will be no residual strain
left due to unloading. Point A is called the proportional limit. Once the load is little
increased, and the stress-strain curve reaches point A′, the material remains still
elastic, but if the material is unloaded from this point, a very small residual strain
remains. This point A′ is called the elastic limit of the material. Once stress increases
more than this, the elastic range is exceeded, and the material reaches yield point,
namely upper yield point B, and lower yield point C. The portion between C and D
is known to be the plastic range of the material, here strain increases at constant
stress. Beyond point D, the material strain hardens up to point E, where the ultimate
stress reaches. After point E, necking starts to develop, and upon further increase in
load, the material gets fractured at point F. F′ is the true fracture stress, where the
change in cross section is considered. Plastic analysis is based on this elastic plastic
stress-strain curve. In most situations, the portion CD is approximated into a flatten
line, and an ideal zone for plastic behavior is taken for analysis and design purpose.
Thus, the ideal stress-strain curve considered for plastic analysis will look similar to
the one shown in Figure 24.2.
For ductile material, upper yield points generally not considered for calculating
strength of material. It is considered that the material is elastic stage up to lower yield
point and then it enters plastic state.
DOI: 10.1201/9781003081227-27 437
438 Introduction to Structural Analysis

FIGURE 24.1  Stress strain diagram of ductile material.

FIGURE 24.2  Ideal stress strain diagram of ductile material for plastic analysis.

24.3  PLASTIC MOMENT

Let us consider a prismatic cantilever beam with an applied increasing moment M
at the end of the beam as shown in Figure 24.3. When moment intensity is such
that bending stress in the extreme fiber does not attain yield stress, the beam will
remain within the elastic zone of deformation. In this zone, the variation of stress
will be zero at neutral axis up to the maximum at the extreme fiber level. With fur-
ther increase in moment, the extreme fiber will reach the yield point, but the internal
fibers of the beam will still have less stress intensity than the yield stress. So, the
bending stress diagram will be triangular with zero intensity at the neutral axis.
Introduction to Plastic Analysis of Structure 439

FIGURE 24.3  Progressive development of the plastic state for a rectangular beam under the
application of increasing bending stress.

In this situation, the moment will be a yield moment, and the same is related to the
yield stress of the material by the following equation:

bd 2
M y =   fy ×
6
where f y is the yield stress of the beam material, b is breadth, and d is the depth of
the beam.
With further increase in moment, the stress in extreme fiber will be unchanged
while the stress in the inner fibers will be increased and the fiber at the immediate
vicinity of the extreme fiber will attain the yield stress. So, the stress diagram will
be slightly rectangular in nature and then there will be triangular portion reaching to
zero at the neutral axis. With further increase in moment, all the fibers of the beam
above and below the neutral axis will attain yield stress resulting in rectangular stress
distribution above and below the neutral axis as shown in the last stress diagram in
Figure 24.3. When all the fibers attain yield stress f y , then the beam is said to attain its
full plastic state. In Figure 24.4, the stress distribution of a beam having rectangular
cross section in partially yielded (Figure 24.4 (b)) condition and fully yielded condition
(Figure 24.4 (c)) is shown. In this same figure (Figure 24.4 (d)), the stress resultant at
the fully yielded condition is also shown. Now, in equilibrium condition, the net force
acting in the cross section will be zero. Let us consider the fully yielded condition,

FC = Ft
or, f y Ac = f y At

FIGURE 24.4  Stress distribution for rectangular section in partially and fully yielded
condition.
440 Introduction to Structural Analysis

where Ac and At are the areas under compression and tension, respectively. So, from
the above expression, we can conclude that, in the fully yielded condition, the neu-
tral axis divides the area exactly into two parts. Now, let us find out the moment of
resistance (M.R.) in a partially yielded condition, where p is the depth of penetration
of the yield zone as shown in Figure 24.4 (a) and (b).
M.R. of the section in partially yielded condition = M.R. of the yielded zone +
M.R. of the elastic zone.
Therefore,

 d p  1
M.R. = M =  2 × f y bp  −   +  f y × × b ( d − 2 p ) 
2

  2 2    6 

or,
1
 f y bp ( d − p )  +  f y × × b ( d − 2 p ) 
2

 6 

Let us consider, p = α d ; M =  f y bα d ( d − α d ) +  f y × 16 × b ( d − 2α d ) 

2

or,
1
M=
6
(
f y bd 2 1 + 2α − 2α 2 )
The above expression can be written as,

1 2
M=
4 3
(
f y bd 2 × 1 + 2α − 2α 2 )
For fully yielded condition, p = 0.5d , i.e., α = 0.5; In that condition,

1
M = MP = f y bd 2 = f y S
4

where S is the plastic modulus. The ratio of fully plastic moment to elastic moment
capacity of any section is called shape factor of the section. So, for our case, the
shape factor for this rectangular beam will be:

MP f y bd 2 /4
=  = 1.5
My f y bd 2 /6

24.4  METHODS OF ANALYSIS

If we keep on increasing the load, a structure will attain fully plastic state in some
regions at some point of time. The resisting bending moment cannot be increased at
these fully plastic sections anymore, but the structure can rotate around these. Thus,
we model this by placing a hinge at that point where structure has attained fully plas-
tic state. So, a fully plastic section of structure forms a hinge, and it is called plastic
hinge. With the plastic hinge formation, the structure attains a mechanism, and it
Introduction to Plastic Analysis of Structure 441

FIGURE 24.5  Plastic hinge formation and mechanism for a simply supported beam.

reaches on the verge of collapse. If the degree of static indeterminacy of a structure

is DS , plastic hinge required (n) to form mechanism is,
n = DS + 1
a. Simply supported beam loaded with concentrated load: To understand
this concept, let us take the example of a simply supported beam with a
concentrated load P applied at a distance a from the support A as shown
in Figure 24.5. The degree of static indeterminacy ( DS ) in this case will
be zero. Then number of plastic hinge (n) required to form mechanism is,
n = 0 + 1 = 1.
As the applied point load intensity increases and reach β P, plastic hinge
forms lead to a mechanism at the location of the applied load at some point
in time, and the structure will collapse. At this situation, the structure can
undergo any rotation about the plastic hinge, and the acting moment at
that location will be fully plastic moment M p. This can be related to initial
bending moment at the same section with the following equation:

Pab
MP = β
l

or,
MPl
β=
Pab

This factor β is called the collapse load factor of the beam we are dealing
with. In case the point load is acting at the midpoint of the beam, then col-
lapse load factor will be (a = b = l /2):

4M P
β=
Pl
442 Introduction to Structural Analysis

Thus, from the above equation, we can determine the collapse load factors
under varying intensity of applied loads. If M y is the moment when first
yielding appears in the structure due to some load intensity of β ′ P , we can
write:

4M y
β′ =
Pl

Now, the ratio ββ′ = MMPy is called as the shape factor.

b. Propped cantilever loaded with concentrated load: Let us consider a
propped cantilever with a concentrated load acting on its span as shown
in Figure 24.6. The degree of static indeterminacy ( DS ) in this case will
be one. Then number of plastic hinge (n) required to form mechanism is,
n = 1 + 1 = 2.
In this case, maximum negative bending moment occurs at the fixed
end, and maximum positive bending moment occurs under the load. If
we keep on increasing the load, the first plastic hinge will occur either at
the fixed end or under the applied load depending on where the numerical

FIGURE 24.6  Plastic hinge formation and mechanism for a propped cantilever.
Introduction to Plastic Analysis of Structure 443

value of bending moment will be maximum. The beam will not still col-
lapse because we need two plastic hinges to form, in this case, to create the
mechanism. So, at the collapse state, the beam will form a mechanism as
shown in Figure 24.6 (c) under the load magnitude of β P . In this condition,
the internal work done by the plastic moments, M P should be equal to the
external work done by the applied force β P . We can write,

M Pθ A + M P (θ A + θ B ) = β P × δ

or,
M P ( 2θ A + θ B )
β= ×
P δ

Here, δ = aθ A = bθ B; So, we can rewrite the earlier expression as,

(2b+ a)
β = MPP ×  a2θθ AA + bθθBB  = M PPab .
If the point load is applied at the middle, a = b = L2
Therefore,

L
MP  L + 
M P ( 2b + a )  2  = 6M P .
β= =
Pab  L L PL
P ×
 2 2

Let yielding first occur for the load magnitude of β ′ P . From Figure 24.6 (b), we can
understand that yielding will first commence in the fixed end A, for a = b = L2 . So, in
the first yield condition, we can write,

L L L
β ′P  ×  ×  L + 
 2 2  2  = β ′ × 3PL
My =
2 L2 16
or,
16 M y
β′ = ×
3 PL
So,
6M P
β PL 9 M 9
= = × P = × Shape factor
β′ 16 M y 8 My 8
×
3 PL

As it is clear from the above discussion, that plastic hinge formation may happen
anywhere over the span of a structural element depending upon the intensity of load-
ing at that region. Thus, possible collapse mechanism will vary in number, par-
ticularly if the number of indeterminacy is more, and it is practically impossible to
carry out all types of mechanisms and determine the minimum load under which the
444 Introduction to Structural Analysis

collapse may occur. To identify the correct load factor in plastic analysis, there are
three important criteria as below:

1. Mechanism: Mechanism is formed when the ultimate load or collapse load

is reached in the structure. To form a mechanism, the number of plastic
hinges developed should be just sufficient.
2. Equilibrium: The internal bending moment must be in equilibrium with
the external loading. The conditions of equilibrium for 2D case are:
∑ Fx = 0,  ∑ Fy = 0,  ∑ M z = 0 .
3. Yield criteria: The bending moment at a section should not exceed the plas-
tic bending moment at that section of the structure.

To overcome the analysis hurdle and based on these three criteria, we have the fol-
lowing theorems that need to be studied and understood well.

24.4.1 The Maximum Principle, Static Theorem, Lower Bound

Theorem, or Safe Theorem
If there exists a bending moment distribution ( M ) for a structure and loading which
is safe ( M < M P ) and statically admissible (equations of statics are applicable) with
a set of loads W then,
W ≤ WC

where WC is the collapse load of the structure.

This theorem is safe, because the load factor will be less than or equal to the col-
lapse load factor once the equilibrium and yield criteria are met. As the load factor
comes out to be less than the actual collapse load, this theorem either gives a wrong
and safe result or gives a right and safe result. Since the elastic analysis always meets
the equilibrium and yield conditions, an elastic analysis will always be safe. As the
load obtained by this theorem is always to be lesser than or equal to the collapse
load, this theorem is also known as lower bound theorem.

24.4.2 The Minimum Principle, Kinematc Theorem, Upper Bound

Theorem, or Unsafe Theorem
This theorem states that of all the mechanisms formed by assuming plastic hinge
locations, the correct collapse mechanism is the one that requires the minimum
loads. So, any choice of plastic hinge location will provide collapse load that is
greater than or equal to the correct collapse load.
For a given structure subjected to a set of loads W , the value of W found corre-
sponding to any assumed mechanism must be either greater or equal to the collapse load.
i.e.,
W ≥ WC

In this theorem, we always assume a mechanism is forming. This theorem is unsafe

because as the load factor from this theorem comes out to be precisely correct or
Introduction to Plastic Analysis of Structure 445

larger than the actual collapse load, it may lead the designer to think that the struc-
ture is capable of taking more load than the actual, which may lead to a dangerous
situation. As the load obtained by this theorem is always to be greater than or equal
to the collapse load, this theorem is also known as upper bound theorem.

24.4.3  The Uniqueness Theorem

This theorem states that the bending moment distribution, which fulfills all the three
conditions mentioned above, i.e., mechanism, conditions of equilibrium, and yield,
is unique. In other words, we can say if we obtain a bending moment distribution
satisfying static theorem, in this distribution, the bending moment is equal to the
plastic bending moment at sufficient sections to form mechanism then that load is
equal to the collapse load.
i.e.,
W = WC
Using these theorems, we can find the collapse load by the following two methods
namely,
a. Static method
b. Kinematic method
Some examples on both these methods are discussed in the following sections.

24.5  STATIC METHOD FOR DETERMINING COLLAPSE LOAD

In this method, collapse load is computed based on the geometry of an assumed
equilibrium bending moment diagram. Here the bending moment of a section is not
greater than the plastic moment (less than or equal to) of that section. Some illustra-
tive examples are provided below to clarify the concept.

Example 24.1:  Determine the collapse load for the beam and loading
condition shown in Figure 24.7 by static method.

SOLUTION:  In this case, corresponding to the collapse condition, plastic

hinge should develop at the center of the beam under the applied load. The
bending moment diagram corresponding to the collapse condition is shown in
Figure 24.8.

FIGURE 24.7  Example problem 1 on static method.

446 Introduction to Structural Analysis

FIGURE 24.8  Bending moment diagram of Example 24.1.

From geometry of the collapse bending moment diagram, we have,

PL
MP =
4
4MP
Therefore, collapse load, PC =
L

Example 24.2:  Determine the collapse load for the beam and loading
condition shown in Figure 24.9 by static method.

SOLUTION:  In this case, corresponding to the collapse condition, plastic hinges

should develop at the center of the beam under the applied load and at the fixed
ends. The bending moment diagram corresponding to the collapse condition is
shown in Figure 24.10.
From geometry of the collapse bending moment diagram, we have,

PL
MP + MP =
4
8MP
Therefore, collapse load, PC =
L

FIGURE 24.9  Example problem-2 on static method.

FIGURE 24.10  Bending moment diagram of Example 24.2.

Introduction to Plastic Analysis of Structure 447

FIGURE 24.11  Example problem-3 on static method.

Example 24.3:  A beam of span 8 m is fixed at both ends and carries a

uniformly distributed load of 70 kN/m over the left half of the span as
shown in Figure 24.11. Design the section considering static method
of plasticity. Allow a load factor of 1.5. Yield stress for steel is
250 N/mm2.

SOLUTION:  Collapse load = load factor × safe load = 1.5 × 70 kN/m = 105 kN/m

Considering the beam as simply supported and carrying a distributed load of
105 kN/m on the left half of the span, right end reaction,

105 × 5 × 2.5
RB =   = 131.25 kN
10

∴ left end reaction,

RA = 105 × 5 − 131.25 = 393.75 kN

At any section in AC distant x meters from A, free bending moment,

2
Mx = 393.75x − 105 x2
For the condition Mx is maximum, dM
dx = 0
x

dMx
∴  = 393.75 − 105x = 0
dx

Therefore, x = 3.75 m

2
∴ Maximum free bending moment = 393.75 × 3.75 − 105 × 3.75 2 = 738.28 kNm
At the collapse condition, plastic hinges will be developed at A, B and the
point of maximum sagging bending moment. The bending moment diagram cor-
responding to collapse condition is shown in Figure 24.12.
From the geometry of this diagram, we find

MP + MP = 738.28 kNm

∴ MP = 369.13 kNm

Again, MP = fy S
448 Introduction to Structural Analysis

FIGURE 24.12  Bending moment diagram of Example 24.3.

or,
369.13 × 106 = 250  S
6
Therefore, plastic modulus required = S = 369.13
250
×10
mm3 = 1476000 mm3
Assuming a shape factor of 1.15,
Section modulus required = 1476000
1.15 mm = 1283.5 × 10 mm .
3 3 3

Example 24.4:  Determine the collapse load for the beam and loading
condition shown in Figure 24.13 by static method.

SOLUTION:  Figure 24.12 shows the bending moment diagram corresponding

to the collapse condition. From the geometry of the collapse bending moment
diagram, we have,

Pab b  b b + L
= EF + FC = MP + MP = MP  1+  = MP
L L  L L

Therefore, collapse load, PC = b+L

ab MP

FIGURE 24.13  Example problem-4 on static method.

FIGURE 24.14  Bending moment diagram of Example 24.4.

Introduction to Plastic Analysis of Structure 449

24.6 KINEMATIC METHOD FOR DETERMINING

COLLAPSE LOAD
The static method is suitable for simple structures, where the bending moment dia-
gram is a simple one. But if the structure gets complicated, kinematic method of
analysis is more suitable. In this method, work performed by the external loads is
equated to the internal work absorbed by the plastic hinges to find out the collapse
load. As this method is based on the assumed mechanisms of the structure, a load
computed based on this method is always greater than or equal to the actual ultimate
load. Some illustrative examples are provided below to clarify the concept.

Example 24.5:  Determine the collapse load for the beam and loading
condition shown in Figure 24.7 as given earlier by kinematic method.

SOLUTION:  Figure 24.15 shows the collapse mechanism of the simply supported
beam shown in Example 24.1. Let us now provide a virtual displacement δ at
point C. Under the influence of this small displacement the beam will rotate about
the plastic hinge. The internal plastic moment MP will try to oppose this rotation.
On the verge of collapse, the beam will be in equilibrium. In this equilibrium
condition, work done by the external force will be equal to the work done by the
internal force.
The external virtual work = Pδ = P L2θ  (since the displacement is small).
Internal virtual work = 2MPθ
Equating both we get,
Lθ
P = 2MPθ
2

or, collapse load, PC = 4MLP ; It can be noted that the amplitude of the collapse load
in this method is same as the earlier static method.

Example 24.6:  Determine the collapse load for the beam and loading
condition shown in Figure 24.9 as given earlier by kinematic method.

SOLUTION:  Figure 24.16 shows the collapse mechanism of a fixed beam shown
in Example 24.2. Let us now provide a virtual displacement δ at point C. Under the

FIGURE 24.15  Collapse mechanism of beam in Figure 24.7.

450 Introduction to Structural Analysis

FIGURE 24.16  Collapse mechanism of beam in Figure 24.9.

influence of this small displacement, the beam will rotate about the plastic hinge.
The internal plastic moment MP will try to oppose this rotation. As the degree
of indeterminacy is two, a total of three hinges need to form for the collapse
mechanism. On the verge of collapse, the beam will be in equilibrium. In this
equilibrium condition, work done by the external force will be equal to the work
done by the internal force.
The external virtual work = Pδ = P L2θ  (since the displacement is small).
Internal virtual work = 2MPθ + MPθ + MPθ = 4MPθ
Equating both we get,

Lθ
P = 4MPθ
2

or, collapse load, PC = 8MLP ; it can be noted that the amplitude of the collapse load
in this method is same as the earlier static method.

Example 24.7:  Determine the collapse load for the beam and loading
condition shown in Figure 24.17 as given earlier by kinematic method.

SOLUTION:  Figure 24.18 shows the collapse mechanism of the fixed beam under
uniformly distributed load. As the degree of indeterminacy is two, a total of three
hinges, as shown in Figure 24.18, need to form for the collapse mechanism.
Total internal work = 2MPθ + MPθ + MPθ = 4MPθ
L L L
wθ L2
∫ ∫ ∫
2 2 2
Total external work = 2 wdxy = 2 wdxxθ = 2wθ xdx =
0 0 0 4

FIGURE 24.17  Example problem-3 on kinematic method.

Introduction to Plastic Analysis of Structure 451

FIGURE 24.18  Collapse mechanism of the beam.

Equating both we get,

wθ L2
= 4MPθ
4
16MP
or, w = L2
16MP
Total collapse load will be, WC = wL = L

Example 24.8:  A beam AB of span L fixed at both ends has to carry a

point load at a distance L/4 from the left end as shown in Figure 24.19 (a).
Find the value of the load at the collapse condition if the plastic moment
of resistance of the left half of the beam is 3MP while the plastic moment
of resistance of the right half of the beam is MP .

SOLUTION:  Figure 24.19 (a) shows the fixed beam subjected to collapse load P.
There are two possible collapse mechanisms, shown in Figure 24.19 (b) and (c).
In the first case shown in Figure 24.19 (b), plastic hinges are formed at A, B and
under the load. Let a small virtual displacement, CD = δ be given.
3L
Total external virtual work = Pδ = P θ
4
Total internal virtual work = 3MP ( 3θ ) + 3MP ( 3θ ) + 3MP (θ ) + MP (θ ) = 22MPθ
3L
Equating both we get, P θ = 22MPθ
4
29.33MP
Therefore, P =
L
In the second case shown in Figure 24.19 (c), plastic hinges are developed at A, B,
and E. Let a small virtual displacement, EF = δ be given.
L
Here, δ = θ
2 L
Total external virtual work =  P θ
4
Total internal virtual work = 3MPθ + MPθ + MPθ + MPθ = 6MPθ
L
Equating both we get, P θ = 6MPθ
4
24MP
Therefore, P =
L 24MP
Therefore, actual collapse load, PC =
L
452 Introduction to Structural Analysis

FIGURE 24.19  Example problem-3 on kinematic method.

24.7  PLASTIC ANALYSIS OF PORTAL FRAMES

A single portal frame can fail in three types of mechanisms, namely, beam mecha-
nism, sway mechanism, and combined mechanism, as shown in Figure 24.20 (a),
(b), and (c), respectively. The portal frame shown in this figure is indeterminate to
degree three. For the total collapse of the structure, we need four plastic hinges to
form. But in the case of beam mechanism, only partial collapse takes place. In this
case, three plastic hinges are developed for the partial collapse of this member. Sway
mechanism is a complete collapse of the frame. So, four hinges need to form for the
total collapse. In the case of a sway mechanism, no plastic hinges are created due
to vertical load. Only the horizontal load creates a mechanism here. The combined
mechanism also led to the total collapse of the structure. So, four plastic hinges need
to form here also. Here, the plastic hinge will not form in the left-hand side beam-
column joint as both the column and the beam will rotate in the clockwise direction
Introduction to Plastic Analysis of Structure 453

FIGURE 24.20  Collapse mechanism of portal frame.

454 Introduction to Structural Analysis

FIGURE 24.21  Frame example problem.

under the influence of applied load, as shown. So, orthogonality of this joint will
remain unaffected. Some illustrative examples are provided below to clarify the
concept.

Example 24.9:  Find the plastic moment required for the portal frame
subjected to the collapse load system shown in Figure 24.21. All members
are of the same section.

SOLUTION:  As discussed earlier, there can be three types of failure mechanism

for these frames: beam, sway, and combined mechanisms. Three types of failure
mechanisms are drawn one by one below for ease of understanding.
Let the load factor be β . Then from Figure 24.22, we get,

100β × 4θ − MPθ − MPθ − 2MPθ = 0

FIGURE 24.22  Beam mechanism.

Introduction to Plastic Analysis of Structure 455

FIGURE 24.23  Sway mechanism.

or,

MP
β= = 0.01MP
100

Then we will deal with sway mechanism as shown in Figure 24.23.

As per Figure 24.23, the equation of sway mechanism will be:

30β × 6θ = MPθ + MPθ + MPθ + MPθ

or,

2
β= MP = 0.022MP
90

Finally, we will draw the combined mechanism as shown in Figure 24.12.

FIGURE 24.24  Combined mechanism.

456 Introduction to Structural Analysis

The equation as per this mechanism will be:

100β × 4θ + 30β × 6θ − MP × θ − MP × θ − MP × θ − MP × 2θ − MP × θ = 0

or,

3MP
β= = 0.01034MP
290

So, comparing among all three cases, final collapse factor will be β = 0.01MP.
Appendix A
Areas and Centroids
of Geometric Shapes
Shape Area Centroid

bh 2b
A= x=
2 3

2h a+b
A= x=
2 3

b(h1 + h2 ) b (h1 + 2h2 )

A= x=
2 3(h1 + h2 )

2bh 3b
A= x=
3 8

(Continued)

457
458 Appendix A

Shape Area Centroid

bh 3b
A= x=
3 4

2bh b
A= x=
3 2

3bh 2b
A= x=
4 5

bh 4b
A= x=
4 5
Appendix B
Review of Matrix Algebra
B.1  INTRODUCTION TO MATRIX ALGEBRA
A matrix is a rectangular array of figures arranged in rows and columns. Algebraically,
matrix A is represented as aij in which i represents the row number and j represents
the column number. In tabular form of matrix, the same can be expressed as:

 a11  a1 j 
 
aij =     
 ai1  aij 
 

Sometimes, it is easier to understand the matrix operation using the index notation,
rather than using the cumbersome tabular forms. A matrix can be only column or
row matrix. So, a matrix having a single row is called row matrix, and a matrix hav-
ing only one column is called column matrix.
A matrix is said to be a square matrix when it has an equal number of rows and
columns.
Two matrices are said to be equal if and only if all the elements at each row are
same. Two matrices can only be added together when both have the same rows and
columns. In index notation, we can write:

A+ B = D

i.e.,

aij + bij = dij

In tabular form, the abovementioned matrix operation can be expressed as follows:

 a11  a1 j   b11  b1 j   d11  d1 j 

     
     +        =     
 ai1  aij   b  bij   di1  dij 
   i1   

where d11 = a11 + b11 etc.

If a matrix has all the entries as 0, only then it is called null matrix and is repre-
sented by 0.
If we add any matrix with null matrix, then the matrix remains unchanged:

aij + 0 = aij
459
460 Appendix B

i.e.,

A+ 0 = A

Two matrices can be multiplied together by row and column multiplication rule. It is
customary to note that two matrices can only be multiplied when number of columns
in the first matrix is same as the number of rows of the second matrix.

C = A × B =  aij × b jk =  cik

Matrix addition is commutative, i.e.,

A+ B = B+ A

aij + bij = bij + aij

Matrix multiplication is noncommutative, i.e.,

A× B ≠ B× A

Transpose of a matrix is the process in which rows of the matrix are transformed into
column or vice versa. It is as follows:

A = aij  implies AT = a ji

Example:
If,

 1 2 
A=  
 3 4 
Then,
 1 3 
AT =  
 2 4 

Example of matrix multiplication:

If,
 1 2   5 7 
A=   and B =   then A × B is given by,
 3 4   6 8 

 1 × 5 + 2 × 6 1 × 7 + 2 × 8   17 23 
A× B =  = 
 3 × 5 + 4 × 6 3 × 7 + 4 × 8   39 83 
It is instructed to readers that they should carry out the product B × A and check that:

A × B  ≠ B × A
Appendix B 461

If we multiply matrix by a scalar, then each element of the matrix is multiplied by the
same scalar. In index notation, this can be written as:

α A = α aij

A matrix is said to be a symmetric matrix if it is symmetric about its diagonal:

aij =  a ji

A matrix is said to be a diagonal matrix if all the elements except its principal diago-
nal are 0. In index notation, this can be written as:

aij = 0  iff   i ≠ j

Unit matrix is the one that has 1 along its main diagonal and all other elements are 0.
Index notation for the same is as follows:

aij = 1 when  i = j

aij = 0  when  i ≠ j

This matrix is also called identity matrix, because if we multiply any matrix with
unit matrix, then the original matrix remains unchanged after multiplication. Identity
or unit matrices are denoted as I.

B.2  MATRIX INVERSE

The inverse of a square matrix is the one that when multiplied with the original matrix
gives unit or identity matrix as an outcome of multiplication. Mathematically it is
expressed as:

A ×  A−1 = A−1 × A = I

B.3  PARTITIONING OF MATRICES

Sometimes, larger matrices are partitioned into smaller sub-matrices to perform
operation in a much easier way. To understand this, the partitioning can be done
in many ways. One of the partitioning methods is shown next to understand this
concept.

 a d g j 
 
A=  b e h k 
 c f i l 
 
462 Appendix B

now, by carrying the partitioning as per the dotted line, we get:

 C11 C21 
A=   
 C12 C22 

 a d g   j 
where C11 =   , C12 =  c f i , C21 =   , and C22 = [ l ]
 b e h   k 

B.4 MATRIX INVERSION BY GAUSS-JORDAN

ELIMINATION PROCESS
The Gauss-Jordan Elimination process necessarily involves multiple row and col-
umn operation within the same matrix. In a matrix, we can add, multiply, and divide
each element of row or column elements and add or subtract between rows and
columns without causing any changes in the matrix. We need to carry out this type
of operations multiple times so that the original matrix can be converted into a
diagonal form. Then the augmented matrix written at the side of the main matrix
will produce the inverse of the main matrix. To understand this process fully, see
Example B.1.

Example B.1:  Calculate the inverse of the following matrix.

 2 −5 4 
 
A= 3 1 8 
 4 −7 −1 

SOLUTION:  To apply Gauss-Jordan Elimination process, we first form the aug-

mented matrix by taking the 3 × 3-unit matrix at the side of the main matrix as
shown next:

 2 −5 4  1 0 0
 
 3 1 8  0 1 0
 4 −7 −1  0 0 1

Now, our main task is to convert the main matrix into diagonal unit matrix by
doing fundamental operations on rows of the total augmented matrix. Let us first
divide the first row by 2, and we will get:

 1 −2.5 2  0.5 0 0
 
 3 1 8   0 1 0
 4 −7 −1  0 0 1
Appendix B 463

Next, we multiply row 1 by 3 and subtract it from row 2, we will get:

 1 −2.5 2  0.5 0 0
  
 −3 × 1 + 3 −3 × 2.5 − 1 −2 × 3− 8  3 × 0.5 − 0 −3 × 0 + 1 3 × 0 −0
 4 −7 −1  0 0 1

 1 −2.5 2  0.5 0 0
 
which gives:  0 8.5 −14    1.5 1 0
 4 −7 −1  0 0 1

Then we multiply row 1 by 4 and subtract it from row 3 to get:

 1 −2.5 2  0.5 0 0
 
 0 8.5 −14    1.5 1 0
 0 3 −9  −1.5 0 1

Now, dividing row 2 by 8.5, we get:

 1 −2.5 2  0.5 0 0
 
 0 1 −1.647    0.176 0.118 0
 0 3 −9  −1.5 0 1

Now, multiplying row 2 by −2.5 and subtracting it from row 1, we get:

 1 0 2.118  −0.94 0 0
 
 0 1 −1.647    0.176 0.118 0
 0 3 −9  −1.5 0 1

Then multiply row 2 by 3 and subtracting it from 3, we get:

 1 0 2.118  −0.94 0 0
 
 0 1 −1.647    0.176 0.118 0
 0 0 −4.059  −3.528 −0.354 1

Now, dividing row 3 by −4.059, we get,

 1 0 2.118  −0.94 0 0
 
 0 1 −1.647    0.176 0.118 0
 0 0 1  −0.869 −0.09 1

Then, on multiplying row 3 by −2.118 and subtracting it from row 1, we get:

 1 0 0  0.9 −0.017 2.118

 
 0 1 −1.647    0.176 0.118 0
 0 0 1  −0.869 −0.008 1
464 Appendix B

Now, multiply row 3 by −1.647 and subtract it from row 2 to get:

 1 0 0  0.9 −0.017 2.118

 
 0 1 0    −1.255 0.104 1.647
 0 0 1  −0.869 −0.008 1

Thus, we have converted the main matrix into the diagonal form by carrying out
several fundamental row operations. So, the 3 × 3 obtained at the right side of the
augmented matrix will be inverse of main matrix A.

 0.9 −0.017 2.118 

 
Hence, A−1 =  −1.255 0.104 1.647 
 −0.869 −0.008 1 

It is customary to check that,

AA−1 =  A−1A = I
Appendix C
Three-Moment Equation
Three-moment equation was developed by Clapeyron in 1857. This method provides
a very convenient way for analyzing continuous beams. The three-moment equa-
tion produces, in its general form, the compatibility condition that the slope of the
elastic curve be continuous at an interior support of the continuous beam. Since the
equation involves three moments – the bending moments at the support under our
observation and at the two supports adjacent to it – is commonly referred to as three-
moment equation. While applying this method, the bending moments at the interior
(and any fixed) supports of the continuous beam are taken as the redundants. The
three-moment equation is then applied at the location of each redundant locations to
obtain a set of compatibility equations that can be solved for the unknown redundant
moments. We will learn this method of analysis with the help of Example C.1.

Example C.1: Analyze the following continuous beam as shown in

Figure C.1 using three-moment equation.

SOLUTION:  The beam shown in Figure C.1 has one degree of redundancy. The
moment at support B is the redundant moment. Hence, we will start at B and its
adjacent two supports, i.e., A and C as shown previously.
The general form of three-moment equation is given by,

M ABLAB L L  M L
+ 2MB   AB + CB  + CB CB
I AB  I AB ICB  ICB

∑ P LI ) ∑ P LI
2 2
w ABL3AB
=  −
k
A AB AB

AB
(
  1− kAB
2
−
CB
k
C CB CB
( )
  1− kCB
2
−
4I AB
w L3  ∆ − ∆ B ∆C − ∆ B 
−   CB CB − 6E  A +
4ICB  LAB LCB 

FIGURE C.1  Example problem of three-moment equation.

465
466 Appendix C

It is to be noted that the redundant moments at support B have been written in

the abovementioned equation as MB. kAB and kCB are the ratios of distances of
point loads from left and right supports, respectively. All the other terms are self-
explanatory. Since in span AB we have two-point loads and along span CB, only
one uniformly distributed load is acting, we have,

LAB = 24 m

LCB = 20 m

I AB = 2I
ICB = I

1
PAB1 = 30,  kAB1 =
3

2
PAB 2 = 20,  kAB 2 =
3
w AB = 0
wCB = 2.5 kN/m

All other parameters in the abovementioned equation are zero.

Hence, once we substitute the above values in the abovementioned equation,
we will get:

M AB (24)  24 20  MCB (20)

+ 2MB  + +
2I  2I I  I

30 ( 24 ) (1/3) 20 ( 24 ) ( 2/3) 2.5 ( 20 )
( ) ( )
2 2 3

1− (1/3) − 1− ( 2/3) −
2 2
=  −
2I I 4I

Now, since A and C supports are hinge supports, by inspection we have:

M AB =  MCB = 0

Substituting these values in the abovementioned equation, we have:

MB =   −151.5 kNm
Once this redundant reaction is known, we can form the free-body diagrams of
each span separately, and by applying equilibrium equations, we can determine
the unknown support reactions at these three supports as well. Since we have
carried out these steps several times in the text, the same is left as an exercise for
the readers.
Appendix D
Solved Examples of
Selected Problems
Here we will provide a few solved examples of different types of problems. After
completing the texts in previous chapters it is suggested to students to do these prob-
lems on their own and then check the solutions provided here to bridge any gap
between computational and conceptual issues.

Example D.1: Determine the reactions at the supports for the three-

hinged arch as shown in Figure D.1.

SOLUTION:  Free-body diagram: See Figure D.1(b)

Static Determinacy: The arch is internally unstable. It is composed of two rigid
portions, AB and BC, connected by an internal hinge at B.
The arch has,
No. of support reactions, r = 4; Equation of condition due to internal hinge,
ec = 1
i.e., Degree of external indeterminacy, ie = r − ( 3 + ec ) = 4 − ( 3 + 1) = 0
The arch is statically determinate.
Support reactions:

+ ∑M C =0

10 1 2
− Ay × 20 + 4 × 10 × + × 20 × 12 × × 20 = 0
2 2 3

Ay = 90 kN ↑

FIGURE D.1  Example problem of a three-hinged arch.

467
468 Appendix D

↑+ ∑F = 0
y

1
90 + Cy − × 20 × 12 = 0
2
Cy = 30 kN ↑

+ ∑M BC
B =0

10 1 1
30 × 10 − C x × 10 − 4 × 10 × − × 10 × 6 × × 10 = 0
2 2 3
C x = 0 kN

→+ ∑F = 0
x

− Ax − 0 − 4 × 10 = 0

Ax = −40 kN →

Check:
To check the computation, we apply the equilibrium  + ∑ MB = 0 for the entire
structure.
+ ∑
MB = 0

10 1  1 
40 × 10 − 90 × 10 + 30 × 10 − 4 × 10 × + × 20 × 12 ×  10 − × 20 = 0
2 2  3 

Hence, o.k.

Example D.2:  Analyze the complex truss as shown in Figure D.2.

SOLUTION:  Inspect the truss properly. If a joint is reached, where there are
three unknowns, remove one of the members at the joint and replace it with an
imaginary member elsewhere in the truss. Here, in this case, let us remove the
member BE and instead add the member AC. Now we can analyze the truss by
the method of joints with actual loading present in the truss.
The member forces are obtained by this, let us name them as F ′ set of forces.
Calculation of F′ set of forces:
Joint E: See Figure D.3 (a)

↑+ ∑F = 0
y

FED + 20 = 0

FED = −20 kN (C )

→+ ∑F = 0
x

FEF = 0
Appendix D 469

FIGURE D.2  Problem of a complex truss.

FIGURE D.3  Example problem of a complex truss: various joint equilibriums for F ′ set of
forces.

Joint F: See Figure D.3 (b)

FFA = 0

FFC = 0
Joint D: See Figure D.3 (c)

↑+ ∑F = 0
y

5 6
20 + × FDC − × FDA = 0
6.4 10
470 Appendix D

→+ ∑F = 0
x

8 4
−FDA × − FDC × =0
10 6.4
Solving the abovementioned two equations,

FDA = 12.5 kN (T )

FDC = −16 kN (C )

Joint A: See Figure D.3 (d)

↑+ ∑F = 0
y

6 11
20 + × 12.5 + × FAC + FAB = 0
10 11.7

→+ ∑F = 0
x

8 4
12.5 × + FAC × =0
10 11.7
FAC = −29.23 kN (C )

Replacing the value of FAC in the abovementioned equation, we get FAB = 0 .

Joint A: See Figure D.3 (e)

FBC = 0 and FBA = 0

Now, consider the simple truss without the external load of 40 kN. Place equal,
but opposite collinear unit loads on the truss at the two joints from which the
member was removed. Due to these unit loads at joint B and E, let’s assume that
a U set of forces develops in the truss members.
If these forces develop a force Ui in the ith truss member, then by proportion
an unknown force X in the removed member would exert a force, XUi in the ith
member.
Calculation of U set of forces:
Joint E: See Figure D.4 (a)

→+ ∑F = 0
x

8
−1× − FEF = 0
10

FEF = −0.8 kN (C )

↑+ ∑F = 0
y
Appendix D 471

FIGURE D.4  Example problem of a complex truss: various joint equilibriums for U set of
forces.

6
1× + FED = 0
10

FED = −0.6 kN (C )

Joint F: See Figure D.4 (b)

↑+ ∑F = 0
y

FFC = 0

→+ ∑F = 0
x

−FFE − FFA = 0

FFA = −0.8 kN

Joint D: See Figure D.4 (c)

↑+ ∑F = 0
y

5 6
FDC × + 0.6 − × FDA = 0
6.4 10

→+ ∑F = 0
x

4 8
−FDC × − FDA × =0
6.4 10
472 Appendix D

Solving the abovementioned two equations, we get:

FDC = −0.48 kN (C )

FDA = 0.375 kN (T )

Joint D: See Fig. D.4 (d)

→+ ∑F = 0
x

4 8
FBC × + 1× =0
6.4 10

FBC = −1.28 kN (C )

↑+ ∑F = 0
y

5 6
FBC × − FBA − 1× =0
6.4 10

Substituting the value of FBC, we get, FBA = −1.6 kN (C ).

Joint A: See Figure D.4 (e)

→+ ∑F = 0
x

8 4
−0.8 + 0.375 × + FAC × =0
10 11.7

FAC = 1.46 kN (T )

Now, if the effects of the abovementioned two loadings are considered, the forces
in the ith member of the truss will be, Fi = Fi′+ XUi .
The member AC does not exist in the actual truss. So, we have to choose the
magnitude of X in such a way so that Fi = 0 in member AC.
For member AC we get,

′ + XU AC = 0
FAC

−29.23 + 1.46 × X = 0

X = 20.02

Finally, we get the value of the member forces in the actual truss from Table D.1.
Appendix D 473

TABLE D.1
Calculating Member Forces in the Actual Truss
Truss Members F ′ Force (kN) U Force (kN) Fi = Fi′+ XUi (kN)
AB 0 −1.6 −32.03
BC 0 −1.28 −25.62
CD −16 −0.48 −25.61
DE −20 −0.60 −32.01
EF 0 −0.80 −16.02
FA 0 −0.80 −16.02
AD 12.5 0.375 20.01
BE 0 1 20.02
FC 0 0 0

Example D.3:  Analyze the following multistory portal frame as shown

in Figure D.5 by moment distribution method.

SOLUTION:  As all the applied loads are nodal loads only, there will be no fixed-
end moments induced in the members. Only moments will be generated due to
sidesway of the portal frame.
Distribution factor at C will be:

2I /6 2
DFCD =   =
(I /6) + ( 2I /6) 3

1
DFCB =
3

Distribution factor at D will be:

2
DFDC =
3

1
DFDE =
3
Distribution factor at B will be:

2I /6 1
DFBE = =
(I /6) + ( 2I /6) + (I /6) 2

I /6 1
DFBC = =
( ) ( ) ( )
I /6 + 2I /6 + I /6 4

I /6 1
DFBA = =
(I /6) +  ( 2I /6) + (I /6) 4
474 Appendix D

FIGURE D.5  Example on multistory portal frame.

Distribution factor at E will be:

2I /6 1
DFEB = =
(I /6) +  ( 2I /6) + (I / 6) 2
I /6 1
DFEF = =
( ) ( ) ( )
I /6 +   2I /6 + I /6 4
I /6 1
DFED = =
(I /6) +  ( 2I /6) + (I /6) 4

Sway Correction for Top Story

Let us assume that δ be the sway at the top story toward right side. For this sway,
the induced moments at joints will be:

6EIδ
MBC = MCB = −
62
6EIδ
MDE = MED = −
62
MBC : MDE =  1: 1
Appendix D 475

Let us assume that

MBC = MCB = MDE = MED = −1000 Nm

The moment distribution for top-story sway toward right is provided in the follow-
ing tabular form:

Joints A B C D E F

Member AB BA BE BC CB CD DC DE ED EB EF FE
DF – 1/4 1/2 1/4 1/3 2/3 2/3 1/3 1/4 1/2 1/4 –
FEM −1000 −1000 −1000 −1000
BAL +250 +500 +250 +330 +667 +667 +333 +250 +500 +250
COM +125 +250 +107 +125 +334 +334 +125 +167 +250 +125
BAL −104 −209 −104 −153 −306 −306 −153 −104 −209 −104
COM −52 −105 −77 −52 −153 −153 −52 −77 −105 −52
BAL +46 +90 +46 +68 +137 +137 +68 +46 +90 +46
COM +23 +45 +34 +23 +69 +69 +23 +34 +45 +23
BAL −20 −39 −20 −31 −61 −61 −31 −20 −39 −20
COM −10 −20 −16 −10 −31 −31 −10 −16 −20 +10
BAL +9 +18 +9 +14 +27 +27 +14 +9 +18 +9
COM −5 +9 +7 +5 +14 +14 +5 +7 +9 −5
BAL −4 −8 −4 −6 −13 −13 −6 −4 −8 −4
COM −2 −4 −3 −2 −7 −7 −2 −3 −4 −2
BAL +2 +3 +2 +3 +6 +6 +3 +2 +3 +2
COM +1 +2 +2 +1 +3 +3 +1 +2 +2 +1
BAL −1 −2 −1 −1 −3 −3 −1 −1 −2 −1
FINAL +90 +178 +530 −708 −683 +683 +683 −683 −708 +0.530 −178 +90
MOMENT

Let the actual sway moment is x times the moment calculated for top-story sway
toward right side.

Sway Correction for Bottom Story

Let δ 2 be the sway toward right of the bottom sway:

6EIδ 2
M AB = MBA = −
62

6EIδ 2
MEF = MFE = −
62

M AB : MEF = 1: 1

Let us assume,

M AB = MBA = MFE = MEF = −1000 Nm

476 Appendix D

The moment distribution of the assumed sway moments for the bottom story is
provided in the below table:

Joints A B C D E F

Member AB BA BE BC CB CD DC DE ED EB EF FE
DF – 1/4 1/2 1/4 1/3 2/3 2/3 1/3 1/4 1/2 1/4 –
FEM −1000 −1000 −1000 −1000
BAL +250 +100 +250 +250 +500 +250
COM +125 +250 +125 +125 +250 +125
BAL −63 −124 −63 −42 −83 −83 −42 −63 −124 −63
COM −32 −62 −21 −32 −42 −42 −32 −21 −62 −32
BAL +21 +41 +21 +25 +49 +49 +25 +21 +41 +21
COM +11 +21 +13 +11 +25 +25 +11 +13 +21 +11
BAL −9 −16 −9 −12 −24 −24 −12 −9 −13 −9
COM −5 −8 −6 −5 −12 −12 −5 −6 −8 −5
BAL +4 +6 +4 +6 +11 +11 +6 +4 +6 +4
COM +2 +3 +3 +2 +6 +6 +2 +3 +3 +2
BAL 2 −2 −2 −3 −4 −5 −3 −2 −2 −2
COM −1 −1
FINAL −900 −700 +609 +190 +75 −75 −75 +75 +190 +601 +799 −900
MOMENT

Let the actual moment be y times the moment obtained from assumed moment
values. Thus, total moment in various members will be:

M AB = 90 x − 900y

MBC = −708 x − 190y

MCB = −683x − 75y

MDC = 683x − 75y

MED = −708 x − 190y

MEF = 178 x − 799y

MBA = 178 x − 799y

MBE = 530 x + 600y

MCD = 683x − 75y

MBD = −683x + 75y

MDE = 530 X + 609y

MFE = 90 x − 900y
Appendix D 477

Now, shear force at the base of top story must be zero to maintain the equilibrium
if forces,

HB + HE + P = 0

MBC + MCB MDE + MED

+ + 1000 = 0
6 6

Also, total summation of moment will also need to be zero to maintain moment
equilibrium conditions.

MBC + MCB + MDE + MED = −6000

−708 x + 190y − 683x + 75y − 683x + 75y − 708 x + 190y = −6000

or,

−2782x + 530 x = −6000

− x + 0.1905y = −2.157

Similarly, shear at bottom of the story needs to be zero to maintain force equilib-
rium conditions.

HA + HE + P = 0

or,

M AB + MBA MEF + MFE

+ + 1000 + 3000 = 0
6 6

M AB + MBA + MEF + MFE = −24,000

90 x − 900y + 178 x − 799y + 90 x − 900y + 178 x − 799y =   −24,000

536 x − 3398y =   −24,000

x − 6.339y =   −44.27

On addition of these final equations, we get:

−6.1485y =   −46.927

or,

y =   +7.633

x =   −44.77 + 63.339 × 7.633 = +3.63

478 Appendix D

Once we got the values of unknown x and y, we need to just substitute back them
in the abovementioned total moment expressions in terms of these parameters.
The final moments are provided in a table below:

Joints A B C D E F
Member AB BA BE BC CB CD DC DE ED EB EF FE
3.63 times +325.8 +645 +1918 −2563 −2473 +2473 +2473 −2473 −2563 +1918 +645 +325.8
sway
moment
for top
story
7.633 −6869.7 −6099 +4648 +1451 +572.5 −572.5 −572.5 +572.5 +1451 +4648 −6098 −6869.7
times
sway
moment
for
bottom
story
Final −6543.9 −5454 +6566 −1112 −1900.5 +1900.5 +1900.5 −1900.5 −1112 +6566 −5454 −6543.9
moment

Example D.4: Analyze the following frame as shown in Figure D.6

by strain energy method. Support A is hinged and support D is fixed.

SOLUTION:  We have shown all the support reactions at the support points A
and D in Figure D.6. Let MD be the fixed-end moment at the support D. Taking
moment about D, we get:

MD + RA × 4 + HA × 3 − 4000 × 2 − 6 × 1000 × 3 = 0

3 M
RA = 6500 − H− D
4 4

As per the support conditions, the supports A and D will not yield. Hence, partial
derivative of strain energy with respect to HA and MD will be zero separately. Refer to
the following table for complete analysis of the given frame by Castigliano’s method.

∂M ∂M
Origin at Portion Bending Moment Limits
∂H ∂ MD
A AB H×y y y 0→3
 3 3  x
B BE  6500 − H −  x + 3 − 0→ 2
4 4  4
M 
− D  x − 3H
4 
3  x  x
B BC 6500 x −  x + 3 H −  x + 3 − 2→ 4
4  4  4
MD
− x − 4000 ( x − 2)
4
D DC (6000 − H ) y −y −1 0→6
y2
−  MD − 1000
2
Appendix D 479

FIGURE D.6  Example of frame analysis problem.

Now, let us first take partial derivative of strain energy with respect to horizontal
reaction HA at A:

6
∂U 1
∂H EI ∫
=   Hy × ydy
0
2
1  3  M   3 

+
2EI 
0
∫
  6500 x −  x + 3 H − D x     −  x + 3  dx
4  4   4 
4
1  3  MD   3 
+
2EI ∫ 6500x −  4 x + 3 H −
2
4
− 4000 ( x − 2)     −  x + 3  dx
  4 
6
1  y2 
+
EI 
0
∫
  (6000 − H ) y −  MD − 1000   − y  dy   = 0 
2

So,

4
1  
3
 y3  3  3   M 3
2

 3 0 2 0  ∫
H   −   6500  x 2 + 3x  − H   x + 3  − D  x 2 + 3x   dx
4  4 0  4 4 

4 6 6 5
3 3   y3   y2   y4 
4∫
+ 2000   x 2 + x − 6 dx − (6000 − H )   + MD   +  500    = 0
2
2   3 0  3 0  4 0
480 Appendix D

or,

4 4 4
H 4  3  
3
6500  x 3 3 2  M  x3 3 
9H −    + x  +      x + 3  + D    + x 2 
2  4 2 0 2  9  4  
 8  4 2 0
0

4
 x 3 3x 2 
+ 2000   + − 6 x  − (6000 − H ) × 72 + 18MD + 162,000 = 0
4 4 0

So, on simplification we get:

9H −
6500
2
[
H 4
(M
)
16 + 24 ] + × 63 − 32 + D (16 + 24 )
2 9 8

+ 2000  
(
 43 − 23
+
)
3 2
( )

4 − 22 − 6 ( 4 − 2) 
 4 4 

− 432,000 + 72H + 18MD + 162,000 = 0

or,

5.347H + MD = 16,434.8

Second, we take partial derivative of strain energy with respect to MD to get:

4
∂U 1  3  M  x 
=
∂MD 2EI 
0
∫
  6500 x −  x + 3 H − D x   −  dx
 4  4  4
4
1  3  M   x
+
2EI 
2
∫
  6500 x −  x + 3 H − D x − 4000 ( x − 2)    −  dx  
4  4   4
6
1  1000y 2 
+
EI ∫ (6000 − H ) y − M
0
D −
2 
 ( −1) dy = 0 

So,

4 4
1  6500 2 H  3 2  M x 
2

∫ (x )
4800
−
∫ 
2  4
0
x −  x + 3x  − D  dx +

4 4  16  8
0
2
− 2x dx

∫
−   (6000 − H ) y − MD − 5000y 2  dy = 0
0

Upon carrying out the integrations and simplification, we finally get:

52,000 2  56 
− + 5H + MD + 500  − 12 − 108,000 + 18H + 6MD + 36,000 = 0
3 3  3 
Appendix D 481

or,

5.347H + MD = 16,434.8

Second, we take partial derivative of strain energy with respect to MD to get:

4
∂U 1  3  M  x 
=
∂MD 2EI 
0
∫
  6500 x −  x + 3 H − D x   −  dx
4  4  4
4
1  3  M   x
+
2EI 
2
∫
  6500 x −  x + 3 H − D x − 4000 ( x − 2)    −  dx  
4  4   4
6
1  1000y 2 
+
EI ∫ (6000 − H ) y − M
0
D −
2 
 ( −1) dy = 0 

So,

4 4
1  6500 2 H  3 2  M x 
2

∫ (x )
4800
−
∫ 
2  4
0
x −  x + 3x  − D  dx +
4 4  16  8
0
2
− 2x dx

∫
−   (6000 − H ) y − MD − 5000y 2  dy = 0
0

Upon carrying out the integrations and simplification, we finally get:

52,000 2  56 
− + 5H + MD + 500  − 12 − 108,000 + 18H + 6MD + 36,000 = 0
3 3  3 

20
23H + MD = 86,000
3

3.45H + MD = 12,900

Solving these two final equations, we get:

H = 1863 N

MD = 6472 Nm

MBA = −H × 3 = −1863 × 3 = −5589 Nm

Taking moment about C, we get:

MCD + (6000 − H ) × 6 − MDC − 1000 × 6 × 3 = 0

or,

MCD = 18,000 + 6472 − 4137 × 6 = −340 Nm

482 Appendix D

Example D.5:  A Continuous beam ABC as shown in Figure D.7, having

uniform cross section has two equal spans of AB and BC each of length l .
During application of imposed loads, the support B sinks by an amount
δ 1 and support C by an amount δ 2. Find the reaction at the supports at
this situation in terms of settlement parameters and EI.

SOLUTION:  Let MB be the moment at B.

Taking moments about B,

RA × l + MB = 0

or,
MB
RA = −
l

Similarly, taking moment about B for BC, we will get:

MB
RC = −
l

RA = RC = R

RB = −2R

And for AB,

M =R×x

Strain energy of beam ABC,

l
M 2dx
=2
∫
0
2EI

or,
l
(Rx )2 dx
2
∫
0
2EI

FIGURE D.7  Example problem of a continuous beam with support settlement.

Appendix D 483

or,

l
R2  x3 

EI  3  0

i.e.,

R 2l 3

EI

So, the total work done on the structure,

R 2l 3
W= − 2R × δ 1 + R × δ 2
EI

For work to be maximum,

∂W
=0
∂R

Upon carrying out the partial derivative, we will get:

 2δ 1 − δ 2
R = 3EI
2l 3

Hence,

 2δ 1 − δ 2
RA = RC = 3EI
2l 3

And,

 2δ 1 − δ 2
RB = −2R = −3EI
2l 3

Example D.6: All the members of the below frame have same cross
section A and same EI. Find the force in each member due to the applied
loading as shown in Figure D.8.

SOLUTION:  The given frame is a redundant frame with more members than
required for stability. Let us take member BC as the redundant member. We will
remove this redundant member from the original frame to make it a deterministic
structure. The same is shown in Figure D.9.
In Figure D.9, we have marked all the unknown support reactions at the sup-
ports A and D of the determinate frame. In the next step, we will apply unit loads
at nodes B and C in the direction of the redundant member BC. The same is also
shown in Figure D.9.
484 Appendix D

FIGURE D.8  Example problem of redundant frame with loading.

FIGURE D.9  Determinate frame with redundant member BC removed.

The force in various members due to this unit load is provided in the following
tabular form.

Length Final Force

Member in cm. Force F k k2 Fkl k 2l F + kX
AB 300  0 +3/4 9/16  0  2700/16 −2568
CD 300 +7500 +3/4 9/16 +6,750,000/4  2700/16  4432
DA 400  0 +1 1  0  400 −3425
AC 500 −12,500 −5/4 25/16 +31,250,000/4  12,500/16 −8220
BD 500  0 −5/4 25/16  0  12,500/16 +4280
∑ +38,000,000/4 +36,000/16
Appendix D 485

where k is the force due to unit load in various members in the determinate frame
after removing the redundant member from the original structure. And force F
is the force in the determinate frame due to applied load of 10,000 N at joint C.
We have:

∑ (Fkl /AE )
X=
( )
∑ k l /AE + ( l0 /A0E )
2

where l0 and A0 are the length and area of cross section of redundant member
that has been removed from the original structure to convert it into a determinate
structure.
Since all the member having same cross-sectional area a and length of redun-
dant member as per Figure D.8 is 400 cm, we have,

∑ Fkl
X=
∑ k 2l + l0

Putting the values, we will get:

X =   −3425 N

Final force in each member will be F + kX , and the same has already been pro-
vided in table’s last column.

Example D.7:  Find the fixed-end moment for the fixed beam as shown
in Figure D.10 using column analogy method.

SOLUTION:  The basic determinate structure of the above fixed beam will be a
cantilever beam AB. Bending moment diagram and loading on analogous column
of the same is drawn sequentially in Figure D.11 for reference. Please note that
load acting in analogous column is −Ms , where Ms is the bending moment in
determinate structure.

FIGURE D.10  Fixed beam with point load.

486 Appendix D

FIGURE D.11  (a) Determinate cantilever, (b) bending moment diagram, (c) loading on
analogous column, and (d) section of analogous column.

Distance of CG from the left support point A is given by,

2 × (1/E ) × 1+ 4 × (1/ 2EI ) × ( 2 + 2) 2 + 8

x =  = = 2.5 m
2 × (1/EI ) + 4 × (1/ 2EI ) 2+ 2

Moment of inertia of the analogous column is given by,

1 1 2 1 1 4
Iy = × × 23 + × 1.52 × + × × 43 + × 1.52
12 EI EI 12 2EI 2EI

or,

12.333
Iy =
EI
Appendix D 487

Area of the analogous column is given by,

1 1 4
A= × 2+ ×4=
EI 2EI EI

Net force acting on the analogous column is given by,

1 6000 6000
P= × × 2=
2 EI EI

Eccentricity,

2 5.5
e = 2.5 − =
3 3

Hence, net stress acting in analogous column at A,

P Pex
fA = +
A Iy

where in abovementioned equation of stress, x = 2.5 m

So, we have:

P Pex
fA = + = 3729.7
A Iy

Hence, net stress acting in analogous column at B,

P Pex
fB = +
A Iy

Here x = 6 − 2.5 = 3.5 m

Thus,

fB = 1621.6

So, the final moments at A and B will be,

M A =   −Ms +   fA =   −6000 + 3729.7 =   −2270.3 Nm

MB = 0 − fB = −1621.6 Nm
Bibliography
1. ASCE Standard Minimum Design Loads for Buildings and Other Structures. (2010)
ASCE/SEI 7-10, American Society of Civil Engineers, Virginia.
2. Arbabi, F. (1991) Structural Analysis and Behavior. McGraw-Hill, New York, NY.
3. Bathe, K.J., and Wilson, E.L. (1976) Numerical Methods in Finite Element Analysis.
Prentice Hall, Englewood Cliffs, NJ.
4. Beer, F.P., and Johnston, E.R.., Jr. (1981) Mechanics of Materials. McGraw-Hill,
New York, NY.
5. Betti, E. (1872) II Nuovo Cimento. Series 2, Vols. 7–8.
6. Boggs, R.G. (1984) Elementary Structural Analysis. Holt, Rinehart & Winston,
New York, NY.
7. BS 8110-1. (1997) Structural use of Concrete-Part 1: Code of practice for design and
construction.
8. Chajes, A. (1990) Structural Analysis, 2nd ed. Prentice Hall, Englewood Cliffs, NJ.
9. Colloquim on History of Structures. (1982) Proceedings, International Association for
Bridge and Structural Engineering, Cambridge, England.
10. Cross, H. (1930) “Analysis of Continuous Frames by Distributing Fixed-End Moments.”
Proceedings of the American Society of Civil Engineers 56, 919–928.
11. Elias, Z.M. (1986) Theory and Methods of Structural Analysis. Wiley, New York, NY.
12. Gere, J.M., and Weaver, W. Jr. (1965) Matrix Algebra for Engineers. Van Nostrand
Reinhold, New York, NY.
13. Glockner, P.G. (1973) “Symmetry in Structural Mechanics.” Journal of the Structural
Division, ASCE 99, 71–89.
14. Goldstein, H., Poole, C.P., and Safko, J. (2011) Classical Mechanics. Pearson Education,
India.
15. Hibbeler, R.C. (2012) Structural Analysis, 8th ed. Prentice Hall, Englewood Cliffs, NJ.
16. Holzer, S.M. (1985) Computer Analysis of Structures. Elsevier Science, New York, NY.
17. Hourani, M. (2002) Mathematical Model of Influence Lines for Indeterminate Beams,
Proceeding of the 2002 American Society for Engineering Education Annual Conference
& Exposition, American Society for Engineering Education.
18. International Building Code. (2012) International Code Council, Chicago, IL.
19. IS-875 Part – I (1997) Indian Standard Code of Practice for Design Loads (Other
Than Earthquake) for Buildings and Structures Part 1 – Dead Loads: Unit Weights of
Building Materials and Stored Materials, 2nd ed.
20. IS-875 Part – II, (1987) Indian Standard Code of Practice for Design Loads (Other
Than Earthquake) for Buildings and Structures Part 2 – Imposed Loads, 2nd revision,
Reaffirmed 2008.
21. IS 875 Part – III (2015) Indian Standard Code of Practice for Design Loads (Other Than
Earthquake) for Buildings and Structures Part 3 – Wind Loads, 3rd revision.
22. IS1893 Part – I (2002) Indian Standard Criteria for Earthquake Resistant Design of
Structures Part – 1 General Provisions and Buildings, 5th revision.
23. Kassimali, A. (2011) Matrix Analysis of Structures, 2nd ed. Cengage Learning,
Stamford, CT.
24. Kassimali, A. (2014) Structural Analysis, 5th ed. Cengage Learning, Boston, MA.
25. Kennedy, J.B., and Madugula, M.K.S. (1990) Elastic Analysis of Structures: Classical
and Matrix Methods. Harper & Row, New York, NY.
26. Laible, J.P. (1985) Structural Analysis. Holt, Rinehart & Winston, New York, NY.

489
490 Bibliography

27. Langhaar, H.L. (1962) Energy Methods in Applied Mechanics. Wiley, New York, NY.
28. Laursen, H.A. (1988) Structural Analysis, 3rd ed. McGraw-Hill, New York, NY.
29. Leet, K.M. (1988) Fundamentals of Structural Analysis. Macmillan, New York, NY.
30. McCormac, J. (1984) Structural Analysis, 4th ed. Harper & Row, New York, NY.
31. McCormac, J., and Elling, R.E. (1988) Structural Analysis: A Classical and Matrix
Approach. Harper & Row, New York, NY.
32. McGuire, W., and Gallagher, R.H. (1979) Matrix Structural Analysis. Wiley, New York,
NY.
33. Maney, G.A. (1915) Studies in Engineering, Bulletin 1. University of Minnesota,
Minneapolis, MN.
34. Manual for Railway Engineering. (2011) American Railway Engineering and
Maintenance of Way Association, Maryland, VA.
35. Maxwell, J.C. (1864) “On the Calculations of the Equilibrium and Stiffness of Frames.”
Philosophical Magazine 27, 294–299.
36. Marium, J.L., and Craig, L.G. (2017) Engineering Mechanics (Part-1 – Statics). Wiley,
India.
37. Noble, B. (1969) Applied Linear Algebra. Prentice Hall, Englewood Cliffs, NJ.
38. Norris, C.H., Wilbur, J.B., and Utku, S. (1976) Elementary Structural Analysis, 3rd ed.
McGraw-Hill, New York, NY.
39. Parcel, J.H., and Moorman, R.B.B. (1955) Analysis of Statically Indeterminate
Structures. Wiley, New York, NY.
40. Petroski, H. (1985) To Engineer Is Human—The Role of Failure in Successful Design.
St. Martin’s Press, New York, NY.
41. Popov, E.P. (1968) Introduction to Mechanics of Solids. Prentice Hall, Englewood
Cliffs, NJ.
42. Rao, D.S.P. (1997) Graphical Methods in Structural Analysis. University Press, India.
43. Ramamrutham, S., Narayan, R. (2003) Theory of Structures. Dhanpat Rai Publishing
Company (P) LTD., New Delhi.
44. Timoshenko, S. (1930) Strength of Materials, D. Van Nostrand Company Inc., Toronto,
NY, London.
45. Uniform building code. (1997), Vol. II.
46. Vazirani, V.N., Ratwani, M.M. (2002) Analysis of Structures, Vol. II. Khanna
Publishers, New Delhi.
47. Wang, C.K. (1983) Intermediate Structural Analysis. McGraw-Hill, New York, NY.
Index
Note: Locators in italics represent figures and Bow’s notation, 101–7, 102
bold indicate tables in the text. Boyle’s law of gas, 41, 41
Bridge girder, 192
A Building frames; see also Lateral (horizontal)
loads; Vertical loads
Accumulation phase (1900–1925), 7 example problem, 263
Active earth pressure coefficients, 24 free body diagram, 264
Analytical model, 25, 27 multi-bay simplified 261
Ancient Egyptian (3150–323 BCE) builders, 3 shear force distribution, 262
Angular displacements, 414
Antisymmetric loading C
boundary conditions, 246, 247
continuous beam, 244, 245 Cables
displacement behavior, 246, 247 concentrated loads, 217–21, 218–20
examples, 244, 245 external load, 217
indeterminate beam, 247, 248 structural analysis, 217
substructures, 248 support and transfer loads, 217
superposition of,, 246 Cable-stayed bridge girder, 224, 224–6
Application phase (1700–1775), 6 Cantilever beams, 125–8, 126, 127, 152, 152–4,
Applied force, 121 174, 176
Approximate analysis Cantilever method, 265–9, 266, 267, 269
forces and reactions distribution, 258 Carry over factor, 333, 414
point of inflection, 256, 257 Cartesian coordinates, 39, 40
redundants, 255 Castigliano’s theorem, 173–7, 174, 176, 273, 274
unknown reactions and/or internal forces, 256 Catenary, 14
Arches, 14–15 Center of gravity (CG), 35–6
components of, 226, 226 Classical phase (1875–1900), 7
funicular, 226 Closed-loop formations, 69, 70
types of, 226, 227 Coefficient of thermal extension, 33–4
vertical reactions development, 226 Collapse load factor, 441, 445–8, 449–52, 450–2
Archimedes of Syracuse (287–212 BCE), 4 Collapse mechanism, 453
Areas and centroids, 457–8 Column analogy method, 485–7, 485, 486
Axial force, 61, 109–11, 122 curved beams, 401, 401
Axial load, 161 indeterminate structure, 402–6
Axial/longitudinal strain, 33 moment distribution, 399
moment of inertia, 400–1
B short column section, eccentric loading, 399,
400
Babylonian clay tablet, 4 stiffness and carry over factors, 406,
Beam analysis, 430–2, 431, 454 406–7
Beam-column elements, 14 Combined mechanism, 455
Beam elements, 16 Compatibility conditions/equations, 251, 271
Beams deflection, 170, 170–1 Compatibility method 293
Bending moment (BM) diagram, 52, 56–7, 58, Complex truss, 87, 87, 468–73, 471, 471,
60, 60, 61, 122–8, 125, 134, 134–5, 473; see also Henneberg’s method
135, 138, 152, 152, 158, 183, 184, 185, Compound truss, 87, 87, 111–13
193, 278, 279 Compression structures, 14–15, 15
Bending structures, 16–17 Concentrated loads, 197–204, 199, 200, 202, 203,
Bend strength, 33 208–13,–209–13
Betti’s Law, 177, 177–9 components, 218
Biaxial bending effects, 16 description, 217, 218

491
492 Index

point loads, 218–20, 219, 220 Equivalent static forces analysis, 22

Pythagorean theorem, 218 Equivalent static method, 22
Configuration, 40–1, 41 Establishment phase (1850–1875), 7
Conjugate beam method, 150–4, 151, 151–3 Euler-Bernoulli beams, 17, 157
Conservation space laws, 46 Euler-Lagrange equation of motion, 41, 43–5
Consolidation period (1900–1950), 7 Evaluation periods, 5, 6
Constitution phase (1825–1850), 7 Exact analysis of structures, 259
Continuous beams, 251, 316, 316–18, 318, 339, External and internal forces, 53–8, 54–8
339–43, 341 External determinacy, 90
Coplanar trusses External indeterminacy, 75
compound and complex, 87, 87 External stability of structures, 74–5, 75
internal stability, 89 External static indeterminacy, 64, 72
imple, 86, 87
Correct load factor, 444 F
Cross-sectional area, 35
Fabrication errors, 25, 254, 282, 282–4, 283
D First moment area theorem, 148
Fixed arch, 226, 227
D’Alembert’s force law, 47 Fixed end moments (FEMs), 327–8, 391, 407–8,
Dead loads, 17, 19, 19–20 408, 409, 414
Deflected shapes, 143–5, 144, 148 Fixed-fixed condition, 310
Deflection line diagram, 129, 150, 174 Fixed-hinged condition, 311
Deflection of structure, 143 Flexibility coefficients, 179, 277, 278
Deformable bodies Flexural load, 53
angles of inclination, 167 Flexural strength, 33
principle, 166 Floor girders, defined, 190
virtual forces, 166 Flooring system, 194
Deformed shape, 158 Force method, 271
Degree of external indeterminacy, 90 Force transformation matrix, 424, 424–5
Degree of indeterminacy, 65–8, 70–1, 71, 72–3, Former, defined, 34
73, 74 Foundation settlement effect, 253
DF, see Distribution factor Frame analysis, 434–6
Diffusion phase (from 1975 to date), 8 Frames deflection, 171–3, 172
Discipline formation period (1825–1900), 7 Framing systems/frameworks, 17
Displacement coefficient, 391 Free-body diagram, 50–1, 56, 94, 96, 122, 122,
Displacement contributions, 397–9 124, 130, 184, 187, 221, 222, 228,
Displacement transformation matrix, 423, 229–31, 314, 330, 467,
423–4 Funicular arches, 226
Distribution factor (DF), 329, 331–2, 391–7, 473 Funicular polygon, 104, 104, 105
Double integration method, 146–7, 146
Ductile material, 437, 438 G

E Gauss-Jordan elimination process, 462–4

Generalized coordinates, 39
Earthquake loads, 22–3 Generalized Maxwell’s law, 177, 177–9
Elastic assumptions, 34 Generalized momentum, 40
Elastic constant of material, 29 Generally curved lines, 285
Elastic curve, 143–5, 144 Geometric nonlinearity, 35
Elastic line derivatives, 128, 160 Geometric shapes, 457–8
Elastic neutral axis, 36, 36 Geometry-independent method, 176
Elastic section modulus (S), 38 Girders, 190
Energy principles, 177 Global equilibrium, 48, 50
‘Envelope of maximums’, 206, 206, 207, 208 Global moment equilibrium conditions, 181, 186
Equilibrium of structures Graphical analysis, 107
external forces and moments, 47 Gravity load path, 17
global and local, 48–50, 49, 50 The ‘Great Bath’ of Mohenjo-daro (3300–1300
mutually perpendicular axes, 48 BCE), 3, 4
Index 493

H members without relative lateral

displacement, 353–54, 354, 355, 358,
Hamiltonian mechanics, 39, 41 358
Hamilton’s equation of motion, 45–6 members with relative lateral displacement,
Hardy Cross method, 327 362–5, 363, 364, 365
Henneberg’s method, 107–11, 108, 109 Karnak Temple (2000–1700 BCE), 3
Hooke’s law, 6, 159, 160, 162 Kinematic theorem, 444–5
Horizontal loading, 17, 18, 370, 378, 380, 381–5, Kinematic indeterminacy, 63
386 defined, 77
Horizontal support reactions, 388 pin-jointed structures/truss, 78–9, 79
Hydrostatic/water pressure, 24, 24–5 rigid-jointed structures, 79–83, 80–2, 82
types of degrees of freedom, 78
I 2D pin-jointed structure, 77, 77
Kinematic method, 449–52, 450–2
Ideally fixed girder, 258
Imhotep (Pharaoh Djoser’s official), 3
Indeterminacy, see Structural indeterminacy
L
Indeterminate beams Lagrangian mechanics, 39, 41
computer programs, 298 Lateral load path, 17
flexibility method, 298 Lateral (horizontal) loads
mathematical model, 298–9, 299 cantilever method, see Cantilever method
vertical reactions, 299–302, 300, 302 portal method, see Portal method
Indeterminate set, 63 Live loads, 20
Indus Valley Civilization, 3 Load combinations, 25, 26
Influence line diagram, 82, 182–5, 183, 185–7, Load path, 17
192, 193, 195, 197, 214; see also Floor Loads, 417
girders; Trusses Local equilibrium, 48, 50
Influence lines; see also Müller-breslau’s Lower bound theorem, 444
principle
beams
indeterminate, 285, 286
M
Maxwell’s law, 287 Macaulay’s method, 154–6, 154
unit load, 288–9, 289 Material nonlinearity, 35
trusses Matrix algebra, 459–64
member force, 293 Matrix inverse, 461
primary, 292, 293 Matrix stiffness method
Initial phase (1775–1825), 7 analysis process, 419–20
Innovation phase (1950–1975), 8 local coordinates, 420–2, 421
Integration period (from 1950 to date), 8 Maximum influence
Internal forces, 122, 122 concentrated loads, 197–204, 199, 200, 202,
Internally unstable structures, 76 203
Internal reactions, 53 uniformly distributed live load, 204, 204–5,
Internal stability of structures, 75, 75–6, 76, 89 205
Internal static indeterminacy, 64, 68–70, 72–4, Maximum principle, 444
73, 74 Maxwell Law, 177, 177–9
Internal stresses, 233–4 Member forces of structures, 53
International codes, 23, 25 Member global stiffness matrix, 425–8, 426,
Invention phase (1925–1950), 8 432–4, 433
Member relative stiffness factor, 332
J Member stiffness factors, 328, 328–9
Method of consistent deformations
Joint stiffness factor, 329, 329–31, 330 multiple degrees of indeterminacy, 276, 276
three-span continuous beam, 277
K unit load diagrams, 277
Method of joints, 92–95, 93, 94
Kani’s method Method of least work, 273–6
history of, 353 Method of sections, 95, 95–796
494 Index

Method of tension coefficients, 97–100, 97–101 Plane frame with external loading, 136, 137
Minimum principle, 444–5 Plane truss
Modulus of elasticity, 29, 30 analysis example, 93
Modulus of rigidity, 32 defined, 85
Modulus of rupture, 33 fundamental triangular element, 85, 86
Mohr’s Theorem, 150 Plastic analysis, 437
Moment-area method, 147–50, 148, 149 Plastic hinge formation, 440, 441, 442
Moment distribution method, 7, 473–6 Plastic moment, 438, 438–40, 439
beam pin supported at far end, 415, 415 Point load, 128, 146, 151, 151, 154, 174, 176, 190
carry over factor, 333 Point of inflection, 256, 257
DF, 331–2 Poisson’s ratio, 33
example problem, 333, 333–4, 335 Polar diagram, 103, 103, 105
FEMs, 327–8 Portal frames, 386–7, 409, 409, 410, 412, 452–6,
oint stiffness factor, 329, 329–31, 330 453–5
member relative stiffness factor, 332, 328, Portal method
328–9 building frame analysis, 259, 259
sign convention, 327 with internal hinges, 259, 260
support settlement, 416–17 symmetric beam member forces, 262–5
and loading, see Symmetric beams multi-bay simplified building frame, 261
Moment of inertia, 36–8, 37 sectional force free body diagram, 260
Müller-Breslau principle, 7, 188, 188–90, 189, shear force distribution, 262
294–7, 296, 297 Post and Lintel-type constructions, 4
Multistorey frames, 350–351, 349, 350, 473–74, Preparatory period (1575–1825), 5, 7
474 Primary beams, 17, 18
Primary structure, 271, 272
N Principle of elasticity, 29
Principle of superposition, 83–4, 84, 273
Neutral layer/neutral axis, 36, 157, 158, 159, 159 Propped cantilever, 442, 442
Neutral plane, 36 P system, 178
Newton’s Laws of motion, 47, 177 Pure bending, 157, 158, 161
Nodal points, 370, 378, 380, 381–5, 386 Pythagoras of Samos (about 582–500 BCE), 4
Nonconcurrent forces, 102, 102, 104 Pythagorean theorem, 218
Nonprismatic members; see also Moment
distribution Q
applications, 413
deflections and loading properties, 413–14, Q system, 178
414 Qualitative influence lines
large span structures, 413 building frame, 296–7
procedure, 295
O shear force/bending moment, 295

Orientation phase (1575–1700), 5 R

Orthogonal coordinates, 39
Radius of curvature, 160
P Real-world behaviors, 35
Reciprocal deflections, 177, 177–9
Panel shear, 191 Redundant members, 251
Parthenon of Athens (447–432 BCE), 5 Redundant restraints, 271
Partitioning of matrices, 461–2 Relative joint translation, 417
Passive earth pressure coefficients, 24 Release of internal reactions, 58, 60, 60
Phase space of particles, 40 Rigid bodies, 163–6
Pin-jointed structures/trusses, 13, 57 free body diagram, 163, 165
frictionless pin joints, 72 principle, 163
kinematic indeterminacy, 78–9, 79 virtual displacement, 163, 163, 164
member forces and external support Rigid-jointed structures, 13, 79–83, 80–2, 82
reactions, 72 defined, 64
shortcut method, 73, 73–4, 74 shortcut method, 68, 69, 73
Index 495

S Static indeterminacy; see also Pin-jointed

structures/trusses
Safe theorem, 444 dimensions, 64
Safety factors, 25 of frame, 65–8
Secant modulus (Es), 31, 31 rigid structures, see Rigid structures
Secondary beams, 17 Static method, 444–8
Second moment of area, 36–8, 37, 148, 149 Stiffening girders
Seismic mass of structure, 22 bending moments and shear forces, 235, 238,
Shear, 122–8 238–9
bending moment diagram, 311–15, 312, 314 equilibrium of, 235, 236
coefficient of materials, 29 example problem, 239, 239
deformable beams, 17, 157 reaction forces, 240
force diagrams, 278, 278 three-hinged, 235, 236
force (SF), 52, 55, 61, 122, 125, 131–4, 135, two load systems, 236, 237
138, 153 uniform/equal load, 235
modulus, 32 Stiffness coefficients, 414
structures, 16, 16 Stiffness factor, 414
Shortcut method, 68–70, 69, 70, 73, 73–4, 74 member pin supported at far end, 335–6, 336
Sidesway, 341–9, 365, 366–78, 366, , 369, 370, symmetric beam, see Symmetric beams
370 Strain energy method, 161, 162, 478–480, 479
degrees of freedom, 318–26, 319, 321–6 Stress distribution, 159
and unequal columns, 388 Stress-strain curve, 437, 438
Sign conventions, 51–3, 52, 53 Stringers, 190
Simple stress-strain relationship, 29–30, 30 Structural engineering method, 8, 9
Simply supported beam, 130, 146, 154, 181, 182, Structural indeterminacy, 63
252, 441 Structural members, 17
Simply supported girder, 258 Structure stiffness equation, 428–9
Single degree of indeterminacy Support moment, 187
energy principle, 274 Support point, 418
propped cantilever, 271, 272 Support reaction forces, 58, 59, 187, 232–3,
reverse direction, 272 467–468
Slope deflection method Support settlements, 279–81, 280, 407–8, 408,
angular displacement, 306, 307 409, 416–17
displacement method of analysis, 305 Sway correction, 476–8
general loading with deflection diagrams, Sway mechanism, 455
305, 306 Symmetric beams
loading conditions, 309, 310 with antisymmetric loading, 416
moments, 309 and loading, 336–7, 337, 416
superposition principle, 306 unsymmetric loading, 337, 338
support settlement, 307–8, 308 Symmetric loading
uniformly distributed load, 309, 309 frame structure, 243, 244
Snow loads, 21–2 internal loadings and deflections, 243
Space truss, 430 loading conditions/geometry, 243
defined, 85 structure, truss, 243, 244
fundamental building element, 112, substructures, 247
112, 113 Symmetry concept, 46
joint equilibriums, 115
support reaction, 114–18, 117
3D version, plane trusses, 112 T
Statically indeterminate structures; see also Tangent modulus of elasticity (ET ), 31
Approximate analysis Temperature change effect, 254, 282, 282–4,
advantages of, 252, 252–3 283
disadvantages of, 253, 253–4, 254 Tension structures, 13–14, 14
force and displacement methods, 255 Theory of plates, 8
member sizes and orientations, 251 Three-hinged arch, 465, 465, 467
optimized economic results, 255 description, 227, 227
unknown reactions and internal forces, 251 method of sections, 227, 228
496 Index

stiffening girders, 235–41, 236–40 Ultimate shear strength, 33

support reactions,228, 229 Ultimate tensile strength, 33
Three-moment equation, 465, 465 Uniform Building Code (UBC), 21
Three-span continuous beams, 278, 279 Uniformly distributed load (UDL), 201, 204,
Tie arch, 227, 227 204–5, 205, 207, 217, 221, 221–6, 224,
Timoshenko beams, 17, 157 309, 309
Torsion, 33 Uniqueness theorem, 445
Transverse/lateral strain, 33 Unit load application, 168
Transverse rupture strength, 33 Unit point load, 181, 182, 185, 192, 193
Trial-and-error method, 198, 199 Unsafe theorem, 444–5
Truss, 15; see also Coplanar trusses; Plane truss; Unstable trusses, 90–1
Space truss Upper bound theorem, 444–5
assumptions, 88
deflection, 167–9, 168, 169
V
global equilibrium conditions, 87
flooring system, 194 Vertical compression, 16
for graphical analysis, 105 Vertical loads, 258, 258–9, 365, 365–78, 366,
free-body diagram, 196, 196 369, 370, 370
gusset plates, 88, 88 Virtual work method; see also Deflection of
structural elements, 85 beams
support reactions, 195, 195 angular rotation, 163
tension/compression convention, 89 deflection of truss, 167–9, 168, 169
types of, 85, 86 deformable bodies, 166–7
typical deck/floor slab supporting rigid bodies, 163–6
arrangements, 194, 194 total angular displacement, 162
Two-dimensional Cartesian plane, 42–3 Volume vs. pressure curve, 41, 41
Two-dimensional (2D) truss system
with axial force, 49, 50
W
coordinate transformation, 422, 422–3
displacement transformation matrix, 423, Wind loads, 20–1
423–4
with external loading, 48–9, 49
external loading and support reactions, 49, 50
Y
local coordinates, 420–2, 421 Yield criteria, 444
member global stiffness matrix, 425–8, 426 Yield strain, 32, 32
Two-hinged arch, 227, 227 Yield strength, 32, 32
Typical girder, 258 Yield stress, 32, 32
Young’s modulus (E), 29, 30
U
UBC, see Uniform Building Code Z
UDL, see Uniformly distributed load Zero-force members, 20, 118, 118–20
Ultimate compressive strength, 33