DESIGN OF STEEL

STRUCTURES
(VOLUME I)

[S.I. UNITS]

By
Dr. Ram Chandra
B.E., M.E. (Hons.), M.I.E., Ph.D. (Roorkee)
Professor and Head
Department of Structural Engineering
Faculty of Engineering
M.B.M. Engineering College
University of Jodhpur, Jodhpur

STANDARD BOOK HOUSE

unit of : RAJSONS PUBLICATIONS PVT. LTD.
1705-A, Nai Sarak, PB.No. 1074, Delhi-110006 Ph.: +91-(011)-23265506
Show Room: 4262/3, First Lane, G-Floor, Gali Punjabian, Ansari Road, Darya Ganj,
New Delhi-110002 Ph.: +91-(011) 43751128 Tel Fax : +91-(011) 43551185,
Fax: +91-(011)-23250212
E-mail: [email protected] www.standardbookhouse.in
Published by:
RAJINDER KUMAR JAIN
Standard Book House
Unit of: Rajsons Publications Pvt. Ltd.
1705-A, Nai Sarak, Delhi - 110006
Post Box: 1074
Ph.: +91-(011)-23265506 Fax: +91-(011)-23250212
Showroom:
4262/3, First Lane, G-Floor, Gali Punjabian
Ansari Road, Darya Ganj, New Delhi-110002
Ph.: +91-(011)-43751128, +91-(011)-43551185
E-mail: sbhl0@ hotmail.com
Web: www.standardbookhouse.in

First Published : 1970

Second Edition : 1975
Third Edition : 1978
Fourth Edition : 1981
Fifth Edition : 1982
Sixth Edition : 1984
Seventh Edition : 1986
Eight Edition : 1988

DSS Vol–I
Ninth Edition : 1989
Tenth Edition : 1992
Eleventh Edition : 1995
Twelveth Edition : 1998
Thirteenth Edition : 2001
Fourteenth Edition : 2004
Fifteenth Edition : 2007
Sixteenth Edition : 2010
Seventeenth Edition : 2013
Eighteenth Edition : 2016

© Publishers

All rights are reserved with the Publishers. This book or any part thereof, may not be
reproduced, represent, photocopy in any manner without the prior written permission of
the Publishers.

` 495.00

ISBN: 978-81-89401-40-5

Typeset by:
C.S.M.S. Computers, Delhi.

Printed by:
R.K. Print Media Company, New Delhi
 Foreword
Tables and clauses from the Indian Standard Specifications have been
reproduced in the book with the kind permission of the Indian Standards
Institution.
It is desirable that for complete detail, reference be made to the latest versions
of the Standards Institution, Manak Bhavan, 9, Bahadur Shah Zafar Marg,
New Delhi-1, or from its branch offices at Mumbai, Kolkata, Kanpur and
Chennai.
SYSTEM INTERNATIONAL d’ UNITES
(SI System of Units)
In order to avoid the conversion of results obtained by engineers working with
the Foot Pound Second (FPS) System (gravitational) of units in terms of
centimetre-gram second absolute system of units used by the scientists, a need
of common system of units was realised. The General Conference on Weights
and Measures held at Paris in 1960 finalised the System International d’ Unites
(SI). It is an absolute system of units. The mass is considered as fundamental
unit and not the force. BIS has included a comment of transition in IS 3616–
1966. ‘Recommendation on the International System (SI) Units’ that this system
has begun to replace older system of units in several branches of science and
technology. The SI is a universal system of units and it has been adopted in
France as a legal system and it is likely to become common in many countries.
SI units have the following six basic units.
Unit of length (metre, m)
The length equal to 1,650, 763.73 wavelengths, in vacuum, of the radiation
corresponding to the transition between 2p19 and 5d5 levels of the krypton η
atom of mass 86 is known as one metre.
Unit of mass (kilogram, kg)
The mass of platinum-indium cylinder deposited at the International Bureau
of Weights and Measures and declared as the international prototype of the
kilogram by the First General Conference of Weights and Measures is called as
one kilogram.
Unit of time (second, s)
1131, 566, 925, 974.7 of the length of the tropical year for 1900, the year
commencing at 1200 hours universal time on the first day of January, 1900 is
termed as one second.
Unit of electric current (ampere, A)
The constant current which flows in two parallel straight conductors of infinite
length of negligible circular cross-section and placed at a distance of one metre
from each other in vacuum producing a force of 2 × 10–7 New tons per metre
length between the conductors is defined as an ampere.
Unit of thermodynamic temperature (degree Kelvin, K)
The degree interval of the thermodynamic scale on which the temperature of
triple point of water is 273.16 degrees, is known as one degree Kelvin.
Units of luminous intensity (candela, cd)
One sixtieth part of luminous intensity normally emitted by one hundred
millimetre square of integral radiator (black body) at the temperature of solidifi-
cation of platinum is called as one candela.
The SI units make the use of multiples and sub-multiples 1000 times or 1/
1000 times the unit quantity and in powers of 103 (kilo) or 10–3 (milli) in respect
of still larger and smaller quantities respectively. The lengths are measured
usually in kilometre (1 km = 1000 m), metre and millimetre (1 mm = 10–3 m).The
symbols of units are not to be suffixed with V for plural.
The force is a derived quantity and physical law connecting the quantity to
the fundamental quantities or previously obtained derived quantities is force =
mass × acceleration. It is defined as that force which produces unit acceleration
i.e., 1 m per sec2 in a unit mass of 1 kg. Its unit is Newton (N). Though, the
Newton is a small unit, a still larger unit kN may be used. The intensity of force
(viz., stress) due to 1 Newton over a unit area of one metre square is known as
one pascal. It is denoted by symbol, Pa. (1 Pa = 1 N/m2 and 106. Pa = 1 N/mm2,
viz. 1 MPa = 1 N/mm2).
SI system of units have many advantages. The units are very handy. The
burden of non-decimal coefficients in foot-pound second system is avoided. It
has relatively large main units in contrast to centimetre-gram-second system.
At the same time, it is closely related to centimetre-gram-second system of units.
In practice, it results in perfectly reasonable number when the value of g = 10
m/sec2 is used instead of 9.806 m/sec2.
(Professor V.S. Mokashi, Visvesvaraya Regional College of Engineering,
Nagpur in his paper titled as International System (SI) Units and their
Application to Engineering published in Journal of Institution of Engineers,
India, Vol. 19, March 1970 has highlighted the advantages and discussed SI
units. A reference has been made to this paper).

Structural Engineering is the science and art of planning, design,

construction, operation, maintenance and rehabilitation of structures. The term
“structures” includes bridges, buildings and all types of civil engineering
structures (towers, shells, etc.) composed of any structural material.
(Reference: Brochure, International Association for Bridge and Structural
Engineering, Final Invitation to the International Conference, Structural
Eurocodes, Daros, Switzerland September 14–16, 1992).
Author
 Preface to the Eleventh Edition
In the subsequent editions of this book, since first edition published in 1970
uptil now, the author enhanced the text by adding useful matter, fresh topics
such as column formulae for axial stress in compression, design of built-up and
perforated cover plate columns, modified and adjusted interaction formulae,
equivalent axial load method of design of eccentrically loaded columns,
approximate method of design of combined footing, graphical method of
curtailment of flange plates, corrugated aluminium sheets used for roof covering
and several examples. The author also added further text of design of high
strength friction grip bolts.
The eleventh edition of the book itself is a fourth edition in S.I. system of
units (viz., system international d’ unites) and revised, rewritten and updated
as per the latest code (viz., ‘Code of Practice for General Construction in Steel.
IS : 800–1984) incorporating the revision of permissible stresses, effective length
of the columns with idealised support conditions and columns in framed
structures and Merchant Rankine formula for the allowable stresses. The concept
of shear lag, design of semi-rigid connections, their behaviour (linear and non-
linear) and methods of analysis have also been included.
The abbreviated symbols for Rolled Steel Sections as recommended
in IS: 808–1989 have been used throughout the text of the book. Various
definitions relating to the new and rational concept of Wind-Load as per IS:
875 (Part III)–1987 have been given in Chapter 2. Accordingly Chapter 9 (viz.
Design of Roof Trusses) has been completely revised and determination of wind
load has been thoroughly described and illustrated.
Author expresses his sincere thanks to his colleagues, members of staff in
various engineering colleges and students for appreciating the efforts made by
them. The author also expresses his personal thanks for the Publisher Shri
Rajinder Kumar Jain and Shri Sandeep Jain for getting the book prepared by
latest technique and bringing out the book in such a nice getup.
Author shall welcome the suggestions from the readers for the further
improvement of the book in fothcoming editions.
August 2013 Dr. Ram Chandra
Jodhpur
 Preface to the First Edition
In this book, the author with his long teaching experience in the subject has
made an attempt to present the subject matter of design of steel structures in a
way which lays emphasis on the fundamentals, keeping in view the difficulties
experienced by the students. Every basic principle, method, equation or theory
has been presented in simplified manner. Metric system of units has been used
throughout the text. Indian Standards Specifications have been followed. The
book is intended for the use of degree, diploma and A.M.I.E. students in various
branches of engineering. The book deals with design of structural members and
their connections.
Each topic introduced is thoroughly described. A number of design problems
including problems for examinations of the University of Jodhpur, and A.M.I.E..
has been solved to illustrate the theory and practice. Slide-rule computation
accuracy is adequate for the design and has been followed. The chapters have
been so arranged that it facilitates self-understanding of the subject, during
study. In spite of careful scrutiny of the manuscript, it is possible that some
typographical and computational errors are still left. The author shall be highly
obliged to any one who brings these errors to his notice.
The author is thankful to Shri J.N. Srivastava and other colleagues who have
very generously helped with their suggestions. The author is also thankful to
the University of Jodhpur, Jodhpur and the Institution of Engineers, India, for
following the use of their examinations, problems.
Suggestions from the readers for the improvement of the book are welcome.
September 11,1970 Ram Chandra
Contents

PART 1 DESIGN OF STEEL STRUCTURES

CHAPTER 1 GENERAL 3–51

1.2 Structures 4
1.3 Classification of Structures 4
1.3.1 One Dimensional Supporting System 5
1.3.2 Two Dimensional Supporting System 5
1.3.3 Three Dimensional Structures 5
1.4 Classification of Structural Members 5
1.5 Advantages and Disadvantages of Steel Members 6
1.6 Structural Steel 6
1.6.1 Mild Steel 7
1.6.2 High Tensile Steel 8
1.7 Production of Steel 11
1.8 Recent Developments in Material 11
1.9 Rolled Structural Steel Sections 12
1.10 Rolled Steel Beam Sections 12
1.10.1 Designation 13
1.11 Rolled Steel Channel Sections 14
1.11.1 Designation 14
1.12 Rolled Steel Tee Sections 15
1.12.1 Designation 15
1.13 Rolled Steel Angle Sections 16
1.13.1 Designation 16
1.14 Rolled Steel Bars 17
x x Index
CONTENTS

1.14.1 Designation 17
1.15 Rolled Steel Tubes 17
1.16 Rolled Steel Flats 18
1.16.1 Designation 18
1.17 Rolled Steel Sheets and Strips 18
1.17.1 Designation 18
1.18 Rolled Steel Plates 18
1.18.1 Designation 18
1.19 Recent Developments in Sections 19
1.20 Specification 19
1.21 Loads and Stresses 22
1.22 Dead Load 23
1.23 Live Loads 27
1.24 Wind Load 37
1.25 Snow Load 40
1.26 Seismic Load (Earthquake Force) 40
1.27 Soil and Hydrostatic Pressure 40
1.28 Erection Effects 40
1.29 Dynamic Effects (Impact) 41
1.30 Temperature Effects 41
1.31 Load Combinations 41
1.32 Stresses 41
1.33 Stress–strain Relationship for Mild Steel 42
1.33.1 Significance of Upper Yield Point 43
1.34 Tensile Stress 43
1.35 Compressive Stress 43
1.36 Bearing Stress 44
1.37 Working Stress 44
1.38 Increase in Permissible Stresses 44
1.39 Factor of Safety 45
1.40 Methods of Design 46
1.40.1 Simple Design 47
1.40.2 Semi-rigid Design 47
1.40.3 Filly-rigid Design 47
1.40.4 Plastic Design 48
1.41 Definitions 48
1.41.1 Modulus of Elasticity 48
1.41.2 Shear Modulus of Elasticity 48
1.41.3 Bulk Modulus of Elasticity 48
1.41.4 Tangent Modulus of Elasticity 49
1.41.5 Poisson’s Ratio 49
 INDEX
CONTENTS xi xi

1.41.6 Yield Point 49

1.41.7 Yield Strength 49
1.41.8 Tensile Strength 49
1.41.9 Fatigue Strength 49
1.41.10 Impact Strength 49
1.41.11 Strain Hardening Modulus 50
1.41.12 Ductility 50
1.42 Stability of Structure 50
1.42.1 Overturning 50
1.42.2 Sliding 51
1.43 Stiffness of Structure 51

CHAPTER 2 DESIGN OF RIVETED, BOLTED AND

PIN CONNECTIONS 52–124
2.1 Introduction 52
2.2 Rivets 52
2.3 Rivet Heads 54
2.4 Rivet Holes 54
2.5 Definitions of Terms Used in Riveting 55
2.5.1 Nominal Diameter of Rivet 55
2.5.2 Gross Diameter of Rivet 55
2.5.3 Pitch of Rivet (p) 55
2.5.4 Gauge Distance of Rivets (g) 55
2.5.5 Gross Area of Rivet 55
2.5.6 Rivet Line 55
2.5.7 Staggered Pitch 56
2.6 Working Stresses in Rivets 56
2.7 Riveted Joint 56
2.7.1 Lap Joint 57
2.7.2 Butt Joint 59
2.8 Transmission of Load in Riveted Joints 60
2.9 Failure of a Riveted Joint 61
2.9.1 Shear Failure of Rivets 61
2.9.2 Shear Failure of Plates 61
2.9.3 Tearing Failure of Plates 62
2.9.4 Bearing Failure of Plates 62
2.9.5 Splitting Failure of Plates 62
2.9.6 Bearing Failure of Rivets 63
2.10 Strength of Riveted Joint 63
2.10.1 Strength of a Riveted Joint against the
Shearing of the Rivets 63
2.10.2 Strength of Riveted Joint against the Bearing of the Rivets 64
2.10.3 Strength of Plate in Tearing 65
2.10.4 Strength of Riveted Lap Joint 66
xiixii Index
CONTENTS

2.10.5 Strength of Riveted Butt Joint 66

2.11 Efficiency or Percentage Strength of Riveted Joint 67
2.12 Rivet Value 67
2.13 Assumptions for Design of Riveted Joint 67
2.14 Arrangement of Rivets 68
2.14.1 Chain Riveting 68
2.14.2 Diamond Riveting 68
2.15 Specifications for Design of Riveted Joint 69
2.16 Design Procedure for Riveted Joint 71
2.17 Eccentric Riveted Connections 79
2.18 Bolt and Pin Connections 98
2.19 Advantages of Bolted Connections 98
2.20 Disadvantages of Bolted Connections 99
2.21 Bolts 99
2.21.1 Unfinished Bolts 99
2.21.2 Turned Bolts 103
2.21.3 High Strength Bolts 107
2.22 Definitions of Terms used in Bolts and Bolting 108
2.23 Bolt, Nut and Washer Assembly 108
2.24 Stresses in Bolts 109
2.25 Permissible Stresses in Bolts 111
2.26 Design of High Strength Friction Grip Bolts 112
2.26.1 Shear Connections 112
2.26.2 Tension Connections 113
2.26.3 Combined Shear and Tension 113
2.26.4 Repeated Variation of Stress 113
2.27 Advantages of Pin Connections 114
2.28 Disadvantages of Pin Connections 114
2.29 Pins 114
2.29.1 Forged Steel Pin 114
2.29.2 Undrilled Pins 115
2.29.3 Drilled Pins 115
2.30 Pin Connection 115
2.31 Stresses in Pins 117
2.31.1 Shear Stress 117
2.31.2 Bearing Stress 117
2.31.3 Bending Stress 117
2.32 Permissible Stresses in Pins 118
Problems 122

CHAPTER 3 DESIGN OF COLUMNS AND COMPRESSION

MEMBERS 125–241
3.1 Introduction 125
3.2 Axially Loaded Columns 125
3.3 Effective Length of Compression Member 128
 INDEX
CONTENTS xiii
xiii

3.3.1 Position Restraint 128

3.3.2 Direction Restraint 128
3.4 Effective Sectional Area 132
3.5 Radius of Gyration 132
3.6 Slenderness Ratio of Compression Member 133
3.7 Column Formulae for Axial Stress in Compression 134
3.8 Design Formula 147
3.9 Common Sections of Compression Members 151
3.10 Strength of Compression Members 151
3.11 Angle Struts 152
3.11.1 Continuous Members 152
3.11.2 Discontinuous Members 152
3.12 Built-Up Compression Members 155
3.13 Thickness of Elements in Compression Members 157
3.14 Design of Axially Loaded Compression Member 166
3.15 Laced and Battened Columns 174
3.16 Batten Plates 189
3.16.1 Design of Batten Plates 189
3.17 Perforated Cover Plates 198
3.18 Eccentrically Loaded Columns 200
3.19 Design of Eccentrically Loaded Columns 205
3.20 Modified and Adjusted Interaction Formulae 211
3.21 Equivalent Axial Load Method of Design of Eccentrically Loaded
Columns 217
3.22 Bending Factor Method of Design of Eccentrically Loaded Column 219
3.23 Column Splice 225
3.24 Design of Column Splice 226
3.25 Encased Columns 233
Problems 237

CHAPTER 4 DESIGN OF COLUMN BASES AND COLUMN

FOOTINGS 242–284
4.1 Introduction 242
4.2 Slab Base 243
4.3 Gusseted Base 247
4.4 Column Bases Subjected to Moment 251
4.5 Independent Column Footing 257
4.5.1 Grillage Footing 257
4.6 Combined Column Footing 265
4.6.1 Steel Grillage Footing to Support two Equal Column Loads 265
4.6.2 Steel Grillage Footing to Support two Unequal Column Loads 266
4.7 Approximate Method of Design of Combined Footing 270
Problems 283
xivxiv Index
CONTENTS

CHAPTER 5 DESIGN OF TENSION MEMBERS 285–312

5.1 Introduction 285
5.2 Type of Tension Members 285
5.2.1 Wires and Cables 286
5.2.2 Rods and Cables 286
5.2.3 Single-structural Shapes and Plates 287
5.2.4 Built-up Members 287
5.3 Net Sectional Area 288
5.4 Net Effective Section for Angles and Tees in Tension 290
5.5 Allowable Stress in Axial Tension 292
5.6 Strength of a Member in Axial Tension 292
5.7 Design of Tension Members Subjected to Axial Tension 296
5.8 Design of Tension Members Subjected to both
Bending and Axial Tension 297
5.9 Tension Member Splice 303
5.10 Lug Angles 305
5.11 Shear Lag 308
Problems 311

CHAPTER 6 DESIGN OF BEAMS 313–417

6.1 Introduction 313
6.2 Bending Stress 315
6.2.1 Allowable Stress in Bending 317
6.3 Shear and Bearing Stresses 332
6.4 Effective Span and Deflection Limitation 334
6.4.1 Limiting Vertical Deflection 334
6.4.2 Limiting Horizontal Deflection 334
6.5 Laterally Supported Beams 335
6.6 Design of Laterally Suppoted Beams 337
6.7 Laterally Unsupported Beams 347
6.7.1 Where the Flanges have Equal Moment of Inertia
about yy-Axis 348
6.7.2 Maximum Permissible Bending Compressive Stress
in Beams and Plate Girders 353
6.7.3 Elastic Critical Stress 358
6.8 Effective Length of Compression Flange 363
6.9 Web Crippling 370
6.10 Web Buckling 371
6.11 Built-up Beam 375
6.12 Design of Built-up Beams 376
6.13 Design of Rivets Connecting Cover Plates with the Flanges of Beam 385
 INDEX
CONTENTS xv xv

6.14 Lintels 387

6.15 Jack Arch 390
6.16 Crane Gantry Girder 393
6.16.1 Limitation of Vertical Deflection 400
6.16.2 Limitation of Horizontal Deflection 400
6.17 Filler Joist 407
6.17.1 Bending Moment 407
6.17.2 Moment of Resistance 408
6.17.3 Spacing 408
6.17.4 Thickness of Concrete 409
6.17.5 Bending Stress 409
6.17.6 Shear Stress 409
6.17.7 Bearing Stress 409
6.17.8 Span Depth Ratio 409
6.18 Encased Beams 409
6.18.1 Design of Encased Beam 410
Problems 414

CHAPTER 7 DESIGN OF BEAM CONNECTIONS 418–476

7.1 Introduction 418
7.1.1 Flexible Connections 418
7.1.2 Semi-rigid Connections 419
7.1.3 Rigid Connections 419
7.2 Framed Connections 421
7.3 Unstiffened Seated Connections 424
7.4 Stiffened Seated Connections 425
7.5 Design of Framed Connections 426
7.6 Design of Unstiffened Seated Connections 428
7.7 Design of Stiffened Seated Connections 430
7.8 Small Moment Resistant Connections 441
7.7.1 Choice of Design Method 443
7.9 Large Moment Resistant Connections 444
7.9.1 Bracket Connections 444
7.9.2 Split Beam Connection 447
7.10 Semi-rigid Connections 460
7.11 Behaviour of Semi-rigid Connections 462
7.12 Beam-line Method 465
7.13 Modified Slope-deflection Method 468
7.14 Modified Moment Distribution Method 472
Problems 476

CHAPTER 8 DESIGN OF PLATE GIRDER 477–561

8.1 Introduction 477
xvixvi Index
CONTENTS

8.2 Elements of Plate Girder 478

8.3 Area of Flanges of Plate Girder 479
8.4 Depth of Plate Girder 481
8.5 Economical Depth of Plate Girder 481
8.6 Self-weight of Plate Girder 483
8.7 Impact 484
8.8 Assumptions for Design for Plate Girder 485
8.9 Allowable Bending Stress 485
8.10 Allowable Shear Stress 485
8.11 Design of Web Plate 490
8.11.1 Web Buckling 491
8.12 Design of Flanges 497
8.12.1 Flange Area Method 498
8.12.2 Moment of Inertia Method 499
8.13 Length of Flange Plates : (Curtailment of Flange Plates) 507
8.13.1 Graphical Method of Curtailment of Flange Plates 509
8.14 Connection of Flange Angles to Web 517
8.15 Connection of Flange Plates to Flange Angles 519
8.16 Diagonal Buckling of Web 523
8.17 Bearing Stiffeners 525
8.18 Intermediate Stiffeners 527
8.18.1 Vertical Stiffeners 528
8.18.2 Horizontal Stiffeners 529
8.19 Web Splice 537
8.19.1 Web Splice (Rational Splice) 537
8.19.2 Web Splice (Moment Splice) 543
8.19.3 Web Splice (Shear Splice) 547
8.20 Flange Splices 551
8.20.1 Splice of Flange Angles 552
8.20.2 Splice of Flange Plates 553
Problems 558

CHAPTER 9 DESIGN OF ROOF TRUSSES 562–671

9.1 Introduction 562
9.2 Types of Trusses 563
9.3 Various Terms Used in Roof Trusses 565
9.3.1 Top Chord 565
9.3.2 Bottom Chord 565
9.3.3 Span 565
9.3.4 Rise 566
9.3.5 Pitch 566
9.3.6 Slope 566
 INDEX
CONTENTS xvii
xvii

9.3.7 Purlins 567

9.3.8 Sub-purlins 567
9.3.9 Rafters 567
9.3.10 Sheathing 567
9.3.11 Panel 567
9.3.12 Bay and Spacing of Purlins 567
9.3.13 Spacing of Trusses 568
9.3.14 Ridge Line and Eaves 568
9.3.15 Sag Tie 568
9.3.16 Principal Rafter 568
9.4 Economical Spacing of Roof Trusses 568
9.5 Roof Coverings 569
9.5.1 Slates 569
9.5.2 Tiles 570
9.5.3 Lead 570
9.5.4 Zinc 570
9.5.5 Glass 570
9.5.6 Galvanized Corrugated Iron Sheets 570
9.5.7 Asbestos Cement Sheets 571
9.5.8 Corrugated Aluminium Sheets 572
9.6 Loads on Roof Trusses 572
9.6.1 Dead Load 572
9.6.2 Live Loads 573
9.6.3 Snow Loads 574
9.6.4 Wind Loads 574
9.7 Design Wind Speed 579
9.7.1 Risk Coefficient (Probability Factor), k1 579
9.7.2 Terrain, Height and Structure Size Factor, k2 581
9.7.3 Topography Factor k3 583
9.8 Design Wind Pressures 585
9.9 Wind Forces (Loads) on Buildings/Structures 585
9.10 Wind Load on Individual Members 587
9.11 External Pressure Coefficients 588
9.11.1 Walls 588
9.11.2 Pitched Roofs of Rectangular Clad Buildings 588
9.11.3 Mono-slope Roofs of Rectangular Clad Buildings 591
9.11.4 Canopy Roofs with [(L/4) < (h/w) < 1 and 1(< L/W) < 3] 593
9.12 Internal Pressure Coefficients 599
9.12.1 Building Having a Small Degree of Permeability 600
9.12.2 Buildings Having Medium Degree of Permeability 600
9.12.3 Buildings Having Large Degree of Permeability 600
9.12.4 Buildings Having High Degree of Permeability 600
9.13 External Pressure Coefficients for some other Roofs 601
xviii
xviii Index
CONTENTS

9.13.1 Pitched and Saw-tooth Roofs of Multi-span Building 602

9.13.2 Pressure Coefficients on Overhangs from Roofs 602
9.13.3 Combined Roofs and Roofs with a Sky-light 603
9.14 Combination of Loads on Roof Trusses 604
9.15 Bracing of Roof Trusses 605
9.16 Assumptions in Analysis of Roof Trusses 606
9.17 Purlins 607
9.18 Principal Axes of Section 607
9.18.1 Analytical Method 609
9.18.2 Graphical Method 611
9.19 Maximum Bending Stress Due to Unsymmetrical Bending 614
9.19.1 By Resolving Bending Moment into two Components
along the Principal Axes 614
9.19.2 By Resolving Bending Moment into two Components
along any two Perpendicular Axes, Passing through
the Centroid of Section 615
9.19.3 By Locating the Neutral Axes of Beam Section 617
9.20 The Z-polygon 621
9.21 Deflection of Beams under Unsymmetrical Bending 626
9.22 Design of Purlins Subjected to Unsymmetrical Bending 629
9.22.1 Design of Angle Iron Purlins 631
9.23 Design of Roof Trusses 638
9.24 Typical Details of Roof Trusses 641
Problems 671

CHAPTER 10 DESIGN OF WELDED JOINTS 672–745

10.1 Introduction 672
10.1.1 Fusion Welding 673
10.1.2 Pressure Welding 673
10.1.3 Thermit Welding 673
10.1.4 Union Melt Welding 673
10.2 Advantages of Welding 675
10.3 Disadvantages of Welding 676
10.4 Types of Welded Joints 676
10.5 Butt Weld 676
10.5.1 Square Butt Weld 676
10.5.2 Single-V Butt Weld 677
10.5.3 Double-V Butt Weld 677
10.5.4 Single-U Butt Weld 677
10.5.5 Double-U Butt Weld 678
10.5.6 Single-J Butt Weld 678
10.5.7 Double-J Butt Weld 678
 INDEX
CONTENTS xixxix

10.5.8 Single-Bevel Butt Weld 679

10.5.9 Double-Bevel Butt Weld 679
10.5.10 Specifications of the Butt Weld 680
10.6 Fillet Weld 681
10.6.1 Specifications of Fillet Weld 683
10.6.2 Intermittent Fillet Welds 685
10.7 Slot Weld and Plug Weld 685
10.7.1 Slot Welding 686
10.7.2 Seam Welding 686
10.7.3 Pipe Welding 687
10.8 Imperfections in Weld 687
10.8.1 Surface Imperfections in Weld 687
10.8.2 Internal Imperfections in Weld 689
10.9 Stress in Welds 691
10.9.1 Butt Weld 691
10.9.2 Fillet Weld 691
10.9.3 Direct Stress Formula 692
10.9. 4 Flexure Formula 694
10.9.5 Torsion Formula 695
10.9.6 Beam Shear Formula 697
10.10 Combined Stresses in Welds 698
10.10.1 Butt Weld 698
10.10.2 Fillet Weld 699
10.11 Working Stresses in Welds 701
10.12 Design of Welded Joints Subjected to Axial Load 704
10.13 Design of Welded Joints from Unsymmetrical Sections
Subjected to Axial Load 708
10.14 Design of Welded Joints Subjected to Eccentric Load 713
10.15 Welded Plate Girder 728
Problems 740

CHAPTER 11 WELDED BEAM CONNECTIONS 746–764

11.1 Introduction 746
11.2 Direct Welded Connections 746
11.3 Welded Framed Connections 749
11.4 Welded Seat Connections 755
11.4.1 Unstiffened Welded Seat Connections 755
11.4.2 Stiffened Welded Seat Connections 756
11.5 Moment Resistant Welded Connections 760
Problems 764
xx xx Index
CONTENTS

CHAPTER 12 DESIGN OF ROUND TUBULAR STRUCTURES

765–784
12.1 Introduction 765
12.2 Round Tubular Sections 766
12.3 Permissible Stresses 771
12.4 Tube Columns and Comression Members 773
12.5 Crinkling 775
12.6 Tube Tension Members 775
12.7 Tubular Roof Trusses 775
12.8 Joints in Tubular Trusses 776
12.9 Design of Tubular Beams 779
12.10 Design of Tubular Purlins 780
Problems 784

PART 2 DESIGN OF TIMBER STRUCTURES

CHAPTER 13 DESIGN OF TIMBER STRUCTURES 787–871

13.1 Introduction 787
13.2 Deffects in Timber 787
13.2.1 Natural Defects 788
13.2.2 Other Defects 788
13.3 Decay of Timber 790
13.4 Preservation of Timber 790
13.5 Seasoning of Timber 792
13.6 Grading of Timber 792
13.7 Sap Wood and Heart Wood 794
13.8 Grains of Timber 794
13.9 Strength of Timber 795
13.10 Factor of Safety in Timber 796
13.11 Working Stresses in Timber 796
13.12 Bearing Stress in Timber 800
13.13 Solid Wood Columns 803
13.14 Built-up and Box Columns 804
13.15 Spaced Columns 805
13.16 Columns Subjected to Combined Stress 807
13.17 Tension Members 815
13.18 Wooden Beams 816
13.18.1 End Bearing of Beams 820
13.18.2 Check for Deflection 820
 INDEX
CONTENTS xxixxi

13.19 Framed Joints 828

13.20 Lap, Scarf and Fish Plate Joints 829
13.21 Bolted Joints in Timber 831
13.21.1 Gauge of Bolts 835
13.21.2 End Distance 835
13.21.3 Edge Distance 835
13.22 Nailed Joints in Timber 841
13.23 Disc Dowelled Joints in Timber 853
13.24 Metal Connectors in Timber 858
13.24.1 Split Rings 858
13.24.2 Toothed Ring 860
13.24.3 Bulldog Plates 861
13.24.4 Shear Plate Connectors 861
13.24.5 Claw-Plate Connectors 862
13.24.6 Spike Grids 862
13.25 Timber Roof Trusses 863
Problems 868

PART 3 DESIGN OF MASONRY STRUCTURES

CHAPTER 14 DESIGN OF MASONRY STRUCTURES 875–895

14.1 Introduction 875
14.2 General Condition of Stability of Masonry Structures 876
14.3 Earth Pressure 878
14.4 Lateral Earth Pressure on Retaining Walls with Dry and Moist
Backfill with No Surcharge 879
14.5 Lateral Earth Pressure on Retaining
Walls with Submerged Backfill 881
14.6 Lateral Earth Pressure on Retaining
Walls with Uniform Surcharge 881
14.7 Lateral Earth Pressure on Retaining Walls with
Backfill with Sloping Surface 882
14.8 Lateral Earth Pressure on Retaining Wall with
Inclined Back and Surcharge 883
14.9 Hydrostatic Pressure of Water 885
14.10 Masonry Dams 888
14.10.1 Width of Base of Trapezoidal Dam 891
Problems 894

CHAPTER 15 DESIGN OF MASONRY CHIMNEYS 896–905

15.1 Introduction 896
xxii
xxii Index
CONTENTS

15.2 Lining For Masonry Chimneys 896

15.3 Various Forces Acting on Masonry Chimneys 897
15.3.1 Self-weight of the Masonry Chimney 897
15.3.2 Weight of Lining 898
15.3.3 Wind Pressure 898
15.3.4 Seismic Forces 899
15.4 Stability of Masonry Chimneys 900
15.5 Architectural Treatment of Masonry Chimneys 903
15.6 Specifications for Design of Masonry Chimneys 903
Problem 905
Index 907
 IS CODES
Useful IS codes used in the Text of the book
1. 226–1975 Structural Steel (Standard quality)
2. 227 Galvanised (Plain and Corrugated) sheets
3. 459 Unreinforced Corrugated Asbestos Cement Sheets
4. 723 Mild Steel Wire Nails
5. 800–1984 Code of Practice for General Construction in Steel
6. 806 Use of Steel Tubes in General Building Construction
7. 808 Rolled Steel Beams, Channels and Angle Sections
8. 812–1957 Glossary of Terms Relating to Welding and Cutting Metals
9. 813–1961 Scheme for Symbols of Welding
10. 816–1969 Use of Metal Arc Welding for General Construction in Mild
Steel
11. 819–1957 Resistance Spot Welding for Light Assemblies in Mild Steel
12. 875–1984 Loading Standard for Structural Safety of Buildings
13. 961–1975 Structural Steel (High Tensile)
14. 1148–1973 Rivets Bars for Structural Purposes
15. 1149–1982 High Tensile Rivet Bars for Structural Purposes
16. 1161 Steel Tubes for Structural Purposes
17. 1173 Rolled Steel Sections, Tee-Bars
18. 1252 Rolled Steel Sections, Bulb-Angles
19. 1261–1959 Seam Welding in Mild Steel
20. 1323–1962 Code of Practice for Oxy-Acetylene Welding for Structural
Works in Mild Steel
21. 1730 Dimensions for Steel Plates, Sheets and Strips for
Structural and General Engineering Purposes
22. 1731 Dimensions for Steel Flats for Structural General
Engineering Purposes
23. 1732 Dimensions for Round and Square Steel Bars for
Structural and General Engineering Purposes
24. 1911 Schedule of Unit Weight of Materials
25. 1977–1975 Structural Steel (Ordinary quality)
26. 2062–1984 Weldable Structural Steel
27. 2585 Black Square Bolts and Nuts and Black Square Screws
28. 3139 Dimensions for Screw Threads for Bolts and Nuts
29. 3757–1972 High Tensile Friction Grip Fasteners for Structural
Engineering Purposes
30. 4000–1967 Assembly for Structural Joints using High Tensile Friction
Grip Fasteners
31. 456–1978 Code of Practice for Plain and Reinforced Concrete (Third
Revision)
32. 6623–1972 High Tensile Friction Grip Nuts
33. 6639–1972 Hexagon Bolts for Steel Structures
34. 6649–1972 High Tensile Friction Grip Washers
 1. General

2. Design of Rivited, Bolted and Pin Connections

3. Design of Columns and Compression Members

4. Design of Column Bases and Column Footings

5. Design of Tension Members

6. Design of Beams

7. Design of Beam Connections

8. Design of Plate Girder

9. Design of Roof Trusses

10. Design of Welded Joints

11. Welded Beam Connections

12. Design of Round Tubular Structures

 Chapter

1 General

1.1 INTRODUCTION
The Structural Engineering is that branch of engineering which deals with
Structural Analysis and Structural Design. The structural engineering covers
a wide range. The structural engineering plays an important and necessary
part in mechanical engineering, electrical engineering, civil engineering, naval
engineering, aeronautical engineering and in all the specialized phases of
engineering. The Structural Analysis deals with the development of suitable
arrangement of structural elements for the structures to support the external
loads or the various critical combinations of the loads which are likely to act on
the structure. The analysis also deals with the determination of internal forces
in the various members (viz., axial forces, bending moments and shear forces),
state of stresses or critical combination of the stresses at the various points
(which includes the nature, magnitude and direction of these stresses) and the
external reactions due to the worst possible combination of the loads. The
external reactions are transmitted to the foundation. The methods of structural
analysis and the principles involved in them remain independent of the materials
used for all types of structures, whether the structures are built of plastics,
aluminium, timber, reinforced concrete or steel.
The Structural Design deals with the selection of proper material, proper
sizes, proportions and shape of each member and its connecting details. The
selection is such that it is economical and safe. It satisfies all the stress
requirements imposed by the most severe combination of the loads to which the
structure is required to transmit or resist including its self-weight. The structural
design further deals with the preparation of final layout of the structure. The
finished design drawings with all pertinent or controlling dimensions for all the
members, parts and connections are made. The finished design drawings are
necessary for fabrication and construction.
4 DESIGN OF STEEL STRUCTURES

The structural design in a limited sense also deals with the design of various
parts or members of a structure. The structural design is governed with the
standard specifications. The hand-books are used as working tools in the
structural design. When the structural steel is used as the material for the
structure, the structural design is known as design of steel structures.
The steel structures and the structural elements are analysed and designed
by elastic method (also termed as allowable-stress method or working stress
method), plastic method (also called as ultimate load method or load-and-
resistance factor method), and the methods based on experimental investigations.
The elastic method is a conventional method and it is in use since long. The
stress–strain relationship for the material is assumed to be linear and the
deflections are assumed to be small in this method. The equilibrium equations
between the loads applied to the structures and the internal forces of the
members meeting at a joint are formulated on the basis of the given geometry of
the structure. The load-deflection relation for the structure is also considered
linear. In the elastic method, a member is selected so that under critical
combinations of the service loads, the calculated maximum stress, based on the
elastic analysis shall not exceed a specified permissible stress. The permissible
stress in the material is calculated by dividing the yield stress by factor of safety.
In general building construction, all the steel structures and the structural
steel components are designed as per IS : 800–1984 Code of Practice for General
Construction in Steel. The provisions of the code are thoroughly followed,
intelligently interpreted and carefully applied.

1.2 STRUCTURES
When any body is subjected to a system of loads and deformation takes place,
and the resistance is set up against the deformation, then, the body is known as
structure. The structures are means of transferring forces and moments. The
structures may be classified as statically determinate structures and statically
indeterminate structures. When the equations of statics (i.e., the equations of
statical equilibrium) are enough to determine all the forces acting on the
structures, and in the structures, then, the structures are known as statically
determinate structures. When the equations of statical equilibrium are not
sufficient to determine all forces acting on the structures, and in the structures,
then the structures are known as statically indeterminate structures. The
equations of consistent deformations are added to the equations of equilibrium
in order to analyse the statically indeterminate structures.

1.3 CLASSIFICATION OF STRUCTURES

The structures are characterized by their supporting systems. There are one
dimensional, two dimensional and three dimensional supporting systems (i.e.,
line, surface and three dimensional structures). When a supporting system is
subjected to only one type of stresses, then, it is known as basic system. When a
system is subjected to simultaneously several types of stresses, then it is known
as mixed system.
 GENERAL 5

1.3.1 One Dimensional Supporting System

This system can also be termed as line supporting structures. The line supporting
structure is large in one dimension and small in the other two dimensions. The
material remains concentrated along a straight, curved or angular line. A flexible
cable is an example for such a line supporting structure.

1.3.2 Two Dimensional Supporting System

This system can also be called as surface structure. The surface structure is
large in two dimensions and small in third dimension. The material remains
distributed along a surface. The surface may be plane or curved. The curved
surface may possess single or double curvature. A surface structure may also
consist of line elements that form a continuous structure. The frame structures
are examples of surface structures made of line elements. The frame structures
are built by assemblies of elongated members. The shell structures are rigid
curved surface structures. The shell structures are built of sheets or plates. The
shell roof coverings for large buildings, air planes, rail road cars, ship bulls,
tanks etc. are the examples of shell structures.

1.3.3 Three Dimensional Structures

These are large in three dimensions and structures may have any shape. The
three dimensional structures are also called as space structures. The three
dimensional frame structures are also space structures. :

The structures may be further classified depending on the materials used as

plastic structures, aluminium structures, timber structures, reinforced concrete
structures and steel structures. The structural steel is used for the steel
structures. The high strength and ductility are the properties of particular
importance for the structural use of the steel.
The ductility is the unique property of steel by virtue of which it is able to
deform substantially either in tension or compression before failure.
The structure can also be divided into the following three categories :
1. Skeletal structures
2. Stressed skin structures
3. Solid structures.
The skeletal structures are those structures in which the member may be
represented by a line and the intersections of lines correspond to joints of the
structures.

1.4 CLASSIFICATION OF STRUCTURAL MEMBERS

A structural member is also classified as one-dimensional, two dimensional and
three-dimensional element. Actually a structural member has always three
dimensions. But, in case, the width and thickness are small in comparison with
length, then, such members are classified as one dimensional elements.
6 DESIGN OF STEEL STRUCTURES

In case, the thickness is small in comparison with length and width, then
such members are classified as two dimensional elements. Hence, the plates
are considered as two dimensional elements.
As far as the ratios between length, width and thickness are concerned there
is no clear cut limit to them in accordance with which members are classified as
one-two- or three-dimensional elements.

1.5 ADVANTAGES AND DISADVANTAGES OF STEEL MEMBERS

The steel is a versatile material for structural and other uses. It has many
advantages.
The advantages of steel members are as follows:
1. The steel members have high strength. Therefore, the steel members
can resist high loads with comparatively light weight and small size of
members. The steel members can be conveniently handled and
transported because of their small size.
2. The steel members are gas and water-tight, because of high density of
steel.
3. The steel members have long service life. This is because of high and
homogeneous strength and density properties of steel.
4. The steel members can be used as pre-fabricated members, because of
ease of handling, fabrication and erection.
5. The steel members can be readily disassembled or replaced.
6. The existing steel structures and structural components may be
strengthened by connecting additional sections or plates.
7. The steel structures may be inspected quickly and conveniently.
The disadvantages of steel members are as follows:
(a) The steel members are susceptible to corrosion. The corrosion
necessitates their painting or the use of other methods of their
protection.
(b) The steel members are costly.

1.6 STRUCTURAL STEEL

The structural steel is the steel used for the manufacture of rolled structural
steel sections, fastenings and other elements for use in structural steel works.
This material steel is an alloy of iron and carbon (small percentage) and other
elements in varying percentages. The strength, hardness and brittleness of steel
increase and ductility of steel decreases with the increase of percentage of carbon.
Depending on the chemical composition, the different type of steels are classified
as mild steel, medium carbon steel, high carbon steel, low alloy steel and high
alloy steel. The mild steel, medium carbon steel and low alloy steel are generally
used for steel structures. The copper bearing quality of steel contains small
percentage of copper contents. The corrosive resistance of such steel is increased.
 GENERAL 7

1.6.1 Mild Steel

The mild steel is used for the manufacture of rolled structural steel sections,
rivets and bolts. Following operations can be done easily on mild steel :
1. Cutting
2. Punching
3. Drilling
4. Machining
5. Welding and
6. Forging when heated.
The mild steel cannot be used for manufacture of cutting tools.
All structural steels used in general construction, coming within the purview
of IS : 800–1984 shall, before fabrication, comply with one of the following Indian
Standard Specifications.
1. IS : 226–1975 structural steel (standard quality)
2. IS : 1977–1975 structural steel (ordinary quality)
3. IS : 2062–1984 weldable structural steel
4. IS : 961–1975 structural steel (high tensile)
5. IS : 8500–1977 weldable structural steel (medium and high strength
qualities)
Table 1.1 gives chemical composition of structural steel IS : 226–1975 as per
ISI.

Table 1.1 Chemical composition

Constituents Percent (Max.)

Carbon (For thickness/diameter 20 mm and below) 0.23
Carbon (For thickness/diameter over 20 mm) 0.25
Sulphur 0.055
Phosphorus 0.055
1. IS : 226–1975 structural steel (standard quality). The mild steel is
designated as St 44-S for use in structural work. This steel is also available in
copper bearing quality in which case it is designated as St 44-SC. The copper
content is between 0.20 and 0.35 per cent. All designs and examples have been
worked out for the steel, conforming to IS : 226–1975.
The physical properties of structural steel are given as under:
(a) Unit weight of steel 78.430 to 79.000 kN/m3
(b) Young’s modulus of elasticity, E = 2.04 to 2.18 × 105 N/mm2
(c) Modulus of rigidity, G = 0.84 to 0.98 × 105 N/mm2
(d) Coefficient of thermal α = 12 × 10–6/°C
expansion (or contraction), α = 6.7 × 10–6/°F.
The tensile strength, yield stress and percentage elongations for IS : 226–
1975 structural steel standard quality, determined in accordance with IS : 1608–
1960. Method for tensile testing of steel products other than sheets, strips, wires
and tubes are given in Table 1.2 as per IS : 226–1975.
8 DESIGN OF STEEL STRUCTURES

Table 1.2 Mechanical properties

Class of steel Nominal thickness Tensile Yield Percentage

product diameter strength stress min. elongation
mm N/mm2 N/mm2 min.
Plates, Sections 6 upto and
e.g., angles, tees, including 20 410–530 260 23
beams, channels
etc., and flats over 20 upto and
including 40 410–530 240 23
over 40 410–530 230 23
Bars (Round, 10 upto and
squares and including 20 410–530 250 23
hexagonals) over 20 410–530 240 23
The steel conforming to IS : 226 is suitable for all types of steel structures
subjected to static, dynamic and repeated cycles of loadings. It is also suitable
for welding upto 20 mm thickness. When the thickness of elements is more than
20 mm, it needs special precautions while welding.
2. IS : 1977–1975 structural steel (ordinary quality). The steel which
did not comply with IS : 226, was formerly called as steel of untested quality.
The standards for such steel have been laid down in IS : 1977–1975 (Ordinary
quality). There are two grades in this standard which are designated as St 44.0
and St 32.0. The steel St 44.0 is intended to be used for structures not subjected
to dynamic loading other than wind loads e.g., platform roofs, office buildings,
foot overbridge. The copper bearing quality is designated as St 44.OC.
The steel conforming to IS : 1977 is not suitable for welding and for the
structures subjected to high seismic forces. The steel structures using steel
conforming to IS : 1977 must not be analysed and designed by plastic theory.
3. IS : 2062–1984 weldable structural steel. This structural steel is
intended to be used for members in structures subjected to dynamic loading
where welding is employed for fabrication and where fatigue (wide fluctuation
of stress, reversal of stress), and great restraint are involved e.g., crane gantry
girder, road and rail bridges etc. It is designated as St 42–W and copper bearing
quality is designated as St 42–WC.
It is suitable for welding the elements of thickness between 28 mm and 50
mm. When the thickness of elements is less than 28 mm, it may be welded
provided the limiting maximum carbon content is 0.22 per cent.

1.6.2 High Tensile Steel

The high tensile steel forms a specific class of steel in which enhanced mechanical
properties and in most of the cases increased resistance to atmospheric corrosion
are obtained by the incorporation of low proportions of one or more alloying
elements, besides carbon. These steels are generally intended for application
where saving in weight can be effected by reason of their greater strength and
atmospheric corrosion resistance. Standards of high tensile steel have been given
 GENERAL 9

in IS : 961–1975. It has been classified into two grades designated as St 58–HT

and St 55–HTW. St 58–HT is intended for use in structures where fabrication is
done by methods other than welding. St 55–HTW is intended for use in structures
where welding is employed for fabrication. The high tensile steel is also available
in copper bearing quality and two grades are designated as St 58–HTC and St
55–HTWC.
4. IS : 961–1975 (High tensile steel). Table 1.3 (a) gives chemical composition
of structural steel as per BIS :
Table 1.3 (a) Chemical composition

Steel Designation
Constituents
St 58–HT St 55–HTW
Carbon 0.27% max. 0.20% max.
Sulphur 0.055% max. 0.055% max.
Phosphorus 0.055% max 0.055% max.
The tensile strength, yield stress and percentage elongation for IS : 961–
1975 high tensile steel, determined in accordance with IS : 1608–1960 method
for tensile testing of steel products other than sheets, strips, wires and tubes
are given in Table 1.3 (b) as per IS : 961–1975.
The steel conforming to IS : 961 is suitable for bridges and general building
construction.

Table 1.3 (b) Mechanical properties

Class of steel Nominal thickness Tensile Yield stress Percentage

products diameter strength min. min. elongation
(mm) (N/mm2) (N/mm2) min.
Structural steel
(St. 58–HT)
For Plates, 6 upto and includ–
Sections, in 28 570 350 20
Flats and Over 28 upto and
Bars including 45 570 340 20
Over 45 upto and
including 63 570 320 20
Over 63 540 290 20
Structural steel
(St. 55–HTW)
For Plates, 6 upto and includ-
Sections ing 16 540 350 20
Flats and Over 16 upto and
Bars including 32 540 340 20
Over 32 upto and
including 63 510 330 20
over 63 490 280 20
10 DESIGN OF STEEL STRUCTURES

The tensile and yield properties of steel may be increased by using increasing
contents of carbon and manganese, but its ductility will be reduced. The welding
is also difficult for such steel. The increased sulphur and phosphorus beyond
0.06 percent will give brittleness and affect weldability and fatigue strength.
The chrome and nickel (if added) give corrosive resistance and resistance to
high temperature.
The tensile and yield strengths are also increased and the ductility is
maintained by other alloying elements.
The tensile strength, yield strength, ductility and weldability are the
properties of steel which depend on the chemical composition of steel. The fatigue
strength of steel members depends on their shape and surface characteristics.
These five properties are the important properties of steel and while selecting
the material, the effect, of different alloys must also be borne in mind.
Any structural steel other than those specified above may also be used provided
that the permissible stresses and other design provisions are suitably modified
and the steel is also suitable for the type of fabrication adopted.
5. IS: 8500–1977 Weldable structural steel (medium and high
strength qualities). Various medium and high strength qualities of weldable
structural steel i.e., Fe 440 (HT1 and HT2) Fe 540 (HT and HTA and HTB), Fe
570 HT, Fe 590 HT and Fe 640 HT have been developed. Table 1.4 (a) gives the
chemical composition of various qualities of IS: 8500–1977 steel.
Table 1.4 (a) Chemical composition

Constituents Steel Designation

Fe 440 HT Fe 490 HT Fe 590 HT
Carbon 0.25 0.25 0.25
Sulphur 0.055 0.040 0.050
Phosphorus 0.055 0.040 0.050
Manganese 1.5 1.5 1.5
Table 1.4 (b) gives the mechanical properties of various medium and high
strength qualities of weldable structural steel IS: 8500–1977.
Table 1.4 (b) Mechanical properties (IS : 8500–1977)

Class of steel Nominal thickness Tensile Yield stress Percentage

products diameter strength min. min. elongation
(mm) (N/mm2) (N/mm2) min.
Fe 440 HT1
Plates, sections, Upto–6 440–560 300 –
angles Over 6 440–560 300 –
beams Upto 20
Fe 440 HT2
channels Over 20 440–560 290 –
etc. Upto 40
Contd.
 GENERAL 11

Table 1.4 (b) Contd.

Class of steel Nominal thickness Tensile Yield stress Percentage
products diameter strength min. min. elongation
(mm) (N/mm2) (N/mm2) min.
bars, flats Over 40 440–560 280 –
upto 63
Fe 540 HT
Plates, Upto 6 540–660 410 20–25
sections
Fe 540 HTA
angles, Over 6 20–25
beams, Upto 20 540–660 400 –
channels . Over 20 540–660 390 20–25
etc. Upto 40
Fe 540 HTB Over 40 540–660 380 –
bars, flats Upto 63
Fe 570 HT
Plates, Upto 6 570–720 450 20–25
sections Over 6
angles Upto 20 570–720 440 20–25
beams Over 20 570–720 430 20–25
upto 40
channels Over 40 570–720 420 20–25
etc.
bars, flats Upto 63
Fe 590 HT Upto 6 590–740 490 20–25
Over 6 590–740 480 20–25
Upto 20
Fe 640 HT Upto 6 640–790 540
over 6 Upto 20 640–790 530 20–25

1.7 PRODUCTION OF STEEL

The steel is produced in the form of ingots and converted to different shapes in
number of plants in our country. Tata Iron and Steel Company, Indian Iron and
Steel Company, Mysore Iron and Steel Company and Hindustan Steel produce
steel at their plants. Hindustan Steel has its plants at Durgapur, Rourkela and
Bhilai. The capacity of each plant is in some million tonnes.

1.8 RECENT DEVELOPMENTS IN MATERIAL

A number of developments in material such as steel have been made recently.
The weldable qualities of steel (IS : 2062) designated as St 42–W) and (IS : 961
designated as St–55–HTW) are developed with the large scale use of welding.
12 DESIGN OF STEEL STRUCTURES

IS : 961 has been developed with high tensile strength and there is saving in
weight due to enhanced mechanical properties. Its weldable quality is
advantageous for composite construction.

1.9 ROLLED STRUCTURAL STEEL SECTIONS

The steel sections manufactured in rolling mills and used as structural members
are known as rolled structural steel sections. The steel sections are named
according to their cross-sectional shapes. Many steel sections are readily available
in the market and have frequent demand. Such steel sections are known as
regular steel sections. Some steel sections are rarely used. The manufacture of
such steel sections is not usual. Such sections are produced on special requisition
and are known as special sections.
‘ISI Handbook for Structural Engineers I’ gives nominal dimensions, weight
and geometrical properties of various rolled structural steel sections. This
handbook also gives other additional data required by the designers and
architects.
The various types of rolled structural steel sections manufactured and used
as structural members are given below :
1. Rolled Steel I Sections (Beam sections)
2. Rolled Steel Channel Sections
3. Rolled Steel Tee Sections
4. Rolled Steel Angle Sections
5. Rolled Steel Bars
6. Rolled Steel Tubes
7. Rolled Steel Flats
8. Rolled Steel Sheets and Strips
9. Rolled Steel Plates.

1.10 ROLLED STEEL BEAM SECTIONS

The rolled steel beams are classified into the following four series as per BIS :
(IS: 808–1989)
1. Indian Standard Junior Beams ... ISJB
2. Indian Standard Light Beams ... ISLB
3. Indian Standard Medium Weight Beams ... ISMB
4. Indian Standard Wide Flange Beams ... ISWB
The rolled steel columns/heavy weight beams are classified into the following
two series as per BIS (IS: 808–1989)
1. Indian Standard Column Sections ... ISSC
2. Indian Standard Heavy Weight Beams ... ISHB
The cross-section of a rolled steel beam has been shown in Fig. 1.1. The beam
section consists of web and two flanges. The junction between the flange and
the web is known as fillet. These hot rolled steel beam sections have sloping
flanges. The outer and inner faces are inclined to each other and they intersect
 GENERAL 13

1
at an angle varying from 1 to 8° depending on the section and rolling mill
2
practice. The angle of intersection of ISMB section is 8°.
Abbreviated reference symbols (JB, LB, MB, WB, SC and HB) have been
used in designating the Indian Standard Sections as per BIS (IS 808–1989).

1.10.1 Designation
The rolled steel beams are designated by the series to which beam sections
belong, (abbreviated reference symbols) followed by depth (mm) of the section
and weight (kN) per metre length of the beam, e.g., MB 225, @ 0312 kN/metre.
The weight per metre length of beam section should always be specified as in
Indian Standard wide flange beams and Indian Standard column sections–H-
Beams sections of equal depths have different weights per metre length and
also different properties e.g.,
WB 600, @ 1.340 kN/m
WB 600, @ 1.450 kN/m
HB 350, @ 0.674 kN/m
HB 350, @ 0.724 kN/m

b = W idth

Flan ge
h = D epth

W eb

Fillet

Fig. 1.1 Beam section

The rolled steel beams are mainly employed to resist bending. The rolled
steel beams are used as independent sections to resist axial forces (compression
or tension). These are also used in the built up sections of columns.
An efficient and a rational series of hot rolled beam and column sections for
use in structures has been developed first time by our country. Out of the five
above I-sections, only MB sections are being produced. Because of the rolling
mill limitations, other sections have not yet been produced. Many countries are
installing universal mills, which are special rolling mills and perfected to roll
parallel flange I–sections as compared to conventional sloping flange sections.
14 DESIGN OF STEEL STRUCTURES

The beams of parallel flange sections have higher efficiency (represented by

ratio of section modulus to wt.) and are easier to fabricate than the conventional
sloping flange sections.

1.11 ROLLED STEEL CHANNEL SECTIONS

The rolled steel channel sections are classified in the following four series as
per ISI :
1. Indian Standard Junior Channels ... ISJC
2. Indian Standard Light Channels ... ISLC
3. Indian Standard Medium Weight
Channels ... ISMC
4. Indian Standard Medium Weight Parallel
Flange Channels ... ISMCP
The cross-section of rolled steel channel section has been shown in Fig. 1.2.
The channel section consists of web and two flanges. The junction between the
flange and the web is known as fillet.

1.11.1 Designation
The rolled steel channels are designated by the series to which channel sections
belong, (abbreviated reference symbols) followed by depth (mm) of the section
and weight (kN) per metre length of the channel, e.g., MC225, 0.261 kN/m @
0.259 kN/m.

b = W idth

Flan ge
h = D epth

W eb

Fillet

Fig. 1.2 Channel section

The channels are employed as elements to resist bending (e.g., as purlins in

the rolling of industrial buildings). It is to note that they are subjected to twisting
or torsion because of absence of symmetry of the section with regards to the axis
parallel to the web (i.e. yy-axis). Therefore, it is subjected to additional stresses.
The channel sections are commonly used as members subjected to axial
 GENERAL 15

compression in the shape of built-up sections of two channels connected by

lattices, or batten plates or perforated cover plates. The built-up channel sections
are also used to resist axial tension in the form of chords of truss girders.
Note : As per IS: 808–1989, following channel sections have also been additionally
adopted as Indian Standard Channel Sections
1. Indian Standard Light Channels with parallel flanges. ... ISLC(P)
2. Medium weight channels ... MC
3. Medium weight channels with parallel flanges ... MCP
4. Indian Standard Gate Channels ... ISPG
In MC and MCP channel sections, some heavier sections have been developed
for their intended use in wagon building industry. The method of designating
MC and MCP channels is also same as that for IS–channels described above.

1.12 ROLLED STEEL TEE SECTIONS

The rolled steel tee sections are classified into the following five series as per
ISI :
1. Indian Standard Normal Tee Bars ... ISNT
2. Indian Standard Wide Flange Tee Bars ... ISHT
3. Indian Standard Long Legged Tee Bars ... ISST
4. Indian Standard Light Tee Bars ... ISLT
5. Indian S tandard Junior Tee Bars ... ISJT
The cross-section of a rolled steel tee section has been shown in Fig. 1.3. The
tee section consists of web and flange. The junction between the flange and the
web is known as fillet.

b = W idth

Flan ge
Fillet

W eb

Fig. 1.3 Tee-section

1.12.1 Designation
The rolled steel tee sections are designated by the series to which the sections
belong, (abbreviated reference symbols) followed by depth (mm) of the section
and weight (kN) per metre length of the Tee e.g., HT 125, @ 0.274 kN/m.
16 DESIGN OF STEEL STRUCTURES

The tee sections are used to transmit bracket loads to the columns. These are
also used with flat strips to connect plates in the steel rectangular tanks.
Note : As per IS: 808–1984, following T-sections have also been additionally
adopted as Indian Standard T-sections.
1. Indian Standard deep legged Tee bars ... ISDT
2. Indian Standard Slit medium weight Tee bars ... ISMT
3. Indian Standard Slit Tee bars from I-sections ... ISHT
It is to note that as per IS 808 (Part II) 1978, H beam sections have been
deleted.

1.13 ROLLED STEEL ANGLE SECTIONS

The rolled steel angle sections are classified into the following three series.
1. Indian Standard Equal Angles ... ISA
2. Indian Standard Unequal Angles ... ISA
3. Indian Standard Bulb Angles ... ISBA
The cross-section of a rolled equal angle section has been shown in Fig. 1.4 (a),
unequal angle section in Fig. 1.4 (b) and that of bulb angle section in Fig. 1.4 (c).

W eb
h d ep th

L L1
Flan ge

L L2 b = W idth
L = L en gth o f leg L = L en gth o f leg
(a ) E q ua l an gle (b ) U n eq ua l a n gle (c) B u lb a ng le

Fig. 1.4 Angle sections

The lengths of the legs in case of equal sections are equal and in case of unequal
angle section, length of one leg is longer than the other. The thickness of legs of
equal and unequal angle sections are equal. The bulb angle shown in Fig. 1.4 (c)
consists of web flange and a bulb projecting from end of web. The thickness of
web of bulb angle may or may not be equal to the thickness of flange.

1.13.1 Designation
The rolled steel equal and unequal angle sections are designated by abbreviated
reference symbols (∠) followed by lengths (mm) and thickness of legs e.g.,
∠ 130 130 × 8 mm (∠130 130, @ 0.159 kN/m)
∠ 200 100 × 10 mm (∠ 200 100, @ 0.228 kN/m)
The rolled steel bulb angles are designated by BA, followed by depth (mm) of
the section and weight (kN) per metre length of bulb angle.
 GENERAL 17

The angle sections are used as independent sections consisting of one or two
or four angles designed for resisting axial forces (tension and compression) and
transverse forces as purlins. The angle sections are used as connecting elements
to join together sheets or plates or to form a built-up section. The angle sections
are also used as construction elements for connecting beams to the columns,
and purlins to the chords of trusses in the capacity of beam seats, stiffening ribs
and cleat angles. The bulb angles are used in the ship buildings. The bulb helps
to stiffen the outstanding leg when the angle is under compression.
Note : As per IS: 808–1984, some supplementary angle sections have also
additionally adopted as Indian standard angle sections. However prefix ISA has
been droped. These sections are designated by the size of legs followed by thickness
(e.g ∠ 200 150 × 15).

1.14 ROLLED STEEL BARS

The rolled steel bars are classified into the following two series :
1. Indian Standard Round Bars ... ISRO
2. Indian Standard Square Bars ... ISSQ
S id e w idth

D ia m e te r

(a ) S q ua re ba r (b ) R o un d ba r
S tee l b ar se ctio ns

Fig. 15 Steel bars

The rolled steel bars are used as ties and lateral bracing.

1.14.1 Designation
The rolled steel bars are designated by abbreviated reference symbol RO followed
by diameter in case of round bars, and ISSQ followed by side width of bar sections.
The bars threaded at the ends or looped at the ends are used as tension
members.

1.15 ROLLED STEEL TUBES

The rolled steel tubes are used as columns und compression members and tension
members in tubular trusses. The rolled steel tubes are efficient structural sections
to be used as compression members. The steel tube sections have equal radius
of gyration in all directions.
18 DESIGN OF STEEL STRUCTURES

S te el tu be s se ctio n
Fig.1.6 Steel tubes

1.16 ROLLED STEEL FLATS

The rolled steel fiats are used for lacing of elements in built-up members, such
as columns and are also used as ties.

W idth

Th ickne ss

S tee l fla ts se ctio n

Fig. 1.7 Steel flats

1.16.1 Designation
The rolled steel flats are designated by width (mm) of the section followed by
letters (abbreviated reference symbol) F, and thickness in (mm), e.g., 50 F 8.
This means a flat of width 50 mm and thickness 8 mm.
The rolled steel flats are used as lattice bars for lacing the elements of built-
up columns. The rolled steel flats are also used as tension members and stays.

1.17 ROLLED STEEL SHEETS AND STRIPS

1.17.1 Designation
The rolled steel sheet is designated abbreviated reference symbol SH followed
by length (mm) × width (mm) × thickness (mm) of sheet. The rolled steel strip is
designated as ISST followed by width (mm) × thickness (mm) e.g.,
SH 2,000 mm × 600 mm × 8 mm
ST 250 mm × 2 mm.

1.18 ROLLED STEEL PLATES

1.18.1 Designation
The rolled steel plates are designated by abbreviated reference symbol PL
followed by length (mm) × width (mm) × thickness (mm) of the plates e.g., PL
2,000 mm × 1,000 mm × 6 mm.
 GENERAL 19

The rolled steel sheets and plates are widely used in construction. Any sections
of the required dimensions, thickness and configuration may be produced by
riveting or welding the separate plates. The rolled plates are used in the web
and the flanges of plate girders, plated beams and chord members and web
members of the truss bridge girders. The rolled steel plates are used in special
plate structures e.g., shells, rectangular and circular steel tanks and steel
chimneys.

1.19 RECENT DEVELOPMENTS IN SECTIONS

The rolled steel beam sections with parallel faces of the flanges are recently
developed. These beam sections are called as parallel flange sections. These
sections have increased moment of inertia, section modulus and radius of gyration
about the weak axis. Such sections used as beams and columns have more
stability. These sections posses ease of connections to other sections as no packing
is needed as in beams of sloping flanges.
The parallel flange beam sections are not yet rolled in our country.
New welded sections using plates and other steel sections are developed
because of welding. The development of beams with tapered flanges and tapered
depths is also due to welding. The open web sections and the castellated beams
(Fig. 6.2) were also developed with the rapid use of welding.

1.20 SPECIFICATION
The construction of steel structures involves four parties, viz., the owners, the
designers, (engineers and architects), the fabricators and the constructors. The
specifications are prepared by the designers. The acceptable standard of quality
of materials and of workmanship in fabrication are defined by the specifications.
The designers prepare three types of specifications, viz., project specifications,
material standards and design codes. The object of drafting specifications and
preparing codes is to achieve economy in the use of material and the cost of
construction without sacrificing the safety. The project specifications describe
complete information about owner’s and the engineer’s precise requirements
for the completed structures. The material Standards describe the quality of
materials and their economic use. The design codes incorporate:
1. General guidance as regards various loads.
2. Allowable stresses.
3. Design principles and procedures generally accepted throughout the
country.
For complex and time consuming formulae and provisions of designs, the
tables are included in the code, which reduce the design calculations. The various
professional associations, institutes and government agencies prepare codes and
specifications.
In America, the American Institute of Steel Construction specifications, briefly
known as AISC specifications (Specifications for the Design, Fabrication and
Erection of Structural Steel Buildings) are followed by steel buildings. For the
special aspect of design, the following specifications are followed:
20 DESIGN OF STEEL STRUCTURES

AISI Specifications : American Iron and Steel Institute Specifications

AWS Specifications : American Welding Society Specifications
SJI Specifications : Steel Joist Institute Specifications
AASHTO Specifications : American Association of State Highway and
Transportation Officials Specifications
USBPR Specifications : U.S. Bureau of Public Roads Specifications
AREA Specifications : American Railway Engineering Association
Specifications
USASI Specifications : U.S. American Standard Institute.
AASHO, USBPR and AREA specifications are for the highways and railway
bridges.
In Britain, the British Standard Institute prepares the specifications. These
specifications are known as British Standard Specifications (BSS). In Germany,
the German Standard Institution (DIN) prepares the specifications. In USSR,
USSR State Standards (GOST) prepares the specifications.
In India the following organisations have prepared and published the
specifications:
1. Bureau of Indian Standards (BIS)
2. Indian Road Congress (IRC)
3. Ministry of Railways and Railway Board.
In 1950’s, the steel economy programme was initiated in Indian Standard
Institute. The object of this programme was to achieve economy in the use of
structural steel. The rational, efficient and optimum standards for structural
steel products and their use were established. The code of practice for ‘Use of
Structural Steel in General Building Construction’ IS 800–1956 was published.
It had been first in the series of Indian Standards. After this standard was in
use for some time, the first revision, incorporating certain very important changes
was prepared and published in IS : 800–1962. Considering the comments received
on this standard during its usages and developments taking place in the country
and abroad, the necessary modifications and additions have been incorporated
in the second revision. The standard has been made more useful and IS 800–
1984 with the modified title as ‘Code of Practice for General Construction in
Steel’ has been brought out. This code is applicable to all types of steel structures.
Following IS codes are useful for the design of steel structures : IS :
1. 226–1975 Structural Steel (standard quality)
2. 227 Galvanised (Plain and Corrugated) sheets
3. 459 Unreinforced Corrugated Asbestos Cement Sheets
4. 723 Mild Steel Wire Nails
5. 800–1984 Code of Practice for General Construction in Steel
6. 806 Use of Steel Tubes in General Building Construction
7. 808 Rolled Steel Beams, Channels and Angle Sections
 GENERAL 21

8. 812–1957 Glossary of Terms Relating to Welding and Cutting Metals

9. 813–1961 Scheme for Symbols of Welding
10. 816–1969 Use of Metal Arc Welding for General Construction in Mild
Steel
11. 819–1957 Resistance Spot Welding for Light Assemblies in Mild Steel
12. 875–1984 Loading Standard for Structural Safety of Buildings
13. 961–1975 Structural Steel (High Tensile)
14. 1148–1973 Rivets Bars for Structural Purposes
15. 1149–1982 High Tensile Rivet Bars for Structural Purposes
16. 1161 Steel Tubes for Structural Purposes
17. 1173 Rolled Steel Sections, Tee-Bars
18. 1252 Rolled Steel Sections, Bulb-Angles
19. 1261–1959 Seam Welding in Mild Steel
20. 1323–1962 Code of Practice for Oxy-Acetylene Welding for Structural
Works in Mild Steel
21. 1730 Dimensions for Steel Plates, Sheets and Strips for
Structural and General Engineering Purposes
22. 1731 Dimensions for Steel Flats for Structural General
Engineering Purposes
23. 1732 Dimensions for Round and Square Steel Bars for
Structural and General Engineering Purposes
24. 1911 Schedule of Unit Weight of Materials
25. 1977–1975 Structural Steel (ordinary quality)
26. 2062–1984 Weldable Structural Steel
27. 2585 Black Square Bolts and Nuts and Black Square Screws
28. 3139 Dimensions for Screw Threads for Bolts and Nuts
29. 3757–1972 High Tensile Friction Grip Fasteners for Structural
Engineering Purposes
30. 4000–1967 Assembly for Structural Joints using High Tensile Friction
Grip Fasteners
31. 456–1978 Code of Practice for Plain and Reinforced Concrete (Third
Revision)
32. 6623–1972 High Tensile Friction Grip Nuts
33. 6639–1972 Hexagon Bolts for Steel Structures
34. 6649–1972 High Tensile Friction Grip Washers
In addition to above the following ISI : Handbooks are also useful for Design
of Steel Structures:
1. Handbook for Structural Engineers 1. (Structural Steel Section)
2. Handbook for Structural Engineers 2. (Steel Beams and Plate Girders)
22 DESIGN OF STEEL STRUCTURES

3. Handbook for Structural Engineers 3. (Steel Columns and Struts)

Following Specifications are useful for the design of timber structures :
(a) 883 Use of Structural Timber in Building (Material, Grading and Design)
(b) 2366 Nail Jointed Timber Construction.
The specifications are revised from time to time depending upon the
experience, developments and research works of research institutions and
laboratories.

1.21 LOADS AND STRESSES

The structures and structural members are designed to meet the functional
and structural aspects. Both the aspects are interrelated. A building has to
perform many functions satisfactorily.
The functional aspect takes into consideration the purpose for which the
building or the structure is designed. The functional aspect includes the
determination of the amount and arrangement of space necessary for the
structure. The structure is able to perform satisfactorily the services for which
it is designed. The functional aspect includes the determination of location and
arrangement of operating utilities of the building for the intended use and
occupancy, structural safety, fire safety and compliance with hygienic, sanitation,
ventilation, special equipment, machinery or other features, incident to the
proper functioning of the structures. The preliminary layout of the structure is
made. The preliminary layout of the structure is useful in determining the type
of structure to be used, the general size, shape and arrangement of the various
parts of the structure. The approximate dimensions of the structures are
established. These dimensions are used for the structural design.
In the structural aspect, it is ensured that the building or the structure is
structurally safe, strong, durable and economical. Many structures are built to
shelter human beings and to protect them from the forces of nature. Occupants
of the buildings or the structure should feel safe with regard to their lives and
properties. The safety of many individuals at one time depends on the structural
safety of the structure. The minimum requirements pertaining to the structural
safety of buildings are being covered in codes dealing with loads by way of laying
down minimum design loads which have to be assumed for dead loads, imposed
loads, wind loads and other external loads, the structures would be required to
bear. Unnecessarily, heavy loads without proper assessment should not be
assumed. The structures are designed between two limits, viz., the structural
safety and economy. The structures and structural members are designed to
satisfy three basic principles of design. The structures should be strong, stable
and stiff.
All the sections of the structures should have strength at least equal to the
structural effects of design loads and forces that occur during the construction
and use. All the loads should be taken into consideration which a building or a
 GENERAL 23

structure would be required to bear. A structure may be required to withstand

any loads out of the various loads or combination of the various loads. These
loads include overhead and other cranes, equipment etc., in industrial buildings,
occupancy etc. in buildings ; vehicles on bridges ; water in hydraulic structure;
loose materials for bins and bunkers and so on. For the purpose of determining
the maximum stresses in any structure or member of a structure, the various
loads which are likely to act on the structure are given below :
1. Dead load
2. Live load (Imposed load)
3. Wind load
4. Snow load
5. Seismic load
6. Temperature effects.
In addition to the above loads and effects, following forces and effects are
also considered while designing the structures if these are liable to affect
materially the safety and serviceability of the structures.
1. Foundation movements
2. Elastic axial shortening
3. Soil and fluid pressures
4. Vibrations
5. Fatigue
6. Impact (dynamic effects)
7. Erection loads
8. Stress concentration effect due to point load and the like.
According to the strength requirement, the internal actions produced in the
structural members must be equal to the structural effects caused due to designed
loads, forces and various effects.

1.22 DEAD LOAD

The dead loads (viz., the self-weight of the structural members), superimposed
dead loads and loads due to filling materials are referred as permanent loads.
Dead loads are loads which are constant in magnitude and fixed in position
throughout the life time of the structure. Dead loads in a building comprises the
self-weight of the structure and all other super-imposed loads (viz., all permanent
constructions and installations including weight of walls, partitions, floors, and
roofs). The self-weight of the structure being unknown in the beginning of the
design, it is assumed by the designer in the light of experience and thumb rules
and formula available. The super-imposed dead loads are determined for the
various materials specified for the construction and unit-weight of materials.
On completion of design, total dead load is calculated and checked with the
dead load assumed for design and the design is revised, if necessary. Table 1.5
gives the unit-weights of materials used in building construction. The dead loads
are also called as permanent loads as regards to the duration of their action.
24 DESIGN OF STEEL STRUCTURES

Table 1.5 Unit weight of building materials, building parts and components
IS : 875 (PART I) – 1987
(Extracts from Tables I and II of IS : 1911–1967, schedule of unit–weight of materials)
S.No. Materials Nominal Weight/mass
size or thickness
kN kg per
(1) (2) (3) (4) (5) (6)
Building materials
1. Asbestos natural – 29.8 3040 m3
raw 5.90 to 8.85 600 to 900 m3
2. Asbestos cement sheeting
corrugated (146 mm) 6 mm 0.118 to 12 m2
6 mm 0.130 13.3 m2
Semi-corrugated (240 mm) 6 mm 0.118 to 1.2 m2
6 mm 0.127 13 m2
plain 5 mm 0.09 9.16 m2
3. Bitumen – 0.102 10.4 m3
4. Boards (cork)
Compressed 10 mm 0.04 4 m2
Ordinary 10 mm 0.02 2 m2
5. Bricks, chips and
broken bricks
Fine – 14.2 1450 m3
Coarse – 9.9 1010 m3
6. Brick dust (Surkhi) – 9.9 1010 m3
7. Bricks (Engineering) – 21.2 2160 m3
8. Cement
Ordinary and Aluminious – 14.10 1440 m3
Rapid hardening – 12.55 1280 m3
9. Plain cement concrete
with sand and gravel or
crushed natural stone – 22 to 23.5 2240–2400 m3
10. Reinforced cement concrete
sandstone aggregate with
1 percent steel – 22.75–24.20 2310–2470 m3
2 percent steel – 23.25–24.80 2370–2530 m3
5 percent steel – 24.80–26.50 2530–2700 m3
11. Glass – 24.384 2480 m3
Glass bricks – 24.304 2480 m3
Window glass and 24.304 2480 m3
Looking glass 24.304 2480 m3

Contd.
 GENERAL 25

Tables 1.5 Contd.

(1) (2) (3) (4) (5) (6)
Sheet 2.0 mm 0.049 5.0 m3
2.5 mm 0.062 6.3 m3
3.0 mm 0.074 7.5 m3
4.0 mm 0.098 10.0 m3
5.0 mm 0.123 12.5 m3
5.5 mm 0.134 13.7 m3
6.5 mm 0.167 17.0 m3
12. Gypsum
Mortar – 11.75 1200 m3
Powder – 13.89–17.25 1410–1760 m3
13. Iron
Pig – 70.60 7200 m3
Gray cast – 68.95–69.90 7030–7130 m3
White cast – 74.30–75.70 7530–7720 m3
Wrought – 75.50 7700 m3
14. Mastic asphalt 10 mm 0.215 22 m2
15. Metal sheeting
Galvanised steel sheets
(plain and corrugated)
Casel 1.60 mm 0.131 1331 m2
1.25 mm 0.104 1050 m2
1.00 mm 0.084 860 m2
0.80 mm 0.069 703 m2
0.63 mm 0.056 570 m2
Case 2 1.60 mm 0.129 1316 m2
1.25 mm 0.102 1041 ,m2
1.00 mm 0.083 845 m2
0.80 mm 0.067 688 m2
0.63 mm 0.054 555 m2
Case 3 1.60 mm 0.128 1301 m2
1.25 mm 0.101 1026 m2
1.00 mm 0.081 830 m2
0.80 mm 0.066 673 m2
0.63 mm 0.053 540 m2
Case 4 1.60 mm 0.127 1294 m2
1.25 mm 0.100 1019 m2
1.00 mm 0.081 822 m2
0.80 mm 0.065 666 m2
0.63 mm 0.052 532 m2

Contd.
26 DESIGN OF STEEL STRUCTURES

Tables 1.5 Contd.

(1) (2) (3) (4) (5) (6)
16. Mortar
Cement – 20.40 2080 m3
Lime – 15.70–18.05 1600–1840 m3
17. Plaster
Cement – 20.40 2080 m3
Lime – 17.25 1760 m3
18. Soils and gravels
Alluvial ground (undisturbed) – 15.60 1600 m3
Clay (dry and lump) – 10.20 1040 m3
Dry, (compact) – 14.10 1440 m3
Damp, (compact) – 17.25 1760 m3
Clay (wet and compact) – 20.40 2080 m3
Undisturbed – 18.85 1920 m3
Undisturbed bed gravelly – 20.40 2080 m3
Earth
Dry – 13.85–18.05 1410–1840 m3
Moist – 15.70–19.60 1600–2000 m3
Gravel
Loose – 15.70 1600 m3
Rammed – 18.85–21.20 1920–2160 m3
19. Stone
Agate – 25.50 2600 m3
Aggregate – 15.70–18.85 1600–1920 m3
Basalt – 27.95–29.05 2850–2960 m3
Granite – 25.90–27.45 2640–2800 m3
Lime – 23.55–25.90 2400–2640 m3
Marble – 26.70 2720 m3
Sandstone – 27.45 2800 m3
20. Tiles clay
Mangalore pattern –
21. Timber
Hard wood (e.g., Babul) – 6.27–9.41 640–960
Light wood (e.g., Fir) – 3.92–4.71 400–480
Medium (e.g., Deodar) – 4.71–6.7 480–640
Building parts and components
22. Ceiling (on wire netting) 10 mm 0.265 27 m3
23. Flooring
Asphalt 10 mm 0.22 22 m2
Terrazo paving 10 mm 0.24 24 m2
Contd.
 GENERAL 27

Tables 1.5 Contd.

(1) (2) (3) (4) (5) (6)
24. Roofing
Allahabad tiles
(including battens)
Single – 0.83 85 m2
Double – 1.67 170 m2
Country tiles
(with battens)
Single – 0.69 70 m2
Double – 1.18 120 m2
Mangalore tiles with battens – 0.64 65 m2
(tiles with mortar) – 1.08 110 m2
Slates (on battens) – 0.34–0.49 35–50 m2
Glazed with aluminium
alloy bars for span
upto 3 m 64 mm 0.19 19.5 m2
Glazed with lead covered
steel bars at
0.6 m centres 6.4 mm 0.25–0.28 26–29 m2
25. Wall
Engg. bricks (in masonry) – 23.55 2400 m3
Stone (masonry)
Granite ashlar – 25.90 2640 m3
nibble – 23.55 2400 m3
Lime ashlar – 25.70 2560 m3
Marble dressed – 26.5 2700 m3
Sandstone – 22 2240 m3
The dead loads are also termed as permanent loads as regards to the duration
of their action.

1.23 LIVE LOADS

Live loads are the loads which vary in magnitude and/or in positions. Live loads
are also known as imposed or transient loads. Live loads include any external
loads imposed upon the structure when it is serving its normal purpose. Live
loads are assumed to be produced by the intended use of occupancy in buildings
including distributed, concentrated, impact and vibration and snow loads. Live
loads are expressed as uniformly distributed static loads (U.D.L.). Live loads
include the weight of materials stored, furniture and movable equipments. For
buildings in most cities, the loads imposed on floors, stairs and roofs are specified
in codes. For design purpose, certain live loads on floors have been assumed by
many countries and which are almost in agreement. Some efforts have been
made at the international level do decide live loads on floors and these have
been specified in the International Standards 2103 (Imposed floor loads in
28 DESIGN OF STEEL STRUCTURES

residential and public building and 2633 (Determination of imposed floor loads
in production buildings and warehouses). These codes have been published in
the International Organisation of Standardization.
Code IS : 875 (Part 2)–1987 defines the principal occupancy for which a
building or part of a building is used or intended to be used. The buildings are
classified according to occupancy as under as per IS : 875 (Part 2) 1987. An
occupancy shall be deemed to include subsidiary occupancies which are contingent
upon.
1. Assembly buildings. The assembly buildings including any building
or part of a building where groups of people gather together for amusement,
recreation, social, religious, patriotic, civil, travel and similar purposes (e.g.,
theatres, motion picture houses, assembly halls, city halls, marriage halls,
townhalls, auditoria, exhibition halls, museums, skating rinks, gymnasiums,
restaurants, places of worship, dance halls, club rooms, passenger stations and
terminals of air, surface and other public transportation services, recreation
piers and stadia, etc.).
2. Business buildings. The business buildings include any building or
part of a building, which is used to conduct business (other than that covered in
6) for maintaining of accounts and records for similar purpose, offices, banks,
professional establishments, court houses, and libraries shall be classified in
this group so far as principal function of these is to deal with public business
and the keeping of books and records.
The office buildings are primarily used for office purposes, (e.g., purpose of
administration, clerical work, handling money, telephone and telegraph
operating and operating computers, calculating machines). The clerical work
includes writing, book-keeping, sorting papers, typing, filing, duplicating,
punching cards or tapes, drawing of matter for publication and the editorial
preparation of matter for publication).
3. Educational buildings. The educational buildings include any building
used for school, college or day-care purposes involving assembly for instructions,
education or recreation and which is not covered in 1).
4. Industrial buildings. The industrial buildings include any building
or a part of a building or structure in which products or materials of various
kinds and properties are fabricated, assembled or processed like assembly plants,
power plants, refineries, gas plants, mills, diaries, factories, workshops, etc.
5. Institutional buildings. Institutional buildings include any building
or a part thereof, which is used for purpose, such as medical or other treatment
in case of persons suffering from physical and mental illness, disease or infirmity;
care of infants, convalescents of aged persons and for penal or correctional
detention in which the liberty of the inmates is restricted. Institutional buildings
ordinarily provide sleeping accommodation for the occupants. These include
hospitals, sanitoria, custodial institutions or panel institutions (e.g., jails, prisons
and reformations).
6. Mercantile buildings. These buildings include building or a part of a
building which is used as shops, stores, market for display and sale of
merchandise either wholesale or retail. Office, storage and service facilities
incidental to the sale of merchandise and located in the same building shall be
included under this group.
 GENERAL 29

7. Residential buildings. The residential buildings include any building

in which sleeping accommodation is provided for normal residential purposes
with or without cooking a dining or both facilities (except building covered in 5).
These buildings include one or multi-family dwellings, apartment houses (flats),
lodging or rooming houses, restaurants, hostels, dormitories and residential
hotels.
The dwellings include any building or part occupies by members of single/
multi-family units with independent cooking facilities. These also include
apartment houses (flats).
8. Storage buildings. The storage buildings include any building or
part of building used primarily for the storage or sheltering of goods, wares or
merchandize, (e.g., warehouses, cold storages, freight depots, transity sheds,
store houses, garages, hangers, truck terminals, grain elevators, barns and
stables).
The imposed loads are the loads assumed to be produced by the intended use
or occupancy of a building, including the weight of movable partitions, distributed
loads, concentrated loads, loads due to impact and vibration, and dust load but
excluding wind, seismic, snow and other loads due to temperature changes,
creep, shrinkage, differential settlement, etc. The imposed loads are the largest
loads those probably will be produced by the intended use or occupancy, but
shall not be less than the equivalent minimum loads specified in Table 1.6
subjected to any reductions allowed by code IS : 875 (Part 2)–1987.
The floors are investigated for both the uniformly distributed loads (UDL)
and the corresponding concentrated loads specified in Table 1.6 and designed
for the most adverse effects but these shall not be considered to act
simultaneously. The concentrated loads specified in Table 1.6 may be assumed
to act over an area of 0.3 m × 0.3 m. However, the concentrated loads may not be
considered where the floors are capable of effective lateral distribution of this
load.
Table 1.6 Imposed floor loads for different occupancies
[As per IS : 875 (Part 2)–1987]
Sl. Occupancy Classification Uniformly Concentrated
No. Distributed Load
Load (UDL)
(1) (2) (3) (4)
(kN/m ) 2
(kN)
(I) RESIDENTIAL BUILDINGS
(a) Dwelling houses:
(1) All rooms and kitchens 2.0 1.8
(2) Toilet and bath rooms 2.0 –
(3) Corridors, passages, staircases 3.0 4.5
including fire escapes and
store rooms
Contd.
30 DESIGN OF STEEL STRUCTURES

Table 1.6 Contd.

(1) (2) (3) (4)
(kN/m ) 2
(kN)
(4) Balconies 3.0 1.5 per metre run
concentrated at the
outer edge
(b) Dwelling units planned and
executed in accordance with
IS : 8888–1989* only :
(1) Habitable rooms, kitchens, toilet 1.5 1.4
and bathrooms
(2) Corridors, passages and staircases 1.5 1.3
including fire escapes
(3) Balconies 3.0 1.5 per metre run
concentrated at
outer edges
(c) Hotels, hostels, boarding houses,
lodging houses, dormitories, residential
clubs :
(1) Living rooms, bed rooms and 2.0 1.8
dormitories
(2) Kitchens and laundries 3.0 4.5
(3) Billiards room and public lounges 3.0 2.7
(4) Storerooms 5.0 4.5
(5) Dining rooms, cafeterias and 4.0 2.7
restaurants
(6) Office rooms 2.5 2.7
(7) Rooms for indoor games 3.0 1.8
(8) Baths and toilets 2.0 –
(9) Corridors, passages, staircases 3.0 4.5
including fire escapes, lobbies –
as per the floor serviced (excluding
stores and the like) but not less than
(10) Balconies Same as rooms to 1.5 per metre run
which they give access concentrated at
but with a minimum the outer edge
of 4.0

Contd.
 GENERAL 31

Table 1.6 Contd.

(1) (2) (3) (4)
(kN/m2) (kN)
(d) Boiler rooms and plant room – to be 5.0 6.7
calculated but not less than
(e) Garages :
(1) Garage floors (including parking 2.5 9.0
area and repair workshop) for
passenger cars and vehicles not
exceeding 25 kN gross weight,
including access ways and ramps
– to be calculated but not less than
(2) Garage floors for vehicles not 5.0 9.0
exceeding 40 kN gross weight
(including access ways and ramps)
– to be calculated but not less than
(II) EDUCATIONAL BUILDINGS
(a) Class rooms and lecture rooms (not 3.0 2.7
used for assembly purposes)
(b) Dining rooms, cafterias and
restaurants 3.0† 2.7
(c) Offices, lounges and staff rooms 2.5 2.7
(d) Dormitories 2.0 2.7
(e) Projection rooms 5.0 –
(f) Kitchens 3.0 4.5
(g) Toilets and bathrooms 2.0 –
(h) Store rooms 5.0 4.5
(i) Libraries and archives :
(1) Stack room/stack area 6.0 kN/m2 for a 4.5
minimum height of
2.2 m + 2.0 kN/m2 per
metre height beyond
2.2 m
(2) Reading rooms (without separate 4.0 4.5
storage)
(3) Reading rooms (with separate
storage) 3.0 4.5
(j) Boiler rooms and plant rooms–to be 4.0 4.5
calculated but not less than
(k) Corridors, passages, lobbies, staircases 4.0 4.5
including fire escapes – as per the floor
services (without accounting for storage
and projection rooms) but not less than
Contd.
32 DESIGN OF STEEL STRUCTURES

Table 1.6 Contd.

(1) (2) (3) (4)
(kN/m2) (kN)
(l) Balconies Same as rooms to 1.5 per metre run
which they give access concentrated at the
but with a minimum outer edge
of 4.0
(III) INSTITUTIONAL BUILDINGS 2.0 1.8
(a) Bed rooms, wards, dressing rooms,
dormitories and lounges
(b) Kitchens, laundries and laboratories 30 4.5
(c) Dining rooms, cafeterias and 3.0† 2.7
restaurants
(d) Toilets and bathrooms 2.0 –
(e) X-ray rooms, operating rooms, general 30 4.5
storage areas – to be calculated but not
less than
(f) Office rooms and OPD rooms 2.5 2.7
(g) Corridors, passages, lobbies and 4.0 4.5
staircases including fire escapes – as
per the floor services but not less than
(h) Boiler rooms and plant rooms–to be 5.0 4.5
calculated but not less than
(i) Balconies Same as the rooms to 1.5 per metre run
which they give access concentrated at the
but with a minimum outer edge
of 4.0
(IV) ASSEMBLY BUILDINGS
(a) Assembly areas :
(1) with fixed seals ‡ 4.0 –
(2) without fixed seats 5.0 3.6
(b) Restaurants (subject to assembly), 4.0 4.5
museums and an galleries and
gymnasia
(c) Projection rooms 5.0 –
(d) Stages 5.0 4.5
(e) Office rooms, kitchens and laundries 3.0 4.5
(f) Dressing rooms 2.0 1.8
(g) Lounges and billiards rooms 2.0 2.7
(h) Toilets and bathrooms 2.0 –
Contd.
 GENERAL 33

Table 1.6 Contd.

(1) (2) (3) (4)
(kN/m ) 2
(kN)
(i) Corridors, passages, staircases 4.0 4.5
including fire escapes
(j) Balconies Same as rooms to 1.5 per metre run
which they give access concentrated at the
but with a minimum outer edge
of 4.0
(k) Boiler rooms and plant rooms 7.5 4.5
including weight of machinery
(l) Corridors, passages subject to loads 5.0 4.5
greater than from crowds, such as
wheeled vehicles, trolleys and the like.
Corridors, staircases and passages in
grandstands
(V) BUSINESS AND OFFICE BUILDINGS
(See also 3.1.2)
(a) Rooms for general use with separate
storage 2.5 2.7

(b) Rooms without separate storage 4.0 4.5

(c) Banking halls 3.0 27
(d) Business computing machine rooms 3.5 4.5
(with fixed computers or similar
equipment)
(e) Records/files store rooms and
storage space 5.0 4.5
(f) Vaults and strong room – to be 5.0 4.5
calculated but not less than
(g) Cafeterias and dining rooms 3.0† 2.7
(h) Kitchens 3.0 2.7
(i) Corridors, passages, lobbies and 4.0 4.5
staircases including fire escapes – as
per the floor serviced (excluding
stores) but not less than
(j) Bath and toilet rooms 2.0 –
Contd.
34 DESIGN OF STEEL STRUCTURES

Table 1.6 Contd.

(1) (2) (3) (4)
(kN/m )
2
(kN)
(k) Balconies Same as rooms to 1.5 per metre run
which they give access concentrated at the
but with a minimum outer edge
of 4.0
(l) Stationary stores 4.0 for each metre of 9.0
storage height
(m) Boiler rooms and plant rooms – 5.0 6.7
to be calculated but not less than
(n) Libraries See Sl. No. (II)
(VI) MERCANTILE BUILDINGS
(a) Retail shops 4.0 3.6
(b) Wholesale shops – to be calculated 6.0 4.5
but not less than
(c) Office rooms 2.5 2.7
(d) Dining rooms, restaurants and
cafeterias 3.0† 2.7
(e) Toilets 2.0 –
(f) Kitchens and laundries : 3.0 4.5
(g) Boiler rooms and plant rooms – 5.0 6.7
to be calculated but not less than
(h) Corridors, passages, staircases 4.0 4.5
including fire escapes and lobbies
(i) Corridors, passages, staircases subject 5.0 4.5
to loads greater than from crowds, such as
wheeled vehicles, trolleys and the like
(j) Balconies Same as rooms to 1.5 per metre
which they give access concentrated at
but with a minimum outer edge
of 4.0
(VII) INDUSTRIAL BUILDINGS
(a) Work areas without machi- 2.5 4.5
nes/equipment
(b) Work areas with machinery/equipment§
1. Light duty To be calculated 5.0 4.5
2. Medium duty but not less than 7.0 4.5
3. Heavy duty 10.0 4.5
Contd.
 GENERAL 35

Table 1.6 Contd.

(1) (2) (3) (4)
(kN/m2) (kN)
(c) Boiler rooms and plant rooms –
to be calculated but not less than 5.0 6.7
(d) Cafeterias and dining rooms 3.0† 2.7
(e) Corridors, passages and staircases 4.0 4.5
including fire escapes
(f) Corridors, passages, staircases subject 5.0 4.5
to machine loads, wheeled vehicles –
to be calculated but not less than
(g) Kitchens 3.0 4.5
(h) Toilets and bathrooms 2.0 –
(VIII) STORAGE BUILDINGS ¶
(a) Storage rooms (other than 2.4 kN/m2 per each 7.0
cold storage) warehouses – metre of storage
to be calculated based on the height with a
bulk density of materials minimum of
stored but not less than 7.5 kN/m2
(b) Cold storage – to be 50 kN/m2 per each 9.0
calculated but not less metre of storage height
than with a minimum of
15 kN/m2
(c) Corridors, passages and staircases 4.0 4.5
including fire escapes – as per the
floor serviced but not less than
(d) Corridors, passages subject to loads 5.0 4.5
greater than from crowds, such as
wheeled vehicles, trolleys and the like
(e) Boiler rooms and plant rooms 7.5 4.5

* Guide for requirements of low income housing.

† Where unrestricted assembly of persons is anticipated, the value of UDL should
be increased to 40 kN/m2.
‡ ‘With fixed scats’ implies that the removal of the sealing and the use of the
space for other purposes is improbable. The maximum likely load in this case
is, therefore, closely controlled.
§ The loading in industrial buildings (workshops and factories) varies
considerably and so three loadings under the terms ‘light’, ‘medium’ and ‘heavy’
are introduced in order to allow for more economical designs but the terms
have no special meaning in themselves other than the imposed load for which
36 DESIGN OF STEEL STRUCTURES

the relevant floor is designed. It is, however, important particularly in the

case of heavy weight loads, to assess the actual loads to ensure that they are
not in excess of 10 kN/m2; in case where they are in excess, the design shall be
based on the actual loadings.
¶ For various mechanical handling equipments which are used to transport goods,
as in warehouses, workshops, store rooms, etc. the actual load coming from
the use of such equipment shall be ascertained and design should cater to such
loads.
Note 1. Where no values are given (in Table 1.6) for concentrated load, it may be
assumed that the tabulation distributed load is adequate for design purposes.
Note 2. The loads specified in Table 1.6 are equivalent uniformly distributed
loads on the plan area and provide for normal effect of impact and acceleration.
These loads do not take into consideration special concentrated loads and other loads.
Note 3. Where the use of an area or floor is not provided in the Table 1.6, the
imposed load due to the use and occupancy of such an area shall be determined from
the analysis of the loads resulting from :
(i) Weight of the probable assembly of persons; (ii) Weight of the probable
accumulation of equipment and furnishing; (iii) Weight of the probable storage
materials; and (iv) Impact factor, if any.
While designing columns, abutments, piers, walls, their supports and foundations,
the live loads on floors are reduced as in Table 1.7 as per IS : 800–1984.
Table 1.7 Percent reduction of total live load

Number of floors carried by Percent reduction in total live load on

member under consideration all floors above the member under
consideration
1 0
2 10
3 20
4 30
5 or more 40
Reduction in live load shall not be made in the case of warehouses, garages
and other buildings used for storage purposes and for factories and workshops
designed for 5 kN/m2. However, above reductions are made for buildings such
as factories and workshops designed for a live load more than 5 kN/m2 provided
that the loading assumed for any column, etc., is not less it would have been if
the floors had been designed for 5 kN/m2 with no reduction.
As per IS : 875–1984, where a single span of beam or girder supports not less
than 50 m2 of floor at one general level, the live load may be reduced in the
design of beam or girder by 5 percent, for each 50 m2 supported, subjected to a
maximum reduction of 25 percent. This reduction or that given in Table 1.7,
whichever is greater, may be taken into account in the design of columns,
supporting such a beam. These reductions are not applicable for the floors used
for storage purposes, in the weight of any plant or machinery which is specifically
allowed for.
 GENERAL 37

1.24 WIND LOAD

The wind loads are the transient loads. Relative to the surface of earth, the
natural air remains in motion, (i.e., a current of air flows near the surface of
earth). The difference in solar and terrestrial radiations setting up irregularities
in temperatures (which gives rise to convection either upwards or downwards.
The wind usually blows horizontal to the ground at high wind speeds. The vertical
components of atmospheric motion are relative small. The term wind denotes
almost exclusive the horizontal wind. The vertical winds are always identified
as such. The wind speeds are assessed with the aid of anemometers or
anemographs which are installed at meteorological observatories at heights
generally varying from 10 m to 30 m above ground. Very weak winds (i.e., winds
at low speeds say in the range of 5 kmph to 10 kmph) are called breeze. The
strong winds (i.e., winds at speed say, 50 kmph) are generally associated with
cyclonic storms. Very strong winds (i.e., winds at speed say 80 kmph) are also
associated with cyclonic storms, thunderstorms, dust storms or vigorous
monsoons. The cyclonic storms weaken rapidly after crossing the coast and move
as depressions in land. A severe storm may have wind speed of 60 kmph to 120
kmph. The winds of very high speeds and very short duration (hurricanes) are
called Kal Baisaki or Norwesters occur fairly frequently during summer
months over North East India.
In addition to wind at any time, there are effects of gusts (sudden blast of
wind, which may last for few seconds) which should be considered. These gusts
cause increase in air pressure but their effect on the stability of the building
may not be so important. Often, the gusts influence only part of the building
and the increased local pressures may be more than balanced by a momentary
reduction in pressure elsewhere. Short period gusts may not cause any
appreciable increase in stress in main components of buildings (because of inertia
of the building), although the walls, roof sheetings and individual cladding units
(glass panels) and their supporting members such as purlins, sheeting rails and
glazing bars may be more seriously affected. Gusts may also be extremely
important for design of structures with slenderness ratios.
The liability of a building or a structure to high wind pressures depends not
only upon the geographical location and proximity of other obstructions to air
flow but also upon the characteristics of the structure itself. In general, wind
speed in the atmospheric boundary layer increases with height from zero at
ground level to maximum at a height called the gradient height. The variation
of wind with height depends primarily on the terrain conditions. However, the
wind speed at any height never remains constant and it has been found
convenient to resolve its instantaneous magnitude into an average or mean
value and a fluctuating component around this average value. The average value
depends on the average time employed in analysing the meteorological data,
which varies from a few seconds to several minutes. The magnitude of fluctuating
component of the wind speed is called gust, it depends upon averaging time. In
general, smaller the averaging interval, greater is the magnitude of the gust
speed.
38 DESIGN OF STEEL STRUCTURES

The wind load depends upon terrain, height of the structure, and the shape
and size of structure. For pitched roof, wind load has been considered and
described further in detail in Chapter 9 (Design of Roof Trusses). However, it is
essential to know the following terms to study the new concept of wind as
described in IS: 875 (Part 3) – 1987.
1. Angle of attack. It is defined as the angle between the direction of
wind and a reference axis of the structure
2. Breadth. It means horizontal dimension of the building measured
normal to the direction of wind.
Note. The breadth and depth are dimensions measured in relation to the direction
of wind. Whereas length and width are dimensions related to the plan of the building.
3. Depth. It means the horizontal dimension of the building measured in
the direction of the wind.
4. Developed height. It is height of upward penetration of the velocity
profile in a new terrain.
At larger fetch lengths, such penetration reaches the gradient height, above
which the wind speed may be taken to be constant. At lesser fetch lengths, a
velocity profile of a smaller height but similar to that of the fully developed
profile of that terrain category has to be taken. The velocity at the top of this
shorter profile equals that of the unpenetrated earlier velocity profile at that
height.
5. Effective frontal area. The projected area of the structure normal to
the direction of wind.
6. Element of surface area. The area of the surface over which the
pressure coefficient is taken to be constant.
7. Force coefficient. It is a non-dimensional coefficient such that the
total wind force on a body is the product of the force coefficient, the dynamic
pressure of the incident design wind speed and the reference area over which
the force is required.
In the direction of wind, this coefficient is called as drag coefficient. In the
perpendicular direction of wind, it is called as lift coefficient.
8. Ground roughness. The nature of the earth’s surface as affected by
small scale obstructions such as trees and buildings (as distinct from topography)
is called ground roughness.
9. Gust. A positive or negative departure of wind speed from its mean
value, lasting not more than say, 2 minutes over a specified interval of time.
10. Peak gust. Peak gust or peak gust speed is the wind speed associated
with the maximum amplitude.
11. Fetch lengths. It is the distance measured along the wind from a
boundary at which a change in the type of terrain occurs.
When the changes in terrain types are encountered (such as, the boundary of
a town or city, forest, etc.), the wind profile changes in character but such changes
are gradual and start at ground level, spreading or penetrating upwards with
increasing fetch length.
 GENERAL 39

12. Gradient height. It is the height above the mean ground level at
which the gradient wind blows as a result of balance among pressure gradient
force, coriolis force and centrifugal force.
It is taken as height above the mean ground level, above which the variation
of wind speed with height need not be considered.
13. Mean ground level. It is the level of average horizontal plane of
the area enclosed by the boundaries of the structure.
14. Pressure coefficient. It is the ratio of the difference between the
pressure acting at a point on a surface and the static pressure of the incident
wind to the design wind pressure.
Where the static and design wind pressures are calculated at the height of
the point considered after taking into consideration the geographical location,
terrain conditions and shielding effect. Also

⎡ ⎛ V ⎞2 ⎤
p
Pressure coefficient = ⎢1 − ⎜ ⎟ ⎥
⎢⎣ ⎝ Vz ⎠ ⎥⎦

where, Vp = actual wind speed at any point on the structure at a height

corresponding to that of Vz.
Note. When the sign of pressure coefficient is positive, it shows that the pressure
acts towards the surface and the negative sign shows that the pressure acts away
from the surface.
15. Return period. It is the number of years, the reciprocal exceeding a
given wind speed in any one year.
16. Shielding effect. It refers to the condition where wind has to pass
along some structure(s) or the structural element(s) located on the upstream
wind side, before meeting the structure or structural element under
consideration.
A factor called shielding factor is used to consider such effects in estimating
the force on the shielded structures.
17. Suction. It means the pressure less than the atmospheric (static)
pressure and it is taken to act away from the surface.
18. Solidity ratio. It is equal to the effective area (projected area of all
the individual elements) of a frame normal to the direction of wind divided by
the area enclosed by the boundary of the frame normal to the direction of wind.
Note. It is to be calculated for the individual frame.
19. Terrain category. It means the characteristics of the surface
irregularities of an area which arise from natural or constructed features. The
categories are numbered in the increasing order of roughness.
20. Velocity profile. The variation of horizontal component of the
atmospheric wind speed at different heights above the mean ground level is
termed as velocity ratio.
21. Topography. The nature of the earth’s surface as influenced by the
hill and valley configuration.
40 DESIGN OF STEEL STRUCTURES

1.25 SNOW LOAD

The snow load depends upon latitude of place and atmospheric humidity. The
snow load acts vertically and it is expressed in kilo-Newtons per square metre
of plan area. The actual load due to snow depends upon the shape of the roof
and its capacity to retain the snow. When actual data for snow load is not
available, snow load may be assumed to be 25 N/m2 per mm depth of snow.
It is usual practice to assume that snow load and maximum wind load will
not be acting simultaneously on the structure.

1.26 SEISMIC LOAD (EARTHQUAKE FORCE)

It becomes essential to consider ‘seismic load’ in the design of structure, if the
structure is situated in the seismic areas. The seismic areas are the regions
which are geologically young and unstable parts and which have experienced
earthquakes in the past and are likely to experience earthquakes in future. The
Himalayan region, Indo Gangetic Plain, Western India, Cutch and Kathiawar
are the places in our country which experience earthquakes frequently.
Sometimes these earthquakes are violent also. Seismic load is caused by the
shocks due to an earthquake. The earthquakes range from small tremors to
severe shocks. The earthquake shocks cause movement of ground, as a result of
which the structure vibrates. The vibrations caused because of earthquakes
may be resolved in three perpendicular directions. The horizontal direction of
vibration dominates over other directions. In some cases structures are designed
for horizontal seismic forces only and in some cases horizontal seismic forces
and vertical seismic forces both are taken into account. The seismic accelerations
for the design may be arrived at from seismic coefficient, which is defined as the
ratio of acceleration due to earthquakes and acceleration due to gravity. This
country has been divided into seven zones for determining seismic coefficients.
The seismic coefficients have also been recommended for different types of soils
for the guidance of designers. IS : 1893–1962 Indian Standard Recommendations
for Earthquake Resistant Design of Structure, may be referred to for actual
design.

1.27 SOIL AND HYDROSTATIC PRESSURE

The pressure exerted by soil or water or both should be taken into consideration
for the design of structures or parts of structures which are below ground level.
The soil pressure and hydrostatic pressure may be calculated from established
theories.

1.28 ERECTION EFFECTS

The erection effects include all effects to which a structure or a part of structure
is subjected during transportation of structural members, and erection of
structural member by equipments. Erection effects also take into account the
 GENERAL 41

placing or storage of construction materials. The proper provisions shall be made,

e.g., temporary bracings, to take care of all stresses caused during erection. The
stresses developed because of erection effects should not exceed allowable
stresses.

1.29 DYNAMIC EFFECTS (IMPACT)

The moving loads on a structure cause vibrations and have also impact effect.
The dynamic effects resulting from moving loads are accounted for, by impact
factor. The live load is increased by adding to it the impact load. The impact
load is determined by the product of impact factor and live load. The effect of
impact has been described in Chapter 8 (Design of Plate Girder).

1.30 TEMPERATURE EFFECTS

The variations in temperature results in expansion and contraction of structural
material. The range of variation in temperature varies from localities to localities,
season to season and day to day. The temperature effects should be accounted
for properly and adequately. The allowable stress should not be exceeded by
stress developed because of design loads and temperature effects.

1.31 LOAD COMBINATIONS

All the parts of the steel structures shall be capable of sustaining the most
adverse combination of the dead loads, prescribed live (imposed) loads, wind
loads, earthquake loads where applicable and any other forces or loads to which
the steel structure may reasonably be subjected without exceeding the stresses
specified in this standard.
The load combinations for dsign purposes shall be the one that produces
maximum forces and effects and consequently maximum stresses from the
following combinations.
(i) Dead load + imposed (live) loads,
(ii) Dead load + imposed (live) loads + wind or earthquake loads, and
(iii) Dead load + wind load or earthquake loads.

1.32 STRESSES
When a structural member is loaded, deformation of the member takes place,
and resistance is set up against deformation. This resistance to deformation is
known as stress. The direct stress is defined as force per unit cross-sectional
area. The nature of stress developed in the structural member depends upon
nature of loading on the member. The following are the various types of stresses:
1. Axial stress (Direct stress) :
(i) Tensile stress (ii) Compressive stress
2. Bearing stress
3. Bending stress
4. Shear stress.
42 DESIGN OF STEEL STRUCTURES

The bending stress and shear stress have been discussed in Chapter 6, ‘Design
of Beams’.
A member may be subjected to combined direct and bending stresses. Such
stress is known as combined stress.
Sign convention for stresses. The tensile stresses are taken as positive
and compressive stresses as negative. This sign convention for stresses is
convenient as a structural member elongates on application of tensile load and
shortens on application of compressive load.

1.33 STRESS–STRAIN RELATIONSHIP FOR MILD STEEL

When a mild steel bar is subjected to a tensile load, it elongates. The elongation
per unit length is known as strain. The stress is proportional to strain within
limit of proportionality. The stress-strain relationship for mild steel can be
studied by plotting stress-strain curve. The stress and load on some suitable
scales may be plotted on y-axis, and strain on some suitable scale may be plotted
on x-axis.
When the tensile load increases with increases in strain, stress–strain curve
follows a straight line relationship upto ‘Limit of proportionality’. The limit of
proportionality is defined as stress beyond which straight line relationship ceases
between stress and strain. Beyond the limit of proportionality stress approaches
the elastic limit. The elastic limit is defined as the maximum stress upto which
a specimen regains its original length on the removal of the applied load. There
is hardly any distinct difference in the positions of limit of proportionality and
elastic limit. Practically, position of limit of proportionality coincides with the
elastic limit. When the specimen is loaded beyond the elastic limit, the specimen
does not resume its original length on the removal of applied load and a little
strain is left in the specimen. This little strain is known as residual strain or
permanent set. When the tensile load further increases the stress–reaches ‘yield
stress’ and material starts yielding. The stress–strain curve suddenly falls
showing a decrease in stress. The distinct position from where sudden fall of
curve occurs marks the upper yield point and the position upto which fall of
curve occurs is known as lower yield point. The material of bar stretches suddenly
at constant stress. The adjustment of stress takes place in the elements of
material in between upper yield point and lower yield point. On further increase
of load, stress increases with the increase of strain. However, strain increases
more rapidly. Finally the load reaches the value of ‘ultimate load’. The ultimate
load is defined as maximum load, which can be placed prior strain increases
more rapidly. Finally the load reaches the value of ‘ultimate toad’. The ultimate
load is defined as maximum load, which can be placed prior to the breaking of
specimen. The stress corresponding to the ultimate load is known as ‘ultimate
stress’. The stress–strain curve suddenly falls with rapid increase in strain and
specimen breaks. The load corresponding to breaking position is known as
“breaking load’. The cross-section of specimen decreases. If actual breaking stress
is computed on the basis of decreased (reduced) cross-sectional area, the breaking
stress will be found to be more than the ultimate stress.
 GENERAL 43

4 50 U ltim ate
4 00

3 50 U p pe r
yie ld B re akin g
3 00
S train h arde ning
S tress (N /m m 2 )

2 50
L ow e r yie ld
2 00
E lastic lim it
L im it of pro po rtion ality
1 50

1 00

50

0
5 10 15 20 25
S train (P erce nt)
S tress-strain cu rve fo r m ild stee l

Fig. 1.8 Stress–strain curve

1.33.1 Significance of Upper Yield Point

The boundaries of grains of mild steel are composed of brittle material. This
forms a rigid skeleton. The rigid skeleton prevents plastic deformation of the
grains at low stress, and shows upper yield point in stress-strain curve. At upper
yield point, this rigid skeleton breaks down. As a result of this, the stress in
material drops down without elongation from upper yield point to lower yield
point. This is followed by sudden stretching of the material at constant stress
from lower yield point upto strain hardening.

1.34 TENSILE STRESS

When a structural member is subjected to direct axial tensile load, the stress is
known as tensile stress (σat). The tensile stress is calculated on net cross-sectional
area of the member

⎛P ⎞
σat = ⎜ t ⎟
⎝ An ⎠
where Pt is the direct axial tensile load and An is the net cross-sectional area of
the member.

1.35 COMPRESSIVE STRESS

When a structural member is subjected to direct axial compressive load, the
stress is known as compressive stress, (σac). The compressive stress is calculated
on gross cross-sectional area of the member
44 DESIGN OF STEEL STRUCTURES

⎛P ⎞
σac = ⎜ c ⎟
⎝ Ag ⎠
where, Pc = direct axial compressive load
Ag = gross-sectional area of the member.

1.36 BEARING STRESS

When a load is exerted or transferred by the application of load through one
surface for the another surface in contact, the stress is known as ‘bearing stress’
(σp). The bearing stress in calculated on net projected area of contact
⎛P⎞
σp = ⎜ ⎟
⎝A⎠
where, P = load placed on the bearing surface
A = net projected area of contact

1.37 WORKING STRESS

The working stress is also termed as allowable stress or permissible stress. The
working stress is evaluated by dividing yield stress by factor of safety. For the
purpose of computing safe load carrying capacity of a structural member, its
strength is expressed in terms of working stress. The working stress is the stress
which may be developed or set up in the member without causing structural
damage to it. The actual stresses resulting in a structural member from design
loads should not exceed working stresses. This ensures the safety of the structural
member. The maximum working stresses are adopted from IS : 800–1984.
IS : 800–1984 specifies a number of grades of steel with different yield
strengths, the design parameters and the geometrical properties. There are
also developments in the design of steel structures. In view of these, the
permissible stresses have been revised in general. These permissible stresses
have been expressed to the extent possible in terms of the yield strength of the
material. However, the specific values have also been given in IS : 800–1984 for
steels commonly used.

1.38 INCREASE IN PERMISSIBLE STRESSES

A structure may be subjected to the different combinations of loads. The possible
combinations of the loads with special reference to the roof trusses have been
described in Chapter 9. These loads in such combinations do not act for long
period. Most of the national codes allow some increase in permissible stresses.
Increases in permissible stresses as per IS : 800–1984 is taken as follows :
(i) When the effect of wind or seismic load is taken into account, the
permissible stresses in steel are increased by 33 13 per cent.
(ii) For rivets, bolts and tension rods, the permissible stresses are increased
by 25 per cent, when the effect of wind or seismic load is taken into
account.
 GENERAL 45

The increased values of permissible stresses must not exceed yield stress of
the material.

1.39 FACTOR OF SAFETY

The factor of safety is defined as the factor by which the yield stress of the
material is divided to give the working stress (permissible stress) in the material.
A greater value of factor of safety used to be adopted formerly and as such the
working stresses computed were small. As a result of this a large cross-section
of the member had to be adopted in design. From experience gained and
experiments made in laboratories, it has been seen that such a large section of
the structural member is not essential. Hence, factor of safety adopted now-a-
days is comparatively small. This has resulted in appreciable saving in the
material. The value of factor safety is decided keeping in view of the following
considerations.
1. The average strength of materials is determined after making tests on
number of specimens. The strengths of different specimens of given
structural material are not identical. Some of the specimens may show
relatively large deviations in strength from the average.
2. The values of design loads remains uncertain. The values of dead loads
can of course be determined correctly. But live load, impact load, wind
load, snow load etc. cannot be determined with certainty since these
depend upon statistics available. The probable values of these loads
are only determined.
3. The values of internal forces in many structures depend upon the
methods of analysis. The degree of precision of different method varies.
The methods involving detailed analysis are more precise. In case,
analysis of the structure is done precisely, a small value of factor of
safety may be adopted.
4. During fabrication, structural steel is subjected to different operations.
The punching of a hole in a structural element distorts the surrounding
material and causes high residual stresses. The warping and buckling
of elements may take place during welding. The welding leaves high
residual stresses. Structural elements are subjected to uncertain
erection stress.
5. The variations in temperatures and settlement of supports are
uncertain. Many times, a well designed structure is damaged because
of these effects. The strength of materials decreases because of corrosion.
The extent of corrosion is more, when a structure is located in industrial
areas and exposed to chemical wastes.
6. The failure of some small structures or some elements of a structure is
less serious and less disastrous than the failure of large structure or a
main element of a structure.
The minimum value of factor of safety, mmin may be found by defining it in a
rational method. The factor of safety may be defined as the ratio of computed
46 DESIGN OF STEEL STRUCTURES

strength, P of the structure or the structural member to the respective computed

internal load, F. Therefore,
⎛P⎞
m = ⎜ ⎟ ...(1.1)
⎝F ⎠
The magnitude of lowest probable strength, P is influenced by the uncertainties
in the mechanism failure, properties of material and workmanship. The
magnitude of highest probable internal load, F is influenced by the uncertainties
in loading conditions and the structural behaviour. The probable values deviate
from the computed value because of these uncertainties. The minimum probable
value of strength is given by
Pmin = (P – ∆P) ...(i)
The maximum probable value of the internal load is given by
Fmax = (F + ∆F) ...(ii)
where, ∆P = Maximum probable deviation of actual value from the computed
value of strength
∆F = Maximum probable deviation of actual value from the computed
value of internal load
When the minimum probable value of strength is atleast equal to or slightly
more than the maximum probable value of the internal load, then the failure of
structure is just prevented under most severe conditions. Therefore,
(P – ∆P) ≥ (F + ∆F) ...(iii)
⎛ ∆P ⎞ ⎛ ∆F ⎞
or P ⎜1 − ⎟ ≥ F ⎜1 + ⎟ ...(iv)
⎝ P ⎠ ⎝ F ⎠
The minimum value of factor of safety is given by

⎛⎛ ∆F ⎞ ⎞
⎜ ⎜ 1+ ⎟⎟
⎛P ⎞ ⎝ F ⎠⎟
mmin = ⎜ ⎟=⎜ ...(1.2)
⎝F ⎠ ⎜⎛ ∆P ⎞ ⎟
⎜ ⎜1 − ⎟⎟
⎝⎝ P ⎠⎠
Assuming the values of maximum deviations ∆P and ∆F to 25 percent of
the computed values of P anbd F respectively, the minimum value of factor of
safety is given by
⎛ P ⎞ ⎛ 1 + 0.25 ⎞
mmin = ⎜ ⎟ = ⎜ ⎟
⎝ F ⎠ ⎝ 1 – 0.25 ⎠
= 1.67 ...(v)

1.40 METHODS OF DESIGN

All parts of the steel framework of a structure shall be capable of sustaining the
most adverse combination of the dead loads, the prescribed superimposed roof
and floor loads, wind loads, seismic forces where applicable, supperimpose roof
 GENERAL 47

and floor loads, wind loads, seismic forces where applicable, and any other forces
or loads to which the building may be reasonably subjected without exceeding
the permissible stresses. The following methods may be employed for the design
of the steel framework :
(i) Simple design
(ii) Semi-rigid design
(iii) Fully rigid designs, and
(iv) Plastic design.

1.40.1 Simple Design

This method is based on elastic theory and applies to structures in which the
end connections between members are such that they will not develop restraint
moments adversely affecting the members and the structures as a whole and in
consequence the structure may, for the purpose of design, be assumed to be pin
jointed. The method of simple design involves the following assumptions:
(a) The beams are simply supported.
(b) All connections of beams, girders, or trusses are virtually flexible and
are proportioned for the reaction shears applied at the appropriate
eccentricity.
(c) The members in compression are subjected to forces applied at the
appropriate eccentricities.
(d) The members in tension are subjected to longitudinal forces applied
over the net area of the section.
(e) The plane sections normal to the axis remain plane after bending. The
stress strain relationship for steel is linear.

1.40.2 Semi-rigid Design

This method as compared with the simple design method permits a reduction in
the maximum bending moment in beams suitably connected to their supports,
so as to provide a degree of direction fixity. In the case of triangulated frames, it
permits rotation account being taken of the rigidity of the connections and the
moment of interaction of members. In cases where this method of design is
employed, it is ensured that the assumed partial fixity is available and
calculations based on general or particular experimental evidence shall be made
to show that the stresses in any part of the structure are not in excess of those
laid down in IS : 800–1984.

1.40.3 Filly-rigid Design

This method as compared to the methods of simple and semi-rigid design assumes
that the end connections are fully rigid and are capable of transmitting moments
and shears. It is also assumed that the angle between the members at the joint
does not change, when it is subjected to loading. This method gives economy in
the weight of steel used when applied in appropriate cases. The end connections
48 DESIGN OF STEEL STRUCTURES

of members of the frame shall have sufficient rigidity to hold virtually unchanged
original angles between such members and the members they connect. The design
should be based on accurate methods of elastic analysis and calculated stresses
shall not exceed permissible stresses.

1.40.4 Plastic Design

The method of plastic analysis and design is recently developed. The method
was thought of around 1935. However, even today all the problems related to
this are not decided. In this method, the structural usefulness of the material is
limited upto to ultimate load. This method has its main application in the analysis
and design of statically indeterminate frame structures. This method provides
striking economy as regards the weight of the steel. This method provides the
margin of safety in terms of load factor which one is not less than provided in
elastic design. A load factor of 1.85 is adopted for dead load plus live load and
1.40 is adopted for dead load, live load and wind or earthquake forces. The
deflections under working loads should not exceed the limits prescribed in IS :
800–1984.
The method of plastic analysis and design has been described in Author’s
Vol. II (Design of Steel Structures).

1.41 DEFINITIONS
Following definitions concerning the mechanical properties of the steel are useful
for the study of this subject.

1.41.1 Modulus of Elasticity

The modulus of elasticity is defined as the ratio of longitudinal stress to the
longitudinal strain within the elastic region. It is denoted by letter E, the slope
of elastic portion of the stress–strain diagram gives the value of modulus of
elasticity, E.

1.41.2 Shear Modulus of Elasticity

The shear modulus of elasticity is defined as the ratio of shearing stress to the
shearing strain within the elastic range. It is denoted by letter G. The shear
modulus of elasticity is also known as modulus of rigidity. It is given by
⎛ E ⎞
G = ⎜ ⎟ ...(1.3)
⎝ 2(1 + µ) ⎠
where µ = Poisson’ s ratio

1.41.3 Bulk Modulus of Elasticity

The bulk modulus of elasticity is defined as the ratio of hydrostatic stress to the
volumetric strain within the elastic range. It is denoted by letter K. It is given
by
 GENERAL 49

⎛ E ⎞
K = ⎜ ⎟ ... (1.4)
⎝ 3(1 − 2 µ) ⎠
It is also given by
⎛ 2G(1 + µ) ⎞
K = ⎜ ⎟ ... (1.5)
⎝ 3(1 − 2 µ) ⎠

1.41.4 Tangent Modulus of Elasticity

The tangent modulus of elasticity is defined as the slope of the tangent at a
point on the stress–strain curve above the limit of proportionality. It is denoted
by Et.

1.41.5 Poisson’s Ratio

The Poisson’s ratio is defined as the ratio of transverse strain to the longitudinal
strain under an axial load. It is denoted by letter µ. The value of Poisson’s ratio
for steel within the elastic region ranges from 0.25 to 0.33.

1.41.6 Yield Point

The yield point on a stress–strain diagram is defined as the stress at which the
strain increases to a large value without an increase of stress. For the mild
steel, it is indicated by a flat portion of stress:strain diagram if the stress–strain
diagram is plotted on the exaggerated scale. Some steels show upper yield point
and lower yield point.

1.41.7 Yield Strength

Some steels do not show yield plateau, i.e., the flat portion on the stress–strain
diagram at yield. The stress–strain diagram for such steels show continuous
curve upto the tensile strength. Therefore, the yield strength of such steel is
defined as a stress shown by a specific point on the curve obtained by drawing
0.2 per cent offset of the strain parallel to the initial elastic portion of the curve.

1.41.8 Tensile Strength

The tensile strength of a test piece is defined as maximum axial load divided by
the original area of cross-section.

1.41.9 Fatigue Strength

The fatigue strength is defined as the stress at which steel fails under repeated
applications of load.

1.41.10 Impact Strength

The impact strength of steel is the measure of ability of the steel to absorb
energy at high rates of loading.
50 DESIGN OF STEEL STRUCTURES

1.41.11 Strain Hardening Modulus

The strain hardening modulus is defined as the slope of the stress–strain curve
in the strain–hardening range. It is denoted by Est. It has the maximum value
at the beginning of strain–hardening.

1.41.12 Ductility
The ductility is the unique property of steel. It is the property of steel by virtue
of which, the deformation of structural member/structure occurs before failure.
It is measured quantitatively by percentage of elongation of test specimen at
failure in a uniaxial test. A gauge length, (LGL = 5.65 (A0)1/2, where Ao is the
original cross-sectional area of the bar) is marked in the middle of the bar. The
bar is tested upto failure. Its elongated length, LEL is measured. Then, the

⎛ L − LGL ⎞
percentage elongation = ⎜ EL ⎟ × 100 . For a good ductile steel the
⎝ LGL ⎠
percentage elongation should be 20 to 33 percent. The ductility is also measured

by percentage of reduction in area. It is given by

( Ao − AR ) × 100 where Ao and
Ao
AR are the original and reduced cross-sectional areas of the bars. For good ductile
steel bars, the percentage of reduction may be 65 percent.

1.42 STABILITY OF STRUCTURE

According to the stability requirement, the stability of a structure as a whole
against overturning is ensured so that the restoring moment is greater than the
maximum over-turning moment.

1.42.1 Overturning
The stability of a structure as a whole against overturning is ensured such that
the restoring moment, MR shall be not less than the sum of 12 times the maximum
overturning moment due to the characteristic dead load, Mo.CDL and 1.4 times
the maximum overturning moment due to characteristic imposed loads, Mo.CIL.
i.e.,
MR |< (1.2 Mo.CDL+ 1.4 Mo.CIL)
In cases, where dead load provides the restoring moment, only 0.9 times the
characteristic dead load shall be considered. The restoring moment due to
imposed loads shall be neglected.
The anchorages or counterweights against for overhanging members (during
construction and service) should be such that the static equilibrium should be
maintained, even when the overturning moment is doubled.
 GENERAL 51

1.42.2 Sliding
The structure should have adequate factor of safety against sliding due to the
most adverse combination of the applied loads.
The structure shall have a factor of safety against sliding not less than 1.4
under the most adverse combination of the applied characteristic forces. In case
only dead loads are acting, only 0.9 times the characteristic dead load shall be
taken into account.
To ensure stability at all times, account shall be taken of probable variations
in dead load during construction, repair or other temporary measures. The wind
and seismic loading shall be treated as imposed loading.
In designing the framework of a building, provisions shall be made by adequate
moment connections or by a system of bracings to effectively transmit all the
horizontal fores to the foundations.

1.43 STIFFNESS OF STRUCTURE

In order to meet the requirement of stiffness for the structure as a whole and for
the structural members, the limitations for the deflections are specified. The
deflection of the structure or part thereof should not adversely affect the
appearance or efficiency of the structure or finishes or partitions.
 Chapter Design of Riveted,

2 Bolted and Pin

Connections

2.1 INTRODUCTION
The various types of connections used for connecting the structural members
are given below :
1. Riveted connections
2. Bolted connections
3. Pin connections
4. Welded connections.
These connections are named after the type of fastening (viz., rivets, bolts
and nuts, pins and welds) used for connecting the structural members. In the
riveted connections, rivets are used. These rivets are the permanent fastenings.
Once the rivets are driven, these rivets cannot be opened without considerable
labour and without complete destruction.
The perfect theoretical analysis for stress distribution in riveted connections
cannot be established. Hence a large factor of safety is employed in the design
of riveted connections. The riveted connections should be as strong as the
structural members. No part in the riveted connections should be so over-
stressed. These riveted connections should be so designed that there is neither
any permanent distortion nor any wear. These should be elastic.
In general, the work of fabrication is completed in the workshops where the
steel is fabricated.

2.2 RIVETS
A piece of round steel forged in place to connect two or more than two steel
members together is known as a rivet.
The rivets for structural purposes are manufactured from mild steel and
high tensile rivet bars. A rivet consists of a head and a body as shown in Fig. 2.1.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 53

The body of rivet is termed as shank. The rivets are manufactured in different
lengths to suit different purposes. The size of rivets is expressed by the diameter
of the shank.
The most rivets used in structural steel work are heated uniformly throughout
its length, without burning or excessive scaling, and shall be of sufficient length
to provide a head of standard dimension when these rivets are driven, these
rivets shall be completely fill the holes. These rivets are known as hot driven
rivets.
For driving the rivets, they are heated till they become red hot and are then
placed in the hole. Keeping the rivets pressed from one side, a number of blows
are applied and a head at the other end is formed. The hot-driven rivets are
divided in the following three types, according to the method of rivet-driving:
1. Power driven rivets
2. Hand driven rivets
3. Field rivets.
The rivets of the first category are more satisfactorily driven than the rivets
of the other two categories and rivets of second category are more satisfactorily
driven than rivets of the third category because of the difference in workmanship.
Consequently strength of a rivet is different for distinctly seen in the working
stresses for these rivets given in Table 2.1. The hot driven rivets of 16 mm, 18
mm, 20 mm and 22 mm diameter are used for the structural steel works.

H e ad

B o dy
(S h an k)

R ive t
Fig. 2.1 Rivet

Some rivets are driven at atmospheric temperature. These rivets are known
as cold driven rivets. The cold driven rivets need large pressure to form the
head and complete the driving. The small size rivets ranging from 12 mm to 22
mm in diameter may be cold driven conveniently. The strength of rivet increases
in the cold driving. The heating of rivet is not necessary. The use of cold driven
rivets is limited because of equipment necessary and inconvenience caused in
the field.
54 DESIGN OF STEEL STRUCTURES

2.3 RIVET HEADS

The various types of rivet heads employed for different works are shown in
Fig. 2.2.
The proportions of various shapes of rivet heads have been expressed in terms
of diameter D of shank of rivet. The snap head is also termed as round head
and button head. The snap heads are used for rivets connecting structural
members. The countersunk heads are used to provide a flush surface.

2.4 RIVET HOLES

The rivet holes are made in the plates or structural members by one of the
following two methods :
1. Punching
2. Drilling.
When the rivet holes are made by punching, the holes are not perfect, but
taper. A punch damages the material around the hole. The operation known as
reaming is done in the hole made by punching.

0 .7D 0 .7D 0 .7D

1 .6D 1 .6D 1 .6D

1 5°
D D D

(a ) S n ap h ea d (b ) P a n he ad (c) P a n he ad w ith
ta pe re d n e ck

1 .5 D 1 .5 D

D D
5 5
0. 0.
6 0° 6 0°

D D

(d ) R o un d co un te r- (e ) C o un te r-sun k he ad
sun k he ad
Fig. 2.2 Rivet heads

When the rivet holes are made by drilling, the holes are perfect and provide
good alignment for driving the rivets.
The diameter of a rivet hole is made larger than the nominal diameter of the
rivet by 15 mm of rivets less than or equal to 25 mm diameter and by 2 mm for
diameters exceeding 25 mm.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 55

2.5 DEFINITIONS OF TERMS USED IN RIVETING

2.5.1 Nominal Diameter of Rivet
The nominal diameter of rivet is the diameter of the cold rivet measured before
driving.

2.5.2 Gross Diameter of Rivet

The gross diameter of rivet is the diameter of the rivet measured after driving
and the diameter of the rivet hole is adopted as the gross diameter of a rivet.

2.5.3 Pitch of Rivet (p)

The pitch of rivets is the distance between two consecutive rivets measured
parallel to the direction of the force in the structural member, lying on the same
rivet line. It is the centre to centre distance between the individual fasteners.
The pitch of the rivet is kept constant over as large a length as it is possible. The
pitch of the rivet is not changed from rivet to rivet. It is better to keep a large
number of pitches with an odd pitch cast to each end than to keep all the pitches
the same with each pitch containing an odd figure. It facilitates the marking
and checking the job. It saves time also.

2.5.4 Gauge Distance of Rivets (g)

The gauge distance is the transverse distance between two consecutive rivets of
adjacent chains (parallel adjacent lines of fasteners) and is measured at right
angles to the direction of the force in the structural member.

2.5.5 Gross Area of Rivet

The gross area of a rivet is the cross-sectional area of a rivet calculated from the
gross diameter of the rivet.

2.5.6 Rivet Line

The rivet line is also known as scrieve line or back line or gauge line. The rivet
line is the imaginary line along which rivets are placed. The rolled steel section
(e.g., I-sections, channel sections, tee-sections) have been assigned standard
positions of the rivet lines. These standard positions of rivet lines are conformed
to whenever possible. The standard positions of the rivet lines depend upon the
flange widths in the case if I-sections, channel sections and tee-sections. The
standard positions of the rivet lines for the angle sections depend upon the
length of legs. For the equal angle sections, the rivet lines are at equal distances
from the heel. For the unequal angle sections, the rivet lines are also at unequal
distances. The standard positions of rivet lines for the various sections may be
noted from ISI Handbook No. 1 for the respective sections. The departure from
standard position of the rivet lines may be done if necessary. The dimensions of
56 DESIGN OF STEEL STRUCTURES

rivet lines should be shown irrespective of whether the standard position have
been followed or not.

2.5.7 Staggered Pitch

The staggered pitch is also known as alternate pitch or reeled pitch. The staggered
pitch is defined as the distance measured along one rivet line from the centre of
a rivet on it to the centre of the adjoining rivet on the adjacent parallel rivet
line. One or both the legs of an angle section may have double rivet lines. The
staggered pitch occurs between the double rivet lines.

2.6 WORKING STRESSES IN RIVETS

The working stresses in the rivets as per IS : 800–1984 are given in Table 2.1.

Table 2.1 Working (maximum permissible) stresses in rivets

Description Working stress

In Tension, σtf
Axial stress on gross area of rivets :
(i) Power driven rivets 100 N/mm2 (MPa )
(ii) Hand-driven rivets 80 N/mm2 (MPa)
In Shear, τvf
Shear stress on gross area of rivets :
(i) Power-driven rivets 100 N/mm2 (MPa)
(ii) Hand-driven rivets 80 N/mm2 (MPa )
In Bearing, σpf
Bearing stress on gross diameter of rivets :
(i) Power-driven rivets 300 N/mm2 (MPa )
(ii) Hand-driven rivets 250 N/mm2 (MPa )

Note. For high tensile steel rivets working stresses as given in IS : 800–1984 may
be adopted. For the field rivets, the permissible stress are reduced by ten per cent.
The calculated bearing stress of a rivet on the parts connected by it shall not
exceed (i) value of the yield stress, fy for hand driven rivets and (ii) value of 1.2
times yield stress of the connected parts for power driven rivets.
When the rivets are subjected to both shear and axial tension, these shall be
so proportioned that the shear and axial stresses calculated shall not exceed
the respective allowable stresses, τsf and σtf and the expression
⎛ τvf ⋅cal. σtf ⋅cal. ⎞
⎜ τ + shall not exceed 1.4.
⎝ vf τtf ⎟⎠

2.7 RIVETED JOINT

The riveted joints are of two types :
1. Lap joint 2. Butt joint.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 57

2.7.1 Lap Joint

When one member is placed above the other and the two are connected by means
of rivets the joint is known as lap joint as shown in Fig. 2.3 (a). In case, the lap
joint, lines of forces are eccentric; this causes bending stress, and has the tendency
to deform. The deformation of lap joint has been shown in Fig. 2.3 (b).
O verlap

(a) Lap joint (Single riveted)

O verlap

(b) Deformation of lap joint

Fig. 2.3
58 DESIGN OF STEEL STRUCTURES

These joints are further classified according to the number of rivets used and
the arrangement of rivets adopted. Following are the different types of lap joints :
1. Single riveted lap joint 2. Double riveted lap joint:
(a) Chain riveted lap joint.
(b) Zig-zag riveted lap joint.
The chain riveted and zig-zag riveted lap joints have been shown in Fig. 2.4
(a) and (b), respectively.

P = Pitch

(a) Chain riveted lap joint

(b) Zig-zag riveted lap joint

Fig. 2.4
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 59

2.7.2 Butt Joint

When plates are placed end to end and flushed with each other and are joined
by means of cover plates, the joint is known as butt joint. The butt joints are of
two types :
(a) Single cover butt joint, (b) Double cover butt joint.
In single cover butt joint as shown in Fig. 2.5 (a), cover plate is provided on
one side of main plates. In this type of joint, bending stress may develop which
may cause deformation of joint as shown in Fig. 2.5 (b).

(a ) S ingle cover plate butt-jo in t

(b ) D e form a tio n of jo in t

Fig. 2.5

In case of double cover butt joint as shown in Fig. 2.6, cover plates are used
on either side of the main plates. There is no possibility of the development of

Fig. 2.6 Double cover single riveted butt joint

bending stress and deformation (distortion) of joint in this type of joint. Butt
joints are also further classified according to the number of rivets used and the
arrangement of rivets adopted. Following are the different types of butt joints :
1. Double cover single riveted butt joint
2. Double cover chain riveted butt joint
3. Double cover zig-zag riveted butt joint.
60 DESIGN OF STEEL STRUCTURES

2.8 TRANSMISSION OF LOAD IN RIVETED JOINTS

There are two modes of transmission of load in riveted joints. When the load is
transmitted by bearing between plates and shanks of rivets, the rivets are
subjected to shear. When the shear of rivets is only across one cross-section of
the rivet, it is known as single shear as shown in Fig. 2.7 (a). When the shear of
rivet is across two cross-sections of the rivet, it is known as double shear as
shown in Fig. 2.7 (b).

(a )

(b)

Fig. 2.7 Rivets in shear

R ive t in ten sion

P P

R ive t in ten sion

Fig. 2.8 Rivet in tension

When the load is transmitted by shearing between heads of rivets and plates,
the rivets are subjected to tension as shown in Fig. 2.8.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 61

2.9 FAILURE OF A RIVETED JOINT

The failure of a riveted joint may take place in any of the following ways :
1. Shear failure of rivets
2. Shear failure of plates
3. Tearing failure of plates
4. Bearing failure of plates
5. Splitting failure of plates at the edges
6. Bearing failure of rivets.

2.9.1 Shear Failure of Rivets

The plates riveted together and subjected to tensile loads may result in the
shear of the rivets. The rivets are sheared across their cross-sectional areas.
Single shear occurring in a lap joint has been shown in Fig. 2.9 (a) and double
shear occurring in butt joint has been shown in Fig. 2.9 (b).

(a )

(b )
Fig. 2.9 Shear failure of plates

2.9.2 Shear Failure of Plates

A plate may fail in shear along two lines as shown in Fig. 2.10. This may occur
when minimum proper edge distance is not provided.

Fig. 2.10 Shear failure of plates

62 DESIGN OF STEEL STRUCTURES

2.9.3 Tearing Failure of Plates

When plates riveted together are carrying tensile load, tearing failure of plate
may occur, when strength of the plate is less than that of rivets. The tearing
failure occurs at the net sectional area of plate as shown in Fig. 2.11.

Fig. 2.11 Tearing failure of plates

2.9.4 Bearing Failure of Plates

The bearing failure of a plate may occur because of insufficient edge distance in
the riveted joint. The crushing of plate against the bearing of rivet as shown in
Fig. 2.12 takes place in such failure.

Fig. 2.12 Bearing failure of plates

2.9.5 Splitting Failure of Plates

The splitting failure of a plate may occur because of insufficient edge distance
in the riveted joint. The splitting (cracking) of plate as shown in Fig. 2.13 takes
place in such failure.

Fig. 2.13 Splitting failure of plates

 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 63

2.9.6 Bearing Failure of Rivets

The bearing failure of a rivet occurs when the rivet is crushed by the plate as
shown in Fig. 2.14.

Fig. 2.14 Bearing failure of rivets

The bearing, shearing and splitting failure of plates may be avoided by

providing adequate edge distance. To safeguard a riveted joint against other
modes of failure, the joint should be designed properly.

2.10 STRENGTH OF RIVETED JOINT

The strength of a riveted joint is determined by computing the following
strengths :
1. Strength of riveted joint against shearing of the rivets ... Ps
2. Strength of a riveted joint against bearing of the rivets ... Pb
3. Strength of plate in tearing ... Pt
The strength of a riveted joint is the least strength of the above three strengths
viz., Ps, Pb, and Pt.

2.10.1 Strength of a Riveted Joint Against the Shearing of the Rivets

The strength of riveted joint against the shearing of rivets is equal to the product
of strength of one rivet in shear and the number of rivets on each side of the
joint. It is given by
Ps = strength of a rivet in shearing × Number of rivets
on each side of joint ...(i)
When the rivets are subjected to single shear, then, the strength of one rivet
in single shear
⎛π ⎞
= ⎜ × d 2 × τvf ⎟ ...(ii)
⎝4 ⎠
Therefore, the strength of a riveted joint against shearing of rivets
⎛π ⎞
Ps = N × ⎜ d2 × τvf ⎟ ...(2.1)
⎝4 ⎠
where N = Number of rivets on each side of the joint
d = Gross-diameter of the rivet
τvf = Maximum permissible shear stress in the rivet.
64 DESIGN OF STEEL STRUCTURES

When the rivets are subjected to double shear, then, the strength of one rivet
in double shear
⎛ π ⎞
= ⎜ 2 × × d2 × τvf ⎟ ...(iii)
⎝ 4 ⎠
Therefore, the strength of a riveted joint against shearing of rivets
⎛ π ⎞
Ps = N × ⎜ 2 × d2 × τvf ⎟ ...(2.2)
⎝ 4 ⎠
When the strength of riveted joint against the shearing of the rivets is
determined per gauge width of the plate, then, the number of rivets, n per gauge
is taken into consideration. Therefore
For single shear of rivets
⎛π ⎞
Ps1 = n × ⎜ d2 × τvf ⎟ ...(2.3)
⎝4 ⎠
For double shear of rivets
⎛ π ⎞
Ps2 = n × ⎜ 2 × d 2 × τvf ⎟ ...(2.4)
⎝ 4 ⎠

2.10.2 Strength of Riveted Joint Against the Bearing of the Rivets

The strength of a riveted joint against the bearing of the rivets is equal to the
product of strength of one rivet in bearing and the number of rivets on each side
of the joint. It is given by
Pb = Strength of a rivet in bearing × Number of rivets
on each side of the joint ...(iv)
In case of lap joint, the strength of one rivet in bearing
= (d × t × σpb) ...(v)
where t = Thickness of the thinnest plate
d = Gross diameter of the rivet
σpb = Maximum permissible stress in the bearing for the rivet.
In case of butt joint, the total thickness of both cover plates or thickness of
main plate whichever is less is considered for determining the strength of a

Fig. 2.15 Bearing pressure on rivet

 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 65

rivet in the bearing. The bearing pressure acting on the rivet is radial as shown
in Fig. 2.15. The projected area is used in determining the strength of rivet in
bearing.
The strength of a riveted joint against the bearing of rivets
Pb = N × (d × t × σpb) ...(2.5)
When the strength of riveted joint against the bearing of rivets per gauge
width of the plate is taken into consideration, then, the number of rivets n is
also adopted per gauge. Therefore,
Pb1 = n × (d × t × σpb) ...(2.6)

2.10.3 Strength of Plate in Tearing

The strength of plate in tearing depends upon the resisting section of the plate.
The strength of plate in tearing is given by
Pt = Resisting section × σt f ...(2.7)
When the strength of plate in tearing per gauge width of the plate is found,
then, it is given by
Pt1 = (g – d) × t × σt f ...(2.8)
where g = Width of plate equal to the gauge of the rivets.
The strength of a riveted joint is the least of Ps, Pb, and Pt. The strength of
riveted joint per gauge width of plate is the least of Ps1 , Pb1 and Pt1 .
A riveted lap joint is shown in Fig. 2.16 (a). A riveted but joint is shown in
Fig. 2.16 (b).

b p
P P

p
(a ) L ap jo in t

p b
P P

p p p p

(b ) Bu tt jo in t
Fig. 2.16
66 DESIGN OF STEEL STRUCTURES

2.10.4 Strength of Riveted Lap Joint [Fig. 2.16 (a)]

The strength of riveted lap joint against shearing of the rivets
⎛π ⎞
Ps = N × ⎜ d2 × τvf ⎟
⎝4 ⎠

⎛π 2 ⎞
= 6 × ⎜ d × τvf ⎟ ...(vi)
⎝4 ⎠
The strength of riveted lap joint against bearing of rivets
Pb = N × (d × t × σpb)
= 6 × (d × t × σpb) ...(vii)
The strength of plate in tearing
Pt = (b – 3d) × t × σt f ...(viii)
The strength of riveted lap joint is the least of Ps · Pb and Pt . The strength of
riveted lap joint per gauge width is the least of Ps , Pb1 and Pt , whichever is
1 1
less, where
⎛π ⎞
Ps1 = n × ⎜ d2 × τvf ⎟
⎝4 ⎠
⎛π 2 ⎞
= 2 × ⎜ d × τvf ⎟ ...(ix)
⎝4 ⎠
Pb1 = n × (d × t × σpb)
= 2 × (d × t × σpb) ...(x)
and Pt1 = (g – d) × t × σtf ...(xi)

2.10.5 Strength of Riveted Butt Joint [Fig. 2.16 (b)]

The strength of riveted butt joint against shearing of rivets
⎛ π ⎞
P3 = N × ⎜ 2 d2 × τvf ⎟
⎝ 4 ⎠

⎛ π ⎞
= 9 × ⎜ 2 d2 × τvf ⎟ ...(xii)
⎝ 4 ⎠
The length of riveted butt joint against bearing of rivets
Pb = N × (d × t × σpb)
= 9 × (d × t × σpb) ...(xiii)
The strength of plate in tearing
Pt = (b – 3d) × t × σtf ...(xiv)
The strength of riveted butt joint is the least of Ps, Pb, and Pt. The strength of
rivet butt joint per gauge width is the least of Ps1 , Pb1 and Pt1 , whichever is
less, where
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 67

⎛π ⎞
Ps1 = n × ⎜ d2 × τvf ⎟
⎝4 ⎠

⎛π 2 ⎞
= 3 × ⎜ d × τvf ⎟ ...(xv)
⎝4 ⎠

Pb1 = n × (d × t × σpb)
= 3 × (d × t × σpb) ...(xvi)
and Pt1 = (g – d) × t × σtf ...(xvii)

2.11 EFFICIENCY OR PERCENTAGE STRENGTH OF RIVETED

JOINT
The efficiency of a joint is defined as the ratio of least strength of a riveted joint
to the strength of solid plate. It is known as percentage strength of riveted joint
as it is expressed in percentage.
Efficiency of riveted joint

⎛ Least strength of riveted joint ⎞

n = ⎜ ⎟ × 100
⎝ Strength of solid plate ⎠

⎛ Least of Ps , Pb or Pt ⎞
= ⎜
⎝ ⎟⎠ × 100 ...(2.9)
P
where, P = Strength of solid plate.

2.12 RIVET VALUE

The strength of a rivet in shearing and in bearing is computed and the lesser is
called the Rivet Value (R).

2.13 ASSUMPTIONS FOR DESIGN OF RIVETED JOINT

The procedure for design of a riveted joint is simplified by making the following
assumptions and by keeping in view the safety of the joint.
1. The load is assumed to be uniformly distributed among all the rivets.
2. The shear stress on a rivet is assumed to be uniformly distributed over
its gross area.
3. The bearing stress is assumed to be uniform between the contact
surfaces of plate and rivet.
4. The bending stress in a rivet is neglected.
5. The rivet hole is assumed to be completely filled by the rivet.
6. The stress in plate is assumed to be neglected.
7. The friction between plates is neglected.
68 DESIGN OF STEEL STRUCTURES

2.14 ARRANGEMENT OF RIVETS

The rivets in a riveted joint are arranged into two forms :
1. Chain Riveting; 2. Diamond Riveting.

2.14.1 Chain Riveting

In chain riveting, the rivets are arranged as shown in Fig. 2.17.
1 2 3 3 2 1

P1 b P1

1 2 3 3 2 1
Fig. 2.17 Chain riveting

In Fig. 2.17, 1–1, 2–2 and 3–3 show sections on either side of the joint. Section
1–1 is the critical section as at the other section, strength of rivets prior to that
section adds the strength of the joint at the section. The strength of plate in
tearing can be computed at this section as under:
Pt = (b – 3d) . t . σt
where b = width of the plate
d = gross diameter of the rivet
t = thickness of the plate
When the safe load carried by the joint is known, width of plate can be found
as follows :

⎛ Pt ⎞
b = ⎜ + 3d ⎟
σ
⎝ t × t ⎠

2.14.2 Diamond Riveting

In diamond riveting, rivets are arranged as shown in Fig. 2.18.
All the rivets are arranged symmetrically about the centre line of the plate.
Section 1–1 is the critical section. The strength of plate in tearing in diamond
riveting at the section 1–1 can be computed as under :
Pt = (b – d) . t . σt
When the safe load carried by the joint is known, width of the plate can be
found as follows :
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 69

⎛ P ⎞
b = ⎜ t + d⎟
σ
⎝ t × t ⎠
where b = width of the plate
d = gross diameter of rivet
t = thickness of plate
1 2 3 3 2 1

P1 b P1

1 2 3 3 2 1
Fig. 2.18 Diamond riveting

The strength of the plate at all various sections can be found as follows :
At Section 2–2. All the rivets are stressed uniformly. Hence, strength of the
plate at the section is
Pt = (b – 2d) . t . σt + strength of one rivet
At Section 3–3. Pt = (b – 3d). t . σt + strength of three rivets.
The failure of plate in this form of riveted joint at any other section than
section 1–1, can take place only when all rivets prior to that section have failed.
Usually failure of plate takes place at the section 1–1.
The width of the plate required in case of chain riveting is more than that in
diamond riveting by the twice the gross diameter of the rivet. In diamond riveting
there is saving of material and efficiency is more ; hence it is preferred to chain
riveting. The diamond riveting is used in bridge trusses generally.

2.15 SPECIFICATIONS FOR DESIGN OF RIVETED JOINT

1. Members meeting at joint. The centroidal axes of the members meeting at
a joint should intersect at one point, and if there is any eccentricity, adequate
resistance should be provided in the connection.
2. The centre of gravity of group of rivets should be on the line of action of
load whenever practicable.
3. Pitch. Minimum pitch. The distance between centres of adjacent rivets
should not be less than 2.5 times the gross diameter of the rivet.
Maximum pitch. (i) The maximum pitch should not exceed 12 t or 200 mm
whichever is less in compression member, and 16 t or 200 mm whenever is less
70 DESIGN OF STEEL STRUCTURES

in case of tension member, when the line of rivet lies in the direction of stress.
In the case of compression members in which the forces are transferred through
the butting faces, this distance shall not exceed 45 times the diameter of the
rivets for a distance from the abutting faces equal to 15 times the width of the
member.
(ii) The distance between centres of any two consecutive rivets in a line
adjacent and parallel to an edge of an outside plate shall not exceed (100 mm +
4 t) or 200 mm, whichever is less in compression or tension members.
(iii) When the rivets are staggered at equal intervals and the gauge does
not exceed 75 mm, the distances specified in para (i) above and (ii) between
centres of rivets may be increased by 50 per cent.
(iv) If the line of rivets (including tacking rivets) does lie in the direction of
stress, the maximum pitch should not exceed 32 t or 300 mm whichever is less
where t is the thickness of the thinner outside plate.
4. Edge distance. A minimum edge distance of approximately 15 times
the gross diameter of the rivet measured from the centre of the rivet hole is
provided in the riveted joint. Table 2.2 hereunder gives the minimum edge
distance as per recommendations of BIS in IS : 800–1984.

Table 2.2 Edge distance of holes

Edge Distance of Hole

Gross diameter of rivet Distance to sheared Distance to rolled
or hand flame cut machine flame cut
edge or planed edge
mm mm mm
13.5 and below 19 17
15.5 25 22
17.5 29 25
19.5 32 29
21.5 32 29
23.5 38 32
25.5 44 38
29.0 51 44
32.0 57 51
35.0 57 51
5. Rivets through packings. The rivets carrying calculated shear stress
through a packing greater than 6 mm thick shall be increased above number
required by normal calculations by 2.5 per cent for each 2 mm thickness of
packing. For double shear connections packed on both sides, the number of
additional rivets required shall be determined from the thickness of the thicker
packing. The additional rivets should preferably be placed in an extension of
the packing. When the properly fitted packings are subjected to direct
compression, then, the above mentioned specifications shall not apply.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 71

6. Long grip rivets. When the grip of rivets carrying calculated loads
exceeds 6 times the diameter of the holes, then, the rivets are subjected to bending
in addition to shear and bearing. The number of rivets required by normal
calculations shall be increased by not less than one percent for each additional
1.6 mm of grip, but the grip shall not exceed 8 times the diameter of the holes.
7. Rivet line distance. When two or more parts are connected together,
a line of rivet shall be provided at a distance of not more than 37 mm + 4 t from
the nearest edge where t is the thickness in mm of thinner outside plate. In case
steel work is not exposed to weather, this may be increased to 12 t.
8. Tacking rivets. When the maximum distance between centres of two
adjacent rivets connecting the members subjected to either compression or
tension exceeds the maximum pitch, then, the tacking rivets not subjected to
calculated stresses shall be used.
The tacking rivets shall have a pitch in line not exceeding 32 times the
thickness of the outside plate or 300 mm whichever is less. Wherever the plates
are exposed to the weather, the pitch in line shall not exceed 16 times the
thickness of the outside plate or 200 mm, whichever is less. In both cases, the
lines of rivets shall not be apart at a distance greater than these pitches.
For the design and construction composed of two flats, angles, channels or
tees in contact back to back or separated back to back by a distance not exceeding
the aggregate thickness of the connected parts, tacking rivets with solid distance
pieces where the parts are separated, shall be provided at a pitch in line not
exceeding 1000 mm.

2.16 DESIGN PROCEDURE FOR RIVETED JOINT

For the design of a lap joint or a butt joint the thickness of plates to be joined are
known and the joint is designed for the full strength of the plate. For the design
of a structural steel work, force (pull or push) to be transmitted by the joint is
known and riveted joint can be designed. Following are the usual steps for the
design of a riveted joint :
Step 1. The size of the rivet is determined for the Unwin’s formula as
d = 6.04 (t)1/2 ...(2.10)
where t = thickness of plate in mm
d = nominal diameter of rivet.
The diameter of the rivet computed is rounded off to available size of rivets.
The rivets are manufactured in nominal diameters of 12, 14, 16, 18, 20, 22, 24,
27, 30, 33, 36, 39, 42 and 48 mm.
In structural steel work, rivets of nominal diameter of 16, 18, 20 and 22 mm
are used. The nomial diameter of rivets to be used in a joint is assumed.
Step 2. The strengths of rivets in shearing and bearing are computed. The
working stresses in rivets and plates are adopted as per BIS. The rivet value R
is found. For designing lap joint or butt joint tearing strength of plate is
determined as under :
Pt = (g – d) . t . σtf
72 DESIGN OF STEEL STRUCTURES

where g = gauge of rivets to be adopted

t = thickness of plate
σtf = working stress in direct tension for plate.
The tearing strength of plate should not exceed the rivet value R (Ps or Pb
whichever is less) or
(g – d) . t . σtf ≤ R
From this relation gauge of the rivets is determined.
In structural steel work, force to be transmitted by the riveted joint and the
rivet value are known. Hence number of rivets required to be provided in the
joint can be computed, as follows :

⎛ Force ⎞
No. of rivets required in the joint = ⎜
⎝ Rivet Value ⎟⎠
The number of rivets thus obtained is provided on one side of the joint and an
equal number of rivets is provided on the other side of joint also.
For the design of joint in a tie member consisting of a flat, width/thickness of
the flat is known. The section is assumed to be reduced by rivet holes, depending
upon the arrangement of rivets to be provided. The strength of flat at the weakest
section is equated to the pull transmitted by the joint. For example, assuming
the section to be weakened by one rivet hole and also assuming that the thickness
of the flat is known, we have
(b – d) . t . σtf = P
where b = width of flat
t = thickness of flat
P = pull to be transmitted by the joint
σtf = working stress in tension in plate.
From the above equation width of the flat can be determined.
Example 2.1. A single riveted lap joint is used to connect plate 10 mm thick.
If 20 mm diameter rivets are used at 55 mm gauge, determine the strength of
joint and its efficiency.
Working stress in shear in rivets = 80 N/mm2 (MPa)
Working stress in bearing in rivets = 250 N/mm2 (MPa)
Working stress in axial tension in plates = 0.6 fy
fy = 260 N/mm2.
Solution
Step 1 :
Assume that power driven field rivets are used
Nominal diameter of rivet = 20 mm
Gross diameter of rivet = 21.5 mm
Step 2 : Strength of rivet
π 80
Strength of rivet in single shear = × (21.5)2 × kN
4 1000
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 73

Pt = 29.029 kN
250
Strength of rivet in bearing = 21.5 × 10 × kN
1000
Pb = 53.750 kN
Strength of plate in tension per gauge length
Pt = (g – d) . t . σt
260
= (55 – 21.5) × 10 × 0.6 × kN
1000
Pt = 52.260 kN
Step 3 : Strength of joint
Strength of the joint is minimum of Ps, Pb, or Pt.
Strength of joint = 29.029 kN
Step 4 : Efficiency of joint
Efficiency of joint
⎛ Strength of joint per pitch length ⎞
η= ⎜ ⎟ × 100
⎝ Strength of solid plate ⎠

⎛ 29.029 × 103 ⎞
η= ⎜ ⎟ × 100 = 33.8 %
⎝ 55 × 10 × 0.6 × 260 ⎠
Example 2.2. A double riveted double cover butt joint is used to connect plates
12 mm thick. Using Unwin’s formula, determine the diameter of rivet, rivet value,
gauge and efficiency of joint. Adopt the following stresses :
Working stress in shear in power driven rivets
= 100 N/mm2 (MPa)
Working stress in bearing in power driven rivets
= 300 N/mm2 (MPa)
For plates working stress in axis tension is 0.6 fv
fy = 260 N/mm2 (MPa).
Solution
Step 1 :
Nominal diameter of rivet from Unwin’s formula
d = 6.04(t)1/2
d = 6.04 (t)1/2
= 20.923 mm
Adopt nominal diameter of rivet = 22 mm
Gross diameter of rivet = 23.5 mm
Step 2 : Rivet value. Strength of rivet in double shear
2
π (23.5 ) × 100
2× × = 86.70 kN
4 1000
74 DESIGN OF STEEL STRUCTURES

Strength of rivet in bearing

⎛ 23.5 × 12 × 300 ⎞
⎜ ⎟ = 84.6 kN
⎝ 1000 ⎠
Rivet value R = 84.6 kN.
Step 3: Gauge of rivets
Let g be the gauge of rivets. Strength of plate in tension per gauge length
Pt = (g – d) × t × pt
⎛ ( g − 23.5) × 12 × 0.6 × 260 ⎞
= ⎜ ⎟
⎝ 1000 ⎠
= 1.872 (g – 23.5) kN
In double riveted joint,
Strength of 2 rivets in shear Ps = (2 × 86.70) = 173.4 kN
Strength of 2 rivets in bearing Pb = (2 × 84.6) = 169.2 kN
The gauge of rivets can be computed by keeping Pt = Ps or Pb whichever is
less
1.872 (g – 23.5) = 169.2
⎛ 169.2 ⎞
(g – 23.5) = ⎜ ⎟ = 90.385
⎝ 1.872 ⎠
g = (90.35 + 23.5) = 113.85 mm
Adopt gauge, g = 100 mm
Step 4 : Efficiency of joint :
( g − d) (100 – 23.5) × 100
Efficiency of joint η = × 100 =
g 100
= 76.5 percent.
Example 2.3. A double cover butt joint is used to connect plates 16 mm thick.
Design the riveted joint and determine its efficiency.
Solution
Step 1 : Diameter of rivet :
Size of rivet, using Unwin’s formula
d = 6.04 (16)1/2 = 24.16 mm
Unwin’s formula gives higher value
Adopt nominal diameter of rivet = 22 mm
Gross diameter of rivet = 23.5 mm
Step 2 : Rivet value. In double cover butt joint, rivets are in double shear.
As per IS : 800–84
Shear stress for power driven rivet = 100 N/mm2
Bearing stress for power driven rivet = 300 N/mm2
Strength of plate in tension = 0.6 × 260 N/mm2
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 75

Strength of rivet in double shear

⎛ 2π ( 2 100 ⎞
Ps = ⎜ × 23.5 ) × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of rivet in bearing.

⎛ 23.5 × 16 × 300 ⎞
Pb = ⎜ ⎟ = 112.8 kN
⎝ 1000 ⎠
Step 3 : Gauge distance of rivet
Let g be the gauge of rivets
Strength of plate per gauge length
Pt = (g – d) . t . σt

= ⎛⎜ ( g − 23.5) × 12 × 0.6 × 260 ⎞⎟ kN

⎝ 1000 ⎠
= 2.496 (g – 23.5) kN
Keep strength of plate Pt @ ps or Pb whichever is less
2.496 × (g – 23.5) = 86.70

⎛ 86.70 ⎞
or g = ⎜ ⎟ = 58.235 mm
⎝ 2.496 ⎠
Adopt gauge, g = 55 mm
Adopt thickness of each cover plate

⎛ 5 ⎞
t = 10 mm ⎜ t @ × 16 ⎟ mm
⎝ 8 ⎠
Step 4 : Efficiency of joint

⎛ g −d ⎞ ⎛ 55 – 23.5 ⎞
η = ⎜ × 100 ⎟ = ⎜ ⎟ × 100
⎝ g ⎠ ⎝ 55 ⎠
= 57.273 %
Example 2.4. Determine the strength of a double cover butt joint used to
connect two flats 200 F 12. The thickness of each cover plate is 8 mm. Flats have
been joined by 9 rivets in chain riveting at a gauge of 60 mm (Fig. 2.19). What is
the efficiency of the joint ? Adopt working stresses in rivets and flats as per IS :
800–1984.
Solution
Step 1 :
Size of flat used = 200 F 12
Width of flat = 200 mm
Thickness of flat = 12 mm
Use power driven rivets
76 DESIGN OF STEEL STRUCTURES

Step 2: Strength of rivet

Strength of a rivet in double shear

⎛ π (23.5 )2 × 100 ⎞
2⋅⎜ × ⎟ = 86.70 kN
⎝4 1000 ⎠
Strength of a rivet in bearing
⎛ 12 × 23.5 × 300 ⎞
⎜ ⎟ = 84.6 kN
⎝ 1000 ⎠

40 m m

60 m m
2 00 m m
60 m m

40 m m

Fig. 2.19 Double cover butt joint

Step 3 : Strength of joint

Strength of joint in shear,
Ps = (9 × 86.7) = 780.3 kN
Strength of joint in bearing
⎛ 12×23.5×300 ⎞
⎜ = 84.6 kN ⎟
⎝ 1000 ⎠
pb = (9 × 84.6) = 761.4 kN
Strength of plate in tearing
Pt = (b – 3d) . t . σt
⎛ 200 – 2 × 23.5)× 12 × 6 × 260 ⎞
Pt = ⎜ ⎟
⎝ 1000 ⎠
= 242.42 kN
Strength of joint is Ps, Pb or Pt whichever is less
Pt = 242.42 kN
Step 4 : Efficiency of joint
⎛ P , P or Pt whicnever is less ⎞
η = ⎜ s b ⎟ ×100
⎝ Strength of solid plate ⎠

⎛ 242.42 × 100 ⎞
⎜ 200 × 12 × 0.6 × 260 ⎟ = 64.7489 %
⎜ ⎟
⎝ 1000 ⎠
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 77

Example 2.5. In a truss girder of a bridge, a diagonal consists of a 16 mm

thick flat and carries a pull of 750 kN and is connected to a gusset plate by a
double cover butt joint. The thickness of each cover plate is 8 mm. Determine the
number of rivets necessary and the width of the flat required. What is the efficiency
of the joint ? Sketch the joint.
Take working stress in shear in rivet = 100 N/mm2
Working stresses in bearing in rivet = 300 N/mm2
Working stresses in tension in plate = 0.6 × 260 N/mm2.
Solution

4 00 m m

Fig. 2.20 Double cover butt joint

Step 1 : Rivet value

Use 22 mm diameter of rivets
Strength of power driven shop rivets in double shear
⎛π 2 100 ⎞
2 × ⎜ × (23.5 ) × ⎟ = 86.70 kN
⎝4 1000 ⎠
Strength of power driven shop rivets in bearing
⎛ 300 ⎞
⎜ 23.5 × 16 × ⎟ = 112.8 kN
⎝ 1000 ⎠
Rivet value = 86.7 kN
Step 2 : Number of rivets
Number of rivets required to transmit pull of 750 kN
⎛ 750 ⎞
n = ⎜ ⎟ = 8.7 @ 9 rivets
⎝ 86.70 ⎠
Using diamond group of riveting, flat is weakened by one rivet hole.
Step 3 : Strength of plate
Strength of plate at Sec. 1–1 in tearing
⎛ 16 × 0.6 × 260 ⎞
Pt = (b – 23.5) × ⎜ ⎟
⎝ 1000 ⎠
= 2.496 (b – 23.5) kN
78 DESIGN OF STEEL STRUCTURES

P = 750 kN
2.496 (b – 23.5) = 750
⎛ 750 ⎞
∴ b = ⎜ + 2.496 × 23.5 ⎟ = 359.136 mm
⎝ 2.496 ⎠
Provide 400 mm width of diagonal member
Step 4 : Efficiency of joint

⎛ (b − d ) t ⋅ pt ⎞ ⎛ 400 – 23.5 ⎞
η = ⎜ ⎟=⎜ ⎟ × 100
⎝ b ⋅ t ⋅ pt ⎠ ⎝ 400 ⎠
= 94.125 %
Example 2.6. A bridge truss diagonal carries an axial pull of 500 kN. It is to
be connected to a guseet plate 22 mm thick by a double cover butt joint with 22
mm rivets. If the width of the flat tie bar is 250 mm, determine the thickness of
flat. Design the economical joint. Determine the efficiency of the joint. Adopt
working stress as per IS : 800–1984.
Solution

40 m m

85 m m

85 m m

40 m m

Fig. 2.21 Diamond riveted double cover butt joint

Step 1 : Diameter of rivet

Nominal diameter of rivet = 22 mm
Gross diameter of rivet = 23.5 mm
Step 2 : Rivet value
Adopt stresses from IS : 800–1984
Strength of power driven rivet in double shear

⎛π 2 100 ⎞
Ps = 2 ⋅ ⎜ × (23.5 ) × ⎟ = 86.70 kN
⎝4 1000 ⎠
Strength of power driven rivet in bearing

⎛ 23.5 × 22 × 300 ⎞
Pb = ⎜ ⎟ = 155.1 kN
⎝ 1000 ⎠
Rivet value, R = 86.70 kN
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 79

Step 3 : Number of rivets required

⎛ 500 ⎞
n = ⎜ ⎟ = 5.75 @ 6
⎝ 86.70 ⎠
Provide 6 rivets in diamond group of riveting for efficient joint.
Step 4 : Thickness of flat
Let thickness of flat be t mm
Strength of plate at weakest section (b – d) × t × σt = 500
⎛ (250 – 23.5)× t × 6 × 260 ⎞
or ⎜ ⎟ = 500
⎝ 1000 ⎠
∴ t = 14.151 mm
Adopt 16 mm thickness of flat
Keep 40 mm edge distance from centre of rivet and 85 mm distance between
centre to centre of rivet lines.
Step 5 : Efficiency of joint
⎛b−d ⎞
η = ⎜ ⎟ × 100
⎝ b ⎠

⎛ 250 – 23.5 ⎞
⎜ ⎟ × 100 = 90.6 %
⎝ 250 ⎠
The design of joint is shown in Fig. 2.21.

2.17 ECCENTRIC RIVETED CONNECTIONS

When the centre of gravity of a group of rivets does not lie on the line action of
the load, the connections are known as eccentric riveted connections.
The eccentric riveted connections are of two types :
1. When the line of action of load the group of rivets, are in vertical plane and
the centre of gravity of rivets is the centre of rotation, the eccentric riveted
connections are of type one.
2. When the line of action of load does not lie in the plane of the group of
rivets, and the line of rotation does not pass through the centre of gravity
of the group of rivets, the eccentric riveted connections are of type two.
Type 1. Figure 2.22 shows an eccentric riveted connection. A vertical load P
is acting at distance e from the centre of gravity of the group of rivets.
Consider a load P acting vertically downward and a load P acting vertically
upward, both applied along and line passing dirough the C.G., of the group of
rivets.
This is equivalent to
(i) an axial load (P) and
(ii) a couple (P × e)
This couple causes twisting moment. This is also called as torsional moment :
80 DESIGN OF STEEL STRUCTURES

Let n be the number of rivets in the bracket connection. The rivets are
subjected to direct shear and a moment.

P
e

Fig. 2.22 A bracket connection

The direction of resisting shear force in a rivet is vertically upward. Shear

force in any rivet is

⎛P⎞
F = ⎜ ⎟
⎝n⎠
The force resisting the twisting (torsional) moment in any rivet is proportional
to the distance (r) of the centre of the rivet from the C.G. of the group of the
rivets. The force acts in a direction perpendicular to the line joining the centre
of the rivet to the C.G. of the group of the rivets and gives resisting moments
about C.G. of group of rivets.

⎛F ⎞
F2 ∞ r, ∴ F2 = K . r ∴K= ⎜ 2⎟
⎝ r ⎠
Resisting moment because of force F2 about C.G. of the group of rivets is
F2 . r = K . r . r , or F2 . r = K . r2
Total resisting moment is ∑ F2 . r and is equal to the external moment P × e.
∑ F2 . r = ∑ K . r2, or ∑ F2 . r = K ∑ r2
F2 ⎛ P ⋅e ⋅r ⎞
P.e = ∑r2, ∴ F2 = ⎜ 2 ⎟
r ⎝ ∑r ⎠
The force resisting the twisting (torsional) moment is maximum in the rivet
at the extreme distance (rn) from the centre of gravity. The force F2 in such a
rivet is given by
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 81

⎛ P ⋅ e ⋅ rn ⎞
F2 = ⎜ 2 ⎟
⎝ ∑r ⎠
The direct shear force F1, and the force resisting the moment F2 in a rivet are
inclined at an angle θ. The resultant of these two forces can be found by
determining their vectorial sum as shown in Fig. 2.23.
Resultant force in a rivet because of these two forces is obtained as the
vectorial sum of two forces as under:
1/2
F = ⎡⎣ F12 + F22 + 2 F1 F2 cos θ ⎤⎦

1/ 2
⎡⎛ P ⎞2 ⎛ P ⋅ e ⋅ r ⎞2 P P ⋅ e ⋅ rn ⎤
n
or F = ⎢⎜ ⎟ + ⎜ 2 ⎟ +2 2
⋅ cos θ⎥
⎢⎣⎝ n ⎠ ⎝ ∑ r ⎠ n ∑r ⎥⎦

F2 F1

θ
θ

F2 F F1
γn
θ

G
(a ) (b )
F = re su lta n t o f tw o fo rc es vecto r sum

Fig. 2.23

The resultant force F in a rivet caused by an eccentric load should not exceed
the rivet value R.
When the rivets are arranged in a narrow strip, the vertical distances may be
taken as the distance to the centre of the rivets from the C.G. of the group of
rivets.
Resultant force in a rivet under the above assumption is then
1/2
⎡ ⎛ P ⎞2 ⎛ P ⋅ e ⋅ y ⎞2 P P ⋅ e ⋅ yn ⎤
n
F = ⎢⎜ ⎟ + ⎜ 2 ⎟ +2 2
⋅ cos θ ⎥ ...(2.11)
⎣⎢⎝ n ⎠ ⎝ ∑ y ⎠ n ∑y ⎦⎥
where y1, y2 etc. are the vertical distances between the centres of rivets and
C.G. of group of rivets. yn is the distance to centre of extreme rivets from, the
C.G. of group of rivets.
The direction of the applied load (P) may be at some inclination (θ) with the
vertical as shown in Fig. 2.24.
82 DESIGN OF STEEL STRUCTURES

Fig. 2.24 Inclined load acting in a bracket connection

1 p
2 f´

p ( n –1 ) p

p N e utra l axis
2 np
3
p
np

1 p
2
f

Fig. 2.25 Stress diagram for rivets

The eccentricity is computed. The direction of shear force in a rivet is parallel

and opposite to the direction of the applied load. The direction of force in a rivet
resisting the external moment (P.e) is in a direction perpendicular to the line
joining the centre of rivet and centre of gravity G of the group of rivets. The
maximum stress in the rivet most remotely placed from the C.G. of the group of
rivets is found as already indicated. Its value should not exceed the rivet value
(R).
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 83

1. Design of bracket connection (Type 1)

When number of rivets in a bracket connection is known, the maximum load
which can be applied at a given eccentricity can be determined. For the design
of a bracket connection, the number of rivets required in the bracket connection
for resisting shear and moment under an eccentrically applied load has to be
assumed. A bracket connection can, however, be designed more conveniently by
the procedure discussed hereunder:
Let n number of rivets be uniformly spaced at a distance ‘p’. The stress in a
rivet is assumed to be proportional to its distance from the neutral axis. Then
stress diagram for the rivets would then be same as for rectangular beam of np
depth. In Fig. 2.25 strips have been shown for the stresses in the rivets. The
maximum stress in the extreme rivet should not exceed the rivet value R.
Average stress per unit depth
⎛R⎞
f´ = ⎜ ⎟
⎝ p⎠
Maximum stress on equivalent rectangular beam is
R ⎛ n ⎞
f = ×⎜ ⎟
P ⎝ n −1 ⎠
The total force below the neutral axis is equal to the intensity of stress
multiplied by the area of stress diagram above neutral axis.
1 ⎛ R ⎞ ⎛ n ⎞ ⎛ np ⎞
F = × ⋅ ⋅⎜ ⎟
2 ⎜⎝ p ⎟⎠ ⎜⎝ ( n − 1 ⎟⎠ ⎝ 2 ⎠
The total force below the neutral axis is also equal to this vatae and acts in
opposite direction. These two forces form a couple and resist the external moment
(M).
M = Force × Lever arm n
1 ⎛R n np ⎞ 2
or M = × × × × np
2 ⎜⎝ P (n − 1) 2 ⎟⎠ 3

Rpn2 ⎛ n ⎞
or M = ×⎜ ⎟
6 ⎝ (n − 1) ⎠
1/2
⎡ 6 M (n − 1) ⎤
n = ⎢ ×
⎣ Rp n ⎥⎦

⎛ n −1 ⎞
The factor ⎜ ⎟ may be assumed to be nearly 1. This assumption would be
⎝ n ⎠
always on the safe side.
1/2
⎡ 6M ⎤
n = ⎢ ⎥ ...(2.12)
⎣ Rp ⎦
The number of rivets computed from this expression is essential for resisting
the external moment. For the purpose of resisting shear, a revision will seldom
84 DESIGN OF STEEL STRUCTURES

be found necessary. The size of rivets and pitch is may be assumed and a rivet
value (R) can then be computed. The edge distance in the design of bracket
connection is kept equal to one-half the pitch.
For two vertical lines of rivets, a rivet value of 2R is used and thus number of
rivets is calculated for one row of rivets. These rivets are arranged for the bracket
connection, and design is checked by general method.
2. Design of bracket connection (Type 2)
In this type of eccentric riveted connections, a rivet is subjected to tension, in
addition to direct shear in the bracket connection as shown in Fig. 2.26.
A tee-section has been attached to a stanchion by riveting. The portion of tee-
section above topmost rivet does not provide any resistance. The depth of the
bracket is measured from bottom of the bracket upto the centre of the top rivet.

h h

N e utral N .A .
A xis 2 1h
21
h
7

(a ) (b ) (c)

Fig. 2.26 A bracket connection

The position of the line of rotation depends upon the degree of rigidity of
bracket connection. In practice, the line of rotation (neutral axis) is assumed to
1
lie at a height of th of the depth of the bracket measured from the bottom edge
7
of the bracket. The rivets above the line of rotation are in tension in addition to
being in direct shear, while below this line the tee-section is in compression
against the column flange. The stress diagram has been shown in Fig. 2.26 (c).
The tensile force in a rivet Ft is proportional to its distance y from the line of
rotation.

⎛F ⎞
Ft ∞ y, ∴ Ft = ky :. k = ⎜ t ⎟
⎝ y ⎠
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 85

Moment of resistance due to this tensile force is

Ft . y = k . y . y, ∴ Ft . y = k . y2
Total moment of resistance (M) provided by rivets in tension
⎛ Ft ⎞
M´ = ∑k . y2, ∴ M´ ⎜ ⎟ ∑y 2
⎝ y ⎠

⎛ M´y ⎞
or Ft = ⎜ 2 ⎟
⎝ ∑y ⎠
Let y1, y2, ....... yn represent distances of the centres of the rivets from the line
of rotation. The tensile forces in these rivets will be as given hereunder. It is
obvious that tensile force in the topmost rivet which is at a distance yn, is
maximum.
⎛ M ´ y1 ⎞ ⎛ M ´ y2 ⎞ ⎛ M ´ yn ⎞
Ft = ⎜ 2 ⎟
, Ft2 = ⎜ 2 ⎟
, Ftn = ⎜ 2 ⎟
⎝ ∑y ⎠ ⎝ ∑y ⎠ ⎝ ∑y ⎠
1

(max)
⎛ M ´ ∑y ⎞
Total tensile force, Ft = ⎜ 2 ⎟
⎝ ∑y ⎠
For equilibrium, total tensile force is equal to total compressive force. Total
compressive force
⎛ M ´ ∑y ⎞
C = ⎜ 2 ⎟
⎝ ∑y ⎠
This compressive force acts at the centroid of stress distribution diagram C
viz., at a distance
⎛2⎞ ⎛h⎞ ⎛ 2h ⎞
⎜ ⎟×⎜ ⎟ = ⎜ ⎟
⎝3⎠ ⎝7 ⎠ ⎝ 21 ⎠
Taking the moment about the line of rotation (neutral axis).
External moment = Moment resisted by rivets in tension
+ Moment resisted by area of tee-section
in compression.
⎛ 2h ⎞ ⎡ M ´∑ y 2h ⎤
M = ⎜ M ´+C ⋅ ⎟, ∴ M = ⎢ M ´+ ⋅ ⎥
⎝ 21 ⎠ ⎣ ∑ y2 21 ⎦

⎡ M ⎤
or M´ = ⎢ ⎥ ...(2.13)
2h ∑ y
⎢1 + ⋅ ⎥
⎢⎣ 21 ∑ y2 ⎥⎦
This equation gives the moment resisted by the rivets in tension from which
maximum tensile force in the extreme rivet Ft can be calculated. Tensile stress:
86 DESIGN OF STEEL STRUCTURES

⎛F ⎞
σtf.cal = ⎜ t ⎟
⎝A⎠
where A = gross cross-sectional area of rivet. Direct shear force in any rivet
is
⎛ p⎞
F = ⎜ ⎟
⎝n⎠

⎛F ⎞
and shear stress fs = ⎜ ⎟
⎝A⎠
The rivets subjected to shear and externally applied tensile force, should be
so proportioned that the quantity

⎡ τvf .cal τtf .cal ⎤

⎢ + ⎥ does not exceed 1.4
⎣ τvf τtf ⎦

where
τvf.cal = Actual shear stress in rivet
σtf.cal = Actual tensile stress in rivet
τvf = Working shear stress in rivet.
τtf = Working tensile stress in rivet.
The stress in rivet is checked from the above equation.
Design procedure for bracket connection (Type 2)
In the design of bracket connection of this type, following are the usual steps
for the design.
Step 1. The nominal diameter of rivet to be used in the bracket connection is
assumed. Working stress in the rivet in tension is adopted as per IS : 800–1984
and the rivet value is calculated.
Step 2. The pitch of 2.5 times to 3 times the nominal diameter of rivet
approximately is adopted for the rivets. The rivets are provided in two vertical
rows. The number of rivets necessary for one row is computed from Eq. 2.12, by
substituting rivet value of 2R in the expression, as the rivets are accommodated
in two vertical rows. The rivet value ‘R’ is calculated in direct tension.
Step 3. The number of rivets provided in the design is checked by calculating
actual shear and tensile stresses in the rivets and quantity

⎡ τvf .cal σtf .cal ⎤

⎢ + ⎥ does not exceed 1.4.
⎣ τvf σ ft ⎦

Example 2.7 A riveted steel bracket connection has 22 mm diameter rivets 12

in number arranged as shown in Fig. 2.27. Determine the load P so that allowable
stress in the extremely loaded rivet is just reached. The safe permissible stress in
bearing in rivet is 300 N/mm2 (MPa) and safe permissible stress in shearing in
rivet is 100 N/mm2 (MPa).
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 87

Solution
Step 1. Diameter of rivets. Nominal diameter of rivet = 22 mm Gross
diameter of rivet = 23.5 mm
Step 2. Rivet value (R). Strength of power driven shop rivet in single shear

⎛ π (23.5)2 × 100 ⎞
⎜ × ⎟ = 43.35 kN
⎝4 1000 ⎠
Assume that strength of rivet in bearing is greater than strength of rivet in
single shear
Rivet value = 43.35 kN

P
2 50 m m
4 00 m m

Fig. 2.27 Rivets in a bracket connection

Step 3. Direct shear in each rivet. Number of rivets in bracket connection

is 12
⎛P⎞
F1 = ⎜ ⎟ kN
⎝ 12 ⎠
External moment acting on bracket connection
M = (P × 250) kN-mm
Force in extreme rivet due to twisting (torsional) moment

⎛ P ⋅ 250 ⋅ rn ⎞
F2 = ⎜ ⎟ kN
⎝ ∑r 2 ⎠
Distance of centre of extreme rivet from C.G. of group of rivets
rn = (502 + 2002)1/2 = 206.155 mm
Horizontal distance of each rivet from C.G. of group of rivtts
= 50 mm
Vertical distances of rivets
Rivets in first row above C.G.= 40 mm
88 DESIGN OF STEEL STRUCTURES

Rivets in second row C.G. = 120 mm

Rivets in topmost row C.G. = 200 mm
Other rivets are symmetrically placed
∑r2 = [ 4 (52 + 42) + 4 (52 + 122) + 4 (52 + 202] × 100 m .
= 2540 × 100 mm2

F2 = ⎛ P × 250 × 206.2 ⎞ = 0.205 P kN

⎜ ⎟
⎝ 2540 × 100 ⎠
Resultant of two forces
1/2
F = ⎡⎣ F12 + F22 + 2 F1 F2 cos θ ⎤⎦

⎛ 50 ⎞
From Fig. 2.27, cos θ = ⎜ ⎟
⎝ 206.2 ⎠
1/2
⎡ ⎛ P ⎞2 P 0.205P × 50 ⎤
∴ F = ⎢⎜ ⎟ + (0.205P )2 + 2 × × ⎥
⎣⎝ 12 ⎠ 12 206.2 ⎦
∴ F = 0.239P kN
As the resultant force in the extreme rivet is not to exceed the rivet value.
Therefore the resultant force, F may be equated to rivet, R
or 0.239 P = 43.35 kN, P =181.38 kN
Hence the maximum load which can be applied is 181.38 kN.
Example 2.8 Design riveted connections for a bracket as shown in Fig. 2.28
carrying an eccentric load of 200 kN at a distance of 350 mm from the centre
line. Adopt working stresses for rivets as per IS : 800–1984.
Solution
Design :
Step 1. Rivet value. Use 22 mm normal diameter rivet. Gross diameter or
rivet
= 23.5 mm
Adopt working stresses given for the power driven rivets from IS : 800–1984.
Strength of rivet in single shear
⎛ π (23.5)2 × 100 ⎞
⎜ × ⎟ = 43.35 kN
⎝4 1000 ⎠
Assume thickness of gusset plate as 14 mm
Thickness of flange of stanchion, HB 300, @ 0.588 kN/m is 10.6 mm.
Bearing strength of rivet
⎛ 23.5 × 10.6 × 300 ⎞
⎜ ⎟ = 74.73 kN.
⎝ 1000 ⎠
Rivet value R = 43.35 kN
Step 2. Number of rivets. Provide rivet in two vertical rows and at 55 mm
pitch then rivet value is 2R
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 89

1/2 1/2
⎡ 6M ⎤ ⎡ 6 × 200 × 350 ⎤
n = ⎢ ⎥ = ⎢ = 9.3
⎣ P (2R ) ⎦ ⎣ 55 × 2 × 43.35 ⎥⎦
Provide 10 rivets in one vertical row.
Distance between centre to centre of rivets in vertical rows for HB 300,
@ 0.588 kN/m from steel section tables in 140 mm.
Step 3. Check. C.G. of group of rivets is midway.
As rivets are in a narrow strip,consider the vertical distance as the distance
of the centre of rivet from the C.G. of group of rivets.
∑y 2 = [4(2.752 + 8.252 +13.752 +19.25 + 24.752] × 102
or ∑ y2 = 4989.37×102 mm2
Force in the extreme rivet, resisting the torsional moment

⎛ P × e ⋅ yn ⎞ ⎛ 200 × 350 × 24.75 × 10 ⎞

F2 = ⎜ 2 ⎟ = ⎜⎝ ⎟
⎝ ∑y ⎠ 4989.37 × 100 ⎠
= 34.72 kN.

2 00 kN
3 50 m m

IS H B 3 00
@ 0 .58 8 kN /m

Fig. 2.28

Direct shear stress in the rivet

⎛ 200 ⎞
F1 = ⎜ ⎟ = 10 kN
⎝ 20 ⎠
Resultant stress is obtained by vectorial sum of two stresses
90 DESIGN OF STEEL STRUCTURES

1/2
⎡ 2 2 70 ⎤
F = ⎢10 + 34.72 + 2 × 10 × 34.72 ×
⎣ 206.2 ⎥⎦
= 39.258 kN < 43.35 kN (Rivet value)
Hence, the design is satisfactory.
Example 2.9 The flange of a tee-section of 200 mm × 200 mm is riveted to the
flange of a rolled steel column of I-section to form a bracket. It carries a vertical
load of 280 kN at a distance of 200 mm from the face as shown in Fig. 2.29. Find
a suitable depth for the bracket and design the riveted connections, connecting
the tee-section with the flange of column.
4 00 m m 2 80 kN

2 00 m m

Tee -section
I - S ectio n

10 m m

10 m m
20 0 m m

10 m m

4 00 m m

Fig. 2.29

Solution
Design :
Step 1. Rivet value. Use 22 mm nominal diameter power driven rivets.
Gross diameter of rivet
= 23.5 mm ...(i)
Rivets are subjected to tension and hence rivet value shall be the strength of
rivet in tension. Gross area of rivet
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 91

⎛π 2⎞
⎜ × 23.5 ⎟ = 433.52 mm2 ...(ii)
⎝4 ⎠
Allowable stress in axial tension for power driven rivets
= 100 N/mm2 (MPa) ...(iii)
Rivet value, R = ⎛⎜ 433.52 × 100 ⎞⎟ 43.35 kN ...(iv)
⎝ 1000 ⎠
Step 2. Number of Rivets. The rivets in the bracket connection are used in
the two vertical rows. Adopting a pitch of 65 mm the number of rivets in one
vertical row is given by
1/2 1/2
⎡ 6M ⎤ ⎡ 6 × 280 × 200 ⎤
n ⎢⎣ P × 2 R ⎥⎦ = ⎢⎣ 65 × 2 × 43.35 ⎥⎦ = 7.72
= ...(v)

The number of rivets provided in each vertical row

n = 9 rivets ...(vi)
The edge distance in bracket connection is kept equal to half the pitch, i.e.,
32.5 mm.
Step 3. Check. Depth of bracket
h = (32.5 + 8 × 65) = 552.5 mm ...(vii)
h
The line of rotation is assumed at a height above the bottom edge of the
7
bracket
h ⎛ 552.5 ⎞
= ⎜ ⎟ = 78.9 mm ...(viii)
7 ⎝ 7 ⎠
The position of line of rotation is between first and second horizontal rows of
rivets from the bottom. For both the rows of rivets.
∑y = [2 (1.86 + 8.36 +14.86 + 21.36 + 27.86 + 34.36 +.40.86 + 47.36)] × 10 mm
...(ix)
or ∑y = 3937.6 mm
∑y2 = [2(1.862 + 8.362 + 14.862 + 21.362 + 27.862 + 34.362 + 40.862
+ 47.362)] × 102 mm2 ...(x)
or ∑y2 = 13234.92 × 100 mm2
Moment to be resisted by the rivets subjected to tension

⎡ M ⎤ ⎛ 280 × 200 ⎞
M´ = ⎢ ⎥ = ⎜
2h ∑ y 2 × 552.5 3937.6 ⎟
⎢1 + ⋅ ⎥ ⎜⎝ 1 + × ⎟
⎣⎢ 21 ∑ y2 ⎦⎥ 21 13234.92 × 100 ⎠
M´ = 48500 mm-kN ...(xi)
Tensile stress in the topmost rivet
⎛ M ´ × yn ⎞ ⎛ 48500 × 473.6 × 1000 ⎞
σvf.cal = ⎜ = ⎜ ⎟
⎝ A × ∑ y2 ⎟⎠ ⎝ 433.52 × 13234.94 × 100 ⎠
92 DESIGN OF STEEL STRUCTURES

= 40.93 N/mm2 ...(xii)

Actual shear stress in the rivet
⎛ 280 × 1000 ⎞
τvf.cal = ⎜ ⎟ = 35.88 N/mm2 ...(xiii)
⎝ 433.52 × 2 × 9 ⎠
The rivets are subjected to combined stress (i.e., tensile stress and shear
stress). Substituting the values of these stresses

⎡ τtf .cal τvf .cal ⎤ ⎡ 40.03 35.88 ⎤

⎢ + ⎥ = ⎢ + = 0.7591 < 1.4
⎣ σtf τvf ⎦ ⎣ 100 100 ⎥⎦

Hence, the design is satisfactory.

Example 2.10 A bracket connection is shown in Fig. 2.30. Check the design
of all fasteners in this bracket.
2 IS A 12 5 x 75 x 1 0 m m
80 80 80 80 40 m m

80
mm
80

80

80
1 4 m m p la te
80

80 2 2 m m p ow e r
d riven sho p rivets
80
2 IS A 12 5 x 75 x 1 0 m m

IS H B 4 50
@ 11 78 .9 N /m

Fig. 2.30

Solution
Step 1. Rivet Value
Nominal diameter of rivets = 22 mm
Gross diameter of rivets = 23.5 mm
Strength of power-driven rivets in double shear
⎡ π 2 100 ⎤
⎢⎣2 × 2 × (23.5) × 1000 ⎥⎦ = 86.70 kN
Strength of power driven shop rivets in bearing
⎛ 300 ⎞
⎜ 23.5 × 14 × ⎟ = 98.70 kN
⎝ 1000 ⎠
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 93

Rivet value R = 86.70 kN

Step 2. Rivets in line AA
Consider rivets in line AA.
These rivets are subjected to direct shear and bending and can be considered
as subjected to an axial load of P = 125 kN and a moment M = 125 × 80 mm-kN
as shown in Fig. 2.31.
P

Fig. 2.31

1. Direct shear in the rivet

P ⎛ 125 ⎞
=⎜ ⎟ = 25 kN
5 ⎝ 5 ⎠
2. Maximum force in the extreme rivet due to torsional moment

⎛ P ⋅ e ⋅ rn ⎞ ⎛ 125 × 80 × 160 ⎞
F2 ⎜ 2 ⎟ = ⎜ 2 2 ⎟ = 25 kN
⎝ ∑r ⎠ ⎝ 2(8 + 16 ) × 100 ⎠
3. Resultant force acting in the rivet (by vector sum of two forces)
R = (25 + 25) = 50 kN< 86.70 kN Rivet value. Hence satisfactory.
Step 3. Rivets in line BB
Consider rivets in line BB.
The eccentric load is acting in the plane of the rivets. The rivets are subjected
to direct shear and bending, and can be considered as subjected to an axial load
of P1 = 125 kN and a moment M1 = 125 × 240 mm-kN as shown in Fig. 23.2.
1. Direct shear force in the rivet
125
F = = 15.625 kN
8
2. Maximum force in the extreme rivet due to bending moment

⎛ P ⋅e ⋅r ⎞
F = ⎜ 1 12 n ⎟
⎝ ∑r ⎠

⎛ 125 × 240 × 280 1 ⎞

F = ⎜ 2 2 2 2
× ⎟
⎝ 2(4 + 12 + 20 + 28 ) 100 ⎠
= 31.25 kN
94 DESIGN OF STEEL STRUCTURES

3. Resultant force in the rivet

R = (15.6252 + 31.252)1/2 = 34.94 kN
Rivet value. Hence, satisfactory.
Consider rivets, connecting angle and the column section. These rivets are
subjected to direct shear and tension. The rivets have been provided on both the
sides.
Height of bracket = 600 mm

CC

Fig. 2.32

Neutral axis of the rivet line is assumed at 1/7th the height of bracket above
the bottom of the bracket.
Height of neutral axis
1
× 600 = 85.7 mm
7
For both rows of rivets
∑y = [2(3.43 + 11.43 +19.43 +27.43 + 35.43 + 43.43 + 51.43)] × 10
= 2 × l92.01× 10 mm
∑y 2 = [2(3.432 + 11.432 + 19.432 + 27.432 + 35.432 + 43.432 + 51.432)] × 100
= 2 × 7058.78 ×100 mm2
Moment to be resisted by the rivets in tension from Eq. 2.13

M
M´ =
⎛ 2h ∑y ⎞
⎜1 + 21 ⋅ ⎟
⎝ ∑ y2 ⎠
The gauge distance for rivet for ISA 125 mm × 75 mm × 10 mm 125 mm
length of leg from ISI Handbook No. 1 is 75 mm.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 95

The eccentricity of the load measured from the face of column

e = (75 + 80 + 80 + 80) = 315 mm
⎛ 125 × 315 ⎞
M´ = ⎜
2 × 600 2 × 192.01 × 10 ⎟
⎜1 + × ⎟
⎝ 21 2 × 7058.78 × 100 ⎠
= 35650 mm-kN
Tensile force resisted by top rivet
⎛ M ´× yn ⎞ ⎛ 35650 × 514.3 ⎞
Ft = ⎜ 2 ⎟ = ⎜ ⎟
⎝ ∑ y ⎠ ⎝ 2 × 7058.78 × 100 ⎠
= 12.950 kN
Actual tensile stress in the rivet

⎛ 12.950 × 1000 ⎞
σtf.cal = = 29.872 N/mm2
⎜ π 2 ⎟
⎜⎝ × (23.5) ⎟
4 ⎠
Allowable axial tensile stress in the power driven rivets
σtf = 100 N/mm2 (MPa)
Actual shear stress in the rivet
⎛ 125 × 1000 ⎞
τvf.cal = ⎜ ⎟ = 18.02 N/mm
2
π
⎜ 2 × 8 × × 23.52 ⎟
⎝ 4 ⎠
Allowable shear stress in the power driven rivets is
τvf = 100 N/mm2
⎛ σtf .cal τvf .cal ⎞ ⎛ 29.872 ⎞ ⎛ 18.02 ⎞
∴ ⎜ σ + = ⎜ ⎟+⎜ ⎟ = 0.479 < 1.4
⎝ tf τvf ⎟⎠ ⎝ 100 ⎠ ⎝ 100 ⎠
Hence, satisfactory.
Example 2.11 A bracket connection is shown in Fig. 2.33. The thickness of
plate attached to the flange of a column section MB 200, @ 25.4 kg/m is 10 mm.
The plate carries a load of 20 kN inclined 60° with the horizontal at a distance
160 mm from the vertical through C.G. of group of the rivets. Design the bracket
connection.
Solution
Design :
Step 1. Distance of line of action of load from C.G. of rivets
The inclination of load P with the horizontal is 60°. Its inclination with the
vertical is 30°. The prolonged direction of load P intersects the vertical through
C.G. of group of the rivets at O. Consider the triangle ABC.
∠ABC = 60°, ∠BAC = 90°, ∠ACB = 30°, AB = 160 mm
∴ AC = AB tan 60° = 160 × 3 = 276.8 mm ...(i)
96 DESIGN OF STEEL STRUCTURES

The eccentricity, e of the load P from the C.G. of group of rivets O, is OD.
Consider the triangle OCD. The angle ODC is 90°
OC = AC – AO = 2768.40 = 236.8 mm ...(ii)
1
OD = OC sin 30° = 236.8 × 2 = 118.4 mm ...(iii)
Distance of line of action of load from C.G. of group of rivets
∴ e = 118.4 mm ...(iv)

P = 2 0 kN

60 m m
55
mm
6 0°
30 m m
2 7.5 m m

2 7.5 m m

30 m m
1 0 m m th ick
55
mm
1 00 m m

M B 2 00
@ 2 5.4 kg/m
C

1 00 m m

Fig. 2.33

The rivets are subjected to direct shear and twisting moment. The number of
rivets in the bracket, N is 4.
Step 2. Force due to direct shear
Force in each rivet due to direct shear
⎛ P 20 ⎞
⎜ F1 = = ⎟ = 5 kN ...(v)
⎝ 4 4 ⎠
Step 3. Force due to twisting moment
All the corner rivets are at equal distance from the C.G. of group of rivets, O.
The distance of corner rivet from O
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 97

r = 27.5 2 = 27.5 ×1.414 = 38.89 mm ...(vi)

∑γ2 = (4 × 38.892) = 6050 mm2 ...(vii)
Forces in the corner rivets due to twisting moment are equal. Force in each
corner rivet due to twisting moment
⎛ Mr P ⋅ e ⋅ r ⎞
F2 = ⎜ 2 = ⎟
⎝ ∑r ∑ γ2 ⎠

⎛ 20 × 118.4 × 38.89 ⎞
or F2 = ⎜ ⎟ = 15.22 kN ...(viii)
⎝ 6050 ⎠
The direction of force F1 is exactly opposite to the direction of load P. The
direction of force F2 is perpendicular to the radius vector and opposite to the
twisting moment. The resultant force on a rivet is equal to vector sum of F1 and
F2. The vector sum of F1 and F2 is maximum when the included angle θ, between
these two forces is the least. The directions of force F1 and F2 on each rivet are
marked as shown in Fig. 2.34.

F1 F2 F1

9 0° 6 0° 9 0° 6 0°

F2
2 7.5 m m

F1 F1
F2
2 7.5 m m
4 5° 3 0° 1 5°
6 0° 6 0°
2 7.5 m m 2 7.5 m m
55 m m
F2

Fig. 2.34

Step 4. Resultant of forces

It is seen that the lower right rivet is subjected to maximum force. The angle
θ1 between F1 and F2 is least
θ1 = 15°, cos θ1 = 0.9659
Resultant of the force F1 and F2 is given by
F = [52 + 15.222 + 2 × 5 × 15.22 × 0.9659]1/2
= 20.09 kN
Strength of 16 mm nominal diameter rivet in single shear

⎛ π (17.5)2 × 100 ⎞
⎜ × ⎟ = 24.05 kN
⎝4 1000 ⎠
98 DESIGN OF STEEL STRUCTURES

Thickness of the plate is 10 mm. The thickness of flange of MB 200, @ 25.4 kg/
m is 10.8 mm. Therefore, the strength of rivet in bearing
⎛ 300 ⎞
⎜17.5 × 10 × ⎟ = 52.5 kN
⎝ 1000 ⎠
Rivet value of 16 mm nominal diameter rivet
R = 24.05 kN > 20.09 kN. Hence, unsafe.
Provide 16 mm diameter power driven rivets for the back connection.

2.18 BOLT AND PIN CONNECTIONS

In Sec. 2.1, it was mentioned that various types of connections are used for
connecting the structual members. The choice of type of fasteners depends on
connection strength required, space limitations of the connections, available
technicians to fabricate and erect the structure, service conditions and finally
the total cost of installation. These fasteners serve essentially the same function
in transferring loads from one component to another. The connections are named
after the type of fasteners used. In bolted connections, bolts and nuts are used.
In tension or compression members, bolts may be used like rivets for end
connections. The bolts may also be used for semi-rigid connections. The
requirements for bolts as regards the pitch and minimum edge distance are
same as those for the rivets.
In pin connections, pins are used. Ordinarily, the bolts are used for temporary
fastenings and erection. Occasionally bolts are used for permanent connections.
When structural members are subjected to shock or vibrations, then, bolted
connections are not used. When structures support machinery or rolling loads,
then also, bolted connections are not used for beam and beam or column
connections. The pin connections simplify the analysis of indeterminate
structures. The pin connections are also occasionally used in buildings to carry
extremely heavy column loads across auditoriums. The pin connections are used
in hinged arches and few other structures. The pin connection acts as hinge
where the moment is zero.

2.19 ADVANTAGES OF BOLTED CONNECTIONS

The advantages of bolted connections are as follows:
1. There is silence in preparing bolted connection.
In riveting, hammering is done. The hammering causes noise in the
riveting.
2. There is no risk of fire in bolted connection.
The rivets are made red hot in riveting and there is risk of fire.
3. The bolted connections may be done quickly in comparison to the riveting.
4. Though the cost of bolts is more than the cost of rivets, the bolted
connections are economical to use them than rivets because less persons
are required for installation, and the work proceeds quickly.
5. The bolted connections facilitate the erection because of ease with which
these connections can be done.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 99

2.20 DISADVANTAGES OF BOLTED CONNECTIONS

The following are disadvantages of bolted connections:
1. If bolted connections become loose, their strength reduces considerably.
2. The unfinished bolts are not uniform in diameter and they have less
strength.
3. The bolted connections have less strength when they are subjected to
axial tension, because area at root of thread is less.
4. Generally, the diameter of hole is kept 16 mm more than the nominal
diameter of black bolt. The bolt does not fill the hole and there remains
a clearance in bolted connections.

2.21 BOLTS
A bolt is a metal pin with a head formed at one end and the shank threaded at
the other in order to receive a nut.
The bolts for structural purposes are manufactured from mild steel and high
strength steel. A bolt consists of head and body as shown in Fig. 2.35. The body
of a bolt is termed as shank. The bolts are manufactured to different lengths to
suit different purposes. The size of bolt is expressed by the diameter of shank.
The shank is threaded at one end. A nut can be provided on the threaded end of
the bolt. The threaded portion of each bolt shall project through the nut at least
by one thread. In all cases, when the full bearing area of the bolt is to be developed,
the bolt shall be provided with a washer of sufficient thickness under the nut to
avoid any threaded portion of the bolt being within the thickness or the parts
bolted together. The structural bolts are classified according to type of shank
(unfinished or turned); material and strength (ordinary structural, or high
strength steel); shape of head and nut (square or horizontal, regular or heavy
duty); and pitch and fit of thread (standard, coarse or fine).
H e ad B o dy (sh an k)

Th re ad ed en d
Fig. 2.35 A bolt

The following are the various types of bolts used for structural purposes.

2.21.1 Unfinished Bolts

The finished bolts are also termed as ordinary bolts, common bolts, rough bolts
and back bolts. The unfinished bolts are manufactured from low carbon steel.
The unfinished bolts have square heads. The heads of unfinished bolts are made
by forging. The shanks of unfinished bolts are not uniform in diameter. The
unfinished bolts are designed for lower permissible stresses in shear and bending
than the rivets and turned bolts. The unfinished bolts are not used for structural
connections of members, which are subjected to shock or vibrations. The
unfinished bolts are used for ordinary field work and light load. The unfinished
or black bolts are used with square nuts. These are also used with hexagonal
100 DESIGN OF STEEL STRUCTURES

nuts. These bolts, screws and nuts are designated as illustrated in Table 2.3 as
per IS : 2585–1968.

Table 2.3 Designation system for black square bolts, screws and nuts

Description
Fasteners Thread Length Designation
size mm
Square bolts with square nut M 10 30 Square bolt M 10 × 30 N
Square bolt only M 10 30 Square bolt M 10 × 30
Square screw M 10 30 Square screw M 10 × 30
Square nut M10 — Square nut M 19
Square bolt with hexagonal nut M 10 30 Square bolt M 10 × 30 NH

A large portion of site connections are made with black bolts. From the point
of view of the erector, this makes for simplicity as the erection can be completed
in almost one operation. The additional cost of special labour for riveting and
the plant required is also provided.
The black bolts cannot, however, carry loads equal to those of a rivet with the
same diameter. In such cases, the turned bolts are used.

3 0° 3 0°

S q ua re bo lt S q ua re nu t

3 0°

Fig. 2.36 Square screw

The various dimensions of bolts, screws and nuts are as shown in Fig. 2.36.
The dimensions are as per IS : 2585–1968 and have been given in Tables 2.4
and 2.5. The diameters range from 6 to 39 mm.
Table 2.4 Dimensions for black square bolts, screws and nuts (as per IS : 2585–1968)
(Diamter range M-6 to M-18)
Size M6 M8 M10 M12 (M14) M16 (M18)
Nom. 6 8 10 13 14 16 18
d Max. 6.48 8.90 10.90 12.10 15.10 17.10 19.10
Min 5.70 7.64 9.64 11.57 13.57 15.57 17.57
s <19-h14 Nom. 10 13 17 9 22 24 27
Max. 10.00 13.00 17.00 19.00 22.00 24.00 27.00
>19-h15 Min. 9.64 12.67 16.57 18.48 21.16 23.16 26.16
c Min. 12.53 16.34 21.54 24.02 27.51 30.11 34.01
Nom. 4 5.5 7 8 9 10 12
k js 16 Max. 4.38 5.88 7.45 8.45 9.45 10.45 12.55
Min. 3.62 5.12 6.55 7.55 8.55 9.55 11.45
da Max. 7.2 10.2 12.2 15.2 17.2 19.2 21.2
r Min. 0.25 0.4 0.4 0.6 0.6 0.6 0.6
a Max. 4.0 4.5 5.0 6.0 7.5 7.5 9.0
b1 Min. 18 22 26 30 34 38 42
b b2 Min. — 28 32 36 40 44 48
DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS

b3 Min. — — — 40 53 57 61
Nom. 5 5 8 10 11 13 15
m fs16 Min. 5.38 6.95 7.55 10.45 11.55 13.55 15.55
Max. 4.62 6.05 8.45 9.55 10.45 12.45 14.45

Notes. 1. Sizes shown in brackets are of second preference.

2. Use b1 for l <130, b2 130 < l < 200, and b3, for l >200.
101
 102
Table 2.5 Dimensions for black square bolts, screws and nuts (as per is : 258 5–1968)
(Diamter range 20 to M-39)
Size M 20 (M 22) M 24 (M 27) M 30 (M 33) M 36 (M 39)
d Nom. 20 22 24 27 30 33 36 39
d Max. 21.50 23.30 25.30 28.30 31.30 34.60 37.60 40.60
Min. 19.481 21.48 23.48 26.48 29.48 32.38 35.38 38.38
s < 19-h14 Nom. 30 32 36 41 46 50 55 60
Max. 30.00 32.00 36.0 41.00 46.00 50.00 55.00 60.00
> 19-h51 Min. 29.16 31.00 35.00 40.00 45.00 49.00 53.80 58.80
e Min. 37.91 40.30 45.50 52.00 58.50 63.70 69.94 76.44
Nom. 13 14 15 17 19 21 23 25
k js 16 Max. 13.55 14.55 15.55 17.55 19.65 21.65 23.65 25.65
Min. 12.45 13.45 14.45 16.45 18.35 20.35 22.35 24.35
da Max. 24.4 26.4 28.4 32.4 35.4 38.4 42.4 45.4
r Min. 0.8 0.8 0.8 1.0 1.0 1.0 1.0 1.0
a Max. 9.0 9.0 1.1 — — — — —
DESIGN OF STEEL STRUCTURES

b b1 Min. 46 50 54 60 66 72 78 84
b2 Min. 52 56 60 66 72 78 84 90
b3 Min. 65 69 73 79 85 91 97 100
Nom. 16 18 19 22 24 26 29 31
m js Max. 16.55 18.55 19.55 22.65 24.65 26.65 29.65 31.80
Min. 15.45 17.45 18.35 21.35 23.35 25.35 28.35 30.20

Notes. 1. Sizes shown in brackets are of second preference.

2. Use b1 for l < 130, b2 for 130 < l < 200.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 103

2.21.2 Turned Bolts

Turned bolts are also termed as turned and fitted bolts. For these bolts, the
holes are made slightly larger than the diameter of the bolts. These bolts are
force fitted. Turned bolts are manufactured from hexagonal stock. The bars are
turned to required diameters. The threads in turned bolts are usually cut with
a die. If turned bolts are used in holes which are reamed, then turned bolts are
equally satisfactory as rivets for many connections. The flat face of nut, inner
face of head of bolt and both sides of washers are machined for better fitting.
The turned or black hexagonal bolts, nuts and lock nuts and screws, are
designated as illustrated in Table 2.6 per IS : 1363–1967.

Table 2.6 Designation system for bolts, screws, nuts and lock nuts

Description
Designation
Fastener Size Length
Belts with nuts M 16 70 Hex. bolt M 16 × 70 NL
and a lock nut
Bolts with a nut M 16 70 Hex. bolt M 16 × 70 N
Bolt only M 16 70 Hex. bolt M 16 × 70
Screw M 16 70 Hex. screw M 16 × 70
Nuts M 16 — Hex. nut M 16
Lock nut M 16 — Hex. lock nut M 16
Double chamfered nut M 16 — Hex. nut DCM 16
The various dimensions of black hexagonal bolts, screws, nuts and lock nuts
are as shown in Fig. 2.37. These dimensions are as per IS : 1363–1967 and have
been given in Tables 2.7 and 2.8. The diameters range from 6 mm to 24 mm.

3 0° 3 0° 3 0°

3 0°

H e xa go n bo lt H e xa go n D o ub le cha m fer
3 0° n ut h exag on n ut
3 0°

x 3 0°

(a ) H e xa go n scre w (b ) H e xa go n lo ck n ut

Fig. 2.37
Table 2.7 Dimensions for black grade hexagonal bolts, screws, nuts and lock nuts
(diamter range M-6 to m-18) as per is : 1363–1967 104
Size M6 M7 M8 M10 M12 M14 M16 M18
d Nom. 6 7 8 10 12 14 16 18
Max. 6.48 7.58 8.90 10.90 13.10 15.10 17.10 19.10
Min. 5.70 6.44 7.64 9.64 11.57 13.57 15.57 17.57
< 19-h14 Nom. 10 11 13 17 19 22 24 27
s Max. 10.00 11.00 13.00 17.00 19.00 22 .00 24.00 27.00
> 19-h15 Min. 9.64 10.57 12.57 16.57 18.48 2.16 23.16 26.16
e Min. 10.89 11.94 14.70 18.72 20.88 23.91 26.17 29.56
Nom. 40 50 55 70 80 90 10 12
k js16 Max. 4.38 5.38 5.88 7.45 8.45 945 10.45 12.55
Min. 3.62 4 62 5.12 6.55 7.55 8.55 9.55 11.45
da Max. 7.2 8.2 10.2 12.20 15.2 17.2 19.2 21.2
a Max. 40 45 45 50 60 75 75 90
b1 Min 18 20 22 26 30 34 38 42
b2 Min. — — 28 32 36 40 44 48
DESIGN OF STEEL STRUCTURES

b3 Min. — — — — 49 53 57 61
Nom. 5.0 5.5 6.5 8.0 10 11 13 15
m js 16 Max. 5.38 5.88 6.95 8.45 10.45 11.55 13.55 15.55
Min. 4.62 5.12 6.05 7.55 9.55 10.45 12.45 14.45
Nom. 3 3 4 5 7 8 8 9
f js 16 Max. 3.30 3.30 4.38 5.38 7.45 8.55 8.45 9.45
Min. 2.70 2.70 3.62 4.62 6.45 7.45 7.55 8.55
r Min. 0.25 0.25 0.4 0.4 0.6 0.6 0.6 0.6
Notes. 1. Sizes shown in brackets are second preference.
2. Use b1 for l < 130, b2 for 130 < l < 200, and b3 for l < 200.
Table 2.8 Dimensions for black grade hexagonal bolts, screws, nuts and lock nuts
(Diamter range M-20 to M-30) as per IS : 1363–1967
Size M20 M22 M 24 M27 M30 M33 M36 M39
Nom. 20 22 24 27 30 33 36 39
d Max. 21.30 23.30 25.30 28.30 31.30 34.00 37.60 40.00
Min. 19.48 21.48 23.48 26.48 29.48 32.38 35.38 38.38
< 19–h14 Nom. 30 32 36 41 46 50 55 60
s Max. 30.00 32.00 36.00 41.00 46.00 50.00 55.00 60.00
> 19–h15 Min. 29.16 31.00 35.00 40.00 45.00 49.00 53.80 58.80
e Min. 32.95 35.03 39.55 45.20 50.85 55.37 60.79 66.44
Nom. 13 14 15 17 19 21 23 25
k js16 Max. 13.55 14.55 15.5 17.55 19.65 21.65 23.65 25.65
Min. 12.45 13.45 14.45 16.45 18.35 20.35 22.35 24.35
da Max. 24.4 6.4 28.4 32.4 35.4 38.4 42.4 45.4
a Max. 9.0 9.0 11 — — — — —
b1 Min. 46 50 54 60 66 72 78 84
b2 Min. 52 56 60 66 72 78 84 90
b3 Min. 65 69 73 79 85 91 97 103
Nom. 16 18 19 22 24 26 29 31
m js 16 Max. 16.55 18.55 19.65 22.65 24.65 26.65 29.65 31.80
Min. 15.45 17.45 18.35 21.35 23.35 25.35 28.35 30.20
DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS

Nom. 9 10 12 12 12 14 14 15
f js 16 Max. 9.45 10.45 10.45 12.55 12.55 14.55 14.55 16.55
Min. 8.55 9.55 11.45 11.45 13.45 13.45 13.45 15.45
r Min. 0.8 0.8 1.0 1.0 1.0 1.0 1.0 1.0

Notes. 1. Sizes shown in brackets are of second preference.

105

2. Use b1 for l < 130, b2 for 130 < l < 200.

Table 2.9 Dimensions for high strength (tensile) friction grips bolts (as per IS : 3757–1966) 106

d M12 M16 M20 M22 M 24 M27 M 30 M33 M 36 M39

d1 Min. 20 25 30 34 39 44 48 53 58 63
b 1 30 38 46 50 54 60 66 72 78 84
2 — 44 52 56 60 66 72 78 84 90
c Nom. 0.4 0.6 0.8 0.8 0.8 0.8 1.0 1.0 1.0 1.0
e Max. 25.4 31.2 36.9 41.6 47.3 53.1 57.7 63.5 69.3 75.0
Nom. 8 10 13 14 15 17 19 21 23 25
k j6 Max. 8.45 10.45 13.55 14.55 15.15 17.55 19.65 21.65 23.65 25.65
Min. 7.55 9.55 12.45 13.45 14.45 16.45 18.35 20.35 22.35 24.35
Max. 1.6 1.6 2 2 2 2.5 2.5 2.8 2.8 2.8
r Min. 1.2 1.2 1.5 1.5 1.5 2.0 2.0 2.2 2.2 2.2
Nom. 22 27 32 36 41 46 50 55 60 65
DESIGN OF STEEL STRUCTURES

e h15 Max. 22.00 27.00 32.00 36.00 41.00 46.00 50.00 55.00 60.00 65.00
Min. 21.16 26.16 31.00 35.00 40.00 45.00 49.00 53.80 53.80 63.80
(1) For length upto 90 mm
(2) For length over 90 mm upto 200 mm.
Note. The dimensions d1 shall not exceed the actual width across the flat.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 107

2.21.3 High Strength Bolts

The high strength bolts are also called high strength friction grip bolts.
Briefly, these are called HSFG bolts. These bolts are developed as a result of
recent developments in the field of connections for the steel structures. These
bolts are in use since 1938. Bureau of Indian Standards Institution has published
a special publication SP : 6(4)–1969 (use of high strength friction bolts), which
deals with the principles of design for such bolts. Use of high strength friction
grip bolts has become very popular as a field fastener for structural connections.
The term high strength friction grip bolts relates to bolts of high tensile steel.
These bolts are used along with high tensile steel nuts and hardened steel washer.
These bolts are tightened to predetermined shank tension which are equal to
proof load. As a result of this, the structural elements connected together are
clamped between the nuts and the head of bolt. The load is transferred from one
structural element to other by friction between the parts as shown in Fig. 2.43
and not by shear in, and bearing on the bolts. These bolts do not allow slip to
occur. When high strength friction grip bolts are used, all connected surface,
including those adjacent to the bolts heads, nuts, or washers should be free of
scale, burns, dirt and other foreign material that would prevent solid seating of
parts. The contact surfaces of parts to be joined should be free of oil, paints,
lacquer or galvanizing.
The high strength friction grip bolts have many advantages over rivets.
Whereas the rivets are subjected to shear and bearing stresses, the bolts are
subjected to uniform tensile stresses only. The high strength friction grip bolts
have higher fatigue strength because there is no concentration of the stress in
the hole. The bolts do not bear against the plates. Therefore, the uneven
distribution of stress does not occur. These bolts are advantageously used in
bridges and machine foundations subjected to vibrations. These bolts also
simplify the problem of alterations and additions to structures as they can be
assembled more easily than rivets. The various dimensions of high strength
3 0°

a dθ θ

z
e c x b
s
k I

Fig. 2.38

friction grip bolts are as shown in Fig. 2.38. The dimensions are as per IS :
3757–1966 and have been given in Table 2.9. The lengths of these bolts are from
30 mm to 80 mm with variationn of 10 mm. These bolts are designated by
expressing type of bolts, followed by the size, the length and symbol representing
108 DESIGN OF STEEL STRUCTURES

the mechanical properties and IS number e.g., friction grip bolts M 16 × 100
IS : 3757 : 10 K . The size of this bolt is M 16 (16 mm) and, the length of bolt is
100 mm. The bolt conforms to the mechanical properties of 10 K.

2.22 DEFINITIONS OF TERMS USED IN BOLTS AND BOLTING

1. Nominal diameters of bolt. The nominal diameter of a bolt is diameter of
unthreaded shank of bolt.
2. Gross diameter of bolt. The gross diameter of a bolt is the nominal
diameter of the bolt.
3. Gross area of bolt. The gross area of a bolt is the nominal area of the bolt.
It is the cross-sectional area of the bolt calculated from gross (or nominal) diameter
of the bolt.
4. Net area of the bolt. The net area of a bolt is the area at the root of
threaded part of cross-sectional area of the unthreaded part whichever is less. It is
not cross-sectional area of the bolt calculated from diameter at the root of threads
(or core diameter).
5. Diameter of the bolt hole. The diameter of a bolt hole is the nominal
diameter of the bolt plus 1.5 mm unless specified otherwise.
The following definitions pertain to high strength friction grip bolts.
6. Grip. The grip of bolt is equal to total thickness of steel sections to be held
together excluding washers.
7. Ply. The ply is equal to a single thickness of steel forming part of a structural
joint.
8. Length of bolt. The length of bolt is equal to the distance from the underside
of the bolt head to the extreme end of the shank, including any camber or radius.
9. Effective interface. The effective intersurface is a common contact
surface between two load transmitting plies, excluding packing pieces, through which
the bolt passes.
10. Slip factor. The slip factor is defined as ratio of the load per effective
interface, required to produce slip in a pure shear joint to the total shank tension
induced in the bolts.
11. Factor of safety. The factor of safety is the numerical value by which the
load which would cause slip in a joint is divided to give the permissible working
load on the joint.

2.23 BOLT, NUT AND WASHER ASSEMBLY

Figure 2.39 shows complete assembly of a bolt, nut and two washers. The
complete assembly provides bolted connections.
The washers are used below the nut and inner face of a head of bolt. The
washers distribute the load to be transmitted over a large area. The washers
provide smooth seating for the nut to turn on. The washers prevent the nut
from cutting into the plates.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 109

H e ad
W a sh e r

A ctual len gth

le ng th
G rip
Nut
W a sh e r

Fig. 2.39. Bolt, nut and washers assembly

Two plates are connected by bolts. The combined thickness of plates connected
is called grip length. The actual length of bolt is kept greater than the grip
length. The nut and washers are placed in length which is in excess over the
grip length. A small threaded portion remains projecting outside the nut.

2.24 STRESSES IN BOLTS

Unfinished bolts are used in bolted connections in lap joint as shown in Fig.
2.40 (a) and in butt joint as shown in Fig. 2.40 (b). The plates are carrying an
axial pull, P. The diameter of hole is 16 mm greater than the diameter of bolt.
Firstly, there occurs a slip. Then plate elements come in contact with the bolts.
The bolts are subjected to single shear in case of lap joint and double shear in
case of butt joint. In addition to shear, bolts are also subjected to bearing.

U n fin ishe d
b olts

(a )
L ap jo in t
U n fin ishe d
b olts

P P

(b ) B u tt jo in t

Fig. 2.40 Bolted connections

110 DESIGN OF STEEL STRUCTURES

When turned bolts are used, the holes are drilled and reamed. The turned
bolts fit tightly in the holes as shown in Fig. 2.41. The pull is transmitted directly.
The bolts are subjected to shear and bearing.

Tu rne d bo lt

Fig. 2.41 Bolted connection

Figure 2.42 shows a bolted connection. Pull acts along the longitudinal axes
of bolts A–A. The bolts are subjected to tension. The tension in bolt is resisted
by the cross-section of bolt at the root of thread (i.e., the net area of bolt).

Bo lt in te nsio n
A

P P

Bo lt in te nsio n

Fig. 2.42 Bolted connections

Figure 2.43 shows bolted connections in which high strength friction grip
bolts are used. There is no slip or movement between the connected parts in
these type of connections. This type of connection is useful where rigid joints
are required. The surfaces in contact must remain free from oil, grease, scale
and paint. H.S.F.G. bolts are tightened by applying a torque to the head of bolt.
These are tightened to shank tension equal to the proof load F. In case, any part
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 111

will remain loose, then slip will occur and the joint will act as an ordinary bolt
joint. The plates are clamped together. As a result of this, friction µF, is developed
between the plates as shown in Fig. 2.43. The load is transferred by friction.
These bolts do not act in shear or in bearing. When these bolts are tightened,
longitudinal tensile stresses, and torsional shear-stresses are developed in the
shank of bolt. The shank tension reduces slightly after completion of tightening.
The torsion drops appreciably. Later on, the shank tension remains more or
less constant for the whole of its working life, irrespective of whether, the
connections remain loaded or unloaded.

P µF
F
µF P
H ig h streng th
frictio n g rip b olt

Fig. 2.43 Bolted connection

There are three methods to see that correct shank tension has been reached
in the bolt. These methods are as follows :
1. Part turning. In this method, the nut is screwed up tight and then
tightened further by half to three-fourth of a turn depending on the length and
diameter of bolt.
2. Torque control. In this method, a calibrated power operated or hand
operated torque wrench is used to give a suited torque to bolt.
3. Patented load-indicating bolts and washers. In this, the nuts
have projections that flatten out when the required tension is reached.

2.25 PERMISSIBLE STRESSES IN BOLTS

The permissible stresses in bolts are adopted as per IS : 800–1984 as given in
Table 2.10.

Table 2.10 Permissible stresses in bolts

Description Permissible stress N/mm2 (MPa)

In close tolerance and turned bolts
(i) Axial tension σtf 120
(ii) Shear τvf 100
(iii) Bearing σpt 300
In bolts in clearance holes
(i) Axial tension σtf 120
(ii) Shear τvf 80
(iii) Bearing σpf 250
112 DESIGN OF STEEL STRUCTURES

Note. The permissible stress in a bolt (other than a high strength friction grip
bolt) of property class higher than 4.6 shall be those given in Table 2.10 multiplied by
the ratio of its yield stress or 0.2 per cent proof stress or 0.7 times its tensile strength,
whichever is less, to 235 N/mm2 (MPa).
The calculated bearing stress of a bolt on the parts connected by it shall not
exceed,
(i) the value of yield stress, fy for bolts in clearance holes, and
(ii) the value of 1.2 fy (fy is the yield stress) for close tolerance and turned
bolts.

2.26 DESIGN OF HIGH STRENGTH FRICTION GRIP BOLTS

Indian Standard Institution has specified specifications for high tensile friction
grip bolts in IS : 3757–1967 (High tensile friction grip fasteners for structural
engineering purposes). The code of practice IS : 4000–1967 deals with assembly
of structural joints using high tensile friction grip fasteners.
The connections made with high strength friction grip bolts may be subjected
to shear, tension in the direction of the axis of bolt, combined shear and tension
or repeated variation of stresses. The principles for design of these connections
are as following.

2.26.1 Shear Connections

When the bolt connections are subjected to shear in the plane of the friction
faces, the force in the members, P as shown in Fig. 2.43 does not exceed the
resulting frictional forces, that is, the maximum value of P may be expressed as
follows :
1
P = ⋅ (µ ⋅ F ) ...(2.14)
F ⋅S
where, µ is the coefficient of friction between the interfaces and F.S. is appropriate
factor of safety. The coefficient of friction is also termed as slip factor. The slip
factor is adopted as 0.45. For all loads except wind, the factor of safety is adopted
as 1.4. When the effect of wind forces is considered, its value is adopted as 1.2.
In case, the frictional forces are developed in N interfaces instead of one as
shown in lap joint Fig. 2.43, Eq. 2.14 is modified as under
1
P = ⋅ N ⋅ µ⋅ F ...(2.15)
F ⋅S
The common contact surfaces subjected to forces in the opposite directions give
the number of effective interfaces. The value of proof load F depends on the
diameter and the proof stress of the material.
The friction grip bolts are manufactured as conforming to mechanical
properties of 10 K or 8 G. The proof loads F for the different diameters of bolts
as per IS : 3757–l966 have been given in Table 2.11.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 113

The values of proof loads for 10 K and 8 G bolts are based on 700 N/mm2 and
600 N/mm2, respectively.
Table 2.11 Proof loads for bolts

Proof load (kN)

Diameter of bolts
(mm ) 10 K bolts 8 G bolts
12 59.00 50.58
14 80.50 75.00
16 107.90 91.20
18 134.40 115.20
20 171.50 147.00
22 212.10 181.80
24 237.10 211.80
27 321.30 274.50
30 392.70 336.60
33 485.80 416.40
36 571.90 490.20
39 683.20 585.60

2.26.2 Tension Connections

In case, the bolt connections are subjected to external tension in the direction of
the axis of bolts, then, the clamping force is likely to be decreased. Very large
values of external tension may separate interfaces and the clamping force may
be decreased to zero and cause the failure of the connections. Therefore, the
maximum external tension of any bolt should not exceed 0.6 of the proof load of
the bolt used.

2.26.3 Combined Shear and Tension

In beam to column connections and in some bracket connections, the connections
are subjected to combined shear and tension. IS : 4000–1967 specifies that the
bolts are so proportioned that the following expression is satisfied.
⎡⎛ Calculated shear ⎞ ⎤ ⎛ Proof load ⎞
⎢⎜ Slip factor ⎟ +1.2 × (calculated tension)⎥ = ⎜⎝ F × S ⎟⎠
⎣⎝ ⎠ ⎦
The effective clamping action of a bolt will cease when the externally applied
tension reaches 0.6 of its proof load.

2.26.4 Repeated Variation of Stress

The nut and washer of the bolt remain in position by spring action due to tension
in the bolt. As such these nuts and washers do not become loose even if these
are subjected to repeated variation of stress. Also there is very little vibrations
114 DESIGN OF STEEL STRUCTURES

in the tension of the bolt. The fatigue strength of friction grip connections has
been found to be 25 per cent higher than the ordinary riveted joints. However,
the external tension is limited to 0.5 of the proof load so that the separation of
friction surfaces does not occur.

2.27 ADVANTAGES OF PIN CONNECTIONS

The following are the advantges of pin connections :
1. The pin connections simplify analysis of indeterminate structure. The
moment at a pin-connection is zero.
2. The pin connections have freedom of rotation of connected members.
3. Only one pin is used in a pin-connection. Whereas in a rivet or bolted
connection, more than one rivets or bolts are used.
4. The pin connections reduce secondary stress occurring at the joints.

2.28 DISADVANTAGES OF PIN CONNECTIONS

The following are the disadvantages of pin connections :
1. The pin cannot resist the longitudinal tension. Because, longitudinal
tension produces friction which prevents free turning of pins. Free turning
of pins is one of its principal function.
2. The pins and pin-holes require expensive machine work on them.
3. The pin connections have lack of rigidity. As a result of this, there is noise
in bridges under traffic vibrations (specially in light structures).

2.29 PINS
The pins for structural connections are manufactured from mild steel. The pins
can freely turn in a pin-connections. The sizes of pins range from 9 mm diameter,
used for connecting strap, iron bars, to railway bridge pins 330 mm or more in
diameter.
The following are various types of pins used for structural connections.

2.29.1 Forged Steel Pin

Figure 2.44 (a) shows a forged steel pin used for structural connections. When
forged steel pins are used, these are secured by means of a cotter pin or a tape
pin, or a thin nut screwed upto shoulder on the end of pin. Fig. 2.44 (b) and Fig.
2.44 (c) show washer and cotter pin, respectively.

P in
C o tte r
(a ) Fo rg ed stee l pin (b ) W a sh e r p in
Fig. 2.44
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 115

2.29.2 Undrilled Pins

Figure 2.45 (a) shows undrilled pins. Undrilled pins are used in bridges. Pilot
nut, driving nut and holding nut shown in Fig. 2.45 (b), (c) and (d), respectively,
are used for securing the pin when it is placed in position.
P ilo t H o ld in g
n ut n ut

D riving
U n drilled n ut
p in
(a ) (b ) (c) (d )

Fig. 2.45 Undrilled pin

2.29.3 Drilled Pins

When a diameter of pin is more than 200 mm, then drilled pin as shown in Fig.
2.46 (a) is used. A hole is drilled in this type of pin. Bolt and cap along with nut
and washers as shown in Fig. 2.46 (b), (c) and (d) are used for securing the pin,
when it is placed in position.

(a ) D rille d pin

B o lt
Nut Nut
Cap W a sh e r
(b ) (c) (d )
Fig. 2.46

2.30 PIN CONNECTION

In a pin-connected structure, the connection between the various members
meeting at a joint is made by means of a pin. One end of a bar is forged in the
form of fork and hole is drilled in it. One end of other bar is forged and an eye is
made. A hole is also drilled in it. The eye-bar can be inserted within the jaws of
the fork. A forged steel pin is driven through the holes in them. The pin is
secured by means of a cotter pin or a screw. The ends of both parts are made
octagonal for good grip. Such type of pin connection is shown in Fig. 2.47. This
116 DESIGN OF STEEL STRUCTURES

type of pin-connection is used in structures such as braced girders, links for

suspension chains (chain-link cables of suspension bridges used for medium
spans) etc.

Fig. 2.47 Pin connection

When two pairs of parallel eye bars are connected together then an undrilled
pin is driven through the holes in them as shown in Fig. 2.48. In such type of

40
mm P

40 3 mm
P mm

80 m m S p ace r

40
mm 3 mm

40
mm P

Fig. 2.48 Pin connection

 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 117

connection, a spacer is inserted between inner pair of eye bars. Such type of
connections are used in structures such as truss bridge girders. A clearance of
3 mm is kept between adjacent elements to allow free rotation at the joint.

2.31 STRESSES IN PINS

The pins are subjected to the following stresses in pin-connections.

2.31.1 Shear Stress

A pin is fact, is a deep beam. The distribution of shear stress in a pin is the same
as for beams of circular section. The maximum shear stress in a pin is
⎛V A y ⎞
τvf (max) = ⎜ ⎟
⎝ I .b ⎠
where, V is shear force at the section, and A is area of cross-section above the
fibre under consideration, and
y = Distance of C.G. of the above area from fibre under consideration
I = Moment of inertia of cross-section
b = Width of fibre under consideration.
For circular section of pin, the maximum shear stress occurs at the neutral
axis. The maximum shear stress is, therefore,

⎡ ⎛ 1 2 ⎞⎛ 4 r ⎞⎤
⎢ V ⎝⎜ 2 π r ⎠⎟ ⎝⎜ 3 π ⎠⎟ ⎥
τvf (max) = ⎢ ⎥
⎢ ⎛ π ⋅ r 4 ⎞ (2 π r ) ⎥
⎜ ⎟
⎣⎢ ⎝ 4 ⎠ ⎦⎥

4 ⎛ V ⎞
= ⋅
3 ⎜⎝ π r 2 ⎟⎠

⎛4⎞
τvf (max) = ⎜ ⎟ . τvf (ave)
⎝3⎠
where τvf (max)= Radius of pin, τvf (ave) = Average shear stress.
The allowable shear stress in pins may be adopted same as that for power
driven shop rivets.

2.31.2 Bearing Stress

The pins are also subjected to bearing stress. The allowable bearing stress in
pins may be adopted same as that for power driven rivets.

2.31.3 Bending Stress

Pins are subjected to bending stress. It may be necessary to proportions the pin
for bending. The design of pin is likely to be governed by bending rather than by
118 DESIGN OF STEEL STRUCTURES

shear or by bearing. This is a significant difference between the design of a pin

connection and the design of riveted joints.

2.32 PERMISSIBLE STRESSES IN PINS

Following values of permissible stress may be adopted for pins. The values of
permissible stress in pins may be adopted as those for rivets.

Table 2.12 Permissible stresses

Description Permissible stress N/mm2 (MPa)

1. Shear 100
2. Bearing 300
3. Axial tension 100
4. Bending
(a) Upto and including
20 mm diameter 0.66 fy
(b) Over 20 mm diameter 0.66 fy
Example 2.12 The uplift on windward side of a steel chimney is 276 N/mm of
periphery at the base. Design foundation bolts.
Solution Design :
Step 1. Tensile strength of bolt
The bolts are subjected to axial tension
Tensile force at the base of steel chimney = 276 N/mm
Use 37 mm diameter bolts
Area at the root of threads = 840 mm2
Allowable stress in axial tension for bolts 37 mm diameter
= 120 N/mm2
⎛ 1.33 × 120 × 840 ⎞
Strength of one bolt in tension = ⎜ ⎟ = 134.064 kN
⎝ 1000 ⎠
Step 2 : Spacing of the bolts
⎛ 134.064 × 1000 ⎞
= ⎜ ⎟ = 485.74 mm
⎝ 276 ⎠
Provide 37 mm diameter bolts at 480 mm spacing along the periphery at the
base of chimney.
Example 2.13. Two plates have been connected in bolted joint as shown in
Fig. 2.49. The width of plate is 200 mm. Design bolted connection to transmit
pull equal to the strength of plate. Thickness of plate is 10 mm.
Solution Design :
Step 1. Bolt value
Let 20 mm diameter turned and fitted bolts are used. Two bolts are used in
one row as shown in Fig. 2.49.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 119

Strength of a bolt is double shear

⎛ π (20)2 × 120 ⎞
2. ⎜ × ⎟ = 75.36 kN
⎝4 1000 ⎠
When turned and fitted bolts are used, then, holes are drilled and reamed,
the diameter of hole is kept equal to the diameter of bolt.
⎛ 20 × 10 × 300 ⎞
Strength of a bolt in bearing = ⎜ ⎟ = 65 kN
⎝ 1000 ⎠
Bolt value = 65 kN
Step 2 : Strength of plate in axial tension
⎛ 0.6 × 260 (200 – 2 × 20) × 10 ⎞
⎜ ⎟ = 249.6 kN
⎝ 1000 ⎠
Number of bolts required to transmit this pull
⎛ 249.6 ⎞
⎜ ⎟ = 3.84
⎝ 65 ⎠

1 0 m m th ick
6
mm
P P
6
mm
2 00 m m

P P

30 6 mm 60 m m 60 m m 60 m m 60 m m
Fig. 2.49 Bolted joint

Provide 6 bolts 20 mm in diameter on each side as shown in Fig. 2.49.

Example 2.14. Two plates of 16 mm thickness have been connected in a lap
joint using high strength friction grip bolts. Design the joint so as to transmit a
pull equal to full strength of the plate.
Solution Design :
Step 1 : Strength of plate
In the lap joint connecting two plates, the number of interfaces N is equal to
one only. Let p be the pitch of the bolts. The strength of plate per pitch.
F = p × t × σt
⎛ p× 16 × 0.6 × 260 ⎞
or P = ⎜ ⎟ = 2.496 p kN
⎝ 1000 ⎠
120 DESIGN OF STEEL STRUCTURES

Step 2. Force in one bolt

Assuming that the plates are connected using bolts in two rows only. The
force to be transmitted by one bolt
1
× 2.496 p = 1.248 p kN
2
From Eq. 2.14
1
P = .(N × µF) (3 µ = 0.45)
F ⋅S
1
∴ 1.248 p = (1 × 0.45 × F )
1.4
⎛ 1.248 × 1.4 p ⎞
∴ F = ⎜ ⎟ = 3.883 p kN
⎝ 0.45 ⎠
Assuming that the bolts are provided at 60 mm pitch. Then,
F = (3.883 × 60) = 232.98 kN
Example 2.15. Two pairs of parallel eye bars of size 150 mm × 40 mm are
connected by a pin. The inner pair of eye is spaced 80 mm apart. Design the pin
connection.
Solution Design :
Step 1
Pull transmitted by each eye bar
⎛ 0.6 × 260 × 150 × 40 ⎞
P = ⎜ ⎟
⎝ 1000 ⎠
= 936 kN
Shear force at X1 or X2 = 936 kN
Maximum bending moment at X1 or X2
= 936 × 43
= 40248 kN-mm
Step 2. Section modulus of pin required
⎛ 40248× 1000 ⎞
Z = ⎜
⎝ 0.66 × 260 ⎟⎠
= 234.545 × 1000 mm3
Step 3. Diameter of pin
Section modulus of a cylindrical pin
⎛ π ⎞
= ⎜ × d3 ⎟ mm3
⎝ 32 ⎠
where, d = Diameter of pin in mm
π 3
d = 234.545 × 1000
32
d = 133.70 mm
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 121

Provide 140 mm diameter pin.

E ye-ba r

E ye-ba r

S p ace r

E ye-ba r

(a ) P in co nn ection

P X1

P X2

(b )

Fig. 2.50

Step 4. Check :
1. Average shear stress in the pin

⎛ 936 × 1000 ⎞
τva = ⎜ ⎟ = 60.834 N/mm
2
π ( )2
⎜ × 140 ⎟
⎝ 4 ⎠
< 100 N/mm2. Hence, satisfactory.
2. Bearing stress in pin
⎛ 936 × 1000 ⎞
σp = ⎜ ⎟ = 167.143 N/mm2
⎝ 140 × 40 ⎠
< 300 N/mm2. Hence satisfactory.
Provide 140 mm diameter undrilled pin for pin connection.
122 DESIGN OF STEEL STRUCTURES

PROBLEMS
2.1. A single riveted lap joint used to connect plates 12 mm thick. If 22 mm
diameter power driven field rivets are used at 70 mm, determine the
strength of joint and its efficiency.
2.2. A double riveted double cover butt joint is used to connect plates 10 mm
thick. Using Unwin’s formula determine the diameter of rivet, rivet value,
pitch and efficiency of joint.
2.3. A double cover joint is used to connect plates 14 mm thick. Design the
riveted joint and determine its efficiency.
2.4. Determine the strength of a double cover butt joint used to connect two
flats 150 F 10. The thickness of each cover plate is 8 mm. Flats have been
joined by 22 mm power driven to 6 rivets in two rows in chain riveting.
The distance between centre to centre of rows of rivets is 70 mm and edge
distance is 40 mm. What is the efficiency of the joint ?
2.5. In a truss girder of a bridge, a diagonal consists of a 14 mm thick flat and
carries a pull of 600 kN and is connected to a gusset plate by a double
cover butt joint. The thickness of each cover plate is 8 mm. Determine the
number of rivets necessary and the width of the flat required. Arrange the
rivets in diamond riveting. What is the efficiency of the joint ? Sketch the
joint.
2.6. A bridge truss diagonal carries an axial pull of 400 kN. It is to be connected
to a gusset plate 22 mm thick by a double cover butt joint with 22 mm
diameter power driven rivets. If the width of the flat tie bar is 200 mm,
determine the thickness of flat. Design an economical joint. Determine
the efficiency of the joint.
2.7. An engineer is investigating an exciting crane runway to see whether the
crane load may be increased. The connection of the gantry girder
P

+ + +

+ + +
+ + +

+ +
50 m m W eb
+ +
10 m m
+ +
A ll rive ts 1 8 m m
d iam e te r

Fig. P 2.7

to the column is shown in Fig. P 2.8. Determine the maximum allowable

load when the permissible shear stress in rivet is 90 N per mm square.
Rivet in double shear may be allowed 2 × single shear value.
 DESIGN OF RIVETED, BOLTED AND PIN CONNECTIONS 123

2.8. Design a bracket connection as shown in Fig. P 2.8 carrying an eccentric

load of 150 kN at a distance of 300 mm from the centre line.

1 50 kN

IS H B 3 00
@ 0 .58 8 kN /m

Fig. P 2.8

2.9. Design the riveted connections for the column bracket carrying an eccentric
load of 100 kN at a distance 250 mm from the face of the column as shown
in Fig. P 2.9.

1 00 kN

+ + + + + + +

10 m m

Fig. P 2.9

2.10. Determine the resultant stress on the rivet B of the eccentricted connections
shown in Fig. P 2.9 and compare this stress with the allowable rivet value
if 20 mm rivets are used. The bracket plate and column flange are both 10
mm thick and allowable stresses of 100 N/mm2 in shear and 300 N/mm2 in
bearing may be assumed.
124 DESIGN OF STEEL STRUCTURES

5 0 kN
2 50 m m

R ive ts 22 m m φ
1 00
mm
1 00
mm
1 00
mm
B

C o lu m n

Fig. P 2.10

2.11. The bracket shown in Fig. P 2.11 consists of a pair of M.S. plates riveted to
the flange of a 305 mm × 152 mm I–section column. If the resultant force
on the critical rivet is limited to 45 kN, determine the load P, the bracket
can support.

3 00 m m

40 m m

75 m m

75 m m

75 m m

40 m m

I - Se ctio n co lu m n
(3 05 m m × 15 2 m m )

89 m m

Fig. P 2.11

2.12. The uplift on a windward side of a steel chimney is 324 N/mm of periphery
at the base. Design the foundation bolts.
2.13. Two pairs of parallel eye bars of size 240 mm × 60 mm are connected by a
pin. The inner pair of eye bar is spaced 100 mm apart. Design pin-
connections.
 Chapter
Design of Columns and

3 Compression Members

3.1 INTRODUCTION
A column is defined as a structural member subjected to compressive force in a
direction parallel to its longitudinal axis. The terms stanchions and posts are
also used for columns. The columns, stanchions and posts are general terms
used in building construction. In truss bridge girders, end compression members
are termed as end posts.
A strut is defined as a structural member subjected to compression in a
direction parallel to its longitudinal axis. The term strut is commonly used for
compression members in roof trusses. A strut may be used in a vertical position
or in an inclined position in roof trusses. The compression members may be
subjected to both axial compression and bending.
When compression members are overloaded then their failure may take place
because of one of the following :
1. Direct compression
2. Excessive bending
3. Bending combined with twisting.
The failure of column depends upon its slenderness ratio. The load required
to cause above mentioned failures decreases as the length of compression member
increases, the cross-sectional area of the member being constant. Therefore,
columns are commonly classified as short and long columns. This classification
is arbitrary and there is no absolute way to determine the exact limits for each
classification.

3.2 AXIALLY LOADED COLUMNS

In an axially loaded column, the load is applied at the centroid of the section
and in a direction parallel to the longitudinal axis of the column. The terms
126 DESIGN OF STEEL STRUCTURES

centrally loaded and concentrically loaded are also used for axially loaded
columns.
An axially loaded column as defined by the structural engineers transmits a
compressive force without an explicit design requirement to carry lateral loads
or end moments. There exist accidental end eccentricity, initial crookedness,
initial curvature and the residual stresses within the tolerance limits in the
columns. By having appropriate modifications in the analytical expressions and
using suitable factor of safety, the design formula makes the account for these
factors. Due to these factors, the strength of an actual column used to be less
than that of a perfectly axially loaded column.
An ideal column is assumed initially to be perfectly straight. It is centrally
loaded. Consider a case of a slender ideal column. The column is vertically fixed
at the base and free at the upper end and subjected to an axial load P as shown
in Fig. 3.1. The column is assumed to be perfectly elastic.

P PCR
δ

( δ– y )

Fig. 3.1

When the value of load P is less than critical load, and stress is within the
limit of proportionality, the column remains straight. The column is in stable
equilibrium that is, if a small lateral load is applied at the free end, the column
deflects. On withdrawal of the lateral load, the column resumes its vertical
position and deflection vanishes. When the axial load P is gradually increased,
a stage will be reached when the vertical position of the column is in the unstable
equilibrium that is, if a small lateral load is applied, a deflection will be produced,
which will not vanish on withdrawal of lateral load. The axial load which is
sufficient to keep the column in such a slight deflected shape is called critical
load. Critical load is also called as buckling load or crippling load. The buckling
load is defined as the load at which a member or a structure as a whole collapses
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 127

in service (or buckles in a load test). The buckling is defined as the sudden
bending, warping, curling or crumpling of the elements or members under
compressive stresses. The direction of buckling of a column depends upon flexural
rigidity, EI, of the column. It buckles in a direction perpendicular to the axis,
about which the moment of inertia of the section is minimum.
In about 1759 Prof. Leonhard Euler (a Swiss mathematician) derived the
most popular column formula. The critical load for the column as shown in
Fig. 3.1 was determined as under.
The differential equation of the deflected shape of the column is

⎛ d2 y ⎞
⎜ EI 2 ⎟ = + M (Hoging moment + ve) ...(i)
⎝ dx ⎠
The bending moment at any point on the deflected shape
M = +P(δ – y)

⎛ d2 y ⎞
Therefore, ⎜ EI 2 ⎟ = P(δ – y) ...(ii)
⎝ dx ⎠

d2 y P
2
− (δ – y) = 0 ...(iii)
dx EI

1
⎛ P ⎞2
Let n = ⎜ ⎟
⎝ EI ⎠
Then the differential equation becomes
d2 y
+ n2y – n2δ = 0 ...(iv)
dx 2
The general solution of this equation is
y = A . sin (nx) + B . cos (nx) + δ ... (v)
⎛ dy ⎞
At ⎜⎝ x = 0, y = 0, = 0⎟ , B = – δ, A = 0
⎠
dx
∴ y = δ (1– cos nx)
At x = l, y = δ
This condition is satisfied when
δ cos (nl) = 0
From this, either δ = 0 or cos nl = 0 ...(vi)
If δ is zero, buckling of column does not occur. If cos nl = 0, then
π
nl = (2π – 1) where n = 1, 2, 3 .....
2
π
For n = 1, the values of nl is smallest. It is equal to
2
128 DESIGN OF STEEL STRUCTURES

⎛ π 2 EI ⎞
∴ Pcr = ⎜ ...(3.1)
⎝ 4l2 ⎟⎠
The values of critical loads for other end conditions can be determined from
this case. For a column, hinged at both ends,
⎛ π 2 EI ⎞
Pcr = ⎜ 2 ⎟ ...(3.2)
⎝ l ⎠
where Pcr = Critical load (or Buckling load).
Since each half of the column is in the same position as the whole of the
column. This is called fundamental case of buckling of a bar.

3.3 EFFECTIVE LENGTH OF COMPRESSION MEMBER

The effective length of a compression member depends upon end restraint
conditions. The end restraint conditions are of two types as given below :
1. Position restraint; 2. Direction restraint.

3.3.1 Position Restraint

In position restraint end of the column is not free to change its position but
rotation about the end of the column can take place e.g., hinged end of column
as shown in Fig. 3.2 (a).

PCR P CR

R o tatio n
fre e

PCR PCR

(a ) Po sition re straint (b ) Po sition an d

d ire ctio n restra in t

Fig. 3.2

3.3.2 Direction Restraint

In direction restraint, end of the column is free to change its position but rotation
about the end of the column cannot take place.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 129

When an end of a column is having restraint in position and direction both

the end is not free to change its position, and the rotation about the end of the
column also cannot take place as shown in Fig. 3.2 (b).
There are various possible combinations of restraints about either or both
axes. The restraint conditions at the two ends of a column may be different or
may be same. Following are the ideal cases of the end conditions :
1. Both ends of column hinged
2. Both ends of column fixed
3. One end of column fixed and the other end hinged
4. One end of column fixed and the other end free.
The deflected shapes of columns under critical loads have been shown in
Figs. 3.3 (a), (b), (c) and (d) respectively. The actual lengths have been indicated
by ‘L’. Cases one i.e., both ends of the column hinged has been considered as
standard case. The effective length (l) of a column is expressed in terms of
equivalent length of compression members, hinged at both the ends. It is the
length of column between two adjacents points of zero moments, and is
represented by ‘l’. It is also called as unsupported length.

PCR P CR PCR
PCR

PCR PCR P CR
(a ) (b ) (c) (d )

Fig. 3.3 Ideal (standard) cases of end conditions

The ideal end conditions cannot be achieved in actual practice. The effective
length of a compression member is adopted as per Table 3.1 as recommended by
BIS in IS : 800–1984 for different types of compression members. The effective
length as given in Table 3.1 will be adequate in most of the cases. The effective
length as given in this table may also be adopted where the columns directly
form part of the frame structures.
130 DESIGN OF STEEL STRUCTURES

PCR PCR PCR

P CR PCR PCR
(a ) (b ) (c)

Fig. 3.4 Ideal (Standard) cases of end conditions

Table 3.1 Effective length of compression members

Sl. Type Effective length

No. of member (l)
1. Effectively held in position and restrained against
rotation at both ends [Fig. 3.3 (b)] 0.65 L
2. Effectively held in position at both ends and restrained
against rotation at one ends [Fig. 3.3 (c)] 0.80 L
3. Effectively held in position at both ends but not
restrained against rotation [Fig. 3.3 (a)] 1.00 L
4. Effectively held in position and restrained against
rotation at one end and at the other end restrained
against rotation but not held in position [Fig. 3.4 (a)] 1.00 L
5. Effectively held in position and restrained against
rotation at one end and at the other end partially
restrained against rotation but not held in position 1.5 L
6. Effectively held in position at one end but not restrained
against rotation, at the other end restrained against
rotation but not held in position 2.00 L
7. Effectively held in position and restrained against
rotation at one end but not held in position [Fig. 3.3 (d)]
or restrained against rotation at the other end. 2.0 L
Note. For battened struts, the effective length l is increased by 10 per cent.

Where the exact frame analysis is not done, the effective length of columns in
the frame structures my be found from the ratio of effective length to the
unsupported length (l/L) from Fig. 3.5 when the relative displacements of the
column is prevented (i.e., when there is no sway) and from Fig. 3.6 when the
relative lateral displacement of the ends is not prevented (i.e., without restraint
against sway viz., the sway occurs), when sway occurs, IS : 800–1984 recommends
that the effective length ratio, (l/L) may not be taken to be less than 1.2.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 131

H in g ed

1.
0
E f t io
fe 1
ra

c t /L
iv =
e
le
ng
th
β1

Fixe d
0
0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 .0
Fixe d β2 H in g ed

Fig. 3.5 Effective length ratios for a column in a frame with no sway

In Figs. 3.5 and 3.6.

⎛ Σkc ⎞
β1 = ⎜ ⎟ ...(i)
⎝ Σkc + Σkb ⎠

⎛ Σkc ⎞
So also, β2 = ⎜ ⎟ ...(ii)
⎝ Σkc + Σkb ⎠
where, the summation is to be done for the members framing into a joint at top
and bottom respectively, and
kc = flexural stiffness of the column, and
k b = flexural stiffness of the beam
Figures 3.6 and 3.7 are from the paper titled as Effective lengths of columns
in multistorey buildings by Professor R.H. Wood, published in the Structural
Engineer Vol. 52, No. 7 July 1974.
It is worthwhile to note that IS : 800–1984 ‘Code of Practice for General
Construction in Steel’ and IS : 456–1978 ‘Code of Practice for Plain and Reinforced
Concrete’ have recommended the same effective lengths for the columns with
similar support conditions.
132 DESIGN OF STEEL STRUCTURES

H in g ed

00
E f t io
fe 1
ra

c t /L
iv =
e
le 2 .0
ng
th
β1

Fixe d
0
0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 .0
Fixe d β2 H in g ed

Fig. 3.6 Effective length ratios for a column in a frame without

restraint against sway

3.4 EFFECTIVE SECTIONAL AREA

The gross cross-sectional area is the area as calculated from the specified size of
the member or part thereof.
The effective sectional area of a compression member is the gross cross-
sectional area of the member. The deduction is not made for members connected
by rivets, bolts and pins. If the holes are not filled by the fastening material,
then deduction is made for unfilled holes. The effective area is, however, modified
when the ratio of the outstand to thickness exceeds the limits specified by BIS.
This point has been discussed further in Sec. 3.13. The deduction is also made
from the gross cross-sectional area for excessive effective plate width (if any) to
determine the effective sectional area.

3.5 RADIUS OF GYRATION

The radius of gyration of a section is a geometrical property of the section and it
is denoted by r
1/ 2
⎛I ⎞
r = ⎜ ⎟
⎝A⎠
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 133

where I = moment of inertia of the section about the axis

r = radius of gyration of the section about the axis
A = effective sectional area of the section.
Along with the cross-sectional shapes of various columns used in practice as
shown in Fig. 3.8, approximate radii of gyration for sections are given. In the
cases of the rectangular and circular cross-sections, the values indicated are
closely approximate to the correct values, but for the built-up sections, there
may be considerable fluctuation because of the variation in relative cross-
sectional dimensions.

3.6 SLENDERNESS RATIO OF COMPRESSION MEMBER

The slenderness ratio of a compression member is defined as ratio of effective
length of compression member (l) to appropriate radius of gyration (r)
l
Slenderness ratio, λ = .
r
The radii of gyration about various axes of rolled steel sections can be obtained
from structural steel section tables (ISI Handbook No. 1). The minimum radius
of gyration is used for computing the maximum slenderness ratio. For built-up
compression members, value of radius of gyration is calculated. The slenderness
ratio for a compression member should be as small as possible so that the material
may be stressed to its greatest possible limit. The maximum slenderness ratio
of compression members should not exceed the values given in Table 3.2. These
limits have been laid down by BIS in IS : 800–1984.
The end restraints of columns arc often different in the two principal planes.
The different moments of inertia of the column cross-section in these planes are
sometimes desirable to achieve approximately equal slenderness ratios. The
intermediate supports are provided to the columns for this purpose.

l
Table 3.2 Maximum slenderness ratio of compression member λ =
r

l
Sl.No. Type of member λ =
r
1. A member carrying compressive loads resulting from
dead loads and imposed loads 180
2. A member subjected to compression forces resulting from
wind/earthquake forces provided that the deformation of
such members does not adversely affect the stress in any
part of the structure 250
3. A member normally acting as a tie in a roof truss or a bracing
system but subjected to possible reversal of stress resulting
from the action of wind or earthquake forces 350
134 DESIGN OF STEEL STRUCTURES

The intermediate supports reduce the unsupported length of the columns.

When the unsupported lengths of columns are reduced, then, the smaller sections
may be used at a higher average stress.
Sometimes, the intermediate supports are furnished only in one direction.
For example, a rolled steel I-section column is having its continuous length
upto two storeys. At the level of one storey, intermediate support is provided by
connecting beams with the web. The radius of gyration, ryy of the section, about
yy-axis (axis parallel to the web) is much smaller than the radius of gyration rxx
of the section, about xx-axis. By providing the intermediate support, the effective
lengths of the column become different in two different directions. The effective
length of columns, lyy for bending about yy-axis is found by considering the length
of column between one-storey only. The effective length of column lxx for bending
about xx-axis is found by considering the length of column between two storeys.
It is seen that the effective length of column lyy is much smaller than that of lxx.
⎛ lyy ⎞ ⎛l ⎞
The values of slenderness ratio ⎜ and ⎜ xx ⎟ about two directions may be
⎜ ryy ⎟⎟ ⎝ rxx ⎠
⎝ ⎠
made approximately equal. As such the use of sections with different values of
radii of gyration in two directions may be made economical. When a column is
subjected to different bending moments in two directions then, the greater value
of r may be kept in the direction of greater moment. The intermediate supports
in the weak direction make the use of I-section and channel section economical.

3.7 COLUMN FORMULAE FOR AXIAL STRESS IN COMPRESSION

The strength of a column depends upon large number of variables. The efforts
are made to obtain a design formula by fitting a curve to experimentally found
buckling loads for the intermediate range of the slenderness ratio. It is tried to
draw a curve which may merge with the Euler hyperbola in the very slender
column range on one side and with the material yield strength for the zero
length on the other side.
A perfectly straight column of perfectly homogeneous material (i.e., an ideal
column) is subjected to an axial load. The primary object is to find the average
axial stress in compression, which corresponds to the allowable load. The average
axial stress is uniform across the section. It is given by

⎛P ⎞
σca = ⎜ a ⎟ ...(i)
⎝ A⎠
where, Pa = allowable load
A = cross-sectional area of the column
The required cross-sectional area for a given design load may be found
conveniently in case σc is known

Areqd. ⎛P ⎞ ...(ii)
= ⎜σ ⎟
⎝ c⎠
where P is the design load.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 135

It is observed that the average axial stress in the column at the time of failure
of column is less than the yield strength of the material. The difference in the
stress depends on the type of failures of the column. The various types of failures
of the column other than the direct compression or direct crushing are as follows :
1. Primary laterally buckling
(a) Inelastic buckling
(b) Elastic buckling
2. Gradual yielding due to excessive local stress
3. Local buckling, and
4. Torsional buckling.
The failure by local buckling and by torsional buckling are prevented by
selecting the rolled steel sections of proper shape and proportions. Therefore,
the failure of a column may occur by lateral buckling or by gradual yielding due
to an excessive local stress. The average allowable axial stress for use in the
design is influenced by amount of lateral buckling. It depends upon the
slenderness ratio and the compressive strength of the material. Therefore, the
columns may be classified in two categories viz., (i) those columns which fail in
elastic buckling, and (ii) those columns which fail in inelastic buckling. The
columns having very large slenderness ratios fail in elastic buckling. Lconhard
Euler predicted that the lateral buckling of the column occurs when the value of
load, P, exceeds the critical or ultimate load, Pcr either in the elastic range or in
the inelastic range, the critical stress in the column for elastic buckling is given
by Euler’s formula as follows :

P ⎛ π2 E ⎞
σcr = ⎛⎜ cr ⎞⎟ = ⎜ ⎟ ...(3.3)
⎝ A ⎠ ⎜ ⎛ KL ⎞2 ⎟
⎜⎜ ⎟ ⎟
⎝⎝ r ⎠ ⎠
where, K = Factor, which depends upon the end conditions of the column
E = Modulus of elasticity
L = Actual length of the column
r = Radius of gyration of the section.
Equation 3.3 may be written as

⎛ P ⎞ ⎛ π2 E ⎞
σcr = ⎜ cr ⎟ = ⎜ ⎟ ...(3.4)
⎝ A ⎠ ⎜ ⎛ l ⎞2 ⎟
⎜⎜ ⎟ ⎟
⎝⎝r⎠ ⎠
where, l = K.L. = Effective length of the column.

⎛ l ⎞
1. Euler’s formula is suitable for large slenderness ratio ⎜ i.e., > 200⎟ .
⎝ r ⎠
It was considered appropriate to apply a variable factor of safety to account for
136 DESIGN OF STEEL STRUCTURES

this stress level and additionally to consider the effect of eccentricity, residual
stresses and the several other factors which complicate the column theory. The
maximum average allowable stress in the axial compression may be found by
dividing σcr by factor of safety (F.S.). Therefore,

⎛ σ cr ⎞ 1 ⎛ π2 E ⎞
fa = ⎜ ⎟ = ⋅ ...(3.5)
⎝ F .S ⎠ F .S . ⎜ ⎛ l ⎞ 2 ⎟
⎜⎜ ⎟ ⎟
⎜⎝ ⎝ r ⎠ ⎟⎠

The relationship between Euler’s formula and A.I.S.C. (American Institute

of Steel Construction) 1953 Specifications formula may be established.
The critical stress by Euler’s formula may be expressed as

⎛ π2 E ⎞
σ cr = ⎜ 2⎟ ...(i)
⎜ ⎛⎜ l ⎞⎟ ⎟
⎜⎝ ⎝ r ⎠ ⎟⎠

⎛l⎞
Differentiating (i) with respect to ⎜ ⎟
⎝r⎠

⎛ d (σ cr ) ⎞ ⎛ −2 π 2 E ⎞
⎜ ⎛ l⎞ ⎟ = ⎜ 3 ⎟ ...(ii)
⎜ d⎜ ⎟ ⎟ ⎜ ⎛⎜ l ⎞⎟ ⎟
⎝ ⎝ r⎠ ⎠ ⎜
⎝ ⎝ r ⎠ ⎟⎠

The maximum value of Euler’s stress or any critical buckling stress is limited
to yield stress fy. The Euler’s formula is not valid for small values of slenderness
ratio. For such a region, the critical stress may be obtained by assuming the
following expression (which one is also based on experimental column test data).
2
⎛l⎞
σ cr = fy — m. ⎜ ⎟ ...(iii)
⎝r⎠

⎛l⎞
Differentiating (iii) with respect to ⎜ ⎟
⎝r⎠

⎛ d( σ cr ) ⎞ ⎛ ⎛ l ⎞⎞
⎜ ⎛ l ⎞ ⎟ = ⎜ −2m ⋅ ⎜ r ⎟ ⎟ ...(iv)
⎜ d⎜ ⎟ ⎟ ⎝ ⎝ ⎠⎠
⎝ ⎝ r⎠ ⎠

l
At = 0, the slope may be assumed as zero. The two curves defined by the
r
expressions (i) and (iii) will have a common tangent. At that point, the slopes
given by (ii) and (iv) will be equal. Equating the two slopes at a common point.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 137

⎛ ⎛ l ⎞⎞ ⎛ −2 π 2 E ⎞
⎜ −2m ⋅ ⎜ ⎟ ⎟ = ⎜ 3 ⎟ ...(v)
⎝ ⎝ r ⎠⎠ ⎜ ⎛⎜ l ⎞⎟ ⎟
⎜⎝ ⎝ r ⎠ ⎟⎠

The slenderness ratio at this common point be arbitrarily defined by a

parameter
⎛l⎞
⎜ ⎟ = Cc ...(vi)
⎝r⎠
Substituting this in Eq. (v),

⎛ −2 π 2 E ⎞
(– 2m . (Cc) = ⎜ ⎟
⎜ (C )3 ⎟
⎝ c ⎠

⎛ π2 E ⎞
∴ m = ⎜ 4 ⎟ ...(vii)
⎝ Cc ⎠
Thus, the expression (ii) may be written as
2
⎛ π2 E ⎛ l ⎞ ⎞
σcr = ⎜ fy − 4 ⎜ ⎟ ⎟ ...(viii)
⎜ Cc ⎝ r ⎠ ⎟⎠
⎝
By substituting, the values of σcr by Euler’s formula (on left hand side)

⎛ π2 E ⎞ ⎡ π2 E ⎛l⎞
2 ⎤
⎜ 2⎟ = ⎢ fy − 4 ⋅⎜ ⎟ ⎥ ...(ix)
⎜ ⎛⎜ l ⎞⎟ ⎟
⎜⎝ ⎝ r ⎠ ⎟⎠
⎢
⎣ Cc( ) ⎝r⎠ ⎥
⎦

⎛l⎞
At ⎜ ⎟ = Cc
⎝r⎠

π2 E π2 E 2
= fy − ⋅ Cc ...(x)
(Cc )2 Cc4

1/ 2
⎛ π2 ⋅ E ⎞
∴ Cc = ⎜ (xi)
⎜ f ⎟⎟
⎝ y ⎠
Substituting the value of Cc in Eq. (x)
⎡ π2 E ⎤
σcr = ⎢ fy − 2 ⋅ fy ⎥ = 0.5 fy ...(xii)
⎣ 2 π E ⎦

l
Thus, the critical buckling stress becomes 0.5 fy at = Cc.
r
138 DESIGN OF STEEL STRUCTURES

In general, the expression (viii) may be written as

2
⎡ ⎛ l⎞ ⎤
⎢ 0.5 ⎜ ⎟ ⎥
⎝ r⎠ ⎥
σcr = fy ⎢1.0 − ...(3.6)
⎢⎣ Cc2 ⎥⎦

The lowest of slenderness ratio upto which the elastic buckling occurs is
denoted by Cc.
⎛ l ⎞
The minimum slenderness ratio ⎜ = Cc ⎟ distinguishes the elastic buckling
r
⎝ min ⎠
from inelastic buckling. When the slenderness ratio is greater than Cc, then
failure of column occurs by elastic buckling. A.I.S.C. Specifications recommended
the following for the determination of allowable stress in axial compression for
⎛l⎞
⎜ r ⎟ > Cc .
⎝ ⎠
⎡ π2 E ⎤ ⎡ 9.87 × 2.047 × 106 10 ⎤
σcr = ⎢ 2 ⎥=⎢ 2
× ⎥ N/mm2
⎢ ⎛ l ⎞ ⎛ l ⎞ 100
× ( F .S .) ⎥ ⎢ ⎜⎝ r ⎟⎠ × 1.92
⎥
⎢⎣ ⎜⎝ r ⎟⎠ ⎥⎦ ⎢⎣ ⎥⎦

⎛ 10,52,000 ⎞
σcr = ⎜ 2 ⎟ N/mm
2 ...(3.7)
⎛ l⎞
⎜ ⎜ ⎟ ⎟
⎝ ⎝ r⎠ ⎠
When the slenderness ratio for the column is less than or equal to Cc, then,
the failure of column occurs by inelastic buckling. The allowable stress in axial
⎛l⎞
compression for ⎜ ⎟ ≤ Cc is given by
⎝r⎠
2
1 ⎡ ⎛l⎞ ⎛ 1 ⎞ ⎤
2

σc = ⎢1 − 0.5 ⎜ ⎟ ⋅ ⎜ 2 ⎟ ⎥ ⋅ fy ...(3.8)
F .S . ⎢ ⎝ r ⎠ ⎝ Cc ⎠ ⎥⎦
⎣
3 3
5 3⎛ l ⎞ ⎛ 1 ⎞ 1⎛l ⎞ ⎛ 1 ⎞
where F.S. = + ⎜ ⎟⋅⎜ ⎟− ⎜ ⎟ ⋅⎜ ⎟ ...(3.9)
3 8 ⎝ r ⎠ ⎝ Cc ⎠ 8⎝r ⎠ ⎝ Cc ⎠

l l
When = 0, F.S.=1.67, and = Cc, F.S.=1.92.
r r
⎛l⎞
The factor of safety of 1.92 remains constant for ⎜ ⎟ greater than Cc.
⎝r⎠
Equations. 3.7 and 3.8 are used for main compression members. For bracing
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 139

⎛l⎞
and secondary compression members for which ⎜ ⎟ exceed 120, the factor of
⎝r⎠
⎡ ⎛ l ⎞⎛ 1 ⎞ ⎤
safety is reduced by multiplying it by a factor ψ equal to ⎢1.6 − ⎜ ⎟⎜ ⎟⎥ . A
⎣ ⎝ r ⎠⎝ 200 ⎠ ⎦
more conservative value of factor of safety is used (F.S. = 2.12) for the members
of the bridge structures since these members are in a hostile atmosphere than
those in buildings.
2. Engesser suggested the tangent modulus formula for the buckling
load of an axially load column.
⎛ π2 E I ⎞
Pcr = ⎜ 2 t ⎟ ...(3.10)
⎝ l ⎠

⎛ Pcr π 2 Et ⎞
or σcr = ⎜ = 2⎟ ...(3.11)
A ⎛ l⎞ ⎟
⎜ ⎜⎝ r ⎟⎠ ⎟⎠
⎜⎝

where Et = Tangent modulus of the elasticity.

Shanley and Von–Karman defined the tangent modulus, load as the smallest
value of the axial load at which the bifurcation of the equilibrium position can
occur regardless of whether the transition to the bent position-requires an
increase of the axial load. Beedle and Tall pointed out that Eq. 3.11 cannot be
applied to a steel column if the stress-strain relationship is determined from a
small coupon from the section. The buckling strength of the column is reduced
due to the existence of residual stress in the cross-section, since there is early
localized yielding in the certain portions of the cross-section where the
compressive residual stresses prevails. The basic equation for the critical strength
of a column containing residual stresses is given by
⎛ π2 E ⋅ I e ⎞
Pcr = ⎜ 2 ⎟ ...(3.12)
⎝ l ⎠

⎛ ⎞
⎜ π2 E ⋅ I e ⎟
⎛ Pcr ⎞ ⎜ I ⎟
or σcr = ⎜
A ⎟=⎜ 2 ⎟ ...(3.13)
⎝ ⎠ ⎛l⎞
⎜⎜ ⎜ ⎟ ⎟⎟
⎝ ⎝r⎠ ⎠
3. Beedle and Tall suggested the following relationships for the critical
stresses for the buckling about the strong axis and the buckling about the weak
axis respectively.
1/ 2
l ⎛E⎞
For < π⎜ ⎟
r ⎝ σp ⎠
140 DESIGN OF STEEL STRUCTURES

2
⎡ ⎤
σcr(xx) = ⎢ f − f ( σ − σ ) ⋅ ⎛ l ⎞ ⎥ ...(3.14)
y 2 y P ⎜ ⎟
⎢⎣ πE ⎝ r ⎠ ⎥⎦

1/ 2
⎡ ⎤
σcr(yy) = ⎢ f − ( σ y − σ p ) ⎛ σ p ⎞ ⋅ ⎛ l ⎞ ⎥ ...(3.15)
y ⎜ ⎟ ⎜ ⎟
⎝ r⎠⎥
⎢⎣ π ⎝E⎠
⎦
1/ 2
l ⎛E⎞
For < π⎜ ⎟
r ⎝ σp ⎠

⎡ π2E ⎤
σ
⎢ cr ( xx ) = σ cr ( yy ) = ⎥ ...(3.16)
⎢ ⎛ l ⎞⎥
⎜⎝ r 2 ⎟⎠
⎢⎣ ⎥⎦
The difference between (σy – σp) gives the magnitude of residual stress, where
σp= stress at the limit of proportionality.
The effect of stresses was recognised as an important factor in calculating
the strength of a centrally loaded column in a report published by Column
Research Council.
The residual stresses develop in the hot rolled structural steel sections due to
differential cooling. The residual stresses result from cooling after hot rolling,
welding or fabrication operations such as plane cutting, cold bending or
combining. The flange tips and interior web parts being thinner and more
exposed, always cool more quickly than the other parts. The junctions of flange
and web are the thickest and most protected parts and always cool last. Residual
tensile stresses develop in those parts of the section that cool last as the metal
tends to contract but it is restrained by colder metal. These residual tensile
stresses produce residual compressive stresses in the adjacent metal which has
cooled earlier.
The welding also produces residual stresses as the hot metal and flame cutting
in the vicinity is restrained from contraction by the cooler-to-cold surrounding
metal.
Because of high heat penetration due to welding in built-up column, the
residual stresses tend to be large. In most cases, the tensile residual stresses at
the weld reach the value of yield stress. The distribution of residual stress in a
heavy shape (the thickness of thinnest element of which is more than 38 mm)
differs from that in a small shape in two major respects. The magnitude of
residual stresses is very large and they vary considerably along the thickness.
Nearly all steel members contain both tensile and compressive residual
stresses.
4. AREA (American Railway Engineering Association) recommends
the following for the determination of allowable stress in axial compression.
l
For |
> 140 (Rivet ends)
r
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 141

2
1 ⎛ l⎞
σc = 15000 − ⎜ ⎟ ...(psi units) ...(3.17)
4 ⎝ r⎠
l
For |
> 140 (Pin ends)
r
2
1 ⎛ l⎞
σc = 15000 − ⋅ ⎜ ⎟ ...(psi units) ...(3.18)
3 ⎝ r⎠
5. AASHTO (American Association of State Highway and Transport
Officials) also recommends the above-mentioned expressions. In bridges, the
compression members may have one end riveted and the other end pin-connected.
For example, the end post of a riveted bridge truss is sometimes pin-connected
at its lower end. Bridge specifications do not have any provisions for this
condition. In the absence of the specific instructions, the allowable stress in
axial compression for the member pinned at one end and riveted at the other
end may be found from the following :
2
7 ⎛ l⎞
σc = 15000 − ⋅ ⎜ ⎟ ...(psi unit) ...(3.19)
24 ⎝ r ⎠
Equations 3.17, 3.18 and 3.19 are the forms of Johnson’s parabolic formula
2
⎛l⎞
σc = f – k ⎜ ⎟ ...(3.20)
⎝r⎠
where, σc = Permissible stress in axial compression
f = Crushing stress at failure
k = Constant
⎛l⎞
AISC specifications recommend Johnson’s parabolic formula for ⎜ ⎟ |
> 120
⎝r⎠
2
⎛ l⎞
σc = 17000 − 0.485 ⎜ ⎟ (psi units) ...(3.21)
⎝ r⎠
The allowable stress in axial compression may be found by a straight line
formula
⎛l⎞
σc = f – k ⎜ ⎟ ...(3.22)
⎝r⎠
⎛l⎞
AREA specifications recommend this straight line formula for ⎜ ⎟ |
> 120
⎝r⎠
⎛ l⎞
σc = 18000 − 70 ⋅ ⎜ ⎟ (psi units) ...(3.23)
⎝ r⎠
6. Rankine–Gorden formula or simply known as Rankine formula
⎛l⎞
is adopted for ⎜ ⎟ greater than 120 and less than 200
⎝r⎠
142 DESIGN OF STEEL STRUCTURES

⎛ f ⎞
σc = ⎜ 2⎟ ...(3.24)
l
⎜ 1 + a ⋅ ⎛⎜ ⎞⎟ ⎟
⎝ ⎝ r⎠ ⎠
where a = constant. Its value depends on the material.
It is a semi-empirical formula. AISC recommended this formula in 1949
l
for > 120 and < 200
r

⎡ 18000 ⎤
σc = ⎢ 2 ⎥ ...(psi units) ...(3.25)
⎢1 + 1 ⋅ ⎛ l ⎞ ⎥
⎢⎣ 18000 ⎜⎝ r ⎟⎠ ⎥⎦
Rankine–Gorden formula, straight line formula and Johnson parabolic
formula were extensively used in the past. The curves given by Rankine–Gordon
formula and Johnson parabolic formula when plotted are seen to be the same
curve. Johnsons parabolic formula is simple and it is more generally used.
7. British Standard Specifications for steel and girder bridges
recommended Perry–Roberston formula for the determination of allowable stress
in axial compression
1/ 2
1 ⎡⎢ fy + (η + 1)C0 ⎛ fy + (η + 1)C0 ⎞ ⎤
2

σc = − ⎜⎜ − f C
y 0 ⎟
⎥ ...(3.26)
K⎢ 2 2 ⎟ ⎥
⎣ ⎝ ⎠ ⎦
where, K = Factor of safety, 1.7, fy = Yield stress.
⎛l⎞
Professor Perry used the value of η as 0.0015 ⎜ ⎟ . Later on Roberston
⎝r⎠
examined the experimental data and suggested the following.
⎛l⎞
η = 0.003 ⎜ ⎟
⎝r⎠
The struts are considered as secondary members. These members do not carry
primary loads. These members are used to brace other structural members or
the structures as a whole. AISC specification 1963 recommended that for the
⎛l⎞
struts having slenderness ratio ⎜ ⎟ greater than or equal to 120 and less than
⎝r⎠
200, the allowable stress is increased by factor, K.

⎡ 1 ⎤
K = ⎢ ...(3.27)
1 ⎛ l⎞ ⎥
⎢1.6 − ⎜ ⎟⎥
⎢⎣ 200 ⎝ r ⎠ ⎥⎦
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 143

The factor K for end-conditions is adopted as unity.

In the above column formulae, the columns were considered perfectly straight
and uniform. The load was considered perfectly axial. In actual practice, the
ideal column and loading do not exist. The load supported by the column is
eccentric with respect to the longitudinal axis of the column. The initial
crookedness and other imperfections of column and imperfectness of the service
load distribution give small eccentricity. The secant formula accounts for the
initial crookedness of the column and imperfectness of the axial loading. Indian
Standard Institution (ISI) adopted the secant formula for the determination of
allowable stress in axial compression in 1962. The secant formula was considered
as design formula in IS : 800–1962 as per Indian Standard Specification.
8. Secant Formula. Consider an initially straight column. Suppose a load
P is applied to it at a distance e from the centre line of the column. The deflected
shape of the column would be as shown in Fig. 3.7.
The bending moment at any point on the deflected shape of the column
M = –P . y ...(i)
The negative sign is adopted since the bending moment is sagging. The
differential equation to the elastic curve is as under:
d2 y
EI = M ...(ii)
dx 2

PCR
e
y

e
PCR

Fig. 3.7
144 DESIGN OF STEEL STRUCTURES

d2 y
EI = – P. y ...(iii)
dx 2

⎛ d2 y ⎞ ⎛ P ⎞
or ⎜ 2 ⎟+⎜ ⎟⋅y = 0
⎝ dx ⎠ ⎝ EI ⎠

⎛ d2 y ⎞
or ⎜ 2 ⎟ + n2 . y = 0 ...(iv)
⎝ dx ⎠
The general solution of this differential equation is
y = A . sin (nx) + B . cos (nx) ...(v)
1/ 2
⎛ P ⎞
where n= ⎜ ⎟
⎝ EI ⎠
The constants of this equation can be found by applying the boundary values.
At x = 0, y = e and x = l, y = e

⎛ 1 − cos(nl ) ⎞
∴ A = ⎜ ⎟ ⋅ e and B = e
⎝ sin(nl ) ⎠
The expression (v) reduces to
⎡1 − cos(nl ) ⎤
y = e. ⎢ ⋅ sin(nx ) + cos(nx ) ⎥ ...(vi)
⎣ sin ( nl) ⎦
The maximum deflection in the column occurs at centre i.e., when x is equal
l
to
2
⎡1 − cos(nl ) (nl ) (nl ) ⎤
ymax = e ⋅ ⎢ ⋅ sin + cos
⎣ sin( nl ) 2 2 ⎥⎦

⎡ 2 nl ⎤
⎢ 2sin 2 nl nl ⎥
ymax = e⋅⎢ sin + cos ⎥
⎢ 2sin nl cos nl 2 2⎥
⎢⎣ 2 2 ⎥⎦

⎡ 2 nl 2 ⎛ nl ⎞ ⎤
⎢ sin 2 + cos ⎜ 2 ⎟ ⎥
⎝ ⎠⎥
ymax = e⋅⎢
⎢ nl ⎥
⎢ cos ⎥
⎣ 2 ⎦

⎡ 2 ⎛ nl ⎞ 2 ⎛ nl ⎞ ⎤
⎢3 sin ⎜ 2 ⎟ + cos ⎜ 2 ⎟ = 1⎥
⎣ ⎝ ⎠ ⎝ ⎠ ⎦
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 145

⎛ nl ⎞
ymax = e ⋅ sec ⎜ ⎟ ...(vii)
⎝ 2⎠
The maximum bending moment in the column
⎛ nl ⎞
Mmax = P . e sec ⎜ ⎟ ... (viii)
⎝2⎠
The maximum stress in the column

⎛P M ⎞
σmax = ⎜ + max ⋅ c ⎟ ...(ix)
⎝ A 1 ⎠
where c = Distance to the extreme fibre from centre line of column

⎡ P P ⎛ ec ⎞ nl ⎤
σmax = ⎢ + ⋅ ⎜ 2 ⎟ sec ⎥
⎣ A A ⎝ r ⎠ 2⎦
1 /2
⎛P⎞
where I = Ar2, n = ⎜ ⎟ , and = σc
⎝ EI ⎠
1/ 2
P ⎡ ec l ⎛ P ⎞ ⎤
Then, σmax = ⎢1 + sec ⎜ ⎟ ⎥
A ⎣⎢ r 2 2r ⎝ EA ⎠ ⎦⎥

⎛ fmax ⎞
σc = ⎜ 1/ 2 ⎟
...(x)
⎡ ⎤
⎜ 1 + ec sec ⎢ l ⎛⎜ P ⎞⎟ ⎥ ⎟
⎝⎜ ⎣⎢ 2r ⎝ EA ⎠ ⎦⎥ ⎠
⎟
2
r

This expression is known as secant formula.

Let m be the factor of safety, and therefore, σc = m . σa
where σa = Allowable average stress in axial compression, as it was
recommended in IS : 800–1962.
The maximum stress fmax reached can be yield stress, fy.
Then

⎛ ⎛ fy ⎞ ⎞
⎜ ⎜ ⎟ ⎟
σa = ⎜ ⎝m⎠ ⎟ ...(xi)
⎜ ec l mf
1/ 2 ⎟
⎜⎜ 1 + 2 sec ⎛⎜ a ⎞⎟ ⎟⎟
⎝ r r ⎝ 4E ⎠ ⎠
Indian Standard Code of Practice for general construction steel IS : 800–
1984 has dropped this secant formula for calculating the allowable average stress
⎛ ec ⎞
in axial compression. In secant formula, the value of ⎜ 2 ⎟ was taken as 0.20.
⎝r ⎠
146 DESIGN OF STEEL STRUCTURES

The formula takes into account any initial crookedness of the column and
imperfectness of axial loading.
Therefore, for

⎛ l⎞
⎜⎝ r ⎟⎠ = 0 to 160

⎛ ⎛ fy ⎞ ⎞
⎜ ⎜ ⎟ ⎟
⎜ ⎝m⎠ ⎟
σa = ⎜ 1/ 2 ⎟ ...(3.28)
⎜ 1 + 0.20 sec ⎡ l ⎛ mfa ⎞ ⎤ ⎟
⎜ ⎢ r ⎜ 4E ⎟⎥ ⎟
⎝ ⎣ ⎝ ⎠⎦ ⎠
where fy = Guaranteed minimum yield stress
m = Factor of safety taken as 1.68 on the yield point

⎛l⎞
⎜ r ⎟ = Slenderness ratio
⎝ ⎠
l = Effective length of the member
r = Appropriate radius of gyration
E = Young’s modulus of elasticity, 20,47,000 kg/cm2
For values of

⎛ l⎞
⎜⎝ r ⎟⎠ = 160 and above

⎛ l ⎞
σ′a = σa ⋅ ⎜1.2 − ...(3.29)
⎝ 800r ⎟⎠

where σ′a = Allowable average axial stress in compression as it was

recommended in IS : 800–1962
⎛l⎞
for ⎜ ⎟ > 160.
⎝r⎠
The secant formula for axial compression as it was recommended to determine
permissible stress in IS : 800–1962 has been dropped in IS : 800–1984.
Professor Ashok K. Jain of University of Roorkee made the comparison of
permissible compressive stresses as given by different national standards in his
paper titled as ‘compressive stresses in columns’ published in International
Journal of Structures, Vol. 4, No. 2, April 1984. The disadvantages of secant
formula were pointed out. The secant formula provides a conservative estimate
of the average column stress at maximum load for an eccentrically loaded steel
column. The secant formula is not convenient for hand calculations. It needs
number of iterations to calculate the permissible axial stress as this term appears
on both the sides of the equation.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 147

The secant formula does not show an yield plateau for the columns of
slenderness ratio upto 30. The secant formula gives very conservative values
for columns of slenderness ratio upto 90 as compare to those given by AISC, BS
449, IS 802 and AS : 1250. Further, this secant formula gives higher values of
permissible stresses for the columns having slenderness ratios beyond 90.
Instead of the secant formula, Merchant–Rankine formula has been specified
in IS : 800–1984.

3.8 DESIGN FORMULA

Indian Standard Code of Practice for general construction in steel IS : 800–1984
has recommended Merchant–Rankine formula as below
n n
⎛ 1 1 ⎞ n
⎛ 1 ⎞ ⎛1⎞
⎜⎝ σ ⋅ F ⋅ S. ⎟⎠ = ⎜ ⎟ +⎜ ⎟
⎜ ⎟
...(i)
ac
⎝ fac ⎠ ⎝ fy ⎠

⎛ fcc ⋅ fy ⎞
or σac = ⎜ ⎟ ...(ii)
n 1/ n
⎜⎝ F .S. ⎡⎢( fcc ) + fy
⎣
n
( ) ⎤
⎥⎦ ⎟⎠

The direct stress in compression on the gross-sectional area of axially loaded

10
compression member was found with a factor of safety of . It should not
6
exceed 0.60 fy nor the permissible stress σac, calculated using the following
formula.

⎛ fcc ⋅ fy ⎞
σac = 0.6 ⋅ ⎜ ⎟ ...(iii)
n 1/n
⎜⎝ ⎡⎢( fcc ) + fy
⎣
n
( ) ⎤
⎥⎦ ⎟⎠

where, σac = Permissible stress in axial compression in N/mm2 (MPa)

fy = Yield stress in steel in N/mm2 (MPa)

⎛ π2 E ⎞
fcc = Elastic critical stress in compression, and ⎜ fcc = 2⎟
⎜ ⎛ l⎞ ⎟
⎜⎝ ⎜ ⎟
⎝ r ⎠ ⎟⎠

E = Young’s modulus of elasticity of steel 2×105 N/mm2 (MPa)

n = Imperfection factor. It is assumed as 1.4.
The values of permissible stresses for axial compression, σac for some of the
Indian standard structural steels (having different values of yield stresses
varying from 220 N/mm2 upto 540 N/mm2) are given in Tables 3.3 (a) and (b),
for convenience as per IS : 800–1984.
148 DESIGN OF STEEL STRUCTURES

Table 3.2 (a) Permissibles stress sac, N/mm2 (Mpa) in axial compression for steel with
l
various yield stress , fy (slendemess ratio, l = )
r

fy →
220 230 240 250 260 280 300 320 340
λ↓
10 132 138 144 150 156 168 180 192 204
20 131 137 142 148 154 166 177 189 201
30 128 134 140 145 151 162 172 183 194
40 124 129 134 139 145 154 164 174 183
50 118 123 127 132 136 145 153 161 168
60 111 115 118 122 126 133 139 146 152
70 102 10 109 112 115 120 125 130 135
80 93 96 98 101 103 107 111 115 118
90 85 87 88 90 92 95 98 101 103
100 76 78 79 80 82 84 86 88 90
110 68 69 71 72 73 74 76 77 79
120 61 62 63 64 64 66 67 67 69
130 55 55 56 57 57 58 59 60 61
140 49 50 50 51 51 52 53 53 54
150 44 45 45 45 46 46 47 47 48
160 40 41 41 41 42 42 42 42 43
170 36 36 37 37 36 37 38 38 38
180 33 33 33 33 33 34 34 34 34
190 30 30 30 30 30 30 31 31 31
200 27 27 28 28 28 28 28 28 28
210 25 25 25 25 25 25 26 26 26
220 23 23 23 23 23 23 23 24 24
230 21 21 21 21 21 21 22 21 22
240 20 20 20 20 20 20 20 20 20
250 18 18 18 18 18 18 18 18 18

Table 3.2 (b) Permissibles stress sac, N/mm2 (Mpa) in axial compression for steel with
l
various yield stress , fy (slendemess ratio, l = )
r

fy 340 360 380 400 420 450 480 510 540

10 204 215 227 239 251 269 287 305 323
20 201 212 224 235 246 263 280 297 314
30 194 204 215 225 236 251 266 280 295
40 183 192 201 210 218 231 243 255 267
50 168 176 183 190 197 207 216 225 233
Contd.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 149

Table 3.2 (b) Contd.

fy 340 360 380 400 420 450 480 510 540

60 152 158 163 168 173 180 187 193 199
70 135 139 142 147 150 155 160 164 168
80 118 121 124 127 129 133 136 139 141
90 103 105 108 109 111 114 116 118 119
100 90 92 93 94 96 97 99 100 101
110 79 80 81 82 83 84 85 86 87
120 69 70 71 71 72 73 73 74 75
130 61 61 62 62 63 63 64 64 65
140 54 54 54 55 55 56 56 56 57
150 48 48 48 49 49 49 49 50 50
160 43 43 43 43 43 44 44 44 44
170 38 38 39 39 39 39 39 39 39
180 34 35 35 35 35 35 35 35 35
190 31 31 31 31 32 32 32 32 32
200 28 28 28 28 28 28 28 28 28
210 26 26 26 26 26 26 26 26 26
220 24 24 24 24 24 24 24 24 24
230 22 22 22 22 22 22 22 22 22
240 20 20 20 20 20 20 20 20 20
250 18 19 19 19 19 19 19 19 19
Professor D. Allen suggested the use of this formula for the design of columns
in his paper titled as ‘Merchant–Rankine Approach to Member Stability’ a
technical note published in Journal of Structural Division, ASCE, Vol. 104. No.
ST 12, December 1978. Depending on the degree of residual stresses, the value
of parameter takes the value of 1, 1.4 or 2. The value of n is taken as 14 for
riveted and normal welded members. Its value is adopted as 1 where the high
residual stresses are expected.
A constant stress for the members of very small slenderness ratio is not
indicated by Merchant–Rankine formula. A cutoff point below which the
permissible stress is constant, is specified by other standards to include this
feature. A correction factor in the Merchant–Rankine expression was suggested
by Maquoi and J. Rondal, in their discussions on ‘Merchant–Rankine Approach
to Member Stability’ by D. Allen published in Vol. 105, No. ST 11, November,
1979 as below

1 ⎛ fcc ⋅ fy ⎞
σac = .⎜ 1 ⎟ ... (iv)
F .S. ⎜
2n ⎤ n ⎟
⎝⎣
⎡ n
(n
)
⎜ ⎢( fy ) + ( fcc ) 1 − 0.15 ⎥ ⎟
⎦ ⎠
This expression (iv) includes the constant yield plateau for the columns of
slendemess ratio less than about 30. It gives about one to four per cent difference
in values given by the expressions (iii) and (iv). IS : 800–1984 does not specify
the expression (iv).
150 DESIGN OF STEEL STRUCTURES

Ashok K. Jain in his paper referred above has mentioned in the conclusion
that Merchant–Rankine formula is very simple. It has lot of potential for use in
different situations. It is more convenient for use in case the multiple column
curves are to be developed. The value of imperfection factor is simply changed.
Only three parameters, namely, yield stress, critical stress and imperfection
factor appear in Merchant–Rankine formula. A computer-program may easily
be written for use of Merchant–Rankine formula. The values of allowable stresses
for columns of slenderness ratio less than 90 obtained by Merchant–Rankine
formula are higher than those found by the secant formula. However, these
higher values closely agree with those determined by other national standards.

y
γx = 0.28 9 d γx = 0 .36 d γx = 0 .36 d
x γy = 0 .45 d γy = 0 .53 b
γy = 0.28 9 d

γx = 0.40 dm γx = 0 .36 d γx = 0 .39 d

x γy = 0 .6 0 b γ = 0 .55 b
y

x γx = 0 .25 d γx = 0 .41 d γx = 0 .42 d

γy = 0 .32 b

x γx = 0 .35 d m γx = 0 .38 d γx = 0.35 d

γy = 0 .22 b
y
γx = 0 .3 d γx = 0 .42 d γx = 0 .31 d
x γy = 0 .3 d γy = 0 .25 d

γx = 0 .30 d γx = 0 .40 d γx = 0 .39 d

x x γy = 0 .21 b γy = 0 .21 b

γx = 0.42 d γx = 0 .31 b γx = 0 .42 d

γy = 0.42 d

γx = 0 .42 b
γx = 0 .29 d γx = 0 .42 d
γy = 0 .42 b γy = 0 .25 b γy = 0 .23 b

γx = 0 .39 d γx = 0 .29 d γx = 0 .30 d

γy = 0 .21 b γy = 0 .24 b γy = 0 .17 b

γx = 0.45 d γx = 0 .25 d
γy = 0.24 b γy = 0 .21 b

Ap pro xim ate rad iu s of gyra tio n for co m m on

section s for co m pre ssion m em b ers

Fig. 3.8 Common sections for compression members

(Reference : J.A.L. Waddell ‘Bridge-engineering’ John Wiley and Sons, Inc.

New York 1925).
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 151

3.9 COMMON SECTIONS OF COMPRESSION MEMBERS

The common sections used for compression members are shown in Fig. 3.8 with
their approximate radii of gyration.
A column or a compression member may be made of many different sections
to support a given load. Few sections satisfy practical requirement in a given
case. A tubular section is most efficient and economical for the column free to
buckle in any direction. The radius of gyration r for the tubular section in all
the directions remains same. The tubular section has high local buckling
Strength. The tubular sections are suitable for medium loads. However, it is
difficult to have their end connections. A solid round bar having a cross-sectional
area equal to that of a tubular section has radius of gyration, r much smaller
than that of tube. The solid round bar is less economical than the tubular section.
The solid round bar is better than the thin rectangular section or a flat strip.
The radius of gyration of flat strip about its narrow direction is very small.
Theoretically, the rods and bars do resist some compression. When the length of
structural member is about 3 m, then, the compressive strengths of the rods
and bars are very small.
Single angle sections are rarely used except in light roof trusses, because of
eccentricity at the end connections. Tee sections are often used in roof trusses.
The single rolled steel I-section and single rolled steel channel section are
seldom used as column. The value of radius of gyration r, about the axis parallel
to the web is small. The intermediate additional supports in the weak direction
make the use of these sections economical. Sometimes the use of I-sections and
channel sections are preferred because of the method of rolling at the mills,
since, the out-to-out dimensions remain same for a given depth. This failure is
not there with other rolled steel sections.
The costs of single rolled steel sections per unit weight are less than those of
built-up sections. Therefore the single rolled steel sections are preferred so long
as their use is feasible.

3.10 STRENGTH OF COMPRESSION MEMBERS

The strength of a compression member is defined as its safe load carrying
capacity. The strength of a centrally loaded straight steel column depends on
the effective cross-sectional area, radius of gyration (viz., shape of the cross-
section), the effective length, the magnitude and distribution of residual stresses,
annealing, out of straightness and cold straightening. The effective cross-
sectional area and the slenderness ratio of the compression members are the
main features, which influence its strength. In case, the allowable stress is
assumed to vary parabolically with the slenderness ratio, it may be proved that
the efficiency of a shape of a compression: member is related to A/r2. The efficiency
of a shape is defined as the ratio of the allowable load for a given slenderness
ratio to that for slenderness ratio equal to zero. The safe load carrying capacity
of compression member of known sectional area may be determined as follows :
Step 1. From the actual length of the compression member and the support
conditions of the member, which are known, the effective length of the member
is computed.
152 DESIGN OF STEEL STRUCTURES

Step 2. From the radius of gyration about various axes of the section given in
section tables, the minimum radius of gyration (rmin) is taken. rmin for a built-up
section is calculated.
Step 3. The maximum slenderness ratio (l/rmin) is determined for the
compression member.
Step 4. The allowable working stress (σac) in the direction of compression is
found corresponding to the maximum slenderness ratio of the column from IS :
800–1984.
Step 5. The effective sectional area (A) of the member is noted from structural
steel section tables. For the built-up members it can be calculated.
Step 6. The safe load carrying capacity of the member is determined from
P = (σac . A)
where P is the safe load

3.11 ANGLE STRUTS

The compression members consisting of single sections are of two types :
1. Continuous members; 2. Discontinuous members.

3.11.1 Continuous Members

The compression members (consisting of single or double angles) which are
continuous over a number of joints, are known as continuous members. The top
chord members of truss girders and principal rafters of roof trusses are
continuous members.
The effective length of such compression members is adopted between 0.7
and 1.0 times the distance between the centres of intersections, depending upon
degree of end restraint provided. When the members of trusses buckle in the
plane perpendicular to the plane of the truss, the effective length shall be taken
as 1.0 times the distance between the points of restraint.
The working stresses for such compression members is adopted from IS :
800–1984 corresponding to the slenderness ratio of the member, and yield stress
for steel.

3.11.2 Discontinuous Members

The compression members which are not continuous over a number of joints,
i.e., which extend between two adjacent joints only, are known as discontinuous
members. The discontinuous members may consists of
(i) Single angle strut, or (ii) Double angle strut.
When an angle strut is connected to a gusset plate or to any structural member
by one leg, the load transmitted through the strut, is eccentric on the section of
the strut. As a result of this, bending stress is developed along with direct stress.
While designing or determining strength of an angle strut, the bending stress
developed because of eccentricity of loading is accounted for as follows :
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 153

(A) Single Angle Strut. (i) When single angle discontinuous strut is
connected to a gusset plate with one rivet as shown in Fig. 3.9 (a) its effective
length is adopted as centre to centre of intersection at each end and the allowable
working stress corresponding to the slenderness ratio of the member is reduced
to 80 per cent. However, the slenderness ratio of such single angle strut should
not exceed 180.

(a ) S ing le an g le strut con ne cted w ith on e rivet

(b ) S ing le a ng le strut con ne cted tw o o r m ore rivets

Fig. 3.9 Single angle strut

(ii) When a single angle discontinuous strut is connected with two or more
number of rivets or welding as shown in Fig. 3.9 (b) its effective length is adopted
as 0.85 times the length of strut centre to centre of intersection of each end, and
allowable working stress corresponding to the slenderness ratio of the member
is not reduced.
(B) Double Angle Strut. (i) A double angle discontinuous strut with angles
placed back to back and connected to both sides of a gusset or any rolled steel
section by not less than two rivets or bolts or in line along the angles at each
end or by equivalent in welding as shown in Fig. 3.10 (a), can be regarded as an
axially loaded strut. Its effective length is adopted as 0.85 times the distance
between intersections, depending on the degree of restraint provided and in the
plane perpendicular to that of the gusset, the effective length T shall be taken
as equal to the distance between centres of intersections.
The tacking rivets should be provided at appropriate pitch.
(ii) The double angles, back to back connected to one side of a gusset plate or
a section by one or more rivets or bolts of welds as shown in Fig. 3.10 (b), these
are designed as single angle discontinuous strut connected by single rivet or
bolt.
154 DESIGN OF STEEL STRUCTURES

+ + +
Tackin g rivets

(a ) D o ub le ang le strut (a ng le s placed b ack to b ack)

con ne cted to b oth side s of gu sse t

+ + + + +
Tackin g rivets

(b ) D o ub le a ng le strut (a ng le s placed b ack to ba ck)

con ne cted to o ne s id e of gu sse t
Fig. 3.10

If the struts carry, in addition to axial loads, loads which cause transverse
bending, the combined bending and axial stresses shall be checked as described
for the columns subjected to eccentric loading.
The tacking rivets should be provided at appropriate pitch.
The tacking rivets are also termed as stitching rivets. In case of compression
members, when the maximum distance between centres of two adjacent rivets
exceeds 12 t to 200 mm whichever is less, than tacking rivets are used. The
tacking rivets are not subjected to calculated stress. The tacking rivets are
provided throughout the length of a compression member composed of two
components back to back. The two components of a member act together as one
piece by providing tacking rivets at a pitch in line not exceeding 600 mm, and
such that minimum slenderness ratio of each member between the connections
is not greater than 40 or 0.6 times the maximum slenderness ratio of the strut
as a whole, whichever is less.
In case where plates are used, the tacking rivets are provided at a pitch in
line not exceeding 32 times the thickness of outside plate or 300 mm, whichever
is less. Where the plates are exposed to weather the pitch in line shall not exceed
16 times the thickness of the outside plate or 200 mm, whichever is less. In both
cases, the lines of rivets shall not be apart at a distance greater than these
pitches.
The single angle sections are used for the compression members for small
trusses and bracing. The equal angle sections are more desirable usually. The
unequal angle sections are also used. The minimum radius of gyration about
one of the principal axis is adopted for calculating the slenderness ratios. The
minimum radius of gyration of the single angle section is much less than the
other sections of same cross-sectional area. Therefore, the single angle sections
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 155

are not suitable for the compression member of long lengths. The single angle
sections are commonly used in the single plane trusses (i.e., the trusses having
gusset plates in one plane). The angle sections simplify the end connections.
The tee-sections are suitable for the compression members for small trusses.
The tee-sections are more suitable for welding.

3.12 BUILT–UP COMPRESSION MEMBERS

The built-up compression members are needed when the single rolled steel
sections are not sufficient to furnish the required cross-sectional area.

(a ) (e ) (i)

(b ) (f) (j)

(c) (g ) (k)
cyy
cyy

y
(d ) (h ) (l)
Fig. 3.11 Built-up compression members

A built-up compression member may consist of two or more rolled structural

steel sections connected together effectively and acts as one compression member.
The built-up compression members are given effective column cross-sections.
The built-up sections are used for one or more of the reasons mentioned below.
156 DESIGN OF STEEL STRUCTURES

(i) The built-up sections provide large cross-sectional area which cannot
be furnished by single rolled steel sections.
(ii) The built-up sections provide special shape and depth. The special shape
and depth facilitate connections between the different members.
(iii) The built-up sections provide sufficient large radius of gyration of more
desirable ratio of the radii of gyration in two different directions. In the single
rolled steel section, the ratio of radii of gyration in two directions cannot be
altered.
The element of built-up members are placed at farthest possible distance
from centroid of the section. It gives the largest radius of gyration and the
smallest slenderness ratio for constant cross-sectional area. The material of the
column is used at the greatest possible stress. As the size of compression member
increases with constant area, the slenderness ratio becomes less and less due to
increase of radius of gyration. The compressive stress increases accordingly. At
the higher compressive stress the local buckling of the member may take place.
The local buckling of the member limits the size of the compression member.
The material of the column is used most economically when radii of gyration of
the compression members about both the axes are kept approximately equal.
The various shapes of built-up members are shown in Fig. 3.11.
When two channel sections are used back to back as shown in Fig. 3.11 (h) the
maximum spacing between face to face of web can be computed as given below.
yy-axes are the vertical axes of individual channel sections, and Cyy represents
the distance between yy-axis and external face of web as shown in Fig. 3.11 (h).
For most economical section, the radius of gyration about xx-axis (rxx) should be
approximately equal to the radius of gyration about yy-axis (ryy) of built-up section
(i.e., Ixx @ Iyy). The moment of inertia about xx-axis is twice the moment of
inertia of individual sections, and therefore it is known.
⎛S ⎞
IYY = 2Iyy + 2A ⎜ + C yy ⎟ @ Ixx
⎝2 ⎠
where, A = effective sectional areas of one component member
S = spacing between back to back of channel section.
The spacing (S) may be computed and rounded off in millimetres which can
be adopted practically. The moment of inertia Iyy is recomputed for the built-up
section for the adopted spacing.
In ISI Handbook No. 1 ‘Structural Steel Sections’, tables are given for various
values of spacing and the corresponding moment of inertia about yy-axis and
radius of gyration ryy. Comparing the values of Iyy with Ixx, that spacing (S) is
adopted which gives Iyy = Ixx.
The primary object of built-up section is to furnish large radius of gyration
and to support heavy loads. However, the arrangement of different sections
must allow case of fabrication, connection and maintenance (e.g., painting). The
lattice material used should be small.
The two-angle sections placed back to back are most frequently used in the
roof trusses. The built-up sections consisting of such sections are particularly
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 157

economical when single gusset plates are used. The angle sections are tied by
rivets at suitable spacings with filler plates between the angles. The unequal
angle sections are preferred for such built-up section, in order to obtain radii of
gyration about two centroidal axes approximately equal.
Four angle sections are used in a built-up section to maintain the overall
depth of the section. In one case, two pairs of four angles are formed, each pair
is made of two angles. Two pairs are connected by lacing, when the load is
small. Two pairs are connected with a solid plate forming a built-up I-section
when the load is large. The long legs of angle sections are kept outstanding. It
gives higher value of radius of gyration. The four angle sections are also used to
form a box section. It provides large value of radius of gyration. Such a built-up
section is economical to carry small load and when the length of the member is
long. Such a built-up section needs large amount of lacing. A section built-up of
four angles is used for compression members having large length (e.g., masts,
crane boom etc.). It provides equal stiffness in both the directions.
A built-up section consisting of two channels back to back is occasionally used.
Such a built-up section has small radius of gyration about y-axis. A built-up
section consisting of two channel sections placed at a distance apart is more
frequently used. The flanges of channels are kept outward. It gives large value
of radius of gyration. When the large value of radius of gyration is not necessary,
then, the flanges are kept inward. It reduces the amount of lacing. A built-up
section is also made of two channel sections with a solid plate on one side or on
both the sidess. The built-up section with solid plate on one side permits the
access for inspection and painting. It is not possible, when solid plates are used
on both the sides. Two channel sections with solid plates constituting built-up
section, provide large cross-sectional area. The bracing between the parts is
also stiff.
The built-up sections are also made of solid plates and angle sections to have
very large cross-sectional area. When the channels of maximum size do not
provide sufficient area, then, such built-up sections are used. Such built-up
sections are used in the bridges.
It is to note that the cost of built-up section per unit weight is high because of
fabrication and additional material needed for lacing the compression members.
The single rolled steel sections are more economical for the equal area and equal
slenderness ratio, if the normal connections may be designed for the member.

3.13 THICKNESS OF ELEMENTS IN COMPRESSION MEMBERS

When any plates element is subjected to direct compression bending, or shear
stresses or to combination of these stresses, the plate element may buckle locally
before the entire member fails. In local buckling, waves (single wave or series of
waves) or wrinkles are formed in the elements of compression members.
ISI in IS : 800–1962 recommended the thickness of element in compression
members in terms of outstanding lengths of compression members. These
thicknesses have been recommended to avoid the failure of elements owing to
local buckling.
The outstanding length of compression member in case of plated beam is
measured from the free edge to the first row of rivet, as shown in Fig. 3.12 (a).
158 DESIGN OF STEEL STRUCTURES

In case of channels, angles and stem of tee-section, it is measured as nominal

width of sections as shown in Fig. 3.12 (b) and in case of flanges of beams and
tee-section, it is measured as half the nominal width, as shown in Fig. 3.12 (c).
First ro w o f rive ts

O utstan d

(a ) P late d be am se ction

O utstan d O utstan d

O utstan d

(b ) C h an ne ls A n gle S te m o f te e-
section section section

O utstan d O utstan d

(c) Flan ge s of b ea m se ctio n Flan ge s of te e-se ctio n

Fig. 3.12 Out standing lengths compression members

The thickness is taken as 1/16 of outstand of any member in compression for

unstiffened outstand and 1/20 of outstand of any member in compression for
stiffened outstand. The thickness is taken as 1/90 of outstand for unsupported
width of unstiffened plate forming any part of a compression member, measured
between adjacent line of rivets. But for the purpose of calculating effective
sectional area and radius of gyration only 50 times the thickness of plate will be
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 159

taken instead of 90 times the thickness in case single plate is used. However, it
is total thickness of two or more plates effectively tacked together.
All hot rolled steel sections as per IS : 808–1964 (beams, channels and angle
sections) conform to outstand requirements as per IS : 800–1962 described above.
As per the recommendations in IS : 800–1984, the projection of a plate or
flange beyond its connection to a web, or other line of support or the like, exceeds
the relevant values given in (i), (ii) and (iii) below, the area of excess flange
shall be neglected when the effective geometrical properties of the section are
determined :
(i) Flanges and plates in compression with unstiffened edges.
The excess projection if exceeds 256t/(fy)1/2 subjected to a maximum of 16t
shall be neglected.
(ii) Flanges and plates in compression with stiffened edges.
The excess projection if exceeds 20 t to the innermost face of the stiffening.
(iii) Flanges and plates in tension the excess projection if exceeds 20 t.
Note. 1. The stiffened flanges shall include flanges composed of channels or I-
sections or of plates with continuously stiffened edges.
2. t denotes the thickness of the flange of a section or of a plate in compression, or
the aggregate thickness of the plates, if connected together appropriately.
Example 3.1 A rolled steel beam section HB 350 @ 0.674 kN/metre is used as
a stanchion. If the unsupported length of the stanchion is 4 metres, determine
safe load carrying capacity of the section.
Solution
Step 1 : Properties of I-section
HB 350, @ 0.674 kN/m section is used as a stanchion.
From steel section tables, the geometrical properties of the section are as
follows :
Sectional area A = 8591 mm2
Radius of gyration rxx = 149.3 mm
Radius of gyration ryy = 53.4 mm
Step 2 : Slendernes ratio, rmin = 53.4 mm
Unsupported length l = 4 metres
l ⎛ 4 × 1000 ⎞
Slenderness ratio of the stanchion =⎜ ⎟ = 75
rmin ⎝ 53.4 ⎠
Step 3 : Safe load
l
From IS : 800–1984 for = 75.0, and the steel having yield stress, fy = 260 N/
r
mm2, allowable working stress in compression
σac = 109 N/mm2 (MPa).
The safe load carrying capacity of the stanchion
⎛ 109 × 8591 ⎞
P = (σac – A) = ⎜ = 936.42 kN
⎝ 1000 ⎟⎠
160 DESIGN OF STEEL STRUCTURES

Example 3.1 (a) In Example 3.1, in case standard column section SC 25°, @
85.6 kg/m is used as a column determine the safe load-carrying capacity of the
section.
Solution
Step 1 : Properties of I-section
From IS : 808–1964, the geometrical properties of the section are as follows.
Sectional area A = 109 × 102 mm2
Radius of gyration ra = 107 mm
Radius of gyration ryy = 54.6 mm
Step 2 : Slenderness ratio
Minimum radius of gyration
rmin = 54.6 mm
Unsupported length of the stanchion
l = 4000 mm
Slenderness ratio
⎛ l ⎞ ⎛ 4000 ⎞
⎜ ⎟ = ⎜ = 73.26
r
⎝ min ⎠ ⎝ 54.6 ⎠⎟
Step 3 : Safe load
⎛l⎞
From IS : 800–1984 for ⎜ ⎟ = 73.26 and the steel having yield sress, fy = 260
⎝r⎠
N/mm2, allowable working stress in compression
⎛ 12 ⎞
σac = ⎜115 − × 3.26 ⎟ = 111.088 N/mm2
⎝ 10 ⎠
The safe load carrying capacity of the stanchion
⎛ 111.088 × 109 × 100 ⎞
P = (σac . A) = ⎜ ⎟
⎝ 1000 ⎠
= 1210.86 kN
Example 3.2 A single angle discontinuous strut ISA 150 mm× 150 mm ×
12 mm (ISA 150 150, @ 0.272 kN/m) with single riveted connection is 3.5 metres
long. Calculate safe load carrying capacity of the section.
Solution
Step 1 : Properties of angle section
ISA 150 mm × 150 mm × 12 mm (ISA 150 150, @ 0.272 kN/m) is used as
discontinuous strut. From steel section tables
Sectional area A = 3459 mm2
ryy = rxx = 46.1 mm
ruu = 58.3 mm, rvv = 29.3 mm
Step 2 : Slenderness ratio
Effective length of strut l = 3.5 m
rmin = 29.3 mm
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 161

Minimum radius of gyration

l ⎛ 3.5 × 1000 ⎞
Slenderness ratio of strut = ⎜ ⎟ = 119.5
rmin ⎝ 29.3 ⎠
Step 3 : Safe load
l
From IS : 800–1984, the allowable working stress in compression for =
rmin
119.5 and the steel having yield stress 260 N/mm2
σac = 64.45 N/mm2
For single angle discontinuous strut with single riveted connection, allowable
working stress
0.80 σac = (0.8 × 64.45) = 51.56 N/mm2
The safe load carrying capacity
⎛ 51.56 × 3459 ⎞
P = ⎜ ⎟ = 178.346 kN
⎝ 1000 ⎠
Example 3.2 (a) In case in Example 3.2, a discontinuous strut 150 × 150 × 15
angle section is used, calculate the safe load carrying capacity of the section
Solution
Step 1 : Properties of angle section
Angle section 150 × 150 × 15 is used as a discontinuous strut. From IS : 808–
1984 cross-section area. A = 43 × 100 mm2
Radius of gyration rxx = ryy = 45.7 mm
ruu = 57.6 mm
rxv = 29.3 mm
Step 2 : Slenderness ratio
Minimum radius of gyration
rmin = 2.93 mm
Effective length of strut
l = 3.5 × 1000 mm
Slenderness ratio of strut
⎛ l ⎞ ⎛ 3500 ⎞
⎜ ⎟ = ⎜ ⎟ = 119.5
r
⎝ min ⎠ ⎝ 29.3 ⎠
Step 3 : Safe load
⎛l⎞
From IS: 800–1984, the allowable working stress in compression for ⎜ ⎟ = 119.5
⎝r⎠
and steel having yield stress 260 N/mm2
σac = 64.45 N/mm2
For single angle discontinuous strut with single riveted connection, allowable
working stress
0.80 σac = (0.80 × 64.45) = 51.56 N/mm2
162 DESIGN OF STEEL STRUCTURES

The safe load carrying capacity

⎛ 51.56 × 4300 ⎞
P = ⎜ ⎟ = 221.708 kN
⎝ 1000 ⎠
Example 3.3 In Example 3.2, if single angle discontinuous strut is connected
with more than two rivets in line along the angle at each end, determine safe
load carrying capacity of the section.
Solution
Step 1 : Properties of angle section
Discontinuous strut ISA 150 mm × 150 mm × 12 mm (ISA 150 150, @ 0.272
kN/m) is used with double riveted connections.
Minimum radius of gyration
rmin = 29.3 mm
Sectional area, A = 3459 mm
Length of strut between centre to centre of intersection
L = 3.50 m
Step 2 : Slenderness ratio
Effective length of discontinuous strut double riveted
0.85 × L = (0.85 × 3.50) = 2.975 m
Slenderness ratio
l ⎛ 2.975 × 1000 ⎞
= ⎜ ⎟ = 101.5
rmin ⎝ 29.3 ⎠
Step 3 : Safe load
l
Allowable working stress, for = 101.5 and the steel having yield stress
rmin
as 260 N/mm2
σac = 71.65 N/mm2 (MPa)
Allowable working stress for discontinuous strut double riveted is not reduced.
Safe load carrying capacity
⎛ 71.65 × 3459 ⎞
P = ⎜ ⎟⎠ = 247.84 kN
⎝ 1000
Example 3.4 A double angle discontinuous strut ISA 125 mm × 95 mm × 10
mm (ISA 12595, @ 0.165 kN/m) long legs back to back is connected to both the
sides of a gusset plate 10 mm thick with 2 rivets. The length of strut between
centre to centre of intersections is 4 metres. Determine the safe load carrying
capacity of the section.
Solution
Step 1 : Properties of angle section
The double angle discontinuous strut 2 ISA 125 mm × 95 mm × 10 mm (ISA
12595, @ 0.165 kN/m) is shown in Fig. 3.13. Assume the tacking rivets are used
along the length.
From steel section tables (properties of two angles back to back) Sectional
area,
A = 4204 mm2
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 163

rxx = 39.4 mm
Angles are 10 mm apart
ryy = 40.1 mm
Step 2 : Slenderness ratio
Minimum radius of gyration
∴ rmin = 39.4 mm
Effective length
l = 0.85 L = (0.85 × 4) = 3.40 m
Slenderness ratio
⎛ 3.40 × 1000 ⎞
⎜ 39.4 ⎟ = 86.4
⎝ ⎠

95 m m 95 m m

X X
12 5 m m

10 m m

Fig. 3.13
Step 3 : Safe load
l
From IS : 800–1984 safe allowable stress in compression for = 86.4 and
rmin
the steel having yield stress as 260 N/mm2
σac = 95.96 N/mm2
Safe load carrying capacity
⎛ 95.96 × 4204 ⎞
P = ⎜ ⎟ = 403.416 kN
⎝ 1000 ⎠
Example 3.5 In Example 3.4 if double discontinuous strut is connected to
one side of a gusset, determine safe load carrying capacity of the strut.
Solution
Step 1 : Properties of angle sections
A double angle discontinuous strut 2 ISA 125 mm × 95 mm × 10 mm (ISA
12595, @ 0.165 kN/m) connected to one side of a gusset is as shown in Fig. 3.14.
Assume that tacking rivets are used along its length.
Effective length of strut whether single riveted or double riveted
l = L=4m
164 DESIGN OF STEEL STRUCTURES

From steel section tables.

Section area A = 4204 mm2 rxx = 39.4 mm
Distance back to back of angles is zero
ryy = 36.7 mm

12 5 m m
95 m m 95 m m

Fig. 3.14

Step 2 : Slenderness ratio

Minimum radius of gyration
∴ rmin = 36.7 mm
Slenderness ratio of strut
l ⎛ 4 × 1000 ⎞
= ⎜ ⎟ = 109
rmin ⎝ 36.7 ⎠
Step 3 : Safe load
l
From IS : 800–1984, allowable working stress in compression for = 109
rmin
and the steel having yield stress as 260 N/mm2
σac = 73.9 N/mm2
For above strut σac = 0.8 × 73.9 N/mm2
Safe load carrying capacity

⎛ 0.8 × 73.9 × 4204 ⎞

P = ⎜ ⎟⎠ = 248.54 kN
⎝ 1000
Example 3.6. In Example 3.4 double angle strut is continuous and connected
with a gusset plate with single rivet, determine safe load carrying capacity of
strut.
Solution
Step 1 : Properties of angle sections
Assume that tacking rivets are used along the length.
The double angle continuous strut 2 ISA 125 mm × 95 mm × 10 mm (ISA
12595, @ 0.165 kN/m) is singly riveted as shown in Fig. 3.13.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 165

rxx = 39.4 mm,

ryy = 40.1 mm
Sectional area = 4204 mm2
Length of strut between centre of intersection is 4 m.
Step 2 : Slenderness ratio
Effective length = l = L = 4 m
rmin = 39.4 mm
l ⎛ 4 × 1000 ⎞
Slenderness ratio = ⎜ ⎟ = 101.5
rmin ⎝ 39.4 ⎠
Step 3 : Safe load
l
Allowable working stress in compression for = 101.5 and the steel having
rmin
yield stress as 260 N/mm2
σac = 71.65 N/mm2
Safe load carrying capacity of strut
⎛ 71.65 × 4204 ⎞
P = ⎜ ⎟⎠ = 301.22 kN
⎝ 1000
Example 3.7 A built-up column consists of three rolled steel beam sections
WB 450 @ 0.794 kN/m, connected effectively to act as one column as shown in
Fig. 3.15. Determine the safe load carrying capacity of built-up section, if
unsupport length of column is 4.25 m.
Solution
Step 1 : Properties of built-up section
The built-up column consists of 3 WB 450 @ 0.794 kN/m
From steel section tables
Area of section = (3 × 101.15 × 100) = 30345 mm2
Moment of inertia of built-up section about xx-axis
Ixx = [2 × 35057.6 + 1706.7] × 104 = 71821 × 104 mm4
Moment of inertia of built-up section about yy-axis
Iyy = [35057.6 × 104 + 2 × 1706.7 × 104 + 2Ah2]
= 38471.0 × 104 + 2Ah2
where A = sectional area of one I-section
1
Iyy = [38471.0 + 2 × 101.15 (22.5 + 2 × 0.92)2] × 104 mm4
= 145115.79 × 104 mm4
Step 2 : Slenderness ratios
Ixx is less than Iyy. Therefore, rxx is minimum
1/2
⎛ 71821.9 × 104 ⎞
rmin = ⎜ ⎟ = 154 mm
⎝ 30345 ⎠
166 DESIGN OF STEEL STRUCTURES

Unsupported length of column

l = L = 4.25 m
Slenderness ratio of column
l ⎛ 4.25 × 1000 ⎞
= ⎜ ⎟ = 27.6
rmin ⎝ 154.0 ⎠

IS W B 45 0 @ 0.7 94 kN /m

x x

Fig. 3.15

Step 3 : Safe load

From IS : 800–1984, allowable working stress in compression for steel having
yield stress as 260 N/mm2
σac = 151.72 N/mm2
Safe load carrying capacity of built-up column
⎛ 151.72 × 303.45 ⎞
P = ⎜ ⎟ = 46.0394 kN
⎝ 1000 ⎠

3.14 DESIGN OF AXIALLY LOADED COMPRESSION MEMBER

When a column or compression member is designed, for given load, actual length
of the member and its support conditions, the cross-sectional shape of the member
is determined.
The cross-sectional shape of axially loaded compression member depends
largely on whether the compression member is long or short and whether it
carries a small load or a large load. It is difficult to decide, whether a column is
short or long. It is arbitrarily decided.
When the slenderness ratio of a column is less than 60, then, it may be
considered as a short column. When the slenderness ratio is between 60 and
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 167

180, the columns may be considered as long column. Following are the length
and load categories arbitrarily made for design of compression members :
1. Short compression members with small loads.
2. Short compression members with large loads.
3. Long compression members with small loads.
4. Long compression members with intermediate load.
The strength of axially loaded compression member depends upon slenderness
⎛ l ⎞
ratio ⎜ ⎟ . For the design of axially loaded compression member load to be
⎝ rmin ⎠
carried, the length of compression member and end conditions are known. The
effective length of the compression member for the given end conditions is
computed. The radius of gyration of compression member is not known as the
cross-sectional shape of the compression member is not known. The allowable
working stress in compression can be found when the slenderness ratio is known.
There is no direct method of designing a compression member. The compression
member is designed by trial and error method. The design of compression member
is also done by using safe load tables, if available.
ISI Handbook No.1 provides tables for safe concentric loads on rolled steel
column sections (HB-sections) for bending about xx-axis and yy-axis. The effective
length of column is determined knowing the end conditions. The value of safe
concentric loads corresponding to respective effective lengths are given for
various sizes of HB-sections. A column section having safe axial load equal to or
slightly greater than the required load on the column is selected.
Design procedure. Following are the usual steps in design of compression
members.
Step 1. The slenderness ratio for the compression member and the value
of yield stress for the steel are assumed.
For the rolled steel beam section compression members, the slenderness ratio
varies from 70 to 90. For struts, the slenderness ratio varies from 110 to 130.
For compression members carrying large loads, the slenderness ratio is about
40.
Step 2. The effective sectional area (A) required for compression member is
determined
⎛ P ⎞
A = ⎜ ⎟
⎝ σac ⎠
where, P = load to be carried by the member.
Step 3. From the steel section tables, section for the compression member of
the required area is selected.
The section for the compression member is selected such that it has the largest
possible radius of gyration for the required sectional area. It should also be
most economical section.
Step 4. Knowing the geometrical properties of the section, slenderness ratio
168 DESIGN OF STEEL STRUCTURES

is computed and allowable axial stress in compression is found from IS : 800–

1984 for the quality of steel assumed.
Step 5. The safe load carrying capacity of the compression member is
determined.
The section selected for the compression member is revised in case the safe
load carrying capacity of the compression member is less than or much larger
than the load to be carried by it.
Example 3.8 Design a rolled steel beam section column to carry an axial
load 1100 kN. The column is 4 metre long and adequately restrained in position
but not in direction at both ends.
Solution
Design : The slenderness ratio for the column and the value of yield stress
for the steel to be used may be assumed as 80 and 260 N/mm2 respectively.
Step 1 : Selection of trial section
Allowable stress as per IS : 800–1984
σac = 103 N/mm2
Effective sectional area required
⎛ 1100 × 1000 ⎞
⎟ = 10679.61 mm
2
⎜ 103
⎝ ⎠
Effective length of the column is 4 m
Step 2 : Properties of trial section
From steel section tables, try HB 450, @ 0.872 kN/m section
Sectional area A = 11114 mm2
rxx = 187.8 mm, ryy = 51.8 mm
Step 3 : Slenderness ratio
∴ rmin = 51.8 mm
l ⎛ 4 × 1000 ⎞
Slenderness ratio = ⎜ ⎟ = 77.2
rmin ⎝ 51.8 ⎠
Step 4 : Check for safe load
From IS : 800–1984, allowable axial stress in compression, for the steel having
yield stress as 260 N/mm2
σac = 106.36 N/mm2
Safe load carrying capacity of the column
⎛ 106.36 × 11114 ⎞
P = ⎜ ⎟⎠ = 1182.08 kN
⎝ 1000
The column section lighter in weight than this is not suitable. Hence the
design is satisfactory.
Provide HB 450, @ 0.872 kN/m section for the column.
Alternatively :
Step 2 : Properties of trial section
From IS : 808–1984, try SC 250, @ 85.6 kg/m (standard column section)
A = 109 × 100 mm2,
rxx = 107.0 mm ryy = 54.6 mm
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 169

Step 3 : Slenderness ratio

Effective length of column is 4000 mm. Minimum radius of gyration, rmin is
54.6 mm slenderness ratio

⎛ l ⎞ ⎛ 4000 ⎞
⎜ ⎟ = ⎜ ⎟ = 73.26
r
⎝ min ⎠ ⎝ 54.6 ⎠
Step 4 : Check for safe load
From IS: 800–1984, allowable stress in axial compression, for the steel having
yield stress as 260 N/mm2
⎛ 12 ⎞
σac = ⎜115 − × 3.26 ⎟ = 111.088 N/mm2
⎝ 10 ⎠
Safe load carrying capacity of the column
⎛ 111.088 × 109 × 100 ⎞
P = ⎜ ⎟ = 1210.86 kN
⎝ 1000 ⎠
Example 3.9 Design a single angle discontinuous strut to carry 110 kN load.
The length of the strut between centre to centre of intersections is 3.25 metres.
Solution
Design :
Step 1 : Selection of trial section
Assuming that the angle strut is connected to the gusset plate with two or
more than two rivets.
Effective length of strut
l = 0.85 L = (0.85 × 3.25 × 1000)
= 2762.5 mm
The slenderness ratio for the single angle discontinuous strut and value of
yield stress for the steel may be assumed as 130 and 260 N/mm2, respectively.
Therefore,
Allowable stress in compression for strut
σac = 57 N/mm2
Effective sectional area required
⎛ 110 × 1000 ⎞
⎟ = 1929.82 mm
2
⎜ 57
⎝ ⎠
The equal angle section is suitable for single angle strut. It has maximum
value for minimum radius of gyration.
Step 2 : Properties of trial section
From steel section tables, try ISA 110 mm × 110 mm × 10 mm (ISA 110110 @
0.165 kN/m).
From steel section tables, try ISA 110 mm × 110 mm × 10 mm (ISA 110110 @
0.165 kN/m).
170 DESIGN OF STEEL STRUCTURES

Sectional area A = 2106 mm2

rxx = ryy = 33.6 mm
rmin = 42.5 mm, ryy = 21.4 mm
Step 3 : Slenderness ratio
⎛ 2762.5 ⎞
Slenderness ratio = ⎜ ⎟ = 129.2
⎝ 21.4 ⎠
Step 4 : Safe load
From IS : 800–1984, allowable working stress in compression for the steel
having yield stress as 260 N/mm2
σac = 57.56 N/mm2
Safe load carrying capacity
⎛ 57.56 × 2106 ⎞
P = ⎜ ⎟ = 121.52 kN
⎝ 1000 ⎠
The angle section lighter in weight than this is not suitable. Hence the design
is satisfactory.
Step 4 : Check for width of outstanding leg
Width of outstanding leg to thickness ratio
⎛ Nominal width ⎞ 110
⎜ Thickness ⎟ = 10 = 11 < 16
⎝ ⎠
Hence, satisfactory.
Provide ISA 110 mm × 110 mm × 10 mm (ISA 110110, @ 0.165 kN/m) for
discontinuous strut.
Alternatively :
Step 2 : Properties of trial section
From IS : 808–1984, try angle section 120 × 120 × 10 (@ 182 kg/m)
Sectional area, A = 2320 mm2
rxx = ryy = 36.7 mm
ruu = 46.3 mm, ryy = 23.6 mm
Step 3 : Slenderness ratio
Effective length of strut is 2762.5 mm
Minimum radius of gyration
rmin = 23.6 mm
Slenderness ratio
⎛ l ⎞ ⎛ 2762.5 ⎞
⎜ ⎟ = ⎜ ⎟ = 117.55
⎝ rmin ⎠ ⎝ 36.6 ⎠
Step 4 : Safe load carrying capacity
From IS : 800–1984, allowable stress in axial compression for the steel having
yield stress as 260 N/mm2
⎛ 9 ⎞
σac = ⎜ 73 − × 7.55 ⎟ = 66.205 N/mm2
⎝ 10 ⎠
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 171

Safe load carrying capacity

⎛ 66.205×2320 ⎞
P = ⎜ ⎟ = 153.596 N/mm
2
⎝ 1000 ⎠
The angle section lighter in weight than this trial section is not suitable.
Hence, the design is satisfactory.
Example 3.10 Design a double angle discontinuous strut to carry 150 kN
load. The length of strut between centre to centre of intersections is 4.00 metre.
Solution
Design :
Step 1 : Selection of trial section
Assuming that the strut is connected to both sides of gusset 10 mm thick by
two or more than two rivets
Length of strut L = 4.00 m
Effective length l = 0.85 L = (0.85 × 4) = 3.40 m
The slenderness ratio of a double angle discontinuous strut and the value of
yield stress for steel may be assumed as 120 and 260 N/mm2, respectively.
Therefore,
Allowable working stress in compression
σac = 64 N/mm2
Effective sectional area required
⎛ 150 × 1000 ⎞
⎟ = 2343.75 mm
2
⎜ 64
⎝ ⎠
Step 2 : Properties of trial section
From steel section tables (properties of two angles back to back) try 2 ISA 100
mm × 65 mm × 8 mm (2 ISA 10065, @ 0.099 kN/m)
Sectional area A = 2514 mm2
rxx = 31.6 mm
For angles having 10 mm distance back to back and long legs vertical
ryy = 27.5 mm
Step 3 : Slenderness ratio
rmin = 27.5 mm
⎛ l ⎞ ⎛ 3.40 × 1000 ⎞
Step 4 : Safe load ⎜ ⎟ = ⎜ ⎟ = 123.6
r
⎝ min ⎠ ⎝ 27.5 ⎠
From IS : 800–1984, allowable working stress in compression for the steel
having yield stress as 260 N/mm2
σac = 61.48 N/mm2
Safe load carrying capacity of the strut
⎛ 61.48 × 2541 ⎞
P = ⎜ ⎟ = 154.50 kN
⎝ 1000 ⎠
Hence, the design is satisfactory.
172 DESIGN OF STEEL STRUCTURES

Provide 2 ISA 100 mm × 65 mm × 8 mm for the strut. Provide tacking rivets

18 mm in diameter at 500 mm spacing.
Example 3.11. A column 5 metre long is to support a load 4500 kN. The ends
of the column are effectively held in position and direction. Design the column if
rolled steel beams and 18 mm plates are only available.
Solution
Design :
Step 1 : Selection of trial section
Length of column = 5 m
Effective length of column is (0.65 × 5) = 3.25 m
In order to support large load, the slenderness ratio for the built-up column
and the value of yield stress for the steel may be assumed as 40 and 260 N/mm2,
respectively.
Allowable working stress in compression
σac = 145 N/mm2
Effective sectional area required
⎛ 4500 × 1000 ⎞
⎟ = 31034.48 mm
2
⎜ 145
⎝ ⎠
Step 2 : Properties of trial section
From steel section tables, try HB 450 @ 0.925 kN/m
rxx = 185.0 mm, ryy = 50.8 mm
For the columns carrying large loads
rxx = ryy

1 40 m m
18 m m

IS H B 4 50 ,
@ 0 .92 5 kN /m
450 mm

18 m m
2 50 m m
7 00 m m

Fig. 3.16
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 173

Step 3 : Slenderness ratio

rmin for the plated built-up column may be estimated as 185 mm.
Therefore the slenderness ratio for the plated column shall be
l ⎛ 3.25 × 1000 ⎞
= ⎜ ⎟ = 17.567
rmin ⎝ 185 ⎠
Step 4 : Area of plates
From IS : 800–1984, allowable working stress in compression for steel having
yield stress as 260 N/mm2
σac = 154.486 N/mm2
Sectional area required
⎛ 4500×1000 ⎞
⎟ = 29128.74 mm
2
⎜
⎝ 154.486 ⎠
Area of HB 450, @ 0.925 kN/m is 11789 mm2
Area to be provided by two cover plates
= 17339.74 mm2
Area to be provided by one plate
= 8669.87 mm2
Plates available = 18 mm
⎛ 8669.87 ⎞
Width required = ⎜ ⎟ = 481.66 mm
⎝ 18 ⎠
Provided 700 mm width of cover plate.
Step 5 : Check for outstanding width; thickness ratio for cover plate
1 560
Outstanding width = (700 – 140) = = 280 mm
2 2
Thickness = 18 mm
⎛ Outstanding width ⎞ ⎛ 280 ⎞
⎜ ⎟ = ⎜ ⎟ = 15.55 < 16. Satisfactory.
⎝ Thickness ⎠ ⎝ 18 ⎠
Step 6 : Check for load carrying capacity
1. Properties of section
1
Iyy of plates = 2× × 1.8 × 703 × 104 mm4
12
= 102900 × 104 mm4
Iyy of H-section
= 3045 × 104 mm4
Iyy of compound section
= 105945 × 104 mm4
Area of section = 117.89 + 2 × 1.8 × 100 mm2
174 DESIGN OF STEEL STRUCTURES

= 369.89 × 100 mm2

1
⎛ 105945 × 104 ⎞2
ryy = ⎜⎜ ⎟⎟ = 169.2 mm
⎝ 369.9 × 100 ⎠
rxx = 185 mm (for I-section alone)
2. Slenderness ratio of column section

⎛ 1 ⎞ ⎛ 3.25 × 1000 ⎞
⎜ ⎟ = ⎜ ⎟ = 19.21
⎝ rmin ⎠ ⎝ 169.2 ⎠
3. Safe load
From IS : 800–1984, allowable working stress in compression for steel having
yield stress as 260 N/mm2
σac = 154.158 N/mm2
Safe load carrying capacity
⎛ 154.158 × 36989 ⎞
⎜ 1000 ⎟ = 5702.15 kN
⎝ ⎠
Hence, the design is satisfactory.
Provide HB 450, @ 0.925 kN/m with two plates 700 mm × 18 mm. One plate
is connected with each flange of I-section.

3.15 LACED AND BATTENED COLUMNS

In built-up columns when rolled steel sections are not connected by plates (viz.
load sharing elements) suitable lateral system is needed to connect different
load carrying elements of column. The lateral system holds the load carrying
elements of the built-up column in their relative positions and it does not share
the load. The object of providing lateral system is to carry the transverse shear
force which occurs when the column deflects. The following lateral systems are
used:
1. Lacing. 2. Batten plates. 3. Perforated plates.
1. Lacing. The lacing is also termed as ‘latticing’ and it is most commonly
used. The rolled steel flats, angles and channels are used for lacing. The rolled
steel sections or tubes of equivalent length may also be used instead of flats.
The lacing is of two types :
1. Single lacing 2. Double lacing.
1. Single lacing. In single lacing as shown in Fig. 3.17 (a), the lacing flats are
placed to form a single-triangular system. The single lacing on opposite sides of
the main components shall preferably be in the same direction so that one may
be shadow of the other, instead of being mutually opposed in direction.
2. Double lacing. In double lacing as shown in Fig. 3.14 (b), the lacing flats
are used to intersect one another and connected at the points of intersections by
rivets.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 175

The compression members comprised of two main components laced and tied
should, where practicable, have a radius of gyration about the axis perpendicular
to the plane of the lacing not less than the radius of gyration at right angles to
that axis (i.e., ryy, should not be less than rxx, see Fig. 3.17).
As far as it is practicable, the lacing system should not be varied throughout
the length of strut. The tie plates are attached at the top and bottom of a laced
column.
The single laced system and double laced system on opposite sides of the
main components should not be combined with cross-members (except the plates)
perpendicular to the longitudinal axis of the strut unless all forces resulting
from deformation of the strut are calculated and accounted for in the lacing and
its fastening.
When the components of built-up column are connected by a lateral system,
the effect of shearing forces to the deflection of the column is more. The reduction
in buckling strength due to the shear deflection is more than that of solid built-
up columns. The buckling strength of laced or battened column is given by

⎛ π2 E ⋅ I ⎞
t
P = ⎜ ⎟
⎜ ( KL )2 ⎟
⎝ ⎠
S p acin g. l
S pa cing . l

y y

x x x x

y y
(a ) S ing le la cing (b ) D o ub le la cing

Fig. 3.17 Lacing of compression members

176 DESIGN OF STEEL STRUCTURES

where Et = Tangent modulus

I = Moment of inertia of the column section, (I0 + 2I1)
I1 = Moment of inertia of each component of column about their own
axes
1
I0 = 2 Ah2
A = Cross-sectional area of each component of column
h = Distance between yy-axis of components of column
For single lacing the factor K is given by
1
⎛ π2 Et I 0 d2 ⎞2
K = ⎜⎜1 + ⋅ ⎟⎟ ...(3.30)
⎝ L2 2Ec ⋅ h2 ⋅ Ad ⎠
and for double lacing
1
⎛ π2 Et I 0 d2 ⎞2
K = ⎜⎜1 + 2
⋅ ⎟⎟ ...(3.31)
⎝ L 2Ec ⋅ h2 ⋅ Ad ⎠
where c = Panel length, as shown in Fig. 3.18
n = Number of panels
d = Length of lacing bars as shown in Fig. 3.18
Ad = Cross-sectional area of lacing bars.
The bending moment at any point on the deflected shape of the column at
buckling
Ma = – Pc .y
The shear force in built-up column is given by
dM 2 dy
V = = −Pc ⋅
dx dx
From Sec. 3.7,
⎡1 − cos nl ⎤
y = e⎢ ⋅ sin (nx ) + cos (nx )⎥
⎣ sin nl ⎦

dy ⎡ nl ⎤
= e ⎢n tan cos (nx ) − n sin (nx )⎥
dx ⎣ 2 ⎦

⎛ dy ⎞
At x = 0, ⎜ ⎟ is maximum.
⎝ dx ⎠

⎛ dy ⎞ nl
∴ ⎜ dx ⎟ = e.n tan
⎝ ⎠max 2
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 177

1
⎛ Pc ⎞ 2
and n = ⎜ ⎟
⎝ EI ⎠
Therefore,
1 1
⎛ Pc ⎞ 2 1 ⎛ Pc ⎞ 2
V = −Pc ⋅ e ⎜
EI ⎟ ⋅ tan 2 ⎜ EI ⎟
⎝ ⎠ ⎝ ⎠

1 1
e ⎛ Pc ⎞ 2 1 ⎛ P ⎞2
= −Pc ⋅ ⎜ ⋅ tan ⎜ c ⎟ ... (3.32)
r ⎝ EI ⎟⎠ 2r ⎝ EI ⎠
This expression for shear force is lengthy. IS : 800–1984 recommended that
lacing bar resists a transverse shear S equal to 2.5 per cent of the axial load in
the member.
For battened column, the factor K is given by
1
⎡ π2 I 2
0 ⎛c⎞ π Et I 0 c ⋅ h ⎤ 2
K = ⎢1 + ⋅⎜ ⎟ + ⋅ ⎥
⎣⎢ 24l1 ⎝L⎠ L2 12EIb ⎦⎥
where, Ib = Moment of inertia of batten plate about the centroidal axis of
the cross-section of column perpendicular to plane of buckling.
For simplification (Et/E) is considered as unity.
Therefore,
1
⎡ π2 I ⎛ c ⎞2 π2 I c ⋅ h ⎤ 2
0 0
K = ⎢1 + ⋅ + ⎥
⎢⎣ 24l1 ⎜⎝ L ⎟⎠ 12 Ib L2 ⎥⎦
The second term of the square root depends on flexibility of the component
moments of built-up column, and last term depends upon flexural rigidity of
batten plates. The last term is small as compared with the sum of first two
terms for a properly designed column, and hence neglected. The expression for
K becomes as under :
1
⎡ π2 I ⎛ c ⎞2 ⎤ 2
0
K = ⎢1 + ⋅ ⎥ ... (3.33)
⎢⎣ 24l1 ⎜⎝ L ⎟⎠ ⎥⎦
The design of lacing or batten plates of built-up column is done in such a way
that the lacing or batten plates do not fail before the load carrying capacity of
whole column is reached. A laced or battened column is considered to collapse if
any element of the column begins to yield locally prior to the buckling load Pcr is
reached. Therefore, IS : 800–1984 specified rules for design of these elements of
built-up column to avoid premature failure of the column.
178 DESIGN OF STEEL STRUCTURES

When perforated plates are used as lateral system then, the factor K for the
column is given by

⎡ 1⎤
⎢ π2 ⎛ a ⎞2 ⎥
K = ⎢1 + ⋅⎜ 2 ⎟ ...(3.34)
2
⎝r ⎠ ⎥
⎢ 12 ⎛⎜ L ⎞⎟ ⎥
⎣⎢ ⎝r⎠ ⎦⎥
where, a = Spacing of perforations
r = Radius of gyration of over-all section of the column
L = Length of the column between supports
rx = Radius of gyration of main segment about its centroidal axis.
Design of lacing. Following are the usual steps required for the design of
lacing as per IS : 800–1984.
Step 1. For lacing the components of built-up column, adopt single or double
lacing. The angle of inclination of lacing bars with the longitudinal axis of the
component member shall not be less than 40° and not more than 70° to the axis
of the member.
Step 2. The spacing of lacing (l) as shown in Fig. 3.17 is computed. The
maximum spacing should be such that the following condition is fulfilled :
(a) Compression members composed of two I-sections. When the
compression members are composed of two I-sections, these shall be connected
together by riveting so that.
Minimum slenderness ratio λ = (l/r) | > 0.7 times most unfavourable
of component of column between slenderness ratio of the ratio
consecutive connections of the column as a whole.
or |
> 50, whichever is less
where, ‘l’ is the distance between the centres of connection of the lattice bars to
each component.
(b) Compression members composed of two channels (back to back).
When the compression members are composed of two angles, two channels or
two tees, back to back in contact or separated by a small distance, these shall be
connected together by riveting so that
Slenderness ratio of component |> 0.6 times most unfavourable
of column slenderness ratio of the column
as a whole
|> 40
Step 3. The lacing bars resist a total transverse shear ‘V’ at any point in the
length of the member where V equal to 2.5 per cent of the axial load in the
member.
The transverse shear V is equally distributed in all transverse systems of
parallel planes. Any other shear due to end moment or transverse loading is
respectively accounted for.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 179

L on gitud in a l a xe s
θ θ

(a ) S ing le la cin g (b ) D oub le la cin g

Fig. 3.18 Transverse shear force in lacing

Step 4. The force in the lacing bar is computed as given below : In single
V
lacing as shown in Fig. 3.18 (a) transverse shear in each lacing is
n

⎛V d ⎞
Force in each lacing, F = ⎜ × ⎟
⎝ n n⎠

⎛V ⎞
or F = ⎜ ⎟ . cosec θ
⎝n⎠
where, θ = Inclination of lacing with the longitudinal axis of member
n = Number of transverse systems in parallel planes
h = Horizontal distance between rivet lines.
In the double lacing as shown in Fig. 3.18 (b) the transverse shear is to be
resisted by two lacing bars.
1 ⎛V ⎞
Force in each lacing = ⋅ ⋅ coses θ⎟
2 ⎜⎝ n ⎠
Step 5. The section for all lacing elements is adopted as below :
Let t be the thickness of flat lacing :
1
(i) In the single lacing, t </ th (one-fortieth) length between inner end
40
rivets.
1
(ii) In the double lacing t </ th (one-sixtieth) length between inner end
60
rivets.
IS : 800–1984 specifies the width of lacing flats as given in Table 3.4.
180 DESIGN OF STEEL STRUCTURES

Table 3.4 Width of lacing bars

Nominal diameter of rivets Width of lacing bars

22 mm 65 mm
20 mm 60 mm
18 mm 55 mm
16 mm 50 mm
The diameter of the rivets connecting lacing bars with components of column
is assumed.
Step 6. The minimum radius of gyration for lacing bar is computed. The
slendemess ratio of the lacing bar for compression member shall not exceed
145.
In the riveted construction, the effective length of lacing bar in single lacing
is adopted as the length between inner end rivets, and in double lacing as 0.7 of
this length. In the welded construction, this effective length is taken as 0.7
times the distance between the inner ends of welds connecting the lacing bars
to the members.
Step 7. The rivet value for the rivet adopted is computed and should be
greater than load coming to the rivet (twice the transverse shear in each lacing.)
Step 8. The lacing bar designed is checked for compressive strength and
tensile strength. These should be more than the respective forces in each lacing
bar.
The end tie plates are provided at the ends of lacing system in the laced
compression members. These tie plates are also provided at the points where
the lacing system is interrupted.
The lacing bars are attached to the main members by rivets or welding. The
riveting or welding of lacing bars to the main members shall be sufficient to
transmit the load in the bars. Where the welded lacing bars overlap the main
members, the amount of lap measured along either edge of the lacing bar shall
be not less than four times the thickness of the bar or the members, whichever
is less. The welding should be sufficient to transmit the load in the bar and
shall, in any case, be provided along each side of the bar for the full length of
lap.
Where the lacing bars are fitted between the main members, these shall be
connected to each member by fillet welds on each side of the bar or by full
penetration butt welds. The lacing bars shall be so placed as to be generally
opposite the flange or stiffening elements of the main members.
Example 3.12. A steel column 12 m long carries an axial load of 1000 kN.
The column is hinged at both the ends. Design an economical built-up section
with double lacing. Design the lacing also.
Solution
Design :
Step 1 : Design of built-up column
1. Selection of trial section
The length of built-up column, L is (12 × 1000) = 12000 mm
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 181

Effective length of built-up column

l = L = (12 × 1000) = 12000 mm ...(ii)
Axial load carried by the column
P = 1000 kN ...(iii)
The slenderness ratio for the laced built-up column and the value of yield
stress for steel may be assumed as 80 and 260 N/mm2, respectively.
Allowable stress in axial compression for the column
σac = 103 N/mm2 ...(iv)
Effective sectional area required
⎛ 1000 × 1000 ⎞
A = ⎜ ⎟ = 9708.74 ...(v)
⎝ 103 ⎠
2. Properties of trial section
From ISI Handbook No. 1 double channels laced used as columns, try 2 MC
350, @ 0.421 kN/m
A = 2 × 5366 = 10732 mm2 ...(vi)
Ixx = 20016 × 104 mm4 ...(vii)
Ixx = 136.6 mm ...(viii)

3 40 m m

2 20
1 00 mm 1 00

60 m m 60 m m

Fig. 3.16 Laced column

3. Spacing of channels
For the built-up column, the spacing between back to back of the channels
should be such that Iyy @ Ixx.
From steel section tables adopt spacing of the channels
S = 200 mm ...(ix)
182 DESIGN OF STEEL STRUCTURES

Iyy = 20246.8 × 104 mm4 ...(x)

ryy = 137.4 mm ...(xi)
4. Maximum slenderness ratio

⎛ 1 ⎞ ⎛ 12000 ⎞
⎜ ⎟ = ⎜ ⎟ = 87.848 ...(xii)
⎝ rmin ⎠ ⎝ 136.6 ⎠
5. Safe load
From IS : 800–1984, maximum permissible stress in axial compression for
the steel having yield stress as 260 N/mm2
σac = 94.367 N/mm ...(xiii)
Safe load carrying capacity of the section
⎛ 94.367×10732 ⎞
P = ⎜ ⎟ = 1012.75 kN
⎝ 1000 ⎠
>1000 kN. Hence, the design is satisfactory.
Step 2 : Design of lacing
1. Spacing of lacing bar
The double lacing is provided at inclination of 45° with the longitudinal axis
of the column member as shown in Fig. 3.19. The spacing of channels S is equal
to 220 mm. The gauge distance, g for rivet lines for MC 350, @ 0.421 kN/m is 60
mm.
∴ S + 2g = (220 + 2 × 60) = 340 mm ...(xiv)
Inclination of lacing is 45°.
Spacing of the lacing is centre to centre of adjacent rivets
= 340 mm ...(xv)
2. Slenderness ratio of component column
Minimum radius of gyration of one channel MC 350, @ 0.421 kN/m
rmin = 28.3 mm ...(xvi)
The column is composed of two channels Slenderness ratio of component of
the column

⎛ 1 ⎞ ⎛ 340 ⎞
⎜ ⎟ = ⎜ ⎟ = 12.014
⎝ rmin ⎠ ⎝ 28.3 ⎠

⎛ 1 ⎞
< 40 and < 0.6 ⎜ ⎟ of column as a whole (i.e., 0.6 × 87.8)
⎝ rmin ⎠
22 mm nominal diameter power driven rivets are used for the connection of
lacing. Width of lacing flat
= 65 mm ...(xvii)
Length of lacing = 340 2 mm ...(xviii)
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 183

1
Thickness of lacing = × 340 2 = 8.1 mm ... (xix)
60
Adopt lacing flat 65 F 10.
Sectional area of flat
65 × 10 = 65 mm2 ...(xx)
Minimum radius of gyration
1
⎡ 1 bt3 ⎤ 2 10
⎢ ⎥ = 1 = 2.88 mm ...(xvi)
⎣12 b ⋅ t ⎦ (12)2
Step 3 : Check
1. Force in lacing bars
Transverse shear to be resisted is 2.5 per cent of axial load
⎛ 2.5 × 1000 ⎞
V = ⎜ ⎟ = 25 kN ...(xxii)
⎝ 100 ⎠
Force in each lacing of double lacing
⎛1 V ⎞ ⎛ 1 2.5 ⎞
⎜ 2 × n × coses θ ⎟ = ⎜ 2 × 2 × 2 ⎟ = 8.84 kN ...(xxiii)
⎝ ⎠ ⎝ ⎠
2. Compressive strength of lacing bars
The effective length of lacing bar in double lacing is 0.7 of its length
= 0.7 × 340 2 mm ...(xxiv)
Slenderness ratio for the lacing bar
⎛ 0.7 × 340 2 ⎞
⎜⎜ ⎟⎟ = 117 < 145
⎝ 2.88 ⎠
Hence the design is satisfactory.
Form IS : 800–1984, allowable stress in axial compression for the steel having
yield stress as 260 N/mm2
σac = 57.7 N/mm2
Compressive strength of the lacing bar
⎛ 57.7 × 65 × 10 ⎞
⎜ 1000 ⎟ = 37.51 kN > Force in each flat ...(xxv)
⎝ ⎠
3. Tensile strength of the lacing flat
⎛ (b − d ) × t × 0.6 fy ⎞
Ft = ⎜ ⎟ kN
⎝ 1000 ⎠

⎛ (65 – 23.5) × 10 × 0.6 × 260 ⎞

Ft = ⎜ ⎟
⎝ 1000 ⎠
= 64.74 kN
> Force in the lacing flat.
184 DESIGN OF STEEL STRUCTURES

4. Rivet value
Strength of power driven rivet in single shear
⎛π 2 100 ⎞
⎜ 4 × (23.5) × 1000 ⎟ = 43.35 kN
⎝ ⎠
The thickness of flange of channel is 135 mm. It is less than that of flat.
Therefore, strength of power driven shop rivet in bearing
⎛ 23.5 × 13.5 × 300 ⎞
⎜ 1000 ⎟ = 95.175 kN
⎝ ⎠
Rivet value, R = 43.35 kN
5. Load on the rivet form both the sides
(2 × 8.84 × cos 45°) = 12.37 kN < (Rivet value).
Hence, the design is satisfactory.
Provide 2 MC 350, @ 0.421 kN/m with lacing flat 65 F 10 and 22 mm diameter
power driven rivets. The tie plates are provided at the ends of lacing system.
Example 3.13 A building column is made of four ISA 100 mm × 100 mm × 12
mm (ISA 100 100, @ 0.177 kN/m) angles with their backs 350 mm apart as
shown in Fig. 3.17. The lacing of column consists of 60 mm × 10 mm flat bars
arranged in a single laced system and inclined to the axis of the column at an
angle of 45°. The effective length of the column is 8 metres. Using the data given
below and using any rational procedure such as recommended in Indian
Standard Code of Practice IS : 800, find the safe load on the column. Also check
the lacing system for safety against (a) local buckling of column angles, (b)
strength of lacing bare in tension and compression against shear load, (c) strength
of lacing rivets. Assume the shear load on the column as 2.5 per cent of the axial
load. Rivets used have 20 mm nominal diameter (hole diameter = 21.5 mm).
Properties of ISA 100 mm × 100 mm × 12 mm (ISA 100 100, @ 0.177 kN/m);
angle area = 2259 mm2, lx = Iy= 270 × 104 mm4, minimum radius of gyration
= 19.4 mm; distance of centroid from back of angle = 29.2 mm; distance of rivet
hole from back of angle = 60 mm.
Adopt permissible stresse in axial compression as per IS : 800.
Solution
Step 1: Properties of ISA 100 mm × 100 mm × 12 mm (ISA 100 100,@ 0.177
kN/m)
Area = 2259 mm2
Ixx = Iyy = 207 × 104 mm4
Cxx = Cyy = 29.2 m, rmin = 19.4 mm
Step 2: Slenderness ratio. Moment of inertia of the whole column section
about xx-axis
Ixx = [(4 × 207) + 4 × 22.59 (17.5 – 2.92)2]× 104 mm4
= 20038 × 104 mm4
Moment of inertia of the whole column section about yy-axis
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 185

Iyy = 20038 × 104 mm4

Cross-sectional area of the whole column section
A = (4 × 22.59 × 100) = 9036 mm2
Radius of gyration of the whole column section

⎛ 20038 × 104 ⎞
rxx = ryy = ⎜⎜ ⎟⎟ = 148.9 mm
⎝ 9036 ⎠
The effective length of the column, l is 8 m
Maximum slenderness ratio of the column

⎛ 1 ⎞ ⎛ 8 × 1000 ⎞
⎜ ⎟ = ⎜ ⎟ = 53.8
⎝ rmin ⎠ ⎝ 148.9 ⎠

(4 IS A 1 00 m m × 1 00
3 50 m m

@ 0 .17 7 kN /m )
x x
2 9.2 m m

1 75 – 29 .2 m m

y
3 50 m m

Fig. 3.20

Step 3 : Safe load

From IS : 800–1984, allowable stress in axial compression for the steel having
yield stress as 260 N/mm2.
σac = 132.2 N/mm2
Safe load on the column
132.2 × 9036
P = σac × A = = 1194.56 kN
1000
Angles have been laced in single laced system as shown in Fig. 3.21. The size
of lacing flats is 60 mm × 10 mm.
Step 4 : Check for local buckling of column angles
Spacing of lacing = 460 mm
For angles rmin = 19.4 mm
The compression member is composed of angles
For the component of the column
186 DESIGN OF STEEL STRUCTURES

⎛ l ⎞ ⎛ 460 ⎞
⎜ ⎟ = ⎜ ⎟ = 23.7 >/ 40
⎝ rmin ⎠ ⎝ 19.4 ⎠
and |
> (0.6 × 53.8)
= 32.28 for column as a whole
Hence, the local buckling of the angles does not occur as per IS : 800–1984.

2 30
mm
4 5°

4 60 m m
2 30
mm

60 m m 60 m m
2 30 m m

3 50 m m
y
35 0 m m

x x

Fig. 3.21

Step 5 : Check for local buckling of lacing flats

Width of lacing flat is 60 mm
Thickness of lacing flat is 10 mm

⎛ t ⎞ ⎛ 10 ⎞
rmin = ⎜ ⎟=⎜ ⎟ = 2.89 mm
⎝ 12 ⎠ ⎝ 12 ⎠
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 187

Length of lacing flat = 230 2 = 325 mm

⎛ t 325 ⎞
⎜ = ⎟ = 112.7 >/ 145
⎝ rmin 2.89 ⎠
Hence satisfactory.
Strength of lacing flat in compression. Slenderness ratio of the lacing flat
⎛ l ⎞
⎜ ⎟ = 112.7
⎝ rmin ⎠
From IS : 800–1984, the allowable stress in axial compression, for the steel
having yield stress as 260 N/mm2
σac = 70.57 N/mm2
Compressive strength of the lacing flat
70.57 × 60 × 10
P1 = = 42.34 kN
1000
Step 6 : Strength of lacing flat in tension
Rivet diameter = 20 mm
Gross diameter = 21.5 mm
Net area of the lacing flat = (60 – 21.5) × 10 = 38.5 mm2
Tensile strength of the lacing flat
⎛ 0.6 × 260 × 385 ⎞
P2 = ⎜ ⎟ = 60.06 kN
⎝ 1000 ⎠
Strength of rivet in single shear
⎛π 2 100 ⎞
⎜ 4 × (21.5) × 1000 ⎟ = 36.287 kN
⎝ ⎠
Strength of rivet in bearing
⎛ 21.5 × 10 × 300 ⎞
⎜ 1000 ⎟ = 64.5 kN
⎝ ⎠
Rivet value, R = 36.287 kN
Transverse shear resisted in two parallel planes of lacing
⎛ 2.5 ×1194.56 ⎞
⎜ 100 ⎟ = 29.86 kN
⎝ ⎠
Transverse shear resisted in one plane of lacing
1
× 29.86 = 14.932 kN
2
Force in the, rivet due to transverse shear is 29.86 kN
< Rivet value. Hence, the design is safe.
Force in the lacing flat
⎛1 ⎞
F1 = ⎜ × 29.86 × cosec 45°⎟ = 21.117 kN
⎝2 ⎠
It is less than compressive strength and tensile strength of the flat.
188 DESIGN OF STEEL STRUCTURES

Example 3.14. Design a square column consisting of four angles to support

an axial load of 1600 kN. The effective length of column is 10 m.
Solution
Design :
Step 1 : Design of built-up column
1. Selection of trial section
The effective length of built-up column is 10 m (i.e., 1000 mm). The axial load
supported by the column, P is 1600 kN. The square column consists of four
angles as shown in Fig. 3.22. The approximate radii of gyration for the built-
column from Fig. 3.8.
rx = ry= r = 0.42 h
2. Slenderness ratio
Assuming the slenderness ratio for the built-up column as 40
⎛ L ⎞
⎜3 r = 40 ⎟
⎝ ⎠
⎛ 10000 ⎞
∴ 0.42 h = ⎜ ⎟ , h = 595.24 mm
⎝ 40 ⎠
y
C yy
C xx

B u ilt-u p
colum n
con sistin g of
fo ur an gles

Fig. 3.22

Adopt the size of square column as 600 mm × 600 mm. The slenderness ratio
of the built-up column with
rx = ry = rmin = 0.42 × 600 = 252 mm
L ⎛ 10000 ⎞
= ⎜ ⎟ = 39.68
r ⎝ 252 ⎠
The permissible stress in axial compression from IS : 800–1984 for the steel
having yield stress as 260 N/mm2
σac = 145.192 N/mm2
The area required for the built-up column
1600 × 1000
A = = 11019.89 mm2
145.192
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 189

3. Properties of trial section

From ISI Handbook No. 1, try
4 ISA 150 mm × 150 mm × 12 mm (4 ISA 150 150, @ 0.272 kN/m). The cross-
sectional area of four angles is 13836 mm2. The moment of inertia about the
reference axes
Ixx = Iyy = 4(735.4 + 34.59 × 25.2) × 104 mm4
= 95468.41 × 104 mm4
The actual radii of gyration of built-up column
1
⎛ 95648.41 × 104 ⎞2
rxx = ryy = ⎜⎜ ⎟⎟ = 262.67 mm
⎝ 13836 ⎠
> 252 mm (approx. value)
Hence, it is satisfactory. Provide 4 ISA 150 mm × 150 mm × 12 mm.
Step 2 : Design of lacing
The double lacing may be provided at an inclination of 45° with the
longitudinal axis of the built-up column. It may be designed as it has been done
in Example 3.12.

3.16 BATTEN PLATES

The batten plates are also called as the plates, and these are also used in lateral
system. The angle sections, channels and I-sections are also used as battens.
The components of built-up column sharing the load are connected together by
batten plates. The battening of columns shall not be done where the columns
are subjected in the plane of battens to eccentric loading.
In the case the battened compression members are subjected, in the plane of
battens, to eccentric loading, applied moments or lateral forces shall be designed
according to the exact theory of elastic stability or empirically from the
verification tests, so that, these have a load factor of not less than 1.7 in the
actual structure. It is note that if the column section is subjected to eccentricity
or other moments about yy-axis (axis perpendicular to the plane of battens), the
batons arid the column section should be specially designed for such moments.
The design includes design of built (main) column and design of batten plates.
The batten plates are designed as follows.

3.16.1 Design of Batten Plates

Following are the usual steps in design of batten plates. The specifications for
the design of plates as per IS : 800–1984 have also been given. The batten plates
used for connecting the component members of a built-up column have been
shown in Fig. 3.23. The batten shall be placed opposite to each other at each end
of the member and at points where the member is stayed in its length and shall,
as far as practicable be spaced and proportioned uniformly throughout.
The number of battens shall be such that the member is divided into not less
than three bays within its actual length from centre to centre of connection.
190 DESIGN OF STEEL STRUCTURES

Step 1. A battened column is designed with its effective length ten per cent
in excess of the usual standard cases.
Step 2. The compression members composed of two main components
battened should preferably have their two main components of the same cross-
section and symmetrically disposed about their xx-axis. The compression
members composed of two main components battened, should, where practicable,
have a radius of gyration about the axis perpendicular to the plane of batten not
less than radius of gyration at right angles to that axis (i.e., ryy |
< rxx).
Step 3. The spacing of battens is centre to centre distance between adjacent
battens, as shown in Fig. 3.23.

S p a cing l

Fig. 3.23 Battened column

The spacing of battens is kept equal to length of lesser main component of

built-up column. The lesser main component of built-up column is the portion of
column between consecutive connections centre to centre of end fastenings. It is
computed as below :
(a) Compression member is compressed of two I-sections
(i) In the battened compression members not specifically checked for shear
stress and bending moment as specified above, the spacing of battens centre to
centre of end fastenings shall be such that the slenderness ratio ‘λ’ of the lesser
main component over that distance shall be not greater than 50 or greater than
0.7 times the slenderness ratio of the main member as a whole about its xx-axis
(axis parallel to the battens).
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 191

(b) Compression member is composed of two channels (back to back)

When the compression members composed of two angles, channels, or tees
back to back in contact or separated by a small distance shall be connected
together by riveting, bolting or welding so that the ratio of slenderness of each
member between the connections is not greater than 40 or greater than 0.6
times the most unfavourable ratio of slenderness of the strut as a whole
whichever is less.
The minimum radius of gyration of the component member of column is
known. The slenderness ratio of the lesser main component is equated to the
minimum value as the case may be, length of the lesser main component is
computed. The spacing of batten plates is taken equal to this length. The number
of batten plates should be such that it divides the column longitudinally in at
least three parts.
In no case shall the ends of the strut be connected together with less than
two rivets or bolts or their equivalent in welding, and there shall be not less
than two additional connections spaced equidistant in the length of strut. Where
the members are separated back-to-back, the rivets or bolts through these
connections shall pass through solid washers or packings, and where the legs of
the connected angles or tables of the connected these are 125 mm wide or over,
or where the webs of channels are 150 mm wide or over, not less than two rivets
or bolts shall be used in each connection, one on line of each gauge mark.
The battens shall be designed to carry the bending moments and shears arising
from the transverse shear force ‘V’ of 2.5 per cent of the total axial force on the
whole compression member, at any point in the length of the member divided
equally between parallel planes of the battens. The main members shall also be
checked for the same shear force and bending moments as for the battens.

M om e nts
S 1
M1 = M2 = . 2 d
n
S 1 S 1
n 2 n 2
S . 1
n 2 S . 1
n 2 M1 M2
B a tte n
S . 1 S . 1 p la te
n 2 n 2

S 1 S 1
n 2 n 2
L on gitud in al she a r
1 Sd
F1 = × ( M 1 = M 2) =
2 nd
(a ) B a tte ne d colu m n (b ) Free -bo dy d ia gra m

Fig. 3.24
192 DESIGN OF STEEL STRUCTURES

Step 4. The batten plates are designed to resist simultaneously the moments
and longitudinal force arising due to transverse shear force. The expression for
which may be derived as under :
A battened column carrying an axial load may deform as shown in Fig. 3.24.
The transverse shear, S at any point in the length of member is taken as 2.5 per
cent of the total axial force on the whole compression member. It is divided
equally between n parallel planes of the battens. It may be assumed that there
are points of contraflexures, where the curvatures of the column change. Thus,
⎛S 1⎞
a transverse shear force ⎜ ⋅ ⎟ shall be acting as shown in Fig. 3.24 (a). Let S
⎝ n 2⎠
be equal to V and n be equal to N.
Consider the free-body diagram of the portion of battened column as shown
in Fig. 3.24 (b). The bending moments arising at each end of the batten plate
due to transverse shear
⎛S 1⎞ ⎛ S ⋅d ⎞
M1 = M2 = ⎜ ⋅ ⎟ × d = ⎜ ⎟
⎝ n 2⎠ ⎝ 2n ⎠

⎛ V .C ⎞
∴ M1 = M2 = ⎜ ⎟ (C = d)
⎝ 2n ⎠
The longitudinal shear arising due to these moments
1 ⎛ S.d ⎞
V1 = (M1 + M2) = ⎜ ⎟
a ⎝ na ⎠

⎛ V ×C ⎞
∴ V1 = ⎜ ⎟ (a = S)
⎝ N×S ⎠
where C = Distance centre to centre of battens longitudinally
N = Number of parallel planes of batten, and
S = Minimum distance across between C.G. of rivets or welding
Step 5. The tie plates shall be designed by the same method as the other
sections used for the battens. In no case shall a tie plate and its fastenings be
incapable of carrying the forces for which the lacing are designed. When the
plates are used as battens, then, the size of batten plates is computed as below :
1. Effective depth. The effective depth shall be taken as the longitudinal
distance between end rivets or welds.
(a) The effective depth of end battens and those at points where the member
is stayed in its length shall be as below :
It shall not be less than the perpendicular distance between the centroids of
the main members.
(|
< the distance between C.G. of component members of column).
(b) The effective depth of intermediate battens shall be not less than three
quarters of this distance.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 193

3
(|
< 4 ithe distance between C.G. of component members of column.)
But in no case the effective depth of any batten shall be not less than twice
the width of one member in the plane of the battens.
(|
< twice the width of one component member of column.)
The total depth is found by adding edge distance on both the sides.
2. Thickness. Thickness of battens (t) should not be less than 1/50th (one-
fiftieth) the distance between innermost connecting rivets.
(This clause does not apply for angles, channels, I-sections, used as batten
their legs or flanges perpendicular to the main (member) and sufficient to resist
longitudinal shear (F1) and moment (M) arising from transverse shear. However,
it should be ensured that the ends of the compression members are tied to achieve
adequate rigidity.
Step 6 : The rivets are provided to connect batten plates with components of
column to resist shear force (F1) and moment (M) computed above.
The rivets, bolts or welds in these connections shall be sufficient to carry the
shear force and moments, if any, specified for the battened struts. The diameter
of rivets shall not be less than 16 mm for members upto and including 10 mm
thickness ; 20 mm diameter for members upto and including 16 mm thick; and
22 mm diameter for members over 16 mm thick.
The compressive members connected by such riveting, bolting or welding
shall not be subjected to transverse loading in plane perpendicular to the washer
riveted, bolted or welded surfaces.
Example 3.15 Design a built-up column consisting of two channels connected
by battens to carry an axial load of 800 kN. The effective length of the column is
6 metres.
Solution
Design :
Step 1 : Design of built-up column
1. Section of trial section
Effective length of the column is 6 m.
The effective length of battened column is increased by 10 per cent.
Therefore, the effective length of battened column shall be 6.6 mm.
The maximum slenderness ratio for the built-up battened column and the
value of yield stress for the steel may be assumed at 80 and 260 N/mm2,
respectively. Permissible stress in axial compression.
σac = 103 N/mm2
Effective sectional area required
⎛ 800 × 1000 ⎞
A = ⎜ ⎟ = 7766.99 mm
2
⎝ 103 ⎠
From steel section tables (Double channels battened to be used as columns),
try 2 LC 300 @ 0.331 kN/m
194 DESIGN OF STEEL STRUCTURES

Effective sectional area of two channels

A = 8422 mm2
Ixx = 12095.8 × 104 mm4, rxx = 119.8 mm
Iyy should be approximately equal to Ixx and ryy, <
| rxx.
From steel section tables, the spacing between back to back of channels (S) is
adopted as 200 mm.
ryy = 128.7 mm
2 A. Slenderness ratio of the battened column
∴ rmin = 119.8 mm

l ⎛ 6.6 × 1000 ⎞
= ⎜ ⎟ = 55.2
rmin ⎝ 119.8 ⎠
From IS : 800–1984, allowable working stress in compression for the steel
having yield stress as 260 N/mm2
σac = 130.8 N/mm2
3 A. Safe load carrying capacity

⎛ 130.8 × 8422 ⎞
⎜ 1000 ⎟ = 1101.60 kN > 800 kN
⎝ ⎠
It is not satisfactory.
Try 2 MC 230, @ 0.304 kN/m
Effective sectional area of two channels,
A = 7734 mm2
Ixx = 7633.6 × 104 mm4
rxx = 99.4 mm, cyy = 23.0 mm
Adopt the spacing between back to back of channels as 160 mm
Iyy = 8643.2 × 104 mm4
ryy = 105.7 mm |< rxx
∴ rmin = 99.4 mm
2 B. Slenderness ratio of battened column

l ⎛ 6.6 × 1000 ⎞
= ⎜ ⎟ = 66.4
rmin ⎝ 99.4 ⎠
From IS : 800–1984, allowable working stress in compression for the steel
having yield stress as 260 N/mm2
σac = 107.96 N/mm2
3 B. Safe load carrying capacity

⎛ 107.96 × 7734 ⎞
⎜ 1000 ⎟ = 834.96 kN
⎝ ⎠
Hence, the design is satisfactory.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 195

Step 2 : Design of battens

1. Spacing of batten plates
The battened column is composed of two channels back to back separated by
small distance.
The maximum slenderness ratios of column about xx-axis is as follows :

⎛ l ⎞
⎜ ⎟ = 66.4
⎝ rmin ⎠
Minimum radius of gyration of one channel
rmin = 23.8 mm
The column is composed of two channels.
Slenderness ratio of component member of column (slenderness ratio of each
member between the connections)

⎛ l ⎞
⎜ ⎟ < 40
⎝ rmin ⎠

⎛ l ⎞ l
⎜ ⎟ = (0.6 r of column as a whole 0.6
⎝ rmin ⎠ min
× 66.4 = 39.84)

⎛ l ⎞
∴ ⎜ ⎟ = 39.84
⎝ 23.8 ⎠
∴ l = 39.84 × 23.8 = 948.192 mm
Adopt the spacing of battens as 900 mm
Transverse shear due to axial load
⎛ 2.5 ⎞
V = ⎜ × 800 ⎟ = 20 kN
⎝ 100 ⎠
Longitudinal shear arising from transverse shear
VC ⎛ 20 × 900 ⎞
V = = = 36 kN
NS ⎝⎜ 2 × 205 ⎠⎟
Moment arising from transverse shear
VC ⎛ 20.0 × 900 ⎞
M = = = 4500 mm-kN
2N ⎜⎝ 2 × 2 ⎟⎠
2. Size of batten plates
2 (A) Effective depth of end batten
= the distance between C.G. of component members
= (160 + 2 × 23.0) = 206 mm
|
< Twice the width of one component member of the column
196 DESIGN OF STEEL STRUCTURES

Assume 22 mm nominal diameter of rivets connection

S pa cin g l

45 m m 45 m m

23 1 60 23
mm

Fig. 3.22 Battened column

Overall depth of end batten

= 206 + Edge distances on both sides
= 206 + 2 × (35) = 276 mm
Effective depth of intermediate batten
3
= distance between C.G. of component
4
member
⎛3 ⎞
= ⎜ × 206 ⎟ = 154.5 mm
⎝ 4 ⎠
< twice the width of one component member of the column (i.e., 2 × 80 = 160
mm).
2 (B) Effective depth of the intermediate battens
= (2 × 80) = 160 mm
Overall depth of intermediate battens
[1600 + 2 × (35)] = 230 mm
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 197

Thickness of batten plates

1
t× th × distance between innermost connecting line of rivets
50

⎛ 1 ⎞
t = ⎜ × 250 ⎟ = 5 mm
⎝ 50 ⎠
Adopt 6 mm thickness of batten plates as a provision against corrosion. If the
depth of intermediate batten plate required for resisting moment is d mm, then
⎛1 ⎞
Moment of resistances = ⎜ t × d2 × σbc ⎟ = M
⎝ 6 ⎠

1 ⎛ 6 × d 2 × 0.66 × 260 ⎞
×⎜ ⎟⎟ = 45000, ∴ d = 161.94 mm
6 ⎜⎝ 1000 ⎠
Adopt the size of end batten plates as 300 mm × 6 mm and the size of
intermediate plates as 230 mm × 6 mm. The actual depth of batten plate is more
than that needed.
3. Check for longitudinal shear
Shear strength of intermediate plate
⎛ 0.4 × 260 ⎞
230 × 6 × ⎜ ⎟ = 143.52 kN
⎝ 1000 ⎠
> longitudinal shear in plate 36 kN. Hence the design is satisfactory.
4. Design of connection
Shear strength of power driven rivets
⎛π 2 100 ⎞
⎜ 4 (23.5) × 1000 ⎟ = 43.35 kN
⎝ ⎠
Bearing strength of power driven rivets
⎛ 300 ⎞
⎜ 6 × 23.5 × 1000 ⎟ = 42.30 kN
⎝ ⎠
Rivet value, R = 42.30 kN
Adopt 3 rivets in a vertical row as shown in Fig. 3.25
Shear force in each rivet
⎛ 36 ⎞
⎜ ⎟ = 12 kN
⎝ 3 ⎠
⎛ My ⎞ ⎛ 4500 × 80 ⎞
Force due to moment = ⎜ 2⎟ = ⎜ ⎟ = 28.125 kN
⎝ Σy ⎠ ⎝ 2 × 80 × 80 ⎠
Resultant force in rivets
(122 + 28.1252)1/2 = 30.578 kN < Rivet value.
Hence the design is satisfactory.
Provide 2 MC 250, @ 0.304 kN/m with batten plates as shown in Fig. 3.25.
198 DESIGN OF STEEL STRUCTURES

3.17 PERFORATED COVER PLATES

The perforated cover plates may be used instead of lacing, as a lateral system
with the columns. The perforated cover plates are particularly suitable for a
built-up box section consisting of four angle sections. The interior of the column
remains accessible for painting. Bureau of Indian Standards has not given
specifications for the design of perforated plates. Guide to design criteria for
metal compression members published by the Column Research Council, U.S.A.
has given the following specifications for the design of perforated cover plates.
(1) The perforation of plate should be of ovaloid shape (i.e., two semi-circles
connected with straight sides) or they may be elliptical or circular. The long
axis of the perforations for the first two cases should be in the direction of the
axis.
(2) The spacing of the perforations (i.e., the centre to centre distance between
the perforations) should not be less than 1.5 times the length of perforation, C.
(3) The clear distance between perforations should be not less than the
distance between the nearest lines of longitudinal fasteners [i.e., (a – c) ≥ d].
(4) The net section of the column (defined as the section at the perforations)
should be used in computing the axial rigidity, EA, and the column moments of
inertia about x-axis and y-axis.
The net area of perforated plate is used as the part of cross-sectional area of
the column. It is the advantage of the perforated plates as compared with the
lacing.
c
(5) If the slenderness ratio (length of perforation divided by radius of
rf
gyration of flange, see Fig. 3.26 is 20 or less, and also not greater than one-third
l
of the slenderness ratio , the appropriate specification for column stress,
rx
applied to the column net area can be used to find the allowable load.
(6) For columns built-up of plates, the net area of each web at the perforation
1
should be sufficient to resist times the transverse shear force, where n is the
n
number of perforated plates. Perforated plates designed in accordance with
provision (3) need not be checked for shear introduced as a specified percentage
of the column load.
(7) The transverse distance from the edge of a perforation to the nearest
line of longitudinal fastener, divided by the plafe thickness, i.e., the b/t ratio of
plate adjacent to a perforation should, conform to minimum specification
requirements for plates in main compression members.
A member with staggered perforations should be treated as if the perforations
were opposite to each other. The perforated cover plates should be detailed in
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 199

such a way that the axial compressive stresses do not cause the failure by local
buckling before the strength of column is reached.

b1 b1

( a – c) b2 a1

(a )
f
x

y y

x
(b )

Fig. 3.26 Perforated cover plates

The perforated cover plates are designed by the criterion obtained by equating
the plate buckling stress (fcr)plane and the column strength (fcr)column. The plate
buckling stress is given by
2
⎛t ⎞
(fcr)plate = kE . ⎜ ⎟ ...(3.35)
⎝b⎠
where, E = Effective modulus of elasticity corresponding to the critical stress
t
= Thickness to width ratio of plate
b
k = Coefficient.
The coefficient k depends upon the type of loading, the conditions of edge
support and the proportion of plate.
The length of column depends upon the slenderness ratio, yield stress, fy, and
effective modulus of elasticity, E. Therefore, the proportion of perforated plate
200 DESIGN OF STEEL STRUCTURES

t
may be expressed as a function of . E and fy. The function is of complex form.
r
The following expressions give the design criteria for all types of steel. These
functions approximate to the complex function with sufficient accuracy.
l
For < 60
r
b
= 19.2 ...(3.36)
(t ⋅ k )
l
For > 60
r
b ⎛l⎞
= 0.32 ⎜ ⎟ ...(3.37)
(t ⋅ k ) ⎝r⎠

⎛b⎞
When the values of k are known, then, the width thickness ratio ⎜ ⎟ may be
⎝t⎠
found. Two portions of the perforated plates are of interest. The portion b1 is
free along one edge and restrained along the other edge. The other edge of
portion b1 is riveted or welded to the segment of the column. The restraint
action along the edge varies between the two cases. In one case, it is equivalent
to a longitudinal hinge, which allows the plate to rotate about the axis of the
edge. In the other case, it is equivalent to that of continuous clamp, which does
not allow rotation. The value of k for above two cases are 0.38 and 0.15,
respectively. In practice, the type of restraint gives the lower value out of two
values.
The portion of plate with width b2 is restraint along both the edges. The
restraint of edges is also equivalent to two above mentioned cases. The values
of k for which vary from 3.6 to 6.3. In practice, the type of restraint gives the
lower value.

3.18 ECCENTRICALLY LOADED COLUMNS

When a load is applied at an eccentric distance from the centroid of the column
section, as shown in Fig. 3.27, the column is called eccentrically loaded.
A column may carry an eccentric load in addition to axial load. The
eccentrically loaded column is subjected to bending stress besides the axial
compression. The strength of the column is very much reduced.
When the column is subjected to axial loading only, then, the cross-sectional
area, Aa required is given by
⎛ P ⎞
Aa = ⎜ ⎟ ...(i)
⎝ σac ⎠
where, σac = Allowable working stress in compression on the member
subjected to axial load only.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 201

When the column is subjected to bending only then cross-sectional area

required is given by
⎛ P ⋅e ⋅ y⎞
Ab = ⎜
⎝ σbc ⋅ r 2 ⎟⎠

Fig. 3.27 Eccentrically loaded column

where, σbc = Maximum allowable bending compressive stress on the extreme

fibre
y = Distance to the extreme fibre of the section measured from the
neutral axis
Ab . r2 = Moment of inertia of the section about the appropriate axis
r = Radius of gyration of the section about the appropriate axis
P.e = Moment due to the eccentricity, e of the load P, and

⎛ P ⋅ e σbc⋅cal ⎞ ⎛ P ⋅ e ⋅ y P ⋅ e ⋅ y⎞
⎜⎝3 I = y ⎟⎠ , ∴ ⎜⎝ σbc⋅cal = I
= ⎟
Ab ⋅ r 2 ⎠
When the column is subjected to both axial load and bending, then, the total
area required
⎛ P P ⋅e⋅ y ⎞
A = (Aa + Ab) = ⎜⎜ + ⎟
2 ⎟ ...(iii)
⎝ σac σbc ⋅ y ⎠
Dividing both the sides of expression (iii) by A and writing the expression as
below
⎡⎛ P ⎞ ⎛ P ⋅ e ⋅ y ⎞ ⎤
⎢⎜ 2 ⎟⎥
= 1
⎟ ⎜
⎢⎜ A ⎟ + ⎜ Ar ⎟ ⎥
⎢⎣⎝ σ ac ⎠ ⎝ σbc ⎠ ⎥⎦
202 DESIGN OF STEEL STRUCTURES

⎛ σ ac.cal σ ac.cal ⎞
⎜⎝ σ + = 1 ... (3.38)
ac σbc ⎟⎠

⎛P⎞
where, σac.cal = ⎜ ⎟ (calculated average axial compressive stress)
⎝A⎠

⎛ P ⋅e⋅ y⎞
σbc.cal = ⎜ ⎟ (calculated bending stress in the extreme fibre)
⎝ Ar 2 ⎠
Equation 3.38 gives the fundamental interaction formula for the compression
member subjected to the axial load and bending. It is a straight line when plotted

⎛σ ⎞ ⎛ σ ac.cal ⎞
graphically, ⎜ ac.cal ⎟ along y-axis and ⎜⎝ σ ⎟ along x-axis. Bureau of Indian
⎝ σ ac ⎠ bc ⎠

Standards adopted this interaction formula. The column subjected to both

bending and axial compression shall be so proportioned that the quantity

⎡⎛ σ ac.cal ⎞ ⎛ σ ac.cal ⎞ ⎤
⎢⎜ ⎟ +⎜ ⎟ ⎥ should not exceed only.
⎣⎝ σ ac ⎠ ⎝ σbc ⎠ ⎦
When the bending occurs about both the axes of the member, then the
interaction formula may be written as below :

⎛ σ ac.cal σ acx .cal σ acy.cal ⎞

⎜ σ + + ≤ 1.0 ...(iv)
⎝ ac σbcx σbcy ⎟⎠

The value of σcbx and σcby to be used in the above formulae shall be lesser of
the values of the maximum permissible stresses σbc for bending about the
appropriate axis.
At a support and using the values σbcx and σbcy at the support.

⎛ σ ac.cal σ acx .cal σ acy.cal ⎞

⎜ 0.60 f + σ + ≤ 1.0
⎝ y bcx σbcy ⎟⎠

IS and BS codes give the equivalent stress formula for combined bending,
bearing and shear stress as under. It is based on von Mises–Hencky’s theory
σe.cal = (σ2bt.cal + 3τ2vm.cal)1/2
or σe.cal = (σ2bc.cal + 3τvm.cal)1/2
Where a bearing stress is combined with tensile stress or compressive stress
bending and shear stresses under the most unfavourable condition of loading
the equivalent stress σe.cal obtained from the following formulae, shall not exceed
σe = 0.9fy
σe.cal = [σ2bt.cal + σbt.cal σp.cal + 3τvm.cal]1/2
or σe.cal = [σ2bc.cal + σ2p.cal σbe.cal σp.cal + 3τvm.cal]1/2
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 203

where σac.cal = calculated average axial compressive stress

σbc.cal = calculated bending compressive stress in the extreme fibre
σbt.cal = calculated bending tensile stress in the extreme fibre
τvm.cal = calculated shear stress
σbc.cal ; σbc.cal ; τvm.cal and σp.cal are the numerical values of the co-existing
bending (compressive or tensile stress) shear and bearing stresses. When the
bending occurs about both the axes of the member, σbt.cal and σbc.cal shall be
taken as the sum of the two calculated fibre stresses. σe is the maximum
permissible equivalent stress. (σe = 0.9fy).
The compression flange of the column subjected to bending is laterally
unrestrained. As such the allowable bending compressive stress is reduced. The
column section is adopted depending upon the effective length of column.
Allowable bending compressive stress for compression flange of a section
unrestrained, has been discussed in Chapter 6, in design of laterally unrestrained
beams.
For design of a column, the eccentricity of loading shall be taken as the distance
from the assumed point of application of load to the centroid of the column. For
the purpose of determining the stress in a stanchion or column section, the
beam reactions or similar loads shall be assumed to be applied 100 mm from the
face of the section or at the centre of the bearing whichever dimension gives the
greater eccentricity except in the following two cases.
(i) In the case of cap connections, the load shall be assumed to be applied
at the face of the column shaft or stanchion section ; or edge of packing if used,
towards the span of the beam ; and
(ii) In the case of a roof truss beating on a cap, no eccentricity need be
taken for single bearing without connections capable of developing an appreciable
moment.
In continuous columns, the bending moment due to eccentricities of loading
on the columns at any floor may be taken as
(i) ineffective at the floor levels above and below that floor and
(ii) divided equally between the column’s length above and below that floor
level, provided that the moment of inertia of either column section, divided by
its effective length does not exceed 1.5 times the corresponding value of the
other column. In case where this ratio is exceeded, the bending moment shall be
divided in proportion to the moment of inertia of the column sections divided by
their respective effective length.
The load was assumed to be applied as given in Table 3.5 as per IS : 800–
1962. The connections are designed as simple connections to transmit shear
only.
Table 3.5 Assumed eccentricity of loads in columns

Type of connections Assumed point of application of load

(i) Stiffened bracket Mid point of stiffened application of load
(ii) Unstiffened bracket Outer face of vertical leg of bracket
(iii) Clats to the web Face of strut
204 DESIGN OF STEEL STRUCTURES

The type of connections have been shown in Figs. 3.25 (a), (b) and (c)
respectively.

C le a t a ng le

w B e am
w

Se at an g le
Stiffen er
a ng le

Filler pla te

(a) Stiffened bracket

C le a t a ng le

B e am
C o lu m n

S e at an gle

(b) Unstiffened bracket

B e am w e b clea t

S ta n ch io n

(c) C lea ts to w e d o f b e am typ es o f b ea m an d co lu m n co nn ection s

Fig. 3.28
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 205

3.19 DESIGN OF ECCENTRICALLY LOADED COLUMNS

Following are the usual steps required for the design of eccentrically loaded
columns :
Step 1. Assume the allowable working stress in compression for the column
and compute the effective sectional area required for the load acting axially or
eccentrically or both.
Step 2. Adopt the effective sectional area of eccentrically loaded column about
one and half to two times the area computed above.
Step 3. The slenderness ratio for the column is computed from the effective
length of the column and minimum radius of gyration. Allowable working stress
in compression corresponding to the slenderness ratio is found from IS : 800–
1984. The average axial compressive stress is calculated by dividing total vertical
load by the sectional area of the column.
Step 4. The actual bending stress is calculated by computing the moment
and dividing it by the section modulus of the section. The allowable bending
compressive stress on the extreme fibre is noted from IS : 800–1984 or calculated
as discussed in Chapter 6, in design of laterally unrestrained beams. The design
of the column is checked by computing the quantity.

⎡⎛ σ ac⋅cal ⎞ ⎤ ⎛ σbc⋅cal ⎞
⎢⎜ ⎟⎥ + ⎜ ⎟ which should not exceed unity, otherwise design is
⎣⎝ σ ac ⎠ ⎦ ⎝ σbc ⎠
revised.
Example 3.16 A LB 500, @ 0.750 kN/m simply supported beam carries a
uniformly distributed load of 280 kN (inclusive of self-weight) over a span of 4
metres. The beam is attached at each end to a stanchion of similar section by a
ISA 150 mm× 75 mm × 12 mm (ISA 15075 @ 0.201 kN/m) angle iron riveted to
one of the flange of the stanchion. In addition to the above, the stanchion is
carrying an axial load of 200 kN. If the effective length of the stanchion is 4 m,
find whether the stanchion is strong enough to carry the loads. Adopt stress as
per IS : 800–1984.
Solution
Step 1 : Properties of section
Axial load of the stanchion = 200 kN
Load from simply supported beam on the stanchion is 140 kN
Total vertical load supported by the stanchion is 340 kN
From steel section tables for LB 500, @ 0.750 kN/m
Sectional area A = 9550 mm2
rxx = 201. 0 mm
ryy = 33. 4 mm
Zxx = 1543–2 × 103 mm3
Step 2 : Slenderness ratio
Effective length of the stanchion is 4 m
206 DESIGN OF STEEL STRUCTURES

Minimum radius of gyration, rmin = 33.4 mm

Slenderness ratio
l ⎛ 4 × 1000 ⎞
= ⎜ ⎟ = 119.8
rmin ⎝ 33 ⋅ 4 ⎠
Step 3: Direct compressive stresses
From IS : 800–1984 allowable working stress in compression for the steel
having yield stress as 260 N/mm2
σac = 64.18 N/mm2
Average axial compressive stress
⎛ 340 × 1000 ⎞
σac.cal = ⎜ ⎟ = 35.568 N/mm2
⎝ 9550 ⎠
Step 4: Bending stresses
Assuming the beam connection with the flange of the stanchion as stiffened
bracket connection, the point of application of load from the beam is at mid-
point of stiffened seating.
Width of stiffened seating = 75 mm
Eccentricity of the load from the axis of column
⎛ 75 ⎞
⎜ 250 + ⎟ = 287.5 mm
⎝ 2 ⎠
Moment due to eccentric load
M = (287.5 × 140) = 40250 mm-kN
Actual bending stress

⎛ M ⎞ ⎛ 40250 × 1000 ⎞
σbc.cal = ⎜ ⎟ = ⎝⎜ 1543 ⋅ 2 × 103 ⎠⎟ = 26.082 N/mm
2
⎝ Z xx ⎠
From IS : 800–1984 allowable bending compressive stress
σbc = 0.66 fy = (0.66 × 260) = 171.6 N/mm2
Step 5: Check for interaction expression
For the stanchion subject to the axial compressive stress and bending stress
⎛ σ ac⋅cal σbc⋅cal ⎞
⎜⎝ σ + < 1
ac σbc ⎟⎠
Substituting these values,
⎛ 35 ⋅ 068 26 ⋅ 082 ⎞
⎜ + ⎟ = 0 706 < 1.
⎝ 64 ⋅ 18 171 ⋅ 6 ⎠
Hence, the stanchion is safe
Example 3.17 A column effectively held in position but not in direction at
their end is 4 metres long and carries an axial load of 700 kN and an end moment
of 35000 mm-kN. Design the column if only rolled steel beam sections are
available. Adopt stress as per IS : 800–1984.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 207

Solution
Design :
Step 1 : Selection of trial section
The slenderness ratio for the column and the value of yield stress for steel
may be assumed as 80 and 260 N/mm2, respectively. Permissible stress in axial
compression.
σbc = 103 N/mm2
Effective sectional area required for axial load
⎛ 700 × 1000 ⎞
A = ⎜ ⎟ = 6796.116 mm2
⎝ 103 ⎠
Assume effective sectional area required for moment in addition to the axial
load
1.5 × 6796.116 = 10194.174 mm2
Step 2 (A) : Properties of trial section
From steel section tables, try MB 500, @ 0.869 kN/m
Effective sectional area
A = 11074 mm2
rxx = 202.1 mm
ryy = 35.2 mm
Zxx = 18087 × l03 mm3
Step 3 (A) : Slenderness ratio
Effective length of columns is 4 m
rmin = 35.2 mm
Maximum slenderness ratio
⎛ l ⎞ ⎛ 4 × 1000 ⎞
⎜ r ⎟ = ⎜⎝ 35 ⋅ 2 ⎟⎠ = 113.5
⎝ min ⎠
Step 4 (A) : Check for interaction expression
Average axial stress in compression
⎛ 700 × 1000 ⎞
σac.cal = ⎜ ⎟ = 63.21 N/mm
⎝ 11074 ⎠
Allowable working stress in compression for
⎛ l ⎞
⎜ r ⎟ = 113.5
⎝ min ⎠
and the steel having yield stress as 260 N/mm2
σac = 69.85 N/mm2
End moment = 35000 kN-mm
Bending compressive stress
⎛ 3500 × 1000 ⎞
σbc.cal = ⎜ ⎟ = 19.351 N/mm
2
⎝ 1808 ⋅ 7 × 103 ⎠
208 DESIGN OF STEEL STRUCTURES

From IS : 800–1984 allowable bending compressive stress is 0.66 fy

σbc = 171.6 N/mm2
⎛ σ ac⋅cal σbc⋅cal ⎞ 63 ⋅ 21 19 ⋅ 351 ⎞
⎜⎝ σ + ⎟ = ⎛⎜ + ⎟
ac σbc ⎠ ⎝ 69 ⋅ 85 171 ⋅ 6 ⎠
= 1.0177 > 1 not satisfactory.
Hence redesign is necessary.
Step 2 (B) : Properties of second trial section
Try WB 500, @ 0.952 kN/m
A = 12122 mm2
rxx = 207.7 mm
ryy = 49.6 mm
Zxx = 2091 .6 × 103 mm3,
rmin = 49.6 mm
Average axial stress in compression
⎛ 700 × 1000 ⎞
σac.cal = ⎜ ⎟ = 57.746 N/mm2
⎝ 12122 ⎠
Step 3 (B) : Slenderness ratio

⎛ l ⎞ ⎛ 4 ⋅ 0 × 1000 ⎞
⎜ r ⎟ = ⎜⎝ 49 ⋅ 6 ⎟⎠ = 80.645
⎝ min ⎠
Step 4 (B) : Check for interaction expression
From IS : 800–1984 allowable working stress in compression
σac = 102.29 N/mm2
⎛ l ⎞
For ⎜ r ⎟ = 80.645
⎝ min ⎠
From IS : 800–1984 allowable bending compressive stress for effective length
of 4.0 m
σbc = (0.66 × 260) = 171.6 N/mm2
⎛ 3500 × 1000 ⎞
σbc.cal = ⎜ ⎟ = 16.734 N/mm
2
⎝ 2091 ⋅ 6 × 103 ⎠
Hence, the design is satisfactory. Provide WB 500, @ 0.952 kN/m.
Alternatively
Step 2 : Properties of trial section
From IS: 808–1984, try Medius weight beam section MB 550, @ 104 kg/m
cross-sectional area
A = 13200 mm2
rxx = 222 mm
ryy = 37.3 mm
Zxx = 2360 × 103 mm3
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 209

Step 3 : Slenderness ratio

Effective length of column is 4000 mm
Minimum radius of gyration
rmin = 37.3 mm
Maximum slenderness ratio
⎛ l ⎞ ⎛ 4000 ⎞
⎜ r ⎟ = ⎜⎝ 37 ⋅ 3 ⎟⎠ = 107.24
⎝ min ⎠
Step 4 : Check for interaction expression
Average axial stress in compression
⎛ 700 × 1000 ⎞
σac.cal = ⎜ ⎟ = 53.03 N/mm2
⎝ 13200 ⎠
Allowable working stress in axial compression for slenderness ratio, 107.24
and the steel having yielding stress as 260 N/mm2
⎛ 9 ⎞
σac = ⎜ 82 – × 7 ⋅ 24 ⎟ = 75.484 N/mm2
⎝ 10 ⎠
End-moment = 35000 kN-mm
Bending compressive stress
⎛ 3500 × 1000 ⎞
σbc.cal = ⎜ ⎟ = 14.831 N/mm2
⎝ 2360 × 1000 ⎠
From IS: 800–1984 allowable bending compressive stress is 0.66fy
σbc = (0.66 × 260) = 171.6 N/mm2

⎛ σ ac⋅cal σbc⋅cal ⎞ ⎛ 53 ⋅ 03 14 ⋅ 831 ⎞

⎜⎝ σ + ⎟ = ⎜ + ⎟
ac σbc ⎠ ⎝ 75 ⋅ 484 171 ⋅ 6 ⎠
= (0.7025 + 0.086)
= 0.78896 < 1.00
Hence, the design is satisfactory. Provide MB 550, @ 104 kg/m (Medium weight
beam section).
Example 3.18 A column is subjected to an axial load of 500 kN. The beam
connected to the flange of a column, has an eccentric load of 100 kN and the
beam connected to the web of a column has an eccentric load 50 kN. If the effective
length of the column is 4 metres and only rolled steel H-sections are available,
design the column. Adopt stresses as per IS : 800–1984.
Solution
Design :
Step 1 : Selection of trial section
Total vertical load acting on the column
(500 + 50 + 100) = 650 kN
The slenderness ratio of the column and the value of yield stress for the steel
may be assumed as 80 and 260 N/mm2, respectively.
210 DESIGN OF STEEL STRUCTURES

Therefore, permissible stress in axial compression

σac = 103 N/mm2
Effective sectional area required
⎛ 650 × 1000 ⎞
A = ⎜ ⎟ = 6310.68 mm2
⎝ 103 ⎠
Assume effective sectional area required
1.5 × 6310.682 = 9466.023 mm2
From steel section tables, try HB 400, @ 0.822 kN/m

2 50 m m

IS H B 4 00
@ 0 .82 2 kN /m
A xia l
4 00 m m

lo ad 5 00 kN

E ccen tric
lo ad 5 0 kN

E ccen tric
lo ad 1 00 kN

Fig. 3.29

Step 2 : Properties of trial sectional

Effective sectional area and other properties are as follows :
A = 10466 mm2 lyy = 2783.0 × 104 mm4
rxx = 166.1 mm ryy = 51.6 mm
Zxx = 1444.2 × 103 mm3 Zyy = 221. 3 × 103 mm3
tw = 10.6 mm tf = 12.7 mm
Average compressive stress
⎛ 650 × 1000 ⎞
σac.cal = ⎜ ⎟ = 62.11 N/mm2
⎝ 10466 ⎠
Step 3 : Slenderness ratio
Effective length of column is 4 m and rmin = 51.6 mm
Slenderness ratio of the column

⎛ l ⎞ ⎛ 4 × 1000 ⎞
⎜ r ⎟ = ⎜⎝ 51 ⋅ 6 ⎟⎠ = 77.5
⎝ min ⎠
Step 4 : Check for interaction expression
From IS : 800–1984 allowable working stress in compression for the steel
having yield stress as 260 N/mm2
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 211

σac = 106 N/mm2

Eccentric load about xx-axis is 100 kN
This load can be assumed to be acting at an eccentric distance of half the
depth of section plus distance upto mid-point of stiffened seating. (Size of stiffened
angle may be assumed as ISA 150 mm × 75 mm × 10 mm (ISA 15075, @ 0.169
kN/m) with long leg vertical.)
Moment in plane of yy-axis
= (200 + 37.5) ×100 = 23750 mm-kN
Eccentric load about yy-axis is 50 kN
Minimum eccentricity may be assumed as 100 mm
Moment in plane of xx-axis
= 50 × 100 = 5000 mm-kN
Actual bending stress in plane of yy-axis (i.e., bending is about xx-axis)

⎛ 23750 × 1000 ⎞
σbc1 ⋅cal = ⎜ =16.445 N/mm2
⎝ 1444 ⋅ 2 × 103 ⎟⎠
Actual bending stress in plane of xx-axis (i.e., bending is about yy-axis)

⎛ 5500 × 1050 ⎞
σbc2 ⋅cal = ⎜ ⎟ = 22.594 N/mm2
⎝ 221 ⋅ 3 × 103 ⎠
∴ Allowable bending compressive stress is
σbc = 0.66 fy= (0.66 × 260) = 171.6 N/mm2
There is no possibility of lateral bending in case the bending of column is
about xy-axis as shown in Fig. 3.29.

⎛ σ ac⋅cal σbc1⋅cal + σbc 2⋅cal ⎞ ⎛ 62 ⋅ 11 16 ⋅ 445 + 27 ⋅ 514 ⎞

⎜⎝ σ +
σ ⎟⎠ = ⎜⎝ 106 + 171 ⋅ 6
⎟
⎠
ac bc

= 0.9813 < 1.
Hence design is satisfactory. Provide HB 400. @ 0 .822 kN/m as shown in
Fig. 3.29.

3.20 MODIFIED AND ADJUSTED INTERACTION FORMULAE

The interaction formula discussed in Sec. 3.18, Eq. 3.36 is the fundamental
interaction formula for the column subjected to axial load and bending. The
formula takes into account for the additional stress caused by column buckling
under an axial load by reducing the value of allowable axial stress, aac as the
⎛l⎞
maximum slenderness ratio X = ⎜ ⎟ increases. The formula does not take into
⎝r⎠
account for the additional stresses for the lateral deflection caused by induced
end moment.
212 DESIGN OF STEEL STRUCTURES

Mo P
Mo

Mo Mo
P
(a ) (b )

Fig. 3.30 Column carrying equal end moments

Consider the sequential column loading as shown in Fig. 3.30. A column is

subjected to equal end moments as shown in Fig. 3.30 (a). The dotted line shows
the deflected shape of the column on exaggerated scale. The maximum deflection
and moment at the centre of column are ∆m and M0 respectively. An axial load
P is applied to more deflected column. The column is subjected to additional
moment at the centre equal to P. ∆m. The column is subjected to more deflection,
due to this, which causes more moment and so on. Final bending stress at the
centre of column is given by

⎛ M ⋅ y ∆mP ⋅ y ⎞
σbc.cal = ⎜ 0 + ⎟ ...(i)
⎝ I I ⎠
It is seen that when the column is subjected to equal end moments at the
ends, the lateral deflection depends on the slenderness ratio of the column in
the direction in which the bending occurs. A large lateral deflection occurs when
the slenderness ratio is large. The bending stresses from the deflection increase
with the increasing value of the axial load. AISC specification recommends a
method, which simplifies the design procedure. The method is based upon the
application of interaction formula. The interaction formula is modified as
necessary. The strength of column subjected to bending combined with
compressive axial load may be expressed conveniently by interaction formula
P M
in terms of the ratios and 0 .
Py My

⎡P M0 ⎤
⎢P + (1 – P ) ⎥ = 1.0 ...(ii)
⎢ y My ⎥
⎣⎢ Pe ⎦⎥
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 213

where Py = Axial load causing yielding if it alone occurred

My = Bending moment causing yielding, if it occurred in the absence
of axial load
Pe = Elastic buckling load of the column for buckling in the plane of
applied moment.
The various terms in the expression (ii) may be interpreted as follows :
P = σac.cal A Py = σac A

⎛ σac.cal I ⎞ ⎛ σbc. I ⎞
M0 = ⎜⎝ My = ⎜ y ⎟
y ⎟⎠ ⎝ ⎠

⎛ π2 EI ⎞
Fć A =
⎜ 2 ⎟
⎝ L ⎠
The expression (ii) reduces to

⎡ I⎤
⎢σ σ ac⋅cal ⎥
A 1 y
⎢ ac⋅cal + ⋅ ⎥ = 1.0 ... (iii)
σ
⎢ ac A ⎛ σ ac⋅cal A ⎞ I ⎥
1– σbc
⎢ ⎜ Fc´ A ⎟⎠ y ⎥
⎣⎢ ⎝ ⎦⎥

⎡ σ ac⋅cal 1 σ ⎤
or ⎢ σ + ⋅ ac⋅cal ⎥ = 1.0 ... (3.39)
⎛ σ ⎞ σ
⎢ ac 1 – ac⋅´cal bc ⎥
⎢ ⎜ Fc ⎟⎠ ⎥
⎣ ⎝ ⎦

1
where = Amplification factor. It is also known as magnification
⎛ σ ac⋅cal ⎞
⎜⎝1 – F ⎟⎠
e

factor ...(3.40)
The stress Fc′ is defined as the limiting Euler stress (divided by a factor of
safety). The stress Fe′ may be found as follows. From Eq. 3.7.

⎡ π2 E ⎤ ⎛ 9 ⋅ 87 × 2 ⋅ 047 × 105 ⎞
Fc´ = ⎢ 2 ⎥ = ⎜ 2 ⎟ N/mm2
⎢⎛ l ⎞ ( ) ⎥ ⎜ ⎛ l ⎞ ⎟
⎢⎣ ⎜⎝ r ⎟⎠ ⋅ F ⋅ S ⎥⎦ ⎜ ⎜ ⎟ × 1 ⋅ 92 ⎟
⎝ ⎝ r ⎠ ⎠

⎛ 10,52,000 ⎞
Fe´ =
⎟ N/mm
2 ...(iv)
⎜ l
2
⎛
⎜ ⎜ ⎟ ⎞ ⎟
⎝ ⎝ r⎠ ⎠
214 DESIGN OF STEEL STRUCTURES

It is to note that the slenderness ratio is measured with respect to the axis
about which bending takes place. When the magnitude of stress Fe is very large
and/or the magnitude of axial stress fc is very small, the amplification factor
may be neglected. Equation 3.39 gives the modified interaction formula for the
equal column end moments causing a single curvature deflection.
The amplification factor given by Eq. 3.40 is based upon equal column end
moments. The modified interaction formula is further modified depending upon
any other combination of end conditions. It introduces a reduction factor, Cm.
When the interaction formula is further modified, then, it is known as adjusted
interaction formula. AISC 1963 specification recommends the adjusted
interaction formula as below :

⎡ σ ac⋅cal Cm σ ⎤
⎢ σ + ⋅ bc⋅cal ⎥ = 1.0 ...(3.41a)
⎢ ac (1 – σ ac⋅cal ) σ bc ⎥
⎢⎣ Fc´ ⎥⎦

When the load is eccentric with respect to x-axis, and also with respect to
y-axis, then, Eq. 3.41 (a) may be written as

⎡ σ ac⋅cal Cmx σbcx ⋅cal Cmy ⋅ σmy⋅cal ⎤

⎢ + + ⎥ < 1.0 ...(3.41b)
⎢ σ ac ⎧ σ ac⋅cal ⎫ ⎧⎪ σ ac⋅cal ⎪⎫ ⎥
⎨1 – ´ ⎬ ⋅ σbcx ⎨ 1 – ´ ⎬ ⋅ σbcy ⎥
⎢ ⎩ F ⎭ ⎪ F ⎪
⎣ ex ⎩ ey ⎭ ⎦
´ ´
∴ Fex = 0.60 fce.x ; Fey = 0.60 fce.y
Therefore,

⎡ σ ac⋅cal Cmx σbcx ⋅cal Cmy ⋅ σmy⋅cal ⎤

⎢ + + ⎥ < 1.0
⎢ σ ac ⎧ σ ac⋅cal ⎫ ⎧ σ ac⋅cal ⎫ ⎥
⎨1 – ⎬ ⋅ σbcx ⎨1 – ⋅ σbey
⎢⎣ ⎩ 0 ⋅ 60 fbcx ⎭ ⎩ 0 ⋅ 60 fecy ⎬⎭ ⎥⎦

⎛ π2 E ⎞
where Fcc´ =⎜ 2⎟
...(3.41c)
⎜⎛ l⎞ ⎟
⎜⎝ ⎜⎝ ⎟⎠ ⎟⎠
r
As per IS : 800–1984, the structural members subjected to axial compression
and bending about both the axes shall be proportioned to satisfy Eq. 3.41 (c) as
above.
The adjusted interaction formula is satisfied for all points on the column
⎛σ ⎞
between lateral supports so long as ⎜ ac⋅cal ⎟ is equal to or less than 0.15.
⎝ σ ac ⎠

⎡ σ ac⋅cal σbcx .cal σbcy⋅cal ⎤

⎢ + + ⎥ < 1.0 ...(3.41d)
⎣ σ ac σbcx σbcy ⎦
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 215

The value of reduction factor, Cm depends on the relative size and the direction
of column end moments. The AISC 1963 specification recommends the following
for the value of reduction factor Cm (M1< M2) (as per IS : 800–1984 also)
where, M1 = Smallest end moment and
M2 = Largest end moment.
(i) For a member in frames where side sway is not prevented
Cm = 0.85
(ii) For members in frames where side sway is prevented and not subjected to
transverse loading between their supports in the plane of bending

⎛M ⎞
Cm = 0.6 – 0.4 ⎜ 1 ⎟ > 0.4
⎝ M2 ⎠
or Cm = 0.6 – 0.4β > 0.4
It is note that the β is the ratio of smaller to the larger moments at the ends
of that portion of the unbraced member in the plane of bending under
consideration. .

⎛M ⎞
It is to further note that β = ⎜ 1 ⎟ is negative for single curvature bending
⎝ M2 ⎠
and positive for reversed-curvature bending.
(iii) For the members in frames where the sidesway is prevented in the
plane of loading and subjected to transverse loading between their supports,
the value of Cm may be determined by rational analysis. In the absence of rational
analysis, IS : 800–1984 recommends the following values.
(a) For members whose ends are restrained against rotation
Cm = 0.85
(b) For members whose ends are unrestrained against rotation
Cm = 1.00
When the end moments act in the same direction, then, these give single
⎛M ⎞
curve deflection and the ratio ⎜ 1 ⎟ is negative. When the end moments acts in
⎝ M2 ⎠

⎛M ⎞
the opposite directions these give double curve deflection and the ratio ⎜ 1 ⎟
⎝ M2 ⎠
is positive.
The AISC specification recommends the values of reduction factor, Cm for
the following end conditions which are frequently met in the steel building as
shown in Fig. 3.31.
It is to note that the value of Cm should not be less than 0.4. The end conditions
of columns carrying transverse loads in addition to axial load is shown in Fig.
3.31 (d). In Fig. 3.31 (d) the bending stress is found at the point using larger
moment M.
216 DESIGN OF STEEL STRUCTURES

M2 P M2 P M2 P M P

M1 M
P P
P P
M1 = M2 M1 = 0 M 1 =1 /2 M 2 M=M
Cm = 1 C m = 0 .6 C m = 0 .4 Cm = 1
(a ) (b ) (c) (d )

Fig. 3.31

The effective length of column KL is larger than the actual length of a column,
in case, there is no bracing against sideway buckling. The value of Cm may be
larger than or equal to 0.85 in such cases. The straight line interaction formula
⎛σ ⎞
may be directly used when ⎜ ac⋅cal ⎟ is less than or equal to 0.15.
⎝ σ ac ⎠
The modified and adjusted interaction formula have been discussed above
for the central section of the column. It is also possible (depending upon the
slenderness ratio of a column unbraced in the plane of bending) that the combined
stress determined at one end may exceed that at all the points where the lateral
deflection is created by the end moments, even when the bending stress at these
points have been modified. The AISC specification recommends that the straight
line interaction formula may be used with fc = 0.6fy.

⎛ σ ac⋅cal σbcx .cal ⎞

+ < 1.0 ...(3.43a)
⎜⎝ 0 ⋅ 6 f σbc ⎟⎠
y

When the load is eccentric with x-axis and also with y-axis the Eq. 3.43 (a)
may be written as

⎡ σ ac⋅cal σbcx .cal σbcy⋅cal ⎤

⎢ + + ⎥ < 1.0 ...(3.43b)
⎣ 0 ⋅ 6 fy σbcx σbcy ⎦

Therefore, the following interaction formula may be used as per the

recommendations of AISC 1963 specifications
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 217

⎛ σ ac⋅cal ⎞
(i) For ⎜⎝ σ ⎟ < 0.15
ac ⎠

⎛ σ ac⋅cal σbcy.cal ⎞
⎜⎝ σ + < 1.0 ...(3.44)
ac σbc ⎟⎠

⎛ σ ac⋅cal ⎞
(ii) For ⎜⎝ σ ⎟ > 0.15
ac ⎠

⎡ σ ac⋅cal Cm σ ⎤
⎢ σ + ⋅ bc ⎥ < 1.0 ...(3.45)
⎢ ac (1 – σ ac⋅cal ) σbc ⎥
⎣⎢ Fc′ ⎦⎥
Equation 3.43 should be satisfied in addition to the above formulae at points
of lateral support in the plane of bending.
Instead of end moments, in case transverse loads are acting on the beam
column in between end supports, then the interaction formula in different forms
shall be used.

3.21 EQUIVALENT AXIAL LOAD METHOD OF DESIGN OF

ECCENTRICALLY LOADED COLUMNS
The columns are frequently subjected to eccentric loads in addition to the axial
loads. Therefore, the columns are subjected to direct compressive stresses and
bending stress (i.e., to combined stress). The method of design of eccentrically
loaded columns is to arrive at a trial section and then to check the trial section
for the combined stress. The trial section of an eccentrically loaded column may
be found conveniently by converting the eccentric load to an equivalent axial
load. The equivalent axial load is also termed as equivalent concentric load. The
equivalent axial load is the load of sufficient magnitude to produce a stress equal
to the maximum stress produced by the eccentric load. The magnitude of
equivalent axial load found is greater than the eccentric load. Consider a column
subjected to an eccentric load, PE as shown in Fig. 3.32.
The eccentricity of load PE is e x with respect to x-axis. The column is subjected
to direct compressive stress and bending stress. The maximum compressive
stress at the extreme fibre of the column is given by

⎛P P ⋅e ⎞
fc.1 = ⎜ E + E x ⎟
⎝ A Zxx ⎠

PE ⎛ A ⋅ ex ⎞
or fc.1 = ⋅ ⎜1 – ...(i)
A ⎝ Zxx ⎟⎠
218 DESIGN OF STEEL STRUCTURES

PE
ex

(a )
X

Y Y

X
(b )

Fig. 3. 32

Let Pequiv be the equivalent axial load. The equivalent axial load Pequiv is
equivalent to the eccentric load, PE. The equivalent axial load produces the
average compressive stress equal to the maximum compressive stress at the
extreme fibre of the column. The average compressive stress due to equivalent
axial load is given by

⎛ Pequiv ⎞
fc.2 = ⎜ ...(ii)
⎝ A ⎟⎠
Since fc.2 = fc.1

⎛ Pequiv ⎞ = PE ⎛1 + A ⋅ ex ⎞
⎜⎝
A ⎠
⎟ A ⎜⎝ Z xx ⎟⎠

⎛ A ⋅ ex ⎞
Pequiv = PE ⎜1 + ...(3.46)
⎝ Z xx ⎟⎠
Equation 3.46 gives the equivalent axial load. When the eccentric load is
having eccentricity, e x with respect to xx-axis, and eccentricity, e y with respect
yy-axis, then, the equivalent load is given by
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 219

⎛ A ⋅ ex A ⋅ e y ⎞
Pequiv. = PE ⎜⎜1 + + ⎟⎟ ... (3.47)
⎝ Z xx Z yy ⎠
When the column is also carrying an axial load in addition to an eccentric
load, then the total axial load acting on the column may be found as the sum of
axial load and the equivalent axial load.

3.22 BENDING FACTOR METHOD OF DESIGN OF

ECCENTRICALLY LOADED COLUMN
When the column is subjected to an eccentric load, then, it is subjected to a
moment M, due to eccentric load. The column moment is given by
Mxx = PE . ex
where PE = Eccentric load on the column
ex = Eccentricity of the column with respect of xx-axis.
The equivalent axial load P equiv. may be found in this method by multiplying
the moment due to eccentric load by the bending factor. The bending factor may
be obtained by dividing the cross-sectional area of the column. A, by the section
modulus, Z of the section. It is denoted by B. The compressive stress at the
extreme fibre of the column is equal to the average compressive stress due to
equivalent axial load, Pequiv. Therefore,

⎛ Pequiv ⎞ = ⎛ PE + M xx ⎞
⎜⎝ ⎟ ⎜ A Z xx ⎟⎠
A ⎠ ⎝

⎛ A ⎞
or Pequiv = ⎜ PE + M xx Z ⎟ ...(i)
⎝ xx ⎠

or Pequiv = (PE + Mxx . Bx) ...(3.48)

⎛ A ⎞
where, Bx = ⎜ ⎟ , bending factor with respect to axis of x.
⎝ Z xx ⎠
When the column is subjected to eccentric load having eccentricity e x, with
respect to x-axis and eccentricity, e y with respect to y-axis then, the column is
subjected to moments in two directions. Therefore,

⎛ Pequiv ⎞ = ⎛ PE + M xx + M yy ⎞
⎜⎝ ⎟ ⎜⎜ ⎟
A ⎠ ⎝ A Z xx Z yy ⎟⎠

⎛ A A ⎞
or Pequiv = ⎜ PE + M xx + M yy ⎟
⎝ Z xx Z yy ⎠

or Pequiv = (PE + M xx ⋅ Bxx + M yy ⋅ By ) ...(3.49)

220 DESIGN OF STEEL STRUCTURES

where Mxx = PE . ex,

Myy = PE . ey and
Bx = Bending factor with respect to x-axis
By = Bending factor with respect to y-axis.
The column may be subjected to moment instead of the eccentric load. It is
seen that the moments are reduced to the equivalent axial load. The column
may be carrying axial load also in addition to the eccentric load. The equivalent
axial load may be found by having the sum of axial load (if any) and eccentric
load, and the products of the bending moments due to eccentric loads and the
approximate bending factor.
It is to note that equivalent axial loads found in Eq. 3.46 and Eq. 3.48 are
same, and that found in Eq. 3.47 and Eq. 3.49 are also same.
The usual method to determine the approximate total equivalent load is based
upon the requirement of the stability check (i.e., the modified interaction
formula). The total equivalent axial load is equal to the sum of total axial load
which is used in determining the average axial stress, fc´, and the equivalent
axial load which accounts for the effect of eccentricity. The effect of the
⎛ f´ ⎞
eccentricity in the stability check is represented in the bending stress ratio ⎜⎜ bc ⎟⎟
⎝ fbc ⎠
´
⎛ fbc ⎞
⎜
The bending stress ratio ⎜ ⎟⎟ is modified by multiplying by the reduction factor,
⎝ fbc ⎠
Cm. In the similar manner, the equivalent axial load is multiplied by the reduction
factor Cm, prior to adding its effect to the axial load. Hence, the total equivalent
axial load is given by
Pequiv = PE + B . Cm .(PE .e) ...(3.50)
The trial section obtained for the total equivalent load is checked by Eq. 3.44
or Eq. 3.45. It is to note that when the moments at the ends of a column are
either equal to zero, value of Cm is unity and Eq. 3.45 assumes the same form as
Eq. 3.48 (since PE . e = M).
Example 3.19 A column HB 300, @ 0.630 kN/m is subjected to an eccentric
load. The eccentricity of the load from x-axis of the section is 28 mm. The
eccentricity of the load from y-axis of the section is 54 mm. The effective length of
column is 4 m. Determine the maximum and safe magnitude of the eccentric-
load for the section. Use equivalent axial load method.
Solution
Step 1 : Properties of section
From ISI Handbook No. 1 for HB 300, @ 0.630 kN/m
Zxx = 863.3 × 103 mm3
rxx = 127.0 mm
Zyy = 178.4 × 103 mm3
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 221

ryy = 52.9 mm
A = 8025 mm2
Step 2: Slenderness ratio
Minimum radius of gyration of the section
rmin = 52.9 mm
Effective length of the column l = 4 m
Maximum slenderness ratio of the section
⎛ l ⎞ ⎛ 4 × 1000 ⎞
⎜r ⎟ = ⎜ ⎟ = 75.7
⎝ min ⎠ ⎝ 52 ⋅ 9 ⎠
Step 3: Equivalent load
From IS: 800–1984, for the steel having yield stress as 260 N/mm2 and the
⎛ l ⎞
maximum allowable stress in axial compression for ⎜ r ⎟ = 75 .7 is σac = 108.16
⎝ min ⎠
2
N/mm .
Maximum safe axial load, which can be carried by the section
⎛ 108 ⋅ 16 × 8025 ⎞
pequiv = σac × A = ⎜ ⎟ = 867.984 kN ...(i)
⎝ 1000 ⎠

Let PE be maximum and safe eccentric load for the section. From Eq. 3 47
⎛ A ⋅ ex A ⋅ e y ⎞
pequiv = PE ⎜⎜1 + Z +
Z yy
⎟⎟ ...(ii)
⎝ xx ⎠
ex = 28 mm
ey = 54 mm
Substituting the numerical values
⎛ 8025 × 28 8025 × 54 ⎞
867.984 = PE ⎜1 + + ⎟
⎝ 863 ⋅ 3 × 103
178.4 × 103 ⎠

⎛ 867 ⋅ 984 ⎞ ⎛ 867 ⋅ 984 ⎞

∴ PE = ⎜ ⎟=⎜ ⎟ = 235.23 kN
⎝ 1 + 0 ⋅ 26 + 2 ⋅ 43 ⎠ ⎝ 3 ⋅ 69 ⎠
The maximum and safe eccentric load which may be carried by ISHB 300,
@ 0.630 kN/m at the given eccentricities is 235.23 kN.
Alternatively
Step 1 : Bending factors
Bending factor for HB 300, @ 630 N/m with respect to x-axis
⎛ A ⎞ ⎛ 8025 ⎞
Bx = ⎜ Z ⎟ = ⎜ 3 ⎟
⎝ xx ⎠ ⎝ 863 ⋅ 3 × 10 ⎠
= 0.0093 mm–1 ...(iii)
222 DESIGN OF STEEL STRUCTURES

Bending factor HB 300, @ 0.630 kN/m with respect to y-axis

⎛ A ⎞ ⎛ 8025 ⎞
Bx = ⎜ ⎟ =⎜ ⎟ = 0.045 mm
–1 ...(iv)
⎝ Z xx ⎠ ⎝ 178.4 × 103 ⎠
Mxx = PE .ex= 28PE ...(v)
Myy = PE .ey = 54PE ...(vi)
Step 2 : Equivalent load
From Eq. 3.49, the equivalent axial load is given by
Pequiv = (PE + Mxx Bx + Myy . By)
or Pequiv = (PE + 28PE × 0.0093 + 53 PE × 0.045)

⎛ Pequiv ⎞
or PE = ⎜
⎝ 1 + 0 ⋅ 26 + 2 ⋅ 43 ⎟⎠

⎛ 867 ⋅ 984 ⎞
= ⎜ ⎟ = 235.23 kN.
⎝ 3 ⋅ 69 ⎠
The maximum safe eccentric load found for the given section for given
eccentricities by using bending factors if also 235.23 kN.
Example 3.20 A column carries an axial load of 400 kN and an eccentric
load of 150 kN at 100 mm from x-axis. The actual length of column is 4 m. The
column is hinged at both the ends. Design the column by equivalent axial load
method.
Solution
Design :
Step 1: Selection of trial section
Axial load carried by the column, P = 400 kN ....(i)
Eccentric load carried by the column, PE = 150 kN ...(ii)
Eccentricity of the column from x-axis, e x = 100 mm ...(iii)
Bending moment
Mxx = PF .ex= (l50 × 100) = 15000 mm-kN ...(iv)
Equivalent axial load
Pequiv = (P + PE + Mxx . Bx) ...(vi)
Since, the column section is not known, the exact bending factor, Bx is also
not known. From ISI Handbook No. 1, HB 200, @ 0.400 kN/m is selected
tentatively. The bending factor for this column section
⎛ A ⎞ ⎛ 5094 ⎞
Bx = ⎜Z ⎟ =⎜ 3 ⎟ = 0.0137 mm
–1 ...(via)
⎝ xx ⎠ ⎝ 372 ⋅ 2 × 10 ⎠
Mxx Bx = (l5000 ×0.0137) = 205.5 kN ...(vii)
Pequiv = (400 + 150 + 205.5) = 755.5 kN ...(viii)
The slenderness ratio for the column and the value of yield stress are assumed
as 80 and 260 N/mm2, respectively. Therefore,
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 223

Allowable axial stress in compression as 103 N/mm2, cross-sectional area

required

Areqd. = ⎛⎜ 755.5 × 1000 ⎞⎟ = 7334.95 mm2 ... (ix)

⎝ 103 ⎠
From ISI Handbook No. l, select the column section HB 300, @ 0.630 kN/m
A = 8025 mm2
rxx = 127.0 mm
Zxx = 863 .30 × 103 mm3
ryy = 52.9 mm
Minimum radius of gyration of the section
rmin = 52.9 mm
Step 2 : Revised equivalent load
Revised brending factor
⎛ A ⎞ ⎛ 8025 ⎞
Bx = ⎜ ⎟ =⎜ 3 ⎟
Z
⎝ xx ⎠ ⎝ 863 ⋅ 3 × 10 ⎠
= 0. 0093 mm–1 ...(x)

Mxx . Bx = (l500 × 0.0093) = 139.5 kN ...(xi)

Revised equivalent axial load
Pequiv = (400 + 150 + 139.5) = 689.5 kN
Area required,
⎛ 689 ⋅ 5 × 1000 ⎞
A = ⎜ ⎟ = 6694.17 mm2 ...(xiii)
⎝ 103 ⎠
Area provided by HB 300, @ 0.630 kN/m is greater than area required.
Effective length of column
l = 4m ... (xiii)
Maximum slenderness ratio for the column
⎛ l ⎞ ⎛ 4 × 1000 ⎞
⎜r ⎟ = ⎜ ⎟ = 75.7 ...(xiv)
⎝ min ⎠ ⎝ 52 ⋅ 9 ⎠
From IS : 800–1984, the maximum allowable stress in axial compression for
the steel having yield stress as 260 N/mm2 and
⎛ l ⎞
⎜ r ⎟ = 75.7 is σac =108.16 N/mm2
⎝ min ⎠
Check :
The column section may be checked by interaction formula. The average axial
stress

⎛ Total vertical load ⎞ ⎛ (400 + 150 ) × 1000 ⎞

σ ac⋅cal = ⎜ = ⎟ = 68.536 N/mm2
⎝ Cross-sectional area ⎟⎠ ⎜⎝ 8025 ⎠
224 DESIGN OF STEEL STRUCTURES

Actual bending stress

⎛ M xx ⎞ ⎛ 15000 × 1000 × 150 ⎞

σ ac⋅cal = ⎜ I ⎟y= ⎜ ⎟ = 17.374N/mm2
⎝ xx ⎠ ⎝ 12950 ⋅ 2 × 104 ⎠
Allowable bending stress in compression from IS : 800–1984
0.60fy = (0.60 × 260) = 156 N/mm2

⎛ σ ac⋅cal σbc.cal ⎞ ⎛ 68 ⋅ 536 17 ⋅ 374 ⎞

⎜⎝ σ + =⎜ + ⎟ = 0.745 < 1.00
ac σbc ⎟⎠ ⎝ 108 ⋅ 16 156 ⎠
Hence, the design is satisfactory. Provide HB 300, @ 0.630 kN/m column
section.
Example 3.21 Verify that the column designed in Example 3.20, satisfy the
AISC requirement.
Solution :
Step 1 : Properties of section
From ISI Handbook No. 1 for the column section HB 800 @ 0.630 kN/m
designed in Example 3.20.
A = 8025 mm2 rxx = 127 mm
Zxx = 863.3 × 103 mm3 ryy = 52.9 mm
Step 2 : Check by interaction expression
Average axial stress in compression

⎛ (400 + 150 ) × 1000 ⎞

σac.cal = ⎜ ⎟
⎝ 8025 ⎠
= 68.536 N/mm2 ...(i)
Allowable axial stress in compression
σac.cal = 108.16 N/mm2 ...(ii)

⎛ σ ac⋅cal ⎞ ⎛ 68 ⋅ 536 ⎞
⎜⎝ σ ⎟⎠ = ⎜⎝ 108 ⋅ 16 ⎟⎠ = 0.634 > 0.15 ...(iii)
ac

From Eqn. 3.45

⎡ σ ac⋅cal Cm σ ⎤
⎢ σ + ⋅ bc⋅cal ⎥ < 1.00 ...(iv)
⎢ ac (1 – σ ac⋅cal ) σbc ⎥
⎢⎣ Fc′ ⎥⎦

⎛ σ ac⋅cal ⎞ ⎛ 17 ⋅ 374 ⎞
⎜⎝ σ ⎟⎠ = ⎝⎜ 156 ⎠⎟ = 0.1114 ...(v)
ac

The column is hinged at both ends. The column end moments are M1 = M2
= 0. Therefore,
Cm = l ...(vi)
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 225

⎛ 1052000 ⎞
Fc′ = ⎜ N/mm2 ... (vii)
⎝ (l / r )2 ⎟⎠
The column has eccentricity about x-axis. The bending takes place about x-
axis .
⎛ l ⎞ ⎛ 4 × 1000 ⎞
⎜ r ⎟ = ⎜⎝ 127 ⎟⎠ = 31.6 ...(viii)
⎝ xx ⎠
⎛ 1052000 ⎞
Fe′ = ⎜ = 1052 N/mm2 ...(ix)
⎝ 31 ⋅ 6 × 31 ⋅ 6 ⎟⎠

⎛ σ ac⋅cal ⎞ ⎛ 68 ⋅ 536 ⎞
⎜⎝ F ′ ⎟⎠ = ⎜ ⎟ = 0.065 ...(x)
e ⎝ 1052 ⎠

⎛ σ ac⋅cal ⎞
⎜⎝1 – F ⎟⎠ = (1– 0.065) = 0.935 ...(xi)
e

⎡ Cm ⎤ ⎛ 1 ⎞
⎢⎛ = ⎜ ⎟ = 1.07 ...(xii)
σ ac⋅cal ⎞ ⎥ ⎝ 0 ⋅ 935 ⎠
⎢ ⎜1 – ⎥
⎣⎢ ⎝ Fc′ ⎟⎠ ⎦⎥

⎡ σ ac.cal Cm σ ⎤
∴ ⎢ σ + ⋅ bc ⎥ < 1.00
⎢ ac (1 – σ ac⋅cal ) σ bc ⎥
⎢⎣ Fe′ ⎥⎦
(0.634 + 1.07 × 0.1114) = 0.753 < 1.00
⎛σ ⎞
when ⎜ ac⋅cal ⎟ > 0.15, a point braced in the plane of bending, Eq. 3.43 should
⎝ σ ac ⎠
also be satisfied. Therefore,
⎛ σ ac⋅cal σbcx .cal ⎞ ⎛ 68 ⋅ 536 ⎞
+ = ⎜ + 0 ⋅ 1114 ⎟ = 0.551< 1.00
⎜⎝ 0 ⋅ 6 fy σbc ⎟⎠ ⎝ 0 ⋅ 6 × 260 ⎠
Hence, the design is satisfactory. The yield stress fy, steel has been adopted
as 260 N/mm2.

3.23 COLUMN SPLICE

A joint in the length of a column provided, when necessary, is known as column
splice. It is also described as column joint. The rolled steel sections are
manufactured upto a certain maximum length. When length of the column
required is more than the length of rolled steel sections manufactured, in that
case, column splice becomes necessary. Secondly, when the columns is used in
multi-storey buildings then the sectional area for the column in the upper storey
is less than that for the columns in upper storey and lower storey separately. In
practice, ends of the column are cut by the following two methods:
226 DESIGN OF STEEL STRUCTURES

1. Ends of column cut by ordinary method.

2. Ends of column cut and milled.
When end of the column are cut by ordinary method, the load is transferred
to the lower column through rivets. When the ends of the columns are cut and
milled, then column has complete bearing over the whole area and ends of the
column are flush ends. Different types of riveted column splices have been shown
in Fig. 3.33.

(a ) (b ) (c)
Fig. 3.33 Column splices

The column splice, as shown in Fig. 3.33 (a) is used when the depth of columns
sections are equal. The column splice, as shown in Fig. 3.33 (b), is used; when
the depth of upper column is smaller than the lower column, then filler plates
are placed in between column splice plates and column. When the depth of the
section of upper column is much smaller than the lower column and it does not
provide full bearing, bearing plate is placed as shown in Fig. 3.33 (c).

3.24 DESIGN OF COLUMN SPLICE

Following are the usual steps in the design of column splice. The specifications
for the design as per IS : 800–1984 have also been given :
Step 1. (i) When the ends of the column are cut by ordinary methods, or not
faced for complete bearing, the column splice is designed for full axial load and
other forces to which the joint is subjected.
(ii) When the ends of the column are faced for bearing over the whole area,
the joint is designed to hold the connected members accurately in place and to
resist tension (if any) and bending if present. The ends of compression members
faced for bearing shall invariably be machined to ensure perfect contact of
surfaces in bearing. The rivets essential to hold the connected members may be
designed for 50 per cent axial load (as advised in IS : 800–1950) and other force
acting on the column.
(iii) When possible, splices shall be proportioned and arranged so that the
centroidal axis of the splice coincides as nearly as possible with the centroidal
axes of members joined in order to avoid eccentricity ; but where eccentricity is
present in the joint the resulting stress shall be provided for.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 227

Step 2. The column splice plates may be assumed to act as short column of
zero slenderness ratio. Allowable working stress in compression in the splice
plates may be found for the steel to be used ; and the sectional area of splice
plates required is computed.
Step 3. The width of the splice plates is kept equal to the width of flange of
the column and thickness is computed.
Step 4. The nominal diameter of rivets used in the joint is assumed and rivet
value is computed. The number of rivets essential to resist force is computed by
dividing the force by rivet value.
Step 5. When moment and shear force are also acting in addition to the axial
load, column splice plates attached with the flanges are assumed to resist axial
force and moment. Column splice plates to resist maximum shear force are
provided with the web. Additional force due to moment is computed by dividing
moment by lever arm between splice plates attached with the flanges. The rivets
and splice plates are checked for tension due to bending.
Step 6. When upper column does not provide full bearing area over lower
column, bearing plate provided is designed as below :
(i) The bearing plate may be assumed as short beam to transmit the axial
load to the lower column section.
(ii) The axial load of the column is assumed to be taken by flanges, neglecting
the load taken by the web. The load transmitted from flanges of upper column
andreaction from flanges of lower column are equal and form a couple as shown
in Fig. 3.34.
P
(iii) Moment due to couple, M is ×a
2
1 2
(iv) Moment of resistance of plate, is bt × σbs
6

P /2 P /2

P /2

Fig. 3.34
228 DESIGN OF STEEL STRUCTURES

where, b = width of bearing plate and it is equal to the width of flange of

column
t = thickness of bearing plate
σbs = allowable bending stress in slab or bearing plates.
By equating moment of resistance and moment due to couple, the thickness
of bearing plate is given by
1/2
⎛ 6M ⎞
t = ⎜ ⎟
⎝ b × σbs ⎠
Example 3.22 A column section HB 250, @ 0.510 kN/m carries an axial load
of 600 kN. The ends of the column are cut by ordinary method. Design the column
splice.
Solution
Design :
Step 1 : Allowable stress in axial compression
The slenderness ratio for the splice plates (being of small length) and the
value of yield stress for the steel to be used may be assumed as zero and
260 N/mm2, respectively.
Therefore, the allowable stress in axial compression may be adopted as 0.6 fy
= 156 N/mm2

IS H B 2 50 ,
@ 0 .51 0 kN /m

S p lice p late
1 0 m m th ick

10
R ive ts on
e ith er
sid e o f
splice

IS H B 2 50 ,
@ 0 .51 0 kN /m

Fig. 3.35

Step 2 : Area of splice required

⎛ 600 × 1000 ⎞
A = ⎜ ⎟ = 3846.154 mm2
⎝ 156 ⎠
Area of one splice plate required = 1923.07 mm2
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 229

Width the splice plate is kept equal to the width of flange of column = 250
mm.
Required thickness of splice,
⎛ 1923.07 ⎞
t = ⎜ ⎟ = 7.692 mm
⎝ 250 ⎠
Adopt thickness of splice plate = 10 mm
Step 3 : Rivet value
Provide 22 mm nominal diameter of rivet. The strength in single shear for
power driven rivets
⎛ π (23 ⋅ 5)2 × 100 ⎞
⎜⎝ × ⎟⎠ = 43.35 kN
4 1000
Thickness of flange of column is 9.7 mm.
Strength in bearing for power driven rivets
⎛ 23 ⋅ 5 × 300 ⎞
9 ⋅7× ⎜ ⎟ = 68.385 kN
⎝ 1000 ⎠
Rivet value R = 43.35 kN
⎛ 600 ⎞
Number of rivets required is ⎜ = 13.84⎟
⎝ 43.35 ⎠
Provide 16 rivets, 8 rivets are provided on one side and 8 rivets are provided
on the other as shown in Fig. 3.35.
Example 3.23 A column section BH 400, @ 0.822 kN/m is carrying an axial
of 500 kN and a moment of 225000 mm-kN and shear force 45 kN. Design a
column splice.
Solution
Design :
Step 1: Allowable stress in axial compression
It is assumed that the ends of column are faced for complete bearing over
whole area (milled or flush ends). Therefore, 50 per cent of axial load is
transmitted through the splice plates and 50 per cent by direct bearing.
Axial load to be transmitted through one splice plate.

⎛1 ⎞
= ⎜ × 500⎟ = 125 kN
⎝4 ⎠
It is assumed that the thickness of splice plates is 6 mm and the moment is to
be resisted fully by flanges and shear force by web only
Lever arm
(400 + 6) = 406 mm
Force to be resisted by splice plate because of moment

⎛ 22500 ⎞
⎜⎝ 406 ⎟⎠ = 55.42 kN
230 DESIGN OF STEEL STRUCTURES

Total load
(125 + 55.42) = 180.42 kN
The slenderness ratio of the splice plates and the value of yield stress for
steel to be used are assumed as zero and 260 N/mm2 respectively.
The allowable working stress in compression is splice plate
0.6 fy = 156 N/mm2
Step 2: Area of one splice plate required
⎛ 180.42 × 1000 ⎞
⎜⎝ 156 ⎟⎠ = 1156.54 mm2
Provide width of splice plate equal to 250 mm
:. Thickness of splice plate required
⎛ 1156.54 ⎞
⎜⎝ 250 ⎟⎠ = 4.626 m
Adopt thickness of splice plate as 6 mm
Step 3 : Rivet value
Provide 22 mm diameter power driven rivets
Strength of rivet in single shear
⎛ π (23.5)2 × 100 ⎞
⎜4 × 1000 ⎟ = 43.35 kN
⎝ ⎠

IS H B 4 00 ,
@ 0 .82 2 kN /m

S p lice p late
6 m m thick

8
R ive ts on
e ith er
sid e o f
splice

IS H B 4 00 ,
@ 0 .82 2 kN /m

Fig. 3.36

Strength of rivet in bearing

⎛ 23.5 × 6 × 300 ⎞
⎜⎝ 1000 ⎟⎠ = 42.30 kN
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 231

Rivet value R = 42.30 kN

⎛ 180.42 ⎞
Number of rivets required on one side is ⎜ = 4.26⎟
⎝ 42.30 ⎠
Provide 8 rivets.
The tension due to moment in splice plate 55.42 kN is less than axial
compression in splice plate 125 kN.
Splice plate for S.F. is designed for maximum shear
Maximum shear force in web = l.5 × 45 = 67.5 kN
Maximum allowable shear stress
0.45fy = (0.45 × 260)
= 117 N/mm2
⎛ 675 × 1000 ⎞
Area of plates required = ⎜⎝ ⎟⎠ = 576.92 mm2
117

⎛ 45 ⎞
Number of rivers required = ⎜⎝ = 1.038
43.35 ⎟⎠
Provide two rivets. Provide splice plates 6 mm thick to accommodate two
rivets and to give above areas as shown in Fig. 3.36.
Example 3.24 A column section HB 350, @ 0 674 kN/m is carrying an axial
load of 900 kN. It is to be supported over a column section HB 450, @ 0.872 kN/m.
Design the column splicing.
Solution
Design :
Step 1: Allowable stress in bending in compression
The column section HB 350, @ 0.674 kN/m does not provide full bearing area
over lower column section HB 450, @ 872 N/m. The bearing plate 450 mm × 250
mm × t mm is placed as shown in Fig. 3.37.
The column load is assumed to be transferred by the flanges.
The load on each flange is (900/2) = 450 kN
Distance between C.G. of flanges in HB 350, @ 0674 kN/m
(350–116) = 338.4 mm
Distance between C.G. flanges in HB 450, @ 0.872 kN/m
(450–13 7) = 436.3 mm
Distance between lines of action of forces
1 1
(436.3 − 338.4) = × 97.9 = 48.95 mm
2 2
Moment due to couple
(450 × 48.95) = 22000 mm-kN
Moment of resistance of bearing plate

⎛1 2 ⎞
⎜⎝ 6 bt × σbc ⎟⎠ = 22000
232 DESIGN OF STEEL STRUCTURES

Allowable stress in bending in compression

σbc = (0.66 × 250) N/mm2
Step 2 : Thickness of splice plate
⎛1 2 ⎞
or ⎜⎝ 6 × 250 × t × 0.66 × 260⎟⎠ = 22000 × 1000
∴ t = 55 mm
Adopt thickness of 56 mm (as manufactured) and provide filler plate of 50
mm thick on either side.
Splice plate. The column ends are faced for complete bearing over whole
area. The slenderness ratio of splice plates and the value of yield stress for steel
to be used are assumed as zero and 260 N/mm2. Therefore allowable working
stress in compression in splice plates 0.6 fy = 156 N/mm2.
Splice plates are designed for 50 per cent load, viz. 450 kN
Area of splice plates required
⎛ 450 × 1000 ⎞
⎜⎝ 156 ⎟⎠ = 2884.62 mm2

IS H B 3 50 ,
@ 0 .67 kN /m

Filler pla te
5 0 m m th ick
S p lice
flate
8 mm
P /2 P /2 th ick

P /2 P /2 B e aring
p la te
56 m m
th ick

IS H B 4 50 ,
@ 0 .87 2 kN /m

Fig. 3.37

Area of one splice plate required

= 1442.31 mm2
The width of splice plate is kept equal to the width of flange of column section
= 250 mm
⎛ 1445.31 ⎞
Thickness of splice plate is ⎜ = 5.769 mm⎟
⎝ 250 ⎠
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 233

Provide 8 mm thickness for splice plate.

Step 3 : Rivet connections. Use 22 mm diameter power driven rivets.
Strength of rivet in single shear

⎛π 2 100 ⎞
⎜⎝ 4 (23.5) × 1000 ⎟⎠ = 43.35 kN

Strength of rivet in bearing

⎛ 23.5 × 8 × 300 ⎞
⎜⎝ 1000 ⎟⎠ = 56.4 kN

⎛ 450 ⎞
Number of rivets required is ⎜ = 10.38
⎝ 43.35 ⎟⎠
In HB 350, @ 0.674 kN/m, rivets are placed through filler plates as shown in
Fig. 3.37.
Thickness of filler plate is 44 mm. It is excess than 6 mm
Rivets are increased by 25% for each 200 mm i.e., 27.5%
= 10.38 × 1.275 = 13.23
Provide 16 rivets. The additional rivets are provided in each flange in filler
plates.

3.25 ENCASED COLUMNS

When a steel column is encased in cement concrete, it is called encased.
A steel column is embedded in concrete for two purposes. It may be necessary
to make a steel frame building fire resisting. The columns may be encased in
cement concrete for architectural requirements. As per IS : 800–1984 a member
may be designed as an encased column when the following conditions are fulfilled:
(a) The member is of symmetrical I-shape or a single I-beam or channels
back to back, with or without flange plates.
(b) The overall dimensions of the steel do not exceed 750 mm × 450 mm
over plating where used, the larger dimensions being measured parallel
to web.
(c) The column is unpainted and is solidly encased in ordinary dense
concrete with 20 mm aggregate (unless solidity can be obtained with a
larger aggregate) and grade designation of concrete minimum M 15.
(d) The minimum width of solid casing is equal to (b0 + 100 mm) where
b0 is the width of the steel flanges in millimetres.
(e) The surface and edges of the steel column have a concrete cover of not
less than 50 mm.
(f) The casing is effectively reinforced with steel wires. The wires shall be
at least 5 mm of diameter and the reinforcement shall be in the form of stirrups
or bending at not more than 150 mm pitch so arranged as to pass through the
centre of the covering of the edges and outer faces of the flanges and supported
by longitudinal spacing bars not less than four in number.
234 DESIGN OF STEEL STRUCTURES

(g) The steel cores in encased columns shall be accurately machined at

splices and provisions shall be made for the alignment of the column. At the
column base the provision shall be made to transfer the load to the footing at
safe unit stresses in accordance with IS : 456–1978.
The encased columns are designed as follows :
The steel column section shall be considered as carrying the entire load but
allowance may be made for stiffened effect of the concrete. This allowance should
be made by assuming the radius of gyration r of the column section about the
axis in the plane of its web to be 0.2 (b0+ l00) mm, where b0 is the width of the
steel flanges in mm. The radius of gyration about its other axis shall be taken as
that of the uncased section.
The axial load on the cased column shall not exceed 2 times that which would
be permitted on the uncased section, nor shall the slenderness ratio of the
uncased section for its length centre to centre of connection exceed 250.
In computing the allowable axial load on the cased strut, the concrete shall
be taken as assisting in carrying the load over its rectangular cross-section, any
cover in excess of 75 mm from the overall dimension of section of the cased strut
being ignored. The allowable compressive load, P in case of encased columns
shall be determined as follows:
P = (Asc . σsc + Ac . σc)
where, Asc, Ac= cross-sectional area of steel and concrete, and
σsc, σc = permissible stress in steel and concrete
It is to note that above clause does not apply to steel struts of overall
dimensions greater than 1000 mm × 500 mm, the dimension of 1000 mm being
measured parallel to the web or to box sections.
Design procedure. Following are the steps in design of cased columns :
Step 1. The slenderness ratio for column and the value of yield stress for
steel to be used are assumed as about 70 and 260 N/mm2 (or any other value).
And the allowable axial compressive stress, σac, is noted from IS : 800–1984.
Step 2. The effective sectional area A, required for the compression member
is determined

⎛ P ⎞
A = ⎜
⎝ σ ac ⎟⎠
where, P = load to be carried by the member.
Step 3. From IS Handbook No. 1, a section for the compression member of
the required area is selected.
The section for the compression member is selected such that it has the largest
possible radius of gyration for the required sectional area. It should also be
most economical section.
Step 4. The geometrical properties of section are noted. The effective length
of column is found as per end conditions. The slenderness ratio of uncased column
is found. The allowable axial compressive stress for uncased column is found
from IS : 800–1984.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 235

Step 5. The safe load carrying capacity of the column is found.

Step 6. For the cased column, radius of gyration about yy-axis is found. yy-
axis is kept parallel to plane of web,
ryy = 0.2 (b0+100) mm
where, b0 = width of flange in mm.
Step 7. The slenderness ratio of cased column is determined. The allowable
axial compressive stress is determined from IS : 800–1984.
Step 8. The load carrying capacity of the cased column is determined. The
load carrying capacity of the column should be equal to or slightly greater than
the load on the column. The design may be revised if the load carrying capacity
is less than or much greater than load on the column.
Step 9. Checks for slenderness ratio and axial load on the cased column are
provided.
Provided sufficient longitudinal steel bars to support stirrups. Provide stirrups
of 5 mm diameter at a pitch of 150 mm. The cement concrete is provided to
provide a cover of 50 mm on all sides.
Example 3.25 Design a cased column to carry an axial load of 1100 kN. The
column is 4 m long and adequately restrained in position but not in direction at
both the ends.
Solution
Design :
Step 1: Selection of trial section
The slenderness ratio for the column and the value of yield stress for the
steel to be used are assumed as 70 and 260 N/m2, respectively. Therefore,
Allowable axial compressive stress =115 N/mm2
Effective length of column, l = 4 m
Effective sectional area required
⎛ 1100 × 1000 ⎞
A = ⎜ ⎟⎠
⎝ 115
= 9565.22 mm2
From ISI Handbook No. 1, try HB 450, @ 0.872 kN/m
Sectional area, A = 11114 mm2
rxx = 187.8 mm, ryy = 51.8 mm,
∴ rmin = 51.8 mm
Step 2 : Maximum slenderness ratio of uncased column
⎛ l ⎞ ⎛ 4 × 1000 ⎞
⎜⎝ r ⎟⎠ = ⎜⎝ 51.8 ⎟⎠ = 77.2
min
From IS : 800–1984, allowable axial compressive stress
= 106.36 N/mm2
Step 3 : Safe load carrying capacity of uncased column

⎛ 106.36 × 11114 ⎞
⎜⎝ 1000 ⎟⎠ = 1182.08 kN
236 DESIGN OF STEEL STRUCTURES

The column is encased in cement concrete. The cement concrete provides a

50 mm cover on all sides. 5 mm diameter stirrups are provided at 100 mm pitch
(|
> 150 mm). The stirrups are kept in position by 4 longitudinal bars 10 mm in
diameter as shown in Fig. 3.38.

S tirru ps
5 m m diam eter
a t 1 00 m m
p itch

4 50 m m

5 50 m m
L on gitud in al
b ars 1 0 m m
d iam e te r
50 2 50 m m 50
mm mm
3 50 m m
Fig. 3.38 Cased column

Step 4 : Radius of gyration for cased column

ryy = 0.2 (b0 + 100) mm, b0 = 250 mm
ryy = 0.2 (250 + 1000) = 70 mm
Step 5 : Maximum slenderness ratio of cased column
⎛ l 4 × 1000 ⎞
⎜r = = 57.14
⎝ yy 70 ⎟⎠
From IS : 800–1984, allowable axial compressive stress in steel
= 128.86 N/mm2
Area of concrete = Total area less area of steel core
(350 × 550 – 11114) mm2 = (192500 – 11114) = 181386 mm2
Let the grade concrete is M 15 for which αac is 4 N/mm2.
Step 6: Safe load carrying capacity of the cased column as per IS :
800–1984
⎛ 181386 × 4 128.86 × 11114 ⎞
P = ⎜ + ⎟⎠ kN
⎝ 1000 1000
P = (725.54 + 1432.15)
= 2157.69 kN
Check : Slenderness ratio of uncased column is 77.2 (< 250).
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 237

PROBLEMS
3.1. A rolled steel beam HB 300, @ 0.588 kN/m is used as a column. The column
is fixed in position but not in direction at both ends. Determine the safe
load carrying capacity of the column if the length of column is 4.50 m.
3.2. A single angle discontinuous strut ISA 130 mm × 130 mm × 10 mm (ISA
130 130, @ 0.197 kN/m) with single riveted connection is 2.5 m long.
Calculate safe load carrying capacity of section.
3.3. In Problem 3.2, if single discontinuous strut is double riveted, determine
safe load carrying capacity of the section.
3.4 A double angle discontinuous strut ISA 150 mm × 75 mm × 10 mm (ISA
150 75, @ 0.165 kN/m) long legs back to back, is connected to both sides of
a gusset plate 10 mm thick with 2 rivets.
The length of strut between centre to centre of intersections is 3.50 m.
Determine the safe load carrying capacity of the section.
3.5. In Problems 3.4, if double angle discontinuous strut is connected to one
side of a gusset, determine safe load carrying capacity of the strut.
3.6. In Problem 3.4, if double angle strut is continuous and connected with the
gusset plate with single rivet; determine safe load carrying capacity of
strut.
3.7. A built-up column consists of two MC 400, @ 0.494 kN/m and two plates
500 mm × 10 mm. The clear distance between back to back of channels is
200 mm. One plate is connected to each flange side. Determine the safe
load carrying, capacity of built-up section if the effective length of column
is 5 m.
3.8. Design a rolled steel beam section column to carry an axial load of 1120
kN.
The column is 6.20 m long and adequately restrained in position and
direction at both ends.
3.9. Design a single angle discontinuous strut to carry 80 kN load. The length
of the strut between centre of intersection is 2 .75 m.
3.10. Design a double angle discontinuous strut to carry 120 kN load. The length
of strut between centre to centre of intersection is 3.80 m.
3.11. A built-up column consists of two channels MC 250, @ 0.304 kN/m placed
back to back with 147 mm gap between them. Determine the safe load on
the column when (i) channels are effectively laced together, (ii) channels
are effectively battened together. Take effectively length of the column to
be 6000 mm.
3.12. Design a built-up column to support 1200 kN axial load. The length of
column is 18.5 m. The column is fixed at both ends. Design (i) single lacing,
(ii) double lacing for the built-up column.
3.13. Design a built-up column of effective length 5 m to carry an axial load of
1000 kN. Also, design a suitable lacing system for the above column.
3.14. An axially loaded built-up strut is made up of two channels back to back
with a gap of 200 mm. The column is 3 metres long and is pinned at both
ends. The individual channels of the column are MC 300, @ 0.858 kN/m.
238 DESIGN OF STEEL STRUCTURES

Design a suitable lacing for the column. The properties of the channels are
Area = 4564 mm2
Depth of section = 300 mm
Width of flange = 90 mm
Thickness of flange = 13.6 mm
Thickness of web = 7.6 mm
Centre of gravity Cyy = 23.6 mm
Ixx = 6362.6 × 104 mm4 Iyy = 310.8 × 104 mm4
rxx = 118.1 mm, ryy = 26.1 mm
3.15. Design a mild steel built–up column to carry an axial load of 2000 kN with
an effective height of 6.10 m. Use 4 m.s. angle sections laced together to
form the column. Design also a suitable system of lacing bars.
3.16. A column 400 mm × 400 mm consists of 4 angles of ISA 80 mm × 80 mm ×
10 mm (ISA 80 80, @ 0118 kN/m) as shown in Fig. P3.16. The column is 8
m long and is hinged at both ends. Find the maximum safe load for the
column and design a suitable lacing system for it. Draw a neat sketch of
the column showing connection details (Properties of each angle are: Area
= 1505 mm2, C.G. line Cxx = Cyy = 23.6 mm, Ixx = Iyy = 87.7 × 104 mm4, rmin
= 15.5 mm, rivet gauge distance = 45 mm.

ISA
8 0 m m × 80 m m × 10 m m
(IS A 8 08 0, @ 0 .118 kN /m )

x x

Fig. P 3.16

3.17. Design a built-up column to carry an axial load of 3600 kN. The effective
length of column is 5 m. Provide two channel sections and plates.
3.18. Design a steel column 3 metre long carrying an axial load of 500 kN. Assume
both the ends to be pin-jointed. Only channel sections are available.
3.19. Design a battened column to carry an axial load of 1100 kN. The effective
length of column is 7.00 m.
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 239

3.20. The ground floor column of a multistorey building is 5 metres high and is
subjected to an axial load of 1000 kN. The column height of the floor next
to the ground floor is 4 metres. The beams connected to the flanges of the
column have an eccentric load of 10 kN. The beams connected to the web
of the column have an eccentric load of 6 kN. Design and sketch the section
of the column. Adopt bending stress as per IS : 800–1984.
3.21. The cross-section of a 6 m long, pin ended column consists of 4 ISA 100
mm × 100 mm × 10 mm (4 ISA 100 100, @ 0.149 kN/m) suitably connected
with lacing bars. The angles face inwards and the outside dimensions of
the cross-section are 350 mm × 350 mm as shown in Fig. P 3.21.
(i) Determine the safe axial compressive load for the column.
(ii) Also design the lacing bars and their connection to angles.

3 50 m m

35 0 m m

(4 IS A 1 00 10 0.× 0 .1 49 kN /m )

4 IS A 1 00 × 10 0 × 1 0 m m

Fig. P 3.21

3.22. A column section HB 350, @ 0.674 kN/m is carrying an axial load of 400
kN and a moment of 18000 mm-kN and shear force 36 kN. Design a column
splice.
3.23. A column of section WB 250 mm × 200 mm @ 0.409 kN/m carries an axial
load of 800 kN. If the allowable bearing stress in concrete base in 4 N/
mm2, design a suitable base plate. Also design a suitable grillage foundation
if the allowable ground pressure is 100 kN/m2.
3.24. Design a column to carry a load of 800 kN at an eccentric distance of 6 mm
from the centroidal axes of the column along the web. The effective length
of column is 6.5 m.
3.25. A column of section WB 300, @ 0.481 kN/m (A = 6133 mm2 ; bf = 200 mm ;
tf = 100 mm ; Ix = 9821.6 × 104 mm4, Iy = 9901.1 × 104 mm4) along with one
plate of 300 mm × 10 mm on each flange as shown in Fig. P3.25 is used as
a column over an effective length of 5 metres. Determine the safe load the
column can carry and also design a suitable riveted connection between
flange and the flange plates. Use 20 mm φ rivets. Assume reasonable values
of any other data required.
240 DESIGN OF STEEL STRUCTURES

10 m m

IS W B 30 0
@ 0 .48 1 kN /m 3 00 m m

10 m m

Fig. P 3.25

3.26. A column made of an I-beam is 8 metres long and has hinged ends. The
column has to carry an axial load of P1 = 250 kN and a bracket load of P2 =
100 kN, the latter applied eccentrically on the y-axis of the column at a
distance of 250 mm from the centroid of the section. Check the safety of
the column.
Properties of the I-beam ; Depth = 350 mm, flange width = 250 mm, area =
9221 mm2, Ix = 19803 × 104 mm4 minimum radius of gyration = 52.2 mm.
Adopt permissible stress (as per IS : 800–1984).
3.27. Design a compound column to carry an axial of 500 kN. The column is
built-up of two channels placed back to back and laced together. The
equivalent free bending length is 6 m.
Details of channels LC 200, @ 0206 kN/m and MC 200, @ 0221 kN/m are
given. Use permissible stress in compression as per IS : 800–1984.
3.28. A steel column to take a central load of 1600 kN is to be built-up of four
equal angles forming a square 500 mm × 500 mm. The height of the column
is to be 6 m with hinged ends. Design
(a) a suitable column section
(b) a lacing system, and
(c) a steel base plate for the column if it is to rest on a concrete block
which can satisfy take a load of 100 kN/m2.
Give a dimensional sketch for the design. Use permissible as per IS : 800–
1984. The distance of the centre of gravity x for the angle is given in
Fig. P 3.28.

K = S ide o f e qu al an g le
x = D ista nce o f C .G . fro m fa ce
CG = 0.27 K

Fig. P 3.28
 DESIGN OF COLUMNS AND COMPRESSION MEMBERS 241

3.29. The compression member of a bridge is composed of two channel sections,

the width over the back of the channel sections being 400 mm. The member
carries a load of 100 kN, the length being 10 metres. Using MC 350
channels, design
(a) the compression member, and
(b) the lacing system together with details of connections,
Adopt permissible stresses as per IS : 800–1984.
Consider the location of the centroid of channel section at approximately
0.24 times the width of the flange from the back surface of the web.
3.30. Select a suitable H-section for a column carrying an axial load of 350 kN
and a moment 25000 mm-kN about major axis. The effective length of
column is 6 m.
3.31. A mid steel column 6 m high has its ends firmly built-in. The column is
built-up with two channels MC 300 placed back to back with 180 mm gap
between them. The channels are effectively placed together. Determine
the safe load on the column. Design also the lacing for it.
 Chapter
Design of Column Bases

4 and Column Footings

4.1 INTRODUCTION
The columns are supported on the column bases. The column bases transmit
the column load to the concrete or masonry foundation blocks. The column load
is spread over large area on concrete or masonry blocks. The intensity of bearing
pressure on concrete or masonry is kept within the maximum allowable bearing
pressure. The safety of the structure depends upon stability of foundation. The
column bases should be designed with utmost care and skill. In the column
bases, intensity of pressure on concrete block is assumed to be uniform as shown
in Fig. 4.1. The column bases shall be of adequate strength, stiffness and area
to spread the load upon the concrete, masonry, other foundation or other supports
without exceeding the allowable stress on such foundation under any
combination of the load and bending moments. The column bases are of two
types :
1. Slab bases ; and
2. Gusseted bases.
The column footings are designed to sustain the applied loads, moments and
forces and the induced reactions. The column load is spread over large area, so
that the intensity of bearing pressure between the column footing and soil does
not exceed the safe bearing capacity of the soil. It is ensured that any settle-
ment which may occur shall be as nearly uniform as possible and limited to an
accepted small amount. The column load is first transmitted to the column foot-
ing through the column base. It is then spread over the soil through the column
footing. The column footings are of two types :
1. Independent footings; and
2. Combined footings.
The designs of column base (slab and gusseted bases) and the column foot-
ings (independent and combined footing) have been described in this chapter.
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 243

4.2 SLAB BASE

The slab base as shown in Fig. 4.1 consists of cleat angles and base plate. The
column end is faced for bearing over the whole area.
The gussets (gusset plates and gusset angles) are not provided with the column
with the slab bases. The sufficient fastenings are used to retain the parts securely
in plate and to resist all moments and forces, other than the direct compression.
The forces and moments arising during transit, unloading and erection are also
considered.

C o lu m n

C le a t
S lab a ng le
b ase

P re ssu re distrib ution slab b a se

Fig. 4.1 Slab base

When the slab alone distributes the load uniformly the minimum thickness
of a rectangular slab is derived as below :
The column is carrying an axial load P. Consider the load distributed over
area h × w and under the slab; over the L × D as shown in Fig. 4.2.
Let t = Thickness of the slab
w = Pressure or loading on the underside of the base
a = Greater projection beyond column
b = Lesser projection beyond column
σbs = Allowable bending stress in the slab bases for all steels, it shall
be assumed as 185 N/mm2 (MPa)
Consider a strip of unit width.
Along the xx-axis,
⎛ wa2 ⎞
Mxx = ⎜ ⎟
⎝ 2 ⎠
Along the yy-axis,
⎛ wb2 ⎞
Myy = ⎜ ⎟
⎝ 2 ⎠
244 DESIGN OF STEEL STRUCTURES

S lab b ase (P la n)

Fig. 4.2

1
If Poisson ratio is adopted as 4 , the effective moment for width D

w ⎛ 2 b2 ⎞
= ⎜a − ⎟
2⎝ 4 ⎠
Effective moment for width L

w ⎛ 2 a2 ⎞
=⎜b − ⎟
2⎝ 4 ⎠
A is the greater projection from the column. Effective moment for width D is
more. Moment of resistance of the slab base of unit width
1
M.R. = ⎛⎜ × 1 × t 2 × σbs ⎞⎟
⎝6 ⎠

⎛1 2 ⎞ w ⎛ 2 b2 ⎞
⎜ × 1 × t × σbs ⎟ = ⎜a − ⎟
⎝6 ⎠ 2⎝ 4 ⎠
Thickness of the slab base
1/2
⎡ 3w ⎛ 2 b2 ⎞ ⎤
t = ⎢ ⎜ a − ⎟⎥ (as per IS : 800–1984)
⎣⎢ σbs ⎝ 4 ⎠ ⎥⎦
...(4.1)
For solid round steel column, where the load is distributed over the whole
area, the minimum thickness of square base as per IS : 800–1984 is given by
1/2
⎡ 90W ⎛ B ⎞ ⎤
t = 10 ⎢ ⎜ ⎟⎥ ...(4.2)
⎣16σbs ⎝ B − d0 ⎠ ⎦
where t = Thickness of plate in mm
W = Total axial load in kN
B = Length of the side of base of cap in mm
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 245

d0 = Diameter of the reduced end (if any) of the column in mm

σbs = Allowable bending stress in steel
σbs is adopted as 185 N/mm2 (MPa).
The allowable intensity of pressure on concrete may be assumed as 4 N/mm2.
When the slab does not distribute the load uniformly or where the slab is not
rectangular, separate calculation shall be made to show that stresses are within
the specified limits.
When the load on the cap or under the base is not uniformly distributed or
where end of the column shaft is not machined with the cap or base, or where
the cap or base is not square in plan, the calculations are made on the allowable
stress of 185 N/mm2 (MPa).
The cap or base plate shall not be less than 150 (d0 + 75) mm in length or
diameter.
The area of the shoulder (the annular bearing area) shall be sufficient to
limit the stress in bearing, for the whole of the load communicated to the slab to
the maximum value 0.75fy and resistance to any bending communicated to the
shaft by the slab shall be taken as assisted by bearing pressures developed
against the reduced end of the shaft in conjunction with the shoulder.
The bases for bearing upon concrete or masonry need not be machined on the
underside provided the reduced end of the shaft terminate short of the surface
of the slab, and in all cases the area of the reduced end shall be neglected in
calculating the bearing pressure from the base.
In cases where the cap or base is fillet welded direct to the end of the column
without boring and shouldering, the contact surfaces shall be machined to give
a perfect bearing and the welding shall be sufficient to transit the forces speci-
fied above. Where the full length T-butt welds are provided no machining of
contact surfaces shall be required.
Example 4.1 A column section HB 250, @ 0.510 kN/m carries an axial load of
600 kN. Design a slab base for the column. The allowable bearing pressure on
concrete is 4 N/mm2. The allowable bending stress in the slab base is 185 N/
mm2 (MPa).
Solution
Design : Step 1 : Area of slab base required
Axial load of column = 600 kN.
It is assumed uniformly distributed under the slab
Area of the slab base required
⎛ 600 × 1000 ⎞
⎜ ⎟ = 15 ×104 mm2
⎝ 4 ⎠
The length and width of slab base are proportioned so that projections on
either side beyond the column are approximately equal.
Size of column section HB 250, @ 0.510 kN/mm
= 250 mm × 25 mm
Area of slab base = (250 + 2a) (250 + 2b ) mm2
246 DESIGN OF STEEL STRUCTURES

Step 2 : Projections of base plate

Let the projections a and b be equal
Area of slab
(250 + 2a)2 = 15 × 104, ∴ a = 68.45 mm
Provide projections a = b = 70 mm
Provide slab base = (250 + 2 × 70) (250 + 2 × 70)
= 390 mm × 390 mm
Area of slab base provided
(390 × 390) = 152100 mm2
Intensity of pressure from concrete under the slab

⎛ 600 × 1000 ⎞
w = ⎜ = 3.945 N/mm2
⎝ 1521000 ⎟⎠
Step 3: Thickness of slab base
1/ 2
⎡ 3 × 3.95 ⎛ 2 702 ⎞ ⎤
Thickness of slab base = ⎢ ⎜ 70 − ⎟⎥ = 15.34 mm
⎣ 185 ⎝ 4 ⎠⎦
Provide 16 mm thick slab base. The fastenings are provided to keep the column
in position.
Example 4.1 (a) A column section SC 250, @ 85.6 carries an axial load of
600 kN. Design a slab base for the column. The allowable bearing pressure on
concrete is 4 N/mm2. The allowable bending stress in slab base is 185 N/mm2.
(MPa)
Solution
Design :
Step 1 : Area of slab base required
Axial load of column is 600 kN. It is assumed uniformly distributed under
the slab.
Area of slab base required

⎛ 600 × 1000 ⎞
⎜⎝ ⎟⎠ = 15 × 104 mm2
4
The length and width of slab base are proportioned so that the projections on
either side beyond the column are approximately equal.
Size of column section SC 250, @ 85.6 kg/m
= (250 mm × 250 mm)
Area of slab base
= (250 + 2a ) (250 + 2b) mm2
Step 2: Projections of base plate
Let the projections a and b be equal
Area of slab
(250 + 2a)2 = 15 × 104, ∴ a = 68.45 mm
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 247

Provide projections a = b = 70 mm
Provide slab base = (250 + 2 × 70) (250 + 2 × 70)
= (390 mm × 390 mm)
Area of slab base provided
(390 × 390) = 152100 mm2

⎛ 600 × 1000 ⎞
w = ⎜⎝ ⎟ = 3.945 N/mm2
152100 ⎠
Step 3: Thickness of slab base
Thickness of slab base
1/2
⎡ 3 × 3.95 ⎛ 702 ⎞ ⎤
= ⎢ ⎜⎝ 702 − ⎟ = 15.34 mm
⎣ 185 4 ⎠ ⎥⎦
Provide 16 mm thick slab base. The fastenings are provided to keep the col-
umn in position.
The length and width of slab base are proportioned so that the projections on
either side beyond the column are approximately equal.

4.3 GUSSETED BASE

In the gusseted base, gusset plates and gusset angles are used on either side of
the column. In addition to the gusset plates, cleat angles are used to connect the
column with the slab base. The gusseted base plate has been shown in Fig. 4.3.
The columns with gusseted bases, the gusset plates, angle cleats, stiffeners,
etc., in combination with the bearing area of the shaft should be sufficient to
take the loads, bending moments and reactions to the base plate without
exceeding the specified stresses. All the bearing surfaces are machined to ensure
perfect contact.
When the ends of the columns and gusset plates are not faced for complete
bearing, the fastenings connecting them to the base plate shall be sufficient to
transmit all the forces to which the base plate is subjected.
Following are the usual steps for the design of gusseted base.
Step 1. The load carried by the column, and intensity of bearing pressure on
concrete or masonry block are known. The area required by the gusseted base
plate is computed by dividing the load by allowable intensity of pressure in
concrete.
Step 2. The gusset materials used in the gusseted base are assumed. The
thickness of gusset plate is assumed as 16 mm. The size of gusset angle is assumed
such that its vertical leg can accommodate two rivets in one vertical line and
corresponding to that leg length, other leg is assumed in which one rivet can be
provided. The thickness of cleat angle is kept approximately equal to the
thickness of gusset plate (i.e., ISA 150 mm × 115 mm × 15 mm is assumed to be
used as cleat angle).
248 DESIGN OF STEEL STRUCTURES

P = 2 80 0 kN

G usset G usset
a ng le p late
1 50 m m × 16 m m
11 5 m m × th ick
15 m m ×
B a se
+ + p late
+ +

22 m m
(B – 2 a )
B = 61 0 m m
(a )
P = 2 80 0 kN

+ + + +
+ + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + + 1 50 m m
22 m m
11 50 m m
(b )
A n ch o r A n ch o r
b olt 2 2 m m φ R ivets b olt

G usset an gle
IS A 1 50 x 11 5 m m IS H B 3 00 ,
x 15 m m + + @ 0 .63 0 kN /m
(IS A 1 50 11 5,
@ 0 .23 8 kN /m ) + +
4 00 m m X 2 0 m m

A n ch o r 2 2 m m φ R ivets A n ch o r
b olt b olt
(c) G u sseted b ase

Fig. 4.3 Gusseted base plate

Step 3. The depth of column section, thickness of gusset plate and length of
leg of angle being known, the width of gusset plate for these distances as shown
in Fig. 4.3 (a) is known.
Step 4. The length of gusset plate is computed by dividing area required for
gusset plate by the width of the gusset plate.
Step 5. The fastenings of the gusseted base and the end of column faced for
complete bearing over whole area shall be sufficient to take loads, bending
moments and reactions to the base plate without exceeding specified stresses.
(As advised in IS : 800–1956 fastenings having flush ends may be designed for
50 percent axial load and for other forces).
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 249

Where ends of the column section and gusset plates are faced for complete
bearing, the fastenings connecting them to the base plate shall be sufficient to
transmit all the forces to which the base plate is subjected.
A 22 mm nominal diameter is assumed for rivets to be used and rivet value
and thus number of rivets to transmit the forces as specified above are com-
puted. The horizontal shear component is equal to the vertical shear. The num-
ber of rivets connecting the gusset plate and gusset angle is adopted equal to
the number of rivets connecting the gusset plate and the column as shown in
Fig. 4.3 (b).
Step 6. The thickness of base plate is computed by equating moment of resis-
tance with the moment due to intensity of pressure at the underside of base
plate at two critical sections X and Y shown in Fig. 4.3 (a).
The thickness whichever is more and safe at both the sections is adopted.
The thickness computed is reduced by the thickness of the gusset angle in case
the thickness at XX section governs the value. It is usual to make the base
plates at least as thick as the gussets or as angles.
In the above computation, base plate is assumed to have supports only at
gusset plates, and bending of plate takes place in one plane only. But the bend-
ing of central portion actually occurs in two planes mutually at right angles. It
is also having some support from web of column. The thickness of the plate is
made one-half of that calculated. The computation of thickness with the said
assumption is on safer side. The thickness of base plate determined at the sec-
tion XX or that determined at the section YY whichever is maximum is adopted
as the thickness of base plate.
When the end of the column is connected directly to the base plate by means
of full penetration butt welds, the connection should be deemed to transmit to
the base all the forces and moments to which the column is subjected.
Example 4.2 A column section HB 300, @ 0.630 kN/m with one cover plate
400 mm × 20 mm on either side is carrying an axial load of 2800 kN inclusive of
self-weight of base and column. Design a gusseted base. The allowable bending
pressure in concrete is 4 N/mm2. The allowable bending stress in base plate is
185 N/mm2 (MPa).
Solution
Design :
Step 1: Area of base plate required
Axial load on the column = 2800 kN
⎛ 2800 ⎞
Area required for gusseted base = ⎜ × 1000 ⎟ = 70 ×104 mm2
⎝ 4 ⎠
Assume gusset plate 16 mm thick and gusset angles ISA 150 mm × 115 mm ×
15 mm. (ISA 150115, @ 0295 kN/m)
Step 2: Width of gusseted base (in the direction parallel to the web of
column section)
= [35.00 + 2 × 2.00 + 2 × 1.60 + 2 × 11.5] × 10 = 602 mm
Adopt width of gusseted base = 610 mm
250 DESIGN OF STEEL STRUCTURES

Step 3 : Length of gusseted base

⎛ 70 × 104 ⎞
= ⎜ ⎟ =1147.54 mm
⎝ 610 ⎠
Adopt length of gusseted base = 1150 mm
Area of gusseted base provided = 610 × 1150 = 701500 mm2
Intensity of pressure between plate and concrete
⎛ 2800 × 1000 ⎞
⎜ ⎟ = 3.99 N/mm2
⎝ 701500 ⎠
Assume ends of column section, cover plates, gusset plates and angles faced
for complete bearing over whole area.
Load transmitted through fastening of both faces of column section is 50 per
cent of axial load.
Step 4 : Connection of gusset angle and plate
= 1400 kN
Use 22 mm diameter power driven rivets.
The column section and cover plates attached with the flanges act as one
unit. The rivets connecting gusset plate and column section are in single shear.
Strength of rivets in single shear

⎛ π (23.5)2 × 100 ⎞
⎜ × ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of rivet in bearing
⎛ 23.5 × 16 × 300 ⎞
⎜ ⎟ = 112.8 kN
⎝ 1000 ⎠
Rivet value R = 43.35 kN
Number of rivets required
⎛ 1400 ⎞
⎜ ⎟ = 32.29
⎝ 43.35 ⎠
Provide 40 rivets on both faces
i.e., 20 rivets on each face in four vertical rows.
The number of rivets that connecting gusset angles with gusset plate is equal
to that connecting gusset plate with column.
Step 5 : Thickness of base plate.
Consider cantilever portion upto critical section XX. The length of cantilever
portion
x = (115 – 15 ) + 4 = 104 mm
Consider width of 10 mm
Bending moment at the critical section XX
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 251

⎛ 3.99 × 1042 ⎞
⎜ ⎟ = 216 mm-kN
⎝ 2 × 1000 ⎠
Let t 1be the thickness of base plate inclusive of thickness of cleat angle.
Moment of resistance of strip
⎛1 2 ⎞
M.R. = ⎜ × 10 × t1 × 185 ⎟ mm-N
⎝6 ⎠

⎛ 216 × 6 × 1000 ⎞
t12 = ⎜ ⎟ ∴ t 1 = 26.47 mm
⎝ 185 × 10 ⎠
Thickness of base plate
= (26.47–15) = 11.47 mm
Consider central portion of unit width strip.
Length = (B – 2a) = (610 – 2 × 115 – 2 × 4) = 372 mm
Bending moment at critical section YY

⎛ w( B − 2a )2 wa2 ⎞ ⎛ 3.99 × 3722 3.99 × (104 − 15)2 ⎞

⎜ − ⎟ = ⎜ − ⎟
⎝ 8 2 ⎠ ⎝ 8 2 ⎠
= 407.5 mm-kN
Let t 2 be the thickness of plate
Moment of resistance of strip
⎛1 2 ⎞
M.R. = ⎜ × 10 × t2 × 185 ⎟ N-mm
⎝6 ⎠

⎛ 407.5 × 6 × 1000 ⎞
t22 = ⎜ ⎟ mm2 ∴ t = 36.35 mm
⎝ 185 × 10 ⎠
As central portion of plate in fact has bending in two directions and also
supported at web; thickness of plate is made about one half
1
∴ Thickness of plate = × 36.35 = 18.177 mm >16 mm
2
Adopt 22 mm plate thickness (manufactured). The base plate should be at
least as thick as gusset angles. Provide 2 ISA 150 mm × 115 mm × 15 mm (2 ISA
150115, @ 0.295 kN/m)(cleat angles) and connect with the web of column. The
design of gusseted base has been shown in Fig. 4.3.

4.4 COLUMN BASES SUBJECTED TO MOMENT

A column is carrying an axial load ‘P’ and moment ‘M’ as shown in Fig. 4.4.
The intensity of bearing pressure between column base and concrete (footing)
because of axial load ‘P’ has been assumed to be uniform as shown in Fig. 4.4.
The intensity of pressure is given by
P = (P/BL)
252 DESIGN OF STEEL STRUCTURES

where, B = Width of the column base

L = Length of the column base
The stress developed between column base and concrete (footing) because of
moment ‘M’ as shown in Fig. 4.4 (b) is given by
M
fb = ±
Z

⎛1 ⎞
where, Z = Section modulus of column base, ⎜ BL2 ⎟
⎝6 ⎠
Depending upon the direction of moment, stress due to moment acts upward
on one side and downward on other side. The combined stress due to axial load
‘P’ and moment ‘M’ is given by
⎛ P 6M ⎞
(P ± fb) = ⎜ ± ⎟
⎝ BL BL2 ⎠
The combined stress varies linearly. When intensity of bearing pressure due
to axial load is greater than the stress developed due to moment, i.e.,
p > fb
⎛ P 6M ⎞ ⎛M L⎞
or ⎜ > ⎟ ⎜ < ⎟
⎝ BL BL2 ⎠ ⎝P 6⎠
then, the intensity of bearing pressure between column base and concrete is
compressive throughout the length of column base as shown in Fig. 4.4 (c).
When the intensity of bearing pressure due to axial load is equal to the stress
developed due to moment, i.e.,

⎛ P 6M ⎞
p = fb, ⎜ = ⎟
⎝ BL BL2 ⎠
⎛ PL ⎞ ⎛M L⎞
or M = ⎜ ⎟ or ⎜ = ⎟
⎝ 6 ⎠ ⎝P 6⎠
then, intensity of bearing pressure between column base and concrete is com-
pressive, throughout the length of column base, and it is zero at one edge, as
shown in Fig. 4.4(d).
When the intensity of bearing pressure due to axial load is less than the
stress developed due to moment, i.e.,

⎛ P 6M ⎞
p < fb, ⎜ < ⎟
⎝ BL BL2 ⎠
⎛ PL ⎞ ⎛M L⎞
or M > ⎜ ⎟ or ⎜ > ⎟
⎝ 6 ⎠ ⎝P 6⎠
then, intensity of bearing pressure between column base and concrete is maxi-
mum at one edge and it is zero at a point (o) within the base. The portion ‘mo’
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 253

will not remain in contact with the base and the intensity of bearing pressure,
will, therefore, be zero for all points between ‘m’ and ‘o’. The intensity of bearing
pressure acting upward (compressive) has been shown in Fig. 4.4 (e). The re-
L
sultant passes at a distance greater than from centre of base (i.e., outside
6
the middle third portion). It acts at the C.G. of the triangle at a distance y from
the edge, and is equal to and coincides with the line of action of total vertical
load acting downward. Length of base of triangle is 3y. The ordinate of triangle
(p1) represents maximum intensity of bearing pressure. This should not exceed
allowable bearing pressure of concrete (footing).
Area of base giving upward pressure
= 3y × B
Total upward pressure (resultant)
1
R = × p1 × 3 y × B (But R = P)
2
1
∴ P = × p1 × 3 y × B ...(4.3)
2
and taking moment about centre of the base
⎛L ⎞
R ⎜ − y ⎟ = M, ⎛⎜ L − y ⎞⎟ = M
⎝2 ⎠ ⎝2 ⎠ R
⎛L M⎞ ⎛L M ⎞
⎜ −y= ⎟, y = ⎜ − ⎟ ...(4.4)
⎝2 P ⎠ ⎝2 P ⎠
Length of column base is determined by trial and error and the distance y can
be computed from Eq. 4.4. Width of column base can be found from Eq. 4.3.
Thickness of the base plate is found by equating the moment of resistance of
base plate with the moment due to intensity of bearing pressure.
The above analysis is for the base plate fixed to the column, and anchor bolts
do not resist any moment. Anchor bolts are provided to keep the column base in
position. When anchor bolts also resist overturning moment, then resultant of
upward pressure
R = P + P1 ...(4.5)
where, P1 = Tension induced in the anchor bolt in one side due to overturn-
ing moment.
Taking moment about C.G. of triangle in Fig. 4.4 (e),
⎛L ⎞
P1 (a − y ) + P ⎜ − y ⎟ = M ...(4.6)
⎝2 ⎠
where a = distance of anchor bolt from edge under maximum pressure.
The position of anchor bolt is known and hence distance a is also known.
Value of P1 is substituted from Eq. 4.5 ; and the distance y is found.
254 DESIGN OF STEEL STRUCTURES

P
M

C /L C /L
A n ch o r bo lt A n ch o r bo lt

P P
BL BL

6M (a )
BL2
P – 6M 6M
B L B L2 (b ) BL2

P 6M
+
BL BL2
(c)

2P
(d ) BL
3y

P1
(e )
Y

Fig. 4.4 Distribution of stresses in column carrying load and moment

The length and width of column base are determined by trial and error method.
Example 4.3 Design a suitable base for column carrying an axial load of 300
kN and a moment on 40000 mm-kN in the plane of the web. The action of column is
HB 300, @ 0.588 kN/m. The allowable bearing pressure on the footing is 4 N/mm2.
Solution
Design :
Step 1: Force/Load transmitted
Provide a gusseted base plate
Provide gusset plate 16 mm thick, and gusset angles 2 ISA 150 mm × 115 mm
× 15 mm, (2ISA 150115, @ 0.295 kN/m)
Assume, moment is transferred to the gusset plate by flanges only.
⎛ Moment ⎞
Force due to moment = ⎜
⎝ Lever arm between flanges ⎟⎠
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 255

⎛ 40,000 ⎞
F = ⎜ kN
⎝ (300 − 10.6) ⎟⎠
10.6 mm is thickness of flange of HB 300, @ 0.588 kN/mm
⎛ 40,000 ⎞
F = ⎜ = 138.217 kN
⎝ 298.4 ⎟⎠
Assume that the gusset plates, gusset angles and end of the column are hav-
ing flush end, i.e., having complete bearing over whole area.
50 per cent of axial load is transferred through bearing and 50 per cent of
axial load is transferred through both the faces.
Load transferred through one face
⎛1 ⎞
⎜ × 300 + 138.217 ⎟ = 213.217 kN
⎝4 ⎠
Step 2 : Connection of cleat angle and base plate
Use 22 mm power driven rivets. Strength of rivet in angle shear

⎛ π (23.5)2 × 100 ⎞
⎜ × ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of rivet in bearing
⎛ 23.5 × 16 × 300 ⎞
⎜ ⎟ = 112.8 kN
⎝ 1000 ⎠
Rivet value = 43.35 kN
Number of rivets required
⎛ 213.217 ⎞
⎜ ⎟ = 4.918
⎝ 43.35 ⎠
Provide 6 rivets on each face in two vertical rows, and 4 rivets are provided in
horizontal on either side of centre line to connect the cleat angle and the gusset
plate.
Step 3: Base plate. When anchor bolts are not resisting the over-turning
moment.
Length of base (dimensions parallel to web)
L = [(3000 + 2 × 16 + 2 × 11.5)]– ×10
= 562 mm
Provide length of the base plate equal to 570 mm
From Eq. 4.4
⎛L ⎞ M ⎛ 570 ⎞ ⎛ 4,00,00 ⎞
⎜ − y⎟ = ,⎜ − y⎟ = ⎜ ⎟
⎝2 ⎠ P ⎝ 2 ⎠ ⎝ 300 ⎠
∴ y = 151.667 mm, 3y = 455 mm
Allowable bearing pressure on footing is 4 N/mm2
256 DESIGN OF STEEL STRUCTURES

Then from Eq. (4.3)

⎛1 ⎞
⎜ p1 × 3 y × B ⎟ = P
⎝2 ⎠

⎛1 ⎞
or ⎜ × 4 × 455 × B ⎟ = 300 × 1000, ∴ B = 329.67 mm
⎝2 ⎠
Provide width of base plate B as 400 mm
Actual maximum intensity of pressure p′1 , then

1
p′1 × 455 × 400 = 3.0 × 1000
2
p1 = 3.296 N/mm2
Length of cantilever portion
(115 – 15) + 4 = 104 mm
The pressure intensity P1´ reduces to zero in a length 455.1 mm from right
side. The intensity of pressure at a section XX at a distance 104 mm
⎛ 455 – 105 ⎞
p = 3.296 × ⎜ ⎟ = 2.543 N/mm2.
⎝ 455 ⎠
The bending moment at section XX, in cantilever portion of 1 mm strip
1 1 2
= × 2.543 × 1042 + × 104 (3.296 – 2.543) × × 104
2 2 3
= (13752.544 – 2714.816) = 16467.36 mm-N
Moment of resistance of 1 mm strip
⎛1 2 ⎞
⎜ × t × 0.66 × 260 ⎟ = 16467.36, ∴ t = 23.995 mm
⎝6 ⎠
Step 4: The thickness of base plate :
= t–thickness of gusset angle
= (23.995 –15) = 8.995 mm < thickness of gusset plate

5 70 m m
4 55 m m
x
3 51 m m
1 51 m m
C .G . + p 2 .54 3 3 .29 6
N /m m 2 N /m m 2
x

Fig. 4.5 Stress distribution diagram

Provide 570 mm × 400 mm × 16 mm thick base plate.

It is fixed to the column. Provide two anchor bolts of 20 mm diameter each.
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 257

Example 4.4 Design available base in Ex. 4.3, when anchor bolts also resist
overturning moment.
Solution
Design :
Step 1 : Forces transmitted
Design of riveted connections remains same. Provide a gusseted base plate of
570 mm × 400 mm size. Provide two anchor bolts in gusset angle on each side at
a distance 54 mm from the edge of the base plate upto centre line of anchor bolt.
From Eq. (4.5)
(P + P1) = R
⎛1 ⎞
or (300000 + P1) = ⎜ p1 × 3 y × B ⎟
⎝2 ⎠

⎛1 ⎞
or P1 = ⎜ × 4 × 3 y × 400 − 300 × 1000 ⎟
⎝2 ⎠
or P1 = (2400 y – 300000) N
From Eq. (4.6)
⎛L ⎞
P1(a – y) + P ⎜ − y ⎟ = M
⎝2 ⎠
Substitute values of P1, p, a L, and M
⎛ 570 ⎞
or (2400y – 300000) (570 – 54 – y) + 30,0000 ⎜ − y⎟
⎝ 2 ⎠
2.4y2 – 1238 y +1093 × 100 = 0

1238 ± (12382 − 4 × 2.4 × 1093 × 100)1 / 2

y= = 113.02 mm
2 × 2.4
P1 = (2400 × 113.02 – 300000) = –28752 N = –28.752 kN
This shows, that the vertical load is larger than that due to overturning.
Two bolts each of 20 mm diameter is provided to maintain the position of the
column.

4.5 INDEPENDENT COLUMN FOOTING

The independent footing is used to distribute the load over large area from
individual column. When the bearing capacity of the soil is not poor, then con-
crete block is used as column footing. When the bearing of the soil is poor, then
grillage footing is used.

4.5.1 Grillage Footing

The steel grillage is a frame-work of rolled steel beams forming foundation in
the soil of poor bearing capacity. In the grillage footing steel beams are used in
258 DESIGN OF STEEL STRUCTURES

one tier or more depending upon spread of the base. In case, two or more tiers
are used, each tier is placed perpendicular to the tier above it. The beams are
embedded in cement concrete and are protected from corrosion. As far as possible,
square base is provided in the independent footing.
The column load P is distributed through a column base of length a to the
grillage beam of length L as shown in Fig. 4.6.

P
2 a
4

L L L
4 4 2
L

Fig. 4.6 Shear force and bending moment in steel grillage beams

It is assumed that each tier distribute the load uniformly to the lower tier
P a
and intensity of bearing pressure is also uniform. A load at a distance ,
2 4
P L
from the centre line acts downward, and resisting force at a distance
2 4
from centre line acts upward.
The maximum bending moment occurs at the centre of beam. Maximum bend-
ing moment
P ⎛L a⎞ P (
M = ⎜ − ⎟= L − a) ... (4.7)
2 ⎝4 4⎠ 8
The maximum shear force occurs at the edge of the base plate.
Maximum shear force,
P ⎛L−a⎞ 1 P (
F = ⎜ ⎟= L − a) ...(4.8)
L⎝ 2 ⎠ 2L
The grillage beams are kept unpainted and solidly encased in ordinary dense
concrete with 100 mm aggregate.
The pipe separators or equivalent are used to keep the grillage beams prop-
erly spaced and the beams are spaced apart so that the distance between edges
of adjacent flanges is not less than 75 mm.
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 259

The thickness of the concrete cover on the upper flange, at the ends, and the
outer edges of the sides of the outermost beam is not less than 100 mm.
The concrete is properly compacted solid around all beams.
The distribution of load, soil pressure, shear force and bending moment dia-
grams are shown in Fig 4.7.

1 (L – a ) 1 (L – a )
2 2

P re ssu re (P / L )
(a ) D istribu tion o f loa d a n d so il pre ssu re

P ⎛L −a ⎞ P (
F F = ⎜ ⎟= L −a)
L ⎝ 2 ⎠ 2L
F
E1 B1
A1 D1
F O

(b ) S h ea r force diagram

A B

MD = ME
P ⎛L a ⎞
2 M m ax = ⎜ − ⎟
1 P ⎛L −a ⎞ D E 2 ⎝4 4 ⎠
M = ⋅ ⎜ ⎟
2 L⎝ 2 ⎠ P
M m ax = (L − a )
1 P( 8
L −a)
2
MD = ⋅ C M m ax = MD + M2
8 L
1 P 2 1P (
M1 = ⋅ a (c) B e nd in g m om ent d ia g ram M2 = a L −a)
8 L ste el grillag e fo un da tio n 8 L

Fig. 4.7

Example. 4.5 Design two tier grillage foundation to carry a stanchion,

designed to carry 1200 kN. The base plate of the stanchion, is 600 mm square in
size. The bearing pressure of earth is limited to a value of 200 kN per square
metre. Take the allowable bending stress and the allowable average shear stress
as per IS : 800–1984.
260 DESIGN OF STEEL STRUCTURES

Solution
Design :
Step 1 : Area of footing required
Direct axial load on stanchion P = 1200 kN
Add 10 per cent of direct axial load for self-weight of the grillage foundation.
Design load to calculate bearing area for grillage foundation
= (1200 + 120) = 1320 kN
Bearing area required for grillage base
⎛ 1320 ⎞
⎜ ⎟ = 6.6 mm2 = 6.6 × 103 × 103 mm2
⎝ 20 ⎠
Area provided = 2600 mm × 2600 mm = 6760000 mm2
Step 2: Design of beams in upper tier
The beams are designed for given column load, P.
The length of base plate of stanchion, a is 600 mm
The length of beam in upper tier, L is 2600 mm
The maximum bending moment occurs at the centre of the beam
P(
M = L − a)
8
Maximum bending moment,
⎡1200(2600 − 600 ⎤
M = ⎢ ⎥⎦ = 300000 kN-mm
⎣ 8
Allowable bending stress is 0.66fy
(0.66 × 260) = 171.6 N/mm2
Section modulus required
M ⎛ 300000 × 1000 ⎞
Z= = ⎜ ⎟ 1748 × 1000 mm3
fb ⎝ 171.6 ⎠
The number of grillage beams in upper tier depends upon minimum clear
spacing of 75 mm between the flanges, which are accommodated in the width of
the base plate given.
Provide three MB 350, @ 0524 kN/m
Width of flange = 140 mm .
Width of base plate = 60 mm
Clear spacing between flanges of beams is 90 mm
Section modulus provided
(3 × 778.9 × 1000) = 2336.7 × 1000 mm3
Step 3: Check for shear force
The maximum shear force in the beam is at the edge of the base plate.
Maximum shear force at the edge
1P( 1
= L − a)
2L n
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 261

where, n is the number of beams in upper tier

Maximum shear force
⎡ 1 1200 1⎤
⎢⎣ 2 × 2600 (2600 − 600) × 3 ⎥⎦ = 153.846 kN
From steel tables thickness of web of beam section
tw = 8.1 mm
The depth of beam section h is 350 mm
Average shear stress
⎛ 153.846 × 1000 ⎞
fs = ⎜ ⎟ = 54.267 N/mm2
⎝ 8.1 × 350 ⎠
It is less than allowable average shear stress
(0.4 × fy) = (0.4 × 260) = 104 N/mm2 as per IS : 800–1984.
Step 4: Check for crippling stress
The crippling of web may occur in beams in upper tier (Discussed in Chapter 6)

⎛ P ⎞
Crippling stress = ⎜ ⎟
⎝ ( a + 2 3h )t
2 w ⋅ n ⎠
where, a = length of base plate
n = number of beams
h2 = depth of the root of fillet from top of flange.
From steel section table, h2 = 31 mm
⎛ 1200 × 1000 ⎞ 2
Crippling stress = ⎜ ⎟ N/mm
⎝ 10 × (60 + 2 3 × 3.1) × 8.1 × 3 ⎠
= 69.727 N/mm2
It is less than allowable bearing stress
σp = 0.75 fy = (0.75 × 260) = 195 N/mm2
Step 5: Design of beam in lower tier
The beams are designed for given column load P.
Maximum bending moment
P ⎛ 1000 ⎞
M = (L – a) = ⎜ × (2600 − 600 ⎟ = 300000 mm-kN
8 ⎝ 8 ⎠
Section modulus required
⎛ 300000 ⎞
Z = ⎜ × 1000 ⎟ = 1748 ×1000 mm3
⎝ 171.6 ⎠
The number of beams depends upon length of beams in upper tier and mini-
mum clear spacing of 75 mm.
Provide 9 number of beams LB 250, @ 6 279 kN/m in lower tier.
Section modulus of one beam is 297.4 × 103 mm3
262 DESIGN OF STEEL STRUCTURES

Thickness of web = 8.2 mm

Width of flange = 125 mm
Section modulus provided
(297.4 × 103 × 9) = 2676.6 ×103 mm
Clear distance between flanges
1
× (2600 − 8 × 125) = 208 mm > 75 mm
8

P = 12 00 kN

IS LB 25 0 @ 27 9 kN /m
1 00 m m 1 00 m m

10 0 m m

S e pa ra tors

6 00 m m

A xia l
60 0 m m

colum n
lo ad
P = 1 20 0 kN

Fig. 4.8 Steel grillage foundation

Step 6: Check for shear force: Shear force

1P( 1 ⎛ 1 1200 (2600 − 600) ⎞
L − a) × = ⎜ × × ⎟ = 51.28 kN
2L n ⎝ 2 2600 9 ⎠

⎛ 51.28 × 1000 ⎞
Average shear stress = ⎜ ⎟ = 25 N/mm2
⎝ 8.2 × 250 ⎠
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 263

It is less than the allowable average shear stress 104 N/mm2 as mentioned
above.
The web crippling does not occur in lower tier.
Design of grillage foundation has been shown in Fig. 4.8.
Example 4.6 Design a foundation footing for a column carrying 1530 kN of
vertical load along with a moment in the same vertical plane equal to 200 m-kN.
The bearing capacity of the soil is 200 kN/m2. The width of foundation should
not exceed 2 metres.
Solution
Design :
Step 1: Area of footing required
Provide a steel grillage footing 2.0 m wide × 5.0 m long.
Area of foundation footing
A = 2.0 × 5.0 = 10.0 m2
Step 2: Check for combined stress
The vertical load of the column, P is 1530 kN
Add 10 per cent for self-weight for the foundation.
Total vertical load = (1530 +153) = 1683 kN
Moment M = 200 metre-kN
Intensity of bearing pressure on soil
⎛P M ⎞ ⎛ 1683 200 ⎞
fb = ⎜ ± ⎟=⎜ ± ⎟
⎝A Z ⎠ ⎜ 10 1
× 2 × 52 ⎟
⎝ 6 ⎠
= (168.3 ± 24) kN/m2
Maximum intensity of bearing pressure on soil
= 192.3 kN/m2 < Bearing capacity of the soil
Minimum intensity of bearing pressure on the soil
= (168.3 – 24.0) = 143.3 kN/m2 compressive
Provide steel beams in two tiers.
Step 3: Design of beams in upper tier
The beams are designed for given load and moment.
Provide 3 steel beams in upper tier as shown in Fig. 4.8
Length of beams L = 2 m
⎛ 1530 ⎞
Vertical load on each beam is ⎜ = 510 kN
⎝ 3 ⎟⎠
In addition to this, two outer beams take a additional load because of mo-
ment. Assume size of base plate of 800 mm × 600 mm and arrange the beams as
shown in Fig. 4.9.
Additional load on each beam
⎛ 200 × 1000 ⎞
⎜ ⎟ = 333.3 kN
⎝ 600 ⎠
264 DESIGN OF STEEL STRUCTURES

Maximum vertical load on outer beam

(510 + 333.3) = 843.3 kN
Maximum bending moment
W (L – a ) ⎛ 843 ⋅ 3 (2000 – 600 ) ⎞
M= = ⎜ ⎟ = 147577 mm-kN
8 ⎝ 8 ⎠
Section modulus required for one beam
M ⎛ 147577 × 1000 ⎞
Z= = ⎜ ⎟ = 860 × 103 mm3
σb ⎝ 171.6 ⎠
Provide MB 400, @ 0.616 kN/m
Section modulus = 1022.9 × 104 mm3
Thickness of web = 8.9 mm

8 00 m m

3 00 m m 3 00 m m

Fig. 4.9 Steel grillage beams in upper tier

Step 4: Check for shear force

Maximum shear force
W (L − a) ⎛ 843.3 × (2000 − 600) ⎞
= ⎜ ⎟ kN
2L ⎝ 2 × 2000 ⎠

⎛ 843.3 × 1400 ⎞
⎜ ⎟ = 295.1 kN
⎝ 4000 ⎠

⎛ 295.1 × 1400 ⎞
Average shear stress = ⎜ ⎟ = 82.893 N/mm2
⎝ 400 × 8.9 ⎠
It is less than allowable average shear stress 104 N/mm2
= (0.4 × 260) N/mm2 as per IS : 800–1984.
Step 5 : Design of beam in lower tier
The beams are designed for given load and moment
Provide 8 beams in lower tier
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 265

Length of beams = 5 m
Moment due to vertical load
⎛ 1530(5000 − 800) ⎞ ⎛ 1530 × 4200 ⎞
⎜ ⎟ = ⎜ ⎟ = 8032.5 × 100 mm-kN
⎝ 8 ⎠ ⎝ 8 ⎠
Total moment = 100 × (8032.5 + 2000) = 10032.5 × 100 mm-kN
Moment shared by each beam
⎛ 10032.5 × 100 ⎞
⎜ ⎟ = 1254.06 ×100 mm-kN
⎝ 8 ⎠
Section modulus required for one beam
⎛ 1254 ⋅ 06 × 1000 × 100 ⎞
⎜ ⎟ = 730.804 × 133 mm3
⎝ 171 ⋅ 6 ⎠
Provide LB 350, @ 0.495 kN/m
Section modulus = 751.9 × 103 mm3.

4.6 COMBINED COLUMN FOOTING

The combined footing is used to distribute loads from two or more columns over
large area. The steel grillage footing used as combined footing to support two
column loads, has been described here. Two column loads may be equal and
may be unequal.

4.6.1 Steel Grillage Footing to Support two Equal Column Loads

A steel grillage footing supporting two equal column loads has been shown in
plan in Fig. 4.10.
The size of the column bases, and centre to centre distance between columns
are known. The rectangular shape for the base of the footing is used. The projec-
tions of beams in lower tier are kept such that bending moments under the
columns and centre of span of beams are approximately equal. It gives economi-
cal design. Length of base of footing is thus known. The width of base of footing
is computed from area of base required to spread the two column loads. In case

P P

Fig. 4.10 Steel grillage footing supporting two equal loads

266 DESIGN OF STEEL STRUCTURES

width of base of footing is restricted comparative more projections are provided

in beams in lower tier, and beams are designed for maximum bending moment.

4.6.2 Steel Grillage Footing to Support two Unequal Column Loads

A steel grillage footing supporting two unequal column loads P1 and P2 has been
shown in plan in Fig. 4.11.
The line of action of the resultant of the two unequal column loads is made to
coincide with the centre of gravity of base of the footing. The trapezoidal shape
is used for the base of the footing. The centre to centre distance between the two
columns K is known. The projections of beams in lower tier m and n are assumed,
such that, as far as possible, bending moments under the columns and between
two columns are approximately equal. However, it is by trial to fix up projections
most satisfactory. In case, there are restrictions for projections, the projections
are fixed accordingly. The length of the base of the footing L is known, knowing
the centre to centre distance of columns K and fixing the projections of beams in
lower tier. The area of base of footing, A, required to spread the two column
loads is known, knowing the allowable bearing pressure of soil. The widths of
sides of trapezoidal base, b and c as shown in Fig. 4.11 are determined as given
below.

a1 a a2
b a2 d
P2 CG C e ntre P2 a2

m k n
L

Fig. 4.11 Steel grillage footing supporting two unequal loads

Let d be the distance between centre of the base and centre of gravity of base
(C.G.). The distance of C.G. of base from large width b of base is given by

x = L ⎛⎜1 + c ⎞⎟
3⎝ b+c⎠

L L⎛ c ⎞
then d = – ⎜1 + ⎟
2 3⎝ b+c⎠
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 267

⎛ c ⎞
or 6d = 3L – 2L ⎜1 + ⎟
⎝ b+c⎠

6d ⎛ 2c ⎞ 6d ⎛ 2c ⎞
or = ⎜3 – 2 – ⎟, = ⎜1 – ⎟
L ⎝ b+c⎠ L ⎝ b+c⎠

6d ⎛ b + c – 2c ⎞ 6d (b – c )
or = ⎜ ⎟, =
L ⎝ b + c ⎠ L (b + c )

A
Multiply both sides by
L
6d A (b – c ) A
× = × , (but (b + c) . L = 2A)
L L (b + c ) L
6d A (b – c ) A
× =
L L 2A
6A ⋅ d (b – c )
or = ...(i)
L2 2
A (b + c )
and = ...(ii)
L 2
⎛ A 6A ⋅ d ⎞
then from (i) and (ii) b = ⎜ + ⎟ ...(4.9)
⎝L L2 ⎠
⎛ A 6A ⋅ d ⎞
and c = ⎜ – ⎟ ...(4.10)
⎝L L2 ⎠
A ⎛ 6⋅d ⎞
or b and c are given by ⎜1 + ⎟
L ⎝ L ⎠
where plus sign gives b, and minus sign gives c.
In case the combined grillage footing is of two tiers, then maximum bending
moments for beams in upper tier under the heavy column load P1 and the other
column load P2 are given below.
Maximum bending moment for beams in upper tier under heavy column load:

P1 ( L1 – a1 )
M1 =
8
where, L1 = length of the beam under the heavy column
a1 = Width of the base plate under the heavy column along the length
of beams
Maximum bending moment for beams in upper tier under other column load:

P2 ( L2 – a2 )
M2 =
8
268 DESIGN OF STEEL STRUCTURES

where, L2 = Length of the beam under the other column

a2 = Width of the base plate under the other column along the length
of beams
The beams are designed for maximum bending moments and checked for
shear.
For the design of beams in lower tier maximum bending moments are deter-
mined at three sections as below :
Let x be the distance from large end of trapezoidal base upto corresponding
section as shown in Fig. 4.12. Then, bending moment due to intensity of pres-
sure of soil w acting upward is
From Fig. 4.12,
y ⎛b – c ⎞ 1
= ⎜ ⎟⋅
(L – x ) ⎝ 2 ⎠ L
⎛ L – x ⎞⎛ b – c ⎞
y = ⎜ ⎟⎜ ⎟
⎝ L ⎠⎝ 2 ⎠

⎛ b – c⎞
2y = (L – x ) ⎜
⎝ L ⎟⎠

A
C
1 (b – c)
2 y
E
x ( L –x )

F
1 (b – c)
2
D

Fig. 4.12

Therefore,
⎛b – c ⎞
CD = (c + 2y) = C + (L – x) ⎜ ⎟
⎝ L ⎠
⎛ b − c⎞
∴ CD = b – x ⎜ ... (ii)
⎝ L ⎟⎠
Area of trapezium ABCD
2
⎛ AB + CD ⎞
= ⎜ ⎟ ⋅x
⎝ 2 ⎠
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 269

1⎡ ⎛ b – c ⎞⎤
= ⎢b+b– x⎜ ⎟ ⋅x
2⎣ ⎝ L ⎠ ⎥⎦

x⎡ x ⎤
= 2b – (b – c )⎥ ⋅ x ...(iii)
2 ⎢⎣ L ⎦
Total load = (intensity) × area
x⎡ x ⎤
= w⋅ 2b – (b – c )⎥ ...(iv)
2 ⎢⎣ L ⎦
Distance of centroid from right side

x⎡ x ⎤
2b + (b – c )⎥
3 ⎢⎣ L ⎦
= ...(v)
⎡ x( )⎤
⎢⎣b + b – L b – c ⎥⎦

Moment = Load × Distance of centroid

⎡ x( ⎤
⎢⎣3b – b – c )⎥
wx ⎡ x( ) ⎤ x L ⎦
2b – b – c ×
2 ⎢⎣ L ⎥⎦ 3 ⎡ x( ⎤
⎢⎣2b – b – c )⎥
L ⎦

wx 2 ⎡ x ⎛ b – c ⎞⎤
= . ⎢b – ⎜ ⎟ ...(vi)
2 ⎣ 3 ⎝ L ⎠ ⎥⎦

wx 2 ⎡ x (b – c ) ⎤
= ⎢b – 3 L ⎥ ...(4.11a)
2 ⎣ ⎦
Bending moment under the heavy column load
2
⎡ ⎛ a ⎞⎤
2 ⎢ x − ⎜m − 1 ⎟ ⎥
wx ⎡ –
x ⎛ b c⎞ ⎤ P ⎝ 2 ⎠⎦
M1′ = ⎢b– ⎜ ⎟ ⎥ − 1×⎣ ...(4.11b)
2 ⎣ 3 ⎝ L ⎠ ⎦ a1 3
Bending moment between the two columns

wx 2 ⎡ x ⎛ b − c⎞ ⎤
M 2′ =
2 ⎢b − 3 ⎜⎝ L ⎟⎠ ⎥ − P1 ( x − m ) ...(4.12)
⎣ ⎦
Bending moment under the other column load

⎡ ⎛ a2 ⎞ ⎤
2 ⎢ x − ⎜⎝ m + K − ⎟⎥
M 3′ = wx ⎡b − x ⎛⎜ b − c ⎞⎟ ⎤ − P ( x − m ) P2 ⎣ 2 ⎠⎦ ...(4.13)
⎢
2 ⎣ 3⎝ L ⎠⎦ ⎥ 1
a2 2
270 DESIGN OF STEEL STRUCTURES

M1′ , M 2′ and M 3′ are differentiated with respect to x, and are equated to zero
separately. The values of x are found for maximum bending moments in the
above expressions. The maximum bending moments from the above expres-
sions are determined by substituting values of x. The maximum of maximum
bending moment is found and it is used for the design of beams in lower tier.
In the trapezoidal shape of the base of the footing, the beams in lower tier are
arranged at close spacing at smaller end than that at the other end.
The rectangular shape is also adopted for combined footing to support two
unequal column loads. The line of action of the resultant of two column loads is
made to coincide with centre of gravity of the base. The cantilever projection
near the other column. The length of the base is kept such that bending mo-
ments under the heavy column load and between two columns (though these
are of opposite sign) are approximately equal.

4.7 APPROXIMATE METHOD OF DESIGN OF COMBINED

FOOTING
The exact method of design of combined footing for two columns has been
discussed in Sec. 4.5. The approximate method of design of combined footings is
discussed here. The first point will be kept in view in dealing the combined
grillage footing is that the line of action of the resultant of the two column loads
is made to coincide with the centre of gravity of the base of footing.

P2

(a ) Fo rces actin g o n fo oting

q o .n

q o .m
Sh ea r force
C h an ge s sign
(b ) Sh ea r force diag ram

+ 1 q 2 1 q 2 +
2 o .m 2 o .m
(c) B e nd in g m om en t d ia g ram

Fig. 4.13 Design of combined footing by approximate method

 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 271

The size of column bases and centre to centre distance between columns are
found by the least projection required from a consideration of the connection of
a column base to the grillage. The beams in the lower tier in the grillage may be
considered to be inverted beams loaded with the uniformly distributed load
equal to the intensity of pressure under the grillage footing and supported at
two points (i.e., at the columns). Thus, the beams have overhanging portions at
both the ends. The beams are subjected to the cantilever moment at each end
and the span moment in between the two columns. The cantilever moments
and the span moments are of opposite sign. The cantilever moments are assumed
to be maximum at the centre line of the columns. The maximum span moment
is assumed to occur at the point, where the shear force changes sign. The
maximum cantilever moments and the maximum span moments are determined.
The beams are designed for the absolute maximum bending moment. The shear
force and the bending moment diagrams are drawn. The beams are suitably
spaced such that the clear distance between the flanges is not less than 75 mm.
The beams designed are checked for shear, web buckling and web crippling.
Example 4.7 Design a combined foundation for two columns carrying an
axial load of 1000 kN and 1500 kN respectively and placed 4 metre apart centre
to centre. The foundation can bear 200 kN/m2 of stress safely.
Solution
Design :
Step 1 : Area of base required
Provide a steel grillage for combined foundation
Load from column 1,
P1 = l500 kN
Load from column 2,
P2 = 1000 kN
Total load P = 2500 kN
Maximum allowable soil pressure
= 200 kN/m2
⎛ 2500 ⎞
Area of base required = ⎜ ⎟ = 12.5 m2
⎝ 200 ⎠
Step 2 : Length and width of grillage base
The distance of line of action of resultant of two columns loads from 1500 kN
load
⎛ 1000 × 4 ⎞
x1 = ⎜ ⎟ = 1.60 m
⎝ 2500 ⎠
Provide rectangular base of the foundation and cantilever projection of 1.72
m near column 1.
Centre of gravity of base is kept coinciding with the line of action of resultant
of column loads.
272 DESIGN OF STEEL STRUCTURES

Length of column base

2(1.60 + 1.72) = (2 × 3.32)
= 6.64 m
Width of grillage base
⎛ 12 ⋅ 10 ⎞
⎜ ⎟ = 1.88 m
⎝ 6 ⋅ 64 ⎠
Provide grillage beams in two tiers.
Step 3: Design of beam in upper tier
Length of beams = 1.88 m
= 1880 mm
Assume size of base plate under column 1 as 650 mm × 650 mm
Maximum bending moment for beams under column 1
1500 (
1880 – 650 ) = 2305 × 100 mm-kN
8
Total section modulus required
⎛ 2305 × 1000 × 100 ⎞
Z = ⎜ ⎟
⎝ 171 ⋅ 6 ⎠
= 1343.24 × 1000 m3
Provide 3 LB 300, @ 0.377 kN/m.Total section modulus provided
(3 × 488.9× 103) = 1466.7 × 103 mm3
Width of flange of LB 300, @ 0.377 kN/m
= 150 mm
Clear distance between flanges
= 100 mm >75 mm
Assume size of base plate under column 2
= 600 mm × 600 mm
Maximum bending moment for beams under column 2
1000 (
1880 – 600 ) = 1538 × 100 mm-kN
8
Total section modulus required
⎛ 1538 × 1000 × 100 ⎞
⎜ ⎟ = 896.27 × 103 mm3
⎝ 171 ⋅ 6 ⎠
Provide 3 LB 250, @ 0.279 kN/m
Width of flange = 125 mm
Clear distance between flanges
= 112.5 mm
Total section modulus provided
3 × 297.4 × 103 = 898.2 × 103 mm3
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 273

The beams under both the sections are checked for shear and web crippling.
Step 4: Check for S.F.
S.F. for beams under column 1
⎛ 1500 (1880 – 650 ) ⎞⎟ = 163.8 kN
⎜
⎝ 2 × 1880 × 3 ⎠
From steel section tables for LB 300, @ 0.377 kN/mm2
h = 300 mm,
tw = 9.4 mm,
h2 = 27.45 mm
⎛ 163 ⋅ 8 × 1000 ⎞
Shear stress, τvf = ⎜ ⎟ N/mm2
⎝ 300 × 9 ⋅ 4 ⎠
58.085 N/mm2 < (0.4 × 260 = 104 N/mm2)
S.F. for beams under column 2
⎛ 1000 (1880 – 600 ) ⎞⎟ = 113.4 kN
⎜
⎝ 2 × 1880 × 3 ⎠
From steel tables for LB 250, @ 0.279 kN/m
h = 250 mm,
tw = 6.1mm,
h2 = 23.7 mm
⎛ 113 ⋅ 4 × 1000 ⎞
Shear stress, τvf = ⎜ ⎟ N/mm2
⎝ 250 × 6 ⋅ 1 ⎠
= 74.36 N/mm2 < 104 N/mm2. Safe.
Step 5 : Check for web crippling
Beams under column 1
⎛ 1500 × 1000 ⎞
Bearing stress, σp = ⎜ ⎟ N/mm
2
⎝ (650 + 2 3 × 27 ⋅ 45 ) × 9 ⋅ 4 × 3 ⎠
71.389 N/mm2 < (0.75 × 260 = 195 N/mm2). Safe.
Beams under column 2

⎛ 1500 × 1000 ⎞
Bearing stress, σp = ⎜ ⎟ N/mm
2
( )
⎝ 650 + 2 3 × 23 ⋅ 7 × 6 ⋅ 1 × 3 ⎠
80.113 N/mm2 < 195 N/mm2. Safe.
Step 6 : Design of beams in lower tier
Intensity of soil pressure acting upward
= 200 kN/m2
The column loads acts downward as uniformly distributed load through the
column bases.
274 DESIGN OF STEEL STRUCTURES

The beams in lower tier act as inverted overhanging beams. Let x be the
distance from left end of base. The bending moment under column 1 (see
Fig.4.14)
4m 0 .42 m
1 00
0 .65 6 0 .6 m
3 .37 5 m mm
4m
1 00 m m 1 00 m m
P 1 = 1 50 0 kN P 2 = 1 00 0 kN

x (4 – x )
1 00
1 .72 m 1 .60 m 2 .40 m mm
1 00 m m 1 00 m m
3 .32 m 3 .32 m
(a ) C o m b in ed colum n fo otin g in plan
(1 50 0/0.6 5) kN -m x4 (1 00 0/0.6 ) kN/m
x1 x2
A B D F J

x 1 (2 00 X 1 .8 8) kN /m x2 x4
(b ) L oa ds o n foo tin g
A 1 C 1 = 1 .67 2 m A 1 G 1 = 5 .82 0 m
A 1 E 1 = 3 .99 0 m
D1 E1 G 1 H1 J1
A1 B1 C1 F1

(4 .0 = – 0 .3 25 – 0.3) 0 .6 m 0 .42 m
0 .65 0
m 3 .37 5 m

B 1 C 1 = 0 .27 7 m D 1 E 1 = 2 .04 5 m F 1 G 1 = 0 .40 0

C 1 D 1 = 0 .37 3 m E 1 F 1 = 1 .33 0 m G 1 H 1 = 0 .20 0

(c) S h ea r force diag ram

4 20 kN -m

A2 B2 C2 D2 F 2 G 2H 2 J2

E2
9 7 kN -m
4 38 .8 kN -m
(d ) B e nd in g mo m en t d ia gra m
d esig n of co m bine d foo tin g
Fig 4.14

⎛ 200 × 1 ⋅ 88 × x 2 1500 ( x – 1 ⋅ 395 )2 ⎞

M1´ = ⎜ – ⎟ kN-m
⎝ 2 0 ⋅ 65 2 ⎠
Differentiate w.r.t. x and equate to zero
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 275

⎛ dM1 ⎞
⎜ ⎟ = 0 x = 1.672 m
⎝ dx ⎠
Substitute the value of x
⎛ 200 × 1 ⋅ 88 × 1 ⋅ 6722 1500 (1 ⋅ 672 – 1 ⋅ 395 )2 ⎞
M1 = ⎜ – ⎟ kN-m
⎝ 2 0 ⋅ 65 × 2 ⎠
M1 = 438.8 kN-m
The bending moment between two columns
⎛ 200 × 1 ⋅ 88 × x 2 ⎞
M2́ = ⎜ – 1500 ( x – 1 ⋅ 72 ) ⎟ kN-m
⎝ 2 ⎠
Differentiate w.r.t. x and equate to zero
⎛ dM 2´ ⎞
⎜ ⎟ = 0 ∴ x = 3.99 m
⎝ dx ⎠
Substitute the value of x
⎛ 200 × 1 ⋅ 88 × 3 ⋅ 992 ⎞
M2́ = ⎜ – 1500 (3 ⋅ 99 – 1 ⋅ 72 ) ⎟ kN-m
⎝ 2 ⎠
= 420 kN-m (negative)
Bending moment under column 2
⎛ 200 × 1 ⋅ 88 × x 2 2
1000 ( x − 5 ⋅ 42 ) ⎞
M3́ = ⎜ − 1500 ( x − 1.7 2 ) − ⎟⎠
⎝ 2 0 ⋅ 60 2
Differentiate w.r.t. x and equate to zero
⎛ dM 3´ ⎞
⎜ ⎟ = 0 ∴ x = 5.82 m
⎝ dx ⎠
The distribution of load, shear force diagram and bending moment diagram
are shown in Fig. 4.14
Substitute the value of x
⎛ 200 × 1 ⋅ 88 × 5 ⋅ 822 2
1000 (5 ⋅ 82 – 5 ⋅ 42 ) ⎞
M3́ = ⎜ ( )
– 1500 5 ⋅ 82 – 1 ⋅ 72 – ⎟
⎝ 2 0 ⋅ 60 2 ⎠
= 97 kN-m.
Alternatively,
Step 1: Shear force
The positions of maximum bending moments may be located by writing the
expressions for shear forces and equating them to zero. The shear force under
the column 1 (see. Fig. 4.11).
1500
F1 = 200 × 1.88x – ( x – 1 ⋅ 395 ) = 0, :. x = 1.672 m
0 ⋅ 65
The shear force between the two columns
276 DESIGN OF STEEL STRUCTURES

F2 = 200 × 1.88 x – 1500 = 0, ∴ x = 3.99 m

The shear force under the column 2
⎛ 1000 ( ⎞
F2 = ⎜ 200 × 1 ⋅ 88x – 1500 – x – 5 ⋅ 42 ) ⎟ = 0, ∴ x = 5.82 m
⎝ 0 ⋅ 60 ⎠
Step 2 : Bending moments
The values of bending moment at these locations have been found as above.
Maximum of maximum bending moments occur under column 1
M1´ = 438.8 kN-m
Step 3 : Selection of section
Total section modulus required
⎛ 438 ⋅ 8 × 1000 × 1000 ⎞
⎜ ⎟ = 2557.11 ×103 mm3
⎝ 171 ⋅ 6 ⎠
Provide 8 LB 275, @ 0.330 kN/m. Total section modulus provided
(8 × 392.4 × l03) = 1319.2 × l03 mm3
Width of flange of LB 275, @ 0.330 kN/m
= 125 mm
Clear distance between flanges
⎛ 1 ⎞(
⎜ ⎟ 1880 – 8 × 140 ) mm = 108.57 mm >75 mm
⎝7⎠
The web crippling does not occur in lower tier. The beams in lower tier may
check for shear and these would be found safe.
The design of combined footing for two columns is shown in Fig. 4.15.

P 1 = 1 50 0 kN P 2 = 1 00 0 kN

ISL B 30 0, @ 0.37 7 kN/m

1 00 m m 1 00 m m
6 50 m m sq ua re P 1 6 .64 m 6 00 m m sq ua re P 2
1 00 m m 1 00 m m
1 .8 8 m

1 .8 8 m

P1 P2

1 00 m m 1 .39 5 m 6 00 1 00 m m
1 .72 m 4 .0 m
6 .64 m

Fig. 4.15 Rectangular combined footing

 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 277

Example 4.8 Design a combined steel grillage foundation for two columns
carrying loads of 2250 kN and 1800 kN spaced 3 m apart. The available space
by the side of 1800 kN column is only 1400 mm. The maximum allowable soil
pressure is 195 kN/m2. Sketch the arrangement of upper and lower tiers with
concrete encasing the joints.
Solution
Design :
Step 1 : Area of base required
The steel grillage foundation is to be designed to support two column loads.
Column load P1 = 2250 kN
Column load P2 = 1800 kN
Total load = 4050 kN
Maximum allowable soil pressure
= 195 kN/m2
Bearing area of base required
⎛ 4050 ⎞
A = ⎜ ⎟ = 20.769 m2
⎝ 195 ⎠
A = (20.769 ×1000 ×1000) mm2
Provide bearing area of base
Step 2 : Length and width of base
A = (21 × 1000 × 1000) mm2
Distance between centre to centre of column, K is 3 m
Available space by the side of 1800 kN column load is 1.40 m
∴Cantilever projection, n as 1400 mm
Provide cantilever projection, m is 1100 mm
Length of column base
L = (1100 + 3000 + 1400) = 5500 mm
Provide trapezoidal base or combined grillage footing
Distance of line of action of resultant of two column loads from 2250 kN
⎛ 1800 × 3000 ⎞
x = ⎜ ⎟ = 1333 mm
⎝ (2250 + 1800 ) ⎠
The centre of gravity of base of footing coincides with the line of action of
resultant of two column loads.
Distance of C.G. of base from larger width of trapezoidal base
(1100 + 1333) = 2433 mm
Distance of centre of base
L ⎛ 5500 ⎞
= ⎜ ⎟ = 2750 mm
2 ⎝ 2 ⎠
Distance between centre of base and C.G. of base
d = (2750 – 2433) = 317 mm
278 DESIGN OF STEEL STRUCTURES

Larger width of trapezoidal base is given by

A ⎛ 6d ⎞ ⎛ 21000000 ⎞ ⎛ 6 × 317 ⎞
b= = ⎜1 + ⎟=⎜ ⎟ ⎜1 + ⎟
L ⎝ L ⎠ ⎝ 5500 ⎠ ⎝ 5500 ⎠
= 5140 mm
Smaller width of trapezoidal base is given by
A ⎛ 6d ⎞ ⎛ 21000000 ⎞⎛ 6 × 317 ⎞
c= = ⎜1 − ⎟=⎜ ⎟⎜1 − ⎟
L ⎝ L ⎠ ⎝ 5500 ⎠⎝ 5500 ⎠
= 2500 mm
Let y1 be the width of base at the centre of column supporting 2250 kN
⎛ K +n⎞
y1 = ⎜ c + (b − c ) ⎟
⎝ L ⎠

⎡ 3000 + 1400 ⎤
= ⎢2500 + (5140 − 2500) × ⎥⎦
⎣ 5500
= 4610 mm
Let y2 be the width of base at the centre of column supporting 1800 kN
⎛ n⎞
y2 = ⎜ c + (b − c ) ⎟
⎝ L⎠

⎡ 1400 ⎤
= ⎢2500 + (5140 − 2500) ×
⎣ 5500 ⎥⎦
= 3172 mm
Step 3 : Design of beams in upper tier
Beams under column supporting 2250 kN
Length of central beam, y1 is 460 mm
Assume size of base plate under 1250 kN column load as 800 mm × 800 mm.
∴ a1 = 800 mm
Maximum bending moment
P1
M1 = ( y1 − a1 ) = 2250 (4610 − 800) = 10.71 × 105 kN-mm
8 8
Permissible bending stress for beams embedded in concrete
σb = (066 × 260) = 171.6 N/mm2
Section modulus required
5 3
⎛ M ⎞ ⎛ 10.71 × 10 × 10 ⎞
Z1 = ⎜ 1 ⎟ = ⎜ ⎟
⎝ σb ⎠ ⎝ 171.6 ⎠
= 6241 × 103 mm3
From ISI Handbook No. 1, provide 3 LB 600, @ 0.995 kN/m
Section modulus of each beam section
= 2428.2 × 103 mm3
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 279

Width of flange of beam, bf is 210 mm

Total section modulus provided
3 × 2428.9 × 103 = 7286.7 × 103 mm3
Clear distance between flanges
1
(800 – 3 × 210) = 85 mm > 75 mm
2
Step 4 : Beams under column supporting 1800 kN
Length of central beam, y2 is 3172 mm
Assume size of base plate under column supporting 1800 kN as 800 mm ×
800 mm.
Maximum bending moment

P2 1800
M3 ( y2 − a2 ) = × (3170 − 800)
8 8
= 5.34 ×105 kN-mm
Section modulus required
5 3
⎛ M 2 ⎞ ⎛ 5.34 × 10 × 10 ⎞
Zy = ⎜ σ ⎟ = ⎜⎝ 171.6
⎟
⎠
⎝ b ⎠
= 3111.89 × l03 mm3
From ISI Handbook No. 1, provide 3 LB 450, @ 0653 kN/m
Section modulus of three beams section
(3 ×12238 × 103) mm3 = 3671.4 × 103 mm3
Width of flange of beam, fy is 170 mm
Clear distance between flanges
1
(800 – 3 ×170) = 145 mm > 75 mm
2
The beams under both the columns may be checked in shear and web crip-
pling and will be found safe.
Step 5 : Design of beam in lower tier
Bearing area of base provided
= 21000000 mm2
= 21 m2
Total load = 4050 kN
Intensity of soil pressure acting upward

⎛ 4050 ⎞
w = ⎜ ⎟ = 192.5 kN/m2
⎝ 21 ⎠
Let x be the distance from larger width of base.
280 DESIGN OF STEEL STRUCTURES

The column loads acts downward as uniformly distributed load through the
base plates. However, bending moment under column supporting 2250 kN load
2
wx 2 ⎡ x b – c ⎤ 2250 ( x – 0 ⋅ 7 )
M1′ = ⎢⎣b – 3 ⋅ L ⎥⎦ – 0 ⋅ 80
2 2
2
192 ⋅ 5 × x 2 ⎡ h 5 ⋅ 14 – 250 ⎤ 2250 ( x – 0 ⋅ 7 )
= ⎢⎣51 ⋅ 4 – 3 × –
2 5 ⋅ 50 ⎥⎦ 0 ⋅ 80 2

⎛ dM1´ ⎞
Equating ⎜ ⎟ = 0 for M1′ to be maximum, we get x = 1.0 m
⎝ dx ⎠
2
192 ⋅ 5 × 1 ⎡ 1 2 ⋅ 64 ⎤ 2250 (1 – 0 ⋅ 7 )
∴ M1′ = ⎢51 ⋅ 4 – × – ×
2 ⎣ 3 5 ⋅ 50 ⎥⎦ 0 ⋅ 8 2
= + 373.5 kN-m (Hogging)
Bending moment between two column loads

⎛ wx 2 ⎞ ⎡ x b−c⎤
M 2′ = ⎜
⎝ 2 ⎠
⎟ ⎢⎣b − 3 ⋅ L ⎥⎦ − 2250 (x − 1 ⋅ 10)

192 ⋅ 5 × x 2 ⎡ x 5.14 − 2.50 ⎤

= ⎢⎣5 ⋅ 14 − 3 × − 2250 ( x − 1 ⋅ 10 )
2 5 ⋅ 50 ⎥⎦

⎛ dM 2´ ⎞
Equating ⎜ ⎟ = 0 for M 2′ to be maximum, we get x = 2.60 m
⎝ dx ⎠

192 ⋅ 5 × 2 ⋅ 62 ⎡ 260 2 ⋅ 60 ⎤
M 2′ = (
⎢⎣5 ⋅ 14 – 3 × 5 ⋅ 60 ⎥⎦ – 2250 2 ⋅ 6 – 1 ⋅ 10
)
2
= –275 kN-m (Sagging)
Bending moment under column supporting 1800 kN
2
⎛ wx 2 ⎞ ( )
M 3′ = ⎜ ⎟
⎡ x b – c⎤ ( ) 1800 x – 3 ⋅ 70
⎢⎣b – 3 ⋅ L ⎥⎦ – 2250 x – 1 ⋅10 – 0 ⋅ 80
⎝ 2 ⎠ 2
2
192 ⋅ 5x 2 ⎡ x 5 ⋅ 14 – 2 ⋅ 50 ⎤ 1800 ( x – 3 ⋅ 70 )
⎢⎣5 ⋅ 14 – 3 × ⎥ – 2250 ( x – 1 ⋅ 10 ) –
2 5 ⋅ 50 ⎦ 0 ⋅ 80 2

⎛ dM 3´ ⎞
Equating ⎜ = 0 for M 3′ to be maximum, we get x = 4.15 m
⎝ dx ⎟⎠

192 × 41 ⋅ 52 ⎡ 4 ⋅ 15 2 ⋅ 64 ⎤
∴ M 3′ =
2 ⎢⎣5 ⋅ 14 – 3 × 5 ⋅ 50 ⎥⎦
 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 281

2
1800 (4 ⋅ 15 – 3 ⋅ 70 )
– 2250 (4 ⋅ 15 – 1 ⋅ 10) –
0 ⋅ 80 2
= +272. 5 kN-m (Hogging)
The distribution of load, shear force diagram, and bending moment diagram
are shown in Fig. 4.16.
The maximum of maximum bending moment is 275 kN-m (Sagging).

1 00 m m

P 1 = 2 25 0 kN
P 2 = 1 80 0 kN

1 .1
m
0.8m

B D C .G . O F H J

1 .33 3
m
2 .75 m 2 .75 m

(a ) Trap ezoida l foo ting (P lan )

kN -m
A2 B2 C2 D2 F2 G2 H2 J2
E2
3 73 .5 kN -m 2 75 kN -m
2 72 .5 kN -m
(b ) B e nd in g m o m en t d ia gra m
A 1 C 1 = 1.0 m A 1 G 1 = 4.15 m
A E = 2.60 m
C1 D1 1 1 E1 G H J
A1 B1 F1

0 .7m 0 .8m 0 .6 m
2 .3 m 1 .1 m
B 1 C 1 = 0 .30 0 m D 1 E 1 = 1 .10 m F 1 G 1 = 0 .35 0 m
C 1 D 1 = 0 .50 0 m E 1 F 1 = 1 .20 m G 1 H 1 = 0 .25 0 m
(c) S h ea r force diag ram co m b in ed (trad ezo da l) d ia gram

Fig. 4.16
282 DESIGN OF STEEL STRUCTURES

Step 6 : Selection of beam section

The section modulus required

⎛ 275 × 103 × 103 ⎞

Z = ⎜ ⎟
⎝ 171 ⋅ 6 ⎠
= 1602.56 × 103 mm3
Provide 10 ISLB 200, @ 0.198 kN/m. The section modulus of each beam
= 169.7 × 103 mm3
Width of flange of beam, by is 100 mm
Total section modulus provided
10 × l69.7 × l03 = 1697 × 103 mm3
Clearance between beams on larger width side
1
(5140 – 10 × 100) = 459 mm > 75 mm
9
Clearance between beams on smaller width side
1
= (2500 – 10 ×100)
9
= 166 mm > 75 mm
The beams may be checked in shear and web crippling and these will be
found safe. The complete design is shown in Fig, 4.17.

P 1 = 2 25 0 kN P 2 = 1 80 0 kN

IS LB 45 0 @ 0.65 3 kN /m
IS LB 60 0 @ 0.99 5 kN /m

1 00 m m 1 00 m m
1 00 m m
1 00 m m

800 mm 900 mm
sq u a re sq u a re
P1 P2

1 00 m m

1 00 m m m = 1.10 m 1 00 m m

Fig. 4.17 Trapezodial combined footing

 DESIGN OF COLUMN BASES AND COLUMN FOOTINGS 283

PROBLEMS
4.1. A steel column HB 300 @ 0.630 kN/m resting on a steel base plate is
supported on concrete foundation. The column carries a central load of
3000 kN. Design the base plate and gusseted connection. The safe bearing
capacity of concrete is 4 N/mm2.
4.2. Design the base plate and its anchor bolts for a 300 mm × 140 mm @ 0.442
kN/m I-scction steel stanchion under an axial load of 350 kN and a moment
of 4250 kN-mm in the plane of web.
4.3. Design a suitable riveted base plate connection for WB 200 column subjected
to an axial load for 300 kN and a moment of 450 kN-mm at the bottom in
the plane of web.
4.4. Design a suitable base for a column carrying an axial load of 240 kN and a
moment of 25000 kN-mm in the plane of the web. The section of the column
is HB 250, @ 0.547 kN/m.
4.5. Design a grillage foundation for a compound column consisting of HB 400,
@ 0.822 kN/m with flange plate 300 mm × 18 mm one on each flange and
carrying a load of 2300 kN. Draw a neat drawing.
4.6. Design a grillage foundation for a column carrying an axial load of 1500
kN inclusive of self-weight. The bearing capacity of soil is 150 kN/m2. The
column base plate resting on the grillage is 600 mm × 600 mm.
4.7. Design a two tier grillage foundation to carry a stanchion designed to carry
1040 kN. The base plate of the stanchion is 700 mm square. The bearing of
earth is limited to a value of 150 kN/m2.
4.8. A combined steel grillage foundation for two columns carrying loads of
1500 kN and 1000 kN spaced at 6100 mm centres is to be designed. The
available space by the side of 1000 kN column is only 460 mm.
The maximum soil pressure is limited to 110 kN/m2.
(a) Draw the bending moment and shear force diagrams for the upper
tier of the grillage foundation.
(b) Work out the sections for the upper and lower tiers of the grillage.
(c) Sketch the arrangement of upper and lower tiers with concrete encas-
ing.
4.9. A column is subjected to an axial load of 2000 kN. Design a grillage foun-
dation for the column assuming that the bearing capacity of the soil is 100
kN/m2. The size of the column base plate is 800 mm × 600 mm.
4.10. Design a combined grillage foundation for the two columns HB 450, @
0.925 kN/m and HB 400, @ 0.774 kN/m spaced at centre to centre distance
of 4 m. The columns carry axial loads of 2000 kN and 1500 kN respec-
tively. The safe bearing capacity of the soil is 150 kN/m2. Give neat sketches
to show design details.
4.11. A column made of an HB rolled steel section carries an axial load of 400
kN and a bending moment of 36000 kN-mm at the bottom end acting about
the major principal axis of the column section. The base plate is connected
to the column flange by two 150 mm × 75 mm × 12 mm (2 ISA 15075, @
0.201 kN/m) angles with 20 mm rivets, the 75 mm legs being outstanding
284 DESIGN OF STEEL STRUCTURES

from the column flanges. The end of the I-section is faced for complete
bearing on the base plate. Using the data given, find (a) the required hori-
zontal dimensions of the base plate, (b) the required thickness of the base
plate and (c) the number of 20 mm rivets necessary to connect each angle
to the column flange as governed by the 3600 kN-mm moment.
Properties of sections :
ISHB 250 column section : depth = 250 mm, flange width = 250 mm. flange
thickness = 9.7 mm, area = 6971 mm2.
Rivets : Nominal diameter 20 mm, diameter of rivet holes = 21 .5 mm.
Gauge distance of the rivet holes :
Column flange = 140 mm, 150 mm leg of angles distance from back to first
row = 55 mm, from first row to second row = 65 mm.
Adopt permissible stresses as per IS : 800–1984.
4.12. It is required to design a two tier grillage foundation for a steel column
carrying a load of 1000 kN. The dimensions of the base plate are about 800
mm × 850 mm. The footings is to be supported on concrete base which in
turn rests on solid rock. The allowable pressure on the concrete base is 4
N/mm2. Design the grillage using I-section girder built-up of suitable rect-
angular plates. Plates 12 mm and 20 mm thick and of any width and length
may be assumed to be available. Assume standard working stresses as
per IS : 800–1984.
Sketch the arrangement of the built-up beams forming the grillage.
4.13. Design a two tier grillage foundation footing for a HB 350 @ 0.674 kN/m
column supporting a load of 1000 kN. The footing is to be supported on soil
which has a bearing capacity of 100 kN/m2.
4.14. Design a grillage foundation for a column consisting of HB 450 section
with flange plate 400 mm × 40 mm each carrying an axial load of 4500 kN.
The concrete can take a safe bearing pressure of 4 N/mm2. Sketch the
arrangement.
4.15. A column consisting of HB 200, @ 0.373 kN/m with cover plates 300 mm ×
20 mm (one on each flange) is fixed at the base and free at the top. It
carries a load of P at an eccentricity of 200 mm from the centroidal axis of
member giving rise to bending moment about the major axis of the mem-
ber. If the height of column is 4 m, design a suitable attached base for the
column. Permissible bearing pressure on concrete is 4 N/mm2.
 Chapter
Design of Tension

5 Members

5.1 INTRODUCTION
A tension member is defined as a structural member subjected to tensile force in
the direction parallel to its longitudinal axis. A tension member is also called as
a tie member or simply a tie. The term tie is commonly used for tension mem-
bers in the roof trusses. When a tension member is subjected to axial tensile
force, then the distribution of stress over the cross-section is uniform. The com-
plete net area of the member is effectively used at the maximum permissible
uniform stress. Therefore, a tensile member subjected to axial tensile is said to
be efficient and economical member. In the building design, the interest in the
use of tension members is increasing. The tension members are used as hang-
ers for floors and cables for roof.
The various members used as tension members are not perfectly straight.
The initial crookedness and imperfections of the member result in the small
eccentricity. The distribution of stress in the eccentrically loaded tension mem-
ber is not uniform. The member is subjected to combined stress, (i.e., direct
tension and bending). Besides this, a tension member may be subjected to axial
tension and bending.

5.2 TYPE OF TENSION MEMBERS

The types of structure and method of connection decide the type of tension
member. The compact tension members are generally used. As far as possible, a
large portion of the tension member is connected directly to the gusset plates or
splice plates. This minimizes stress concentration in the members. The various
types of tension members used may be grouped into following four groups:
(i) Wires and cables
(ii) Rods and bars
286 DESIGN OF STEEL STRUCTURES

(iii) Single structural shapes and plates

(iv) Built-up members

5.2.1 Wires and Cables

The wire ropes are used for hoists, derricks, rigging slings, guy wires and hangers
for the suspension bridges. The untwisted parallel wires are used for the
manufacture of main cables of the suspension bridges. The wire ropes have the
advantages of flexibility and strength. The wire ropes require special fittings
for proper end connections. The wire ropes are used as guy wires or guy lines
with the steel towers. The wire ropes are given initial tension. The initial tension
prestresses the frame. The effectiveness of frame in resisting the external load
is increased when the frames are prestressed.

5.2.2 Rods and Cables

The square and round bars as shown in Fig. 5.1 (b) and (c) are quite often used for
the small tension members. The round bars are threaded at the ends and are
used with nuts. The area of cross-section at the ends and are used with nuts. The
area of cross-section at the root of the thread is the net area which resists

(a ) (f)

(b ) (g ) (k)

(c) (h ) (l)

(d ) (i)

(e ) (j) (m )

Fig. 5.1 Various forms of tension members

the tension. The round bars with threaded ends are used with pin-connections
at the ends with standard clevises. Sometimes, the bars are bent to form loops
 DESIGN OF TENSION MEMBERS 287

at the ends instead of threads. The round bars with clevises and loop rods are
made in two portions. These two portions are joined with turn buckles. The
tension members are tightened with these turn buckles. Such members are
commonly used for lateral bracing. The ends of rectangular bars or plates are
enlarged by forging and then, these are bored to form eye bars. The eye bars are
used with pin-connections. The rods and bars have disadvantages of inadequate
stiffness resulting in noticeable sag under the self-weight.

5.2.3 Single-structural Shapes and Plates

The single structural shapes, viz., angle sections and tee-sections as shown in
Fig. 5.1 (d) and (e) are used as tension members. The angle sections are consid-
erably more rigid than the wire ropes, rods and bars. When the length of ten-
sion member is too long, then, the single angle section is also too flexible. The
equal angle sections are commonly used when the angles are used as single
structural members. The single angle members are also used for bracing. These
members can resist small compression if reversal of stress takes place. The
single angle sections have the disadvantages of eccentricity in both the planes
in the riveted connections.
Sometimes, the single channel sections are also effectively employed as the
tension members. The channel sections have eccentricity only in one direction.
The single channel sections have high rigidity in the direction of web and low
rigidity in the direction of flange.
Occasionally, the I-sections are also used as tension members. I-sections have
more rigidity. The single I-sections are more economical than the built-up
sections.

5.2.4 Built-up Members

Two or more than two members are used to form built-up members. When the
single rolled steel sections cannot furnish the required area, then, the built-up
sections are used. The built-up sections may be made more rigid and more stiff
than the single structural shapes. The moment of inertia of the built-up sections
may be increased. The built-up sections can resist compression if the reversal of
stresses takes place.
The double angle sections of unequal legs shown in Fig. 5.1 (f) are exten-
sively used as tension members in the roof trusses. The angle sections are placed
back to back on two sides of a gusset plate. When both the angle sections are
attached on the same sides of the gusset, then, built-up section has eccentricity
in one plane. The built-up section is subjected to tension and bending simulta-
neously. The two angles sections may be arranged in the star shape (i.e., the
angles are placed diagonal opposite to each other with legs on outer sides). The
star shape angle sections may be connected by batten plates. The batten plates
are alternately placed in two perpendicular directions. The star arrangement
provides a symmetrical and concentric connection. Two angle sections as shown
in Fig. 5.1 (h) are used in the two plane trusses, where two parallel gussets are
288 DESIGN OF STEEL STRUCTURES

used at each connection. Two angle sections as shown in Fig. 5.1 (h) have the
advantage of adjusting the distance between them. Four angle sections as shown
in Fig. 5. 1 (i) are also used in the two plane trusses. The angles are connected to
two parallel gussets. Four angle sections connected by plates as shown in Fig.
5.1 (j) are used as tension members in the bridge truss girders.
A built-up section may be made of two channels placed back to back with a
gusset plate in between them. Such sections are used for medium loads in single
plane truss. In two plane trusses, two channels are arranged at a distance with
their flange turned inward. It simplifies the transverse connections. It also mini-
mizes lacing. The flanges of two channels are occasionally turned outward in
order to have greater lateral rigidity.
The heavy built-up tension members in the bridge truss girders are made of
angles and plates. Such members can resist compression if the reversal of
stresses occurs.

5.3 NET SECTIONAL AREA

The net sectional area is used for tension member. The net sectional area of a
tension member is the gross-sectional area of the member less the maximum
deduction for holes. The projected area for each hole is a rectangle, one side of
which is the thickness of the member and the other side is the diameter of the
hole.
In chain riveting, the net sectional area of a member at a section is the gross-
sectional area minus the area to be obtained from the product of thickness of
the plate and the sum of the diameters of all rivet and bolt holes in that section.
In zig-zag riveting, the rivets used to connect the tension member are stag-
gered as shown in Fig. 5.2.
The failure of plate may occur on a zig-zag section passing through all the
three rivets instead of a straight right angle section passing through first two

b
g

Fig. 5.2

rivets. In case all the three rivets are in one line along a right angle section
then, the deduction is made for all the three rivet holes in determining the net
area. When one or more than one rivet hole is off the line, the failure of the plate
depends upon the staggered pitch, p the gauge distance g, and the diameter of
 DESIGN OF TENSION MEMBERS 289

the hole d. When the gauge distance g, is large as compared to the staggered
pitch, p then, the failure may occur in zig-zag line. When the gauge distance in
small, as compared to the staggered pitch, p then, the failure may occur along a
straight right angle section passing through rivet holes in Fig. 5.2. When the
staggered pitch, p and the gauge, g are fixed, then, the zig-zag failure become
more likely as the diameter of hole, d increases. From the theoretical investi-
gations and laboratory tests, it is found that the net area of the plate for Fig. 5.2
may be found as below:

⎡ ⎛ p2 ⎞ ⎤
Anet = ⎢ Agross – 2dt – d ⋅ t ⎜1 – 2 × ⎥ ...(i)
⎣ ⎝ 4 g ⋅ d ⎟⎠ ⎦

⎡ ⎛ p2 ⎞ ⎤
or Anet = ⎢bt – 2dt – d ⋅ t ⎜1 – 2 × ⎟⎥ ...(ii)
⎢⎣ ⎝ 4 g ⋅ d ⎠ ⎥⎦

⎡ ⎛ p2 ⎞ ⎤
or Anet = ⎢b – ⎜ 3d – 2 × ⎟⎥ ...(iii)
⎣⎢ ⎝ 4 g ⋅ d ⎠ ⎦⎥

⎡ ⎛ p2 ⎞ ⎤
In general, Anet = ⎢b – ⎜ nd – n´ ⎟⎥ ...(5.1 a)
⎢⎣ ⎝ 4 g ⎠ ⎥⎦
where b = width of plate
t = thickness of plate
p = staggered pitch i.e., the distance between any two consecutive
rivets in zig-zag chain measured parallel to the direction of stress
in the number as shown in Fig. 5.2.
g = gauge space, that is, the distance between the same two con-
secutive rivets in a zig-zag chain at right angles to the direction
of stress in the number as shown in Fig. 5.2. It is transverse
spacing between parallel adjacent lines of fasteners.
n = number of rivet holes in zig-zag line.
n´ = number of gauge spaces.
d = gross diameter of rivet.
Net sectional area of plate as shown in Fig. 5.2

⎡ ⎛ p2 ⎞ ⎤
Anet = ⎢b – ⎜ 3d – 2 ⎟⎥ ⋅ t ...(iv)
⎣⎢ ⎝ 4 g ⋅ d ⎠ ⎦⎥
Equation 5.1 is known as Steinman’s formula.
In case, the rivets are connected in different staggered pitches, P1,p2, etc.
and the gauge distances g1, g2, then, the deductions shall be as below :

⎡ ⎛ p2 p2⎞ ⎤
Deduction = ⎢(Sum of section areas of holes) − ⎜ 1 + 2 ⎟ t ⎥
⎢⎣ ⎝ 4 g1 4 g2 ⎠ ⎥⎦
290 DESIGN OF STEEL STRUCTURES

In case, the staggered rivets are used in the two legs of an angle section,
then, the net area can also be determined by Eq. 5.1. Then, the riveted connec-
tions are in two different planes, viz., in non-planer sections, such as angles
with the holes in both the legs. The gross width of angle is the sum of lengths of
two legs less the thickness of angle. The gauge distance for the rivets in two legs
is equal to the sum of gauge distance for the rivets in each legs less the thick-
ness. It is to note that the standard rivet gauge distances for legs of angles are
given is ISI Handbook No. 1.
The effective net area at approximately the root of the thread of a threaded
tension member may be found from the following equation. This is as per AISC
specification.
2
π⎛ 0 ⋅ 9743 ⎞
Ae.net = ⎜D − ⎟ mm2 ...(5.1 b)
4⎝ n ⎠
where D = nominal outside diameter of threads in mm
n = number of threads per mm

5.4 NET EFFECTIVE SECTION FOR ANGLES AND TEES IN

TENSION
When an angle section is used as tension member, it is connected with the gusset
plate by one leg. When a tee-section is used as tension member, it is connected
with the gusset plate by the flange. In such cases the rivets connecting
tension member and the gusset plate do not lie on the line of action of the load.

O utstan ding
le g

C o nn ecte d leg

G usset plate

Fig. 5.3

This results in an eccentricity moment on the connection. The distribution of

the stress becomes non-uniform because of eccentric connection. The net sectional
area of an angle or a tee member is reduced to account for this non-uniform
distribution of stress. The reduced net sectional area of such member is called
net effective area. The net effective section for angles and tees in tension are
adopted as specified in IS: 800–1984 as below :
(i) In the case of single angle in tension connected by one leg only, the net
effective section of the angle is taken as
An = (A1+ A2 . k) ...(5.2)
 DESIGN OF TENSION MEMBERS 291

where, An = net effective sectional area

A1 = effective cross-sectional area of the connected leg
A2 = gross cross-sectional area of the unconnected (outstanding) leg,
and
⎛ 3 A1 ⎞
k = ⎜ ⎟ ...(5.3)
⎝ 3 A1 + A2 ⎠
where the leg angles are used, the effective sectional area of the whole of the
angle member shall be considered.
(ii) In the case of a pair of angle back to back (or a single tee) in tension
connected by only one leg of each angle (or by the flange of a tee) to the same
side of a gusset, the net effective area is taken as
An = ( A1+A2 . k) ...(5.4)
where, A1 = effective sectional area of the connected legs (or flange of the tee)
A2 = gross cross-sectional area of the unconnected (outstanding) legs
(or web of the tee), and
G usset plate

C o nn ecte d leg

O utstan ding
le g

Fig. 5.4 Pair of angles (back to back) connected by one leg of each
angle to same side of gusset plate

⎛ 5 A1 ⎞
k = ⎜ ⎟ ...(5.5)
⎝ 5 A1 + A2 ⎠
The tacking rivets are used with the appropriate spacing to connect the angles
along their length.
(iii) For double angles or tees carrying direct tension placed back to back
and connected to each side of a gusset as shown in Fig. 5.5 or to each side of a
rolled section, sectional area to determine the mean tensile stress shall be the
full gross area less the deductions for holes, provided the tacking rivets are
used with the appropriate spacing to connect the angles along their length.

Fig. 5.5 Two angles connected to each side of gusste plate

292 DESIGN OF STEEL STRUCTURES

The area of the leg shall be taken as the product of the thickness by the
length from the outer corner minus half thickness, and the area of the leg of a
tee as the product of the thickness by the depth minus the thickness of the
table.
When the angles are back to back and the tacking rivets are not used to
connect the angles along their length, then, each angle shall be designed as
single angle connected through one leg only as described above.
When two tee sections are placed back to back but the tacking rivets are not
used to connect them along the length, then each tee-section shall be designed
as single tee-section connected to one side of a gusset as specified above.

5. 5 ALLOWABLE STRESS IN AXIAL TENSION

The permissible stress in axial tension, σat on the net effective area of the section
shall not exceed.
σat = 0.6 fy
where, fy is the minimum yield stress of steel is N/mm2 (MPa).

5. 6 STRENGTH OF A MEMBER IN AXIAL TENSION

The tensile strength of a member subjected to axial tension Pt is found as below:
Pt = (An × σat)
where, An = Net sectional effective area of the member
σat = Allowable stress in axial tension.
Example 5.1 In a roof truss, a diagonal consists of an ISA 60 mm × 60 mm ×
8 mm and it is connected to gusset plate by one leg only by 18 mm diameter rivets
in one chain line along the length of the member. Determine tensile strength of
the member.
Solution
Step 1 : Net area of section
ISA 60 mm × 60 mm × 8 mm (ISA 6060,-@ 0.070 kN/m) is used as a tension
member.
Diameter for rivet hole = (18 +1.5) =19.5 mm
The rivets are used in one chain line along the length of the member. Net
sectional area of connected leg
⎛ 1 ⎞
A1 = ⎜ 60 − 19 ⋅ 5 − × 8⎟ × 8 = 292 mm2
⎝ 2 ⎠
Area of outstanding (unconnected) leg,
⎛ 8⎞
A2= ⎜ 60 − ⎟ × 8 = 448 mm2
⎝ 2⎠

⎛ 3 A1 ⎞ ⎛ 3 × 292 ⎞
k = ⎜ ⎟ = ⎜⎝ 3 × 292 + 448 ⎟⎠ = 0.6616
⎝ 3 A1 + A2 ⎠
Net effective sectional area
An = (292 + 0.6616 × 448) = 588.41 mm2
 DESIGN OF TENSION MEMBERS 293

Step 2 : Tensile strength of section

It is assumed that value of yield stress for steel shall be 260 N/mm2. The
permissible stress in axial tension
σat = 0.6 × 260 = 156 N/mm2
Tensile strength of the angle section
⎛ 588 ⋅ 41 × 156 ⎞
P1 = ⎜ ⎟ = 91.792 kN.
⎝ 1000 ⎠
Example 5.2 A double angle tie ISA 150 mm × 75 mm × 10 mm (ISA
15075, @ 0.169 kN/m) (short legs back to back) of a roof truss is connected to
the same side of a gusset, with rivets 18 mm in diameter, such that each angle
is reduced in section by one rivet hole only. Determine the tensile strength of
the member. Stitch rivets have been provided at suitable spacing.
Solution
Step 1 : Net area of section
Tie member consists of 2 ISA 150 mm × 75 mm × 10 mm (2 ISA 15075, @
0.169 kN/m)
Diameter of rivet hole
(18 +1.5) = 19.5 mm
The long legs of angles have been connected to the same side of gusset and
the short legs have been kept back to back.
Net sectional area of the connected legs.
A1 = 2 (150 – 19.5 – 5) × 10 = 2510 mm2
Area of outstanding (unconnected) legs,
⎛ 1 ⎞
A2 = 2 × ⎜ 75 – × 10 ⎟ × 10
⎝ 2 ⎠
= 1400 mm 2

⎛ 5 A1 ⎞ ⎛ 5 × 2510 ⎞
k = ⎜ ⎟ = ⎜ ⎟
⎝ 5 A1 + A2 ⎠ ⎝ 5 × 2510 + 1400 ⎠
= 0.8996
Net effective area of the angle
Step 2 : Tensile strength of section
An = (A1 + A2 . k)
= 2510 + 1400 × 0.8996
= 3769.493 mm2
It is assumed that the value of yield stress for the steel used is 260 N/mm2.
The permissible stress in axial tension
σat = (0. 6 × 260)
= 156 N/mm2
Tensile strength of the member in case the angles are connected with tacking
rivets along their length at suitable spacing
294 DESIGN OF STEEL STRUCTURES

Example 5.3 In Example 5.2 if the angles are connected to each side of a
gusset, determine the tensile strength of the member.
Solution
Step 1 : Net area of section
From steel section tables, for 2 ISA 150 mm × 75 mm × 10 mm (2 ISA 15075,
@ 0.169 kN/m)
Cross-sectional area
2 × 2156 = 4312 mm2
Deductions for rivet holes
2 × 19.5 × 10 = 390 mm2
Net effective sectional area
(4312 – 390) = 3922 mm2
Step 2 : Tensile strength of section
It is assumed that the value of yield stress for the steel used is 260 N/mm2.
The permissible stress in axial tension
σat = (0.6 × 260) = 156 N/mm2
Tensile strength of the member in case the angles are connected with the
tacking rivets along their length at suitable spacing
⎛ 3922 × 156 ⎞
Pt = ⎜ ⎟ = 611.832 kN
⎝ 1000 ⎠
Example 5.4 Long leg of an ISA 150 mm × 75 mm (ISA 15075) is connected
to a gusset plate by 20 mm diameter rivets in two rows. The gauge space is 55
mm and the staggered pitch is 40 mm. Determine the thickness of the angle
which would be sufficient to transmit a pull of 250 kN. Allowable tensile stress =
0.6 fy.
Solution
Step 1: Net area of section

1
55 m m
2

1 50 m m
55 m m 55 m m 3

40 m m
4
40 40
mm mm
Fig. 5.6

In the zig-zag chain line of rivet holes 1–2–3–4, Fig. 5.6 the net effective width
⎛ p2 ⎞
b1 = b – ⎜ nd – n´ ⎟
⎝ 4g ⎠
 DESIGN OF TENSION MEMBERS 295

⎛ 1 × 402 ⎞
= 150 – ⎜ 2 × 21 ⋅ 5 – ⎟ = 114.273 mm
⎝ 4 × 55 ⎠
Let t be the thickness of the angle
Net effective sectional area of the connected leg
⎛ t⎞
A1 = ⎜114 ⋅ 273 – ⎟ × t mm2
⎝ 2⎠
Area of the outstanding leg
⎛ t⎞
A2 = ⎜ 75 – ⎟ × t mm2
⎝ 2⎠
⎛t⎞
Neglecting ⎜ ⎟ within the parenthesis
⎝2⎠
∴ A1 = 114.273 t and A2 = 75t
⎛ 3 A1 ⎞ ⎛ 3 × 114 ⋅ 273t ⎞
k= ⎜ ⎟ = ⎜ ⎟ = 0.82
⎝ 3 A1 + A2 ⎠ ⎝ 3 × 114 ⋅ 273t + 75t ⎠
Net effective sectional area
An = (Al + A2 . k ) = (114.273t + 0.82 × 75 × t)
= 175.8 t mm2
Step 2 : Thickness of section
Pull transmitted by the angle
= 175.8 t × 156 N
175.8 t × 156 = 250 × l000
t = 9.12 mm
Thickness of the angle section (required) is 10 mm.
Example 5.5 Both legs of an ISA 100 mm ×100 mm × 10 mm (ISA 100,100 @
0.149 kN/m) are connected to the gusset plates by 20 mm diameter rivets in
staggered chain line as shown in Fig. 5.7. Determine the staggered rivet pitch so
that the angle section may transmit a pull of 230 kN.
Solution :
Step 1 : Net effective sectional area
Diameter of rivet hole
30 + 1.5 = 31.5 mm
Gross width of angle = Sum of width of two legs less thickness of legs
(100 + 100 – 10) = 190 mm
Gauge space, g = 60 + 60 – 10 = 110 mm
⎛ p2 ⎞
Net effective width = 190 – ⎜ 2 × 21 ⋅ 5 – ⎟ mm
⎝ 4 × 110 ⎠
⎛ p2 ⎞ ⎛ p2 ⎞
⎜190 – 430 + ⎟ = ⎜147 ⋅ 0 + ⎟ mm
⎝ 440 ⎠ ⎝ 440 ⎠
296 DESIGN OF STEEL STRUCTURES

1
2

IS A 1 00 m m x 10 0 x 10 m m

Fig. 5.7

⎛ p2 ⎞
Net effective sectional area = ⎜147 ⋅ 0 + ⎟ ×10 mm2
⎝ 440 ⎠
Step 2: Staggered pitch of rivets
Pull transmitted by the angle

⎛ p2 ⎞
= ⎜147.0 + × 156 N
⎝ 440 ⎟⎠

⎛ p2 ⎞
⎜147.0 + 440 ⎟ × 156 = 230 × 1000, ∴ p = 764.22 mm
⎝ ⎠
Adopt staggered pitch of the rivets equal to 760 mm

5 .7 DESIGN OF TENSION MEMBERS SUBJECTED TO AXIAL

TENSION
A tension member subjected to axial tension is designed on the basis of its net
sectional area. The end connections of a tension member should be such that
the deduction for rivet and bolt holes from the gross-sectional area is minimum.
The rivets connecting the tension member with the gusset plate or other sections
should be arranged on the line of action of load as far as possible.
The following are the usual steps for the design of tension members. The
specifications for the design of tension members as per IS : 800–1984 have also
been given.
Step 1. The axial pull to be transmitted by the member and allowable stress
in axial tension are known for the steel with yield stress fy. This determined
sectional area of the member required.
⎛ Axial pull ⎞
Net sectional area = ⎜ ⎟
⎝ Allowable stress in axial tension ⎠
 DESIGN OF TENSION MEMBERS 297

Step 2. The gross-sectional area of the member is determined by making

suitable allowable for rivet and bolt holes. Allowance for holes cannot be made
correctly as the end connection are designed after the design of tension member.
Allowance for holes in the net area should not be so small that it may require
redesign of the member or may result in long connections. This allowance for
holes should not be as large that the design may become uneconomical. In
general, the net area of a tension member should not be less than 80% of the
gross area of the member. In case, a tension member is made of thin plates, the
deductions for hole shall be less, but the width of plate needed should be more
(which will result in more secondary stresses). The allowance for different cases
has been indicated in the solved examples that follow.
Step 3. The suitable section from steel section tables is selected.
Step 4. The end connections of tension member are designed and again the
net area of the member is calculated. The net area actually provided is com-
pared with the required net area. If the net area provided is less or much larger
than the net area required, the member is redesigned.
If the tension member is normally acting as a tie but subjected to possible
reversal of stress due to loads other than wind or seismic forces, the member
shall have slenderness ratio not greater than 180.
A member normally acting as a tie truss or a bracing system but subjected to
possible reversal of stress resulting from the action of wind or earthquake forces
shall have slenderness ratio, λ not more than 350.
Note. In ISI Structural Engineer’s Handbook No. 1, tables are directly given for
single angles and double angles used as ties. The sections can be selected directly
from these tables.
The gross cross-sectional area is determined after calculating net sectional
area by making suitable allowance for rivet and bolt holes. The suitable allowance
for area for rivet and bolt holes may be made according to the following guidelines.
When a tension member consists of single angle section, then, the allowance
may be made for one rivet hole. When the tension member consists for two
angle sections, then the allowance may be made as one hole for each angle.
When a tension member consists of two channel sections then the allowance
may be made as two holes from each web or one hole from each flange whichever
is more. When a tension member consists of four angles then the allowance may
be made as one hole for each angle. When a tension member is made of four
angles and a plate as a web, then the allowance may be made as one hole for
each angle and two holes for the web. When the tension member comprises of
two channels laced then the allowance may be made as two hole from each
channel web or one hole from each flange, which is more. When a built-up tension
member consists of four angles with or without plates, then, the allowance may
be made as two rivet holes for each angle and one rivet hole for every 150 mm
width of the plate.

5.8 DESIGN OF TENSION MEMBERS SUBJECTED TO BOTH

BENDING AND AXIAL TENSION
In actual structures, the tension member are seldom subjected to pure axial
load. The tension members may be subjected to bending stresses along with
298 DESIGN OF STEEL STRUCTURES

direct tensile stress because of various reasons (e.g., the eccentricity in the
connections, the member itself may not be straight, resulting in eccentric load
at the section the member may not be vertical and it may be subjected to bending
due to self-weight). The member may be bending due to wind, vibrations or
earthquake forces. The combined stress at any section may be determined, if
the axial load and bending moment due to combined effects of eccentricity and
transverse forces are known. The combined stress at any section is given by

⎛P M ⎞
f = ⎜ + xx ⋅ y ⎟ ...(i)
⎝ A I xx ⎠
where, Mxx = Bending moment above xx-axis
y = Distance to the extreme fibre from the neutral axis (xx-axis)
Ixx = Moment of inertia of the section about xx-axis
P = Axial tensile load
A = Net cross-sectional area of the section
When the section is subjected to bending about the both the axes, then the
combined stress is given by

⎛P M M yy ⎞
f = ⎜ + xx ⋅ y + ⋅ x⎟ ... (ii)
⎝A lxx lyy ⎠
where, Myy = Bending moment about yy-axis
Ixx = Moment of inertia about yy-axis
x = Distance to the extreme fibre from the neutral axis ( yy-axis)
When the member is subjected to pure axial tension load P, then the net area
required is given by
⎛ P ⎞
Aa = ⎜ σ ⎟ ...(iii)
⎝ at ⎠
where, σat = Maximum allowable stress in axial tension (0.6 fy)
fy = Minimum yield stress of the steel.
When the member is subjected to pure bending then the bending stress is
given by
⎛M ⋅y M ⋅y⎞
σbt = ⎜ = ⎟ ...(iv)
⎝ I Ab⋅r 2 ⎠
The required cross-sectional area is given by

⎛ M ⋅ y⎞
Ab = ⎜ 2 ⎟ ...(v)
r
⎜ ⎟
⎝ σbt ⎠
where, I = Moment of inertia of the section about the axis about which
bending takes place
 DESIGN OF TENSION MEMBERS 299

r = Radius of gyration of the section

σbt = Maximum allowable bending stress in tension
When the member is subjected to axial tensile load and bending then the net
cross-sectional area required is given by
⎛ M⋅y⎞
⎜ P 2 ⎟
A = ( Aa + Ab ) = ⎜⎜ + r ⎟⎟
⎝ σat σbt ⎠

⎛ P M⋅y⎞
⎜ 2 ⎟
or ⎜⎜ A + Ar ⎟⎟ = 1.0
⎝ σat σbt ⎠

⎛ σ at⋅cal σbt⋅cal ⎞
or +
⎜⎝ σ
at σbt ⎟⎠ < 1.00 ...(vi)

where, σat.cal = P = Actual axial tensile stress

A
⎛M ⋅y M ⋅y⎞
σbt.cal = ⎜ = ⎟
⎝ Ar 2 I ⎠
= Actual bending tehsile stress in the extreme fibre
When the tension member is subjected to both axial tension and bending
then it is so proportioned so that
⎛ σ at⋅cal σ ⎞
+ bt⋅cal < 1.00 ...(vii)
⎜⎝ 0.60 f 0.66 fy ⎟⎠
y

When the bending occurs about both the axes of members actual bending
tensile stresses σbtx.cal and σbty are calculated.
The member is so proportioned that
⎛ σ at⋅cal σbtx ⋅wcal σbty.cal ⎞
⎜ 0.6 f + σ + < 1.00 ...(viii)
⎝ y bt .x σbr . y ⎟⎠
where, σat.cal = Calculated average axial tensile stress
σbt.cal = Calculated bending tensile stress in the extreme fibre
σat = Permissible bending tensile stress in the member subjected to
axial load
σbt = Permissible bending tensile stress in extreme fibre.
AISC specification recommends that the member should be so proportioned
that

⎛ σt σbt ⎞
⎜ 0.6 f + σ ⎟ should not exceed unity
⎝ y bt ⎠

where fy = yield stress of the steel.

300 DESIGN OF STEEL STRUCTURES

The trial section of a tension member subjected to axial tension and bending
may be found by determining the equivalent axial tensile load. The equivalent
axial tensile load is the load, which produces the average axial tensile stress in
the section equivalent to the maximum combined stress at the extreme fibre of
the section. The equivalent axial load Pequiv is given by
⎛ M ⎞
Pequiv = ⎜ P + × A ⋅ y⎟ ...(vii)
⎝ I ⎠

⎛ M ⋅A⎞
or Pequiv = ⎜ P + ...(viii)
⎜⎜ ⎛ I ⎞ ⎟⎟
⎜ y⎟ ⎟
⎝ ⎝ ⎠ ⎠

⎛A⎞
or Pequiv = P + M ⎜ ⎟ ...(ix)
⎝Z⎠
or Pequiv = P + M.(B) ...(5.6)
where, B = (A/Z) = Bending factor
= Ratio of net area of the section to the section modulus of the
area
When the member is subjected to bending about both the axes then the equiva-
lent axial tensile load is given by
Pequiv = (P + Mxx Bx + Myy.By ...(5.7)
where, Bx = Bending factor of the section about the strong axis (i.e., xx-axis)
By = Bending factor of the section about the weak axis (i.e., yy-axis)
Example 5.6 Design a single angle section for a tension member of a roof
truss to carry a pull of 100 kN. The member is subjected to possible reversal of
stress due to action of wind. The length of the member from centre to centre of
intersection is 3.50 metres.
Solution
Design :
Step 1 : Selection of trial section
The steel having yield stress fy as 260 N/mm2 shall be used. The permissible
bending stress for the member subjected to axial tension
σat = (0. 60 × 260) = 156 N/mm2
Net effective sectional area required
⎛ 100 × 1000 ⎞
⎜ ⎟ = 641.03 mm2
⎝ 156 ⎠
In case of single angle section the gross area is assumed to 40 to 50 per cent
in excess of the net area required.
Gross area = (l.40 × 641.03) = 897.44 mm2
From steel section tables try ISA 65 mm × 65 mm × 10 mm
Step 2 : Net effective sectional area
Gross area = 1200 mm2
rmin = 12.5 mm
 DESIGN OF TENSION MEMBERS 301

Net effective sectional area

An = (A1+ A2 . k)
Assume 18 mm diameter rivet used for end connections. Diameter of rivet
whole
(18 + 1.5) = 19.5 mm
Net sectional area of connected leg,
Ax = (6.5 – 19.5 – 0.5 × 10) × 10 = 405 mm2
Area of outstanding leg,
⎛ 10 ⎞
A2 = ⎜ 65 − ⎟ × 10 = 60 mm2
⎝ 2 ⎠

⎛ 3 A1 ⎞ ⎛ 3 × 405 ⎞
k = ⎜ ⎟ = ⎜⎝ 3 × 405 + 600 ⎟⎠ = 0.669
⎝ 2 A1 + A2 ⎠
Net effective sectional area provided
An = (405 + 0.669 × 600) = 806.65 mm2
> Net sectional area required (641.03 mm2)
Step 3 : Check for slenderness ratio
Assume two or more than two rivets are used for end connections.
⎛ l ⎞ ⎛ 0.85 × 3.50 × 1000 ⎞
Slenderness ratio ⎜ ⎟ = ⎜⎝ ⎟ = 238 < 350
⎝ rmin ⎠ 12.5 ⎠
Hence design is satisfactory.
Provide ISA 65 mm × 65 mm × 10 mm (ISA 6565 @ 0 094 kN/m) as tension
member.
Example 5.7 Design a tension member consisting of a pair of angles (back to
back) and connected by the short legs to the same side of gusset plate. The mem-
ber is to carry a pull of 250 kN.
Solution
Design :
Step 1 : Selection of trial section
It is assumed that the steel having yield stress, fy as 260 N/mm2 shall be
used.
The permissible stress in axial tension
σat = (0.6 × 260) = 156 N/mm2
Pull to be carried by tension member = 250 kN
⎛ 250 × 1000 ⎞
Net area required = ⎜ ⎟ = 1602.56 mm2
⎝ 156 ⎠
Assume gross area 40 per cent in excess of the net area required.
Gross area = (1.40 × 1602.56) = 2243.59 mm2
From steel section tables, try 2 ISA 90 mm × 60 mm × 10 mm
302 DESIGN OF STEEL STRUCTURES

Step 2 : Net effective sectional area

Gross area = 2802 mm2
Use 20 mm diameter rivets. Diameter of rivet hole
= 21.5 mm
Net sectional area of connected leg,
⎛ 10 ⎞
A1 = 2 × ⎜ 60 − 21.5 − ⎟ × 10 = 690 mm2
⎝ 2 ⎠
Area of the outstanding leg,
⎛ 10 ⎞
A2 = 2 × ⎜ 90 − ⎟ × 10 = 1700 mm2
⎝ 2 ⎠

⎛ 5 A1 ⎞ ⎛ 5 × 699 ⎞
k1 = ⎜ ⎟ = ⎜⎝ 5 × 690 × 1700 ⎟⎠ = 0.6727
⎝ 5 A1 + A2 ⎠
Net effective sectional area
(A1+ A2 . k) = (699 +1700 × 0.6727)
= 1842.696 mm2
> Net area required (1602.56 mm2)
Hence, the design is satisfactory.
Provide, 2 ISA 90 mm × 60 mm × 10 mm (2 ISA 9060, @ 0.110 kN/m) section
and stitch rivets at pitch of 1.0 metre.
Example 5.8 If in Example 5.7 the angles are connected to each side of the
gusset plate, design the section.
Solution
Design :
Step 1 : Selection of trial section
⎛ 250 × 1000 ⎞
Net area required = ⎜ ⎟ = 1602.56 mm2
⎝ 156 ⎠
The angles are connected to both the sides of the gusset plate.
Use 20 mm diameter rivets for end connections. Diameter of rivet hole
= 21.5 mm
Assume one rivet hole in each angle and 6 mm thickness of angle.
Gross area = (1667 + 2 × 2.15 × 6) = 1925 mm2
From steel section tables
Try 2 ISA 100 mm × 75 mm × 6 mm (2 ISA 10075, @ 0.080 kN/m)
Step 2 : Net area
Gross area = 2028 mm2
Net area provided = (2028 –258) = 1770 mm2
Hence the design is satisfactory. Provide 2 ISA 100 mm × 75 mm × 6 mm (2
ISA 10075 @ 0.080 kN/m).
 DESIGN OF TENSION MEMBERS 303

5.9 TENSION MEMBER SPLICE

When a joint is to be provided in a tension member, then splice plates are used.
Splice plates and rivets are designed for the pull required to be transmitted by
the tension member. If the tension members are of unequal thickness, then, the
packings are used to have surfaces of tension members in one level. The rivets
or bolts carrying calculated shear stress through a packing greater than 6 mm
thick shall be increased above the number required by normal calculations by
2.5 per cent for each 20 mm thickness of packing. For double shear connections
packed on both sides, the number of additional rivets or bolts required shall be
determined from the thickness of the thicker packing. The additional rivets or
bolts should preferably be placed in an extension of the packing.
Example 5.9 In a truss girder of a bridge, a diagonal consists of mild steel
flat 400 F16 and it carries a pull of 750 kN. Design a suitable splice for the
member.
Solution
Design :
Step 1 : Design of splice

4 00 m m

Fig. 5.8

Provide two 400 ISF 8 as splice plates to join the flats.

Step 2 : Rivet value
Provide 22 mm normal diameter, power driven rivets
Gross diameter of rivet = 23.5 mm
Strength of rivet in double shear
⎛ π 2 100 ⎞
⎜ 2 × (23 ⋅ 5 ) × ⎟ = 86.70 kN
⎝ 2 1000 ⎠
Strength of rivet in bearing
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 16 × ⎟ = 112.8 kN
⎝ 1000 ⎠
Rivet value, R = 86.70 kN
304 DESIGN OF STEEL STRUCTURES

Step 3 : Number of rivets required

⎛ 750 ⎞
⎜ ⎟ = 8.65
⎝ 86 ⋅ 70 ⎠
Provide 9 rivets in diamond group of riveting as shown in Fig. 5.8.
The strength of splice plates is equal to the strength of main member as the
total thickness of splice plates is equal to the thickness of the main member.
Example 5.10 A bridge truss diagonal carries an axial pull of 300 kN. Two
mild steel flats 250 ISF 10 and 260 ISF 18 of the diagonal member are to be
joined together. Design a suitable splice.
Solution
Design :
Step 1 : Rivet value
Provide 22 mm nominal diameter power driven rivets.
Gross diameter of rivet = 23.5 mm
Strength of rivet in double shear

⎛ π 2 100 ⎞
⎜⎝ 2 × (23 ⋅ 5) × ⎟ = 86.70 kN
2 1000 ⎠
Strength of rivet in bearing

⎛ 23 ⋅ 5 × 10 × 300 ⎞
⎜ ⎟ = 70.5 kN
⎝ 1000 ⎠
Rivet value, R = 70.5 kN
Step 2 : Number of rivet required

⎛ 300 ⎞
⎜⎝ 70.5 ⎟⎠ = 4.25

Provide 6 rivets.
Thickness of packing required
(18 –10) = 8 mm > 6 mm
Provide additional rivets @ 2.5 per cent for each 2.0 mm thickness of packing.
Additional percentage of rivets
⎛ 2 ⋅ 5 × 8⎞
⎜⎝ ⎟ = 10 per cent.
2 ⎠
Number of additional rivets
⎛ 6 × 10 ⎞
⎜ ⎟ = 0.61 Ω 1 rivet
⎝ 100 ⎠
Provide additional rivet in the extension of packing as shown in Fig. 5.9.
 DESIGN OF TENSION MEMBERS 305

Fig. 5.9

Provide splice plate 6 mm thick. The strength of splice at the weakest section
⎛ 0 ⋅ 6 × 260 ⎞
= ⎜ 2 (250 – 2 × 23 ⋅ 50) × 6 × ⎟ kN
⎝ 1000 ⎠
= 380.02 kN > Design load. Hence, the design is satisfactory.

5.10 LUG ANGLES

Length of the end connections of angle or channel section to a gusset plate is
reduced by using lug angles. Lug angle used with single angle member is shown
in Fig. 5.10. Rivets connecting lug angle to the gusset plates are less effective

A n gle

L ug an g le

G usset plate
Fig. 5.10

because of possible deformation of outstanding leg of the lug angle. As such,

rivets are provided in excess (as per specification) of the number of rivets required
by computation.
Specifications. The following specifications for the design of lug angle may
be followed:
1. In the case of angle members, the lug angles and their connections to
the gusset or other supporting member shall be capable of developing
strength not less than 20 per cent in excess of the force in the outstand-
ing leg of the angle, and the attachment of the lug angle to the angle
member shall be capable of developing 40 per cent in excess of that
force.
306 DESIGN OF STEEL STRUCTURES

2. The effective connection of the lug angle shall as far as possible

terminate at the end of the member connected and the fastening of the
lug angle to the member preferably start in advance of the direct con-
nection of the member to the gusset or other supporting member.
3. Where lug angles are used to connect an angle member, the whole area
of the member shall be taken as effective rather than net effective sec-
tion for angles as per Sec. 5.3. (The whole area of the member is the
gross area less deduction for rivet and bolt holes).
4. In no case, less than two rivets or bolts shall be used for attaching the
lug angle to the gusset or other supporting member. The lug angle used
with channel section has been shown in Fig. 5.11.

L ug
a ng le

L ug
a ng le

G usset plate

Fig. 5.11

5. In the case of channel members and the like, the lug angles and their
connection to the gusset or other supporting member shall be capable of
developing a strength of not less than 10 per cent in excess of the force
not accounted for by the direct connection of the member and the at-
tachment of the lug angles to the member shall be capable of developing
20 per cent in excess of that force.
Example 5.11 Design a riveted end connection for the full strength of ISA
100 mm × 116 mm × 10 mm. Provide a lug angle in the connection. Adopt stress
as per IS : 800 –1984.
Solution
Design :
Step 1 : Strength of angle section
Provide 22 mm nominal diameter power driven rivets.
Gross diameter of rivet = 23.5 mm
 DESIGN OF TENSION MEMBERS 307

From steel section tables, gross area of ISA 110 mm × 110 mm × 10 mm

10 110, @ 0.165 kN/m) = 2106 mm2
Assume, the angle section will be reduced by one rivet hole only. Net area of
angle section
(2106 – 23.5 × 10) = 1871 mm2
Tensile strength of the angle section
⎛ 1871 × 0 ⋅ 6 × 260 ⎞
⎜ ⎟ = 291.876 kN
⎝ 1000 ⎠
Strength of outstanding leg of the main angle

⎛ (110 – 10 ) ⎞ ⎛ 100 ⎞
⎜( × 221 ⋅ 876 ⎟ = ⎜ × 291 ⋅ 876 ⎟ = 138.988 kN
)
⎝ 110 – 10 + 110 ⎠ ⎝ 210 ⎠
Strength of connected leg of the main angle
= 152.888 kN
Provide a gusset plate 12 mm thick
Step 2 : Strength of rivet
Strength of rivet in angle in single shear
⎛π ( 2 100 ⎞
⎜ × 23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of rivet in bearing on 10 mm thickness
⎛ 23.5 × 10 × 300 ⎞
⎜⎝ 1000 ⎟⎠ = 70.50 kN

Assume size of lug angle ISA 110 mm × 110 mm × 8 mm

Strength of rivet in bearing on 8 mm thickness
⎛ 23 ⋅ 5 × 8 × 300 ⎞
⎜ ⎟ = 56.4 kN
⎝ 1000 ⎠
Step 3 : Number of rivets joining connected leg of main member with the
gusset plate
⎛ 152.888 ⎞
⎜ ⎟ = 3.525 @ 4
⎝ 43 ⋅ 35 ⎠
Number of rivets joining connected leg of the lug angle with gusset
⎛ 1 ⋅ 2 × 138 ⋅ 988 ⎞
⎜ ⎟ = 3.847 @ 4
⎝ 43 ⋅ 35 ⎠
Number of rivets joining outstanding legs of lug and main angles
⎛ 1 ⋅ 4 × 138 ⋅ 988 ⎞
⎜ ⎟ = 4.488 = 5
⎝ 43 ⋅ 35 ⎠
The design has been shown in Fig. 5.12.
308 DESIGN OF STEEL STRUCTURES

IS A 11 0 m m × 11 0 m m
× 10 m m

IS A 11 0 m m ×
11 0 m m × 10 m m G usset plate
1 2 m m th ick

Fig. 5.12

Step 4: Selection of lug angle section

The lug angle should develop strength 20 per cent in excess of the force in the
outstanding leg of the angle.
Required strength of lug angle
(l.20 × 138.988) = 166.78 kN
Net area required for lug angle
⎛ 166 ⋅ 78 × 1000 ⎞
⎜ ⎟ = 1069.138 mm2
⎝ 156 ⎠
From steel section tables (single angle used as ties)
Provide ISA 110 mm × 110 mm × 8 mm (ISA 110110 @ 0.165 kN/m) as lug
angle.

5.11 SHEAR LAG

In case a tension member is built-up of plates and thickness of plates are small,
then, the deduction for areas of rivet holes (number of holes × diameter of rivet
× thickness of plate) will be less and widths of the plates needed shall be large.
The large widths of plates results in more secondary stresses. In the design of
riveted connections for the wide plates in tension, certain precautions are nec-
essary, in the absence of which, the strength of such plates becomes less due to
effect of shear lag. The effect of shear leg may be seen by applying two tensile
forces P1 and P1 at the edges of a wide rubber sheet as shown in Fig. 5.13. The
stretch at the edges AA1 and CC1 is large as compared to that at the middle line
BB1. Except at A and A1 and C and C1, the stresses in longitudinal direction, at
all the points on edges ABC and A1B1C1 are zero. The small elements at D and
D1), on the edges ABC and A1B1C1, respectively, deform as shown. It shows that
the shear stresses are developed in the material of the sheet. These shear stresses
transmit the tensile forces from the edges towards the middle line in a certain
length approximately equal to the width of sheet. The stresses in tension in this
length shall be not equal. However, the stress in tension beyond this length
 DESIGN OF TENSION MEMBERS 309

may be treated as uniform over the complete cross-section of the rubber sheet,
the transfer of tensile forces from the edges to the complete width through shear
stresses is termed as shear lag.
A A1
P
R u bb er sh ee t
C C1

B P
B1

Fig. 5.13

In case, a tension member (angle section or channel section) is connected to

the gusset plates, it is necessary to keep length of the joint sufficiently long (viz.
equal to the length of transfer or length of shear lag), so that the stresses in the
connected member at the beginning remain uniform over its complete cross-
section. If the length of joint is smaller, then, there will be stress concentration
near the edges of member and its effectiveness will be reduced as shown in
Fig. 5.14. The axial force in the tension member shown lags the transfer to the
gusset plate, because of its distribution across the member cross-section. If the
length of joint is too large, the portion of axial tensile force transmitted by each

G usset plate
I-se ctio n P
2

P
2
G usset plate

σ= P σ1 > σ1 σ2 < σ1 σ Ω
A

Fig. 5.14

rivet or bolt shall not be equal. In case, N is number of rivets or bolts used in the
connection, the rivets in first line carry more than P/N and the rivets in the last
line will carry practically zero force. Due to high concentration of stress, the
web plate may fall in tearing and tearing of plate progresses across the section.
In case, the web plate is adequate in the net section, the tearing of plate does
not occur, but it stretches (PL/AE) depending upon the value of tensile strain.
The rivets in first line will either undergo compatible shear strain or will shear-
off if the strain and resulting forward displacement of the hole is too large. The
failure of rivets in second line will also fail and so on. As such, it produces a
310 DESIGN OF STEEL STRUCTURES

progressive joint failure. This process is called as unbuttoning. It is to note that

when the axial force in tension to be transmitted is large, this process is almost
instantaneous.
If the length of joint is short, then all rivets or bolts carry load. The rivets or
bolts in first line strain with the plate. The strain in these rivets increases
corresponding to the shear yield stress. On further increase of strain, the rivets
continue to strain with no increase in load and the rivets in second line will take
up the transferred load. The ultimate load for the joint shall reach when all the
rivets or bolts have yielded.
The analysis of joint for strain compatibility is rarely done. The factors of
safety used together with the property of steel ductility are such that except for
long joint, only the rivets or bolts in first line in a connection (if any) may yield
or reach close to the yield.
This is to note that the factor of safety for the connection is kept higher than
that used for the members. This is so because, a failure of member (if it occurs)
should be earlier than the failure of joint. A joint failure used to be catastrophic,
whereas, the failure of member permits time for safety measures to be under-
taken.
Professors Munsc and Chesson, in their paper titled as, ‘Riveted and Bolted
Joints : Net Section Design’, Proceedings, ASCE, ST1, February 1963 have given
the following equation for shear lag efficiency.
x
η = ⎛⎜1 – ⎞⎟ × 100 ...(5.8)
⎝ L⎠
where X = distance from gravity axis to fastener plane
L = length of the joint.
Professor Bendigo, Hansen and Rumpf in their paper, titled as ‘long bolted
joints’, Proceedings, ASCE. Vol. 84, ST6, December 1963, has mentioned that
the efficiency of joint is given by the following equation (assuming spacing as 3
times diameter)
η = 0.85 – C1 . (L –C2) ...(5.9)
where C1 = 000275 in SI units, and
C2 = 406 mm > (L > C)
Equation 5.9 indicates that the connections with joint lengths upto 406 mm
have an efficiency of 85 per cent. The effect of shear lag may be accounted by
calculating the effective area
⎛ x⎞
Aeff = Anet ⎜1 – ⎟ ...(5.10)
⎝ L⎠
For a single plane connection, x is the centroidal co-ordinate of the entire
cross-section, while for a symmetrical double plane connection, it is the dis-
tance from each connector plane to the centroid of the half cross-section as shown
in Fig. 5.15.
 DESIGN OF TENSION MEMBERS 311

G G G G G x x G

x x x

(a ) (b ) (c) (d )

x x
+ +

x + + x

x
(e ) (f) (g )

Fig. 5.15

PROBLEMS
5.1 In a roof truss, a diagonal consists of ISA 80 mm × 80 mm × 10 mm (ISA
8080, @ 0.118 kN/m) and is connected to a gusset plate by one leg only by
20 mm diameter power driven rivet in one line along the length of the
member. Determine tensile strength of the member.
5.2 A double angle tie ISA 125 mm × 95 mm × 10 mm (viz., ISA 125, 95,
@ 0.165 kN/m) (short leg back to back) of a roof truss is connected to the
same side of a gusset with rivets 20 mm in diameter, such that each angle
is reduced in section by one rivet hole only. Determine the tensile strength
of the member. Stitch rivets have been provided at suitable spacing.
5.3 In Problem 5.2, if angles are connected to each side of a gusset, determine
the tensile strength of the member.
5.4 Long leg of an ISA 125 mm × 95 mm (ISA 12595) is connected to a gusset
plate by 22 mm diameter rivets in two rows. The gauge distances in long
legs of angle are 45 mm and 55 mm and staggered pitch is 45 mm. Deter-
mine the thickness of the angle which would be sufficient to transmit a
pull of 180 kN.
5.5 Design a single angle section for a tension member of a roof truss to carry
a pull 100 kN. The member is subjected to possible reversal of stress due
to action of wind. The length of the member from centre to centre of inter-
section is 3.50 m.
5.6 Design a tension member consisting of a pair of angles and connected by
their short legs to the same side of a gusset plate. The member is to carry
a pull of 180 kN.
5.7 If in Problem 5.6, the angles are connected to each side of the gusset plate,
then, design the section.
312 DESIGN OF STEEL STRUCTURES

5.8 A bridge truss diagonal carries an axial of pull of 240 kN. Two mild steel
flats 200 F 12 and 200 F 20 of the diagonal member are to be joined to-
gether. Design a suitable splice.
5.9 Design a riveted end connection for the full strength of ISA 130 mm × 130
mm × 8 mm. (ISA 130130, @ 0.159 kN/m). Provide a lug angle in the con-
nection.
5.10 The main tie of a roof truss has to carry a load of 800 kN. Design a suitable
double angle section for it. Determine also the number of 18 mm diameter
rivets required to connect the member to the gusset plates at ends.
5.11 A tension member consists of an ISA 125 mm × 75 mm × 10 mm (ISA
12575,@ 0.149 kN/m). The member is to be connected at its ends to a 12
mm thick gusset plate using a zig-zag pattern of 20 mm diameter rivets as
shown in Fig. P 5.11. Determine the axial tensile load that the angle can
safely carry.

50 50 50 50
mm mm mm mm

45 m m

IS A 1 25 x 7 5 m m x 1 0 m m (IS A 12 5 75 , @ 0 .1 49 kN/m )
Fig. P 5.11
 Chapter

6 Design of Beams

6.1 INTRODUCTION
A beam is defined as a structural member subjected to transverse loads. The
plane of transverse load is parallel to the plan of symmetry of the cross-section
of the beam and it passes through the shear centre, so that the simple binding
occurs. The transverse loads produce bending moments and shear forces in the
beam at all the sections of the beam.
The term joist is used for beams of light sections. Joists support floor con-
struction; they do not support other beams. The term subsidiary beam or sec-
ondary beam is also used for the beam supporting floor construction. Main beams
are supporting joists for subsidiary beams: these are called floor beams in build-
ings. The term girder is most commonly used in buildings. Any major beam in a
structure is known as a girder.
In the roof trusses, horizontal beams spanning between the adjacent trusses
are known as purlins. The beams resting on the purlins are known as common
rafter or simply rafters. In the buildings the beams spanning over the doors,
windows and other opening in the walls are known as lintels. The beam at the
outside wall of a building, supporting is share of the floor and also the wall up to
the floor above it are known as spandrel beams. The beams framed to two beams
at right angle to it, and usually supporting joists on one side of it; used at opening
such as stair wells are known as headers. The beam supporting the headers is
termed as trimmers. The beam supporting the stair steps are called as stringers.
In the bridge floors, the longitudinal beams supported by the floor beams are
also called as stringers. In the mill buildings, the horizontal beams spanning
between the wall columns, and supporting wall covering are called as girts.
The beams are also called simply supported; overhanging cantilever, fixed
and continuous depending upon nature of supports and end conditions.
314 DESIGN OF STEEL STRUCTURES

The rolled steel I-sections, channel, sections, angle sections, tee-sections, flat
sections and bars as shown in the Fig. 6.1 are the regular sections, which are
used as beams. The rolled steel I-sections as shown in the Fig. 6.1(a) are most
commonly used as the beams, and as such these sections are also termed as
beam sections. The rolled steel I-sections are symmetrical sections. In these
sections, more material is placed near top and bottom faces, i.e., in the flanges
as compared to the web portion. The rolled steel I-sections provide large moment
of inertia about xx-axis with less cross-sectional area. The rolled steel I-sections
provide large moments of compared to the other sections and such as these are
most efficient and economical beams sections. The rolled steel wide flange beams
as shown in Fig. 6.1 (b) provide additional desirable features. As the name
indicates, the flanges of the sections are wide. These sections provide greater
lateral stability and facilitate the connections of flanges to other members. I-
sections and wide flanges beam sections have excellent strength. The rolled
steel channel sections as shown in Fig. 6.1(c) are used as purlins and other small
structural members. The channel sections have reasonably good lateral strength
and poor lateral stability. The channel sections are unsymmetrical sections about
yy-axis. When the channel sections are loaded and supported by vertical forces
passing through the centroid of the channel, then the channel sections bend
and twist if these are laterally unsupported, except for the special case, wherein
the loads act normal to the plane of web, causing bending in the weakest direction.
The rolled steel angle sections as shown in Fig. 6.1 (d) are also used as purlins
y y y

x x x x x x

y y y
(a ) (b ) (c)
y y
y y y

x x x x
x x x

y y y y y
(d ) (e ) (f) (g ) (h )
Fig. 6.1 Rolled steel sections used as beams

and so other small structural members. The angle sections act as unsymmetri-
cal sections about both xx-axis and yy-axis. The rolled steel tee-sections as shown
in Fig. 6.1 (e) are used as beams in the rectangular water tanks. The angles and
tee-sections are used for light loads. The rolled steel flats and bars as shown in
6.1 (f), (g) and (h) are very rarely used. These sections are weak in resisting
bending. Most commonly the beams are loaded in the direction perpendicular
 DESIGN OF BEAMS 315

to xx-axis, so that the bending of beams occurs about strong (xx-axis) and xx-
axis becomes neutral axis. The beams are very rarely loaded in the direction
perpendicular to yy-axis. In such cases, yy-axis becomes neutral axis.
In general, in cases of bending of the beam about one (single) axis the load is
considered to be applied through the shear centre of the beam sections. In case,
the loading passing through the shear centre, the sections may be analyzed for
simple and binding and shear. The shear centre for the beam sections is at the
centre of area, and this load position produces simple bending about either axis.
When the load does not pass through the shear centre as in channels, angles
and some built-up sections, a torisonal moment is produced along with the bend-
ing moment and both are considered to avoid over stressing of the member. For
such sections, a special load device may be used so that the load passes through
shear centre of the section and the torisonal may be avoided.
In addition to the above, expanded or castellated beams as shown in Fig. 6.2
(b) are used. The castellated beams are light and these are economically used

(a ) S p litting pa tte rn

(b ) C a stellated b ea m s

Fig. 6.2

for the light construction. The castellated beams are made by splitting the web
of rolled steel I-sections in a predetermined pattern as shown in Fig. 6.2 (a). The
splitted portions are rejoined in such a manner as to produce a regular pattern
of opening in the web. The modulii of sections of castellated beams increase
without the increase of material and weight.

6.2 BENDING STRESS

The bending stress is also termed as flexural stress. When beams are loaded,
they bend and bending stress are set up all the sections. The established theory
of bending is expressed in the following formula:
⎛ M σ E⎞
⎜⎝ I = y = R ⎟⎠ ...(6.1)

where, M = Bending moment

I = Moment of inertia
σb = Bending stress at any point
316 DESIGN OF STEEL STRUCTURES

y = Distance from the neutral axis to the point under consideration

R = Radius of curvature of the beam.
Equation 6.1 holds good when the plane of bending coincides with the plan of
symmetry of the beam section. The bending of beam occurs in the principal
plane of the beam section. The simple bending of beam occurs i.e., the bending
is produced by the application of pure couples at the ends of the beam. In such
bending the deflection of beam does not occur due to shear. In Eq. 6.1, further,
it assumed that the vertical section of the beam plane before bending remain
plane after bending. Then stress in any given fiber is proportional to its strain
i.e., Hooke’s law holds good. For the material of beam, the value of E is same for
the complete beam.
When the load is acting downward in a simply supported beam, then the
distribution of bending stress, for any sections of a beam is as shown in Fig. 6.3.

N A
N e utral axis

fbt
(a ) Se ctio n o f be am (b ) D istribu tion o f b en ding stress

Fig. 6.3

The bending stress varies linearly. The bending stress is zero at the neutral
axis. When the load is acting downward, the bending stress is compressive above
the neutral axis of the section and tensile below it and these are denoted by σbc.cal
and σ bt.cal, respectively. The bending stress is maximum at the extreme fibre.

⎛ ⎞
⎛M ⎞ ⎜ M ⎟
σb.max = ⎜ × ymax ⎟ = ⎜ ⎟ ...(6.2)
⎝ I ⎠ ⎜ I ⎟
⎜⎝ y ⎟
max ⎠

⎛M⎞
or σb.max = ⎜ ⎟ ...(6.3)
⎝Z⎠
where, Z is the section modulus (Z = I/ymax) is the distance from the neutral
axis to the entrance fibre, and σmax is the maximum bending stress.
The maximum bending stress in the beam section (if compressive) should be
less than the allowable bending compressive stress and (if tensile); should be
less than the allowable bending tensile stress. When the section of beam is
 DESIGN OF BEAMS 317

symmetrical about the neutral axis then the value of ymax is equal to half the
depth of section, and the maximum bending stress in compression and in the
tension at the extreme fibres are equal. When the beam section is not symmetrical
about the neutral axis, then there are two distance y1 and y 2 to the two extreme
fibres from the neutral axis. The bending stress at the extreme top and bottom
⎛ I⎞ ⎛ I ⎞
are not equal. Then, the values of z1 = ⎜ ⎟ and z2 = ⎜ ⎟ both should be
⎝ y1 ⎠ ⎝ y2 ⎠
calculated and compared with the section modulus, Z of the beam section
provided.
The total compressive force C above the neutral axis is equal to the tensile
force T, for the beam is equilibrium. These two forces act in the opposite direction
and form a couple. This couple resists the bending moment and this moment is
known as moment of resistance Mr.
The moment of resistance of a beam section is the moment of the couple
which is set up at the section by the longitudinal forces C and T created in the
beam due to the bending.
Mr = (C × Lever arm) = (T × Lever arm)
For the beam in the equilibrium, the moment of resistance Mr would be
equal to the maximum bending moment M at any section (Mr = M).

6.2.1 Allowable Stress in Bending

The allowable bending stress in the design of rolled steel beam sections consid-
erably depends on the geometrical properties of the section and the lateral sup-
⎛ 1 bf ⎞
port. In case of flange width/flange thickness ⎜ ⎟ and the depth of section/
⎝ 2 tf ⎠

⎛ h⎞
thickness of web ⎜ ⎟ ratio are not adequate, the elements of beam section will
⎝ tw ⎠
tend to buckle at low compressive stresses which will be due to bending com-
bined with axial loads. If the compression flange is not laterally supported (i.e.,
supports at intervals or uniformly) along the compression zone, it will either
buckle in plane or cut-of plane coupled with twisting.

⎛ 1 bf ⎞ ⎛ h⎞
The rolled steel sections are produced with adequate ⎜ ⎟ and ⎜ ⎟ ratios
2
⎝ f⎠t ⎝ tw ⎠
such that the buckling of the flange or web does not occur. The designers may
provide supports at intervals or uniformly along the compression flange such
that its bucking is avoided.
The calculated bending compressive stress σbc.cal and bending tensile stress
σbt.cal in the extreme fibres should not exceed the maximum permissible bend-
ing stress in compression (σbc) or in the tension σbt as below.
σbc or σbt = 0.66 fy ...(i)
318 DESIGN OF STEEL STRUCTURES

The structural steel used in general construction may have yield stress as
220, 230, 240, 250, 260, 280, 300, 340, 360, 380, 400, 420, 450, 480, 510 or 540 N/
mm2 (MPa). The structural steels having these values of yield stress are also
used in flexural members.
The maximum permissible bending compressive stress in beams and channel
with equal flanges has been given separately in IS: 800–1984.
For an I-beam or channel with equal flanges bending about the axis of the
maximum strength (xx-axis), the maximum bending compressive stress on the
extreme fibre, calculated on the effective section shall not exceed the value of
maximum permissible bending compressive stress, σbc. The maximum permis-
sible values of stress are given directly in Table 6.1 (a) or 6.1 (b), Table 6.1 (c) or
6.1 (d) and Table 6.1 (e) or 6.1 (f), as appropriate steels with yield stress, fy =
250 N/mm2 (Mpa), and 400 N/mm2 (Mpa), respectively as per IS: 800–1984.
⎛T ⎞
It is to note that Table 6.1 (a), (c) and (e) are applicable for ⎜ ⎟ ratio greater
⎝ tw ⎠

⎛d ⎞
than 2 (two) and ⎜ 1 ⎟ ratio greater than 85,75 and 67, respectively. Whereas
⎝ tw ⎠

⎛T ⎞ ⎛d ⎞
Table 6.1 (b), (d) and (e) are used for ⎜ ⎟ ratio less than 2 (two) and ⎜ 1 ⎟ ratio
⎝ tw ⎠ ⎝ tw ⎠
less than 85,75 and 67, respectively. The Tables 6.1 (a) to (f) have been prepared
by deriving the given values in accordance with the expression recommended
by IS: 800–1984.
For the steels having yield stresses other than these included in Tables 6.1
(a) to (f) may be obtained from such recommended expressions by IS: 800–1984.
Following symbols have been used in Tables 6.1 (a) to (f)
D = overall depth of the beam
d1 = Depth of the beam
(i) For the web of a beam without horizontal stiffeners, the clear distance
between the flanges, neglecting fillets or the clear distance between the inner
loss of the flanges as appropriate.
(ii) For the web of a beam with horizontal stiffeners, the clear distance
between the horizontal stiffener and the tension flange, neglecting fillets or the
clear distance between the inner toes of the flanges as appropriate.
l = Effective length of the compression flange.
ry = Radius of gyration of the section about its axis of minimum
strength (yy-axis)
T = Mean thickness of the compression flange. It is equal to the area
of horizontal portion of flange divided by the width; and
tw = Web thickness.
For the rolled steel sections, the mean thickness is that which one is given
in ISI Handbook No. 1.
 DESIGN OF BEAMS 319

It is to note that for all the rolled steel beam sections given in ISI Handbook
No.1, the depth of web/thickness of web ratio is less than 67. At the same time
mean thickness of compression flange/thickness of web ratio is less than 2.0
(two) and it is at the most equal to 2.0 (two) for only WB 600, @ 1.451 kN/m
section in which t f = 23.6 mm t w = 11.8 mm.
As such, the maximum permissible bending stress σbc in equal flange I-beam
is to be adopted from Table 6.1 (b), Table 6.1 (d) and Table 6.1 (f) depending
upon the value of yield stresses fy for the structural steel to be used.
In case of compound girders with the curtailed flanges, D shall be taken as
the overall depth of the girder at the point of maximum bending moment, and T
shall be taken as the effective thickness of the compression flanges. It shall be
calculated as follows:
T = k 1× mean thickness of the horizontal portion of the compression
flange at the point of maximum bending moment.
The values coefficient k 1 are given in Table 6.2. It allows for the reduction in
thickness or breadth of flange between points of effective lateral restraints. Its
value depends upon ψ the ratio of total area of both flanges at the point of least
bending moment to the corresponding area at the point of greatest bending
moment between such points of restraint).
Table 6.1 (a) (i) Maximum Permissible Bending Stress, σbc N/mm2 (MPa). In Equal
⎛T d ⎞
Flange Beams or Channels with fv =250 N/mm2 (Mpa). ⎜ > 2.0 and 1 > 85 ⎟
⎝ tw tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 160 160 159 159 158 158 158 158 158
45 159 158 157 157 156 156 156 155 155
50 158 157 156 155 154 154 153 153 152
55 157 155 154 153 152 151 150 149 149
60 156 153 152 150 149 148 148 146 145
65 154 152 150 148 147 145 144 143 145
70 153 150 148 146 144 142 141 139 138
75 152 148 145 143 141 139 138 136 134
80 180 147 143 , 141 138, 136 135 132 130
85 149 145 141 138 136 133 132 128 126
90 147 143 139 136 133 130 128 124 122
95 146 141 137 133 130 128 125 121 118
100 145 140 135 131 128 125 122 117 114
110 142 136 131 127 123 119 116 111 107
120 139 133 127 122 118 114 111 104 100
Contd.
320 DESIGN OF STEEL STRUCTURES

Table 6.1 (a) (i) Contd.

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
130 137 130 124 118 113 109 106 99 94
140 134 127 120 114 109 105 101 93 88
150 132 124 117 110 105 100 96 88 83
160 129 121 113 107 101 96 92 84 78
170 177 118 110 104 98 93 88 80 74
180 124 115 107 100 94 89 85 76 70
190 122 113 104 97 91 86 82 73 66
200 120 110 102 94 88 83 78 70 63
210 118 108 99 92 86 80 76 67 60
220 116 105 97 89 83 78 73 64 58
230 113 103 94 87 80 75 70 62 55
240 111 101 92 84 78 73 68 59 53
250 109 99 90 82 76 70 66 57 51
260 107 97 88 80 74 68 64 55 49
270 106 95 86 78 72 66 62 53 47
280 104 93 84 76 70 65 60 51 45
290 102 97 82 74 68 63 58 50 44
300 100 99 80 72 66 61 57 48 48

Table 6.1 (a) (ii) Maximum permissible bending stress, σbcN/mm2 (MPa). in equal flange
⎛T d ⎞
I-beams or channels with f = 250 N/mm2 (MPa), ⎜ > 2.0 and 1 > 85 ⎟
⎝ tw tw ⎠

D /T →
l / ry 20 25 30 35 40 50 60 80 100
↓
40 158 158 158 157 157 157 157 157 157
45 156 155 155 155 155 155 154 154 154
50 153 153 152 152 152 151 151 151 151
55 150 149 149 148 148 148 148 147 147
60 148 146 145 145 144 144 144 143 143
65 144 143 142 141 140 140 139 139 139
70 141 139 138 137 136 135 135 135 134
75 138 136 134 133 134 131 130 130 129
80 135 132 130 128 128 126 126 125 125
Contd.
 DESIGN OF BEAMS 321

Table 6.1 (a) (ii) Contd.

D /T →
l / ry 20 25 30 35 40 50 60 80 100
↓
85 132 128 126 124 123 122 121 120 120
90 128 124 122 120 119 117 116 115 115
95 125 121 118 116 114 112 111 110 110
100 122 117 114 112 110 108 107 105 105
110 116 111 107 104 102 99 98 96 95
120 111 104 100 97 94 91 90 88 87
130 106 99 94 90 88 84 82 80 79
140 101 93 88 84 81 78 75 73 72
150 96 88 83 79 76 72 69 67 65
160 92 84 78 74 71 66 64 61 60
170 88 80 74 69 66 62 59 56 55
180 85 76 70 65 62 58 55 52 50
190 82 73 66 62 58 54 51 48 46
200 78 70 63 59 55 50 48 44 43
210 76 67 60 “56 52 47 44 41 40
220 73 64 58 53 49 45 42 38 37
230 70 62 55 51 47 42 39 36 34
240 68 59 53 48 45 40 37 31 32
250 66 57 51 46 43 38 35 32 30
260 44 55 49 .44 41 36 35 30 28
270 62 53 47 43 39 35 32 28 26
280 60 51 45 41 38 33 30 27 25
290 58 50 44 39 36 32 29 25 24
300 57 48 42 38 35 30 27 24 22

Table 6.1 (b) (i) Maximum permissible bending stress, σbcN/mm2 (MPa). in equal flange
⎛T d ⎞
I-beams or channels with fy = 250 N/mm2(MPa), ⎜ ≤ 2.0 and 1 ≤ 85 ⎟
⎝ tw tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 161 161 160 160 160 160 160 159 159
45 161 160 159 159 158 158 158 157 157
50 160 158 158 157 156 156 156 155 155
Contd.
322 DESIGN OF STEEL STRUCTURES

Table 6.1 (b) (i) Contd.

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
55 159 157 156 155 154 154 153 153 152
60 168 156 154 153 152 152 151 150 149
65 156 153 153 151 150 149 148 147 146
70 155 153 151 149 149 147 146 144 143
75 154 152 149 147 146 144 143 141 140
80 153 150 148 145 143 142 140 138 136
85 152 149 146 143 141 139 138 135 133
90 151 147 144 141 139 137 135 131 129
95 150 146 142 139 137 134 132 128 126
100 149 145 141 137 134 132 129 125 122
110 147 142 137 133 130 127 124 119 115
120 144 139 134 129 126 122 119 113 109
130 142 136 131 126 121 118 114 108 103
140 140 133 128 122 118 113 110 103 97
150 138 131 124 119 114 109 105 98 92
160 136 128 121 115 110 106 101 93 87
170 134 126 119 112 107 102 98 89 83
180 131 123 116 109 104 99 94 85 79
190 129 121 113 106 101 95 91 82 75
200 127 118 111 104 98 92 88 79 72
210 125 116 108 101 95 90 85 76 69
220 123 114 106 99 92 87 82 73 66
230 122 112 103 96 90 84 80 70 63
240 120 110 101 94 87 82 77 68 61
250 118 108 99 92 85 80 75 65 59
260 116 106 97 89 83 77 73 63 57
270 114 104 95 87 81 75 71 61 55
280 113 102 93 85 79 73 69 59 53
290 111 100 91 84 77 72 67 58 51
300 109 98 89 82 75 70 65 56 49
 DESIGN OF BEAMS 323

Table 6.1 (b) (ii) Maximum permissible bending stress, σbc N/mm2 (MPa). in equal flange
⎛T d ⎞
I-beams or channels with f = 250 N/mm2 (MPa), ⎜ ≤ 2.0 and 1 > 85 ⎟
⎝ tw tw ⎠

D /T →
l / ry 20 25 30 35 40 50 60 80 100
↓
40 160 159 159 159 159 159 159 159 159
45 158 157 157 157 157 157 157 157 157
50 156 155 155 155 154 154 154 154 154
55 153 153 152 152 152 151 151 151 151
60 151 150 149 149 149 148 148 148 148
65 148 147 146 146 145 145 144 144 144
70 146 144 143 142 142 141 141 140 140
75 143 141 140 139 138 137 137 136 136
80 140 138 136 135 134 133 132 132 132
85 138 135 133 131 130 129 128 127 127
90 135 131 129 127 126 125 124 123 123
95 132 128 126 124 122 121 120 119 118
100 129 125 122 120 118 116 115 114 113
110 124 119 115 113 111 108 107 105 105
120 119 118 109 106 104 101 99 97 96
130 114 108 103 99 97 94 91 89 88
140 110 103 97 94 91 87 85 82 81
150 105 98 92 88 85 81 78 76 74
160 101 93 87 83 80 75 73 70 68
170 98 89 83 79 75 70 68 64 63
180 94 85 79 74 71 66 63 60 58
190 91 82 75 71 67 62 59 55 54
200 88 79 72 67 63 58 55 51 50
210 125 76 69 64 60 55 52 48 46.
220 82 73 66 61 57 52 49 45
230 80 70 63 58 55 49 46 42 40
240 77 68 61 56 52 47 43 40 38
250 75 65 59 54 50 44 41 37 35
260 73 63 57 52 48 42 39 35 33
270 71 61 55 50 46 41 37 33 31
280 69 59 53 48 44 39 35 32 30
290 67 58 51 46 42 37 33 31 28
300 65 56 49 45 41 36 32 29 27
324 DESIGN OF STEEL STRUCTURES

Table 6.1 (c) (i) Maximum permissible bending stress, σbcN/mm2 (MPa). in equal flange

⎛T d ⎞
I-beams or channels with fy = 340 N/mm2 (MPa), ⎜ > 2.0 or 1 > 75 ⎟
⎝ tw tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 215 214 212 212 211 211 210 210 209
45 213 211 209 208 207 206 206 205 204
50 210 208 205 204 203 202 201 199 199
55 208 204 202 200 198 197 196 194 191
60 205 201 198 195 193 191 190 188 186
65 203 198 194 191 188 186 185 181 180
70 200 195 190 186 183 181 179 175 173
75 198 192 186 182 179 176 173 169 166
80 195 188 183 178 174 170 168 163 159
85 193 185 179 174 169 165 162 157 153
90 190 182 175 169 165 161 157 151 147
95 188 179 172 165 160 156 152 145 141
100 185 176 168 162 156 151 147 140 135
110 180 170 162 154 148 143 138 130 124
120 176 165 155 147 141 135 130 121 115
130 171 159 149 141 134 128 122 113 106
140 167 154 144 135 127 121 116 105 98
150 163 150 139 129 122 115 110 99 92
160 158 145 134 124 116 110 104 93 86
170 155 141 129 120 111 105 99 88 80
180 151 137 125 115 107 100 94 83 76
190 147 131 121 111 103 96 90 79 72
200 144 129 117 107 99 92 86 75 68
210 140 125 113 103 95 88 83 72 64
220 137 122 110 100 92 85 79 69 61
230 134 119 107 97 89 82 76 66 58
240 131 116 104 94 86 79 74 63 56
250 128 113 101 91 83 76 71 61 53
260 125 110 98 88 80 74 68 58 51
270 122 107 95 86 78 72 66 56 49
280 120 105 93 83 76 69 64 54 47
290 117 102 90 81 73 67 62 52 46
300 115 100 88 79 71 65 60 51 44
 DESIGN OF BEAMS 325

Table 6.1 (c) (ii) Maximum permissible bending stress, σbcN/mm2 (MPa). in equal flange

⎛T d ⎞
I-beams or channels with fy = 340 N/mm2 (MPa), ⎜ > 2.0 and 1 > 75 ⎟
⎝ tw tw ⎠

D /T →
l / ry 20 25 30 35 40 50 60 80 100
↓
40 210 210 209 209 209 209 209 209 209
45 206 205 204 204 204 203 203 203 203
50 201 199 199 198 198 197 197 197 197
55 196 194 193 192 191 191 190 190 190
60 190 188 186 185 185 184 183 183 183
65 185 181 180 178 177 176 176 175 175
70 179 175 173 171 170 169 168 167 167
75 173 169 166 164 163 161 160 159 159
80 168 163 159 157 156 154 153 151 151
85 162 157 153 150 149 146 145 144 143
90 157 151 147 144 142 139 138 136 136
95 152 145 141 137 135 132 131 129 128
100 147 140 135 131 129 126 124 122 121
110 138 130 124 120 117 114 112 109 108
120 130 121 115 110 107 103 101 98 97
130 122 113 106 101 98 93 91 80 97
140 116 105 98 93 90 85 82 79 78
150 110 99 92 87 83 78 75 72 70
160 104 93 86 80 77 72 68 65 63
170 99 88 80 75 71 66 63 59 58
180 94 83 76 70 66 61 58 54 53
190 90 79 72 66 62 57 54 50 48
200 86 75 68 62 58 53 50 46 44
210 83 72 64 59 55 50 46 43 41
220 79 69 61 56 52 47 43 40 38
230 76 66 58 53 49 44 41 37 35
240 74 63 56 51 47 42 38 35 33
250 71 61 53 48 44 39 36 32 31
260 68 58 51 46 42 37 34 31 29
270 66 56 49 44 41 36 32 29 27
280 64 54 47 42 39 34 31 27 25
290 62 52 46 41 37 32 29 26 24
300 60 51 44 39 36 31 28 25 23
326 DESIGN OF STEEL STRUCTURES

Table 6.1 (d) (i) Maximum permissible bending stress, σbcN/mm2 (MPa), in equal
⎛T d ⎞
flangeI-beams or channels with fy = 340 N/mm2 (MPa), ⎜ ≤ 2.0 and 1 ≤ 75 ⎟
⎝ tw tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 217 216 215 214 214 213 213 213 212
45 215 214 212 211 211 210 210 209 208
50 213 211 209 208 207 206 206 205 204
55 211 209 206 205 203 202 201 200 199
60 209 206 203 201 199 198 197 195 193
65 207 203 200 197 195 193 192 189 188
70 205 201 197 194 191 189 187 184 182
75 203 198 194 190 187 184 182 178 176
80 201 195 190 186 183 180 177 173 170
85 199 193 187 183 179 175 173 168 164
90 197 190 184 179 175 171 167 162 158
95 195 187 181 175 171 167 163 157 153
100 193 185 178 172 167 163 159 152 147
110 188 180 172 165 159 155 150 142 137
120 184 175 166 159 152 147 142 133 127
130 180 170 161 153 146 140 135 125 119
140 177 165 156 147 140 134 128 118 111
150 173 161 151 142 134 128 122 112 104
160 169 157 146 137 129 122 117 106 98
170 166 153 142 132 124 117 111 100 92
180 162 149 137 128 120 113 107 95 87
190 159 145 133 124 115 108 102 91 82
200 155 141 130 120 111 104 98 86 78
210 152 138 126 116 108 100 94 83 74
220 149 135 123 113 104 97 91 79 71
230 146 132 119 109 101 94 88 76 68
240 143 128 116 106 98 91 85 73 65
250 141 126 113 103 95 88 82 70 62
260 138 123 110 100 92 85 79 68 60
270 135 120 108 98 89 82 77 65 58
280 133 117 105 95 87 80 74 63 56
290 130 115 103 93 84 78 72 61 54
300 128 112 100 90 82 76 70 59 52
 DESIGN OF BEAMS 327

Table 6.1 (d) (ii) Maximum permissible bending stress, σbcN/mm2 (MPa), in equal
⎛T d ⎞
flangeI-beams or channels with fy = 340 N/mm2 (MPa), ⎜ ≤ 2.0 and 1 > 75 ⎟
t
⎝ w tw ⎠

D /T →
l / ry 20 25 30 35 40 50 60 80 100
↓
40 213 213 212 212 212 212 212 212 212
45 210 209 208 208 208 208 208 207 507
50 206 205 204 203 203 203 203 202 202
55 201 200 199 198 198 197 197 197 197
60 197 195 193 193 195 191 191 191 190
65 192 189 188 187 186 185 184 184 184
70 187 184 182 181 180 178 178 177 177
75 182 178 176 174 173 172 171 170 169
80 177 173 170 168 167 165 164 163 162
85 173 168 164 162 160 158 157 156 155
90 167 162 158 156 154 151 150 148 148
95 163 157 153 150 148 145 143 142 141
100 159 152 147 144 142 138 137 135 134
110 150 142 137 133 130 126 124 122 121
120 142 133 127 123 120 116 113 110 139
130 135 125 119 114 110 106 103 100 99
140 128 118 111 105 102 97 94 91 89
150 122 112 104 99 95 89 86 83 81
160 117 106 98 92 88 82 79 75 74
170 111 100 92 86 82 76 73 69 67
180 107 95 87 81 77 71 67 63 61
190 102 91 82 76 72 66 63 59 56
200 98 86 78 72 68 62 58 54 52
210 94 83 74 69 64 58 54 50 48
220 91 89 71 65 61 55 51 47 45
230 88 75 68 62 58 52 48 44 42
240 85 73 65 59 55 59 45 41 39
250 82 70 62 57 52 45 43 38 36
260 79 68 60 54 50 44 40 36 34
270 77 65 58 52 48 42 38 35 32
280 74 63 56 50 46 40 36 32 30
290 72 61 54 48 44 38 35 31 29
300 70 59 52 46 42 37 33 29 27
328 DESIGN OF STEEL STRUCTURES

Table 6.1 (e) (i) Maximum permissible bending stress, σbcN/mm2 (MPa), in equal flangeI-

⎛T d ⎞
beams or channels with fy = 400 N/mm2 (MPa), ⎜ > 2.0 and 1 > 67 ⎟
⎝ tw tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 250 248 247 245 245 244 243 243 242
45 247 244 242 240 239 238 237 236 235
50 244 240 237 234 233 231 230 228 227
55 240 235 232 229 226 224 223 221 219
60 236 231 226 223 220 217 216 212 210
65 233 226 221 217 213 210 208 204 202
70 229 222 216 211 207 203 201 196 193
75 226 217 211 205 200 186 193 188 184
80 222 213 206 199 194 190 186 180 175
85 2f9 209 201 194 188 183 179 172 167
90 216 205 196 188 182 177 173 165 160
95 212 201 191 183 177 171 166 158 152
100 209 197 187 178 171 165 160 151 145
110 203 189 178 169 161 155 149 139 133
120 196 182 170 160 152 145 140 129 121
130 191 176 163 153 144 137 131 119 112
140 185 169 156 i46 137 129 123 111 103
150 179 163 150 139 130 122 116 104 96
160 174 158 144 133 124 116 109 97 89
170 169 152 139 127 118 110 104 92 83
180 165 147 134 122 113 105 97 86 78
190 160 143 129 117 108 100 94 82 74
230 156 138 124 113 104 96 90 78 70
210 152 134 120 109 100 92 86 74 66
220 148 130 116 105 96 88 82 71 63
230 144 126 112 101 92 85 79 67 60
240 141 123 109 98 89 82 76 65 57
250 137 119 106 95 86 79 73 67 53
260 134 116 103 92 83 76 70 60 52
270 131 113 100 89 81 74 68 57 50
280 128 110 97 86 78 71 66 55 48
290 125 107 94 84 76 69 64 53 46
300 122 105 92 82 74 67 62 52 45
 DESIGN OF BEAMS 329

Table 6.1 (e) (ii) Maximum permissible bending stress, σbcN/mm2 (MPa), in equal flangeI-
⎛T d ⎞
beams or channels with fy = 400 N/mm2 (MPa), ⎜ > 2.0 and 1 > 67 ⎟
⎝ tw tw ⎠

D /T →
l / ry 20 25 30 35 40 50 60 80 100
↓
40 243 243 242 242 242 241 341 241 241
45 237 236. 235 235 234 234 234 233 233
50 230 228 227 227 226 226 225 225 225
55 223 221 219 218 217 216 216 216 215
60 216 212 210 209 208 207 206 206 205
65 208 204 202 200 199 197 197 196 195
70 201 195 193 191 189 188 187 186 185
75 193 188 184 182 180 178 177 175 175
80 186 180 176 173 171 168 167 166 165
85 179 172 167 164 162 159 158 156 155
90 173 165 160 165 154 151 149 147 146
95 166 158 152 149 146 142 140 138 137
100 160 151 145 141 138 135 133 130 129
110 149 139 133 128 125 121 118 115 114
120 140 129 121 116 113 108 106 103 101
130 131 119 112 106 103 98 95 92 90
140 123 111 103 98 94 88 85 82 80
150 116 104 96 90 86 81 77 74 72
160 109 97 89 83 79 74 71 67 65
170 104 92 83 78 73 68 64 61 59
180 97 86 78 72 68 63 59 55 54
190 94 82 74 68 64 58 55 51 49
200 90 78 70 64 60 54 51 47 45
210 86 74 66 60 56 51 47 43 41
220 82 71 63 57 53 47 44 40 38
230 79 67 60 54 50 45 41 37 36
240 76 65 57 52 47 42 39 35 33
250 73 62 54 49 45 40 37 33 31
260 70 60 52 47 43 38 35 31 29
270 68 57 50 45 41 36 33 29 27
280 66 55 48 43 39 34 31 27 26
290 64 53 46 41 38 33 30 26 24
300 62 52 45 40 36 31 28 25 23
330 DESIGN OF STEEL STRUCTURES

Table 6.1 (f) (i) Maximum permissible bending stress, σbcN/mm2 (MPa), in equal flangeI-
⎛T d ⎞
beams or channels with fy = 400 N/mm2 (MPa), ⎜ ≤ 2.0 and 1 ≤ 67 ⎟
t
⎝ w tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 253 252 250 249 249 248 248 247 247
45 251 248 246 245 244 ‘ 243 243 242 241
50 248 245 242 240 239 238 237 235 234
55 245 241 238 236 234 232 231 229 227
60 242 237 234 231 228 226 228 222 220
65 239 234 229 225 222 220 218 215 212
70 236 230 225 220 217 214 212 207 205
75 233 226 220 215 211 208 205 200 197
80 230 223 216 .210 206 202 199 193 189
85 227 219 212 205 200 196 192 186 181
90 225 215 207 201 195 190 186 179 174
95 222 212 203 196 190 185 180 172 167
100 219 208 199 191 185 179 175 166 160
110 213 202 191 183 176 169 164 154 148
120 208 195 184 175 167 160 154 144 136
130 203 189 177 167 159 152 146 134 126
140 198 183 171 160 152 144 138 126 117
150 193 178 165 154 145 137 131 118 109
160 188 172 159 148 139 131 124 111 102
170 183 167 154 142 133 125 118 105 96
180 179 172 149 137 127 119 112 99 90
190 175 158 144 132 122 114 108 94 85
200 171 153 139 128 118 110 103 90 80
210 167 149 135 123 114 105 99 86 77
220 163 145 131 119 110 102 95 82 73
230 159 141 127 115 106 98 91 79 70
240 156 138 123 112 102 94 88 75 67
250 152 134 120 108' 99 91 85 72 64
260 149 131 117 105 96 88 82 70 61
270 146 128 114 102 93 85 79 67 59
280 143 125 111 99 90 83 77 65 57
290 140 122 108 97 88 80 74 63 55
300 137 119 105 94 85 78 72 61 53
 DESIGN OF BEAMS 331

Table 6.1 (f) (ii) Maximum permissible bending stress, σbcN/mm2 (MPa), in equal flange
⎛T d ⎞
I-beams or channels with fy = 400 N/mm2 (MPa), ⎜ ≤ 2.0 and 1 ≤ 67 ⎟
⎝ tw tw ⎠

D /T →
l / ry 8 10 12 14 16 18 20 25 30
↓
40 248 247 247 246 246 246 246 246 246
45 243 242 241 241 240 240 240 240 239
50 237 235 234 234 233 233 233 232 232
55 231 229 227 227 226 225 225 225 224
60 225 222 220 219 218 217 217 216 216
65 218 215 212 211 210 209 208 207 207
70 212 207 205 203 202 200 199 198 198
75 205 200 197 195 193 191 190 189 188
80 199 193 189 186 185 182 181 180 179
85 192 186 181 178 176 174 172 171 170
90 186 179 174 171 168 165 164 162 161
95 180 172 167 163 161 157 155 153 152
100 175 166 160 156 153 150 148 145 144
110 164 154 148 143 140 135 133 130 129
120 154 144 136 131 127 123 120 117 115
130 146 134 120 141 117 111 108 105 103
140 138 126 117 111 107 102 98 95 93
150 131 118 109 103 99 93 89 86 84
160 124 111 102 96 92 85 82 78 76
170 118 105 96 90 85 79 75 71 69
180 112 99 90 84 79 73 69 65 63
190 108 94 85 79 74 68 64 60 88
200 103 90 80 75 70 63 60 85 83
210 99 86 77 70 66 59 55 51 49
220 95 82 73 67 62 56 52 48 45
230 91 79 70 64 59 53 49 44 42
240 88 75 67 61 56 50 46 42 39
250 85 72 64 58 53 47 43 39 37
260 83 70 61 55 51 45 41 37 34
270 79 67 59 53 49 43 39 35 32
280 77 65 57 51 47 41 37 33 30
290 74 63 55 49 45 39 35 31 29
300 72 61 53 47 43 37 33 29 27
332 DESIGN OF STEEL STRUCTURES

6.3 SHEAR AND BEARING STRESSES

When the beams are subjected to loads, then, these are also required to trans-
mit large shear forces either at supports or at concentrated loads. For simply
supported beams, the shear forces is maximum at the supports. The values of
shear forces at the concentrated loads also remain large. Due to shear force, the
shear stresses are set up along with the bending stresses at all sections of the
beams. The shear stress at any point of the cross-section is given by
⎛ F ⋅Q⎞
τv = ⎜ ...(6.4)
⎝ I ⋅ t ⎟⎠
where, τv is the shear stress and F is the shear force at cross-section
Q = static moment about the neutral axis of the portion of cross-
sectional area beyond the location at which the stress is being
determined.
I = Moment of inertia of the section about the neutral axis
t = Thickness of web (width of section at which the stress is being
determined).
The distribution of shear stresses for rectangular section beam and I-beam
section are shown in Fig. 6.4 (a) and Fig. 6.4. (b), respectively.

N τm a x A N τm a x A
d

R e ctan gu lar Sh ea r stre ss I-B ea m Sh ea r stre ss

b ea m
(a ) (b )

Fig. 6.4

The maximum shear stress occurs at the neutral axis of the section.
The maximum shear stress in a member having regard to the distribution of
stresses in conformity with the elastic behaviour of the member in the flexure
(bending) should not exceed the value of maximum permissible shear stress,
τvm found as follows:
τvm = 0.45 fy ...(i)
where, fy is the yield stress of structural steel to be used.
It is to note that in the case of rolled beams and channels, the design shear is
to be found as the average shear.
The average shear stress for rolled beams or channels calculated by dividing
the shear force at the cross-section of the web is defined as the depth of the
beam or channel multiplied by its web thickness.
 DESIGN OF BEAMS 333

Average shear stress for rectangular beam is given by

⎛ F ⎞
τv = ⎜ ...(6.5)
⎝ b × d ⎟⎠
Average shear stress for I-beam is given by
⎛ F ⎞
τv = ⎜ ...(6.6)
⎝ h × tw ⎟⎠
For rolled steel beams and channels, it is assumed that shear force is resisted
by web only. The portion of shear resisted by the flanges is neglected.
The average shear stress τva.cal, in a member calculated on the gross cross-
section of web (when web bucking is not a factor) should not exceed in case of
unstiffened web of the beam,
τva = 0.4 fy ...(ii)
The allowable shear stresses as per AISC, AASHTO and AREA specification
are as follows:
Specifications Allowable shear stress
AISC 0.40 fy
AASHTO 0.33 fy
AREA 0.35 fy
When the beams are subjected to co-existence bending stresses (tension or
compression) and shear stress, then the equivalent stress, σe.cal is obtained from
the following formula
2 2 1/ 2
σe.cal = [ σbt .cal + 3τvm.cal ] ...(iii)
2 2 1/ 2
σe.cal = [ σbc .cal + 3τvm.cal ] ...(iv)
The equivalent stress σe.cal due to co-existence bending (tension or compres-
sion) and shear stresses should not exceed the maximum permissible equiva-
lent stress, σe found as under
σe = 0.90fy ...(v)
When the bearing stress σp is combined with tensile or compressive bending
and shear stresses under the most unfavourable conditions of loading the equiva-
lent stress σe.cal obtained as below should not exceed
σe = 0.90fy ...(vi)
σe.cal = [σbt2 + σbt2 .cal + σbt2 .cal + 3 τvm2.cal]1/2 ...(vii)
σe.cal = [σbt2 + σbt2 .cal + σbt2 .cal + 3 τvm2.cal]1/2 ...(viii)
σbc.cal, σbt .cal,τvm.cal and σp.cal are the numerical values of the co-existence bend-
ing (compression or tension), shear and bearing stresses. When the bending
occurs about both the axes of the member,σbt .cal, and σbc.cal should be taken as
the sum of the two calculated fibre stresses,σe is the maximum permissible
equivalent stress.
Bearing stress. The bearing stress in any part of a beam when calculated
on the net area of contact should not exceed the value of σp calculated as below
σp = 0.75fy ...(ix)
where σp is the maximum permissible bearing stress and fy is the yield stress.
334 DESIGN OF STEEL STRUCTURES

6.4 EFFECTIVE SPAN AND DEFLECTION LIMITATION

The effective span of a beam shall be taken as the length between the centres of
the supports, export in cases where the point of application of the reaction is
taken as eccentric to the support, then, it shall be permissible to take the effec-
tive span as the length between the assumed points of applications of reaction.
The stiffness of a beam is a major consideration in the selection of a beam
section. The allowable deflections of beams depend upon the purpose for which
the beams are designed.
The large deflections of beams are undesirable for the following reasons:
(i) When the loads are primarily due to human occupants especially in the
case of public meeting places, large deflections result in noticeable to
the occupants.
(ii) The large deflections may result in cracking of ceiling plaster, floors or
partition walls.
(iii) The large deflection indicates the lack of rigidity. It may cause vibra-
tions and over-stresses under dynamic loads.
(iv) The large deflection may cause the distortions in the connections. The
distortions cause secondary stresses.
(v) The large deflection may cause poor drainage, which will lead to ponding
of water, and therefore increase the loads.

6.4.1 Limiting Vertical Deflection

The deflection of a member is calculated without considering the impact factor
or dynamic effect of the loads causing the deflection. The deflection of a member
shall not be such as to impact the strength of efficiency of the structure and lead
to damage to finishings.
Generally, the maximum deflection should not exceed (1/325) of the span.
This limit may be exceeded in cases where greater deflection would not impair
the strength or efficiency of the structure or lead to damage to finishings. The
deflection of the beams may be decreased by increasing the depth of beams,
decreasing the span, providing greater end restraint or by any other means.

6.4.2 Limiting Horizontal Deflection

At the caps of columns in single storey buildings the horizontal deflection due to
lateral force should not ordinarily exceed (1/325) of the actual length l of the
column. This limit may be exceeded in cases where the greater deflection would
not impair the strength and efficiency of the structure or lead to damage to
finishings.
According to AISC specification, the deflection of beams and girders for live
load and plastered ceiling should not exceed (1/360) of the span.
 DESIGN OF BEAMS 335

6.5 LATERALLY SUPPORTED BEAMS

The laterally supported beams are also called laterally restrained beams. When
lateral deflection of the compression flange of a beam is prevented by providing
effective lateral support, (restraint) the beam is said to be laterally supported.
The effective lateral restraints is the restrains which produces sufficient resis-
tance in a plane perpendicular to the plane of bending to restrain the compres-
sion flange of a beam from lateral buckling to either side at the point of applica-
tion of the restraint. The concrete slab encasing the top flange (so that the
bottom surface of the concrete slab is flush with the bottom of the top flange) is
shown in Fig. 6.5 (a). It provides a continuous lateral supports to the top flange
of the beam. When other beams frame at frequent intervals into the beam in
questions as shown in Fig. 6.5 (b), lateral support is provided at each point of
connection but main beam should still be checked between the two supports.

(a) Continuous la teral support (b) Local lateral support

Fig. 6.5 Adequate lateral support

In the laterally supported beams, the value of allowable bending compres-

sive stress remains unaltered and the reduction in its value is not made. Bend-
ing compressive stress is taken equal to the allowable bending tensile stress,
(σbc = σbt = 0.66fy). The adequate lateral support is provided to safeguard against
the lateral-torsional buckling. In case of doubt for adequate lateral support, the
beams should be designed as laterally unsupported. In case the concrete slab
holds the top flange (compression flange) of the beam from one side only, then,
the lateral support is not credited. The concrete slab simply resting over the top
flange of beam without shear connectors also does not provide a lateral support.
Sometimes, the plank or bar grating is attached to top flange of beam by means
of bolts. When the bolts are firmly fastened, then, they provide adequate lateral
support temporarily. Even then, bolts have temporary nature of connections. It
is possible that the bolts might be omitted or removed. As such, the top flange
should not be considered laterally supported fully. The beams having lateral
support from other members may buckle between points of lateral supports.
Therefore, the laterally unsupported length of beam is kept short.
Example 6.1 The effective length of compression flange of simply supported
beam MB 500, @ 0.869 kN/m is 8 m. Determine the safe uniformly distributed
336 DESIGN OF STEEL STRUCTURES

load per metre length which can be placed over the beam having an effective
span of 8 metres. Adopt maximum permissible stresses as per IS 800–1984. The
ends of beam are restrained against rotation at the bearings.
Solution
Step 1 : Permissible bending stress
MB 500, @ 0.869 kN/m has been used as simply supported beam. The effec-
tive span of beam is 8 m. The effective length of compression flange is also 8 m.
From the steel section table, the section modulus of beam
Z = 1808.7 × 103 mm3
Means thickness of compression flange,
tf = T = 17.2 mm
Thickness of web,
tw = 10.2 mm
It is assumed that the value of yield stress, fy for the structural steel of MB
500, @ 0.869 kN/mm2 (MPa).
⎛ T 17.2 ⎞
Ratio ⎜ = ⎟ = 1.686 < 2.00
⎝ tw 10.2 ⎠

⎛ dt ⎞ ⎛ h − 2h2 ⎞ ⎛ 500 − 2 × 37.95 ⎞

Ratio ⎜⎝ t ⎟⎠ = ⎜⎝ t ⎟⎠ = ⎜⎝ ⎟⎠
w w 10.2

⎛h ⎞
= ⎜ 1 ⎟ = ⎛⎜ 424.1 ⎞⎟
⎝ tw ⎠ ⎝ 10.2 ⎠
= 41.578 < 85

⎛ D⎞ ⎛ 500 ⎞
Ratio ⎜⎝ T ⎟⎠ = ⎜⎝ 17.0 ⎟⎠ = 29.07

The effective length of compression flange is 8 m

l ⎛ 0.7 × 8 × 1000 ⎞
Ratio = ⎜ ⎟⎠ = 159.1
ry ⎝ 35.2
From IS: 800–1984, from Table 6.1 (b), the maximum permissible bending
stres, for above ratios (by linear interpolation)
σbc = 65.121N/mm2 (MPa)
Step 2: Moment of resistence of beam
⎛ 88.566 × 1808.7 × 103 ⎞
Mr =(σbc × Z) = ⎜ 1000 × 1000 ⎟
⎝ ⎠
= 160.189 m-kN
MB 500, @ 0.869 kN/m can resist maximum bending moment equal to moment
of resistance
 DESIGN OF BEAMS 337

∴ Maximum bending moment.

M = 160.189 m-kN
Step 3: Load supported over beam
The effective span of the beam is 8 metres. Let w be the uniformly distrib-
uted load per metre length.
The maximum bending moment, M for the beam occurs at the centre,
⎛ w ⋅ l2 ⎞
M = ⎜ ⎟
⎝ 8 ⎠

⎛ w × 8 × 1000 ⎞
⎜⎝ 8 ⎟⎠ = 160.189

⎛ 160.189 × 8 ⎞
w = ⎜ × 1000⎟ = 20.02 kN/m
⎝ 8 × 8 × 1000 ⎠
The self-weight of the beam is 0.869 kN/m. Therefore, the safe uniformly
distrbuted load which can be placed over the team.
(20.02 – 0.869) = 19.15 kN.

6.6 DESIGN OF LATERALLY SUPPOTED BEAMS

The design of beams is generally governed by the maximum allowable bending
stress and the allowable deflection. Its design is controlled by shear only when
the spans are short and loads are heavy. The members are selected such that
the sections are symmetrical about the plane of loading and the unsymmetrical
bending and torsion are eliminated.
The design of beams deals with proportioning of members, the determina-
tion of effective section modulus, maximum deflection and the shear stress. In
general, the rolled steel sections have webs of sufficient thickness such that the
criterion for design is seldom governed by shear. The following are the usual
steps in the design of laterally supported beams:
Step 1. For the design of beams, load to be carried by the beam and effective
span of the beam are known. The value of yield stress, fy for the structural steel
to be used is also known. For the rolled steel beams of equal flanges as given in
ISI Handbook No. 1, the ratio of mean thickness of the compression flange (T =
t f) to the thickness of web used to be less than 2.00. So also, the ratio of the
depth of web d1 to the thickness of web is also smaller than 85. The ends of
compression flange of a laterally supported beam remain restrained against
lateral bending (i.e., not free to rotate in plan at the bearings).
In the beginning of design, the permissible bending stress in tension, σbt are
in compression, σbc may be assumed as below:
σbt or σbc = 0.66 fy
The bending compressive stress, σbc and the bending tensile stress, σbt are
equal for the laterally supported beam.
338 DESIGN OF STEEL STRUCTURES

Step 2. The maximum bending moment M and the maximum shear force F
in the beam are calculated. The required section modulus for the beam is deter-
mined,

⎛ M⎞
Z = ⎜
⎝ σbc ⎟⎠
Step 3. From the steel section tables, a rolled steel beam section, (which
provides more than the required section modulus) is selected. The steel beam

⎛ D⎞ ⎛ l⎞
section shall have ⎜ ⎟ and ⎜ ⎟ ratios more than 8 and 40, respectively. As
⎝T ⎠ ⎝ ry ⎠
such the trial section of beam selected may have modulus sections, Z more than
that required. Some of the beam section of different groups (categories, have
almost the same value of the section modulus Z. It is necessary to note the
weight of beam per metre length and the section modulus, Z. The beam section
selected should be such that it has minimum weight and adequate section modu-
lus, Z.
Step 4. The rolled steel beam section is checked for the shear stress. The
average and maximum shear stresses should not exceed the allowable average
and maximum values of shear stresses.
Step 5. The rolled steel beam is also checked for deflection. The maximum
deflection should not exceed the limiting deflection.
ISI Handbook No. 1 provides tables for allowable uniform loads on beams
and channels used as flexural members with adequate lateral support for com-
pression flange. The values of allowable uniform loads corresponding to respec-
tive effective spans are given for various beams and channel sections. For given
span and total uniformly distributed load found, rolled steel beam or channel
section may be selected from these tables.
The rolled steel I-sections and wide flange beam sections are most efficient
sections. These sections have excellent flexural strength and relatively good
lateral strength for their weights.
Example 6.2 Design a simply supported beam to carry a uniformly distrib-
uted load of 44 kN/m. The effective span of beam is 8 metres. The effective length
of compression flange of the beam is also 8 m. The ends of beam are not free to
rotate at the bearings.
Solution
Design :
Step 1 : Load supported, bending moment and shear forces
Uniformly distributed load = 44 kN/m
Assume self-weight of beam = 1.0 kN/m
Total uniformly distributed load
w = 45 kN/m
 DESIGN OF BEAMS 339

The maximum bending moment, M occurs at the centre

⎛ w ⋅ l2 ⎞ ⎛ 45 × 8 × 8 × 1000 ⎞
M = ⎜ ⎟ =⎜ ⎟⎠
⎝ 8 ⎠ ⎝ 8 × 1000
= 360 m-kN
The maximum shear forces, F occurs at the support
⎛ w ⋅ l ⎞ ⎛ 45 × 8 ⎞
F = ⎜⎝ 8 ⎟⎠ = ⎜⎝ 2 ⎟⎠ = 180 kN

Step 2: Permissible bending stress

It is assumed that the value of yield stress, fy for the structural steel is 250 N/
⎛T ⎞ ⎛d ⎞
mm2 (MPa). The ratio ⎜ ⎟ and ⎜ 1 ⎟ are less than 2.0 and 8.5, respectively.
⎝ tw ⎠ ⎝ tw ⎠
The maximum permissible stress in compression or tension may be assumed as
below:
σbc = σbt = (0.66 × 250)
= 165 N/mm2
Section modulus required,

⎛ M ⎞ ⎛ 360 × 1000 × 1000 ⎞

Z = ⎜ =⎜ ⎟⎠
⎝ σbc ⎠⎟ ⎝ 165
= 2181.82 × 103 mm3

⎛ D⎞ ⎛ l⎞
The steel beam section shall have ⎜ ⎟ and ⎜ ⎟ ratios more than 8 and 40,
⎝T ⎠ ⎝ ry ⎠
respectively. The trial section of beam selected may have modulus of sections, Z
more than that needed, about 25 to 50 per cent more).
Step 3: Trial section modulus
1.50× 2181.82 × 103 mm3 = 3272.73 × 103 mm3
From steel section tables, try
WB 600, @ 1.337 kN/m
Section modulus, Zxx = 3540.0 × 103 mm3
Moment of inertia, lxx = 106198.5 × 104 mm4
Thickness of web, t w = 11.2 mm
Thickness of web, T = tf = 21.3 mm
Depth of section, h = 600 mm
Step 4 : Check for section modulus

⎛ D 600 ⎞
⎜⎝ T = 21.3 ⎟⎠ = 28.169
340 DESIGN OF STEEL STRUCTURES

⎛ T 21.3 ⎞ ⎛d ⎞
⎜⎝ t = 11.2 ⎟⎠ = 1.901 < 2.0, also ⎜ 1 < 85⎟
w t
⎝ w ⎠
The effective length of compression flange of beam is 8 m.

⎛ l⎞ ⎛ 0.7 × 8 × 10.0 ⎞
⎜ r ⎟ = ⎜⎝ 52.5 ⎟⎠ = 106.66
⎝ y⎠
From Table 6.1 (b), IS 800–1984, maximum permissible bending stress
σbc = 118.68 N/mm2 (MPa)
Section modulus required
⎛ 360 × 1000 × 1000 ⎞
⎜⎝ 118.68 ⎟⎠ = 3033.34 × 103 mm3

< 3540 × 103 mm3 provided

Further trial may give more economical section.
Step 5: Check for shear force
⎛ F ⎞
Average shear stress, τv.cal = ⎜ h × t ⎟
⎝ w⎠

⎛ 180 × 1000 ⎞
τv.cal = ⎜⎝ = 26.78 N/mm2
600 × 11.2 ⎟⎠
Permissible average shear stress
0.4 × fy = (0.4 × 250) = 100 N/mm2
> Actual average shear stress
Step 6: Check for deflection
Maximum deflection of the beam

⎛ 5 w.l4 ⎞
ymax = ⎜ 384 ×
⎝ EI ⎟⎠

5 ⎛ 4.5 × 8 × 83 × (1000)3 × 1000 ⎞

ymax = × ⎜ 5 4⎟
384 ⎝ 2.047 × 10 × 106198.5 × 10 ⎠
= 24.53 mm
Allowable deflection

1 ⎛ 1 ⎞
× span = ⎜ × 8000⎟
325 ⎝ 325 ⎠
= 24.60 mm
The maximum deflection is less than allowable deflection, hence the beam is
safe. Provide WB 600, @ 1.337 kN/m.
 DESIGN OF BEAMS 341

Example 6.3 A class-room is 7 m × 15 m. It is provided with 120 mm thick

stone patties over rolled steel beams spaced 3 m centre to centre. A wearing coat
of 20 mm thick cement concrete is provided over 160 mm thick lime concrete. The
effective length of compression flange would be equal to the effective span of
beam. Design rolled steel beams. The ends of beams shall not be free to rotate at
the bearings.
Live load = 4 kN/ m2
Unit weight of stone patties = 24 kN/m3
Unit weight of plain cement concrete = 24 kN/m3
Unit weight of lime concrete = 18 kN/m3
Solution
Design :
Step 1 : Load supported, bending moment and shear force
The rolled steel beam would carry a load from a strip of 3 m width.
Uniformly distributed load per metre length of beam.

⎛ 120 ⎞
Weight of stone patties = ⎜⎝ 3 × 1 × × 24⎟ = 8.64 kN
⎠
1000

⎛ 20 ⎞
Weight of plain cement concrete ⎜⎝ 3 × 1 × × 24⎟ = 1.44 kN
⎠
1000

⎛ 160 ⎞
Weight of lime concrete ⎜⎝ 3 × 1 × × 18⎟
⎠ = 8.64 kN
1000
Assume self-weight of beam = 0.60 kN
Live load on the floor = 3 × 4 = 12.00 kN
Total uniformly distributed load = 31.32 kN/m
Say = 31.50 kN/m
Clear span of beams = 7 m
Assume, 300 mm wide bed block would be provided.
The effective span of beam is adopted as the distance centre to centre of
bearing or clear span + depth of beam
Adopt effective span = (7000 + 300) mm = 7.30 m
The maximum bending moment, M occurs at the centre,

⎛ w ⋅ l2 ⎞ ⎛ 31.50 × 7.30 × 1000 ⎞

M =⎜ ⎟ = ⎜⎝ ⎟⎠
⎝ 8 ⎠ 8 × 1000

= 209.829 m-kN
The maximum shear, F occurs at the support,

⎛ w ⋅l⎞ ⎛ 31.50 × 730 ⎞

F= ⎜ ⎟ = ⎜⎝ ⎟⎠ = 114.975 kN
⎝ 2 ⎠ 2
342 DESIGN OF STEEL STRUCTURES

Step 2 : Permissible bending stress

It is assumed that the value of yield stress, fy for the structural steel is 250 N/
⎛T ⎞ ⎛d⎞
mm2 (MPa). The ratios ⎜ ⎟ and ⎜ ⎟ are less han 2.0 and 85, respectively.
⎝ tw ⎠ ⎝ tw ⎠
The maximum permissible stress may be assumed as below.
σbc = σbt= (0.66 × 250) = 165 N/mm2
Step 3 : Section modulus required
⎛ M⎞ ⎛ 209.829 × 1000 × 1000 ⎞
Z = ⎜ σ ⎟ = ⎜⎝ ⎟⎠
⎝ bc ⎠ 165
= 127.69 × 103 mm3

⎛ D⎞ ⎛ l⎞
The steel beam section shall have ⎜ ⎟ and ⎜ ⎟ ratios more 8 and 40,
⎝T ⎠ ⎝ ry ⎠
respectively. The trial section of beam selected may have modulus of section
more than that required (about 25 to 50 per cent more).
Trial section modulus
1.50 × 1271.69 × 103 = 1907.535 × 103 mm3
Step 3 : Properties of trial section
From steel section tables, try MB 550, @ 1.037 kN/m
Section modulus provided, Zxx = 235.8 × 103 mm3
Moment of inertia, Zxx = 235.8 × 104 mm4
Thickness of web, t w = 11.2 mm
Depth of section, h = 600 mm
Means thickness of flange, t f = 19.3 mm
Step 5 : Check for section modulus
⎛ D 600 ⎞
⎜⎝ T = 19.3 ⎟⎠ = 31.09

⎛ T 19.3 ⎞ ⎛ d1 ⎞
⎜⎝ t = 11.2 ⎟⎠ = 1.723 and also ⎜⎝ t < 85⎟⎠
w w

The effective length of compression flange of beams is 7.30 m (equal to effective

span of beam)
⎛ l⎞ ⎛ 0.7 × 7.30 × 1000 ⎞
⎜r ⎟ = ⎜⎝ ⎟⎠ = 136.99
⎝ y⎠ 37.3
From Table 6.1 (b), IS: 800–1984, maximum permissible bending stress
σbc = 98.086 N/mm2 (MPa)
Section modulus required
⎛ 209.829 × 1000 × 1000 ⎞
= ⎜⎝ ⎟⎠
98.086
 DESIGN OF BEAMS 343

= 2139.235 × 103 mm3

< 2359.8 × 103 mm3 provided
Further trial may give more economical section.
Step 6 : Check for shear stress
Average shear stress
⎛ F ⎞ ⎛ 114.975 × 1000 ⎞
τva.cal = ⎜ h × t ⎟ = ⎜⎝
⎝ w⎠ 600 × 11.2 ⎟⎠
= 17.11 N/mm2
< (0.4 × 250 = 100 N/mm2). Hence safe.
Step 7 : Check for deflection
Maximum deflection of beam
⎛ 5 × 31.50 × 7.30 × (7.30)3 × (1000)3 × 1000 ⎞
ymax = ⎜⎜ 384 × 2.047 × 105 × 91813 × 104
⎟
⎟⎠
⎝
= 6.197 mm

⎛ 1 ⎞
Allowable deflection = ⎜⎝ × 7300⎟ = 22.45 mm
325 ⎠
The maximum deflection is less than allowable deflection. Hence, the section
of beam adopted is safe. Provide MB 550, @ 1.037 kN/m.
Example 6.4 The floor of an assembly hall is supported by main beams and
secondary beams as shown in Fig. 6.6. The secondary beams would be connected
to the web of main beams. The floor consists of 100 mm thick reinforced concrete
slab. Design the beams.
Solution
Design :
The unit weight of reinforced concrete slab is 24 kN/m3, and the live load on
the floor is 5 kN/m2.
Step 1: (i) Design of secondary beams
Each secondary beam supports load from strip 2 m wide. Uniformly distrib-
uted load per metre length of the beam:
1. Load supported
Weight of reinforced concrete slab

⎛ 100 ⎞
= ⎜2 ×1 × × 24⎟ = 4.80 kN
⎝ 1000 ⎠
Live load on the floor = (2 × 1 × 5) = 10.00 kN
Assume self-weight of the beam = 0.50 kN
Total uniformly distributed load = 15.30 kN/m say, 15.50 kN/m
2. Bending moment and shear force
The effective span of the beam is 5 m
344 DESIGN OF STEEL STRUCTURES

The maximum bending moment, M occurs at the centre,

⎛ w ⋅ l2 ⎞ ⎛ 15.50 × 5 × 5 × 10.00 ⎞
M = ⎜ ⎟= ⎜ ⎟⎠
⎝ 8 ⎠ ⎝ 8 × 100

= 48.438 m-kN
The maximum shear force, F occurs at the support,

⎛ w ⋅l⎞ ⎛ 15.50 × 5 ⎞
F = ⎜ ⎟ = ⎜⎝ ⎟⎠ = 38.75 kN
⎝ 2 ⎠ 2

S e co n da ry be am s
M ain be am s

Fig. 6.6

3. Permissible bending stress

It is assumed that the value of yield stress fy for the structural steel is 250 N/
⎛T ⎞ ⎛d ⎞
mm2 (MPa). The ratio of ⎜ t ⎟ and ⎜ 1 ⎟ are less than 2.0 and 85, respectively.
⎝ w⎠ ⎝ tw ⎠
The maximum permissible stress in compression or tension may be assumed as
below
σbc = σbt = (0.66 × 250)
= 165 N/mm2
4. Section modulus required
⎛ M⎞ ⎛ 48.438 × 1000 × 1000 ⎞
Z = ⎜ σ ⎟ = ⎜⎝ ⎟⎠
⎝ bc ⎠ 165
= 293.56 ×103 mm3

⎛ D⎞ ⎛ l⎞
The steel beam section shall have ⎜ ⎟ and ⎜ ⎟ ratios more than 8 and 40,
⎝T ⎠ ⎝ ry ⎠
respectively. The trial section of beam selected may have modulus of section, Z
1.5 times more than that needed.
 DESIGN OF BEAMS 345

Trial section modulus

1.50 × 293.56 × 103 = 440.345 × 103 mm3
5. Check for section modulus
⎛ D 300 ⎞
⎜⎝ T = 9.4 ⎟⎠ = 31.915

⎛ T 9.4 ⎞ ⎛ d1 ⎞
⎜⎝ t = 6.7 ⎟⎠ = 1.043 < 2.00, also ⎜⎝ t < 85⎟⎠
w w

The effective length of compression flange of beam may be assumed equal to

effective span,
⎛ l⎞ ⎛ 0.7 × 5 × 1000 ⎞
⎜r ⎟ = ⎜⎝ ⎟⎠ = 125
⎝ y⎠ 28

From Table 6.1 (b), IS: 800–1984, maximum permissible bending stress
σbc = 104.66 N/mm2
Section modulus required
⎛ 48.438 × 1000 × 1000 ⎞
Z = ⎜⎝ ⎟⎠
104.66
= 462.815 × 103 mm3
< (488.9 × 103 mm3) provided.
6. Properties of trial section
From steel section tables, try LB 300. @ 0.377 kN/m
Section modulus provided, Zxx = 488.9 × 103 mm3
Moment of inertia, Ixx = 7322.9 × 104 mm4
Thickness of web, t w = 6.7 mm
Depth of section, h = 300 mm
Means thickness of flange, t f = 9.4 mm
7. Check for shear force

⎛ F ⎞ ⎛ 38.75 ⎞
Average shear stress = ⎜⎝ h.t ⎟⎠ = ⎜⎝ 300 × 6.6 ⎟⎠ = 19.29 N/mm2
w

< (0.4 × 250 = 100 N/mm2). Hence, safe.

8. Check for deflection

5 ⎛ wl4 ⎞
ymax = 384 . ⎜ ⎟
⎝ EI ⎠

( )
⎛ 5 × 15.50 × 5 × 53 × 10003 × 1000 ⎞
ymax = ⎜⎜ 384 × 2.047 × 105 × 7332.9 × 104 ⎟⎟
⎝ ⎠
346 DESIGN OF STEEL STRUCTURES

= 8.4 mm

⎛ 1 ⎞
Allowable deflection = ⎜ × 5000⎟ = 15.39 mm
⎝ 325 ⎠
The maximum deflection is less than allowable deflection.
Hence, design is satisfactory.
Step : 2 (ii) Design of main beams
1. Load supported
Clear span = 12 m
Provide bed blocks = 300 mm wide
The effective span is taken as the distance between centre to centre of bearings.
Therefore,
Effective span = 12.30 m
Load transferred from each secondary beam
= (15.50 × 5 ) = 7.50 kN
Assune self-weight of beam = 2 kN/m
2. Bending moment
The maximum bending moment (occurs at centre) due to uniformly distrib-
uted load
⎛ w ⋅ l2 ⎞ ⎛ 2 × 12.30 × 12.30 × 1000 ⎞
M1 = ⎜ ⎟ =⎜ ⎟⎠
⎝ 8 ⎠ ⎝ 8 × 1000
= 37.8225 m-kN
End reaction due to concentrated loads
= (2.5 × 77.500) = 193.75 kN
The maximum bending moment (occurring at centre) due to concentrated
loads
M1 = (193.75 × 6.15 – 77.50 × 4 – 77.50 × 2)
= 726.5625 m-kN
Total bending moment
= (726.5625 + 37.8225)
= 764.384 m-kN
The secondary beams are connected to the web at 2 m centre to centre. The
compression flange is assumed to the fully supported against lateral deflection.
3. Permissible bending stress
It is assumed that the value of yield stress, fy for the structural steel is 250 N/
⎛T ⎞ ⎛d ⎞
mm2 (MPa). The ratio ⎜ ⎟ and ⎜ 1 ⎟ are less than 2.0 and 85, respectively.
⎝ tw ⎠ ⎝ tw ⎠
The maximum permissible stress in compression or tension may be assumed as
under (for lateral laterally supported beam)
σbc = σbt = (0.66 × 250 ) = 165 N/mm2 (MPa)
 DESIGN OF BEAMS 347

4. Section modulus required

⎛ 764.3850 × 103 × 103 ⎞

Z = ⎜ 165 ⎟
⎝ ⎠
= 4632.6 × 103 mm3
From steel section tables (Tale XIV). try WB 500, @ 2.522 kN/m with 25 mm
thick plate in each flange, 250 mm wide.
Section modulus provided
Zxx = 6916.6 × 103 mm3
Moment of inertia of plated beam
lxx = 190207.6 × 104 mm4
Thickness of web, t w = 9.9 mm
Depth of section, h = (500 + 2 × 25) = 550 mm
Total load on the girder inclusive of its own weight
= (15.50 × 5 × 5 + 2.522 × 12.30)
= 418.52 kN
5. Maximum shear force
= 209.26 kN
6. Check for average shear stress
Average shear stress
⎛ 209.26 × 1000 ⎞
τva.cal = ⎜
⎝ 500 × 9.2 ⎟⎠
= 42.27 N/mm2
< (0.4 × 250 = 100 N/mm2). Hence, safe.
7. Check for moment
⎛ 165 × 6916.6 × 103 ⎞
σbt.Z = ⎜⎝ ⎟⎠ kN-m
1000 × 1000
= 1141.14 kN-m
< 764.385 kN-m. Hence, safe.
Provide MB 500, @ 0.869 kN/m with 25 mm thick and 250 mm wide plate in
each flange. This section may be checked for deflection. It will be found safe.

6.7 LATERALLY UNSUPPORTED BEAMS

The laterally unsupported beams are also termed as laterally unrestrained beams.
The symmertrical beams even if loaded in the plane containing principal axis
may bend out of the plane of load under certain conditions. The rolled steel
beams and channels have moment of inertia about xx-axis. Ixx (i.e., the other
principal axis). The shapes of beas are made economical in the manner. Such
shapes have less resistance to bending and torsion in the direction of yy-axis
(i.e., weak axis). The beams may bend in the weak direction if it is not held or
supported in a direction normal to the weaker axis. The beam is weakend further
348 DESIGN OF STEEL STRUCTURES

if a horizontal load is applied in the direction normal to yy-axis. The bending of

beams is accompanied by twisting.This phenomenon is termed as lateral buckling
or lateral torisonal buckling. When the torsional resistance and moment of inertia
of any beam for the weak axis are small as compared with the moment of inertia
about the strong axis, the beam may buckle laterally wheather the beams are
symmetrical or not. The buckling tendency of compression flange increases as
the stress increases and this tendency is maximum when the stress attains the
value of critical stress. The compression flange buckles side-ways or laterally
and it is also accompanied by twisting. The hypothetical buckling of a beam is
shown in Fig. 6.7. The tendency of lateral buckling of compression flange also
increases as the ratio (Ixx/Iyy) increases. This tendency further increases when
the supported length of compression flange increases.

(a ) S e ctio n at m id-sp an (b ) P lan o f com p ressio n fla ng e la tera l bu cklin g

Fig. 6.7

In order to avoid lateral buckling due to a load in the plane of one of the
principal axis of the member, the beam should have enough torsional stiffness
to resist torsional forces induced and the beam should be laterally supported in
the direction perpendicular to the weaker principal axis. The compression flange
of beam tends to buckle if not held in line by attached construction. Sometimes,
the special conditions may require that the beams may be loaded in the plane of
the web without continuous or intermittent lateral suport at sufficient frequent
intervals.
The bending stress in compression for sectional shapes with Iyy smaller than
Ixx (where Iyy is moment of inertia of the whole section about the axis lying in
the plane of bending. yy axis, and Ixx is moment of inertia of the whole section
about the axis normal to the plane of bending, xx-axis) is reduced for lateral
buckling in proportion to elastic critical stress, fcb found as follows for sections
of single web.

6.7.1 Where the Flanges have Equal Moment of Inertia about yy-Axis
If symmetrical I-beam is subjected to couples M, at the ends and beam is
restraints against torsion at the ends, the lateral buckling of beam takes place
when the value of M attains critical value. The critical value of moment M is
given by
1/ 2
⎛ π⎞ ⎡ ⎛ C1 .π2 ⎞ ⎤
Mcr = ⎜⎝ ⎟⎠ ⎢ ElyC. ⎜1 + ⎟⎥ ... (6.7)
l ⎢⎣ ⎝ C2 .l2 ⎠ ⎥⎦
 DESIGN OF BEAMS 349

where, l = Effective length of compression flange

C = Torsional rigidity of beam = GK
G = Shear modulus of elasticity
E = Young’s modulus of elasticity
K = Torsional constant of beam
C1 = Warping rigidity = ECw
⎛1 3 2 1 2⎞
Cw = Warping constant = ⎜⎝ tf b h = I f ⋅ h ⎟⎠
24 2

1 1
= I y ⋅ h2 (considering If = I f ).
4 2
If = Moment of inertia of compression flange about yy-axis of the
grider
I = Moment of inertia of beam about yy-axis at the point of maxi-
mum bending moment
b = Width of flange of I-beam
tf = Thickness of flange
h = Distance between the centre of gravity of compression flange ad
centre of gravity of tension flange. It may be taken approxi-
mately as depth of section minus thickness of flange, ( h – t f).
Substituting the values of C and C1 in Eq. 6.7
1
2
⎛ π⎞ ⎡ 1 4GK ⋅ l2 ⎤
Mcr = ⎜⎝ ⎟⎠ ⎢1 + 2 . ⎥ ...(6.8)
l ⎢ π EI ⋅ h 2
⎥⎦
⎣ y

⎛ π2 EI y ⋅ h2 ⎞
By rearranging the terms and taking the terms ⎜ ⎟ common from
⎝ 4GK ⎠
the parenthesis, Eq. 6.8 is written as follows:
1/2
π2 EI y h ⎡ 1 4GK ⋅ l 2 ⎤
Mcr = ⋅ ⎢1 + 2 ⋅ ⎥ ...(6.9)
2 l2 ⎢⎣ π EI y ⋅ h2 ⎥⎦
From Eq. 6.8 it may be observed that the critical bending moment depends
not so much on the material, since value of E is constant for all steel, but on
torsional rigidity, GK and elastic stiffiness, EIy for the bending in the perpen-
dicular to the load.
Subustituting.
G = 0.4 E,
⎛ 2 × t f ⋅ bf3 ⎞
K = (0.9 bf . bf3 ) and Iy = ⎜ ⎟
⎝ 12 ⎠
350 DESIGN OF STEEL STRUCTURES

where t f and bf are the thickness and width of the flange, respectively.
1
⎡ 2
1 4 × 0.4 E × 0.9bf tf l2 ⎤
3
π2 EI y h
Mcr = . 2 ⎢1 + 2 × . ⎥ ...(i)
2 l ⎢⎣ π E × 2 × tf × bf 2 /12 h2 ⎥⎦

1
π2 EI y h ⎡ 8.64 tf 2 l2 ⎤ 2
Mcr = . 2 ⎢1 + 2 × 2 . 2 ⎥ ...(ii)
2 l ⎢⎣ π bf h ⎥⎦
For an I-section and I-section of wide flanges, the approximate radii of gyra-
tions are given by ry = 0.22b (viz., b = 4.545 ry ) and ry = 0.25 b (viz., b = 4.000 ry),
respectively. As an approximation bf @ 4.2 ry), then
1
2
π ⋅ EI yh ⎡⎛
2 8.64tf 2 l2 ⎞ ⎤
Mcr = ⎢ ⎜ 1 + . ⎟⎥ ...(iii)
2l2 ⎢⎝⎜ π 2
× 17.64 r 2
h2 ⎠⎟ ⎥
⎣ y ⎦
1
⎡⎛ ⎤ 2
1 ⎛ l ⋅ tf ⎞ ⎞ ⎥
2 2
π ⋅ EI yh ⎢
Mcr = ⎜ 1 + ⎜ ⎟ ⎟ ...(iv)
2l2 ⎢⎜⎝ 20 ⎜⎝ ry ⋅ h ⎟⎠ ⎟⎠ ⎥
⎣ ⎦
The elastic critical for bending for beams with Iy smaller than Ixx is given by
(∴ Mcr = fcb. Zxx)
1
⎡ 2 2
⎛ l ⋅ tf ⎞ ⎤
π2E EI h ⎢1 + 1 ⎥
fcb = · ⎢ 20 ⎜ r ⋅ h ⎟ ⎥ ...(v)
2.Z xx .l2 ⎣ ⎝ y ⎠
⎦
The values of modulus of elasticity, E is taken as 2 × 105 N/mm2 (ΜPa) and
Iy = A ⋅ ry2 . Substituing these values in the expression (v)
1
2 2
9.8526 × 105 A ⋅ h ⎡ ⎛ l ⋅ tf ⎞ ⎤
fcb = ⎢1 + 1 ⎜ ⎥ ...(vi)
(
2
)
l / ry Z xx ⎢ 20 ⎝ ry ⋅ h ⎟⎠ ⎥
⎣ ⎦

⎛ A ⋅ h⎞
The value of ⎜ for I- section be taken as 2.688 apporximately.
⎝ Z xx ⎟⎠
Then
2 1/ 2
26.5 × 105 ⎡ 1 ⎛ l ⋅ tf ⎞ ⎤
fcb = ⎢1 + ⎥ ...(6.10)
⎛ l⎞
2
⎢ 20 ⎜⎝ r ⋅ h ⎟⎠ ⎥
⎣ ⎦
⎜r ⎟
⎝ y⎠
The value of critical stress, Cs is increased by 20 per cent for rolled steel
beams, channel and plate girders provided
 DESIGN OF BEAMS 351

⎛ te ⎞ ⎛ d1 ⎞
⎜⎝ t ⎟⎠ >/ 2 and ⎜⎝ t ⎟⎠ >/ 85
w w

where l = Effective length of compression flange

ry = Radius of gyration about yy-axis of the gross-section of the whole
girder at the point of maximum bending moment
h = Overall depth of the girder, at the point of maximum bending
moment
tf = Effective thickness of the compression flange
= k 1 × Mean thickness of the horizontal portion of the compres-
sion flange at the point of maximum bending moment.
k 1 makes the allowance for reduction in thickness and breadth of flanges
between points of effective lateral restraints and depends on ψ the ratio of the
total area of both flange at the point of least bending moment to the correspond-
ing point of greater bending moment between such points of restraints.
The values of k 1 have been given in Table 6.2.

Table 6.2 Values of k1 for Beams with curtailed flanges (As per IS: 800–1984)

ψ 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
k1 1.0 1.0 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.3

Note : The flanges shall not be reduced in width to give a value y lower than 0.25
where the value of ψ calculated for compression flange alone is smaller than that
when both the flanges are combined, this smaller value of ψ shall be used.
The values of critical bending stress fcb are increased by per cent for all rolled
⎛T ⎞
beam and channels and built-up beams for which ⎜ ⎟ is not greater than 2 and
⎝ tw ⎠

⎛ d1 ⎞ ⎛ 1344 ⎞
⎜⎝ t ⎟⎠ is not greater than ⎜⎜ f ⎟⎟ where, T is the mean thickness of the compression
w ⎝ y ⎠
flange and d1 is the clear depth of the web. The effective thickness of compression
flange is equal to k1 times the mean thickness of the horizontal portion of the com-
pression flange at the point of maximum bending moment. The mean thickness of
flange may be found by dividing area of horizontal portion of the compression flange
by width of flange.
In order to provide safety against lateral buckling, allowable bending com-
pressive stress σbc is reduced in proportion to critical bending stress fbc at which
lateral buckling of compression flange occurs.
The critical bending stress in compression for section symmertical about the
xx-axis may also be found from the following expression.
12
π ⎡ ELyGK ⎛ C π2 ⎞ ⎤
fcb = ⎢ ⎜ 1+ 1 2 ⎟⎥ ...(6.11)
Zx ⋅ l ⎢ µ ⎝ C2 l ⎠ ⎥⎦
⎣
352 DESIGN OF STEEL STRUCTURES

⎛ Ix − I y ⎞ ⎛ Ix − I y ⎞
where, µ= ⎜ L ⎟⎠ ⎜⎝ I ⎟⎠
⎝ z x

Ix = Moment of inertia of whole section about xx-axis

The values C1 and C are substituted, then
12
π ⎡ ELyGK ⎛ π2 ECw ⎞ ⎤
fcb = ⎢ . ⎜1 + 2 ⎟ ⎥ ...(6.12)
Zx ⋅ l ⎢⎣ µ ⎝ l GK ⎠ ⎥⎦
The values of E,G and Cw mentioned above may be subsititued. For I-sec-
tion, Eq. (6.14) reduces to
12
4.094 × 105 ⎡ ⎛ I f ⋅ h2 ⎞ ⎤
fcb = ⎢ I y K ⎜ 1 + 12.3 ⎟⎥ N/mm2 ...(6.13)
Zz ⎢⎣ ⎝ Kl ⎠ ⎥
⎦
For channels, Eq. (6.13) gives conservative values since value of warping
constant for channels is more than that for I-sections.
The values of torsional constant ‘K’ for Indian Standard rolled steel beams
and Indian Standard rolled steel channels sections are given in Appendix E in
IS: 800–1962. For the sections composed of approximately rectangular sections
⎛ b´t3 ⎞
K = ∑⎜ 3 ⎟
⎝ ⎠
Where, b´ and t´ are the width and average thickness of the rectangular
components of the section, respectively.
For box members, conservative values of ‘fcb’ are obtained. The value of K is
given by
⎛ 4A 2 ⎞
K = ⎜ approximately
S ⎟
⎜ ⎟
⎝ t ⎠
where, A = Total enclosed area of section
S = Lengthalong the periphery along which t is constant
t = Thickness of box member
t = Effective thickness (in the case of curtailed flanges)
2b2 ⋅ d2t
K= approximately
(b + d)
where, b = Width of box members
d = Depth of box members
For a plate of flat in bending in a plane parallel to its surface, the value of
elastic critical stress in bending is given by

⎛ 4.094 × 105 ⋅ t 2 ⎞
fbc = ⎜ ⎟ N/mm
2 ...(6.14)
⎝ l⋅D ⎠
 DESIGN OF BEAMS 353

where t = Thickness of plate or flat

D = Depth of the plate
t = Effective length of part in compression.

6.7.2 Maximum Permissible Bending Compressive Stress

in Beams and Plate Girders
For the beams and plate girders which are likely to bend about the axis of
maximum strength (xx-axis),the maximum bending compressive stress on the
extreme fibre, determined by using the effective section σbc.val should not exceed
the maximum permissible bending compressive stress σbc in N/mm2 (MPa)
obtained by using the following formula as recommended by IS: 800–1984.

⎛ fcb ⋅ fy ⎞
σbc = ⎜ 0.66 1/n ⎟ ...(6.15)
⎜
⎝ ⎢⎣ cb ( )
⎡( f )n + f n ⎤ ⎟
y ⎥⎦ ⎠
where, fcb = Elastic critical stress in bending, calculated as described below
or an elastic flexural-torsional buckling analysis in N/mm2 (MPa)
fy = Yeild stress of the steel in N/mm2 (MPa) and
n = A factor asumed as 1.4
The values of maximum permissible bending compressive stress,σbc in
N/mm2 (MPa) calculated from the above formula recommended by IS: 800–1984
for the structural steels (having the value of yield stress, fy as 220, 230, 240,
250, 260, 280, 300, 320, 340, 360, 380, 400, 420, 510, and 540 N/mm2 (MPa) are
given in Tables 6.3 (i) and (ii).
Table 6.3 (i) Values of σbc calculated from fcb for different value of fy All units in N/mm2
(MPa)

fy →
200 230 240 250 260 280 300 320 340
fcb ↓
20 13 13 13 13 13 13 13 13 13
30 19 19 19 19 19 19 19 19 19
40 25 25 25 25 25 25 25 25 26.
50 30 30 31 31 31 31 31 31 31
60 36 36 36 36 36 37 37 37 37
70 41 41 41 41 42 42 42 43 43
80 45 46 46 46 47 47 48 48 48
90 50 50 51 51 51 52 53 53 54
100 54 54 55 55 56 57 57 58 59
110 58 58 59 60 60 61 62 63 64
120 61 62 63 64 64 65 67 67 68
130 65 66 67 67 68 70 71 72 73
Contd.
354 DESIGN OF STEEL STRUCTURES

Table 6.5 (i) Contd.

fy →
200 230 240 250 260 280 300 320 340
fcb ↓
140 68 69 70 71 72 73 75 76 77
150 71 72 73 74 75 77 79 80 81
160 74 75 77 78 79 81 82 84 85
170 77 78 80 81 82 84 86 88 89
180 79 81 82 84 85 87 89 91 93
190 82 84 85 87 88 90 93 95 97
200 84 86 88 89 91 93 96 98 100
210 86 88 90 92 93 96 99 101 103
220 89 90 92 94 96 99 102 104 106
230 90 93 94 96 98 101 104 107 110
240 92 94 97 88 100 104 107 110 113
250 94 96 99 101 103 106 110 113 115
260 96 98 100 103 105 108 112 115 110
270 97 100 102 104 107 111 114 118 121
280 99 101 104 106 108 113 116 120 123
290 100 103 105 108 110 115 119 122 126
300 102 104 107 110 112 116 121 125 128
310 103 106 108 111 114 118 123 127 130
320 104 107 110 113 115 120 125 129 133
330 105 108 111 114 117 122 126 131 135
340 106 110 113 115 118 123 128 133 137
350 108 111 114 117 120 125 130 134 139
360 109 112 115 118 121 126 131 136 141
370 110 113 116 119 122 128 133 138 143
380 111 114 117 120 123 129 135 140 144
390 111 115 118 121 125 130 136 141 146
400 112 116 119 122 126 132 137 143 148
420 114 118 121 124 128 134 140 146 151
440 115 119 123 126 130 136 142 148 154
460 117 121 124 128 132 138 145 151 157
480 118 122 126 130 133 140 147 153 159
500 119 123 127 131 135 142 149 155 162
520 120 125 129 133 136 144 151 158 164
Contd.
 DESIGN OF BEAMS 355

Table 6.3 (i) Contd.

fy →
200 230 240 250 260 280 300 320 340
fcb ↓
540 121 126 130 134 138 145 153 160 166
560 122 127 131 135 139 147 154 161 168
580 123 128 132 136 140 148 156 163 170
600 124 129 133 137 141 150 157 165 172
620 125 129 134 138 143 151 159 166 174
640 126 130 135 139 144 152 160 168 175
660 126 131 136 140 145 153 161 169 177
680 127 132 136 141 145 154 163 171 178
700 128 132 137 142 146 155 164 172 180
720 128 133 138 143 147 156 165 173 181
740 129 134 139 143 148 157 166 174 182
760 129 134 139 144 149 158 167 175 184
780 130 135 140 145 149 159 168 176 185
800 130 135 140 145 150 159 169 177 186
850 131 137 142 147 152 161 171 180 188
900 132 138 143 148 153 163 172 182 191
950 133 138 144 149 154 164 174 183 193
1000 134 139 145 150 155 165 175 . 185 195
1050 135 140 145 151 156 167 177 187 196
1100 135 141 146 152 157 168 178 188 198
1150 136 141 147 152 158 168 179 189 199
1200 136 142 147 153 159 169 180 190 200
1300 137 143 149 154 160 171 182 192 203
1400 138 144 149 155 161 172 183 194 205
1500 139 144 150 156 162 173 184 195 206
1600 139 145 151 157 163 174 185 197 208
1700 140 146 151 157 163 175 186 198 209
1800 140 146 152 158 164 176 187 199 210
1900 140 146 152 158 164 176 188 200 211
2000 141 147 153 159 165 177 189 200 212
2200 141 147 154 160 166 178 190 202 213
2400 142 148 154 160 166 179 191 203 215
2600 142 148 154 161 167 179 191 204 216

Contd.
356 DESIGN OF STEEL STRUCTURES

Table 6.3 (i) Contd.

fy →
200 230 240 250 260 280 300 320 340
fcb ↓
2800 142 149 155 161 167 180 192 204 216
3000 143 149 155 161 168 180 193 205 217
3500 143 149 156 162 168 181 194 206 218
4000 143 150 156 163 169 182 194 207 219
4500 144 150 157 163 169 182 195 208 220
5000 144 150 157 163 170 183 195 208 221
5500 144 151 157 163 170 183 196 208 221
6000 144 151 157 164 170 183 196 209 222

Table 6.3 (ii) Values of σbc calculated from fcb for different value of fy All units in N/mm2
(MPa)

fy →
340 360 380 400 420 450 480 510 540
fcb ↓
20 13 13 13 13 13 13 13 13 13
30 19 19 19 19 19 19 20 20 20
40 26 26 26 26 26 26 26 26 26
50 31 32 32 32 32 32 32 32 32
60 37 37 38 38 38 38 38 38 38
70 43 43 43 44 44 44 44 44 44
80 48 49 49 49 49 50 50 50 50
90 54 54 54 55 55 55 56 56 56
100 59 59 60 60 60 61 61 62 62
110 64 64 65 65 66 66 67 67 67
120 68 69 70 70 71 71 72 72 73
130 73 74 74 75 76 76 77 78 78
140 77 78 79 80 80 81 82 83 84
150 81 82 83 84 85 86 87 88 89
160 85 87 88 89 90 91 92 93 94
170 89 91 92 93 94 95 97 98 99
180 93 94 96 97 98 100 101 102 103
190 97 98 100 102 102 104 106 107 108
200 100 102 103 105 106 108 110 111 113
210 103 105 107 109 110 112 114 116 117
220 106 109 111 112 114 116 118 120 121

Contd.
 DESIGN OF BEAMS 357

Table 6.3 (ii) Contd.

fy →
200 230 240 250 260 280 300 320 340
fcb ↓
230 110 112 114 116 118 120 122 124 126
240 113 115 117 119 121 124 126 128 130
250 115 118 120 122 124 127 130 132 134
260 110 121 123 126 128 131 133 136 138
270 121 124 126 129 131 134 137 139 142
280 123 126 129 132 134 137 140 143 145
290 126 129 132 135 137 141 144 147 149
300 128 131 135 137 140 144 147 150 153
310 130 134 137 140 143 147 150 153 156
320 133 136 140 143 146 150 153 157 160
330 135 138 142 145 148 152 156 160 163
340 137 141 144 148 151 155 159 163 166
350 139 143 147 150 153 158 162 166 169
360 141 145 149 152 156 161 166 169 172
370 143 147 151 155 158 163 168 172 175
380 144 149 153 157 160 166 170 174 178
390 146 151 155 159 163 168 173 177 181
400 148 152 157 161 165 170 175 180 184
420 151 156 160 165 169 175 180 185 189
440 154 159 164 169 173 179 185 190 195
460 157 162 167 172 177 183 189 194 200
480 159 165 170 175 180 187 193 199 204
500 162 168 173 178 183 190 197 203 209
520 164 170 176 181 187 194 201 207 213
540 166 172 178 184 189 197 204 211 217
560 168 175 181 187 192 200 208 215 221
580 170 177 183 189 195 203 211 218 225
600 172 179 185 192 198 206 214 222 229
620 174 181 187 194 200 209 217 225 232
640 175 183 189 196 202 211 220 228 235
660 177 184 191 198 204 214 222 231 238
680 178 186 193 200 207 216 225 234 242
700 180 187 195 202 209 218 228 236 244
720 181 189 196 204 210 220 230 239 247
740 182 190 198 205 212 222 232 241 250
760 184 192 199 207 214 224 234 244 253

Contd.
358 DESIGN OF STEEL STRUCTURES

Table 6.3 (ii) Contd.

fy →
200 230 240 250 260 280 300 320 340
fcb ↓
780 185 193 201 208 216 226 236 246 255
800 186 194 202 210 217 228 238 248 257
850 188 197 205 213 221 232 243 253 263
900 191 200 208 216 224 236 247 258 268
950 193 202 211 219 227 240 251 262 273
1000 195 204 213 222 230 243 255 266 277
1050 196 206 215 224 233 246 258 270 281
1100 198 207 217 226 235 248 261 273 285
1150 199 209 219 228 237 251 263 276 288
1200 200 210 220 230 239 253 266 279 291
1300 203 213 223 233 243 257 270 284 297
1400 205 215 225 236 246 260 274 288 302
1500 206 217 228 238 248 263 278 292 306
1600 208 219 229 240 250 266 281 295 309
1700 209 220 231 242 252 268 283 298 313
1800 210 221 232 243 254 270 285 301 316
1900 211 222 234 245 256 272 287 303 318
2000 212 226 235 246 257 273 289 305 321
2200 213 225 237 248 259 276 292 309 325
2400 215 226 238 250 261 278 295 312 328
2600 216 227 239 251 263 280 297 314 331
2800 216 228 240 252 264 282 299 316 333
3000 217 229 241 253 265 283 300 318 335
3500 218 231 243 255 267 286 303 321 339
4000 219 232 244 257 269 287 306 324 342
4500 220 233 245 258 270 289 307 326 344
5000 221 233 246 259 271 290 309 327 246
5500 221 234 247 259 272 291 310 328 347
6000 222 234 247 260 273 291 310 329 348

6.7.3 Elastic Critical Stress

The elastic critical stress in bending may be calculated by an elastic flexural -
torsional buckling analysis. In case, this analysis is not carried out, the elastic
critical stress in bending, fcb for the beams and plate girders with the moment of
inertia about axis of minimum strength (viz. yy-axis), Iyy smaller than that about
axis of maximum strength (viz. xx-axis), Ixx may be determined from the
expression derived as follows:
A paper titled as the basis for design of beams and plate girders in the revised
 DESIGN OF BEAMS 359

BS: 153 (British Standards) structural paper 48, proceedings of the Institution
of Civil Engineers, London, August 1956 was published by Kerensky, D.A., Flint,
A.R., and Brown, W.C. Professor Kerensky gave the expression for critical
bending moment for beams and plate girders as under:
1 1
⎡⎛
1 π2 EIy ⋅ h ⎞ ⎤2
2
π h ⎛ E ly ⎞ 2
π
Mcr = ( EI y .GK ) ⎜
⎢2 1 + ⎟ ⎥ + (2 λ − 1 )
l ⎢⎣ ⎝ l 2 4G K ⎠ ⎥ 2 .l ⎜⎝ G K ⎟⎠
⎦
where, λ is the ratio of moment of inertia of compression flange about yy-axis, If
to the moment of inertia of the whole section about yy-axis Iy.( 2λ − 1) may be
replaced by k 2. For the values of λ = 1 or 0, k 2 is equal to + 1.0 and – 1.0. Above
expression (i) may be written as below [as expression (ii) derived in Eq. 6.10].
1
2 ⎤2
π EI y ⋅ h ⎡⎢
2
1 ⎛ lt y ⎞ ⎥ ⎛ π2 EI y .h ⎞
Mcr = 1+ + k . 2 ⎟ ...(ii)
2 ⋅ l2 ⎢ 20 ⎜⎝ ry ⋅ h ⎟⎠ ⎥ 2⎜
⎝ 2 l ⎠
⎣ ⎦
1
2 ⎤2
π DI y′ ⋅ h ⎡
2
1 ⎛ ltf ⎞ ⎥ ⎛ π2 EI y ⋅ h ⎞
Mcr = ⎢1 + ⎜ ⎟ + k ⋅ ...(iii)
2 ⎜ ⎟
2 ⋅ l2 ⎢ 20 ⎝ ry ⋅ h ⎠ ⎥ ⎝ 2 l2 ⎠
⎣ ⎦
where Iy´ is the modified moment of inertia. It is equal to k 1· Iy´,(Iy is the moment
of inertia of the whole section about yy-axis at the section of maximum bending
moment).
Since, Mcr = fcb . Zxx, the critical bending stress is given by
1
2
π2 EI y′ ⋅ h ⎡ ⎛ l ⋅ tf ⎞ ⎤ 2 ⎛ π2 EI y′ h ⎞
fcb = ⎢1 + 1 ⎜ ⎟ ⎥ + k2⎜ ⎟ ...(iv)
2Z xx ⋅ l2 ⎢ 20 ⎝ ry ⋅ h ⎠ ⎥ ⎝ 2.Z xx ⋅ l2 ⎠
⎣ ⎦
This expression (iv) may be written as below : (as Eq. 6.10 has been derived).
1
2 ⎤2 ⎛ 26.5 × 105 × k1 ⎞
5 ⎡ ⎛ l ⋅ t ⎞
26.5 × 10 ⋅ k1 ⎢ 1 f ⎜ 2 ⎟
fcb = 2 ⎢
1+
20 ⎜ r ⋅ h ⎟ ⎥⎥ + k2 ⋅ ⎜ ⎛ l⎞ ⎟ ...(6.16)
⎛ l⎞ ⎝ y ⎠ ⎜ ⎜r ⎟ ⎟
⎣ ⎦ ⎝ ⎝ y⎠ ⎠
⎜r ⎟
⎝ y⎠
where the moment of inertia of the tension flange about yy-axis exceed that of
the compression flange, Eq. 6.16 may be written as below:
⎡ 1 ⎤
⎧ 2 ⎫2
⎢ 26.5 × 105 × k ⎪ 1 ⎛ l ⋅ tf ⎞ ⎪ 26.5 × 105 × k1 ⎥ . C2
1
fcb = ⎢ ⎨1 + ⎜ ⎟ ⎬ + k2 ⎥ N/mm2
⎢ ⎛ l⎞
2
⎪⎩ 20 ⎝ ry ⋅ D ⎠ ⎪⎭ ⎛ l⎞
2
⎥ C1
⎢ ⎜r ⎟ ⎜r ⎟ ⎥
⎣⎢ ⎝ y⎠ ⎝ y⎠ ⎦⎥
...(6.17)
360 DESIGN OF STEEL STRUCTURES

C2
fcb = k 1 . [X + k 2 . Y] . ...(6.18)
C1
IS: 800–1984 recommends Eq. 6.18 for calculating elastic critical stress in
bending
1
⎡ 1 ⎛ l ⋅ T ⎞ ⎤2
where X = Y ⎢1 + ⎜ ⎟ ⎥ N/mm2 (MPa) ...(6.19)
⎢⎣ 20 ⎝ ry ⋅ D ⎠ ⎥⎦

⎛ 26.5 × 105 ⎞
Y =⎜ 2
2
⎟ N/mm (MPa) ...(6.20)
⎜ ⎛ l ⎞ ⎟
⎜ ⎜r ⎟ ⎟
⎝ ⎝ y⎠ ⎠

k 1, l, ry, D and T symbols as used in Tables 6.1 (a) to (f) have been defined
above.
k1 = a coefficient. It allows for the inequality of the flanges. It depends
on ω . ω is the ratio of moment of inertia of the compression
flange alone to that of the sum of the moments of inertia of the
flanges, each calculated about its own axis parallel to the y-y of
the girder, at the point of maximum bending moment.
The values of co-efficient, k2 for different values of ω are given
in Table 6.4.

Table 6.4 Values of k2 for beams with unequal flanges (As per IS: 800–1984)

w 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
k2 0.5 0.4 0.3 0.2 0.1 00 –0.2 –0.4 –0.6 –0.8 –0.10
c1 = lesser distance to the exterme fibre of the section
c2 = maximum (greater) distance to the extreme fibre of the section
Iy = moment of inertia of the whole section about the axis lying in
the plane of bending (yy-axis); and
Ix = moment of inertia of the whole section about the axis normal to
the plane of bending (xx-axis).
(a) For flanges of equal moment of inertia
ω = 0.5, k2 = 0
(b) For tees and angles bending in the plane of web with the xx-axis (neu-
tral axis) nearer the extreme fibre of compression zone.
ω = 1.0, k2 = 0.5
(c) The bending stress in the leg when loaded with the flange or table in
compression shall not exceed 0.66 fy. When loaded with the leg in compression,
 DESIGN OF BEAMS 361

the permissible bending stress shall be calculated from Eq. 6.17 and Eq. 6.18
with k 2 = –1.0 and T as the thickness of leg.

⎛ D⎞
The values of X and Y are given in Table 6.5 for appropriate values of ⎜ ⎟
⎝T ⎠

⎛ l⎞
and ⎜ ⎟ .
⎝ ry ⎠
The values of elastic critical stress in bending, fcb may be increased by 20 per
⎛T ⎞ ⎛d ⎞
cent when ⎜ ⎟ ratio is not greater than 2.0 and ⎜ 1 ⎟ is not greater than
⎝ tw ⎠ ⎝ tw ⎠
1344
where d1 is the depth of web and t w is the thickness of web.
( fy )1 / 2
When the bending of beams occur about axis of minimum strength (yy-axis),
the maximum permissible bending stress in tension σbc or in compression σbc in
beams should not exceed 0.66 fy, where fy is the yield of structural steel used.
The principal distinction between the design of laterally supported beams
and laterally unsupported beams is the method of dealing with the bending in
the laterally unsupported beams. The allowable bending stress in compression
is reduced where necessary, depending upon the beams susceptibility of buck-
ling. The reduction in allowable bending stress in compression provide adequate
safety against the lateral torisonal buckling. The reduction in allowable stress
increases with the unsupported length of compression flange to the width of
flange ratio and depth of beam to the thickness of flange ratio.
Table 6.5 (i) Values of x and y for calculating fcb in N/mm2 (MPa)

D/T→ X Y

l / ry
8 10 12 14 16 18 20 25 30
↓
40 2484 2222 2066 1965 1897 1849 1814 1759 1728 1656
45 2103 1856 1708 1612 1546 1499 1465 1411 1380 1309
50 1822 1590 1449 1357 1293 1248 1214 1161 1131 1060
55 1607 1389 1254 1166 1105 1061 1028 976 947 876
60 1437 1232 1104 1020 961 918 886 835 806 736
65 1301 1107 985 904 847 806 775 726 697 627
70 1188 1005 889 811 757 717 687 638 610 541
75 1094 920 810 735 682 644 615 567 540 471
80 1014 849 743 672 621 584 556 509 482 414
Contd.
362 DESIGN OF STEEL STRUCTURES

Table 6.5 (i) Contd.

D/T→ X Y

l / ry
8 10 12 14 16 18 20 25 30
↓
85 945 788 687 618 570 533 506 461 434 367
90 886 735 639 573 526 491 464 420 394 327
95 833 689 597 534 488 454 428 385 360 294
100 787 649 560 499 455 423 398 356 331 265
110 708 582 449 443 402 371 347 307 283 219
120 644 527 451 398 359 330 308 270 247 184
130 591 482 411 361 325 298 277 240 218 157
140 546 444 378 331 297 271 251 217 195 135
150 508 412 350 306 274 249 230 197 177 118
160 474 385 326 284 254 230 212 181 161 104
170 445 360 305 265 236 214 197 167 148 92
180 420 339 286 249 221 200 184 155 137 82
190 397 320 270 235 208 188 172 145 137 73
200 376 304 256 222 197 177 162 136 119 66
210 358 288 243 210 186 168 153 128 112 60
220 341 275 231 200 177 159 145 121 105 55
230 326 262 220 191 169 152 138 115 99 50
240 312 251 211 182 161 145 132 109 94 46
250 299 241 202 175 154 138 126 104 90 42
260 288 231 194 167 148 133 121 99 85 39
270 277 222 186 161 142 127 116 95 82 36
280 267 214 180 155 137 122 111 91 78 34
290 257 207 173 149 132 118 107 88 75 32
300 249 200 167 144 127 114 103 84 72 32

Table 6.5 (ii) Values of x and y for calculating fcb in N/mm2 (MPa)

D/T→ X Y

l / ry
20 25 30 35 40 50 60 80 100
↓
40 1814 1759 1728 1709 1697 1683 1675 1667 1662 1656
45 1465 1411 1380 1362 1349 1335 1327 1319 1315 1309
50 1214 1161 1131 1113 1101 1086 1078 1070 1067 1060
Contd.
 DESIGN OF BEAMS 363

Table 6.5 (ii) Contd.

D/T→ X Y

l / ry
20 25 30 35 40 50 60 80 100
↓
55 1028 976 947 929 917 902 894 886 883 876
60 886 835 806 788 776 762 754 746 743 736
65 775 726 697 679 667 653 645 637 634 627
70 687 638 610 592 581 567 559 551 547 541
75 615 567 540 522 511 497 489 481 478 471
80 556 509 482 465 454 440 432 424 421 414
85 506 461 434 417 406 392 385 377 373 367
90 464 420 394 377 366 353 345 337 327 327
95 428 385 360 343 332 319 311 304 300 294
100 398 356 331 314 304 290 283 275 227 265
110 347 307 283 268 257 244 237 229 226 219
120 308 270 247 232 222 209 202 194 191 184
130 277 240 218 204 194 181 174 167 163 157
140 251 217 195 181 172 160 153 145 142 135
150 230 197 177 163 154 142 135 145 124 118
160 212 181 161 148 139 127 121 113 110 104
170 197 167 148 135 126 115 109 102 98 92
180 184 155 137 125 116 105 98 92 88 82
190 172 145 127 115 107 96 90 83 80 73
200 162 136 119 107 99 89 83 76 73 66
210 153 128 112 101 93 82 76 70 66 60
220 145 121 105 94 87 77 71 64 61 55
230 138 115 99 89 82 72 66 60 56 50
240 132 109 94 84 77 67 62 55 52 46
250 126 104 90 80 73 64 58 52 49 42
260 121 99 85 76 69 60 55 48 45 39
270 116 95 82 72 66 57 52 46 42 36
280 111 91 78 69 63 54 49 43 40 34
290 107 88 75 66 60 52 46 41 38 33
300 103 84 72 64 57 49 44 38 35 32

6.8 EFFECTIVE LENGTH OF COMPRESSION FLANGE

In the laterally unsupported of beams, allowable bending compressive stress σec
is reduced in proportion to critical bending stress fbc. The critical bending stress
can be computed from the effective length of common pression flange ‘l’ and
other properties of the section.
364 DESIGN OF STEEL STRUCTURES

In the lateral buckling of compression flange, lateral bending of the flange

occurs with twisting. The lateral buckling of the compression flange depends on
the effective length of the compression flange, which depends on the effective
lateral supports. The lateral supports should provide torsional restraints. The
torsional restraints can be provided in the following ways:
1. By providing web or flange cleats in the end connection.
2. By providing bearing stiffeners acting in conjuction with bearing of the
beam.
3. By providing lateral and frames or any other external support to the
end of the compression flange.
4. By fixing the ends of the beam into the walls.
ISI recommends effective length of compression flange in ISI: 800–1984 as
follows:
1. For simply supported beams and girders where each end of the
beam is restraint against torsion.
(i) With ends of compression flange unrestrained (unsupported) against
lateral bending
l = span
(that is, free to rotate in plane at the bearings)
(ii) With ends of compression flange partially restrained against lateral
bending
l = 0.85 × span
(that is, not free to rotate in plane at the bearings)
(iii) With ends of compression flange fully restrained against lateral bending
l = 0.7 × span
(that is, free to rotate in plane at the bearings)
Where the ends of the beam are not restrained against torsion, or where the
load is applied to the compression flange and both the load and the flange are
free to move laterally, the above value of effective length should be incresed by
20 per cent.
Note. It is to note that the end restraint element shall be capable of safely resisting,
in addition to wind and other applied external forces, a horizontal forces acting at
the bearing in a direction normal to compression flange of the beam at the level of the
centriod of the flange and having a value equal to, not less than 2.5 per cent of the
maximum force occurring in the flange.
2. For beams which are provided with members giving effective
lateral restraint to the compression flange at interval along the span
in addition to the end torsional restraint.
The effective length ‘l’ shall be taken as the maximum distance centre to
centre of the restraints members.
3. For cantilever beams builts-in at the support with projecting
length ‘L’.
(i) Built-in at the support, free at the end
l = 0.85 L
 DESIGN OF BEAMS 365

(ii) Built-in at the support, restrained against torsion at the end by

continuous construction
l = 0.75 L
(iii) Built-in at the support, restrained against lateral bending and torsion
at the free end
l = 0.5 L
4. For cantilever beams continuous at the support with project-
ing length ‘L’.
(i) Unrestained against torsion at the support and free at the end
l = 3L
(ii) Partially restrained against torsion at the support and free at the end
l = 2L
(iii) Restained against torsion at the support and free at the end
l = L
where, L = length of cantilever
If there is a degree of fixidity at the free end, the effective length shall be
⎛ 0.5 ⎞ ⎛ 0.75 ⎞
multiplied by ⎜ in 3 (ii) and 3 (iii) above, and by ⎜ in 4 (i), 4 (ii) and 4
⎝ 0.85 ⎟⎠ ⎝ 0.85 ⎟⎠
(iii) above.
Where the beams supported slab construction, the beam shall be deemed to
be effective restrained laterally if the frictional or positive connection of the
slab to the beam is capable of resisting a lateral force of 2.5 per cent of the
maximum force in the compression flange of the beam, considered as distributed
uniformly along the flange. Furthermore, the slab construction shall be capable
of resisting this lateral force in lateral flexure and shear.
For the beams which are provided with members giving effective lateral
restraint of the compression flange at intervals along the span, the effective
lateral restraint shall be capable of resisting a force of 2.5 per cent of the
maximum force in the compression flange taken as divided equally between the
number of points at which the restraint members occur.
In a series of such beams, with soild webs, which are connected together by
the same system of restraint members, the sum of the restraining force shall be
taken as 2.5 per cent of the maximum flange force in one beam only.
In case of a series of latticed beams, which are connected together by the
same system or restraint members, the sum of the restraining forces required
shall be taken as 2.5 per cent of the maximum force in the compresssion flange
plus 1.25 per cent of the forces of 7.5 per cent.
The maximum slenderness ratio of the compression flange that is, the ratio
of the effective length to the appropriate radius should not exceed 300 as in case
of column members.
Example 6.5 MB 550, @ 1.037 kN/m has been used as simply supported beam
over a span of 4 metres. The ends of beam are restrained against torsion but not
366 DESIGN OF STEEL STRUCTURES

against lateral bending. Determine the safe uniformly distributed load per cent
metre length which the beam carry.
Solution
Step 1 : Properties of given section
From ISI Handbook No. 1, for MB 550, @ 1.037 kN/m
Overall depth D = h = 550 mm
Means thickness of flange T = t f = 19.3 mm
Radius of gyration, ry = 37.3 mm
Section modulus, Zxx = 2359.8 × 103 mm3
Thickness of web, tw = 11.2 mm
h2 = 41.25 mm
Step 2 : Elastic critical stress in bending , fcb
The effective length of the compression flange is 4 m. Slenderness ratio
⎛ l⎞ ⎛ 4 × 1000 ⎞
⎜ r ⎟ = ⎜⎝ = 107.238
⎝ y⎠ 37 ⋅ 3 ⎟⎠

⎛ D⎞ ⎛ Overall depth ⎞
Ratio ⎜⎝ T ⎟⎠ = ⎜⎝
Means thickness of flange ⎟⎠

⎛ D⎞ ⎛ 550 ⎞
⎜⎝ T ⎟⎠ = ⎜⎝ 19.3 ⎟⎠ = 28.497
From IS: 800–1984, Table 6.5
X = 303.555, and Y = 231.705
c2
From Eq. 6.18, fcb = k 1 . ( X + k2 ⋅ Y ) ⋅ ...(i)
c1
c1 and c2 of the beam and greater distance from the neutral axis to the extreme
fibres of the beam section which are equal
k 1 = 1.0 ψ = 1.0 and k 2 0.0 for w = 0.5
Substituting the respective values in the expression (i)
fcb = 1.0 (303.555 + 0.0 × 230.705) × 1.0
= 303.555 N/mm2 (MPa)
Let the value of yield stress for the structural steel be 260 N/mm2.
⎛T ⎞ ⎛ 19.3 ⎞
Ratio ⎜⎝ t ⎟⎠ = ⎜⎝ 11.2 ⎟⎠ = 1.723 >
|2
w

⎛ d1 ⎞ ⎛ 550 − 2 × 41 ⋅ 25 ⎞
Ratio ⎜⎝ t ⎟⎠ = ⎜⎝ ⎟⎠ = 41.74
w 11 ⋅ 2

⎛ 1344 ⎞ ⎛ 1344 ⎞
⎜ 1⎟ = ⎜ 1 ⎟ = 83.35
( )
⎜⎝ fy 2 ⎟⎠ ⎜
⎝ (260) 2 ⎟⎠
 DESIGN OF BEAMS 367

⎛ d1 ⎞ ⎛ 1344 ⎞
⎜⎝ t ⎟⎠ |
> ⎜ 1⎟
w
( )
⎜⎝ fy 2 ⎟⎠

As such, the value of fcb shall be increased by 20 per cent. Therefore,

fcb = (1 ⋅ 20 × 303 ⋅ 555)
= 364.266 N/mm2
Step 3 : Permissible compressive stress in bending
From Table 6.2, IS: 800–1984,
Maximum permissible bending compressive stress
σbc = 121.4 N/mm2 (MPa)
Step 4 : Moment of resistance of beam section
⎛ 121 ⋅ 4 × 2359 ⋅ 8 × 103 ⎞
Mr = ⎜ ⎟
⎝ 1000 × 1000 ⎠
= 286.48 m-kN
Step 5 : Uniformly distributed load
Let w be the uniformly distributed load in kN/m
The maximum bending moment, M occurs at the centre
⎛ w ⋅ l2 ⎞
M = ⎜ ⎟
⎝ 8 ⎠

⎛ w × 4 × 4 × 1000 ⎞
⎜⎝ 8 ⎟⎠ = (286 ⋅ 48 × 10 )
3

w = 143.24 kN/m
The self-weight of beam is 1.037 kN/m
The safe uniformly distributed load which can be placed over the beam
= ( 143.24–1.037)
= 142.203 kN
Example 6.6 Design a rolled steel I-section for a simply supported beam with
a clear span of 6 m. It carries a uniformly distributed load of 50 kN per metre
exclusive of self-weight of the girder. The beam is laterally unsupported.
Solution
Design :
Step 1 : Total load to be supported
The clear span of the simply supported beam is 6 m.
Assuming that 300 mm wide end bed blocks are used
Effective span of the beam l = 6.30 m
Uniformly distributed load carried by the beam exclusive of self-weight
= 50 kN/m
Assuming self-weight of the beam = 1 kN/m
Total uniformly distributed load inclusive of self-weight
= 51 kN/m
368 DESIGN OF STEEL STRUCTURES

Step 2 : Bending moment and shear force

The maximum bending moment occurs at the centre

⎛ w ⋅ l2 ⎞ ⎛ 51 ⋅ 00 × 6 ⋅ 30 × 6 ⋅ 30 × 1000 ⎞
M =⎜ ⎟ = ⎜⎝ ⎟⎠
⎝ 8 ⎠ 8 × 1000
= 253 kN-m
The maximum shear force occurs at the ends

⎛ wl ⎞ ⎛ 51 ⋅ 00 × 6 ⋅ 30 ⎞
F = ⎜ ⎟ = ⎜⎝ ⎟⎠ = 160.65 kN
⎝ 2⎠ 2
Step 3 : Permissible stress
Let the value of yield stress fy for the structural steel to be used be 250 N/
mm2 (MPa). For the maximum value of fcb, the maximum permissible bending
compressive stress should not exceed 0.66 fy. Therefore, assume
σbc = (0 ⋅ 66 × 250) = 165 N/mm2 (MPa)
Step 4 : Section modulus required

⎛ 253 × 1000 × 1000 ⎞

Z = ⎜ ⎟⎠
⎝ 165
= 1533.33 × 103 mm3
However, the actual value of permissible bending compressive stress,σbc shall
be less than 0.66 fy. The modulus of section Z needed for the beam section shall
be more. The trial section modulus may be adopted 25 percent to 50 percent in
excess of above calculated.
Step 5 : Properties of trial section
Trial section modulus = 1.50 × 1533.33 × 103 = 2299.995 × 103 mm3
From ISI Handbook No. 1, try WB 550, @ 1.125 kN/m
Overall depth D = h = 550 mm
Mean thickness of flange
T = t f = 17.6 mm
Radius of gyration ry = 51.1 mm
Section modulus
Zxx = 2723.9 × 103 mm3
Thickness of web,
t w = 10.5 mm, h2 = 38.30 mm,
Ixx = 74906.1 ×104 mm4
Step 6 : Check for section modulus
The effective length of compression flange,
l = 6.30 m
Slenderness ratio
⎛ 6 ⋅ 30 × 1000 ⎞
⎜⎝ 51 ⋅ 1 ⎟⎠ = 123.287
 DESIGN OF BEAMS 369

⎛ D⎞ 550
Ratio ⎜⎝ T ⎟⎠ = = 31.25
17.6
From IS: 800–1984, Table 6.5
X = 233.799 and Y = 175.125
From Eq. 6.18,
c2
fcb = k 1 . (X + k 2.Y).
c1
c1 and c2, the smaller and larger distance from the neutral axis to the extreme
fibres are equal. Therefore,
k1 = 1.0 for ψ = 1.0 and
k2 = 0.0 for ω = 0.5
Substituting the respective values in the expression (i)
fcb = 1.0(233.799 + 0.0 × 175.125 ×) ×1.0
fcb = 233.799 N/mm2
⎛T ⎞ ⎛ 17.6 ⎞
Ratio ⎜⎝ t ⎟⎠ = ⎜⎝ 10.5 ⎟⎠ = 1.676 < 2.0
w

⎛ d1 ⎞ ⎛ 550 − 2 × 38.50 ⎞
⎜⎝ t ⎟⎠ = ⎜⎝ 10.5 ⎟⎠ = 45.08
w

⎛ 1344 ⎞ ⎛ 1344 ⎞
⎜ 1⎟ = ⎜ 1 ⎟ = 85
( )
⎜⎝ fy 2 ⎟⎠ ⎜⎝ (250) 2 ⎟⎠

The value of fcb shall be increased by 20 per cent. Therefore,

fcb = (1.20 × 233.799)
= 280.559 N/mm2
From IS: 800–1984, Table 6.2
σbc = 106.112 N/mm2 (MPa)
Section modulus required
⎛ 253 × 1000 × 1000 ⎞
⎟⎠ = 2384.27 × 10 mm
3 2
⎜⎝ 106 ⋅ 112
2723.9 × 103 mm2 provided.
Hence, safe.
Step 7: Check for shear stress
Average shear stress
⎛ 160 ⋅ 65 × 103 ⎞
τva.cal = ⎜ ⎟
⎝ 550 × 11 ⋅ 2 ⎠
= 26.079 N/mm2
< (0.4 × 250 = 100 N/mm2). Hence, safe.
Step 8: Check for deflection

5 ⎛ w ⋅ l4 ⎞
ymax = ⋅⎜ ⎟
384 ⎝ EI ⎠
370 DESIGN OF STEEL STRUCTURES

5 ⎛ 51 ⋅ 00 × 6 ⋅ 3 × 6 ⋅ 303 × 10003 × 1000 ⎞

ymax = × ⎜ ⎟ mm
384 ⎝ 2.047 × 105 × 74906.1 × 104 ⎠
= 6.82 mm
Allowable deflection
⎛ 1 ⎞
⎜⎝ 325 × 6300⎟⎠ = 19.4 mm > y max. Hence, safe.

Provide WB 550 @ 1.125 kN/m

6.9 WEB CRIPPLING

The web of rolled steel beam is usually thin. The web is relatively weak when
placed under direct compression.
In rolled steel beams, the failure of the web may occur at concentrated loads,
and at supports, where it has reaction. The failure of web in direct crushing
under concentrated load is known of web crippling or web crimpling. The web
crippling may occur if the in-plane compressive stress in the web is too large. It
may also occur if (b + 2k) distance (Fig. 6.9) used to deliver load from the beam
flange to the web is too narrow or (b + k) distance used to deliver load from the
web to the flanges is too small.
The web crippling may also occur if the uniformly distributed load on the
flange is too large for web thickness.
The web crippling means the stress concentration due to bottleneck condi-
tion at the web toe of the fillet under or over heavy load concentrations. The
web crippling is a localized bearing stress caused by the transmission of com-
pression from the comparatively wide flange to the narrow web. The crippling
may be brought about by a failure in bearing of the metal at the web toe of the
fillet and the resulting tendency of the flange and web to fold over on each other
at the plane.
In web crippling, local buckling of the web occurs immediately adjacent to a
concentration of stress as shown in Fig. 6.8.

h2
1
2
1 –1 C on cen tra tio n
o f stre ss
2 –2 Lo cal bu cklin g
of web
2
1
h2

(a ) At po in t o f su pp ort (b ) At po in t loa d

Fig. 6.8 Web crippling

 DESIGN OF BEAMS 371

For safety against web crippling adequate bearing length along the beam
may be provided so that the bearing stress ‘σp’ for beams without stiffeners may
not exceed the allowable bearing stress σp = 0.75 fy (0.75 times yield stress of IS:
800–1984). As per IS: 800–1962 dispersion of load or reaction through the flange
to the web occurs uniformly at an angle of 30 degrees to the horizontal as shown
in Fig. 6.9.
k k
h2
3 0° 30°
b + 2k tw
h

b +k 3 0°
h2

b k
Fig. 6.9 Bearing length of web

AISC specification recommend an angle of dispersion of 45° instead of 30 ° as

recommended in IS: 800–1984.
The bearing length of the web under concentrated load is given by

(
B = (b1 + 2k) = b1 + 2h2 3 ) ...(6.17)
where b1 = length of bearing plate
h2 = depth of root of fillet from the outer surface of flange

⎛ ⎞
Bearing stress σp = ⎜ Concentrated load ⎟ N/mm2
(
⎜⎝ b1 + 2h2 3 ⋅ tw ⎟⎠ )
This should be less than the allowable bearing stress 0.75 N/mm2.
The bearing length of web at the support is given by

(
B = (b + k) = b + h2 3 ) ... (6.18)

⎛ Reaction ⎞
Bearing stress, σp = ⎜ ⎟
⎝ b + h2 3tw ⎠
This should be less than the allowable bearing stress σp = 0.75 fy.
This depth of root of fillet from the outer surface of flange ‘h2’ for the beam is
given in steel section tables under connection details.

6.10 WEB BUCKLING

The web buckling means that type of failure of web in which the web, vertically
above the bearing plate and the lower flange at the reaction or below a
concentrated load (bearing plate and the upper flange) is subjected to the column
372 DESIGN OF STEEL STRUCTURES

action and tends to buckle under it. It is out-of plane distortion of web of a
beam. It is also called as vertical web buckling. It reults due to combination
⎛h ⎞
of large value of ⎜ 1 ⎟ ratio and the bending stress. The unbraced length of
⎝ tw ⎠
compression flange may also contribute to the web buckling.
In adding to local crushing or crippling of the web, buckling may also occur
above a support or below a concentrated load. The web is considered to act as a
column. The web may buckle as shown in Fig. 6.10 depending upon restraint of
flanges against lateral deflection and rotation. The web buckling will occur
depending on the type of support, the top flange receives from the surrounding
construction.

(a ) (b ) (c) (d )
Fig. 6.10 Web buckling

The bottom flange is restrained against lateral deflection and rotation in all
cases. The restraint conditions for top flange are as below:
1. Restrained agaist lateral deflection and rotation as shown in Fig. 6.10
(a);
2. Restrained agaist lateral deflection and rotation as shown in Fig. 6.10 (b);
3. Restrained agaist lateral deflection and rotation as shown in Fig. 6.10 (c);
and
4. Restrained agaist lateral deflection and rotation as shown in Fig. 6.10 (d).
IS: 800–1984 recommends that bearing stiffners should be provided at points
of concentrated load and point of support where the concentrated load reaction
exceeds the value of ( σac . tw . B)
⎛ d1 3 ⎞
where, σac = Axial stress for column for a slenderness ratio ⎜ ⎟
⎝ tw ⎠
d1 = (h – 2 · h2)
tw = Thickness of web
B = The length of the stiff portion of the bearing plus the additional
length by dispersion at 45° to the level of the neutral axis, plus
the thickness of the scating angle, if any.
(The stiff portion of bearing is that length which cannot deform
appreciably in bending and shall not be taken as greater than
half the depth of beam for simply supported beams and the full
depth of beams continuous over a bearing.)
 DESIGN OF BEAMS 373

For concentrated load at any intermediate position on the beam, B is the

length of bearing plate under the load plus the depth of beam plus thickness of
flange plates if any, e.g., in plated beams (i.e., B = b + h) as shown in Fig. 6.11,
d1 = Clear depth of web between root fillets
where, d1 = (h – 2h2)
h2 = Depth of root of fillet from the outer surface of flange.
The selenderness ratio for the portion of web acting as a column, when the
two flanges are restrained against lateral displacement and rotation may be
found as follows :

⎛ Effective length ⎞ ⎛ 0 ⋅ 5d1 ⎞

Slenderness ratio = ⎜ ⎟ = ⎜ ⎟
⎝ Minimum radius of gyration ⎠ ⎝ ry ⎠

⎛ d1 ⎞ ⎛ d1 ⎞
⎜ 2⎟ ⎜ 2 ⎟ d1 / 2 ⎛d ⎞
⎜r ⎟ = ⎜ 1/ 2 ⎟ =
= ⎜ 1 ⋅ 3⎟
⎝ y⎠ ( )
⎜ I yy 3
⎟ ⎡ 1 B ⋅ tw ⎤
1/2
⎝ tw ⎠
⎜⎝ ⎟⎠ ⎢ ⎥
A ⎣12 B ⋅ tw ⎦
The selenderness ratio for the portion of web acting as a column, when the
top flange is held in position only while the bottom flange is restrained against
lateral displacement and rotation.
d1
2 ⎛d ⎞
Slenderness ratio = 1/2
= ⎜ 1 ⋅ 6⎟
⎡⎛ 1 ⎞ B ⋅ tw3 ⎤ ⎝ tw ⎠
⎢⎜ ⎟ ⎥
⎣⎝ 12 ⎠ B ⋅ tw ⎦
Note. A beam that is safe in web crippling will usually be safe in web buckling, as
well.
The web crippling and web buckling may be prevented by adopting increased
thickness of the web, or by adopting the increased length of bearing plates. The web
⎛d ⎞
buckling is controlled by limiting either ⎜ 1 ⎟ ratio. The web buckling may be avoided
⎝ tw ⎠
by stiffening the web. The bearing stiffners may be provided under the heavy loads.
The diagonal buckling of the web has been discussed in Chapter 8.
Example 6.7 Check the beam section WB 500, @ 1.451 kN/m against web
crippling and web buckling if reaction at the end of beam is 179.6 kN. The length
of bearing plate at the support is 120 mm. Design bearing plate. The bearing
plate is set in masonry wall.
Solution
Step 1 : Properties of given section
Assume allowable bearing stress in masonry as 5.5 N/mm2
End reaction R = 179.6 kN
The length of bearing plates at the support is 120 mm
374 DESIGN OF STEEL STRUCTURES

From the steel section tables for WB 600, @ 1.451 kN/m

Thickness of web t w = 11.8 mm
Distance between root of fillet and flange of beam
h2 = 46.05 mm
Depth of beam between root of fillets
h1 = 507.9 mm
Step 2 : Check for web crippling
Bearing stress at the support, σp.cal

⎛ ⎞ ⎛ ⎞
End reaction 179 ⋅ 6 × 1000
⎜ ⎟ = ⎜ ⎟
( )
⎜⎝ b + h2 3 tw ⎟⎠ ( )
⎜⎝ 120 + 46 ⋅ 05 3 × 11.8 ⎟⎠

= 76.3 N/mm2
< ( 0.75 × 260 = 195 N/mm2). Hence, safe.
Step 3 : Check for web buckling
Reaction should not exceed the value (σac× tw × B)
⎛ h1 ⎞ ⎛ 507.9 3 ⎞
Slenderness ratio ⎜⎝ t 3 ⎟⎠ = ⎜ 11 ⋅ 8 ⎟ = 74.55
w ⎝ ⎠
Let the value of yield stress, fy for the steel be 260 N/mm2 (MPa)
Axial stress, σbc = 109.54 N/mm2
The length of stiff portion of bearing
⎛ 600 ⎞
B = ⎜120 + = 420 mm
⎝ 2 ⎟⎠

⎛ (109 ⋅ 54 × 11 ⋅ 8 × 420) ⎞
Allowable reaction = ⎜ ⎟⎠ = 542.88 kN.
⎝ 1000
> 179.6 kN and reaction. Hence, safe.
Step 4 : Design of bearing plate
End reaction = 179.6 kN
Allowable bearing stress in masonry
= 5.5 N/mm2
Bearing area required IS W B 6 00

⎛ 179 ⋅ 6 × 1000 ⎞
⎜⎝ 5⋅5 ⎟⎠ = 32654.54 mm2 b = 17 1 .5 m m

Length of bearing plate provided

= 120 mm
Width of bearing plate, 3 0°

⎛ 32654 ⋅ 54 ⎞
B = ⎜ ⎟⎠ = 272.12 mm ( say 280 mm)
⎝ 120 Fig. 6.12
 DESIGN OF BEAMS 375

Step 5 : Thickness of bearing plate

The dispersion of reaction occurs uniformly at 30º with the horizontal
b = (11.8 + 2 × 46.05 × cot 30º) mm
= (11.8 + 2 × 46.05 × 3 ) = 171.5 mm
The maximum bending moment, M occurs at the centres
R ⎛ 179 ⋅ 6 ⎞
M = ( B – b) = ⎜⎝ × (280 – 171 ⋅ 5)⎟
8 8 ⎠

= 2435.825 mm-kN
The moment resistance of plate section
⎛ 0 ⋅ 75 × 260 × 280 × t 2 ⎞
(σbc· Z) = ⎜ 1000 × 6 ⎟ kN-mm
⎝ ⎠

⎛ 0 ⋅ 75 × 260 × 210 × t 2 ⎞
⎜ 1000 × 6 ⎟ = 2435.825
⎝ ⎠

⎛ 2435 ⋅ 825 × 6 × 1000 ⎞

t2 = ⎜ (3 t = 16.36 mm)
⎝ 0 ⋅ 75 × 260 × 280 ⎟⎠
Provide base plate 120 mm × 280 mm ×18 mm.

6.11 BUILT-UP BEAM

The built-up beam are also termed as compound beams or compound
girders.The built-up beam are used when the span, load and corresponding
bending moment are of such magnitudes that rolled steel beam section become
inadequate to provide required section modulus. The uilt-up beams are also
used when rolled steel beams are inadequate for limited depth. In building
construction, the depth of beam is limited by the space provided by the architect´s
drawings. The beam of small depth do not provide required section modulus.
Therefore plates are attached to the beams. The strength of rolled steel beam is
increased by adding plates to its flanges. Which is therefore one method of
forming built-up section. The other method is to compound a number of rolled
steel sections,themselves. The built-up beam section commonly used are shown
in Fig. 6.13. The built-up sections shown in Fig. 6.13 (a) and (b) are used for
heavy loads and small spans. The built-up section shown in Fig. 6.13 (c) is also
used for heavy loads and large spans. In this built-up section, the neutral axis
passes in between the two flanges in contact. The bending stresses in these two
flange are very small. These two flanges in contact practically do not provide
any moment of resistance. Hence, such section is provided where deflection is a
major consideration. The built-up section shown in Fig. 6.13 (e) and (f) are
provided for the gantery girders. The top flange in these sections have been
strengthend. This increases the resistance of the section for the lateral loads.
376 DESIGN OF STEEL STRUCTURES

(a ) (b ) (c)

(d ) (e ) (f)

Fig. 6.13 Built-up beam sections

In the riveted built-up beams, the area of tension flange is reduced by rivet
holes. The deduction is made for rivet holes in tension flange while determining
moment of inertia and section modulus of the built-up beam section. In actual

Fig. 6.14
practice, it is usual to make allowance for rivet holes in the both flanges. The
rivets connecting flanges plates are staggered so that each flange is reduced by
one rivet hole as shown in Fig. 6.14.
The curtailment of cover plates of built-up beams is similar to curtailment of
cover plates of plate girder. It is discussed in chapter 8, Design of Plate Girders.

6.12 DESIGN OF BUILT-UP BEAMS

The built-up beams are designed by trial and error method. The built-up section
is selected and it is checked for the stresses. The built-up beams with soild webs
are proportioned in general on the basis of the moment of inertia of the gross
cross-section with neutral axis taken at the centroid of the section.
 DESIGN OF BEAMS 377

The following are the usual steps in design of built-up beams.

Step 1. The effective span and load required to be carried by the built-up
beam are known. The bending moment M and maximum shear F in the built-up
beam are calculated.
Step 2. In the beginning, the value of yield stress fy, for structural steel to be
used is assumed. The permissible bending stress σbc is assumed as 0.66 fy. The
required section modulus, Z for the beam section is calculated

⎛ M⎞
Required section modulus, Z = ⎜
⎝ σbc ⎟⎠
where,σbc is allowable bending compressive stress.
For laterally supported beams, allowable bending compressive stress is equal
to the allowable bending tensile stress. For laterally unsupported beams,
allowlable bending compressive stress is determined by the condition of lateral
restraints and the section of built-up beams.
From ISI Handbook No. 1, a trial section for beam for beam is adopted having
the modulus of section, Z about 25 to 50 percent in excess of that required. The
geometrical properties of beam section, are noted. The permissible bending
compressive stress corresponding to the geometrical dimensions of bean section
and restraint conditions is found from IS: 800–1984. As a check, the section
modulus, Z needed corresponding to actual permissible stress is found and it is
shown that is less than the section modulus of the beam provided.
When the depth of beam is no limited, then the rolled steel beam is selected
for the maximum section modulus. The plates are provided for the difference of
section modulus i.e., between that required and that provided by the rolled
steel beam. When the depth of beam is limited, then, the usual practice is to
select from ISI Handbook No. 1, the strongest rolled steel beam that will allow
for necessary thickness of cover plates, at top and buttom. The araea of covers
plates required is then found by trial and error method. The moment of inertia
of trial plated beam section is found, and then checked. It involves a consider-
able amount of work. The required area of cover plates may be found from the
following guidance, which will result in satisfactory selection of the required
area of cover plates at the first trial.
Let the moment of inertia of built-up plated beam (symmetrical section) be
given by

⎡ 2
⎛ h⎞ ⎤
I
I = ⎢ beam + 2 ⋅ Ap⎜ ⎟ ⎥ ...(i)
⎢⎣ ⎝ 2⎠ ⎥
⎦
where, Ibeam is the moment of inertia of the rolled beam section, Ap is the area of
cover plates in one flage and h is the distance between the centroids of the top
and bottom flanges plates.
The distance between the centroids of the top and bottom flange plates, h
may be adopted equal to the depth of rolled steel beam itself, since the thickness
of flange plates is small compared with the depth of beam section.
378 DESIGN OF STEEL STRUCTURES

From the expression (i)

⎛ I ⎞ ⎛ I beam h⎞
+ 2 Ap .
⎜ ⎛ h⎞ ⎟ = ⎜ ⎛ h⎞ 2⎟
...(ii)
⎜⎜ ⎟⎟ ⎜ ⎜ ⎟ ⎟
⎝
⎝ 2 ⎠⎠ ⎝ 2⎝ ⎠ ⎠
or Zreqd = (Zbeam + Ap . h)
⎛ Z reqd – Z beam ⎞
or Ap = ⎜ ⎟⎠ ...(iii)
⎝ h

⎛ I ⎞ ⎛ I beam ⎞
where, Zreqd = ⎜ and Zbeam = ⎜ ⎛ h⎞ ⎟
⎛ h⎞ ⎟
⎜⎜ ⎟⎟ ⎜ ⎜ ⎟ ⎟
⎝ ⎝ 2⎠ ⎠ ⎝ ⎝ 2⎠ ⎠
The expression (iii) gives the area of cover plates in one flange, when there
are not rivet holes in the plates. In the riveted built-up beams, an allowance of
about 13 percent of the area of cover plates found is added. The required area of
cover plates in one flange is then provided by one or two plates and preferably
not more than three plates. The rolled steel beam and cover plates provide the
required section modulus. The width of cover plates and thickness are decided
with the restrictions of the outstand.
Step 3. From steel section tables, select a suitable section. The outstand of
flange plates that is their projection beyond the outer line of connection shall
not exceed the relevant values given in IS: 800–1984 in (a), (b) and (c) below.
The area of excess flange shall be neglected while calculating the effective
geometrical properties of the section.
These values have also been expressed in terms of yield stress of the struc-
tural steel as it has been done in recommending the various permissible stresses
and other.
256t
(a) Flanges and plates in compression with the unstiffened edges :
( fy )1 2
subject to maximum of 16 t.
(b) Flanges and plates in compression with the unstiffened edges : 20 t to
the innermost face of the stiffening.
(c) Flanges and plates in tension 20 t.
Step 4. The built-up section is checked for bending stresses in the extreme
fibres. The gross moment of inertia of the built-up section is found on the basis
of gross cross-section of the built-up section with neutral axis assumed at the
centriod of the section. The maximum tensile stress in bending is found from
the maximum compressive stress in bending by increasing the maximum com-
pressive stress in bending in the ratio of gross area to the net area of tension
flange. The effective sectional area for the compression flange in the gross area
with deductions for excessive widths of the plates as discussed for compression
 DESIGN OF BEAMS 379

members and for open holes occurring in a plane perpendicular to the direction
of stress at the section being considered. The effective sectional area for the
tension flanges is the gross area with deductions for the rivet holes.
Actual bending compressive stress

⎛ Max Bending Moment ⎞ ⎛ Distance of extreme fibre in ⎞

σbc.cal = ⎜ ×
⎝ Gross Moment of Inertia ⎟⎠ ⎜⎝ compression, from neutral axis⎟⎠

⎛M⎞
σbc.cal = ⎜ ⋅ y1
⎝ I xx ⎟⎠

The actual bending tensile stress σbt.cal is determined by increasing the actual
bending compressive stress in the ratio of gross area of the tension flange to the
net area of the tension flange.

⎛ Gross area of the tension flange ⎞

σbt.cal = σbc.cal × ⎜
⎝ Net area of the tension flange ⎟⎠
The actual bending stresses should not exceed allowable bending stresses.
Step 5. The built-up section is checked for shear force and deflection.
Example 6.8 Design a simply supported plated rolled steel beam section to
carry a uniformly distributed load 50 kN/m inclusive of self-weight of beam.
The effective span of beam is 10 metres. The depth of beam should not be more
than 500 mm. The compression flange of the beam is laterally supported by floor
construction.

1 50 m m 1 50 m m

P late
4 00 m m × 1 6 m m
2 33 m m

IS W B 4 5 0 @ 0 .7 94 kN /M
48 2 m m

Fig. 6.15

Solution
Design :
Step 1 : Effective span of beam
= 10 m
380 DESIGN OF STEEL STRUCTURES

Step 2: Maximum bending moment,

⎛ w ⋅ l2 ⎞
M = ⎜ ⎟
⎝ 8 ⎠

⎛ 50 ⋅ 00 × 10 × 10 ⎞
= ⎜ ⎟⎠ = 625 kN-m
⎝ 8
Step 3: Maximum shear force,
⎛ wl ⎞ ⎛ 50 ⋅ 00 × 10 ⎞
F = ⎜ ⎟ =⎜ ⎟⎠ = 250 kN
⎝ 2⎠ ⎝ 2
Step 4 : Permissible bending stress
Let the value of yield stress for the structural steel to be used be 250 N/mm2.
The permissible bending stress in compression may be assumed as
(0.66 ×250) = 165 N/mm2 (MPa)
Step 5 : Section modulus required
⎛ M⎞ ⎛ 625 × 1000 × 1000 ⎞
Z= ⎜ ⎟ = ⎜ ⎟⎠
σ
⎝ bc ⎠ ⎝ 165
= 3787.88 ×103 mm3
Step 6 : Trial section
From the steel section table, try WB 450, @ 0.794 kN/m with one cover plate
400 mm ×16 mm connected to each flange by 22 mm diameter rivets as shown
in Fig. 6.15.
For WB 450, @ 0.794 kN/m
Ixx = 35057.6 × 104 mm4, tw = 9.2 mm,bf = 200 mm
Zxx= 1558.1×103 mm3, h = 450 mm, t f = 15.4 mm
Projection of cover plate beyond the rivet line in compression flange
= 150 mm < 16 × 16 = 256 mm
(Maximum allowable outstand)
Step 7 : Check for fibre stress
Gross area of tension flange
= (200 × 15.4 × 400 × 16) = 9480 mm2
Net area of tension flange
= [9480 – 23.5 × (15.4 + 16)] = 8742.1 mm2
Gross moment of inertia of built-up beam
Ixx = [35057.6 + 2 × 40 × 1.6 × (23.3)2] × 104 mm4
= 104457.6 × 104 mm4
(Neglecting moment of inertia of plates about their own axes)
Calculated bending tensile stress
⎛ 625 × 1000 × 1000 × 241 ⎞
σbc.cal = ⎜ ⎟⎠
⎝ 104457 ⋅ 6 × 104
= 144.197 Ν/mm2
 DESIGN OF BEAMS 381

Calculated bending tensile stress

⎛ 9480 ⎞
σbt.cal = 144.1 × ⎜ = 156.36 N/mm2
⎝ 8743 ⎟⎠
The calculated bending stress in compression and that in tension do not exceed
(0.66 × 250) = 165 N/mm2 (MPa)
Step 8 : Check for shear force
Average shear stress
⎛ F ⎞ ⎛ 250 × 1000 ⎞
τbc.cal = ⎜ =⎜ ⎟ = 60.386 N/mm
2
⎝ h ⋅ tw ⎟⎠ ⎝ 450 × 9 ⋅ 2 ⎠
< (0.4 × 250 = 100 N/mm2. Hence safe.
Step 9 : Check for deflection
Maximum deflection in the beam

5 ⎛ wl4 ⎞
ymax = ⋅
384 ⎜⎝ EL ⎟⎠

⎛ 5 × 50 ⋅ 00 × 103 × 10003 × 1000 ⎞

ymax = ⎜ ⎟
⎝ 384 × 2 ⋅ 047 × 105 × 104457.6 × 104 ⎠
= 30.44 mm
Allowable deflection
⎛ 1 ⎞
⎜⎝ 325 × 10 × 1000⎟⎠ = 30.57 mm

The actual deflection is less than the allowable deflection. Hence, the section
of the beam selected is safe.
Provide WB 450 @ 0.794 kN/m and one 400 mm × 16 mm plate on each flange.
Example 6.9 A simply supported beam is to support a uniformly distributed
load 70 kN/m excluding weight of the beam, over a clear span of 8 metres. Design
a plated rolled steel beam if MB 500, @ 0.869 kN/m and 10 mm thick plates are
only available. The compression flange of the beam is laterally restrained.
Solution
Design :
Step 1 : Effective span of beam
Clear span of beam = 8
Provide 300 mm wide bed blocks on each side to support the beam
Effective span = Centre to centre of bearing = 8.30 m
Step 2 : Total uniformly distributed load
uniformly distributed load on the beam = 70 kN/m
Assume self-weight of beam = 1.5 N/m
Total uniformly distributed load = 71.50 kN
382 DESIGN OF STEEL STRUCTURES

Step 3 : Bending moment and shear force

The maximum bending moment, M occurs at the centre,
⎛ w ⋅ l2 ⎞ ⎛ 71.50 × 8 ⋅ 30 × 8 ⋅ 30 ⎞
M = ⎜ ⎟ =⎜ ⎟⎠
⎝ 8 ⎠ ⎝ 8
= 615.7 m-kN
The maximum shear force, F occurs at the support,
⎛ 71.50 × 8 ⋅ 30 ⎞
F = ⎜ ⎟⎠ = 296.725 kN
⎝ 2
Let the value of yield stress for the structural steel to be used be 250 N/mm2
(MPa).
Step 4 : Permissible bending stress
The maximum permissible bending stress in compression and that in tension
are equal for the laterally supported beam and it may be assumed as
σbc = ( 0.66 × 250) = 165 N/mm2
Step 5 : Section modulus required
⎛ M ⎞ ⎛ 615 ⋅ 7 ⎞
Z = ⎜ = = 3731.51 × 103 mm3
⎝ σbc ⎟⎠ ⎜⎝ 165 ⎟⎠
Step 6 : Properties of trial section
From the steel section tables, for MB 500, @ 0.869 kN/m
Ixx = 45218.3 × 104 mm4, t w = 10.2 mm
tf = 17.2 mm
Zxx = 1808.7 × 103 mm3, h = 500 mm
bf = 180 mm, ry = 35.2 mm
h2 = 37.95 mm
Provide two cover plates 340 mm × 10 mm on each flange.
Provide 22 mm diameter rivet to connect flange plates.
1 20 1 00 1 20
mm mm mm
2 60 m m

2 P la te s
3 40 m m x 10 m m
54 0 m m

IS M B 50 0 @ 0.8 6 9 kN

Fig. 6.16
 DESIGN OF BEAMS 383

Step 7 : Check for fibre stress

Gross area of tension flange = (180 × 17.2 + 2 × 340)
= 9932 mm2
Net area of tension flange = [9932 – 23.5 (172 +20)]
= 9057.8 mm2
Gross moment of inertia of built-up section
Ixx = [45218.3 + 2 × 2×2 ×34 × (26)2] ×104 mm4
= 137418.3 × 104 mm4
Actual bending compressive stress

⎛ 615 ⋅ 7 × 106 × 270 ⎞

σbc.cal = ⎜ ⎟ = 120.973 N/mm
2
⎝ 137418 ⋅ 3 × 104 ⎠
Actual bending tensile strength

⎛ 9932 ⎞
σbt.cal = 120 ⋅ 973 × ⎜ = 132.648 N/mm2
⎝ 9057.8 ⎟⎠
The calculated bending stress in compression and that in tension do not exceed
(0.66 × 250) = 165 N/mm2 (MPa)
Step 7 : Check for shear force
Average shear stress

⎛ F ⎞ ⎛ 296 ⋅ 725 × 1000 ⎞

τca.cal = ⎜ =⎜ ⎟
⎝ h ⋅ tw ⎟⎠ ⎝ 500 × 10 ⋅ 2 ⎠
= 58.18 N/mm2
< (0.4 × 250 = 100 N/mm2 ). Hence, safe.
Step 8 : Check for deflection
Maximum deflection of the beam

5 ⎛ 71 ⋅ 50 × 8 ⋅ 30 × (8 ⋅ 30)3 × (1000)3 × 1000 ⎞

ymax = ×⎜ ⎟ mm
384 ⎜⎝ 2 ⋅ 047 × 105 × 137418 ⋅ 3 × 104 ⎟⎠

= 15.71 mm

⎛ 1 ⎞
Allowable deflection = ⎜ × 8300⎟ = 25.538 mm
⎝ 325 ⎠
The maximum deflection is less than the allowable deflection. Hence, the
section of beam selected is safe.
Provide MB 500, @ 0.869 kN/m with two cover plates 340 mm × 10 mm on
each flange as shown in Fig. 6.16.
Example 6.10 In Example 6.9, the designed built-up beams restrained against
torsion only. Determine the maximum uniformly distributed load per metre length
which could be placed over this beam.
384 DESIGN OF STEEL STRUCTURES

Solution
Step 1 : Properties of given section
Effective length of the beam restrained against torsion,
l = Span = 8 m
The built-up section of beam consists of BM 500, @ 0.869 kN/m and two cover
plates 340 mm × 10 mm connected to each flange as shown in Fig. 6.16.
Ixx of MB 500, @ 0.869 kN/m is 1369.8 × 104 mm4
tf = 17.2 mm
⎛ 1 ⎞
Iyy of built-up section = ⎜⎝1369.8 + 2 × × 2 × 343 ⎟ × 104 mm4
12 ⎠
= 14471.1 × 104 mm4
Ixx of built-up section = 137154.3 × 104 mm4
Zxx of built-up section
⎛ 137154 ⋅ 3 × 104 ⎞
⎜ ⎟ = 5079.9 × 103 mm3
⎝ 270 ⎠
Total cross-sectional area of the built-up beam
(11074 + 4 × 340 × 10) = 24674 mm2
Radius of gyration of the section
1/ 2
⎛ 14469 ⋅ 8 × 104 ⎞
ry = ⎜ ⎟⎠ = 76.579 mm
⎝ 24674
Slenderness ratio
⎛ l ⎞ 8000 ⎞
⎜ r ⎟ = ⎛⎜ ⎟ = 104.467
⎝ ⎠y ⎝ 76 ⋅ 579 ⎠
Overall depth D = 540 mm
Mean thickness of flange = (20 + 17.2) = 37.2 mm
⎛ D⎞ ⎛ 540 ⎞
Ratio ⎜⎝ T ⎟⎠ = ⎜⎝ 37.2 ⎟⎠ = 14.516

From IS: 800–1984, Table 6.5

X = 412.55 and Y = 2445.6
Step 2 : Critical bending stress
c2
From Eq. 6.18, fcb = k1(X + k2Y) .
c1
c1 and c2 the lesser and greater distances from the neutral axis to the extreme
fibres of the beam section are equal
k1 = 1.0 for ψ = 1.0 and k 2 = 0.0ω = 0.5
Substituting the respective value in the expression (i)
fcb = 1.0(412.55 + 0.0 × 244.45) 1.0
= 412.55 N/mm2
 DESIGN OF BEAMS 385

Let the value of yield stress for the structural steel be 250 N/mm2.
⎛T ⎞ ⎛ 37 ⋅ 2 ⎞
Ratio ⎜ t ⎟ = ⎜⎝ 10 ⋅ 2 ⎟⎠ = 3.647 > 2.0
⎝ w⎠

⎛ d1 ⎞ ⎛ 540 – (200 + 2 × 37.95 ) ⎞

Ratio ⎜ t ⎟ = ⎜⎝ ⎟ = 43.54
⎝ w⎠ 10 ⋅ 2 ⎠

⎛ 1344 ⎞ ⎛ 1344 ⎞
⎜⎜ ⎟⎟ = ⎜ ⎟ = 85
⎝ f y ⎠ ⎝ 250 ⎠

⎛ d1 ⎞ ⎛T ⎞
⎜ t > 85 ⎟ . But ⎜ t > 2.0 ⎟
⎝ w ⎠ ⎝ w ⎠
Therefore fcb = 412.55 N/mm2
Step 3 : Permissible bending stress
From IS: 800–1984, Table 6.2
σbc = 123.255 N/mm2
Step 4 : Moment of resistance
Mr = (σbc.Zxx)
⎛ 123.255 × 5089.56 × 103 ⎞
Mr = ⎜ ⎟
⎝ 1000 × 1000 ⎠
= 627.314 kN-m
Maximum bending moment
2
M = Mr = ⎛ w ⋅ l ⎞
⎜ 8 ⎟
⎝ ⎠
Step 5: Load which may be supported
⎛ w × 8 × 8 × 1000 ⎞
⎜⎝ 8 ⎟⎠ = (627.314 ×1000)
w = 78.414 kN/m
Self weight of beam = (869 + 1078) N/m
= 1.947 kN/m
The uniformity distributed load which can be placed over the beam
= (78.414 – 1.947) = 76.467 kN/m.

6.13 DESIGN OF RIVETS CONNECTING COVER PLATES WITH

THE FLANGES OF BEAM
The pitch of rivets connecting cover plates with the flanges of rolled steel beam
is determined knowing the shear force at any section of beam.
The shear stress at any point of the cross-section is given by
386 DESIGN OF STEEL STRUCTURES

⎛ F ⋅Q ⎞
fx = ⎜ ⎟
⎝ I ⋅b ⎠
where, F = Shear force at the cross-section of beam.
Q = Static moment about the centroidal axis of the portion of cross-
sectional area beyond the location at which the stress is to be
found.
I = Moment of inertia of the section about the centroidal axis.
b = Width of section at which the stress is to be found.
The shear force per unit depth of the beam
The shear force per unit depth of the beam
Fs = (F·Q/I)
The horizontal shear force per unit length of the beam is, equal to the verti-
cal shear force per unit depth of the beam. The horizontal shear force per unit
length is found at the level where plates are connected to the flange.
The horizontal shear force per unit length of beam

Fs´ = ⎛⎜ FQ ⎞⎟
⎝ I ⎠
The horizontal shear force per pitch length
⎛ FQ ⎞
= p × Fs´ = p × ⎜ ⎟
⎝ I ⎠
where, p = pitch of the rivets
The horizontal force per pitch length is resisted by strength of rivet 9 i.e.,
rivet value R,
⎛ FQ ⎞ ⎛ R⋅I ⎞
p. ⎜ ⎟ = R, or p = ⎝⎜ F ⋅ Q ⎠⎟
⎝ I ⎠
In chain riveting, rivet value R is taken as strength of two rivets. In zig-zag or
staggered riveting, staggered pitch of rivet is found and rivet value R is taken as
strength of one rivet. The pitch of rivets in chain riveting should not exceed 16 t or
200 mm whichever is less in tension flange and it should not exceed 12 t or 200
mm whichever is less in compression members where t is the thickness of cover
plate. When rivets are staggered at equal intervals and the gauge space does not
exceed 75 mm, the above limits of pitch may be increased by 50 per cent.
Example 6.11 In Example 6.9, design rivets connecting flange of rolled steel
beam and cover plates.
Solution
Design :
Step 1 : Static moment
Maximum shear force
F = 296.725 kN
Gross-moment of inertia of the built-up section
Ixx = 137418.3 × 104 mm4
Provide 22 mm diameter power driven rivets in staggered manner. Static
moment of cover plates about xx-axis.
Q = Ay
 DESIGN OF BEAMS 387

where, A = Gross-sectional area

y = Distance of C.G. of cross-sectional area from xx-axis
Q = (340 × 20 – 23.5 × 20) × 260 = 1645.8 × 103 mm3
Step 2 : Rivet value
Strength of power driven rivet in single shear
⎛π 2 100 ⎞
⎜⎝ 4 (23 ⋅ 5) × 1000 ⎟⎠ = 43.35 kN
Strength of power driven rivet in bearing
⎛ 23 ⋅ 5 × 17.2 × 300 ⎞
⎜⎝ 1000 ⎟⎠ = 121.26 kN
Rivet value R = 43.35 kN
Step 3 : Pitch of rivets
⎛ R ⋅ I xx ⎞
p = ⎜
⎝ F ⋅ Q ⎟⎠

⎛ 43 ⋅ 35 × 137418 ⋅ 3 × 104 ⎞
⎜ ⎟ = 122.04 mm
⎝ 29 ⋅ 725 × 1645 × 103 ⎠
Pitch of rivets in one line = (2 × 122.04)
= 244.08 mm
Gauge space is 100 mm
Maximum pitch of rivet in compression flange
= (l2 × 10) = 120 mm
Maximum pitch of rivets in tension flange
= (l6 × 10) = 160 mm
Provide a pitch of 120 mm in compression flange and 160 mm in tension
flange throughout the length of built-up beam.

6.14 LINTELS
The brick masonry or stone masonry over doors, windows and other openings in
walls is supported by beams. Such beams are known as steel lintel or simply as
lintels. The various types of lintels are shown in Fig. 6.17.

(a ) (b ) (c) (d )

Fig. 6.17 Steel beams used as lintels

388 DESIGN OF STEEL STRUCTURES

It is considered that the lintel-supports the masonry contained in an equilat-

eral triangle as shown in Fig. 6.18. The masonry should have well bond and
should have an arch action. The masonry should extend at least half the effec-
tive span to either side of the opening. The masonry above the lintel should be
upto a height of 1.25 times the effective span.Then, the load is considered as
uniformly distributed load.

6 0° 6 0°

+ + + + +

I-S ection lintel

Fig. 6.18

When the height of masonry above the lintel is not sufficient it is considered
to support whole masonry about it. The floor or roof loads which are immedi-
ately above the opening, are considered as separate load to be supported by the
lintels.
Example 6.12 A lintel consist of two LC @ 0.206 kN/m connected by pipe
and bolt separator as shown in Fig. 6.17 (b), The effective span is 3.50 m. Deter-
mine the maximum uniformly distributed load for the lintel.
Solution
Step 1 : Properties of given section
Two LC 200, @ 0.206 kN/m are used as a lintel.
From the steel section tables, the section modulus of two LC 200, @ 0.206
kN/m
Zxx = (2 × 172.6 × 103)
= 345.2 × 103 mm3
Step 2 : Moment of resistance of the lintel
⎛ 165 × 345 ⋅ 2 × 103 ⎞
Mr = (σbc × Z ) = ⎜ ⎟
⎝ 1000 ⎠
= 56.958 m-kN
Step 3 : Maximum bending moment
⎛ wl2 ⎞
M = ⎜ ⎟
⎝ 8 ⎠
Step 4 : Load supported by beam

⎛ w × 3 ⋅ 5 × 3 ⋅ 50 × 1000 ⎞
∴ ⎜⎝ 8 ⎟⎠ = 56.958 × 1000 × 1000
 DESIGN OF BEAMS 389

w = 37.197 kN/m
Self-weight of lintel = 2 × 206 = 412 N/m
Load imposed on lintel = (37.197 – 0.412)
= 26.785 kN/m
Example 6.13 A 300 mm thick wall bonded brick well is to be supported over
an opening of 2.75 m. An additional load of 15 kN/m is imposed over the wall
from the roof. Arch action exists. Design a suitable section for the lintel. Weight
of brick masonry = 21.60 kN/m3.
Solution
Design :
Step 1 : Effective span of lintel
Width of the opening = 2.75 m
Provide 150 mm wide bed blocks to support the lintel.
The effective span of lintel is the distance between centre to centre of bear-
ings
l = (2.75 + 0.15) = 2.90 m
Step 2 : Load supported by beam
Weight of brick masonry in equilateral triangle
⎛1 3 ⎞
⎜ 2 × 2 ⋅ 90 × 2 ⋅ 90 × 2 × 0 ⋅ 30 × 21 ⋅ 60⎟ = 23.598 kN
⎝ ⎠
Additional load imposed over the lintel from roof
= (2.90 × 15.00) = 43.50 kN
Self-weight of lintel = 2.85 kN
Total uniformly distributed load
– (23.598 + 43.50 + 2.85) = 69.948 kN
Step 3 : Bending moment
The maximum bending moment, M occurs at centre
⎛ wl2 ⎞ ⎛ 69 ⋅ 948 × 2 ⋅ 90 ⎞
M = ⎜ ⎟ =⎜ ⎟⎠ = 25.356 kN-m
⎝ 8 ⎠ ⎝ 8
Step 4 : Shear force
The maximum shear force, F occurs at the centre
⎛ wl ⎞ ⎛ 69 ⋅ 948 ⎞
F = ⎜ ⎟ =⎜ = 34.974 kN
⎝ 2⎠ ⎝ 2 ⎟⎠
Step 5 : Section modulus required
⎛ 25 ⋅ 356 × 1000 × 1000 ⎞
Z = ⎜ ⎟⎠
⎝ 165
= 153.67× 103 mm3
Step 6 : Trial section
From steel section tables, try two JC 175, @ 0.112 kN/m
390 DESIGN OF STEEL STRUCTURES

Section modulus provided

= (2 × 82.3 × 103)
= 164.6 × 103 mm3
Thickness of web, tw = 3.6 mm
Depth of section = 175 mm
Moment of inertia of section
Ixx = 719.9 × 104 mm4
Step 7: Check for shear force

⎛ F ⎞ ⎛ 34 ⋅ 974 × 1000 ⎞
τva.cal = ⎜ =⎜ ⎟
⎝ h ⋅ tw ⎟⎠ ⎝ 2 × 175 × 6 ⎠
= 27.757 N/mm2
Step 8 : Check for deflection
Maximum deflection of lintel
2
⎛ 5 ⎞ ⎛ wl ⎞
ymax ×
= ⎜⎝ 384 ⎟⎠ ⎜ ⎟ (3wl = 69.948)
⎝ EI ⎠

⎛ 5 × 69 ⋅ 948 × (2 ⋅ 90)3 × (1000)3 × 1000 ⎞

= ⎜ ⎟
⎜⎝ 384 × 2 ⋅ 047 × 105 × 2 × 719 ⋅ 9 × 104 ⎟⎠

= 7.537 mm

⎛ 1 ⎞
Allowable deflection = ⎜ × 290⎟
⎝ 325 ⎠
= 8.923 mm
The actual deflection is less than the allowable deflection. Hence design is
safe. Provide 2 JC, @ 0 .112 kN/m.

6.15 JACK ARCH

Jack-arches are constructed with bricks laid on edge, springing from and sup-
ported on the lower flanges of rolled steel beams. The spacing of rolled steel
beams shall be generally 1.0 to 1.4 metres, but shall never exceed 20 metres.
The beams are designed to support the portion of floor coming over them. Jack
1 1
arches are arches of small rise. Rise of the arch is kept about
th to th of the
8 12
span. The intermediate beams receive thrust from arches on either side and
thrust is balanced. The end beams receive side thrust from the arch. The rods
are provided to resist the side thrust of end beams. The tie rods are properly
anchored with stout mild steel nut and washers. The details of jack arch floor
construction are shown in Fig. 6.19.
 DESIGN OF BEAMS 391

C e m e n t con crete
L evelling cou rse o f lim e con cre te , 6 0 m m m in thick a t cro w n

Tie ro d R ise
S tee l b ea m s en ca sed in ce m e nt con crete

Fig. 6.19 Jack arch

Specification. ISI recommends in IS: 2118–1962 the following specifica-

tions for the design and construction of jack arch floor or roof:
1
1. The deflection of the beams shall not exceed of the span.
480
2. The thickness of arch shall be not less than 100 mm.
3. The tie rods shall be not less than 12 mm in diameter.
4. The tie rods shall be placed generally not less than 75 mm above the
bottom of the joist. The spacing of the tie rods shall not exceed 20 times
the width of rolled steel beams.
The tie rods are designed to resist side thrust found as follows :

⎛ WL ⎞
T = ⎜
⎝ 8 R ⎟⎠
where, F = Side thrust per metre width of arch
W = Live and dead load per metre width of the arch
L = Span of the arch
R = Rise of the arch.
Design of jack arch consists of design of
(i) thickness of arch, (ii) the rods, and; (iii) rolled steel beams.
Example 6.14 Design a jack arch roof for a room of 6 metres × 4.5 metres.
Total uniformly distributed load including self weight of the roof is 9.750 kN/
square metre. Allowable compressive stress is masonry is 0.3 N/mm2.
Solution
Design of Jack arch :
Step 1 : Load supported
Provide rolled steel beams on span of 4.5 m, at a regular interval of 1.5 m.
Span of jack arch = 1.5 m

⎛1 ⎞ ⎛1 ⎞
Provide rise of the arch = ⎜⎝ 12 × span⎟⎠ = ⎜⎝ 12 × 1 ⋅ 5 × 1000⎟⎠
392 DESIGN OF STEEL STRUCTURES

= 125 mm
Load per metre width for one span of arch
W = (1 ⋅ 2 × 1 × 9 ⋅ 75) = 14.625 kN/m
Step 2: Side thrust from the arch
⎛ WL ⎞ ⎛ 14 ⋅ 625 × 1 ⋅ 50 × 1000 ⎞
T = ⎜ =
⎝ 8 R ⎟⎠ ⎜⎝ 8 × 125 ⎟⎠
= 21.938 kN
Let t be the thickness of arch at the crown, A be the area of arch ring per
metre width and ac be the allowable compressive stress in masonry
A × σc = T
(1000 × t × 0.3) = (21.938 ×1000)
∴ t = 73.125 mm
Provide minimum thickness = 100 mm
Step 3 : Rolled steel beam
Clear span of beam = 4.5 m
Assume bearing of 300 mm on either sides of beam
Effective span of the beam is the distance between centre of bearing
= 4.80 m
Uniformly distributed load w = 4.625 kN/m
Step 4 : Maximum bending moment, M occurs at the centre.
⎛ wl2 ⎞ ⎛ 14 ⋅ 625 × 4 ⋅ 8 × 4 ⋅ 8 ⎞
M = ⎜ = ⎟⎠ = 42.12 m-kN
⎝ 8 ⎟⎠ ⎜⎝ 8
Step 5 : Maximum shear force, F occurs at the support
wl ⎛ 14 ⋅ 625 × 4 ⋅ 8 ⎞
F =
= ⎟⎠ = 35.1 kN
2 ⎜⎝ 2
Step 6 : Section modulus required
⎛ 42 ⋅ 12 × 1000 × 1000 ⎞
Z = ⎜ ⎟⎠ = 255.273 × mm
3
⎝ 165
Step 7 : Properties of trial section
From steel section Table, try LB 250, @ 0.279 kN/m
Thickness of web tw = 6.1 mm
Depth of section h = 250 mm
Width of section b = 125 mm
Section modulus Zxx = 297.4 × 103 mm3
Moment of inertia Ix = 3717.8 × 104 mm4
Step 8 : Check for shear
⎛ F ⎞ ⎛ 35 ⋅ 10 × 1000 ⎞
Average shear stress = ⎜ =⎜ ⎟ = 23.02 N/mm
2
⎝ h ⋅ tw ⎟⎠ ⎝ 251 × 61 ⎠
< (0.4 × 250 = 100 N/mm2). Hence, safe.
 DESIGN OF BEAMS 393

Step 9 : Check for deflection

Maximum deflection of beam

5 ⎛ 14 ⋅ 625 × 4 ⋅ 8 × (4 ⋅ 8)3 × (1000)3 × 1000 ⎞

ymax = ×⎜ ⎟ mm
384 ⎜⎝ 2 ⋅ 047 × 105 × 3717.8 × 104 ⎟⎠

= 13.283 mm

⎛ 1 ⎞
Allowable deflection = ⎜⎝ 480 × 4800⎟⎠

= 10 mm
LB 250, @ 0.279 kN/m fails in deflection
Increase the moment of inertia of beam
Moment of inertia of beam required

⎛ 3717 ⋅ 8 × 13 ⋅ 283 × 104 ⎞

⎜ 10 ⎟ = 4860 × 104 mm4
⎝ ⎠
Provide LB 275, @ 0.330 kN/m
Step 10 : Design of tie rods
Spacing of the tie rods shall not exceed 20 times width of the beam
(i.e., 20 × 140) = 2800 mm
Provide the rods at 2.0 m. Thrust resisted by tie rod

= (21.938 × 2.0)
= 43.876 kN
Allowable tensile stress in tie rods upto 20 mm diameter
= (0.6 × 250) = 150 N/mm2

⎛ 43 ⋅ 876 × 1000 ⎞
Net area required = ⎜ ⎟⎠
⎝ 150
= 292.51 mm2
Provide tie rods 25 mm in diameter (ISRO 25).

6.16 CRANE GANTRY GIRDER

The overhead travelling cranes are used in factories and workshops to lift the
heavy materials, equipments etc., and to carry them from one place to the other.
These cranes are either hand operated or electrically operated. The crane consists
of bridge spanning the bay of the shop. A trolley or crab is mounted on the
bridge. The trolley moves along the bridge. The bridge as a whole moves
longitudinally on rails provided at the ends. The rails on either side of the bridge
rest on crane gantry girders. The rails, supporting the crane and resting over
gantry girders are shown in Fig. 6.20 (a) and in partial plan in Fig. 6.20 (b).
394 DESIGN OF STEEL STRUCTURES

R a ils
W h e els for lo ng itu dina l m o ve m en t

Tro lley B ridg e

G an try g ird er

C o lu m n C o lu m n
(a ) C ra ne tra n sverse brid ge

C o lu m n
C o lu m n

Tro lley

D ista nce b etw e en c/c o f cran e w he e ls

S p an fo r g an try g ird er
C o lu m n C o lu m n

(b ) C ra ne g an try gird er

Fig. 6.20

The gantry girders are the girders which support the loads transferred through
the travelling (moving) wheels of the cranes.
The effect of cranes to be considered under the imposed loads shall include
the following:
1. The vertical loads from the crane.
2. The eccentricity effects induced by the vertical loads, and the impact
factors.
3. Lateral (surge) thrust across the crane rail, and
4. Longitudinal horizontal thrust along the crane rail.
The crane loads to be considered shall be as indicated by the customer. In the
absence of any specific indications, the load combinations shall be as follows:
(a) The vertical loads with full impact from loaded crane or two cranes in
case of tandem operation together with the vertical loads, without impact, from
as many loaded cranes as may be positioned for maximum effect along with
maximum horizontal thrust (surge) from one crane only or two cranes in case of
tandem operation;
(b) For the multibay multicrane gantries the loads specified above, subject
to consideration of cranes in maximum of any two bays of the building cross-
section;
(c) The longitudinal thrust on a crane track rail shall be considered for the
maximum of two loaded cranes on the track; and
 DESIGN OF BEAMS 395

(d) The lateral thrust (surge) and the longitudinal thrust acting respectively
across and along the crane rail shall not be assumed to act simultaneously. The
effect of both the forces, shall however be investigated separately.
While investigating the effect of earthquake forces the resulting effect from
dead loads of all the cranes parked in each bay positioned for the maximum
effect shall be considered.
The crane runaway girders supporting bumpers shall be checked for bumper
impact loads.
The stress developed due to secondary effects such as handling, erection,
temperature effects, settlement of foundations shall be appropriately added to
the stresses calculated from the combination of loads as recommended by IS:
800–1984. The total stresses thus calculated shall be within the permissible
limits.
While considered the simultaneous effects of vertical and horizontal surge
loads of cranes, the permissible stresses for the design of crane gantry girders
and their supporting structures may be increased by 10 percent.
Where the wind load is the main acting on the structure, no increase in the
permissible stresses is allowed.
The gantry girder is to be designed for the following additional loads given in
Table 66 as per IS: 875–1984.
The gantry girder is designed on the assumption that either of the horizontal
forces transverse to rails or along the rails act at the same time as the vertical
loads including the impact load. The horizontal forces act at the rail level. The
gantry girder is subjected to bending in vertical plane as well as in horizontal
plane along with twisting. The design calculations are simplified by providing a
channel at the top flange of the girder, and neglecting the bottom flange for
transverse load computations. The transverse loads are comparatively small
and this simplification in design calculations does not result in serious error.
The design example discussed here is for simply supported gantry girder, the
compression flange is subjected to horizontal and vertical loads. The horizontal
forces act in the plane of the top flange and in the direction perpendicular to it.
Therefore, the top flange is strengthened by providing a channel section as shown
in Fig. 6.20 to take the horizontal forces in the plane of this flange. The channel
section provides flange areas to resist bending in horizontal plane due to hori-
zontal forces acting in transverse direction. It increases moment of inertia about
xy-axis. The flanges of channel section resist the bending in the horizontal plane.
The bending of the crane gantry girder occurs about vertical axis as well as
about the horizontal axis of the member. The actual bending stresses for bend-
ing of the girder in the vertical and horizontal planes are computed. The com-
bined bending compressive stress is taken as the sum of the two calculated fibre
stresses. The combined bending compressive stress should be less than or equal
to the allowable bending compressive stress.
(σbc.x.cal + σbc.y.cal) < σbc
where,
sbc.x.cal = Actual bending compressive stress in vertical plane
σbc.y.cal = Actual bending compressive stress in horizontal plane
σbc = Allowable bending compressive stress.
396 DESIGN OF STEEL STRUCTURES

Table 6.6

Type of load Additional load

1. VERTICAL LOAD
(a) for electric overhead cranes 25 percent of maximum static wheel loads
(b) for head operated cranes 10 percent of maximum static wheel loads
2. HORIZONTAL FORCES TRANSVERSE TO RAILS
(a) for electric overhead cranes 10 percent of weight of trolley and weight
lifted on the crane
(b) for hand operated cranes 5 percent of weight of trolley and weight
lifted on the crane.
3. HORIZONTAL FORCES ALONG THE RAILS
5 percent of weight of trolley and weight
lifted on the crane
For detailed classification of cranes, the code of practice for design,
manufacture, erection and testing (structural portion of cranes and hoists IS:
807–1976) may be referred.
The additional load for overhead travelling cranes has been given in Table
6.6 as the percentage of the maximum static wheel load. The maximum wheel
load is the reaction on a wheel due to the total load given by weight of crane
plus crab and the lifted load.
The term ‘maximum static wheel load’ needs explanation. The minimum
approach of crane hook is the distance from the gantry girder upto which the
crab or trolley may approach. This extreme position of the crab with respect to
the span of crane gives the maximum reaction on one of the gantry girders. This
maximum reaction is distributed equally among the crane wheels. The reaction
on each wheel is termed as the maximum static wheel load.
The allowable bending compressive stresses for bending in horizontal plane
is equal to the allowable bending stresses in tension. The allowable bending
compressive stress for bending in vertical plane is reduced in proportion of critical
stress in bending.
IS: 800–1984 recommends that the allowable stress in axial tension, axial
compression, and bending stresses, and allowable stresses for rivets are increased
by 10% for the design of gantry girder for the combination of vertical and
horizontal loads as discussed above. This increase in allowables stresses is not
in addition to that allowed for erection loads with or without wind or seismic
forces.
The bridge of a crane as a whole moves longitudinally on rails provided at the
ends and mounted on the crane gantry girders. The sections of rails are known
as crane rails sections. These sections are designated by the letter ISCR followed
by the head width of the rail section in millimeters. The dimensions of crane
rail sections are given in Table 6.7 and are shown in Fig. 6 .21 (a) as per IS:
3443–1966 (Indian Standard Specifications for Crane Rail Sections). The
 DESIGN OF BEAMS 397

sectional properties of the crane rail sections are given in Table 6.8. The
dimensions corresponding to Table 6.7 are shown in Fig. 6.21 (b). The dimensions
of crane rail section ISCR 140 are shown in Fig. 6.21 (c).

C
D

E
R5 R4
R1
R2
F A

R3

R6 G
(a )
B R7
b
y

ex

x x h
tw
cx

y
(b ) b1
140

10 R
6R 35
64
13 R
38 R
152 5 82

40 38 R
15

37 35 35

140
All d im en sio ns in m m
(c) C ra ne ra il se ctio n (cr 14 0)

Fig. 6.21 Crane rail section

In selecting a crane gantry track rail for a given application on the basis of
the crane manufacturer’s specified maximum wheel load (and the wheel diam-
eter) and by working to a table of maximum permissible wheel loads as recom-
mended by some authority for a range of rails, it is desirable to remember that
the maximum wheel load occurs rarely in service. The maximum wheel load is
reached when the load on the hook of the crane is at full rated capacity and
when the trolley is at the limit of its traverse motion. These two conditions
398 DESIGN OF STEEL STRUCTURES

rarely occur together. Actual wheel loads in service conditions (and with the
crane on its long travel motion) normally lie in the range of 50 to 75 percent of
the specified possible maximum wheel load. The various formulae are used in
determining maximum wheel load. One such formula in use in U.K. for steel
mill cranes converted to metric unit is as follows:

⎛ 4.90a ⋅ D ⎞
W = ⎜⎝ 1000 ⎟⎠

where, W = Maximum permissible wheel load in kN

a = Rail head width in mm, and
D = Wheel tread diameter in mm
The crane gives the satisfactory services when the gantry girder is properly
designed and the foilowing matters are ensured in its subsequent maintenance:
(i) The laid track is reasonably straight. The deviation should not be more
than 5 mm for every 10 m length of long travel track.
(ii) The track span dimension is fairly constant. The variation should not
be more than 10 mm on either side.
(iii) The ends of the rails at the joints are held in close and hard contact
except where expansion joints are necessary in long gantries which themselves
have definite expansion joint details.
Where rails and wheels do not give a satisfactory life it will generally the
consequence of shortcomings in one or more of (i), (ii) and (ii) and/or defect in
the crane wheel bearing, or end carriage which make the wheels out of alignment
in plan or displaced from the vertical in elevation. These defects and also
shortcomings in (i) and (ii) above cause excessive wear of the wheel flanges and
also on the sides of rail head. Poorly designed and poorly maintained rail joints
will not normally adversely affect the life of the wheels or even the wheel bearings
if these later are suitably lubricated plain bush type.

Table 6.7 Dimensions of Crane Rails as per IS : 3443–1996

Desig- Dimension in mm
nation A B C D E F G H R1 R2 R3 R3 R5 R6 R7
CR 50 90 90 55 50 25 20 20 9.70 300 26 18 6 6 5 1.5
CR 60 105 105 65.5 60 27.5 24 22 9.82 350 32 20 6 6 5 1.5
CR 80 139 130 87 80 35 32 26 9.75 400 44 26 8 8 6 1.5
CR 100 150 150 108 100 40 38 30 11.25 450 50 30 8 8 8 2.0
However, bed rail joints shorten considerably the life of antifriction bearings
in crane-wheels. Also bad rail joints shorten the life of the rails. Quite often
poorly designed and poorly maintained rail joints deteriorate to a point where it
is deemed desirable to replace along lengths of rails in which there is no defect
whatsoever, except at the joints.
Table 6.8 Sectional properties of crane rail section as per IS : 3443–1966

Moment Section
of Inertia modulus
Design Area Weight Head Bottom Overall Wcb c2 ex Ixx Iyy Zx1 Zx2 Zy
nation width width Height thickness
(b) (bw) of Narrowest ⎛ I xx ⎞ ⎛ I xx ⎞ ⎛ I yy ⎞
section Point (tw) = ⎜ c ⎟ = ⎜ c ⎟ =⎜b ⎟
⎝ x ⎠ ⎝ x ⎠ ⎝ 1/ 2 ⎠
(h)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)
mm2 N/m mm mm mm mm mm mm cm4 cm4 cm3 cm3 cm3
CR 50 3802 298.5 50 90 90 20 432 468 357.54 111.42 82.76 76.40 24.76
CR 60 5099 4003 60 105 105 24 483 567 654.60 195.88 13552 115.45 37.31
DESIGN OF BEAMS

;
CR 80 8113 6389 80 130 130 32 643 657 1547.40 482.39 240.65 235.52 74.21
CR 100 11332 889.6 100 150 150 38 760 740 286473 940.98 376.94 387.12 125.46
399
400 DESIGN OF STEEL STRUCTURES

6.16.1 Limitation of Vertical Deflection

When the crane structure is subjected to vibration or shock impulses, it may be
desirable to maintain reasonable deflection limits such as will produce a stiff
structure less apt to vibrate and shake appreciably. For example, the excessive
deflection in crane runway support girder will lead to uneven up and down
motion of the crane as it proceeds. The impact stresses in such cases would be
increased.
In case of crane runway girder, the maximum vertical deflection under dead
and imposed loads as per the recommendations of IS: 800–1984, shall not ex-
ceed the following values:
⎛ L ⎞
(i) Where the cranes are manually operated and to similar loads ⎜
⎝ 500 ⎟⎠

⎛ L ⎞
(ii) Where the cranes are overhead travelling and operated ⎜ electrically
⎝ 750 ⎟⎠
upto 500 kN
⎛ L ⎞
(iii) Where the electric overhead travelling cranes operated over ⎜
⎝ 1000 ⎟⎠
500 kN
⎛ L ⎞
(iv) Other moving loads, such as charging cars, etc. ⎜
⎝ 600 ⎟⎠
where, L is the span of crane runway girder.

6.16.2 Limitation of Horizontal Deflection

The horizontal deflection at the column cap level of columns supporting crane
runway girders in the building shall not exceed as may be specified by the pur-
chaser.
Example 6.15 Design a simply supported gantry girder to carry one electric
over head travelling crane.
Crane capacity = 300 kN
Weight of crane excluding trolley = 190 kN
Weight of trolley = 100 kN
Minimum approach of crane hook = 1.2 metres
Distance between centres of crane wheel = 3.5 metres
Distance between centres of crane wheel = 18 metres
Span of gantry girder = 6 metres
Weight of rail section = 0.300 kN/m
Height of rail section = 75 mm
Solution
Design :
Step 1 : Maximum wheel load
Weight of trolley + lifted load = (300 + 100) = 400 kN
 DESIGN OF BEAMS 401

The weight of crane (excluding trolley) 190 kN acts as uniformly distributed

live load as shown in Fig. 6.22.
The vertical reaction on each wheel of crane would be maximum, when trol-
ley is at nearest distance to trolley girder as shown in Fig. 6.22.
4 00 kN

1 90 kN w t of cra ne

1 .2 m

A B
Fig. 6.22

Take moment about B, then reaction at A

1 ⎡ 18 ⎤
RA = ⎢ 400 × 16 ⋅ 8 + 190 × ⎥ = 468 kN
18 ⎣ 2⎦
This vertical load at one end of the crane bridge is transferred to the gantry
girder through two wheels.
Maximum vertical load on each wheel of crane

⎛1 ⎞
= ⎜⎝ 2 × 468⎟⎠ = 234 kN

Step 2 : Maximum bending moment (due to D.L. + L.L. + J.L.)

The maximum bending moment in the gantry girder under a moving load
occurs when the line of action of that load and C.G. of the loads are at equal
distance from the centre of span. That is,
EC = CF = 0.875 (Fig. 6.23)
C .G . of lo ad s
2 34 kN 2 34 kN
0 .37 5
1 .75 m 1 .75 m 2 .12 5 m
A D E C F B

0 .87 5 0 .87 5
3m 3m

Fig. 6.23

The reaction at the supports A and B are as follows :

1
RA = 234 × ⎡(6 − 0.375) + 2.125 ⎤⎦
6⎣
= 302.23 kN
RB = (2 × 234 – 302 ⋅ 23)
= 165.77 kN
402 DESIGN OF STEEL STRUCTURES

Maximum bending moment due to moving load

MF = (l65.77 × 2. 125) = 352.3 kN
Add 25 per cent impact moment viz., 88.1 kN-m
(1) Live load moment
= (352.3 + 88.1) = 440.4 kN-m
Assume self-weight of the girder as 2 kN/m
Weight of rail section is 0.300 kN/m
Total dead load
= 2.3 kN/m
Maximum bending moment due to dead load
⎛ wl2 ⎞ ⎛ 2 ⋅ 3 × 6 × 6⎞
⎜ 8 ⎟ = ⎜⎝ 8 ⎟⎠ =10.35 kN-m
⎝ ⎠
(2) Dead load moment = 10.35 kN-m
(3) Total vertical moment = (440.4 + 10.35) = 450.75 kN-m
Assume allowable bending compressive stress
= (0.66 × 250) = 165 N/mm2
The section modulus required for bending moment is vertical plane (approxi-
mately)
⎛ 450 ⋅ 75 × 1000 × 1000 ⎞
Z = ⎜⎝ ⎟⎠
165
= 2731.8 × 103 mm3
From steel section Tables, try WB 600, @ 1.337 kN/m and LC 300, @ 0.331
kN/m.
The section of the gantry is shown in Fig. 6.24.

3 00 m m
2 50 m m
1 40 m m
2 5.5 m m 6 .7 m m

IS LC 3 00 ,
@ 0 .33 1 kN /m
2 8.2 m m 3 00 m m
N A
5 5.6 m m

IS W B 6 0 0,
@ 0 .33 7 kN /m
3 00 m m

Fig. 6.24 Gantry girder

Sectional area of beam section is 17038 mm2

Section area of channel section is 4211 mm2
 DESIGN OF BEAMS 403

∴ Total section area is 21249 mm2

Thickness of flange of beam section, t f is 21.3 mm
Let y be the distance of neutral axis of built-up section from neutral axis of
beam section
Moment of inertia of built-up section about xx-axis
Iyy (gross) = [106198.5 + 170.38 × 5.562 + 346 + 42.11
× (28.12 – 5.56)2] × 104 mm4
= 133334.5 × 104 mm4
Moment of inertia of built-up section about yy-axis
Iyy (gross) = [4702.5 + 6047.9] × 104
= 10750.4 × 104 mm4
Bending stress due to vertical loading
Actual bending compressive stress for vertical loading
⎛ 450 ⋅ 75 × 1000 × 1000 × 251 ⋅ 1 ⎞
σbcx.cal = ⎜ ⎟⎠
⎝ 133334 ⋅ 5
= 84.8867 N/mm2
Actual bending tensile stress for vertical loading
⎛ 450.75 × 1000 × 1000 × 355.6 ⎞
σbcx.cal = ⎜ ⎟⎠
⎝ 122224.5 × 104
= 119.4 N/mm2
< (1.10 ×165 =181.5 N/mm2)

C .G . of lo ad s
2 0 kN 2 0 kN
0 .37 5 m
1 .75 m 1 .75 m 2 .12 5 m
A C
B
0 .87 5 0 .87 5
3m 3m

Fig. 6.25

Step 3 : Maximum bending moment due to horizontal (transverse)

force
Horizontal force transverse to the rail
= 10 per cent of (weight of trolley + lifted load)
1
× (300 + 100) = 40 kN
=
10
Horizontal force transverse to the rail on each wheel or crane
= 20 kN
Horizontal reaction at support A (Figs. 6.23 and 6.25)
20
= × 302 ⋅ 33 = 25.83 kN
234
404 DESIGN OF STEEL STRUCTURES

Horizontal reaction at support B

= 14.17 kN
Horizontal moment
14 .17 × 2.125 = 30.1 kN-m
Step 4 : Bending moment in horizontal plane
Horizontal moment = 30.10 kN-m
The moment of inertia of compression flange about yy-axis (considering Iyy of
compression flange of beam section as half of that for beam section)

⎡ 1 ⎤
Iyy = ⎢6047 ⋅ 9 + 2 × 4702 ⋅ 5 ⎥ × 10
4
⎣ ⎦
= 8399.6 × 104 mm4
Bending compressive stress in horizontal plane (Bottom flange is neglected)

⎛ 30 ⋅ 1 × 1000 × 1000 × 150 ⎞

σbc.y.cal = ⎜ ⎟⎠
⎝ 8399 ⋅ 6 × 104
= 53.58 N/mm2
Step 5 : Allowable stress in horizontal plane
Let y1 be the distance of compression flange from top fibre

⎡ 4211 × 25 ⋅ 5 + 250 × 21 ⋅ 3 (6 ⋅ 7 + 10 ⋅ 65) ⎤

y1 = ⎢ 4211 + 250 × 21 ⋅ 3
⎥
⎣ ⎦
= 20.9 mm
Distance between C.G. to C.G. of top and bottom flanges
h = (605.7– 20.9 – 10.6) = 575.2 mm
Section modulus about xx-axis reference to the compression flange

⎡ 133334 ⋅ 5 × 104 ⎤
Zxx = ⎢ ⎥ = 5308.8 × 103 mm3
⎣ ( 300 + 6 ⋅ 7 – 55 ⋅ 6 ) ⎦
⎛ Moment of inertia of comp. flange about yy-axis ⎞
ω = ⎜⎝ Moment of inertia of built-up section about yy-axis ⎟⎠

⎛ 8399 ⋅ 6 × 104 ⎞
ω = ⎜ ⎟ = 0.78
⎝ 10750 ⋅ 4 × 104 ⎠

From IS: 800–1984, k1 = 0.28

Effective length of compression flange = 6000 mm
Radius of gyration of the complete section about yy-axis

⎛ 10750 ⋅ 4 × 104 ⎞
ry = ⎜ ⎟ = 79.58 mm
⎝ 21249 ⎠
 DESIGN OF BEAMS 405

⎛ 6000 ⎞
Slenderness ratio = ⎜ = 75.39
⎝ 79 ⋅ 58 ⎟⎠
Overall depth, D = 606.7 mm
Mean thickness of flange
T = (t f = 21.3 + 6.7) = 28.0

⎛ D⎞
Ratio ⎜⎝ T ⎟⎠ = 21.668

From IS: 800–1984, Table 6.5

X = 632.02 and Y = 503.27
From Eq. 6.18, the elastic critical stress

⎛c ⎞
fcb = k 1 (X + k 2.Y) ⎜ 2 ⎟
⎝ c1 ⎠

⎛ 306 ⋅ 7 ⎞
= 1.0 (632.02 + 0.28 × 503.27) × ⎜
⎝ 300 ⎟⎠
= 790.20 N/mm2 (MPa)
Let the value of yield stress for the structural steel be 250 N/mm2

⎛T ⎞ ⎛ 28 ⎞
Ratio ⎜⎝ t ⎟⎠ = ⎜⎝ 11 ⋅ 2 ⎟⎠ = 2.5 > 2.0
w

∴ fcb is not increased by 20 per cent From IS: 800–1984, Table 6.2
σbc = 145 N/mm2
Step 6: Check for combined bending compressive stress in extreme
fibre
(σbcx.cal + σbcy.cal) = (84 .498 + 53.58)
137.98 N/mm2 < 1. 1 × 145 = 159.5 N/mm2
Hence design is safe and satisfactory.
Step 7 : Horizontal (longitudinal) force along the rails
⎛ 1 ⎞
5% of the static wheel load = ⎜⎝ 20 × 2 × 234⎟⎠ = 23.4 kN

Height of rail = 75 mm
Bending moment in the longitudinal direction
= 23.4 × (75 + 251.1) = 7630.74 mm-kN
Stress in longitudinal direction
⎛P M⎞ ⎛ 23 ⋅ 4 × 1000 7630 ⋅ 74 × 1000 ⎞
⎜⎝ A + Z ⎟⎠ = ⎜⎝ 21249 +
5308 × 104 ⎟⎠
N/mm2

(1.10 + 14.376) = 2.538 N/mm2 (Very small)

406 DESIGN OF STEEL STRUCTURES

Shear force
Maximum shear force in the gantry girder
⎛ 2 ⋅ 5⎞
⎜⎝ 234 + 234 × 6 ⋅ 0 ⎟⎠ = 331 kN
Add 25% for impact = 82.75 kN
⎡ (1337 + 331) 6 ⎤
Dead load shear = ⎢ 2 × 1000 ⎥ = 5.61 kN
⎣ ⎦
Total shear = 419.36 kN
Intensity of horizontal shear stress per mm length
⎛ FQ ⎞ Q = Ay
fy = ⎜ ( )
⎝ I ⎟⎠
Consider the portion of web of flange only.
Area = (6.7 × 300) = 2000 mm2

From NA, y = 251 ⋅ 1 – 1 × 6 ⋅ 7 = 247.75 mm

2
⎛ 419 ⋅ 36 × 2000 × 247 ⋅ 75 × 1000 ⎞
τva = ⎜⎝ ⎟⎠
13334 ⋅ 5 × 104
= 155.84 N/mm2
Step 8 : Rivet value
Use 22 mm diameter power driven rivets.
Strength of power driven rivets in single shear

⎛ π (23 ⋅ 5)2 × 100 ⎞

⎜ ⎟ = 43.35 kN
⎜⎝ 4 1000 ⎟⎠

Strength of rivet in bearing

⎛ 6 ⋅ 7 × 300 ⎞
⎜⎝ 23 ⋅ 5 × 1000 ⎟⎠ = 47.235 kN

Rivet value, R = 43 .35 kN

⎛ 43 ⋅ 35 × 1000 ⎞
Pitch of rivets = ⎜
⎝ 155 ⋅ 84 ⎟⎠
= 278.17 mm
Rivets are provided in two lines
2.p = 556.34 mm
Maximum allowable pitch in compression
= (12 × 6.7) = 80.4 mm
Provide rivets at 80 mm pitch throughout the length of gantry girder.
 DESIGN OF BEAMS 407

6.17 FILLER JOIST

The steel beams of light sections are termed as joists. The joists placed in plain
cement concrete are called filler joists. When filler joists are continuous over
more than two supports, then, these are called continuous fillers. Whenever
possible, continuous fillers may extend over three spans. The joists are sup-
ported by main steel beams as shown in Fig. 6.26.

C e m e n t con crete sla b C o ve r

25 m m
IS LB 20 0 @ 0.1 9 8 kN /m filler jo ist a t 1 .4 0 m sp a cing
25 m m
C o ve r
M ain be am s
S tirru ps 5 m m φ
1 00 pitch

4 .0 m

Fig. 6.26 Continuous fillers

The joists in continuous fillers are made over supports in such a manner to
preserve considerable measure of continuity. The joints are staggered so that
full continuity of the fillers occurs in 50 per cent of those crossing any support,
such a continuity being evenly displaced.
When filler joists spans between two main steel beams only, then, these are
called discontinuous fillers. Discontinuous filler joists are shown in Fig. 6.27.
These are connected to main beams by means of cleat angles. Overall depth of
C e m e nt con cre te sla b C o ve r

Filler joist

C le a t a ng le s
M ain b e am s

Fig. 6.27 Discontinuous filler

construction in case of discontinuous filler joist is small than that of continuous

filler joist construction.
The bottom of filler joists remains flush with cement concrete slab.
The design of filler joists is done as per following specifications.

6.17.1. Bending Moment

The bending moments on slabs of which filler joists form part, shall be calcu-
lated to satisfy the following :
408 DESIGN OF STEEL STRUCTURES

(a) As slabs spanning continuously over supports and subjected to those

combinations of dead and live loads producing the maximum positive and nega-
tive moment.
(b) In the case of three or more approximately equal spans of continuous
filler joists described above, as slabs designed for uniformly distributed loading,
satisfying moment value given in Table 6.9.
Table 6.9 Bending moment values for filler joists

Locations Values of B.M.

1. Near Middle of end span + wL2/10
2. At support, next to end span – wL2/10
3. At middle of intermediate span + wL2/12
4. At other interior supports – wL2/12
where w is the dead load plus live load per unit length of span and L is the span,
centre to centre of supports.
Note : For single spans, the slab shall be assumed to be freely supports.

The spans are considered approximately equal when the longest span does
not exceed the shortest span by more than 15 percent.

6.17.2 Moment of Resistance

The moment of resistance of the slab shall be calculated from the section prop-
erties of the filler joists only except that where the filler joists are completely
embedded in a solid concrete slab, the moment of resistance of slab may alter-
natively be calculated as a composite reinforced concrete section.

6.17.3 Spacing
The spacing of filler joist centre concrete slab is given in Table 6.10 unless the
concrete is reinforced to span as slab or function as an arch between the filler
joists.

Table 6.10 Spacing of filler joists

Imposed load per m2 Volume of n

1. Upto 2.50 kN 9
2. Above 2.50 kN upto 5.0 kN 8
3. Above 5.0 kN upto 7.50 kN 7
4. Above 7.50 kN upto 10 kN 6
5. Above 10 kN 5
Where a slab of concrete is designed to function as an arch between the filler
joist, the thrust from the arch section shall be taken up by steel tie or by other
means so as not to cause appreciable increase of stress in the joists. The concrete
shall have a working strength of not less than 15 N/mm2 at 28 days.
 DESIGN OF BEAMS 409

6.17.4 Thickness of Concrete

Where the underside of the concrete is arched between the filler joists, the
thickness at the crown shall be not less than 50 mm. The thickness of structural
concrete over hollow blocks shall nowhere be less than 30 mm when the filler
joists exceed 450 mm centre to centre and nowhere less than 50 mm when the
spacing of filler joists is greater than 450 mm centre to centre.

6.17.5 Bending Stress

The bending stress in filler joists other than those designed as part of reinforced
concrete slab, shall not exceed (165 + 0.62 t) N/mm2, where t is an allowance,
the value of which is equal to the thickness in mm of the structural concrete
cover to the compression flange of the filler joists, provided that the allowance t
is applied only in cases where the filler joists are embedded at least flush with
underside of their bottom flanges in a solid concrete slab throughout and that
any cover less than 25 mm and in excess of 75 mm shall be neglected.
In cases where the underside of the slab is flush with the bottom flanges of
the filler joists, the allowance t shall not apply in respect of support moments;
nevertheless if the top flange is covered, the allowance may be made in calculating
to the resistance sagging moment.

6.17.6 Shear Stress

The shear stress shall be calculated as taken entirely on the filler joists and
stress shall not exceed the allowable stress in shear, i.e., 0.4 fy.

6.17.7 Bearing Stress

The bearing stress shall be calculated as taken entirely on the filler joist and
stress shall not exceed the allowable stress in bearing i.e., 0.75 fy.

6.17.8 Span Depth Ratio

The span of filler joist centre to centre of supports, shall not exceed 35 times the
depth from the underside of the joist to the top of the structural concrete or 12
times this depth in the case of cantilever fillers.

6.18 ENCASED BEAMS

When a steel beam is encased in cement concrete throughout the entire length,
it is called encased beam. The cased beam is shown in Fig. 6.28.
A steel beam is embedded in cement concrete for two purposes. It may be
necessary to make a steel frame building fire resisting. The beams may be encased
in cement concrete for architectural requirement. As per IS: 800–1984 beams
may be designed as encased beams when the following conditions are fulfilled :
(a) The section is of single web and I-shape or of double open channel
formed with the web not less than 40 mm apart.
410 DESIGN OF STEEL STRUCTURES

(b) The overall dimensions of the steel section do not exceed 750 mm ×
450 mm overplating where used, the larger dimensions being measured paral-
lel to web (as per IS: 800–1962).

IS W B 6 00 ,
@ 1 .45 5 kN /m

5 mm θ
S tirru ps
a t 1 00 m m
p itch

Fig. 6.28 Cased beam

(c) The beam is unpainted and it is solidly encased in ordinary dense

concrete, with 10 mm aggreagate (unless solidity can be obtained with a large
aggregate) and of grade designation M 15, min (IS: 800–1978).
(d) The minimum width of solid casing is equal to (b0 + 100) mm where b0
is the width of the steel flanges in mm.
(e) The surface and edges of the flanges of the beam have a concrete cover
not less than 50 mm.
(f) The casing is effectively reinforced with steel wire of at least 5 mm
diameter and the reinforcement shall be in the form of stirrups or binding at
not more than 150 mm pitch, and so arranged as to pass through the centre of
the covering to the edges and soffit of the lower flange.

6.18.1 Design of Encased Beam

The steel beam section shall be considered as carrying the entire load but allow-
ance may be made for the effect of concrete on the lateral stability of the com-
pression flange. This allowance should be made by assuming for the purpose of
determining the permissible stress in compression, that the equivalent moment
of inertia, Iy about yy-axis is equal to Ary2, where A is the steel section and ry may
be taken as 0.2 (b0 + 100 mm). The width of flange b0, is taken in mm. The other
properties required for calculating critical bending stress for Eq. 6.11 may be
taken as for uncased section. The permissible bending stress so determined
shall not exceed 15 times that permitted for the uncased sections.
It is noted that this does not apply to beams and girders having a depth
greater than 1000 mm or a width greater than 500 mm or to box sections.
 DESIGN OF BEAMS 411

Example 6.16 A filler joist is continuous over three equal spans. The dis-
tance between centre to centre of main beams is 4 m. The floor consists of 250
mm thick plain cement concrete slab. Unit weight of plain cement concrete is 24
kN/m3. The live load on floor is 5 kN/m2. Design filler joist.
Solution
Design :
Step 1 : Load supported by joist
Distance between centre to centre of main beam = 4 m
Thickness of plain cement concrete slab = 250 mm
Assume, thickness of top cover = 25 mm
Thickness of bottom cover = 25 mm
Therefore, thickness of structural concrete slab = 200 mm
Live load on floor = 5 kN/m2
Therefore, spacing of filler joist centre to centre
n × thickness of structural slab = 7 × 20 = 1400 mm
Dead load
Weight of plain cement concrete inclusive of cover at top and bottom
⎛ 24 × 4 × 1 ⋅ 400 × 200 ⎞
⎜⎝ 1000 ⎟⎠ = 26.88 kN

Self-weight of filler joist assuming 0.200 kN/m

= (0.200 +27.70 × 4) = 0.800 kN
Total dead load = (26.81 + 0.80) = 27.680 kN (say Ω 27.70)
Live load
Total live load = (5 × 4 ×1.40) = 28 kN
Total dead load plus live load
= (28 + 27.70) = 55.70 kN
Step 2 : Maximum bending moment
The maximum bending moment occurs at middle of end span
⎛ 55 ⋅ 70 × 4 ⎞
= ⎜⎝ ⎟⎠ = 22.28 kN-m
10
Allowable bending stress
(165 + 0.62 t) = (165 + 0.62 × 25) = 180.5 N/mm2
Step 3 : Required section modulus for filler joist
⎛ 22 ⋅ 28 × 106 ⎞
Z = ⎜ ⎟ = 126.32 × 103 mm3
⎝ 180 ⋅ 5 ⎠
From ISI Handbook No. 1, try LB 200 @ 0. 198 kN/m
Zxx = 169.7 × 103 mm3
Step 4 : Check
Span of filler joist = Centre to centre of main beams (4 m)
Depth from the underside of the joist to the top of the structural concrete
= 225 mm
412 DESIGN OF STEEL STRUCTURES

⎛ 4 × 100 ⎞
Span depth ratio = ⎜⎝ 22 ⋅ 5 ⎟⎠ = 17.75

< 35. Hence design is satisfactory.

Provide LB 200, @ 0.1908 kN/m filler joist
5 mm in diameter stirrups are provided at 100 mm pitch.
Complete design is shown in Fig. 6.26.
Example 6.17 An encased beam carries total uniformly distributed load of
40.00 kN/m. The effective span of beam is 8 m. The ends of beams are restricted
against torsion, and not restrained against lateral bending. Design the encased
beams.
Solution
Design :
Step 1 : Load supported by beam
Total uniformly distributed load is 40 kN/m
Step 2 : Maximum bending moment, M occurs at the centre
⎛ 40 ⋅ 0 × 8 × 8 ⎞
M = ⎜⎝ ⎟⎠ = 32 kN-m
8
Step 3 : Maximum shear force, F occurs at the support
F = 160.00 kN
Step 4 : Selection of trial section
Select an uncased rolled steel beam section such that the section of beam
provides required section modulus corresponding to the allowable bending com-
pressive stress for the uncased section for the given span.
Select WB 600, @ 1.451 kN/m
σbc = (0.66 × 250) = 165 N/mm2
Step 5 : Section modulus required
⎛ 32 ⋅ 00 × 1000 × 1000 ⎞
Z = ⎜⎝ ⎟⎠
165
= 193.94 × 103 mm3
Step 6 : Properties of trial section
From ISI Handbook No. 1 for WB 600, @ 1.451 kN/m
Sectional area A = 18486 mm2
Depth of section = 6000 mm
Thickness of flange tf = 23 .6 mm
Width of flange b = 250 mm
Section modulus Z = 3854.2 × 103 mm3
Section modulus provided is greater than section modulus required. Hence,
safe.
For encased beam, ry = 0.2 (b0 + 100) mm
= 0.2 (250 + 100) = 70 mm
 DESIGN OF BEAMS 413

Moment of inertia of cased beam

Iy = Ary2 = (18486 × 702)
= 9020 × 104 mm4
Distance between C.G. of flanges
= (600 – 23.6) = 576.4 mm
Step 7: Elastic critical stress in bending
From IS: 800–1984, K = 249.3 × 104 mm4
⎛ D⎞ ⎛ 600 ⎞
⎜⎝ T ⎟⎠ = ⎜⎝ 23 ⋅ 6 ⎟⎠ = 25.42
T = t f = 23.6 mm
⎛T ⎞ ⎛ 23 ⋅ 6 ⎞
Ratio, ⎜⎝ t ⎟⎠ = ⎜⎝ 11 ⋅ 8 ⎟⎠ = 2.0
w

⎛ l⎞ ⎛ 8000 ⎞
Ratio, ⎜ r ⎟ = ⎜⎝ = 114.28
⎝ y⎠ 70 ⎟⎠

⎛ d1 ⎞ ⎛ 600 – 2 × 46 ⋅ 05 ⎞
Ratio, ⎜⎝ t ⎟⎠ = ⎜⎝ ⎟⎠ = 43
w 11 ⋅ 8

⎛ 1344 ⎞ ⎛ 1344 ⎞ ⎛ d1 ⎞ ⎛ 1344 ⎞

⎜ ⎟ = ⎜ = 85, ⎜⎝ t ⎟⎠ > ⎜⎜ f ⎟⎟
⎜⎝ fy ⎟⎠ ⎝ 250 ⎟⎠ w ⎝ y ⎠

The value of fcb is to be increased by 20 percent From IS: 800–1984, Table 6.5
X = 289.18 and 7 = 204.2
⎛c ⎞
From Eq. 6.18 fcb = k 1(X + k 2 T) ⎜ 2 ⎟ ...(i)
⎝ c1 ⎠
c1 and c2 the smaller and the larger distances from the neutral axis to the
extreme fibres are equal
k1 = 1.0 for ϕ=1.0, and k 2 = 0.0 for ω = 0.5
Substituting the respective values in expression (i)
fcb = 1.0 (289.18 + 0.0 × 204.2) 1.00
= 289.18 N/mm2
fcb = (1.20 × 20 × 289.18)
= 347.02 N/mm2 (MPa)
Step 8 : Permissible stress in bending. The permissible stress in bend-
ing in compression from Table 6.2
σfc = 116.4 N/mm2
From IS: 800–1984, allowable bending compressive stress
= 165 N/mm2
> (1.5 × 116.4 = 174.6 N/mm2). Hence, safe.
|
414 DESIGN OF STEEL STRUCTURES

Provide WB 600, @ 1.451 kN/m. The beam is cased in concrete. The cement
concrete provides a 50 mm cover on all sides. 5 mm diameter stirrups are provided
at 100 mm pitch (| > 150 mm). The section of cased beam is shown in Fig. 6.28.
The overall dimensions of steel section are not greater than the allowable
dimensions.

PROBLEMS
6.1. A laterally supported beam LB 350, @ 0.495 kN/m has been placed on two
supports. Determine the safe uniformly distributed load per metre length
which can be placed over the beam for an effective span of 6.30 m.
6.2. Design a beam of 5 metres span carrying a uniformly distributed load of
30 kN per metre including self-weight and a concentrated load of 10 kN at
the centre. The beam is simply supported at both ends.
6.3. A simply supported beam of 12 metres span is made of MB 600, @ 1.226
kN/m. The associated properties of the beam are
Depth = 600 mm
Width of flange = 210 mm
Thickness of flange = 20.8 mm
Thickness of web = 12 mm
Ixx = 91813 ×104 mm4, Iyy= 2651 × 104, Zxx = 3060.4 × 103 mm3
Determine that total uniformly distributed load that this beam can carry.
The top flange is fully restrained laterally.
6.4. A beam carries a uniformly distributed load 44.00 kN/m. The effective
span of beam is 6.40 m. The ends of beam are restrained against torsion
and not restrained against lateral bending. Design a suitable section of a
rolled steel beam.
6.5. Two LB 600, @ 0.995 kN/m joists at 300 mm with one 36 mm thick plate,
600 mm wise on each flange is used as a simply supported beam on a span
of 12 m. Find the maximum uniformly distributed load per metre run which
can be allowed on this built-up girder. Neglect rivet holes.
6.6. Design a simply supported plated rolled steel beam section to carry a
uniformly distributed load 36.00 kN/m inclusive of self-weight of beam.
The effective span of beam is 7.20 m, the depth of beam is limited to 500
mm. The beam is restrained against lateral deflection.
6.7. Design a beam of 6 m span carrying uniformly distributed load of 10 kN/m
and two concentrated loads of 80 kN each at 2 m and 4 m, respectively
from the left support. Only available rolled steel section is ISMB 350, @
0.524 kN/m. The compression flange of the beam is laterally supported.
6.8. A beam of 6 m span carries uniformly distributed superimposed load of 75
kN/m and laterally unsupported. (i) Design the central section of the beam
using an MB 600 and suitably chosen cover plates attached to both the
flanges.
6.9. Design a beam of 5.4 m span carrying a uniformly distributed load of 30
kN/m and two concentrated loads of 50 kN each at third points. The depth
 DESIGN OF BEAMS 415

of the section is restricted to 350 m. Assume the compression flange to

have lateral restraint.
6.10. A freely supported beam spanning 12 m is made up of an I-section MB
500, @ 0.869 kN/m with two pates in each flange of 200 mm × 12 m each.
Calculate the maximum uniformly distributed load the beam can carry if
the compression flange is fully restrained. Assume the flange connections
to be made with 18 mm φ rivets. Show calculations for the curtailment of
the outermost plate in each flange.
6.11. A plated steel beam of span 12 metres effective is to carry a uniformly
distributed load of 20 kN per metre inclusive of its own weight. Design a
suitable plated beam including minimum spacing of rivets using any one
of the three section listed below. Plate of 10 mm thickness is available in
any width. The flange of beam should not be wider than 300 mm. Assume
that 22 mm diameter rivets are used for the connection.
MB 500, @ 0.869 kN/m. LB 600, @ 0.995 kN/m
LB 500, @ 07.50 kN/m.
6.12. A simply supported beam of effective span 8 metres is required to carry
a.u.d.l. of 50 kN/m including self-weight over the entire span. Check
whether the LB 450, @ 0.653 kN/m with the following properties is suitable
for the purpose. (Area = 8314 mm2; bf = 170 mm; tj = 13.4 mm; t w = 8.6
mm; Ixx = 27566.1 × 104 mm4). If not, design a safe section using 150 mm ×
10 mm thick plates.
Design a suitable welded connection between top plate and flange of the
joint.
Assume that the compression flange is adequately restrained laterally.
6.13. Design a jack arch roof for a hall of size 8 m × 4 m. The self-weight of the
roof may be taken as 8.5 kN/m2. The permissible compressive stress in the
brick masonry is 0.3 N/mm2.
6.14. A jack arch roof is to be constructed for a room measuring 5.5 m × 4.2 m.
The live loaded acting is 4 kN/m2. Design the roof assuming appropriate
self-weight for the roof.
6.15. Design a gantry girder to carry an electric overhead travelling crane to
suit following data:
Crane capacity ... 200 kN
Wt. of crab alone ... 70 kN
Wt. of crane ... 150 kN
Minimum approach of crane hook ... 1.2 m
Dist. between centres of crane wheels ... 3.5 m
Dist. between cranes of gantry girders ... 15.0 m
Span of gantry girder ... 7m
Wt. of rail section ... 0.300 kN/m
Height of rail section ... 80 mm
416 DESIGN OF STEEL STRUCTURES

18 m

C ra ne g ird er
C ra b ho ok
C ra ne tra n sverse b rid ge

18 m

C /L of colum ns
W h e els
C ra ne g ird er

C /L of colum ns

Fig. P 6.16
6.16. A gantry girder is provided for mill building to carry an electric overhead
travelling crane with the following data as shown in Fig. P 6.16.
(i) Centre to centre distance between columns (span of gantry girder) = 8
metres
(ii) Centre to centre distance between gantry girders = 18 metres
(iii) Crane capacity = 300 kN
(iv) Self-weight of crane hook = 30 kN
(v) Self weight of crane hook = 30 kN
(vi) Allowable approach of crane hook from the vertical axis of the gantry girder
= 1.2 metres
(vii) Distance between centres of wheels, moving on the gantry girder and sup-
porting the crane girder = 3 .5 metres
(viii) Weight of rail section attached on the top of the gantry girder = 0.300 kN/
metre
Determine :
(a) The maximum B.M. and S.F. produced by the vertical loads
(b) The maximum B.M. and S.F. produced by the lateral loads
(c) The maximum longitudinal force exerted along the rails.
Also design the section of gantry girder.
6.17. A gantry girder is composed of an MC 550 × 150 mm, @ 1.037 kN/m and a
channel MC 300 × 90 mm, @ 0 .358 kN/m placed on the top of the joist with
its flange down. Calculate the maximum flange and web stresses in the
girder section, given the following:
 DESIGN OF BEAMS 417

(i) Effective span of gantry girder = 7.3 m

(ii) Effective span of crane girder between centres of gravity of rails = 17 m
(iii) Distance between the pair of carriage wheels moving on each gantry rail =
3.6 m
(iv) Capacity of overhead crane = 150 kN
(v) Total weight of crane girder excluding the crab = 180 kN
(vi) Weight of crab = 50 kN
(vii) Weight of gantry girder inclusive of rail = 1.80 kN/m
(viii) Lateral load due to horizontal surge = 10 percent of lifted weight and weight
of crab
(ix) Impact for vertical wheel loads = 25 percent
(x) Closest position of crane hook from the centre of gantry rails = 1 mm
(b) What would have been the effect on the vertical moment and shear diagrams
if an allowance for longitudinal load were also taken.
6.18. A beam of simple framed building is made up of LB 550, @ 0.863 kN/m (bf
= 190 mm, tf = 15 mm, A = 10.97 mm2, tw = 9.9 mm) and carries a total load
of 400 kN uniformly distributed over the entire span. It is connected to the
flange of column at ends of section WB 600, @ 04.81 kN/m bf = 200 mm,
tf = 10 mm, A = 6133 mm2. tw = 7.4 mm). Using 20 mm φ rivets, design a
suitable connection for the joints. Draw a neat sketch of the joint.
6.19. A beam has an effective span of 10 m and carries a uniformly distributed
load of 100 kN. Design the cross-section of the beam if its depth is limited
to 45 mm.
6.20. The cross-section of a built-up beam consists of an LB 325 with a flange
plate of 1200 mm × 10 mm provided at the top as well as at the bottom. It
uses 18 mm diameter rivets.
It is simply supported over an effective span of 6 m. Calculate the maxi-
mum uniformly distributed load the beam can carry, if the beam is later-
ally supported throughout its length along the compression flange. Con-
sider bending moment only.
418 DESIGN OF STEEL STRUCTURES

Chapter
Design of Beam

7 Connections

7.1 INTRODUCTION
A steel structure is a framework of assembly of structural members. The beams
are connected to beams or columns (stanchions). The connections are necessary
to attach the members and to allow an orderly flow of the load to the foundation
by continuing the transfer of loads from the adjoining members. The connec-
tions are designed adequately. The joints are made safe economical and practi-
cal. More practical connections are not more economical as fabrication cost
greatly influences the economy of both connections and members. The connec-
tions must be designed with almost care, so that the failure of joint does not
occur. A joint may fail earlier than the failure of a member. When the failure of
member occurs, it is likely to allow time for safety to be undertaken. A failure of
a joint will be more catastropic. The factor of safety used for the design of con-
nections is kept more than that for the members to be connected. The transfer
of loads through the joint is made gradual and the joint is not overloaded. Vari-
ous factors for example, length of the joint, edge distance, distribution of fas-
teners, gauge distance, distribution of stress in the fasteners, length of fasten-
ers themselves, factor of safety, etc. influence the design of connections.
The frame connections are classified according to their rotational character-
istics as flexible connections, semi-rigid connections and rigid connections.

7.1.1 Flexible Connections

Flexible connections permit large angles of rotation and transmit no negligible
moment. Actually a small moment may be developed, but it is ignored in the
design. Any joint eccentricity less than about 63 mm is neglected. The
deformations and moments in beams with flexible connections are treated as
for ideally simply supported beams. The ends of beams are connected to resist
shear only. The frameworks with flexible connections are referred as simple
frame constructions.
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 419

7.1.2 Semi-rigid Connections

These connections allow small end rotation and transmit appreciable moment
(moment less than the full moment capacity of connected members). Design of
these connections needs an assumption regarding moment capacity. Arbitrarily,
the moment capacity is considered as twenty, thirty, or seventy five percent of
member capacity. The frameworks with semi-rigid connections are referred as
semi-rigid constructions.

7.1.3 Rigid Connections

Rigid connections do not allow any end rotation and retain a constant relative
angle between the connected members under any joint rotation and transmit
moment equal to full moment capacity of the members connected. The original
angles between the connected members remain unchanged. The connections
are fully restrained. The frameworks with rigid connections are referred as rigid
frame connections.
The simple beam end connections are of two types :
1. Frame connections and 2. Seated connections.

W eb cle at

B e am

B e am

(a ) B e am co nn ecte d to be am

B e am

S ta n ch io n

(b ) B e am co nn ecte d to sta n ch io n fram e d co nn ection s

Fig. 7.1 Framed connections

7.1.3.1 Framed Connections

When a beam is connected to a beam or a stanchion by mean of two angles
riveted to them as shown in Fig. 7.1, the connection is known as a framed
connection.
420 DESIGN OF STEEL STRUCTURES

In case of framed connections, when the flanges of the beams are to be kept
as the same level, the connecting beams are cut as shown in Fig. 7.2.

Fig. 7.2 Flanges of beams at same level (framed connections)

The framed connections are normally used to connect beams to beams. When
the beams are to be connected to webs of stanchions, or to flanges of stanchions,
depth of the web or width of the flange of stanchion may be insufficient to ac-
commodate the connecting angles. In such cases, the framed connections are
not possible, and therefore, the seated connections are employed.

7.1.3.2 Seated Connections

When a beam is connected to a beam or a stanchion by means of an angle at the
bottom of the beam, which is shop riveted to the beam or stanchion and an
angle at the top which is field riveted as shown in Fig. 7.3, the connections are
known as seated connections. The field riveted angle at the top, prevents the

B e am

S e at an gle
B e am
(a ) B e am co nn ecte d to b e am

Flan ge cle at

B e am

S e at an gle
S ta n ch io n
(b ) B e am co nn ecte d to sta nch io n se ate d co nn e ction s

Fig. 7.3

lateral deflection of the beam. The connection as shown in Fig. 7.3 (b) is known
as unstiffened seated connection. When the end reaction of the beam is large
then one or two stiffener angles are provided to support the outstanding leg of
the seat angle, as shown in Fig.7.4. The stiffening angles should be tightly fitted
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 421

under seating angle. Such connections are known as stiffened seated connections.
The stiffener angles are also used when the vertical leg of seat angle is inadequate
to accommodate the required number of rivets. Beams connected to flange of a
stanchion are shown in Fig. 7.3 (b) and in Fig. 7.4. Beams can also be connected
to webs of stanchions.

Seat angle
Stiffener
angles

Packing

Fig. 7.4 Stiffened seated connection

The seated connections are normally used to connect beams to stanchions.

When the beams are connected to beams, the depth of beam sections may be
insufficient to accommodate both the beams to be connected and the seating
angle. In such cases the seated connections are not possible, and therefore framed
connections are employed.

7.2 FRAMED CONNECTIONS

In Sec. 1.39 simple design method has been discussed. In the simple design
method, it is assumed that all connections of beams to columns are virtually

1
2
3

(a ) (b )

C o nn ection s no t stre sse d D e form e d sh ap e fra m e d con ne ctio ns

Fig. 7.5

flexible and are proportioned for the reaction shears applied at the appropriate
eccentricity. The connections do not offer resistance to rotation of the end of the
beam in the vertical plane. If the rotation of end of the beam does not take place
422 DESIGN OF STEEL STRUCTURES

fully, moment is induced at the end of the beam and bending stress is caused in
the connecting angle. Figure 7.5 shows a section through a framed connection.
Figure 7.5 (a) shows the connections when these are not stressed. Figure 7.5 (b)
shows the deformed shape of connecting angles under severe loading on the
tension side of connections. Connecting angles deform more at the top than at
the bottom of the connection. This deformation decreases gradually from top
towards bottom. It is assumed that the tension in the rivet is sufficient to keep
the centre line of legs, at points 1 and 2, parallel to their unstressed positions.
For analysis of stress in the connecting angles due to bending moment, consider
Fig. 7.6 (a) and Fig. 7.6 (b) before and after stressing, respectively. The frame
shown in Fig. 7.6 (a) is pulled by a horizontal force P. It results in a horizontal
reaction P acting in the opposite direction at point 1 as shown in Fig. 7.6 (b). Two
equal and opposite forces P acting at a distance g1 form a couple Pg1.

P 1

θ
3 2 P

3 2
F F
(a ) C o nn ection s no t stre ssed (b ) C o nn ection s stresse d

Fig. 7.6

This couple acts in anticlockwise direction. The original direction of frame at 1

and 2 remains same. This joint 3 is displaced towards right by an amount δ, in
the direction of force P, and it rotates by an angle θ in anticlockwise direction.
Let M1, M2 and M3 be the moments at 1, 2 and 3 respectively
1
M2 = M3 ...(i)
2

⎛ M 3 ⋅ g2 ⎞
For moment 3–2 θ = ⎜ ⎟ ...(ii)
⎝ 4EI ⎠
where, I = Moment of inertia of member 3–2,
g2 = Length of member 3–2
We have (M1 + M3) = P . g1
Ml = (P . g1 – M3) ...(iii)
For member 3.1, by area moment method
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 423

⎛M g M ⋅g ⎞
θ = ⎜ 1 1 − 3 1⎟
⎝ 2EI EI ⎠
Substitute the value of θ from (iii) and M1 from (iv)
⎛ M 3 ⋅ g2 ⎞ g2 ⎛M ⋅g ⎞
⎜ ⎟ = ⋅ ( Pg1 – M 3 ) – ⎜ 3 1 ⎟
⎝ 4EI ⎠ 2 EI ⎝ 2EI ⎠

⎛ 2 g1 ⎞
M3 = Pg1. ⎜ ⎟ ...(7.1)
⎝ 4 g1 + g2 ⎠

⎛ 2 g1 ⎞
Therefore, M1 = Pg1 – Pg1 ⎜ 4 g + g ⎟
⎝ 1 2 ⎠

⎛ 2 g + g2 ⎞
M1 = Pg1. ⎜ 1 ⎟ ...(7.2)
⎝ 4 g1 + g2 ⎠
where, g1 = Length of member 3–1
I = Moment of inertia of member 3–1, same as moment of inertia of
3–2
Thus, the moment induced at the points 1 and 3 in Fig. 7.5 have the same
values as given by Eq. 7.2 and Eq. 7.1, respectively. In case g1 and g2 are equal,
then, the values of moments induced are given by
M1 = (0.6 P.g1), M2 = (0.2P.g1), and M3 = (0.4P.g1) ...(7.3)
where, g1 = g2
If g is gauge distance for leg of angle, then
⎛ 1 ⎞
∴ g1 = ⎜ g – ⋅ t ⎟
⎝ 2 ⎠
The difference between the values gx and g2 is not significant. The value of
bending moment M1 is maximum in the angle. The connecting angle should be
able to resist M1. The value of pull P should not be greater than maximum
allowable tensile force in the rivet.
The tensile force in the rivet connecting the leg 3–2, in Fig. 7.6 (b) is F. This
force is given by
⎛ 3 M3 ⎞ 1
∴ F. g2 = (M2 + M3), F = ⎜2⋅ g ⎟ (∴ M2 = M3 )
⎝ 2 ⎠ 2
Substituting the values of M2, and taking, g = g2
⎛1 M g 2 1 M g 1 ⎞
F = 0.6 P, δ = ⎜ ⋅ 1⋅ 1 ⋅ g1 ⋅ ⋅ 3 1 ⋅ g ⎟ ...(7.4)
⎝ 2 EI 3 2 EI 3 ⎠
Substituting the values of M1 and M3 from Eq. 7.3
⎛ P ⋅ g3 ⎞
δ = 0.133 ⎜ ⎟
⎝ EI ⎠
424 DESIGN OF STEEL STRUCTURES

7.3 UNSTIFFENED SEATED CONNECTIONS

In case of the unstiffened seated connections, seat angle supports a beam. The
action of seat angle is approximately similar to that of a cantilever beam. In
case, thicker seat angles are used, then, they have a tendency to concentrate
the reaction at the toe of outstanding leg. While if thinner seat angle is used,
they have a tendency to distribute the reaction over more length. The stiffness
of flanges of beam and thickness of web also affect the distribution of reaction.
If the connection of beam is not done with the seat angle the bending of seat
angle takes place about a critical section. The critical section is at the top of row
of rivets. When the beam is loaded, the bottom flange of the beam elongates. It
pushes the back of seating angle against the column. As a result of this, the
vertical leg of seating angle is relieved from bending moment. The connection of
seat angle and beam flange restrains the bending of horizontal leg as a simple
cantilever. Therefore, it is difficult to locate the critical section about which
bending moment is to be found. Location of critical section is assumed at the
edge of the fillet in the horizontal leg. The end reaction is assumed to be uni-
formly distributed on the outstanding leg of seat angle, which is sufficient to
meet the web crippling requirement of beam. The length of beam resisting the
web crippling has been found in Sec. 6.9.

1
C le a t a ng le B e am
B e am 3 2

C o lu m n C o lu m n

S e at an gle

(a ) (b )
Fig. 7.7

Figure 7.7 (a) shows an unstiffened seated connection before loading. Figure
7.7 (b) shows an unstiffened seated connection after loading. When the rotation
of end of beam takes place, then the seat angle is pressed against the column,
and cleat angle is deformed as shown in Fig. 7.7 (b). If the rotation of end of
beam does not take place fully, moment is induced at the bending stress is caused
in the connecting angle.
The analysis of the stress in the connecting angle due to bending is same as
that for connecting in framed connection. When the section of connecting angle
is considered, it is horizontal plane. The section for the connecting angle (cleat
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 425

angle) is in the vertical plane. The values of moment M1, M2 and M3 found by
Eq. 7.3 hold good at same respective point as in Fig. 7.7 (b).

7.4 STIFFENED SEATED CONNECTIONS

When the end reaction of the beam is large, then one or two stiffener angles are
used to support the outstanding leg of the seat angle. The length of outstanding
leg of stiffening angle is extended upto the toe of seating angle as far as possible.
The beam reaction acts on the outstanding leg of stiffening angle. In case, a
single stiffener angle is used, the reaction is eccentric with respect to the rivets
connecting the stiffener to the column as shown in Fig. 7.8. The end reaction R,
through the stiffener has eccentricity e xx along the xx-axis and e yy along the yy-
axis. In addition to direct shear, rivets are subjected to twisting moment, R . eyy
and bending moment, R. exx. The twisting moment acts in a plane parallel to
YZ- plane, (the plane in which rivets are connected). The rivets are subjected to
additional shear due to twisting. The bending moment acts in a plane parallel
to XZ plane. The rivets above the neutral axis (the line of rotation) are subjected
to tension, and rivets below the neutral axis are subjected to compression against
column flange. The rivets connecting stiffener angle to column flange are
designed to resist direct shear, twisting moment and bending moment.

R
x
e yy

z
xx

y
e

y
x

Fig. 7.8 Single stiffener angle

In case two stiffener angles are used, their outstanding legs are connected
together, by means of tacking rivets. Two stiffener angles act as one unit. The
426 DESIGN OF STEEL STRUCTURES

reaction R, transmitted by stiffening angles has eccentricity along xx-axis only

as shown in Fig. 7.9. The rivets connecting stiffener angles to the flange of

R
x

y z
e xx
x
y

Fig. 7.9 Two stiffener angles

column are subjected to bending moment R . exx. The bending moment acts in
XZ-plane. The design of stiffened seated connections are discussed in Sec. 7.7.
If the two stiffener angles are connected by tacking rivets they act as a single
stiffening angle as discussed above and each carries half reaction.

7.5 DESIGN OF FRAMED CONNECTIONS

The framed connections are designed to transmit the end reaction of connected
beam. The following are the usual steps in the design of framed connections.
Step 1. Compute end reaction for the connected beam for the loading on the
beam.
Step 2. The cleat angles are connected to the web of the connected beam in
the workshop by means of power driven shop rivets. Assume the diameter of
the rivets and determine the rivet value for them. Compute number of rivets
required to transmit the end reaction of the connected beam as under :
⎛ End reaction of connected beam ⎞
Number of rivets = ⎜ ⎟⎠
⎝ Rivet value
The size of cleat angles depends upon whether the rivets can be accommo-
date in one line or in two lines.
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 427

Step 3. The power driven field rivets are used to connect the cleat angles to
the supporting beam. Assume the diameter of the rivets and determine rivet
value of them. Compute the number of rivets required.
Step 4. The cleat angles are checked for shear stress. The maximum shear-
stress should not exceed 0.45 fy N/mm2.
Maximum shear stress is found as below.

t v.max = 3 × ⎛ F ⎞
⎜ ⎟
2 ⎝ 2d × t ⎠
where, F = End reaction of the connected beam
d = Depth of cleat angles
t = Thickness of cleat angles.
Example 7.1 An LB 350 @ 0.495 kN/m transmits an end reaction of 290 kN
to the web of an MB 500, @ 0.869 kN/m. Design a framed connection and give a
neat sketch. Use 22 mm diameter rivets.
Solution
Design :
Step 1 : End reaction
End reaction of connecting beam = 290 kN
From steel section tables,
Thickness of web of connecting beam = 7.4 mm
Step 2: Rivet value
Strength of power driven rivets in double shear
⎛ π 2 100 ⎞
⎜ 2 × × (23.5) × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of power driven rivets in bearing
⎛ 300 ⎞
⎜ 23.5 × 7.4 × ⎟ = 52.17 kN
⎝ 1000 ⎠
Rivet value = 52.17 kN
Step 3 : Design of connections
⎛ 290 ⎞
⎜ ⎟ = 5.558. (Say 8 rivets)
⎝ 52.17 ⎠
Accommodate the rivets in two rows. Pitch of rivets is 65 mm
Edge distance = 40 mm
Use 2 cleat angles ISA 150 mm × 115 mm × 10 mm (2 ISA 150 115 @ 0.200
kN/m).
Thickness of web of supporting beam, t w is 10.2 mm
Step 4 : Rivet value
Strength of power driven rivets in single shear
⎛ π (23.5)2 × 100 ⎞
⎜⎝ × ⎟⎠ = 43.35 kN
4 1000
428 DESIGN OF STEEL STRUCTURES

Strength of power driven rivets in bearing

⎛ 300 ⎞
⎜ 23.5 × 10.2 × ⎟ = 71.91 kN
⎝ 1000 ⎠
Rivet value = 43.35 kN
Step 5 : Design of connections
⎛ 290 ⎞
Number of rivets = ⎜ ⎟ = 6.689. Say 8 rivets.
⎝ 43.35 ⎠
Provide 4 rivets for each cleat angle.
Pitch of rivets = 65 mm
Edge rivets = 40 mm
Step 6 : Check for maximum shear stress
Maximum shear stress
⎛ 3 290 × 1000 ⎞
⎜ × ⎟ = 79.09 N/mm2
⎝ 2 2 × 10 × 275 ⎠
< (0.5 × 250 = 112.5 N/mm2)
Hence design is satisfactory.
Design of framed connection is shown in Fig. 7.10.

IS L B 35 0, @ 0.49 5 kN /m

65 mm

65 mm

65 mm

IS M B 50 0, @ 0.86 9 kN /m

Fig. 7.10 Beam connected to beam (Framed connection)

7.6 DESIGN OF UNSTIFFENED SEATED CONNECTIONS

The unstiffened seated connections are used to trasmit end reaction of beam
upto 200 kN. The following are the usual steps for the design of unstiffened
seated connections:
Step 1. The seat angle is designed to transmit end reaction, F of connected
beam. The end reaction, ‘F’ is considered as uniformly distributed on the out-
standing leg of seat angle over a length b as shown in Fig. 7.11. Compute the
bearing length b.
The length b is such that web cripping of beam does not occur. The length of
beam ‘B’ resisting the web cripping at the end is found as below :
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 429

⎛ F ⎞
B = ⎜ ⎟ , and also B = b + 3h2( )
⎝ σ p.tw ⎠
where, ap = Allowable bearing stress in web (0.75fy N/mm2)
t w = Thickness of web of beam
h2 = Depth of root of fillet from extreme fibre of flange.

h2

C le a ran ce 1 0 m m
3 0°
h2

Fig. 7.11 Beam connected to Stanchion unstiffened seated connections

The bearing length

⎛ F ⎞ 1 ⎛1 F ⎞
b = ⎜ − 3h2 ⎟ |< B,i.e., ⎜ ⎟
f ⋅
⎝ b w t ⎠ 2 ⎝ 2 σ p ⋅ tw ⎠
Step 2. Select a trial section for seat angle.
Step 3. Provide a clearance of 10 mm between the beam and stanchion. Com-
pute the distance of end reaction from the critical section of the outstanding leg
of the seating angle. Calculate bending moment, ‘M’ at the critical section, XX.
The critical section is the section at the edges of the fillet as shown in Fig. 7.12.
C ritical section
F

t
r1
Ed ge o f fillet
x

r 1 = R adiu s at th e roo t

Fig. 7.12 Critical section for angle section

Step 4. Assume the length of the seating angle equal to the width of the flange
of the beam. Compute moment of resistance of seating angle, which should be greater
than bending moment M. Then trial section of seating angle is satisfactory.
430 DESIGN OF STEEL STRUCTURES

Step 5. The seating angle is connected to the stanchion in the workshop by

power driven rivets. Assume diameter of rivets and determine the rivet value.
Compute the number of rivets required.
Step 6. Connect a cleat angle at the top by two rivets in either leg. The
diameter of the rivets should be equal to the diameter of rivets used to connect
the seating angle.

7.7 DESIGN OF STIFFENED SEATED CONNECTIONS

The stiffened seated connections are used, when end reaction of beam exceeds
200 kN. The following are the usual steps in the design of stiffened connections :
Step 1. Compute the bearing length b as for unstiffened seated connections.
In stiffened seated connections, the bearing length is measured from the end of
stiffening leg.
Step 2. Compute the bearing area required to be provided by stiffener angles.
Select the stiffener angles to furnish the bearing area. The outstand of the
stiffener angles to furnish the bearing area. The outstand of the stiffener angles
should not exceed 16 times its thickness to avoid local buckling of the angles.
Step 3. Provide a seating angle depending upon the length of outstanding
leg of stiffener angles.
Step 4. The rivets connecting the stiffener angles to the stanchion are
subjected to direct shear and tension. It is better to provide two stiffener angles
and connect their outstanding legs by stitching rivets. The distance of end
reaction is found from the face of the stanchion and bending moment, ‘M’ is
computed. The number of rivets is computed from Eq. 2.12 as below :
1/ 2
⎛ 6M ⎞
n = ⎜ ⎟
⎝ p⋅R ⎠

⎡ τvf .cal σtf ⋅cal ⎤

The rivets are checked for shear and tension. The quantity ⎢ + ⎥
⎣ τvf σtf ⎦
should be less than or equal to 1.4
where, τvf.cal = Actual shear stress in the rivet
τvf = Allowable shear stress in the rivet
σtf.cal- = Actual tensile stress in the rivet
σtf = Allowable tensile stress in the rivet.
Example 7.2 An MB 500, @ 0.869 kN/m transmits an end reaaction of 130
kN to the flange of stanchion HB 250, @ 0.510 kN/m. Design a unstiffened seated
connections. Use 22 mm diameter rivets.
Solution
Design :
Step 1 : End reaction
End reaction of beam, F = 130 kN
Required bearing length of seat angle
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 431

⎛ F ⎞
b = ⎜ − 3h2 ⎟
σ ×
⎝ p w f ⎠
From steel section tables, for MB 500, @ 0.868 kN/m
Width of flange = 180 mm
Thickness of web t w = 10.2 m
h2 = 37.95 mm
Permissible bearing stress = 0.75 fy
σp = (0.75 × 250) = 187.5 N/mm2
⎛ 130 × 1000 ⎞
b = ⎜ − 3 × 37.95 ⎟ = 2.24 mm
⎝ 187.5 × 10.2 ⎠

⎛ 1 F ⎞ ⎛ 1 130 × 1000 ⎞
But b |
< |⎜ ×
< ⎟ mm
⎝⎜ 2 σ p × tw ⎠⎟ ⎝ 2 187.5 × 10.2 ⎠
< 33.987 mm
|
The bearing length of seating angle
= 33.987 mm
Step 2 : Selection of trial section
Try seating angle ISA
150 mm × 115 mm × 12 mm (ISA 150115 @, 0.238 kN/m)
Radius at roof = 11 mm
10 m m 3 3.9 87 m m
15 0 mm

11 m m
12 m m
Fig. 7.13 Bearing length of seat angle section

The distance of end raction from critical sections

⎡⎛ 1 ⎞⎤
⎢⎜⎝10 + 2 × 33.987) − (12 + 11 ⎟⎠ ⎥ = 3.994 mm
⎣ ⎦
Step 3 : Bending moment at the critical section
(3.994 × 130) = 519.155 mm-kN
432 DESIGN OF STEEL STRUCTURES

Length of seating angle

Width of flange of beam = 180 mm
Moment of resistance of seating angle
(σbc = 0.66 × 250 = 165 N/mm2)

⎛ 165 × 180 123 ⎞

⎜ × ⎟ = 712.8 mm-kN
⎝ 1000 6 ⎠
Hence seating angle may be provided.
Step 4. Rivet value
Strength of 22 mm power driven rivets in single shear

⎛ π (23.5)2 × 100 ⎞
⎜ ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of power driven rivet in bearing
⎛ 23.5 × 10.2 × 300 ⎞
⎜ ⎟ = 71.91 kN
⎝ 1000 ⎠
Rivet value = 43.35 kN
Step 5 : Design of connections
⎛ 130 ⎞
Number of rivets required is ⎜ = 2.999 ⎟
⎝ 43.35 ⎠
Provide 4 rivets in two rows.
Connect top cleat angle ISA 100 mm × 75 mm × 10 mm (ISA 10075 @ 0.130
kN/m with two rivets 22 mm diameter in either leg of the angle).
Example 7.3 An WB 550 beam transmits an end reaction of 280 kN to the
web of a column HB 250, @ 0.547 kN/m. Design and sketch a suitable connection
using 22 mm diameter rivets. Note that the flange of the beam will not fit inside
the flange of the beam will not fit inside the flange of the column as such.
Solution
Design :
Step 1: End reaction
End reaction of beam = 280 kN
The stiffened seated connection would be used to connect the beam with the
stanchion.
Step 2 : Properties of given section
From the steel section tables, for WB 550, @ 1.125 kN/m
Width of flange = 250 mm
tw = 10.4 mm
h2 = 38.30 mm
For HB 250, @ 0.547 kN/m
Width of flange, bt = 250 mm
tw = 8.8 mm
tf = 9.7 mm
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 433

The flanges of WB 550, @ 1.125 kN/m are cut and fitted as shown in Fig. 7.14.

IS W B 5 50 @ 1 .1 25 kN /m

22 0 m m

25 0 m m
Fig. 7.14 Beam connected to Stanchion (Stiffened seated connection)

Step 3 : Permissible bearing stress

(0.75 × 250) = 187.5 N/mm2
⎛ F ⎞
The bearing length b = h
⎜⎝ σ p × tw 3 2 ⎟⎠

⎛ 280 × 1000 ⎞
⎜⎝ − 3 × 38.3⎟ = 75.885 mm
187.5 × 10.5 ⎠

⎛ 1 F ⎞ ⎛ 1 280 × 1000 ⎞
But b |
< ×
⎜⎝ 2 σ p × tw ⎟⎠ ⎜⎝ 2 187.5 × 10.5 ⎟⎠

= 71.11 mm < 75 mm
Provide a clearance of 10 mm
Length of outstanding leg of stiffener angle
= (75.885 + 10)
= 85.885 mm
Provide ISA 100 mm × 75 mm × 12 mm (ISA 10075, @ 0.154 kN/m) for seating
angle. Outstanding leg of seating angle is 100 mm.
Step 4 : Bearing area required for stiffener angles
⎛ 280 × 1000 ⎞
⎜ ⎟ = 1493.33 mm2
⎝ 187.5 ⎠
Outstanding legs of stiffener angles furnish bearing area.
Provide 2 ISA 90 mm × 90 mm × 10 mm (2 ISA 9090, @ 0.1703 kN/m)
Bearing area furnished
2 (90 –10) × 10 =1600 mm2 > Bearing area required.
The distance of end reaction of the beam from end of stiffener angle
⎛1 ⎞
⎜ × 75.885 ⎟ = 37.94 mm
⎝2 ⎠
The distance of end reaction from face of web of stanchion
(90 + 12 – 37.3) = 64.7 mm
434 DESIGN OF STEEL STRUCTURES

B e am

b
3 0°
S titch rivets

C o lu m n

Fig. 7.15 Beam connected to Stanchion (Stiffened seated connection)

Step 5 : Bending moment about the face of web

⎛ 64.7 × 280 ⎞
⎜ ⎟ = 18.116 m-kN
⎝ 1000 ⎠
Step 6 : Rivet value
Strength of 22 mm power driven rivets in single shear
⎛ π (23.5)2 × 100 ⎞
⎜⎝ × ⎟⎠ = 43.35 kN
4 1000
Strength of rivets in bearing
⎛ (23.5) × 8.8 × 300 ⎞
⎜ ⎟ = 62.04 kN
⎝ 1000 ⎠
Strength of rivets in tension
⎛ π (23.5)2 × 18 ⎞
⎜⎝ × ⎟⎠ = 34.68 kN
4 1000
Rivet value, R = 34.68 kN
Step 7: Design of connections
The rivets are provided in two rows at pitch of 70 mm.
Number of rivets required in one row
1/2
⎛ 6 × 18.11 × 1000 ⎞
n = ⎜ ⎟ = 4.73
⎝ 2 × 70.0 × 34.68 ⎠
Provide 5 rivets in each row. Force due to direct shear
280
= 28 kN
10
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 435

Fig. 7.16

Step 8 : Shear stress

⎛ 28 × 1000 ⎞
Actual shear stress τav.cal = ⎜ ⎟ = 64.59 N/mm
2
π ( )2
⎜ × 23.5 ⎟
⎝4 ⎠
Allowable shear stress in power driven rivet
τvf = 100 N/mm2
Step 9 : Moment resisted by rivets in tension
⎛ M ⎞
M´ = ⎜
2h ∑ y ⎟
⎜⎜ 1 + ⎟
⎝ 2t ∑ y2 ⎟⎠
∑y = 2[60 + 130 + 200 + 270] = 1320 mm
∑y2 = 2[(62 + 132 + 202 + 272) × 102]
= 2668 × 100 mm2
⎛ 18.11 ⎞
M´ = ⎜ = 15.785 m-kN
2 × 315 × 320 ⎟
⎜1 + ⎟
⎝ 21 × 2668 × 100 ⎠
Step 10 : Tensile force in the top rivets
⎛ 15.785 × 1000 × 270 ⎞
Ft = ⎜ ⎟ = 15.974 kN
⎝ 2668 × 100 ⎠
Tensile stress in the rivet,
⎛ 15.974 × 1000 ⎞
σtf.cal = ⎜ ⎟ = 36.847 N/mm
2
π 2
⎜ × (23.5) ⎟
⎝ 4 ⎠
Allowable tensile stress in power driven rivets
= 100 N/mm2
436 DESIGN OF STEEL STRUCTURES

Step 11 : Check for interaction expression

⎡ τvf .cal σtf .cal ⎤ ⎛ 64.59 36.847 ⎞
⎢ + ⎥ = ⎜ + ⎟ = 1.0144 |
> 1.4
τ
⎣ vf σ tf ⎦ ⎝ 100 100 ⎠
Hence, the design is satisfactory.
Provide top cleat angle ISA 100 mm × 75 mm × 10 mm (ISA 10075, @ 0.130
kN/m) and connect it by 22 mm rivets in either leg.
Example 7.4 A LB 400, @ 0.569 kN/m transmits an end reaction of 250 kN to
the flange of an HB 300, @ 0.588 kN/m column. Design stiffened seated
connections. Provide (i) a single stiffener angle, (ii) two stiffener angles.
Solution
Design :
Step 1: End reaction
The end reaction for the beam is 250 kN
Step 2 : Properties of given section
From ISI Handbook No. 1, LB 400, @ 0.569 kN/m
Width of flange –165 mm
Thickness of web, t w = 8.0 mm, h2 = 31.90 mm
For HB 300, @ 0.588 kN/m
Width of flange, bf = 250 mm
Thickness of web, t w = 7.6 mm
Thickness of flange, t f = 10.6 mm
Step 3: Permissible bearing stress
σp = (0.75 × 250) = 187.5 N/mm2
The bending length, b
⎛ F ⎞ ⎛ 250 × 1000 ⎞
⎜ σ ⋅ t − 3h2 ⎟ = ⎜ − 3 × 31.90 ⎟|
⎝ p w ⎠ ⎝ 187.5 × 8.0 ⎠
= 111.414 mm
⎛1 F ⎞ ⎛ 1 250 × 1000 ⎞
But, b |
< ⎜ ⎟|
<⎜ < 83.33 mm
⎟ |
⎝ 2 σ p ⋅ tv ⎠ ⎝ 2 187.5 × 8 ⎠
Provide a clearance of 10 mm
Length of outstanding leg of stiffener angle
= (111.414 + 10)
= 121.414 mm
Provide ISA 130 mm × 130 mm × 10 mm ISA 130130, @ 0197 kN/m for seat-
ing angle.
Step 4 : Bearing area required for stiffener angle/angles
⎛ 250 × 1000 ⎞
⎜ ⎟ = 1333.33 mm2
⎝ 187.5 ⎠
(i) Provide a single stiffener angle ISA 110 mm × 110 mm × 15 mm. (ISA
110 110, @ 0.242 kN/m.
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 437

The outstanding leg provides bearing area.

Bearing area provided
(110 – 15) × 15 = 1425 mm2
> Bearing area required. Hence, safe.
Length of outstanding leg
= (110 –15) = 95 mm
Thickness of angle, t is 15 mm
⎛ 95 ⎞
Outstanding ratio ⎜ ⎟ = 6.33 < 16. Hence, safe.
⎝ 15 ⎠
Width of seating angle is 130 mm
Point of application of reaction from face of column
⎛ 130 ⎞
⎜ ⎟ = 65.0 mm
⎝ 2 ⎠
∴ Eccentricity, exx = 65.0 mm.
From ISI Handbook No. 1, the gauge distance for ISA 110 mm × 110 mm × 15
mm for 110 mm leg is 65.0 mm.
Eccentricity, eyy = 65.0 mm
Step 5 : Bending moment and twisting moment
⎛ 250 × 65.0 ⎞
Mxx = R.exx ⎜ ⎟
⎝ 1000 ⎠
= 16.250 kN-mm
The twisting moment
⎛ 250 × 65.0 ⎞
Myy = R.exx = ⎜ ⎟
⎝ 1000 ⎠
= 16.250 m-kN
Step 6 : Rivet values
Use 22 mm power driven rivets. The strength of rivets in single shear
⎛ π (23.5)2 × 100 ⎞
⎜ ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of power driven rivets in bearing
⎛ 23.5 × 10.6 × 300 ⎞
⎜ ⎟ = 74.73 kN
⎝ 1000 ⎠
Strength of power driven rivet in tension
⎛ π (23.5)2 × 100 ⎞
⎜⎝ × ⎟⎠ = 43.35 kN
4 1000
The rivet value for rivets subjected to shear and twisting,
R = 43.35 kN
The rivet value for rivets subjected to shear and bending,
R = 43.35 kN
438 DESIGN OF STEEL STRUCTURES

The rivets are provided in one row at a pitch of 80 mm

1/ 2
⎛ 6M ⎞
Number of rivets n = ⎜ ⎟
⎝ p⋅R ⎠
1/2
⎛ 6 × 16.25 × 1000 ⎞
⎜ ⎟ = 5.30. Provide 8 rivets.
⎝ 80 × 43.35 ⎠
Step 7 : Forces in rivets
250
The direct shear force in a rivet = 31.25 kN
8
∑y2 = 2 × [42 + 122 + 202 + 282] × 100
= 2688 × 100 mm2
The shear due to twisting
⎛ My ⎞ ⎛ 16.25 × 1000 × 280 ⎞
⎜ 2⎟ = ⎜ ⎟ = 16.927 kN
⎝ ∑y ⎠ ⎝ 268800 ⎠
The resultant force in a rivet
(31.252 + 16.9272)1/2 = 35.54 kN
Hence, rivets are safe in direct shear and twisting
Shear stress in rivets
⎛ 31.25 × 1000 ⎞
= 72.08 N/mm2
⎜ ⎛ π⎞ 2
⎟
⎜ ⎜⎝ ⎟⎠ × (23.5) ⎟
⎝ 4 ⎠
Step 8 : Moment resisted by rivets in tension
⎛ ⎛ 2h Σy ⎞ ⎞
M´ = ⎜ M / ⎜1 +
⎝ ⎝ 2l Σy2 ⎟⎠ ⎟⎠
Height of line of rotation above the bottom of stiffener angle
⎛1 ⎞
⎜ × 600 ⎟ = 85.5 mm
⎝2 ⎠
Distance to extreme rivet is 514.5 mm
∑y = [3.45 + 11.45 + 19.45 + 27.45 + 35.45 + 43.45 + 51.45] × 10 = 1921.5 mm
∑y2 = 2 × [3.452 + 11.452 + 19.452 + 27.452 + 35.452 + 43.452 + 51.452] × 100
= 7053 × 100 mm2

⎛ 16.25 ⎞
M´ = = 14.06 kN-m
⎜ 2 × 600 × 1921.5 ⎟
⎜⎝ 1 + ⎟⎠
21 × 7053 × 100
The tensile force in the top rivet,
⎛ 14.06 × 1000 × 514.5 ⎞
Ft = ⎜ ⎟ = 10.275 kN
⎝ 7053 × 100 ⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 439

The tensile stress in the rivet,

⎛ 10.257 × 1000 ⎞
σtf.cal = ⎜ ⎟ = 23.66 N/mm
2
π 2
⎜ × (23.5) ⎟
⎝ 4 ⎠
Step 9 : Check for interaction expression
As per IS: 800–1984
⎡ τvf .cal σtf .cal ⎤ ⎡ 72.08 23.66 ⎤
⎢ + ⎥ = ⎢ + | 1.4
= 1.9574 >
⎣ τvf σtf ⎦ ⎣ 100 100 ⎥⎦
Hence, the rivets are safe for shear tension.
Complete design of seated connection, using a single stiffener angle is shown
in Fig. 7.17.

C le a t a ng le
IS A 9 0 m m x 9 0 m m x 1 0 m m

IS LB 40 0 @ 0.5 69 kN /m
C o lu m n B e am

S e at an g le
IS A 1 30 m m x 130 m m
10 m m
S tiffen e r an g le
IS A 11 0 m m x 110 m m x 15 m m

8 rive t, 2 2 m m φ
a t 8 0 m m p itch
IS H B 3 00 , @ 0 .568 kN /m

Fig. 7.17 Beam connected to stanchion (Stiffened seated connections) single angle section

(ii) Provide, two stiffener angle ISA 90 × mm × 90 mm × 10 mm (2 ISA

9090,@ 0.134 kN/m)
Bearing area provided = 2 × (90 – 10) × 10 = 1600 mm2
>1322 mm2. (Bearing area required). Hence, safe.
Length of outstanding leg
(90 – 10) = 80 mm
Thickness of angle t is 10 mm
80
Outstanding ratio = 8 < 16. Hence, safe.
10
440 DESIGN OF STEEL STRUCTURES

Point of application of reaction from face of column

⎛ 130 ⎞
⎜ ⎟ = 65 mm
⎝ 2 ⎠
Eccentricity exx = 65.0 mm
The bending moment
⎛ 250 × 65 ⎞
⎜ ⎟ = 16.25 kN-m
⎝ 1000 ⎠
Step 10 : Design of connection
Number of rivets in each row
1/2
⎛ 6 × 16.25 × 1000 ⎞
n = ⎜ ⎟ = 3.75
⎝ 2 × 80 × 43.35 ⎠
Provide 5 rivets in each row
Step 11 : Shear stress in rivet
⎛ 250 ⎞
Shear force in each rivet ⎜ ⎟ = 25 kN
⎝ 10 ⎠

⎛ 25 × 1000 ⎞
Shear stress in rivet = ⎜ ⎟ = 57.668 N/mm
2
π
⎜ × (23.5)2 ⎟
⎝4 ⎠
Distance of line of rotation from bottom of stiffener angles
1
× 360 = 51.4 mm
2
Distance to top-most rivet = 308.6 mm
∑y = 2 [6.86 + 14.86 + 22.86 + 30.86] × 10 = 1508.8 mm
∑y2 = 2 [6.862 + 14.862 + 22.862 + 30.862] ×100 = 3502 × 100 mm
Step 12 : Moment resisted by the rivets in tension

⎡ M ⎤ ⎡ 16.25 ⎤
M´ = ⎢ ⎥= ⎢ = 14.159 kN-m
2h ∑ y 2 × 360 × 1508.8 ⎥
⎢1 + ⋅ ⎥ ⎢1 + ⎥
⎢⎣ 2l ∑ y2 ⎥⎦ ⎣ 21 × 3502 × 100 ⎦
Tensile force in top-most rivet
⎛ 14.159 × 1000 × 308.6 ⎞
Ft = ⎜ ⎟ = 12.476 kN
⎝ 350200 ⎠
Tensile stress in the rivet

⎛ 12.476 × 1000 ⎞
Ft = ⎜ = 28.779 N/mm2
⎝ ( π ,4 ) × (23.50)2 ⎟⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 441

Step 13 : Check for interaction expression

As per IS: 800–1984

⎡ τvf .cal σtf .cal ⎤ ⎡ 57.66 28.779 ⎤

⎢ + ⎥ = ⎢ + = 0.8644 |
> 1.4
τ
⎣ vf σ tf ⎦ ⎣ 100 100 ⎥⎦
Hence the rivets are safe.
The complete design of seated connections using two stiffener angles are shown
in Fig. 7.18.

C le a t a ng le
IS A 9 0 m m x 9 0 m m x 10 m m

IS LB 40 0 @ 0.5 6 9 kN /m
B e am

S tiffen e r a ng le
2 S A 9 0 m m x 90 m m x 10 m m
1 0 rivets, 22 m m φ
in tw o row s a t 8 0 m m
p itch
IS H B 3 00 @ 0 .5 68 kN /m

Fig. 7.18 Beam connected to stanchion (Stiffened seated connections) two angle sections

7.8 SMALL MOMENT RESISTANT CONNECTIONS

When the values of moments are small, then the clip-angle connections as shown
in Fig. 7.19 are used.
The beam is connected to column by four angles. The web angles are used to
connect the web of beam. One clip angle at the top and one clip angle at the
bottom are used to connect the flanges of beam. These angles are called flange
clip angles. It is assumed that web connection resists no moment and flange
connections resist no shear. When the connections are subjected to clockwise
moment then the rivets at 1–1 are subjected to tension and rivets at 2–2 are
subjected to shear. When the connections are subjected to anticlockwise moment
then, the rivets at 1´–1´ are subjected to tension and rivets at 2´–2´ are subjected
to shear. Two rivets are used in one gauge line to connect the vertical leg of the
clip angle. If instead of two rivets, four rivets are used in one gauge line, then
the distribution of tension is not uniform. If instead of one gauge line rivets are
provided in two guage lines in the vertical leg of the clip-angle, then the
distribution of tension is not uniform. It is seen that number of rivets limits
442 DESIGN OF STEEL STRUCTURES

moment resisting capacity of the connections. Therefore, these types of

connections are used for small moments.

C lip an gle
1 1 2

2
d
2'

2'
1' 1'
C lip an gle

(a ) (b )

(c)

Fig. 7.19 Clip-Angle connections

When the tensile stress in the rivet is above the yield stress, then, the out-
standing leg of angle bends in simple flexure. The outstanding leg of angles acts
as a simple cantilever as shown in Fig. 7.20.
If P is pull on the rivets, g is gauge distance for connected leg of angle,
and t is the thickness of angle, the bending moment to be resisted by angle is
M = P. (g – t) ...(7.5)

P
P
1 1
1 (9 – 1)
(9 –1 ) 2
P o in t o f
con tra fle xure
1 (9 – 1)
2
2
f
P P
2

Fig. 7.20 Fig. 7.21

When the initial tension in the rivets is sufficient to keep the centre lines of the
angle legs at 1–1 and 2–2 parallel to their original position as shown in Fig. 7.21
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 443

the angle bends in double flexure. The rotation of angle at 1–1 and 2–2 is
prevented. There exists a point of contraflexure halfway between the rivet and
face of angle.
The bending moment is given by
⎛ g −t ⎞
M = P. ⎜ ⎟
⎝ 2 ⎠
M = 0.5 P. (g – t) ...(7.6)
The value of bending moment obtained by Eq. 7.6 is half that obtained of Eq.
7.5 for single cantilever flexure. For safety, the point of contraflexure is assumed
at a distance 0.6 (g – t). This gives a little conservative design of clip angles. The
bending moment is given by
M = 0.6 P . (g – t) ...(7.7)

7.7.1 Choice of Design Method

The connections shown in Fig. 7.19 are designed for small moment. There is no
clear limitation to the value of moment. The values of moments to be resisted by
clip angles, found under different assumptions are given below:
From Eq. 7.3 M = (0.6 P. g1) = 0.6 P . (g – ½ t)
From Eq. 75 for single cantilever flexure
M = P. (g – t)
From Eq. 7.7 for double cantilever flexure
M = 0.6 P . (g – t)
The design of clip angle based on simple cantilever flexure is conservative
and neglects the initial tension of rivets. A thick section of clip angle is obtained
by this method, which increases the weight and cost of the structure. The val-
ues of moments obtained by Eq. 7.3 and Eq. 7.7 both account for the initial
tension in the rivet. The difference between two values is small. The value of
moment obtained by Eq. 7.3, viz., M = 0.6 P . g1 is maximum and it is based on
detail analysis. The design of clip angle is done on the basis of this Eq. 7.3.
If, l is the length of connected leg of angle and it is equal to rivet spacing, the
moment of resistance is given by
⎛1 ⎞
MR = (σbt . Z) = ⎜ l ⋅ t 2 ⋅ σbt ⎟ ...(7.8)
⎝6 ⎠
where, Z = Section modulus
σbt = Allowable bending stress in tension.
The moment of resistance of angle leg is equal to moment to be resisted by
the clip angle.
Therefore
⎛1 2 ⎞
⎜ l ⋅ t ⋅ σbt ⎟ = M1
⎝6 ⎠
1/ 2
⎛ 6 M1 ⎞
t = ⎜ ⎟ ...(7.9)
⎝ σbt ⎠
444 DESIGN OF STEEL STRUCTURES

⎛ 1 ⎞
where, Mx = 0.6 P . g1 = 0.6 R. ⎜ g − t ⎟
⎝ 2 ⎠
The thickness of the angle is determined by Eq. 7.9. The allowable bending
stress in tension, 0.66fy used for the design of beam results in over design. The
value of bending stress in tension, σbc in Eq. 7.9 is adopted as 185 N/mm2, which
is permissible bending stress for slab bases as per IS : 800–1984.

7.9 LARGE MOMENT RESISTANT CONNECTIONS

The maximum allowable value of tension in rivets limits the use of connections
shown in Fig. 7.19. In case, the connections are subjected to large moment,
bracket connections or split beam connections are used instead of flange clip
angles. The bracket connections and split beam connections considered as heavy
connections.

7.9.1 Bracket Connections

The bracket connections shown in Fig. 7.22 are composed of gasset plate and
two sets of angles. One set of angles is connected to the column. The other set
consisting of two pairs of angles is used to connect the flange of beam. In the
bracket connections shown in Fig. 7.22 (a) the angles which are connected to
the column are continuous. In the bracket connections shown in Fig. 7.22 (b),
the angles which are connected to the column, are discontinuous.

2 IS A
1 00 m m x 10 0 m m x 1 2 m m
A1
G usset plate G usset plate
40 m m
92 m m
3 20 m m
B1
1 60 m m C1

B e am C o lu m n
4 00 m m B e am
1 S LB 5 00 , @ 0 .7 50 kN/m

1 60 m m C1
B1
3 20 m m
G usset plate G usset plate
40 m m 92 m m
A1
C o lu m n
IS H B 3 00 , @ 0 .5 88 kN /m
(a ) (b )
Fig. 7.22 Bracket connections

Consider bracket connection shown in Fig. 7.22 (a). Rivets on line A1 A2 resist
moment and shear. The other rivets are in double shear. The bearing strength
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 445

of the outer rivets is found, taking thickness of bracket plate. (The thickness of
the bracket plate is kept equal to the thickness of web of beam). It is assumed
that rivets are provided at uniform pitch. The number of rivets is found from
Eq. 2.12.
1/2
⎛ 6M ⎞
n = ⎜ ⎟
⎝ pR ⎠
The rivets are not provided opposite to the flanges of beam. Therefore, it is
not possible to provide uniform pitch for these rivets. The resultant of force due
to direct shear and bending should be less than the rivet value.
The rivets connecting angles to the flanges of column are provided in two
rows. The rivets are subjected to direct shear and tension. The rivet value is
governed by strength of rivet in tension. The number of rivets is found from
Eq. 2.12.
1/ 2
⎛ 6M ⎞
n1 = ⎜ ⎟
⎝ 2 pR1 ⎠
As per IS : 800–1984 these rivets are so proportioned that the quantity.
⎡ τvf .cal σtf .cal ⎤
⎢ + ⎥ does not exceed 1.4
⎣ τvf σtf ⎦
where, τvf.cal = Actual shear stress in the rivet
σvf.cal = Actual tensile stress in the rivet
τvf = Allowable shear stress in the rivet
σtf = Allowable tensile stress in the rivet.
The depth of angles connected to the column is kept to accommodate the
rivets n or n1 whenever is more. The thickness of angle is determined for required
moment of resistance. The angles are subjected to maximum bending moment
given by Eq. 7.3 (viz., M1 = 0.6 P . g1). The thickness of angles, t is found from
Eq. 7.9.
The resultant of horizontal forces of rivets 4 to 8 in line A1 A1 Fig. 7.23 acts at
some distance apart from the line B1B1. Therefore, the rivets in line B1B1 are
subjected to horizontal force and bending moment. The number of rivets in line
B1B1 is found for this moment and checked as discussed above. The allowable
tension in the extreme rivets if found from allowable bending moment in the
flange angle. The allowable bending moment in the flange angle is given by Eq. 7.3.
The rivets connecting flange angles to flanges of beam are provided in two
rows. These are subjected to horizontal forces as for rivets in line B1B1 in Fig.
7.23 and bending moment due to this horizontal force at level C1C1.
The thickness of bracket plate is kept equal to the thickness of web of beam.
This provides ease in construction. The bracket plates and web of beam act as a
rectangular beam. These must be strong enough to resist bending moment act-
ing at line A1A1 in Fig. 7.22. The bending stress at the extreme fibre of this
rectangular beam is given by
446 DESIGN OF STEEL STRUCTURES

⎛ M ⋅ y Agross ⎞
σb = ⎜ × ...(7.10)
⎝ I Anet ⎟⎠
where, I = Gross-moment of inertia of beam formed by bracket plates and
web of beam
y = Distance to the extreme fibre of rectangular beam from neutral
axis as shown in Fig. 7.23 (b)
Agross = Gross area of flanges of the beam
Anet = Net area of flanges of the beam
The strength of bracket plate (gusset plate) may be checked for bending
moment in line B1B1.
2 IS A 1 50 m m x
3 2.3 0 kN
7 5 m m x 12 m m
8 p G usset plate
7 p
y = 68 0 m m

6 p
5 p B1
4 C1 B1
2p C1
1p 3 p C1
2 2 p p1 p 1 p1 p 1 p1 p 1 p 1 p 1 p 1
1 N A
1 p p1 = 6 0 m m
1p 2 p
2 3 2p C1 C1
C1
4 p B1 B1
p = 80 m m 5 p
rive t 6 p
22 m m 7 p
8 2 IS A 7 5 m m × 50 m m G usset plate
× 8 mm
(a ) (b ) (c)
Fig. 7.23

The inclined edge of top or bottom bracket plate will be in tension or com-
pression depending on whether the moments are clockwise and anti-clockwise.
When the inclined edge of the beam will be in compression, the lateral buckling
of plate may take place. The lateral buckling of plate may be prevented either
by providing sufficient thickness of the plate or by providing stiffening angles
as shown in Fig. 7.24.

l= L
l = 12 L L

(a ) (b )
Fig. 7.24
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 447

In Fig. 7.24 (a), it is assumed that inclined edge of the bracket plate acts as a
column and its edges are free to rotate. The lateral buckling of bracket plate
⎛l⎞
will not occur, in case ⎜ ⎟ is less than 60.
⎝r⎠

⎛ l ⎞
⎜ t ⎟ ⎛ t ⎞
⎜⎜ ⎟⎟ < 60, ⎜⎝ For rectangular plate r = ⎟
⎝ 12 ⎠ 12 ⎠

⎛l⎞
⎜ ⎟ < 17.3 ...(7.11)
⎝r⎠

where, l = Effective length of inclined edge of plate

t = Thickness of bracket plate.
In Fig. 7.24 (b), the edges are assumed to be restrained against rotation, then
l ⎛ l ⎞ l
< 60 , ∴ ⎜ ⎟ < 60 , ∴ t < 34.6 ...(7.12)
r 2t
⎜⎜ ⎟⎟
⎝ 12 ⎠

7.9.2 Split Beam Connection

The split beam connection shown in Fig. 7.25 (a) consists of a pair of web angles
and two split beams. A beam-section is cut at the middle of the web. The two cut
sections of a beam are called split beams. Split beams are used as flange clips to
connect the flanges of beam to flange of column. Web clip angles connect the
web of beam to the column flange. Tee-sections are also used instead of split
beams in these types of connections. In split beam connections, it is assumed
that split beams resist moment only and web angles resist shear only.

(S p lit b ea m section )
Flan ge clip
1 (g – t )
P 2 2P w
B
P
W eb clip
C o lu m n a ng le s be am ( h + tw )
M

2P
Flan ge clip
(S p lit b ea m section )

(a ) (b )

Fig. 7.25 Split beam connections

448 DESIGN OF STEEL STRUCTURES

Figure 7.26 (a) and Fig. 7.26 (b) show modified split beam connections. The
depth of connection is increased by providing flange clip seat in between flange
clips and beam. The value of pull transmitted by flange clip is reduced. There-
fore, the number of rivets connecting the split beams to column is reduced.
Figure 7.25 (a) shows split beam connections before loading. Fig. 7.25 (b)
shows split beam connections after loading. If the connections are subjected to a
clockwise moment, the top of split beam undergoes flexure. Because of symme-
try, the stem of split beam does not bend. If one split beam is to transmit a pull
equal to 2P, the rivets connecting split beam to column in each gauge line are
subjected to pull equal to P.

Flan ge clip
Flan ge clip
Flan ge clip Flan ge clip
sea t sea t

W eb clip a n gle s W eb clip a n gle s

b ea m C o lu m n b ea m
C o lu m n

Flan ge clip

Flan ge clip
sea t
Flan ge clip
(a ) (b )

Fig. 7.26 Modified split beam connections

2P × (h + tw) = M
1 M
P = ... (7.13)
2 (h + tw )
where, M = Moment of connections
h = Depth of beam
tw = Thickness of stem of split beam.
The values of moment at points 1 and 3 are equal
( g − tw )
Therefore Ml = M3 = P . ...(7.14)
2
If t f is the thickness of flanges of split beam, and l is the length between
centre to centre of rivets, then
⎛1 2 ⎞
⎜ ⋅ l ⋅ t f ⋅ σbt ⎟ = M1
⎝6 ⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 449

1/2
⎛ 6 M1 ⎞
tf = ⎜ ⎟ ...(7.15)
⎝ l ⋅ σbc ⎠
where, σbt = Allowable bending stress in tension (185 N/mm2)
From Eq. (7.14), M1= 0.5 P . (g – tw)
The thickness of stem of split beam, t w is kept sufficient to transmit the pull,
2P and it provides bearing strength to rivets greater than the strength of rivets
connecting split beams and flange of beam in single shear. The web clip angles
are designed to resist end shear.
Example 7.5 In a framed connection, a LB 400, @ 0.569 kN/m transmits an
end reaction 170 kN and a moment 14.0 kN-m to a column HB 300, @ 0.588 kN-m.
Design a clip connection.
Solution
Design :
Step 1: End moment and end reaction
End moment = 140 kN-m
End reaction =170 kN
Step 2 : Properties of given section
From ISI Handbook No. 1, for LB 400, @ 0569 kN/m
Thickness of flange t f = 12.5 mm
Thickness of web t w = 8.0 mm
For HB 300, @ 0.588 kN/m
Thickness of flange tf1 = 10.6 mm
Thickness of web tw1 = 1.6 mm
Step 3 : Rivet values
22 mm power driven rivets are used
Strength of rivets in single shear
⎛π 2 100 ⎞
⎜ × (23.5) × ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of power driven rivets in double shear
⎛ π 2 100 ⎞
⎜ 2 × × (23.5) × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of power driven rivets in bearing (on t w = 8.00 mm)
⎛ 300 ⎞
⎜ 23.5 × 8.0 × ⎟ = 56.4 kN
⎝ 1000 ⎠
Strength of power driven rivets in bearing (on t f1 = 10.6 mm)
300
3.5 × 10.6 × = 74.73 kN
1000
Strength of power driven rivets in tension
450 DESIGN OF STEEL STRUCTURES

⎛π 2 100 ⎞
⎜⎝ 4 × (23.5) × 1000 ⎟⎠ = 43.35 kN

Rivet values
The rivet value for connecting flange to column is governed by tension
R1 = 43.35 kN
The rivet value for connecting flange clip angles to beam is governed by single
shear
R2 = 43.35 kN
The rivet value for connecting web angles to column is governed by bearing
R3 = 74.73 kN
The rivet value for connecting web angles to beam is governed by bearing
R4 = 56.4 kN
Step 4 : Flange clip angles
Two power driven rivets 22 mm in diameter are used to connect flange clip
angles.
Maximum allowable tension for rivets
= (2 × 43.35) = 86.70 kN
Provide flange ISA 200 mm × 100 mm
The gauge distance for rivets for 100 mm leg from ISI Handbook No. 1 is 60
mm.
Distance between gauge lines of rivets
= (400 + 2 × 60) = 520 mm
Pull transmitted by two rivets
⎛ 14 + 1000 ⎞
P =
⎜ ⎟ = 26.923 kN
⎝ 520 ⎠
< Maximum allowable tension. Hence, rivets are safe.
The maximum moment induced in clip angles is given by Eq. 7.3
1
M1 = (0.6 × g1) = 0.6 P.(g – t)
2
= [0.6 × 26.923 (60 – 7.5)]
= 848.07 kN-mm
(t is assumed as 15 mm)
From ISI Handbook No. 1, rivet spacing for HB 300, @ 0.585 kN/m column
l = 140 mm
Thickness of clip angles of clip angle is given by Eq. 7.9
1/2 1/2
⎛ 6 × M1 ⎞ ⎛ 6 × 848.07 × 1000 ⎞
t = ⎜ ⎟ =⎜ ⎟
⎝ l × σbc ⎠ ⎝ 140 × 185 ⎠
= 14.02 mm
Adopt thickness of clip angle 15 mm
Provide flange clip angle ISA 200 mm × 100 mm × 15 mm (ISA 200 100, @
0.536 kN/m)
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 451

Horizontal shear in rivets connecting flange clip angle to beam

⎛ Moment at the connection ⎞ ⎛ 14 × 1000 ⎞
P1 = ⎜ ⎟⎠ = ⎜⎝ 400 ⎟⎠ = 35 kN
⎝ Depth of beam
Number of rivets required to connect flange clip to beam
⎛ 35 ⎞
⎜ ⎟ = 0.807
⎝ 43.35 ⎠
For the rigid connections, at least 4 rivets are needed. These rivets are pro-
vided in two rows.
Web angles
The number of rivets required to connect web angles to column
170
= 2.275
74.73
Four rivets are provided in two vertical rows. The number of rivets required
to connect web angles to beam
⎛ 170 ⎞
⎜ ⎟ = 3.014
⎝ 50.4 ⎠
Four rivets are provided at a pitch of 80 mm, 2 ISA 100 mm × 75 mm × 8 mm
are provided. The long leg is connected to the beam. The clip angle connection
design, is shown in Fig. 7.27.

IS H B 3 00 @ 0 .5 88 kN /m
IS A 2 0 0 m m x 1 00 m m x
15 m m

40 m m IS LB 40 0
80 m m @ 0 .56 9 kN /m
80 m m 22 m m φ
80 m m rive ts
40 m m

IS A 2 00 m m x 10 0 m m x
15 m m
1 40 m m

Fig. 7.27 Clip angle connection

Example 7.6 In a framed connection, an LB 500, @ 0.759 kN/m transmits

an end reaction 100 kN and an end moment 140 0 kN-m to a column HB 300, @
0.588 kN/m. Design a bracket connection.
Solution
Design
Step 1: End moment and end reaction
End moment = 140.0 kN-m
End moment = 100 kN
452 DESIGN OF STEEL STRUCTURES

Step 2: Properties of given section

From ISI Handbook No. 1, for LB 500, @ 0.759 kN/m
Thickness of flange tf1 = 14.1 mm
Thickness of web tw1 = 9.2 mm
For HB 300, @ 0.538 kN/m, thickness of flange
tf1 = 10.6 mm
Thickness of web twl = 7.6 mm
Thickness of bracket plate is kept equal to the thickness of web of beam = 9.2
mm
Step 3: Rivet values
Use 22 mm power driven rivets.
Stength of rivets

⎛π 2 100 ⎞
In single shear ⎜ × (23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝4 1000 ⎠

⎛ π 2 100 ⎞
In double shear ⎜ 2 × × (23 ⋅ 5 ) × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
In bearing (on t w = 9.2 mm)

⎛ 300 ⎞
⎜ 23 ⋅ 5 × 9 ⋅ 2 × ⎟ = 64.86 kN
⎝ 1000 ⎠
In bearing (on t f1 = 10.6 mm)

⎛ 100 ⎞
⎜ 23 ⋅ 5 × 10 ⋅ 6 × ⎟ = 74.73 kN
⎝ 1000 ⎠

⎛π 2 100 ⎞
In tension ⎜ × (23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝4 1000 ⎠
Rivet values
The rivet value for connecting angles to the flanges of column is governed by
tension,
R1 = 43.35 kN
The rivet value for connecting angles to bracket plate and web of beam is
governed by bearing
R2 = 64.86 kN
The rivet value for connecting angles to the flanges of beam is also governed
by tension
R3 = 43.35 kN
The rivet value for connecting anges to the bracket plate is governed by
bearing,
R4 = 64.86 kN
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 453

Step 4: Rivets
Let number of rivets connecting angles to column in each row at a pitch of 80
mm be n1. There are two vertical rows of rivets.
From Eq. 2.12
1 1
⎛ 6 × M ⎞ 2 ⎛ 6 × 140 × 1000 ⎞ 2
n1 = ⎜ ⎟ =⎜ ⎟ = 11.0
⎝ 2 p × R1 ⎠ ⎝ 2 × 80 × 43 ⋅ 35 ⎠
Let number of rivets connecting angles to bracket plate and web of beam at
a pitch of 80 mm be n. These rivets are provided in one row. From Eq. 2.12
1 1
⎛ 6M ⎞ 2 ⎛ 6 × 140 × 1000 ⎞ 2
n1 = ⎜ ⎟ = ⎜⎝ 80 × 64 ⋅ 86 ⎟⎠ = 12.723
⎝ p × R2 ⎠
Sixteen rivets are provided to connect angles to column, and angles to bracket
plate and beam in each vertical row as shown in Fig. 7.22 (a).
Step 5: Check for rivets
The rivets connecting angles to column are subjected to direct shear and
tension,
Shear for in rivet
100
= 3.125 kN
(2 × 16 )
Shear stress in rivet
⎛ 3 ⋅ 125 × 1000 ⎞
τvf.cal = ⎜ ⎟ = 7.208 N/mm
2
⎛ π⎞ ( 2
⎜⎜ ⎜ ⎟ × 23 ⋅ 5 ) ⎟⎟
⎝⎝4⎠ ⎠
Allowable shear stress in the power driven rivets
τvf = 100 N/mm2
Distance to the extreme rivet from neutral axis, y is 680 mm
⎡ ⎛ 1 ⎞2 ⎛ 3 ⎞2 ⎛ 5 ⎞2 ⎛ 9 ⎞ 2
2 = 4 × 82 ⎢⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟
∑y ⎣⎝ 2 ⎠ ⎝ 2 ⎠ ⎝ 2 ⎠ ⎝ 2 ⎠

2 2 2 2
⎛ 11 ⎞ ⎛ 13 ⎞ ⎛ 15 ⎞ ⎛ 17 ⎞ ⎤
+ ⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟ ⎥ × 100
⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠ ⎦
= 64 × 920 × 100 = 58880 × 100 mm2
Maximum tension in extreme rivet
⎛ 140 × 1000 × 680 ⎞
F1 = ⎜ ⎟ = 16.168 kN
⎝ 5888000 ⎠
Tensile stress in rivet
⎛ 16 ⋅ 168 × 1000 ⎞
τvf.cal = ⎜ ⎟ = 37.296 N/mm
2
π ( 2
⎜ × 23 ⋅ 5 ) ⎟
⎝ 4 ⎠
454 DESIGN OF STEEL STRUCTURES

Step 6: Check for interaction expression

⎡ τvf ⋅cal σvf ⋅cal ⎤ ⎡ 7 ⋅ 208 37 ⋅ 396 ⎤

⎢ + ⎥ = ⎢ + = 0.445 < 1.4 (Safe)
τ τ ⎣ 100 100 ⎥⎦
⎣⎢ vf vf ⎦⎥
The rivets connecting angles to web are in direct shear and bending.

⎛ 100 ⎞
Shear force in rivet = ⎜ = 6.25 kN
⎝ 16 ⎟⎠

∑ y12 = ⎛⎜ 1 × 58880 × 10 ⎞⎟
⎝2 ⎠
= 29440 × 100 mm2
Force in the rivet due to bending moment
⎛ 140 × 1000 × 680 ⎞
Fb = ⎜ ⎟ = 32.337 kN
⎝ 29440 × 100 ⎠
1
2 2
Resultant force, R = ⎡⎣( Fb ) + ( Fb ) ⎤⎦ 2
[6.752 + 32.3372]1/2 = 32.935 kN < Rivet value. Hence, safe.
Angles connecting bracket plate to the column :
Depth of angles = (2 × 720 ) = 1440 mm
Maximum tension in one rivet,
P = 16.188 kN
Use 2 ISA 100 mm × 100 mm
For 100 mm leg, gauge distance is 60 mm
Assume thickness of angle as 12 mm
Step 7: Maximum moment induced in the angle is given by
Eq. 7.3
⎛ 1 ⎞
M1 = 0.6 P × g1 = 0.6 P . ⎜ g – t ⎟
⎝ 2 ⎠
= (0.6 × 16.168 × 54)
= 523.84 kN-mm
The spacing for rivets for HB 300, @ 588.84 kN/m is 140 mm
Thickness of angle is given by Eq. 7.9
1 1
⎛ 6M ⎞2 ⎛ 6 × 523 ⋅ 84 × 1000 ⎞ 2
⎜ l × σ ⎟ = ⎜⎝ 140 × 185
⎟ = 11.016 mm
⎠
⎝ bc ⎠

Adopt thickness of angles as 12 mm

Provide 2 ISA 100 mm × 100 mm × 12 mm to connect bracket plate to column
as shown in Fig. 7.22 (a).
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 455

Step 8: Design of riveted connection

Rivets in the line B1B1
Force in extreme rivet in line A1A1
= 32.935;kN
Force in rivet in line A1A1 at the level of B1B1
⎛ 32 ⋅ 935 + 360 ⎞
= ⎜ ⎟ = 17.436 kN
⎝ 680 ⎠
Similarly, force in other rivets can be found. Resultant of horizontal force
rivets from 4 to 8
⎛ 32 ⋅ 935 + 17 ⋅ 436 ⎞
Figure 7.23 (a), F = ⎜ ⎟ × 5 = 125.927 kN
⎝ 2 ⎠
Line of action of horizontal force above B1B1
320 ⎛ 2 × 32 ⋅ 935 + 17 ⋅ 436 ⎞
×⎜ ⎟ = 176.41 mm
3 ⎝ 32 ⋅ 935 + 17 ⋅ 436 ⎠
Moment at the level B1B1
⎛ 125 ⋅ 927 × 176.41 ⎞
⎜⎝ ⎟⎠ = 22.215 kN
1000
Moment at the level C1C1

⎛ 125 ⋅ 927 [176.41 + (360 250 )]⎞

⎜⎝ ⎟⎠ = 36.0667 kN-m
1000
Let n be number of rivets in line B1B1 at the pitch of 60 mm
These rivets are in one row
1
6 × 22 ⋅ 215 × 1000 ⎞ 2
n1 = ⎛⎜ ⎟ = 5.85
⎝ 60 × 64 ⋅ 86 ⎠
Let n1 be number of rivets connecting angles to flange of beam at a pitch of 60
mm. These rivets in two rows
1
6 × 36 ⋅ 0667 × 1000 ⎞ 2
n1 = ⎛⎜ ⎟ = 6.45
⎝ 2 × 60 × 43 ⋅ 35 ⎠
Provide 9 rivets in each row, connect angles to bracket plate and angles to
beam as shown in Fig. 7.22 (a) and Fig. 7.23 (a).
Step 9: Check for rivets
The rivets in line B1B1 are in direct shear and bending
Direct shear in rivets
⎛ 125 ⋅ 927 ⎞
Fs1 = ⎜ ⎟ = 13.99 kN
⎝ 9 ⎠
456 DESIGN OF STEEL STRUCTURES

∑ y12 = 2 × 62 [12 + 22 + 32 + 42] × 100

= 2160 × 100 mm2
Distance to the extreme rivet y1 is 240 mm
Force in rivet due to bending
⎛ 22 ⋅ 215 × 1000 ⎞
Fb1 = ⎜ × 240 ⎟ = 24.683 kN
⎝ 21600 ⎠
Resultant force in the rivets
1
⎡ Fs2 + Fb2 ⎤ 2 = [13.992 + 24.6832]1/2 = 28.372 kN
⎣ 1 1⎦
< 64 .86 kN, rivet value. Hence, safe.
The rivets in the C1C1 are in direct shear and tension
⎛ 125 ⋅ 927 ⎞
Direct shear Fs2 = ⎜ ⎟ = 13.99 kN
⎝ 2×9 ⎠
⎛ 6 ⋅ 996 × 1060 ⎞
Shear stress τvf.cal = ⎜ ⎟ = 16.138 N/mm
2
⎛ π⎞ ( 2
⎜⎜ ⎜ ⎟ × 23 ⋅ 5 ) ⎟⎟
⎝⎝4⎠ ⎠

∑ y22 = 4320 × 100 mm2

Tension in rivets
⎛ 36 ⋅ 0667 × 1000 ⎞
Ft2 = ⎜ × 240 ⎟ = 20.037 kN
⎝ 432000 ⎠
Tensile stress in rivets
⎛ 20 ⋅ 037 × 1000 ⎞
σtf.cal = ⎜ ⎟ = 46.22 N/mm
2
π
⎜⎜ ⎛⎜ ⎞⎟ × (23 ⋅ 5 )2 ⎟⎟
⎝ ⎝4⎠ ⎠
Step 10: Check for interaction expression
⎡ τvf ⋅cal σvf ⋅cal ⎤ 16 ⋅ 138 46 ⋅ 22 ⎤ = 0.624 < 1.4
⎢ + ⎥ = ⎡⎢ +
σ
⎣⎢ vf
σ tf ⎦⎥ ⎣ 100 100 ⎥⎦
Hence, the rivets are safe.
Angle connecting bracket plate to beam
Depth of angle = (9 × 60) = 540 m
Tension in extreme rivet is 20.37 kN
Use 2 ISA 150 mm × 75 mm. The gauge distance for 70 mm leg from ISI
dbook No. 1 is 40 mm.
Moment induced in angle

⎛ 0 ⋅ 6 × 20 ⋅ 037 × (40 – 6 ) ⎞
M1 = ⎜ ⎟ = 0.4088 kN-m
⎝ 1000 ⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 457

Rivet space for ISLB 500, @ 0.750 kN/m is 100 mm

Thickness of angle is obtained from Eq. 7.3
1
6 × 0 ⋅ 4088 × 1000 × 1000 ⎞ 2
t = ⎛⎜ ⎟
⎝ 100 × 185 ⎠
= 11.514 mm

Provide 2 ISA 150 mm × 75 mm 12 mm (2 ISA 15075, @ 0.201 kN/m) shown in

Fig. 7 .22 (a).
The long leg is connected to bracket plate.
Bracket plate. Thickness of bracket plate is kept equal to the thickness web
of beam = 9.2 mm. It will be prepared by matching from 10 mm thickness of
plate.
At section A1A1. Distance to the extreme fibre of bracket plate
= 720 mm
⎛1 2⎞
Moment of inertia = ⎜ × 9 ⋅ 2 × (720 ) ⎟ = 28650 × 104 mm4
⎝ 12 ⎠
Gross area, A (gross) = (9.2 × 720) = 64.9 × 100 mm2
Net area, A (net) = (64.9 × 100 – 16 × 23.5 × 9.2)
= 3030.8 mm2
Tensile stress in extreme fibre of bracket plate from Eq. 6.28
⎛ M ⋅ y Agross ⎞ ⎛ 140 × 103 × 103 6490 ⎞
σb = ⎜ ⎟ =⎜ × ⎟
⎝ I xx Anet ⎠ ⎝ 28650 × 103 3030 ⋅ 8 ⎠
= 10.4 N/mm2 < Allowable stress. Hence, safe.
2 ISA 75 mm × 50 mm × 8 mm (2 ISA 7550, @ 0.074 kN/m) are connected to
inclined edge of the bracket plate to avoid lateral buckling.
Example 7.7 In a framed connection, an LB 600, @ 0 .995 kN/m transmits an
end reaction 200 kN and an end moment of 40 kN-m to HB 300 @ 0.588 kN/m
column. Design split-beam connections.
Solution
Design
Step 1: End moment and end shear
End moment = 40 kN-m
End shear = 200 kN
Step 2: Properties of given sections
From ISI Handbook No. 1, for LB 600, @ 0.995 kN/m
Thickness of flange tf = 15.5 mm
Thickness of web tw = 10.5 mm
Width of flange b = 210 mm
For BH 300, @ 0.588 kN/m, thickness of flange
tf1 = 10.6 mm
458 DESIGN OF STEEL STRUCTURES

Thickness of web tw1 = 7.6 mm

Step 3: Rivet value
Use 22 mm power driven rivets.
Strength of rivets
⎛π 2 100 ⎞
In single shear ⎜ × (23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝4 1000 ⎠

⎛ π 2 100 ⎞
In double shear ⎜ 2 × × (23 ⋅ 5 ) × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
In beating (on t w = 10.5 mm)
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 10 ⋅ 5 × ⎟ = 74.03 kN
⎝ 1000 ⎠
In bearing (on t f = 10.6 mm)
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 10 ⋅ 6 × ⎟ = 74.73 kN
⎝ 1000 ⎠

⎛π 2 100 ⎞
In tension ⎜ × (23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝4 1000 ⎠
Rivet values. The rivet value for rivets connecting split beams to the flanges
of column is governed by tension
R1 = 43.35 kN
The rivet value for rivet connecting split beam to the flanges of beam is gov-
erned by single shear
R2 = 43.35 kN
The rivet value for rivets connecting web angles to flanges of column is gov-
erned by single shear
R3 = 43.35 kN
The rivet value for rivets connecting web angles to web of beam is governed
by beaing
R4 = 74.3 kN
Split beams. Two rivets 22 mm is diameter are used to connect spit beam to
column.
Maximum allowable tension
= 2 × 43.35 = 86.70 kN
Two rivets are subjected to a pull acting through the centroid of the rivets.
The rivets are assumed to be equally stressed.
Assume thickness of split beam as 10 mm
From Eq. 7.13, pull in split beam
1 M ⎛ 1 40 × 1000 ⎞
P = =⎜ × ⎟ = 32.787 kN
2 (h + tw ) ⎝ 2 (600 + 10 ) ⎠
< Maximum allowable tension
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 459

Assume gauge distance for rivet in split beam as 140 mm

Moment induced in split beam is given by Eq. 7.14.
1
M1 = P ⋅ ( g – tw )
2

⎛ 1 32 ⋅ 787 (140 – 10 ) ⎞
= ⎜ × ⎟ = 2.13 kN-m
⎝2 1000 ⎠
From ISI Handbook No. 1, rivet spance for ISHB 300, @ 0.585 kN/m column,
l is 140 mm
The thickness of flange of split beam is given by Eq. 6.33.
1 1
⎛ 6 M1 ⎞ 2 ⎛ 6 × 2 ⋅ 13 × 1000 × 1000 ⎞ 2
tf = ⎜ ⎟ =⎜ ⎟
⎝ l × σbc ⎠ ⎝ 140 × 185 ⎠

= 22.22 mm
From ISI Handbook No. 1 select WB 600 @ 1.451 kN/m.
This beam section is cut at half the depth. This gives two split beams.
Thickness of flange tf = 23.0 mm
Thickness of web tw = 11.8 mm
Length of stem = 300 mm
Minimum gross width of stem = Width of flange of beam = 210 mm
Net width of stem = (210 – 2 × 23.5) = 163.0 mm
The pull transmitted by split beam
2P = (2 × 32.787) = 65.574 kN
Strength of split beam in axial tension

⎛ 0 ⋅ 6 × 250 ⎞
⎜⎝163 ⋅ 0 × 11 ⋅ 9 × ⎟ = 288.51 kN > 2P . Hence, safe
1000 ⎠
Horizontal shear at the flanges of beam
⎛ 40 × 1000 ⎞
Fs = ⎜ ⎟ = 66.667 kN
⎝ 43 ⋅ 35 ⎠
Number of rivets required to connect split beam
⎛ 66 ⋅ 667 ⎞
⎜ ⎟ = 1.538
⎝ 43 ⋅ 35 ⎠
Four rivets are provided in two rows for rigid frame connections.
Web clip angles and connections
Number of rivets to connect web angles to beam
⎛ 200 ⎞
⎜ ⎟ = 2.676
⎝ 74 ⋅ 73 ⎠
460 DESIGN OF STEEL STRUCTURES

Four rivets are provided at 80 mm pitch.

Number of rivets required to connect web angles to column
⎛ 200 ⎞
⎜ ⎟ = 4.616
⎝ 43 ⋅ 35 ⎠
Eight rivets are provided in two vertical rows.
2 ISA 100 mm × 75 mm × 10 mm (2 ISA 10075, @ 0.130 kN/m) are provided.
The long leg is connected to the beam.
A complete design of split beam connection is shown in Fig. 7.28.

S p lit b ea m
IS W B 6 0 0 @ 1 40
3 00 m m 1 .45 kN /m mm

6 00
80 mm
mm
B e am

80 mm
80 mm
40 mm

IS L B 60 0 @ 0 .9 95 kN /m
IS H B 3 00 @ 0 .58 8 kN /m

R ive ts 22 m m φ

Fig.7.28 Split beam connection

7.10 SEMI-RIGID CONNECTIONS

Flexible, semi-rigid and rigid connections of a beam with columns at its are
shown in Fig. 7.29 (a), (b) and (c), respectively. In order to simplify the calculations
of moments in the members of a frame, in the analysis of structures, the
connections are generally idealized as either flexible rigid. With the flexible end
connections, a beam shall be simply supported and the maximum bending
⎛ wL2 ⎞
moment occurs at the centre, Mc = ⎜ ⎟ where w is the intensity of uniformly
⎝ 8 ⎠
distributed load and L is the span of beam. The beam is then designed for moment,
Mc. The design of beam shall be not economical. With the infinitely rigid
connections, a beam shall be fixed beam and the moments at the fixed ends
⎛ wL2 ⎞
shall be MA = MB = ⎜ ⎟ . By having some appropriate choice of connections
⎝ 12 ⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 461

(say, semi-rigid), the end moments of beam MA, MB and moment at the centre of
beam, Mc may be made equal and equal to maximum bending moment any
⎛ wl2 ⎞ wL2
where in the beam M = 0.5 times ⎜⎝ ⎟ = In that case, the modulus of
8 ⎠ 16
section for a beam needed with semi-rigid connection shall be 50% of that required
for beam with flexible connections, and 75% of that necessary for beam with
rigid end connections.
φ0 φ0
φ φ
U n ifo rm lo a d w
A B A B
C C

wL2 w L2
Mc =
B 16
w L2
ME =
L 16

(a ) Flexib le con n ectio n s (b ) Se m irigid co nn ection s

A B
C

wL2
24
wL2
12
(c) R igid co nn ection s

Fig. 7.29 Flexible semi-rigid and rigid connections

The rigidity of connections is defined as the ratio of resulting end-

moment with semi-rigid connections to the fixed end moment with rigid
connections. For above suggested semi-rigid connections, the rigidity of
⎡⎛ wL2 ⎞ ⎛ wL2 ⎞⎤
connections becomes ⎢⎜ ⎟ divided by ⎜ ⎟ ⎥ equal to 75%. It means, the
⎣⎝ 16 ⎠ ⎝ 12 ⎠⎦
connections provide 75 percent fixity and the moment developed at the end
2
⎛ wL2 ⎞ wL
shall be ⎜ 0 ⋅ 75 ⎟= . In other words it permits 25 percent of rotation for
⎝ 12 ⎠ 16
simply-supported beam. (The flexibility of connections shall be® unity minus
rigidity of connections).
When the semi-rigid connections are used in the frame structures, the
redistribution of moments occurs and the values of maximum bending moment
is reduced. The modulus of section necessary for beam also becomes smaller as
462 DESIGN OF STEEL STRUCTURES

compared with other types of connections. However, in practice, all the riveted
connections act as semi-rigid connections.

7.11 BEHAVIOUR OF SEMI-RIGID CONNECTIONS

The behaviour of semi-rigid connections may be studied by plotting its moment-
rotation characteristics. Professor B.G. Johnston and E. Mount have shown in
their paper titled as ‘Analysis of building frames with semi-rigid connections,
transactions of ASCE 107, pp 993, 1942 that the moment-rotation characteris-
tics (M-φ curve) may be approximately by a straight line representing elastic
behaviour and another line representing post-yielding plastic behaviour. A typi-
cal plate connection as shown in Fig. 7.30 was considered.

L
To p
e
p lat e
T

φ
kh h
v
T

Fig. 7.30 Semi-rigid connection with to plate

A beam of depth, h is connected to a column by using a seat angle at the

bottom and a plate of length, Lp at the top. The beam carries uniformly distrib-
uted load over its span, L. The end of beam rotates by an angle φ, and the plate
is stretched by e, due to tension T developed in the plate. Due to end rotation, a
thrust, C (equal to tension T) also acts against the face of the column. Two
forces tension, T and thrust, C are equal in magnitude and act in opposite direc-
tion and form a couple, M = T.h at the connection. The end rotation φ may be
expressed as under:
e
φ = ( ...(i)
k ⋅ h)

T ⋅ Lp
where, e=
( Ap ⋅ E )
Ap = Area of the plate
k.h = Depth from the top plate to the centre of rotation.
The position of line of rotation depends upon the deformation of connection
between flange of beam and the support. The value of k varies from 0.5 to 1.00.
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 463

The behaviours of all elements of the connections may be assumed elastic and
the deformation of support may be neglected. The value of e may be rules
substituted in expression (i), then

⎛ T ⋅ Lp ⎞
φ = ⎜ ⎟ ...(iii)
⎝ A p ⋅ E ⋅ kh ⎠
The slope of moment-rotation (M – φ) line, θ as shown in Fig. 7.31 may be
stated stated as below:

⎛ θ ⎞⎛ T ⋅ Lp ⎞ ⎛ T ⋅ Lp ⎞
θ= ⎜ ⎟ ⎜ A ⋅ E ⋅ kh ⋅ T ⋅ h ⎟ = ⎜⎜ A ⋅ E ⋅ kh2 ⎟⎟ ...(iv)
⎝M ⎠⎝ p ⎠ ⎝ p ⎠
When the stress in top plate becomes yield stress, σy, the rotation at the end
of beam, θ will increase and the moment, M remains constant. The moment-
rotation curve (M– φ curve) may be idealised as bi-linear (consisting of two
straight lines) as shown in Fig. 7.31.
M L ine a r m om e nt-relatioship
o f a sem i-rigid co nn ection
M om en t

φ
My
M

φ
R o tatio n

Fig. 7.31 Linear moment-rotation diagram (semi-rigid connection)

The value of end momemt, M and rotation at the end of beam, φ also depend
on the intensity of load, w, the span of beam. L and the geometrical properties
of beam section. The value of end moment M for the beam AB, Fig. 7.29 (b) may
be written by slope-deflection as follows. In general,
MAB = MFAB – 2E.K (2φΑB – φBA) ...(v)
MBA = MFBA – 2E.K (2φBA – φAB) ... (vi)
where, MAB and MBA are the end moments and MFAB and MFBA are the fixed end
moments for a beam, K is the relative stiffness of the beam (ratio of moment of
inertia of the cross-section and length of beam). By adding the expression (v)
and (vi)

⎛ M AB + M BA ⎞ ⎛ M FAB + M FBA φ + φBA ⎞

⎜ ⎟ = ⎜⎝ − 2EK AB ⎟⎠ ...(vii)
⎝ 2 ⎠ 2 2
464 DESIGN OF STEEL STRUCTURES

In case, the load is uniformly distributed load on the beam, then

MAB = MBA = M ...(viii)
MFAB = MFBA = MF ...(ix)
The expression (vii) may be rewritten as under:
⎛ 2EI ⎞
M = ⎜ MF – ⋅ (φ ) ⎟ ...(xi)
⎝ L ⎠

⎛ w ⋅ L2 2EI ⎞
or M = ⎜ – ⋅ (φ ) ⎟ ...(7.16)
⎝ 12 L ⎠
From Eq. 7.16, the value of end rotation, φ may be written as below:

⎛ w ⋅ L2 M ⋅ L ⎞
φ = ⎜ – ⎟ ...(7.17)
⎝ 24 EI 2EI ⎠
Since, M = T.h

⎛ w ⋅ L3 T ⋅ h ⋅ L ⎞
φ = ⎜ – ⎟ ...(7.18)
⎝ 24 EI 2EI ⎠
It is to note that if the beam AB would have been simply supported, the slope
⎛L ⎞ T ⋅h⋅L
at end of beam shall be ⎜ 2 ⋅ El⎟ . Second term in Eq. 7.l8, viz., ( represents
⎝ 24 ⎠ 2EI )
the effect of end moment, M. Equation 7.18 is illustrated in Fig. 7.32.
w
T
T
h
T
φ T
L

wL2
8
T h

Fig. 7.32

In the analysis and design of semi-rigid connections of a structure, the

moments at the ends of members for the known loads are determined and the
linear rotation characteristics of a particular connection is found.
The moment-rotation relationship of the connection may be assumed carried.
The analysis of frame structure with semi-rigid connection may then be carried
out either by graphical method (known as beam-line method) or by deal methods,
(namely ; modified slope-deflection method and moment-distribution procedure).
Actually, the moment rotation relationship for the semi-rigid connection is
near as shown in Fig. 7.33. It shows M-φ relationship for semi-rigid connections
1, 2 and 3 and beam lines QR and SV for two beams AB and CD, respectively.
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 465

The semi-rigid connection-1 is moment resisting connection with

connection on web and moment connection by using Tee-sections or bracket
type. This type of connection is used when the span moment is larger than the
end moment. The semi-rigid connection-2 is moment resisting connection with
shear connection with the web and moment connection by seat lip angles. In
this type of connection, the difference between the end moment and span moment
is smaller. Whereas, the semi-rigid connection-3 is frame type or seat type
connection with ordinary cleat angle at the top. In this type the end moments
are bigger than the span moment. An arrangement of connection would be
economical in case the span moments and end moments are equal. In such a
case, a beam is designed for mimimum moment.

S Se m i-rigid
con ne ctio n-1

Se m i-rigid
con ne ctio n-2
M o m en t m

Q
B e am lin es

Se m i-rigid
con ne ctio n-3

O R V φ
R o tatio n φ

Fig. 7.33 Non-linear moment-rotation diagram semi-rigid connection

In case, the sign of moment M as well as its magnitude are considered, the
analysis becomes extremely complex. Further, when the lateral forces are
considered, the analysis by analytical method becomes impractical.

7.12 BEAM-LINE METHOD

Professor Batho, C and Rowan, H.C. in their paper titled as ‘The analysis of the
moments in the members of a frame having rigid or semi-rigid connections’
published in second report, Steel Structures Research Committee, London,
H.M.S.O., 1934 and Professor Batho, C in his paper titled as ‘The analysis and
design of beams under given end restraints published in final report, Steel
Structures Research Committee, London, H.M.S.O. 1936 gave graphical method
applicable to single-bay frames in which there is no away or sideways deflection.
The relation between end moment, M and end rotation, φ for a beam AB of
uniform section, loaded symmetrically with identical semi-rigid connections at
its ends, Fig. 7.29 is given by the straight line
466 DESIGN OF STEEL STRUCTURES

⎛ 2EI ⋅ (φ ) ⎞
M = ⎜ MF – ⎟ ...(i)
⎝ L ⎠
where, MF is the fixed end moment for the beam AB with fixed ends. The straight
line represented by the expression (i) is called as beam-line. This light line is
plotted graphically as QP as shown in Fig. 7.34. Point Q is plotted for the beam
wL2
AB fixed at its both the ends for which end-moment M = and end-rotation
12
φ = φ0 is zero. Point P is plotted for the beam AB simply supported at its both the
ends for which end-moment M is zero and end rotation φ = φ0. Points Q and P
are joined, which represents the beam-line QP the beam AB, having semi-rigid
connections at its ends. When the beam AB designed for simply supported ends,
then, maximum bending moment is Mc.

wL2 ⎛σ I ⎞ ⎛ I⎞
Mc = = ⎜ b⋅ ⎟ = ⎜ σb⋅ ⎟ ...(ii)
8 ⎝ y ⎠ h
⎜ ⎟
⎝ 2⎠

⎛ w ⋅ L2 ⎞ 2 ⎛σ ⎞ ⎛L⎞
and φc = ⎜ ⎟ = = ⎜ b⎟ = ⎜ ⎟ ...(iii)
⎝ 24 EI ⎠ 3 ⎝ E ⎠ ⎝h⎠

M
S
2M F
B e am lin es

Q
MF
B e am
line s

φ
R o tatio n φ0 2 φ0 4 φ0
P R V
⎛ w ⋅ L2 w ⋅ L3 ⎞
⎜ M F = 1 2 φ0 = 2 4 E I ⎟
⎝ ⎠

Fig. 7.34 Moment-rotation diagram (Semi-rigid connections)

⎛L⎞
This expression (iii) shows that (φ0 is directly proportional to ratio ⎜ ⎟ for
⎝h⎠
simply supported beam for given bending stress, σb, but the moment, Mc =
⎛ w ⋅ L2 ⎞ ⎛I⎞ ⎛L⎞
⎜⎝ ⎟⎠ = σb ⎜ ⎟ is not dependent on the ratio ⎜⎝ h ⎟⎠ . The end rotation φ0
8 ⎝ y⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 467

⎛ wL2 ⎞ ⎛ w ⋅ L2 ⎞
⎜ ⎟ and the fixed end-moment M = MP = ⎜ ⎟ are directly portional to
⎝ 24El ⎠ ⎝ 12 ⎠
the intensity of uniformly distributed load, w. The influence of rotation span/
depth and load, w on M–φ relationship for a beam having semi-rigid connections
at its ends is also shown in Fig. 7.34.
⎛w ⎞
The straight line QR represents a beam-line when the load w1 = ⎜ ⎟ and
⎝4⎠
span Lx = 2L. From Eqs. 7.16 and 7.17

⎛ w ⋅ L2 ⎞ 1 ⎛w ⎞
MF1 = ⎜⎝ 1 1 ⎟⎠ = . ⎜ ⎟ . × (4 L2) =MF ...(iv)
12 12 ⎝ 4 ⎠

⎛ w ⋅ L3 ⎞ 1 ⎛ w ⎞ ⎛ 8 L3 ⎞
φ0.2 = ⎜ 1 1 ⎟ = ⋅⎜ ⎟⎜ ⎟ = 2φ0 ...(v)
⎝ 24 EI ⎠ 24 ⎝ 4 ⎠⎝ EI ⎠
The straight line SR represents a beam-line when the load w2 = (2w) and the
span L2 = I. From Eqs. 7.16 and 7.17

⎛ w L2 ⎞ (2w ) ⋅ L2 2 ⋅ wL2
MF2 = ⎜ 2 2 ⎟ = = 2× = 2M F ...(vi)
⎝ 12 ⎠ 12 4

3 3
φ0.2 = ⎛⎜ w2 L2 ⎞⎟ = ⎛⎜ (2w) ⋅ L ⎞⎟ = 2φ ...(vii)
0
⎝ 24 EI ⎠ ⎝ 24 EI ⎠
The straight line SV represents the beam-line, when load w2 = (0.5 w) and
the span L3 = L. From Eqs. 7.16 and 7.17

⎛ w L2 ⎞ 1 ⎛ (2w) ⋅ L3 ⎞ 2 ⋅ wL2
MF3 = ⎜ 3 3 ⎟ = ⎜ ⎟ = 2φ0 × = 2M F ...(viii)
⎝ 12 ⎠ 2 ⎝ 24EI ⎠ 4

2 3
φ0.2 = ⎛⎜ w3 L3 ⎞⎟ = 1 ⋅ ⎛⎜ w ⎞⎟ 8L = 4 φ ...(ix)
0
⎝ 24 EI ⎠ 24 ⎝ 2 ⎠ 24 EI
The beam lines QP, QR, SR and SV as shown in Fig. 7.34 are distinctly differ-
ent from the moment rotation (M–φ) relationship of a semi-rigid connection shown
in Fig. 7.31.
For a particular structure, the line may be defined and the end moment and
end rotation (viz. M–φ) characteristics of the beam and column connections are
known, then the corresponding values of end moment, M and end rotation, φ
may be found graphically as shown in Fig. 7.35.
Figure 7.35 shows beam lines QR and SV for beams AB and CD, respectively.
It also shows moment and end rotation characteristics OAB and OCB (assuming
the relationship to be linear) for the semi-rigid connections 1 and 2, respectively,
having different rigidities. The points of intersection of the beam line QR and
M–φ curve for the semi-rigid connection-1, OAB, (viz., point E) gives the values
468 DESIGN OF STEEL STRUCTURES

of end moment M and end rotation φ. These values for the beam AB and the semi-
rigid connection-1 remains same. The point of intersection of the beam-line, SV
and M–φ curve OAB (viz., point F) gives the values of end moment M,

S A B Se m i-rigid
con ne ctio n-1

C D Se m i-rigid
con ne ctio n-2
H
Q
B e am lin e
E (b ea m C D )
G

O R V φ

Fig. 7.35 Beam-lines and moment-rotation characteristics (semi-rigid connections)

and φ end rotation φ1. It is seen that the semi-rigid connection is suitable and
adequate for both the beams AB and CD. Whereas, the semi-rigid connection-2
is suitable and adequate only for beam AB. This connection does not have reserve
capacity for overload for the beam CD.

7.13 MODIFIED SLOPE-DEFLECTION METHOD

Batho’s graphical method (beam line method) is based on the true moment-
rotation characteristics curves for the semi-rigid connections. It was not
possible to extend his graphical method for the analysis of multi-storey multi-
bay frames. The relationship between end moment and end rotation may be
made linear by replacing these curves by their chords. Then, the analysis of
semi-rigid connections may be carried out by modified slope-deflection method
discussed as under. Following sign convention and assumption are made for
the analysis.
The clockwise rotations of joints are considered positive. The end-moments
applied by the joints to the members are positive when the rotation due to
moment is clockwise. The deflection of one end of a member relative to the other
is positive when the line joining the ends rotates clockwise.
The frame structures are considered two dimensional and consist of lines of
vertical columns, fixed at supports, and connected by lines of horizontal beams.
Any change in the section at a column is made at the floor level and the splices
are perfectly rigid. The section of each beam is uniform throughout its length.
The width of column is taken into account. However, the beams are represented
by their neutral axes. The relative rotation of the members at a beam to column
connection is proportional to the moment transmitted by the connection, (viz.
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 469

the moment-rotation relationships of the connections known and assumed linear

i.e., M. = stiffness × rotation). The deformations resulting from shear and direct
stresses in the members, and the strut effect in the columns are neglected. In any
particular problem, many of these assumptions may be amended to represent
practical conditions more closely. It is desirable to keep the general expressions
which follow as simple as possible. Therefore, it is assumed that only uniformly
distributed vertical loads, w per unit length, act on the structure.

U n ifo rm lo ad w

φB
φA M BA
B
VAB

VAB

(a ) S im p ly sup p orte d be am

θ θ
A B

A B

φ φB
A

(b ) P a rt o f a loa de d fram e w ith se m i-rigid b ea m con ne ctio n

Fig. 7.36

Consider a simply supported horizontal beam of length, L and relative stiffness,

K with end couples MAB and MBA applied to its ends as shown in Fig. 7.36 (a). The
end B of beam AB settles down by a height, δ relative to its end A. Let φA and φB be
the rotations at the ends. Then
⎡ 3δ ⎤
MAB = 2EK ⎢2 A + φ B – ⎥ + M F ⋅ AB ...(i)
⎣ L⎦
⎡ 3δ ⎤
MBA = ⎢2 A + 2φ B – ⎥ + M F ⋅ AB ...(ii)
⎣ L⎦

where MF.AB and Mf.BA are the fixed end moments due to loads acting over the
beam, taken with correct sign. A part of a loaded frame structure having semi-
rigid connections of beams with the columns is shown diagrammetically in Fig.
7.36 (b). The tangents and rotations at the ends of beams and columns are also
470 DESIGN OF STEEL STRUCTURES

shown. Total rotations at the joints A and B are (θA – φA) and (θΑ – φB) respectively.
Since, the moments at the ends are proportional to the end rotation.
MAB = KA.(QA – φA) ...(iii)
MBA = KB.(θB – φΒ) ...(iv)
where KA and KB are the rigidity co-efficients of the connections at A and B.
Equating the moments in the expressions (i) and (ii) and moments in the
expressions (ii) and (iv) and rearranging the terms
⎡ 3δ ⋅ α α ⎤
φA = ⎢(2α + 1 ) θ A + α ⋅ θ B – + ⋅ M F ⋅ AB ⎥ ...(v)
⎣ L 2EK ⎦

⎡ 3δ ⋅ β β ⎤
φB = ⎢β ⋅ φ A (2β + 1) ⋅ θ B – + ⋅ M F ⋅BA ⎥ ...(vi)
⎣ L 2EK ⎦

⎛ 2EK ⎞ ⎛ 2EK ⎞
where α = ⎜ ⎟ and β = ⎜ ⎟
⎝ KA ⎠ ⎝ KB ⎠
Solving the expression (v) and (vi) for φA and φB,

⎛ 1 ⎞
φA = ⎜⎝ 3αβ + 2α + 2β + 1 ⎠⎟ ⎡⎣(2β + 1) φ A – α ⋅ θ B

δ α
+ 3α (β + 1 ) – {(2β + 1) M F ⋅ ABβ ⋅ M F ⋅BA }⎤⎥ ...(vii)
L 2EK ⎦

⎛ 1 ⎞
φB = ⎜ ⎡( – β ⋅ θ A + 1) ⋅ φ B – α ⋅ θ B
⎝ 3αβ + 2α + 2β + 1 ⎠⎟ ⎣

δ α
+ 3β (α + 1 ) – {– α ⋅ M F ⋅ AB β ⋅ (2α + 1) M F ⋅BA }⎤⎥ ...(viii)
L 2EK ⎦
Values of θA and θB are substituted in the expressions (i) and (ii) and the
terms are rearranged. The end moments MAB and MBA may be found as below:

⎛ (2β + 1 ) M F ⋅BA – β ⋅ M F ⋅ BA ⎞ ⎛ 2EK ⎞

MAB = ⎜ ⎟+⎜ ⎟
⎝ (3α ⋅ β + 2α + 2β + 1) ⎠ ⎝ (3α ⋅ β + 2α + 2β + 1 ) ⎠

⎡ 3δ ⎤
× ⎢(3β + 2 ) θB – (β + 1) ⎥ ...(7.19)
⎣ L⎦

⎛ – α.M F ⋅ AB (2β + 1) M F ⋅BA ⎞ ⎛ 2EK ⎞

MBA = ⎜ ⎟+⎜
⎝ (3α ⋅ β + 2α + 2β + 1 ) ⎠ ⎝( 3α ⋅ β + 2α + 2β + 1 ) ⎟⎠
⎡ 3δ ⎤
× ⎢θ A + (3α + 2 ) θ B – (α + 1 ) ⎥ ...(7.20)
⎣ L⎦
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 471

In case, there is no settlement of end B, and for uniformly distributed load, w,

⎛ –wL2 ⎞ ⎛ wL2 ⎞
MF.AB = ⎜ ⎟ and MF.BA = ⎜ ⎟
⎝ 12 ⎠ ⎝ 12 ⎠

⎛ 2EK ⎞ ⎡ (3β + 1)wL2 ⎤

MAB = ⎜ (3α ⋅ β + 2α + 2β + 1 ) ⎟ ⎢(3β + 2 ) θ A + θ B ⎥ ...(7.21)
⎝ ⎠ ⎣ 24 E ⋅ K ⎦

⎛ 2EK ⎞ ⎡ (3 + 1 ) wL2 ⎤
MBA = ⎜ (3α ⋅ β + 2α + 2β + 1 ) ⎟ ⎢θ A (3α + 2 ) θ B + θ B ⎥ ...(7.22)
⎝ ⎠ ⎣ 24 E ⋅ K ⎦
The shears at the ends of beam are given by
⎛ M + M BA ⎞
VAB = VS.AB+ ⎜ AB ⎟ ...(7.23)
⎝ L ⎠
⎛ M + M BA ⎞
VBA = VS.BA+ ⎜ AB ⎟ ...(7.24)
⎝ L ⎠
Where VS.AB and VS.BA are the shears at the ends of beam as a simply supported
beam considering the shear to the right of section as positive.
Consider the equilibrium of any joint, say B. Figure 7.37 (a). The end shears
eccentric with respect to the joint. Let the eccentries of end shears be e BA and
e BC as shown in Fig. 7.37 (b) for the equilibrium of joint B.

B B
A C

eAB eB C

E
(a ) B e am s con ne cted to co lu m n (b ) Fre e -bo dy d ia gra m

Fig. 7.37 Equilibrium of joint

MBA + MBC + MBD + MBE + MBE + VBA.eBA + VBC.eBC = 0

or ∑MB + VS.BA . eBA + VSBC.eBC + eBA ( M + M ) + eBC ( M + M ) = 0

AB BA BC CB
LBA LBC
...(7.25)
For the equilibrium of a storey
∑ (column moments of a storey) + PH h = 0 ...(7.26)
where PH is the total horizontal shear and h is the storey height.
472 DESIGN OF STEEL STRUCTURES

7.14 MODIFIED MOMENT DISTRIBUTION METHOD

Analysis of the bending frames with semi-rigid beam to column connections may
also be done by the modified moment distribution method. The expressions for
fixed end moments, stiffness and carry-over factors for the members will be modi-
fied because of end rotations in the semi-rigid connections.
The joints are considered first fixed in position. The deformations (viz., joint
rotations θA and θB, and the displacement, δ) shall be zero. The modified fixed
end moments are now called as initial end moments. The initial end moments
MI.AB and MI.BA may now be written from Eqs. 7.19 and 7.20 below

⎛ (2β + 1 ) M F ⋅ AB – β ⋅ M F ⋅ BA ⎞
Mi.AB = ⎜ ...(7.27)
⎝ (3α ⋅ β + 2α + 2β + 1) ⎟⎠

⎛ – α ⋅ M F ⋅ AB + (2α + 1) M F ⋅BA ⎞
Mi.AB = ⎜ ⎟ ...(7.28)
⎝ (3α ⋅ β + 2α + 2β + 1) ⎠
The modified stiffness may be defined as the moment required for a unit
rotation (viz., 1 radian) at the joint when the far end is kept fixed. Then, MIBA,
θB and δ shall remain zero, and θΑ = 1 from Eq. 7.19.

⎛ 2EK ⋅ (3β + 2) ⎞
K ´AB = ⎜ ...(7.29)
⎝ (3α ⋅ β + 2α + 2β + 1) ⎟⎠
Similarly, when MIBA, θA and δ shall remain zero, and θB = 1 from Eq. 7.20.
⎛ 2EK ⋅ (3β + 2 ) ⎞
K ´AB = ⎜ ⎟ ...(7.30)
⎝ (3α ⋅ β + 2α + 2β + 1 ) ⎠
The modified carry over factor may be defined as the ratio of moment at the
far end (being fixed) to that at the near end (being rotating without transla-
tion). The initial end moment, MiBA, end rotating, θB and the displacement, δ
are zero. The modified carry-over factor from A to B and from B to A (θB = 8 = 0,
and MiAB, = 0) are as below :

⎛ MBA ⎞ ⎛ 1 ⎞
CAB = ⎜ M ⎟ = ⎜ (3β + 2 ) ⎟ ...(7.31)
⎝ BA ⎠ ⎝ ⎠

⎛M ⎞ ⎛ 1 ⎞
CBA = ⎜ AB ⎟ = ⎜ ⎟ ...(7.32)
⎝ M BA ⎠ ⎝ (3β + 2 ) ⎠
The moment developed at fixed ends due to settlement of any support, d
relative to the other may be determined by considering initial end moments
MiAB, and MiBA,, and end rotations θA and θB as zero. Then

⎛ 6 (β + 1 ) EK ⎞
MAB.δ = ⎜ (3α ⋅ β + 2α + 2β + 1 ) ⎟ ...(7.33)
⎝ ⎠
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 473

⎛ 6 (α + 1) EK ⎞
MBA.δ = ⎜ (3α ⋅ β + 2α + 2β + 1) ⎟ ...(7.34)
⎝ ⎠
The analysis may be simplified by having advantage of symmetry or anti-sym-
metry, if any. For symmetry, the end rotations θΑ = –θΒ and α = β. The initial end
moments and displacement, δ shall be zero. From Eq. 7.19

KAB . sym = KBA . sym = ⎛⎜ 2EK ⎞⎟ ...(7.35)

⎝ ( α + 1) ⎠
And for initial end moments
MF . AB = – MF . BA
⎛ M F ⋅ AB ⎞
Mi . AB . sym = Mi . BA . sym = ⎜ ( ⎟ ...(7.36)
⎝ α + 1) ⎠
For anti-symmetry, the end rotations θA = θB =1 and α = β. The initial end
moments and displacement, δ shall be zero. From Eq. 7.19
⎛ 6EK ⎞
KAB . sym = KBA . sym = ⎜ ( ⎟ ...(7.37)
⎝ 3α + 1) ⎠
And for initial end moments
MF . AB = + MF . BA
⎛M ⎞
Mi . AB . asym = Mi . BA . asym = ⎜ F ⋅ AB ⎟ ...(7.38)
⎝ ( 3α + 1 ) ⎠
For uniformly distributed load
⎛ wL2 ⎞
Since, ⎜ – M F ⋅ AB = M F ⋅BA = ⎟ , the initial end moments from Eq. 7.19 are
⎝ 12 ⎠
as follows:
⎛ – (3α + 1 ) ⎞ wL2
Mi . AB = ⎜ 3α ⋅ β + 2α + 2β + 1 ⎟ ⋅ 12 ...(7.39)
⎝( )⎠

⎛ (3α + 1 ) ⎞ wL2
Mi . AB = ⎜ ⎟⋅ ...(7.40)
⎝ (3α ⋅ β + 2α + 2β + 1 ) ⎠ 12
Example 7.8 A beam of uniform section 6 m long carries a uniformly distributed
load 40 kN/m. The end moments and span moment are equal. Determine the
stiffness of end connections and end rotation.
Solution
The end moments and span moment are equal. It is possible when
⎛ wL2 ⎞
–MAB = MBA = ⎜ ⎟ ...(i)
⎝ 16 ⎠
And, the similar semi-rigid connections having equal rigidities are used at
both the ends. Therefore,
474 DESIGN OF STEEL STRUCTURES

⎛ 2EK ⎞
α = β= ⎜ ⎟
⎝ KA ⎠
Fixed end moments
⎛ wL2 ⎞
– M F ⋅ AB = M F ⋅BA = ⎜ ⎟ ...(ii)
⎝ 12 ⎠
The initial end moment from Eq. 7.27
⎛ (2β + 1 ) M F ⋅ AB – β ⋅ M F ⋅ BA ⎞
Mi . AB = ⎜ (3 α = β)
⎝ (3α ⋅ β + 2α + 2β + 1) ⎟⎠
wL2 ⎛ – (3α + 1) ⋅ wL2 ⎞ 1
– = ⎜ ∴α=
16 ⎟
⎝ (3α + 1)( α + 1) ⋅12 ⎠ 3
The stiffness of semi-rigid connection at A
⎛ 2EK ⎞
KA = ⎜ = 6 E.K
1 ⎟
⎜ ⎟
⎝ 3 ⎠
The bending moment diagram shall be as shown in Fig. 7.29 (b). From the
area-moment theorem, area of bending moment diagram from end to half the
span.
1 ⎛ –wL2 L 2 wL2 L ⎞
End rotation θA = ⋅ + ⋅ ⋅ ⎟
EI ⎜⎝ 16 2 3 8 2⎠

⎛ wL3 ⎞ ⎛ 40 × 63 × (1000 )3 ⎞
θA = ⎜ ⎟ = ⎜ ⎟ rads.
⎝ 96EI ⎠ ⎝ 96EI ⎠
3 –1
θA = 90 × (1000) × (EI) rads. ... (iii)
For simply supported span, the end rotation

⎛ wL3 ⎞ ⎛ 40 × 6 × (1000 )3 ⎞
θA = ⎜ ⎟ = ⎜ ⎟ rads
⎝ 24 EI ⎠ ⎝ 24EI ⎠
3
θA = 360 × (1000) × (EI) rads–1

From the expressions (iii) and (iv) it is seen that the end rotation for semi-
rigid connection is 25 percent of that for simply supported span. The rigidity of
end connections is 75 percent.
Example 7.9 A single bay single storey portal frame carries uniformly distributed
load 40 kN/m over its beam 6 m long. The rigidity of semi-rigid connection at
either end of beam with the column is 80 percent. Determine the design moments.
The columns and beam have uniform cross-section.
Solution
Step 1: End moment and end rotation
The rigidity of semi-rigid connection at end B, (and 50 also of end C) of beam
BC of portal frame ABCD as shown in Fig. 7.38 is 80 percent. Therefore, the
 DesignOF
DESIGN of BEAMS
Beam CONNECTIONS
Connection 475

fixed end moment developed shall be 80 percent and free span rotation at the end
shall be 20 percent.
End moment, (MBC = MCB)
⎛ 0 ⋅ 8 × wL3 ⎞ ⎛ 0 ⋅ 8 × 40 × 6 × 6 ⎞
MBC = ⎜ ⎟ = ⎜ ⎟ = 96 kN-m
⎝ 12 ⎠ ⎝ 12 ⎠
End rotation, (θBC = θCB)
⎛ 0 ⋅ 20 × wL3 ⎞ ⎛ 0 ⋅ 20 × 40 × 63 ⎞
θBC = ⎜ ⎟ = ⎜ ⎟ 12 × 6 (EI)–1 rads.
⎝ 24 EI ⎠ ⎝ 24 ⎠

4 0 kN /m

B C

1 1

A D

6m

Fig. 7.38

Step 2: Stiffness of members

Modified stiffness of joint KB = KC

⎛ End moment ⎞ ⎛ 90 ⎞ 8 EI
KB = ⎜ ⎟ = ⎜ ⎟ = = rad.
–1
⎝ End rotation ⎠ ⎝ 12×6 (EI ) ⎠ 6

1
Relative stiffness of beam KBC =
6

⎛ I⎞
⎛ 2EK BC ⎞ ⎜ 2E ⋅ 6 ⎟
By symmetry α = β = ⎜ ⎟ = ⎜ ⎟ = 0.25
⎝ KB ⎠ ⎜⎜ 8E I ⎟⎟
⎝ 6 ⎠
Fixed end moment (MF . BC = – MF . CAB)
2
MF . BC = ⎛⎜ wl ⎞⎟ = ⎛⎜ 40 × 6 × 6 ⎞⎟ = 120 kN-m
⎝ 12 ⎠ ⎝ 12 ⎠
For symmetry
Initial end moment (Mi . BC , Mi . CB)
476 DESIGN OF STEEL STRUCTURES

⎛ M F ⋅BC ⎞ ⎛ 120 ⎞
Mi . BC = ⎜ ⎟ = ⎜ ⎟ = 96 kN-m
⎝ 1 + 0 ⋅ 25 ⎠ ⎝ 1 + 0 ⋅ 25 ⎠
Relative stiffness of end B
⎛ I ⎞
⎛ 2E ⋅ K BC ⎞ ⎜ 2E ⋅ 6 ⎟
KBC . sym = ⎜ ⎟ = ⎜ ⎟ = 0.267 EI
⎝ 1+α ⎠ ⎝ 1 + 0 ⋅ 25 ⎠
For column BA fixed at A
⎛ 4 EI ⎞
KBA = ⎜ ⎟ = EI
⎝ 4 ⎠
Step 3: Distribution factors at joint B
KBC .KBA = 0.211:0.789
Step 4: Moments at sections
Moment at B at the top of the column
MBA = 0.789 × 96 = 75.744 kN-m
Moment at the centre of beam

= ⎛⎜ 40 × 6 × 6 ⎞⎟ – 75.256 kN-m
⎝ 8 ⎠

PROBLEMS
7.1. AN MB 200, @ 0.254 kN/m transmits an end reaction of 186 kN to the
web of an MB 450, @ 0.794 kN/m. Design a framed connection. Give a
neat sketch.
7.2. An LB 325, @ 0.431 kN/m transmits an end reaction of 124 kN to the
flange of a column HB 300, @ 0.630 kN/m. Design an unstiffened seated
connection.
7.3. An LB 600, @ 0.995 kN/m transmits an end reaction of 286.0 kN to the
flange of an HB 400, @ 0.774 kN/m column. Design stiffened seated con-
nections.
Provide (i) a single stiffener angle (ii) double stiffener angles.
7.4. In a framed connection, an LB 325, @ 0.431 kN/m transmits an end reac-
tion 148.0 kN and a moment 21.0 kN-m to a column HB 300, @ 0.530 kN/
m. Design a clip angle connection.
7.5. In a framed connection, an MB 450, @ 0.724 kN/m transmits an end reac-
tion 120 kN and an end moment 156.0 kN-m to a column HB 400, @ 0.774
kN/m. Design a bracket connection.
7.6. In a framed connection, an MB 500, @ 0.869 kN/m transmits an end reac-
tion 162 kN and an end moment 36 kN-m to HB 400, @ 0.822 kN/m. Design
a split beam connection.
7.7. Design an unstiffened seat connection to support a beam LB 350, @ 0.495
kN/m transmitting end reaction of 100 kN to the flange of column HB 200,
@ 0.400 kN/m. Use 16 mm dia. rivets. Show the details in a neat sketch.
 Chapter

8
Design of Plate Girder

8.1 INTRODUCTION
When the span and load combination is such that the rolled steel beams become
insufficient to furnish the requirement, and built-up beams become uneconomi-
cal, then plate girders are used. The built-up beams are used where overall
depth is limited. As such the use of beams, built-up beams and plate girder is a
step by step approach for the increase in loads and the spans. The object of
design is to achieve overall economy, which involves the cost of fabrication in
addition to the cost of material. The cost of fabrication is more for built-up beams
as compared to beams, and it is still higher for plate girders as compared to
both. Attempt is made to provide deep sections for economy as regards material
and cost of fabrication.
The plate girders are deep structural members subjected to transverse loads.
They consist of plates and angles riveted together. Plates and angles form an I-
section. They are used in building constructions and also in bridges. The plate
girders are economically used for span upto about 30 m in building construc-
tion. The depth of plate girder may range upto 5 m or more, 1.5 m to 2.5 m
depths are very common. Common sections used for plate girders are shown in
Fig. 8.1.
Figure 8.1 (a) shows the simplest form of a plate girder. In case, this simple
section cannot furnish sufficient flange material, additional plates are riveted
to outstanding legs of angles as shown in Fig. 8.1 (b) and (c). When number of
cover plates becomes excess, then the section of plate girder is modified as shown
in Fig. 8.1 (d). In case of deck girders of rail-road bridges, sometimes a form
shown in Fig. 8.1 (e) is used. Sometimes for long span bridges carrying heavy
loads, heavy sections shown in Fig. 8.1 (f) and (g) are used. Section shown in
Fig. 8.1 (h) is adopted for crane runway girders.
478 DESIGN OF STEEL STRUCTURES

Sometimes in building construction, the depth of a girder is limited to provide

necessary head room. Limitations of head room require the depth of girder much
less than the economical depth. It becomes necessary to provide large

(a ) (b ) (c) (d )

(e ) (f) (g ) (h )

(i) (j)

Fig. 8.1 Common section of plate girder

web area to resist shear forces. In such cases, sections shown in Fig. 8.1 (i) and
(j) are often used in which two or more webs are provided. These are called box
girders. In addition to shear, webs resist bending moment to a considerable
extent.

8.2 ELEMENTS OF PLATE GIRDER

The vertical plate of the plate girder is termed as web plate. The angles connected
at the top and bottom of the web plate are known as flange angles. The horizontal
plates connected with the flange angles are known as flange plates or cover
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 479

plates. The web plate, flange angles and flange plates are shown in Fig. 8.2. The
bearing stiffeners, intermediate stiffeners and horizontal stiffeners used with
the plate girder are shown in Fig. 8.3.

Flan ge p la te s

Flan ge a ng le s

W eb p la te

Flan ge a ng le s

Flan ge p la te s

Fig. 8.2 Elements of plate girder

A web plate is kept unstiffened when the ratio of clear depth to thickness of
web is less than 85. It does not require stiffeners. A web plate is called stiffened,
when the ratio of clear depth to thickness of web is greater than 85 and stiffen-
ers are provided to contribute additional strength to web.
H o rizon ta l Flan ge a ng le s Flan ge p la te
stiffe ne r W eb p la te

B e aring stiffen er In te rm ed ia te B e aring stiffen er

stiffe ne rs
Fig. 8.3 Elements of a plate girder

8.3 AREA OF FLANGES OF PLATE GIRDER

When a plate girder is carrying transverse load, its top flange is in compression
and bottom flange is in tension. The area of flange comprises of areas of flange
angles, flange plates (if flange plates are attached) and part of the web plate.
The parts of web plate sandwiched between flange angles also resist bending
moment. The part of web plate which is available for flange area is computed as
below.
⎛ I ⎞
Bending moment, M = σbc ⎜ xx ⎟
⎝ ymax ⎠
where, Ixx = Gross moment of inertia of plate girder
σfc = Allowable bending stress
ymax = Vertical distance of extreme fibre of section from neutral axis
480 DESIGN OF STEEL STRUCTURES

The gross moment of inertia of plate girder

⎡ 2
t ⋅d ⎤
3
⎛d ⎞
Ixx = ⎢2 A f ⎜ e ⎟ + w e ⎥ approximately ...(i)
⎣ ⎝ 2 ⎠ 12 ⎦
where, Af = Gross area of flange
de = Effective depth
tw = Thickness of web
The moment of inertias of flange plates and flange angles about their own
axes are neglected. Then

⎛ σb × 2 Af × de 2 ⎞ ⎛ σ × t × d 3 ⎞
M = ⎜ ⎟+⎜ b w e
⎟ (ii)
⎜ de ⎟ ⎜ de
⎜ ⎟ ⎜ ⎟⎟
⎝ 2×4 ⎠ ⎝ 2 × 12 ⎠

de
M = ( Af × σb × de ) + (tw × dd ) × σb × 6
(iii)

⎛ de ⎞
M = ( Af × σb × de ) + ⎜⎝ Aw × σb × 6 ⎠
⎟ (iv)

⎛ 1 ⎞
∴ M = ⎜ Af + Aw ⎟ × σb × de (8.1)
⎝ 6 ⎠
The effective depth is considered approximately equal to the depth of web, where,
Aw is the area of web.
In the riveted plate girder, the vertical stiffeners are connected by using the
rivets. The splice plates are also connected to the web by the rivets. As such
rivet holes are made in the web. In the tension flange, net area is used. The net
area is found after making deductions for the rivet holes. It is assumed that 25
mm diameter holes are made for rivets at 100 mm spacing. Then the net area of
web is three-fourth of the gross-area.
Therefore, net area of tension flange
⎛ 3 1 ⎞ ⎛ 1 ⎞
Af net = ⎜ A´f + × Aw ⎟ = ⎜ A´f + Aw ⎟ ...(v)
⎝ 4 6 ⎠ ⎝ 8 ⎠
where, Af́ = Net area of flange excluding web.
1
From this, it is seen that approximately th web area acts as flange area if
6
1
the gross area is considered, and th web area (on account of reduction in
8
flange area for rivet holes) acts as flange area if the net area is considered. The
portion of web which acts as flange, is termed as web equivalent.
The effective sectional area of compression flange shall be the gross area
with deductions for excessive width of plates as it was specified for compression
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 481

members and open holes occurring in a plane perpendicular to the direction of

stress at the section being considered.
The effective sectional area of tension flanges shall be the gross-area with
the deduction for holes where the rivets are also used.

8.4 DEPTH OF PLATE GIRDER

The depth between outer surfaces of flanges is termed as over-all depth d0 or
1 1
depth of the plate girder. In general, the depth of plate girder is kept to of
10 12
span. The distance between C.G. of compression flange and C.G. of tension flange
is known as effective depth de of the plate girder. The distance between vertical
legs of flange angles at the top and at the bottom is known as clear depth d of
plate girder, in case horizontal stiffeners are not used. The overall depth, effective
depth and clear depth of plate girder are shown in Fig. 8.4. The depth of plate
girder is an essential factor for the design of plate girder. Sometimes, structural
C o m p re ssio n fla ng e

C .G .

C .G .

Ten sion flan g e

Fig. 8.4 Different depths of plate girder

requirements limits the depth of plate girder. When the depth of plate girder is
less than 750 mm, then, such girders are known as shallow plate girders.
The shallow plate girders resemble rolled steel beams. When the depth of plate
girder is more than 750 mm or when the depth is atleast eight times the depth
of vertical leg of the flange angles, then, such girders are known as deep plate
girders.

8.5 ECONOMICAL DEPTH OF PLATE GIRDER

The economical depth of the plate girder is the depth which gives the minimum
weight of plate girder. The economical depth of plate girder is found as below:
Let M = Maximum bending moment
de = Effective depth of plate girder
(Depth of web plate)
482 DESIGN OF STEEL STRUCTURES

tw = Thickness of web plate

σb = Allowable bending stress
l = Length of plate girder
ρ = Unit weight of steel
M
Area of flange Af =
( b ⋅ de )
σ
When two or more cover plates are used in plate girder in each flange, then
cover plates are curtailed. The depth of plate girder varies as the cover plates
are curtailed. The part of total weight of plate girder (comprising of flanges,
web plates, splice plates, stiffeners and filler plates) varies with the variation in
the depth of plate girder. As such, the average section of flange is approxi-
mately reduced to 80 per cent. The weight of stiffeners is approximately 60 per
cent of web plate.
The total weight of plate girder
W = Average weight of flanges + weight of web plate inclusive of
stiffener
⎡ M ⎤
W = ⎢2 × 0 ⋅ 80 × σ × d × l × ρ⎥ + [1 ⋅ 6 × tw × de × l × ρ] ...(i)
⎣ b e ⎦

⎡ M ⎤
W = ⎢1 ⋅ 60 × σ × d × l × ρ⎥ + [1 ⋅ 6 × tw × de × l × ρ] ...(ii)
⎣ b e ⎦
In case, the portion of web plate in resisting the bending moment is consid-
ered for, then, the total weight of plate girder is given by
⎡ M ⎤ ⎛ 1 ⎞
W = ⎢1 ⋅ 60 σ × d × l × ρ⎥ – ⎜⎝ 2 × 8 tw × de × l × ρ⎟⎠
⎣ b e ⎦
+ [1.6 × t w × de × l× ρ] ...(iii)
Differentiating the expression (iii) for W with respect to de
⎛ dW ⎞ ⎡ M ⎤
⎜ dd ⎟ = – ⎢1 ⋅ 60 2
× l × ρ⎥ + [1.6 × t w × l × ρ]
⎝ e⎠ ⎣ σ b × de ⎦
⎡1 ⎤
– ⎢ × tw × de × l × ρ⎥ ...(iv)
⎣4 ⎦
⎡⎛ dW ⎞ ⎤
For the economical depth, ⎢⎜ ⎟⎥ = 0
⎣⎝ d (de ) ⎠ ⎦
1/2
⎛ M ⎞
∴ de = 1.1 ⎜ ...(8.2)
⎝ σb × tw ⎟⎠
In case the portion of web plate in resisting the bending moment is not
considered, then the total weight of plate girder is given by the expression (i).
Differentiating the expression (ii), and equating to zero, the economical depth
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 483

⎛ M ⎞
of plate girder works out to be ⎜ . When the span of plate girder is
⎝ σb × tw ⎟⎠
small, then the flange plates may not be curtailed, and the section is uniform
throughout the span. For such plate girder, the economical depth works out to
1/2
⎛ M ⎞
1.21 ⎜ .
⎝ σb × tw ⎟⎠
The weight of plate girder varies 1 percent for 10 percent variation in the
depth of the girder. In practice, economical depth of plate girder is assumed 15
to 20 percent less than that computed.

8.6 SELF-WEIGHT OF PLATE GIRDER

From Sec. 8.5, the total weight of plate girder
⎡ M ⎤
W1 = ⎢1.60 × l × ρ + 1.60 tw × l × ρ⎥ ...(i)
⎣ σbde ⎦
The economical depth of plate girder, from Eq. 8.2
1/2
⎛ M ⎞
de = 1.1 ⎜
⎝ σb ⋅ tw ⎟⎠

⎛ M ⎞
∴ tw = 1.21 ⎜ ⎟ ...(ii)
⎝ σb ⋅ de2 ⎠
Therefore,
⎡⎛ M ⎞ ⎛ M ⎞
W 1 = ⎢⎜1.60 × l × ρ⎟ + ⎜1.60 × 1.21 × × de × l × ρ⎟
⎢⎣⎝ σb de ⎠ ⎝ σb ⋅ de2 ⎠

⎛1 M ⎞⎤
− ⎜ × 1.21 × 2
× de × l × ρ⎟ ⎥ ...(iii)
⎝4 σb ⋅ de ⎠ ⎥⎦

⎡⎛ M ⎞ ⎛ M ⎞⎤
W 1 = ⎢⎜1.60 d × l × ρ⎟ + ⎜1.35 × 1.21 × × l × ρ⎟ ⎥ ...(iv)
⎣⎝ σb e ⎠ ⎝ σb ⋅ de ⎠⎦

⎛ M ⋅ lρ ⎞
W 1 = 3.2 ⎜ σ d ⎟ ...(v)
⎝ ( b e )⎠
The expression (v) is for the weight of plate girder with gross-area of the
flanges without making deductions for rivet holes. The weight of plate girder is
about 10 to 20 percent greater than that given by the expression (v). The unit
weight of steel ρ is 78450 N/m3. Therefore, the weight of plate girder is given by
484 DESIGN OF STEEL STRUCTURES

⎛ 1 ⋅ 2 × 3 ⋅ 2 × 78450 × Ml ⎞
W1 = ⎜ × 100 × 10⎟
⎝ 100 × 100 × 100σb × de ⎠

⎛ 300 × Ml ⎞
or W1 = ⎜ ...(vi)
⎝ σb ⋅ de ⎟⎠
where, W1 = Weight of the plate girder in N
M = Bending moment in N-mm
l = Span of the plate girder in metre
σb = Bending stress in N/mm2
de = Effective depth of plate girder in mm.
The expression (vi) is useful when the bending moment M and the depth, de
are known. Therefore, this expression is used to check the self-weight of the
plate girder.
For all practical purpose, self-weight of girder is taken as
⎛W ⋅l⎞
W1 = ⎜ ...(8.3)
⎝ 300 ⎟⎠
where W = Total superimposed load
l = Span of plate girder in metres
The expression for self-weight of plate girder is approximate since it is assumed
that web plate resists shear only and bending moment is resisted by flanges.

8.7 IMPACT
The dynamic action of the moving load on a structure is called impact.
The impact is caused due to sudden application of load, unbalanced driving
wheel of locomotives, uneven track, wheel tyres worn out of truth, etc. The
stresses and strains (deformations) due to sudden application of load are twice
than those due to gradual application of load simultaneously and the actual
increase in deformation is generally less than twice that due to gradually applied
load because of damping effect of other construction. Therefore, the effect of
impact is considered less than 100 percent. The effect of impact depends upon
nature of moving load, speed of moving load, type of structure, loaded length of
structure etc. Provision for effect of impact is made by impact allowance or
impact load. The effect of impact is generally expressed as the percentage of the
moving load. Impact load is determined as a product of impact factor and live
(moving) load. The impact factor is specified by different authorities separately
for different type of moving loads and different type of structures. Indian Railway
Board has specified for steel girders for single track span, for metre gauge and
broad gauge impact factor as under.
⎛ 20 ⎞
Impact factor i = ⎜ |
> 1.00 ...(8.4)
⎝ 14 + L ⎟⎠
where, L is the loaded length of span in metres.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 485

8.8 ASSUMPTIONS FOR DESIGN FOR PLATE GIRDER

The plate girders are designed by trial and revised until the actual bending stresses
are less than or equal to the permissible bending stresses. The stiffeners are attached
as needed, until the shear stress is less than or equal to the allowable shear stress.
The web plate is designed on the basis of economical depth of the web. The initial
design of flange size is made by the flange area method and it is finally checked by
moment of inertia method. The moment carried by the flange is
Mf = (A f (
⋅ σb′ d0 − tf )) ...(i)
and the moment shared by the web is
⎛ σ ′′ t ⋅ d 2 ⎞
Mw = (σb′′ ⋅ Zw ) = ⎜ b w e ⎟
6
...(ii)
⎝ ⎠
The total moment resisted by the plate girder
M = (Mf + Mw) ...(iii)
However, to simplify, the following assumptions are made for the design of
the plate girder:
1. The web plate resists the shear force, and the shear stress is uniformly
distributed over whole cross-sectional area of web.
2. The flanges resist the bending moment. The distribution of bending
stress in the flanges in uniform.

8.9 ALLOWABLE BENDING STRESS

The maximum bending compressive stress calculated for the plate girders bend-
ing about the axis of maximum strength (xx-axis), σbc.cal shall not exceed the
maximum permissible bending compressive stress, σbc in N/mm2 (MPa) obtained
by the following formula
⎛ fcb⋅ fy ⎞
σbc = 0.66 ⎜ 1 ⎟
( )
n⎤ n
⎜⎜ ⎡( f )n + f ⎟⎟
⎝ ⎢⎣ cb y
⎦⎥ ⎠
where fcb = Elastic critical stress in bending
fy = Yield stress in the steel in N/mm2 (MPa); and
n = A factor assumed as 1.4.
The expression to calculate the elastic critical stress, fcb has been given in
Chapter 6 ‘Design of Beam’.
The values of σbc as derived from the above formula for some of the Indian
Standard structural steels have also been given in Table 6.2 in Chapter 6, ‘Design
of Beam’. Same values as given in Table 6.2 are used for the plate girders.

8.10 ALLOWABLE SHEAR STRESS

The allowable average shear stress for unstiffened web of plate girder calculated
on the cross-section of the web τva.cal shall not exceed, the permissible
average shear stress as recommended by IS: 800–1984, τva as below :
486 DESIGN OF STEEL STRUCTURES

τva = (0.4 fy)

where, fy is the yield stress for the structural steel to be used.
The web plate of plate girder may be classified as (i) unstiffened (ii) stiffened
for shear, and (iii) stiffened for shear and bending. The unstiffened web plate
must be sufficiently thick to resist the shear with an adequate factor of safety
against shear buckling. The stiffened web plate is designed to resist the shear
with some margin of the respect to shear buckling. The longitudinal stiffeners
are used for web stiffened for both shear and bending. The longitudinal stiffen-
ers increase the buckling resistance of the web in the region of the compressive
bending stress.
For the stiffened webs, the average shear stress in a member calculated on
the cross-section of the web, τva.cal shall not exceed the permissible average shear
stress as recommended by IS: 800 –1984. The value of permissible average shear
stress τva are given in Tables 8.1 (a), (b) and (c) for appropriate values of yield,
stresses 250, 340 and 400 N/mm2 (MPa), respectively.
The values τva for stiffened webs for steel for which the yield stress is not
given in Table 8.1 (a), (b) and (c) shall be found by using the following formulae,
provided that the average stress t va shall not exceed 0.4 fy.
1. For the webs where the distance between the vertical stiffeners is less
than the clear depth d.

⎡ 1 c ⎤
⎢ fy 2 ⋅ ⎥
tw
τva = 0.4 fy ⎢1 ⋅ 3 – 2
⎥ N/mm2 ...(8.5)
⎢ ⎪⎧ 1 ⎛ c ⎞ ⎪⎫ ⎥
⎢ 4000 ⎨1 + ⎜ ⎟ ⎬ ⎥
⎝ ⎠
⎢⎣ ⎩⎪ 2 d ⎭⎪ ⎥⎦

2. For the webs where the distance between the stiffeners is more than, d.

⎡ 1 c ⎤
⎢ fy 2 ⋅ ⎥
tw
τva = 0.4 fy ⎢1 ⋅ 3 – ⎥ N/mm2 ...(8.6)
⎢ ⎧ 1 ⎛ c ⎞⎫⎥
⎢ 4000 ⎨1 + ⎜ ⎟ ⎬ ⎥
⎢⎣ ⎩ 2 ⎝ d ⎠ ⎭ ⎥⎦

where τva = maximum permissible average shear stress

c = distance between the vertical stiffeners
tw = thickness of web.
For the vertically stiffened webs without horizontal stiffeners
d is the clear distance between flange angles, or where there are no flange
angles, the clear distance between the flanges ignoring the fillets.
For the vertically stiffened webs with the horizontal stiffeners
d is the clear distance between the tension flange (angles flange plates) and
the horizontal stiffeners.
Table 8.1 (a) Permissible average shear stress τva in stiffened webs of steel with σy = 250 N/mm2 (MPa)

⎛d⎞ Stress τva (MPa) for Different Distance c Between Stiffners

⎜⎝ t ⎟⎠
w 0.3d 0.4d 0.5d 0.6d 0.7d 0.8d 0.9d 1.0d 1.1d 1.2d 1.3d 1.4d 1.5d
90 100 100 100 100 100 100 100 100 100 100 100 100 100
95 100 100 100 100 100 100 100 100 100 100 100 100 99
100 100 100 100 100 100 100 100 100 100 100 99 99 98
105 100 100 100 100 100 100 100 100 100 99 98 97 96
110 100 100 100 100 100 100 100 100 99 98 96 95 94
115 100 100 100 100 100 100 100 100 98 96 95 94 93
120 100 100 100 100 100 100 100 98 96 95 93 92 91
125 100 100 100 100 100 100 98 97 95 93 92 91 90
Design

130 100 100 100 100 100 99 97 96 94 92 90 89 88

DESIGN

135 100 100 100 100 100 98 96 94 92 90 89 87 86

of OF

140 100 100 100 100 99 96 95 93 91 89 87 86 85

Beam

150 100 100 100 100 97 94 92 90 88 86 84 83 81

PLATE

160 100 100 100 98 94 92 89 88 85 83 81 80 78

170 100 100 100 96 92 89 87 85 82 80 78 76 75
GIRDER

180 100 100 98 94 90 87 84 82 80 77 75 73 72

Connection

190 100 100 97 92 88 84 82

200 100 100 95 90 86 82 _81_
210 100 99 93 88 83 81
220 100 98 91 86 81 _80_
230 100 96 90 84 79 Non-applicable zone.
240 100 95 88 83 77
250 100 93 86 82 _74_
260 100 92 85 81
270 99 90 84 81
Note : Intermediate values may be obtained by linear interpolation.
487
Table 8.1 (b) Permissible average shear stress τva in stiffened webs of steel with fy = 340 N/mm2 (MPa)
488
Stress τva (MPa) for Different Distance c Between Stiffners
(d/tw) 0.3d 0.4d 0.5d 0.6d 0.7d 0.8d 0.9d 1.0d 1.1d 1.2d 1.3d 1.4d 1.5d
75 136 136 136 136 136 136 136 136 136 136 136 136 136
80 136 136 136 136 136 136 136 136 136 136 136 136 136
85 136 136 136 136 136 136 136 136 136 136 136 134 133
90 136 136 136 136 136 136 136 136 136 135 133 132 131
95 136 136 136 136 136 136 136 136 135 133 131 129 128
100 136 136 136 136 136 136 136 135 132 130 128 127 126
105 136 136 136 136 136 136 135 133 130 128 126 124 123
110 136 136 136 136 136 135 133 131 128 126 124 122 120
115 136 136 136 136 136 133 131 129 126 123 121 119 118
120 136 136 136 136 135 131 129 127 124 121 119 117 115
125 136 136 136 136 133 129 127 125 121 119 116 114 113
130 136 136 136 135 131 127 125 122 119 116 114 112 110
135 136 136 136 134 129 126 123 120 117 114 109 108 108
140 136 136 136 132 127 124 121 118 115 112 109 107 105
150 136 136 135 129 124 120 117 114 110 107 104 102 100
160 136 136 132 126 120 116 113 110 106 102 99 97 95
170 136 136 129 123 117 112 109 106 101 98 95 92 90
DESIGN OF STEEL STRUCTURES

180 136 135 127 119 113 108 105 102 97 93 90 87 84

190 136 133 124 116 110 105 100
200 136 130 121 113 106 101 96
210 136 128 118 110 103 97
220 136 126 116 107 99 93 Non-applicable zone
230 135 123 113 103 96
240 134 121 110 100 92
250 132 119 107 97 89
260 130 116 104 94
270 128 114 102 91
Note. Intermediate values may be obtained by linear interpolation.
Table 8.1 (c) Permissible average shear stress τva in stiffened webs of steel with fy = 400 N/mm2 (MPa)

⎛d⎞ Stress τva (MPa) for Different Distance c Between Stiffners

⎜⎝ t ⎟⎠
w 0.3d 0.4d 0.5d 0.6d 0.7d 0.8d 0.9d 1.0d 1.1d 1.2d 1.3d 1.4d 1.5d
70 160 160 160 160 160 160 160 160 160 160 160 160 160
75 160 160 160 160 160 160 160 160 160 160 160 160 159
85 160 160 160 160 160 160 160 160 160 160 159 157 156
90 160 160 160 160 160 160 160 160 160 158 156 154 152
95 160 160 160 160 160 160 159 157 154 152 149 147 146
100 160 160 160 160 160 160 157 155 151 149 146 144 143
110 160 160 160 160 159 155 152 149 146 143 140 138 136
115 160 160 160 160 156 152 149 147 143 140 137 135 133
Design

120 160 160 160 159 154 150 147 144 140 137 134 132 129
DESIGN

125 160 160 160 157 152 147 144 141 137 134 131 128 126
of OF

130 160 160 160 155 150 145 141 139 134 131 128 125 123
135 160 160 160 153 147 143 139 136 132 128 125 122 120
Beam
PLATE

140 160 160 158 151 145 140 136 133 129 125 122 119 116
150 160 160 155 147 141 135 131 128 123 119 115 112 110
160 160 160 151 143 136 130 126 123 117 113 109 106 103
GIRDER

170 160 158 148 139 132 126 121 117 112 107 103 100 97
Connection

180 160 155 144 135 127 121 116 112 106 101 97 93 90
190 160 152 140 131 123 116 111
200 160 149 137 127 118 111 106
210 160 146 133 123 114 106
220 157 143 130 119 109 101
230 155 140 126 114 105 Non-applicable zone.
240 153 137 123 110 100
250 151 134 119 106 96
260 148 131 116 102
270 146 128 112 98
Note. Intermediate values may be obtained by linear interpolation.
489
490 DESIGN OF STEEL STRUCTURES

It is to note that the allowable shear stresses given in Table 8.1 (a), (b) and (c)
apply provided any reduction in the web cross-section is due to only rivet holes.
Where the large apertures are cut in the web, a special analysis is to be carried
out to ensure the maximum permissible average shear stresses laid down in IS:
800 –1984 are not exceeded.
The maximum shear stress having regard to the distribution of stresses in
confirmity with the elastic behaviour in bending, shall not exceed the value τvm,
found as under
τvm = (0.45 fy)
where, fy is the yield stress of the structural steel to be used.

8.11 DESIGN OF WEB PLATE

In the design of web plate of plate girder, the economical depth and thickness of
web plate are determined. Approximately, the economical depth of web plate is
given by the following formula:
Economical depth of web plate
1
⎛ M ⎞ 2
= 1.1 ⎜ ⎟
⎝ σbc × tw ⎠
where, M = Maximum bending moment
σbc = Allowable bending stress
tw = Thickness of web plate.
For the economy, the depth of web plate selected should be such that web
plates are manufactured and are available for that depth. The depth of plate
1 1
girder varies from th and th the span. The thickness of web plate should
10 12
be such that maximum shear stress in the web does not exceed maximum
allowable shear stress and average shear stress in the web does not exceed
average allowable shear stress. The thickness of web plate should also provide
necessary bearing area for rivets connecting web plate and flange angles, so
that rivets are provided at proper spacing. A minimum thickness of 6 mm is
adopted to provide for corrosion. The thickness of web plate is fixed keeping in
view the buckling of web in shear, and shear along with bending.
The economical depth of web plate may be approximately determined by the
use of Rawter and Clark formula also.
Economical depth of web plate
1
⎡ M ⎤2
= K. ⎢ ⎥
⎣ óbc ⎦
where, K = A parameter, which is adopted as 4.5 for the riveted plate girder
M = Maximum bending moment in N-mm
σbc = Allowable bending stress in N/mm2 (MPa).
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 491

8.11.1 Web Buckling

The web plate of simply supported plate girder may be considered as subjected
to pure shear close to the support, to pure bending at the centre, and to com-
bined shear and bending at the intermediate positions, when subjected to uni-
formly distributed load. The web plates used are deep and thin. The web plate
may fail by buckling.
Buckling is defined as sudden bending, warping, curling or crumpling of
the elements under compressive stresses.

a
Fig. 8.5 Uniformly distributed shear along edges

8.11.1.1 Web Buckling in Pure Shear

When the web plate is subjected to pure shear uniformly distributed along the
edge as shown in Fig. 8.5, then, the critical shear stress is given by the expres-
sion given below :

⎛ π2 ⋅ E ⎞
τc = ⎜ 2
⋅ Ks ⎟ ...(8.7)
⎜ 2 ⎛ b ⎞ ⎟
⎜ 12 (1 – µ ) ⎜ t ⎟ ⎟
⎝ ⎝ w⎠ ⎠
where µ = Poisson’s ratio
Ks = Constant, for the plate subjected to pure shear
⎛a⎞
⎜ ⎟ = Aspect ratio
⎝b⎠
a = Length of the plate
b = Width of the plate
tw = Thickness of the plate

⎛ a⎞
The value of constant, Ks depends upon aspect ratio ⎜ ⎟ and the support
⎝ b⎠
conditions for the edges of plate. The concept of deep plate girders was developed
by considering the importance of this aspects ratio.
Plate simply supported along all four edges
The values of constant Ks are given by the following expressions:
492 DESIGN OF STEEL STRUCTURES

⎡ 4 ⋅ 00 ⎤ ⎛ ⎛ a⎞ ⎞
Ks = ⎢5 ⋅ 34 + 2 ⎥⎜
for ⎜ ⎟ > 1⎟ ...(i)
⎛ a⎞ ⎥ ⎝ ⎝ b ⎠ ⎠
⎢ ⎜⎝ ⎟⎠ ⎥
⎢⎣ b ⎦

⎡ 5 ⋅ 34 ⎤ ⎛ ⎛ a⎞ ⎞
and Ks = ⎢4 ⋅ 00 + 2 ⎥⎝⎜ for ⎜⎝ ⎟⎠ < 1⎠⎟ ...(ii)
⎛ a⎞ ⎥ b
⎢ ⎜⎝ ⎟⎠ ⎥
⎢⎣ b ⎦
In the web plate for the plate girder, the dimension a may always be selected
⎛a⎞
as the larger dimension. Therefore, the aspect ratio ⎜ ⎟ will be greater than
⎝b⎠
⎛a⎞ ⎛a⎞
unity. From the expression (i), for ⎜ ⎟ = ∞, the value of Ks is 5.34. For ⎜ ⎟ = 1,
⎝b⎠ ⎝b⎠
the value of Ks is 9.34. For different values of the aspect ratio, the values of Ks
can be found from the expression (i). The values of Ks for some values of the
aspect ratio have been given in Table 8.2
Table 8.2

⎛a⎞
⎜ ⎟ 1.0 1.2 1.4 1.6 1.9 2.0 3.0 5.0 ∞
⎝b⎠
KS 9.34 8.12 7.38 6.90 6.58 634 5.78 5.36 5.34
As the values of aspect ratio increase from 5.0 to ∞ the change in value of is
Ks is insignificant. For unstiffened web, value of Ks can be assumed equal to
5.34 for all practical purposes. For structural steel value of Poisson’s ratio is 0.3
and Young’s modulus of elasticity E is 2.11 × 105 N/mm2. Substituting these
values in above expression Eq. (8.7) reduces to
⎛ 10 ⋅ 05 × 105
⎞
σc = ⎜ ⎟ N/mm2
2
...(8.8)
⎜ ⎛b ⎞ ⎟
⎜ ⎜t ⎟ ⎟
⎝ ⎝ w⎠ ⎠
The value of width of web plate may be substituted equal to clear depth of
web viz. d,
⎛ 10 ⋅ 05 × 105
⎞
∴ σc = ⎜ 2⎟ N/mm2 ...(8.9)
⎜ ⎛d⎞ ⎟
⎜ ⎜t ⎟ ⎟
⎝ ⎝ w⎠ ⎠
It is seen that the critical shear stress in the web is inversely proportional
d
ratio . When the value of critical shear stress reaches the value of yield in
tw
shear, viz. 150 N/mm2, then
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 493

2
⎛d⎞ ⎛ 10 ⋅ 05 × 105 ⎞
⎜t ⎟ = ⎜ ⎟⎠
⎝ w⎠ ⎝ 150
2
⎛d⎞ ⎛d ⎞
⎜t ⎟ = 6700, ⎜ = 82 ⎟
⎝ w⎠ t
⎝ w ⎠
d
When the ratio is less than 82, the failure of web plate takes place by
tw
d
yielding. When the ratio of is greater than 82, the buckling of web takes
tw
d
place. Therefore the allowable shear stress is decreased. As ratio increases,
tw
value of critical shear stress decreases rapidly. The value of critical shear stress
may be increased by decreasing the aspect ratio. When the aspect ratio is
decreased, the value of constant Ks increases. The aspect ratio may be decreased
reducing the length a, by providing transverse stiffeners on the web plate, as
shown in Fig. 8.6. The value of constant Ks increases as the length a (i.e., the
spacing between the transverse stiffeners, c) reduces. The value of critical stress
increases. When the spacing between the transverse stiffeners c, becomes less
than the clear depth of web, d, then, value of critical stress further increases
since, b is the smaller dimensions in Eq. 8.7, and it is universely proportional to
⎛b⎞
square of ⎜ ⎟ ratio. Sub-dividing a simply supported plate of length, a, by
⎝ tw ⎠
sufficiently rigid transverse stiffeners of spacing c, is shown in Fig. 8.6, smaller
panels are formed. The smaller panels may be considered approximately as simply
supported along the four edges. The buckling strength of web plate is increased,
and economical design of web may be done.

Fig. 8.6 Panel dimensions of plate girder

8.11.1.2 Web Buckling in Pure Bending

When the web plate is subjected to pure bending, then, the critical bending
stress is given by

⎛ π2 ⋅ E ⎞
σc = ⎜ ⋅ Kb ⎟ ...(8.11)
⎜ 12 (1 – µ2 ) ⎛ b ⎞ ⎟
⎜ ⎜t ⎟ ⎟
⎝ ⎝ w ⎠ ⎠
where, K b = constant for the plate subjected to pure bending.
494 DESIGN OF STEEL STRUCTURES

For the plates simply supported on all the four edges, the minimum value
⎛a⎞ ⎛2⎞
constant Kb is 23.9 as integer multiples of ⎜ ⎟ = ⎜ ⎟ . It is the common practice
⎝b⎠ ⎝3⎠
to use this minimum value of Kb for plates with ab > 0.67. For the plates simply
supported on loaded edges and clamped on the other edges, the minimum value
⎛a⎞
of Kb = 3.96 as integer multiples of ⎜ ⎟ = 0.47.
⎝b⎠

8.11.1.3 Web Buckling in Combined Shear and Bending

The web plate of simply supported plate girder is subjected to combined shear
and bending at the positions in between the support and the centre. The plate is
subjected to non-uniform longitudinal bending stress in compression above
neutral axis, along with shear. The bending stress in compression causes the
local buckling of the web plate above the neutral axis. The value of constant K is
given in Table 8.3 for simply supported plates and for plate fixed or clamped at
the longitudinal edges.
Table 8.3

⎛ a ⎞ ratio 0.4 0.47 0.5 0.6 0.667 0.7 0.8 1.0 1.5
⎜ ⎟
⎝b⎠
K simple 29.1 — 25.6 24.1 23.9 — 24.4 25.6 24.1
K fixed 42.7 39.6 39.7 41.8 — 45.8 — — —
For simply suppbrted edges, the minimum value of Kb is 23.9. The critical
stress for bending is given by

⎡ 45 ⋅ 4 × 105 ⎤
σc = ⎢ 5 ⎥ N/mm2 ...(8.12)
⎢ ⎛b⎞ ⎥
⎢ ⎜t ⎟ ⎥
⎣ ⎝ w⎠ ⎦
For the plate clamped at the longitudinal edges, the critical stress for bend-
ing is given by

⎡ 75 ⋅ 2 × 105 ⎤
σc = ⎢ 2 ⎥ ...(8.13)
⎢ ⎛b⎞ ⎥
⎢ ⎜t ⎟ ⎥
⎣ ⎝ w⎠ ⎦
The critical stress for combined bending and shear lies in between these two
values. For combined bending and shear, the web plate may be checked by
calculating the equivalent stress, σe.cal due to co-existent bending (tension or
compression) and the shear stresses obtained as below. σe.cal shall not exceed
the value, σe = 0.9 fy (the maximum permissible equivalent stress):
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 495

1
2 2
σe.cal = ⎡⎣σbt⋅cal + 3τvm⋅cal ⎤⎦ 2 ...(8.14 a)
1
2 2
σe.cal = ⎡⎣σbc⋅cal + 3τvm⋅cal ⎤⎦ 2 ...(8.14 b)
The theoretical values of critical stress for the buckling of web have been
determined by small-deflection buckling theory making the following
assumptions:
(i) The thickness of web plate is small compared with its surface
dimensions.
(ii) The plate is perfectly straight before loading.
(iii) The plate deflections are small compared with the plate thickness.
The small-deflection buckling theory provides a safe but often over conserva-
tive design. When the web plate is adequately supported then, it will carry
stresses substantially exceeding theoretical buckling stresses.
1. The thickness of the unstiffened web plate shall be not be less
than the following:

⎛ 1⎞

(i) t w . min = ⎜
d1 ⋅ ( τ )
va⋅cal 2 ⎟
and
⎝ 816 ⎠

⎛ 1 ⎞
⎜ d1 ⋅ ( fy )2 ⎟
(ii) t w . min = ⎜ ⎟
⎝ 1344 ⎠
whichever is more, but
⎛d ⎞
(iii) | ⎜ 1⎟
t w . min <
⎝ 85 ⎠
where, d1 is the depth of the web, and τva.cal is the calculated average shear
stress in the web due to shear force.
2. The thickness of the vertically stiffened web plate shall be not
be less than the following
⎛ 1 ⎞
(i) ⎜ ⎟ of the smaller clear panel dimension and
⎝ 180 ⎠

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟
(ii) ⎜ ⎟ whichever is more in (i) and (ii)
⎝ 3200 ⎠

⎛ d ⎞
(iii) ⎜ tw⋅min < 2 ⎟
⎝ 200 ⎠
3. The thickness of webs stiffened both vertically and horizontally with a
horizontal stiffener at a distance from the compression flange equal to 2/5 of the
distance from the compression flange to the neutral axis.
496 DESIGN OF STEEL STRUCTURES

⎛ 1 ⎞
(i) ⎜ ⎟ of the smaller dimension in each panel and
⎝ 180 ⎠

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟
(ii) ⎜ ⎟ whichever is more in (i) and (ii)
⎝ 4000 ⎠

⎛ d ⎞
(iii) ⎜ tw⋅min < 2 ⎟
⎝ 250 ⎠
4. The thickness of webs stiffened additionally with a horizontal
stiffener at the neutral axis.
⎛ 1 ⎞
(i) ⎜ ⎟ of the smaller dimension in each panel, and
⎝ 180 ⎠

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟
(ii) ⎜ ⎟ whichever is more in (i) and (ii)
⎝ 6400 ⎠

⎛ d ⎞
(iii) ⎜ tw⋅min < 2 ⎟
⎝ 400 ⎠
where, d2 in the above expressions is twice the clear distance from the compres-
sion flange angles or plate to the neutral axis.
The minimum thickness of web plates for the different values of yield stress
are given in Table 8.4 (i) and (ii) as specified by IS: 800–1984.
Table 8.4 (i) Minimum thickness of web plate

Minimum thickness of web for yield stress fy N/mm2 (in MPa) of

fy→ 220 230 240 250 260 280 300 320 340

⎛ 1 ⎞
⎜ d1 ⋅ ( fy )2 ⎟ d1 d1 d1 d1 d1 d1 d1 d1 d1
⎜ ⎟ 85 85 85 85 83 80 78 75 73
⎝ 1344 ⎠

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟ d2 d2 d2 d2 d2 d2 d2 d2 d2
⎜ ⎟ 200 200 200 200 198 191 185 179 174
⎝ 3200 ⎠

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟ d2 d2 d2 d2 d2 d2 d2 d2 d2
⎜ ⎟ 250 250 250 250 248 239 231 217
⎝ 4000 ⎠ 224

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟ d2 d2 d2 d2 d2 d2 d2 d2 d2
⎜ ⎟ 400 400 400 400 396 382 370 358 348
⎝ 6400 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 497

Table 8.4 (ii) Minimum thickness of web plate

Minimum thickness of web for yield stress fy N/mm2 (in MPa) of

fy → 340 360 380 400 420 450 480 510 540

⎛ 1 ⎞
⎜ d1 ⋅ ( fy )2 ⎟ d1 d1 d1 d1 d1 d1 d1 d1 d1
⎜ ⎟ 73 71 69 67 66 63 61 60 58
⎝ 1344 ⎠

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟ d2 d2 d2 d2 d2 d2 d2 d2 d2
⎜ ⎟ 174 169 164 160 156 151 146 138
⎝ 3200 ⎠ 142

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟ d2 d2 d2 d2 d2 d2 d2 d2 d2
⎜ ⎟ 217 205 200 195 189 183 177 172
⎝ 4000 ⎠ 211

⎛ 1 ⎞
⎜ d2 ⋅ ( fy )2 ⎟ d2 d2 d2 d2 d2 d2 d2 d2 d2
⎜ ⎟ 348 338 328 320 312 302 292 284 276
⎝ 6400 ⎠
The effective cross-sectional area of the web plate in the plate girder shall be
calculated as the product of the full depth of the web plate and the thickness of
web plate.
For the riveted plate girder, in the exposed situation and which do not have
flange plates for their entire length, the top edge of the web plate shall be flush
with the angles and the bottom edge of the web plate shall also be flush with the
flange angles.
The thickness of web plate selected should be such that web plates are manu-
factured and are available for that thickness. For the purpose of economy, the
thickness of web is kept minimum practicable and stiffeners are provided if
required.

8.12 DESIGN OF FLANGES

In the riveted plate girder, the flange section consists of flange plates, flange
angles and that part of the web which is between the flange angles. The effec-
tive sectional area of compression flange is the gross area with deductions for
excessive width of plates and for open holes, occurring in a plane perpendicular
to the direction of stress at the section. The effective area of tension flange is the
gross area with deduction of holes.
In practice, the gross area of compression flange and tension flange are kept
equal. The tension flange of the plate girder is designed on the net area basis
and the corresponding gross area is provided for compression flange. After
selecting thickness and depth of web plate, flanges are designed by the follow-
ing methods:
1. Flange area method; 2. Moment of inertia method.
498 DESIGN OF STEEL STRUCTURES

8.12.1 Flange Area Method

The trial section of plate girder is determined by flange area method. The flange
area method is also termed as approximate method. The bending stress in
compression flange and in tension flange are assumed uniformly distributed as
shown in Fig. 8.7, apart from triangular stress distribution. This assumption is
satisfactory, so long as the depth of the flange is small in compression to overall
depth of the plate girder.
The forces in the compression flange C and in the tension flange T are equal
and act at the C.G., of flanges in opposite direction. They form a couple which
resists the bending moment.
M = (C × de ) ...(i)
M = (T × de ) ...(ii)
M = (A f́ × σb́ × de ) ...(iii)
∴ A f́ = (M /(σb × de )) ...(8.15 a)
where, M = Maximum bending moment
σb́ = Average bending stress
de = Effective depth of plate girder
A f́ = Net area of tension flange.
C le aran ce

N A

Fig. 8.7 Distribution of bending stresses over flanges

In the design of net area of tension flange, a large part of the area of the
flange is provided by the flange angles and the number of flange plates is kept
to a minimum. Preferably not less than one-third net area of the tension flange
is provided by angles. One-eighth area of web plate acts as flange area. The
remaining flange area is provided by flange plate. It is also possible that the
flange of a plate girder consists of only flange angles and web equivalent and
flange plates. The number of flange plates in each flange should not be more
than three. When more than three plates are used, then, the length of rivets
connecting flange plates and flange angles becomes large. The large abjected to
bending in addition to shearing and bearing. As such the rivet grip becomes less
effective. The net area of flange provided should be little more than that required.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 499

Corresponding gross area to the net area adopted for compression flange. The
unequal angle sections are used for flange angles, and the long legs are kept
horizontal. The outstand of flange plate, i.e., their projection beyond the outer
line of connections to flange angle is provided as recommended for the
compression to flange angles.
In order to transmit load through flanges and not by direct bearing through
web plate clearance of 5 mm is kept between cover plates and web plate as
shown in Fig. 8.7.

8.12.2 Moment of Inertia Method

The moment of inertia method of design of plate girder section is BIS method of
design of plate girder section. This is also known as exact method. A trial section
of plate girder, designed by flange area method is checked by moment of inertia
method. The moment of resistance of trial section is computed. If it is equal to or
greater than the maximum bending moment,then the trial section designed is
satisfactory.
Moment of resistance,

⎛ Igross ⎞
M.R. = ób × ⎜ ...(i)
⎝ ymax ⎟⎠

⎛d⎞
where, ymax = Distance of extreme fibre from neutral axis ⎜ ⎟ .
⎝2⎠
The neutral axis of the plate girder is assumed at the C.G. of gross area of
plate girder. However, the net moment of inertia of plate girder may be calcu-
lated and used to determine the bending stress. The gross moment of inertia of
the plate girder section is found as follows :

⎡ 2 ⎤
⎛d⎞ 1
Igross = ⎢2 I f + 2 Af 1 ⋅ ⎜ ⎟ + tw d3 ⎥ ...(ii)
⎣ ⎝ 2 ⎠ 12 ⎦
where, Af1 = Flange area excluding web equivalent.
It is to note that the difference between the overall depth, the effective depth
and the depth of the web is very small. Therefore, the depth of web and the
effective depth have been assumed to be equal to the depth of plate girder sec-
tion. The moment of inertia of flange plates about their own axes, If is very
small. Hence, If is neglected. Therefore, the gross moment of inertia is given by

⎡1 1 ⎤
Igross = ⎢ Af 1 ⋅ d2 + tw d3 ⎥ ...(iii)
⎣2 12 ⎦

2
⎛ 1 ⎞d
or Igross = ⎜ A f 1 + Aw ⎟ ...(iv)
⎝ 6 ⎠ 2
500 DESIGN OF STEEL STRUCTURES

The moment of resistance is found by substituting this in the expression (i)

σbc ⎛ 1 ⎞d
∴ M.R. = ⎜ A f 1 + Aw ⎟ ...(v)
d
⎛ ⎞⎝ 6 ⎠2
⎜ ⎟
⎝2⎠

⎛ 1 ⎞
M.R. = σbc . ⎜ Af 1 + Aw ⎟ . d ...(vi)
⎝ 6 ⎠
When the moment of resistance is equal to or greater than the maximum
bending moment, then, the trial section designed is satisfactory. The moment of
resistance of the plate girder section may also be written as

M.R. = σb1 ⎛ A + 1 A ⎞ . d ...(vii)

⎜ f1 w⎟
⎝ 8 ⎠
In case, σb́ and σb´´ in the expression (i) and (ii) are taken equal and (d0 – tf) is
adopted as de , then, the bending moment resisted by flanges, then
Mf = (M – Mw)
Substituting the values of these moments.
The compression and the tension flanges are checked for bending stresses in
extreme fibre.
Equating (vi) and (vii)
⎛ 1 ⎞ ⎛ 1 ⎞
σbt . ⎜ A f 1 + Aw ⎟ . d = σbc . ⎜ Af 1 + Aw ⎟ . d
⎝ 8 ⎠ ⎝ 6 ⎠
σbt . Anet = σbc . Agross
⎛ Agross ⎞
σbt = σbc × ⎜
⎝ Anet ⎟⎠

⎛ 1 ⎞
then M.R. = σb́ ⎜ Af + tw ⋅ dw ⎟ ⋅ de ...(viii)
⎝ 6 ⎠

⎛ 1 ⎞
= σb́ ⎜ Af + Aw ⎟ ⋅ de
⎝ 6 ⎠

⎛ t ⋅ d2 ⎞
(Af × σb́ × de ) = ⎜ M – σ′b w e ⎟ ...(ix)
⎝ 6 ⎠
The bending stress in the extreme fibre in compression flange is given by

⎛ M ⎞
σbc.cal = × ymax ...(8.15)
⎜⎝ I gross ⎟⎠

The bending stress in the extreme fibre in tension flange is computed by

increasing σbc in proportion of gross area and net area of flanges.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 501

The bending stress in the extreme fibre in tension flange

⎛ Gross area of flange ⎞
σbt.cal = σbc.cal × ⎜ ⎟ ...(8.16)
⎝ Net area flange ⎠
σbt should be less than allowable bending stress.
Example 8.1 A plate girder simply supported at ends is composed of web
plate 1000 mm depth × 12 mm thickness, and two flange angles ISA 200 mm ×
100 mm × 15 mm (ISA 200100, @ 0.336 kN/m) and two flange plates 500 mm
wide × 20 mm thickness in each flange. The effective span of the plate girder is
12 m. The diameter of rivets used for connecting flange plates to flange angles
and flange angles to web plate is 22 mm. Determine the maximum uniformly
distributed load (inclusive of self weight) which can be carried by the plate girder.
Assume maximum allowable stress as per IS : 800 –1984.
Solution
Step 1 : Moment of inertia of given plate girder section
The gross-moment of inertia of the plate girder Ixx is found about xx-axis. xx-
axis is assumed to be located at the centre. The moment of inertia of the flange
plates about their own axis is neglected.
Ixx (gross) = Moment of inertia of web plate + moment of inertia of flange
angles about their own axis + Moment of inertia of flange angles
about xx-axis + Moment of inertia of flange about xx-axis.

5 00 m m
20 m m
20 m m

22 m m
D ia m eter rivets 2 15 A
2 00 m m ×
1 000 m m

1 00 × 15 m m

2 15 A
2 00 m m ×
1 00 × 15 m m

20 m m
20 m m

Fig. 8.8

From ISA Handbook No. 1

502 DESIGN OF STEEL STRUCTURES

Moment of inertia of one flange about its own axis

= (298.1 × 104) mm4
Area = 4278 mm2
Distance of its C.G. is 22.2 mm
Distance between C.G. and xx-axis = (500 – 22.2) = 477.8 mm

⎡1 ⎤
Ixx (gross) = ⎢ × 1 ⋅ 2 × 1003 + 4 × 298.1 + 4 × 42.78 × 47.782 + 2 × 50 × 4 × 522 ⎥
⎣12 ⎦
× 104 mm4
= 1573992 × 104 mm4
Net flange area = 26776 mm2
Gross flange area = 30556 mm2
Overall depth of the plate girder
D = (1000 + 40 + 40) = 1080 mm
Mean thickness of the compression flange
T = (20 + 20 + 15) = 55 mm
⎛D⎞ ⎛ 1080 ⎞
Ratio ⎜ ⎟ = ⎜ ⎟ = 19.636
⎝T ⎠ ⎝ 55 ⎠
Effective length of compression flange
l = (l2 × l000) = 12000 mm.
Moment of inertia of the plate girder about its axis of minimum strength (yy-
axis)

⎡1 1 ⎤
Iyy = ⎢ × 100 × 1 ⋅ 23 + 4 × × 2 × 503 + 4 × 42 ⋅ 78 × 7 ⋅ 182 ⎥ × 104 mm4
⎣12 12 ⎦
= 99172.377 × 104 mm4
1
⎡ 99172 ⋅ 377 × 104 ⎤ 2
ry = ⎢ ⎥ = 180.155 mm
⎣ 30556 ⎦
Step 2 : Slenderness ratio
⎛ l ⎞ ⎛ 12000 ⎞ ⎛D⎞
Ratio, ⎜ ⎟ = ⎜ ⎟ = 66.61, ⎜ ⎟ = 19.636
⎝ ry ⎠ ⎝ 180 ⋅ 155 ⎠ ⎝T ⎠
From IS: 800–1984, Table 6.5
X = 186.727 and Y = 81.86
Step 3 : Elastic critical bending stress
From IS : 800 –1984
c
fcb = ⎛ k1 (X + k2 .Y ) 2 ⎞
⎜⎝ c1 ⎟⎠
k1 = 1 00 for ϕ = 1.0 and k 2 = 0.0 for ω = 0.5
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 503

c2
and = 1.00 (N.A. assumed at centre)
c1
fcb = 1.00 (186.726 + 0.0 × 81.86) × 1
= 186.726 N/mm2
Step 4 : Allowable bending stress. From IS : 800 –1984, Table 6.2 for the
value of yield stress for the structural steel as 250 N/mm2.
Bending stress in compression is 86.02 N/mm2.
Flange Area
Description Gross Area Deduction for Net area (mm)2
(mm)2 rivet hole (mm)2
Flange angles
2 ISA 200 mm 2 × 4278 4 × 235 × 15 7146
× 100 mm × 15 mm = 8556 = 1410
Flange Plates 2 × 500 × 20 4 × 235 × 20
2 × 500 mm × 20 mm = 20000 = 1880 18120
1 1 1 1
Web equivalent Aw = × 1000 — Aw = × 1000
6 6 6 6
× 12 = 2000 × 12 = 1500
Total 30556 — 26766
From theory of bending

⎛M σbc ⎞
⎜I = y ⎟
⎝ xx ⎠
Step 5 : Moment of Resistance of plate girder section

⎛ σ × I xx ⎞ ⎛ 86 × 15373992 ⋅ 4 × 104 ⎞
M = ⎜ bc ⎟= ⎜ ⎟
⎝ y ⎠ ⎝ 540 ⎠
= 2507.25 kN-m
Step 6 : Load supported by plate girder
Let w be the uniformly distributed load inclusive of self weight of plate girder
per unit length

⎛ w × l2 ⎞
Maximum bending moment = ⎜ ⎟
⎝ 8 ⎠
Equating bending moment to moment of resistance of plate girder
⎛ w × 12 × 12 ⎞
⎜ ⎟ = 2507.25,
⎝ 8 ⎠
∴ w = 139.29 kN/m
The plate girder can carry 139.29 kN/m load inclusive of self weight.
504 DESIGN OF STEEL STRUCTURES

Example 8.2 A mild steel plate girder simply supported at two ends has an
effective span of 20 m. It carries a dead load of 50.00 kN/m and uniformly
distributed live load of the same intensity, longer than the span. Design the
maximum section of the plate girder. Allow for impact.
Solution
Design :
Step 1: Impact load
The effective span of the plate girder is 20 m
⎛ 20 ⎞
Impact factor i = ⎜ ⎟
⎝ 14 + L ⎠
where L = Effective span in metres
⎛ 20 ⎞
i = ⎜ ⎟ = 0.588
⎝ 14 + 20 ⎠
Intensity of live load = 50.00 kN/m
Impact load = (0.588 × 50.00) = 29.40 kN/m
Step 2: Dead load = 50.00 kN/m
Superimposed load over the plate girder
W = 129.40 kN/m
Self-weight of the plate girder
⎛ WL ⎞
W1 = ⎜ ⎟ (W 1 = Self-weight of the plate girder)
⎝ 300 ⎠
where, W = Superimposed load over the plate girder
L = Effective span of plate girder
⎛ 129.40 × 20 ⎞
W1 = ⎜ ⎟⎠ × 20 = 175.52 kN.
⎝ 300
Self-wcight of the plate girder per unit length is 8.626 kN/m
Step 3: Total uniformly distributed load
(50.00 + 50.00 + 29.40 + 8.626) = 138.026 kN/m
(Say = 138.10 kN/m)
Step 4: Maximum bending moment
⎛ wL2 ⎞ ⎛ 138 ⋅ 10 × 20 × 20 ⎞
M = ⎜ =
⎝ 8 ⎟⎠ ⎜⎝ 8
⎟
⎠
= 6905 kN-m
⎛ wL ⎞ ⎛ 138 ⋅ 10 × 20 ⎞
Maximum S.F., F = ⎜ = ⎟ = 1381 kN
⎝ 2 ⎟⎠ ⎜⎝ 2 ⎠
Step 5: Design of web plate
Let the value of yield stress for the structural steel to be used be 250 N/mm2.
The maximum permissible stress in compression flange may be assumed as
σbc = 0.66 × 250 = 165 N/mm2
Assume thickness of web plate, t w= 12 mm
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 505

Economical depth of the plate girder

1 1
⎛ M ⎞2 ⎛ 6905000 × 1000 ⎞ 2
d = 1.1 ⎜ σ × t ⎟ = 1.1 ⎜ ⎟ = 1867.45 mm
⎝ bc w ⎠ ⎝ 165 × 12 ⎠
Adopt depth of the web plate as 2000 mm
Step 6 : Design of flanges (Flange area method)
Net flanges area required for tension flange
⎛ M ⎞ ⎛ 6905000 × 1000 ⎞
Af.net = ⎜ ⎟ = ⎜⎝ 165 × 2000 ⎟⎠ = 20924.24 mm
2
σ
⎝ bc × de ⎠
Assume 22 mm diameter rivets for connecting flange plates to flange angles
and flange angles to web plate.
Net flange area provided in tension flange = 24642 mm2
Gross flange area provided in compression flange
= 28556 mm2
Flange Area
Description Gross Area Deduction for Net area (mm)2
(mm)2 rivet hole (mm)2
Flange angles
2 ISA 200 mm 2 × 4278 4 × 23.5 × 15 7146
× 100 mm × 15 mm = 8556 =1410
Flange Plates 2 × 500 × 16 4 × 23.5 × 16
2 × 500 mm × 16 mm = 16000 = 1504 14496
1 2000 × 12 1 2000 × 12
Equivalent web area Aw = — Aw =
6 6 6 6
= 4000 = 3000
Total 28556 — 24642
The maximum section of the plate girder is shown in Fig. 8.9.
The maximum section selected by the flange area method is checked by
moment of inertia method. The neutral axis (xx-axis) of the plate girder is
assumed to be located at the centre. The moment of inertia of plates about their
own axis is neglected.
Step 7 : To check the designed section by (moment of inertia method)
From ISI Handbook No.1
The moment of inertia of the flange angle about its own axis
= 298.1 × 104 mm4
2
Area A = 42.78 × 100 mm . Distance of C.G. is 22.2 mm
⎡1 2
Ixx(gross) = ⎢ × 1.2 × 2003 + 4 × 298.1 + 4 × 42.78 × (100 – 2.22)
⎣12

+ 2 × 50 × 2 × 1.6 × 101.62 ⎤⎦ × 104 mm4

= 5736642.2 × 104 mm4
506 DESIGN OF STEEL STRUCTURES

Step 8 : Stress in compression flange :

⎛M ⎞ ⎛ 6905000 × 1000 × 1032 ⎞

σbc.cal = ⎜ × y⎟ = ⎜ ⎟
I
⎝ xx ⎠ ⎝ 5736642 ⋅ 4 × 104 ⎠
= 124.2 N/mm2

5 00 m m
16 m m
16 m m

2 IS A
2 00 m m ×
1 00 × 15 m m
2 00 0
1 60 0
mm
mm
x x
12 m m

2 IS A
2 00 m m ×
1 00 × 15 m m

16 m m
16 m m

Fig. 8.9

Step 9: Stress in tension flange

⎛ 28556 ⎞
σbt.cal = 124 ⋅ 2 × ⎜⎝ ⎟
24642 ⎠
= 143.8 N/mm2 < 165 N/mm2. Hence, safe.
Step 10: Check for shear stress
⎛ 1381000 ⎞
τv.cal = ⎜ ⎟
⎝ 2000 × 12 ⎠
= 57.6 N/mm2
< (0 4 × 250) = 100 N/mm2. Hence, safe.

⎛d⎞ ⎛ 1800 ⎞
⎜ t ⎟ = ⎜⎝ 12 ⎟⎠
⎝ w⎠
= 150.
It requires intermediate stiffeners.
The maximum section of the plate girder shown in Fig. 8.9 is satisfactory for
plate girder. It requires the intermediate stiffeners.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 507

8.13 LENGTH OF FLANGE PLATES : (CURTAILMENT OF

FLANGE PLATES)
When the plate girder is subjected to loading, then, the maximum bending
moment occurs at one section usually. When the plate girder is simply supported
at the ends, and subjected to the uniformly distributed load, then, maximum
bending moment occurs at the centre. The values of bending moment decrease
towards the supports. The flanges of plate girder or the plate girder section is
designed to resist the maximum bending moment. The flange area designed to
resist the maximum bending moment is not required at other sections. The
flange area is reduced as the bending moment decreases. The reduction in flange
area is done by curtailing the flange plates. Each flange plate is curtailed or
made shorter than the neighbourhood flange plate nearer to the flange angles.

2 x1
2x2
B .M . D ia gra m fo r U .D .L.

P late g ird er

Fig. 8.10 Curtailment of flange plates

The curtailed length of the flange plates suits the variation in the bending
moment. It gives economy as regards the material and cost. The actual lengths
of flange plates curtailed are found as follows :
When the plate girder is carrying uniformly distributed load, the bending
moment diagram has parabolic curve.
Let Af = Total flange area including web equiva-
lent
A1, A2,..., An = Area of individual cover plates, counted
from outer side
2x1, 2x2.... 2xn = Theoretical length of individual plates
as shown in Fig. 8.10
l = Span of plate girder.
The flange area is given by
M
Af = ...(i)
(σbc × de )
From this, it is seen that flange area varies as the moment. As such, flange
area also varies following parabolic curve. The effective depth of plate girder is
assumed as constant.

⎛ A1 ⎞ ⎛ 4x12 ⎞
Then ⎜A ⎟ = ⎜ 2 ⎟ ...(ii)
⎝ f⎠ ⎝ l ⎠
508 DESIGN OF STEEL STRUCTURES

1 1
l ⎛ A1 ⎞ 2 l ⎛ A + A2 ⎞ 2
x1 = , x2 = ⎜ 1 ...(iii)
2 ⎜⎝ Af ⎟⎠ 2 ⎝ Af ⎟⎠

1
l ⎛ A1 + A2 + .... An ⎞ 2
xn = ...(8.17)
2 ⎜⎝ Af ⎟⎠

Alternatively. Equation 8.17 may also be found by equating the bending

moment at any section to the moment of resistance of the section. The
maximum bending moment for the uniformly distributed load of intensity w
per unit length over the simply supported plate girder
⎛ wL2 ⎞
Mmax = ⎜ ⎟ ...(iv)
⎝ 8 ⎠
The moment of resistance of the section resisting the maximum bending
moment
⎛ 1 ⎞
M.R.max = ⎜ Af 1 + Aw ⎟ ⋅ σbc ⋅ d ...(v)
⎝ 6 ⎠
where, Af1 = Area of flange excluding web equivalent
Equating the expressions (iv) and (v)
⎛ 1 ⎞ ⎛ wL2 ⎞
⎜ Af 1 + Aw ⎟ ⋅ σbc ⋅ d = ⎜ ⎟ ...(vi)
⎝ 6 ⎠ ⎝ 8 ⎠
In case, one flange plate is curtailed from the top, two flange plates are
curtailed from the top and n flange plates are curtailed from the top, then, the
expressions similar to the expression (vi) may be written as follows :

⎡⎛ 1 ⎤ ⎛ wL2 w (2x )2 ⎞
⎞ = ⎜ – 1 ⎟ ...(viii)
⎜ A +
⎢ f1 6 w A ⎟ – A σ
1 ⎥ bc ⋅ d
⎣⎝ ⎠ ⎦ ⎝ 8 8 ⎠

⎡⎛ 1 ⎞ ⎤ ⎛ wL2 w (2x )2 ⎞
⎢⎜ Af 1 + Aw ⎟ – ( A1 + A2 )⎥ σbc ⋅ d = ⎜
2 ⎟
– ...(viii)
⎣⎝ 6 ⎠ ⎦ ⎝ 8 8 ⎠

⎡⎛ 1 ⎞ ⎤ ⎛ wL2 w (2x )2 ⎞
⎢⎜ Af 1 + Aw ⎟ – ( A1 + A2 + An )⎥ σbc ⋅ d = ⎜
n ⎟
– ...(ix)
⎣⎝ 6 ⎠ ⎦ ⎝ 8 8 ⎠

By dividing the expressions (vii), (viii) and (ix), respectively, by the expres-
sion (vi)

⎛ A1 ⎞ ⎧⎪ (2x )2 ⎫⎪
⎜
1–
⎟ = ⎨1 – 21 ⎬ ...(x)
1 ⎪⎩ l ⎪⎭
⎜⎝ Af 1 + Aw ⎟
6 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 509

⎧⎪ 2
⎛ A1 + A2 ⎞
⎨1 –
(2x2 ) ⎫⎪⎬
⎜ 1 – ⎟ = ⎪ ...(xi)
1 ⎩ l 2 ⎪⎭
⎜ Af 1 + Aw ⎟
⎝ 6 ⎠
⎛ A1 , A2 ,.... An ⎞ ⎪⎧
⎨ 1 –
(2xn )2 ⎪⎫⎬
⎜ 1 – ⎟ = ...(xii)
1 ⎪⎩ l 2 ⎭⎪
⎜ Af 1 + Aw ⎟
⎝ 6 ⎠
1
l ⎛ A1 ⎞ 2
Therefore, x1 = ...(xiii)
2 ⎜⎝ A2 ⎟⎠
1
l ⎛ A1 + A2 ⎞ 2
x2 = ...(xiv)
2 ⎜⎝ Af ⎟⎠
1
l ⎛ A1 + A2 + .... An ⎞ 2
x3 = ...(xv)
2 ⎜⎝ Af ⎟⎠

⎛ 1 ⎞
where, Af = ⎜ Af 1 + Aw ⎟
⎝ 6 ⎠
The computations are done for the curtailment of flange plates in tension
flange. Net areas of plates and flanges are used in calculations. The curtailment
of plates in compression flange is done at the same positions as that in tension
flange, for economy in fabrication.

8.13.1 Graphical Method of Curtailment of Flange Plates

For other system of loading, other than that of uniformly distributed load,
analytical expression for the theoretical lengths of cut-off the flange plates
becomes complicated. In such a situation, the graphical method of curtailment
of flange plates is adopted. The graphical method is suitable for all types of
loading including the system of uniformly distributed load. In the graphical
method, the bending moment diagram for the given system of loading is drawn
to the scale. Let the bending moment diagram for the given system of loading
be as shown in Fig. 8.11. The maximum section of the plate girder is designed
for the maximum value of the bending moment DD1.
The moment of resistance of the section resisting the maximum bending
moment from expression (v) above.
⎛ 1 ⎞
M.R. = ⎜ Af 1 + Aw ⎟ ⋅ σbc ⋅ d
⎝ 6 ⎠
⎛ 1 ⎞
or M.R. = ⎜ Afa + A1 + A2 + .... An + Aw ⎟ σb ⋅ d
⎝ 6 ⎠
⎛ 1 ⎞
or M.R. = ⎜ Afa + Aw + A1 + A2 + .... An ⎟ σb ⋅ d ...(xvi)
⎝ 6 ⎠
510 DESIGN OF STEEL STRUCTURES

where, Afa = Area of the flange-angles

A1, A2,..., An=Area of individual cover plates, counted from outer side
The expression (xvi) gives moment of resistance of the section. The moments
of resistance of the section with one cover plate curtailed, with two cover plates
curtailed, with n cover plates curtailed are given by the following expressions
⎡⎛ 1 ⎞ ⎤
M1.R1 = ⎢⎜ Afa + Aw ⎟ + A2 + .... An ⎥ σb ⋅ d ...(xvii)
⎣⎝ 6 ⎠ ⎦

⎡⎛ 1 ⎞ ⎤
M2.R2 = ⎢⎜ Afa + Aw ⎟ .... + An ⎥ σb ⋅ d ...(xviii)
⎣⎝ 6 ⎠ ⎦

⎡⎛ 1 ⎞⎤
Mn.Rn = ⎢⎜ Afa + Aw ⎟ ⎥ σb ⋅ d ...(xix)
⎣⎝ 6 ⎠⎦
The moments of resistance of the sections, M1.R1,M2.R2,...Mn.Rn are found
and then plotted on the same bending moment diagram as shown in Fig. 8.11.
The horizontal lines intersecting with bending moment diagram given the theo-
retical lengths of curtailment of plates, EF, GH,..., and JK.
Q D

E F
C

G H

J K
( A fa + 1 A w )
6 A B
( L = sp an )
Fig. 8.11

Alternatively, from the expression (xvi) it is seen that the total moment of
resistance comprises of those due to flange angles and equivalent of web, i.e.,
⎛ 1 ⎞
⎜⎝ Afa + Aw ⎟⎠ σbd , due to one cover plate, i.e., A1 σbd , due to second cover plate,
6
i.e., A2 σbd due to second cover plate, i.e., An σbd . It is to note that the moments
of resistance of different flange elements are proportional to their respective
areas.
A line AQ is drawn perpendicular to AB at A to represent the moment of
resistance of the maximum section, (AP = DD2, maximum bending moment). A
line AP is drawn at any inclination from A. On this line, AP, APn, P2 Pn, P1P2
⎛ 1 ⎞
and PP1 are marked equal to ⎜ Afa + Aw ⎟ , A , A and A , respectively to some
n 2 1
⎝ 6 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 511

scale. The points P and Q are joined. Then, from P1, P2, Pn lines P1 Q1, P2Q2 and
PnQn are drawn parallel to PQ. The horizontal lines QtE, Q2 GH and QnJK and
drawn through Q1, Q2 and Qn, respectively. The intersections of these horizon-
tal lines with the bending moment diagram give the theoretical lengths of cut-
off as EF, GH and JK. In this method, it is not necessary to calculate the mo-
ments of resistance. Thus, the theoretical lengths of cut-off of the cover plates
are found graphically.
The flange plates are extended beyond the theoretical cut-off points. IS: 800–
1984 recommends that the extension shall contain sufficient rivets to develop
in the plate, the load calculated for bending moment and girder action (taken to
include the curtailed plate) at the theoretical cut-off points. When the plate
girder is used in the exposed situation, at least one plate of top flange shall
extend the full length of girder, unless the top edge of the plate is machined
flush with the flange angles. Where two or more flange plates are used on the
one flange, then, the tacking rivets shall be provided, if necessary.
The actual lengths of curtailed flange plates shall be theoretically lengths of
cut-off plus the extended lengths on both the sides of the plates.
The outstand of flange plates with unstiffened edges, that is, their projection
beyond the outer line of connections to flange angles shall not exceed the values
specified in the design of compression members.
Example 8.3 A plate girder simply supported at ends is composed of flange
plates, flange angles and web plate as shown in Fig. 8.12. Effective span of plate
girder is 20 m. The plate girder is subjected to maximum bending moment of
9000 kN-m due to uniformly distributed load. Determine actual lengths of the
flanges if they are curtailed.

16 m m
16 m m
16 m m
25 00 m m

x x
16 m m

16 m m
16 m m
16 m m

Fig. 8.12
512 DESIGN OF STEEL STRUCTURES

Solution
Design :
Step 1 : Area of flanges
The bending moment diagram due to uniformly distributed load is parabola
as shown in Fig. 8.13. The maximum bending moment occurs at the centre of
plate girder. From table below :
Net area of tension flange
= 377784 mm2
Gross area of compression flange = 43117 mm2

2x 3
2 x2
2 x1
9 00 0 kN /m
1 1
2 2
3 3

20 m

Fig. 8.13 Bending moment diagram

Flange Area
Description Gross Area Deduction for Netarea (mm)2
(mm)2 rivet hole (mm)2
Flange angles 2 × 5025 4 × 23.5 × 15
2 ISA 200 mm = 10050 = 1410 8640
× 100 mm × 15 mm
3 Flange Plates
550 mm × 16 mm 550 × 160 = 8800 2 × 23.5 × 16 = 752 8048
550 mm × 16 mm 550 × 160 = 8800 2 × 23.5 × 16 = 752 8048
550 mm × 16 mm 550 × 160 = 8800 2 × 23.5 × 16 = 752 8048
1 1
Equivalent web area Aw — Aw area
6 6

1 2500 × 12
= × 2500 × 16 — =
6 8
= 6667 = 5000
Total 43117 37784
Step 2 : Gross moment of inertia
The neutral axis (xx-axis) of the plate girder is assumed to be located at the
centre. The moment of inertia of flange plates about their own axis is neglected.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 513

Gross moment of inertia

Number Web plate Flangee ∠s Flanges ∠s Flanges Ixx
of flange 104 × mm4 about about about (gross)
plates own axis xx-axis xx-axis 104 × mm4
104 × mm4 104 × mm4 104 × mm4
Nil 2083333.3 4 × 969.9 4 × 50.25 × 50472129
= 3879.6 (12.5 – 372)
= 2960000
One 2083333.3 3879.6 2960000 2 × 55 × 1.6 × 125.82 7837212.9
= 2790000
Two 2083333.3 3879.6 2960000 2 × 55 × 3.2 × 12562 13487212.9
= 5650000
Three 2083333.3 38796 2960000 2 × 55 × 48 × 12742 22137212
= 8650000
Step 3: Curtailments of plates in tension flange (Theoretical length
of cut-off of plates)
Let the theoretical length of cut-off flange plates be 2x1, 2x2, and 2x3 as shown
in Fig. 8.13.
1
1 ⎛ A1 ⎞ 2
Length x1 = L
2 ⎜⎝ A fn ⎟⎠

where, A1 = Net area of the plate curtailed

Afn = Net flange area
1
20 ⎛ 8048 ⎞ 2
Length x1 = ⎜ ⎟ = 4.62 m
2 ⎝ 37784 ⎠

1
1 ⎛ A1 + A2 ⎞ 2
Length x2 = L
2 ⎜⎝ Afn ⎟⎠

1
20 ⎛ 2 × 8048 ⎞ 2
x2 = ×⎜ ⎟ = 6.52 m
2 ⎝ 37784 ⎠

1
L ⎛ A1 + A2 + An ⎞ 2
Length x3 =
2 ⎜⎝ Afn ⎟
⎠
1
20 ⎛ 3 × 8048 ⎞ 2
x3 = ⎜ ⎟ = 8.00 nm
2 ⎝ 37784 ⎠
The plates to be extended beyond their theoretical cut-off points. The value
of bending moments at the theoretical cut off points are found as follows:
514 DESIGN OF STEEL STRUCTURES

Equation of parabolic bending moment diagram with A as origin, is given by

4a
y = x ⋅ (L – x )
L2

L
At x = , y = 9000 kN-m
2
∴ a = 9000
(L – x )
∴ y = 36000 x
L2
Step 4: Distances to sections for theoretical cut-off
Distance of cross-section 1,
from A = (10 – 4.62) = 5.38 m
Distance of cross-section 2,
from A = (10 – 6.52) = 3.48 m
Distance of cross-section 3,
from A = (10 – 8) = 2 m
Step 5: Bending moment at sections for theoretical cut-off
Bending moment at section 1,
36000
M1 = × 5 ⋅ 38 × 14 ⋅ 62 = 7019 kN-m
400
Bending moment at section 2,
36000
M2 = × 3 ⋅ 48 × 16 ⋅ 52 = 5174 kN-m
400
Bending moment at section 3,
36000
M3 = × 2 × 18 = 3240 kN-m
400
The extensions of the plates beyond the theoretical cut-off points contain
vets to develop in the plate, the load calculated for the bending moments and
rder section (taken to include the curtailed plate) at the theoretical cut-off )
points. The bepding stresses in compression are found considering the gross
moment of inertia of the sections as follows:
Step 6: Bending stress in compression in the top most cover plate
The gross moment of inertia of section cut-off the top-most plate
M1 = 7079 kN-m
The distance upto the centre of top-most cover plate
⎛ 1 ⎞
y1 = ⎜⎝125 + 2 × 1.6 + × 1.6⎟⎠ = 129 × 10 mm
2
The bending stress in compression flange at section, 1
⎛ 7079 × 106 × 1290 ⎞
σbc.cal.1 = ⎜ ⎟ = 41.25 N/mm
2
⎝ 22137212.9 × 104 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 515

Step 7: Bending stress in compression in the second cover plate

The gross moment of inertia of section including the second plate to be
curtailed
Ixx(gross)2 = 13487212.9 × 104 mm4
The bending moment at the theoretical cut-off of the second plate
M2 = 5174 kN-m
The distance upto the centre of second cover plate

⎛ 1 ⎞
y2 = ⎜125 + 1 ⋅ 6 + × 1 ⋅ 6 ⎟ = 127.4 × 10 mm
⎝ 2 ⎠
The bending stress in compression flange at section, 2

⎛ 5174 × 106 × 1274 ⎞

σbc.cal.2 = ⎜ ⎟ = 48.87 N/mm2
⎝ 13487212 ⋅ 3 × 104 ⎠
Step 8: Bending stress in the third cover plate
The gross moment of inertia of section including the third cover plate to be
curtailed
Ixx(gross).3 = 7837212.9 ×104 mm4
The bending moment at the theoretical cut-off of the third cover plate
M3 = 3240 kN-m
The distance upto the centre of third cover plate

⎛ 1 ⎞
y3 = ⎜125 + × 1 ⋅ 6 ⎟ = 125.8 × 10 mm
⎝ 2 ⎠
The bending stress in compression flange at section, 3

⎛ 3240 × 106 × 1258 ⎞

σbc.cal. = ⎜ ⎟
3 ⎝ 7837212 ⋅ 9 × 104 ⎠
= 52 N/mm2
Step 9: Bending stress in tension upon the centre of respective cover
plate at the respective sections

⎛ 41 ⋅ 25 × 43117 ⎞
σbc.cal. = ⎜ ⎟
1 ⎝ 37784 ⎠
= 47.072 N/mm2

⎛ 48 ⋅ 87 × (43117 – 8800 ) ⎞
σbc.cal. = ⎜ ⎟
2 ⎝ (37784 – 8048 ) ⎠
= 56.396 N/mm2
516 DESIGN OF STEEL STRUCTURES

⎛ 52 × (43117 – 2 × 8800) ⎞
σbc.cal. = ⎜
3 ⎝ (377.84 – 2 × 8048) ⎟⎠
= 61.18 N/mm2
Step 10 : Tensile force in each cover plate at the respective sections
⎛ 8048 ⎞
F1 = ⎜ 47 ⋅ 072 × ⎟ = 378.834 kN
⎝ 1000 ⎠

⎛ 8048 ⎞
F2 = ⎜ 48 ⋅ 87 × ⎟ = 393.304 kN
⎝ 1000 ⎠

⎛ 8048 ⎞
F3 = ⎜ 61 ⋅ 18 × ⎟ = 492.37 kN
⎝ 1000 ⎠
Step 11 : Rivet value
Use 22 mm diameter power driven rivets. The strength of rivet in single
shear
⎛ π (23 ⋅ 5 )2 × 100 ⎞
⎜ × ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of rivet in bearing
⎛ 16 × 300 ⎞
⎜ 23 ⋅ 5 × ⎟ = 112.8 kN
⎝ 1000 ⎠
Rivet value, R = 43.35 kN
Step 12 : Design of riveted connections
Number of rivets at section, 1–1
⎛ 378 ⋅ 834 ⎞
⎜ ⎟ = 8.739
⎝ 43 ⋅ 35 ⎠
Number of rivets at section, 2–2
⎛ 393 ⋅ 304 ⎞
⎜ ⎟ = 9.07
⎝ 43 ⋅ 35 ⎠
Number of rivets at section, 3–3
⎛ 492 ⋅ 37 ⎞
⎜ ⎟ = 11.36
⎝ 43 ⋅ 35 ⎠
Provide 16 rivets at each section. These rivets are provided in two rows in
extended portion of the plate beyond the theoretical sections with 1.5 times
leter of rivets as edge distance and 3 times diameter as pitch.
Step 13 : Actual length of curtailment of top-most cover plate
⎛ 2 ⎞
⎜ 2 × 4 ⋅ 62 + × 8 × 3 × 23 ⋅ 5 ⎟ = 10.36 m
⎝ 100 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 517

Actual length of curtailment of second cover plate

⎛ 2 ⎞
⎜ 2 × 6 ⋅ 52 + × 8 × 3 × 23 ⋅ 5 ⎟ = 14.16 m
⎝ 100 ⎠
Actual length of curtailment of third cover plate
⎛ 2 ⎞
⎜2× 8 + × 8 × 3 × 23 ⋅ 5 ⎟ = 17.12 m
⎝ 100 ⎠
The flange plates in compression flange are curtailed at the same positions,
where the flange plates in the tension flange are curtailed. If the plate girder is
used in the exposed conditions, one plate in top flange is continued throughout
the length of girder.

8.14 CONNECTION OF FLANGE ANGLES TO WEB

The flange angles of plate girder are connected to the web by sufficient rivets to
transmit the maximum horizontal shear force resulting from bending moments
in the girder, combined with any vertical loads which are directly applied to the
flanges.
The design of connection of flange angles to web depends upon horizontal
shear. A vertical component of load is added in the horizontal shear. The rivets
connecting flange angles to web resist the resultant of vertical component of
load and horizontal shear as shown in Fig. 8.14. The pitch of rivets is found as
under :
The shear stress at any point of the cross-section is given by
⎛ VQ ⎞
τvf.cal = ⎜ ⎟ ...(8.18)
⎝ Ib ⎠
where, tvf.cal = Calculated shear stress
V = Shear force at the cross-section
Q = Static moment about the neutral axis of the portion of cross-
sectional area beyond the location at which stress is determined
l = Moment of inertia of section
b = Width of section, at which stress is determined.

v
τv w
p p

v
N A

Fig. 8.14
518 DESIGN OF STEEL STRUCTURES

Shear per unit depth

⎛ VQ ⎞
τvf.cal = ⎜ × b ×1⎟ ...(i)
⎝ Ib ⎠

⎛V ⋅Q ⎞
τvf.cal = ⎜ ⎟ ...(ii)
⎝ I ⎠
Moment of section

⎛ de de ⎞
I = ⎜ 2 Af × × ⎟ ...(iii)
⎝ 2 2 ⎠

⎛ de ⎞
and Q = ⎜ A ⋅ y = Af × ⎟ ...(iv)
⎝ 2 ⎠
Shear per unit depth

⎡ ⎛ de ⎞ ⎤
⎢ V × Af × ⎜⎝ 2 ⎟⎠ ⎥ ⎛ V ⎞
τvf.cal = ⎢ ⎥= ⎜ ⎟ ...(v)
⎢ 2 A × ⎛ de ⎞ × ⎛ de ⎞ ⎥ ⎝ de ⎠
⎢⎣ f ⎜ ⎟ ⎜ ⎟
⎝ 2 ⎠ ⎝ 2 ⎠ ⎥⎦

The shear stress per unit depth is accompanied by horizontal shear per unit
length of equal intensity. The horizontal shear per unit length is therefore,
given by

V Af 1
τvf.cal = × ...(8.19)
de ⎛ 1 ⎞
⎜ Af 1 + ⋅ A ⋅ w ⎟
⎝ 6 ⎠
where Af1 is the gross area of flange excluding web equivalent. The resultant of
horizontal shear per unit length and vertical component of load w is given by
1
2 2
= ⎡⎣( τvf .cal ) + (w ) ⎤⎦ 2 ...(vii)

The resultant force per pitch length should be less than or equal the rivet
value.
1
Then, p × ⎡( τvf .cal )2 + (w )2 ⎤ 2 = R
⎣ ⎦

⎛ R ⎞
p = ⎜ ⎟ 1
...(viii)
⎜⎜ ⎡ 2 ⎟⎟
)2 ⎤ 2
⎝ ⎣( vf .cal )
τ + (w ⎦ ⎠
Substitute the value of τvf.cal The pitch of rivets connecting flange angles to
web for loaded flange is given by
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 519

R
p = 1
...(8.20)
⎡V Af 1 ⎤2
⎢d × 2 ⎥
⎢ e ⎛ A + 1 A ⎞ (w )2 ⎥
⎢⎣ ⎜
⎝ f 1 w ⎟
⎠ ⎥⎦
8

The pitch of rivets connecting flange angles to web for unloaded flange (w = 0)
is given by

R
p = ...(8.21)
⎛V Af′1 ⎞
⎜d × ⎛ 1 ⎞ ⎟
⎜ e ⎜⎝ Af′1 + Aw ⎟⎠ ⎟
⎝ 8 ⎠

where, Af́1 is net area of tension flange. The bottom flange is in tension and it
is unloaded flange.
Where a load is directly applied to the top flange, it is considered as
dispersed uniformly at an angle of 30 degrees to the horizontal.

8.15 CONNECTION OF FLANGE PLATES TO FLANGE ANGLES

The flange plates of the plate girder are connected to the flange angles by suffi-
cient rivets.
The design of connection of flange plate to flange angles is designed for the
horizontal shear. The horizontal shear per unit length (for tension flange)

V ⎛ A1 + A2 + .... + An ⎞
τvf.cal = × ...(i)
de ⎜⎜ 1
Af 1 + An
⎟
⎟
⎝ 8 ⎠
where, A1 A2.. .An = Areas of flange plates, counting from outside.
The pitch of rivets for tension flange, if rivets are to be provided in one straight
line, is given by

⎛ 1 ⎞
A´ f 1 + Aw
R × de ⎜ 8 ⎟
p = × ⎜⎜ ⎟⎟ ...(8.22a)
V A´ A
⎝ 1+ 2+ + ´ .... A´ ⎠
For the compression flange

⎛ 1 ⎞
A´ f 1 + Aw
R × de ⎜ 6 ⎟
p = × ⎜⎜ ⎟⎟ ...(8.22b)
V ⎝ A´1 + A´2 +.... + A´ ⎠
But the rivets are provided in two parallel lines, as such rivets are provided
at double the pitch computed above.
520 DESIGN OF STEEL STRUCTURES

Example 8.4 A simply supported plate girder spans 24 m and supports a

uniformly distributed load of 3840 kN excluding self-weight. The section of plate
girder at supports is shown in Fig. 8.15.

4 50 m m
y 20 m m

22 00 m m
x x
12 m m

2 ISA
2 00 m m ×
1 50 m m × 1 8 m m

20 m m

Fig. 8.15

Design the connection of flange angles to web and flange plates to flange angles.
Solution
Design:
Step 1: Total load supported plate girder
⎛ 3840 × 24 ⎞
Self-weight of plate girder = ⎜ ⎟ 307 @ 310 kN
⎝ 300 ⎠
Total uniformly distributed load = (3840 + 310) = 4150 kN
⎛ 4150 ⎞
Shear force at support V = ⎜ ⎟ = 2075 kN
⎝ 2 ⎠
Step 2: Effective depth of plate girder
Let y be the distance of C.G. of compression flange from top. From ISI Hand-
book No. 1, cross-sectional area of one flange angle
= 59.75 × 100 mm2
Distance of its C.G. = 38.4 mm
⎛ 45 × 2 × 1 + 2 × 59 ⋅ 76 (2 + 3 ⋅ 84 ) ⎞
y = ⎜ ⎟
⎝ 45 × 2 + 2 × 59 ⋅ 76 ⎠
= 3.75 × 10 mm
Effective depth of plate girder at supports
de = (224 – 2 × 3.75)
= 216.50 × 10 = 2165 mm
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 521

Step 3: Area of flanges

Net area of tension flange = 21622 mm2
Gross area of compression flange = 25352 mm2
Flange Area
Description Gross Area Deduction for Net area (mm)2
(mm)2 rivet hole (mm)2
Flange angles 2 × 5976 4 ×23.5 ×18
2 ISA 200 mm = 11952 = 1690 10262
× 150 mm × 18 mm
Flange Plates 450 × 20 2 × 23.5 × 20
450 mm × 20 mm = 9000 = 940 8060

1 1
Equivalent web area Aw — Aw area
6 8

1 2200 × 12
= × 2200 × 12 =
6 8
= 4400 — = 3300
Total 25352 — 21622
Step 4 : Connection of flange angles to web in compression flange
Load on compression flange per mm length
⎛ 3840 1000 ⎞
w = ⎜ × ⎟ = 160 N/mm
⎝ 24 1000 ⎠
Use 22 mm diameter power driven rivets
Strength of rivet in double shear
⎛ 2
π (23 ⋅ 5 ) × 100 ⎞
⎜2× × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of rivet in bearing
⎛ 23 ⋅ 5 × 12 × 300 ⎞
⎜ ⎟ = 84.6 kN
⎝ 1000 ⎠
Rivet valued R = 84.6 kN
Let p be the pitch of rivets connecting flange angles to web. From Eq. (8.20)

R
p = 1
mm
⎡⎛ V 2 ⎤2
Af 1 ⎞
⎢⎜ × ⎟ + w2 ⎥
⎢⎜ de A + 1 A ⎟ ⎥
⎢⎣⎝ f1 w ⎥⎦
6 ⎠
522 DESIGN OF STEEL STRUCTURES

84 ⋅ 6
p = 1
= 106.42 mm
⎡⎛ 2075 × 1000 11952 + 9000 ⎞ 2 ⎤2
2
⎢⎜ × ⎟ + (160 ) ⎥
⎣⎝ 2165 ⋅ 0 25352 ⎠ ⎦
Provide 100 mm pitch for rivets connecting flange to web plate.
Step 5: Connection of flange plates to flange angles in compression
flange
Use 22 mm diameter power driven rivets
Strength of rivet in single shear
⎛ π (23 ⋅ 5)2 × 100 ⎞
⎜⎝ × ⎟⎠ = 43.35 kN
4 1000
Strength of rivet in bearing
⎛ 23 ⋅ 5 × 18 × 300 ⎞
⎜ ⎟ = 126.9 kN
⎝ 1000 ⎠
Rivet value R = 43.35 kN
Let p be the pitch of the rivets, if rivets arc provided in one row. From
Eq. 8.22 (b)
⎛ 1 ⎞
Af 1 + Aw
R × de ⎜ 6 ⎟
p = × ⎜⎜ ⎟⎟
V ⎝ A1 + A2 + .... + An ⎠
43 ⋅ 35 × 217 ⋅ 50 ⎛ 20952 + 4400 ⎞
p = ×⎜ ⎟
2075 × 1000 ⎝ 9000 ⎠
= 128.248 mm
Rivets are used in two rows
∴ 2p = 256.496 mm
Provide 230 mm pitch for rivets in two rows connecting flange plates to flange
angles.
Step 6: Connection of flange angles to web in tension flange
Rivet value, R = 84.6 kN
Tension flange is unloaded flange
∴ w = 0
Let p be the pitch of the rivets connecting flange angles to web. From Eq. 8.21
R
p =
⎛V A´ f ⎞
⎜ × 1 ⎟
⎜ de A´ f + Aw ⎟
⎝ 8 ⎠
84 ⋅ 6
p= = 105.75 mm
⎛ 2075 × 1000 10262 + 8060 ⎞
⎜ × ⎟
⎝ 2165 ⋅ 0 21622 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 523

Provide 100 mm pitch for rivets connecting flange angles to web plate.
Step 7 : Connection of flange plates to flange angles in tension flange
Use 22 mm diameter power driven rivets.
Rivet value R = 43.35 kN
Let p be the pitch or rivets if rivets are provided in one row. From Eq (8.22a)

⎛ 1 ⎞
A × de ⎜ A´ f 1 + Aw
8 ⎟
p = ×⎜
V ⎝ A1′ + A2′ + .... + An′ ⎟⎠

⎛ 43 ⋅ 35 × 217 ⋅ 50 21622 ⎞
p = ⎜ × ⎟ = 121.88 mm
⎝ 2075 ⋅ 0 × 1000 8060 ⎠
Rivets are used in 2 rows
∴ 2p = 243.77 mm
Provide 230 mm pitch for rivets in two rows connecting flange plates of angles.

8.16 DIAGONAL BUCKLING OF WEB

The diagonal buckling of web occurs when the ratio of clear depth to thickness
⎛d⎞
of web ⎜ ⎟ exceeds 85. It occurs because of diagonal compression as shown in
⎝ tw ⎠
Fig. 8.16. A small element of the web plate is subjected to the bending stress σb
acting horizontally and shear stress, acting vertically and which one is also

B u ckling D ia g on al
com p ression

4 5° 4 5° W eb p la te

Fig. 8.16 Diagonal buckling of web

accompanied by complimentary shear stress, τvf acting horizontally.The princi-

pal stress which small be acting on this element is under :
1/ 2
σ ⎡⎛ σ ⎞ 2 2⎤
σ1 = b + ⎢⎜ b ⎟ + ( τvf ) ⎥ ...(i)
2 ⎣ 2⎝ ⎠ ⎦
1/ 2
σ ⎡⎛ σ ⎞ 2 2⎤
σb = b – ⎢⎜ b ⎟ + ( τvf ) ⎥ ...(ii)
2 ⎣ 2⎝ ⎠ ⎦
524 DESIGN OF STEEL STRUCTURES

At the end, the plate girder is simply supported, where, the bending moment
and thereby bending stress are zero. Therefore, σ1 = τvf and σ2 = τvf (numerically).
One principal stress is compressive, while the other is tensile. The inclination
of the principal stress is given by

⎛ –τ τ ⎞
tan 2θ = ⎜ vf = vf ⎟ = ∞
⎛ σb ⎞ zero
⎜⎜ ⎟ ⎟
⎝⎝ 2 ⎠ ⎠
2θ = 90°, ∴ θ = 45°
Both these stresses σ1 and σ2 act along the diagonals inclined 45° with the
horizontal line. The compressive stress acting along the diagonal causes buckling
of the web plate and phenomenon is known as diagonal buckling. In this
phenomenon, the diagonal compression and the diagonal tension of equal
intensities, acting mutually at right angles are set up in the web at the neutral
axis at an inclination of 45° with horizontal. At the neutral axis, the intensity of
diagonal compression is equal to that of shear. Intensity and inclination of
diagonal compression and of diagonal tension are changed by bending
compressive stress above the neutral axis, and bending tensile stress below the
neutral axis.

Fig. 8.17 Truss like behaviour of plate girder

The strength of web plate is increased by providing stiffeners. When the

suffeners are provided, and diagonal buckling of web occurs, then flanges and
vertical stiffeners form imaginary panels in the plate girder. Each panel of the
plate girder behaves like a panel of a Pratt truss. The flanges behave like the
top and bottom chords of the truss and the web plate constitutes the web mem-
bers. The vertical stiffeners act as struts and diagonal strips of web act as ties
as shown Fig. 8.17. The buckling strength of a plate girder web that has regu-
larly spaced transverse stiffeners is increased by its capacity because of truss
like behaviour. The diagonal carrying tension are considered as effective. The
contribution of those diagonals carrying compression to the shear resistance of
the web is neglected (although each would, in fact, contribute its small buckling
load). Therefore, the shear resistance between any two transverse stiffeners of
an adequated stiffened web is the sum of the shear buckling strength of the web
and the vertical component of the web yield strength in direct diagonal tension.
This latter is sometimes known as tension field strength.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 525

8.17 BEARING STIFFENERS

The bearing stiffeners are attached with the web plate of the plate girder to
avoid local bending failure of the flange, local crippling or buckling and crush-
ing of web (web crippling). The bearing stiffeners are provided under concen-
trated loads and at the points of supports. The bearing stiffeners strengthen the
web and transmit heavy concentrated loads or reactions to the flanges of plate
girder. When these are provided at ends, these are also termed as end bearing
stiffeners. They should be joggled or crimped. They are provided straight as
shown in Fig. 8.18. The bearing stiffener extend the full dept between flange to
flange and have a tight bearing against the flange transferring the load. The
filler plates of thickness equal to the thickness of flanges angles are inserted
between web plate and stiffeners.

B e aring
stiffe ne r

Filler
p late

Fig. 8.18 Bearing stiffener

The various types of bearing stiffeners used are shown in Fig. 8.19. The angle
sections in the end bearing stiffeners are arranged with reference to the point
of points of maximum pressure on the flanges. In the fixed bearing (plate bear-
ings and cast-iron shoe bearings), the maximum pressures are likely to occur at
the outer and inner edges of the sole plate (the plate attached to the bottom
flange of plate girder at the support). At such plates, the arrangements of the
angles as shown in Fig. 8.19 (b) and (d) are suitable. In the pin or centre-bear-
ing, the maximum pressure occurs at the centre of bearing. Therefore, the ar-
rangement of the angles as shown in Fig. 8.19 (c) i.e., in the form of cluster of
four angles is suitable. Sometimes, two angles (one on either side of the web
plate) are attached to the free or outer edge of the web, and four angles in the
form of a cluster are arranged at the point of maximum pressure.
526 DESIGN OF STEEL STRUCTURES

2 0 tw 2 0 tw
(a ) (b )

2 0 tw 2 0 tw

(c) (d )

A n gle se ctio ns use d as b ea rin g stiffe ne rs

Fig. 8.19 Bearing stiffeners

The end of bearing stiffeners are machined or ground, and are fitted lightly
between top and bottom flange angles. They
are provided symmetrically about the web.
The outstanding width of the bearing
stiffener is kept such as to extend as nearly R o ot of
flan ge a ngle
as practicable upto the outer edges of flange
angles. The bearing stiffeners are initially
selected on the basis of bearing area
B e aring
required. The intensity of bearing pressure
Flan ge a ngle
should not exceed σp =0.75 fy N/mm2. The stiffe ne r
bearing areas outstanding legs are taken
clear of root of the flange angles as shown in
Fig. 8.20. The load bearing capacity Filler pla te

calculated using the effective area should be

greater than or equal to the applied load or Fig. 8.20
the reaction.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 527

The bearing stiffeners are designed as columns of sectional area consisting of

stiffeners together with a length of web in each side of centre line of stiffeners
as shown in Fig. 8.19 equal to 20 times the thickness of web, if available. This
area is used to determine the radius of gyration and to check the column stresses.
The effective length is assumed, 0.7 of length of the stiffeners because of secured
connection with the web. The radius of gyration is computed about the centre
line of the web. The rivets connecting stiffeners to the web under concentrated
loads are designed to transmit these loads. At supports the rivets connecting
stiffeners to the web are designed to transmit the reaction. The filler plates are
used in between the stiffeners and the web plate. The thickness of filler plate is
equal to the thickness of the flange angles. The rivets are connected through
the filler plates. The rivets carrying calculated shear stress through packing
greater than 6 mm thick are increased above the number required by normal
calculations by 2.5 percent for each 2 mm thickness of packing. For double shear
connections packed on both sides, the number of additional rivets required are
determined from the thickness of thicker packing. The additional rivets are
preferably be placed in an extension of the packing. When torsional restraint at
the ends of plate girder is provided only by stiffeners, then the moment of inertia
of the stiffeners about the centre line of the web shall not be less than
⎛ D 3T R ⎞
⎜ × ⎟
⎝ 250 W ⎠
where D = Overall depth of the girder
T = Maximum thickness of compression flange
R = Reaction on the bearing
W = Total load on girder.
The load carrying capacity of the bearing stiffeners as a column should be
greater than or equal to the applied load or the reaction.

8.18 INTERMEDIATE STIFFENERS

The intermediate stiffeners are used for the economical design of the web plate
of the plate girder.
The intermediate stiffeners are used to avoid diagonal buckling of the web.
⎛d⎞
Depending upon the ratio of clear depth to the thickness of web ⎜ ⎟ , vertical
⎝ tw ⎠
stiffeners, or vertical and horizontal stiffeners are provided throughout the length
of the girder. In Sec. 8.16, the diagonal buckling of web, it is seen that a complex
and variable stress pattern is set in the web due to combined bending moment
and shear. The inclined compressive stresses are set up in the direction at right
angles to the compressive stresses and at the neutral axis, the intensities of the
stresses are numerically equal. The direction of inclined compressive stresses
varies from 45° at the neutral axis to a considerably flatter angle at the edge of
the flange. Due to this, the web plate tends to buckle in one direction and tends
to straighten in the right angle direction. The intermediate stiffeners prevent
528 DESIGN OF STEEL STRUCTURES

the web plate from buckling under the inclined compressive stress. The vertical
intermediate stiffeners divide the web plate into small panels. These panels are
supported along the lines of stiffeners. The resistance of web plate to buckling
is measurably increased. The intermediate vertical stiffeners also have a sec-
ond function. When the vertical stiffeners are fitted against the top and bottom
flanges, then, they maintain the original 90° angle between the flanges and the
web. When the dimensions of web are very large, then the panel dimensions are
reduced by providing the horizontal stiffeners on the compression side of the
web.
Unless the vertical stiffeners are closely spaced, these are not sufficient in
increasing the resistance to buckling in bending. The horizontal stiffeners placed
on the compression zone of the web, effectively, increase its resistance to buck-
ling due to bending. A combination of both vertical and horizontal web stiffen-
⎛d⎞
ers is often used depending upon ⎜ ⎟ ratio. When the web plate is thoroughly
⎝ tw ⎠
stiffened, it becomes effective upon permissible stresses in shear.

8.18.1 Vertical Stiffeners

The vertical stiffeners are also termed as transverse stiffeners. The vertical stiff-
eners are provided throughout the length of the girder when the thickness of
web is less than the limits specified for the minimum thickness of the web plate.
They are joggled or crimped as shown in Fig. 8.21. They may be provided
straight. In that case, filler plate of thickness equal to that of flange angles is
inserted between the stiffener and web plate. They are fitted tightly between
outstanding legs of top and bottom flange angles. When two angles are used for
the vertical stiffeners, they are provided on either side of web. When single angle
is used for the vertical stiffeners, then they are placed alternately on opposite
sides of the web.

Jog gled
(crim p ed )

Ve rtica l
in te rm ed ia te
stiffe ne r

Fig. 8.21 Intermediate (vertical) stiffeners

 Design of OF
DESIGN Beam Connection
PLATE GIRDER 529

The vertical stiffeners are provided at spacing not greater than l.5 d and not
less than 0.33 d, where, d is the distance between flanges angles or where there
are no flange angles, the clear distance between flanges ignoring fillets. When
the horizontal stiffeners are provided, d shall be taken as the clear distance
between the horizontal stiffener and the tension flange (farthest flange) ignoring
fillets. The vertical stiffeners divide the web plate into number of panels of
dimensions c and d as shown in Fig. 8.6. The panels are supported along the
lines of stiffeners. The greater unsupported clear dimension of web panel should
not be greater than 270 times the thickness of web, and the lesser unsupported
clear dimension of the same web panel should not be greater than 180 times the
thickness of web. These limiting values of panel dimensions have been adopted
a
by keeping the aspect ratio as 1.5 and to avoid the web buckling. The spacing
b
of stiffeners is kept less near the supports, where the shear force is more. The
spacing in increased towards the centre since the shear force decreases towards
the centre. Actual average shear stress in the web should not exceed the allowable
average shear stress in the stiffened web.
The length of outstanding leg of vertical stiffener may be taken equal to 1/30
of the clear depth of girder plus 50 mm. The outstand of stiffeners from the web
256 ⋅ t
shall not be more than (where, t is the thickness of the section, so long as
fy1 / 2
the outer edge of each stiffener is not continuously stiffened, or 12 t for the flat
sections. The length of connected leg of vertical stiffener should be sufficient to
accommodate the rivets connecting the stiffener to the web. The moment of
inertia I of the stiffener selected should not be less than
⎛ 1 ⋅ 5d 3tw3 ⎞
| ⎜
I < ⎟
⎝ c2 ⎠
where, I = The moment of inertia of a pair of stiffeners about the centre
line of web, or of a single stiffener about the face of the web
tw = The minimum required thickness of the web
c = The maximum permitted clear distance between vertical stiff-
eners

8.18.2 Horizontal Stiffeners

The horizontal stiffeners are also termed as longitudinal stiffeners. When
the thickness of web is less than the limits specified (for minimum thickness of
web plate), are provided on the web at a distance from compression flange equal
2 ⎛1 ⎞
to th of distance of compression flange to the neutral axis i.e., at ⎜ ⎟ th × d,
5 ⎝5⎠
where d is the clear depth of the plate girder. The moment of inertia I, of the
horizontal stiffener should not be less than 4c1. t w2.
where c1 is the actual distance between the transverse (vertical) stiffeners.
530 DESIGN OF STEEL STRUCTURES

When the thickness of web is less than the limits specified, second horizontal
stiffeners in addition to vertical stiffeners and horizontal stiffeners described
above are provided at the neutral axis of the girder. The moment of inertia ‘I’ of
this horizontal stiffeners should not be less than d.tw3. The moment of inertia is
computed about the line as described for vertical stiffeners. The outstand of the
256 ⋅ t
stiffeners from the web shall not be more than 1
(where, t is the thickness
fy 2
of the section) or 12 t for the flat sections.
The requirements for moment of inertia for vertical stiffencrs are based on
the theoretical and experimental work. Timoshcnko used a value of 0.3 in the
expression of moment of inertia for the vertical stiffeners. Whereas Moore, R.L.
proposed a value of 1.33 based on his investigation on effectiveness of the
stiffeners on shear resistance of plate girder webs. Moore’s this value was found
to be adequate by ultimate load test, but it has been raised to 1.5. The expressions
for moment of inertia of horizontal stiffeners are based on theoretical work and
it is due to Blcich, F.
Connection of intermediate stiffeners to web plate. The intermediate
vertical stiffeners and horizontal stiffeners (when they are not subjected to
external load) are connected to the web by rivets to withstand shear force not
less than
⎛ 125 tw2 ⎞
⎜⎝ ⎟ kN/m
h ⎠
where, h is the outstand of stiffener in mm and t w is the thickness of web in mm.
External Forces on intermediate stiffeners. In case, the vertical loads
are acting eccentric or transverse forces are acting on the vertical intermediate
stiffeners causing bending moment and shear, the moment of inertia of the
stiffeners mentioned above shall be increased as under :
(a) The vertical load acting eccentric with respect to the vertical axis of the
web causing bending moment

⎛ 150 MD2 ⎞
Increase of I = ⎜⎜ ⎟⎟ × 104 mm4
⎝ E ⋅ tw ⎠
(b) The lateral loading of stiffener

⎛ 0 ⋅ 3VD2 ⎞
Increase of I = ⎜⎜ ⎟⎟ × 104 mm4
⎝ E ⋅ tw ⎠

where, M = Applied bending moment in kN-m

D = Overall depth of girder in mm
E = Young’s modulus of elasticity, 2.047 × 105 N/mm2 (MPa)
tw = Lateral force to be taken by the stiffener and deemed to be
applied at the compression flange of girder in kN.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 531

Example 8.5 A simply supported plate girder spans 20 m and carries a

uniformly distributed load, (including its own weight), of 3000 kN. The section
of plate girder at supports is shown in Fig. 8.22. Design end bearing stiffeners.
Also design the necessary intermediate stiffeners.
Solution
Design :
Step1: Design of bearing stiffener (End reaction)
The uniformly distributed load including own weight of plate girder is 3000 kN.
Support reaction = 1500 kN
Allowable bearing stress (yield stress for steel 250 N/m2)
σp = (0.75 ×250)
= 187.5 N/mm2
Step 2: Bearing area required

⎛ 1500 × 1000 ⎞
⎟ = 8000 mm
2
⎜
⎝ 187 ⋅ 5 ⎠

5 00 m m
20 m m

2 ISA 200 m m
× 15 0 mm
× 18 mm
2500 m m

x x
8 mm

2 ISA 200 m m
× 1 50 m m
× 18 mm

20 m m

Fig. 8.22

From ISI Handbook No. 1, select 4 ISA 150 mm × 115 mm × 15 mm (4 ISA

150 115, @ 0.394 kN/m)
Radius at root, r1 for the flange angle is 13.5 mm
Bearing area provided
= 4 (150 – 13.5) × 15
= 8190 mm2
The bearing provided is grater than bearing are a required. Provide 18 mm
thick filler plates, as shown in Fig. 8.23.
532 DESIGN OF STEEL STRUCTURES

4 IS A 15 0 m m
x 11 5 x 1 5 m m

18 m m
8 mm
18 m m

Filler
p late

Fig. 8.23 Bearing stiffener

Step 3 : Check for load carrying capacity

The bearing stiffener acts as a column
Actual length of bearing stiffener
= (250 – 2 ×18)
= 2464 mm
Effective length of column
l = 0.7 × 2464
= 1724.8 mm
Cross-sectional area of column section
A = [4 × 37.52 + (40 × 0.8) × 0.8] × 100
= 17568 mm2
The moment of inertia of column section about the centre line of web,
1
I = [4 × 823.5 + 4 × 37.52(4.76 + l.8 + 0.4)2 + × 0.83] 104 mm4
2
= 10595.36 × 104 mm4
The radius of gyration of column section about the center line of web,
1
1
⎛ 10595 ⋅ 36 × 104 ⎞ 2
r = ⎛⎜ I ⎟ = ⎜
⎞2 ⎟
⎝A⎠ ⎝ 17568 ⎠
= 77.66 mm
Slenderness ratio of column section
⎛I⎞ ⎛ 1724 ⋅ 8 ⎞
⎜ ⎟ = ⎜ ⎟
⎝r⎠ ⎝ 77 ⋅ 66 ⎠
= 22.23
From IS : 800–1984, allowable stress in axial compression, for the steel
having yield stress as 250 N/mm2
σ = 147.331 N/mm2
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 533

Load carrying capacity of stiffener

⎛ 147 ⋅ 331 × 17568 ⎞
⎜ ⎟ = 2588.31 kN > 1500 kN
⎝ 1000 ⎠
(Support reaction)
Hence design of bearing stiffener is safe.

H o rizon ta l
stiffe ne r

IS A 9 0 m m
2 2 m m rive ts x 60 m m x 8 m m
p = p itch
H o rizon ta l
stiffe ne r

N e utral axis IS A 9 0 m m
x 60 m m x 6 m m

B e aring Ve rtica l
stiffe ne r stiffe ne r

4 IS A 1 50 m m IS A 9 0 m m
x 11 5 x 1 5 m m x 85 m m x 8 m m

Fig. 8.24 Details of bearing and vertical stiffeners

Step 4: Connection of bearing stiffener to web plate

Use 22 mm diameter power driven rivets
Strength of rivet in double shear
⎛ 2
π (23 ⋅ 5 ) × 100 ⎞
⎜2× × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of rivet in bearing
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 8 × ⎟ = 56.4 kN
⎝ 1000 ⎠
Rivet value, R = 56.4 kN
Number of rivets required to transmit reaction
⎛ 1500 ⎞
= ⎜ ⎟ = 26.59
⎝ 56 ⋅ 4 ⎠
534 DESIGN OF STEEL STRUCTURES

The filler plates are provided on both the sides of web plate.
Thickness of filler plate = 18 mm
The filler plate (packings) are properly fitted with the bearing stiffeners.
These filler plates are subjected to direct compression only.
Provide 30 rivets in 2 rows at pitch
p = 130 mm
Step 5: Design of Intermediate stiffeners
Clear depth between flange angles of plate girder
d = (2500 – 2 × 150) = 220 mm
Thickness of web tw = 8 mm
In case, the web plate is to be unstiffened, the minimum thickness of web
needed is found as under. Calculated average stress in the web plate due to
shear force.
⎛ 1500 × 100 ⎞
τva.cal = ⎜ ⎟ = 75.00 N/mm2
⎝ 2500 × 8 ⎠

⎛ d1⋅τva ⋅cal1 / 2 2200 × 751 / 2 ⎞

(i) t w.min = ⎜ = ⎟
⎝ 816 816 ⎠
= 23.34 mm, or

⎛ d2⋅ fy1 / 2 2200 × 2501 / 2 ⎞

(ii) t w.min = ⎜ = ⎟
⎝ 1344 1344 ⎠
= 25.88 mm, or
⎛ d1 2200 ⎞
(iii) t w.min = ⎜ = ⎟ = 25.88 mm
⎝ 85 85 ⎠
Actual thickness of the web 8 mm is less than the above values of t w.min, as
such the vertical stiffeners are required.
In case, the vertical stiffeners are used only, then, the thickness of web
required is as under (d2 = 2200 mm).

⎛ d2⋅ fy1 / 2 2200 × 2501 / 2 ⎞

(i) t w.min = ⎜ = ⎟ = 10.870 mm
⎝ 3200 3200 ⎠

⎛ d2 2200 ⎞
(ii) t w.min = ⎜ = ⎟ = 11 mm
⎝ 200 200 ⎠
Actual thickness of the web 8 mm is less than t w.min, then, both, the vertical
and horizontal stiffness become necessary.
Therefore, thickness required shall be as below, (d2 = 2200 mm)

⎛ d2⋅ fy1 / 2 2200 × 2501 / 2 ⎞

(i) t w.min = ⎜ = ⎟ = 8.696 mm
⎝ 4000 4000 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 535

⎛ d2 2200 ⎞
(ii) t w.min = ⎜ = ⎟ = 8.8 mm
⎝ 250 250 ⎠
Since, actual thickness of web 8 mm is still less than t w.min, a horizontal stiff-
ener is also necessary at the neutral axis, in which case, the minimum thick-
ness of web needed is as follows :
(d2 = 2200 mm)
⎛ d2⋅ fy1 / 2 2200 × 2501 / 2 ⎞
(i) t w.min = ⎜ = ⎟
⎝ 6400 6400 ⎠
= 5.435 mm
⎛ d2 ⎞
(ii) t w.min = ⎜ ⎟ = 5.5 mm
⎝ 250 ⎠
Therefore, the web of 8 mm thickness has to be stiffened using vertical and
2
horizontal stiffeners at a distance from the compression equal to th of the
5
⎛2 ⎞
distance from the compression flange to the neutral axis ⎜ ⋅ 1100 = 440 mm⎟
⎝5 ⎠
and also at the neutral axis of the plate girder.
Step 6 : Design of vertical stiffeners
At support, shear force = 1500 kN
Actual average shear stress in web plate
⎛ 1500 × 1000 ⎞
τva.cal = ⎜ ⎟
⎝ 2500 × 8 ⎠
= 75 N/mm2
d ⎛ 2200 ⎞
Ratio, = ⎜ ⎟ = 275
tw ⎝ 8 ⎠

The smaller clear panel dimensions for the actual thickness of web
= 180 × 8 = 1440 mm
The greater clear panel dimension for the actual thickness of web
= 270 × 8 = 2160 mm
The vertical stiffeners may be provided at spacing smaller than 1440 mm.
Let the spacing of vertical stiffeners be
= (0.6 × d) = 0.6 × 2200 = 1320 mm
From IS: 800 –1984, Table 6.6 (a), the permissible average shear stress, τva in
d
the stiffened web plate of steel with fy = 250 N/mm2 and 0.6 d spacing and
tw
ratio
τva = 81 N/mm2 > (τva.cal = 75 N/mm2)
536 DESIGN OF STEEL STRUCTURES

Length of outstanding leg of the vertical stiffener

⎛ 1 ⎞
⎜ × clear dept of girder + 50 ⎟ mm
⎝ 30 ⎠

⎛ 2200 ⎞
⎜ + 50 ⎟ = 123.33 mm
⎝ 30 ⎠
Provide ISA 125 mm × 95 mm × 8 mm (ISA 125 95, @ 0.133 kN-m). The
length of outstanding leg of the angle section is 90 mm.

IS A 1 25 m m × 95 m m × 8 m m

W eb 8 m m th ick

1 26 0 m m
IS A 1 25 m m × 95 m m × 8 m m

1 32 0 m m
Fig. 8.25

Clear distance between vertical stiffeners

c = (1320 – 60) = 1260 mm
Depth of plate girder = 2500 mm
Minimum required thickness of web
⎛ 1500 × 1000 ⎞
tw1 = ⎜ ⎟ = 7.41 mm
⎝ 2500 × 81 ⎠
Required moment of inertia,
⎛ 1 ⋅ 5d 3tw3 ⎞ ⎛ 1 ⋅ 50 × 22003 × 7 ⋅ 413 ⎞
I = ⎜ 1
⎟⎠ = ⎜ ⎟
⎝ c2 ⎝ 12602 ⎠
= 409 × 104 mm4
Moment of inertia about the face of web plate provided
(266 + 3.802 × 1698) × 104 = 511 × l04 mm4.
Step 7: Connection of vertical stiffener to web plate

⎛ 125t 2 ⎞ ⎛ 125 × 82 ⎞
Shear force = ⎜ ⎟ =⎜ ⎟
⎝ h ⎠ ⎝ 90 ⎠
= 88.89 kN/m
Use 22 mm diameter power driven field rivets
Strength of power driven rivet in single shear
⎛π ( 2 100 ⎞
⎜ × 23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝ 4 1000 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 537

Strength of rivet in bearing

⎛ 300 ⎞
⎜ 23 ⋅ 5 × 8 × ⎟ = 56.4 kN
⎝ 1000 ⎠
Rivet value, R = 43.35 kN
⎛ 43 ⋅ 35 ⎞
Pitch of rivets = ⎜ ⎟ = 0.487 m = 487 mm
⎝ 88 ⋅ 89 ⎠
Provide rivets at a pitch of 200 mm
Provide ISA : 125 mm × 95 mm × 8 mm (ISA 125 95, @ 0.133 kN/m) and 22
mm rivets to connect the stiffener with the web at 200 mm pitch. The vertical
stiffeners are shown in Fig. 8.24.

8.19 WEB SPLICE

A joint in the web plate provide to increase the length is known as web splice.
The plates are manufactured upto a limited length. When the maximum
manufactured length of plate is insufficient for full length of plate girder, web
splice becomes essential. It also becomes essential when the length of plate
girder is too long to handle conveniently during transportation and erection.
As far as possible, web splices may be located at sections where excess flange
areas are available. The excess flange areas are available at sections prior to
the curtailment of flange plates. Preferably, splices may be located under
stiffeners. The stiffeners provide additional strength to the splice. The splices
should not be located at the sections, where maximum bending moments occur.
In case, only one splice may suffice for full length of the girder, it may be located
at such sections.
The web splice are designed to resist the shears and moments at the spliced
sections. The splice plates are provided on each side of the web. There are
following three types of web splices, which are commonly used :
1. Rational splice
2. Moment splice
3. Shear splice.

8.19.1 Web Splice (Rational Splice)

This type of web splice is shown in Fig. 8.26. The stresses are transmitted directly
in this type. This type is most satisfactory than other two types. The splice
plates A as shown in Fig. 8.26 are provided between flange angles. A clearance
of 6 mm is left between splice plates and flange angles. The splice plates B are
provided for portion of web underneath the flange angles. If sufficient excess
flange area is available at the splice section, the splice plates B need not be
provided. These plates are designed for shear and moment, which would be
resisted by the portion of the web, if the web was not spliced. The rivets are
provided at uniform spacing in this type. The pitch of rivets connecting splice
plates to the web is found as under:
538 DESIGN OF STEEL STRUCTURES

Vertical shear per unit depth

⎛V ⎞
τv.cal = ⎜ ⎟ ...(i)
⎝ de ⎠
The bending stress upto the level of rivets connecting flange angles,
⎛M ⎞
σb.cal = ⎜ ⎟ × y1 ...(ii)
⎝ I ⎠
where, M = Bending moment at the splice section
l = Moment of inertia of the girder
y1 = Distance upto the level of rivets connection flange angles from
neutral axis.

6 mm
C le a ran ce

W ed splice ( ra tio n al sp lice m e tho d)

Fig. 8.26 Web splice

The horizontal force per unit length due to moment

= (σbc. t w)
If the rivets are provided in one vertical row, and p is the pitch of rivets, the
resultant of vertical and horizontal forces per pitch should not exceed the rivet
value, R.
2 1/ 2
⎡ 2 ⎛V ⎞ ⎤
Then p × ⎢(σ cal × tw ) + ⎜ ⎟ ⎥ = R
⎢⎣ ⎝ de ⎠ ⎥⎦

R
∴ p = 1/ 2
...(8.23)
⎡ 2
2 ⎛V ⎞ ⎤
⎢(σb⋅cal × tw ) + ⎜ ⎟ ⎥
⎣⎢ ⎝ de ⎠ ⎦⎥

If the computed pitch of rivets is less than minimum pitch, rivets are provided
in two or three vertical rows. The rivets are provided at spacing of twice pitch
computed above, if rivets are provided in two vertical rows. The rivets are
provided at spacing of three times the pitch computed above if rivets are provided
in three vertical rows. The thickness of splice plates A is kept equal to half the
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 539

thickness of web, but not less than 6 mm. The width of splice plates A is kept
sufficient to accommodate the rivets.
The area of portion of web beneath the flange angles is found. The splice
plates B provide this area. The rivets connecting splice plates B to the flange
angle area designed to resist horizontal shear per pitch length between splice
plates and angles, and horizontal force in the portion of web beneath flange
angles due to moment.
The horizontal shear force per pitch length

⎛V Af ⎞
P1 = ⎜ × × p⎟
d 1
⎜⎜ e Af Aw ⎟⎟
⎝ 6 ⎠
The pitch of rivets is assumed
Af = Gross area of flange excluding web equivalent
p = Pitch of rivets, which one is assumed
The horizontal force in the portion of web beneath flange angles due to
moment
M ⋅ y ⎛ Area of portion of web⎞
P2 = ×⎜
I ⎝ beneath flange angles ⎟⎠
where, y is the distance of rivets connecting splice plates B to the flange angles
from the neutral axis.
If n is the number of rivets required
n . R = n . P1 + P2
P2
n = ...(8.24)
( R – R1 )
The rivets connecting splice plates B to flange angles are provided at close
spacing, so that their length is small.
Example 8.6 The bending moment and shear force at a particular section of
a plate girder are 5760 kN-m and 1080 kN respectively. The cross-section of
plate girder at that section is as shown in Fig. 8.27. Design the web splice. 22
mm power driven rivets are used.
Solution
Design :
Step 1: Effective depth of plate girder
Gross flange area = 25517 mm2
Net flange area = 21688 mm2
Let y be the distance of compression flange from top

⎡ 2 × 50 ⋅ 25 (1 ⋅ 6 + 3 ⋅ 27 ) + 1 ⋅ 6 × 5 ⋅ 5 × 0 ⋅ 8 ⎤
y =⎢ ⎥ × 10 = 32.1 mm
⎣ 2 × 50 ⋅ 25 + 1 ⋅ 6 × 55 ⎦
Effective depth of the plate girder
de = (253.2 – 2 × 3.21) × 10 = 2467.8 mm
540 DESIGN OF STEEL STRUCTURES

Step 2: Gross moment of inertia of the plate girder about xx-axis.

⎡1 3
Ixx = ⎢ × 1 ⋅ 6 × (250) + 4 × 969.2 + 4 × 50 ⋅ 25
⎣12
2
× (125 – 3 ⋅ 72) + 2 × 55 × 1 ⋅ 6 × 125 ⋅ 82 ⎤⎦ × 104 mm4
Ixx = 7829002.1 × 104 mm4
Step 3 : Area of flanges
The short legs of flanges are connected with the web plate. From ISI Hand-
book No. 1, the rivets are provided at 90 mm from top of the flange angles.

5 50 m m
16 m m

2 IS A
2 00 m m × 1 50 m m
× 15 m m

2 50 0
mm

16 m m

2 IS A
2 00 m m × 1 50 m m
× 15 m m
16 m m

Fig. 8.27

Flange Area
Description Gross Area Deduction for Net flange area
(mm)2 rivet hole (mm)2 (mm)2
Flange angles 2 × 5025 4 ×23.5 ×15
2 ISA 200 mm = 10050 = 1410 8640
× 150 mm × 15 mm
Flange plates 550 × 16 2 × 23.5 × 16
550 mm × 16 mm = 8800 = 752 8048
1 2500 × 16 1 2500 × 16
Equivalent web area Aw = — Aw =
6 6 8 8
= 6667 = 5000
Total 25517 21688
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 541

The distance upto the level of rivets connecting flange angles to web
(1250 – 90) = 1160 mm
Step 4: Bending moment
M = 5670 kN-m
Step 5 : Bending stress at the level of rivets connecting flange angles
6
⎛ 5760 × 10 × 1160 ⎞
sbc.cal = M ⋅ y1 = ⎜ ⎟
I ⎝ 7829002 ⋅ 1 ⎠
xx
= 85.3 N/mm2
Thickness of web = 16 mm
Step 6: Design of riveted connection
Use 22 mm diameter power driven rivets
Strength of rivets in double shear
⎛ 2
π (23 ⋅ 5 ) × 100 ⎞
⎜2× × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of rivet in bearing
⎛ 23 ⋅ 5 × 16 × 300 ⎞
⎜ ⎟ = 112.8 kN
⎝ 100 ⎠
Rivet value, R = 86.70 kN
Let p be the pitch of rivets, in case rivets are used in one row
R
p = 1/ 2
⎡ 2
2 ⎛V ⎞ ⎤
⎢(σ × tw ) + ⎜ ⎟ ⎥
⎣⎢ ⎝ de ⎠ ⎦⎥

86 ⋅ 70 × 1000
p = 1/ 2
⎡ 2
2 ⎛ 1080 × 1000 ⎞ ⎤
⎢(85 ⋅ 2 × 16 ) + ⎜ ⎟ ⎥
⎣ ⎝ 2467 ⋅ 8 ⎠ ⎦
= 60.49 mm
Provide rivets in rows at 165 mm pitch, using 13 rivets in each row.
Depth of splice plates A = (2200 – 2 × 6) = 2188 mm
1 1
Thickness of splice plate A = tw = × 16 = 8 mm
2 2
Width of splice plates A to accommodate rivets = 423.0 mm
This provides distance between adjacent rows of rivets = 3d and edge distance
= 1 .5 d, where d is the gross diameter of rivets used.
Step 7 : Splice plates B
Area of web plate under flange angles = (150 × 16) = 2400 mm2
Provide 130 mm wide plates on both sides
⎛ 2400 ⎞
Thickness of plates = ⎜ ⎟ = 9.23 mm
⎝ 130 × 2 ⎠
542 DESIGN OF STEEL STRUCTURES

Provide two plates 130 mm ×10 mm

The horizontal shear force per pitch length between flange angles and splice
plates
⎛V Af ⎞
P1 = ⎜ × × p⎟
1
⎜⎜ de Af Aw ⎟⎟
⎝ 6 ⎠
Assume pitch of rivets p as 60 mm (p > 2.5 d minimum pitch)
⎛ 1080 × 1000 18850 60 ⎞
P1 = ⎜ × × ⎟
⎝ 2467 ⋅ 8 18850 + 6667 1000 ⎠
= 19.40 kN
The horizontal force in the portion of web beneath the flange angles due to
moment
M
P2 = ⋅ y × ( Area of web beneath flange angles )
I
⎛ 5760 × 106 1160 × 2400 ⎞
or = ⎜ × ⎟ = 204.83 kN
⎝ 7829002 ⋅ 1 104 × 1000 ⎠
Number of rivets required with 60 mm pitch, from Eq. 8.24
⎛ P2 ⎞ ⎛ 204.50 ⎞
n = ⎜ ⎟ = ⎜ ⎟ =4
⎝ R1 – P1 ⎠ ⎝ ( 86 ⋅ 70 – 19 ⋅ 40 ) ⎠
The length of splice plates is 480 mm
The complete web splice is shown in Eq. 8.28.

90 m m
10 m m
6 mm
22 m m
R ive ts
11 60 @ 1 65
m m m m pitch

2 19 8
mm

1 3 N o s.
in a ro w

8 mm
Th ick

6 mm
10 m m

4 23 m m
4 80 m m

Fig. 8.28 Web splice (Rational Splice method)

 Design of OF
DESIGN Beam Connection
PLATE GIRDER 543

8.19.2 Web Splice (Moment Splice)

This type of web splice is shown in Fig. 8.29. There are four moment plates, (two
on each face), marked as splices A plates and two shear plates (one on each
face), marked as splices plate B. It is assumed that moment plates resist
moment resisted by web, and shear plates resist shear resisted by web. In fact,
each set of plate resist shear as well as moment. But in case of deep girders,
shear resisted by splice plates A is small compared with plates B. Similarly the

Sp lice
p la tes
A

d1
Sp lice p la tes B

Sp lice
p la tes
A

Fig 8.29 Web splice (Moment splice method)

moment resisted by splice plates B is small compared with splice plates A. This
type of splice may be used for girders about 2 metres deep. A clearance of 6 mm
is provided between splice plates A and flange anges, and between splice plates
A and splice plates B. The web splice (moment splice) is designed as under :
The moment resisted by the web plate is as under

⎛ Iw ⎞
Mw = ⎜ ×M⎟ ...(8.25)
⎝ I ⎠
where, Iw = Gross moment of inertia of the web
I = Gross moment of inertia of the girder
M = Bending moment at the web splice section

8.19.2.1 Moment Plates

The moment of resistance of four moment plates (splice plates A) is equal to
moment resisted by web.
The moment of resistance of four moment plates is As. σ1. d1
544 DESIGN OF STEEL STRUCTURES

where, As = Net area of two moment plates

d1 = Distance between centre to centre of splice plates A (Moment
plates)
σ1 = Bending stress at the centre of splice plate A, and assumed
uniform in these plates.
Therefore Mw = As. σb1. d1 ...(i)
⎛d ⎞
σb1 = ⎜ 1 × σbc ⎟ ...(ii)
⎝ dw ⎠
where, σbc.cal is the bending stress at the extreme fibre of the web plate
⎛ M ⎞
σbc.cal = ⎜ ⎟ ...(iii)
⎝ An⋅dw ⎠

⎛ d ⎞
∴ Mw = ⎜ As ⋅ d1 ⋅ 1 σbc⋅cal ⎟
⎝ d w ⎠

⎛ M w ⋅ dw ⎞
As = ⎜ 2⎟
...(8.26)
⎝ σbc⋅cal ⋅ d1 ⎠
Let t s be the thickness of the moment plate
1 As
ts = ⋅ ...(8.27)
2 (ds – n1d )
where, ds = Depth of moment plate
n1 = Number of rivets in one vertical row
d = Diameter of rivets.
The horizontal force in the two moment plates A
= (A1 × σb1)
The number of rivets required to connect moment plates A to the web plate
on each side of web splice is given by
⎛ σb1 ⎞
n = ⎜ As ⋅ ⎟ ...(8.28)
⎝ R ⎠
where, R = Rivet value.

8.19.2.2 Shear Plates

The shear plates B resist shear at the web splice section. The combined thick-
ness of these plates is designed to resist shear at web splice section. The width
of the splice plates B is kept sufficient to accommodate rivets. The number of
rivets, required to resist shear
⎛V ⎞
n = ⎜ ⎟
⎝R⎠
where, V is the shear at the web splice and R is the rivet value.
These rivets are provided on each side of the web splice.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 545

Example 8.7 Design the web splice given in Example 8.6.

Solution
Design :
Step 1: Moment resisted by web plate
Gross moment of inertia of plate girder
Ixx = 7829002.l × l04 mm4
Gross moment of inertia of web plate
Iw = 2083334.3 × 104 mm4
The bending moment resisted by the web plate from Eq. 8.25

⎛ Iw ⎞ ⎛ 2083333 ⋅ 3 ⎞
Mw = ⎜ ×M⎟ = ⎜ × 5760 ⎟
⎝ I ⎠ ⎝ 7829002 ⋅ 1 ⎠
= 1535 kN-m
The depth of web plate, dw is 2500 mm
Net flange are Afn = 21688 mm2
Step 2: Moment plates
The depth of moment plates is kept sufficient to accommodate rivets in 3
rows. 22 mm diameter rivets are used at 100 mm spacing centre to centre and
50 mm edge distance.
The distance between centre to centre of moment plates,
d1 = (2500 – 2 × 152 – 2 × 150 – 2 × 6)
= 1888 mm
The bending stress at the extreme fibre of web plate,
⎛ 5760 × 1000 × 1000 ⎞
σbc.cal = ⎜ ⎟
⎝ 21688 × 2500 ⎠
= 106.23 N/mm2
The net area of two moment plates from Eq. (8.22)

⎛ M w ⋅ dw ⎞ ⎛ 1535 × 106 × 2500 ⎞

As = ⎜ ⎟ = ⎜⎝ ⎟
2
⎝ sbc ⋅cal d1 ⎠ 106 ⋅ 23 × 18882 ⎠
= 10134.38 mm2
The thickness of moment plates as per Eq. (8.23)
1 ⎛ As ⎞
ts =
2 ⎜⎝ ds – n1d ⎟⎠

1 ⎛ 10134 ⋅ 38 ⎞
= ×⎜ ⎟
2 ⎝ (300 – 3 × 23 ⋅ 5 ) ⎠
= 22.08 mm
Provide 18 mm thick plates.
Step 3 : Design of riveted connections
Strength of power driven rivets in double shear
546 DESIGN OF STEEL STRUCTURES

⎛ 2
π (23 ⋅ 5 ) × 100 ⎞
⎜2× × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of rivets bearing
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 18 × ⎟ = 126.9 kN
⎝ 1000 ⎠
Rivets value, R = 86.70 kN.
Number of rivets required to connect moment plates to web
⎛ 1800 ⎞
Bending stress σb1 = ⎜106 ⋅ 23 × ⎟
⎝ 2500 ⎠
= 80.22 N/mm2
⎛ As × σb1 ⎞ ⎛ 18 × 2 (300 – 3 × 23 ⋅ 5 ) ⎞ 80 ⋅ 22
From Eq. 8.28, n = ⎜ ⎟ =⎜ ⎟×
⎝ R ⎠ ⎝ 86.70 ⎠ 1000
= 7.64
Provide 12 rivets on each side of web splice section
The distance between rows of rivets is 70 mm
Edge distance = 35 mm
Width of moment plates = 560.0 mm
Step 3 : Shear plates
6 mm clearance is kept between moment plates and shear plates.
Depth of shear plates = (2500 – 2 × 150 – 2 × 300 – 4 × 16) mm
= 1536 mm
Shear force at web splice = 1080 kN
Allowable shear stress = 0.4fy = 0.4 × 250 = 100 N/mm2
Thickness of two shear plates
⎛ 1080 × 1000 ⎞
⎜ ⎟ = 7.03 mm
⎝ 1576 × 100 ⎠
Thickness of one shear plates
= 3.515 mm
Provide 6 mm thick shear plate
Step 4: Design of riveted connection
Number of rivets required
⎛ 1080 ⎞
⎜⎝ ⎟ = 12.45
86.70 ⎠
Provide 14 rivets in two rows at 225 mm pitch.
Width of splice plate = 280.0 mm
The distance between rows of rivets in 70 mm
Edge distance = 35 mm
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 547

The complete web splice of the web plate is shown in Fig. 8.30.

6 mm 50 mm
C lea ran ce 3 00 1 00 mm
mm 1 00 mm
50 mm
11 5 m m
2 25 m m
2 25 m m
1 57 6 m m
2 25 m m
2 25 m m
2 25 m m 1 88 8 m m

2 25 m m
6 mm 11 5 m m
C lea ran ce 50 mm
3 00 1 00 mm
mm 1 00 mm
50 mm

2 80 m m
5 60 m m

Fig. 8.30 Web splice (moment splice method)

8.19.3 Web Splice (Shear Splice)

This type of web splice is shown in Fig. 8.31. The splice plates are provided
between flange angles. A clearance of 6 mm is left between splice plates and
flange angles. The web splice (shea splice) is designed as under :
The moment of resistance of splice plate is kept equal to moment of resistance
of web plates
⎛ 1 2⎞ ⎛ 1 2⎞
⎜ σb1 × × ts × (d ) ⎟ = ⎜ σbc × × tw × (dw ) ⎟
⎝ 6 ⎠ ⎝ 6 ⎠
where, σb1 is the bending stress at the extreme fibre of splice plate and σbc the
bending stress at the extreme fibre of web plate. From the triangular distribution
of bending stress,
⎛ d ⎞
σb1 = ⎜ σbc ⋅ e ⎟
⎝ d w ⎠
⎛ de 1 2⎞ ⎛ 1 2⎞
∴ ⎜ σbc ⋅ d ⋅ 6 ⋅ ts ⋅ (ds ) ⎟ = ⎜⎝ σbc ⋅ 6 ⋅ tw ⋅ (dw ) ⎟⎠
⎝ w ⎠
⎛ dw ⎞
Thus As = Aw ⋅ ⎜ ⎟ ...(8.29)
⎝ ts ⎠
where, As and ds = Area and depth of splice plates
548 DESIGN OF STEEL STRUCTURES

Aw and dw = Area and depth of web plates

6 mm
S p lice
p late ds

6 mm

W eb p la tes (sh ea r splice m e th od )

Fig. 8.31 Web plates

Total thickness of splice plates

⎛ A ⎞ ⎛ Area of the splice plate ⎞
ts = ⎜ s ⎟ = ⎜ ⎟
⎝ bs ⎠ ⎝ Width of the splice plate ⎠
The width of splice plate is kept sufficient to accommodate the rivets. The
splice plates are designed to resist shear and moment which would be resisted
by web, if web was not spliced
⎛ Iw ⎞
Mw = ⎜ × M⎟
⎝ I ⎠
where, Iw = Gross moment of inertia of the web
I = Gross moment of inertia of the girder
M = Bending moment at the splice section.
The vertical shear V at the splice section is shown in Fig. 8.32 (a). It acts at
an eccentric distance from C.G. of rivets, connecting splice plates to the web. It
is considered equivalent to a vertical force V and an eccentric moment Me = V ×
e as shown in Fig. 8.32 (b).
The splice plates resist a total moment Ms
Ms = (Me+Mw)

v
v M e = Vxe
e

C .G . of C .G . of C .G . of
rive ts rive ts rive ts

(a ) (b )
Fig. 8.32
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 549

The rivets connecting splice plates to the web are designed to resist a vertical
force V and a moment Me . This becomes design of eccentric riveted connections,
explained in Art. 2.27, Type 1.
Example 8.8 Design the web splice given in Example 8.6.
Solution
Design :
Step 1 : Area of splice plate
The depth of web plate, dw is 2500 mm
Depth of splice plates
de = (2500 – 2 × 150 – 2 × 6) = 2188 mm
Thickness of web plate, t w = 1.6 mm
Area of web plate
Aw = dw.tw
= (2500 × 16) = 40000 mm2
Area of splice plate required, from Eq. 8.25
2
⎛d ⎞
As = Aw ⎜ w ⎟
⎝ ds ⎠
2
⎛ 2500 ⎞
= 4000 × ⎜ ⎟ = 52221 mm2
⎝ 2188 ⎠
Step 2 : Total thickness of splice plates required
⎛A ⎞ ⎛ 52221 ⎞
ts = ⎜ s ⎟ = ⎜ ⎟ = 23.887 mm
⎝ ds ⎠ ⎝ 2188 ⎠
Provide two splice plates 12 mm thick one on each side of web plate. The
width of the splice plate is kept to accommodate three rows of rivets on either
side of splice section.
Step 3 : Width of splice plate
Use 22 mm power driven rivets
∴Width of splice plates, bs = (6 × 3d)
where, d the gross diameter of rivet
bs = (6 × 3 × 23.5) = 423.0 mm
From Example 8.6,
The gross moment of inertia of plate girder,Ixx is 7829002.1 × 104 mm4
The gross moment of inertia of web plate 2500 mm deep × 16 mm thick from
ISI Handbook No. 1.
Iw = 2083333.3 × 104 mm4
Bending moment at web splice section
M = 5760 kN-m
Bending moment resisted by the web plate
⎛ Iw ⎞ ⎛ 2083333 ⋅ 9 ⎞
Mw = ⎜ ×M⎟ = ⎜ × 5760 kN-m
⎝ I ⎠ ⎝ 782900 ⋅ 12 ⎟⎠
550 DESIGN OF STEEL STRUCTURES

Vertical shear at web splice section,

V = 1080 kN
Eccentricity from C.G. of group of rivets
e = (4.5 d) = (4.5 × 23.5) = 105.0 mm
Eccentric moment
⎛ 1080 × 105 ⋅ 5 ⎞
Me = ⎜ ⎟ = 113.4 kN-m
⎝ 1000 ⎠
Splice plates resist a total moment
Ms = (Me + Mw) = (113.4 + 1535)
= 1648.4 kN-m
Step 5: Design of riveted connections
Group of rivets resist a moment
= 1648.4 kN-m
and a vertical force = 1080 kN
Strength of power driven rivet in double shear
⎛ 2
π (23 ⋅ 5 ) × 100 ⎞
⎜2× × ⎟ = 86.70 kN
⎝ 4 1000 ⎠
Strength of power driven rivet in bearing
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 16 × ⎟ = 112.8 kN
⎝ 1000 ⎠
Rivet value, R = 86.70 kN
Let n be the number of rivets to be provided in one row. The rivets are pro-
vided in 3 rows. Assume pitch of rivets p in each row at 90 mm.
1/ 2 1/ 2
⎛ 6M s ⎞ ⎛ 6 × 1648 ⋅ 4 × 1000 ⎞
n = ⎜ =⎜ ⎟
⎝ (3 p) × R ⎟⎠ ⎝ 3 × 90 × 86 ⋅ 70 ⎠
= 20.55
Provide 24 rivets in each row.
Step 6: Check for forces in rivets
The rivets are arranged in a narrow strip. The vertical distance to the
extreme rivet, is as below.
yn = 1035 mm
∑ 2y = 6. [4.52 + 13.52 + 22.52 + 31.52 + 40.52 + 49.52
+ 58.52 + 67.52 + 76.52 + 85.52 + 94.52 + 103.52] × 100 = 6 × (46557.20 × 100) mm.
The force due to direct shear in each rivets is as under
⎛ 1080 ⎞
F1 = ⎜ ⎟ = 15 kN
⎝ 24 × 3 ⎠
Force in an extreme rivet resisting the moment
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 551

⎛ M ⋅ yn ⎞ ⎛ 1648 ⋅ 4 × 1000 × 1000 × 1035 ⎞

F2 = ⎜ =
2 ⎟ ⎜ ⎟
⎝ ∑ y ⎠ ⎝ 6 × 46557 ⋅ 20 × 100 × 1000 ⎠
= 61 kN
70 ⋅ 5
cos θ = = 0.068
(70 ⋅ 5
2
+ 103 ⋅ 502 )
Resultant of two forces
F = (F12 + F22 + 2F1F2 cos θ)1/2
= ((15)2 + (61)2 + 2 × 15 × 61 × 0.068)1/2
= 63.90 kN < 867.0 kN, rivet value.
Hence, the design of rivet is satisfactory.
The complete web splice is shown in Fig. 8.33.

6 mm

2 2 m m R ive ts
@ 9 0 m m p itch
2 4 N o s.
in a ro w
2 19 8 m m

2 S p lice
p lates
e ach 12 m m
th ick

6 mm

4 23 m m

Fig. 8.33 Web splice (shear splice method)

8.20 FLANGE SPLICES

A joint in the flange element provided to increase the length of flange angle or
flange plate is known as flange splice.
The flange splices should be avoided as far as possible. In general, the flange
angles and flange plates can be obtained for full length of the plate girder. In
spite of availability of full length of flange angles and flange plates, sometimes,
552 DESIGN OF STEEL STRUCTURES

it becomes necessary to make flange splices, for example, the transportation

facilities may not permit transportation of plate girder for the entire span as
one piece. The flange splices should be located at the section where some excess
of flange area is available and not at the points where web splice is done. It
should preferably not be located at points of maximum stress. Only one element
of the flange, viz., one flange angle or one plate should be spliced at one section.
Where splice plates are used, their area shall be not less than 5 percent in
excess of the area of flange element spliced. The centre of gravity of splice plates
should coincide as nearly as possible with that of the element spliced. The rivets
are designed for force 5 percent in excess to force in the element spliced. The
strength of splice should be equal to or greater than 50 percent of the strength
of splice elements.

8.20.1 Splice of Flange Angles

When one flange angle is spliced at a section, a single splice angle may be
provided in case it provides sufficient area. The splice angle is shaped at the
heel and made to fit in the fillet of flange angle. The splice angle should not
project beyond the flange angles specially in exposed positions. When a single
splice angle cannot provide required cross-sectional area, then two splice angles,
one on each side may be provided or one splice angle on one side and one plate
on the other side may be provided. Figure 8.34 shows splices of flange angles of
three types.

(a ) (b ) (c)
S ing le splice a n gle D o ub le splice a n gle S p lice a ng le a nd plate
Fig. 8.34 Flange angle splices

In Fig. 8.34 (a), a single splice angle is used for flange angle splice, and splice
angle is in direct contact with the flange angle. Force from the flange angle
shifts to splice angle on the same side of web. As a result of this, horizontal
shear between flange angle and web plate is not affected. The splice and is
connected with the flange angle with the same rivets and same spacing as used
for connecting flange angle with the web. In order to minimize the length of
splice angle, the rivets may be provided as less spacing. The length of splice
angle is kept sufficient to accommodate rivets required to transmit force from
flange to splice angle and back. In Fig. 8.34 (c), a single splice angle and a splice
plate is used for flange angle splice. The plate is not in direct contact with the
element spliced. The part of the force from flange angle shifts to opposite side of
the web through the web and unspliced flange angle. As a result of this, hori-
zontal shear between flange angle and web is affected. It requires additional
rivets for connections. The force in the flange angle is assumed to be distributed
to the elements of the splice in proportion to their cross-sectional area. The
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 553

strength to be transmitted by rivets connecting splice plate is equal to force in

splice plate plus force due to horizontal shear. Therefore,
(n . Rs = P + Horizontal shear)
where, n = Number of rivets required to connect splice plate on each side of
splice
P = Force to be carried by splice plate
Rs = Strength of rivet in single shear
The strength of rivets is considered single shear because P is force to be
carried in one plane only.
The horizontal shear per unit length

⎛V ⎞
τvf.cal = ⎜ ⎟
⎝ de ⎠
The horizontal shear per unit length in one plane
1 ⎛1 V ⎞
.τ = ⎜ ⎟
2 vf.cal ⎝ 2 de ⎠
The horizontal shear per unit length in one plane
⎛1 V ⎞
Fs = ⎜ × n ⋅ p⎟
⎝ 2 ds ⎠
where, p = Pitch of rivets
V = Shear force at the splice section
de = Effective depth of the girder
⎛ 1V ⎞
∴ n . Rs = ⎜ P + × n ⋅ p⎟
⎝ 2 ds ⎠

P
∴ n = ...(8.30)
⎛ 1V ⎞
⎜ Rs – 2 d ⋅ p ⎟
⎝ ⎠

8.20.2 Splice of Flange Plates

In case, it becomes necessary to splice an outer flange plate, a splice plate of
same cross-section as the plate is provided. The length of splice plate is kept
sufficient to accommodate necessary number of rivets. The strength of rivet is
found in single shear. The splice plate is in direct contact. Therefore, the force
to be transmitted by splice plate is not affected.
In case, it becomes necessary to splice an inner flange plate, the splice may
be located at the theoretical cut-off of the next outer plate. The outer plate is
extended. This serves as a splice plate. Splice of an inner plate is also done by
providing an extra plate, which is placed outside of all flange plates. In this,
splice plate is not in direct contact with the flange plate splices. The area of
554 DESIGN OF STEEL STRUCTURES

cross-section of splice plate is kept equal to the cross-sectional area of flange

plate cut. The rivets may be designed to take full strength of flange plate cut
and the shearing stress due to transmission of flange increment is neglected.
Example 8.9 A plate girder is composed of a 1600 mm × 12 mm web plate, 4
ISA 150 mm × 150 mm ×18 mm flange angles and two 500 mm × 12 mm thick
flange plates. The gross moment of inertia of plate girder is 2355120.2 × 104
mm4. The web of bending moment at a section is 2100 kN-m and shear force
1600 kN. One flange angle is spliced at this section. Design flange angle splice.
Provide a single splice angle.
Solution
Design :
Step 1: Design of splice of flange angle
Size of flange angle ISA 150 mm × 150 mm × 18 mm (ISA 150 150, @ 0.399
kN/m)
Use 22 mm diameter power driven rivets for connections
Net area of flange angle (5079 – 2 × 235 × 18) = 4233 mm2
∴ Add 5 per cent = 212 mm2
Total area required = 4445 mm2
From ISA 200 mm × 200 mm × 25 mm (ISA 200 200, @ 0.736 kN/m) cut a
splice angle 130 mm × 130 mm × 25 mm. The splice angle is cut and shaped to
clear root of flange angle, and it does not project beyond flange angle.
Gross area of splice angle
[13 × 25 × (13 – 2.5) × 2.5] = 58.75 × 100 mm2
Deduct 2 rivet holes
2 × 23.5 × 25 = 1175 mm2
Net area provided by splice angle
(5875 –1175) = 4700 > 4445 mm2
Strength of power driven rivet in single shear
⎛π ( 2 100 ⎞
⎜ × 23 ⋅ 5 ) × ⎟ = 43.35 kN
⎝4 1000 ⎠
Strength of rivets in bearing
⎛ 300 ⎞
⎜ 23 ⋅ 5 × 12 × ⎟ = 84.6 kN
⎝ 1000 ⎠
Rivet value = 43.35 kN
Net flange area = 16501 mm2 (See Table on page no. 555)
Gross flange area = 19358 mm2
The stress at the level of C.G. of flange angles
⎛ 2100 × 106 ⎞ ⎛ (812 – 12 – 43 ⋅ 8 ) × 19358 ⎞
σb = ⎜ 4⎟×⎜ ⎟ = 78.1 N/mm2
⎝ 2355120 ⋅ 2 × 10 ⎠ ⎝ 16501 ⎠
Force in flange angle
⎛ 78 ⋅ 1 ⎞
⎜ × 4445 ⎟ = 347.15 kN
⎝ 1000 ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 555

Number of rivets required

⎛ 347 ⋅ 15 ⎞
⎜ ⎟ = 8.008.
⎝ 43 ⋅ 35 ⎠
Step 3 : Area of flanges
Flange Area
Description Gross Area Deduct for Net area (mm)2
(mm)2 rivet hole (mm)2
Flange angle 2 × 5071 4 ×23.5 ×18
2 ISA 150 mm = 10158 = 1692 8466
× 150 mm × 18 mm
Flange Plate 500 × 12 2 × 23.5 × 16.5
550 mm × 12 mm = 6000 = 565 5635
1 1
Equivalent web area × 1600 × 12 — × 1600 × 12
6 8
= 3200 = 2400
Total 19358 16501
Provide 10 rivets on each side of flange angle, 5 rivets are provided in each leg at
60 mm spacing. The length of splice angle
= 60 × 10 = 600 mm
The complete flange splice is shown in Fig. 8.35.

Flan ge P P P P
S p la ce 2 2 2 2
a ng le a ng le
IS A cut size 6 00 m m
1 50 m m x 1 30 m m x p = 60 m m
1 50 m m x 1 30 m m x
18 m m 25 m m
Sing le an g le sp lice (fla ng e a n gle sp lice )

Fig. 8.35 Flange angle splice

Example 8.10 Design flange angle splice in Example 8.9. Provide a splice
angle and a rectangular plate.
Solution
Design :
Step 1: Design of splice angle
From Example 8.9
Net area of flange angle ISI 150 mm × 150 mm × 18 mm (ISA 150 150, @
0.399 kN/m)
= 4233 mm2
556 DESIGN OF STEEL STRUCTURES

Add 5 per cent = 212 mm2

Total area of splice elements required
= 4445 mm2
Provide ISA 130 mm × 130 mm × 15 mm (ISA 130 130, @ 0.289 kN/m) as
splice angle
The splice angle does not project beyond flange angle
New area provided by splice plate
(36.81 – 2 × 2.35 × 1.5) × 100 = 29.76 × 100 mm2
Step 2 : Design of splice plate
Area required from splice plate
= (4445 – 2986) = 1469 mm2
Provide 120 mm × 16 mm splice plate
Net area provide by splice plate
= (120 × 16 – 23.5 × 16) = 1544 mm2
Total area provide by splice elements
= (2976 + 1544) = 4520 mm2
From Example 8.9, the stress at the level of C.G. of flange angle
σb = 78 1 N/mm2
Force to be transmitted by flange angle
= (78.1 × 4445) = 347.15 kN
Force transmitted by splice angle
⎛ 347.15 × 2976 ⎞
⎜⎝ ⎟⎠ = 228.56 kN
4520
Rivets required to connect splice angle
⎛ 228.56 ⎞
n = ⎜⎝ ⎟ = 5.27
43.35 ⎠
Provide 6 rivets on each side of flange splice, 3 rivets in each leg at 6 mm
spacing. The force transmitted by splice plate
⎛ 347.15 × 1544 ⎞
p = ⎜⎝ ⎟⎠ = 118.584 kN
4520
Step 3: Rivets required to connect splice plate
The distance of C.G. of compression flange from outside
⎛ 50 × 1.2 × 0.6 + 2 × 50.79 × (1.2 + 4.38) ⎞
y = ⎜⎝ × 10⎟ = 37.1 mm
50 × 1.2 + 2 × 50.79 ⎠
Effective depth of the plate girder
de = (1624 – 2 × 37.1) = 1549.8 mm
The number of rivets needed for splice plate, from Eq. 8.26
P
n =
⎛ 1 V ⎞
⎜ Rs – 2 × d ⋅ p ⎟
⎝ e ⎠
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 557

118 ⋅ 584
n =
⎛ 1 1600 × 1000 ⎞
⎜ 43 ⋅ 35 – × × 600 ⎟
⎝ 2 1549 ⋅ 8 ⎠
= 12.11 mm
Provide 14 rivets on each side of flange splice. The length of splice plate.
= 28 × 1680 mm
The complete flange splice is shown in Fig. 8.36.

Flan ge S p la ce 1 68 0 m m
a ng le p la te
IS A 1 4 rivets on ea ch sid e of fla n ge splice
1 30 m m × 1 20 m m × a t 6 0 m p itch
1 30 m m × 16 m m
15 m m

Fig. 8.36 Flange angle splice

Example 8.11 It is required to splice flange plate only in Example 8.9. De-
sign splice for flange plate.
Solution
Design :
Step 1: Force in flange plate
From Example 8.9, net area of flange plate
= 5635 mm2
Add 5 per cent = 282 mm2
Net area required by splice element
= 5197 mm2
Provide 50 mm × 14 mm splice plate
Net area provided = (50 × 1.4 – 2 × 2.35 × 1 6) ×100
= 6245 mm2 > 5917 mm2
The stress in spliced flange plate
⎛ 2100 × 106 × 19558 ⎞
⎜ ⎟ = 84.31 N/mm2
⎝ 2355120 ⋅ 2 × 104 × 16501 ⎠

⎛ 84 ⋅ 31 × 5635 ⎞
Force in flange plate = ⎜ ⎟ = 475.1 kN
⎝ 100 ⎠
Force 5 percent in excess of force in flange plate
= 1.05 × 475.1 = 498.855 kN
Step 2 : Design of riveted connections
Rivet value, R = 43.35 kN
558 DESIGN OF STEEL STRUCTURES

The number of rivets required

⎛ 498 ⋅ 855 ⎞
⎜ ⎟ = 11.50
⎝ 43 ⋅ 35 ⎠
Provide 12 rivets in two rows on each side of flange plate splice at 6 mm
spacing. The length of splice plate
= 12 × 60 = 720 mm
The complete flange plates splice is shown in Fig. 8.37.

P P P P
2 2 2 2

Sp lice p la te 14 m m th ick pitch o f

rive ts 60 m m

7 20 m m

Fig. 8.37 Flange splice

PROBLEMS
8.1. A plate girder simply supported at ends is composed of a web plate 1250
mm depth × 16 mm thickness, and, two flange angles ISA 200 mm 100 mm
× 15 mm (ISA 200 100, @ 0.336 kN/m wide and two flange plates 550 mm
wide × 16 mm thickness in each flange. The effective span of the plate
girder is 15 m. The diameter of power driven rivets used for connecting
flange plates to flange angles and flange angles to web plate is 22 mm.
Determine the maximum uniformly distributed load (inclusive of self
weight) which can be carried by the plate girder.
8.2. A plate girder, having an effective span 12.20 m is required to carry a
uniformly distributed load of 90 kN/m inclusive of its own weight, ends
being simply supported. Compression flange has got adequate lateral
support. Design the plate girder and draw :
(i) An elevation showing the plate girder with details
(ii) A plan of the flange showing curtailment of flange plates
(iii) A cross-section showing details of the make-up.
8.3. Design the central section of a plate girder for an effective span of 20 metres
if the dead and live loads amount to 30 kN/m and 60 kN/m, respectively.
Show the curtailment of flanges on a diagram.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 559

8.4. A plate girder consists of a web plate 1000 mm × 10 mm, flange angles 100
mm × 100 mm × 12 mm (ISA 100 100, @ 0.177 kN-m) and cover plates 300
mm wide by 1.20 mm thick (one each at top and bottom). Design rivets
connecting the flanges angles with the web and the cover plates at a sec-
tion subjected to 730 kN shear.
8.5. A simply supported plate girder spans 13.60 m and carries a uniformly
distributed load of 80 kN/m inclusive of self weight of girder. The plate
girder consists of 1600 mm deep × 12 mm thick web plate, 2 ISA 200 mm ×
100 mm ×15 mm (2 ISA 200 100, @ 0.336 kN/m) flange angles and 500 mm
× 12 mm cover plate in each flange. Design end bearing stiffeners, and
necessary intermediate stiffeners.
8.6. Design a web splice for a plate girder fro the following data : Web plate :
2000 mm ×12 mm
Flange angles : ISA 200 mm × 100 mm × 15 mm (ISA 200 100, @ 0.336 kN/m)
Net area of tension flange : 24642 mm2
Gross area of compression flange : 28556 mm2
Maximum bending moment at section : 6900 kN-m
Corresponding shear force at section : 1380 kN
Flange plates in each flange : 2 × 500 mm × 16 mm.
8.7. A plate girder has an effective span of 16 m and is simply supported at its
ends. It carries a uniformly distributed load of 100 kN/m exclusive of its
self-weight. Design (i) the central section (ii) the rivets connecting the
components, using at least two flange plates on each flange.
Determine the theoretical points of cut-off for the flange plates of the plates
girder.
8.8. A riveted plate girder section is designed to carry a uniform load of 50 kN/
m excluding the self-weight of the girder. In additions, the girder has to
support two concentrated loads of 400 kN each at one-third points. The
effective span is 18 m. The following section has been provided.
Web plate 2000 mm × 6 mm
Flange angles ISA 150 mm × 150 mm × 12 mm (ISA 150 150, @ 0.272 kN/m)
Flange plates 2 Nos. 400 mm × 12 mm
Design the bearing stiffeners to be provided at the supports and the verti-
cal intermediate stiffeners.
8.9. A plate girder section consists of the following components
Web plate 1500 mm × 10 mm
Flange angles ISA 150 mm × 150 mm × 10 mm (ISA 150 150, @ 0.228 kN/m)
Flange plates 2 Nos. 500 mm × 10 mm on each flange
If the end reaction produced by the applied loads is 1000 kN design the
bearing stiffener for it.
For the plate girder above, design all the necessary intermediate stiffeners.
560 DESIGN OF STEEL STRUCTURES

8.10. It is required to design a suitable mid-span section of built-up plates girder,

supporting a uniformly distributed load of 90 kN/m including its own
weight.
The girder has a simply supported span of 23 m.
Use 22 mm diameter power driven rivets.
8.11. A plate girder of span 20 m has to carry a uniformly distributed load of 90
kN/m inclusive of its own weight. Design the maximum section of the girder
if the web depth is not to exceed 1600 mm, calculate the spacing of 22 mm
diameter power driven rivets connecting the flange angles and the flange
plates near the ends.
8.12. A simply supported plate girder having a span of 4 metres has to support
floor beams that frame at 2 metres centre to centre as showfi in Fig. P.8.12.

7 @ 2 m = 14 m

Fig. P.8.12

Each floor beam introduces a concentrated load of 1000 kN on the girder.

In addition, the girder has to carry a uniformly distributed load of 18 kN/
m including its own weight. The top flange of the girder is to be restrained
effectively and the girder is to be provided with vertical stiffeners. Assum-
ing that the depth of web plate is to be 1000 mm and using the data given,
find (a) the suitable thickness of the web plate, (b) the section of the flange
at the centre of span. Show by a dimensioned sketch the cross-section of
the girder at the centre span.
8.13. Evaluate the girder cross-section of a riveted plate girder to carry a super
load of 120 kN per metre on a simple span of 2 metres. Furnish typical
connections of:
(i) flange plates
(ii) web and flange angle connection.
(N.B: Determine the preliminary section and check by rigorous moment of
inertia method).
The girder is laterally restrained throughout.
C.G. of equal angles = 0.28 × Leg length from the heel.
8.14. A simply supported plate girder spans 24 m and carries a total uniformly
distributed load of 4000 kN inclusive of self-weight of the girder. The plate
girder consists of 1600 mm deep × 12 mm thick web plate, 2 ISA 200 mm ×
150 mm × 18 mm (2 ISA 200 150, @ 0.469 kN/m) angles and 550 mm × 12
mm cover plate in each flange. Design the riveted connection of flange
angles to web and flange plates to flange angles.
 Design of OF
DESIGN Beam Connection
PLATE GIRDER 561

8.15. A plate girder section is made up of a web of 200 mm × 12 mm and flange

angles 150 mm × 150 mm × 12 mm (ISA 150 ×150, @ 0.279 kN/m) and one
cover plate in each flange of 45 mm × 12 mm. The girder is supported at
either end on bearing plates 500 mm × 500 mm. If the maximum end
reaction is 1640 kN, design end stiffener using a cluster of 4 angles ISA
130 mm × 130 mm × 12 mm. (ISA 130 130, @ 0.234 kN/m).
8.16. It is required to a design a triple plate, i.e., with moment and shear plates
web splice for a plate girder where the shear is 480 kN and moment 54,00
kN-m. The size of web plate is 1800 mm × 10 mm. The available clear
distance between toes of flange angles is 1520 mm. The total moment of
inertia of the girder section is 466 × 108 mm4.
Use 22 mm diameter power driven rivets.
8.17. Design a suitable web splice for a plate girder at a section at which the
bending moment is 5000 kN-m and shear force is 1000 kN. The plate girder
is built-up from a web plate 2500 mm deep × 16 mm thick and two ISA 200
mm × 150 mm × 15 mm (ISA 200 150, @ 0.394 kN/m) angles and a 600 mm
× 16 mm cover plate in each flange. The rivets used are of 22 mm
diameter. Draw a sketch of the splice. The long legs of the flange angles
are placed horizontally.
 Chapter
Design of Roof

9 Trusses

9.1 INTRODUCTION
The roof trusses are the frame structures in which separate straight members
are so arranged and connected at their ends that the members form triangles.
The axes of members meeting at one joint interest at a common point. The
riveted joints used for the connections of the members are considered to act as
pin-joints. The external loads are applied to the joints of the truss so that the
members carry direct forces only. When beams are subjected to bending, then
bending stress varies from zero at the neutral axis to maximum towards the
extreme fibres. The different fibres in a cross-section of the beams are subjected
to different intensity of stress. The entire section of beam is not utilized. Whereas
the bending of trusses is quite distinct from the beams. In roof trusses, the
entire section of each member is subjected to uniform stress. The strength of
member is fully utilized. The bending action in roof trusses is provided by
elongation and shortening of the members of the truss. It further results in
deflection of truss. The forces in various members are either compressive or
tensile. The members carrying compressive forces in a roof truss are called struts,
and those carrying tensile forces are called ties. In a roof truss, members are so
arranged that the length of members in compression are small while the lengths
of members in tension are long. As far as possible, it is seen that the length of a
member is not more than 3 metres. The members of a truss are classified as
main members and secondary members. The main members are the structural
members which are responsible for carrying and distributing the applied loads
and stability of the truss. The secondary members are the structural members
which are provided for stability and or restraining the main members from
buckling or similar modes of failure.
The roof trusses are used at places which require sloping roofs. The sloping
roofs are necessary at places where rainfall is more and at places where snow
 DESIGN OF ROOF TRUSSES 563

fall occurs. The roof trusses are also used in many single storeyed industrial
buildings, workshops, godowns, warehouses, residential buildings and schools,
where large column free spaces are required for operational purposes. The roof
trusses are suitable for relatively light loads and large spans. The roof trusses
are also used to span the distance between walls or end supports when the least
dimension of the building becomes large and unsuitable and economical to span
with simple beams or joists, columns and girder. The roof trusses have the
advantages of permitting a wider variety of roof shapes and greater unobstructed
interior floor area at less cost. The trusses are specially advantageous where
greater depth is required to satisfy the condition of adequate stiffness.
The primary function of a roof truss is to support the roof covering, external
loads carried by roof covering and ceiling load if any. These loads are transmitted
as reactions to the walls or to the supporting stanchions. In general, the external
loads are applied at the joints of truss. Sometimes, it becomes necessary to
apply loads at intermediate points. In such cases, the members are subjected to
bending in addition to direct forces.

9.2 TYPES OF TRUSSES

The various types of roof trusses used are shown in Fig. 9.1. The suitable spans
for these trusses have also been shown for each type of truss. King post truss
shown in Fig. 9.1 (a) is a wooden truss. It can also be built of a combination of
wood and steel. Queen post truss shown in Fig. 9.1 (b) is also a wooden truss.
Howe truss had been named after Willium Howe. Howe trusses shown in Fig. 9.1
(c) are made of combination of wood and steel. The vertical members are tension
members and made of steel. Pratt trusses shown in Fig. 9.1 (d) are made of steel.
The Pratt truss had been named after Thomas W. and Caleb Pratt. The diagonal
members near the top are highly stressed because of their sharp inclination.
These are less economical than fink trusses. The vertical members are
compression members and diagonal members are tension members. The modified
shape of Pratt roof truss is also used. Fink roof trusses shown in Fig. 9.1 (e) are
made of steel. They are very economical form of roof trusses. The lengths of
compression members are small in these types of trusses. Fink trusses are also
called French roof trusses. The Fink truss had been named after its originator
Albert Fink. A Fink truss is constructed by drawing perpendicular to and from
the centre of the top-left and right hand chords and extending until they meet
the lower chord. The remaining two web members are drawn from these
intersections to the peak as shown in Fig. 9.1 (e). The number of panels in half
the truss may be increased to four by sub-dividing each panel into 2, as shown in
Fig. 9.1 (e). This process may be repeated again resulting in eight panels as
shown in Fig. 9.1 (e). A disadvantage of this type of truss is that the number of
panels can be increased only by doubling the previous number. A modification of
fink truss that permits greater flexibility in number of panels is the fan truss.
Fan roof trusses shown in Fig. 9.1 (f) are made of steel. Fan roof trusses are form
of fink roof trusses. In fan trusses top chord is divided into small lengths, in order
to provide supports for purlins which would not come at joints in trusses shown
in Fig. 9.1 (e). The web numbers do not intersect the top chord at right angles.
564 DESIGN OF STEEL STRUCTURES

This feature of fan truss may be used or combined with the fink truss. The
number of panels in the combined fink-fan truss may be made six instead of four,
i.e., in between four and eight. The sky light may be mounted on top of most of
trusses when desired. For unsymmetrical layouts, the natural light is received
by using saw tooth trusses or north light trusses. The roof trusses shown in Fig.
9.1 (g) are used for industrial buildings and drawing rooms. Such type of trusses
provide north light. These trusses are called North light roof trusses. The trusses
shown in Fig. 9.1 (h) are used for roofs. The trusses shown in Fig. 9.1 (j) are used
for very large span such as railway sheds and auditoriums.

(a ) K ing p ost spa n s u p to 8 m (b ) Q ue e n p ost sp ans up to 1 0 m

(c) H o w e tru ss w ith 4 a n d 8 p an els sp an s 6 m to 30 m

(d ) P ra tt tru ss spa ns 6 m to 1 00 m m o difie d P ratt truss

(e ) Fink o r F ren ch roof tru sse s sp an s up to 1 0 m

C o m p ou n d F ren ch truss C o m p ou n d F ren ch ro o f tru ss

spa n 20 m to 3 0 m spa n 20 m to 3 0 m

(f) Fa n truss spa ns 10 m to 15 m (g) N orth lig ht roo f truss spa ns 8 m to 1 0 m

(i) (a ) Truss used for la rge spa n (h ) Q ua d ran gu la r truss

(i) (b ) Truss used for larg e sp a ns

Fig. 9.1 Various types of trusses

In general, the lower chord of the truss is kept straight. In large rooms,
series of trusses are used one behind the other. In such cases, trusses with
straight chords appear to sag. Then, the lower chord is cambered. It is done for
 DESIGN OF ROOF TRUSSES 565

better appearance. As a result of this, the depth of truss is reduced. Therefore,

the cost of trusses arrangement will be economical and pleasing.

9.3 VARIOUS TERMS USED IN ROOF TRUSSES

The various terms used in roof trusses are given below and are shown in Fig. 9.2.

R id g e
Top cho rd
R a fte rs
P u rlin P a el p oint
P rincip a l ra fte r
P a ne l R ise
E a ve s

θ S a g tie θ
S lop e P a ne l B o tto m cho rd S lop e
M ain tie
S p an
E L E VATIO N

C o lu m n P u rlin R id g e

S p ac ing
of C /L of tru sse s
purlins

S p acin g of trusse s Bay

PLAN
Fig. 9.2 Elements of roof truss

9.3.1 Top Chord

The top chord is defined as the uppermost line of members extending from one-
support to the other and that passes through the peak of truss. The top chord is
also known as the upper chord of the roof truss.

9.3.2 Bottom Chord

The bottom chord is defined as the lower most line of members of truss extend-
ing from one support to the other. The bottom chord is also known as the lower
chord of the roof truss.

9.3.3 Span
The span of a roof truss is defined as the distance between centre to centre of
supports. The span of a roof truss is decided by the dimensions of area to be
kept free of columns.
566 DESIGN OF STEEL STRUCTURES

9.3.4 Rise
The rise of a roof truss is defined as the distance from the highest point to the
line joining supports.

9.3.5 Pitch
The pitch of a symmetrical truss is defined as the ratio of rise to the full
span. Where the roofs are to carry snow loads in addition to wind load, a pitch
1
of is most common and economical. It corresponds to a slope of 1 in 2 or an
4
1 1 1
angle of 26 ° . The pitches , and that corresponding to an angle of inclina-
2 3 5
tion of 30° are also commonly employed. The wind force is less when the pitch of
roof truss is less.

9.3.6 Slope
The slope of a symmetrical truss is defined as ratio of rise to half the span. The
minimum slope depends to a great extent on the type of roof covering material.
The slope provided must be sufficient to drain off the rain water without allow-
ing leakage at the joints of covering material. The steep slopes remain better for
architectural effect. But the wind pressure becomes large on such roofs. It will
need strong roof trusses and large quantity of roof material. The pitches of sym-
metrical trusses used, their respective slope and the angles of inclination with
the horizontal are given in Table 9.1.

Table 9.1

Pitch Slope Angle of inclination with horizontal

1
1 45° 00´ 00"
2
1 2
33° 41´ 00"
3 3
1 1
20° 34' 00"
4 2
1 2
21° 48' 00'
5 5
1 1
18° 26' 00"
6 3
1 1
14° 02' 00"
8 4
 DESIGN OF ROOF TRUSSES 567

The slope of a roof is defined as the tangent of the angle that the plane of the
roof makes with horizontal; or the tangent of the angle between the top and
bottom chords of the roof trusses provided the bottom chord is horizontal. The
slope of the roof therefore is not equal to the pitch and great care must be exercised
to see that two terms are not used synonymously. The slope of the roof is equal to
twice the numerical value of pitch in all the cases whether truss is symmetrical
or unsymmetrical.

9.3.7 Purlins
The purlins are structural members subjected to transverse loads and rest on
the top chords of roof trusses. The purlins support the sheathing that carries
roof covering or roof covering directly. The purlins are horizontal beams spanning
between the two adjacent trusses.

9.3.8 Sub-purlins
The sub-purlins are secondary system of purlins resting on the rafter. The sub-
purlins are spaced to support the tiles or slate coverings.

9.3.9 Rafters
The rafters are beams and rest on the purlins. The rafters support the sheathing.
These may support sub-purlins directly. These are called common rafters to
distinguish from principal rafter.

9.3.10 Sheathing
The sheathing are covering of boards or reinforced concrete. The sheathings are
supported on purlins or rafters. The sheathings provide support for the roof
covering.

9.3.11 Panel
The panel is defined as a distance between two adjacent joints in a principal
rafter of a roof truss. It is also defined as the distance between the two adjacent
purlins.

9.3.12 Bay and Spacing of Purlins

The bay is defined as the distance between the adjacent trusses. The spacing of
purlins is defined as the distance between two adjacent panel points, if purlins are
placed at panel points only. It depends upon type of roof covering material and on
the slope of the truss. The spacing of purlins should be less than or equal to the safe
span of roof covering material if these are placed directly over purlins. In case, the
roof covering material is supported on common rafters, then, purlins must be placed
at any desired spacing. As far as possible, the purlins must be placed at the panel
points so that in the top chord of truss, there is no bending moment due to purlin
load. Generally, the spacing of purlins varies from 2 m to 3 m.
568 DESIGN OF STEEL STRUCTURES

9.3.13 Spacing of Trusses

The spacing of trusses is decided by local conditions and requirements. The
positions of doors, windows and columns, also control the spacing of trusses. It
is desirable to have a uniform spacing of trusses in one portion or in a wing of a
structure, so that as many trusses as possible may be made identical.

9.3.14 Ridge Line and Eaves

The ridge line or a ridge is a line joining the vertices of the trusses. The bottom
edges of an inclined roof surface or a pitched roof is termed as eaves.

9.3.15 Sag Tie

As shown in Fig. 9.2, a tie member is provided to join the peak of truss and
middle tie member. The length of middle tie member used to be large. The
deflection of this member due to self weight may be large. When a sag tie is
provided, it decreases the deflection of middle tie member.

9.3.16 Principal Rafter

The top chord members as shown in Fig. 9.2 are called principal rafters. The
principal rafter carries compressive force.

9.4 ECONOMICAL SPACING OF ROOF TRUSSES

For a roof structure, the most economical spacing of trusses depends on many
factors. It is not possible to give a simple formula. The economic spacing of
trusses depends upon the weight per square metre of floor area of trusses, purlins
columns, roof covering etc. It also depends on relative cost of each type of material.
In case, the spacing of trusses is large, the cost of purlins will be large. If the
spacing of trusses is kept small, the cost of purlins is small, but the cost of
trusses will be large. Therefore, the spacing of trusses should be kept such that
the overall cost of roof structure as a whole is minimum. The economic spacing
of trusses is defined as the spacing which makes the cost of trusses, purlins,
columns, roof covering etc., a minimum.
Let s, be the spacing of trusses ; t, be the cost of trusses ; p, be the cost of
purlins : and r, be the cost of roof covering material.
The cost of trusses, t, is inversely proportional to the spacing of trusses, s.
Therefore,

⎛c ⎞
t = ⎜ 1⎟ ...(i)
⎝s⎠
The size of purlins depends on the bending moment. The bending moment
carried by purlin varies as the square of the span of purlin. Hence, the cost of
purlins, p, is directly proportional to the square of spacing of trusses.
 DESIGN OF ROOF TRUSSES 569

Therefore,
p = (c2 . s2) ...(ii)
The cost of roof covering materials, r, is directly proportional to the spacing
of trusses. Therefore,
r = (c3 . s) ...(iii)
where c1, c2 and c3 are constants
The overall cost of roof structure
X = (t + p + r)
⎛c ⎞
Therefore, X = ⎜ 1 + c2 ⋅ s2 + c3 ⋅ s ⎟ ...(iv)
⎝s ⎠
For, economic spacing of the trusses, the overall cost of these items should be
minimum.
⎛ dX ⎞
Therefore, ⎜ ⎟ = 0 ...(v)
⎝ ds ⎠
⎛ dX ⎞ ⎛ c ⎞
⎜ ⎟ = ⎜ − 1 + c2 ⋅ s + c3 ⎟
⎝ ds ⎠ ⎝ s ⎠

⎛ c1 2 ⎞
⎜⎝ − s + 2c2 ⋅ s + c3 ⋅ s⎟⎠ = 0,

or (–t + 2p + r) = 0
t = (2p + r) ...(9.1)
Therefore, for economic spacing of trusses the cost of trusses should be
equal to twice the cost of purlins plus cost of roof covering material. In this
expression the factor s, spacing of trusses does not occur. Therefore, it is not
possible to determine the spacing directly from this expression. This expression
provides only a check to verify that the spacing provided is economical or not.
The economical spacing will usually be kept between 3 m to 5 m for spans
upto 15 m and 45 m to 7.5 for spams 15 m to 30 m.

9.5 ROOF COVERINGS

The following materials are used as roof covering materials.

9.5.1 Slates
The slates are most durable roof covering material. The slates are not affected
by heat, rain or other atmospheric actions. The slates are heavy in weight. The
slates are attached to battens by slating nails. The slating nails are made of
1
copper or non-rusting material. A minimum pitch of is required for roof trusses
2
using slates as roof covering materials. The slopes of 27° and 33° are provided
for 500 mm × 250 mm size and 400 mm × 200 mm size of slates, respectively.
570 DESIGN OF STEEL STRUCTURES

9.5.2 Tiles
The tiles are used as roofing material as substitute for slates. The tiles are
manufactured from asbestos, clay and many compositions. The tiles are available
in market in various forms. Tiles are cheap in comparison to slates. The tiles
1
are not durable as slates. A minimum pitch of is required for roof trusses
4
using tiles as roof covering materials.

9.5.3 Lead
The lead-sheets were formerly used as roof covering materials specially on flat
roofs. The lead is heavy in weight. The lead is also a costly material. The lead
sheets are very durable. But these are not suitable for steep slopes.

9.5.4 Zinc
The zinc is also one of the materials used for roof covering. The zinc sheets are
comparatively light in weight and cheap in cost in comparison to lead. Under
atmospheric action, the corrosion and decay of zinc sheets may take place. The
zinc sheets can be used on steep slopes.

9.5.5 Glass
The glass sheets are manufactured in various sizes and thickness. The glass
sheets are mainly used for light from sloping roof.

9.5.6 Galvanized Corrugated Iron Sheets

The galvanized steel sheets are made by galvanizing black sheets, rolled from
good quality low-carbon mild steel, free from cracks, pittings, blisters,
laminations, twists, scales and other surface defects. The black sheets used for
galvanizing are annealed. The galvanizing is carried out by the hot-dip process
by first pickling the black sheets, and then dipping them in bath of molten zinc,
at a temperature suitable to produce a complete and uniform adhesive coating
of zinc. The zinc coating contains not less than 97.5 percent of pure zinc. The
galvanizing corrugated sheets used for roof covering should be free from twist
or buckle and should have uniform corrugations, true in depth and pitch and
parallel to the sides of sheets.
The depth of corrugation is 18 mm (nominal), and pitch of corrugations is 5
mm. The number of corrugations in a galvanised corrugated sheet is 8 or 10.
The nominal overall width, of galvanized corrugated sheets measured between
the crown of outside corrugations in case of 8 corrugations per sheet, is 750 mm
before corrugations and 600 mm after corrugation and that in case of 10
corrugations per sheet, is 900 mm before corrugation and 800 mm after
corrugations. The galvanized sheets are manufactured in lengths of 18 m, 2.2
m, 2.5 m, 2.8 m and 3.2 m. The dimensions of galvanized corrugated sheets are
given as per IS : 277–1962.
 DESIGN OF ROOF TRUSSES 571

These sheets are used for slopes of roof 1 in 4 and above. These sheets are
very light in weight, and can be fixed to purlins by J-bolts. The holes in these
sheets should be made on ground. Bitumen washers are used below. G.I.
washers to make the joint water-tight. The decay of galvanized sheets starts
rapidly at holes due to atmospheric action. The decay of these sheets takes
place due to atmospheric action as soon as the galvanizing has worn off. The life
of galvanized corrugated sheet is comparatively short.

9.5.7 Asbestos Cement Sheets

A.C. sheets for roof covering are available in two varieties, namely, ‘Asbestos
cement corrugated sheet’ and ‘Asbestos cement Trafford sheet’. A.C. sheets are
good insulators against temperature. A.C. sheets do not decay because of atmo-
spheric action. These are not perishable. The asbestos cement corrugated sheets
are shown in Fig. 9.3 (a) and asbestos cement trafford sheets are shown in Fig.
9.3 (b). The trafford sheets are also termed as semi-corrugated sheets.

r
29 m m R
25 m m R

(a ) A sb esto s-ce m e n t corru g ate d sh eets

C
B

25 m m R
35 m m R
4 5°

(b ) A sb esto s-ce m e n t sem i-co rru ga ted (traffo rd ) she ets

a sb esto s ce m en t sh ee ts (ro of co vering s)

Fig. 9.3 Asbestos cement sheets

Both the varieties of A.C, sheets are available in lengths of 1.50 m, 1.75 m,
2 m, 2.25 m, 2.50 m, 2.75 m and 3 m. Both the sheets are manufactured in
nominal thickness, T, as 6 mm and 7 mm. For corrugated sheets, the pitch P is
146 mm, and depth of corrugation D, is 48 mm. The overall width B of these
sheets is 1.50 m and effective width C is 1.01 m. For trafford sheets, the pitch P
is 338 mm and depth of corrugation D is 45 mm. The overall width B of the
572 DESIGN OF STEEL STRUCTURES

trafford sheets is 1.01 m and effective width is 1.014 m. The dimensions of these
A.C. sheets are given as per IS : 459–1962.
The asbestos cements sheets are used for 30° slope and above. A.C. sheets
are fixed with purlins with J-bolts. The holes are drilled on the top of corrugations.
The bitumen washers are placed under G.I. washers in order to make it water-
tight.

9.5.8 Corrugated Aluminium Sheets

The corrugated aluminium sheets find extensive application where use of
aluminium is justified due to its corrosive resistance, lightness and thermal
efficiency. The dimensions of corrugated aluminium sheets may be noted from
IS : 1254–1965. The corrugated aluminium sheets are known as general purpose
sheets. The thickness of sheet is kept not less than 0.45 mm. The widths available
are 650 mm and 800 mm and lengths, 1800, 2400 and 3600 mm.

9.6 LOADS ON ROOF TRUSSES

The roof trusses are subjected to dead load and live load. In addition to these
loads, the roof trusses are also subjected to some special loads such as ceiling or
suspended floors or heavy machinery.

9.6.1 Dead Load

In Chapter 1; loads and stresses, in Sec. 1.22, dead load has been described in
general. Dead load on roof trusses includes the weight of roof covering, the
weight of purlins, the weight of bracing and the self weight of trusses.
(i) Weight of roof covering. The unit weights of building materials have
been given in Table 1.7. It includes the weights of asbestos cement
corrugated and semi-corrugated sheets, C.G.I, sheets, tiles, glass and
slates. The weights of these materials are given in kN per square metre
of plan areas.
(ii) Weight of purlins. The weight of purlins is assumed as 0.070 to 0.150
kN per square metre of plan area.
(iii) Weight of bracings. The weight of bracings is assumed as 0.015 kN of
plan area.
(iv) Weight of trusses. For the design of roof trusses, the weight of truss
is assumed. The weight of truss varies with the span, and the rise of
truss, the spacing of trusses, the type of roof covering material, the
geographical situation of the roof structure. It is not possible to have an
exact formula which includes all these factors. The self weight of truss
is a small part of the total design loads for the roof truss. The self weight
of truss may be assumed as 0.90 to 0.150 kN per square metre of plan
area. The self weight of truss can also be found by empirical formula
given below :
 DESIGN OF ROOF TRUSSES 573

The self weight of truss in kN per square metre of plan area

1 ⎛l ⎞
w = ⎜ + 5⎟⎠ kN/m2 ...(9.2)
100 ⎝ 3
where, l is the span of truss in metres.
This formula is applicable for roof trusres with C.G.I, sheets and pitch equals
to 1 to 4 and for 4 m spacing of trusses. For other spacing of trusses, a propor-
tionate value for self weight is determined.
The weight of trusses w in N/m2 of area covered, depending upon the design
loads q in N/m2 may be found from the following formulae for the spans men-
tioned:
1 ⎛ q ⎞
Span L = 18 m, w = ⎜ 2.2 + ⎟ kN/m2 ...(i)
100 ⎝ 12.5 ⎠
1 ⎛ q ⎞
Span L = 24 m, w = ⎜ 2.78 + ⎟ kN/m2 ...(ii)
100 ⎝ 5.42 ⎠
1 ⎛ q ⎞
Span L = 30 m, w = ⎜ 4.44 + ⎟ kN/m2 ...(iii)
100 ⎝ 3.47 ⎠
1 ⎛ q ⎞
Span L = 36 m, w = ⎜ 5.27 + ⎟ kN/m2 ...(iv)
100 ⎝ 2.1 ⎠
The weight of steel roof trusses may also be estimated from the following
expressions. These expressions include the plan area covered by the steel roof
trusses. The weight of riveted steel trusses are found as follows:
For sheeted roof trusses.
1
w = (4.88 + 0.075 × A) kN/mm2 ...(v)
100
For partly glazed roof trusses
1
w = (4.88 + 0.088 × A) kN/m2 ...(vi)
100
where A = plane area in m2
The weight of welded roof trusses area approximated as follows :
For sheeted roof trusses
1
w = (5.37 + 0.053 × A) kN/m2 ...(vii)
100
For partly glazed roof trusses
1
w = (5.37 + 0.064 × A) kN/m2
100

9.6.2 Live Loads

In Chapter 1, the loads and stresses : Sec. 1.23 live loads have been described in
general. The imposed (live) loads on various types of roofs other than wind load
and snow load, as per IS : 875–1987, for roofs with slopes upto and including
574 DESIGN OF STEEL STRUCTURES

10 degrees, is adopted as 1.5 kN per square metre of plan area where access is
provided to roof. The minimum live load measured on plan shall be 3.75 kN
uniformly distributed over any span of one metre width of the roof slab and 9.0
kN uniformly distributed over the span in the case of all beams. Where the
access is not provided, except the maintenance, live load on roofs is adopted as
0.750 kN per square metre of plan area. In ths case, the minimum live load
measured on plan shall be 1.9 kN uniformly distributed over any span of one
metre width of the roof slab and 4.50 kN uniformly distributed over the span in
the case of beams. The live load for sloping roof with slopes greater than 10
degrees is adopted as 0.750 kN per square metre of plan area less 0.020 kN/m2
for every degree increase in slope over 10 degrees subjected to minimum of
0.400 kN/m2 per square metre of plan area.
The live loads for curved roofs with slopes at springing greater than 10
degrees shall be (0.75–52 γ2) × kN/m2,
⎛h⎞
where, r= ⎜ ⎟
⎝l⎠
h = the height of the highest point of the structure measured from
springing
l = the chord width of the roof if singly curved and shorter of the
two, if doubly curved.
This is subjected to minimum of 0.400 kN/m2. The live load also includes
snow load and wind load.

9.6.3 Snow Loads

The snow loads depends upon geographical situation of roof structure. It
depends upon latitude of the place and atmospheric humidity. If a roof is
subjected to snow load, it should be designed for the actual load due to snow or
for the live load mentioned above whichever is more severe. The actual load due
to snow will depend upon the shape of roof and its capacity to retain the snow
and each case shall be treated on its own merits. In the absence of any specific
information, the bending due to the collection of snow may be assumed to be 25
N/m2 per mm depth of snow. The possibility of total or partial snow load should
be considered, that is, one half of the roof fully loaded with the design snow load
and the other half loaded with half the design snow load. In the case of roof with
slopes greater than 50°, snow load may be disregarded. Where, however, there
are possibilities of formation of snow pockets, these should be taken into
accounts.

9.6.4 Wind Loads

The wind load is one of the most important loads that an engineer has to deal
with and is also one of that is most difficult to evaluate properly. The magni-
tude of wind pressure depends on wind velocity and the shape of the struc-
ture. The magnitude of wind velocity varies with the geographical location of
the structure and the height of the structure.
 DESIGN OF ROOF TRUSSES 575

Sir Isaac Newton established the relationship between the wind pressure
and the wind velocity and is known as Newtonian theory. When a stream of air
flows around the object, then, the wind creates a pressure on the nose of object.
The pressure developed may be determined from the consideration of energy
involved. According to Bernoullis theorem, the sum of energies at all points.
Therefore. (K.E. + P1. E1 + P2 . E2) = Constant ...(v)
Wu2
where, K.E. = Kinetic energy, which a mass M possesses,
2g
W = Weight of that mass
g = Acceleration due to gravity
u = Linear velocity of the wind
P1 E1 = Potential energy, W.h
h = Height above the datum plant
W⋅p
P2E2 = Pressure energy,
w
P = Pressure of the fluid (wind)
W = Weight this pressure can lift
w = Unit weight of the fluid (wind)
The expression may be written in terms of the expressions for these energies.
Therefore,
⎛ W ⋅ u2 W ⋅ p⎞
⎜ +W ⋅h + ⎟ = Constant ... (vi)
⎝ 2g w ⎠
The expression (vi) is divided by W
⎛ v2 p⎞
∴ ⎜ + h + ⎟ = Constant ... (vii)
⎝ 2g w⎠
The sum of three terms in the expression (vii) is known as total head. It is
represented by H. These terms are known as velocity head, gravity head and
pressure head, respectively.
⎛ v2 p⎞
H = ⎜ +h+ ⎟
⎝ 2g w⎠

⎛ w ⋅ v2 ⎞
or H.w = ⎜ +w⋅h + p⎟ ...(viii)
⎝ 2g ⎠
The expression (viii) gives total energy
⎛ w ⋅ v2 ⎞
E = ⎜ + w ⋅ h + p⎟ ...(ix)
⎝ 2 g ⎠
The unit of total energy is now N per sq. metre. The energy per unit volume
given by the expression (ix) remains constant. The energy at the nose of object
and the stagnation point (the point where wind velocity is zero) are equal
576 DESIGN OF STEEL STRUCTURES

⎛ w ⋅ u02 ⎞
⎜ p0 + = ps ⎟
⎝ 2g ⎠
where, p0 = Pressure at the nose of object
ps = Pressure at the stagnation point of object
u0 = Velocity at the nose
From the expression (x)

⎛ w ⋅ u02 ⎞
(ps – p0) = ⎜ ⎟ ...(xi)
⎝ 2g ⎠
When the wind is brought to rest, then it results in rise in pressure (ps – p0).
It is termed as impact pressure, (kinetic pressure) of the air and is represented
by q. Hence

⎛ w ⋅ v02 ⎞ ⎛ w ⋅ v2 ⎞
q = ⎜ ⎟=⎜ ⎟ ...(xii)
⎝ 2g ⎠ ⎝ 2g ⎠
When the temperature and the barometric pressure change, then, the weight
of wind per unit volume also changes. The weight of wind at sea level at a
temperature of 17°C is 0.347 N. At 0°C, the weight of wind is 0.365 N. It increases
with the decreases of temperature. The value of acceleration due to gravity may
be adopted at 9.81 m/sec2. The expression (xii) may be written as

⎛K ⎞
q = ⎜ 2⎟ ...(xiii)
⎝v ⎠
where, K is the constant or a coefficient. It takes into account the unit of wind
velocity in km per hour.
The basic wind pressure p is adopted equal to q × CD and the value of K is
taken as 0.06 in IS : 875–1964. Therefore,
p = (0.06 × v2) × 10 N/m2
= 0.6 v2 N/m2 ...(xiv)
where, CD is the shape factor.
The velocity of wind (wind speed) varies with the height above ground
elevation. The relationship showing the variation of wind is as follows :
0.225
v1 ⎛H ⎞
= ⎜ 1⎟ ...(xv)
v2 ⎝ H2 ⎠
where, v1is the velocity of wind (wind speed) in km per hour at height H1 in
metres and v2 is the velocity of wind in km per hour at height H2 in metres.
The wind velocities (wind speed) are measured by meteorological observatories
with the help of instrument known as anemometers. These instruments are
kept at standard height of 10 m in the different parts of our country and the
wind velocities (speeds) are measured at this height.
 DESIGN OF ROOF TRUSSES 577

Basic wind speed: The basic wind speed, Vb is the wind speed measured in
a 50 year return period. The basic wind speed is based on peak gust velocity
averaged over a short interval of time of about 3 (three) seconds and it corre-
sponds to mean heights above groud level in an open terrain (category 2).

G IL G IT

S R IN A G A R Z ON E 1
L EH
P U N CH
C H A BA

MANDI
A M R IT S A R S H IM L A Z ON E 5
C H A N D IG A R H
D E H RA D U N
PA TIA LA S A HA R N P U R
K A RN A L A LM O D A
R O H TA K Z ON E 2
M E E RU T A LO N G
B IK A N A E R D E LH I Z ON E 3 T E ZU
R A M PU R D E BR U A R H
N A G A U R S IK A R A LW A R G A N G TO K
J A IS A L M E R J A IP U R AGRA JO RH AT
LU CK N O W G O R A NH P U R
JO DH P U R
A JM E R G W A L IO R D IS P U R K O H IM I
B A RM E R R A E B A R E L I
B U N DI S H IL LC N H
S IR O H IE H A M IR P U R PA TN A
K O TA A LL A H A B A D IM P H A L
U D A IP U R N AW A D D U M K A Z ON E 2 B IL C H A R
A U R AN G
PA L A N P U R R E W A A B AD A G R TA L A
B H AL I SAGR R A N CH I
ZONE 2 B H O PA L
A H M D AB A D JAB L P U R Z ON E 1
A M B IK A P U R R A R K E L A C A LC U T TA
JAM N A G A R ZONE S E O NI B IL A S P U R Z ON E 2
M E D N PU R

S U R AT JAL G A O N N A G PU R R A IP U R
D H U LE A M R AVA
B H U RA N E S H W AR
N A S H IK A K O LA
C H A ND R A P U R
A U R AN G A B A D C H H AT R A P U R
B O M B AY JAG D A L PU R
A H M D NA G A R
K A RN A LO R
PUNE
S ATA R A S O LA B B A D A L
W A R A N A P U R V IS H A K H A PA T NA M
H Y D ER A B A D
Z ON E 4
R A N CH U R M A C H LIPAT N A M
B E LG A U M
K A RN O O L B A S IC W IN D
S P E E D IN M / S Z O N ES
B E LA R Y
S H IM O G A N E LL O R E 55 1
DS

50 2
C O B A R IS L A N

C A N NA M G R E B A NS A LO R E
M A D R AS
M Y S O RE 47 3
P O N DC H E R RY 44 4
P O R T B LA IR
LA K SH D IA )

S A LE M
39 5
(IN D IA )
N & NI

T R IC H H R K A R A IK A L
33 6
(IN

K AV A R A TT I Z ON E 3
M A D U RA
A DWE

AM A

R A M ES W AR A M
D
AN
EP

K A N N IYA K U M A R I IN D IR A
P O IN T

Fig. 9.4

Figure 9.4 shows basic wind speed map of our country, as applicable to 10 m
height above mean ground level for different zones. As per IS : 875 (Part 3 –
wind loads) – 1987, six wind zones have been formed which correspond to basic
wind speed of 55, 50, 47, 44, 39 and 33 metre per second, respectively, as shown
in Fig. 9.4. The basic wind speeds for some important cities/towns are given in
Table 9.2.
578 DESIGN OF STEEL STRUCTURES

Table 9.2 Basic wind speeds in m/sec (As per IS : 875 (Part 3)–1987)

City/Town Speed City/Town Speed City/Town Speed

(m/sec) (m/sec) (m/sec)
Agra 47 Dehradun 47 Mysore 33
Ahmadabad 39 Delhi 47 Nagpur 44
Ajmer 47 Durgapur 47 Nainital 47
Almora 47 Gangtok 47 Nasik 39
Amritsar 47 Gauhati 50 Nellore 50
Asansol 47 Gaya 39 Panjim 39
Aurangabad 39 Gorakhpur 47 Patiala 47
Bahraich 47 Hydrabad 44 Patna 47
Bangolore 33 Imphal 47 Pondicherry 50
Barauni 47 Jabalpur 47 Portblair 44
Bareilly 47 Jaipur 47 Pune 39
Bhatinda 47 Jamshedpur 47 Raipur 39
Bhilai 39 Jhanshi 47 Rajkot 39
Bhopal 39 Jodhpur 47 Ranchi 39
Bhubaneshwar 50 Kanpur 47 Roorkee 39
Bhuj 50 Kohima 44 Rourkela 39
Bikaner 47 Kumool 39 Simla 39
Bokarao 47 Lakshadweep 39 Surat 44
Mumbai 44 Lucknow 47 Tiruchchirrappalli 47
Kolkata 50 Ludhiana 47 Trivendrum 39
Calicut 39 Chennai 50 Udaipur 47
Chandigarh 47 Madurai 39 Vadodara 44
Combatore 39 Mandi 39 Varanasi 47
Cuttak 50 Mangalore 39 Vijaywada 50
Darbhanga 55 Moradabad 47 Visakapatnam 50
Darjeeling 47
It is to note that the occurrence of a tornado is possible in virtually any part
of India. They are particularly more severe in the northern parts of our country.
The recorded number of these tornados is too small to assign any frequency.
The devastation caused by a tornado is due to exceptionally high winds about
its periphery, and the sudden reduction in atmospheric pressure at its centre,
resulting in an explosive outward pressure on the elements of the structure.
The regional basic wind speeds do not include any specific allowance for tornado.
It is the usual practice to allow for the effect of tornados unless special
 DESIGN OF ROOF TRUSSES 579

requirements are called for as in the case of important structures, such as, nuclear
power reactors and satellite communication towers.
It is to further note that the total number of cyclonic storms that have struck
different sections of east and most coasts are included in Fig. 9.4 based on
available records for the period from 1877 to 1982. The figures above the lines
(between the stations) indicate the total number of severe cyclone storms with
or without a core of hurricane winds (speeds above 87 kmph) and the figures in
the brackets below the lines indicate the total number of cyclone storms. Their
effect on land is already reflected in the basic wind speeds specified in Fig. 9.4.
These have been included only as an additional information.
The cyclonic storms gradually reduce in speed as they approach the sea coast.
The cyclonic storms generally extend upto 60 km inland after striking the sea
coast. Their effect onland is already reflected in basic wind speed specified in
Fig. 9.4. The influence of wind speed off the coast upto a distance of about 200
km may be taken as 115 times the value on the nearest coast in the absence of
any definite wind data.

9.7 DESIGN WIND SPEED

The basic wind speed, Vb for any site shall be noted from Fig. 9.4 and it is
modified to include the effect of risk level, terrain roughness, height and size of
the structure, and local topography. The design wind speed, Vz may be
mathematically expressed as under. The design wind speed is the wind speed
for which the structure is designed.
Vz = (k 1 . k2 . k3) .Vb ...(9.3)
where,
Vz = design wind speed at any height z in m/sec,
k1 = risk coefficient (probability factor),
k2 = terrain, height and structure size factor, and
k3 = topography factor
Above factors k1, k 2 and k 3 have been described in IS : 875 (Part 3) –1987,
which are as follows: It is to note that the design wind speed upto 10 m height
from the mean ground level shall be considered constant.

9.7.1 Risk Coefficient (Probability Factor), k1

The basic wind speed is based on 50 years mean return period. Figure 9.4 gives
basic wind speeds for terrain (category 2) as applicable at 10 m ground level.
The design life of a structure is based on the function aspect and importance
of a structure. A structure is designed to withstand the frequency
distribution of the wind over that period. The values of risk coefficients based
on probability concept are given in Table 9.3.
580 DESIGN OF STEEL STRUCTURES

Table 9.3 Risk coefficients for different classes of structures in different wind speed zones
(As per IS : 875 (Part 3)–1987)

Class of structures Meam Probable k1 factors for basic wind

life speeds (m/s)
years 33 39 44 47 50 55
1. Temporary structures 5 0.82 0.76 0.73 0.71 0.70 0.67
2. Structures presenting 25 0.94 0.92 0.91 0.90 0.90 0.89
a low degree of hazard
to life and property
3. All general buildings 50 1.0 1.0 1.0 1.0 1.0 1.0
and structures
4. Important building and 100 l.05 1.06 1.07 1.07 1.08 1.08
structures

Note: It is to note that the temporary sheds and structures include those used
during construction operations (e.g., form work and falsework), structures during
construction stages and boundary walls. Isolated towers in wooded areas, farm build-
ings other than the residential buildings present a low degree of hazard to life and
property in the event of failure. Hospitals, communication buildings/towers, power
plant structures are important buildings and structures.
The risk factor k 1 is based on statistical concepts which take into consider-
ation of the degree of reliability needed and period of time in years daring which
these will be exposed to wind, (i.e., life of the structures). Whatever wind speed
is adopted for the design purposes, there is always a probability (however small)
that it may be exceeded in a storm of exceptional voilance; the greater the pe-
riod of years over which these will be exposed to wind, the greater is the prob-
ability.
Higher return periods ranging from 100 years to 1000 years (implying lower
risk level) in the association with greater periods of exposure may have to be
selected for exceptionally important structures, such as, nuclear power reactors
and satellite communication towers.
The expression given below may be used in such cases to estimate k1 factors
for different periods of exposure and chosen probability of exceedance (risk level).
The probability level of 0.63 is normally considered sufficient for the design of
buildings and structure against wind effects and the values of k 2, corresponding
to this risk level are given below:

⎛ X N .P ⎞
k1 = ⎜ ...(9.4)
⎝ X 50, 0.63 ⎟⎠

⎛ ⎡ 1 ⎤⎞
⎜⎝ A − B ⎢ log( − log(1 − PN )⎥⎟⎠
⎣ N ⎦
k1 = ...(9.5)
( A + 4B )
 DESIGN OF ROOF TRUSSES 581

where,
N = mean probable design life of a structure in years
PN = risk level in N consecutive years (probability that the design
wind speed is exceeded atleast once in N successive years), nor-
mal value is 0.63.
(X50, 0.63) = extreme wind speed for (N = 50 years and PN = 0.63)
The values of A and B coefficients for different basic wind speed zones.

Values of Wind zones of speeds (m/sec)

33 39 44 47 50 55
A 83.2 84.0 88.0 88.0 88.8 90.8
B 9.2 14.0 18.0 20.5 22.8 27.3

9.7.2 Terrain, Height and Structure Size Factor, k2

Terrain category means the characteristics of the surface irregularities of an
area which arise from natural or constructed features. The terrain categories
are numbered in the increasing order of roughness. The effect of obstructions in
the surrounding of the structure creates the turbulence and roughness, which
also depends upon the direction of wind. The terrain is classified into following
four categories:
(a) Category 1 : The exposed open terrain with few or no obstruction and
in which the average height of any object surrounding the structure is less than
15 m above ground surface. It is to note that open sea coasts and flat plains
having no trees are included in this category.
(b) Category 2 : The open terrain with scattered obstructions having
heights usually between 15 m to 10 m above ground surface. It is to note that
air fields, open parklands and undeveloped sparsely built-up out skirts of towns
and suburbs are included in this category. This category may also include open
land adjacent to sea coast due to roughness of large sea waves at high winds.
(c) Category 3 : The terrain having numerous closely spaced obstractions
(e.g., building structures upto 10 m in height with or without a few isolated tall
structures.
It is to note that well wooded areas, and shrubs, towns and industrial areas
full or partially developed are included in this category. It is likely that the next
higher category than this will not exist in most design situations and that
selection of a more severe category will be deliberate. The particular attention
must be given to performance of obstructions in areas affected by fully developed
tropical cyclones. The vegetation which is likely to be blown down or defoliated
cannot be relied upon to maintain the conditions of category 3. Where such
situation may exist, either an intermediate category with velocity multipliers
midway between the values for category 2 and 3 given or category 2 should be
selected having due regard to local conditions.
(d) Category 4 : The terrain with numerous large high closed spaced
structures. It is to note that large city centres, (generally with obstructions above
25 m and well developed industrial complexes) are included in this category.
Table 9.4 Values of factors, k2 (as per IS: 875 (Part 3)–1987)
582
Height (m) Terrain categories and classes of buildings
(1) (2) (3) (4)
A B C A B C A B C A B C
10 1.05 1.03 0.99 1.00 0.98 0.93 0.91 0.88 0.82 0.80 0.76 0.67
15 1.09 1.07 1.03 1.05 1.02 0.97 0.97 0.94 0.87 0.80 0.76 0.67
20 1.12 1.10 1.06 1.07 1.05 1.00 1.01 0.98 0.91 0.80 0.76 0.67
30 1.15 1.13 1.09 1.12 1.10 1.04 1.06 1.03 0.96 0.97 0.93 0.83
50 1.20 1.18 1.14 1.17 1.15 1.10 1.12 1.09 1.02 1.10 1.05 0.95
100 1.26 1.24 1.20 1.24 1.22 1.17 1.20 1.17 1.10 1.20 1.15 1.05
150 1.30 1.24 1.20 1.24 1.22 1.17 1.20 1.17 1.10 1.20 1.15 1.05
200 1.32 1.30 1.26 1.30 1.28 1.24 1.27 1.24 1.18 1.27 1.22 1.13
250 1.34 1.32 1.28 1.32 1.31 1.26 1.29 1.26 1.20 1.28 1.24 1.16
300 1.35 1.34 1.30 1.34 1.32 1.28 1.31 1.28 1.22 1.30 1.26 1.17
DESIGN OF STEEL STRUCTURES

350 1.37 1.35 1.31 1.36 1.34 1.29 1.32 1.30 1.24 1.31 1.27 1.19
400 1.38 1.36 1.32 1.37 1.35 1.30 1.34 1.31 1.25 1.32 1.28 1.20
450 1.39 1.37 1.33 1.38 1.36 1.31 1.35 1.32 1.26 1.33 1.29 1.21
500 1.40 1.38 1.34 1.39 1.37 1.32 1.36 1.33 1.28 1.34 1.30 1.22

Note. Intermediate values may be obtained by linear interpolation, if desired. It is permissible to assume constant wind speed
between two heights for simplicity. The height mentioned above is the height of the structure.
 DESIGN OF ROOF TRUSSES 583

The terrain and building sizes classified into following three classes which
depend on their sizes.
(i) Class A : The structures and/or their components such as cladding,
glazing, roofing etc. having maximum dimensions (horizontal/vertical) is less
than 20 m above ground surface).
(ii) Class B : The structures and/or their components such as cladding,
glazing, roofing etc. having maximum dimension (horizontal/vertical) between
20 m and 50 m above ground surface.
(iii) Class C : The structures and/or their components such as cladding,
glazing, roofing etc. having maximum dimension (horizontal/vertical) more than
50 m above ground surface.
Table 9.4 gives multiplying factor, k 2 by which the basic wind speed noted
from Fig. 9.4 must be multiplied to get the wind speed at different heights, in
each terrain category for different sizes/classes of buildings/structures.
Depending on the direction of wind, the terrain category used in the design
of a structure may vary. The velocity profile for a given terrain category does
not develop to full height immediately with the commencement of that terrain
category. It develops gradually to height hx which increases with the fetch or
upward distance. Fetch is the distance measured along the wind from a bound-
ary at which a change in the type of terrain occurs. The developed height, hx in
metres may be noted from Table 9.5.

Table 9.5 Fetch length and develope height, hx (As per IS : 875 (Part 3)–1987)

Fetch length x km Developed height, hx (m) (Terrain category)

1 2 3 4
0.2 12 20 35 60
0.5 20 30 35 95
1 25 45 80 130
2 35 65 110 190
5 60 100 170 300
10 80 140 250 450
20 120 200 350 500
50 180 300 400 500

For the structures of heights more than the developed height hx as noted or
least rough terrain.

9.7.3 Topography Factor k3

The basic wind speed, Vb as given in Fig. 9.4 takes into consideration the
general level of site above sea level. This does not allow for local topographic
features such as hills, valleys, cliffs, escarpments, or ridges which may
significantly influence the wind speed in their surrounding. Near the summits
of hills or crests of cliffs, escarpments or ridges, the wind gets accelerated, where
584 DESIGN OF STEEL STRUCTURES

as near the foot of cliffs, steep escarpments or ridges, the wind gets decelerated.
The wind is parallel to the ground in the planes. The wind is assumed to be
normal to the face of the face of the structure. The topography factor k 3 is given
by
k3 = (1 + C · s) ...(9.6)
For 3° < φ < 17°

⎛ z⎞
C = 12 . ⎜ ⎟ ...(9.7)
⎝ L⎠
Le = L ...(9.8)
For φ > 17°,
C = 0.36 ...(9.9)

⎛ z ⎞
L = ⎜ ...(9.10)
⎝ 0.3 ⎟⎠
where φ = upward slope of the ground in the wind direction
S = a factor taken from Figs 9.5 and 9.6
L = actual length of the upward wind slope in the direction of wind
z = effective height of the crest (feature)
Le = effective horizontal length
The effect of topography becomes significant at a site when the upward wind
slope, φ is more than 30°, and below that, the value of k3 may be taken equal to
unity. The value of k3 lies in the range of 1.0 to 1.36 for slopes greater than 3°.

Region affected by topographical feature

5 Le 2.5 L e
Crest
H H
Wind p e Do
s lo w nw
nd ard
Z i s lo
Average ground U pw pe
level e

L
5 km
–ve Upwind +ve Downwind

(a) General notations

Crest Wind Crest
Wind

Downwind
Slope >3° Downwind
Slope <3°
e e

(b) Cliff and escarpment (c) Hill and ridge

Fig. 9.5 Topographical dimension

 DESIGN OF ROOF TRUSSES 585

From the expressions (9.8) and (9.10), it is seen that the values of factors s
depend upon the effective horizontal length, Le. For the design of all the struc-
ture within the effected zone (1.5 Le on the windward side and 2.5 Le on the
loeward side) as shown in Fig. 9.5, this coefficient has to be considered.

9.8 DESIGN WIND PRESSURES

The design wind pressure, pz depends upon the basic wind speed, Vb, the struc-
ture above ground level, the terrain categories, the local topography, the aspect
ratio (viz., length and breadth of the building/structure, i.e., size of structure),
the shape of structure and the solidity ratio or openings the structures.
The design wind pressure at any height above mean ground level shall be
obtained from the following expression
pz = 0.6 (Vz)2 N/m2 ...(9.11)
pz = 0.0006 (Vz)2 kN/m2 ...(9.12)
where,Vz is the design wind speed in m/sec at height z. This coefficient 0.6 (or
0.0006) (in S.I.units) in the above expressions depends on a number of factors
and mainly on the atmospheric pressure and temperature.
It is to note that wind pressures and wind forces (loads) acting over the struc-
tures shall be calculated for the building as a whole, the individual structurals
elements as roofs and walls and the individual cladding units (glazing and their
fixing).

9.9 WIND FORCES (LOADS) ON BUILDINGS/STRUCTURES

For a particular surface or part of the surface of a building, the pressure coef-
ficients are always given. The wind load acting normal to a surface is deter-
mined by multiplying the area of that surface or its appropriate potion by the
pressure coefficient, Cp and the design wind pressure at the height of the sur-
face from the ground. The averages values of the these pressure coefficients
(external pressure coefficients and internal pressure coefficients) for
some shapes of buildings have been given in subsequent sections.
The average values of pressure coefficients are given for critical wind direc-
tions in one or more quadrants. In order to calculate the maximum wind load on
the building, the total wind load should be found for each of the critical direc-
tions shown from all the quadrants. In case, the considerable variation of pres-
sures occurs over a surface, it has been subdivided and the mean pressure coef-
ficients have been mentioned for each of its several parts.
In addition, the areas of high local suction (negative pressure concentration)
frequently occurring near the edges of walls and roofs are separately noted.
These coefficients for the local effects should only be used for calculation of
forces on these local areas influencing roof sheeting, glass panels, individual
cladding units including their fixtures. These should not be used for calculating
forces on entire structural elements such as roof walls or structure as a whole.
It is to note that the pressure coefficients have been found from measurements
on models in wind tunnels, and the great majority of data available has been
586 DESIGN OF STEEL STRUCTURES

used in conditions of relatively smooth flow. In case of rectangular buildings,

sufficient field data are available. For these building; turbulent flow has been
used. The design of wall glazing and cladding has become a source of major
concern in recent years. The damage to glass may be hazardous and cause
considerable financial losses.

Cr est Cr est
2 .0 S 2 .0
=
0.
05

1 .5 1 .5
09
0.

0.
1
=
S

1 .0 1 .0
0.
1
0.

2
2
0.

0 .5 0 .5
(H / L e )

(H / L e )
0 .4
0 .6
0 .8
0 1 .0 0
–1 .5 –1 .0 –0 .5 0 0 0 .5 1 .0 1 .5 2 .0 2 .5
Up w in d ( X / Le ) D o w n w in d ( X/ L e )
(a) Fact o r S f o r cli f f an d escarp h en t

Cr est Cr est
2 .0 S= 2 .0
0 .0
5

1 .5 1 .5
09

0 .1
0.
=
S

1 .0 1 .0
0 .2
1
0.

2
0.

0 .5 0 .5
(H / L e )
(H/ Le )

0 .4
0 .6
0 .8
0 1 .0 0
–1 .5 –1 .0 –0 .5 0 0 0 .5 1 .0 1 .5 2 .0 2 .5
Up w in d ( X/ Le ) D o w n w in d ( X/ Le )
(b ) Fact o r S f o r r id g e an d h il l

Fig. 9.6

In Chapter 1, loads and stresses Sec. 1.24, the wind load has been described.
In order to determine the wind load on roofs and other inclined surfaces, the
direction of wind is assumed as horizontal. The roof surfaces are also inclined to
the horizontal. The normal component and tangential component of wind on
inclined roof surfaces may be found. It is to note that these components of
 DESIGN OF ROOF TRUSSES 587

horizontal wind pressure cannot be found by resolving. The tangential component

of wind pressure is small and it is neglected. The normal component of wind
pressure can be found from empirical formula.
In the beginning in 1829, Col. Duchemin, a French Army Officer derived an
empirical formula from experiments. This empirical formula bears his name.
The Duchemin’s formula is

⎛ 2sin α ⎞
pn = p ⋅ ⎜ ⎟ ...(9.5 a)
⎝ 1 + sin2 α ⎠
where, p is the wind pressure on surface in the direction of the wind.
pn = Component of wind pressure normal to the inclined roof sur-
face.
α = Angle which the inclined roof surface makes with the horizon-
tal direction of wind. The angle α is shown in Fig. 9.6
pv = pn . cos α
⎛ 2sin α + cos α ⎞
or pv = p ⋅ ⎜ ⎟ ...(9.5 b)
⎝ 1 + sin2 α ⎠
and ph = pv . sin α
⎛ 2 sin α cos α ⎞
or ph = p ⋅ ⎜ ⎟ ...(9.5 c)
⎝ 1 + sin 2 α ⎠
It is considered that 1.5 kN/m2 wind pressure on surface normal to the direc-
tion of wind is sufficient for roof structure in an exposed position. For the roof
structures, in protected position, 1.00 to 1.25 kN/m2 of wind pressure is suffi-
cient.
The wind pressures are expressed in terms of basic wind pressures described
in Art. 1.24. The design wind load on the roof surface is determined as the
resultant effect of internal air pressure and external air pressure.

9.10 WIND LOAD ON INDIVIDUAL MEMBERS

In order to find the wind load on individual structural elements/members (e.g.,
roofs and walls, and individual cladding units and their fittings, it is essential
to consider the pressure difference between the opposite faces of such elements
or units. For cladding, it is therefore, necessary to know the internal pressure
as well as the external pressure. The wind load, W L acting in a direction
normal to the individual structural element or cladding unit may be determined
from the following expression.
WL = (Cpe – Cpi) . A . pd ...(9.13)
where,
Cpe = external pressure coefficient
Cpi = internal pressure coefficient
A = surface area of structural element or cladding unit, and
pd = design wind pressure
588 DESIGN OF STEEL STRUCTURES

The external pressure coefficients and internal pressure coefficients

are described in the subsequent sections.
It is note that the external pressure coefficient is positive if it is acting out-
side to inside. However, the internal air pressure is treated positive if it is act-
ing inside to outside.

9.11 EXTERNAL PRESSURE COEFFICIENTS

Average external pressure coefficients shall be considered for walls, pitched
roofs, monoslope roofs and canopy roofs, as described hereunder:

C
C
A B A B
D
0 .25 W (a )
C
0 .25 W D
(d ) 0 .25 W
C

C
A B
A B A B

D (e )
(b )
C

C D
A B
A B
D

D
Elevatio n (Plan ) Elevatio n (Plan ) Elevatio n (Plan )
(c) (f) (g )

Fig. 9.7 Walls of rectangular clad buildings

9.11.1 Walls
The average external pressure coefficients for the walls of clad buildings of
rectangular plan as shown in Fig. 9.7, are given in Table 9.6.

9.11.2 Pitched Roofs of Rectangular Clad Buildings

The average external pressure coefficients and pressure concentration
coefficients for pitched roofs of rectangular clad building as shown in Fig. 9.8
 DESIGN OF ROOF TRUSSES 589

are noted from Table 9.7, where, the pressure coefficients are not given, the
average coefficients shall be used.
It is to note that the pressure concentration is assumed to act outward (suc-
tion pressure) at the ridges, eaves, cornices and 90 degree corners of roofs. This
pressure concentration is not included with the net external pressure while
calculating overall loads.

Y Y

Y
1

2 4
E G

3 4
W ind
F H

(K e y p la n)

y = h or 0.15 W
(W hich eve r is less)
(a ) (b )

Fig. 9.8 Pitchd roof rectangular clad buildings

 590
Table 9.6 External pressure coefficients. Cp for walls of rectangular buildings (as per IS : 875 (Part 3) – 1987)

Building Building plan Wind CPc for surfaces Local, Cpe

Height Ratio Ratio Angle, θ A B C D shaded area
(h/w) < 0.5 1 < (L/W) < 1.5 0° +0.7 –0.2 –0.5 –0.5 –0.8
Fig. 9.7 (a) 90° –0.5 –0.5 +0.7 –0.2
1.5 < (L/W) < 4.0 0° +0.7 –0.25 –0.6 –0.6 –1.0
Fig. 9.7 (b) 90° –0.5 –0.5 +0.7 –0.1
1 < (L/W) < 1.5 0° +0.7 –0.25 –0.6 –0.6 –1.1
0.5 (h/w) < 1.5 Fig. 9.7 (c) 90° –0.6 –0.6 +0.7 –0.251
1.5 < (L/W) < 4.0 0° +0.7 –0.3 –0.7 –0.7 –1.1
Fig. 9.7 (d) 90° –0.5 –0.5 +0.7 –0.1
1 < (LIW) < 1.5 0° +0.8 –0.25 –0.8 –0.8 –1.2
1.5 (h/w) < 6 Fig. 9.7 (e) 90° –0.8 –0.8 +0.8 –0.25
1.5 < (L/W) < 4.0 0° +0.7 –0.4 –0.7 –0.7 –1.2
Fig. 9.7 (f) 90° –0.5 –0.5 +0.8 –0.1
DESIGN OF STEEL STRUCTURES

(L/W) = 1.5 0° +0.95 –1.85 –0.9 –0.9 –1.2

Fig. 9.7 (g) 90° –0.8 –0.8 +0.9 –0.85
(L/W) =1.0 0° +0.95 –1.25 –0.7 –0.7 –1.25
(h/w) > ∞ Fig. 9.7 (g) 90° –0.7 –0.7 +0.95 –1.25
(L/W) = 2 0° +0.85 –0.75 –0.75 0.75 –1.25
Fig. 9.7(g) 90° –0.75 –0.75 +0.85 –0.75
Note. h is the height to caves or parapet
L is the greater horizontal dimension of building and
W is the lesser horizontal dimension of building.
 DESIGN OF ROOF TRUSSES 591

Table 9.7 External pressure coefficients, Cpe for pitched roof of rectangular clad build-
ings (As per IS: 875 (Part 3)–1987)

Building Angle of Wind Wind Local

height Roof Angle, θ Angle, θ coefficients
ratio α EF GH EG FH 1 2 3 4
0° –0.8 –0.4 –0.8 –0.4 –2.0 –2.0 –2.0
5° –0.9 –0.4 –0.8 –0.4 –1.4 –1.2 –1.2 –10
10° –1.2 –0.4 –0.8 –0.6 –1.4 –1.4 –1.2 –1.2
(h/w) ≤ 0.5 20° –0.4 –0.4 –0.7 –0.6 –1.0 –1.2
30° 0 –0.4 –0.7 –0.6 –0.8 –1.1
45° +0.3 –0.5 –0.7 –0.6 –1.1
60° +0.7 –0.6 –0.7 –0.6 –1.1
0° –0.8 –0.6 –1.0 –0.6 –2.0 –2.0 –2.0
5° –0.9 –0.4 –0.9 –0.6 –20 –20 –2.0 –1.0
10° –1.1 –0.6 –0.8 –0.6 –2.0 –2.0 –1.5 –1.2
0.5 (h/w) ≤ 1.5 20° –0.7 –0.5 –0.8 –0.6 –1.5 –1.5 –1.5 –1.0
30° –0.2 –0.5 –0.8 –0.8 –1.0 –1.0
45° +0.2 –0.5 –0.8 –0.8
60° +0.6 –0.5 –0.8 –0.8
0° –0.7 –0.6 –0.9 –0.7 –2.0 –2.0 –2.0
5° –0.7 –0.6 –0.8 –0.8 –2.0 –2.0 –1.5 –1.0
10° –0.7 –0.6 –0.8 –0.8 –2.0 –2.0 –1.5 –1.2
0.5 (h/w) ≤ 1.5 20° –08 –0.6 –0.8 –0.8 –1.5 –1.5 –1.5 –1.2
30° –1.0 –0.5 –0.8 –0.7 –1.5
40° –0.2 –0.5 –0.8 –0.7 –1.0
50° +0.2 –0.5 –0.8 –0.7
60° +0.5 –0.5 –0.8 –0.7
Note . h is the height to eaves or parapet and w is the lesser horizontal dimension
of the building. Where the local coefficients are not given the overall coefficients apply.
For hipped roofs, the local coefficients for the hip ridge may be conservatively taken
as the appropriate ridge value.

9.11.3 Mono-slope Roofs of Rectangular Clad Buildings

The average pressure coefficients and pressure concentration coefficient for
mono-slope (lean – to) roofs of rectangular clad buildings as shown in Fig. 9.9
may be noted from Table 9.8.
Table 9.8 External pressure coefficients, Cle for mono–slope roofs for rectangular clad buildings with (h/w) < 2 (as per IS : 875 (Part 3)– 592
1987)

Angle of roof Angle of wind, θ Local Cfe

α 0° 45° 90° 135° 180°
(degrees) H L H L H&L H&L H L H L H1 H2 L1 L2 He Le
5 –1.0 –0.5 –1.0 –0.9 –1.0 –0.5 –0.9 –1.0 –0.5 –1.0 –2.0 –1.5 –2.0 –1.5 –2.0 –2.0
10 –1.0 –0.5 –1.0 –0.8 –1.0 –0.5 –0.8 –1.0 –0.4 –1.0 –2.0 –1.5 –2.0 –1.5 –2.0 –2.0
15 –0.9 –0.5 –1.0 –0.7 –1.0 –0.5 –0.6 –1.0 –0.3 –1.0 –1.8 –0.9 –1.8 –1.4 –2.0 –2.0
20 –0.8 –0.5 –1.0 –0.6 –0.9 –0.5 –0.5 –1.0 –0.2 –1.0 –1.8 –0.8 –1.8 –1.4 –2.0 –2.0
25 –0.7 –0.5 –1.0 –0.6 –0.8 –0.5 –0.3 –0.9 –0.1 –0.9 –1.8 –0.7 –0.9 –0.9 –2.0 –2.0
30 –0.5 –0.5 –1.0 –0.6 –0.8 –0.5 –0.1 –0.6 0 –0.6 –1.8 –0.5 –0.5 –0.5 –2.0 –2.0

Note. These values apply to length (w/z) from windward end and
** these values apply to remaining length
h is the height to eaves at lower side
DESIGN OF STEEL STRUCTURES

L is the greater horizontal dimension of a building and

W is the lesser horizontal dimension of the building.
 DESIGN OF ROOF TRUSSES 593

Y
He Le

L1
H1

H L
W ind

θ
H2

L2

y = h or 0.15 w

W h ich ever is less

N o te : A rea H a n d a re a L
R e fer to w h ole q ua dran t

(a ) (b )

Fig. 9.9 Mono slope roofs for rectangular clad buildings with (h/w) < 2

9.11.4 Canopy Roofs with [(L/4) < (h/w) < 1 and 1(< L/W) < 3]
The pressure coefficients are given in Tables 9.9 and 9.10 separately for mono-
pitch and double-pitch canopy roofs (as shpwn in Figs 9.10 and 9.11, respectively)
such as open-air parking garages, shelter areas, outer areas, railway platforms,
stadiums and theatres. These coefficients take into consideration the combined
effect of the wind exerted on and under the roof for all wind direction; the
resultant is to be taken normal to the canopy.
In case the local coefficients overlap, the greater of the two given values should
be considered. However, the effect of partial closure of one side and/or both
sides, such as those due to trains, buses and stored materials shall be foreseen
and considered.
The solidity ratio φ is equal to the area obstruction under the canopy
divided by the gross area under the canopy, both areas normal to wind direction
(φ = 0 represents a canopy with no obstruction under ncath) and (φ) =1 represents
the canopy fully blocked with contents to the downward caves).
594 DESIGN OF STEEL STRUCTURES

Table 9.9 Pressure coefficients for mono-slopes free roofs (as per IS : 875 (Part 3) – 1987)

Angle of roof Solidity Maximum (largest +ve) and Minimum (argest –ve)
(degrees) ratio Pressure coefficients
Overall Local coefficients
coefficients 1 2 3
0 +0.2 +0.5 +1.8 +1.1
5 +0.4 +0.8 +2.1 +1.3
10 All +0.5 +1.2 +2.4 +1.6
15 values +0.7 +1.4 +2.7 +1.8
20 of +0.8 +1.7 +2.9 +2.1
25 φ +1.0 +2.0 +3.1 +2.3
30 +1.2 +2.2 +3.2 +2.4
0 0 –0.5 –0.6 –1.3 –1.4
1 –1.0 –1.2 –1.8 –1.9
5 0 –0.7 –1.1 –1.7 –1.8
1 –1.1 –1.6 –2.2 –2.3
10 0 –0.9 –1.5 –2.0 –2.1
1 –1.3 –2.1 –2.6 –2.7
15 0 –1.1 –1.8 –2.4 –2.5
1 –1.4 –2.3 –2.9 –3.0
20 0 –1.3 –2.2 –2.8 –2.9
1 –1.5 –2.6 –3.1 –3.2
25 0 –1.6 –2.6 –3.2 –3.2
1 –1.7 –2.8 –3.5 –3.5
30 0 –1.8 –3.0 –3.8 –3.6
0 –1.8 –30 –3.8 –3.6

Note. For mono-pitch canopies, the centre of pressure should be taken to act at
0.3W from the windward edge.

2
( L/10)
α 3
–C p
1
h
+C p

(L/10)
2

( W /10) ( W /10)

(a) (b)
Fig. 9.10 Mono-slope free roofs
 DESIGN OF ROOF TRUSSES 595

The values of Cp for intermediate solidities may be linearly interpolated

between these two extremes and apply upwind of the position of maximum
lockage only. Downwind of the position of maximum blockage, the coefficients
for φ = 0 may be used.
Table 9.10 Pressure coefficients for free standing double sloped roofs (as per IS : 875 (Part
3) – 1987)

Angle of roof Solidity Maximum (largest] +ve) and Minimum (argest –ve)
(degrees) ratio Pressure coefficients
Overall Local cofficients
coefficients 1 2 3 4
–20 +0.7 +0.8 +1.6 +0.6 +1.7
–15 +0.5 +0.6 +1.5 +0.7 +1.4
–10 +0.4 +0.6 +1.4 +0.8 +1.1
–5 +0.3 +0.5 +1.5 +0.8 +0.8
+5 All +0.3 +0.6 +1.8 +1.3 +0.4
+10 values +0.4 +0.7 +1.8 +1.3 +0.4
+15 of +0.4 +0.9 +1.9 +1.4 +0.4
+20 φ +0.6 +1.1 +1.9 +1.5 +0.4
+25 +0.7 +1.2 +1.9 +1.6 +0.5
+30 +0.9 +1.3 +1.9 +1.6 +0.7
–20 0 –0.7 –0.9 –1.3 –1.6 –0.6
1 –0.9 –1.2 –1.7 –1.9 –1.2
–15 0 –0.6 –0.8 –1.3 –1.6 –0.6
1 –0.8 –1.1 –1.7 –1.9 –1.2
–10 0 –0.6 –0.8 –1.3 –1.5 –0.6
1 –0.8 –1.1 –1.7 –1.9 –1.3
–5 0 –0.5 –0.7 –1.3 –1.6 –0.6
1 –0.8 –1.5 –1.7 –1.9 –1.4
+5 0 –0.6 –0.6 –1.4 –1.4 –1.1
1 –0.9 –1.3 –1.8 –1.8 –2.1
+ 10 0 –0.7 –0.7 –1.5 –1.4 –1.4
1 –1.1 –1.4 –2.0 –1.8 –2.4
+ 15 0 –0.8 –0.9 –1.7 –1.4 –1.8
1 –1.2 –1.5 –2.2 –1.9 –2.8
+ 20 0 –0.9 –12 –1.8 –1.4 –2.0
1 –1.3 –1-7 –2.3 –1.9 –3.0
+ 25 0 –1.0 –1.4 –1.9 –1.4 –2.0
1 –1.4 –1.9 –2.4 –2.1 –3.0
+ 30 0 –1.0 –1.4 –1.9 –1.4 –2.0
1 –1.4 –2.1 –2.6 –2.2 –3.0
Due to wind pressure on any fascia and to friction over the surface of the
canopy, the horizontal loads also act in addition to the pressure forces normal to
596 DESIGN OF STEEL STRUCTURES

the canopy. For any wind direction, only the greater of these two forces need be
considered. Fascia loads should be found on the area of the surface facing the
wind, using a force coefficient of 1.3.
–CP –CP

α α L
4 4 10
+CP +C P
1 1
L
3 3 10

A A A
(a ) N e ga tive roo f an gle (b ) P o sitive ro of an gle
W
( A = W / 10 )

Fig. 9.11 Free standing double-sloped roofs

D E D
α D' E' α D' E'

C ' an d G ' C ' an d G '

C' C'

G' G'
W ind W ind
θ θ

D D

R 9 0° R 9 0°
b = 5d

b = 5d

E E

G G

C C
R o of S lo pe α = 3 0° R o of S lo pe α = 3 0°
θ = (0 °– 45 °), D , D ,'E , E ' θ = (0 °– 45 °), or (13 5°–1 80 °)
full len gth D , D ,'E , E ' fu ll le ng th
θ = 9 0° D ,D ', E , E ' θ = 9 0° D ,D ', E , E '
P art le ng th b ' P art le ng th b '
Th ere a fte r C p = 0 Th ere a fte r C p = 0
(Effe cts of tra in s/store d m a te ria l)
(a ) (b )

Fig. 9.12 Pitched free roofs α = 30°

 DESIGN OF ROOF TRUSSES 597

For pitched and toughed free roofs for some specific cases for which aspect
ratios and roof slopes have been specific as shown in Fig. 9.12 (a), and (b), Fig. 9.13
(a) and (b) and Fig. 9.14 (a) and (b) the internal and external pressure coef-
ficients may be noted from Tables 9.11, 9.12, 9.13 (a), 9.13 (b), 9.14 (a) and 9.14
(b), respectively. However while using these Tables, any significant departure
from it should be investigated carefully. No increase shall be made for the local
effects except as indicated.

Table 9.11 Pressure coefficients (top and bottom) for pitched roofs, a = 30°
(As per IS : 875 (Part 3) – 1987)

Angle of wind Pressure coefficients, Cp

θ D D´ E E´ End Surfaces
C C´ G G´
0° 0.6 –1.0 – 0.5 – 0.9 — — — —
45° 0.1 –0.3 –0.6 –0.3 — — — —
90° –0.3 –0.4 –0.3 –0.4 –0.3 0.8 0.3 0.4
For j : Cp (top) = –1.0; Cp (bottom) = – 0.2
Tangentially acting friction : R 90° = 0.05 pd . bd
D E D E
α D' E' α D' E'

C ' an d G ' C ' an d G '

C' C'

G' G'

W ind
θ
W ind
D D E
O' E O' θ

R 9 0° R 9 0°

f=
0 .2 d
G G

C C
R o of S lo pe α = 1 0° R o of S lo pe α = 1 0°
θ = (0 °– 45 °), D ,D ,'E , E ' θ = (0 °– 45 °), or (13 5°–1 80 °)
fu ll le ng th D , D ,'E , E ' fu ll le ng th
θ = 9 0° D ,D ', E , E ' θ = 9 0° D , D ', E , E '
P art le ng th b ' Pa rt len gth b '
Th ere a fte r C p = 0 Th ere a fte r C p = 0
(Effe cts of tra in s/sto re d m a te ria l)
(a ) (b )
Fig. 9.13 Pitched free roofs, α = 10°
598 DESIGN OF STEEL STRUCTURES

Table 9.12 Pressure coefficients (top and bottom) for pitched roofs, a = 30°
With effect of train or stored material (As per IS : 875 (Part 3) – 1987)
Angle of wind Pressure coefficients, Cp
θ D D´ E E´ End Surfaces
C C´ G G´
0° 0.1 +0.8 + 0.7 + 0.9 — — — —
45° 0.1 +0.5 –0.8 +0.5 — — — —
90° –0.4 –0.5 –04 –0.5 –0.3 0.8 0.3 0.4
180° –0.3 –0.6 –04 –0.6 — — –– —
45° For f Cp (top) = –1.5; Cp (bottom) = – 0.5
Tangentially acting friction : R 90° = 0.05 pd . bd

Table 9.13 (a) Pressure coefficients (top and bottom) for pitched roofs, a = 10°
(As per IS : 875 (Part 3) – 1987)
Angle of wind Pressure coefficients, Cp
D D´ E E´ End Surfaces
C C´ G G´
0° –1.6 +0.3 – 0.5 – 0.2 – – – –
45° –0.3 +0.1 –0.3 –0.2 – – – –
90° –0.3 0 –0.3 0 –0.4 0.8 0.3 0.6
0° For f : Cp (top) = –10; Cp (bottom) = + 0.4
0° – 90° Tangentially acting friction : R 90° = 0.01 pd . bd
R o° R o°
D E D E
h =0 .5 d

h =0.8 h

D' E' D' E'

W ind
b' = d

D E W ind D E
θ θ

R 9 0°
R 9 0°
(b = 5d )

b = 5d

f=
0 .2 d

R o of S lo pe α= 1 0° R o of S lo pe α= 1 0°
θ = (0 °– 45 °), D , D ' , E , E ' fu ll le n gth θ = (0 °– 45 °), or (13 5°–1 80 °) D , D ', E , E ' full len gth
θ = 9 0° D , D ', E , E ' Pa rt len gth b ' θ = 9 0° D , D ', E , E ' Part le ng th b '
Th ere a fte r C p = 0 Th ere a fte r C p = 0 (Effe cts o f trains/sto red m aterial)
(a ) (b ) Effe cts of tra ins/sto re d m a terials
Fig. 9.14 Troughed free roofs α = – 10°
 DESIGN OF ROOF TRUSSES 599

Table 9.13 (b) Pressure coefficients (top and bottom) for pitched roofs, a = 10°
With effect of train or stored material (As per IS : 875 (Part 3) – 1987)

Angle of wind, Pressure coefficients, Cp

θ D D´ E E´ End surfaces
C C´ G G´
0° –0.3 +0.8 – 0.6 +0.7 — — — —
45° –0.5 +0.4 –0.3 +0.3 — — — —
90° –0.3 0 –0.3 0 0.4 +0.8 +0.3 –0.6
180° –0.4 –0.3 0.6 0.3 — — — —
0° For f : Cp (top) = –1.6; Cp (bottom) = + 0.9
0° – 180° Tangentially acting friction : R 90° = 0.01 pd . bd

Table 9.14 (a) External pressure coefficients for troughed free roofs, α = 10°
(As per IS : 875 (Part 3) – 1987)

Angle of wind, Pressure coefficients, Cp

θ D D´ E E´
0 0.3 –0.7 0.2 –0.9
45 0 –0.2 0.1 –0.3
90 –0.1 +0.1 – 0.1 0.1
0° For f : Cp (top) = 04, Cp (bottom) = –1.5
0°–90° Tangentially acting friction : R 90° = 0.1 pdd .bd

Table 9.14 (b) Pressure coefficients (top and bottom) for troughed free roofs, a = 10° with
effects of train or stored materials (As per IS : 875 (Part 3) – 1987)

Angle of wind, Pressure coefficients, Cp

θ D D´ E E´
0 –0.7 0.8 –0.9 0.6
45° –04 0.3 –0.2 0.2
90° –0.1 0.1 –0.1 0.1
180° –0.4 –0.2 –0.6 –0.3
0°
0°–180°

9.12 INTERNAL PRESSURE COEFFICIENTS

The internal air pressure in a building depends upon the degree of
permeability of the structure (cladding) to the flow of air. The internal air
pressure may be positive or negative (that is, a positive pressure or suction)
depending upon the direction of flow of air in relation to the openings in the
buildings. Four types of building having different degrees of permeability and
internal pressures shall be considered.
600 DESIGN OF STEEL STRUCTURES

9.12.1 Building Having a Small Degree of Permeability

In the case of buildings where the cladding permits the flow of air with open-
ings not more than about 5 percent of the wall area but where there are no large
openings, it is necessary to consider the possibility of the internal pressure
being positive or negative. Following two design conditions shall be examined.
1. A building with an internal pressure coefficient of + 0.2, and
2. A building with an internal pressure coefficient of – 0.2.

9.12.1.1 Design Wind Pressure on Roofs

The internal pressure coefficient is algebraically added to the external pressure
coefficient and the analysis which indicates more distress of the member shall
be adopted. In most situations, a simple inspection of the sign of external pres-
sure will at once indicate the proper sign of the internal pressure coefficient to
be taken for design.
The term normal permeability relates to the flow of air commonly afforded
by the structures (claddings). It includes the flow of air not only through open
windows and doors, but also through the slits round the closed windows, doors,
ventilators, chimneys and through the joints between the roof coverings, the
total open area being less than 5 percent of area of the walls having the open-
ings.

9.12.2 Buildings Having Medium Degree of Permeability

Buildings with medium openings may also show either positive or negative
internal pressure depending upon the direction of wind. The buildings with
medium openings between about 5 to 20 percent of wall area shall be designed
for an internal pressure coefficient of + 0.5 and again with an internal pressure
coefficient of – 0.5. The analysis shall be done for maximum distress of the
members.

9.12.3 Buildings Having Large Degree of Permeability

Buildings with large openings, that is, openings larger than 20 percent of the
wall area shall be designed once with an internal pressure coefficient of + 0.7
and again with an internal pressure coefficient of – 0.7. The analysis shall be
carried out for maximum distress on the members.

9.12.4 Buildings Having High Degree of Permeability

Buildings with one open side or having opening exceeding 20 percent of wall
area may be assumed to be subjected to internal positive pressure or suction
(internal negative pressure) similar to those for buildings with large openings.
The buildings with one sided openings are shown in Fig. 9.15. The values of
internal pressure coefficients with respect to the direction of wind are also
indicated.
 DESIGN OF ROOF TRUSSES 601

The values of intenal pressure coefficients for all the four above categories of
buildings have been noted briefly in Table 9.15.
Table 9.15 Internal pressure coefficients, Cpi (as per IS : 875 (Part 3) –1987)

S. No. Categories of buildings Cpi

1. Buildings with small degree of +0.2
permeability (openings |> percent) – 0.2
2. Buildings with medium degree of +0.5
permeability (openings 5 to 20 percent) – 0.5
3. Buildings with large degree of +0.7
permeabilities (opening > 20 percent) – 0.7
4. Buildings with one open side or Fig. 9.15
(opening > 20 percent)

Note. For (B/L) equal to unity, average values may be used.

W ind

W ind W ind
+0 .8 +0 .5 +0 .4

(a ) Fo r ( B / L ) Le ss tha n un ity
W ind
W ind W ind
+0 .8 +0 .7 +0 .3

( b ) Fo r ( B / L ) m ore th a n u nity

(Note : for (B/L) = 1 average values are used)

Fig. 9.15

9.13 EXTERNAL PRESSURE COEFFICIENTS FOR SOME OTHER

ROOFS
The external pressure coefficients for the following more types of roofs are
adopted as specified in IS : 875 (Part 3)–1987.
602 DESIGN OF STEEL STRUCTURES

9.13.1 Pitched and Saw-tooth Roofs of Multi-span Building

For the pitched and saw-tooth roofs of multispan buildings, as shown in Figs.
9.16 and 9.17 (see page 602–603) the external average pressure coefficients and
pressure concentration coefficients may be noted from Tables 9.16 and 9.17,
respectively provided all the spans are equal and the height to eaves shall not
exceed the span.

9.13.2 Pressure Coefficients on Overhangs from Roofs

The pressure coefficients on the top overhanging portion of the roof are taken to
be the same, as that of the nearest top portion of the non-overhanging portions
of the roofs. The pressure coefficients for the underside surface of the portions
are taken as follows and shall be taken as positive in case the overhanging
portion is on the windward side.
Overhangs External pressure coefficients, Cpc
(a) sloping downward 1.25
(b) horizontal 1.00
(c) loping upward 0.75

Table 9.16 External pressure coefficients, (Cpe) for pitched roofs of multispan buildings
(all spans equal) with (h > w)(As per IS : 875 (Part 3) –1987)

Angle of Angle of First First Other End Local

roof, a wind θ span interspan intermediate span coefficients
(degrees) (degress) a b c d m n x z 1 2
5 0 –0.9 –0.6 –0.4 –0.3 –0.3 –0.3 –0.3 –0.3 –2.0 –1.5
10 –1.1 –0.6 –0.4 –0.3 –0.3 –0.3 –0.3 –0.4 –2.0 –1.5
20 –0.7 –0.6 –0.4 –0.3 –0.3 –0.3 –0.3 –0.5 –2.0 –1.5
30 –0.2 –0.6 –0.4 –0.3 –0.2 –0.3 –0.2 –0.5 –2.0 –1.5
45 +0.3 –0.6 –0.6 –0.4 –0.2 –0.4 –0.2 –0.5 –0.2 –1.5
Distance
Angle of Angle of h1 h2 h3
roof, α wind θ
upto 45° 90° –0.8 –0.6 –0.2

Note : It is to note that when the angle of wind q is 0°, the horizontal forces due to
frictional drag are allowed for in the above values. And when the angle of wind is
90°, the above values are allowed for frictional drag as specified in IS : 875 (Part 3)
1987.
 DESIGN OF ROOF TRUSSES 603

Table 9.17 External pressure coefficients, (Cpe) for saw-tooth roofs of multi-span buildings
(all span equal) with (h |> w) (as per IS : 875 (Part 3) – 1987)

Angle of First First Other End Local

wind, θ span intermediate intermediate spans coefficients
(degrees) span span
a b c d m n x z 1 2
0 +0.6 –0.7 –0.7 –0.4 –0.3 –0.2 –0.1 –0.3 –2.0 –1.5
180 –0.5 –03 –0.3 –0.3 –0.4 –0.6 –0.6 –0.1 –2.0 –1.5
Distance
Angle of wind, θ h1 h2 h3
90° –0.8 –0.6 –0.2
270° similarly, but handed
Note : It is to note that when the angle of wind, θ is zero, the horizontal forces due
to frictional drag are allowed for in the above values. And when the angle of wind, 8
is 90°, the frictional drag is allowed as recommended in IS : 875 (Part 3) – 1987.
It is to note that for the overhanging portions on the sides othe than the
windward side, the average pressure coefficients on adjoining walls may be
used.

9.13.3 Combined Roofs and Roofs with a Sky-light

The average external pressure coefficients for combined roofs with a sky light as
shown in Figs. 9.18 and 9.19 may be noted from Tables 9.18 and 9.19, respectively.

0 .1 W ´
y
0 .1 W ´
h1
W
in d
θ h2
W
h3
y

W' W´ W´ W´ W´ W´
R o of plan
y = h or 0.1 w h ich ever is le ss (in de x. 1 . 2. )

a b c d m n m n x z

S e ctio n

Fig. 9.16 Pitched roofs of multispan buildings (all spans equal with h ( |
> w')
604 DESIGN OF STEEL STRUCTURES

0 .1 w ´
y
0 .1 w ´
Wi h1
nd
θ h2
w
h3

w´ w´ w´ w´ w´ w´
R o of plan
y = h or 0.1 w w hich ev e r is less h 1 = h 2 = h
(In de x: 1 . 2. )
a b c d m n m n m n x z

Fig. 9.17 Saw-tooth roofs of multispan buildings (all spans equal with (h > w´)

Table 9.18 External pressure coefficients, Cpe for combined roofs and roofs with sky-light
per IS : 875 (Part 3) – 1987) values of Cpe

Potion Direction 1 Direction 2

a From Fig. 9.18
b Cpe = –0.5 (h1/h2) < 1.5
Cpe = –0.7 (h1/h2) > 1.5 – 0.4
c and d See Table 9
e See Sec. 9.13 (2)

Table 9.19 External pressure coefficients, Cpe for combined roofs and roofs with sky-light
(as per IS : 875 (Part 3)– 1987) pressure coeffients, Cpe

b1 > b2 b1 ≤ b2
Portions Portions
a b a b
–0.6 +0.7 Table 9.18 Table 9.18

9.14 COMBINATION OF LOADS ON ROOF TRUSSES

A judicious combination of loads on roof trusses specified in Sec. 9.6, keeping
in view the possibility of (i) their action together and (ii) their disposition in
relation to other loads and the severity of stresses or deformation caused by
the combination of various loads is necessary to ensure by the combination
of various loads is necessary to ensure the required safety and economy in the
design of roof trusses.
 DESIGN OF ROOF TRUSSES 605

The combination of loads on roof trusses is used to determine the maximum

stresses in the members of a truss. The following combination of loads are used:
(i) Dead load + snow load, if any.
(ii) Dead load + partial or full live load whichever causes the maximum
stresses in the members.
(iii) Dead load + wind load + internal pressure.
(iv) Dead load + wind load + internal suction.

D irectio n 2 α=
0 .8 30 °
D irectio n 1 c d 1
b1 = b2
b
0 .6 a
h1
h2
a
0 .4
b1 b2
h1
C p e =0 .4 h -.0 6
0 .2 2
0 .13
1 .2
C pe 0 .5 1 .0 1 .5 1 .82 0 2 .5 3 .0 3 .5 h 1
0
h2
– 0.2 a h1
C p e h 1.7
2
– 0.4 h 1d D irectio n 2
D irectio n 1 c

– 0.6 b
h1 a
C p e h 2 .9 h1
2 e h2
– 0.8
a
– 1.0

Fig. 9.18 Combined roofs and roofs with a sky light

W ind W ind – 0.8

+0 .4 – 0.6 b – 0.6
– 0.6 – 0.5 a
– 0.5
h2
h1

b1 b2
Fig. 9.19 Combined roofs and roofs with a sky light

9.15 BRACING OF ROOF TRUSSES

The bracing of roof trusses is done to provide a stiff rigid structures so that
structure is not subjected to vibrations due to rolling loads of machinery. In
building structures, the roof trusses are supported on masonry walls. The pur-
lins at ends also rest on end masonry walls. For such structures, lateral bracing
606 DESIGN OF STEEL STRUCTURES

is not necessary. The purlins provide sufficient lateral support. The roof trusses
are also securely fixed with anchor bolts in the masonry walls.
When the roof trusses are supported over steel columns, then lateral bracing
should be provided for stability against lateral forces. In absence of lateral
bracing, the structure will collapse in a high wind storm. The structure may
collapse due to stress and vibrations due to rolling loads. The lateral bracing
provided should be such that it transmits lateral forces directly as possible to
the walls or foundations of the buildings.
The roof trusses should be braced in the plane of top chord and also in the
plane of bottom chord. The bracings should also be provided in the plane of
column bracing is provided in every third or fourth panel. The bracing of a truss
in top and bottom chords is shown in Fig. 9.20. The bracings are provided in
such a manner that their diagonals form angles about 45° with the loads to be
carried. The bracings may consist of single angle sections.

(a ) R o of tru ss

(b ) B ra cin g in plan e o f top cho rd

(c) B ra cin g in plan e of bo tto m cho rd

Fig. 9.20 Bracing of roof trusses

9.16 ASSUMPTIONS IN ANALYSIS OF ROOF TRUSSES

The analysis of roof trusses consist of two parts. Part one consists of
determination of loads and reactions. The other part consists of determination
of the internal forces in the members of roof trusses. The following assumptions
are made for determining the forces (stresses) in the members :
 DESIGN OF ROOF TRUSSES 607

1. The roof truss is not restrained by the reactions.

2. The axes of members meeting at a joint intersect at a common point.
3. The riveted joints in roof trusses act as frictionless hinges.

9.17 PURLINS
The rolled steel sections and rectangular wooden beams are used as purlins in
the roof trusses. The purlins are placed with their webs or sides perpendicular
to the top chords of roof as shown in Fig. 9.21. The principal axes of rolled steel
I-section, channel section, rectangular wooden section shown in Fig. 9.21 (a),
(b) and (c) respectively are parallel to the sides of sections. Whereas Fig. 9.21 (d)
shows rolled steel angle section. One leg of angle section is normal to the top
chord of roof truss. It supports the normal component of loading. The principal
axes of angle section are not parallel to its sides. In general, when a structural
member is subjected to transverse loading, the plane of bending (or plane of
loading coincides with one of the principal axes). The neutral axis of the section
coincides with other principal axes. Whereas in case of purlin, the line of action
of resultant of vertical loads and normal component of wind load do not coincide
with any of the principal axes. Therefore, the bending of purlins occurs in a
plane other than the principal plane of section. Such a bending is known as
unsymmetrical bending. The purlins are subjected to unsymmetrical bending.
Plan e of Plan e of u
v

lo a din g lo a din g
v

x
v
u
u

v
u

x
u
u

u
v
v

Princip a l ra fte r S e ctio n Princip a l ra fte r S e ctio n

(a ) (b ) (c) (d )
Fig. 9.21 uu, vv principal axes

I-scction shown in Fig. 9.21 (a), rectangular section shown in Fig. 9.21 (c)
have two axes of symmetry. The channel section shown in Fig. 9.21 (b) has only
one axis of symmetry. Whereas angle section shown in Fig. 9.21 (d) has no axes
of symmetry.

9.18 PRINCIPAL AXES OF SECTION

The moment of inertia of a section about any axis is defined as the second
moment of an area about that axis. Figure 9.22 shows any plane section. The
point O represents centre of gravity of the section. xx-axis and yy-axis represent
two co-ordinates axes. The co-ordinates axes pass through the point O, and
these are perpendicular to each other. (x, y) represent co-ordinates of any
elementary area δA.
Then, Ixx = Σy2 δA, Iyy = Σx2 δA,Ixy = Σxy ΣA
608 DESIGN OF STEEL STRUCTURES

where, Ixx = Moment of inertia of the section about xx-axis

Iyy = Moment of inertia of the section about yy-axis
Ixy = Product of inertia about xx-axis and yy-axis of section.
The product of inertia of a section is determined by taking moment of the
area about one axis, and then taking the moment of this moment about other
axis.

Y
V

U in ∝
xs
ys in ∝
∝

∝ V y = co s ∝
Y U
∝ ∝
x s in

x ∝ x
x

Y V

Fig. 9.22 Principal axes of a section

Let UU and VV be the two perpendicular axes and these axes pass through
point O as shown in Fig. 9.22. (u,v) represent co-ordinates of the elementary
area ΣA with reference to UU-axis and VV-axis.
Then; Iuu = Σv2δA, Ivv = Σu2δA, Iuv = Σuvδ4
where, Iuu = Moment of inertia about UU-axis
Ivv = Moment of inertia about VV-axis
Iuv = Product of inertia about UU-axis and VV-axis.
The principal axes of a plane section are defined as two perpendicular axes in
its plane passing through the centre of gravity of the section, such that the
product of inertia is zero.
If the product of inertia, Iuv is zero, then, UU-axis and VV-axis are known as
principal axes of the plane section. The angle of inclination of principal axes
with XX-axis and YY-axis is as shown in Fig. 9.22. If any plane section has an
axis of symmetry, then the product of inertia about that axfs is zero. The axis of
symmetry represents one of the principal axis. The other principal axis is at
right angle to this axis of symmetry. If any plane section has two axes of
symmetry, then both the axes of symmetry represent principal axis of the plane
section. I-section and rectangular section are shown in Fig. 9.21 (a) and
Fig. 9.21 (c). UU-axis and VV-axis are two axes of symmetry. These axes are
 DESIGN OF ROOF TRUSSES 609

principal axes of the sections. The channel section shown in Fig. 9.21 (b) has
one axis of symmetry. The axes of symmetry UU is principal axis. The other
principal axis VV is at right angle to this axis and passes through C.G. of section.
Angle section shown in Fig. 9.21 (d) has no axis of symmetry. Such I sections
are called unsymmetrical section. The principal axes for such sections are
located either by analytical method or by graphical method.

9.18.1 Analytical Method

The principal axes of a plane section shown in Fig. 9.22 can be located by deter-
mining angle of inclination, a with UU-axis and XX-axis as below. The co-ordi-
nates (u, v) of the elementary δA with reference to UU-axis and W-axis may be
expressed in terms of co-ordinates (x, y) and α
u = (x . cos α + y . sin α) ...(i)
v = (y . cos α –x . sin α) ...(ii)
The product of inertia about UU-axis and VV-axis,
IUV = ΣuvδA ...(iii)
= Σ (x . cos α + y . sin α) (y . cos α – x . sin α) δA
= (Σxy . cos2 α . δA) – (Σx2 . sin α cos α . δA)
– (Σxy . sin2 α δA) + (Σx2 . sin α cos α . δA)
IUV = (cos2 α – sin2 α) Σxy . δA + sin α . cos α (Σy2 . δ4 – Σx2 . δA)
1
= (cos 2 α . Σxy . δA) + sin 2α . (Σy2 . δA – Σx 2 . δA)
2

⎛ I xx – I yy ⎞
= cos 2α°. Ixy + ⎜ ⎟ ⋅ sin 2α
⎝ 2 ⎠
For the principal axes, the product of inertia of the plane section about those
axes is zero.
Therefore, IUV = 0 ... (iv)
⎛ I xx – I yy ⎞
cos 2α . Ixy + ⎜ ⎟ ⋅ sin 2α = 0 ...(v)
⎝ 2 ⎠

⎛ –2 I
xx ⎞
tan 2α = ⎜ ⎟ ...(9.14)
⎝ xx I yy ⎠
I –
From this, two values, (α and 90° + α) are obtained. UU-axis is inclined to
XX-axis by an angle α, and VV-axis is inclined to the same axis by an angle
(90°+ α). Thus, two principal axes UU and VV of the plane section are located
with reference to XX-axis.
The values of principal moment of inertia lUU and IVV can be found in terms
of Ixx, Iyy and Ixy as below :
IUU = Σv2 – δA
= Σ (y . cos α – x . sin α)2 . δA
= (Σ cos2 α . y 2 δA) + (Σ sin2 α x2 δA) – (Σ2 sin α cos α . xy δA)
610 DESIGN OF STEEL STRUCTURES

= Ixx . cos2 α+ Iyy . sin2 α – Iyy . sin 2α ...(9.15)

⎛ 1 + cos 2α ⎞ ⎛ 1 – cos 2α ⎞
= I xx ⋅ ⎜ ⎟ + I yy ⋅ ⎜ ⎟ – I ay sin 2α
⎝ 2 ⎠ ⎝ 2 ⎠

⎛ I xx + I yy ⎞ ⎛ I xx – I yy ⎞
= ⎜ ⎟⎜ ⎟ cos 2α – I xy sin 2α
⎝ 2 ⎠⎝ 2 ⎠
From value of tan 2α, Eq. 9.6, we have
– I xy
sin 2 α = 1/ 2
...(i)
⎡⎛ I – I ⎞ 2 ⎤
⎢⎜ xx yy
⎟
2
+ I xy ⎥
⎣⎝ 2 ⎠ ⎦

1
2
( I xx – I yy )
cos 2 α = 1/ 2
...(ii)
⎡⎛ I – I ⎞2 ⎤
2
⎢⎜ xx yy
⎟ + I xy ⎥
⎣⎝ 2 ⎠ ⎦
Substituting the values of sin 2α and cos 2α, we have
1/ 2
2
⎛ I + I yy ⎞ ⎡⎛ I xx – I yy ⎞ ⎤
IUU = ⎜ xx ⎟ ⎢ ⎜ ⎟
2
+ I xy ⎥ ...(iii)
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
Similarly,
IVV = Σu2 δA
= Σ(x . cos α + y . sin α)2 δA
= (Σ cos2 α . x2 δA) + (Σ sin2 α y2 δA) – (Σ 2 sin α cos α . x . y · δA)
= Iyy . cos2 α + Ixx . sin2 α – Ixy . sin 2α ...(9.16)
⎛ 1 + cos 2α ⎞ ⎛ 1 – cos 2α ⎞
= I yy ⋅ ⎜ ⎟ + I xx ⋅⎜ ⎟ – I xy sin 2α
⎝ 2 ⎠ ⎝ 2 ⎠

⎛ I xx + I yy ⎞ ⎛ I xx – I yy ⎞
= ⎜ ⎟–⎜ ⎟ cos 2α + I xy sin 2α
⎝ 2 ⎠ ⎝ 2 ⎠
1/ 2
2
⎛ I – I yy ⎞ ⎡⎛ I xx – I yy ⎞ ⎤
= ⎜ xx ⎟ – ⎢ ⎜ ⎟
2
+ I xy ⎥ ...(iv)
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
Thus, the principal moment of inertias are
1/ 2
2
⎛ I + I yy ⎞ ⎡⎛ I xx – I yy ⎞ ⎤
IUU = ⎜ xx ⎟ + ⎢⎜ ⎟
2
+ I xy ⎥ ...(9.17)
⎝ 2 ⎠ ⎣ ⎝ 2 ⎠ ⎦
1/ 2
2
⎛ I xx + I yy ⎞ ⎡⎛ I xx – I yy ⎞ 2
⎤
IVV = ⎜ ⎟ ⎢
– ⎜ ⎟ + I xy ⎥ ...(9.18)
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
 DESIGN OF ROOF TRUSSES 611

If, the values of IUU and IVV are added, then, we have
(IUU + IVV) = (Ixx + Iyy) ...(919)
Thus, it is seen that the sum of moment of inertias about any two perpen-
dicular axes of a plane section, passing through the centre of gravity of section
remains unchanged.

9.18.2 Graphical Method

The principal axes UU and VV of a plane section shown in Fig. 9.23 can also be
located by graphical method. The values of principal moment of inertiasIUU and
IVV can also be found graphically. Following are the graphical methods for
locating the principal axes and determining the values of IUU and IVV.
(i) Mohr’s circle method, and (ii) Circle of inertia method.
1. Mohr’s circle method :
Let OX and OY be the two co-ordinate axes, passing through point O, as
origin, as shown in Fig. 9.23.
Plot OP = Ixx and OQ = Iyyon x-axis.
Draw PR = Ixx perpendicular to OX-axis at point P. PR is plotted upwards to
OX-axis, if the value of Ixy is positive and it is plotted downwards to OX-axis, if
its value is negative. The point C is obtained by bisecting the distance PQ. Joint

I xy
I yy 2∝ X
O V O C P U
I xx
a v = I vv
a v = Ivv

Fig. 9.23 Mohr circle method

C and R . With C as the centre, and CR as radius draw a circle as shown in Fig,
9.23. The circle is known as Mohr’s circle. The circle intersects OX-axis at two
points U and V.
Distance PQ = (Ixx – Iyy)
In the triangle PCR,

⎛ I xx – I yy ⎞
∴ PC = ⎜ ⎟ , PR = Ixy
⎝ 2 ⎠

1/ 2
⎡⎛ I – I ⎞ 2 ⎤
xx yy 2
CR = ⎢⎜ ⎟ + I xy ⎥
⎣⎝ 2 ⎠ ⎦
612 DESIGN OF STEEL STRUCTURES

PR I xx ⎛ 2 I xy ⎞
∴ tan (∠ PCR) = = = ⎜⎜ I – I ⎟⎟
PC 1
2
( I xx – I yy ) ⎝ xx yy ⎠
= tan (–2α)
∴ ∠ PCR = 2α
where, α represents the angle of inclination of principal axis UU of the plane
section. The negative sign shows that angle is to be plotted in the negative
direction, i.e., clockwise. In Mohr’s circle, if CR appears anti-clockwise to OX-
axis, then angle is plotted clockwise. If CR appears clockwise to OX-axis, then
angle is plotted anti-clockwise.
Distance OU = (OC+ CU) = (OC+ CR) (3 CU = CK)
1/ 2
2
⎛ I xx + I yy ⎞ ⎡⎛ I xx + I yy ⎞ 2
⎤
⎜⎝ ⎟⎠ + ⎢ ⎜⎝ ⎟⎠ + I xy ⎥ = IUU (3 CV = CR)
2 ⎣ 2 ⎦
Distance OV = (OC – CV) = (OC – CR) (3 CV = CR)
1/ 2
2
⎛ I xx + I yy ⎞ ⎡⎛ I xx – I yy ⎞ 2
⎤
=⎜ ⎟ ⎢
– ⎜ ⎟ + I xy ⎥ = IVV
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
Thus, OU and OV in Mohr’s circle method give principal moment of inertias
IUU and IVV, respectively.
(ii) Circle of inertia method :
Let OX and OY be the two co-ordinate axes passing through point O, as origin
as shown in Fig. 9.24.
Plot OP = Ixx on y-axis and PQ = Iyy on x-axis.
Draw PR = Ixy perpendicular to CY-axis at point P. PR is plotted towards
positive direction i.e., right if Ixy is positive and towards negative direction, i.e.,
left if Ixy is negative. The point C is obtained by bisecting the distance OQ. With
C as the centre and OC as the radius, draw a circle as shown in Fig. 9.24. The
circle is known as circle of inertia. Join C and R and produce it on both sides. It
intersects the circle at points U and V.
Distance OQ = (Ixx + Iyy)

⎛ I xx + I yy ⎞
OQ = CQ = ⎜ ⎟ = Radius of circle
⎝ 2 ⎠
Then, distance PC = (CQ + PQ)

⎛ I xx + I yy ⎞ ⎛ I – I yy ⎞
⎜ – I yy ⎟ = ⎜ xx ⎟
⎝ 2 ⎠ ⎝ 2 ⎠
In ∆PCR,

⎛ I xx – I yy ⎞
CP = ⎜ ⎟ and PR = Iyy
⎝ 2 ⎠
 DESIGN OF ROOF TRUSSES 613

1/ 2
⎡⎛ I – I ⎞ 2 ⎤
CR = ⎢⎜ xx yy
⎟
2
+ I xy ⎥
⎣⎝ 2 ⎠ ⎦
I xx
PR I
tan ∠PCE = = ⎛ xx – I yy ⎞
PC ⎜ ⎟
⎝ 2 ⎠
I xx
tan (∠–PCR) =
( I xx – I yy )
∴ ∠PCR = 2 α
where, α represents the angle of inclination of principle axis UU of the plane
section. The negative sign shows the angle is to be plotted in the negative direc-
tion, i.e., clockwise. OU and OV give the directions of the principal axes UU and
VV.

Q Y

I VV
= V Iyy
R
U R
P
IY X
∝
I UU C
=
R
U I xx

∝
U

O ∝ X
V1 U1

Fig. 9.24 Circle of inertia method

Distance UR = (UC + CR)

= (OC + CR) (3 UC = OC)
1/ 2
2
⎛ I xx + I yy ⎞ ⎡⎛ I xx – I yy ⎞ 2
⎤
= ⎜ ⎟ ⎢
+ ⎜ ⎟ + I xy ⎥ = IUU
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
Distance VR = (VC – CR) = (OC – CR) (3VC = OC)
1/ 2
2
⎛ I xx − I yy ⎞ ⎡⎛ I xx – I yy ⎞ 2
⎤
= ⎜ ⎟ – ⎢ ⎜ ⎟ + I xy ⎥ = IVV
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
Thus, UR and VR in circle of inertia method give principal moment of iner-
tias IUU and IVV respectively.
614 DESIGN OF STEEL STRUCTURES

9.19 MAXIMUM BENDING STRESS DUE TO UNSYMMETRICAL

BENDING
When a structural member is subjected to transverse loading such that the
bending of beam occurs in the principal plane of section, then the neutral axis of
the section of beam is perpendicular to the plane of bending. The neutral axis
passes through the centroid of section. If the transverse loading is such that the
bending of beam occurs in a plane other than the principal plane of section, the
beam section is subjected to unsymmetrical bending. The direction of neutral
axis is not perpendicular to plane of bending. The maximum bending stress in a
beam-section subjected to unsymmetrical bending may be determined by
the following methods:
1. By resolving bending moment into two components along the principle
axes.
2. By resolving bending moment in two components along any two
perpendicular axes passing through the centroid of the section.
3. By locating the neutral axis of section.

9.19.1 By Resolving Bending Moment into two Components Along

the Principal Axes
Let the beam section is subjected to a bending moment, M. The plane of bending
(i.e., the plane of loading) is inclined at θ with the principal VV-axis as shown in
Fig. 9.25. It is noted that angle θ is considered positive when measured in anti-
clockwise direction from OV-axis.

M V N
P Q
Plan e of
b en ding
θ (L oa ding )

V (9 0– θ)
U β
U U
β O
V

N S R
V

Fig. 9.25 Unsymmetrical bending

The bending moment, M may be resolved into two components along the
principal axes UU and VV. The component of bending moment M along UU-
axis is M sin θ, and that alongVV-axis M cos θ. After resolving the bending
 DESIGN OF ROOF TRUSSES 615

moment, into its two components, the theory of simple bending can be applied.
For the component, M sin θ, the bending occurs in the principal plane passing
through UU-axis, and VV-axis becomes a neutral axis. For the component M cos
θ, the bending occurs in the principal plane passing throughVV-axis and UU-
axis becomes a neutral axis.
The bending stress at any point X, may be found by algebraic sum of the
stress due to M . sin θ and M . cos θ. The bending stress at point X is given by
⎛ M ⋅ cos θ M ⋅ sin θ ⎞
σb = ⎜ v+ u⎟ ...(9.20)
⎝ IUU IVV ⎠
where, IUU = Principal moment of inertia of the section about UU-axis
IVV = Principal moment of inertia pf the section about VV-axis
u, v = Co-ordinates of point, X in the beam section with reference to
principal axes UU and VV.
The co-ordinates (u, v) are assumed positive as shown in Fig. 9.25. Due to
M.cos θ the portion of beam section above UU-axis is subjected to compressive
bending stress and that below UU-axis is subjected to tensile bending stress.
Due to M . sin θ, the portion of beam section to the left of VV-section is subjected
to compressive bending stress and that to the right of VV-axis is subjected to
tensile bending stress. Thus, it is seen that point P is subjected to maximum
tensile stress. The co-ordinates of point P and R are known. The maximum
compressive and maximum tensile bending stresses in a beam section can be
found by Eq. 9.20.
The method is suitable for beam section which has at least one axis of
symmetry. The axis of symmetry is also a principal axis. The points of extreme
distances from principal axes may be seen by visual inspection. When a beam
section has no axes of symmetry as shown in Fig. 9.21 (d), it is not possible to
locate the points which are at extreme distances by visual inspection, and this
method is not convenient to determine the maximum bending stress for such
sections.

9.19.2 By Resolving Bending Moment into two Components along

any two Perpendicular Axes, Passing through the Centroid of
Section
Let the beam section shown in Fig. 9.26 is subjected to a bending moment M.
XX and YY are two perpendicular axes and pass through the centroid of the
beam section, O. The bending stress at any point can be determined as under:
The plane of loading is inclined at θ with YY-axis as shown in Fig. 9.26. The
bending moment M may be resolved into two components
M · cos θ M . sin θ along YY and XX axes respectively.
The bending of beam due to M . cos θ takes place about XX-axis, and that due
to M . sin θ along YY and XX axes, respectively.
Then, Mxx = M . cos θ ...(i)
616 DESIGN OF STEEL STRUCTURES

and Myy = M . sin θ ...(ii)

where, Mxx is the component of bending moment, M, about XX-axis, and Myy is
the component of bending moment, about YY-axis.

M Y
θ A
P

X X
O

R S
Y

Fig. 9.26 Angle-section (maximum bending stress) on unsymmetrical bending

The variation of bending stresses at any point whose co-ordinate (x, y), is
linear
then σb = (a1 . x+ a2 . y) .. .(iii)
where, σb = Bending stress at point, (x, y)
a1 a2 = Any constants
then Mxx = ∫ σb ⋅ dA ⋅ y ...(iv)
where, dA = Any elementary area of beam section

∫ (a1x + a2 y) ydA = a1 ∫ xydA + a2 ∫ y dA

2
∴ Mxx =
= (a1.Ixy + a2.Ixx) ...(v)
where, Ixx is the moment of inertia of beam section about XX-axis and Ixx is the
product of inertia of beam section about XX-axis and YY-axis.
And Myx = ∫ σb ⋅ dA ⋅ = ∫ (a1x + a2 y ) xdA ...(vi)

= a1 ∫ x 2dA + a2 ∫ xydA
= (a1 . Ixy + a2 . Ixx) ...(vii)
where, Iyy is the moment of inertia of beam section about YY-axis.
From (v) and (vii), we have
⎛ M yy I xx – M xx I xy ⎞
a1 = ⎜ 2 ⎟ ...(viii)
⎝ I xx I yy – I xy ⎠

⎛ M xx I yy – M yy I xy ⎞
a2 = ⎜ 2 ⎟ ...(ix)
⎝ I xx I yy – I xy ⎠
 DESIGN OF ROOF TRUSSES 617

Therefore, substituting from (viii) and (ix) and values of a1 and a2 in (iii),

⎛ M xx I yy – M xx I xy ⎞ ⎛ M xx I yy – M yy I xy ⎞
σb = ⎜ 2 ⎟ ⋅x +⎜ 2 ⎟⋅y
⎝ I xx I yy – I xy ⎠ ⎝ I xx I yy – I xy ⎠

⎛ M xx I yy – M yy I xy ⎞ ⎛ M yy I xx – M xx I xy ⎞
σb = ⎜ ⎟ ⋅y+⎜ ⋅x ...(9.21)
2
⎝ I xx I yy – I xy ⎠ ⎝ I yy I xx – I x2 ⎟⎠

Equation (9.21) is specially useful for beam sections, in which the web and
flanges or legs of angle sections are parallel to XX and YY axes. The values of
bending stress at different corner points are found. The maximum bending stress
in the beam section, then be obtained.

9.19.3 By Locating the Neutral Axes of Beam Section

When a beam section is subjected to unsymmetrical bending moment M,
inclined at θ, with the principal axis VV as shown in Fig. 9.25, then the neutral
axis is not perpendicular to the plane of bending. The neutral axis of beam
section may be located by
(i) Analytical method and (ii) Graphical method.
(i) Analytical method. The bending stress at any point X, is given by Eq.
9.12.The bending stress at any point X1 on the neutral-axis is zero. Therefore,
from Eq. (9.12), we have

⎡ M ⋅ cos θ M ⋅ sin θ ⎤
⎢ I ⋅V + ⋅ u⎥ = 0
⎣ UU IVV ⎦

⎛ IUU ⎞
v = – u .⎜ I ⎟ . tan θ ...(9.22)
⎝ VV ⎠
Equation (9.22) is an equation of straight line, and represents the equation of
neutral NN-axis of the beam section. From this equation it is seen that it passes
through the centroid of the section O.
Let β be the angle of inclination of neutral NN-axis with the principal xx-
axis. Then, from Fig. 9.25
⎛v⎞
tan β = ⎜ ⎟ ...(i)
⎝u⎠
From Eq. 9.21, we have

⎛v⎞ ⎛ IUU ⎞
–⎜ ⎟ = ⎜ ⎟ . tan θ ...(i)
⎝u⎠ ⎝ IVV ⎠
Substituting the value from this, we have

⎛ IUU ⎞
tan β = ⎜ I ⎟ . tan θ ....(9.23)
⎝ VV ⎠
618 DESIGN OF STEEL STRUCTURES

From Eq. 9.23, it is seen that angle β in general, is not same as θ. If the angle
θ is zero, the bending moment M will be acting along the principal plane, and
the neutral axis will become perpendicular to the plane of bending. The angle β
will also be equal to θ in case
∴ (IUU = IVV) ∴ (tan β = tan θ), ∠ β = ∠ θ
In such cases, the ellipse of inertia of the beam section becomes a circle. Any
pair of two perpendicular axes passing through the centroid can be considered
as principal axes. Therefore, the neutral axis remains perpendicular to the plane
of bending.
UU and VV axes are the principal. Therefore, for product of inertia IUV about
UU-axis and VV-axis is zero. The moment of inertia of beam section about any
axis inclined β with the axis-UU is obtained from Eq. 9.15.
Therefore, INN = (IUU . cos2 β + IVV . sin2 β) ...(9.24)
The angle of inclination of plane of loading with the neutral axis-NN, is equal
to (90 – θ + β) or (90 + β – θ).
The angle of inclination of plane of loading with the line perpendicular to
neutral axis-NN will therefore be equal to (β – θ).
The component of bending moment, M, with the axis
= M . cos (β – θ)
If, any point is at a maximum distance, YN, from the neutral, axis-NN, then,
the maximum bending stress

⎡ M ⋅ cos (β – θ ) ⎤
σb = ⎢ ⎥ ⋅ YN ...(9.25)
⎣ I NN ⎦
The bending stress will be compressive or tensile depending upon the posi-
tion of point relative to the neutral axis and direction of loading.
(ii) Graphical method. The radii of gyration about principal axes are known
as principal radii of gyration. Let rUU and rVV be the principal radii of gyration
about UU and VV principal axes of a beam-section. The moment of inertia
about neutral axis INN may be found by determining the radius of gyration rNN
about it. The radius of gyration rNN about neutral axis-NN is found graphically
by constructing momental ellipse.The momental ellipses is also known as ellipse
of inertia.
Draw UU and VV principal axes of a beam section passing through point O
as shown in Fig. 9.27. The point O represents centroid of beam-section. Plot OA
= rUU on UU-axis and OB = rVV on VV-axis on some suitable scale. Draw two
concentric circles with point O as centre and, OA and OB as radii, respectively.
Draw MM at angle θ, with VV-axis. MM represents the plane of loading. Draw
momental ellipse with OA as length of major axis and OB is length of minor axis
as shown in Fig. 9.27. Draw a line M´ M´ tangential to the momental ellipse
such that it is parallel to MM. This line is tangent to the moment ellipse at point
S. Join points O and S, and produce it on both sides.
 DESIGN OF ROOF TRUSSES 619

V
M M'

θ
B

O M'
YVV P N
P=β V
U U
O B A
(– U )
N
M'

YUU
M*
Tan ge nt to
e llipse at S
M
V

Fig. 9.27 Momental elllipse

This locates neutral axis-NN of beam section. The radius of gration rNN about
the neutral axis, AW, is given by OS.
Then, INN = A.r2NN
where, A is the area of beam section. This can be proved as under :
Draw M´M´ perpendicular to MM and passing through point O as shown in
Fig. 9.27. M´ M´ intersects the concentric circles at points P and Q, respec-
tively. From point Q, draw a perpendicular QR to OA. From the point P draw a
perpendicular to PS to QR.
Let (–u, v) be the co-ordinates of point S with reference to UU and VV princi-
pal axes. The co-ordinates of points S are as (–u, v). Since co-ordinates of point
X in Fig. 9.25 are assumed as (u, v) in the positive direction of axes, i.e., u and v
are positive in second quadrant. Then
–u = OR = OQ . cos θ = OA . cos θ
= rUU . cos θ (OQ = OA)
U2 = rUU2. cos2 θ
y = SR = PT = OP . sin θ = OB . sin θ
(3 OP = OB)
= rVV . sin θ
v2 = (rVV2 sin2 θ )
From (i) and (ii), we have

⎛ u2 v2 ⎞
⎜ 2
+ ⎟ = (COS2 θ + sin2 θ)
⎝ rUU rVV 2 ⎠
620 DESIGN OF STEEL STRUCTURES

⎛ u2 v2 ⎞
∴ ⎜ 2 + 2 ⎟ = 1 ...(9.26)
⎝ rUU rVV ⎠
Equation 9.26 represents the equation of an ellipse, with rUU and rVV as lengths
of semi-major and semi-minor axes, respectively. Equation 9.26 also represents
the locus of S. The ellipse represented by Eq. 9 26 is same as momental ellipse
which is drawn to find rNN. The lengths of semi-major and semi-major axes are
the same.
Let the angle SOA be β´
⎛ –v ⎞
tan β´ = ⎜ ⎟
⎝ u ⎠
From Eq. 9.23 (i), we have
⎛ –v ⎞
tan β´ = ⎜ ⎟
⎝ u ⎠
Therefore, β´ = β
Therefore, NN represents the neutral axis of the beam section. The line MM
may be considered as direction of one of the diameter of momentral ellipse. The
line NN passes through middle point of MM and point, S, the point on tangent
drawn parallel to MM. Line NN shows the direction of conjugate diameter, there-
fore, MM and NN form conjugate diameter of the ellipse.
The slope of MM with UU-axis is tan (90° + θ) or cot θ and that of NN with
UU-axis is tan β. From the properties of conjugate diameters of an ellipse, we
have
⎛ rUU 2 ⎞
– cot θ . tan β = ⎜⎜ ⎟⎟ ...(i)
⎝ rVV ⎠

⎛ A ⋅ rUU 2 ⎞ ⎛ I 2 ⎞
tan β = ⎜⎜ + ⎟⎟ ⋅ tan θ = ⎜⎜ + UU tan è ⎟⎟ ...(ii)
⎝ A ⋅ rVV ⎠ ⎝ IVV ⎠

⎛ –v ⎞
∴ tan β = ⎜ ⎟
⎝ u ⎠

⎛I 2⎞
Therefore, v = –u ⎜⎜ UU ⎟⎟ ⋅ tan è ...(iii)
⎝ IVV ⎠
This relation obtained is same as Eq. 9.22 and it is equation of the neutral
axis. Further from Fig. 9.27,
(OS)2 = (OR)2 + (RS)2 = (OQ. cos β)2 + (OP . sin β)2 ... (i)
(3 OQ = OA, OP = OB)
(OS)2 = (OA . cos β)2 + (OB . sin β)2
= (rUU2. cos2 β + rVV2 . sin2 β) ...(ii)
 DESIGN OF ROOF TRUSSES 621

Multiplying both sides by A, the area of cross-section of beam section, we

have
A.(OS)2 = (A.ruu2 COS2 β + A.rvv2 sin2 β)
= (IUU . cos2 β + IVV . sin2 β) ..(iii)
IVV = A.rUU2, IVV = A.rVV2 and IVV = 0
From Eq. (9.16), therefore
A . (OS)2 = INN
⎛I ⎞
(OS)2 = ⎜⎝ NN ⎟⎠ = r2NN ...(iv)
A
OS = rNN ...(v)
After determining the value of radius of gyration graphically the moment of
inertia INN about neutral axis is determined. The maximum bending stress at
the extreme distance from the neutral axis may be found from Eq. 9.25.

9.20 THE Z-POLYGON

The strength of a beam section in bending depends upon the modulus of section
Z. Along with the geometrical properties of a beam section, the value of modulus
of section Z also depends upon plane of bending (or plane of loading). The value
of modulus of section Z, varies as the position of the plane of bending changes.
The maximum and minimum values of modulus of section, Z, may be found by
determining the corresponding positions of plane of bending.
Consider any point A, whose co-ordinates are (uA, vA) with reference to
principal axes UU and VV. The beam section is subjected to bending moment
M. The plane of bending is inclined at θ with VV-axis, as shown in Fig. 9.28.
From Eq. 9.20 the bending stress at point is given by
V
M P2
P3
P4
Z-L in e
θ1 Iuu
VA
θ
VA
U U
o
P1
Iu u
UA

V
Z-L in e
Fig 9.28
622 DESIGN OF STEEL STRUCTURES

⎛ M cos θ ⎞ ⎛ M sin θ ⎞ ⎡ v ⋅ cos θ u A ⋅ sin θ ⎤

σb = ⎜ ⋅ vA + ⎜ ⋅ uA = M ⎢ A +
⎝ IUU ⎟⎠ ⎝ IVV ⎠ ⎟
⎣ IUU IVV ⎥⎦

⎡ IVV ⋅ (vA ⋅ cos θ) + IUU ⋅ (uA ⋅ sin θ) ⎤

σb = ⎢ ⎥
⎣ IUU ⋅ IVV ⎦

⎡ M ⎤
σb = ⎢ ⎥
IUU ⋅ IVV
⎢ ⎥
⎢⎣ IVV ⋅ (vA ⋅ cos θ) + IUU ⋅ u A ⋅ sin θ ⎥⎦

⎛M ⎞
∴ σb = ⎜ ⎟ ...(9.27)
⎝Z ⎠
where, Z is the modulus of section of the beam section, and

⎡ UU I VV ⋅I ⎤
Z = ⎢ ...(9.28)
I
⎣ VV ⋅ (V A ⋅ cos θ ) + IUU ⋅ U A ⋅ sin θ ⎥⎦

In Eq. 9.28, IUU and IVV are principal moment of inertias of the beam section
about UU and VV axes, respectively. For a given beam section, the value of IUU
and IVV are constant. For a given point, A on the beam section uA and vA are also
constant. Thus, it is seen that the value of Z depends upon θ only.

⎛ IUU ⎞
When (θ = 0°), Z = ⎜ I ⎟ ...(i)
⎝ VV ⎠

⎛ IUU ⎞
and when (θ = 90°), Z = ⎜ U ⎟ ... (ii)
⎝ A ⎠
Expressions (i) and (ii) are usual form of modulus of section. For any given
point of a given beam section, with θ variable, Eq. 9.28 represents polar form of
the equation of a straight line. Equation 9.28 may be expressed in the rectangu-
lar co-ordinates as under:
1 ⎡ IVV ⋅ (V A ⋅ cos θ ) + IUU ⋅ U A ⋅ sin θ ⎤
= ⎢ ⎥ ...(iii)
Z ⎣ IUU ⋅ IVV ⎦

1 ⎡ V A ⋅ cos θ U A ⋅ sin θ ⎤
or = ⎢ + ⎥ ...(iv)
Z ⎣ IUU IVV ⎦

⎡ V A ⋅ (Z cos θ ) U A ⋅ ( Z sin θ ) ⎤
or ⎢ + ⎥= 1 ...(v)
⎣ IUU IVV ⎦

⎡⎛ Z cos θ ⎞ ⎛ Z cos θ ⎞ ⎤
or ⎢⎜ I ⎟+⎜ I ⎟⎥ = 1 ...(vi)
⎢⎜⎜ UU ⎟⎟ ⎜⎜ VV ⎟⎟ ⎥
⎢⎣⎝ V A ⎠ ⎝ U A ⎠ ⎥⎦
 DESIGN OF ROOF TRUSSES 623

Put Z . cos θ = v, and Z . sin θ = u

⎡⎛ u ⎞ ⎛ u ⎞⎤
Then ⎢⎜ I ⎟ + ⎜ I ⎟⎥ = 1 ...(9.29)
⎢⎜⎜ VV ⎟⎟ ⎜⎜ UU ⎟⎟ ⎥
⎣⎢⎝ uA ⎠ ⎝ V A ⎠ ⎦⎥
This represents equation of a straight line, which is known as Z-line. The length
⎛ IVV ⎞
of interception on UU-axis is ⎜ u ⎟ and the length of interception on VV-axis as
⎝ A ⎠

⎛ IUU ⎞
⎜ u ⎟.
⎝ A ⎠
Equation 9.29 can also be expressed as below :

⎛ v ⎞ ⎛ u ⎞
⎜I ⎟ = –⎜ I ⎟+1 ...(i)
UU
⎜⎜ ⎟⎟ ⎜⎜ VV ⎟⎟
v
⎝ A ⎠ ⎝ uA ⎠

⎛u ⎞ ⎛ I ⎞ ⎛I ⎞
v = –u ⎜ A ⎟ ⋅ ⎜ UU ⎟ + ⎜ UU ⎟ ...(ii)
⎝ vA ⎠ ⎝ IVV ⎠ ⎝ vA ⎠
When the plane of bending passes through the point A,

⎛ uA ⎞
⎜ v ⎟ = tan θ
⎝ A⎠

⎛I ⎞ ⎛I ⎞
Therefore, v = –u ⎜ UU ⎟ .tan θ+ ⎜ UU ⎟ ...(9.30)
⎝ IVV ⎠ ⎝ vA ⎠
By comparing Eq. 9.22 and Eq. 9.30 for the plane of bending passing through
given point, it is seen that Z-line is parallel to the neutral axis.
⎛ IUU ⎞
The Z-line represented by Eq. 9.29 is plotted by taking OP = ⎜ u ⎟ and OP2=
⎝ A ⎠

⎛ IUU ⎞
⎜ u ⎟ along UU-axis and VV-axis respectively. The points P1 and P2 are joined.
⎝ A ⎠
The line P1P2 represents Z-line. The Z-line intersects the plane of bending at
P3. OP3 represents the modulus of section, Z. the bending stress at point A is
given by
⎛ M ⎞
σb = ⎜ ⎟
⎝ OP3 ⎠
Let OP4 represent perpendicular from point O, the centroid of beam section
to the Z-line. The minimum distance from point O to Z-line is given by
perpendicular distance OP4. Then OP4 represents the minimum value of Z. Let
624 DESIGN OF STEEL STRUCTURES

θ1 be the angle of OP4 measured from OV-axis. The bending stress at point A is
maximum when the plane of loading is inclined at 1 to principal axis W.
The maximum bending stress at the point A
⎛ M ⎞
σb(max) = ⎜ where θ = θ1.
⎝ OP4 ⎟⎠
The maximum distance of any point on Z-line from the centroid O, of the beam
section, depends, upon the lengths of intercepts on UU-axis and VV-axis. If OP2
is greater than OP1, then OP2 is the maximum distance. The maximum value of
modulus of section Z is given by OP2. The bending stress at point O is minimum.
The plane of bending coincides with the principal axis VV.
Minimum bending stress at point A
⎛ M ⎞
σb(min) = ⎜ ⎟
⎝ OP2 ⎠
where θ= θ

OP2 = ⎛ IVV ⎞
⎜⎝ v ⎟⎠
A
It is necessary to draw Z-lines for critical points, (the points which are at
extreme distance from C.G. of beam section) in order to determine maximum
bending stress. The Z-lines drawn critical points form a polygon. This polygon
is called Z-polygon. The values of Z for these critical points for a given plane of
bending may be compared. The maximum bending stress occurs at a critical
point having least value of Z. The Z-polygon for a rectangular beam section is
drawn as below:
The principal axes UU and VV are drawn passing through point O, the centroid
of rectangular beam. The rectangular beam section ABCD is drawn as shown in
Fig. 9.29. Let the width of section be b and depth of section be h. The rectangular
beam section ABCD is drawn as shown in Fig. 9.29. The rectangular section is
symmetrical section. Therefore, co-ordinate axes XX and YY coincide with the
principal axes of the section.
Z-line for point A is drawn first. The co-ordinates of point A, lying in first
1 1
quadrant are uA = 2 b.vA = h.
2
⎛1 ⎞ ⎛1 ⎞
IUU = Ixx= ⎜ ⎟ bh3 and IVV =Iyy= ⎜ ⎟ hb3
⎝ 12 ⎠ ⎝ 12 ⎠
The length of intercepts of Z-line along UU-axis is
⎡⎛ 1 ⎞ 3 ⎤
⎛ IVV ⎞ ⎢ ⎜⎝ 12 ⎟⎠ hb ⎥ ⎛ 1⎞
OP1 = ⎜ ⎟ and OP =
1 ⎢ ⎥ = ⎜⎝ ⎟⎠ hb2
⎝ uA ⎠ ⎢ 1b ⎥ 6
⎣⎢ 2 ⎦⎥
= Zyy = ZVV (3 Zyy = ZVV)
 DESIGN OF ROOF TRUSSES 625

The length of intercepts of Z-line along VV axis is

⎡⎛ 1 ⎞ 3 ⎤
⎢ ⎜⎝ 12 ⎟⎠ bh ⎥ ⎛ 1 ⎞
OP2 = ⎢ ⎥ = ⎜ ⎟ bh2 = Zxx= ZUU
⎢ 1h ⎥ ⎝6⎠
⎢⎣ 2 ⎥⎦
(3 Zxx = ZUU)
Equation 9.29 becomes

⎡ u v ⎤
⎢Z + Z ⎥ = 1
⎣ VV UU ⎦

⎛1⎞ ⎛1 ⎞
On UU-axis, plot OP1= ⎜ ⎟ hb2 and on VV-axis plot OP2= ⎜ ⎟ bh2 to a suit-
⎝6⎠ ⎝6⎠
able scale. Points P1 and P2 are joined. P1P2 represents Z-line for the point A.
Similarly, Z-lines P2P3, P3P4 and P4P1 are drawn for points B, C and D respec-
tively. P1P2P3P4 represents Z-polygon. The Z-polygon for a rectangular beam
section is a rhombus.
The position of plane of bending for maximum and minimum strength of a
section can be found by inspection of Z-polygon. For example, for minimum
strength of the section, the plane of bending is along VV-axis.
V

P2
1 bh2
6

b
E A 2 B F

h
2 θ1
P1 P3
U u
1 b h2 O h
6

H D C G
b

P4
V
Z- P o lygo n

Fig. 9.29
626 DESIGN OF STEEL STRUCTURES

The value of Z is maximum along this line. Similarly, for minimum strength
of the section, perpendiculars OE, OF, OG and OH are drawn from point, O, to
Z-lines as shown in Fig. 9.29. the strength of beam section is minimum along
EG and FH planes. The values of Z for plane bending along these planes is
minimum. The inclination of plane EG is θ1. The value of θ1 is obtained by

⎛1 ⎞
⋅ bh2 ⎟ ⎛h⎞
⎛ OP2 ⎞ ⎛ ZUU ⎞ ⎜ 6 h
tan θ1 = ⎜ ⎟ =⎜ ⎟ =⎜ 1 ⎟= , ∴θ1 = tan–1 ⎜⎝ b ⎟⎠
Z
⎝ OP1 ⎠ ⎝ VV ⎠ ⎜⎜ ⋅ hb ⎟⎟ b
2
⎝6 ⎠

9.21 DEFLECTION OF BEAMS UNDER UNSYMMETRICAL

BENDING
The plane of bending or the plane of loading is inclined at an angle θ, with
principal axis-VV as shown in Fig. 9.30. The bending moment M1, acting along
the plane of bending may be resolved into two components along W-axis and
UU-axis, respectively. The deflection of beam subjected to unsymmetrical bend-
ing may be found as under :
From unit load method, the deflection of beam subjected to moment M1, is
given by
V
M N
M cos θ

β = β + 9 0° N
θ
M sin θ β' β
U U

6
N δv

V δv N'
Fig. 9.30 Deflection of beams (unsymmetrical bending)

L
⎛ M1m1 ⎞
δ = ∫ ⎜⎝ ⎟
EI ⎠ dx
o

where, m1 = Moment due to unit load acting in the direction of desired

deflection. Unit load is applied at the point where deflection is
to be found
dx = Elementary length of beam, measured along span of beam
L = Span of beam.
The deflection in the direction of VV-axis is given by
 DESIGN OF ROOF TRUSSES 627

L
⎛ M cos θ ⎞
δV = ∫ ⎜⎝EIUU ⎟⎠
⋅ mv ⋅ dx ...(i)
o
The deflection in the direction of UU-axis is given by
L
⎛ M sin θ ⎞
δV = ∫ ⎜⎝EIVV ⎟⎠
⋅ mu ⋅ dx ...(ii)
o
The resultant deflection is found by adding these deflections vectorially. There-
for, resultant deflection
2 2 1/ 2
δ = ⎡⎣(δU ) + (δV ) ⎤⎦ ...(iii)

2 2 1/ 2
L
⎛ M ⎞ ⎡⎛ cos θ ⎞ ⎛ sin θ ⎞ ⎤
δ = ∫ ⎜⎝ Em ⎟⎠ ⎢⎢⎜⎝ IUU ⎟⎠ + ⎜⎝ IVV ⎟⎠ ⎥⎥ ...(iv)
o ⎣ ⎦
Since mV = mU =m
L 2 2 1/ 2
⎛ M ⎞ ⎡I cos θ + I sin θ ⎤
or δ = ∫ ⎜⎝ Em ⎟⎠ ⎢⎢ UU I 2 I UU2 ⎥ dx
o ⎣ UU VV ⎦⎥
...(9.31)
Let β´ be the angle which the resultant deflection makes with the UU-axis.
L
⎛ M m sin θ ⎞
⎛ –δ ⎞
∫ ⎜⎝ EIVV ⎟⎠
dx
tan β´ = ⎜ U ⎟ = o
L
⎝ δV ⎠ ⎛ M m cos θ ⎞
∫ ⎜⎝ EIUU ⎟⎠
dx
o

⎛I ⎞
or tan β = – ⎜ UU ⎟ tan θ ...(9.32)
⎝ IVV ⎠
From Eq. 9.23, we have
⎛I ⎞
tan β = + ⎜ UU ⎟ tan θ
⎝I ⎠ VV

By comparing Eq. 9.23, and Eq. 9.23, it is seen that

tan β´ = tan (90° + β) and β´ = 90° + β
Therefore, it is seen that the resultant deflection occurs in the direction N´N´,
which is perpendicular to neutral axis-NN for any given plane of loading.
If the beam is subjected to uniformly distributed load w, then

5 ⎛ w ⋅ sin θ ⋅ L4 ⎞
δUU = ...(i)
384 ⎜⎝ ElVV ⎟⎠

5 ⎛ w ⋅ cos θ ⋅ L4 ⎞
δVV = ...(ii)
384 ⎜⎝ E ⋅ lUU ⎟⎠
628 DESIGN OF STEEL STRUCTURES

The resultant defletion of the beam

2 2 1/ 2
δ = ⎡⎣(δUU ) + (δVV ) ⎤⎦ ...(iii)
1/ 2
5 wL4 ⎡ sin 2 θ cos2 θ ⎤
δ = ⎢ 2 + 2 ⎥
384 E ⎣ I VV I UU ⎦

2 1/ 2
5 ⎛ wL4 cos θ ⎞ ⎡ ⎛I ⎞ ⎤
= ⎢1 + ⎜ UU tan θ⎟ ⎥
384 ⎜⎝ E ⋅ IUU ⎟⎠ ⎢⎣ ⎝ IVV ⎠ ⎥⎦
...(iv)
⎛ I ⎞
From Eq. 9.23, tan β = ⎜ + UU tan θ⎟
⎝ IVV ⎠

5 ⎛ wL4 cos θ ⎞ ⎡ 2
1/ 2
Therefore, δ = ⎣1 + tan β⎤⎦ ...(v)
384 ⎜⎝ E ⋅ IUU ⎟⎠

5 ⎛ wL4 ⎞ 5 ⎛ wL4 ⎞ cos θ

or = ⋅⎜ ⋅ cos θ ⋅ sec β = ⋅⎜ ⎟.
384 ⎝ EIUU ⎟⎠ 384 ⎜⎝ EI uu ⎟⎠ cos β

5 ⎛ wL4 ⎞ ⎛ cos θ ⎞ ⎛ cos (β – θ ) ⎞

or = ⋅⎜ ⎟ .⎜ ⎟ .⎜ ⎟
384 ⎜⎝ EI uu ⎟⎠ ⎝ cos β ⎠ ⎝ cos (β – θ ) ⎠

5 ⎛ wL4 ⎞ ⎛ cos θ.cos (β – θ ) ⎞

or = ⋅ ⎜⎜ ⎟⎟ . ⎜ ⎟
384 ⎝ EI uu ⎠ ⎝ cos β [cos β cos θ + sin β sin θ] ⎠

5 ⎛ wL4 ⎞ ⎛ cos (β – θ ) cos θ ⎞

= ⋅⎜ . ⎜ cos β ⋅ cos β cos θ 1 + tan β tan θ ⎟ ...(vi)
⎟
384 ⎝ EIUU ⎠ ⎝ [ ] ⎠

⎛I ⎞
From Eq. 9.23, tan β = ⎜ UU ⎟ ⋅ tan θ
⎝ IVV ⎠
Value of tan θ is substituted in (vi)
5 ⎛ wL4 ⎞ ⎛ cos (β – θ) ⎞
∴ δ = ⋅
384 ⎜⎝ EIUU ⎟⎠ ⎜ 2 ⎡ IUU 2 ⎤
⎟
⎜ cos β ⋅ ⎢1 + tan β ⎥ ⎟
⎝ ⎣ IVV ⎦⎠

⎛ cos (β – θ) ⎞
5 ⎛ wL4 ⎞ ⎜ ⎡ IUU cos β + IVV sin β ⎤ ⎟
2 2
or δ = ⋅⎜ ⎜ 2
EI ⎟
384 ⎝ UU ⎠ ⎜ cos β ⋅ ⎢ ⎥ ⎟⎟
⎝ ⎣ IUU ⋅ cos2 β ⎦⎠

⎛ wL4 cos (β – θ ) ⎞
δ = ⎜ 2 2 ⎟ ...(viii)
⎝ E ⎡⎣ IUU ⋅ cos β + IVV sin β ⎤⎦ ⎠
 DESIGN OF ROOF TRUSSES 629

From Eq. 9.24, INN = ( IUU ⋅ cos2 β + IVV sin2 β)

5 ⎛ w ⋅ cos (β – θ ) L4 ⎞
∴ δ = ⎜ ⎟⎟ ...(9.33)
384 ⎜⎝ E ⋅ I NN ⎠
In Eq. 9.33, w cos (β – θ) is the component of resultant uniformly distributed
load along the direction N´N´ perpendicular to the neutral axis NN, for plane
bending (or plane loading) inclined θ with VV-axis.

9.22 DESIGN OF PURLINS SUBJECTED TO UNSYMMETRICAL

BENDING
The rectangular section, I-section, channel section and angle section are used
for purlins of roof trusses. The rectangular section and I-section have two axes
of symmetry. The channel section has at least one axis of symmetry. The axes of
symmetry locate the principal axes in case of rectangular section, I-section and
channel section.
All the purlins are designed in accordance with the requirements for uncased
beams. However, the limitations of bending stress based on lateral instability of
the compression flange and limitation of deflection are waived for the design of
purlins. The calculated deflections should not exceed those permitted for the
type of roof cladding used. In calculating the bending moment, the advantage
may be taken of the continuity of the purlin over the supports. The bending
stresses about the two axes should be determined separately and checked by
the combined interaction equations.
The design of purlins, in cases where principal axes are located directly from
axes of symmetry of a section, is done as follows :
The span of the purlin is centre to centre distance between adjacent trusses,
(i.e., it is equal to the spacing of trusses).
The spacing of purlins should should not exceed 1.4 m so that ACC sheets do
not fail.
Step 1. Roof sheets are supported by the purlins. These root sheets transfer
the loads to the purlins. The various loads acting over the purlins are weight of
roof-sheets and fixtures, self-weight of the purlins, live load from the roof-sheets
and the wind load. Except the wind loads, other loads are the gravity loads.
These gravity loads act vertically. The vertical loads and wind load acting over
the purlins are shown in Fig. 9.31. The vertical load acts downward and the
wind load acts normal to the sheets either inward or outward depending upon
the wind pressure or suction. The purlins are connected to the rafters of the roof
trusses. The major principal axis of the purlin-UU remains parallel to the
principal rafter. Therefore, the vertical load remains inclined with the major
principal axes. As a result of which unsymmetrical bending (that is, biaxial
bending) of the purlins occur.
The appropriate combination of loads on purlins are resolved into two
components perpendicular to and parallel to the roof slope.
630 DESIGN OF STEEL STRUCTURES

Step 2. The principal axes UU and VV of the section coincide with the co-
ordinate axes XX and YY respectively. The maximum bending moments are
found about these axes. Let the maximum bending moment about UU-axis (i.e.,
XX-axis) be MUV and that about VV-axis (i.e.,YY-axis) be MVV.
Step 3. The maximum bending stress at any point on a purlin section is
given by
⎛M ⎞ ⎛M ⎞
σb = ⎜ UU ⎟ ⋅ u + ⎜ VV ⎟ ⋅ u ...(i)
⎝ IVV ⎠ ⎝ IVV ⎠

Ve rtica l loa d
V
U
W ind loa d

P rincip al
ra fte r
U
V

B o tto m ch o rd

Fig. 9.31 Channel section purlin

If M is the maximum bending moment caused due to resultant loading, and if

a plane of bending (i.e., plane of loading) is inclined at θ then,
MUV = M cos θ and MVV = M sin θ
Expression (i) is in the same form as Eq. 9.20. In above expression
IUU = IXX and IVV =IYY
Expression (i) can be written as

⎡M ⎛ ⎞ ⎛M ⎤
⎞ M M
σb = ⎢ I UU + I VV ⎥ = ⎜ UU ⎟ + ⎜ VV ⎟
⎢ UU VV
⎥ ⎝ ZUU ⎠ ⎝ ZVV ⎠
⎣ v u ⎦

MUU ⎛ M VV IUU ⎞
or σb = . ⎜1 + ...(9.34)
ZUU ⎝ MUU ZVV ⎟⎠

MUU ⎛ M Z ⎞
or ZUU = ⎜ 1 + VV UU ⎟ ...(9.35)
σb ⎝ MUU ZVV ⎠
In Eq. 9.35 it is to note that ZUU and ZVV, the moduli of section about UU-axis
and VV-axis, respectively, are the same as Zxxand Zyy, the moduli of section
about XX-axis and YY-axis, respectively. In Eq. 9.35 suitable value of ratio
⎛ Zuu ⎞
⎜⎝ Z ⎟⎠ is assumed in order to determine, modulus of section ZUU. For rectangular
vv
 DESIGN OF ROOF TRUSSES 631

⎛ ZUU ⎞
section, the value of ⎜ may be assumed as 5. For I-sections, this ratio may
⎝ ZVV ⎟⎠
be assumed 5 to 7 for lightly loaded purlins, and 7 to 10 for heavily loaded
⎛Z ⎞
purlins. For channel section, the value of ⎜ UU ⎟ may be assumed as 8. From
⎝ ZVV ⎠
Eq. 9.35 the required modulus of section ZUU is determined by assuming σb,
equal to permissible bending stress. A trail section may be selected from ISI
Handbook No. 1.
Step 4. This trial section for purlins is checked for maximum bending stress.
The maximum bending stress is determined from Eq. 9.34. The maximum
bending stress compressive or tensile in beam section should not exceed the
maximum permissible bending stress in compression or in tension.
The weight of purlins as compared to the total weight of the steel structure
may vary from 10 percent to 25 percent. The economic impact (i.e., cost),
durability of roof, increase or decrease of the fixtures and the vibrations of roof
sheets are the various factors, which should be kept in view while designing the
purlins. The spacing of roof trusses may vary from 3 m to 12 m. The economical
spacing of roof trusses depends on the various factors (e.g., type of soil, bearing
capacity of soil, wind zone, inclination of the roof and the functional aspects).
The various sections (e.g., channels, angles, tubes, Z-shape or cold formed,
Warren or N-type truss, triangular or open section) may be used for the purlins.
The channels or angles should be so placed that the accumulation of dust,
moisture and water is minimum or avoided. The purlin sections should be
connected with the principal rafters of the roof trusses in their correct
orientations. For example, the flanges in case of channels may be kept sloping
downward with the web normal to the rafter and upward. So also, in case of the
angle sections, one leg is kept and connected normal to the rafter while the
other leg is kept sloping downward. The durability, that is, the life of purlin
reduces due to incorrect position (i.e., wrong orientation) of the purlin sections.

9.22.1 Design of Angle Iron Purlins

The angle-section are most commonly used for purlins. The angle sections do
not have any symmetric axis. Therefore it becomes essential to locate the
principal axes of the angle-section. The principal axes may be located, if the
geometrical properties of angle section are known. Depending upon the resultant
load, the trial section is selected for purlins. Then, general procedure for design
of purlins mentioned above may be followed.
If the slope of roof truss is not greater than 30° and the steel is conforming to
Grades Fe 410-O or Fe 410-W, then, the angle purlins may be designed as an
alternate to the general design procedure, as recommend by IS: 800–1984.
However, the following requirements which are based on a maximum imposed
load of 0.75 kN/m2 are fulfilled.
632 DESIGN OF STEEL STRUCTURES

Step 1. The depth of angle purlin in the plane approximate to the incidence
⎛ 1 ⎞
of the maximum load or maximum component of the load is not less than ⎜ ⎟ th
⎝ 45 ⎠
of the length of purlin.

⎛ 1 ⎞
Step 2. The width of angle purlin is not less than ⎜ ⎟ th of the length of
⎝ 60 ⎠
purlin.
Knowing the depth of purlin, and width of purlin, a trial angle selection is
selected from ISI Hand-book No. 1.
Step 3. The maximum bending moment in a purlin is taken as

⎛ WL ⎞
Mmax = ⎜ ⎟
⎝ 10 ⎠
where, W is total uniformly distributed load on the purlin including wind load
and L is the distance centre to centre of the rafters or other supports of the
purlins.
The loads are assumed as acting normal to the roof surface. The bending of
the purlin about the minor axis is neglected. In calculating the bending moment,
the advantage of continuity of the purlins over roof trusses has been taken.
Step 4. The maximum bending stress calculated for Mmax should not exceed
the maximum permissible bending stress in compression or in tension. The design
or purlins takes into account the wind load. Therefore, the permissible stress in
bending in compression or tensile are 1.33 times the usual permissible values,
(viz, dead load plus live load, dead load plus half the live load plus the wind load
or the dead load and wind load) shall be considered. It is to note that the dead
load and live load cause biaxial bending, whereas the wind load produces bending
about major principal axis. The bending moments due to dead loads, due to live
load and due to wind load are calculated.
These purlins may be treated as continuous beams supported over principal
rafters of the roof trusses. The moments at the supports and mid-spans of the
end and intermediate spans may be calculated as recommended in the code.
For end span (one end simply supported and other end continuous)
(a) Bending moent due to dead loads

⎛ 2 ⎞
mid-span M = + ⎜ wd ⋅ L ⎟ , and .. .(i)
⎝ 16 ⎠

⎛ 2 ⎞
support M = ⎜ wd ⋅ L ⎟ ...(ii)
⎝ 9 ⎠
 DESIGN OF ROOF TRUSSES 633

(b) Bending moment due to live loads

⎛ 2 ⎞
mid-span M = + ⎜ wL ⋅ L ⎟ and ....(iii)
⎝ 9 ⎠

⎛ 2 ⎞
support M = ⎜ wL ⋅ L ⎟ ...(iv)
⎝ 9 ⎠
It is to note that for the intermediate spans, the bending moments are less
⎛ 2⎞
than these values. It is recommended to use the value ⎜ wL L ⎟ .
⎝ 10 ⎠
Following guidelines may be kept in view while selecting a particular section
for the purlins. These guide lines will be advantageous in selecting the purlin
sections.
1. The angle or channel sections may be used suitably and economically
upto 5 m span without sag bars except in heavy snow zones.
2. The tubular sections may be used upto 6 m span. A minimum of 3 mm
thickness of the tube is essential for the structural welding. In case the
thickness of tube is small, it will be rusted soon. As far as possible,
tubes with thickness more than 3.5 mm may be used.
3. In case of snow zones, sag bars are used for purlins for 4 m and more
spans for angle sections and channel sections. The channel sections
even sag bars for purlins of spans beyond 8 m are not suitable.
4. For purlins of spans 6 m and more, the triangulation or truss purlins
may be used economically. However, their fabrication is slightly diffi-
cult.
The durability, maintenance aspect fabrications and the erection convenience
should also be considered.
Example 9.1 A roof truss-shed is to be built in Jodhpur city area for an
industrial use. Determine the basic wind pressure. The size of shed 18 m × 30 m.
Solution
Step 1: Basic wind speed
From IS : 875 (Part 3) – 1987 (from wind zone map of the country or Table
9.2, the basic wind speed in Jodhpur,
Vd = 47 m/sec ...(i)
Step 2: Risk coefficient (factor – k1)
The design life for roof shed for industrial use may be assumed 50 years.
From IS : 875 (Part 3) – 1987 or Table 9.3, the risk coefficient (factor – k 1), for Vb
= 47 m/sec,
k1 = 100 ...(ii)
Step 3 : Terrain, height and structure size factor (factor-k2)
The terrain is Jodhpur city industrial area. Therefore, the area is of type
category 3. The size of shed shall be 18 m × 30 m. Therefore, the maximum
horizontal dimension is between 20 m and 50 m. Therefore, the roof shed
634 DESIGN OF STEEL STRUCTURES

structure is of type class B. In category 3, the terrain with numerous closely

spaced obstructions having the size of building structures upto 10 m in height
with or without a few isolated tall structures, therefore, from IS : 875 (Part 3)
1987 or Table 9.4,
k2 = 0.98 ...(iii)
Step 4 : Topography factor, k3
Near the roof-shed, the ground is assumed to be plain. Therefore, the topog-
raphy factor from Eq. 9.6,
k3 = ( 1 + C . S)

⎛Z ⎞
where C = ⎜ ⎟ =0
⎝L⎠
k3 = 1.00 ...(iv)
Step 5 : Design wind sped (Eq. 9 3)
Vz = k 1 . k2 . k3 .Vb
Vz = ( 1 × 0.98 × 1 × 47)
= 46.06 m/sec.
Step 6 : Basic wind pressure (from Eq. 9.12)
pz = 0.0006 × Vz2 kN/m2
= 0.0006 × (46.06)2
= 1.2729 kN/m2
Example 9.2 An industrial roof shed of size 20 m × 30 m is proposed to be
constructed at Manglore near a hillock of 160 m and the slope is 1 in 2.8. The
roof shed is to be built at a height of 120 m from the base of the hill. Determine
the design wind pressure on the slope. The height of roof shed shall be 12 m.
Solution
Step 1: Basic wind speed
From IS : 875 (Part 3) – 1987 or Table 9.2, tne basic wind speed Vb is 39 m/sec.
Step 2: Risk coefficient (factor-k1)
For 50 years of design life of the structure and all general buildings for Vb
= 39 m/sec, risk coefficient,
k = 100 ...(i)
Step 3: Terrain factor, k2
Near the hillock side, the terrain shall be open with well scattered obstructions
having heights generally batween 15 m to 10 m. Therefore, the terrain is of type
category 2. The size of roof shed is 20 m × 30 m. Therefore, it is of type class B.
(since maximum horizontal dimension is between 20 m and 50 m). The height of
roof-shed shall be 12 m. Therefore, the terrain, height and structure size factor,
from IS : 875 (Part 3) – 1987 or Table 9.2

⎡ (1 ⋅ 02 – 0 ⋅ 98 ) × 2 ⎤
k2 = ⎢0 ⋅ 98 + ⎥⎦ = 0.996
⎣ 5
 DESIGN OF ROOF TRUSSES 635

Step 4 : Topography factor k3

The effective height of the hill, z is 160 m.
The slope is 1 in 2.8. Therefore,
⎛ 1 ⎞
θ = tan–1 ⎜ ⎟ = 19.654°
⎝ 2⋅8 ⎠
Length of upward slope
L = (160 × 2.8) = 448 m
For θ > 17°, effective horizontal length of the hill
⎛ Z ⎞ ⎛ 160 ⎞
Le = ⎜ ⎟ =⎜ ⎟ = 533.33 m
⎝ 0 ⋅3⎠ ⎝ 0 ⋅3⎠
For (θ = 19.654°) > 17°, from IS : 875 (Part 3) – 1987, Appendix C.
C = 0.36
k3 = (1 + C . S)
Let x be the horizontal distance of the building from the crest. It is measured
positive towards leeward side and negative towards the windward side.
x = – (L – 120)
= – (448 – 120) = – 328 m
x ⎛ –328 ⎞
= ⎜ ⎟ = –0.615
Le ⎝ 533 ⋅ 33 ⎠

H ⎛ 12 ⎞
= ⎜ ⎟ = 0.0225
Le ⎝ 533 ⋅ 33 ⎠

⎛ x ⎞ ⎛H⎞
From Fig. 9.6 (b), for ⎜ ⎟ = – 0.615 and ⎜ ⎟ = 0.0225
⎝ Le ⎠ ⎝ Le ⎠
S = 0.2
From Eq. 9.6, for (b) = 19.654°) > 17; c = 0.36
k3 = (1 + C.S)
k3 = (l + 0.36 × 0.2) = 1.072
Step 5: Design wind speed (from Eq. 9.3)
Vz = k1.k2.k3.Vb
Vz = (1 × 0.996 × 1.072) × 39
= 41.641 m/sec
Step 6: Design wind pressure (from Eq. 9.12)
pz = 0.0006 V22 kN/m2
pz = (0.0006 × 41.6412) = 1.04 m/sec
Example 9.3 A hospital building of size 50 m × 100 m and a height 10 m is
proposed to be built at Poona on a hill top. Determine the design wind pressure
on the building. The height of hill is 320 m with a slope of 1 in 4. The hospital is
proposed at a distance 120 m from the crest on the downward slope.
636 DESIGN OF STEEL STRUCTURES

Solution
Step 1: Basic wind speed: From IS : 875 (Part 3) – 1987 or Table 9.2, the
basic wind speed at Poona is 39 m/sec.
Step 2: Risk coefficient (factor –k1)
For important buildings and structures, such as hospitals for which the de-
sign life be assumed 100 years, the risk coefficient from IS : 875 (Part 3) –1987
or Table 9.3, for Vb = 39 m/sec
k1 = 1.06 ...(i)
Step 3: Terraine, height and structure size factor, k2
The height of hospital shall be 10 m. The open terraine with well scattered
obstruction having heights generally between 1.5 m to 10 m belongs to category
2. The size of hospital dimension is more than 50 m. As such, it belongs to class
(c). From IS : 875 (Part 3) – 1987 for Table 9.4, for 10 m height
k2 = 0.93 ...(ii)
Step 4: Topography factor k3
The effective height of the hill, Z is 320 m. The slope is θ = 1 in 4
⎛1 ⎞
θ = tan–1 ⎜ ⎟ = 14.036°
⎝4⎠
Length of upward slope
L = (320 × 4)= 1280 m
Effective horizontal length of the hill, for (θ = 14.036°) < 17°,
Le = L= 1280 m
Value of C for θ 17°
⎛Z ⎞ ⎛ 1 ⋅ 2 × 320 ⎞
C = 1.2 × ⎜ ⎟ = ⎜⎝ ⎟ = 0.3 ...(iv)
⎝L⎠ 1280 ⎠
Height of the hospital building shall be 10 m. Let x be the distance from the
crest, (then, x = + 120 m). Non-dimensional factors
⎛ x ⎞ ⎛ 120 ⎞
⎜ ⎟ = ⎜ ⎟ = + 0.09375
⎝ Le ⎠ ⎝ 1280 ⎠

⎛H⎞ ⎛ 10 ⎞
⎜ ⎟ = ⎜ ⎟ = + 0.0078
⎝ Le ⎠ ⎝ 1280 ⎠
From Fig. 9.6 (b), the value of S for ridge and hill
S = 1.00 ...(v)
From Eq. 9.6,
k3 = (l + C.S) = (l + 0.3 × 1.00) = 1.3 ...(vi)
Step 5: Design wind speed (from Eq. 9.3)
Vz = k 1. k 2 . k3. Vb
Vz = [(1.06) . (0.93) . (1.3) . (39)]
= 49.98 m/sec
 DESIGN OF ROOF TRUSSES 637

Step 6 : Design wind pressure (from Eq. 9.12)

pz = 0.0006 . Vz2
pz = 0.0006 × (49.98)2 = 1.4988 kN/m2
Example 9.4 A communication tower of 80 m height is proposed to be built
over hill top of height 520 m with a gradient of 1 in 5. The horizontal approach
distance is 2.8 km from the level ground. The tower is proposed at Abu mount.
Determine the design wind pressure.
Solution
Step 1 : Basic wind speed: From IS : 875 (Part 3) – 1987, or Table 9.2. the
basic wind speed at Abu mount (wind zone 5),
Vb = 39 m/sec ..(i)
Step 2 : Risk coefficient (factor – k1)
For important buildings and structures such as the communication towers,
the design life may be assumed 100 years. From IS : 875 (Part 3) – 1987, or
Table 9.3, for Vb = 39 m/sec,
k1 = 1.06 ...(ii)
Step 3 : Terraine, height and structure size factor, k 2
The height of communication tower is 80 m. The structure shall be having
maximum vertical dimension more than 50 m. The structure belongs to class
⎛C⎞
⎜⎝ 1 ⎟⎠ . The exposed open terraine with few or no obstruction and in which the
average height of any object surrounding the structure is less than 15 m. The
terraine belongs to category-1. The terraine factor
⎡ (1 ⋅ 20 – 1 ⋅ 14 ) × 30 ⎤
k 2 = ⎢1 ⋅ 14 + ⎥ = 1.176
⎣ 50 ⎦
Step 4 : Topography factor k3
The effective height of the hill, Z is 520 m. The slope is 1 in 5
⎛ 1⎞
θ = tan–1 ⎜⎝ ⎟⎠ = 11.31°
5
The length of upward slope, L is 2.800 km, that is, 2.800 m. The effective
length of hill,
for θ = ll.31° < 17°, Le =L = 2800 m
Value of C for θ 17°
⎛ Z ⎞ ⎛ 1 ⋅ 2 × 520 ⎞
C = 1.2× ⎜ ⎟= ⎜ ⎟ = 0.222857
⎝ L ⎠ ⎝ 2800 ⎠
For H = 80 m, and x = 0,
⎛ x ⎞
⎜ ⎟ = 0 and
⎝ Le ⎠

⎛H ⎞ ⎛ 80 ⎞
⎜ ⎟ = ⎜ ⎟ = 0.02857
⎝ Le ⎠ ⎝ 2800 ⎠
638 DESIGN OF STEEL STRUCTURES

From Fig. 9.6 (b), or IS : 875 (Part 3) – 1987,

S = 1.00
k3 = (1+C.S)
k3 = (1 + 0.22857 × 1) = 1.222857
Step 5: Design wind speed (from Eq. 9.3)
Vz = k 1 . k2 . k3 . Vb
Vz = (l.06) . (1.176) . (1.222857) × (39))
= 59.45 m/sec
Step 6: Design wind pressure (from Eq. 9.12)
pz = 0.0006 . Vz2 kN/m2
pz = (0.0006 × 59.452) = 2.121 kN/m2

9.23 DESIGN OF ROOF TRUSSES

The following are usual steps in the design of roof trusses :
Step 1. The type of truss which suits the requirements of span and roof
covering to be used is selected.
Step 2. The general proportion for the type of truss selected is fixed as
discussed in Art. 9.3. The line diagram of roof truss is drawn.
Step 3. The dead and live loads acting on the roof truss are determined as
discussed in Art. 9.6. The wind load on roof truss depending upon slope of the
truss and other conditions is found as discussed in Sec. 9.6. Thus, various loads
acting on the roof truss are estimated.
Step 4. The purlins are ordinarily provided at the panel points. The purlin
section is designed as discussed in Sec. 9.22.
Step 5. After the design of purlins, the loads acting at the panel points of the
roof truss are determined.
Step 6. The forces in various members of the roof truss are determined for
various combinations of loads as discussed in Sec. 9.14. The forces in roof truss
may be found analytically or graphically by drawing force (stress) diagram. A
design table for forces in various members of the roof truss is prepared. In this
design table, the forces in various members for different combinations of loads
are tabulated. The design forces for various members of the roof truss are noted.
Step 7. The compression members of roof truss are designed as discussed in
Chapter 3. The principal rafters of roof truss are designed as continuous struts
as discussed in Sec. 3.11. Other comspression members, have their lengths in
between two joints and these are not continuous. Such members are designed
as discontinuous struts as discussed in Sec. 3.11. The double angle sections will
be designed for principal rafter and single angle sections are designed as far as
possible for other members. The unequal angle sections are selected for double
angle sections. The long legs of angles sections are connected together. The
equal angle sections are selected for single angle section. The angle sections
smaller in size than 50 mm × 50 mm × 6 mm will not be used. The members are
so designed that their slenderness ratios are not greater than 180. The effective
 DESIGN OF ROOF TRUSSES 639

length of compression members in roof truss is adopted as 0.7 to 1.0 times the
distance between centre to centre of intersections of longitudinal axes of members
at the joint.
Step 8. The tension members are designed as discussed in Chapter 5. The
double angle sections are used for all tension members except those which are
lightly stressed. The single angle sections are objectionable. The single angle
sections have a tendency to twist the truss and produce eccentric forces at the
joint. The long legs are kept outstanding. The angle sections smaller in size
than 50 mm × 50 mm × 6 mm will not be used. If reversal of force takes place in
tension members, their slenderness ratio should be less than 350.

P rincip al
ra fte r

M ain tie

G usset plate
S h oe a ng le s

B a se p la te

B e aring p la te

A n ch o r bo lt

Fig. 9.32 Elements at end support of a roof truss

Step 9. The members meeting at a joint should have their centroidal axes
intersecting at one point so as to avoid eccentricity effect. The joints are
designed for the forces in the members. When the members are continuous over
a joint, the forces on both sides of the joints act in opposite directions. The rivets
640 DESIGN OF STEEL STRUCTURES

are provided for the difference of their values. The rivets are placed symmetrical
to the joints. The rivet of nominal diameter less than 16 mm will not be used.
The thickness of gusset plate at any joint will not be less than 6 mm.
Step10. The lateral bracings for roof trusses are provided if necessary.
Step11. Design of end support. The various elements used at the end
supports of truss are shown in Fig. 9.32. The design ni end supports consists of
(i) design of shoe angle, (ii) design of bearing plate, and (iii) design of anchor
bolt.
(i) Design of shoe angles. The shoes angles are designed to accommodate
number of rivets necessary to transmit maximum end reactions. The number of
rivets are found by dividing maximum reaction by rivet value. If the number of
rivets works out small, then, rivets are provided on the basis of practical
considerations. Two shoe angles are used. The shoe angles are connected with
the gusset plate. The horizontal legs of shoes agles are connected with base
plate. The number of rivets for connecting horizontal legs of shoe angles is kept
equal to the number of rivets used for connection of vertical legs. The nominal
size of 75 mm × 50 mm × 6 mm may be used for the shoe angles.
(ii) Design of bearing plate. The bearing plate is provided to distribute
end reaction so that the bearing pressure on masonry or concrete does not exceed
the allowable bearing pressure. Let R be the maximum end reaction. The area
of bearing plate required

⎛ R ⎞
A = ⎜ ⎟
⎝ σp ⎠
where σp is the allowable bearing pressure in masonry or concrete and it is
taken as 4 N/mm2.
The width of bearing plate b be kept equal to twice the length of horizontal
leg of shoe angle plus the thickness of gusset plate.
The length of bearing plate.

⎛ Area of bearing plate ⎞

l = ⎜ ⎟
⎝ Width of bearing plate ⎠
If σP1 is actual bearing pressure in masonry or concrete, then σP1 should be
less than σp.
The thickness of bearing plate t, is designed to resist the bending moment M,
at xx caused due to actual bearing pressure acting at the bottom of bearing
plate, as shown in Fig. 9.33. The moment of resistance of bearing plate.

⎛ 1 2 ⎞
M1 = (σBS × z) = ⎜ σbs × × x × t1 ⎟ (3M = M1)
⎝ 6 ⎠

1
⎛ 6M ⎞ 2
Therefore, t1 = ⎜ ...(9.36)
⎝ x × σbs ⎟⎠
 DESIGN OF ROOF TRUSSES 641

where, Z = Modulus of section of bearing plate

σbs = Allowable bearing pressure in steel bearing plate
= 185 N/mm2 (MPa)

S h oe a ng le

G usset plate

x
W idth o f be aring plate
Fig. 9.33 Shoe angles and bearing plate

∴ t = (t thickness of angle)
A base plate is provided under the bearing plate. The thickness of base plate
is kept equal to the thickness of bearing plate.
(iii) Design of anchor bolt. The anchor bolts are designed to resist net
uplift pressure acting on the roof truss.
One end of the truss is kept fixed, while slots of elliptical shape are provided
at the other end. The elliptical slots are provided in the horizontal legs of those
angles and in the bearing plate. This provides allowance for the expansion or
contraction of the truss. The expansion or contraction of truss may take place
due to rise or fall of temperature. The bearing plate can slide over the base
plate.
Step 12. After the design of all the elements of roof truss, a complete drawing
is prepared. The complete drawing is necessary for fabrication, The complete
drawing is also useful for estimating the weight of truss.

9.24 TYPICAL DETAILS OF ROOF TRUSSES

In Fig. 9.34 details at the shoe joint of a roof truss have been shown. The roof
trusses have horizontal movement at their support due to expansion and contraction
of its members under change of temperature. In general provision for horizontal
movement at support is made at one support either by providing a sliding plate as
shown in Fig. 9.34 (a) or by providing a roller as shown in Fig. 9.34 (c). The joints
642 DESIGN OF STEEL STRUCTURES

in roof trusses are so designed that lines of action of forces meet at joint to avoid
effect of eccentricity. There is eccentricity in the connections shown in Fig. 9.34
(a), (b) and (c). If these connections are made as shown in Figure 9.34 (g), (h)
and (j) ; this eccentricity may be avoided. Figure 9.34 (b), (d) and (f) show trusses
supported on columns. When the roof trusses are supported over columns,
provision for horizontal moment is not made.

S lid in g p la te R o lle rs R o cker

(a ) (b ) (c)

C o lu m n con ne ctio ns
(d ) (e ) (f)

Fixe d su p po rts
(g ) (h ) (j)
Fig. 9.34 Details at shoe of a roof truss

The typical details at the ridge and eves of roof trusses are shown in Fig. 9.35.
A.C.C. sheets are shown as roofing material over the trusses. A.C.C. sheets are
connected to the angle purlins by hook bolts. The bitumen washers are provided
under G.I. washers to make the joint tight.
 DESIGN OF ROOF TRUSSES 643

R id g e
H o ok b olt

A .C .C . G .I. w a sh er
she ets B itum e n
w a sh er

(a )
D e tail of rid ge jo in t

A .C .C .
she ets

E a ve s
g utter

(b )
D e tail of ea ve s jo in t
Fig. 9.35 Typical details of A.C.C. sheeting

Figure 9.36. shows the details of northlight glazing.

G .I. W a she r A .C . n orth

H o ok b olt A .C .C . ligh t cu rve
she et
B itum e n
w a sh er A .C .C .
she ets
N .L . g la zing

G utte r

Fig. 9.36 Detail at north light glazing

644 DESIGN OF STEEL STRUCTURES

Example 9.5 Locate the principal axes of ISA 200 mm × 100 mm × 10 mm

(ISA 200 100, @ 0.228 kN/m). Determine principal moment of inertia and radii
of gyration about the principal axes for this angle section.
Solution
Step 1 : Geometrical properties of given section
From ISI Handbook No, 1, ISA 200 mm × 100 mm × 10 mm (ISA 200 100, @
0.228 kN/m)
Sectional area, A = 2903 mm2
cxx = 69.6 mm
cyy = 20.1 mm
exx = 130.4 mm
eyy = 79.9 mm
Ixx = 1210 × 104 mm4
Iyy = 209.2 ×104 mm4
Step 2 : Product of inertia of section
The angle section is divided into two rectangle. Let A1 and A2 be the areas of
these rectangles
A1 = 100 × 10 = 1000 mm2 and A2 = (200 – 10) × 10 = 1900 mm2
x1 = (50 – 20.1) = 29.9 mm and y1 = (69.6 – 5.0) = 64.6 mm
x2 = – (20.1 – 5.0) = –15.1 mm and y2 = – (130.4 – 95) = 35.4 mm

Y
V
N
e yy = 2 0.1 m m

1 00 m m
B
A
10 m m
C xx = 6 0 .6 m m

C
U ∝

β= 70 °2 5
X X
O ∝= 15 °3 0
200 m m

exx = 13 0.4 m m

C D
N V
10 m m y

Fig. 9.37
 DESIGN OF ROOF TRUSSES 645

Product of inertia Ixx about xx-axis and yy-axis

Ixy = (A1 . x1y 1 + A2 . x2y2)
Ixy = (10 × 2.99 × 6.46)+ {19 ×(–1.51) × (3.54)}
× 104 mm4
= 295.3 × 10 mm4

From Eq. 9.6.

⎛ –2 I xy ⎞
tan 2 α = ⎜
⎜ I xx – I yy ⎟⎟
⎝ ⎠

⎛ ( –2 × 295 ⋅ 3 ) × 104 ⎞
= ⎜ = – 0.59
4 ⎟
⎝ (1210 – 209 ⋅ 2 ) × 10 ⎠
∴ 2α = – 31°, α = – 15° 30´
tan α = – 0.275
(From ISI Handbook No. 1, tan α = – 0.27)
The principal axes are as shown in Fig. 9.37.
The negative sign indicates that angle α is to be measured clockwise from xx-
axis.
Step 3 : Principal moment of inertia of section (From Eq. 9.9),
1/ 2
2
⎛ I xx + I yy ⎞ ⎡⎛ I xx – I yy ⎞ 2
⎤
IUU = ⎜ ⎢
⎟+ ⎜ ⎟ + I yy ⎥
⎝ 2 ⎠ ⎢⎣⎝ 2 ⎠ ⎥⎦

2 1/ 2
⎛ 1210 + 209 ⋅ 2 ⎞ ⎡⎛ 1210 – 209 ⋅ 2 ⎞ ( 2
⎤
=⎜ +
⎟ ⎜⎢ ⎟ + 295 ⋅ 3 ) ⎥ × 104 mm4
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
IUU = 1291.6 × l04 mm4
(From ISI Handbook No. 1, IUU = 1286.7 × 104 mm4)
Step 4: Radius of gyration of section
1/ 2
2
⎛ I – I yy ⎞ ⎡⎛ I xx – I yy ⎞ ⎤
IVV = ⎜ xx ⎟ – ⎢ ⎜⎝
2
⎟ + I yy ⎥
⎝ 2 ⎠ ⎣ 2 ⎠ ⎦

2 1/ 2
⎛ 1210 + 209 ⋅ 2 ⎞ ⎡⎛ 1210 – 209 ⋅ 2 ⎞ ( 2
⎤
=⎜ ⎟ – ⎢⎜ ⎟ + 209 ⋅ 3 ) ⎥ × 104 mm4
⎝ 2 ⎠ ⎣⎝ 2 ⎠ ⎦
= 127.6 × 104 mm4
(From ISI Handbook No.1 IVV= 132.5 × 104 mm4)
1/ 2 1/ 2
⎛I ⎞ ⎛ 1291 ⋅ 6 × 104 ⎞
rUU = ⎜⎝ UU ⎟⎠ = ⎜ ⎟ = 66.6 mm
A ⎝ 2903 ⎠
(From ISI Handbook No. 1, rUU = 66.6 mm)
646 DESIGN OF STEEL STRUCTURES

1/2 1/ 2
⎛I ⎞ ⎡127 ⋅ 6 × 104 ⎤
rVV = ⎜⎝ VV ⎟⎠ = ⎢ ⎥ = 21.0 mm
A ⎣ 2902 ⎦
(From ISI Handbook No. 1, rVV = 21.4 mm)
The actual angle section is rounded at the toe and at the root. In above calcu-
lations, angle section is not rounded. As a result of this, there are small differ-
ences in the values calculated and the values noted from Handbook No. l.
Example 9.6 In Example 9.5, the angle section is subjected to bending
moment 6 kN-m acting in the vertical plane through the centroid of section.
Determine the maximum bending stress induced in the angle section.
Solution :
Step 1: Principal moment of inertias of section
From Example 9.5,
IUU = 1291.6 × 104 mm4
IVV = 127.6 × 104 mm4
The plane of loading is vertical and passes through the centroid of the angle
section. From Eq. 9.15 for neutral axis

⎛I ⎞ ⎛ 1291 ⋅ 6 × 104 ⎞
tan β = ⎜ uu ⎟ tan θ ⎜ × tan15° 30´ = + 2.81
⎝ Ivv ⎠ ⎝ 127 ⋅ 6 × 104 ⎟⎠
∴ β = + 70° 25´
The value of β is positive. It is plotted anti-clockwise from principal axis UU
is shown in Fig. 9.37. From Fig. 9.37, it is seen that the point D is at extreme
distance from the neutral axis.
Step 2 : Co-ordinates of farthest point D
xD = – (20.1 –10) = –10.1 mm
yD = – 130.4 mm, α = – 15° 30'
uD = (– XD COS α + yD sin α)
= [–10.1 × cos (–15° 30') + (– 1 30.4) × sin (15° 30')]
= + 25.1 mm
vD = (y D cos α – xD sin α)
= [–130.4 cos (–15° 30') – (10.1) × sin (– 15°30')]
= – 128.5 mm
MVV = M . cos θ and MVU – M . sin θ
Step 3 : Maximum bending stress at point D (From Eq. 9.12),
⎛ M cos θ ⎞ ⎛ M sin θ ⎞
σb = ⎜ ⋅ vD + ⎜ ⋅ uD
⎝ IUU ⎟⎠ ⎝ I vv ⎟⎠
Both the components of bending stresses are additive
⎛ 6 × 106 cos15°30´ ⎞ ⎛ 6 × 106 sin15°30´ ⎞
σb = ⎜ ⎟ × 128 ⋅ 5 + ⎜ ⎟ = 88.8 N/mm2
⎝ 1291 ⋅ 6 × 104 ⎠ ⎝ 127 ⋅ 6 × 104 ⎠
The maximum bending stress at point D is tensile.
 DESIGN OF ROOF TRUSSES 647

(i) Alternative. The moment of inertia of the section about neutral axis
fromEq. 9.16
INN = (IUU . cos2 β + IVV . sin2 β)
= [1291.6 × cos2 70° 25´+ 127.6 sin2 70°
25´] × 104 mm4
= 258.3 × l04 mm4
(β – θ) = (70° – 25´ – 15° 30´)
= 54º55'
YN = (– vD . cos β - uD . sin β)
∴ YN = (– 128.5 × 0.335 – 25.1 × 0.942)
= – 66.6 mm
From Eq. 9.17, maximum bending stress at point D

⎡ M ⋅ cos (β – θ ) ⎤
σb = ⎢ ⎥ .YN
⎣ I NN ⎦

⎛ 6 × 106 cos 54°55´ × 66.6 ⎞

=⎜ ⎟⎠ = 88.7 N/mm
2
⎝ 258 ⋅ 3 × 104
The maximum bending stress at the point D is tensile.
(ii) Alternative. The maximum bending stress at point D may be found by
Eq. 9.13. Eq. 9.13 is useful specially for beam sections in which less of angle
section are parallel to xx and yy axes. From Eq. 9.13, the bending stress at point
D,

⎛ M xx I yy – M yy I xy ⎞ ⎛ M yy I xx – M xx I xy ⎞
σb = ⎜ 2 ⎟ ⋅ yD ⎜ 2 ⎟ x D, and φ = 0°
⎝ I xx I yy – I xy ⎠ ⎝ I yy I xx – I xy ⎠
Mxx = M cos φ = 6 × 106 kN-m and Myy = M sin φ = Zero.

⎛ M xx I yy – M yy I xy ⎞ ⎛ 6 × 106 × 209 ⋅ 2 × 104 ⎞

a2 = ⎜ 2 ⎟ = ⎜ 2 ⎟ = 0.754
⎝ I xx I yy – I xy ⎠ ⎝ ⎣⎡1210 × 209 ⋅ 2 – (29 ⋅ 3) ⎦⎤ × 108 ⎠

⎛ M yy I xx – M xx I xy ⎞ ⎛ –6 × 106 × 295 ⋅ 3 × 104 ⎞

a2 = ⎜ 2 ⎟ = ⎜ 2 ⎟ = –1.06
⎝ I yy I xx – I xy ⎠ ⎝ ⎣⎡1210 × 209 ⋅ 2 – (295 ⋅ 3) ⎦⎤ × 108 ⎠

σb = (0.754 yD – 1.06 xD)

= [0.754 (– 130.4) – (1.06)(– 10.1)] = – 87.8 N/mm2.
The maximum bending stress at point D is tensile and it has been worked out
from Eqs. 920, 921 and 925. The values differ slightly.
Example 9.7 LB 200, @ 0.198 kN/m is subjected to bending moment 12 kN-
m. The plate of loading passes through centroid of beam section, and it is inclined
8° with the YY–axis in the anti-clockwise direction. Locate the neutral axis.
Determine maximum bending stress induced in the beam section.
648 DESIGN OF STEEL STRUCTURES

Solution
Step 1. Geometrical properties of section
From IS I Handbook No. 1, for LB 200, @ 0.198 kN/m
Depth of section, h = 200 mm
Width of section, b = 100 mm
IUU = Ixx= 1696.6 × 104 mm4
IVV = Iyy =115.4 × 104 mm4
It is to note that I-section shown in Fig. 9.33 is a symmetrical section.
The axes of symmetry XX and YY are principal axes UU and VV
From Eq. 9.23, for neutral axis
⎛I ⎞
tan β = + ⎜ UU ⎟ tan θ
⎝ IVV ⎠

or tan β = + ⎛⎜ 1696 ⋅ 6 ⎞⎟ tan 8°

⎝ 115 ⋅ 5 ⎠
β = 64° 7´. The value of β is positive. It is plotted anti-clockwise from princi-
pal axis-UU as shown in Fig. 9.38.

M Y N
(V)
B

2 00 m m

β = 64 °7 '
x x
(U) (U)

1 00 m m
Y
N (V)

Fig. 9.38 I-section purlin

Step 2. Moment of inertia of beam section about neutral axis (from
Eq. 9.24)
INN = (IUU .cos2 β + IVV . sin2 β )
or INN = [1696.6 cos2 64° 7´+ 115.4 × sin2 64° 7´]
× 104 mm4
4
= 417.5 ×10 mm 4

(β – θ) = [(64°7´) – 8°] = 56°7´

Step 3. Co-ordinates of points A and D are at extreme distance from
the neutral axis.
 DESIGN OF ROOF TRUSSES 649

∴ yN = [100 cos β –(–50) sin β]

= [(100 cos 64° 7´+ 50 sin 64° 7´)]
= 88.6 mm
Step 4. Maximum bending stress at point A (from Eq. 9.1)
⎡ M ⋅ cos (β – θ ) ⎤
σb = ⎢ ⎥ ⋅ yN
⎣ I NN ⎦

⎛ 12 × 106 cos 56°7´ ⎞

= ⎜ ⎟ × 88.6
⎝ 417 ⋅ 5 × 104 ⎠
= 141.8 M/mm2
The maximum bending stress at point A is compressive and that at point D is
tensile.
Alternative. The maximum bending stress at point A, can be found by
resolving bending moment along VV-axis and UU-axis. From Eq. 9.20, bending
stress at point A,
⎛ M cos 8° ⎞ ⎛ M sin 8º ⎞
σb = ⎜ ⎟ ⋅ vA + ⎜ uA
⎝ 1UU ⎠ ⎝ IVV ⎟⎠

⎛ 121 × 106 × 0 ⋅ 99 × 100 ⎞ ⎛ 12 × 106 × 1 ⋅ 39 × 50 ⎞

or σb = ⎜ ⎟+⎜ ⎟ N/mm2
⎝ 1696 ⋅ 6 × 104 ⎠ ⎝ 115 ⋅ 54 × 104 ⎠
= 141.8 N/mm2.
The maximum bending stress at point A is compressive and that at the point
D is tensile.
Example 9.8 In Example 9.7, the beam section is carrying uniformly
distributed load 6 kN/m. The effective span is 4 m. The beam is simply supported
at both ends. The plane of loading passes through centroid of beam section and
it is inclined 8 ° with the yy-axis. Determine the magnitude and direction of
maximum deflection. Take E = 2.04 × 105 N/mm2.
Solution
Step 1. Properties of section
From Example 9.7, the inclination of neutral axis from UU-axis
β = + 64.7'
Moment of inertia about neutral axis
INN = 417.5 × 104 mm4
(β – θ) = [(64° 7´) – 8°] = 56°7´
Step 2. Maximum deflection for the beam section (from Eq. 9.33),

5 ⎛ w ⋅ l4 cos(β − θ) ⎞
δ = ⋅⎜ ⎟ mm
384 ⎝ E ⋅ I NN ⎠

5 ⎛ 6 × 103 × 4 × (4000)3 × cos 56° 7´ ⎞

δ = ×⎜ ⎟
384 ⎝ 2.04 × 105 × 417.5 × 104 ⎠
= 13.2 mm
650 DESIGN OF STEEL STRUCTURES

The maximum deflection occurs in the direction perpendicular to neutral

axis NN. It occurs in the direction β´measured from UU principal axis,
β´ = (β + 90°)
= (64° 7´+ 90°) = 154°7´
Example 9.9 Draw Z polygon for ISA 200 mm × 100 mm × 10 mm. The angle
section is subjected to bending moment 6 kN-m acting in the vertical plane through
the centroid of section. Determine the maximum bending stress induced in the
angle section. Also determine the absolute maximum bending stress for the given-
moment and corresponding inclination of plane of loading with YY-axis.
Solution
Step 1 : Geometrical propertion of given section
From ISI Handbook No. 1 for ISA 200 mm × 100 mm × 10 mm. (ISA 200 100,
@ 0.228 kN/m)
Sectional area, A = 2903 mm2
cxx = 69.6 mm2, cyy = 20.1 mm
exx = 130.4 mm, cyy = 79.9 mm
Ixx = 1210 × 104 mm4, Iyy = 209.2 × 104 mm4
4
Iuu = 132.5 × 10 mm 4 Ivv = 1286.7 × 104 mm4,
tan α = – 0.270, ∴ α = –15.7´

Plan e of
lo ad in g
P 6 α= θ
Z-lin e for B
A B
E
U Z-lin e for A
P8 P2
P9
x x
O α = 1 5°7 '
Z-lin e for E U
P3
Z-lin e for D
P7 P P P P P P
1 2 3 4 5 1
P4 Z-p o lygo n

C ISA 2 00 m m × 10 0 m m
P5 D × 10 0 m m
Z-lin e for C V Y

Fig. 9.39

It is to note that angle α is negative. It is plotted in clock-wise direction from

XX-axis as shown in Fig. 9.39. UU and VV are principal axes.
 DESIGN OF ROOF TRUSSES 651

Step 2 : Co-ordinates of corner points with reference to UU and VV-

axes.
(i) Corner point A,
xA = –20.1 mm, yA = + 69.6 mm
uA = (xA . cos α + yA . sin α)
= [(–20.1) cos (–15.7´) + 69.6 sin (–15.7´)]
= – 37.6 mm
vA = (yA . COS α – xA . sin α )
= (69.6 × 0.956 – (– 20.1)(– 0.261))
= 51.9 mm
(ii) Corner point B,
xB = + 79.6 mm, yB = + 69.6 mm
uB = (x B . COS α – y B . sin α )
= (79.9 × 0.965 – 69.6 × 0.261) = + 58.8 mm
vB = (y B . COS α – xB . sin α )
= [(69.6 × 0.965)–(79.9)(– 0.261)] = 88.0 mm
(iii) Corner point C,
xD = – 20.1 mm, yc = – 130.4 mm
uC = (x C . COS α – x C . sin α )
= [(–20.1 × 0.965) + (–130.4)(– 0.261)]
= + 14.7 mm
vC = (y C . COS α – x C . sin α )
= [–130.4 × 0.965 – (– 20.1)(– 0.261)]
= –131.3 mm
(iv) Corner point D,
xD = 10.1 mm, yD = –130.4 mm
uD = (x D . COS α – y D . sin α )
= [(–10.1)(0.965) + (–130.4)(– 0.261)]
= + 24.4 mm
vD = (y D . COS α – xD . sin α )
= [(–130.4)(0.965) – (– 10.1)(– 0.261)]
= 128.6 mm
(v) Corner point E,
xE = (+ 79.9 mm, yE = (69.6 – 10))
= + 59.6 mm
uE = (x E . COS α – y E . sin α )
= 79.9 × 0.965 + (59.6)(– 0.261)
= + 61.4 mm
vE = (y E . COS α – x E . sin α )
= [59.6 × 0.965 – (79.9)(– 0.261)]
= 78.4 mm.
652 DESIGN OF STEEL STRUCTURES

Step 3. Equation of Z-lines for corner points. It is to note that, now co-
ordinates (u, v) are taken positive in second quadrant.
(i) Corner point A, from Eq. 9.29

u v u v
+ = 1, + = 1 × 104
⎛ IVV ⎞ ⎛ IUU ⎞ ⎛ 132.5 ⎞ ⎛ 1286.7 ⎞
⎜ ⎟ ⎜ ⎟ ⎜⎝ ⎟ ⎜ ⎟
⎝ uA ⎠ ⎝ VA ⎠ 37.6 ⎠ ⎝ 51.9 ⎠
or v = ( –7.06u + 24.8 × 104) ...(i)
(ii) Corner point B

u v u v
+ = 1, + = 1 × 104
⎛ IVV ⎞ ⎛ IUU ⎞ ⎛ 132.5 ⎞ ⎛ 1286.7 ⎞
⎜ ⎟ ⎜ ⎟ ⎜⎝ ⎟ ⎜ ⎟
⎝ uB ⎠ ⎝ VB ⎠ −58.8 ⎠ ⎝ 88.0 ⎠
or v = ( 6.5u + 14.6 × 104) ...(ii)
(iii) Corner point C
u v u v
+ = 1, + = 1 × 104
⎛ IVV ⎞ ⎛ IUU ⎞ ⎛ 132.5 ⎞ ⎛ 1286.7 ⎞
⎜ u ⎟ ⎜V ⎟ ⎜ ⎟ ⎜ ⎟
⎝ C ⎠ ⎝ C ⎠ ⎝ −14.7 ⎠ ⎝ –131.3 ⎠

or v = ( 1.08u – 9.78 × 104) ...(iii)

(iv) Corner point D
u v u v
+ = 1, + = 1 × 104
⎛ IVV ⎞ ⎛ IUU ⎞ ⎛ 132.5 ⎞ ⎛ 1286.7 ⎞
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ uD ⎠ ⎝ VD ⎠ ⎝ −24.4 ⎠ ⎝ –128.6 ⎠

or v = ( 1.84u – 10.9 × 104) ...(iv)

(v) Corner point E
u v
+ = 1 or v = ( 7.46 u – 16.4 × 104) ...(v)
⎛ 132.5 ⎞ ⎛ 1286.7 ⎞
⎜ ⎟ ⎜ ⎟
⎝ −61.4 ⎠ ⎝ 78.4 ⎠
Step 4. Z-polygon
(i) The intersection of Z-lines for points A and B locales point P1 of Z-polygon.
Z – line for point A, v = (– 7.06u + 24.8 × 104) [from Eq. (i)]
4
Z – line for point Bv,v = (6.5u + 14.6 × 10 ) [from Eq. (ii)]
Solving equations (i) and (ii),
Co-ordinates of P1 are (75.4, 1948).
(ii) The intersection of Z-lines for points B and E locates point P2 of Z poly-
gon.
Z – line for point B, v = (6.5u+ 14.6 × 104) [from Eq. (ii)
Z-line for point E, u = (7.4u+ 16.4 × 104) [from Eq. (v)
Solving Eqs., (ii) and (v), co-ordinates of point P2 are (–188, 240).
 DESIGN OF ROOF TRUSSES 653

(iii) The intersection of Z-lines for points E and D, locates P3 of Z-polygon.

Z-line for point E, v = (7.46u + 16.4 × 104) [from Eq. (v)]
.
Z-line for point D, v = (–1 84u –10.0 × 10 )4 [from Eq. (iv)]
Solving Eqs. (v) and (iv), co-ordinates of point P4 are (–283, –580).
(iv) The intersection of Z-lines for points D and C, locates point PA of Z-poly-
gon.
Z-line fur point D, v = (–1.84 u – 10.0 × 104) [from Eq. (iv)]
Z-line for point C, v = (–1.08 u – 9.78 × 104) [from Eq. (iii)]
Solving Eqs. (iv) and (iii), co-ordinates of point P5 are (–289, 946.8).
(v) The intersection of Z-lines for points C and A, locates point P5 of Z polygon.
Z-line for point C, v = (–1.08 u – 9.78 × 104) [from Eq. (iii)]
Z-line for point A, v = (–7.06 u + 2.48 × 104) [from Eq. (i)]
The points P1, P2, P3, P4 and P5 are plotted with reference to UU and VV
principal axes. Join P1 and P2, P2 and P3, P3 and P4, P4 and P5 and P5 and P1.
P1P2P3P4P5P1 represents Z-polygon.
The bending moment M is acting in the vertical plane passing through the
centroid of section. The vertical plane through the centroid intersects Z-line of
point A at P6 and Z-line of point D at P7.
From Fig. 9.39, OP6 = 88 × 103 mm3, OP7 = 68 × 103 mm3
Therefore, point D is subjected to maximum bending stress
Step 5. Maximum bending stress
6
⎛ M ⎞ ⎛ 6 × 10 ⎞
σb = ⎜ ⎟=⎜ 3⎟
⎝ OP7 ⎠ ⎝ 68 × 10 ⎠
= 88.235 N/mm2
The bending stress at point D is tensile.
For absolute maximum bending stress draw perpendiculars OP8 and OP9
from centroid O. OP9 gives minimum modulus of section. OP9 = 24 × 103 mm3.
Absolute maximum bending stress for the given moment

⎛ 6 × 106 ⎞
σb = ⎜ ⎟ = 250 N/mm2
⎝ 24 × 103 ⎠
For absolute maximum bending stress, the inclination of plane of loading is
96° clockwise from vertical.
Example 9.10 A roof truss-shed is to be built in Jodhpur city area for an
industrial use. The size of shed shall be 18 m × 30 m. Determine the basic wind
pressure (Example 9.1).
Roof trusses of 18 m span and 4.5 rise shall be provided at 4 m spacing centre
to centre. The height of cladding or eave level is 6 m. Determine the wind load
parallel to the ridge causing a total outward normal wind pressure on the interior
purlins, for (i) small permeability (ii) medium permeability and (iii) high
permeability.
654 DESIGN OF STEEL STRUCTURES

Solution
Step 1 : From Exmaple 9.1,
The basic wind speed, Vb is 47 m/sec. The design wind speed (as worked out),
Vz is 41.641 m/sec. The design wind pressure. pz is l.04 kN/m2.
Step 2 : Geometry of truss
Span of the roof-trusses to be used is 18 m its rise shall be 4.5 m. Slope of the
roof
⎛ 4 ⋅5 ⎞
θ = tan–1 ⎜ ⎟ = 26.565° ...(i)
⎝ 9 ⎠
Length of the principal rafter
= [(4.5)2 + (9)2]1/2 = 10.06 m ...(ii)
Panel length should not be more than 1.4 m, so that the failure of roof cover-
ing sheets does not occur. A compound French roof truss may be used with 8
(eight) panels @ 1.25 m, so that there shall be 9 (nine) panel joints. The remain-
ing 6.23 mm shall be left near the ridge and near the eaves.
Total length covered on the principal rafter
= (2 × 31.15 + 8 × 1250)
= 10062.3 mm
The spacing of purlins shall be 1.25 m.
Step 3 : Wind load
Let the height of plinth be 600 mm and the projection of eave girder behond
centre line be 400 mm. The eave height of the roof shed above ground level.
h = (6 + 0.600) = 6.6 m
Gross-width of the roof-shed
w = (18 + 2 × 0.400) = 18.8 m
Ratio of height to width of roof shed
⎛h⎞ ⎛ 6⋅6 ⎞
⎜ ⎟ = ⎜ ⎟ = 0.351 < 0.5
⎝w⎠ ⎝ 18 ⋅ 8 ⎠
(A) External air pressure coefficients. The wind pressure coefficients
⎛h⎞
are to be adopted from a set ⎜ ⎟ < 0.5 and for slope θ = 26.565°. The wind
⎝w⎠
pressure coefficients are noted from IS : 875 (Part 3) – 1987, or from Table 9.7
(External pressure coefficients, Cpe for pitched roofs of rectangular clad build-
ings.

Side of Wind angle, θ

pitched roof θ = 0° * θ = 90° *
Windward – 0.1374 –.8687 – 0.7 –
Leeward – 0.4 –1.1344 – 0.7 –
(*Local effects)
 DESIGN OF ROOF TRUSSES 655

(B) Internal pressure coefficients. Internal air pressure in a building

depends upon the degree of permeability of cladding to the flow or air.
(i) For small permeability Cpi = ± 0.2
(ii) For medium permeability Cpi = ± 0.5
(iii) For high permeability Cpi = ± 0.7
Note. The internal pressure coefficient is added algebraically to the external pres-
sure coefficient and the worst combination shall be considered.
Step 4. Total wind load parallel to ridge
(i) For small permeability :
External + internal pressure coefficients
(Cpe–Cpi) = [– 0.7 – (+0.2)] = 0.9
Total wind load
Ww = (Cpe– Cpe) .pd · A
Ww = – 0.9 pd . A
(ii) For medium permeability :
External + internal pressure coefficients
(CPe–Cpi) = [–0.7– (+0.5)] = –1.2
Total wind load
Ww = – 1.2 pd . A
(iii) For high permeability :
External + internal pressure coefficients
(Cpe–Cpi) = [–0.7 –(+0.7)1 = –1.4
Total wind load
Ww = – 1.4 pd . A
Note. pd (design wind pressure) is 104 kN/m2. A (effective exposed area, that is,
spacing of purlins multiplied by the length of purlin is (1.25 × 4) = 5 m2. Therefore,
total wind load parallel to ridge for
(i) small permeability:
Ww = [(– 0.9).( 1.04).(5)] = – 4.68 kN
(ii) medium permeability:
Ww = [(– 1.2).( 1.04).(5)] = – 624 kN
(iii) high permeability:
Ww = [(–1.4). (1.04).(5)]= – 7.28 kN
Figures 9.40 (a), (b) and (c) show external air pressure coefficients, Cpe for
pitched roofs, and internal air pressure coefficients (positive and negative) for
small permeability buildings, medium permeability buildings and high
permeability buildings, respectively.
Example 9.11 Design an I-section purlin to support galvanised corrugated
iron sheet roof. The purlins are 1.25 m apart over roof trusses spaced 5 m centre
to centre. The roof surface has an inclination of 30 ° to the horizontal. The weight
of corrugated iron sheet is 0.1331 kN/m2. The weight of fixtures etc. 0.053 kN/m2.
656 DESIGN OF STEEL STRUCTURES

The design wind pressure for medium permeability is 1.20 kN/m2 (outward)
parallel to ridge.
Solution
Design :
Step 1. Dead load on purlin per metre length
Weight of galvanized corrugated iron sheets
(0.1331 × 1.25 × 1) = 0.1664 kN/m
Extra load for overlap and fixtures
(0.053 × 1.25 × 1) = 0.0663 kN/m

W ind – C pe – C pe – C pe – C pe

C pe C p i = + 0.2 C pe C p i = + 0.2 C pe

(a ) Sm a ll pe rm ea bility bu ilding s
(o pe ning s in cla dd in g 5 p erce nt)

W ind – C pe –C pe – C pe – C pe

C pe C p i = + 0 .5 C pe C p i = – 0.5 C pe

(b ) M ed iu m pe rm ea bility bu ilding s
(O p en in gs in cla dd in g 5 to 20 p ercen t)

W ind – C pe – C pe – C pe – C pe

C pe C p i = + 0.7 C p i = – 0.7 C pe

(c) H igh p erm ea b ility b uilding s

(O p en in gs in cla dd in g s > 2 0 pe rce nt)

Fig. 9.40 External and internal wind pressure coefficients

 DESIGN OF ROOF TRUSSES 657

Self-weight of purlin (assumed)

= 0.120 kN/m
The dead load = 0.3527 kN/m (say 0.360 kN/m)
Dead load acts vertically downward
Component of dead load normal to the roof
⎛ ⎞
(0.360 × cos 30°) = ⎜ 0 ⋅ 360 × 3 ⎟
⎝ 2 ⎠
= 0.312 kN/m
Component of dead load parallel to the roof
⎛ 0 ⋅ 360 × 1 ⎞
(0.360 × sin 30°) = ⎜ ⎟ = 0.180 kN/m
⎝ 2 ⎠
Step 2. Imposed (live) load
The imposed (live) load for sloping roof with slope grater then 10°
[0.75 – (30 – 10) (0.020)] = 0.35 kN/m2
It is subjected to a minimum of 04 kN/m2
Live load
(0.4 × 1.25 × 1) = 0.5 kN/m
Component of live load normal to the roof

⎛ 0 ⋅5× 3 ⎞
0.5 × cos 30° = ⎜ ⎟ = 0.433 kN/m
⎝ 2 ⎠
Component of live load parallel to the roof
⎛ 0 ⋅ 5 ×1 ⎞
(0. 5 × sin 30° ) = ⎜ ⎟ = 0.25 kN/m
⎝ 2 ⎠

V
N o rm a l Ve rtica l
lo ad in g 3 0° lo ad in g
U
P u rlin

U
V

3 0°

Fig. 9.41 I-Section purlin

Step 3. Wind load on Purlin

Given design pressure for medium permeability (parallel to ridge) is 1.20 kN/
m2 (outward or negative). That is, it acts upward. Wind load on purlin parallel
to the ridge.
658 DESIGN OF STEEL STRUCTURES

Step 4. Combination of loads

The purlins are designed for worst combination of the load and wind from
any direction. The worst combinations of loads for design of purlin are
(i) dead load plus live load, and
(ii) dead load plus live load plus wind load (paralle) to the ridge). In case,
the design wind pressure acts outward (negative), the imposed live load
shall not be considered.
Step 5. Design of purlin for dead load plus live load
Bi-axial bending of the purlin occurs due to dead load plus live load. The
purlins remain continuous and the end span has the maximum support (nega-
tive) moment.
1. Bending moment due to dead load plus live load parallel to the major
principal axis (UU-axis, Fig. 9.40)
⎡⎛ 0 ⋅ 312 × 5 ⎞ ⎛ 0 ⋅ 25 × 5 ⎞ ⎤
MUU = ⎢⎜ ⎟×5 + ⎜ ⎟ × 5⎥
⎣ ⎝ 10 ⎠ ⎝ 9 ⎠ ⎦
= (0.78 + l.2028) = 1.9828 kN-m
2. Bending moment due to dead load plus live load parallel to the minor
principal axis (VV-axis, Fig. 9.40)
⎡⎛ 0 ⋅ 180 × 5 ⎞ ⎛ 0 ⋅ 25 × 5 ⎞ ⎤
MVV = ⎢⎜ ⎟×5 + ⎜ ⎟ × 5⎥
⎣⎝ 10 ⎠ ⎝ 9 ⎠ ⎦
= (0.45 + 0.694) = 1.144 kN-m
Section modulus required for purlin Eq. 9.35
MUU ⎛ M VV ZUU ⎞
ZUU = ⎜⎝1 + M ⋅ ⎟ mm
σb UU ZVV ⎠

⎛Z ⎞
Assuming ⎜ UU ⎟ 7 for I-section purlin required modulus of section for purlin
⎝ ZVV ⎠
(σb = (0.66 × 250) = 165 N/mm2)
⎛ 1 ⋅ 9828 × 106 ⎞ ⎛ 1 ⋅ 144 ⎞
ZUU = ⎜⎝ ⎟⎠ ⋅ ⎜⎝1 + × 5.61⎟
⎠
165 1 ⋅ 9828
= 60.55 × 103 mm3
From ISI Handbook No.1, select LB 125, @ 0.119 kN/m
Modulus of section
ZUU = ZXX = 65.11 × 103 mm3 and
ZVV = ZYY = 11.6 × 103 mm3
It is to note that XX and YY axes are also the axes of symmetry for I-section.
Therefore, these are also principal axes of the section.

⎛ ZUU ⎞ ⎛ 65 ⋅ 1 × 103 ⎞
⎜⎝ Z ⎟⎠ = ⎜ ⎟ = 5.61
VV ⎝ 11 ⋅ 6 × 103 ⎠
 DESIGN OF ROOF TRUSSES 659

Step 6. Check for bending stress

Maximum bending stress in purlin from Eq. 9.34
MUU ⎛ M VV ZUU ⎞
σb = ⎜⎝1 + M ⋅ ⎟
ZUU UU ZVV ⎠

⎛ 1 ⋅ 9828 × 106 ⎞ ⎛ 1 ⋅ 144 ⎞

σb = ⎜ 3 ⎟
⋅ ⎜1 + × 5 ⋅ 61⎟
⎝ 65.11 × 10 ⎠ ⎝ 1 ⋅ 9828 ⎠
= 129.02 N/mm2
< 165 N/mm2. Hence satisfactory
Step 7. Design of purlin for dead load plus wind load
The wind load (given) acts outward. Therefore, the live load is not considered
as worst combination of load. From step 5, moment due to dead load parallel to
the major axes, UU
MU1U1 = + 0.780 kN-m
Moment due to dead load parallel to the minor axis, VV
MV1V1 = + 0.45 kN-m
Bending moment due to wind load acting outward about major principal
-axis, UU
⎛ 1 ⋅ 50 × 5 ⎞
⎜ ⎟ × 5 = –3.750 kN-m
⎝ 10 ⎠
Total design moments
MUU = (+ 0.780 – 3.750) = – 2.97 kN-m
Section modulus required for purlin from Eq. 9.35
MUU ⎛ M VV ZUU ⎞
ZUU = ⎜⎝1 + M ⋅ ⎟ mm
3
σb UU ZVV ⎠

⎛ ZUU ⎞
Ratio ⎜ is assumed as 7 for I-section
⎝ ZVV ⎟⎠

⎛ 6 ⎞
ZUU = ⎜ 2 ⋅ 97 × 10 ⎟ ⋅ ⎜⎛1 + 0 ⋅ 450 × 7 ⎟⎞
⎝ 1 ⋅ 33 × 165 ⎠ ⎝ 2 ⋅ 97 ⎠
= 27.888 × 103 mm3
Step 8. Check for bending stress
Maximum bending stress in the purlin from Eq. 9.34
MUU ⎛ M VV ZUU ⎞
σb = ⎜⎝1 + M ⋅ ⎟
σb UU ZVV ⎠

⎛ 2.97 × 106 ⎞ ⎛ 1 ⋅ 144 ⎞

σb = ⎜ ⎟ ⋅ ⎜1 + × 5 ⋅ 61 ⎟
⎝ 65 ⋅ 11 × 103 ⎠ ⎝ 1 ⋅ 9828 ⎠
= 84.388 N/mm2
< 165 N/mm2. Hence satisfactory
660 DESIGN OF STEEL STRUCTURES

Step 9. Local effects

It is to note that the local effects are for the design of fixtures and not for the
design of trusses or even purlins. The net wind pressure at the ridge due to local
effects as the wind is parallel to the ridge (From Example 9.10)
Ww = –(1.1344 + 0.5) pd . A
= – 1.6344 . pd . A
⎛ 1 ⋅ 6344 ⎞
This is ⎜ ⎟ = 1362 times that on the normal purlin.
⎝ 1 ⋅ 20 ⎠
Example 9.12. In Example design a channel-section purlin.
Solution
Design : In Example 9.11, the dead load and its components normal to and
parallel to the roof have been calculated in step 1. The imposed (live) load and
its components normal to and parallel to the roof have also been determined in
step 2. The wind load parallel to the roof acting normal to the roof have been
found in Step 3. The worst combinations of loads for the design of purlin have
been described in step 4. The section modulus of section for the purlin for dead
load and live load combinations from Step 5, Example 9.11
ZUU = 60.55 × l03 mm3
From IS : 808, select MC 125, @ 0.131 kN/m section modulus
ZUU = ZXX = 68.1 × 103 mm3
ZVV = ZYY = 13.4 × 103 mm3
It is to note that XX and YY axes are also the axes of symmetry for channel-
section. Therefore, there are also principal axes of the action.

⎛ ZUU ⎞ ⎛ 68 ⋅ 1 × 103 ⎞
⎜⎝ Z ⎟⎠ = ⎜ ⎟
VV ⎝ 13 ⋅ 4 × 103 ⎠
= 5.082
Step 1: Check for the bending stress
Maximum bending stress in the purlin from Eq. 9.34

MUU ⎛ M Z ⎞
σb = ⋅ ⎜1 + VV ⋅ UU ⎟
σVV ⎝ M UU ZVV ⎠

⎛ 1 ⋅ 9828 × 106 ⎞ ⎛ 1 ⋅ 144 ⎞

σb = ⎜ ⎟ ⋅ ⎜1 + × 5 ⋅ 082⎟
⎝ 68 ⋅ 1 × 103 ⎠ ⎝ 1 ⋅ 9828 ⎠

= 144.489 N/mm2
< 165 N/mm2 . Hence satisfactory
Step 2. Design of purlin for dead load plus wind load
From Example 9.11 total design moment
MUU = – 2.97 kN-m
MVV = 0.45 kN-m
 DESIGN OF ROOF TRUSSES 661

Step 3. Check for bending stresses

Maximum bending stress in the purlin from Eq. 9.34
MUU ⎛ M VV ZUU ⎞
σb = ⎜⎝1 + M ⋅ ⎟
σb UU ZVV ⎠

⎛ 2 ⋅ 97 × 106 ⎞ ⎛ 0 ⋅ 450 ⎞
σb = ⎜ ⎟ ⋅ ⎜1 + × 5 ⋅ 082 ⎟
⎝ 68 ⋅ 1 × 165 ⎠ ⎝ 2 ⋅ 97 ⎠
= 77.194 N/mm2
< 165 N/mm2. Hence satisfactory
Provide MC 125, @ 0.131 kN/m.
Step 4. Local effects
It is to note that the local effects are for the design of fixtures and not for the
design of trusses and even for the purlins. The net wind pressure at the ridge
due to local effects as the wind as parallel to the ridge, from Example 9.11
Ww = – (1.1344 + 0.5) pd . A
= – 1.6344 . pd . A
⎛ 1 ⋅ 6344 ⎞
This is ⎜ ⎟ = 1.362 times that on the normal purlin.
⎝ 1 ⋅ 20 ⎠
Example 9.13 Design an angle iron purlin for a trussed roof from the following
data :
Span of roof truss = 12 m
Spacing of roof trusses = 5m
Spacing of purlins along the slope of roof truss = 1.2 m
Slope of roof trass = 1 vertical to 2 horizontal
Wind load on roof surface normal to roof = 1.04 kN/m2
Vertical loadfrom roof sheeting etc. = 0.200 kN/m2.
Solution
Design :
Step 1: Slope of roof truss
Slope of roof truss = 1 vertical to 2 horizontal
1
∴ tan θ = , θ = 26.565° = 26° 34´ <30°
2
The angle iron purlin may be designed according to IS: 800–1984.
Step 2. Vertical load on purlin per metre length
Vertical load from roof sheeting etc. = (0.200 × 1.2 × 1) = 0.240 kN/m
Assume self-weight of angle purlin = 0.120 kN/m
Total vertical load = 0.360 kN/m
Step 3. Wind load on roof surface normal to roof
(1.04 × 1.2 × 1) = 1.248 kN/m
662 DESIGN OF STEEL STRUCTURES

Step 4: Total load (assumed to act normal to roof)

Therefore, load on purlin normal to roof
(1.248 + 0.360) = 1 608 kN/m
The maximum bending moment in purlin,
⎛ WL ⎞ ⎛ 1 ⋅ 608 × 5 × 5000 ⎞
M = ⎜ ⎟= ⎜ ⎟ = 4.02 kN-m
⎝ 10 ⎠ ⎝ 10 ⎠
The required modulus of section

⎛ 4 ⋅ 02 × 106 ⎞
Z = ⎜ ⎟ = 18.318 ×103 mm3
⎝ 1 ⋅ 33 × 165 ⎠
The depth of angle purlin
L ⎛ 5000 ⎞
= ⎜ ⎟ = 111.11 mm
45 ⎝ 45 ⎠
The width of angle purlin
L ⎛ 5000 ⎞
= ⎜ ⎟ = 83.33 mm
60 ⎝ 60 ⎠
From ISI Handbook No. 1, select ISA 125 mm × 95 mm × 6 mm. (ISA 125 95,
@ 0.129 kN/m)
Modulus of section provided
= 23.4 × 103 mm3. Hence, the section is satisfactory.
Provide ISA 125 mm × 95 mm × 6 mm (ISA 125 95, @ 0.129 kN/m) for angle
purlin for trussed roof.
Example 9.14 A 10 m × 40 m godown is to be constructed. The steel roof
trusses will be used for roofing. The trusses will be supported over masonry
walls 300 mm thick. Galvanized corrugated iron sheets will be used for cover-
ing. Propose a suitable type of roof truss.
The basic wind pressure is 1.04 kN/m2, and there is no snowfall. Determine
the load at each point.
Solution
Design :
Step 1. Preliminary geometry of truss
Clear spacing between the masonry walls = 10 m
Thickness of masonry = 0.30 m
Centre to centre spacing between the masonry walls = 10.30 m
Therefore, span of roof truss = 10.30 m
For galvanized corrugated iron sheet,
1
Provide pitch of roof truss =
4
1
Rise of roof truss = × span = ⎛⎜ 1 × 10.30⎞⎟ = 2.575 m
4 ⎝4 ⎠
 DESIGN OF ROOF TRUSSES 663

1
Let θ be slope of roof truss for pith of
4
1,
tan θ = ∴ θ = 26.565° = 26° 34´
2
Length along the sloping roof
1/ 2
⎡⎛ 10.30 ⎞2 2⎤
= ⎢⎜ ⎟ + (2.575 ) ⎥ = 5.76 m
⎣⎝ 2 ⎠ ⎦
The length along the sloping roof may be divided into six equal panels.
⎛ 5.76 ⎞
Length of each panel = ⎜ ⎟ = 0.960 m ( > 1.40 m)
⎝ 6 ⎠
Therefore, for above proportions, a French roof truss as shown in Fig. 9.41 (a)
may be selected. These trusses may be provided at 4 m spacing. 11 trusses will
be provided over the entire length of godown. The purlins will be provided at
each panel point.
Step 2 : Load at each panel point is determined as under :
1. Dead load
Weight of galvanized corrugated iron sheets = 0.133 kN/m2
Weight of purlins (assumed) = 0.150 kN/m2
Weight of bracing (assumed) = 0.015 kN/m2
Self weight of roof truss
1 ⎛l ⎞ 1 ⎛ 10.30 ⎞
w = ⎜ + 5⎟ = ⋅⎜ + 5⎟
100 ⎝ 3 ⎠ 100 ⎝ 3 ⎠
= 0.084 kN/m2
Total dead load = (0.133 + 0.150 + 0.015 + 0.084)
= 0.382 kN/m2
Spacing of trasses = 4m
Panel length in plan = 0.960 cos (26° 34´) = 0.858 m
Load at each intermediate panel due to dead load, w1
(0.382 × 4 × 0.858) = 1.311 kN
⎛ w1 ⎞
Load at end panel due to dead load ⎜ ⎟ = 0.656 kN.
⎝ 2 ⎠
2. Live load
Access is not provided for roofing except for maintenance. Slope of roof truss
= 26° 34´ = 26.565°
∴ Live load = (0.750 – 0.020 × 16.565) = 0.4187 kN/m2
Load each intermediate panel due to the live load, w2
0.4187 ×4 × 0.960 = 1.61 kN
⎛ w2 ⎞
Load at each end panel due to live load ⎜ ⎟ = 0.804 kN
⎝ 2 ⎠
664 DESIGN OF STEEL STRUCTURES

S S
S S
S S
S S 2 .57 5 m
S S
S S a g tie S
θ θ
1 0.3 0 m
(a ) A fa n and fren ch roo f truss (com bine d)

W1 W1 W1
W1 W1
W1 W1
W1
W1 W1
( W 1 /2 ) ( W 1 /2 )

1 0.3 0 m
(b ) P a ne l load d ue to d ea d loa d ( W 1 )
W2 W2 W2
W2 W2
W2 W2
W2 W2
W2
( W 2 /2 ) ( W 2 /2 )

1 0.3 0 m
(c) P a n el lo ad d ue to live loa d ( W 2 )
W3 W3
W3 W3
W3 W3
W3 W3
W3 W3
( W 3 /2 ) ( W 3 /2 )

1 0.3 0 m
(d ) P a ne l load d ue to w in d lo ad ( W 3 )

Fig. 9.42 Dead live and wind loads at panel points of roof truss

3. Wind load
Slope of roof trass = 26.565°
Intensity of design wind pressure = 1.4 kN/m2
For pitched roof, the height of eave, and the width of cladding, w are also
necesary data. Let the height of plinth be 600 mm and the projection of eave
girder beyond the centre line be 400 mm. The gross width of the roof-shed.
w = (10 + 2 × 0.400)
= 10.80 m assumed
 DESIGN OF ROOF TRUSSES 665

The height of eave of the roof-shed above ground level

h = (3.18 + 0.60) = 3.78 m assumed
Ratio of height to the width of roof-shed
⎛h⎞ ⎛ 3.78 ⎞
⎜ ⎟ = ⎜⎝ ⎟ = 0.35 < 0.5
⎝w ⎠ 10.8 ⎠
(A) External air pressure coefficients
⎛h⎞
The wind pressure coefficients are to be adopted from a set ⎜ ⎟ < 0.5 and for
⎝w ⎠
slope θ = 26.565°. The wind pressure coefficients are noted from IS : 875 (Part 3)
–1987 or from Table 9.7 (External presure coefficients, Cpe for pitched roofs of
rectangular clad buildings).

Wind angle, θ
Side of pitched roof θ = 0° * θ = 9° *
Wind ward –0.1374 –0.8687 –0.7 —
Lee ward – 0.4 –1.1344 –0.7 —

(* local effects)
(B) Internal pressure coefficients. The godown shall be a building (clad-
ding) having large opening. Let the large opening be in 40 m length and direc-
tion of wind be in the direction perpendicular to this length having the opening
(i.e., as shown in Fig. 9.43).
Total wind load C pi = + 0.8 parallel to the ridge
For buildings having large, opening
(cpe – cpi) = [– 0.7 – (+ 0.8)] = – 1.5
Total wind load (let the spacing of roof trusses be 4 m)
W3 = (Cpe – Cpi). pd . A
W3 = –1.5 . Pd . A
W3 = –1.5 × 1.04 × (0.960 × 4) = 5.99 kN

⎛ W3 ⎞
⎜ ⎟ = 2.995 kN
⎝ 2 ⎠

10 m
– C pe – C pe

W in d C p i = + 0.8 40 m
W in d
C p i = +0 .8
C pe C pe

10 m
(a ) Elevatio n (b ) Plan
Fig. 9.43 Buildings with large areas of openings on one side
666 DESIGN OF STEEL STRUCTURES

The dead load, live load and wind load at each panel point of truss are shown
in Fig. 9.42 (b), (c) and (d) respectively.
Note. For the design of various members, the stresses in members can be found for
each loading either graphically or analytically. The maximum force in each member
may be found for the combination of loads whichever is severe. The members may be
designed for maximum force in them.
Example 9.15 At the apex joint A of a Fink roof truss, four members AB, AC,
AD and AE meet so that ∠BAC = ∠EAD = 30° and ∠CAD = 60°. The lengths of
members between the panel points are AB = AE = 3 metres, AC = AD = 3.4 metres
and the stresses carried are AB = AE : 170 kN compression, and AB = AD :
100 kN tension. Design the members and detail the joint A.
Solution
Design :
Step 1 : Design of compression members :
Force in AB = AE = 170 kN compression.
AB and AE are rafter members in Fink roof truss. The rafter members in a
truss are continuous struts.
Length of members AB = AE = 3 m
Effective length AB = AE = 0.85 × 3 m, (l = 2.55 m)
Let the value of yield stress for the structural steel be 260 N/mm2, and the
slenderness ratio of the compression member be 100. Therefore, the permis-
sible stress in axial compression from Table 5.1, IS: 800–1984
σbc = 82 N/mm2
Area required for the compression member
⎛ 170 × 1000 ⎞
⎜ ⎟ = 2073.17 mm2
⎝ 82 ⎠

2 IS A 90 m m x 6 0 m m 2 IS A 90 m m x 6 0 m m
x 8 mm x 8 mm

3 0° 3 0°
6 0°
B E
1 70 kN 1 70 kN
C o m p re sion C o m p re ssio n
2 IS A 50 m m ×
C 50 m m × 6 mm D
1 00 kN 1 00 kN
Ten sion Ten sion

Fig. 9.44 Design of a ridge joint

From ISI Handbook No. l, select 2 ISA 90 × mm × 60 mm × 8 mm (2 ISA 90 60,

@ 0.089 kN/m). The long legs are kept back to back. 6 mm thick gusset plate is
used for connections.
Sectional area provided = 2274 mm2
Radius of gyration rxx = 284 mm
 DESIGN OF ROOF TRUSSES 667

Radius of gyration ryy = 24.5 mm

Minimum radius of gyration rmin = 24.5 mm
Maximum slenderness ratio
⎛ l ⎞ ⎛ 2.55 × 1000 ⎞
⎜ r ⎟ = ⎜⎝ 24.5
⎟ = 104
⎠
⎝ min ⎠
From IS : 860–1984, allowable stress in axial compression
= 78.4 N/mm2
Safe force carrying capacity of each compression member
⎛ 78.4 × 2274 ⎞
⎜ ⎟ = 178.28 kN > 170 kN
⎝ 1000 ⎠
(Compression in members). Hence, safe.
Provide 2 ISA 90 mm × 60 mm × 8 mm (2 ISA 90 60, @ 0089 kN/m) for AB and
AE.
Step 2: Design of tension members
Force in members
AC = AD = 100 kN
The allowable stress in axial tension is (0.6 × 260) N/mm2
= 156 N/mm2
Net area required for each member
⎛ 100 × 1000 ⎞
⎜ ⎟ = 641.025 mm2
⎝ 156 ⎠
From IS1 Handbook No. 1, select 2 ISA 50 mm × 50 mm × 6 mm. (2 ISA 50 50,
@ 0045 kN/m). The angles are provided on both sides of a gusset plate, 16 mm
diameter rivets are used in one row.
Gross area of angle sections = 1136 mm2
Net area provided = (1136 – 2 × 23.5) = 776 mm2
Allowable axial tension in each member
⎛ 156 × 776 ⎞
⎜ ⎟ = 121.056 kN > 100 kN
⎝ 10 ⎠
(Tension in member). Hence, safe.
Provide 2 ISA 50 mm × 50 mm × 6 mm (2 ISA 50 50, @ 0.045 kN/m) for
AC and AD.
Step 3 : Design of Joint A
16 mm diameter power driven rivets are used for connections. Gross diameter
of rivet
(16 + 1.5) = 17.5 mm
Strength of rivet in double shear
⎡ π ( )2 100 ⎤
⎢⎣2 × 4 × 17.5 × 1000 ⎥⎦ = 48.l kN
668 DESIGN OF STEEL STRUCTURES

Strength of rivet in bearing

⎛ 300 ⎞
⎜17.5 × 6 × ⎟ = 31.5 kN
⎝ 1000 ⎠
Rivet value R = 31.5 kN
Number of rivets required to connect AB and AE individually
⎛ 170 ⎞
⎜ ⎟ = 5.397 (say 6)
⎝ 31.5 ⎠
Number of rivets required to connect AC and AD individually 3.17 (say 4)
The complete design of joint A is shown in Fig. 9.44.
Example 9.16 The sliding end of a roof truss rests on 450 mm brick wall
through a concrete bearing pad. The maximum normal reaction on the bearing
is 125 kN. The principal rafter is inclined at 30 ° to the main tie which is hori-
zontal. If the panel length of the principal rafter is 1.38 metre, design the rafter,
the tie and the sliding joint. Also design the anchor bolt for a pull of 7.50 kN.
Solution
Design :
Normal reaction on bearing = 125 kN
Force in principal rafter = 125 sec 60° = 250 kN
Force in member = 250 cos 30° = 216.50 kN
Step 1 : Design of compression member
Force in principal rafter = 250 kN
Let the value of yield stress for the structural steel be 260 N/mm2 and the
slenderness ratio for the compression member be 100. Therefore, the permis-
sible stress in axial compression from Table 5.1, IS: 800–1984.
σbc = 82 N/mm2
Area required for the compression member
⎛ 250 × 1000 ⎞
⎜ ⎟ = 3048.78 mm2
⎝ 82 ⎠
From ISI Handbook No. l, select 2 ISA 90 mm × 60 mm × 12 mm. (2 ISA 90 60,
@ 0.130 kN/m) 10 mm thick gusset plate is used.
Sectional area provided = 3314 mm2
Radius of gyration rxx = 27.9 mm
Minimum radius of gyration rmin = 27.0 mm
Actual length of member = 2.60 m
The principal rafter is a continuous strut.
Effective length of member = 0.85 × 1.38
= 1.173 m
⎛ 1.173 × 1000 ⎞
Maximum slenderness ratio = ⎜ ⎟ = 43.44
⎝ 27.0 ⎠
The allowable stress in axial compression in 141.9 N/mm2
 DESIGN OF ROOF TRUSSES 669

The safe force carrying capacity of member

⎛ 3314 × 141.9 ⎞
⎜ ⎟ = 470.27 kN > 250 kN
⎝ 1000 ⎠
(Actual force in member). Hence, sale.
Provide 2 ISA 90 mm × 60 mm × 12 mm (2 ISA 9060, @ 0.130 kN/m) for
principal rafter.
Step 2 : Design of tension member
Force in the member = 216.50 kN

P rincip al ra fte r
2 IS A 9 0 m m ×
6 0 m m × 12 m m

3 0°

A n ch o r bo lt Tie m e m b e r
2 0 m m d ia m e te r 2 IS A 7 5 m m ×
50 m m × 8 mm
G usset plate 1 0 m m thick

S h oe a ng le s
2 IS A 8 0 m m ×
80 m m × 8 mm
B a se p la te 10 m m

B e aring p la te 10 m m

Fig. 9.45 Design of sliding end of a roof truss

Allowable stress in axial tension = (0.6 × 260) = 156 N/mm2

Net area required for each member
⎛ 216.50 × 1000 ⎞
⎜ ⎟ = 1387.82 mm2
⎝ 156 ⎠
670 DESIGN OF STEEL STRUCTURES

From ISA Handbook No. 1, select 2 ISA 75 mm × 50 mm × 8 mm. (2 ISA 75

50, @ 0.074 kN/m).
The angles are provided on both sides of gusset plate. 22 mm diameter rivets
are used in one row for connections.
Gross area provided = 1876 mm2
Net area provided = (1876 – 2 × 8 × 23.5)
= 1516 mm2
Allowable axial tension in member
⎛ 156 × 1516 ⎞
⎜ ⎟ = 236.496 kN > 21650 kN. Hence, safe.
⎝ 1000 ⎠
Provide 2 ISA 75 mm × 50 mm × 8 mm (2 ISA 75 80, @ 0.074 kN/m) for main
tie.
Step 3 : Design of joint
22 mm diameter power driven rivets are used for connections.
Gross diameter of rivet = (22 + 1.5) = 23.5 mm
Strength of rivet in double shear
⎡ π 2 100 ⎤
⎢⎣2 × 4 (23.5) × 1000 ⎥⎦ = 86.70 kN
Strength of rivet in bearing
⎛ 300 ⎞
⎜ 23.5 × 10 × ⎟ = 70.5 kN
⎝ 1000 ⎠
Number of rivets required to connect principal rafter
⎛ 250 ⎞
⎜ ⎟ = 3.545 (Provide 5 rivets)
⎝ 70.50 ⎠

⎛ 216.50 ⎞
Number of rivets required to connect the member ⎜ ⎟
⎝ 70.50 ⎠
= 3.07 (Provide 4 rivets)
Step 4: Design of sliding joint
Maximum normal reaction of bearing
= 125 kN
Number of rivets required for connection of shoe angles
⎛ 125 ⎞
⎜ ⎟ = 1.77
⎝ 70.50 ⎠
4 rivets are provided to connect shoe angles with gusset plate. 4 rivets are
also provided to connect shoe angles with base plate. 2 ISA 80 mm × 80 mm × 8
mm, 450 mm long are used for shoe angles.
1. Bearing plate
Normal reaction = 125 kN
Length of base plate = 45 mm
 DESIGN OF ROOF TRUSSES 671

Width of bearing base plate

(80 + 80 + 10) = 170 mm
Bearing pressure on concrete bearing pad
⎛ 125 × 1000 ⎞
⎜ ⎟ = 1.634 N/mm2
⎝ 450 × 170 ⎠
< Allowable bearing pressure in concrete pad
Consider 1 mm strip of base plate. Bending moment about xx as shown in
Fig. 9.38

⎛ 1.634 × (80 − 8 )2 ⎞
M = ⎜ ⎟
⎝ 2 ⎠
= 4235.33 N-mm
Moment of resistance of the base plate

⎛ t2 ⎞
⎜185 × 1 × ⎟ = 4235.33,
⎝ 6⎠
∴ t = 11.72 mm
Therefore, the thickness of base plate required
t1 = (11.72 – 8) = 3.72 mm
Provide 6 mm thick base plate 450 mm × 170 mm × 6 mm bearing plate below
the base plate. An elliptical hole is kept on each side of shoe angles and base
plate. The base plate can slide over bearing plate.
2. Anchor bolt
Pull in anchor bolt = 7.50 kN
Allowable axial tension in anchor bolt is 0.6 × 260 N/mm2
= 156 N/mm2
Area required at the roof of thread
⎛ 7.50 × 1000 ⎞
⎜ ⎟ = 48.07 mm2
⎝ 156 ⎠
Two 20 mm nominal diameter anchor bolts are provided on each side of shoe
angles. A complete sliding joint is shown in Fig. 9.43.

PROBLEMS
9.1. Locate the principal axes of ISA 90 mm × 60 mm × 10 mm. [ISA 9060, @
0.110 kN/m]. Determine the principal moment of inertias and radii of
gyration about the principal axes for the angle section. Compare these
values with those given in ISI Handbook No. 1. Other geometrical properties
may be noted from handbook.
9.2. In Problem 9.1, the angle section is subjected to bending moment 2 kN-m
acting in the vertical plane through the centroid of section. Determine the
maximum bending stress induced in the section.
672 DESIGN OF STEEL STRUCTURES

9.3. An MC, 350. @ 0.421 kN/m is subjected to bending moment 8 kN-m. The
plane of loading is inclined 10° with the YY-axis in anti-clockwise direc-
tion, and it passes through the centroid of channel section. Determine the
maximum bending stress induced in the section. Locate the neutral axis.
9.4. In Problem 9.3, the channel section is carrying uniformly distributed load
5 kN/m. The effective span is 5 m. The section is simply supported at both
ends. The plane of loading is inclined 10° with YY-axis and it passes through
the centroid of beam section. Determine the magnitude and direction of
maximum deflection. Take E = 2.04 × 105 N/mm2.
9.5. Draw Z-polygon for MC, 350 @ 0421 kN/m. Determine the maximum bend-
ing stress induced in the section for data given in Problem 9.3. Also deter-
mine the absolute maximum bending stress for the given moments and
corresponding inclition of plane of loading with YY-axis.
9.6. It is proposed to construct a shed in Jodhpur. The average height of roof
above ground level is 12 m. Determine the wind load for the following :
(i) When degree of permeability is small. (ii) When shed is having normal
permeability. (iii) When shed is having large opening.
9.7. A gable with corrugated asbestos roofing is supported on steel trusses of
span 9 m, rise 2.25 m, spaced at 3 m c/c. The dead load is l.220 kN/m2
normal to roof. Taking purlins one at each end of the slope and the others
equally spaced, design an unequal angle purlin with long leg perpendicu-
lar to the slope of roof.
9.8. Design an angle iron purlin for a roof with the following data :
Span of truss = 8 m
Spacing of trusses = 4 m c/c
Pitch of truss = 1/4
Spacing of purlins along the slope of truss = 120 m c/c
Roof covering = Asbestos sheets
Basic wind pressure = 1.04 kN/m2.
9.9. A 10 m × 50 in godown is to be coverted by roof trusses. Propose a suitable
type of roof truss for the purpose and design the various members of the
truss.
9.10. The forces in the members of the roof truss at joint A , AB compression 500
kN and AD tension 600 kN. Design the members AB and AD and the joint
A as shown in Fig. P.9.10. Sketch the joint A including shoe detail.

B
m
1 .3 6
9 0°
3 0°
A
D
Fig. P 9.10
 DESIGN OF ROOF TRUSSES 673

9.11. In steel roof truss, the vertical reaction at the left hand support is 500 kN.
The principal rafter at this support is inclined at an angle of 30° to the
horizontal. Design the main tie, the principal rafter, and the riveted joint.
The panel length of the principal rafter may be assumed to be 1.28 m.
Draw a fully dimensioned neat sketch.
9.12. For a steel roof with an asbestos cement sheet covering design an I-sec-
tion purlin from the following data :
Spacing of trusses = 4 m
Spacing of purlins of principal rafter = 1.32 m
Inclination of principal rafter to the main tie = 30°
Wind pressure of roof = 1.000 kNm2.
 Chapter

10 Design of Welded Joints

10.1 INTRODUCTION
The welding is one of the methods of connecting the structural members. In the
welding, a metallic link is made between the structural members. The weld is
defined as a union between two pieces of metal at faces rendered plastic or
liquid by heat or by pressure or both. A typical weld is shown in Fig. 10.1. This
shows the various zones of the weld. In the fusion zone, the portion of the
structural members has been fused (i.e., made plastic or liquid by heat). In the
weld metal zone, the portion of structural members to be connected and filler
metal are melted or made plastic in making a weld. The portion of the structural
members to be connected and filler metal melted or made plastic is termed as
weld metal. The metal added in the welding is termed as filler metal. In the
heat-affected zone, the parent metal is metallurgically affected by the heat of
welding.
W eld zo ne
W eld m e ta l H e at affe cted
zon e zon e
W eld face Fu sio n zo ne

Fig. 10.1 Various zones of a typical weld

The welding process in which the weld is made between metals in a state of
fusion without hammering or pressure is known as fusion welding. The welding
process in which the weld is made by sustained pressure while the surfaces to
 DESIGN OF WELDED JOINTS 675

be united are plastic is known as pressure welding. The welding is done by

several processes. These processes are listed below :

10.1.1 Fusion Welding

(i) Arc Welding
(a) Metal arc welding (Shielded/or unshielded)
(b) Carbon arc welding (Shielded/or unshielded)
(ii) Gas welding
(a) Oxy-Acetylene welding
(b) Oxy-Hydrocarbon welding

10.1.2 Pressure Welding

(i) Resistance welding
(ii) Pressure welding

10.1.3 Thermit Welding

10.1.4 Union Melt Welding
In general, the process of “metal arc welding” is used for the structure steel
connections. The procedure for metal arc welding is covered by IS : 823 –1964.
In the metal arc welding, the electrodes are used. When considering electrods,
it is necessary always to bear in mind the fact that heated metal combines
chemically with oxygen and nitrogen from the atmosphere forming oxides and
nitrides, which result in brittle poor quality weld metal. For this reason, it is
essential to provide some means of preventing the atmosphere from reacting
the hot weld area. This can be done either by shrouding the area with an inert
gas or by the use of a suitable flux. The flux is fusible material used in welding
or oxygen-cutting to dissolve and facilitate removal of oxides and other
undesirable substances.
The metal arc welding was originally done with bare electrodes consisting of
a piece of wire or rod of the same composition as to the metal to be welded. For
the reasons mentioned above, the welds were of poor quality, and thus bare
electrodes are now rarely encountered. Bare wire is of course used for automatic,
welding, but in this case, the weld area is protected from the atmosphere by
powdered flux or inert gas.
Most modern electrodes are of the coated or covered type consisting of a metal
core wire surrounded by a thick coating applied by extrusion, winding or other
process. The success of welding operation depends on the composition of coating,
which is varied to suit different conditions and metals. The coating on the
electrodes play the following function :
1. To stabilise the arc and enable the use of alternating current (if
necessary)
2. To flux away any impurities present on the surface being welded.
676 DESIGN OF STEEL STRUCTURES

3. To form a slag over the weld to

(a) Protect the molten metal from the atmospheric contamination,
(b) Slow the rate of cooling of the weld and to reduce chance of
brittleness, and
(c) Provide a smooth surface by reducing the ripples caused by welding
operations.
4. To add certain constituents to the weld metal to compensate for the
loss of any volatile alloying elements or any constituent lost by oxidation.
5. To speed up the welding operation by increasing the rate of melting.
The various types of electrodes are :
(i) the normal penetration electrodes,
(ii) the deep penetration electrodes,
(iii) the automatic electrodes,
(iv) shielded arc electrode, and
(v) low hydrogen electrodes.
The electrodes are used to suit the different conditions. The electrodes have
the following types of flux covering. The numbers indicating the digit refer to
type of covering.
Digit (type of covering)
1. Having a high cellulose content.
2. Having a high content of titania and producing a fairly viscous slag.
3. Containing an appreciable amount of titania and producing a fluid slag.
4. Having a high content of oxides/or silicates of iron and manganese and
producing an inflated slag.
5. Having an high content of iron oxide and/or silicates and producing a
heavy solid slag.
6. Having a high contents of calcium carbonate and fluoride.
7. Having a high content of iron powder.
8. A covering of any type not classified above.
The code number of covered electrodes shall consist of seven essential and
one or two additional symbol, each indicating a specific characteristic of the
electrode. The essential symbol shall consist of the letter M indicating the
suitability of the electrode for metal arc welding and six digits in Hindu – Arabic
numerical each denoting the following :
(a) Type of covering material (mentioned above).
(b) Welding position in which the electrode may be used.
(c) Welding current conditions.
(d) Ultimate tensile strength of deposited metal.
(e) Percentage elongation in tensile test of disposited metal, and
(f) Impact test values of deposited metal.
The essential symbols are followed by letter P or A or both, denoting the
electrode for deep penetration, or automatic welding or for both purposes. The
size of electrode is designated by the diameter of the core wire in mm.
 DESIGN OF WELDED JOINTS 677

It became possible to provide the perfect rigid joints in the rigid frame
structures for the designers with the advent of welding. There is no loss of cross-
section of a member in a welded joint. In riveted joint, the cross-section area is
reduced due to rivet holes. The welded joints are capable to transmit axial forces,
shears and moments.

10.2 ADVANTAGES OF WELDING

The use of welding is increasing rapidly because of its several advantage over
riveting. The advantages of welding are as under:
1. There is silence in the process of welding.
When the riveting is done, the rivet head is prepared by hammering.
The hammering causes noise in the process of riveting.
2. There is safety of welding operator in the welding.
The riveting is done cold as well as hot. While hammering, rivets may
fly away and injure the persons. When the hot riveting is done, the
rivets are made red hot and handed over to place them in position.
During transactions, the red hot rivets may fall down and strike the
workers.
3. The welding may be done quickly in comparison to the riveting.
4. The welded joints have better appearance than riveted joints.
5. The welded joints are more rigid than the riveted joints.
6. The welding is more adaptable than riveting.
In the complicated structures, riveting may be extremely difficult or
not possible but the welding may be done easily. For circular steel pipes
and tube connections, riveting is practically impossible, but welding
may be adopted conveniently. The welding is also convenient for
additions in the existing structures.
7. The welded joints have more efficiency than riveted joints. The efficiency
of a joint is the ratio of strength of the joint to the strength of the solid
plate. In the welded joints, holes are not made as it is done in riveted
joints. The strength of joint even in a tension member may be obtained
equal to the strength of solid member. As such, in welded joint efficiency
of joint may be achieved equal to 100 per cent. In the riveted joint, the
efficiency of a joint in general is about 80 percent only.
8. The welding is economical in comparison to the riveting.
In the welding connecting angles, gusset plates, splice plates and rivet
head are not required. Thus, it reduces the weight of structure also. In
the riveted connections, the minimum size is governed by the diameter
of the rivet besides the requirement of force to be transmitted. Even if
a smaller size of a member is sufficient to transmit the force, but the
minimum size of member is to be used, which may provide minimum
edge distance and minimum pitch for connecting rivets. In the welding,
in absence of such restrictions, small size members may be used.
678 DESIGN OF STEEL STRUCTURES

In the welding, the cost of labour is less than the riveting. The cost of painting
and other maintenance is also less.
Overlapping of mild steel bars at joints in reinforced cement concrete
structures is not necessary, if the bars are welded.

10.3 DISADVANTAGES OF WELDING

Besides the several advantages, there are disadvantages of welding. The
disadvantages of welding are as under:
1. The members are likely to distort in the process of welding.
Uneven heating and cooling are caused due to welding. As a result of
this, the distorsion of members may take place.
2. A welded joint fails earlier than riveted joint, if the structure is under
fatigue stresses.
3. There is a greater possibility of brittle fracture in welding than the
riveting.
4. The inspection of welded joint is more difficult and more expensive
than the riveted joint.
The inspection of riveted joint may be done simply by tapping with a
hammer. For inspection of a welded joint, a stethoscope and a light
hammer is required. Some of the defects in a weld may be detected by
this. The inspection of welded joint may also be done by the magnetic
power method, radiography by means of X-rays, Gamma rays or by
trepanning.
5. A more skilled person is required in the welding than in the riveting.
6. The welded joints are over rigid.

10.4 TYPES OF WELDED JOINTS

The various types of welded joints are as follows :
1. Butt weld 4. Spot weld
2. Fillet weld 5. Seam weld
3. Slot weld and Plug weld 6. Pipe weld

10.5 BUTT WELD

A butt weld is also termed as groove weld. The butt weld is used to joint structural
members carrying direct compression or tension. It is used to make tee-joint
and butt-joint. Various types of butt joint are as under. These are named
depending upon shape of the grove made for welding.

10.5.1 Square Butt Weld

A square butt weld is a weld in the preparation which the fusion faces lie
approximately at right angles to the surfaces of the components to be joined
and are substantially parallel to one another. The square butt weld is shown in
Fig. 10.2.
 DESIGN OF WELDED JOINTS 679

(a) A square butt weld (one side)

(b) A square butt weld (both sides)

Fig. 10.2

10.5.2 Single-V Butt Weld

A single-V butt weld is a weld in the preparation of which the edges of both
components are prepared so that in the cross-section, the fusion faces form a ‘V’
as shown in Fig. 10.3.

Fig.10.3 Single-V butt weld

10.5.3 Double-V Butt Weld

A double-V butt weld is a weld in the preparation of which the edges of both
components are double bevelled so that (in cross-section, the fusion faces form
two opposing ‘V ’s as shown in Fig. 10.4.

Fig. 10.4 Double-V butt weld

10.5.4 Single-U Butt Weld

A single-U butt weld is a weld in the preparation of which the edges of both
components are prepared so that in cross-section, the fusion faces form a ‘U’ as
shown in Fig. 10.5.
680 DESIGN OF STEEL STRUCTURES

Fig. 10.5 Single-U butt weld

10.5.5 Double-U Butt Weld

A double-U butt weld is a weld in the preparation of which the edges of both
components are prepared so that in cross-section, the fusion faces form two
opposing ‘U’s’ having a common base, as shown in Fig. 10.6.

Fig. 10.6 Double -U butt weld

10.5.6 Single-J Butt Weld

A single-J butt weld is a weld in the preparation of which the edges of one
component are prepared so that in cross-section, the fusion face is in the form of
a T and the fusion face of the other component is at right angles to the surface
of the first component as shown in Fig. 10.7.

Fig. 10.7 Single-J butt weld

10.5.7 Double-J Butt Weld

A double-J butt weld is a weld in the preparation of which the edges of one
component are prepared so that in cross-section, the fusion face is in the form of
two opposing ‘J’s and the fusion face of the other component is at right angles to
the surfaces of the first component as shown in Fig. 10.8.
 DESIGN OF WELDED JOINTS 681

Fig. 10.8 Double-J butt weld

10.5.8 Single-Bevel Butt Weld

A single-bevel butt weld is a weld in the preparation of which the edge of one
component is bevelled and the fusion face of the other component is at right
angles to the surfaces of the first component as shown in Fig. 10.9.

Fig. 10.9 Single-bevel butt weld

10.5.9 Double-Bevel Butt Weld

A double-bevel butt weld is a weld in the preparation of which the edges of one
component are double bevelled and the fusion face of the other component is at
right angles to the surfaces of the first components as shown in Fig. 10.10.

Fig. 10.10 Double-bevel butt weld

682 DESIGN OF STEEL STRUCTURES

10.5.10 Specifications of the Butt Weld

(i) Size of butt weld. The size of a butt weld is specified by the effective throat
thickness.
The effective throat thickness in case of complete penetration butt weld is
taken as the thickness of thinner part joined. The effective throat thickness of T
or L butt-joints is equal to the thickness of abutting part. The double-V, double-
U, double-J and double-bevel butt welds are the examples of complete penetration
butt weld.
The effective throat thickness in case of incomplete penetration butt weld is
taken as 7/8th of the thickness of the thinner part joined. But for the purpose of
stress calculation, a required effective throat thickness not exceeding 5/8th of
the thickness of thinner part joined should be used. An incomplete penetration
butt-weld is also termed as unsealed single butt weld. Single-V, single-U, single-
J and single- bevel butt joints are the examples of incomplete penetration butt
weld. In incomplete penetration butt weld, the weld metal is not deposited
intentionally through the full thickness of the joint. The unwelded portion in
incomplete penetration butt weld welded from both sides shall not be greater
than 1/4th of the thickness of thinner part joined, and should be central in the
depth of the weld.
The unsealed butt welds V, U, J and bevel types and incomplete penetration
butt welds should not be used for highly stressed joints and joints subjected to
dynamic, repeated or alternating forces. They shall also not be subjected to a
bending moment about the longitudinal axis of the weld other than that normally
resulting from the eccentricity of the weld metal relative to the parts joined.
(ii) Effective length of butt weld. The effective length of butt weld is the
length for which the specified size (throat thickness) of the weld exists.
(iii) Effective area of butt weld. The effective area of a butt weld is taken
as the product of the effective throat thickness and the effective length of butt
weld.
(iv) Reinforcement. The extra metal deposited proud of the surface of
the parent metal as shown in Fig. 10.11, is called reinforcement. This
reinforcement is provided to provide sufficient surfaces convexity and to ensure
full effectiveness at the joint. This requires a minimum practical surface convexity
of 1.0 mm. This reinforcement should not exceed 3.0 mm. This is not considered
as part of throat thickness. This reinforcement may also be removed if a flush
surface is desired.
W eld face
R e in force m e nt

S ize Th ro at
th ickne ss

R e in force m e nt

Fig. 10.11
 DESIGN OF WELDED JOINTS 683

Intermittent butt welds are used to resist shear only. The effective length
of intermittent weld is taken not less than four times the thickness of thinner
part joined. The longitudinal space between the effective lengths of welds is
taken not more than 16 times the thickness of thinner part joined. The inter-
mittent weld is not used in position subjected to dynamic, repetitive and/or
alternate stresses.
When the structural members of unequal thickness are butt welded and
difference in thickness of members exceeds 25 per cent of the thinner part or 3.0
mm in metal are welding and 6.0 mm
Taper no t exce edin g t in 5
or more in oxy acetylene welding, the
thicker part is bevelled so that the slope
of the surface from one part to the
other is not steeper than one in five as
(a )
shown in Fig. 10.12 (a). Where this
arrangement is not practicable, the
weld metal should be built-up at the
junction with the thicker part to (b )
dimension at least 25 per cent greater
than that of the thinner part (in metal 1 .5 t m a x
are welding) as shown in Fig. 10.12
(b), alternatively, the weld metal should (c)
be built-up to the dimensions of thicker Fig. 10.12
members as shown in Fig. 10.12 (c).
In case of complete penetration butt weld, generally, design calculations are
not necessary, as these will usually provide the strength at the joint equal to
the strength of the member connected.

10.6 FILLET WELD

A fillet weld is a weld of approximately triangular cross-section joining two
surfaces approximately as right angles to each other in lap joint, or tee-joint. A
fillet weld is shown in Fig. 10.13 (a).

Fu sio n
fa ce

Toe
W eld face

R o ot Toe
Fu sio n face
L eg Fu sio n zo ne

(a ) Fillet w eld
684 DESIGN OF STEEL STRUCTURES

When the cross-section of fillet weld is 45°, isosceles triangle as shown in Fig.
10.13 (b) (i), it is known as a standard fillet weld. The standard 45° fillet welds
are generally used. When the cross-section of the fillet weld is 30° and 60° triangle
as shown in Fig. 10.13 (b) (ii), it is known as a special fillet weld.

4 5° 6 0°
L eg S h ort le g 3 0°
4 5°

L eg L on g leg

(i) S ta n dard fille t w e ld (ii) S pe cial fille t w eld

(b ) Fillet weld
Fig. 10.13

A fillet weld is termed as concave fillet weld or convex fillet weld or mitre
fillet weld depending on the weld face is concave or convex or approximately
flat as shown in Fig. 10.14, respectively.
Effective th ro at
E ffe ctive th roa t

E ffe ctive th roa t

S ize S ize S ize

(a ) C o ncave fille t w e ld (b ) C o nvex fillet w eld (c) M iter fille t w e ld

Fig. 10.14 Fillet welds

A fillet weld is termed as normal fillet weld or deep penetration fillet weld
depending upon the depth of penetration beyond the root is less than 2.4 mm or
(2.4 mm or more), respectively.

D ia g on al
fille t w eld

P P

End
Fillet
w e ld
S ide fille t w eld

Fig. 10.15 Types of fillet weld

The fillet welds are of three types as shown in Fig. 10.15.

1. Side fillet weld
 DESIGN OF WELDED JOINTS 685

2. End fillet weld

3. Diagonal fillet weld.
1. Side fillet weld is a fillet weld stressed in longitudinal shear, i.e.,
a fillet weld, the axis of which is parallel to the direction of these applied load.
It is also termed as longitudinal fillet weld.
2. End fillet weld is a fillet weld stressed in transverse shear, i.e., a
fillet weld, the axis of which is at right angles to the direction of the applied
load. It is also termed as transverse fillet weld.
3. Diagonal fillet weld is fillet weld the axis of which is inclined to
the direction of the applied load.
A fillet weld is placed on the sides or end of the base metal and it is subjected
to shear along with tension or compression and usually bending.

10.6.1 Specifications of Fillet Weld

10.6.1.1 Size of Fillet Weld
The size of normal fillet weld is specified as minimum leg length of a convex or
mitre fillet weld or 1.414 times the effective throat thickness of a concave fillet
weld. The size of deep penetration fillet weld is specified as minimum leg length
plus 2.4 mm. The length of leg is the distance from the root to the toe of a fillet
weld, measured along the fusion face as shown in Fig. 10.13 (a). The minimum
size of single run fillet weld is adopted as per Table 10.1 as recommended in IS :
816 –1969.
Table 10.1 Minimum size of first run or of a single run fillet weld

Thickness of thicker part Minimum size

Over (mm) Upto and including (mm) (mm)
— 10 3
10 20 5
20 32 6
32 50 8 First run
10 mm size of fillet

Notes 1. When the minimum size of the fillet weld is greater than the thickness of
the thinner part, the minimum size of the weld should be equal to the thickness of
thinner part. The thicker part shall be adequately preheated to prevent cracking of
the weld.
2. Where the thicker part is more than 50 mm thick, special precaution like
preheating will have to be taken.

10.6.1.2 Effective Throat Thickness

The effective throat thickness of a fillet weld is the perpendicular distance from
the root to the hypotenuse of the largest isosceles right-angled triangle that can
be inscribed within the weld cross-section as shown in Fig. 10.14. The effective
throat thickness of a fillet weld shall not be less than 3 mm and shall generally
686 DESIGN OF STEEL STRUCTURES

not exceed 0.7t and 1.0t under special circumstances where t is the thickness of
thinner part.
1
Effective throat thickness = × size of weld = 0.7 × Size of weld
2
In general, for the purpose of stress calculations, the effective throat thickness
= K . Size of weld
where, K is a constant.
The value of constant K for different angles between fusion faces is adopted
as per Table 10.2 as recommended in IS : 816 –1969.
Table 10.2 Value of K for different angles between fusion faces

Angle between fusion Value of constant K

60°–90° 0.70
91°–100° 0.65
101°–106° 0.60
107°–113° 0.55
114°–120° 0.50
A fillet weld is not used for joining parts, if the angle between fusion faces is
greater than 120° or less than 60°.

10.6.1.3 Effective Length

The effective length of the weld is the length of the weld for which the specified
size and throat thickness i.e., correctly proportioned cross-section of the weld,
exists. It is taken as the actual length minus twice the size of weld, since the
specified size and throat thickness do not exist at the ends. The effective length of
the weld is shown on the drawings. In practice the actual length of weld is made
equal to the effective length shown on the drawing plus twice the weld size.
The effective length of fillet weld should not be less than four times the size
of the weld.
When the ends are returned, as shown in Fig. 10.16 then the ends should be
carried continuous around the corners for distance not less than twice the size
of weld. This should be applied particularly to side and top fillet welds in tension.

R e turn e nd
P P

R e turn e nd

Fig. 10.16

10.6.1.4 Effective Area

The effective area of a fillet weld is effective is taken as the product of effective
length and effective throat thickness.
 DESIGN OF WELDED JOINTS 687

10.6.2 Intermittent Fillet Welds

The intermittent fillet welds may be used to transfer calculated stress across a
joint when the strength required is less than that developed by continuous fillet
weld of the smallest allowable size for the thickness of the parts joined. Any
section of intermittent fillet welding shall have an effective length of not less
than four times the weld size with a minimum of 40 mm except for plate girder
stiffeners.
The clear spacing between the effective lengths of intermittent fillet welds
carrying calculated stress shall not exceed the following number of times the
thickness of the thinner plate joined and shall in no case be more than 200 mm.
(i) 12 times for compression, and
(ii) 16 times for tension.
The longitudinal fillet welds at the ends of built-up members shall have an
effective length of not less than the width of the component part joined unless
end transverse welds are used, in which case, the sum of the end longitudinal
and end transverse welds shall be not less than twice the width of the component
part.
The chain intermittent welding is to be preferred to staggered intermittent
weld. Where the staggered intermittent welding is used, the ends of the
component parts shall be welded on both the sides.
In a line of intermittent fillet welds, the welding should extend to the ends of
parts connected for welds staggered about two edges, this applies generally to
both the edges.
The intermittent fillet welds are not used in the case of main members of
structures directly exposed to weather.

10.7 SLOT WELD AND PLUG WELD

The slot weld is the weld used to join two touching contiguous components by a
fillet weld round the periphery of a slot in one component so as to join it to the
surface of the other component exposed through slot as shown in Fig. 10.17.

S lot w eld

S lot w eld

Fig. 10.17 Slot weld

688 DESIGN OF STEEL STRUCTURES

The plug weld is the weld made by filling in a hole in one part with filler
metal so as to join it to the surface of contiguous part exposed through a hole as
shown in Fig. 10.18.
In case where there is insufficient space to accommodate the required length
of fillet weld, plug weld or slot weld is used, to provide extra strength. These are
also used in addition to the fillet weld when plates are welded. These welds
balance the stresses in the plates and avoid buckling.

P lug w e ld s

Fig. 10.18 Plug welds

Specifications
The dimensions of slot or hole should comply with the following limits in terms
of the thickness of the part in which the slot or hole is made :
(i) The width or diameter should not be less than three times the thickness
or 25 mm whichever is greater.
(ii) The corners at the enclosed ends of slots should be rounded with a
radius not less than 1 12 times the thickness or 12 mm whichever is
greater.
(iii) The distance between the edge of the part and the edge of the slot or
hole or between the adjacent slot or holes should not be less than twice
the thickness and not less than 25 mm for holes.
(iv) The effective area of plug weld shall be considered as the nominal area
of the hole in the plane of facing surface. The plug welds shall not be
designed to carry stress.

10.7.1 Slot Welding

The spot welding is done by the process of resistance welding. In a resistance
welding process, a weld is produced by the heat obtained from resistance to the
flow of electric current through the work parts held together under pressure by
electrodes. The size and shape of the individually formed welds are limited
primarily by the size and contour of the electrodes.

10.7.2 Seam Welding

The seam welding is also done by the process of resistance welding. The welding
in which the pressure is applied continuously, and current impulsively to produce
 DESIGN OF WELDED JOINTS 689

a linear weld, the work piece being between two electrodes wheels or between
an electrode wheel and electrode bar. The electrode wheels apply the pressure
and may be rotated continuously or stopped during the passage of the current.

10.7.3 Pipe Welding

The pipe welding is used in pipe manufacture, in pipe fabrication and in pipe
joint. The steel pipe is manufactured by five different processes (viz., butt weld,
lap weld, electric weld, hammer weld and seamless weld).

10.8 IMPERFECTIONS IN WELD

The welding should be done carefully so that there are no defects or imperfections
in it. The strength of an imperfect or a defective weld is less than that of a
perfect weld. Imperfect weld does not attain the strength equal to that of a
properly designed weld. Imperfect weld remains weak and unsafe. Therefore,
necessary precautions should be taken to avoid imperfections in welding. The
various imperfections in welding may be classified in two categories:
1. Surface imperfections in the weld or adjacent metal.
2. Internal imperfections in the weld or adjacent metal.

10.8.1 Surface Imperfections in Weld

The surface imperfections are generally removed in final dressing of the surface
of weld. Sometimes, surface imperfections cannot be eliminated completely. The
following are main surface imperfections in the weld or adjacent metal:
1. Edge of the plate melted off 2. Exposed inclusion
3. Exposed porosity 4. Incompletely filled groove
5. Overlap 6. Undercut

10.8.1.1 Edge of the Plate Melted Off

This type of imperfection occurs in a fillet weld, the free edge of plate may be
melted off as shown in Fig. 10.19. This may happen because the free edge may
not be sufficiently built-up.

Fig. 10.19 Edge of plate melted off

10.8.1.2 Exposed Inclusion

A slag or any other foreign matter may be entrapped at the surface. This type of
defect is called exposed inclusion.
690 DESIGN OF STEEL STRUCTURES

10.8.1.3 Exposed Porosity

A group of gas pores may remain at the surface of weld. These gas pores may
not be completely removed while final dressing of the surface of weld.

10.8.1.4 Incompletely Filled Groove

This type of defect occurs in butt weld. A continuous or an intermittent channel
may remain in the face of a butt-weld as shown in Fig. 10.20. As a result of this,
the thickness of throat of the weld remains less than that of the parental metal.

Fig. 10.20 Incompletely filled groove

10.8.1.5 Overlap
When the overflow of weld metal takes place over the surface of weld and fusion
of the parent metal does occur, then, this defect is caused at toe or root of the
weld as shown in Fig. 10.21.

O verlap O verlap

Fig. 10.21 Overlap

10.7.1.6 Undercut
A groove may form in the base metal along the toe of a weld because of excess
heat of arc in the metal arc welding. The groove may remain unfilled by weld
metal as shown in Fig. 10.22. This defect is known as undercut. This defect may
be removed in final dressing of the surface of the weld by depositing weld metal.

U n de r cut U n de r cut

Fig. 10.22 Under cut

 DESIGN OF WELDED JOINTS 691

10.8.2 Internal Imperfections in Weld

The internal imperfections in the weld or adjacent metal cannot be removed.
The following are main internal imperfections in the weld :
1. Gas pore 2. Inclusion
3. Lack of fusion 4. Incomplete penetration
5. Porosity

10.8.2.1 Gas Pore

While welding, gas may be entrapped in the weld. As a result of this, a small
cavity occurs in the weld as shown in the Fig. 10.23. The cavities upto 1.6 mm in
diameter included in the defects are known as gas pore.
G as p ore

Fig. 10.23 Gas pore

10.8.2.2 Inclusion
While welding, gas or other foreign matter may be entrapped in the weld. This
effect is known as inclusion. The inclusions may appear generally in the form as
large isolated inclusions, clusters of small inclusion or line inclusion. This defect
is more irregular in shape than gas cavity.

10.8.2.3 Lack of Fusion

The lack of union in a weld is known as lack of fusion. The lack of union in a
weld may remain between weld metal and parent metal, or between parent metal
and parent metal or between weld metal and weld metal as shown in Fig. 10.24.

L ack of
fu sion

L ack of
fu sion L ack of fusio n

Fig. 10.24 Lack of fusion

The lack of fusion may occur in the following forms.

(i) Lack of side fusion. The lack of union between weld metal and parent
metal at a side of a weld outside the root is called lack of side fusion. It is shown
in Fig. 10.25.
692 DESIGN OF STEEL STRUCTURES

No
fu sion
Fig. 10.25 Lack of fusion

(ii) Lack of root fusion. The lack of union between weld metal and parent
or between adjacent faces of the parent metal at the root is called lack of root
fusion. It is shown in Fig. 10.26.

No
fu sion
Fig. 10.26 Lack of root fusion

(iii) Lack of inter-run fusion. In a multi-run weld, the lack of union between
adjacent runs of weld metals is called lack of inter-run fusion. It is shown in
Fig. 10.27.

No
fu sion

Fig. 10.27 Lack of inter-run fusion

10.8.2.4 Incomplete Penetration

In case of multi-run weld, small fissures may form by previous runs or run.
These small fissures may usually occur at the toes of the underlying runs or
run.
These small fissures may remain unfilled. At such positions, a gap may occur.
This defect is known as complete inter-run penetration. It is shown in Fig. 10.28.

N o pe ne tra tion

Fig. 10.28

Sometimes, at the root of a groove, the gap or groove angle is insufficient. As

a result of this, weld metal fails to fill root and a gap is left. Such defect is
known as incomplete roof penetration.

10.8.2.5 Porosity
A group of gas pores may occur in a weld. Such defect is known as porosity in
weld. Depending upon size of pores, it is called fine or medium or coarse. The
porosity may occur as clusters or chains or it may be scattered.
 DESIGN OF WELDED JOINTS 693

10.9 STRESS IN WELDS

10.9.1 Butt Weld
A complete penetration butt weld provides strength equal to that of the member
itself. In general, there is no necessity to determine stress in the weld. There is
also no necessity to determine the size of weld so long as the weld metal possesses
the physical properties of the plate. In case of incomplete penetration, the stress
in direct compression or tension is given by
⎛P⎞
p = ⎜ ⎟ ...(10.1)
⎝A⎠
where, P = Pull or thrust to be transmitted by the weld
l = Effective length of butt weld
t = Effective throat thickness
Stress in binding
⎛M ⎞
σ = ⎜ ⎟× y ...(10.2)
⎝ I ⎠
where, M = Bending moment at the weld section
I = Moment of inertia of weld group
y = Distance of the fibre of weld under consideration
Stress in torsion
⎛T ⎞
τ = ⎜ ⎟×r ...(10.3)
⎝J ⎠
where, τ = Shear stress due to torsion
T = Torsion at the weld section
b = Distance to the fibre of weld under consideration
J = Torsional moment of inertia
Stress in beam shear
⎛ FAy ⎞
τs = ⎜ ⎟ ....(10.4)
⎝ I ⋅b ⎠
where, τs = Shear stress in the weld
I = Moment of inertia of weld group
b = Width of the weld at the weld section under consideration
A y = Statistical moment of area of weld above the fibre under
consideration.
The above expressions for stresses in butt weld in direct compression or
tension, or bending or torsion or in beam-shear are based upon simple analysis,
and are satisfactory for purpose of design.

10.9.2 Fillet Weld

The stress in a fillet weld is considered as shear on the throat regardless of the
direction of the applied load. In design, it is convenient to use longitudinal weld.
The distribution of stress in weld is shown in Fig. 10.29. The stresses in weld
near the ends are high.
694 DESIGN OF STEEL STRUCTURES

S tress distrib ution in w eld

Fig. 10.29

For the structural welds, the heavily coated electrodes are generally used.
The heavily coated electrodes produce more ductile weld. Because of plastic
deformation near the ends, the shear stress in the weld becomes more or less
uniform before failure. The following formulae are considered sufficient and
satisfactory for purpose of design :
⎛ P⎞
1. Direct stress formula, ⎜ σ = ⎟
⎝ A⎠

⎛ M ⎞
2. Flexure formula, ⎜ σb = × y⎟
⎝ I ⎠

⎛ ⎛T ⎞ ⎞
3. Torsion formula, ⎜ τ = ⎜ ⎟ × r⎟
⎝ ⎝J⎠ ⎠

⎛ FAy ⎞
4. Beam-shear formula τs = ⎜ ⎟
⎝ I ⋅b ⎠

10.9.3 Direct Stress Formula

It is assumed that when the weld is subjected to direct compression or tension,
the stress in weld is obtained by dividing the load by effective throat area. The
effective throat area is product of effective length and effective throat thickness
of weld. The heavily coated electrodes are used. These electrodes produce ductile
weld. The ductility of weld allows equalization of stress in such weld before
failure. Therefore, the use of direct stress formula is satisfactory for purpose of
design.

P
P

P P

Fig. 10.30 Transverse fillet weld

 DESIGN OF WELDED JOINTS 695

Figure 10.30 shows fillet welds joining two plates in lap joint. The loads acting
are eccentric by distance t. Along with direct stress, welds are subjected to
bending and shear.
Figure 10.31 shows fillet weld joining two plates at right angles. The resisting
P t
loads are eccentric by the distance on either side of the plate. Along with
2 2
direct stress, welds are also subjected to the bending and shear.

1 t
2
P

t P
1
2
P

Fig. 10.31

Figure 10.32 shows fillet welds joining two plates in lap joint. The longitudinal
fillet weld have been used.

P P

Fig. 10.32 Longitudinal fillet weld

It is common practices to treat welds shown in Figs. 10.30, 10.31 and 10.32,
stressed uniformly.
In Fig. 10.30, direct stress

⎛P⎞ ⎛ P ⎞
σ = ⎜ ⎟= ⎜ ⎟
⎝A⎠ ⎛ 1 ⎞
⎜⎜ ⋅ 2l ⋅ s ⎟
⎜ 2⎟ ⎟
⎝⎝ ⎠ ⎠
where, l is the effective length of weld and s is the size of weld In Fig. 10.31,
direct stress

⎛P⎞ ⎛ P ⎞
σ = ⎜ ⎟= ⎜⎛ 1 ⎞ ⎟
⎝A⎠ ⎜⎜ ⎟
⎜ 2⎟ ⋅ 2l ⋅ s ⎟
⎝⎝ ⎠ ⎠
In Fig. 10.32, direct stress
696 DESIGN OF STEEL STRUCTURES

⎛P⎞ ⎛ P ⎞
σ = ⎜ ⎟= ⎜ ⎟
⎝A⎠ ⎛ 1 ⎞
⎜⎜ ⋅ 2l ⋅ s ⎟
⎜ 2⎟ ⎟
⎝⎝ ⎠ ⎠
The direct stresses in all these three cases are equal in intensities. Since the
allowable stress in shear is less, it controls the design.
In Fig. 10.33, the thickness of two plates connected by transverse fillet weld
in lap joint are not equal. The strengths of weld per unit length are equal.
Therefore, the load between welds is spread between plates is in proportion to
their respective thicknesses.

s2 t1
s1
t2 P

Fig. 10.33

⎛ P ⋅ t1 ⎞
Load in plate A, P1 = ⎜ t +t ⎟
⎝ ( 1 2)⎠
where, t 1 is the thickness of plate A and t 2 is the thickness of plate B
⎛ P ⎞
Stress in weld, σ = ⎜ ⎟
⎛ 1 ⎞
⎜⎜ ⎟ ⋅ l ⋅ s ⎟
⎜ 1 ⎟
⎝⎝ 2 ⎠ ⎠

⎛ P ⋅ t1 2 ⎞
σ = ⎜⎜ ⋅ ⎟ ...(10.5)
⎝( 1 2)
t + t l ⋅ s1 ⎟⎠
wbere, l is the effective length of weld and s1 is the size of weld in plate A
⎛ P2 ⋅ t2 ⎞
Load in plate B, P2 = ⎜ t + t ⎟
⎝ ( 1 2)⎠

⎛ P2 ⎞ ⎛ P ⋅ t2 ⎞ 2
Stress in weld, σ = ⎜ ⎟ = ⎜ t + t ⎟ ⋅ l ⋅ s ...(10.6)
⎜⎜⎛ 1 ⎞ (
⎟ ⎝ 1 2 ⎠ ) 2
⎜ 2 ⎟ ⋅ l ⋅ s2 ⎟
⎝⎝ ⎠ ⎠
where, s2 is the size of weld in plate B.

10.9. 4 Flexure Formula

Figure 10.34 shows fillet weld subjected to longitudinal flexure. The throat section
is subjected to critical stress. The failure will occur at this section. The moment M,
s
is resisted by width of weld 2. and depth of weld as l. The bending stress
2
 DESIGN OF WELDED JOINTS 697

s s

+ l
M

Fig. 10.34 Longitudinal flexure

M ⎡ M ⎤ l ⎡ 6M ⎤
σ= ⋅y = ⎢ ⎥⋅ 2 = ⎢ 1 2 ⎥ ...(10.7)
I 1 ⎛ s 3⎞
⋅ l 2s ⎥
⎢ ⎜2⋅ ⋅ (l ) ⎟ ⎥ ⎢
⎢⎣ 12 ⎝ 2 ⎠ ⎥⎦ ⎣⎢ 2 ⎦⎥
Figure 10.35 shows fillet welds subjected to cross flexure. The foot of weld is
subjected to maximum shear stress and maximum tensile stress.
The fillet weld subjected to cross flexure may be analysed by steel theory.
The moment M, is resisted by a couple C (h + s) or T (h + s). The top weld is
subjected to tension T, while bottom weld is subjected to thrust C, and both
forces are equal.

T s

h +
m

Fig. 10.35 Cross flexure

⎛ M ⎞ ⎛ M ⎞
Force in weld, C = ⎜ ⎟ and T = ⎜ ⎟
⎝ (h + s ) ⎠ ⎝ (h + s ) ⎠
where, (h + s) is the lever arm.
⎛ Force ⎞ ⎛ M ⎞
Sttess in weld f = ⎜ ⎟= ⎜ ⎟ ...(10.8)
⎝ Throat Area ⎠ ⎜ ( ) ⎛ s ⎞ ⎟
⎜ h + s ⋅⎜ 2 ⎟⋅l ⎟
⎝ ⎝ ⎠ ⎠
where, l is the effective length of weld and s is the size of weld.

10.9.5 Torsion Formula

Figure 10.36 shows a circular pipe connected to a plate by a fillet weld. The
weld is subjected to torsion, T. The throat of weld is subjected to critical stress.
698 DESIGN OF STEEL STRUCTURES

⎛T ⎞
The shear stress due to torsion in weld, x is equal to ⎜ ⎟ ⋅ r
⎝J ⎠

Fig. 10.36 Torsion of circular fillet weld

2
⎛ s ⎞⎛ d ⎞ ⎛ π ⋅ s ⋅ d3 ⎞ d
J = πd ⎜ ⎟ ⎜ ⎟ = ⎜ ⎟ (3 r = )
⎝ 2 ⎠⎝ 2 ⎠ ⎝ 4 2 ⎠ 2

⎛ ⎛d⎞ ⎞
⎜ T ⋅ ⎜⎝ ⎟⎠ ⎟ ⎛
2 ⎟ 2 2T ⎞
∴ τ = ⎜ 3
= ⎜ ⎟ ...(10.9)
⎜ ⎛ π ⋅ s ⋅ d ⎞ ⎟ ⎝ π ⋅ s ⋅ d2 ⎠
⎜⎜ ⎟⎟
⎝⎝ 4 2 ⎠⎠
It is seen that the torsional formula is exact for fillet weld of circular section.
Figure 10.37 shows a bar of rectangular section connected to a plate by fillet
welds. The fillet welds are short and are widely spaced. The welds are subjected
to torsion, T. The torsion is resisted by a resisting couple.
Force in each weld
⎛ Torsion ⎞ ⎛T ⎞
= ⎜ ⎟ = ⎜ ⎟
⎝ Lever Arm ⎠ ⎝L⎠

Fig. 10.37 Torsion of widely spaced short fillet weld

Shear stress in weld

T ⎛ 1 ⎞ ⎛ T 2 ⎞
τs = = ⎜ ⎟ ...(10.10)
L ⎛ s ⎞ ⎟ ⎝ L ⋅ s ⋅b ⎠
⎜
⎜⎜ ⋅
⎟ ⎟b ⎟
⎜
⎝⎝ 2 ⎠ ⎠
 DESIGN OF WELDED JOINTS 699

Figure 10.38 shows a plate of rectangular cross-section. It is connected to

another plate by two adjacent long and narrow fillet welds. The welds are
subjected to torsion T. The polar moment of inertia J is approximately equal to
moment of inertia Ixx. The distance from centre of rotation (e.g., of weld group)
to the extreme fibre of weld is approximately equal to half the depth of weld.

h
2

b
2

Fig. 10.38 Torsion of adjacent fillet weld

Therefore, shear stress in such weld

T ⎛h⎞ ⎡ T ⎤ h ⎛ 3 2T ⎞
fs = ⋅⎜ ⎟ = ⎢ ⋅ ⎜ ⎟ ...(10.11)
I xx ⎝ 2 ⎠ 1 s ⎞ 3 ⎥ 2 ⎝ s ⋅ h2 ⎠
⎢ ⋅ 2 ⎛⎜ ⎟ h ⎥
⎢⎣ 12 ⎝ 2 ⎠ ⎥⎦
It is seen that the torsional formula for long and narrow fillet weld becomes
equivalent to the flexural formula.

10.9.6 Beam Shear Formula

Figure 10.39 shows a rectangular bar connected to a plate by fillet weld. The
bar carries vertical load at its free end. The weld is subjected to shear stress and
bending stress. The distribution of shear stress is parabolic. The shear stress is
maximum at the neutral axis and it is 1.5 times the average shear stress.
P

Tv . m ax

Sh ea r stre ss
d istrib utio n
Fig. 10.39

⎛ FAy ⎞
The shear stress, τs = ⎜ ⎟
⎝ I ⋅b ⎠
Statical moment of area of the weld above neutral axis about xx-axis
⎛ s ⎞ ⎛ h ⎞ ⎛ h ⎞ s ⋅ h2
A⋅ y = 2 × ⎜ ⎟ ×⎜ ⎟×⎜ ⎟=
⎝ 2⎠ ⎝2⎠ ⎝4⎠ 4 2
700 DESIGN OF STEEL STRUCTURES

1 ⎛ s ⎞( 3 ) 1
Ixx = ⋅ 2⎜ ⎟ h s ⋅ h3
12 ⎝ 2 ⎠ 6 2
⎛ 2s ⎞
Width of weld at xx = ⎜ ⎟
⎝ 2⎠

⎡ sh2 ⎤
⎢ F⋅ ⎥ 3
4 2 F
τs(max) = ⎢ ⎥= ×
⎢ 1 ⎛ 2s ⎞ ⎥ 2 ⎛ 2s ⋅ h ⎞
⎢ sh3 ⎜ ⎟⎥ ⎜ ⎟
⎣6 2 ⎝ 2 ⎠⎦ ⎝ 2 ⎠
= 1.5 fs (average)

10.10 COMBINED STRESSES IN WELDS

10.10.1 Butt Weld
A structural material may be subjected to any possible combined stress. When
two structural members are connected by butt weld, it takes the place of
equivalent amount of base metal. Therefore, the butt weld may also be subjected
to any possible combined stress. In butt weld, the allowable stresses for
compression, tension and shear are different. Therefore, maximum values of
each stress are found to determine the critical stress. When butt weld is subjected
to tensile or compressive stress (axial and/or bending) along with direct shear
stress, the welds are proportioned so that quantity.
2 2
⎡ ps ⎤ ⎡ pt c ⎤
⎢ p ⎥ + ⎢ P ⎥ does not exceed unity.
⎣ s⎦ ⎣⎢ t c ⎥⎦
where, ps = Actual shear stress in the weld
Ps = Maximum permissible shear stress in the weld
pt c = Actual tensile/compressive stress (Axial or bending) in the weld

Pt c = Maximum permissible tensile/compressive stress (axial or

bending) in the weld
When the butt weld is subjected to following combined stresses, then, the
combined stress shall not exceed allowable stress for the parent metal or
equivalent stress, σe cal which shall not exceed 110 N/mm2.
(A) For shear combined with bending compressive or tensile.
1
σe.cal = ⎡⎣σ 2bt⋅cal + 3τ 2vm.cal ⎤⎦ 2 ...(10.12 a)

1
or σe.cal = ⎡⎣σ 2bc⋅cal + 3τ 2vm.cal ⎤⎦ 2 ...(10.12 b)
 DESIGN OF WELDED JOINTS 701

(B) For the combined bearing, bending and shear stresses

1
σe.cal = ⎡σ 2bt⋅cal + σ 2p⋅cal + σbt⋅cal σ p⋅cal + 3τv2mcal ⎤ 2 ...(10.12 c)
⎣ ⎦
1
or σe.cal = ⎡σ2bc⋅cal + σ2p⋅cal + σbc⋅cal σ p⋅cal + 3τv2mcal ⎤ 2 ...(10.12 d)
⎣ ⎦
where, σbc.cal; τvmcal and τvmcal , are the numerical values of the co-existing
bending (tensile or compressive), shear and bearing stress, respectively. σe.cal
should not exceed σe.max permissible equivalent stress.

10.10.2 Fillet Weld

When a fillet weld is subjected to two load systems, which produce collinear
stress, then these are added directly to determine combined stress. When a
fillet weld is subjected to two load systems which produce stresses in two
perpendicular direction, the maximum combined stress is computed as the vector
sum and that should not exceed the maximum permissible stress in shear.
Figure 10.40 shows a rectangular bar connected to plate by fillet welds. The
fillet weld is subjected to direct stress due to load P, and bending stress due to
moment M. The two load systems produce collinear stresses. The combined stress
in weld is found by adding these stresses directly.

s s

+ P h
M

Fig. 10.40 Direct stress and flexure

⎡ h ⎤
M ⋅
⎛ P My ⎞ ⎢ p + 2 ⎥
σb(max) = ⎜ + =
⎟ ⎢ ⎥
⎝ A I xx ⎠ ⎢ 2 s h 2 ⋅ 1 ⋅ s 3⎥
h
⎢⎣ 2 2 2 ⎥⎦

⎛ P 2 3 2M ⎞
or σs(max) = ⎜ + ⎟ ... (10.13)
⎝ 2s ⋅ h s ⋅ h2 ⎠
Figure10.41 shows rectangular bar connected to a plate by fillet welds. The
weld is subjected to an eccentric load. The load system produces shear stress
and bending stress. The stresses produced are collinear. The combined stress is
obtained by two stresses
702 DESIGN OF STEEL STRUCTURES

⎛ ⎛ h⎞ h ⎞
⎜ P P ⎜L + ⎟ ⋅ ⎟
⎛ P My ⎞ ⎝ 2⎠ 2
fs(max) = ⎜ + ⎟= ⎜ s +
⎝ A I xx ⎠ ⎜⎜ 2 1 s 3 ⎟⎟
h 2× × h ⎟
⎝ 2 12 2 ⎠

s s

h h

Fig. 10.41 Cross shear and flexure

⎛ ⎛ L + h⎞ ⎞
⎜ P 2 3 2P ⋅ ⎜⎝ 2 ⎟⎠ ⎟
fs(max) = ⎜ + ⎟ ...(10.14)
⎝ 2s ⋅ h s ⋅ h2 ⎠

Figure 10.42 shows a circular bar connected to a plate by fillet weld around
the periphery of the bar. The bar is subjected to tangential load P at its free end.
The weld is subjected to shear due to torsion and cross shear. The shear stress
due to torsion is uniform around the entire periphery. At the upper end of the
bar, the two stresses are collinear and the combined stress is found by adding
them directly.

Fig. 10.42 Cross shear and torsion

Figure 10.43 shows a rectangular bar connected to a plate by fillet welds. The
bar is carrying a vertical load, P, at its free end. The weld is subjected to shear
stress and bending stress. The stresses are acting in two perpendicular directions.
The combined stress is obtained by vectorial sum of two stresses.
P
Shear stress, τs =
A
 DESIGN OF WELDED JOINTS 703

⎛ P ⎞ ⎛ P 2⎞
= =
⎜ ⎛ s ⎞ ⎟ ⎜⎝ 2h ⎟⎠
2
⎜ ⎜ h ⎟
⎝ ⎝ 2 ⎟⎠ ⎠

P
e s s

Fig. 10.43

The maximum bending stress

⎛ h ⎞
P ⋅e⋅
⎛ My ⎞ ⎜ 2 ⎟
σb = ⎜ ⎟=⎜ ⎟
⎝ A ⎠ ⎜ 2 ⋅ 1 ⋅ s h3 ⎟
⎜ 12 ⎟
⎝ 2 ⎠

⎛ P ⋅ e (3 2 ) ⎞
= ⎜ ⎟
⎝ s ⋅ h2 ⎠
The combined stress, σ = [τs2 + σb2]1/2
When the fillet weld is subjected to following combined stress, then the
equivalent stress, σe . cal is calculated, which should not exceed 110 N/mm2.
(A) Combined bending and shear in fillet weld
1/2
σe.cal = ⎡⎣ σ 2e⋅cal + 3τvm
2 ⎤
⋅cal ⎦ ...(10.14 a)
1/2
σe.cal = ⎡⎣σ 2bc⋅cal + 3τvm
2 ⎤
⋅cal ⎦ ...(10.14 b)

σbt⋅cal
Equations (10.14 a) and (10.14 b) are used when the stress and τvm.cal
σbc⋅cal
are mutually perpendicular to each other.
(B) Combined axial and shear in fillet weld
σe = [σ2 +1.8 τ2]1/2 ...(10.14 c)
where, σ is the axial stress in the weld, σbt, σbc and τ have been defined in (10.12
a), (10.12 b) and (10.12 c), respectively.

10.11 WORKING STRESSES IN WELDS

Working stresses in welds, when welded joints are constructed with mild steel
confoming to IS : 226–1962 as parent metal and with electrodes conforming to
IS : 814–1974 are adopted as per Table 10.3 are recommended in IS : 816–1969.
704 DESIGN OF STEEL STRUCTURES

Table 10.3 Working stresses in welds

S.No. Kind of stresses Max. permissible

values
1. Tension on section through throat of butt weld 142 N/mm2
2. Compression on section through
throat of butt weld 142 Nm2
3. Fibre stresses in bending
(a) tension 157.5 N/mm2
(b) compression 157.5 N/mm2
4 Shear on section through throat of
butt and fillet welds 110 N/mm2
5. Plug welds 110 N/mm2

Note. IS 816–1969 is yet to be revised in S.I. units. as such the values of the
maximum permissible stresses have been converted from those given in M.K.S. units.
The maximum permissible value of stresses of shear and tension are reduced
to 80 per cent of those given in Table 10.3, in case, the welding in done at site,
(field).
When the effects of wind or earthquake forces are considered, then, maximum
permissible values of stresses are increased by 25 per cent.
It is to note that maximum permissible stresses given in Table 10.3 are same
as for the parent metal (mild steel IS : 226 –1962).
Example 10.1 Two plates 16 mm thick are joined by (i) a double-U butt weld,
(ii) a-single-U butt weld. Determine the strength of the welded joint in tension in
each case. Effective length of weld is 150 mm. Allowable stress in butt weld in
tension is 142 N/mm2.
Solution
(i) In case of double-U butt weld, complete penetration of weld takes place.
Effective throat thickness of weld = 16 mm
Effective length of weld= 150 mm
Strength of single -U butt weld
⎛ 142 × 150 × 16 ⎞
⎜ ⎟ = 340.8 kN
⎝ 1000 ⎠
(ii) In case of single-U butt weld, incomplete penetration of butt weld
takes place.
5
Effective throat thickness = × 16 = 10 mm
8
Effective length of weld =150 mm
Strength of single-U butt weld
⎛ 142 × 150 × 10 ⎞
⎜ ⎟ = 213.0 kN
⎝ 1000 ⎠
 DESIGN OF WELDED JOINTS 705

Example 10.2 In a truss girder of a bridge, a tie as shown in Fig. 10.44 is

connected to the gusset plate by fillet weld. Determine the strength of the weld.
The size of the weld in the fillet weld is 6 mm.
Solution
Size of weld = 6 mm
Effective throat thickness
= 0.7 × 6 = 4.2 mm
Effective length of fillet weld
= (200 + 200 + 200) = 600 mm
Strength of fillet weld
⎛ 110 × 600 × 4 ⋅ 2 ⎞
= ⎜ ⎟ = 277.2 kN
⎝ 1000 ⎠

2 00 m m
20 0 m m

20 0 m m

2 00 m m

Fig. 10.44

Example 10.3 In Example 10.2, the pull to be transmitted by the tie is 300
kN. Determine the necessary overlap of the tie.
Solution
Size of weld = 6 mm
Effective throat thickness is 4.2 mm
The pull transmitted by the end fillet weld
⎛ 110 × 200 × 4 ⋅ 2 ⎞
⎜ ⎟ = 9.4 kN
⎝ 1000 ⎠
Let l be the necessary overlap required. The pull transmitted by the side
fillets
⎛ 2 × 110 × l × 4 ⋅ 2 ⎞
⎜ ⎟ = 0.924 l kN
⎝ 1000 ⎠
(92.4 + 0.924 . l) = 300, ∴ l = 224.7 mm
Necessary overlap of the tie is 224.7 mm
Example 10.4 The web plate of a built-up welded I-section is 200 mm × 12
mm, and the flange plates are 100 mm × 12 mm. The size of fillet weld is 6 mm.
Compute the maximum shear force that may be allowed at any section, if the
average allowable shear in the web is 0.4 fy and maximum allowable shear in
the weld is 110 N/mm2.
706 DESIGN OF STEEL STRUCTURES

Solution
Step 1: Moment of inertia of the built-up section (about xx-axis)
1
Ixx = [10 × 22.43 – 8.8 × 203] × 104 = 3508 × 104 mm4
12
Step 2: Intensity of shear stress (at the weld section)
⎛ F ⋅ Ay ⎞
τs = ⎜ I ⋅ (2t ) ⎟
⎝ xx ⎠
where, Ay = Moment of the area above the section about xx-axis
F = Shear force at the section
t = Effective throat thickness of one weld
⎛ F × 100 × 12 × (100 + 6 ) ⎞
τs = ⎜ ⎟
⎝ 3508 × 104 × 2 × 0 ⋅ 7 × 6 ⎠
∴ F = 238 kN
The average shear stress in the web is 0.4 × 250 N/mm3
1 00 m m

12 m m
2 00 m m

12 m m

12 m m

Fig. 10.45

Step 3: Allowable shear force in the web

⎛ 0 ⋅ 4 × 240 × 200 × 12 ⎞
F1 = ⎜ ⎟ = 240 kN
⎝ 1000 ⎠

10.12 DESIGN OF WELDED JOINTS SUBJECTED TO AXIAL LOAD

The complete penetration butt weld as mentioned earlier, does not require design
calculations. In case of incomplete penetration butt weld, effective throat
thickness of the weld is computed, and welding is done upto the required length.
In case of fillet weld, size of the weld is fixed keeping in view the minimum
size of the weld as per Table 10.1. IS : 816 –1969 recommends that when fillet
 DESIGN OF WELDED JOINTS 707

weld is applied to the square edge of member, the maximum size of weld should
be less than the edge thickness by at least 15 mm, as shown in Fig. 10.46. This
avoids the washing down of edges of weld. When fillet weld is applied to the
round toe of rolled steel sections, the maximum size of the weld should not
3
exceed of the thickness of the section at the toe. When fillet weld is used for
4
lap joint, then overlap of the members connected as shown in Fig. 10.46, should
not be less then five times thickness of thinner part.

1 .5 m m
P S ize o f w e ld

1 .5 m m P
O verlap < 5 t
Fig. 10.46

The strength of the fillet weld is determined per mm length for the size of the
weld adopted. The effective length of the weld is then computed for the full or
thrust to be transmitted by the weld. In case, only side fillet welds are applied,
the length of the each weld should not be less than perpendicular distance
between them, and spacing between them shall not be more than 16 times the
thinner part.
Example 10.5 Design a suitable longitudinal fillet weld to connect the plates
as shown in Fig. 10.47, and to transmit a pull equal to the full strength of thin
plate. Allowable stress in the weld is 110 N/mm2 and tensile stress in the plate
0.6fy Nlmm2. The plates are 10 mm thick.

1 95 m m

1 20 m m 1 50 m m

Fig. 10.47

Solution
Design :
Step 1 : Size of weld
The maximum size of weld required for thickness upto 19 mm is 5 mm.
The maximum size of fillet weld is limited by the thickness of the plate
i.e.,(10 –1.5) = 8.5 mm
Provide 6 mm fillet weld
708 DESIGN OF STEEL STRUCTURES

Step 2 : Pull transmitted by 1 mm weld

⎛ 110 × 1 × 0 ⋅ 7 × 6) ⎞
⎜ ⎟ = 0.462 kN
⎝ 1000 ⎠
Step 3 : Tensile strength of thin plate
⎛ 120 × 10 ⎞
⎜ 0 ⋅ 6 × 250 × ⎟ = 180 kN
⎝ 1000 ⎠
Step 4 : Necessary length of the weld
⎛ 180 ⎞
⎜ ⎟ = 389.61 mm
⎝ 0 ⋅ 462 ⎠
Provide 195 mm longitudinal weld on each side.
Check: (a) Length of the weld 195 mm is greater than perpendicular distance
12 mm between welds.
(b) Spacing between weld = 120 mm < 16 × 10 mm.
Example 10.6. Two plates 120 mm × 10 mm are overlapped and connected
together by transverse fillet weld to transmit pull equal to full strength of the
plate. Design the suitable welding. Allowable stress in the weld is 110 N/mm2.
Allowable stress in tension in the plate is 0.4 fy N/mm2.
Solution
Design :
Step 1: Size of weld
Minimum size of weld = 5 mm
Maximum size of weld = (10 –1.5) = 8.5 mm
Total length of two welds = 240 mm
Step 2: Total load transmitted by 6 mm weld
⎛ 240 × 0 ⋅ 7 × 6 ⎞
⎜110 × ⎟ = 110.88 kN
⎝ 1000 ⎠
Maximum pull that can be transmitted by the plate
⎛ 120 × 10 ⎞
⎜ 0 ⋅ 6 × 250 × ⎟ = 180 kN
⎝ 1000 ⎠
10
mm
10
mm
1 55 m m
10 m m 20 m m
30 m m
1 20 m m

20 m m
30 m m
20 m m

Fig. 10.48
 DESIGN OF WELDED JOINTS 709

To transmit the pull equal to the full strength of plate, provide additional
weld by plug weld. Provide two rectangular plug welds 30 mm × 15 mm as
shown in Fig. 10.48 which satisfies the specification. Strength of two plug welds.
⎛ 30 × 15 ⎞
⎜ 2 × 110 × ⎟ = 99 kN
⎝ 1000 ⎠
Total pull now transmitted
(110.88 + 99) = 209.88 kN >180 kN
Hence, satisfactory.
Example 10.7 A tie member consists of two MC 225, @ 0.259 kN/m. The
channels are connected to either side of a gusset plate 12 mm thick. Design the
welded joint to develop the full strength of the tie. The overlap is limited to 400 mm.
Solution
Design :
Step 1 : Properties of section
From ISI Handbook No. 1, for MC 225, @ 0.259 kN/m
Thickness of web = 6.4 mm
Thickness of flange = 12.4 mm
Sectional area = 3301 mm2
Step 2 : Tensile strength of each channel section
⎛ 0 ⋅ 6 × 250 × 3301 ⎞
⎜ ⎟ = 495.15 kN
⎝ 1000 ⎠
Provide 4 mm weld.
Step 3 : Strength of weld per mm length
⎛ 110 × 0 ⋅ 7 × 4 ⎞
⎜ ⎟ = 0.308 kN
⎝ 1000 ⎠
Step 4 : Total length of fillet weld
Total length of fillet weld necessary to connect one channel section
⎛ 495 ⋅ 15 ⎞
⎜ ⎟ = 1607.6 mm
⎝ 0 ⋅ 308 ⎠
The overlap of channels is limited to 400 mm.

4 00 m m

60 m m
20 m m
IS M C 22 5, @ 2 25
0 .25 0 kN /m 65 m m m m

Fig. 10.49
710 DESIGN OF STEEL STRUCTURES

Step 5: Arrangement of slots

The width of slots should not be less than 3 times thickness (3 × 6.4 mm).
Provide two slots 20 mm wide. The distance between edge of the slot and edge of
channel or between adjacent slots also should not be less than twice the thickness
(i.e. 2 × 6.4 mm).
Provide these distances as shown in Fig. 10.49. Let x be the length of the slot.
Total length of the weld.
(800 + 225 + 4x – 2 × 20) = 1607.6 mm
∴ x = 155.65 mm
160 mm long fillet welding is done as shown in Fig. 10.49.

10.13 DESIGN OF WELDED JOINTS FROM UNSYMMETRICAL

SECTIONS SUBJECTED TO AXIAL LOAD
In order to avoid the effect of eccentricity, the load is applied along the neutral
axis of unsymmetrical sections (e.g., angle-section, tees). When the
unsymmetrical sections are connected by welding, then, fillet welding is so
applied that centre of gravity of the welds lies on the line of action of action of
the load i.e., it coincides with the neutral axis. An angle section connected to a
gusset plate is shown in Fig. 10.50.
The angle section carries and axial pull p. Let l1 and l2 be the lengths of welds
applied at the sides. These welds resist P1 and P2 forces as shown in 10 .50. The
lengths l1 and l2 may be determined as under.
Equating the moment of forces about the line of action of P2
P1 . (a + b) = P.b
P ⋅b
∴ P1 = ( ...(10.15)
a + b)

I1
P1
a
P
b
P2
I2

Fig. 10.50

Equate the moment of forces about the line of action of Pl

P2.(a + b) = P . a
P ⋅b
∴ P2 = ( ...(10.16)
a + b)
Let τvmbe the maximum permissible stress in the fillet weld, and s be the size
of the weld.
 DESIGN OF WELDED JOINTS 711

The strength of weld per mm length

= (τvm× 0.7 s)
⎛ p1 ⎞ ⎛ P ⋅b ⎞
l1 = ⎜ ⎟ =⎜ ⎟ ...(10.17)
( )
⎝ τvm × 0 ⋅ 7s ⎠ ⎝ τvm × 0 ⋅ 7s a + b ⎠

⎛ p2 ⎞ ⎛ P ⋅a ⎞
l2 = ⎜ ⎟ =⎜ ⎟ ...(10.18)
τ
⎝ vm × 0 ⋅ 7 s τ
⎠ ⎝ vm × 0 ⋅ 7 s ( a + b ) ⎠
The lengths of the welds should further satisfy the specifications of the filler
weld. The total length of the weld required to apply is computed by dividing the
load transmitted by the strength of the weld per mm length. In case the lengths
available at the sides become insufficient or to make the joint more effective,
end fillet weld is also applied as shown in Fig. 10.51. In such cases, the lengths
of welds l1 and l2 to be applied at the sides are computed for the load to be
transmitted less the strength of end fillet weld.

l1

l2

Fig. 10.51

Example 10.8 A tie member of a roof truss consists of 2 ISA 90 mm × 60 mm

× 10 mm. The tie member is subjected to put of 250 UN. The angles are connected
either side of a gusset plate 10 mm thick. Design the welded connection.
Solution
Design:
Step 1: Size of weld
3
The size of weld should not exceed th thickness of the rolled steel section
4
3
at the toe i.e., × 10 = (7.5 mm)
4
Step 2: Lengths of welds
The fillet weld of 6 mm size is provided on both the sides. Each angle section
carries a pull of 125 kN. From Eq. 10.17,
⎛ P ⋅ b × 1000 ⎞ ⎛ 125 × 61 ⋅ 3 × 1000 ⎞
l1 = ⎜ ⎟=⎜ ⎟ = 184.28 mm
⎝ ( a + b ) × 0 ⋅ 7 × 6 × 110 ⎠ ⎝ 90 × 4 ⋅ 2 × 110 ⎠
From Eq. 10.18,
⎛ 125 × 28 ⋅ 7 × 1000 ⎞
l2 = ⎜ ⎟ = 86.279 mm
⎝ 90 × 0 ⋅ 7 × 6 × 10 ⎠
712 DESIGN OF STEEL STRUCTURES

l1

l2

= 86 .2 8 m m

Fig. 10.52

These are the effective lengths of welds. Add twice the size of weld to have
actual lengths of the welds.
The effective lengths of welds are shown in Fig. 10.52.
Example 10.9 In a roof truss, a tie members ISA 110 mm × 110 mm × 8 mm
carries a pull of 210 kN. The tie is connected with the gusset plate 8 mm thick.
Design the welded joint. Apply end fillet weld also.
Solution
Design :
Step 1 : Size of weld
3
The size of weld at the toe in case of rolled steel section should not exceed
4
3
to thickness (|> × 8 mm).
4
The maximum size of weld connecting vertical leg of the angle with gusset
(8 – 1.5) = 6.5 mm
Provide a uniform size of 6 mm for the weld.
Distance of neutral axis is 30 mm from top.
Provide an end fillet weld of 60 mm effective length, symmetrical to the neutral
axis.
Step 2 : Strength of end fillet weld
⎛ 110 × 0 ⋅ 7 × 6 × 60 ⎞
⎜ ⎟ = 27.72 kN
⎝ 1000 ⎠
Step 3 : Pull resisted by side welds
(210 – 27.72) = 182.28 kN
Step 4 : Length of welds
FromEq. 10.17,
⎛ 182 ⋅ 28 × 80 × 1000 ⎞
l1 = ⎜ ⎟ = 286.94 mm
⎝ 100 × 0 ⋅ 7 × 6 × 110 ⎠
From Eq. 10.18
⎛ 182 ⋅ 28 × 30 × 1000 ⎞
l2 = ⎜ ⎟ = 107.5 mm
⎝ 100 × 0 ⋅ 7 × 6 × 110 ⎠
 DESIGN OF WELDED JOINTS 713

30
mm

80
mm

Fig. 10.53

The effective lengths of side welds are shown in Fig. 10.53.

Example 10.10 A joint in a weld truss is as shown in Fig. 10.54. Design the
welded connections.
Solution
Design :
Step 1 : Size of weld
Force in member = 50 kN
3
Size of weld should not exceed th thickness of rolled steel-section at the toe.
4
Thickness of angle section 8 mm
3
∴ Size of weld |> × 8 mm |> 6 mm
4
Fillet weld of 6 mm size is provided as shown in Fig. 10.54. From ISI Handbook
No. 1, for ISA 75 mm × 50 mm × 8 mm (ISA 7550, @ 0.074 kN/m)
a = 25.2 mm, and b = 49.8 mm
(a + b) = 75.0 mm
Step 2 : Effective length of weld, from Eq. 10.17
⎛ P ⋅b ⎞ ⎛ 50 × 1000 × 49 ⋅ 8 ⎞
l2 = ⎜ τ × 0 ⋅ 7 s ( a + b ) ⎟ = ⎜ ⎟ = 71 .86 mm
⎝ vm ⎠ ⎝ 110 × 0 ⋅ 7 × 6 × 75 ⎠
Effective length of weld from Fig. 10.18.
⎛ P ⋅b ⎞ ⎛ 50 × 1000 × 25 ⋅ 2 ⎞
l2 = ⎜ τ × 0 ⋅ 7 s ( a + b ) ⎟ = ⎜ ⎟ = 71.86 mm
⎝ vm ⎠ ⎝ 110 × 0 ⋅ 7 × 6 × 75 ⎠
Step 3 : Actual length of welds
l1´ = (71.86 + 2 × 6) = 83.86 mm
l2´ = (36.36 + 2 × 6) = 48.36 mm
Effective lengths of weld are shown in Fig. 10.54.
Force in member = 80 kN
The thickness of angle section is 8 mm
3
Size of weld |
> 8 |> 6 mm.
4
714 DESIGN OF STEEL STRUCTURES

6 0 kN
5 0 kN

5 9.0 9 m m
7 1.8 6 m m IS A 6 0 m m × 5 0 m m × 8 m m

IS A 7 5 m m × 5 0 m m × 8 m m G usset 10 m m thick
11 4.0 7 m m
3 6.3 6 m m
1 2.3 5 m m 1 2.3 5 m m
6 0°
6 0° 6 0°
1 00 kN 80 m m 3 5 kN

2 2.8 2 m m 2 2.8 2 m m
Fig. 10.54

The fillet weld of 6 mm size is provided as shown in Fig. 10.54.

From ISI Handbook No. 1, for ISA 80 mm × 50 mm × 8 mm (ISA 8050, @ 0.077
kN/m)
a = 27.2 mm and b = 52.7 mm
(a + b) = 80.0 mm
Step 4 : Effective length of weld, from Eq. 10.17,
⎛ P ⋅b ⎞ ⎛ 80 × 1000 × 52 ⋅ 7 ⎞
l1 = ⎜ ⎟ = ⎜⎝ 110 × 0 ⋅ 7 × 6 × 80 ⎟⎠ = 114.07 mm
τ
⎝ vm × 0 ⋅ 7 s ( a + b ) ⎠
Effective length of weld, from Eq. 10.18,
⎛ P ⋅b ⎞ ⎛ 80 × 1000 × 27 ⋅ 3 ⎞
l2 = ⎜ τ × 0 ⋅ 7 s ( a + b ) ⎟ = ⎜ ⎟ = 59.09 mm
⎝ vm ⎠ ⎝ 110 × 0 ⋅ 7 × 6 × 80 ⎠
Step 5 : Actual length of welds
l1´ = (114.86 + 2 × 6) = 126.07 mm
l2´ = (59.09 + 2 × 6) = 71.09 mm
The effective lengths of welds are shown in Fig. 10.54.
The horizontal member is a continuous tie member. It carries pull of 100 kN
on one side and 35 kN on the other side. The weld is designed for difference of
forces i.e., for (100 – 35) = 65 kN
There are two angles sections. The welding is done on both angle sections.
Force in weld on one face
1
×65 = 32.5 kN
2
From ISI Handbook No 1, for ISA 80 mm × 50 mm × 10 mm (ISA 8050, @
0.094 kN/m)
a = 28.1 mm and b = 51.9 mm
(a + b) = 80 mm
 DESIGN OF WELDED JOINTS 715

3
Size of weld |> × 10 mm |> 7.3 mm.
4
The fillet weld of 6 mm size is provided as shown in Fig. 10.54
Step 6: Effective length of weld l1 from Eq. 10.17,
⎛ P ⋅b ⎞ ⎛ 32 ⋅ 5 × 1000 × 51 ⋅ 9 ⎞
l1 = ⎜ ⎟=⎜ ⎟ = 45.627 mm
( )
⎝ τvm × 0 ⋅ 7s a + b ⎠ ⎝ 110 × 0 ⋅ 7 × 6 × 80 ⎠
1
l = 22.82 mm
2 2
Effective length of weld l2 from Eq. 10.18,
⎛ P ⋅b ⎞ ⎛ 32 ⋅ 5 × 1000 × 28 ⋅ 1 ⎞
l2 = ⎜ ⎟ = ⎜⎝ 110 × 0 ⋅ 7 × 6 × 80 ⎟⎠ = 24.71 mm
τ
⎝ vm × 0 ⋅ 7 s ( a + b ) ⎠
1
l2 = 12.35 mm
2
Step 7: Actual length of welds
l1´= (45637 + 2 × 6) = 57.637 mm
l2´= (24.71 +2 × 6) = 36.71 mm
Half-length of welds are placed symmetrically as shown in Fig. 10.54.

10.14 DESIGN OF WELDED JOINTS SUBJECTED TO ECCENTRIC

LOAD
When the C.G. of group of the weld does not lie on the line of action of applied
load, then the welded joint is subjected to eccentric load. Similar to the riveted
bracket connections, the welded joints subjected to eccentric load are also of two
types as under:
1. The C.G. of group of the weld lies in the plane of line of action of the
applied load.
2. The C.G. of group of the weld does not lie in the plane of line of action
of the applied load.

Y P
e
b

d
x θ x
G

x
b
Y

Fig. 10.55
716 DESIGN OF STEEL STRUCTURES

Type 1. Figure 10.55 shows a fillet welded bracket connection. A vertical

load, P is acting at a distance e from the C.G. of the group of the weld.
The eccentric load P is considered equivalent to direct load P passing through
C.G. of the group of the weld and twisting moment ‘P × e’.
Let a weld of uniform size be applied throughout, and t be the effective throat
thickness.
The direct shear stress in the weld
⎛ P ⎞
ps = ⎜ ⎟
⎝ ( 2b + d ) × t ⎠
The stress in the weld due to twisting moment is maximum in the weld at the
extreme distance from the C.G. of group for the weld and acts in the direction
perpendicular to the radius vector. The maximum stress due to twisting
⎛ r ⎞
pb = ⎜ P ⋅ e ⋅
⎝ I zz ⎟⎠
where, r = Distance to the extreme weld from the C.G., of weld group
Izz = Polar moment of inertia
Izz = (Ixx + Iyy)
The vector sum of two stresses is given by
p = [(ps)2 + (ps)2 + 2 ps ps cos θ]1/2 ...(10.19)
The resulting stress p, is not to exceed the maximum permissible stress in
the weld. In this type of eccentric welded joint, the stresses ps and pb are not
normal to each other. As such, the combined stress is determined by using Eq.
(10.19).
The thickness of gusset plate (bracket plate) should not be less than the size
of the fillet weld. It should also be thicker than the thickness of flange of the
column section.
Type 2. In this type C.G. of group of weld does not lie in the plane of action of
the applied load. In such cases, butt weld can be applied or filled weld can also
be used.
Figure 10.56 shows a bracket connection in which complete penetration butt
weld has been applied.

e P

Fig. 10.56
 DESIGN OF WELDED JOINTS 717

The load P is acting at an eccentric distance e. The weld is subjected to direct

shear stress as well as bending stress. Let t be the effective throat thickness of
the weld.
P
Direct shear stress, ps =
(h ⋅ t )
where, h = Effective depth of the weld.
The actual bending stress at the extreme distance of the weld
⎛M M 6 M 6P ⋅ e ⎞
pt/e = ⎜ = = = ⎟
Z 1 2 t h2 t ⋅ h2 ⎟
⎜ t ⋅h
⎝ 6 ⎠
where, pt/e = Actual tensile/compressive bending stress in the weld,
2 2
⎡p ⎤ ⎡p ⎤
For the combined stress in the butt weld, the expression ⎢ s ⎥ + ⎢ t / e ⎥ should
p
⎣ s⎦ ⎣ pt / e ⎦
not exceed unity.
For the purpose of design, the effective depth weld is computed from the
following approximate depth formula
1/2
⎡ 6M ⎤
d = ⎢ ⎥ ...(10.20)
⎣ t × Pt / e ⎦
where, P t/c is the maximum permissible tensile or compressive bending stress
in the weld.
The effective throat thickness of the weld is know, knowing the thickness of
the member transmitting the load. The butt weld so designed is checked for
combined stresses as above.
Figure 10.57 shows a bracket connection in which fillet weld has been applied
on both the sides of the plate.

P
e

Fillet w eld

Fig. 10.57

The fillet welds are subjected to direct shear stress and bending stress. Let t
be the effective throat thickness of the weld.
718 DESIGN OF STEEL STRUCTURES

The direct shear stress in the weld

⎛ P ⎞
τv = ⎜ ⎟
⎝ 2t ⋅ h ⎠
The bending stress in the weld
⎛M 6P ⋅ e ⎞
pb = ⎜ = ⎟
⎝Z 2 ⋅ t ⋅ h2 ⎠
The combined stress in the fillet weld is obtained by the vector sum of these
stresses
P = [(τV)2 + (pb)2]1/2 ...(10.21)
The resulting stress p should not exceed the maximum permissible shear
stress in the weld.
For the purpose of the design, size of the weld is fixed depending upon
thickness of the member transmitting the load. The effective throat thickness
of the weld is thereby known.
The depth of fillet weld applied on both the sides is obtained by
1/2
⎡ 6M ⎤
h = ⎢ ⎥ ...(10.22)
⎣ 2 ⋅ t ⋅ Ps ⎦
where, Ps is the maximum permissible shear stress in the weld.
In case, the depth of weld is limited, the combined stress in the weld for
shear and bending is computed assuming the effective throat thickness of the
fillet weld. The combined stress in the fillet weld is not to exceed maximum
permissible shear in the weld. Equating the combined stress in the weld with
the maximum permissible shear stress in the weld, effective throat thickness of
the weld is found. The size of the weld is computed knowing the effective throat
thickness of the weld.
Example 10.11 Design : (i) a suitable butt weld, (ii) a fillet weld for the bracket
connection as shown in Fig. 10.58.
Allowable stress on the butt weld in bending is 157.5 N/mm2.

1 20 kN
1 20 m m

1 2 m m th ick p la te

IS H B 3 00 ,
@ 0 .63 0 kN /m

Fig. 10.58
 DESIGN OF WELDED JOINTS 719

Solution
Design :
(i) Butt weld
Step 1: Depth of weld
Provide a double V butt weld. The effective throat thickness of weld is 12 mm.
Let h be the approximate depth of the weld. From Eq. 10.20,
where, M = Bending moment
t = Effective throat thickness of weld
P t/c = Allowable tensile or compressive stress in the weld in bending
1/2
⎡ 6 × 120 × 1000 × 120 ⎤
∴ h =⎢ ⎥⎦ = 213.81 mm
⎣ 12 × 157.5
Try 230 mm depth of the weld
Step 2: Stresses in welds
Direct shear stress in the weld
⎛ 120 × 1000 ⎞
τv = ⎜ ⎟ = 43.478 N/mm2
⎝ 230 × 12 ⎠
Actual bending stress
⎛ 6 × 120 × 1000 × 120 ⎞
P t/c = ⎜ ⎟ = 136.11 N/mm
2
⎝ 12 × 2302 ⎠
Check
Step 3: For combined stress in the butt weld,
2 2 2 2
⎡ ps ⎤ ⎡ pt / e ⎤ ⎡ 43.478 ⎤ ⎡136.1 ⎤
⎢p ⎥ + ⎢p ⎥ = ⎢ ⎥ +⎢ = 0.903 < 1.
⎣ s⎦ ⎣ t/e ⎦ ⎣ 110 ⎦ ⎣157.5 ⎥⎦
Hence, safe.
Step 4: Check for equivalent stress
For butt weld, the equivalent stress σe cal due to coexisting bending stress
(tension or compression) and shear stress is given by
σe .cal = [σbc.cal2 + 3τvm.cal2]1/2 = [(136.1)2 + 3 × (43.478)2]1/2
= 155.54 N/mm2
As per IS 816–1969, the equivalent stress σe.cal shall not exceed the values
allowed for the parent metal. For the parent metal (structural steel having yield
stress, fy as 250 N/mm2), the equivalent stress, σe (as per IS : 800–1984) is given
by
σe = (0.9 × 250)
= 225 N/mm2; (σe.cal |> σe).
Hence, the design is safe.
Provide 230 mm depth of double V butt weld.
(ii) Fillet weld
Step 1: Depth of weld
Apply 8 mm fillet weld on each side of plate.
720 DESIGN OF STEEL STRUCTURES

Effective throat thickness of fillet weld

(0.7 × 8) = 5.6 mm
Strength of 1 mm fillet weld
⎛ 110 × 5 ⋅ 6 × 1 ⎞
⎜ ⎟ = 0.616 kN
⎝ 1000 ⎠
Approximate depth of fillet weld applied on two faces
1
⎡ 6 × 120 × 120 ⎤ 2
h = ⎢ = 264.82 mm
⎣ 2 × 0 ⋅ 616 × 1 ⎥⎦
Try 300 mm depth for fillet welds.
Step 2: Stresses in welds
Direct shear stress in fillet weld
⎛ 120 × 1000 ⎞
τv = ⎜ ⎟ = 35.71 N/mm2
⎝ 2 × 300 × 5 ⋅ 6 ⎠
Bending stress in fillet weld
⎛ 6 × 120 × 1000 ⎞
pb = ⎜ ⎟ = 85.71 N/mm2
⎝ 2 × 5 ⋅ 6 •× 3002 ⎠
Step 3: Check for combined stress in fillet weld
p = [35.712 + 85.712]1/2
= 92.85 < 110 N/mm2.
Hence, safe.
Step 4: Check for equivalent stress
For fillet weld, the equivalent stress σe .cal due to co-existing bending stress
and shear stress is given by
1
σeca.cal = ⎡⎣σbc⋅cal 2 + 1 ⋅ 8tvm⋅cal 2 ⎤⎦ 2
= [ 85.712 + 1.8 × 35.712]1/2
= 98.19 N/mm2 |> (σe = 110 N/mm2)
Hence, the design is safe.
Provide 300 mm depth of 8 mm fillet weld.
Example 10.12 Plates have been connected with the flanges of an I-section by
applying 8 mm fillet weld as shown in Fig. 10.59. Compute the maximum load
which may be placed at a distance of 100 mm from the flanges.
Solution :
Step 1: Size of weld
Consider one face,
Size of weld = 8 mm
Effective throat thickness = (0.7 × 8)
= 5.6 mm
 DESIGN OF WELDED JOINTS 721

Step 2: Properties of welds

Let x be the distance of centroid ‘G’ of weld group from left hand edge of the
plate
(2 × 200 × 5 ⋅ 6 × 100 ) + (200 × 5 ⋅ 6 × 0 )
x =
(2 × 200 × 5 ⋅ 6 ) + (200 × ⋅6 )
= 66.7 mm
Moment of inertia of weld group about xx-axis
⎡ 1 3⎤ 4
Ixx= ⎢2 × 20 × 0 ⋅ 56 × 100 + × 0 ⋅ 56 × 20 ⎥ × 10
⎣ 2 ⎦
= 4480 × 104 mm4
Moment of inertia of weld group about yy-axis
⎡1 2 ⎤
Iyy = ⎢ × 2 × 0 ⋅ 56 × 203 + 2 × 20 × 0 ⋅ 56 (20 – 6 ⋅ 67 ) + 20 × 0 ⋅ 56 × 62 ⎥ 104 mm4
⎣2 ⎦
= 1493.2 ×104 mm4
2P

2 00 m m 1 00 m m

1 00
mm
θ
G
1 00
mm

IS H B 3 00 ,
6 6.7 1 33 .3 @ 0 .63 0 N /m
mm mm

Fig. 10.59

Polar moment of inertia of weld group

Ixx = (Ixx+Iyy) = (4480 + 1493.2)
= 5973.2 × 104 mm4
Distance to the extreme weld from the centroid of weld group
r = [1002 + 133.32]1/2
= 166.64 mm
⎛ 13 ⋅ 33 ⎞
cos θ = ⎜ ⎟
⎝ 16 ⋅ 65 ⎠
722 DESIGN OF STEEL STRUCTURES

Let 2P be the maximum load which can be placed on the bracket. Load
transmitted by each face = P
Step 3 : Stresses in welds
Direct shear stress
⎛ P ⎞ ⎛ (0 ⋅ 0297P ) ⎞
τv = ⎜ ⎟ = ⎜ ⎟ Nmm2
⎝ (2 × 200 + 200 ) × 5 ⋅ 6 ⎠ ⎝ 100 ⎠
Twisting moment resisted by the weld group
T = P × (133.3 + 100)
= 233.3 P N-mm
Maximum shear stress due to twisting
⎛ 233 ⋅ 3P × 166 ⋅ 64 ⎞ ⎛ 0.0650P ⎞
pb = ⎜ ⎟= ⎜ ⎟
⎝ 5973 ⋅ 2 × 104 ⎠ ⎝ 100 ⎠
Combined stress in the weld group, from Eq. 10.19,
2 ⎞
p = ⎛⎜
⎝  ⎟⎠

⎛ P ⎞ ⎛ 13.33 ⎞ 1/2
⎜ ⎟ ×[(0.0297)2 +(0.650)2 +2 ×0.0297×0.0650 ⎜⎝ ⎟]
⎝ 100 ⎠ 16.65 ⎠

⎛ 0 ⋅ 0905P ⎞
= ⎜ N/mm2
⎝ 100 ⎟⎠
Step 4: Check for combined stress
Combined stress should not exceed maximum permissible stress 110 N/mm2
⎛ 0 ⋅ 0905P ⎞
∴ ⎜⎝ ⎟ = 110, P = 121488 N
100 ⎠
∴ P = 121.488 kN, and 2P = 242.97 kN
Maximum load which may be placed is 242.97 kN.
Example 10.13 A bracket plate is welded to the flange of a column as shown
in Fig. 10.60. Calculate the size of the weld required.
Solution :
Step 1: Properties of welds
Let t be the effective throat thickness of the weld group, and x be the distance
of its centroid from left hand edge of the plate ,
⎛ 2 × 180 × t × 90 + 250 × t × 0 ⎞
x =
⎜ ⎟
⎝ 2 × 180 × t + 250 × t ⎠
= 53.1 mm
Moment of inertia of weld group above xx-axis
⎡ 2 1 3⎤ 4
Ixx = ⎢⎣2 × 18 × t × 12 ⋅ 5 + 2 × t × 25 ⎥⎦ × 10 mm
4

= 6942t ×104 mm4

 DESIGN OF WELDED JOINTS 723

Moment of inertia of weld group about y-axis

⎡1 2 ⎤
Iyy = ⎢ × 2 × t × 183 + 2 × 18 × t (9 – 5 ⋅ 31) + 25t × 5 ⋅ 312 ⎥ × 104
⎣2 ⎦
= 1884.5 t × l04 mm4
Polar moment of inertia of weld group
Ixx = (6942 + 1884.5)t × 103 mm4
= 8826.5t × 103 mm4
Distance to the extreme weld from the centroid of weld group
r = [(125)2 + (126.9)2]1/2
= 178.8 mm
⎛ 126 ⋅ 9 ⎞
cos θ = ⎜ ⎟ = 0:71
⎝ 178 ⋅ 8 ⎠
Step 2: Stres in welds
Direct shear stress
⎛ 100 × 1000 ⎞ ⎛ 163 ⋅ 93 ⎞
τv = ⎜ ⎟= ⎜ ⎟ N/mm2
⎝ (2 × 180 × 250 ) ⋅ t ⎠ ⎝ t ⎠

1 00 kN

2 46 .9 m m
1 80 m m 1 20 m m

2 50 m m θ
X X
G

ISH B 3 00
5 3.1 1 26 .9 @ 0 .63 0 kN /m
mm mm

Fig. 10.60

Maximum shear stress due to twisting

⎛ 100 × 1000 × 246 ⋅ 9 × 178 ⋅ 8 ⎞
pb = ⎜ ⎟ N/mm2
⎝ 8826 ⋅ 5t × 103 ⎠

⎛ 500 ⋅ 15 ⎞
= ⎜ ⎟ N/mm2
⎝ t ⎠
724 DESIGN OF STEEL STRUCTURES

Step 3: Combined stress in the weld group, from Eq. 10.19,

1
⎡⎛ 163 ⋅ 93 ⎞ 2 ⎛ 500 ⋅ 15 ⎞ 2 ⎛ 163 ⋅ 93 ⎞ ⎛ 500 ⋅ 15 ⎞ ⎛ 500 ⋅ 15 ⎞ ⎤2
P = ⎢⎜ ⎟ +⎜ ⎟ +2×⎜ ⎟ ⎜ ⎟ ⎜ ⎟ × 0 ⋅ 73 ⎥
⎣⎝ t ⎠ ⎝ t ⎠ ⎝ t ⎠⎝ t ⎠⎝ t ⎠ ⎦
627 ⋅ 25
= N/mm2
t
Combined stress in the fillet weld should not exceed 110 N/mm2.
⎛ 627 ⋅ 25 ⎞
∴ t = ⎜ ⎟ = 5.7 mm
⎝ 110 ⎠
⎛ 5⋅7 ⎞
Size of weld required = ⎜ ⎟ = 8.146 mm
⎝ 0 ⋅7 ⎠
Provide 9 mm fillet weld.
Example 10.14 Calculate the size of the weld required for the joint cutting
used as a bracket loaded as shown in Fig. 10.61. The flange welds are of double
the size of welds.
Solution :
Step 1 : Properties of welds
Let 2 t be the effective throat thickness of flange welds.
Effective throat thickness of web weld = t mm
Moment of inertia of weld group about xx-axis
1
Ixx = [2 × 12.5 × 2t × 202 + 2 × × t × 353] × 103
12
= (27166 × t ×103) mm4

1 50 kN
2 00 m m

IS H B 4 00
@ 0 .77 4 kN /m
3 50
mm

C o lu m n
IS H B 4 00 1 25
@ 0 .77 4 kN /m mm

Fig. 10.61

Step 2 : Stresses in welds

1. Direct shear stress in weld
Direct load shared by welds in the proportion to the areas of welds. Total
area of welds = (2 × 125 × 2t + 2 × 350 × t)= 1200 t mm2
 DESIGN OF WELDED JOINTS 725

⎛ 500t ⎞
Load shared by flange welds = ⎜ × 150 ⎟ = 62.50 kN
⎝ 1200t ⎠
Load shared by web weld = 87.50 kN
Shear stress in the flange weld
⎛ 62 ⋅ 50 × 1000 ⎞ ⎛ 125 ⎞
ps = ⎜ ⎟=⎜ ⎟ N/mm2
⎝ 2 × 350 × t ⎠ ⎝ t ⎠
Shear stress in the web weld
⎛ 87 ⋅ 50 × 1000 ⎞ ⎛ 125 ⎞
ps1 = ⎜ ⎟ =⎜ ⎟ N/mm2
⎝ 2 × 350 × t ⎠ ⎝ t ⎠
2. Bending stress in welds
The bending stress in the welds is in proportion to the distance from the
neutral axis of weld group.
Bending stress in the weld
⎛M ⎞
Pb = ⎜ ⎟⋅ y
⎝ I xx ⎠

⎛ 150 × 1000 × 200 × 200 ⎞

Pb = ⎜ ⎟
⎝ 27166 × t × 103 ⎠

= ⎛ 220 ⋅ 86 ⎞ N/mm2
⎜ ⎟
⎝ t ⎠
Bending stress in the extreme fibre of web welds
⎛ 150 × 1000 × 200 × 175 ⎞
Pb1 = ⎜ ⎟
⎝ 27166 × t × 103 ⎠
⎛ 193 ⋅ 26 ⎞
= ⎜ ⎟ N/mm2
⎝ t ⎠
Combined stress in flange welds. From Eq. 10.21.
1
⎡ 125 ⎞2 ⎛ 220 ⋅ 86 ⎞2 ⎤ 2
p = ⎢⎛⎜ ⎟ +⎜ ⎟ ⎥
⎣⎝ t ⎠ ⎝ t ⎠ ⎦

⎛ 253 ⋅ 78 ⎞
= ⎜ ⎟ N/mm2
⎝ t ⎠
Combined stress in web weld
1
⎡ 2 2 ⎤2
p1 = ⎢⎛⎜ 125 ⎞⎟ + ⎛⎜ 193 ⋅ 26 ⎞⎟ ⎥
⎣⎝ t ⎠ ⎝ t ⎠ ⎦

⎛ 230 ⋅ 16 ⎞
= ⎜ ⎟ N/mm2
⎝ t ⎠
726 DESIGN OF STEEL STRUCTURES

Combined stress in the weld should not exceed 110 N/mm2

⎛ 253 ⋅ 78 ⎞
∴ t = ⎜ ⎟ = 2.307 mm
⎝ 110 ⎠
⎛ 2 ⋅ 307 ⎞
Size of web welds = ⎜ ⎟ 3 296 mm. Adopt 5 mm weld.
⎝ 0 ⋅7 ⎠
Size of flange welds
2s = 6.59 mm. Adopt 2 × 5 = 10 mm weld.
Example 10.15 A bracket plate is welded to the flange of a column as shown
in Fig. 10.62. The bracket plate carries a load of 250 kN at a distance of 500 mm
from the flange of column. Provide intermittent fillet weld on both sides of the
bracket plate. Design the welded connection.
Solution
Design :
Step 1 : Number of intermittent fillet welds
The number of weld lengths is found in the same manner as the number of
rivets necessary to resist the moment caused. From Sec. 2.17, eccentric riveted
connections, (design of bracket connections type 1) relation
1
⎡ 6 M (n – 1 ) ⎤ 2
n = ⎢ ⎥
⎣ p⋅ R ⋅n ⎦
(n – 1)
may be used. Factor is assumed to be nearly unity, therefore
n
1
⎡ 6M ⎤ 2
n = ⎢ ⎥
⎣ p⋅R⎦
The intermittent fillet welds are provided on both the sides of bracket plate.
Therefore,
1
⎡ 6M ⎤ 2
n = ⎢ ⎥
⎣2p ⋅ R ⎦
where, n = Number of weld lengths in one line
p = Strength of one weld length.
Let the size of weld for 12 mm thick bracket plate be 8 mm
The effective length of intermittent weld should be four times the size of weld
or 38 mm whichever is more.
Therefore, effective length of weld = 38 mm
Step 2: Spacing between welds
Assume pitch of welds, i.e., distance between centre to centre of adjacent
intermittent welds
p = 80 mm
 DESIGN OF WELDED JOINTS 727

2 50 kN

5 00 m m

P P P P P P P P P P P P
1 2 m m th ick
b racket plate
e ffe ctive
le ng th o f
e ach w eld
= 38 m m

IS H B 4 00
@ 0 .77 4 /m

Fig. 10.62

Clear spacing between the effective length of welds

⎛ 38 ⎞
= ⎜ 80 – 2 × ⎟ = 42 mm.
⎝ 2 ⎠
The clear spacing does not exceed 16 t for compression and 24 t for tension.
The moment at weld section
⎛ 500 ⎞
= ⎜ 250 × ⎟ = 125 kN-m
⎝ 100 ⎠
Strength of one 38 mm weld
⎛ 38 × 110 ⎞
R = ⎜0 ⋅7× 8× ⎟ = 23.41 kN
⎝ 1000 ⎠
Number of weld lengths in one line
1
⎡ 6 M ⎤ ⎡ 6 × 125 × 1000 ⎤ 2
n = ⎢ ⎥= = 14.15
⎣ 2 p ⋅ R ⎦ ⎢⎣ 2 × 80 × 23 ⋅ 41 ⎥⎦
Provide 15 intermittent weld lengths on either face of bracket plate as shown
in Fig. 10.62. the effective length of each weld
= 38 mm
The pitch of weld length is 80 mm.
Example 10.16 A 150 mm diameter mild steel pipe 0.50 metre long is welded
to a vertical plate 12 mm thick, with the axis of pipe at right angles to the face of
the plate. The pipe is subjected to a twisting moment of 1.5 kN-m and carries a
vertical load of 5 kN acting at the end of pipe. Design the welded connection.
Assume thickness of pipe 6 mm. Allowable stress in the weld is 110 N/mm2.
728 DESIGN OF STEEL STRUCTURES

Solution
Design :
Step 1: Properties of weld
The welding is subjected to
(i) a direct load = 5 kN
(ii) a bending moment
⎛ 5 × 500 ⎞
⎜ ⎟ = 2.5 kN-m
⎝ 1000 ⎠
and (iii) a twisting moment = 1.5 kN-m
1 2 m m th ick
5 kN 1 .5 kN -m
5 00 m m

Fig. 10.63

Let t mm be the effective throat thickness of weld.

The approximate polar moment of inertia of weld group
Ixx = ∑δa × r2
where, δa is the elementary area of weld and r is the distance from the centre
Izz = (πrt × r2 = 2 π r3t)
= (2π × 753 × t × 103)
= 2640 t × 103 mm4
Moment of inertia of weld about xx-axis
1
Ixx = I = 1320t × 303 mm4
2 zz
Step 2 : Stresses in welds
Shear stress in the weld due to twisting
⎛ 1 ⋅ 5 × 106 × 75 ⎞ ⎛ 42 ⋅ 6 ⎞
ptw = ⎜ ⎟= ⎜ ⎟ N/mm2
⎝ 2640t × 103 ⎠ ⎝ t ⎠
Stress in the weld due to bending
⎛ 2 ⋅ 5 × 106 × 75 ⎞ ⎛ 142 ⎞
pb = ⎜ ⎟= ⎜ ⎟ N/mm2
⎝ 1320 × t × 103 ⎠ ⎝ t ⎠
The stress in the weld due to shear is maximum at the centre. But average
shear stress is considered. Average shear stress remains uniform.
 DESIGN OF WELDED JOINTS 729

⎛ 5 × 1000 ⎞ ⎛ 10 ⋅ 62 ⎞
ps = ⎜ ⎟=⎜ ⎟ N/mm2
⎝ 2π × 75t ⎠ ⎝ t ⎠
Combined stress in the weld group at the top
1
⎡⎛ 10 ⋅ 62 ⎞2 ⎛ 142 ⎞2 ⎛ 42 ⋅ 6 ⎞2 ⎤ 2 148 ⋅ 63
⎢⎜ ⎟ +⎜ ⎟ +⎜ ⎟ ⎥ = N/mm2
⎣⎝ t ⎠ ⎝ t ⎠ ⎝ t ⎠ ⎦ t
The combined stress in the fillet weld is not to exceed maximum permissible
stress in shear
⎛ 148 ⋅ 63 ⎞
∴ t = ⎜ ⎟ = 1.35 mm
⎝ 110 ⎠
⎛ 1 ⋅ 38 ⎞
Step 3 : Size of weld ⎜ ⎟ = 193 mm
⎝ 0 ⋅7 ⎠
Provide 3 mm size of fillet weld.
Example 10.17 An ISA 125 mm × 95 mm × 12 mm (ISA 12595, @ 0.196 kN/
m) is welded with the flange of a column HB 300, @ 0.630 kN/m as shown in Fig.
10.64. The bracket carries a load of 100 kN at a distance of 60 mm from the face
of column. Design the bracket connection.
Solution
Design :
Step 1: Throat thickness of weld
Width of flange of column HB 300, @ 0.630 kN/m
b = 250 mm
The horizontal fillet welds of 250 mm length each, are provided as shown in
Fig. 10.64.
Let t mm be the throat thickness of weld. The weld is subjected to direct
shear and bending.
Stpe 2: Stresses in welds
Shear force in weld
⎛ 100 × 100 ⎞
τvm.cal = ⎜ ⎟ N/mm2
⎝ 2 × t × 250 ⎠

⎛ 200 ⎞
= ⎜ N/mm2
⎝ t ⎟⎠
Moment about the faces of column
⎛ 60 ⎞
= ⎜100 × 100 × 6 ⎟
⎝ 10 ⎠
= 6 kN-m
The bending moment is resisted by a couple. The weld at the top is subjected
to pull and that at the bottom is subjected to thrust. This is a case of discontinuous
weld i.e., cross flexure
730 DESIGN OF STEEL STRUCTURES

1 00 kN
60
mm

125 m m
IS A 1 25 m m x 95 m m
x 12 m m

Fillet w eld
5 mm
IS H B 3 00
@ 0 .63 0 kN /m

Fig. 10.64

From Eq. 10.8,

⎛ M ⎞
Stress in weld, fs = ⎜ s ⎞ ⎟
⎜ (h + s ) ⋅ ⎛⎜ ⎟ ⋅ l ⎟⎟
⎜
⎝ ⎝ 2⎠ ⎠

⎛ 6 × 106 ⎞
ft = ⎜ ⎟
⎝ 125 × t × 250 ⎠
(h+s) is taken as h only
⎛ 192 ⎞
ft = ⎜ ⎟ N/mm2
⎝ t ⎠
Step 3: Resultant stress in the weld
f = [fs2 + ft2]1/2
1
110 = [(200)2 + (192)2]1/2
t
1
t = [(200)2 + (192)2]1/2 = 2.52 mm
100
2 ⋅ 52
Size of weld s = = 3.6 mm
0 ⋅7
The size of weld is very small. Provide 5 mm weld as shown in Fig. 10.64.

10.15 WELDED PLATE GIRDER

In the welded plate girders, whole of the section is effective in resisting the
loads. Therefore, it is efficient than the riveted plate girder. Figure 10.65 (a)
and (b) show the sections of welded plate girder commonly used in section. Figure
10.65 (c) shows the elevation of a plate girder with bearing stiffeners at the
ends and intermediate stiffeners. In the welded plate girder, flange angles are
 DESIGN OF WELDED JOINTS 731

not used. The flange plates are directly welded to the web of the girder as shown
in Fig. 10.65 (a). The cut sections from wide flange beams are also used as shown
in Fig. 10.65 (b). The stacked flange plates are not used for welded plate girders
because of multiple welding involved, but it is possible to vary the cross-sectional
area by flanking the flange plate required for maximum bending moment with

C u t from w ide
Flan ge flate flan ge b ea m

B u tt w eld

W eb p la te

Fillet w eld

(a ) (b )

B e aring B e aring
stiffe ne r stiffe ne r

In te rm e d ia te S tiffen ers
(c)
E lem e nts o f a w e ld ed p la te girde r

Fig. 10.65 Welded plate girder

one or more successively thinner plates, butt-welded end to end. In welded plate
girder, it is not necessary to make an allowance for rivet holes to determine the
net section for tension flange. There is saving in the section of tension flange.
Generally, the welded plate girders weigh less than riveted plate girders. The
saving may vary from 5 to 15 per cent depending on the governing specifications.
The welded plate girder is economical in material and cost. The section of welded
plate girder is proportioned by the same principles as in the case of riveted
plate girder. Instead of rivets, welds are used for connection.
1 1
(i) Design of web. The depth of plate girder is adopted as to of
8 12
the span of girder. The web plate is designed for maximum shear. The web plate
designed should also satisfy the requirements of local buckling. The economical
depth of web is given by
732 DESIGN OF STEEL STRUCTURES

1
⎡ M ⎤3
de = 5⎢ ⎥ ...(10.23)
⎣ fb ⎦
where, M is the maximum bending moment and fb is the maximum permissible
bending stress (150 N/mm2).
(ii) Design of flanges. The area of flange is determined by approximate
flange area method and then it is checked by moment of inertia method.
Allowance is not made for rivet holes to determine the area of tension flange.
The gross area of flange is effective to resist the moment. The flange
⎛1 ⎞
includes ⎜ ⎟ th the web area in the tension flange. In order to avoid the local
⎝6⎠
buckling, the outstanding of flange plates, with unstiffened edges, that is, their
projection beyond the face of the web shall not exceed 16 t for the flanges in
flexural compression and 20 t for the flanges in flexural tension, where t is the
thickness of the plate.
In general, one plate is used in each flange. The thickness of flange plate is
maximum at the section, where the bending moment is maximum. The value of
bending moment reduced towards the end. Therefore, the maximum thickness
of flange plate provided is not needed throughout the entire span. Either the
thickness or width of flange plate may be reduced as the value of bending moment
decreases. For different lengths of plate girder, either different thicknesses are
used or width of flange is reduced by tapering it. The plates are joined by butt
weld at the junction of two plates and the flange is made continuous. When the
difference in thickness of two plates exceeds 25% thickness of thinner plate or
3.2 whichever is more, the thickness of thick plate is reduced to that of thin
plate at the butt joint. This is done by providing a slope in the thicker plate.
This slope should not be greater than one in five. This is shown in Fig. 10.12 (a).
In case, the thick plate cannot be reduced, the weld metal is built up 25% greater
than the thickness of thinner plate as shown in Fig. 10.12 (c). The thickness of
thicker plate in such case should not be greater than 50% the thickness of thiner
plate.
(iii) Design of connection of flange plate and web. The welds connecting
flange plates with the web plate may be applied continuously or intermittently
on both the sides of web. When continuous weld is used, the size of fillet weld is
changed in different portions of girder. The fillet weld is designed for horizontal
shear per unit length. The horizontal shear per unit length for the loaded flange
is given by
⎡ F ⋅ Ay ( ⎤
fsh = ⎢ × 2t × 1)⎥
(
⎣ I xx ⋅ 2t ) ⎦

⎡ F ⋅ Ay ⎤
∴ fsh = ⎢ ⎥ ...(10.24)
⎣ I xx ⎦
 DESIGN OF WELDED JOINTS 733

where, F = Shear force at the section

Ay = Moment of the area above the horizontal section considered
about xx-axis (neutral axis)
t = Effective throat thickness of the weld
Ixx = Moment of inertia of the welded plate girder about xx-axis
(neutral axis).
Let w be vertical load per unit length over the loaded flange. The combined
stress per unit length is given by
1
2
f = ⎡⎣( fsh ) + w2 ⎤⎦ 2 ...(10.25)
For the unloaded flange, w is zero.
The size of fillet weld to be applied should be such that the strength of weld
per unit length is equal to the combined stress per unit length. If s is the size of
fillet weld in mm, strength of weld per mm length
0.707 × (2s)× 110 = 155.54S N
Therefore, 155.54 s = fsh

⎛ f ⎞
s = ⎜ sh ⎟
⎝ 155 ⋅ 4 ⎠
The size of welds should not be less than minimum size of weld required as
per Table 10.1.
In case intermittent fillet welds are provided, the pitch of weld is found similar
to that or rivets connecting flanges with the web. The minimum size of weld
depends upon thickness of thicker plate to be connected. The thickness of plate
is maximum near the centre where the bending moment is maximum. Near the
supports, the thickness of the plate is small, but shear force is maximum. A
suitable uniform size of weld is selected for the entire span.
The effective length of intermittent fillet weld should not be less than four
times the size of weld or 38 mm, whichever is more. The clear spacing between
the weld should not be greater than 12 t or 200 mm for compression flange and
16 t or 200 mm for tension flange, where t is thickness of thinner plate.
If R is the strength of one weld, the pitch of weld for loaded flange
R
p = 1
...(10.26)
⎡( f )2 + w2 ⎤ 2
⎣ sh ⎦
For the unloaded flange, w = 0.
The weld is provided on both faces, therefore, the weld is provided at a distance
equal to twice the pitch calculated. The size of intermittent weld adopted used
to be large. It is economical to use continuous fillet weld for connections.
The gap between the web plate and the flange plates shall be kept to a
minimum, and for fillet welds shall not exceed 1 mm at any point before welding.
The dispersion of loads through the flange is assumed at 30° to the horizontal.
734 DESIGN OF STEEL STRUCTURES

(iv) Design of bearing stiffener. The stiffeners consist of flat plates instead
of angles as in case of riveted plate girder as shown in Fig. 10.65 (c). The ends of
plates are machined to provide direct bearing so that the load is transmitted by
bearing. The welds connecting the stiffeners to the web are designed to transmit
full load or support reaction. Besides this, the bearing stiffeners are designed
similar to those in riveted plate girder.
(v) Intermediate stiffeners. The stiffeners consist of flat plates instead
of angles as incase of riveted plate girder as shown in Fig. 10.65 (c). The outstand
of stiffeners of flat sections shall be not more than 12 t, where t is the thickness
of flat section used. The intermediate stiffeners are designed similar to those in
riveted plate girder. The weld connecting stiffeners and web are designed to
⎛ 1 ⋅ 5tw2 ⎞
resist horizontal shear ⎜ ⎟ kN/m length, where t w is the thickness of web in
⎝ h ⎠
mm and h is the outstand of the stiffener in mm. The minimum size of weld
depends upon the thickness of stiffener or thickness of web.
In case intermittent welds are used, the instance between the effective lengths
of any two welds, even if staggered on opposite side should not exceed 16 times
the thickness of stiffener nor 300 mm. In case intermittent welds are provided
on one side of stiffener only or both sides but taggered of where single plate
stiffeners are butt welded to the web, the effective length of each weld should
not be less than 10 times the thickness of stiffener. In case, intermittent welds
are provided in pairs on both sides, the effective length of each weld should not
be less than four times the thickness of the stiffener.
Example 10.18 A welded plate girder simply supported at two ends has an
effective span of 28 m. It carries a uniformly distributed load of 25 kN per metre.
Design the maximum section of welded plate girder. Show the reduction of flange
plate.
Solution
Design :
Step 1: Maximum bending moment and shear force
The effective span of welded plate girder is 28 m and the uniformly distributed
load, w is 25 kN/m.
Self weight of girder,
⎛ wL ⎞ ⎛ 25 × 28 ⎞
w1 = ⎜ ⎟ =⎜ ⎟ = 2.33 kN-m
⎝ 300 ⎠ ⎝ 300 ⎠
Total uniformly distributed load
(25 + 23.3) = 27.33 kN/m
Maximum bending moment,
⎛ 27 ⋅ 33 × 28 × 28 ⎞
M = ⎜ ⎟ = 2678.34 kN-m
⎝ 8 ⎠
Maximum shear force,
⎛ 27 ⋅ 33 × 28 × 28 ⎞
F = ⎜ ⎟ = 382.62 kN
⎝ 2 ⎠
 DESIGN OF WELDED JOINTS 735

Stpe 2 : Economical depth of web plate from Eq. 10.23.

1
1/3
⎛M⎞ ⎛ 2678 ⋅ 34 × 106 ⎞ 3
ds = 5 ⎜ ⎟ = 5⎜
⎝ fb ⎠ ⎝ 0 ⋅ 66 × 250 ⎟⎠
= 1265.99 mm
Adopt depth of web = 1200 mm
Let t w be the thickness of web
F ⎛ 382 ⋅ 62 × 1000 ⎞
Thickness of web, =
fs d = ⎜⎝ (0 ⋅ 4 × 250 ) × 1200 ⎟⎠
= 3.1885 mm
The thickness obtained is very small. Adopt 10 mm thickness of web.

5 00 m m

32 m m

6 00 m m 1 20 0 m m

x x

6 00 m m

32 m m

Fig. 10.66 Welded plate girder

Stpe 3 : Design of flange plate

Assume thickness of flange 30 mm. Distance between centre of gravity of
flanges
= (1200 + 15 + 15 = 1230) mm
Area of flange inclusive of web equivalent
⎛ 2678 × 106 ⎞
Af́ = ⎜ ⎟ = 13197 mm2
⎝ (0 ⋅ 66 × 250 ) × 1230 ⎠
Web equivalent
1 1 (
Aw = × 10 × 1200 ) = 2000 mm2
6 6
Area of flange = (13197 – 2000) = 11197 mm2
Thickness of flange plates
= 32 mm
736 DESIGN OF STEEL STRUCTURES

11197
Therefore, width of flange plate required = = 349.906 mm
32
Provide 500 mm × 32 mm flange plates one at the top and one at the bottom
of web plate as shown in Fig. 10.66.
Area of flange plate provided
= 500 × 32 = 16000 mm2
Step 4: Check for bending stress
1
Moment of inertia Ixx = × 1 × (120)3 + 2 × 3.2 × 50 × (61.60)2] × 104
12
= 1358259.2 × l04 mm4
Maximum bending stress

⎛ 2678 ⋅ 34 × 106 × 630 ⎞

fb = ⎜ ⎟
⎝ 1358259 ⋅ 2 × 104 ⎠
= 124.228 N/mm2
Hence, satisfactory.

C e ntre lin e

20 m m 32 m

9 .41 5 m
1m
(a ) R e du ctio n o f fla ng e thickn ess
(e le va tio n )

5 00 m m
2 40

1m
(b ) R e du ctio n o f w idth of fla ng e plate (P la ne )

Fig. 10.67

Step 5: Reduction of flange plate

The thickness of flange plate is reduced to 20 mm.
Moment of inertia of welded plate girder with 500 mm × 20 mm flange plates
1
Ixx = [ × 1 × (120)3 + 2 × 2 × 50 × (61)2] × 104
2
= 888200 × 104 mm4
Moment of resistance
⎛ (0 ⋅ 66 × 250 ) × 888200 × 104 ⎞
M1 = ⎜ ⎟
⎝ 610 × 106 ⎠
= 2402.51 kN-m
 DESIGN OF WELDED JOINTS 737

The equation of parabolic bending moment diagram with left hand support
as origin is y = k . x . (L – x)
The maximum bending moment occurs at centre
∴ ymax = 2678.34 kN-m
At x = 14 m, ∴ k = 13.665
∴ y = 13.665 . x . (L – x)
At y = 239.15, x = 9.415 m
Alternative. Instead of reducing the thickness of flange plate, it is convenient
to reduce the width of flange plate. The tapering width is provided as shown in
Fig. 10.67 (b). The equation of parabolic bending moment diagram curve is
At y = 13.665 x (28 – x)
x = 6.5 m,
y = (13.665 × 6.5 × 21.5)
= 1909.68 kN-m
The bending moment at x equal to 6.5 m
M2 = 1909.68 kN-m
Let b be the width of flange at x equal to 6.5 m
Moment of resistance
I (0 ⋅ 66 × 250 ) ⎡ 1 3 2⎤
Mx = f ⋅ = × ⎢ × 1 × (120 ) + 2 × 3 × b (61 ⋅ 6 ) ⎥ × 104
y 615 ⎣12 ⎦
Equating 1909.68 × 106
(0 ⋅ 66 × 250 ) ⎡ 1 3 2⎤
= × ⎢ × 1 × (120 ) + 2 × 3 × b (61 ⋅ 6 ) ⎥ × 104
615 ⎣12 ⎦
∴ b = (23.65 × 10) = 236.5 mm
Provide, width of flange = 240 mm
Note. The width of flange may be found at different sections. The flange plate may
be tapered on both edges from one end such that the taper accommodates the required
width at those sections.
Example 10.19 A welded plate girder is made of a web 2000 mm deep and 20
mm thick and flange 500 m wide and 40 mm thick. The span of the girder is 28
m and total load per metre inclusive its own weight is 27 kN per metre. Design a
suitable welded connection between the web and the flange.
Solution
Design :
Step 1: Maximum shear force
Total uniformly distributed load is 27 kN/m
Maximum shear force,
27 × 28 ⎞
F = ⎛⎜ ⎟ = 378 kN
⎝ 2 ⎠
Load on compression flange (loaded flange) per mm length
⎛ 27 ⎞
W = ⎜ ⎟ = 0.0270 kN/mm
⎝ 1000 ⎠
738 DESIGN OF STEEL STRUCTURES

Step 2: Properties of plate girder

Area of flange plate
A = (500 × 40)
= 20000 mm2
Moment of this area about xx-axis (neutral axis)
Ay = (20000 × 1020)
= 20.4 × 106 mm3
Moment of inertia of welded plate girder
⎛1 ⎞
Ixx = ⎜ × 1 × 2003 + 2 × 50 × 4 × 1022 ⎟ × 104 mnm4
⎝ 12 ⎠
= 5494900 × 104 mm4
Step 3: Horizontal shear per mm length

⎛ F ⋅ Ay ⎞ ⎛ 328 × 103 × 20 ⋅ 4 × 106 ⎞

fsh = ⎜ ⎟ = ⎜ ⎟
⎝ I xx ⎠ ⎝ 549400 × 104 ⎠
= 140.33 N/mm
Step 4: Combined stress for per 1 mm
f = [(0.14033)2 + (0.0270)2]I/2
= 0.1429 kN/mm

5 00 m m
40 m m
6 mm
6 mm
Fillet w eld
20 m m
2 00 0 m m
x x

Fillet w eld
6 mm

40 m m

Fig. 10.68 Welded connection (flange and web)

Step 5 : Continuous weld

⎛ 0 ⋅ 1429 × 1000 ⎞
Size of weld, s = ⎜ ⎟
⎝ 2 × 0 ⋅ 7 × 110 ⎠
= 0.928 mm
Minimum size of filet weld = 6 mm
Provide 6 mm fillet weld for connection of the flange plate and the web plate
as shown in Fig. 10.68.
 DESIGN OF WELDED JOINTS 739

5 00 m m
40 m m
p p
6 m m Fille t w e ld
3 8 m m E ffe ctive
2 00 0 le n gth p = 22 5 m m
x x

p p
40 m m

Fig. 10.69 Intermittent welded connection flange and web

Step 6 : Intermittent weld

Let 6 mm size of fillet weld be used
Effective length = 4 × Size of weld or 38 mm
= 24 mm or 38 mm
(whichever is more)
Hence the effective length of fillet weld is adopted as 38 mm
Strength of weld on both the faces
⎛ 2 × 38 × 0 ⋅ 7 × 6 × 110 ⎞
R = ⎜ ⎟ = 35.11 kN
⎝ 1000 ⎠

R ⎛ 35 ⋅ 11 ⎞
Pitch of weld, p = =⎜ ⎟ = 245.71 mm
fsh ⎝ 0 ⋅ 1429 ⎠
Provide intermittent weld at 225 mm pitch
Allowable clear spacing
= 12 × t = 12 × 20 = 240 mm
Hence the clear spacing provided is satisfactory.
The intermittent fillet weld of 6 mm size is provided at a pitch of 225 mm as
shown in Fig. 10.69.
Example 10.20 Design the end and intermediate stiffeners for the girder in
Example 10.10. Assumed shear in weld is not to exceed 110 N/mm2.
Solution
Design of end stiffeners
Step 1: Bearing area required
The uniformly distributed load inclusive of self weight of the welded plate
girder is 27 kN/m and the span of girder is 28 m.
⎛ 27 × 28 ⎞
Support reaction = ⎜ ⎟ = 378 kN
⎝ 2 ⎠
740 DESIGN OF STEEL STRUCTURES

The allowable bearing stress is 185 N/mm2

⎛ 378 × 1000 ⎞
Bearing area required = ⎜ ⎟ = 2043.24 mm2
⎝ 185 ⎠
The flat plates are provided for end bearing. The ends are machined so that
plates fit tightly between the flange plates
Provide 2 flat plates 160 mm ×10 mm
% Bearing area provided = (2 × 120 × 10) = 2400 mm2 > 204324 mm2
Width of flat plate = 120 mm
Thickness of flat plate = 120 mm
⎛ Width ⎞ 120
⎜⎝ ⎟ = = 12 |
> 12 t (As per IS : 800–1984)
Thickness ⎠ 10
Step 2 : Check for bearing stiffener
The bearing stiffener is checked as a compression member. This column section
consists of the pair of stiffeners together with a length of wed on each side of the
centre line of the stiffeners and equal to 20 times the thickness of web, if available.

5 00 m m
40 m m

10 m m
1 20 m m

1 20 m m

2 00 0
mm 1 20 m m
x x 5 00
mm
1 20 m m E n d stiffen e r
20
mm

40 m m

Fig. 10.70 Bearing stiffener (Welded plate girder)

Area of cross-section
= (2 × 20 × 20 + 2 × 120 × 10)
= 3200 mm2
Moment of inertia of the stiffeners about centre line
1
2× × 10 × 1703 × 104 = 818.83 × 104 mm4
2
1
⎡ 818 ⋅ 83 ⎤2
Radius of gyration r = ⎢ × 104 ⎥ = 50.58 mm
⎣ 3200 ⎦
Effective length = (0.7 × 2000) = 1400 mm
⎛ 1400 ⎞
Slenderness ratio = ⎜ ⎟ = 20.7
⎝ 50 ⋅ 58 ⎠
 DESIGN OF WELDED JOINTS 741

From IS : 800–1984, allowable stress in axial compression for the structural

steel having yield stress as 250 N/mm2
Safe load carrying capacity
⎛ 147 ⋅ 79 × 3200 ⎞
⎜ ⎟ = 472.93 kN > 378 kN
⎝ 1000 ⎠
Hence the design as satisfactory.
Provide 120 mm × 10 mm two flat plates for end stiffeners. The flat plates
are connected by 66 mm minimum size of fillet weld as shown in plates,
Fig. 10.70.
Step 3. Design of intermediate stiffener
The clear distance between flanges of welded plate girder is 2000 mm and
the thickness of web, t w is 20 mm.
In case, the web plate of the welded plate girder remains unstiffened, the
thickness of web plate shall be not less than t w.min found as under :
Average shear stress at the support
⎛ 378 × 1000 ⎞
τva.cal = ⎜ ⎟
⎝ 20 × 2000 ⎠
= 9.45 N/mm2
⎛ d ⋅t 1/2 ⎞
⎛ 2000 × 9 ⋅ 451 / 2 ⎞
(i) τw.min = ⎜ 1 va ⋅cal ⎟ = ⎜ ⎟
⎝ 816 ⎠ ⎝ 816 ⎠
= 7.5 mm
⎛ d1 ⋅ fy1 / 2 ⎞ ⎛ 200 × 2501 / 2 ⎞
(ii) τw.min = ⎜ ⎟⎜ ⎟
⎝ 1344 ⎠⎝ 1344 ⎠
= 23.528 mm
⎛d 2000 ⎞
(iii) τw.min = ⎜ 1 = ⎟ = 23.53 mm.
⎝ 85 85 ⎠
Actual thickness of the web plate is 20 mm and it is less than 23.53 mm.
Therefore, it is necessary to provide vertical stiffeners to stiffener the web plate.
The shear stress is very small. The intermediate stiffeners may be provided
at maximum spacing i.e., 1.5 d
l.5 × 2000 = 3000 mm
Moment of inertia required for intermediate stiffener
⎛ 1 ⋅ 5 × (20003 × 203 ) ⎞
⎜ ⎟ = 1066.66 × 104 mm4
⎝ 30002 ⎠
Provide 2 flat plates 10 mm ×110 mm, as shown in Fig. 10.71
Moment of inertia provided
1
× 10 × (110+ 110 + 20)3 = 1152 × 104 mm4
12
> 1066.66 × 104 mm4. Hence, satisfactory.
Shear force on the weld connecting intermediate stiffeners
742 DESIGN OF STEEL STRUCTURES

⎛ 125 tw2 ⎞ ⎛ 125 × 20 × 20 ⎞

⎜ ⎟ = ⎜ ⎟
⎝ h ⎠ ⎝ 110 ⎠
= 454.545 kN
where, h is outstand of stiffeners in mm

5 00 m m
40 m m
11 0 m m

110 m m

2 00 0 10 m m
mm 11 0 m m
20 m m
5 00
x x
mm
11 0 m m

3 00 0 m m 3 00 0 m m

40 m m

In te rm ed ua te (vertical) stiffen ers (w e ld e d p la te g ird er)

Fig. 10.71 Welded plate girder

Size of weld required

⎛ 454.54 × 1000 ⎞
⎜ ⎟ = 2.95 mm
⎝ 2 × 0.7 × 100 ⎠
The size of weld obtained is very small. The intermittent fillet welds of 3 mm
size are provided for the connection.

PROBLEMS
10.1. Two plates 14 mm thick are joined by (i) a double V butt weld, (ii) a single-
V butt weld. Determine the strength of the welded joint in tension in each
case. The effective length of weld is 200 mm.
10.2. Design a suitable side fillet weld to connect two plates 100 mm × 10 mm
and 120 mm × 12 mm, and to transmit pull equal to the full strength of
thin plate.
10.3. Two plates 100 mm × 10 mm are connected in a lap joint by means of end
fillet weld. Design suitable weld to transmit pull equal to full strength of
the plate.
10.4. A truss joint is as sketched in Fig. P.10.4. Design and detail the welds
connecting the members to the gusset.
 DESIGN OF WELDED JOINTS 743

4 2 kN 3 0 kN
IS A 4 0 m m × 2 5 m m
× 6 mm

4 5° 60

5 0 kN C o ntinu o us cho rd 7 0 kN

U n eq ua l an ge l IS A 50 m m × 30 m m × 6 m m

Fig. P.10.4
10.5. A welded bracket is of the design shown in Fig. P.10.5. Design the welded
connections completely.

5 0 kN

2 0 kN

G usset plate
2 2 m m th ick

C o lu m n flan ge
8 m m thick

Fig. P.10.5
10.6. An ISA 100 mm × 75 mm × 10 mm (ISA 100 75, @ 0130 kN/m) is welded
with the flange of a column HB 350, @ 674 N/m as shown in Fig. P.10.6.
The bracket carries a load of 80 kN at a distance of 40 mm from the face of
column. Design the bracket connection.

8 0 kN

1 00 IS A 1 00 m m × 7 5 m m
mm × 10 mm

IS H B 3 50
@ 0 .67 4 kN /m

Fig. P.10.6
744 DESIGN OF STEEL STRUCTURES

10.7. Figure P.10.7 shows an I-section bracket 350 mm × 140 mm connected to

a steel stanchion by 12 mm fillet welds. The weld lengths are indicated on
the figure. Determine the load P which the bracket can support if the stress
in the welds is not to exceed 110 N/mm2.

P
1 20 m m

1 20 m m

Fig. P.10.7

10.8. A bracket is loaded as shown in Fig. P.10.8. Design the welded connection.

1 20 kN 1 20 kN
40 80
mm mm

G usset plate G usset plate

th ickne ss th ickne ss
10 m m 1 00 16 m m
mm

IS H B 3 00 , 40 IS H B 3 00
mm
@ 0 .58 8 kN /m @ 0 .58 8 kN /m

Fig. P.10.8 Fig. P.10.9

10.9. A welded bracket is loaded as shown in Fig. P.10.9. Design the suitable
weld.
10.10. A 100 mm diameter mild steel pipe 0.40 m long is welded to a vertical
plate 10 mm thick with the axis of pipe at right angles to the face of the
plate. The pipe carries a vertical load of 10 kN at its free end. Design the
welded connection. Assume thickness of pipe 6 mm.
10.11. A welded plate girder is simply supported at both the ends. The effective
span of plale girder is 16 m. It carries a uniformly distributed load of 30
kN/m of exclusive of its own weight. Design the maximum section of welded
plate girder, end bearings and intermediate stiffeners.
10.12. An I-section is built-up by welding 600 mm × 15 mm (thick) web plate and
two flange plates 150 mm × 12 mm (thick). Design the welded joint to
develop full strength of the section. Adopt safe stresses as per I.S. code.
 DESIGN OF WELDED JOINTS 745

10.13. A circular shaft of diameter 100 mm is welded to a rigid plate by a fillet

weld 6 mm size. If a torque of 6 kN-m is applied to the shaft, find the
maximum stress in the weld.
10.14. An angle section ISA 40 mm × 25 mm × 6 mm (ISA 4025, @ 0028 kN/m) is
to be joined to a gusset plate with the longer leg attached. The joint is to be
designed for maximum load carrying capacity of the angle iron in tension.
The thickness of the gusset plate is 8 mm. Design the welded joint.
10.15. Two angles of ISA 90 mm × 90 mm × 8 mm (ISA 9090, @ 0108 kN/m)
transmit at tensile force of 250 kN. The angles are connected one on each
side of a gusset plate 10 mm thick by welding. Design the joint using 6 mm
weld. The C.G. line for each angle Cxx = Cyy = 28.7 mm.
10.16. A welded plate girder of effective span 25 m is made of web 2000 mm deep
and 20 mm thick and flanges 500 mm wide and 30 mm thick. It carries a
uniformly distributed load of 30 kN/m including its own weight. Design
the intermittent weld connection between the web and the flange.
10.17. A welded plate girder has a uniform section as shown in Fig. P.10.17. The
plate girder carries a uniformly distributed load of 40 kN/m over the entire
span of 25 m. The girder is simply supported at both the ends. Design the
welded connection between flanges and web plate.
3 00 m m

10 5 0 kN
10 m m

3 00
6 00 m m mm

G usset
2 0 m m th ick

10 Flan ge
1 2.7 m m th ick

Fig. P.10.17 Fig. P.10.18

10.18. Design the bracket connection (Fig. P.10.18) assuming that the loads shown
are for one gusset plate only.
10.19. Design a single angle tension member of a roof truss carrying an axial
load of 100 kN. The member is connected to a gusset by welds on end and
sides. Design the welded connection.
10.20. A bracket web plate 12 mm thick transmits a load of 200 kN at an
eccentricity of 350 mm to flange of a column. Design the fillet welds on
either side of web.
10.21. A 12 m span welded plate girder is subjected to a uniformly distributed
load of 30 kN/m along with a concentrated load of 150 kN at 4 m from one
of the supports. Design
(i) the cross-section of the plate girder assuming the web to be 6 mm thick,
(ii) the welded joint for connecting the flange plates with the web
(iii) the web stiffeners including bearing stiffeners at supports.
 Chapter
Welded Beam

11 Connections

11.1 INTRODUCTION
The welded beam connections are similar to the riveted connections in many
respects. The welded beam connections are generally more compact and free
than the riveted connections. The beams are connected to beams or columns at
their ends by direct fillet or butt welds, or through the plates and angles. The
welded beam connections are of four types :
1. Direct welded connections
2. Welded framed connections
3. Welded seat connections
4. Moment-resistant welded connections.

11.2 DIRECT WELDED CONNECTIONS

The beams are connected to beams or columns at their ends by direct fillet
welds or butt welds. The fillet welds are applied on both the sides of the web of
beam as shown in Fig. 11.1. The web is connected directly to the face of support-
ing member. The fillet welds are designed for vertical reaction at the end of
beam. The restraining moment existing at the weld section is not considered.
As far as possible, the fillet welds are applied near the top. The minimum size of
fillet weld is kept equal to the thickness of web or 7 mm whichever is more. The
welds, web of beam and element of supporting member are checked for shear.
Let V be the vertical reaction at the end of beam. Then, for shop welds, the
shear stress
⎛ V ⎞
⎜ ⎟ < 110 N/mm2 ...(11.1)
⎝ 2 × 0 ⋅ 7s ⋅ h ⎠
Average shear stress in web of beam
 WELDED BEAM CONNECTIONS 747

⎛ V ⎞
⎜ t h ⎟ < 0.4fy N/mm2 ...(11.2)
⎝ w⋅ ⎠
The shear at supporting element is resisted at two sections. Therefore, the
shear stress in the element of supporting member
⎛ V ⎞
⎜ ⎟ < 0.4fy N/mm2 ...(11.3)
⎝ 2t ⋅ h ⎠
From Eq. (11.1) and Eq. (11.2)

h B e am

Fillet welds

Fig. 11.1

⎡ V ⎤ ⎛ 110 ⎞
⎢ 2 × 0 ⋅ 7s ⋅ h ⎥ < ⎜ 0 ⋅ 4 f ⎟
⎢ ⎥ ⎝ y⎠
⎢ V ⎥
⎢⎣ tw⋅h ⎥⎦

⎛ 0 ⋅ 4 fy ⎞
s < ⎜ ⋅ tw ⎟
⎝ 11 × 1 ⋅ 4 ⎠

⎛ 0 ⋅ 4 fy ⋅ tw ⎞
s < ⎜ ⎟ ...(11.4)
⎝ 154 ⎠
From Eq.(11.1) and Eq.(11.3)

⎡⎛ V ⎞⎤
⎢ ⎜⎝ 2 × 0 ⋅ 7s ⋅ h ⎟⎠ ⎥ ⎛ 110 ⎞
⎢ ⎥ < ⎜ 0 ⋅ 4f ⎟
⎢ ⎛ V ⎞ ⎥ ⎝ y ⎠
⎢⎣ ⎜⎝ 2s ⋅ h ⎟⎠ ⎥⎦

⎛ 0 ⋅ 4f × 2 0.8 fy ⎞
⎜s < ⋅ t; s < ⋅t ⎟ ...(11.5)
⎝ 110 × 1 ⋅ 4 154 ⎠
748 DESIGN OF STEEL STRUCTURES

The rotation of end of beam takes place. This rotation may take place by
yielding of web or yielding of weld. The size of weld is made larger than the
minimum size necessary, so that yeilding will occur in the web plate. In order to
provide adequate flexibility, the depth of weld should not be greater than one-
half to two-third the depth of beam, and size of the weld is kept at least four-
fifth the thickness of web. However, size of weld equal to two-third the thick-
ness of web is considered effective to determine depth of weld.
When the beams are connected to beams or columns at their ends directly or
columns at their ends directly by butt weld, a backing strip is used as shown in
Fig. 11.2. The backing strip is welded with the web in the shop. Then the web of
beam is connected to the supporting member by butt weld.

C o lu m n B u tt w eld

B e am

Fillet w eld
B u tt
w e ld

B a ckin g strip

Fig. 11.2

The extra weld metal is deposited on the butt weld. This provides extra
strength to butt weld. The depth of butt weld is kept small as far as possible in
order to allow end rotation. The butt weld is designed to resist the vertical
reaction at the end of beam. If V be the vertical reaction, then, the shear stress
in butt weld.
⎛ V ⎞
⎜ d × t ⎟ < fsh(110 N/mm2)
⎝ w ⎠

The direct fillet weld or butt weld needs preparation of accurate length of
beam. The rotation of end of beam takes place by yielding of web. Therefore, as
far as possible, these types of connections are not used.
Example 11.1 A MB 500, @ 0.869 kN/m transmits an end reaction of 130 k/N
to the flange of a column MB 250, @ 0.510 kN/m. The fillet welds are applied
directly on both the sides of web of the beam. Design the connections.
Solution
Design:
Step 1: Size of weld
From ISI Handbook No. 1, for MB 500, @ 0.869 kN/m beam
The thickness of web, t w is 10.2 mm
For MB 250, @ 0.510 kN/m column,
 WELDED BEAM CONNECTIONS 749

The thickness of flange, t f is 9.7 mm

4
Size of fillet weld = × 10.2 = 8.16 mm
5
Provide 9 mm fillet yield.
Effective size of fillet weld from Eq. 11.4

⎛ 0 ⋅ 4 fy ⋅ tw ⎞ ⎛ 0 ⋅ 4 × 250 × 10 ⋅ 2 ⎞
s = ⎜ ⎟= ⎜ ⎟
⎝ 154 ⎠ ⎝ 154 ⎠
= 6.62 mm
⎛ 130 × 1000 ⎞
Step 2: Depth of weld = ⎜ ⎟ = 272.31 mm
⎝ 0 ⋅ 7 × 6 ⋅ 62 × 10 ⎠
Provide 140 mm deep fillet welds on each side of web.
Effective size of weld

⎛ 0 ⋅ 8 fy ⋅ t ⎞ ⎛ 0 ⋅ 8 × 250 × 9 ⋅ 7 ⎞
< ⎜ ⎟ = ⎜ ⎟ = 12.6 mm
⎝ 154 ⎠ ⎝ 154 ⎠
Therefore, the flange of column is not overstressed.

11.3 WELDED FRAMED CONNECTIONS

The direct fillet weld or butt weld needs preparation of accurate length of beam.
The gap between end of beam and supporting member must be short. For fillet
weld, the end of beam must remain completely flush with the element of sup-
porting member. Butt weld may be provided even if the gap is upto the extent of
3 mm. But, when beams are manufactured for ordered length, the rolling mill
allows a tolerance of 12 mm either way. The gap may be greater at one flange of
beam, than at the other. The beam end may be prepared by milling, cold sawing
or flame cutting. All these operations are fairly expensive. To avoid this diffi-
culty, welded framed connections are used. In the welded framed connections,
two plates or two angles are used. The plate connection as shown in Fig. 11.3
may be considered as a special case. The plates make the connection somewhat
flexible, than the direct welded connection.
One plate is used on one side of web of the beam and other plate is attached
on the other side. For connection, one plate is welded in the shop with the beam
and other plate is welded in the shop with the element of supporting member.
When both the members are kept in position, erection bolts are inserted and
welding is done. To allow rotation at the end of beam, the lengthof plates is
limited to two-third the depth of beam. The size of weld connecting plates to
the web of beam is kept equal to four-fifth of thickness of web.
The welds connecting plate to the supporting member are designed to resist
the vertical reaction V, at the end of beam, Whereas, the weld connecting
plate to the web of beam are designed to resist vertical reaction, V, moment,
V.l where l is distance between two welds. The length of plates is kept about
750 DESIGN OF STEEL STRUCTURES

5 mm. The plates may be thicker than the welds connecting plates and web of
beam by 16 mm. In that case, the plates will also be safe, if the welds connecting
plates and web ofbeam are safe.

Fig. 11.3

Example 11.2 In Example 11.1, two plates are used for connection on both
the sides of web of a beam. Design the welded connections
Solution
Design
Step 1: Moment to be resisted
The vertical reaction at the end of beam, V is 130 kN
Take two plates 50 mm long
⎛ 130 × 50 ⎞
Moment M =⎜ ⎟ = 6.5 kN-m
⎝ 1000 ⎠
Step 2: Strength of 1 mm weld : weld value
1. Weld connecting plates and web in the shop
For web thickness of 10.2 mm, 9 mm fillet welds are adopted.
⎛ 0 ⋅ 7 × 9 × 110 ⎞
R = ⎜ ⎟ = 0.693 kN/m
⎝ 1000 ⎠
The fillet welds are provided on both the sides of web
The depth of weld
1 1
6 M ⎤ 2 ⎡ 6 × 6 ⋅ 5 × 1000 ⎤ 2
d = ⎡⎢ =⎢ = 167.75 mm
⎣ 2R ⎥⎦ ⎣ 2 × 0 ⋅ 693 ⎥⎦
Try 250 mm deep welds
Vertical shear per mm length,
⎛ 130 × 1000 ⎞
fsv = ⎜ ⎟ = 260 N/mm
⎝ 2 × 250 ⎠
 WELDED BEAM CONNECTIONS 751

Horizontal shear per mm length,

⎛ 6 × 6 ⋅ 5 × 106 ⎞
fsv = ⎜ ⎟ = 312 N/mm
⎝ 2 × 250 × 250 ⎠
Resultant shear, f = ((260)2 + (312)2)1/2 = 406 N/mm
< 693 N/mm. Hence, safe.
2. Weld connecting plate to the supporting member in the field
Let s be the size of weld
Then 0.7 × s × (0.1 × 110) × 2 × 250 = 130 × 1000
⎛ 130 × 1000 ⎞
s = ⎜ ⎟ = 4.22 mm
⎝ 0 ⋅ 7 × 0 ⋅ 8 × 110 × 500 ⎠
Provide 6 mm size for the welds
Thickness of plates = (9 + 1.6) =10.6 mm
Provide 12 mm × 50 mm plates.
The connections of two framing angles are similar to the riveted frame
connections. The welded framed connections with two framing angles are more
flexible than other type of welded connections. The bending of outstanding legs
of the angles provides flexibility to the connection. During the erection, the
erection bolts are provided through the connection angles.
A welded framed connection is shown in Fig. 11.4. In order to provide flexibility
to the connections, thickness of angle is kept less than the size of weld.
B olts

C o lu m n B e am

Fig. 11.4

For economical consideration, the length of leg of angle connecting the web of
beam should be short. The welds are usually returned to the end of beam on
this leg, as shown in Fig. 11.4. This welding is done in the shop. This will now be
referred as shop weld. The outstanding legs are connected to the supporting
member in the field and the weld will be referred as field weld. The field welds
are returned at the top for short distance.
Figure 11.5 shows a beam with two frame angles attached to it. The beam is
assumed simply supported at the face of outstanding legs of connecting angles.
Therefore, the field welds are not subjected to any moment. Both the angles are
752 DESIGN OF STEEL STRUCTURES

connected individually to the web of beam. These angles are not connected
R
together. Each field weld is subjected to the vertical reaction as shown in
2
⎛ R⎞
Fig. 11.5. Therefore, each shop weld is subjected to vertical reaction ⎜ ⎟ , and
⎝ 2⎠

⎛ R⎞
a twisting moment, ⎜ ⎟ . l2.
⎝ 2⎠

R R R R R R
h 2 2 2 2 2 2 B e am

l1 l1 l2

Fig. 11.5

R .
Each angle is also subjected to a moment . l Though this moment is resisted
2 1
by both shop weld and field weld, it is assumed that this moment is resisted by
field weld only. This moment is resisted by bearing pressure between angles
and web at the top and between angles and weld for the remaining length.
h
It is assumed that the neutral axis is situated atfrom the top of angles as
6
shown in pressure distribution diagram, Fig. 11.5. The maximum horizontal
shear in the field may be obtained as under:
⎛1 5 ⎞ 2 R
⎜⎝ 2 fsh ⋅ 6 h⎟⎠ ⋅ 3 h = 2 ⋅ l1

5 ⎛ R ⋅ l1 ⎞
fsh = ⋅ ...(11.6)
9 ⎜⎝ h2 ⎟⎠

R
The resultant of horizontal shear and shear due to vertical reaction is
2
found by vector sum. The resultant shear stress must be less than the allowable
shear. The depth of angle is kept about one-half to two-third the depth of the
beam, 2 ISA 90 mm × 60 mm are used for framing angles. The thickness of
angles is determined after designing the welds.
Example 11.3 In Example 11.1, two framing angles are used for connections
on both the sides of web of beam. Design the welded connections.
 WELDED BEAM CONNECTIONS 753

Solution
Design
Step 1 : Properties of section
Depth of beam = 500 mm
Half depth of beam = 250 mm
2
depth of beam = 333.3 mm
3
The depth of angles is adopted between one-half to two-thirds the depth of
beam.
Assume depth of angles,
h = 300 mm
2 ISA 90 mm × 60 mm are used for framing angles.
1. Design of field weld vertical reaction on each weld
R 130
= = 65 kN
2 2
Horizontal shear per mm length

⎛ 9 R ⋅l ⎞
fsh = ⎜ × 2 1 ⎟
⎝5 h ⎠

10 m m

3 00 m m

Fig. 11.6

5 ⎛ 130 × 1000 × 90 ⎞
fsh = × = 234 N/mm
9 ⎜⎝ 300 × 300 ⎟⎠
Vertical shear per mm length

⎛ 65 × 1000 ⎞
fsv = ⎜ = 216.6 N/mm
⎝ 300 ⎟⎠
754 DESIGN OF STEEL STRUCTURES

The resultant shear stress

t s = [fsv)2 + (fsh)2]1/2 = [(216.6)2 + (234)2]1/2 = 320 N/mm
Let s be the size of field weld. The strength of 1 mm weld is 0.7s × (0.8 × 0.8)
× 1 = 320
⎛ 330 ⎞
s = ⎜⎝ 0.7 × 0.8 × 110 ⎟⎠ 5.195 mm

Provide 6 mm size for fillet weld.

2. Design of shop weld

⎛ 50 ⎞
2 × 50 × ×t
⎜ 2 ⎟
x = ⎜ = 6.25 mm
2 × 50 × t + 300 × t ⎟
⎜ ⎟
⎝ ⎠
Therefore, the twisting moment
130
M = × (60 – 6.25) = 3493.75 kN-m
2
Step 2: Properties of welds
Let t mm be the throat thickness
1
Ixx = [t × 303 + 2 × 5 × t × 152] ×103 mm4 = (2250 + 2250) × t× 103
12
= 4500 t × 103 mm4
1
Iyy =[30 × t × 0.6252 + 2 × × t × 53 + 2 × 5 × t (2.5 – 0.625)2] × l03 mm4
12
= (514 t × 103) mm4
J = (4500t + 51.4t) = (4551.4 t × 103) mm4
Step 3: Stresses in welds
1. Horizontal shear in the weld
⎛ 3493.75 × 103 × 150 ⎞ 115
fsh = ⎜ ⎟ = t kN/mm
2
⎝ 4551.4t × 103 ⎠
2. Vertical shear in the weld
⎛ 65000 3493.75 × 103 × 43.75 ⎞ 196.1
fsv = ⎜ + 3 ⎟ =
⎝ 2 × 50 × t + 300t 4551.3t × 10 ⎠ t
3. Resultant shear stress
1
1 2 227 ⋅ 33
f = ⎡⎣115 + (196 ⋅ 1 )⎤⎦2 = N/mm2
t t
227 ⋅ 33
∴ = 110, t = 2.066 mm ∴ s = 2.98 mm
t
 WELDED BEAM CONNECTIONS 755

Provide 6 mm size for shop welds. The thickness of angle is kept equal to
Ithe size of weld i.e., 6 mm.

11.4 WELDED SEAT CONNECTIONS

When a beam is connected to a beam or a column by means of an angle at the
bottom of the beam, which is shop welded to the supporting member and an
angle ISA 100 mm × 100 mm × 6 mm at top which is field welded, the connections
are known as seated connections. The top angle provides stability to the beam.
The welded seat connections provide some flexibility as the welded framed
connections. The welded seat connections provide satisfactory and efficient
support to the beams. Similar to the riveted connections, the welded seat
connections are also unstiffened and stiffened.

11.4.1 Unstiffened Welded Seat Connections

The unstiffened welded seat connections are shown in Fig. 11.7. In Fig. 11.7,
one beam is supported at the flange of a column and other beam is supported at
the web. The unstiffened seat connections are used to support light loads. The
Fillet w eld 6 m m
IS A 1 00 m m x
1 00 m m x 6 m m

B e am

B u tt w eld

S e at an gle

Fig. 11.7 Unstiffened welded connections

vertical welds are used to connect seat angles. These welds are returned at the
top of seat angles. The return ends of weld are not considered to carry-loads.
The top angles are welded along the toss by 6 mm welds. Therefore, it remains
free to bend and provides flexibility to the connection.
The bearing length b of seat angle is determined in the manner seat similar
to that of the riveted connection.
The bearing length
⎛ F ⎞ 1 1
b = – 3 ⋅ h2 <|
⎜⎝ fb ⋅ tw ⎟⎠ 2 fb ⋅ tw

The end reaction F is considered as uniformly distributed on the outstanding

leg of seat angle over the length b. A clearance of 10 mm is provided between
the end of beam and column. The critical section is assumed at the edge of filler.
756 DESIGN OF STEEL STRUCTURES

The distance of end reaction from critical section of the outstanding leg of seat
angle is determined, and bending moment M is found at the critical section. The
length of seating angle is assumed equal to width of the flange of beam. The
thickness of seating angle is found by equating moment of resistance to the
moment at the critical section.
The neutral axis for vertical welds is assumed to be situated at mid-height,
The horizontal shear per mm and vertical shear per mm of weld are determined
and combined resultant shear per mm length should be less than or equal to
allowable shear per mm of welds.

11.4.2 Stiffened Welded Seat Connections

The design of stiffened seat connections is somewhat simpler than that of riveted
connections, In the stiffened welded seat connections, two plates welded together
forming a tee-section are used. A split-beam section is also used to form a seat
in stiffened welded seat connections. The stiffened welded seat connections are
shown in Fig. 11.8. The thickness of stiffening plate is kept equal to thickness of

IS A 1 00 m m x 10 0 x 6 m m
6 m m Fille t

B e am

S e at plate
S tiffen in g plate
Fillet w eld

Fig. 11.8 Stiffened welded seat connection

web of the beam. The thickness of seat plate is kept equal to flange thickness.
The stiffening plate is welded along both the vertical sides. The horizontal welds
are provided along the under-side of the seat plate. The length of each horizon-
tal weld is kept one-fifth to one-half the length of vertical leg. The horizontal
welds and vertical welds form a double angle section as shown in Fig. 11.8.
The bearing length b is determined as for the unstiffened welded seat con-
nections. For the stiffened seat connections the eccentricity of beam reaction is
more than the unstiffened one. The position of beam reaction is assumed to lie
at the mid-point of bearing length. The bearing length is measured from outer
end of seat plate.
Example 11.4 A MB 500, @ 0.869 kN/m transmits an end reaction of 130 kN
to the flange of column HB 250, @ 0.510 kN/m. Design an unstiffened welded
seat connection.
 WELDED BEAM CONNECTIONS 757

Solution
Design
Step 1: Bearing length of seat connection
The end reaction of beam, F is 130 kN
From ISI Handbook No. 1, for MB 500, @ 0.869 kN/m
Width of flange, bt = 180 mm and thickness of web,
tw = 10.2 mm
Thickness of flange, tf = 17.2 mm, h2 = 37.95 mm
The bearing length

⎛ F ⎞ ⎛ 130 × 1000 ⎞
b= ⎜ – 3 ⋅ h2 ⎟ = ⎜ – 3 × 37 ⋅ 95 ⎟ mm = 3.16 mm
f t
⎝ b⋅ w ⎠ ⎝ 185 × 10 ⋅ 2 ⎠

⎛1 F ⎞ 1 ⎛ 130 × 1000 ⎞
But b <
|⎜ ⎟ or <
| ×⎜ ⎟<| 34.45 mm
⎝ 2 fb⋅tw ⎠ 2 ⎝ 185 × 10 ⋅ 2 ⎠
Let the bearing length of seating angle b be 34.45 mm
Try seating angle ISA 150 mm × 115 mm × 12 mm. (ISA 150 115, @ 0.238
kN/m) The long leg is kept vertical.
Radius at root, r1 = 11.0 mm
The distance of end reaction from critical section
⎡⎛ 1 ⎞ ⎤
⎢⎜⎝1 ⋅ 0 + 2 × 3 ⋅ 37 ⎟⎠ – (1 ⋅ 2 + 1 ⋅ 1)⎥ × 10 = 3.85 mm
⎣ ⎦
Step 2: Bending moment at the critical section
(3.85 × 130 × 1000) = 500.5 kN-mm
The width of flange of beam is 18 mm
The length of seating angle is 18 mm
The moment of resistance of seating angle
⎛ 0 ⋅ 66 × 250 × 180 × 122 ⎞
MR = ⎜ ⎟ = 712.8 kN-mm
⎝ 6 ⎠
Hence, the seating angle is safe.
Step 3: Design of welds
The length of vertical welds is 150 mm. The vertical shear per mm welds
⎛ 130 × 1000 ⎞
⎜ ⎟ = 433.33 N/mm
⎝ 2 × 150 ⎠
The distance of end reaction from the weld
⎛ 34 ⋅ 45 ⎞
⎜ + 10 ⎟ = 27.225 mm
⎝ 2 ⎠
Step 4: Bending moment at the weld,
M = 130 × 27.225 = 3539.25 kN-mm
758 DESIGN OF STEEL STRUCTURES

Step 5: Horizontal shear per mm,

fsh = ⎛ 3539 ⋅ 25 × 1000 ⎞ = 471.9 N/mm

⎜ 1 ⎟
⎜ 2 × × 450 × 150 ⎟
⎝ 6 ⎠
Step 6: Resultant shear stress per mm,
1
2 2
fs = ⎡⎣(433 ⋅ 33) + (471 ⋅ 9 ) ⎤⎦ 2
= 640.675 N/mm
Let s be the size of the weld
Then, (0.7 × s × 1 × 110) = 640.675
∴ s = 8.235 mm
Provide 9 mm fillet weld and ISA 150 mm × 115 mm × 12 mm (ISA 150 115,
@ 0.238 kN/m) seating angle. ISA 100 mm × 100 mm × 6 mm (ISA 100 100, @
0.092 kN/m) is used at the top and 6 mm weld is applied to the toes.
Example 11.5 A MB 500, @ 0.869 kN/m transmits an end reaction of 240 kN
to the flange of a column HB 250, @ 0.510 kN/m. Design stiffened seat connections.
Solution
Design
Step 1: Bearing length of seat connection
End reaction of beam
F = 240 kN
From ISI Hand book No. 1.
For MB 500, @ 0.869 kN/m
Width of flange, bf = 180 mm
Thickness of web, tw = 10.2 mm
Thickness of flange, tf = 17.2 mm
h2 = 37.95 mm

IS A 1 00 m m × 1 0 0 m m × 6 m m
Fillet w eld 6 m m

B e am
IS M B 50 0,
@ 0 .86 9 kN /m

S e ating p la te 18 m m
S tiffen in g plate 1 2 m m

Fillet w eld 1 0 m m

Fig. 11.9 Stiffened welded seat connections

 WELDED BEAM CONNECTIONS 759

The bearing length required

⎛ F ⎞
b = ⎜ – 3 ⋅ h2 ⎟
f t
⎝ b⋅ w ⎠

⎛ 240 × 10 ⋅ 2 ⎞
= ⎜ – 3 × 37 ⋅ 95 ⎟
⎝ 185 × 10 ⋅ 2 ⎠
= 61.455 mm

⎛1 1 ⎞ 1 ⎛ 240 × 1000 ⎞
b<
|⎜ or <
| ×⎜ ⎟<| 63 ⋅ 59 mm
⎝ 2 fb⋅tw ⎟⎠ 2 ⎝ 185 × 10 ⋅ 2 ⎠
∴ The bearing length = 63.59 mm
10 mm clearance is provided between the end of the beam and flange of the
column
Width of seat plate = (63.59 + 10)
= 73.59 mm
Adopt width of a seat plate = 80 mm
Thickness of a seat plate = Thickness of flange of beam
= 17.2 mm

1 80 m m
18 m m

6 1.4 m m

1 80
mm
11 8.6 m m

84 m m 12 84 m m

Fig. 11.10

Provide 18 mm thick and 80 mm long and 180 mm wide seat plate.

Thickness of stiffening plate
Thickness of web of beam = 10.2 mm
Provide 12 mm thick stiffening plate,
The distance of end reaction from outer end of seat plate
⎛ 1 ⎞
⎜ 80 × × 63 ⋅ 59 ⎟ = 48.20 mm
⎝ 2 ⎠
Step 2: Bending moment,
M = (240 × 48.20) =11568 kN-mm
760 DESIGN OF STEEL STRUCTURES

Step 3: Properties of welds

⎛ 2 × 180 × 90 ⎞
y1 = ⎜ ⎟ = 61.4 mm,
⎝ 2 × 180 + 168 ⎠
and y2 = 118.6 mm

⎡ 1 2⎤
Ixx = ⎢16 ⋅ 8 × 6 ⋅ 142 + × 183 + 2 × 18 × (11 ⋅ 86 – 9) ⎥ × 104 = 1898 × 104 mm4
⎣ 12 ⎦
Step 4: Stresses in welds
The vertical shear per mm of weld
⎛ 240 × 1000 ⎞
fsh = ⎜ ⎟ = 454.54 N/mm
⎝ 2 × 180 × 168 ⎠
The horizontal shear per mm of weld

⎛ 11568 × 1000 × 614 ⎞

fsh = ⎜ ⎟ = 374.22 N/mm
⎝ 1898 × 104 ⎠
Since the connection will not allow failure of weld in thrust.
y = 61.4 mm
Step 5: Resultant shear stress per mm
fs = [(454.54)2 + (374.22)2]1/2
= 588.77 N/mm
Step 6: Size of weld. Let the size of weld be
0.7 × s ×1 mm ×110 = 588.77
⎛ 588 ⋅ 77 ⎞
s = ⎜ ⎟ = 7.646 mm
⎝ 0 ⋅ 7 × 110 ⎠
Provide 10 mm fillet welds. ISA 100 mm × 100 mm × 6 mm (ISA 100 100, @
0.092 kN/m) is used at top. A complete welded stiffened seat connection designed
is shown in Fig. 11.9.

11.5 MOMENT RESISTANT WELDED CONNECTIONS

When the ends of beam are subjected to moments and shears, moments resis-
tant welded connections as shown in Fig. 11.11 are used. The end moments of
beam are resisted by top plate and bottom flange. When the moment of beam,
say, on the left side of connection is clockwise, the top plate carries tensile force
and the bottom flange of beam carries compressive force. The forces are equal in
magnitude, and opposite in direction. These forces form a resisting couple. When
the direction of moment is anticlockwise, these elements of connections carry
equal forces of opposite sign. The top plate is butt welded to the flange of col-
umn. The butt weld needs a 5 mm minimum gap between plate and the flange
of column. It requires a backing strip. The backing strip is tack welded to the
 WELDED BEAM CONNECTIONS 761

flange. The length of top plate is kept large. The entire length of top plate is not
welded. A length equal to one-half to one times the width of the plate is kept

B u tt w eld
Fillet w eld

B u tt w eld

S tiffen er plate B u tt w eld

Fig. 11.11

unwelded. This portion of plate is kept unwelded so that if yielding occurs, it

occurs in this, rather than in weld. This provides ductility to the connections.
The size of fillet weld should be such that the length of side welds remains
about one-half to one times the width of plate. In case, the moment resistance
connections are made with the web of column, the top plate is flared. This allows
side welds to the column flange.
The welded seat connections are designed to resist the vertical end reactions
of beam. This welded seat connections are kept unstiffened or stiffened depend-
ing upon the magnitude of end reactions.
Figure 11.12 shows moment resistant welded connections for beams to beams.
In case, the top flanges of two beams to be connected are at the same level of
supporting beam, then a plate is welded as shown in Fig. 11.12 (a). In case, the
top flange of two beams to be connected are not at the same level as that of the
flange of supporting beam, then, the top plates are butt welded to the web as
shown in Fig. 11.12 (b). As an alternative to this a slot may be cut in a channel
to clear the flange of supporting beam and it is connected to the flanges as
shown in Fig. 11.12 (c).
762 DESIGN OF STEEL STRUCTURES

B e am
B e am b utt w e ld

(a )

B u tt w eld

(b )

B u tt w eld

(c)

Fig. 11.12

Example 11.6 A MB 500, @ 0.869 kN/m transmits an end reaction of 130 kN

and an end moment of 120 kN-m to the flange of a column HB 250, @ 0.510 kN/
m. Design the welded connections.
Solution
Design
Top plate
Step 1: Propertion of section
From ISI Handbook No. 1, for MB 500, @ 0.869 kN/m
Depth of beam, h = 500 mm
Width of flange, bf = 180 mm
Thickness of web, tw = 10.2 mm
Thickness of flange, tf = 17.2 mm
Step 2: Thickness of plate
⎛ 120 × 103 ⎞
Force in the top plate = ⎜ ⎟ = 240 kN
⎝ 500 ⎠
Area required for the top plate
⎛ 240 × 1000 ⎞
⎜ ⎟ = 1454.54 mm2
⎝ 0 ⋅ 66 × 250 ⎠
Width of flange of beam
bf = 180 mm
 WELDED BEAM CONNECTIONS 763

Adopt width of the top plate

= 160 mm
Thickness of the top plate required
⎛ 1454 ⋅ 54 ⎞
⎜ ⎟ = 8.08 mm
⎝ 180 ⎠
Provide 160 mm × 10 mm top plate. The top plate is connected to the flange of
column HB 250, @ 0.510 kN/m. 10 mm thick stiffener plates are provided, so
that the flange of the column does not deform.
Step 3: Fillet weld
For 10 mm thick top plate, adopt 8 mm size of weld
Strength of 1 mm fillet weld
(0.7 × 8 × 1 × 110) = 616 N/mm
Length of the fillet weld
⎛ 240 × 1000 ⎞
⎜ ⎟ = 389.61 mm
⎝ 616 ⎠
Provide 120 mm side fillet weld on each side of the top plate and 160 mm end
fillet weld.
The total length of weld provided is 400 mm. The length of unwelded portion
is kept equal to the width of plate (viz., 160 mm).
Therefore, the total length of top plate is 290 mm.
The bottom flange of beam may be directly welded to the flange of column.
The welded unstiffened seat connections for end reaction 130 kN designed in
Example 11.4 are adopted. The connections designed are shown in Fig. 11.13.

B u tt w eld
1 60 m m x 10 m m top p la te

IS M B 50 0 @ 0.8 69 kN /m

IS H B 2 50 @ 0.51 0 kN /m

Fig. 11.13
764 DESIGN OF STEEL STRUCTURES

PROBLEMS
11.1 A LB 300, @ 0.377 kN/m transmits an end reaction of 80 kN to the
flange of a column HB 200, @ 0400 kN/m. The fillet welds are applied
directly on both the sides of web of the beam. Design the connections.
11.2 In Problem 11.1 if two plates are used for connections on both the
sides of web of beam, then, design the welded connections.
11.3 In Problem 11.1, if two framing angles are used for connections on
both the sides of web of a beam, then, design the connections.
11.4 A LB 400, @ 0.569 kN/m transmits and end reaction of 110 kN to the
flange of a column HB 350, @ 0.674 kN/m. Design unstiffened welded
seat connections.
11.5 A HB 450, @ 0.653 kN/m transmits an end reaction of 220 kN to the
flange of a column HB 300, @ 0.630 kN/m. Design stiffened welded seat
connections.
11.6 A MB 300, @ 0.442 kN/m transmits an end reaction of 11 kN and an
end moment of 80 kN-m to the fladge of a column HB 300, @ 0.630 kN/
m. Design the welded connections.
 Chapter
Design of Round

12 Tubular Structures

12.1 INTRODUCTION
The round tubular sections are used as structural components since long. A
large number of tubular structures have been constructed in the past. The
tubular sections have been used for columns, compression members, tension
members, roof trusses, mill buildings, air plane hangers, cross bracings and
beams. The tubular sections are also used for single storey industrial buildings,
warehouses, and shopping centres where long span column free areas are
required. The tubular sections are used to advantages in structures designed
for material handling equipments (e.g., a bridge, derrick and tower cranes) where
weight savings may be very substantial economic consideration. The tubular
sections are effectively used in large space frame, lattice structures for arenas,
stadium exhibition halls where appearance as well as weight become an
important design consideration. The masts and transmission towers are other
examples where tubular sections are utilised effectively. In the past, the use of
tube was hampered because of connection details. A major contribution to the
solution of this problem has undoubtedly been the development of fully automatic
oxyacetylene tube cutting machines, which not only cut tubes to fit flat surfaces
but also cut them to fit cylindrical surfaces such as tube as well. The machine
cuts the tubes to the correct profile with a bevelled edge to simplify the welding
process at the joint.
The round tubular sections have as much as 30 to 40 percent less surface
area than that of an equivalent rolled steel shape. Therefore, the cost of main-
tenance, cost of painting, fire proofing and/or other protective coatings reduce
considerably. The moisture and dirt do not collect on the smooth external sur-
face of the tubes. Therefore, the possibility of corrosion also reduces. The ends
766 DESIGN OF STEEL STRUCTURES

of tubes are sealed. As a result of this, the interior surface is not subjected to
corrosion. The interior surfaces do not need any protective treatment.
The use of round tubular members is becoming more generally adopted for
structures. The tubes are of special interest to the architects from an effective
view point and to the engineers from a structural effectiveness view point.
The tubular sections have more torsional resistance than the other sections
of the equal weights. The tube sections have a higher frequency vibrations under
dynamic loading than the other sections including the solid round one. The
round tubular sections offer less resistance to wind. Some of the codes allow a
one-third reduction in wind load compared to equivalent projected area. IS : 875
–1984 recommends a shape factor of 0.7 for round tubular sections.
The tubular sections are most efficient sections which are adoptable to many
different situation. The tubular sections cannot be surprised in its efficiency by
other sections.

12.2 ROUND TUBULAR SECTIONS

The round tubular sections are manufactured from steel which when analysed
in accordance with the methods in IS : 228, Methods of Chemical Analysis of Pig
Iron, Cast Iron and Plain Carbon and Low Alloy Steels shall show not more
than 0.06 percent sulphur, not more than 0.06 percent phosphorus. The plain
carbon steel tubes as specified in IS : 1161–1963 are used for the structural
purpose. The sizes and properties of steel tubes for structural purpose are given
in Tables 12.1 (a) and (b) as per IS : 1161–1963. These tubes are of following
types :
1. HFW : Hot finishes welded
2. HFS : Hot finished seamless
3. ERW : Electric resistance or induction butt welded.
The hot rolled carbon steel structural tubes are made in round shapes from a
continuous welding or seamless process. In the continuous weld process coils of
steel called skelp are welded and to end to form a continuous band of steel and
passed through furnaces. As the skelp exists from the furnace, it is formed into
a round pipe and welded by pressure. The pipe is then placed into sketch reduc-
ing mill, where, it is brought to the desired diameter and wall thickness.
In seamless tubes, a solid round bar of redetermined size is treated and then
pierced by a mandrel while rotating at high speed. The tube is then placed to
other rolling operations which bring it to the proper diameter and wall thick-
ness.
The structural members consisting of round steel tubes are designed as per
specification given in IS : 806 –1957 (Code of Practice for Use of Steel Tubes in
General Building Construction).
Table 12.1 (a) Sizes and properties of steel tubes for structural purposes

Nominal Outside Class Thickness Weight Area of Internal Surface M.I. Z r r2

bore diameter X-section volume
Ext. Int.
2 3 2
mm mm mm N/m mm mm /m mm /m mm 2 /m m4 cm3 cm cm2
1 2 3 4 5 6 7 8 9 10 11 12 13
15 21.3 Light 2.0 96.2 121 23500 66900 54300 0.57 0.54 0.69 0.47
Medium 2.6 12.1 153 20400 50600 0.68 0.64 0.67 0.45
Heavy 3.2 14.4 182 17400 46800 0.77 0.72 0.65 0.42

20 26.9 Light 2.3 14.1 178 39100 84500 70100 1.36 1.01 0.87 0.76
Medium 2.6 15.7 198 37000 68200 1.48 1.10 0.86 0.75
Heavy 3.2 18.9 238 32000 64400 1.70 1.27 0.85 0.71

25 33.7 Light 2.6 20.1 254 63800 105900 89500 3.00 1.84 1.10 1.22
Medium 3.7 24.2 307 58500 85800 3.60 2.14 1.08 1.18
Heavy 4.0 29.5 373 51900 80700 4.19 2.49 1.06 1.12

32 42.4 Light 2.6 25.7 325 108700 133200 116900 6.46 3.05 1.41 1.99
DESIGN OF ROUND TUBULAR STRUCTURES

Medium 3.2 31.1 394 101800 113000 8.99 4.24 1.36 1.93
Heavy 4.0 38.1 483 91900 108100 8.99 4.24 1.36 1.86
40 40.3 Light 2.9 32.7 414 141900 151700 133500 10.70 4.43 1.61 2.59
Medium 3.2 35.9 453 137900 131600 11.59 4.80 1.60 2.56
Heavy 4.0 44.1 550 127600 126600 13.77 5.70 1.57 2.47

Contd.
767
Table 12.1 (a) Contd.
768
1 2 3 4 5 6 7 8 9 10 11 12 13
50 60.3 Light (1) 2.9 41.4 523 233300 189400 171200 21.59 7.16 2.03 4.13
Light (2) 3.2 45.4 574 228200 169300 23.47 7.18 2.02 4.09
Medium 3.6 50.7 641 221500 166800 25.87 8.58 2.01 4.03
Heavy 4.5 61.7 788 206700 161200 30.90 10.20 1.98 3.92

65 76.1 Light 3.2 58.0 733 381600 239100 219000 48.78 12.8 2.58 6.66
Medium 3.6 64.9 820 372800 216500 54.01 14.2 2.57 6.59
Heavy 4.5 79.7 1010 353600 210800 65.12 17.1 2.54 9.48

80 88.9 Light 3.2 68.1 862 534600 279300 259200 79.21 17.8 3.03 9.19
Medium 4.0 84.3 1070 514000 254200 96.34 21.7 3.00 8.03
Heavy 5.0 103.0 1320 488900 247900 116.04 26.2 2.97 8.83
(where M.I. is moment of inertia, Z is modulus of section, r is the radius of gyration and r2 square of radius of gyration)
DESIGN OF STEEL STRUCTURES
Table 12.1 (b) Sizes and properties of steel tubes for structural purposes

Nominal Outside Class Thickness Weight Area of Internal Surface M.I. Z r r2

Bore diameter X-section volume
Ext. Int.
2 3 2
mm mm mm N/m mm mm /m mm /m mm 2 /m m4 cm3 cm cm2
1 2 3 4 5 6 7 8 9 10 11 12 13
90 101.6 Light 3.6 87.6 1110 699900 319200 296600 133.2 26.2 3.47 12.02
Medium 4.2 97.0 1230 688100 294100 146.3 28.8 3.45 11.98
Heavy 5.0 119.0 1520 659000 287800 177.5 34.9 3.42 11.70
100 114.3 Light 3.6 99 1250 900900 359100 336500 192.0 33.6 3.92 15.33
Medium 4.5 121 1550 870900 330800 234.3 41.0 3.89 15.10
Heavy 5.4 145 1850 841300 335200 274.5 48.0 3.85 14.86

110 127.0 Light 4.5 122 1730 1093600 399000 370700 325.3 51.2 4.33 18.78
Medium 5.0 150 1920 1075100 367600 357.1 56.2 4.32 18.64
Heavy 5.4 162 2060 1060500 865100 382.0 60.2 4.30 18.52

125 139.7 Light 4.5 149 1910 1341700 438900 410600 437.2 62.6 4.78 22.87
Medium 5.0 166 2120 1321200 407500 480.5 68.8 4.77 22.71
DESIGN OF ROUND TUBULAR STRUCTURES

Heavy 5.4 179 2280 1305000 405000 514.5 73.7 4.75 22.58

135 152.4 Light 4.5 164 2090 1615100 478800 450500 572.2 75.1 5.23 27.37
Medium 5.0 182 2320 1592500 447400 629.5 82.6 5.21 27 .19
Heavy 5.4 195 2490 1574800 447900 674.5 88.5 5.20 27.05
150 165.3 Light 4.5 178 2270 1913800 518700 490400 732.6 88.7 5.78 32.27
Medium 5.0 197 2510 1889400 487300 806.6 97.7 5.66 32.07
Heavy 5.4 212 2710 1869900 484700 864.7 105.0 5.65 31.92

Contd.
769
Table 12.1 (b) Contd.
770
1 2 3 4 5 6 7 8 9 10 11 12 13
150 168.3 Light 4.5 181 2320 1993100 528700 500500 777.2 92.4 5.79 33.56
Medium 5.0 201 2570 1968100 497300 855.8 102.0 5.78 33.36
Heavy (1) 5.4 217 2760 1948300 494800 917.7 109.0 5.76 33.21
Heavy (2) 6.3 253 3210 1904000 489100 1053.4 125.0 5.73 32.85

175 193.7 Light 5.0 233 2960 2650400 608500 577100 132.0 136 6.67 44.54
Medium 5.4 250 3190 2627400 574600 141.7 146 6.66 44.36
Heavy 5.9 273 3480 2598700 571600 153.6 159 6.64 44.13

200 219.1 Light 5.0 264 3360 3434000 688300 650000 191.8 176 7.57 57.33
Medium 5.6 294 3760 3894700 653100 214.2 195 7.55 57.02
Heavy 5.9 310 3950 3375100 651300 224.7 205 7.54 56.86

225 244.5 Heavy 5.9 342 4420 4252900 768100 731100 314.9 258 8.44 71.71
DESIGN OF STEEL STRUCTURES
 DESIGN OF ROUND TUBULAR STRUCTURES 771

12.3 PERMISSIBLE STRESSES

The direct stress in axial tension on the net cross-sectional area of tubes shall
not exceed the values of σt given in Table 12.2 as per IS : 806 –1957.

Table 12.2 Permissible axial stress in tension

IS : 1161 Grade σt in N/mm2

St. 35 (Yst. 22) 125.0
St. 40 (Yst. 25) 150.0
St. 55 (Yst. 32) 190.0
Note. The values of permissible stress in axial tension have been converted in S.I.
units
The direct stress in compression on the cross sectional area of axially loaded
steel tubes shall not exceed the values of σc given in Table 12.3 as per IS : 806 –
1957. (Note : These values have been converted in S.I. units).

Table 12.3 Permissible axial stress in compression

σc in N/mm2

⎛l⎞
Slenderness ratio ⎜ ⎟ IS : 1161 IS : 1161 IS : 1161
⎝r⎠
Grade St. 35 Grade St. 40 Grade St. 55
(Yst. 22) (Yst. 25) (Yst. 32)
0 125.0 150.0 190.0
10 121.7 144.8 182.1
20 117.5 140.0 175.0
30 113.1 135.2 167.9
40 108.8 130.3 161.0
50 104.6 125.5 153.9
60 100.2 120.7 146.8
70 97.0 115.5 137.5
80 92.9 108.8 126.3
90 87.6 103.3 112.8
100 81.4 91.0 98.9
110 74.5 81.3 86.9
120 67.4 72.1 75.8
130 60.3 63.8 65.5
140 54.0 56.5 58.4
150 49.0 50.3 51.7
160 43.2 44.3 45.0
170 38.1 39.2 39.8
180 33.9 34.8 35.3
Contd.
772 DESIGN OF STEEL STRUCTURES

Table 12.3 Contd.

190 30.4 31.1 31.5
200 27.1 27.8 28.0
210 24.3 24.9 25.0
220 21.9 22.5 22.7
230 19.8 20.4 20.5
240 18.0 18.5 18.7
250 16.2 16.7 16.7
300 10.6 10.6 10.6
350 7.1 7.1 7.2
Notes. 1. Intermediate values may be obtained by linear interpolation.
2. The values of allowable stress in axial compression has been found
from secant formula.
The tensile and compression bending stress in the extreme fibres of the tubes
shall not exceed the values ab given in Table 12.4 as per IS : 806–1953.
(Note : These values have been converted in S.I. units).
Table 12.4 Permissible axial stress in extreme fibres in tension and compression

IS : 1161 Grade σt in N/mm2

St. 35 (Yst. 22) 140.0
St. 40 (Yst. 25) 155.0
St. 55 (Yst. 32) 205.0
The maximum shear stress in a tube calculated by dividing the total shear by
an area equal to half the net cross-sectional area of tube shall not exceed the
value of σb given in Table 12.5 as per IS : 806 –1957. The net cross-sectional
area shall be derived by deducting areas of all holes from the gross cross-sec-
tional area (Note : These values have been converted in S.I. units).

Table 12.5 Permissible axial stress in tension

IS : 1161 Grade σt in N/mm2

St. 35 (Yst. 22) 190.0
St. 40 (Yst. 25) 110.0
St. 55 (Yst. 32) 135.0
The average bearing stress on the net projected area of contact shall not
exceed the values σb given in Table 12.6 as per IS : 800 –1987.
(Note : These values have been converted in S.I. units).
Table 12.6 Permissible axial stress in tension

IS : 1161 Grade σt in N/mm2

St. 35 (Yst. 22) 170.0
St. 40 (Yst. 25) 190.0
St. 55 (Yst. 32) 250.0
 DESIGN OF ROUND TUBULAR STRUCTURES 773

The members subjected to both bending and axial stresses shall be so propor-
⎡⎛ fa ´ ⎞ ⎛ fb ´ ⎞ ⎤
tioned that the quantity ⎢⎜ ⎟+⎜ ⎟ ⎥ does not exceed unity.
⎣⎝ fa ⎠ ⎝ fb ⎠ ⎦
where, fa´ = Calculated axial stress
= Axial load divided by appropriate area of member
fa = Permissible stress in member for the axial load
fb´ = Calculated bending stress in the extreme fibre
fb = Permissible bending stress in the extreme fibre
When the bending occurs about both the axes of the members, then fb shall be
taken as
fb = [(fbx)2 + (fby)2]1/2 ...(12.1)
where fbx and fby are the two calculated unit fibre stresses.The equivalent stress,
σe due to co-existing bending shear stresses shall not exceed the value given in
Table 12.7 as per IS : 806 –1957.
(Note : These values have been converted in S.I. units).

Table 12.7 Permissible axial stress in tension

IS : 1161 Grade σt in N/mm2

St. 35 (Yst. 22) 190.0
St. 40 (Yst. 25) 228.0
St. 55 (Yst. 32) 290.0
The equivalent stress σb is obtained from the following formula
σe = [σb2 + 3τs2]1/2 ...(12.2)
where, σb is the calculated bending stresses at the point under consideration
and τS is the calculated actual shear stress at the point under
consideration.

12.4 TUBE COLUMNS AND COMRESSION MEMBERS

The round tubular sections provide the most efficient cross-sectional shape for
the columns and compression members having lateral restraint in all directions

d = 2R
Fig. 12.1
774 DESIGN OF STEEL STRUCTURES

normal to the axis of member. The diameter of such a member should be as

large as possible with the additional requirement that d/t (see Fig. 12.1)
should be small enough to assure that pressure failure by local buckling will not
occur.
The local buckling strength of very short perfect tubes depends primarily on
L/d ratio. In case of the medium tubes, the local buckling strength is prima-
⎛d⎞
rily a function of ⎜ ⎟ ratio. For long tubes, the local buckling failure should be
⎝t⎠
avoided so that the general column buckling is the primary problem. The
medium length tubes are most frequently used in practice and the theoretical
local buckling stress based on small deflection theory for solution of Donell’s
equation
C ⋅E
σb = ...(12.3)
⎛R⎞
⎜ ⎟
⎝t ⎠
where, C = [3(1 – v2)]–1/2 ( ≈ 0.6 when v = 0.3) and
R and t are defined in Fig. 12.1. ...(12.4)
The medium length tube solution defined by Eq. 12.3 is applied for value of Z
greater than 2.85, whereas Z is given by Batdorf
1
L2 ⎡(
⎣ 1 – v )⎤⎦ 2
2
Z = ...(12.5)
R ⋅t
where, Z is the limiting constant for medium length tubular, columns.
⎛d⎞
The experiments show that as ⎜ ⎟ increases, failure usually occurs at a
⎝t⎠
stress increasing lower than that obtained from Eq. 12.3. This is because re-
sidual stresses and/or small irregularities in shape have an extremely detri-
mental effect on the compressive strength of round tubular section. The value
of constant C depends to some degree on the tube manufacturing process and
should be based on the recommendations made by the particular manufacturer.
In the design of steel tubular columns and compression members the atten-
tion must be paid to two points : crinkling and heat treatment. The yield strength
of mild steel may be considerably reduced by any heat treatment which it re-
ceives such as welding. On this account, the precautions should be taken to
prevent heat treatment or the strength must be taken as that of the annealed
material.
For the round tubular columns, and compression members, the effective length
of member is adopted as given in Table 3.1, Chapter 3. The limits specified for
maximum slenderness ratio in Table 3.2 are also applicable for the tube members.
In addition to this, the members should satisfy the minimum thickness
requirement. The steel tubes used for construction exposed to the weather shall
be not less than 4 mm thick and for construction not so exposed shall be not less
 DESIGN OF ROUND TUBULAR STRUCTURES 775

than 3.2 mm thick when the structures are not readily accessible for
maintenance, the minimum thickness shall be 5 mm. Notwithstanding the above,
the thickness of tube subjected to compressive stress shall not be less than that
given by the formula
t = 0.865 [D]1/3 ...(12.6)
where, D is the outside diameter of tube in mm. The ends of the members are
sealed.

12.5 CRINKLING
When a steel tubular member is subjected to compression, then, the tube may
crinkle (i.e., the walls of the tubes may cave in and form folds after the manner
of concertina). These folds may be circular, oval or polygonal and they may
occur after or before the longitudinal stress reaches the yield point.
Professor R.V. Southwell investigated the matter and deduced the following
formula analytically
1
t ⎡ m2 ⎤2
p = E⋅ ⎢ ⎥ ...(12.7)
R ⎣ 3 (m – 1 ) ⎦
2

where, p = Stress causing the collapse

t = Thickness of the tube
R = Mean radius of tube
1/m = Poisson’s ratio.
⎡⎛ K 2 – 1 ⎞ ⎤
The stress which will produce the lobbed failures is ⎢⎜ 2 ⎟ ⎥ times that
⎣⎝ K + 1 ⎠ ⎦
given by Eq.12.7, where K is the number of lobes formed. This formula applies
only if p is less than the yield stress. Thus, for a mild steel tube having a yield
⎛t ⎞
point of 315.0 N/mm2 the formula will only apply when ⎜ ⎟ ratio is less than
⎝R⎠
⎛ 1 ⎞
about ⎜ ⎟.
⎝ 400 ⎠

12.6 TUBE TENSION MEMBERS

The effectiveness of a member subjected to axial tension depends on its cross-
sectional area, grade of steel and its method of connection. Therefore, a round
tube or any rolled steel shape with the same net cross-sectional area and the
same material would have the same equivalent resistance to the force. The
tubular section do not have any advantage as tension member. The tubular
sections have higher cost of production than other rolled steel sections.

12.7 TUBULAR ROOF TRUSSES

The design of roof trusses has been discussed in Chapter 9. The roof trusses are
subjected to external loads. The external loads. The external loads, in general,
776 DESIGN OF STEEL STRUCTURES

are applied at the joints. Therefore, the different members of the roof trusses
are subjected to axial compressive and tensile forces only. The tubular
compression members have been discussed in Sec. 12.4. The tubular tension
members have been discussed in Sec. 12.6. The members of tubular trusses are
generally joined by welding. A tubular roof truss is shown in Fig. 12.2.

G H

E L F

A B J K C D
Fig. 12.2

In order to facilitate the erection of the trusses, the complete truss is splitted
in different parts (viz., AEGB, DFHC and JK). These parts are then assembled
after the erection. These parts are then connected by bolts and nuts. The
arrangements for bolt connections made are as shown in Fig. 12.2 at J, K and L.
The rectangular pieces of plates are welded to the projected end of the parts of
the truss. The bolts are placed in the holes in both the plates and then, the nuts
are tightened. The different parts of the truss are then assembled and a complete
truss is built. The rectangular pieces with holes are also welded with the principal
rafters at the joints and these pieces are kept projected as shown in Fig. 12.2 at
E, K, G and H. These projected pieces are used to provide tubular purlins. The
tubular purlins are provided with holes at the ends with internal threads. The
tubular purlins are placed from one side. The studs are used for the connections
of the purlins. The studs are placed through the holes of projected plates and
tightened in the internally threaded ends of the purlins. The purlins are placed
and fixed with the other ends of the studs. Similar arrangements are also made
for tubular bracings with the bottom tie of the truss as shown in Fig. 12.2 at B
and C.
In addition to the type of truss shown in Fig. 12.2, the tubular trusses of
various other types are also made with suitable arrangements for the erection.

12.8 JOINTS IN TUBULAR TRUSSES

The welding is generally adopted for connections in tubular steel construction.
The welded connections provide rigid joints. The rigid joints results in saving in
materials and greater overall economy. The actual condition of rigidity should
be taken into consideration while designing these types of joints.
The welded connections in the tubular structures may, however, be designed
on a simple design basis comparable with IS : 800. In such cases, the secondary
stresses may be neglected in the design of roof trusses except where the axes of
 DESIGN OF ROUND TUBULAR STRUCTURES 777

the members do not meet at a point where there is such eccentricity, the effects
of the eccentricity should also be considered.
In joints in compression members, the ends of the members are faced for
complete bearing over their whole area. The welding and joining materials are
kept sufficient to retain the members accurately in place and to resist all forces
other than direct, compression including those arising during transit, unloading
and erection.
Where such members are not faced for complete bearing, the welding and
joining material shall be sufficient to transmit all the forces to which the joint is
subjected. Wherever possible, the joints shall be proportioned and arranged so
that the gravity axes of the members and the joints are in line, so as to avoid
eccentricity.
Initially, the tube members in tubular trusses were welded to the gusset
plates as shown in Fig. 12.3.
The welded joint with the gusset plates in the tubular trusses were found to
be unsatisfactory for severe load conditions. The difficulty with this type of con-
nection arises from the concentrated stress flow between the tube walls and the
gusset plate even though the gusset plate is capable of transferring the forces
effectively.
As far as possible, tube connections between the members are made directly
tube to tube without gusset plates and other attachments. A weld connecting
two tubes end to end shall be full penetration butt weld. The effective throat
thickness of the weld shall be taken as the thickness of the thinner part joined.

Fig. 12.3

A weld connecting the end of one tube (branch tube) to the surface of another
tube (main tube) with their axes at an angle of not less than 30° shall be one of
the following :
I. A butt weld throughout.
II. A fillet weld throughout.
III. A fillet-butt weld, the weld being a fillet weld in one part and a butt
weld in another with a continuous change from the one form to the
other in the intervening portions.
778 DESIGN OF STEEL STRUCTURES

Type I may be used whatever the ratio of the diameter of the tubes joined
provided complete penetration is secured either by the use of the backing
material, or by deposing a sealing run of metal on the back of joint or by some
special method of welding. Type I weld is shown in Fig. 12.4. When Type I is not
employed then, Type II should be used whether the diameter of branch tube is
1
less than rd of the diameter of the main tube. Type II weld is shown in Fig.
3
1
12.4. When the diameter of branch tube is equal or greater than rd of the
3
diameter of the main tube, then, the Type III weld as shown in Fig. 12.4 is used.
For the purpose of stress calculation, the throat thickness of the butt weld
portion shall be taken as the thickness of the thinner part joined. The throat
thickness of the fillet weld and fillet-butt weld is taken as the minimum effective
throat thickness of the fillet or fillet butt weld.
A weld connecting the end of one tube to the surface of another with the axes
of the tubes intersecting at an angle less than 30° shall be permitted only if
adequate efficiency of the junction has been shown.
Typ e I butt w e ld

DB

e m a x = 1 (D m – D B )
2
Typ e II F ille t w e ld

DB

e m a x = 1 (D m – D B )
2 3
Typ e III fille t b utt w e ld

e m a x = 1 (D m – D B )
2
Fig. 12.4
 DESIGN OF ROUND TUBULAR STRUCTURES 779

The length of curve of intersection of the tube with another may be

determined from the following formula :

L = (a + b + 3 a2 + b2 ) ...(12.8)
where
⎛d⎞
a = ⎜ ⎟ . cosec θ
⎝2⎠
For intersection with a tube :

⎡ 2
⎛d⎞ ⎤
⎢ 3 – ⎜ ⎟ ⎥
d ⎝D⎠ ⎥
b = ⎢ ⋅ ...(12.9)
⎢3 ⎛d⎞ ⎥
2
⎢ 2–⎜ ⎟ ⎥
⎢⎣ ⎝ D ⎠ ⎥⎦

⎛d⎞
Alternatively, b = ⎜ ⎟ .φ ...(12.10)
⎝4⎠

⎛φ⎞ ⎛d⎞
where, sin ⎜ ⎟ = ⎜ ⎟ ...(12.11)
⎝2⎠ ⎝D⎠
where φ is measured in radians.
For intersection with flat plates :
⎛d⎞
b = ⎜ ⎟ ...(12.12)
⎝2⎠
where θ = Angle between branch and main tubes
d = Outside diameter of branch tube
D = Outside diameter of main tube.
The above parameters of tubes are shown in Fig. 12.5.

12.9 DESIGN OF TUBULAR BEAMS

The tensile and compressive stresses in bending in the extreme fibres of tubes
shall not exceed the values given in Table 12.4. The maximum shear stress in
tubes in flexure calculated by diving the total shear by an area equal to half the
net cross-sectional area of the tube shall not exceed the values given in Table 12.5.
When the tubular steel beam rests on abutment or other supporting member,
it shall be provided with a shoe adequate to transmit the load to the abutment
and to stiffen the end of tube.
Where a concentrated load is applied to a tubular member transverse to its
length or the effect of load concentration is given by the intersection of triangular
truss member, consideration shall be given to the local stresses set up and the
method of application of the load and stiffening shall be provided as necessary
to prevent the local stresses from being excessive. The increase in the intensity
780 DESIGN OF STEEL STRUCTURES

of local bending stresses caused by concentrated loads particularly marked if

either the diameter of connected member or the connected length of a gusset or
the like is small in relation to the diameter of the tubular member to which it is
connected.

Fig. 12.5

The deflection of a member shall not be such as to impair the strength,

efficiency or appearance of the structure or lead to damage to fittings and
1
finishings. Generally, the maximum deflection should not exceed of the
325
span for simply supported beams. This requirement may be deemed to be satisfied
⎛D⎞
if the bending stress in compression or in tension does not exceed 3153 ⎜ ⎟ N/
⎝ l ⎠
mm2, where D is the outer diameter of the tube in mm and l is the effective
length of the beam in mm.

12.10 DESIGN OF TUBULAR PURLINS

The requirement mentioned in Sec. 12.9 regarding limiting deflection may be
waved in the design of simple tubular purlins provided the following require-
ments are satisfied:
1. Name of end fixing Simply Effectively
supported continuous
2. Maximum value of section
modulus (mm3)
⎛ WL ⎞ ⎛ WL ⎞
(a) IS : 1161 grade St. 35 (Yst. 22) ⎜ ⎟ × 103 ⎜ ⎟ × 103
⎝ 11200 ⎠ ⎝ 16800 ⎠

⎛ WL ⎞ ⎛ WL ⎞
(b) IS : 1161 grade St 40 (Yst. 25) ⎜ ⎟ × 103 ⎜ ⎟ × 103
⎝ 13230 ⎠ ⎝ 19840 ⎠
 DESIGN OF ROUND TUBULAR STRUCTURES 781

3. Minimum outside diameter

IS : 1161 grade St. 35 or
⎛ L ⎞ ⎛ L ⎞
St. 40 (mm) (Yst 32) ⎜ ⎟ × (10) ⎜ ⎟ × (10)
⎝ 40 ⎠ ⎝ 70 ⎠
where, W is the total distributed load in N on the purlins arising from dead load
and snow load excluding wind, and L is the distance in mm between the centres
of the steel principals or other supports.
A purlins is considered as effectively continuous at any intermediate point of
support, if it is actually continuous over that point or it has there a joint able to
⎛ WL ⎞
provide a fixing moment of not less than ⎜ ⎟.
⎝ 2 ⎠
Example. 12.1 A tubular column consists of IS : 1161 grade St. 35 steel. The
column is hinged at both the ends. The outside diameter of tube is 219.1 mm.
The weight of 1 metre length of tube is 310 N. The length of column is 4.5 metre.
Determine the safe load carrying capacity of the column.
Solution
Step 1: Slenderness ratio
The column is 4.5 m long and it is hinged at both ends. The effective length of
column
l = (4.5 × 1000) = 4500 mm
The outside diameter of tube is 219.1 mm, the weight of 1 metre length of
tube is 310 N. The column belongs to heavy class. The radius of gyration of the
tube from IS : 1161–1963.
r = 75.4 mm
The slenderness ratio of the column

l ⎛ 4500 ⎞
= ⎜ = 59.68 |> 180
r ⎝ 75.4 ⎟⎠
Step 2: Allowable stress in axial compression
The steel tube is IS : 1161 grade St. 35. The maximum allowable stress for
⎛l⎞
⎜ ⎟ = 59.58 from IS : 806 –1957
⎝r⎠
σc = 101.282 N/mm2
The cross-sectional area of the tube column from IS : 1161–1963
A = 3950 mm2
Step 3: Safe load carrying capacity of the column
⎛ 101 ⋅ 282 × 3950 ⎞
P = ⎜ ⎟ = 400.064 kN
⎝ 1000 ⎠
Example 12.2 The principal rafter in a round tubular truss carries a
maximum force of 84 kN. A tension member meeting at right angles to the
782 DESIGN OF STEEL STRUCTURES

principal rafter carries a force of 20 kN. Design the member using IS : 1161
grade St. 35 steel for the tube. The panel length along the principal rafter is
1.80 m.
Solution
Step 1: Slenderness ratio
The maximum force in the principal rafter of a tubular truss
Pc = 84 kN
Assuming the maximum allowable stress for the tubular compression member
σc = 800 N/mm2
The cross-sectional area required for the column

⎛ 80 × 1000 ⎞
A = ⎜ ⎟ = 1050 mm2.
⎝ 80 ⎠
From IS : 1161–1963, provide a tube of outside diameter 88.9 mm, and 84.3
N/m weight. The properties of this tube member are as follows:
Cross-sectional area, A = 1070 mm2
Radius of gyration, r = 30 mm
Thickness of tube, t = 4 mm
Internal diameter of the tube = (88.9 – 2 × 4) = 80.9 mm
Pannel length of principal rafter = 1.80 m
The effective length of the member
l = (0.85 × 1.8 × l000) = 1530 mm
The slenderness ratio of tube member

l ⎛ 1530 ⎞
= ⎜ ⎟ = 51 |
> 180
r ⎝ 30 ⎠
Step 2: Allowable stress in axial compression
The tube consists of IS : 1161 grade St. 35 steel. The allowable stress for tube
column from IS : 806 –1967
σc = 10416 N/mm2
Step 3: Safe load carrying capacity of member

⎛ 1070 × 104 ⋅ 16 ⎞
Pc1 = ⎜ ⎟ = 111.45 kN > 14 kN
⎝ 1000 ⎠
Hence, the design is satisfactory.
The thickness of tube 4 mm is equal to minimum thickness required for
exposed situation. Hence satisfactory. Provide tube of outside diameter 88.9
mm and 86 5 N/m weight.
Design of tension member
The allowable stress in steel IS : 1161 grade St. 35 in tension
 DESIGN OF ROUND TUBULAR STRUCTURES 783

σa= 125 N/mm2

Since, the welding is used for the joint, the net area of tube is equal to gross
area. Therefore, the area required for the tube

⎛ 20 × 1000 ⎞
At = ⎜ ⎟ = 160 mm2
⎝ 125 ⎠
Provide tube of 26.9 mm outside diameter and 14.1 N/m weight. The cross-
sectional area provided is 178 mm2 > 160 mm2.
Example 12.3 Design the welded joint for the tubular truss members in
Example 12.2.
Solution
Step 1: Length of curve of intersection
The principal rafter is the main tube member. The outside diameter of main
tube from Example 12.2, D is 88.9 mm.
The outside diameter of branch tube (i.e., tension member)
d = 26.9 mm
The tension member is meeting the principal rafter at right angles. There-
fore, the angle between main tube and branch tube
θ = 90°

⎛d⎞ ⎛ 26 ⋅ 9 ⎞
∴ a = ⎜ ⎟ . cosec θ = ⎜ × 1 ⎟ = 13.45 mm
⎝2⎠ ⎝ 2 ⎠
From Eq. 12.9

⎡ 2
⎛d⎞ ⎤
⎢ 3–⎜ ⎟ ⎥
d ⎝ D ⎠ ⎥ mm
b = ⎢ ×
⎢3 ⎛d⎞ ⎥
2
⎢ 2–⎜ ⎟ ⎥
⎣⎢ ⎝ D ⎠ ⎦⎥

⎡ 2
⎛ 26 ⋅ 9 ⎞ ⎤
⎢3 – ⎜ ⎟ ⎥
269 ⋅ 3 ⎢ ⎝ 88 ⋅ 9 ⎠ ⎥ = 13.68 mm
or b = ×
3 ⎢ ⎛ 26 ⋅ 9 ⎞ ⎥
2
⎢2 – ⎜ ⎟ ⎥
⎢⎣ ⎝ 88 ⋅ 9 ⎠ ⎥⎦

The length of curve of intersection from Eq. 12.8,

L = a + b + 3(a2 + b2)1/2
or L = (13.45 + 13.68 + 3 [(13.452 + 13.682)]1/2
= 84.67 mm
Let t be the throat thickness of fillet weld. The allowable stress in shear in
weld from IS: 816 –1969
784 DESIGN OF STEEL STRUCTURES

fs = 110 N/mm2
Force to be transmitted is 20 kN. Therefore, throat thickness required

⎛ 20 × 1000 ⎞
t = ⎜ = 2.147 mm
⎝ 110 × 84 ⋅ 67 ⎟⎠
The size of weld required

⎛ 2 ⋅ 147 ⎞
s = ⎜ × 10 ⎟ = 3.06 mm
⎝ 0 ⋅7 ⎠
Provide 4 mm fillet weld for the joint.

PROBLEMS
12.1 Two members meeting at a joint at 30° inclination in a tubular truss
carry 120 kN compression and 36 kN tension. The length of compres-
sion member from centre to centre of joint is 2 m. The members are
made of IS : 1161 grade St. 35 steel. Design the members.
12.2 Design the welded joint for the tubular truss in Problem 12.1.
13. Design of Timber Structures
 Chapter
Design of Timber

13 Structures

13.1 INTRODUCTION
The wood is one of many building materials used in building construction and
various types of structures. The wood is an organic material. It is generally
used in its natural state. When the wood is used as a structural material, it is
called timber. The timber is used for temporary structures mainly because of
its low cost. The temporary structures are those structures which are dismantled
immediately after their use. The concrete form work, scaffolding, shuttering,
and structures of exhibitions are the examples of temporary structures. The
timber is also used for the permanent structures. The timber is used for various
structural components such as beams, stringers, purlins, columns etc. The
various types of timber structures are roof trusses, electric transmission posts,
trestles and other types of frames. The various types of timbers which are used
for structural purposes are teak, sal, deodar, rosewood etc. In addition to these,
there are many other types of timbers which are also used for structural purposes.
There are eighty-seven varieties of timbers which are used for structural
purposes. The life of timber structure is long if it is maintained either
continuously dry or wet. The life of a timber structure is short if it remains
alternately wet and dry.
All the structural members, assembles or framework in a building in
combination with the floors, walls and other structural parts of the building
shall be strong (viz., capable of sustaining the whole dead and superimposed
loads including other types of loads referred to IS : 875 –1984, with due stability
and stiffness without exceeding the limits of stress specified.

13.2 DEFFECTS IN TIMBER

Any irregularity which occurs on or in a specimen of a timber and affects its
strength or durability or appearance is known as defect. The timber is a natural
788 DESIGN OF STEEL STRUCTURES

product. It is seldom entirely free from defects. The defects in timber may be
classified in two following groups :
1. Natural defects 2. Other defects.

13.2.1 Natural Defects

There are some factors which influence the growing tissues of a living tree. As
a result of this, the defects are developed. Such defects are called natural defects.
The natural defects are described below.
(i) Knots. When the portion of a branch or a limb embedded in the body of a
tree is cut, then knot is formed. Sometimes lower branches of a tree die and fall
down because of insufficient supply of sunlight. As a result of this, knot is formed
in the body of a tree as shown in Fig. 13.1. Ordinarily, the knots are composed

K n ot

Fig. 13.1

of inferior wood than the normal wood. In the vicinity of knots, the grains are
irregular and distorted. The knots reduce the strength of timber. The reduction
in the strength of timber depends upon size, position and type of knots.
The strength of beams is reduced considerably, when the knots are situated
in the bottom of a beam (the portion of beam in tension). When the knots are
situated near the top of a beam (the portion of beam in compression) then it has
less weakening effects. Knots have still less weakening effects when these occur
near the centre of depth. The strength of timber is considerably decreased in
tension because of knots. The strength of timber reduces to a small extent in
compression because of knots. The knots do not affect the stiffness of timber
appreciably.
(ii) Wane. A wane is bark or lack of wood on the edge or corner of a piece of
timber. It is shown in Fig. 13.2. Wane affects the strength of a timber member
only because the cross-sectional area is reduced. It can be avoided by properly
squaring the edge.

13.2.2 Other Defects

Some defects develop because of the activity of external agents and subsequent
treatment of sawn timber. As a result of faulty seasoning techniques, permanent
distortion of timber may take place. The permanent distortions result in various
types of warping and rupture of tissues. These types of defects are described
below.
(i) Warping. The warping is caused as a result of permanent distortion of
timber. The warping includes cupping, twisting and bowing. When the warping
 DESIGN OF TIMBER STRUCTURES 789

takes place across the width of a timber board, then it is known as cupping.
When the warping takes place in spiral form in a longitudinal direction, then it
is known as twisting. The twisting in extreme cases renders the timber useless.
W a ne

C h eck

Fig. 13.2 Fig. 13.3

When the warping takes place in the form of sagging from one end to the other
along the longitudinal direction, then it is known as bowing. The warping reduces
the strength of timber. The applied loads do not remain parallel with or
perpendicular to the grain. This type of defect is most serious in long columns,
loaded parallel to the grains.
(ii) Rupture of tissues. The rupture of tissues takes place as a result of
permanent distortion of timber. The rupture of tissues includes checks, shakes
and splits. A check is a crack or lengthwise separation of the timber. It usually
occurs across the annual rings. It is shown in Fig. 13.3. During drying from
outside towards centre, the outer fibres shrink more than the inner fibres. As a
result of this cracks develop across the annual rings.
A shake is separation along the grains of the timber. It usually occurs between
and parallel to the annual rings, as shown in Fig. 13.4. The shakes do not develop
because of drying. The shakes originate because of careless felling. When the
living trees are felled, the internal stresses are released which result in
separation along the grains.

S p lit

S h ake

Fig. 13.4 Fig. 13.5

A split is separation of the timber which extends from one face to the other.
When it occurs at the end, it is known as end split. It is shown in Fig. 13.5.
The rupture of tissues (checks, shakes and splits) reduce the resistance of
timber against shear. The strength of a beam is reduced considerably, if these
occur in a horizontal plane and specially at mid-height. At mid-height of a beam,
790 DESIGN OF STEEL STRUCTURES

the shear stress is more than those at other heights. These have little effect,
when these occur in a vertical plane.

13.3 DECAY OF TIMBER

The disintegration of timber substance due to the action of timber destroying
agents is known as decay. The various timber destroying agents are fungi, marine
borers, and insects. The fungi is most destructive. The fungi develops in damp
places. The fungi is a kind of plant life. It grows in the form of hair like threads.
The fungi travels through the timber. It renders the timber ultimately to powder.
When the timber is maintained either completely dry or completely saturated,
then, it lasts long. If the timber remains alternately wet and dry, and used in
dark, damp and unventilated conditions, then it deteriorates very soon. If timber
is used in ground it also deteriorates soon. The timber structures which are
used for marine works, are situated in sea-water. The marine borers attack
these structures. They bore into the timber structures and reduce or entirely
destroy the strength of timber. The decay of timber also takes place by rot.
There are two types of rots, dry rot and wet rot. When the timber is used in
unventilated places where air is confined, then, the decay of timber takes place
by dry rot. Wet rot attacks living trees. The timber also deteriorates by the
attack of white ants or termites. The white ants get into structures and destroy
the timber.

13.4 PRESERVATION OF TIMBER

The preservation of timber is essential to avoid the decay of timber. The pores
of timber are filled up by preservatives. The preservatives exclude moisture
from penetrating below the surface. The various antiseptic chemical compounds
are injected to fill up the pores of timber. The coal tar creosote is most important
preservative. The coal tar creosote is also known as creosote oil or dead oil of
coal tar or creosote coal tar solutions. The composition of coal tar varies widely.
The coal tar creosote used for preservation of timber should conform to IS : 218
–1952. The process of preservating the timber with coal tar creosote is known
as creosoting. The timber is thoroughly impregnated with coal tar creosote. The
creosoting is specially useful for timber for exterior use, for example railway
sleepers, piles, poles etc.
The coal tar is also a good preservative. It is less effective than the coal tar
creosote. The coal tar is more suitable for surface application. The coal tar is
applied hot. All timbers placed in contact with the masonry should be well tarred.
The mercuric chloride is also used as preservative for timber. It is injected into
the pores of timber. The mercuric chloride is used in the form of corrosive
sublimate, one part of sublimate to 150 parts of water. It prevents the destruction
of timber by ants and sea-worms. The mercuric chloride is also effective against
dry rot. The process of preserving the timber by mercuric chloride is known as
kyanizing. Sodium flouride, zinc chloride and copper sulphate are also used as
timber preservatives. When the timber is thoroughly impregnated with copper
 DESIGN OF TIMBER STRUCTURES 791

sulphate, then it is prevented from dry rot. Copper sulphate is not suitable for
timber exposed to the action of water. It does not prevent timber from white
ants.
A new preservative, called Ascu has been developed at the Forest Research
Institute and College, Dehradun. It is made up of three chemicals proportioned
as under:
One part by weight of As2O5.2H2O
Three parts by weight of CuSO4.5H2O
Four parts by weight of K2Cr2O7.
Ascu is available in the powder form. A solution if prepared by mixing six
parts by weight of the powder in a hundred parts by weight of water. The solution
is odourless. The wood treated with it can be painted, varnished, polished or
waxed. Ascu solution can be applied or sprayed on in two coats or the wood
pieces can be soaked in the solution tank or the wood pieces can be impregnated
with the solution under pressure. The method to be adopted depends on the
degree of immunity required, and the nature of wood. The last method is most
effective.
When the untreated timber is used contact with the masonry foundations,
then, termite shields are provided. These are provided at the top of masonry
foundation and below the timber work as shown in Fig. 13.6 as per IS : 883–
1961. The termite shield is provided by galvanized iron sheet or copper sheet

Tm b er
m em b er
G .I. o r
cop pe r
te rm ite
50 m m
shield
50
m

4 5°
m

M aso nry
fo un da tion

Fig. 13.6 Termite shield

not less than 0.45 mm thick. The galvanized iron sheets are laid on a damp
proof course. When the copper sheets are used, damp proof course is not
necessary. It prevents termite and white ants to enter the timber structures.
792 DESIGN OF STEEL STRUCTURES

13.5 SEASONING OF TIMBER

The seasoning is the process of either expelling or drying up the sap remaining
in the green timber. Seasoning of timber is done by exposing it in air, or placing
it in water or in hot air. Seasoning of timber by exposing it in air, is known as
natural seasoning. In natural seasoning, bulks of timber are stacked in layers
under cover. Free circulation of air is allowed to pass around each bark. Natural
seasoning takes considerable time. Seasoning of timber by placing it in water is
known as water seasoning. In water seasoning, timber is placed in water for 15
days. It is preferable to place the timber in streams. Timber is then stacked
under cover. It is allowed to dry thoroughly. The sap of timber is diluted and
carried away by water. It reduces the time necessary to season, when stacked.
Seasoning of timber by keeping it in hot air is known as dessicating seasoning.
Timber is placed in chambers. Hot air is circulated around the timber. This
method also reduces the time necessary for seasoning when stacked. The strength
of timber increases because of seasoning. Timber becomes more durable and
stiff than green timber.

13.6 GRADING OF TIMBER

The method of designating the quality of piece of timber is known as grading.
While grading the timber, the defects in timber, their size, their number, and
their locations are taken into consideration. The effect of density of timber and
effect of slope of grains of timber are also taken into account while grading the
timber. The timber is graded into three categories, selected timber, standard
timber and common timber. The cut sizes of structural timber is graded after
seasoning into three grades, selected grade, grade I, and grage II in IS : 883–
1994. The selected timber is either free from the defect or it has minimum defects.
The standard timber has defects within specified limits. The common timber is
inferior to standard timber. The safe working stresses in compression, shear
and bending for the selected limber are 1.16 times the safe working stresses for
timber of the standard grade. The safe working stresses for timber of the common
grade in compression, shear and bending are 0.84 times the safe working stresses
for timbers of the standard grade.
All the grades of timber with the following defects are prohibited for the
structural use:
(a) The timber with loose grains, splits, compression wood in coniferous
structural timber, heartwood rot, saprot and warp.
(b) The worm holes made by powder post beetles and pitch pockets.
The following defects are permissible for all the grades of timber :
(a) The wanes are permitted provided they are not combined with knots
and the reduction in strength on account of the wanes is not more than the
reduction with the maximum allowable knots. The wanes may also be permitted
provided there is no objection to its use as bearing area, nailing edge and affects
general appearance.
(b) The worm holes other than those due to powder post beetles located
 DESIGN OF TIMBER STRUCTURES 793

and grouped to reduce the strength of timber shall be evaluated in the same
way as the knots.
(c) All other defects which do not affect any of the mechanical properties of
timber.
In addition to the permissible defects mentioned above, the defects listed in
Table 13.1 (as per IS : 883–1994) are permitted for the selected grade, grade I
and grade II timber for structural purposes.
Table 13.1 Defects permissible for different grades of timber for structural purposes (As
per IS : 883–1994)

Selected I Grade Grade I Grade II

B C D θ
C1 C2
1 2 3 4 5 6 7
mm mm mm mm
Knots, checks and 75 19 19 25 1 in Knots, check
shakes shall be of and shakes
half the size for 100 25 25 35 15 shall be 1.5
grade I the size for
150 38 38 50 for grade
Slope of grain shall
be not more than 200 44 50 65 all Slope of grain
1 in 20. shall be not
250 50 57 81 more than 1
in 12.
300 54 75 100
350 57 81 115
400 63 87 131
450 66 93 150
500 69 100 165
550 72 103 181
600 75 106 200
where, B = Width of face
C = Location and sizes of live knots
⎛1 ⎞
C1 = Narrow face and ⎜ ⎟ of the width of wide face close to top and
⎝4⎠
bottom edges
C2 = Central half of the width of wide face
D = Depth of checks and shakes
θ = Slope of grain.
794 DESIGN OF STEEL STRUCTURES

13.7 SAP WOOD AND HEART WOOD

The size of a tree increases by the growth of new wood on the outer surface
below the bark. A layer of new wood is formed every year on the periphery of
the trees. The layer is called annual ring. The trunk of a young tree consists
entirely of light colour wood. Light colour wood is known as sap wood. The sap
wood carries sap from roots of the tree to leaves. The sap wood contains the
living cells. With the increase of age of tree and addition of new layers every
year, the sap wood gradually changes to heart wood and new sap wood is formed.
The older wood ceases to take part in the growth activity. The cells in older
wood die. The various substances are formed in the older wood. As a result of
these substances, the heart wood has distinct dark colour, and heart wood is
more resistant to decay. The heart wood is more durable than sap wood. The
heart wood is less porous than the sap wood. The sap wood is stronger and
heavier than the heart wood, provided the moisture content is same. The sap
wood can be impregnated with preservatives more readily because of its
absorptiveness. Properly treated with preservative, the sap wood also becomes
more durable than even the most durable wood like teak, sal and deodar, heart
wood.
There are two choices, as regards suitability in respect of durability and
treatability for the structural timber. The timber spices in the first choice shall
be of any one of the following :
(a) Untreated heart wood of high durability.
(b) Treated heart wood of moderate and low durability and class A and
class B treatability.
(c) Heart wood of moderate durability and class C treatability after pressure
impregnation.
(d) Sap wood of all classes of durability after thorough treatment with
preservatives.
The durability and treatability for the various species of timber have been
given in a table in IS : 883 –1994.
The timber species in the second choice shall be of heart wood of moderate
durability and Class D treatability.

13.8 GRAINS OF TIMBER

The grain is a term used to denote the direction of fibre of timber in relation to
the main axis of a tree or member. It also denotes the relative width of annual
rings. When the direction of fibres of a tree is more or less parallel to the axis of
tree (or longitudinal axis), then the grains are called straight grains. The straight
grains are also known as parallel grains. The straight or parallel grains of a
limber member are shown in Fig. 13.7. The strength of timber along the straight
grains is more than other directions. The directions along the grains and across
the grains are also shown in Fig. 13.7. When the direction of fibres is at varying
and irregular inclinations in relation to the main axis, it is known as irregular
 DESIGN OF TIMBER STRUCTURES 795

grain. The irregular grains occur locally. These reduce the strength of timber.
When the fibres take a more or less winding or spiral course, the grains are
called spiral grains. The spiral grains give the whole tree a twisted appearance.
The spiral grains are considered as serious structural defect. The spiral grained
D irectio n acro ss (p erp e nd icular to) the g rains
S traigh t
g rains
D irectio n
p ara llet to
(A lon g the )
g rains

Fig. 13.7

timber is weak. Some trees have spiral grains in one direction for number of
years. The slope of grains reverses in direction around the trees in the succeeding
annual layers of growth. Such fibres are called interlocked grains. The inter-
locked grained timber is not so weak as spiral grained timber.
When the fibres take the form of waves and undulations, then these are called
wavy grains. When the fibre are distorted to that they have a curled direction,
then these are termed as curly grains. When the grains are straight but inclined
to the main axis, these are termed at diagonal grains. In the diagonal grains,
fibres extend at an inclination (i.e., diagonally across a piece). It is a sawing
defect. This defect is produced when a straight grained timber is not cut parallel
to the grains. All the deviations from straight or parallel grains (including spiral
and diagonal grains) are termed as cross grains.
The slope of grains is defined as the ratio of unit deviation to the distance
within which this deviation occurs. In case of spiral and diagonal grains, the
effective slope is found by taking square root of sum of the squares of’two types
of cross-grains.

13.9 STRENGTH OF TIMBER

The strength of timber is influenced by moisture content, density, presence of
defects, type of grains and size. There is free moisture in the green wood within
the cell spaces. There is also absorbed moisture in the cell walls. When green
wood dries, free moisture evaporates. On the evaporation of all the free moisture,
the fibre saturation point is reached, and absorbed moisture from cell wall starts
evaporating. The fibre saturation point differs for different timbers. The fibre
saturation point is the point at which the weight of moisture in the wood is
between 21 percent and 28 percent of oven dry weight of wood. When the moisture
content in the wood is more than this point, the strength of wood is practically
constant. When the moisture content in the wood is less than this point and
decreases, the strength of wood increases and wood starts to shrink. The strength
of wood continues to increase until the wood is bone dry.
The density of wood measures amount of wood substance in a given volume.
The strength of wood depends on the amount of wood substance. The strength
796 DESIGN OF STEEL STRUCTURES

of wood is more for dense wood than for light wood. The strength of timber
reduces due to the presence of defects. The influence of particular type of defect,
and position of defects have been described in Sec. 13.2. The strength of timber
also depends upon type of grain of timber. The strength of timber along the
grains is more than that across the grains. The effect of each type of grain has
been described in Sec. 13.8. The strength of timber also depends upon size of
timber. The defects developed during seasoning in large pieces are more than in
smaller pieces.
Table 13.2 Factors of safety to be applied to basic stress to obtain safe working stresses

Standard Grade

S.No. Stress N/mm2 Inside Outside Wet

location location location

1. Fibre stress in beams for broad 5 6 7.5

leaved species, max.
2. Fibre stresses in beams in conifers, max. 6 7 8.5
3. Shear along grain 7 7 7
4. Horizontal shear in beams 10 10 10
5. Compressive stress parallel to grain 4 4.5 5.5
6. Compressive stress perpendicular 1.75 2.25 2.75
to grain, max.
7. Modulus of elasticity for calculation
of allowable deflection in beams 1.25 1.75 2

13.10 FACTOR OF SAFETY IN TIMBER

The factor of safety depends upon strength of timber, effect of moisture, the
presence of defects and location of structure. The timber structure may be
situated in inside location, or in outside location or in wet location. When the
timber structures are situated in inside location they remain continuously dry
and protected from moisture. When the timber structures are situated in outside
location, they remain occasionally wet but dry quickly, for example, timber
bridges, trestles, framework of open sheds etc. Sometimes the timber structures
are situated in such a location, where they remain usually wet, for example,
timber structures exposed to tide water or waves, or in contact with water of
moist ground. The factor of safety for different locations to be applied to basic
structures to obtain safe working stresses for standard grade of timbers is given
in Table 13.2 as per IS : 883–1961.

13.11 WORKING STRESSES IN TIMBER

The working stresses for the timber depend on many factors. The working
stresses are based upon the basic stresses and appropriate factor of safety. The
basic stresses are obtained by tests on small clear specimen of standard grades
Table 13.3 Safe working stresses for indian timbers

Working stress in N/mm2 for Standard Grade (Grade I)

S.No. Trade Name Bending and Tension Shear Compression
of Timber along grade
(Extreme fibre stress)
Hori- Along Parallel Perpendicular
zontal1 grain2 to grain to grain
Inside Outside Wet All All Inside Out Wet Inside Out Wet

Ave. unit wt at
12% moisture
loca- loca- loca- loca- loca- loca- side loca- loca- side loca-

All grades, all location

Modulus of Elas (kN/mm2)
tion tion tion tion tion tion loca- tion tion loca tion
tion tion

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Group A
1. Bullet wood 11.05 17.40 22.8 19.0 15.2 1.5 2.1 14.4 12.6 10.2 11.2 8.8 7.1
DESIGN OF TIMBER STRUCTURES

2. Ballagi 11.35 16.30 22.6 18.6 15.2 1.5 2.2 14.8 13.0 10.6 8.7 6.7 5.5
3. Dhaman 7.85 14.80 18.2 15.2 12.4 1.3 1.9 12.0 10.6 8.8 6.0 4.6 3.8
4. Kala Siris 7.35 13.60 18.6 15.4 12.6 1.5 2.7 13.4 12.0 9.4 7.2 5.6 4.6
5. Sal 8.65 12.70 16.8 14.0 11.2 0.9 1.3 10.6 9.4 7.8 4.5 3.5 2.9

Notes.1. The timber of group A (viz. super group) have E greater than or equal to 12600 N/mm2 and those of group B (standard
group) have E greater than 9800 N/mm2. Whereas the timbers of group C (viz. ordinary group) have E greater than
5600 N/mm2.
2. Above values of the working stresses have been converted in S.I. units.
797
Table 13.3 Contd.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 798
Group B
1. Indian Oak 8.65 12.5 14.8 12.4 9.8 1.2 1.7 9.2 8.0 6.6 4.5 3.5 2.9
2. Bullet wood 8.80 12.4 17.2 14.4 11.6 1.3 1.8 10.8 9.8 8.8 5.5 4.3 3.5
3. Dhaman 7.70 12.0 15.5 13.0 10.2 1.4 2.0 9.1 8.1 6.7 4.1 3.2 2.6
(U.P. M.P.)
4. Eucalyptus 8.50 11.5 16.6 13.8 10.8 1.2 1.7 11.2 10.2 8.0 7.6 5.9 4.8
5. Jaman 8.50 11.2 15.1 12.6 10.2 1.2 1.7 9.1 8.4 6.7 5.8 4.5 3.6
6. Babul 7.85 10.8 18.2 15.4 12.4 1.5 2.2 11.2 10.2 8.0 6.5 5.0 4.1
7. Teak 6.40 9.6 14.0 11.6 9.4 1.0 1.4 8.8 7.8 6.4 4.0 3.1 2.5
Group C
1. Indian chestnut 6.40 9.8 10.6 8.8 7.0 0.8 1.2 6.4 5.6 4.6 2.6 2.0 1.7
DESIGN OF STEEL STRUCTURES

2. Deodar 5.45 9.5 10.7 8.8 7.0 0.7 1.0 7.8 7.0 5.6 2.6 2.1 1.7
3. Rose wood 7.55 9.3 16.8 14.0 11.2 1.2 1.7 10.6 9.2 7.8 6.4 5.0 4.1
4. Sissoo 7.85 8.9 15.2 12.6 10.7 1.7 1.7 9.4 8.4 6.6 4 .6 3.6 2.9
5. Safed Siris 6.40 9.0 13.4 11.2 8.8 1.0 1.4 8.4 7.8 6.4 4 .3 3.3 2.7
6. Chir 5.75 9.8 8.4 7.0 6.0 0.6 0.9 6.4 5.6 4.6 2.7 1.7 1.4
Notes. 1. The values of horizontal shears are to be used for beams.
2. In all other cases, the shear along grains are to be used.
3. Above values of the working stresses have been converted in S.I. units.
 DESIGN OF TIMBER STRUCTURES 799

in the laboratories for standard conditions of service. The safe working stresses
for 72 species of timbers of standard grade are given in IS : 883 –1994. The
actual values of safe working stresses for the few species of timber are given
(for standard or grade I) as per IS : 883 – 1994 in Table 13.3 for reference purpose,
for inside, outside and wet locations.
The permissible stresses for groups A, B and C for different locations applicable
to grade I structural timber are given in Table 13.4 as per IS : 883 –1994 provided
that the following conditions are fulfilled :
(a) The timber should be of high or moderate durability and be given the
suitable treatment where necessary. They may be used in any location. If the
location is inside and not in contact with the ground, low durability timber may
be used after proper seasoning and preservative treatment are given, and
(b) The loads should be continuous and permanent.
The permissible stresses for timbers of selected grade, are 1.16 times the
stresses for grade I timbers given in Table 13.4. The permissible stresses for
timber of grade II are 0.84 times the stresses for grade I timbers given in Table
13.4. When the low durability timbers are used on outside locations, the
permissible stresses for all grades of timber found are multiplied by 0.80.
In inside location, the timbers inside the buildings remain continuously dry
or protected from the weather. In outside locations, the timbers either in open
sheds or in outside exposed structure are occasionally subjected to wetting and
drying. Whereas in wet locations, the timbers are almost continuously damp or
wet or in contact with earth or water such as piles and timber foundation.

Table 13.4 Permissible stresses for grade I timber (As per IS : 883–1994)

S. No. Types of stresses Permissible stress in N/mm2¶

Group A Group B Group C
1 2 3 4 5
1. Bending and tension along grain
(i) Inside location 18.2 12.3 8.4
(ii) Outside location 15.2 10.2 7.0
(iii) Wet location 12.0 8.1 6.0
2. Shear
(i) Horizontal* all locations 1.2 0.9 0.6
(ii) Along grain** all locations 1.7 1.3 0.9
3. Compression parallel to grain
(i) Inside location 12.0 7.0 6.4
(ii) Outside location 10.6 6.3 4.6
(iii) Wet location 8.8 5.8 4.6
* The values of horizontal shears are to be used for beam. Contd.
** In all other cases, the values of shear along the grains are to be used.
¶ Above values have been converted in S.I. units.
800 DESIGN OF STEEL STRUCTURES

Contd. Table 13.4

Group A Group B Group C
1 2 3 4 5
4. Compression perpendicular to grain
(i) Inside location 6.0 2.2 2.2
(ii) Outside location 4.6 1.8 1.7
(iii) Wet location 3.8 1.5 1.4
Modifications factors for permissible stresses
The permissible stresses given in Table 13.4 are modified by multiplying
with the modification factors for the following:
(i) For change in slope of grain. When the timber has major defects
like slope of grains, knots, and checks or shakes (but not beyond permissible
value), the permissible stresses given in Table 13.4 are multiplied by the
modification factors K1 for different slopes of grain as given in Table 13.5 as per
IS : 883 –1994.
Table 13.5 Modification factor K1 to allow for change in slope of grain IS : (883–1994)

K1
Slope Strength of beams ; joints and Ties Strength of Posts or Column
1 2 3
1 in 10 0.80 0.74
1 in 12 0.90 0.82
1 in 14 0.98 0.87
1 in 15 1.00 1.00

Table 13.6 Modification factor K2 to allow for change in slope of grain (883–1994)

S.No. During of loading K2

1 2 3
1 Continuous 1.0
2 Two months 1.15
3 Seven days 1.25
4 Wind and earthquake 1.33
5 Instantaneous or impact 2.00
(i) For change in the duration of load. For the duration of design load
other than continuous, the permissible stresses given of Table 13.4 are multiplied
by the modification factors K2 given in Table 13.6 as per IS : 883 –1994.

13.12 BEARING STRESS IN TIMBER

The bearing stress or compressive stress on a surface depends upon the
inclination of surface with the direction of grain, bearing length, and distance
 DESIGN OF TIMBER STRUCTURES 801

from the end of a structural member. At any bearing of the side grain of timber,
the permissible stress in compression perpendicular to the grain, fcn is dependent
on the length and position of bearing. The permissible stresses given in Table
13.4 for compression perpendicular to the grain are also the permissible stresses
for bearings of any length at the ends of members and for the bearings 150 mm
or more in length at any other position. For the bearings less than 150 mm in
length and located 75 mm or more from the end of a member, the permissible
compressive stress perpendicular to the grain may be multiplied by the
modification factor K7 given in Table 13.7 as per IS : 883 –1994. The allowance
is not made for difference in intensity of the bearing stress due to bending of a
beam. For the bearing stress under washer or a small plate, the same co-efficient
recommended in Table 13.7 may be taken for bearing with a length equal to the
diameter of the washer or the width of the small plate.
Table 13.7 Modification factor K7,for bearing stresses

Length of bearing in mm Co-efficient

15 1.67
25 1.40
40 1.25
50 1.20
75 1.13
100 1.10
150 1.00

When the direction of stress is at an angle to the direction of grain in any

structural member as in roof trusses, as shown in Fig. 13.8, the permissible

9 0°

Fig. 13.8

bearing stress in that member is calculated from Hankinson’s formula given

below
⎛ fcp ⋅ fcn ⎞
fce = ⎜ ...(13.1)
⎜ fcp sin θ + fcp cos θ ⎟⎟
2 2
⎝ ⎠
where, fce = permissible bearing stress in the direction of the line of action
of load
802 DESIGN OF STEEL STRUCTURES

fcp = permissible compressive stress parallel to the grain

fcn = permissible compressive stress perpendicular to the grain
θ = Angle of load with the direction of grain
Example 13.1 The safe working stress in compression of a standard timber
(dhaman) for inside location, parallel to the grain is 12 N/mm2 and that
perpendicular to the grain is 6 N/mm2. Determine the safe working stress for
this timber, if the timber is of selected grade and if the timber is of common
grade (i.e., grade II).
Solution
For standard grade. The safe working stress in compression parallel to the
grain is 12 N/mm2. The safe working stress in compression perpendicular to the
grain is 6 N/mm2.
For selected grade. The safe working stress in compression parallel to the
grain
(1.16 × 12) = 13.92 N/mm2
The safe working stress in compression perpendicular to the grain
(1.16 × 12) = 6.96 N/mm2
For common grade. The safe working stress in compression parallel to the
grain
(0.84 × 12) = 10.08 N/mm2
The safe working stress in compression perpendicular to the grain
(0.84 × 6) = 5.04 N/mm2.
Example 13.2 The safe working stress in compression of a timber (dhaman)
for inside location parallel to the grain is 12 N/mm2 and that perpendicular to
the grain is 6 N/mm2. Determine the permissible bearing stress in the direction
of line of action of the load acting at 30° to the grain direction.
Solution The safe working stress in compression parallel to the grain
fcp = 12 N/mm2
The safe working stress in compression perpendicular to the grain
fcn = 6 N/mm2,
θ = 30°.
The permissible bearing stress in the direction of line of action of the load as
per Eq. 13.1

⎛ fcp ⋅ fcn ⎞
fce = ⎜ 2 2 ⎟
⎝ fcp ⋅ sin θ + fcp ⋅ cos θ ⎠

⎛ 12 × 6 ⎞
= ⎜ = 9.6 N/mm2
1 3⎟
⎜ 12 × + 6 × ⎟
⎝ 4 4⎠
 DESIGN OF TIMBER STRUCTURES 803

13.13 SOLID WOOD COLUMNS

The columns are defined as the structural members which support load primarily
by inducing compressive stress along the grain.
The solid wood columns consist of a single piece of wood. Generally, these are
of rectangular cross-section. The columns of round or circular cross-section are
also known as solid wood columns. The round wood columns are not commonly
used. The load carrying capacities of columns of round and square wood columns
s
of equal cross-sectional area are same. The slenderness ratio in case of solid
d
wood columns is defined as the ratio of unsupported length (effective length), s,
of the column to the dimension of least side, d. For a rectangular section, the
dimension of least side, d is the dimension of narrowest of two faces. For a
circular section, the dimension of least side, d is the dimension of side of a
square column of an equivalent cross-sectional area. For tapered column, the
least dimension, d is taken as sum of the corresponding least dimensions at the
small end of column and one-third of the difference between the least dimensions
at the large end and small end. If D is the least dimension at the large end and
D2 is the least dimension at the small end, then the least dimension of tapered
1
column, d is [D2 + (D – D2)]. The least dimension of tapered column should
3 1
not be greater than one and a half times the least dimension at the small end
(i.e. d |> 1.5 D2). The induced stress at the small end tapered column should not
exceed the safe compressive stress in the direction parallel to the grain. The
solid wood columns are classified into the following three categories:
1. Short columns 2. Intermediate columns 3. Long columns.
The solid columns are formed of any section having a solid core throughout.
⎛s⎞
1. Short columns. When the slenderness ratio, ⎜ ⎟ is not greater than
⎝d⎠
11, then the columns are known as short columns. For short columns, the
permissible stress is calculated as follows :
⎛P⎞
fc = ⎜ ⎟ = fc.p ...(13.2)
⎝A⎠
where, P = Load of the column in N
A = Area of cross-section of column in mm2
fc .p = Stress in compression parallel to the grains.
⎛s⎞
2. Intermediate columns. When the slenderness ratio, ⎜ ⎟ is greater
⎝d⎠
than 11, and less than or equal to K8 then the columns are known as intermediate
columns. For intermediate columns, the permissible stress is found as follows.
4⎤
⎛P⎞ ⎡
fc = ⎜ ⎟ = fc.p ⎢1 – 1 ⎛ s ⎞ ⎥ ...(13.3)
⎝A⎠ ⎜ ⎟
⎣⎢ 3 ⎝ k8 d ⎠ ⎦⎥
804 DESIGN OF STEEL STRUCTURES

where, K8 = Constant. It depends upon E and fcp

1
⎡ E ⎤2
K8 = ⎢ ⎥
⎣ fc ⋅ p ⎦
E = Modulus of elasticity in N/mm2.
⎛s⎞
3. Long columns. When the slenderness ratio, ⎜ ⎟ is greater than K8,
⎝d⎠
then the columns are called long columns. For long columns, the permissible
stress is determined as follows :

⎛ P ⎞ ⎛ 0.329E ⎞
fc = ⎜ ⎟ = ⎜ ...(13.6)
⎝ A ⎠ ⎜ ⎛ s ⎞2 ⎟⎟
⎜ ⎜ ⎟ ⎟
⎝ ⎝d⎠ ⎠

s
In solid wood columns, the slenderness ratio should not exceed 50.
d

13.14 BUILT-UP AND BOX COLUMNS

Built up wooden columns consist of wooden pieces joined together with spikes,
bolts, nails, screws or glue or with other
mechanical fasteners. Built-up wooden
columns are shown in Fig. 13.9. The strength
of a built-up column isnot equal to that of a
solid wood column. The built-up column
should be fastened together at intervals not
exceeding six times the thickness of each
individual compression member by bolts and
a limber connector or which are braced in
direction parallel to the least dimension of Fig.13.9 Built-up and box columns
the individual member.
A rectangular box column is formed of members and a hollow core. The
members arc joined with one another forming box provided with solid blocks at
ends and at intermediate points.
The box columns are also classified into short, intermediate and long columns
as follows :
(a) The built-up columns are called as the short columns when

⎡ s ⎤
⎢ 2 ⎥ is less than 8.
( )
1/2
2
⎢⎣ d1 + d2 ⎥⎦
where, d1 = Least overall width of box column in mm
d2 = Least overall dimension of core in box column in mm
(b) The built-up columns are called as the intermediate columns when
 DESIGN OF TIMBER STRUCTURES 805

⎡ s ⎤
⎢ 2 ⎥ is between 8 and K9 and
(
⎣⎢ d1 + d2
2
)
1/2
⎦⎥
(c) The built-up columns are called as the long columns when

⎡ s ⎤
⎢ 2 ⎥ is greater than K9
( )
1/2
2
⎢⎣ d1 + d2 ⎥⎦
where, K9 is a constant
1/ 2
π ⎡ U ⋅E ⎤
or K9 = ...(13.6)
2 ⎢⎣ 5 ⋅ q ⋅ fcp ⎥⎦

For the short columns, the permissible compressive stress is calculated as

follows :
fs = q . fc.p ...(13.7)
For the intermediate columns the permissible compressive stress is calculated
as follows :

⎡ 4
fs = q . fcp ⎢1 – 1 ⎛ s ⎞ ⎤ ...(13.8)
⎥
⎢ 3⎜ 1⎟
⎜⎝ K ⋅ ⎡d2 + d2 ⎤ 2 ⎟⎠ ⎥
⎣⎢ 9 ⎣ 1 2⎦ ⎦⎥
For the long columns, the permissible compressive stress is calculated by
using the following formula :

0 ⋅ 329 U ⋅ E
fc = 2
...(13.9)
⎡ s ⎤
⎢ 1 ⎥
⎢ ⎡d 2 + d 2 ⎤ 2 ⎥
⎣⎣ 1 2⎦ ⎦

where q and U are constants for the particular thickness of the plank. The
following values of constant q and U are used depending upon plank thickness, t
t U q
25 mm 0.80 1.00
50 mm 0.60 1.00

13.15 SPACED COLUMNS

The spaced columns consist of two or more wooden members with their
longitudinal axes parallel joined at their ends and at intermediate points by
block pieces. The members are separated at ends by end blocks and at middle
by spacer block. These are connected at the ends by timber connectors and bolts.
Spaced columns are used as compression members in trusses in which timber
connectors are used. Spaced column is shown in Fig. 13.10.
806 DESIGN OF STEEL STRUCTURES

End
b lock

L ea st
d im en sion W idth

S p acer
b lock l1 l

End
b lock

Fig. 13.10 Spaced column

The spacer block used for separation of members is placed at middle 10 per
cent of length. The timber connectors are not required for a single spacer block.
In case more than one block is used, the distance between two adjacent spacer
blocks should not be greater than half the length between the end connectors.
The timber connectors are necessary in case more than one spacer block are
used. The end connectors may be provided either within 5 per cent of actual
length. The dimensions of end blocks arc kept to accommodate the timber
connectors.
In case the end blocks are provided within 5 per cent of actual length of
l1
individual members, then the effective length of columns is . In case the
2⋅5
end blocks are provided at a distance 5 to 10 per cent of actual length of individual
l1
members, then the effective length of column is . The slenderness ratio of a
3
spaced column is the ratio of its effective length to the least dimension of its
individual members. The safe load carrying capacity of spaced column is the
sum of safe load carrying capacity of individual members.
The formulae for solid column specified in Sec. 13.13 are applicable to spaced
column with a restraint factor of 2.5 or 3, depending upon the distance of end
 DESIGN OF TIMBER STRUCTURES 807

connectors in the column. For intermediate spaced column, the permissible

compressive stress is given by

⎡ 1 ⎛ s ⎞4 ⎤
fc = fc.p ⎢1 – ⎜ ⎥ ...(13.10)
⎣⎢ 3 ⎝ k10 d ⎟⎠ ⎦⎥
where, K10 is a constant
1/ 2
⎡ 2 ⋅ 5E ⎤
or K10 = 0.702 ⎢ ⎥ ...(13.11)
⎣ fc⋅ p ⎦
For the long spaced columns the permissible compressive stress shall be

⎛ 0 ⋅ 329 E × 2 ⋅ 5 ⎞
fc = ⎜ 2 ⎟ ...(13.12)
⎜ ⎛s⎞ ⎟
⎜ ⎜ ⎟ ⎟
⎝ ⎝d⎠ ⎠
The compression members shall be notched, where it is necessary to pass
service line through such a member. This shall be affected by means of a bored
hole not larger in diameter than one-quarter the width of the face through which
the hole is bored provided that the local stress is calculated and found to be
within the permissible stress specified. The distance from the edge of the hole
to the edge of member shall not be less than one-quarter of the hole of the face.
The above recommendations and formulae are as per IS : 883 – 1970.

13.16 COLUMNS SUBJECTED TO COMBINED STRESS

Generally, the purlins are placed at panel points of the roof trusses. Sometimes,
the purlins are placed at the points other than panel points. In such cases,
principal rafters of the roof truss are subjected both to axial compression and
bending. When a structural member is subjected both to axial compression and
bending, then the compression,
⎡ ⎧ fac fab ⎫ ⎤
⎢ ⎨ f + f ⎬ ⎥ should not be greater than one.
⎢⎣ ⎩ c⋅ p b ⎭⎥⎦

⎡ ⎧ fac fab ⎫ ⎤
i.e., ⎢⎨ f + f ⎬⎥ | > 1.00 ...(13.13)
⎣⎢ ⎩ c⋅ p b ⎭⎦⎥
where, P = Total axial load in N
A = Area of cross-section in mm2
⎛P⎞
fac = ⎜ ⎟ = Actual average stress in compression in the member
⎝A⎠
fcp = Permissible stress in compression parallel to the grain
M = Bending moment
808 DESIGN OF STEEL STRUCTURES

Z = Section modulus of the member in mm2

M
fab = = Actual maximum bending stress in the member
Z
fb = Maximum permissible stress in member.
Equation 13.13 is applicable to spaced column only when the bending moment
is acting in the direction parallel to the mutually contacting surfaces of the
blocks and individual members.
Example 13.3 A column 150 mm × 150 mm is made of babul wood. The
unsupported length of column is 1.50 m. Determine safe axial load on the column.
Solution
Step 1: Slenderness ratio
Unsupported length of column, s = 1.50 m
Least dimension of column, d = 150 mm
Maximum slenderness ratio
s ⎛ 10 × 1000 ⎞
= ⎜ ⎟ = 10 < 11
d ⎝ 150 ⎠
Step 2: Safe working stress
The column is a short column. Assuming that the column is used for inside
location and wood is of standard grade. Safe working stress in compression
parallel to the grain for babul wood
fcp = 11.2 N/mm2
Step 3: Safe axial load in column

⎛ 11 ⋅ 2 × 150 × 150 ⎞
P = ⎜ ⎟ = 252.5 kN
⎝ 1000 ⎠
Example 13.4 In Example 13.3. If the unsupported length of column is 3 m,
then, determine safe axial load on the column.
Solution
Step 1: Slenderness ratio
Unsupported length of column, s = 3 m
Least dimension of column, d = 150 mm
Maximum slenderness ratio
s ⎛ 3 × 1000 ⎞
= ⎜ ⎟ = 20
d ⎝ 150 ⎠
Step 2: Safe working stress
For babul wood, safe working stress in axial compression parallel to the
grain,
fcp = 11.2 N/mm2
and E = 10800 N/mm2
 DESIGN OF TIMBER STRUCTURES 809

1
⎡ E ⎤2
K8 = 0.702 ⎢ ⎥
⎣ fcp ⎦

1
= 0.702 ⎡10800 ⎤ 2 = 21.8
⎢⎣ 11.2 ⎥⎦

The slenderness ratio of the column is greater than 11 and it is less than K8.
The column is considered as an intermediate column,
⎡ 1 ⎛ 3000 ⎞ ⎤
fc = 11 ⋅ 2 ⎢1 – ⎜ ⎟ = 8.55 N/mm2
⎣ 3 ⎝ 21 ⋅ 8 × 150 ⎠ ⎥⎦
Step 3: Safe axial load on the column
⎛ 8 ⋅ 55 × 150 × 150 ⎞
P = ⎜ ⎟ = 192.5 kN
⎝ 1000 ⎠
Example 13.5 In Example 13.4. If the effective length column is 3.60 m, then
determine the safe axial load on the column.
Solution
Step 1: Slenderness ratio
Effective length of column, s = 3.6 m
Least dimension of column, d = 150 mm
Maximum slenderness ratio,
s ⎛ 3 ⋅ 60 × 1000 ⎞
= ⎜ ⎟ = 24
d ⎝ 150 ⎠
For Example 13.4 Ks = 21.8
The slenderness ratio of the column is greater than K8. The column is treated
as a long column.
Step 2: Permissible stress on the column (from Eq. 13.5)

0 ⋅ 329E ⎛ 0 ⋅ 329 × 10800 ⎞

fc = 2
= ⎜ ⎟
⎛s⎞ ⎝ (24 )2 ⎠
⎜ ⎟
⎝d⎠
= 6.18 N/mm2
Step 3: Safe axial load on the column
6 ⋅ 18 × 150 × 150 ⎞
P = ⎛⎜ ⎟ = 139 kN
⎝ 1000 ⎠
Example 13.6 A column 150 mm in diameter is made of deodar wood. The
effective length of column is 1.20 mm. Determine the safe axial load of the round
column. The column in situated in outside location and subjected to alternate
wetting and drying.
810 DESIGN OF STEEL STRUCTURES

Solution
Step 1: Slenderness ratio
Sectional area of round column
π
× (150)2 = 16900 mm2
4
Let d be the dimension of a square column of equivalent cross-sectional area
to that of round column
Sectional are of square column = d2 mm2
d2 = 16900 ∴ d = 130 mm
The effective length of column, s is 1.20 mm
The least dimension of the column, d is 130 mm
Maximum slenderness ratio
s ⎛ 120 × 1000 ⎞
= ⎜ ⎟ = 9.24 < 11
d ⎝ 130 ⎠
Step 2: Safe working stress
The column is a short column. For deodar wood, safe working stress in axial
compression parallel to the grains for outside location
fcp = 7 N/mm2
Step 3: Safe axial load for the round column
⎛ 7 × 130 × 130 ⎞
P = ⎜ ⎟ = 118.3 kN
⎝ 1000 ⎠
Example 13.7 A column carries an axial load of 500 kN inclusive of self
weight. The effective length of column is 3.50 m. Design the wood column.
Solution
Design :
The effective length of column, s is 3.50 m
⎛s⎞
Let d be the least dimension of the column. For a short column, ⎜ ⎟ < 11
⎝d⎠

⎛ 3 ⋅ 50 × 1000 ⎞
d > ⎜ ⎟ > 318.1 mm
⎝ 11 ⎠
Adopt the least dimension of the column as 320 mm
Let fir wood be used for the solid column
Step 1: Safe working stress in compression parallel to the grain for
firewood
fcp = 5.2 N/mm2
Step 2: Sectional area required for column
⎛ 500 × 1000 ⎞
⎜ ⎟ = 96153.846 mm2
⎝ 5⋅2 ⎠
 DESIGN OF TIMBER STRUCTURES 811

The least dimension of the column, d is 320 mm

Other dimension of the square column
⎛ 96153 ⋅ 846 ⎞
⎜ ⎟ = 300.48 mm
⎝ 320 ⎠
Adopt 320 mm × 320 mm fir wood column
Step 3: The load carrying capacity of the column
⎛ 52 × 320 × 320 ⎞
P = ⎜ ⎟ = 533 kN > 500 kN
⎝ 1000 ⎠
Hence, safe.
Example 13.8 A built-up sal wood column consists of a solid core 200 mm ×
200 mm and four planks 50 mm × 50 mm. All the pieces are spiked together. The
effective length of column is 3 m. Determine the safe axial load on the column.
Solution :
Step 1: Slenderness ratio
The built-up column is shown in Fig. 13.11. All the elements of built-up column
are joined together. The least overall width of box column, d1 is 300 mm and the
least overall dimension of core, d2 in the box column is 200 mm. The slenderness
ratio of the column
s ⎛ 300 ⎞
1
= ⎜ 1 ⎟ = 8.32
⎜( ⎟
( )
d12 + d22 2 2
⎝ 300 + 200
2 )2
⎠
1
π ⎡U ⋅ E ⎤ 2
K9 =
2 ⎢⎣ 5qfcp ⎥⎦

1/2
π ⎡ 0 ⋅ 60 × 127 × 1000 ⎤
=
2 ⎢⎣ 5 × 1 × 10 ⋅ 6 ⎥⎦
= 18.825 (8 > 8.32 < K9)
50 50
mm 2 00 m m mm

50 m m
2 00 m m

50 m m

Fig.13.11 Built-up column

812 DESIGN OF STEEL STRUCTURES

Step 2: Safe working stress

Hence, the built-up column is an intermediate column. The permissible
compressive stress for the built-up intermediate column

⎡ 1⎛ 4⎤
s ⎞
fc = q ⋅ fcp ⎢1 – ⎜ 1 ⎟ ⎥
⎢ 3⎜ ⎥
⎢⎣ k (
⎝ 9 1 d 2
+ d2 )
2 2 ⎟
⎠ ⎥⎦

⎡ 1 ⎛ 8 ⋅ 32 ⎞4 ⎤
fc = 1 ⋅ 00 × 10 ⋅ 6 ⎢1 – ⎜ ⎟ ⎥
⎣ 3 ⎝ 18 ⋅ 825 ⎠ ⎦
= 10.465 N/mm2
Step 3: Safe axial load, which may be carried by the built-up column
⎛ 10 ⋅ 465 × 300 × 300 ⎞
P = ⎜ ⎟ = 941 kN
⎝ 1000 ⎠
Example 13.9 A column carries an axial load of 1000 kN inclusive of self
weight. The effective length of column is 3.60 m. Design a built-up sal wood
column.
Solution
Design:
Step 1: Slenderness ratio
The effective length of column, s is 3.60 m. Let the least overall width of box
column be d1 and the least overall dimension of core d2 be 250 mm in the box
column. In case, the column is designed as a short column, then
⎛ s ⎞
⎜ 1 ⎟ < 8
⎜ (d 2 + d 2 )2 ⎟
⎝ 1 2 ⎠

⎛ 3600 ⎞
⎜ 1 ⎟ < 8
⎜ (d2 + 2502 )2 ⎟
⎝ 1 ⎠
∴ d1 = 374 mm
Let d1 be 350 mm. The slenderness ratio for the column
The slenderness ratio for the column

⎛ s ⎞ ⎛ 300 ⎞
⎜ 1 ⎟ = ⎜ 1 ⎟ = 8.37
⎜ (d 2 + d 2 )2 ⎟ ⎜( 2 2 )2 ⎟
⎝ 1 2 ⎠ ⎝ 350 + 250 ⎠

1
π ⎡ U ⋅ E ⎤2
K9 =
2 ⎢⎣ 5q ⋅ fcp ⎥⎦
 DESIGN OF TIMBER STRUCTURES 813

Let the sal wood be used for the column,

1
π ⎡ 0 ⋅ 60 × 127 × 1000 ⎤ 2
K9 = ⎥⎦ = 18.825
2 ⎢⎣ 5 × 1 × 10 ⋅ 6
(8 > 8.37 > 18.825 )
Step 2: Safe working stress
Hence, the built-up column shall be intermediate column. The permissible
compressive stress for the built-up intermediate column

⎡ 1⎛ 4⎤
s ⎞
fc = q ⋅ fcp ⎢1 – ⎜ 1 ⎟ ⎥
⎢ 3⎜ ⎥
⎢⎣ k (
⎝ 9 1 d 2
+ d2 )
2 2 ⎟
⎠ ⎥⎦

⎡ 1 ⎛ 8 ⋅ 37 ⎞ 4 ⎤
fc = 1 × 10 ⋅ 6 ⎢1 – ⎜ ⎟ ⎥
⎣ 3 ⎝ 18 ⋅ 825 ⎠ ⎦
= 10.4619 N/mm2
Step 3: Area required for the column
⎛ 1000 × 1000 ⎞
A = ⎜ ⎟ = 95584.8 mm2
⎝ 10 ⋅ 4619 ⎠
Provide 2 planks 350 mm × 50 mm and 7 planks 250 mm × 50 mm as shown
in Fig. 13.12. The area provided is 350 mm × 350 mm (i.e., 122500 mm2)

3 50
mm

Fig. 13.12 Built-up column

Step 4: Safe axial load

⎛ 122500 × 10 ⋅ 4619 ⎞
P = ⎜ ⎟
⎝ 1000 ⎠
= 1281.59 kN > 1000 kN
814 DESIGN OF STEEL STRUCTURES

Example 13.10 A spaced column is 3.6 m long. It carries an axial compression

of 45 kN. Design the column. Use deodar wood for the spaced column.
Solution
Design
Step 1: Slenderness ratio
Actual length of spaced column,
l = 3.60 m
Provide end connectors at 5 per cent distance
Distance of end connectors
⎛ 5 ⎞
⎜ × 3 ⋅ 60 × 1000 ⎟ = 180 mm
⎝ 100 ⎠
Distance between centre to centre of connectors
l1 = (3600 – 2 × 180) = 3240 mm
Effective length of column
⎛ l1 ⎞ ⎛ 3240 ⎞
s = ⎜ ⎟=⎜ ⎟ = 2050 mm
⎝ 2⋅5 ⎠ ⎝ 2⋅5 ⎠
Let the least dimension of the spaced column be 60 mm.
Slenderness ratio of column
⎛ 2050 ⎞
⎜ ⎟ = 34.16 >11
⎝ 60 ⎠
Step 2: Safe working stress
From code, for deodar wood, safe working stress in axial compression parallel
to the grain
fcp = 7.8 N/mm2, and E = 9.5 kN/mm2
1
⎡2 ⋅ 5× E ⎤2
K10 = 0 ⋅ 702 ⎢ ⎥
⎣ fcp ⎦
1
⎡ 2 ⋅ 5 × 9 ⋅ 5 × 103 ⎤ 2
= 0 ⋅ 702 ⎢ ⎥ = 38.74
⎣ 7⋅8 ⎦

⎛s⎞
⎜ ⎟ < K10. The column from Eq. 13.10 for intermediate spaced column
⎝d⎠

⎡ 1 ⎛ 34 ⋅ 16 ⎞4 ⎤
fc = ⎢1 – ⎜ ⎟ ⎥ = 6.228 N/mm 2
⎣ 3 ⎝ 38 ⋅ 76 ⎠ ⎦
Step 3: Area required for column
⎛ 45 × 1000 ⎞
⎜ ⎟ = 7225.24 mm2
⎝ 6 ⋅ 228 ⎠
 DESIGN OF TIMBER STRUCTURES 815

Use 2 planks. Area required for one plank

⎛ 7225 ⋅ 24 ⎞
⎜ ⎟ = 3612.62 mm2
⎝ 2 ⎠
Least dimension assumed = 60 mm
Step 4: Width of plank required
⎛ 3612 ⋅ 62 ⎞
⎜ ⎟ = 60.21 mm
⎝ 60 ⎠
Provide 2 planks 80 mm × 60 mm as shown in Fig. 13.13. The spacer block is
placed at the centre. The spacer block does not require connectors. The spacer

End
b lock
0 .18
m

S p acer 3 .24 m
3.6 0 m

b lock

2 × 140 mm
× 60 m m

0 .18
m
End
b lock

Fig. 13.13 Spaced column

block is kept in position by a bolt. The end blocks are provided at the ends with
suitable connectors.

13.17 TENSION MEMBERS

Ordinarily, the wood is not used for tension members because steel tie rods are
more economical. With the use of modern timber connectors it is possible to
have wooden tension members. The safe working stresses in axial tension along
the grain for different timbers are given in IS : 883–1970.
When a structural member is subjected both to axial tension and bending,
then, the expression
⎡ fat fab ⎤
⎢ f + f ⎥ is not greater than one,
⎣ t b ⎦
816 DESIGN OF STEEL STRUCTURES

where, fat is actual average stress in tension in the member and ft is permissible
stress in axial tension
⎡ fat fab ⎤
⎢ f + f ⎥ |> 1.0 ...(13.14)
⎣ t b ⎦

The other notations are same as mentioned in Sec. 13.6.

The net section is used for the tension member. The net section is obtained
by deducting from the gross-section the projected area of all material removed
by boring, grooving or other means. The net area used in calculating load carrying
capacity of a member shall be the least net section determined as above passing
a plane for a section or a series of connected planes transversely through the
members.
The notches shall in no case remove more than one quarter of the section.
Example 13.11 A tension member is made of 150 mm × 50 mm deodar wood.
Determine the safe axial pull in the member. Nails are used for connections.
Solution
From IS : 883 –1970, for deodar wood safe working stress in tension along the
grain for inside location
= 10.2 N/mm2
Cross-sectional area of tension member
= (l50 × 50) = 7500 mm2
Safe axial pull in the member
⎛ 10 ⋅ 2 × 7500 ⎞
P = ⎜ ⎟ = 76.5 kN
⎝ 1000 ⎠

13.18 WOODEN BEAMS

The beams are defined as the structural members which support the load
primarily by its internal resistance to bending.
Wooden beam usually consists of a single piece of rectangular section as shown
in Fig. 13.14.

b = W idth

d = D e pth

Fig. 13.14 Wooden beam

 DESIGN OF TIMBER STRUCTURES 817

The effective span of beams and other flexural members shall be taken as
the distance from face of the supports plus one half of the required length of
bearing at each end except that for continuous beams and joists, the span may
be measured from centre of bearing at those
supports over which the beam is continuous.
When the span of beam and load combination
is such that the beams of large dimensions are
required. It is not possible to have wooden beams
of large dimensions. The built-up wooden beams
are formed by connecting a number of smaller
beams together by bolts, screws or spikes as shown
in Fig. 13.15. Bolts are staggered and spaced
longitudinally at a distance less than four times
the depth of beams. The large dimension of
individual beam is kept vertical.
The following form factors are applied to the
Fig. 13.15 Built-up beam
bending stress for the following cross-sections of
the beams as per IS : 883 –1970.
(a) Rectangular section. For rectangular sections for different depth of
beams, the form factor K3 shall be taken as

⎛ D 2 + 89400 ⎞
K3 = 0.81 ⎜ 2 ...(13.15)
⎝ D + 55000 ⎟⎠
It is to note that form factor K3 shall not be applied for beams having depth
less than or equal to 300 mm.
(b) Box beams and I-beams. For the box beams and I-beams the form
factor K4 shall be obtained by using the formula

⎛ D2 + 89400 ⎞
K4 = 0 ⋅ 8 + 0 ⋅ 8 γ ⎜ 2 – 1⎟ ...(13.16)
⎝ D + 5500 ⎠
where y = p21 (6 – 8p1 + 3p21) (1 – q1) + q1 ...(13.17)
p1 = Ratio of thickness of the compression flange to the depth of the
beam.
q1 = Ratio of the total thickness of web or webs to the overall width
of the beam.
(c) Solid cross sections. For the beams of solid circular cross-sections,
the form factor K5 shall be taken as 1.18.
(d) Square cross sections. For the beams of square cross-section where
the load is in the direction of diagonal, the form factor K6 shall be taken as
1.414.
The minimum width of the beam or any flexural member shall not be less
1
than 50 mm or of the span whichever is greater. All the flexural member
50
818 DESIGN OF STEEL STRUCTURES

having a depth exceeding three times its width and/or a span exceeding fifty
times its width shall be laterally restrained from twisting or buckling and the
distance between such restraints shall not exceed 50 times its width.
The wooden beams are designed for maximum bending stress. The bending
stress in beams due to dead load and super-imposed load should not exceed the
maximum permissible bending stress. The beams are considered laterally
supported if the depth of beam, d is less than three times its width b, (i.e., d <
3b). When the depth of beam d, is greater than three times its width b, (i.e., d >
3d) or span is greater than fifty times its width b, (i.e., l > 50b), then, the beams
are laterally unrestrained. The lateral buckling occurs in such beams. In order
to prevent lateral buckling of such beams, lateral restraints are provided at a
distance fifty times its widths.
Check of shear. The beams designed for bending are checked for shear. The
maximum horizontal shear stress occurs at the neutral axis and it can be obtained
by
⎛ V ⋅Q ⎞
fsh = ⎜ ⎟ ...(13.18)
⎝ I ⋅b ⎠
where, fsh= Horizontal shear stress in beams in N/mm2
V = Vertical shear at the section in N
b= Width of beam section in mm
D = Depth of beam section in mm
I= Moment of inertia of section in mm4
Q = Statical moment of area above the level under consideration
about neutral axis in mm3
Equation 13.18 gives the general formula for calculating horizontal shear.
For the following types of beams, the maximum shear stress may be found by
using the following formulae :
(a) Rectangular beams
3 ⎛ V ⎞
fsh = ⋅⎜ ⎟ ...(13.19)
2 ⎝b⋅D ⎠
(b) Notched beams, notched at tension face at the support

3 ⎛V ⋅D⎞
fsh = ⋅ ...(13.20)
2 ⎜⎝ b ⋅ D12 ⎟⎠

where D1 = Depth of beam at notch in mm

D2 = Depth of notch.
(c) Notched at the compression face (where, e > D)
3 ⎛ V ⎞
fsh = ⋅ ...(13.21)
2 ⎜⎝ b ⋅ D1 ⎟⎠
 DESIGN OF TIMBER STRUCTURES 819

(d) Notched at compression face (where, e > D)

⎛ 3V ⎞
fsh = ⎜ ⎟ ...(13.22)
⎜⎜ 2b ⎡⎢ D – ⎛⎜ D2 ⎞ ⎤⎟
⎟ e⎥ ⎟
⎝ ⎣ ⎝ D ⎠ ⎦⎠
where, e = Length of the notch along the beam span from the inner edge of
the support to the farthest edge of the notch in mm.
The maximum horizontal shear when the load on a beam moves from the
support towards the centre of the span, and the load is at a distance of three to
four times the depth of beam from the support shall be calculated from the
formula
⎛ V ⋅Q ⎞
fsh = ⎜ ⎟
⎝ I ⋅b ⎠
For rectangular beam
⎛1 D⎞ 1 2
Q = ⎜ × b × D × ⎟ = b ⋅ D and
⎝2 4⎠ 8
1
I = b ⋅ D3
12
⎛ V ⋅Q ⎞ 3 ⎛ V ⎞
∴ fsh = ⎜ ⎟= ⎜ ⎟
⎝ I ⋅b ⎠ 2 ⎝ b ⋅ D ⎠
The value of V shall be calculated from the following formula.
For concentrated loads

⎛ 2
⎛x⎞ ⎞
⎜ 10C(l – x ) ⎜ ⎟ ⎟
⎝D⎠ ⎟
V = ⎜ ...(13.23)
⎜ ⎡ 2⎤ ⎟
⎜⎜ 9l ⎢2 + ⎛⎜ x ⎞⎟ ⎥ ⎟⎟
⎝ ⎣ ⎝D⎠ ⎦ ⎠
where, C = Concentrated load in N
l = Span of beam in mm
x = Distance from reaction to load in mm
For uniformly distributed loads
W⎛ 2D ⎞
V = ⎜l – ⎟ ...(13.24)
2⎝ l ⎠
The maximum horizontal shear stress should not exceed the maximum
allowable shear stress in the wood.
The maximum shear in a simply supported beam carrying uniformly
distributed load occurs at the edge of the support. When the beams are carrying
concentrated loads, then these concentrated loads which are acting in the vicinity
of the supports, are reduced by the percentage mentioned in Table 13.8.
820 DESIGN OF STEEL STRUCTURES

Table 13.8 Percentage of reduction for concentrated loads in the vicinity of supports

Distance of load from the 1.5 D 2D 2.5D 3D or

nearest support or less more

Percentage reduction 60 40 20 No reduction

D = Depth of beam
All the uniformly distributed loads within a distance equal to the depth of
the beam from edge of the nearest support may be neglected.

13.18.1 End Bearing of Beams

The ends of beams are supported in recess and these are not enclosed. The
recess provides proper ventilation. The bearing stress perpendicular to the grain
at supports and under concentrated loads should not be greater than safe working
stress in compression across the grain. When the beams are supported directly
on masonry or concrete’, then a minimum of 75 mm bearing length should be
provided. The timber joists or floor planks supported on top flange of steel beams
should rest on and be secured to a timber plate having a minimum cross-section
of 75 mm × 40 mm. The timber joists should not be supported on the top flange
of steel beams unless the bearing stress calculated on the net bearing as shaped
to fit the beam, is not greater than the permissible compressive stress
perpendicular to the grain.

13.18.2 Check for Deflection

The deflection in case of beam joists, purlins, battens, and other flexural members
supporting brittle material like gypsum ceiling slates, tiles and asbestos sheets
⎛ 1 ⎞
should not be greater than ⎜ ⎟ of the span. The deflection in case of other
⎝ 360 ⎠

1
flexural members should not be greater than of span. For the cantilever
240

⎛ 1 ⎞
beams, the deflection should not be greater than ⎜ ⎟ or span. For checking
⎝ 180 ⎠
up the deflection in case of beams and joists, the loads taken shall be twice the
dead load plus 0.75 times live load.
Notched beams. When a groove is cut either at the ends or at the middle of
span or anywhere in between support in the timber beams, then beams are
known as notched beams. The beams are cut or notched at the ends to reduce
the depth of floors. The ends of beams are also notched to bring top surfaces
level with adjacent beams. Sometimes, the beams are notched to increase the
room clearance. The beams are notched at the middle or anywhere between the
 DESIGN OF TIMBER STRUCTURES 821

supports to provide space for pipes or support to other beams and frames. The
notched beams are shown in Fig. 13.16. The cross sectional area of beams at
notches are reduced.
When the beams are carrying uniformly distributed loads only, then it is not
necessary to calculate the effect of notches if their depth and situations are
within the limits specified by IS : 883 –1970. Unless the local stress is calculated
and found to be within the permissible stress, flexural member shall not be cut,
notched or bored except as follows :
The notches cut at the top or bottom of the beams should not be deeper than
one-fifth of depth of the beam. The notches should not be cut at distances more
than one-sixth of the span from the edges of support. If notches occur at a distance
greater than three times the depth of beam from the edge of nearest support

(a )

(b )

(c)

Fig. 13.16 Notched beams

the net depth should be used in determining the bending strength, since the
modulus of section at the notch reduces. The holes at larger in diameter than
one-quarter of the depth may be bored in the middle third of the depth and
length.
The shear strength of notched beams decreases. The decrease in strength of
notched beams depend upon the shape of notch and on the relation of the depth
of notch to the depth of beam. Concentration of stress occurs at the notch. The
shear stress in case of square notch at the ends as shown in Fig. 13.16 (a) is
given by Eqs. 13.20, 13.21 and 13.22.
Flitched beams. The flitched beams consist of wooden beams and steel beams
joined together by means of bolts or screws. The flitched beams are shown in
Fig. 13.17. When the timber sections are joined together with steel plates, then
the deformations in the fibres of timber and steel in flitched beams are equal.
Consider a flitched beam shown in Fig. 13.17 (b). The modulus of elasticity of
a material is defined as stress divided by unit deformation.
822 DESIGN OF STEEL STRUCTURES

(a ) (b ) (c)
Fig. 13.17 Flitched beams

BI B
Then Es = and Ew = M
∆I ∆M
where, Es = Modulus of elasticity of steel
Ew = Modulus of elasticity of timber
fs = Bending stress in the extreme fibre of steel
fw = Bending stress in the extreme fibre of steel
∆s = Deformation per unit length on the extreme fibre of steel
∆w = Deformation per unit length on the extreme fibre of timber
⎛ f ⎞ ⎛ f ⎞
or ∆s = ⎜ s ⎟ and ∆w = ⎜ w ⎟
E
⎝ s⎠ ⎝ Ew ⎠
The unit deformation on extreme fibre of steel and that on extreme fibre of
timber are equal
⎛ f f ⎞
∆s = ∆w, ⎜ s = w ⎟ ...(13.25)
⎝ Es Ew ⎠

⎛E ⎞
∴ fw = fs × ⎜ w ⎟
⎝ Es ⎠

⎛E ⎞
The ratio of ⎜ w ⎟ is known as the modular ratio.
⎝ Es ⎠
From Eq. 13.25, it is seen that bending stress on the extreme fibre of timber
⎛ Ew ⎞
is equal to the bending stress on the extreme fibre of steel multiplied by ⎜ E ⎟ .
⎝ s ⎠
When the bending stress in extreme fibre of steel is maximum, then the bending
stress in timber on the extreme fibre should be less than or equal to the maximum
allowable bending stress in the timber.
Example 13.12 A deodar timber beam carries a uniformly distributed load
16 kN/m inclusive of self-weight of the beam. The beam is simply supported at
both ends. The clear span of beam is 5 m. Design the timber beam.
 DESIGN OF TIMBER STRUCTURES 823

Solution
Design
Step 1: Effective span
Clear span of beam = 5m
Assume width of bearing at each end
= 300 mm
1 6 kN /m

5m

5 .30 m

Fig. 13.18

Effective span of beam

⎛ 1 1 ⎞
= ⎜ 5 + × 0 ⋅ 30 + × 0 ⋅ 30 ⎟ = 5.30 mm
⎝ 2 2 ⎠
Step 2: Maximum bending moment
⎛ 5 ⋅ 30 × 5 ⋅ 30 ⎞
M = ⎜16 × ⎟ = 56.18 kN-m
⎝ 8 ⎠
The maximum allowable bending stress along the grain for inside location
for deodar wood of standard grade
= 10.2 N/mm2
From factor for rectangular section : K3
Assuming the depth of beam as 400 mm, from Eq. 13.15

⎛ D 2 + 89400 ⎞
K3 = 0.81 ⎜ 2
⎝ D + 55000 ⎟⎠

⎛ 160000 + 89400 ⎞
K3 = 0.81 ⎜ ⎟ = 0.9396
⎝ 160000 + 55000 ⎠
Maximum allowable bending stress, (since D > 300 mm)
= (0.9396 × 10.2) = 9.584 N/mm2
Step 3: Section modulus required
⎛ 5 ⋅ 618 × 1000 × 1000 ⎞
Z = ⎜ ⎟
⎝ 9 ⋅ 584 ⎠
= 5861.85 × 103 mm3
The beam is considered laterally supported if

1 d
b > × span , and b >
50 3
824 DESIGN OF STEEL STRUCTURES

Width of beam required

⎛ 1 ⎞
b =
⎜ × 5 ⋅ 30 × 1000 ⎟ = 106 mm
⎝ 50 ⎠
Adopt b = 240 mm
The modulus of section of rectangular beam
1 ⎛1 ⎞
b ⋅ d2 = ⎜ × 240 × d2 ⎟ = 40 d2
6 ⎝6 ⎠
Therefore,
40 d2 = 586.185 × 103
d = 382.81 mm
Adopt d = 400 mm
d ⎛ 400 ⎞
= ⎜ ⎟ = 133.3 mm < b
3 ⎝ 3 ⎠
Step 4: Check for shear
Maximum shear force at the edge of the support
⎛ 16 × 5 ⋅ 30 ( ⎞
V = ⎜ – 0 ⋅ 15 + 0 ⋅ 4 ) × 16 ⎟
⎝ 2 ⎠
= 33.6 kN
Maximum shear stress in the beam, from Eq. 13.19
3 ⎛ V ⎞
fsh = ×⎜ ⎟
2 ⎝b⋅d ⎠

3 ⎛ 33 ⋅ 6 × 1000 ⎞
= ×⎜ ⎟ = 0.5166 N/mm2
2 ⎝ 240 × 400 ⎠
Step 5: Check for deflection
Maximum deflection

5 ⎡ (16 × 5 ⋅ 30 )(5 ⋅ 30 )3 × (1000 )3 ⎤

ymax = ×⎢ ⎥ = 13.518 mm
384 ⎢ 1 3
9 ⋅ 5 × 1000 × × 240 × (400 ) ⎥
⎣⎢ 12 ⎦⎥
Allowable deflection
⎛ 1 ⎞
⎜ × 5 ⋅ 30 × 1000 ⎟ = 22.1 mm > y . Hence, satisfactory.
⎝ 240 ⎠ max

Step 6: Check for bearing

Reaction at the support
⎛ 16 × 5 ⋅ 30 ⎞
⎜ ⎟ = 42.40 kN
⎝ 2 ⎠
 DESIGN OF TIMBER STRUCTURES 825

Bearing stress at the support

⎛ 42 ⋅ 40 × 1000 ⎞
⎜ ⎟ = 0.578 N/mm2
⎝ 300 × 240 ⎠
Safe working stress in compression perpendicular to the grain
= 2.6 N/mm2
> Bearing stress. Hence, satisfactory.
Provide a rectangular beam 240 mm × 400 mm.
Example 13.13 A simply supported timber beam carries a total uniformly
distributed load of 50 kN inclusive of self weight. The effective span of beam is 8
m. The timber beam is made of sal wood. 300 mm × 50 mm planks are only
available. Design a built-up beam.
Solution
Design :
Step 1: Effective span
The strength of a built-up timber beam is equal to sum of the strength of each
element of the built-up beam.
Total uniformly distributed load inclusive of self weight
= 50 kN
Effective span = 8m
Step 2: Maximum bending moment
⎛ WL ⎞ ⎛ 50 × 8000 ⎞
M = ⎜ ⎟= ⎜ ⎟ = 50 kN-m
⎝ 8 ⎠ ⎝ 8 × 1000 ⎠
From IS : 883 –1970, safe working stress in bending for inside location and
standard grade sal wood = 16.8 N/mm2.
Step 3: Sectional modulus required

⎛ 50 × 106 ⎞
50 50 50 50 Z = ⎜ ⎟ = 2976.19 × 103 mm3
mm mm mm mm ⎝ 16 ⋅ 8 ⎠
Section modulus of 300 mm × 50 mm plank
⎛1 ⎞
Z1 = ⎜ × 50 × 300 × 300 ⎟
⎝6 ⎠
= 750 × 103 mm3
30 0 m m

Number of planks required

Z 3
= 2976 ⋅ 19 × 10 = 3.968
Z1 750 × 103
Provide 4 planks of 300 mm × 50 mm as
shown in Fig. 13.19. The planks are bolted
together. The bolts are staggered. The bolts
Fig. 13.19 are spaced 1.10 m apart (less than four
times the depth of the beam).
826 DESIGN OF STEEL STRUCTURES

Example 13.14 A 150 mm × 300 mm timber beam has an effective span of

5.30 m. It carries a concentrated load of 25 kN at the centre. The beam has 150
mm × 50 mm notches at the ends as shown in Fig. 13.20. Verify that whether the
beam is safe in bending and shear or not. The beam consists of dhaman wood of
Group A.
Solution
Step 1: Effective span and load
Effective span of beam = 5.30 m
Concentrated load at the centre = 25 kN

⎛ 150 × 300 × 5 ⋅ 30 × 7 ⋅ 85 ⎞
Self weight of beam = ⎜ ⎟ = 1.872 kN
⎝ 1000 × 1000 ⎠
Step 2: Bending moment
Maximum bending moment due to concentrated load

⎛ 25 × 5 ⋅ 30 ⎞
M1 = ⎜ ⎟ = 38.125 kN-m
⎝ 4 ⎠
Maximum bending moment due to self-weight

⎛ 1 ⋅ 872 × 5 ⋅ 30 ⎞
M2 = ⎜ ⎟ = 1.24 kN-m
⎝ 8 ⎠
Total bending moment at the centre
M = M1 + M2 = (33.125 + 1.24) = 34.365 kN-m

2 5 kN

5 .30 m

Fig. 13.20 Notched beam

Step 3: Section modulus of beam

⎛1 ⎞
Z = ⎜ × 150 × 300 × 300 ⎟ = 2250 × 103 mm3
⎝6 ⎠
Actual stress in bending

M ⎛ 34 ⋅ 365 × 106 ⎞
fb = = ⎜ ⎟
Z ⎝ 2250 × 103 ⎠
= 15.27 N/mm2 < 18.2 N/mm2
(Safe working stress in bending in dhaman wood)
Hence, the beam is safe in bending.
 DESIGN OF TIMBER STRUCTURES 827

Step 4: Maximum shear force

⎛ 25 1 ⋅ 872 ⎞
V = ⎜ + ⎟ = 13.436 kN
⎝ 2 2 ⎠
Total depth of beam d is 300 mm.
Size of notch = 150 mm × 50 mm
Depth of beam above notch
ds = 250 mm
Step 5: Shear stress in beam, from Eq. 13.20
⎛3V D ⎞
fs = ⎜ ⋅ 2⎟
⎝ 2 b D1 ⎠
3 ⎛ 13 ⋅ 436 × 1000 × 300 ⎞
fs = ×⎜ ⎟
2 ⎝ 150 × 250 × 250 ⎠
= 0.645 N/mm2 < 1.3 N/mm2
(Safe working stress in shear in shear in dhaman wood)
Hence, the beam is safe in shear. The beam is found safe in bending and
shear both.
Example 13.15 A beam is simply supported at its both the ends. The effective
span of beam is 6 mm. It consists of 200 mm × 300 mm teak wood with 300 mm
× 12 mm steel plates bolted to its sides as shown in Fig. 13.21. Determine the safe
uniformly distributed load, which the beam will support.
Solution :
Step 1: Modulus of section of two steel plates
⎛ 1 2⎞
Z = ⎜ 2 × × 12 × 300 ⎟ = 360 × 103 mm3
⎝ 6 ⎠
It is assumed that the beam is laterally supported. For the laterally supported
beam, allowable stress in bending at the extreme fibre, in steel
fb = (0.66 × 250) = 165 N/mm2
Step 2: Moment of resistance of steel plates
⎛ 360 × 103 ⎞
M = fb.Z = 165 × ⎜ ⎟
⎝ 1000 × 1000 ⎠
= 59.4 kN-m
Step 3: Load supported by beam
Let W be the uniformly distributed load, supported
by the steel plates.
WL
∴ = 59.4 ×106
8
Fig. 13.21 Flitched beam
⎛ 59 ⋅ 4 × 106 × 8 ⎞
∴ W = ⎜ ⎟ = 79.2 kN
⎝ 6000 × 1000 ⎠
From IS : 883 –1970, for teak wood, modulus of elasticity = 9600 N/mm2.
Modulus of elasticity for steel = 2.06 × 105 N/mm2.
828 DESIGN OF STEEL STRUCTURES

Step 4: Stress in the extreme fibre of wood, (from Eq. 13.25)

Ew ⎛ 9600 ⎞
fw = f × = 165 × ⎜ ⎟
Es ⎝ 2 ⋅ 06 × 105 ⎠
= 7.689 N/mm2 < 14.0 N/mm2
(Allowable stress in bending for teak-wood)
Modulus of section of wooden beam
⎛1 2⎞
Z1 = ⎜ × 200 × 300 ⎟ = 3000 × 103 mm3
⎝6 ⎠
Moment of resistance of wooden beam

⎛ 7 ⋅ 689 × 3000 × 103 ⎞

M1 = fw × Z1= ⎜ ⎟
⎝ 1000 × 1000 ⎠
= 23.067 kN-m
Let W 1 be the uniformly distributed load, supported by wooden beam
W1 L
= 23.067 × 106
8

⎛ 23 ⋅ 067 × 106 × 8 ⎞
or W1 = ⎜ ⎟ = 30.756 kN
⎝ 6000 × 1000 ⎠
The uniformly distributed load inclusive of the self-weight of flitched beam
W + W1 = (79.2 + 30.756) = 109.956 kN

13.19 FRAMED JOINTS

Two pieces of timber are connected by framed joints as shown in Fig. 13.22. In
framed joints, when two pieces are joined at their ends, then the joints are
known as corner joints or angle joints as shown in Fig. 13.22 (a). When two
pieces are joined in between the length of one or both the members at right
angles, then, the joints are known at T joints as shown in Fig.13.22 (b).
In butt joint, Fig. 13.22 (a) (i), one piece comes squarely against another piece.
In mitre joint, Fig. 13.22 (ii), two pieces are bevelled and joined. The plane of
joint bisects the angle between them as at the cornei of photo frame. In dovetail
joint, Fig. 13.22 (a) (v), one piece has the shape of wedge. It fits into the groove
of corresponding shape in another piece. Wedge shape piece cannot be withdrawn
in the direction of length.
In halving, Fig. 13.22 (b) (i), both pieces are cut to one-half of their depth.
When both pieces are joined, the lower and upper surfaces remain flush. In
bevelled halving, Fig, 13.22 (b) (ii), both pieces are cut and their surfaces of
contact are kept inclined. In tenon and mortise joint, Fig. 13.22 (b) (iii), one
piece has a tenon or rectangular projection on its end. It is inserted into a socket
or mortise in another piece. In double notching, Fig. 13.22 (b) (iv) notches are
 DESIGN OF TIMBER STRUCTURES 829

(i) B u tt joint H a lvin g

(ii) S ho ulde r Joint B evelle d h a lvin g

(iii) M itre join t M ortise an d ten on

(iv) Ton gu e an d gro ove jo ijnt D o ub le n otch in g

(v) D o ve tail joint C o gg in g

(a ) C o rne r joints (b ) Te e-joints
Fig. 13.22 Framed joints

prepared by cutting grooves in the pieces. On piece is inserted into another

piece. In cogging, Fig. 13.22 (b) (v), groove does not extend entirely across the
width of one piece. These joints are used in the frames of doors, windows and
ventilators.

13.20 LAP, SCARF AND FISH PLATE JOINTS

Two pieces of timber are joined along the longitudinal direction by lap joint or
scarf joint or fish plate joint. When joint is provided along the length of a member,
then it is known as splice or lengthening joint. In lap joint, one piece of timber is
placed over another piece and the pieces are joined together by nails or bolts as
shown in Fig. 13.23 (a). Lap joint is used to transmit tension or compression of
small values only. Lap joint is not commonly used. In scarf joint, two timber
pieces are cut away at their ends on opposite sides so that these may be lapped
and joined together by nails or bolts as shown in Fig. 13.23 (b). In scarf joint, the
830 DESIGN OF STEEL STRUCTURES

thickness at joint does not increase. There are various types of scarf joints
depending upon the method of cutting the ends of timber pieces. Figure 13.23 b
(i) shows a simplest type of scarf joint. It is also known as half lap joint. Figure
13.23 b (ii), keys are inserted in between half-lap joint. In Fig. 13.23 (iii) shows
oblique scarf joint. Fig. 1323 b (iv) and (v) show straight tabled scarf joint without
and with keys respectively. Similarly, Fig. 13.23 b (vi) and (vii) show oblique
tables scarf without and with keys respectively.

(a ) L ap jo in t

(i) H a lf lap (scarf) jo in t

(ii) H a lf la p (sca rf) w ith keys

(iii) O bliq ue sca rf joint

(iv) S tra ig ht ta bled sca rf joint

(v) Ta bled sca rf w ith keys

(vi) Ob liqu e ta bled sca rf joint

(vii) O bliq ue ta b led scarf w ith ke ys

(b ) S carf join t
Fig. 13.23

The fish plate joint is also known as butt-joint. In the fish plate joints, two
pieces of timber are placed end to end and cover plates or fish-plates are placed
on them, and these are joined together as shown in Fig. 13.24. The fish-plates
are made of timber. The steel fish-plate are also used for connection. Figure
 DESIGN OF TIMBER STRUCTURES 831

13.24 (i) shows plane fish plate joint. Figure 13.24 (ii) shows tabled fish-joint.
Figure 13.24 (iii) shows indented fish-plate joint. The fish-plate joint is commonly
used or splice in timber member.

(i) P lain fish-p la te jo in ts

(ii) Ta bled fish-late joints

(iii) In de nte d fish plate joints

Fig. 13.24 Fish-plate joints

13.21 BOLTED JOINTS IN TIMBER

The bolts are used for connecting timber members. The bolts are used for
connecting timber-to-timber fastenings. The bolts are also used for connecting
plate fastenings to timber members. The distribution of pressure under the
bolts depend upon the direction of grain of timber.

P
2

P
2

d1 D istribu tion of b ea rin g stress

u nd er the b olt
b1

d1

Fig. 13.25 Bolted joints in timber

832 DESIGN OF STEEL STRUCTURES

Figure 13.25 (a) shows a fish-plate joint. The distribution of pressure under
the bolts is parallel to the grains of timber. The surfaces of member under the
bolts are subjected to bearing pressure.
The bearing pressure under the bolts is not uniform as shown in Fig. 13.25
(b). The bearing pressure is maximum at the edges of the members. If this fish-
plate joint is tested, and slip is measured for successive increase of load, it will
be seen that the slip is proportional to the load upto a certain point. If the load
is increased beyond this point, slip increases more rapidly. This point is known
as proportional limit. The basic bearing pressure in timber parallel to the grain
under the bolt is given by
⎛ P ⎞
p = ⎜ ⎟ ...(13.26)
⎝ LD ⎠
where, P = Load transmitted
L = Length of the bolt in the central member
D = Diameter of the bolt
The allowable bearing pressure in timber parallel to the grain under the bolt
L L
depends upon ratio. As the ratio increases the allowable bearing pressure
D D
parallel to grain decrease. The effect of concentration of stress is taken into
account by decreasing allowable bearing pressure. The allowable bearing
pressure in timber parallel to the grain with metal cover plates is given by
p1 = K 1 . p ...(13.27)
L
where, K 1 = Constant. The values of K1 depend upon ratio.
D
The values of K1 are given in Table 13.9.
Table 13.9 Values of K1

L
Ratio Values of K1
D
1 1.00
2 1.00
3 1.00
4 0.90
5 0.86
6 0.75
7 0.64
8 0.58
9 0.51
10 0.46
11 0.42
12 0.38
13 0.34
14 0.30
 DESIGN OF TIMBER STRUCTURES 833

It is to note that when wooden cover plates are used instead of metal cover
plates, each being half the thickness of the main member, then the allowable
bearing stress in the direction parallel to the grain is taken as 80 per cent of
calculated stress.
For the loading parallel to the grain, the net sectional area should be greater
than 80 per cent of bearing area of bolt in case of soft wood, and 100 per cent of
bearing of bolt in case of hard wood.

c1
b1

a1

P P P
2 2

Fig. 13.26

The distribution of pressure under the bolt in the vertical member is shown in
Fig. 13.26 is parallel to the grain and that in the horizontal member is
perpendicular to the grain. The allowable bearing pressure under the bolt in
timber perpendicular to the grain with metal cover plates is obtained by
q1 = K2.K.q ...(13.28)
where, q = Basic stress in timber across the grain
L
K2 = Constant. It depends upon ratio. The values of K2 are given
D
in Table 13.10.
Table 13.10 Values of K2

L
Ratio Values of K1
D
1 1.00
2 1.00
3 1.00
5 1.00
6 1.00
7 1.00
8 0.93
9 0.88
10 0.80
Contd.
834 DESIGN OF STEEL STRUCTURES

Contd. Table 13.10

L
Ratio Values of K1
D
11 0.68
12 0.61
13 0.55
14 0.51
K 2 = Constant. It is known as diameter factor. It depends upon diameter of
the bolt. The values of K3 are given in Table 13.11.
Table 13.11 Values of K3 (Diameter factor)

L
Ratio Values of K1
D
6 2.50
10 1.95
13 1.68
16 1.52
19 1.41
22 1.33
25 1.27
32 1.19
38 1.14
42 1.10
50 1.07
64 1.03
75 and above 1.00
It is to note that when wooden cover plates are used instead of metal cover
plates, each being half the thickness of the main member, then the allowable
bearing stress in the direction perpendicular the grain is not reduced.
The distribution of pressure under the bolt in the inclined members as shown
in Fig. 13.27 is parallel to the grain, and that in the horizontal member is neither
perpendicular nor parallel to the grain. The allowable bearing pressure under
the bolt in the horizontal member is obtained by Hankinson’s formula

⎛ p1 ⋅ q1 ⎞
N1 = ⎜ ...(13.29)
⎝ p1 ⋅ sin 2 θ + q1 ⋅ cos2 θ ⎟⎠
where, p1 = Allowable bearing pressure in the timber parallel to the grain
in bolted joints
q1 = Allowable bearing pressure in the timber perpendicular to the
grain in bolted joints.
 DESIGN OF TIMBER STRUCTURES 835

Pitch of bolts. When the direction of loading is parallel to the grain, then
the minimum distance a1, (the centre to centre distance between the bolts in
the direction parallel to the loading, i.e., pitch) is 4 times the diameter of bolt.
When the direction of loading is perpendicular lo the grain, then, the minimum
distance as shown in Fig. 13.26 is also 4 times the diameter of bolt.

Fig. 13.27

13.21.1 Gauge of Bolts

When the direction of loading is parallel to the grain, the maximum distance b1,
(the centre to centre distance between the bolts in the direction perpendicular
to the loading, i.e., gauge) as shown in Fig.13.25, depends upon allowable bearing
pressure and net sectional area. The distance, b1 should be such that net sectional
area should be less than 80 per cent of gross sectional area. When the direction
of loading is perpendicular to the grain, then the minimum distance, b1 as shown
in Fig. 13.26 is 2.5 times the diameter of bolt when L/D ratio is upto 2 and 5
times the diameter of bolt when L/D ratio is three or more.

13.21.2 End Distance

The distance from end of timber to the centre of bolt is knows as end distance.
When the load is tensile and parallel to the grain, then the end distance 7 times
the diameter of bolt for soft wood and 5 times the diameter of bolt for hard
wood. When the load is compressive, then, the end distance is 4 times the
diameter of bolt. When the direction of loading is perpendicular to the grain,
then the end distance is 4 times the diameter of bolt.

13.21.3 Edge Distance

The distance from edge of timber to the centre of bolt is known as edge distance.
When the load is parallel to the grain whether it is compressive or tensile, then
the edge distance d1, as shown in Fig 13.25 is 1.5 times the diameter of bolt for
L L
ratio less than or equal to 6. When ratio is greater than 6, edge distance
D D
is taken as half gauge distance. When the direction of loading is perpendicular
to the grain, then, the edge distance is 4 times the diameter of bolt.
836 DESIGN OF STEEL STRUCTURES

Example 13.16 Two 50 mm × 150 mm wooden fish plates are used to splice a
100 mm × 150 mm timber tension member. Deodar wood is used for all elements.
Four 22 mm diameter bolts are used on each side of the splice. Determine the
maximum load that the splice will carry.

50 m m

1 00 m m

50 m m

1 50 m m

Fig. 13.28

Solution
Design :
Length of bolt, L = Thickness of main timber member
= 100 mm
Diameter of bolt, D = 22 mm
L ⎛ 100 ⎞
ratio = ⎜ ⎟ = 4.55
D ⎝ 20 ⎠
⎡ 0 ⋅ 55
From Table 13.9, K1 = ( )⎤
⎢⎣0 ⋅ 90 – 1 ⋅ 00 × 0 ⋅ 90 – 0 ⋅ 86 ⎥⎦ = 0.88
Step 1: Safe working stress in compression
From IS : 883 –1970, safe working stress in compression (bearing) parallel to
the grain for deodar wood
= 7.8 N/mm2
Allowable bearing stress parallel to the grain with metal cover plates
p1 = (0.88 × 7.8) = 6.864 N/mm2
Since wooden cover plates are used,
p1 = (0.80 × 6.684) = 5.49 N/mm2
Load carrying capacity of one bolt
⎛ 22 × 100 ⎞
⎜ 5 ⋅ 49 × ⎟ = 12.078 kN
⎝ 1000 ⎠
Step 2: Load carrying capacity of 4 bolts
(4 × 12.078) = 48.312 kN
The bearing area of 4 bolts
(4 × 22 × 100) = 8800 mm2
80 per cent of bearing of bolts
 DESIGN OF TIMBER STRUCTURES 837

= 0.80 × 88 = 7040 mm2

Diameter of bolt hole = (22 + 3) = 25 mm
Net sectional area of member
100 × (150 – 2 × 25) = 10000 mm2 > 7040 mm2
Step 3: Safe working stress in tension
From IS : 883 –1970, safe working stress in tension along the grain for deodar
wood
= 10.2 N/mm2
Step 4: Safe load carrying capacity

⎛ 10 ⋅ 2 × 10000 ⎞
= ⎜ ⎟⎠ = 102 kN
⎝ 1000
Therefore, the splice will carry maximum load equal to load bearing capacity
of the bolts (being minimum)
= 48.312 kN
Example 13.17 A 150 mm tension member carrier a load of 150 kN. Design
suitable splice for the member. Use 22 mm diameter bolts. The tension member
is made of babul wood.
Solution
Design :
Length of bolt, L = Thickness of main timber member
= 150 mm
Diameter of bolt, D = 22 mm
L ⎛ 150 ⎞
ratio = ⎜ = 6.83
D ⎝ 22 ⎟⎠

⎡ 0 ⋅ 83 ⎤
From Table 13.9, K1 = ⎢0 ⋅ 75 – × (0 ⋅ 75 – 0 ⋅ 64 )⎥ = 0.66
⎣ 1 ⋅ 00 ⎦

80 m m
1 50
mm
80 m m

11 0 11 0
mm mm
= 88 m m = 88 m m

2 50
mm

Fig. 13.29
838 DESIGN OF STEEL STRUCTURES

Step 1: Safe working stress in compression

From IS : 883 –1970, safe working stress in compression (bearing) parallel to
the grain for babul wood = 11.2 N/mm2
Allowable bearing stress parallel to the grain with metal cover plates
(0.66 × 11.2) = 7.392 N/mm2
Since, wooden cover plates are used
(0.80 × 7.392) = 5.914 N/mm2
Step 2: Load carrying capacity of one bolt
⎛ 5 ⋅ 914 × 150 ⎞
⎜ ⎟ = 19.514 kN
⎝ 1000 ⎠
Number of bolts required
⎛ 150 ⎞
⎜ ⎟ = 7.686
⎝ 19 ⋅ 514 ⎠
Provide 8 bolts in 2 rows on either side of splice as shown in Fig. 13.29.
Bearing area of 8 bolts = 8 × 22 × 150 = 26400 mm2
Diameter of bolt hole = (22 + 3) = 25 mm
Net sectional area of member
(250 – 2 × 25) × 150 = 30000 mm2
> 80% Gross-sectional area
Net sectional area is greater than 100% of total bearing area of the bolt. It
suits bolt requirement for hard wood (Babul).
Pitch of bolts, a1 = 4 × D = 4 × 22 mm = 88 mm
Provide a1 = 90 mm
Gauge distance d1 = 124 mm
1
Edge distance d1 = × 124 = 62 mm
2
Provide d1 = 63 mm
End distance = 5D for hard wood = 5 × 22 = 110 mm
From IS : 883 –1970, Safe working stress in tension along the grain for
babul wood = 18.2 N/mm2
Step 3: Safe load carrying capacity of member
⎛ 30000 × 18 ⋅ 2 ⎞
⎜ ⎟ = 516 kN >150 kN. Hence, safe.
⎝ 1000 ⎠
Fish-plates. Thickness of main member =150 mm
Half-thickness = 75 mm
Provide 2 fish-plates 80 mm × 250 mm
Complete design of timber splice is shown in Fig. 13.29.
Example 13.18 Two 50 mm × 250 mm wooden vertical member transmit a
compressive force of 25 kN to a wooden horizontal member 100 mm × 250 mm.
The members consists of deodar wood. Design the bolted joint. Use 19 mm
diameter bolt.
 DESIGN OF TIMBER STRUCTURES 839

Solution
Design

77 m m
96 2 50
mm mm
77 m m
77 m m
96 m m
77 m m

100 m m
50 m m

50 m m
2 50
mm

Fig. 13.30

Length of bolt, L = Thickness of central member = 10 mm

Diameter of bolt, D = 19 mm
L ⎛ 100 ⎞
ratio = ⎜ ⎟ = 5.26
D ⎝ 19 ⎠

L
From Table 13.10, For ratio = 5.26
D
Value of K2 = 1.00
From Table 13.11 for 19 mm diameter of bolt, diameter factor
K2 = 1.41
Load transmitted is perpendicular to the horizontal member.
Step 1: Safe working stress in compression
From IS : 883 –1970, the safe working stress in compression (bearing across
the grain in deodar wood)
q = 2.65 N/mm2
Allowable bearing pressure under the bolt
q1 = K2K3q = (1.00 × 1.41 × 2.65)
= 3.737 N/mm2
Step 2: Load carrying capacity of one bolt
⎛ 3 ⋅ 737 × 19 × 100 ⎞
⎜ ⎟ = 7.099 kN
⎝ 1000 ⎠
Number of bolts required
⎛ 25 ⎞
⎜ ⎟ = 3.052
⎝ 7 ⋅ 099 ⎠
840 DESIGN OF STEEL STRUCTURES

Provide 4 bolts as shown in Fig. 13.30

Step 3: Load carrying capacity of the member
⎛ 2 ⋅ 65 × 100 × 250 ⎞
⎜ ⎟ = 66.25 kN > 25 kN
⎝ 1000 ⎠
Pitch of bolts a1 = 4 D = 4 × 19 mm = 76 mm
Provide a1 = 96 mm
Gauge distance b1 = 5 × D × 19 = 95 mm
Provide b1 = 96 mm
Edge distance C1 = 4 × D = 4 × 19 = 76 mm
Provide c1 = 77 mm
The complete design is shown in Fig. 13.30.
Example. 13.19 Two 50 mm × 250 mm wooden inclined members are
connected to a wooden horizontal member 100 mm × 250 m. The member consist
of deodar wood. Four 19 mm bolts are used for connections. Determine the total
load transmitted by sloping member if the inclination of members with the
horizontal is 45 °

4 5°
50 m m

50 m m

2 N os 5 0 m m x 2 50 m m

Fig. 13.31

Solution
Length of bolt, L = 100 mm
Diameter of bolt, D = 19 mm
L ⎛ 100 ⎞
ratio = ⎜ ⎟ = 5.26
D ⎝ 18 ⎠

⎡ 0 ⋅ 26
From Table 13.9, value of K1 = ( )⎤
⎢⎣0 ⋅ 86 – 1 ⋅ 00 × 0 ⋅ 86 – 0 ⋅ 75 ⎥⎦ = 0.83
From Table 13.10, value of K 2 is 100 From Table 13.11, diameter factor K3 is
1.41
Step 1: Safe working stress in compression
From IS : 883 – 1970.
Safe working stress in compression along the grain for deodar wood
p = 7.8 N/mm2
Safe working stress in compression across the grain for deodar wood
p = 2.65 N/mm2
 DESIGN OF TIMBER STRUCTURES 841

Allowable bearing pressure along the grain

p1 = K1 . p = 0.83 × 7.8 = 6.474 N/mm2
It is reduced to 80 per cent since wooden cover plates are used
(6.474 × 0.8) = 5.18 N/mm2
Allowable bearing pressure across the grain
q1 = K2.K3 · q = (1.00 × 1.41 × 2.65) = 3.74 N/mm2
Allowable bearing pressure along the sloping member

5 ⋅ 18 × 3 ⋅ 74
= 8.69 N/mm2
⎛ 1 1⎞
⎜ 5 ⋅ 18 × + 3 ⋅ 74 × ⎟
⎝ 4 4⎠
Step 2: Load transmitted by one bolt
⎛ 8 ⋅ 69 × 19 × 100 ⎞
⎜ ⎟ = 16.5 kN
⎝ 1000 ⎠
Step 3: Load transmitted by 4 bolts
4 × 16.5 = 66 kN

13.22 NAILED JOINTS IN TIMBER

The nailed joints in timber are suitable for light timber frames such as roof
trusses for spans upto 15 m. The nailed joints in timber are designed as per IS :
2366 –1963, IS : 2363 –1963 gives general guidance and furnishes the strength

O vera ll Le ngth
H e ad B o dy (sh ank) P o in te d e n d

(a ) 1 0.0 0 , 8,00 6 ,3 0,5 ,6 0,5,0 0 ,4,50 ,a nd 4,00 m m

S h an k diam e te r
o ve rall len gth

(b ) 3 .55 , 3 ,15,2,8 0 an d 2,50 m m sh a nk d ia m e te r

O vera ll le ngth O vera ll le ngth

D 75D

(c) 2 .2 4, 2.00 ,1,80 a nd 1,60 m m (d ) 1 .40 a nd 1 .2 5 m m

sha nk d ia m ete r sha nk d ia m ete r
Fig. 13.32 Plain head nails

data. The strength data were made available by Forest Research Institute and
Colleges, Dehradun. The nailed joints in soft timber are safe and economical to
842 DESIGN OF STEEL STRUCTURES

transmit load upto 25 kN and in hard wood upto 50 kN. The nails are used in
nailed joints to connect two pieces of timber. A nail consists of a head at one end
a slender body either tapering or pointed at other end in Fig. 13.22 (a). The
head of nail is used to drive the nail through and into the wood. The nails as
classified as cut nails, wrought nails and wire nails. Cut nails are of rectangular
cross-section and are cut from a metal strip. Wrought nails are forged. Wire
nails are made directly from wire by machine which forms both head and point.
Wire nails are made of mild steel. Wire nails used for nailed joints in timber are
plain-head nails specified in IS : 723 –1961 as shown in Fig. 13.32. Wire nails
may be diamond pointed, or blunt and tapered to blunt-pointed or long sharp
pointed. Diamond pointed nails are used in nailed joints. A diamond point nails
has a point which commonly resembles as octahedron and less commonly a
tetrahedron.
The dimensions of plains head round mild steel wire nails are given in Table
13.12.
Table 13.12 Dimensions of round mild steel wire nails plain head nails

Diameter of shank Diameter of head Length Corresponding

mm* mm* mm** SWG**
10.0 20.00 250 –
8.00 16.00 250, 200 –
6.30 12.00 175, 150 –
5.60 11.60 150, 125 –
5.00 10.20 125, 100 –
4.50 9.00 100, 90 6
4.00 8.00 90, 80 7
3.55 8.00 90, 80, 70, 60 8
3.15 7.10 70, 60,50 9
2.80 6.30 70, 60, 50 10
2.50 5.60 70, 60,50 11
45, 4012
2.24 5.60 70, 60, 50 13
45, 40, 35 41
2.00 5.00 50, 45, 40 51
35, 30,25 61
1.80 4.50 30, 2571
1.60 4.00 25, 20, 15
1.40 3.80 20
1.25 3.40 20
* Dimensions are as per IS : 723 – 1961.
**Values are as per IS : 2366 – 1963.
Note. Rationalized metric values of diameter of shank of nails approximately
equivalent to SWO numbers are shown in Table 13.12.
 DESIGN OF TIMBER STRUCTURES 843

When a nail is double shear, the depth of penetration of the nail in the main
central member shall be not less than two-thirds of the total penetration in the
side members. For nails in single shear, the depth of penetration in the member
containing the nail point shall be not less than two-thirds length of the nail
driven into the joint. For penetration of
nails in soft woods. pre-boring is not
necessary. For penetration of nails woods,
pre-boring is necessary. In hard woods,
holes are drilled and pre-boring is done and
then, the nails are driven. The diameter of
pre-bore should not be greater than 4/5th
diameter of the nail.The nail points after
nailing are finished either by clenching the (a ) N a ils cle n ch ed a cro ss the g rain
nails across the grain as shown in Fig.
13.33 (a) or clenching the nails along the
grains as shown in Fig. 13.33 (b) or by
protruding from the surface of the joint and
cut so as to be flush with the joint face. (b ) N a ils cle n ch ed a lo ng th e gra in
Turning over the projecting point of a nail
Fig. 13.33
so as to be flush with the surface of member
is known as clenching. Clenching increases the holding power of nails and makes
withdrawal more difficult. Clenching perpendicular to (across the) grain gives
greater power to nails than parallel to the grain.
The nailed joints in timber used for structural joints are of various types. Figure
13.34 shows monochord type and split-chord type lap joints, used for lengthening

L en gth en in g jo in t N o de jo in t
(a ) M on och ord typ e lap jo in t

L en gth en in g jo in N o de jo in t
(a ) S p lit cho rd type la p jo in t
Fig. 13.34
844 DESIGN OF STEEL STRUCTURES

joints and node joints. Figure 13.35 shows monochord type and split-chord type
butt joints used for lengthening joints and node joints. In monochord type joint,
the main member transmitting load consists of a single member. In split chord
type joint, the main member consists of more than one member.

L en gth e nin g jo in t N o de jo in t
(a ) M on och ord typ e b utt joint

L en gth e nin g jo in t N o de jo in t

Fig. 13.35

The nails are arranged in a lengthening joint as shown in Fig. 13.36. The
spacing of nails in the direction of grains of wood (pitch) shall not be less than l0
d where, d is the shank diameter of the nail. The spacing between the rows of
nails perpendicular to the grains of wood (gauge) shall not be less than 3d. The

E ffe ctive e d ge distan ce 5 d m in

E ffe ctive e n d d ista nce 10 d m in
Fig. 13.36 Spacing of nails in a lengthening joint subjected to
either tension or compression

effective edge distance for the nails (the distance measured perpendicular to
the grain from the centre of nail to the edge of number) shall not be less than
5d. The loaded or unloaded end of any of the members in a lengthening joint
subjected to either tension or compression shall have an effective end distance
(the distance measured parallel to the grain from the centre of nail to the square
 DESIGN OF TIMBER STRUCTURES 845

end of member) of not less than 10d. These recommendations are as per IS :
2366 – 1963.
The nails are arranged in a node joint members are at right angles to one
another as shown in Fig. 13.37 as per IS : 2366 –1963.

U n lo ad ed
e dg e L oa de d
e dg e
5 d m in 1 0 d m in
1 0 d m in 1 0 d m in
1 0 d m in 1 0 d m in
1 0 d m in 5 d m in

U n lo ad ed
e dg e
m in
m in
m in
m in
m in
m in

m in
m in
m in
m in
3 d m in
3 d m in
L oa de d
e dg e
d
d
d
d
d
d

d
d
d
d
5
3
3
3
3
3

5
3
3
3

Fig. 13.37. Spacing of nails where members at right angles to one another

The nails are arranged in a node joint where members are inclined to one
another at angles other than 90° as shown in Fig. 13.38 as per IS : 2366 –1963.

U n lo ad ed L oa de d
e dg e e dg e

5 d m in 1 0 d m in
1 0 d m in 1 0 d m in
1 0 d m in
1 0 d m in
1 0 d m in
1 0 d m in
1 0 d m in 5 d m in

L oa de d U n lo ad ed
5 d d m in

5 d d m i in

e dg e e dg e
3

3
3 d mi n

3 d m in
m i in

mi n
3 d mi n

3 d m in n
3 d mi
3 d mi

n
5d

5d
m n

d = S h an k diam e te r of n ail
Fig. 13.38 Spacing of nails where members are inclined to
one another at angles other than 90°
846 DESIGN OF STEEL STRUCTURES

Strength of nailed joints. Where number of nails are used in a joint, the
allowable load in withdrawal or lateral resistance is sum of the allowable loads
for the individual nails, provided that the centroid of the group of fixing nails
lies on the axis of the member and the spacing between nails, end distances,
and edge distances are sufficient to develop the full strength of each nail. As far
as practicable, the nails shall be arranged so that the line of force in a member
passes through the centroid of the group of nails transmitting load to it. Where,
this is not practicable, suitable allowance shall be made for any eccentricity in
computing the maximum load on the fixing nails as well as the load and bending
moment in the member. The permissible lateral strength of mild steel-nails
shall be as given in Table 13.13 as per IS : 2366–1963 for Indian species of
timber. These values are for nails with their points cut flush with the faces.
In addition to the above, following specifications as per IS : 2366 – 1963 are
adopted for the design of nailed joints in timber :

Table 13.13 Permissible lateral strength (in double shear), 9 SWG (3.55 mm diameter)
nails of size 80 mm to 100 mm driven in timber
(Note : The values are converted in S.I. units)

For Permanent For Temporary

Construction Construction strength
per nail. (For both
lengthening and node
S. No. Species of wood Lengthening Node joint joint (kN)
joint (kN) (kN)

1* Fir (S) 0.8 0.2 1.2

2 Babul (H) 1.5 1.1 3.4
3 Kala Siris (H) 1.4 0.5 2.2
4* Deodar (S) 0.9 0.4 1.5
5 Eucalyptus (H) 1.0 1.0 3.0
6* Dhaman (H) 1.3 0.5 2.4
7* Mango (S) 1.1 0.9 1.6
8* Chir (S) 1.1 1.0 1.6
9 Sandan (H) 1.7 1.1 1.8
10 Oak (H) 1.1 1.1 2.7
11 Sal (H) 1.0 0.5 0.9
12 Jamun (H) 1.5 1.2 2.5
13 Teak (H) 1.4 0.8 1.3
14 White Chuglam(H) 1.8 0.9 2.1
15* Black Chuglam (H) 2.3 1.0 3.3
16 Sain (H) 1.6 1.6 2.2
Notes. 1. For nails clenched across the grain, the strength may be increased by
20 per cent over the values given in above Table.
* 2. Species required no pre-boring for nail penetration.
3. H = Hard wood S = Soft wood
 DESIGN OF TIMBER STRUCTURES 847

Dimensions of member
1. The minimum thickness of any individual piece of timber (that, is any
single member) shall be 15 mm.
2. The maximum thickness of any individual piece of timber (other than
spacer block) shall be 100 mm.
3. Normally the width of any individual piece of timber shall not exceed
eight times the thickness of member.
Fish plates. Wooden plates which are used in joint to hold the members in
alignment and also to stiffen the joints, are known as fish plates. Fish plates
ransmit load from one member to another. Fish-plates are also known as timber
gussets or timber splice plates.
1. The total combined thickness of the gusset or splice plates (fish-plates)
on either side of a joint in a monochord type construction shall not be less than
5 times the thickness of the main members.
2. The total combined thickness of all spacer block and/or plates including
outer splice plates, at any joint in a split-chord type construction shall not be
less than 15 times the total thickness of all the main members at that joint.
Nails
1 1
1. The diameter of nails shall be between and of the thickness of the
6 4
main member.
2. In a lengthening joint, in order to avoid eccentricity, if any, due to the
hinge action, a minimum member number of 8 nails (i.e., 4 nails on either side
of the joint) shall be provided.
3. In a node joint, a minimum number of 2 nails shall be provided to keep
the member in position.

(a ) W ith dra w a l o f n ail from side g rain

(b ) W ith dra w a l o f n ail from en d g ra in

P

(c) L atera l loa d on n ail o n side g rain P

(d ) L ate ra l loa d on n ail o n en d g ra in

Fig. 13.39
848 DESIGN OF STEEL STRUCTURES

The holding power (strength) of nail depends upon many factors including
kind of nails, its diameter, depth of penetration, the density of timber, the
moisture content of timber, its state of seasoning, thickness of the actual rings.
the proportion of hard wood, the mode of driving nails, the mode of withdrawing
the nails relative to the direction of grain, use of pre-bore holes. and manner of
clenching the nails. Figure 13.39 (a) and (b) show the withdrawal of nails from
side grain and grain, respectively. Figure 13.39 (c) and (d) show the lateral load
on nail on side grain and on end grain respectively.
Withdrawal resistance of nails from side grain and lateral resistance of nails
on side grain are important for structural considerations. The roof covering
materials are fixed to the purlins by nails then, the suction tends to withdraw
the nails. The timber members in roof trusses carry axial forces. The nails in
nailed joints in timber members are subjected to lateral loads. The permissible
withdrawal resistance per nail is given by
R = K . G25 . D ...(13.30)
where, G = Specific gravity of oven dry timber
D = Diameter of nails in mm
⎛ 972 ⋅ 5 ⎞
K = Constant = ⎜ ⎟ = 0.9725
⎝ 1000 ⎠
R = Permissible withdrawal resistance per nail in kN per mm of
penetration. Permissible withdrawal resistance of nail for
various timber species are given in Table 13.14.
Table 13.14 Permissible withdrawal resistance for nail per 10 mm penetration

S. No. Species of Timber Specific gravity Withdrawal Resistance

(G) 0.9725 G25 (kN)
1 Fir (S) 0.465 0.143
2 Babul (H) 0.835 0.619
3 Kala Siris (H) 0.735 0.449
4 Deodar (S) 0.561 0.228
5 Eucalyptus (H) 0.850 0.646
6 Dhaman (H) 0.755 0.482
7 Mango (S) 0.655 0.339
8 Chir (S) 0.575 0.244
9 Sandan (H) 0.865 0.674
10 Oak (H) 0.865 0.674
11 Sal (H) 0.800 0.556
12 Jamun (H) 0.850 0.647
13 Teak (H) 0.625 0.300
14 White Chuglam (H) 0.690 0.382
15 Black Chuglam (H) 0.835 0.619
16 Sain (H) 0.880 0.705
 DESIGN OF TIMBER STRUCTURES 849

The permissible lateral strength (in double shear) for nail that has its point
cut flush with the surface is given by the following :
1. For lengthening joint for permanent construction
⎛k ⋅D⎞
R1 = ⎜ 1 ⎟ ...(13.31)
⎝ 1000 ⎠
2. For node joint permanent construction
⎛ k2 ⋅ D ⎞
R2 = ⎜ ⎟ ...(13.32)
⎝ 1000 ⎠
3. For lengthening joint and node joint for temporary construction
⎛ k3 ⋅ D ⎞
R3 = ⎜ ⎟ ...(13.33)
⎝ 1000 ⎠
where, R1,R2,R3 = Permissible lateral strengths per nail (in double-shear)
K1 K2, K3 = Constants depending upon the type of timber
D = Diameter of nails in mm
The nails are cut flush with the surface of timber.
In case the nails are driven in unseasoned timber, which will dry and season
under the load, the permissible withdrawal resistance per nail as obtained by
Eq. 13.30 is reduced to 75 percent and the permissible resistance per nail as
obtained by Eqs. 13.31,13.32 and 13.33 are reduced to 25 percent.
The values given in Table 13.13 for permissible lateral strength (in double
shear) are applicable for 9 SWG (3.55 mm) diameter nails. In general the
permissible lateral strength (in double shear) for nails may be found from Eqs.
13.31, 13.32 and 13.33. The values of constants K 1 K2 and K3 for some species of
wood have been given in Table 13.15.
Example 13.20 A 35 mm × 150 mm tension member carries an axial load of
50 kN. Design the suitable splice for the member. The tension member is made of
babul wood. Use standard wire nails.
Solution
Design :
Thickness of tension member = 35 mm
1 1
Diameter of nail = and th thickness of member
6 4

⎛ 35 35 ⎞
= ⎜ to ⎟ = (5.83 to 8.75) mm
⎝ 6 4 ⎠
Use 8 mm diameter wire nails, with 200 mm length. The excess length of nail
is clenched across the grain. Assume that lengthening joint is used in permanent
construction. From Eq. 13.31, the permissible lateral strength (in double shear)
⎛ K1 ⋅ D ⎞
R1 = ⎜ ⎟ kN.
⎝ 1000 ⎠
850 DESIGN OF STEEL STRUCTURES

Note. For nails clenched across the grain, the values of constants may be increased
by 20 per cent over the values given in Table 13.15.

Table 13.15 Values for constant K1, K2 and K3 or lateral strength of nails in double shear

For Permanent For Temporary

S. No. Species of Construction Construction strength
wood per nail. (For both
Lengthening Node joint lengthening and node
joint (kN) (kN) joint (kN)
1 Fir (S) 2.26 5.65 3.38
2 Babul (H) 4.22 3.10 9.59
3 Kala Siris (H) 3.94 1.41 6.20
4 Deodar (S) 2.54 1.125 4.22
5 Eucalyptus (H) 4.79 2.82 8.43
6 Dhaman (H) 3.66 1.41 6.76
7 Mango (S) 3.10 2.54 4.52
8 Chir (S) 3.10 2.82 4.52
9 Sandan (H) 4.79 3.10 5.08
10 Oak (H) 3.10 3.10 7.62
11 Sal (H) 2.82 1.41 5.36
12 Jamun (H) 4.22 3.38 7.04
13 Teak (H) 3.95 2.26 3.66
14 White Chuglam (H) 5.08 2.54 7.92
15 Black Chuglam (H) 6.49 2.82 9.30
16 Sain (H) 4.54 4.50 8.18
H = Hard wood S = Soft wood
From Table 13.15, for babul wood,
K1 = 4.220
R1 = (4.220 × 0.8) = 3.376 kN
The nails are clenched across the grain. Therefore, the permissible lateral
strength is increased by 20 percent.
Step 1: Lateral strength per nail in double shear
(l.20 × 3.376) = 4.05 kN
Force in member = 50 kN

⎛ 50 ⎞
Number of nails = ⎜ ⎟ = 12.34
⎝ 4 ⋅ 50 ⎠
Provide 15 nails on either side in three rows as shown in Fig. 13.40.
 DESIGN OF TIMBER STRUCTURES 851

S p lice p la te

28 m m

35 m m

28 m m
M ain m e m be r S p lice p la te
P = 5 0 kN

P P

Fig. 13.40

Step 2: Arrangement of nails

Effective end distance = 10 × Diameter of nail (Min.)
= (10 × 8) = 80 mm
Effective edge distance = 5 × Diameter of nail (Min.)
Pitch of nails = 10 × Diameter of nail (Min.)
= (10 × 8) = 80 mm
Gauge distance for nails
3 × Diameter of nail (Min.) = 3 × 8 = 24 mm
Provide 40 mm gauge distance of nails. Thickness of splice plates
= 28 mm
Step 3: Safe working stress in tension
Babul wood does not require preboring. From IS : 883 –1970, safe working
strength in tension along the grain for babul wood
= 18.2 N/mm2
Step 4: Strength of tension member
⎛ (18 ⋅ 2 × 35 × 160 ) ⎞
⎜ ⎟ = 101.92 kN > 50 kN
⎝ 1000 ⎠
The complete design of splice (lengthening joint) is shown in Fig. 13.40.
Example 13.21 Two 45 mm × 250 mm wooden vertical members transmit a
compression force 25.4 kN to a wooden horizontal member 60 mm × 250 mm.
The members consist of babul wood. Design the nailed joint.
Solution
Design:
Thickness of member carrying compression perpendicular to the grain
= 60 mm
852 DESIGN OF STEEL STRUCTURES

1
Diameter of nail = th thickness of member
6
1
= × 60 = 10 mm
6
Use 10 mm diameter wire nails with 250 mm length. The excess length of
nail is clenched across the grain.
Assume that the node joint is used in permanent construction. From Eq.
13.32, the permissible lateral strength of nails (in double shear)
R2 = K2 . D
From Table 13.15, K2 = 3.100 for babul wood
R2 = (3.100 × 10) = 31 kN
The nails are clenched across the grains. Therefore, the permissible lateral
strength is increased by 20 percent.
Step 1: Lateral strength per nail in double shear
= (1.20 × 310) = 3.72 kN
Force in member = 25.40 kN
⎛ 25 ⋅ 40 ⎞
Number of nails = ⎜ ⎟ = 0.683
⎝ 37 ⋅ 2 ⎠
Provide 8 nails as shown in Fig. 13.41.

1 00 m m
2 50
1 00 m m mm
50 m m
45 m m

60 m m

45 m m
mm
mm
mm
mm
mm
50
50
50
50
50

2 50 m m

Fig. 13.41

Step 2: Arrangement of nails

Effective edge distance
5 × Diameter of nails (Min.) = 5 × 10 = 50 mm
Pitch of nails = 10 × diameter of nail (Min.)
= 10 × 10 = 100 mm
Gauge distance of nails
3 × Diameter of nails (Min.) = 3 × 10 = 30 mm
 DESIGN OF TIMBER STRUCTURES 853

Provide 50 mm gauge distance for nails. The horizontal member is subjected

to compression perpendicular to the grain. From IS : 883 –1970, safe vorking
stress in compression perpendicular to the grain for babul wood = 6.55 N/mm2
Step 3: Strength of member
⎛ 60 × 250 ⎞
⎜ 6 ⋅ 55 × ⎟ = 98.25 kN > 25.40 kN
⎝ 1000 ⎠
The complete design of nailed joint is shown in Fig. 13.41.

13.23 DISC DOWELLED JOINTS IN TIMBER

In disc-dowelled joints in timber, circular grooves are made in the members
transmitting the load and a circular disc is placed in the grooves as shown in
Fig. 13.42.
Figure 13.42 (a) shows a lap joint using a circular disc, and Fig. 13.42 (b)
shows a butt-joint using circular disc in disc-dowelled joints in timber. Hard
timber like babul is used for preparing the discs. The directions of grains of
discs are kept parallel to the direction of loading. The force is transmitted in
disc-dowelled joint partly by shear along the middle surface and partly by bearing
along the end. Bolt is used to keep the members together. Bolt carries tensile
load. Disc-dowelled joints are used to transmit large forces.

(a ) L ap jo in t u sin g d isc-do w el

(b ) B u tt jo in t u sing d isc-d o w el

Fig. 13.42 Disc-dowelled joints

Consider a lap joint using a disc-dowel as shown in Fig. 13.43, to transmit

force from one member to the other. The forces, P, P in members tend to rotate
the dowel in anti-clockwise direction. As a result of this, dowel tends to separate
the members. Bolt prevents the separation of members and keep the member
together. Surface AB and CD of the dowel are subjected to bearing pressure.
The bearing pressure is zero at the toe A of surface AB and toe D of surface CD.
The points B and C of surfaces AB and CD are subjected to maximum bearing
pressure fb max. The bolt is subjected to tensile load. The maximum bearing
pressure in dowel and tensile load in bolt can be found as under.
854 DESIGN OF STEEL STRUCTURES

Consider an elementary strip dx along EF at a distance X from F as shown in

⎛D ⎞
Fig. 13.43 (b). The distance of elementary strip from centre, O is ⎜ − x ⎟
⎝2 ⎠

P A m ax
+ 38 D + 28 D + 38 D +
F D
C P

(a )

M
1/ 2
D ⎡ 2 2⎤
2 2 ⎢⎛⎜ D ⎞⎟ − ⎛⎜ D − x ⎞⎟ ⎥
⎣⎝ 2 ⎠ ⎝ 2 ⎠ ⎦
P P
E O F
x
D
( – x)
2
N
dx
(b )

Fig. 13.43

Therefore, the width of strip MN

1/2
⎡ D2 ⎛ D 2
⎞ ⎤
= 2 ⎢ −⎜ −x⎟ ⎥ ...(i)
⎣ 2 ⎝2 ⎠ ⎦
The bearing pressure on the elementary strip
⎛D−x⎞
= ⎜ ⎟ ⋅ fb max ...(ii)
⎝ D ⎠
The bearing pressure force on the elementary strip
1/2
⎛D−x⎞ ⎡ ⎛ D ⎞2 ⎛ D 2
⎞ ⎤
δF1 = ⎜ ⋅ f 2 ⎢
⎟ b max ⎜ ⎟ ⎜ − − x ⎟ ⎥ dx ...(iii)
⎝ D ⎠ ⎣⎝ 2 ⎠ ⎝ 2 ⎠ ⎦
Total bearing force acting on the surface AB
1/2
2 fb max
D ⎡ ⎛ D ⎞2 ⎛ D 2
⎞ ⎤
F1 =
D ∫ ( D − x ) ⎢ ⎜ ⎟ ⎜ −
⎣⎝ 2 ⎠ ⎝ 2
− x ⎟ ⎥
⎠ ⎦
dx ...(iv)
0

D 1/2
2 fb max 2
=
D ∫ ( D − x ) ⎡⎣( Dx − x )⎤⎦ dx
0
 DESIGN OF TIMBER STRUCTURES 855

D
2 fb max 1/ 2
= ∫ (D − x ) ⋅ ( D − x )x1 / 2 ⋅ dx
D 0

D
2 fb max 3/2
= ∫ (D − x) ⋅ x1 / 2 ⋅ dx ...(v)
D 0

Substitute x = D sin2 θ, dx = (2D sin θ . cos θ).dθ

x1/2 = (D1/2 sin θ), (D – x) = D. cos2 θ
Therefore,
π/2
2 fb max
F1 = ∫ D3 / 2 (cos2 θ ⋅ D1 / 2 sin θ ⋅ 2D sin θ ⋅ cos θ) ⋅ dθ
D 0

π/2
4 fb max
= ∫ D3 / 2 sin θ ⋅ cos4 θ) ⋅ dθ
D 0

2
= 4f ⎛ 2 π ⎞ ⎛ π ⋅ fb max⋅ D ⎞ ...(vi)
b max ⋅ ⎜ D ⎟ = ⎜ ⎟
⎝ 32 ⎠ ⎝ 8 ⎠
Similarly a force F1 is also acting in opposite direction on the surface CD.
⎛3⎞
Two forces act at a distance ⎜ ⎟ Z from the edges. These forces act at a distance
⎝8⎠
2D 2D
apart. Two equal and opposite forces acting at a distance apart from a
8 8
couple M.

2D ⎛ π ⋅ fb max ⋅ D2 ⎞ 2D
M = F1 ⋅ =⎜ ⎟⋅ ...(vii)
8 ⎝ 8 ⎠ 8

⎛ π ⋅ fb max ⋅ D3 ⎞
∴ M = ⎜ ⎟ ...(viii)
⎝ 32 ⎠
This couple is balanced by bearing forces, F, F acting on the ends of dowel at
t
a distance apart.
2

1 ⎛ π ⋅ fb max ⋅ D3 ⎞
∴ F· = ⎜ ⎟⎠ ... (ix)
2 ⎝ 32

⎛ 16 F ⋅ t ⎞
∴ fb max = ⎜ ⋅ 3 ⎟ ... (13.34)
⎝ π D ⎠
where t = Thickness of dowel.
856 DESIGN OF STEEL STRUCTURES

The tensile load in the bolt is equal to F1

From expression (ix)
⎛ π ⋅ fb max ⋅ D3 ⎞ 2
⎛ F ⋅ t ⎞ ⎛ π ⋅ fb max ⋅ D ⎞ 2 F ⋅ t
⎜ ⎟ = ⎜ ⎟⎜ ⎟= ...(x)
⎝ 32 ⎠ ⎝ 2 ⎠⎝ 8 ⎠ D
For expression (vi)
⎛ π ⋅ fb max ⋅ D3 ⎞ ⎛ 2F ⋅ t ⎞
F1 = ⎜ ⎟ , ∴ F1 = ⎜ ⎟ ...(13.35)
⎝ 32 ⎠ ⎝ D ⎠
In lap joint, F = P
In a butt joint, dowels are placed on both the sides of the members. The force
F
transmitted by each dowel F = . The maximum bearing pressure on dowel
2
and tension on the belt are given by Eqs. 13.34 and 13 .35, respectively.
The maximum bearing stress on the dowel should not exceed the safe bearing
stress of the timber. To utilize the full strength of dowel, the strength of dowel
in shear and in bearing are kept equal. For this purpose, the diameter of dowel
is kept 3.0 to 3.5 times the thickness of dowel. The minimum thickness of dowel
kept is 25 mm. The following specifications are followed for disc-dowelled joint
in timber in addition to above :
(1) The effective edge distance (the distance measured from the edge of
⎛D ⎞ ⎛D ⎞
member to centre of dowel) is ⎜ + 12 ⎟ mm to ⎜ + 20 ⎟ mm depending upon
⎝2 ⎠ ⎝2 ⎠
the size of dowel.
(2) The effective end distance (the distance measured from the end of
member to dowel) is 1.5D in case of tension member and 1.25D in case of
compression members. For timber, which are weak in shearing strength, the
effective end distance is determined from safe working stress in shear for those
timbers.
⎛1 ⎞
(3) The minimum thickness of a member in case of lap joint is ⎜ t + 12 ⎟
⎝2 ⎠
mm and in case of butt joint, it is (t + 12) mm, where t is the thickness of the
dowel.
Example 13.22 A teak wood member 80 mm × 240 mm carries an axial pull
of 40 kN. Design a suitable disc-dowelled joint for the member.
Solution
Design :
A butt joint as shown in Fig. 13.44 with dowels, is used for the splice in the
member.
Thickness of member = 80 mm
Maximum thickness of member in case of butt joint
= (t + 12) mm
where, t = Thickness of dowel
Let the thickness of dowel
t = 40 mm
 DESIGN OF TIMBER STRUCTURES 857

Minimum thickness of member

(40 + 12) = 52 mm < 80 mm
The diameter of dowel
D = 3 × thickness = (3 × 40) = 120 mm

40 m m

4 0 kN 4 0 kN 80 m m

40 m m

D o w els 40 m m x 12 0 m m B o lts 1 6 m m diam eter

120 m m
2 40 m m
120 m m

2 00 2 00 2 00 2 00
mm mm mm mm
Fig. 13.44

Step 1: Safe working stress

Let the dowel be made of babul wood. From IS : 883–1970 safe working stress
in shear along the grain in babul
= 2.22 N/mm2
Strength of dowel in shear along the grain
⎛ π (120)2 × 2.22 ⎞
⎜ × ⎟ = 25.095 kN
⎝4 1000 ⎠
From IS : 883–1970,
Safe working stress in bearing in teak wood
= 8.8 N/mm2
Strength of dowel in bearing
⎛ 40 8.8 ⎞
⎜120 × × ⎟ = 21.12 kN
⎝ 2 1000 ⎠
Dowel value = 21.12 kN
Step 2: Number of dowels required
⎛ 40 ⎞
⎜⎝ ⎟ = 1.894
21.12 ⎠
Provide two dowels on each side as shown in Fig. 13.44.
Force in each dowel
40
F =
= 20 kN
2
Maximum bearing pressure in the dowel from Eq. 13.34,
⎛ 16 F ⋅ t ⎞ ⎛ 16 × 20 × 1000 × 40 ⎞
fb max = ⎜ × 3 ⎟ = ⎜ ⎟
⎝ π D ⎠ ⎝ π × 1203 ⎠
= 2.36 N/mm2
858 DESIGN OF STEEL STRUCTURES

From IS : 883–1970, safe bearing pressure perpendicular to the grain in teak

wood
= 4 N/mm2 > 2.36 N/mm2. Hence, safe.
Tensile force in the bolt from Eq. 13.35,
⎛ 2F ⋅ t ⎞ ⎛ 2 × 20 × 40 ⎞
F1 = ⎜ ⎟ = ⎜ ⎟ =13.33 kN
⎝ D ⎠ ⎝ 120 ⎠
⎛ 13.33 × 1000 ⎞
Net area required = ⎜ ⎟ = 88.89 mm
2
⎝ (0.6 × 250) ⎠
Provide 16 mm diameter bolts
Net area provided = 156 mm2 > 88.89 mm2
Step 3: Arrangement of dowels
Edge distance provide =120
Effective edge distance required
⎛D ⎞ ⎛ 120 ⎞
⎜ + 20 ⎟ = ⎜ ⎟ + 20 = 80 mm < 120 mm
⎝2 ⎠ ⎝ 2 ⎠
Effective end distance required for tension member
1.5D = 15 × 120 =180 mm
Provide effective end distance = 200 mm
Step 4: Tensile strength of member
⎛ 14(240 × 80 − 40 × 120) ⎞
⎜ ⎟ = 201.6 kN > 40 kN
⎝ 1000 ⎠
Provide 2 fish-plates 40 mm × 240 mm on both the sides as shown in Fig.
13.44.

13.24 METAL CONNECTORS IN TIMBER

The metal connectors in timber are also known as modern timber connectors.
The metal connectors are basically metal rings or fabricated plates. The metal
rings or fabricated plates are embedded in the faces of contact of two adjacent
members. The members are held together by small diameter bolts. These
connectors are used to transmit the load from one member to the other. The
various metal connectors in timber are described below.

13.24.1 Split Rings

The split rings used as metal connectors in timber as shown in Fig. 13.45. The
split rings are most commonly used. The split rings are made of low-carbon steel.
The split rings are placed between two pieces of timber. The circular grooves for
split rings are pre-cut in the members by a special tool. The diameter of grooves
are slightly more than diameters of rings.
The rings are spread and then inserted in the grooves. The rings remain
tight in the grooves. The half width of ring remains in one member and other
half width remains in the other member. The inner surface of ring has bearing
against the core left in the groove. The outer surface of ring has bearing against
 DESIGN OF TIMBER STRUCTURES 859

Fig. 13.45 Split rings

the outer wall of groove. After embedding the rings into the grooves, the timber
members are held together with small diameter bolts and washers. The bolts
are kept axial with the rings. The single split ring and multiple split rings used
to transmit the load are shown in Fig. 13.46 (a), (b), respectively.

(a ) S ing le split rin g

(b ) M ultip le split rin gs

Fig. 13.46 Multiple split rings

Figure 13.47 shows a section through split ring connectors joining two pieces
of timber. The distribution of bearing pressure of the ring against the core inside

(a )

(b )
Fig. 13.47
860 DESIGN OF STEEL STRUCTURES

the ring and against the wood outside the ring is shown in Fig. 13.47 (a). The
bearing area against the wood outside the ring is at the compression side of the ring.
The shaded area shown in Fig. 13.47 (b) is subjected to shear. The permissible
loads for split ring connectors are based upon information obtained by tests.
The permissible load for split ring connectors depends upon species of timber,
size of timber members, moisture content in timber, defects in timber, size of
split rings, diameter of bolt, spacing of split rings, edge distance, end distance,
and direction of load relative to the grain of timber.

13.24. 2 Toothed Ring

The toothed ring used as metal connector in timber is shown in Fig. 13 .48. The
toothed ring transmits the load in the manner similar to the split-ring. The
groove is placed in between surfaces of contact of two adjacent timber members.
The timber members are pressed by applying pressure by an alloy steel bolt of
high tensile strength. The bolts are kept axial with the ring. The sharp teeth of

Fig. 13.48

toothed ring penetrated into the timber members. The hole of bolt is made 2
mm over size. So that the alloy steel bolt can be removed easily. An ordinary
steel bolt is then placed in position. The bolt keeps the members together. The
single toothed ring and multiple toothed ring used to transmit the load are
shown in Fig. 13.49 (a) and (b) respectively.

(a ) S ing le to othe d rin g

(b ) M ultip le to o the d rin g

Fig. 13.49
 DESIGN OF TIMBER STRUCTURES 861

13.24.3 Bulldog Plates

The bulldog plates used as metal connectors are shown in Fig. 13 50. The teeth
in bulldog plates are distributed over its entire surface. The bulldog plates are

Fig. 13.50

embedded in timber member by bolt pressure in the manner similar to that of

toothed ring. The bolt is kept in the centre of the connector.

13.24.4 Shear Plate Connectors

The shear plate connectors used as metal connectors are shown in Fig. 13.51.
These are also called flanged plate connectors. The shear plate-connectors are

C a st iro n P re sse d stee l

Fig. 13.51

made of pressed steel or cast iron. A hub is provided in the shear plate connector
for bearing against the bolt. The grooves are precut in timber members to embed
the shear connectors.
The shear plate connectors are used singly in a timber to metal joint or back
for a timber to timber joint as shown in Fig. 13.52 (a) and (b), respectively.

(a )

(b )
Fig. 13.52
862 DESIGN OF STEEL STRUCTURES

13.24.5 Claw-Plate Connectors

The claw-plate connectors used as metal connectors are as shown in Fig. 13.53.
These are made of cast iron. These connectors transmit the load in the same
manner as shear plate connectors. The grooves are precut in the surfaces of

Fig. 13.53 Claw-plate connectors

timber members. The teeth of claw plates are pressed by bolt pressure. The
connectors are made as male type connector and female type connector.
The male type connector has a hub. It is used for timber to metal joint as
shown in Fig. 13.54 (a). The shear force is transmitted by the hub and not by the

M etal plate

(a)

(b)

Fig. 13.54

bolt. The male connector is used with the female or plain connector for timber to
timber joint as shown in Fig. 13.54 (b).

13.24.6 Spike Grids

The spike grids used as metal connectors are shown in Fig. 13.55. Fig 13.55 (a)
shows flat spike grid. Figures 13.55 (b) and (c) show single curved spike grid,
respectively.
 DESIGN OF TIMBER STRUCTURES 863

(a ) (b )

(c)

Fig. 13.55

The spike grids are used with piles, poles of trestles, piers whavers, jetties
and transmission lines. The flat spike grid is used to connect sawn faces of
timber pieces as shown in Fig. 13.56. Single curved spike grid is used to join flat
and a curved face, and a double curved spike grid is used to joint two curved
faces. The spike grids are pressed in timber members.

Fig. 13.56 Spike grid

13.25 TIMBER ROOF TRUSSES

The timber roof trusses are built principally of wood. In the past, timber roof
trusses were used to employ steel rods for its tension web members. Now-a-
days timber trusses can be built-in which all the members are of timber. King
864 DESIGN OF STEEL STRUCTURES

post truss and Queen post truss shown in Fig. 9.1 (a) and (b) are wooden trusses.
Howe truss shown in Fig. 9.1 (b) is made of combination of wood and steel. The
vertical members of Howe roof truss are tension members and are made of steel.
As a result of developments in timber engineering and work carried out at Forest
Research Institute and College, Dehradun, it is possible to built all types of
trusses with wood. The timber roof trusses are classified in three categories,
namely, light weight, medium weight and heavy weight roof trusses. The timber
roof trusses using corrugated galvanized iron sheets as roof covering material
are called light weight trusses. The timber roof trusses using asbestos cement
sheets as roof covering material are termed as medium weight roof trusses. The
timber roof trusses using earthen tiles are known as heavy weight roof trusses.
The spans are classified into the following three categories:
1. Small spans (Upto 6 m)
2. Medium spans (> 6 m, but < 12 m)
3. Large spans (>12 m)
Indian timbers are classified into the following three groups depending upon
the various mechanical strength properties of wood which influence the design.
1. Super group E > 12.600 kN/mm2, ft , > 18.0 N/mm2
2. Standard group E > 9.800 kN/mm2,
α < 12.600 kN/mm2, ft , > 12.0 N/mm2
3. Ordinary group E > 5.600 kN/mm2,
α < 9.800 kN/mm2, ft , > 8.5 N/mm2
where, E = modulus of elasticity of timber species
α = modulus of rupture of timber species
ft = extreme fibre stress in bending and tension along the grain.
The timbers of ordinary group are recommended for small spans. The timbers
of standard group are recommended for medium spans and those of super group
are recommended for large spans.
The various joints used for connections in timber roof trusses are framed
joints, bolted joints, nailed joints, and disc-dowelled joints. The metal connectors
are also used for connections at the joints in timber roof trusses. The nailed
joints are suitable for loads and span upto 14 m. The bolted joints are suitable
for medium to large spans. The disc-dowelled joints or metal connection with
small diameter bolts are used for large spans. The web compression members
(discontinuous struts) are made of single members and are inserted in between
the other members at the joints in trusses. The web tension members are made
of double members. The members in principal rafter (continuous struts) and
lower chord members (main ties) are also made of double members. The members
in principal rafters and in main tie members are spliced if necessary in between
the panel points. The various types of joints at the support used for timber
trusses are shown in Fig. 13.57.
 DESIGN OF TIMBER STRUCTURES 865

(a )
P rincip a l
B o lts ra fte r

L ow e r
cho rd
M aso nry C o rbe l
w a ll
(b )

B o lt P rincip a l
ra fte r
Fo rg ed sho e
B o lt B o lt
Key

Key L ow e r
Key
cho rd
M aso nry B o lt
w a ll
(c)

Fig. 13.57

The principal of design of trusses and various loads which will be acting on
the roof trusses were discussed in Chapter 9. The order of design of various
elements of roof trusses was also discussed in Sec. 9.16.
Example 13.23. At the apex A of a wooden roof truss, three member AB, AC
and AD meet as shown. Forces in AB and AD arc 42.26 kN (compressive) and in
AC 11.31 kN tensile. The unsupported length of the members AB and AD are 2
metres. Design and sketch the wooden members the joint.
Take the following allowable stresses in timber :
Compression along the grain = 10.0 N/mm2
Compression across the grain = 4.0 N/mm2
Tension along the grain = 16.0 N/mm2
Shear stress = 1.6 N/mm2
Solution Design
Step 1: Principal Rafters
Members AB and AD are principal rafters. Forces in AB and AD = 42.36 kN
(compression).
866 DESIGN OF STEEL STRUCTURES

Unsupported length of AB and AD = 2 m

Provide a spaced column. Let the least dimension of member d be 50 mm.
A

6 0° 6 0°

B C

Fig. 13.58

Slenderness ratio of members,

s ⎛ 2000 ⎞
= ⎜ ⎟ = 40 > 11
d ⎝ 50 ⎠
Assuming, the modulus of elasticity of wood,
E = 10000 N/mm2
Allowable compression along the grain
C = 10.0 N/mm2
1/2
⎡10000 ⎤ s
K = 0.702 ⎢ = 22.2, >K
⎣ 10.0 ⎥⎦ d
Permissible stress on column

P 0.329 E ⎛ 0.329 × 10000 ⎞

= 2
=⎜ ⎟ = 2.056 N/mm2
A ⎛s⎞ ⎝ 40 × 40 ⎠
⎜ ⎟
⎝d⎠
Area required for column
⎛ 42.36 × 1000 ⎞
⎜ ⎟ = 20603.11 mm2
⎝ 2.056 ⎠
Area required for plank is 10301.555 mm2
Least dimension required is 50 mm
Width of plank required
⎛ 13301.555 ⎞
⎜ ⎟ = 206.03 mm
⎝ 50 ⎠
Provide 2 planks 50 mm × 210 mm
Step 2: Tension members
Force in member AC = 11.31 kN (Tensile). The nailed joint is used for
connection. Assuming that preboring is not necessary for driving the nails.
Therefore, the gross area and net area of members are same.
 DESIGN OF TIMBER STRUCTURES 867

40 m m

40 m m
40 mm

40 mm
40 m m

40 m m
40 m m

40 m m
40 m m

40 m m
30 m m 40 mm 30 mm
30 m m 40 mm 30 mm
30 m m 40 mm 30 mm
30 m m 1 0 m m th ick 30 mm
40 mm
30 m m splice plate 30 mm
18 mm
18 mm
18 mm
2 × 50 m m × 21 0 m m 2 × 50 m m × 21 0 m m
spa ce d m em b er spa ce d m em b er

Fig. 13.59

Allowable tension along the grain is 16.0 N/mm2

Area required for member
11.31 × 1000
= 706.88 mm2
16.0
Provide 50 mm × 20 mm member
Area provided = 50 × 20 = 1000 mm2 > 706.88 mm2
Step 3: Design of joint A. Thickness of member AB, and
AD = 20 mm
1 1
Diameter of nails = to the thickness
4 6

⎛ 20 20 ⎞
⎜ to ⎟ mm = ( 5 to 3.33) mm
⎝ 4 6 ⎠
Provide 3.55 mm diameter (9 SWG) wire nails. Assuming the lateral strength
of nails of size 9 SWG in double shear for permanent construction = 1000 N
Number of nails required for members
⎛ 42.36 × 1000 ⎞
AB (and AD) = ⎜ ⎟ = 42.36
⎝ 1000 ⎠
Provide 48 nails
Step 4: Arrangement of nails
Spacing of nails (pitch)
10 × diameter of nail (Min.) = 10 × 3.55 = 3.55 mm
Provide nails at 40 mm pitch
Edge distance 5 × diameter of nail (Min.)
(5 × 3.55) = 17.75 mm
868 DESIGN OF STEEL STRUCTURES

Provide nails at 30 mm from edge

Gauge distance 3 × diameter of nail (Min.)
(3 × 3.55) = 10.65 mm
Provide gauge distance as 30 mm
Number of nails required for member AC

⎛ 11.31 × 1000 ⎞
⎜ ⎟ = 11.31
⎝ 1000 ⎠
Provide 12 nails
Spacing of nails = 40 mm
Edge distance =180 mm
Gauge distance =14.0 mm
10 mm thick splice plates are provided. A sketch of the joint is shown in
Fig. 13.59.
PROBLEMS
13.1 The safe working stress in compression for deodar wood for inside
location, parallel to the grain is 7.8 N/mm2 and that perpendicular to
the grain is 2.65 N/mm2. Determine the permissible bearing stress in
the direction of line of action of the load. The load is acting at 40° to the
grain direction. The wood is of common grade.
13.2 Determine the safe axial load on a deodar wood column, size 120 mm ×
120 mm if its unsupported length is (i) 2.80 m, (ii) 4.20 m, (iii) 6.20 m
and it has an inside location.
13.3 A column 100 mm in diameter is made of sal wood. The column is
situated in outside location and subjected to alternate wetting and
drying. The effective length of column is 1.65 m. Determine the safe
load carrying capacity of the round column.
13.4 A sal wood column is to be designed for an outside location with a
selected grader timber. If the axial load on the column is 200 kN and
the height of the column is 3 m, determine the size of the section to be
provided.
13.5 A column carries an axial load of 750 kN inclusive of self-weight. The
effective length of column is 3 m. Design a built-up dhaman wood
column.
13.6 A spaced column carries an axial compression of 36 kN. The effective
length of column is 2.50 m. Design the column using babul wood.
13.7 Design a simply supported deodar wood beam for the following data :
Grade of timber : Common
Location : Inside
Span : 4m
Load u.d.l. : 15 kN/m.
 DESIGN OF TIMBER STRUCTURES 869

13.8 A simply supported timber beam carries a total uniformly distributed

load of 42 kN inclusive its own weight. The effective span of beam is
7’20 m. The beam is made of deodar wood. Design a built-up beam. 60
mm × 240 mm planks are available.
13.9 A 120 mm × 300 mm timber beam has an effective span 4.80 m. It
carries a connected load of 20 kN at the centre. The beam has a 120
mm × 40 mm notch at the ends. Verify that whether the beam is safe in
bending and shear or not. The beam consists of deodar wood.
13.10 A simply supported beam is 560 m long. It consists of two 100 mm ×
300 mm wooden planks and one 10 mm × 300 mm steel plate in the
centre. Determine the safe load carrying capacity of the beam.
13.11 A 120 mm × 240 mm tension member carries an axial pull of 80 kN.
The tension member is made of dhaman wood. Design the lengthening
joint. Use 19 mm diameter bolt.
13.12 Two 40 mm × 240 mm wooden vertical members transmit a compression
of 20 kN to a wooden horizontal 80 mm × 240 mm. The members consists
of dhaman wood. Design the bolted joint.
13.13 Two 40 mm × 240 mm wooden inclined members are connected to a
wooden horizontal member 80 mm × 240 mm. The members consists of
dhaman wood. Determine the load transmitted by the sloping member
if four 22 mm diameter bolts are used for connections. The inclination
of the members with the horizontal is 30°.
13.14 In Problem 13.11, design the lengthening joint using standard wire
nails.
13.15 In Problem 13.12, if standard wire nails are used, then, design the
nailed joint.
13.16 A dhaman wood member 60 mm × 180 mm carries an axial pull of 36
kN. Design a suitable disc-dowelled joint for the member.
13.17 Design a timber joint for the ridge of a king-post truss where the rafters,
having a slope of 30° to the horizontal, transmit a load of 300 kN axially
and the vertical member 100 kN axially.
13.18 A timber king-post truss has a span of 6 metres and a rise of 2 metres.
It is supported on 450 mm walls. The maximum forces in the principal
rafter and the horizontal tie are 47.50 kN and 50.00 kN, respectively.
Design the principal rafter, the tie and their joint. Allowable stresses
in timber are :
Compressive stress along the grains = 10.0 N/mm2
Compressive stress across the grain = 4.0 N/mm2
Tensile stress along the grain = 16.0 N/mm2
Shear stress = 1.0 N/mm2
13.19 A cycle stand consists of timber frame as shown in Fig. P. 13.60. The
frames are spaced at 2.44 m centres and the roof covering consists of
A.C. sheets over wooden purlins.
870 DESIGN OF STEEL STRUCTURES

0 .91 4 m
0 .91 4 m

1 .82 8 m 1 .82 8 m

2 .28 6 m

Fig. P 13.60

(a) Design the members of the frame including vertical support if the
wind pressure is equal to 0.735 kN/m2 on the projected surface
vertically or horizontally.
Permissible stress the nail diameter in mm. Show the details of the
joint by means of neat sketch.
13.21 A deodar sal wood tension member 50 mm × 150 mm in section carries
a pull of 35 kN. Design a double cover bull joint to splice the member,
using the standard wire nails. The coefficient for lateral resistance in
double shear with clenching is 300 N/mm of nail diameter. What is the
efficiency of the joint designed.
Nail data
Gauge of Diameter Available length
nail (mm) (mm)
2 7.0 228
3 6.5 178
4 5.9 152
13.22 The members AB and AE of a timber truss as shown in Fig. P. 13.61 are
1.16 m long and carry axial compression of 10.30 kN. The members AC
and AD carry axial pull of 4.90 kN. Design the members and the joint A.
A

B E

3 0°
C D

Fig. P 13.61
 DESIGN OF TIMBER STRUCTURES 871

13.23 A compression member of a wooden roof truss has unsupported length

equal to 2 m and carries an axial compression of 30 kN. Design the
member and the nailed connection of the member to a fish plate. Shown
the design details. Assume suitable working stress.
13.24 Design a teak wood beam 5 metres effective span.
13.25 (a) A timber column 200 mm × 200 mm in section is made from selected
grade of sal wood and is used for an outside location. If the effective
length of the column is 3.00 m, determine the safe axial load for the
column.
(b) What will be the safe load if the effective length is increased to
5.00 m.
13.26 Design a sal wood beam simply supported over a clear span of 2.4 m to
carry a dead load of 0.80 kN/m including self-weight, and a live load of
0.40 kN/m. The deflection should not exceed 1/240 of span. Permissible
stresses are :
Bending stress = 14.0 N/mm2
Horizontal shear stress = 0.7 N/mm2
Bearing stress = 0.40 N/mm2
E = 1.30 × 105 N/mm2
Unit weight of sal = 8.70 kN/m3
13.27 A beam of 8 m span has to support a load of 30 kN/m. Using 200 mm ×
300 mm sal wood pieces design a depend section of not more than 600
mm depth for the beam. Also design suitable connectors for it so that
the pieces act as one unit. Assume for sal wood.
Permissible shear stress along the grain = 1.5 N/mm2
Permissible compressive stress along the grain = 10.0 N/mm2
Permissible bending stress along the grain = 15.0 N/mm2
Permissible tension in steel bolts = 0.6 fy
where, fy, is the yield stress for the structural steel to be used in
N/mm2.
14. Design of Masonry Structures
15. Design of Masonry Chimneys
 Chapter
Design of Masonry

14 Structures

14.1 INTRODUCTION
The masonry walls, masonry retaining walls, masonry dams, and masonry
chimneys are some of the examples of masonry structures. The masonry retaining
walls are the masonry structures which are used for maintaining the ground
surfaces at different elevations on either side of it. The material supported by
the retaining wall is known as the backfill. The backfill of a retaining wall may
have its surface horizontal or inclined. The position of backfill lying above a
horizontal plane at the level of top of a retaining wall is termed as surcharge.
The inclination of the backfill with the horizontal plane is known as the angle of
surcharge.
The masonry dams are the masonry structures which are used to retain water
on one side. The masonry chimneys are the masonry structures which are used
to escape fuel gases in the atmosphere at high levels. The masonry chimneys
have been discussed in Chapter 15. The various forces acting on these masonry
structures are the vertical forces and lateral forces. The vertical forces are mainly
due to their self-weights. The wind forces or earthquake forces and the earth or
the water pressure act as lateral forces on structures. The wind forces have
been discussed in Chapter 1 and Chapter 9. The magnitudes and directions of
wind forces are adopted from IS : 875 –1984. The earthquake forces have been
discussed in Chapter 1. The earthquake forces are adopted from IS : 1893–
1962. The earth pressure and water pressure acting on the different masonry
structures have been discussed in the subsequent articles.
The masonry structures are generally designed so that there is only
compressive strength between the blocks of which the structure is composed.
Although, the mortar has some tensile strength, it is usual to assume that the
mortar bear no tensile stress (i.e., the tensile strength of the mortar is neglected)
and the adhesion between the masonry and the mortar is negligible. Therefore,
876 DESIGN OF STEEL STRUCTURES

the shear of tangential force on the masonry at any level must not be greater
than the natural friction between masonry and masonry. The stability of the
masonry structures depends on their self-weight only. The dimensions of the
masonry structures are designed, so that the masonry structures remain stable
when subjected to lateral forces.

14.2 GENERAL CONDITION OF STABILITY OF MASONRY

STRUCTURES
Following general conditions should be satisfied in the masonry structures for
their stability :
1. There must be no tension across the cross-section.
2. The maximum compressive stress must be within the safe stress for the
material.
3. The shear force must not be greater than the natural friction between
the masonry.
4. The restoring moment must be greater than overturning moment.

W
D E C B

dc dt

Fig. 14.1

Condition 1. Figure 14.1 shows the cross-section of a masonry structure

The point C is the centroid of the cross-section. Let the lateral pressure acting
be P ; and gravity force be W, and line of pressure cuts the cross-section at the
load point E. The resultant force acting on the cross-section is R. Let the area of
cross-section of the masonry structure be A.
The direct compressive stress at the section under consideration
⎛W ⎞
fc = ⎜ ⎟ ...(i)
⎝A⎠
The cross section is also subjected to bending moment M, due to eccentricity
of the gravity force W
M = W . EC = W . x ...(ii)
The compressive stress due to bending moment
 DESIGN OF MASONRY STRUCTURES 877

M ⎛ W ⋅ x ⋅ dc ⎞
fbc = ⋅ yc = ⎜ ...(iii)
I ⎝ Ak2 ⎟⎠
The tensile stress due to bending moment

M ⎛ W ⋅ x ⋅ dc ⎞
fbt = ⋅ yc = ⎜ ...(iv)
I ⎝ Ak2 ⎟⎠
where, k = Radius of gyration of the cross-section
The combined compressive stress

⎛ W W ⋅ x ⋅ dc ⎞
f1 = fc + fbc = ⎜ + ⎟
⎝A Ak2 ⎠

W ⎛ x ⋅ dc ⎞
or f1 = ⎜1 + 2 ⎟ ...(14.1)
A ⎝ k ⎠
The combined tensile stress

⎛ W ⋅ x ⋅ dc W ⎞ W ⎛ x ⋅ dc ⎞
f2 = ⎜ – ⎟= ⎜ – 1⎟
⎝ A A ⎠ A ⎝ k2 ⎠
...(14.2)
According to first condition, there must be no tensile stress. Therefore,

⎛ x ⋅ dt ⎞
⎜ 2 – 1 ⎟ must be negative
⎝ k ⎠

x ⋅ dt ⎛ k2 ⎞
or < 1, or ⎜ x > ⎟
k2 ⎝ dt ⎠
The maximum possible value of x is given by

⎛ k2 ⎞
x = ⎜ ⎟ ...(14.3)
⎝ dt ⎠
Now consider the following cross-sections.
(a) A solid rectangular cross-section. It is the most general case
in the masonry structure. Let the width of structure DB = B, then

B2 B
k2 = and dt =
12´ 2

⎛ B2 2 ⎞ B
∴ x = ⎜ × ⎟= ...(14.4)
⎝ 12´ B ⎠ 6
The point E may lie anywhere between middle third of cross-section.
(b) A solid circular cross-section. Let the diameter of the circular cross-
section be D, then
878 DESIGN OF STEEL STRUCTURES

D2 D
k2 = and dc = dt =
12 2

2 D
x = ⎛⎜ D × 2 ⎞⎟ = ...(14.5)
⎝ 16 D ⎠ 8
The point E may lie anywhere between middle quarter of the cross-section.
The value of x may be found similarly for hollow rectangular, hollow circular
and any other cross-section.
Condition 2. When the line of pressure is in the limiting position, and the
section is symmetrical, so that, dc = dt, then the combined bending stress is
given by

W ⎛ x ⋅ dc ⎞ x ⋅ dc
f1 = ⎜1 + 2 ⎟ [3 = 1]
A ⎝ k ⎠ k2

W 2W
∴ f1 = (1 +1) = ...(14.6)
A A
The compressive stress f1 should be less than the maximum allowable
compressive stress for the masonry.
Condition 3. Let µ be the coefficient of friction for the material. Then, the
frictional force
F = µW, F > P or µW > P
⎛P⎞
or µ > ⎜ ⎟ or µ > tan θ ...(14.7)
⎝W ⎠
But if tan φ = µ then, the condition may be expressed as follows :
The angle of friction for the masonry on masonry must be greater than θ.
Condition 4. The lateral force P has a tendency to overturn the masonry
structure about the edge. The moment of the forces tending to over-turn the
structure is called as overturning moment. The moment of the forces tending
to restore the position or restraining the overturning is known as stability
moment. The stability factor or factor of safety against overturning is defined
as the ratio of the stability moment to the overturning moment. When the line
of pressure is just at the middle third, then, this factor is 3.

14.3 EARTH PRESSURE

The retained mass of backfill or earth exerts lateral pressure on the retaining
walls. The lateral pressure exerted by the earth used as backfill material is
known as lateral earth pressure. The problem of determining the lateral earth
pressure against the retaining wall is one of the oldest in the civil engineering
field. A lot of theoretical and experimental work has been conducted in this
field. Many theories and hypothesis have been developed to determine the
magnitude of lateral earth pressure. Rankine’s theory of earth pressure is most
 DESIGN OF MASONRY STRUCTURES 879

commonly used to find the lateral earth pressure. The earth pressure does not
act equally in all the directions.
The lateral earth pressure is determined as per Rankine’s theory of earth
pressure in its original form. Rankine’s theory of earth pressure is applicable to
uniform cohesionless soil only. The following assumptions were made in
Rankine’s theory:
1. The soil mass is cohesionless, homogeneous and semi-infinite.
2. The ground surface in the backfill is a plane. It may be horizontal or
inclined.
3. The back of retaining wall is vertical and smooth. There are no other
shear stresses between the wall and the soil. The stress relationship for any
element adjacent to the wall is the same as for any other element far away from
the wall.
4. The wall yields at the base.
The retaining walls are made in masonry or cement concrete. The back of
retaining wall does not remain smooth. As a result of this, the frictional forces
develop between the soil and the wall surface. According to Rankine’s
assumptions of non-existence of frictional forces at the wall surface, the resultant
pressure must be parallel to the surface of backfill. The existence of friction
makes the resultant pressure inclined to the normal to the wall at the angle
that approaches the friction angle between the soil and the wall.
The various cases of cohesionless backfill have been considered for the
determination of lateral earth pressure by Rankine’s theory in the subsequent
articles.
If a bank of earth be left to itself, it will crumble down under the action of
weather, until it has taken up a certain slope. The angle of inclination at which
such crumbling cases is known as angle of repose. It depends on the nature of
earth and its wetness.

14.4 LATERAL EARTH PRESSURE ON RETAINING WALLS

WITH DRY AND MOIST BACKFILL WITH NO
SURCHARGE
Consider an element of earth in the form of a cube at a depth z, below the
ground surface as shown in Fig. 14.2 (a). Let p be the normal pressure on the
A

p z
Ka. w z
H p' p'

p
H p
3
B Ka. w H
(a ) (b )
Fig. 14.2
880 DESIGN OF STEEL STRUCTURES

two horizontal faces. The horizontal intensity of pressure developed on the

vertical face of cube is p´ is less than p. Rankine gave the relationship between
p and p´as below :
⎛ 1 – sin φ ⎞
p´ = p . ⎜ 1 + sin φ ⎟ ...(14.8)
⎝ ⎠
where, φ = Angle of repose of the backfill material
Equation 14.8 may also be written as below :
p´ = Kp . p ...(14.9)
where, Ka = Coefficient of active earth pressure.
The total earth pressure on the retaining wall may be determined by
considering 1 metre length of the retaining wall retaining earth at its vertical
face. The normal intensity of pressure varies with its depth from the surface
p = w.z ...(14.10)
where, w = Unit weight of the material
Therefore, Eq. 14.8 may be written as
⎛ 1 – sin φ ⎞
p´ = w . z . ⎜ 1 + sin φ ⎟ ...(i)
⎝ ⎠
or p´ = K a . w . z ...(ii)
The expression (ii) shows that the horizontal intensity of pressure varies
linearly with the depth z of the element. At the top, at z = 0,
p´ = 0, and
At z = H, p´ = Ka . w . H ...(iii)
The distribution of horizontal intensity of earth pressure is shown in Fig.14.2
(b). The distribution diagram is a triangle ABC. The total earth, pressure P on 1
metre length of the retaining wall is given by the area of triangle ABC.
Therefore,
1
P = × H × (Ka ⋅ w ⋅ H )
2
1
or P = Ka ⋅ w ⋅ H 2 ...(14.11)
2

1 ⎛ 1 – sin φ ⎞
or P = wH 2 ⋅ ⎜ ...(14.12)
2 ⎝ 1 + sin φ ⎟⎠

H
The centroid of lateral earth pressure diagram is at from the base. The
3
H
total lateral earth pressure acts at from the base.
3
 DESIGN OF MASONRY STRUCTURES 881

14.5 LATERAL EARTH PRESSURE ON RETAINING

WALLS WITH SUBMERGED BACKFILL
When the backfill of retaining wall is saturated with water then, it is known as
submerged backfill. In this case, the lateral pressure is made of two components:
1. Lateral pressure due to submerged weight of the soil, w.
2. Lateral pressure due to water.
The lateral pressure at the base of retaining wall with submerged backfill as
shown in Fig. 14.3 (a) is given by
p´ = (Ea . w´ . H + ww . H) ...(14.13)
where, w´ = Unit weight of submerged soil
ww = Unit weight of water
A A

S u bm e rge d so il
Su bm e rge d so il

W a te r

H H

B CD B C
W w .H k a . w 'H
(a ) (b )
Fig. 14.3

The lateral pressure distribution diagram is shown in Fig. 14.3 (a) by triangles
ABC and ACD. When the water stands on both the sides of wall as shown in Fig.
14.3 (b), then, water pressure is not considered. Then,
p´ = Kaw´. H ...(14.14)

14.6 LATERAL EARTH PRESSURE ON RETAINING

WALLS WITH UNIFORM SURCHARGE
When the backfill is horizontal and it carries a surcharge of uniform intensity q
per unit area as shown in Fig. 14.4 (a), then, the vertical intensity of pressure at
any depth z increases by q. The increase in the lateral pressure is equal to Ka .q.
The lateral pressure at any depth z is given by
p´ = (Ka .w . z + KA.q) ...(14.15)
The lateral pressure at the base of retaining wall is given by
p´ = (K a .w.H + Ka. q) ...(14.16)
The lateral pressure distribution diagram is (ABC + BCDE) as shown in Fig.
14.4 (a). Figure 14.4 (b) shows the alternative method of plotting the lateral
pressure diagram. The lateral pressure increment K a, q due to surcharge is the
same at every point of the backfill of the wall and does not vary with the depth
z. The surcharge intensity q may be expressed by equivalent height of backfill z,
such that
882 DESIGN OF STEEL STRUCTURES

Ze = q
q w
A E A
A E
E
k a .w .Z e
= ka . q

B C D D B C B C D

k a ( H + Ze ) w
(a ) (b ) (c)
Fig. 14.4

(ka.w.ze) = (Ka.q)
∴ ze = (q/w) ...(14.17)
Therefore, the effect of surcharge of intensity q is the same as that of a fill of
height ze, above the ground surface as shown in Fig. 14.4 (c).

14.7 LATERAL EARTH PRESSURE ON RETAINING

WALLS WITH BACKFILL WITH SLOPING SURFACE
The backfill of retaining wall is inclined by angle with the horizontal as shown
in Fig. 14.5. The angle β is called the angle of surcharge. In determining
the active earth pressure for this case by Rankine’s theory, an additional
assumption is made. It is assumed that the vertical lateral stresses are conjugate.
It may be shown that, if the stress a given point is parallel to another plane,
then, the stress on the later plane at the same point must be parallel to the first
plane. Such planes are termed as conjugate planes and the stresses are termed
as the conjugate stresses.

A β

B β

Fig. 14.5

The lateral earth pressure at any depth z in this case is given by

 DESIGN OF MASONRY STRUCTURES 883

p´ = Ka´ . w . z ...(14.18)

⎛ 1 ⎞
⎜ cosβ– ⎡⎣cos2β – cosh 2 φ ⎤⎦ 2 ⎟
where, Ka´ = cos β. ⎜ 1 ⎟
⎜ ⎡ 2 2 ⎤2 ⎟
⎝ cos β – ⎣ cos β – cosh φ ⎦ ⎠
The lateral earth pressure at height H, i.e., at the base is given by
p´ = Ka´w . H ...(14.19)

⎛ 1 – sin φ ⎞
when β = 0, Ka´ = Ka= ⎜ ⎟
⎝ 1 + sin φ ⎠
The total earth pressure is given by
1
P´ = Ka´ . wH2 ...(14.20)
2

14.8 LATERAL EARTH PRESSURE ON RETAINING WALL

WITH INCLINED BACK AND SURCHARGE
Figure 14.6 shows a retaining wall with an inclined back supporting a backfill
with horizontal ground surface. The total lateral pressure calculated on a vertical
plane BC passing through the heel, (point B). The total pressure is the resultant
of the horizontal pressure P and the weight of the wedge ABC.

A C

H P1

Fig. 14.6

∴ P = (P2 · W 2) ... (14.21)

⎛1 2⎞
where, P = ⎜ K a ⋅ wH ⎟
⎝2 ⎠
Example 14.1 A retaining wall is 6 m high and the level of earth is horizontal.
The wall is 2 m wide at the base and 1 m wide at the top. The unit weight of earth
is 16 kN/m2 and that of the masonry is 24 kN/m3 and the back of wall is vertical.
If the angle of repose for the earth is 30 °, determine the stability of wall.
884 DESIGN OF STEEL STRUCTURES

Solution
Step 1: Lateral earth pressure
The lateral earth pressure is given by
wH 2 ⎛ 1 – sin φ ⎞
P = [ 3 φ = 30°]
2 ⎜⎝ 1 + sin φ ⎟⎠

⎛ 16 × 9 × 6 ⎞ ⎛ 1 – 0 ⋅ 5 ⎞
or P = ⎜ ⎟×⎜ ⎟ = 96.00 kN
⎝ 2 ⎠ ⎝1 + 0 ⋅5 ⎠
Step 2: Gravity forces
The trapezoidal section of retaining wall as shown in Fig. 14.7 is divided into
a rectangle ABCE and a triangle AED. The area of masonry section of retaining
wall
1
= 1× 6 +× 1 × 6 = 9 m2
2
The weight of masonry of 1 metre length of the retaining wall
W = 1 × 9 × 24 = 216 kN
Let the vertical line of section of W pass through point F. Let this point be at
a distan x from C, horizontally. Then by taking the moment of areas about C.
⎛ 1⎞
6 × 0.5 + 3 × ⎜1 + ⎟ = 9 × x
⎝ 3⎠
7
x = = 0.777 m
9

A 1m B

6m
P G

R x 2m
W
D
H E F C
1m 1m
2m

Fig. 14.7

Step 3: Resultant-force
The resultant force R is found as follows :
R = (P2+W2)1/2
or R = [(96.00)2 + (216.00)2]1/2 = 236.20 kN
 DESIGN OF MASONRY STRUCTURES 885

The resultant force R passes through the point H. The force P acts at a height
⎛H⎞
⎜⎝ ⎟⎠ = 2 m above the base.
3
The distance FH is found as follows :
FH ⎛P ⎞ ⎛ 96 ⋅ 00 ⎞
= ⎜ ⋅ FH ⎟ = ⎜ × 2 ⎟ = 0.888 m
FG ⎝W ⎠ ⎝ 216 ⋅ 00 ⎠
∴ HC = (EC + FH) = 0.777 + 0.888 = 1.665 m
The distance of point H from the centre of section
⎛ 1 ⎞
= ⎜1 ⋅ 655 – × 2 ⎟ = 0.665 m
⎝ 2 ⎠
The width of base of cross-section B is 2 m
⎛B⎞
⎜ ⎟ = 0.6667 m
⎝2⎠
The resultant R passes within the middle third of the cross-section. Hence,
the retaining wall is stable.

14.9 HYDROSTATIC PRESSURE OF WATER

The force transmitted between water and a solid surface is considered as a
normal pressure. The force exerted on any given area is known as water pressure.
When the water is at rest, then the water pressure is known as hydrostatic
pressure of water. The intensity of pressure exerted within a fluid or on the
walls of a structure is measured by the force per unit area. The intensity of
pressure is given by
⎛P⎞
p = ⎜ ⎟ ...(14.22)
⎝A⎠
where, P = Force exerted by water
A = Area
The stationary water transmits only normal pressure. Therefore the normal
pressure is equal in all the directions at a given depth in the water. The water
remains under pressure due to self-weight. The water pressure increases with
the increase of depth. Let w be the density of water. The relation between
pressure and depth in a liquid may be found as follows :
Consider a vertical cylindrical column of water at rest. The surface of water
is horizontal. The cylinder is in the equilibrium. The cross-sectional area of
cylinder is A. The base of cylinder is h unit below the free water surface. The
volume of this cylindrical column
V = (A × h) ...(i)
The weight of this cylindrical column
W = (w.V) = (w.A.h) ...(ii)
The upward force exerted by water at the base
R = p.A ...(iii)
The forces exerted on the sides of cylindrical column are self-balanced.
Therefore
886 DESIGN OF STEEL STRUCTURES

R = W
or p.A = w.A.h
or p = w.h ...(14.23)
Equation 14.23 show that the intensity of pressure at a given depth h below
the free surface is given by the weight of a cylindrical column of height h on a
unit area. It is to further note that the intensity of pressure in a liquid at rest is
equal in all the directions.
The total pressure on any horizontal surface such as the bottom of a tank of
area A (above atmospheric pressure) is given by
P = w.h ...(iv)
When any plane surface is inclined and it is not horizontal, then, the intensity
of pressure varies continuous across the surface according to varying depth.
Consider a vertical submerged surface of total area A as shown in Fig. 14.8
This surface is divided into a large member of parallel strips. The areas of the
strips are a1, a2, a3 etc. The depths of these strips below the free water surface
are h1, h2, h3 etc respectively. Total pressure on the vertical surface is given by
P = w.(h1a1 + h2a2 + h3a3 +.....+ etc.)
...(14.24)
The terms h1a1, h2a2, h3a3 constitute the moment of area A about an axis
which is the line of intersection of the vertical plane of the submerged surface
with the surface of water. Therefore,
h1a1 + h2a2 + h3a3 +.....+ etc. = h . A1

P = w . A .h ...(14.25)
where, h = Depth of the centroid of area
A

h1
h3 h2 b
h h

H
dh
CG

WH
(a ) (b )

Fig.14.8

The total pressure in a vertically immersed surface may also be found by

considering elementary strip of thickness dh and width b at depth h below the
free surface of water. The pressure on this strip
δP = (wh × b × dh)
 DESIGN OF MASONRY STRUCTURES 887

The pressure on the surface

∫ dP = ∫ (wh × b × dh )
∴ P = w ∫ h ⋅ bdh or P = w . Ah ...(v)
Figure 14.8 (b) shows a vertical surfaces submerged in the water. The top of
surface is at the free surface of water. The intensity of pressure at the top
p1 = 0 ...(vi)
The intensity of pressure at the bottom
p2 = w.H ...(vii)
The pressure distribution diagram is a triangle ABC as shown in Fig. 14.8 (b)
Total pressure on the plane is equal to the area of this triangular diagram

1 ⎛ wH 2 ⎞
P = wH ⋅ H = ⎜ ⎟ ...(14.26)
2 ⎝ 2 ⎠
From Eq. 14.25
2
⎛H ⎞ wH
p = w . Ah = w (H × 1) × ⎜ ⎟= ...(viii)
⎝ 2 ⎠ 2
This resultant pressure acts through the centre of gravity of triangle ABC in
a direction perpendicular to the surface AB. The centre of pressure is situated
at two-thirds of the depth AB below the free surface or at one-third, of the depth
AB above the bottom.
The total pressure on any inclined surface submerged in the water is also
given by Eq 14.25, i.e.,
P = w . Ah ...(ix)
But this water pressure also acts perpendicular to the inclined surface. The
centre of pressure may be found by determining its depth from free water surface,
which is given by

⎛ Second moment of area ⎞

y = ⎜ ⎟
⎝ First moment of area ⎠

⎛ I 00 ⋅ sin2 θ ⎞ ⎛ Ak00
2
⋅ sin2 θ ⎞ ⎛ K 00 ⋅ sin2 θ ⎞
∴ y =⎜ ⎟=⎜ ⎟=⎜ ⎟ ...(x)
⎝ Ah ⎠ ⎝ Ah ⎠ ⎝ h ⎠

⎛ K 2 + h2 ⎞ ⎛ K 2⎞
C ⋅G 2 2
or y = ⎜ ⎟ ⋅ sin θ = ⎜ h + C ⋅G ⎟ ⋅ sin θ ...(14.27)
⎝ h ⎠ ⎝ h ⎠
where, k 00 = Radius of gyration about free water surface
kCG = Radius of gyration about the axis passing through CG
888 DESIGN OF STEEL STRUCTURES

14.10 MASONRY DAMS

The masonry dams are further classified as gravity dams and arched dams. The
gravity dams resist the water pressure by action of the dead or self-weight only.
The arched dams are constructed on the sites where a reliable rock abutment is
provided by the sides of valley across which the dam is built. The arched dams
are curved in plan and resist the pressure behind them by acting as the horizontal
arches and are therefore of much lighter section than the gravity dams. The
arched dams are not often constructed and are subjected to complicated stress
distributions. The stability of gravity dams has only been discussed here. A
dam is virtually a retaining wall for the water. But the magnitude of water
pressure may be accurately determined as compared to that of earth pressure
on the retaining wall. Therefore, the design of dam is capable of more rigid
treatment than that of a retaining wall. Further, the dams are work of much
greater magnitude than the retaining walls and it is more necessary to provide
the minimum quantity of material consistent with the safety. The dams
possessing the minimum section consistent with the safety are said to be designed
of economical section.

PH G
wH 2
=
2
H/3 R
W
W
C
B D E
x
x
B

Fig. 14.9

The stability of any proposed dam is to be examined under two conditions :

1. When the reservoir is empty, and material of dam is subjected to the
action of self-weight only.
2. When the reservoir is filled with water and it is subjected to combine
effect of water pressure and weight of masonry.
The line of action of resultant under both the conditions should pass within
the middle-third at any horizontal section through the dam, including at the
base.
A right angled-triangle ABC as shown in Fig. 14.9 is an ideal theoretical
section. It gives the basis to deduce a practical section. Let the water level be
 DESIGN OF MASONRY STRUCTURES 889

upto the top as shown in Fig. 14.9. Consider 1 metre length of dam. The total
water pressure acting on the face AB of dam
wH 2
PH = ...(i)
2
H
The line of action of water pressure is at height from the bottom of dam.
3
The weight of masonry of dam
1
W = ⋅ p⋅B⋅H ...(ii)
2
where, p = Unit weight of masonry
B = Width of the dam at the bottom
1
The weight of masonry acts at rd width of the base from B. When the
3
reservoir is empty, then, the water pressure is zero. The eccentricity at the base
of dam
⎛B B⎞ B
e = ⎜ – ⎟= ...(iii)
⎝2 3⎠ 6
The weight of dam acts at the middle third point from the face (i.e., on the
heel side). The pressure at the toe, from Eq. 11.2

W ⎛ x ⋅ dt ⎞ ⎛ k2 ⎞
f = – 1⎟ ⎜⎝3 x =
⎜
A ⎝ k2 ⎠ dt ⎟⎠
∴ f = zero ... (iv)
The pressure at the heel is given by Eq. 14.1

W ⎛ x ⋅ dc ⎞ ⎛ k2 ⎞
f = ⎜1 + 2 ⎟ ⎜3 x = d ⎟
A ⎝ k ⎠ ⎝ c⎠

⎛ 2W ⎞ ⎛ 2W ⎞ ⎛ 2W ⎞
∴ f = ⎜ ⎟=⎜ ⎟=⎜ ⎟ ...(v)
⎝ A ⎠ ⎝ B ×1 ⎠ ⎝ B ⎠
⎛2 1 ⎞
or f = ⎜ × × p × B × H ⎟ = ρ.H ...(vi)
⎝B 2 ⎠
When the reservoir is full, then, the resultant of water pressure PH and the
weight of the masonry of dam W passes through E. The point E may be located
by considering the triangle of forces GDE.
DE ⎛ PH ⎞ ⎛ PH ⎞
GD = ⎜⎝ W ⎟⎠ , x = DE = ⎜⎝ W × GD ⎟⎠

⎛ PH H ⎞
x = ⎜ × ⎟
⎝W 3⎠
890 DESIGN OF STEEL STRUCTURES

⎛ WH 2 H ⎞ ⎛ 1 ⎞
or x = ⎜ × ⎟×
⎝ 2 3 ⎠ ⎜ 1 ⎟
⎜⎝ × P × B × H ⎟⎠
2

⎛1 w H2 ⎞
or x = ⎜ ⋅ ⋅ ⎟ ...(14.28)
⎝3 p B ⎠
Alternatively. The distance x may be found by taking the moments of both
the forces about E. Therefore,

H ⎛1 w H2 ⎞
PH × = W. x, or x = ⎜ ⋅ ⋅ ⎟ ...(viii)
3 ⎝3 p B ⎠
B
The point E may be at a distance on the toe side at the most, so that the
3
resultant is at the middle third point.

B ⎛1 w H2 ⎞
∴ = ⎜ ⋅ ⋅ ⎟
3 ⎝3 p B ⎠
1/2
⎡w ⎤
∴ B = ⎢ ⎥ ...(14.29)
⎣ρ⎦
Equation 14.29 gives the minimum width necessary for an idea section.
Taking w = 10 kN/m2 and ρ = 20 kN/m3
∴ B = 0.71 H ...(14.30)
The resultant pressure at the bottom, R is proportional to side GE of the
triangle of forces GDE. The vertical and horizontal components of this resultant
force R are W and PH, respectively. These components also act at E. The vertical
component gives normal pressure at the bottom. The horizontal component PH
tends to slide the dam horizontally. The maximum pressure at the toe
⎛ 2W ⎞
fmax = ⎜ ⎟= ρ . H ...(ix)
⎝ B ⎠
The co-efficient of friction, from Eq. 14.7
PH 1 ⎛ 1 ⎞
µ> , or µ > wH 2 ×
W 2 ⎜ 1 ⎟
⎜⎝ ρ ⋅ B ⋅ H ⎟⎠
2

⎛w H⎞
or µ>⎜ ⋅ ⎟ ...(14.31)
⎝ p B⎠

⎛ 10 0.71B ⎞
or µ>⎜ × ⎟ , µ > 0.355 ...(x)
⎝ 20 B ⎠
 DESIGN OF MASONRY STRUCTURES 891

Alternatively. The position of point E may be found by equating the moments

of PH and W and to the moment of resultant, R about the heel i.e., about toe
point B in Fig. 14.9. In this method, the distance of point E, EB = x is found
directly Therefore,
H B ⎛ B⎞
PH + +W × = W × ⎜ x + ⎟ + PH × O
3 3 ⎝ 3⎠

⎛ wH 2 H ⎞ ⎛ ρ ⋅ B ⋅ H B ⎞
or ⎜⎝ × ⎟ +⎜ × ⎟ = W ⋅x ...(14.32)
2 3⎠ ⎝ 2 3⎠
In a practical section, the following modifications of the theoretical triangular
section become necessary. A free board is always necessary above the supply
level. The upper edge is given a reasonable practical width to the triangular
section at the top. Many dams carry a roadway along the top. The base width is
increased by providing a sloping outer surface or a series of batters or a
continuous curve.

14.10.1 Width of Base of Trapezoidal Dam

Let ABCD be a trapezoidal dam section with a vertical face as shown in Fig
14.10 (a). Let width of top of dam AB be a, and width of bose of dam CD be b, and
the height of dam be h. The width of the base of the dam section is determined

A a B

H
PH G

H/3 R
W W
D J E PH F C
x
b
(a ) (b )

Fig. 14.10

such that the line of action of resultant R strikes the base within the middle
third of the section. The point G represents the centroid of the trapezoidal dam
section. The vertical through G intersects the base at the point E. Let the distance
DE be a from D. The trapezoidal dam section is divided into a rectangle ABJD
and a triangle BJC. Taking the moments of these areas about D.
892 DESIGN OF STEEL STRUCTURES

(a + b ) a 1 ⎡ 1 ⎤
x ⋅h = k ⋅a ⋅ + h (b – a ) ⎢a + (b – a )⎥
2 2 2 ⎣ 3 ⎦

(a + b ) h ⎡ 2 (b – a )(b + 2a ) ⎤
or x = a +
2 2 ⎢⎣ 3 ⎥⎦

(a + b ) h( 2
or x = b + ab + a2 )
2 6

⎡ a2 + ab + b2 ⎤
x = ⎢ ⎥ ...(xi)
⎣ 3 (a + b ) ⎦
Total water pressure per metre length

wH 2
PH = ...(xii)
2
Total weight of masonry per metre length

H
W = ρ(a + b) ...(xiii)
2
From the triangle of forces GEF

EF PH wH 2 ⎛ 2 ⎞
= = ×⎜
GE W 2 ⎝ ρ (a + b) H ⎟⎠

EF ⎛ w⋅H ⎞ H
or = ⎜ (3 GE = )
GE ⎝ ρ (a + b) ⎟⎠ 3

⎛ w ⋅ H2 ⎞
or EF = x´ = ⎜ ...(xiv)
⎝ 3ρ (a + b) ⎟⎠
The line of pressure GF is to be within the middle third of the base, so that
there is no tensile stress in the base. Therefore,

2b
x + x´ =
3

a 2 + ab + b2 wH 2 2b
or + =
3 (a + b ) 3 p (a + b ) 3

wH 2
a 2 + ab + b2 + = 2b (a + b)
p
 DESIGN OF MASONRY STRUCTURES 893

wH 2
or a 2 + ab + b2 + = 2 ab + 2 b2
ρ

⎛ wH 2 ⎞
or b2 + ab – ⎜ a 2 + ⎟ = 0
⎝ 2 ⎠

1
⎡ ⎛ wH 2 ⎞ ⎤ 2
– a + ⎢a2 + 4 ⎜ a2 + ⎥
⎣ ⎝ ρ ⎟⎠ ⎦
∴ b =
2

1
a ⎡ 5a2 wH 2 ⎤ 2
∴ b = – +⎢ + ⎥ ...(14.33)
2 ⎣ 4 ρ ⎦

It is to note that Eq. 14.33 gives the minimum width of the base required for
the trapezoidal dam section. The maximum compressive stress is given by

⎛ 2W ⎞ ⎛ ρH (b – a ) ⎞
fmax = ⎜ ⎟= ...(14.34)
⎝ b × 1 ⎠ ⎜⎝ p ⎟⎠

In actual practice, a steep batter is also given on the upstream side of the
dam section as shown in Fig 14.10 (b).
Example 14.2 A masonry dam of trapezoidal section is 75 m high and 12 m
thick at the top. The water face of dam is vertical. The free water level is upto the
top of the dam. The unit weight of masonry is 24 kN/m3. Determine the minimum
width of the base of the dam, so that there is no tensile stress anywhere at the
base. Also determine maximum compressive stress at the base.
Check the stability of dam against sliding. The coefficient of friction between
the base of dam and earth is 0.7.
Solution
Step 1. Minimum width of base
The minimum width of base is given by Eq. 14.33 as
1
a ⎡ 5a2 wH 2 ⎤ 2
b = – +⎢ + ⎥
2 ⎣ 4 ρ ⎦

1
12 ⎡ 5 2 1000 ⎤2
or b = – + ⎢ × (12 ) + × 75 × 75⎥ m
2 ⎣4 2400 ⎦
894 DESIGN OF STEEL STRUCTURES

or b = 44.23 m
Step 2. Max. Compressive stress
From Eq 14.34, the maximum compressive stress at the base

ρ ⋅ H (b + a ) ⎛ 24 × 75 (44 ⋅ 23 + 12) ⎞
fmax = =⎜ ⎟⎠
b ⎝ 44 ⋅ 23
fmax = 2288.36 kN/m2 or fmax = 2.288 N/mm2
Step 3. Check for stability of dam against sliding

PH ⎛ wH 2 2 ⎞ ⎛ w H ⎞ ⎛ 10 × 75 ⎞
=⎜ × = =⎜ ⎟ = 0.556 < 0.7.
W ⎝ 2 ρ ⋅ H (b + a ) ⎟⎠ ⎜⎝ ρ (b + a ) ⎟⎠ ⎝ 24 × 56 ⋅ 23 ⎠
Hence, safe
PROBLEMS
14.1. The height of a masonry retaining wall is 9 m. The width at top of
retaining wall is 1 m and that at bottom is 7 m. The wall is used to
retain earth. The face of retaining wall is vertical. The unit weight of
earth is 16 kNm2 and that of masonry is 24 kN/m3. The angle of friction
of earth particles is 30°. The coefficient of friction between the masonry
and the masonry is 0.5. The bearing capacity of earth is 120 kN/m2.
Determine the total lateral pressure, the total vertical load and the
eccentricity of resultant force. Check the stability of the retaining wall
against (a) tension, (b) compression, (c) sliding, and (d) overturning.
Also find the maximum and minimum stresses at the base of the
retaining wall.
14.2. A masonry dam of trapezoidal section is 50 m high and 8 m thick at
the top. The water face of the dam is vertical. The free water level is at
the top of dam. The unit weight of masonry is 24 kN/m3. Determine the
minimum width of the base of the dam, so that there is no tension
anywhere at the base. Also determine the maximum compressive stress
at the base. Check the stability of dam against sliding. The coefficient
of friction between the base of dam and earth is 0.65.
14.3. A masonry retaining wall of trapezoidal section has its top width equal
to 0.75 m and height 5 m. Its face which is in contact with the retained
earth is vertical. The earth retained is level at top. The soil weighs 16
kN/m3 and its angle of internal friction is 30°. The masonry weighs 24
kN/m3. Determine the minimum width of the base to avoid tensile
stresses and determine the maximum compressive stress for this base
width.
If the coefficient of friction between base and soil is 0.60 check the
stability of retaining wall.
14.4. Check the stability of the masonry retaining wall with concrete base
loaded as shown and determine maximum pressure under the base.
 DESIGN OF MASONRY STRUCTURES 895

Take unit weight of masonry as 20 kN/m3 of concrete as 24 kN/m3 and

that of soil as 16 kN/m. The angle of internal friction of soils is 30° and
coefficient of friction between base and soil is 0.5.

1 .6 m
0 .3 m 1 0 kN 6 kN /m 2

2
2m
M ason ry
0 .6 m

0 .5 m

2 .4 m

Fig. P 14.4

14.5. A masonry dam has a triangular cross-section with 15 m base width as

shown in Fig. P 14.5. The reservior is filled with water. The water level
is 1 m below the top. Determine the maximum height of the dam such
that the dam remains structurally safe.
Take unit weight of masonry = 20 kN/m3
Permissible stress in compression in masonry = 1.5 N/mm2
No tensile stresses are permitted.

W a te r le vel 1m

15 m

Fig. P 14.5
 Chapter
Design of Masonry

15 Chimneys

15.1 INTRODUCTION
The chimneys are used to escape the gases to such a height that the gases do
not contaminate surrounding atmosphere. The cross-sectional area of chimney
is kept large enough to allow the passage of burnt gases. The cross-sectional
area of a chimney for the passage of burnt gases depends on the type of fuel and
quantity of fuel to be used in a plant, available draft for carrying the burnt
gases set up in the chimney and losses due to friction within the chimney. The
height of chimney is kept to provide required draft. The draft is defined as the
pressure available for producing a flow of burnt gases. The draft depends on the
height of chimney above sea level, the type of fuel to be burnt, the type of furnace
and the temperature of burnt gases. When the gases in the chimney are heated,
then the hot gases expand. The hot gases occupy larger volume than before.
The weight of gases per cubic metre becomes less. As a result of this, the pressure
at the bottom of chimney due to weight of hot gases also becomes less than the
pressure due to weight for cold air outside the chimney. The difference between
the two pressure result in the flow of the burnt gasses up the chimney. For the
purpose of the structural design, the height and diameter of a chimney at the
top are given.
The masonry chimneys are generally built in brick masonry. The brick
masonry chimneys are built in common bricks or of special radial bricks, which
are usually perforated and moulded to different sizes suitable for use in various
diameters. The masonry chimneys may have square, hexagonal, octagonal or
circular shapes in plan. The masonry chimneys of circular shape in plan are
usually preferred over the other shapes.

15.2 LINING FOR MASONRY CHIMNEYS

The material used for lining should be capable of withstanding high temperature
 DESIGN OF MASONRY STRUCTURES 897

upto 2000° F. The lining is required from below the breech opening to the height
where the heat of gases does not damage the chimney. The fire bricks are used
for lining in the brick masonry chimneys. The fire brick lining must be free to
expand and contract independently of the main chimney. The height of lining
depends on the particular purpose of chimney. The height of chimney may be as
low as 6 m to 10 m but it is commonly 16 m to 26 m high in lofty chimneys. The
1 1
height of lining is kept th to th as a common rule.
4 5
An efficient air space should be provided between the fire brick line and the
outer shaft, suitably covered or protected from corebelled course of brickwork
projecting from the inside of the outer shaft and leaving sufficient clearance for
maximum expansion of the lining.

15.3 VARIOUS FORCES ACTING ON MASONRY CHIMNEYS

The various forces acting on the masonry chimney are as follows :
1. Self weight of the masonry chimney.
2. Weight of lining.
3. Wind pressure.
4. Seismic force.

15.3.1 Self-weight of the Masonry Chimney

The self-weight of masonry chimney W s acts vertically. The masonry chimneys
of circular cross-section in plan are built tapering from top towards bottom. The
portion of a circular masonry chimneys between two horizontal sections may be
considered to be frustum of a hollow cone. The volume of a hollow frustum of a
cone is given by
π⋅H
VP = [(R12 + r12 + 2R1r1) – (R22 + r22 + 2R2r2) ]] ...(15.1)
3
where, R1 = Outer radius of the cone at the bottom,
r1 = Outer radius of the cone at the top
R2 = Inner radius of the cone at the bottom
r2 = Inner radius of the cone at the top
H = Height of the frustum of hollow cone.
The weight of masonry of the chimney
Ws = ρ . V
π⋅H
or Ws = ρ [(R21 + r21 + 2R1r1) – (R22 + r22 + 2R2r2)]
3
where, ρ = unit weight of the material used in masonry
H
or Ws = ρ [π (R21 –r12) + π (r21 – r22) + (2 πR1r1 – 2 πR2r1)
3
898 DESIGN OF MASONRY CHIMNEYS

H
or Ws = ρ
3
( A1 + A2 + A1 A2 ) ...(15.2)

The compressive stress due to weight of masonry of the chimney at any

horizontal cross-section
Ws
f1 =
A

⎛ W ⎞
or f1 = ⎜⎜ π ( R2 – R2 ) ⎟⎟ ...(15.3)
⎝ 1 2 ⎠

where, R1 = Outer radius of the chimney

R2 = Inner radius of the chimney.

15.3.2 Weight of Lining

The weight of lining in the masonry chimney, W L also acts vertically. The weight
of lining is given by
WL = (ρ2 . π . d. tL . HL) ...(15.4)
where, ρ2 = Unit weight of brick lining = 20 kN/m3
d = Mean diameter of the lining
tL = Thickness of lining
HL = Height of the lining.
The lining is supported directly at the base.

15.3.3 Wind Pressure

The wind pressure acts horizontally. The wind pressure acting on a structure
depends on the shape of the structure, the height of structure, the location of
structure, and the climatic conditions. The wind pressure per unit area increase
with the height of structure above the ground level. In order to simplify the
design, the chimney is divided into number of segments. The intensity of wind
pressure over each segment may be assumed as uniform. The intensity of wind
pressure corresponding to mid-height of each segment may be noted from IS :
875 –1984. The wind pressure acting on the chimney may be found as below :
P = k . p. (Projected area of chimney)
...(15.5)
where, k = Shape factor
p = Intensity of wind pressure.
The shape factors for various shapes in plan of the chimney shaft are given in
Table 15.1 as per IS : 875 –1984.
Therefore, the wind pressure acting on the chimney circular in plan
P = 0.7 × p × (Projected area) ...(15.6)
 DESIGN OF MASONRY STRUCTURES 899

Table 15.1 Shape (in plan) factors

Factor, k
Plan shape Ratios of Ht to Ratio of Ht to Ratio of Ht to
base with base with base with
(0 – 4) (4 – 8) 8 or over
Circular 0.7 0.7 0.7
Octagonal 0.8 0.9 1.0
Square (wind perp. to
diagonal) 0.8 0.9 1.0
Square (wind perp. to
face) 1.8 1.15 1.3
The projected area of a masonry chimney circular in plan and tapering from
top to the bottom is a trapezium.
The wind pressure has an overturning effect on the chimney. The overturning
moment Mw due to wind pressure causes compressive stress on the leeward
side of the chimney and the tensile stress on the windward side of the chimney.
The maximum compressive stress and tensile stress on the extreme fibre of
chimney are equal and these are given by

⎛ Mw ⎞ ⎛M ⎞
fw = ⎜ ⋅ R1 ⎟ = ⎜ ⎟ ...(15.7)
⎝ I ⎠ ⎝Z ⎠
where, R1 = Outer radius of the chimney at the section under
consideration
I = Moment of inertia of the section
π π
I=
64
( )
R14 − R24 =
64
( D14 − D24 ) ...(15.8)

It is to note that the circular shape in plan for the masonry chimney realizes
the greatest stability with the minimum quantity of brickwork. It is the most
effective section of flue. It is equally resistant to the wind pressure applied in
any direction.
In addition to the overturning effect due to wind pressure, the wind has also
aerodynamic effect. This effect of wind has not been taken into consideration
for the design of chimney.

15.3.4 Seismic Forces

The seismic forces also act horizontally. The seismic forces act on a structure,
when the structures are located in the seismic areas.
The worst effect out of effect due to seismic (earthquake) forces and wind
effect is only taken into consideration. Only one effect is considered for the
design of structure out of these two effects.
900 DESIGN OF MASONRY CHIMNEYS

15.4 STABILITY OF MASONRY CHIMNEYS

A masonry chimney of circular shape in plan tapering from top towards bottom
is shown by a diagrammatic sketch in Fig. 15.1.
The stability of tall masonry chimney depends on the weight of the masonry
work of the chimney shaft W s, and the wind pressure PH acting on the outer
face. In the calm atmosphere, the centre of pressure at any horizontal section

C h im ne y
sha ft

PH G

L ining

R
W

B C PH E D Fo un da tion

Fig. 15.1 Fig. 15.2

coincides with centre of gravity of the section and the uniform intensity of
compression exists over the entire section. The application of lateral wind
pressure PH causes the displacement of the centre of pressure at every horizontal
section. The displacement should be restricted within such units as to prevent
the development of tensile stress in the mortar joints.
Let W s, be the weight of the chimney and PH be the lateral wind pressure
action on the chimney. In Fig. 15.1 the line of action of force PH, is produced to
meet the line of action of W,s. In the triangle of forces, GC represents the weight
W s, and CE represents the lateral wind pressure. Then, GE represents the line
of action of the resultant of these two forces, R. When the point E falls outside
the core (middle-quarter in case of circular shape in plan) then, there will be
tension in the mortar at the point R on the windward side. When the point F
falls inside the core, the chimney is quite stable.
Table 15.2

Diameter Area Volume Weight Direct

∑W Compression
W/A2
⎡H ⎤
Section Outside Inside A1 A2 A1 A2 ( )
⎢⎣ 3 A1 + A2 A1 A2 ⎥⎦
1 2 3 4 5 6 7 8 9
m m m2 m2 m2 m2 kN N/mm2
XX 290 2.50 1.68 1.83 1.76 10.54 210.80
X1 X1 3.15 2.75
2.55 0.1153
2.68 2.92 2.80 16.80 336.00
X2 X2 3.40 2.80
2.60 0.1873
3.77 4.08 3.92 29.54 470.80
X3 X3 3.65 2.85
2.65 0.2495
4.94 5.34 5.14 30.84 616.80
X4 X4 3.90 2.90 0.3063
DESIGN OF MASONRY STRUCTURES

2.70
6.22 6.69 6.45 38.72 774.40
X5 X5 4.15 2.95
2.75 0.3602
7.59 8.12 7.85 47.12 942.40
X6 X6 4.40 3.00
2.80 0.4128
9.05 9.68 9.80 57.06 1141.20
X7 X7 6.65 3.05
2.85 0.4641
10.60 11.30 10.94 65.68 1313.60
X8 X8 4.90 3.10 0.5159
901
Table 15.3
902
Moments of Modulus of Projected Wind Height Bending
P M P M
Inertia Section Area Force from top Stress + − Remarks
A Z A Z
π 4 M P ⋅h
Section I= ( D1 + D24 ) Z(I/R 1 ) A PH = 0.7 h fb = =
4 Z Z
× p × ∑A
1 2 3 4 5 6 7 8 9 10
m4 m3 m2 kN m N/mm 2 N/mm N/mm 2
XX — — — — — — — — —
18.15 19.06 6
X1X1 2.16 1.336 0.0413 0.1566 0.074 No tension
19.65 39.69 12
X2X2 3.59 2.112 0.1127 0.3000 0.0746 –do–
21.15 61.90 18
X3X3 5.56 3.047 0.1827 0.4322 0.0668 –do–
22.65 85.68 24
DESIGN OF MASONRY CHIMNEYS

X4X4 8.06 4.113 0.2492 0.5555 0.571 –do–

24.15 120.97 30
X5X5 10.82 5.216 0.3192 0.6794 0.4410 –do–
25.65 137.97 36
X6X6 14.32 6.509 0.3815 0.7933 0.0313 –do–
27.15 166.48 42
X7X7 18.90 8.128 0.4405 0.8946 0.0313 –do–
28.15 166.48 42
X8X8 23.50 9.583 0.4669 0.9828 0.0490 –do–

Notes : 1. The maximum compressive stress at the base is 0.9828 N/mm2. It is less than allowable value of 1 N/mm2.
2. There is no tension on the windward face. Hence, the design is safe.
 DESIGN OF MASONRY STRUCTURES 903

The combined compressive stress at any horizontal section is given by

⎛ Ws M w ⎞
f = ⎜ + ⋅ R1 ⎟ ...(15.9)
⎝ A I ⎠
The combined compressive stress is maximum at the extreme fibre at the
leeward side. The maximum combined compressive stress should be less than
maximum allowable compressive stress in the masonry.

15.5 ARCHITECTURAL TREATMENT OF MASONRY

CHIMNEYS
The chimneys are the structures which are emitting clouds of black smoke.
Therefore, the chimneys are not handsome structures from their nature.
However, the appearance of the chimneys may be improved by giving
consideration and attention to the details. It is generally accepted that the shaft
should be tapering. Panelling and projecting the arches are the devices which
are frequently used in the brick chimneys. The enlarged top of a chimney may
take one of various forms, but it should bear some suitable proportions to the
size of the entire structure. The excessive ornamentation in a chimney is
objectionable.
The strongly projecting coping caps string courses are undesirable, since,
they offer increased resistance to the wind, the former being at the maximum
leverage above the base.

15.6 SPECIFICATIONS FOR DESIGN OF MASONRY CHIMNEYS

The heights of a square chimney shall not exceed ten times the width the base
section measured immediately above the footings. The height of circular
chimneys must not exceed twelve-times the diameter clear of the footings. The
height of octagonal chimneys should not exceed eleven-times the diameter.
The shaft is to batter (as shown in Fig. 15.2) 1 in 48 and not less than 1 in 50.
The projections should not extend beyond the face of the shaft to a greater
distance than the thickness of wall at that level.
The minimum thickness of the uppermost section is to be one brick thick and
the thickness is to increase by half the brick length at least every 6 m from the
top downward.
The footings are to project beyond the basal section in all sides to a minimum
distance equal to the thickness of the wall at the base.
The masonry chimneys are designed by following the above specifications
and the stability of the chimney dimensions is then checked.
Example 15.1 Design a brick masonry chimney of 48 metres height with 2.5
m diameter at the top. The horizontal wind pressure acting on the chimney is
1.50 kN per square metre. The unit weight of bricks is 20 kN/m3. The safe
compressive stress for brick masonry is 1.0 N/mm2.
904 DESIGN OF MASONRY CHIMNEYS

Solution
Design. The diameter of brick masonry chimney at top is 2.5 m. Let the
thickness of masonry at top be 200 mm. The chimney shaft is given a batter of
1 in 48.

2 00 m m 2 00 m m

6m
X1 X1
6m
X2 X2
6m
X3 X3
6m
X4 X4 48 m
6m
X5 X5

6m
X6 X6
6m
X7 X7
6m
X8 X8

Fig. 15.3

The projections at every 6 m height from top are given. The projections of 100
mm are given. The thickness of lining is kept 200 mm. The height of lining is
1 ⎛1 ⎞
kept th the height of chimney (i.e., ⎜ ⎟ × 48 = 12 m]. The dimensions and
4 ⎝4⎠
data for the calculations are given in Tables 15.2 and 15.3. The dimensions of
brick masonry chimney are shown in Fig. 15.3.
The combined stress due to vertical loading and moment due to wind forces
⎛P M ⎞ ⎛P M ⎞
have been found on leeward side ⎜ + ⎟ and on windward side ⎜ – ⎟ in
⎝A Z ⎠ ⎝A Z ⎠
Table 15.3, columns 8 and 9, respectively. The maximum compressive stress at
the base does not exceed the allowable value and there is no tension at any
section. Hence, the design is safe. The lining is supported on foundation directly.
 DESIGN OF MASONRY STRUCTURES 905

The load is transferred to the foundation by pedestral footing. The cylindrical

concrete foundation is provided under the chimney. The foundation is designed
such that the compressive stress does not exceed the bearing capacity of the soil
and there is no tension.
PROBLEM
15.1 Design a brick masonry chimney with the following data :
Height of chimney = 42 m
Diameter at top = 2.5 m
Unit weight of bricks = 20 kN/m3
Wind pressure = 1.50 kN/m2
The safe compressive stress for brick masonry is 1.0 N/mm2.
Index

web crippling, 370

A Beam connections, 418
Allowable Stresses riveted, 312
beam, 318 framed, 441
bending, 316, 318 moment resistant, 441
shear, 332 stiffened seated, 428, 441
tension members, 296 unstiffened seated, 476
Anchor bolt, 606 welded, 573
Angle of attack, 36, 38 framed, 426, 864
Appropriate radius of gyration, 133 moment resistant, 752
Approximate radii of gyration, 150 sealed, 426, 755
Asbestos cement sheets, 571 Beam line method, 463, 464
corrugated, 671 Bearing stiffeners, 479
trafford, 571 allowable stress, 317
Axially loaded columns, 125 critical stress, 353
unsymmetrical, 607
B Bolted connections, 98
advantages, 98
Bases for columns, 244 disadvantages, 99
gusseted, 241 Bolts, 99
slab, 243 high strength, 107
Batten plates, 189 permissible stresses, 99, 111
Beams, 311 turned, 103, 110
built-up, 289 unfinished, 109
crane gantry girder, 393 Bracing, 133, 606, 640
encased, 409 roof trusses, 606
girder, 349 Bracket connections, 80
laterally supported, 361 riveted, 63
laterally unsupported, 361, 367 welded, 737, 745
main, 395 Buckling, 271
subsidiary, 313 diagonal, 523
web buckling, 271
908
908 Index
INDEX

local, 157 effective sectional area, 132

Butt weld, 180, 677 radius of gyration, 132
slenderness ratio, 133
C strength, 157
Cased 234 Connections 151
beams, 407 flexible, 418
columns, 225 frame, 419
Chain riveting, 68 rigid, 418, 543
Chemical composition, 6, 7, 10 seated, 424
high tensile steel, 7, 8 semi-rigid, 460
mild steel, 7 stiffened, 420
Chimneys, 600, 873 unstiffened, 419
masonry, 873 beam, 416
architectural treatment, 903 behaviour, 46, 334
forces acting, 897 bolted, 94
lining, 896 Crane gantry girder, 393
specifications, 856 Crane overhead, 391
stability, 876 Curtailment of flange plates, 507
Column, 129
D
axially loaded, 125
bases, 242 Deflections 4, 126
eccentrically loaded, 146, 200 Design of Fr. connection, 424
end restraint conditions, 128 stiffened, 427
footing, 234, 242 unstiffened, 426
independent, 242, 257 Large moment resistant, 444
combined, 263 small moment resistant, 441
laced, 175, 181 semi-rigid, 460
solid wood, 803 pin, 93, 95
spaced (wooden), 805 welded, 670
splice, 225 Deflections 4, 126
trusses, 602 beams, 332
Combined footing, 261 allowable (steel), 333
bending and compression, 125 allowable (wooden), 804
bending and tension, 197 Diagonal buckling, 523
welds, 696 Diamond riveting, 68
Compression members, 125 Duchemin’s formula, 587
allowable stress, 143
axially loaded, 123 E
built-up, 151 Earth pressure, 863
common sections, 150 Eccentrically loaded columns, 198
design, 162 Eccentric riveted connections, 66 , 79
design formula, 143, 147 Effects 22, 23
eccentrically loaded, 198 dynamic, 38, 41
effective length, 122, 128 erection, 39, 40
 INDEX 909

temperature, 37, 39, 41 load, 480

Effective length factor, 480
butt weld, 681 Independent footing, 254
compression flange, 361 Intermediate stiffeners, 479
compression member, 126 Intermittent butt weld, 681
fillet weld, 681, 683 Internal air pressure, 587
Efficiency of riveted joints, 63
External wind pressure, 600 J
Jack arch, 390
F
Joints, 60
Factor of safety, 43 riveted, 49
general, 44 timber, 785
timber, 785, 788 bolted, 825
Fetch length, 37, 38 disc-dowelled, 853
Force coeff, 36 framed, 419
Filler joists, 405 welded, 670
Fillet weld, 680
Framed conections, 419 K
riveted, 419
welded, 742 Kyanizing, 790

G L
Galvanized corrugated iron, 655 Lacing (latticing), 174
sheets, 570, 571 design, 188
Galvanizing, 564 570, 571 double, 174
Gantry girder, 391 single, 176
Girder (see beams, crane gantry Lateral earth pressure, 878
girder, plate girder), 311 , 313 retaining walls, 875
Grains of timber, 782 Lintels, 387
Grillage footing, 254, 242, 257 Loads, 21
combined, 263 breaking, 43
independent, 254 dead, 45
Gussested base, 243 impact, 484
live, 27
H seismic, 46
snow, 40
Hankison’s formula, 790 ultimate, 42
Heart wood, 788 wind, 37
High strength bolts, 108 Local buckling, 156
High tensile steel, 07 Longitudinal stiffeners, 529
Horizontal stiffeners, 530 Lug angles, 305
Hydrostatic pressure, 885
M
I
Masonry structures, 860
Impact, 484 chimneys, 896
910
910 Index
INDEX

dams, 875 depth, 493

general conditions of stability, 876 design of flanges, 497
retaining walls, 878 design of web plate, 498
Mechanical properties, 108 diagonal buckling of web, 523
high tensile steel, 8 economical depths, 481
mild steel, 6 elements, 478
Metal (modern) connectors in, 858 flanges, 496
Mild steel, 4 self-weight, 484
fusion welding quality, 6 splice, 557
ordinary quality, 8 flange, 551
standard quality, 7 web, 537
Modified, 20 web stiffeners, 528
slope-deflection-method, 464 bearing, 525
moment distribution method, 472 horizontal, 527
Moment resistant connection, 444 vertical, 529
riveted, 443 Plate girder (welded), 730
welded, 760 Permeability, 599
normal, 421
N small, 568
Net area, 289 Principal axes of section, 607
tension flange, 512 Purlins, 567
tension members, 288
R
P Rafter, 567
Perforated cover plates, 199 common, 567
Pins, 115 principal, 568
Pin connections, 98 Return period, 39
advantages, 114 Riveted connections, 52
disadvantages, 114 beams, 418
Plate girder (riveted), 477 eccentric, 57
allowable stress in bending, 231, 827 Riveted joint, 56
allowable stress in shear, 485 assumptions, 67
area of flanges, 479 butt, 59
gross, 446 design procedure, 86
net, 480 edge distance, 61
assumption in design, 485 efficiency, 69
buckling of web, 491 failure, 61
connections, 444 lap joint, 65
flange angles to web, 519 percentage strength, 67
flange plates to flange, 522 pitch, 55
angles, 478 specifications, 69
curtailment of plates, 509 strength, 66
 INDEX 911

transmission of load, 60 fillet weld, 685

Rivets, 44 lug angle, 303
arrangements, 68 plug and slot weld.
holes, 53 riveted joint, 53
types, 52 Splice, 225
working stress, 53 columns, 222
Rolled steel section, 21 flange, 546
Roof trusses, 562 web, 529
bearing plate, 641 Stanchion, 125
bracing, 605 Steel, 3, 6
covering, 600, 638 high tensile, 7
dead load, 572 mild steel, 4, 6
economical spacing, 569 structural steel, 4
live load, 574 Stiffeners 477
pitch, 569 angle, 430
purlins, 567 bearing, 521, 522
rise, 566 intermediate, 527
shoe angles, 639 Stress, 317
slope, 565 bearing, 42
snow load, 574 bending, 315
spacing, 569 breaking, 42
timber, 863 compressive, 42
wind load, 574 shear, 330
Round tubular structures, 765 tensile, 42
tube columns and compression ultimate, 40
members, 801 working, 44
tubular sections, 151, 633 yielding, 42
Stress-strain relationship, 40
S Strut, 22
Sap wood, 794 angle, 148
Seat connections, 476 continuous, 152
riveted, 423 discontinuous, 152
stiffened, 430
unstiffened, 476
T
Secant formula, 143, 145 Tension member, 282
Shear lag, 308 allowable stress, 289
Shielding effect, 39 axial tension, 282
Slab base, 242 axial tension and bending, 299
Slot weld, 687 common sections, 151, 477
Solidity ratio, 39 net sectional area, 288
Specifications, 69 Terrain category, 38
butt weld, 678 Tie, 283
912
912 Index
INDEX

Timbers, 787 V
decay, 790
defects, 788 Vertical stiffeners, 530
grading, 792 Velocity profile, 38, 39
grains, 788
W
metal connections, 846
preservation, 790 Web
roof trusses, 852 buckling, 370
seasoning, 788 crippling, 370
strength, 784 splicing, 530
working stress, 788 stiffeners, 520, 522
Timber joints, 816 bearing, 520
bolted, 818 horizontal, 535
disc-dowelled, 853 vertical, 523
fish-plate, 847 Weight
framed, 816 bracing, 567
lap, 829 plate girder, 480
nailed, 841 purlin, 567
scarf, 829 roof covering, 567
Topography, 38 trusses, 567
Timber structures, 785 unit weight of materials,
Torsional constant, 349 Weld, 674
Transverse stiffeners, 524 butt, 681
Trusses, fillet, 684
roof, 556 imperfections, 689
tubular roof trusses, 775 plug, 686
types, 563 slot, 688
Tubular beams, 779 stresses, 696
Tubular columns and compression working, 792
members, 774 Welding, 677
Tubular roof trusses, 767 Welded connections, 746
Tubular sections, 758 framed, 748
Tubular structures, 757 moment resistant, 752
joints in trusses, 864 seated, 747
permissible stress, 771 Welded joints, 671
Tube-tension members, 775 subjected to axial load, 710
subjected to eccentric load, 715
U Wind load, 35
Ultimate load, 42 angle of attack, 38
Unbuttoning, 310 breadth, 38
Unsymmetrical bending, 607 depth, 36
deflection, 626 development, 36
maximum bending stress, 632 eff. frontal area, 36
 INDEX 913

of surface area, 36 topography, 39

force coefficient, 38 Wooden beams, 821
fetch length, 38 built-up, 805
ground roughness, 38 flitched, 821
gust, 38 notched, 807
gust (peak), 38 Working stress, 44
gradient height, 39
mean ground level, 39 Y
pressure coeff., 37
return coeff., 37 Yield point, 42
shielding effect, 39 lower, 42
factor, 39 upper, 42
suction, 39
Z
solidity ratio, 39
terrain category, 39 Z-line, 621
velocity ratio, 38 Z-polygon, 625
profile, 39