Concrete Structure by Zahid Ahmed Siddique (Part-1)

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 395

ABCMITTHEAIIW

hikJ Hi-dd Mi1


.Udiu md Srfslhiin, kin j, J.,

IIH. ml 1 nKi;iv rur ] h'piiiiianmih d1 ।


i^
fc'OlylH'L'1 Il1||i, ,l lull '1 ui.hiHi II jdWllI.
i
th
||>|W1^
CONRET COIICIIEIE STRBCTURES

STRUC ES
Hi.iilhii.il' nd im-ii i, dUixim vd

die bulb^i-n. uded "Wl<v'U Nifti-rlim^


"I. XJ*h Wluri Pvi.igiu A ii.W ,11.1111,11 "Ildidii-Mi-mj;
lhhj.!m.'i.;r^ ijm He Iiuj-i iiriwlniaiiii,,
wsi-Jia'di di id |,q iiiilii, d da.iiiiij.il a iiiii:|'Wi.^^ iail“
id!ii:ii;i:d JX wmih lib: lun
-t'^wirim,^
11. «diiiji|i|i J(ri1 11 -i।r< I? j.J 1' '।i>i i-i 11idi,i || ku ,i in !,
IIrdciip mid <। i,n ihr Sli.:idi.-j^ an.
jA'Itviily rfiwnln,|j,| UH Jui!d dj,<M|i|ii ed:
Uiuecvillll hlll^lllili^L

I1: uujJ * i.'du |i+i

/All III) A HIM Ali) SliHIDIIQII

® Help Cwi Engineering Publisher, Lahore, Pakistan


Published in 2012 by

Help Civil Engineering Publisher


97-C1, NESPAK Colony, Lahore

CONCRETE STRUCTURES
Zahid Ahmad Siddiqi
E-mail: zasiddiq@uet.edu.pk Part - I, 2nd Edition

Based on ACI 201 1

Concrete Structures
Part - I, 2nd Edition
ISBN 978-969-8633-07-3

This book was composed and produced by Zahid Ahmad ZAHID A. SIDDIQI
Siddiqi, Professor, Civil Engineering Department, Lahore Professor
and Published by Help Civil Engineering Publisher, Lahore. Civil Engineering Department
University of Engineering and Technology
Lahore
E-mail: zasiddiq@uet.edu.pk

Copyright © Zahid Ahmad Siddiqi 2012

AU rights are reserved. No part of this publication may be reproduced,Stored in a retrieval


system, or transmitted in any form or by means of electronic, mec ical, photocopying,
recording or otherwise, without the prior written permission of the authqf.

Price in Pakistan: Rs.


Help Civil Engineering Publisher
97 C-I, NESPAK Colony, Lahore
In the Name of Allah, the Most Beneficent, the Most Merciful

Recite in the Name of the Allah, Who had created (Al-Quran)

© Copyright by Zahid Ahmad Siddiqi


All Rights Reserved
vii

CONTENTS

Preface xv
1. MATERIALS AND DESIGN PROCESS 1
1.1 Introduction, 1
1.1.1 Concrete, 1
1.1.2 Size of Bricks, 2
1.1.3 Mechanism of Load Transfer, 2
1.2 Merits of Concrete Construction, 3
1.2.1 Control over Cross-Sectional Shape, 3
1.2.2 Easy And Universal Availability of Constituents, 3
1.2.3 Economy, 3
1.2.4 Better Insulation Properties, 3
1.2.5 Good Bond with Steel, 3
1.2.6 Ductility and Warning Before Failure, 4
1.2.7 Lesser Chances of Buckling, 4
1.2.8 Aesthetics, 3
1.3 Demerits of Concrete Construction, 4
1.3.1 Weaker in Tension, 4
1.3.2 Greater Self Weight, 4
1.3.3 Cracking, 4
1.3.4 Relatively Unpredictable Behavior, 4
1.3.5 Limited Industrial Behavior, 5
1.3.6 Inelastic Behavior, 5
1.3.7 Creep and Shrinkage, 5
1.4 Specifications, 5
1.5 Types of Loads, 5
1.5.1 Self Load, 6
1.5.2 Imposed Load, 6
1.5.3 Service or Characteristic Loads, 6
1.5.4 Factored Loads, 6 "

1.5.5 Dead Load, 6 ;


1.5.6 Live Load, 6
1.6 Limit States, 9
1.7 Various Design Methods, 10
1.7.1 Basic Design Equation, 10
1.7.2 Factor of Safbty (F.O.S), 10
1.7.3 Strength Design, 11
1.7.4 Allowable Stress Design (ASD), 11
1.7.5 Plastic Design, 12
CONCRETE STRUCI URES PART - 1 AUTHOR: ZAHID A. SIDDIQI

17 6 12
3.3 Flexural Strength, 63
1.8 design Steel is Yielding, 64
Depth of Rectangular Stress Block when TensionOne-Way
1.9 3.4 Slabs, 65
1.10
Analysis of Structures, 13
3.5 Minimum Depth of Non-Prestressed Beams and
yeS’gn of Structures, 13 65
3.6 Maximum Permissible Computed Deflections,
1.10.1 Objectives of Structural Designer, 13
1.10.2 Economy in Design, 1 5 3.7 Balanced Steel Ratio, 65 Behavior, 68
3.8 Types of Sections Depending on Flexural
1.10.3 Procedure of The Structural Design, 15 Tension-Controlled Sections, 68
1.10.4 General Design Flow Chart, 1 6 3.8.1
1.11 3.8.2 Compression-Controlled Sections, 69
Commonly Used Load Factors and Load Combinations, 16
1.12 Shrinkage of Concrete, 19 3.8.3 Transition Sections, 69
1.13 Creep of Concrete, 20 3.9 Strength Reduction Factors (^), 69
1.14 Fatigue in Concrete, 20 3.9.1 Tension-Controlled Sections, 70
1.15 3.9.2 Compression-Controlled Sections, 70
Concrete Compressive Strength, 21
1.16 Reinforcing Steel, 23 3.9.3 Transition Sections, 70
1.17 Modulus of Elasticity, 25 3.9.4 Shear and Torsion, 70
1.18 Requirements for Contract Documents, 25 3.9.5 Bearing on Concrete, 70
1.19 Duties of Inspector, 26 3.9.6 Strut-And-Tie Models, 70
1.20 3.9.7 Development Length, 71
I 21
1.22
Legal Standing of Design Codes, 26
Two-Cycle Moment Distribution, 27
Portal Frame Method for Lateral Loads, 28
I 3.9.8
3.10 Maximum
Bearing
Steel
on Concrete, 71
Ratio, 71
Problems, 33 3.11 ACI Minimum Reinforcement, 73
3.12 Types of Failures, 74
2. FLEXURAL BEHAVIOR OF BEAMS UNDER
SERVICE LOADS 35 3.12.1 Under-Reinforced or Tension Failure; 74
2.1 Introduction, 35 3.12.2 Over-Reinforced or Compression Failure, 78
2.2 Assumptions Regarding Flexural Behavior, 3.12.3 Balanced Failure, 79
35
2.3 Flexural Behavior in General, 36 3.13 Analysis of Singly Reinforced Sections, 80
2.4 Terms Used to Study Flexural Behavior, 38 3.14 Distribution of Flexural Reinforcement, 85
2.4.1 " ‘
Modular Ratio (w), 39 3.15 Skin Reinforcement in Deep Beams, 85

2.4.2 Uncracked Transformed Section, 40 3.16 Requirements for Structural Integrity, 86


2.4.3 Modulus of Rupture and Split Cylinder Strength,
41 3.17 Selection of Steel Bars For Beams, 87
2.4.4 Tensile Strength of Concrete, 41 3.18 Concrete Cover to Steel Reinforcement, 87
2.4.5 Cracked Transformed Section, when 3.19 Number of Bars in One Layer,, 88
Concrete Stress is Lesser Thanje '/2, 43 3.20 Loads Supported by Beams, 88
2.4.6 General Allowable Stress Design Procedure, 46 3.20.1 Beams Supporting One-Way Slabs, 89
2.4.7 ACI Minimum Reinforcement in Beams, 3.20.2 Beams Supporting Two-Way Slabs, 89
2.4.8 47 Walls Supported by Beams, 91 . .
Maximum Allowable Concrete and Steel Stresses, 3.20.3
2.4.9 47
Design Values of Factor K, 47 3.21 Walls Supported by Lintels, 92
2.5 Reduced Moment Of Inertia Due to Cracking, 3.22 Calculation of Beam Loads, 92
57
2.6 -
Long Term Deflections, 57 3.22.1 Slab Loads, 92
Problems, 59 3.22.2 Calculation of Self Weight of Beams, 93
3 DESIGN OF BEAMS 3.23 Bar Bending Schedule, 93
61 3.24 Design of Singly Reinforced Section, 96
3.1 Introduction, 61 3.24.1 Using Trial Beam Dimensions, 96
3.2 Stress-Strain Curve for Concrete, 61 Steel Ratio is Known, 98
3.24.2
Problems, 117
xi
X CONCRETE STRUCTURES PART -I
AUTHOR: ZAHID A. SIDDIQI
4. SHALLOW, TEE AND
CONTINUOUS BEAMS 119 5.14 T russ Analogy for Shear, 184
4,1 Doubly Reinforced Beams, Problems, 187
119 189
4.2 Behavior of Doubly Reinforced Sections, 120 6. ONE-WAY AND EDGE SUPPORTED SLABS
4.2.1 Case I: Both Steels are Yielding, 120
6.1 One-Way Slabs, 189
4.2.2 Case H: One or Two Steels are Not Yielding, 123 Procedure of Slab Design, 189
4.3 6.1.1
Balanced Steel Ratio for Doubly Reinforced Sections, 123 6.1.2 Depth of One-Way Slabs, 190
4.4 Maximum Steel Ratio for Doubly Reinforced Sections, 125
4.5 -6^ Minimum Clear Cover, 190
Minimum Tension Steel Ratio for Compression Steel Yielding, 126 Provision of Slab Steel, 191
4.6 Another Check for Yielding of Compression Steel, 127 6.1.4
4.7 Analysis of Doubly Reinforced Sections, 128 6.1.5 Maximum Spacing of Bars, 191 Steel, 192
4.8 6.1.6 Distribution, T emperature and Shrinkage
Design of Doubly Reinforced Sections, 133
4.9 Tee and Ell Beams, 138 6.1.7 Minimum Flexural Steel, 192
4.9.1 Effective Flange Width (b) [ACI 8.12], 139 Continuous One-Way Slabs, 192
4.9.2 Requirements for Flarige Reinforcement, 6.1.9 Check for Shear, 193 Slabs, 193
4.10 Joist Construction, 140 140 61.10 Design Flow Chart for One-Way
4.11 Flexural Behavior of T-Beams, 141 6-1-11 Approximate Amount of Steel Slabs, 194
in
4.12 6.1.12 Design of Stair Slab, 204
Tension-Controlled Failure of T-Beams, 143 Two-Way Edge Supported Slabs, 208
4.13 Analysis
of T-Beams, 144 6.2
4.14 Design of T-Beams, 146 6.2.1 Basic Behavior of Two-Way Slabs, 215
4.15 6.2.2 Minimum Slab Thickness for Two-Way Slabs, 216
Continuous Beams, 152 Design Procedure for Two-Way Slabs, 216
4.16 AC1 Moment Coefficients for Continuous Beams, 6.2.3
154 Problems, 227
4.16.1 Conditions for ACI approximate analysis to be applicable, 229
154 " '

7. BASICS OF SHORT COLUMN DESIGN


4.16.2 Moment Coefficients, 154 Introduction, 229
4.17 Approximate Curtailment of Bars, 155 7.1
7.2 Types of Column Reinforcement, 230
Problems, 165
7.3 Sources of Moment in Columns, 232
5. DESIGN FOR SHEAR Behavior of Concentrically Loaded Short Columns
at Service Loads,
7.4
5.1 Introduction, 167 233
234
5.2 Concrete Shear Strength, 168 7.5 Steel Yields Earlier than Crushing of Concrete in Columns,
7.6 Transformed Uncracked Section of Column, 234
5.2.1 Concrete Compressive Strength, 168
5.2.2 7.7 Capacity Reduction Factors for Columns, 236
Applied Moment to Shear Ratio and Effective Columns, 236
5.2.3 Longitudinal Reinforcement Ratio, 170
Depth, 169 7.8 Ultimate Strength of Concentrically Loaded Short
7.9 Minimum Concrete Compressive Strength, 236
5.3 ACI Code Provisions for Concrete Shear Strength,
5.4 ,
Shear Strength Provided by Vertical 170 7.10 Minimum and Maximum Steel Ratios in Columns, 237
5.5 Reinforcement,
Strength Reduction or Resistance Factor in 17? 7.11 Minimum Cover and Splicing of Bars, 238
5.6 Minimum Web Reinforcement, 172
Shear, 172 7.12 Amount of Transverse Ties, 238
5.7 Location of Maximum Shear for Design Of Beams, 7.13 Amount of Spiral Reinforcement, 239
5.8 Minimum Section for Design Shear Capacity, 174 173 7.14 Trial Column Size, 243
59 Maximum Spacing Requirements, 174 7.15 Amount of Steel in Columns, 243
5.10 Design of Web Reinforcement, 7.16 Eccentrically Loaded Columns, 246
5,11 Hanger Reinforcement, 175
174 7.17 Procedure for Design of Eccentrically Loaded Columns, 247
5.12 Typical Shear Force Diagrams, Problems, 273
5.13 General Procedure for Shear
176
Design, 176
xiri
xii CONCRETE STRUCTURES PART - J
AUTHOR: ZAHID A. SIDDIQI 10.3 Wind Load, 365
8 RETAINING WALLS 10.4 UBC 97 Wind Loads, 367
8.1 10.4.1 Combined Height, Exposure and Gust Factor (Ce), 368
Introduction, 255 Pressure Coefficient (Cq), 368
8.2 Important Parameters for Retained Soils, 255 10.4.2
10.4.3 Other UBC Wind Load Provisions, 368
8.3 Components of RC Cantilever Retaining Wall, Loads, 372
8.4 256 10.5 Dynamic
Loads / Forces on Retaining Wall, 259 Types of Prescribed Loading, 374
8.5 Required Retaining Wall Strengths, 263 10.5.1
10.5.1.1 Periodic loading, 374
8.6 Load Factors for Wali Design, 266
8.7
10.5.1.2 Non-periodic loading, 375
Placement of Joints, 267 Earthquakes, 376
8.8 Seismic Lateral Earth Pressure, 287 10.6
10.6.1 Structure of Earth, 376
8.9 Lateral Force on Separate Structural Elements, 291 Plate Tectonics, 377
8.10 Bending Moment Coefficients for 10.6.2
Walis Subjected to Triangular Loads, 10.6.2.1 Plate boundaries, 378
293 10.6.3 Focus and Epicenter, 380
8.10.1 Upper One-Third of Wall, 294 10.6.4 Shallow and Deep Earthquakes, 380
8.10.2 Middle-Third of Wall, 294 10.6.5 Earthquake / Seismic Waves, 380
8.10.3 Lower One-Third of Wall, 294 10.6.6 Destruction Caused by Earthquakes, 382
Problems, 307 10.6.6.1 Earthquake magnitude and Richter scale, 382
9. DESIGN OF FOOTINGS 10.6.6.2 Earthquake intensity and Mercalli scale, 382
10.6.7 Measuring Earthquakes, 383
9.1 Introduction, 311 Example Earthquakes, 384
9.2 10.6.8
Types of Footings, 311 10.6.9 Types Of Earthquakes, 384
9.3 Bearing Pressure, 315 10.6.8.1 Types depending on duration, 384
9.4 Allowable Bearing Capacity, 316 10.6.8.2 Types depending on intensity, 384
9.5 Net Contact Pressure, 316
9.6
10.6.10 Main Considerations for Seismic Design, 384
Failure Behavior of Isolated Footings, 318
9.7
J 0.6.11 Brief Description of Methods of Analysis and Design for
Design Bending Moment and One-Way Earthquake Loading, 385
Shear for Isolated Footings,
320 10.6.11.1 Free vibration analysis, 385
9.8 Transfer of Load from Column to Footing, 321 10.6.11.2 Response history analysis (RHA), 385
9,9 Distribution of Footing Steel, 322 10.6.11.3 Response spectrum analysis (RSA), 385
9.10 Development and Splicing of Steel, 322 10.6.11.4 Base isolation method, 386
9.11 Concrete Protection for Reinforcement, 323 10.6.11.5 Use of special energy dissipating devices, 386
9.12 Procedure for Design of Isolated Footings 10.6.11.6 Equivalent static or pseudo-static load method,
323 (Concentrically Loaded),
387
9.13 Eccentrically Loaded Isolated Footings, 332
10.7 Free Vibration Analysis, 387
9.14 Kem of Contact Area, 333
10.8 Spring Constant or Stiffness Coefficient for Columns , 392
9.15 Eccentric Punching Shear, 336
10.9 Equation of Motion for Viscous Damping, 393
9.16 Combined Footing Design for Two Columns,
345 10.10 Free Vibration Analysis for Multiple Degrees of Freedom Structures,
9.17 Typical Types of Combined Footings, 347
396
9.18 Foundation Beam Footing, 356
10.11 Time History Analysis, 401
9.19 Raft Footing, 360
Problems, 363
10.11.1 Single Degree of Freedom Damped System Subjected to
Dynamic Force, 401
10. LATERAL LOADS AND MOMENT
ENVELOPES 10.11.2 Dynamic Effect of Earthquakes on Rigid Structures, 403
10.1 Loads, 365 10.1 1.3 Dynamic Effect of Earthquake on Flexible Structures, 403
10.2 Rain and Snow Loads, 365 10.12 Response Spectrum Analysis (RSA), 405
Xiv CONCRETE STRUCTURES PART- 1 AUTHOR: ZAHID A. SIDDIQI XV
10.12.1 Procedure to Use Response Spectra for S DOF Systems, 405
10.12.2 Procedure to Use Respon.-e Spectra for MDOF Systems, 407 11.11 Development of Bundled Bars, 459
10.12.3 UBC Design Response Spectrum, 407 11.12 Development Length with Standard Hooks, 459
10.13 Equivalent Static Load Method, til 11.13 Bar Cut-Off Requirements, 460
10.13.) Seismic Zone Factor (27. 411 11.14 Procedure for Curtailment in Continuous Beams, 466
10.13.2 Soil Profile Types, III Problems, 485
10.133 Seismic Source Types, 412 487
12. TWO-WAY COLUMN SUPPORTED SLABS
10.13-4 Near-Source Factors, 412
10.13-5 Ground Response Coefficients, 412 12.1 Introduction, 487
10.13-6 Fundamental Time Period of a Structure, 4] 3 12-2 Code Provisions for Joist Construction, 488
10.13.6.1 Approximate Method, 413 1 2.3 Flat Slab Vs. Coefficient Method of Slab Design, 489
10.13.6.2 Method A of UH C, 413 12.4 Fraction of Load to be Carried in Two Mutually Perpendicular
10.13.6.3 Method B of UBC, 413 Directions, 490
10.13.7 Ductility, 415 12.5 Total Moment to be Distributed Amongst Mid-Span and Supports, 492
10.13.8 Response Modification Factor (R), 415 12.6 Variation of Bending Moment in a Slab Panel, 492
10.13.9 Seismic Importance Factor (7), 416 1 2.7 Design Strip, Column Strip and Middle Strip, 495
10.13.10 Seismic Response Coefficient (Cs), 416 12.8 Approximate Steel for a Typical Slab Panel, 496
10.13.11 Seismic Dead Load (IF), 417 12.9 Torsional Strength of Slab, 496
10.13.12 Magnitude of Base Shear (F), 417 12.10 Direct Design Method, 497
10.13.12.1 UBC refined formula, 417 12.11 Equivalent Frame Method, 537
10.13.12.2 UBC simplified formula, 418 12.11.1 Slab Beams, 538
12.13.13 Distribution of Base Shear at Various Story 12.11.2 Torsional Member and Torsional Stiffness (At), 540
Levels 418
10.14 General Considerations for Seismic Design, 425 ’ 12.11.3 Equivalent Column Stiffness (Aec), 543
10.15
Pattern Loads, 430
428’
ACI Definitions for Seismic Design, 1 2. 11 .4 Pattern Loading, 544
10 16 12.11.5 Correction of Moments to Face of Supports, 545
10.17 Correction of Beam Moment at Face of Supports, 433 12.12 Analysis of Slab Systems for Lateral Loads, 550
10.18 Moment Envelope, 434 12.13 Seismic Provisions for Two-Way Slabs Without Beams, 553
Problems, 445 Problems, 555
1]
1 DEVELOPMENT OF REINFORCEMENT 13. YIELD LINE AND STRIP METHODS FOR SLABS 557
. 451
111 Introduction to Bond, 451 13.1 Introduction, 557
11.2 Mechanism of Force Transfer by Bond, 451 13.2 Collapse Mechanism and Yield-Line Pattern, 557
11.3 Bond Strength, 451 13.3 Types of Analysis Methods, 558
11.4 Pullout Test, 452 13.3.1 Lower Bound Method, 558
11,5 Failure Modes, 453 13.3.2 Upper Bound Method, 559
11.5.1 Side Split Mechanism, 453 13.4 Advantages of Yield-Line Analysis, 560
11.5.2 Face and Side Split Mechanism 13.5 Demerits of Yield-Line Analysis, 560 .
or Bottom Inside
u^ue
Mechanism, 454 13.6 Conventions to Show Boundary Conditions and Yield Lines, 561
11.5.3 V-Notch Mechanism, 454 13.7 Rules to Locate Yield Lines, 561
11.5.4 Pullout Mechanism, 454 13.8 Yield Line Patterns, 563
11.6 Effect of Transverse Reinforcement, 455 13.9 Fundamental Assumptions in Yield Line Theory, 565
U.7 Development of Reinforcement in Tension, 455 13.10 Contribution of a Steel (Afi) Along and Across a Skewed Yield Line,
11.8 Development Length in Compression, 458 566
11.9 Splices of Deformed Bars in Tension, 458 13.11 Contribution of Steel Placed in Grid Resolved Along and Across a
11.10 Splices of Deformed Bars in Compression, 459 Skewed Yield Line, 568
13.12 Major Steps for Analysis, 569
xvii
XVI CONCRETE STRUCTURES PART - 1
Segment Equilibrium Method, 570
AUTHOR: ZAHID A. SIDDIQI

13.14 Virtual Work Method, 572


13.15 Special Conditions at Edges and Comers, 583
13.15.1 Edges, 583
PREFACE
13.15.2 Comers, 584 buildings and AASHTO LRFD
13.16 Affinity Rules, 591 This book is prepared according to the 2011 ACI Code for are the SI units,
13.17 Strip Method of Slab Design, 591 Specifications for bridges. The units used throughout the presentation
in the starting
13.17.1 Basic Assumptions, 592 however, the expressions and examples are also given in US Customary units
tried that the three main
13.17.2 Simply Supported Square Slab, 592 chapters to keep continuity with die traditional system of units. It is detailing are
1 3.17.3 Slabs with Continuous Edges, phases of structural design, namely load determination, design
calculations and
596 book. After the
13.17.4 Rectangular Slab with Two Adjacent Edges introduced to the beginner. This book is useful with the 2nd part of engineers
the same
and students are
Other Two Simply Supported, 597 Continuous and printing of the first edition, the comments send by colleagues, fellow
acknowledged with thanks. Suggestions for further improvement of the presentation will be
13.17.5 Slabs with Short Unsupported Edges, 601
13.17.6 Slabs with Longer Unsupported Edges, 604 highly appreciated arid will be incorporated in the future editions.
PrnhlRmc A1 1

Appendix-I: Steel Tables


613
Appendix-II: Beam and Slab Design Aids
623 Author
Appendix-Ill: Column Interaction Curves 2012
653
Appendix-IV: Simplified Wall and its Footing Design
719
Selected Formulas
729
References
751
Chapter - 1

MATERIALS AND DESIGN PROCESS


1.1 INTRODUCTION

The main ingredients of reinforced concrete are steel reinforcing bars and concrete. The
important properties of both of these methods are discussed in the following sub-sections.

1.1.1 Concrete
Concrete is a mixture of stone particles and some suitable binding material like cement
While in fresh state, concrete is plastic or fluid-like and may be molded in any shape,
but with time, it hardens and becomes an artificial stone-like material. Because small
. stone particles may be “assembled” at site to provide any desired architectural shape
: and because of the relatively lesser cost, concrete is used for most of the construction.

The main constituents of concrete are described as under:


(a) Binding Material: It is usually a paste of cement in water and is the relatively
costly constituent of the cement concrete. Lime and some other materials may also be
used as binding materials along with the cement.
(b) Filler Material: It is required to reduce the cost of the concrete and to provide
the natural strength of stone particles. Most commonly filler material is composed of a
combination of various sizes of crushed stone but other materials such as round gravel,
brick ballast, bloated clay and iron filings may also be used in certain applications.
Filler material may also be termed as “aggregate”.
The stone aggregate should be well graded meaning that all the particle sizes must be
present in suitable proportions so that the smaller particles fill in the spaces between the
bigger particles to give a dense mass. Depending on the particle size, the aggregate
may be classified into two categories, coarse and fine aggregates. The portion of
aggregate having particle sizes greater than 5 mm (3/16 in) is called coarse aggregate
and the portion having particle sizes lesser than 5 mm is called fine aggregate. Fine
aggregate is commonly called sand.
Concrete is hence a mixture of cement, fine aggregate and coarse aggregate. Water is
also added in the fresh concrete to make it workable and to allow the hardening of the
concrete by the chemical reaction. Concrete is a brittle material and is very strong in
compression but very weak in tension. Further, its properties are sufficiently reliable in
compression but are highly unpredictable in tension. Whenever, there are chances of
development of significant tension in concrete, it is reinforced with steel bars. This
combination of concrete to resist compression and steel to resist tension is called
reinforced concrete. The simple concrete without any steel reinforcement is called
Plain Cement Concrete (PCC) whereas the concrete with embedded steel bars is called
Reinforced Concrete (RC).
The ratio between the cement and various types of aggregates is specified as parts of the
cement taken as unity (reference), parts of fine aggregate compared to cement and parts .
2 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI MATERIALS AND DESIGN PROCESS 3
of coarse aggregate relative to cement. For example, a PCC or RC of ratio 1:2:4 means CHAPTER 1: in order as application of the load, member
cement content by weight is one out of 7 parts, sand is 2 out of 7 parts and coarse structured member is categorized members, production of stresses in the
is out parts. The water content to be added in the start is deformation, appearance of strains in the
specified as the fraction of the weight of water with respect tb the weight of
separately
materials, accumulation of the stresses to
provide an equal and opposite reaction to the
the cement equilibrium condition where all the external
For example, a water-cement ratio (W/C ratio) of 0.45 means
that 45 kgs applied loads and establishment of a final Each of the above phenomena takes place
(approximately 45 litres) of water is to be added in a concrete forces are balanced by the internal forces.
mix having the weight of just before it in order. Main deformation of the
the cement equal to 100 kgs. The W/C ratio is defined
with respect to saturated surface because of the occurrence of the event
is associated with elongation of the one side of
dry aggregate. In practice, extra water equal to
the water absorbed by the aggregate beam with the loads is bending, which
during mixing transportation and compaction of concrete the beam and shortening of the other
side. Hence, compression is developed on one
is also to be added. side. The layers of concrete material
side and tension is developed on the opposite because of the excessive
1.1.2 Size Of Bricks subjected to compressive loads can easily resist the stresses
layers subjected to tension are to be
The standard size of the burnt clay brick may be
available compressive strength. However, the
x 3 in.). This size includes average
taken as 228 x 114 x 75 mm (9 x 4% reinforced by the steel bars.
mortar thickness on both sides. Hollow block or
stone masonry may have different sizes. The sizes of
beams and columns are usually 1.2 MERITS OF CONCRETE
CONSTRUCTION
decided keeping in view the brick wall thickness,
which may vary in multiples of 114
mm (4l/j in.). Shape
1.2.1 Control Over Cross-Sectional
which it is poured. It is possible to
Fresh concrete adopts the shape of the formwork inchanging the shape of the mold. All
1.1.3 Mechanism Of Load Transfer give any cross-sectional shape to the concrete
by
The function of most of the structures is to carry the proportions like depth over width ratio, width over thickness ratio and other shape
various types of loads applied on them is not true in case of steel
and to safely transfer these loads to the soil
or rock underneath. A number of different parameters are under the control of the designer. Thisalready made in the factories.
types of members contribute in resisting the applied structures where design is a selection out of the sections and is one of the main
Shape adoptability is very good for the architectural reasons
loads including the slabs, beams,
walls, columns and foundations. The term slab is
generally employed for the actual
horizontal roof surface used to provide access at a height from factors for the popularity of the concrete structures.
beams are used to provide support to the slabs and other upperthe ground level. The
leaving a parts of the building
clear space underneath. The load is transferred to the ends the 1.2.2 Easy And Universal Availability Of Constituents
of beam and available, leaving only the steel
the member undergoes bending in resisting the load.
Slabs also provide clear space Concrete consists of the cheapest materials that may be
produced in abundance.
underneath the load but the difference between the beam and slab
is that the beams are reinforcement that is costly in some places where steel is not
stone particles. Even
one-dimensional members and the slabs are two-dimensional members, Most of the concrete material is a mixture of naturally available
dimensions being negligibly smaller. Columns, walls and foundations the remaining the cement is manufactured from limestone and other
stones. The cost of
provide a direct grinding of the
support to the applied loads. manufacturing of cement only includes the cost for heating and
The gravity load passes from top to bottom through materials. These materials are locally available in most parts of the world.
all the members of the structure
until it reaches the underneath soil. The load acts at
the
underneath slab and transfers to the beams and walls. This isfloor finish, goes to the 1.2.3 Economy
then accumulated in the structures, especially in
columns, moves to the foundations and then finally
dissipates in the soil. The terms Concrete construction becomes economical in most types of
ingredients.
member and support are defined relative to each other. There
arc ho separate supports areas where steel is relatively costly. This is because of very cheap
in the structure as is normally seen in the structural
analysis books. For the roof slab,
beams and walls are supports. For the beams, columns are 1.2.4 Better Insulation Properties
the columns, foundations are acting as supports. acting like supports, and for
Similarly, the underneath soilacts as Concrete is a porous material and exhibits fairly good heat and
sound insulation
support for the foundations. steel construction. The fire resistance
As already stated, beam is a structural member
/ properties especially when compared to structures.
used to provide support to upper parts of properties are also much better than the steel
the building leaving clear space underneath and
to transfer the imposed loads to
supports at its ends. Usually the loads (or at 1.2.5 Good Bond With Steel
least their components) act in a transverse
direction to the longitudinal axis of the member. The strength, the
general behavior of any When steel bars are introduced in the concrete to improve its tensile
together
concrete makes a fairly strong bond with the steel and the combination behaves
4 CONCRETE STRUC TURES PART - 1 AUTHOR: ZAHID A. SIDDIQI MATERIALS AND DESIGN PROCESS 5
within acceptable limits. The coefficient of expansion of steel and that of CHAPTER):
concrete are reinforcing steel and mixing placing,
also fairly closer giving no problem when the reinforced members are cement, fine aggregate, coarse aggregate,
subjected to affect the resulting strength and other structural
temperature changes. The coefficient of linear expansion for
steel is 11.7xl0-6 per compaction, curing of concrete greatly
degree centigrade and for concrete it is 9.9xl0"6 per degree
centigrade. properties of the concrete.
1.2.6 Ductility And Warning Before Failure
The property of a material by which it can withstand
failure is said to be its ductility. Although concrete is excessive deformation up to
I 1.3.5 Limited Industrial Behavior
The industrial behavior in concrete construction
joining the precast concrete units and the
is limited because of the difficulty in
resulting appearance. The construction time
a brittle material, reinforced
concrete if properly designed may exhibit large deformations,
ductility and warning
is also accordingly greater for the in-situ construction.
before failure.
1.3.6 Inelastic Behavior
1.2.7 Lesser Chances Of Buckling at very low stress level. After
Concrete behavior is inelastic right from the beginning becomes much more
Concrete structures usually have larger thickness of members compared cracking and by the addition of the steel, the behavior
of the
widths and hence the chances of buckling are much reduced as compared with their complicated. This makes the application of the formulas derived on the basis
structures. with the steel assumption that the material is elastic not fully exact.

1.2.8 Aesthetics 1.3.7 Creep And Shrinkage


Creep is extra strain produced in the concrete members
when subjected to loads for
Various architectural shapes and surface textures may be given
to the concrete
structures. The appearance of these structures is hence usually much better prolonged durations. Shrinkage occurs mainly due to the volume changes within the
occurs early in the
types of constructions. than other concrete paste due to expulsion of water. Majority of the shrinkage cured or if
hardening process and causes cracking of concrete if it is not properly
1.3 DEMERITS OF CONCRETE CONSTRUCTION sudden large change in temperatures takes place.
1.3.1 Weaker In Tension 1.4 SPECIFICATIONS
Concrete is a brittle material and has very less and unpredictable tensile set of design rules, called
strength. In The adequacy of a structural member is in part determined by achecking strength, stiffness,
most of the designs, the concrete tensile strength is neglected and specifications, which include formulas that guide the designer in
steel bars are added to
resist tension. This increases the cost and makes the construction proportions and other criteria that may govern the acceptability of the member.
little more difficult.
and structures.
1.3.2 Greater Self Weight There are a variety of specifications that have been developed for both materials usage.
Each is based on years of research and experience gained through actual structural
Concrete structures are generally bulky and contribute in generating Following specifications will be used in this book quite often:
and dead loads compared with steel structures. In more self-weight
may become more than the imposed live and other concrete structures, the dead load
loads. I- ACI American Concrete Institute
2- AASHTO American Association of State Highway and Transportation
1.3.3 Cracking Officials.
Concrete is very weak in tension and hence, even if steel 3- ASTM American Society for Testing and Materials.
concrete members crack on the tension side. These
reinforcement is provided,
cracks may be invisible to the
naked eye but they reduce the moment of inertia, 1.5 TYPES OF LOADS
increase deflections, change the
structural behavior of the member and make the member
susceptible to penetration of the
water. The load that almost retains its magnitude and point of application throughout the life of
weight of the
structure is called Dead Load (denoted by D). This load is usually the self
1.3.4 Relatively Unpredictable Behavior structure (not only this member but all other members resting on it) which is estimated by
multiplying volume of a member with the unit weight of the material of construction. This
The concrete properties vary by a large amount due to the
involvement of a number of load constitutes a larger part of the total load on a structure. The load due to persons occupying
constituents and a number of processes involved in the construction. Properties the structure and their belongings is called Live Load (denoted by L) whose magnitude and
of the .
MATERIALS AND DESIGN PROCESS 7
PART - 1 AUTHOR: ZAHID A. SIDDIQI
6 CONCRETE STRUCTURES CHAPTER I:
point of application changes with time. In case of bridges, live load consists of weight of Materials.
vehicles moving over the bridge. Other types of loads are Wind Load (W), Earthquake Load Table 1.1. Unit Weights of Common
(E), Retained Water Load, Retained Soil Load, and Thermal Load due to temperature Unit Weights nf Common Materials
variations.
Material kN/m3 lbs/ftJ
22 6 144
1.5.1 Self Load 2300
p^in concrete 23.6 150
This is a type of dead load comprising of self weight of the member under
consideration. For design, a reasonable value of self load depending on past experience
Reinforced concrete
Burnt clavmasonDL-
—— -- 1600 to 1800
2400
1900 to 1930 18.9
15.7 to 17.7
120
100 to 110
is assumed in the start which is then compared with the actual self weight at the end. Earth, sand or brick ballas
Corrections in design are made if necessary. fillinp
Looseaggregate _ 1450 to 1600
13.9
90 to 100
88 to 90
1.5.2 Imposed Load
All external loads, leaving the self weight, acting on the member to be designed are
Loose cement
Light weight concrete
1415

1040 to 1520 10.2 to 14.9 —


65 to 95 —
called imposed or superimposed loads, including live load, wind load, earthquake load,
etc. Part of dead load may also act as imposed load.

1.5.3 Service Or Characteristic Loads


Clinker (1:8)
Pumice n

Expanded clay or shale



Foamed blast luiiid.ee stag — — 720 to 1120
1680 to 2080
1360 to 1840
1600 to 1920
7.1 to 11.0
16.5 to 20.4
13.4 to 18-1
15.7 to 18.9
45 to 70 _
105 to 130
85 to 115
100 to 120
No finesgravel .
14 1 to 15 7 i 90 to 100
The maximum intensity of load expected during the life of the structure depending or gas concrete
oiinlar, aerated
1440 to 1600
2560 25J . 160
upon a certain probability of occurrence is called service or characteristic load. No
Water hound macadam 96 1.0 6
additional factor of safety is included in the service loads. ^nshivfallen snow 19
300 3.0
1.5.4 Factored Loads Accumulated snow _J
500 4.9 31
Compacted snow 26.7 170
Lt f
Service loads increased by some factor of safety or overload factor are called factored Glass 19.6 125
2000
loads. All-in stone aggregate | 490
1.5.5 Dead Load ?

Dead load includes the self-weight of the structure and weights of any permanent
Steel
Copper
Brass
—— 8940
8500
77.0
87.7
83.3

558
530
558
fixtures, partitions, finishes and superstructure. These loads remain almost constant in
magnitude and fixed in location throughout lifetime of the structure. Dead load may be
Bronze
Aluminum
__ — — ..
8940
2770
87.7
27.2
111.0
173
707 ’

:— _
calculated with good accuracy from the geometric configuration, dimensions of the
11330
Lead 446
structure, and density of the material. Unit weights of some common materials are
given in Table 1.1.
Rolled zinc
Petrol
Cotton (in bales)
— . 7150
705
240 to 560
70.0
6.9
2.4 to 5.5
71
.
_ 44
15 to 35
45
1.5.6 Live Load Flour
720
1^0 1.3 8 _
The load due to persons occupying the building and their belongings like furniture, Hay pressed in bales 9.0 57
915
dresses, books and kitchenware, etc., is called live load. Its magnitude and the point of 560 5-5
application change with time. Nonpermanent partition loads (usually 100 kgs/m2, 1900 OldJS-VU IJLUlv
9.4 60
N/m2 or 20 psf) are also considered as live loads. The typical values of service live
loads for various uses are given in Table 1.2.
Packed paper
Waste paper (pressed)
— 960
560
800
3.J
7.9
35
50
ougar ^xoosoj_ 7.9 50
Average timber 800
8 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI
R 1: MATERIALS AND DESIGN PROCESS 9
Table 1.1 (Continued). Unit
Weights of Common Materials. reduced live load cannot be less than 50% of the original live load for columns
supporting one floor or for flexural members, and not less than 40% for other
Material L Unit Wei ghts of Common
Materials members.
, —
polid stone aggregate block — |I kgs/m2
Ibs/ft2- Table 1.2. Typical Minimum Uniformly Distributed Live Loads.
^OOmmorJjn. thick |
Hollow 200 mm or 4.31 Live Load
thick TVf[ 295~~”
’ 2.87
90
60
S.
Occupancy or Use Pa or Ibs/ft2
_

Brick work, 114 mm No. kgs/m2 N/m2


nr a s 111 1 265
thick 2.6 ~ 200 1900 40
54 1- Private apartments, school class rooms.
raving Slabs, 50 mm Or 2 m. "ik - 2- Offices;-': 250 to 425 2400 50 to 85
thick
.Roofing tiles
Gypsum piaster
.thick

I2mm or % in.
H 77
———
22 0
-
1.15

y-o to 0.9
215
24

~1Z5 to 19~
4.5
3-

4-
Fixed-seats, assembly halls, library
reading rooms.
Corridors in public buildings.
300

400
to 4000
2900

3800
60

80
Windows (metal or wooden
en 5- Movable seats assembly hall. 500 4800 100
24-4
.frames)
r ——
(Doors (wooden)
—1 39 0
kgs/m2 per
.
380
kN/m2 per
240

8
5 6-

7-
Wholesale stores, light storage
warehouses.
Library stack rooms.
610

730
1200
6000

7200
12000
125

150
250
I 1 mm thickness mm thickness Ibs/ft2 per 8- Heavy manufacturing, heavy storage
inch warehouses, sidewalks and driveways
.Asphalt ~
H
—— __ thickness subject to trucking.
Hardboard F
J
77
.
~~
in a

7'5
1 ——
4.0
12
5.5
9- Stairs, general.
10- Stairs, up to two-family residences,
50% more than specification.
500
300
4800
2900
100
60

61 [ 3.25
^8.9 to 23.6 “
10 to 12.5 Important floor slabs must also be checked for concentrated loads of magnitude 10 to
Industrial cfonl
mu-uiji steel ,
stairs, Im or-t— r 1
3 ft' — kgs/m

N/m Ibs/ft 35 kN (2 to 8 kips) spread over an area of 0.75x0.75 m2 (30x30 in2) to represent heavy
wide 820 56 items. The given live loads include the impact effects produced by normal use. Special

,oad
———
-
for this member may i 3e
impact factors are to be used for elevators, machinery, large reciprocating or rotating
machines and cranes.

400 sft by the folloi^ expression ““ ** is 8™ ter than 40 m2 ar 1.6 LIMIT STATES

L = 4(0.25 + < t
SI Units, 4, fa m2,
Limit state is defined as the stage in the loading after which the structure cannot fulfill its
Ja^ 5
intended function due to strength or serviceability considerations. The term “limit state” is
preferred compared with ‘Tailure” because in most cases of limit states, the actual failure or
L = 4(0.25 + 4^) collapse does not occur. Limit states are generally divided into two categories, strength and
^2 < T
~ L° EDCrr •
FPS Units, A, in sft
serviceability limit states. Strength or safety limit states means conditions of loading
To determine the influence 1
I
area of a given member, the corresponding to maximum ductile flexural strength (i.e., plastic strength), ultimate ductile
raised by a unit amount member is imagined to shear strength, buckling, fatigue, fracture, progressive collapse, formation of plastic
done, keeping the adjacent The portion of the loaded
area that is raised when this be is mechanism, overturning, and sliding, etc. Rupture refers to complete separation of structural
similar type of members unmoved, is
area. For example, for an interior member into two or more pieces causing collapse. Rupture may be due to flexural, shear,
beam, it is twice the tributary
called the influence
load to this member) of the
beam. For a column it is four times area (area contributing torsional of other types of stresses. Progressive collapse means spreading of failure once one
of the members collapses. Plastic mechanism produces infinitely large deformations in the
the tributary area. The
MATERIALS AND DESIGN PROCESS ] I
10 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
CHAPTER 1 •

than or equal to one to provide reserve


strength as a margin
e fatigue produces reduction in material strength Its value is always greater
of toads Serviceability limit states are those
due to repeated cycles of reversal
of safety; This factoradditionally covers the following aspects:-
concerned with occupancy of the building,
as excessive deflections, undesirable vibrations, permanent such should be able to
fire, explosions, corrosion deformation, accidental effects like in applied forces or loads. The structure
stee^ deterioration of
deflections may be visually unacceptable and may concrete and excessive cracking. More
1- Uncertainties loading up to certain extent without failure
cause disturbance in parts of the withstand the variations in expected
resting on the member. Excessive crack structure of the structure.
deterioration of concrete. widths lead to leakage, corrosion of the steel be small at service load conditions.
and 2- The deflections should strengths, particularly when the material
used
According to some specifications, 3- To cover uncertainties in material
serviceability limit states are the limiting
affecting the ability of a structure to preserve is under-strength.
conditions workmanship.
comfort of its occupants and function of its appearance, maintainability,
durability, or the 4- To cover, in part, poor developed.
machinery in case the theory is not fully
The structure should not cross any
under normal usage. 5~ To cover unexpected behavior
strength or serviceability limit for a perfect
applicable limits are to be checked by design. All the 6- To cover natural disasters.during fabrication and erection due to storage of
stresses produced safety
using the available procedures.
materials, movement of machinery and
labor may be very high. Factor of
1.7 VARIOUS DESIGN METHODS construction.
may take care of these loads during the level
of residual stresses and stress concentrations beyond
There are three main methods of 8- Presence
design, namely, allowable stress theoretically expected.
LRFD and plastic design. Before discussing design, strength
importance of factor of safety these methods, the basic design equationdesign or
in design are explained. and the
1.7.3 Strength Design loads
of safety is applied on the service loads to get increased
1.7. 1 Basic Design Equation If the major part of factor divided by the minor
Loads, and the material strengths are
In design, the applied forces called Ultimate or Factored design method is called Strength Design, Load
and moments due to external loads remaining part of factor of safety, the Limit State Design. The
maximum resistive forces and moments that
are equated to the
and Resistance Factor Design
(LRFD), Ultimate Design, or Design is called
member with a factor of safety which can be developed Wi thin the any load is increased for the
Strength
is always greater than or equal to material of the factor of safety by which factor by which the
produce external bending one. Loads can factor of safety or reduction
moments,
internal stresses are developed withinshear forces, torques and axial forces. Overload Factor and the reciprocal of is called Capacity Redaction
or Resistance Factor
the material to balance all the Similarly material strength is slightly adjusted the structure
forces and moments but with a above types of limiting stage in the loading after which
maximum (<$. Limit state is defined as the considerations.
summarized by the following design equation:resistance value. This concept may be due to strength or serviceability to be within
cannot fulfill its intended function performed considering the structure
Analysis of structures for loads is of
behavior, ultimate failure modes and redistribution as
Extern^
Forces/Pressures
x Factor of Safety (FOS)
7 Maximum Internal elastic range. However, inelastic
forces after elastic range are considered
in this method. This is more realistic
design
due to Loads Resistance Offered by Design.
Material of the Structure compared with the old Allowable Stress
/ Material Strength
1.7.4 Allowable Stress Design (ASD) given
right side of the basic design equation
A simply supported beam of
span (Z) subjected to a point load (P) method, the FOS is taken on
get the maximum bending
moment of PL
can be analyzed to For this
be produced if the material of However, this bending moment will only earlier. The modified equation is as
under:
the beam is strong enough to develop
strength. This means that the
answer of analysis may be true for biggerthe required Forces / Forces Per Unit Area =
Material Resistive / FOS
concrete beam but may not be true reinforced Forces / Stresses
for small wooden batten. Due to Loads
1.7.2 Factor Of Safety (FOS) strengths and service loads are
by reducing the material
Factor of safety is defined as the If the factor of safety is applied Stress
the design method is called Allowable more
value of the expected failure load
maximum service load. It is required to bring divided by the considered as such without any increase,
Design. This method is
conservative, giving
Design (ASD) or Working Stress compared with
a usable state. It is defined as the structure from the state of collapse loads are higher in magnitude
the failure load divided by the
maximum service loads.
to safe designs with greater costs when dead
allowable material strengths, the structure can be
the other loads. For the reduced or
MATERIALS AND DESIGN PROCESS J 3
12 CONCRETE STRUCTURES PART-1
assumed to act perfectly within the elastic ranee for iLa
„a,ysK for toads and riastie material behavior become
«7™°^ ZAH1D A‘
J
CHAPTER 1:
exists with a fixed value. The applied load may have a different value with only one
or equal to the member capacity for
stability.
condition

that the applied load must be lesser than


This method is now gradually replaced by Strength Design or LRFD for । may be denoted by and the design shear strength
may be
the structures flexural strength
where the behavior near collapse is fully understood. However, it I The design strengths and fa and fa are the
some engineers for important structures like atomic is still preferred by denoted by and are the nominal flexural and shear
reactors and pre-stressed concrete. J capacity reduction factors for flexure and
shear, respectively.
|
1.7.5 Plastic Design
I 1.9 ANALYSIS OF STRUCTURES
The third method of design is called Plastic Design, of the member, finding the maximum
loads
Strength Design but here the analysis for which is somewhat similar to the Knowing the material properties and dimensions called Capacity Analysis or
loads is also performed considering the using the design equation is
collapse mechanism of the structure. Full reserve that can be applied on tire member
strength due to indeterminacy and simply Analysis
inner elastic portion of the structure is utilized.
in the analysis and design. Deflections and Inelastic material behavior is considered
more important along with the strength requirements.other serviceability conditions become
I 1.10 DESIGN OF STRUCTURES
the members in the basic design equation,
Knowing the expected loads and span lengths of
1.7.6 Advantages Of Using Strength Design finding the appropriate cross-sectional
Or LRFD deciding the required material properties and
design may be defined as a mixture of
1. Strength design is another tool for proportioning of
concrete members, which dimensions is called Design of Structures. Structural feeling for the behavior of a structure
provides a flexibility of options intuitive
art and science, combining the experience and statics, dynamics, mechanics of materials, and
to the designer in selecting the design of
methodology. with a sound knowledge of the principles
structure which will serve its intended
2. Strength design becomes economical when dead
load is relatively larger structural analysis, to produce a safe economical
compared with live load. purpose.
3. Strength design may take into account the behavior of the building. He, sometimes, also
at the
ductility, warning before failure and strain-hardening, etc. collapse including An architect generally provides the overall layout dimensions. In majority of
possible in ASD because there the structure is This is not directly provides the sizes of structural members and insists on his selected and the structural designer.
considered at service stage, not the cases, the member sizes are negotiated amongst the
architect
approaching close to collapse. designer is to provide structural stability against the worst
The main responsibility of a
4. Plastic design concepts may conveniently be employed loads acting on roof surfaces
method.
in Strength Design combination of expected loads on the structure. In other words, soil underneath through the
and other parts of the structure are to be safely carried to the
5. Every type of load may be given a different factor path through the structure to
of safety depending upon its foundations. It is very important to understand the load flow
of a column does not act on
probability of overload, number of severe occurrences,
and changes in point of establish loads on individual members. For example, dead weight Further, as stated earlier, it is
application. a beam whereas dead weight of a roof slab acts on the beam.
6. for a member is
Safer structures may result under Strength Design
because of a better awareness necessary to understand the concept of support that is a relative term. Support
reaction to that
of structural behavior, especially close to collapse. a part of the overall structure that can provide constraint and correspondingacts as a support
1.8 DESIGN STRENGTH particular member. For example, beam acts like a support for slab, column support the top
for beam, soil provides support to foundations and underlying layers of soil other member
In Strength Design, design strength of all layers of the soil. The reaction provided by a member acting as support for
some
elements is obtained as capacity reduction factor direction.
(resistance factor) multiplied with
maximum stress that can be developed multiplied with acts as a load on the first member in the opposite
sectional area or section modulus. The design
strength is also called load capacity, or
sometimes, only capacity of a member. An
example to explain the difference between the 1.10.1 Objectives Of Structural Designer
due to any
member capacity and the applied load is that of a bottle.
This bottle may have a fixed liquid The primary objective of a structural design is the safety against loss of life
retaining capacity of suppose 1 litre. The safe failure. Further, the structure should perform satisfactorily at the service loads. In
retaining capacity may be considered equal to design process
0.9 litres to avoid spitting, etc. This allowance addition to these objectives, an optimum solution is obtained during the
may be considered as the capacity reduction
factor. However, the bottle may be empty at times satisfying certain criteria. Some typical criteria are:-
meaning that the amount of liquid retained
in it is zero litres but the capacity of the bottle
still remains the same. Any amount of liquid a) minimum cost
may be poured in this bottle that is not exceeding
1 litre. Similarly, load capacity of a member b) minimum weight
PART -1 . CHAPTER 1:
MATERIALS AND DESIGN PROCESS 15
14 CONCRETE STRUCTURES AUTHOR- ZAHID A. SIDDIQI
A
y
. • • , .
c) minimum construction time
d) minimum labor 1,10.2 Economy In Design
e) maximum efficiency of operation to owner, etc. Tire most economical design can be determined only by comparing the approximate
If a specific objective criterion can be expressed mathematically in the form of an costs of different designs. The prices of cement, aggregate, steel, formwork and labor
objective function, then optimization techniques may be employed to achieve the goal. determine the overall cost of a reinforced concrete structure. In beams and slabs, much
The structural designer must learn to arrange and proportion the parts of his structures of the concrete is in tension and is neglected in calculations and hence the use of
so that they can be practically erected and will have sufficient strength and reasonable moderate concrete may be economical than a rich concrete. In columns, the use of rich
economy. The important items, namely safety, cost and practicality, are briefly concrete is economical because all the concrete is in compression and because this
discussed below. concrete will require smaller size of the formwork. If good quality Grade 420 steel is
available at rate only slightly higher than Grade 300 steel, it may cause economy in
1. The structure must safely support the loads to which it is subjected. The construction. However, it is to be noted that Grade 420 steel increases the depth
deflections and vibrations should not be so excessive , as to frighten the requirements of the members. Doubly reinforced and shallow members are generally
occupants or cause unsightly cracks. uneconomical. The extra costs may offset the advantages resulting from reduction in
2. The designer must keep the construction, operation, and maintenance costs at sizes. When the cost of steel is high in comparison to that of the concrete, the most
the lowest level without sacrificing the strength. economical slab is that which has amount of steel well below the balanced steel ratio
3. Designers need to understand fabrication methods and should try to fit their (which will be explained later).
work to the available fabrication facilities, available materials and the genera]
construction practices. Some designers lack in this very important aspect and Formwork is cheaper if it is repeatedly used and is relatively simple. Cylindrical tanks
' their designs cause
problems during fabrication and erection. are economical than rectangular tanks for the same capacity. If formwork cost is lesser,
domed roofs and curved tank bottom may be more economical. Steel forms for beam-
Designers should leant everything possible about the detailing, the fabrication of steel slab-and-column construction are cheaper than timber formwork if twenty or more uses
reinforcement, and the field construction techniques besides the loads, mechanics, and can be assured. Cost also depends upon the design life, quality and the type of use of a
the expected material strengths. The designer must have information concerning the structure.
transportation of the materials to site, labor conditions, equipment for erection,
problems at site, field tolerances and the required clearances at the site. This 1.10.3 Procedure Of The Structural Design
knowledge helps to produce reasonable, practical and economical designs.
The structural framework design is the selection of the arrangement and sizes of
Objectives of a reinforced concrete designer may be summarized as under. structural elements so that service loads may be safely carried. Structural designer has
1. The arrangement of structural components (design layout) should be selected to complete the following steps to get a successful design:
such that the structure can fulfill its intended purpose, should have better a) The general layout of structures.
appearance, harmony and symmetry, and should be economical. • b) Studies of the possible structural forms that can be used.
2. The structure should be the most economical satisfying all requirements and the c) Consideration of loading conditions.
budget of the client. d) Analysis of stresses, deflections, etc.
3. The structure should be adequate to safely support all anticipated loadings and e) Design of parts.
should not have excessive deformations, vibrations or cracking. f) Design of assembly and connections.
4. The required structural maintenance should he a minimum. g) Preparation of design drawings.
The primary task of reinforced concrete design is to determine the amount of steel The above design procedure requires iterations and is repeated below to incorporate the
required in a particular member. However, after calculation of mount of steel required, trial method:
it is more important to properly detail this steel avoiding congestion and to provide
structural ductility. The designer must be familiar with the field practices, machinery at 1- The functions to he performed by the structure and the criteria for optimum
site, type of available formwork, type of materials, soil conditions, atmospheric solution of the resultiog design must be established. This is referred to as the
conditions and other site requirements. A perfectly calculated design may sometimes planning stage.
be impossible to Construct at site. Detailing of steel at the joints is very important 2- The general layout of the structure is decided.
where there re much more chances of congestion. 3- Different arrangements of various elements to serve the functions in step 1 are
considered. The possible structural forms that can be used are studied and an
16 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI MATERIALS AND DESIGN PROCESS 17
arrangement appearing to be best is selected for the first trial, called preliminary CHAPTER!:
structural configuration. Only in very rare cases, it has to be revised later on.
4- Loading conditions are considered and the loads to be carried by the structure 6. U= 0.9 0 + 1.0 W
are estimated. 7 u= 0.9 0+1.0 0
5- Based on the decisions of earlier steps, trial selection of member sizes is carried
out depending on thumb rules or assumed calculations to satisfy an objective
criterion, such as least cost.
6- Structural analysis involving modeling the loads and the structural framework
to obtain internal forces, stresses and deflections is carried out.
7- All strength and serviceability requirements along with the predetermined
criteria for optimum are checked. If any check is not satisfied, the member sizes
are revised. This stage is called evaluation of the trial member sizes.
8- Repetition of any part of the above sequence found necessary or desirable as a
result of evaluation is performed in this stage called redesign.
9- It is determined whether or not an optimum design has been achieved, and the
final decision is made.
10- Drawings are prepared to show all design details. An estimate for the required
quantities is also made. This stage of design is called preparation of design
documents.
1.10.4 General Design Flow Chart
The important sub-steps in design of parts (point no.7 above) are shown in the form of a
flow chart in Figure 1.1. Objectives of the design must always be kept in mind while
using this flow chart. The selection of trial section in step-2 depends on the
main
objective, availability of material, construction requirements and compatibility with
other members of the structure.

1.11 COMMONLY USED LOAD FACTORS AND LOAD COMBINATIONS Figure 1.1. General Design Flow Chart.

It is almost impossible that all loads like live load, snow load, wind load and earthquake occur The nominal service load symbols have the following meanings:
together with their maximum intensity. A load combination combines different types of
loads
depending on the probability of occurrence of these loads acting simultaneously, considering O Deadload
their expected intensity in the combination compared with the maximum load Intensity. The
factors of safety are also included in these combinations and hence the output of the F Loads due to weight and pressures of fluids with well-defined densities and
expressions is a factored design load. The alphabets used in the combinations mean different controllable maximum heights. Where F is present, it must be included with the
types of nominal service loads and the numerical values with them are the load factors. The same factor as O in Eqs. 1, 5 and 7.
last two combinations, given below, are very important for uplift of structure or reversal of T Cumulative effect of temperature, creep, shrinkage and differential settlement
forces. The wind load on roof is upwards in majority of the cases and if the downward gravity L Live load
load is less, the structure may be blown up or sagging bending may change into hogging H Loads due to weight and pressures of soil or water in soil. Vertical 77 is included
bending. A list of most commonly used combinations is as under: in O. When lateral 77 acts alone or adds to the effects of the loads, include it as
1. O=1.4D 1.677. Where lateral H is permanent and counteracts, include it as 0.977.
2. U= 1.2D + 1.6Z + 0.5(L,orSorR) However, lateral O is not included if it is not permanent and it counteracts other
3. U = 1.2 D + 1.6 (Lr or S or R) + (1.0 Lr or 0.5 W) loads.
4. O = 1.2D + 1.0 17+ l.Oi + 0.5 (£rorSorR) Lr Roof live load
5. 0=1.20 + 1.00+1.07+0.25 5 Snow load
MATERIALS AND DESI GN PROCESS 19
18 CONCRETE STRUCTURES PART - I AUTHOR: ZAHID A. SIDDIQI
CHAPTER L
R Rain load
stresses is to be used with these load combinations.
W Wind load No increase in allowable the following alternate
E Load effects of seismic forces allowable load combinations, UBC-97 also allows
In place of the above
The load factor on L in Eqs. (3) to (5) is permitted to be reduced to 0.5
combinations:
areas occupied as places of public assembly, and all areas where the live
except for garages, I. D + L + (Lr or S)
500 kgs/m . The wind load equation in ASCE 7-98 and IBC 2000 load is greater than 2. D + L + (WorE/lA)
directionality, which is equal to 0.85 for buildings. includes a factor for wind D + L + W+S/2
Where wind load W is based on service D + L + S+W72
level wind, \.6Wis to be used in Eqs. 4 and 6 and0.8JPis
used in Eq. 3. 5. D + L + S + E/IA
The basic LRFD load combinations according to UBC-97 0.9D + EAA
are as under: 6.
in
1. 1.4D basic load combinations, a one-third increase is allowed
2. When using these alternate includingW or E. Design snow loads of 1.44 kN/m2 or
3.
1.2D + 1.6Z + 0.5 (Z, or S)
allowable stresses for all combinations
seismic loads. Where design snow loads exceed 1.44 kN/m2,
4.
1.2D+1.6(ZrorS) + (/iZor0.8IT)
1.2D + 1.31F+/1Z + 0.5 (Zr or S)
lessneed not be combined with included with seismic loads, but may be reduced up to 75 percent
the design snow load is to be
5. based on the site conditions. Where F, H, P or Tare
1.2D + 1.05 + (/iZ +/1*5) where approved by the building official added to the combinations.
6. 0.9D±(1.05orl.3fr) also acting, each applicable load is to be
1.0 for floors in places of public assembly, for
live loads in excess of 4.9 kN/m2, and
for garage live load. 1.12 SHRINKAGE OF CONCRETE
0.5 for other live loads. during hardening and drying under constant
0.7 for roof configurations (such as saw tooth) that do not Shrinkage is the shortening of concrete
the shrinkage is unrestrained, the
0.2 for other roof configurations.
shed snow off the structure. temperature. For concrete that can dry completely and where28 days and 0.00035 at 3 months.
0.00025 at
The above load combinations should be multiplied by linear coefficient of shortening is approximatelymay approach a maximum of 0.0005 at the end
1.1 for concrete and this the shrinkage change is less rapid and
After
combinations include seismic forces. Where F, H, P or Tare acting, masonry where load of 12 months. The primary type of shrinkage is the
drying shrinkage or simply shrinkage that
each applicable load
should be added to the above combinations factored as 1.35, 1.6H,
1.25 and 1.2T. from the surface of the gel particles. If but
occurs due to the loss of a layer of adsorbed waterwater
the
(needed for workability of concrete
The ASD load combinations according to ASCE 7, Section concrete is exposed to air, the larger part of free
2.4 are as under:
not for hydration of cement) evaporates with time. As the concrete dries, it shrinks in volume.
1. D+ F
If the water escapes earlier before sufficient gain of
the concrete strength, the shrinkage
2. in water, it expands and
3.
D+H+F+L+T produced becomes more. Conversely, if dry concrete is immersed
D + H+F+^otS or R) regains part of the volume loss due to shrinkage.
4. D + H+F+0.75(L + T) + 0.75 (Zr or S or 5) undergoes shrinkage. The
The aggregates do not shrink and it is the cement paste only that
'

5. D + H+F+(Wor0.7E) with larger aggregate-to-


aggregates act to restrain the shrinkage and hence shrinkage reduces
6. D + £?+ 5+ 0.75(15or 0.75) + 0.75Z + 0.75 (Zr OrS or.fi) shrinkage and the contribution
cement ratio. High water-cement ratio increases the amount of
7. 0.6 Z) + 1F+ An effective means of
8. of this factor is significantly greater than the effect of cement content.
0.6 D + 0.7E + H fresh concrete to a minimum
reducing shrinkage is to reduce the water content of the for the
The factor of 0.7 with the earthquake load may be replaced
maintaining the required workability. Prolonged and careful curing is beneficial
with 1.0 if this load is calculated by
some earlier code giving the service load in place of the shrinkage control.
ultimate load, The ASD load
combinations of UBC-97 are as under: in carbon dioxide
A secondary form of shrinkage, known as the carbonation shrinkage, occurs of shrinkage.
1. D rich atmospheres such as in parking garages and may double the amount in structural
2. cracks
D + L + (Lr or S) Shrinkage of concrete may cause unattractive and often harmful
3. D + (lkorZ71.4) components. In indeterminate structures, it can cause large and harmful stresses.
If a concrete
place, tensile
4. 0.9D± 5/1.4 member is restrained so that a reduction in length due to shrinkage cannot takeMPa (500 psi)
5. D + 0.75[Z + (Z, or S) + (J7or £71.4)] stresses are developed. A strain of 0.0002 may correspond to a stress of 3.5
20 1
'

when restrained. In such cases, it is important to AUTHOR: ZAHID A. SIDDIQI MATERIALS AND DESIGN PROCESS 21
reduce or CHAPTER I:
strong concrete, proper curing, extra steel and
Approximate shrinkage in 70 years in
neutralize these stresses by using
properly designed joints.
normal weight structural
a -
; isture condition, age and
compressive strengths is
rate of loading. The maximum reduction in bending and
generally 45 percent of the corresponding static strength.
strength of 20 to 50 MPa (3000 to concrete having cylinder
humidity of about 50% is as
7500 psi) for dry atmospheric
STRENGTH
follows: conditions with relative gj 1.15 CONCRETE COMPRESSIVE
For effective thickness of 150
mm, shrinkage strain = 0.00056 specified on the construction drawings and used in the
For effective thickness of 6
10 mm, shrinkage strain 1 The concrete compressive strength (//)
= 0.00047 a calculations is measured by compression tests on 150 x 300 mm (6 x 12 in) cylinders tested
1.13 CREEP OF moist curing. According to ACI 5.6.2.4, the standard strength test means the
CONCRETE after 28 days of
two 150 x 300 mm cylinders or at least three 100 x 200 mm
4 average of the strengths of
Creep is the property of a cvlinders, from the same sample, tested at 28 days age. Alternately, for the local conditions,
stress. Creep strains occurmaterial in which it continues to deform with time under a may taken identical to 80% (75% is more safe value) of the average
fl this strength test
become thinner. This changebecause the adsorbed water layers
in thickness occurs rapidly
constant
between gel particles tend to compressive strength of two 150x150x150 mm cubes or 80% of the average compressive
Bonds form between the gel at first and slowing down
with time. strength of three 100x100x100 mm cubes (not an ACI provision).
even after the removal of particles in their new position and hence a residual strain «
load. The time-dependent J|j
stress after initial elastic strain is strain produced with time due remains The apparent compressive strength of cubes is higher because of more restraining effect of
called specific creep to a unit platens of the machine. The local compressive stresses developed near the platens prevent the
proportional to the magnitude strain. Creep deformations are
concrete. Majority of the creep
of the applied stress and directly
inversely proportional to the strength 4 splitting of concrete. Cylinders give strengths closer to the actual crushing strength of concrete
from about 1.2 to 3 times the occurs in 2 to 5 years and the of because of possible failure pattern resembling concrete in most of the structural members.
instantaneous maximum creep strain ranges
strain. Typical values of
corresponding to various concrete strengths
are given in Table 1.3. maximum specific strains Samples for strength tests shall be taken in accordance with ASTM C 172 (Method of
Sampling Freshly Mixed Concrete). Cylinders for strength tests shall be molded and
Table 1.3. Typical Creep laboratory-cured in accordance with ASTM C 31 (Practice for Making and Curing Concrete
Strains. Test Specimens in the Field). These shall be tested in accordance with ASTM C 39 (Test
Method for Compressive Strength of Cylindrical Concrete Specimens).
MPa Specific Creep
si % per MPa Strength of the finished concrete varies considerably due to variations in the properties or
20 % per psi proportions of its constituents, transporting, placing, compaction and discrepancies in the tests.
3000 0:0145
)
4000 0.000100 Hence, ACI 5.6.2 requires that samples for strength tests of each class of concrete placed each
0.0116 0000080 day shall be taken not less than all of the following:
6000 0.0080 ~
55 8000 0.000055^ > once a day.
00058 0.000040
Creep strain for 1:2:4 concrete > once for a batch of 110 m3 (150 yd3) of concrete.
loaded at 28 days with a sustained > once for every 460 m2 (5000 sft) of surface area for slabs and walls.
0.0003 at 28 days after loading stress of 4 MPa (600 psi) is
approximately of the same degree ofand 0.0006 at one year time interval. Thus, creep is Strength level of a certain concrete is considered satisfactory if both of the following are
magnitude as shrinkage. satisfied:
1.14 FATIGUE IN Every arithmetic average of any three consecutive strength tests should not be less than
CONCRETE
the minimum required compressive strength,/.' .
Fluctuating or repeatedly applied
are greater than a certain loads reduce the. strength No individual strength test should give strength less than // - 3.5 MPa (500 psi) for
limit. One cycle of loading is of concrete if the cycles of loading
certain minimum value to a defined as the variation of load from // < 35 MPa. This limit is considered as 0.10 // when // > 35 MPa.
certain maximum value and a
well-defined tensile stress level to a well then back, or, variation from
some The specified concrete strength (//) is always lesser than the mean strength of the concrete
original tensile stress. This defined compressive stress level and
phenomenon of reduction then back to
cycles of loading is
called fatigue and is directly in concrete strength due to repeating (//) that is used for mix proportioning. According to the ACI Code 5.3.2, if previous
micro cracks. Inhibiting the linked to the development and growth statistical data is not available, the required average compressive strength (//) of concrete
growth of such cracking
may reduce fatigue. Fatigue of
depends on range of
variation of stress, number of cycles of loading, frequency of
strength should be taken as under:
variation.
CONCRETE STRUCTURES PART -I MATERIALS AND DESIGN PROCESS 23
22 AUTHOR: ZAHID A. SIDDIQI chapter l:

Specified compressive strength Required average compressive strength Table 1.3. Grades of Concretes.
4, MPa 4, MPa Grade fc (MPa)
<21 Grade 4
4 + 7.0 MPa CTO 10 C30 30
21 to 35 C35 35
4 + 8.3 MPa C12 12
C40 40
>35 15
1.104 + 5.0 MPa C15
18 C45 45
If statistical data is available and Cl 8 50
4 is less than or equal to 35.0 MPa (5000 psi) then is to C20 20 C50
be taken larger of the following: 22
C22
25
= f: + 1.34 s, C25
28
C28
fa = 4 + 2.33 s.T - 3. 5 MPa (500 psi)
Where ss is the sample standard deviation of at least 30 consecutive
tests.
The first equation gives lowest value of
/^ required for the probability of occurrence of 1.16 REINFORCING STEEL Ribs
average of any three consecutive strength tests below is very less
4' less than 1 in 100. In this way,
probability of any individual test giving strength below f'c Because the tensile strength of concrete
usually neglected in
becomes lesser than 1 in IT. The as well as unpredictable and is
second equation makes sure that probability of any the design of reinforced members, steel bars or wires
individual strength test having strength tensile
lesser than //- 3. 5 MPa (// - 500 psi) remains at the most 1 in
100. are to be embedded in concrete to resist the
If statistical data is available and stresses. The most common type of reinforcement
4 is greater than 35.0 MPa (5000 psi) then is to be taken for non-prestressed members is hot-rolled deformed
or
larger of the following: bars. These are round in cross section with lugs Bar Used as
deformations, called ribs, rolled into the surface to Fig. 1.2. Deformed
. fa ~ ;

help in anchoring the bars in the concrete (Fig. 1.2).


Reinforcement in Concrete.
4 = 0.904 + 2.33 the concrete between two closely
Square bars are not preferable in concrete because placing
Concrete is a brittle material. However, the stress¬ spaced square bars is significantly difficult than between two round bars. Further, bending of
compared with possible bending of round bars in
strain curve is nonlinear and appears to be somewhat such bars is only possible in two directions rolling and must conform to
by hot
ductile due to development of micro-cracking within all the directions. The bars are made from billet steel in three grades with
are usually available
the concrete and the resulting redistribution of stress ASTM A 615M, A706 or A996. Reinforcing bars MPa (60 ksi) and 520
(50 ksi), 420
within the concrete mass. A typical stress-strain yield strengths of 280 (or 300) MPa (40 ksi), 350 MPa Grade 52O (Grade 40,
Grade 420 and
curve for concrete is shown in Fig. 1.1. MPa (75 ksi) known as Grade 280 (or 300), Grade 350,
Nominal size of a bar is
The ultimate strength is larger than the breaking Grade 50, Grade 60 and Grade 75 in US Units), respectively.
5 or 10 mm in SI units or in multiples of
strength. The breaking strain for high strength specified in millimeters that are usually multiples of and 57). Generally sizes
13, 16, 19, 22, 25, 29, 32, 36, 43
concretes is close to 0.002, whereas, it varies up to '/sd! of an inch translated in mm (10,280, up to 60 mm in Grade 420 and 20 to 60 mm in Grade
0.0045 or even more for medium and low strength up to 20 mm are available in Grade diagrammatically shown
520. Stress-strain curves of these three types of reinforcing steel are
concretes. ACI Code proposes conservative value of
crushing strain for all except very high strength in Fig. 1.3.
strengths of 460 MPa for 6 to 16
concretes equal to 0.003. BS 4461:1978 cold worked bars are available in characteristic
the specification). The
According to ACI 1.1.1, the minimum specified compressive strength mm bars and 425 MPa for over 16 mm bars (BS has withdrawn
6 and 50 mm are not a
17 MPa (2500 psi).
is not to be lesser than available size range is 6, 8, 10, 12 16, 20 25, 32, 40 and 50 (although and grades as simple
preferred sizes). BS 4449:1997 prescribes steel bars of the same sizes
at least 15% higher than
The following grades of concrete may be defined based on the
minimum concrete compressive deformed steel bars. The ultimate strength of 250MPa steel must be strength for
characteristic
strength: the characteristic strength and it should be 5% or 8% higher than the replaced with Thermo
460 MPa depending on grade A or B. Cold worked bars are now
8

24 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI MATERIALS AND DESIGN PROCESS 25
Mechanically Treated (TMT) bars replacing the laborious twisting operation. The residual
CHAPTER!:
stresses are not developed, the corrosion resistance is improved, the ability to accommodate FOR STEEL
welding is better and the loss of strength at low temperatures is avoided. This steel can be LOCAL ACCEPTANCE CRITERIA
used of the relevant
in the humid and coastal areas and in seismic zones 2 and above. In TMT process, the bars are supposed to qualify the tension test if all the requirements
passed through a specially designed water cooling system so that the temperature of the A steel sample is The most general requirements are as under:
surface of the bars drops considerably, while the core remains hot. This rapid quenching
outer ASTM or BS standard are satisfied.
of the
steel bare creates a temperature gradient in the bars. Outside the based on the nominal area must be equal to or larger
from the core to the outer surfaces of the bars causing further
cooling system, the heat flows (a) The yield strength calculated strength.
tempering of steel bars. This than the specified minimum yield
increases the yield strength of steel with better elongation. BS on the nominal area must be equal to or
steel and it is manufactured in grades 415 to 550 according
4449:2005 recognizes TMT (b) The ultimate strength calculated based strength.
to various standards and local larger than the specified minimum ultimate
practices. equal to or larger than the minimum specified limit,
(c) Percentage elongation must be than or
According to ACI 9.4, yield strength of reinforcement greater seismically more active regions (Zones 3 and 4),fjfy must be greater
than 550 MPa (80 ksi) should not
be used for flexure and greater than 420 MPa (60 ksi) equal to 1.25.
should not be used for shear and torsion. that the total yield
However, may be used up to 550 MPa if the steel is maximum of 6% under-weight steel bar is allowed provided nominal specified
497. ACI 3.53.2 allows fy to be equal to or more manufactured according to ASTM A than the
load and the total ultimate load are equal to greater actual less weight.
than 420 MPa if the yield is taken The contractor is to be paid for the
corresponding to a strain of 0.35 percent. Otherwise, if fy is values (fyAb
taken corresponding to a strain of 0.5 percent As already
less than 420 MPa; the yield is over-weight steel bars, the contractor is to be paid for the standard
stated, the bars must satisfy ASTM (f) In case of
A615M for carbon steel, A706M for low-alloy steel or A996 for weight..
stainless steel.
1.17 MODULUS OF ELASTICITY
and is equal to 200,000 MPa or
Modulus of elasticity of steel (Es) remains nearly constant and hence the slope of tangent at
29,000 ksi. The stress-strain curve for concrete is non-linear
point to point. Also it may
any point (called tangent modulus of elasticity) is different from
using the modulus of elasticity
vary with the rate of loading during the test ACI recommends
of the stress-strain curve. The value
of concrete (£J as the slope of the initial straight portion
computed by the empirical equation
for concretes up to strength of 42 MPa (6000 psi) can be
given below:
Ec (MPa) = 0.043 w‘J 777 (SI) Ec(psi) = 33 w’s^ (FPS)
(pcf), between 1440 and 2560
Strain where wc is the unit weight of the hardened concrete in kgs/m3
Fig. 1.3. Typical Experimental Stress-Strain Curves for kgs/m3, and is its strength in MPa (psi).
Steels.
may be computed as follows:
A bilinear stress-strain curve is used in design that is a For normal weight concretes with wc = 2300 kgs/m3 (144 pcf), Ec
simplified form of the experimental stress-strain curve Ec(MPa) = 4700 777 (SI) £c (psi) = 57,000 777 (FPS)
(refer to Fig. 1.4). The first line starts from the origin
and goes up to the yield or equivalent yield stress. The
1.18 REQUIREMENTS FOR CONTRACT DOCUMENTS
second line is a horizontal line after yielding
neglecting the strain hardening in steel. Yield point is registered engineer or
measured in tests using halt of machine needle, strain
All copies of drawings and specifications have to bear the seal of a
architect. Following details are to be provided on these documents:
offset or particular strain values. The percentage
elongation of steel after fracture measured over a J . Details of Code according to which the design is carried out.
gauge length of 200 mm (8 in) should not, in general, 2. Live load, earthquake load and wind intensity.
stages.
be less than 12% for Grade 280 or 300, 9% for Grade Fig. 1.4. Idealized Stress- 3. Specified compressive strength of concrete at various construction
420 and 7% for Grade 520. Strain Curve. 4. Specified grade of steel.
26 CONCRETE STRUCTURES PART - 1
5. Reinforcement details, size and AUTHOR; ZAHID A.
location of all SIDDIQI MATERIALS AND DESIGN PROCESS 27
6. Designed dimensional changes structural elements. CHAPTER 1:
as a result of creep, danger without perhaps specifying the best
shrinkage and thermal stresses. requirements necessary to protect the public from
1.19 DUTIES OF
INSPECTOR practice The mam objective of a code
is usually to eliminate the most common mistake*
practice that is
Following are the minimum especially those involving safety. A recommended practice defines the best
The assumptions, procedures, construction
responsibilities of the inspector:
1. Quality and mix proportions
of concrete materials are
Ty t0
difficulties, and thumb rules for design are contained in these practices.
practice.
ACI code iswritten
of concrete is to be
verified. to be checked and final strength the form of a code but is not far away from a recommended
2. Formwork design
and its
should only be allowed afterconstruction should be checked. Removal of 1.21 TWO-CYCLE MOMENT DISTRIBUTION
3. Amount and placing the specified strength gain formwork
and in proper sequence.
of reinforcement must be
4. Mixing, placing and checked. At the time of initial proportioning of the members, the member sizes are not available. In the
5. The inspector must curing of concrete must be carefully monitored. absence of member stiffnesses, the exact analysis for loads becomes impossible. In order to
keep a record of the general
6. Construction loading progress of the work.
should be carefully controlled. speed up the process of determining the trial sizes, the stiffness ratios may be assumed and
approximate and quick analysis may be performed. For gravity loads, one of the approximate
1.20 ’^“STANDING OF DESIGN CODES methods is the two-cycle moment distribution. The major steps involved in this method are
listed below:
There are three closely
Specifications are set of related terms namely 1. Assume the stiffness of each member meeting at a joint to be equal. The stiffness of
rules given by individuals specifications, codes and standards. members is not known for the preliminary design. Calculate distribution factors for the
construction in a safe, efficient and or organizations to cany
economic way. out design and members accordingly.
quality control, workmanship
and testing of different Standards are generally related with the 2. The 2-cycle moment distribution is equal to performing ordinary moment distribution
detailed verification for the materials. A set
Specifications have veiy corresponding performance may be of specifications after for a total of 2 cycles and then terminating the analysis.
less or no legal standing as compiled 3. It is used to find moment at a single required joint out of a big frame.
Codes are developed by a even individuals may propose as a code. 4. Apply fixed end moments at the joint under consideration and the joints adjacent to it
number of different organizations these.
The provisions of the code such as AISC, ACI 5. Unlock the joints adjacent to the joint under consideration. Calculate balancing
suppliers / fabricators arid are finalized by a voluntary consensus between and AASHTO. moments and distribute into the members. Carry-over these moments to the joint under
law but have no legal university teachers / students. The codes are written the designers, consideration.
standing until a particular in the form of
value comes through the code is selected for 6. Re-lock the adjacent joints.
and through promoting free Government laws to protect life and to provide construction. Its legal 7. Sum up the moments at the joint.
trade and commerce. Every Government Department,safety to the people
Department and Federal / 8. Unlock the joint of interest. Apply balancing moment and distribute into the members.
already available codes. Provincial Agency may issue their own Codes or Provincial 9. Sum up the moments again.
accept one of the
10. For fixed end moments on the joints, use moments acting from the members on the
Model codes are those
Departments may adopt codes that are compiled by independent organizations joints, counter-clockwise positive.
them.
State / Government Department.These codes do not have legal jurisdiction unless and State
Examples of some codes published adopted be a Example 1.1: Calculate the bending moment at point B for the given continuous beam for
1. Uniform Building Code (UBC), in USA are as under: initial proportioning. The span lengths and the loading are shown in Fig. 1.5.
International Building Code (IBC). better for its seismic provisions,
now replaced with the
2. Southern Standard Building Code 10 kN/m
3. National Building Code (N.B.C.)
4. American Building Code “2
5. Building Officials Cbhsress nf Am—
Model codes may rely upon
... w/ry‘-
provision of the model other documents or
codes. standards. Cities are free to change
have to decide what provisions Engineers have to critically look into the any
are satisfactory and Building Codes and 7.5 m 7.5 m 7.5m
written in the form of a law
for enactment by publicwhich ones are to be modified. A code is
bodies. It usually contains the
minimum Fig. 1.5. Continuous Beam for Approximate Analysis.
materials and DESIGN PROCESS 29
PART - 1 CHAPTER 1:
28 CONCRETE STRUCTURES AUTHOR: ZAHID A. SIDDIQI
in Fig. 1.8, using the Portal
and shear for the frame, shown
Esample 1.2: Find moments
Solution? frames are on 6m centers.
Frame Method. The
The fixed end moment for each beam segment is calculated as follows: Wind pressure = 700 Pa
10x7.52 Half height = 2m
Mf = = 46.88 kN-m C/C distance of frames = 6m
12 12
For the moments acting on the joints, Solution:
counter-clockwise direction may be
Wind load on roof = 1000
= 8 4 kN
considered positive. This corresponds
to clockwise moments acting on the = 16.8 kN
members from the joints being Fig. 1.6. Fixed End Moment for a Wind load on level-2 = 2x8.4
positive (Fig. 1.6). Beam Segment having UDL.

From the two-cycle moment distribution, as shown m Fig. 1.7, it is clear that the
approximate moment at joint B will be 58.60 kN-m. Considering actual stiffness after
actual beam design, where the central panel may be less stiff, the difference between the
actual and approximate results is not more than 5 to 10 percent.

F.E.M. -46.88
U~
Balancing &
Distribution Points for Example 1.2.
C.O. Fig. 1.8. Data and Position of Inflection
Sum I +70.32 -46.88 are explained in Fig. 1.9 to 1.11. The
Balancing &
-11.72 -11.72 The calculations for the shears and moments Figs. 1.12 and 1.13 and the shear force
Distribution bending moment diagrams are plotted in
diagram for beams is plotted in Fig. 1.14.
Sum +58.60 -58.60

Fig. 1.7. Two-Cycle Moment Distribution.


8.4 kN- — Roof
2m
1.22 PORTAL FRAME METHOD FOR LATERAL LOADS

The approximate method that may be used for the approximate analysis of frames for the ^(8.4) «1.4 kN ~(8-4) =4.2 kN |(8.4) = 2.8kN
lateral loads is called portal frame method. The salient features of this method are explained
below: Fig. 1.9. Columns Shears for Upper Floor.
1.9, as
1. This method gives good results for moderate-rise buildings. the column shears, given in Fig.
2. Assume inflection points at column mid-heights and at beam mid-spans. The column moments may be calculated from
3. Assume horizontal shear of each column equal to the total horizontal stoty-shear follows:
distributed proportional to its tributary area. For example, if adjacent spans are equal, Column moment at A = 2(1.4) = 2.8kN-nx
interior columns carry twice the shear than the exterior columns. Consider this shear to
be acting at the mid-height or inflection points level.
Column moment at B = 2(4.2) = 8.4kN-m^
Column moment at C - 2(2.8) = 5.6 kN-m)
4. Due to column shear at the level of the inflection points, find the moments in the
columns at the beam level.
5. Start calculating the beam moments from one end to maintain joint equilibrium.
34 -
CONCRETE STRUCTURES PART 1
AUTHOR: ZAHID A. SIDDIQI
Space For Reader’s Notes
Chapter - 2 I

FLEXURAL BEHAVIOR OF BEAMS


UNDER SERVICE LOADS I

2.1 INTRODUCTION I
To resist applied bending moments, flexural stresses are developed within the member normal I
to the cross-section. These normal stresses are tensile on one side of the neutral axis and I
compressive on the other side. The overall resultant of these normal stresses is zero if there are |
no axial loads applied on the member. However, the individual resultant tensile and
compressive forces, acting in opposite directions, produce a resisting couple or moment equal
and opposite to the applied moment. To resist applied shear forces, shear stresses are
developed within the member parallel to the cross-section. The formulas and the related design
of beams presented in this chapter are based on ACI 1999 alternate method of design.

2.2 ASSUMPTIONS REGARDING FLEXURAL BEHAVIOR


Following assumptions are generally considered to simplify the flexural behavior, to study the
basic concepts and to derive the basic formulas for analysis and design.
L Plane sections remain plane after bending. This may not be exactly true closer to the
ultimate conditions.
2. There exists a perfect bond between the steel and concrete and hence the strain in steel
is exactly equal to the strain in the surrounding concrete. This assumption may not be
exactly satisfied after the cracking of the member.
3. The external forces acting at any cross-section are fully balanced by the internal
resistive forces. This condition remains valid up to the failure of the structure.
4; It is assumed that Hooke’s law is applicable. The stresses in concrete and steel may be
estimated from the corresponding strains using stress-strain curves and modulus of
elasticity of the materials.
5. After the appearance of first hairline or visible cracks, the concrete strength in tension is
neglected. Actually, the concrete in-between two cracks and near the neutral axis may
provide a limited amount of tensile strength. Similarly, even after development of
cracks, concrete may resist shear (and its associated tensile component) up to certain
extent in the presence of the usual steel reinforcement.
& The stress-strain relationships of concrete and steel are simplified to study the complex
interaction of the two materials in the reinforced concrete members, particularly closer
to the collapse when the materials are in their inelastic ranges. Strain-hardening of steel
is ignored and empirical modulus of elasticity of concrete is utilized in the
formulations. The effect of these two simplifications is only indirectly incorporated in
tothedesign.
7. Concrete is assumed to be crushed at an assumed value of ultimate strain, which is
usually taken equal to 0.003.
FLEXURAL BEHAVIOR OF
36 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI beams under service loads 37
CHAPTER 2:
machine and it increases with its distance from the
2.3 FLEXURAL BEHAVIOR IN GENERAL the strain increases with time in a
The neutral axis further moves upwards due to
neutral axis in the flexural member.
The formulas to predict the maximum resistive moment are developed by using the strain cracks penetrating up to greater height.
distribution diagram, stress distribution diagram and resultant force diagram along the depth of
the cross-section. The above mentioned assumptions are employed in construction of these
diagrams and deriving the formulas. The steps in the derivation of formulas are explained
below: N. A.
1. A plane cross-section remains straight and plane after bending. However, it may be
inclined to its original vertical orientation. This means that strain distribution diagram Strain Diagram Stress Diagram
after bending remains straight line, as shown in Figs. 2.1 and 2.2. Steel Bars
Fig. 2.3. Concrete Stress Diagram after Cracking.

Original Orientation of Cross-Section

N. A.
—-Position of Cross-Section after Bending
N. A.

Fig. 2.1. Strain Diagram Due to Bending. Strain Diagram Stress Diagram
Steel Bars
The axis or line along the member which does not change in length or which has zero
strain is called neutral axis. Fig. 2.4. Concrete Stress Diagram Closer to Ultimate.
2. If the member is made up of an elastic material having same properties in tension and center of gravity of the
compression, the stress remains proportional to the strain and the stress diagram also 3. Before cracking of the member, the N.A. passes through the
by the usual bending formula
remains a straight line with zero stress at the N.A. (Fig. 2.2). cross-section and the intensity of stress can be determined
asunder:
My
I J
I
'
N. A. where M at the section
= applied bending momentneutral
I axis
= moment of inertia aboutstress is required from the NA.
y = distance of fiber where
Strain Diagram Stress Diagram
Fig. 2.2. Concrete Strain and Stress Diagrams Before Cracking.
M
~
_
~
M_
max S
Where,S elastic section modulus of the section
Concrete is weak in tension and hence its fracture or cracking will appear at the tension
face when the tensile stress at the face exceeds the modulus of rupture. At this stage, Similarly, shear stresses can be determined from the basic shear stress formula.
all tension is transferred to steel. However, if the loads are such that the maximum
compressive stress is lesser than approximately 50% of its ultimate value, the stress V
diagram in compression remains straight line with no stresses on the tension side of the lb
neutral axis, as illustrated in Fig. 2.3. The neutral axis significantly moves upwards due vyhere V = applied shear at the section
to this cracking. Q = first moment (called statical moment) of area of cross-section
Closes to the ultimate collapse stage, the stress diagram resembles the stress-strain outside the point where shear stress is to be determined about
curve in a compression test because, away from the N.A., the strain is linearly N.A.
increasing exactly as in the testing machine (refer to Fig. 2.4). Only difference is that b = width of member at the point under consideration
38 CONCRETE STRUCTURES PART - I FLEXURAL BEHAVIOR OF
AUTHOR: ZAHID A. SIDDIQI BEAMS UNDER SER VICE LOADS 39
CHAPTER 2:
At any point within the member,
the principal stresses are assumed concrete strain corresponding o crushing
expression: given by the
following Scu =
= 0.003
,2 Sc = concrete strain
2~V 4 S* = steel strain
The angle of these stresses with P steel ratio = A5 1 bd
= resultant compressive force
respect to horizontal is given =
by:
T = resultant tensile force
« - T-
2 The cracks that appear in the regions having very less shear and high moments are vertical and
these extend from tension face toward the neutral axis. These cracks are shown in Fig. 2.5 and
2.4 TERMS USED TO STUDY FLEXURAL are called flexural cracks. In the regions where both shear and moment are high, vertical
BEHAVIOR cracks appear first reducing the cross-section available to resist shear. Shear then becomes
The terms and parameters used in understanding critical and the cracks are turned at an angle and may even cross the neutral axis if the section
diagrammatically explained in Fig. 2.5 and are given the flexural ofr ubeams are is weak in shear. This cracking depends on the magnitude of principal stresses acting due to
below: combined action of flexural and shear stresses. However, only shear stresses are present closer
4 = area of reinforcement on tension face to NA. These cracks that are transverse at the face but inclined near the N.A. are called
b = width of the compressiontite face of the beam flexural-shear cracks. Pure shear cracks originate at the NA. and quickly propagate to the
d = distance of centroid of tension farec at almost 45° angle. These cracks appear in the region of high shear but very less
called effective depth of beam steel from extreme fiber in compression; moment such as near the simple supports and are termed web-shear cracks. It is important to
h = total depth of the member remember that most of service load cracks for a correct design are hairline cracks not visible to
= specified 28-day compressive cylinder naked eye.
= allowable stress in concrete, equal to f'strength of concrete
divided by the factor of safety As already explained, at low loads not causing cracking, the entire concrete is active in
-specified yield strength of
steel
resisting compressive as well as tensile stresses.: The concrete stress will be proportional to the
=allowable stress in steel, equal toj^
divided by the factor of safety
distance from the N.A. The strain in steel will also be equal to the adjoining concrete.
ft = lever arm between the resultant compressive
However, stress in steel will be higher than in concrete due to its higher modulus of elasticity.
j = ratio of lever arm to the and tensile forces in the The N.A. will be shifted from the geometric centroid of the section towards the tension steel
effective depth member
hi = depth of NA from the extreme compression
for equilibrium. The ratio between the stress in steel and stress in concrete can be determined
k = ratio of depth of NA to the effective depth force at any stage using the respective modulus of elasticity. In practice, it becomes more convenient
to convert the steel area into an equivalent assumed concrete area. The resulting section
consisting only of concrete but having equivalent properties of both the materials is called
transformed section.

2.4.1 Modular Ratio (n)


The ratio between the modulus of elasticity of reinforcing steel and the modulus of
elasticity of concrete is called modular ratio denotedby n.
n = Eft Ec
According to AC1 Code, this ratio is used as a nearest whole number for convenience,
but not less than 6. The difference of rounding is insignificant as the concrete modulus
Longitudinal View of Beam of elasticity is a variable quantity and only an estimate is used for its calculation.
Cross Section Modular ratio is less for concretes of higher strengths. For very high strength concretes,
Fig. 2.5. Basic Terminology the formula for Ec estimates a value of modulus of elasticity of concrete that may not
Used in Reinforced Beam Behavior. practically exist. Thus, the code has imposed the condition for a minimum value of 6
on the modular ratio.
FLEXURAL BEHAVIOR OF
40 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI BEAMS UNDER SERVICE LOADS 41
CHAPTER 2:
2.4-2 Uncracked Transformed Section
Of Rupture And Split Cylinder Strength
2 4.3 Modulus
For equal strain in steel and the surrounding concrete, the following may be written:
Modulus of rupture, denoted by/, is the tensile flexural stress ai which a concreteagainst
beam
cracking. Split cylinder strength iff) is the strength of a standard cylinder
starts
longitudinal splitting by the application of two longitudinal line loads at the opposite
fJE, = fc!Ec
sides. This is also an indication of the tensile strength of concrete. Mean split cylinder
. f* = "TT 7c = ^fc strength is approximately. 0.53^/77 MPa psi) and mean modulus of rupture is

This means that, before cracking of the concrete, the steel stress is equal to the stress
approximately 0.69^/77 MPa (8.3 psi).
in
the adjoining concrete multiplied with the modular ratio. This further means that if the
steel area is to be replaced with an equivalent concrete area, an additional concrete area
ACI uses modulus of rupture equal to O.62X-777MPa (7.5k psi) in calculating

equal to (n - 1) times the area of steel is to be used. After assuming such a deflections (ACI Eq. 9-10) and 0.50k Jf^ MPa (6k psi) indirectly in strength
replacement, the resultant tensile force becomes equal to that actually developed calculations. The parameter k is a modification factor for light weight concrete and its
together by the steel and the concrete. The resulting section is called
uncracked value is equal to 1.0 for normal weight concrete. According to ACI Commentary
transformed section. The fact that an additional area of (n - 1) A, is to be added to
get 10.2.5, the modulus of rupture is approximately 10 to 15 % of the concrete compressive
an equivalent concrete section may also be proved in an other way. It in the original strength.
section without transformation, Ac denotes the area of concrete alone at the level of
steel, Ag denotes the gross area of section at the level of steel including both
steel and 2.4.4 Tensile Strength Of Concrete
concrete and As denotes the area of steel, then we get:
The tensile strength of concrete has more variation than the corresponding compressive
Ag = Ac + As strength of concrete. It usually varies between 8 to 15 percent of the compressive
Total tensile force, T = fc Ac +fs A, strength. It is very difficult to measure the tensile strength directly due to difficulty of
— fc(Ag—AdA-nfcAs eccentricity of samples in the machine, their gripping and large variation due to voids.
= fc(Ag-As + nAs) Although actually no direct relationship exists between the tensile and compressive
= fc{Ag + {n~Y)As} strengths of concrete, the tensile strength may still be approximately related with the
Hence for transformed section, sn additional area of concrete equal to (n - 1)AS must compressive strength of the concrete. For design, the true safe tensile strength of
be added at the level of the actual steel reinforcement, as sown in Fig. 2.6. concrete is generally taken varying from 0.25^77 MPa (3 psi) to 0.42 .J77 MPa
(5^f^ psi) [Shear strength « 0.172^77)- However, experimentally the concrete
tensile strength is usually measured indirectly by modulus of rupture, split cylinder or
other suitable tests.
Example 2.1: A rectangular beam of size 250 x 650 mm, with effective depth equal to 590
mm, is reinforced with three No. 25 US customary bars, C 28 concrete and Grade 420 steel are
to be used. Determine the stresses at the top, bottom and level of reinforcement caused by a
a) Actual Cross Section b) Uncracked Transformed Section bending moment of 50 kN-m. The member is within its elastic range.
Fig. 2.6. Formation of Uncracked Transformed Section. Solution:
The transformed section may sometimes be conveniently used to calculate the sectional
properties, particularly the moment of inertia, of tire member comprising of two 1530 mm2 /,= 200,000 MPa
different materials. After obtaining the dimensions of transformed section, the usual
As = 3 x 510 = :
methods may be employed to find the location of centroid (or N.A.), moment of inertia fc' = 28MPa :
Ec = 4700^ « 24,870 MPa
and stresses. The neutral axis for such a section will lie below half depth of the beam. n = ES!EC = 8
Additional steel area = (n-l)A = 10,710 mm2
FLEXURAL BEHAVIOR OF
42 CONCRETE STRUCTURES PART -I AUTHOR ZAHID A. SIDDIQI BEAMS UNDER SERVICE LOADS 43
CHAPTER 2:
Referring to Fig. 2.7, following results may be obtained:
Solution:
„6 = 250 A, = 3 x 0.79 = 2.37 in2
, Jr — = 29,000 ksi
f.' = 4 ksi
e;- = 57777
o' = 590 (value of/c 'must be in psi)
h = 65G
= 3605 ksi
n — ESI Ec = 8
Additional steel area = (n - 1) As
= 16.59 in2
Fig. 2.7. Transformed Area for Example 2.1
Referring to Fig. 2.8, following results may be obtained:

(250)(650)(325) + (10,710)(590) (12)(24)(12) + (16.59)(21 5)


y 341 mm = = 12 52in
173,210 304.59
(12X247
I = + (12)(24X0.52)2 + (16.59)(8.98)2 = 15240 in4
(250)(650) 12
= 12
- + (250)(650)(16)2 + (10,710)(249)2 = 642,700 x 104 mm4 _ My 30x12000x12.52 = Compressive)
_ = 7S

-
(/c)top
(ft, = 50 x10® x 341
= 2.65 MPa (compressive) My
I
_ 15240
30x12000x11.48 _ 271.I 8 psi (tensile)
642,700 xlO4 Oc)bot
_ I
_ 15240

_
50 x 10® x 309
642,700 x IO4
= 2.40 MPa (tensile) f
Js n
My

fr for strength =
I—6777oX

= 380 psi
15240— ——
30x 12000x(21.5 -12.52) = 1697 psi

, My 50x10® x (590
v -341)
f* = = 8x- = 15.50 MPa
/ 642,700xl04 The bottom concrete stress in tension is lesser than its modulus of rupture indicating
that the section is really uncracked. Further, the stresses are much lesser than fc 7 2 and
/r for strength = 0.5777 = 2.65 MPa fy / 2 showing elastic behavior. Hence, the assumption of uncracked transformed
section is justified.
The bottom concrete stress in tension is lesser than its modulus of rupture indicating
that the section is really uncracked. Further, the stresses are much lesser than fc 7 2 and 2.4.5 Cracked Transformed Section, When
/ 2 showing elastic behavior. Hence, the assumption of uncracked transformed Concrete Stress Is Lesser Than /c 72
section is justified. : This situation occurs under the action of service loads when the tensile stress in
concrete exceeds the modulus of rupture. The condition that the section is cracked may
be checked by first considering the section to be uncracked and by determining the
Example 2.2: A rectangular beam of size 12 x 24 in, with effective
depth equal to 21.5 in, maximum tensile stress. If the tensile stress is greater than the modulus of rupture, the
is reinforced with three No. 8 US customary bars. The concrete has ultimate cylinder strength section will be cracked. The cracked concrete is neglected in calculations. The neutral
/c 'equal to 4000 psi. The yield strength of steel fy is 60,000 psi. Determine the stresses caused axis moves upwards due to this cracking and now it lies above the d/2 depth level. The
by a bending moment of 30 k-ft transformed area of steel in this case will be n As and this fictitious concrete area is in
tension. The strain, stress and resultant force diagrams due to flexure of a beam under
such conditions are shown in Fig. 2.9. The stress diagram is redrawn separately in
isometric view to make the calculation of forces more clear.
FLEXURAL BEHAVIOR OF
CONCRETE STRUCTURES PART -I BEAMS UNDER SERVICE LOADS 45
44 AUTHOR: ZAHID A. SIDDIQI CHAPTER 2:
root term because
Plus sign is considered in the above expression with the square
of deriving the same
way
otherwise the whole answer will be negative. Another equating the total
relationship is by using the strain diagram (Fig. 2.10) and
compression equal to the total tension as under:
E.
_ d-kd _ 1-k
E, kd k
1-k
Cracked Strain Stress Resultant k
Transformed Diagram Diagram
Section
Forces G -T d-kd
I

J Yifcbkd =
ViEuX Ecbkd = pbdx —— xecxEi
1-k
k
Fig. 2.10. Strain Diagram

Vik = np
1-k
k
k2 + 2npk~2np = 0

k = fipn)2 +2pn ~ pn
After calculating k,j can easily be calculated as follow:
Fig- 2.9. Internal Forces Produced Due to Flexure.

The depth of N.A. from the top may be calculated by taking first moment
of all the
jd =d
1
— - rf(i-^)
3
k
3
shaded area of transformed section and dividing it by the total area. The
centroid of the 3
section also lies along the same depth.
After locating the neutral axis and finding the lever arm, all the internal forces and
b(kd)(kd /2)+nAd resisting couple may easily be calculated.

b(kd)2 + nAshi
—-
bkd+nA,
b(kd^ / 2 + nAjd
b(kd)2 / 2 = nA^d kd)
The same result may also be obtained by equating the first moment
Total tensile force, T = A/s
Total compressive force, Cc = volume of compressive stress diagram
= bxL
fcxkd
Jc

2
'
'

of tension and
compression areas about the neutral axis.
. = ^xbkd
2
The steel area A* in the above relation may be replaced with pbd, Internal moment depending on steel strength = couple formed by T and its equal
ratio defined earlier. The above equation is simplified as follows:
where pis the steel and opposite force in concrete
bd\k2/2) = npbd2-npbd2k M
k2 + 2wpt— 2»p = 0 Mr = Asfsjd
AsJd
k = ^(pn)1+ 2pn - pn Where, Afis the applied moment and Mr is the resisting moment These two must be
equal for an economical design (M M^. -
FLEXURAL BEHAVIOR OF
46 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDD1Q[ BEAMS UNDER SERVICE LOADS 47
CHAPTER 2:
Internal moment depending on concrete strength = Ccjd Reinforcement In Beams
2.4.7 ACI Minimum
to be provided in the beams to avoid their
Mr ~x bkd x jd ACI requires certain minimum reinforcement
excessive cracking. The amount of this reinforcement is as
under:
2M
= ^-kjbd2 Igbd1
2.4.6 General Allowable Stress Design Procedure
A™ = ^^d ^bwd > (SI Units)

The trial depth of section is first selected on the (FPSUnits)


basis of deflection
architectural aspects. Effective depth and trial width of the section are control and
4,,min =
Jy Jy
The depth of N.A. is determined first in terms of then decided.
i-value by using the formulas SI units.
discussed in the following sections. The value of j may be
calculated easily The second part of formula is critical fork's 31.4 MPa for
calculated value of k. The concrete compressive strength is then indirectly from the Where, bv = width of the beam web
providing minimum depth for the singly reinforced sections. checked by
This minimum depth is Further, this steel is sufficient to prevent shrinkage and thermal
cracking in the beams.
calculated from the moment of resistance for concrete as follows: tension:
For a statically determinate part of a T-beam having its flange in
fef = allowable maximum stresses in concrete and steel, respectively
A™ = (SI Units)
2 24
yosf^-b Aa = 6~b„d (FPSUnits)

For k = 0.333,7 = 0.889, dmin I- - l~-


- - = y^b 2.4.8 Maximum Allowable Concrete And Steel Stresses
when using the
Every section has two strengths, one based on the concrete The following allowable stresses for concrete and steel will be used
in compression and the other Allowable Stress Design (ASD) method:
based on the tensile steel. For design, the concrete strength is always
made larger than
the steel strength so that if failure has to take place it should be by
yielding of the steel. Permissible extreme fiber stress in compression fc = 0.45 f/
In case concrete crushes before steel, a brittle and sudden failure
Permissible maximum stress for Grade 300 steel f = 0.5 fy - 150 MPa
without warning is (SI units)
obtained. The section having this type of collapse behavior is called
(SI units)
Permissible maximum stress for Grades 420 and 460 steels/ = 0.4/
over-reinforced
section, which is not allowed for design. Steel yielding produces
deformations before collapse and results in providing more ductility in the excessive fs for 10 mm or less diameter bars in one-way slabs
Greater concrete strength, compared with steel strength, may be provided by structure. of less than 3.5 m span = 0.5 fy
effective depth larger than dmm required when full concrete strength is selecting
the trial depth has to be increased if the corresponding effective utilized. Hence,
depth is lesser than In FPS units, the permissible steel stresses are as under:
dmin. The section having failure by yielding of steel is called under-reinforced section.
Permissible maximum stress for Grade 40 steel/ = 0.5 fy = 20,000 psi
The required area of steel is then calculated from the expression
resistance of the steel as follows:
for moment of Permissible maximum stress for Grade 60 steel/ = 0.4/, = 24,000 psi
/ for 3/8 in or less diameter bars in one-way slabs -
Mr =M~Asf,jd => As = of less than 12 ft span 0.5/
fs}d
The number and diameter of the required steel bars may then be
decided by using the 2.4.9 Design Values Of Factor k
tables of area for the reinforcing bars. steel area is not
The depth of neutral axis (kd) for design may be calculated, where of maximum
known in the start, by using the assumption of simultaneous occurrence
FLEXURAL BEHAVIOR OF
48 CONCRETE STRUCTURES PART - 1 BEAMS UNDER SERVICE LOADS 49
AUTHOR: ZAHID A. SIDDIQI CHAPTER 2:
permissible concrete and steel stresses or by using a maximum allowed steel
the first case, where the simultaneous occurrence of maximum stresses is
ratio. In selected. The value of k is then calculated using the expression developed
considered for earlier, relating the value of k with the steel ratio.
both file materials, under-reinforced behavior is ensured by the code. The
specifies suitable values for the allowable stresses considering the code
under-reinforced
behavior, which means that yielding of steel should occur before the crushing A™ -0.852x|i (SI and US Customary units)
concrete.
1. Simultaneous Occurrence Of Maximum Permissible Concrete And Steel
Stresses
of
k
- ^prif+2pn-pn
The values of the factor ‘P obtained by any of the above methods are
corresponding to the maximum possible depth of neutral axis and will predict
The strain diagram is used to calculate the required fc-value. It is assumed less than actual lever arm in deep and lightly reinforeed sections. Thus, the
the concrete reaches its maximum permissible strain (&) at the same time when
that calculated value of the factor ‘F may conservatively be used for both fully or
the steel reaches its permissible strain (&). A typical beam section and its
lightly reinforced sections.
corresponding strain diagram are shown in Fig. 2.11. KSromple 2.3: Determine flexural stresses for a cracked section beam whose data is given
stresses. Also calculate
below. Check these stresses against the maximum permissible ACI
moment of inertia of the cracked transformed section.
b = 250 mm
h = 650 mm
d = 590 mm
As = 1530 mm2
C 28 concrete
Grade 420 steel
M = 120kN-m
Fig. 2.11. Strain Diagram to Calculate k. Solution:
For similarity of triangles ABC and AED, we have, Example 2.1 shows that if uncracked section is considered for the moment given in this
example, the stresses will become twice that of the Example 2.1 exceeding the modulus
of
ec _
Id "
d-kd rupture. Cracked section must therefore be considered for this example.
Ec(d-kd) = e,(kd)
n = 8 as in Example 2.1
^-(d-kd) - P
1530
= 0.01037
(250)(590)
f^d-kd) = ^-(kd)
n
pn = 0.08298
kd{nfc+f^ = nfd k = ^(pn*1 ' 2pn- pn
k
-
This approach gives less factor of safety against brittle failure and hence is not a
= 7(0-08298)2 + 2 (0.08298) - (0.08298) = 0.333
= 1-1/3 = 0.889
better option for design. 120x10s 149.5 MPa
2. Using Already Selected Steel Ratio (1530)(0.889X590)
The maximum steel ratio permitted by the ACI-2002 code is given below. A 2M . (2)(120xl0s) _ = 932MPa
steel ratio (p ) equal to pkn or a fraction of p^s, but greater than is Jgbd2 (0.333)(0.889X250)(590)2
FLEXURAL BEHAVIOR OF
50 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI BEAMS UNDER SERVICE LOADS 51
CHAPTER 2:

OOponussible == 0-4 fy 168 MPa
^permissible = 0.457;' = 0.45 x 28 = 12.6 MPa
Ec
m
= 570007^ = 3,604,997 psi
= EJ Ec = 8
Hence both the steel and concrete stresses are within the ACI limits.
Additional steel area = n As = 18.96 in2
/r for strength = 0.5 Jff = 2.65 MPa 2.37 '

p = = 0.00919
As the concrete stress at the bottom of beam exceeds modulus of rupture, the cracked (12)(21.5)
transformed section is actually to be considered. The cracked transformed section
shown in Fig. 2.12. is pn = 0.07349
k = yj(pn)2 +2pn - pn
= 7(0-07349)’ +2 (0.07349) - (0.07349) = 0.317
= 1 -it/3 = 0.894
90x12000
= 23,708 psi
(2.37X0.894)(21.5)
2M (2X90x12000)
1 1
I = 1374psi
kjbd1 (0.317)(0.894)(12)(21.5)’
Fig. 2.12. Cracked Transformed Section For Example 2.3.
j GOpcnnissible = 0.4 fy = 24,000 psi
(/blpeimissibfc = 0.45^'= 0.45 x 4000 = 1800 psi
/ = 12
+ 250 x 196.5 x 98.252 + 12240 x 393.502 I
। Hence both the steel and concrete stresses are within the ACI limits.
= 252,800 x 104 mm4
J 777 - 380 psi
fi at bottom

„ = My~ 120xl06 x (650 -196.5)
I= -
252,800x10“
The section is actually cracked.
- = 21.53 MPa > rfr ?t
I
:
/ for strength = 6

As the concrete stress at the bottom of beam exceeds modulus of rupture, the cracked
transformed section is actually to be considered. The cracked transformed section is
shown in Fig. 2.13.
Example 2«4; Determine flexural stresses for a cracked
section beam whose data is given i
below. Check these stresses against the maximum permissible ACI stresses. Also I = + 12 x 6.8155 x 3.40782 + 18.96 x 14.6845
2
= 5354.8 in4
calculate ;
12
moment of inertia of the cracked transformed section.
*
b ~ 12 in
h =
I
24in j
d = 21.5 in i
A, = 2.37 in2 1
f' = 4000 psi
fv = 60,000 psi '
M = 90k-ft I
Solution;
Assume that the section is cracked and cracked section must therefore be considered for this
example. This assumption will be checked at the end.
I Fig- 2.13. Cracked Transformed Section for Example 2.4.
1

I
52 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI FLEXURAL BEHAVIOR OF
BEAMS UNDER SERVICE LOADS 53
CHAPTER 2:
Example 2.5: Design a rectangular section for a simply supported beam of 5m clear span,
subjected to a uniformly distributed load of 30 kN/m, using allowable stress design. Use C 30 permissible Stresses
concrete. Grade 420 steel and SI bars. Permissible 0.4/ = 168 MPa
=

Solution: permissible = 0.45//= 0.45x30 = 13.5 MPa


Check For Minimum Effective Depth
M= w x Zn2 / 8 = (30)(5)2 / (8) = 93.75 kN-m evaluated to ensure
The moment of resistance depending on the concrete strength is always the
For trial size of a beam, the total depth is normally taken closer to span/12 and width is taken under-reinforced behavior. This will ensure that concrete will not crush in any case and
approximately equal to depth / 3 for large beams to 2/3n) of depth for smaller beams. However^ collapse is always by yielding of steel giving lot of warning.
both must conform to the sizes of other components like columns and masonry walls.
Mr = M=
h = L /12 =5 x 1000/12 = 416.7 mm
Say the selected depth be 6 x 75 = 450 mm in terms of multiples of brick height.
b = 228 mm (in terms of brick length) = (13.5 / 2)(0.423X0.86) bf
4
fc'

A-75 = 375 mm
= 30MPa
= 2.456 bd2 for moment is in N-mm units

fy = 420 MPa dmtn = y ——2—


2.4566 :
for moment is in kN-m units
fc = 0.45// = 13.5 MPa
/ = 168 MPa
Ec = 4700 777
= 25,743 MPa
n = ESIEC = 8 Hence, the already selected depth is to be revised. Let the depth be slightly more than
A-Value For First Criterion this minimum and preferably in multiples of the brick height (75 mm).

k
-
=
l-k/3 = 0.870
j
. (8X13-5)
(8X1 3.5) + (168)
. 0.391
d
h
b
= 450 mm
= d+75 = 450 + 75 = 525mm
= 228 (length of one brick)
Calculation Of Reinforcement
k-Value For Second Criterion -
The required area of steel is calculated from the expression for moment of resistance of
3 f » the steel as follows:
= 0;8S^
8 Mr = M = A,fjd for M is in N-mm units
3 30 , AfxlO6 for Afis in kN-m units
= 0.85 x-x 0.85 x— = 0.01935 As =
8 420
pn =0.1548 93.75x10s = 1442 mm2
= (168)(0.86)(450)
'

k = J[pn)2 +2pn - pn •
Three # 25 (SI bars having 25 mm diameter) may be used giving .4, - 1500 mm2. The
= V(0.1548)2 + (2)(0.1548) -(0.1548) = 0.423 reinforcement details are shown .in Fig.2.14. A maximum of half of the steel may be
; curtailed at £a 1 20 distance from inner edge of the support (no curtailment according to
-
) =0.86 the ACI Code), where £„ is the clear distance between the supports. Similarly, if bent-
. '
"
,
This value is preferable and will be used in the present example. However, for practical

up bars are used, a maximum of half the bars may be bent up at a distance of ia!7 from
j
designs, any one value should be calculated and used for a particular design. the inner edge of the support.
''111i = 5000-228 = 4772 mm
FLEXURAL BEHAVIOR OF
BEAMS UNDER SERVICE LOADS 55
54 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 2:
h =L / 12 = 20 x 12/ 12 = 20 in. say 21 in,
£n / 20 ~ 235 mm # = 12 in.
Length of straight main bars « 5000 + 228 = 5228 mm d = A -2.5 = 18.5 in.
Length of curtailed bar = 4772 - 2 x 235 = 4302 mm = 2500 psi
Weight of tension steel = (2 x 5.228 +4.302) x 1.05 x 3.925 = 61 kg fc
fy = 40,000 psi
(including 5% wastage) f. = 0.45 fc' = 1125 psi
f = 24,000 psi
228 Ec = 57,000 777 = 2.85 x 106 psi
n = A / A = 29.0 x 106 / 2.85 x IO6 10 =
A-Value For Second Criterion

r max = 0.85x—Q
3
'3
”1

f
-f

2500
0.85 x-- = 0.01693
= 0.85 x-x
8 40,000
pn =0.1693
k - f(Pn)2 +2 pn - pn
= 7(0,1815)’ + (2)(0.1815) - (0.1815) = 0.489
Longitudinal Section j = 0.837
Check For Minimum Effective Depth
Fig. 2.14. Reinforcement Details for Beam of Example 2.5. -M =
Mr
AC1 requires certain minimum reinforcement to be provided in the beams to avoid their
excessive cracking. The amount of this reinforcement is as under: = (H25 / 2X0.437)(0.854)&/
= 210 b<f Moment is in in-lbs units
14 14
Amin = ~bd =^(228X450) = 342 mm2 lbfxl2,000 Moment is in k-ft units
d„m =
The provided steel is more than this amount, hence the condition is satisfied. V 2106
p = 1500/ (228)(450) = 0.0146 < /w =0.01935 OK = 21.8 in.
Example 2.6 (FPS Units): Design a rectangular section for a simply supported beam of > 210x12
depth be slightly more than
20 ft span subjected to a uniformly distributed service load of 2.0 k/ft using allowable stress Hence, the already selected depth is to be revised. Let the
design. fc' = 2500 psi and = 40,000 psi. this minimum and preferably in multiples of the brick
height (3 in.).
Solution: h = 27 in.
3f=(2.0)(20)2/(8) = lOOk-ft d
b
= 27-2.375 24.6 in.
= 12 in.
=
The total depth may be taken closer to span/12 while the width may be taken approximately
equal to depth/3 to depth/4.
56 CONCRETE STRUCTURES PART -I FLEXURAL BEHAVIOR OF
AUTHOR: ZAHID A. SIDDIQI
CHAPTER 2: BEAMS UNDER SERVICE LOADS 57
Calculation Of Reinforcement
The required area of steel is calculated
2.5 REDUCED MOMENT OF INERTIA DUE TO CRACKING
the steel as follows:
from the expression for moment of resistance
of The moment of inertia after cracking varies all along the length of the member and depends on
the number of cracks, height and angle of cracks and the crack width. In place of
Mr
A,
_
~ M = Asfsjd
Mx 12,000
M is in in-lbs units

M is in k-ft units
comprehensive analysis, ACI 9.5.2.3 allows the calculation of immediate deflections based on
the effective moment of inertia.
fsJd
4
100*12,000 _ 2‘38in
~
(24,000)(0.854)(24j5) Where,
A steel of 2 #8 + 2 #6 may be used
giving^, = 2.46 in2. The reinforcement
shown in Fig.2.15. ACI requires certain details are
minimum reinforcement to be provided in the
beams to avoid their excessive cracking. The
Ma = cracking moment =
y, —
fr1.

amount of this reinforcement is as under: = modulus of concrete for deflections


^m- = ™bd = 0.42^ (SI) 7.5777 (US Customary units)
_yt = distance of extreme fiber in tension from the centroidal axis of gross section,
neglecting the reinforcement.
fg = moment of inertia of gross section about the centroidal axis, neglecting the
reinforcement.
The provided steel is more than this amount,
hence the condition is satisfied. = moment of inertia of cracked section transformed to concrete, usually
p = 2.46 /(12)(24.6) = 0.0083 < =0.01693
cracked section at the service load stage and not at the ultimate stage.
OK Mi = maximum moment in the member at the stage at which deflection is to be
calculated.
Hanger Bars 2.6 LONG - TERM DEFLECTIONS
Cross Section Stirrups To Be Designed Later i
In place of more exact calculations, additional long - term deflection caused by sustained load
2- #8 + 2- #6 and due to creep and shrinkage of flexural members is calculated by multiplying the immediate
I deflection due to this sustained load by the factor 2a defined as under:

I l + 50p'
Where p' is the compression reinforcement ratio, calculated at midspan for simple and
continuous spans and at support for cantilevers and ^is a time dependent parameter having the
following values:

Elapsed Time £
5 years or more 2.0
Longitudinal Section 12 months 1.4
6 months 1.2
Fig. 2.15. Reinforcement Details for Beam 3 months 1.0
of Example 2.6.
The immediate deflections may be calculated by using lt and then the effect of long term
deflections may be added by multiplying with the result with (1 + 2a).
58 CONCRETE STRUCTURES PART - 1 FLEXURAL BEHAVIOR OF
AUTHOR: ZAHID A. SIDDIqi 2: BEAMS UNDER SER VICE LOADS 39
Space For Reader’s Note. CHAPTER
PROBLEMS

Note: N = Registered number of the student.


No = A fixed positive or negative number assigned by
j the instructor to get R between 600 and 1000.
R = N - No, a parameter to get numerical data for the
problems.

1- A rectangular beam of size 300 x 650 mm, with effective depth equal to 590 mm, is
reinforced with four No. 25 US customary bars.. The concrete has ultimate cylinder
strength (fc') equal to R / 33 MPa. The yield strength of steel (fy is 420 MPa.
Determine the stresses at the top, bottom and level of reinforcement caused by a
bending moment of R 1 20 kN-m. The member is within its elastic range.

2- A rectangular beam of size 300 x 650 mm, with effective depth equal to 590 mm, is .
reinforced with four No. 25 US customary bars. The concrete has ultimate cylinder
strength (fc } equal to R I 33 MPa. The yield strength of steel is 420 MPa.
Determine the stresses at the top, bottom and level of reinforcement caused by a
bending moment of J? 1 8 kN-m. Assume the member to be within its elastic range.

3- Determine flexural stresses for a cracked section beam whose data is given below.
Check these stresses against the maximum permissible ACT stresses. Also calculate
moment of inertia of the cracked transformed section.

b = R / 2.5 rounded to nearest 25 mm multiple


h = R mm
d = A - 75 mm
As = 1530 mm2
fc' = J?/ 33 MPa
fy = 420 MPa
M = ^/6kN-m
4- Design a rectangular section for a simply supported beam of R I 150 m span subjected

to a uniformly distributed load of R 1 20 kN/m using allowable stress design. fc' RI
33 MPa and/, = 280 MPa.
60 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI Chapter - 3
Space For Reader’s
Notes
DESIGN OF BEAMS
3,1 INTRODUCTION
In Chapter 2, the flexural behavior of members is considered under service loads. In such
cases, the steel and concrete behave almost elastically. However, close to the ultimate, the
steel may start yielding while the concrete definitely becomes highly inelastic. In other word,
the stresses do not remain proportional to the strains. The shape of the stress diagram is
significantly changed and the NA. is further shifted up. The concrete tensile strength is still
neglected and hence all the concrete below the N.A. becomes ineffective. The structural
pngineer is interested in predicting the ultimate strength of the structural members with
reasonable accuracy. To achieve this goal, the materials are to be studied in their inelastic
range and formulations are to be developed, with some simplifying assumptions, to calculate
the ultimate fl exural strength of the members.

3.2 STRESS-STRAIN CURVE FOR CONCRETE

A typical stress-strain curve for concrete, when tested in a compression-testing machine, is


shown in Fig. 3.1. The stress and strain are approximately proportional to each other up to
about half of the ultimate strength. At higher stresses, the behavior is considerably inelastic.
The final breaking or crushing takes place at a stress level lesser than the maximum/ultimate
stress, which is called the breaking stress. This value is approximately 15% lesser than the
ultimate strength. The strain at which Crushing of concrete takes place is 0.0025 for high-
strength concretes to more than 0.0045 for low-strength concretes. For design, a conservative
value of 0.003 is considered as the ultimate strain, for all concretes except very high strength
ones. The area under the curve divided by this ultimate strain gives the average value of the
stress. The average stress is 0.72/ 'for concretes up to ultimate strength of 28 MPa (4000 psi).

_
The factor /indicates the ratio of breaking strength to the ultimate strength.

/ =
Breaking stress
Ultimate Stress —
7
(/')
r = 0.85
,
(average value)x
or r

The ratio of average to crushing stress of concrete is denoted by /?i and is found for concrete of
fa 28 MP,a as under:

fa =
f<nsi
= °'0.85/
72^ = 0.85 (for/' 28 MPa)

The concrete on the compression side of a flexural member is also subjected to compressive
stresses and strains. The related stress-strain diagram within the member must be known in
order to determine its resistance against the applied loads. It may be noted in abeam that away
from the neutral axis, the strain is linearly increasing just as in the compression-testing
machine. Hence, the stress-strain behavior in such a member must be identical to that obtained
in a separate compression test The stress-strain curve of Fig. 3.1 is developed within the
62 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
CHAPTERS: DESIGN OF BEAMS 63
member with the difference that the diagram is rotated counterclockwise through 90° until
its 3 3 FLEXURAL STRENGTH
strain axis becomes vertical. Further, the strain axis is replaced by the depth of section as it is
directly proportional to the strain. This resulting diagram is shown in Fig. 3.2, considering
As pointed out earlier, close to the ultimate, the stress diagram resembles the stress-strain
strain on the vertical axis and stress on the horizontal axis. diagram in a compression test. Away from the N.A., the strain is linearly increasing exactly as
Ilie actual parabolic variation of concrete stress is approximated by a rectangular stress
I in the testing machine (refer to ultimate stress diagram of Fig. 3.3). Only difference in the
distribution in.beams to make the calculations simple. The uniform stress is taken equal to testing machine is that the strain increases with time in a machine and it increases with its
0.854 'and the distance along the depth of beam is taken equal to a, which is lesser than the | distance from the neutral axis in the flexural member. The neutral axis further moves upward
depth of N.A. denoted by c. This is done to approximate the position of the resultant
I
due to the propagation of cracks towards the top edge.
compressive force in the actual stress-strain curve with this idealized curve. The value of this j
dimension, a, is found such that the magnitude of the resultant force remains same in both
actual and idealized curves. Assuming the width of beam on which the concrete stress acts to
I
be b, we get, J
j
Jav x & x c (actual curve) = x fix a (idealized curve) O
X f
a —~ c - P\c (equal to 0.85 cfor^' < 28 MPa)
J crush <

Steel Elastic Stress Strain Stress Diagram Resultant


Diagram Diagram At Ultimate Forces

Fig. 3.2. Concrete Stress Distribution at Ultimate Load.

c = depth of N.A. from the top at the ultimate stage


a = depth of rectangular stress block from the top

It is difficult to find area and centroid of the actual stress strain diagram and hence it is
assumed to be replaced with an equivalent fictitious stress block. This stress block is called
Whitney’s equivalent rectangular stress block and is shown in Fig. 3.4. Depth of equivalent
stress block, a, is so adjusted that the total compression force Ce remains same as in the actual
situation. Further the point of application of this compression resultant is not significantly
Fig. 3.1. Experimental Stress-Strain Curve for Concrete. changed from its true location. Considering stress equal to fm over full depth up to the N.A. for
the assumed stress block is no good because of two reasons. Firstly, the centroid will not be
close to the centroid of the actual diagram. Secondly, the top stress will be wrongly
represented by 0.724' at the ultimate strain, which must be equal to the breaking stress
(0.854').

For actual parabolic stress block, Cc = fmbc - Q.^fc'bc


For equivalent rectangular stress block, Cc - 0.854' ba
For both of these to be equal, 0.854 = O.Hfc'ba

a=
a=
——
0.72
0.85
flic
c => a = 0.85 c
or fli = ale
Fig. 3.2. Concrete Stress Block in a Flexural Member.
QIAPTER3: DESIGN OF BEAMS 65
crushed, we
At the above final stage when the steel is yielding and the concrete is about to be
jhay calculate the depth of the N.A. from the compression face (c) and the depth of the
rectangular stress block (a) as follows:

T = Cc => Asfy = 0.85/'b a


AJ f
a= —
0X5f’cb
and c - al Pi
The lever arm (la) is then equal to d-al 2 and the resisting moment (A/) may be calculated by
multiplying the resisting force Tor Cc with this lever arm.
Mr = Tla = A^fy (d-al 2) for steel strength
-
and M, = Ccla 035fc'b a(d-a! 2) for concrete strength
This Mr is denoted by M and is called the nominal flexural strength of the section in the
Strength Design or LRFD.
Pi —'
ratio of depth of equivalent rectangular stress block to the depth of N.A.
0-85 ; for fc' < 28 MPa (4000 psi) 3.5 MINIMUM DEPTH OF NON-PRESTRESSED
(decreases by 0.05 for every 7 MPa in excess of 28 MPa)
BEAMS AND ONE-WAY SLABS

= 1.05-0.00714 /' 0.85 for /'> 28 MPa (SI Units) In order to keep the deflections within limits, we may check the expected deflections by using
= 1.05 - 0.00005// 0.85 for /'> 4000 psi (FPS Units) the code formulas, However, this is usually quite complicated process due to the unknown
Note: Maximum value is 0.85 and minimum value is 0.65 in the above formulas. pattern of cracking and the corresponding change in the moment of inertia. Further, if the
deflections are more than allowable, we have to repeat the design for the second time. To
3.4 DEPTH OF RECTANGULAR STRESS BLOCK avoid such repetitions, the depth of flexural members may be kept larger than or equal to the
WHEN TENSION STEEL IS YIELDING Code specified minimum depths, which ensure that the deflection check will be satisfied in
majority of the cases. Table 3.1 provides the ACI Code minimum depths for non-prestressed
beams and one-way slabs, when normal weight concrete is employed for the construction.
When steel is yielding at failure, the depth of the rectangular stress block depends upon the
amount of reinforcement. The yield strain in steel for various grades is as under: The required minimum depth is more for 420-grade steel than 280-grade (or 300 grade) steel
because, at the ultimate load, the steel will be yielding. The yield strain is higher for higher-
% = // £, = 300 / 200,000 = 0.0015 for Grade 300 steel grade steel meaning more rotation of the critical sections associated with more deflections.
Sy =fy!Es~ 280 /200,000 = 0.0014 for Grade 300 steel The depth is increased in such cases to reduce the rotation angle for the high yield strain at the
(Grade 40 in US Customary Units) critical sections.
Sy - fyl Ei = 420/200,000 = 0.0021 for Grade 420 steel
(Grade 60 in US Customary Units) 3.6 MAXIMUM PERMISSIBLE COMPUTED DEFLECTIONS
Sy = fy I Es = 460 / 200,000 = 0.0023 for Grade 460 steel
Sy = fy I Es = 520/200,000 = 0.0026 for Grade 520 steel According to ACI Table 9.5(b), for flat roofs not supporting or attached to noh-structural
(Grade 75 in US Customary Units) elements likely to be damaged by large deflections, the immediate deflection due to live load
As soon as the steel reaches this strain, the surrounding concrete cracks to a very large extent should not exceed 1 1 180. However, this deflection must not cause ponding. For intermediate
and the N.A. moves up. The yielding continues at a constant stress level, the strain and floors with the same conditions, the value must be 1 1360.
cracking still increases, causing the N.A. to shift up by greater amounts. The final failure
occurs when the concrete strain at the top reaches a value of 0.003. At this strain the concrete 3.7 BALANCED STEEL RATIO
crushes and the member ceases to resist any bending moment.
Steel ratio is defined as the ratio of the total area of tension steel and the effective area of the
section. Balanced failure is a special condition at ultimate loads in which the yielding in
PART " 1 AUTHOR: ZAHID A. SIDDIQI
tE STRUCTURES 67
DESIGN OF BEAMS

e crushing of concrete starts simultaneously. In other words, in a balanced


,
stee' an reaches yield strain (^) (or steel stress,^, reaches yield stress, fy
teh5’ . the stee s . 1 reaches 0.003 at the same time.
0

,«efibersW

^enc®’
a and Sc — 0.003 Fig. 3.5. Isometric View of
Force Diagram.
Table 3.1. Minimum Depth of Beams and One-Way Slabs

.| Grade Simply One End Both Ends Cantilever


considered.
Supported Continuous Continuous To evaluate the ratio add, the strain diagram at ultimate condition of Fig. 3.6 is

or 300 £/25 £130 £/35 £/12


colid One' 420 £/20 . £/24 £/28 : £ /10

460 £/\i.9 £/22.7 < J /265 ,


' V/^f s ?

) or 300 £/20 £726 £/16


420 £H6 £/18.5 £/21 £/8
pea”15 Steel
460 £l\5.\ £/17.5 £/19.9 £/7.6

520 £/14 £ /16 £ /18.4 £/7 Fig. 3.6. Ultimate Strain Distribution at Balanced Conditions.

^g effective span of the beam or the slab. For similarity of triangles ABC and CDE, we have,
^otes; 1' structural lightweight concretes having unit weights in the range 1440 - 1840 BC DC
2. ^3
k6Q9
values shall be multiplied by (1.’65 - 0.0003 wr) but not less than
above
weight of concrete in kgs/m3.
AB DE
where is the unit cb d-c.

other than for /y = 420 are obtained by multiplying the values for = 420 0.003

Cb 0.003 d - 0.003 cb
E,
He and compressive forces at a cross section at ultimate condition are shown
result te [jgiing this figure and the rectangular stress block shown earlier, the values (fy + 0.003) = 0.003 d
Cb{~E-
Fig f rces may be calculated. In
the following expressions, the subscript b refers to

of^Se balanced
theVaUe f=0.85//^
condition.
Cb = -—
0.003
d -
0.003 E,
d (ID

_ fy/E,+0.003 fy +0.003 E,
(Pbbd)fy
f = Asfy ~
For Et = 200,000 MPa, the value of cb becomes:
T 0.85//J ab = Pb b dfy 600 ,
Cb = —fy+SM
— d (SI Units) QU)
chapter. 3;
DESIGN OF BEAMS 69
68 CONCRETE STRUCTURES PART - 1 AUTHOR- ZAHID A. SIDDIQI I d, —c
J 0.003 -4 0.005
Cb = —4+87,000 d
87,000
(FPS
V Units) {
;
0.003
c
0.005 + 0.003
Note: The value of 600 ( or 87,000 in FPS units) in the above expression is the stress in a c
'
fictitious high strength steel corresponding to a strain of 0.003.
It is already known that the depth of neutral axis at ultimate conditions is related with the depth >
of the equivalent rectangular block as follows: ; The net tensile strain (fit ) is the tensile strain in the extreme tension steel at the nominal
strength not including the strains due to prestress, creep, shrinkage and temperature and
ai> = Pi Cb (IV) | its limit of 0.005 for tension-controlled sections is same for all types of steel
Using equations ID and IV, the depth of the rectangular stress block becomes, (prestressed and non-prestressed). The flexural members are usually tension-controlled
j
and the compression members are usually compression-controlled.
ab = d (SI UmtS^ W '
Jy + oUU
87,000
3.8.2 Compression-Controlled Sections
ab = Pi - d (FPS
1 Units) These are the sections where the net tensile strain in the extreme tension steel is less
4+87,000
than or equal to its yield strain (£0, when the corresponding concrete strain in
Plugging the value of add from Eq. V into Eq. I, the following result may be obtained: ; compression just reaches a strain of 0.003. The compression-controlled strain limit is
the net tensile strain in the reinforcement at the balanced condition. The value of
(SIUniB)
- ;
may be taken equal to 0.002 for Grade-420 and all prestressing reinforcement.

0.003
d, —c
-4
Pb = 0.85 Pi 8^° (FPS Units) < 1 c . .

4 4+87,000 < s ?
i
3.8 c 0.003
TYPES OF SECTIONS DEPENDING ON FLEXURAL BEHAVIOR
J
The sections may be classified into three types depending upon how much ductility is provided |
d
-4. < t
c —
£,+0.003
0.003
or —d, c
—sy
0.003
+0.003
- —
by them. The strength reduction factor is accordingly different for these types of sections. ;

3.8.3 Transition Sections


3.8. 1 Tension-Controlled Sections
When the net tensile strain in the extreme tension steel is between the limiting values
These are the sections at which the net tensile strain in the extreme tension steel is for the compression-controlled and the tension-controlled sections, the section behaves
equal to or greater than 0.005, when the corresponding concrete strain in compression as a transition between the two main types of sections. The ^-factor is usually linearly
just reaches a strain of 0.003. varied for such sections for a smooth transition from compression-controlled to tension-
Let dt = depth of steel closest to tension controlled sections.
face from the compression face, j

and 4 - strain in steel closest to the 3.9 STRENGTH REDUCTION FACTORS |


tension face, on outer side of bar.
et _ d, —c The strength reduction factor (also called resistance factor) is reciprocal of the minor part of the
0.003 c~~ overall factor of safety that is applied on the strength of a member to obtain its design strength.
This part of factor of safety covers the following aspects:
If this strain is greater than or equal to 0.005,
the section is tension controlled. ! The 1. It covers the probability of under-strength due to variations in material strengths and
Ultimate Strain
following simplification may be performed dimensions.
Diagram
to get other useful results: 2. It covers minor discrepancies in the design equations.
Fig. 3.7. Ultimate Strain Distribution.
yp
STRUCTURES PART - 1
AU7Ho* 2Ahid a

3 It
depends upon the degree of ductilitv a CHAPTERS: DESIGN OF BEAMS 71
depends upon the importance
4 It dso
ACI Code gives the following
ln

^cture.
ofsecdons:
i 3.9.7 Development Length
= 1.00
3g,l Tension-Controlled
Sections
3.9.8 Bearing On Concrete
= 0.90 = 0.65
impression-Controlled Sections 3.10 MAXIMUM STEEL RATIO
3.92 spiral reinforcement
a) Members with 0.75
bj Other reinforced
members 0- =0.65 To make sure that every flexural member fails by yielding of steel giving sufficient warning
before failure, ACI 10.3.5 requires that, if the axial load on a non-prestressed member is less
Sections than 0.10/c'Ag, the net tensile strain ft at nominal strength should not be less than 0.004.
3 transition
linearly increased from compression-controlled
The ^factor is Rvalue to 0.90 as the > 0.004
‘tion changes from compression-controlled to
tension-controlled section. As already
fit
d, —c
mentioned, the value of may be taken equal to 0.002
for Grade-420 and all 0.003 -t 0.004
prestressing reinforcement 4
aj Members with spirals = 0.75 + - Q

0.005-f, ' '
c
d, 7
3
2 a

— or „3
or
= 0.75 + ---
0.005-ay[
0.003) 11-sv
.
c 3 7
F or s ingiy reinforced beams with one layer of steel reinforcement,
dt d.
The value must be between 0.75 and C =T
0.90.
y Members with ties = 0.65 +
= 0.65+ -
— -r/
0.005
[>, -fj
'
0.003|^—
0.85/c' b a
P - 0.85 -
= pbdfy

0.005-4,[ Vc ) 'j 0.85 AxAx2


The value must be between 0.65 A 7
and 0.90.
Torsion
j 4 Shear And Pmax = 0.364 px
= 0.75
It is to be noted that, for doubly reinforced and T beams, the effects of compression
Concrete
j9 5 Bearing On reinforcement and flanges are automatically considered during the computation of the net
tensile strain through the value of the depth of neutral axis (c).
For design, it is better to keep the minimum strain equal to 0.005 to
Tie Models give some margin against
3 9,6 Strut-And
the possibility of being below the strain limit of 0.004 while
the steel bars are
= 0.75 for strut-and-tie models, struts, ties nodal31 201165 due to reduced factor of safety at a strain of 0.004, no considerable economy selected. Also,
X
F zones and
311(1 k
bearing areas in maximum steel ratio for singly reinforced sections with the limiting strain of is obtained. The
such models. 0.005 may be
found as under:
ft 0.005
72 CONCRETE STRUCTURES PART - 1
AUTHOR: Z AHID A. SIDDIqj DESIGN OF BEAMS 73
CHAPTER 3:
0.003 - 0.005 3,11 ACI MINIMUM REINFORCEMENT

Another brittle type of failure in addition to the over-reinforced one, without any warning, may

——
c
8
> T
3
or
— <
c3 _ a
A* 8
3
occur in very lightly reinforced beams.
tlian the bending moment
immediately
that
upon
produced
formation of
If the flexural strength of the cracked section is less
cracking
the first
of the previously uncracked section, the beam
crack. This type, of behavior is undesirable.
d, 8 will fail
For singly reinforced beams with one layer

p
Ofsted Kjnfateteent.4
0.85 A x4A
- d.
To ensure against this

the cracked section.


type of failure, a lower limit may be set for the steel ratio by equating the
corking moment, computed from the concrete modulus of rupture, to the flexural strength of

o x ACI 10.5 requires certain minimum reinforcement to be provided in the beams to avoid their
excessive cracking. The amount of this reinforcement is as under:
Ptnsct = 0.318 A
(SI Units)
Similarly, the maximum steel ratio for
singly
0.0075 for redistribution of moments (ACI 8.4.2)reinforced sections with the limiting strain of The second expression is critical when// 31.4 MPa.
'
may be found as under;
£t 0.0075
0.003 > 0.0075
^s,nun = ^&bvd
fy ^b„d
fy
(US Customary Units)

This condition of minimum steel may be waived off if the area of reinforcement provided at
every section, positive or negative, is at least one-third greater than that required by analysis.
In T-beams, the ratio p to be checked for minimum steel ratio is to be computed using the
width of the web (p = As I bvd). This value of ACI minimum steel ratio is very conservative
for rectangular beams but is also fractionally conservative for the T-beams. For statically
determinate members with a flange in tension, the area A^n is to be calculated by using
P = 0.85 - smaller of and width of flange in place of A, in the above expressions.
4 d
For roof slabs and foundation slabs, a minimum area of steel expressed as the following ratios
≤0.85Ax-x^.
7 f of the reinforcement area to gross concrete area, should be provided. This steel provides 1.1 to
Jy
1.5 times the cracking moment in slabs. The minimum steel ratio on the basis of the gross
P^ = 0.243 A area may be converted to p^ based on the effective area of concrete by multiplying it with hid.
This minimum steel is also to be provided for temperature and shrinkage control and to
According to section 8.4.2 of the ACI Code, distribute the loads uniformly over the perpendicular reinforcement.
allowed if the extreme tensile strain is equal redistribution of negative moments is only a) Slabs where Grade 300 or any other deformed
to or greater than 0.0075 at the section 0.0020
moment is reduced. Usually this section is where the steel bars less than Grade 420 are used
doubly reinforced and the value of above
(applicable only for singly reinforced sections) steel ratio b) Slabs where Grade 420 deformed
changes.
The same effect was covered in the ACI bars or welded wire fabric are used 0.0018
the steel ratio should not exceed
1999 and earlier codes by specifying the
75% of the balanced steel ratio for design. condition that c) Slabs where Grade 460 deformed
under-reinforcea behavior in all the flexural members. This ensured 0.0017
bars or welded wire fabric are used
P™ = 0.75a d) Slabs where reinforcement with yield strength exceeding
The portion of steel 0.0018x420
. by the factor of 0.75.ratio A equalized by the compression reinforcement was not to be redubed 420 MPa measured at a yield strain of 0.35% is used
4
> 0.0014
DESIGN OF BEAMS 75
^STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI cHAPTER3:
As the N.A. moves up and more cracking occurs, the area of concrete in
OF FAILURES
compression reduces. This causes a rise in the concrete compressive stresses.
<1 I2
3’ When compressive stress in concrete reaches its ultimate value, concrete
aijsed by crushing of the concrete, with the steel stresses less than its yield
crushes at the top and the flexural capacity is dropped. Hence, the final failure,
failf16 ned compression failure or over-reinforced failure and the resulting sections are even in case of under-reinforced members is by the crushing of concrete.
^ugih, ^^pressiort'Controlled
0
sections. The failure caused by the yielding of steel before
However, this failure is accompanied by large deformations giving warning.
of concrete is called under-reinforced failure. The resulting section is called
section, which may further be classified into either a tension-controlled The failure that is initiated by the yielding of steel is known as tension failure and the
^^ef-t^^^gnsilion
°r a
section depending on the steel strains at the failure. section is called under-reinforced section. As already stated, this type of section may
further be divided into a tension-controlled section or a transition section. The capacity
Or Tension Failure of such a member is derived below:
3-12* .^ernal couple developed by the resultant compressive stresses in concrete and the Maximum strain in concrete = 0.003
A° steel tension resists the applied bending moment. Flexural strength is defined Extreme fiber concrete stress = 0.85/'
juaximum moment of resistance that can be developed in a member. Stress in steel = fy
35
= (Cor7)x/a Strain in steel —e> %
= fyl&
applied load is increased on an under-reinforced beam in which the final failure Refering to Fig. 3.9, the following results are
As .gj by yielding of the steel, the resulting bending moment is resisted in three
stages, shown in Fig. 3.8 and explained below. obtained:

Lever aim is defined as the distance between the resultant tensile and C = T => 0.85/'6a = Atfy
L compress^6 forces, which remains nearly constant and only the compressive
gnd tensile forces increase. This is applicable when the section is elastic and
uncracked. . The nominal moment capacity of a beam
failing in tension can only be determined from
the steel strength, because it is lesser, and not
from the concrete strength. The resisting;
internal couple will be equal to the resultant Fig. 3.9. Stress and Internal Force
tensile force, when the steel is having yield Diagrams for a Beam.
stress, multiplied with the corresponding lever
arm.
Stage 1: Un cracked Stage 2: Cra eked Stage 3: Ultim ate M„ = Asfy(d-al2} (ii)
Section Section Condition
The value of a from Eq. (i) may be put in Eq. (ii) to get the following:
Figure 3.8. Various Stages of Loading an Under Reinforced Beam.
A,f> (iii)
4s the bottom concrete starts cracking, the depth of N.A. from the compression 2x0.85/' 6
fiber starts decreasing and hence the lever arm increases. At the same time, the
compressive force in concrete and tensile force in the steel also increase. Both Knowing that As = p bd, Eq. Ill becomes:
ofthese phenomena increase the moment of resistance of the section.
This third stage starts after the yielding of steel in tension. The tensile force in
M„ = pbdfy \d-~—
< 'JcO )
I
(iv)

steel (T) thus becomes constant in magnitude. The resultant compressive force
in the concrete (C) has to remain equal in magnitude to the tensile force,
however, due to more and more cracking, the N.A. is shifted upward and the
Mn = pbd2fy[\-^
K 7^ J
(V)

|eVer arm increases. Hence, moment of resistance increases in this stage only by
increase of the lever arm.
S1B.UC TURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
70 CHAPTER 3: DESIGN OF BEAMS 77
rfect and most economical design, Mv = ^Mn
ForaP60^ 21?
.
capacity reduction or resistance factor, and its value for flexural
Where, .
concrete member is 0.9, where the extreme tensile steel strain isresistance II 0-765/;,
ofre^n than 0.005.
orgreater
equal to

applicable when 0.005 (ix)


1-^ (vi)
The minimum depth of rectangular section for under-reinforced conditions may be
0^ R= determined by calculating the moment of resistance from the concrete side when s =
then,
Let & "
/ bd1 0.005. The same relationship is applicable for both SI and US Customary units and is
also approximately valid for si = 0.004.
0.9pM7Jl-^- - —83 pxd, where # = 0-85
^ Wfy^P-P1)
=
\ IO J At si > 0.005, a fox fc'<28 MPa.

2S)R~ Mu = ^Mn = 0.9 x 0.85/c'i a


j. 2<oR 085/;
l&p-P = x
0.94 0.85/c' = 0.765/ '6 (0.375# d)
2 2 ~ 0.205 fc'bd2
2<Dp"P - x
0.9x0.85
(X)
^R
0.765 x Jf'c
0 “1^205^
If the effective depth of beam is selected greater than or equal to this the behavior
of beam will be that of under-reinforced section and p will be lesser than
2®± JW - —0.765// '
3.12.2 Over-Reinforced Or Compression Failure
p^.

P 2
The first two stages produced by increasing the load are exactly the same as in case of
2R the tension failure described above. However, in the third stage, concrete being weaker
= ® 1+1
1± 1- (vii) than the steel crushes first before the steel reaches its yielding point. The concrete
I V 0.765/;
crushing is assumed to start when the extreme fiber concrete strain reaches a value of
0.003. The stress in steel, /, remain lesser than its yield stress, /. The concrete
'11 give answer more than 1.0® if positive sign is used for the radical quantity,
maximum steel ratio is much lesser than this limit To prove this, the crushing occurs suddenly and chunks of concrete in the maximum compression region
H^stepsmaybecarriedout: are blown off as the load just exceeds the ultimate load capacity.- Hence the over-
reinforced or compression failure is a sudden failure without any warning. A beam
P should never be designed as over-reinforced. However, for the analysis of an accidental
n | 600 ) over-reinforced beam, the capacity may be determined by the formulas derived here.
[600+/J (viii)
Extreme fiber concrete strain, £c — 0.003

Awtnr
The fector
A is I®8 one and the factor
— —
^qo +j
is also less than one. This Extreme fiber concrete stress
/ </ and £ = %
= 0.85/

“d »« be ln Bq Because the steel stress is unknown at the ultimate conditions, it is to be calculated


from the strain distribution diagram of Fig. 3.10.
78 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTERS: DESIGN OF BEAMS 79
The only unknown in this equation is ‘a’ and the quadratic equation may be solved to
find the value of this depth of equivalent rectangular stress block. The resisting
moment capacity and the design flexural capacity may be determined as follows:
Mn = Mrc = Ccxla
= 0.85 f'b a (d-ah) (viii)
Mu = ^>Mn (ix)
The factor in case of over-reinforced sections will be 0.65. It is to be noted that these
formulas are only used for analysis of some wrongly designed beams. Otherwise, the
Steel Cross Ultimate Strain design of beams is not allowed as over-reinforced section by the ACI Code.
Section Diagram
3.12.3 Balanced Failure
Fig. 3.10. Ultimate Strain Distribution for Compression Failure.
The first two stages produced by increasing the load are exactly the same as in the
From similarity of triangles CBA and CDE, previous two cases. However, in the third stage shown in Fig. 3.8, concrete crushes and
steel starts yielding simultaneously. The stress in steel, fs, becomes equal to its yield
g, d —c stress, fy: The depth of neutral axis and the depth of rectangular stress block are
0.003 denoted by ck and at, respectively.
es = 0.003—— —
d c
(i) Extreme fiber concrete strain, £c 0.003

c Extreme fiber concrete stress = 0.85
d -r
fs =EsEs = 0.003 Es~^- (ii) ft = fy and £s = Ey = fy/Es
c
c = at pt Using strain diagram of Fig. 3.11,

°!p> , g, _
fs “
0.003
al Px
= 0003.^ ° (iii) 0.003 cb
Cc = 0.85 f'b a (iv) EyCb — 0.003 d— 0.003 Ct (0
cb(fy + 0.003) = 0.003 d
- As f, = Ar 0.003 Es p^d-a
T

Cc = T => 0.85// ba = As 0.003 Es a


(v)
cb = — -
0.003
fy!Es+0.003
d

0.85 f' ba2


cb =
.
— — —d
0.003 Es
4+0.003^
(ii)
= pxd-a
——
'

0.003AsEs a 0.003 E ,
ak = Pt cb => ak = Pt r— (iii)
( 0-85 f'b 4 0.003^
[o.OO3AtE. \ a2 + a
-
— ptd =0 (vi)
+
The balanced steel ratio, pb, may be found by equating the resultant tensile and
compressive forces, as done earlier. Moment capacity may either be determined by
. ,,
As = pbd =>
( 0.85 f'b 1 2
[a^a-ptd
^0.003pW£j =0 using the tension or the compression formula. The value of ^-factor is to be taken equal
to 0.65.
Multiplying the equation with d,
Mu = ^Asfy(d-ahl2) or Mu = <j)0.85f'bah (d-ab/z) (iv)
( 0.85/ > a2, + ad-Ptd2
, ,
" =0 (vii) A comparison of strain diagrams for the three types of failure is presented in Fig. 3.11.
^0:003^p) The steel strain at failure is the main factor that determines the type of failure and the
CHAPTER 3:
DESIGN OF BEAMS 81
n
d”
CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
Moment Capacity, Mn
ductility of the member. In all cases, the final failure involves the crushing of concrete may be used to determine the flexural capacity of a member:
The following procedure
in the compression zone.
j Calculate the depth of neutral axis based on the assumption that the section is under¬
reinforced.
JA-
° = 0.85 fib
c = a!Pi (Not valid value for compression failure)

2 Calculate the steel strain. Also calculate q, if different.


= 0.003 c
If , the section will be under-reinforced as assumed, go to the next step.
Fig. 3.11. Ultimate Strain Distribution for Various Types of Failures.
If < Sy , the section will be compression-controlled, revise the value of a by using
The comparison of extreme tensile steel strain, depth of neutral axis and the maximum the following equation:
steel ratio is given in Table 3.2. ;
J + ad-Pid2 = 0
Table 3.2. Basic Properties of Sections Based on Flexural Behavior. 600 p )
i Alternatively, write equation for fs in terms of a using the strain diagram.
'Extreme Type of Ratio c / d Ratio did
5
p
Steel
Tensile Section * fs = 0.003 Es
0,d-a
ai
drain, q


Compression 600 See compression 0.65 i
This value is then used in the equation for equilibrium of Cc and T.
> j

-controlled 600+4 600 + 4 members j



T = Cc => AJS = 0.85fc'ba
Under- 600 pZPb Use
reinforced
<
" formula Simplification of this equation will give the same quadratic equation as above.
600-k/, Pb^
0.854 Determine the ^-factor depending upon the value of £t-
5
600 3.
P1
4 600+4 st > 0.005 =>^ = 0.90
'Tow4 With code
VI
m]
Use St Sy => ^, = 0.65
Mnax A formula
minimum r- f
Jy 7' 0.002 <Et < 0.005 => Calculate using the transition formula
steel strain
4. Calculate the flexural capacity, ^34 as under:
"^0005 Tension-
VI )
m VI
m 0.90
oo = fMJy (d—ali) for under-reinforced sections
|

controlled oo Jy 5
?W4i = ^0.85^ 'fi a (d-a/z) for compression-Controlled sections
^00075 Negative ^81
. =0^1_
?

moment 0.90 Alternate Method


IA1

Mnax Pi
redistribution "5 y
Jy
possible 1. Calculate steel ratio (p) for the given section.
Calculate the balanced steel ratio.
313 analysis of singly reinforced sections p> Pl znn ,
i) Dimensions like b,d and I s
ii)fc'Jyan<iEs
iii) Area of tension steel, As
cgETE STRUCTURES
PART 1 - AUTHOR: ZAHID A. SIDDlQj DESIGN OF BEAMS 83
CHAPTERS:
82 ftriine the type of failure as follows:
tension failure 1. a
P compression failure 0.85/c'&

C

ulate the value of ‘a’ from one of the following equations:
1900x420
0.85x18x228 —
_nooo„_
228.o mm

2- •, •
AJy
a = i 0.003^^ °'8^x4^0 -228 8
TeOsionfaliure- 2. 5 = = 0.003

-
0.85//6 a 228.8
<0.85/' ,
[ 6(W; J -° = 0.00202 = 420/200,000 = 0.0021
Revise a by considering the expression for the compression-controlled section.
_ A singly reinforced rectangular section has a width of 228 mm and an p = As! bd = 228x450
= 0.0185
450 mm. Use C-18 concrete and Grade 420 steel. Calculate the design
0-85/;
for 311 3163 of stee! equal to 1116 following; a2 + ad- ptd2 = 0
-q3-#25 + 2-#15 (SI) ui) steel for balanced failure
0.003^ p
0.85x18
a2 + 450 a -0.85 x4502 =0
$2^ = 228 mm
1^600x0.0185} "

b 1.378 a2 + 450 a - 172,125 = 0


d = 450 mm
ft
fy
Es
= 18 MPa
= 420 MPa
= 200,000 MPa 3.
“ z
2x1.378 ————
_ - 450 + ^450z + 4 x 1.378 x 172,125 226 mm
The section is compression-controlled
=
=> = 0.65
=?
4. = 0.85 fc'ba (d-a/i)
= 1000 mm2 I = (0.65)(0.85)(l 8X228X226X450 - 226 / 2) / 106
= 1500 + 400= 1900 mm2 = 172.7 kN-m
balanced failure i Case flit)
600
i
1. Pb = 0.85 P,
a 0.85//i s

1.

0^x18x228
1000 x 420
= 120.4 mm 1
r
= 0.85 x 0.85

420 1020
= 0.01821

I 2. Let As = pb bd = 0.01821 x 228 x 450 = 1869mm2


g0C03fe= 0:00^x450-120.4 , i
. 120.4 (1869X420)
3. Ob 225 mm
= 0-00653 > £y = 420/200,000 = 0.0021 ! 0-85 /Ji (0.85)(18)(228)
!
=0 tension failure, ^, = 0;90 4. = ^Aifyid-ah)
3. ah) = (0.65)(l 869)(420X450 - 225 / 2) / 106
, >/ = $>A,fy(d i = 172.2kN-m
4. 000)(420)(450
(0-9X1 - 1 20.4 / 2) / 106 i
= 1473 kN-rn
Example 3.2 (FPS Units): A singly reinforced rectangular section has a width of 12 in
First assuming the under-reinforced behavior, we get. and an effective depth of 15 in. fe'= 3,000 psi and fy = 40,000 psi. Calculate the design
flexural strength for an area of steel equal to the following:
84 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI
CHAPTER 3: DESIGN OF BEAMS 85
i) 6-#8(US) ii) 6-#10(US) 0.85x3,000
a2 + 15 a -0.85x 152 = 0
Solution: 87,000x 0.0422
b - 12 in 0.695 a2 + 15 a - 191.25 = 0
d = 15 in -15 + V152 + 4x0.695x19125
a = 9.0 in
fc' = 3,000 psi 2x0.695
fy = 40,000 psi
Et =
=?
29 x IO6 psi 3
4.
The section is compression-controlled
^,Mn = ^>6.S5fc'ba (d-a/2)
=>
^ = 0.65
Case(i) As = 4.71 in2 Case(ii) As =7.59 in2 = (0.65)(0.85)(3,000)(12)(9.0X15 - 9.0 /2) / 12,000
Cased) = 156.63 k-ft

3.14 DISTRIBUTION OF FLEXURAL REINFORCEMENT


O.&f’b
For better distribution of flexural reinforcement in beams and one-way slabs (ACI 10.6.4)
4.71x40,000
related with control of the crack width, the maximum center-to-center spacing (s) of
0.85x3,000x12 reinforcement closest to a surface in tension must be as under:
2. % 0.003-—'--
= 0.00321
a
a

>
= 0.003
$
— 6.16
6-16

= 40/29,000 = 0.00138
(40,000) -2.5 , J 40,000
(SI)

>3. > Sy => tension failure, 0b = 0.90 Smax = 15 CC < 12 - (US Customary)
4. faMn = ^Asfy(d-al2)
= (0.9)(4.71)(40,000)(l 5 —6.16/2)/ 12,000 When there is only one bar or wire nearest to the extreme tension face, s is the width of the
= 168.43 k-ft extreme tension face. In this expression, f is the calculated stress in reinforcement at service
Case (ii) First assuming the under-reinforced behavior, we get, load, which is found by dividing the un-factored moment by the product of steel area and
internal moment arm. It is permitted to take f = 2/3 fy in place of detailed calculations. Hie
term cc denotes the clear cover from the nearest surface in tension to the outer surface of the
0.85//5 flexural reinforcement. This provision limit the surface cracks to a generally acceptable width
7.59 x 40,000 but actual value may vary widely in a given structure. The previous code gave expressions
0.85x3,000x12
based on a calculated crack width of 0.4 mm (0.016 in). The following values are Obtained in
some of the typical cases:
2. = 0.003^^= 0.0032^1^92 fy = 280 MPa, ce = 40 mm, smax = 450 mm
a 9.92 fy = 420 MPa, = 40 mm, smai = 280 mm
= 0.000856 < Ey= 40/29,000 = 0.00138 Jy = 460 MPa, Ct = 40 mm, = 245 mm
Revise a by considering the expression for the compression-controlled section. = 280 MPa, ce = 20 mm, Smax = 450 mm
7 59 fy = 420 MPa, cc - 20 mm, = 300 mm
p =AJbd=~ = 0.0422 = 460 MPa, cc = 20 mm, = 270 mm
12x15
3.15 SKIN REINFORCEMENT IN DEEP BEAMS

For relatively deep flexural members, skin reinforcement is to be provided, which is to be


distributed along the side faces of the tension zone to control cracking in the web. If this
88 CONCRETE STRUCTURES PART -1 DESIGN OF BEAMS 89
AUTHOR: ZAHID A. SIDDIQ] CHAPTERS.-
distributed over the length of the beam. This increased
1. To protect the steel reinforcement against corrosion. increased by some factor and is then for the beam.
2. To provide sufficient bond strength between steel and concrete. load may not give the true shear force
3. To protect steel against fire, up to certain extent, and to improve the
fire-rating of the
structure. t 3.20.1 Beams Supporting One-Way Slabs
4. To reduce abrasion and wear of the steel reinforcement a one-way
If a slab panel has its length-over-width ratio greater than two, it is called
The following minimum concrete cover is to be provided for all the reinforcement: ’ slab. In case of one-way slabs supported on beams, the width of slab supported
shown in Fig. 3.12.
by each
Similarly, the
interior beam is equal to the spacing of the beams as
a) Concrete permanently exposed and cast against earth be equal to half of the spacing
75 mm (3 in.) width supported by an exterior beam in such cases will
b) Concrete occasionally exposed to earth or unsheltered concrete between the beams.
No. 19 (#6 US) bars through No. 57 (#18 US) bars
No. 16 (#5 US) and smaller bars
c) Sheltered concrete not in confArt with amnnH
50 mm (2 in.)
40 mm (1 % in.)
Width supported by an interior beam Zj
Width supported by an exterior beam =
where -
— 2 + cantilever portion, if any
spacing between the beams and
Slabs, walls and joists, No. 36 (#11 US) and smaller bars 20 mm (3/4 in.) i
= span of the beams
Beams and columns 38 mm (I’/j in.) I
3.20.2 Beams Supporting Two-Way Slabs
d) Shells and folded plates is called a
If a slab panel has its length-over-width ratio lesser than or equal to two, it
No. 1 6 (#5 US) and smaller bars 13 mm (3/4 in.) , two-way slab. The beams are usually provided in the two mutually perpendicular
beam, may be
directions for such slab systems. The area, contributing load to each
:

is will
3.19 NUMBER OF BARS IN ONE LAYER assumed to have 45° angles from all the beams, as shown in Fig. 3.13. This
cause some difference from the actual situation in case one end is simply supported.
The maximum number of bars in a layer of beam reinforcement depends on the
diameter of
main reinforcement diameter of the shear reinforcement the clear concrete cover and
the
required minimum spacing between the bars. Using No. 19 (#6 US) bars
as main
reinforcement and No. 10 (#3 US) bars as the shear stirrups, Ilie approximate number of bars
per layer may be calculated asunder:

Nb = approximate number of bars in a single layer


bw = width of web of the beam
Clear spacing
50 Nb = hw-70
= — -2x40-2x10- AL5 x20
Nb — 1
= 30

Nb = 0.02 b„- 1.4 rounded to lower whole number (SI units)


Fig. 3.12. Width of One-Way Slabs Supported by Beams.
3.20 LOADS SUPPORTED BY BEAMS

In a structural system, slabs are to be supported by beams at regular spacing provided in Slab width supported by interior smaller beams (Bi)
one or Fig. 3.14.
two mutually perpendicular direction. Each beam has to support a particular area and The area, contributing load to one interior smaller beam, is shown in
width of
the slab that determines the total load acting on that beam. If the load acting on the beam not Actual area supported = t? ! 2
is
UDL, the calculations become very lengthy especially in case of continuous beams. In such To get an
cases, usually an equivalent UDL is calculated which gives the same This area exerts a triangular load on the beam as shown in Fig. 3.15.
maximum bending bending moment in simply
moment as in case of the actual load for a simply supported beam. For triangular loads equivalent UDL on the beam that gives the same maximum 4
with
maximum ordinate at the mid-span and trapezoidal loads, the actual load on the beam is to be supported conditions, it is to be multiplied with the factor /3.
90 CONCRETE STRUCTURES PART 1
-
AUTHOR: Z AHID A SIDDIQI APTERS: DESIGN OF BEAMS 91
This area exerts a trapezoidal load on the beam as shown in Fig. 3.16, and to
approximately convert it in to an equivalent UDL, the load is to be increased by a factor
given below:

Fig. 3.16. Slab Loads Supported by Interior Longer Beams.

Equivalent slab width for calculation of bending moment

For first interior beams, the above width is to be increased by 10%, if exterior beams

———— = 2/?
.4.' do not have a significant cantilever slab portion.
Equivalent slab width for : ' X
calculation of bending moment Slab width supported by exterior longer beams (B4)
. '
=
For first interior beams, the X
/3
do not have a significant above width is to be increased by 10%, if exterior beams Equivalent slab width for calculation of bending moment = 1,
cantilever slab portion.
Slab width supported by + cantilever width, if any
exterior smaller beams (Ba)
The width supported by the The effective slab width supported by a beam multiplied with the load per unit area of
beam. exterior beam will simply be half the slab gives the load on the beam per unit length.
as that of the interior
Equivalent slab width for 3.20.3 Walls Supported By Beams
calculation of bending moment
= 1/3 1* The load of wall on the beams may easily be calculated by multiplying the unit weight
+ cantilever width, if any of brickwork with the volume of brick wall per unit length as follows:
Slab width supported by
interior longer beams (63) tw = thickness of wall in mm
The area, contributing load to
one interior longer beam, is
shown in Fig. 3.13. H = height of wall in m
Actual area supported = 1,1
f-~
UDLonthebeam
t
= 1930x 1000 —
9 81
1000
kN/m

= 0.019 tw H kN/m (SI)


= 0.12 k/ft (FPS)
92 CONCRETE STRUCTURES PART -I CHAPTERS: DESIGN OF BEAMS 93
author.- zahid a. siddiqi
3.21 WALLS SUPPORTED BY LINTELS
Brick tiles: x 1930 = 74 kgs/m2
The portion of wall load supported by a 1000
remaining load is supported by the arch lintel is a 60° triangle as shown in Fig. 3.17. Total dead load: = 554 kgs/m2
action within the wall. The Live load: = 200 kgs/m2
tw = thickness of wall in Total factored load: = 1.2 D + 1.6 L
£ = span of the lintel in mmm = [1.2(554) + 1.2(200)] x 9.81 / 1000
Weight of the triangular wall portion = 9.66 kN/m2
-J3 £x £x
= 18.9x—1 x— t 1
Intermediate Floor: For 150 mm R.C. slab, 75 mm brick ballast, 40 mm thick P.C.C.,
20 mm thick terrazzo finish, and the occupancy type of an office, the slab load may be
8.18^2/, calculated as under:
1000

Equivalent UDL for bending


moment =- — —— ——
4 8.18£2f
x
R.C. slab:

Brick ballast:
1000
75
1000
x 2400

x 1800
= 360 kgs/m2
= 135 kgs/m2
=
=
0.01 If /„ kN/m (SI)
0.0058 £ tv k/ft (US Customary)
P.C.C. + terrazzo:
Total dead load:

1000
x 2300 = 138 kgs/m2
= 633 kgs/m2
Live load: = 250 kgs/m2
Total factored load: = 1.2D + 1.6L
= [1.2(633) + 1.6(250)] x 9.81 / 1000
= 11.38 kN/m2

3.22.2 Calculation Of Self Weight Of Brains

After selection of trial dimensions, self weight per unit length may be approximated as
follows:
Fig. 3.17. Wall Loads Supported
by a Lintel. Self load = 2400 x--* -x x
1000 1000 1000
kN/m
3.22 CALCULATION OF BEAM LOADS
3.22.1 Slab Loa «1^
The slab load is always
Most general value of unfactored self weight = 2400 x
= 0.1 1£2
——x x

kN/m (SI)
Hence, the dead load may calculated as the load acting on a unit area (1
be m2) of the slab. = 0.0007f2 k/ft (FPS)
weight of various materials calculated by multiplying the thickness with the unit
explain the calculation of the exerting load on the slab. The following two examples i 3.23 BAR BENDING SCHEDULE
roof loads:
Top Roof: For 125 mm R.C. slab,
the slab load may be 100 mm earth filling and brick After calculation of steel requirement for bending moment, shear force and torque at important
calculated as under: tiles (38 mm thick),
sections of a beam, detailing for reinforcement is carried out. Detailing means selection of the
R-C. slab:
-^-x2400
1000 = 300 kgs/m2
bar sizes, number of bars, bar spacing, bar positions and bar cut-offs and showing the results on
a neat sketch. A bar bending schedule is then prepared which helps the cutting, bending and II
placing of bars. The number, length, diameter, steel grade and shape of each bar are separately ft
Earth filling:
-^xl800
1000 = 180 kgs/m2 Written. Each type of bar is given a separate designation on the drawings and the
94 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS: DESIGN OF BEAMS 95
corresponding values are entered in the bar-bending schedule. This table gives the
total
amount of steel required for a particular project. The columns of a typical bar bending
schedule are given in Table 3.3. 8
2-#10(M-3)
The details of bent-ups and standard hooks are given below:
375 #10@180mm C/C (S-l)

2 - #20(M-l)

1 - #15, Bent-Up (M-2)

a) Mid-span Cross Section

Fig. 3.19. Beam Reinforcement Details for Example 3.3.

Solution:

Table 3.4. Bar Bending Schedule for Example 3.3. Steel Grade: 420
s. Bar No. Len. Dia. Weight Of Bars Shape Of Bar
No. Desig¬ Of Of Of No.10 No.15 No.20 No.25
nation Bars Bar Bar
(m) (mm)

1 M-l 2 5.056 20 23.9 4436


Jf- 2860
1 S . 7?.. . 260


/
2 M-2 1 5.092 15 10.0
wet
b) Standard jonou
90° Hook
i
c) Standard 180° Hook J4 4436 /
1 1
Fig. 3.18. Details of Bent-Ups and Standard Hooks. 3 M-3 2 5.056 . 10 8.0
4436
Approximate amount of steel in beams = 0.0002 bh. kgs per meter (± 20 %).
Example 3.3: Prepare bar bending schedule for the reinforcement of the beam shown in
Fig. 3.19. Clear cover is 40 mm and the bar sizes are in SI units.
4 S-l 25 1.206

With 5% margin for wastage £=


10 23.7

33.3 10.5 25.1


138“
£285
=
Total Steel Required 69 kgs
96 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI DESIGN OF BEAMS 97
CHAPTERS:
3.24 DESIGN OF SINGLY REINFORCED SECTION
5. Deciding Overall Depth Of Beam
Data: i) Bending moment or load and span of the beam Following considerations must be given while deciding the overall depth of the beam:
ii) fc\fy andEj There is usually an architecturally proposed maximum depth, which is denoted
iii) Required architectural depth by ha. It is better to satisfy this requirement, but it is not a must.
Required: ii. The Code minimum depth for deflection control is usually satisfied, which is
i) Beam dimensions b and d
denoted by hmin.
ii) Area of tension steel, As iii. For singly reinforced section, the effective depth should not be lesser than dmin
iii) Detailing and bar bending schedule defined above. This means that the total depth, /i, must be approximately dmi„ +
75 mm or greater.
The beam may be designed by one of the following two methods: iv. Consideration must be given to the most economic depth. Further, the most
a) Selecting the trial dimension of the beam (b and d). common depth of beams is taken as 1 1 12.
b) Selecting steel ratio p (which should be lesser than or equal to PmaA as a The selected depth is rounded either to the nearest brick height or some other
percentage of the maximum steel ratio. architectural dimension.
vi. It is tried to keep same beam depths across a floor. At least, smaller difference
The first method is relatively more common. After making a decision about the type of the in heights must be avoided.
method to be used, the further procedure becomes different. vii. For multistory buildings, try to minimize the floor-to floor height, as far as
possible, and to maximize the headroom.
3.24.1 Using Trial Beam Dimensions
The selected depth must be a compromise between all the seven requirements.
1. Calculate the loads acting on the beam. of steel
6. One of the following four methods may be employed to calculate the amount
2. Calculate the maximum factored load using the applicable load combination and required for the applied bending moment The selected method depends upon the
determine the maximum bending moment using any load analysis procedure. For the convenience of use, availability of design tables or charts for the selected concrete and
simple gravity load combination, the following expression is used: steel strengths and the choice of the designer.
wu = 1.2 w& + 1.6 wi
A. DIRECT SOLUTION
3. Select the beam width, which usually varies from 4? / 15 for smaller beams to 1 1 20 for
larger beams. The width must be sufficiently larger to avoid the congestion of steel. It
, \
Calculate the factors, co - ———
0.85/; and J?
=
bd2
then,
is to be noted that more shear reinforcement may be required for less-width beams.
This width must also match with the dimensions of the supporting columns and walls.
The selected dimension is generally rounded to the nearest 75 mm or multiples of half P
_
= . .
«,11

2.614R
brick length.
4. Calculate the minimum depth required for the tension failure using the formula derived As = pbd
earlier:

dmin I
V 0.205/;b— (Both SI and US Customary Units)

Select beam dimensions such that d is greater than or equal to dmi„.


B. USING DESIGN TABLES
Calculate I bd2 and see the value of p directly from the tables developed between
!bd2 and p for particular values of /' and /,. The steel area then can easily be
calculated using the following expression:
h dmin + 60 mm for single layer of steel, and,
h = ‘inin + 75 mm for more layers of steel. A.t - pbd i
98 CONCRETE STRUCTURES PART - I
AUTHOR- ZAHID A. SIDDIQI DESIGN OF BEAMS 99
C. USING DESIGN GRAPHS CHAPTERS:
3. Calculate the maximum steel ratio as follows:
Design curves are also developed between
/bd2 and p for particular values of
and fy. Using these curves, the value of p^ = 0.85 x 0.375#
then the steel area is calculated as usual.
required steel ratio may be directly read
and
As = pbd Select any steel ratio (p ) lesser than
D. USING TRIAL METHOD 4. The quantity bd2 is calculated from the following equation:

> Assume a reasonable value of the depth of rectangular stress block, a A , ( 1


d/4 or d/3 may be selected in the start. value close to Mu - $Mn = ^pbd2fy 1 p

> Determine the steel area required for the above assumed value of a. bd2 =
- M-
A
'
0.94^-^] 5. The beam dimensions b and d are decided such that bd2 value is satisfied. The depth
> Revise the value of a by using the above value should not be lesser than the minimum depth specified by the ACI Code for deflection
of 4S- control.
"
a= 6. Area of steel is then calculated as follows:
^f'cb -

A3 = pbd
> Revise the above two steps until 4S becomes nearly
constant. 7-10. Same as in first method.
E. USING APPROXIMATE METHOD
Example 3.4: In a slab system, longer beams have a single simply supported effective
span of 8 m and shorter beams have three spans of 5 m each. Slab thickness is 160 mm and
For beams, if d> I floor finish consists of 75 mm of brick ballast and 50 mm of floor finish, the longer beams
x= —
0.756* dxf? support a 228 mm thick and 3 m high brick wait The structure is to be used as an office
building. Use C-20 concrete and Grade 420 steel. Selecting US bars, design the interior longer
For beams, if d> I K
As=-^~ beam having a rectangular section with width of 300 mm for flexure under the following
conditions;
7. If area of steel is lesser than the i) Minimum depth obtained with maximum steel ratio, even if it violates the ACI
_ 1.4 minimum, provide the minimum value. minimum depth for deflection control.
A .
•zxs,jnin
—— fory£' 31.4 MPa and SI units. ii)
iii)
Overall depth equal to 835 mm.
Overall depth of 910 mm using trial and error method. Also prepare bar bending
8. Proper detailing of the steel is schedule for this option.
carried out including the selection of bar diameters,
number of bars, cover, spacing between the bars, Solution;
bar cut-offs and bar bent-ups.
9. Detailed sketches are drawn to Data
completely represent the design and detailing.
10. Bar bending schedule is prepared, Z = 8m
if required.
Slab thickness - 160 mm
3.24.2 Steel Ratio Is Known Floor brick ballast = 75 mm
1. Same as in first method. Floor finish = 50 mm
fc' = 20 MPa : fy = 420 MPa
2. Same as in first method. b = 300 mm
100 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI DESIGN OF BEAMS 1Q1
CHAPTER 3:
Slab Load
factored Slab Load Acting On Beam
R.C. slab:

Brick ballast:
—75
160
x 2400 = 384 kgs/m2 Factored slab load on beam = width of slab x slab load per unit area
- 4.78 x 11.39 = 54.49 kN/m
1000
xl 800 = 135 kgs/m2 Total Factored Load
50
P.C.C. + terrazzo:
j^x2300 = 115 kgs/m2 = 54.49 + 15.54 + 5.65 = 75.68 kN/m
Total dead load: = 634 kgs/m2 Total Factored Bending Moment
Live load: = 250 kgs/m2 75.68 x82
Total factored load: = 1.2D + 1.6 L = 605.44 kN-m
8 8
= [1.2(634) + 1.6(250)] x 9.81 / 1000
Minimum Effective IDepth For Singly Reinforced Section
Brick Wall Load
= 11.39 kN/m2
_ 228 „„ 9.81
_ I 605A4xlQ~ = 702 mm
Service dead load - 1930x x3.0x 0.205 X Xb V0.205 x 20 x 300
1000 1000
= 12.95 kN/m i = 1.4/^, = 1.4/420 = 0.00333
Factored dead load = 1.2 x 12.95 = 15.54 kN/m > Depth For Deflection Control
Approximate Self Weight Minimum depth of beam for deflection control = ^/I6
C = 8000/16
Service dead load

- o/mn
2400 x x x
——
300 8 9.81
1000 12 1000
= 4.71 kN/m
:
Cased)
= 500 mm

steel ratio.
Factored dead load = 1.2x4.71 = | The minimum depth will be obtained by using the maximum permissible
5.65 kN/m However, this will be equal to the already calculated dmi„.
i
Equivalent Width Of Slab Supported By Beam
Bl
The beams B| are shown in Fig. 3.20. [ = 0.375 x 0.85 A
Jy
V= 8 m : 4 = 5 m : R = ^/ty = 5/8 = 0.625 •
i"
Equivalent slab width supported = (1 - R^/S) h = = 0.375 x 0.85 x 0.85 x— = 0.0129
420
= (l-0.625z/3)x5 x 1.1 = 4.78m
(10% extra for first interior beam)

s 300x300 mm Columns I
d2 = 492,328
V —
^2^9x420
605.44 X 106 = 0.9x0.0129 x 300 x d2 * 420 x [1 1.7 20 J
J
d = 702 min
= pbd
= (0.0129)(300)(702) = 2717 mm2
6 No.25 provide 3060 mm2 (12.6 % higher)
3 @ 5 m Each 5 No.25 + 1 No.22 provide 2937 mm2 (8.1 % higher)
Fig. 3.20. Layout of Beams for Example 3.4.
The reinforcement details are shown in Fig. 3.21.
102 CONCRETE STRUCTURES PART - 1 AUTHOR; ZAHID A. SIDDJQj
design of beams 103
CHAPTER. 3:
As = pbd
= 0.01065 x 300 x 760 = 2428 mm2
Clear Cover = 40 mm —
2 #13
#10@300mm C/C
-
4 - #22 + 2 #25 will be sufficient (2568 mm2) (OK)

a) Mid-span (assumed) Method 2


Cross Section
2 #25 +1 #22 Bent Up R = 3.494 MPa as before.
From the table corresponding to fc'= 20 MPa and fy = 420
MPa,
;
3-#25 p =0.0108 (by approximate interpolation)
Almost the same reinforcement as in the first method.
-#13 3- #25

Method 3
#10@300mm C/C R ~ 3.494 MPa as before. to R = 3.494
Using design curves for fc'~ 20 MPa and fy = 420 MPa, corresponding
? MPa, the following steel ratio is obtained:
3 + 2 - #25 + 1 - #22
7 7700/20^ r
B
p = 0.0107
Almost the same reinforcement as in the first method.
= 385 mm '

। ^=1100 mm i
' Method 4
300 7700 300
b) Longitudinal Section Trial 1: Assume a = d / 3 = 760/3 = 253 mm
Fig. 3.21. Reinforcement Details of Beams for Example 3.4 (Case-i).

When using bent-up reinforcement, the bars may be bent at a distance of


_
simply supported end, if at least 50% of the total steel is continued beyond
point For curtailed bars, this distance for simply supported end may
4 / 20 (no curtailment according to the ACI Code).
/7 from a
the bent-up
be taken equal to — — —I 605.44 x10s
- (
0.9x 420 760-
""

—J
"
253
2
— , 2
2528 nun

c •
Spacing u 4.
between u
bars = 300-2x40-2x10-3x25
= 62.5c mm (OK) Trial 2:
0.85/;#
Case (li)
h = 835 mm
= —— =
2528 x 420
0.85x20x300
208 mm

d & 835-75 = 760 mm > dmin =


Mv = 605.44 kN-m
Methodi
0-85x20 = = 2442 mm2
at

R
=
_
— M
4
=
= 420
605.44 x10s
= 0.0405
— 3 494 2442x420
/
0.9x 420 760-
k —
70R
2 )
i

V- 300x760’ Trial 3: = 0.85x20x300


= 201 mm
P = Jl-
> Pnin
I V Z J - 0.040Sf1-^^14x3.494
( V 20
= 0.01065
(OK)
1 04 CONCRETE STRUC TURES P ART - 1
AUTHOR.- ZAHID A. SIDD[qi DESIGN OF BEAMS 105
CHAPTER 3:
„ 605.44x10s
As =
0.9x420 760-
(7
I —
2
v
201)
J
(sufficiently close to the previous answer)
Almost the same reinforcement is obtained
- 2429 mm2

as in the first method, shown in Fig.


605.44 x10s
(
0.9x420 835
I
2164x420
190
2
= 2164 mm2

Case (Hi) 3.22. 1’rial 3: = 178 mm


0.85x20x300
h = 910mm 605.44x10s
d = 910-75 = 835 mm
178
= 2147 mm2
Trial 1: Assume a ~ d!3 0.9x420 835
= 835/3 = 278 mm I 2 .
As _
~ 605.44x10s (sufficiently close to the previous trial)
- 2301 mm2
0.9 f d--
k 2
•.
0.9x 420 835-—
I 2
(

—J The area of steel is provided by 5 #22 + 1 #19 give 2219 mm2, as shown in Fig. 3.23.

300
2- #13
Clear Cover = 40 mm ’"Jr —
2 #13
Clear Cover = 40 mm
#10@300mm C/C

a) Mid-span
Cross Section
760 — #10@300mm
(assumed)
C/C a) Mid-span
Cross Section
(assumed)
2 #22 +1 #19 Bent Up
P-Q-4 - 3- #22 3 - #22
~ 2 - #25 +1 #22

4/5 = 1540
4 4 -#13 2-#I3 + l-#22
/ 1 - #19 (M-3) —
2 #13(H-1)
3-#22(M-l)

#10@300mm C/C (S-l)


#10@300mm C/C
2 - #22 (M-2) + 1 - #19 (M-3) t f J

2 #25 + 1 - #22 +3 #22
— 7700 / 20 1
। ^=1100
mm
7700 / 20 !
= 385 mm
1
300
2-#25+l-#22
7700
—385 mm
M 300 7700
b) Longitudinal Section
M
300
300
b) Longitudinal Section
Fig. 3.23. Reinforcement Details of Beams for Example 3.4 (Case-iii).
Fig. 3.22. The bar bending schedule is given in Table 3.5.

A*fy Example 3.5: In a slab system, longer beams have a single simply supported effective
Trial 2: a 2301x420
= 190 mm span of 25 ft and shorter beams have three spans of 15 ft each. Slab thickness is 6 in and floor
°-8V> O^5?2O^5o “
finish consists of 3 in of brick ballast and 2 in of floor finish. The longer beams support a 9 in
thick and 7 ft high brick wall. The structure is to be used as an office building. fc'~ 3.0 ksi
and fy = 60 ksi. Selecting US bars, design the interior longer beam having a rectangular section
with width of 12 in for flexure under the following conditions:
1 06 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 3: DESIGN OF BEAMS 107
i) Minimum depth obtained with maximum steel ratio, even if it violates the ACI
minimum depth for deflection control. P.C.C. + terrazzo: — x 144
12
= 24.0 PSf
ii) Overall depth equal to 30 in. = 126.5 psf
Total dead load:
Live load: = 50 psf
Table 3.5. Bar Bending Schedule. Steel Grade: 420 Total factored load: = 1.2 D + 1.6 L
s. Bar No. Len. Dia. Weight Of Bars Shape Of Bar = [1.2(126.5) +1.6(50)] /1000
No. Desig¬ Of Of Of No. 10 No. 13 No. 19 No. 22 = 0.232 ksf
nation Bars Bar Bar Brick Wall Load
(m)
9 1
1 M-l 3 8.200 #22 74.84 n Service dead load = 120x— x7x = 0.63 k/ft
8200 ' 12 1000
Factored dead load = 1.2x12.95 = 15.54 k/ft
2 M-2 2 6.930 #22 42.16 / 6930 z Approximate Self Weight

4 M-3 1 9.500 #19 21.23


F
S
5500^r
8200 %
1
744 Service dead load
=
12
= 150x — x —
25 x 1
12 12 1000
0.31 k/ft
5 H-l f 1
2 8.434 #13 16.77 K- ——*/ Factored dead load = 1.2 x 0.31 = 0.38 k/ft
8200

6 S-l 27 2.420 #10


Equivalent Width Of Slab Supported By Beam Bi
36.59 210

• 820
The beams are shown in Fig. 3.24.
S= 38.5 | 17.6 22.3 122.9
Total Steel Required = 202 kgs
ty = 25 ft : tx = 15 ft : R = 4 / ty = 15/25 = 0.6
Equivalent slab width supported = (1 - R2^) h
Note: Clear cover on both longitudinal ends is taken as 50 mm. = (1 -0.62 /3) x 15 x 1.1 = 14.52 ft
(10% extra for first interior beam)
Solution;
Data
f = 25 ft
Slab thickness = 6 in


Floor brick ballast = 3 in
Floor finish 2 in
,4'=3.0ksi
fy = 60 ksi
b = 12 in
Wall thickness = 7 ft
Wall thickness = 9 in
Slab Load
R.C. slab:

12
x 150 =
.
75.0 Hpsf
Factored Slab Load Acting On Beam
Brick ballast: — x 110
IO
=27.5 rpsf Factored slab load on beam = width of slab x slab load per unit area
14.52 x 0.232 = 3.37 k/ft
=
:a
DESIGN OF BEAMS 109
108 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS:
12-2x1.5—2x0.375-3x1

Total Factored Load Spacing between bars
wB = 3.37 + 0.76 + 0.38 = 2.625 in (OK)
= 4.51 k/ft
Total Factored Bending Moment 12
2-#4
Clear Cover = 1.5 in
#3@12 in C/C (assumed)
Minimum Effective Depth For Singly Reinforced Section a) Mid-span
Cross Section 2 #7 + 1 #7 Bent Up
, I K I 352.4x12000 „..
3— #8
} 0.205 xfc’xb V 0.205x3000x12
p„i„ = 200 //, = 200 / 60,000 = 0.00333
Depth For Deflection Control

Minimum depth of beam for deflection control = t / 16


= 25 x 12/16 = 19 in
Case (i)
The minimum depth will be obtained by using the maximum permissible steel ratio.
However, this will be equal to the already calculated dmm.

pm^ = 0.375 x0.85 Pi


—fy b) Longitudinal Section
(Case-i).
= 0.375 x 0.85 x 0.85 x— = 0.0135 Fig. 3.25. Reinforcement Details of Beams for Example 3.5
60
, ( X pfj\ Case (di
Mu = <t>bpbd2fy\\-~^-\ h = 30 in
d »30-3.0 = dmin
352.4 x 12 = 0.9 x 0.0135 x 12 x d2 x 60 x (1
d2 = 574.7 J = 24 in
I —1-7 3
|
J Mu
Method 1
= 352.4 k-ft

As = pbd a>
_ 0.85 /; 0.85x3
= 0.0425
= (0.0135)(12)(24) = 3.89 in2 60
3 No.8 + 3 No.7 provide 4.17 in2 (7.2% higher) 352.4x12000 = m si
R ‘
The reinforcement detai Is are shown in Fig. 3.25. bd 12x272
When using bent-up reinforcement, the bars may be bent at a distance of /„ / 7 from a =
(
fit 1-1-
r 2.6147?~ 1
P z,

simply supported end, if at least 50i% of the total steel is continued beyond the bent-up
point For curtailed bars, this distance for simply supported end may be taken equal to
4 / 20 (no curtailment according to the ACI Code). - 0 0425
f 2.614x484

— 0.0102 pmu (OK)


110 CONCRETE STRUCTURES PART - 1 AUTHOR: Z AHID A. chapter 3: DESIGN OF BEAMS UI
SIDDlQj
Xt = pbd Nearly same reinforcement is obtained as in the first method and is shown in Fig. 3.26.
= 0.0102 x 12x27 = 3.31 in2
6 - #7 provides 3.6 in2 (OK)
Method 2
R = 484 psi
From the table corresponding to fc 3 ksi and/ = 60 ksi,
p =0.0102 (by approximate interpolation)
Almost the same reinforcement is obtained as in the first method.
Method3
7? = 484 psi
Using design curves for fc'= 3 ksi and fy = 60 ksi, corresponding to R = 484 psi, the
following steel ratio is obtained: \|
p. =0.0102
Method 4
Trial !: Assume a — d/3 = 27/3 = 9.0 in 3-#7 = 15 in
12 in 25 ft 12 in
b) Longitudinal Section

352.4x12
/ = 3.48 in2 Fig. 3.26. Reinforcement Details of Beams for Example 3.5 (Case-ii).
gY
0.9x60 27-— I

3fy
< 2)
3.48x60
Example 3.6: Design a rectangular singly reinforced flexural member to be used as an
interior simply supported short beam. The slab system has a longer span of 5 m and a shorter
Trial 2: a
0.85 f^b
= span of 3.5 m. Factored slab load is 10.00 kN/m2,/'= 17.25 MPa and/ = 300 MPa. Select SI
0.85 x3x
reinforcing bars and keep the width of member equal to 228 mm. Architecturally, maximum
permitted depth is 375 mm.
Solution.:
Data
352.4x12 £ = 3.5 m
= 3.32 in2
0.9x60f27
< 2 )
factored slab load = 10.00 kN/m2
/'=17.25 MPa
3.32x60 . fy = 300MPa
Trial 3: a 3.25 in b = 228 mm
0.85x6x12
A0 = 375 mm
352.4x12
( 3 25)
= 3.09 in2 Approximate Self Weight
0.9x60 27-—
I 2 )
(sufficiently close to the previous answer)
Service dead load = 2400 x
228
x— x : — ——
3 5 9 81
1000 12 1000
= 1.57 kN/m

Factored dead load = 1.2 x 1.57 = 1.88 kN/m


11 2 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
CHAPTER 3: DESIGN OF BEAMS 113
Calculation Of Steel
Equivalent Width Of Slab Supported By Beam
Trial 1: Assume a = d! 3 = 240/3 = 80 mm
ly = 5 m : lx = 3.5 m i
Equivalent slab width supported - % lz = 2/3 x 3.5 = 38.74x10’
2.34 m 4 80'
= 717 mm2
f
Factored Slab Load Acting On Beam 0.9x300 240
Factored slab load on beam = width of slab x slab load per unit area
I 2

Total Factored Load


= 2.34x 10.00 = 23.4 kN/m Trial 2: a
O.S5f'b
717x300
- — ——
0.85x17.25 x 228
= 64 mm

wu = 23.4+ 1.88 = 25.3 kN/m = - 38.74x10s


- - - ,
( 64) — 690
,Qn
mm2
Total Factored Bending Moment 0.9x300 240 ---
I 2 J
690x300
Trial 3: a = 62 mm
0.85x17.25x228
Minimum Effective Depth For Singly Reinforced Section 38.74x10s
A, = 687 mm2
If
hmin = /iin + 60 (assuming one layer of steel) 0.9x300 240--- I
2)

yI0.205/; xb +60 =
38.74 xlO6 (sufficiently close to the previous answer)
= + 60 = 279 mm
0.205x17.25x228
Depth For Deflection Control A reinforcement of 2 #20 + 1 #15 provides the required area of steel (Fig. 3.27).

Minimum depth of beam for deflection control (h„„n) = f/20 228


= 3500/20 = 175 mm
Maximum Architectural Depth
375 mm
Most General Depth
h = // 12 = 3500/12 = 292 mm
The depth may be selected in multiples of the brick height, if possible.
Selected Depth
h = 300 mm
d = A - 60 = 240 mm
The figure of 60 mm is corresponding to expected one layer of steel.
Minimum Steel Ratio
pmm = 1.4// = 1.4/300 = 0.00467
Maximum Steel Ratio
b) Longitudinal Section
pmax = 0.375 x 0.85 rBi
f'
f
0.375 x 0.85 x 0.85 x— —
17 25
inn
= 0.0156
Fig. 3.27. Reinforcement Details of Beams for Example 3.6.

Ia
DESIGN OF BEAMS H5
114 CONCRETE STRUC TURES PART -1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS:
Depth For Singly Reinforced Section
o L .
Spacing between bars =
2

228-2x40 2x10-2x20—15 Mtrtimimt Effective
hmn ~ <4nin + 2.5 (assuming one layer of
steel)

= 36.5 mm (OK)
+ 2.5, 17 in.
fZKZ + 2.5 =
Provided steel ratio = —
228x240 — = 0.01462 > p^u,

Pmax (OK)
=
\ 0.205 fcxb
Depth For Deflection Control
V 0.205 x 2500 x9

(/imin) = //20
Example 3.7 (<ZS
Minimum depth ofbeam for deflection control = 15x12/20 = 9 in.
CustomanA: Design a rectangular singly reinforced flexural
member to be used as an interior simply supported short beam. The slab system has a longer
span of 20 ft and a shorter span of 15 ft. Factored slab load is 270 Ibs/ft2,/ '= 2500 psi and/ = Maximum Architectural Depth
40,000 psi. The width of member is to kept equal to 9 in. Architecturally, maximum permitted ha,max . “ 1? in.
depth is 18 in.
Most General Depth
Solution: = 15 in.
h = II 12 = 15x12/12 multiples of the brick height, if possible.
Data The depth may be selected in
f = 15 ft Selected Depth
wu = 270psf h - 18h-2.5
in.
fc' = 2500 psi = 15.5 in.
fy = 40,000 psi
d = expected one layer of steel.
The figure of 2.5 in is corresponding to
b = 9 in
ha= 18in. ... . . . Minimum Steel Ratio
Approximate Self Weight fe 200 !fy = 200/40,000 = 0.005
9 15 Maximum Steel Ratio
Service dead load - 0.150 x—x
12 12
= 0.14 k/ft
Factored dead load =1.2x0.14 = 0.17 k/ft Pm = 0.375 x 0.85 pt Jy

= 0.375 x 0.85 x 0.85 x-25—


Equivalent Width Of Slab Supported By Beam B<
40,000
= 0.0169
/„ = 20 ft : 1, = 15 ft
Equivalent Slab width supported = 2/3Zx = 2/jx 15 = 10 ft Calculation Of Steel
Factored Slab Load Acting On Beam
Trial 1: Assume a = d!3 = 15.5/3 = 5.17 in.
Factored slab load on beam = width of slab x slab load per unit area S0-72*12 = 2.08 in2
= 10 x 270/1000 = 2.70 k/ft A, = - - =
0.9x40 [15.5-— |
Total Factored Load 0.9/ (d--2|
7 2 J
’ I k
= 2.70 + 0.17
Total Factored Bending Moment
= 2.87 k/ft
Trial 2; a =

=
— ^^^l2
Atf,
=-2—
0.85 fcb
2.08x40 .
= 0.85x2.5x9 = 4.35 m.

= 2.02 in2
As
2.87xl52
= 80.72 k-ft
(
0.9x40 15.5
I -—J
4 35
2
i

8
116 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A.
SIDDIQi CHAPTER 3: DESIGN OF BEAMS H7
Trial 3: a 2.02x40 PROBLEMS
“422ilt
0.85x2.5x9
80.72x12

0.9x40
r155-—-^ = 2-01 in'.2 H Note: N
No
=
=
Registered number.
A fixed positive or negative number assigned by
l 2 J the instructor to get R between 600 and 1000.
(sufficiently close to the previous answer) < R = -
.N No, a parameter to get numerical data for the
A reinforcement of 2 #8 + 1 #6 provides the ‘ problems.
required area of steel (Fig. 3.28).
% z 9 z

q i. A singly reinforced rectangular section has a width of 300 mm and an effective depth of
Clear Cover = 1.5 in 18
2-#4
#3@7 "C/C (assumed)
0.75R mm. Use C — 20 concrete and Grade 280 steel. Calculate the ideal flexural
strength for an area of steel of:
W P. » 2 #8 *■ (i) 2.5 R mm2 (ii) 5 R mm2 (iii) Maximum allowed by the Code.
1 -#6, Bent Up
a) Mid-span Cross Section Q.2. A 375 mm wide singly reinforced rectangular section has to carry a service load
bending moment of R 1 7 kN-m from dead load and R / 5 kN-m from live load. Using C
. — 20 concrete and Grade 420 steel, design the section for:
(i) Minimum depth obtained by maximum permissible steel.
(ii) Effective depth of R mm.
(iii) Foran overall depth of 1.252? mm using trial and error method.

Q.3. Determine the cross-section of concrete and area of steel for a simple reinforced
concrete beam of R / 100 m span to carry a computed dead load of R I 20 kN/m and
service live load of R I 30 kN/m. Use C - 20 concrete and Grade 280 steel. Also
prepare the bar bending schedule.
b) Longitudinal Section
L Q.4. In a slab system, longer beams have a single simply supported effective span of R 1 125
Fig. 3.28. Reinforcement Details of Beams for Example 3.7. - m and shorter beams have three spans of R I 200 m each. Slab thickness is R I 6 mm
and floor finish consists of 75 mm of brick ballast and 50 mm of floor finish. The
$ longer beams support a 228 mm thick and R / 300 m high brick wall. The structure is
Spacing between bars = nrHL (OK) g to be used as an office building. Use C-20 concrete and Grade 280 steel. Selecting US
2 =
bars, design the interior longer beam having a rectangular section with width of 300
Provided steel ratio 2 02 mm for flexure under the following conditions:
= -9x15.5 = 0 0145 >\
a. Minimum depth obtained with maximum steel ratio, even if it violates the ACI
Pmax (OK) minimum depth for deflection control.
b. Overall depth equal to £/12.
c. Overall depth of £/10 mm using trial and error method. Also prepare bar bending
schedule for this option.

Q.5. Design a rectangular singly reinforced flexural member to be used as an interior simply
supported short beam. The slab system has a longer span of R I 160 m and a shorter
span of R I 200 m. Factored slab load is 10.00 kN/m2,^'= 17.25 MPa and fy = 280
11 8 CONCRETE STRUCTURES PART - 1
AUTHOR- ZAHID A. SfDDlQj
MPa
- teHtectaaUy,
-
to ™.
Chapter - 4

SHALLOW, TEE AND CONTINUOUS BEAMS

4.1 DOUBLY REINFORCED BEAMS

Doubly reinforced beam is defined as the beam that has both tension and compression
reinforcement to allow the depth of beam to be lesser than the depth required for the
corresponding singly reinforced beam. The comparison with singly reinforced beams is as
under:
1. Sometimes, the beam depth is limited because of the architectural or other
considerations as compared with the depth required for a singly reinforced section.
These shallow depth beams cannot develop compression force required to resist the
given bending moment. This not Ohly makes the beams having under-strength but also
show over-reinforced and brittle failure. In such cases, compression steel is placed to
improve the compressive strength of the beam. An approximately equal amount of
steel is also provided on the tension side to balance the extra force developed in the
compression steel. Hence, the overall moment of resistance is increased and the
behavior of the beam remains ductile same as that of the under-reinforced sections. The
overall cost of the beam may be increased if the cost of steel is much greater than the
cost of the concrete.
2. Compression steel increases ductility of the beams by making the beams under¬
reinforced and also by improving the compression behavior of the surrounding
concrete. Larger plastic deformations are produced at the plastic hinges produced near
the failure loads. This provides proper warning before failure and allows the
redistribution of stresses and the use of reserve strength of other less-stressed sections.
3. Hanger bars are used in most of the beams to support the shear reinforcement during
construction and to provide resistance against some accidental negative moments.
Considering this steel as compression reinforcement in the calculations may provide
economy in design.
4. In case the cost of steel is much higher than the cost of concrete, use of shallow depth
beams with more compression reinforcement makes the construction uneconomical.
However, the savings obtained by efficient use of the ceiling heights and reduction in
overall building heights must also be considered during the economic comparison.
5. Sometimes, compression steel is provided for reasons other than strength. The
provision of minimum compression steel reduces long term deflections of members due
creep and shrinkage.
6. With the widespread use of the Strength Method of design, the percentage use of the
doubly reinforced beams has reduced. The reason is that strength design utilizes the
full strength of the concrete on the compression side of the neutral axis.

J
SHALLOW, TEE AND
CONTINUOUS BEAMS 121
120 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
1 chap^r4: compression steels ere actually
strai„ diagram is used to check whether the tension end
4.2 BEHAVIOR OF DOUBLY REINFORCED SECTIONS t
yielding or not.
There are two distinct cases for the behavior of doubly reinforced sections that are used for the *
analysis and design of such sections. These are as under: i
Case I: In this case, both the tension and compression steels are yielding at the ultimate load
level. ’
<

Case II: Both or any one of the steels may not be at yield stress level at the ultimate moment
stage.
The strain, stress and resultant internal forces diagrams are shown in Fig. 4.1. The additional
parameters used in the discussion of doubly reinforced sections are defined as under: Ultimate Strain
d' = depth of centroid of compression steel from the extreme compression face
Cr^s
Section Diagram
Cs = resultant compression in the compression steel Fig. 4.2. Strain Diagram for Doubly Reinforced Beams.
fl = compressive stress in the compression steel
s' = compressive strain in the compression steel of Fig. 4.2,
From similar triangles is ABC and FGC

=
c-<<
0.003 c
= 0.003-^"^
- a! (3)
= 0.003
° DEC of Fig. 4.2,
Also, from similar triangles As ABC and
Cross Strain Actual Parabolic Whitney’s Resultant es
_
Section Diagram Ultimate Stress Equivalent Forces 0.003 c
Block Rectangular ~
Stress Block £ = 0.003 °
Fig. 4.1. Internal Forces for Doubly Reinforced Beams. (4)
s, = 0M3^'d~a
a
4.2.1 Case I: Both Steels Are Yielding
For both the steels to be yielding, f = f/ = f, I* Method Of Checking
stresses as under:
Compressive force in compression steel C, = At'fy Calculate the tension and compression steel
Compressive force in concrete = QXSf/ba f' ~ < fy
Total tensile force T = Aify ft Es
~ X Eg fy
For equi librium along the axial direction of the beam, are less than or equal to
Now check to see that whether these stresses
Ce + C, = T f' fy compression steel is yielding
0.85£'ba+ A/fy = A,fy f>fy tension steel is yielding
a = 0.85/;* — (1)

Also, a = Qi c (2)
122 CONCRETE STRUCTURES PART -I SHALLOW, TEE AND
AUTHOR: ZAHID A. SIDDIQI CONTINUOUS BEAMS 123
2nd Method Of Checking CHAPTER 4:
4.2-2 Case II: One Or Two Steels Are Not Yielding
'

Calculate Ey and then check the strains as under:


; The stress corresponding to that steel which is not yielding may be used in terms of its
f'~ e/xEs < fy expression in terms of ‘a’. However, it is important to remember that if the steel is
fs & X Ex fy I yielding, these expressions may not give the right answer.
Now check to see that whether these stresses are
f,' > fy
f fy
less than or equal to fy.
compression steel is yielding
tension steel is yielding
;

- 6(X>AL^
a
and //- 600?^
a
(11)

(12)
formula formoment 0.85 fob
Equation-12 will be a quadratic equation in terms of ‘a’ that may be solved to get the
Moment due to compression steel, = value of ‘a’. Equation-11 is then used to calculate the stresses in the steels and the
M„2 =C,(d-d') moment capacity is determined as usual by using the following equation:
~ Afftd-d')
(5) M„ = A/f ' (d-d')+ 0.85 f 'ba (d-a/2)
T = total tensile force in the tension
or A,'f '(d-d') + (Asf-As'f')(d-aty (13)
reinforcement
Ti ~ tensile force in
tension steel to 4.3 BALANCED STEEL RATIO FOR
balance compression force in the DOUBLY REINFORCED SECTIONS
compression steel
= C, Whitney’s Resultant The following symbols may be used in the development of the expression for the balanced steel
Ti = tensile force in tension steel to Equivalent Forces ratio in case of doubly rein forced sections.
balance compressive force in the Rectangular pt = total balanced tension steel ratio in case of doubly reinforced beams,
concrete Stress Block
limiting steel ratio dividing the under-reinforced and the over-reinforced
= 0.85 f 'ba behavior
Fig. 4.3. Internal Forces Diagram
Moment due to concrete, for Doubly Reinforced Pb = balanced tension steel ratio in case of corresponding singly reinforced beams
M/ = Cc(d~a/2)
= 0.85 fc'b a (d—aty
Beams.
(6)
p' = compression steel ratio - —-
bd
Also, Total force in Force in tension steel Force in tension steel
Mr = (T-TiMd-ah) tension steel to balance concrete to balance steel
s = (T-Cx)(d~a/2) " compression + compression
~ (Asf -Ax'f/Xd-a/l) (7) t = Ti + r2
iff - f' = fythen, or
Asfy
T
=
= Cc + Cs
0.85 f'b a + Af f,
Mr = (A,-,4/)/,(d-a/2) (8) i
M„ = Mnl +M„2 t In the above equation, the compression steel is assumed to be yielding. If it is not actually
!
X yielding, adjustment is made later on.
= 0.85 fc' b a (d-a/2f+ A/ f/(d-d'^
or Divide the above equation throughout by bdfy.
(9) s
Mu = ^M where = 0.9 (10)
As = 0.85— — A’,
bd bd

Pb = 0.85 P‘ (14)
SHALLOW, TEE AND
CONTINUOUS BEAMS 125
1 24 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 4:
From the strain diagram shown in Fig.4.4 for the balanced failure, the following may be
written:

c d—Ct
0.003

0.003 d- 0.003 c
E,

<

AU
—+0.003
J=
0.003 d
Fig. 4.4. Ultimate Strain Diagram.
a 0.003 Es
(I5)
d 'V, +0.003 E,
Put this value of aid in Eq. 14:
Fig. 4.5. Strain Diagram To Calculate Compression
Steel Strain.
Pt = 0.85/1,4-7—; (SI)
For£i'>/ 0,003 - —x(0.003
:
d
+ £ ) > £y

fy fy+^,^
(FPS) d’ 0.003- e, _W0-fy
Pb = pb + p' (16) d ~ 0.003 + Ey* E, 600 + 4

The above formula is only valid if both the steels are yielding. Tension steel is always yielding 4.4 MAXIMUM STEEL RATIO FOR
at the balanced condition by the definition. However, the compression steel may or may not be DOUBLY REINFORCED SECTIONS
yielding. The effectiveness of the compression steel of magnitude As' is reduced in the ratio/’
/ fy when the compression steel is not yielding. The balanced steel ratio, for the cases where From the strain diagram shown in Fig.4.6 for the
tension-controlled failure, the. following may
compression steel is not yielding, is written as follows: be written:
d-c
Pi, = Pb + p' * (17)
0.003 0.005
A formula for/' in terms of d'/d may also be derived using Fig. 4.5 for the balanced condition
as follows:
From the similarity of triangles HBG and AJE, we get,

5
3
c

c = 3-d
HB
HG —
AE
- or HB
d’
0.003 + e
d
or HB
0.003 + £
~
d“ Xd' a
8

= ^,x-d
8
(19) Fig. 4.6. Ultimate Strain Diagram.

Now, = 0.003 -HB = 0.003 -


Putting this value of aid in Eq. 14, the following is obtained:
= Es[ 0.003 ~4x(0.003
a
+£ )]
P^ 0.85 A *
= 600- —d
x(600+/ ) Maximum value is/ (SI) (18) p = pm<a for singly reinforced sections with
any extreme strain + p
SHALLOW, TEE AND
126 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. CONTINUOUS BEAMS 127
SIDDIqi
If the compression steel is not yielding, the formula may be written as
follows:
amount of minimum steel in doubly reinforced flexural members that
This steel ratio givesinthethe compression steel. If given steel ratio is lesser than this limit, the
max
(20) will ensure yielding
compression steel will not be yielding.
4.5 MINIMUM TENSION STEEL RATIO FOR
46 ANOTHER CHECK FOR YIELDING OF COMPRESSION STEEL
COMPRESSION STEEL YIELDING
the maximum value for a tensile strain of 0.005
Consider that the compression steel is yielding at This check is valid when tension steel equal to
failure, as in Fig. 4.7. The tension steel will is provided.
always be yielding due to the requirement of Let p ~ p' = /w for singly reinforced sections
tension-controlled design. From the strain
diagram, we have,

_
= 0.85^, x '

——
8 4
for extreme tensile strain of 0.005 (24)

c 0.003 For compression steel to be yielding,


c~d’ ey &-P'Wy (25)
= = ;
CEy .= 0.003 c -0.003 <7 ^^f'cb 0.854
Fig.4-7. Ultimate Strain Diagram To
c (0.003 - ) = 0.003 d 3 f’ df 3
c = 0.003
0.003-^ “
Determine Minimum Steel
Ratio For Compression
Steel Yielding.
From Eq. 24: a = 0.85^
-^— =£ d (26)

- 0.005 is the same as that


a
- A 0.003-c
J

Summing the forces in the horizontal direction:


(21)
Equation 26 shows that ‘a’ for a doubly reinforced section with
for the corresponding singly reinforced section.
For compression steel yielding, fy I Es

j Asfy ~ O.Hfc'ba + A/fy 0.003 [1-^—1


V a
≥fy/Es
/
Divide the above equation throughout by bdfy. ( 8 I

.0.85^ + A
E, x 0.003
I
fyd bd
By putting the value of a from Eq. 21 in to Eq.22, we get,
(22)
'
^xQ.003-4 x 3
Es* 0.003 8 — (27)

600-4 3
(SI)
0.85 600 8
did <
87000-4
- Z V 3
(US Customary)
0.85 A £. [
a °-003 87000 8
fy d ( 0.003 — e The condition for yielding of the compression steel when maximum tensile steel ratio
is
to be used may be evaluated for different grades of steel as follows:
600
0.85^-^ 600-4 ?
(SO (23)
For/y = 280 MPa, d I d < 0.2
For/y = 300 MPa, d Id < 0.1875

4—
87,000 For/y = 420 MPa, did 0.1125
0.85— A. (US Customary)
4/^(87,000-4 For/ = 460 MPa, d Id < 0.0875
For/y = 520 MPa, did 0.05
SHALLOW, TEE AND
CONTINUOUS BEAMS 129
TgR 4;
128 - = 600^^^'
CONCRETE STRUCTURES PART 1 AUTHOR: ZAHID A- + 0.85 fc'b a
4.7 ANALYSIS OF DOUBLY REINFORCED SECTIONS
SODnj ^,600^^
s a a
(3)
+ (As+As')a - ^{A^A^d) =
0
Data: i) Dimensions like 5, d, rf'and £ or f~600 y Eqs. (1) and (2).
ii) .£>4 and Es
a from Eq.3 and then calculate^' and^ from
iii) Areas of tension and compression steels, Xr and As' Calculate the value of the depth of N.A.
the strength reduction factor (<))b) depending on
Required; Moment capacity, M, Calculate
and the extreme tensile strain. yielding or
The following procedure may be used to determine tile flexural capacity of a doubly reinforced strength depending upon the whether the steels are
flexural member: Calculate the flexural
not.
Check the beam as a singly reinforced section to see whether the provided compression
steel will be used for strength or not. For this purpose, pmay be checked and it must be Both Steels Are Yielding
more than for singly reinforced sections. ft ~ fs ~ ft
Mu = (Ai-A^)fy(d-a/2)+As,fy(d-d')
Ptaxx ~ 0.85^ — —— for singly reinforced sections Mu ~ Mn
Any Steel Is Not Yielding
2. Assume both tension and compression steels to be yielding and calculate the depth of
N.A. (c) and depth of equivalent rectangular stress block (a). fs’ = 600^^
a

3.
a ~
(As-A^fr
QSSf'b
Check for yielding of both the steels.
andc = a! Pi f
Ma
- 600^^
a
= (Ms-A^Md-^
Edu Edu
Es following section
rectangular section has the
£, =
d C
0.003—— —
c
Example 4.1:
properties:
A doubly reinforced

= 0.003^^ A/ = 568 mm2 [2 - #19(US)]


c As = 3060 mm2 [6 - #25(US)]
If £y => f=f b = 300mm
If Sj > £y fs ~ fy d' = 60 mm
fy = 300 MPa
Alternatively calculate p9 and make sure that p for yielding of the compression for the following two conditions:
Calculate the design flexural strength
steel. Similarly d I d ratio may be evaluated for checking of the compression steel, as mm.
explained earlier. i) C-20 concrete and d = 525
ii) C— 35 concrete and d= 225 mm.
4. If the tension and/or the compression steels are not yielding, write the expressions for
,
and/or_/s' in terms of unknown a and solve to get its new value. Solution: \

Case (i)
= 600^^ (i) A/ = 568 mm2
As = 3060 mm2
/s = 600^^
a
(2) b = 300 mm
d' = 60mm
mm
T = Cs + Cc => tension failure d = 525
SHALLOW, TEE AND
130 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI
CONTINUOUS BEAMS 131
fy = 300 MPa M = (As - As) fy(d- a/ 2)+ (d- d)
fc = 20 MPa = [(3060 - 568) x 300 x (525 - 146.6 / 2)
Es
-
= 200,000 MPa
?
+ 568 x 300 x (525 - 60)] / 106
= 416.9 kN-m
— —= =—
(3060) <|>b Mn
P = 0.0194 M
(300X525) 0.9x416.9 = 375.2 kN-m
p' = -r3— — =
(568)
- Casefii)

Pnani -
_ (300)(525)
3/'
0.85/,— ~
0.0036

for singly reinforced sections


A/
As
= 568 mm2
= 3060 mm2
= 300 mm
= 0.852x|x22. =0.0181
d' = 60 mm
8 300 d = 225 mm
p^ ~ Pmas p' assuming the compression steel to be yielding
fy = 300 MPa
fc' = 35 MPa
= 0.0181+0.0036 = 0.0217 Es = 200,000 MPa
&M„ = ?
p* = 0.85^^/,f—^L_]+ P
/ dP\600-/J
_ °-85 60 f 600 ^36 P = = 0.0453

p
___2 20
xwxmJx|.6oo-3ooJ+
)
-
> Paas for singly reinforced section, analyze as doubly reinforced
00146
p' =
(300)(225)
^68)-
(300)(225)
= 0.00841

section. For/' > 28 MPa, # = 1.05 - 0.00714/' 0.85


p 2 P^ compression steel is expected to be yielding. = 0.8
3 f
p tension steel is expected to be yielding and = 0.90. Paas = 0.85/, for singly reinforced sections
’ 8A
Considering both the tension and the compression steels to be yielding, calculate depth
of equivalent rectangular stress block (a) and the depth of NA. (c). The yielding
tension and compression steels may then be confirmed.
of
3
= 0.85 x 0.8 x-x
8 300 —
35
= 0.0298

assuming the compression steel to be yielding


-
a
(Ar-A’)f
= 0.85
//6 —
(3060- 568) x 300
0.85 x 20 x 300
= 146.6 mm
Pm„
=
A«r + p'
0.0298 + 0.00841 = 0.03821

c = ai = 146.6/0.85 172mm — Pb 0.85 A


600

Sy

ft
=
= 0.003
fy/Ez

c
= 300 / 200,000 = 0.0015

= 0.003 525-1
172 —
= 0.00616
0.85 x 0.8 x 21x

d'f
(
6°- +
300 \ 600 +300/
600
-) 0?00841 - 0.0613

> ft and 0.005, tension steel is yielding and = 0.90. Pty 0.85^ —/J 600- f,
+ P'
ft' = 0.003^— = 0.003 ^-22 = 0.00195 „ 35 60 x <
600 A + 0.00841
c '
172 •
x
.

> ft, the compression steel is yielding.


.
0.85 x 0.8 x
1
300 225 600 -300 J
= 0.0507
132 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SlDD[Qi
SHALLOW, TEE AND
p > ZWw for singly reinforced section, CONTINUOUS BEAMS 133
section.
analyze as doubly
reinforced CHAPTER 4:
p < p^,, compression steel is not yielding. Mi = (Asfy-Ai fi)(d-a! 2)+As' fi (d-d)
p > 5 extreme tensile strain is not
= [(3060 x 300 - 568 x 265.12) x (225 - 86 / 2)
greater than 0.005.
p , tension steel is yielding. -
+ 568 x 265.12 x (225 60)] / IO6 = 164.5 kN-m
Considering both the tension and the compression JWu Ida
of equivalent rectangular stress steels to be yielding, calculate depth
block (a) and the depth of N.A. (c). The = 0.78 x 164.5 = 128.3 kN-m
tension and compression steels may then be

a
(4 -4)4
0.85 f'tb
-
confirmed.
(3060 568) x 300 __ _ yielding of
4.8 DESIGN OF DOUBLY REINFORCED SECTIONS
0.85x35x300 Data: i) Loads and span length
c = a! pi ~ 83.8 / 0.80 = 104.8mm ii) ^/.and-fi)
= ft! E* = 300/200,000 = 0.0015 iii) Architectural depth restriction
= 0.003 = 0.003 = 0 00344 Required: i) Beam cross-sectional dimensions such as b, d, d'and h
c 104.8
> Sy, tension steel is yielding ii) Areas of tension and compression steels, At and Ai
iii) Detailing
= 0.003 c = 0.003 104 j- 60 = 0.00128
^- The following procedure may be used to design a doubly reinforced flexural member:
104.8
< the compression steel is not yielding. 1. Calculate all types of loads acting on the structure and hence calculate the total factored
Calculation of revised value of a: load.
2. Calculate the total maximum factored moment, M„, acting on the member. Also decide
the tentative cross-sectional dimensions of the member.
a a 3. Calculate minimum effective depth required for singly reinforced section. If the
T ~ Cs + Cc => tension failure
available effective depth is lesser than this depth, design the member as a doubly
reinforced section.
A,fy = Ai ft + 0.85^'b a
3060 x 300 = 568 x 600
a
+ 0.85 x 35 x 300 x a d •
= I
yO.ZOSftb
918,000 a = 340,800 a- 16,358,400 + 8925 a2
4. Calculate the maximum moment that can be resisted by the tension-steel according to
a2 -64.67 a- 1832.9 = 0 the maximum steel ratio for singly reinforced section for an extreme tensile strain of
a = 86 mm 0.005.
c ~ al fa = 86/0.80 = 107.5 mm P = Paa* - 0.85 X 0.375/71 —4
= 0.003-^-^
107.5 Asi ~ Pnax bd
> Sy = 0.0015, tension steel is still yielding a = P^td
= <|>bMi
& = o.65 +
— 0 25
0.005-^
- - (Ej- fy) = 0.78 !
Mi = ^Asify(d-a/2)
In the above expressions, Mi is the moment capacity provided by the maximum
compression in the concrete and Asi is the area of tension steel to balance the maximum
compression developed in the concrete according to the code allowed value of extreme
ft = 600 = 600 8-- 4- = 265.12 MPa ! tensile strain.
SHALLOW, TEE AND
CONTINUOUS BEAMS 135
134 CONCRETE STRUCTURES PART - 1 AUTHOR Z AHI P A «IDDIQI
chapter41
5. Calculate the excess moment (Mi) that must be resisted by the moment capacity i) P'P’ = 0.5 Pa
provided by the compression in the compression steel.
ii) Area of compression steel is minimum
= M„-Mi be
Mi ratio, the amount of compression steel will
Note: Keeping maximum tension steel
6. Assume that the compression steel is yielding (fs -
fy, to be checked later).
the amount of compression steel for the above assumption.
Calculate the minimum.

A' = Mz = 6m
I
wB - 70 kN/m
The total amount of tension steel may be calculated by adding an extra amount of b = 300 mm steel from the extreme compression face
tension steel equal to the above required compression steel in to the already calculated Assume the depth of centroid of compression
tension steel. to be:
d' = 58 mm
As — Asi +A&
Calculate the tensile steel ratio and check for yielding of compression steel by any
fy = 420 MPa
// =20 MPa
method.
'
= ZZ 12 = 6000/12 = 500mm
6000/16 = 375 mm
"

4, Amin for deflection control = ^/16 =


~bd hsd = 500 mm
d = - 75 = 425 mm
0.85/^-™ 1 -J ran
M .^021-315^
——
8 8
If , compression steel is yielding, as assumed before, l~M
for singly reinforced section = 4—
and amount of compression steel = As V 0.205 fb ~~
9. If the steel ratio (p) is lesser than p^ (compression steel is not yielding), stress in I 315x10* =
compression steel is determined from the strain compatibility. When we compensate V 0.205 x 20 x 300
the internal compressive force in steel as in the next step, the depth of rectangular stress d dnun, design as doubly reinforced section.
block (a) remains the same as in step 4.
Case (i)
600°-^'
a = 0.85 A
600
A
If the compression steel is not yielding, compression steel area (//) is increased to
-
10.
compensate the reduction in compression steel stress to get the required compressive 0.852
20 < 600 J)
= q o2O2
force. 1
420 600 +420 J

As ,revised As , trial p- p' * — 0.5 Pt, = 0.0101


Example 4,2: Design a simply supported rectangular beam having a span of 6m, subjected r As As x _ = Go-p'x w
to a total factored load including the self weight of 70 kN/m. The depth of the beam should not
exceed its span over 12 and its width is to be 300 nun. Use C—20 concrete and Grade 420 = 0.0101x300x425 = 1288 mm2
steel. Design the beam for the following two options selecting SI bars:
CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A SIDDIQI CHAPTER 4:
SHALLOW, TEE AND
CONTINUOUS BEAMS 137
P'P'x~<Pb
Jy
=> p<pa+p’*^ 2- #15
Jy
=> tension steel is yielding with sufficient margin, <|>b

{As-A'^)fy
— 0-9.
Clear cover = 40 mm
2- #25

#10@210mm C/C (assumed)


a fy 1288x420
0.85 feb
i
= 0.85 x 20 x 300 = 106 nun 3- #25
3- #20
c ~a!p\ = 106 / 0.85 = 125 mm
a) Mid-Span Cross Section
= fy/E^ 420/200,000 = 0.0021


=
< £y
0.003^-
c =
0.003^^ 125
=> compression steel is not yielding
- 0.00161

// = 0.00161 x 200,000 = 322 MPa

^-A’x^A = 1288

Multiplying the above expression with^, we


get^
A/y'^’ = 12«^ =4» A'/s’ = (A - 1288)/y b) Longitudinal Section

Mu = = ^(A-A'x y-)fy{d-al2')+ ^Asfs(d~d) Fig. 4.8. Reinforcement Details for Beam of Example 4.2 [Case (i)].
Jy
315 * 106 = (0.9X1288X420X425 106 / 2) + (0.9)(A
- Ay = p^bd = 0.0129x300x425 = 1645 mm2
- 1288X420X425 - 58) = /3] % d = 0.85 x 0.375 x 425 = 135 mm


138,726 A = 3.1257 x 108 a
A = 2253 mm2 [3425 + 3420] Mi - (|)bA4n
=

Asijy (d a / 2)
(0.9)(l645)(420)(425 - 135 / 2) / 106 = 222.3 kN-m
& = (A - 1288)/y /// = (2253 - 1288) x 420 / 322
Design For Balance Moment Capacity
= 1259 mm2 [2425 + 2415]
The reinforcement details are shown in Fig. 4.8. = 315.0-222.3 = 92.7 kN-m
Assuming compression steel to be yielding,
= 58/425 = 0.1365 > 0.1125, compression steel will not
be yielding A'
^fy(d-d')
Moment Capacity As Singly Reinforced Section 92.7 x10s
p = /W = 0.85 *0.375/3]
—A
= 0.85 X 0.85 X 0.375 x (26 / 420) = 0.0129
' .

4j2
(0.9)(420)(425-58)
= A' = 668 mm2
= 668 mm2

— Ai T-
= 1645 + 668 = 2313 mm2 [3 #25 + 3-#20]
138 CONCRETE STRUCTURES PART - 1 SHALLOW, TEE AND
AUTHOR; ZAHIDA. SiDDlQ1 cHAPTER4: CONTINUOUS BEAMS 139
a = 135 mm (as calculated before)
= 600
a
= 600 -- 135 = 380.9 MPa
< fl, compression steel is not
yielding
revised - ^s', trial —~

= 668 x 420 / 380.9 = 737 nun2 [3-#20]


The corresponding
reinforcement details are shown in Fig. 4.9.
Fig. 4.10. Typical Tee and Ell Beams.

Following notation is used to study the behavior of T-beams:


3- #20
Ar or b = width of flange or the portion of slab acting with the beam
#10@210mm C/C (assumed) hf = thickness of flange / slab
= width of web / rib
3-#25 and h = total depth of beam including the flange.
3 -#20 It is to be noted that if the beams of Fig. 4.10 are subjected to negative or upward bending, the
a) Mid-Span Cross Section flange concrete will be in tension and is to be ignored while studying the behavior of such
beams. In such cases, the same beams will behave as having rectangular sections of width
_^3-#20 3- #25 equal to b^ and total depth equal to h.
>
#10@210mm c/C 4.9. 1 Effective Flange Width (b) [ACI 8.12]
3 -#25 + 3- #20 The compressive stresses developed in the slab are the maximum at the top of the web
and these gradually reduce towards the panel centerlines. , In place of considering the

300

Fig. 4.9.
—— 5700
b) Longitudinal Section 300
actual variation of stresses, it is convenient to consider uniform stresses with a reduced
width so that area of the stress diagrams in both cases are approximately the same. The
ACI Code specifies these width values as under:
T-Beam
Reinforcement Details for Beam of Example 4.2 [Case The effective flange width, b, is the minimum of the following three dimensions:
(ii)].
4.9 TEE AND ELL BEAMS i) t / 4 where I = span of the beam
ii) 16hf + bw
In case slab and the supporting iii) S = bw + one-half the clear distance to the next web on both sides
compression is produced beam are monolithically cast together,
towards the slab side, a portion of the slab acts such that the For regularly spaced beams, 5 = center-to-center spacing of the beams.
The portion projecting from the as part of the beam.
the beam is called flange of
slab is called web, rib or stem, and
the resulting beam. If slab is present the slab portion acting with L-Beam
beam is called T-beam, and if the slab on both
(Fig. 4.10).
is present on one side only, the sides of the web, the The effective flange width, b, is the minimum of the following three dimensions:
beam is called L-beam
SHALLOW, TEE AND
CHAPTER 4: CONTINUOUS BEAMS 141
140 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI vii) Minimum cover for in-situ joists, not exposed to weather or in contact with ground,
using bars lesser than or equal to #35 is to be 20 mm (0.75 in).
i) 11 12 where = span of the beam
ii) 6Af+&w
iii) 4.11 FLEXURAL BEHAVIOR OF T-BEAMS
b* + Sc / 2 where Sc = clear distance to the next web
Isolated T-Beam Case -I: Flange is on the tension side.
The following requirements must be satisfied while proportioning the isolated
T-beams: The analysis and design of such a beam is identical to a rectangular section of width, bw, and
i) depth of the T-beam including the thickness of the slab.
hf bv/tl
ii) 5f < 4hw Case -II: Neutral axis at failure lies within the flange and slab is on compression side.
The depth of neutral axis (N.A.) is calculated for a rectangular section having width equal to
4.9.2 Requirements For Flange Reinforcement flange width, b, and depth equal to the total depth of the T-beam. If this NA. lies within the
The reinforcement in the slab to act as flange of T-beam perpendicular to the beam is to flange (Fig. 4.12), the flange alone can provide the full compressive resistance required for
be provided on the top (if not already available as regular slab reinforcement) balancing the tensile force in the tension steel. The concrete on the tension side is neglected in
depending on the following criteria: all beam designs. Hence; the presence of less or more dimension of concrete on tension side of
i) This reinforcement is to be calculated for the overhanging slab portion to act as NA. does not change the flexural capacity of the member. The beam under such conditions
may be analyzed and designed as a rectangular beam of dimensions b x h.
flange taken as cantilever carrying the factored slab load.
ii) The spacing of this reinforcement is not to exceed 5hf not 450 mm (18 in).

4.10 JOIST CONSTRUCTION

Joists are usually smaller size T-beams (Fig. 4.11) that are closely spaced and provide in one or
two mutually perpendicular directions. The following requirements must be satisfied for the
joist construction:

*
~\r Fig. 4.12. T-Beam with Neutral Axis within Flange.

Case -III: Neutral axis lies outside the flange and slab is on the compression side.
Fig. 4.11. Cross Section through Typical One-Way Joist Construction.
In case the neutral axis, determined on the basis of b x h rectangular section, lies outside the
i) Minimum width of ribs - b^in = 100 mm (4 in). flange, the compressive resistance is provided in two ways. Firstly, the overhanging flanges
produce their full capacity denoted by Cf in Fig. 4.13 and, secondly, the web develops its
ii) Maximum depth of ribs = 3.5 hw^in.
compression carrying capacity denoted by Cw.
iii) Maximum clear spacing between ribs = 750 mm (30 in). Cw = compression developed in the web portion
iv) If permanent burnt clay or concrete block fillers are used, the minimum slab = 0.85 fc' bw a
thickness over fillers is kept larger of the following two: Cf = compression developed in the overhanging flanges
a) '/]2th the clear distance between the ribs =0.85/c'^4(b-bw)
b) 40 mm (1.5 in) Total compression = C = Cw + Cr
v) The concrete shear strength is considered 10% = 0.85/c'bwa+ 0.85/c' AAf(b- bw) (1)
more than given by the ordinary
formula. 4sw = area of tension steel to balance compression in the web
vi) In one-way joists, the minimum temperature and shrinkage reinforcement must be Aa = area of tension steel to balance compression in the overhanging flanges
provided in the slab normal to the ribs.
SHALLOW, TEE AND
142 CONCRETE STRUCTURES PART -I AUTHOR: ZAH1D A. CONTINUOUS BEAMS 143
SIDDjqj *
= A^fy (d~a/2) + Axfy (d- a/2 + a/2- d')
^A^fy (d— a/2 + a/2 — pxhf! 2)
= (A^ + Asz + A^fy (d- a 1 2) - Axf (d'- a/2)
+ 0.85 // pxh((b- bw) (a/2 - Pi hf/ 2)
-
= Asfy (d - a/2) Axfy (d'~ a/2) + 0.85/c' Pi hf (b - bw) (a/2 - Pi hf I 2)
(9)

= area of tension steel to balance compressive force in compression steel.


4.12 TENSION-CONTROLLED FAILURE OF T-BEAMS
discussing the tension-controlled
Fig: 4,13 T-Beam with Neutral Axis within Web, The following steel ratios are required to be defined before
failure:
As = total area of tension steel

Tw. =force in tensile steel to balance Cw


(2)
p* = web steel / total tension steel ratio = —— (10)

Tf -
—Asw Xjy = (As Asf) fy —
= force in tensile steel to balance Cf
(3)
pi = overhanging flange steel ratio = A* ——bvd (11)

= Asffy (4) Pb = balanced steel ratio for rectangular singly reinforced beam
Tf = Cf A^ffy — 0.85 fz' pihf(b — bv) p w, max = maximum steel ratio allowed for T-beams
Pwa. = maximum steel ratio for rectangular singly reinforced section
A^ ~ Q.85pthf^-(b-bw)
J y
(5) and at = limiting value of ‘a’ producing tension-controlled failure
~ Cy/
Tension-controlled sections are those at which the net tensile strain in the extreme
Tvi ^swfy = 0.85 fc O tension steel is equal to or greater than 0.005, when the corresponding concrete strain in
compression just reaches a strain of 0.003. The derivation of expressions may be
a — ^f'bw
(6) performed just like for singly reinforced sections considering Fig. 4.14.
Lever ann between Cf and 7f = d-Pih(/2 Let dx = depth of steel closest to
tension face from the
Lever arm between Cw and Tw - d- a f 2 compression face.
Moment of resistance depending on compression forces is: and Ea = strain in steel closest to the
Me = Cw(d-a/2) + Cf(d-/W2) tension face.
= 0.85ft'bwa(d-a/2) + 0.85f'pi h^b-b^d- pxh(/2) (7)
=
Moment of resistance depending on tensile forces is: 0.003 c
Mt = -Wnw +Mif If this strain is greater than or equal to

-
= Tw(d-a/2) + Tf(d-/?ihf/2)
(As-A^)fy(d-al2) + AsSfy(d-pxh(/2) (8)
Moment capacity in case of T-beams having compression reinforcement (although less
0.005, the section is tension controlled.
The following simplification may be
performed to get other useful results:
Diagram
Fig. 4.14. Ultimate Strain Distribution.

common) may be written as follows: 0.003 - > 0.005


c

= A^fy (d-a/2)+ Axfy (d-d') + A^fy (d- A hf / 2)


144 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQj I SHALLOW, TEE AND
I chapters continuous beams 145
0.003 > 0.005 + 0.003
c 1 Case-ii:
— —
/8
c 3
or —
c
d,
-83 <■

If c Af, treat the beam as a rectangular beam of width ‘b’ and accordingly calculate the
capacity of the beam. This next part of this flow chart is only to be used for Case-iii.
c = ~d
8
Case-tit:
i If c > A& analyze as T-beam by following the following steps:
at = Px~d (12)

1 3. Calculate the area of tension steel to balance the compression in overhanging part of
o
flange.
According to ACI R 10.3.3, for beams with compression reinforcement or T-beams, the effects
of compression reinforcement arid flanges are automatically accounted for in the computation
of the net tensile strain,
Aa = 0.85AAf^(6-M
A
For equilibrium, , 4. Calculated as follows:
T ~ C^ + Cf
or —
T Cw + 7f
Asfy = 0-85/' bv at + Asffy
Asw ~ ~Asf

5. Calculate ‘a’ for the T-beam as follows:


Putting the value of at and dividing by bv dfy, the following is obtained: i _
0-85/; bv
= 0.85 A
b,.d f, 8 bwd 6. Check that the tension steel is yielding and the extreme tensile strain is greater than or
P W, max = P ma* for rib alone + /2T
i equal to 0.005.
(13)
The minimum steel ratio is the same as for the rectangular section and is already known. ! If a < at, the limiting tensile strain is produced in the steel and <1% = 0.9.
7. Calculate the flexural capacity as follows:
4.13 ANALYSIS OF T-BEAMS
= faAsw£(d-a/2)+ faA!ffy(d-/hhf/2)
Paia: i) Dimensions like b, d,bv, hf and I
Example 4.3: A T-beam has the following data:
In place of c/c spacing of beams and span of beam may be given
ii) /^/andX
iii) Area of tension steel, A3 b =800 mm
d = 450 mm
Required: Moment capacity, fa M„ bv = 350 mm
The following procedure may be used to determine the flexural capacity of a tee or ell shaped hf = 125 mm
flexural member: 1
Use C-20 concrete and Grade 420 steel.
Casei: ! Slab is on the compression side. s
1. Design the beam as a rectangular section if the slab is in tension (Case-i). Calculatethe I Calculate the design flexural strength for the following two cases:
effective flange width, b, if the slab is in compression. | i) As = 3900 mm2
2. Assume N.A. to be lying within the flange and tension steel to be yielding. Calculate !
I
ii) Aj = 3000 mm2
depth of equivalent rectangular stress block (a) and depth of N.A. (c). Solution:
I

a Case (i)
= and c = a!
M5f;b Considering the N.A. within the flange.
SHALLOW, TEE AND
146 CONCRETE STRUCTURES PART -I AUTHOR; ZAHID A. SIDDIQI CHAPTER 4: CONTINUOUS BEAMS 147
1. Select the depth of the beam depending on the same factors as are to be considered for
A/,- 3900x420
a =
0.85/; b
= —
0.85x20x800
= 120.5 mm rectangular sections. T-beam behavior is especially useful when the selected width and
depth are insufficient for design as singly reinforced section. For continuous beams, if
c = at p = 141.7mm > hf, N.A. lies outside the flange. T-beam behavior is possible close to the mid-span, the ends of the beams usually are to
be designed as doubly reinforced rectangular sections because here the same beam acts
A* = 0.85 hf (b - bj as a rectangular section. However, at these ends, compression steel is already present
as an extension of the mid-span bottom steel. Proper overlap in compression must be
20 provided to utilize this steel as compression reinforcement.
= 0.85 x 0.85 x 125 x x (800 - 350) = 1935 mm2
Certain minimum width and depth of web are also required for the efficient shear
= AS-A^ = 3900 - 1935 = 1965 mm2 design.
-^sw
_ _
a =
A„fy
-Z
0.85 —
/X
1965x420
= - 138.7
0.85x20x350
mm 2. Calculate the effective flange width, b, of the T- or L- beam if the slab is in
compression.
3. The beam is first designed as a rectangular section of dimensions b and d, assinning that
a. = p —83 d = 0.85 x 0.375 x 450 = 143.4 mm the N.A. lies within the flange. Calculate depth of equivalent rectangular stress block
at the limiting tensile strain is produced in the steel and <|>b = 0.9 (a) and depth of N.A. (c).

= $bAsvlfy(d-al'Z) + §bAitfy(d- phf/2) Assume a = px hf


= [0.9 x 1965 x 420 (450 - 13 8.7/2)
+ 0.9 x 1935 x 420 (450 - 0.85 x 125 /2)] / 106 = 573.0 kN-m
Case(ii) &&^fy
Considering the N.A. within the flange,
Calculate the revised value of a and c as follows:
AJy 3000x426 Asfyl

=
0.85/; 6
= 0.85x20x800
a! p\ = 109mm < hf
= 92.6 mm a =

0.85 /.6
,
andc = at P\
tn

Case-ii: If c hf, design the beam with a rectangular beam of width ‘b’. The next part
NA. lies within the flange and the T-beam behaves like a rectangular beam of of this flow chart is only to be used for Case-iii.
dimension b x d. Case-iii: Uc> hf, perform design as a T-beam by following the given steps:
= <^Asfy(d-a/2) 4. Calculate the area of tension steel to balance the compression in overhanging part of
= [0.9 x 3000 x 420 (450 - 92.6/2)] / 10s = 457.8 kN-m flange.
4.14 DESIGN OF T-BEAMS
Af = 0.85^^^^-^)
A
Data: i) Slab design and dimensions (hi), c/c spacing and span (£) of the beam.
ii)//,/-and£, 5. Calculate the moment capacity for the flange portion as follows:
iii) Loads or beam ultimate moments = <|>b Mf ~
M 4*b Asffy (d- phf!2)
Required: i) Beam dimensions 6. Calculate the remaining moment capacity to be resisted by the web:
ii) Area of tension steel, As
iii) Detailing of steel Mw — Ma — Mf
iv) Bar bending schedule, if required 7. Assume the value of ‘a’ equal to the already calculated value and calculate A*
The following procedure may be used to design a T- or L- beam:
SHALLOW, TEE AND
CONTINUOUS BEAMS 149
148 CONCRETE STRUC TURES PART -1 AUTHOR: ZAHID A. SIDRiq] CHAPTER 4:

"SWA = Width of slab supported by the beam =


(

11 R2 1 ,

Now calculate the new value of a using the following expression:


=[ fi-221L3.5
J
3
= 2.93 m

Factored slab load = 2.93 x 15.00 = 43.95 kN/m


^f'cbv Total factored load = wu = 43.95 + 3.82 = 47.77 kN/m
8. Revise step 7 one or two times more until A^ becomes nearly constant Calculate the w 47 77 x 5Z
total required tension steel as follows: M = 8
= 149.3 kN-m
8
As Asw + A$f following:
Effective slab width, b, is the smallest of the
~

9. Calculate the steel ratio p„ = ——


b„d
A iv)
v)
/4 = 5000 / 4 = 1250mm
16 Af + 6W= 16 x 125 + 300
+ 0.5 Scl + 0.5 Scr
= 2300mm
vi)
10. Calculate p#, max and p*. mm and make sure that p^ is within the limits. If pw p^ maXj
revise the beam dimensions. Otherwise, ‘a’ may be checked against at and beam
= 300 + 0.5 x (3500 - 300) x 2 = 3500 mm
dimensions may be revised if a > at. b = 1250 mm
Example 4.4: Design a T-beam section to be used as an interior simply supported beam of
Assuming a = Pi h{
0.85 x 125 = 106.3 mm
span 5 m. The slab panels on both sides of the beam are 5m x 3.5m. Factored slab load is to
be 15 kN/m2. fc' = 17.25 MPa, fy = 420 MPa, hf = 125 mm, bw = 300 mm and h - 575mm.
Use SI bars. The slab is on the compression side. Trial A =
&fy&-~) 0.9x420x(500--^~) - 884 mm2

i) Design the beam for the calculated loading.


ii) Design the beam if the factored positive moment is 800 kN-m. a
AA
= 0.85
_ 884x420
//6 0.85x17.25x1250
Solution.:
= 23.9 mm < /if, N.A. lies within the flange.
^y — 5 in
= 3.5 m
a! p\

Factored slab load = 15 kN/m2 Revised As = 806 mm2


h = 575mm a = 18.5 mm
d ~ h -75 = 500 mm As = 805 mm2
b„ = 300 mm 3 = 159.4 mm
ht = 125 mm at =pi-d= 0.85 x 0.375 x 500
fc' = 17.25 MPa < at, the limiting tensile strain is produced in the steel
and <J>b ~ 0.9
fy = 420 MPa a
Case (i)

Factored self weight


& of the beam = 1.2 x 0.300 x —
575 125
1000
x 2400 x
—-—
9 81
1000
I
j


= bwd = 0.00545
PV.min = 1-4 //y = 1.4 / 420 =
0.00333
= 3.82 kN/m Pw > Pwt min (OK)
,
Fig. 4.15.
R = 4 /^y = 3.5/5.0 = 0.7 As = 805 mm2 [3-#20] The reinforcement details are shown in
SHALLOW, TEE AND
150 CONCRETE STRUC TURES PART - 1 CONTINUOUS BEAMS 151
AUTHOR: ZAHID A.
SIDDIqi CHAPTER 4:
A/v AL AL
== 800.0 - 595.3 = 204.7 kN-m
—■
2
— #15 Assume a = 106.3 mm
- 204.7x10s _
= 1212 mm2
575 #10@250mm C/C assumed t-u
TnaMsw -=
2- #20
^4 (d-^) 0.9x420x(500-^)
^fy 1212x420 _ 115 7mni
1 — #20 a
o.85/x
-
0.85x17.25x300
a) Cross Section XX
A^ = 1225 mm2
2- #20 a =117 mm
Xsw = 1227 mm2
j4s “ ^sw ^sf
= 1227 + 3524 = 4751 mm2 [10-#25]
ae = ft ±3 d = 0.85 x 0.375 x 500 = 159.4 mm
300 4700 8
300
0.9
b) Longitudinal Section < at, the limiting tensile strain is produced in the steel and

Fig. 4.15. Beam Reinforcement Details for Example 4.4 (i). The reinforcement details are shown in Fig. 4.16.
Case (u)
Afu = 800 kN-m
Assuming a = fah{
= 0.85 x 125 = 106.3 mm

Trials =
AZ/rf-y)
= ——
800x10s
0.9x420x(500-^)
4736 nun2

_- _ X A
a) Cross Section XX
a
— AJy
0.85// b
4736 x 420
0.85x17.25x1250 =
.....
108.5 mm
6 - #25 -#15 .
I 1175 2- #15
+ 2-#20
c - al fa = 128 mm > Af, N.A. lies outside the flange.
#10@250mm C/C
d
= 0.85 fa ht^tb-bj
= 0.85 x 0.85 x 125 (1250 - 300)

6 #25 + 4 #25
kx — 4-4
= 3524 mm2 4700 300
300 b) Longitudinal Section
M = 4>b Mif = <|»b {d- fa hf 12)
Fig. 4.1 6. Beam Reinforcement Details for Example
4.4 (ii).
= -
0.9 X 3524 X 420 x (500 0.85 x 125 / 2) / 106 = 595.3 kN-m
SHALLOW, TEE AND
CONTINUOUS BEAMS 153
4-.
152 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDiqi CHAPTER
1.2D + 1.6L 1.2D + 1.6L
300-2x40-2x10-4x25
Space between bottom layer bars = = 33
= (OK for 19 mm aggregate)
4.15 CONTINUOUS BEAMS

Continuous beams are those beams that have two or more spans built monolithically. These
beams are indeterminate and require detailed procedures for plotting shear force and bending
moment diagrams. The beams in frames are to be analyzed for various load combinations of
different loads such as dead, live, wind and earthquake loads. For each combination of these
loads involving live load, pattern loading is to be considered to get maximum force effect at
any point. In pattern loading, live load may be applied in adjacent panels or alternate panels in
order to produce critical magnitude of a particular moment or shear. This means that the
analysis is to be performed many times for each combination involving different pattern loads.
Maximum values at various points are then picked out of these results. The details of pattern 1.2D+1.6L 12D+1-6L
loadings are given in a later chapter but a brief view is presented below.
According to ACI 8.11.2, it is allowed to consider only the following two arrangements of live
loads: - .

a. Factored dead load acts on all spans with full factored live load on two adjacent spans. Pattern Loading for Af bc andW de
This arrangement gives the maximum negative moment at the central support. 1.2D + 1.6L 1.2D + 1.6L
b. Factored dead load acts on all spans with full factored live load on alternate spans. This
arrangement gives the maximum positive moment within the fully loaded spans.
The actual number of loading arrangements becomes greater when the adjacent two-span- Af max
loaded condition is applied for each support. Similarly, the alternate spans loaded condition is Pattern Loading for Me
reversed once to load those panels that were previously without any live load. Some examples
of application of live load to get maximum force effects at selected points are shown in Fig.
4.17.
It is to be noted that to get the influence line diagram with negative moment ordinates on the
lower side of the beam, bending moment and rotation at the section are to be applied in the ILDForMc
positive direction. 1.2D + 1.6L
1 2D + 1.61.
The negative moments at the supports obtained by the analysis are the centerline moments that
are very high in magnitude but quickly diminishes away from the supports. The design based
on these moments will be unrealistically conservative because the support regions are almost M max
infinitely strong. The ACI Code allows the design of beams at the supports to be based on the Pattern Loading for Mb
moments at the edges of supports, which are to be calculated for each load combination and
pattern. The negative moments calculated in the above way are still much higher in magnitude
than the positive moments. This makes the design of prismatic beams somewhat more
difficult. Selecting the beam size for negative moments make the beam uneconomical for the
larger part, which is subjected to positive moments. Similarly, the size selected for positive ILD For M b
moment is usually insufficient for negative moment, making the design uneconomical and Force Effects.
causing congestion of steel reinforcement. I Fig. 4.17. Pattern Loading to Get Maximum

II
SHALLOW, TEE AND
154 CONCRETE STRUCTURES PART - 1 CONTINUOUS BEAMS 155
AUTHOR: ZAHID A.
SIDDiqi cfiAPTER4:
ACI Code allows a redistribution of moments up to 20% in
which the negative momeiits
be reduced and the positive moments are increased accordingly. Hence, in m
and practically design a continuous beam, one has to consider order to accurately
all of the following aspects: *
> Load combinations.
> Pattern loading.
> Both directions of wind and earthquake loads.
> Correction of negative moments at faces of supports.
> Redistribution of moments.
In place of all the above detailed procedure, ACI moment
coefficients may
associated limitations are satisfied. It is important to note that if all the above be used if the
followed for analysis of beams and frames, the ACI coefficients will give procedure is not Fig. 4.18. Critical Sections for Moments in Continuous Beams.
realistic design as compared insufficient analysis. The general
more accurate and
concept of solving the frame by
computer only once for all the expected loads applied together using a
single combination to Than
get accurate results is incorrect and may lead to unsafe and
uneconomical design. Table 4.1. Moment Coefficients for Slabs Having Spans Lesser
3.0 m OR Beams Having Ratio of Sum of Column Stiffness
The continuous beam, with monolithic construction of beam
designed as a T-beam for positive moment at mid-spans and
with slab on top of it, may be to Beam Stiffness More Than 8 at Each End of the Span.
as rectangular section (singly or
doubly reinforced) at the supports.
1. Negative moments at all supports, integrally built with
beams.
XwX
12
2

4.16 ACI MOMENT COEFFICIENTS FOR CONTINUOUS BEAMS


2. Positive moment in end panel. X
14
2

4.16.1 Conditions For ACI Approximate Analysis To Be Applicable


The ACI moment coefficients for continuous beams are only applicable if 3. Positive moment in central panels. X Wu^, 2
the following 16
conditions are satisfied:
1. There must be at least two spans.
2. The adjacent spans must not vary by more than 20% of the shorter span. 4.17 APPROXIMATE CURTAILMENT OF BARS
3. Only uniformly distributed loads are applied. For other by using moment envelopes and
loads, separate analysis and In place of locating the actual cut-off points of bars of the bars may be
coefficients must be used. development length evaluation procedures, approximate but safe curtailment
4. The live to dead load ratio must not exceed 3. of reducing the steel where
carried out by using thumb rules. There are two different methods where these are not
5. All the beams must be prismatic. portions
it is not required. The first method is curtailment of bars in are originally
these
6. The beams must have an extreme tensile strain of
0.0075 at the negative moment required for strength or serviceability on the same face of the beam where positive bars are
sections to allow redistribution of moments. bars, where the
placed, Fig. 4.19(a). The second method is the use of bent-up
7. The beam must be present in a braced frame without significant act as negative reinforcement, Fig. 4.19(b). Further, the bent up
moments due to lateral bent up near the supports to bent-up is present
loads. bars can increase the shear capacity of a local area in which the actual
8. These coefficients only give moments due to gravity loads.
in simple members
According to ACI 12.11.1, at least one-third of the positive reinforcement
4.16.1 Moment Coefficients must extend along the same
and one-fourth the positive reinforcement in continuous members curtailment methods,
The ACI Code moment coefficients are given at six critical sections along face of the member in to the support However, to use the approximate
the length of steel in to the support.
a constant span continuous beam, as shown in Fig. 4.18. These design engineers prefer to extend one-half or two-third of the positive at least 150
sections are inner face must extend
of exterior support, exterior face of first interior support, interior face
of first interior
According to the code, such reinforcement carried in to the supports it in case of
to utilize
support, face of interior support, exterior / end panel mid-span and interior
mid-span. mm in to the support. However, it is better to fully anchor this steel
The moment coefficients for one-way slabs and beams are given in Tables 4.1 hogging due to lateral loads or as compression reinforcement.
and 4.2.
SHALLOW, TEE AND
156 CONCRETE STRUCTURES PART -1 CONTINUOUS BEAMS 157
author 4:
zahida..^ CHAPTER Design the first
Table 4.2. Moment and Shear Values for Beams and the columns having more than two spans in each direction.
over
directionslong beam (Sz) if the factored slab load is 12.00
kN/m2, thickness of slab = 125 mm,
Slabs Having Spans Greater Than 3.0 m. interiorconcrete, Grade 420 steel, and live load 3 x dead load. Select US customary bars.
I C-20
1. Positive Moment 1 5 if
4 / 4, may be reduced to 0.3 4, 50% out of this steel may be
the end is not monolithic with RC curtailed at 4 / 4, provided this distance
End spans: column (not ACI provision). is more than 4 (not ACI provision).

If discontinuous end is unrestrained 1 7 1

If discontinuous end is integral with the support I


2 150 mm extension without
4 /20 if 1/3"* steel is overlap for non-perimeter beam
1 curtailed, no curtailment for
Interior spans: 0
steel more than 1/3 (not
ACI provision).
2. Negative moment at exterior face of first interior support 90° hook at least for one-fourth positive moment steel.
span for curtailment of
Two spans: j 4 = clear distance of the respective two adjacent spans for
9
^n2 bottom bars and larger of the
the top steel.
More than two spans: J wu42 a) Curtailed Bars for Beams
id
3. Negative moment at other faces of interior supports

(4 m no. 3 is the average of clear spans of the two


adjacent panels.)
—1 ,2
^4
»

c
4. Negative moment at interior faces of exterior supports
for
members built Integrally with their supports: b) Curtailed Bars for Slabs, Distances are Same as for Beams.
The support is a spandrel beam or girder: 2 Larger of top additional and bent-up
4 / 4, may be reduced to / 5 steel may be curtailed at 4 / 3 while
The support is a column: 1 ;2 if the end is not monolithic with the other may be curtailed at 4 / 4,
~ ™u4u RC column. not less than tj (AC! value is / 3).
16 °
The support is not monolithic:
5. Shear in end members at first interior support
Zero L
1.152^
2
6. Shear at all other supports wu£n
2 („17 if less than 50%
4/4 if less than % of steel is bentupup,
of the steel is bent 14). more than 50%must not be bent for
In case adjacent spans are different, the distance from which bent-up
considered according to that clear span from where the bars originate.
or curtailment starts is approximate detailing.
However, tire extension may be reduced to 0.224
of the bars on the top in both directions is provided according to
the larger of the adjacent clear Note: For slabs, the distances, 4/4 and 1^3 for top bar extensions
spans. may be reduced to 0.224-
and 0.34, respectively. Similarly, bottom 4/4 distance
c) Bent-Up Bars
Example 4.5: A roof system consists of
228 x 228 mm columns at a spacing of 5m and
3.5m in two mutually perpendicular directions, as shown in Fig. Fig. 4-19. Approximate Curtailment of Steel in Beams
and Slabs.
4.20. Beams run in both
SHALLOW, TEE AND
158 CONCRETE STRUCTURES PART -I 4: CONTINUOUS BEAMS 159
AUTHOR: ZAHID A. SIDDIQI CHAPTER
Solution: jo^Factored Bending Moments
4 = 5m
4 = 3.5 m 4= 5- 0.228 = 4.772 m
Factored slab load =12 kN/m2 w ^2 40.5x4.7722 = 57.7 kN-m
Exterior support Mu =
= 228 mm, equal to the size of the column 16 16
Af = 125 mm Exterior span Mu+ = =
40.5x4.7722 = 65.9 kN-m
// =20 MPa 14 14
fy = 420 MPa w f2 40.5x4.7722 =
Beam Bi is to be designed. i First interior support Mu~ = 92.3 kN-m
10 10

=
w 40.5x4.7722 = 83.9 kN-m
Interior support Mu
11 11
228x228 mm Columns 40-5x4.772 2
Interior span Mu =
16
= 16
= 57.7 kN-m

More than 2 @ Selection Of Beam Depth


3.5 m Each Many different options are available for sizing the beam. One option may be to size the
beam as a singly reinforced section for exterior span positive moment and then design
as doubly reinforced section for negative moments. Second option may be to
proportion the section for maximum negative moment, which may significantly
! More than 3 @ 5 m each increase the beam depth. Third option may be to select dimensions in-between the
earlier stated two options.

Slab Load
Fig. 4 20. Beam Layout for Example 4.5. Minimum depth of beam for deflection control for exterior panel / 18.5 —^
= 5000 / 18.5
= 270 mm
Total factored load: = 12.00 kN/m2 ^/12 = 5000/12 = 417 mm
Approximate Self Weight

q gi AT,
Factored dead load = 1.2 x 2400 x 0.228 x (5/12 - 0.125) x = 1.9 kN/m dmm for Mu in the exterior panel 0.205 x/Jxh
1000
Equivalent Width Of Slab Supported By Beam
B^ ' 65.9x10*

= 266 mm
4 = 5m : 4 = 3.5m : R = = 3.5/5 = 0.7 0.205x20x228
Equivalent slab width supported = (1 ^ R2/!) hmin dmin + 75 341 mm
= (1 - 0.72 /3) x 3.5 = 2.93 m ' AT,
Factored Slab Load Acting On Beam dmin for -Mu , max ?
0.205 x f^b
Factored slab load on beam = 1.1 x width of slab x slab load per unit area 1
= 1.1 x2.93 x 12.00 = 38.6 kN/m 92.3x10* = 315 mm
Note: First interior beam will have to support 10% more width than the interior one. 0.205x20x228
Total Factored Load hmin = dmm + 75 = 390 mm
wu = 38.6 + 1.9 = 40.5 kN/m Let h = 4 x 75 + 125 = 425 mm => d ~ 425 - 75 = 350 mm
SHALLOW, TEE AND
I CONTINUOUS BEAMS 161
160 CONCRETE STRUCTURES PART -I AUTHOR: Z AHID A. SIDDIQI ’ CHAPTER 4;
Maximum Capacity As Singly Reinforced Rectangular Section At Support
For negative moment at support,
0.0075 may be used.
p™^ corresponding to an extreme tensile strain of
i As —
=. 0.9 x 420 x (350 -10/2). = 505mm2
Design For Positive Moment In Interior
Span
> A^ [2-#16 + 1^13]

/W = 0.85 Pi
2 f’
7 fy
7
= 0.85 x 0.85 x-x—
7 420
20

= 0.00983 x 228 x 350 = 784 mm2


- 0.00983
Mu = 57.7 kN-m
Assume a = 10 mm
57.7x10s 2
^si "
Psnm bd a = , t = 0.9x420x(350-10/2) = 442 mm
A^f, 784x420 s . a
a =
0.85/J b
=
0.85x20x228
= 85 mm
Asfy 442x420
=
__
8.7 nun
Mi = = <*b^4siA (d - a/ 2) = 0.9 x 784 x 420 x (350 - 85 /2)/ 10s a = 0.85/; b
= 0.85x20x1250
= 91.1 kN-m
57.7 x10s 442 mm2 > Amin [2-#13 + l-#l 6]
I As = =
Effective Flange Width For T-Beam Behavior 0.9x420x(350-8.7/2)
The effective flange width, b, is the minimum of the following three dimensions: Increased to 2-#16 + 1—#13 for compatibility.
i) f/4 = 5000/4 = 1250mm Design For Negative Moment At Exterior Support
ii) - l6Af+bw = 16 x 125 + 228 = 2228 mm = 57.7 kN-m
iii) S — center-to-center spacing of the beams = 3500 mm
M„~
The T-beam will act like a rectangular section of dimensions
228 x 425, because the
b = 1250 mm
flange comes under tension.
Minimum Steel Ratio
Anin = 1.4// = 1.4/420 = 0.00333
57-7x10^ = 2.066 MPa
= 228x3502
bd
A, min = 0.00333 x 228 x 350 = 266 mm2
p ~ 0.0059 < p™ (OK)
Design For Positive Moment In Exterior Span

For positive moment, the flange of the T-beam will be in compression. However, it is | As = 0.0059 x 228 x 350 = 471 mm2 > As,mm
more likely that for this smaller moment the N.A. will lie within the flange. The beam Design For Negative Moment At First Interior Support
will act like a rectangular section of dimensions 1250 x 425 mm. = 91.1 kN-m
। = 92.3 kN-m
M," >
Assume a = 75 mm Design as a doubly reinforced rectangular section of dimensions
228 x 425.
65.9x10s 558 mm2
A3 = ! Let d* = 60 mm

a
A,fr
=
0.9x420x(350-75/2)

558x420
= 11 mm
Mi — —
Mu Mi
= 92.3-91.1 = 1.2 kN-m
'

-
0.85/; b 0.85x20x1250 '
_
Assuming compression steel to be yielding,
_ ——— —
< = 106 mm
= __6^9xl(X__ =
0.9 x 420 x (350 -11/2)
Assumption is correct.
506 mm2
,, =
,4s

Aa
hf,
M2* -
——
(d-d')
= A^ = 11 mm2
=:
1.2x10s
(0,9)(420X350-60)
11 mm

506 x 420
0.85x20x1250
= 10 mm
; As ,4s] + As2

= 784 + 11 = 795 mm2


SHALLOW, TEE AND
CONTINUOUS BEAMS 163
162 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDlQj CHAPTER4:
a = 85 mm (as calculated before)

= 600^^-
a
-— o(JU
85-0.85x60


A' revised
•As,
85
240.00 MPa < fy, compression steel is not yielding
= 4', trial 4
= 11 x 420 / 240 = 20 mm2
Design For Negative Moment At Interior Supports

Afu~ = 83.9 kN-m < <|)bMn = 91.1 kN-m


Design as a singly reinforced rectangular section of dimensions 228 x 425. Longitudinal Section

83.9x10^ = 3 Q04
= 228x3502
M2
P = 0.0090
A = 0.0090 x 228 x 350 = 718 mm2
Top steel available at exterior support = 3-#13 (387 mm2)
Extra steel required - 471 -387 = 84 mm2 [1-#13]
Top steel available at first interior support = 24*13 + 1-#13 + 14*13
(516 mm2)
Extra steel required = 795 - 516 = 279 mm2 [2-#16]
2-#19 + 1-#16
Bottom steel available at first interior support (exterior face)
= 2-#16 (398 mm2)
Extra steel required -
= 20 398 = Code Minimum
Bottom steel available al first interior support (Interior face)
= 24*16 (398 mm2)
Extra steel required = 0 - 398 = (Nil)
I
Top steel available at interior support 2-#13 + 2-#13 (516 mm2)
Extra steel required 718 -516 = 202 mm2 [2-#16]

Bottom steel available at interior support = 2-#16 (398 mm2) 90° Hook
Extra steel required = 0-398 = (Nil) Using Curtailment
Better Reinforcement Details
Example 4.5.
The reinforcement details, using bent-up bars, are shown in Fig. 4.21. Fig. 4.21. Reinforcement Details for 1
164 CONCRETE STRUC TURES PART -1 AUTHOR: ZAHID A.
SIDDIQ[ SHALLOW, TEE AND
Space For Reader’s Notes
cHAPTER4: CONTINUOUS BEAMS 165

PROBLEMS

Note: N = Registered Number.


Nq = A fixed positive or negative number assigned by
the instructor to get J? between 600 and 1000.
R - N - No, a parameter to get numerical data for the
problems.

1, For a doubly reinforced section with As = 580 mm2, As = 2390 mm2, b = 300 mm, d =
490 mm, d = 50 mm and/ = 280 MPa. Calculate the idealized flexural strength if:

(i) /' = R/33MPa


(ii) /' = R/25MPa
2. For a doubly reinforced section with b = 300 mm, d = 450 mm, d = 60 mm,/ = 280
MPa, service live load bending moment = R/4 kN-m and service dead load bending
moment - R/3 kN-m, calculate the steel areas required for:

(a) -
p p' x f's/f = 0.7 times the maximum permissible steel ratio for singly
reinforced section.
(b) Area of compression steel is to be the minimum. Keeping the maximum steel
ratio for tension, compression steel will be the minimum.

Repeat the design for the following two cases:

(i) /' = R/33MPa


(ii) /' = 5Z25MPa
3. Calculate the flexural strength for a T-beam (if the data is for singly reinforced
condition) with As = 1840 mm2, b = 760 mm, bw = 190 mm, d = 320 mm,/' = R/33
MPa and/ ~ 420 MPa, if:

(i) hf = R/6 mm.


(ii) hf = R/20 mm (violate the AC1 provision for isolated T-beam).
4. A T-beam section with b = R mm, tw = 228 mm, d= 0.75R mm,/' = 20 MPa,/ = 420
MPa, and hf = R/6 mm is to carry a design moment of R kN-m. Calculate the required
steel area. The slab is on the compression side.

5. A roof system consists of 300 x 300 mm columns at a spacing of JE/MOm and 7E/200m
in two mutually perpendicular directions, as shown in the figure. Beams run in both

ii
166 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI
directions over the columns having more than two spans Chapter - 5
first interior long beam (ft) if the factored slab in each direction. Design
load is 12.00 kN/m2, thickness of the
live load <3 x dead load. Select =
125 nun,// = RJ33 MPa, Grade 420 steel, and slab DESIGN FOR SHEAR
customary bars. US

bTT— 5,1 INTRODUCTION


r 300x300 mm Columns t Shear stress is defined as the magnitude of the internal force acting parallel to the area divided
by the area. This stress is produced due to the sliding of various layers of the material on one
!
another. Shear stress along the depth of cross-section for homogeneous, elastic and uncracked
More than 2 @ ; beams may be calculated by using the following equation:
£/200 m each
lb
Where V = applied shear force at the section,
j Q = first moment of area of the part of the section lying between the point where
More than 3 @ £/140 m each the shear stress is required and the nearest outermost fibers about the neutral
; axis,
:

I = moment of inertia of the cross section about the corresponding axis of
bending,
J And b = width ofthe section at the location where the shear stress is required.
j The shear stresses acting on the vertical sections are complementary to the horizontal shear
stresses in the longitudinal direction of the beams. At the neutral axis, a differential element
has only the shear stresses and hence the principal axes are oriented at 45° to the longitudinal
I axis ofthe member, as shown in Fig. 5.1.

a) Differential Element at Neutral b) Differential Element Away From


Axis Closer to Right Support. Neutral,Axis towards Compression

Fig. 5.1. Differential Element at Neutral Axis Showing State


of Stresses Due to Shear Acting on a Beam.
Away from the neutral axis, the principal stress angle increases until it becomes horizontal on
the tension face, and decreases until it becomes vertical on the compression face. The tensile
stress trajectories appear as shown in Fig. 5.2(a). Cracks also develop on the same pattern
along the tensile principal stress paths, Fig. 5.2(b). Near the supports, cracks appear at almost
45° originating from the neutral axis, which quickly spread towards the two faces. These shear
168 CONCRETE STRUCTURES PART- 1 AUTHOR: ZAHID A
design for shear 169
SJDDlQ1 chapters-
cracks are called pure shear, web or inclined shear cracks. When
these cracks extend Moment To Shear Ratio And Effective Depth
the supports, they weaken the anchorage of the main steel.
Away from the supports buttowards 5 2.2 Applied
at a particular section may simply be
length of the beam having shear force, flexural cracks appear
first on the tensionwithin The ratio of the applied moment (M) and shear (F)
calculated as M/V. Greater is this ratio at a particular section of the member, flexural
penetrating vertically upwards. When the effective cross-section
reduces due to these side resist shear. Consequently, the
shear becomes dominant and the crack become inclined at angles
closer to 45°. These
cracks cracks appear quite earlier reducing the net area to
initial flexural cracks may be
are termed as flexural shear cracks. The average shear stress along cracks effective shear strength is reduced. The opening of the
the depth of the beam (ri) and in turn the shear strength may
=
given by the formula: v V'bvd, where V is the applied shear
force, bw is the width of is controlled by providing greater effective depth
beam and ri is its effective depth. the vary as directly proportional to the effective depth.
be considered as inversely
This means that the shear strength of concrete mayusually subjected to uniformly
proportional to the factor (MiVd). In case of field beams cross-section of concern.
distributed loads, the ratio M/V is to be calculated at the applied symmetrically, as
However, for experiments, usually concentrated loads are
shown in Fig. 5.3. In such cases, the M/Vd ratio
simplifies to the shear span over
effective depth ratio (a/ri).

Cracks
a) Stress Trajectories in a Beam. b) Shear Cracks in a Beam.
Fig. 5.2. Shear Cracking in a Simple Beam. Fig. 5.3. Test Setup for Beams.

For the beam in Fig. 5.3,


5.2 CONCRETE SHEAR STRENGTH
M = maximum moment ~ Va
As already explained, the cracks due to shear are either inclined M!V = a
or become inclined starting
from the flexural cracks. An inclined crack approximately opens M a
perpendicular to its
longitudinal axis. A combination of longitudinal flexure
and transverse reinforcement is Vd d
mostly employed to restrain it from opening. This transverse
reinforcement is provided by Where a = the shear span
vertical stirrups and is called shear or web reinforcement. The stirrups
closed but must be properly anchored at the two ends. Shear strength for shear may not be The shear span to effective depth ratio (a/ri ratio) in experimental
beams have the same
of a member depends loads. Very short shear
upon the following factors: effect as M!Vd ratio in beams subjected to any other types of
action as in Fig. 5.4.
spans (aid 1.0) show tied arch action at failure in place of beam damaging the horizontal
5.2. 1 Concrete Compressive Strength Inclined cracks joining the load and the support are produced The arch is tied at the
Greater concrete strength usually corresponds to more tensile and shear flow from the longitudinal steel to the compression zone.
shear bottom by the flexural reinforcement and the final failure
mode is commonly the
remains true even after initial cracking of the concrete at certain strengths. This
locations. The anchorage failure at the ends.
maximum value of ^[f^ allowed by the ACI code is 8.3 MPa (100
psi), except in after the formation of
prestressed concrete beams and joists. This means that the Short shear spans (aid from 1.0 to 2.5) also develop arch action and partly by the
by this arch action
up to 69 MPa.
expressions are valid for/ ' inclined cracks but the total load is partly carried mechanical interlocking
with the
dowel action of the main reinforcement combined or dowel
takes place by bond
between the cracked inclined surfaces. The failure In shear
failure along the flexural reinforcement or by the shear
compression failure.
beam than a flexural crack,
compression failure, the inclined crack rises higher into the
design for shear 171
170 CONCRETE STRUCTURES PART - I AUTHOR: ZAHID A. SIDDIQI chapter 5
reduces the compression area, and causes crushing of concrete over the crack.
yd < 1.0, the maximum
„„
value to be useo
used
M be in the formula must be
1.0, if H is
Slender
shear span {aid from 2.5 to 6.0) do not develop arch action and failure occurs purely by M.
--

the flexural shear cracks. The resistance offered by concrete in shear after the initial actually more
concrete, 1.0 for normal weight
concrete
modification factor for light-weight
crack is only due to mechanical interlocking of concrete surfaces at cracks and dowel 2 =
action of horizontal steel. Very slender shear span beams {aid > 6.0) usually fail in = 0.75 of ¥u and Mv at each section and
hence the manual
flexure without the formation of inclined cracks. involves calculation easy equation to estimate
above equation lengthy. The ACI Code gives an alternate and
i calculations become follows:
strength of concrete, which is as (FPS)
the shear
g: jz= 0.17277; b^d (SI) Fc= ^^b^d
weight concrete, not given by the ACI Code,
to the first equation for normal
Another alternate
is:
। V d^
if y
Fig. 5.4. Shear Behavior of Beams with Lesser aid Ratio.
v'^\f^r\b*d
4.68 \
<sl> )

to the factored applied shear according to


shear stress correspondingusing approximate equation, for which no shear
5.2.3 Longitudinal Reinforcement Ratio The average safe compressive strengths
of the
particular values required,
Smaller steel ratio (p^) causes flexural cracks to extend higher into the beam and to 1 is given in Table 5.1.
open more. This reduces the shear capacity of the remaining smaller section. reinforcement is
Longitudinal steel provides dowel action and prevents relative vertical movement of No
Stress for
two parts of the beam formed by the inclined cracks. It also helps to provide more Table 5.1. Safe Ultimate Shear
interlocking between the cracked surfaces in case of hairline cracks. The dowel action
ends after splitting of concrete horizontally at the level of the main reinforcement. Average Shear
Longitudinal steel also acts as a tie if arch action has to develop for smaller aid ral ios. MPa
MPa 265
17.25
20
J
~~ 00.285
0,299
5.3 ACI CODE PROVISION’S FOR CONCRETE SHEAR STRENGTH 22
0.319
25 A 1^*7
For members without axial force, the shear strength {Vc) provided by the concrete alone is 28
given by:
0.349
30
Average Shear Stress
psi
psi J
bwd < 0.29^bwd (SI)

37.5


2500 41.1
3000 ’
44,4
'
l.9A^+2500pw (FPS) 3500 ’’
47 4
4000 53.0
5000
where Vc = Concrete shear strength the
taken equal to 0.80 times
Vy ~ Factored shear at the section, N or lb
to ACI 11.2.3, the area of
circular sections in shear is
According
Mu - Factored moment at the section, N-mm or Ib-in square of the diameter.
d = Effective depth, mm or in
DESIGN FOR SHEAR 173
172 CONCRETE STRUCTURES PART -I ffXAPTER5:
AUTHOR: Z ARID A.
5.4 SHEAR STRENGTH PROVIDED
BY VERTICAL REINFORCEMENT
SffiDlQl
^»=0.062^^ 0.35^ J yt
>
J yt
(SI)

The shear reinforcement commonly consists of 10


to 15 mm (No. 10 to 16) bars ^=0^5^^.A*
>
J yt
(FPS)
against the flexural reinforcement and hanger
bars. These stirrups may be single, anchored within distance, s
four legged having the shapes as shown in Fig. 5.5. double or iVhereA,m« = total cross-sectional area of web steel
reinforcement.
Let = combined area of all transverse legs of the stirrup at a particular andfn ~ specified yield strength of transverse
cross¬ in the SI equation governs for fcr < 31.9 MPa.
This minimum
section, Hie second expression 0.35 MPa (50 psi). It is important to note that
s = longitudinal spacing of the stirrups, : provides a shear stress capacity of
reinforcement spacing for this minimum
and = shear strength provided by the stirrups at a particular cross-section. the maximum limit of d!2, given later, is also required to be satisfiedrequired are the
this minimum reinforcement is not
Knowing that both the applied shear force and the
stirrups are in the transverse web reinforcement. The exceptions where stresses across the width of the member or in some
expected that, after the formation of cracks, one
shear stirrup can resist A/y force.direction^
it is members that have capacity to redistribute
resistive force provided per unit length of beam then The average • casesto adjacent members. The examples are as follows:
becomes A^fy / s. If the dimension of a
crack is assumed to be equal to d in the longitudinal : 1. Slabs and footings.
direction (for a 45° crack), the shear
strength provided by the stirrup (Vs) becomes: 2. Concrete joist construction.
3. Shallow beams with a total depth not
greater than 250mm (10 in.) and beams integral
i larger of 2V2 times the thickness
; with slabs, having depths not greater than 600 mm and
। of the flange and one-half the width of the web.
BEAMS
5.7 LOCATION OF MAXIMUM SHEAR FOR DESIGN OF
less than a distance d from the
According to ACI 11.1.3.1, the non-prestressed sections located
that computed at a distance d., as
the support may be designed for the same shear, V„, as
face of
used in case of inverted beam with load
shown in Fig. 5.6. This provision should not be This is due to the presence of local
1 applied on the slab acting as flange of the beam.reaction, which reduces the chances of
compressive bearing stresses due to the support
Used to Cover-up within a distance d from the
development of shear cracks. The loads applied to the beam
the Deficiency an inverted wedge of concrete in
>1 support will be transferred directly to the support by
compression formed within 45° cracks.
Fig. 5.5. Various Types of Transverse Reinforcement for
Shear.
5.5 STRENGTH REDUCTION OR RESISTANCE
FACTOR IN SHEAR
The strength reduction factor for shear (^,) is taken equal to
0.75. It is relatively lesser than the
corresponding factor in bending, giving more factor of safety,
of failure and secondly the larger scatter of the test due to firstly the sudden nature
results.
5.6 MINIMUM WEB REINFORCEMENT
If Vu is lesser than theoretically no web reinforcement is required. However, if
WJ2, the ACI Code requires the provision of a minimum area of Vu
web reinforcement for extra
safety as given below: Fig. 5.6. Critical Section for Design Shear Force.
174 CONCRETE STRUCTURES PART -I DESIGN FOR SHEAR 175
AUTHOR: ZAHID A CHAPTER 5:
5.8 MINIMUM SECTION FOR DESIGN SHEAR CAPACITY SlDDi^
Smax
According to the ACI 11.4.7.9, if required shear to be resisted by
the transverse steel
making the
exceeds 0.66^fcbwd (N) (y^ of legs of stirrups should be selected to avoid a spacing
Gb) > the cross-sectional dimensions The size and numberdifficult, should commonly be not less than 80 to 90 mm. The
increased. This is required for— development of concrete compression
_J
must be pouring of concrete a certainwhich
shear should be continued for some number of spaces towards
truss analogy for shear. However, in most practical cases, required struts during the spacing calculated for is to be evaluated there.
Vs should not be more than thelower shear region and then new spacing
given as
0.33^bwd (N) 4^bwd Gb) - For inclined stirrups to be used as
shear reinforcement, the available shear strength is
- 1 under:
5.9 MAXIMUM SPACING REQUIREMENTS A, (sin a + cos a)d
'5
s
The maximum spacing requirements for the transverse axis of the member, and
reinforcement may be combined with
the minimum web reinforcement criterion to get the following Where, a= the angle between stirrups and the longitudinal
results: s spacing of stirrups parallel to the
longitudinal reinforcement.
=
bent-up longitudinal bars at same location are used
If a single bar or a single group of parallel
the maximum taken equal to lesser of Vs = ^4^sin® and
as shear reinforcement, their shear contribution is
spacing is given by bent-up bars and longitudinal axis of the
0.25^[f~ b^d (N), where a is the angle between the
Smaz — smaller of the following three for SI units:
member.
i)
0.35b„ 5.11 HANGER REINFORCEMENT
beam-2. Compression fens
Imagine that Beam-1 in Fig. 5.7 is supported on a perpendicularthe compressive forces in the
ii) dl l
iii) 600 mm component of
form in the supporting beams, and the downward
= smaller of the following three for FPS units: by the hanger reinforcement provided at the joint
Smaz members of the fan are to be balanced the supporting beam. In addition to
Otherwise these forces will separate the bottom portion of
i) -z~ ii) d/1 iii) 24 in. having a tensile capacity of (1 - h 7hs) times the
the normal shear stirrups, hanger reinforcement
h 'is the vertical distance between
end shear of the supported beam must be provided. Where
beam. About two-third of this
When required exceeds 0.33-^bwd (N) (lb) , the maximum spacing as the bottoms of the beams and hs is depth of the
supporting
the remaining in the supported
reinforcement should be provided in the supporting beam and of the supported beam is less
given by the second and third of above conditions must be end shear
halved. beam. Hanger reinforcement is not needed when the
5.10 DESIGN OF WEB REINFORCEMENT •
than 0.25-^^ (N) (lb)]-
Total nominal shear strength of a section is the sum of the shear strength of
the shear strength provided by the steel. For design this nominal shear concrete itself and ;
the strength reduction factor must be larger than or equal to the
capacity multiplied with I Compression
factored applied shear. i
Beam-1 Fans
Kn ~VC + VS
For design, Vu where = 0.75
' Hanger
_ Beam-2
Reinforcement
A Af d
Fig. 5.7. Compression Fans Formed at Junction of Two Beams.
DESIGN FOR SHEAR 177
1 76 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A.
SlDDlQj (FPS units)
5.12 TYPICAL SHEAR FORCE
DIAGRAMS
Smm = the smallest out of the following four values:
The shear force diagram in case of simply supported
beams may easily be plotted.
continuous beams subjected to uniformly distributed loads and designed For
for flexure using the
° 50b„
ii) dll
ACI moment coefficients, approximate shear force diagrams, iii) 24 in
as in Fig. 5.8 may be used.
5.13 GENERAL PROCEDURE FOR SHEAR DESIGN iv)J
{Vu-m
1. Plot exact or ACI Code shear force diagram for
the beam. Also draw bending
diagram if the exact equation for Vc is to be used. moment
2. Find maximum design shear, Vu> and factored
moment, M,, at d distance from conditions (ii) and (iii) to one-half of the above values.
face of the support. inner
multiples. If this value is lesser than
3. Calculate concrete shear strength preferably using the
exact equation. However, if the 8. Round the calculated spacing to the nearest 10mmeasy placing of concrete), increase the
data for exact equation is not available, the approximate equation
may be used. about 80mm (or any other convenient value for
i
stirrup steel (if feasible) or the
4. If Vu < faVJl, then shear reinforcement is not required
according to the code. A | number of legs of stirrups (if possible), yield strength of
minimum amount of stirrups may be provided to keep the longitudinal 1 diameter of the stirrup.
steel in position. face of the support and decide how
5. Decide whether cross-sectional dimensions are all right
or the size has to be increased. F 9. Place first stirrup at s / 2 distance from the inner beam. Usually the
many of such spaces are to be continued towards the center of the
in-between 4/5 to 4
Generally, if F, > (N) (lb) s spacing for the most critical section is continued up to a distance
the size is io be that the spacing may be
/4. Calculate Vv and Mu at this new location. It is to be noted
is carried out in a factory
l_

increased. However, we may go to double of this value in exceptional


cases recalculated after placement of each stirrup if the constructionfor the calculations.
6. Decide a trial diameter and shape of the stirrup and calculate under controlled conditions and a computer program is used
Av. steps to calculate spacing up to the point
: 10. Repeat steps 3 onwards leaving the irrelevant for one more point in¬
where no shear reinforcement is required or make calculations
between.
by the ACI Code can be
, 11. Location beyond which shear stirrups are not required shear in terms of
approximately found by equating the expression for the external
<
using the approximate relation for the concrete
I distance from the support, x, to
shear strength.
S. F. Diagrams Example 5.1 (SI Units): A simply supported rectangular beam of size
300x525 mm,
span
130 kN/m. The clear
having an effective depth of 450mm, carries a total factored load of
of the beam is 4.5m (effective span 4.8m). The flexural reinforcement
= consists of 8-#25 US
customary bars (Aa = 4080 mm2), four of which are curtailed
at 450mm short of face of
the
Fig. 5.8. Typical Shear Force Diagrams. supports. Use C-20 concrete and Grade 280 steel. Using double legged stirrups, design
web reinforcement.
7. Calculate the maximum required spacing as follows:
Solution:
Smm = the smallest out of the following four values: (SI units) I. The shear force and bending moment diagrams are shown in Fig. 5.9. The shear force
as follows:
and bending moment at x distance from the support may be calculated
f
i) ~ 0.35 is replace with 0.062 if fc '>31.9 MPa
ii) dll Vt «312-130»
iii) 600 mm M, «312» -65?
iv)M^Z(^-^X)
178 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. DESIGN FOR SHEAR 179
SIDDIQ| cHAPTER5:
2. The shear and moment at the critical section, d distance
from inner edge of the sun<v«
are as follows: bUPPPrt, V, = (Vu - ^Vc) /^ = 144.97 / 0.75 = 193.3 kN
Vu = 234 kN Mu = 163.8 kN-m 0.33y[Zbwd
v = 0.33^20 x-Ox-—°
1000
= 199.2 kN
V„d 234x0.45 4080/2
=0.643 < 1.0
300x450 °’°15 Cross-sectional dimensions are OK.

Vc b„d < 0.29 A bwd Vs > 0.33-/fc bvd and the spacing limits are to be used accordingly.
6. Let diameter of bar - #10 US
Then, area of double legged stirrup, A„ = 142 mm2
300x450
1000
029x1x^20x300x450 7. S/nax = the smallest out of the following four values:
1000 142x280 = 379 mm
= 118.7 175.1 kN 9 i)
0.35x300
0.35b,
= 118.7 kN
ii) d! 2 = 225 mm
iii) General minimum = 300 mm
130 kN/m
0.75x142x280x450 = 92 mm
’V)
(^, 144.97x1000
= 92 mm
8. s = say 90 mm > 80 mm (OK)
9. First stirrup is placed at s/2 = 45mm distance from face of
support. 4/5 = 900mm and
at a spacing of 90mm.
4/4 = 1125mm. Eleven intervals of stirrups may be provided
The location of the last stirrup is defined as follows:
= 0.150 + 0.045+10 x 0.090 = 1.095 m
10. Atx = 1.095 m:
Vu = 169.65 kN Mu = 234.06 kN-m
= 0.326 1.0 = 0.030
M„
f Vvd bA
Mu
Fig. 5.9. S. F. And B.M. Diagrams For Example 5.1.
300x450
1000
4. = 0.75x118.7 = 89.03 kN ; ^K/2 = 44.51 kN 0.29 x 1 x V20 x 300 x 450
Vu > t^VJl Transverse reinforcement is required. 1000
5. Vu - 234.00 - 89.03 = 144.97 kN = 119.04 175.08
= 119.04 kN
180 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
DESIGN FOR SHEAR 181
^rc = 0.75x119.04 = 89.28 kN ; CHAPTERS:
= 44.64 kN beam of size 12x2 1 in,
Vu > trample 5.2 (FPS Units}: A simply supported rectangular
Transverse reinforcement is required. factored load of 9 k/ft. The clear span of the
having an effective depth of 18 in, carries a total
= the smallest out of the following four values: span = 16ft). The flexural reinforcement consists of 8-#8 US
Smax
beam is 15ft (effective at 18 in short of face of supports. fc'
customary bars = 6.32 in2), four of which are curtailed
legged stirrups, design the web reinforcement.
•x
V
0.35^
= 142x280
0.35x300
- 379 mm = 3000 psi and = 40,000 psi. Using double
Solution:
ii) d/2 - 225 mm 1, The shear force and bending moment diagrams are shown
in Fig. 5.11. The shear force
calculated as follows:
iii) 600 mm and bending moment atx distance from the support may be
. . 0.75x142x280x450
(169.65 -89.28)x 1000 - 167 mm
V* =72-9x
M, =72x-4.5x2
inner edge of the support,
= say 160 mm 2. The shear and moment at the critical section, d distance from
1 1. To find the location after which stirrups are not required are as follows:
Approximate relation for Vc may be used here.
theoretically, let Vx = F„ = 54k ; Mu = 126.0 k-ft
Vx =^/2xO.V7^,^bwd 9 k/ft

312

130 x
— 01

= 312-38.49
=

03?^'217 1000 x 300x450
x = 2.10 m /. Continue a spacing of 150 mm for 7 intervals.
Further it is better to provide some stirrups in the portion
are not required. For example, in this case, a stirrup mayin which theoretically these
The results are shown in Fig. 5.10. be provided at the center.

Fig. 5.11. S. F. And B.M. Diagrams For Example 5.2.

, 0.0146
YA „ 54xL5 i W
Fig, 5.10, Shear Design Result for Example 5,1. 126 12x18
M.
182 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. DESIGN FOR SHEAR 183
8^ CHAPTERS:
V,d
Vc = [1 + 25°0pw b*d 35^ b-d

4.
= [1.973000 + 2500 x 0.0146 x 0.643]12x18
= 27.55
= 27.55 k
41.41 k

Wc = 0.75x27.55 = 20.66 k
1000
3.5-73000 x 12 x 18
1000
r
4.W3000
.
= 26.28 41.41 k
112x18
+ 2SMX0.0292 X 0.241]— —
3.573000x12x18

^VJ2 = 10.33 k = 26.28 k


Vu > Transverse reinforcement is required. ^E/2 = 9.85 k
^Vc = 0.75x26.28 = 19.71k ;reinforcement is required.
5. Vu - ^Vc = 54.00-20.66 = 33.34k Vu > ^K/2 Transverse
V3 = (K, - = 33.34/0.75 = 44.45 k Smax = the smallest out of the following four values:
i) 14.7 in
4^6^
v = 4^000 x^i^
1000 =
47.32 k ii) 9 in
iii) 24 in
4^f^bvd .
(35.25 -19.71)x 1000—
.-. Cross-sectional dimensions are OK. .1V),
— =
0.75x0.22x40000x18
-7 r — „
7.64 in
6. Let diameter of bar = #3 US (F,-^)
Then, area of double legged stirrup, Ar = 0.22
in2 = say 7.5 in
7. $tnac the smallest out of the following four To find the location after which stirrups are not
theoretically required, let Vx = faVJl.
values: 11.
Approximate relation for Vc may be used here.
= 0.22x40000
i)
506w
ii) d/2
iii) 24 in
50x12
= 9 in
14.7 in Vx
4^2^^
72-9x = 0.75x 73000 x12 x 18 / 1000
= 24 in 9x = 72-8.873
0.75 x 0.22x40000x18
= 7.01 ft Continue a spacing of 7.5 in for 5 intervals.
33.34x1000
- 3.56 in x
portion in which theoretically these
= 3.56 in Further it is better to provide some stirrups in the
may be provided at the center. The
are not required. For example, in this case, a stirrup
8. s = say 3.5 in > 3 in (OK) results are shown in Fig. 5.12.
.9. First stirrup is placed at sh = 1.75 in (say 1 in)
distance from face of support. ^5 = 36
in and £74 = 45 in. Twelve intervals of stirrups may.
The location of the last stirrup is defined as follows: be provided at a spacing of 3.5 in.
x 1 + 6 + 12 x 3;5 = 49 in (4.083ft)
10. Atx = 4.083ft:
= 35.25 k M = 218.96 k-ft
Yd
P* = 0.0292

Fig. 5.12. Shear Design Result for Example 5.2.


184 CONCRETE STRUCTURES PART - 1 DESIGN FOR SHEAR 185
AUTHOR: ZAHID A.
SFDDjqj CHAPTERS:
5.14 TRUSS ANALOGY FOR SHEAR to be wrapped around the longitudinal tension
The transverse shear reinforcement is compression zone by providing hooks around the
The general shape of the beam after shear cracking is shown anchored in the
reinforcement and is to befunctions
in Fig. 5.13. This shape suggests longitudinal bars. The of transverse steel may be summarized as under:
that an arch action is developed within the beam, with top
compression at the top and at
within the cracks and longitudinal tension ties at the bottom sides . It carries part of the applied shear, denoted by K.
reinforcement. The concrete portion between two consecutive cracks
compression by the arch action and is terms “compression
in the form of longitudinal
is subjected to
. the interface of a crack due to interlocking of the aggregates
Hence, the shear strength at
It prevents excessive widening of the inclined shear cracks. (Ka) maintained.
is
strut”. In the presence of
reinforcement, this arch action after initial cracking is modified transverse
into an analogous truss . It keeps the longitudinal bars in position after the formation
steel in
of
resisting
the shear cracks. This
shear in the form of
mechanism, consisting of bottom chord of longitudinal steel,
vertical tension members of increases the contribution of the longitudinal
stirrups, diagonal concrete compression struts, top compression dowel action (Fd).
chord of concrete, compression
. the compressive strength of
fans and joints at the intersection of the truss members (assumed
as pins). The confining action of the stirrups on concrete increases
the concrete to resist a component of shear (K).
in Fig. 5.13, the direct shear strength of the
In the truss model of shear behavior, shown aggregates and the dowel action of
concrete, the shear carried by the mechanical interlocking of
the longitudinal reinforcement are neglected.
In other words, it is assumed that the shear
of shear reinforcement. In this
strength of a member can only be mobilized in the pr4esence
model, the top chord compression and bottom chord tension
provide the resisting couple to
balance the applied bending moment. Near the
concentrated loads and reactions, the load
angles. The zone near the point of
spreads as a fan made of compression struts at varying change in member depth is called
application of a concentrated load or near an abrupt -

discontinuity / disturbance /D region. In such region, the
order elastic beam theory is not applicable. The region
plane section theory and first
where normal beam theory is applicable

is called the beam/ Bernoulli /B region.
The angle of the compression field / strut depends
on the size and spacing of the stirrups, which
value of 45°. While forming the
generally varies between 25° and 65° with the most common of Fig. 5.13b is replaced by
simplified truss model for calculation purposes, the actual modelvertical section are combined
through one
that of Fig. 5.13c. All the diagonal struts passing
stirrups crossing one-half of the adjacent struts are lumped into
together, while all the vertical
the force in one compression strut
one vertical stirrup at all ends of the struts. In Fig. 5.13d, horizontal and vertical components,
(Ce) acting near the bottom reinforcement is resolved into
where,
Cc = compression in the concrete diagonal strut,
V = shear force at distance ‘x’ from the support,
and,
JVb in the longitudinal steel for shear equilibrium,
= additional tensile forcethe
0 = angle of inclination of diagonal struts.
stirrups becomes known equal to and
At ultimate condition near failure, the force in the longitudinal tension due to shear
the truss becomes determinate. The magnitude of additional
can be determined as under:
Nb = 7/tan0 = Kcot0
element and the bottom ties, effective
= lever arm between the top compression between
d) Equilibrium of Compression Strut Force.
Let dv the centroids of tensile and
depth in shear equal to the distance
Fig. 5.13. Truss Analogy for Shear Behavior. compressive forces but not less than 0.9d,
DESIGN FOR SHEAR 187
186 CONCRETE STRUCTURES PART -1 AUTHOR: CHAPTERS:
ZARIBA. SIJjDjqj
bv - web width = bw, PROBLEMS
ft = stress in the diagonal compression strut.
Cc = F/sin# Note: N = Registered number of the student.
fA *
C
Zi =
_
V No = A fixed positive or negative number assigned by
the instructor to get R between 600 and 1000.
b„dv cos# bvdv sin# cos# ’ R = N No, a parameter to get numerical data for the
-
= dj^0 problems.

.
Afa
.
= —d~
Kstan#
j_ A cantilever beam has to carry a concentrated service
live load of 80,000 1 R kN acting
at a distance of R I 600 m from the edge of the support,
b = R 1 2.4 mm and d = R / 1.4
and grade 280 steel.
If a lower value of #is used, less vertical and more
horizontal steel is the result. Similarly, if a mm. Design the shear reinforcement using C-22 concrete
higher value of #is selected, more vertical and less vertical
steel is obtained. Usually the value lengths equal to R / 120 m and is
of # slightly less than 45° is used for economical design.
For a safe design the stressed must 2- A two-span continuous beam has equal spans of clear and a service live load
not exceed the safe compression carrying capacity of the j
subjected to a superimposed service dead load of R I 100 kN/m
diagonal struts. According to dead load of R I 5 kN and a
AASHTO Specification, the following expressions are proposed: ;
; of R / 80 kN/m. A superimposed service concentrated mid-section of one span only.
service concentrated live load of R / 4.5 kN are applied at
K = pjfM : Design the shear reinforcement using C-35 concrete
and Grade 420 steel. Width of the
depth of beam is R/l .5 mm rounded
-v ^d,(cot0+cota)sina beam is AZ2.2 mm rounded to 75 mm multiples and

s to 25 mm multiples.
Where, a is the angle of the transverse reinforcement with : multiples) and effective
For a rectangular beam of width R 1 2 mm (rounded to mm for the following three
respect to the longitudinal axis and 20
3-
P is the concrete tensile stress factor. Both # and P are determined from AASHTO tables ; depth 0.9 R mm, determine the maximum factored shear
depending on vu I fc', longitudinal strain in concrete and presence
minimum transverse steel. of minimum or less than ’■ cases using C-25 concrete and Grade 420 steel:
I
a. No shear reinforcement is provided.
legged stirrups is provided.
b. AC1 Code minimum steel using #10 double
c. Maximum shear reinforcement is provided.
shear of R / 7 kN and
4- A 1.5 m length of beam is subjected to a constant dead service
live service shear of R 1 6 kN. b - RI 1.5 mm (rounded to 20 mm multiples) and d =
at an angle of 45°. Check
1.17? mm. Two #25 bars are bent up at every R mm distance
stirrups, if required. Use C-25
the adequacy of bent-up bars for shear and design she^r
concrete and Grade 420 steel.

A beam is constructed using C-30 concrete and Grade


420 steel, having effective depth
5-
equal to 2 x its width, and is expected to have a maximum
factored design shear of R /
only for the following conditions:
4 kN. Proportion the section based on shear design

a. No web reinforcement is used.


b. AC1 Code minimum web reinforcement is provided.
c. Vs = 2 x Vc.
188 CONCRETE STRUCTURES PART 1
- AUTHOR: Chapter - 6
ZAHlDA.
Space For Reader’s Notes
ONE-WAY AND EDGE SUPPORTED SLABS

61 ONE-WAY SLABS
jhe loads acting on a slab panel, bounded by slab edges, walls or beams, are to be transferred
to the supports at the ends. In case the slab is supported on all the four edges and the longer to
shorter span ratio is lesser than 2.0, the load is distributed parallel to shorter and longer
directions to all the four supports. This is called two-way action of slabs.
One-way slabs are defined as those slabs that are supported along one or two opposite edges
or supported on four edges but longer to shorter spans ratio is greater than two. Almost the
entire load is transferred in one direction only. In case only one or two opposite edges of the
slab panel are supported, the span of slab is considered perpendicular to the supporting walls or
beams/ For a slab panel supported on all the four edges, the span is considered as the shorter
dimension of the slab panel. Full slab load is considered for this span to calculate the bending
moments. Main steel is only provided in the direction of span for the one-way slabs.
Examples of one-way slabs are cantilever slabs, slab panels with much larger aspect ratios, stair
slabs, strip footing slabs and high retaining walls.

6.1>1 Procedure Of Slab Design


For the design of a one-way slab, a unit width (equal to Im, 1mm or I ft) strip is
considered that extends along the full span of the slab, Fig. 6.1. This strip has a
rectangular cross-section with the considered unit width and depth equal to the
thickness of the slab. The load per unit length acting on this strip is calculated as the
slab load per unit area multiplied by the width of the strip. This strip exactly behaves
like a wide beam and hence its design is carried out by using the same formulas as for
the beams. If the slab is continuous beyond the supports into other panels, this strip
may act like a continuous beam and ACI moment coefficients may be used to calculate
the moments at the critical sections.
Depths of the slabs are usually decided based on deflection control and fire rating.
These depths are generally much larger than that required for the flexural and the shear
strengths. The load acting on the strips are smaller because these have to support loads
directly acting on them and not the loads acting on other parts of the slab as in case of
beams. Hence, majority of slabs are only lightly reinforced with steel ratios
significantly lesser than the maximum steel ratio allowed by the ACI Code. The steel
reinforcement calculated for the unit width strip is continued over foil width of the
slab panel, perpendicular to the span.
It is important to note that if a slab panel has longer to shorter spans ratio larger than
two but is supported on all the four sides than the slab behaves somewhat like a two-
way slab near the shorter edges. If these shorter edges of the slab are continuous or
ONE-WAY AND EDGE
SUPPORTED SLABS 191
190 CONCRETE structures p ART - 1 AUTHOR: ZAHID A CHAPTERS:
siddiqi
partially fixed, some negative steel is to be provided over the
cracking. supports to
avoid Table 6.2. Fire Rating Based on Clear Cover.

Clear Cover (mm / in) Fire Rating (Hours)


Unit Width 1^4
20
J
l
1
i Thickness 25 1 l’/2
of Slab
40 l*/2 3
Unit Width Section 1-1
Fig. 6.1. Design Strip for One-Way Slabs.
6,1.4 Provision Of Slab Steel
the slab width.
6.1.2 Depth Of One-Way Slabs The steel calculated for the unit width strip is to be provided all along
This is done by providing bars at a regular spacing (s). If 4 is the area of a single bar
The minimum depth of slabs depends upon span length, end area of steel required for a Im wide strip,
conditions and yield effective for a width equal to s and As is the
strength of the steel to be used. Slab thickness also influences the
during the incident of fire and the fire rating is sometimes measuredtransmission of haaf
as the time in hours
the following may be written:
required to raise the temperature on the slab side opposite to the fire by
140° C. A 90 4 4 or 5
10004 (SI)
mm slab gives a 1-hour fire rating, a 125 mm slab provides a 2-hours 1000 s 4
and a 160 mm
slab possesses a 3-hour fire rating.
A, or s „ 124 (FPS)
The minimum thickness of non-prestressed one-way slabs for the 12 (in) s 4
may be found by using Table 6.1. deflection control
bar sizes
Spacing tables are available, such as those given in Annexure, having various
along the columns and different spacing values along the rows and giving area of steel
Table 6.1. Minimum Thickness of Non-Prestressed One-Way Slabs For selected first and the
Normal Weight Concrete. per unit width. Usually, smallest bar size (10mm) is a
corresponding spacing is seen from the table. If this spacing is too less (< 100 mm), the
spacing is
End Conditions Steel Grade second larger size is selected and the procedure is repeated. This minimum
280 & 300 420 fy maintained for easy cutting, bending and placing of steel.
Simply supported ^/25 //20 / /20 x (0.4 + 4/690)
One end continuous //30 I! 24 // 24 x (Q.4 +4/690) 6.1.5 Maximum Spacing Of Bars
Both ends continuous
The ACI Code limits the maximum spacing of main slab steel bars to lesser of
/ 35 //28 //28 x (0.4 +4/690) three
Cantilever However, as a local field practice, it is
//12 //10 //10x(0.4 + fy / 690) times the slab thickness and 450 mm (18 in).
Notes: 1) I - c/c span of the slab. preferred to keep this spacing within 2 times the slab thickness and 300 mm.
2) The calculated thickness usually may not be lesser than 90mm.
The maximum spacing of the shrinkage and temperature reinforcement, to be described
in the next sub-article, is limited to lesser of 5 times the slab thickness and
450 mm (18
6.1.3 Minimum Clear Cover and mm.
in). Local practice is to limit this spacing to lesser of 2.5h 375
The concrete cover to the reinforcement is required to provide resistance against
Further, the flexural reinforcement must be well distributed to get better control over
corrosion, fire and wearing of the surface. Further, cover is also required to develop closest
bond between steel bars and concrete. According to the ACI Code, the minimum clear the width of cracking. According to ACI 10.6.4, the spacing, s, of reinforcement
to a surface in tension is not to be more than the following:
cover required for slabs not directly exposed to weather or in contact with soil is to be
20 mm. Greater concrete cover may prevent the loss of steel strength due to the rise in 280
temperature due to fire for a longer duration. The approximate relationship that exists Smax = 380 ^-2.5C< 300—- (SI)

between the clear cover and the fire rating is given in Table 6.2.
192 CONCRETE STRUC TURES P ART - 1 AUTHOR: ZAHID A. SIDDIQI
g ONE-WAY AND EDGE
J CHAPTER6: SUPPORTED SLABS 193
I 6.1.9 Check For Shear
Where, cc is the clear cover from the nearest surface in tension to the outside Shear is not generally critical in floor slabs supported on walls or beams. The
flexural tension reinforcement, -- of the
|| x x foundation slabs are to be checked for punching and one-way shear. If required, the
one-way shear may be checked in a slab just like a beam.
^factored moment
and £ =
6.1.10 Design Flow Chart For One-Way Slabs
, / 40,000^I- „ „
Smax =
15^—-— 2.5 c 12 I — —I
( 40,000^
(FPS) 1. Check that the slab panel falls in the category of one-way slab. The span length is
taken as the clear slab dimension plus half the slab depth on either side.
These provisions are not sufficient for watertight structures.
2. Find the minimum thickness of slab required for the deflection control according to
the steel strength, span of the slab and the end conditions. Round this depth to 10
6.1.6 Distribution, Temperature And Shrinkage Steel
mm multiples keeping a minimum of HO mm, except for shades arid shelves.
As already stated, in case of one-way slabs, the main reinforcement is only placed along
3. Calculate the dead load acting on the slab per unit area and estimate the live load
the shorter span. However, some minimum reinforcement must be provided in the according to the use of the floor.
perpendicular direction to control cracking due to temperature and shrinkage volume
changes. Further, this steel causes better distribution of loads over the main 4. Calculate factored or ultimate load per unit area. Multiply this load with the unit
reinforcement and hence sometimes termed distribution steel. It is always placed on the width to change it into load per unit length acting on the design strip.
inner side of the main reinforcement with respect to the slab depth. The amount of this I , 5. Calculate maximum bending moments at the critical sections by direct analysis or
minimum reinforcement is given in Table 6.3. by using the ACI coefficients.
Table 6.3. Distribution, Temperature And Shrinkage Steel in Slabs. 6. The unit strip is designed just like a beam to calculate the area of steel which
will be in the units mm2 per meter width of the slab. The cross-section of this strip
Main Steel Yield Strength Fraction Of Distribution, Temperature will be rectangular of size 1000 mm hmm.
x
MPa psi And Shrinkage Steel With Respect To
Gross Concrete Area d = h - 27 mm
280 & 300 40,000 0.0020 7. Calculate the minimum depth as a singly reinforced section for the maximum of all
420 60,000 0.0018 moments at the critical sections and make sure that this is lesser than the selected
>420 > 60,000 depth. Almost always this check will be satisfied for tbe slabs.
> 0.0014 (SI)
8. Calculate area of steel required per unit width using trial procedure, by using
^1^0.000
A
, 0 0014
graphs, by using tables or by using the quadratic formula for all the critical sections.
9. Calculate the amount of minimum reinforcement and adjust the steel in the previous
step, if required.
6.1.7 Minimum Flexural Steel 10. Using tables of steel areas, select diameter of bars and their spacing for all the
ACI 10.5.4 specifies minimum flexural steel in slabs lesser than that of beams, which is critical sections having bottom steel.
equal to the temperature and shrinkage reinforcement. 11. Check for maximum spacing allowed by the code and adjust the diameter of bars
and the spacing, is required.
6.1.8 Contlnuoue One-Way Slabs
12. Curtail this bottom steel or bent it up at the same distances as for beams. Decide
The ACI moment coefficients given earlier for the continuous beams may be used to the diameter and spacing of top additional steel at all critical sections having
calculate the bending moments at all the critical sections. The curtailment of bars may negative moments.
also be carried out just like continuous beams.
13. Calculate the steel required in the perpendicular direction, along the longer
dimension of the panel, as the amount of distribution, temperature and shrinkage
reinforcement. Decide the bar diameter and the spacing and check for the
ONE-WAY AND EDGE
SUPPORTED SLABS 193
194 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A.
SIDD1Q[ CHAPTER 6:
maximum spacing. This steel is to be provided on the inner side of the peadjAS^
compared with the main reinforcement
slab x 2400 = 384 kgs /m2
R. C. slab: 1000
14. Check shear at d-distance from edge of the support in case of that slabs
which x 1800 = 135 kgs /m
appear to be excessively loaded. If the slab is considered safe, the beam 75 mm screed of brick ballast: 1000
shear condition K / 2 is not applicable for slabs.
15. Detail all the main and distribution steel and show the reinforcement on drawings or P. C. C. + terrazzo:
.62- x2300 = 138 kgs /m2
1000
at least sketches. = 657 kgs Zm
At least 50% steel of a cantilever slab is to be extended for a minimum distance of
1.5 times the cantilever length into the slab from where the cantilever is taken. = 200 kgs/ m2
Theoretically, this distance may be calculated by equating the moment due to all For residential building:
loads on the cantilever slab about the support to the moment due to only dead load
on the main slab about the same support.
Factored Load
qu = 1.2^d + 1.6?l 10.87 kN Im2
For the drawing, the following rales are considered: 657 + 1.6 x 200) x 9.81 / 1000 =
= (1.2 x kN
= 10.87 /m per meter width
• The outer edge of slab is shown by full line in the plan View.
• The inner edges of walls are shown by dashed lines. rytored Bending Moment
• The bottom steel is shown by thick continuous lines and the top steel is (approximation on safe side)
= 1 / 8 q0 fn = 1 / 8 qa 20.78
indicated by thick dashed lines. Afu kN-m per meter width
= 1/8 x 10.87 x 3.912 =
• Towards a simply supported end, every two out of three bars may be bent
up. However, generally every alternate bar is bent up and then top extra bars Main Reinforcement
of #10 diameter are provided between these bars. Assuming o = d!6 = 22 mm
• The top steel along the longer direction of a one-way slab may be extended
20.78x10*
by considering its length with respect to the shorter span.
A = = -
0.9 x 280x(133 22/2)
16. Make bar bending schedule, if required.

6.1.11 Approximate Amount Of Steel In Slabs


= 676 mm2 per meter width
A,fy 676x280
= [2.4 mm
Approximate amount of steel in slabs is 0.07 kg per mm depth per 1.0 m2 of area. a = 0.85 =
x//b 0.85x18x1000
Example 6.1: Design a slab of 10 m x 3.75 m clear dimensions supported over 342 mm
thick walls on all the four sides. This slab is part of a residential house. Use C-18 concrete 20.78x10* = 650 mm2 per meter width
and Grade 280 steel. Use US Customary bars and prepare bar bending schedule. As "
0.9x280x(133-12.4/2)
‘Solution: 650x280 _ 119 mm
The slab is supported on all the four sides and longer to shorter dimensions is 10 / 3.75 = 2.67.
Hence, the slab is one-way along the 3.75 m side. For the calculation of slab depth, the span
length may be considered equal to the clear dimension plus half brick length bearings on the
0.85x18x1000
20.78x10*
_ $49 mm2 per meter width
twosides. (k9x280x(133—11-9/2)

Amin = £125 = — = 159 mm (say 160 mm)


Diameter And Spacing
Using #10 bars: #10 @ 100 mm c/c provides As
#13 @ 190 mm c/c provides As = 679
mm2
= 710 mm1
d = h-27 = 133 mm Using #13 bars:
£ = clear span + h = 3.91 m
ONE-WAY AND EDGE
I SUPPORTED SLABS 197
196 CONCRETE STRUCTURES PART -I AUTHOR ZAHID A. SIDUiqj CHAPTER 6:
Maximum preferred spacing: least of Number of bars for M-2 = 10,456/360x2 = 55 51)
Number of bars for D-l = 3750/200 = 19 (say
i) 2A ~ 320 mm (Code value is 3/i)
Bottom length of bars D-l = 10,000-535 + 175 = 9650 mm
ii) 300 mm (Code value is 450 mm) = 10,000 + 2 x 180 = 10.36 m
.... _159,600 „ .
Finished length of bars D-l 10,360 + 0.414x86+18x 10 = 10.58 m
in)
159,600 ,
2.5 c. = 2.5x20 = 520 mm Total length D-l for estimation =
fy 280 Number of bars for D-2 = 3750 / 400 x 2 = 20
.iv), 126,000 126,000 Length of bars D-2 and M-2 = 750+180+18 x 10 = 1.110 m
= = 450 mm
4 280
Smax ~ 300 mm
Selected main reinforcement: #13 @ 190 mm c/c
Temperature Reinforcement

Temperature steel: 0.002 x b x h = 0.002 x 1000 x 160 — 320 mm2


Using #10 bars: #10 @200 mm de provides As = 355 mm2
Maximum preferred spacing: least of
i) 2.5h = 400 mm (Code value is 5h)

....
iu)
159,600
_ _ cr =
ii) 375 mm (Code value is 450 mm)

2.5
159,600 „.
2.5x20 - 520 mm
fy 300

^niax
..
iv)

375 1HI11

126,000
t
fy
= ——— =
126,000
280
450 mm

Selected temperature reinforcement: #10 @ 200 mm c/c


Check For Shear
= g0(4/2-d)
= 10.87 x (3.75 /2 - 0.133) = 18.94 kN
= 0.75 x 0.17 b^d
= 0.75 x 0.17x1x718 x 1000 x 133 /1000 = 71.94kN
The applied shear force is significantly lesser than even / 2.
Sketch Of Reinforcement The reinforcement details are shown in Fig. 6.2.
Bar Bending Schedule The bar bending schedule is given in Table 6.4.
Number of bars for M-l = 10,456 / 190 = 55.03 (say 55)
Total length of bars M-l = 3750 + 2 x 180 = 4110 mm
Length M-l for estimation = 3750 + 2 (200 - 20) + 0.414 x 109 + 18 x 13
= 4.389 m
ONE-WAY AND EDGE
SUPPORTED SLABS 199
1 98 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
CHAPTERS:
Hanger bars of # 10 for M-l and M-2 are provided at a spacing of 375 or 450 mm c/c 1
factored Wall Load / 1000 x Im x Im x 1920 x
9.81 1 1000 = 5.65 kN/ m
424 Pa = 1.2 x 250
Amount of slab steel Fper unit area = = 9.82 kg / mi
10.40x4.15 Factored Bending Moment
The cross-section of the slab marked in Fig. 6.2 is shown in Fig. 6.3.
M = 1 / 2 qa t2 + Pa x t
12.6 kN-m per meter width
= 1/2 x 11.46x 1.072 + 5.65x 1.07 =
Main Reinforcement
Assuming a = d / 6 = 20 mm
12.6x10*
A = hf^-a/2} = 0.9 x 280 x (98 -20 /2)

_- 44—
= 568 mm2 per meter width
Fig. 6.3. Section 1-1 for One-Way Slab.
iI a
— -
0.85x/cb
~
568x280
= 0.85x18x1000 = 10.4 mm
Example 6.2: Design cantilever projecting out from a room slab extending 1.0 m to be
used as part of balcony (live load = 300 kg / m2). A brick Wall of 250 mm thickness (inclusive
of wall finishes) of Im height is provided at the end of the cantilever. Use C-18 concrete and ; as
.
- —- —— ————
12.6XI06
0.9 x280x (98 -10.4/2) —
= 539 mm per meter width
2

— —
i
Grade 280 steel. The thickness of slab of room is 125 mm and the slab bottom steel in the
539x280
direction of cantilever is #13 @ 190 mtn c/c, out of which alternate bars are bent up. j
a -
= 0.85x18x1000 = 9.9 mm
Solution:
Cantilever slab is to be designed as a one-way slab.

d ~ h- 27
=
A™ =H 12 (1000 + 114) / 12 = 93 mm
(Use h = 125 mm, same as the room slab)
~ 98mm
1

1
A
.

Diameter And Spacing



- ——————
12.6x10*
0.9 x 280 x (98 -9.9 /2)

Steel already available


= 538 mm
2

= #13 @ 190 mm c/c


per meter width

/2 = 679 / 2 = 339mm2
t = clear span + A =
= 1 +0.125 /2 1.07 m = 199 mm2
= 538 -339 mm2or increase to #13 @ 380 mm c/c)
Extra steel required
Dead Load (Use #10 @350 mm c/c
(OK)
R. C. slab:

125
1000
75
x 2400 = 300 kgs / m2 2A = 250 mm c/c
Temperature Reinforcement
xh = 0.002 x 1000 x 125 = 250 mm2
75 mm screed of brick ballast: x 1800 = 135 kgs / m2 Temperature steel: 0.002 x b
258 mm2
#10 @ 275 mm c/c provides As =
1000
P. C. C. + terrazzo:

9d

1000
x 2300 = 138 kgs /m2
= 573 kgs / m2
Maximum preferred spacing: least
of
i) 2.5A = 312 mm (Code
value is 5A)
Live Load mm)
ii) 375 mm (Code value is 450
For residential building: q^ = 300 kgs /m2 159,600 = 520 mm
....
m)
159,600
• 2.5c = ! 2.5x20
Factored Slab Load 280
4
= 1.2^d + 1-6?l
q^
= (1.2 x 573 + 1.6 x 300) x 9.81 / 1000 = 11.46 kN / m’
= 11.46kN/m per meter width —
.iv). 126,000
4
=
126,000
280
= 450 mm
ONE-WAY AND EDGE
SUPPORTED SLABS 201
CONCRETE STRUCTURES PART - 1 6:
200 AUTHOR: ZAHID A. SIDDlQi CHAPTER
= 1 1 30 = 9999 = 127 mm (say 130 mm)
•• ^max 3 12 mm Amin for end panel 30
Selected temperature reinforcement: #10 @ 275 mm c/c d = h- 27 = 103 mm
The reinforcement details are shown in Fig. 6.4,
Dead Load
R. C. slab:
-129- x 2400 = 312 kgs /m2
1000 .
= 300 kgs /m
Floor finish: = 200 kgs /m1
Partition load:
gD = 812 kgs /m2
For office building: oL = 250 kgs / m2
Live Load
Factored Load
“ 1.2 ^d+ 1.6 qt
= (1.2x812 + 1.6x250)wi x i.81 /1000 = 13.48 kN / m2
= 13.48 kN /m per meter

Example 6.3: Design a slab consisting of eight panels of 8 m x 3. 5 m clear dimensions, = 3.5 m
that are continuous along their longitudinal edges and are supported on 300 mm wide beams.
Office live load is to be used along with a floor finish load of 300 kg/m2 and 200 kg/m2
4
Exterior support Mi" = 1/24 - -F^3- = MH"*'"
immovable partition load. Use C-20 concrete. Grade 280 steel and US Customary bars.
Solution: Exterior span Mi - - -48^L -
11

A unit strip of slab, taken along the shorter direction, acts as a continuous beam and is shown
in Fig. 65. First interior support M” - - -
l/10»,4! 1M1 W-/-

r -1 I —1 - 4 -p-H-
Interior support Mu" =1/11«.4Z - - 15.01 kN«/m

J Interior span Mi+ = 1 / 16 = 10.32 kN-m / m


Main Reinforcement: The amounts of steel are given in Table 6.5.
j | 1 m strip | | | Check For Maximum Spacing Of Main Steel
i i i i i 8“
Maximum preferred spacing: least of
j„_ _ i i i i i) 2k = 260 mm (Code value is 3h)
l ii) 300 mm (Code value is 450 mm)
.... __
38m
'
in)
fy —
-—2.5 = — 280
„c
159,600 , f c, 159,600 -2.5x20 = 520 mm
Fig. 6.5. Continuous One-Way Slab of Example 6.3. .iv),

= 260 mm
126,000
fy —-— — —
126,000
= 280- = 450 mm

t = 3500 + 300 = 3.8 m


STRUCTURES PART I - adTHor.- 24hjd E CHAPTER 6:
ONE-WAY AND EDGE
SUPPORTED SLABS 205
204 CONCRETE
The
reinforcement
__ —
«io@32L^s!
details, using curtailed ban

1

#10 @300
— /
J 1. The span length and loading with respect to the horizontal plan are considered
for the calculation of bending moments.
2. The self-weight of the stair slab is first calculated in the inclined plane and is
then multiplied with 4r2 +T2 IT or approximately 1.22 to calculate the load
on the horizontal plan, where R is the riser and T is the tread.
3. In case only one or no edge of steps is supported on wails, the stair is considered
#10®250 c/ci to span longitudinally. However, the slab may be assumed to span along the
width of the steps if there is newel wall towards the inner side and both edges of
1 the slab are supported.
4. As a conservative estimate, the waist of stairs may be selected according to one-
||& '
edge continuous one-way slabs. However, due to the inclined nature and
I availability of more stiffness, the waist dimension may be selected equal to both
ends continuous one-way slab (L/35 for Grades 280 and 300 and Z/28 for Grade
ts 420). In case the landing is also supported along the other edges, the span of
I- #10 @160 c/c u #10 @160 c/c ; stair may be considered up to the center of the landing. However, in this case,
- #13 @ 180 c/c #13 @200 c/c
. the landing must be designed to cany all die corresponding loads along a
- #10 @25°
dQ
: direction perpendicular to the stair.

67 Reinforcement Details for Slab of Example 6.3 Using Curtailed Bars. 5. If the steps are made up of reinforced concrete, some minimum steel is to be
* provided within these steps.
6. A small and usually concealed beam, in-between the landing and the flight of
Remforcemenl Value stair, is beneficial to keep the depth of stair slab and the required reinforcement
T^pe of Steel Length
(mm) J
i
in the economical range.
^extension from face of exterior
T°P d for short and long direction.
4/5 700 ; 7. Tension steel making an angle less than 180° and present on the inner side of
: this angle may cause falling of the concrete cover and loss of tensile force (Fig.
suPP0^ bars curtailed from face of
Altenl snorts, in short direction.
4/8 435 j 6.8). The detailing must be carried out to eliminate this situation.
Example 6.4', Design the first flight of the stair shown in Fig. 6.9, having a reinforced
bars curtailed from face of same as 435
in short —
concrete footing at the bottom. Use C 18 concrete and Grade 280 steel. R = 180 mm and T
direction = 260 mm. Select US Customary reinforcement
inn of top extra steel
on interior supports, 0.34 1050 Solution:
60 of top extra steel on
intenor supports,
0.224 770
I = 1.2 + 3.12 = 4.32 m
remaining alternate
face of supports, for Am™ considering both ends continuous/fixed = £ / 35
bars.
= 4320 / 35 = 124 mm (say 125 mm)
d = h-21 =98 mm
npsi2Ji Of Stair Stab Dead Load
6.1’^ ath the stairs is designed as one-way slab for the expected live loading,
The slab jea£| load of steps and dead load of
floor Anishes. The R. C. slab: x 2400 x 71802 + 2602 /260
dead I03** for stair is called its waist dimension. Following points are to be 1000
= 365 kgs / m2
& 2400
nz u ofr steps:
Weight —x
2 1000
.
= 216 kgs /m2
ONE-WAY AND EDGE
206 CONCRETE STRUCTURES PART -1 AUTHOR ZAHID A.
I CHAPTER 6: SUPPORTED SLABS 207
SIDDIQI
15 mm floor finish: 15
} factored Bending Moment
1000
x 2300 = 35 kgs / m Mu = 1 / 10 qv (one end continuous)
gD = 616 kgs /m2 = 1/10 x 11.96x4.322 = 22.4 kN-m per meter width

4in for singly reinforced section

'0.205x18x1000
22.4 xlO6
'

=

78 mm

Main Reinforcement
Assuming a — d! 6 = 16 mm
22.4xl06
S
fhfy(d-al2) 0.9 x 280x(98-16/2)
= 988 mm2 per meter width
a = —
A,fy1
0.85 x/cA
=
988x280
0.85x18x1000
=18 mm

_ ggg mm2
As = — 22.4x10
0.9 x 280 x (98 -18/2)
meter width
999 x 280
Fig. 6.8. Incorrect Ways of Placing Stair Slab Steel. a = = 18.3 mm, A* = 1001 mm2 per meter width
0.85x18x1000
Diameter And Spacing
Selected Steel = #13 @ 120mmc/c
2A = 250 mm (OK)
Temperature Reinforcement
Temperature steel: 0.002 x b x h = 0.002 x 1000 x 125 = 250 mm2
2.5 h = 312 mm
Selected temperature reinforcement: #10 @ 275 mm c/c
Check For Shear

Fig. 6.9. Plan View of Stair for Example 6.4.


Fil = q* (4 / 2
=

d)
11.96 x (4.32 /2 -0.098) = 24.66 kN
Live Load
= 0.75 x 0.17 Jfc Aw d = 0.75 x 0.17718 x 1000 x 98 / 1000 = 53.0 kN
For stairs:
= 300 kgs /m2 The applied shear force is significantly lesser than even ^Kc/ 2.
Factored Slab Load Curtailment Distances
qa = 1.2 gp + 1.6 4/7 = 4320 / 7 = 617 mm (say 610 mm)
= (1.2x616 + 1.6x300) x 9.81 /1000
= 11.96 kN/m per meter width
= 11.96 kN/m2 4/5 = 4320/7
Inclined 0.22 4
— 864mm (say 870 mm)
= 0.22 x 4320 x 1.22 = 1160 mm
Inclined 0.30 4 = 0.30 x 4320 x 1.22 = 1580 mm
ONE-WAY AND EDGE
208 CONCRETE STRUCTURES PART -I C}IAPTER6: SUPPORTED SLABS 209
AUTHOR: MH1Da
Reinforcement Details Whereas, foe continuous slab edge is shown with hatching (Fig. 6.11). Nine different
The reinforcement details are combinations are possible for different combinations of the edge conditions.
shown in Fig. 6.10.
#13 @ 120 c/c
Simply Supported Edge

#13 @ 240 c/c Waist =125


Continuous Edge
#13@120e/c 1 - #10 (typical)
#6 @ 300 c/c (typical)
-
1200 260
= 940>4/5 Fig. 6.11. Two-Way Slab Panel with Edge Conditions.
#10 @275 c/c
No negative moment #13 @ 120 c/c
capacity at this section, 2. The ratio of longer to shorter span lengths should not exceed 2.0. The longer span may
only allowed if after / 5 be denoted by and the shorter by 4-
4
If the distance is lesser than / 5
4 3. Unit width strips are taken in both shorter and longer directions to design the slab. The
or if more rigidity / continuity is #13 @ 240 c/c strip along shorter direction supports more load and hence its steel is greater than the
present at this end, the detail
may longer strips, which is placed towards the outer edges of the slab to get more effective
be made as follows:
depth. Each strip has positive moment and bottom steel near the mid-span and negative
- #13 @120 c/c moment and top steel near foe continuous end.
#13 @ 120 c/c 4. Moment for each direction strip may be calculated by the following general expression:
#13 @240 c/c Af= C?u/2
Where is foe factored slab load, t is foe clear span length of foe strip and C is a
coefficient, which may be obtained from foe Tables 6.6 to 6.10. These coefficients are
separately given for dead load positive moments, live load positive moments (BS8110
are combined for all foe positive moments) and negative moments.
Fig- 6.10. Reinforcement Details for Stair. The coefficients give foe approximate slab analysis results taking into account foe
62 TWO-WAY EDGE SUPPORTED SLABS pattern loading, redistribution of moments and correction to face of foe supports. Each
table has cases of edge conditions along the horizontal direction and the shorter to
. If a slab panel is supported on all the four longer span ratios along the vertical direction, foe body of the table giving the
than 2.0, the load is distributed parallel edges and the longer to shorter span
to ratio is lesser corresponding moment coefficients.
This is called two-way action of slabs. shorter and longer directions to all the four
supports. The original ACI positive moment coefficients are too low and the negative moments
be the main steel. A unit width The steel in both the short and long directions will now
may be are relatively high because of consideration of lateral and torsional stiffness of edge
corresponding steel required to provide foe considered in both the directions to find die beams. Further, the positive moments are significantly lesser than those given by
may be used with some flexural strength. Method-3 of the ACI Code 1963
modifications
sections. The remaining procedure for to estimate foe bending moments at various various other analysis techniques such as solution of governing differential equations or
the rectangular beams. The design is exactly same as that for foe one-way critical yield line theory. For local conditions, the positive moment coefficients are increased
following nomenclature and conditions are slabs and by 25% and a 10% reduction is made in the negative moment coefficients. Similarly,
slabs by this method: used in foe design of
the British Standards positive moment coefficients for slabs simply supported on all the
1. Each edge of foe slab is either four sides are significantly larger and are replaced by the actual solution for
supported by a wall or a beam of depth
equal to minimum depth of beam greater than or homogeneous plates. The negative moment coefficients of this specification are
for
including the edge where foe slab is deflection control. The simply supported edge, adjusted up by a margin of 10%.
discontinued at a beam, is denoted by a line alone.
ONE-WAY AND EDGE
SUPPORTED SLABS 211
210 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. chapter 6:
SIDDIQI

Table 6.6. ACI 1963 Coefficients For Dead Load Positive Moments In
Slabs Increased by 25%.

Case Case Case Case Case Case Positive Moments In
Ratio 1 2 3 4 5 6
Case
7
Case Case Table 6.7. ACI 1963 Coefficients For Live Load
9 Slabs Increased by 25%.
m
a 1 I Lt
1.00 0.045 0.023 0.023 0.034 0.034 0.041 0.034 0.025 0.029
— : Case
1
Case
2
Case
3
Case
4
Case
5
Case
6
Case
7
Case
8
Case
9
1
Cy 0.045 0.023 0.034 0 034 0.023 0.034
0.95
Cy 0.050 0.025 0.026 0.038 0.035 0.045
0.038
0.039
0.029
0,028
0.025
0.030
o ^£3
G 0.041 .0.020 0.031 0.030 0.019 0.030 0039 0.026 0.040 0.044 0.040 0.035 O.Ujo

0 90 J 0.056 0.028 0.031 0.041 0.036 0.049 0.044 0.031


0.021
0.033 1.00 cx 0.045 0.034
0.034
0.034
0.040
0.040
0.040 0.034 0.040 0.044 0.038 0,035
Cy 0.045
c 0.036 0.018 0.030 0.028 0.016 0.026 0.035 0.024 0.019 Cy 0.050 0.038 0.039 0.044 0.043 0.048 0.045 0.039.
0.85 cx 0.063 0.030 0.036 0.045 0.039 0.053 0.050 0.036 0.035 0.95
Cy 0.041 0.031 0.036 0.036 0.030 0.036 0,040 0.034 0.031
0.033 0.015 0.028 0.024 0.014 0.021 0.031 0.021 0.049 0.046 0.053 0.050 0.044 U.V4J
0.016 Cy 0.056 0.043 0.044
0.80 cx 0.070 0.033 0.043 0.049 0.040 0.056 0.056 0.040 0.036 0.90 r 0.036 0.028 0.034 0.033 0.026 0.031 0.036 0.030 0.028
Cy 0.029 0.014 0.025 0.020 0.011 0.019 0.028 0.019 0.013 0.050 0.054 0.051 0.058 0.056 0.050 0.04y
Cy 0.063 0.046
0.75 cx 0.076 0.035 0.050 0.054 0.041 0.060 0.064 0.045 0.039 0.85 c, 0.033 0.024 0.030 0.029 0.024 0.028 0.033 0.028 0.025
Cy 0.024 0.011 0.023 0.016 0.009 0.015 0.025 0.016 0.009 0.051 0.056 0.060 0.055 0.064 0.064 0.055 0.053
Cy 0.070
0.70 Cy 0.085 0.038 0.058 0.058 0.044 0.064 0.073 0.050 0.041 0.80 0.028 0.025 0.020 0.024 0.029 0.024 0.021

__
0.029 0.021
cvp 0.020 0.009 0.020 0.014 0.006 0.011 0.021 0.014 0.008
Cv
Cy 0.076 0.056 0.064 0.065 0.059^ 0.069 0.070 0.061 U.u3o
0.65 0.093 0.040 0.068 0.063 0.045 0.068 0.081 0.055 0.043 0.75 0.024 0.020 0.016 0.020 0.025 0.020 0,016
p 0.016 0.008 0.018 0.011 0.005 0.009 0.018
c 0,024 0.018
0.064 0.075 0.079 0.068 0.063
0.011 0.006 I Cy 0.085 0.061 0.071 0.071
0.60 UX 0.101 0.043 0.078 0.066 0.046 0.070 0.091 0.060 0.045 0.70 0.020 0.018 0.014 0.016 0.021 0.018 0.014
0.020 0.015
U> 0.013 0.005 0.014 0.009 0.004 0.008 ^Y
0.069 0.080 0.088 0.074 U.UDO
0.55 cx 0.110 0.044 0.089 0.070 0.048 0.073
0.015
0.101
0.009
0.065
0.005
0.046 0.65 cx 0.093 0.066 0.080 0.078
0.014 0.011 0.013 0.018 0.014 0.011
Cy 0.010 0.004 0.011 Uy 0.016 0.013 0.018 0.096 0.081 0.0/4
0.006 0.003 0.005 0.011 0.006 0.004 Cy 0.101 0.073 0.089 0.084 0.074 0.085
0.5 G 0.119 0.046 0.100' 0.074 0.049 0.076 0.111 0.070 0.048 0.60 0.011 0.009 0.010 0.014 0.011 0.009
Cv 0.013 0.009 0.014
Cy 0.008 0.003 0.009 0.005 0.001 0.004 0.009 0.005 0.003
^Y
0.090 0.079 0.091 0.106 0.088 o.u /y
Cy 0.110 0.078 0.100
0.55 0.009 0.006 0.008 0.011 0.009 0.008
r 0.010 0.008 0.011
0.096 0.084 0.098 0.115 0.095 U.0o4
0.119 0.083 0.110 0.009 0.063 1 0.005
0.5 0.008 0.005 0.009 0.006 0.005 0.006
212 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A.
SIDD[Q] 1
ONE-WAY AND EDGE
SUPPORTED SLABS 213
; cHAPTER 6:

Table 6.8. ACI 1963 Coefficients For Negative Moments In


Slabs Decreased by 10%. Table 6.9. BS8110 Coefficients For Positive Moments In Slabs Decreased.

Case Case Case Case Case Case Case Case Case


Case Case Case Case Case Case Case Case 8 9
Case 1 2 3 4 5 6 7
1 2 3 4 5 6 7 8 9
MVXM

0.043 0.030 0.029


cx
— — ——
0.034 0.044
1.00 C* 0.041 — 0.045 0.068 0.064 0.030 0.055 1.00
0.048 0.024 0.034 0.036
0.034 0.044 0.043 0.030 0.029
Cy — 0.041 0.068 Cy 0.048 0.024 0 034 0.036
0.95
Cy
— 0.045
0.037

0.065
0.045
0.050 0.071
— —
0.068
0.064 0.055
0.034
0.030
0.059 0.95 C*
cv
0.052
0.044
0.026
0.022
0.040
0.031
0.039
0.032
0.036
0.031
0.046
0.040
0.048
0.039
0.033
0.026
0.031
0.026

———
0.041 0.060 0.050

— — —
0.026 0.054 0.037 0.033
0.90 0.050 — 0.054 0.072 0.071 0.039 0.061 0.90 cx 0.057 0.028 0.047 0.043 0.038
0.028
0.049
0.035 0.035 0.023 0.024
c 0.033 0.063 Cy 0.040 0.020 0.028 0.028
0.036
0.85 C*
c. — 0.054
0.028

0.059
'

0.059
0.031
— — ——
0.074 0.075
0.056

0.051
0.047
0.044
0.041
0.023
0.065
0.019
0.85 Cx
Cy
Cx
0.061
0.036
0.067
0.031
0.018
0.034
0.054 0.046
0.025 0.025
0.061 0.050
0.040
0.025
0.042
0.052
0.031
0.055
0.061
0.031
0.068
0.041
0.021
0.045
0.035
0.021
0.037
0.80 Cl 0.059 0.064 0.075 0.077 -
0.050 0.068 0.80 0.028 0.019 0.019
Cv —

0.024
0.062
0.055
— 0.026
0.068

0.077— 0.079
. •"
0.046 0.037
0.055
0.015
0.070 .75
Cy
CK
0.033
0.072
0.016
0.036
0.022 0.022
0.067 0.053
0.022
0.044
0.028
0.058 0.074 0.049 0.039
0.75
° 0.016 0.016

0.025 0.025

0.70
Cy
G*

—— 0.020
0.067
0.050
'
0.022
0.073
— —
0.077 0.082
0.040 0.032
0.061
0.013
0.073 0.70
Cy
Cx
0.029
0.078
0.014
0.038
0.019 0.019
0.074 0.056
0.019
0.046 0.061 0.080
0.022
0.053 0.042
0.014 0.014
— — — 0.017 0.021
.

Cy 0.025 0.012 0.016 0.017


& 0.015 0.045

——
0.017 0.034 0:026 0.010 0.056 0.044
cx 0.080 0.059 0.047 0.065 0.086
0.65 C* 0.069 — . 0.077 0.078 0.084 . . 0.067 0.075 0.65
0.084 0.041
0.021 0.010 0.015 0.015 0.015 0.018 0.018 0.012 0.013
Cy 0.013 0.039 0.014 — — 0.028 0.022 0.007
Cy
0.087 0.063 0.049 0.067 0.093 0.060 0.046
0.60 Gx 0.073 — 0.080 0.079 0.086 0.072 0.077 0.60 Cx 0.090 0:043
0.013 0.016 0.015 0.010 0.011
Cy — 0.009 0.032 0.010 0.022 Cy 0.018 0.008 0.013 0.013

— 0.016 0.005 0.063 0.048


'
'

0.070 0.099
Gx — 0.076 Cx 0.096 0.045 0.095 0.066 0.052

———
0.083 0.080
"

0.086 0.077 0.077


0.55
Cy
C*
0.006
0.077
0.025

0.007 — —— — 0.017 0.013 0.005
0.55
Cy
Cx
0.015 0.007
0.102 0.048
0.011 0.010
0.100 0.069
0.011
0.053
0.014 0.013
0.074 0.105
0.008 0.008
0.066 0.050
0.5
... l Cy
— 0.005 0.020
0.085
0.005
0.081

0.087
"• 0.013
0.080
0.009
0.079
0.003
0.5
Cy 0.012 0.006 0.009 0.008 0.009 0.011 0.011 0.007 0.007
214
CONCREresraucruRESPMT , ONE-WAY AND EDGE
chaffer* SUPPORTED SLABS 215
’ MDDiqj
5. After calculation of moments at all the critical sections, the design is performed exactly
like beams or one-way slabs.

g_2.1 Basic Behavior Of Two-Way Slabs


When a two-way slab is divided into series of strips in tbe two directions, the question
arises that how much load out of the total load is taken by shorter direction strips and
how much contribution of load resistance is provided by the longer strips. To study this
aspect, the slab of Fig. 6.12 may be considered where two central strips are shown in
Table 6.10. BS8110 each direction.
Coefficients For Negative Moments
Slabs Increased by 10% h
Ca« 1 Case Case Case
2 Case Case Case
6
Case

0.036 0.052
0.036 0.051 0.064
0.050 0.050 0.044
0.95 0.039 0.063
0.057 0.053 0.041
0.033 0.045 0.068
0.90 0.045 0.049 0.046
0.041 0.057 0.037
0.029 0.040
0-062 J 0.055 0.072 0.037 Fig. 6.12. Design Strips in a Two-Way Slab.
0.85 0.040 0.055 0.049
0.046 0 067 0.051 0.033
0.026 0.036
J 0.058 0 033
The shaded portion is jointly supported by both the strips. One of the simplest
0.036 0.060 0.052
0.80 I 0.050 methods, which is the basis of similar coefficients in the British Standards is to find the
0.072 0.061 0.082 0 046 I 0.030 0.030
0.023 0.032 I 0.032 0.065 0 055 deflections at the common point for both the strips and evaluating their contribution by
0.75 0.053 equating these deflections.
0.70
0.020 0.028
0.056
0.077 0.064 0.086
0.028 °
0.027
0.071 0.058
0.036 0.023 0.024
Let = load taken by shorter strip
0.082
0.018 0.024 0.024 0.067 0.090
qy = load taken by longer strip
0.65
0.076 0.061 = total load on the slab
0.060 I 0.031 0.019 0.020
0.016 0.021
0.087 0.069
0.094 = + 9y
0.021 0.082 0.065
0.60 0.063
0.092 0.072 0.098 0.027 0.018 0.018
and m = 47-4
0.014 0.018 0.087 0.068 Then,
0.55 0.018
0.067 5^22 0.015 0.015
0.096 °-074
1.011 0.015 0.015 JOJ03J 0.093 0.072 =
0.5 0.069 384F7 384E7
0.102 0.077 0.107 0.019 0.013 0.013
0.009 0.013 0.012 i
0.097 0.074 xm4 = qt- q*
0017 I 0.010 0.011 qx x m4 = qy or q*

1 + m1
For square panel, m = 1, qx = 0.5 qt
For£y = 2x4, m = 0.5, qx = 0.941 qt
For m =1, = qx x 78 = 0.0625 x qt x : Cx = 0.0625
Form = 0.5, = qx^&xH = 0.118 x^txfx2 ; Cx = 0.118
ONE-WAY AND EDGE
216 CONCRETE STRUCTURES PART - 1 SUPPORTED SLABS 217
AUTHOR: ZAHID A.
SIDDIQI CHAPTER 6:
steel and increase the above
6.2.2 Minimum Slab Thickness For Two-Way Slabs 6. Calculate steel for each moment. Check for the minimum
According to the ACI 1963 Code, the minimum thickness for two-way steels if these are less than the minimum steel.
curtailed bars, alternate bottom
taken equal to 1 / 180 times the inner panel perimeter. However, slabs is to be 7. Decide bending up and curtailment of bars. In case of
the center
for steels steel may be curtailed at every continuous edge at a distance of 0.154 from(ACI) or it

Amin
inner £perimeter _ _
strengths greater than 300 MPa, it is better to increase this
thickness by 10%.
for Grade 300 steel with a minimum of 90 mm
of yield
of support. At discontinuous supports, either there should be no curtailment
may be done at 4 1 20 for one-third of the bars. A minimum of one-half of the
adjacent panel by a distance
negative
of 0.3 4
180 steel at a continuous edge is to be carried into the
from the face of support, while the other half may be curtailed at a distance of 0.22 4
Amtn for Grade 420 steel with a minimum of 90 mm extend 4by
165 from the face of support. The negative steel at discontinuous edges should
Usually it is preferred to keep the minimum depth larger than or equal 1 5 from the face of support.
to 110 mm
8. For bent-up bars option, one-third bars may be bent
up at a distance of 4 / 7 from face
According to the latest code, if 4 is the clear span in the short direction, / 5 from face
depth is taken as follows: the minimum of the discontinuous supports and the negative top steel is extended by 4
by 4 / 5 from the
of supports. The extra top steel at a discontinuous edge must extend continuous
_ 4(0-8+4/1400)
-
face of the support. A minimum of one-half of the negative steel at a
edge
the face of support,
is to be carried into the adjacent panel by a distance of 0.3 4 from
"mn ' ~ not less than 90 mm
36m + 9
while the other half may be curtailed at a distance of 0.22 4 from the face of support
i reinforcement is in
to the top slab
6.2.3 Design Procedure For Two-Way Slabs < 9. Most common type of temporary support provided
1. Draw the slab panel with its internal dimensions and and the form of steel chairs as shown in Fig. 6.13.
4
the case of slab with respect to the edge conditions. 4 end conditions. Decide
2. Determine the aspect ratio, m = 4 / 4, and decide whether the
slab is one-way or two-
way. If mis between 0.5 and 1.0, the slab is two-way. Fig. 6.13. A Typical Steel Chair.
3. Find the minimum thickness (hmjn) required for the slab and decide
the
after necessary rounding. If the slab consists of more than one panels, final thickness is provided for the
is usually selected and used for all the panels. Calculate effective
the largest depth ' 10. At a support that is in-between two different size panels, the steel
depth d for the short larger bending moment. Also the top steel is to be extended on both
sides according to
and long direction steels approximately, equal to h 27 and h 38,
- -
4. Calculate factored dead load, factored live load and factored totalrespectively.
load for the slab shorter panel becomes significantly less, extend the negative
of opposite edges for a
the larger span direction. If distance between negative steels steel throughout the
top
panel.
5. According to the aspect ratio and the case number according to i span.
the edge conditions, dimensions of the
11. In case a panel is shown to be continuous on any edge but the present adjacent
find coefficients for negative moments, dead load positive
moments and live load J adjacent slab are not given, always assume that similar type of panel is
positive moments for both the directions using the tables of
moment coefficients.

Further calculate the corresponding moments. I the panel to be designed.
12. Make a sketch or drawing showing all the reinforcement.
Short Direction 13. Complete bar bending schedule, if required.
Example 6.5: Design the marked four panels of monolithic RC slab, shown
in Fig. 6.14.
M~ per meter width Ux,neg X X 4 bar sizes are to be
Af dl per meter width Gx> dl X 1.2 ^dl X i The building is to be used for residential purposes and Us Customary
ATu, per meter width j selected. Use C-l 8 concrete and Grade 280 steel.

GxLL X 1.6 ^ll X 4
M* per meter width = Af*DL + Af LL Solution:
Long Direction
Check For Type Of Slab
= 0.64 * 0.65 (two-way)
AC per meter width
A4dl per meter width
— Cy^neg X X Panel-1: £x = 4.5 m, 4 = 7.0 m, m = Ly
(two-way)
CytDL X 1 .2 ^DL X 4 Panel-2: £x = 6.0 m, 4= 7.0 m, m = 4 / 4 = 0.86 “ 0-85
AT^ll per meter width Gy^LL X 1.6 ^LL X 4^ Panel-3: £x = -
3.5m, 4 6m, m = 4/4 = 0.58 ® 0.60 (two-way)
M+ per meter width = M DL + M +LL Panel-4: £x = 6.0m, 4 = 7.0m, m = 4/4 = ^-00 (two-way)
218 CONCRETE STRUCTURES FART -I ONE-WAY AND EDGE
AUTHOR; SUPPORTED SLABS 219
CHAPTER 6:
,, , (3500) (0.8 + 280/1400)
'
36^9 ’ I17n,m
„ . (6000) (0.8 + 280/1400) „
133mm

For equal depth, h — 150 nun


= 150 - 27 = 123 mm
+

Loads
d for short direction
d for long direction
— 150-38 = 112 mm

R. C. slab:

75 mm screed of brick ballast;



1000
75
x 2400

x 1800
= 360 kgs / m2
=
7
135 kgs / m
1000
P. C. C. + terrazzo: -
1000
- x 2300 = 138 kgs / m2
9d = 633 kgs /m2
For residential building: = 200 kgs /m2
1.2 qD = 1.2 x 633 x 9.81 /1000 x 1 = 7.45 kN/m
1.6 qt = 1.6 x 200 x 9.81 / 1000 x 1 = 3.14 kN/m
qv = 1.2 go + 1-6
= 7.45 + 3.14 = 10.59 kN/m
Fig. 6.14. Part-Plan of a House for Minimum Steel
Slab Design.
Edge Conditions: The four slab panels have the edge
6.15. conditions as shown in Fig.
A,nuii — = 0.002 x 1000 x 150 = 300 mm2
0.002 x 6 x h
Panel-1, x-direction m = 0.65, Case = 4, d= 123 mm

AT^ = x 1.2?«, x 42 = 0.063 x 7.45 x 4.52 = 9.51 kN-m

Af*

Q,u x 1.69u, x/x2 = 0.078 x 3.14 x 4.52 = 4.96 kN-m
= Af*DL + Af* ~ 14.47 kN-m
ll

Fig. 6.15. Various Cases ofEdge AT = CXjne8 x q^ x = 0.077 x 10.59 x 4.52 = 16.51 kN-m
Conditions for Slab of Example 6.4.
Slab Thickness For A^, R = / bd2 = 14.47 x 10s / (1000 x 1232) = 0.96 MPa,
p=0.0040
Panel-1-. ^0-8 + /y/140Q) 4, = 0.0040 x 1000 x 123 = 492 mm2 > Amin (#10 @ 140 mm c/c < 2A)
367,7+9 not less than 90 nim
For AT, R-M^/bd2 = 16.51 x 106 / (1000 x 1232) = 1.09 MPa,
= (4500)(0.8+28Q/1400j~ 140 mm
& 0.0046
Panel-2. Amin
- 36 x0.64+ 9
(6000) (0.8 +280/1400}
36x086 + 9” = 150 mm
A = 0.0046 x 1000 x 123 = 566 mm2 >A,niin
ONE-WAY AND EDGE
SUPPORTED SLABS 221
220 CONCRETE STRUCTURES PART - 1 CHAPTER 6:
AUTHOR: ZAHID A. / (1000 x 1122) = 0.79 MPa,
SIDDIqi ? For M~, R = MJbd2 = 9.86 x 106
Panel-1, v-direction m = 0.65, Case 4, d p=O.OO33 @ 190 mm c/c)
= = 112 mm As = 0.0033 x 1000 x 112 = 370 mm2 >4s,min (#10
ATdl = Cy>DL x 1-2?dl x ^y2 ~ 0.011 x 7.45 x 7.02 = 4-02 kN-m Panel-3, x-direction tn = 0.60, Case = 8, d= 123 mm
AAl = Cy>LL x 1.6?ll x ^2 = 0.014 x 3.14 x 7.02 = 2.16 kN-m A/^pl =CXjDl x 1.2? pt x /x2 = 0.060 x 7.45 x 3.52 = 5.48 kN-m
At = AA>l +AAl ~ 6.18 kN-m
AAl = Au 1.6?u. x x V = 0.081 x 3.14 x 3.52 = 3.12 kN-m
Af = Cy^g x x £2 = 0.014 x 10.59 x 7.02 = 7.27 kN-m At = AAl+AAl = 8.60 kN-m
AC = Cx^eg x ?u x t2 = 0.072 x 10.59 3.52 =
x 9.34 kN-m
For At, R = Ma 7 bd2 = 6.18 x 106 /
(1000 x 1122) = 0.49 MPa,
p= 0.0025
x 1232) = 0.57 MPa,
As = 0.0025 x 1000 x 112 = 280 mm2 =4^™. For At, R^M^ I bd2 = 8.60 x 106 / (1000
Use4s = 300 mm2 (#10 @ 225 mm c/c < 2ft) /7=0.0025 225 mm c/c < 2ft)
As = 0.0025 x 1000 x 123 = 308 mm2 >^Sjnin (#10 @
For AT, R = MJbd2 = 7.27 x 106 / (1000 x U22) x 1232) = 0.62 MPa,
p= 0.0025 = 0.58 MPa, For AC, R^MJbd2 = 9.34 x 10* /(1000
/7=0.0026
- 0.0026 x 1000 x 123 = 320 mm2 > A,mtn (#10 @ 200 mm
As = 0.0025 x 1000 x 112 = 280 mm2 (Use A,min = 300 mm2) c/c)
As

Panel-2, x-direction tn = 0.85, Case 9,
d- 123 mm
[#10 @ 225J

ATdl = CX,DL x 1.2? DL x 41 = 0.035 x 7.45 x 6.02 = 9.39 kN-m


Panel-3, y-direction m = 0.60, Case = 8, d=
112 mm

At pl = Cy>DL x 1.2?DL x I2 = 0.009 x 7.45 x 6.02 = 2.41 kN-m


AAl = Cx>Ll x 1-6?ll x i2 = 0.049 x 3.14 x 6.02 = 5.54 kN-m A(+u. = CyiLL x 1.6? ll * ^y2 0.011 x 3.14 x 6.02
- = 1.24 kN-m
At = AT^bl + Af +u ~ 14.93 kN-m At = At + M +LL ~ 3.65 kN-m
AF = Cx^eg x qa x £2 = 0.065 x 10.59 x
6.02 = 24.78 kN-m At = Cy^eg x ?u x t2 = 0.016 x 10.59 x 6.02 = 6.10 kN-m
ForAf*, R= Mu/bd2 = 3.65 x 106/(1000 x 1122) = 0.29
For At, R = Ma I bd2 MPa,
p=0.0042
= 14.93 x 106 / (1000 x 1232) = 0.99 MPa,
As = 0.0042 x 1000 x 123 = 517 mm2 > /7=0.0025 280 mm2
(#10 @ 130 mm c/c < 2ft) As = 0.0025 x 1000 x 112225= mm c/c < 2h)
For At, R = Afu / bd2 = 24.78 x IO6 / (1000 x 1232) Use4s = 300 mm2 (#10 @
x 122) = 0.49 MPa,
/»=0.0070 = 1.64 MPa, For AT, R = Afu / bd2 = 6.10 x 106 / (1000 1
- As = 0.0070 x 1000 x 123 = 861 mm2 >Am,n (#13 @ 150 mm c/c) /7=0.0025
As = 0.0025 x 1000 x 112 = 280 mm2 (Use 300
mm2) (#10 @ 225 mm c/c]
Panel-2, v-direction m = 0.85, Case
= 9, d ~ 112 mm Panel-4, x-direction tn = 1.0, Case = 2, d— 112 mm
Af*dl = Cy>DL x 1.2?dl x ^y2 = 0.016 x 7.45 x 7.02 Same as the y-direction below.
= 5.84 kN-m
ATu. = Cy.u. x 1.6? IX x ^y2 = 0.025 x 3.14 x 7.02 = 3.85 kN-m Panel-4, y-direction tn = 1.00, Case = 2, d= 112 mm
Af^ = AAl+AAl = 9.69 kN-m pt x ^y2 = 0.023 x 7.45
x 6.02 = 6.17 kN-m
At pl =
CyjDL x1.2?
Atu. = CyiU x 1.6?u x = 0.034 x 3.14 x 6.02 =
AC - Cy_neg x ?u x ^y2 = 0.019 x 10.59 x 3.84 kN-m
7.02 = 9.86 kN-m
For At, R = I bd2 = 9.69 x 106 / (1000 x
1122) = 0.77 MPa, At = A/+dl + A/\l = 10.01 kN-m
= Cy^g x ?y x ^,2 = 0.041 x 10.59 x 6.02 = 15.63
p= 0.0032 kN-m
At
As = 0.0032 x 1000 x 112 = 359 mm2 >Asjmia (#10 @ 190 mm c/c
< 2ft)
ONE-WAY AND EDGE
SUPPORTED SLABS 223
222 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A.
SIDDIQ!
ForAT, R
/7 = 0.0034
— Mu/bd1 = 10.01 x 10s / (1000 x 1 122) = 0.80 MPa,

As =0.0034 x 1000 x 112 = 381 mm2 (#10 @ 180 mm c/c < 2h)
For AT, R = MJbd2 = 15.63 x 106 / (1000 x 1122) = 1.25 MPa,
p = 0.0053 #10 @ 420 c/c #10 @140 c/c #10 @ 130 c/c
I #13 @ 400 c/c
A* = 0.0053 x 1000 x 112 = 594 mm2 >^in
P 1 r— #13 @ 350 c/c

Reinforcement Details:
The reinforcement details along the horizontal direction of the paper are shown
separately for clarity in Fig. 6.16. Similarly, the reinforcement details along the
vertical direction of the paper are shown in Fig. 6.17. The combined reinforcement
details, using bent up bars, are shown in Fig. 6.18. Figure 6.19 shows the
reinforcement details for the same example if curtailed bars are used.
Top Extra Steel:

Edges A&B = 861 -#10@140/2-#10@130/2 =335 #13@350


EdgeC - =594- #10@225/2- #10@190/2 =249 #10@275
EdgeD = 594-#10@190/2 -#10@190/2 =220 #10@300
EdgesE&G = 320-#10@225 / 2-#10@225 /2 = 004 No need
EdgesF&H = 594-#10@190/2-#10@180/2 =210 #10@300

Curtailment And Bent-Up Distances:

Span, 4

7000
4/8
875
4/7
1000
4/5
1400
0.22 4

1540
0.30 4

2100 770|
Ap18-^ 1800
6000
4500
750
560
860
640
1200
900
1320
990
1800
1350
Ai- 1320|-
3500 435 500 700 770 1050
trZ J 320
1320
Notes Related With Fig. 6.19. 132(

Type of Steel Reinforcement Value


#10 @ #10 J-t- #10 @300 c/c
Top steel extension from face of exterior support, for both . 4/5 450 c/c 225 c/c #10 @275 c/c i90c/c
directions.
Alternate bottom bars curtailed from face of inner supports, for 4/8
both directions. Fig. 6.16. Reinforcement Details in Horizontal Direction of Raper.
Extension of top extra steel on interior supports, from the face 0 34
of supports, for alternate bars.
Extension of top extra steel on interior supports, from the face 0.224
of supports, for remaining alternate bars.
224 CONCRETE STRUCTURES PART - 1 ONE-WAY AND EDGE
AUTHOR;
ZAIHDA CHAPTER 6: SUPPORTED SLABS 225

Fig. 6.17. Reinforcement Details in Vertical Direction of Paper.


Fig. 6.18. Combined Reinforcement Details using Bent-Up Bars.
STRUCTURES Part.; ONE-WAY AND EDGE
226 CONCRETE AUTHOR;
ZAHIDA.SJDDIQI CjIAPTER 6: SUPPORTED SLABS 227
PROBLEMS

Note: N = Registered number of the student.


No =
'

A fixed positive or negative number assigned by


the instructor to get R between 600 and 1000.
N~ No, a parameter to get numerical data for the
problems.

[, Design a slab of 72/70 m x J?/180 m clear dimensions supported over 342 mm thick
walls on all the four sides. This slab is part of a residential house. Use C-20 concrete
and Grade 420 steel. Use US Customary bars and prepare bar bending schedule.

2 Design the flight of a stair consisting of 10 steps with 1.3 m wide landings on both
ends, having a reinforced concrete beam with continuous slab at one end and wall on
the other end. The width of stair is 1.3 m. Use C-20 concrete and Grade 420 steel.
Riser is to be 180 mm and tread is to be 260 mm. Select US Customary reinforcement.

3 Design the panel of RC slab having length of 72/1 10 m and width of (1600 -72)/110 m.
The building is to be used for office purposes and SI bar sizes are to be selected. One
long edge is continuous, one short edge is continuous and other edges are
discontinuous. Use C-18 concrete and Grade 280 steel.

4. Design a slab consisting of eight panels of 72/75 m x 72/200 m clear dimensions,


continuous along their longitudinal edges, which are supported on 450 mm wide beams.
Office live load is to be used along with a floor finish load of R/2 kg/m2 and 200 kg/m2
immovable partition load. Use C-20 concrete, Grade 280 steel and US Customary
bars.

Fig. 6.19. Combined Reinforcement Details using Curtailed


Bars.
228 CONCRETE STRUCTURES PART - 1
AUTHOR; ZAHID A. SlD0lQl
Space For Reader’s Notes
Chapter - 7

BASICS OF SHORT COLUMN DESIGN

7,1 INTRODUCTION

Columns are the structural members that are subjected to axial compressive loads as the main
force, as shown in Fig. 7.1. However, these may also be subjected to simultaneous bending
moments. The cross-sectional dimensions of a column are generally considerably less than its
height. Columns support vertical loads from the floors and roof and transfer these loads to the
foundations. Depending upon the presence or absence of bending moment, columns maybe of
{ the following two types:
1 Concentrically loaded columns are those columns that are subjected to only axial load with
| only some accidental eccentricity of load producing considerably smaller bending moment.
Eccentrically loaded columns are those columns that are subjected to load at an eccentricity
resulting in a combined action of axial compression and bending moment
The column behavior mainly depends on its slenderness ratio, denoted by Kijr, where K is the
effective length factor depending on the presence or absence of side-sway and the end
conditions, £n is the unsupported length of column and r is the radius of gyration of the column
cross-section along the axis of bending. Side-sway is the lateral movement between the top
and the bottom ends of the column in the deformed shape. Depending upon the slenderness
ratio, the columns may be classified into two categories as under:
Short Columns: The slenderness ratio of such columns is so low that the instability, chances
of buckling and second order effects is eliminated. Second order affects usually means the
magnification of first order moments produced as a result of product of the first order column
[ deflections and the corresponding axial forces. The strength of these columns depends only on
i the material strengths and the cross-sectional dimensions.
Slender Columns: The strength of such columns may be significantly lesser than the short
columns depending upon their slenderness ratio and the first order lateral deflection. Moment
magnification, may take place due to second order effects. This type of column will be
discussed in detail in a later chapter.
The effects of slenderness for a braced frame ratio may be ignored and the column may be
considered to behave as a short column if:

where, Afj = magnitude of smaller end moment: Afe = magnitude of larger end moment
zero if moment is absent, positive for single curvature and negative for
M2
reverse curvature.
basics of SHORT column design 231
230 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. CHAPTER?:
SlDDlQl column where
P
amount of longitudinal steel is to be provided for concentrically loaded
concrete cross-section is sufficient to carry the loads.
2. To provide certain minimum ductility in the columns. movement by reducing the
3. To prevent reduction in column stiffiiess against lateral
crack widths.
4. To provide resistance against bending moments.
5. To increase the axial load carrying capacity.
transverse reinforcement. This transverse
The longitudinal bars are held together by the or spirals. The functions performed by
reinforcement may be in the form of closely spaced ties
Fig. 7.1. A Simple Compression Member. the ties are as under:
bars that may occur by breaking of
1. To prevent the outward buckling of the longitudinal
According to ACI 10.10.1, a compression member may be the concrete cover.
considered
elements have a total stiffness effective in resisting lateral movement of braced if the bracing 2. To provide some confinement to the inner concrete, preventing
its sudden collapse and
12 times the gross stiffness of all columns of that astory at least equal to
story. Lateral braces and shear walls are hence increasing ductility.
typical bracing elements. 3. To provide resistance against shear and shrinkage cracking.
The radius of gyration ‘r’ may approximately be taken 4. To hold the column steel at its proper position during casting of concrete.
provide the level of
Due to their relatively larger spacing, lateral ties do not gives relatively less
equal to 0.3ft for rectangular sections
and 0.25^ for circular sections, where ft is the 5.
column dimension in the direction of confinement to concrete as that provided by the spirals.
This
bending/buckling and d is the diameter of the column. For a concentrically tied column fails at
column with shorter side equal to ft having K 0.75, the limiting loaded rectangular ductility. The failure of a tied column is abrupt and complete. A
'
slenderness ratio is: outward along the inclined
the ultimate load when concrete fails by crushing and shears


0.75x£,
0.3ft =
34 ~
n
This means that if the unsupported height for a partially fixed
13.6
planes, and the longitudinal steel buckles outward between
shown in Fig. 7.2.
ties as is diagrammatically

braced
times its least lateral dimension, the column may be considered as column is less than 13.6
a short column. In other
words, if the center-to-center height for a partially fixed
braced column is Jess than 15 times its
least lateral dimension, it is a short column. Similarly, for a
concentrically loaded rectangular
column with least lateral dimension equal to ft and K = 1, / ft 10, and
for a concentrically
loaded circular column, £u / d= 8.5.
A circular section is allowed to be used in place of actual square,
octagonal or any other shape
having a diameter equal to the least lateral dimension of the
actual member. All the design,
including the minimum steel ratio, is performed using this
compression member is larger than that required
circular section. If the size of a
for the loads, due to architectural or other
reasons, it is allowed to base the design on reduced effective area, but
gross area. not less than half of the

7.2 TYPES OF COLUMN REINFORCEMENT Fig. 7.2. Failure of a Tied Column.

The main steel in columns is provided along the length of The spiral reinforcement consists of a continuous or
lapped wire wrapped around the
the
to the outer periphery, called longitudinal steel. The members in the comers or closer longitudinal steel in the form of spiral at a closer pitch (s), as shown
in Fig. 7.3.
functions of this main column
reinforcement are as under:
The spiral reinforcement provides the following functions:
1. To reduce creep and shrinkage in the columns.
Short or long term shortening in 1. It more effectively brace the longitudinal bars against their
outward buckling compared
columns acts like settlement of supports for the upper parts of the building
damage them. This is the main function of the main steel and and may with the ties.
hence a minimum
232 CONCRETE STRUCTURES
PART -1
2. ft provides AUTHOR: ZAHID A.
much more confinement SIDDIQ} cHAPTER7; BASICS OF SHORT COL UMN DESI GN 233
loads, the concrete to the inner core of
cover spalls concrete.
down. However, the spiral keeps Closer to the ultimate
and the inner 2. Unequal Spans On Both Sides Of Column
concrete core intact The the longitudinal Presence of unequal spans on sides of a column also produces resultant eccentricity and
becomes sufficiently larger and load carrying steel
The loads are may offset the decrease incapacity of confined concrete moment.
strength
lot of ductility.sustained for much longer times without final collapse,
due to loss of
cover. 3. Rigid Frame Action
3. It can hold the which provides If the joint between the concrete beams and the column is monolithic, it behaves as a
4. Hoop tension steel firmly during casting.
is developed in the spirals rigid joint and the end moment of the beam is transferred to the column.
due to internal pressure of
the concrete. 4. Eccentricity Of Load
Sometimes the load is applied on a column at an eccentricity through brackets or other
mechanisms as in case of crane girders producing considerably larger moments.

Fig. 7.5. Eccentric Column Load through Bracket.

e
5 pitch of the spiral

Fig. 7.3. Layout of Spiral


7.3 Reinforcement. 5. Out —Of -Plumbness Of Column
SOURCES OF MOMENT IN If the column is not exactly vertical, the load at the top produces some eccentricity at
COLUMNS the bottom.
Bending moments may
which are given below: be developed in the columns due to many 6. Out -Of- Straightness Of Column
different reasons, some of In case the column is not perfectly straight, some eccentricity and hence moment is
I. Unequal Loading
generated.
When load on both On Both Sides Of Column
sides
eccentricity with respect to of a column is unequal, tire resultant load 7.4 BEHAVIOR OF CONCENTRICALLY LOADED
Comer columns of a building the column centroid and will have
hence moment will be generated SHORT COLUMNS AT SERVICE LOADS
bending. Edge columns always have
have moment at leastmoments about the two possible axes
about one axis of bending of
(Fig. 7.4). At the service loads, the steel and the concrete behaves almost elastically. Further, there is no
cracking due to absence of tension. When the load is applied on a column, having both steel
and concrete in it, both the materials deform by same amount due to bond between the steel
and the concrete. The strain produced in steel («0 is equal to the strain produced in the
concrete (ij. If the concrete stress f is within the maximum allowable value, it behaves
almost elastically. For the section compatibility, considering the concrete stress fc and denoting
the steel stress by f, the following may be written:

f = nf
234 CONCRETE STRUCTURES PART - 1 235
BASICS OF SHORT COL UMN DESI GN
Where n is the modular ratio, defined c(IAPTER7:
as the extra concrete area
modulus of elasticity of concrete rounded to ratio of modulus of elasticity of ste However, after making compensation for the already present Ast, the
the nearest whole e 10 in Fig. 7.6(a) and its uncracked
-1) Ast. The actual column cross-section is shown


number. (n
Let Ag = gross or total area of the column becomes section for axial load and bending / buckling about axis AA is shown in Fig.
Am area of longitudinal steel in the column transformed multiplied with the concrete stress gives the load
Ac -
area of concrete in the column cross-section 7.6(b). The area of this transformed section strength of the actual reinforced concrete section.
P = total load capacity of the column cross-section carrying capacity of the column equal to the actually the
concrete strength in compression is considered to be 0.85 fc, which is
Ps = load resisted by the concrete The ultimate its ultimate strength). In an actual
column, shrinkage
and Pc = load resisted by the concrete crushing strength of the concrete (and not
the flexural rigidity (EI) of the column.
cracking, creep and bending significantly reduce
Ps fs X Ast

Pc fc* Ac fc (Ag Ast)

P ^Pc+Ps
~ fc Ac +fs ^st
= fc (Ag — ^jt) + nfc Ast

= fc[(Ag- Ast)+ n Ast]
fc[AB + (n-})Ast]
Uis formula is applicable only e|Mjc (o

7.5 STEEL YIELDS EARLIER THAN


CRUSHING OF CONCRETE IN COLUMNS
When the column load is gradually increased, the question
arises that whether the steel will
yield first or the concrete will
crush first. To answer this question, Fig. 7.6. Uncracked Transformed Column Section.
consisting of concrete strength^' 20 MPa and consider a column
= steel of strength of 280, 420, 460 or 520 concentrically loaded short column has a cross-
MPa. Example 7.1: A reinforced concrete 7.7). If
Ec = 4700^ =21019 MPa by 8 # 19 (US) bars of grade 420 (Fig.
=
Es 200,000 MPa sectional size of 450 x 450 mm and is reinforced 20 MPa), what will be the corresponding load
at any stage, the concrete stress/^ is 9 MPa
(fc'~
Ga, = 0.003 carrying capacity?
gy = 280 / 200,000 = 0.0014 for grade 280 steel 450
Sy = 420 / 200,000 = 0.0021 for grade Solution.:
420 steel
Sy = 460 / 200,000 = 0.0023 for grade
460 steel = 20 MPa
• Sy = 520 / 200,000 = 0.0026 for
grade 520 steel
This means that the steel will yield at a
f, = 420 MPa
lesser
When the load is less than that causing yielding strain or earlier than the crushing of concrete.
Ast = 2280 mm2 8# 19
the concrete per unit area. After of steel, the steel takes n-times more load than Es = 200,000 MPa
yielding of steel, the additional load is Ec
concrete. transferred to the Example 7.1.
Fig. 7-7- Column Cross-Section for
= 21,019 MPa
7.6 TRANSFORMED UNCRACKED SECTION OF COLUMN n = EstEc = 10
elastic.
72 Both concrete and steel will be approximately
Before cracking of the column concrete, the transformed uncracked section may be
an imaginary section made up of concrete
only but providing the same axial load
defined as = /^ + (*-1)^]
capacity and flexural resistance about any
particular axis of bending as that of
carrying = 9 x [450 x 450 + 9 x 2280] 1 1000
reinforced concrete section. The steel area the actual
transformed into an equivalent concrete area is n = 2007 kN
BASICS OF SHORT COL UMN DESIGN
t CHAPTER. 7: 237
236 CONCRETE STRUCTURES PART -I
author- zahid
7.7 CAPACITY REDUCTION FACTORS FOR
COLUMNS
The capacity reduction factors (fl for columns are lower
than the h
relative importance of columns and relatively less ductility giving of more
A column failure may cause the collapse of a complete
structur/ wh^ of
affect a local portion. The reduction factor is 0.65 for tied cnmnrp ’ • hereas’ a ^am niav Oni,
is 0.75 for spirally reinforced compression controlled P C°ntrolIed columns and
columns. Fig. 7.8. Column Cross-Section for Example 7.2.
Additional reduction factor is also applied on nominal strength
nf
columns to take care of accidental eccentricities, which is 0 80 tor H C0”centricany loaded
7.10 MINIMUM AND MAXIMUM STEEL RATIOS IN COLUMNS
spirally reinforced columns (ACI 10.3.6.1). These values approximate
at e/h ratios of 0.05 and 0.10, specified in the earlier codes. P
°85
* Wa ioad
strengths jlie steel ratio in columns is defined as the total amount of longitudinal steel in the cross¬
7.8 ULTIMATE STRENGTH OF CONCEN¬ section divided by the gross area of the section.
TRICALLY LOADED SHORT COLUMNS

For a slow rate of loading, much lesser than prescribed for a cylinder
reliable strength of concrete is approximately 0.85 f'c As the load' '
t r
lest’ max’mum The minimum steel ratio in columns is 1%, which is provided to strengthen concrete against
excessive creep. If steel is provided less than this amount, creep occurs in concrete and strain
'becomes conXTf
the steel yields first and its contribution to strength 'S
increased> increases without increase of stress. This transfers load to the steel at the service load stage up
Hence, the total nominal strength of a column becomes the yield
its area and 0.85 /c times the concrete area.
st™ JJ
r s^e
d stress leveL
multiplied by to which the steel may not have yielded. This phenomenon continues until steel yields and
both materials undergo large strains. The lower limit is also necessary to ensure resistance to
= 0.85/Ue +fy Aa = 0.85/' Ag+(fy- 0.85 bending moments not accounted for in the analysis.
The design strength becomes the following: The maximum steel ratio given by the ACI Code is 8% to avoid congestion of steel,
Axially Loaded Spiral Short Columns particularly where it must be spliced. The practical values of steel ratio range from 3 to 6% as
the upper limit for easy placement of concrete. The most economical tied column section
= 0.85 x0.75 [0.85/'c (.4 -4,.^ generally involves pu from 1 to 2%. As a result, tied columns seldom have pA greater than 3%.
Because spiral columns are used for high axial loads, they generally have steel ratios between
Axially Loaded Tied Short Columns
2.5 and 5%.
»0.«0x0.65[M5AW,-4.)+/MJ According to ACI 10.9.2, the minimum number of longitudinal bars in compression members
is to be 4 for bars within rectangular or circular ties, 3 for bars within triangular ties, and 6 for
“elate the maximum load capacity ofthe tW co|mn bars enclosed by spirals. Almost universally, an even number of bars is used in a rectangular
of
column so that the column is symmetrical about the axis of bending and usually all the bars are
Solution: of the same size. According to ACI 7.6.3, the clear distance between longitudinal bars must
not be less than larger of 1 .5d/, and 40 mm.
fc = 20 MPa : fy = 420 MPa
Au = 2280 mm2 : Ag = 450 x 450 mm2 The minimum steel ratio required for a beam is significantly different from a column because it
is provided for a totally different purpose than for the later as explained above. In case of a
= *0.65 [0.85 f'c(Ag~Aa)+fyA^
M0
-0-52 x [0.S5 x 20 (450 x 450 - 2280) + 420 x
22S0] / 1000 = 226SKN
beam, if the flexural strength of the cracked section is less than the amount that caused this first
cracking, the beam will fail immediately and without any warning. Hence, a minimum value of
steel is to be provided with larger margin of safety. Similarly the purpose of minimum steel in
7.9 MINIMUM CONCRETE COMPRESSIVE case of roof and footing slabs is totally different from beams and columns. A minimum steel
STRENGTH
of 0.18 to 0.20% of gross area is provided in slabs to control temperature and shrinkage
In order to keep creep and shrinkage of columns within acceptable stresses. Further this steel may be used to evenly distribute the load on main steel in the
compressive strength should be kept not less than 20mp a for limits, the concrete
buildings.
important and multistory
238 CONCRETE
STRUCTURES PART -I AUTHOR; BASICS OF SHORT COLUMN DESI GN 239
ZAHID A SIDD[Qi I CHAPTER7:
jicular direction of the slab.
perp®11 beams have only one-wayThe minimum steel in slabs is lesser than that of
action and are subjected to simultaneous action beams
c) Least lateral dimension of the column.
may be provided at
of d) According to ACI R7.10.5, circular ties in a circular column
whereas the slabs have two-way and have considerably lesser shear higher spacing, calculated just like for rectangular columns, and only one tie in this spacing is
shear sjabs only support loads acting directlyaction forces,
above them as compared with beams that sufficient These circular ties may also be provided in the form of helix with pitch
further^ joaxjs over a width. However, the minimum steel in slabs still have
equal to the required tie spacing. According to ACI 7.10.5.3, the ends of the circular

sU^t capacity of 1.1 to 1.5 times the flexural cracking moment and is capable of provides a
ffioin .jen failure at the first cracking. preventing ties must overlap by a minimum of 150 mm and must terminate in a standard hook
thes°° engaging a longitudinal bar. These overlaps are staggered along the height of the
. column.
7,11 MINIMUM CLEAR COVER
ANp SPLICING °f BARS e) 300 mm (not an ACI requirement).

clear cover for sheltered columns not in


c°n
Sround should be at least
(1.5 in). For unsheltered columns
4®^ng bars greater than No. 16 (US), clear Spacing < 150mni
is to be increased to 50mm (2 in). For
c°y in contact with earth, the minimum
_-
1 ar co^
should be 75mm (3 in).

xVhen detailed
u calculations
..
. are not made
, for _
ing hi compression, the overlap of
SPr1
-wdinal barsbe at the base and at each floor
1 should more than 35 times the
^’aflieterof the bar, as shown in Fig. 7.9. Fig. 7.9. Bar Splicing in Columns.
absence of detailed calculations for splicing in tension, the overlap of
c-milarly> 111
? ^tudinal
long11
bars at the base and at each floor level should be more than 50 times the diameter
of the bar-
7!2 AMOUNT of transverse ties
p r th® de51®11 of transverse ties, three parameters are to be decided, which are: (a) diameter of
shape of ties and (c) spacing of ties. The diameter of ties should be at least No. 10(US
ties (b) longitudinal
\’si) for bars upto No. 32 US (No. 30 SI) and at least No. 13 US (No. 15 SI) for
Fig. 7.10. Typical Shapes of Column Transverse Ties.
larger diam®161
longitudinal bars.
pot th® different criteria are to be satisfied. (I) The ties must be placed such
every comer longitudinal bar, every alternate bar and bars having clear spacing greater than
7.13 AMOUNT OF SPIRAL REINFORCEMENT
that (6 in) should be
enclosed by a tie having an included angle of not more than 135°. (2)
150mm ties should have Figure 7.11 shows the behavior of concentrically loaded short columns as the load is gradually
-pjje lateral proper hooks at the ends. (3) The ties should not have any bend increased. The first maximum load' is reached after the nominal capacity of the column
is
uere there is no longitudinal bar (ACI 7.10.5). According to these criteria, some typical tie- exhausted. The concrete shell around the longitudinal bars and ties or spiral spalls and the load
shap£s are shown in Fig. 7.10. carrying capacity is reduced due to reduction in the concrete area. A tied column collapses
column,
spacing of the ties should be smaller of the following four quantities: only after a little strain after the above stage. However, in case of spirally reinforced
the inner core concrete starts taking more loads due to lateral confinement. The longitudinal
a) 16 times the diameter of the longitudinal bars.
48 times the diameter ofthe tie bars. steel and the concrete within the core are prevented form outward failing by the spiral. The
concrete cover in the outer shell does fall down. At this stage, the confining action of the spiral
240 CONCRETE STRUCTURES PART -t BASICS OF SHORT COLUMN DESIGN 241
AUTHOR; ZAHID 3
has a significant effect, and the failure
is caused by the yielding or
A
SIDDiqi CHAPnR^ stress, is given by
collapse load much larger than
that at which the shell spalled off. fracture of spiral steel at concrete strength, as increased by the lateral confinement
at the stage of failure of column will Furthermore, confined core
toughness of the be much greater than a tied column; the axial
column has been significantly increased.
stra*3
meaning that the the
followingempirical equation;
(I)
/>4.0/2' + 0.85/;

Turn of Spiral.
Fig. 7,12. Free Body Diagram of Half
evaluated. The
lateral confinement pressure must be equilibrium of
the confined strength, the
To estimate of a spiral is shown in Fig. 7.12.
For the force
Axial Shortening free body diagram of half turn stresses,
direction of the shown
Fig. 7.11. Load-Deflection Behavior of Various Types of Columns. this free body diagram along the
Core-concrete under compressive load 24^ = (ID
against the spiral. The spiral will thus will have a tendency to expand laterally and
be under tension and it exerts will bear ”

core. This increases the load carrying capacity of confining action on the Dchs
the core in spite of the loss of shell of the spiral
concrete. ratio is defined as the ratio of volume
Let,
confined core concrete strength The volumetric spiral reinforcement out-to-out of the spirals as under:
fl reinforcement to volume of the core measured
lateral confinement stress produced by the spiral ftp

ft
the maximum value of
fa when the spiral is yielding
concrete cylinder strength
specified yield strength of transverse
Ap
Ac s
Asp & D^, 44y_ (HI)

^sp reinforcement (^D^/4)s D^s


diameter of spiral
4sp From Eqns. I and II:
area of one spiral bar, equal to (n/4) (VI)
djp =
A volumetric spiral reinforcement ratio 2
A* That is, ~fi
diameter of core, out-to-out of the spirals When ~fyt, the maximum confinement is obtained.
length of one turn of spiral
4ch
= for smaller pitch = ' (V)
area of cross section of core, equal to (n/4) /2'
lateral confinement stress in core concrete produced by spiral is equal to fa* - 0.85/', and
the ultimate
and 5 spiral The extra strength provided by the
pitch of the spiral. strength is given by the following expression:
Ultimate strength for spirally reinforced
column = -
{fc 0.85//) 4ch
4.0/2’
= 2 (VI)
242 CONCRETE STRUC TURES PART 1 - BASICS OF SHORT COLUMN DESIGN 243
AUTHOR: ZAHID A.
SIDDIQI CHAPTER?:
Strength contribution of the outer shell
before spalling = 0.85/c' Gfg-Ah) (VII) 0.45Drt/c'(^/4*-D
The amount of spiral reinforcement may be so adjusted that the increased strength
spiral becomes exactly equal to the loss of strength due to spalling of shell concrete.
due to the 2. To bind the core effectively, the maximum clear pitch should be 75 mm
(3 in).
This gives be larger of
a second maximum load in Fig. 7.11 exactly equal to the first maximum load.
This type of 3. To avoid problems in placing of concrete, the minimum clear pitch should
spiral is called a balanced spiral. The following may be written to derive 1.5 times the size of coarse aggregate and 25 mm (1 in).
formula for the
balanced spiral:
^Ap/t-^ch “ 0-85/' (4g — 4^) 7.14 TRIAL COLUMN SIZE

or Ap = 0.425

(A y f'
-1 U?-
J fyt
(VIII)
In case the factored column load (P^) and moments (Ma and
known, the trial column size may be estimated for an assumed steel
in kN-m or k-ft units) are
ratio of 1.5% as follows:

The ACI 10.9.3 recommends a minimum spiral steel ratio


value, as unden
a little more than the above J.8 (trial) >
P + 2Af + 'I
0.43// + 0.0084 —for tied column (SI)

JP
(Ap)min — 0.45
Az
14*
[
— } f
1 ~~
JJ*
(IX) and Ae (trial) >
k)5f 01/
+0.6M +0.6Min,
~ sP’ra^ (SI)

Hence, an ACI spiral has a minimum spiral reinforcement to provide axial load
capacity
slightly larger than that of the shell concrete. A spiral that provides indirect axial strength •
Ao (trial)
0.43/; + 0.0084 — for tied column (FPS)
more
than the loss of shell concrete strength (spiral steel is more than given by Eq. IX) is
heavy spiral and a spiral giving indirect axial strength less than the shell concrete is called
spiral. The behavior of light, balanced and heavy spiral is shown in Fig. 7.1 1. The advantage
called and Xg (trial) — —
P+0.6M+0.6Mm~
QSf’ o 01/
for spiral column (FPS)

precisely known
of spiral reinforcement is only significant in axially loaded columns like piles, bridge piers and The constant with moment has units of 1/m or 1/ft. If the column load is not
concentrically loaded columns but is much lesser when bending moment is also present. at a preliminary stage, the column size may be approximated by thumb rules.
Spirals should consist of a continuous or properly spliced bar of at least 10mm diameter. If
exact calculations are not performed, the lap should be at least 40 times the diameter with the 7.15 AMOUNT OF STEEL IN COLUMNS
hooks anchored into the core concrete. The minimum ratio of spiral steel is specified so that However, as
the structural performance of the column is improved with respect to both the ultimate load and The total amount of steel in columns is calculated from the bar bending schedule.
a rough estimate, the amount of steel in columns may be estimated as 175
kgs/m3 (± 25%).
the type of failure. To keep the spirals firmly in place and true to line, one-and-a-half extra
turns of spiral are provided at each end of spiral.
Example 7.3 (SU: An axially loaded short column has a length of 3.0 m and has a
factored load of 1500 kN. //=20 MPa, maximum aggregate size = 19mm and / =
The following rules may be used to select the pitch of spiral reinforcement: 300 MPa.
1. The maximum center-to-center pitch should be as under: Calculate the suitable dimensions and reinforcement for the following two options:
A) Square tied column, and
_ B) Circular spirally reinforced column.
Ap
Solution:
fc' = 20 MPa
=
fy = 300 MPa
Pu = 1500 kN
= L = 3.0 m
D^S
244 CONCRETE STRUCTURES PART - 1 BASICS OF SHORT COLUMN DESIGN 245
AUTHOR: ZAHIDA CHAPTER?:
Part-A

4 +2M,a + 2May
SIDoiqi . —
Length / least dimension - 3000 / 3 /□ =

j- innn / 175 80 (really a short column)

— no = 0.85 x 0.75 [0.85 f'c (Ag-AJ +/ AJ


A (trial)
0.43 /J + 0.008
4 4/0.6375- 0.85 4'4
1500x1000 4-0-854
0.43x20 + 0.008x300 -
1500,000/ 0.6375 0.85 x 20 x 4
x 375’ - = 1680 mm
7
136,364 mm2 375 x 375 mm
Length /least dimension = 3000/375 * 300-0.85x20
= 8.0 (really a short column)
A = ^nO = 0.80 x 0.65 [0.85/;(4-4) +fy A
J Ast,min = 0.01 x it/4 x 3752 = 1105 mm2
0.80x0.65 = 0.85/; Ag + (fy 0.85/; ) 4«
- Fig. 7.14.
Use 6 - #19 (1704 mm2) as shown in
4
4 -0-85/J Diameter of the spiral = 10 mm

= 1500,000/0-52-0.85 x20x3752 Clear cover = 40 mm

tin
300-0.85x20
= 0.01 x 3752 = 1406 mm2
1746 mm2

= 375 2 x 40 = 295 mm
Ag = it/4 x 3752 = 110,447 mm2
Use 4 - #25 (2040 mm2) or 8 - #19 (2272 mm2)
or 4 - #19 + 4 - #16 (1932 mm2) Agh = tc/4x2952 = 68,349 mm2
Note: Compared with four bars, eight bars Example 7.3.
may require more ties and hence the Fig. 7.14. Spiral Column for
in steel by using eight bars may be saving *<4
somewhat compensated.
Smax
Diameter of ties for #19 (US) bars = #10 (US) 0.45 D,* 4(4
C
Spacing of ties = least of 1) 16x

16 256 mm
2) 48 x 10 = 480 mm
0 45x295x20
^•x102 x 300
(110,447 / 68,3 49-1)
= 57 mm
3) 375 mm ton
a dear spacing of 45 mm that is larger
4) 300 mm Provide spire. « a pitch of 55 mm. giving
= 256 mm 375 25mm and 1.5x19 = 29mm and Less
than 75mm.
Clear spacing between the bars has a
column has a length of 12 ft and
Example 7.4 (FPS): An axially loaded short ~ 60,000 psi.
= 375-2x40-2x10-2x19-16 kips. fc'~ 4000 psi, maximum aggregate size = % in and 4
factored load of 350 column option.
— 110mm < 150 mm
2 3751
Calculate the suitabledimensions and reinforcement for the tied
Additional ring of ties is not required. Solution:
The resulting placing of reinforcement
is fc' = 4ksi
shown in Fig. 7.13. = 60 ksi
Part-B Fig. 7.13. Column Cross-Section
.
fy
Pu
L

=
350 kips
12 ft
4 (trial) P.+2M„+2MV
^54+0.014
for Example- 7.3.
- Ag (trial) =
Pk+0.6M^+0.6M^
0.434 +O.OO84 —
1500x1000 2^2 = 159.1 in2 «13.5 x 13.5 in
0-5x20 + 0.01x300
~ II5>385mm2 * 375 mm diameter 0.43x4 + 0.008x60
246 concrete structures part - 1
Length/faMdimension = 12x12/13.5 AUTHOR: ZAHID f CHAPTER 7: BASICS OF SHORT COLUMN DESIGN
SlDDiQj 247
- ^no - 0.80 x 0.65 [0 = 10.7 (really a short column) -4g = gross area of the column

u
85/,(^ h column dimension in the direction of eccentricity or perpendicular to the
axis of bending
b = width of column perpendicular to the eccentricity
; and yh = center-to-center distance between the extreme face bars parallel to h
= 2^5£^85x4x13.5= =
dimension.
60^015^4 in2
= 0.01 x 13.52 = 1.823 in2
‘Stjnin

-
Use 4 #7 (2.40 in2)
or 4 - #5 + 4 #4 (2.04
- in2)
Note: Compared with
in steel by using four bars, eight bars may require
eight bars may be more ties ®d hence
somewhat compensated.
Diameter of ties for #7 (US) bars
Spacing of ties = least of = #3 (US)
1) 16 x 0.875
= 14 in
2) 48 x 0.375
= 18 in
3) 13.5 in
4) 12 in
= 12 in Fig- 7.15. Column
Cross-Section
7.17 PROCEDURE FOR DESIGN OF
ECCENTRICALLY LOADED COLUMNS
toFIt 7*"*
placing of Enforcement for Example 7.4.
is shown
1. Select trial cross-sectional dimensions b and h, considering the load and the
?16 approximate effect of the bending moment, as discussed for the axially loaded column.
‘^WCALLY loaded COLUMNS 2. Calculate the ratio y using the clear cover and the assumed diameter of the ties and the
When load on a main reinforcement. The value may be rounded to any lower value of y for which the
moment (A^ = Pu column
x e) we.
is applied at an
eccentricity or when axial load (PJ curve is available.
axial load and simultaneously
bending moment interact applied on a column, the stress and bending
one side and less with each other. This produces distributions ue to 3. Curve is selected for the required values of y,^',^ and shape of cross-section.

may be represented
compression or a
interaction are explained small tension on the other
in a later chapter on
by a curve between
more compression on
side. The etai s o t ts
columns. In short, the capacity of sue a co umn
4. The parameters
p
—A ——
and
Af
Agh
are calculated for each combination of Pu and Mu such as

(Pul and Afoi) and (PU2 and Afa), etc.


and , as shown in Fig.
7.16.
corresponding to
safe and any pointthe applied combination of Pu and
A series of
outside of the curve is
unsafe.
Ma that is lying on or inside the curve
Any point
is
5. The point corresponding to each
P
Ag
and ——
M
is located on the diagram and the

such interaction required steel ratio is approximately interpolated between the two adjacent interaction
superimposed on one curves corresponding to various curves.
in Appendix. diagram for particular steel ratios are
These curves are provided values ofy? and/y, some of such curves areusually 6. The maximum steel ratio (/%) is found for all the pairs of values.
for various values of the give
factor /defined below. 7. The total steel area^st is calculated as follows:
At = Pe bh
9‘
^^hinm AUTHOR.- ZaHU) BASICS OF SHORT COLUMN DESIGN 249
Ue SIDDIQI CHAPTER?:
ties ^cement is excessive
r = 450-2x40-2x10-25 = 0.72 « 0.75
C^bina^
lnati°ns of loads™ which « short and
tied, is to
450

Check For 1st Combination: A. 1080x1000


= 6.40 MPa
3
c°mbinatiOn. = 300 k£’ A =225 = 58 M = 85 kN 375x450
205,6x10®
20 MPa, maxi^ be of rectangular 58 kN"m’ = 40 kN-m 375x4502
= 2.71 MPa
r Agh
reinforce^ = 19^ not larger than
375 mm Using the interaction curve, pg 0.03.
720x1000
Check For 2nd Combination: = 375x450
= 4.27 MPa
20MPa Ag
300 Mpa Mu = 205,6x10®
The factored
Joad,,s = 2.71 MPa
A*h 375x4502
^^mbination. _
calculated below:
1-2 x 300+1.6 x
Using the interaction curve, pg < 0.03 (OK).
7»d
2

3ri
.
combmation- 1-2 x 58 + 1.6 x 85 = 1080 kN
2 1-2x300
450

_ 1-2x300+1.6x600
= 205.6 kN-m
+ 1.6x225 ~
720 kN
Check For 3rd Combination: —Ag
2-
= 1320x1000
375x450
= 7.82 MPa
combination. = 1-2 x 58 + j 6 x 85 133.6x10®
= = 1.76 MPa
The trial size
Ofthe column is
~
1-2 x 58 + 1,6 x
205.6 kN-m
= 1320 kN
40 = 133.6 kN-m
- Arh 375x4502
Using the interaction curve, pg < 0.03 (OK).
decided as under:
As (trial) =pg*bh = 0.03 x 375 x 450 = 5063 mm2 (10- #25)
Diameter of ties for #25 (US) bars = #10 (US)
0-43//+Oj0087~ Spacing of ties = least of 1) 16 x 25 = 400 mm
-®t2xl316)x]000 2) 48 x 10 = 480 mm
°’43x 3) 375 mm
F°r 10.mm i44,291 mm2
20+0^008^300- 4) 300 mm
diam 375 x 375 nun = 300 mm
steel,
375-2x40-2x10-4x25
375 ~ 0.67 * Q,6o si = 58.3 mm < 150 mm
Check For 1st n 3
^binafj^. 1080x1000 450-2x40-2x10-3x25
= 7.68 MPa S2 = 137.5 mm < 150 mm
'V 375*~~ 2
205.6x10® Additional ring of ties is not required in the shorter direction.
Using Agh = ~~375^~ ~ 3.90 MPa
r The resulting placing of reinforcement in the designed column is shown in Fig. 7.17.
J”
* SoX1*0 ”d/’ ° 3W
tetter to revrse the
“»
column size to
Example 7.6 (FPS): An 11 ft long column, which is short and tied, is to be used for the
following combinations of loads:
4314 251
BASICS OF short COLUMN DESI
250 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A cgAPTER7:
SIDDlQl
cbeck For 1st Combination: —
P.
M„
244
12x15
158x12
= 1.36 ksi
= 0.70 ksi
12xT?
Arh
curve for y— -
0.60, fz 3 ksi and/.
Using the relevant interaction 3%. It is better to revise the column
than

60 ksi, the required
size to 15 x 18 in.
steel ratio becomes greater
18-2x1.5-2x0375^2 = 0.74 0.75
18
Fig. 7.17. Column Cross-Section for Example 7.5. P. 244 = 0.90 ksi
Check For 1st Combination: Ag 15x18

——
1st combination: Pd ~ 70 kips, Pl = 100 kN, Md = 45 k-ft, A/l 65 k-ft M„ 158x12 = 0.39 ksi
2nd combination: Pd ~ 70 kips, Pl = 50 kN, Md = 45 k-ft, AtL 65 k-ft Ah 15x18*
3rd combination: Pd = 70 kips, Pl = 130 kN, Md = 45 k-ft, Md = 30 k-ft curve, ft s 0.023.
Using the interaction
The column is to be of rectangular section with at least one side not larger than 15 in. f.'~ 3
ksi, maximum aggregate size = 0.75 in and fy = 60 ksi. Calculate the suitable dimensions and
reinforcement
Solution:
Check For 2nd Combination: —
Pu
M„
164
15x18
158x12
15x18*
= 0.61 ksi
- 0.39 ksi
Afi
fc' = 3 ksi : fy = 60 ksi
curve, ft < 0-023 (OK).
The factored loads are calculated below: Usingthe interaction
1st combination:
2nd combination:
Pu
Afu
Pu
-
=
=
1.2 x 70 + 1.6 x 100 = 244 kips
1.2x45 + 1.6x65 = 158 k-ft
1.2 x 70 + 1.6 x 50 = 164 kips
Check For 3rd Combination: —p«

M,
292
15x18
102x12
= 1.08 ksi
= 0.25 ksi
A/ = 1.2x45+ 1.6x65 = 158 k-ft Agh is^is*'
3rd combination: Pu = 1.2x70 + 1.6 x 130 = 292 kips curve, ft < 0-023 ^O
)
Ma = 1.2 x 45 + 1.6 x 30 = 102 k-ft Using the interaction
~xbh = 0.023 x 15 x 18 = 6.21 mm2
(8 #8) -
^4^
The trial size of the column is decided as under:
P, +0.6AFm + 0.6A^
Fb

-
(US) bars #3(US) " 16 *n
Diameter of ties for #8= least of 1) 16 x 1
. Ag (trial) Spacing of ties 2) 48 x0.375 = 18
in
0.54+0.014 3) 15 mm
(292 + 0.6x102)
4) 12 mm
0,5x3 + 0.01x60
= 168.20 in2 « 12x15 in = 12 mm
For #3 diameter ties and #8 diameter longitudinal steel,
si
15-2x1.5-2x0.37523x1
2
= 4.13 in
15- 2x1.5 - 2 x0375-1
18- 2 x1.5-2 x OJTSji3^!
-

Y
15
= 0.68 « 0.60 = 5.63 in
32 2
m shorter
is in
nf ties is direction not required.
Additional nng of
1
252
CONCRETBSTRUcturesrart.t
zahr,., QIAPTER7: BASICS OF SHORT COLUMN DESIGN 253
PROBLEMS

Note; N Registered number of the student.


Ao A fixed positive or negative number assigned by
the instructor to get R between 600 and 1000.
R N~ Nq, a parameter to get numerical data for the
problems.

Column 1- A concentrically loaded RC short column has a cross-sectional size of 300 x 300 mm
and is reinforced by 8 # 19 (US) bars. fc'= 20 MPa, fy = 420 MPa. Show the elastic
uncracked transformed section. What will be the load carrying capacity when the
concrete stress reaches a level of/c = 0.30 fc'.

2- An axially loaded column has a length of 3.0 m and has to cany a factored load of 2R
(kN). fc'~ R I 35 (MPa) and fy = 300 (MPa). Calculate suitable dimensions and
reinforcement for the following two options:
a) Square tied column
b) Circular spirally reinforced column

3- Design a 3.0 m long short column for the following combinations of loads:

a)
b)
PB
Pu

1.2 R (kN) and Afu
=
0.8 R (kN) and Mu
=
=
R / 4 (kN-m)
R / 4 (kN-m)
a) PB= R (kN) and
1.2 = R 1 6 (kN-m)
a) Pa- 1.4 R (kN) and = 0

The column is to be of rectangular cross-section with fc'~ 20 MPa and fy = 420 (MPa).
254 CONCRETE STRUC TURES PART . I
J
^BgceFor Reader’s Notes
Chapter - 8

RETAINING WALLS

s.l INTRODUCTION

Retaining walls are the walls used to hold wafer, other fluids, earth and other loose / granular
material having a vertical abutting edge. For example, the basement retaining walls of the
buildings must retain the soil surrounding the structure. Similarly, the embankments must be
retained vertically for a highway or railway bridge. The failure of these walls may be by
overturning due to lateral loads, sliding at the foundation level due to lateral loads, failure of
foundations and lack of sufficient strength. The self-weight of the wall along with the weight
of soil directly resting on the foundation provides resistance against overturning. Similarly, the
total vertical load at the base multiplied with the coefficient of friction of the contact surface
provides the friction to prevent the sliding of the wall. These walls have the major types
discussed below:
Free Standing Retaining Walls: These are the retaining wails with no extra load or restraint
of super-structure at their top, unlike the basement walls of the buildings.
Gravity Retaining Walls: These retaining walls are made without steel reinforcement to carry
the tension. Hence, either no tensile stresses should be developed in the structural members or
these stresses should be much less. The stability and less-tension conditions are satisfied by
the self-weight of the wall. The thickness is greater even at the top and it increases with the
depth (Fig. 8.1a). These walls may be constructed by using brick masonry, stone masonry or
concrete. These are economical for low heights, generally up to 3 m.
Cantilever Retaining Walls: In this case, a relatively thin stem of reinforced concrete projects
out of the foundation, as shown in Fig. 8.1b. The tension developed, due to large moment at
depths of fill, is to be resisted by steel and the stability is provided by a horizontal base rigidly
connected to the stem and by the weight of soil over this base. These walls can be used
economically for heights from 3 to 6 m.
Counterfort Retaining Walls: When height of a cantilever retaining wall exceeds about 6 m,
counterforts are used to provide additional support or bracing at regular intervals (Fig. 8.2).
Counterforts are commonly provided on the fill side but occasionally may also be used on the
opposite-to-fill side (called buttresses). This later situation is used for temporary arrangement
in some structures and, after few years, the other side is also filled with soil.

8.2 IMPORTANT PARAMETERS FOR RETAINED SOILS


The following parameters of the retained soil are important for the design of retaining walls:
Angle Of Repose Of Soil: The angle with horizontal of the slope that becomes stable with
time is called angle of repose of that soil Table 8.1). For sandy soils, the tangent of the angle
of repose is approximately equal to the coefficient of internal friction, The clays have angle
RETAINING WALLS 257
256 CONCRETE STRUCTURES PART - 1 AUTHOR-• 7 AUm .
''^A-SIDD/q, CHAPTERS: not generally compensated by the
savings in
Of repose value depending on lts cohesi°n and decreases by increasing providing tapered formwork is
content. For saturated clays, the angle of repose value may approach to zero. &
.
m°isture ^tra costforofsuch heights.
Coefficient Of Friction: This factor gives the ratio offeree develop
concrete often recommended for walls higher than 3.5 m. The taper on to
and
the
the interface Of th. taper or batter of 1 in 50 is tends to counteract the deflection up to some extent
wall base with the underlying soil to prevent sliding with resnect £ I* outside face is preferable,
as it is
contact surface (Table 8.1). In case sliding between the two smfaJf «
tOtaI on
wall due to the lateral pressure. The steel ratio (p) lesser than A
friction becomes equal to tangentofthe angle of internal friction of fee s of induce tilting of the will be
not usually be critical for shear and lateral deflections
stem will
considered better as the
under control.

(a) Gravity Retaining Wall (b) Cantilever Retaining


Wall
Fig. 8.1. Gravity and Cantilever Retaining Walls,

Components of attaining Woll.


Fig. 8.3. Typical Structural
retaining wall, which is further
divided into two
slab of the a minimum
2. Base: This is the foundation The thickness ranges from h / 14 to h 1 10 with height of
toe.
parts, namely, the keel and the total width of the base depends upon the type of fill, of the
value of 250 to 300 mm. The and the bearing capacity
of friction at the foundation interface % of the overall
soil retained, coefficient
estimate for thewidth of the base is 40 to 75
underneath soil. The preliminary standing wall and when coefficient of friction is low.
height; more size is required for earthquake is also considered, the base width may approach
free
to soil, it may be
Fig. 8.2. Counterfort Retaining Walls. When lateral earth pressure due is present over the retained
In case surcharge load
the height of retained soil. soil.
8. 3 COMPONENTS OF RC CANTILEVER RETAINING WALL included as an additional depth of be left while
bond of the stem concrete with the base, a groove must
helpful for its easy
In order to get good purpose, a beveled timber may be
The main components of a cantilever retaining wall are shown in Fig. 8.3 and are described casting concrete for the base. For this
in Fig. 8.4.
the stem into the base, as shown
below: removal and to get a proper key of
1 Stem or Arm: This is the actual cantilever wall vertically projecting out of a
horizontal Beveled timber
base at the bottom, where it is ngidly connected. Due to the horizontal soil pressure the stem
is forced to slide opposite to fill along with the base, overturn suddenly in case of oveiloads
tilt due to differential settlement of sod on
The excessivelvXX
one side of the base, or bend
rige
Base
lesser flexural strength
14 to A / 8j where A is
or
theJfe
height of fill.
thickness
The
of arm at the base is taken
minimum
in the
Im
thickness at the top may be 200
of h /
but
a value of 300 mm is preferable because a gnd of steel is to be provided at both the faces of the Construction Joint of Stem with Base.
Fig. 8.4.
because
wall. The stem is usually made of constant thickness for total height under 3.5 m, fee
260 CONCRETE STRUCTURES PART -1 RETAINING walls 261
AUTHOR: ZAHID A. CHAPTER8: and
component that acts as lateral SIDDIQ] angle of internal friction, which
can exactly be determined for sands
earth pressure on the retaining wall. Where (|) = for clayey soils.
a heap, it maintains a
stable slope having an angle of repose When pure sand is placed
in cohesion is approximately added in this value
of this angle is roughly with the horizontal.
may be determined
equal to the coefficient of inter-granular The tangent and ft = the angle of backfill.
experimentally
its internal cohesion. However,
becomes nearly zero in
maintain a larger angle with
friction for that sand,
content than its saturation point, mayin the laboratory. The clay, with much less which
moisture
as the soil water increases, this respect to horizontal due to
clays. In this case, the soil acts cohesion greatly reduces and
For horizontal

Aa
_
soil surfaces with p- 0, the above
1-sin^
~ : -
equation reduces to:

lot of lateral pressure onsaturated like a thick liquid and 1+ smp


then is
the wall. exerts is to be excavated for placing the base of the wall and
Usually soil adjacent to the wall the type of
Regardless of the type of soil
earth pressure in this case depends upon
slide down or flow under the retained vertically, some portion of soil has a natural tendency to be backfilled. The coefficient of
depositing and compacting it and the
drainage
action of gravity, leaving behind the soil of
stable slope. The weight of this
unstable portion
mass that has its naturalto backfill, method
material used for the may become saturated during rains or due to high water table and
acts as lateral earth pressure of soil (called wedge) has a
on the retaining wall. The magnitude component that facilities provided. Soils design of retaining walls for such
walls is prohibitively
varies with depth, exactly as for the of the lateral soil pressure exert excessive pressures. The are often provided
regular spacing or continuous back-drains
may
liquid retaining wall, as follows: Weep holes placed at accumulating behind
expensive.
P - Ky3h to avoid the heavy pressure
clean
condition
sands and
that can be caused by ground
gravels
water
because of less
are considered better for the backfill the passage of
Where ys is the unit" weight of the the wall. Further,
draining nature, no stability loss with
soil and K is a constant called
pressure. coefficient of static earth pressures in case of wet conditions, free
The coefficient of static time and less severe frost action.
earth pressure
developed by soil with respect to its (K) is defined as the ratio of the lateral pressure The approximate soil properties
are provided in Tables 8.1 to 8.3.
under the action of earth pressure unit weight. Retaining walls usually
deflect elastically Properties.
away
base takes place due to unequal pressurefrom the soil. Also some tilting of the wall along its Table 8.1. Approximate Soil
on the underlying soil. A sliding plane Coefficient
as shown in Fig. 8.6, and the be is formed, Unit Weight
the wall has a tendency to wedge abc exerts soil pressure on the wall. In these Sr. kg/mJ degrees of Friction,/
move away form the soil, the corresponding pressure cases where No. Type of Soil
33-40 0.5 -0.6
earth pressure and the is called active 1750 - 1920
Sand or gravel without fine
angle 8 with the
corresponding X-value is denoted by K^. If the soil surface
horizontal and denotes the angle of
coefficient for active earth pressure is given
by:
subtends an
internal friction of the soil, the
ri
particles.
Sand or gravel mixed with —silt.^——
- 1920 - ~
Silty sand, sand and gravel with 1750-
1920
2080~ 25^35
23^30
'
0,4- 0.5
0.3 -0.4
~

high clay content. 25^35 0.2 - 04


Medium or stiff clay. -
1600 1920
Zero for fully
saturated
5. Soft clay. -
1440 1750 20-25
Zero for fully
U.Z

no n q

saturated

of Friction Angles For


Table 8.2. Approximate Values
(b) Clayey Soil Various Densities of Sand.
Fig. 8.6. Failure Surface through
Soil Retaining By Wall.

„ cos^- Jcos2 /7-cos2


Aa = cos^— ~ L
cos/? +^/cos2 /5- cos2
262 CONCRETE STRUCTURES PART -I RETAINING WALLS 263
AUTHOR: ZAHID A
2. Passive Earth Pressure: In case, SfDDlqi CHAPTERS:
due to external loads, the wall is pushed against
bigger soil wedge will be affected,
which will have the tendency to slide the soil a 8.5 REQUIRED RETAINING WALL STRENGTHS
resisting pressure developed by soil in upwards.
The lateral
the soil placed above the wall footing,
this way is called passive earth pressure.
For example, wall, three resultant forces, namely, the resultant Fig.
opposite to the retained fill, will For design of any type of retaining soil bearing resistance (S), shown in
phenomenon. Soil can offer greater experience this pressure (0), the total developed weight (P) and the of safety. The value f) is the angle of
pressure for the retaining wall knownresistive pressure, which will be a helpful
as passive earth pressure. The lateral earth 8.7, have to be brought in equilibrium, with some margin
angle. In
and the value J is the wall friction
_—
earth pressure in case of horizontal soil top
1 + sin^
surface is given by:
coefficient for passive the backfill surface with respect to the horizontal
addition, all internal stresses in the structure and
all contact pressures of footing should be
. Ap
1-sm^ - within the permissible limits.
The retaining wall must be safe against all possible
modes of failure that are briefly explained
Table 8.3. Maximum Allowable Soil Pressure. below:
wall, including stem, toe and heel, has to be
Sr. Type of Soil 1. Structural Stability: Every part of the retaining at the junction with the other members
No. Maximum Allowable designed independently as a cantilever fully restrained
1.
un-reinforced gravity retaining wall, the tensile
and carrying all the applied loads. In case of wall, cracks should not open enough to expose
Rock.
20% of ultimate crushing stresses should be within limits. For reinforced strengths of all the structural elements must
2.
_ strength. the steel to rusting. Both the flexural and the shear requirements fbr development of bars and
Compact sand, hard clay or sandy clay. Further, the
3. Medium rw
clay.
385 sir be checked for the worst conditions.
4. Compact inorganic sand miypd with c.;n 290 j dowels must also be satisfied.
5. Loose sand. 190
6. Soft sandy 145
— clay sand
~ *•’ * or clav

— Jrnixed with slit
______
;
7 r 95-
Loose inorganic
8. Loose organic sand mixed with silt, muck. . so .

0
Modulus of sub-grade reaction (g) ~ 260 qa 0.25 q^
- kPa/m or kN/m3
3. Surcharge Load: The traffic load
or load
retained by the wall, is called surcharge load. of goods in a storage yard carried by the fill,
traffic loading is In case of highway and railway bridges, the
usually represented by an additional soil
height is used to calculate soil pressures height of 1 m (3 ft). This extra soil
but is not added in height to measure the moment
4. Self Weight Of Wall: The self-weight arm.
is useful for the wall in providing the
stability, and better resistance against sliding overturning
(produced due to bending moments)
and reduction of tensile stresses in the wall
(a) Gravity Retaining Wall 0» Cantilever Retaining Wall
is obtained. In case of gravity retaining walls, this §
becomes more important. weight
Fig. 8.7. Forces Acting on a Retaining Wall.
5. Weight Of Fill On Heel-. This weight
is also desirable because it provides the balancing
moment to prevent overturning of the wall. may overturn as a single unit around the maximum
Further, it improves the sliding resistance. 2. Overturning Stability: The wall
limit state, the restoring / resisting moment g
6. Upward Soil Reaction On Footing:
This compressed comer (edge of the toe). For service / driving moment (A^), giving a factor of
footing as a reaction to the applied vertical loadsoil reactive pressure is developed below the (M) should not be less than 2.0 times the
overturning
and the resulting overturning moment at the according to some codes). These calculations
base. The maximum pressure that can be safety against overturning of at least 2.0 (1.78 are r
developed depends upon the bearing capacity of soil. According to UBC-1997-1611.6, the retaining walls
A more uniform contact pressure improves
the settlement and overturning, while tensile are performed using the service loads. 1.5 against sliding and at least 1.5 against jj
stresses on one side reduces the resistance against to be designed for a factor of safety of at least
from the soil on the tension side. overturning due to separation of footing overturning, using allowable stress design loads. |
using a factor of 0.9 with all forces
producing
At ultimate stage, the ACI code requires overturning moments and then 0.9M must
restoring moments and 1.6 with all forces producing
RETAINING WALLS 265
264 CONCRETE STRUCTURES PART - 1 CHAPTERS: on the total vertical
AUTHOR: ZAHID A. SIDDIQI
Bearing: The pressure under the footing dependsabout the centroid of
be greater than equal to 1.6M1. In other words, the factor of safety, as previously defined 4. Safety Against Soil moment of all loads except soil
bearing forces
of inertia
for load on footing (W) and the
footing in contact with soil and If is the moment
service loads, becomes 1.78. AASHTO specifications require that the resultant of all
factored If A is the area of the base centroid
forces acting on the retaining wall, except the bearing pressure, must be within middle half the footing (A0.
area, then the contact pressure (o) at a distance x from
the footing. The overturning moment is due to horizontal component of the active earth force
of of the same contact usual formula, as under:
can be determined by using the

— — ——
and the restoring moment is due to the wall self weight, weight of fill, vertical component . positive)
of
active earth force and passive horizontal earth pressure. If the resultant of all forces except the W + Mfx i (Compressive stress is considered
bearing pressure passes through the middle third of base, adequate safety against overturning cr A footing by B and
exists. retaining wall, denoting the width of the
Considering unit length of a long
becomes:
3. Sliding Resistance: Lateral resistance against the horizontal component of the active earth
contact of the base with the soil, the above equation
pressure has to be developed by the static friction at the footing base and by the passive earth assuming fill!
pressure against the embedded front portion of the wall, with a factor of safety of at least 1.5. B
fy 6M j
The value of this factor calculated in LRFD method but applied to the service forces becomes W 1 V
Oinax lx#3 B B
1.78. The total static friction can be calculated by multiplying the resultant vertical load on the B
footing with the coefficient of friction (£ as in Table 8.1) between the concrete and soil. These 12
calculations are also performed using the service loads. Where this resistance is insufficient,
extending a key or lug into the soil below the footing can increase the passive pressure and and Oinin
hence the resistance against sliding.
In case less sliding resistance of gravity retaining wall is available, requiring an excessive base
width, the base may be sloped to provide extra passive earth pressure adding into the sliding
resistance (Fig. 8.8). Otherwise, if the sliding resistance is less, a key can be provided. The
sliding resistance is increased in the presence of a key by the passive pressure acting along the
key and because of the fact that the sliding plane below the toe has a soil - soil friction factor
of tan<f> (significantly greater than concrete - soil friction factor). The second type of increase
is usually neglected to get a conservative design. The resistance provided by the key is shown
in Fig. 8.9 and is mathematically equal to the following:

A = yk where or = Kp yy, <r2 = Ap/O-yk)


= Ap/ykO-jk/Z)

Provided by Key.
Fig. 8.9. Sliding Resistance
soil and the
will be contradiction to the assumption that the all loads,
Negative or tensile value of
crmjn
intersection of the resultant of
to keep the
a) Concrete Retaining Wall. b) Brick Masonry Retaining Wall. footing are in foil contact. It
is desirable
the kern or middle-third of the base to avoid tensile
within
except the soil bearing pressure, may be achieved by adjusting the dimensions of
the heel
Fig. 8.8. Gravity Retaining Walls with Better Sliding Resistance. sides would result in
contact pressure (Fig. 8.10a). Thisbetween contact stresses on toe and heel
and the toe. Greater difference
264 CONCRETE STRUCTURES PART -1 RETAINING WALLS 265
AUTHOR: ZAHID A. CHAPTER 8:
SIDDIQI vertical
the footing depends on the total
4. Safety Against Soil Bearing: The pressure under soil bearing forces about the centroid of
be greater than equal to 1.6Mj. In other words, the factor of
safety, as previously defined
service loads, becomes 1.78. AASHTO specifications require rOr except
that the resultant of all load on footing (W) and the moment of all loads with soil and If is the moment of inertia
forces acting on the retaining wall, except the bearing pressure, factored
must be within middle half the footing (M)- If A is the area of footing in contact
the footing. The overturning moment is due to horizontal (o) at a distance x from the base centroid
and the restoring moment is due to the wall self weight,
component of the active earth of
force of the same contact area, then the contact pressure
weight of fill, vertical component
active earth force and passive horizontal earth pressure. If the
can be determined by using the usual formula, as under:

— ——
of
resultant of all forces except
bearing pressure passes through the middle third of base, adequate the W Mfx . .
(Compressive stress is considered positive)
exists. safety against
overturning cr = ± I
J Ij
of the footing by B and
Considering unit length of a long retaining wall, denoting the width
3. Sliding Resistance: Lateral resistance against the
horizontal component of the active
pressure has to be developed by the static
friction al the footing base and by the passive earth assuming full contact of the base with the soil, the above equation
becomes:
pressure against the embedded front portion of the wall,
The value of this factor calculated in LRFD method
1.78. The total static friction can be calculated by multiplying
earth
with a factor of safety of at least 1.5.
but applied to the service forces
becomes
the resultant vertical load on the
W
+
Mf 2
_w +
footing with the coefficient of friction (f as in Table 8.1) B lx B3 B B2
between the concrete and soil. These
calculations are also performed using the service loads. Where this 12
resistance is insufficient
extending a key or lug into the soil below the footing can
hence the resistance against sliding.
In case less sliding resistance of gravity retaining wall is
increase the passive pressure and

available, requiring an excessive base


and Omin ~
B B1 —
width, the base may be sloped to provide extra passive earth pressure
resistance (Fig. 8.8). Otherwise, if the sliding resistance is less, a adding into the sliding
key can be provided. The
sliding resistance is increased in the presence of a key by
the passive pressure acting along the
key and because of the fact that the sliding plane below the
of tan<() (significantly greater than concrete -
toe has a soil soil friction factor
- soil friction factor). The second type of increase
is usually neglected to get a conservative design. The
in Fig. 8.9 and is mathematically equal to the following:
resistance provided by the key is shown

where cq - Kvyy, = Kvy(y-y$

Fig. 8.9. Sliding Resistance Provided by Key.

assumption that the soil and the


Negative or tensile value of cw will be contradiction to the of the resultant of all loads,
footing are in full contact. It is desirable to keep the intersection
of the base to avoid tensile
Fig. 8.8. Gravity Retaining Walls with Better Sliding
Resistance. except the soil bearing pressure, within the kern or middle-third
adjusting the dimensions of the heel
contact pressure (Fig. 8.10a). This may be achieved by toe and heel sides would result in
and the toe. Greater difference between contact stresses on
266 CONCRETE STRUCTURES PART - 1 RETAINING WALLS 267
AUTHOR.- ZAHIDa. CHAPTERS:
required on
much larger settlement of the toe relative
to
SlbujQj
developed on the fill side and hence steel is
required here. For the toe, the steel is
Due to practical reasons, if the resultant the heel with a corresponding tilting of the heel, steel is required on the upper side.
does not pass through the middle wall the lower side, whereas for the
consider a pressure distribution
extending only over a part of the footing third, we have to
width of this contact is then taken equal to (Fig. 8.10b).
location of resultant at the bottom of base 3 times the horizontal distance from toe toThe 8.7 PLACEMENT OF JOINTS
Improvement of underneath soil tb» more than
required in certain cases where the
underneath soil is- riot firm. may -.h 3 According to the AASHTO specifications,
expansion joints must not have spacing Further,
m (30 ft).
If Rv is the vertical component of the 27 m (90 ft) and construction joints
must not have spacing more than 13wall (1/8 in2 per foot)
resultant and <4 is the distance of the less than 265 mm2 per meter depth of
toe at the bottom, the contact pressures may
also be found as under: resultant from the horizontal temperature steel of not exposed surface.
should be provided adjacent to the
Diagram (a): o’1 = (45-6<4)A.
o', = (6<4 -2B) A
B2 B
Diagram (b): o-, =
3^
8.6 LOAD FACTORS FOR WALL
DESIGN
Lateral earth pressures will be considered
like live load with a load factor of
factored loads are required. Dead loads 1.6 wherever
design forces, such as for the toe slab, should be multiplied with 0.9 where they
and with 1.2 where they increase design reduce
for the heel slab. Surcharge load is forces, such
also considered as live load. Heel slab
factored loads acting downwards, not including
as
is designed for (a) Deformed Shape of Retaining Wall
0» Main steel Requirement
acting in case of lateral pressure more than the the upward soil pressure because it may not be Requirements.
design service pressure. Fig. 8.11. Deformed Shape and Steel
using (1:6) c/s
Example 8.1: A gravity retaining wall is
to be constructed in masonry
8.12) and is to retain 2 m height of soil
besides heavy traffic
mortar and burnt clay bricks (Fig. / stiff clay. The
wall is additionally embedded by 0.5 m into a soil that is medium
loads. The compressive stress in
is expected to have a density of 1750 kg/m3. The allowable factor of
backfill the safe bearing capacity is 75 kPa. The
masonry is to be taken equal to 0.54 MPa and
safety against sliding should be a minimum of
1.5. Ignore passive resistance of soil.

_ ——— —_
Surcharge, load of 1 m height of soil, h' = 1m

OR Soil Pressure
For medium / stiff clay, = 30°
Aa ~
'
l-sin^
l + sin^
, IW
1-sin^
l-sin30
-

1 + sin 30
1 + sin30
l-sin30

_

3 000

% = 17.17 kN/m3
= 1750 x 9,81 /1000 stiff
(a) Resultant Load within the Kern (b) Resultant Load outside the Kern / = 0.30 (for medium / caly)
Fig. 8.10. Foundation Soil /1000 = 18.93 kN/m3
Unit weight of brickwork, ft = 1930 9.81
Pressure Distributions. x
The bent shape of main components
of a Allowable compressive stress in masonry, Fa = 0.54 MPa
steel required is shown in Fig. 8.1 lb. The retaining wall is shown in Fig. 8.11a and the main
stem tries to bend away from the backfill,
tension is
269
RETAINING WALLS
268 CONCRETE STRUCTURES PART - 1 AUTHOR-
ZAHIDA.SIDD1Q! CHAPTERS:
top width of wall Overturning moment, Af0 ~ Qy
= 2 bricks wide = 456 mm 32.77kN-m
Let
bottom width of footing =B = 32.163 x 1.071 = sub-weight (P) from the toe, Mr
the
of each (2?) from the
2nd length of wall = 1.0 m x represent the distance of the line of action of line of action of resultant way in Table
Let and a the distance performed in a systematic
Horizontal thrust at the top of wall = Xah x /s x h' corresponding resisting moment stability analysis may be
The calculations for the
= 0.333 x 17.17 x 1.0 = 5.718kN/m toe.
S.4.
Trial Bottom Width
Horizontal thrust at the bottom of footing = 0.333 x 17.17 x 3.5
= 20.012 kN/m

The trial bottom width may be decided based on sliding resistance. Assuming the combined
density of bricks and soil above footing equal to 18 kN/m3, the weight (W) acting on the
foundation can be calculated. The surcharge load is considered to be extended only up to start
of the base, producing full horizontal thrust but no weight to stabilize the wall.

, Retaining Wall
Retaining of Example 8.1.
Analysis for
Fig. 8.13. Stability

Table ».4. Calculi Sl»«


For Slabl^^^
Sub-weights (m) (kN-m)
Forces (kN) _ 4.920
21.580 . 0.228
Fig. 8.12. Data for Retaining Wall of Example 8.1. Pl 0.456 x 2,5 x 1,0 x 18.93
x 18.93 75.530 1.520
P2 0.5 x 3.192 x 2.5 x 1.0 7R508- 2.584
D xl.Qx 17.17
W = 18 X 2.5 X
.

X l.o = 0.045 B kN P3 0.5x3.192x2.5 165718


1000 Total: Mr
25
Q = 5.718 x 2.5 + (20.012 - 5.718) x -y = 32.163
Now, fW = FOS x Q
0.3 x 0.045 B = 1.5x32.163
B = 3574 mm (16 bricks wide, say 3648 mm) 296.751-32.77
_ ^59401
165.618
Stability Against Overturning 2.432 m
2B/3 =
BI3 = 1.216 m
against overturning.
Various applied forces acting on the wall are shown in Fig. 8.13.
lies within the middle
third and the wall is safe / M>) is 9.06, which
Hence the resultant resisting moment to overturning moment (M
Height of force Q from toe, y = Further the ratio of wall is stable.
further confirms that the
= 1.019 m
270 CONCRETE STRUCTURES PART - 1 AUTHOR; ZAHID A SlnjjjQi CHAPTERS: RETAINING WALLS 271
Check For Bearing On Soil Solution:
Moment at base = Mf = algebraic sum of X and mo
03 about centroid
Height of soil retained, H = 4.5 m
of base.
-
= Mo [clockwise] 2P x (B/2 - r™ *
XP to act on left side of centroid of thTbX'e assuming

_
Height of surcharge soil, Hs
l-sin^ _
~
=
1- sin 30
0.5 m
_
— w,33
= 32.77 -
165.618 x (1.824 -1 594)
_ 1+sin^
_ l + sin30
= 5.32 kN-m (counter-clockwise, tension on left side)
Section modulus of the contact area = Sf =
An ~ 1+sin^
1-sm^ — l + sin30
l-sin30
= ,J.UU
6 Lateral Pressure Due To Soil
Contact stress = —
W
A
±
Sf <Ta ~ 7 H Ka -2c* J~K~
(165.618+3.648x1.0x17.17) 5.32 0.333 x 18 x 4.5 -2 x 0 x V033 = 27 kPa
=
= 1x3.648 2.218 P3 = 0.5 x Ha^ = 0.5 x 4.5 x 27 = 60.75 kN/m
= 62.57 ±2.40 kPa Location of Pa from bottom = Hl 3 =4.5/3 = 1.5 m
= 64.97 kPa (C) and 60.17 kPa (C) < 75 kPa (OK) Surcharge additional
Check For Compression In Masonry depth = 0.6 m
Maximum compressive stress 64.97 kPa (as above)
< 0.54 MPa or 540 kPa
(OK) r= 18 kN/m3
Because of the large margin of strength available in masonry, it is possible 3m $5=30°
width of wall above the base, both towards the toe and the heel. to reduce the
c = 0MPa
Check For Sliding y= 16 kN/m3
1.5 m
Sliding resistance, F„ = f^P- °-30 x 165.618 = 49.69 kN |
Lateralload, Q = ®
- k 1k
FOS against sliding ~p3rl Q 1,54 1.2 m 1.925 m
(OK) 0.275 m
pyample 8.2: Design a cantilever reinforced concrete retaining wall used to Fig. 8.14. Retaining Wall for Example 8.2.
•TT^r^L with an additional burial depth of 1.5 m. The trial dimensions ofretain 3 m soil
Noting are shown in Fig. 8.14. The other related data as under:
is
the stem and
Overall height of fill = 3 + 0.6 (surcharge) + 1.5-0.5 = 4.6 m
y 18 kN/m3 (Active side) B « 75% of height = 3.4 m
= Toe width B /3 = 1.2 m
^.^OMPa Stem thickness « h / 14 = 4000 / 14 = 285 mm (rounded later)
y « 16 kN/m3 (Passive side) Lateral Pressure Due To Surcharge
allowable bearing capacity = 100 kPa
Friction coefficient,/ = 0.45 o3 = Ka y = 0.333 xl 8 x 0.6 m = 3.6 kPa
^1friction angle, 8 = 0 Ps = H<ts = 4.5 x 3.6 = 16.2 kN/m
Surcharge additional depth - 0.6 m
Location of P5 from bottom = Hl 2 = 4.5/2 = 2.25 m
Overburden depth to ignore = 0.3 m
r_20 concrete and Grade 420 steel are to be used Passive Pressure
Take passive earth pressure coefficient equal to one-half of its value, to be on conservative Kp = 0.5 x original value = 1.5
side.
RETAINING WALLS 273
CHAPTER 8:
272 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A.
SIDDIQI
<Tp - Kp 7 dp + 2c x ^Kp
= -
1.5 x 16 x (1.5 0.3) + 2 x 0 x -7^00 = 28.8 kPa
Top 0.3 m top soil on passive side is ignored.
Pp - 0.5 x dp x Op = 0.5 x 1.2 x 28.8 = 17.28 kN/m
Location of Pp from bottom = dp /3 = 1.2/3 = 0.4m
These pressures are shown in Fig. 8.15 and are tabulated in Table 8.5 to get the value of
the disturbing moment.

0.275 m
Retaining Wall for Example 8.2
Fig. 8.16. Gravity Forces Acting on

Resisting Moments.
Table 8.6. Vertical Loads and
Magnitude Distance
W from toe Mr
Symbol Expression
S. X
No. kN/m m kN-m/m
25.91 1.338 34.67
Fig. 8.15. Gravity Forces Acting on Retaining Wall of Example 8.2. 0,275 x 4.0 * 23.55
1. IFi 40.04 1.700 ' 68.07
2. W2 0.5 x 3.4 x 23.55 2.438 337.91
138.60
3. Wi 1.925x4.0x18 19.20 0.600 11.52
Table 8.5. Horizontal Loads and Disturbing Moments. Wa 1.2x1,0x16
4. Rv
Magnitude Height 223.0 452.17
s. Symbol Expression from Md Summation Without surcharge
No. toe,y
kN/m m kN-m/m
1. A 0.5 x Han = 0.5 x 4.5 x 27 60.75 1.5 91.13
P. Ho, = 4.5 x 3.6 16.20 2.25 36.45
• P, -0.5tZocrD = - 0.5 x 1.2 x 28.8 -17.28 0.4 -6.91
Summation 59.67 120.67
Rh Ma
The loads providing resistance against over-turning and sliding are shown in Fig. 8.16 and the
resisting moment is calculated in Table 8.6. The position of the resultant of applied forces is
shown in Fig. 8.17.
For the over-turning check, the surcharge load is considered outside of the base. This means
that it will be included in the driving moment but not in the resisting moment.
FOS against overturning = Mtl Ma Acting on Retaining Wall for Example 8.2.
= 452.17/120.67 = 3.75 > 2.00 Fig. 8.17. Resultant Force
Overturning check is satisfied.
274 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A.
Resultant moment about the toe, M SIDDlQI CHAPTER 8: RETAINING WALLS 275
= Mx (CCW) -Ma (CW)
= 452.17- 120.67 = 331.50 CCW Friction force = /x^ = 0.45 x 223.75 = 100.69 kN/m
Horizontal distance of the resultant from the toe, Force resisting sliding, F = fxRv+Pp = 100.69 + 17.28 = 117.97 kN/m
d^ = Ml Rv
= 331.50/223.75 = 1.482 m Sliding force, 5a + A = 60.75 + 16.20 = 76.95 kN/m
One-third of base width = 5/3 = F/CRa+P^ 117.97/76.95 = 1.533 > 1.5
Two-third of base width = 1.133 m FOS against sliding
One-fourth of base width
= 25/3 = 2.267 m Sliding check is satisfied.
= 5/4 0.850 m
Three-fourth of base width = 35/4 = 2.550 m Stem Design
Let A = The toe about which the over-turning is checked.
wiateileKl0llalwiainae B = The point at the junction of the stem with the base.
Tie .oil pressure ar toe may be y = Vertical distance from top of fill up to any point.
denoted by ?I „d ar hee! may be denoted by ymax = Maximum vertical distance from top of fill up to point B.
hs = Thickness of the stem.
- 5 lx53
12
B 1x5’ B
1+31-2—^
d = hs- 50-13 = 212 mm
Backfill soil pressure at point 5 — =
= x 4.0 24 kPa
. 223.75 F -
”12 4.5
1 +’ 1.482 Al Ignoring the passive forces due to toe side fill,
3.4 ^-1 = 91.13 kPa

—_ R^
.
6Mf
= r
-*s» 7,
Afu at y depth = 1.6
27
4.5
x yx
——- —
y y
2 3
r yx y
x + 3.6x
2.
= (l.by3 + 2.88/) kN-m/m
-1-31-2^-
_ B B2
223.75 f (
b
1.482 Al
B K> at depth y = 1.6
27
•— xy
y i

x^ + 3.6x y = (4.8 y1 + 5.76 y) kN /m


3.4 1-31-2-j— J = 40.49 kPa Mi at point B = (1.6 x 43 + 2.88 x 42) = 148.48 kN-m/m
These pressures are plotted in Fig. Fu at distance d higher than point B = (4.8 x 3.788z + 5 76 x 3.788) = 90.69 kN /m
8.18.

Without
\m,n for flexure
M
V 0.205/Jb — +50+ 13

Surcharge
1
148.48 x10s +
63 = 253.3 mm

h^m for shear =


0.205 x 20x1000

' il „
_— +50+13
0.75x0.17777x6
90.69x1000
91.13 = j— +63 = 222.1 mm
0.75x0.17720x1000
Selected depth of stem is OK.
Fig. 8.18. Base Pressures for Example
8.2.
?max = 91.13 kPa Allowable bearing capacity of 100 kPa
fAFfaAB’M'Jbd1 = = 3.304 MPa
1000x212
Bearing pressure check is satisfied. From tables, p = 0.0102, As - 2163 mm2 /m
276 CONCRETE STRUCTURES PART- 1 AUTHOR: ZAHID A. SIDDlQi CHAPTERS:
RETAINING WARS 277

^s^nin — 1.4— b^d — 1.4 x 1000 x 212 / 420 = 707 mm2 /m for main steel. Development Length For #19 Bars
1
£d = d
Use# 19 @ 130 mm c/c. T1^jfc ( Cb + ^tr
The minimum vertical and horizontal steel ratio on both faces of
walls is 0.0025. If the
reinforcement is provided in two curtains, the amount on each face becomes
I A J
0.00125.
fdh
0-24^4 dh > 84 > 150 mm
Amin for one face = 0.00125 x 1000 x 275 = 344 mm2 /m
Use # 13 @ 350 mm c/c for the minimum curtain reinforcement (on
toe side). Where,
Amin for distribution steel = 0.0018 x 1000 x 275 = 495 mm2 /m
Use # 13 @ 250 mm c/c for the distribution steel. 2 =1.0 for normal weight concrete.
below the development
IPt =1.3 if more than 300 mm of fresh concrete is cast
Curtailment of Bars length, otherwise equal to 1.0.
Alternate bars may be curtailed at the location where the capacity of =1.0 for uncoated reinforcement.
becomes greater than or equal to the applied moment plus larger of half of the bars =1.0 for #22 and larger bars, otherwise equal to 0.8.
12<4 and d, two- center of a bar to nearest concrete surface and one-
(

third shear capacity becomes greater than or equal to


applied shear and the development Cb = smaller of the distance ofspacing of bars.
length from the critical section is satisfied. half the center-to-center
Kb = transverse reinforcement index, (q, + A) / 4 must not be taken greater than
Alternate bars 'are curtailed here at reasonable height from the
section giving
section, diviner an
an amount
amniint of
maximum rhoment 2.5.
steel equal to # 19 @ 260 mm c/c or As = 1092.5 mm',2 f 1 420 13x1.0x0.8 xl9 = 675 mm
/m. The moment capacity for this steel may be evaluated
as under: fd fortop #19 bars
AJy _ 1092.5x420
.1.11.0x720 2.5
a ~ 26‘" mm
'

1 420 1.0xl.0x0.8 xl9 = 520 mm


0.85/;/> 0.85x20x1000 ^d for other #19 bars .1-11,0x720 2.5
AM
Class - B splice length in tension for #19 bars = 1.3 £d =
675 mm
= 0.9x 1092.5 x 420 (212 -26.99 /2)/ 106 = 81.98 kN-m/m development length with hook for #36
AM = M => 1.6? +2.88? = 81.98 An additional reduction of 70% is allowed in the plane of hook and at least
By trial solution, y = 3.2 m
and smaller bars with at least 65 mm side cover normal to
50 mm cover on bar extension beyond the hook.
-
12 d\, 228 mm and d = 212 mm, larger of 12
db andd = 228 mm 0.24x1.0x420 xl9 = 300 mm
fdhfor# 19 bars 0.7 x
Hence, y for flexure = 3.2 0.228 - = 2.972 m 1.0x720 .

Development of Dowels in to Base


For shear, Fu —2 A^>
Available length = 500 - 50 - 25 - 25 = 400 mm
4.8 y2 + 5.76y = 2/3 x 0.75x fdh for# 19 bars = 300 mm
4.8 ? + 5.76y- 80.59 = 0
0.17^x W/ 1000 The dowels can easily be anchored.
342 mm (say 350 mm)
y = (-5.76 ±39.76) /9.6 - 3.54m i Minimum extension of> dowels at the bottom = 18 =
Final selected value ofy

For development at critical section, y
least of 2.972 m, 3.54 m and 3.325 m = 2.972 m
= ymax - splice length 1,3fd laps for dowels and curtailment may be combined.
at very low depth. Hence, the
In this example, the curtailment of the bars is possible The dowels may provide full
the base. At 1100 mm from the
moment capacity at the junction of the stem with may be curtailed, while the other half
= 4.0- 675/1000 bottom (2.9 m from the top), half of the dowels
= 3.325 m steel for upper portion
may be spliced with the reduced steel for the upper portion. The
The development length is taken from the calculations given below: may be brought down to the base for additional safety.
RETAINING WALLS 279
278 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS:
1 3] - 0.9 x [At x 23.55 L?
x / 2]
-pi) x L2
Mi = 1.6 x [pi Lt2 +
x / 2 (pi
Heel Design 73.26) x 1.22 / 3]
[73.26 x 1.22 / 2 + (91.13 -
The heel is designed for the condition that it is just lifted up to mobilize full resistance Without surcharge, Mu = 1.6 x
x [0.5 x 23.55 x 1.22/ 2]
against overturning. The self weight of heel (m), weight of backfill (m) and the - 0.9
surcharge load (ws) [as an additional margin] will act on the heel, while the soft = 90.49 kN-m/m
pressure from below is not considered. The moment may not be taken less than the
maximum moment at the base of the stem. The shear force will be critical at the
-
junction with the stem (not at d distance from it) because the load causes tension in
the supporting members and not compression as in regular supports. Here, the load
Pu at d - distance from stem
=
Without surcharge,
-
1.6 x [pj x (Lt ^) + (pi - pi) x {Lt-d) / 2]
= 1.6 x [79.77xx23.55

0.9 x [hf x 23.55 x (Lt

x 0.763]
-
0.763 + (91.13 79.77) x 0.763
- d)]
/ 2]

factor with retained soil weight is taken equal to 1.2, which may be increased to 1.6 if -0.9 x [0.5
more additional safety is desired. = 96.23 kN/m
— 1.2 Wh+ 1.2 m + 1-6 ws
= 1.2 x 0.5 x 23.55 + 1.2 x 4.0 x 18.00 + 1.6 x 0.6 x 18.00
= 117.81 kN/m
Mi = w»*Lh2/2 = 117.81 x 1.9252 /2 = 218.28 kN-m/m
This moment is greater than the stem moment at its base.
= wuxZh = 117.81 x 1.925 = 226.78 kN/m

fibrin for flexure


M
0.205/c'A— +50+13
1 218.28 x10s
—,
+ 63

= 293.7 mm Example 8.2.

_
0.205x20x1000 Fig. 8.19- Forces on Toe for

Amin for shear + 50+13


I 90.54 x10s +63 = 211.6 mm
, «
r flexure = J
226.78x1000 , Ah mtn for 10.205 x 20x1000

d
=
Selected depth of heel is OK
= 500 - 50 - 13 = 437 mm
==-r
0.75 x 0.17720x1000
+63 = 460.7 mm
, r shear
Ah min for
u •
96.23x1000
= ——
0.75 x 0.17720x1000
Selected depth of toe is OK
+ 63 = 231.8 mm

218.28x10'
Mjbd2
1000x437’
1.143 MPa d = 500-50- 13 = 437 mm
From tables,/? = 0.0034, As = 1486 mm2 /m bd2 = 1000x437’ = 0.474 MPa
Amin = 0.0018 bh = 0.0018 x 1000 x 500 = 900 mm2/m
= 0.0025, As = 1093 mm2900 mm2 /m
/m
Use# 19 @ 190 mm c/c for the calculated steel. From tables, p
Use # 16 @ 200 mm c/c for the minimum and the distribution steel. = 0.0018x 1000x500 =
Amin - 0.0018 A A c/c.
Use # 19 @ 250 mm steel.
the minimum and the distribution
Toe Design
Use # 16 @ 200 mm c/c for
The soil fill over the toe is usually neglected in calculating the critical moment and an
overload factor of 1.6 may be applied on the soil pressure. The overload factor on dead
The reinforcement details are shown in Fig. 8.20.
load of base slab may be taken equal to 0.9. The soil pressures acting on the toe are
shown in Fig. 8.19.
280 CONCRETE STRUCTURES PART -I RETAINING WALLS 281
CHAPTERS:
retaining wall used to retain 3 m soil
Example 8.3: Design a cantilever reinforced concrete
1.5 m. The trial dimensions of the stem and
above NSL with an additional burial depth of is as under:
footing are shown in Fig. 8.21. The other related data
y = 18 kN/m3 (Active side), 0=30°, c = 0MPa
y = 16 kN/m3 (Passive side)
Allowable bearing capacity = 200 kPa
=
Friction coefficient,/ 0.35
Wall friction angle, <5=0
Surcharge additional depth = 0.6 m
Overburden depth to ignore = 0.3 m
C-20 concrete and Grade 420 steel are to be used
to one-half its value, to be on conservative
side.
Take passive earth pressure coefficient equal
Solution:
Surcharge additional
depth = 0.6 m
I cjOOOCXa-

y= 18 kN/m3
3m 0=30°
c = 0MPa

y= 16 kN/m3 1.5 m [ 0.4 m


0.6 m

1 1.525 m
0.275 m

Fig. 8.21. Retaining Wall for Example 8.3.

1- sin 1- sin 30
= 0.33
Ks 1 + sin^ l + sin30
1 + sin^ l + sin30
= 3.00
1— sin^ —
1 sin30
Lateral Pressure Due To Soil
(b) PLAN OF FOOTING SHOWING (c)
ONLY THE NECESSARY ELEVATION OF = y#^-2cx
STEM SHOWING SOIL
REINFORCEMNT SIDE REINFORCEMNT = 18 kN/m3 x 4.5 x - 2 x 0 r0.33 =27 kPa
Fig. 8.20. Reinforcement Details for Example 8.2. Pa = 0.5 x Haa = 0.5 x 4.5 x 27 = 60.75 kN/m
Location of -Pa from bottom = #73 = 4.5/3 =
1.5 m
RETAINING WALLS 283
CHAPTERS:
282 CONCRETE STRUCTURES PART - I AUTHOR: ZAHID A. SIDDIQI Surcharge

Lateral Pressure Due To Surcharge


0s = 0.33 xl8 x 0.6 m = 3.6 kPa
Ps = = 4.5 x 3.6 = 16.2 kN/m
Location of Ps from bottom = H12 = 4.5/2 = 2.25 m
Passive Pressure
Kp = 0.5 x original value =1.5
ap = Kpydp + 2cx

Pp

= 1.5 x 16 x (1.5 0.3) + 2 x 0 x ^00 = 28.8 kPa
= 0.5 x dp x ap = 0.5 x 1.2 x 28.8 = 17.28 kN/m
Location of Pp from bottom = dpll - 1.2/3 = 0.4 m 0.275 m
8.3.
Retaining Wall of Example
These lateral pressures, loads and their points of application are shown in Fig. 8.22. The
Fig. 8.23. Gravity Forces Acting on
disturbing moment is evaluated in Table 8.7.

S. Symbol
..
Table 6.6-

Expression
— and Resisting Moments.
Magnitude
W
Distance
from toe
X
m
Mr
kN-m/m
No. 27.56
26.55 1.038
~O275 x 4Tx 2355_ 34.33
nr Wx
0 4 x 2.7 x 23.55
25-43 J 1.350
1938 “"2i8J2~
NJ W2 112.55

Fig. 8.22. Lateral Pressures on Retaining Wall of Example 8.3.


IjpJT Wa

summas^
1,525 x4.1 x 18
0.9 x 0.8x16
— 11.52
Rv
176^05^
0.450 5.18
Mr
285.19

Table 8.7. Horizontal Loads and Disturbing Moments.


Magnitude Height
s. Symbol Expression H from Md
No. toe,j
kN/m m kN-m/m
1. Pa 0.5 x H aa = 0.5 x 4.5 x 27 60.75 1.5 91.13
Ps Has = 4.5 x 3.6 16.20 2.25 36.45
Pp -OSdpffp = - 0.5 x 1.2 x 28.8 - 17.28 0.4 - 6.91
Summation 59.67 120.67
Ph Md
The resisting forces are show in Fig. 8.23 and the resisting moment is evaluated in Table 8.8.
Figure 8.24 gives the position of resultant of the applied forces.
RETAINING WALLS 285
284 CONCRETE STRUCTURES PART -1 CHAPTERS:
AUTHOR: ZAHID A. SIDDIQI
FOS against overturning = Mr! Ma Pp = 17.28 kN/m
= 285.19/120.67 = 2.36 > 2.00 Friction force, F = fx.Rv~ 0.35 x 176.05 = 61.62 kN/m
+ 17.28 = 78.90 kN/m
Overturning check is satisfied. Total force resisting sliding = fxRv + Pp = 61.62 kN/m
Resultant moment about the toe, M = Mx (CCW) - (CW) Sliding force, & + Ps = 60.75 + 16.20 =78.90
76.95
/ 76.95 = 1.025 < 1.5
M^ FOS against sliding = F/^+Ps) =
= 285.19- 120.67 = 164.52 CCW
Horizontal distance of the resultant from the toe, Sliding check has failed, key is to be provided.
dR = Ml Bv
= 164.52/176.05 = 0.935 m Depth of key, yt = 600 mm
One-third of base width Thickness of key, hk = 275 mm
= BJ3 = 0.900 m Depth of bottom of key from effective level of soil
overburden on toe, y = 1.8 m
Two-third of base width = 2B/3
One-fourth of base width = 1.800 m mathematically equal to the following:
= B/4 = 0.675 m The resistance provided by the key is
Three-fourth of base width = 3B/4 = 2.025 m
The resultant lies within the middle third Pk =- Pk where ~ Kp yy, Gi = Kp y(y-yk)
and no tension will be developed within
(y-yj2) = 3.0x 16 x 0.6x(1.8 - 0.6 =
base. the / 2) 43.20 kN/m
The soil pressure at toe may be denoted by yr and at
= Kpry^
resisting sliding, F = 78.90 + 43.20 = 122.1 kN/m
Total force 1.5
Without surcharge - - .
heel may be denoted by yn-
FOS against sliding = P/(Pa+Ps) = 122.10/76.95 = 1.587 >
Sliding check is satisfied.
=A
B
^R Stem Design
B
176.05 Same as for the previous example.
2.7
l+ 3fl-2i«5'
I 2.7 ,
= 125.34 kPa Heel Design
wa = 1.2 wj, + 1.2 Wb + 1.6 ws
= 4B 1-3G-24 d,
B = 1.2x0.4x23.55 + 12x4.1x18.00 + 1.6x0.6x18.00
176.05
= 117.15 kN/m
kN-m/m
2.7 Ma = woxZh2/2 = 117.15 x 1.525z/2 = 136.22
at its base.
This moment is greater than the stem moment
K = w„ x Zb = 117.15 x 1.525 = 178.66 kN /m
Afonin for flexure
A
0.205<'b— +50 + 13

136.22x10* +63 = 246 mm


0.205x20x1000
Vu r
+ 50+13
Ah,min for shear
0.75 x 0.17^ xb
178.66x1000 + 63 = 377 mm
Fig. 8.25. Base Pressure of Example 8.3.
0.75 x0.17>/20x 1000
Qmax = 125.34 kPa Allowable bearing capacity of 200 kPa
Bearing pressure check is satisfied. Selected depth of heel is OK
RETAINING WALLS 287
286 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
CHAPTERS:

MJbd2
— 400 -50 13—
136.22x10s
= 337 mm
= 1.200 MPa
K, aid- distance from stem
= 1.6 x [p3 x - d) + (pi -pd *(Lt~d) / 2] “ °-9 x fa x 23'55 x
- 2]
1000x337* Without surcharge, Va = 1.6 x {100.26 x 0.563 + (125.34 100.26) x 0.563 /
From tables, p = 0,0033, = 1112mm2/m
- 0.9 x [0.4x23.55x0.563]
A,min = 0,0018 bh = 96.84 kN /m
= 0.0018 x 1000 x 400 = 720 mm2 /m
Use # 19 @ 250 mm c/c for the calculated steel. ^xl°_+63 = 193 mm
Use # 16 @ 275 mm c/c for the minimum and the distribution steel. Kmin for flexure Y 0.205 x 20 x 1000
The development length is taken from the calculations given below: 96.84x1000 +63 = 233 mm
Mijnin for shear =
Development Length For #19 Bags 0.75x0.17720x1000
As in the previous example, the development lengths are as under: Selected depth of toe is OK,
fortop #19 bars 1 420 1.3x1.0x0.81 d - 400-50- 13 = 337 mm
.1.1
1
1.0 x 720
420
25
1.0xl.0x0.8
*19
J = 675 mm
Mu / bd2 69.13x10^ = 0609MPa
1000x337*
for other #19 bars =
1-11.0x720 2.5
xl9 = 520 mm
Class -B splice length in tension for #19 bars = 1.3 £d = 675 mm
From tables, p
= 0.0018 b =h

= 0.0023, As = 776 mm
0.0018 x 1000 x400
mm2 /m
=
/m
720 mm /m
Use# 19 @350 mm c/c.
^dh for # 19 bars = 0.7 x 0.24x1.0 x 420 xl9 = 300 mm Use # 16 @275 mm c/c for the minimum and the distribution
steel.
1.0x720 Fig. 8.27.
Toe Design The reinforcement details within a typical cross-section are shown in
The pressure acting on the toe is reproduced in Fig. 8.26. The soil fill over the
toe is 8.8 SEISMIC LATERAL EARTH PRESSURE
usually neglected in calculating the critical moment and an overload factor of 1.6
may loads, the seismic lateral
be applied on the soil pressure. The overload factor on dead load of
base slab may be For retaining walls that deflect significantly under the action of lateral
equation. In case of
taken equal to 0.9. earth pressure can be estimated by the Mononobe - Okabe pseudo-static
to 1 / 1000 of the wall
sandy soils, a forward displacement of the top of the retaining wall equal
the inertial forces of
height may cause transition from at rest to active earth pressure. However
the wall are to be added separately. The Mononobe - Okabetheory method is an extension of the
is preferable in the sense
This soil load may be Coulomb’s sliding wedge theory. Coulombs’s earth pressure the wall and the soil, the
ignored in designing the toe. that it considers additional parameters including the friction between
the wall.
inclination of the wall and the inclination of the ground surface behind
The Mononobe - Okabe theory is based on the following assumptions:
sliding wedge
a) The basic concept of this theory is the same as that of the Coulomb
theory.
Pi = 125.34 P2 = 85.25 Without surcharge create active earth pressure
b) The wall will deflect sufficiently in an elastic way to
conditions.
c) The backfill is of uniform properties and is granular and cohesion-less.
Fig. 8.26. Forces for Toe Design of Example 8.3.
d) The surface of the backfill is in the form of a smooth plane.
Mi =1.6 x [pz x L2 / 2 + (pi -pz) x Lt / 3] - 0.9 x [At x 23.55 x if / 2] of the soil.
Without surcharge, Mu = 1.6 x [85.25 x 0.92 / 2 + (125.34 - 85.25) x 0.92 / 3] e) Ifthe surcharge is present, it is applied over considerably larger surface
-
0.9 x {0.4 x 23.55 x0.92/2]
= 69.13 kN-m/m
288 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
RETAINING WALLS 289
CHAPTERS:

#19 @ 260 nun c/c


iff

0
= seismic inertial angle

( k
= tan'1 I 1 ~
kv J
.1

= inclination of wall on soil side with respect to the vertical. 90 6.


ft with respect to the horizontal = -
= inclination of wall on soil side following
#13 @350 mm c/c general values:
= wall friction angle, having the
#13 @ 250 mm c/c
Gravel
Coarse sand
Fine sand
27 30
20 28
15-25

-
#19 @ 130 mm c/c Stiff clay 15-20
Silty clay 12-16
= inclination of soil surface with respect to the horizontal.
wall with respect to the vertical are
1100 The expressions for earth pressures in terms of angle of the
600 j- #19 @ 250 nun c/c as follows:
#16 @275 mm c/c| TT" F

Ik
#19 @ 350 mm c/<
#13 @ 250 mm c/c

550
SECTION OF THE WALL
Fig. 8.27. Reinforcement Details for Example 8.3.
Lateral Earth Pressure.
The following notation is used in the formulas: Fig. 8.28. Geometric Parameters for Defining Seismic

= active earth pressure coefficient due to earthquake, including static and dynamic cos2^-^-^)

=
effects.
passive earth pressure coefficient due to earthquake, including 0008^ + 0 + ^1+ ^;^^
r sin(^ + 5)sin(^ - - P)
effects.
static and dynamic cosmos
= active static pressure coefficient due to soil. cos2(^-^ + ^)
= unit weight of the backfill. Xpe
[ I sin(^ + <5)sin(^ - yz + ^)
Av = vertical earthquake acceleration in terms of g. cos cos2 0COS^ - 0 + <5) 1
Ah = horizontal earthquake acceleration in terms of g.
H = total height of soil, including the soil above the ground and the depth of the footing. sin2(a + ^)cos J
0 = internal friction of the soil. '
'

rsin(^ + <5)sin(^-/?)
sinasin(a-5) 1 +
sinG0 + a)sin(a-5)
RETAINING WALLS 291

290 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS:

The expression for active earth pressures in terms of angle of the wall with respect to the
vertical is as follows:
r
_ sin2(a + ^-{/)
cos y sin2 a sin(a - - 5) yI sinQ?
sin(0 + 3)sin(^ -y-P)
+ a) sin(a - 3-
Pae =
Fpe = (I
Fig. 8.29. Seismic Lateral Earth Pressure.
If k? - 0, 3= 0, p- 0 and Q~ 0 (or a~ 90°), then the above expressions are simplified to:
cos2(^-^)
cos2 p< /sin^sin(^ ~
y cos^-
cos2(^-^)
cos2 ~y 1-1 Dynamic Active
Static Active
Pressure
Pres'sure
1 -sin^
K*
1+sin^ a) Earth Pressures if <5=0 and cr = 90°.
= tan'(Ah)
P^ = ^yH2 = ^yH2
The Mononobe - Okabe theory gives results only when the following are satisfied:
kb (1 Av) tan - p) for active wedge

kb (1 - ky) tan P for passive wedge


When Ah 0.2, A, may be neglected.
This method predicts very high earth pressures when At > 0.2, hence a maximum value of Ah ~ Dynamic Active
Static Active Pressure
0.25 is usually taken. Pressure
Net dynamic force = NP^ = Wall on Seismic Earth Pressure.
Fig. 8.30. Effect of Slopping
The location of is to be taken equal to 0.6H from the bottom. To satisfy this location of
the resultant force, the dynamic pressure may be assumed to act as an inverted prism, as shown SEPARATE STRUCTURAL ELEMENTS
8.9 LATERAL FORCE ON elements, non-
in Fig. 8.29. The earth pressures for various conditions are diagrammatically shown in Figure
8.30. UBC97-1632, the total lateral force on separate structural is evaluated as
According to equipment
in buildings) and supported
Important Note: When the seismic earth pressure is included in combination for the design of structural components (like parapets
retaining walls, the factor of safety against sliding and overturning may be reduced from 1.778 under:
and 0.75x1.778 = 1.35, considering the short time effect of such loads. Fp =4.0CaZplFp
RETAINING WALLS 293
CHAPTERS:
292 CONCRETE STRUCTURES P ART - 1 AUTHOR: ZAHID A. SIDDIQI ..
sou,
,
p = 2.5 -0-3 = 22
Effective depth of passive
Where, kN/m
~ — dP Cp = 115.7
Fp ~ seismic force having the units of Wp,
= dp/ 3 -14.04°
2 0.73 m
Ca = seismic coefficient from table 16-Q of UBC, Location from the bottom
Ip
Wp
= importance factor, 1.0 for standard occupancy structures, and,
= weight of the com ponent. = tanl(fo) = tan'1(0.25) =
cos2(0-^)
This force is applied in the horizontal direction so as to create the most critical loading. K.
Example 8.4: Calculate seismic lateral earth pressure for the retaining wall shown in cos2!/ 1+
Figure 8.31 and having the following data: cos1(30s14;04) = 0.518
g = 10kPa ^30sin^0-^j^.
cos2 14.04 1 + cosl4.04
/= 18 kN/m3
^
5d 0=30°
'

P*
x 18 x 7.52 (1 - 0) =
c = 0MPa 262.4 kN/m
= 1x 0.518
53533^
/=16 kN/m3 2.5 m = = P^-P, kN/m
Net dynamic force = 262.4-168.8 = 93.66
Figure 8.31. Retaining Wall for Example 8.4. 4.5 m
= 0.6 H = 19.98
Location from bottom = kPa
Top pressure = 1.6/HxAPm kPa
a) Use surcharge for sliding, over-tuming and bearing. ZfxAPae = 5.00
b) Use passive force for over-turning, but after leaving top 0.3 m of soil. Bottom pressure = 0.4/ pressures are shown in Figure 8.32.
seismic lateral earth
c) FOS for overturning = 2.00 and FOS for sliding = 1.50. The resulting additional
d) Allowable bearing capacity = 200 kPa. 19.98 kPa
e) Friction coefficient, /= 0.45.
f) Wall friction angle, 0°.
g) Coulomb analysis is to be used for the active pressure.
APac= 93.66 kN

_
h) th = 0.25 and k, = 0.
i) Ignore top 0.3 m soil for passive resistance.
Solution: h-7.5 0.6H-4.5
„ 1-sM l-ih.30
= 03M
l + sin0 l + sin30
5.00 kPa
Bottom pressure, = X» /H = 0.333 x 18 x 7.5 = 45 kPa for Example 8.4.
Figure 8.32. Seismic Lateral Earth Pressure
Pa |
= H aa = 0.5x7.5x45 = 168.8 kN/m FOR WALLS
8.10 BENDING MOMENT COEFFICIENTS
LOADS
Height of Pa from the bottom = thickness of footing + 13 SUBJECTED TO TRIANGULAR
=15+6/3 = 3.5 m
= height of wall
Xp = tan2 (45° + 0/ 2) = 3.000 Let H panel
= c/c width ofthe wall
ob = yc^Kp + lc^ = 105.4 kPa S
294 CONCRETE STRUCTURES PART RETAINING WALLS 295
-1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS:
= aspect ratio = width / height
= intensity of maximum pressure= S/ H - 0.5 for given coefficients Burial depth = 3m
Center-to-center spacing of counterforts = 5.5 m
= vertical moment coefficients y = 18 kN/m3 (Active side).
= horizontal moment coefficients
= vertical moment = Kv x q S2 0 =30°
c =0MPa
= horizontal moment = Ar x q S2
/ = 16 kN/m3 (Passive side).
8.10.1 Upper One-Third Of Wall Allowable bearing capacity = 300 kPa.
Friction coefficient,/ = 0.50.
Ah at support = 0.0323 for triangular Wall friction angle, 3 = 0.
load, 0.0954 for uniform load, 0.0858 for
triangular load (tension on fill side) inverted Surcharge additional depth = 0 m.
Ak at midspan = 0.0166 for triangular Overburden depth to ignore = 0.3 m.
load, 0.0488 for uniform load, C-20 concrete and Grade 420 steel are to be used.
triangular load (tension opposite to fill side). 0.0436 for inverted equal to one- half its value, to be on conservative
Av span Take passive earth pressure coefficient
= 0.00326 for triangular load, 0.00968 for uniform load, side.
0.00753 for inverted sliding = 1.35.
triangular load (tension opposite to fill side). FOS for overturning = 1.35 and FOS for pressure,
Coulomb analysis is to be used for the active
8-10.2 Middle-Third Of Wall th = 0.25 and iv = 0-05. earth load plus minimum stem seismic load
of
Ak at support = 0.0545 for triangular load, 0.0908 for uniform load, Consider seismic load greater of seismicwith
0.0627 for inverted 2.4 kPa and separate stem seismic load Ca = 0.4A4 andNa = 1.3.
triangular load (tension on fill side) +1.677 and U = 0.9D + 1.0A+ 1.6F/'
Ah at midspan = 0.0268 for triangular For strength design, U = 1.2 D
For service conditions, S' = D ± 0.8E H such that 0.9D is changed to D).
load, 0.0468 for uniform load, +
0.0323 for inverted
triangular load (tension opposite to fill side). (Obtained by adjusting the UBC combination
Av span = 0.0132 for triangular load, 0.0162 for
uniform Solution;
triangular load (tension opposite to fill side). load, 0.00687 for inverted
8.10.3 Lower One-Third Of
Wall
At at support = 0.0545 for
triangular load, 0.0852 for uniform load, 0.0311 for inverted
triangular load (tension on fill side)
Ah at midspan = 0.0268 for triangular
load, 0.0426 for uniform load, 0.0159 for inverted
triangular load (tension opposite to fill side).
Av at bottom = 0.0496 for
triangular load, 0.0616 for uniform load, 0.0120 for inverted
triangular load (tension on fill side)
Av span = 0.0145 for triangular load, 0.0230 for uniform load,
triangular load (tension opposite to fill side). 0.00363 for inverted Example 8.5.
Figure 8.33. Selected Given Data for
^Aampie 8,5: Design
a cantilever reinforced concrete retaining
shown in Fig. 8.33, used to retain
The other related data is
wall with
9 m soil above NSL with an additional burial counterforts,
depth of 3.0 m.

_ 1-sin^ _ l-sin30
J + sin30
as under: 1 + sinp
Trial heel width ~ 6 m. Ap = l+_sm£ = l^sm30 = 3 00 or tan2 (45° + <4/ 2)
Trial toe width = 2 m. 1-sinfi 1- sin 30
Trial stem thickness 450 mm. = 1.50
=
Trial footing thickness
Effective Ap
Bottom pressure, o"s = K^yH ~ 0.333 x 18 x 12 = 71.93 kPa
= 1000 mm.
296 CONCRETE STRUCTURES PART -I RETAINING WALLS 297
AUTHOR: ZAHID A. SIDDIQI CHAPTERS:
Pi = -11^= 0.5 x 12x71.93 = 431.6 kN/m Table 8.9. Horizontal Loads and
Disturbing Moments.
Height of Pa from the bottom

Op '"
K^yd^ +
'

lc^

— — HI'S 12/3 = 4.0 m
Effective depth of passive soil, dp = 3.0 -0.3
- = 2.7 m
= 1.50x 16x2.7 + 0 = 64.8 kPa
s.
No.
Symbol Expression
Magnitude
H
Height
from
toe,y
m
Md
kN-m/m
kN/m
p
|
= d9 = 0.5 x 2.7 x 64.8 = 87.48 kN/m 1 p* O.SxZfo'a = 0.5 X 12x71.93_ 431.6 4.0 1 mo.4
- /».»
-0.5^ = - 0.5 x 2.7 x 64.8 - 87.5 0.9
Location from the bottom = dp/S = 2.7/3 pn
x p
= 0.9 m 177.8 7.2 1280»2
0.8xAPae 0.8x(APae-Pa)
= 0.8x(653.9- 431.6) 6.5 137.3
U-frJ (1-0.05J 0.8xFp n 8 y 76 4 21.17
543.02 3065.1
= cos2(^— (y) Snmm ation

cos2^ 1+
Ph —
- Md

8.10, while Fig. 8.36 shows the resultant


The vertical loads are given in Fig. 8.35 and Table
COS2 (30 -14.74) force acting on the retaining wall and its location.
p = 0-531
cos2 14.74 4 + /sin30sin(30- 14^74)
cos14.74
P» 1
= = |x 0.531 x 18 x 12.02 (1-0.05) 653.9 kN/m
Net dynamic force = APae - P* - P-
= 653.9-431.6 = 222.3 kN/m
Location from bottom = 0.6 H = 0.6 x 12
Top pressure = 7.2 m
=
Bottom pressure 1.6//7xAPM = 29.64 kPa
= 0.4 /77 x
APae 7.41 kPa
Minimum seismic load of stem, -
Pp,min = 2.4 kPa
Resultant force = 2.4 x 11 = 26.4 kN/m 0.45 m
Retaining Wall of Example 8.5.
Fp = 4.0 CAIP Wp = 4.0 x 0.4 x 1.3 x 1.0 x 0.450 x 23.55 = 22.04 kPa Figure 8.35. Gravity Forces Acting on
Hence seismic earth load with
loads are shown in Fig. 8.34 and minimum stem load is more critical. The horizontal
are presented in Table 8.9.
Table 8.10. Vertical Loads and Resisting Moments.
Magnitude Distance
Expression IP from toe Mr
S. Symbol
No. kN/m m kN-m/m
1 Wt 0.45x 11.0 x 2335 _^6___^
199.0 4.225 ow.o
0 W, 1.0 x 8.45 x 23.55
5.450 04/4.0
i W. 6.0x11.0x18
7 —^-130x1.7x16 ^4 L00O

Figure 8.34. Pressures Acting


on Retaining Wall of Example 8.5. Summation —— pv

— —1558.0 l— -1—
Mr
/ozv.z .
RETAIMING W ALLS 299
298 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS:

Example 8.5.
Figure 8.37. Base Pressure for
Figure 8.36. Resultant Force Acting on Retaining Wall of Example 8.5.
„ = 353 40 kPa Allowable bearing capacity of 400 kPa
FOS against overturning Bearing pressure check is satisfied.
= 7629.2/3065.1 = 2.49 > 1.35
Pp = 87.48 kN/m kN/m
Overturning check is satisfied. Friction force, F = = 0.50 x 1558.0 = 779.0866.48 kN/m
sliding, Fa = f*Rv + Pp = 779.0 + 87.48 =
Resultant moment about the toe, M = Mr (CCW) - Md (CW) Total force resisting
Sliding force, all except Pp, = 630.52 kN/m / 630.52 = 1.374 > 1.35
= 7629.2-3065.1 = 4564.1 CCW = F^IF^ = 866.48
FOS against sliding
Horizontal distance of the resultant from the toe, dg = Ml Rv
= 4564.1 /1558.0 = 2.929 m Stem Design
= The toe about which the over-turning is checked.
One-third of base width = 5/3 = 2.817 m Let A of the stem with the base.
Two-third of base width - 25/3 = 5.633 m B = The point at the junction top of fill up to any point,
One-fourth of base width = 5/4 = 2.113 m
y = Vertical distance fromdistance from top of fill up to point
B.
Three-fourth of base width = 35/4 = 6.338 m yma* = Maximum ofvertical mm
= Thickness the stem = 450 vertical steel)
The resultant lies within the middle third and no tension will be developed within the d -
= hs 50 - = mm (for horizontal
13 387 mm (for
steel)
base. = 387-19 = 368
——
d 71.93 x 11.0 = 65.94 kPa
Backfill soil pressure at top of base, q =
'
The soil pressure at toe may be denoted by qj and at heel may be denoted by q^.

qy = ^ri+3fi-2^
I J B
H
S
= height of wall =wall
= c/c width of the
11.0m

-
-
panel 5.5 m
H 5.5/11= 0.5
it = aspect ratio = SIcoefficients
8.45 [ 8.45 J
= 354.05 kPa Kv - vertical moment
Ak = horizontal moment coefficients
Mv - vertical moment
9h 1-3 fl-2—
l B Mb
= Kv x q S2
= horizontal moment
1558 R = Kb^qS2
8.45[ 3 flf 22-929!
8.45 J
= 14.71 kPa
I naris for 0.9P + 1.0# + 1 -6H Combination
The base pressure is shown in Fig. 8.37. , = 1l 6x65
bx 94 = 105.50 kPa
q for lower triangle (earth pressure)
z
300 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS: RETAINING WALLS 301
q for upper triangle (seismic earth pressure) = 1.0x20.38 = 20.38 kPa Development Length For #19 Bars
q for uniform load (seismic earth pressure) = 1.0x2.4 + 1.0x9.26 1 fy
= 11.66 kPa db
Minimum Steel
Minimum wall steel on one face (ratio = 0.00125) = 563 mm2/m
I J
Minimum beam / cantilever slab steel (ratio = 0.00333) = 1229 mmz/m > 150 mm
Minimum slab steel (ratio = 0.0018)
^dh
= 810mm2/m
Where,
The calculations for the moment produced in the stem and — the corresponding required
reinforcement are presented in Table 8.11. For the calculation of steel, b = 1000 mm, d = 368 A = 1.0 for normal weight concrete.
mm,//=20 MPaand/^ = 420MPa. = 1.3 if more than 300 mm of fresh concrete is cast below the development
length, otherwise equal to 1.0.
= 1.0 for uncoated reinforcement.
Table 8. 11. Stem Bending Moments. = 1.0 for #22 and larger bars, otherwise equal to 0.8.
Upper Third Height Cb = smaller of the distance of center of a bar to nearest concrete surface and one-
Middle Third Height Lower Third Height
Item Support I Midspan half the center-to-center spacing of bars.
Support 1 Midspan Support | Midspan
Horizontal Direction Xt, = transverse reinforcement index, (cb + X^) / db must not be taken greater than
Xh,A 0.0323 0.0166 0.0545 0.0268 0.0545 2.5.
0.0268
M 103.08 52.98 173.93 85.53 173.93 85.53 1 420 1.0xl.0x0.8
Xh, UDL 0.0954 0.0488 0.0908 0.0468 0.0852 for other than top #19 bars = xl9 = 520 mm
M,
0.0426 _1.11.0xV20 2.5
33.65 17.21 32.03 16.51 30.05 15.03
Ah, V 0.0858 0.0436 0.0627 0.0323 0.0311 0.0159 Class - B splice length in tension for #19 bars = 1.3 fd = 675 mm
Mi 52.90 26.88 38.65 19.91 19.17 9.80 An additional reduction of 70% is allowed in development length with hook for #36
Total 189.63 97.07 244.61 121.95 223.15 1’10.36 and smaller bars with at least 65 mm side cover normal to the plane of hook and at least
Horiz. R = 1.400 A = 0.717 R= 1.806 R = 0.901 R= 1.648 R = 0.815
50 mm cover on bar extension beyond the hook.
Moment p =0.0039 p =0.0039 p =0.0039 p =0.0039 p =0.0039 p =0.0039
As W/m 1435 810 1950 810 1767 810 0.24x1.0x420
Steel ^dh for# 19 bars = 0.7 x xl9 = 300 mm
#19®190 #13®150 #19@140 #13®150 #19®160 #13®150 l.Ox 720
. Vertical Direction
Xh,A —— 0.00326 — 0.0132 0.0496 0.0145
Development Length For #13 Bars

——
M 10.40 42.13 158.29 46.27 1 420 l.Ox 1.0x0.8 „ =
Xh, UDL — 0.00968 0.0162 0.0616 0.0230
for other than top #13 bars = ^-=-
2.5
x!3 356 mm
1.11.0x720
—— ———
M, 3.41 5.71 21.73 8.11

'

*h,V 0.00753 0.00687 0.0120 0.00363 Class B splice length in tension for #13 bars = 13 fd = 462 mm
M,
Total
— 4.64
18.45
4.24
52.08
7.40
187.42
2.24
56.62
An additional reduction of 70% is allowed in development length with hook for #36
and smaller bars with at least 65 mm side cover normal to the plane of hook and at least
Vertical R = 0.136 R = 0.385 X= 1.384 R = 0.418 50 mm cover on bar extension beyond the hook.
Moment Distr. /?=min Distr. p- min p =0.0040 p~ min 0.24x1.0x420
As mm2/m 810 810 810 810 1472 810 £dhfor# 13 bars = 0.7 x xl3 = 206 mm
Steel #13@150 #13®150 #13®150 #13@150 #19®190 #13®150
Counterfort Design
The forces acting on a typical interior counterfort are shown in Fig. 8.38.
302 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS: RETAINING WALLS 303 '

Let thickness of the counterfort = 600 mm


Load at depth of 9 m from top = 559.26 kN/m Steel area in tension to cany the above load —
= -^££/1
0.9x420
= 3898 mm2
= 1949 mm2 per face
#19@140mmc/c
Horizontal force at junction of wall and counterfort per meter width of height
= 598.44 kN
, . . . . , , , 559.26x1000 = 1480 mm2
Steel area m tension to cany the above load =
0.9x420
= 740 mm2 per face
#13 @ 170 mm c/c
Heel Slab Design
The thickness of footing is increased to 1300 mm for safety against shear.
wu =1.2x11x18+1.2x1.3x23.55
= 274.34 kN/m per meter width

Top Steel
Figure 8.38. Base Pressure for Example 8.5.
At. = wux52/ 11 = 274.34 x 5.52/ 11 = 754.5kN-m/m
= 1,300-50-13 = 1237 mm
,,
Momentt at depth y =M

- 176.222 x/ f 468.16
= 88.11 / + 7.093/
116
/
x v x i-
d
At. / bd2 = JTS^SxJO^ - q 4p2g MPa
1000x1237*
(minimum steel)

Shear at depths = V = 176.22 y + 21.28 y 2 ^in = 0.0018 b h = 0.0018 x 1000 x 1300 = 2340 mm2 /m
Use # 25 @ 200 mm.
Afu at the top of buttress = 12,308 kN-m
Fu at the top of buttress = 3,310 kN-m Bottom Steel
d = 5359 x sin 671.39 - 126 = 4578 mm
R = 0.9788 => p = 0.0034 At. = wuxS2/ 16 = 274.34 x 5.5 2 / 16 = 518.7 kN-m/m
As = 0.0034 x 600 x 4578 = 9340 mm2
= 0.75 x 0.17-720 x 600 x 4578 / 1000 = 1566 kN
Fu . > ^Fc => Provide shear reinforcement.
19- #25 At. / bd2
518-7*10
= .1000x1237 — = 0.3390 MPa
^min = 0.0018 bh = 0.0018 x 1000 x 1300 = 2340 mm2 /m
(minimum steel)

Using 2-legged #13 stirrups, Av ~ 258 mm2 Use # 25 @ 200 mm.


= least of Maximum Shear
"k
i)

0.356,
= 516 mm
Cl
K. =274.34 x (5.5 / 2 - 0.3) = 672.14 kN /m
0.75 x 0. 1 7 -720 x 1000 x 1237 / 1000 = 705.33 kN
=

—_
ii) d! 2 = 2289 mm
.....
UI) 0.75x258x420x4578
: ~ ..
_
““ 1OO
Hence, the slab is safe in shear.

(3310- 1566) xl000 Development Length For #25 Bars


= 175 mm 1 420 1.3xl.0x0,8
Downward force at junction of heel and counterfort per meter width of counterfort for top #25 bars x 25 = 888 mm
1.1 1.0x720 2.5
= 1.2 x 10.7 x 18 x 5.5 + 1.2 x 1.3 x 23.55 x 5.5 = 1473.22 kN
Class - B splice length in tension for #25 bars = 1.3 Id - 1155 mm
CHAPTERS: RETAINING WALLS 305
304 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

Toe Slab Design


The maximum upward unfactored pressure on the toe is 354.05 kPa.
wu = 1.6 x 354.05 + 0.9 x 1.3 x 23.55
= 538.93 kN/m per meter width

Bottom Steel
Afu = lesser of wu x 52 / 11 (top and bottom) and wu x L 2 / 2 (only bottom)
= 538.93 x 5.5 2 /11 [1482] and 538.93 x22/ 11 [1078] kN-m/m
= 1078 kN-m/m (Cantilever action is dominant)
d = 1,300-75- 13 = 1212 mm

M / bd1 = - 1078x106
1000x1212"
= Q.7339 MPa (minimum steel)

.4, min as a cantilever = 0.00333 bd = 0.00333 x 1000 x 1212 = 4036 mm2 /m


Use #25 @ 125 mm.
yls min (distribution) = 0.0018 bh = 0.0018 x 1000 x 1300 = 2340 mm2 /m
Use # 25 @ 200 mm.

Maximum Shear
Ku = 566.48 x (2 - 0.937) = 602.17 kN /m
^Vc = 0.75 x 0.17720 x 1000 x 1237 / 1000 = 705.33 kN
Hence, the slab is safe in shear.

Development Length For #25 Bars


1 420 1.0xl.0x0.8
for bottom #25 bars= x25 = 683 mm
1.11.0x720 2.5
Class - B splice length in tension for #25 bars = 1.3 Lt = 888 mm
Figure 8.39. Reinforcement Details for Example 8.5.
The reinforcement details are shown in Fig. 8.39. The plan 3-3 is drawn above the buttresses
level and it gives the layout of the counterforts and the reinforcement within the wall and the
counterforts. Section 1-1 clarifies the details of the stem, heel and toe reinforcement while
Section 2-2 gives details of the counterfort and buttress reinforcement in cross-section.
306 CONCRETE STRUCTURES PART - I AUTHOR: ZAHID A. SIDDIQI CHAPTER 8: RETAINING WALLS 307
PROBLEMS

Note: N = Registered number of the student.


TVo = A fixed positive or negative number assigned by
the instructor to get R between 600 and 1 000.
R = -
N No, a parameter to get numerical data for the
problems.

1- A gravity retaining wall is to be constructed in masonry using (1:6) c/s mortar and burnt
clay bricks (see figure) and is to retain 22/400 m height of soil besides heavy traffic
loads. The wall is additionally embedded by 0.4 m into a soil that is medium / stiff
clay. The backfill is expected to have a density of 1750 kg/m3. The allowable
compressive stress in masonry is to be taken equal to 0.54 MPa and the safe bearing
capacity of the soil is 72/8 kPa. The factor of safety against sliding should be a
minimum of 1.5. Ignore passive resistance of soil.

1.0 m

R/400m

0.4 m

2- Design a cantilever reinforced concrete retaining wall, without key, used to retain /2/200
m soil above NSL with an additional burial depth of /2/500 m. The other related data is
as under:

y = 18 kN/m3 (Active side)


^=30°
c = 0 MPa
y = 16 kN/m3 (Passive side)
Allowable bearing capacity = /2/8 kPa
Friction coefficient,/ = 0.3 + 22/4000
Wall friction angle, 8 = 0
Surcharge additional depth = 0.6 m
Overburden depth to ignore = 0.3 tn
Figure 8.39 (Continued). Reinforcement Details for Example 8.5. C-20 concrete and Grade 420 steel are to be used
308 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI
CHAPTERS: RETAINING WALLS 309
3- Design a cantilever reinforced concrete retaining wall used to retain 72/200 m soil above
NSL with an additional burial depth of 22/500 m. The other related data is as under: <0 = 30°
c = 0 MPa
/ = 16 kN/m3 (Passive side).
y = 18 kN/m3 (Active side), =30°, c =0 MPa Allowable bearing capacity = 72/3 kPa.
y = 16 kN/m3 (Passive side) Friction coefficient,/ = 0.35 + 72/4000.
Allowable bearing capacity = 200 kPa Wall friction angle, <5=0.
Friction coefficient,/ = 0.15 + 72/4000 Surcharge additional depth = 0 m.
Wall friction angle, 3 = 0 Overburden depth to ignore = 0.3 m.
Surcharge additional depth = 0.6 m < C—20 concrete and Grade 420 steel are to be used.
Overburden depth to ignore = 0.3 m FOS for overturning = 1.35 and FOS for sliding = 1.35.
C—20 concrete and Grade 420 steel are to be used Coulomb analysis is to be used for the active pressure.
Key may be provided , if required ith = 0.25 and ky = 0.05.
Consider seismic load greater of seismic earth load plus minimum stem seismic load of
4- Calculate seismic lateral earth pressure for the retaining wall shown in the figure and 2.4 kPa and separate stem seismic load with Ca = 0.4Wa and 7/ = 1.3.
having the following data: For strength design, U = 1.2D + 1.6 77 and U = 0.9 D+ 1.0 E + 1.6 77
For service conditions, S = D ± 0.85 + H
a) Use surcharge for sliding, over-turning and bearing.
b) Use passive force for over-turning, but after leaving top 0.3 m of soil.
c) FOS for overturning = 2.00 and FOS for sliding = 1.50.
d) Allowable bearing capacity = 72/4 kPa.
e) Friction coefficient,/= 0.35 + 72/4000.
f) Wall friction angle, 3= 0°.
g) Coulomb analysis is to be used for the active pressure.
h) kh = 0.25 and kv = 0.
i) Ignore top 0.3 m soil for passive resistance.

q = 10 kPa

y= 18 kN/m3
72/200 m ^=30°
c = 0MPa

/=16 kN/m3 z.u m

Design a cantilever reinforced concrete retaining wall with counterforts used to retain
72/100 m soil above NSL, with an additional burial depth of 72/400 m. The other related
data is as under:

Center-to-center spacing of counterforts = 72/400 + 1.0 m


/ = 18 kN/m3 (Active side).
310 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI '

Space For Reader’s Notes Chapter - 9

DESIGN OF FOOTINGS

9.1 INTRODUCTION

Foundations are the structural elements that are used to transfer the structural loads to the soil
giving sufficient margin of safety against its failure, without excessive settlement and rotation.
The loads on the columns and walls are concentrated over smaller areas and the function of
foundations is to spread this load over larger areas until it becomes equal to the safe bearing
capacity of soil.

9.2 TYPES OF FOOTINGS

Footings can be made in various shapes and types depending upon the use and the underlying
soil characteristics (Fig. 9.1). Strip or wall footings are used under the walls and these consist
of one-way slabs cantilevered out from the bottom of wall in one or both the lateral directions.
The wall footing may also be a stepped wall footing, where the slab thickness decreases in
steps away from the wall. Isolated column footings can be in the form of reinforced concrete
pads under the columns, which may be in the form of spread footings, stepped footings or
tapered footings. The isolated footing for comer column of a building, where there is site
restriction on two sides, may be in the form of a slab extending as a cantilever only in two
perpendicular directions. Similarly, for the edge column, the isolated footing extends on both
sides of the column parallel to edge of the building but is cantilevered only on inner side in the
perpendicular direction. Two or more columns can have combined footing in the form of
either a rectangular base or two individual bases connected by a narrow strip (The latter type is
called strap footing). In some cases, the combined footing can have a foundation beam
running along the column centerline and is called foundation beam footing.
If the bearing capacity under the footing is too low compared with the applied load, the base
may be supported by piles driven into the ground. The base slab in this type of foundation is
designed as a pile cap and the resulting footing is called pile cap footing. The piles may carry
load to a deeper underground strata, if available, may cany the load by surface friction at the
outer surface of the piles or by a combined action. The first type of pile is called end-bearing
pile, the second is termed as friction pile and the third is known as a mixed pile. Caissons may
also be used in place of piles, which are large diameter hollow well foundations that transfer
the load to deep soil strata. The wall of the caissons is built at ground level with a reinforced
ring at the bottom and then the soil from inside is excavated. This drives the caisson into the
ground, called sinking of the well. In case the isolated column footings for a building come
closer to each other and reinforcement is required on both top and bottom of the pad, it is
advantageous to combine all the footings in the form of a raft, mat or floating foundation.
The raft may be a base slab of constant thickness or a slab supported by foundation beams at
each column centerline. Raft is also effective in reducing the chances of differential settlement
312 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A SIDDIQI
of different parts of the building. A bridge footing is used to erect a column over a place that
is already occupied.

Tapered Spread Footing

Edge Column Footing

Stepped Column Footing

Fig. 9.1. In continuation of the previous page,


Fig. 9.1. Types of Footings,
(continued on next page)
(continued on next page)
314 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTERS: DESIGN OF FOOT! NGS 315
considered better than the isolated column footings for the edge and the comer columns, as
isolated footings for these latter columns produce high bending moments in the columns. This
type is also preferred when the isolated footings of any two columns come close to each other.
The suitable foundation type is selected and designed for a building not only to reduce total
settlement, differential settlement and chances of failure of soil but also to eliminate the
chances of sliding and overturning of the building under gravity and lateral loads. As a rough
guideline, if the area covered by isolated footings for a particular building exceeds half of the
area within the perimeter of the building, it is usually better to use raft. However, if the steel is
provided only at the bottom for the isolated footings, the total amount of steel required in case
of raft may become larger due to double layer of steel in larger part of the raft.
Generally, raft is constructed at deeper levels such that the total weight of the excavated
material becomes equal to the total expected weight of the building. This will only apply load
on the soil equal to what actually existed before the excavation and the structure will be
floating on the foundation. According to ACI 1.1.4, ACI 332 report is allowed to be used for
the design of footings for one and two family dwellings.

9.3 BEARING PRESSURE

When structure loads are concentrically applied on isolated footings, the bearing stresses are
not actually uniformly distributed over the area of the footing due to deflected shape of the
foundation and uneven behavior of the soil underneath (Fig. 9.2). Cohesive soils produce shear
resistance at the edges of the footing and hence more bearing pressure is developed near the
perimeter as compared with the bearing pressure inside the footing, as shown in Fig. 9.2a. In
case of granular soils, the soil particles near the end of the foundation tend to slide towards the
less stress area (sideward, outside the footing) and hence a distribution shown in Fig. 9.2b is
obtained with more pressure near the center and less pressure near the ends of the footing. The
Grid or Grillage Foundation in Plan determination of actual pressure distributions and the resulting moments and shears for
practical designs becomes quite cumbersome and hence simplified pressure distribution shown
in Fig. 9.2c is usually utilized for design of isolated footings and other simple types of
foundations. The assumption of uniform pressure under the footings produces insignificant
difference in shears and moments within the acceptable limits of tolerance. For mats and other
sophisticated foundations, it is better to use more refined estimates of bearing pressures in
place of the common practice of using the uniform pressure distribution.

Fig. 9.1. In continuation of the previous page.

Differential settlement of foundations is very dangerous for the super-structure as when some
a) Actual Distribution b) Actual Distribution c) Assumed
columns sink down relative to the others, very heavy forces are transferred to the beams and
for Cohesive Soils for Granular Soils Distribution
slabs through the beam-column joints. This may cause excessive cracking and yielding of the
steel reinforcement that may even lead to final failure of the structure. Combined footings are Fig. 9.2. Bearing Pressure Distributions.
CHAPTER 9: DESIGN OF FOOTINGS 317
316 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

The pressure at the contact surface of the foundation with the soil is obtained by dividing the
9.4 ALLOWABLE BEARING CAPACITY total foundation load with the area of the footing. However, the load that directly acts over the
foundation slab counteracts the contact pressure up to certain extent. This reduced pressure to
The allowable bearing capacity of a soil, denoted by at a certain level is defined as the be used for the structure design of the foundation components is called net contact pressure,
amount of maximum external uniform pressure that may be applied with sufficient margin of whose unfactored value is denoted by qa and factored value is denoted by q^.
safety against failure of soil and against excessive settlement. The bearing capacity of a
shallow foundation depends on the depth, shape, type and width of foundation along with the Net contact pressure = total foundation load per unit area of footing
type of soil, the natural level of compaction and consolidation, the ground water table level, the - load directly acting on the foundation slab
depth of frost penetration and the drainage properties of the area. The allowable settlement,
the chances of differential settlement and the required factor of safety are also important
Table 9.1. Approximate Allowable Bearing Capacities.
factors in the final establishment of the bearing capacity. The factor of safety against the
rupture of soil is taken from 2 to 3, the most common value is usually equal to 3. Sr. Type of Soil Allowable Bearing Capacity
It is estimated that architectural damage to buildings occurs when the amount of differential No. kPa tons/tf
settlement divided by the distance between the two points where the settlement is noted (called 1. Solid crystalline bedrock. 10700 100
differential settlement per unit length or distortion) exceeds 1/300, and structural damage 2. Foliated rocks, such as schist or slate. 4300 40
occurs when this value reaches 1/150. The value of distortion normally allowed in design is 3. Sedimentary rocks such as sandstones, 1600 15
1/3300 for LIH 3.0 in sandy soil, 1/2500 for L/H < 3.0 in clayey soil, 1/2000 for L!H > 5 in limestones and siltstones.
sandy soil and 1/1400 for L!HJ> 5 in clayey soil, where L is the width of the building and H is 4. Gravel-sand mixtures (loose, medium and 320-540 3-5
the height of the building. However, a good isolated foundation is that which does not settle dense).
more than 25mm under the application of allowable bearing pressure. The allowable bearing 5. Well-graded sands and gravely sands (loose, 215-375 2.0 - 3.5
capacity based on settlement is usually taken as the applied pressure producing a maximum medium and dense).
settlement not to exceed 25mm and differential settlement not to exceed 20mm. For sandy 6. Poorly graded sands and gravely sands (loose, 160-320 1.5 -3.0
soils, the settlement may be predicted by the blow counts of the standard penetration test, the medium and dense).
larger the blow counts; the less will be the settlement. 7. Gravel-sand-silt mixtures and silty gravels 160-270 1.5 -2.5
(loose, medium and dense).
The allowable bearing capacity at a certain depth (A) with the soil backfilled is called net 8. Silt-sand mixtures. 215 2.0
allowable bearing capacity; and the allowable bearing capacity with no backfill is called gross 9. Clayey mixtures with sand and gravel. 215 2.0
bearing capacity.
Gross bearing capacity = net bearing capacity +
10. Mixture of inorganic silts, fine sands and
inorganic clays.
80 110 -
. 0.75 - 1.0

An approximate idea about the allowable soil bearing capacity of various types of soils may be 11. Organic materials, clays of high plasticity, etc. — —
obtained from Table 9.1. However, it is important to note that the design must be based on the
actual geo-technical investigations at a particular site. Modulus of sub-grade reaction (Ks) as 260 q„ - 0.25 q„
When backfill load is applied on any soil layer at certain depth, we need not to provide full Unfactored loads are generally used for sizing of the foundations. Accordingly, unfactored
factor of safety of 3.0, but a reduced value of 1.5 may be used. This means that half of the load combinations are to be utilized for this purpose, these combinations given by UBC-97 are
backfill load is assumed to consume part of the bearing capacity. For example, 1.2m of as under:
backfill with a soil of density 1600 kg/m3 may be considered to consume 1. D 2. D + L + (LrorS)
0.5x1600x1.2x9.81/1000 = 9.4 kPa of bearing capacity. Most building codes permit a 33 3. D + (JTor*E/1.4) 4. 0.9Z>±£/1.4
percent increase in the allowable bearing pressure for certain service load combinations 5. ZJ + 0.75[i + (Z,orS) + (^or£/1.4)]
(but not all) involving wind or earthquake.
No increase in allowable stresses is to be used with these load combinations.
In the case the soil pressures are determined by detailed computer programs and small pockets
Where, D = dead load, L = live load, Lr = roof live load, S = snow load, W = wi nd load and E =
under the footing or near its edges have up to 15 percent more pressure than the allowable
earthquake load.
bearing pressure, these may be considered acceptable based on plastic redistribution of
confined soil and the footing above.
9.5 NET CONTACT PRESSURE
4136, 4137
Ph: 042-3 5 90 5 000 Ext:

SHAUKAT KHANUM MEMORIAL CANCER


HOSPITAL & RESEARCH CENTRE
DIAGNOSTIC SERVICES
RADIOLOGY
MAMC- /1 MR MCI'
i

Ph: 042-35905000’Ext: 4136, 4137

SHAUKAT KHANUM MEMORIAL CANCER


HOSPITAL & RESEARCH CENTRE
DIAGNOSTIC SERVICES
. RADIOLOGY
.0 ^f^y
SHAUKAT KHANUM MEMORIAL CANCER
CO HOSPITAL & RESEARCH CENTRE
DIAGNOSTIC SERVICES
. RADIOLOGY

PROCEDURE: .
APPHIMTMPMT DATO TIMO
AI

Ph: 042-35905000 'Ext: 4136, 4137

SHAUKAT KHANUM MEMORIAL CANCER


HOSPITAL & RESEARCH CENTRE
DIAGNOSTIC SERVICES
r RADIOLOGY

NAME:

PROCEDURE:

APPOINTMENT DATE: TIME:


jo

PROCEDUCE EST. COST UNIT COORDINATOR


CHAPTER 9:
318 CONCRETE STRUCTURES PART I - AUTHOR: ZAHID A. SIDDIQI design of footings
319
In place of the above allowable load combinations, UBC-97 also allows the following alternate Due to the presence of simultaneous high vertical
compressive
combinations: stresses at the top of footing and lateral tensile stresses at the stresses, lateral compressive
1. D + L + (Lr or S) 2. D + L + (IT or £/1.4) section for calculation of strength against this type of shear bottom of footing, the critical
distance from the face of the column compared with d failure is to be considered at dil
3. D+L + W+S/2 4, D + L + S+W/2 distance taken in case of one-way shear.
The critical sections for one-way and two-way
5. D + L.+ S + E/X.A 6. 0.9D±£/1.4 shears are shown in Fig. 9.4.
When using these alternate basic load combinations, a one-third increase is allowed in
allowable stresses for all combinations including W or E.
For concentrically loaded footings, the required area is determined as follows: Critical section for
one-way shear Critical section for
S!qt Where S is the service load two-way shear
After sizing of the foundation, the factored contact pressure is calculated by the usual factored
load combinations and design is carried out according to these factored actions. To check the
overturning of a foundation, only those parts of the loads causing overturning should be Fig. 9.4. Critical Sections for Shear failure.
included and these must be balanced by dead loads multiplied with 0.9, maintaining a
minimum factor of safety against overturning of 1.5. It is not a common practice to provide shear
reinforcement in the footings and hence these are
usually designed by sufficiently increasing the depth
of the footing such that the concrete shear
9.6 FAILURE BEHAVIOR OF ISOLATED FOOTINGS strength alone is greater than the applied punching
shear. In single footings, the total and
effective depth (d) of the footings is mostly governed by
A sufficiently thick isolated rectangular column footing may failure by flexural cracks parallel design process.
shear and is decided in the start of the
to each side close to the column or by one-way shear cracks also at the same location.
However, the most common failure of such a column is by punching through the relatively thin The concrete strength against two-way shear is
estimated as follows:
slab. The failure occurs around the perimeter of the column by a special type of shear called
lesser of the following three expressions
the 2-way or punching shear, as shown in Fig. 9.3. As a consequence, a truncated pyramid in
case of rectangular columns and a truncated cone in case of circular columns, with sides
sloping outward at an angle of approximately 45°, tend to separate from the slab and move
1. 0.75x 0.33^7/7M
0.75x0.17^1
with the column.

where b0
d
2.

3. 0.75x 0.083
+

I bo ^+2
2.^ bBd
|2.^bod
J
= critical section perimeter, at a distance d/2 from the
= effective depth of the footing column perimeter
P = ratio of the long to short sides of the column
a* = a constant, 40 f6r interior loading, 30 for edge loading and 20 for comer
loading of a footing.
and A = 1.0 for normal weight concrete.
According to ACT 15.7, the depth of footing above bottom
than 150mm for footings on soil and not reinforcement should not be less
less than 300mm for footings on piles.
The applied load causing punching shear is
calculated by taking sum of all the contact pressure
acting outside the critical perimeter of the column,
as shown shaded in Fig. 9.5.
c) Failure Mechanism in Sectional Elevation Punching shear, Fu = qu x area of footing outside critical
perimeter
Fig. 9.3. Punching Type Failure of Foundations. -
= qa x (B x L bi x bi)
320 CONCRETE STRUCTURES PART - r AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 321
Bending moment for longer steel =• qa x (L - ci)2 / 8

Bending moment for shorter steel = qa x (B - c2)2 / 8


Maximum one-way shear for flexure = qu x (L 1 2 - ci / 2 - d)
If bending moment in a column produces uplift on one side of the footing and soil above
prevents this uplift, top steel in the footing must be provided for the bending moment
calculated on the basis Of the weight of the soil.

9.8 TRANSFER OF LOAD FROM COLUMN TO FOOTING

Fig. 9.5. Area Affective in Producing Punching Shear. The concentrated load in the columns is transferred to the footing by direct bearing of the
column over the footing (producing bearing stresses at the interface) and by forces in the
9.7 DESIGN BENDING MOMENT AND dowels or column main steel bars that cross the joint. The loaded area or the area of cross-
ONE-WAY SHEAR FOR ISOLATED FOOTINGS section of column is denoted by At, whereas, a larger area of footing may be considered
effective in resisting the bearing stresses and is denoted by ^2 The area /1? at a certain depth
The critical section for calculating the inside the footing is found by spreading the area At at a slope of lesser of 2 horizontal to 1
pedestal or wall. When the footing supports a mas
midway between the center of the footing and
all is°tX vertical for solid footing and actually available steeper slope on each side. The concrete
present in the larger area ^2 around the loaded area At provides lateral confinement to the

between^ ^^^of
base plates, the
concrete and causes increase in the bearing strength. The increase in the bearing strength is
critjalsection is taken halfway steel base plate. Referring
to Fig. 9.6, it can be seen that a unit strip acts jus
uniformly distributed load, qw, which is equal to the net con
p
beamP subjected t0 a considered equal to I Ax with a maximum value of 2.0. The bearing strength (^fn) is thus
estimated by using the following expression with the value of equal to 0.65:
0Pa = ^0.85/^^/^ < 0.85/^ x 2.0
Where, fc' is the cylinder strength of the footing concrete.
The above expression is used to check the concrete in the footing just below the column,
without dowels having development length beyond this region.
Excess force to be resisted by dowels, having development length beyond the bottom of the
-
footing = Pu $Pa
P -dP
Area of steel required for dowels within the footing = —
- , where = 0.65

ACI minimum area of steel required for dowels = 0.005 AE, where, As = gross area of the
supporting member.
The above check and amount of dowels is required to make sure that that the footing concrete
is capable of taking the full load of the column. However, if some of the column steel is
discontinued at the top of footing, the portion of column just above the footing will be lacking
fully developed bars. The bearing strength of this bottom portion of the column is to be
checked. The continuing steel must provide a dowel steel area to develop a force equal to Pu -
a) Critical Section for Bending b) Critical Section for Bending 00.85/X.
Along Longer Dimension Along Shorter Dimension

Fig. 9.6. Critical Sections for Failure.


Area of steel required for dowels within the column = —— ——
P„-^0.85//rt
tfy
, = 0.65
322 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 323
According to ACI 15.8.1.2, reinforcement, dowels or mechanical connectors between
supported and supporting members must be adequate to transfer all compressive force that fdc = 0.24 > 200mm for fc < 31 MPa
exceeds the concrete bearing strength and any computed tensile force across the interface. ACI
15.8.2.1 requires that for cast-in-place columns and pedestals, area of reinforcement across = 0.043/ d/b 200mm for/' > 31 MPa
interface must not be less than 0.005Ag. Dowels must have diameter less than or equal to No. The lap length in case of a Class-B tension splice should not be less than 1.3 and 300mm.
35 and must extend into the supported member by a distance equal to larger of development The lap length for compression splices is evaluated as follows:
length in compression of the column main steel bars and the splice length of the dowels.
Similarly, these dowels must extend into the footing by a minimum distance equal to the Compression splice length = 0.071 fydbF 300mm, for/ 420 MPa
development length of the dowels. ACI 15.8.2.3 permits the use of No. 44 and No. 57 bars in = (0.13/ - 24) db F > 300mm, for/, >420 MPa
columns and their splicing with the lesser diameter dowels, if the bars are in compression.
Where, if/'> 20 MPa, F- 1.0, otherwise, F= 1.3.
9.9 DISTRIBUTION OF FOOTING STEEL Where bars of different diameters are lap spliced in compression, the required splice length is
to be taken larger of the development length in compression for the larger diameter bar and
The steel bars required along the longer side of the footing are uniformly distributed across the splice length for the smaller diameter bar.
entire width of the footing. However, the reinforcement in the short direction is concentrated
more in the central band of dimension equal to width of the footing. The ratio of reinforcement 9.11 CONCRETE PROTECTION FOR REINFORCEMENT
to be placed in the central band with respect to the total required reinforcement is taken equal
to where p is the. ratio of long to short side of footing. The remainder of the The minimum clear cover for concrete exposed to earth should be 20mm for slabs, walls and
joists and 40mm for other members. Concrete cast against and permanently exposed to earth
reinforcement is to be uniformly distributed outside the central band of reinforcement should have a minimum clear cover of 75mm.

9.10 DEVELOPMENT AND SPLICING OF STEEL 9.12 PROCEDURE FOR DESIGN OF ISOLATED FOOTINGS
(CONCENTRICALLY LOADED)
The development and splice lengths for dowels are to be checked. The development of footing
steel is to be checked at the critical sections for moment, changes of section and sections where 1. Gather all input information, such as the allowable bearing capacity of soil, foundation
part of the reinforcement is terminated. The column steel must also be folly developed within depth, type of foundation, support reactions for applicable load combinations and soil
the footing, if tensile forces are present in it. Considering normal weight concrete, steel refill, etc.
without any coating, no transverse reinforcement and the clear cover at least equal to 1.5^, the 2. Using service loads, calculate the required area of footing. Select the trial size and
development length in tension may be written as follows: shape of the footing depending on the area and other site-specific requirements.
3. Estimate the depth of footing based on punching shear requirements in an approximate
fd > 300mm for top bars and «4 No. 20 way, which must not be lesser than 250mm according to the ACI code. The
approximate depth may be found by using the following expression:
= o.485^=rfb 300mm for bottom bars and db

> 300mm for top bars and db > No. 20


No. 20
J F+l.5^
V
d
I + 1.5^
0.0\asfc
= H-60
— + 60 mm > 250 mm

Where moments are in meter units and <zs is is a constant equal to 40 for interior
300mm for bottom bars and db > No. 20 loading, 30 for edge loading and 20 for comer loading on a footing.
4. Using the factored loads, estimated self-weight of the footing and the weight of the
The development length in compression, without considering the modifications for presence of backfill, calculate the design contact pressure, qu.
confining steel and area of steel in excess of that required, is calculated as follows: 5. Calculate the factored bending moment, one-way shears and two-way shear at the
critical sections.
324 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI
CHAPTER 9: DESIGN OF FOOTINGS 325
6. Check the effective depth of footing for punching shear. Similarly, check the slab for P+1.5M+1.5M.
one-way shear. One-way shear may be critical in case the footing is made slopping Approximate depth of footing, H + 60 mm
towards the ends.
250 mm
7. Calculate the amount of steel required in both the perpendicular directions. The
minimum steel must be 0.002d/z for Grade 300 and 0.001 ^bh for Grade 420 steel. 980x1000 + 60 250 mm
=
8. In case of rectangular footing, determine the amount of steel required in the central } 0.01x40x20
band. Decide the diameter and spacing of bars satisfying the maximum spacing = 410 mm (say 450 mm)
requirements.
d = H- 60 = 390 mm
9. Check bearing strength of the footing and decide the required number and size of dowel
bars.
bo = 2 x (C1 + c2 + 2d) = 2 x (450 + 450 + 2 x 390) = 3360 mm
P = 450/450 = 1.0
10. Check and satisfy all development length and lap splice requirements.

Example 9.1: A 450 mm square interior column, reinforced with eight No. 25 bars of
<|>Kc = lesser of 0.75x 0.332^ bad, 0.75x 0.17^1 -^4^
+ b°d 311(1

Vjfl bod
_
Grade 420, supports a dead load of 700 kN and a live load of 400 kN. Assume that a live load
0.75x 0.083 +2
reduction of 30% may be allowed at the foundation level. The foundation is to be placed at a I bo J
depth of I m where the gross allowable bearing capacity is 110 kPa. The average density of the lesser of the following:
backfill material and the footing concrete together may be taken as 2100 kg/m3, while the
filling from ground to plinth level is to be 0.5m of material with average density of 1800 „„„720x3360x390 = 1450 kN
1. 0.75x0.33—
kg/m3. Design a square footing using// = 20 MPa,/ for foundation steel = 300 MPa and / for 1000
dowel steel = 420 MPa.
Solution:
31 = 110 kPa
2.

3.
——————
0.75x0.1t(1+—1——
1J -
1000

o.75x0.08/“^+21^^^»
2241 kN

= 2423kN
Pd = 700 kN t 3360 J 1000
Pl = 400 kN = 1450 kN
Live load reduction = 30 % for punching shear = qu x {Af- (ci + d) (c2 + 7)}
/' =20 MPa
= 118.3 x {3.32 - (0.84) (0.84)}
/ = 300 MPa
= 1204.8 kN < ^)Kc ««)
/y for dowels = 420 MPa
Depth of footing = 1.0 m For I m wide strip:
Average density of fill and foundation =2100 kg/m3
Density of surcharge = 1800 kg/m3 Cantilever length in short direction, /
= (3.3 -0.45) /2 = 1.425 m
P = PD + Pl = 700 + (1 - 0.3) x 400 = 980 kN
= 110 - (1.0 x 0.5 x 2100 + 0.5 x 1800) x 9.81/1000 = 90.87 kPa Cantilever length in short direction, = (L-ci)/2
fret
req ~ P ! ^net
= (3.3 -0.45) /2 = 1.425 m
= 980 / 90.87 = 10.79 m2 (say 3.3 m x 3.3 m foundation) Bending moment for longer steel, Mu = ?ox^z2/2
Pa = 1.2PD + 1-6FL = 1.2x700 + 1.6x0.7x400 = 1288 kN
qu = PJAt
Bending moment for shorter steel = same as above
= 1288/ 3.32 = 118.3 kPa
Maximum one-way shear for flexure = q^LIZ-Cill-d)
= 118.3 x (3.3/2 - 0.45/2 - 0.39)
= 122.44 kN
CHAPTER 9: DESIGN OF FOOTINGS 327
326 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
Development length in compression for #25 bars,/, = 420 MPa, is:
= 0.75 X 0.1 M
4>KC = 0.24^=f db
y c
200 mm
0.75 x0.17 = 222.4 kN > applied shear (OA)
420
1000 = 0.24 x x 25 = 564 mm
12°^ = 0.790 <20
Development length in compression for #1 9 bars, fy = 420 MPa, is:

= pxbd = 0.0030 x 1000x390 1170 mm2 /meter width
Aj, min = 0.002 *bh = 0.002 x 1000 x 450 = 900 mm2 / meter width ^dc = 0.24^= db
V ft
200 mm
Selected steel: #13 @ 1 10 mm c/c or #16 @ 170 mm c/c
Ai = Ac - 202,500 mm2
= 0.24 x
420
<20
— x 19 = 429 mm

A2 is the area of an enclosed figure consisting of column size plus an additional Revised = fdc x required area of dowels I provided area of dowels
dimension of minimum of actual projection beyond column and four times total depth = 429 x 1013 /2272 = 192 mm
of slab on ail four sides. The lest projection out of the four sides must be used on all the Compression splice length for #19 bars = 0.071 Fdb 300mm
sides.
= 0.071x420x1.3x19 = 737 mm
Let, A2 = 4741 = 810,000 mm2 (900mm x 900mm) The selected dimensions of the footing and the reinforcement details are shown in Fig. 9.7.
a +4h = 2250 mm and actual projection is 2850 mm
A2 >4 Ai is available ; ^A27Al = 2.0
^Pn = fix 2 x 0.85J7 Aj
= 0.65 x 2 x 0.85 x 20 x 202,500 / 1000 = 4475 kN > Pu (OK)
Foundation concrete is capable of resisting the column load even without the
dowels.
Now we check the dowels required to provide development length for the column steel.

Area of steel required for dowels within the column


Pu ~^5f:Ag

1 288 x 1000 - 0.65 x 0,85 x 20 x 4502


= zero
0.65x420
Minimum area of steel required for dowels = 0.005 Ac
= 0.005 x4502 = 1013 mm2

4 - #19 bars are sufficient as dowel reinforcement. However, 8 #19 bars may also be
used for a conservative design.
Development length in tension for #16 bars,,4 = 300 MPa, is: Fig. 9.7. Reinforcement Details for Example 9.1.

= 0.485 300mm for db No. 20 bottom bars The lap splice length for #25 column bars and #19 dowels is to be taken larger of the
development length in compression for #25 bar (564 mm) and splice length for #19 bar (737
mm), that is, approximately 0.75 m. This lap will be provided above the footing level. The
= 0.485 x 16 = 521 mm length of dowels into the footing must not be less than ^dc for #19 bars equal to 429 mm, which
<20
328 CONCRETE STRUCTURES PART 1 - AUTHOR; ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 329
is greater than the effective depth of the footing slab. Four solutions are possible: increase area 1450x1000 , „cn
of dowels, use smaller diameter bars, increase thickness of footing or provide pedestal. The +60 250 mm
first requirement is satisfied here.
0.01x40x25

The required development length for flexure of footing steel at the face of column is 0.521 m = 441 mm (say 500 mm)
and the available length is (3.3 - 0.45) /2= 1 .425 m. Hence the steel is fully developed at the d '
- H-60 = 440 mm
critical section. The footing steel may not be curtailed near the edges of the footing in this b0 = 2 x (ci + c2 + 2d) = 2 x (450 + 300 + 2 x 440) = 3260 mm
example because it is greater than the minimum steel only by a small margin. However, if the P = 450 / 300 = 1.5
footing thickness is tapered towards the ends, half of the steel may also be curtailed at a well-
_ f 0 1

determined location. If the option of reducing the foundation slab thickness towards the ends <Fc = lesser of 0.75x 0.332^V, 0.75x0.171 1+- W77 V a™*
is utilized, it is important that this reduction must not start from within <H2 distance from the
face of column. Similarly, the development length of continuing bars from the theoretical cut¬
off point must be checked against the available length. The placing and compaction of
concrete may become significantly difficult for the latter option.

Example 9.2: A 450 mm x 300 mm interior column, reinforced with eight No. 25 bars of
0.75x0.083
^
I b„
+2

lesser of the following:


J12777

^X326OX 440 = 1775 kN


V

1. 0.75x0.33
Grade 420, supports a dead load of 1000 kN and a live load of 600 kN. Assume that a live 1000
load reduction of 25% may be allowed at the foundation level. The foundation is to be 2)725x3260
rectangular (having longer to shorter side ratio of approximately 1.2) and placed at a depth of 2. 0.75x0.17
A^
1H = 2134 kN
Im where the gross allowable bearing capacity is 240 kPa. The average density of the backfill
1.5 J 1000
material and the footing concrete together may be taken as 2100 kg/m3, while the filling from
ground to plinth level is to be 0.7m of material with average density of 1900 kg/m3. Design
3. 0.75x0.083f«^t2'|^
I 3260 ) 1000
= 3303 kN
the footing using// = 25 MPa and fy for all steel = 420 MPa. = 1775 kN
Solution: Fu for punching shear = q^ x {Af- (ci + d) {c2 + d)}
q^ = 240 kPa = 285.72 x {2.4x2.8 - (0.89) (0.74)}
PD = 1000 kN PL = 600 kN = 1732 kN {OK)
Live load reduction = 25 %
For 1 m wide strips:
f! = 25 MPa : L = 420 MPa
Depth of footing = 1.0 m Cantilever length in short direction, 4 = (S-c2)/2
Density of fill = 2100 kg/m3 = (2.4 -0.30) /2 = 1.05 m
Density of surcharge = 1900 kg/m3
Depth of surcharge = 0.7 m Cantilever length in short direction, = (L-ci)/2
= (2.8 -0.45) /2 = 1.175 m
P = Pv + Pl = 1000 + 0.75x600 = 1450 kN
Bending moment for longer steel, = qu x t2 ! 2
285.72 xl.1752 _
?net = 240 - (1.0 x 0.5 x 2100 + 0.7 x 1900) x 9.81/1000 = 216.65 kPa
req = P/ = 2
= 197.24 kN-m
= 1450/216.65 = 6.69 m3 = 1.2 x2
(say 2.4 m x 2.8 m) Bending moment for shorter steel, = 9u x V/2
= 1.2Pd+ 1.6Pl = 1.2 x 1000 + 1.6 x 0.75 x 600 = 1920 kN 285.72 x 1.052
- = 157.50 kN-m
= PjAf 2
= 1920/(2.4x2.8) = 285.72 kPa

Approximate depth of footing, H = J


P + 1.5Mx + 1.5Mr
—- + 60 mm > 250 mm I Maximum one-way shear for flexure = qa^{L /2 - ci 1 2~d)
= 285.72 x (2.8/2 - 0.45/2 - 0.44)
= 210.00 kN

>1.
330 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 331

0VZ =0.75x1^^
6 ^A2 /A, = 5.807 (consider maximum value of 2.0)

-
Longer Steel:
p/75 x 1^*
6
10011 * 440
1000
= 275 kN > applied shear (OA-)
0Pn = 0*2 *0.85 fc Ai
= 0.65x2x0.85x25x 135,000/1000 = 3729 kN >
Foundation concrete is capable of resisting the column load even without the
(OK)

dowels.
p = 0.0028 Now we check the dowels required to provide development length for the column steel.
bd1 1000 x4402
_
— pxbd —
0.0028 x 1000 x 440 = 1232 mm2 / meter width
-
As, min - 0.0018 xbh 0.0018 x 1000 x 500 = 900 mm2 / meter width
Area of steel required for dowels within the column -
_

P„-^0.85/X
ff,
1920x1000-0.65x0.85x25x450x300
Selected steel:
Shorter Steel:
#16 @ 160 mm c/c = — 0.65x420
= 202.61 mm

Mm 157.50 xlO6 Minimum area of steel required for dowels = 0.005 Ac


bd2 1000 x4402
p = 0.0023 0.005 x 135,000 = 675 mm2
=

A^niin
-

pxid,— 0.0023 x 1000 x 440 = 1012 mm2 /meter width
0.0018 *bh —
0.0018 x 1000 x 500 = 900 mm2/ meter width

used for a conservative design.

4 #16 bars are sufficient as dowel reinforcement. However, 8 #16 bars may also be

Development length in tension for #13 bars, fy = 420 MPa, is:


Central band width: 2400 mm
P — ratio of long to short side of footing = 1.17 = 0.485 -^=<7b
V ft
300mm forJb No. 20

Reinforcement in central band 420


= x total amount of reinforcement = 0.485—i= x 13 = 530 mm
P +\ 725
= 0.923 x 1012 x 2.8 / 2.4
Development length in tension for #16 bars, fy - 420 MPa, is:
= 1090 mm2 per meter width
Selected steel: #16 @180 mm c/c 420
U = 0.485— x 16 = 652 mm
( 2
725
Reinforcement in outer band = 1 x total amount of reinforcement Development length in compression for #25 bars.^y = 420 MPa, is:
I 0+V
= (1-0.923) x 1012x2.8/0.4
= 0.24 db > 200 mm
= 546 mm2 per meter width
Selected steel: 1 #13 in each 0.2m outer band
Whenever depth of footing exceeds 450mm, it is better to provide minimum steel (0.001%!2xbh
= 0.24 x —725= x
420
25 = 504 mm
— 450 mm /m) on the top also. In this example, #13 @ 275 mm c/c in both the directions may
also be provided at the top. Development length in compression for #16 bars, fy = 420 MPa, is:
420
Ai = Ac = 135,000 mm2 ^dc = 0.24 x —X
725
x 16 = 323 mm
A2 = (lesser of 2800 and ci + 4tf) x (lesser of 2400 and c2 + ^d)
— (lesser of 2800 and 450 + 4x440)
x (lesser of 2400 and 300 + 4x440)
= 4,552,600 mm2
Compression splice length for #16 bars = 0.071 Ty f
=
300mm
0.071 x 420 x 1 x 16 = 477 mm
332 CONCRETE STRUCTURES PART - 1
CHAPTER 9: DESIGN OF FOOTINGS 333
AUTHOR: ZAHID A. SIDDIQI

The lap splice length for #25 column bars and #16 dowels is to be
taken larger of the
development length in compression for #25 bar (504 mm) and splice length
for #16 bar (471
mm), that is, approximately 0.55 m. This lap will be provided
above the footing level. The
length of dowels into the footing must not be less than for #16 bars equal
to 325 mm.
The required development length for flexure of longer footing steel at the
face of column is
-
0.652 m and the available length is (2.8 0.45) / 2 0.05 = 1.125 m.
-
development length for flexure of shorter footing steel of #16 diameter
Similarly, the required
at the face of column is
-
0.53 m and the available length is (2.4 0.30) / 2 - 0.05 1.00 m.
= Hence, the steel is fully
developed at the critical sections. The reinforcement details are shown in Fig. 9.9. Position of Load and Contact Area of Footing.
Fig. 9.8.

The resulting soil pressure (q) may be calculated as follows:

where P = vertical load, positive in compression


L = dimension of footing parallel to x-axis
B = dimension of footing parallel to y-axis
A = contact area of footing with the soil = A x B
Mx = moment about x-axis
y = distance of point parallel to y-axis from centroid, where pressure is required
Zx = moment of inertia of the contact area about the x-axis
= LB"
12
My = moment about y-axis
x = distance of point parallel to x-axis from centroid, where pressure is required
Iy = moment of inertia of the contact area about the y-axis

= ^L
12
Fig. 9.8. Reinforcement Details for Example 9.2. The contact pressure variation due to axial load and moment about one of the axes is shown in
Fig. 9.10. The axial load produces uniform compression throughout while the moment
9.13 produces tension on one side and compression on the other side. The resultant value at one of
ECCENTRICALLY LOADED ISOLATED FOOTINGS the edges adds and maximum compression is obtained there. On the other edge, there may be a
1
The footing x-axis may be defined as the centroidal axis parallel lesser magnitude tension, no stress or a less compressive pressure. Similarly, if moments are
to the longer
dimension and y-axis may be defined as the centroidal axis parallel to the shorter footing applied in both the directions, one comer becomes most heavily stressed in compression while
footing the diagonally opposite comer has the least compressive pressure or tension.
dimension. The isolated footings may be subjected to a combination of axial load (P), moment
about x-axis (M = P x ey) and moment about y-axis (My = P
* e^. The eccentricity e* is the
eccentricity of load parallel to the x-axis and ey is the eccentricity
9.14 KERN OF CONTACT AREA
parallel to the y-axis. The
position of load with respect to the footing centroid and the
centroidal axes are shown in Fig. The kern or core of the contact area is defined as that area around the centroid of the footing
9.9.
inside which if load is applied no tension is developed any where in the area (Fig. 9.11). For a
334 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 335
rectangular area, the shape of the kem is diamond shape and its dimensions are the middle third
of the area at the centroidal axes. Hence if the eccentricity ex, present alone, is lesser than or
equal to Z / 6 or the eccentricity ey, present alone, is lesser than or equal to B I 6, no tension is
Where ekx = —6^1 6
j
and = —6er1 —
6
developed within the section. I Bet L Ley B
I ...
The values eja and eky are the x and y ordinates of the point of intersection of line from centroid
in the direction of resultant eccentricity and the boundary of the kem of the contact area.
Referring to Fig. 9.12, the contact pressures at any point, such as gi to q^, may be calculated
using the previously presented expression. The most critical bending moment may then be
determined for the end strip. However, the one-way shear may be evaluated considering the
average on one side of the column.

x-axis

Fig. 9.12. Specific Points over Contact Area.

Fig. 9.10. Pressure Distribution Diagrams under Eccentrically Loaded Footing.

Fig. 9.11. Kem for Rectangular Areas.

In case load is applied at simultaneous eccentricities in both the directions, one of the following
conditions is to be satisfied for no tension:
Bending moment for longer steel = ^3 x (Z - ci)2 / 8
6» Clot OR By Cky + (?i -^3)x (Z-q)2/ 12
Bending moment for shorter steel
- ?jx(B-c2)2/8
+ (?i-?j)x(5-c2)2/12
H
336 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
CHAPTERS: DESIGN OF FOOTINGS 337
Maximum one-way shear for flexure = qav x (L / 2 - ci / 2 - d) And b2 same in a direction parallel to the moment vector, mm.
Where, ® (71 + q2 + q& + 77) / 4
~ Yf = 1 L
Top steel must be provided in these footings, if the bending moment acting on the footing
produces uplift on one side, whereas, the weight of soil above the footing provides the required
balancing moment.

9.15 ECCENTRIC PUNCHING SHEAR

If the footing -width perpendicular to the span for moment is lesser than or equal to the column
size in the same dimension plus 1.5 times the footing slab thickness on each side, all the
unbalanced moment is automatically transferred to the column by flexure. However if this
width is greater, the moment created by the outer strips changes into torque for the
perpendicular strips and then is transferred to the column through eccentric shear. The
unbalanced moment is the difference of moment produced at two opposite edges of the
column. For a symmetric footing with column in the center having no external moment,
unbalanced moment will be zero. However, in case external moment is present in the column, Centroidal C
axis
the same will be the unbalanced moment.
Eccentric shear is the two-way shear developed due to transfer of moment from the footing to b) Direct Shear
the column through torsion. According to ACT 11.2.6.2, the shear stress resulting from a) Critical Section For Shear
moment transfer by eccentricity of shear shall be assumed to vary linearly about the centroid of
the critical section. Consider a typical interior slab-column connection where an unbalanced
moment must be transferred between the footing and the column, as shown in Fig. 9.13.
At the connection of footing slab with the interior column subjected to axial load and
unbalanced moment (Afu), yfMa part of moment is transferred by flexure; y, M part of moment
is transferred by eccentric shear and direct applied shear (KJ is transferred to the column as an
axial load. The moment y„ Mu acts as torque on the outer sides of the column perpendicular to c) Eccentric Shear
the moment vector. Figure 9.13 (a) shows the critical section for two-way shear according to
the ACI Code. The expression for perimeter of the critical section (d0) is also shown in this
figure. Figures 9.13 (b), (c) and (d) represent the variation of direct shear stresses due to shear d) Resultant Shear
Ku, linear variation of shear stresses due to twisting moment and the resultant shear stresses, Fig. 9.13. Two-Way Shear Acting on Critical Slab Section
respectively. The direct shear stress (considering the direction of the applied loads) is upwards around Column.
on all the faces, while the shear stress due to twisting moment is upwards on one side and In case the above values of ^and are used for design, the
downwards on the other side having linear variation in-between. The resultant shear stress on in the central band of column size plus 3h for central columns.reinforcement is to be concentrated
one side adds up and becomes critical for the punching of the slab. In case uniform reinforcement
is to be provided, the factor y, may be calculated as follows;
The values of fractions, jy and are calculated by using the expressions presented here.
B-c. -3h C» 3A
Y*
B
and - L
However, one- way shear must not exceed 0.75|yKc for interior and edge
for corner column if the values of and yv are changed columns and O-S^Fc
I from their code values. The resultant
shear stress acting on the critical perimeter, considering
Where b1 = width of the critical section for punching shear perpendicular to the moment directions, may be written as follows: moment acting from both the
vector, mm.
338 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI A CHAPTER 9: DESIGN OF FOOTINGS 339

PL = 350 kN
vu at face AB = —— ±
Jc[ —- ±
^C2 —
-
"4
M^
MLy
=
=
150 kN-m
100 kN-m
1 Live load reduction = 20 %
Where, Ac and Jc are defined below for an interior column and are calculated for the faces of a = 20 MPa
box-like shape defined by the assumed vertical failure section, as shown in Fig. 9.14. The Jy = 420 MPa
values for edge and comer columns may be found in the chapter on column-supported slabs. Depth of footing = 1.5 m
Density of fill = 2100 kg/m3
Ac = perimeter area of the critical section
Density of surcharge = 1900kg/m3
= 'Xb' + bdd Depth of surcharge = 0.5 m
Jc = torsional constant, like polar moment of inertia of the area Ac
A +4 P = Pd + Pl = 500 + 0.8x350 ® 780 kN
b,d3 ^db? । db^ My = Moy +My = 150 + 0.8 x 100 = 230 kN-m
6 6 2
Qua - 300 - (1.5 x 0.5 x 2100 + 0.5 x 1900) x 9.81/1000 = 275.23 kPa
^4 req = {P^-\.5M^\.5My)!q^
= (780 + 0 + 1.5x230) /275.23 = 4.09 m3 = 1.4 x2
(say 1.8 m x 2.3 m)
= = 4.14 m2

_
Af 1.8 x 2.3
_ _ _
h — B?

12 — 1.8x2.33
12
1,^25 m4

- -L+ M .
A, I
780
4.14
+
23OXL15
1.825
>

I? = 2.783 m4
Revised size’. 1.9 x 2.6m, A = 4.94 m2,
Fig. 9.14. Critical Section Perimeter over Interior Column. 780 230x1.3
= 494 + ^78F = 26533 „ row
(°K)
9net
Example 9.3: A 450 mm x 300 mm interior column, reinforced with eight No. 25 bars of
Grade 420, supports a dead load of 500 kN and a live load of 350 kN along with Afpy = 150 ex = Myy IP = — = 0.295 m
kN-m and A/Ly = 100 kN-m. Assume that a live load reduction of 20% may be allowed at the 780
foundation level. The foundation is to be rectangular (having longer to shorter side ratio of = 2.6 / 6 = 0.433 m
L/ 6 No tension is developed.
approximately 1.4) and placed at a depth of 1.5m where the gross allowable bearing capacity is
300 kPa. The average density of the backfill material and the footing concrete together may be Pv = 1.2PD + 1-6Pl = 1.2 x 500 + 1.6 x 0.8 x350 = 1048 kN
taken as 2100 kg/m3, while the filling from ground to plinth level is to be 0.5m of material with M^ = 1.2Afoy + 1.6A£y = 1.2 x 150 + 1.6 x 0.8 x 100 = 308 kN-m
average density of 1900 kg/m3. Design the footing using// = 20 MPa and / for all steel = 420 av Pit I Af
MPa. = 1048/4.94 = 212.15 kPa
Solution:
qa = 300 kPa P+1.5M + 1.53/
Approximate depth of footing, H = + 60 mm 250 mm
Pd = 500 kN o.oia,/;
340 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
DESIGN OF FOOTINGS 341
ri
_
—j
I (780 +1.5 x 230) x 1000 + ou 2ju mm
CHAPTER 9:

} 0.01x40x20 = 9U x {^4f- (ci + d) (c2 + </)}


= 435 mm (say 450 mm) = 212.15 x {4.94 -(0.84) (0.69)} = 925.1 kN
d = H-60 = 390mm
B = 1.9m > c2 + 3h = 0.3 + 3 x0.45 = 1.65m Vu
4
Eccentric shear is required to be considered.
=
925.1x1000 (q.424x308x106)(420)
b0 = 2 x (cj + c2 + 2d) = 2 x (450 + 300 + 2 x 390) = 3060 mm 1 193400 14176890X104
= 450/300=1.5 = 0.775 + 0.387 = 1.162 MPa (Revise)

^>vc is lesser of 0.75x0.33//? , 0.75x0.1711 + 0)|/^ — and 0.75x0.0831/^^ + 2)|//? Revised depth: H = 475 mm
I '
l
<j»vc = lesser of the following: d = H- 60 = 415 mm
1. 0.75x0.33/20 = 1.107 MPa B = 1.9m c2 +3A = 0.3 + 3 x 0.475 = 1.725m
Eccentric shear is required to be considered.
( — = 1.331 MPa
2. 0.75x0.17 1+
I —1-5J
2
/20
j

by = 450+415 = 865 mm
bi = 300+415 = 715 mm
3. 0.75x0.083| 40x390 + 2 |/20 = 1.976 MPa 1
k 3060 J Yv = 1- = 0.423
= 1.107 MPa
3V715
B-c2-3h = 2.8-0.3-3x0.45 A. = 2(bi + bi) d = 2 (865 + 715) x 415 = 1,311,400 mm2
B 2.8 865x415’ 415x865’ 415x715x865’
Using the code formula: 6 6 2
*i = 450 + 390 = 840 mm = 16,607,829 xlO4 mm4
*2 = 300 + 390 = 690 mm Ku = 212.15 x {4.94 - (0.865) (0.715)}
1 = 916 8 kN
= 1- = 0.424
2 /840
=
916.8x1000 (q.423x308x106)(432.5)
Vu
3 V690 1311400 16607829 xlO4
Use larger value, =0.424
= 0,699 + 0.339 = 1.038 MPa < K (OK)

Ac = Kbr + bi) d = 2 (840 + 690) x 390 = 1,193,400 mm2 Ultimate / Factored Contact pressures:
.
J = 6
+ —
bd3 db*1_+ db2b?
6
L_L

2 9> A
+
I.
1048 308xL3
4.94
+
2.783
=
840x390’. _j 390x840’ -l_ 390x690x8402
— 6 6 2
= 14,176,890 xlO4 mm4
92
A r =
4.94— -;
2.783
= 7^2.98 kPa
342 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI t CHAPTER 9: DESIGN OF FOOTI NGS 343

Ultimate / Factored Contact pressures: #VO = 0.75 x 0.17^ bd


The contact pressures are calculated over four different points shown in Fig. 9.15.
= 0.75 xQ.17^2° xl°°-x415 = 236.63 kN > applied shear {OK)
1000
Longer Steel:

bd2
1^80x10^
= 1000x4152 = 1.0614 => p = 0.00295
As = pxbd = 0.00295 x 1000x415 = 1224 mm2 / meter width
-As, min = 0.0018 xbh = 0.0018 x 1000x475 = 855 mm2 / meter width
Selected steel: #16 @160 mm c/c
Shorter Steel:
Fig. 9.15. Points Considered Over Contact Area.
113.93xl06 p= 0.0017
= = 066}5
bd2 1000 x4152
= pxbd = 0.0017 x 1000 x 415 = 706 mm2 / meter width
4,
-l2>
L
= ^+308x(0.45/2)
4.94 2.783
= 237()5kpa
As
As, min = 0.0018x6/2 = 0.0018 x 1000x475 = 855 mm2 / meter width
Central band width: 1900 mm
P. = 1048 = 212.15 kPa P = ratio of long to short side of footing = 1.368
A, 4.94 2
Reinforcement in central band = x total amount of reinforcement
?av for one-way shear = 0.5 x (^ + g2) z?+i
= 0.5 x (356.02 + 282.98) = 319.5 kPa = 0^845 x 855 x2.6/1.9
For 1 m wide strips: = 989 mm2 per meter width
ci= 450mm, C2 = 300mm, B = 1.9m, Z = 2.6m, d = 415mm Selected steel: #13 @ 140 mm c/c (#16 has lesser development length available)

Cantilever length in short direction, = (B-c2)/2


- (1.9 -0.30) /2 = 0.8 m
Reinforcement in outer band = (1
=

/7+1
) x total amount of reinforcement

(1 - 0.845) x 855 x 2.6 / (2.6 - 1.9)


Cantilever length in short direction, tt = (Z-ci)/2 = 492 mm2 per meter width
= (2.6 -0.45) /2 = 1.075 m (for both sides outer band)
Use minimum of 855 mm2 / meter width
Bending moment for longer steel, Mi = 73 x / 2 + (qi q3) x / 3- Selected steel: #13 @ 150mm c/c in each 0.3 5m outer band (3 - #3)
237.05 xl.0752 (356.02-237.05) xl.0752
2 3
- 182.80 kN-m
Whenever depth of footing exceeds 450mm, it is better to provide 0.001bh minimum steel (475
Bending moment for shorter steel, M2

=
= ?ix/s2/2
356.02 x 0.82 _
= 113.93 kN-m
mm2 / m) also on the top.
be provided at the top.
In this example, #10 @ 160 mm c/c in both the directions may also

2
Maximum one-way shear for flexure = q„x (Z / 2 - ci / 2 - d) - Az = 135,000 mm2
= 319.5 x (2.6/2 - 0.45/2 - 0.415) A2 ~ (lesser of 2600 and 450 + 4x415)
= 210.87 kN x (lesser of 1900 and 300 + 4x415) = 4,009,000 mm2
344 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 345
will be provided higher than the footing level. The length of dowels into the footing must not
1 A^ - 5.449 (consider maximum value of 2.0)
be less than 4c for #16 bars that is equal to 365 mm.
= ^x 2 x 0.85 fc' Ai The required development length for flexure of longer footing steel at the face of column is
= 0.65 x 2 x 0.85 x 20 x 135,000 / 1000 = 2983 kN > A (OK) 0.729 m and the available length is (2.6 - 0.45) / 2 = 1.075 m. Similarly, the required
Foundation concrete is capable of resisting the column load even without the development length for flexure of shorter footing steel at the face of column is 0.592 m and the
dowels. available length is (1.90 - 0.30) / 2 = 0.80 m. Hence the steel is fully developed at the critical
sections.
Now we check the dowels required to provide development length for the column steel.
Area of steel required for dowels within the column 9.16 COMBINED FOOTING DESIGN FOR TWO COLUMNS
^4 The combined footing behaves as a longitudinal beam along the length of the footing resting
- -
1048x1000 0.65 x 0.85 x 20 x 450 x 300
= Zero on columns, and the width of beam being equal to the full width of the footing. The transverse
0.65x420 strips shown in Fig. 9.17 expand the support effect over the full width of the footing. The
Minimum area for dowels = 0.005 Az = 0.005 x 135,000 = 675 mm2 width of this strip may be considered equal to column width in the same direction plus lesser
O- of d I 2 distance and the available size on both sides of the column. The salient features of
4 - #16 bars are sufficient as dowel reinforcement However, 8 - #16 bars may also be
design of this type of footing are discussed below:
used for a conservative design.
Development-length in tension for #13 bars,^ = 420 MPa, is:
4 = 0.485
~^=db > 300mm fort4 No. 20

420
= 0.485-^
^20
x 13 = 592 mm

Development length in tension for #16 bars,^ = 420 MPa, is:


420
4 = 0.485 x 16 = 729 mm
V20
Development length in compression for #25 bars, fy = 420 MPa, is:
f
- 0.24 - dL d^ > 200 mm

420
, = 0.24 x x 25 = 564 mm
V20
Development length in compression for #16 bars,jy = 420 MPa, is:
420
4c - 0.04 x x 16 = 361 mm i
V20
Fig. 9.16. Reinforcement Details for Example 9.3.
Compression splice length for #16 bars = 6.071^, F db 300mm
= 0.071 x 420 x 13 x 16 = 621 mm
The reinforcement details are presented in Fig. 9.16. The lap splice length for #25 column bars 1. The centroid of the combined footing should coincide with the location of the resultant
and #16 dowels is to be taken equal to larger of the development length in compression for #25 load on the footing. This makes the resultant moment of the footing transferred to soil
bar (564 mm) and splice length for # 16 bar (621 mm), that is, approximately 0.625 m. This lap equal to zero, which produces uniform pressure under the footing and avoids tilting of
the foundation.
346 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI ER9: DESIGN OF FOOTINGS 347
transverse direction will resist bending in the perpendicular direction due to more
2. Depending upon the clearances available, the footing is to be extended beyond the edge
| effective moment of inertia of the bent footing.
of the column near the boundary. This makes the footing safer against punching shear
and helps in economizing the depth of footing. ! 9. The width of the footing multiplied with the net contact pressure gives the load per unit
length acting on the footing slab in the longitudinal direction acting like a beam. The
shear force and bending moment diagrams may then be plotted to determine the forces
at the critical sections. To plot these diagrams, the column loads may be considered as
point loads acting at their centerlines if the spacing between the columns is sufficiently
large. This will make insignificant difference from the actual results considering the
column load to be uniformly distributed over their width. However, if the column
spacing is lesser, we have to consider the column loads as uniformly distributed load
acting over the width of the columns.
10. It is found that a transverse width of the footing under the columns having a width
Fig. 9.17. Typical Combined Footing with Transverse and Longitudinal Strips. equal to approximately c + d also undergoes transverse bending. Hence, transverse
bending moment is calculated for this strip just like an isolated footing and the
3. The distance between the shorter edge of the footing towards the column closer to the corresponding transverse steel is decided. The load per unit length of the strip is
boundary and the resultant of the load is evaluated. The footing is longitudinally calculated by dividing the total column factored load with the width of the footing.
extended on the other side of the resultant of the load by the same distance. This fixes
the length of the fpoting.
9.17 TYPICAL TYPES OF COMBINED FOOTINGS
4. The required area of the footing is calculated by dividing the total service load on the
footing with the net allowable bearing capacity. This area is then divided by the There are three commonly used types of the combined footings for two columns, as shown in
decided length of the footing to establish the width of the footing. Fig. 9.18. The first type (Fig. 9.18b) has a constant width and is the most popular type. The
5. It is perhaps preferable to first decide the slab depth for one-way shear. dimensions may be decided in the following sequence:
6. The depth is then checked for punching of both the columns.
7. Pattern loading for the live load is not considered for any footing design even if live
load is included in the total load. Reasons are that the live loads are usually a smaller
part of the total load for concrete structures and secondly the effect of patterns loads for Xi = decided earlier
a multi-story building dies out at the foundation level.
L = 2(x+x0

8. A longitudinal section of the footing is then considered, with columns acting as


supports over the full width of the footing, to determine the moments for design of
longitudinal steel and one-way shears. In case the columns are widely spaced and there The second type (Fig. 9.18d) is used when the outer column has more load than the inner
are sufficiently larger cantilever lengths at the two ends, the moments near the columns column. The dimensions may be decided in the following sequence:
produce tension at the bottom fibers, requiring bottom steel, while the moments at the
mid-span will produce tension on the upper side due to upward contact pressure,
requiring top steel. In case the columns are widely spaced but the cantilever lengths at
both sides are only equal to half the column dimensions, the moment everywhere will
be producing tension on the upper side requiring top steel. For the situation when the X] = decided earlier
columns are closely spaced but the cantilever lengths on both sides are excessive, the Li = x + xi
moment everywhere may produce tension on the lower side requiring bottom steel L2 = greater than S - x , decided earlier longer>than or equal to L\. ]£L2 = L\,
throughout footing of the first type is obtained. If L2L\, comes out to be larger than
Bi
It is to be noted that the contact pressure bend the footing slab just like a plane steel B2 and vice versa.
sheet is bend to give corrugations. The column loads help to get these corrugations but L = L\ + L2
need not to be present throughout the length. The deformed pattern in longitudinal or
B. = 2^(21-31,)
CHAPTER 9: DESIGN OF FOOTINGS 349
348 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

Ln = this length is also decided arbitrarily such that the second column lies within
Bz this length
L — Ly + Li
Li = L-Li
R 2L,-L3
Bi
Q„

Bi =
R L4B2
Qn^i L]
Example 9.4: An exterior column of size 300 mm x 450 mm having a dead load of 700
kN and a live load Of 400 kN and an interior column of size 450 mm x 450 mm having a dead
load of 1000 kN and a live load of 650 kN are 5.0m apart center-to-center. Design a
rectangular combined footing only (not the dowels or splice) carried to a depth of 1.5m, where
the gross allowable bearing capacity is 165 kPa. The outer edge of the exterior column is the
property line: The average density of the backfill material and the footing concrete together
may be taken as 2100 kg/m3, while the filling from ground to plinth level is to be 0.75m of
material with average density of 1900 kg/m3. Design the footing using ft = 20 MPa and for
all steel = 280 MPa.
Solution:
Qi = 165 kPa
Pd for exterior column = 700 kN
Pl for exterior column = 400 kN
Pd for interior column = 1000 kN
Pl for interior column - 650 kN
S, spacing of columns - 5.0 m
xi = 0.15 m
Live load reduction = 20 %
A' = 20 MPa
fy = 280 MPa
Depth of footing = 1.5 m
Fig. 9.18. Types of Two-Columns Combined Footings. Density of fill = 2100 kg/m3
Density of surcharge = 1900 kg/m3
Similarly, the third type (Fig. 9.18e) is also used when the outer column has more load than the Depth of surcharge = 0.75 m
inner column. The dimensions may be decided in the following sequence:

x
_ . = P2*S—
Pi for exterior column= Pd + Pl = 700 + 0.8x400 = 1020 kN
Pz for interior column = Pd + Pl = 1000 + 0.8 x 650 = 1520 kN
?nei = 165 - (1.5 x 0.5 x 2100 + 0.75 x 1900) x 9.81/1000 = 135.57 kPa
*1 = decided earlier R = Pi +P2 = 1020+ 1520 = 2540 kN
Zi = x +xj
Li = decided earlier and kept lesser than Zi x P2*S 1520x5
= 2.992 m
R 2540
CHAPTER 9: DESIGN OF FOOTINGS
350 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI 351

0,75x0.17
A ,„V20 x3000xtZ - 1057 => r7 = 618 mm
= 0.15 m
L = 2(x +xj) = 2(2.992 + 0.15) = 6.284 m (say 6.3 m)
:Letrevised77 = 700 mm
d = H — 60 = 640 nun
B = —q„L 135.57x6.284
= 2.981 m (say 3.0 m) Maximum one-way shear for at (0.225+0.64) = 0.865 m just left of the right side load
Af = 6.3 x 3.0 = 18.9 m2 = 1414.27 - 0.865 x 537.4 = 949.42 kN
Ai for exterior column = 1.2Pd+1.6Pl
= 1.2 x 700+ 1.6 x 0.8 x 400 = 1352 kN
P^2 for interior column = 1.2Pd+1.6Pl
= 1.2 x 1000 + 1.6 x 0.8 x 650 = 2032 kN
qa ~ (A + A) /A (a) Loading
= = 179.05 kPa
(1352 + 2032)/ 18.9
ttttt + t + 'S + tTT 537.14 kN/m
Bending Moments And Shears Along Longer Direction
The column -loads, may be considered as uniformly distributed loads spread over the
dimension of the column in those cases where spans are smaller and greater accuracy in
calculations is required. As an approximation for longer spans, the column loads may
be considered as point loads acting at the centerlines of the columns with only a minor
difference in result. The load along the length of the footing per unit length will be qu
x B = 179.05 x 3.0 = 537.15 kN/m. The resulting shear force and bending moment
diagrams are shown in Fig. 9.19.
Trial Depth Of Footing 1498.72

P + 1.5M + 1.5A7
Trial depth of footing, H =
0.01aj; —
+60 mm 250 mm
Fig. 9.19. Shear Force and Bending Moment Diagrams
along Length of Combined Footing.

1020x1000 Check For Punching Shear


77 for exterior column = J — +60
y 0.01x30x20
250 mm
For Exterior Column:
= 473 mm b0 = 2 x (ci + d 1 2) + (ci + d)
= 2 x (300 + 640 / 2) + (450 + 640) = 2330 mm
. .
7/ for interior column = J
1520x1000
V 0.01x40x20 — + 60 250 mm
= 450/300 = 1.5
,0.75x0.17| 1 +— ]jf!
\^
= 496 mm is lesser of 0.75 x 0.33Jy; and 0.75x0.083| + 2 iff7
First trial depth, H~ 500 mm, d = 440 mm I fl) bo /
c

Depth For One-Way Shear H = lesser of the following:


Maximum one-way shear for at (0.225+0.44) 0.665 m just left of the right side load 1. 0.75x0.33720 = I.107MPa
= 1414.27 - 0.665 x 537.14 = 1057 kN
0.75x0.171
#VC = 0.75 x 0.1iff) bd = A =>
2.
1+—JV20 = 1.330 MPa
352 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
CHAPTER 9: DESIGN OF FOOTINGS 353
Design Of Longer Bottom Steel Under Interior Column
3. 0.75x0.0831^^
k 2330
+ 2 k/20 = 2.851
)
MPa 1
= 1.107 MPa |
— 237.76 kN-m/m
K 237.76 xlO6
0FC =(KM = 1.107x2330x640/ 1000 = 1650.8 kN bd1 3000 x6402
= 0.1935 P Poon

Fu = Column load - x { (ci + d 1 2) (c2 + d)}


= 1352- 179.05 x (0.62 x 1.09) = 1231 kN
4 As -
0.002 x bh = 0.002 x 3000 x 700 4200 mm2
Selected steel: 11 - #22 (spacing = 272 mm c/c)
-
K < (OK) j Development length in tension for #22 bars, f, = 280 MPa, is:
For Interior Column: ? > 300mm for db > No. 20
b0 = 2 x (ci + C2 + 2d) 1
= 2 x (450 + 450 + 2 x 640) = 4360 mm | 280
= 0.606-7= x 22 = 835 mm (say0.S4m)
/7 = 450 / 450= 1.0 1 V20
(|)vc = lesser of the following: Available length on cantilever side = 1150-450/2
1. 0.75 x 0.33^20
= 1.107 MPa 4 = 925 mm > (OK)

——
'

J Distance of inflection point on inner side from face of the column plus larger of d and
f 2 4k/20 '4
2. 0.75x0.17 1+
k 1.0 J
I
= 1.710 MPa 3 —
12 di, = 271 225 + larger of 640 and 12 x 22 = 686 mm (say 0.7 m).

3. 0.75x0.083
f 40 x 640

k 4360
A
+2 M20
)
i
—= 2.191 MPa
1
a4
Design Of Longer Bottom Steel Under Exterior Column
Moment is almost negligible; provide some steel only for additional safety. For
example, the same steel as for the interior column may be provided.
= 1.107 MPa
Selected steel: 11 - #22 (spacing = 272 mm c/c)
4»^
Fu
= 4>vc bod = 1.107x4360x640/ 1000 = 3089.0 kN
- Column load - qa x { (ci + d) (c? + d)}
| Provide standard 90° hooks on the column side to get the full development length at the
interior face of the column.
= 2032- 179.05 x (1.09 x 1.09) = 1819 kN < Design Of Transverse Beam Under Exterior Column
Fu < *KC (OK)
Approximate width of beam = ci + d!2
Design Of Longer Top Steel = 300 + 640/2 = 620 mm
Bending moment for longer steel, Afui = 1498.72 kN-m/m Load on transverse beam = P*\IB = 1352/3.0 = 450.67 kN/m
ll^KT^xlO- = 450.67 xl.2752
= j 220 0.0051 Bending moment = 366.31 kN-m/m
bd2 3000x6402 2

= pxbd = 0.0051 x 3000x640 = 9792 mm2 366.31x10®


As
bd1 620x6152
= 1.5621 p = 0.0067
v4s min = 0.002 x bh = 0.002 x 3000 x 700 = 4200 mm2
Selected steel: -
20 #25 (spacing = 150 mm c/c)
4 = pxbd — 0.0067 x 620 x 615 = 2555 mm2
Curtailing half of the bars may be allowed near the columns.
A^ mb, = 0.002 x bh = 0.002 x 620 x 700 = 868 mm2
Selected steel: 5 - #25 (spacing = 136 mm c/c)
35 4 CONCRETE STRUCTURES PART - I AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 355

Design Of Transverse Beam Under Interior Column

Approximate width of beam


= 450 + 640 = 1090 mm
Load on transverse beam = P^/B = 2032/3.0 = 677.33 kN/m
677.33 x 1.275*
Bending moment, = = 550.55 kN-m / m
2
550.55 xlO6
= 1.3354 p = 0.0057
bd 1090 x6152
A, = px bd = 0.0057 x 1090 x 615 = 3821 mm*
As, min = 0.002 xbh = 0.002 x 1090 x 700 = 1526 mm*
Selected steel: 8 - #25 (spacing = 152 mm c/c)
Distribution Steel
As, min = 0.0O2 x bh = 0.002 x 1000x700 = 1400 mm2 /m
Selected steel: #19 @ 200 mm c/c 6300

Check For Development Length


Fig. 9.20. Main and Distribution Steels for Combined Footing of Example 9.4.
Development length in tension for #25 bottom bars, = 280 MPa, is:
Solution:
> 300mm fords > No. 20
/d =
0.606—^dh 280
(a) = 2.992 m as in Example 9.4.
= 0.15 m
= 0.606S= x25 = 949 mm < 1275mm Ly = x +xi = 3.142 m
V20
The transverse steel is fully developed. . Li Li (say 3.858 m)
One Face Minimum Steel
L = Lx+L2 = 3.142 + 3.858 = 7.0m
This steel is provided on all faces in both the directions where there is no other steel By = _^(2jr-3L1)
and the total depth of the footing is greater than 450mm.
A,. mm = 0.001 x bh = 0.001 x 1000 x 700 = 700 mm2 /m = „2x2540 (2x7_3x3 142) = 3 50m
Selected steel: #13 @180 mm c/c 135.57x72
1R
Sketch Of Reinforcement Bi - R

The reinforcement details for the main and the distribution steel are shown in Fig. 9.20.
An extra steel equal to #13 @ 150 mm c/c is to be provided on all other faces in both 2x2540
= 3.50 = 1.85 m
the directions having proper overlap with the other reinforcement. 135.57x7
Examtile 9.5: For the combined footing of Example 9.4, select the size of the foundation (b) = 2.992 m, xi = 0.15 m, L\ - 3.142 m
for the following two types: Let Li = 2.5 m and Lt = 4.5 m
a) Gradually slopping in plan. = Li + L4 = 7.0 m
b) Having sudden change in width. Lj = L-Ly = 3.858 m
II
J CHAPTER 9: DESIGN OF FOOTINGS 357
356 CONCRETE STRUCTURES PART I - AUTHOR: ZAHID A. SIDDIQI
I are drawn and values of these quantities are calculated at the critical sections. This case
52
R ^21,-^ j of continuous beam is determinate, as the factored column loads are already known.
L*L<
2540 2x3.142-2.5 | 7. If the bending moments do not change in sagging - hogging sense along the length of
X = 2.25 m the beam at mid-span and column sections, this type of footing must seriously be
135.57 7.0 x 4.5 5 analyzed for its applicability in terms of expected differential settlement and non-
R ? uniform bearing pressure.
51
^3 ; 8. The minimum width of the foundation beam is to be taken equal to column width plus
2540 4.5x2.25 f 75 mm on each side.
135.57x2.5 2.5 9. The depth of a foundation beam usually lies between t 1 10 and 1 1 8, as compared with /
J
• / 12 for beams of super-structure due to the presence of heavy loads.
9.18 FOUNDATION BEAM FOOTING
e 10. The beams may either be designed as rectangular sections as a conservative approach or
In this type of footing, a slab strip is provided at the bottom for more than one columns to resist • the mid-span sections may be considered as T-beams.
the contact pressure, which in turn is supported by a beam running along the column centerline.
In most of the cases, where upturn beams may be accommodated in the footings, this becomes Example 9.6: Design a combined footing using foundation beam for the arrangement of
an economical option. The following points are important to be remembered for the design of
foundation beam footings:
columns shown in Fig. 9.21. -
20 MPa, = 300 MPa and net allowable bearing capacity is
equal to 80 kPa. The size of all columns is 300 x 300 mm.
1. The point of application of the resultant load (5) must coincide with the centroid of the Pi P2 Pi Pi Pi
contact area. The length of footing (L) is decided based on this requirement and the , 4.75m , 4.75m , 4.75m , 4.75m
available dimensions.
~
I
2. Width of footing (B) is then decided as follows:

B = —
I
>1 L= 19.3m — 1

Fig. 9.21. Length and Loading for Footing of Example 9.6.


3. The first requirement may only be satisfied for more or less symmetric arrangement of
columns and the corresponding loads. When there is no cantilever length beyond the
exterior load, this load must be lesser than half of the interior load (~ 5/12 times the L = 19.3 m
interior load). For the case where the exterior and interior loads are almost equal, the = 80 kPa
cantilever length is taken approximately equal to 0.4 times the spacing of the columns. Pi, Pi = 300 kN
Where the exterior load is half of the interior load, the cantilever length may be kept Pi, Pi, Pi = 650 kN
close to 8 % of the column spacing. Similarly, when there is no cantilever length and Pul, Pul = 400 kN
outer column has a load equal to 60% of the inner column load, the outer column must Pu2, Pul, Pui = 875 kN
have a moment approximately equal to (0.15 to 0.18) x inner column load x column 2.P = 2550 kN
spacing, acting towards the inner side. The service inner column load is equal to
service UDL x column spacing parallel to footing x column spacing perpendicular to
footing.
4. The soil pressure may approximately be assumed to be uniform.
%Pu
ft
. _ req
= 3425 kN
= 20 MPa
_
— 2550
__
and fy =300 MPa
_ 31»o75 ID2
80
5. Only one-way shears in the beam and in the overhanging slab are to be satisfied and
there are no chances of two-way shear failure.
Width of footing (5) = = 1.651m (say 1.7 m)
6. The longitudinal section of the beam is then considered with uniformly distributed load
equal to qBfl x B acting over the full length. Shear force and bending moment diagrams
358 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 359

-4f,prov = BxL = 32.81 m2


3 M = 242.88 kN-m, bd2 = 1.9582 MPa, p = 0.0080, As = 1890 mm2
= 21?! = 104.39 kPa [Use 4 - #25]
X™ 32.81
=
= 0.75 x 0.17^20 x 450 x 525 /1000 134.71 kN
Load per unit length of beam, wu = q^ x B = 177.46 kN/m
The shear force and bending moment diagrams for the foundation beam are shown in Using 2-legged #13 stirrups, Av = 258 mm2
Fig. 9.22. Smax = least of the following four values:

400 kN 875 kN 875 kN 875 kN 400 kN i)2


A.v^L_ = 25lx321 = 491 mm
0.15m 0.35bu 0.35x450
ii) d! 2 = 262.5 mm
iii) 600 mm
_
. ^f^d = 0.75x258x300x525
t\7l
2
VU~WC
.

(349.77 -134.71)x 1000
1“1 Unix

Uses - 140mmc/c
Shear force at a distance of / 5 from edge of support (x = 1.1 m) is:
= 469.55 - 177.46 x 1.1 = 274.35 kN
^A^f^d 0.75x142x300x525
Smaxfor 2-legged #10 stirrups - “

^77^7 (274.35 -134.71)x 1000


Uses = 120mmc/c
Slab Design
_
Fig. 9.22. Length and Loading for Footing of Example 9.6.
,
Cantilever length, I = —— B-b 1.7-0.45
- 0.625 m

For estimation the trial depth of the footing slab, the one-way shear for some unknown
Let, Width of beam, b - 300 + 75 x 2 =
450 mm
effective depth may be equated to the applied Shear at the edge of the beam (giving
Depth of beam, h = £/8 = 4750/8 = 594 mm (say 600 mm)
some margin of safety) with an additional 20% margin.
d = h-15 = 525 mm
0.8 x = qa x
I 390.81x10s 0.8 x 0.75 0.17720 x 1000 x dl 1000 = 104.39 x 0.625
t/min for singly reinforced beams = 460 mm x
0.205x450x20 d = 143 mm, = 143 + 75 = 218 mm (say 250 mm)
Design shear force, Kn 469.55 - 177.46 x (0.15 + 0.252) = 349.77 kN
= d = 175 mm
For design, consider all sections to be rectangular.
= 104.39x0.625^ = 2QJ9 m
1.4 i 2
Ann = — = 0.00467
= 390.81 kN-m, —bd = 3.1509 MPa, p = 0.0135, As = 3190 mm2 bd2
As
1000xl752
= 0.0026 x 1000 x 175 = 455 mm2/m
[Use 7- #25]
360 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTER 9: DESIGN OF FOOTINGS 361
has weak pockets (there are chances of differential settlement) and the isolated footings cover
— 0.002x1000x250 = 500mm2/m more than 50 percent of the area. The mat may be supported by piles in case of higher ground
Use #13 @ 250 mm c/c. water table, weak soils or heavy loads. In case of many basements under a building, the weight
of the soil removed may be equal to the load of the building applied on the raft, and the system
for #13 = 0.485 x x 13 / 1000 = 0.423 m < 0.625 m (OK) may “float” on the soil surface.

The reinforcement details are shown in Fig. 9.23.


Total factored resultant load and factored moment acting on the raft centroid is calculated.
Similarly, if moments are acting at the bottom of columns and walls, their resultant moment is
#13, 2-legged @ 140 min also calculated at the raft centroid. It is tried to adjust the raft dimensions such that its centroid
4/10 c/c within 4|/5 distance coincides with the point of application of the resultant with no moment. However, if it is not
4-#25 4/5
possible both load and moment are evaluated at the raft centroid. The upward pressure on the
1 4-#25 + \ ,2 — #25 + raft slab is calculated by dividing the resultant load on the raft by its total contact area.
\2-#25 + 2-#19 : /2-#25
Additional upward pressure is also calculated on one side or on the comer due to moment in
#10, 2-legged
@ 120 mm c/c
_ one direction or in both directions, respectively. The combined pressure may conservatively be
taken as the applied load on the raft for its design. It is to be remembered that the self weight
of the slab is not included in the applied load as it is directly balanced by the soil.
2-#19 + 2- #25 2 + 2-#25 2-#25 :2 + 2-#25 After calculation of the loads, the raft slab may be considered to be upturned and may be
2-#25 :
JOX designed like a regular slab. Sometimes beams projecting upwards from the raft slab are
4/3 4/3 provided along the column center-lines or between the columns as secondary beams. The loads
4/4 Longitudinal Section of Foundation Beam acting on these beams are calculated just like superstructure slabs. While detailing the
reinforcement, it must be kept in mind that the load is acting upwards and the tension will be
produced at the critical sections accordingly. In case no beams are provided, it is better to
design the slab on computer by modeling the soil as area springs applied below the footing.
The depth of footing is selected such that no punching failure is expected for any column. The
foundation supported by independent or uncoupled springs is termed as Winkler foundation,
where the deflection of any spring is not dependent on the adjacent springs. Increasing the
number of springs near the edges of the mat can approximately model the coupled effect of
springs that is closer to the actual situation. The spring constant or the modulus of sub-grade
reaction (ks in kN/m3 units) for central portions is calculated by dividing the allowable bearing
capacity by the allowable settlement (usually 50 mm for mats). To account for the long term
consolidation settlement, this value may be multiplied with the ratio of the settlement due to
load and the total settlement including the consolidation settlement.

Fig. 9.23. Detailing for Footing of Example 9.6.

9.19 RAFT FOOTING

A raft or mat is a two-dimensional combined footing that extends over the entire area under a
structure and which transfers all the loads from walls and columns to the soil underneath. It is
used when most isolated footings of the columns tend to overlap with each other or if the soil
362 -
CONCRETE STRUCTURES PART 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER?: DESIGN OF FOOTINGS 363
Space For Reader’s Notes
PROBLEMS

Note: N = Registered number of the student


No = A fixed positive or negative number assigned by
the instructor to get R between 600 and 1000.
= N - .No, a parameter to get numerical data for the
problems.

1- A R/2 mm (rounded to nearest 75 mm multiple) square interior column, reinforced with


2 percent of Grade 420 steel, supports a dead load of 0.77? kN and a live load of 0.47?
kN. Assume that a live load reduction of 30% may be allowed at the foundation level.
The foundation is to be placed at a depth of Im where the allowable bearing capacity is
77/6 kPa. The average density of the backfill material and the footing concrete together
may be taken as 2100 kg/m3, while the filling from ground to plinth level is to be 0.5m
of material with average density of 1800 kg/m3. Design a square footing using/c' = 20
MPa, fy for foundation steel = 300 MPa and fy for dowel steel - 420 MPa.

2- A R/2 mm (rounded to nearest 75 mm multiple) x 300 mm interior column, reinforced


with 2 percent of Grade 420 steel, supports a dead load of 0.77? kN and a live load of
0.47? kN. Assume that a live load reduction of 25% may be allowed at the foundation
level. The foundation is to be rectangular (having longer to shorter side ratio of
approximately 1.2) and placed at a depth of Im where the allowable bearing capacity is
77/4 kPa. The average density of the backfill material and the footing concrete together
may be taken as 2100 kg/m3, while the filling from ground to plinth level is to be 0.7m
of material with average density of 1900 kg/m3. Design the footing using/' = 25 MPa
and fy for all steel = 420 MPa.

3- A R/2 mm (rounded to nearest 75 mm multiple) x 300 mm interior column, reinforced


with 2 percent of Grade 420 steel, supports a dead load of 0.47? kN kN and a live load
of 0.257? kN kN along with Mr>y = R/5 kN-m and MLy = R/7 kN-m. Assume that a live
load reduction of 20% may be allowed at the foundation level. The foundation is to be
rectangular (having longer to shorter side ratio of approximately 1.4) and placed at a
depth of 1.5m where the allowable bearing capacity is 77/4 kPa. The average density of
the backfill material and the footing concrete together may be taken as 2100 kg/m3,
while the filling from ground to plinth level is to be 0.5m of material with average
density of 1900 kg/m3. Design the footing using// = 20 MPa and fy for all steel = 420
MPa.

4- An exterior column of size R/2 mm (rounded to nearest 75 mm multiple) x 300 mm


having a dead load of 0.77? kN and a live load of 0.47? kN and an interior column of size
450 mm x 450 mm having a dead load of 7? kN and a live load of 0.67? kN are 77/150 m
apart center-to-center. Design a rectangular combined footing only (not the dowels or
364 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
splice) carried to a depth of 1.5m, where the gross allowable bearing capacity is RJ^
kPa. The outer edge of the exterior column is the property line. The average density of Chapter -10
the backfill material and the footing concrete together may be taken as 2100 kg/m3,
while the filling from ground to plinth level is to be 0.75m of material with average LATERAL LOADS AND MOMENT ENVELOPES
density of 1900 kg/m3. Design the footing usings' = 20 MPa and fy for all steel = 300
MPa.
10.1 LOADS
5- For the combined footing of Q. 9.4, design the foundation of the following two types:
The basic types of loads were discussed in the first chapter and lateral loads and their
a) Gradually slopping in plan. estimation will be discussed here. The load that almost retains its magnitude and point of
b) Having sudden change in width. application throughout the life of the structure is called Dead Load (denoted by D). This load
is usually the self-weight of the structure and the permanent fixtures, which is estimated by
multiplying volume of a member with the unit weight of the material of construction. This
6- Design a combined footing using foundation beam for the arrangement of columns load constitutes a larger part of the total load on a structure. The load due to persons occupying
shown in the figure. 7c' = 20 MPa, = 300 MPa and net allowable bearing capacity is the structure and their belongings is called Live Load (denoted by L) whose magnitude and
equal to R/7 kPa. The size of all columns is R11 mm square (rounded to nearest 75 mm point of application changes with time. Other types of loads are Wind Load (W), Earthquake
multiple). Load (E), Retained Water Load, Retained Soil Load, and Thermal Load due to temperature
variations,

10.2 RAIN AND SNOW LOADS


For rain load, maximum ponding during the rain is observed and it is multiplied with the
density of water to get the rain load per unit area. The snow load is calculated by taking
ground snow load from the maps of the area and multiplying this with some coefficients.
1 Snow pressure = Coefficients x Pg (Ground snow load from Maps)
Figure. Length and Loading for Footing of Q. 9.6.
These coefficients depend upon the following factors:
Flat vs. sloped roofs
= R/3 kN Drifting of snow
Pt, Pa, = 0.82? kN Heated or unheated building
= 2?/2kN Simultaneous wind speed
Py2, P u3j Pu4 = 1.12? kN
= 25 MPa and fy =280 MPa The snow load on slopping roof becomes less as some of the snow drifts down, as is clear in
Fig. 10.1. However, if this snow accumulates on some other part of the building, the load there
may become quite excessive. Heated building may attract less snow load, as there are chances
that the bottom part of snow melts and drifts down. The presence of simultaneous heavy winds
reduces the snow load.

10.3 WINDLOAD
Wind load is produced due to change in momentum of an air current striking the surface of a
building. A building is less likely to experience the other design loads in its life but it is almost
certain that the building is likely to be subjected to the design wind loads. If the building is
very tall, the wind velocity varies along the height and sophisticated codes account for this
effect. Basic wind speed is defined as the fastest wind speed in km/hr having a probability of
occurrence of 0.02 and measured at a point 10m high above the ground under exposure
LATERAL LOADS AND
CONCRETE STRUCTURES PART - 1 CHAPTER 10: MOMENT ENVELOPS 367
366 AUTHOR: ZAHID A. SIDDIQI
10.4 UBC-97 WIND LOADS
categoiy-C conditions, defined later. The factors affecting the wind pressure in addition to the
basic wind velocity are: According to UBC 1621.2 and 1621.3, there are two methods for the calculation of wind

Accumulation

pressure. Method 1 Normal Force Method may be used for any structure but is the only
| method for the design of gabled rigid frames. In this method, the wind pressures are
snow simultaneously applied normal to all exterior surfaces. Hence, the wind loads will be inclined
for parts of the structure at an angle to horizontal and vertical planes. Method 2 - Projected
Area Method has two restrictions: Firstly, it is not applicable to gable rigid frames. Secondly,
it is not applicable if the height of the structure is more than 60m. In this method, only
Fig. 10.1. Snow Load on Slopping Roof. horizontal and vertical pressures are applied. The horizontal pressure is applied over the full
vertical projected area of the structure and vertical pressure is applied over the full horizontal
a) Gradient of wind velocity with height above ground. projected area.
b) Local variations of pressure due to vortices. The pressure is the highest at the
comers, relatively high at the edges and low at the center of the building, as
shown in Fig. 10.2. This affects cladding or minor element design. Wind Windward „
. . .... Leeward
Building .,
side side

Highest' (a) Windward and leeward sides.

Lowest
Higher

Fig. 10.2. Local Variation of wind Pressure. wind

c) Exposure of the structure. For example, the coastal areas will have more wind
loads. Buildings surrounded in other tall buildings will experience less wind
pressures. The wind can be just a gust of wind or long wind periods. There are
(b) Internal pressure. (c) Internal suction.
three Exposure Categories defined in the UBC-97 Code. Exposure B has
terrain with buildings, forest or surface irregularities, covering at least 20 per
cent of the ground level area and extending 1.6 km or more from the site. Fig. 10.3. Various Types of Negative Wind Pressures.
Exposure C has terrain that is flat and generally open, extending 0.8 km or more
from the site in any full quadrant. Exposure D is the most severe exposure in The minimum wind load should be 480 Pa acting over the projected area of the building
areas with basic wind speeds of 129km/hr or greater and has terrain that is flat normal to wind, normal to surface in either direction for components and cladding or area At
and unobstructed facing large bodies of water over 1.6 km in width relative to for open buildings. A building is considered open if each wall is at least 80 percent open.
any quadrant of the building site. Exposure D extends towards the land from
the shoreline 0.4 kin or 10 times the building height, whichever is greater. Wind load, P = qsACDIw
d) Internal pressure. When the wind enters the building from the windward side gs -= wind stagnation pressure
and the leeward side is relatively closed, internal pressure is developed that acts 'A pF2
like negative pressure or pressure acting away from the structure, as shown in * 0.0475 P2
Fig. 10.3. Similarly, when high-speed wind passes by a building, it produces a P = air density
vacuum on the leeward side. This vacuum results in internal suction producing K = basic wind speed (km/hr)
negative pressure fro the structure. A = effective exposed area
LATERAL LOADS AND
MOMENT ENVE LOPS 369
CHAPTER JO;
- 10.5, the overturning of structure
368 CONCRETE STRUCTURES PART 1 AUTHOR ZAHID A. SIDDIQI 3 dead load resisting moment. Referring to Fig.
about point A may be investigated as follows:
Cd = drag coefficient
= Ce X Cq
Table 10.2. Pressure Coefficient (C9).
Ce ~ combined height, exposure and gust factor coefficient (Table 16- ?|
G of UBC) I Part of Structure Angle CB
Cq - pressure or shape factor coefficient for the structure or its portion "Windward roof 0° to 9.5° 0.7 outward
under consideration (Table 16-H of UBC) 9.5° to 37.0° 0.9 outward or 0.3 inward,
Iw = importance factor (Table 16-K of UBC) which ever is more critical
= 1.15 for essential and hazardous facilities like hospitals, fire and 37° to 45° 0.4 inward
police stations, disaster centers and water tanks for fire-fighting. >45° 0.7 inward
= 1.0 for all other buildings. Leeward or flat roof 0.7 outward
0,8 inward
10.4.1 Combined Height, Exposure And Gust Factor (Ce)
Windward walls ,

Leeward walls
The values of this coefficient are given in Table 10.1. Chimneys, tanks and solid
towers
Table 10.1. Values of Coefficient (Ce). square or rectangular 1.4 any direction
hexagonal or octagonal 1.1 any direction
Height Above Average Level of round or elliptical 0.8 any direction
Adjoining Ground (m) Exposure C Exposure B Signs, flagpoles, light poles
UBC Approximate 1.4 any direction
or minor structures
0-4.57 0-4.5 1.06 0.62 Roof eaves without <9.5° 2.3 upward
6.10 6 1.13 0.67 overhangs 9.5° to 30° 2.6 upward
7.62 7.5 1.19 0.72 30° 1.6 upward
9.14 9 1.23 0.76 Overhangs at roof eaves and 0.5 added to above values
12.19 12 1.31 0.84 canopies
18.29 18 1.43 0.95
24.38 24 1.53 1.04
30.48 30 1.61 1.13
36.58 36 1.67 1.20
48.77 50 1.79 1.31
60.96 60 1.87 1.42
91.44 90 2.05 1.63
121.92 120 2.19 1.80

10.4.2 Pressure Coefficient (Cq)


The values of this coefficient for various parts of the building are given in Table 10.2
and some typical values are reproduced in Fig. 10.4.
Elevation
10.4.3 Other UBC Wind Load Provisions Fig. 10.4. Typical Values of Pressure Coefficient
1. Wind should be assumed to come from any horizontal direction. No reduction in
overturning moment due to wind
wind pressure shall be taken for the shielding effect of adjacent structures. Mv =
2. The base overturning moment for the entire structure, or for any one of its
= H* a
individual primary lateral resisting elements, should not exceed two thirds of the
LATERAL LOADS AND
370 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI
CHAPTER 10: MOMENT ENVE LOPS 371
Mm - resisting moment due to DL for overturning
= Wxb/2 Example 10.2: Determine the wind forces for an office building in downtown area with
Mw < F.O.S. = 1.5 or 3/2 against overturning
exposure type B. The basic wind speed for the area from charts is 160 km/h. The bay length is
5m and the frames, shown in Fig. 10.7, are at 7.5m at centers perpendicular to the plane of the
frames.

6 @ 3.4m
= 20.4m

Fig. 10.5. Safety against Overturning.

where, H = resultant horizontal wind load


a = height of resultant 77 from the base
Fig. 10.7. Frame for Example 10.2.
W = resultant of dead loads of the structure
and b = width of the structure Solution:
For ah entire structure with a height-to width ratio of 0.5 or less in the wind For office building: Iv = 1.0
direction and a maximum height of 18m, the combination of the effects of uplift Exposure category: B, as the building is downtown
and overturning may be reduced by one third. Basic wind speed: 160 km/h
3. For pressures on roofs and leeward Walls, Ce is evaluated at the mean roof height. = 0.0475 V2
4. For outward acting pressures on roofs and leeward walls, Ce is evaluated at the = 0.0475(160)2 = 1216 Pa
mean roof height and it is applied for the entire height of the structure for the Cg = - 0.7 outward for the roof
leeward walls. Cq = 0.8 inward for the windward wall
5. A building structure or story is considered open when 15% or more of the area of Cg = - 0.5 outward for the leeward wall
the exterior wall on any one side is open (doors and windows, etc.) The values of coefficients Ce and Cq at locations other than roof are given in Table
10.3.
Example 10.1: Determine wind load on the windward and the leeward side, as well as on
the roof, if the wind pressure is 700 Cg (MPa). Table 10.3. The Coefficients Ce and C, at Various Heights.
Solution:
The solution is given in Fig. 10.6. ca
A(m) Average Ce Windward Leeward
Wall Wall
0-4.5 0.62 0.80. 0-50
4.5 -6.0 0.65 0.80 0.50
6.0 -7.5 0.70 0.80 0.50
7.5 -9.0 0.74 0.80 0.50
9.0-12.0 0.80 0.80 0.50
12.0-18.0 0.90 0.80 0.50
18.0-20.4 0.97 0.80 0.50
20.4 0.99 0.80 0.50
Fig. 10.6. Wind Load Calculation.
The pressures are calculated in Table 10.4 and are shown in Fig. 10. 8.
LATERAL LOADS AND
372 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTER 10: MOMENT ENVELOPS 373
statically applied forces. Dynamic forces are those that are very much time dependent and
these either act for small interval of time or quickly change in magnitude or direction.
Table 10.4. Calculation of Wind Pressures. Earthquake forces, machinery vibrations and blast loadings are examples of dynamic forces.
Pressure Calculation Magnitude Structural response is the deformation behavior of a structure associated with a particular
loading. Similarly, dynamic response is the deformation pattern related with the application of
Pl (0.62)(0.8)(1216)(1.0) 603
(0.65)(0.8)(1216)(1.0) 632 dynamic forces. In case of dynamic load, response of the structure is also time-dependent and
P2
hence varies with time. Dynamic response is usually measured in terms of deformations
P3 (0.70)(0.8)(1216)(1.0) 681
(displacements or rotations), velocity and acceleration.
P4 (0.74)(0.8)(1216)(1.0) 720
P5 (0.80)(0.8)(1216)(1.0) 778 Table 10.5. Calculation of Wind Loads.
P6 (0.90)(0.8)(1216)(1.0) 876
P7 (0.97)(0.8)(1216)(1.0) 944 Force Calculation Magnitude
P8 (0.99)(-0.5)(1216)(1.0) -602 Fl (2.8)(7.5)(603) + (0.6)(7.5)(632) + 30.86 kN
P9 (0.99)( -0.7)(1216)(1.0) -843 (3.4X7.5X602)
F2 (0.9)(7.5)(632) + (1.5)(7.5)(681) + 32.68 kN
(1.0)(7.5)(720) + (3.4)(7.5)(602)
F3 (0.5)(7.5)(720) + (2.9)(7.5)(778) + 34.97 kN
(3.4X7.5X602)
F4 (0.1)(7.5)(778) + (3.3)(7.5)(876) + 37.62 kN
(3.4X7.5X602)
F5 (2.7)(7.5X876) + (0.7)(7.5)(944) + 38.05 kN
(3.4)(7.5)(602)
F6 (1,7)(7.5X944) + (1.7X7.5X602) 19.71 kN
Roof (7.5X843) 6.32 kN/m
Load

Fig. 10.8. Wind Pressures Acting on Structure of Example 1 0.2.


The lateral loads acting at the story levels are calculated in Table 10.5 and are shown in
Fig. 10.9. Fig. 10.9. Final Lateral Wind Loads for Example 10.2.

10.5 DYNAMIC LOADS As already stated, dynamic force, F[f), is defined as a force that changes in magnitude,
direction or sense in much lesser time interval or it has continuous variation with time, as
There are two types of forces/loads that may act on structures, namely static and dynamic shown in Fig. 10.10. Impact load is the other extreme where the load is applied only for an
forces. Static forces are those that are gradually applied and remain in place for longer infinitesimal interval of time with some momentum and is considered separate from the
dynamic loads; The variation of a dynamic force with time is called history of loading.
duration of time. These forces are either not dependent on time or have less dependence on
time. Live load acting on a structure is considered as a static load because it usually varies Prescribed dynamic loading is regularly varying loading in which well-defined cycles of
loading are repeated after equal intervals of time. Example of prescribed loading is a regular
gradually in magnitude and position. Similarly moving loads may also be considered as
LATERAL LOADS AND
374 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
CHAPTER 10: MOMENT ENVE LOPS 375
vibration of machinery with a certain amplitude and frequency. Amplitude of vibration is the
maximum structural displacement during one complete cycle of load and frequency is the ii) Complex Variation Loading: The variation of this type of loading with respect to
number of loading cycles in a unit time (usually one second). Dynamic response of a structure time is shown in Fig. 10.13. By carrying out Fourier series analysis, complex variation
may be estimated in two different ways: may be changed in to a sinusoidal variation.

Fig. 10.10. A Typical Dynamic Force.

Deterministic Estimate Of Dynamic Response


In case of deterministic analysis for dynamic response, time variation of loading is fully
known whether in case of prescribed oscillatory motion or in case of already recorded
earthquake. The response to such dynamic force may be determined exactly.

Non-Deterministic Estimate Of Dynamic Response


The analysis for random dynamic loading to estimate the structural response is called 3
non-deterministic estimation of dynamic response. Random dynamic loading is a 3
loading in which the exact variation of force with time is not frilly known but can only J
be approximately defined in a statistical way with some probability of occurrence.

10.5.1 Types Of Prescribed Loading


The prescribed dynamic loading may be periodic or non-periodic. Periodic loading is
the loading that repeats itself after equal intervals of time. The interval of time during
which one cycle of loading is completed is called its time period. Non-periodic
loading is not repeated in a fixed pattern and magnitude. The examples of these two
types of loading are given in the following sub-sections.

10.5.1.1 Periodic loading


i) Sinusoidal Loading: This loading corresponds to simple harmonic motion and the
force in one cycle of loading varies as a sine wave, as shown in Fig. 10.11. This is 10.5.1.2 Non-periodic loading
usually produced by running of any machine / motor installed on a structural member. Non-periodic loading usually acts for short duration of time and can be of the following
In simple cases of machine foundation designs, in place of designing for the actual types:
dynamic forces, static loads are increased to include the effect of vibrations.
i) Impulsive Loading: This loading acts for very less time duration but the
ii) Stepped Loading: The pattern of loading is shown in Fig. 10.12. magnitude of load is very high. Most common example is a blast loading shown in Fig.
10.14.
LATERAL LOADS AND
CHAPTER 10: MOMENT ENVELOPS 377
average of approximately 30 km. Under the big mountain ranges, the thickness of crust
can approach a value of 1 00 km.
b. Mantle: This layer has an approximate thickness of 3000 km and consists of dense and
hot semi-solid material. It is composed of more iron, magnesium and calcium. The
temperature ranges from 1000 to 3500° C.
c. Outer Core: Outer core is a thickness of approximately 2200 km and consists of liquid
at a temperature of 3500 to 4000° C.
Fig. 10.14. Typical Impulsive Loading. d. Inner Core: The inner core has a radius of approximately 1250 km and is a layer of
solid material at temperatures higher than 4000° C. The core mainly consists of iron¬
ii) Earthquake Loading: The duration of the loading is relatively larger
than nickel alloy.
impulsive loading but the variation is higher, as shown in Fig. 10.15.

Fig. 10.15. Typical Earthquake Loading. Fig. 10.16. Various Earth Layers.
10.6 EARTHQUAKES
In plate tectonics, we are interested with the layers lithosphere and asthenosphere. The
Majority of the information given in this section is taken from U.S. Geological Survey outer layer of earth having an average thickness of at least 80 km is relatively rigid and
Website, more specifically by the report of W. Jacquelyne Kious and Robert I.
(USGS) is called lithosphere. This means that the lithosphere includes the crust and some more
part is also taken from Wikipedia through internet. An earthquake
Tilling. Some rigid part of mantle. The layer of earth below lithosphere having a thickness of 100 km
is the vibration of earth is softer / more mobile and is called asthenosphere. The rigid lithosphere actually
produced by rapid release of energy from within itself. This extra energy
may be stored in floats over the mobile asthenosphere.
earth and released at intervals due to many different phenomena, some of
which are as under:
1. Plate tectonics. 10.6.2 Plate Tectonics
2. Volcanic eruptions.
3. Atomic explosions. Plate tectonics deals with large scale movement and deformation of the earth’s
4. Collision of massive meteorites with the surface of earth. outermost layer (lithosphere) and proposes this movement to occur as a result of the
movement of rigid plates over a soft layer (asthenosphere). The lithosphere acts as
rigid plate that moves over partly molten asthenosphere. According to this theory,
10.6.1 Structure Of Earth.
lithosphere is cracked in places or broken in to smaller pieces or plates. This may have
The earth is divided in to four main layers called crust, mantle, outer core and
inner happened during initial drying of the earth from a molten state. There are seven large
core, as shown in Fig. 10.16. and several small plates. The largest plates are the Pacific plate, the North American
a. Crust: Crust is the outermost rigid layer of earth consisting of plate, the Eurasian plate, the Antarctic plate, the Indo-Australian plate and the African
solid material varying in plate.
temperature from surface temperature to a maximum temperature of 1000°
C. Its
thickness under deep oceans is close to 5 km and the thickness under continents has an All plates consist of either oceanic or continental lithosphere or a combination of both.
This means that a plate is either completely in ocean or continent or a combination of
LATERAL LOADS AND
378 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI | CHAPTER 10: MOMENT ENVELOPS 379
4
both. Over a time period of approximately 200 million years, the Indian Plate (Indo- '
Pak sub-continent) separated from Australian Plate and moved towards the Eurasian
Plate. When it touched the Eurasian Plate, the great Himalayas were formed.
'4
10.6.2. 1 Plate boundaries < Fig. 10.18. Divergent Plate Boundary.
There are two types of plate boundaries depending on their physical appearance as
explained below:
a- Mid-Oceanic Ridge: At the junction of two plates under the ocean, due to y
movement of ridges away from each other, molten lava rises up and after /1
solidification forms elevated lithosphere at the plate boundaries, which is known |
as mid-oceanic ridge. More earth surface if formed, and to maintain the
diameter and size of the earth the crust should disappear at some other locations
(actually happens at the subduction zones). J
b. Subduction Zone: This zone is a trench several kilometers deep frequently
found along the margins of the continents. Here the plates converge and one Fig. 10.19. Transform Plate Boundary.
plate moves under the other plate. The crust moves down and after melting
becomes part of the asthenoshere. d. Plate Boundary Zones: These are wide belts or zones in which the plate
Further, there are four types of plate boundaries depending on the relative movement boundaries are not well defined. The effects of the plates interaction are not
between the two adjoining plates. clear in these zones.

a- Convergent Plate Boundary: This type boundary is generally present where Each plate is bounded by a combination of these types of boundaries, that is, every edge
we see Subduction zones. The two adjoining plates move towards each other can be of a different type. The convergent plate boundaries are further of three different
and lithosphere of one plate moves over the lithosphere of the other plate. The types depending on whether the continental or oceanic lithosphere is at the leading edge
lithosphere that descends beneath an overriding plate melts and changes in to of each plate. During an ocean-continental convergence, the oceanic lithosphere sinks
asthenosphere. If a continental plate moves against an oceanic plate, it in to the asthenosphere. During this movement, sediments eroded from the continents
overrides and oceanic plate sinks below, as shown in Fig. 10.17. may accumulate in the trench formed by the down-going plate and some of these
sediments may be carried down in to the asthenosphere, which melt there. Volcanoes
may erupt in the continents where this material, due to its low density, rises up through
the continental plate.
An ocean-ocean convergence also gives rise to the formation of volcanoes in a manner
similar to the ocean-continent collision. However, here the volcanoes form on the
ocean crust.
Fig. 10.17. Convergent Plate Boundary. In a continent-continent convergence, the two landmasses come together, crumble and
deform, but neither sinks in to the asthenosphere due to their low density. The extreme
b- Divergent Plate Boundary: This usually occurs at mid-oceanic ridges where
heights of the Himalayan Mountains are attributed to this sort of collision. During the
two plates move away from each other. The fractures created by these
early movement of the Indian plate, the oceanic region in between the Indian and
movements are immediately filled with molten lava that oozes up from the hot
Eurasian plates was subducting under the Eurasian plate. However, when the
asthenosphere (Fig. 10.18). In this way same amount of sea floor is created as is
continents collided, a continent-continent convergence with rising of the material
consumed at a convergent plate boundary. At Mid-Atlantic Ridge, a new crust
started.
of about 25 mm is created eveiy year.
According to continental drift theory, all the continents were once part of a huge
c. Transform Plate Boundary: At transform plate boundary, the two plates slide landmass, which have slowly moved apart. The Indian sub-continent was not a part of
against each other without creating or destroying lithosphere (Fig. 10.19).
Asia. It drifted over millions of years from Australia to Asia and the collision produced
LATERAL LOADS AND
380 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI t CHAPTER 10: MOMENT ENVELOPS 381
Himalayas. Modem techniques such as GIS and GPS prove such movements of damage. There is usually an after-shock at an interval during which the other more
continents. | damaging waves approach the area.

The past record of earthquakes shows that highly seismic regions of earth are situated (2) S-Waves or Secondary Waves or Shear Waves: These waves involve
on the boundaries of the tectonic plates. Most of the earthquakes occur along the edges particle movement perpendicular to the direction of propagation of the wave (Fig.
of the interacting plates than within the plates and are called inter-plate earthquakes. J, 10.20(b)).
However, sometimes earthquakes may also occur within the plates and are called intra- 3
plate earthquakes. There can be several small plates within large plates with active / | >. -Wave Direction
inactive faults on their boundaries. These earthquakes are also produced due to large | < > -> Wave Direction
plate movements or other geological changes in the asthenosphere, which cause 3 Particle Movement
movements of the small plates along the local faults. In our area, earthquakes are J Particle Movement
concentrated over Main Karakoram Thrust (MKT), Main Boundary Thrust (MBT),
Main Mantle Thrust (MMT) and many other major and local faults. (a) P-Waves (b) S-Waves

Convection flow in the mantle occurs when less dense rocks rise and cooler and denser 5 Fig. 10.20. Types of Seismic Waves.
material sinks. Large convection cells in the mantle rotate just like wheels and cany the
lithosphere in a conveyer belt fashion. Hence thermal convection is the driving force When body waves reach the ground surface, part of these is reflected back while other
that propels the rigid lithosphere plates across the globe. part produces surface waves. Surface waves are the waves produced on the earth’s
surface due to an earthquake and are of following two types:
10.6.3 Focus And Epicenter
'
'
(1) R-Waves or Rayleigh Waves: These waves produce a circular motion
The point within the earth along the rupturing geological faults where an earthquake analogous to the motion of ocean waves. Hence, rotation along with vertical movement
originates is called the focus or hypocenter. The point on the earth’s surface directly takes place in case of Rayleigh waves (Fig. 10.21).
above the focus is called the epicenter. Earthquake waves radiate out from the focus.
The focal depth is the depth of the hypocenter below the epicenter. Focal distance is
the distance from the hypocenter to a given reference point.

10.6.4 Shallow And Deep Earthquakes Fig. 10.21. Movement of Particles in Rayleigh Waves.
Shallow focus earthquakes are produced if focus or hypocenter is near the surface with
a focal depth of up to 7 km. These earthquakes tend to be severe and more damaging. (2) L-Waves or Love Waves: These waves produce horizontal motion along the
This is because the rocks closer to the surface are stronger and build up more energy. ground surface transverse to the direction of propagation.
Deep focus earthquakes are produced at a focal depth of more than 7 km. One The amplitude of both R and L waves reduces with depth from the surface. Knowing
additional reason for more damage associated with the shallow earthquakes is that the the time interval (AT) between the arrival of P and S waves at a given site, the focal
energy is released closer to the buildings and is dispersed to a lesser area having more distance x/may be found as follows:
' intensity. The waves produced are usually much stronger producing severe shaking of
buildings. ^AT
10.6.5 Earthquake / Seismic Waves
where vp = velocity of the primary waves.
The waves originated at the rupture zone are called body waves and are of die following
two types: Recordings of the P-S time interval at three or more non-collinear sites enable the
epicentral position to be located. As distance from the epicenter increases, the duration
(1) P-Waves or Primary Waves or Dilation Waves: These waves involve of shaking at a given site increases and it becomes more complex. This is because of
particle movement parallel to the direction of propagation of the wave, as shown in Fig. the increase in time between the arrival of P and S waves and also due to scattering
10.20(a). The speed of travel of these waves is approximately 1.73 times greater than effects resulting from reflection of P and S waves from the surface.
the other waves. These waves are felt earlier in an earthquake and cause relatively less

1
LATERAL LOADS AND
CHAPTER 1 0: MOMENT ENVELOPS 383
382 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI

10.6.6 Destruction Caused By Earthquakes


Mercalli
Earthquake vibrations produce waves. When these waves reach the earth’s surface, the
Scale Effect
ground is shaken in different directions causing destruction. To measure the amount of
I. Felt by almost no one.
the destruction caused by an earthquake, two terms namely earthquake magnitude and
II. Felt by veiy few people.
earthquake intensity must be understood.
in. Tremor noticed by many, but they often do not realize it as an
10.6.6.1 Earthquake magnitude and Richter scale earthquake.
Earthquake magnitude is a measure of the energy released during an earthquake. It IV. Felt indoor by many. Feels like a truck has struck the building.
defines the size of the seismic event but is not related with damage or effect of v. Felt by nearly everyone; many people awakened. Swaying trees and
earthquake at a given location. The magnitude of earthquake is usually measured on poles may be observed.
Richter scale; the corresponding value is calculated on log scale as follows: VI. Felt by all; many people run outdoors. Furniture moved, slight damage
occurs.
Richter magnitude, M= logl(M(mm) + distance correction factor (log Ao) VII. Everyone runs outdoors. Poorly built structures considerably damaged;
Where, A = amplitude of seismic waves measured by the seismograph slight damage elsewhere.
vin. Specially designed structures damaged slightly, others collapse.
and A„ = empirical function depending on the distance of the measuring station ix. All buildings considerably damaged, many shift off at foundations.
from the epicenter. Noticeable cracks in ground.
In other words, the Richter scale is a logarithmic based scale that utilizes the amplitude x. Many structures damaged. Ground is badly cracked.
of seismic vibrations, recorded on a standard seismograph, to determine the strength of XL Almost all structures fall. Bridges wrecked. Very wide cracks in ground.
an earthquake. A magnitude of M4 Richter scale is ten-times greater than a magnitude xn. Total destruction. Waves seen on ground.
of M3 and is associated with an increase in energy release of (1010)1’5 = 31.6 times than
that of M3. A magnitude of M5 is 100 times greater than a magnitude of M3 scale and
the energy released is (1020)1'5 times greater than that of M3. 10.6.7 Measuring Earthquakes
The magnitude is related to the maximum trace deformation to the surface wave portion A typical seismograph is shown in Fig. 1 0.22, which is securely mounted on to the
of seismograms recorded by a standard Wood-Anderson seismograph at a distance of surface of the earth so that when the earth shakes, the entire unit shakes with it.
100 km from the epicenter. As such, it can be sensitive to the focal depth of the However, the mass on the spring, which has inertia, remains in the original position.
earthquake, and magnitudes computed from the body wave portions of seismograms are As the seismograph shakes under the action of the earthquake, the recording device on
often used to refine estimates of the magnitude. the mass records the relative motion between itself and the rest of the instrument, thus
Although Richter scale has no direct relationship with the destruction, following recording the ground motion. Accelerographs record ground acceleration in optical or
approximate relationship may be valid: digital form as a time-history record known as accelerograms.
generally not felt but recorded on seismographs
M2.5
M3.5 felt by many people
M4.5 some local damage may occur
M6.0 a destructive earthquake
M7.0 a major earthquake
M8.0 and up great earthquake
Presently, the Richter scale has been improved to moment magnitude scale (MMS).
10.6.6.2 Earthquake Intensity And Mercalli Scale
Intensity is an assessment of the effect of the earthquake at a given location and is not
directly related to the earthquake magnitude. This is determined not by reading Fig. 10.22. A Typical Seismograph.
instruments but by observing the effects on structures, human life and disturbance to
the ground surface. Modified Mercalli index is based on the observed effects of an
earthquake at a specific site.
LATERAL LOADS AND
384 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
CHAPTER 10: MOMENT ENVELOPS 385
10.6.8 Example Earthquakes
1. The magnitude, intensity and other characteristics of future earthquakes are not
1. Strongest earthquake of magnitude 9.5 occurred off the coast of Chile (South
precisely known.
America) in I960, killing at least 2,000 people and displacing 2 million people.
2. Stiffer structures attract more earthquake loads. These structures cannot
2. The financial damage occurred by Kobe (Japan) earthquake of 1995 was over US dissipate energy and all the energy is stored in them making them unstable.
$100 billion. 3. Heavier design means more mass of the structure. Due to larger mass, more
3. The most life damage was due to earthquake in Chinese province of Shaanxi in inertial forces are produced during the ground excitation.
1556 causing 830,000 fatalities.
4. Most destructive earthquakes of Pakistan are earthquake of Quetta in 1935 having a The most common method to design earthquake resistant structures is to design for
magnitude of M7.7 (killing 30,000 to 60,000 people) and Kashmir earthquake of mild earthquakes of expected common occurrence in the elastic range or in the inelastic
2005 having amagnitude of M7.6/7.8 (killing at least 80,000 people). range with less or no permanent deformations. Ductility is then provided for maximum
5. El-Centro earthquake of 1940, Mexico, is the most studied earthquake for structural expected intensity of earthquakes. Ductility is a measure of inelastic deformations that
analysis and design purposes. may be produced in a structure before its collapse. Inelastic deformations release
energy in the form of heat and make the structure stable. Permanent deformations may
10.6.9 Types Of Earthquakes be produced in the structure with considerable cracking and structure may not be
useable after a severe earthquake. However, the life is saved as the people may escape
10.6.8. 1 Types depending on duration out of the building.
1. Short Duration Earthquakes: These are usually more severe and occur closer to
the epicenter. The duration is 2 to 3 seconds. In essence the main aim of earthquake resistant design is to avoid loss of life and then
less loss to property is the second criterion. It may be tried that the damage is
2. Long Duration Earthquakes: These have low intensity but the duration can be up repairable for moderate earthquakes.
to 30 seconds or more. ‘

10.6.8.2 Types depending on intensity 10.6.11 Brief Description Of Methods Of Analysis


1. Single Shock Earthquakes: These are have short duration and occur in area
And Design For Earthquake Loading
around the fault zone (Fig. 10.23). 10.6.11.1 Free Vibration Analysis
2. Irregular Earthquakes: These occur in areas 400-500 km away from the fault If a structure is displaced by a considerable amount and is then released suddenly, it
zone (Fig.10.15). starts vibrating without the action of any external forces except the first excitation. The
study of response of a structure when it is vibrating without any external force is called
3. Earthquakes Causing Earth Deformations: These earthquakes cause large-scale free vibration analysis of structure. This analysis gives information about natural time
permanent deformation of the earth’s surface. period and frequency of the structure. Information is obtained about resonance and
other dynamics related parameters such as damping, etc.
10.6.11.2 Response History Analysis (RHA)
Response history analysis (RHA) is the evaluation of response against the elapsed time
of a system throughout a known record of earthquake by solving the actual equations of
motion considering all the dynamic forces. The fundamental problem in dynamic or
seismic analysis is to determine the deformation response (u) of the structure with
respect to time during an earthquake by solving the differential equation of motion.
Fig. 10.23. A Typical Single Shock Earthquake. 10.6.11.3 Response Spectrum Analysis (RSA)
The fundamental time period of a structure is the time taken to complete one complete
10.6.10 Main Considerations For Seismic Design cycle of vibration according to its fundamental mode (deformed shape during vibration)
Design of structures to withstand the maximum intensity earthquake is highly expensive of vibration. The natural circular frequency of the structure is the reciprocal of
and may not even be possible due to the following factors: fundamental natural time period for the structure. In other word, natural frequency is
LATERAL LOADS AND
386 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI ER10. MOMENT ENVELOPS 387
the number of cycles of vibration according to first mode of vibration per second.
When the maximum displacement of the structure during an earthquake is multiplied 10.6.11.6 Equivalent Static Or
with square of natural circular frequency, a term called Pseudo-Acceleration is Pseudo-Static Load Method
obtained. Similarly, Pseudo-Velocity is the maximum displacement during the In this method, some equivalent static forces are applied to approximately get the effect
earthquake multiplied by the natural circular frequency. Using these parameters, of vibrations according to fundamental and higher modes of vibration. This method is
response of various structures may be studied independent of the geometric dimensions recommended by almost all the design codes including Uniform Building Code (UBC),
of the structure. National Earthquake Hazard Reduction Program (NEHRP), National Building Code
If response history analysis (RHA) is carried out for a wide range of time periods and (NBC) of Canada and Building Code of Pakistan for easy and sufficiently accurate
ground motions; maximum responses for every case may be obtained. When these design of buildings.
maximum responses are plotted on a graph having four coordinates, natural period, As a general procedure, a base shear is first determined for the building, which is then
deformation, pseudo-velocity and pseudo-acceleration, the resulting charts are called distributed in to a set of static forces along the height of the structure depending upon
Response Spectra. Spectral response means maximum displacement, velocity or the stiffness and mass of the stories of the structure. The base shear depends upon
acceleration response. These charts may then be used for a wide range of structures for seismic hazard zone, type of structure, geometry and material of structure and
their quick evaluation. The graph having four coordinate axes, usually having log scale, importance of structure.
is called Tripartite Graph.
There can be major design deficiencies if the building code is applied to structures
Dynamic method of seismic analysis consists of Response History Analysis (RHA) and where dynamic properties differ significantly from those of ordinary buildings. For
Response Spectrum Analysis (RSA). These methods must be used under the following critical projects such as high-rise buildings, dams, nuclear power plants, long span
conditions: - - bridges, special earthquake considerations are necessary.
a. For all structures that are more than 73 m in height. The equivalent static method is only applicable if the following conditions are satisfied:
b. For all structures having a stiffness, weight or geometric vertical irregularity. a. All the types of structures in Zone 1 and regular or irregular structures, but not
Stiffness irregularity means the presence of a soft story. A soft story is defined for essential, hazardous or special use, in Zone 2.
as a story where the lateral stiffness is less than 70 percent of the upper story or
b. All regular structures up to 73 m in height with any lateral load resisting system.
less than 80 percent of the average stiffness of three stories immediately above
the story under consideration. Weight irregularity means the presence of a story c. All irregular structures up to 5 stories or 20 m height may be analyzed.
with weight more than 150 percent of the effective weight of an adjacent story, d. Regular structures having a flexible upper portion supported over a rigid lower
except top roof. portion satisfying the UBC provisions.
c. Structures of height more than five stories or 20 m present in Zones 3 and 4.
d. All structures with a time period greater than 0.7 seconds located on Soil Profile 10.7 FREE VIBRATION ANALYSIS
Type SF.
Lumped Mass: During an earthquake the earth only produces acceleration in the structure.
10.6.11.4 Base Isolation Method The force is produced when the mass of structure undergoes acceleration, which is called the
Inertial Force. In case of buildings, most of the mass is concentrated at the floor level.
This method is used in construction to reduce the earthquake demand on the structure Hence, for convenience in analysis, it is assumed that whole of the mass is present at a
and is not directly an analysis tool. In this technique, the base of a structure is partially particular height (at a point), known as Lumped Mass.
isolated from the foundations so that there may be relative movement between the
ground and the base of the building. In this way, less energy and vibrations are Degrees Of Freedom: The number of displacement components that must be considered in
transferred to the structure. order to represent the effects of all significant inertial forces of a structure are termed as the
10.6.11.5 Use Of Special Energy Dissipating Devices number of dynamic degrees of freedom of the structure. Most commonly only one degree of
freedom, that is, the horizontal movement of the mass, is considered for each lumped mass.
This technique is also related with construction and is not directly an analysis method. For example, three DOF may be considered for the structure shown in Fig. 10.24.
In some designs, special energy dissipating devices may be installed at locations of
maximum deformations. These devices are replaced after each considerable
earthquake. This makes the maintenance of the remaining part of the structure very
easy.
LATERAL LOADS AND
388 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 10: MOMENT ENVELOPS 389
Inertial Force: This force tries to retain the original shape or direction of motion of the
structure.
F . = mass x acceleration
Fig. 10.24. Lumped Masses and Dynamic DOFs. Lumped = m ii
Masses
Elastic Restoring Force: This is the resisting force that tries to restrict the deformation or
tries to regain the original shape. For a particular deflected shape, this acts as potential energy.
It acts as spring constant in the dynamic model.
Fe = ku
where k = stiffness of all columns of the story.
Three different ways of representing the models for the dynamic analyses are shown in Fig.
10.25. In this figure, F(f) is the dynamic force and is the stiffness of each story. Damping: Damping is the process by which free vibration steadily diminishes. This is due to
release of energy from the structure, usually in the form of heat. Damping is produced by
opening and closing of micro-cracks, friction between different components and deformations
within the inelastic range, etc.
Damping Force: The value of this force is equal to the product of the damping constant (c)
and the velocity. The damping constant represents the frictional characteristics and energy
losses of the structure.
Fd =cu
(a) Single Degree of Freedom System. Excitation Force: Excitation force is the external dynamic force acting on the system as a
function of time and is denoted by Ft.
Single Degree Of Freedom Damped Model:

A model to represent single degree of freedom


structure has the following components:
a) A mass element (m) having inertial forces.
b) A spring constant (k) representing the structural
stiffness. Fig. 10.26. Single Degree of Freedom
c) A damping element (c) representing the Damped Model.
frictional characteristics and energy losses of the
structure.
d) An excitation force Ft.
(b) Multiple Degrees of Freedom System.
Differential Equation Of Motion: The differential equation of motion may be written for the
Fig. 10.25. Various Methods of Showing Dynamic Models. equilibrium in horizontal direction, for the general case, as follows:
Displacement, Velocity And Acceleration: The following notation is used to represent F, + Fd + Fe ~Ft
displacement, velocity and acceleration: mii + cu + ku = Ft
Displacement = u For free vibration, the excitation force Ft and damping force Fd are not considered and the
Velocity = —=
di
u resulting differential equation becomes:
mu +ku = 0
Acceleration =
dt
= u or
k
m —
u + u = 0-
LATERAL LOADS AND
CHAPTER 10: MOMENT ENVELOPS 391
390 CONCRETE STRUCTURES PART - 1 AUTHOR ZAHID A. SIDDIQI

This is a second order differential equation of homogeneous type with constants coefficients. m = 8000 kg
A trial solution of this type of equation is:

u A cosat (simple harmonic function with angular speed <a) Angular frequency, a = I
[k
————
— - 12000x10’
J - 15.81
,CO1
rad/sec

then m=
— —
du
dt
= j4d»sinot
Time period, T - — =
Vm V 8000

= 0.397 sec
- ——
K
„ d^u a 15.81
and u = —Aa 2 cosat Frequency,/ = 1/7’ = 2.52 cycles/sec
dr
Put the values of u and w in the differential equation to get: i) The cosine function has the maximum value at an angle of zero. In the first case, as
2
shown in Fig. 10.27, the cosine function is starting with no lagging, meaning that u =
cos ait) + k A coses' = 0 umax when a t = 0.
or (-ma2 + k)A cosat = 0
The term cosat can not be zero at all times. Also, A = 0, represents a trivial solution.
-ma2+k - 0
_
= I
-im —
fF [kg as —
j

VW
r
v —
I force / length x length/time
-
force
' .
2
— ,
rao/sec

The value of k in the above expression must be in N/m as g is taken in m/sec2 units and m in
kgs. Comparing with the simple harmonic motion, the positive root of a? is known.as the
natural frequency of the system in radians per second units. It is to be noted that the period of
both the sine and cosine functions is 2n radians. Hence the expressions for the time period and Fig. 10.27. Plot of Vibrations for Case (i).
frequency may be determined as follows:
Total angle covered in one cycle = 2n rad ii) The cosine function is lagging by an angle equal to as shown in Fig. 10.28. Its
Angular speed, = a rad/sec value is not maximum at a t = 0.
Time required for one cycle to
complete, called time period, T = 2n I a
Natural frequency of the system,/ = a Hit
In the general solution u = A cos at, A is the maximum displacement with respect to the mean
position and may be called the amplitude of vibration, u0.
u = u0cosot
Example 10.3: For free vibration of mass of 8000 kgs, having a stiffness factor (spring
constant, k) of 2000 kN/m, calculate the time period and natural frequency. Also plot the
vibration for the following two initial conditions:
Fig. 10.28. Plot of Vibrations for Case (ii).
1. At t = 0, u = 0.01m,
du
dt —=0 When lagging is considered, the displacement function becomes:
2. At * = 0, u = 0.008m, — .= 0.08
dt
m/s u = u0 cos (at - 0)

Solution:
k = 2000 kN/m = 2000 x 103 N/m
and

dt
= -u0 a sin (at - . 0D
LATERAL LOADS AND
CHAPTER 10: 1 ENVELOPS 393
392 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI

For most of the buildings, the horizontal floors are assumed to be rigid (no bending or rotation)
Case (i): At / —
= o, dt = o => whereas the columns are assumed to be flexible with respect to the floors.

= - uo osin (- 0 and io* 0


So,
0
sin (^= 0 or ^= 0 u0
meaning no lagging
Example 10.4; A rigidly connected plane frame consists of rigid floor of mass 6000kg
supported by two flexible columns, 3.5m long, as shown in Fig. 10.30. The flexural rigidity of
At t = 0, u = 0.01m
0.01 -uD cos (0)
=>
or uo- 0.0 Im
the frame.

each column is El 4.0 x 1 012 N-mm2. Calculate natural time period and natural frequency of

m u
6000 kg u —
a ’
Case (ii): At t = 0, —
dt
= 0.08 m/s => 0.08 = - u0 sin (- {0

or 0.08 = u0<z>sin(^ (HI) EZ = 4.0x 1012 k


.5m
N-mm2
At t = 0, u = 0.008m => 0.008 = uo cos (- 0)
or 0.008 = uB cos (^ (TV)

Eq. HI /IV: uomsin^ _ 0.08 Fig. 10.30. Data for Frame of Example 10.4.
uo cos^ 0.008 Solution:
EI of each column = 4.0 x 1012 N-mm2 = 4.0 x 106 N-m2

<0 = 32.31° or 0.564 rad 12£7


lvalue for each column = kj = —— for rigid floor
From IV: u„ = 0.008 / cos d> = 0.01m

10.8 SPRING CONSTANT OR STIFFNESS


COEFFICIENT FOR COLUMNS
= —
12x4.0xl06 .1.12x10
3.53
= in6xT.
N/m

Total k-value = 2 x kj = 2.24 x 106 N/m


The spring constant (A?) is defined as the force required to produce unit elongation or m = 6000kg
contraction. This is analogous to the stiffness coefficient of a structure corresponding to a
particular degree of freedom in case of stiffness method of analysis. Stiffness coefficient is 2.24x10s „
a = 19.32 rad/sec
defined as the force required to produce unit deformation in the direction of the DOF. It is Vm V 6000
important to note that a vertical column behaves like a horizontal spring in these models. The
Time period, T= litl a> = = 0.325 sec
19.32
Natural frequency of the system,/ = 1 / 7 = 3.075 cycles per second
10.9 EQUATION OF MOTION FOR VISCOUS DAMPING

Damping means the presence of frictional forces in the structure, which transforms the
mechanical energy of system in to other forms of energy, such as, heat. If damping is
completely absent in an ideal system, a structure once excited will oscillate indefinitely with
constant amplitude at its natural frequency.
Mass Allowed Not Allowed Mass Allowed Not Allowed A viscous damper model considers damping behavior to be similar to a piston moving in a
viscous fluid. The viscous damping forces produced are proportional to the velocity of the
Fig. 10.29. Stiffness of Columns for Various End Conditions.
LATERAL LOADS AND
394 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 10: MOMENT ENVELOPS 395
piston. However, the actual damping in a structure may result from looseness of joints, dry ,
friction between components (called Coulomb damping), material damping (or internal
damping found by examining the area within the hysteresis loop between stresses and
acceleration), structural damping (general term for all types of damping in a structure) and
1.
—c \2
2m)
1 ki
m
=0
„ or
/

\2m)— =—
। c 1 k
m
called critically damped system

many other complex causes that would lead to nonlinear behavior of the structure. Structural
damping is proportional to displacement but is in-phase with the velocity of a harmonically 2.
c
2m J
y
>
k
m
called over-damped system
oscillating system. Usually, an attempt is made to express all types of damping together in
terms of equivalent viscous damping.
3.
c V<k called under-damped system
In case of viscous damper of Fig. 10.31, the damping force produced is directly proportional to 2m J m
the magnitude of velocity and opposite in direction to the direction of motion. The constant of
proportionality is known as viscous damping coefficient, denoted by c, and having units FT/L Critical Damping (cq-): It is defined as that amount of damping due to which a freely
such as N-sec/mm. excited system does not oscillate but returns to its original position in the shortest
possible time. An oscillating system is said to have critical damping if:

c#
2m J
y _ k_
m
c„ =2mJ— = 2jkm
\m
ft” ,

m if

Ccr = 2m ~ => = 2m co
Vm
Fig. 10.31. Forces Acting on Mass in a Viscous Damper Model. 2k _2k
or c^ = 2-7^ = 2k^km/k2 - 'klm co
For equilibrium of the mass,
mii + cu + ku = 0 Where Cq. = critical damping coefficient.
In case of critically damped systems, the roots of the characteristic equation are equal,
The exponential function u = Cep‘ satisfies the differential equation, giving ii-Cpep' and
ii = C p2ep! . From the differential equation, we get:
mCp2ep' + cCpep' +kCepl =0
called repeated roots. pi~p2~—p, where p =
c
—— ,
. In such case, the first solution is:
2m

or mp+cp + k-Q

—mc p + m£ = 0
This equation is called the characteristic equation of the system. The two roots of this
The other independent solution, in case of repeated roots, can be found by multiplying
the above function with t, as follows:

u2 (t) = constant x t x u, (r) = C2t e~pl


characteristic equation are:
The total solution becomes:
_ c/ c ] ,
±,
2m \\2m) —
- k
m
«(0 = uffy + uff)
ulf) - + C2te,-p>
Hence, the general solution of the equation is given by the superposition of the two possible
solutions and is given below: = (q yC^e^ where p = —2m—
w(0 = C1ep',+C2ew The graphical plot of this solution is given in Fig. 10.32.
Where, C/ and C2 are the constants of integration to be determined from the initial conditions. Damning Ratio Of Svstem Gf): Damping ratio of a system is defined as the ratio of
The following three distinct cases of solution for pi and p2 are possible depending on whether damping present in a system to its critical damping. £ = c / ca. Critical damping
the quantity under the radical in the characteristic equation is zero, positive or negative:
LATERAL LOADS AND
396 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 10: MOMENT ENVELOPS 397
coefficient usually ranges between 2 to 10% of ccr (^ = 0.02 to 0.10) for actual The above system of equations may conveniently be written in matrix form as follows:
structures. (IV)
1 => critically damped response
f > 1 => over-damped response
£ < 1 => under- damped response

Fig. 10.32. Response of Critically Damped System.

10.10 FREE VIBRATION ANALYSIS FOR MULTIPLE


DEGREES OF FREEDOM STRUCTURES

A structure in which There is no rotation of a horizontal section at the level of the floors is
termed as Shear Building. The following assumptions apply when modeling the structure (a) Single Bay (b) Single Column (a) Free Body Diagrams
using shear-building concept: Representation Model of Each Floor
1. The total mass of the structure is concentrated at the levels of the floors. In this way the Fig. 10.33. Various Models to Represent Shear Buildings.
actual structure with infinite number of degrees of freedom due to distributed mass is
changed in to a lumped mass model with degrees of freedom equal in number to the
lumped masses at the floors.
2. The floors are considered infinitely rigid as compared to columns. Thus, the joints
between the floors and the columns are fixed against rotation.
3. The axial deformation of the columns is neglected. This means that the horizontal
floors remain horizontal under the action of lateral loads. Fig. 10.34. Multi-mass Spring Model of a Shear Building.
Various models to represent the shear buildings in the form of sketches are shown in Figs.
10.33 and 10.34. Where, [Af] and [X] are the mass and stiffness matrices, respectively, given by:
Considering horizontal dynamic equilibrium of the free body diagrams of each of the three
floors, gives:
0
k2 +k3 (V)
+ (^ + F2)ut -k2u2 - F, (t) = 0 “^3
miWi+^i«1-*2(u2-“l)-^i(0 = 0 (I)
and {«}, {ii} and {f} are the displacement, acceleration and force vector given by:
m2u2 - k2ut + (k2 + £j)u2 - k3u3 - F2(t) = 0
m2ii2+k2(u2~i^-k3(u3-U2)~F2(t)=G (H)
(VI)
m3u3-k3u2+k3u3-F3(t) = 0
m3u3+k3(u3~u2)-F3(t) = 0 (III)
LATERAL LOADS AND
STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 10; MOMENT ENVELOPS 399
398 CONCRETE
coefficient (element ky of matrix |X|) is defined as the force produced at floor-i
The stiffaessdisplacement is given to floor-,; all other floors being fixed at zero displacement For each value of a2 satisfying the characteristic equation, Eq. XI can be solved for a;, in terms
when a unit of one reference value for any one constant out of the n-values. This is because that one of the
- *2+ is die force required at the second floor to produce a unit
J22 £
For example,of the same floor when 1st and 3ri floors do not move. Using the force vector of equations is already used to calculate the value of c? and hence two of the equations will
displacement may be converted into the equation of free vibration, as follows: become similar out of the set of n-equations. Usually the amplitude of first story is taken equal
zero forces, Eq. IV to unity and all other amplitudes are calculated with respect to it. Using this procedure, each
set of a, defines the relative amplitude and deformed shape of the frame with respect to a
of the undamped system tno
• •
be s™Ple particular frequency and time period value, known as normal mode or modal shape of
Knowing that the vibration
general solution of this equation mav be m tenDS °f amPlitude vibration. The term fundamental mode is used to refer to the mode associated with the lowest
motion, the frequency, while the other modes are called harmonics or higher harmonics. The normal
harmonicvelocity m, time t and phase angle a; as follows
angular modes or modal shapes represent the n possible ways of simple harmonic motions of the
u. =a,sin(at a) - i =1,2, ,n (VDI) structure that can occur in such a way that all the masses move in phase at the same frequency.
The amplitude at the floor level-i for mode-/ may be denoted by ay. For example, a3t denotes
Eq. VIII becomes: the relative amplitude of the second story when the structure vibrates freely at the fundamental
Tn matrix notation
(IX) natural frequency according to the fundamental mode. Typical normal modes of a two-story
{u}= {a}sin(at-a) frame are shown in Fig. 10.35.
is the amplitude of motion of the ith coordinate and n is die number of degrees of
*x"'s“bst^ w + MW™1 - ®) - W
MMsin ={0} a)since
-
*
-
sin {at -a) cannot be zero at all the times. (X)
* = 1.0 a„ 1.0
= a1 similar to general equation form A x = 2 B x (XI)
having n+1 unknowns, i.e., 2 and xi to x„. The n different values of
8^ of n equations
J^be evaluated that give n sets of x2 to x„ in terms of xb
E1££E
^^for —•
P!C And Eigen Vectors: Let A = [q,*] be a given matrix and consider the vector

it is clear that the zero vector x - 0 is a solution for any value of 2. A value
which the equation has a non-trivial solution x * 0 is called eigen-value or
tjc vaiue or latent root of the matrix-^. The solutions x * 0 corresponding to n
of the equation are called eigen-vectors or characteristic vectors of A
(a) Fundamental Mode (b) Second Mode
Fig. 10.35. Normal Modes for a Two-Story Building.

Hie total motion of the system is the superposition of the modal harmonic vibrations for all the
modes with varying contribution; the maximum contribution is usually of the fundamental
^611 values
ding to particular eigen-values 2. The set of all eigen-values is called the spectrum mode. The normal modes are orthogonal to each other. Orthogonality means that if all masses
Cf A^Thepyomwncies And Normal Modes: The non-trivial solution of Eq. X requires that the
largest of the absolute eigen-values of A is called the spectral radius of A. in a structure are multiplied with their amplitudes in two different modal shapes and are added,
azero answer is obtained. Modal Orthogonality has the following properties:

1? " . Ha crinal tn i ft 1. The work done by the wth mode inertia forces in going through the rth mode
determinant ui w displacements is zero.
|[X]-®2M=O (XII)
2. The work done by the equivalent static forces associated with displacements in the
mode in going through the r01 mode displacements is zero.
the above equation results in a polynomial of degree n in terms of m2 which
When expanded, fbe system. This equation can be solved to get n
the characteristic equation of
is known as of 0 (r»i , a2), the positive square roots of which are called the
Example 10.5: Determine kt and k2 for the frame given in Fig. 10.36.
real distinct values of the structure. These frequencies may then be Solution:
angular natural frequencies
(n?j, ai,
having units of cycles per second. Further we
changed in to natural frequencies Tn) as follows:
EI of bottom story = 14.4 x 1011 N-mm2 = 1 440000 N-m2
may also find the natural time periods (Tj, T2, EZoftopstory = 7.2 x 10" N-mm2 = 720000N-m2
2ff
T, =4 (XIII)
LATERAL LOADS AND
CHAPTER 10: MOMENT ENVELOPS 401
400 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

F13100xl03 -7750x10’
-7750x10’ 7750x10’ j
^/23751 °1
[ 0 11519^
Q

13,100,000 -2375 Im2 -7,750,000 =Q


-7,750,000 7,750,000 -11519m2

1.015 x 1014 - 3.350 x 1011 m2 + 2.736 x 108 m4 - 6.006 x 10!3 = 0


2.736 m4 - 3350 x 108 m2 + 414400 = 0
3350 ±2586 m?2= 1084.79
2
= 2x2.736 ^2= 139.62 and
Fig. 10.36. Data for Frame of Example 10.5. mz= 11.82rad/sec and m, = 32.94 rad/sec
f,= m7/2n = 188 cps and/3= mj/2n = 5.24 cps
3EI 3x1440000 Tj= 1 // = 0.532 sec and F2= l//3 = 0.191 sec
t-value for bottom story, ki : = 47407 N/m
L3 4.53 Now using the equation,
12EI 12x720000
k-value for top story, k2
L3 3.75’
: = 163840 N/m ([^]-m2[M]){a}= {0}
'13,100,000 -2375 Im2 -7,750,000 J0}
7,750,000 -11519m2 J [a2J
Example 10.6; Determine the natural frequencies and normal modes for die frame given |0j
-7,750,000
in Fig. 10.37.
The first and the second rows become the same equation as one equation has already
Solution;
been used in the determination of m. Considering the first row for m = m;, we get,
= wl/g= 233,000/9.81 9,783,885 a„- 7,750,000 a}J = 0
= 23751 kg Considering a,, = 1.000, a21 = 1.262.
m2 = w2/g = 113,000/9.81 Similarly, considering the first row for m= ah, we get,
= 11519 kg w2=113kN •- -12,664,847 aI2 - 7,750,000 a22 = 0
k2 = = 5350 x 103N/m
5350 N/mm Considering al2 = 1.000, a22 = - 1.634.
k2 = 7750 N/mm = 7750 x IO3 N/m i3 = 7750 N/mm
The normal mode shapes are shown in Fig. 10.38.
_ kj + k2 -k2 w; = 233kN «/ 10.11 TIME HISTORY ANALYSIS
-k, k.
13l00xl03 -7750xl03' = 5350 N/mm 10.11.1 Single Degree Of Freedom Damped System
Subjected To Dynamic Force
-7750xl03 7750x10’
The equation of motion for a linear single degree of freedom (SDF) system subjected to
0 23751 0 Fig. 10.37. Data for Frame of Example external force is the general equation derived earlier and reproduced below.
0 0 11519 10.6. mil + cu + ku = Ft
|[<|-c4Af]|=0
402 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI LATERAL LOADS AND
CHAPTER 10: MOMENTENVELOPS 403

Duhamel’s integral provides an alternative method to the classical solution if the


applied force F(t) is defined analytically by a single function that permits analytical
evaluation of the integral. For complex excitations that are defined only by numerical
values of F(t) at discrete time instants, Duhamel’s integral can be evaluated by
numerical methods. However, more efficient numerical procedures not using the
Duhamel’s integral are also available to determine dynamic response.

Example 10.7: Using Duhamel’s integral, determine the response of a SDF system,
assumed to be initially at rest, subjected to a step force, F(t) = Fo at t > 0.
Solution:
The force function F( r) = Fo is shown in Fig. 10.39.

1 f,
= F(r)sin[0„(Z-r)]Jr
mat*
, k
= f F.sin^-r)]^
Jo
t
maK
Fig. 10.39. Dynamic Force for
(a) Fundamental Mode = —ma>n
F j z
{ - sin[®n (r - ) dr Example 10.7.

Fig. 10.38. Normal Modal Shapes for Frame of Example 1 0.6.

The initial displacement u(0) and initial u (0) at time zero must be specified to define
u(l) = —5-cos[©„(/-?)];
p

"
= —— [cos
F
m
0- cos

the problem completely. Typically, the structure is at rest before the onset of dynamic
excitation and = ^-[l-cos(<v)]
k
u = u =0 at t = 0
One technique that is used to solve such linear differential equations is the use of 10.11.2 Dynamic Effect Of Earthquakes On Rigid Structures
Duhamel's Integral. In this approach, the applied force is represented as a sequence of A rigid structure is that structure where k ~ 00 and hence there is no relative movement
infinitesimally short impulses. The response of the system to an applied force Ft, at of top of the building relative to the base or ground. In such cases, u = ug, where ug is
time t, is obtained by adding the responses to all impulses up to that time. For the ground displacement at any time. The lateral force becomes m iig = Z (m g) = ZxW.
undamped SDF system, The factor Z is the zone factor of the locality and is defined as the peak ground
w(0 - —^JBF(r)sin[#n(/-r)j4/r where = Vi/m
acceleration divided by the acceleration due to gravity. This is an ideal case and does
not practically exist practically for most of the buildings.

This equation is based on “at rest” initial conditions. The variable “r” is an 10.11.3 Dynamic Effect Of Earthquake On Flexible Structures
intermediate time variable. Similarly, using the symbols for the damped angular In this case, the base of the structure or ground is subjected to earthquake motion,
velocity and £ for the critical damping coefficient, the solution can be written as whereas, each floor of building or lumped masses have a different pattern of motion
follows: due to flexibility of the structure. It is to be noted here that only the relative motion u

u(t) = —— [,F(r)e_f“"('"r)sin[©„(r-r)]
Jo
dr
between the mass and the base produce structural deformations along with elastic and
damping forces. The rigid body component of displacement of the structure produces
no inertial forces.
LATERAL LOADS AND
CHAPTER 10: MOMENT ENVELOPS 405
404 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
critical damping coefficient and is equal to 2^1km or 2m a>. Dividing the above
equation of motion by m, the following is obtained:
Let ug (t) = displacement of ground at any time

and
(t) = total or absolute displacement of mass at any time
u (?) = relative displacement between mass and ground
.. £ .
u + —ccru
m —
k
+ w = -u
m
= w,(0-Wg(0 U + 2co^ii + a>2u = —u^
U =Ut—Ug
u = Ut—ug The solution of this equation is:
Referring to Fig. 10.40, the equation of motion is:
Fj + Fd + Fe = 0
mil, +c(u, ~ug)+k(ul -ug) = 0
u(l) = —
mto:a
sin[«0-r)] dr
For undamped systems with £= 0, the expression becomes:
m(ii + m ) + cu + ku = 0
mil + cu + ku = —mil?
u(0 = —f
mco^
O) sin[m(z - r)] dr

Fs=ku =k(ut-Ug) 10.12 RESPONSE SPECTRUM ANALYSIS (RSA)


Fi=mut Spectral response means the maximum displacement, velocity or acceleration response.
Fd = cu ~ c(ut - Ug) Pseudo-Acceleration is the maximum displacement of the structure multiplied with square of
Fig. 10.40. Earthquake Forces Acting on a Flexible Structure.
natural circular frequency. Pseudo-Velocity is the maximum displacement of the structure
during an earthquake multiplied with the natural circular frequency.
This shows that the equation of motion subjected to ground motion u/t) using relative Sd = spectral displacement
displacement between the structure and the ground is exactly same as that of equation Sv = spectral velocity
of motion of a fixed base structure subjected to dynamic external force of mil g. This = co Sd
dynamic force in the second case may be denoted by F^f) and is called the effective Sa = spectral acceleration
earthquake force shown in Fig. 10.41. = co Sv = co2 Sd
^0= Because of this inter-relationship, all three spectra may be plotted on the same graph using
tripartite axes and logarithmic scales. A response curve or response spectrum is a graph of
w. the spectral (or maximum) response of a range of single-degree-of-freedom oscillators to a
specified ground motion, plotted against the frequency or period of the oscillators. The
(lamping ratio of oscillator may be varied and the response recorded may be displacement,
velocity or acceleration. Sometimes, the maximum absolute acceleration is called spectral
acceleration while the terms spectral velocity and spectral distance are used for values relative
to the maximum acceleration. A typical simple response spectrum, not according to scale or
particular site and earthquake, is shown in Fig. 10.42.
Fixed Base
Fig. 10.41. Fixed Base Idealization of Earthquake Motion. 10.12.1 Procedure To Use Response Spectra For SDOF Systems
The procedure to use response spectrum to calculate the earthquake lateral forces for
This force is equal to the mass times the ground acceleration acting opposite to the single degree of freedom systems is summarized as under:
acceleration. It is important to recognize that the effective earthquake force is 1. Calculate angular speed co and time period T for the structure.
proportional to the mass of the structure. Thus if the structural designer increases the 2. Estimate the damping ratio £
effective earthquake force, the mass of members is increased making the inertial forces
still greater. It is already explained that the damping ratio £ = c 7 where c„ is the
406 CONCRETE STRUCTURES PART - I AUTHOR: ZAHID A. SIDDIQI LATERAL LOADS AND
CHAPTER 10: MOMENT ENVELOPS 407

a) For £= 0, from the response spectrum, Sv = 300 mm/sec = 0.3 m/sec


Sd - Sv / a = 0.0116 m
SF for each column = 0.0116 x 2000 = 23.2 kN
Total lateral force = 2 x 23 .2 = 46.5 kN
b) For £=0.10, from the response spectrum, = 70 mm/sec = 0-07 m/sec
Sd = Sv I a - 0.0027 m
SF for each column - 0.0027 x 2000 = 5.42 kN
Total lateral force = 2 x 5.42 = 10.84 kN

10.12.2 Procedure To Use Response Spectra For MDOF Systems


The procedure to use response spectrum for the calculation of the earthquake lateral
forces in case of multiple degrees of freedom systems is summarized as under:
1. Calculate angular speed co and time period T for the desired mode of vibration
of the structure.
2. Find the mode shape a,.
3. Estimate the damping ratio £.
4. Find Sd, Sv and Sa from the response spectrum or calculate others after knowing
Fig. 10.42. A Typical Out-of-Scale Response Spectrum. one out of these.
5. Calculate the effective weight as follows:
3. Use applicable response spectrum for a particular area and find Sd or Sv or Sa
WE = (Z'W
(the values are inter-convertible) against the time period.
4. Find shear force in each column as:
S.F = Sd x k 6. Calculate the base shear as follows:
5. Find the total lateral force by adding shear forces in all the columns.
6. The lateral forces may be scaled to the values given by the equivalent static K=
method. This scaling depends upon the types of frames and their ductility. 7. Find the lateral force at each level as follows:
Example 10.8: Calculate the shear force in m = 6000 kg F, = V
each column and the total lateral force for tire frame
given in Fig. 10.43 by using the response spectrum
k = 2000 kN/m 5 The lateral forces may be scaled to the values given by the equivalent static
of Fig. 10.42 for the following damping ratios:
method. This scaling depends upon the types of frames and their ductility.
4=2000 kN/m
a) £=0 b) £=10% 10,12.3 UBC Design Response Spectrum
Solution:
Fig. 10.43. Frame for Example 10.8.
k for each column = 2,000,000 N/m
Although the response spectrum should be developed for a particular site based on geologic,
m = 6000 kg
seismological and soil properties at a specific site. In the absence of such response spectrum,
the UBC design response spectrum may be used. According to UBC 1631.4.1, response
fk
= J— = J
fl?
12x2,000,000
= 25.82 rad/sec spectrum analysis is defined as an elastic dynamic analysis of a structure utilizing the peak
Vm V 6000 dynamic response of all modes having a significant contribution to total structural response.
T = In! at = 0.243 sec
LATERAL LOADS AN
CHAPTER 10: MOMENT ENVELOPS 409
408 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI

According to UBC 1633, the following requirements for seismic design of buildings must be
. Maximum modal contributions are combined in a statistical manner to obtain an
approximate total structural response.
satisfied:
. This condition is considered satisfied by demonstrating that for the modes considered,
1. The earthquake design forces are to be resisted only by the elements of the dedicated at least 90 percent of the participating mass of the structure is included in the
seismic-force-resisting system. calculation of response for each principal horizontal direction.
2. All building components in Seismic Zones 2, 3 and 4 are to be designed to resist the . The peak member forces, displacements, story forces, story shear’s and base reactions
effects of all the loads including the seismic and the gravity loads. for each mode should be combined by recognized methods.
3. Uplift produced as a result of seismic loads must always be considered in design. . When three-dimensional models are used for analysis, modal interaction effects shall
4. The effects of earthquake forces must additionally be considered at oblique angles in be considered when combining modal maxima.
Seismic Zones other than Zones 0 and 1 for the following cases: Reduction Of Elastic Response Parameters For Design
a. When the structure has plan irregularity Type 5 (UBC Table 16-M) and there are According to UBC 1631.5.4, the Elastic Response Parameters obtained by a Response
nonparallel systems. The orientation of vertical lateral-ioad-resisting elements
Spectrum Analysis may be reduced for design as explained below, with the condition
(columns and shear walls) is not parallel to or symmetric about the major
that the design base-shear by these methods should not be less than die Elastic
orthogonal axes of the lateral-force-resisting system.
Response Base Shear divided by the value of response modification factor, R.
b. When the structure has plan irregularity Type 1 or torsional irregularity (UBC
1. For all regular structures, Elastic Response Parameters determined according to
- Table 16-M) for both major axes. Torsional irregularity is present when the
UBC Response Spectrum may be reduced such that the corresponding design base
maximum story drift at one end of the structure transverse to an axis is more
shear is not less than 90 percent of the base shear determined according to the Static
than 1.2 times the average of the story drifts Of the two ends of the structure.
Force Procedure.
A column is common to two or more intersecting lateral-force-resisting systems.
2. For all regular structures using site-specific design spectrum, Elastic Response
d. If the axial load in the column due to seismic forces acting in either direction is Parameters may be reduced such that the corresponding design base shear is not
less than 20 percent of the column axial load capacity, these effects need not to less than 80 percent of the base shear according to the Static Force Procedure.
be considered.
3. For all irregular structures, Elastic Response Parameters may be reduced such that
These effects may be included in the design by satisfying the following requirements: the corresponding design base shear is not less than 1 00 percent of the base shear
a. Design the elements for 100 percent of the code design seismic forces in one according to the Static Force Procedure.
direction plus 30 percent of the code design seismic forces in the perpendicular 4. The vertical component of ground motion is taken equal to two-thirds of the
direction. horizontal accelerations. Where the Near Source Factor, Na, is greater than 1.0,
b. The design is to be based on the combination giving the greater force effect. site-specific vertical response spectra is to be used in place of the above procedure.
As an alternate to the above procedure, the effects of the two orthogonal The UBC Design Response Spectrum is shown in Fig. 10.44.
directions may be combined on a square root of the sum of the squares (SRSS)
basis. Example 10.9: Determine the base shear for the fundamental mode of the Frame shown

d. When the method given in point c is used, the sign of each term should be
considered in such a way to obtain the most conservative result.
in Fig. 10.45 by using response spectrum of Fig. 10.42 with £ = 1 %. Also determine the
lateral load at each level for the fundamental mode.
5. The strength and stiffness of the framing between the base and the foundation shall not 1.00
be less than that of the superstructure.
6. The concrete lateral-force-resisting frames in Seismic Zones 3 and 4 should be special m = 15.1 rad/sec a, = <1.95
moment-resisting frames. V
2.86
7. The concrete lateral-force-resisting frames in Seismic Zone 2 should be a minimum of Solution:
intermediate moment-resisting frames. Dead Loads:
The procedure of using the Response Spectrum method is as follows:
» Peak modal responses are calculated using the ordinates of the appropriate response
spectrum curve which correspond to the modal periods.
Level 2 -
Level 1 & 3 = (880)(30)(30) x 9.81 / 1000 = 7770 kN
= (590)(30)(30) x 9.81 / 1000 209 kN
LATERAL LOADS AND
CHAPTER 10: MOMENT ENVELOPS 411
410 CONCRETE STRUCTWF.S PART -I AUTHOR: ZAHID A. SIDDIQI

we = (40150)
(91132)
=17689kN y^
17689x5.285
9.81
= 9530kN

10.13 EQUIVALENT STATIC LOAD METHOD

The parameters discussed in the following sub-sections are required to be evaluated to get the
values of equivalent static loads according to UBC-97.

10.13.1 Seismic Zone Factor (Z)


Table 16-1 of UBC-97 categorizes land areas into six zones depending upon the Code
estimate of peak ground acceleration at a particular site. The zone factor (2) is given as
a factor of peak acceleration with respect to acceleration due to gravity (g) and it varies
from 0.075 to 0.40. The suggested values correspond to recurrence interval of 475
years giving a 10 percent probability of being exceeded in a 50 years period. These
zone factors are reproduced in Table 10.7.
Fig. 10.44. UBC Design Response Spectrum.
Table 10.7. Seismic Zones and Effective Peak Ground Accelerations.

a> = 15.1 rad/sec Zone 4 3 2B 2A 1


T = 2tU a> ' 0.416 sec
Effective Peak Ground 0.40 0.30 0.20 0.15 0.075
From the response spectrum, = 350mm/sec Acceleration (EPA)
So = <dSv = 5285 mm/sec2 = 5.285 m/sec2
The calculations are made in Table 10.6 using 10.13.2 Soil Profile Types
the following expressions:
The ground vibrations traveling through the soil may be amplified or reduced
depending upon the fundamental period and type of strata. UBC classifies soils into six
profile types, as given in Table 10.8. This classification depends on the average shear
wave velocity in the top 30m of material.
V- WESa / g and
Table 10.8. Soil Profile Types.
Ft = V
YWia- Soil Profile Type Description of Soil Shear Wave Velocity
(m/s)
Fig. 10.45. Frame for Example 10,9. Sa Hard rock > 1500
Sb Rock 760 to 1500
Table 10.6. Calculation Sheet for Example 10.9. Sc Soft rock 360 to 760
Sd Stiff soil 180 to 360
Level Wi at Se Soft soil < 180
kN mm
Ft Sf Veiy soft clayey soil Detailed investigations required
kN-mm kN-mm2 kN
1 7770 1.00 7770 7770 1844
2 5209 1.95 10158 19807 2411
3 7770 2.86 22222 63555 5275
Total 20749 40150 91132 9530
LATERAL LOADS AND
412 CONCRETE STRUCTURES PART - 1 AUTHOR- ZAHID A. SIDDIQI CHAPTER 1 0: MOMENT ENVELOPS 413
10.13.3 Seismic Source Types When soil parameters are unknown, soil profile type So may be assumed in seismic
The seismic source types are decided based on the maximum moment magnitude zones 3 and 4 and profile Se may be assumed in other zones. For a regular structure,
potential of a fault (identical to the Richter scale value) and its slip rate per year. Type the near source factor needs not exceed 1.3.
C represents almost an inactive fault in Table 10.9.
10.13.6 Fundamental Time Period Of A Structure
Table 10,9. Seismic Source Characteristics. The time period of a structure may exactly be calculated by performing free vibration
Seismic Source i Source Characteristics
analysis of the structure, which involves lengthy calculations. Following empirical
Type | Maximum Moment Potential Slip Rate (mm/year) methods are also available to reasonably guess the fundamental time period of a
A
|
I 7.0 > 5.0 structure:
10.13.6.1 Approximate method
B The fault which is not A or C
C | < 6.5 2.0
Fundamental time period, T = sec
10
10.13.4 Near-Source Factors
10.13.6.2 Method A of UBC
Two factors, No and Nv, are used to consider increased ground motions near a fault.
The factor Na is the acceleration-based factor that is important for short-period
structures and velocity-based factor Nv that is important for periods exceeding one
Ta = C,^
second. The values of these factors are given in Table 10.10. where hn = height of the roof above the base in meters, not including the height
of parapets.
Table 10.10. Near Source Factors (Na and AQ for Various Seismic Source Types. Ct = 0.085 for steel moment resisting frames
I = 0.073 for reinforced concrete moment resisting frames and eccentric
Seismic Distance From Fault
braced steel frames
Source 2 km 5 km 10km 15 km
Type Na Nv Na Nv Nv X = 0.050 for all other buildings
A 1.5 2.0 1.2 1.6 1.0 1.2 1.0 1.0
10.13.6.3 Method B of UBC
B 1.3 1.6 1.0 1.2 1.0 1.0 1.0 1.0

10.13.5 Ground Response Coefficients Tb = 2x1^=; 1.47^ for Zones 1,2 and 3
The two ground response coefficients, Ca and Cv, give indication of the vibration
amplification capacity of a soil depending on zone factor (Z), soil profile factor (5) and 1.3 Ta for Zones 4
the near-source factors (Na and Vv). The fundamental period of a structure determines where <5, = static elastic deflection at level “i” due to the forces applied at all
whether Ca or Cv is more important for design of a structure. Table 10.11 gives the levels, increasing in a linear way with height. The value of
values of these factors. deflection must be with respect to the base in mm.
total lateral force at i - th floor
Table 10.11. Ground Response Coefficients, Ca and Cv.
<
= shear stiffness of columns under floor“F
Soil Zone 1 Zone 2A Zone 2B Zone 3 Zone 4 kt
ft = lateral force at level “z”, N
Profile ca cv G cr Ca c. Ca cv Ca G
wt = dead load located at level“F, N
Sa 0.06 0.06 0.12 0.12 0 16 0.16 0.24 0.24 0.32 Na 0.32M
Sb 0.08 0.08 0.15 0.15 0.20 0.20 0.30 0.30 0.40 0.40 a;
and g = acceleration due to gravity
Sc 0.09 0.13 0.18 0.25 0.24 0.32 0.33 0.45 0.40 Na 0.56
= 9810 mm/sec
Sd 0.12 0.18 0.22 0.32 0.28 0.40 0.36 0.54 0.44 Na 0.64 Example 1O.1O: Calculate the empirical time period for the steel moment resisting frame,
Se 0.19 0.26 0.30 0.50 0.34 0.64 0.36 0.84 0.36 Na 0.96 A, shown in Fig. 10.46, located on a site in zone-3.
LATERAL LOADS AND
CHAPTER 10: MOMENT EN VE LOP S 415
AUTHOR: ZAHID A. SIDDIQI
414 CONCRETE STRUCTURES PART -
1

TB In -- 1.47x for Zones 1,2 and 3


Solution:
1. Approximate Method:
7.421x10’
~~ 27C -V 9810x2.669x10
1 - - < 1.4x0.517
T = number of stories 1 10 - 0.3 sec = 1.058 < 0.724
2. Method A: = 0.724 sec
h„ = 3 x 3.7 = 11.1 m
Ta = = 0.085(11.1)^ = 0.517 sec 10.13.7 Ductility
Ductility of an element shows its capacity to deform in the inelastic range without
collapse. Due to these inelastic deformations, the energy is dissipated making the
m} = 455,000 kg structure relatively stable against earthquake forces. If these deformations successfully
/ = 525 kN occur in the two opposite directions causing reversal of stresses in the members,
hysteresis loops are produced dissipating energy in each cycle of loading, unloading and
loading in the opposite direction.

»350 kN 10.13.8 Response Modification Factor (R)


The response modification factor of a structure [R) is ratio of the seismic base shear of
an elastic system to a reduced design base shear depending upon ductility, energy
/ = 175 kN absorbing capacity, increase in natural period due to yielding and increase in damping
ratio of the structure. If shear walls or braced frames provide support to gravity loads
and all the lateral loads, the structural system is a Bearing Wall System (BWS). In
other words, the gravity loads are resting on walls. If separate systems are provided to
resist lateral and gravity loads, the structural system is called Building Frame System
Fig. 10.46. Frame for Example 10.10.
(BFS). No special detailing is required for gravity load supporting frames. Special
Moment Resisting Frames (SMRF) are frames specially detailed to provide high
3. Method B: ductility and support for lateral and gravity loads by flexural action. Moment Resisting
= 70000 N/mm and fa = 52500 N/mm Frames With Masonry Shear Walls are called MRWF systems. Dual Systems are
kj = 105000 N/mm, k?
those in which more than one systems are used together. Some typical values of
= W+Z+Z)/^ response modification factor are given in Table 10.12.
= (525 + 350 + 175) (1000) / 105000
- 10mm The value of the response modification factor (R) is determined from consideration of a
^</3+f2}lk, + St structure’s over-strength capacity beyond the point at which the elastic response of the
4 structure is exceeded. The value of R always exceeds unity, which indicates that all
= (525 +350) (1000) / 70000 + 10 structures are designed for forces less than would be produced in a completely elastic
= 22.5 mm
structure. This reduced force level is made possible by the energy absorption and
$ -fjl^ + Sz dissipation capacity of the structure at displacements in excess of initial yield.
= 525000 1 52500 + 22.5
= 32.5 mm Lightly damped structures constructed of brittle materials are unable to tolerate
455000x32.52) x9.812 appreciable deformation in excess of initial yield and are assigned low values of R.
= (455000xl02 + 455000x22.52 + Highly damped structures constructed of ductile materials are assigned larger values of
= 7.421 xlO9 N-mm2 x 22.5 + 525000 x 32.5 R.
YfA = 175000 x 10 + 350000
= 2.669 xlO7 N-mm2
LATERAL LOADS AND
416 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 10: MOMENT ENVELOPS 417

Example 10.11: Calculate the value of the seismic force coefficient (G) for an ordinary
Table 10.12. Response Modification Factor, J?.
steel moment resisting frame of Fig. 10.47 on stiff soil. T= 0.517 sec.
Structural System R Height Limit
4-
(m)
1. BWS with concrete or masonry shear walls 4.5 49
2. BWS with steel braced frames 4.4 49
3. BFS with steel eccentrically braced frames 7.0 73
4. BFS with concrete shear walls 5.5 73
5. BFS with masonry shear walls 5.5 49
6. BFS with steel ordinary braced frames 6.4 73
7. Steel or concrete SMRF 8.5 None
8. Masonry MRWF 6.5 49
9. Concrete shear walls with SMRF 8.5 None
Many other types given in UBC
Fig. 10.47. Frame for Example 10.11.
10.13.9 Seismic Importance Factor (I) Solution:
Stiff soil corresponds to soil Profile SD (Table 10.8).
Depending on the occupancy category, value of importance factor for different types of
G = 0.36 and Cv = 0.54 (Table 10.11)
buildings is given in UBC-97. The factor is equal to 1.25 for essential and hazardous
facilities and 1.00 for special occupancy, standard occupancy and miscellaneous
-
For special moment resisting steel frames, R 8.5 (Table 10.12)
For ordinary building, 7=1.0
structures. Time period, T = 0.517 sec

10.13.10 Seismic Response Coefficient (Cs) CI


G = -±-
KI
(G)^ = 2.5 CaHR
The seismic response coefficient (Cs) is the fraction of total dead load of the structure
(G)min = 0.11 GZ
that is acting as base shear on the structure. This means that tire base shear (V) is: K=
0.54 x 1.0
Cs W. This factor depends upon velocity or acceleration based ground response G = - 5xQ517 =0.123 (G)^ = 2.5x0.36x1.0/8.5 = 0.106
coefficient (G or CD), importance factor (T), response modification factor (/?) and time
period (7). The relationship between these coefficients may be explained as under: (G)min = 0.11x0.36x1.0 =0.040
i) Larger is the ground response coefficient for a specific zone and soil profile: Hence, G = 0.106
larger is the Cs factor and the base shear.
10.13.11 Seismic Dead Load (W)
ii) Larger is the importance factor; larger is the seismic response coefficient.
iii) Larger is the response modification factor depending upon ductility and better The seismic dead load (IF) consists of the following:
detailing; smaller is the base shear. i) Dead load of the structure.
iv) Larger is the time period of the structure; smaller are the design forces. ii) 25 percent of the floor live load for storage and warehouses.
iii) A minimum allowance of 50 kg/m2 for movable partitions.
G
Let, Response time Ts = and Tu = 0.27j iv) The total weight of permanent equipment and fittings.

C7 10.13.12 Magnitude Of Base Shear (V)


C = (ifT^TD subjected to maximum and minimum values
RT 10.13.12.1 UBC refined formula
Maximum value = 2.5 CaH R (Controls when T- To to Ts)
Base shear V = C3W
Minimum value = 0.11 GZ (OR) Q.^ZNvI / R for zone-4
LATERAL LOADS AND
418 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
CHAPTER 10: MOMENT ENVELOPS 419
calculations is 245 kg/m2. Determine the seismic base shear and compare the results with
Maximum inelastic displacement Am = 0.7 R As simplified formula.
=
Where As the displacement corresponding to the shear V, given above.

10.13.12.2 UBC simplified formuJ?


Base shear V = (3.0 Ca / R)W •

This is a conservative formula having the following restrictions: Wi


i) Ordinary occupancy type.
ii) Light-frame construction not exceeding 3 stories.
iii) Any construction, except bearing wall systems, but not exceeding two
stories.

10.13. 1 3 Distribution Of Base Shear At Various Story Levels

Shear at a particular story, Fx = x 1


ri Wi
Where E'
fV
a,
— modal base shear
= seismic dead load at level-;
= mode shape component at level-;' for the given mode
Fig. 10.48. Plan of Building for Example 10.12.

and
wt
ax

=
seismic dead load located at level-x
mode shape component at level-x for the given mode
Solution:
Roof load = WR = 245x5.5x4.5x9.81 /1000
Considering half height of walls to act with the roof:
= 59.49 kN
For uniform distribution of mass over height and for first linear mode, the distribution
of base shear may be simplified as follows:
Longer side walls
Right side wall
-
= W] 2x440x3.75x2.0x9.81 / 1000
= W2 = 440x2.0x2.0x9.81 / 1000
= 32.37 kN
= 17.27 kN
Lcli side wall = IF? = 440x2.0x2.0x9.81 / 1000 = 17.27 kN

Where Vt = base shear corresponding to first mode


Total Load = W
Ca for stiff clay (soil profile Sd), from Table 10.11 = 0.36
- 126.40 kN

ht — height above the base to level-/ Cv for stiff clay (soil profile Sd), from Table 10.11 = 0.54
I = 1.0 for a standard occupancy
and hx = height above the base to level-x
In order to account for higher mode effects in the above expression for long period Ta = C,(\)^ = 0.050(4.0)% = 0.141 sec
buildings, an additional force Ft is added at the top of the structure.
Ft = 0.07 TV when T >0.7 sec
R for a masonry bearing wall system, from Table 10.12 = 4.5
Where V - total base shear Cv (C,)^ = 2.5 COIIR
= Ft + LFX Ki
= 0.11 Cal
Wh
In such cases: Fx = (V-F,) ~ — C, =
0 54*1
1V0
=0.851 (Climax
V 2 = 2.5x0.36x1.0 / 4.5 ~ 0.20
4.5x0.141
Example 10.12: The floor plan of a single story commercial building located at a site in
(Qmin = 0.11x0.36x1.0 =0.040
seismic zone 3 on stiff clay is given in Fig. 10.48. The 4.0 m high masonry shear walls are Hence, Cs = 0.20
load bearing and have a seif-weight of 440 kg/m2. The weight of roof to be included in seismic For such single story wall bearing structures usually (Cj)max = 2.5 Ca I! R is critical.

5
420 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDlQ[ 1 LATERAL LOADS AND
MOMENT ENVE LOPS 421
CHAPTER HE
Seismic base shear = G x IF | jrvnmple 10.14: A steel moment resisting frame, with fundamental period of 2.5 seconds
= 0.20 x 126.40 = 25.28 kN load of 90,000 kN, is located at a site in zone 4 with a soil profile type Sd- The
Seismic base shear according to die simplified formula is
। ^nd a total dead 7.5 km from a potential seismic source type A. Calculate the value of the
as under: 5 site is situated
V = (3.0 Ca/ R)W seismic base shear.
= (3.0 x 0.36 /4.5) x 126.4 = 30.34 kN ? Solution^
j Near Source Factors (N„ and M):
Example 10.13: A three story
steel moment
located at a site in zone 4 with a soil profile type resisting frame, shown in Fig. 10. 49, is
Sg. The site is situated 7.5 km
; Soil profile type = SD
potential seismic source type A. Calculate
the value of the seismic base shear.
from a f Distance from source = 7.5 km
period of the frame by method-A is 0.517 sec. The time j Na = 1.1 and = 1.4 by interpolation from Table 10.10.
Solution: . Ground Response Coefficients (C„ and CyY
Near Source Factors (Ng and NJ: Zone =4
Soil profile type =SD
Soil profile type = Sg
Using Table 10.11,
Distance from source = 7.5 km
Na = U andNv = 1.4 by interpolation from Table 10.10. Ca = 0.44 Na = 0.44 x 1.1 = 0.484
G = 0.64 Nv = 0.64x1.4 = 0.896
Ground Response Coefficients (G and
C^: = 455,000 kg Response Coefficient (G):
Zone = 4
Soil profile type = SB Ts = G / (2.5 Ca)
= 0.896 / (2.5 x 0.484) = 0.740 sec
UsingTablelO.il,
C„ = 0.4 N„ = 0.4 x 1.1 = 0.44 qL = 455,000 kg T = 2.5 sec
G = 0.4 Av = 0.4 x 1.4 = 0.56 R =8.5 for steel moment resisting frame
Response Coefficient (CX
Ts = CJ (2.5 Ca)
455,000 kd As Ta > Ts CJ
RT
. 0.896x1.0
8.5 x 2.5
= 0 M22
= 0.56 / (2.5 x 0.44) The minimum value of Cs for zone-4 is;
= 0.509 sec
G = 0.SZNJ/R = 0.8 x0.4 x 1.4 x 1.0 / 8.5 = 0.0527
Ta = 0.2 Ts Hence, G = 0.0532
= 0.102 sec
Fig. 10.49. Frame for Example 10.13.
Base Shear:
R - 8.5 for steel moment resisting frame W = 90,000 kN
As Ta > T, CVI _ 0.56x1.0 F = Cs W

Base Shear:
RT 8.5x0.517
~
° 127 = 0.0532 x 90,000 = 4788 kN

Example 10.15: A two story masonry bearing wall structure of plan area 8m x 8m and
W = total weight height of each story equal to 4.5m consists of walls on all the four sides having a mass of 450
= 3 x 455,000 x 9.81 / 1000 = 13390 kN kg/m2. The building is located at a site in seismic zone 3 with an undetermined soil profile and
K = CSW the floors have a mass of 100 kg/m2. Find the vertical earthquake force distribution and
compare it with the approximate distribution.
= 0.127 x 13390 = 1701 kN
LATERAL LOADS AND
422 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI MOMENT ENVELOPS 423
CHAPTER 10:
Solution:
Time Period: +: 112.5 kN 91.3 kN
Roof height = 9m
Ct = 0.05 for bearing wall system
Ta = Ct(h„f4 = 0.05 x93'4 = 0.260 103.3 kN 167.6 kN
Time Period < 0.7 sec, so Ft = 0
Dead Load At Level-1:
215.8 kN 258.9 kN
Roof = 100 x 82 = 6400 kg :
Four walls = 4 x 450 x 8 x 4.5 / 2 = 32400 kg (b) Using Approximate Expression
« (a) Using Exact Expression
W2 = (6400 + 32400) x 9.81 / 1000 = 380.6 kN \
Fig. 10.50. Lateral Earthquake Forces for Structure of Example 10.15.
Dead Load At Level-1:
Roof = 6400 kg 1
Four walls = 4 x 450 x 8 x 4.5 = 64800 kg Fx = y—
W
Wr = 698 5
Fi = (258.98) 1079.1 -
(6400 + 64800) x 9.81 / 1000 = 698.5 kN = 167.6 kN
For such building, the following formula for C, governs:
Cs = 2.5 CaI/R F2 = (258.98) = 91.3 kN
1079.1
R = 4.5 for a masonry bearing wall system
I 1.0 for standard occupancy - Example 10.16: An eight story hospital
Ca = 0.36 for soil profile type Sq, zone3. I building R. C. frame is shown in Fig. 10.51, which
W =W, + W1 is to be built in Abbotabad. The columns are 4m
= 698.5 + 380.6 = 1079.1 kN ? C/C perpendicular to the plane of paper.
Determine earthquake base shear and its
Cs = 2.5CaI/R = 2.5x0.36x1.0/4.5 = 0.20 distribution over each story level considering floor
Base shear = V= Cs W = 0.20 x 1079.1 = 215.8 kN finish equal to 200 kg/m2 and partitions load of
= 698.5 x 4.5 + 380.6 x 9 = 6569 kN-m 100 kg/m2. Outside walls are assumed to have a
Distribution Of Vertical Forces: mass, distributed over the plan area, of 300 kg/m2
for all floors except the roof where it is 175 kg/m2.
Slab thickness for floors 1 and 2 is 200mm and for
Fx = ^y-F^ all other floors is L50mm.
Fig. 10.51. Frame of Example 10.16.
698.5x4.5 Solution:
Fi = (215.8) = 103.3 kN
6569 According to Building Code of Pakistan, 2007, consider the given part of Abbotabad to be
F2 = (215.8) 380.6x9.0 = 112 5 kN placed in Zone 2B for the present example. The soil profile type may be assumed to be Se.
6569 hn = 25m
The distribution of lateral forces is shown in Fig. 10.50. Ct = 0.073
Approximate Distribution Of Vertical Forces: As column stiffness is not known, Method B cannot be used to determine the time
period.
Seismic base shear according to the simplified formula is as under: Ta
3/4
= Cth™ = 0.073 x 25 = 0.816 sec
V =(3.0Co/R)W WA = W2 = (0.2 x 2400+ 100 + 200 + 300) x 8 x4 x9.81 /1000 = 339 kN
= (3.0 x 0.36 / 4.5) x 1079.1 = 258.98 kN
424 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQi

W3 to W, = (0.15 x 2400 + 100 + 200 + 300) x 8 x 4 x 9.81 1 1000


= 301.4 kN
= (0.15 x 2400 + 200 + 175) x 8 x 4 x 9.81 / 1000
= 230.7 kN
Total Load, W= 2 x 339 + 5 x 301.4 + 230.7
= 2415.7 kN
Zone = 2B
Soil profile type = SE
UsingTablel0.il, Ca = 0.34 and G = 0.64
= 1.25
T
R
Ts

0.816 sec
= 8.5 if special moment resisting frame is designed
= Cp/(2.5 = 0.64/ (2.5 x 0.34) = 0.753 sec
As Ta > Ts Cs CJ _ 0.64x1.25
" °'1153
RT 8.5x0.816
The minimum value of C3: Fig. 10.52. Individual and Cumulative Story Shears for Example 10.16.
Cs = O.HCaJ = 0.11 x 0.34 x 1.25 = 0.0468
Hence, C3 0.01153 10.14 GENERAL CONSIDERATIONS FOR SEISMIC DESIGN
Base shear = V = Cs W = 0.1153 x 2415.7 = 278.5 kN As already explained, buildings are designed to withstand moderate earthquakes without
Additional to force, Ft = 0.07 x 0.816 x 278.5 = 15.9 kN damage and severe earthquakes without collapse. Earthquake movements impose deformations
on the structures. We find inertial forces due to these earthquake movements depending upon
the structure. Dynamic effects like resonance are also important to be considered. Due to
'IM availability of limited data, the design is generally based on statistical studies of the previous
earthquakes. As more and more earthquake data become available and understanding of the
The values of lateral forces at various levels are calculated in Table 10.13
and are structural behavior is improved, Building Codes undergo modifications to cover the
graphically shown in Fig. 10.52. The force Ft in this figure accounts for the
effects of weaknesses in design criteria of the previous codes. Further, the safety of a structure subjected
the higher modes of vibration.
to earthquake loading also depends on the designer’s understanding of the response of the
structure to ground motion. It is prohibitively expensive to design the structure in the elastic
Table 10.13. Calculation of Lateral Forces For Example 10.15.
range. Overall structural ductility is very important in such designs.
Level fix Following are the general considerations for the seismic design of structures:
(kN)
Fx
(m)
8 230.7
(kn-m) (kN) 1. Design for earthquakes differ from the design for gravity and wind loads particular with
25 5767.5 =
45.1+15.9 61.0
7 respect to greater sensitivity of earthquake-induced forces to the geometry of the
301.4 22 6630.8 51.8
6 structure. Most structures, which are not extremely tall, are designed by the equivalent
301.4 19 5726.6 44.7 static loading (up to about 20 stories). This is applicable for regular buildings with
5
4 __ 301.4
_
301.4
16
13
4822.4
3918.2
37.7
30.6
center of mass and center of resistance very near to each other (Fig. 10.53).
2. Design is made for loads that are a function of weight/mass of the structure. Bulk Of
3 301.4 10 3014.0 23.5
2 339.0 7 2373.0
mass is located at floor levels, so forces are considered at these levels. Triangular
18.5 variation of forces is generally assumed. Without careful design, for an irregular
1 339.0 4 1356.0 10.6
Total | 2415.7 33608.5 _ 278.40
426 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDII1QJ
7
I CHAPTER 10:
LATERAL LOADS AND
MOMENT ENVELOPS 427
structure, forces and displacements can be concentrated in portions of the structure usually exhibit stronger response to long period ground motion and short buildings
may not be capable of providing the adequate strength or ductility. that i usually exhibit stronger response to short period ground motion.
5. The first mode of vibration usually provides the greatest contribution to lateral
displacement.
6. The taller structures are more affected by the higher modes of vibration and their
effect
.Center of Resistance actually adds to the effects of lower modes. The upper stories attract more forces due to
And Center of Mass
the higher modes of vibration.
7. The longer duration of earthquake always has a greater potential for damage
to the
structure.
Fig. 10.53. Center of Resistance against Earthquake and result in
8. Vertical geometric (Fig. 10.55a) and plan irregularities (Fig. 10.55b) may
Center of Mass of a Regular Structure. torsion induced by ground motion.
The configuration of a structure has a major effect on its response to an earthquake.
Structures with a discontinuity in stiffness or geometry can be subjected to very high
displacements and forces. For example, the absence of shear walls, infill walls or even
cladding al a particular story level, as compared to other stories, causes concentration of
displacements at this story. The ground floor of a shopping center generally has this
weakness, as shown in Fig.10.54. This weak story compared with rest of the structure
is termed as open or soft story. The larger displacements require a considerably larger
ductility at the level of soft story. If this amount of ductility is not available, the
a) Vertical Geometric Irregularities b) Plan Irregularities
structure fails locally at this level. Such a design is not recommended and the stiffening
members must be continued to the foundations. Fig. 10.55. Geometrical Irregularities in Structures.

9. The addition of stiff members, such as shear wall, can on one side reduce the
displacements of the structure and hence the damage. On the other side, stiff members
pick up a greater portion of the load. When this effect is ignored in design, unexpected
and often undesirable results can occur.
10. An adequate separation must be left between structures. Large lateral displacements
can cause the structures to come in contact with each other during an earthquake. This
results in major damage due to hammering effect.
Flexible Story
11. Members designed for seismic loading must behave in a ductile manner and should
dissipate energy without compromising the strength. Confinement of concrete is to be
provided to ensure ductility in members subjected to shear and bending. Due to this
Fig. 10.54. Deformation of a Building Having Soft Story. confinement, the beams and columns can undergo nonlinear cyclic bending. Further,
the flexural strength is maintained and no excessive diagonal tension cracking occurs.
3. Steps to strengthen a member for one type of loading may actually increase the forces in Confinement is provided by the use of closed loops or spiral reinforcement that
the member and change the mode of failure from ductile to brittle. encloses the core-concrete of beams and columns. Confined concrete allows the
4. As the frequency of the ground motion becomes closer to one of the natural frequencies formation of ductile hinges that can dissipate energy in case of reinforced concrete
of a structure, the chances of the structure to experience resonance increases. This frames.
4
results in an increase in both displacements and damage to the structure. The frequency ; 12. It must be tried that the plastic hinges are developed in the beams rather than columns.
or time period of a structure basically depends on height of the structure and
the I The weak beam - strong column approach is always preferred for the design of
earthquake response of a structure especially varies with the height. Tall buildings | reinforced concrete frames subjected to seismic loading. This effect is achieved by

I
llSil
i
LATERAL LOADS AND
MOMENT ENVE LOPS 429
428 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIqj CHAPTER 10:
making the columns stronger than the beams at the joints, forcing the hinges to he
formed in the beams. The advantage of this approach is that the overall vertical load
’ 7. Moment Frame: This is a lateral force-resisting frame that resists forces by flexure
carrying capacity is maintained near collapse and smaller portion of the structure is besides shear and axial forces. This may be of three types, namely, ordinary moment
frame, intermediate moment frame, and special moment frame.
affected by the nonlinear behavior.
8. Ordinary Moment Frame: Any frame that satisfies the general ACI Code
13. Transverse reinforcement for the columns is to be carefully designed for the shear force provisions but not necessarily the seismic provisions is called ordinary moment frame.
due to lateral loads in addition to shear force resulting from the dead and live loads. A It may be cast-in-place or precast frame. This type of frame may be used in regions of
smaller length column closer to high stiffness members or shear walls may attract large low seismic risk or where correspondingly low seismic performance is used in the
shear forces and may fail in shear. This type of column, called captive column, is more
calculation of seismic forces. This frame has performance corresponding to Seismic
critical for design in shear than in flexure. Design Category B.
10.15 ACI DEFINITIONS FOR SEISMIC DESIGN 9. Intermediate Moment Frame: This type of frame satisfies the general requirements
of the ACI Code along with provisions of ACI 21 for super-structure and part of the
Chapter 21 of the ACI code deals with special provisions for seismic design. Some of the structure between the base and the foundation. Base of the foundation is the level at
definitions are discussed below: which earthquake motions are transferred to the structure, not necessarily the ground
level. Further, only cast-in-place frames are included in this category. This type is used
1. Ductile Connection: A connection that is capable of yielding up to the level of the in regions of moderate seismic risk or for frames where intermediate seismic
design earthquake displacements. performance is taken in the calculation of the loads. Once this type of frame is
2. Strong Connection: This type of connection remains elastic even if the adjoining I considered for the seismic analysis, all requirements of this type of frame given in
members yield as a result of design earthquake displacements. Chapter 21 of ACI Code must be applied.
3. Crosstie: This is a continuous reinforcing bar satisfying the following requirements: 10. Special Moment Frame: This frame may be either cast-in-place satisfying the ACI
Code 21.1.3 to 21.1.7 and 21.5 to 21.7 or it may be precast additionally satisfying the
a) It should have a seismic hook at one end. ; Code requirement 21.8. The general provisions of ACI must also be satisfied. This
b) It should have a hook not less than 90° at the other end. type of frame must be used in regions of high seismic risk or where high seismic
c) It should have a six-diameter extension at the 90° hook side. performance is considered in the analysis. In such cases special structural walls and
d) Both the hooks must go around and anchored against peripheral longitudinal diaphragms may also be used. For design and detailing, all requirements of special
bars. moment frame must be satisfied if it is used for calculation of tire forces.
e) Two successive cross ties between the same longitudinal bars must alternate in
their hooks.
। 1 1. Plastic Hinge Region: The is the length over which the flexural yielding due to
design displacements is extended, which is not to be taken lesser than the overall depth
of the member from the critical section at which yielding starts.
4 Design Displacement: It is the code prescribed total lateral displacement in case of 'A
earthquake. As the codes for seismic design allow inclusion of the nonlinear response, TI 12. Seismic Hook: This is a hook having a bend greater than or equal to 135° for
the stability of the lateral force resisting system must be studied at displacements larger • crossties, rectangular stirrups and hoops and greater than or equal to 90° for circular
than those obtained by linear analysis. Further, the interaction with other structural and hoops. The hooks after going around the longitudinal reinforcement must have
non-structural members must also be considered. If detailed nonlinear response extension of greater of six-diameters and 75mm, projecting into the interior of the
analysis is not carried out and code specific value is also not available, an approximate stirrup or hoop.
method may be to consider the total displacement equal to double of the displacement 13. Seismic Design Categories (SDC).* The categories of design depend upon the
obtained by linear analysis considering cracked stiffness for the members. performance, structural detailing, provision of ductility and special earthquake design
5. Hoop: This is a closed tie with seismic hooks at both ends, or a spiral with seismic requirements. The requirements become more strict from Category A to F. According
hooks at both ends. , to UBC 1997, SDC A and B are used in Seismic Zones 0 and 1, SDC C is used in
6. Lateral Force Resisting System: The members of the structure designed to resist Seismic Zone 2 and SDC F, E and F in Seismic Zones 3 and 4.
.
the earthquake forces, collectively making the resisting frames. ' ~? 14. The minimum compressive strength of concrete for members resisting seismic
forces must be greater than or equal to 21 MPa.
LATERAL LOADS AND j
430 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI CHAPTER lb: MOMENT ENVELOPS 431 1

10.16 PATTERN LOADS For gravity load analysis, assume far ends of columns to be ACI 8.10.3 &
1.
fixed. ACI 8.11.1b
The various loads are combined according to their probability of occurrence together and the
required overload factor for design. Following load combinations are suggested to be 2. Pattern loads are applied as specified. ACI 8.11.2
investigated by the ACI Code (already given in an earlier chapter): 3. Analyze line representation of structure, and correct moments to ACI 8.9.2
face of supports. and 8.9.3
1.
2.
3.
4.
U
U -= 1.4 D
1.2 5 + 1.6Z + 0.5 (Lr or S or A)
1.2 D +1.6 (Lr or S or K) +(1.0 Lr or 0.5 W)
U = 1.25 + 1.0 W+ 1.0L + 0.5 (Lr or 5 or 2?)
4. Any reasonable assumption for stiffness of members may be
used, but must be consistent throughout the analysis.
ACI 8.7.1

5. U = 1.2D + 1.0£+1.0Z +0.25 The live load changes in magnitude as well as in direction. It is possible that some spans out of
6. U = 0.9 0 + 1.0 W all the structure are loaded by live load. Pattern loading is that position of live load, which
I U=0.9D+1.0E gives maximum force-effect at a particular section. According to ACI 8.11.2, it is allowed to
consider only the following two arrangements of live loads:
The nominal service load symbols have the following meanings:
a. Factored dead load acts on all spans with full factored live load on two adjacent spans.
D Dead load This arrangement gives the maximum negative moment at the central support.
F *

toads due to weight and pressures of fluids with well-defined densities and b. Factored dead load acts on all spans with full factored live load on alternate spans. This
controllable maximum heights. Where F is present, it must be included with the arrangement gives the maximum positive moment within the fully loaded spans.
same factor as D in Eqs. 1, 5 and 7.
T Cumulative effect of temperature, creep, shrinkage and differential settlement The actual number of loading arrangements becomes greater when the adjacent two span
L Live load loaded condition is applied for each support and when the alternate spans loaded condition is
H Loads due to weight and pressures of soil or water in soil. Vertical H is included icversed to load those panels which were previously without any live load. The loading cases,
in D. When lateral H acts alone or adds to the effects of the loads, include it as given in Fig. 10.56, are to be considered for the indicated maximum moments in case of beams.
1.62/. Where lateral H is permanent and counteracts, include it as 0.9H.
However, lateral 27 is not included if it is not permanent and it counteracts other It is to be noted that to get the influence line diagram having negative (hogging) ordinates on
loads. lower side of the beam, applied moment and rotation at the section are still to be applied in the
positive (sagging) direction. The arrow outside a particular segment shows its direction of
Lr Roof live load rotation.
S' Snow load
R Rain load 1 Three loading arrangements are to be checked for the relative magnitudes of axial load (A) and
W Wind load • bending moment (Mu) in case of design of columns. These pattern loads are shown in Fig.
E Load effects of seismic forces 10.57. The arrangement of load in case (b) with minimum eccentricity may be more critical for
lower floors. The arrangement (c) gives the largest eccentricity due to larger Mu / Pu ratio.
. This condition may govern exterior columns design of the top floor of buildings.
In general, the lateral loads are more critical if they produce moment reversal at the negative
moment sections. The sixth combination is critical when during construction there is a
According to ACI 8.10.1, column are to be designed for the axial loads obtained by applying
heaviest windstorm. The uplift acting on the roof may also cause moment reversal. The same
combination may give the condition of least axial load and heavy moment for the columns.
! factored loads on all floors or roof and the moment obtained by applying factored loads on a
; single adjacent span of the floor or roof under consideration. The design must also be checked
ACI Code provisions for development of reinforcement indirectly assure that some strength is for the maximum ratio of moment to axial load and for this the suitable loading condition must
reserved for this moment reversal by requiring that at least 25 percent of the maximum positive • be found out.
reinforcement be extended and anchored at the support if the frame is part of the primary lateral
load resisting system. The following ACI provisions can be used to quickly and correctly
analyze a frame:
LATERAL LOADS AND
MOMENT ENVELOPS 433
chapter 10'
432 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDiqi

1.2D

1.2D

1.2D+ 1.6£

Column Under
Consideration

ILD For M c
c) Less Pu and Large Mu.

Fig. 10.57. Pattern Loading For Columns.

10.17 CORRECTION OF BEAM MOMENT AT FACE OF SUPPORTS


As already mentioned, the beam moments are to be corrected at the face of supports as per ACI
8.9.2 and 8.9.3. This makes the design for negative moments considerably economical, as the
drop of moments near a support is usually very sudden. A typical moment diagram / envelop
shown in Fig. 10.58 can be used to carry out this type of correction.
ILDForMs
Fig. 10.56. Pattern Loading to Get Maximum Force Effects.
JUM 1C-KTXlj LAJCUJU

434 CONCRETE STRUCTURES PART 1


- AUTHOR- ZAHID A. SIDDIQI ’
CHAPTER 10:
MOMENT ENVELOPS 435
Similarly a shear envelope may also be plotted. As stated in the previous
ornent envelope.
loading is defined as the position of live load on a structure that produces
0 article, pattern
maximum at P0^ of interest For example, total load acting on two adjacent

jiffx , —
= M> max —
i ry f
— i
| I nans of a continuous beam with only dead load on the adjoining spans produces maximum
“I2J “12A4J support in-between these two panels. For plotting moment envelope the number
moment at thebending
Considering Vu approximately equal to Wv£/2, of individual moment diagrams depends on the number of load combinations, number
of direction3 in which horizontal loads may act and the number of pattern loads to be
considered.
pygmoZe 10.17: Plot shear and moment envelopes for the two span continuous beam
shown in Fig. 10.59. The beam is subjected only to dead and live loads having magnitudes
given in the figure. Assume EI^b = EIBc.
Wdl = 10 kN/m
IQdi wll = 20 kN/m

Fig. 10.59. Two-Span Continuous Beam of Example 10.17.

Solution:
LOAD CASE - 1 (Total load on both spans!
The loads for this case are shown in Fig. 10.60. Our first purpose of analysis is to
calculate moment at support-B to make the beam determinate. The calculations are
carried out as under and the free body diagrams of the two spans are shown in Fig.
10.60:
w, = 1.2wdl+ L6i*zz
=1.2x10+1.6x20
Fig. 10.58. Moment Envelope near Beam Joint with Column.
I|
||||
|||
||||'| I = 44 kN/m
The term
——
w c2
8
is relatively small and may be neglected to get the following
expression for moment at the face of support:
reduced
Fig. 10.60. Loading of Case-1 for Beam of Example 10.17.

Some designers prefer to include the approximately consider the


F.E.M.= -^M-
12
= 366.67 kN-m
term in the expression by reducing the value of the negative
effect of third positive
term reasonably as under: The calculations and free body diagrams of the beams are shown in Fig. 10.61.
From Fig. 10.61, Vx = 165 — 44 x and

Mx = 165x 22X2
The shear force and bending moment diagrams are given in Fig. 10.62.
10.18 MOMENT ENVELOPE LOAD CASE - P (Total load on left span and dead load on right span)

jf betiding moment diagrams are plotted for each load combination The loading, calculations and free body diagrams of the two spans are given in Fig.
and load pattern on one
diagram, the outline of these diagrams represents the design moment diagram 10.63.
and is called a

j
436 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI

i.ol LATERAL LOADS AND


1.0
CHAPTER 10: MOMENT ENVELOPS 437
FEM -366.6 +366.6 -366.6 +366.6
1+366,6' +183.31-183
+183,3 3 = 25 - 12 xi and M2 = 25 x- 6 x22
0 T +550 -550
c.o. moment
The resulting shear force and bending moment diagrams are sketched in Fig. 10.64.
c.o. moment
i
r PAD CASE - HI (Total load on right span and dead load on left span)
44
This case is identical to Case II due to symmetry of loading and the structure and hence there is
550 kN-m I no need to perform the detailed analysis once again.
550 kN-m f4 Jr J,

165kN 275kN 275kN 165kN F.E.M. =(12X10)z/12


Fig. 10.61. Free Body Diagrams For Beam of = 100 kN-m
Example 10.17.

44 12

3.75m 350 kN-m


|
350 kN-m
_
10 25kN
185kN 255kN 95kN

Fig 10.63. Solution of Example 10.17 for Case-II. I

SHEAR AND MOMENT ENVELOPES


The shear and moment envelopes are plotted in Fig.10.65by superimposing the shear force
diagrams and bending moment diagrams of the three cases.
Fig. 10.62. SF and BM Diagrams for
Case-I of Example 10.17.
Example 10.18: Plot moment envelope for beam AB at level 2 of the frame given in Fig.
The shear force and bending moment at various
sections can be obtained from the 10.66 using approximate analysis and determine the design end beam-moments.
following expressions:
Ki = 185-44xi wd = 25 kN/m
and M/ = I85xi-22xi2 WL = 22 kN/m
Column size = 300mm x 300mm
LATERAL LOADS AND
438 CONCRETE STRUCTURES PART T
1 MOMENT ENVELOPS 439
AUTHOR: ZAHID A. SIDDjqj CHAPTER 10:

Fig. 10.66. Frame and Other Data for Example 10.18.


Solution:
Service DL FEM = 75 kN-m
Service LL FEM = 66 kN-m
Service DL SSM
Service LL SSM

113 kN-m
= 99 kN-m
Considering clockwise joint reactive moments as positive, the analysis at level-2 is
performed below:
Fig. 10.64. SF and BM Diagrams for Case-H
of Example 10.17. i (al Live Load On Left Panel

The analysis for live load on left panel is performed in Fig. 10.67.
Loading
Shear Envelnne rkm FEM

S
Dist.

Fig. 10.67. Analysis for Live Load on Left Panel.

The positive mid-span moment is calculated by adding the average negative moment at
the mid-span and, the simply supported mid-span moment
Shear at end-A = 22 x 3-1.5 = 64.5 kN
-
Bending moment at x-distance from end-A = 49 + 64.5 x 11 x2 -
tb) Live Load On Right Panel

The results may be obtained from Case (a) using symmetry.


’g. 10.65. Shear and Moment Envelopes for Beam of Example
10.17. Shear at end-A = — 4.0 kN
Bending moment at x-distance from end-A = 5.0 - 4.0 x
LATERAL LOADS AND
440 CONCRETE STRUCTURES PART - 1 MOMENT ENVELOPS 441
AUTHOR ZAHID A.
siDDlQj CHAPTER 10:
(c) Dead Load Throughout For service LL on alternate spans:
_ —
The analysis for dead load on both panels is performed in Fig. 10.68.
_
= 49 (?2X0»
58
(6)
_ (23(03X0
(0
42.4 kN-m

For service LL on all spans:


J44_
(22X03X0
" (0
.7?_
(22X03X0
(0
(e) Unfactored Wind Loads
Considering wind moments applied at the joints (not the joint reactions), the
Fig. 10.68. Analysis for Dead Load on Both Panels. calculations for moments are made in Fig. 10.70.
Moment at x-distance from any column centerline = Mx 68
~
-
(68/3) x
Shear at end-A = 25 x 3 - 6.33 = 68.67 kN
Bending moment at x-distance from end-A = -50 + 68.67 x- 12.5 x2
;

Moment at column face = 68 (68/3) (150/1000) = 64.6 kN-m
Mid-span moments are zero for the wind load alone.
j (f) Combinations
(d) Live Load Throughout The values may be scaled from Case (c) by multiplying 1) U = 1.2D+ 1.6L+ 1.6Zr
with 22/25. Ji 2) U = X.2D + 13 W+\.0L + 0.5 Lt
The values for moments for live load on both panels are given in Fig.
10.69.
3) U = 0.9D + 1.3 W

M*

Fig. 10.69. Analysis for Live Load on Both Panels.

Shear at end-A = 60.43 kN


Bending moment at x-distance from end-A = - 44 + 60.43 x 11 x2
Negative Moments At Column Faces

Formula:
— wv—
D

For service DL: M~ = 50 - (?^(Q3^— - 42.5 kN-m


Fig. 10.70. Wind Load Analysis for Example 10.18.
= 88
(6)
= 80.^
LATERAL LOADS AND
442 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI RIO: MOMENT ENVELOPS 443
The factor of 1.3 is considered with the wind load based on the assumption
that
directionality factor is not applied on the wind loads. By plotting all these equations, the outline of the resulting curves will be the required
Due to pattern loading, the first combination will give 2 bending moment diagrams. moment envelope (Fig. 10.71). The design moments at the face of supports are as
The second combination will give 2x2 diagrams for two pattern loads and two under:
directions of wind. Similarly, the third combination will produce 2 diagrams
making a Exterior Mace = - 171.53 and +48.00 kN-m
total of 8 bending moment diagrams. Interior Afecc = —242.2 and +15.53 kN-m
(g) Equations For Bending Moment Diagrams
150
Denoting the distance along the beam from the end-A by x, the equations for bending
moment diagrams maybe written as follows: 100
50
Ext. Int.
Alfaro 0
1. M = 1.2(-50 + 68.67 x- 12.5 x12) 1st comb, LL
+ 1.6(-49 + 64.5x- 11 x2) -50
-111.3 -168.3 on left panel
= -138.4 + 185.6 x- 32.6 x2 -100
2, -M = 1.2(—50 + 68.67 x— 12.5 x2) 1st comb, LL -150
+ 1.6(- 44 + 60.43 x- 11 x2) - 104.3 -198.4 on both
= -130.4 + 179.09 x- 32.6 x2 panels -200
3. M = 1.2(—50 + 68.67 x— 12.5 x2) 2nd comb, -250
+ 1.0(- 49 + 64.5 x- 1 1 x2) -3.57 -223.4 LL on left
+ 1.3(68- 22.67 x) panel, wind -300
= - 20.6 + 117.43 x- 26 x2 on left side
4. Fig. 10.71. Moment Envelope for Beam of Example 10.18.
M = 1.2(-50 + 68.67 x - 12.5 x2) 2nd comb,
+ 1.0(-49 + 64.5 x- 11 x2) -171.53 - 55.4 LL on left The equations for shear force in case of various load combinations and pattern
loads are
+ 1.3(— 68 + 22.67 x) panel, wind as follows:
= - 197.4+ 176.38 x~ 26 x2 on right side
5- M = 1.2(-50 + 68.67 x - 12.5 x2)
+ 1.0(-44 + 60.43 x- 11 x2) -0.82 - 242.2
2nd comb, 1. V = 1.2(68.67 - 12.5 x) + 1.6(64.5 - 11 x) 1st comb, LL on left
panel
LL on both = 185.6 -65.2 x
V = 1.2(68.67 - 12.5 x) + 1.6(60.43 - 11 x) 1st comb, LL on both -
+ 1.3(68- 22.67 x)
= -15.6 + 113.36 x- 26 x2
6. M = 1.2(-50 + 68.67 x - 12.5 x2)
+ 1.0(- 44 + 60.43 x— 11 x2) -167.14 -74.17
panels, wind
on left side
2nd comb,
LL on both
I
!
!
2.

3.
-
V
+ 1.3(22.67)
x
179.09- 65.2
= 1.2(68.67 - 12.5 x) + 1.0(64.5 - 11 x)
panels
2nd comb, LL on left
panel, left wind
+ 1.3(- 68 + 22.67 x) panels, wind j = 117.43 -52 x
= -192.4 +172.31 x- 26 x2 on right side ’ 4. V = 1.2(68.67- 12.5 x)+ 1.0(64.5- 11 x) 2nd comb, LL on left
-
7. M 0 9(-50 + 68.67 x - 12.5 x2) 3rd comb, + 1.3(22.67) panel, right wind
+ 1.3(68- 22.67 x) 48.00 - 152.5 wind on left J = 176.38- 52 x
= 43.4 + 32.33 x- 11.25 x2 side r 5. V = 1.2(68.67 - 12.5 x) + 1.0(60.43 - 11 x) 2nd comb, LL on both
panels, left wind
8. M = 0.9(-50 + 68.67 x - 12.5 x2) 3rd comb, + 1.3(22.67)
+ 1.3(- 68 + 22.67 x) -119.96 15.53 wind on = 113.36- 52 x
= -133.4 + 91.27x-l 1.25 x2 right side 6. -
V = 1.2(68.67 12.5 x) + 1.0(60.43 11 x) - 2nd comb, LL on both
panels, right wind
+ 1 3(22.671
= 172.31- 52 x
LATERAL LOADS AND
MOMENT ENVELOPS 445
chapter- 10:
PROBLEMS
444 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A.
SIDD^
Note: iV = Registered number of the student.
7. = 0.9(68.67 - 12.5 x) + 1.3(22.67)
V 3td comb, left wind No = A fixed positive or negative number assigned by
= 32.33- 22.5 x ?
the instructor to get R between 600 and 1000.
8. V = 0.9(68.67 - 12.5 x) + 1.3(22.67) 3rd comb, right wind
= 91.27 -22.5 x
5
7? = -
N No, a parameter to get numerical data for the
problems.
By plotting all these equations for shear force corresponding to various loading
methods, determine the following for the given frame:
conditions, the outline of the resulting curves will be the required shear envelope (Fig. 1- Using approximate analysis
10.72). moment in span AB.
- Maximum negative moment at B3 and maximum positive at C3 and maximum positive
- Define loading for maximum negative moment
moment in span CD.
- Find the maximum column moments.
paper. The factored
The frames are on 71/200 m centers perpendicular to the plane of
live load is 5 .57? Pa and factored dead load is 67? Pa.

Fig. 10.72. Shear Envelope for Beam of Example 10.18. Tjf total
2- For the frame shown in the accompanying figure and assuming a reasonable
It is observed from this example that moment reversal may occur at negative moment height, cany out the following analysis steps:
sections due to lateral loads. Similarly, uplift on the roof may also cause moment i
reversal. The combination 0.9D + 1.3W or 0.90 + 1.6 IF is usually the most critical for t
such reversal. For columns, this combination may give the minimum axial force due to *
dead load and significantly higher moment due to wind load. This example further
endorses the ACI condition of extending and properly anchoring at least 25% of the
maximum positive reinforcement at the support if the frame is part of primary lateral
load resisting system. This ensures that a positive moment capacity of approximately ;

one-fourth the maximum positive moment capacity exists at the face of supports. .
LATERAL LOADS AND
446 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQj MOMENT ENVELOPS 447
CHAPTER 10:
a. Calculate the wind loads.
b. Analyze frame for beam AB.
W2 = 72/4 kN 9 — u2

c. Plot moment and shear envelops for beam AB using manual procedure. However ki = 87? N/mm
spreadsheet may be used to plot the curve.
m = 1 27? kg
d. Check moment and shear envelopes using any commercial computer software and j
attach computer printout. 1 k =6R kN/m
I
^ = 72/2.4 kN ^=67? kN/m
3- For free vibration of mass of 107? kgs, having a stiffness factor (spring constant, k) of i kx = 107? N/mm
107? kN/m, calculate the time period and natural frequency. Also plot the vibration for Figure for Question No.6.
the following two initial conditions: Figure for Question No.7.

a. At t = 0, u = 72/50,000 m, —
dt
=0 I
:
7- Calculate the shear force in each column and the total lateral force for the frame given
in figure for this question by using the provided response spectrum for the following
damping ratios:
b. At t = 0, u = 72/60,000 m, —
di
= 72/5,000 m/s ‘ a) f=0 b)^=5%

A rigidly connected plane frame consists of rigid floor of mass 107? kg supported by 8- Calculate the empirical time period for the shown steel moment resisting frame, located
two flexible columns, 0.57? m long, as shown in the figure. The flexural rigidity of each on a site in zone-3.
column is El = 72/100 x IO12 N-mm2. Calculate natural time period and natural m3 = 7007? kg
frequency of the frame. = 0.87? kN 7
m u K/2m fe = 7?/12xlO6N/m
mj. = 7007? kg
— U; DAK1N ''
r

k 72/2 m fe = 72/10xl06N/m
= 0.37? kN ?1 r
mi — 7007? kg
72/2 m it, = 72/6x10* N/m
5- Determine the stiffness and for the given frame.
I
9- Calculate the value of the seismic force coefficient (G) for an ordinaiy RC moment
resisting frame on stiff soil (see Figure). Use approximate value of the time period.
*2
EI (total) = R/60 x 10“ N-mm2 0.67? m

Ui
mi
EI (total) = 7?/40 x 1011 N-mm’ 0.57? m
kt

6- Determine the natural frequencies and normal modes for the structure given in the
accompanying figure.
448 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI LATERAL LOADS AND
| CHAPTER 10: MOMENT ENVE LOPS 449
10- The floor plan of a single story commercial
building located at a site in seismic zone 3
on stiff clay is given in the Figure. The RJ2 13. A two story masonry bearing wall structure of plan area R m x R m and height of each
m high masonry shear walls are load
bearing and have a self-weight of 440 kg/m2. The weight story equal to R I 2 m consists of walls on all the four sides having a mass of 0.77?
of roof to be included in
seismic calculations is 0.47? kg/m2. Determine the seismic base kg/m2. The building is located at a site in seismic zone 3 with an undetermined soil
shear and compare the
results with simplified formula. profile and the floors have a mass of 7?/6 kg/m2. Find the vertical earthquake force
distribution and compare it with the approximate distribution.

14. An eight story hospital building R. C.


frame is shown in Figure, which is to be
built in Mansehra. The columns are 0.67?
m C/C perpendicular to the plane of paper.
Determine earthquake base shear and its
distribution over each story level
considering floor finish equal to R/3 kg/m2
and partitions load of R/6 kg/m2. Outside
walls are assumed to have a mass,
distributed over the plan area, of 450 kg/m2
for all floors except the roof where it is 450
kg/m2. Slab thickness for floors 1 and 2 is
7?/3 mm and for all other floors is 72/4 mm.

11. A three story steel moment resisting frame, shown in Figure


is located at a site in zone
4 with a soil profile type Sb. The site is situated
5 km from a potential seismic source
type A. Calculate the value of the seismic base shear.
The time period of the frame by
method-A is 0.6 sec.

m3 = R/7 kg

12. A RC special moment resisting frame, with fundamental period


of
total dead load of 1507? kN, is located at a site in zone 4 with a soil 7?/300 seconds and a
profile type Sd. The
site is situated 2 km from a potential seismic source type
A. Calculate the value of the
seismic base shear.
450 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SJDDIq,
Space For Reader’s Notes
Chapter -11

DEVELOPMENT OF REINFORCEMENT

11.1 INTRODUCTION TO BOND


fo reinforced concrete beams, the flexural compressive forces are resisted by concrete, while
the flexural tensile forces are provided by reinforcement. For equilibrium of these two forces,
a force transfer or bond must exist between the two materials. In case of no bond, the bar will
be pulled out of the concrete and the tensile force, T, will drop to zero, causing the beam to fail.
Bond strength is defined as the resistive stress, against the pulling out of a steel bar from
concrete mass, developed per unit surface area of a reinforcing bar. The bond stress balances
the force present in the bar. Development length is the embedded length of bar that is
sufficient to develop maximum expected force in the bar after strain hardening (generally taken
as 1.15 fy Ab). Greater bond strength would mean smaller required development length.
Development length of a larger diameter bar increases more rapidly due to smaller surface area
compared with the area of cross-section. Splice length is defined as the lap length required to
safely transfer the force from one bar to the surrounding concrete and then back in the other
overlapped bar. It is used to extend length of a bar; although bars of different diameters may
also be spliced.

11.2 MECHANISM OF FORCE TRANSFER BY BOND

Bond is developed in a smooth bar embedded in concrete by the chemical adhesion between
the steel and the concrete and some friction between the two materials. However, when the bar
is pulled it is reduced in diameter due to Poisson’s effect. The chemical adhesion and friction
disappear after certain reduction in diameter. Smooth bars must always be anchored at the ends
by hooks or other mechanical anchorages.
In case of a deformed bar, the deformation provide bearing on the interlocked concrete besides
chemical adhesion and friction between the two materials. The latter two mean of bond are
eliminated upon application of tensile load and bond is only developed by the bearing stresses
produced in the concrete around the steel deformations. For a plain bar, chemical adhesion and
the friction between concrete and steel are the only means of bond development.

11.3 BOND STRENGTH

Each bar at the level of its theoretical cutoff should be extended for a certain length so that the
maximum force Ab fy may be developed in it, where Ab is area of the bar and fy is the yield
strength of the steel The forces are somehow to be transferred from the concrete to the steel.
This is achieved by bond between the reinforcing bars and the concrete in two main ways:
a. Chemical adhesion.
b. Bearing of deformations/ribs of bars on surrounding concrete.
452 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI ] B81SSS
The bond stress varies along WiiSS CHAPT® 11: DEVELOPMENT OF REINFORCEMENT 453 f
average bond stress is used forthe length of the bar as a complex function and generally
past by the pullout test and
design calculations. The bond was
beamtest with heavy transverse
cases is by breaking of the concrete
the
primarily measured in the '
reinforcement. The failure in such
id = mwK
14 A * J
1
0.7974
[!
without any splitting of the present within the deformations
and the bar is pulled 0-46 A 4 0-364 4
concrete around the bars. out
=
11*4 PULLOUT TEST 474 "
74
To investigate the bond Later research showed that the pullout and other tests gave more bond strengths than the actual
strength,
is projecting from the block on a steel bar is cast inside a concrete block of large size, situation because:
one side. After the concrete which a. The bearing of a big block on the surface prevents the splitting of the block due to
platform as shown in Fig.
11.1 and tensile load is applied onhardening, the block is placed on a
the bar. The load (P) at which its bigger size and the local bearing compression produced.
bar is pulled out is noted. The the
bond strength (p) may then be b. The transverse reinforcement in beam tests can actually be less and reduce the
calculated as follows:
bond strength x surface area of bond strength.
=
bar failure load
p X (n db) x Actually tensile splitting of beams is most common type of bond failure because of limited
=P Bond Stress (g)
spacing between bars, lesser cover from outside, limited transverse reinforcement, and initial

Where
p = P

= length of the bar embedded in concrete.
cracks due to other reasons.
11,5 FAILURE MODES

The bond strength (p) is Due to anchorage of deformations on the surface of steel bars, bearing stresses are produced in
found to be directly the concrete, which have a longitudinal and a radial component. When pull acts on the bar, the
proportional to tensile strength above radial component acts like an internal bursting pressure and concrete acts like a thick
of concrete or Jf! .
Y
cylinder subjected to internal pressure. If sufficient cover is not available to steel bar, ft cannot
C
Early tests indicated that p
~
numbers 10 through 29. However,
0.7974 for bar withstand the tensile stresses. Eventually, the concrete splits parallel to the bar and the
the splitting resulting crack extends up to the concrete surface. As shown in Fig. 11.2, a circle can be drawn
strength alone is approximately around a bar in which burst pressure is critical. As the force in bar is increased, this circle
0.55 (MPa).
For an ideal design, the maximum touches with the outside and crack propagates towards that end: Similarly when this circle
must l- force in steel bar touches another circle from some other bar, crack occurs between the bars. The mechanisms of
be eqUai the available ultimate bond Fig. 11-1. Experimental Setup for
strength. bond failure are explained in the sub-sections.
PulloutTest.
T A*fy = Pind^
id
Wdb
-df /
= 4 '

^d* Fig. 11.2. Cracking in Concrete Due to Pull in Steel Bar.

.\
Bond failure is a brittle failure
for 1.15 fy stress in bar which mechanism. To make development safe and ductile, we design 11.5.1 Side Split Mechanism
includes capacity reduction factor and strain
hardening in steel.
This type of mechanism occurs when the spacing between the bars is much lesser and
the splitting extends along the line of steel bars. At last, it breaks the outside cover and
^dt a part of concrete separates exposing the steel (Fig. 11.3). The bottom cover is larger
and hence the initial cracking does not propagate in this direction.
454 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQ] DEVELOPMENT OF REINFORCEMENT 455
H:
Edge cover, ce > CHAPTER
Governing cover parameter, Cb

Fig. 11.6. Pullout Failure Mechanism.

REINFORCEMENT
Fig. 11.3. Side Split Bond Failure Mechanism 11.6 EFFECT OF TRANSVERSE
diagonal cracks adversely affect the bond strength.
Transverse
11.5.2 Face And Side Split Mechanism Or Bottom Shear or flexural-shear improves bond. It also prevents splitting of
This mechanism occurs when the outside
Inside Mechanism
reinforcement reduces these cracks and hence
cover is lesser than half the spacing concrete.
in tension
between the bars. This outside splitting
occurs first weakening the remaining 11.7 development of reinforcement
concrete. Horizontal splitting may then according to ACI
1. The values of used in this chapter should not exceed 8.3 MPa
occur releasing the bars from the concrete
causing failure, as is clear in Fig. 11.4. 12.1.2.
for the required development
Side cover > bottom cover Cracks 2. According to ACI 12.2.2, the approximate expression
cs/2 > ce length is as under:
Fig. 11.4. Face and Side Split Bond
Governing cover parameter et, = ct Failure Mechanism.
id
fyV.V' .
11.5.3 V-Notch Mechanism
In V-notch mechanism, the spacing between the
.
Where, the constant K is taken from Table 11 1.
bars is much larger compared with the outside
cover and hence splitting in line of bars cannot
occur. Here first cracks perpendicular to the No. 20 and > No. 20
surface appear followed by cracks at 45° and deformed wires bars
other angles (Fig. 11.5). Hence, after this 1 1
splitting, the bar looses its anchorage and starts Given bar spacing criteria
satisfied 2.1 1.7
slipping. Fig. 11.5. V-Notch Bond
1 1
Failure Mechanism. Other cases
cs/2 » ce 1.4 1.1
Governing cover parameter = Cb = ce criteria are as under:
11.5.4 Pullout Mechanism Recommended Bars Spacing Criteria: The required bar spacing
not be less than d/,.
i. Clear spacing of bars being developed or spliced should
When concrete cover to bars is sufficiently larger on each side,
no splitting occurs but ii. Clear cover should not be less than dt-
the bars are pulled but after shearing off the concrete the length Id not less than the code
interlocking (Fig. 11.6). This iii. Stirrups or ties should be present throughout
mechanism is exactly similar to that of a simple pullout test. In
are sufficiently large. this case both cs and ce minimum.
OR
456 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SlDDlQj
1
‘ CjjAPTER if
DEVELOPMENT OF REINFORCEMENT 457

i. Clear spacing of bars being developed or spliced should not be less


than 2db. Cb Spacing or cover dimension, mm. Use the smaller of either the
ii. Clear cover should not be less than db. distance from the center of the bar or wire to the nearest concrete
The first row of the table for Ai-values, where recommended surface or one-half the center-to-center spacing of the bars or
bar spacing and cover wires being developed. There are many practical combinations
requirements are satisfied, the factor (ct + Ktr}!db of the
detailed formula explained of side cover, clear cover, and confining reinforcement, which
below is taken as 1.545 and, for the bottom row, a value of 1.0 is used for
the factor (c0 can be used with ACI 12.2.3 to produce significantly shorter
+ Ktr)/db.
t development lengths than allowed by ACI 12.2.2. For example,
3. As an alternate to the ACI 12.2.2, a more accurate method is given using bars or wires with minimum clear cover not less than 2db
in ACI 12.2.3. This
method is applicable when exact cover, spacing of bars and details and minimum clear spacing not less than 4db and without any
of transverse
reinforcement are available. s confining reinforcement, the factor (cb + Kt^ldb has a value of 2.5
and the required development length is only 0.6 times the values
'-a — 1-1 7 ~~ '
> of ACI 12.2.2.
transverse reinforcement index, a factor that represents the
I Z
g-
contribution of confining reinforcement present across the
, . c. K„
where -5 +
— T
I

The limit Of 2.5 on the term (ct +


2.5 potential splitting planes.
404
K^ldt is included to safeguard against pullout type sn
failures. Provision of this limit eliminates the need for the check of
Q-^difyl Where Atr = total area of all transverse reinforcement which is within the
previously required under ACI-89. spacing “s” and which crosses the potential plane of splitting
through the reinforcement being developed, mm2.
a. ¥( = Reinforcement location factor. Top reinforcement always have s = center-to-center maximum spacing of transverse
poorly placed concrete around it, reducing the bond. reinforcement within £4 mm.
= 1.3 if horizontal reinforcement is so placed that more than 300 n = number of bars or wires being developed along the plane of
mm of fresh concrete is cast in the member below the splitting.
development length or splice.
- 1.0 for other reinforcement. => It is permitted to use Ktr = 0 as a design simplification even if
transverse reinforcement is present.
b. 94 = Coating factor he/she
= 1.5 for epoxy-coated bars, zinc and epoxy dual coaled bars or The user may construct simple expressions for groups of reinforcement that
frequently uses. Practical implementation requires that either the user calculate £j
wires with cover less than 34, or clear spacing less than 64- Ktr)/db values be
= 1.2 for all other epoxy-coated bars or wires. based on the actual (ct + Ktr)/db for each case or a range of (cb +
= 1.0 for uncoated and zinc-coated (galvanized) reinforcement. prepared for common cases, like ACI-12.2.2.
c. The produced of ¥( and 94 need not be taken greater than 1.7. For example, in all structures with normal weight concrete (2 = 1.0), uncoated
reinforcement (94 = 1.0), No. 20 or smaller bottom bars (9( = 1.0) with /'c = 20 MPa
d. = Reinforcement size factor. and Grade 420 reinforcement, the equations reduce to:
y r =0.8 for No. 20 and smaller bars and deformed wires.
= 1.0 for larger than No. 20 bars. (420)(l .0X1.0) = 45 4
Minimum spacing and cover satisfied
e. X = lightweight aggregate concrete factor.
(2.1X1-0)720
0.75 when lightweight aggregate is used, unless fa is specified. (420)(l .0)(1.0)
= 68 4
= /«/(0.565/27 )≤ 1-0 f°r lightweight aggregate if the average Otherwise, if spacing is not satisfied
(1.4X1-0)720
splitting tensile strength (fa) is known.
= 1.0 for normal weight concrete.
458 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDD1QJ DEVELOPMENT OF REINFORCEMENT 459
CHAPTER H:
5. Excess reinforcement
ACI 12.2.5 allows reduction in development length where reinforcement in a flexural Bars in column above
member is in excess of that required by analysis, except where anchorage or
development for fy is specially required.

Reduced
— ^rg^'
As provided
11.8 DEVELOPMENT LENGTH IN COMPRESSION Fig. 11.7. Offset of Overlapped Column Bars.

Development length for deformed bars in compression is evaluated by the expression of 11.10 SPLICES OF DEFORMED BARS IN COMPRESSION
ACI 12.3, as under:
in compression is as under:
The minimum lap length for splices of deformed bars
x > 200 mm for/zc <31 MPa = 0.071 / db 300 mm for f, < 420 MPa
> = (0.13/^ -24)/’ 4 300 mm for ^> 420 MPa
/7^x0.043^4 200 mm forA > 31 MPa
= 1 when// > 20 MPa
Where, r-?'

A, provided
where /”
= 4/3 when/c' 20 MPa
splice length is to be taken
and = 0.75 if the reinforcement is enclosed with spiral reinforcement of 6 When bars of different sizes are lap spliced in compression, the
F2
mm minimum diameter at amaximum pitch of 100 mm or with 13 longer of of larger bar and splice length of smaller bar.
mm ties satisfying the code requirements at a maximum spacing of
lOOmmc/c. 11.11 DEVELOPMENT OF BUNDLED BARS
is to be
= 1.0 otherwise. According to ACI 12.4, the development length in tension or compression of bars
calculated for individual bars but increased by the following percentages:
11.9 SPLICES OF DEFORMED BARS IN TENSION
a) Bundle of three bars 20 %
The tension splices are either of Class A or the stronger Class B type. All lap splices in tension b) Bundle of four bars 33 %
must be Class B except in the following cases where Class A splices are allowed:
11.12 DEVELOPMENT LENGTH WITH STANDARD HOOKS
a, Jprov / A^, 2.0 for the entire length of the splice.
b. One-half or less of the total reinforcement is spliced within the required lap length. The ACI standard hooks, as given in an earlier chapter, are reproduced in Fig. 11.8.
The required splice length (/st) for the two types is evaluated as follows:
db
Splice Type Splice Length
124
Class A 1.0 /a 300 mm f
Class B 1 -3 /d 300 mm - 54
44 for bars up to No.25 / Radius same as for 90° Hook
Where, /a is the development length in tension without reduction for Apmv. /
the 300 mm minimum. .4^ and without - for bars No.29 to No.36
a) Standard 90° Hook
b) Standard 180° Hook
Splices in adjacent bars must be staggered by at least 600 mm. Further, the column Fig. 11.8. Standard ACI Hooks.
longitudinal bars must be lapped as illustrated in Fig. 11.7.
,, DEVELOPMENT OF REINFORCEMENT 461
460 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIqi
CHAPTER in
4
According to ACI 12.5, the development length (£<ih) for deformed bars having standard
hook is Table II.2. Bending Moment at Various Locations Along
given as under:
the Span of a Simply Supported Beam Subjected to LIDL.
^dh - FiF] —±=db 8 db Location, x Bending Moment Percentage w.r.t.
-^rnax
> 150 mm
//2 0.125 w? 100%
where = 1.2 for epoxy-coated reinforcement Z/4 0.09375 w f 75%
= 1.0 for ordinary reinforcement
£/5 0.08 w £2 64%
2 = 0.75 for lightweight aggregate reinforcement £/7 0.06122 w £2 49%
= 1.0 for normal weight concrete £/10 0.045 w t1 36%

Ft = A‘ req-
4 provided 2 ACI 12.10.3 requires that reinforcement is to be extended beyond point where it is
theoretically no longer required to resist flexure by at least the dimension, £a, given
and F2 = 0.7 for No.36 or less size bars having hooks with side covers (normal below:
to the plane of hook) at least equal to 65 mm and for 90° hook with
cover on bar extension beyond hook of at least 50 mm. ta = additional embedment length at support, or at point of inflection, mm.
1 - larger of 1) effective depth of member, d
= 0.8 for No.36 or less size bars having 90° hooks enclosed within tics
2) 12 db
or stirrups perpendicular to the bar being developed having a
maximum spacing of 34 along development length (/dt) of the hook A diagonal tension crack in a flexural member without stirrups may shift the location of
or enclosed within ties or stirrups that are parallel to the bar to be the calculated tensile stress approximately a distance d towards the point of zero
developed having a maximum spacing of 3db along the length of the moment. When stirrups are provided, this effect is less severe, although still present up
tail extension. to some extent. This requirement needs not to be satisfied at supports of simple spans
and at free ends of cantilevers.
= 0.8 for 180° hooks of No.36 or less size bars, which are enclosed
within ties or stirrups perpendicular to the bar having a maximum 3. Curtailment Of Negative Reinforcement
spacing of 3db within £dh. At least one-third the M- reinforcement at a support shall have an embedment length
= 1.0 otherwise. beyond the P.I. equal to the following:

11.13 BAR CUT-OFF REQUIREMENTS t p) d'


! greater of < 2) 124
For a simply supported beam subjected to uniformly distributed load, the bending moment | L3) 4 / 16 (4 is the clear span)
expressed as a percentage of maximum moment at some predefined sections is given in Table
At interior supports of deep flexural members, M- reinforcement should be continuous
This table provides the approximate idea about the percentage of bars that may be curtailed at a with that of the adjacent spans (ACI 12.12.4).
:
particular distance. However, extra development must be provided for the curtailed bars. I 4. Minimum Positive Reinforcement At Support
Fraction of M* reinforcement to be carried into the support at least 150 mm from face
1. According to ACI, there are following critical sections for checking the development of of the support (ACI 12.11.1) is as under:
bars: {

Simple Spans Continuous Spans


a. Points of maximum stress. I Fraction of M* reinforcement ’/3 14
b. Points where reinforcement is terminated /bent. |
462 CONCRETE STRUCTURES PART - I DEVELOPMENT OF REINFORCEMENT 463
AUTHOR: ZAHID A. SIDDiqj CHAPTER 11:
If the above member is part of a primary lateral load resisting system, the positive i. Bars-a at face of support must have length greater than or equal to £i-
reinforcement should be anchored in the support to give full fy strength at the face Bars-b must have at least full development length (^d) beyond the point on
Of ii.
support. negative moment envelop corresponding to flexural capacity of only bars-b.
5. ACI 12.10.4 requires that the continuing reinforcement should have an Bars-b must extend beyond point of inflection on negative moment envelop for
embedment
length not less than beyond the point where some reinforcement is terminated
iii.
or a distance greater than or equal to larger of d, 12d* and /16.
bent. The distance must start from the point where the discontinued reinforcement iv. Bars-c must at least have full development length starting from mid-span to the
is
theoretically not required. j point of their curtailment.
Peak stresses exist in the remaining bars wherever adjacent bars are cutoff, or bent, 1 v. Bars-c must be extended by larger of d and 12db from point on positive moment
in envelop corresponding to flexural capacity of bars-d alone.
tension regions. If bars are cutoff as short as the moment diagram allow, these peak
stresses become the full fy, which requires a full extension. vi. Bars-d must have full development length from point on positive moment
6. envelope corresponding to flexural capacity of bard-d alone up to their end
Termination of bars in tension zones is to be avoided. However, if it is compulsory to
terminate the bars in tension zone, at least one of the following conditions (ACI within supports.
i
12.10.5) is to be satisfied: vii. Diameter of positive bars is limited at point of inflection on the positive
moment envelope, as discussed later.
a* F^, cutoff (factored) <
viii. Bars-d must extend from face of the support into the support satisfying ACI
The design shear strength (^F„) includes the contribution of both concrete and
steel stirrups.
12.11.4, ACI 12.11.2 or £dc for compression if these bars are used as
i
compression reinforcement.
b. Excess stirrups (above that required for shear and torsion) are provided within %
distance from the termination point
i

Excess stirrup area Av 0.41^5

^max

Where p^ A3 at cut - off


Total As at section
c. The following conditions are satisfied:
Bar nos. 36
P.I. of positive Point corresponding to
As after cutoff > 2 Alf req. capacity of bars-d
Fil, cutoff % envelope
This code provision does not apply to splices and bars bent up and anchored in Fig. 11.9. Typical Half Span of a Continuous Beam
compression zone. It is important to not that reduced shear strength and loss of Showing Flexural Reinforcement.
ductility in case the bars are cutoff in tension zone have been reported.
7. Limit On Positive Reinforcement Diameter
7. Referring to Fig. 11.9, the following requirements are to be checked:
According to ACI 12.11.3, at simple supports (SS) and inflection points (PI), positive
Let,
reinforcement (for M+) diameter is limited such that
bars present at support for negative moment = a and b
Negative bars curtailed during first step
Negative bars curtailed later
=a
= b 4 K^-+ta
V
Bars at mid-span for positive moment = c and d This means that the rate of development of bond should be more than the rate of
Positive bars curtailed =c
Positive bars continued up to the support =d increase of moment. The value of the constant K is defined as under:
DEVELOPMENT OF REINFORCEMENT 465
JR 11:
464 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDlQi
1. Moment developed between P.I. to the critical section, showing under-strength
for portions in-between in case of parabolic variations of the applied bending
K = 1.3 when the ends of reinforcement are confined by a compressive moments.
reaction (for example at a support). The compressive stresses
present in concrete increases the bond strength as the splitting of 2. Actual moment capacity required.
concrete becomes difficult. 3. Moment capacity required to be developed tangent to the actual B.M. diagram.
K = 1.0 = (from basic mechanics)
otherwise. Slope of B.M.D. required Va
_
at a support =
fo P.I
at =
embedment length beyond center of support.
greater of 1) d 2) 12d6. Further, this should not exceed the
; =
Af
—-
x
„ , . , .
from basic mechanics,
dM „
—ax
actual bar extension provided. K =
X
This condition needs not to be satisfied if positive reinforcement is terminated by a ;
standard book or mechanical anchorage. Af,
or *
Discussion OnACI 12,11.3
Even if there is no cutoff, on one side of P.I. there is positive moment producing Within this distance ‘x’, bars should be fully developed, so,
tension in the same bars whereas on the other side there is negative moment causing
compression. Bars may slip if development is not provided faster than the maximum
possible rate of increase of required moment capacity (K„) to utilize full moment
capacity; ; This development length is definitely required for straight line B. M. diagrams such as
Let, that for concentrated loads (see Fig?. 11.11 and 11.12). For parabolic moment
diagrams, the required rate of development of moment from P.L is not so sharp and the
j

Mn = section capacity at P.I. considering all bars to be yielding. J code allows the minimum distance from the P. I. to a point with fully developed
= section capacity at the critical moment section. ;
Vu = maximum factored S. F. at the P.L : maximum moment capacity to be increased by £a, which is larger of d and 12 dt.
Fig. 11.10 shows positive portion of moment envelop for a typical beam. It is apparent
from the figure that the slope required for the development of full bending moment
according to the shape of the diagram is equal to the slope of tangent to the B. M.
diagram at P.I., which is the shear force Pu. In this figure, x is the required maximum
length in which full development should occur. The three different lines/curves
represent the following:

Same situation arises at the cutoff points. However, if the bar diameter is satisfied at
B.M;D is
P.L, the criterion will automatically be satisfied at the cutoffs because slope of
very steep at P.I. and relatively flat at cutoffs. Code uses the value of in the
Fig. 11.10. Moment Envelop and Rate of Bond Development. numerator to make the expression further on the conservative side.
466 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI DEVELOPMENT OF REINFORCEMENT
CHAPTER 11-' 467
If the given criterion is not satisfied for a particular reinforcement, the bar diameter
is to
be reduced. The criterion is checked once more and the process is repeated
until the b) gtn-tailed ToqBars
criterion is satisfied. j Interior face of exterior support.
ii. Interior face of interior support.

cj rnntinuous Bottom Bars


At supports, if the bars are to be discontinued.
At theoretical section where strength of these continuous bars alone is sufficient
to
ii
resist applied bending moment

d) Curtailed Bottom Bars


i. Maximum positive moment section.
ii. Section at which these bars may theoretically be curtailed (same as
c-ii abdve).

Rule 2: Minimum Extension Of Bars At Ends For Reasons Other Than


Fig. 11.12. Comparison of Rates of Full and ACI Bond Development. Development Length In Tension
to a practical eiitoff
The theoretical location at which the bars may be curtailed must be shifted
moments
In Fig. 11.12, the two numbered lines refer to the following: location towards less moment region to provide safety against shifting of maximum
Curve— 1 = rate of development of bars necessary if bars extend only to P.L due to variation in expected loads. The following points must be kept in mind:
= line tangent to the B.M.D. i. Bottom curtailed bars - extend lesser of ri and 12rib from theoretical cutoff-
Curve- 2 = rate of development required if bars extend by 4 beyond P.l. point (ACI 12.10.3).
This problem does not arise for negative moment. The negative moment is increasing extend lesser of ri and 12rib from theoretical cutoff
ii. Top curtailed bars
very rapidly towards the support, but the bar is fully anchored on that side and no point (ACI 12.10.3).
reversal of forces occur nearby like at PJ. Towards the side away from the support,
required negative moment capacity is reducing very quickly even as compared with iii. Top extended bars - extend lesser of ri, 12rib and 4/16 from the point
straight line assumed for gain of strength due to development of bars. of inflection (Section a-iii in rule 1).
iv. Full-length bottom bars - extend by at least 150 mm in to the
11.14 PROCEDURE FOR CURTAILMENT IN CONTINUOUS BEAMS supports (ACI 12.11.1) or 4 for compression if
used as compression reinforcement or to satisfy
Before discussing the curtailment for continuous beams, two terms need to be defined. requirements of structural integrity in perimeter
Curtailed or discontinued bars are the bars that are discontinued at a particular location along beams (ACI 7.13). This is required for critical
the length of the beam. Continuing bars are the bars left out of a group of reinforcing bars section c-i defined in rule-1.
when certain number of bars is curtailed. Following rules may be used curtail the beam
reinforcement: v. Top bars at exterior support - extend by 4h into the support with standard hook
at the end (ACI 12.12.1). If available space does
Rule 1: Critical Sections To Be Checked not provide the length 4h, extend the length
a) Extended Top Bars beyond hook to make the total length equal to the
development length in tension. See Fig. 11.13.
i. Interior face of exterior support.
ii. Theoretical location at which the capacity of the continuing bars alone is sufficient to vi. In case of primary lateral load resisting frame, bottom bars must develop full _/y in
resist all the applied negative moment. tension at the face of supports (ACI 12.1 1.2).
iii. Point of inflection of negative moment envelope.
468 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDD[qi
DEVELOPMENT OF REINFORCEMENT 469
CHAPTER It;
ji. According to ACI 7.13.2.2, for perimeter beams, at least one-sixth of the maximum top
or must be spliced at or near the
reinforcement at a support must be either continuousRule-4.
mid-span combined with provision of stirrups as m

Rule 6: Limit On Positive Steel Diameter


(SS)
according to ACI 12.11.3, the positive bottom bars at simple supports
As already stated, following condition:
and inflection points (PI) must satisfy the
Fig. 11.13.
H M
Development of Top Steel at Exterior Support.
rate of development of bond is greater than the rate of
Rule 3: Provision Of Full Development Length This condition makes sure thatofthe
the constant K is defined as under:
(ACI 12.10.2 & 12.10.4) increase of moment. The value
reaction (for
The curtailed bars must have full development length in tension (fd) K - 1.3 when the ends of reinforcement are confined by a.compressive
of maximum bending moment and continuing bars must measured from the point example at a support).
have full development length in K = 1.0 otherwise.
tension (£d) measured from the point of theoretical cutoff. The critical
sections are a(i & ii), 6(i length measured from the centerline of
& ii), c(ii) and d(i& ii).
The distance la is defined as the additional embedment
supports or points of inflection.
Rule 4: Minimum Bottom Steel At Supports
la at a support = embedment length beyond center of support.
the
According to ACI 12.11.1, one-third in case of simple supports
and one-fourth in case of ta at P.I = greater of 1) d 2) 12<4. Further, this should not exceed
continuous supports of the positive reinforcement must be continued and actual bar extension provided.
anchored into the
.support. For structural integrity requirement of continuous beams, this
steel must satisfy the is terminated by a standard
following requirements (ACI 7.13.2.3 and 7.13.2.4): This condition needs not to be satisfied if positive reinforcement
satisfied, one of the following solutions
i. The minimum number of bars must be two. book or mechanical anchorage. If this condition is not
is possible:
ii. If U-shaped stirrups enclosing the top bars continuous over the
support by 135° hooks a. The positive reinforcement diameter may be reduced.
or one piece closed stirrups with not less than 135° hook around one of the
top bar, extend the steel into the support by a minimum of 150 continuous or mechanical anchorage.
mm (without b. The positive reinforcement is terminated by a standard hook
requirement of splicing of bottom bars). and carry out
iii. If above types of stirrups are not provided, the bottom steel Example 11.1: Calculate the amount of positive and negative reinforcement
must either be continuous or spliced near the support with
extended into the support
detailing of these steels for the beam shown in Fig. 11.13. Use fc'
- 25 MPa,^, = 420 MPa,
aclass-A tension splice, a
mechanical splice or a welded splice. Further, at non-continuous supports, clear cover = 40 mm, #13 stirrups and minimum transverse
reinforcement throughout the
this steel load = 23 kN/m. Use
must end in a standard hook. length. Dead load including the self-weight = 40 kN/m and live
approximate formulae for development length and consider the
width of the support equal to
The perimeter beams have additional requirements for structural integrity.
300 mm.
Rule 5: Extension Of Top Steel For Structural Integrity
Solution:
i. According to ACI 12.12.3, at least one-third of the total tensile
reinforcement provided wu = 1.20+ 1.6 Z
at a support must have a minimum embedment length by the point
largerof:
of inflection equal to
120
- == 1.2 x 40 + 1 .6 x 23
48kN/m
= 84.8 kN/m
a) d
In Fig. 11.13, x may be
b) 12di, The three possible pattern loadings are shown in Figs. 11.13 to 11.15.
c) f„/16 evaluated as under:
-96 +250.65 x - 84.8/2 x2 =0 => x = 0.412 m
470 CONCRETE STRUCTURES PART- 1 AUTHOR; ZAHID A.
SIDDlQj chapter 1 1; DEVELOPMENT OF REINFORCEMENT 47 1

450

525

Fig. 11.13. First Pattern Loading for Example 11.1.

- 169.60 + 264.04 x - 84.8/2 x2 = 0 => x = 0.727 m


d = 525-53- 10 = 462mm
dmin
525 y 0.205/' b
~
I 274.5x10* 345 mm
V 0.205 x 25 x 450
** dmin OK
Design For Positive Moment
274.5x10*
bd2 450x4622
= 2.858 MPa
p = 0.0082, A+s = 1705 mm2, 4 # 22 + 1 # 19
2 # 22 may be curtailed.
Design For Negative Moment Steel
169.6x10*
bd2 450x4622 = 1.766 MPa
p = 0.0049, A’s = 1019 mm2, 2 # 19 + 3 # 16 (2 # 16 may be curtailed).
Fig. 11.14. Second Pattern Loading for Example 11.1.
Development Lengths According To ACI 12.2.2
- 169.60 + 162.84 x - 48/2x2 = 0 x = 1.285 m The bar spacing of Fig. 11.17 satisfy the minimum values.
472 CONCRETE STRUCTURES PART - 1 DEVELOPMENT OF REINFORCEMENT 473
AUTHOR: ZAHID A. SIDDiqj CHAPTER 1 1:
Check For Bar Spacing
(450 - 2x53 - 16)74 = 82 nun (c/c) 450-2x53-2x19-3x16 = 64mm
At top: clear spacing, s ~ 4
534,
> db and 25mm (OK)
3-#I6 + 2-#19" 450 - 2 x 53 - 4 x 22-19 = 59mm
At bottom: c lear spacing, s = 4
#13 > db and 25mm (OK)
Rvamine Cutting 2 - #16 Bars In W Region

_
(Using Load Pattern #2)
Capacity Of Section With Continuing 2 - #19 + 1 - #16 Bars
-
^(2 #19 + 1 - #16)
</=525-40-13-19/2 =

(0.35X25X450) *
-
(2 #19 + 1 - #16)
= 767 mm2
462 mm

= (0.9)(767)(420)(462 - 34/2) /106 = 129.02 kN-m


Fig. 11.17. Reinforcement Details for Positive and Negative Moment Sections.
Point (T.C.P)
Distance From Maximum Moment To Theoretical Cutoff
The bar spacing of Fig. 11.17 satisfy the minimum values. Left of Support B
Development Lengths According To ACI 12.2.2 Referring to Fig. 11.18, we get.
According to ACI 12.2.2, the approximate expression for the required development
length when code bar spacing criteria are satisfied is as under:
129.02 kN-m (~ I* 'kN 169.60 kN-m

id XL Iff84
Support B.
Where, Fig. 11.18. Free Body Diagram of Span Between Left T.C.P. and
Reinforcement location factor.

A
Coating factor.
Lightweight aggregate concrete factor. 129.02 - 48

2
-
+ 162.84 xL 169.60 = 0 xL = 0.259 m

No. 19 top bare id —®=(1.3X1«X19)'


(2.1)(1)V25V
988 mm
A
Right of Support B
Referring to Fig. 11.19, we get,
84.8 kN/m
No. 16 top bar id 129.02 kN-m f W 44 44 W 4 Freeen(i

832 mm
No. 22 bottom bare id
1.7(1)725 7 Fig. 11.19. Free Body Diagram of Span Between Support and Right T.C.P.
1087 mm Let, x = distance from the free end
development OF REINFORCEMENT 47 5
474 CONCRETE STRUCTURES PART -1 AUTHOR ZAHID A. SIDDjqj CHAPTER U:
d= 0.462
1.285 + larger of 12 4, = 0.228
84.8x2 <|a stance of cut-off from
supports =
2— = 129.02 => x = 1.509 m £^16=^=0.344
16
xr = 2.0- 1.509 = 0.491 m = 1.285 +m0.462
Cutoff Location For Right Side = 1.747 vs.(di = 0.988m
19

The cut-off location may be found by considering Fig. 11.20. I developed at he critical section.
All the bars are fully
Distance of cut-off point from support B = greater ofxR and £d for #16 Sav cutoff is provided at 1.75
m left of support B
+ greater of d and 12 db
of d, Ylds, or £„/16
= greater of 0.491 and 0.839 least 1/3 M reinforcement must continue by a distance larger
Note: At here.
+ greater of 0.462 and 0.228 beyond P.I., which is already satisfied
In M* Region
= 1.301m
Examine Cutting 2 - #22 Bars
Continuing 2 - #22 + 1 - #19 Bars
Capacity Of Section With
A,(2 #25 + 1 - #19)=
1058 mm2
-
£/=525-53-22/2 = 461mm _
(1058X420) = 46.5 mm
a = (0.85X25X450) t

46.5/2)/106
(2 - #25 + 1 - #19) = (0.9X1058X420X461 -
= 175.1 kN-m
Fig. 11.20. Cutoff Location Right of Support B. Point To T.C.P
Distance From
the distance of cut-off moment from the maximum moment point
Say cutoff at 1.35 m right of support B It is assumed that x is order to evaluate x,
maximum moment section, shear force is zero. In force
Check That Continuing Reinforcement Is Developed (Fig. 11.21). At point because it will eliminate the shear
moment may be taken about the cut-off
Distance available on the right side = 2.0 - greater of x« and £d for #16 present there from the calculations.

£d# 19 = 0.988 m <


= 2.0-0.839 1.161 m
1.301 m

OK
ZM = Oat point D =>
175.1 + -274.5 = 0
In case sufficient space is not available, hooks are to be provided. 2
Cutoff Location For Left Side (Beyond P.L) x = 1.531 m on either side of M
Distance of cutoff of 2 - #16 bars = greater ofxx and £d for #16
— greater of 0.259 and 0.832
Distance of PI where all the bars may be curtailed
— 1.285m
= 0.832 m ‘ 175.1 kN-m
84.8 kN/m
444444 4 =CJ 274.5 kN-m
Zero
There is no considerable saving in curtailing 2 - #16 bars on the left side. All the bars
may be curtailed at the P.L
Distance of cut-off from support B = distance up to point of inflection plus the
Any direction
'
minimum allowed extension. Fig. 11.21.
I
development of reinforcement 477
CHAPTER 1 1:
476 CONCRETE STRUCTURES PART - 1 AUTHOR ZAHID A. SIDDiqi
Point Of Inflection
£a =461 mm
Check That Curtailed Steel (2#22) Is Fully Developed Maximum Moment Section
-
M (2 #22) = 143.36 kN-m
= 215.71 kN
Available distance = 1.531 = 250.65 - (84.8)(0.412)


Vu
^d#22 = 1.087 m < 1.531m OK M, + . 143.36x1000 + 461 = 1126 mm
ari > £^22 = 1087 mm
Distance From Mmu Point To Actual Cutoff Location
The bars must extend up to larger of the following distances from the maximum
moment section. Figure 11.22 shows the final detailing of reinforcement for the example beam. Hanger
bars and transverse reinforcement are not shown
for clarity of diagram.
..
1) . . , =
x + larger of-4 12z4
d
_12(22)
— 461mm
= 264mm ,
= 1.531 + 461 = 1.992
,
m 2#19 + L#16
1000
2) £</#22 = 1.109 m
Distance from left support = 2.544-1.992 = 0.552 m
Say 0.55 m from center of left support
Distance from right support = (5.5 - 2.544) - 1.992 = 0.964 m
Say 0,95 m from center of support B
Check That Continuing Steel (2#25) Is Fully Developed At TCPs Fig. 11.22. Final Detailing of Reinforcement.
Distance of theoretical cut-off point from left support = 2.544 1.531
= —
1.013m Example 11.2: Design the beam shown in Fig. 11.23,
loading and correcting moments to face of supports.
without considering any pattern
Propose curtailment of bars and locate the
£d#22 = 1.087 m > 1.013 m It is better not to curtail bars on left side. MPa, fy = 420 MPa, clear cover = 40
=
cut-off points for this curtailment of bars. Use fe' 20 throughout the length.
Distance of theoretical cut-off point from right support = (5.5 - 2.544) - 1.531 mm, #13 stirrups and minimum transverse reinforcement
= 1.425 m Solution:
On the right side, adequate space is available and there is no end of beam problem.
d = 525-75 = 450 mm
Note: At least 1/3 M* reinforcement at simple end and % M* reinforcement at continuous end [ K
must extend into supports by a distance larger than or equal to 150 mm. dain y 0.205/; b
CHECK FOR M* STEEL DIAMETER = I 242 x10s
~
_ 362 mm
Support B \ 0.205x20x450
d > dmin OK
Say£a = 150 mm (embedment beyond center of support)
1 75 1 _
Positive Moment Steel
M(2-#22 + l -#19)

Vu =
=
250.65 kN
= 194.6kN-m

bd2
s_
242x10s
= 450x450 = 2.66 MPa
p = 0.0078, A+s = 1580 mm2, 2 # 25 + 1 # 29
1.3 +4 = 1.3 + 150 = 1159 mm > ^22 = 1087mm 1 # 29 may be curtailed.
OK
478 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI CHAPTER 11: DEVELOPMENT OF REINFORCEMENT 479

J5 kN/m The bar-spacing of Fig. 11.24 satisfy the minimum values.


450 Development Lengths According To AC! 12.2.2
4 4 4 44 ^<44444444444
(420)
B C 525 No. 19 top bars £d (1.3X1.0X19) = 1105 mm
(2. l)(l)5/20

No. 25 bottom bar £d —-^=(1.0X1.0X25)


1.7 (1)5/20 V = A A 7 1382 mm

No. 29 bottom bars £d ,oXl .0X29) = 1602 mm


1.7 (1)5/20 V A A 7
Development Lengths According To AC! 12.2.3 With =0
For #19 bars, ce = 62.5 mm, c,/2 = 54 mm Cb = 54mm
=2.84 > 2.5; take ^-=2.5
db d»

Fig. 11.23. Data for Beam of Example 11.2.


For #25 bars, ce = 65.5 mm, cJ2 = 79.7 mm, —dt =2.62 (take —db =2.5)
Negative Moment Steel
For #29 bars, ce = 67.5 mm, cjl = 79.7 mm, —
dt
= 2.33

170x10s =
1 420
bd
=
TTA
450x450
= 1.87 MPa fj
1.1 (1)5/20 (ct/db) db

p 0.0054, A s = 1094 mm2, 4 # 19 (2 # 19 may be curtailed).
The required steel reinforcement is shown in Fig. 11.24.
For #19 top bars £d = (85.38)(1.3)(0.8)(19)/2.5 = 675 mm
For #25 bottom bars £d = (85.38)(1.0)(1.0)(25)/2.5 = 854 mm
For #29 bottom bars £d = (85.38X1.0)(1.0X29)/2.33= 1063 mm
These values are considerably lesser than the approximate values and will be used here.
Examine Cutting 2 - #19 Bars In M~ Region
Capacity of Section With Continuing 2 - #19 Reinforcement
A/2- #19) = 570 mm2
d— 525 —53 - 19 /2 = 462 mm
(570X420)
(0.85X20X450) =
31.3 mm

- -
(450 - 2x53 25)/2 159.5mm (c/c) M (2 - #19) = (0.9)(570)(420)(462 - 31.3/2) /106 = 96.2 kN-m
Fig. 11.24. Reinforcement Details for Positive and Negative Moment Sections.
480 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI DEVELOPMENT OF REINFORCEMENT 481
j CHAPTER1!:
Distance From Maximum Moment To Theoretical Cutoff Points (T.C.P) f
Left of Supports by a distance larger of d, \2db, or 4/16
Note: At least 1/3 M reinforcement must continue

;
Referring to Fig. 11.25, we get, t ' "
beyond P.L, which is already satisfied here.
Cutoff Location For Right Side

The cut-off location may be found by considering Fig. 11.27.


96.2
kN-ci^J, Xl 264.7 kN
170kN-m

Fig. 11.25. Free Body Diagram of Span Between T.C.P. and Support B.

x2
96.2 - 85
^- + 264.7 xt-l 70 = 0 => xt = 0.29m
Two of the bars may be curtailed to the left of support while the other two may continue
as hanger bars. Alternatively, all bars may be curtailed at the point of inflection and
separate hanger bars may later be provided. This second option is adopted here. Fig. 11.27. Cutoff Location Right of Support B.
Right of Support B
Distance of cut-off point from support B = xR + greater of d & 12 db
Referring to Fig. 11.26, we get, 0.50 + greater of (450 & 228)mm
=
= 0.95 m
170kN-m< + + +l "X 96.2 kN-m
‘ 4 = 0.675 m for #19 bars Say cutoff at 1.0m right of support B
170.0 kN Check That Continuing Reinforcement Is Developed
H * 2.0-0.50 = 1.5m .’.OK
Fig. 11.26. Free Body Diagram of Span Between Support and Right PI.
4« 19 = 0.675 m <
In case sufficient space is not available, hooks are to be provided.


170 170x« +
85x2
2
--96.2 = 0 => xR = 0.50 m —
Examine Cutting 1 #29 Bars In M* Region
Capacity Of Section With Continuing 2 - #25 Reinforcement
Cutoff Location For Left Side (Beyond P.L) ?
4X2-#25) = 1014mm2
Let us curtail all the four bars at the P.I. because there is no significant distance between
the cutoff points for 2 bars and all the steel.

J =0.450
'

f __
d = 525 - 53 - 25 / 2
</=525-53-25/2
a-
=
(1014X420)
(0.85X20X450)
_
- 459 nun
55.7 mm
Distance of cut-off from support B = 0.73 + larger of < 12 <4 = 0.228
M (2 - #25) =
$M„(2-#25) (0.9)(1014)(420)(459 - 55.7/2)/106 = 165.3 kN-m

= 0.73 + 0.45 = 1.18 m


4/16 = —
16
= 0.344 M™.. Point To T.C.P
Distance From MmBX
It is assumed thatx is the distance of cut-off moment from the maximum moment x,
point
uk 4« is = 0.675m
(Fig. 11.28). At maximum moment section, shear force is zero. In order evaluate
to
(OK) it will eliminate the shear force
Sav cutoff at 1.2 m left of support B moment may be taken about the cut-off point because
present there from the calculations.
482 CONCRETE STRUCTURES PART - 1 DEVELOPMENT OF REINFORCEMENT 483
AUTHOR: ZAHID A. SIDDIQj CHAPTER 1 E
Sm=o =>
fV18-3-7xl0-0>|
165.3 +

x = 1.34 m
——
85 x“
2
-242 =0
on either side of M
13 +4 = 1.3

Point Of Inflection
202.8 J
+150 - 1327mm > ^25 =854 mm
OK

- = 459 mm
165.3 kN-m W 4 4 4* : M, = 183.7 kN-m
242 kN-m
= Vu = -
264.7 (85)(0.73) = 202.65 kN

Any direction i +ln = 18-3-7-X


202.65
+459 = 1365mm > =854mm
1 '
V.
OK
Fig. 11.28. Cut-Off Location on Either Side of Maximum Moment Section.
; Note: Inorder to terminate bars in tension zone, ACT 12.10.5
must be satisfied.
Now bars must extend up to larger of the following distances from the maximum example beam. Hanger
moment section. Figure 11.29 shows the final detailing of reinforcement for the
clarity of diagram.
bars and transverse reinforcement are not shown for
n
1) xr + larger
&
nf
ok |a
»
= 12<29) = 348mm
= Am
459mm —
- 1.34
1 14 + 459 - 1l.ouni
1000
“■
«0 m
1.20 m,
l< —»1
2#19

2) frf#29 = 1063mm = 1.063 m


2#19 0
Actual cutoff distance from left support = 2.386- 1.80 = 0.584 m |<r
1#29
Sav 0.5 m from left support
13 m:
Actual cutoff distance from right support = (5.5 - 2.386)- 1.80 = 1.314 m 0.5 m
2#25

Sav 1.3 m from support B 5.5 m 2.0 m


Check That Continuing Steel (2#25) Is Fully Developed At TCPs
Fig. 11.29. Final Detailing Of Reinforcement
Distance of theoretical cut-off point from left support = 2.386-1.34
= 1.046 m
4i#25 - 854 mm = 0.854 m < 1.046 m OK
On the right side, adequate space is available and there is no end of beam problem.
Note: At least 1/3 M4 reinforcement at simple end and % M4 reinforcement at continuous end
must extend into supports by a distance larger than or equal to 150 mm.

CHECKFOR M+ STEEL DIAMETER


Support

Say ta = 150 mm (embedment beyond center of support)


A4(2-#25) = = 183.7 kN-m
P. = 202.8 kN
484 CONCRETE STRUCTURES PART - I
AUTHOR: ZAHID A. SIDDIQ]
CHAPTER 1 1: DEVELOPMENT OF REINFORCEMENT 485
Space For Reader’s Notes
PROBLEMS

Note; N = Registered Number.


JV0 = A fixed positive or negative number assigned by
the instructor to get R between 600 and 1000.
R = N~ No, a parameter to get numerical data for the
problems.

1. Design the beam shown in the Figure, without considering any pattern loading and
correcting moments to face of supports. Locate the cut-off points for the proposed
curtailment of bars. Use // = 28 MPa,^ = 420 MPa, clear cover = 40 mm, #13 stirrups
and minimum transverse reinforcement throughout the length.

kN/m
0.8R
4 4 4 4 4 4 4 4 4444444444
R

Mm
486 CONCRETE STRUCTURES PART -I il
AUTHOR; ZAHID A. SIDDIqi
Space For Reader’s Notes Chapter- 12

TWO-WAY COLUMN SUPPORTED SLABS

12.1 INTRODUCTION
The slab, which rests on supports on all the four sides and has longer to shorter span ratio
lesser than 2.0 is called two-way slab. The supports may be beams cast within the slab. If the
supports are incorporated in both directions within the depth of the slab itself with or without
projected beams, the resulting slab system is called two-way column supported slab system. A
general classification of two-way slab systems is given in Fig. 2.1. If there are no projected
beams arid the slab alone is directly resting on the columns, the resulting slab system is called a
flat plate. If some portion of the slab over the columns and surrounding area is constructed
with greater but constant thickness or if the column is gradually enlarged like a cone or
pyramid at the junction with the slab, the resulting system is called flat slab.
The thickened slab near the column is called drop panel. Flaring up to the top of the column is
called column capital. The drop panel commonly extends about one-sixth of the span each
way from the column, giving extra strength in the column region while minimizing the amount
of concrete at mid-span.
For larger spans, the thickness of slab required to transmit the vertical loads to the columns
exceeds that required for moments in the central region between the columns. Further, for
positive moments alone with very less shear, most of the concrete on the tension side is not
useful. This actually increases self-weight with no strength contribution. A solid flat slab
becomes uneconomical in such cases, as the material is not efficiently utilized. To lighten the
slab, reduce the slab moments and save material, the slab at mid-span can be replaced by
intersecting ribs. However, near the columns, full depth is retained to transmit loads from the
slab to the columns. This type of slab is known as a waffle slab or a two-way joist system,
where the cavities in the slab are made with fiberglass or metal dome forms. Using waffle slab,
the cost of material is reduced but the labor cost and the cost of formwork is increased. The
total cost of waffle slab will approximately become lesser when the total depth of solid slab
required exceeds 180 mm.
sum of slab panelsides(m) x 2 x 1000 >
180
or sum of slab panel sides (m) 17
Uris means that a square panel 9m x 9m or larger and rectangular panel 10m x 7m or larger
may be designed as a waffle slab.
There are two standard types of domes used as formwork for flat slab construction shown in
Fig. 12.2. The slab thickness over top of the dome is usually 75 or 1 15 mm. The approximate
volume of each type of dome is given in Table 2.1.
488 CONCRETE STRUCTURES PART - 1 TWO-WAY COLUMN SUPPORTED SLABS 489
AUTHOR: ZAHID A. SIDDIQI CHAPTER 12:
to get the required shape, which are known as
o 750 x 750 mm square domes with 75 mm flanges. The flanges of adjacent domes
form loads. Usually, removable metal forms are used are used between joists, which are left in place to
150 mm wide joist ribs at 900 mm center-to-center spacing. These are available fillers or pans. In other cases, clay tile fillers
in pans are available in opening sizes of 475, 600
depths of 200, 250, 300, 350, 400 and 500 mm. become part of the ceiling. Standard square
The corresponding center-to-center spacing of
o 475 x 475 mm square domes with 62.5 mm flanges. The flanges of adjacent
domes and 750 mm widths measured at the bottom.
The available depths are 200, 250, 300,
form 125 mm wide joist ribs at 600 mm center-to-center spacing. These are openings becomes 600, 750 and 900 mm, respectively.
available 350, 400 and 500 mm (only for two larger sizes) with sides tapered at the rate of 1 in 12. The
in standard depths of 200, 250, 300, 350 and 400 mm.
to 400 mm, 600 mm
horizontal taper is also present at the endsreduces to 625 mm over areduces
where 475 mm width
length of 900 mm. The top
width reduces to 500 and 750 mm width
slab thickness may be 75, 100, 115 or 125 mm.
given in Table
The standard mold for a waffle slab is shown in Fig. 12.2 and its dimensions are
12.1. The size of 75 mm in Fig. 12.2 reduces to 63 mm for size of 475 mm pans.

Fig. 12.2. Standard Dome Formwork for Flat Slab.

Table 12.1. Volumes of Standard Slab Domes.


Dome Size Dome Depth Volume of Void
mm mm (m3)
200 0.113
250 0.139
750 300 0.165
350 0.191
400 0.215
500 0.263
200 0.0713
250 0.0881
(c) Flat Slab (d) Two-Way Slab with Beams 300 0.104
Fig. 2. 1. Types of Two-Way Slabs. 600 350 0.120
400 0.135
12.2 CODE PROVISIONS FOR JOIST CONSTRUCTION 500 0.164
200 0.0442
The code provisions for the joist construction are given in ACI 8.13. The joist construction 250 0.0541
may be considered to be equivalent to solid one-way or two-way slabs if the following 475 300 0.0637
three 0.0730
conditions are satisfied: 350
400 0.0821
a) There must be a monolithic combination or regularly spaced ribs and a
top slab
arranged to span in one direction or two orthogonal directions.
12.3 FLAT SLAB VS. COEFFICIENT METHOD OF SLAB DESIGN
b) The minimum width of ribs must be 200 mm and maximum depth of
times the minimum width of rib.
ribs must be 3.5 1. When the edges of each panel of the slab are not allowed to deflect down by the
presence of continuous and rigid supports like walls or very rigid beams, the design
c) The maximum clear spacing between the ribs must be 750 mm.
according to the conventional coefficient method is sufficiently accurate. However,
If the above requirements are not satisfied, the joist system must be designed as slabs resting even in case of very rigid beams, there is some interaction between the slab and the
on beams. One-way joist floor system may be used for long span floors subjected to light live beam. The beams and slab system deflects together as a complex two-dimensional
490 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIqj TWO-WAY COLUMN SUPPORTED SLABS 491
CHAPTER 12:
system. A proper finite element or other mathematical model can that the total load is to be first
accurately solve the moment P x xi,thegiving the same situation as above. This shows
problem. This analysis requires lot of efforts and is not
used for ordinary designs The
transferred to thiscolumn centerlines in one direction, producing moment in the respective
deflection of beams may be considered as settlement of supports, which is of total load is to be transferred in the perpendicular direction to the
column,
magnitude along each edge. This significantly affects the unequal direction. Then
required design of the slab. A portion of the slab acts as part
flexural behavior and which also produces bending moment in this direction.
of the beam while serving direction, full load is considered. Similarly, while
its primary purpose. The result is that, while analyzing one This is not equivalent to
2. When the beam depth is reduced, the participation of slab to act as a analyzing the other direction, the full load is once again considered. directions. Another
beam
and a stage comes when the coefficient method cannot be used for design. increases considering double the load. Only the same load is distributed in two
Also,
coefficient method, the bending moment variations along the two directions of thein ihe example of this load distribution is given below for further explanation.
slab simply supported on beams along its
panel show considerable variation along their span requiring
different amounts of steel The example of Fig. 12.4 deals with a rectangular slab
the load is transferred to the edge beams
in different portions. shorter edges, which in turn rest on columns. When
The extreme case is the flat plate, where there are no external beams. (b-part of the figure), the moment produced in the slab
at mid-span of its longer direction is as
A strip
acts as the beam. The general interaction of beams and slabs is approximately of slab under:
by two semi-empirical design methods, direct design method and modeled
method.
equivalent frame »2

8
(per unit width) =
alft
¥ *
® —
(total moment)

12.4 FRACTION OF LOAD TO BE CARRIED IN TWO


MUTUALLY PERPENDICULAR DIRECTIONS Slab simply supported

In case of flat slab design, design, frames / strips are considered in each
of the two mutually
perpendicular directions, including a single line of columns, beams
above these columns and
the slab width up to center-line of panels on each side. The question
is that
percentage of loads is to be taken by one direction strip when the perpendicular how much
strip will also
pass through the same area. The answer is explained as under by
considering on of the slab
panels, as shown in Fig. 12.3.

a) Slab with Uniformly b) Slab After Transferring Load


Distributed Load along The Longer Direction To
Beams

c) Slab With Full Transfer


Fig. 12.3. Distribution of Panel Load in Two Perpendicular Directions. - of Load To Columns

Load at point P can be transferred to the column but then it will be accompanied
(P x x). This moment can be resolved into two components, x
by a moment Fig. 12.4. Example Of Slab Load Distribution.
P xj causing bending of the x-
surface about y-axis and Pxx/ causing bending of the y-surface about x-axis. figure, the beam load is
The load on each beam becomes qfj /2 (N/m). In c-part of the-
The same situation may be considered in another way. Let us first transfer the load shorter direction at
to point-A transferred to columns and this produces the following moments along the
at the column line. The load will be transferred to point-A accompanied by the mid-span of each beam:
the moment P x x/.
Next the load is transferred to the column at B. Again it will be
accompanied by an extra
TWO-WAY COLUMN SUPPORTED SLABS 493
492 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDlQ]
CHAPTER 12:
having moment vector parallel to the base-line
or bending taking place about this line. In other
Moment at the center = —
at
2
-x —£z8- = 16
words, bending according to the shown bending moments
the line on which
bending moment diagram is plotted.
occurs on slab strips perpendicular to

Total moment in both the beams = 8


w / unit length

The reaction on each column is ?£]/4 and the total load transferred to the columns is qt^, a) Simply Supported Beam
which is equal to the total load of the slab. However, the bending moment in each direction is
to be considered for full load. Hence, for every design strip of the slab in both the directions,
the total moment in each direction is calculated as follows:
.. . direction
Momentx-in each j-x-
= —q(mdthof- load^spcm)1
8 —— = x
moment due to total load
, w/ unit length

Mid-span +ve moment


— 0.50
12.5 TOTAL MOMENT TO BE DISTRIBUTED X Mid-span -ve moment = 0.50 MQ
AMONGST MID-SPAN AND SUPPORTS

The example of previous article proves that the moment distributed in each direction is the total
static moment due to full load. Here some more examples are given in Fig. 12.5 to prove that
the negative moment at the mid-span (or the average of the negative support moments) pi us
positive moment at the mid-span is always equal to the total static moment, given as under:

Total static moment, Mo =


8
Considering the example of simply supported beam of Fig. 12.5-a subjected to uniformly
distributed load, the end negative moments are zero and hence the maximum central moment is
Mid-span +ve moment
Mid-span -ve moment
—%
=
equal to total static moment Mo. Figure 12.5-b shows that, for a propped cantilever beam, the
positive moment and average negative moment at mid-span are both equal to half of the total
static moment. In other words, the total moment at the center (A/„) is distributed at various
sections longitudinally; the end moment being 100 percent of Mo and mid-span moment is 50
percent of Mo. Similarly, in case of a fixed ended beam (Fig. 12.5-c), average negative moment Fixed Ended Beam
at mid-span is 2/j Mo and positive moment at mid-span is */3 Mo. This means that the total
moment at mid-span is again equal to same total static moment that is divided into positive and
negative moments m the ratio of 1 : 2. Then the moments at all other sections are calculated Bending Moment
accordingly. The concept of total static moment at mid-span may be generalized as in Fig. Diagram of Typical Span
12.5-d. of a Continuous Beam

For uniform load, total static moment Mo =


8
(independent of boundary conditions)
Fig. 12.5. Total Static Moment in Various Types of Beams.
12.6 VARIATION OF BENDING MOMENT IN A SLAB PANEL
Bending moments along the base-line are those which bend slab strips parallel to it. The
diagram is
When bending of slab is represented by curling of fingers of the right hand, the thumb shows moment vectors are perpendicular to the base-line on which bending moment
the axis of moment called moment vector. The moments across a base-line are the moments plotted. Figure 12.6 shows variation of bending moments both across the longer column and
494 CONCRETE STRUCTURES PART 1 - AUTHOR: ZaHID A. SIDDIQI CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 495
panel centerlines. The bending moment at column centerline is negative throughout meaning length of span transverse to / , , measured c/c of supports.
that all perpendicular strips have hogging at this centerline. The bending moments at panel width of a panel perpendicular to the design strip considered between
centerline are positive showing that all perpendicular slab-strips have sagging at this line. Same centerlines of adjacent panels.
types of variations exist along the shorter side centerlines.

Fig. 12.7. Bending Moment Variation along Shown Slab Strip.

Figure 12.7 shows bending moments along the shorter column and panel centerlines. Both of
these diagrams have negative ordinates at the ends and positive ordinates in-between. This
means that a strip parallel to these centerlines have hogging moments at the ends and sagging Design
moments for the central portion. The longer direction variation of bending moments is also Strip A
similar. The negative moments around columns have the highest magnitude and central
portion of panels have the smallest positive moments.
Design.
The slab system is divided along the mid-span centerlines along both perpendicular directions, Strip B
as shown in Fig. 12.8, to get design strips / frames. Each frame consists of one line of II
columns, beam running on these columns (if present) and portion of slab extending up to mid¬
L-v -y—
Design Strip D Design Strip C
span of adjacent panel or edge of slab. There are at least four design frames for a slab system,
namely, exterior long frame, interior long frame, exterior short frame and interior short frame. Fig. 12.8. Position of Design Strip in Plan.
However, if the slab panel size varies, the number of design frames may increase. The width of
frame ( / 2 ) and span of frame ( , ) are defined as under: 12.7 DESIGN STRIP, COLUMN STRIP AND MIDDLE STRIP
^
Zj = length of span in the direction in which moments are being determined, Each design strip / frame is further sub-divided into column strips and middle strips (Fig.
measured center-to-center of supports. 12.9). According to ACI 13.2.1, column strip is a design strip with width on each side of
= length of a panel parallel to the design strip considered center-to center column centerline equal to £j/4 or t , /4, whichever is less. Column strip includes beams,
ofsupports. column capitals and drop panels, if any. Middle strip is a design strip bounded by two column
DEVELOPMENT OF REINFORCEMENT 481
496 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI chapter 11;

stops (ACI 13.2.2). Half middle stop starts from column strip on one side and extends up to
larger of d, 12<4, or fn/16
the panel mid-span on the other side. Note: At least 1/3 M reinforcement must continue by a distance
"
beyond P.I., which is already satisfied here.
Cutoff Location For Right Side
The cut-off location may be found by considering Fig. 11.27.

Fig. 11.27. Cutoff Location Right of Support B.

Distance of cut-off point from support B - xr + greater of d & 12 dt,


=0.50 + greater of (450 & 228)mm
= 0.95 m
Fig. 12.9. Design, Column and Middle Strips.
id = 0.675 m for #19 bars Say cutoff at 1.0m right of support B
12.8 APPROXIMATE STEEL FOR A TYPICAL SLAB PANEL Check That Continuing Reinforcement Is Developed
The arrangement of steel reinforcement for a typical panel is approximately shown in Fig. j td#i9 = 0 675 m < 2.0-0.50 = 1.5m .’.OK
12.10, without actual cut-offs. A full line indicates bottom or positive steel and a dashed line i In case sufficient space is not available, hooks are to be provided.
indicates top or negative steel. Central portion of the panel has positive steels in both the •
directions while the regions around the columns (comers of the panels) have negative sleel in Examine Cutting 1 - #29 Bars In ML Region
both the directions. The remaining parts have positive steel parallel to the column centerlines j
and negative steel in perpendicular directions. The actual steel detailing requirements are : Capacity Of Section With Continuing 2 - #25 Reinforcement
discussed later. I 4/2- #25) = 1014 mm2
*
d= 525 -53 -25 /2 = 459 mm
12.9 TORSIONAL STRENGTH OF SLAB
(1014X420)
= (0.85X20X450) = 55.7 mm
The formulas used for analysis and design of flat slab most commonly include the torsional
strength or stiffness. A question arises that why torsional strength of slab is so important while (2 - #25) = (0.9)(1014)(420)(459 - 55.7/2)/10s 165.3 kN-m
performing design for flexure? To answer this question. Fig. 12.11 is presented having two
horizontal strip beams joined together in mutually perpendicular directions. The bending Distance From Mmax Point T0 T.C.P
point
moment in one of the beam at the common point acts as torque for the other beam. Hence, part It is assumed that x is the distance of cut-off moment from the maximum moment x,
of the total load may be resisted by torsional strength of beams besides the usual flexural (Fig. 11.28). At maximum moment section, shear force is zero: In order to evaluate
force
strength; Exactly in the same way, perpendicular design strips are connected to each other and moment may be taken about the cut-off point because it will eliminate the shear
loads are distributed in the two directions partially by the torsional strength. Due to this two- present there from the calculations.
way action the bending stiffness at a particular section is increased by the presence of the
perpendicular strips.
482 CONCRETE STRUCTURES PART 1 - DEVELOPMENT OF REINFORCEMENT 483
AUTHOR: ZAHID A. SIDDIQI CHAPTER II:
SM = 0 =>
165.3 +

x = 1.34 m
RS y2
2
242 =0
on either side of
1.3
V„
+4 =

Point Of Inflection
u fl83—
^^
202.8 )
+150 = 1327mm > ^2j =854 mm
OK

. 4 = 459 mm
85 kN/m
Mn = 183.7 kN-m
165.3 kN-m

/ 11
U
Any direction
(^**f.****<t±Lj 242 kN-m
-2 J

Vu = 264.7 (85)(0.73) = 202.65 kN
+459 = 1365mm > 4«5 =854mm
V, 202.65
OK
Fig. 11.28. Cut-Off Location on Either Side of Maximum Moment Section.
must be satisfied.
Now bars must extend up to larger of the following distances from the maximum Note: In order to terminate bars in tension zone, ACI 12.10.5
Hanger
moment section. Figure 11.29 shows the final detailing of reinforcement for the example beam.
of diagram.
..
1) -
_J12^
[rf
_
x + ilarger oW, = 12(29) = 348mm
= 459mm
, ,. 459 . on
= 1.34+-
iqqq
= 1.80 m
bars and transverse reinforcement are not shown for clarity

2) 4#29 = 1063 mm = 1.063 m


Actual cutoff distance from left support = 2.386- 1.80 = 0.584 m
Sav 0.5 m from left support
Actual cutoff distance from right support = (5.5-2.386)- 1.80 = 1.314m
Sav 1,3 m from support B
Check That Continuing Steel (2#25) Is Fully Developed At TCPs
Fig. 11.29. Final Detailing Of Reinforcement
Distance of theoretical cut-off point from left support = 2.386 - 1.34
= 1.046 m
4m = 854 mm = 0.854 m < 1.046 m OK
On the right side, adequate space is available and there is no end of beam problem.
Note: At least 1/3 M4- reinforcement at simple end and !4 M4- reinforcement at continuous end
must extend into supports by a distance larger than or equal to 150 mm.

CHECK FORM4 STEEL DIAMETER


Support

Say 4 = 150 mm (embedment beyond center of support)

M (2- #25) = ^- = 183.7 kN-m


V„ 202.8 kN
484 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDlQj
CHAPTER U: DEVELOPMENT OF REINFORCEMENT 485
Space For Reader’s Notes
PROBLEMS

Note; N = Registered Number.


No = A fixed positive or negative number assigned by
the instructor to get R between 600 and 1000.
R = N— No, a parameter to get numerical data for the
problems.

Design the beam shown in the Figure, without considering any pattern loading and
correcting moments to face of supports. Locate the cut-off points for the proposed
curtailment of bars. Use // = 28 MPa,^ = 420 MPa, clear cover = 40 mm, #13 stirrups
and minimum transverse reinforcement throughout the length.

w„=^7kN/m
X X A X X X X X XX XX XX X X
486 CONCRETE STRUCTURES PART - I AUTHOR: Z ARID A. SIDDlQj
Space For Reader’s Notes Chapter - 12

TWO-WAY COLUMN SUPPORTED SLABS

12.1 INTRODUCTION
The slab, which rests on supports on all the four sides and has longer to shorter span ratio
lesser than 2.0 is called two-way slab. The supports may be beams cast within the slab. If the
supports are incorporated in both directions within the depth of the slab itself with or without
projected beams, the resulting slab system is called two-way column supported slab system. A
general classification of two-way slab systems is given in Fig. 2.1. If there are no projected
beams and the slab alone is directly resting on the columns, the resulting slab system is called a
flat plate. If some portion of the slab over the columns and surrounding area is constructed
with greater but constant thickness or if the column is gradually enlarged like a cone or
pyramid at the junction with the slab, the resulting system is called flat slab.
The thickened slab near the column is called drop panel. Flaring up to the top of the column is
called column capital. The drop panel commonly extends about one-sixth of the span each
way from the column, giving extra strength in the column region while minimizing the amount
of concrete at mid-span.
For larger spans, the thickness of slab required to transmit the vertical loads to the columns
exceeds that required for moments in the central region between the columns. Further, for
positive moments alone with very less shear, most of the concrete on the tension side is not
usefill. This actually increases self-weight with no strength contribution. A solid flat slab
becomes uneconomical in such cases, as the material is not efficiently utilized. To lighten the
slab, reduce the slab moments and save material, the slab at mid-span can be replaced by
intersecting ribs. However, near the columns, full depth is retained to transmit loads from the
slab to the columns. This type of slab is known as a waffle slab or a two-way joist system,
where the cavities in the slab are made with fiberglass or metal dome forms. Using waffle slab,
the cost of material is reduced but the labor cost and the cost of formwork is increased. The
total cost of waffle slab will approximately become lesser when the total depth of solid slab
required exceeds 180 mm.
sum of slab panel sides(m)x2x 1000 on
180
or sum of slab panel sides (m) >17
This means that a square panel 9m x 9m or larger and rectangular panel 10m x 7m or larger
may be designed as a waffle slab.
There are two standard types of domes used as formwork for flat slab construction shown in
Fig. 12.2. The slab thickness over top of the dome is usually 75 or 115 mm. The approximate
volume of each type of dome is given in Table 2.1.
TWO-WAY COLUMN SUPPORTED SLABS 489
488 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 12:
o 750 x 750 mm square domes with 75 mm flanges. The flanges of adjacent domes form loads. Usually, removable metal forms are used to get the required shape, which are known as
150 mm wide joist ribs at 900 mm center-to-center spacing. These are available in fillers or pans. In other cases, clay tile fillers are used between joists, which are left in place to
depths of 200, 250, 300, 350, 400 and 500 mm. become part of the ceiling. Standard square pans are available in opening sizes of 475, 600
o 475 x 475 mm square domes with 62.5 mm flanges. The flanges of adjacent domes and 750 mm widths measured at the bottom. The corresponding center-to-center spacing of
form 125 mm wide joist ribs at 600 mm center-to-center spacing. These are available
openings becomes 600, 750 and 900 mm, respectively. The available depths are 200, 250, 300,
350, 400 and 500 mm (only for two larger sizes) with sides tapered at the rate of 1 in 12. The
in standard depths of 200, 250, 300, 350 and 400 mm.
horizontal taper is also preserit at the ends where 475 mm width reduces to 400 mm, 600 mm
width reduces to 500 and 750 mm width reduces to 625 mm over a length of 900 mm. The top
slab thickness may be 75, 100, 1 15 or 125 mm.
The standard mold for a waffle slab is shown in Fig. 12.2 and its dimensions are given in Table
12.1. The size of 7 5 mm in Fig. 12.2 reduces to 63 mm for size of 475 mm pans.

12 Flange

75 mm

(a) Flat Plate (b) Waffle Slab Fig. 12.2. Standard Dome Formwork for Flat Slab.

Table 12.1. Volumes of Standard Slab Domes.


Dome Size Dome Depth Volume of Void
mm mm (n?)
200 0.113
250 0.139
750 300 0.165
350 0.191
400 0.215
500 0.263
200 0.0713
250 0.0881
300 0.104
Fig. 2.1. Types of Two-Way Slabs. 600 350 0.120
400 0.135
500 0.164
12.2 CODE PROVISIONS FOR JOIST CONSTRUCTION 200 0.0442
The code provisions for the joist construction are given in ACI 8.13. The joist construction 250 0.0541
may be considered to be equivalent to solid one-way or two-way slabs if the following three 475 300 0.0637
350 0.0730
conditions are satisfied: 400 0.0821
a) There must be a monolithic combination or regularly spaced ribs and a top slab
arranged to span in one direction or two orthogonal directions. 12.3 FLAT SLAB VS. COEFFICIENT METHOD OF SLAB DESIGN
b) The minimum width of ribs must be 200 mm and maximum depth of ribs must be 3 5 I. When the edges of each panel of the slab are not allowed to deflect down by the
times the minimum width of rib. presence of continuous and rigid supports like walls or very rigid beams, the design
c) The maximum clear spacing between the ribs must be 750 mm. according to the conventional coefficient method is sufficiently accurate. However,
If the above requirements are not satisfied, the joist system must be designed as slabs resting even in case of very rigid beams, there is some interaction between the slab and the
on beams. One-way joist floor system may be used for long span floors subjected to light live beam. The beams and slab system deflects together as a complex two-dimensional
490 CONCRETE STRUC TURES PART 1 - AUTHOR: ZAHID A. SlDDlQj CHAPTER 12:
TWO-WAY COLUMN SUPPORTED SLABS 491
system. A proper finite element or other mathematical model that the total load is to be first
can accurately solve the moment P xtoXi,thegiving the same situation as above. This shows
problem. This analysis requires lot of efforts and is not
used for ordinary designs. The transferred thiscolumn centerlines in one direction, producing moment in the respective
deflection of beams may be considered as settlement of supports, which is the column,
of unequal direction. Then total load is to be transferred in the perpendicular direction to
magnitude along each edge. This significantly affects
the flexural behavior and
required design of the slab. A portion of the slab acts as part which also produces bending moment in this direction.
of the beam while serving direction, full load is considered. Similarly, while
its primary purpose. The result is that, while analyzing one is once again considered. This is not equivalent to
2. When the beam depth is reduced, the participation of slab to act as a analyzing the other direction, the full load two directions. Another
and a stage comes when the coefficient method cannot be used for design.
beam increases considering double the load. Only the same load is distributed in
Also, example of this load distribution is given below for further
explanation.
coefficient method, the bending moment variations along the two directions of in die
the slab slab simply supported on beams along its
panel show considerable variation along their span requiring different
amounts of steel The example of Fig. 12.4 deals with a rectangular
the load is transferred to the edge beams
in different portions. shorter edges, which in turn rest on columns. When mid-span of its longer direction is as
The extreme case is the flat plate, where there are no external
beams. A strip of slab (b-part of the figure), the moment produced in the slab at
acts as the beam. The general interaction of beams and slabs is under:
approximately modeled
by two semi-empirical design methods, direct design method
and equivalent frame
method. M- -2-L (per unit width) =
8
-8 (total moment)

12.4 FRACTION OF LOAD TO BE CARRIED IN TWO 1


MUTUALLY PERPENDICULAR DIRECTIONS
In case of flat slab design, design frames / strips are considered in
each of the two mutually
perpendicular directions, including a single line of columns,
beams above these columns and
the slab width up to center-line of panels on each side.
The question is that
percentage of loads is to be taken by one direction strip when the perpendicular how much
pass through the same area. The answer is explained as under by considering strip will also
panels, as shown in Fig. 12.3.
on of the slab

a) Slab with Uniformly b) Slab After Transferring Load


Distributed Load along The Longer Direction To
Beams

c) Slab With Full Transfer


of Load To Columns
Fig. 12.3. Distribution of Panel Load in Two Perpendicular Directions.

Load at point P can be transferred to the column but then it will be


accompanied by a moment Fig. 12.4. Example Of Slab Load Distribution.
(P x x). This moment can be resolved into two components, x
P xj causing bending of the x-
surface about y-axis and P xxi causing bending of the y-surface about the- figure, the beam load is
x-axis. The load on each beam becomes q tx!2 (N/m). In c-part of
along the shorter direction at
The same situation may be considered in another way. Let us first transfer the load to point-A transferred to columns and this produces the following moments
at the column line. The load will be transferred to point-A the mid-span of each beam:
accompanied by the moment P x xj.
Next the load is transferred to the column at B. Again it will be accompanied
by an extra
TWO-WAY COLUMN SUPPORTED SLABS 493
12:
492 CONCRETE STRUCTURES PART -1 CHAPTER
AUTHOR: ZAHID A. S!DD[qj
parallel to the base-line or bending taking place about this line. In other
* having moment vector shown bending moments occurs on slab strips
perpendicular to
.. 4
the center = - - x
Moment at 4k4 4

2 — —8
J- = — 16
L

q£ Sp2
j

Words, bending according to
the line on
the
which bending moment diagram is plotted.
w / unit length
Total moment in both the beams = ~ 8-
llllllll^'H'I'J''!''!'^ a) Simply Supported Beam
The reaction on each column is qt j /4 and the total load transferred to the columns is
which is equal to the total load of the slab. However, the bending moment in each direction is wf1 .
to be considered for full load. Hence, for every design strip of the slab in both the directions, =MC
8
the total moment in each direction is calculated as follows:
.Moment
. . x- =
j.
in each direction
q[widthof load^spanf'
— —— =
'

, x
moment due to total load
T
w / unit length
8 ||||||||'ll'l,'l''l''l''l''l' Mid-span +ve moment = 0.50
Mid-span -ve moment = 0.50
12.5 TOTAL MOMENT TO BE DISTRIBUTED
AMONGST MID-SPAN AND SUPPORTS
8
The example of previous article proves that the moment distributed in each direction is the total
static moment due to full load. Here some more examples are given in Fig. 12.5 to prove that 8 b) Propped Cantilever
the negative’ moment at the mid-span (or the average of the negative support moments) plus 16
positive moment at the mid-span is always equal to the total static moment, given as under:
tv / unit length
wf2 |11iI IInK
§|11|||
Total static moment, Mo =
8 £

Considering the example of simply supported beam of Fig. 12.5-a subjected to uniformly
distributed load, the end negative moments are zero and hence the maximum central moment is
to *1
wl2
Mid-span +ve moment
Mid-span -ve moment
— % Mo
= /3 M
equal to total static moment Mo. Figure 1 2.5-b shows that, for a propped cantilever beam, the 24
positive moment and average negative moment at mid-span are both equal to half of the total
static moment. In other words, the total moment at the center (Af0) is distributed at various
sections longitudinally; the end moment being 100 percent of M„ and mid-span moment is 50
percent of Mg. Similarly, in case of a fixed ended beam (Fig. 12.5-c), average negative moment
h - s
c) Fixed Ended Beam
at mid-span is 2/3 Mo and positive moment at mid-span is % Mo. This means that the total 12
moment at mid-span is again equal to same total static moment that is divided into positive and M,
d) Bending Moment
negative moments in the ratio of 1 : 2. Then the moments at all other sections are calculated Diagram of Typical Span
accordingly. The concept of total static moment at mid-span may be generalized as in Fig.
12.5-d.
of a Continuous Beam

For uniform load, total static moment Mn =


(independent of boundary conditions)
8
M.= —o
-=
2
Types of Beams.
Fig. 12.5. Total Static Moment in Various
12.6 VARIATION OF BENDING MOMENT IN A SLAB PANEL slab strips parallel to it. The
Bending moments along the base-line are those which bend bending moment diagram is
which
When bending of slab is represented by curling of fingers of the right hand, the thumb shows moment vectors are perpendicular to the base-line on
the axis of moment called moment vector. The moments across a base-line are the moments plotted. Figure 12.6 shows variation of bending moments both across the longer column and
494 CONCRETE STRUC TURES PART - 1 AUTHOR: ZaHID A. SIDDlQj CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 495
panel centerlines. The bending moment at column centerline is negative throughout meaning Z2 = length of span transverse to measured c/c of supports.
that all perpendicular strips have hogging at this centerline. The bending moments al pane] = width of a panel perpendicular to the design strip considered between
centerline are positive showing that all perpendicular slab-strips have sagging at this line Same centerlines of adjacent panels.
types of variations exist along the shorter side centerlines.

Fig. 12.7. Bending Moment Variation along Shown Slab Strip.

Figure 12.7 shows bending moments along the shorter column and panel centerlines. Both of
these diagrams have negative ordinates at the ends and positive ordinates in-between. This
means that a strip parallel to these centerlines have hogging moments at the ends and sagging Design
moments for the central portion. The longer direction variation of bending moments is also Strip A
similar. The negative moments around columns have the highest magnitude and central
portion of panels have the smallest positive moments.
Design <
The slab system is divided along the mid-span centerlines along both perpendicular directions, Strip B
as shown in Fig. 12.8, to get design strips / frames. Each frame consists of one line of ft -v—
columns, beam running on these columns (if present) and portion of slab extending up to mid¬ Design Strip D Design Strip C
span of adjacent panel or edge of slab. There are at least four design frames for a slab system,
namely, exterior long frame, interior long frame, exterior short frame and interior short frame. Fig. 12.8. Position of Design Strip in Plan.
However, if the slab panel size varies, the number of design frames may increase. The width of
12.7 DESIGN STRIP, COLUMN STRIP AND MIDDLE STRIP
frame (^ 2) and span of frame (/}) are defined as under:
ft = length Of span in the direction in which moments are being determined, Each design strip / frame is further sub-divided into column strips and middle strips (Fig.
measured center-to-center of supports. 12.9). According to ACI 13.2.1, column strip is a design strip with width on each side of
= length of a panel parallel to the design strip considered center-to center column centerline equal to ^/4 or ^t/4, whichever is less. Column strip includes beams,
of supports. column capitals and drop panels, if any. Middle strip is a design strip bounded by two column
TWO-WAY COLUMN SUPPORTED SLABS 497
496 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI CHAPTER 12:
|
strips (ACI 13.2.2). Half middle strip starts from column strip on one side and extends up t0 i
the panel mid-span on the other side. I

Typical Slab Panel.


Fig. 12.10. Approximate Steel Placement for a

Fig. 12.12 having both the torsional and flexural


An element of slab out of a panel is shown instrips, bending moment and twisting moment are
moments at its faces. For the perpendicular the load is distributed in the two
interchanged in the two perpendicular directions and hence
beams is very important in determining
Fig. 12.9. Design, Column and Middle Strips. i directions. It is clear that torsional stiffness of slab and
the overall behavior of the slab system.
12.8 APPROXIMATE STEEL FOR A TYPICAL SLAB PANEL

The arrangement of steel reinforcement for a typical panel is approximately shown in Fig.
12.10, without actual cut-offs. A full line indicates bottom or positive steel and a dashed line : Fig. 12.11. Two Mutually Perpendicular shallow
indicates top or negative steel. Central portion of the panel has positive steels in both the ’ Strip Beams Lying in Horizontal Plane.
directions while the regions around the columns (corners of the panels) have negative steel in
both the directions. The remaining parts have positive steel parallel to the column centerlines
and negative steel in perpendicular directions. The actual steel detailing requirements are
discussed later.

12.9 TORSIONAL STRENGTH OF SLAB

The formulas used for analysis and design of flat slab most commonly include the torsional
strength or stiffness. A question arises that why torsional strength of slab is so important while
and Flexural Moments.
performing design for flexure? To answer this question. Fig. 12.11 is presented having two Fig. 12.12. A Small Element of Slab Having Torsional
horizontal strip beams joined together in mutually perpendicular directions. The bending
moment in one of the beam at the common point acts as torque for the other beam. Hence, part 12.10 DIRECT DESIGN METHOD
of the total load may be resisted by torsional strength of beams besides the usual flexural
strength. Exactly in the same way, perpendicular design Strips are connected to each other and Tire first design method that may be used for two-way
slab system design is the empirical
loads are distributed in the two directions partially by the torsional strength. Due to this two- | (DDM). Because of the empirical
method given in ACI 13.6, called direct design method
way action the bending stiffness at a particular section is increased by the presence of the ! systems and six limitations of the
nature of this method, it is applicable to selected types of slab
perpendicular strips.
498 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDiqi TWO-WAY COLUMN SUPPORTED SLABS 499
CHAPTER 12:
method are given in the code. The design procedure using this method is explained here
terms of distinct steps. in vi) For a panel with beams between supports on all sides, the relative stiffness
(af of beams in two perpendicular directions should not be less than 0.2 nor
STEP 1: Check For Limitations Of DDM
greater than 5.0. This means that the stiffness of beams in the two perpendicular
First five restrictions of the direct design method are checked in this step and the 6th directions should not vary too much.
restriction will be considered later. These restrictions are as under:
i) There is a minimum of three continuous spans in each direction.
ii) Panels are rectangular, with a ratio of center-to-center longer to shorter span ratio
of
each panel not greater than 2. This condition eliminates the possibility of one-way ratio of flexural stiffness of beam section to flexural stiffness of a
action of slabs.
where a/ =
width of slab bounded laterally by centerlines of adjacent panels

short
_ (if any) on each side of the beam.

=
(El)beam
iii) Adjacent center-to-center span lengths in each direction should not differ by more EaI, (El)slab
than one-third of the longer span. vii) Moment redistribution allowed by ACI 8.4 is not applicable to slab systems
designed by the Direct Design Method.
"J-‘
= 0.67 to 1.33 for each ‘s’
Selection Of Slab Depth
STEP 2:
iv) If the columns are not exactly present in a single line, a maximum offset of columns
In this step, the slab depth is decided according to ACI 9.5.3 provisions.
equal to 10% of the span in the direction of offset is allowed from between
centerlines of successive columns (Fig. 12.13). Two-Way Slab Depth Without Interior Beams
The minimum thickness is greater of the following values and that given by Table 12.2:
a)
b)
Slabs without drop panels
Slabs with drop panels
— 125 mm
100 111111

Table 12.2. Minimum Slab Depth Without Interior Beams.


Exterior panel Interior panel
+ either drop + drop panel
Exterior panel panel or edge (OR)
fl beam
+ no drop panel Exterior panel
(MPa) (OR)
+ no edge beam + drop panel
Interior panel + edge beam
Fig. 12.12. Maximum Allowed Column Offset. + no drop panel
280/300 4/33 4/36 4/40
v) The direct design method is only applicable to uniformly distributed gravity loads. 420 4/30 4/33 4 /36
Separate analysis is to be made for concentrated loads or lateral loads. Further, live 520 4/28 4/31 4/34
load should not exceed two times the dead load. The coefficients given in this
method are for pattern loads up to the specified limit
Where 4 = length of clear span in long direction of two-way construction,
a) No lateral loads measured face-to-face of supports in slabs without beams and face-to-
LL face of beams or other supports in other cases.
b) 2.0 Note:- Edge beam is considered to be present if > 0.8.
DL
500 CONCRETE STRUCTURES PART 1 - AU 1HOR: ZAHID A. SIDDjqj
TWO-WAY COLUMN SUPPORTED SLABS 50]
Two-Way Slab Depth With Beams On All Sides CHAPTER 12:
the slab
above or below the slab, whichever is greater, but not greater than four times
Let, Of = thickness (Fig. 12.14).
and afa ~ average value of ‘ a? for all beams on edges of a panel. For L-beams, the effective width ‘6’
For slabs with beams spanning between the supports on all sides and denoting is lesser of
clear i) bv + hb
span in the long direction by the minimum thickness required is determined as ii) bv + ^hf
follows:
a) If 0.2 < < 2.0 (shallow beams), For interior T-beams the effective
width ‘b’ is lesser of b.
0.8+^^ i) b„+ 2hi,
h. = I 1400...Jy, but not less than 125 mm
ii) b^ + Shf Fig. 12.14. Effective Slab Width for L-Beams
36 + 5Xa>-0-
The neutral axis is then located for the resulting beam and its moment of inertia (Ii) is
b) If Ofm > 2.0 (stiff beams), calculated exactly. For normal proportions, lb is approximately equal to twice the
moment of inertia of rectangular portion for interior beams and 1.5 times the moment of
£ 0.8+-^- inertia of rectangular portion for edge beam. The values may optionally be tabulated as
I 1400 in Table 12.3.
"mio but not less than 90 mm
36+93
Where fl = ratio of clear spans in long to short direction. Table 12.3. Calculation of Stiffiiess of Beams.
Exterior Interior Exterior Interior
c) If Ofin < 0.20, the provisions for slabs without interior beams must be applied. Frame Short
Long Long Short
d) For panel with one or more discontinuous edges having edge beam with ctf <
0.8, Amin is to be increased by at least 10% in that panel. This increase is not Web width, bv (mm)
required for slabs without interior beams and is not to be applied to the upper Depth, h (mm) — —

limit of 125 and 90 mm. , 7b (xlO4 mm4)


Conservative Approach For Selection Of Slab Depth
The maximum thickness for slabs without interior beams may be used as under:
STEP 4: t^-Value And 6th Limitation Of DDM
IS Compute the relative stiffiiess of longitudinal beams with respect to slab
(spanning in
the same direction) for the beams included in each design strip. Length for
both the
h 5 125 mm for fy = 280/300 MPa I members is same and hence is cancelled out.
stiffiiess of a width of
h

Zi
30
5 125 nun forty = 420 MPa
s ctf = ratio of flexural stiffiiess of beam section to flexural
slab bounded laterally by centerlines
of the beam. In other words, the width
of adjacent
of slab
panels
should
(if any) on each side
be equal to width of
For slabs with interior beams, ctm may be assumed to be greater than 2.0 and depth may
each design strip (/zw)-
be calculated using the corresponding formula. For very shallow beams, necessary
increase in depth may be required. _ Beamstiffhess __ EaIt
STEP 3: Beam Stiffness Slab stiffiiess EaI^
The stiffness for beams within all the design strips is determined. According to ACI Ofi - ctf- value in direction of : ap = ctf- value in direction of £2
13.2.4, for monolithic or fully composite construction, a beam includes that portion of I
slab on each side of the beam extending a distance equal to the projection of the beam Is = value of slab inertia effective for £1 direction.
I
=
12
502 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SJDDlQj TWO-WAY COLUMN SUPPORTED SLABS 503
The 6th condition for the use of direct design method is that “The
RAFTER 12:
ratio of and H , Calculation Of Factor
a/2^iA2must in-between 0.2 to 5.0 for all combinations of beams in
the The factor pt is defined as the ratio of torsional stiffness
of edge beam section to
perpendicular directions. This condition needs only
to be satisfied if beams are present stiffness of a width of slab equal to span length of the edge beams, center-to-
flexural
on all the four sides of the panel. For example, it is not
applicable for the exterior center of supports.
panels if exterior beams are present but interior
beams are absent”.
$
__ GC of edge beam ~ EaC
a
‘ ^1 of slab having width equal to span of edge beam 2Ea Is
5.0 > 0.2
s2

——
zv
E E
if the Poisson’s effect is neglected.
*

The required values may be written in tabular form as in Table 12.4.


? where G = —7 r
2(1+v) 2
Further, the value i
of Ofa is calculated for each panel and the slab depth is adjusted
if required. width of slab is to be
The calculation is same for exterior and interior frames as the
form of these
1 considered equal to the span of the torsion member. The systematic
Table 12.4. Calculation of Relative Stiffiiess of Beams.
calculations may be done by filling tables like Table 12.5.
Frame Exterior Interior Exterior Interior
Long Long
Table 12.5. Torsion Properties of Beams.
Short Short
Zb (xlO^mm4) Frame Long Short
Z2W(mm) Span of edge beam Shorter Longer
7s(xl04mm4) C(xl04 mm)
Span of torsion member, ^2 (mm)
Is (xlO4 mm) for Z2 width
STEP 5: Torsional Stiffness Of Edge Beam A
The dimensions of edge beams to act as torsional members are
found according to ACL
13.7.5.1, for both directions. The torsional members are considered Calculation Of Factored Static Moment For Each Design Strip
to have a constant STEP 6:
cross section throughout their length consisting of the largest of: The slab is divided into design strips and the total factored static
moment is calculated
and average negative factored moments,
a) A portion of slab having a width equal to that of the column, bracket, for each design strip. Absolute sum of positive
or called total static moment, in each direction is equal to the following:
capital in the direction of the span for which moments are being
determined
b) For monolithic or fully composite construction, the transverse beam
above
and below the slab is added to the portion of slab as above in (a) to get
the
effective slab width
c) The transverse beam as defined in Step 3 (ACI 13.2.4) where, = total width of the design strip, and,
capitals,
= clear span, extending from face to face of columns,
Divide the torsional members obtained into brackets, or walls.
rectangles with smaller dimensions ‘x’ and larger
dimension 'y’, as shown in Fig. 12.15. The 0.65^
square supports with
torsional constant is then evaluated as the Circular or regular polygon shaped supports should be treated as
following summation: the same area, as in Fig. 12.16.
side of the centerline of
*3
ACI requires that where the transverse span of panels on either
C =
^^1-0.63^^^
Compute the torsional constant, C, for all the edge beams.
Fig. 12.15. Torsion Member.

These torsional constants


supports varies, in above equation is to be taken as the average of adjacent
transverse spans. Similarly, for the edge design frame, when
parallel to an edge is being considered, the distance from edge
the span adjacent and
to panel centerline is
are to be used for the perpendicular strips. 12.6.
substituted for (ACI 13.6.2.4). The values are entered in Table
a

504 CONCRETE STRUCTURES PART - 1 TWO-WAY COLUMN SUPPORTED SLABS 505


AUTHOR ZAHID A. SIDDIQj CHAPTER 12:
Table 12.7. Longitudinal Distribution Of Moments For Exterior Slab Panels.
Slab without beams
Exterior edge Slab with between interior
unrestrained supports Exterior
beams edge fully
(Torsion of
between all Part slab restrained
member not supports considered With edge
considered) as torsion beam
member
(1) (2) (3) .
(4) (5)

Fig. 12.16. Equivalent Column Area for Calculation of Clear Span. Int. M 0.75 0.70 0.70 0.70 0.65

NT 0.63 0.57 0.52 0.50 0.35


Table 12.6. Total Static Moments All Design Frames. 0.65
Ext. M 0 0.16 0.26 0.30
Frame Exterior Interior Exterior Interior
Long Long Short Short
Aw(m) Table 12.8. Moments After Longitudinal Distribution.
4(m) Exterior Interior
Exterior Interior
Mo (kN-m) Frame Long Long Short Short
Mo
STEP 7: Longitudinal Distribution Of Moments M at ext. support
M* in ext. span
Longitudinal distribution of moments means the way in which the total static moment
at mid-span is divided into positive and negative moments. Further, the end negative M at first int. support
moments are decided based on the average negative moment at the mid-span. In other M at typical int. support
words, the total static moment for each of the design frames is to be distributed M+ in interior span
longitudinally into negative moments at the supports and positive moments at mid¬
spans. However, the sum of positive moment and magnitude of average negative STEP 8: Transverse Distribution Of Moments
moment at the mid-span should remain equal to the total static moment This
distribution of moments is along the length of the design frame or 11 direction. The moments determined at critical sections of the design frames as above are further
distributed into column strips including the beams (if any) and middle strips. This step
According to ACI 13.6.3.2, in an interior span, total static moment Mo is to be is called the transverse distribution of moments.
distributed according to the given fractions. If spans on both sides of a support are Column Strip Moment Percentages
different, M sections are to be designed to resist the larger of the two interior negative
moments determined for the adjacent spans. Let = 4 0.5≤A≤2.0
pt ~ B If #>2.5, 5 = 2.5
Factored M at supports = 0.65 Mo
Factored M* at mid-span = 0.35 Mo a^= D If an ^->1.0, D= 1.0
ACI 13.6.3.3 says that, in an end span, M is to be distributed according to Table 12.7.
A
The values of the moments after the distribution may be tabulated as in Table 12.8. Note that £2 is the total width of panel, c/c of columns, for both exterior and interior
design strips. As per ACI 13.6.4, column-strip moment is expressed as the following
percentage of total moment at critical section:
506 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDJQj CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 5Q7
Interior negative moment (%age): 75 + 30(1-A)D Middle Strip Moments
Exterior negative moment (%age): 100-105+1250(1 -A) The portion of the total moment at critical sections of design frames not resisted by the
Positive moment (%age): -
60 + 15(3 2A)D column strip is proportionally assigned to the adjacent half middle strips (ACI 13.6.6).
When there are no longitudinal beams, the column strip percentage The middle strip adjacent to an edge supported by a wall should be proportioned to
the aspect ratio and smaller than the cases with beams. is independent of resist twice the moment assigned to its interior half. The resulting values may be
With the increase of the entered in Table 12.10, for all of the critical sections.
torsional rigidity of the edge beam, the column strip moment
percentage out of exterior
negative moment reduces. When edge beam is replaced by
a wall in greater than three-
fourth of the width of the frame, the exterior negative moment is Table 12.1 0. Design Bending Moments For Various Frames.
into column and middle strips. The values for different uniformly distributed
design frames may be entered Frame Location Ext Int Ext Int
in worksheets as in Table 12.9.
Long Long Short Short
Beam moment
Table 12.9. Column Strip And Beam Moments.
ext.span
Frame Ext Long Int. Long Ext. Short Tnt Shnrf
A= Jim int.
B= A -
D= M+ int. span
%age ot M1 to column strip
Column strip slab moment
%age share of beam from above moment
%age of to column strip M* ext. span
ypage share ot beam from above moment Jim int.
%age of Mal to column strip
%age share of beam from above moment int. span

Beam Moment Middle strip moment


M+ ext. span
Percentage of column strip moment that is to be resisted by column
line beam ACI first int.
13.6.5) spanning in direction is equal to:

afl x 85 with a maximum of 85 % (entered in Table If int. span


12.9)
In addition to moments calculated for uniform loads, beams STEP 9: Calculation Of Slab Reinforcement
by concentrated or linear loads applied directly to resist all moments caused
beams, including weight of projecting The slab steel may be calculated from the slab moments by using the usual under¬
beam stem above or below the slab. The total static
moment due to such loads may be reinforced concrete design formulas. For this purpose, trial method, quadratic equation,
distributed in the same ratio as for the slab.
tables or curves may be used. The minimum steel required in slabs for temperature,
Column Strip Slab Moment shrinkage and distribution of loads is given in Table 12.11 and the calculated steel may
In cases where beams are present within column strips, the slab portion be entered m Table 12.12.
are proportioned to resist that portion of column strip of column strips
fTablptOOt moments not resisted by beams
A.
0.9//d-a/2) 0.85//6
CS slab moment = CS moment - beam moment
j CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 509
508 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI
> reinforcement within a drop panel runs at the same level as in the part of slab without
1 drop panel.
Table 12.11. Slab Minimum Steel Ratios For Temperature And Shrinkage.
Ann based on &xA for Grade 280, 300 or 350 deformed bars 0.0020 STEP10: Torque For Edge Beam
Aiin for Grade 420 deformed bars or welded wire fabric 0.0018 Edge beams or edges of slab reserved as a torsion member are to be designed to resist
Anh for measured at 0.3% yield strain 0.0018x420 torsion equal to their due share of exterior M . At middle strip location, the exterior
fy negative moment of middle strip may be transferred to the edge beam as torque, while
for column strip; the part of column strip exterior moment that is not transferred by
• flexure to column may be transferred by torque. ACT 13.6.3.6 recommends that the
Table 12.12. Slab Steel Areas For Various Frames. > gravity load moments to be transferred between slab and edge column in accordance
Frame Location Ext Int Ext. Int. j withACI 13.5.3.1 are to be 0.3 Mo.
Long Long Short Short
CS Width Minus Beam Width STEP 11: Special Corner Reinforcement
As for column strip ACI 13.3.6 requires that, in slabs with beams between supports with a value of a
Dia. and no. of bars for CS greater than 1.0, special top and bottom slab reinforcement should be provided at
As for column strip exterior comers in accordance with the following:
M* ext. span
Dia. and no. of bars for CS i) Moment for this top and bottom steel must be equal to (per meter width) in
A, for column strip the slab.
first int.
Dia. and no. of bars for CS ii) For top steel, this moment is considered about an axis perpendicular to the diagonal
A, for column strip from the comer (steel is placed parallel to the diagonal), and for bottom steel,
^7«. moment is acting about an axis parallel to the diagonal from the comer (steel is
Dia. and no. of bars for CS
placed perpendicular to the diagonal).
Xs for column strip M+ int. span
iii) The special reinforcement shall be provided for a distance in each direction from the
Dia. and no. of bars for CS |< . comer equal to one-fifth the longer span.

MS width The special reinforcement may either be placed parallel to the diagonal in the top of the
slab and perpendicular to the diagonal in the bottom of slab, or, it may be placed
As for column strip Af"
ext. parallel to the edges in both directions in top and bottom of the slab. However, the total
Dia. and no. of bars for MS g steel required will be more in this second option.
As for column strip M + ext. span
Dia. and no. of bars for MS STEP 12: Development Of Flexural Reinforcement
As for column strip first int. For slabs with beams, usual procedure is used to curtail the slab reinforcement.
Dia. and no. of bars for M.S However, for slabs without beams, ACI Fig. 13.3.8 is used for detailing that gives the
following provisions for bar curtailment in slabs without beams:
As for column strip
A. Column Strip Top Steel
Dia. and no. of bars for MS
As for column strip M" int.span i) Half top steel should extend 0.30 tn beyond the face of support and 90° hooks
Dia. and no. of bars for MS are to be provided at ends on exterior supports. The value of 0.30 f „ should be
increased to 0.33 n , if drop panels are present.
The negative reinforcement for a drop panel may be calculated using the total effective
depth available at the drop panel. However, the extra depth of drop panel below the
slab must not be considered greater than the slab depth divided by 4. The positive

4a
510 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SJDDjqj CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 511
ii) The remainder half steel should extend 0.20 I „ past the face of support and I M = 0.07 [(7du + 0.5
end with 90° hooks in exterior supports. : Where q’DU , are dead load, panel width and clear span related to
and l’a
B. Column Strip Bottom Steel '
shorter span.
All bars must be provided throughout the span, with half having 90° hooks Edge Column: As already stated, according to ACI 13.6.3.6, the gravity load moment
in
exterior supports over the columns. to be transferred between slab and edge column is to be 0.3 Mo.
C. Middle Strip Top Steel
* STEP 16: Moment Transferred From Slab To Column By Flexure (ACI 13.5.3)
All bars must extend 0.22 ln past the face of support, with 90° hooks in
exterior Two-Way Shear Or Punching Shear: The shear acting all along the perimeter of a
supports. ; column for a flat slab without beams can punch the column into the slab and is called
D. Middle Strip Bottom Steel , two-way or punching shear. The failure occurs by the successive punching of all
i) Half bottom steel should extend throughout the span. I columns of a slab system forcing the slab to settle down to the bottom of the columns.
ii) The remainder half alternate bars should extend fully to the outer edges but When upper floor after failure strikes the lower floor, it may also undergo similar type
can J? of failure. This causes complete collapse of the structure.
curtailed at a maximum distance of 0.15 f „ from center of the interior supports.
Direct Shear: The two-way shear produced by the vertical loads on the slab and
STEP 13: Shear In Beams having constant stress intensity all along the critical perimeter is called direct shear.

According to ACI 13.6.8, for beams with Concept Of Eccentric Shear Or Combined Shear And Moment: Consider atypical
> 1.0, shear is calculated by 45°
tributary lines area shown in Fig. 12.17. For connection of slab with column, shown in Fig. 12.18, where an unbalanced moment,
< 1.0, linear interpolation should | Mu, is to be transferred between the slab and the column. A part of this unbalanced
be made assuming that shear is zero when afl - 0, Shears produced by factored loads
directly applied on beams must also be added to the above results. h moment, # Mu, is transferred by flexure and the other part, Mu, is transferred by
eccentric shear. An applied direct shear force, Vu, is also to be transferred to the
Area Supported column as an axial load. The moment yy Mu acts as torque on the outer sides of the
column parallel to the span of moment.
To understand the mechanism of moment transfer from the slab to the column, consider
four imaginary cuts around the column in Fig. 12.18. The cuts A are present parallel to
K the span and cuts B are present perpendicular to the span. If cuts B are present in the
slab at the face of the column, but not on the sides (cuts A), all the imbalanced moment
must be transferred by twisting alone (called shear transfer). Similarly, if cuts A are
s' * present on the sides of the column, but not on the faces (cuts B), all the unbalanced
moment must be transferred through flexure alone (called moment transfer). In the
absence of the both the cuts, part of the moment is transferred by flexure and part by the
Area Supported I' x
twisting.
By Beam BD When gravity load, wind, earthquake, or other lateral forces cause transfer of
Fig. 12.17. Slab Tributary Areas for Beam Shears. unbalanced moment Mu between a slab and a colurnn, a minimum fraction of the
unbalanced moment between slab and column, Mu, must be transferred by flexure
STEP 14: Beam Design within an effective slab width, between Unes that are L5h (his the slab or drop panel
Design the beams, if present, both for flexure and shear. thickness) outside opposite faces of the column or capital. The numerical value of
is discussed in the next step.
STEP 15: Column Design Moments
Concentration of reinforcement over the column by closer spacing or additional
Interior Column: As per ACI clause 13.6.9.2, at an interior support, supporting reinforcement is to be used to resist moment on the effective slab width defined above.
elements (columns) above and below the slab must resist the bending moment given
below in direct proportion to their stiffness.
512 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDJqj
CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 513
However, the reinforcement ratio p within the effective slab width
0.375 pi,.
should not exceed

Where bi = width of the critical section for punching shear defined in the
direction of the span for which moments are determined, mm.
and bi = same in a direction perpendicular to bj.

Critical section
inside drop
panel ।
b) Plan

A/4
Critical
S x/4 for calculation
section of flexural top steel
c) Critical Perimeter outside (almost f /30)
drop panel Critical sections

da = effective depth outside the drop panel


Fig. 12.18. Slab - Column Connection with Unbalanced di = effective depth inside the drop panel
Moment. x = the distance from the edge of drop panel to the face of column or
STEP 17: Moment Transferred From Slab To Columns By Eccentric Shear column capital
Moment Transferred By Eccentric Shear: As stated above, after the Figure 12.19. Critical Shear Section for Slabs with Drop Panels.
moment transfer of
by flexure, the remainder of the unbalanced moment is transferred
by eccentricity of shear about the centroid of the critical
section.
Yy = ^~Yf
Critical section: The edges of critical section perimeter for punching shear as
inACl 11.12.1.2 are considered at distance d/2 from the following: defined For square column, bi = bi 40% of moment transfer by shear
If bj is very small nearly all transfer by flexure
a) Edges or corners of columns, concentrated loads, or
reaction areas, or If bi is very small nearly all transfer by shear
b) Changes in slab thickness such as edges of capital or
drop panels. '
The value of critical perimeter, b0, should be a minimum. The critical The critical section for punching shear is considered at a distance of dll from face of
perimeter for
interior columns is shown in Fig. 12.19. For columns closer to the edges, the critical column on each side. For unbalanced moments about an axis parallel to the edge at
section may extend up to the edge of slab if its distance from the ; exterior supports, the value of yr may be increased up to 1.0 provided that the following
column does not conditions are satisfied:
exceed neither half of the critical section dimension parallel to the
edge nor 4h.. as
shown in Fig. 12.20. Vu at an edge support 0.75 ^Vc
The values of fractions, #and yv, are calculated by using the expressions or Vu at an comer support 0.5 /Vc
presented here.
However, the ACI code allows the readjustment in the values according to specific
requirements. i For unbalanced moments, about an axis transverse to the edge at exterior supports, the
value of Yf is allowed to be increased by up to 25% provided that the following
requirement is satisfied:
514 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIqi TWO-WAY COLUMN SUPPORTED SLABS 515
CHAPTER 12:
Pu at the support 0.4 0 Vc

Direct Two-Way Or Punching Shear Force; The direct shear force, Vu, to be resisted
by the slab-column connection can be calculated as the total factored load on the area
bounded by panel centerlines around the column less the load applied within the area
defined by the critical shear perimeter. This is to be calculated both at the column
perimeter and at the perimeter of drop panel, if present, using the critical section
defined in Figs. 12.19 and 12.20.
Edge Of Slab
lesser than forces is shown.
1)^/2 c) Direct Shear
2)4A

by
Fig. 12.20. Critical Section for Edge Column.

Eccentric Punching Shear Force: As already discussed, eccentric shear is the two-
way shear developed due to transfer of moment to the column through torsion.
According to ACI 11.11.7.2, the shear stress resulting from moment transfer by
eccentricity of shear shall be assumed to vary linearly about the centroid of the critical
section. Consider a typical interior slab-column connection where an unbalanced
moment must be transferred between the slab and the column, as shown in Fig. 12.21.
Figure 12.21 (a) shows a typical connection of the slab with interior column, where b„= 2c! + 2ci + Ad
M part of slab moment is to be transferred by flexure, y„ part of slab moment is to
be transferred by eccentric shear and direct applied shear is to be transferred to the b) Critical Section For Shear
column as an axial load. Fig. 12.21 (b) shows the critical section for two-way shear
according to the ACI Code. The expressions for perimeter of the critical section (ba)
and the applied direct shear (Vv) are also shown in this figure. Figures 12.21 (c), (d)
and (e) represent the variation of direct shear stresses due to shear K, linear variation of
e) Resultant Shear
shear stresses due to twisting moment and the resultant shear stresses, respectively. The
direct shear stress (considering the direction of the applied loads) is downwards on all
Fig. 12.21. Two-Way Shear Acting on Critical Slab Section
around Column.
the faces, while the shear stress due to twisting moment is upwards on one side and
of a box-like shape
downwards on the other side haying linear variation in-between. The resultant shear Where, Ac and Jc are defined below and are calculated for the faces
stress on one side adds up and becomes critical for the punching of the slab.
defined by the assumed vertical failure section.
The resultant shear stress acting on the critical perimeter, considering moment acting section
from both the directions, may be written as follows: Aa = perimeter area of2(£>ithe+critical
« x
= = bi) d
Jc = torsional constant, like polar moment of inertia of the area

vu at face AB =
—A

±
Je‘1 —- ± ——Jc— —-
'■
ci
-
:
1
= 4+4
Torsional Constant For Interior Column; The critical area
subjected to punching
shear is a three dimensional area and hence the calculation of its
torsional constant is
516 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIqi ’ two-way column supported SLABS 517
CHAPTER 12:
not as simple as for any planar area. In order to get a reasonably good estimate. ; .
width of the area may be squeezed to zero but it is assumed that the original area is
maintained. This will change the area in Fig. 12.22 (a) to that in Fig. 12.22 (b), such
that the visible area has a numerical value equal to sum of areas for faces AD and BC
and the two edge lines represent the two areas AB and CD. After this simplification
the torsional constant can be considered equal to the polar moment of inertia (as for a
circular section). For the critical section over an interior column, the centroid will be at
the mid-dimension in both the directions.

= Ix for faces AD and BC + A for faces AB and CD


+ Iy for faces AD and BC + 4 for faces AB and CD

(a) Critical Perimeter Section


over Edge Column.
Edge Column.
Fig. 12.23. Critical Section for Two-Way Shear over
area for the comer column, as
Torsional Constant For Corner Column: The critical
and its centroidal distance along the 1-
shown in Fig. 12.24, is like angle section of steel first moment of all the areas about the
direction is again to be determined by taking the
right-most line.

(a) Critical Perimeter Section


Over Interior Column.

Fig. 12.22. Critical Section for Two-Way Shear over Interior Column.

Torsional Constant For Edge Column: The critical area in this case (Fig. 12.23) is
like a channel section of steel and its centroidal distance along the 1-direction is to be
determined by taking the first moment of all the areas about the right-most line;
(a) Critical Perimeter Section
2bid(bj2) bj over Edge Column.
= =
2byd + b2d 2bt+b2 Over Comer Column.
Fig. 12,24. Critical Section for Two-Way Shear
= (2hj + 62) d
b^/2) _b[_
r — o l jf Vl ft jW V
b^d + b2d
= 2bt + b2
two-way column supported slabs 519
518 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDlQj
chapters
At = (61 + d Critical
n n
inr
Section
Je = + +
12 12 1
^2 J
Concrete Punching Shear Strength (K): According to ACI 11.11.2.1, for non-
prestressed slabs and footings, Vc shall be the smallest of:

a) Pc = 0.17
^l+2^^ bad
If the column is very wide, it behaves like a wall and the factor inside the brackets
becomes close to 1, making the expression identical to beam shear.

P = of column, concentrated load, or section area,


shortside
60 = perimeter of critical section for slabs and footings.

(a) Stirrup Shear Reinforcement. of Channel Sections.

Fig. 12.25. Typical Shear Heads.


where a = 40 for interior columns
30 for edge columns
20 for comer columns
c) Pc = O^A^bod
Maximum shear stress due to the factored shear force and moment < <j> vn
Where<J>vn =
P's — 0, when no shear reinforcement is provided in slabs in the
form of shear heads.
— ci/2
Shear Head To Improve Strength Against Punching Shear: Four types of shear
heads, shown in Figs. 12.25 and 12.26, may be designed over the top of columns in case
the concrete strength done is not sufficient to resist the applied punching shear. Other
types are also possible by changing the arrangement and the type of structural steel
section. In these figures, 7V represents the length of shear head arm from centroid of
concentrated load or reaction. The detailed design of these shear heads may be found
from any handbook on concrete design.
According to ACI 11.11.3, shear reinforcement consisting of bars, wires, single leg
(a) Small Interior Shear Head
stirrups and double leg stirrups may be provided in slabs and footings with effective (b) Shear Head of Stud Connectors.
depth greater than or equal to greater of 150 mm and 16 times the shear reinforcement of Channel Section.
bar diameter. This reinforcement must engage the longitudinal flexural reinforcement Fig. 12.26. More Examples of Shear
Heads.
in the direction being considered.
TWO-WAY COLUMN SUPPORTED SLABS 521
520 CONCRETE STRUCTURES PART -I AUTHOR ZAHID A. SIDDIQI CHAPTER 12:
To calculate the shear strength with shear reinforcement, the maximum value of js bi = 600 + 190/2 + 200 = 895 mm
taken equal to 0A7A.JfJbod and Ks is not to be taken greater than 0.33j~fcbod when 2 bi = 1200+ 190 = 1390 mm
= 1.0. In other words, the maximum two-way shear strength cannot exceed ba = 2 bi +bi = 3180 mm
Mjf&d , even if shear reinforcement is provided. The area of shear reinforcement, a* = 30 for the edge column
least out of the following.
The concrete two-way nominal shear strength, vc, is the

—J —
Ay, is equal to area of all legs -of reinforcement on one perimeter of the column section.

I —0™
The distance between the column face and the first line of stirrup legs that surround the ( o'! ( 2 i
column must not exceed d/2. The spacing parallel to the column face between the i) 0.172 1 + = 0.17 1 + 2.0 V25
j
= 1.7 MPa
stirrups in this first line must not exceed 2d. The spacing between successive lines of I
shear reinforcement that surround the column must not exceed d/2 measured in a
direction perpendicular to the column face. ii) 0.083X| ^+2 1J/; =
I b„ /
0.083|^^+2|V25
I 3180 ) - 1.574 MPa

Structural steel I- and channel-shaped sections are also allowed in the slabs. Arms of = 165 MPa
the shear-head must not be interrupted within the column sections. The section should iii) 0.33X757 =0.33725
not be a depth greater than 70 times the web thickness of the steel shape. All
compression flanges of the structural shapes are to be located within 0.3d of 4> vc = 0.75x1.574 = 1.181 MPa
compression surface of the slab and these sections may be considered to be effective in = b0 x d = 3180 x 190 = 604,200 mm2
resisting the moments besides providing the shear strength. The ratio (av) between the
flexural stiffness of each shear-head arm and that of the surrounding composite cracked bi 895: = 252 mm
xi 2x895 + 1390
slab section of width (c2 + d) must not be less than 0.15. 2bl+b1
Example 12.1: Perform check for punching shear of a two-way slab system (Fig. 12.27) b^L + ^+b^2 i+(M)(xi)2
at the given edge column. The panel size is 6m x 8m and all conditions of direct design 12 12 1
method are satisfied. The other related data is as under:
= 2xfl?5xW0!.+ 190x«^.+895x|9oJ»5_252
= 11,000 Pa 2
[ 12 12 I2
(unbalanced) = 200kN-m
fa = 25 MPa ? + (1350xl90)(252)2
= 5,349,563 x IO4 mm4
h = 230 mm
d = 190 mm
Solution:
0 - longer / shorter sides ratio for the column = 2.0
200mm width of ci + 3h, to transfer
More flexural steel is to be provided near the column, in a
Direction of 65.1% of moment.
unbalanced at the inner edge and
moment 8m The direct shear stress and eccentric shear are both downwards
shear.
will add into one another to give the critical value of the applied
600mm Direct shear force, Zu [length x width hi £2] -
11,000
3.5 m
[(3.5X8) -(0.895X1.390)] = 294.3 kN
1000
Fig. 12.27. Slab System for Example 12.1. Direct shear stress = Zu I A< = 294.3 x 1000 / 604,200 = 0.487 MPa
i
i

522 CONCRETE STRUC TURES PART - 1 1i


chapter1* TWO-WAY COLUMN SUPPORTED SLABS 523
AUTHOR: ZAHID A. SIDDJqj

Eccentric shear
ti
I (
_
The column and middle strips are shown in Fig. 12.29.
2.5m
,
| I
,665mm square

0.349 x 200 x IQ6 x252 -& j E


5,349,563x10*
= 0.329 MPa
1.5mj 1.5tn • i
Total applied shear, m, = 0.486 + 0.329 = 0.815 MPa I

1.5m = Half
=> The slab is safe against two-way shear. Middle Strip
If vu > <|> vc, following solutions are possible: I

1. Design a shear head. B


I
2. Provide drop panels or column capitals. -i
3. Slightly increase the depth of slab if the difference Fig. 12.29. Equivalent Columns and Column and Middle Strips.
of vu and <}> Vc is smaller. 1
Example 12.2: Design
reinforcement for the
shown in Fig. 12.28. Assume that direct design interior panel of the flat plate floor system,
method is । 1. E-W Span
satisfied using a depth of 220 mm. Check the depth of slabapplicable and the depth criterion is i ti = 8m
for shear considering the effect of
eccentric shear jequal to 5% of the direct shear for this 4 = 6m
as under: interior panel. The other related data is
4 = 8.0-0.665 = 7.335 m
Live load
Floor finish and partitions —
=
300 kgs/m2
150 kgs/m2
4w
M
= 6.0 m

= (12~69) = 512.1 kN-m


/c' = 25 MPa
fy = 300 MPa
h Support Section (Top steel, E-W Direction)
= 220 mm
i M~ = 0.65 M = 0.65 (512.1) = 332.9 kN-m
| A = 4/4 =0.75
a p = 0 (no beam), D = 0
Percentage column strip moment out of negative moment = 75%
Column strip: 0.75(332.9) = 249.7 kN-m
Middle strip: 0.25(332.9) = 83.2 kN-m
Mid Span Section (Bottom steel, E-W Direction)
M* = 0.35 AT0 = 0.35(512.1) = 179.2 kN-m

Fig. 12.28. Typical Interior Panel of Slab System ! Percentage column strip moment out of positive moment = 60%
for Example 12.2.
Solution: J Column strip = 0.60 (179.2) = 107.6 kN-m
! Middle strip = 0.40(179.2) = 71.6 kN-m
Equivalent square column side, h = J— (750) = 665 mm
= 300 kg/m2
v4 J 2. N-SSpan
4 = 6m
q0 = 0.220 x 2400 + 150 = 678 kg/m2 4 = 8m
?u = [1.2^) + 1.6(?l)] x 9.81 / 1000 4 = 6.0-0.665 = 5.335m
= (1.2(678) + 1.6(300)] x 9.81 / 1000 = 12.69 kN/m2 s 4w = 8m
CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 525
524 CONCRETE STRUCTURES PART -I AUTHOR ZAHID A SIDDlQi
Table 12.13. Calculation of Slab Steel For Example 12.2.
=
(8 °X5-335)2
Mo (12.69) = 361.2 kN-m Design Location Strip Width M. R P As Steel
8 mm2
Frame mm kN-m =MJbd2
Support Section (T op steel, NS Direction) Support cs 3000 249.7 2.627 0.0106 1887 #19@150mm c/c
E-W Top steel MS 3000 83.2 0.875 0.0033 587 #13@200mm c/c
M~ = 0.65 Mq = 0.65 (361.2) = 234.8 kN-m Mid-span CS 3000 107.6 1.132 0.0046 819 #13@150mm c/c
Column Strip = 0.75 M~ = 0.75 (234.8) = 176.1 kN-m E-W Bot. steel MS 3000 71.6 0.753 0.0029 516 #13(ig250mm c/c
Middle Strip = 0.25 M~ = 1.853 0.0076 1353 #19@200mm c/c
0.25 (234.8) = 58.7 kN-m
N-S
Support
Top steel
CS 3000 176.1
0.371 0.0025 445 #10@160mm c/c
MS 5000 58.7
Mid-span CS 3000 75.8 0.797 0.0033 587 #13@200mm c/c
Mid Span Section N-S #10@160mm c/c
Bot. steel MS 5000 50.6 0.319 0.0025 445
M' - 0.35 Mo 0.35(361.2) 126.4 kN-m
Column Strip = 0.60 (126.4) 75.8 kN-m
Middle Strip = Table 12.14. Curtailment of Slab Steel For Example 12.2.
0.40 (126.4) 50.6 kN-m
Span Lengths Column Strip Middle Strip
The moments across the full width of both the column and the middle strips are + Vi Eq. Column Size + V, Eq. Column Size
Design
pictorially shown in Fig. 12.30. Frame
4 0.30 4 0.20 to 0.22 to 0.15 to
*
(mm) (mm) (mm) (mm) (mm) (mm)

E-W 8000 7335 2540 1800 1950 1430

N-S 6000 5335 1940 1400 1510 1130

The concrete two-way nominal shear strength, Vo, is the least out of the following:

i)
— x 2915x178/1000
0.17^[ 1 + 4p)|7£V = 0.17J1v +-2I )1725 i

:i I
= 1323 kN

Jmax
Fig. 12.30. Moments across Full Width of Strips for Example 12.2.

= 2h = 440 mm
! ii)
0.0832^^+2^777^^ 0.083^^^+2^725x2915x178/1000
=
= 956 kN
d = 220 -20-16-6 = 178 mm (lesser value for the inner steel) iii) O.Hljf&d = 0.33725x2915x178/1000 = 856 kN
Adn = 0.0020, = 0.002 1000x220 = 440 mm2/m width
x
$ Vo = 0.75 x 856 = 642 kN > 1.05 Va = 630.57 kN (OK)
The calculation of the steel reinforcement is shown in Table 12.13. The area of steel. The steel reinforcement is shown in Fig. 12.31.
As, is calculated for lm width. The distances for curtailment of bars are calculated in
Table 12.14. Example 12.3: Calculate the design moments for the exterior panel of the flat plate
system given in Fig. 12.32, perpendicular to the edge. The other related data is as under:
b0 = x (750+ 178) = 2915 mm
Ac = b0*d = 2915 x 178 = 518,870mm2 Clear cover = 20 mm
= 12.69x [8 x 6 - it/4 (0.928)2] = 600.54 kN Grade of steel = 420 MPa
Vo
Total shear = 1.05 V, = 630.57 kN
Superimposed q^
Live load ?l
— 150 kgs/m2
= 300 kgs/m’
= 20 MPa

i
CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 527
526
ix375mm Columns
TT

fi=6.0m

4=5.5m
Fig. 12.32. Flat Slab of Example 12.3.
Total Static Moment
0.2 x 2400 + 150 = 630 kgs/m2
300 kgs/m2
<1* [1.2(630) + 1.6(300)] x 9.81 / 1000 = 12.13 kN/m:
4w 5.5 m
_
8
12.13x5.5x5.6251
8 — zoJ.o kN-m

Longitudinal Distribution Of Moments

— Int M“ = 0.70Mo = -184.7kN-m

= 263.8 — M 0.52Mo = +137.2kN-m

— ExtM- = 0.26Mo = -68.6kN-m

Torsional Member
There is no edge beam and 375mm width of slab may be assumed to act as a torsion
member, as shown in Fig. 12.33.
y = 375mm

Fig. 12.31. Detailing For Slab of Example 12.2.


Solution: 5500mm
X-section in
Depth For Control Of Deflections L-section in ly direction
4 direction
4 = (6000 -375) /1000 =5.625 m Fig. 12.33. Torsion Member for Example 12.3.
^min. = 4/30 for fy = 420 MPa > 125 mm
= =
(5.625x1000) /30 187.5 mm 125 mm OK
Try h = 200 mm C Y 1-0.63-^ = 66,400 x 104 mm4
I yj 3

4'
TWO-WAY COLUMN SUPPORTED SLABS 529
528 CONCRETE STRUCTURES PART 1 - AUTHOR ZAHID A. SIDDlQj CHAPTER 12:
heads, and size of columns equal to 375x375mm. Beams are present on the edges and also
For slab width equal to length of torsion member, the moment of inertia is as follows; ’ ‘ ‘ t
along the interior column-centerlines.
' The overall size of long beams is 375x700mm and the
5500 x2003 overall size of short beams is 300x525mm.
= 366,667 x 104 mm4
12
375mm

2.75m
Actual Torque Variation
6M00X104
(2X366,667 xlO4) 2750-375/2 ; Panel E
375mm

= 2562.5mm
A = hHi = 5.5 /6.0 = 0.917
r, [
Assumed linear
variation of torque
B = A 0.09
Torque
(conservative 200mra
accroach)

0 =>

D = ^2 0 0=160mm at “4” distance
ah a{x y- from support Torsion Reinforcement
face
Transverse Distribution Of Moments
The column strip moment percentages are calculated as under:
. Int.AT = 75 + 30(1-^)79 =75%
Ext. AT = 100 - 105+ 12379(1 ->4) =99.1%
Af = 60+15(3-24)79 =60%

— Int. M~
i—
= 0.70M. = -184.7kN-m H*— MS
CS = 0.75(-184.7) = 138.5kN-m
= 0.25(-184.7) = 46.2kN-m

pCS = 0.60(137.2) = 82.3kN-m


= 263.8 M* = 0.52Mo = +137.2kN-m H
*— MS = 137.2-82.3 = 54.9kN-m

— Ext M- = 0.26M„ = -68.6kN-m


1
i— CS


= -68.0kN-m (99.1%)
MS = very small
Torsion In Slab At Edge Of Panel

The unbalanced moment to be transferred to the column acts as torque for the edge
beam. It may be assumed that half of this torque is resisted by edge beam on each side,
having length up to their midspan. Further, the variation of this torque from colunm-
centcrline to midspari may be taken as linear in place of actual higher degree variation Solution:
giving a conservative estimate of the torque, as shown in Fig. 12.34. The total torque at 8.0 m
any section of the torsion member is found just like the estimation of shear force for a Longer span
Shorter span 6.0 m
distributed load acting on a beam. The reinforcement required for shear and torsion are 3.50 m
combined and placed together (Fig. 12.34). However, due to difficulty in fabrication Story height
Column size 375 x 375 mm
and anchorage of stirrups, it is usually better to provide an edge spandrel beam.
Exterior longer span beam size 375 x 700 mm
Example 12.4: Design a flat slab system (Fig. 12.35) for 3 panels @ 6m in one direction, Interior longer span beam size 375 x 700 mm
4 panels @ 8m in the other direction, story height = 3.5m, service live load = 300 kg/m", Exterior shorter span beam size 300 x 525 mm
superimposed dead load = 350 kg/m2,// = 25 MPa, fy = 420 MPa, no drop panels, no column
530 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID AM TWO-WAY COLUMN SUPPORTED SLABS 531
Interior shorter span beam size
CHAPTER > 2:
300 x 525 mm b = 895 mm
Superimposed dead load 350kgs/m7
Live load ‘6’ is lesser of the following:
300 kgs/m2 For shorter L-beams, the effective width
Concrete cylinder strength
Steel yield strength
25 MPa . b .
420 MPa i) bw + hb
Cover to steel 20 mm = 300 + 345 = 645 mm A/
STEP 1: Check For Limitations Of DDM ii) bw + ^hj
A*
The first five restrictions of the direct design method are
checked below: = 300 + 4 x 180
i) There is a minimum of three continuous spans
in each direction. (OK)
= 1020 mm bw
H) = 8/6 = 1.33 <2 b = 645 mm Fig. 12.36. Effective Slab Width for L-Beams.
(OK) the following:
For longer T-beams, the effective width ‘b’ is lesser of
iii) Adjacent span lengths in each direction do
not differ. (OK) i) bv + 2 hb = 375 + 2 x 520 = 1415 mm
iv) All the columns are present in a single line.
(OK) ii) bw + 8 hf = 375 + 8 x 180 = 1815 mm
v) Only uniformly distributed gravity loads are
applied. The live load to dead load
ratio is expected to be less than 2.0 (To be checked later).
b = 1415 mm
following:
STEP 2: - - Selection Of Slab Depth For shorter T-beams, the effective width ‘6’ is lesser of the
In this step, the slab depth is decided according to ACI 9.5.3 provisions. i) bw + 2 hb = 300 + 2 x 345 = 990 mm
ii) 8^ + 44/ = 300 + 8 x 180 = 1740 mm
fn = length of clear span in long direction
b = 990 mm
= 8000 - 375 = 7625 mm beams are given in Table
The moment of inertia (Z/,) and other properties of these
Assume the value of arra to be greater than 2.0. 12.15.
P = ratio of clear spans in long to short direction Table 12.15. Calculation of Stiffness of Beams.
Exterior Interior Exterior Interior
7625/5625 = 1.356 Frame Short
Long Long Short
895 1415 .645 990
Total width, b (mm) 300
Web width, bw (mm) 375 375 300
increased by at least 10%, in the panels with a 700 700 525 525
36 + 9^ Depth, h (mm)
1,563,570 1,861,103 511,056 601,922
discontinuous edge without a beam having «i less Zb (xlO4 mm4)
than 0.8, but not less than 90 mm
(7625) fo.k 8+^-1 STEP 4: o^-Value And 6th Limitation Of DDM
may be entered in
1400/ = 174 mm (say 180 mm) The required values of Zs and a/ may be calculated- as follows and
36 + 9x1.356 Table 12.16. Further, the value of Ofm is calculated for the comer panel.
STEP 3: Beam Stiffness
4 = value of slab inertia effective for £] direction.
For longer L-beams (a typical L-beam is shown in Figl2.36),
the effective width 'b’ is
lesser of the following:
12
i) bw + hn = 375 + 520 = 895 mm
=
Beam stiffness = E^Z**
ii)
^ + 4AZ= 375 + 4 x 180 = 1095 mm df\
Slabstiffness £MZV
532 CONCRETE STRUCTURES PART -I AUTHOR: Z AHID A.
SIDDiQi
ctfin

=
average value of Of for beams at four edges of the
5.13 > 2.0
comer panel CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 533
6: Calculation Of Factored Static Moment For Each Design Strip
The depth already considered is OK.
DL = 350 + 0.180 x 2400 = 782 kgs/m2
• afl ^2
Now the ratio

is calculated using on and fi for the short exterior
and /2 for the long interior beam to get one
extreme answer.
beam and
LL
LL/DL
qa
= 300 kgs/m2
= 300/782 = 0.384 < 2.0
= (1.2 x 782 + 1.6 x 300) x 9.81 / 1000 = 13.91 kN/m2
(OK)

The values of total static moments are calculated for each of the design frame, using the

T2 - 6.38x6.02 - 0.70

The 6th condition of direct design method is


0.2
following formula, and are filled in Table 12.18.

Mo = hhiltL
8
satisfied.
Table 12.18. Total Static Moments All Design Frames.
Table 12.16. Calculation of Relative Sti:
Exterior Interior Exterior Interior
Exterior Interior Frame Long Short Short
Frame Exterior Interior Long
Long Long Short Short (m) 3.188 6.000 4.188 8.000
A (x IQ4 mm4) 1,563,570 1,861,103 511,056 601,922 4(m) 7.625 7.625 5.625 5.625
f 2w (mm) 3188 6000 4188 8000 Mo (kN-m) 322.3 606.7 230.4 440.2
A(xl04 mm4) 154,937 291,600 203,537 388,800
10.09 6.38 2.51 1.55 STEP 7: Longitudinal Distribution Of Moments
STEPS: Torsional Stiffness Of Edge Beam Factored M at interior supports = 0.65 Afo
The dimensions of edge beams to act as ; Factored M+ at interior mid-span = 0.35 M>
torsional members are found to be the same as
the flexural members. Compute the
torsional constant, Cs for all the edge beams. The From Table 12.6, the distribution of moments in the exterior panel must be as under:
factor pt is also calculated by the expression given M at first interior support = 0.70 M>
in Table 12.17.
below and all the results are entered
M+ at exterior mid-span = 0.57 M>
M at exterior support = 0.16 M>
= £ 1-0.63—
I y) 3 The resulting moments at various critical sections after the longitudinal distribution are
shown in Table 12.19.
A = E*c
Table 12.19. Moments After Longitudinal Distribution.

Table 12.17. Torsion Properties of Beams. Exterior Interior Exterior Interior


Frame Long Long Short Short
Frame Long Short (kN-m) (kN-m) (kN-m) (kN-m)
Span of edge beam 322.3 606.7 230.4 440.2
Shorter Longer Mo
C (xlO4 mm) 347,423 894,229 M at ext. support 51.6 97.1 36.9 70.4
Span of torsion member, (mm) 6000 8000 M* in ext. span 183.7 345.8 131.3 250.9
4 (xl 04 mm) for width 291,600 388,800 M at first int. support 225.6 424.7 161.3 308.2
Pt 0.596 1.150
M at typical int. support 209.5 394.4 149.8 286.2
M* in interior span 112.8 212.3 80.7 . 154.1
534 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDlQj
CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 535
STEP 8: Transverse Distribution Of Moments
The column-strip moment percentages are calculated in Table 12.20 using Table 12.21. Design Bending Moments For Various Elements.
following expressions: the
Frame Location Ext. Int. Ext Int
>1 0.5≤^≤2.0 Long Long Short Short
Pt B = If £>2.5, 5=2.5 kN-m kN-m kN-m kN-m
Beam moment 42.0 79.1 26.3 50.2
a fi— = D = If ap— > 1.0, D=1.0
M* ext. span 128.8 242.5 72.5 138.6
Interior negative moment (%age): 75 + 30(1 —
Exterior negative moment (%age): 100-105+ 1 25D (1-^4)
first int.
158.2 297.8 89.1 170.3
146.9 276.5 82:8 158.1
Positive moment (%age): 60 + 15(3 -2A)D
Beam moment out of CS moment (%age): 85 * D A/' int. span 79.1 148.9 44.6 85.1

ACI Code does not clearly specify that which width of slab is to be used in calculation Column strip slab moment 7.4 14.0 4.6 8.9
of a fi. However, as a common practice, the width of the edge panel (^w) is used for M* ext. span 22.7 42.8 12.8 24.5
this purpose. The percentages of column strip moments that are to be resisted by 52.6 15.7 30.0
27.9
column line beams are also given in Table 12.20. The slab portions of column strips first ini.

are proportioned to resist those fractions of column strip moments that are not resisted ^l. 25.9 48.8 14.6 27-9
by beams. The resulting design moments are presented in Table 12.21. 26.3 7.9 15.0
M+ int. span 14.0
Middle strip moment ^ext. 2.2 4.0 5.9 11.3
Table 12,20. Column Strip And Beam Moment Percentages. .
ext. span 32.2 60.5 46.0 87.8
Frame Ext. Int.Ext Int
Long Long Short Short 39.5 74.3 56.5 107.9
first int.
A = /2//| 0.750 0.750 1.333 1.333 36.7 69.0 52.4 100.2
B= p 0.596 0.596 1.150 1.150 19.7 37.2 28.2 53.9
M+ int. span
afi 10.09 6.38 2.51 1.55
D = ctfi t-Jt\ 1.0 1.0 1.0 1.0
%age of M' to column strip 82.5 82.5 65.0 65.0
%age share of beam from above moment 85.0 85.0 STEP 10: Torque For Edge Beam
85.0 85.0
%age of M to column strip 82.5 82.5 65.0 65.0
ACI 13.6.3.6 recommends that the gravity load moment to be transferred between slab
%age share of beam from above moment 85.0 85.0 85.0 85.0 and edge column in accordance with ACI 13.5.3.1 is to be 0.3 Mo. For example, torque
%age of to column strip 95.8 95.8 83.9 83.9 in the edge beam near an edge column along the shorter span may be calculated as
%age share of beam from above moment 85.0 85.0 85.0 85.0 follows:
I 0.3 V, = 0.3x606.7 = 182.0 kN-m
; Moment resistance provided by the beam = 79.1 kN-m
STEP 9: Calculation Of Slab Reinforcement t Half of the balanced moment on each side must be the resultant torque in the edge
The slab steel may be calculated from the slab moments by using the usual under-reinforced fO, beams near the columns.
concrete design formulas. A,min = 180x1000x0.0018= 324 mm2/m, fc'~ 25 MPa,/, = 420
MPa, d = 142 mm and let minimum spacing of steel = 200mm. The results are entered in I
;
Resultant torque in the edge beam = 51.5 kN-m
This torque is uniformly distributed along half span of the torsion member, giving a
Table 12.22.
resultant torque of 51.5 kN-m near the column.
536 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIqi TWO-WAY COLUMN SUPPORTED SLABS
CHAPTER 12: 537
Table 12.22. Slab Steel Areas For Various Frames. iii) The special reinforcement shall be provided for a distance in each direction from the
comer equal to one-fifth die longer span, equal to 1600mm in this example.
Frame Location Ext Int Ext Int
Long Long Short 12.11 EQUIVALENT FRAME METHOD
Half Ct) width minus 1313 1313
beam width 1388 1350
Equivalent frame method (EFM) is described in ACI 13.7 and is a general method for design of
As for column strip 425 425 450 two-way column supported slab systems, without the restrictions of the direct design method.
ext. 437
Dia, and no. of bars
6# 10 6 #10 However, this method is only applicable in case of gravity loads and all general provisions for
4#13 4 # 13 except those of ACI 13.6, are also applied in this method. The three-
A for column strip two-way slabs,
M+ ext. span 425 425 450 437
Dia, and no. of bars dimensional slab systems are first divided into two-dimensional design frames by cutting at the
6# 10 6#10 4# 13 4# 13 panel centerlines. The removal of the torsional links between various design frames makes this
As for column strip 525 493 450 method conservative in nature. The longitudinal distribution of moments for these design
* Jim w. 437
Dia, and no. of bars
4# 13 4 #13 4 # 13
frames is carried out by performing actual 2-D frame analysis. This requires some
4# 13 modifications before implementation. Firstly, equivalent column stiffness is to be calculated
As for column strip 486 457 450 437 combining the effects of actual column stiffness, unsupported edge of slab and the torsion
Dia, and no. of bars 4 #13 4# 13 4# 13 member. Secondly, the variation of moment of inertia of the horizontal member along its
4# 13
As for column strip M+ int. span 425 425 450 length between the column centerlines is to be considered. Thirdly, the variation of moment of
437
Dia. and ho; of bars 6 #10 6#10 Ml# 13 inertia of the column between the centerlines of horizontal members must be considered. The
4 # 13
horizontal member in the equivalent frame consisting of slab, beams (if present) and drop
Half MS width 1500 panels (if present) is termed slab-beam. There are four main steps involved in the use of the
1500 2500 2500 equivalent frame method.
As for column strip ext. 486 486 810 810
Dia, and no. of bars 7# 10 Step 1: The 3-D slab system is represented by four or more 2-D frames, just like in DDM.
7 #10 12 #10 12# 10
As for column strip These design strips (or design frames) are separately considered for analysis and
M+ ext span 605 568 943 900 design.
Dia, and no. of bars
5 #13 5 #13 8# 13 7# 13 Step 2: The stiffness of frame elements is determined considering the facts that the slab is not
As for column strip 748 702 1166
first int. 1112 supported along full width at the edge, torsion member is present but its effect can not
Dia, and no. of bars
6 # 13 6# 13 9 #13 9 # 13 be included directly in the 2-d analysis and the columns and the slab-beam are non-
As tor column strip 492 650 1080 prismatic members. Torsional stiffness of transverse beams and the slab edge
1030
Dia, and no. of bars condition at the junction with the outer column are included in the column stiffness
6 #13 5 # 13 9 #13 8 # 13
A, for column strip and thus the concept of an equivalent column is used. ACI Code allows the analysis
int span 486 486 810 810 of a particular floor of the building by considering a free body of that floor-with the
Dia. and no. of bars | 7# 10 7# 10 12 #10 12# 10 columns below and the columns above (if present), with the far ends of these columns
taken as fixed. This simplification is very useful for hand calculations.
STEPH: Special Corner Reinforcement
Step 3: The 2-D frames obtained in the above step are analyzed for full gravity loads (pattern
If any one beam at the comer is provided with a loading is not considered if live load is within certain percentage of the dead load).
value of a greater than 1.0, special top
and bottom slab reinforcement is be provided This is equivalent to the longitudinal distribution of total static moment in the direct
at exterior corners (ACI 13.3) in
accordance with the following: design method.
i) Moment for this top and bottom steel must Step 4: The negative and positive moments (M- and M+) are distributed laterally to column
be equal to (per meter width) in
the slab. This is equal to 481 / 1.313 or 367 mmVm
[#10@190mm c/c] for the strips and middle strips using coefficients of DDM if the following limitation is
given slab system. satisfied:
ii) The bottom steel is provided perpendicular
to the diagonal and the top steel is
placed parallel to the diagonal.
TWO-WAY COLUMN SUPPORTED SLABS 539
12:
538 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. STDDlQ] CHAPTER
12.11.1 Slab Beams
As already written, the horizontal members in the equivalent frame are referred to as
slab-beams. According to ACI 13.7.3, moment of inertia of slab-beam outside the
supports is taken for gross area of concrete consisting of slab and beam (if present)
This value is to be increased for the portion inside the column, bracket, or capital by
dividing the moment of inertia just outside this region with the quantity (1 -
where C2 and are measured transverse to the direction of the span for which moments
are being determined. The increase in stiffness of slab-beam inside the column is
relatively lesser as the column is not acting as a continuous support throughout the slab
section and a greater portion of slab is hanging free. When a slab is resting on wall, the
support dimension will be equal to the frame width /2 and the slab-beam moment of
inertia within the support will become infinite.
I
Figure 12.37 graphically represents the variation of moment of inertia along the length I Variation of Moment of Inertia
of the slab-beams. In case of left side slab system, the moment of inertia h is to be Variation of Moment of Inertia
calculated for a T-beam of flange width equal to h, thickness of flange equal to the 12.37. Stiffness Variation in Case of Slab-Beams.
Fig.
depth of slab, rib width equal to the width of the drop panel and projected depth of rib
top of slab-beam to its bottom is considered
equal to projected depth of drop panel. The value I\ is to be calculated for a rectangle Moment of inertia of a column from theportion is continuously stiffened by the slab¬
of width ^2 and depth equal to the depth of slab. Similarly, in case of right side slab equal to infinity. The column in this of
beam, which has large dimension in the
direction of column bending. The variationThe
system, the moment of inertia /2 is to be calculated for a rectangle of width h and depth typical cases is shown in Fig. 12.38.
equal to the depth of perpendicular beam and the moment of inertia It is to be column stiffness along the height for some properties may be determined for hand
calculated for a T-beam having flange width of /2, thickness of flange equal to depth of approximate equivalent prismatic column
slab and a rib equal to the beam along the span (1-direction). calculations by using the following expressions:
For hand calculations, the non-prismatic slab-beam member may approximately be 4 = c/c height of the columns
converted into a prismatic member by modifying the fixed end moments, flexural £u = unsupported height of column
stiffness and carryover factors as under:
FEM = mxq£2v<Ii2 r
— slab centerline
= vertical distance starting from themid-height
up to inner end of
of column capital (Fig.
slab, inner end of drop panel or
Ksb - kxEI^J
_ depth at drop panel 12.38)
depth of slab a-end = column end near the slab to be analyzed
analyzed
ci = larger of column or capital width at the top in the direction of b-end = column end away from the slab to be
calculation of moments. ra = r-value at a-end of column
C2 = larger of column or capital width at the top perpendicular to the tb = r-value at b-end of column
direction of calculation of moments.
z \ 0.015 Ke,a = ka* Ele / te
x 0.08 z X2.7
m = 0.09 P-x-^ a024 0.083
= 4.0
z

d >4.0
z 0.05
k&
Id I*J

X

k = 5.3 I *— I a09 > 4.0 for t^l = 0.4 to 2.2 and up to 1.2

z \0.02

COF = 0.57 pi-x-^ a037 > 0.5


CHAPTERS: TWO-WAY COLUMN SUPPORTED SLABS 541
540 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDdiqj of this diagram at the middle of the column has been chosen to give a total area equal to
1.0 and the corresponding ordinates at the column centerline, column edge and at
distance ‘x’ from the free end are also shown in the figure. Figure 12.39(c) gives the
‘f for top column diagram of resultant torque at each section equal to the area of diagram-b to the outside
dividing the
7’ for bottom column of this section. Rotation per unit length at each section can be found by
is the diagram of
ordinates of diagram-c by the torsional rigidity (CG) and the result
Fig. 12.39(d). Note that the relative rotation within the column is zero. The total
rotation at each section considering free end as the reference point is shown in Fig.
12.39(e). The total twist of the end of an arm relative to the column is the summation
per
Variation of of the twists per unit length and is equal to the area of the diagram of twist angle
Moment of Inertia unit length in diagram-d.

Resultant torque at distance x = area of diagram-b from end up to distance x

‘f
1 4x 2x2
= —2 x—xx = Z23-
4 —
Z = oo for bottom = area of diagram-d in distance
Total rotation at distance x
column

Variation of
Moment of Inertia 2
Fig. 12.38. Stiffness Variation of Columns along Their Heights.
3£22CG
12.11.2 Torsional Member And Torsional Stiffness (Kt)
Torsional members are very important component of the slab system, as these are the Total rotation at distance x 6CE
only members that connect the 2-D design frames together to get the
actual 3-D Considering x =
behavior. They also affect the rotational constraint available at the outer end (end For poisson’s ratio approximately considered equal to zero, G = E / 2.
condition) of the 2-D design frames and hence modify the distribution of
moments. 0, the total rotation of the free end with respect to the column may be calculated as
The dimensions of these members have already been explained in the direct design follows:
method and the same are applicable here.
Detailed 3-D analyses of various slab systems suggest that an approximate value of the
^end 6CE
torsional stiffness can be obtained by assuming a torque distribution along the torsional The average rotation, ^v> of the slab end may be found by evaluating the area of
member that varies linearly from a maximum at the center of the column to zero at diagram-e and dividing by the length,
the
middle of the panel, as shown in Fig. 12.39(a). The distance from centerline of the
panel to the edge of column is calculated as under:
Cantilever slab distance = ^/2-c2/2 T \ *2 J

1--
2 I 2 2 I £ 4
In order to find the torsional flexibility of free end of the slab (which is the rotation
produced as a result of unit force), a unit torque is applied at this end, as
shown in F ig. 4 6CE I,
12.39(b), half of which going to each arm. This torque is distributed in such a way that
the torque variation is linear and the ordinate at the panel centerline is zero. The height
1
542 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. S1DDIq( CHAPTER 12: TWO-WAY COLUMN SUPPORTED SLABS 543 I
i

After further simplification, the following result is obtained:

a - ^o-^H3^4 , 4£2J
18CX

= assuming Ci «
,£2/9
/Q
18 CE

Finally, the torsional stiffness of one arm is calculated as


(a) Section of Design Strip
~*™EC
T 9E(J
Xt (one arm) = — = —~ —
G(1-c2/G)
= —

In general, for cases where the arm is present on both sides, the total torsional stiffness
is calculated as summation of the torsional stiffness for each arm.

r . y

Torsional stiffness ‘Xt’ of the edge beam depends on the flexural stiffness of the
longitudinal beam because when torsional member with flanges twist, it causes bending
of the longitudinal beam. According to ACI 13.7.5.2, where beams frame into columns
in the direction of moments (1-direction), the torsional stiffness shall be multiplied by
the ratio of moment of inertia of slab with such beam (7Sb in Fig. 12.40) to moment of
inertia of slab without such beam (4 in Fig. 12.40).

Ka = Kt^-
T
(c) Resultant Torque At Each Section

Fig. 12.39. Modification of Torsional Stiffness, Kh Due to Beam.


(d) Angle Change Per Unit Length
12.11.3 Equivalent Column Stiffness (Kec)
Equivalent column is a fictitious single element consisting of the columns above and
M-c2/i2y below the floor and attached torsional members. The stiffness of the equivalent
6CE column, Kec, represents the combined stiffness of the columns, attached torsional
members and the free slab edge. The value is lesser than actual column stiffness
(e) Total Rotation At Sections Considering because the purpose to get a reasonable estimate of the end condition for the slab-beam
Free End As The Reference Point to be used for analysis. When the end of slab beam rests on a monolithic reinforced
concrete wall, the end of the slab-beam and the wall rotate by equal amount at the joint.
Fig. 12.39. Calculation of Slab Rotation Due to Applied Torque.
544 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A.
SlDD1Qr TWO-WAY COLUMN SUPPORTED SLABS 545
CHAPTER 12:
However, when the end of the slab-beam rests on a column, greater width of the
not directly supported by the column and the torsional member (even if
slab is • In cases where pattern loading is required, three-fourth of the actual live load is
present)
contribute in less proportion to provide the rotational constraint- The result is that
used for analysis. However, the factored moments calculated in this way should not
be less than those occurring with full factored live load on all panels. Reduction of
width of the slab directly connected with the column rotates by amount equal to thatthe
of live load for pattern loading is considered because the equivalent frame method is a
the column. However, the overhanging slab undergoes larger rotation that
away from the column towards the panel centerline. increases conservative method of design.
As is explained above, the average rotation of the edge of the slab is greater than
the 12.11-5 Correction Of Moments To Face Of Supports
rotation of the end of the column. For analysis of slab-beam for moments using 2-D
methods, a fully monolithic joint is to be considered between the column and the For design, the negative moments must be evaluated at the critical sections defined in
horizontal member. Hence, the actual column stiffness is to be reduced to allow more ACI 13.7.7. These critical sections in cases of columns, drop panels and column
rotation at the junction of the two elements. To account for this in slab analysis, an capitals are shown in Fig. 12.4 1.
equivalent column is considered which provides the above average rotation of the slab. Method,
If the slab system to be designed meets the requirements of the Direct Design
#ec - + ^iv the total design moments in a panel can be reduced so that the absolute sum of the
moment, Mo. Referring to Fig.
where 0^c = rotation of equivalent column and the average M” does not exceed die statical
12.41(b), flie computed moments Mi, Mi and M3 are multiplied by ‘F* where:
0c —
rotation of actual column
Osa - average rotation of slab edge K ^1+^2
also,Jet, 8/1 2
^ec — stiffness of equivalent column
YKc = sum of actual stiffness of columns above and below the slab
Kt = sum of torsional stiffness of the perpendicular torsion members
From the basic stiffness equation, we have,
K =M!0^> 0 = M!K
The above equation for may be written as follows:
Af Af Af
^EC
_L = _J_ + _L
^BC
If no torsion member is considered, 1 / Kt becomes infinity and Kec is reduced to zero
(near a roller support).

12.11.4 Pattern Loading


According to ACI 13.7.6, the analysis may be simplified under the following conditions
by avoiding pattern loading:
• If the live load is not exceeding three-fourth of the dead load, no pattern loading is
required and the analysis is carried out for full load on all the panels. This is due to Fig. 12.41. Critical Sections for Negative Bending Moments.
the fact that the slab is sufficiently ductile in flexure to allow moment
Fig.
redistribution. t Example 12.S: Find the properties of frame taken from a slab system and shown in
12.42 required for design of an interior design panel, having frames 8.5m on centers and
columns of size 450mm x 250mm. Service live load is to be 200 kg/m2 for rooms and
ordinary
546 CONCRETE STRUCTURES PART - 1 TWO-WAY COLUMN SUPPORTED SLABS 547
AUTHOR: ZAHID A.
SlDDlQl CHAPTER12:
balconies, while it is to be considered 300 kg/m2 for public 450
height is 2.75m. Separate floor finish and permanent partitionbalconies. Center-to-center sto = 0.067
~

wall load is 150 kg/m2. 6750


Solution:
Slab Loads:
C2 2^1 = 0.029
8500
200mm slab: qD = 200/1000 (2400) + 150 = 630 kg/m2
9« = [1.2(630) + 1.6(200)] x 9.81 / 1000 = 10.55 kN/m2
2.75m
1
9.98x8.5 10.55x8.5 89.7kin/w. 11.83x8.5
2.75m
=84.8k7l/ =89.7kK/ =!00.6kn/

K 6.75m r 6.75m 1.75m


2.5m

Fig. 12.43. Isolated 2-D Frame for Analysis.

89.7 kN/m 89.7 kN/m (100.6X1 -75)2


265.0 k 44 +4 2
kN-m
=154.0 kN-m

W
6.75m 6.75m
Fig. 12.44. Simplified Analysis Model.
Fig. 12.42. Frame Dimensions for Example 12.5.
190mm slab: td = 190/1000 (2400) + 150 = 606 kg/m2
qu = 11.2(606) + 1.6(300)] x 9.81 /1000 = 11.83 kN/m2
180mm slab: qD = 180/1000 (2400) + 150 = 582 kg/m2
qu = [1.2(582) + 1.6(200)] x 9.81 / 1000 = 9.98 kN/m2
L.L.
For left balconies: = 200 = 0.345 : 0.75
D.L. 583
L.L.
For right balconies: = 252 = 0.495 <: 0.75
D.L. 606 Fig. 12.45. Moment of Inertia Diagram for Slab-Beam.
Pattern loading is not required to be considered.
A typical story for hand calculations is shown in Fig. 12.43. The FEM (uniform load, q) = mxq fi2
frame may be simplified for (stiffness) = kxEIaJ / = slab depth
analysis by replacing the cantilevers with the corresponding
Fig. 12.44. The stiffness parameters for the slab-beam
determinate moments, as shown in
under-
of Fig. 12.45 may be calculated as kEt/
12/,
Carryover factor = C.O.F.
548 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDlQi
TWO-WAY COLUMN SUPPORTED SLABS 549
a =
at —j q for s}afo without drop panels CHAPTER 12:
depth of slab prismatic column may be found by using Fig. 12.47.
The properties of equivalent


z xO.015
m = 0.09 x^- a024 0.083 = 2750 mm
= tb
= 0.09 (0.067 x 0.029)° 015 x1.0 ° 24 = 0.083 = 2550 mm
z \Q.05
ra = 100 mm
k = 5.3 [— x^| a0-’ 4.0 tb = 100 nun
1^1 ^2/ 4/4 = 2750 / 2550 = 1.078
= 5.3 (0.067x0.029)005xl.0°’ = 4.0
COF = 0.57 (0.067 x 0.029)omxl.0OJ7 = 0.503
FEM = mxq 42
-4
° &r&r
forta/rb = 0.4to2.2and4/4upto 1.2
"4-°

p-
jKd.h
=

(0.083) (89.7) (6.75)2 =

ki

~
_ 339.2 kN-m
4.0 (8500)(200)’
i E 3358 — — __ _
E kN-mm/rad jq.
== 4.0 (l.O)008 (1.078)27 = 4.899
= k*.EIcl£e
_
12£j 6750x1000 4.899 (250)(450)^
= 12x2750x1000 E 3382 E ^.mm/rad
The torsion member may be considered equal to a width of slab equal to the column dimension
as shown in Fig. 12.46 and the relevant parameters are calculated as under:

C =
lfl
-0.63-1^
yj 3
(U'J
= 0.539
= 1-0.63 -
i
(450)—
a</2°0^(200)’(450)<
3
= 86400 x 104 mm44 (1.0)0

t,=100mm

£c=2750mm' 4=2550mm

l^lOOmm

450mm
Equivalent torsional members
Fig. 12.47. Column Properties for Example 12.5.

Fig. 12.46. Torsion Member for Example 12.5.

V 9^c
= Zi = (2) 9Ecf(86400xl04) : = 2001 E kN-mm/rad + _J—.

z >1

l t.
(8500)Tl—
v
I 8500
1 1000 2(3382 E) 2001E
= 1545 E
550 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A CHAPTER »2: TWO-WAY COLUMN SUPPORTED SLABS 551
The equivalent column stiffness
may
However, the stiffness of these columnsseparately be calculated for the interior j For lateral load analysis, the usual equivalent frame with
lateral loads and add the results.
moments almost neutralize each other. does not appreciably affect the results as the columns -r-x-ked EI values underestimates the lateral deflections and hence the P-A effects.
Here the stiffhess of these columns beam
Considering the frame of Fig. 12.48, the may be taken equa|
to the outer columns stiffhess.
Recording to ACI 13.5.1.2, analysis of unbraced frames for lateral loads should take into
calculated as under. distribution factors are the effects of cracking and reinforcement on stiffhess of frame members. Cracking of
account due to gravity loads and volume changes reducing the flexural stiffness of slab
jlabs occurs
^bd =

^db =
DFm
DFW =
- 3358£~
1545£ + 3358£
3358^
0.68
members. The lateral
usually
quarter
appropriate
of
considered
the
equal to
to
flexibility is further increased when lateral loads are applied. It is
reduce
uncracked
0.35Zg
flexural stiffness of slab to a value ranging from one-half to one-
stiffness.
and the
For cracking alone, the slab-beam stiffness „ may be
actual column stiffness may be considered equal to 0.7Zg.
2x3358£- + 1545jE' 0.404
After analysis of frame of Fig. 12.48, For lateral loads, the frame is to be considered as a whole unlike gravity loads where each floor
the following steps are required to design
the frame: can be considered separately. The drift must be equal for all the frames if there are frames in
• Correct moments to face of the supports. the perpendicular direction and connected together in a 3-D pattern. To obtain equivalent 2-D
frame, various 2-D frames may be arranged in a single plane connected as shown in Fig. 12.49.
• Distribute into column strips and middle strips by
using the expression of the In frames subjected to gravity loads, moments are basically generated in slabs. The average
design method. direct rotation of slab end is more than rotation of the actual column and hence column stiffness is to
• Designthe column and middle strip slabs. be modified to account for this flexibility of the torsion member. When lateral loads are
• Cany out detailing of the reinforcement. applied on frames, moments are generated in columns. Here, the column end condition is to be
• The column moments may be evaluated by satisfying the equilibrium at adiusted and not its stiffness. The concept of equivalent column is not applicable but an
junction with the slab. The their equivalent slab element is to be somehow considered.
unbalanced
and bottom columns according to ratio slab moments are distributed into the top One approach in case of lateral loads is to consider the transfer of moments through torsional
of their stiffness.
links to slab elements. The procedure in this case may be summarized as under:
• Use the cracked stiffhess for the column and the slab-beam.
• Consider torsional links between the slab-beams and the columns. This means that
moments may be transferred from columns to slab-beams only through these links, as
shown in Fig. 12.50.

wind
|

| Frame-3

Fig. 12.48. Final Analysis Frame


for Example 12.5.
U
Frame-2
Frame-1

12.12 ANALYSIS OF SLAB


SYSTEMS FOR LATERAL LOADS
A frame, without shear walls or
other bracing elements, is inefficient in
resulting in significant lateral drifts. The resisting
It is better to brace the slab systems lateral deflections are amplified by the lateral loads
by shear walls and the unbraced P-A moments.
only used for low buildings or for
the top slab-column frames are
lateral drift is possible, it is necessary to few floors of a tall building. In cases where more
analyze equivalent frame structures
for both gravity
Fig. 12.49. Equivalent 2-D Frame to be Considered for Lateral Load Analysis.
552 CONCRETE STRUCTURES PART - 1 TWO-WAY COLUMN SUPPORTED SLABS 553
AUTHOR: ZAHID A. CHAPTER 12:
SDDiqj
carryover operations and calculate tire resulting moments.
• Carryout distribution and Knowing this base
the total shear at the base of columns from their moment
• Calculate the the ratio by
assumed moments and the actual horizontal shear, calculate
shearfor
which the moments are to be corrected.
• Calculate the corrected moments.
stiffness by including the effect of torsional member in
Third method is to increase thethecolumn
torsional link is divided into upper and lower columns in
the
the column. The stiffness of
Fig. 12.50. Model for Lateral Load Analysis
According to First Option.
• Consider carryover factor for slab-beams and the column equal to 0.5 and that
.
torsional link equal to -I This means that torque from one side is for the
ratio of their own stiffness (Ki
respective column stiffness (Fig
K

+ Kt
and Ai

12.52).
K

* , respectively) and is added into the
Kv+ Kt we add
This is opposite of gravity loads, where
the
other with the same magnitude but opposite in sense. transferred to the torsional members. The remaining analysis is just like
flexibilities of actual columns andtitled “Frame Analysis of Concrete Buildings” by Vanderbilt
• Considering the stiffness of columns, torsional links and slab-beams, other two methods. The article 1983, presents details of these methods of solution
for
distribution factors at each joint calculate the and Corley, Concrete International, Dec.
lateral loads.
• Assume some reasonable values of column end moments in multiples loads and the lateral loads and the results are
ratios. of their stiffness The frames are separately analyzed for the gravity
then super-imposed.
• Carryout- distribution and carryover operations and calculate the resulting
moments.
• Calculate the total shear at the base of columns from their moment.
Knowing this base 12.13 SEISMIC PROVISIONS FOR TWO-WAY
shear for the assumed moments and the actual horizontal
shear, calculate the ratio by SLABS WITHOUT BEAMS
which the moments are to be corrected.
ACI, two-way slabs without beams
• Calculate the corrected moments. According to Portland Cement Association (PCA) Notes on
of low or moderate seismic risk, but are not
Second approach may be to calculate equivalent slab are valid lateral force resisting systems in regions less strict requirements for detailing for
stiffness (K^) having the collective effect
ofK^and Ki. The procedure may be outlines as follows: allowed in regions of high seismic risk. There are requirements are to be satisfied for the
ordinary moment resisting frames. The following
• Use the cracked stiffness for the column and the slab-beam. intermediate moment resisting systems (ACI 21.3.6):
• Calculate the reduced slab beam stiffness, Ka by considering its reciprocal equal to sum
of reciprocal of K^a and reciprocal of (Fig.
Kt 12.51).

Fig. 12.51. Concept of Equivalent Slab Stiffness.

• Now consider the frame as usual without the


torsional link.
Load Analysis.
• Assume some reasonable values of column end moments in multiples of their stiffness Fig. 12.52. Concept of Equivalent Column for Lateral
ratine
554 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. StDDlQj TWO-WAY COLUMN SUPPORTED SLABS 555 ft
CHAPTER 12:
. The design moment at support section should be
determined for the PROBLEMS
involving earthquakes. The reinforcement calculated in the slab for load combinations
this moment is to
.
be placed within the column strip. h
I?;
Reinforcement calculated for yjMsiOb is to be placed within the effective Note: N = Registered Number.
width, which
for exterior and comer connections should be a maximum
of ci plus lesser of 1.5ft and No = A fixed positive or negative number assigned by
ct measured perpendicular to the slab span. The parameter indicates the instructor to get R between 600 and 1000.
from the interior face of the column to the slab edge in ct the distance
R = A - Ao, a parameter to get numerical data for the
I
1-direction, but not more than
Cl. problems.
• Further, a minimum of one-half of the reinforcement in
the column strip at support is to
be placed within this width.
• At least one-quarter of the top reinforcement at the support in the column 1- Design a flat plate slab system consisting of 5
panels @ R/200 m (2 decimal places) in
strip should in the other direction.
be made continuous throughout the span. one direction and 4 panels @ R/140 m (2 decimal places) dead load = R/3 (whole
» At least one-third of the bottom reinforcement at the support
in the column strip should ? Service live load = R/2 (whole number) kg/m2, superimposed story height = 2.75m,
be made continuous.
,
,
=
number) kg/m2,//= R/30 (2 decimal places) MPa, fy 420 MPa,
columns is 228x342mm.
• At least one-half of all bottom middle strip reinforcement and all 1 no drop panels, no column heads, no beams, and size of
reinforcement at midspan are to be made continuous. These steels bottom
should
column strip
m (2 decimal places) in one direction, 4
face of support.
develop /y at 2- Design a flat slab system for 3 panels @ R/175 direction, story height = 3.5m,
• All top and bottom reinforcement at support is to be developed at the face of panels @ R/130 m (2 decimal places) in the other
dead load = R/2 (whole
discontinuous edges of the slab. support at service live load = R/3 (whole number) kg/m2, superimposed no drop panels, no
number) kg/m2, fc'~ R/30 (2 decimal places) MPa, fy = 300
MPa,
'
• The two-way shear at the column critical sections caused by Beams are present on the
factored gravity loads column heads, and size of columns equal to 375x300mm.
must not exceed 0.4<p/c. As an alternate, slab shear
reinforcement of amount given in edges and also along the interior column-centerlines. The
overall size of long beams is
ACI 21.13.6 is to be provided for a distance at least equal j
thickness from face of the support.
to four times the slab ; 375x500mm and the overall size of short beams is 300x400mm.
j
3- Design a slab system for the following data:

6 panels @ R/160 m (2 decimal places) in one direction.


4 panels @ R/140 m (2 decimal places) in the other
direction.
400x400 mm with R/1000 m x R/800 m (2 decimal places)
Exterior columns of size bottom of drop panel.
column capital, R/1500 m (whole number) deep from the
places) diameter column
Interior columns of 450mm diameter with R/450 m (2 decimal of drop panel.
capital, R/1200 m (2 decimal places) deep from the bottom
Interior drop panels = R/450xR/400 m (2 decimal places).
drop panel size.
Exterior drop panel to be decided according to the interior
=
Extra thickness of slab at drop panel > h/4.
Story height = 3.5m.
Service live load 0.8R (whole number) kg/m2.
= (whole number) kg/m2.
Superimposed dead load including weight of drop panels = R/3
//= R/30 (2 decimal places) MPa.
No interior or edge beams.
556 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDlQj
Chapter - 13
Space For Reader’s Notes

YIELD LINE AND STRIP METHODS FOR SLABS

13.1 INTRODUCTION
Yield line analysis of slabs is identical to plastic design of frames consisting of skeletal
elements. In frames, plastic hinges are formed at maximum moment sections. This means
that, at these sections, large inelastic rotations may occur at almost constant resisting moments.
This constant resisting moment is called the plastic moment. However, after the formation of
initial hinges without loosing the internal stability of indeterminate structures, more loads may
be applied due to moment redistribution and utilization of strength of less-stressed sections.
Difference of a fictitious plastic hinge from a real hinge is that, in case of the plastic hinge, free
rotations occur at a constant moment level. Below the plastic moment value, the rotations are,
however, locked. When sufficient number of these hinges is formed adjacent to each other,
internal stability of the structure causes infinitely larger deformations causing collapse of the
structure. This condition is referred to as the formation of collapse mechanism or simply
mechanism.
Yield line formation is a similar mechanism that takes place in slabs with the difference that
the plastic hinges concentrated at points are replaced by lines of free rotation at constant
moments. When the slab is loaded beyond a certain limit, a fictitious hinge is formed over a
certain straight length of the slab having maximum moment; this line is called yield line. The
yield line serves as an axis of rotation for the slab segment and large inelastic rotation may
occur at nearly the same moment per unit length measured along a yield line.

13.2 COLLAPSE MECHANISM AND YIELD-LINE PATTERN

When sufficient number of yield-lines is formed such that any further load causes vety large
infinite deformations, exceeding the permissible deflection limit state, the resulting system is
called collapse mechanism or simply mechanism or yield-line pattern. A yield line pattern
indicates how a slab collapses after the formation of such number of yield lines that makes the
slab unstable, just as an arrangement of plastic hinges indicates how a framework collapses.
Plastic moment capacity at a yield line is determined by the usual strength or limit state
method. The plastic design method (yield-line design) differs from the strength design only in
the method of analysis for loads based on plastic behavior of materials compared with elastic
analysis in other methods of analysis and design of slabs.
Yield line forms at the section where the flexural reinforcement yields. The plastic moment
capacity at the yield line is assumed equal to the ultimate strength of a section,
distributed over the length (^m* per unit length of the yield line). The position and orientation
of he yield lines in a yield line pattern depends on the boundary conditions, the nature of
loading and the geometric dimensions.
558 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A YIELD LINE AND STRIP METHODS FOR SLABS 559
^UDlQl CHAPTER 13:
13.3 TYPES OF ANALYSIS METHODS 2. The yield strength of the slab sections is not exceeded anywhere in the
slab
system.
The actual analysis of complex systems is either quite lengthy or is not possible 3. The boundary conditions are satisfied.
and hence
simplifying assumptions are made to reach at reasonably accurate solutions. Similarly The panel is
analysis of determinate and indeterminate slabs is generally very complicated. detailed One example of lower bound solution is the strip method of analysis. moment)
the load (and hence
methods may be used for the approximate solutions depending upon the nature
Two general
j assumed to be divided into strips in both directions. Then point or
of j is distributed in the two directions by equating the deflections at the common
simplifying assumptions, namely, lower bound and upper bound methods. For the ultimate analysis of
elastic or plastic analysis, certain assumptions are to be made in the procedures. approximate 1 by some other approximation. The collapse load obtained from
Either the these strips will be lesser than the actual ultimate load because of two main reasons:
moments are distributed in the start depending upon experience as in direct design
method
j just before collapse
the failure pattern (collapse mechanism) is assumed in the start as in the plastic analysis. or 1. The loads or moments may be redistributed in some other way
Both the ultimate
of these starting assumptions may not give the actual ultimate load and different
trials may be utilizing the strength of the under-stressed sections and thus increasing
required to reach at an answer close to the actual ultimate load. load.
adjacent strips are
According to the general theory of plasticity, the true ultimate collapse load of a ! 2. We neglect the torsional stiffness of the slab by which the
structure situation
between two limits, an upper bound and a lower bound of the collapse load. These limits lies I connected to reduce the bending. Hence, if we consider the complete
be found by well-established methods. A full solution is obtained when both the can ; including the torsional stiffness, the ultimate load will be more.
upper and of longer to
lower bound solutions converge to a single solution, as shown in Fig. 13. 1. Another example is a one-way slab supported on four edges. The ratio carried in
1
* are
J shorter side is greater than 2 and we assume in the start that all the loads
find collapse load depending on this assumption, it will be
} the smaller direction. If we
direction.
I a lower bound as actually at least some load is also carried in the longer

13.3.2 Upper Bound Method


load that is either
For a given slab system, the upper bound method gives an ultimate
correct or higher than the actual value. A reasonable collapse mechanism is assumed in
The ultimate load is calculated
the start depending on experiments and past experience.
from the equilibrium of the slab segments separated by the yield lines.
system at the
The upper bound method gives a collapse mechanism for the slab
Fig. 13.1. Concept of Upper and Lower Bounds. ultimate load such that:
1. The moments at the plastic hinges are not greater than the
ultimate moments of
13.3.1 Lower Bound Method
resistance of the sections.
For a given slab system the lower bound method gives an ultimate load, which is either 2. The collapse mechanism is compatible with the boundary conditions.
correct or low. That is, the ultimate load is never overestimated and there is no to ensure
possibility that the ultimate load is below our calculated collapse load. The distribution The portions of the slab between the lines of plastic hinges are not examined ultimate
available
of loads or moments is decided at the start depending on the past experience. The that the moments there are lesser than or at most equal to the
moments of resistance of those sections. The moment capacity of the slab between the
ultimate load is calculated from the equilibrium equations and the postulated Actually, some
distribution of moments. yield lines is exceeded if an incorrect collapse .mechanism is used.
It is evident
other collapse mechanism is produced at a smaller predicted ultimate load. considered;
is
The lower bound method gives a distribution of moments in the slab system at the that great care is required to ensure that the correct collapse mechanism
ultimate load such that: otherwise, the ultimate load will be overestimated.
method, and consequently
1. The equilibrium conditions are satisfied at ail points in the slab system. In case The yield line method of analysis for slabs is an upper bound may be higher
of the segment equilibrium method of plastic analysis (which is an upper bound the failure load calculated for a slab with known flexural resistances
solution), the equilibrium is satisfied only along the yield lines and not within pattern is not assumed. This is certainly a
than the true values in case a correct failure be on the safe
the slab segments. Hence, this method is not a complete equilibrium solution. concern, as the designer would naturally prefer to be correct or at least to
YIELD LINE AND STRIP METHODS FOR slabs 5^
560 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDjqj cgAPTER13:
side. However, procedures can be incorporated in yield line analysis to help ensure that designs are calculated by varying
amount of reinforcement until a satisfactory
the calculated capacity is correct. arrangement is obtained.
of analysis. Completely
is not valid for any plastic method
13.4 ADVANTAGES OF YIELD-LINE ANALYSIS c. Principle of superposition to be performed for different load
independent analysis is generally required
combinations. compared with
Some of the most significant advantages of the yield-line analysis of slabs are explained below;
slabs having regular geometries, the calculations are lengthy as
a. The common strength design or LRFD method involves elastic analysis of slab systems d. For
for the applied loads. On the other hand, reinforcement for slabs is calculated by
traditional elastic methods.
strength methods that account for the actual inelastic behavior of members at the TO SHOW BOUNDARY
factored load stage. This means that the load actions and the corresponding material 13.6 CONVENTIONSAND YIELD LINES
properties are not fiilly compatible. Yield-line analysis removes this inconsistency^
CONDITIONS
and the
b. The slabs designed by yield-line theory become economical as it accounts for the be selected to conveniently show the slab boundary conditions
may for the
reserve strength characteristics of most concrete structures. This reserve capacity may
Any convention
In this text, the sign convention of Fig. 13.2 will be used throughout a free
type of yield lines. supported edge will be shown by line with
single hatching,
be at the most heavily stresses section or it may be in terms of redistribution of moment edge conditions. A simply continuous edge will be
and utilization of reserve capacity of less stressed sections away from the yield lines. by a simple line and a fixed, built-in, or
edge will be represented
c. Yield-line theory permits, within limits, an arbitrary readjustment of moments found by identified by line with double hatching.
elastic analysis to arrive at design moments that gives more practical reinforcing
arrangements.
d. Yield-line theory gives more general method of analysis and design. It can be used for
round and triangular slabs; slabs with large openings, slabs supported on two or three
edges only, and slabs carrying concentrated loads. 1 I «
or Continuous Edge
For elasticity-based methods, slab panels must be rectangular, supported on at least two a) Simply Supported Edge b) Free edge c) Fixed, Built-in
opposite sides for one-way slabs, supported on four sides for two-way edge supported Edge Conditions.
Fig. 13.2. Conventions to Show
slabs, or supported by a fairly regular array of columns for flat plates. Further, these
Wavy or
methods usually deal with uniformly distributed loads. to show various types of yield lines are shown in Fig. 13.3.
e. In reinforced concrete frames consisting of beams and columns, the rotation capacity or
Similarly, the notations
lines. A positive yield line is formed by sagging curvature
ductility of the member is generally less because of higher steel ratios. Checking the bold lines will show positive yielddeveloped lines will show
on the lower surface. Bold dashed slab and the
of slab and the tension cracks are of
rotation capacity at each section becomes a tedious and time-consuming job. As line is formed by hogging curvature
negative yield lines. A negative yield
compared with this, the slabs are lightly reinforced and their rotation capacities are upper surface.
tension cracks are developed on the
much greater than usually required.
1
13.5 DEMERITS OF YIELD-LINE ANALYSIS or I
I
a. The yield line method is an upper bound approach in determining the ultimate flexural
strength of slabs; the error in the analysis is on the unsafe side. Wrongly assumed b) Negative Yield Lines
collapse mechanism gives more slab strength than the actual one. This is a special a) Positive Yield Lines
Lines.
cause for concern when tiie method is used for design. Fig. 13.3. Convention to Show Yield
b. When applying the yield Une method for design, it is necessary to assume the
distribution of reinforcement over the whole slab, which may be uniform in 13.7 RULES TO LOCATE YIELD LINES
perpendicular directions or a percentage of steel in any one direction is taken in the
they represent the intersection of
two planes. The
other direction. The yield line approach is basically a tool for review of capacity of a 1. Yield lines are straight lines because as a whole with negligible elastic
bending
given slab. It can be used for design only in an iterative sense. The capacities of trial adjacent plate segments are assumed to rotateplane.
within them. Hence, each plate segment is a
562 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. YIELD LINE AND STRIP METHODS FOR SLABS 563
SIDDjqj cHAPTER13:
segments.
Deflection of any point on the yield line be should be equal for both the for the other
Hence if it is zero for one segment it should simultaneously be zero line with
segment. Any point other than b acting as the point of intersection of the yield
Any
one of the edges will have zero deflection for one segment but not for the other.of a rigid
point away from the axis of rotation will have some deflection for the rotation
formed by two adjacent plate segments (common
plane. So, the three axes of rotation
axis, right support axis, and left support axis) must intersect at a single point.
Original
original or Slab
rosiuon of
Position _
Real Hinge

Rotated Plate Segment


Plastic Hinge

Fig. 13.4. Positive Yield Line in One-Way Simply Supported Slab.

2. Yield lines represent axes of rotation. The slab segments are considered to rotate
rigid bodies in space about these axes of rotation. as
3. Axes of rotation will be formed at the supported edges of the slab in the following
different ways: two
a) A negative yield line may form over a fixed edge providing constant b) Incorrect Position of Yield Line
resistance a) Correct Position of Yield Line
to rotation equal to the ultimate flexural capacity of the slab.
b) An existing axis of rotation is considered at the edge that is simply
providing zero restraining moment.
supported Fig. 13.5. Yield Line between Two Trapezoidal Slab Segments.
4. An axis of rotation will pass over any column support. Its orientation depends on other of the two
considerations. For example, if columns are present in a row, yield line may pass 7. At a comer, the positive yield line extends towards the point of intersection which
edges at an angle. Only exception to this is the formation of the comer levers,
through the column centerline. An independent single column may act as a full edge
along any direction or it can act as a point load applied from below. will be discussed later.
yield lines as
5. Yield lines form under concentrated loads, radiating outward from the point of 8. The positive yield lines are always pushed away from the negative
application of the loads. cpmpared with the natural axis of rotation.
6. A yield line between two slab segments must pass through the point of intersection 13.8 YIELD LINE PATTERNS
of
the axes of rotation on the other sides of the adjacent slab segments (Fig. 13.5). This
yield lines are
condition is not required to be satisfied if the three axes of rotation are parallel to each Some typical yield-line patterns developed according to the rules to locate the
other. This means that if axes of rotation ae and dg meet at point b in Fig. 13.5a, then given in Fig. 13.6.
the positive yield line fc or its extension must also pass through point b.
If the axis of rotation for the positive yield line does not pass through the axes of
rotation corresponding to the supports, as in Fig. 13.5b, it can be proved that the yield¬
line pattern will not be’developed without bending of the slab segments. The deflection
at edge bd is zero throughout as gd is a support with zero deflection and bd an
is
extension of line dg. Similarly, the deflection at edge ab is also zero. Rotation of
triangular plane chd about axis bd is possible with zero deflection at the edge hd.
However, when the plane abhc tries to rotate about axis ab, either point h has to deflect a) Simply Supported Square Slab b) Simply Supported Rectangular Slab
or the plane undergoes warping. Since both are not possible, point A must coincide
withpointb.
Fig. 13.6. Typical Yield Line Patterns (Continued on next page).
564 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 565

k) Rectangular Slab Continuous 1) Circular Slab Resting


On Two Sides and Resting On On Four Columns
a Comer Column

m)Rectangular Slab Simply n) Rectangular Slab Simply


Supported On Three Sides Supported On Three Sides
Having Large Aspect Ratio Having Small Aspect Ratio

o) Square Simply Supported


Slab with Comer Levers

g) Triangular Slab Simply Axis of rotation, becomes


Supported On Two negative yield line if the
coiner is held down.

Fig. 13.6. Typical Yield Line Patterns.

13.9 FUNDAMENTAL ASSUMPTIONS IN YIELD LINE THEORY

1. Steel reinforcement is fully yielded along the yield lines at failure. The sections are
under-reinforced with Very small steel ratios allowing large hinge rotations.
i) Triangular Slab
j) Square Slab Continuous On One Side and 2. The resistance per unit width of slab is the nominal flexural strength of the slab; that is,
Continuous On Two mp = m„, where mn is calculated by the usual expression for evaluation of the flexural
Simply Supported On Other Three Sides
Fig. 13.6, Typical Yield Line Patterns (Continued resistance. For design purposes, mp is to be taken equal to ^mn, with - 0.90 for
on next page). flexure.
YIELD LINE AND STRIP METHODS
FORSLABS 567
566 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDjqj CHAPTER 13:
for strength is to be
bars as placed in Fig. 13.10 whose contribution on a single
steel
3. Bending and twisting moments are uniformly distributed along the yield lines and Consider the
along and across the skewed
yield line shown by wavy line. Concentrate
vector mi.
these evaluatedof the steel placed at a spacing‘s’ corresponding to the moment
give the maximum bending moments perpendicular to die yield lines.
bar out
4. Elastic deformations are negligible as compared with plastic deformations.
5. Moment rotation (moment-curvature) curve of critical regions is idealized as elastic¬
plastic bilinear curve with considerable inelastic rotation to allow full redistribution
(Fig. 13.7).
6. Shear failure, bond failure and compression failure are prevented.
7. The position of yield lines in a yield pattern depends upon the boundary conditions,
nature of the load and the geometric dimensions.

Fig. 13.9. Steel Placed In iwo uiuciv...

the moment axis direction,


Let 0 = clockwise angle of the yield line from
the yield line, bending
s' = bar spacing alongmoment (perpendicular to the yield line, called
m„ = normal yield

and

Fig. 13.7. Moment Curvature Relationship.

13.10 CONTRIBUTION OF A STEEL (mi) ALONG


AND ACROSS A SKEWED YIELD LINE

A moment vector (mi) shows that the reinforcement is provided in a perpendicular direction to
resist a bending moment mi given by the right hand rule, as shown, in Fig. 13.8. If thumb
points in the direction of the moment vector, the fingers give the direction of bending and the
corresponding steel.

Along and Across a Skewed Yield Line.


Fig. 13;10. Contribution of Steel
may be resolved in to two components, one parallel and one
The yield force in the bar,
AJy, respectively. Jf the slab
denoted by yf^in^ and ^^cos^,
perpendicular to the yield line,
by mj, the total moment over
a width equal to s is given
yield moment per unit width is denoted
by: (1)
Fig. 13.8. Steel and its Corresponding Resisting Moment Vector. mi s = A/y x z
and compressive forces.
Magnitude m of the moment vector shows the bending moment per unit width of slab and its where z = lever arm between tensile
line for a width equal to s' is:
direction shows the axis of rotation. If steel is provided in two directions, the magnitudes of The total moment along the yield
these will be shown independent of each other even if they provide some strength in the
direction of the other steel (Fig. 13.9). mn x s' = Afy z x cos (2)
= mi s cos
569

568 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHSja CHAPTER l3:


and mt x y = Afy x z x sin
= missing (3)

It can be seen from Fig. 13.10 that


or j = s' cos
cos
^ = —s' (4)
From Eqs. (2) and (4): ma x s' = mi s' cos-
or ma = mi cos2 (5)

mt = mi sin fa cos
From Eqs. (3) and (4): (6)
The directions of normal moment vector and torsional moment vector are shown in Fig, 13.] ] Fig. 13.12.
ContributionofOrthogonalGridofReinforcementataSkewedYieldLine.
both for acute and obtuse angle of yield line from the resultant moment vector. The torsional
moment wt is considered positive when its moment vector points away from the section. bars producing my,
For the x-direction
= - and sin^i = cos a
90 a
cos - sina
cos 2 (90 -a)
ma = znysin2a
= sin(90 a) cos(90 - a)
and mt = my -
= my sina cosa
together,
For both the bars present a+mysm a
a) Yield Line Having Acute b) Yield Line Having Obtuse znn = mxcos cosa sina cosa
and mi -
= mxsina
sinacosa
Fig. 13.11. Sense and Direction of Normal and Torsional Moments. = (my-m^ reinforcement provided in each
with the same
when m* = my = m,
For the special case
13.11 CONTRIBUTION OF STEEL PLACED IN GRID RESOLVED direction,
ALONG AND ACROSS A SKEWED YIELD LINE
= m(cos2 a + sin2 a) = m
= 0
Generally slab reinforcement is placed orthogonally, i.e., in two perpendicular directions and and mt of barsparallel to the yield lines.
yield lines form at an angle with the directions established by the reinforcement. Consider an to 0° and 90°, there is some contributionthe contribution of the bars nearly
For values of a close to neglect all
orthogonal grid of reinforcement^ with angle a between the yield line and the x-axis. Bars in However, for such
cases, it would be conservative resistance per unit length (ma = m) in
The slab having the same
the x-direction are at a spacing v and have moment resistance my per unit length about the y- parallel to the yield line.
isotropically reinforced slab.
axis, while bars in the Y-direction are at spacing u and have moment resistance m* per unit directions is said to be isotropic slab, referred
length about the x-axis. Bars in x-direction produce moment about y-axis and vice versa. can be simplified to that of a related areas is the
analysis of an orthotropic slab negative to positive reinforcement
The
provided that the ratio of to permit
For the y-direction bars producing ?mx, to as the affine slab, dimensions and slab loads must be modified
horizontal
fa = 180 - a (clockwise angle between yield line and moment vector) same in both directions. The
cos = -cosa and sin = + sina this transformation.
FOR ANALYSIS
and, m^ = cos 2 a 13.12 MAJOR STEPS
mt = - m* sina cosa or yield-line pattern is assumed.
a) A possible mechanism
YIELD LINE AND STRIP METHODS FOR SL ABS 571
570 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A CHAPTER 13:
SIDDIQI
b) Segment equilibrium or virtual work method is used to find out the ultimate or collapse
load for this mechanism.
wx2 _ (30 + 30) 0
c) The geometric dimensions within a mechanism are adjusted to get a minimum collapse
load for this mechanism.
d) The above procedure is repeated for all the possible yield line patterns. The minimum
.’. ^--(60) = 0 (D

collapse load is considered as the final answer. Equilibrium of Portion BC

13.13 SEGMENT EQUILIBRIUM METHOD


XMc = 0
-70 0 (II)
i) A suitable yield line pattern is assumed according to previously discussed rules and 2
guidelines producing a collapse mechanism.
ii) For the selected collapse mechanism, rigid body movements of slab segments are
considered along the yield lines maintaining the deflection compatibility.
iii) Each piece or segment is considered as free body and its equilibrium is studied.
iv) Unknowns for the solution are the failure load and location and orientation of the yield
lines.
v) The number of equilibrium equations required is equal to the number of unknowns.
One unknown is always the resisting moment of the slab or the collapse load. Other w kN/m
unknowns are needed to define the locations of yield lines.
vi) The resulting equations are solved simultaneously to evaluate all the unknowns.
vii) Because the yield moments are principal moments (yielding will start at the maximum
moment values), twisting moments are zero along the yield lines, and in most cases the
shearing forces are also zero (maximum moment sections are usually associated with
zero shear force). Hence, only the unit moment Tn’ is generally considered in writing
equilibrium equations. 40
viii) All the other possible yield line patterns are investigated turn by turn. The minimum
collapse load is the final answer.

Example 13.1: A one way, uniformly loaded and continuous slab panel of 4m span
having a positive flexural capacity of 30 kN-m/m and negative flexural capacities of 30 kN-
m/m and 40 kN-m/m at the left and the right supports, respectively. Calculate the ultimate load Fig. 13.13. Slab System for Example 13.1.
capacity of the slab.
Solution: Equations I & n are solved simultaneously for w and x:
The slab system for the given data is shown in Fig. 13.13, where negative yield lines are
formed over the beams. The positive yield line is parallel to the negative yield lines but its
position is unknown in terms of distance x. The free body diagrams of the two slab segments
From I: w=

120

Using this value of w in Eq. H, we get:


for unit width thus formed are also shown in the same figure. At positive yield line, shear force
is almost zero and may be ignored for the free body diagram. “(4 -^-70 = 0
Equilibrium of Portion AB X
60(16 + x2- 8x)-70x2 =0
= 0 - 10x2-480x + 960 =0
x2 + 48x - 96 =0
YIELD LINE AND STRIP METHODS For sl ABS 573
572 CONCRETE STRUCTURES PART 1 - AUTHOR: Z AHID A. SIDDIqj CHAPTER 13:

x = — 48 + 51'85
— = 1.923 m (As x cannot be negative)

w = —xz -
120
32.45 kN/m

13.14 VIRTUAL WORK METHOD

i) A suitable yield line pattern is assumed according to the given loads, dimensions and
edge conditions.
ii) Since the moments and loads are in equilibrium when the yield line pattern has fonncd,
an infinitesimal increase in load will cause the structure to deflect further. The external
work done by the loads to cause a small arbitrary virtual deflection must equal the
internal work done as the slab rotates at the yield lines to accommodate this deflection.
Hence, the slab is given a unit virtual displacement at the yield line joining the two by Virtual Work Method.
plate segments to apply the principle of virtual work. Fig. 13.14. Solution of One-Way Continuous Slab
iii) Due to the above displacement, rotation (0) is calculated at each hinge. The rotation at f i a ( 4 "1 .1
each yield line is calculated by considering the geometry of the section taken
perpendicular to the yield line.
Wi = (30)U (30\^)J + (40) 4^
+

120-30X + 120 + 40x Ox 1 + 240


iv) Rotation multiplied with the constant moment at the hinge gives the internal work.
= x(4-x) x(4-x)
W{ = Z
v) External work is calculated as the product of load and displacement.
We = Wi
10x + 240
= 2w
We YFd x(4-x)
vi) External work done is equated to the internal work done to solve for the unknown. 5x + 120
Differentiation may be used to minimize the collapse load. w
4x-x2
Example 13.2: Solve slab system of Example 13.1 using the virtual work method. To determine the minimum value of the load w,
this expression is differentiated with
respect to x and is equated to zero.

-
Solution:
Give a unit virtual displacement to yield line-B for the slab system shown in Fig. 13.14.
6b fti + ftz I -i-W
dx v*
_

1
x
1
1
4-x
=
4-x + x
x(4 - x)
4
x(4 - x) dw
=
(4x - x2 )(5)- (5x +120)(4 - 2x)
(4X-X2)2
=0
ft = fti + ftz
1 1 4-x + x 4 20x - 5X2 - 20x + lOx2 -480 + 240x = 0
J-
x 4-x
— x(4-x) x(4-x) 5x2 + 240x-480 = 0
w. — x2 + 48x-96 = 0 =>
_ 32.45x= 1.923


=3 y2wxx 1 + 14w(4 x) x 1
as
w.
— (x + 4-x)
'
.
= 2w ..• W> -= 5x +120 ~ kN/m2
2 4x-x2
574 CONCRETE STRUCTURES PART - 1 YIELD LINE AND STRIP METHODS FOR SL AUS 575
AUTHOR: ZAHID A.
SIDDJqj CHAPTER 13:
Example 13.3: A square slab is simply
supported along all the sides and is
isotropically reinforced. Determine the resisting moment m = t0 From above relation: m = per unit width of slab
0 per linear width required
sustain a uniformly distributed factored load of w kN/m2. Solve by using to (10.8% lesser than the case when comer levers
both the methods bv
ignoring the comer lever effects. 3 are considered)
Solution: If comer levers (as explained later) are considered, the mechanism becomes critical
Segment Equilibrium Method: The i giving lesser collapse load for a particular reinforcement. Alternatively speaking,
slab along with the possible yield line pattern without the greater resisting moment is to be provided to cany a given load when comer levers are
comer levers is shown in Fig. 13.15. The collapse load w is treated as a incorporated in the analysis. The moment capacity required in this case becomes:
moment capacity m is the only one unknown and there is one segment type known, the j
(giving one
equation). The free body of the slab segment is also shown in the same figure. wL2
. m = (considered as reference value to compare other answers)

] ACI 1963 moment coefficient for such slab is significantly lesser than the value
s' obtained here, perhaps due to use of lateral stiffness of beams if present on the edges.
’I An arch action may be developed in the slabs due to edge beams, which utilizes the
axial stiffness of the slab and file lateral strength of the beams. In the absence of the
edge beams, this value may be on the unsafe side. ACI moment value is as under:
wr2
m = 27.78
(23.0% lesser)

Fig. 13.15. Isotropic Square Simply Supported Slab of Example 13.3. If deflections are equated in both direction strips, as is done by the British Codes, the
value of required moment is as follows:
For isotropic slab, effect of reinforcement is same in all directions and hence
the
resisting moment is also ‘m’ along the diagonal yield lines. For sum of moments about wL2
m = (33.7% more, safe but uneconomical)
the top support line of the free body diagram to be zero:
The elastic solution obtained by solving the governing differential equation considering
homogeneous isotropic material with v= 0.3 is as follows:

m ~ wL2 (2.4% more)


For two edges.
20~9
The torsional moment required to prevent comer lifting is equal to wL2/I5.38.
Load on one Total moment on To find component
segment one diagonal edge of moment along Virtual Work Method: The assumed yield line pattern is shown in Fig. 13.16. Point E is
the top edge.
displaced downwards for a virtual displacement of unity. To calculate the rotations,
or wL1 sections are to be considered perpendicular to the yield lines. All yield lines are
mL
24 identical in this example and the yield line AE.is considered here as a typical yield line.
The total internal moment about the support-line, mL, can be directly written as the Its rotation will be constant throughout AE length and will be equal to the rotation at
product of the total length and the resisting moment along the top edge. In other words, point E of a perpendicular line such as BD. The section taken along the line BD is also
the moment about any line may always be considered equal to the projected length of shown in the figure.
inclined edge along that line multiplied with the moment capacity per unit length along
the projected direction. &B + @D
1
+
1
= 2V2
l/42 L
CONCRETE STRUCTURES PART - 1 CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 577
576 AUTHOR: ZAHID A. SIDDIQJ

• Rotation at yield line BE can be found by considering the rotation of two


perpendicular lines GE and EH at E.

Fig. 13.16. Analysis of Square Simply Supported slab by Virtual Work Method.

Total JFi

=
mx l/ -Jz x —
4(2 m) =
—— — 2m
8 m (There are four yield lines similar to AE)
Fig. 13.17. Solution of Slab ofExample 13.4 by Virtual WorkMethod.

BE = Jx2 + 4
Load on plate segment BED Using similarity of triangles BEJ and GEK:

Deflection of centroid of plate BED = — EJ KE 2 X 2


r2 1 wL2
We for plate BED = w— x -
IF Triangles BEJ and EHJ are also simitar, so,

Total We = 4x
wZ2
12
wL2
3
BE
x
= —
a
EJ
=> a - Jx2x+ 4 x .
,

We = Wi

Fig. 13.18. Section GH for Slab of Fig. 13.17.


Example 1 3.4: A two-way slab 7m x 4m is simply supported on all four sides and carries
a uniformly distributed ultimate load of w kN/m2. Determine the ultimate moment of resistance Considering a section along GH as shown in Fig. 13.18:
required for the slab if it is to be isotropically reinforced. Solve by both virtual work and 0g = 0g + 0n
segment equilibrium method by ignoring the corner levers. 1 1
Solution:
+
b a
Virtual Work Method: The slab and the expected yield line pattern are shown in Fig. 13.17.
Rotation of yield line BE can take place along a line perpendicular to it (GH in this
=
2
=== 4-
xVx2 +4 Z^X2 +4
x
= —4x2 — f
1
+4
-I
2
F

x'lI
27
case).
• The yield line EF is given a unit virtual displacement.
- 1 f4 + x2^ _ 7x2 +4
• The unknowns are w or m and x.
0^
7x2 + 41 J 2x
578 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A SIDDlQj CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SL ABS 579
The rotation of yield line EF is: m =1.046 w
firtiment Equilibrium Method: There are two unknowns and two distinct segments in. this
Sef — example These slab segments are shown in Fig. 13.19. Considering equilibrium of
segment AEB, we get,
= W{ along BE x 4 + along EF

mx 7x2 +4 x +4
x 4 + [m(7
_ 2x) 11
x
2x

—m(x2+4)+m(7-2x)
We due to AAEB x 2 + We due to ABEJ x 4 + We due to JEFN x 2
A2) +
~ J J
8/3 wx-4wx+ 14 w
- 4/3 w x + 14 w
W,
w — 4
3
x+14
1/
Fig. 13.19. Free Body Diagrams for Slab of Example 13.4.

w[-4x+42] Considering equilibrium of segment BEFC,

2 lx + 24 7m = 2x
(1 \
— x2 xxx
\2 J
w

<3J —
+(2)(7 2x)w
-4x2+42x
The parameter (w/m) should be the smallest for the critical condition. This can be
7m
7m
=
=

4/3 x w + 14 w 4 wx
- 8/3 w x + 14 w (D)
done either by trial method or by using differentiation to minimize {wtm}. Trial From Eq. 1:
method is presented here in Table 13.1.
zx - 8/3 w x + 14 w
Table 13.1. Trial Method to Calculate Collapse Load. 6
w 7wx2 = -16wx + 84w
X

2.0
2.2
m
0.97059
0.96112
7x2 + 16x— 84=
r =
0
-16 ±51.069 _— Z.jUSrn
2.4 0.95679 14
2.5 0.95625 From Eq. (I) fa = 1.046 w|
2.55 0.95634
2.52 0.95626 Example 13.5: Solve the orthotropically reinforced square slab shown in Fig. 13.20 by
using the virtual work method.
Hence, 2.51 m and 0.95625 Solution: Consider the negative and positive yield lines of Fig. 13.21 to produce the yield line
pattern. Furflier, allow a unit virtual displacement at the point E.
VTFI n TINE AND STRIP METHODS EOR SLABS 581
580 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIq
chapter131
Bottom Steel Tod Steel If the positive and negative reinforcements are
moment capacity may be obtained:
equal (a — 1), the following required

am -
wL2
am 48

Rrample 13.6: Calculate collapse load for the


orthotropically reinforced square slab
work method in terms of m, a and L.
shown in Fig. 13.22 by using the virtual
Fig. 13.20. Square Slab for Example 13.5.

Fig. 13.22. Square Slab for Example 13.6.


body
I and H, and two unknowns. The free
Solution: There are two distinct segments,13.22.
Rotation at yield line EB, Oe = as for the simply supported square slab diagrams of these segments are shown in Fig.
L
Segment I: Taking moments about the bottom
line, we get,
Rotation at yield line AB
i
Ln
'
2
l UL-x) (L-x)
14^ X1,1 -
_ wZ
77Z
,
ij t Cl Tfl 1j
.
2 3
Wc = W.D. by the load for segment ADE x 4

— 6
x(Z-x)2 = mL(\ + a)

Wi Wi for yield line EB x 4 + Wi for yield line AB x 4 —6


x(Z-x)2 = m(l

Segment II: Taking moments about the right line, we


get,
+ a) (I)

hZ— x—
2 3
= mL

W9 = —
wx2
6
- = m
Using the value of m from Eq. II in Eq. I, we
get,
(II)

=
24(1 + g) —Z2+x2-2Zx = —
w
o
,r .2
x(Z-x) , =
wx2
o
x2(l
/■,

+g)
,
(1 + «)
582 CONCRETE STRUCTURES PART 1
- AUTHOR: ZAMD A. SIDDlQj
CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 5g3
x - ~ = : 13.15 SPECIAL CONDITIONS AT EDGES AND CORNERS

For x to be positive.
2a
L\ 13.15.1 Edges
Positive yield lines in some of the patterns shown earlier, where these have to intersect
free edges not at right angles as shown in Fig. 13.25b, are assumed to go straight and
= Zi intersect the edges at angles different from the right angle. This is just a simplification
at an edge. At a free edge, both the bending moment (having axis parallel to the edge)
and the twisting moment should be zero, defining the minor principal direction of
= Zi 1-0-^ L moments. Hence, the major principal direction is perpendicular to the edge, the
bending moment having moment vector in this direction should be the maximum and
yield line must be formed in this direction. Consequently, the yield lines should enter
Form Eq. IT: an edge perpendicular to it, as shown in Fig. 13.25a. This is a contradiction with the
„2
original statement that the yield lines must be straight, and is to be treated as a special
Example 13.7: case. Tests confirm that yield lines change their angle very close to the edge and enter
square slab shown in
F^ 321(1 distinct portions available for analysis
for the
the edge in a perpendicular position. The distance ‘x’ in the figure is usually much
small compared to the other slab dimensions.

Fig. 13.23. Square Slab for


Example 13.7. Fig. 13.25. Positive Yield Line at a Free Edge.
Solution: The yield line pattern is
shown in Fig. 13.24 along with the required slab portions. The calculation procedure-may be simplified by still considering straight yield lines
intersecting the edges at angles different from 90°, which increases the collapse load
due to the use of wrong assumption. This collapse load may then artificially be may
made smaller equal to the actual collapse load by introduction of an applied twisting
moment at the edge. This twisting moment not only overload the slab but also creates
the condition of maximum bending moment along the inclined direction of the yield
line. Further, in place of applying this twisting moment, a shear force V acting
downward at the acute comer (circled cross in the figure) and the force P acting upward
at the obtuse comer (circled dot in the figure) together are applied as the static
equivalent of twisting moment. The magnitude of this pair of fictitious shearing forces
Pis given by the following expression:
Fig. 13.24. Yield Line pattern V = tn cot a
for Square Slab of Example 13.7.
Where ‘/n’ is the resisting moment per unit length along the yield line and ‘o’ is the
acute angle between simplified yield line and the edge of the slab. This pair of equal
yield LINE AND strip methods forslabs 585
584 CONCRETE STRUCTURES PART - 1 AUTHOR: Z AHID A. SIDDIQI CHAPTER 13-.
and opposite forces does not enter the virtual work calculations as the work done by
both of them cancels each other. However, when free body diagram of each segment is
separately considered in the equilibrium method, the forces 7 enter in to the
calculations.

13.15.2 Corners
To simplify the calculations at comers, it is commonly assumed that yield line enters
the comer as a simple straight line passing between the two intersecting sides. In
reality, the yield lines fork or divide in to two branches to form a Y before reaching the
comers. The resulting slab segment formed near the comer is called a corner lever or
corner pivot, as shown in Fig. 13.26. If the comer is not held down, the triangular
element adbc will pivot about the axis ab and lift off the supports. If the comer is held
down, the line ab becomes a negative yield line and the segment abd rotates about it. independent of the
of junction of the Y from the pivot, h, is
Hence, the distance and negative
segment. It increases as both the positive
distance x and the width of the respect to the load intensity.
moment capacities increase with critical than those
levers are generally more
Although yield line patterns with comer only approximate adjustment is made for
neglected and
without them, they are often segments is slightly
In square panels, the maximum effect of the comer these can be
their effect.
than 9%. Further, if sufficient comer steel is separately provided,
less
eliminated in the strength calculations.
the collapse
the isotropically reinforced slab shown in Fig. 13.28, find
a) Comer Allowed to Lift-Up Example 13.8: For
using virtual work method.
Fig. 13.26. Formation of Corner Levers. load in terms of the moment

If cracking at the comers of such a slab is to be controlled, top steel more or less
perpendicular to the line ab must be provided. The position of the positive yield lines
near the comer indicates the requirement of supplementary bottom-slab reinforcement
at the comers, placed approximately parallel to the line ab. If the comer is sufficiently b
reinforced for negative moment giving capacity w', the junction of Y is shifted away 7m
from the comer. When the negative moment and the distance of junction of Y increases
Example 13.8.
from the comer, the triangular comer lever is not formed and the ordinary yield line Fig. 13.28. Triangular Slab for
pattern with a single yield line extending up to the comer is developed. In continuous two
slabs, the negative steel at the comers already exists and the formation of the comer yield-line pattern is shown in Fig. 13.29, which involves is
Solution:, The selected A unit vertical deflection
levers depends upon amount of this negative reinforcement. in the form of collapse load and the angle a
unknowns
Consider the free body of the comer lever formed shown in Fig. 13.27. applied at the point ‘tf.
of AB x 1/3
If no shear force acts at the apex of the segment, we get, We w x area of M X 1/3 + W X area
w/3 x total area
SAfab = - 1.414 xm'- 1.414 x m + 0.5 (1.414 xXwXAXA/3) = 0 k
w/3 (baseXheight) 'A
wA2 m m [w/6] (7X5-5 Sin 70°)
6 = + h = ^6(m + m')lw 77/i2 w sin
584 CONCRETE STRUCTURES PART -1 YIELD LINE AND STRIP METHODS FOR SLABS
AUTHOR: ZAHID A. SIDDIQI -HAPTER13: 585
and opposite forces does not enter the
virtual work
both of them cancels each other. However, when freecalculations as the work done by
body diagram of each segment
separately considered in the equilibrium method,
the forces V enter is
calculations. in to the

13.15.2 Corners
To simplify the calculations at comers, it is commonly
the comer as a simple straight line passing assumed that yield line enters
between the two intersecting sides; In
reality, the yield lines fork or divide in to two branches
to form a Y before reaching the
comers. The resulting slab segment formed near the comer
corner pivot, as shown in Fig. 13.26. If the comer is not is called a comer lever
element adbc will pivot about the axis ab arid lift off the held down, the triangular
supports. If the comer is held
down, the line ab becomes a negative yield line and the segment Fig. 13.27. Free Body Diagram of Comer Lever.
abd rotates about it.
h, is independent of the
Hence, the distance of junction of the Y from the pivot, and negative
as both the positive
distance x and the width of the segment. It increases
moment capacities increase with respect to the load intensity.
Although yield line patterns with comer levers are generally
more critical than those
adjustment is made for
without them, they are often neglected and only approximate
the comer segments is slightly
their effect. In square panels, the maximum effect of
separately provided, these can be
less than 9%. Further, if sufficient comer steel is
eliminated in the strength calculations.
a) Comer Allowed to Lift-Up Fig. 13.28, find the collapse
b) Comer Held Down Example 13.8: For the isotropically reinforced slab shown in
Fig. 13.26. Formation of Comer Levers. load in terms of the moment using virtual work method.
If cracking at the comers of such a slab is to be
controlled, top steel more or less
perpendicular to the line ab must be provided. The
position of the positive yield lines
near the comer indicates the requirement of
supplementary bottom-slab reinforcement
at the comers, placed approximately parallel to the
line ab. If the comer is sufficiently
reinforced for negative moment giving capacity m', the junction of
from the comer. When the negative moment and Y is shifted away a
7m
the distance of junction of Y increases
from the comer, the triangular comer lever is not
formed and the ordinary yield line
pattern with a single yield line extending up to the Fig. 13.28. Triangular Slab for Example 13.8.
comer is developed. In continuous
slabs, the negative steel at the comers already exists
and the formation of the comer in Fig. 13.29, which involves two
levers depends upon amount of this negative
reinforcement. Solution: The selected yield-line pattern is shown
angle iz A unit vertical deflection is
Consider the free body of the corner lever formed shown in Fig. unknowns in the form of collapse load and the
13.27. applied at the point 'd.
If no shear force acts at the apex of the segment,
we get, w x area of AX x 1/3 + w x area of AZ? x 1/3
SAfab = - 1.414 xm’— 1.414 xm + 0.5 (1.414 x)(w)(A)(A/3) = 0 w/3 x total area
w/3 (base)(height) 14
= m+ m’ h= + [w/6] (7X5.5 Sin 70°)
6
/12 w sin 70°
586 CONCRETE STRUCTURES PART -1 YIELD LINE AND STRIP METHODS FOR slabs 587
CHAPTER 13:
AUTHOR: ZAHID A. SIDDIqi

e %2 [
_ cos
(70 - a) sin a + sin(70 - a) cos a]
0


sin a cos (70 a) =
tan a = tan (70- a)
sin (70 -a) cos a

2a 70° z = 35°
a = 70- a
m
From Eq. I:
w
77/l2 sin2 35'
b
a) Possible Yield-Line Pattern f m 24.67
b) Geometric Properties of the
Yield-Line Pattern w
Fig. 13.29. Yield Line Pattern for Slab of
Example 13.8. Q.4737~^
Consider a section perpendicular to the Y.L, shown as shown in Fig. 13.30, find the
e/in Fig. 13.29(b). Example 13.9: For the orthotropically reinforced slab
ed - ad tan (70 - a) collapse load using segment equilibrium method.
df = nritan(a)
= Med
ft = Vdf
ft =- . rotation all along the yield line
— x
+ ft
1 1
t
oritan(70-a) ad tana
= -i-(cot(70-a)+cota)
ad
Fig. 13.30. Triangular Slab for Example 13.9.
Wi = moment capacity at Y.L.xlength of Yl.xrotation at
=
=
m x adx

ad
m[cot(70-a)+cota]
[cot(70-a) + cota]
Y.L.
Solution: The selected yield-line pattern is shown
diagrams of the two segments formed.
in Fig. 13.31 along with the free
body

We =
=>
-
w sin 70° = m[cot (70 a) + cot a]

Trigonometric Relation: cot + cot R = ~n


sin^sinB
.. 77, - 70°=
/12 w sin m sin
(70- a +a)
t
sin(70-a)sina
or

_

w = 77/i2 sin (70 - a)(sin a)
=
Fig. 13.31. Yield Line Pattern for Slab of
Example

distance of point of intersection of yield


13.9.

line with free edge measured


. . m , we ,
To maximize
w
have the relation (wj
da ——- =0 *
Let x

and a =
from point B.
acute angle of yield line with respect to the free
edge.
588 CONCRETE STRUCTURES PART 1
-
cos a =
—== x
Vx2+25 and sin a = —==
Vx2
5
AUTHOR: ZAHID A.
SIDDlQj
CHAPTER 13:


YIELD LINE AND STRIP METHODS FOR SLABS

———
539

cot a - —
5
Along the yield line CD, the
9 = tan1)

kej =
following may be calculated:
|
+ 25

39.8°
5w (6-x)
o
„ .2
=
240 + 86.4x
,
(200 + 72 x) +
12x3~ +200x (6 x),

14.4x’+240x
%’+25)(6-x)
-

©
(6-x)1
m = ma = mxcosa +
wysm a Segment DBC

= 60-^-+40~^-
x2 +25 x2+25 =
60x^1000 The bending moment at the yield line is already resolved in to horizontal and vertical
components and moment is to be taken about the vertical line. Hence, there is no need
x’+25
V = th cot a :M to further resolve the moments.

Segment ACD
=m x
—— —
x
5
12x’+200x
x2+25 - = 0 w —xxx 5 x
12 J —
3
+ Kxr = 200

It is to be noted that the


be resolved along the horizontal and vertical moments at the yield line may
13.32. ‘
'
line about which moment
sum is to be taken, as showndirectly 240 (14.4x2+240) (H)
in Fig. x2 (x2+25)
By solving Eqs. I and H simultaneously using trial method, as in Table 13.2, we get,
w = 37.07 kN/m2
Table 13.2. Solution of Example 13.9.
X w from Eq. I w from Eq. II
(m) (kN/m2) (kN/m2)
3.0 66.34 15.80
2.5 44.77 27.84
.3.32. 2.0 30.93 49.74
and Vertical Moments at Y,L.
about Line AC. 2.2 35.79 39.21
Referring to Fig. 13.32, we
have. 2.25 37.14 37.00

= 0
^e-Bcuc^SIAc , Example 13.10: Estimate the collapse concentrated load acting on a large size slab
=> causing fen pattern failure.

^(DE)(AOy^yxDe components of moments


along yield line
Solution: If a concentrated load acts on a reinforced concrete slab at an interior location, away
from any edge or comer, a negative yield line will form in a more-or-Iess circular
pattern with positive yield line radiating outward from the load point (shown in Fig.

— . —— 12x3+200x
w
x^Cx(6 -x)2x
c
xsinO
CD about AC
Let
13.33).
r = radius of the negative yield line

— = 200 sin© + 60 x cosO


-x)(sin0)
5w , P = applied concentrated load
(6 -x) 12x3+200x (6 m -
x2+25 -x) =
positive moment capacity
6 200 + 60 x / tanG m~ = negative moment capacity
and /3 = angular distance between the two positive yield lines in radians.
590 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SlDDlQi J CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 591
I 13.16 AFFINITY RULES
%
? Affine slab is an idealized equivalent slab considered for analysis purposes in place of an
orthotropic (different moment capacities in two mutually perpendicular, directions) and/or skew
I (sides of the slab or reinforcement at an angle different from 90°) slab. The affine slab is
I obtained by transforming the shape and loading of the orthotropic slab according to certain
rules, given below:

Fig. 13.33. Fan Type Yield Line


Pattern and Free Body Diagram of One Segment
Total load P First Coordinate Axis

Load per unit angle in radians= Fig. 13.35. Orthotropic and Corresponding Affine Slab.
2^
1. The original slab must have reinforcement in two directions parallel to the edges of the
Load per angle equal to p = slab (separated by an angle such that the ratio of the ultimate moments due to each
set of reinforcement taken separately is constant throughout the slab.
In Fig. 13.33, the projected length of positive 2. The deflections at the corresponding points of the skew and affine slabs are considered
yield lines AC and CB is the length
rp) and hence the total component AB (or equal.
of positive moment about X-axis is m rp.
negative moment about The total
axis-YYis m rp. Both the negative and the positive 3. The strength of the equivalent isotropic slab at a point corresponding to the actual slab
are counter-clockwise. moments
is considered equal to m in all directions.
4. The affine slab is drawn such that all distances measured in the direction of the m-
reinforcement (perpendicular to the m moment vector) remain the same. This direction
forms one co-ordinate axis for both slabs called the first coordinate axis.
5. The second direction in the skew slab, corresponding to the a m-reinforcement
direction, is taken at right angles to the first coordinate axis in the affine slab, and is
called the secondcoordinate direction.
Fig. 13.34. Sectional Front View of All distances in the affine slab in this coordinate direction are obtained by dividing
a Radial Strip through Fan Pattern.
corresponding lengths in the skew slab by 4a .

_=
Referring to Figs. 13.33 and 13.34, we have,
SAtcr = 0 => 6. All loads in the affine slab are obtained by modifying the original loads as under: -
m

P
- rP +mrP
=
——
PP r
2?r
2 K{m + m~)
a) Uniformly distributed load qu remains qu in the affine slab.

b) Point load Pn is replaced by


-^=-“
P
Pa^Siny
in tire affine slab.

The collapse load P is seen to be 13.17 STRIP METHOD OF SLAB DESIGN


independent of the fan radius r. With only
concentrated load acting, a complete fan of any
collapse load. A fan pattern will be radius could form with no change in
controlling failure mechanism if the collapse Strip method of slab analysis and design is a lower bound technique in which a slab is divided
for this mechanism is the minimum of load into strips based on a reasonably assumed load path and load distribution. Equilibrium is then
collapse loads for all other possible yield lines.
592 CONCRETE STRUCTURES PART - 1 CHAPTER 13: yield lineand strip methods for slabs 593
AUTHOR: ZAHID A. SIDDIQI
satisfied throughout the slab. This method has the following Fig. 13.36c. The lateral distribution of the moment capacity required at x = #2 is
advantages over the other shown in d-part of the same figure.
methods:
1. Strip method is applicable for various geometric shapes of
slabs such as rectangular,
triangular, circular and L-shaped slabs.
2. The method is applicable for slabs having unsupported
edges, skewed supports and
column supports.
3. Different types of loads, such as uniformly distributed
loads, patch loads, linearly
varying loads, line loads and concentrated loads may be considered.
4. Compared with yield line method, strip method is a design
tool.
5. Use of different amounts of reinforcement in various strips usually
makes the design
economical.
6. The designer has full control over the distribution of loads and placement
of
reinforcement.
7. The calculations for slab deflections are easy and straightforward.
Cracking may also
be minimized if care is taken to select the distribution of loads and a) Top View of Slab.
moments. ’

z
8. Due to lower bound approach, this method gives results on
the safe side. M
9. Slab with larger openings can also be designed.
10. Strong bands of steel may be used around openings or over 4
deflections, cracking and economy.
columns improving ^716
Z
13.17.1 Basic Assumptions 4
Kj
The following two assumptions are made in this method to simplify the
solution:
1. The torsional moment capacity is neglected, which gives extra safety in design.
2. The load distribution is reasonably assumed in the start to simplify <4 d) my Across Line x = £/2.
the analysis. The
choice of load transfer is to be selected to get safe and economical design and to
avoid
excessive cracking and deflections.
Strip Method of Andysis.
Fig. 13.36. First Load Distribution of a Square Slab for
13.17.2 Simply Supported Square Slab '4' 1

capacity equal to q^/16,


Consider a simply supported square slab having side length £ and subjected to a This assumption provides an answer for the required moment is that actudly the
uniformly distributed load of intensity q per unit area, as shown in Fig. 13.36. It is to be which is not an economicd or workable solution. The reason portions. For full
end
remembered that the steel for moment capacity vector, curvatures are high for the central portions as compared to the of moments
utilization of the flexurd strength of such a design, extensive redistributionregion of the
is placed in the y-direction.
Let k be the factor of total load carried by the x-direction strips and 1 k be the
of total load carried by the y-direction strips. - factor I would be required that may cause considerable cracking in the
centrd
sM slab.
First Load Distribution Pattern
4 Second Load Distribution Pattern
In this case, whole of the slab may be considered as strips in the two perpendicular segments as shown in Fig.
directions. The simplest pattern is obtained by assuming that half of the load is ’ The slab is considered to be divided into four triangular
regions and are placed dong
distributed in each direction, i.e., the solution is simplified by assuming that k = 0.5.
si 13.37a. The wavy lines indicate the boundaries of various
only in a single direction
Considering a unit width of slab dong section 1-1 of Fig. 13.36a, the load diagram the diagonds. Each slab segment is assumed to transfer load and required moment
this strip is shown in Fig. 13.36b and the required moment capacity (my) is calculated
for and the k-vdue is either zero or 1 for various segments. The load respectively.
the figure,
in capacity dong section 2-2 are shown in b-part and c-part of
594 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIqj YIELD LINE AND STRIP METHODS FOR SLABS 595
CHAPTER 13: on the beams, the beam
The distribution of the required moment capacity at x = Hl along the width is shown in in a single direction. If the slab panel is supported
distributed by calculating the slab segment
reactions. Figure 14.38b
d-part of the figure. The required maximum bending moment at the center will be loads can easily be evaluated the variation of my at x =
q^H. This diagram suggests that the spacing of bars is to be continuously varied along the variation of my along section 3-3, c-part represents
shows
my along section 4-4.
the width, which is practically almost impossible. Even if the reinforcement spacing is Hl and part-d explains the variation of
selected in steps according to average strength required in those steps, the reserve
strength compensates for the loss of strength. However, parts of the structure may go
into the inelastic range and the elastic equilibrium conditions may not be strictly-
satisfied in such cases.

, „ q q ) Variation of Across x f/2.


b) q* and my Along Section 3-3.

Varies From 0 to t at x = //2.


c) Along Section 2-2.
Fig. 13.37. Second Load Distribution of a Square Slab for Strip Method of Analysis.

Third Load Distribution Pattern


In the third pattern, two discontinuity lines are taken parallel to edges in both the
directions. The distance of these discontinuity lines are generally taken equal to /74 (as 4-4.
shown in Fig. 13.38a). The loads in the comer segments and the central portion are d) q^ and my Along Section
Analysis.
equally divided in the two directions, but the loads in the remaining two portions are of a Square Slab for Strip Method of
Fig. 13.38. Second Load Distribution
596 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDiqi FOR SL ABS 597
YIELD LINE AND STRIP METHODS
Calculation Of fOTy)may For Section 3-3 CHAPTER 13: Edges
Rectangular Slab With Two Adjacent
End reaction = qh x 11^ = q f /8 13.17.4 Continuous And Other Two Simply Supported
(^ykw —
= q f/8*£ /4 q £ /8 x£ /8 = q I x 1/8 = q £2/64 adjacent edges continuous and
other two simply supported
rectangular slab with two into strips as shown
A
in Fig. 13.39. The slab may be assumed to be distributed indicated. The
is shown is also
Calculation Of (my)m!,y For Section 4^1
a-part of the figure, where the assumed distribution of loads 5-5 are given in b-part,
in of moments along section
End reaction = qxU4 + q/2x Ha = \q £ loading and longitudinal distributions c-part of
'

moments along section 6-6 is shown in


(%)max = 3/g? f x £ /2
= 2
— —
g £ /4 x (£ /4 + £ /8) 9 f /8 x £ /8
q £ (3/16 - 3/32 - 1/64) = 5 q £2/64
while the longitudinal distribution
the iigure.
of

x-direction middle strip (section


5-5) is:
positive moment in the
In this option, a constant spacing in the two directions in the central strip and different The maximum
a£yl4
spacing in the outer strips may be selected. Out of the three options considered, this is myp = qal^/lx a^h- qa£yhx
acq£^lZ
the best, most practical, economical and safe solution. = qa£y linalyl A) =

13.17.3 - Slabs With Continuous Edges


1. Frequently, in this method, strips that are loaded near the supports and unloaded in the
central region are encountered. It is convenient to keep the unloaded region subjected
to a constant moment having zero shear force. This means that the entire load on a side
go to support on that side as a reaction. By using this treatment, the selection of the
positive reinforcement is simplified.
2. The simply supported moment (if it is made constant) in the central portion evaluates
equal to the cantilever moment at the respective fixed ends of loaded portions, if these
are separately considered as cantilevers. If end moments on both the ends are assumed
to be nearly equal, the positive moment in case of continuous span will be equal to
cantilever moment on any side (equal to simply supported moment) minus the
magnitude of the corresponding end moment. In other words, the sum of the absolute
values of positive span moment and negative moment at one of the ends (considering
full strip length) is equal to the negative moment if the half span beam acts as cantilever
(or only the loaded length on each side) fixed at the corresponding end. This means
that knowing the cantilever moments and selecting the span moment, the corresponding
support moment may be evaluated.
3. As a general rule, the support moment should be kept equal to about 1.5 to 2.5 times the
' span moment in the same strip. For longer, largely unloaded and having more than
minimum reinforcement strips, a ratio of support to span moment of 3 to 4 may be used.
4. If opposite edges of a slab are fixed and simply supported, the distance of discontinuity
lines of the segments from the fixed end is considered larger than the distance from the
simply supported end. The distance towards simply supported end may be taken equal
to a / 2 (fs is the shorter span) and the distance towards the fixed end will then
become (1 - a) £s / 2. Value of a is selected less than or equal to 0.5 and its values
between 0.35 and 0.39 give corresponding ratios of negative to positive moments of
2.45 to 1.45 (desired range).
b) and my Along Section 5-5.
Strip Method of Analysis.
Fig. 13.39. Rectangular Slab for
598 CONCRETE STRUCTURES PART -I
AUTHOR: ZAHID A. SIDDlyi
The cantilever moment at the CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 599
left support is:
mct ~ q(\-a) £y/2 x (}-a) Self weight = 0.19 x 2400 = 456 kg/m2
/4 = (1-a)2 /y2/8
The negative moment at the
left support is: q^ = 200 + 456 = 656 kg/m2
^yn ~ ^yp
q* = (1.2 x 656 + 1.6 x 600) x 9.81 / 1000 = 17.14 kN/m2
= (1-a)2 q ~dq £y/% = (1 2a) q
The ratio of negative to positive
-
moment is:


The moments in the
= l-2a
a2
x-direction
Similarly, y-direction outer strip edge strips are one-half of the middle strip moments.
outer strips (because the central results are exactly same as those for the x-direction
length is not important for
For the y-direction inner strip, calculations).
Alternatively, the ratio between actual results for propped cantilever may be
used.
direction may be employed for thethe positive and the negative
y-direction (the cantilever moments for the x-
distributed in the ratio of a2 and (1 - 2a) moment is approximately
Fig. 13.40. Rectangular Slab for Example 13.11.
giving: between positive and negative moments) ?

77 r
Cantilever moment = q lyll The assumed load distribution is shown in Fig. 13.41.
m%p = a2q£^/2 and z»xn = (1 -2<2) q X-Direction Middle Strip
To calculate the distance of t/h
maximum positive moment point from Cantilever moment = 17.14 x 1.52 / 2 = 19.28 kN-m/m
denoted by z, one might proceed as follows: the simple end,
This moment is to be divided in the ratio of 2 negative : 1 positive.
Propped reaction Negative moment = % x 19.28 = 12.86 kN-m/m
«^xz
Positive moment = a2 q 1^11 = qz x z - qz x'zll

Example 13.11: Design


z = a£y — q^ll
Positive moment = ’/3 x 19.28 = 6.43 kN-m/m
X- And Y- Direction Edge Strips
The moments are half of that for the x-direction middle strip.
a typical interior panel Negative moment = 12.86/ 2 = 6.43 kN-m/m
method, having clear spans as shown in Fig. 13.40 by using strip
indicated
superimposed dead load is 200 kg/m2. in the figure. Service live load is 600 kg/m2 and the Positive moment = 6.43 /2 = 3.22 kN-m/m
moment to span moment of 2.0. /'=20 MPa and fy = 420 MPa. Use a ratio of Y-Direction Middle Strip
support
Solution: Cantilever moment about left or right support considering half span
= 17.14 X3.02/ 2 = 77 13 kN-m/m
For stiff supports, ,
Anun
where/n = clear longer span = 8000mm
— —
/„(0.8 + /71400)
This moment is to be divided in the ratio of 2 negative : 1 positive.
Negative moment = 2/3 x 77.13 = 51.42 kN-m/m
and p
,
Amin
<
_
=
clear short span = 8/6 1.33
8000(0.8 + 420/1400)
- = 184 mm
- . .
Positive moment = ’/3 x 77.13 = 25.71 kN-m/m
A^ia = 0.0018 x 1000 x 190 = 342mm2/m (#10 @ 200 mm c/c)
d in short direction = 190-20-13/2 163 mm®
36 + 9x1.33 ,nn
(say 190 mm)
d in long direction — —
= 190 20— 13 10/2 = 152 mm
600 CONCRETE STRUCTURES PART -1
AUTHOR: ZAHID A. SIDDIQI CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 601

For = 25.71 kN-m/m, = 2221/1121. = 0.9677, p = 0.0027


bd2 1000xl632
As = 440mm2/m (#13@280mmc/c)
Extension Of X-Direction Negative Steel
The distance, x, from the left end up to the point of inflection can be determined as follows:
- 12.86 + 17.14 x 1.5 xx- 17.14 xx2/2 = 0
x2-3x+1.5 = 0

The extension beyond the point of inflection must be larger of d = 143 mm and 12db = 120
mm. Hence actual bar extension from face of support is equal to 0.634 + 0.143 = 0.777 m (say
0.8m).
Extension OfY-Direction Middle Strip Negative Steel
The distance, x, from any end up to the point of inflection can be determined as follows:
-51.42+ 17.14x3 xx- 17.14 xx2/ 2 = 0
xz-6x + 6 = 0
= 736-24
6+
x = 1.268 m
2
The extension beyond the point of inflection must be larger of d = 143 mm and 12db 156 =
mm. Hence actual bar extension from face of support is equal to 1.268 + 0.156 = 1.424 m
(say 1.5m).
b) x-direction middle strip 13.17.5 Slabs With Short Unsupported Edges
c) y-direction middle strip
Fig. 13.41. Load and Moment Distribution for Strip
Method of Analysis. A slab with three edges simply supported and one shorter edge free is shown in Fig.
13.42 and is subjected to uniformly distributed load of q kN/m2. In this method of slab
342 design, a wide and concealed beam, called strong band, is provided along the
pfor^min = = 0.00225 unsupported edge. This band resists some additional load from the perpendicular strips.
1000x152
A strong band is a wide beam having same total depth as the remainder of the slab but
From tables, = 0.8269 containing a concentration of reinforcement. The additional load in the direction
bd2 parallel to the unsupported edge is taken as k x q: In the perpendicular direction, an
= 0.8269 x 1000 x 1522 / 106 = 19.10 kN-m/m opposite load - k x q is considered for the same portion to keep the resultant load equal
This moment capacity is sufficient for all the to the original value.
moments except the y-direction middle strip.
Further, all selected spacing are less than 2ft. Due to one of the shorter side discontinuous and also otherwise having the shorter load
path, a larger fraction of load is carried along the shorter direction. This load directly
ForM = 51.42 kN-m/m,
bd2 = 1000xl632
-^^-12*.= 1.9353, 0.0055
goes to the longer side supports and this is particularly true for the short middle strip.
Some part of the load near the edges is transferred to the short side support on one side
As = 897 mm2/m (#13@140mmc/c) and to the strong band on the other side.
Considering a middle strip along x-direction, as shown in Fig. 13.42b, we have:
602 CONCRETE STRUCTURES PART -I
AUTHOR: ZAHID A. SIDDIQI CHAPTER 13: YIELD LINE AND STRIP MEfhOdsfor SLABS 603
According to one originator of this method, for section 9-9, the r;mn „ /, „ will -
^Mxedend - 0 (clockwise positive) 'YP —
depend on the shape and aspect ratio of the slab. If 4 is larger in comparison with £
the strong band will resist more loads and myn will approach the propped cantilever
m^q - kq (4/4)( 4 -
4 /8) = 0 moment of q£*18. If 4 and 4 ar® close, the strong band will deflect more increasing
the piyn to a value approximately equal to half the cantilever moment, q£*14 (because a
maximum of about half the load can go in one direction).
Knowing my^ and k, the distance of maximum positive moment from the left end (x)
may be found by considering the shear force (taken due to forces on the right hand side)
at maximum positive moment point equal to zero as follows:
~kqx 4/4 + q(tyl4 - x) = 0
x = 4/4 ~ k 4/4
= (1-Jt) 4/4
Taking moment of loads on the right of the maximum moment section,
Wyp = kq x 4/4 (4 - X - £y/i) ~ q (4 /4 - x)2/2
= ^4/4 (4 - 4/4 + W4 - 4/8) - 9 (4 /4 - 4/4 + kiyl^ll
= kqty/32 (84 - 34 + 2k£y) - q£ty / 32
= kq£y -
/ 32 (84 / 4 3 + 2k - k}

The moments in the x-direction edge strips are one-half of those in the middle strip.
The moments in the middle y-direction strip may be found as follows:
q(tjtf
Cantilever moment — —— = qty2 /8
Assuming the negative moment to be 2 times the positive moment.
= z/3 ?42 /8 - qt2 /12
= ’/3 qty 18 = ^y2 /24
The moments in the left edge strip in y-direction can be found as follows:

Cantilever moment = qty /64

Wxn = 2h qty /64 = qt2 /96


= ’/3 qty2 164 = qty /192

6) q* and my Along Section 9-9.


For the edge strip near unsupported end (strong band), the moments may conservatively
be found by considering the load (1 + tyq on the full span. Hence the moments will be
Fig- 13.42. Rectangular Slab with (1 + times the middle strip moments.
Free Shorter Edge.
CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 605

604 CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI

13.17.6 Slabs With Longer Unsupported Edges


A slab with three built-in edges and one longer edge unsupported is shown in Fig.
13.43a and is subjected to uniformly distributed load of q kN/m2. Again, in this case,
larger part of the load still has a natural tendency to go along the shorter path.
However, due to unavailability of a solid support at one edge and the deflections of the
strong band, the transfer of load in this direction is reduced to some extent The strips
along the longer span have to take more share of load as compared to a slab where all
the edges are supported. Even in such circumstances, a significant portion of the load is
carried in the shorter direction and the strong band has to withstand more load than the
other long strips.
For the y-direction strip along section 15-15, as shown in Fig. 13.43b, we have:
- mXn + V2 k^q (1 - a)2 ^2 - kzq aiy /2) = 0
2 x k^q a^y (1 - al 2) = 2 x % k\q (1 - a)1 ly - 2 x
Dividing throughout by qQ and rearranging, we get:

«2 = a(2 - a)
The constant a is reasonably selected to get the strong band designed as singly
reinforced section (preferably with lesser steel ratio). Close to minimum bottom steel is
usually provided for positive moment in x-direction strip supported along longer edge
and the value of ki is accordingly selected. Strips in the x-direction can easily be Fig. 13.43. Rectangular Slab with Free Longer Edge.
analyzed due to uniformly distributed load over the fill! length. Along the y-direction,
the value of may be selected equal to about half of the free cantilever moment up to
Referring to Fig. 3.44, let a = 0.2 to evaluate the strong band width and ii - 0.4.
the center of the strong band, ’A kiq ty (1 - at "2^. The kt value may then be
determined using the developed expression, which is then followed by the calculation For the y-direction strip along main slab, we have:
of positive moment. The assumed value of mxn automatically adjusts the load path in
the two directions. Greater value of means less vertical uplift/reaction provided by
the strong band, which is represented by the kt value. 4
0.4xl4.79x3.52(l-0.2/2/
= = 1468kN^M
Example 13.12: A 3.5 x 6 m slab is continuous on three sides and unsupported along one
of the longer side. Service live load is 600 kg/m2 in addition to the self-weight. fa'=25 MPa
-
and 7^ 420 MPa. Using strip method, design the slab. k
_ W-a}2 -2mjgt\
a(2-a)
4

Solution:

For a flat plate, ^min


i
where = clear long span = 6000 mm
=

0.4 x (1 0.2) 2
L^x3.5L = 0 2g!
0.2 x (2 -0.2)
6000
hmin = 182 mm (say 190
Shear force at the continuous end is equal to 13.863 kN/m.
Self weight “ 0.19 x2400 = 456 kg/m2
qt = 456 kg/m2 : qt = 600 kg/m2
qu = (1.2 x 456 + 1.6 x 600) x 9.81 / 1000 = 14.79 kN/m2
CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS
606 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI 607
Assuming the negative moment to be 2 times the positive moment
"lyn = 2/3 (I - ki)qC /8 = (l-^fx2 /12
= (1 - 0.4) x 14.79 x 62 / 12 = 26.62 kN-m/m
= % (1 - k^qC /8 = (1 - kx)q^ /24
= 14.79 X 62 /24 = 13.31 kN-m/m
(1 - 0.4) X

Similarly, the moments in the strong band are evaluated as follows:

+
Cantilever moment = = (1 + ki)q£x /8

Assuming the negative moment to be 2 times the positive moment,


4 = 6m
Fig. 13.44. Rectangular Slab of Example 13.12. mn = 2/3 (1 + k^qt? /8 = (1 + Wx2 /12
= (1+0.26 l)x 14.79 x62/ 12 = 55.95 kN-m
Assuming y to be the distance from the fixed end to zero shear force section of the y-direction
strip shown in Fig. 13.45, the following may be written:
mp -M+Wx2^ =(l+Wx2/24
= (1 + 0.261) x 14.79 x 62 / 24 = 27.98 kN-m
- 3.860 x 0.7 + 5.916 x (2.8 -y) = 0 (1 +£ }q£
y = 2.343jn Support reaction for the strong band = 7 1

The maximum positive moment is,


= (1 + 0.261) x 14.79 x 6 / 2 = 55.95 kN
=- 14.68 -5.916 x2.3432/2+ 13.863 x 2.343 The distance of point of inflection from the left end (xj) may be found as under:
= 1.563 kN-m/m
- 55.95 - + 55.95 X; = 0
x? - 6 Xi + 6 = 0 Xj = 1.268 m
The location of the point of inflection for the x-direction slab strips will also be evaluated as
above.
^s^nin = 0.0018 x 1000 x 190 = 342mmz/m (#10 @ 200 mm c/c)
d in short direction = 190 -20- 13/2 ~ 163 mm
J in long direction = 190-20- 13 - 10/2 = 152 mm

Fig. 13.45. Y-direction Strip for Slab of Fig. 13.44. /c '=25 MPa and /y = 420 MPa.
/ 342
The distance of point of inflection from the continuous end (y,) may be found as under:
p for ^s^nin = ’
1000x152
= 0.00225

5 916v2 From tables, =0.8316


- 14.68 + 13.863 y = 0 bd2.
y2 - 4.687 y + 4.963 = 0 y = 1.616m = 0.83 16 x 1000 x 1522 / 106 = 19.21 kN-m/m
This moment capacity is sufficient for all the moments except the negative moment for the x-
For the x-direction strips, the cantilever moment may be evaluated as under: direction strip and the moments for the strong band. Further, all spacing are to be selected less
Cantilever moment = —— = (1 - ki)qtx /s
than 2h.
608 CONCRETE STRUCTURES PART -I
AUTHOR ZAHID A. SIDDIQI CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 609
The reinforcement details are
shownin Fig, 13.46, while the
calculated below: amount of the reinforcement is As = 1016 mm2 (6-#16, @117 mm c/c)
x-direction negative moment for regular slab For positive moment of strong band = 27.98 kN-m,
= 26.62 kN-m/m,

As ~
26.62x10®
Mf = 1000x1632r = 1.0019, p 0.0028
mm2/m (#13 @ 275 mm c/c)
bd2 700xl632
= ,.5044, p- 0.0043
As = 491 mm2 (3-#16, @ 233 mm c/c)
Extension of bars equal to 12^4 is 120 mm for #10 bars, 156 mm for #13 bars and 180 mm for
#16 bars.

Fig. 13.46. Reinforcement Details


for Slab of Fig. 13.44.
For negative moment of strong
band = 55.95 kN-m,

bd2 700x163*
CHAPTER 13: YIELD LINE AND STRIP METHODS FOR SLABS 61 1
PROBLEMS

Note: N = Registered number of the student.


Ab = A fixed positive or negative number assigned by
the instructor to get R between 600 and 1000.
R = N - Ao, a parameter to get numerical data for the
problems.

1- Find wu in terms of ‘m’ for the given isotropically reinforced slab.

f R/250m

RHOOm

2- Find Wn in terms of ‘m' for the given isotropically reinforced slab.


1
i
l#ioo
m I
I
R/250m

3- Calculate the collapse load for the slab given in the related figure.

R/lOOkN-m/m R/50kN-m/m R/lOOkN-m/m

R/350 R/250 m R/350


R/100 kN~mAn U50 kN-m/nT "1V100 kN-m/tn
612 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI
613
4- For the orthotropically reinforced slab shown in figure, find the collapse load using
segment equilibrium method. Appendix-I:

5- For the isotropically reinforced slab shown in figure, find the collapse load in terms of
the moment using virtual work method.

R/150
Appendix - I
6- Calculate collapse load for the orthotropically reinforced square slab shown in figure by
using the virtual work method in terms of m, a and L.
Steel Tables

7- Design a typical interior panel of size .R/100 x 22/120 m built-in on all the four edges by
using strip method. Service live load is 400 kg/m2 and the superimposed dead load is
300 kg/m2. ^'=25 MPa and fy = 280 MPa. Use a ratio of support moment to span
moment of 2.0.

8- An 21/200 x 22/120 m slab is continuous on three sides and unsupported along one of the
longer side. Service live load is 450 kg/m2 in addition to the self-weight. ^'=20 MPa
and/y =280 MPa. Using strip method, design the slab.
614 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-I: 615
Space For Reader’s Notes
Standard US Customary Reinforcing Bars
SI US Solid Nominal Nominal
Designation Designation Diameter Area Mass
Number Number (mm) (mm2) (kgs/m)
6* *2 6.4 32 0.248
10 . 3 9.5 71 0.560
13 4 12.7 129 0.994
16 5 15.9 199 1.553
19 6 19.1 284 2.235
22 7 22.2 387 3.042
25 8 25.4 510 3.973
29 9 28.7 645 5.060
32 10 32.3 819 6.404
36 11 35.8 1006 7.907
43 14 43.0 1452 11.38
57 18 57.3 2581 20.24
* This diameter is not a standard bar diameter.

Cross Sectional Area Of US Customary Bars Per Meter Width (mrr^/m)


, 1000
As=~ s xA„A
Bar Bar Number
Spacing
(mm) 6 10 13 16 19 22 25
100 320 710 1290 1990 2840 3870 5100
110 291 645 1173 1809 2582 3518 4636
120 267 592 1075 1658 2367 3225 4250
130 246 546 992 1531 2185 2977 3923
140 229 507 921 1421 2029 2764 3643
150 213 473 860 1327 1893 2580 3400
160 200 444 806 1244 1775 2419 3188
170 188 418 759 1171 1671 2276 3000
180 178 394 717 1106 1578 2150 2833
190 168 374 679 1047 1495 2037 2684
200 160 355 645 995 1420 1935 2550
225 142 316 573 884 1262 1720 2267
250 128 284 516 796 1136 1548 2040
275 116 258 469 724 1033 1407 1855
300 107 237 430 663 947 1290 1700
350 1 91 203 369 569 811 1106 1457
400 80 178 323 498 710 968 1275
450 71 158 287 442 631 860 1133
616 CONCRETE STRUCTURES PART 1

l
J

Joar
1 Spacing
(mm)
100
110
_ _ 120
130
140
150
160
170
-180,
190
200
225
250
Ltd

300
350
400
450
-

29
__6450
5864
5375
4962
_ 4607
4300
4031
3794 j
3583
3395
3225
2867
2580
2345
2150
1843
1613
1433
_
32
8190
7445
6825
6300
5850
5460
11y
4818
4550
J1
4095
3640
3276
2978
2730
2340
2048
1820
Bar Number
36
10060
9145
8383
7738
7186
6707
6288
5918
5589
5295
5030
4471
4024
3658
3353
2874
2515
_2236
43

13200
12100
1 J 169
10371
9680
9075
8541
8067
7642
7260
6453
5808
5280
4840
4149
3630
3227
57
25810
23464
21508
19854
18436
17207
16131
15182
^14339
13584
_12905
11471
10324~
9385
8603
7374
6453
5736
1

r—
i


AUTHOR: ZAHID A. SIDDIQI

- _

i
j.-—
-Z.
Appendix-I:

Bar

No.
U6
U10
U13
U16
UI 9
U22
U25
U29
U32
U36
U43
U57

Bar
No.
U6
U10
UI 3
U16
1
32
71
129
199
284
387
510
645
819
1006
1452
2581
Area of Group of US Customary Bars in SI Units

2
64
142
258
398
568
774
1020
1290
1638
2012
2904
5162
3
96
213
387
597
852
1161
1530
1935
2457
3018
4356
7743
Number of bars
4
128
284
516
796
1136
1548
2040
2580
3276
4024
5808
10324
5
160
355
645
995
1420
1935
2550
3225
4095
5030
7260
12905
6
192
426
774
1194
1704
2322
3060
3870
4914
6036
8712
15486
7
224
497
903
1393
1988
2709
3570
4515
5733
7042
10164
18067

Area of Group of US Customary Bars in SI Units (Continued)

9
288
639
1161
1791
10
320
710
1290
1990
11
352
781
1419
2189
Number of bars
12
384
852
1548
2388
13
416
923
1677
2587
14
448
994
1806
2786
15
480
1065
1935
2985
8
256
568
1032
1592
2272
3096
4080
5160
6552
8048
11616
20648

16
512
1136
2064
3184
617

UI 9 2556 2840 3124 3408 3692 3976 4260 4544


U22 3483 3870 4257 4644 5031 5418 5805 6192
U25 4590 5100 5610 6120 6630 7140 7650 8160
U29 5805 6450 7095 7740 8385 9030 9675 10320
U32 7371 8190 9009 9828 10647 11466 12285 13104
U36 9054 10060 11066 12072 13078 14084 15090 16096
U43 13068 14520 15972 17424 18876 20328 21780 23232
U57 23229 25810 28391 30972 33553 36134 38715 41296
619
Appendix-I:
618 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQ]

Standard US Customary Reinforcing Bars Bars in Inch Units


Area of Group of US Customary
Designation Nominal Nominal Nominal
Number Diameter Area Mass Number of bars , i—

(in) (in2) (Ibs/ft) 5 6 7 8


2 3 4
*2 'A 0.049 0.167 No. 1
0.25 0.29 0.34 0.39
3 3/8 0.11 0.376 0.15 0.20
4 % 0.668 7 2~~ 0.05 0.10
0.33 0.44 0.55 0.66 0.77 : 0.88

_
0.20 0,11 0.22 1.57
3 0.98 1.18 138 _
5 % 0.31 1.043
4 0.20 0.39 _ 0.59 0.79 .
1.84 2.15 2.45
6 % 0.44 1.502
0.31 0.61 0.92 1.23 ~ 1.53 3.53
5 2.21 2.65 3.09
7 7A 0.60 2.044 0.44 0.88 1,33 1.77
4.21 4.81
8 1 0.79 2.67
6
2.41 J.Ui 3 61
J_J
I

0.60 1,20 1.81 5.50 6.29


9. 17» 1.00 3.400 2.36 3.14 3.93 4.72
0.79 1.57 7.00 8.00


LJ coIo
li
10 l'A 1.27 4.303 J 4.00 5.00 6.00
1

1,00 2.00 300 8.86 10.13


11 17» 1.56 5.313 5.06 6.33 7.60
1.27 2.53 3.80 10,94 12,50
14 17« 2.25 7.65 6.25 7.81 9.38
3.13 4.69 _ 18.00
18 2'A 4.00 13.60 <- v- 1.56
6.75 9.00 111.25 13.50 15.75
't 2.25 4.50 28.00 32.00
* This diameter is not a standard bar diameter. 4.0Q— j 12.00 16.00 j 20,00 j 24.00
« T 8.00
1

Cross Sectional Area Of US Customary Bars Per Foot Width (in2/fi)

Bar Bar Number


Spacing
(in) 2 3 4 5 6
3 0.20 0.44 0.79 1.23 1.77
3.5 0.17 0.38 0.67 1.05 1.51
4 0.15 0.33 0.59 0.92 1.33
4.5 0.13 0.29 0.52 0.82 1.18
5 0.12 0.26 0.47 0.74 • 1.06
5.5 0.11 0.24 0.43 0.67 0.96
6 0.10 0.22 0.39 0.61 0.88
6.5 0.09 0.20 0.36 0.57 0.82
7 0.08 0.19 0.34 0.53 0.76
-7.5 0 08 0.18 0.31 0.49 0.71
8 0.07 0.17 0.29 0.46 0 66
9 0.07 0.15 0.26 0.41 0.59
10 0.06 0.13 0.24 0.37 0.53
11 0.05 0.12 0.21 0.33 0.48
12 0.05 0.11 0.20 0.31 0.44
Appendix-I: 621
620 CONCRETE STRUCTURES PART -1
AUTHOR: ZAHID A. SIDDIQI
BS 4449-97 Reinforcing Bars

Designation Number Nominal Nominal


Area Mass
Standard SI Reinforcing Bars (mm2) (kgs/m)
6* 28.3 0.222
Designation Solid Nominal 8 50.3 0.359
Nominal
Number Diameter Area Mass 10 78.5 0.616
(mm) (mm2) 12 113.1 0.888
(kg/m)
S10 11.3 16 201.1 1.579
100 0.785
S15 16.0 20 314.2 2.466
200 1.870
S20 19.5 25 490.9 3.854
300 2.356
S25 25.2 32 804.2 6.313
500 3.925
S30 29.9 40 1256.6 9.864
700 5.495
S35 35.7 50* 1963.5 15.413
1000 7.850
S45 43.7 1500 * This diameter is not a preferred bar diameter.
1 1.775
S55 56.4 2500 1 19.625
Cross Sectional Area Of BS 4449-97 Per Meter Width (mm2/m)

Bar Bar Bar Number


Bar Number Spacing


Spacing 12 16 20 25
(mm) S10 (mm) 6 8 10
S15 S20 S25
100 283 503 785 1131 2011 3142 4909
100 1.000 2.000 3.000 5.000 110 257 457 714 1028 1828 2856 4463
115 0.870 1.739 2.609 4.348 120 236 419 654 943 1676 2618 4091
125 0.800 1.600 2.400 4.000 3776
140 _ 0.714 1.429 2.143 3.571
130
140
218
202
387
359
604
561
870
808
1547
1436
2417
2244 3506
150 0.667 1.333 2.000 3.333 150 189 335 523 754 1341 2095 3273
165 0.606 1.212 1.818 3.030 160 177 314 491 707 1257 1964 3068
175 0.571 1.143 1.714 2.857 170 166 296 462 665 1183 1848 2888
190 0.526 1.053 1.579 2.632 180 157 279 436 628 1117 1746 2727
200 0.500 1.000 1.500 2.500 190 149 265 413 595 1058 1654 2584
225 0.444 0.889 1.333 2.222 200 142 252 393 566 1006 1571 2455
250 0.400 0.800 1.200 2.000 225 126 224 349 503 . 894 1396 2182
300 0333 0.667 1.000 1.667 250 113 20-1 314 452 804 1257 1964
350 0286 0.571 0.857 1.429 275 103 183 285 411 731 1143 1785
400 0.250 0.500 0.750 1.250 300 94 168 262 377 670 1047 1636
450 0.222 0.444 0.667 Lill 350 81 144 224 323 575 898 1403
400 71 126 196 283 503 786 1227
450 63 112 174 251 447 698 1091
622 CONCRETE STRUCTURES PART 1 - AUTHOR: ZAHID A. SIDDIQI Appendix-II: 623

Standard Metric Prestressing Steels

Type Nominal Nominal Nominal


Diameter Area Weight
(mm) (mm2) (kg/m)
Seven Wire Strand 6.35 23.22 0.0182
(Grade 1725) 7.94 37.42 0.294
9.53 51.61 0.405
11.11 69.68 0.548
12.70 92.90 0.730
15.24 139.35 1.094
Seven Wire Strand 9.53 54.84 0.432
(Grade 1860)

Prestressing Wire
(Grades 1620, 1655, 1725)
11.11
12.70
15.24
4.88
4.98
74.19
98.71
140.00
18.7
19.4
0.582
0.775
1.102
0.146
0.149
Appendix — II

Beam and Slab Design Aids


6.35 32 0.253
7.1 39 0.298
Prestressing Bars, Plain 19 284 2.23
(Grades 1035, 1100) 22 387 3.04
25 503 3.97
29 639 5.03
32 794 6.21
35 955 7.52
Prestressing Bars, Deformed 15 181 1.46
(Grades 1035, 1100) 20 271 2.22
26 548 4.48
32 806 6.54
36 1019 8.28
624 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIq]
Space For Reader’s Notes
Appendix-II: 625

Moment Capacity Of Rectangular Sections


fy = 280 MPa
V = 17.25 MPa fc‘ = 20.0 MPa fc' = 25.0 MPa
MJbd2 Mu/bd2 P
Mu/bd1 P
P
MPa MPa MPa
1.1998 0.0050 1.2081 0.0050 1.2185 0.0050
1.2798 0.0054 1.3067 0.0054 1.3512 0.0056
1.3591 0.0057 1.4045 0.0059 1.4829 0.0061
1.4378 0.0061 1.5016 0.0063 1.6135 0:0067
1.5159 0.0064 1.5979 0.0067 1.7431 0.0073
1.5935 0.0068 1.6934 0.0071 1.8716 0.0078
1.6704 0.0071 1.7882 0:0076 1.9990 0.0084
1.7467 0.0075 1.8822 0.0080 2.1253 0.0090
1.8225 0.0078 1.9755 0.0084 2.2506 0.0095
1.8976 0.0082 2.0680 0.0089 2.3749 0.0101
1.9722 0.0085 2.1597 0.0093 2.4980 0.0107
2.0461 0.0089 2.2507 0.0097 2.6201 0.0112
2.1195 0.0092 2.3409 0.0101 2.7412 0:0118
2.1923 0.0096 2.4303 0.0106 2.8611 0.0124
2.2644 0.0099 2.5190 0.0110 2.9800 0.0129
2.3360 0.0103 2.6070 0.0114 3.0979 0.0135
2.4070 0.0106 2.6942 0.0118 3.2146 0.0141
2.4774 0.0110 2.7806 0.0123 3.3304 0.0146
2.5471 0.0113 2.8662 0.0127 3.4450 0.0152
2.6163 0.0117 2.9511 0.0131 3.5586 0.0158
2.6849 0.0120 3.0353 0.0136 3.6711 0.0163
2.7529 0.0124 3.1186 0.0140 3.7826 0.0169
2.8203 0.0127 3.2013 0.0144 3.8930 0.0175
2.8871 0.0131 3.2831 0.0148 4:0023 0.0180
2.9533 0.0134 3.3642 0.0153 4.1105 0.0186
3.0189 0.0138 3.4446 0.0157 4.2177 0.0192
3.0840 0.0141 3.5241 0.0161 4.3239 0.0197
3.1484 0.0145 3.6029 0.0166 4.4289 0.0203 '

3.2122 0.0149 3.6810 0.0170 4.5329 0.0209


3 2754 0.0152 3.7583 0:0174 4.6359 0.0214
3.3381 0.0156 3.8348 0.0178 4.7378 0.0220
3.4001 0.0159 3.9106 0.0183 ' 4.8386 0.0226
3.4615 0.0163 3.9856 0.0187 4.9383 0:0231
3.5224 '0.0166 4.0599 0.0191 5.0370 0.0237
3.5470 0:0170 4.1077 0.0195 5.1269 0.0242
3.5611 0.0173 4.1249 0.0200 5.1499 0.0248
3.5750 0.0177 4.1418 0.0204 5.1724 0.0254
3.5887 0.0180 4.1585 0.0208 5.1945 0:0259
3.6021 0.0184 4.1748 0.0213 5.2162 0.0265
3.6153 0.0187 4.1909 0.0217 5.2374 0.0271
3.6282 0.0191 4.2067 0.0221 5.2583 0.0276
626 CONCRETE STRUCTURES PART - 1 •

627
AUTHOR: ZAHID A. SIDDIQI
App611^"^'

Moment Capacity Of Rectangular Sections


Moment Capacity Of Rectangular Sections
fv = 300 MPa fy = 350 MPa
1 17.25 MPa f.' b 5n n
1 Mjbd2 — ~ * 5.0 Mra
fc' = 17.25 MPa V = 2O.() MPa fc' = 25.( MPa
p MJbd2 M„/bdz M,/bdz
MPa MPa
P
MPa
p Mu/bd2 P
Mu/bd2 P P
| 1996 MPa MPa
0.0047 1.2081 0.0047 1.2185 MPa
1 1.2798 0.0050 1.3067 0.0051 1.3512
0.0047
0.0052 1.1998 0.0040 1.2081 0.0040 1.2185 0.0040
0.0045
1.3591 0.0053 1.3067 0.0043 1.3512
1.4045 0.0055 1.4829 0.0057 1.2798 0.0043 0.0049
1.4378 0.0057 1.5016 0.0046 1.4045 0.0047 1.4829
0.0059 1.6135 0.0063 1.3591 0.0054
1.5159 0.0060 1.5979 0.0048 1.5016 0.0050 1.6135
0.0063 1.7431 0.0068 1.4378
1.5935 1.5979 0.0054 1.7431 0.0058
0.0063 1.6934 0.0067 1 8716 0.0073 1.5159 0.0051
1.6704 0.0066 1.6934 0.0057 1.8716 0.0063
1.7882 0.0071 1.9990 0 0078 1.5935 0.0054
1.7467 0.0070 1.8822 0.0057 1.7882 0.0061 1.9990 0.0067
0.0075 2.1253 0 0084 1.6704
1.8225 0.0073 1.8822 0.0064 2.1253 0.0072
1.9755 0.0079 2.2506 0 008Q 1.7467 0.0060
1.8976 0.0076 1.9755 0.0067 2.2506 0.0076
2.0680 0.0083 2.3749 00094 1.8225 0.0063
1.9722 0.0080 2.1597 0.0065 2.0680 0.0071 2.3749 0.0081
0.0087 2.4980 0 0100 1.8976
2.046.1 0.0083 2.1597 0.0074 2.4980 0.0085
2.2507 0.0091 2.6201 0 0105 1.9722 0.0068
2.1195 0.0078 2.6201 Q.009U
0.0CB6 2.3409 0.0095 2.7412 0 0110 2.0461 0.0071 2 2507
2.1923 0.0089 2.3409 0.0081 2.7412 0.0094
2.4303 0.0099 2.86 11 0 0115 2.1195 0.0074 0.0099
[ 2.2644 0.0093 0.0077 2.4303 0.0085 2.8611
25190 0.0103 2.9800 0 0121 2.1923
2.3360 0.0096 2.5190 0.0088 2.9800 0.0103
2.6070 0.0107 3.0979 0 0126 2.2644 0.0079
2.4070 0.0099 2.6070 0.0091 3.0979 0.0108
2.6942 0.0111 3.2146 0 0131 2.3360 0.0082
2.4774 0.0103 2.6942 0.0095 3.2146 0.0112
2.7806 0.0115 3.3304 0 0136 2.4070 0 0085
I 2.5471
2.6163
0.0106
0.0109
2.8662 0.0119 3.4450 0 0142 2.4774 0.0088
0.0091
2.7806
2.8662
0.0098
0.0102
3.3304
3.4450
0.0117
00122
2.9511 0.0123 3.5586 0.0147 2.5471
2.6849 0.0112 2 9511 0.0105 3.5586 0.0126
3.0353 0.0127 3.6711 0.0152 2.6163 0.0093
2.7529 0 0116 3.0353 0.0108 3.6711 0.0131
3.1186 0.0131 3.7826 0.0158 2.6849 0.0096
28203 3.1186 0.0112 3.7826 0.0135
0.0119 3.2013 0.0135 3.8930 0.0163 2.7529 0.0099
2.8871 3 2013 0.0115 3.8930 0.0140
0.0122 3.2831 0.0139 4.0023 2.8203 0.0102
0.0168 0.0119 4.0023 0.0144
2.9533 0.0125 3.3642 0.0143 2.8871 0 0105 3.2831
4.1105 0.0173 0.0122 4.1105 0.0149
3.0189 0.0129 3.4446 0.0147 2.9533 0.0108 3.3642
4.2177 0.0179 0.0153
3.0840 0.0132 3.0189 0.0110 3.4446 0.0126 4.2177
3.5241 0.0151 4.3239 0.0184 0 0158 ’
3.1484 0.0135 0.0113 3.5241 0.0129 4.3239
3.6029 0.0155 4.4289 0.0189 3.0840
3.2122 0.0139 3.6029 0.0132 4.4289 0.0162
3.6810 0:0159 4.5329 0.0195 3.1484 0.0116 0.0167
3.2754 0.0142 0.0119 3.6810 0.0136 4.5329
3.7583 0.0162 4.6359 0.0200 3.2122
0.0171
3.3381 0.0145 0.0122 3.7583 0.0139 4.6359
3.8348 0.0166 4.7378 0 0205 3.2754 0.0176
3.4001 0.0148 0.0124 3.8348 0.0143 4.7378
3.9106 0.0170 4.8386 0.0210 3.3381 0.0180
3.4615 0.0152 0.0127 3.9106 0.0146 4.8386
3.9856 0.0174 4.9363 0:0216 3.4001
3.5224 0.0155 3 9856 0 0150 4.9383 0.0185
4.0599 0.0178 5.0370 0.0221 3.4615 0.0130
3.5460 4.0599 0.0153 5.0370 0.0189
0.0158 4.1069 0.0182 5.1267 0.0226 3.5224 0.0133
3.5588 4.1049 0.0156 5.1261 0.0194
0.0162 4.1226 0.0186 5.1475 0.0232 3.5432 0.0136
3.5714 4.1161 0.0160 5.1411 0.0199
0.0165 4.1379 0.0190 5.1680 3.5524 0.0139
3.5838 0.0237 4.1271 0.0163 5.1557 0.0203
0.0168 4.1530 0.0194 5.1880 0:0242 3.5614 0.0141
3.5959 4.1379 0.0167 5.1701 0 0208
0.0171 4.1678 0.0198 5.2076 3.5702 0.0144
0.0247 0.0170 5.1840 0.0212
3.6079 0.0175 4.1824 0.0202 3.5789 0.0147 4.1484
5.2269 0.0253
I 3.6196 0.0178 4.1966 0.0206' | 3.5873 0.0150 4.1587 0.0174 5.1977
|
0.0217
5.2458 0.0258 Ill 3.5956 0.0153 4.1688 I 0 0177 5.2110 0 0221
628 CONCRETE STRUCTURES PART - 1 JI 629
AUTHOR: ZAHID A. SIDDIQI Appendix-II:

Moment Capacity Of Rectangular Sections


Moment Capacity Of Rectangular Sections
fy = 420 MPa
fy = 460 MPa
V = 17.25 MPa fe' = 20.0 MPa f/ = 25.0 MPa
f' = 17.25 MPa f/ = 20.0 MPa fc’ = 25.0 MPa
M^bd2 Mu/bd2 M„/bd2 Mu/bd2
MPa
P
MPa
P P MJbd2 Mu/bd2 P P
MPa P
MPa MPa
1.1998 0.0033 1.2081 MPa
0.0033 1.2185 0.0033 0.0030 1.2185 0.0030
1.2798 0.0036 1 1998 0.0030 1.2081
1.3067 0.0036 1.3512 0.0037 0.0033 1.3512 0.0034
1.3591 0.0038 1.4045 1 2798 0.0033 1.3067
0.0039 1.4829 0.0041 0.0036 1.4829 0.0037
1.4378 0.0040 1 3591 0.0035 1.4045
1.5016 0.0042 1.6135 0.0045 1.6135 0.0041
1.5159 1 4378 0.0037 1.5016 0.0038
0.0043 1.5979 0.0045 1.7431 0.0048 1.7431 0.0044
1.5935 1.5159 0.0039 1.5979 0.0041
0.0045 1.6934 0.0048 1.8716 0.0052 0.0048
0.0041 1.6934 0.0043 1.8716
1.6704 0.0047 1.7882 0.0050 1.5935
1.9990 0.0056 0 0043 1.7882 0.0046 . 1.9990 0.0051
1.7467 0.0050 1.8822 0.0053 2.1253 1.6704
0.0060 1.8822 0.0049 2.1253 0.005b
1.8225 0.0052 1.9755 0.0056 1.7467 0.0045
2.2506 0.0064 1.9755 0.0051 2.2506 0.0058
1.8976 0.0054 2.0680 0.0059 1 8225 0.0048
2.3749 0.0067 2.0680 0.0054 2.3749 0.0061
1.9722 0.0057 2.1597 1 8976 0.0050
0.0062 2.4980 0.0071 0.0056 2.4980 0.0065
2.0461 0 0059 1 9722 0.0052 2.1597
2.2507 0.0065 2.6201 0.0075 0.0059 2.6201 0.0068
2.1195 0.0061 2 0461 0.0054 2.2507
2.3409 0.0068 2.7412 0.0079 2.7412 0.0072
2.1923 2 1195 0.0056 2.3409 0.0062
0.0064 2.4303 0.0070 2 8611 0.0082 0.0075
2.2644 0.0058 2.4303 0.0064 2.8611
0.0066 2.5190 0.0073 2.9800 2 1923 0.0079
0.0086 0.0060 2.5190 0.0067 2.9800
2.3360 0.0069 2.6070 0.0076 3.0979 2 2644
2.4070
0.0090
0.0063 2.6070 0.0070 3.0979 0.0082
0.0071 2.6942 0.0079 3.2146 2 3360
0.0094 2.6942 0.0072 3.2146 0.0086
2.4774 0.0073 2.7806 0.0082 3.3304 2 4070 0.0065
0.0097 2.7806 0 0075 3.3304 0.0089
2 5471 0.0076 2.8662 0.0085 2 4774 0.0067
3.4450 0.0101 0.0077 3.4450 0.0092
2.6163 0.0078 2.9511 25471 0.0069 2.8662
0.0088 3.5586 0.0105 0.0080 3.5586 0.0096
2.6849 0.0080 3.0353 2.6163 0.0071 2.9511
0.0090 3.6711 0.0109 0 0083 3.6711 0.0099
2.7529 0.0083 3.1186 2 6849 0.0073 3.0353
0.0093 3.7826 0.0113 0.0085 3.7826 0.0103
2.8203 0.0085 3.2013 2 7529 0.0075 3.1186
0.0096 3.8930 0.0116 3.2013 0.0088 3.8930 0.0106
2.8871 0.0087 3.2831 0.0099 2 8203 0.0078
4.0023 0.0120 0.0090 4.0023 0.0110
2 9533 0.0090 3.3642 2 8871 0 0080 3.2831
0.0102 4.1105 0.0124 0.0093 4.1105 0.0113
3.0189 0.0092 3.4446 2.9533 0.0082 3.3642
0.0105 4.2177 0.0128 0.0096 4.2177 0.0117
3.0840 0.0094 3.5241 3 0189 0.0084 3.4446
0.0108 4.3239 0.0131 0.0098 4.3239 0.0120 ,
3.1484 0.0097 3.6029 3 0840 0.0086 3.5241
0.0110 4.4289 0.0135 0.0101 4.4289 0.0123
3.2122 0.0099 3.6810 3 1484 0.0088 3.6029
0.0113 4.5329 0.0139 0.0103 4.5329 0.0127
3.2754 0.0101 3.7583 3 2122 0.0090 3.6810
0.0116 4.6359 0.0143 0.0106 4.6359 0.0130
3.3381 0.0104 3.8348 3 2754 0.0093 3.7583
0.0119 4.7378 0.0147 0.0109 4.7378 0.0134
3.4001 0.0106 3 3381 0.0095 3.8348
3.9106 0.0122 4 8386 0.0150 4.8386 0.0137
3.4615 0.0108 3 4001 0.0097 3.9106 0.0111
3.9856 0.0125 4.9383 0.0154 4.9383 0.0141
3 5224 0.0111 3 4615 0.0099 3.9856 0.0114
4.0599 0.0127 5.0370 0.0158 5.0370 0.0144
3.5384 3 5224 0.0101 4.0599 0.0116
0.0113 4.1014 0.0130 5.1251 0.0162 5.1244 0.0148
3.5415 3 5352 0.0103 4.0991 0.0119
0.0115 4.1053 0.0133 5.1302 0.0165 5.1227 0.0151
3.5445 3 5341 0.0105 4.0978 0.0122
0.0118 4.1089 0.0136 5.1351 0.0169 0.0154
0.0108 4.0964 0.0124 5.1209
3.5474 0.0120 4.1124 0.0139 5.1398 3.5329 0.0158
0.0173 0.0110 4.0949 0.0127 5.1189
3.5500 0.0122 4.1157 0.0142 3.5316
5.1442 0.0177 4.0932 0:0129 5.1167 0.0161
3.5526 0.0125 4.1188 0.0145 3 5302 0.0112
5.1483 0.0181 0.0114 4.0914 0.0132 5.1143 0.0165
3.5550 0.0127 4.1218 0.0147 5.1522 3 5287
_00184 0.0168
3.5271 _ 0.0116 4.0894 0.0135 5.1118
630 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI 631
Appendix-II:

Moment Capacity Of Rectangular Sections Moment Capacity Of Rectangular Sections


fy = 520 MPa
fc' = 17.25 MPa I fc‘ = 20.0 MPa = 25.0 MPa
M^bd2 Mu/bd* Mu/bd2
MPa P P P
MPa MPa
1.1998 0.0027 -
1.2081 0.0027 1.2185 0.0027
1.2798 0.0029 1.3067 0.0029 1.3512 0.0030
1.3591 0.0031 1.4045 0.0032 1.4829
1.4378
0.0033
0.0033 1.5016 0.0034 1.6135 0 0036
1.5159 0.0035 1.5979 0.0036 1.7431 0.0039
1.5935 0,0036 1.6934 0.0038 1.8716 O.0042
1,6/04 0.0038 1.7882 0.0041 1.9990 0.0045
1.7467 0.0040 1.8822 0.0043 2.1253 0.0048
1.8225 0.Q042 1.9755 0.0045 2.2506 0.0051
1.8976 0.0044 2.0680 0.0048 2.3749 0.0054
1.9722 0.0046 2.1597 0.0050 2.4980 0.0057
2.0461 0.0048 2.2507 0.0052 2.6201 0.0060
2.1195 0.0050 2.3409 0.0055 2.7412 0.0064
2.1923 0.0052 2:4303 0.0057 2.8611 0.0067
2.2644 0.0053 2.5190 0.0059 2.9800 0.0070
2.3360 0.0055 2.6070 0.0061 30979 0.0073
2.4070 0.0057 2.6942 0.0064 3.2146 0.0076
2.4774 0.0059 2.7806 0 0066 3.3304 0.0079
2.5471 0,0061 2.8662 0.0068 3.4450 0.0082
2.6163 0.0063 2.9511 00071 3.5586 0.0085
2.6849 0.0065 3.0353 0.0073 3.6711 0.0088
2.7529 0.0067 3.1186 0.0075 3.7826 0.0091
2.8203 0.0069 3.2013 0.0078 3.8930 6.0094
2.8871 0.0071 3.2831 0.0080 4.0023 0.0097
2.9533 0.0072 3.3642 0.0082 4 1105 0.0100
3.0189 0-0074 34446 0.0085 4.2177 0.0103
3.0840 0.0076 3.5241 00087 4.3239 0.0106
3.1484 0.0078 3.6029 0.0089 4.4289 0.0109
3.2122 0.0080 3 6810 0.0091 4.5329 0.0112
3.2754 0.0082 3.7583 0.0094 4.6359 0.0115
3.3381 0.0084 3,8348 0.0096 4,7378 0.0118
3.4001 0.0086 3.9106 0.0098 4.8386 0.0121
3.4615 0.0088 3.9856 0.0101
3.5224
4 9383 0.0124
0.0089 4.0599 0 0103 5.0370 0.0128
3.5292 0.0091 4.0948 0.0105 5.1231 0.0131
3.5206 0.0093 4.0843 0 0108 5.1092 0.0134
3.5118 0.0095 4.0737 0.0110 5.0952 ,
0 0137
3.5031 0.0097 4.0630 0.0112 5.0811 0.0140
3.4943 0.0099 4.0524 0.0114 5.0670 0.0143
3.4854 0.0101 4.0416 0.0117 5.0528 0.0146
3.4766 0.0103 4.0308 0 0119 5.0385 | 0.0149
632 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-II: 633

Moment Capacity Of Rectangular Sections


Moment Capacity Of Rectangular Sections

(MPa) (MPa)
HUM’ Mu/bda

Steel Ratio (p)


CONCRETE STRUCTURES PART -1 AUTHOR: ZAHID A. SIDDIQI Appendix-II: 635

Moment Capacity Of Rectangular Sections Moment Capacity Of Rectangular Sections


fy = 40 ksi fy = 60 ksi
-
fc' 2.5 ksi fc' = 3.0 ksi L' = 4.0 ksi fj 2.5 ksi f ' = 3.0 ksi f,.' = 4.0 ksi
Mu/bd2 Mu/bd2 Mu/bd2 Mu/bd2 Mu/bd2 Mu/bd2
P P P P P P
ksi ksi ksi ksi
ksi ksi
0.1715 0.0050 0.1729 0.0050 0.1747 0.0050 0.0033 0.1729 0.0033 0.1747 0.0033
0.1715
0.1832 0.0054 0.1880 0.0055 0.1966 0.0056 0.1880 0.0036 0.1966 0.0038
01832 0.0036
0.1948 0.0057 0.2029 0.0059 0.2183 0.2029 0.0039 0.2183 0.0042
0.0063 0.1948 0.0038
0.2062 0.0061 0.2178 0.0064 0.2399 0.0069 0Q041 0.2178 0.0042 0.2399 0.0046
0.2062
0.2176 0.0064 0.2325 0.0068 0.2613 0.0043 0.2325 0.0045 0.2613 0.0051
0.0076 0.2176
0.2289 0.0068 0.2470 0.0073 0.2824 0.0082 0.0045 0.2470 0 0049 0.2824 0.0055
0.2289
0.2402 0.0072 0.2615 0.0077 0.3035 0.0089 0.0048 0.2615 0.0052 0.3035 0.0059
0.2402
0.2513 0.0075 0.2759 0.0082 0.3243 0.0095 0.0050 0.2759 0.0055 0.3243 0.0064
0:2513
0.2624 0.0079 0.2901 0.0086 0.3449 0.0102 0.0052 0.2901 0.0058 0.3449 0.0068
0.2624
0.2733 0.0082 0.3042 0.0091 0.3654 0.3042 0.0061 0.3654 0.0072
0.0108 0.2733 0.0055
0.2842 0.0086 0.3182 0.0096 0.3857 0.0115 0.2842 0.0057 0.3182 0.0064 0.3857 0.0077
0.2950 0.0089 0.3321 0.0100 0.4058 0.3321 0.0067 0.4058 0.0081
0.0121 0.2950 0.0060
0.3057 0.0093 0.3459 0.0105 0.4258 0.0128 0.0062 0.3459 0.0070 0.4258 0.0085
03057
0.3163 0.0097 0.3595 0.0109 0.4455 0.0134 0.3163 0.0064 0.3595 0.0073 0:4455 0.0090
0.3268 0.0100 0.3731 0.0114 0.4651 0.0141 0.3268 0.0067 0.3731 0.0076 0.4651 0.0094
0.3372 0.0104 0.3865 0.0118 0.4845 0.0147 0.0069 0.3865 0.0079 0.4845 0.0098
0.3372
0.3476 0.0107 0.3998 0.0123 0.5038 0.0154 0.0072 0.3998 0.0082 0.5038 0.0103
0.3476
0.3579 0.0111 0.4130 0.0127 0.5228 0.0160 0.3579 0.0074 0.4130 00085 0.5228 0.0107
0.3680 0.0115 0.4260 0.0132 0.5417 0.0167 0.3680 0.0076 0.4260 0 0088 0.5417 0.0111
0.3781 0.0118 0.4390 0.0137 0.5604 0.0079 0.4390 0.0091 0.5604 0.0116
0.0173 0.3781
0.3881 0.0122 0.4518 0.0141 0.5789 0.0180 0.3881 0.0081 0.4518 0.0094 0.5789 0.0120
0.3980 0.0125 0.4645 0.0146 0.5972 0.0186 0.3980 0.0084 0.4645 0.0097 0.5972 0.0124
0.4079 0.0129 0.4771 0.0150 0.6154 0.0193 0.0086 0.4771 0.0100 0.6154 0.0129
0.4079
0.4176 0.0133 0.4896 0.0155 0.6334 0.0199 0.4176 0.0088 0.4896 0.0103 0.6334 0.0133
0.4272 0.0136 0.5019 0.0159 0.6512 0.0206 0.0091 0 5019 0.0106 0.6512 0.0137
0.4272
0.4368 0.0140 0.5142 0.0164 0.6688 0.0212 0.0093 0.5142 0.0109 0.6688 0.0142
0.4368
0.4463 0.0143 0.5263 0.0168 0.6862 0.0219 0.0096 0.5263 0.0112 0.6862 0.0146
0.4463
0.4557 0.0147 0.5383 0.0173 0.7035 0.0225 0.0098 0.5383 0.0115 0.7035 0.0150
0.4557
0.4650 0,0150 0.5502 0.0178 0.7206 0.0232 0.0100 0 5502 0.0118 0.7206 0.0155
0.4650
0.4742 0.0154 0.5620 0.0182 0.7375 0.0103 0.5620 0.0121 0.7375 0.0159
0.0238 0.4742
0.4833 0.0158 0.5736 0.0187 0.7542 0.0245 0.0105 0.5736 0.0124 0.7542 0.0163
0.4833
0.4924 0.0161 0.5852 0.0191 0.7708 0.0251 0.0107 0.5852 0.0127 0.7708 0.0167
0.4924
0.5013 0 0165 0.5966 0.0196 0.7871 0.0258 0.5013 0.0110 0.5966 0.0131 0:7871 0.0172
0.5102 0.0168 0.6079 0.0200 0.8033 0.0264 0.0112 0.6079 0.0134 0.8033 0.0176
0.5102
0.5140 0.0172 0.6159 0.0205 0.8193 0.0271 06152 0:0137 0.8193 0.0180
0.5129 0.0115
0.5161 0.0176 0.6186 0.0209 0.8236 0.0277 0.0117 0.6159 0.0140 0.8209 0.0185
0.5134
0.5182 0.0179 0.6212 0.8219 0.0189
0.0214 0.8274
"J 0 0143
'
0.0284 0.5139 0.0119 0.6166
0.5202 0.0183 0.6238 0.0219 0.8311 0.6172 0.0146 0.8228 0.0193
0.0290 0.5144 0.0122
0.5222 0.0186 0.6264 0.0223 0.8347 0.0297 0.6178 0.0149 0.8237 0.0198
0.5149 0.0124
0.5242 0.0190 0.6289 0.0228 0.8382 0.0303 0.0127 0.6184 0.0152 0.8245 0.0202
0.5154
0.5261 0.0194 0.6313 0.0232 0.8417 0.6190 0.0155 0.8253 0.0206
0.0310 0.5158 0.0129
636 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-II: 637

Moment Capacity Of Rectangular Sections Moment Capacity Of Rectangular Sections

fa = 75 ksi
fc' = 2.5 ksi fo' = 3.0 ksi fc'=4.0 ksi
Mu/bd2 Mu/bd2 Mu/bd2
P P P
ksi ksi ksi
0.1715 0.0027 0.1729 0.0027 0.1747 0.0027
0.1832 0.0029 0.1880 0.0029 0.1966 0.0030
0.1948 0.0030 0.2029 0.0032 0.2183 0.0034
0.2062 0.0032 0.2178 0.0034 0.2399 0.0037
0.2176 0.0034 0.2325 0.0036 0.2613 0.0041
0.2289 0.0036 0.2470 0.0039 0.2824 0.0044
0.2402 0.0038 0.2615 0.0041 0.3035 0.0047
0.2513 0.0040 0.2759 0.0044 0.3243 0.0051
0.2624 0.0042 0.2901 0.0046 0.3449 0.0054
0.2733 0.0044 0.3042 0.0049 0.3654 0.0058
0.2842 0.0046 0.3182 0.0051 0.3857 0.0061
0.2950 ... 0.0048 0.3321 0.0053 0.4058 0.0065
0.3057 0.0050 0.3459 0.0056 0.4256 0.0068
0.3163 0.0052 0.3595 0.0058 0.4455 0.0072
0.3268 0.0053 0.3731 0.0061 0.4651 0.0075
0.3372 0.0055 0.3865 0.0063 0.4645 0.0079
0.3476 0.0057 0.3998 0.0066 0.5038 0.0082
0.3579 0.0059 0.4130 0.0068 0.5228 0.0086
0.3680 0.0061 0.4260 0.0070 0.5417 0.0089
0.3781 0.0063 0.4390 0.0073 0.5604 0.0092
0.3881 0.0065 0.4518 0.0075 0.5789 0.0096
0.3980 0.0067 0.4645 0.0078 0.5972 0.0099
0.4079 0.0069 0.4771 0.0080 0.6154 0.0103
0.4176 0.0071 0.4896 0.0083 0.6334 0.0106
0.4272 0.0073 0.5019 0.0085 0.6512 0.0110
0.4368 0.0075 0.5142 0.0087 0.6668 0.0113
0.4463 0.0076 0.5263 0.0090 0.6862 0.01:17
0.4557 0.0078 0.5383 0.0092 0.7035 0.0120
0.4650 0.0080 0.5502 0.0095 0.7206 0.0124
0.4742 00082 0.5620 0.0097 0.7375 0.0127
0.4833 0,0084 0.5736 0.0100 0.7542 0.0131
0.4924 0.0086 0.5852 0.0102 0.7708 0.0134
0.5013 0.0088 0.5966 0.0104 0.7871 0.0137
0.5102 0.0090 0.6079 0.0107 0.8033 0.0141
0.5116 0,0092 0.6144 0.0109 0.8193 0.0144
0.5104 0.0094 0.6128 0.0112 0.8178 0.0148
0.5091 0.0096 0.6113 0.0114 0.8156 0.0151
0.5079 0.0097 0.6098 0.0117 0.8134 0.0155
0.5067 0.0099 0.6082 0.0119 0.8112 0.0156
0.5055 0.0101 0.6066 0.0121 0.8090 0.0162
0.5042 0.0103 0.6051 0.0124 0.8068 0.0165
638 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-Il: 639

Moment Capacity Of Rectangular Sections Moment Capacity Of Rectangular Sections


= 4.0 ksi

M^bd
640 CONCRETE STRUCTURES PART - 1 641
AUTHOR ZAHID A. SIDDIQ]

I
ApPe^^’^
'

Moment Capacity Of Slab Sections


Moment Capacity Of Slab Sections
/, = 280 MPa
/y = 300 MPa
fc' = 17 25 MPa fc' - 20.0 MPa = 25 0 MPs
Mu/bd Mu/bd2 M„/bd2 fc = 1 b wira *c

MPa P
MPa P P MJbd’ Mu/bdz Mu/bd2 P
MPa p P MPa
0.6150 0.0025 0.6170 MPa MPa
0.0025 0.6196 0 0025 - 0.6631 0.0025
0.6856 0.0028 0.7019 0 6577 0.0025 0.6601 0.0025
0.0029 0.7304 0.0030 0.7725 0.0029
0.7558 0.0031 0.7862 0 7270 0.0028 0.7436 0.0028
0.0032 0.8404 0 0034 0.UU33
0.8256 0.0034 0.0030 0.8266 0.0031 0.8812
0.8700 0.0036 0.9498 0 0039 0 7959 0.0038
0.8949 0.0037 0.0033 0.9091 0.0035 0.9893
0.9533 0.0039 1.0585 0.0043 0 8644 0.0042
0.9639 0.0036 0.9911 0.0038 1.0966
0.0040 1.0361 0.0043 1.1665 0.9325
1.0324 0.0048 0.0041 1.2033 0.0046
0.0043 1.1184 0.0046 1.2738 1 0002 0.0039 1.0727
1.1005 0.0052 0.0044 1.3094 0.0050
0.0046 1.2001 0.0050 1.3804 1 0675 0.0041 1.1537
1.1682 0.0057 0.0048 1.4147 0.0054
0.0049 1.2814 0.0053 1.4863 1 1344 0.0044 1.2342
1.2355 0 0061 0.0051 1.5194 0.0059
0.0052 1.3621 0.0057 1.5915 1 2008 0.0047 1.3142
1.3023 0.0055
0.0066 1.3937 0.0054 1.6234 0.0063
1.4423 0.0060 1.6961 0 0071 1 2669 0.0049 0.0067
1.3688 0.0057 0.0052 1.4727 0.0057 1.7267
1.5220 0.0064 1.7999 0.0075
t
1 3325 0.0071
1.4348 0.0060 0.0055 1.5511 0.0061 1.8293
1.6012 0.0067 1.9031 0.0080 1 3978 0.0076
1.5004 0.0063 0.0058 1.6291 0.0064 1.9313
1.6798 0.0071 2.0056 0 0084 1 4626 0.0080
1.5655 0.0066 0.0060 1.7066 0.0067 2.0325
1.7580 0.0074 2.1073 0 0089 1 5270 0.0084
1.6303 0.0069 1.7836 0.0070 2.1331
1.8356 0.0078 2.2084 1 5911 0.0063
1.6946 0.0093 0.0074 2.2331 0.0088
0.0072 1.9127 0.0081 2.3088 1 6547 0.0066 1.8601
1.7585 0 0098 0.0077 2.3323 0.0092
0.0075 1.9893 0.0085 2.4085 1 7179 0.0068 1.9361
1.8220 0.0102 2.0116 0.0080 2.4309 0.0097
0.0078 2.0654 0.0088 2.5076 1 7807 0.0071
1.8851 0.0107 2.0865 0.0083 2.5288 0.0101
0.0081 2.1410 0.0092 2.6059 1 8431 0.0074
1.9478 0.0084
00112 2.1610 0.0087 2.6260 0.0105
2.2161 0.0095 2.7035 1 9051 0.0077
2.0100 0.0116 2.2350 0.0090 2.7226 0.0109
0.0087 2.2906 0.0099 2.8005 1 9667 0.0079
2.0718 0 0121
2.3085 0.0093 2.8184 0.0113
0.0090 2.3646 2 0278 0.0082
2.1332 0.0093 2.4381
0.0102
0.0106
2.8967
2.9923
0 0125
0 0130
Illi 2 0886 0.0085 2:3814 0.0096
0.0100
2.9136
3.0082
0.0118
0.0122
2.1942 0.0096 2.5111 0.0110 2 1490 0.0087 2.4539
3.0872 0.0134 0.0103 3.1020 0.0126
2.2548 0.0099 2.5836 0.0113 2 2089 0.0090 2.5259
3.1814 0 0139 0.0106 3.1952 0.0130
2.3149 0.0102 2.6556 0.0117 2 2685 0.0093 2.5974
3.2749 0 0144 0.0109 3.2876 0.0135
2.3746 0.0105 2.7271 0.0120 2 3276 0.0096 2.6683
3.3677 0.0148 0.0113 3.3795 0.0139
2.4339 0.0108 2.7980 0.0124 2.3863 0.0098 2.7388
3.4598 0.0153 0.0116 3.4706 0.0143
2.4928 0.0111 2.8684 0.0127 2 4447 0.0101 2.8087
3.5512 0 0157 0.0119 3.5610 ’ 0.0147
2.5513 0.0114 2.9383 0.0131 2 5026 0.0104 2.8782
3.6420 0.0162 0.0122 3.6508 0.0151
2.6093 0.0117 3.0077 0.0134 2 5601 0.0106 2.9472
3.7320 0.0166 0.0126 3.7399 0.0156
2.6670 0.0119 3.0766 0.0136 26172 0.0109 3.0156
3.8214 0 0171 0.0129 3.8284 0.0160
2.7242 0.0122 3.1450 0.0141 2 6739 0.0112 3.0836
3.9101 0.0175 3.9161 0.0164
2.7810 0.0125 2 7302 0.0115 3.1510 0.0132
3.2128 0.0145 3.9980 0 0180 4.0032 0.0168
2.8373 0.0128 27861 0.0117 3.2180 0.0135
3.2802 0.0148 4.0853 0.0185 4.0896 0.0172
2.8933 0.0131 3.3470 28416 0.0120 3.2844 0.0139
0.0152 4.1719 0.0189 4.1753 0.0177
2.9488 0.0134 2 8966 0.0123 3.3504 0.0142
3.4133 0.0155 4.2578 0.0194 4.2603 0.0181
3.0039 0.0137 2 9513 0.0125 3.4158 0.0145
3.4791 0.0159 4.3431 0.0198 4.3447 0.0185
3.0586 0.0140 3.5444 3 0056 0.0128 3.4808 0.0148
0.0162 4.4276 0.0189
.3.1129 0.0143 | 3.6091 | 0.0166 | 4.5114 |
0.0203
0.0207 ill! 3.0594 0.0131
0.0134
3.5452
3.6091
0.0152
0,0155
4.4284
4.5114 0.0194
3.1129
642 soncp.ct structure RART-i 643
author, zahio a. siddiq. Appendix-H-

Moment Capacity Of Slab Sections Moment Capacity Of Slab Sections


/y = 350 MPa jf = 420 MPa
f„' = 1 7.25 MPa 0.0 MPa fc’ = 25.0 MPa
5.0 MPa f/ = 17.2!> MPa fo' = 20.0 MPa
Mj/bd2 Mu/bd2 Mu/bd2
MPa
P P
M„/bd2
P Mu/bd2 P
Mu/bd2 P P
MPa MPa MPa MPa
0.7640 MPa 0.0023
0.0025 0.7672 0.0025 0 7713 0.0025 0.8269 0.0023 0,8316
0,8301 0.8231 0.0023 0.0025
0.0027 0.8474 0.0028 0.8773 0.0029 0.9052 0.0025 0.9358
0.8957 0.8873 0.0024 0.0028
0.0029 0.9272 0.0030 0.9827 0.0032 0.9832 0.0027 1,0393
0.9610 0.9512 0.0026 0.0031
0.0032 1.0065 0.0033 1.0875 0.0036 1.0607 0.0029 1 1423
1.0260 1.0148 0.0028 0.0034
0.0034 1.0853 0.0036 1.1916 1.1377 0.0031 1.2446
0.0039 1.0780 0.0030
1.0905 0.0036 1.1636 0.0038 0.0034 1.3463 0.0037
1.2951 0.0043 1.1408 0.0032 1.2143
1.1547 0.0038 1.2415 0.0041 0.0036 1.4474 0.0040
1.3980 0.0046 1.2032 0.0033 1.2904
1.2185 0.0041 1.3189 0.0044 1.3661 0.0038 1.5478 0.0043 •
1.5001 0.0050 1.2653 0.0035
1.2819 0.0043 1.3958 0.0040 1.6476 0.0046
•0.0047 1.6017 0.0053 1.3271 0.0037 1.4413
1.3450 0.0045 1.4723 0.0049 0.0042 1.7468 0.0049
1.7026 0.0057 1.3884 0.0039 1.5161
1:4076 0.0047 1.5482 0.0052 0.0045 1,8454 0.0051
1 8029 0.0060 1.4495 0.0041 1.5904
1.4699 0.0050 1.6238 0.0055 0.0047 1.9433 0.0054
1;9025 0.0064 1.5101 0.0043 1.6642
1.5318 0.0052 1.6988 0.0057 0.0049 2.0406 0.0057
2.0014 0.0067 1.5704 0.0044 1.7376
1.5934 0.0054 1.7734 0.0060 0.0051 2,1373 0.0060
210998 0.0071 1.6304 0.0046 1.8106
1.6546 0.0056 1.8475 0.0063 0.0053 2.2334 0.0063
2.1975 0.0074 1.6899 0.0048 1.8831
1.7153 0.0059 1.9211 0.0Q65 0.0056 2.3288 0.0066
2.2945 0.0078 1.7492 0.0050 1.9551
1.7758 0.0061 1.9943 0.0068 0.0058 2.4236 0.0069
2.3909 0.0081 1.8080 0.0052 2.0267
1.8358 0.0063 2.0670 0.0071 0.0060 2.5178 0.0072
2.4867 0.0085 1.8665 0.0053 2.0979
1.8955 0.0065 2.1392 0.0073 0.0062 2.6113 0.0075
2.5818 0.0088 1.9247 0.0055 2.1686
0,0077
1.9547 0.0067 2.2109 0.0076 2.6762 0.0064 2.7043
0.0092 1.9825 0.0057 2.2388
2.0137 0.0070 2.2822 0.0067 2.7966 0.0080
0.0079 2.7701 0.0095 2.0399 0.0059 2.3086
2.0722 0.0072 2.3530 0.0082 0.0069 2.8883 0.0083
2.8633 0.0099 2.0969 0.0061 2.3779
2.1303 0.0074 2.4234 0.0084 0.0071 2.9793 0.0086
2.9558 00102 2.1536 0.0063 2.4468 0.0089
2.1881 0.0076 2.4932 0.0087 0.0073 3.0698
3.0477 0.0106 2.2100 0.0064 2.5152
0,0092
2.2455 0.0079 2.5627 0.0090 0.0075 3.1596
3.1389 00110 2.2660 0.0066 2.5832
2.3026 0.0081 2.6316 0.0092 0.0078 3.2488 0.0095
3.2295 0.0113 2.3216 0.0068 2.6507
2.3592 0.0083 2.7000 0.0095 0.0080 3.3373 0.0098
3.31 95 0.0117 2.3769 0.0070 2.7178
2.4155 0.0085 2.7680 0.0098 0.0082 3.4252 0.0101
3.4088 0.0120 2.4318 0.0072 2.7844
2.4714 0.0088 2.8356 0.0100 0.0084 3.5125 0.0104
3.4975 0.0124 2.4863 0.0074 2.8505
2.5269 0.0090 2.9026 0.0103 0.0086 3.5992 0.0106
3.5855 0.0127 2.5405 0.0075 2.9162 0.0109
2.5821 0.0092 2.9692 0,0106 3.6729 0.0131 0.0077 2.9815 0.0089 . 3.6853
2.6368 2.5944 3.7707 0.0112
0.0094 3.0353 0.0108 3.7597 0,0134 3.0463 0.0091
2.6912 2.6478 0.0079 0.0115
0.0097 3.1009 0.0111 3.8458 3.1107 0.0093 3.8555
2.7453 0.0138 2.7009 0.0081 0.0118
0.0099 3.1661 0.0114 3.9312 0.0141 3.1745 0.0095 3.9397
2.7537 0.0083 0.0121
2.7989 0.0101 3.2308 0.0117 0.0097 4.0232
4.0160 0.0145 2.8061 0.0084 3.2380
2.8522 0.0103 3.2950 0.0119 0.0100 4,1062 0.0124
4.1002 0.0148 2.8581 0.0086 3.3010
2.9051 0.0106 3.3588 0.0122 0.0102 4.1885 0.0127
4-1837 0.0152 2.9098 0.0088 3.3635
2.9576 0.0108 3.4221 0.0125 0.0104 4.2701 0.0130
4.2666 0.0155 2.9611 0.0090 3.4256
3.0097 0.0110 3.4849 0.0127 4.3489 0.0106 4.3512 0.0132
0.0159 3.0120 0.0092 3.4872
3.0615 0.0112 3.5473 0.0130 4.4305 3.5484 0.0108 4.4316 0.0135
0.0162 3.0626 0.0094
3.1129 0.0114 3.6091 0.0133 0,0111 4.5114 0.0138
4.5114 _ 0.0166 3.1129 0.0095 3.6091
644 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI Appendix-II: 645

Moment Capacity Of Slab Sections Moment Capacity Of Slab Sections


//=17.25 MPa // = 20 MPa

(MPa) (MPa)
M^bd2 Mu/bd2
646 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIqi Appendix-H: 647

Moment Capacity Of Slab Sections


Moment Capacity Of Slab Sections
/c' = 25 MPa
fy = 40 ksi
f? 2.5 ksi fc’ = 3.0 ksi fc* = 4.0 ksi
MJbd2 Mu/bdz P
Mu/bd2 P
P
ksi ksi ksi
0.0879 0.0025 0.0882 0.0025 0.0887 0.0025
0.0982 0.0028 0.1011 0.0029 0.1067 0.0030
0.1084 0.0031 0.1139 0.0032 0.1247 0.0035
0.1185 0.0034 0.1266 0.0036 0.1425 0.0041
0.1286 0.0037 0.1392 0.0040 0.1602 0.0046
0.1386 0.0040 0.1517 0.0044 0.1778 0.0051
0.1486 0.0043 0.1642 0.0047 0.1952 0.0056
0.1585 0.0046 0.1766 0.0051 0.2126 0.0061
0.1684 0.0049 0.1889 0.0055 0.2299 0.0066
0.1781 0.0052 0.2011 0.0059 0.2470 0.0072
0.1879 0.0055 0.2133 0.0062 0.2640 0.0077
0.1975 0.0058 0.2254 0.0066 0.2809 0.0082
0 2071 0.0061 0.2374 0.0070 0.2977 0.0087
0.2167 0.0064 0.2493 0.0073 0.3144 0.0092
0.2261 0.0067 0.2611 0.0077 0.3310 0.0098
0.2356 0.0070 0.2729 0.0081 0.3474 0.0103
0 2449 0.0073 0.2846 0.0085 0.3638 0.0108
0.2542 0.0076 0.2962 0.0088 0.3800 0.0113
0.2635 0.0079 0.3077 0.0092 0.3961 0.0118
0.2726 0.0082 0.3191 0.0096 0.4121 0.0123
0.2817 0.0085 0.3305 0.0100 0.4280 0.0129
0.2908 0.0088 0.3418 0.0103 0.4438 0.0134
0.2998 0.0091 0.3530 0.0107 0.4594 0.0139
0.3087 0.0094 0.3641 0.0111 0.4750 0.0144
0.3176 0.0097 0.3752 0.0115 0.4904 0.0149
0.3264 0.0100 0.3862 0.0118 0.5057 0.0155
0.3351 0.0103 0.3971 0.0122 0.5209 0.0160
0.3438 0.0106 0.4079 0.0126 0.5360 0.0165 '

0.3524 0.0109 0.4186 0.0129 0.5510 0.0170


0.3610 0.0112 0.4293 0.0133 0.5656 0.0175
0.3695 0.0115 0.4399 0.0137 0.5806 0.0180
0.3779 0.0118 0.4504 0.0141 ' 0.5952 0.0186
0.3863 0.0121 0.4608 0.0144 0.6097 0.0191
0.3946 ‘0.0124 0.4711 0 0148 0.6242 0.0196
0.4029 0.0127 0.4814 0.0152 0.6384 0.0201
0.0130 0.4916 0.0156 0.6526 0.0206
0.4111
0.4192 0.0133 0.5017 0.0159 0.6667 0.0212
0.4273 0.0136 0.5117 0.0163 0.6807 0.0217
0.4353 0.0139 0.5217 0.0167 0.6945 0.0222
0.4433 0.0142 0.5316 0.0170 0.7082 0.0227
0.4511 0.0145 0.5414 0.0174 0.7218 0.0232
648 CONCRETE STRUCTURES PART
-1 AUTHOR: ZAHID A. SIDDIQI

Moment Capacity Of Slab Sections

[1 V = 2.5 ksi
My/bd2 —: = 3.0 ksi I
~ —
f. - 4.0 ksi
ksi P
M,/bd2
ksi
I p M„/bd2
P
0.0023 0.1183 0.0023 I
0.1270 0.0024 0.1191 0.0023
0.1363 0.0026
0.1302 0.0025 I 0.1362 0.0026
0.1456 0.0028
0.1421 0.0027 I 0.1532 0.0029
0.1539 0.0030 0.1701
0.1548 0.0030 0.0032
0.1639 0.0032
0.1656 0.0032 I 0.1869 0.0036
0.1773 0.0034 1
0.1730 1 0.0034 0.2036 0.0039
0.1888 0.0037 0.2201
0.1821 0.0035 0.2004 0.0042
0.1911 0.0037
0.0039 I 0.2366 0.0046
0.2118 0.0041 I 0.2530
0.2000 0.0039 0.0049
0.2089 0.0041
0.2232 0.0044 I 0.2693 0.0052
0.2345 0.0046 0.2854
0.2177 0.0043 0.0056
0.2265 T 0.0045
0.2457 0.0048 I 0.3015 0.0059
0.2353 0.0047
0.2569 0.0051 I 0.3174 0.0062
0.2439 0.0048
0.2680 0.0053 I 0.3333 0.0066
0.2526 0.0050
0.2790 0.0055 I 0.3490 0.0069
0.2900 0.0058 I 0.3647
0.2611 0.0052 0.0072
0.2697
0.3009 0.0060 I 0.3802 0.0075
0:0054 0.3117 0.0062 I 0.3956
0.2781 0.0056 0.0079
0.2865
0.3225 0.0065 I 0.4110 0.0082
0.0058 0.3331 0.0067
0.2949 0.0060 0.4262 0.0085
0 3032
0:3437 0.0069 0.4413
0.0061 0.0089
0.3115
0.3543 0.0072 I 0.4563
0.0063 0.3648 0.0092
0.3197 0.0074 1 0.4712 0.0095
0.0065 0.3752 0.0076
0.3278 0.0067 0.4861 0.0099
0.3359
0.3855 0.0079 I 0.5008 0.0102
0.0069 0.3958
0.3440 0.0081 i 0.5154 0.0105
0.0071
0.3520
0.4060 0.0083 I 0.5298 0.0109
0.0073 0.4161 0.0086 I 0.5442
0.3599 0:0074 0.4261 0.0112
0.3678 0:0076 0.0088 I 0.5585 0.0115
0.4361 0:0090 I 0.5727
0.3756 0.0078 0.4460 0.0118
0.3834 0.0093 0.5868 0.0122
00080 0.4559 0.0095
0.3912 0.0082 0.6008 0.0125
0.4657 0.0097 0.6146
0.3988 0.0084 0.4754 0.0128
0.4065 0.0086
O.O1OO I O.G2G4 0.0132
0-4850 0.0102 I
0.4141 0.0087 0.6421 0.0135
0.4216 0.0089
0.4946 0.0104 I 0.6556 0.0138
0.5041 0.0107 0.6691
0.4290 0.0091 0.0142
0.5135 0.0109 1
0.4365 0.0093 0.6824 0,0145
0.4438 0.0095
0.5229 0.0111 I 0.6957 0.0148
0.5322 0-0114 0.7088
0.4511 | 0.0097 | 0.0152
0.5414 , ,0.0116 | Q.7218
0.0155
650 CONCRETE STRUCTURES PART - 1 651
AUTHOR: ZAHID A. SIDDIQI Appendix-II:
Moment Capacity Of Slab Sections
Moment Capacity Of Slab Sections
ft = 4.0 ksi

MJbd
652 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI Appendix-Ill: 653
Space For Reader’s Notes

Appendix - m

Column Interaction Curves


654 CONCRETE STRUCTURES PART - I
AUTHOR: ZAHID A. SIDDIQI
Space For Reader’s Notes
MPa
Pu/Ag,
659
^pendix-®
66
Appendix-Ill:

Mu/Agh,MPa
f
663
•Appendix-Ill’.
!

COLUMN INTERACTION CURVE


fy = 300 MPa // = 25 MPa / =
0.6

Mu/Agh, MPa
664 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-III:

Mu/Agh, MPa
Appendix-Ill: 669
668 -
CONCRETE STRUCTURES PART 1 AUTHOR: ZAHID A. SIDDIQI
Appendix-HE 671
670 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

COLUMN INTERACTION CURVE


= 420 MPa fc' = 17 MPa y = 0.75

11

7 10

It
2

0 P"1"1
0 0.5 1.5 5
Mu/Agh , MPa
672 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDlyi
Appendix-Ill: 675

COLUMN INTERACTION CURVE


= 420 MPa // = 20 MPa / - 0.6
676 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDh
677
Appendix-Ill:

COLUMN INTERACTION CURVE


b
= 420 MPa /c'=25MPa

25
^ /=0.9 -
Y*

Mu/Agh, MPa
678 CONCRETE STRUCTURES PART - 1 679
AUTHOR. ZAHID A. SIDDlyi
Appendix-HI.

COLUMN INTERACTION CURVE


fy — 420 MPa // = 25 MPa X~

Mu/Agh, MPa
680 CONCRETE STRUCTURES PART -I
AUTHOR: ZAHID A. SIDDjqj

Pu/Ag,MPa
682 CONCRETE STRUCTURES PART - 1 Appcndix-HI: 683
AUTHOR. ZAH.0 A. S1DD,W
Appendix-Ill: 685

Mu/Agh, MPa

A
Appendix-Ill: 687

Mu/Agh
Mu/Agh
689
Appendix-ill:

I
690 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-Ill: 691

8i

Mu/Agh
694 CONCRETE STRUCTURES PART -I 695
AUTHOR: ZAHTOA. SJDDIqj Appendix-Ill:

COLUMN INTERACTION CURVE


= 460 MPa /c' = 35 MPa / - 0.75

2
Mu/Agh
696 concrete structures part
-i
AUTHOR: ZAHID A. SIDoiqj Appendix-Ill: 697
698 CONCRETE STRUCTURES PART -I
AUTHOR: ZAHID A. SIDDTqj
Appendix-Ill: 701
700 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
702 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

Mu/Agh
704 con^stkuciumspart-t author.- zahid a. SIDrai)I Appendix-HI:
706 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDIQI 707
Appendix-Ill:
708 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIyi
Appendix-Ill: 709

COLUMN INTERACTION CURVE


CIRCULAR
fy = 420 MPa f/ = 25 MPa y = 0.9
Appendix-Ill:

COLUMN INTERACTION CURVE


= 40 ksi = 3.0 ksi /=0.75

Pu/A
7 1 2 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI 1
715
Appendix-Ill:
714 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

COLUMN INTERACTION CURVE


= 60 ksi /c'=3.0ksi /-0.6
1
716 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
718 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

COLUMN INTERACTION CURVE


fy = 75 ksi f'~ 3-0 ksi / = 0.6

3.5

Appendix - IV
Simplified Wall and its Footing Design

Mu/Ash
720 CONCRETE STRUCTURES PART - 1
AUTHOR: ZAHID A. SIDDiqj Appendix-IV: 721
Space For Reader’s Notes
BRICK WALLS AND THEIR FOUNDATIONS

Crushing Strength of Bricks in Flat Position

Crushing strength range = 4.3 - 19.3 MPa


(0.62 - 2.80 ksi)

Punjab bricks have 10.7 MPa (1.56 ksi) crushing strength. The strength decreases when soaked
in water by 25%. Crushing strength of brick masonry is only about one-third to one-fourth or
less of the crushing strength of a single brick and depends upon the mortar used.

Allowable Stress for Brick Masonry

Item Fa(MP„) Fa (ksi)


Brickwork in (1:3) C/S mortar 0.97 0.140
Brickwork in (1:4) C/S mortar 0.86 0.125
Brickwork in (1:6) C/S mortar 0.54 0.080
Brickwork in mud 0.16 0.025
Block masonry in 1:3 C/S mortar
3.5 MPa (500 psi) block crushing strength 0.27 0.040
6.9 MPa (1000 psi) block crushing strength 0.65 0.095
13.8 MPa (2000 psi) block crushing strength 1.02 0.150
Solid cement concrete block 1.61 0.230

Height of Walls
The height of a wall is measured from top of the plinth to the highest point of the wall
excluding any parapet. Storey height is the center-to-center height between lateral supports.

Effective Height (H)


s 1. For walls without lateral support at top = 1.05 x actual height
2. For walls with lateral support at top = 0.75 x storey height

Slenderness Ratio for Walls


It is the ratio of the effective height to the effective thickness.

Upper Limits of Slenderness Ratio

Slenderness ratio should not usually exceed 18 for walls built in cement mortar. For residential
buildings of not more than two stories, this value may be increased to 24 for lean C/S mortars
and 32 for 1:3 C/S mortar.
722 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQJ Appendix-IV: 723

Strength Reduction Due to Slenderness Effects


Wall Foundation
following bearing
Considering reduction of loads on various walls, self weight of walls and the
Slenderness Ratio Reduction Factor capacities, the wall foundation size is estimated below:-
1 1.00 vm/™2
2 0.96 Safe bearing capacity =
or 1.7 to 2.8 ksf
4 0.88
Maximum free wall height = 4.5m or 15ft.
6 0.80
8 0.70
10 0.60 SI Units
12 0.50 (rounded to higher whole number with a
14 Fi = l2N3tw+ZL
0.40 minimum of 15)
16 0.35 1050
18 0.30
21 0.25 Where Ns ~ number of stories and tM = wall thickness (mm)
24 0.20 XL is defined as before.
Bottom width of footing Wf~F2x 50 (mm)
Nominal brick size
Size = 228x 114x75 mm Let D = depth of P.C.C. under wall foundation
.
or 9 x 4.5 x 3 in. and W = 2D for inner walls or D for exterior walls.
(N) = WF-W-t# (rounded to higher
Thickness of wall No. of steps in spread footing
114 whole number)
Considering reduction factor due to wall slenderness, distribution of load on various
walls and
self weight of walls, the thickness can be found as follows:- Depth of steps should be multiples of 75 mm.
7=1.0 for inner walls and /= 1.75 for outer walls FPS Units
SI Units
(rounded to higher whole number with a
F2 = N^t^+^L minimum of 10)
Fl = Ix£L (rounded to higher whole number)
4600
Where N, = no. of stories and tw = wall thickness (in)
Where XL is the total width of floor supported by the wall in mm.
< XL is defined as before.
Then, wall thickness t„= Fi x 1 1 4 (mm)
Z Bottom width of footing = F2 x 0.25 ft.
FPS Units
Let D- depth of P.C.C. under wall foundation (in)
7j — Ix£L (rounded to higher whole number) and W= 2D for inner walls or D for exterior walls.
15 (rounded to higher
Where SL is the total width of floor supported by the wall in ft. No. of steps in spread footing (N) =
45 whole number)
-
Then, wall thickness tw F| x4.5 (in)
Depth of steps should be multiples of 3 in.
724 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
Appendix-lV: 725

Franmle -2: Design wall-A and its footing.

I =
SZ =
1.75
2x
x 000 _ 6100 mm.

=
2 x 15.75/2 + 10.75/2 + 5.375
26.5 ft.
2
Fi = 26.5/15 = 1.77 »2.0
2^ 1.75x6100
2.32 » 3 Wall thickness (tw) = 2 x 4.5 = 9 in.
4600
= Fj x 114 342 mm 2x9+265
F2 = 3.5 '
= 35
Foundation
F1 =
1050
= 12x2x342+6100 _ 13.63 »15
Width of foundation (WF)
Let D = 6 in.
= 13x0.25 = 3' -3"

1050
- Fz x 50 = 750 mm JV =
325x12 - 2x6 - 9 = .
4
45
Let D = 100mm .-. w = 100
N = 750—100—342
= 2.70 «3
114
726 CONCRETE STRUCTURES PART - 1 AUTHOR; ZAHID A. SIDDIQI Append ix-IV: 727

SIZE OF COLUMNS AND THEIR FOOTINGS


First floor area = 15x12 ft.
Size of RC column in SI Units Second floor area = 15 x 12 ft.
Roof area =8x12 ft.
column)
Area (mm2) F2xl500x FjxS^^ 85
= 456 ft2, Ns = 3, tw = 9 in, Fx = 1.0, Fi = 1.0 (interior
Fortw = 9 in.
where tw - wall thickness in mm, A - area of a floor slab supported by column, tn2,
Area of column = 021 x 1.0 x 1.0 x
Fi = 1.0 for slabs lesser than 150 mm thick, 1.15 upto 175 mm thick and
1.3 for
greater thicknesses, = 142.4 in2
Fi = 1.0 for interior column, 2.0 for edge column and 4.0 for comer
column. A 12" x 12" column is sufficient.
Area of Footing in SI Units Fortw = 4.5 in.
a of footing
Area

Notes:
+

*
-
8.5
*
—V
(N-l) [XA
ZA f+—
720
Area of column 021 x 1.0 x

= 119.1 in2
1. Area of a storey for footing design to be added in A 10" x 12" column is sufficient.
future may be reduced to A/(NS + I)
where Ns is the no. of stories to be constructed at present. This is
based on the assumption
that bearing capacity will improve with time. However, for design
of footing, the increased
soil pressure (or improved bearing capacity) should be used. 66.70 ft2
=
2. Areas of columns and footings may be multiplied by
(0.97)N,-i to consider reduced Provide 9 x 7'-6" footing.
probable intensity of live load. data:-
3. Architectural requirements must also be considered while
Example: Find size of column and its footing for the following
deciding the area and shape of
2 stories now and one storey to be constructed a year
later.
columns. Ns =
Size of RC Column in FPS Units TA = 60 m2, tw = 114 mm, Fi = 1.0, Fi = 1.0 (interior column)

Area (in2) Area of column = — rr"4


2 [60
1.0xl500x 1.0x 60+ 85 x

= 107,994 mm2
where = wall thickness in inches, A = area of floor slab supported by
column ft2,
Fi = 1.0 for slabs lesser than 6 in thick, 1.15 upto 7 in thick and 1.3 Provide 300 x 375 mm column.
for greater = 47 m2
thicknesses, TA for footing = 40 + 20 x
Fi = 1.0 for interior column, 2.0 for edge column and 4.0 for comer column. 47 2-1 I4!
Area of Footing in FPS Units Area of footing = K5 +
720 V 2
= 5.536 m2
2
^f(ft2) = 172 XA +
4 Use 3m x 2m size for the footing.

Example: Find size of a column and its footing for second wall case supporting
following
roof areas with 9 in or 4.5 in walls on column center-lines.
728 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI

Strip Column Footing


Assuming that length of strip footing is approximately equal to the length of floors supported
and considering safe bearing capacity equal to 80 kN/m2 (1.7 ksf) the width of strip-footing
may be estimated as follows:
SI Units
Let tw = thickness of walls, Ns = no. of storeys,
W = width of floor supported by the column-line per storey, m. This can be reduced for
any future storey as for single column.
iff = width of footing in m.

- 2 (rounded to higher multiples of 0.05 m)


X *X +
17 1000
FPS Units
Let tw = thickness of walls, Ns = no. of stories,
W = width, of floor supported by the column-line per storey, ft. This can be reduced for
any future storey as for single column. Selected formulas
Wi = width of footing in ft.

Ax^x^+l(^-l)/w (rounded to higher multiples of 0.25ft.)

Example; Find width of strip footing for the given data:-


ST
rw = 228mm, N=3, ^=4.5m.


Wr = 17- x 3 x 4.5 +
1000
= 2.044 mm w 2.05m

FPS
tw = 9 in., N = 3, Width of floor supported per storey W= 15 ft.
2 1
= x 3 x I5 + x2x9 = 6.93 ft « 7 ft.
730 CONCRETE STRUCTURES
PART -I
AUTHOR: ZAHID A. SIDDIQI Appendix-V:
SftaceFpr Reader’s Not^ 73 J
1. The coefficient of linear expansion for steel is 11.7xl0-6 per degree centigrade and for
concrete it is 9.9xl0’6 per degree centigrade.
2. The live load (L) for this member may be reduced from its original value (Za) when the
influence area (Jj) is greater than 40 m2 by the following expression:
46 i
L = Lo (0.25 + -== ) < Lo SI Units, Ai in m2
A
3- Important floor slabs must also be checked for concentrated loads of magnitude 10 to 35 kN
(2 to 8 kips) spread over an area of 0.75x0.75 m2 (30x30 in2) to represent heavy items.
4.
a. U = 1.4 D
b. U = 1.2D + 1.6L + 0.5 (Zr or S or £)
c. U = 1.2D + 1.6(Z,.orSor£)
+ (1.0Zror0.5 (?)
d. U = 1.2 Z> 1.0 W+ 1.0 Z 0.5 (Lr or S or R)
+ +
e. U= 1.2 D + 1.0 £+1.0 Z +0.2S
f. U = 0.9 Z)+ 1.0 W
g. U — 0.9D+1.0E
5. For concrete that can dry completely and where the shrinkage is unrestrained, the linear
coefficient of shortening is approximately 0.00025 at 28 days and 0.00035 at 3 months. After
this the shrinkage change is less rapid and may approach a maximum of 0.0005 at the end of
12 months.
6. Approximate shrinkage in 70 years in normal weight structural concrete having cylinder
strength of 20 to 50 MPa for dry atmospheric conditions with relative humidity of about 50%
is as follows:
For effective thickness of 150 mm, shrinkage strain = 0.00056
For effective thickness of 610 mm, shrinkage strain = 0.00047
7. Creep strain for 1:2:4 concrete loaded at 28 days with a sustained stress of 4 MPa is 0.0003
at 28 days after loading and 0.0006 at one year time interval.
-
'

8. : , .

Specified compressive strength Required average compressive strength


//,MPa £,MPa
<21 X' + 7.0 MPa
21 to 35 + 8.3 MPa
>35 1.10/; + 5.0 MPa
9. If statistical data is available and /.' < 35.0 MPa then is larger of the following:
~ fc +
^34^ and = f’ + 2.33 s, - 3. 5 MPa
where is the sample standard deviation of at least 30 consecutive tests.
Appendix-V: 733
732 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. SIDDIQI
10. If statistical data is available and f'c is greater than 35.0 MPa (5000 psi) 24. Ai =
then is to be l + 50p'
taken larger of the following:
fL = f'c + 1-34^ = 0.90/; + 2.33^ Elapsed Time £
5 years or more 2.0
11. Ec(MPa) = 0.043 £c(MPa) = 4700 777 (SI) 1.4
12 months
12. Mean split cylinder strength is approximately 6 months 1.2
0.53777 MPa and mean modulus of rupture 3 months 1.0
is approximately 0.69 777 MPa. ratio of depth of equivalent rectangular stress block to the depth of N.A.
25. pt =
13. ACI uses modulus of rupture equal to 0.62A.
777 MPa in calculating deflections and = 0.85 for // 28 MPa
O.5OA.777 MPa indirectly in strength calculations. According to ACI Commentary 10.2.5, the (decreases by 0.05 for every 7 MPa in excess of 28 MPa)
modulus of rupture is approximately 10 to 15 % of the concrete compressive strength. 0.85 for //> 28 MPa (SI Units)
= 1.05-0.00714 //
Il 14. For design, the tensile strength of concrete is generally taken varying from 0.25
777 MPa 26.
Minimum Depth of Beams and One-Way Slabs
to 0.42 777
MPa.

15. k
— -j(pn')2 + 2pn — pn Member Steel Grade Simply
Supported
One End
Continuous
Both Ends
Continuous
Cantilever

16. Mr - Asfsjd or - /=
M
AJd Solid One- 280 or 300 //25 //30 //35 //12.5
2M Way Slabs
17. Mr = kj bd2 420 //20 //24 Z /28 //10
kjbd2
M 280 or 300 //20 //23 //26 //10
18- d™, =

Beams 420 t /16 //18.5 / /21 Z /8


19. For k = 0.333, j = 0.889, dmb, = / — = /UK
V0.148/c^ \f’b 520 / /14 / /16 / /18.4 tn
20. M
Notes: 1. / = the effective span of the beam or the slab.
21. -4s.mw '77~b»d - J^b*d (SI Units)
2. For structural lightweight concretes having unit weights in the range 1440
kgs/m3, the above values shall be multiplied by (1.65 0.0003 — wc) but not less
1840
than

1.09, where wc is the unit weight of concrete in kgs/m3.
Jy J y
For a statically determinate part of a T-beam having its flange in tension, we have
3. Values other than for / = 420 are obtained by multiplying the values for / =
420
ds.mii ~bvd (SI Units) MPa by (0.4 +/y/700).
lJy
nfc
27. & = 0.85 pi -—
fy fy+^
(SI Units)
22. k

23-
+ f.

= 0.85/7, x-A^o.^ (SI and US Customary units)


28. Members with ties = 0.65 + ————L0.003P—
0.005-^ lc J
l]-s,
The value must be between 0.65 and 0.90.
J
^fy *f,
734 CONCRETE STRUCTURES P ART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-V: 735
29. A,mm 600 -/ 3
(SI Units) 45. d Id (SI)
600 *8
The second expression is critical when// 31.4 MPa. 46. The effective flange width of T-beams is the minimum of the following:
_
m
JU. (D — —-—
0-85/;
fy
and R
—_ — M
bd
n = © fI -
I V
I 2.6142?
1
fl
when Q > 0.005
i) £ 1 4 where £ = span of the beam
ii) 16ftf + iw
iii) 5 = bw + one-half the clear distance to the next web on both sides
47. The effective flange width of L-beams is the minimum of the following:
31 - dmin for/ '<28 MPa
0.205/' h i) £ 1 12 where £ = span of the beam
ii) Oht+bw
f 0.85/' I 7 , iii) b^ + SJl where 5C = clear distance to the next web
\GM3Espj a^ad-^d1
32- = 0, M = 0.85fc'ba (d-ah)
{A,-A^fy
48. a
33.
280
smax = 380 f - I
Js ) \
— - 2.5 cc < 300
i

\ Js
280 l
J
It is permitted to take/ = 2/3/ in place of
49. Mt
0.85 fl b„
= (4; - Af)/ (d- a 1 2) + Af/ (^-/i hf! 2)
detailed calculations.
50. M = AJy (d - a/2) - Axfy (d'~ a/2) + 0.85/' fl hf (b - b^ (a/2 2)
34. Nb — —
0.02 bw 1 .4 rounded to lower whole number (SI units)
51. Pw.max = /7roax for fib alone + /?f
= T1-— 1/
35. Equivalent slab width for interior longer beam 7?2

36. Equivalent slab width for exterior shorter beam = % / + cantilever width, if any 53. Mt = A^fy (d - hr / 2)
37. UDL wall load on the beam = 0.019 H kN/m (SI) ^fy
38. Equivalent UDL wall load on lintel for bending moment = 0.011 kN/m (SI) 0-85

3,. a = 0.85 fcb a


= 0.003^
a
55. Moment Coefficients for Slabs Having Spans Lesser Than 3.0 m OR Beams Having
Ratio of Sum of Column Stiffness to Beam Stiffness More Than 8 at Each End of the
40. M = 0.85 fc'b a (d- a/2) +As'f/(d-d') or (4r- A/) fy(d-a/2)+A/f/(d-d') Span.

41. / = 600^^
a
and // = 600^^
a
1. Negative moments at all supports, integrally built with beams.
— w„42
12
u n
'

42. pb = Pb + P' Pb = pb + p' x ~-


2. Positive moment in end panel.
— wu/2
14

43. Pm^ = Pi^ + p'


J y
3. Positive moment in central panels.
— WUV
16

56. (SI)
44. p„
* ~ 0.85&—A
/ d ^\600-/ J ———
1+ p' P (SI)
V
57. Ve= OAl^^b^d (SI)
737
Appendix-V:
736 CONCRETE STRUCTURES PART -I AUTHOR: ZAHID A. S1DDIQ]
61. Smar = smaller of the following three for SI units:
58. 4,^=0.062757^- 0.35^r
Jyt
.
J
(SI)
i) hi) 600 mm: 0.35 is replaced with 0.0624^ if^'> 31.9 MPa
0 356
59. Moment and Shear Values for Beams and Slabs Having Spans Greater Than 3.0 m. by the second
When required F, exceeds Q33^f~b„d
(N), the maximum spacing as given
1. Positive Moment and third of above conditions must be
halved.

End spans: 62. smax


$vAJd
=- 1

—cosax
If discontinuous end is unrestrained — wu42
11 „ „
63. Ks =
44,(sin a +
- 4A =
VstanO

If discontinuous end is integral with the support



14
w 2

64. Kc =
Interior spans:

16
Wu^n2
65. ^Fs = Va - &VC = 4
444 (cot 6 + cot a)sin a
s
2. Negative moment at exterior face of first interior support Concrete).
One-Way Slabs (Normal Weight
66. Minimum Thickness of Non-Prestressed
Two spans:
9
Wu^2 Steel Grade
A
More than two spans:
— wn42
10
End Conditions
Simply supported
280 & 300
Z/25
420
~£/20 £/ 20 x (0.4 Vy/
690)
3. Negative moment at other faces of interior supports

(4 in no. 3 is the average of clear spans of the two adjacent panels.}


— h42
11 Onp „nd continuous
Rnth ends continuous
Z/30
£ / 35
£/24^ / / 24 x (0.4 +/v / 690)
//28 / / 28 x (0.4 +4 /
690)

Cantilever / 12 // 10 _ l/10x(0.4 +fy/690)J


4. Negative moment at interior faces of exterior supports for
members built Integrally with their supports:
The support is a spandrel beam or girder:
24
67.
The support is a column: ^,^n2 Clear Cover (mm / in)
Fire Rating (Hours)
The support is not monolithic: Zero
20 3/< l’/4
w /
5. Shear in end members at first interior support
2 25 1
3
6. Shear at all other supports 40 V/2
60. If required shear to be resisted by the transverse steel (Pi) exceeds 0.66^4
b„d (N), the
„280
cross-sectional dimensions must be increased. However, in most practical cases, required K „280 300— (SI)
68. Srnax = 380—— -2.5ce < f.
should not be more than Q33-/f^bwd (N) . J Ji
738 CONCRETE STRUCTURES P ART - 1 .
AUTHOR:
rTTUrm
ZAHID A. SIDDIQJ 739
69. Appendix-V;
Dlstnbution, Temperature And Shrinkage Steel in Slabs. 71.
Fraction Of Distribution, Temperature ACI 1963 Coefficients For Live Load Positive Moments In Slabs
psi And Shrinkage Steel With Respect To Increased by 25%.
Gross Concrete Area rCase
-1S0&3O0 ] 40,000 ~
0,0020
Case
1
Case
2
Case
3
Case
4
Case
5
Case
6
Case
7
Case
8 9
W00 Ratio
~~ 0.0018 m
420 >60,000 0.0018x420
" " > 0.0014 (SI) 0.040 0.044 0.040 0.035 0.038
j- Cx 0.045 0.034 0.034 0.040
1.00 C, 0.045 0.034 0.040 0.040 0.034 0.040 0.044 0.038 0.035
0.0018x60,000 0.039 0.044 0.043 0.048 0.045 0.039 0.040
>0.0014 (FPS) (J 0.050 0.038
J V
0.95 u 0.041 0.031 0.036 0.036 0.030 0.036 0.040 0.034 0 031
0.056 0.043 0.044 0.049 0.046 0.053 0.050 0.044 0.045
70. 0.90 0.028 0.034 0.033 0.026 0.031 0.036 0.030 0.028
0.036
ACI 1963 Coefficients For Dead Load Positive Cx 0.063 0.046 0.050 0.054 0.051 0.058 0.056 0.050 0.049
Moments In Slabs 0.85
Increased by 25%. c. 0.033 0.024 0.030 0.029 0.024 0.028 0.033
0.064
0.028
0.055
0.025
0.053
Cx 0.070 0.051 0.056 0.060 0.055 0.064
Case 0.80 0.021 0.028 0.025 0.020 0.024 0.029 0.024 0.021
Ratio
Case Case Case Case Case Case Case Case CY 0.029
0.061 0.058
1 2 3 4 5 Cx 0.076 0.056 0.064 0.065 0.059 0.069 0.070
6 7 8 9 0.75
m Cy 0.024 0.018 0.024 0.020 0.016 0.020 0.025 0.020 0.016
1
wZ&w Cx 0.085 0.061 0.071 0.071 0.064 0.075 0.079 0.068 0.063
Lx 0.045 0.023 0.023 0.70 0.015 0.020 0.018 0.014 0.016 0.021 0.018 0.014
LOO 0.034 0.034 0.041 0.034 0.025 0.029
Cy 0.020
0.045 0.023 0 034 0.034 0.023 Cx 0.093 0.066 0.080 0.078 0.069 0.080 0.088 0.074 0.068
0.034 0.038
0.95 cx 0.050 0.025 0.026 0.038 0.035 0.045 0.039
0.029
0.028
0.025
0.030
0.65
Cv 0.016 0.013 0.018 0.014 0.011 0.013 0.018 0.014 0.011
Cy 0.041 0.020 0.031 0.030 0.019 J 0.101 0.073 0.089 0.084 0.074 0.085 0.096 0.081 0.074
0.030 0.039
0.90 cx 0.056 0.028 0.031 0.041 0.036 1
0 026 A APT 0.60 <J 0.013 0.009 0.014 0.011 0.009 0.010 0.014 0.01 1 0.009

0.85
—CY cx
0.036
0.063 0.030
0.033 0.015
0.018 0.030
0.036
0.028 0.016
0.045 0.039
0.049 0.044
0.026 0.035
0.053 0.050
0.031
0024
0.036
0.033
a mo
0.035
0.55 Cx
Cv
0.110
0.010
0.078
0.008
0.083
0.100
0.011
0.110
0.090
0.009
0.096
0.079
0.006
0.084
0.091
0.008
0.098
0.106
0.011
0.115
0.088
0.009
0.095
0.079
0.008.
0.084
0.028 0.024 0.014 0.021 0.031 0 021 n AM 0.5 Cx 0.119
0.80 ^x 0.070 0.033 0043 0.049 0.040 0.056 0.056 CY 0.008 0.005 0.009 0.006 0.005 0.006 0.009 0.063 0.005
cv 0.029 0.014 0.025 0.020 0.011 J
0.019 0 028
0.040
0 019
0.036
A AT Q
0.75 Cx ^0.076 0.035 r0.050 0.054 0.041 0.060 0.064 0.045 72.
Cv 0.024 0.011 n 0.039
.
0.016 0.009 0.015 0.025 0.016 0.009 ACI 1963 Coefficients For Negative Moments In Slabs Decreased
0.70 0.085 0.038 0.058 0.058 0.044 0.064 0.073 by 10%.
0.050 0.04]
CY 0.020 0.009 0.020 0.014 0.006 0.011 0.021 0 014 A Ans
C* 0.093 0.040 Case


0.65 0.068 0.063 0.045 0.068 0.081 Case Case Case Case Case Case Case Case
Cy 0.016 0.008 0.055 0.043 9
0.018 0.011 0.005 0.009 0.018 Ratio 1 2 3 4 5 6 7 8
Cx 0.101 0.043 0 01 1 A AA£
0.60
A 0.013 0.005
0.078
0.014
0.066 0.046 0.070 0.091
0.009 0.004 0.008 0.015
0.060
0 009
0.045 m ^Z^ZZI
0.55 c* 0.110 0.044 0.089 0.070 0.048 0.073 0.101 Cx — 0.041 0.045 0.068 0.064 0.030 0.055
& 0.065 0.046 LOO
cx
0.010 0.004 0.011 0.006 0.003 0.005 0.011 0 006 A AA/t c. 0.041 0.068 0.045 0.064 0.055 0.030
0.5 0.119 0.046 0.100 0.074 0.049 0.076 0.1 11 — 0.045 0.050 0.071 0.068 0.034 0.059

_
0.008 | 0.003 0.070 0.048 0.95
CY 0.009 0.005 0.001 0.004 0.009 0.005 0.003
0.037 0.065 0.041 0.060 0.050 0.026
Cx 0.050 0.054 0.072 0.071 0.039 0.061
0.90
cv 0.033 0.063 0.036 0.056 0.047 0.023
0.85
u„ M — 0.054 0.059 0.074 0.075 0.044 0.065
u 0.028 0.059 0.031 0.051 0.041 0.019
740 CONCRETE STRUCTURES P ART - 1 AUTHOR: ZAHID A. SIDDIQI 741
Appendix-V:
Case Case Case Case Case Case Case Case Case
Ratio 1 2 3 4 5 6 7 8 9 80.
m 4.= 1 sinp for/ = 0
0.80 cx — 0.059
0.024

0.055
0.064
0.026
— — ——
0.075 0.077
0.046
0.050
0.037
0.068
0.015 Kp
+
=
1- sin
Pk = KPyykfy-yk/^
0.75 cx — 0.062 0.068 0.077 0.079 0.055 0.070
_W
Cy — 0.020 0.050 0.022 — 0.040 0.032 0.013 W GM,
0.70 Cx
cy


0.067
0.015

0.045
0.073
0.017
0.077

0.082


0.034
0.061
0.026
0.073
0.010
81. Olnax = B B2
0.65 Ct
Cy
..

.


0.069
0.013

0.039
0.077
0.014
0.078
— — 0.084

.

0.028
0.067
0.022
0-075
0.007

— ——
Cx .
— 0.073 0.080 0.079 0.086 0.072 0.077
0.60
cy
Ct
—— 0.009
0.076
0.032 0.010
0.083

0.080 0.086
0.022 0.016
0.077
0.005
0.077 = seismic inertial angle — tan
0.55 Cy . — 0.006 0.025 0.007
— — 0.017 0.013 0.005
83. V

73.
0.5

Zimm =
Ct
cv


£„(0.8 +/,/1400)
0.077
0.005 0.020

not less than 90 mm


0.085
0.005
0.081

0.087
1
0.013
0.080
0.009
0.079
0.003 84. Ae
cos2(^~ y ^) — —
I sin(^ + £) sin(^ -y~P)
cosV cos2 0 cos(y/ + 0 + d) 1 + cos(/-0)cos(y/ + 0+£)
“36m + 9
cos1 + #)
85. Kpe = r [sin(^ + £) sin(^ - + /)
cosy, cos2 0 cos^ - 0 + 3) 1 - J^(^)coS(v,^+5)
75. ^>Pno - 0.85 x 0.75 [0.85/^ (4 -At) P/4tI for columns with spirals
no = 0.80 x 0.65 [0.85 f'c (Ag-A^ for tied columns
sin 2 (a + ^) cos <5
86. Ka 2
76. (ApUn = o.45|^-l1A I sin(^ + 5)sin(?i- ^)
)fyt sin«sin(«-5) +
'
77
•/ 5

0.45^ /'(4 /4,-1) 87. ?„■=


the above expressions are simplified to:
78. Ag (trial) >
P+2M + 2Mm
- for tied column (SI) 88. Ift-0. #-O,^-Oand f)-(>(ora=90"), then
0.43/' + 0.008/, cos2 (^ -y^)
cos\4l^2____
T A* . ~1~ ।

and Ag (trial)
P + 2M + 2M
~ -
0.5/; + 0.01/, — for spiral
E column (SI) , W F, lsin^sin(^ - ^)
cos 1+,
L )[ COS(/
cos2 </
jsin^sin^JP)
COSl/

to >r,= 89. FP = 4.0Ca/p Wp


cos/7 + ^cos2 /7-cos2 90. Upper One-Third Of Wall
load, 0.0954 for uniform load, 0.0858 for
inverted
Kk at support = 0.0323 for triangular
triangular load (tension on fill side) inverted
load, 0.0488 for uniform load, 0.0436 for
at midspan = 0.0166 for triangular
triangular load (tension opposite to fill side).
742 CONCRETE STRUCTURES P ART - 1 AUTHOR: ZAHID A. SIDDIQI Appendix-V: 743
Kv span = 0.00326 for triangular load, 0.00968 for uniform load, 0.00753 for inverted
triangular load (tension opposite to fill side). 1
91. Middle-Third Of Wall
Kb at support = 0.0545 for triangular load, 0.0908 for uniform toad, 0.0627 for inverted
decided in the following
triangular toad (tension on fill side) 101. For constant width combined footing, the dimensions may be
Kh at midspan = 0.0268 for triangular load, 0.0468 for uniform toad, 0.0323 for inverted sequence:
Xvspan
triangular toad (tension opposite to fill side).
= 0.0132 for triangular toad, 0.0162 for uniform load, 0.00687 for inverted
triangular load (tension opposite to fill side).
x — P,+P2
x = decided earlier, i = 2(x + xj), B = 7 —
102. For variable width combined footing, the dimensions may be
decided in the following
92. Lower One-Third Of Wall
sequence:
Kb at support = 0.0545 for triangular toad, 0.0852 for uniform load, 0.0311 for inverted
triangular load (tension on fill side)
Kh at midspan = 0.0268 for triangular toad, 0.0426 for uniform load, 0.0159 for inverted
triangular load (tension opposite to fill side).
x = P2 —
xS
, X|= decided earlier, L\ = x +xj

= £|,
= greater than S- x , decided earlier longer than or equal to L\.to beLilarger
If
Kv at bottom = 0.0496 for triangular load, 0.0616 for uniform load, 0.0120 for inverted Li than
triangular load (tension on fill side) footing of the first type is obtained. If Li > Ly, B] comes out
Bi and vice versa.
93,

1.
= lesser of the following three expressions

0.75x0.332777bod
L = L\+Li, B\
op
= -^(21-3^), B2 = —
27?
-B{

may be decided in the


2.
0.75x0.17^1(^-+2
^2/^t M 103. For abruptly changing width combined footing, the dimensions
following sequence:

3. 0.75x0.083
I b„
|llJfib.d
)
x , X) = decided earlier, Li — x + X|

94. $P„ = L3 = decided earlier and kept lesser than Lt


0O.85/e'4x2.O lies within
95. ACT minimum area of steel required for dowels = 0.005 L. = this length is also decided arbitrarily such that the second column
the supporting member.
A& where, Ae = gross area of this length
^3 P

96. Area of steel required for dowels within the column = Pu-^5f^A. ,
— = 0.65 L - I +T T ~ T I R - d
-
97. The ratio of reinforcement to be placed in the central band with respect to the
total required 104. P = q,ACl}^ q. = V3pV2 « 0.0475 V2
reinforcement is taken equal to 2
. C9 = - 0.7 outward for the roof
A+i Cv = 0.8 inward for the windward wall
P'l = - 0.5 outward for the leeward wall
98. Approximate depth = P+1.5MX+1.5MV + 60 mm > 250 mm
0^77 105. x

99. e^ =
—6^
Be,
1
6
L
and
— 6—
!
106. Richter magnitude, M
c
= loglO J(mm) + distance correction factor (tog Ao)

Lev B 107. p1,p2=-~


744 CONCRETE STRUCTURES P ART - 1 Appendix-V:
745
AUTHOR: ZAHID A. SIDDIQI

Jk/m = —
-tB=
108. c„ =27^ = 118. Base shear V= C, W
co = 0.7 R As
Maximum inelastic displacement Am
109.

110.
/= 2^

u(f)

=
/=

f,

—mon— [F(-r)sin[0 (r-r)]dr


Base shear V = (3.0 Ca I R)W

119. Shear at a particular story, Fx =


V'W a
-
\w.a<
T
Fx =
Wh,

120. Ft ~ 0.07 TV when T> 0.7 sec


111. «(/) =
— — fF^e'^’^sin^^-OjJr
-
md)b J> 121.

112. w(0 =

mco
(^COsin^C-Opr
113. UBC Typical Design Response Spectrum
199
122. P ~
H. r— 3
“b

Value of Constant K for Calculation of Development


Length.
< No. 20 and > No. 20
deformed wires bars
1 1
Given bar spacing criteria
satisfied 2.1 1.7
Other cases _1_ J_
1.4 1.1

,23. t,.
11
*»^.*23
d*
I 4 J the contribution of
Kir = transverse reinforcement index, a factor that represents
confining reinforcement present across the potential splitting
planes.
404
0.085 for steel moment resisting frames /
sn
0.073 for RC moment resisting frames and eccentric braced steel
frames 200 mm for /^31 MPa
0.050 for all other buildings 124. ^dc = F,Fi* 0.24 -d

115‘ Tb = > Fi F2 * 0.043 fy db 200 mm for/^ 31 MPa


1.47^ for Zones 1,2 and 3
300 mm for fy < 420 MPa
1.3 TA for Zones 4
125. £sc = 0.071 F fy dh
for 4 > 420 MPa
= (0.134-24)^4 300 mm
116. Response time T, =
——
C
2.5CO
- and Ta = 0.27^ where F =1 when 4' > 20 MPa

> 3 db
and 4/3 when

> 150 mm
20 MPa

117. C,
Maximum value
= (ifT>T,)
= 2.5 Ca I! R
subjected to maximum and minimum values
(Controls when
126. tdh = F/FyO.24^
^d'
Minimum value Ta to Ts)
= 0.11 Ca I (OR) Q.3ZNJ I R for zone-4
Appendix-V: 747
746 CONCRETE STRUCTURES P ART - 1 AUTHOR: Z AHI D A. SJ DDK)!
< 0.20, the provisions for slabs without interior beams
must be applied.
c) If
127. 4 < K^-+ia
K d)
ctfi„

For panel with one or more discontinuous


edges having edge beam with < 0.8,
panel. This increase is not required for slabs
128. Two-Way Slab Depth without Interior Beams is to be increased by at least 10% in that .n-J
• -
’imnnr limit of 1 OS and QO mm

The minimum thickness is greater of the following values and that given by Table:
EchC
a) Slabs without drop panels 125 mm
b) Slabs with drop panels
— 100 mm
131.
Slab Panels.
Minimum Slab Depth Without Interior Beams. Longitudinal Distribution Of Moments For Exterior
Slab without beams
Exterior panel
Interior panel Exterior edge Slab with between interior
+ either drop Exterior
+ drop panel unrestrained supports
Exterior panel panel or edge beams edge fully
fy (OR) (Torsion Part of slab
+ no drop panel beam between all considered With edge restrained
(MPa) Exterior panel
+ no edge beam (OR) member not supports
+ drop panel considered) as torsion beam
- Interior panel
+ edge beam member
+ no drop panel
(2) (3) (4)
280/300 4/33 ta /36 4/40 (1) ....

420 4/30 4 /33 4 /36 0.70 0.70 0.65


Int. M 0.75 0.70
520 4/28 4/31 4/34
0.52 0.50 0.35
0.57
Where 4 - length of clear span in long direction of two-way construction, measured face-to- M+ 0.63
face of supports in slabs without beams and face-to-face of beams or other supports in other 1 Ext. M 0.16 0.26 0.30 0.65
cases. 0
Note:- Edge beam is considered to be present if > 0.8.
= *B
129. Two-Way Slab Depth With Beams On All Sides
L>2. Let IM
£ lfA>25. 2.5
n-ZS
B —
Let, = D If«/,^>10. D-L0
and = average value of ‘a/ for all beams on edges of a panel. Interior negative moment (%age): 75 +
100 105 2BD(
_ )
For slabs with beams spanning between the supports on all sides and denoting clear span in the Exterior negative moment (%age): )
60 + 15(3
long direction by 4, the minimum thickness required is determined as follows: Positive moment (%age):
a) If 0.2 < c^m 2.0 (shallow beams). 133. M = 0.07 [(?du + 0.5 ?lu) f2C “ 1

_ 4 0.8+—- -
L
^min
\ 1400 )
but not less than 125 mm
1
* *
/
36 + 5/?(a>-0.2)
“X
3^2
b) If ctfi„ > 2.0 (stiff beams), db2bf
b^d3 +
135. Je for interior columns 6 6 2
0.8+-±i-
^min \ 1400
but not less than 90 mm
36 + 9/?
Append ix-V: 749
748 CONCRETE STRUCTURES P ART - 1 - mxcos2a + mysm2 a
AUTHOR: ZAHID A. SIDDIQI 143. mn
136,
and mt = - mxsinacosa + mv sinacosa
for edge columns = 2x + + = (my-mx) sinar cosa
144. V = m cot a
137. Jc for comer columns 145. h = ^6(m + m')/w
12
138. p

a) v. = o.n
146. Point load Pu is replaced by —
. "
•J a* Siny
in the affine slab

myn _ 1 — 2«
~
= met ~

0.083^^+2^777^ 147. wiyn ~


b) = m>E a~

where a =40 for interior, 30 for edge and 20 for corner columns 148. For Slabs with Short Unsupported Edges:
c) = 0.331777 1-32^ ,
139. The non-prismatic slab-beam member may approximately be converted into a prismatic
member as under:
af-

8A-_1
X

= (1 £) fy/4 Wyp — 8-
I y
3+ k

FEM =" nrx. q t? = k X E/Sb I f]


_ depth at drop panel 149. For Slabs With Longer Unsupported Edges
a = 0.09 -^x^J > 0.083 k^-aY-lm^lq^
depth of slab
AJ 2 =
/ \0.05 / \0.02 a{2-a)
k = 5.3 (A a09 > 4.0 COF = 0.57 a0-37 > 0.5
AJ \A Ay 150. Deflections.
140. The approximate equivalent prismatic column properties may be determined as: A- For uniformly loaded and simply supported beams,
t - vertical distance starting from the slab centerline up to inner end of
slab, inner end of drop panel or mid-height of column capital
Amax
3S4E1
= k^EIJ tc B- For uniformly loaded continuous beams,
_
5£2 + M.)]
k. = 4.0
/ x0 08 z \ 2.7 z Amidspan =
48 EI ^-O.KAG
where Mc = magnitude of central moment,
and Ma,Mh= magnitude of end moments.
for 4 1 tb = 0.4 to 2.2 and £c / 4 up to 1.2

COFa = 0.5 V
A —,t
V—
XO.08
C- For simply supported beams subjected to point load (refer to Figure),

141. X 7^£ xh

1 1 , 1
142.
EC
CONC RETE STRUCTU RES PA RT - I
Appendix-V: 751
750 AUTHOR: ZAHID A. SIDDIQI

D- For overhanging part of beam subjected to UDL, REFERENCES


f . a .

1. Building Code Requirements for Structural Concrete, ACI 318, American Concrete
Institute, Ml, USA, 20 11.
2. M. N. Hassoun and A. A. Manaseer, Structural Concrete - Theory and Design, 4,h Ed.,
John Wiley & Sons, Inc., 2008.
E- For the above case, with UDL also present within supports, 3. C. K. Wang, C. G. Salmon, and J. A. Pincheira, Reinforced Concrete Design, 7th Ed.,

(4a2£-£3 +3a3) John Wiley & Sons, Inc., 2007.
Amar = 24E7 v ’ 4. E. G. Nawy, Reinforced Concrete - A Fundamental Approach, 5th Ed., Pearson Prentice
F- For overhanging part of beam subjected to point load, Hall, NJ, USA, 2005.
5. PCA Notes on ACI - 2005, Portland Cement Association, 5420 Old Orchard Road,
No load
Skokie, Illinois 60077-1083, 2005.

£
f/S.
6. A. H. Nilson, D. Darwin, and C. W. Dolan, Design of Concrete Structures, 13th Ed.,
1 3E1 McGraw Hill, 2004.
151. Bending moment equal span
in (L) continuous beams subjected to central point load (P). 7. CRS1 Design Handbook, Concrete Reinforcing Steel Institute, 9th Ed., American
First interior support 0.188PZ Institute of Steel Construction, IL, USA, 2002.
Outer span 0.213 PL (Live load) : 0.175 PL (Dead load) 8. J. C. McCormac, Design of Reinforced Concrete, S4 Ed., John Wiley & Sons, Inc.,
Interior supports 0.167 PL (Live load) : 0.11 8 P L (Dead load) 2001.
Interior spans 0.183 P L (Live load) : 0.1 16 PL (Dead load)
9. ASTM Standards, American Society for Testing and Materials, 100 Barr Harbor Drive,
West Conshohocken, PA 19428-2959, United States.
152. Bending moment in equal span (L) continuous beams subjected to two third point loads
(P)- 10. A. K. Chopra, Dynamics of Structures, Prentice Hall, 1998.
First interior support 0.334 P L 11. Uniform Building Code, UBC-97, International Conference of Building Officials, Whittier,
Outer span 0.288 P L (Live load) : 0.244 PL (Dead load) California, USA, 1997.
0.296 P L (Live load) : 0.210 P L (Dead load)
Interior supports
Interior spans 0.222 P L (Live load) : 0.122 PL (Dead load)
12. BS8110, Structural use of concrete — Part 1: Code of practice for design and
construction, British Standard Institute, 1997.
153. Shear force in equal span (Z) continuous beams subjected to central point load (P). 13. ACI Design Handbook, SP-17, American Concrete Institute, MI, USA, 1997.
First interior support 0.688 P 14. J. G. MacGregor, Reinforced Concrete, 3rd Ed., Prentice-Hall, NJ, USA, 1997.
Outer support 0.425 P (Live load) 0.350 P (Dead load)
Interior supports 0.636 P (Live load) 0-500 P (Dead load) 15. J. E. Bowles, Foundation Analysis and Design, 5th Ed., McGraw Hill, 1996.
154. Shear force in equal span (Z) continuous beams subjected to two third point loads (P). 16. A. M. Neville, Properties of Concrete, 4th ed., John Wiley & Sons, Inc., New York,
1996.
First interior support 1.334 P
Outer support 0.866 P (Live load) 0.734 P (Dead load) 17. Design and Performance of Mat Foundations - State of the Art Review, SP-152,
Interior supports 1.242 P (Live load) 1.000 /(Dead load) American Concrete Institute, Detroit, 1995.
18. . ACI Detailing Manual, ACI Special Publication SP66, American Concrete Institute,
MI, USA, 1994.
752 CONCRETE STRUCTURES PART - 1 AUTHOR: ZAHID A. SIDDIQI
19. B. C. Punmia, A. K. Jain, and A. K. Jain, Reinforced Concrete Structures, 7th Ed.,
Laxmi Publications (P) Ltd, 1992.
20. M. Paz, Structural Dynamics - Theory and Computation, 3rd Ed., Chapman and Hall,
1991.
21. J. C. McCormac, Structural Steel Design: LRFD Method, HarperCollins Publishers,
New York, 1989.
22. C. E. Reynolds and J. C. Steedman, Reinforced Concrete Designer’s Handbook, 10th
Ed., W. & F.N. Spon, London, 1988.
23. M. Fintel, Handbook of Concrete Engineering, 2nd Ed., Van Nostrand Reinhold Co.,
New York, 1985.
24. Vanderbilt and Corley, “Frame Analysis of Concrete Buildings”, Concrete
International, Dec. 1983.
25. R. Park and W. L. Gamble, Reinforced Concrete Slabs, John Wiley and Sons, New
York, 1980.
26. R. Park and T. Pauley, Reinforced Concrete Structures, John Wiley, New York, 1975.
27. Code of Practice for the Structural Use of Concrete, (CP 110) Part 1, British Standards
Institution, London, 1972.
28. R. H. Wood and G. S. T. Armer, “The Theory of the Strip Method for the Design of
Slabs,” Proc. Inst. Civ. Eng., Vol.41, 1968.
29. L. L. Jones and R. H. Wood, Yield Line Analysis pf Slabs, American Elsevier, New
York, 1967.
30. A. H. Mattock, “Rotation Capacity of Hinging Regions in Reinforced Concrete
Frames,” ACI Publication SP-12, 1964.
31. Building C 'de Requirements for Structural Concrete, ACI 318, American Concrete
Institute, Detroit, USA, 1964.
32. E. Hognestad, “Yield Line Theory for the Ultimate Flexural Strength of Reinforced
Concrete Slabs,” J. ACI, Vol. 24, No. 7, 1953. ,

You might also like