Statics and Mechanics of Structures

Steen Krenk ! Jan Høgsberg

Statics and
Mechanics of
Structures
Prof. Steen Krenk Prof. Jan Høgsberg
Department of Mechanical Engineering Department of Mechanical Engineering
Technical University of Denmark Technical University of Denmark
Kongens Lyngby, Denmark Kongens Lyngby, Denmark
sk@[Link] jhg@[Link]

ISBN 978-94-007-6112-4 ISBN 978-94-007-6113-1 (eBook)

DOI 10.1007/978-94-007-6113-1
Springer Dordrecht Heidelberg New York London

Library of Congress Control Number: 2013933869

© Springer Science+Business Media Dordrecht 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered
and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of
this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from Springer.
Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations
are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of pub-
lication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any
errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect
to the material contained herein.

Cover photo: Golden Terraces Shopping Mall, Warsaw. Designed by the Jerde Partnership, completed
2007. Photo by Steen Krenk

Printed on acid-free paper

Springer is part of Springer Science+Business Media ([Link])

 Preface

The theory of statics of structures has developed from intuition via grad-
ual refinement to its current state, where the basic principles are put into a
systematic framework that enables precise analysis. Although the basic laws
governing statics of structures have been known for several centuries, the
methods of analysis have developed considerably over the last decades. At
the current state of this development an introductory book on statics should
aim at the dual goal of providing sufficient background for developing an
intuitive understanding of structures, and at the same time lay a solid foun-
dation for modern analysis, typically made by computational techniques. In
this vein the present book makes extensive use of simple but realistic exam-
ples to develop familiarity and understanding of how structures carry and
distribute the loads through the structural members to the supports. This is
then supplemented by a few simple computer programs that illustrate, how
the theories for trusses and frames are implemented, and open up to a more
general approach to computational mechanics as a natural extension of the
present book.

The book is organized as follows. The first five chapters build up a basic
understanding of the statics of structures. It starts with force systems and
reactions in Chapter 1, then proceeding to the intuitively very accessible
theory of trusses, first analyzed by hand calculation procedures and then
reformulated as a small systematic finite element program MiniTruss in
Chapter 2. Chapter 3 develops the statics of beams and introduces the con-
cept of internal forces. The internal forces are then related to deformation
mechanisms of curvature, shear and extension in Chapter 4, and the princi-
ple of virtual work is developed in a concise form and used for calculation
of specific displacements. The introductory part is rounded off in Chapter 5
on the analysis of columns, describing instability as a bifurcation problem,
solved by eigenvalue analysis, and design principles based on the existence of
a characteristic imperfection. This part of the book covers material suitable
for an introductory one-semester course on basic statics of structures.

The remaining six chapters treat various extensions, that are typically in-
cluded in one form or another in a second semester course. The Chapters 6
and 7 deal with analysis of statically indeterminate frame structures. The
vi Preface

first of these chapters gives a systematic development of the force method

and describes how simple structures can conveniently be analyzed by hand.
The following chapter then develops the deformation method in which the
displacements of individual nodes play the key role. This then serves to in-
troduce the idea of the finite element formulation of frame structures. This
development is supported by the small program MiniFrame for internal
forces and displacements, and an extension MiniFrameS for linearized sta-
bility analysis. The Chapters 8 and 9 introduce three-dimensional states of
stress and strain, and present the theory of linear elasticity and some common
failure conditions. This material provides the background for the Chapters 10
and 11, in which the simple two-dimensional beam theory used in the previ-
ous chapters is extended to flexure and torsion of non-symmetric beams, and
the associated shear stress distributions.

The three small computer programs are coded in Matlab. The syntax and
input structure are described in connection with the corresponding theory in
the text, and the code is available from the authors via e-mail.

The authors are grateful for the permission to include photographs provided
by the following companies: Chapter 7, Rafsanjan Bridge, Waagner-Biro AG,
Vienna, Austria; Chapter 8, Test of wind turbine blade, LM Wind Power,
Kolding, Denmark; Chapter 10, Wind turbine, Siemens Wind Power, Brande,
Denmark.

Kgs. Lyngby Steen Krenk

September 2012 Jan Høgsberg
 Contents

1 Equilibrium and Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 The parallelogram rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Parallel forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 Moment from forces in a plane . . . . . . . . . . . . . . . . . . . . . 6
1.2.2 Moment from forces in space . . . . . . . . . . . . . . . . . . . . . . . 8
1.2.3 Force couples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.1 Virtual work of rigid bodies . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3.2 Equilibrium in a plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.3.3 Distributed load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4 Support conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.5 Reactions by equilibrium equations . . . . . . . . . . . . . . . . . . . . . . . 19
1.5.1 Plane beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.5.2 Simple frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.5.3 Three-hinge frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.5.4 Space structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
1.6 Reactions by virtual work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

2 Truss Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.1 Basic principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.1.1 Building with triangles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.1.2 Counting joints and bars . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.1.3 Qualitative tension-compression considerations . . . . . . . 45
2.2 Method of joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.2.1 Planar truss structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.2.2 Space trusses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.3 Method of sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.3.1 Bar forces via the method of sections . . . . . . . . . . . . . . . 55
2.3.2 Special types of planar trusses . . . . . . . . . . . . . . . . . . . . . 57
2.4 Stiffness and deformation of truss structures . . . . . . . . . . . . . . . 64
2.4.1 Axial stress and strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
viii Contents

2.4.2 Linear elastic bars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

2.4.3 Virtual work for truss structures . . . . . . . . . . . . . . . . . . . 67
2.4.4 Displacements of elastic truss structures . . . . . . . . . . . . . 71
2.5 Finite element analysis of trusses . . . . . . . . . . . . . . . . . . . . . . . . . 73
2.5.1 Elastic bar element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2.5.2 Finite Element Method for trusses . . . . . . . . . . . . . . . . . . 77
2.5.3 The MiniTruss program . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

3 Statics of Beams and Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

3.1 Internal forces and moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
3.2 Beams with concentrated loads . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.2.1 Variation of internal forces for concentrated loads . . . . 99
3.3 Beams with distributed load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
3.3.1 Differential equations for internal forces . . . . . . . . . . . . . 107
3.3.2 Maximum moment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
3.4 Combined loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
3.4.1 Superposition of load cases . . . . . . . . . . . . . . . . . . . . . . . . 116
3.4.2 Superimposing the distributed load . . . . . . . . . . . . . . . . . 117
3.5 Internal forces in frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
3.5.1 Influence of load distribution . . . . . . . . . . . . . . . . . . . . . . . 124
3.5.2 Influence of support conditions . . . . . . . . . . . . . . . . . . . . . 128
3.5.3 Three-hinge frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
3.5.4 Principle of the arch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
3.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

4 Deformation of Beams and Frames . . . . . . . . . . . . . . . . . . . . . . . 143

4.1 Bending of elastic beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
4.1.1 Homogeneous bending . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
4.1.2 Linear kinematic relations . . . . . . . . . . . . . . . . . . . . . . . . . 148
4.2 Bernoulli beam theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
4.2.1 Statically determinate beams . . . . . . . . . . . . . . . . . . . . . . 154
4.2.2 Statically indeterminate beams . . . . . . . . . . . . . . . . . . . . . 159
4.3 Shear flexible beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
4.4 Virtual work and displacements of beams . . . . . . . . . . . . . . . . . . 168
4.4.1 Principle of virtual work . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.4.2 Displacements in elastic beams . . . . . . . . . . . . . . . . . . . . . 171
4.4.3 Virtual work and displacements in frames . . . . . . . . . . . 179
4.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

5 Column Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

5.1 Beam with normal force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
5.1.1 Stiffness reduction from normal force . . . . . . . . . . . . . . . 193
5.2 Stability of the ideal column . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
Contents ix

5.2.1 Equivalent column length . . . . . . . . . . . . . . . . . . . . . . . . . 203

5.2.2 Buckling direction and intermediate supports . . . . . . . . 205
5.3 Design of columns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
5.3.1 Column length and slenderness . . . . . . . . . . . . . . . . . . . . . 208
5.3.2 Geometric imperfections . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
5.3.3 Stresses in column cross-sections . . . . . . . . . . . . . . . . . . . 215
5.3.4 Perry-Robertson’s column design criterion . . . . . . . . . . . 218
5.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222

6 The Force Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

6.1 Principle of the force method . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
6.2 The general force method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
6.2.1 Released structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
6.2.2 The basic steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
6.2.3 Summary of the force method . . . . . . . . . . . . . . . . . . . . . . 241
6.3 Application of the Force Method . . . . . . . . . . . . . . . . . . . . . . . . . 242
6.4 The force method for frame structures . . . . . . . . . . . . . . . . . . . . 250
6.4.1 Simply supported frames . . . . . . . . . . . . . . . . . . . . . . . . . . 251
6.4.2 Frames with fixed supports . . . . . . . . . . . . . . . . . . . . . . . . 257
6.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

7 Deformation and Element Methods for Frames . . . . . . . . . . . 267

7.1 Stiffness of beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
7.1.1 Symmetric and anti-symmetric bending . . . . . . . . . . . . . 269
7.1.2 Basic cases of imposed deformation . . . . . . . . . . . . . . . . . 271
7.1.3 Loads on constrained beams . . . . . . . . . . . . . . . . . . . . . . . 277
7.2 Deformation method for frames . . . . . . . . . . . . . . . . . . . . . . . . . . 278
7.3 Beam elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
7.3.1 Beam bending element . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
7.3.2 Beam-column element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
7.3.3 Transformation to global form . . . . . . . . . . . . . . . . . . . . . 305
7.4 Finite element method for frames . . . . . . . . . . . . . . . . . . . . . . . . . 307
7.4.1 The MiniFrame program . . . . . . . . . . . . . . . . . . . . . . . . . . 308
7.4.2 Stability analysis of frames . . . . . . . . . . . . . . . . . . . . . . . . 312
7.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

8 Stresses and Strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321

8.1 Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
8.1.1 The stress vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
8.1.2 General stress components . . . . . . . . . . . . . . . . . . . . . . . . . 324
8.1.3 Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
8.2 Deformation and strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
8.2.1 Strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
8.2.2 Rotation at a point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
x Contents

8.2.3 Displacement decomposition . . . . . . . . . . . . . . . . . . . . . . . 337

8.3 Virtual work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
8.3.1 Equation of virtual work . . . . . . . . . . . . . . . . . . . . . . . . . . 338
8.3.2 Matrix and tensor notation . . . . . . . . . . . . . . . . . . . . . . . . 341
8.4 Special states of stress and strain . . . . . . . . . . . . . . . . . . . . . . . . . 342
8.4.1 Plane stress and plane strain . . . . . . . . . . . . . . . . . . . . . . . 342
8.4.2 Stress and strain transformations . . . . . . . . . . . . . . . . . . . 343
8.4.3 Principal stresses and strains in a plane . . . . . . . . . . . . . 349
8.4.4 Principal stresses in three dimensions . . . . . . . . . . . . . . . 354
8.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358

9 Material Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363

9.1 Elastic materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
9.1.1 Internal elastic energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
9.1.2 Linear isotropic elasticity . . . . . . . . . . . . . . . . . . . . . . . . . . 367
9.2 Mean and deviator components . . . . . . . . . . . . . . . . . . . . . . . . . . 376
9.3 Yield conditions for metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
9.3.1 Von Mises’ yield condition . . . . . . . . . . . . . . . . . . . . . . . . . 380
9.3.2 Tresca’s yield condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
9.4 Coulomb’s theory of friction materials . . . . . . . . . . . . . . . . . . . . . 385
9.4.1 Critical section and stress state . . . . . . . . . . . . . . . . . . . . 386
9.4.2 Coulomb failure surface . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
9.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391

10 General Bending of Beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395

10.1 Bending of non-symmetric beams . . . . . . . . . . . . . . . . . . . . . . . . . 397
10.1.1 Kinematic formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
10.1.2 Stresses and section forces . . . . . . . . . . . . . . . . . . . . . . . . . 400
10.2 Cross-section analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
10.2.1 Elastic center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
10.2.2 Moments of inertia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
10.2.3 Principal coordinate system . . . . . . . . . . . . . . . . . . . . . . . . 417
10.3 Axial stresses and strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425
10.3.1 Neutral axis and line of curvature . . . . . . . . . . . . . . . . . . 434
10.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440

11 Flexure and Torsion of Beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

11.1 Shear stresses in beam flexure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444
11.1.1 Shear flow – Grashof’s formula . . . . . . . . . . . . . . . . . . . . . 445
11.1.2 Shear stress on cross-section . . . . . . . . . . . . . . . . . . . . . . . 449
11.2 Thin-walled cross-sections in shear . . . . . . . . . . . . . . . . . . . . . . . . 455
11.2.1 Shear center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462
11.2.2 Shear flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465
11.3 Torsion of circular cylinders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
Contents xi

11.4 General homogeneous torsion of beams . . . . . . . . . . . . . . . . . . . . 472

11.4.1 The Prandtl stress function . . . . . . . . . . . . . . . . . . . . . . . . 475
11.5 Torsion of thin-walled beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
11.5.1 Open sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
11.5.2 Single-cell sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484
11.5.3 Multi-cell sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490
11.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501
 Equilibrium and Reactions 1

Statics of structures deals with structures that are exposed to loads and
develop reactions and internal forces that leave the structure stationary. The
present book deals with buildings and civil engineering structures that are
supported to prevent motion, as opposed to space structures, trains etc. where
motion is an integral part of the behavior. A fundamental tool of statics is the
concept of equilibrium. In order to remain stationary the total effect of the
loads and the reactions provided by the supports must be in equilibrium. This
applies to the full structure and also to its different parts. In this chapter the
equilibrium conditions for the full structure are used to identify requirements
for the supports and to determine the reactions provided by the supports.
The concept of equilibrium is developed further in the following chapters to
deal with hypothetical parts of the structure, and thereby obtain knowledge
of the distribution of the forces inside the structure.

First the notion of a force is introduced in Section 1.1. A force is specified by

its magnitude and its line of action, and is closely related to the mathemat-
ical concept of a vector. If two forces have intersecting lines of action they
combine as vectors, and act at the point of intersection. However, this con-
cept of intersecting forces is too limited, and it is necessary to introduce the
notion of a moment, as described in Section 1.2. When considering forces and
moments together, the concept of equilibrium takes a precise mathematical
form, discussed in Section 1.3. The direct form of the equilibrium conditions
constitutes two vector equations for the total force and the total moment,
respectively. It is explained in detail, how these equations can be combined

S. Krenk, J. Høgsberg, Statics and Mechanics of Structures,

DOI 10.1007/978-94-007-6113-1 1,
© Springer Science+Business Media Dordrecht 2013
2 Equilibrium and Reactions

into a single scalar equation of virtual work – that is the work that the forces
and moments would perform, if subjected to an arbitrary small virtual dis-
placement. The virtual work is here introduced in its basic form, but appears
in a more advanced form later in connection with deformation of beams and
frames. The concept of virtual work plays a central role in the modern formu-
lation of theories for structures and solid bodies, e.g. in connection with the
formulation of numerical methods. The two last sections of the chapter deal
with the support conditions and the reactions developed in the supports.

1.1 Forces
The notion of a force is fundamental to the theory of structures. A force
is associated with a magnitude, a direction, and a point of action. In the
analysis of forces it is convenient first to focus on the direction and magnitude,
combined in the boldface vector symbol P. The magnitude is represented by
the length of the vector and is denoted by P = |P|. In practice a force often
has a specific point of action, but it is often convenient to consider the force
as acting in a line of action, defined as the line obtained by extending the
force vector in space. This notion permits the force to be translated along
its line of action, and leads to a fairly intuitive formulation of the theory of
equilibrium of a set of one or more forces.

1.1.1 The parallelogram rule

It is an important property of a force P that it can be resolved into com-

ponents according to the parallelogram rule, known from elementary vec-
tor analysis, see e.g. Strang (2001). The parallelogram rule is illustrated in
Fig. 1.1 showing the force P and two directions intersecting the line of action
of P. For convenience and clarity the point of intersection is shown as the
point of action of the force P in the figure. If this is not the case for the initial
location of P, it is translated to the point of intersection along the line of
action.

Fig. 1.1: Decomposition of force P in given direction.

The parallelogram rule for resolving a force P into components P1 and P2

with given directions consists in forming a parallelogram with P along a
Forces 3

diagonal and the components P1 and P2 along adjoining sides as shown in

Fig. 1.1b.

The parallelogram rule can also be used to form the resulting force from two
given forces P1 and P2 , when these forces have intersecting lines of action.
The construction of the parallelogram follows from sliding the forces P1 and
P2 to the point of intersection. They then form the sides of a parallelogram
with the resultant P along the diagonal as shown in Fig. 1.1b.

Example 1.1. Force on a string. Figure 1.2 shows a simple example of the force paral-
lelogram rule. A vertical force is acting on a string, which is stretched, forming two linear
parts. These parts carry constant forces T1 and T2 , in the direction of the respective
strings. !

Fig. 1.2: Cable carries P via the forces T1 and T2 .

In a static analysis of simple structures the forces may be referred directly to

directions in the structure, e.g. as along or transverse to a beam. However, in
larger analyzes, and when using a computer for the numerical computations,
it is often convenient to represent forces by their components in a Cartesian
coordinate system. In this case two forces P1 and P2 are represented by their
xyz-components as
   
Px1 Px2
 1  2
P1 = Py  , P2 = Py  , (1.1)
Pz1 Pz2

and it then follows from the parallelogram composition rule that the resultant
force P has the components
     
Px Px1 Px2
   1  2
Py  = Py  + Py  . (1.2)
Pz Pz1 Pz2

It should be noted that this standard addition rule of vector components

must be accompanied by an account of the resultant’s line of action.
4 Equilibrium and Reactions

1.1.2 Parallel forces

In the case of two parallel forces their lines of action do not intersect, and
thus the parallelogram rule needs an extension. The problem is illustrated in
Fig. 1.3 showing two parallel forces with distance a and magnitude P1 and
P2 , respectively. In principle the magnitude and location of the resulting force
P can be obtained as a limit of the two forces, if inclined slightly with their
original common direction. However, it is more direct to obtain the result by
introducing two auxiliary forces as demonstrated here.

Fig. 1.3: Parallel forces with distance a.

In order to increase the clarity of the geometric construction the two forces
P1 and P2 are first translated along their respective lines of action, until
their points of application A1 and A2 lie on a line orthogonal to the lines of
action as shown in Fig. 1.4a. Two forces of equal magnitude Q but opposite
direction along the connecting line are now added as shown in the figure. As
these forces are opposite with the same line of action they have the sum zero,
and therefore do not change the resulting force of the system. There is now
a force P1 + Q acting at A1 and a force P2 − Q acting at A2 . These forces
are not parallel, and they can therefore be combined by the parallelogram
rule, whereby the resulting force P = P1 + P2 passes through the point of
intersection C of the lines of action.

Fig. 1.4: Composition of parallel forces.

The line of action of the resulting force P is characterized by the distance a1

from P1 and the distance a2 from P2 . The figure contains similar triangles
Moments 5

formed by the force components and geometric distances. Thus, the force
triangle with sides P1 and Q is similar to the geometric triangle with sides h
and a1 . An equivalent relation holds for P2 and a2 . From this the following
two relations are obtained:
P1 h P2 h
= , = . (1.3)
Q a1 Q a2
Elimination of the product hQ between these equations then gives the relation

a1 P1 = a2 P2 . (1.4)

This is the lever rule, used since Antiquity for scales where two weights are
placed on a lever at different distance from a point of fixture.

Fig. 1.5: Composition or resolution of parallel forces.

The magnitude of the resulting force is illustrated in Fig. 1.4b, showing the
resolution of the forces P2 and −Q on top of the resolution of the forces P1
and Q. The double occurrence of ±Q implies that the resulting force retains
the original direction and has the magnitude

P = P1 + P2 . (1.5)

Thus, the vector components can also be found by direct summation as al-
ready indicated by the component summation formula (1.2).

1.2 Moments
The geometric construction of the line of action of the sum of two parallel
forces demonstrates that the two forces are translated in such a way that
the product aj Pj is equal for the two forces. This is a special instance of a
moment. Moments play a central role in the mechanics of structures.

The basic form of a moment generated by a force is illustrated in Fig. 1.6.

Let C be a point and P a force in space. The moment of the force P about
the point C is defined by its magnitude and its direction, and can therefore
be represented by a vector. The magnitude is hP , where h is the distance of
6 Equilibrium and Reactions

Fig. 1.6: Moment of force P about the point C.

C from the line of action of the force. The direction is orthogonal to the plane
defined by the line of action and the point C. The concept of a moment is
considered in the following subsections: first in the plane containing the line
of action and the point C, then by extending the concept to three-dimensional
vector form, and finally the relation between force couples and moments.

1.2.1 Moment from forces in a plane

Many problems within structural analysis can be resolved into one or more
planar problems. It is therefore convenient first to consider the moment of a
force P about a point C as a planar problem. The problem is illustrated in
Fig. 1.7a, showing the force P located in the xy-plane of a Cartesian coor-
dinate system with origin C. The components of the force in this coordinate
system are [Px , Py ]. The moment of the force P about the point C can be
calculated as the sum of the moment of each of the force components Px
and Py .

Fig. 1.7: Moment calculation via components.

The procedure is illustrated in Fig. 1.7, where the point A with components
[ax , ay ] indicates the point of application of the force. The moment MC is
considered positive when acting in a counter clockwise direction about C.
Thus, the force component Px gives the moment −ay Px , while the force
component Py gives the moment ax Py . As a result the moment of the force
P about the point C is
M = ax Py − ay Px . (1.6)
Moments 7

This procedure, where the moment is calculated from suitable components

instead of using lengths and angles of the corresponding vectors, is often the
most convenient, as illustrated extensively throughout the book.

The moment was here defined in terms of the components of a vector [ax , ay ]
from the origin C to the point A of application of the force, here taken as the
origin of the force vector. It is easily demonstrated that the moment remains
unaffected if the point A is replaced by any point A! on the line of action of
the force. Any such point may be represented in terms of the original point
A by a parameter representation of the form
' ! ( ' ( ' (
ax ax Px
= + α , (1.7)
a!y ay Py

where α is a scalar parameter defining the location of the point A! . When

substituting these vector components into the moment definition (1.6) it is
seen that the result is independent of the parameter α. Thus, any point on
the line of action leads to the same moment.

Example 1.2. Moment of a force. This example illustrates two alternative methods of
computing the moment of a force in a plane: a) straightforward calculation using given
components, and b) translating the force along its line of action before computing the
moment.

Fig. 1.8: Moment by horizontal and vertical force components.

Figure 1.8 shows a force P acting at the point A located in the xy-plane with components

[ Px , Py ] = [ 4, 3 ], [ ax , ay ] = [ 6, 3 ],

with dimension of force and length, respectively. Straightforward application of (1.6) gives
the moment about the origin C as

M = ax Py − ay Px = 6 · 3 − 3 · 4 = 6.

It is observed that if the force was translated along its line of action to the intersection
with one of the coordinate axes, then the resulting expression would contain only one
product. In the present case the force can be translated to the point B on the x-axis with
coordinates [ bx , by ] = [ 2, 0 ]. The moment would then be obtained as

M = bx Py = 2 · 3 = 6,
8 Equilibrium and Reactions

where it has been used that by = 0. In the analysis of e.g. truss structures it is quite
common to simplify the calculation of moments by sliding the forces along their line of
action, as illustrated in Chapter 2. !

In the form presented here the moment M appears as a scalar, i.e. a quantity
associated with a numerical value. However, in a more general context the
moment M calculated here is the moment about an axis normal to the plane
through C. Thus, the moment is actually the z-component Mz of a moment
vector, acting as a moment about the z-axis of the xyz-coordinate system.
While this may be of less importance in planar problems, it is important to
recognize the present problem as a special case applicable to forces located
in the xy-plane. The next subsection gives the extension to the fully three-
dimensional case.

1.2.2 Moment from forces in space

The three-dimensional problem of the moment of a force P about a point C is

illustrated in Fig. 1.9. The force vector P and the point C define a plane, and
the moment is defined by a vector M through C orthogonal to that plane,
with magnitude hP . This coincides with the definition of the vector product
between a vector a from the point C to any point on the line of action of the
force and the force vector P,

M = a × P. (1.8)

This relation defines the moment M about C as a vector.

Fig. 1.9: Moment vector M from force P.

The components of the moment vector M defined in (1.8) are given by the
determinant relation    
Mx ∗ ∗ ∗
My  =  ax ay az  . (1.9)
Mz Px Py Pz
The individual moment components are the sub-determinants defined by this
relation,
Moments 9

   
Mx ay Pz − az Py
My  = az Px − ax Pz  . (1.10)
Mz ax Py − ay Px
It is seen that the moment component Mz corresponds to the planar case
considered in (1.6).

1.2.3 Force couples

The construction from Section 1.1.2 for addition of parallel forces gives finite
distances, if the resultant force has a magnitude different from zero. This
leaves a special case, where one force is P while the other is −P with a parallel
line of action. This is called a force couple with two forces of equal magnitude
but opposite direction acting along parallel lines as illustrated in Fig. 1.10.
Let C be a point in the plane, and let a1 and a2 be vectors connecting C to a
point on the line of action of the force P and −P, respectively. This is shown
in Fig. 1.10a. Applying the vector product formula (1.8), the total moment
of the two forces about the point C then is

M = a1 × P + a2 × (−P) = (a1 − a2 ) × P = a × P , (1.11)

where a = a1 − a2 is a vector connecting the two lines of action. Thus,

the moment M of the force couple is independent of the initial reference
point C. It follows from (1.10) that the magnitude of the moment component
orthogonal to the plane of the forces is M = hP , where h is the distance
between the lines of action and P is the length of the force vectors, see
Fig. 1.10b.

Fig. 1.10: Force couple with P and −P along parallel lines.

The force couple plays a central role, when a force is translated to a new line
of action. This problem is illustrated in Fig. 1.11a, showing the force P in its
original position in blue. This force is translated into a new line of action by
placing two new forces P and −P in the new line of action. As they act in
the same line and are of equal but opposite magnitude they are equivalent
to zero force. The original force P and the new force −P constitute a force
couple with moment M = hP , and thus the total effect of the three forces is
a translated force P and the moment M illustrated in Fig. 1.11b.
10 Equilibrium and Reactions

Fig. 1.11: Offset of force by introduction of moment.

1.3 Equilibrium
Statics is the theory of equilibrium of force systems and the use of this theory
to study the equilibrium of solids and fluids. In particular statics of structures
deals with the equilibrium of structures under time-independent loads.

Fig. 1.12: Statics and kinematics of a rigid body.

A basic concept of statics is the notion of equilibrium. Figure 1.12a illustrates

a rigid body acted upon by forces P1 , P2 , · · · at points located relative to a
common point C by the vectors a1 , a2 , · · · and by moments M1 , M2 , · · · . In
order for the system of forces and moments, and thereby the rigid body, to
be in equilibrium the resulting total force and moment vectors must vanish.
The resulting force is found by direct summation of the individual forces
as vectors. The resulting moment consists of the vector sum of the applied
moments plus the sum of the moments of the forces about a point, e.g. the
point C. Thus, equilibrium of a rigid body is expressed by the two vector
equations
) ) )
P = Pi = 0 , MC = Mj + ai × Pi = 0 . (1.12)
i j i

The general equilibrium conditions, stated in vector form, are often used for
numerical computations, where generality is important. On the other hand
most calculations by hand are made by isolating one or more of the com-
ponents and treating the equilibrium conditions sequentially. Efficient proce-
dures for handling this problem is the main issue in the remaining examples
and the exercises in this chapter.
Equilibrium 11

1.3.1 Virtual work of rigid bodies

The equations of equilibrium are closely related to the conditions needed to

prevent motion of the body. In essence the three force components must van-
ish in order not to produce accelerated motion in any of the directions of
three-dimensional space. Similarly, the three moment conditions prevent the
occurrence of rotation with angular acceleration about any of three indepen-
dent directions in space. The theory of motion of bodies described in terms of
translations and rotations is termed kinematics, while equilibrium described
in terms of forces and moments is within the field of statics. There is an in-
teresting and important relation between the kinematic quantities describing
a possible motion of a body and the forces and moments acting on this body.
This relation is called the equation of virtual work. It is called virtual work,
because it deals with a hypothetical situation in which a body is acted upon
by actual forces Pi and moments Mj , and then a small hypothetical motion
in terms of displacements δu and rotations δθ is introduced. The symbol δ is
used on the displacement parameters to indicate that these are infinitesimal
quantities, similar in nature to the notation ‘d’ used for differentials, e.g. as
dx and dy.

A virtual displacement of a rigid body can be described by the virtual trans-

lation δuC of a point C and the virtual rotation δθ about this point as
illustrated in Fig. 1.12b. The virtual rotation is infinitesimal, and the result-
ing displacement at a point described by the position ai relative to the point
C can then be described as

δui = δuC + δθ × ai . (1.13)

The idea is now to consider the work that would result if the forces Pi , located
at ai , were translated via the virtual displacements δui , and the moments Mj
were rotated by the common virtual rotation δθ. Because the displacements
are virtual this work is called the virtual work and denoted δV . The virtual
work of the forces Pi and the moments Mj through the virtual displacements
δui and the common virtual rotation δθ is
) )
δV = δui · Pi + δθ · Mj , (1.14)
i j

where a dot denotes the scalar product of two vectors. It follows from the
representation (1.13) that an infinitesimal virtual rigid body motion can be
described in terms of the virtual translation δuC of the point C and the
virtual rotation δθ about this point. When the representation (1.13) is sub-
stituted, the expression for the virtual work becomes
) ) )
δV = δuC · Pi + (δθ × ai ) · Pi + δθ · Mj . (1.15)
i i j
12 Equilibrium and Reactions

The middle term is a summation of so-called triple vector products. A vector

triple product consists of a vector product of two vectors, multiplied via a
scalar product with a third vector – here in the form of the vector product δθ×
ai multiplied by Pi . The triple product has a simple geometric interpretation
as the volume of the parallelepiped spanned by the three vectors. This volume
does not depend on which of the two vectors are used to form the vector
product, and thus the factors can be interchanged according to the relation

(δθ × ai ) · Pi = δθ · (ai × Pi ) . (1.16)

When this expression is substituted into (1.15) the virtual work takes the
form
*) + *)* ) +
δV = δuC · Pi + δθ · ai × Pi ) + Mj . (1.17)
i i j

The terms in the large parentheses have already been identified in (1.12) as
the resulting force P and the total moment MC about the point C. Further-
more it was found that equilibrium corresponds to the conditions P = 0 and
MC = 0 for an arbitrary point C. With these observations the virtual work
of a rigid body in equilibrium can be expressed as

δV = δuC · P + δθ · MC = 0 . (1.18)

The first equality defines the virtual work as the work of the resulting force
P through the virtual displacement δuC at some reference point C plus the
work of the resulting moment MC about C through the virtual rotation δθ
of the body.

Fig. 1.13: Virtual translation δuC and rotation δθ of a rigid body.

The second equality states that the virtual work, formed through six virtual
displacement components, must vanish. This holds for any choice of the vir-
tual displacement components, and thus the six equilibrium equations are
equivalent to the statement that the virtual work must vanish for any choice
of the virtual displacement components. This is called the principle of virtual
work. It is a fundamental tool in statics and mechanics of structures. The
principle of virtual work establishes a close relationship between the statics
Equilibrium 13

and kinematics of a structure. Thus, the force equilibrium equation P = 0

can be stated as: the work done by the forces by any virtual rigid body trans-
lation δuC must vanish. This is illustrated in Fig. 1.13a. Similarly, moment
equilibrium about a point C originally expressed as MC = 0 can be restated
in the form: the work done by forces and moments via any rotation about C
must vanish. This relation is illustrated in Fig. 1.13b.

The principle of virtual work is important in understanding and analyzing

structures. It will appear in the discussion of support conditions in Section 1.4
and calculation of reactions in Section 1.6. It is extended to deformable bodies
in the following chapters.

1.3.2 Equilibrium in a plane

Many structures can be decomposed into planar parts, i.e. parts that can be
described in a plane and are loaded in this plane. For these structures only
a reduced set of equilibrium equations are needed. The basic form of these
conditions consists of the projection of the force equilibrium condition into
the plane, supplemented by the component of the moment equation along
a normal to the plane. This corresponds to a special case of the principle
of virtual work including only motion in the plane, e.g. in the form of two
translation components and one rotation component.

Fig. 1.14: In-plane equilibrium by a) 2 force components and moment about 1 point, b) 1
force component and moment about 2 points, c) moment about 3 points.

It is often useful to modify the initial formulation of the equilibrium equations

in terms of two projections and a moment equation by replacing one or both
projection equations with moment equations. This results in the following
three possibilities, illustrated in Fig. 1.14,
a) Two force projection equations and one moment equation, e.g. Px = 0,
Py = 0 and MA = 0.
b) One force projection equation and moment equations for two points, e.g.
Px = 0, MA = 0 and MB = 0. The line connecting A and B must not
be orthogonal to the line used in the force projection.
c) Moment equations for three points, MA = 0, MB = 0 and MC = 0. The
points A, B and C must not lie on a common line.
14 Equilibrium and Reactions

The examples and exercises of this chapter illustrate the use of each of these
sets of equilibrium conditions for calculation of reactions: a) for a clamped
structure, b) for a structure with parallel reactions, and c) for structures with
non-parallel reactions.

1.3.3 Distributed load

Structures may be exposed to loads that are distributed over the length,
surface or volume of the structure. These distributed loads typically arise
from e.g. the weight of the structure itself, load from wind or fluid pressure,
or earth pressure on foundations. Distributed loads are characterized by their
intensity p, denoting the force per unit volume, surface area or per unit length.
In plane problems loads are distributed along e.g. the length of a beam, and
will have the unit N/m, Newton per meter.

Fig. 1.15: Distributed load p(x) and equivalent load P .

In the analysis of structures with distributed loads it is often convenient to

represent the load by an equivalent load in the form of one or more concen-
trated forces, or a concentrated force and a moment. In order to be equiva-
lent with the original distributed load the concentrated forces (and moments)
must have the same resulting force and resulting moment. The situation is
illustrated in Fig. 1.15, showing a load distributed along a line with intensity
p. The line is taken as x-axis with origin C, and the load extends from a to b.
The load is to be represented by an equivalent force P , acting at the distance
xc from C. The magnitude of the distributed load and its moment about C
are determined by the integrals
, b , b
P = p(x) dx , MC = p(x) x dx . (1.19)
a a

The force P is located at the distance xC from C that produces the same
moment MC as the distributed load, and thus
MC
xc P = MC ⇒ xc = . (1.20)
P
The calculation is seen to be similar to that of calculating the location of the
‘center of gravity’ of the load distribution.
Equilibrium 15

Fig. 1.16: Distributed load and equivalent load.

The two most common load distributions are the constant intensity and a lin-
ear variation of the intensity, illustrated in Fig. 1.16. For the uniform intensity
load distributed over the interval [0, a] the force and the moment about the
left end point follow from the integrals (1.19) as
, a , a
P = p dx = ap , M = p x dx = 12 a2 p .
0 0

This gives the distance xc of the force from the left end as
1 2
M 2a p 1
xc = = = 2a .
P ap
The result is illustrated in the left half of Fig. 1.16.

The linear load intensity variation can be represented via triangular distri-
butions as illustrated in the right side of Fig. 1.16. With origin C located
at the left end, the load distribution is px/a, where p is used to denote the
maximum intensity as shown in the figure. In this case the force and moment
integrals are
, a , a
P = px/a dx = 12 ap , M = (px/a) x dx = 13 a2 p .
0 0

Thus, in this case the distance xc of the force from the left end is
1 2
M 3a p 2
xc = = 1 = 3a .
P 2 ap

This result is illustrated in the right half of Fig. 1.16.

A common load distribution is a linear variation between p1 at the left end

and p2 at the right end, illustrated in Fig. 1.17. In the representation of the
distributed load it is convenient to associate the load intensities p1 and p2
with each their triangular distributions. Each of these distributions are then
associated with a concentrated force at distance 13 a from the corresponding
16 Equilibrium and Reactions

Fig. 1.17: Distributed and equivalent load.

end of the load distribution. The equivalent load then consists of the two
concentrated forces 12 ap1 and 12 ap2 , located as shown in the figure.

1.4 Support conditions

Structures are typically connected to the surroundings via local supports. A
simple example is shown in Fig. 1.18a, where the rotor of a wind turbine is
exposed to a distributed horizontal load from the wind. The wind turbine
is placed on a foundation that is sufficiently strong and large to secure the
wind turbine against motion. As discussed in the previous section equilib-
rium is associated with the balance of forces and moments. Thus, in plane
problems the reaction forces transmitted to the foundation, and from there
to the soil, can be represented by two orthogonal force components, typically
a horizontal and a vertical force, and a single moment. The wind turbine
must be secured against vertical motion due to its own weight, against hor-
izontal motion due to the wind loading P and against rotation in A due to
the overturning moment P h. This requires a vertical reaction force RA , and
!
horizontal reaction force RA and a reaction moment MA , which are shown in
the so-called free body diagram in Fig. 1.18b, where the foundation and the
rotor have been replaced by the representative loading and reaction forces
acting on the main structure, which in this case is the wind turbine tower.

Fig. 1.18: Reaction forces and reaction moment on wind turbine structure.
Support conditions 17

In the case of the wind turbine all three reactions are required. If for instance
the horizontal reaction is removed the structure is free to move in the hor-
izontal direction, thereby violating the concept of equilibrium. The number
of support conditions matches the number of reactions necessary for general
equilibrium, and the wind turbine structure is therefore statically determi-
nate. As demonstrated in the following this means that the unknown reactions
can be determined directly from the available equilibrium equations.

Figure 1.19a shows a frame structure placed on two small supporting foun-
dations in A and B. The load on the frame is transmitted to the surrounding
soil by the foundation, which is assumed sufficiently large to actually do this
for the reaction forces needed in the present problem. The frame structure is
supported at two separated points A and B. This separation secures the frame
structure against rotation and overturning. Thus, for the frame structure re-
action moments in A and B are not required, whereby the local foundations
are fairly small. The lack of significant local moment capacity implies that a
conservative supporting system only assumes vertical and horizontal reaction
forces, i.e. no reaction moments, as indicated by the free body diagram in
Fig. 1.19b. This leads to a frame structure with four reactions. As discussed in
connection with Fig. 1.14 equilibrium in the plane requires three equilibrium
conditions composed of forces and moments. In Fig. 1.19b it is shown that
the frame structure has four reactions, which is one more than the number of
available equilibrium equations in plane problems. Thus, the frame structure,
with the present support conditions, is statically indeterminate. This is also il-
lustrated by the fact that one reaction force can be removed without allowing
any rigid body motion or mechanisms. For instance, the horizontal reaction
! !
RA can be removed, and the remaining horizontal reaction RB will still se-
cure the structure against horizontal motion. The analytical and numerical
analysis of statically indeterminate structures is presented in Chapters 6–7.

Fig. 1.19: Plane frame: a) Structure with supports, b) Free body diagram with reactions.

In the analysis of structures a number of specific symbols are commonly used

to indicate the various support conditions. For basic static analysis the most
important symbols are presented in Fig. 1.20, illustrated for a beam type
structure. In Fig. 1.20a the circle indicates a hinge that permits free rotation.
18 Equilibrium and Reactions

Thus, a hinge has no moment capacity. This is also called a simple support,
and the reactions are the vertical reaction force R and the horizontal reaction
force R! . If supports are allowed to translate in a particular direction, this is
indicated by rollers. For the simple support this is shown in Figs. 1.20b,c for
unconstrained motion in the horizontal and vertical direction, respectively.
Note, that in these situations the support imposes no reaction force in the
specific direction of the roller, leaving only a reaction force in the orthogonal
direction.

Fig. 1.20: Symbols for various support conditions.

If the support is assumed to be rigidly fixed, it constrains all displacement and

rotation components. The symbol for the fixed support is shown in Fig. 1.20d,
where all three reaction components are in general non-vanishing. Thus, the
wind turbine in Fig. 1.18 has a fixed support. If the moment capacity of a
fixed support is removed, the simple support in Fig. 1.20a is recovered. If the
fixed support can translate this is indicated by rollers showing the direction
of translation as shown in Figs. 1.20e,f. The reaction force is normal to the
direction of translation. Most support conditions can be described by the
combination of a fixed support, a hinge and/or rollers. In the case of flexible
supports these may be represented via spring connections.

Fig. 1.21: Intermediate and internal support conditions.

Structures often have intermediate supports, such as pylons carrying cables

and bridge decks, or columns carrying roof structures. These intermediate
support conditions can often also be represented by the symbols presented
Reactions by equilibrium equations 19

above. Figure 1.21a shows a simple support with horizontal rollers supporting
a beam structure. In this case the only non-vanishing reaction force is the
vertical reaction R. It is important to note that the beam in Fig. 1.21a is rest-
ing on top of the intermediate simple support, whereby the internal moment
capacity of the beam is not changed by the support. Figure 1.21b shows a
hinge in the beam, whereby the structure is free to rotate at the hinge. This
implies that no internal moment can be transmitted through the hinge, and
thus M = 0 in the beam at the location of the hinge. This particular type
of internal support is common in structural engineering, because structural
parts are often connected by hinge type connections with negligible moment
capacity. The notion of internal section forces is considered in more detail in
Chapter 2 for trusses and in Chapter 3 for beams and frames.

1.5 Reactions by equilibrium equations

If the reactions of a structure can be determined by equilibrium conditions
alone, the structure is classified as statically determinate with respect to
reactions. A necessary condition for static determinacy of a rigidly connected
structure then is that the number of reaction components corresponds to the
number of equilibrium conditions. Thus, a rigid structure in space will have
six reaction components, while a rigid plane structure will have three reaction
components – relating to the plane.

Fig. 1.22: Inclined beam with concentrated load.

It is often convenient to organize the description of the structure and the

following analysis of the reactions in a systematic way. The procedure is
illustrated by the simple inclined beam shown in Fig. 1.22, and supplemented
by a number of examples on the reactions of beams and frames in the later
part of this section.

The first step in the analysis of the beam in Fig. 1.22 is to make a descrip-
tion of the structure, including support conditions and loading. This step is
illustrated in Fig. 1.23. It consists of making a representative sketch of the
static system including reactions and loads as follows:
20 Equilibrium and Reactions

Fig. 1.23: Description of structure with system lines and supports.

a) Make a sketch of the static system with internal joints (as appropriate)
and supports. The structure is represented by its system lines, and suf-
ficient dimensions to define the geometry of the structure are included.
b) Indicate all possible reactions. It is important that all reactions, that can
be generated by the supports are included to ensure that the structure
has sufficient support to prevent motion. Compare the number of reaction
components with the available number of equilibrium conditions. In the
present case the reactions consist of the three components RA , RB and
!
RA shown in Fig. 1.23.
c) Introduce the loads. This is the last point, as many structures are an-
alyzed for several load cases, and the previous points are unaffected by
the specific load case to be analyzed.

Fig. 1.24: Sketches for static analysis with loads and reactions.

The second step consists in the calculation of the reaction components. It is

often preferable to compute each reaction component by a relation that only
involves this component, and this can often be accomplished by a suitable
choice of the equilibrium condition to be used.
!
a) There are two reaction components RA and RA at A. Thus RB can be
determined via moment about A. This equation is independent of the
!
components RA and RA ,
! a−c
A b RB − (a − c)P = 0 ⇒ RB = P.
b
Reactions by equilibrium equations 21

!
b) The reactions RA and RB are both horizontal. Thus, RA can be deter-
mined independently by a vertical projection,

↑ RA − P = 0 ⇒ RA = P .

c) The line of action of the reaction components RA and RB intersect at a

!
point C, and RA is therefore determined by moment equilibrium about
this point.
! a−c
! !
C b RA − (a − c)P = 0 ⇒ RA = P.
b
!
Note the notation, in which the symbol A identifies the point as well as the
direction of positive moments, while arrow symbols ↑ and → are used for
vertical and horizontal projection, respectively.

1.5.1 Plane beams

The beam plays an important role in many structures. The beam is a struc-
tural element which is relatively long compared to a characteristic cross-
section. Typical examples are wooden or concrete beams with rectangular
cross-section and steel beams with I or H-section. Beams are often used in
contexts where the statics can be analyzed as one or more two-dimensional
problems. A typical first step in a static analysis is determination of the reac-
tions associated with a certain load on the structural element. In the following
a number of two-dimensional examples involving beams and frames are con-
sidered, and the reactions are determined. The flow of the load through beams
and frames is considered later in Chapter 3, and the ability of the beams to
withstand the effects of the load is discussed in Chapters 8–11.

Example 1.3. Simply supported beam. The horizontal beam shown in Fig. 1.25 is a
common structural element. The beam has a fixed simple support at the left end A and
a moving simple support with horizontal motion at the right end B. The beam carries a
vertical downward concentrated force P acting at the distance a from the left end A. Thus,
the support at A permits a vertical reaction RA and a horizontal reaction RA ! , while the

support at B permits a vertical reaction RB . The reactions are shown in the figure as they
would act on the beam, if positive. The supports and the associated reactions are precisely
sufficient to prevent motion of the beam, and thus the beam is statically determinate.

Fig. 1.25: Simply supported beam with concentrated force.

22 Equilibrium and Reactions

The three reactions are determined from the equilibrium conditions, expressed in the form
of one force projection equation and two moment equations as illustrated in Fig. 1.14b.
It is advantageous to use the equilibrium equations in a form, where each new condition
determines a new reaction. In the present case a simple straightforward procedure consists
in using a horizontal projection, followed by moment about A, and finally moment about B.

The only force with a horizontal component is the reaction RA! , and thus horizontal force
! = 0. When taking moment about A the two reactions through
equilibrium directly gives RA
this point do not contribute, and the resulting clockwise moment about A then takes the
form
! a
A a P − (a + b)RB = 0 ⇒ RB = P.
a+b
Similarly the moment equation about B does not involve the reactions pointing through
B, and thus the counterclockwise moment equation is
" b
B b P − (a + b)RA = 0 ⇒ RA = P.
a+b
It is noted that each of the three reactions have been determined independently, i.e. without
using reactions already determined.

It is often desirable to carry out a simple check of the reactions. In the present case a
simple check is obtained by calculating the sum of the vertical reactions,

b a
↑ RA + RB = P + P = P.
a+b a+b
Thus, the sum of the vertical reactions is equal to the imposed load P . It is noted that the
force is distributed to the supports according to the ‘lever rule’ illustrated in Fig. 1.5. !

Example 1.4. Cantilever beam. A beam supported only at one end is called a cantilever.
An example is illustrated in Fig. 1.26. The cantilever is supported by a fixed rigid support at
A, while the other end B is free. The fixed rigid support prevents vertical and horizontal
motion via the reaction forces RA and RA ! , while rotation is prevented by the reaction

moment MA . The reactions and their positive directions are indicated in the figure.

Fig. 1.26: Cantilever beam with concentrated force.

Equilibrium of the horizontal projection of all loads and reactions directly determine the
! = 0. Similarly, the vertical projection of the load and the reactions
horizontal reaction RA
determine the vertical reaction,

↓ P − RA = 0 ⇒ RA = P .
! and R , determined by projection, are independent
It is seen that the reaction forces RA A
of the location of the force a.
Reactions by equilibrium equations 23

The reaction moment MA is determined by taking the clockwise moment about A,

!
A a P − MA = 0 ⇒ MA = aP .

The moment MA is seen to increase linearly with a, confirming the intuitive feeling that
it requires more to support a load at a large distance.

Note, that the load P and the vertical reaction RA constitute a force couple aP , balanced
by the reaction moment MA . !

Example 1.5. Beams with hinge. Figure 1.27 shows two beams AD and DB, joined at
D by a hinge. The French term ‘charnier’ for a hinge is often used in connection with
structures. The total system consisting of the two beams must be in equilibrium, thus
providing three equilibrium conditions. In addition, the moment at the hinge vanishes,
and thus the moment of all loads and reactions on each side of the hinge must vanish. The
hinge introduces an extra degree of freedom in the motion of the structure, and thereby
a need for an extra reaction, bringing the number of reactions of the present structure
to four. There are three simple supports, of which the support at C is fixed, while the
supports at A and B permit horizontal motion. Thus, there is one horizontal reaction RC!

and three vertical reactions RA , RB and RC , as indicated in the figure. In the present
example the load consists of a concentrated force P acting at the center of the beam DB.

Fig. 1.27: Hinged beams with concentrated force.

When determining the reactions on structures with hinges, the forces transmitted through
the hinges are often included in the analysis. This is illustrated in Fig. 1.28, showing the
parts slightly separated to make room for the force components RD and RD ! acting on the

beam DB through the hinge.

Fig. 1.28: Reactions and hinge forces.

First the reactions on the beam DB are determined. It follows from horizontal force equilib-
rium that RD ! = 0. Thus, the hinge at D only transmits a vertical force R . The beam DB
D
now appears as simply supported, carrying a concentrated force P at its center. Symmetry
implies that the two reactions RD and RB are equal, and it then follows from vertical
force projection that RD = RB = 12 P . These reactions could also have been determined
by moments about the end points of the beam as in Example 1.3.
24 Equilibrium and Reactions

Fig. 1.29: Reactions on left beam.

Finally, the remaining reactions are determined by considering the beam AD as simply
supported and loaded by RD and RD ! as shown in Fig. 1.29. Horizontal force equilibrium

gives
! ! !
→ RC + RD = 0 ⇒ RC = 0.
Counterclockwise moment about A gives
" a+c
A a RC − (a + c) 12 P = 0 ⇒ RC = P,
2a
while clockwise moment about C gives
! c
C a RA + c 12 P = 0 ⇒ RA = − P.
2a

Note, that the beam AD carries half the load, and thus RA + RC = 12 P . However, as RA
is negative, the reaction RC is greater than the load RD = 12 P actually carried by this
part of the structure. !

1.5.2 Simple frames

Frames offer some additional features in the determination of reactions. The

following two examples illustrate how to deal with supports placed at different
heights and supports with rollers at inclined angles. As in the case of beams
the main point is to arrange the analysis in such a way that the individual
reaction components can be computed sequentially.

Example 1.6. Portal frame. This example deals with inclined load and a frame with
supports at different height. Figure 1.30a shows a rigid frame with a fixed simple support
at A and a simple support
√ on horizontal rollers at B. The frame is loaded by a concentrated
force of magnitude 2P with inclination −45◦ at C.

The load, decomposed in horizontal and vertical components of magnitude P , is shown in

! , R and R .
Fig. 1.30b together with the three reaction components RA A B

The reactions are determined sequentially by the following procedure. First the component
! is determined by horizontal projection of all load and reaction force components,
RA
! !
→ RA + P = 0 ⇒ RA = −P .

Counterclockwise moment about A then gives

"
A 2a RB − 2a P = 0 ⇒ RB = P .
Reactions by equilibrium equations 25

Fig. 1.30: Portal frame with inclined load.

The last reaction RA can be determined by moment about the support point B. However,
this would imply that both reaction components RA and RA ! would appear in the equation.

Although the horizontal component RA ! has already been determined it may be advanta-

geous to find an equilibrium condition in which only the unknown reaction component RA
appears together with the given loads. This condition is found by taking moment about
! and R . At this point
the point of intersection E of the two other reaction components RA B
these two reaction forces do not contribute to the moment. Clockwise moment about E
gives
!
E 2a RA + 2a P − 2a P = 0 ⇒ RA = 0 .
! and R
The results are controlled by translating the reactions RA B to the point of inter-
section E, and observing that they combine to a force along the diagonal EC. !

Example 1.7. Angle frame. The distribution of the load to the supports depends on the
support conditions as illustrated in this example. The issue is illustrated by comparing the
angle frame in Fig. 1.31a with vertical reaction at B with the similarly loaded frame in
Fig. 1.32a with a reaction at B inclined by 45◦ . The example also illustrates the use of the
equivalent load in the form of an equivalent concentrated force in the calculation of the
reactions.

Fig. 1.31: Angle frame with distributed load.

26 Equilibrium and Reactions

The distributed load with intensity p is equivalent to a concentrated force P = 2ap acting
at the center of the beam CB. This is independent of the support condition, and thus
applies to both cases as shown in Fig. 1.31b and Fig. 1.32b.

In the case of the frame with vertical reaction in B shown in Fig. 1.31 the calculation
proceeds in a manner quite similar to that of the simply supported beam in Example 1.3.
It follows immediately from horizontal projection that RA! = 0, and the vertical reaction

components then follow from moment about B and A, respectively.

!
B 2a RA − a (2ap) = 0 ⇒ RA = ap ,
"
A 2a RB − a (2ap) = 0 ⇒ RB = ap .
Vertical equilibrium is verified by projection, giving the sum of the reactions RA + RB =
2ap, matching the vertical load.

In the case of the frame with inclined support at B the calculation of reactions is arranged
in a slightly different way. In this case there are no parallel reaction components, and
each reaction component can therefore be determined independently by taking moment
about the intersection point of the two other reaction components. By this procedure
the calculation of each reaction component is independent of the order in which they are
computed.

Fig. 1.32: Angle frame with inclined support.

First the horizontal reaction RA! is calculated by taking moment about the point D defined

by the intersection of the lines of action of RA and RB . The point D is located vertically
above C at distance 2a. Thus, the moment equation is
"
! ! 2
D 3a RA − a (2ap) = 0 ⇒ RA = 3
ap .
! and R
The intersection E of the lines of action of RA B is located at the same height as A
at a distance 3a to the right. This gives the moment equation
!
4
E 3a RA − 2a (2ap) = 0 ⇒ RA = 3
ap .

The final equilibrium equation is a moment

√ equation about A. The distance from the point
A to the line of action of RB is ( 32 a) 2, and thus the moment equation is

" √ √
3 2
A 2
2a RB − a (2ap) = 0 ⇒ RB = 3
2ap .
Reactions by equilibrium equations 27

Alternatively, sliding the reaction RB along its√line of action to the point E at the same
level as A and using its vertical component 12 2RB gives the same equation. Finally, a
check is obtained by horizontal and vertical projection of the reaction forces,
!
√ √
→ RA − 12 2 RB = 0 , ↑ RA + 12 2 RB = 2ap ,

corresponding to zero horizontal load and a vertical load of P = 2ap.

An alternative procedure consists in determining the inclined reaction RB first, and then
obtaining RA and RA ! by vertical and horizontal projection, respectively. However, in that

procedure the determination of the components are not independent. !

1.5.3 Three-hinge frame

A classic form of a statically determinate frame is the three-hinge frame,

sometimes called a three-charnier frame, shown in Fig. 1.33. The frame has
two fixed supports with hinges, and has a third hinge in the frame structure.
As a result of the hinge within the frame it now has two reaction components
at each of the supports. The shape of the frame and the location of the
internal hinge may vary without changing the basic principles of analysis of
the frame. The task here is to determine the reactions at A and B.

Fig. 1.33: Three-charnier frame.

The reactions may be determined by purely graphical means using boldface

notation for vectors, as illustrated in Fig. 1.34 for a concentrated force P
acting on the left part of the frame at the distance x ≤ a from the left side.
The reaction force RB at B is the only force to act on the right half of the
frame, apart from any force transmitted through the hinge at C. Therefore
its line of action must pass through C in order not to produce any mo-
ment about C. Thus, the line of action of RB goes through BC as shown
in Fig. 1.34. By extending the lines of action of the load P and the reaction
RB the intersection point is identified. When the load is translated to this
point, it can be resolved into a force −RB through the support point B and
another force −RA through the support point A by use of the force parallel-
ogram construction introduced in connection with Fig. 1.1 in Section 1.1.1.
This purely geometric construction gives the reaction forces RA and RB as
28 Equilibrium and Reactions

Fig. 1.34: Force resolution for three-hinge frame.

vectors. While the graphic procedure establishes a direct relation between

the geometry of the frame and the resulting reactions, and thereby a good
intuitive understanding of the statics of the frame, it is less convenient for
numerical computations.

Fig. 1.35: Reactions of three-hinge frame.

A straightforward computation of the reactions can be performed by use of

the principles already used in the previous examples. First the reactions are
represented in terms of their vertical components RA , RB and their horizontal
! !
components RA , RB , shown in Fig. 1.35. These components can then be
determined sequentially by moment and force projection equations as follows.

Example 1.8. Reactions of the three-hinge frame. First the vertical reaction RB is
determined by moment equilibrium of the total frame about A,
Reactions by equilibrium equations 29

! x
A x P − 2a RB = 0 ⇒ RB = P.
2a
In a similar way the vertical reaction RA is determined by moment equilibrium of the total
frame about B,
" 2a − x
B (2a − x) P − 2a RA = 0 ⇒ RA = P.
2a
The hinge at C imposes a condition of zero moment of the forces on the right half of the
frame about C,
! x
! !
C h RB − a RB = 0 ⇒ RB = P.
2h
By using the forces on the right half of the frame the moment relation only involves the two
reaction components RB and RB ! . If the left half were used, the moment equation would

involve the similar reaction components at A and in addition the contribution form the
load P . The determination of the reactions is completed by using a horizontal projection
of all forces acting on the frame,

! ! ! x
→ RA = RB ⇒ RA = P.
2h
It is noted that the vertical reactions are independent of the height h of the frame, while
the horizontal components are proportional to x/2h, (x < a). Thus, increasing the height
of the location of the central hinge will reduce the horizontal reactions correspondingly. !

1.5.4 Space structures

Although many structures can be represented via planar problems, some must
be analyzed in three dimensions. In three-dimensional space the number of
equilibrium equations increases to six, e.g. projection in three directions and
moment about three lines. The following example demonstrates the procedure
for a simple space frame.

Example 1.9. Reactions of simple crane. Figure 1.36a shows a space structure repre-
senting a simple crane, carrying a tip load P . The structure is simply supported in A,
B and C, permitting horizontal displacement in the x-direction in A and C and in the
y-direction in B. Thus, the structure is statically determinate. Figure 1.36b shows a top
view of the crane structure. The vertical forces are indicated by circles with a bullet or
a cross inside, indicating positive upward or downward direction, respectively. First the
horizontal reactions are determined. Equilibrium in the x-direction directly gives zero hor-
izontal reaction RB! = 0 in B. Moment about the vertical line through B only receives a

contribution from the horizontal reaction in A, and thus RA ! = 0. The horizontal reaction

in C can be found by equilibrium in the y-direction, or by moment about the vertical line
through A, leading to RC ! = 0. Thus, all horizontal reactions vanish.

The vertical reaction in A is obtained directly by moment about BC:

!
BC a RA + a P = 0 ⇒ RA = − P .

Symmetry implies that RB = RC . This can also be verified by moment about the line DE,
where the only non-vanishing reactions are RB and RC . The magnitude of these identical
30 Equilibrium and Reactions

Fig. 1.36: a) Crane carrying vertical tip load P . b) Top view with forces.

reactions can then be found by vertical equilibrium:

↓ P − RA − RB − RC = 0 ⇒ RB = RC = P .

This shows that the three reactions have the same magnitude, because the support at A
has the same distance from the central plane BCD as the load P . !

1.6 Reactions by virtual work

The determination of statically determinate reactions on a structure was dis-
cussed in Section 1.5 based on direct use of equilibrium conditions. While this
is always possible for statically determinate reactions, it is sometimes conve-
nient to use the principle of virtual work instead. The advantage of this is
partly the ability to concentrate the calculation on individual reaction com-
ponents, even for compound structures with internal hinges, and partly the
visual display of displacement mechanisms that determine the contributions
from loads applied anywhere on the structure.

Fig. 1.37: Angle frame with distributed load.

The basic idea is illustrated in Fig. 1.37 showing a simply supported angle
frame. The frame has a simple support permitting vertical motion of B.
Thus, there is one horizontal reaction component RB at the support B. The
Reactions by virtual work 31

idea behind the use of the principle of virtual work for determination of
reactions is to release the support, and to impose a virtual displacement
δuB of the support. When the structure is statically determinate the release
of one support condition creates precisely one mechanism. The mechanism
created by releasing the horizontal support in B is a rotation about the fixed
support A, illustrated by the dashed line in the figure. The displacements
associated with the mechanism are considered infinitesimally small, and the
magnitudes of the displacements are then proportional to the imposed virtual
displacement δuB . For concentrated loads the corresponding virtual work
equation takes the form
)
δV = δuB RB + δuj Pj = 0 , (1.21)
j

with loads Pj and corresponding virtual displacements δuj , defined by the

mechanism corresponding to releasing the support with reaction RB . The par-
ticular mechanism shown in the figure corresponds to an infinitesimal counter
clockwise rotation of magnitude δuB /b, and thus the virtual displacements
of the force is δu1 = −(a/b)δuB . Hereby the equation of virtual work takes
the form
a a
δV = δuB RB − δuB P1 = 0 ⇒ RB = P1 . (1.22)
b b
The displacement field from the virtual rotation about A is here proportional
to δuB . In some cases it appears more direct to represent the displacement
field in terms of a virtual rotation – in this case the rotation about A, given
by δθA = δuB /b.

Example 1.10. Simply supported beam. This example illustrates the use of the principle
of virtual work to determine the reactions of the simply supported beam AB of length "
with a concentrated force P located at the distance a from the left support A shown in
Fig. 1.38a. The reactions have already been determined by use of force projection and
moments in Example 1.3, and thus the present example mainly illustrates the idea of the
virtual work equation, that will find more general applications later.

Each of the three equilibrium equations for the beam are replaced by an equivalent virtual
motion of the beam, found by releasing a support component and imposing an infinitesimal
positive virtual displacement in its place. The three virtual motions are illustrated in
Fig. 1.38b. The first is a virtual horizontal translation δu!A of the beam. The vertical force
P does not contribute to the virtual work produced by a horizontal motion, and thus the
corresponding virtual work equation is

δV = δu!A RA
!
= 0 ⇒ !
RA = 0.

Clearly, this condition is equivalent to a horizontal force projection.

The second virtual motion consists in lifting the support at A by the virtual displacement
δuA , whereby the beam rotates about B. The corresponding virtual work equation is
32 Equilibrium and Reactions

Fig. 1.38: a) Simply supported beam. b) Virtual displacements.

b b
δV = δuA RA − δuA P = 0 ⇒ RA = P,
" "
where b = " − a is the distance of the force from the right support B. It is seen that by
rotating the beam the contribution of the forces to the virtual work is proportional to
the distance from the center of rotation. This is quite analogous to the contributions in a
moment equation about the center of rotation.

Finally, the last reaction RB is found via the virtual work equation by lifting the support at
B by the virtual displacement δuB , whereby the beam rotates counter clockwise about B,
a a
δV = δuB RB − δuB P = 0 ⇒ RB = P.
" "
This relation is equivalent to a moment equation about A. !

Example 1.11. Beam with distributed load. In the case of distributed loads the deter-
mination of reactions proceeds in the same way via mechanisms obtained by releasing one
support at a time. Figure 1.39 shows a uniformly loaded beam ABC with a fixed simple
support at A and a simple support permitting horizontal motion at B. After releasing the
! = 0.
horizontal support component a virtual horizontal translation immediately gives RA

Fig. 1.39: Simply supported beam with distributed load.

The reaction RA is determined via a vertical virtual displacement δuA of point A. This cor-
responds to a clockwise rotation of the beam with angle δuA /a about B. The corresponding
virtual work equation is determined by considering the load on AB as an equivalent force
Reactions by virtual work 33

of magnitude ap acting at distance 12 a to the left of B, and similarly the load on BC

as an equivalent force of magnitude bp acting at the distance 12 b to the right of B. The
corresponding virtual work equation is

δuA ! 1 " a2 − b2
δV = δuA RA + b(bp) − 12 a(ap) = 0 ⇒ RA = p.
a 2 2a
The reaction RB is determined from the virtual displacement corresponding to lifting the
support B vertically by δuB = 1. Hereby the beam rotates the angle δuB /a about A, and
the virtual work equation becomes

δuB (a + b)2
δV = δuB RB − 1
2
(a + b)2 p = 0 ⇒ RB = p.
a 2a
The sum of the vertical reactions is RA + RB = (a + b)p. It is observed that the procedure
is nearly identical to the use of the similar moment equations. However, this changes when
the principle of virtual work is applied to composite structures connected by hinges as
illustrated in the following example. !

Example 1.12. Beams with hinge. The problem of a composite structure was illustrated
in Fig. 1.27, showing two beams connected by a hinge. The reactions of the composite
structure were determined sequentially in Example 1.5 by use of the equilibrium conditions.
The principle of virtual work offers an alternative, in which each reaction component of
the composite structure can be determined independently of the others. The procedure is
illustrated by application to the vertical reaction RC at the support C. A mechanism is
created by releasing the support C and then imposing a virtual vertical displacement δuC
as shown in Fig. 1.40.

Fig. 1.40: Simply supported beam with distributed load.

There is no moment at the hinge at D, and therefore the different rotations at the two
sides of the hinge do not lead to any contribution to the virtual work. It is seen directly
from the figure that the vertical motion of D is δuD = δuC (a + c)/a, and thus the virtual
work equation is

1 a+c
δV = RC δuC − 2
δuD P = 0 ⇒ RC = P.
2a
The other reactions can be determined similarly from mechanisms created by releasing that
particular support and imposing a virtual displacement in the direction of the reaction. !
34 Equilibrium and Reactions

1.7 Exercises
Exercise 1.1. The figure shows two planar forces
P1 = [−1, 2] and P2 = [1, 2], both acting at the
point A = [3, 2].
a) Determine the moment around C for each of
the two forces P1 and P2 individually and for
the two forces combined.
Exercise 1.2. The figure shows a force Pα =
[2 cos α, 2 sin α], acting at the point A = [3, 2].
a) Determine the angle α such that the moment
about C vanishes.
b) Determine the angle α where the moment
about C attains its largest magnitude.

Exercise 1.3. The top figure shows two parallel

planar forces P1 = −P2 = [0, 2] acting at A1 =
[3, 2] and A2 = [−3, 2], respectively.
a) Determine the moment around C for the force
couple P1 and P2 .
In the lower figure the forces are now inclined sim-
ilar to Exercise 1.1a such that P1 = [−1, 2] and
P2 = [−1, −2].
b) Determine the moment around C for the force
couple P1 and P2 .
c) Locate the points where the moment vanishes.

Exercise 1.4. The figure shows a three-dimensional space with a force P = [−1, 2, 1]
acting at the point A = [3, 1, 2].

a) Sketch the projection on each of the coordi-

nate planes and calculate the moments Mx ,
My and Mz as in Fig. 1.7.
b) Determine the moment vector [Mx , My , Mz ]
about the origin of the coordinate system by
use of (1.10) and compare with a).

Exercise 1.5. Show that the ‘lever rule’ (1.4) for two parallel forces P1 and P2 implies
a1 a2 a
= =
P2 P1 P

where a = a1 + a2 and P = P1 + P2 .

Exercise 1.6. The figure shows a simply sup-

ported beam loaded by two concentrated forces.
a) Show all possible reaction components with in-
dication of the sign convention used.
b) Determine the magnitude of the reactions.
Exercises 35

Exercise 1.7. The figure shows a cantilevered simply supported beam loaded by two
concentrated forces.
a) Show all possible reaction components with in-
dication of the sign convention used.
b) Determine the magnitude of the reactions.
c) Interchange the two forces and determine the
magnitude of the reactions.

Exercise 1.8. The figure shows a cantilevered simply supported beam loaded by a moment
M at the free end.
a) Show all possible reaction components
with indication of the sign convention
used.
b) Determine the magnitude of the reactions.

Exercise 1.9. The figure shows a simply supported beam with a vertical distributed load
with intensity p acting only on part of the beam.
a) Show all possible reaction components with indication of the sign convention used.
b) Determine and show the location and
magnitude of the force P that is equiva-
lent to the distributed load p.
c) Determine the magnitude of the reactions.

Exercise 1.10. The figure shows a beam with a fixed simple support at the left end and a
simple support at the right end permitting motion at an angle of 45◦ . The beam is loaded
by a uniform distributed load with intensity p.
a) Show all possible reaction components
with indication of the sign convention
used.
b) Determine and show the location and
magnitude of the force P that is equiva-
lent to the distributed load p.
c) Determine the magnitude of the reactions.

Exercise 1.11. The figure shows three differ-

ent load cases for a cantilever beam of length
a: i) linearly increasing load, ii) constant inten-
sity load, iii) linearly decreasing load. The fol-
lowing questions should be answered for each
of the three load cases.
a) Show all possible reaction components
with indication of the sign convention
used.
b) Determine and show the location and
magnitude of the force P that is equiva-
lent to the distributed load.
c) Determine the magnitude of the reactions.
36 Equilibrium and Reactions

Exercise 1.12. The figure shows two simply supported beams connected by a hinge at C,
loaded by a uniformly distributed load with intensity p acting over the full length of the
two beams.
a) Show all possible reaction components and interaction force components at C with
indication of the sign convention used.
b) Determine and show the forces
P1 and P2 that are equivalent to
the distributed load p on the two
beams.
c) Determine the magnitude of the
reactions and the interaction
forces at C.

Exercise 1.13. The figure shows a three-hinge frame with the internal hinge located at
the center of the span. The frame is loaded by a uniformly distributed load with intensity
p along the top of the frame.

a) Show all possible reaction components and

interaction force components at E with in-
dication of the sign convention used.
b) Determine and show the forces P1 and P2
that are equivalent with the distributed
load p on BE and EC, respectively.
c) Determine the magnitude of the reactions
and the interaction forces.

Exercise 1.14. The figure shows a simply supported angle frame, with a uniformly dis-
tributed load along CB. The frame has a fixed simple support at B and a simple support
permitting motion at an angle of 45◦ at A.

a) Show all possible reaction components

with indication of the sign convention
used.
b) Determine and show the force P that is
equivalent to the distributed load p.
c) Determine the magnitude of the reac-
tions.
d) Use the force parallelogram rule to illus-
trate, how the load represented by the
equivalent force P is decomposed into
the reaction forces.
Exercises 37

Exercise 1.15. The figure shows a three-

hinge frame with the internal hinge at the top
point C of a straight inclined roof beam. The
roof is loaded by a vertical load of intensity p
per unit horizontal length.
a) Show all possible reaction components and
interaction force components at C with in-
dication of the sign convention used.
b) Determine and show the force P that is
equivalent to the distributed load p.
c) Determine the magnitude of the reactions
and the interaction forces at C.

Exercise 1.16. The figure shows an angle frame ABC located in the x, y-plane. The frame
has a fixed simple support in A, and simple supports permitting motion in the y-direction
and in both directions in C and B, respectively. The length of AB is a, while the length of
BC is B. At the center of BC a transverse force P is acting in the downward out-of-plane
direction.

a) Define the reaction forces in a figure, and

explain why the structure is statically de-
terminate.
b) Show that the three in-plane reaction
forces are zero.
c) Determine the remaining three out-of-
plane reaction forces.

Exercise 1.17. Three beams of equal length AD, BE and CF are placed in a horizontal
plane as shown in the figure. Each beam has a simple support at one end, while the other
end rests on the mid-point of one of the other
beams. The beam AD is loaded by a downward
force of magnitude 14 N, located in the middle
between A and B.
a) Determine the vertical reaction forces at
the supports D, E and F as well as the
forces transferred between the beams at
A, B and C.
It is convenient to formulate the equilibrium
conditions for each individual beam.

Exercise 1.18. The figure shows a simply supported girder ABCD with a hinge at C.
The girder is loaded by a uniform distribu-
tion of vertical forces of intensity p, similar
to Exercise 1.12.
a) Determine all reaction components
RA , RA! , R
B and RD by use of the
principle of virtual work.
38 Equilibrium and Reactions

Exercise 1.19. The girder ACEF DB is simply supported at A, B, C and D. The central
part is connected via hinges at E and F .
The vertical reactions at A, B, C and D
are to be determined by the principle of
virtual work for the following load cases:
a) Uniform vertical load p over the full
length AB.
b) Uniform vertical load p over the cen-
tral part EF .
c) Uniform vertical load p over ACEF .

Exercise 1.20. Determine the remaining reactions in Example 1.12 by the principle of
virtual work.