viii 3118? 023 ee8 blOyREINFORCED CONCRETE DESIGNAlso available from McGraw-Hill Schaum’s Outline Series in Civil Engineering Most outlines include basic theory, definitions and hundreds of example problems solved in step-by-step detail, and supplementary problems with answers. Related titles on the current list include: Descriptive Geometry Dynamic Structural Analysis Engineering Esonomics Engineering Mechanics Fluid Dynamies Fluid Mechanics & Hydraulics Introductory Surveying Mathematical Handbook of Formulas & Tables Reinforced Concrete Design Statics & Mechanics of Materials Strength of Meterials Structural Steel Design (LRFD Method) Schaum's Solved Problems Books Each title in this series is a complete and expert source of solved problems with solutions worked out in step-by-step detail Related titles on the list include: 3000 Solved Problems in Calculus 2500 Solved Problems in Differential Equations 2500 Solved Problems in Fluid Mechanics & Hydraulics 3000 Solved Problems in Linear Algebra 2000 Solved Problems in Numerical Analysis 700. Solved Problems in Vector Mechanics for Engineers: Dynamics 800. Solved Problems in Vector Mechanics for Engineers: Statics Available at most college bookstores, or for a complete list of titles and prices, write to Schaum Division ‘The McGraw-Hill Companies, Inc. LI West 19th Street New York, NY 10011Conforms to 1995 ACI Codes REINFORCED CONCRETE DESIGN THIRD EDITION Kenneth M. Leet Northeastern University Contributing Author Dionisio Bernal Northeastern University ‘THE McGRAW-HILL COMPANIES, INC. New York St. Louis San Francisco Auckland Bogoté Caracas Lisbon London Madrid Mexico City Milan Montreal New Delhi San Juan Singapore Sydney Tokyo TorontoMcGraw-Hill 2 A Division of The McGraw Hill Companies REINFORCED CONCRETE DESIGN Copyright © 1997, 1991, 1982 by The McGraw-Hill Companies, In. All rights reserved Printed in the United States of America. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a data base or retrieval system, without the prior written permission ofthe publisher. ‘This book is primed on acid-free paper. 1234567890SEM BKP909876 ISBN 0-07-037100-8 This book was set in Times Roman by Publication Services, In: The editors were B. J. Clark and John M. Morriss; the production supervisor was Kathryn Porzio. The cover was designed by Carla Bauer. Project superision was done by Publication Services, Ine. Quebecor Semline was printer Quebecor BookPress was binder Cover photo ert: Photo by Ezra Stoller © Esto Library of Congress Cataloging-in-Publication Data Leet, Kenneth Reinforced concrete design: conforms to 1995 ACI codes/Kenneth Lect; contributing author, Dionisio Bernal. —3ed ed. p. om Includes inden. SRN 0.07-097100.8ISRN 04.07-097101-6 fam) 1. Reinforced concrete constuction. I. Bernal, Dionisio. UL. Tile, ‘TAGS3.2.L36 1997 624.1'8341 420 9631173 Itp:/fwww.mnbcollege.comChapter 1 Ld 12 13 14 LS 16 17 18 19 1.10 Chapter 2 21 22 23 Chapter 3 3 32 33 34 35 36 37 38 Preface Introduction Overview Concrete and Reinforced Concrete Units of Measure Codes Service Loads Duetility versus Brittleness Strength and Serviceability Elastic versus Strength Design ‘The ACI Design Procedure Overview of the Design Procedure References Questions Problems Materials Concrete Mechanical Properties of Concrete Reinforcement References Questions Design of Beams for Flexure Introduction The Uncracked Cross Section; Computation of the Cracking Moment Working-Stress Design; Service-Load Stresses Limit of Elastic Behavior for an Underreinforced Beam Deflections under Service Loads Control of Cracking under Service Loads Failure Modes and Flexural Strength of Reinforced Concrete Beams The Plastic Hinge and Moment-Curvature Diagrams CONTENTS 39 42 48 51 63 68 5 v39 3.10 3. 3.12 3.13 3.4 3.15 3.16 3.17 Chapter 4 4 42 43 44 45 46 47 43 Chapter § 34 5.2 33 34 55 56 57 58 Chapter 6 61 62 63 64 65 6.5 Minimum Reinforcement to Prevent a Brittle Failure Balanced Failure of a Beam with a Rectangular Cross Section Strength Design of Rectangular Beams for Moment ‘Trial Method Balanced Steel for Beams with Nonrectangular Compression Zones Design of Beams with Compression Steel Design of T-Beams One-Way Slabs Temperature and Shrinkage Steel References Questions, Problems Shear and Diagonal Tension Introduction Description of a Shear Failure in a Beam Reinforced for Moment Design Procedure for Shear Design of Beams with No Shear Reinforcement Design of Shear Reinforcement Modeling of a Reinforced Concrete Beam as a Truss Punching Shear Shear Friction References Questions Problems Torsion Introduction Distribution of Shear Stresses Created by Torsion on a Rectangular Cross Section Prediction of Shear Stresses on a Rectangular Cross Section Due to Torque ‘Computation of the Cracking Torque Ter Reinforcing for Torsion Design Equations for Torsional Reinforcement Summary of Design Procedure for a Member Stressed by Torsion, Shear, and Moment Equilibrium and Compatibility Torsion References Question Problems Bond, Anchorage, and Reinforcing Details Scope of Chapter Bond Stresses Distribution and Variation of Bond Stresses Mechanics of Bond Strength Nominal Bond Strength Development Length of Tension Steel 1 81 95 100 103 12 8 n9 121 121 122 130 130 132 136 144 146 160 165 167 168 169. 169 12 172 173 175, 178, 192 183, 190 196 199 199 199 202 202 203 204 205 209 267 68 69 6.10 6.11 6.12 Chapter 7 Td 72 73 14 15 16 17 78 19 7.10 7 712 713 714 Chapter 8 8.1 8.2 83 84 85 8.6 8.7 * Chapter 9 OL 9.2 93 94 95 96 Development Length of Compression Steel Standard Hooks Anchorage of Steel at Simple Supports and Points of Inflection Cutoff Points Splicing Reinforcement Comprehensive Design Example References Questions Problems Design of Columns Introduction Fundamentals of Column Behavior Braced and Unbraced Frames Effective-Length Factors for Columns of Rigid Frames ‘The ACI Procedure for Classifying Beam-Columns Strength of Short Axially Loaded Columns Plastic Centroid Strength of Short Columns for Axial Load and Moment. Design Procedure for a Short Column Design of Long Columns Magnification of Moments in Braced (Nonsway) Columns with Unequal End Moments Magnification of Column Moments in Unbraced Frames Stability of Unbraced Frames Supporting Gravity Loads Biaxial Bending References Questions Problems Footing Design Introduction Types of Foundations Soil Pressures Design of Footings for Vertical Load Design of a Wall Footing with a Uniform Load Design of an Axially Loaded Two-Way Footing Design of a Combined Footing for Two Columns Reference Problems Retaining Walls Introduction Earth Pressures on Retaining Walls Behavior of a Cantilever Retaining Wall Design Procedure for a Cantilever Retaining Wall Load Factors and Design Assumptions Design Example References Problems 222 223, 227 230 234 238 243 243 248 248 253 256 258 261 265 270 272 284 291 295 300 303 307 313 313 313 318 318 319 321 325 326 330 336 344 347 347 350 354 355 337 359 37 372 vii ‘CONTENTSvill Chapter 10 16.1 10.2 10.3 10a 10.5 10.6 10.7 10.8 10.9 10.10 Chapter 11 ALL 12 13 na nS 11.6 7 118 119 1.10 Wl Chapter 12 11 12.2, 123 12.4 ins 12.6, 12.7 12.8 12.9 1210 12.11 12.12, 12.13 12.14 1215 1216 12.17 The Design and Analysis of Multistory Building Frames Introduction Factors That Influence the Configuration of Floor Systems Behavior of Rectangular Slabs Supported on Four Sides Lateral Bracing of Buildings Preliminary Design Analysis of Multistory Building Frames Positioning Live Load for Maximum Moment Analysis by ACI Coefficients Moment Redistribution Steel versus Reinforced Concrete for Building Frames Reference: Problems Design of Two-Way Slabs Introduction Forces in Two-Way Slabs Midspan Moments in a Rectangular Slab Supported at the Comers Subdividing a Building into Two-Dimensional Frames The Equivalent Column Minimum Thickness of Two-Way Slabs to Control Deflections Analysis of Two-Way Slabs Design Moment for Columns ‘Transfer of the Floor Load into Columns Shearhead Reinforcement Openings in Slabs References Problems Prestressed Concrete Introduction Description of the Prestressing Procedure Prestressing Methods ‘Materials and Allowable Stresses Flexural Design Requirements for Prestressed Beams Forces Exerted by Stressed Tendons Elastic Analysis of Prestressed Beams Kern Points ‘The Cracking Moment Behavior of a Prestressed Beam Loaded to Failure Strength Design for Flexure Design of a Cross Section for Moment Design for Shear Deflections End-Block Stresses Loss of Prestress Prestressing Columns References Questions Problems 373 373 375 377 319 380 380 384 386 391 394 395 395 397 397 399 403 409 412 430 431 437 438 438 439 442 447 450 450 461 465 466 467 469 474 480 485 490 499 499 499 500Chapter 13 13.1 13.2 133 13.4 13.5 13.6 13.7 13.8 13.9 13.10 moa Introduction to Yield Line Analysis Introduction The Critical Yield Line Pattern Basic Assumptions Yield Line Patterns Analysis Using the Virtual Work Principle Flexural Strength along Yield Lines Rotation at Yield Lines Concentrated Loads Computation of Internal Work Using Plate Element Rotations Design Formulation References Questions Problems Appendixes Conversion Factors Bar Area and Spacing in Slabs. Distribution Factors for Moments in Two-Way Slabs Properties of Standard Prestressed Concrete Beam Cross Sections Notation Answers to Questions ‘Answers to Selected Problems Index 504 504 506 506 507 509 513, 514 517 519 519 521 522 522 523, 523 524 525 526 528 535 539 542 ix CONTENTSPREFACE Engineering and architectural students who complete this book will have acquired the back- ‘ground necessary to design the majority of reinforced concrete structures commonly encountered in professional practice. ‘My goal in writing this text was to clarify the behavior and simplify the design of reinforced concrete—a topic that can be quite complex if not presented in an orderly sequence. NEW FEATURES OF THE THIRD EDITION New ACI Code provisions: Pleased that many instructors have found the second edition of Reinforced Concrete Design “ery teachable,” have retained the underlying framework of the text and have concentrated on updating the topics already presented, incorporating the latest technological changes in materials and design philosophy contained in the most recent edition (1995) of the American Concrete Institute (ACI) Building Code Requirements for Structural Concrete (ACI 318-95). Among the major changes in this third edition, reflecting the new pro- visions of the Code, are design procedures for the following topics: Torsion: Chapter 5, Torsion, has been completely revised to conform to the new Code provisions. The new design procedure for torsion treats members with solid cross sections as hollow thin-walled tubes to establish the internal shear forces. Once these forces are determined, the reinforcement is sized by assuming forces are transmitted through a three-dimensional space truss in which diagonal concrete struts carry compression and steel reinforcement acts as tension ‘members. Development length of bars in tension: Chapter 6 contains the new Code provisions for ‘ensuring safe anchorage of tension steel. Unlike the previous provisions, which mandated three steps to determine the required development length of a bar, the new design procedure has been reduced to a single equation with four factors that account for the influence of thickness of cover or spacing between bars, bar size, epoxy coating, and depth of concrete under bars. Analysis and design of slender columns in unbraced frames: The new Code provisions affect many sections of Chapter 7. These provisions establish the appropriate value of moment of inertia to be used to classify columns as long or short and to analyze indeterminate frames. New procedures are given for the magnification of moments in unbraced frames that are analyzedxil by first-order computer programs to account for secondary (P-delta) moments. Also various ctiteria are established to ensure that columns in unbraced frames have adequate stiffness to prevent excessive deflection under both gravity and lateral loads. Many new examples have been added to illustrate the new Code provisions. Minimum flexural steel for high-strength concrete members: To ensure a ductile failure of lightly reinforced beams constructed of high-strength concrete, the provisions for minimum flexural steel in Section 3.9 have been expanded to express minimum steel requirements in terms of the compressive strength of the concrete. This section also contains new provisions that crease the minimum steel requirements for T-beams whose flanges are stressed in tension. Design of shear reinforcement using the truss analogy: A new Section 4.6 has been added to develop the truss analogy for sizing shear reinforcement. This method is based on the observation that a beam on the verge of failure behaves like a truss. The new procedure, now Utilized to design stirrups for vertical shear in the latest edition of the bridge code of the American Association of State Highway and Transportation Officials (AASHTO), will in the near future probably replace the current ACT design provisions for calculating shear reinforcement. The truss analogy is also utilized in Chapter 5 to establish equations for sizing torsion reinforce- ment. Expanded coverage of SI metric units: Since most engineering firms in the United States work exclusively in U. S. customary units (USCU), the majority of the examples and homework problems in this text use this system. However, to permit the American concrete construction industry to compete globally in future years, the engineering profession can expect increasing numbers of engineering projects to be carried out in metric units. Therefore, in this transition period, I have expanded the use of metric units in the present edition—particularly in the first seven chapters—so that instructors can provide students with the background needed to design basic structural elements (beams, slabs, and columns) in S/ metric units. To this end, tie book now contains tables listing properties of metric bars, provides all basic equations in both USCU and SI metric units, adds examples in which material properties, dimensions, and other design information are given in metric units, and supplies a variety of homework problems in metric units. Based on my own experience, I believe that students who have acquired a solid under- standing of reinforced concrete behavior will feel comfortable working in metric units after a brief exposure. Revised homework problems and design examples: A large selection of homework problems has always been an important feature of earlier editions. Where helpful to the student, Thave incorporated additional design examples and have added new and stimulating homework problems throughout. From my many years of teaching, I have culled design problems and examples that have enabled students to grasp the essential concepts most easily. I have chosen whatever worked best in my own classroom over the years, selecting examples that get to the heart of the concept under discussion. In the third edition, most homework problems in Chap- ter 3 (on bending) have been revised and many additional problems added. Other chapters—on shear, bond, torsion, and columns—also contain many new or revised homework problems. GOALS OF THE TEXT In writing this introductory text, my primary aim was to enable the student to develop a clear understanding of the basic principles—the critical groundwork that must be laid in a first- year course. In a carefully sequenced, step-by-step approach, each topic is broken dowr into itssimplest and most essential components so that students, in turn, learn how to break down and deal with any structure—no matter how complex—by looking at the basic behavior of each ‘component, ‘All design equations are derived from a few fundamental principles of staties and engineer- ing mechanics that should be familiar to the average engineering student with a background in these areas The text, which has been class-tested for over 20 years, is geared toward the average stu- dent's needs, questions, and difficulties, yet the more advanced problems will challenge even those students who wish to specialize in reinforced concrete design. Encouraging students to take into account the basic behavior and characteristics of rein- forced concrete allows them to understand the rationales behind the provisions of the ACI Code and prepares them to deal knowledgeably with design situations not covered by the Code. COURSE LEVEL AND TEACHING SUGGESTIONS This text contains sufficient material for two semesters of reinforced concrete design. The ma- terial for the first course, which provides the essential background for proportioning members and detailing reinforcement, is covered in Chapters | through 7. If time permits, Section 10.8, on the analysis of continuous members by ACI coefficients, may be logically introduced after Section 3.16 has been covered. This permits students who are not familiar with indeterminate analysis to calculate design moments in continuous one-way ‘members. If time is limited, or if the instructor wishes to develop certain topics in more detail, Chapter 5, on torsion, may be omitted from a first course without any loss of continuity. Ina second course, the material of Chapters 8 through 13 will give the student an under- standing of the design of typical reinforced concrete structures and prestressed concrete. Chap- ters 8 and 9, on foundations and retaining wall design, provide the student with the opportunity to apply many of the design procedures introduced in the first seven chapters to a number of relatively simple but common structural elements. Chapters 10 and 11, on building design, de- scribe how the analysis of large, highly indeterminate three-dimensional building frames for gravity load can be simplified by subdividing the structure into smaller two-dimensional sub- systems. Finally, Chapter 12 contains an extensive treatment of prestressed concrete, which provides certain advantages over reinforced concrete because cracking can be eliminated and higher-strength materials can be used more efficiently. TWO TYPES OF EQUATIONS ‘This text contains two types of equations. The first type, derived from principles of mechanics, is used to relate internal stresses to applied forces. The variables in these equations may be expressed in terms of any set of dimensionally consistent units. In the second type of equation, which is usually based on experimental studies, the variables are restricted to specific units. For example, except for the equation predicting crack width under service loads, all ACI equations in U. S. customary units require that the compressive strength of concrete f, and the yield point of steel f, be expressed in pounds per inch squared (Ib/in”).. ACKNOWLEDGMENTS I would like to express my deep appreciation to colleagues, friends, students, and reviewers of the text, who have contributed greatly to the development of this book over the years. PRERCExiv PREFACE In particular, I would like to thank Judith Leet, who edited the initial draft of the first edition; ‘Saul Namyet and Robert Taylor, who reviewed many of the chapters of the first edition and offered many valuable suggestions; Walid Najjar and Chia-Ming Uang for their contributions to the second edition; and J. P. Paniagua who provided the second-order computer analysis of an unbraced frame, in Chapter 7 of the present edition. Chapter 13 of the third edition, on yield line theory, was written by Dionisio Bernal of Northeastem University, who also prepared the initial draft of Chapter 5. I would also like to thank Kris Engberg and Matthew Harris of Publication Services, Inc., for their help in expediting the production of the third edition. Finally, I would like to acknowledge with appreciation the support over the years of both Northeastern University and Mishac Yegian, chairman of the Northeastem Civil Engineering Department. For this third edition I would like to express my thanks to the following reviewers: Dr. P.R. Chakrabarti, California State University at Fullerton; Patrick Findlay, U.S. Military Academy James Nau, North Carolina State University; Irving Oppenheim, Caregie-Mellon University: Andrew Scanlon, Pennsylvania State University; and R. P. Silvan, University of Arkansas. Kenneth Leet Northeastern UniversityCHAPTER I INTRODUCTION Cascade Orchards Bridge over the Wenatchee River, Leavenswonth, Washington, Continuous prestressed concrete bridge with slender beams designed by Arvid Grant Associates. (Photograph by Arvid Grant.) 1.1 OVERVIEW In almost every branch of civil engineering and architecture, extensive use is made of rein- forced conerete for structures and foundations. Engineers and architects therefore require a ba- sic knowledge of reinforced concrete design throughout their professional careers. Much of this text is directly concerned with the behavior and proportioning of the components that make up typical reinforced conerete structures—beams, columns, and slabs, Once the behavior of these individual elements is understood, the designer will have the background to analyze and design a wide range of complex structures, such as foundations, buildings, and bridges, composed of these elements. Since reinforced concrete is a nonhomogeneous material that creeps, shrinks, and cracks, its stresses cannot be accurately predicted by the traditional equations derived in a course in strength of materials for homogeneous elastic materials. Much of reinforced concrete design is therefore empirical; ic., design equations and design methods are based on experimental and time-proven results instead of being derived exclusively from theoretical formulations, 12 REINFORCED CONCRETE Desi A thorough understanding of the behavior of reinforced concrete will allow the designer to convert an otherwise brittle material into tough, ductile structural elements and thereby take ad- vantage of concrete’s desirable characteristics—its high compressive strength, its fire resistance, and its durability. 1.2 CONCRETE AND REINFORCED CONCRETE Concrete, a sionelike material, is made by mixing cement, water, fine aggregate (often sand), coarse aggregate, and frequently other additives (that modify properties) into a workable mix- ture. In its unhardened, or plastic state, concrete can be placed in forms to produce a Lange variety of structural elements. Although the hardened concrete by itself, ie., without any reinforcement, is strong in compression, it lacks tensile strength and therefore cracks easily. Because unrein- forced concrete is brittle, it cannot undergo large deformations under load and fails suddenly without warning. The addition of steel reinforcement to the concrete reduces the negative ef- fects of its two principal inherent weaknesses, its susceptibility to cracking and its brittleness. Although steel isa stiff, high-strength material, it also has several weaknesses that can be min- imized or eliminated by encasing it in concrete. Concrete surrounding steel protects it from corrosion by moist air or salt water. At temperatures over 1200°F, the tensile strength of steel reduces rapidly. Since concrete is a good insulator, steel that is protected by several inches of concrete cover will retain its strength during several hours of exposure to intense heat. Thus steel and concrete form a synergistic relationship; i.e., each material improves the usefulness of the other. When properly combined by the designer, weaknesses of each are largely eliminated while all desirable characteristics are retained. ‘When the reinforcement is strongly bonded to the concrete, a strong, stiff, and ductile con- struction material is produced. This material, called reinforced concrete, is used extensively to construct foundations, structural frames, storage tanks, shell roofs, highways, walls, dams, canals, and innumerable other structures and building products. Two other characteristics of concrete that are present even when concrete is reinforced are shrinkage and creep, but the neg- ative effects of these properties can be mitigated by careful design. 13 UNITS OF MEASURE Although the United States has been committed for several decades to converting its system of measurements from U.S. customary units (foot, pound, second) to SI units (meter, newton, sec- ond), litle change has taken place in the American concrete industry to date. At present (1997), structural engineering offices in the United States work almost exclusively in U.S. customary units. Today the United States is the only developed country in the world that has not converted fully to the metric system. However, itis now clear that metrification can no longer be delayed by the American construction industry if itis to compete globally. To encourage conversion to the metric system in the United States, a number of federal agencies now require that ergineers use metric units exclusively in all new designs under their control Over the next four to five years, the United States will likely be in a major transitional pe- riod during which the construction industry will shift to metric units (SI) for all new designs in order to be compatible with the rest of the world. To avoid fractions and permit modular con- struction, this switch to metric units will also require that manufacturers modify the dimensions of reinforcing bars and other products used in reinforced concrete construction. ‘The American Concrete Institute (ACI) Code, which contains the specifications that govern the design and construction of reinforced concrete structures in the United States, is currently published in two versions. In one edition, equations are expressed in U.S. customary units; in the other edition, equations are formulated in SI units. The normal units used in the equations of each system are listed in Table 1.1. Some conversion factors are given in Appendix A.TABLE 1.1 Units used in ACI Code equations US. customary SI Length in ‘mm (millimeter) ft ‘m (meter) Load Ib N (newton) Stress Tofin® ‘MPa (megapascal) Density Tole (weight) kg/m? (mass) Since engineers are frequently required to establish the strength of existing structures, many of which have been designed in U.S. customary units (USCU), structural designers must be able to work in both USCU and SI units with equal facility. Therefore, in addition to working a number of problems in metric units, we will give basic data, important equations, certain design aids, and a number of examples in both systems, with the SI units typically in parentheses after the U.S. customary units. Because of round-off in the coefficients of the ACI metric equations may give values differing by several percent from those given by equivalent equations in U.S. customary units 14 CODES ‘A code is a set of technical specifications and standards that control important details of design and construction. The purpose of codes is to produce sound structures so that the public will be protected from poor or inadequate design and construction. ‘Two types of codes exist. One type, called a structural code, is originated and controlled by specialists who are concerned with the proper use of a specific material or who are involved with the safe design of a particular class of structures. For the structural engineer we list several important codes: ‘The American Concrete Institute (ACI) Building Code 318-95, covering the design of re- inforced concrete buildings ‘The American Institute of Steel Construction Specifications (AISC), covering the design of steel buildings ‘The American Association of State Highway and Transportation Officials (AASHTO), cov- ering the design of highway bridges ‘The American Railroad Engineering Association (AREA), covering the design of railroad bridges The second type of code, called a building code, is established to cover construction in a given region, often a city or a state. The objective of a building code is also to protect the public by accounting for the influence of local environmental conditions on construction. For example, local authorities may specify additional provisions to account for such regional conditions as earthquake, heavy snow, or tornados. National structural codes generally are incorporated into local building codes. The ACI Code contains provisions covering all aspects of reinforced concrete manufacture, sign, and construction. It includes specifications on quality of materials, details on mixing and placing concrete, design assumptions for the analysis of continuous structures, and equations for proportioning members for design forces. All design procedures used in this text are consistent with specifications of the ACI Code. ‘The specifications of the ACI Code, based on many years of research and field experience with reinforced concrete, represent the minimum standards required to produce safe, serviceable 3 INTRODUCTION4 Esian structures, As research provides additional understanding of behavior, the contents of the ACI Code are continually reviewed and updated. Currently the ACI Code committee makes yearly changes by issuing supplementary provisions. Every six or seven years a comprehensive code revision is made, incorporating all revisions since the last edition. The ACI Code, like most codes, makes provisions for departures from prescribed standards if it can be shown by test or analysis that such changes will produce a safe design. All structures must be proportioned so they will not fail or deform excessively under any possible condition of service. Therefore itis important that an engineer use great care in anticipating all the probable loads to which a structure will be subjected during its lifetime. ‘ACI Code §8.2 contains provisions listing the factors that should be considered when es- tablishing the forces in concrete structures. Although the design of most members is controlled typically by dead and live loads acting simultaneously, consideration must also be given to the forces produced by wind, impact, shrinkage, temperature change, creep, support settlements, earthquake, and so forth The load associated with the weight of the structure itself and its permanent compenents is called the dead load. The dead load of concrete members, which is substantial, should never be neglected in design computations. The exact magnitude of the dead load is not known accurately until members have been sized. Since some figure for the dead load must be used in computations tosize the members, its magnitude must be estimated at first. After a structure has been analyzed, the members sized, and architectural details completed, the dead load can be computed more accurately. Ifthe computed dead load is approximately equal to the initial estimate of its value (or slightly less), the design is complete, but if a significant difference exists between the computed and estimated values of dead weight, the computations should be revised using an improved value of dead load. An accurate estimate of dead load is particularly important when spans are long, say over 75 ft (22.9 m), because dead load typically constitutes major portion of the design load. Live loads associated with building use are specified by city or state building codes. Instead of attempting to evaluate the weight of specific items of equipment and occupants in « certain area of a building, building codes specify values of uniform live load for which members are to be designed. Typical values of live load for standard buildings are listed in Table 1.2. After the structure has been sized for dead and live or gravity load, itis checked for wind in combination with dead and live loads as specified in the code. Wind loads do not usually control the size of members in buildings less than 16 to 18 stories, but for tall buildings wind loads become significant and cause large forces to develop in the structures. Under these conditions, economy can be achieved only by selecting a structural system that is able to transfer herizontal loads into the ground efficiently. In seismic zones, analysis for dynamic effects due to ground motions must be considered. Dynamic forces in the structure are a function of the building’s mass and the acceleration im- parted to the building by ground motions. For buildings located in zones of low to moderate seismic activity, the provisions of the main body of the ACI Code produce sufficient ductility to permit a concrete structure to withstand the shaking due to ground motions. In regions where large earthquakes have a high probability of occurrence, concrete structures must be designed and detailed in accordance with the provisions of Chapter 21 of the ACI Code in order to ensure high ductility and toughness. When members are subjected to large lateral forces, reversal of stfesses may occur, and regions which would normally be in compression under gravity load are stressed in tension, For members subject to reversal of stress, reinforcement must be provided oon both sides of members,5 INTRODUCTION TABLE 1.2 Typical live-load values! imum uniformly Type of use Apartment buildings: Private units 40 1.92 Public rooms 100 4.80 Corridors 80 3.84 Office buildings: Offices 30 2.40 Lobbies 100 4.80 Corridors above first floor 80 3.84 Garages (cars only) 50, 2.40 Stores: First floor 100 48 Upper floors 75 36 Warehouse: Light storage 125 60 Heavy storage 250 120 1.6 DUCTILITY VERSUS BRITTLENESS A major objective of the ACI Code is to design concrete structures with adequate ductility since concrete is brittle without reinforcement. The term ductility describes the ability of a member to undergo large deformations without rupture as failure occurs. A structural-steel girder is an example of a ductile member that can be bent and twisted through large angles without rupture. Ductile structures may bend and deform excessively under load, but they remain, by and large, intact. This capability prevents total structural collapse and provides protection to occupants of buildings. On the other hand, the term brittle describes members that fail suddenly, completely, and with little warning. When a brittle member fractures, it usually disintegrates and may dam- age adjacent portions of the structure or overload other members, bringing on additional failures. The ability of a ductile structure to undergo large deformations before collapse produces visible evidence of impending failure and may give occupants the opportunity to relieve distress by reducing loads. In contrast, brittle failures occur suddenly, without warning, and with no time for measures to be taken to prevent damage. ‘The need to engineer ductility into the entire structure of a building was demonstrated dra- matically in 1968 when 24 stories of the exterior comer of the prefabricated Ronan Point Tower in London collapsed after a gas explosion on the eighteenth floor had blown out a single load- bearing wall panel.?} This type of collapse, in which the effects of a local failure spread to the entire structure or to a significant portion of the structure, is termed a progressive collapse. Design provisions that reduce the likelihood of progressive failures are contained in ACI Code §7.13. These provisions contain detailing requirements that tie together both cast-in-place and precast concrete structural elements (1) by requiring continuity of the reinforcement be- tween members and (2) by providing for the effective anchorage of the ends of reinforcing bars. Ifa local failure ruptures a member or produces settlement of a column or other supports, ‘fNumbered references appear at the end of each chapter.6 ‘REINFORCED CONCRETE well-anchored steel that remains bonded to the body of members permits sagging beams and slabs to transmit load by catenary (cable) action. The provisions of ACI §7.13 include: 1, Requiring continuous top and bottom longitudinal reinforcement in perimeter beams. 2. Encircling longitudinal reinforcing bars with closely spaced hoops of steel (stirrups) to pre- vent longitudinal bars from tearing out of the conerete if a local failure occurs. 3. Ensuring that over supports, bottom reinforcing bars from adjacent beams and slabs are well anchored by splicing them together. 4, Providing tension ties in the transverse, longitudinal, and vertical directions and also around the perimeter of a structure to effectively connect individual precast elements. 5, Eliminating all connection and support details between precast elements that depené solely on friction, created by gravity loads, to prevent slippage. 1.7. STRENGTH AND SERVICEABILITY Members are always designed with a capacity for load that is significantly greater than that required to support anticipated service loads. This extra capacity not only provides a factor of safety against failure by accidental overload or defective construction but also limits the level of stress under service load to provide some control over deformation and cracking. Even if a member can support the design loads, it must not bend to such an extent that its function is impaired. By providing a reserve of strength, the designer recognizes that members may be subjected to loads greater than those assumed in design. For example, during construction, temporary storage of buikling materials may create forces in certain members well above those praduced by normal occupancy. Even after the building has been in use, itis still subject to overload if heavy equipment is introduced into an area that was designed for a smaller load. The extra strength also provides for the possibility that members may be constructed with a lower strength than specified because of understrength materials or poor workmanship. Although it is imperative that structures be designed with adequate strength to reduce the probability of failure to an acceptable level, they must also function effectively under normal service loads. Deflections must be limited to ensure that floors will remain level within required tolerances and do not vibrate, to prevent plaster ceilings and masonry partitions from cracking, and to ensure that sensitive equipment will not be thrown out of alignment. In addition, the width of cracks must be limited to preserve the architectural appearance of exposed surfaces and to protect reinforcement from attack by corrosion, 1.8 ELASTIC VERSUS STRENGTH D 1G ‘Two design approaches for sizing reinforced concrete members are available to the engineer. The first, called elastic design, is based on the prediction of stress in members as they support the anticipated service loads. Service load is the actual or maximum value of load the member is expected to carry, In elastic design, the members are sized so that service-load stresses do not exceed a prescribed, predetermined value of stress. The allowable stress is set as some fraction of the maximum stress that the material can sustain before rupture or yielding occurs. ‘The elastic-design approach requires an understanding of how the member behaves with the service load in place. It typically assumes that materials behave elastically and that stresses induced by loads can be accurately predicted. Elastic design does not take into consiceration the inherent feilure mode (ductile or brittle) of the member or account for the magnitude of the additional strength in a member between service-load capacity and the ultimate-load capacity. ‘Thus the actual factor of safety against failure is, in fact, unknown,The second design approach, called ultimate-strength design or more simply strength de- sign, is based on predicting the load that will produce failure in a member rather than predicting stresses produced by service loads. Using strength design, the designer is concerned with de- termining the load that will bring a structure to complete collapse and with the mode of failure when this load is applied. The preferred mode of failure is to ensure a controlled local failure of the member in a ductile rather than brittle manner. Thus the engineer who uses strength design ives only indirect attention to the state of stress or the deflection that will occur in the member when service loads are in place. Since the design of a ductile structure that will fil locally is a foremost concern in reinforced concrete design, strength design is considered the more desirable approach, By controlling the ultimate strength of each member of a structure, the designer can control the overall mode of failure of a total structural system. In this way it is possible to design structures so that in the unlikely event of unanticipated overload, failures are confined to a limited region instead of causing total collapse of the entire system. In addition, laboratory studies confirm that the controlling failure modes of concrete mem- bers can be accurately and consistently predicted. Although the failure mode can be predicted, the magnitude of the actual stresses in reinforced concrete members cannot be predicted for several reaso 1. Shrinkage stresses produced by the drying of plastic concrete induce a set of self-balancing stresses of unknown magnitude, 2, Under load, concrete cracks in erratic, unpredictable patterns so that the properties of the cross section are not known with certainty. 3. With time, reinforced concrete creeps. The creep causes stress intensities to change; typically stresses increase in the steel and decrease in the concrete. 1.9 THE ACI DESIGN PROCEDURE In strength design, which is the design method recommended by the current edition of the ACI Code, members are sized for factored loads that are greater than the service loads. The factored load is produced by multiplying the service load by load factors, numbers typically greater than 1. The size of the load factor, which represents part of the factor of safety applied to loads, reflects the accuracy with which the design loads can be predicted, A load whose magnitude and distribution can be established with certainty is increased by a smaller load factor than a load whose magnitude is subject to variation or whose exact intensity cannot be predicted with precision, For example, dead load, which can usually be computed very accurately, is increased by a load factor of 1.4; on the other hand, live load, which is subject to greater variation, is increased by a load factor of 1.7. Table 1.3 contains additional load factors for various types and combinations of load. ‘When used in the general sense, the factored load is often denoted by U.+ Factored loads or forces produced by factored loads are subscripted by a lowercase 1. For example, M, represents the moment at a section produced by factored loads. Using factored loads, the designer carries out an elastic analysis of the structure. Structures are usually analyzed both for full gravity loads and for wind in combination with reduced grav- ity loads. Example 1.1 illustrates the loading conditions that must be investigated for a typical building frame. The forces in a member created by factored loads represent the required strength of the member. Example 1.2 illustrates the more extensive computations required when the po- sition of the live load is not fixed. 7 INTRODUCTION8 TABLE 13 ACI load factors REINFORCED CONCRETE DESIGN Dead load D ....... 14 Live load L ........ 17 Earth pressure W... 1.7 Fluid pressure Fo... 14 Earthquake E ...... Substitute 1.1 £ for Win equations below ‘When the force in & member is due to a combination of wind load Win addition to dead and live loads, the factored load U is produced by considering the following combination of Toad factors: Factored load U = 0.75(1.4D + 1.7L + 1.7W) Where tke 0.75 reduction factor accounts forthe improbability of having maximum wind and live loacs acting simultaneously on the structure. Since the dead load is always present, the logic of applying the 0.75 factor to itis not clear. However, since most structures designed for wind by the above equation behave satisfactorily, there is little reason to modify the ex- pression If the absence of live load produces a reversal of stress or increases the likelihood of the structure overturning when the wind acs, the following load factors are used to establish the factored load u =09D+1.3W When the structural effects T of differential settlement, creep, shrinkage, or tempera- ture are to be considered, the factored load is established by U =0.751.4D+ 147 + 1.7L) _ butnotless than U = 14(D +7) Members are sized so that their design strength will be equal to or greater than the required strength. The design strength is a reduced value of the ultimate or nominal strength of the cross section. The nominal strength of a member is evaluated in accordance with provisions and as- sumptions specified by the ACI Code. Nominal strength is evaluated analytically by considering the state of siress associated with the particular mode of failure (cither the stee! yields or the concrete crushes) or experimentally by studies that relate the ultimate strength to the propor- tions of the cross section and the strength of the materials. Nominal strength is designated by the subscript. In order to account for inevitable losses in member strength due to imperfect workmanship, e.g., undersized members, bars placed out of position, or voids in the concrete and understrength materials, the nominal strength of a member is reduced by multiplying by a capacity reduction factor @, a namber less than 1, to give the design strengeh, Reduction factors are listed in Ta- ble 1.4. The magnitude of the reduction factor is also influenced by the ductility of the member, TABLE 1.4 Capacity reduction factors, Reduction factor Nominal strength o Bending with or without axial tension, 09 and for axial tension Shear and torsion 085 Bearing on coneret 00 Bending in plain concrete 0.65 Columns with spirals o.75¢ Columns with ties 0.704 {For members carrying moment and small values of axial toads (less than 0.1 £ Ay) the reduction factors vary linearly from 0.9 to 0.7 or to 0.75 for tied and spiral columns, respectively.the degree of accuracy with which the member's capacity can be predicted, and the importance of the member to the overall strength of the structure. For example, a flexural failure of a beam involves one member and produces a local failure, but failure of a column may result in the collapse of many floors. The reduction factor constitutes the second part of the factor of safety in strength design. Load factors and reduction factors have been selected for the ACI Code so that failure of structures will initiate in beams by yielding of the tension steel. Yielding of the tension steel in properly designed beams produces sagging and heavily cracked members but does not cause total collapse of the structure. Factors of safety against other less ductile modes of failure are ‘much higher. In summary, the design criteria of the strength method can be stated as Required strength = design strength ay or Required strength = (nominal strength) (2) The above criteria applied to a beam that is stressed only by shear and moment would state that at every section Ves $Vn and = My = My where V, and M,, the shear and the moment produced by factored loads, represent the required strength; and V,, and M,, stand for the nominal shear and the nominal flexural strengths at the same section. To clarify the fundamental ideas of strength design, a short steel column carrying an axial load that produces uniform stress on each cross section i sized in Example 1.3. Since the strength method of design is based on behavior at failure, it does not guarantee that behavior will also be satisfactory under service loads; therefore, the ACI Code has established additional criteria to ensure that members will also satisfy the requirements of serviceability. These requirements for beams will be discussed in Secs. 3.5 and 3.6. EXAMPLE 1.1. Computation of Required Strength. Determine the axial forces for which the ‘member CD in Fig. 1-1a should be designed given the following service loads: (a) dead load of 1 kip/f on girder BC, (b) live load of 2 kips/ft on girder BC, (c) wind load of 20 kips horizontally at joint B. Wind may actin either direction. See Table 1.3 for load factors. Solution. Since the frame is supported at D by a roller, only an axial force equal to Ry develops in ‘member CD. The force Ry may be computed by summing about support A moments of forces due to factored loads. CASE: L + D (see Fig. 1-10) we = 14D + LIL 1.4() + 1.7@) = 4.8kips/tt Analysis gives 48 kips compression in member CD. akin > c » c T Asin ny compression F DL 4h > st 3 ~ faa as ge FIGURE 1.1. (a) Frame; (®) Dead and live load. @ © (continued) 9 INTRODUCTION10 ‘REINFORCED CONCRETE Skips/ft e , Saipan 6p 255i cpl > canine} © @ FIGURE 1.1 (CONTINUED) (c) Dead, live, and wind loads create compression in DC; (d) Dead and wind; wind creates tension in DC. CASE 2: L + D+ W (wind induces compression, see Fig. 1.1) (0.75(1.4D + 1.7L) = 0.75{1.4(1) + 1.72)] = 3.6 kips/ft W, = 0.75(1.7W) = 0.75(1.7)20 kips) = 25.5 kips acts to right Analysis gives 48.75 kips compression in member CD. CASE 3: D+ W (wind induces tension, see Fig. 11d) W, = 0.9D = 0.9(1) = 0.9 kipsitt W, = 1.3W = 1.320 kips) = 26 kips acts to left Analysis gives 4 kips tension in member CD. CONCLUSION. Member CD must be designed with a compressive strength of 48.75 kips and a tensile strength of 4 kips. EXAMPLE 1.2. Computation of Nominal Strength for a Beam Supporting a Moving Load. ‘The beam in Fig. 1.2a carties service loads that consist of (1) a uniformly distributed dead load ws = 1.429 kips/t acting over its entire length and (2) a concentrated live load P; = 24 Kips that cean act anywhere on the span. Using the ACI load factors in Table 1.3 and the reduction factors in ‘Table 1.4 determine the required nominal flexural strength Mz, at both point B and at point C. Solution. Compute factored loads by multiplying service loads by the appropriate load factors Wa = L4wg = (1.4)1.429 = 2 kipsft Py = ATP) = 1.7024) = 40.8 kips 1, MAXIMUM POSITIVE MOMENT AT B. Apply factored loads to the beam (see Fig. 1.2b). To ‘maximize the positive moment at B, position the concentrated load at that point. Analysis of the beam, as shown by the shear and moment curves, establishes the required flexural strength M, = 291.5 ft- kips. To compute the nominal flexural strength M,, we set My = dM, — where @ = 0.9 291.5 = 09M, My = 323.9 ft- ips ‘To investigate the possibility of negative moment developing at B, we shift the concentrated load. to the tip af the cantilever—the position that creates maximum negative moment in span ABC. As shown in Fig. 1.2c, M, = —14.5 ft-kips at B, and accordingly M, = —16.11 ft-kips. Thus the beam. ‘must be designed to carry both positive and negative moment, that is, we must reinforce both the top ‘and bottom of the cross section with longitudinal steel 2, MAXIMUM NEGATIVE MOMENT AT C. Position the live load at D (see Fig. 1.2c). Analysis shows that at point 229 fi-kips. Setting My = &M,, we compute M, = ~254.44 ft-kips.TT] Ty + Trsasnips t stesnips tasstips t e225 ups as 508; 10 835 Shear, Vy c (ips) 19.5 41.65 “3145, 2915 {_lo"_} Moment, M, — (f-kips) - a We kips) 14s “209 ® © P= 408 kips = 2hips/fe FIGURE 1,2. (a) Beam dimensions, (4) dead load acts over entire span; live load positioned to maximize positive ‘moment at point B, (c) live load positioned to maximize negative moment at points # and C. EXAMPLE 1.3. Design of a Short Steel Column by the Strength Method. Determine the mini- mum cross-sectional area A of a shor steel column required to support a live load L = 133.44 kN (30 kkips) and a dead load D = 177.93 KN (40 kips); see Fig. 1.3. Failure is assumed to occur when the average stress on the cross section reaches the yield-point stress f,. Use a reduetion factor = 0.7 and load factors of 1.4 for dead and 1.7 for live load. [f, = 344.7 MPa (50 kips/in?),] Solution, Multiply the service loads by the respective load factors to produce Py, the failure load, ‘which also represents the required axial strength Py = LAD + 1.7L = 1.4(177.93) + 1.7(133.44) = 475.96 KN (107 kips)~ ‘The design strength 6, of the column in terms of the area ofthe cross section A and the yield strength of the steel f, is OP, = Af, = 0.7A(344,7) = 241.294 MPa (BSA kips) | ib LEXELEA Suess t Swain FIGURE 1.3 (a) Failure impends @ 0 (©) stress-strain curve for steel.n ENFORCED CONCRETE ‘To establish the required area of the cross section we equate the required strength P, to the design strength $P,, to give 475.96 KN = 241.294 MPa A = 1972.56 mm? (3.06 in?) Of course, when the full service load of 311.37 KN (70 kips) acts, the stress is well below f, and behavior is elastic, The service stresses are P__31137X10°N 5 7 Toresesc Tom 7 157.85 MPa 22.9 kipsin®) Go 1.10 OVERVIEW OF THE DESIGN PROCEDURE Before the analysis and design of individual elements (slabs, beams, and columns) of multistory structures are discussed, the overall design procedure from conception to final drawings will be outlined briefly. Although the following discussion focuses on building design, which tra- ditionally involves close coordination between the architect and engineer, the same procedure is applicable to the design of those structures, such as dams, water-storage tanks, and bridges, which are traditionally the exclusive province of the engineer; i. the engineer conceives of the structural form as well as performs the structural design and analysis. The notable bridges of Maillart or shells of Nervi are well-known examples of an engineer establishing a structural form that is much admired. Preliminary Design ‘The major functional requirements of a building, whether they are a complex mechanical support system for a rower plant or unbroken spaces for a basketball arena, largely determine how the architect arrives at a preliminary concept and a preliminary set of plans. The architect must also integrate the interior functional requirements with the character of the specific site chosen for the project, often making use of any natural features of the location that can contribute to the desirability ard therefore the value of the building. For example, if a site permits a view across valley ora harbor, the architect could take advantage of this feature and design a building that provides a panoramic view for as many occupants as possible. Conversely, if a building faces a blank wall, the architect will work to minimize this negative feature. The architect's preliminary plans will roughly indicate how the floor areas of the building will be divided. These layouts also establish ceiling heights and the location of walls, entrances, stairs, and elevators. Once the geometry of the building has been roughly defined by its main function and by the constraints of the site, the architect will consult with a structural engineer to establish the best-conceived structural solution to the architectural problem. During this preliminary design stage, the designer will investigate and weigh the best al- tematives. Considering all possiblities of materials, structural systems, relative costs, time con- straints, and availability of construction materials, the engineer searches for the structural system that will best suit the architectural scheme and will pick that system which represents 2 rational, efficient, and sound use of materials. The preferred relationship between the engineer and architect is well expressed by Siegel:* ‘The engineer ought to do more than merely make sure that what the architect designs stands up. . ‘The engineer should act as the architect's critical partner, objecting strenuously wherever the design offends structural logic. ...He should try to make the significance of the structure and its behavior clearer to the architect, so thatthe latter can draw upon [the engineer's] understanding for inspiration, in working out the final design.Final Design ‘After the architect has incorporated the details of the structural system into the architectural drawings, and after the electrical and mechanical engineers have established the location of openings in floors and beams for pipes, conduit, ducts, and other building services, the struc- ture and the design loads are clearly defined. In the final design phase the structural engineer will analyze and design each component of the structure for all possible loading conditions. All information—member size, reinforcement, construction joints, floor slopes, and all other details—required to build the structure will then be incorporated into the structural drawings 2B INTRODUCTION and specifications. REFER! NCES 1, Basic Building Code, Building Officials and Code Administrators International, Chicago, 1978, 2. A. Popoft, Jr. “Design against Progressive Collapse,” Prestressed Concer. Inst, vel. 20, no. 2, p. 4, March-April 1975. 3, Cust Siegel: Structure and Form in Modern Architecture. Reinhold, New York, 1962, QUESTIONS+ 1 1.2, What is ductility and why is itimportant? 1.3. What two criteria form the basis for proportioning struc- tural members? |. Why is concrete design termed empirical? 1.4, Whats a progressive collapse? How can the likelihood of such a failure be reduced? 15. Define catenary action. 1.6. Why is strength design considered more desirable than clastic design? 1.7. What is the major deficiency of elastic design? PROBLEMS 1.1, The beam in Fig. P11 carries uniform service loads of wg = 09 kip/ft and wy = 0.6 kip/ft. The uniformly distributed dead load acts continuously; but the concen- trated and the distributed live loads may or may not be P= 32 kips L. Saar eden sven FIGURE P11 @ i A=22kips i E 1.8. What is a load factor? How is a factored load used? 1.9. What is a reduction factor? What determines its size? 1.10. Define nominal strength, design strength, and required strength 1.11. {If suength design is used to size the members of mul- tistory buildings, which members, columns or beams, will fil first ifthe structure is overloaded? Explain, 1.12 Why isthe load factor greater for soil pressure than for fluid pressure? present. Concentrated live loads of 32 kips also act, termittently at the ends of the cantilevers, Determi the required flexural design strength at midspan and at support B. Hint: Different loading conditions pro- duce maximum positive and negative moments at mid- span. 1.2. The frame in Fig. P1.2 (on next page) carries service loads that consist of a wind load W as well as a uni- formly distributed dead and live load on girder BC. At the section of maximum moment in column AB, deter- ‘mine the required flexural and axial strength when all loads act simultaneously. ‘Answers to questions and selected problems are given atthe end of the book.14 rexPoRceD CONCRETE DESIGN FIGURE P12 1.3. The beam in Fig. PI.3 carries a uniformly distributed dead load and two concentrated loads, The dead load is always present, but the Tive loads may act separately or simultaneously. Considering pater loading, determine the required flexural strength M, at section B located at the center of span AC. Also determine the required shear strength to the right of support. P.=10kips ea = 05 kips/fe = 20kips Ise FIGURE P13 1.4, Girder ABC of the rigid frame shown in Fig. P1.4 sup- ports a uniform dead load of 600 Ib/ft over its entire length. During a snowstorm a maximum live load of 500 tb/ft can act over all ot a portion of girder ABC. In addition, when the wind biows from left to right, it creates the equivalent of a lateral force of 4 kips at ry Bc 208 D fas np By @ FIGURE PL4 (a) Frame, (b) forces due to wind only. joint A and an uplift of 300 Ib/ft over the entire length of girder ABC (see Fig. P1.4b). Wind from the oppo- site direction need not be considered. Ifthe dimensions of the girder are controlled by moment, determine the value of the required flexural strength Mf, at the section of maximum moment, Assume that the snow and winds loads are not applied simultaneously; that is, consider separately D + Land D + wind, 1.8. In addition to its own weight, the beam in Fig. PLS ‘must carry a 400-KN concentrated service live load that ccan act anywhere on the span. Determine the required flexural strength M, at points B and C, with w. 2400 kg/m’ P= 400KN Fl peer 4 m—s—4 m- FIGURE PLSConcrete is being placed imo forms during construction of floor slab (Photograph by Portland Cement Association.) 2.1 CONCRETE Concrete is a carefully proportioned mixture of cement, water, fine aggregate, and coarse ag- ‘gregate, To these basic components a variety of admixtures, i.e., chemicals that influence the reaction or modify the physical properties of the hardened concrete, are frequently added. As soon as the components of conerete have been mixed together, the cement and the water react to produce a cementing gel that bonds the fine and the coarse aggregates into a stonelike ma- terial. The chemical reaction between the cement and water, an exothermic reaction producing significant quantities of heat, is termed hydration. During the initial stages of hydration, when only small amounts of gel have formed, the ‘concrete is ina plastic state and flows easily. Throughout this stage the concrete can be deposited in forms, worked into the spaces between reinforcement, and compacted to climinate voids, and the surfaces can be leveled and finished. As hydration continues and larger amounts of gel form, the concrete progressively stiffens, loses its workability, and gains strength. Z CHAPTER MATERIALS 1516 REINFORCED CONCRETE DESIGN FIGURE 2.1 Section through conerete showing aggregate. (Portland Cement Association.) ‘The rate at which the gel forms is influenced by the temperature at which the reaction occurs. At high temperatures the reaction is rapid: only 10 or 15 min may be required for the concrete to stiffen. At low temperatures 10 to 12 hours may be required to produce the same degree of stiffness. The final product (Fig. 2.1) consists of the various-sized aggregates surrounded by a mortar compesed of cement and the fine aggregate. Ideally, engineers would like to produce a dense homogeneous concrete that is free of voids, channels, cracks, and other defects. Such ‘a material would have uniform strength properties, be impermeable to water, and present an at- tractive surface appearance. As a practical matter, however, i is not possible to produce concrete that is free of laws: even the most carefully controlled concretes have a variety of small internal and surface defects. Voids are created by air bubbles produced during the mixing process and by segregation of the coarse aggregate from the mortar when concrete is deposited in forms. Since approximately twice as much water is needed for workability than is required to react with the cement, fine channels develop throughout the concrete as excess water rises to the surface when the concrete is vibrated. The rise of the water to the surface is termed bleeding. Surface defects such as pits and honeycomb (the absence of mortar between aggregates) develop when the concrete contains insufficient mortar or when the concrete is not properly compacted in forms. In addition, x-ray and microscopic studies! show that fine microcracks develop as a result of the rupture of the bond between the gel and the aggregate. ‘The break in bond is primarily attributed to drying and carbonation shrinkage (shrinkage produced by exposure to CO). The presence of these internal flaws explains in part the large difference between the tensile and the compressive strengths of concrete. In addition to these small unavoidable defects, with time, restrained concrete members may develop shrinkage and temperature cracks due to fluctuating environmental conditions. Although these cracks can never be eliminated, careful attention to design details and the use of reinforcement, called temperature and shrinkage steel, will control the size and location of ‘cracks so that neither the function nor the appearance of the structure will be impaired. Concrete Materials Most concrete produced in this country is made with portland cements, Termed hydraulic ce- ‘ments because they react with water, they have the ability to harden under water. Portland ce~ ments are available in five ASTM types (Table 2.1). They are manufactured from the same basic materials, but their properties are altered somewhat by the way the raw materials, often clays and limestones, are blended and fired before they are ground into cement powder. Aggregates, which constitute approximately 75 percent of the concrete volume, are usually composed of well-graded gravel or crushed stone. Aggregates passing ano. 4 sieve [the wires ofTABLE 21 Types of portland cement "7 Type Name Use or characteristics of cement NATTRATS Normal General-purpose; used where no special requirements exist 0 Moderate Moderate resistance to sulfate attack and moderate evolution of heat during hydration TM Hiighearly Rapid gain in strength; permits quicker removal of forms; speeds construction IV Low-heat Low heat generated: used in construction of massive sections, eg. dams v Sulfate-resisting Used where concrete is exposed to high concentrations of sulfates the screen are spaced } in (6.4 mm) apart] are classified as ine aggregates; larger aggregates are classified as coarse aggregates. Since the presence of dust or chemicals can weaken the bond between the cementing gel and the aggregates, strong, durable concretes require aggregates that are clean and hard. Unreinforced coneretes made with stone or gravel are normally assumed to weigh 145 Ib/ft3 (2320 kg/m’), When reinforcement is added, the weight of concrete and ste! is generally taken as 150 Ib/ft® (2400 kg/m") for normal-weight aggregates. If the designer wishes to reduce the weight of concrete members that are not subjected to heavy wear or abrasion, lightweight aggregates, made from a variety of expanded shales and slags, can be substituted for the heavier stone or gravel aggregates. Coneretes that have a 28- day strength equal to or greater than 2500 Ib/in? (17 MPa) and an air-dry weight less than 115 lb/ft (1840 kg/m?) are termed structural lightweight concrete? If all aggregates (both the fine and the coarse) are lightweight, the concrete is called all-lightweight. If the coarse aggregate is lightweight and the fine aggregate is sand, the conerete is termed sand-lightweight concrete. Unless field practice verifies that concrete with a certain size aggregate can be success- fully placed and compacted to prevent honeycomb or voids, ACI Code §3.3.2 requires that the maximum size of coarse aggregate not exceed 1. One-fifth of the smallest dimension of the form 2. One-third the depth of slabs 3. Three-fourths of the minimum clear spacing between reinforcement Strength of Conerete The water-cement ratio is the major variable influencing the strength and durability of concrete. ‘When the water-cement ratio is large, a dilute, high-shrinking, weak gel is produced. Concrete containing such gels is low in strength and lacks resistance to deterioration by weathering. The variation of the compressive strength f, with the water-cement ratio is shown in Fig. 2.2. The rate of strength gain in concrete can be speeded by using finely ground cement, applying heat, e.g. steam curing, or by adding an accelerator such as calcium chloride. A rapid increase in strength of freshly poured conerete permits early removal of forms and speeds construction. If freshly poured concrete is frozen, the free water present in the pores transforms into ice, producing a volume expansion that causes a significant breakup in the structure of the concrete and permanently reduces its ultimate strength. Where concrete is exposed to salt water or cycles of freezing and thawing, air-entraining agents are typically added to entrain small percent (4 to 8% by volume) of air in the form of small bubbles throughout the concrete mass. This entrapped air produces a major increase in the concrete’s durability. In addition, the air bubbles serve as a lubricant to make the concrete more workable.18 ‘REINFORCED CONCRETE soo a7 ‘ K 000 Nonsiremined | 04 55 soon |_——-Aietained — nose £ 2000 B79 FIGURE 2.2 Approximate 1000 689 variation of 28 o| | a & Compression 02 oa 06, os Compression FIGURE 2.9 Biaxial strength of concrete, (Adapted from Ref. 5.) 234 REINFORCED CONCRETE Desicn Spcincs nic atl Rs fee Gs Creep I Elastic recovery | Load removed g | é re c 5 ci > a Seem lem FIGURE 2.10 Elastic and creep deformations of concrete. (Adapted from Ref. 3.) Creep Experimental studies* show that both axial and bending deformations of reinforced concrete members increase with time, The total deformation is often divided into two parts: (1) an it tial, instantaneous deformation that occurs with the application of load, followed by (2) a time- dependent deformation, termed creep, that continues at a decreasing rate for a period of years. To illustrate the effect of creep, the variation of axial deformation over time for a uniformly loaded conerete cylinder is shown graphically in Fig. 2.10, which shows that after the initial deformation has taken place, the creep deformation occurs rapidly at first and then decreases steadily with time. Approximately 75 percent of the ultimate creep deformation occurs during the first year, Although measurements indicate that creep goes on for many years, itis essentially completed after 3 or 4 years. If the load is removed (see the dashed curve BC), immediate elas- tic rebound occurs, followed by a modest increment of creep recovery. However, a permanent deformation remains, denoted by the distance between points D and E. Although creep under the normal levels of stress produced by service loads does net reduce the strength of a reinforced conerete member, it frequently influences behavior. The following ‘cases illustrate how creep affects behavior and how its undesirable effects can be minimized. Case 1. Because of the delayed effects of creep, the long-term deflections of a beam may be 2.or 3 times larger than the initial deflectio»s. ‘Thus the long-term deflections of both reinforced and prestressed concrete beams must be ‘vestigated by the designer to ensure that the struc- ture will remain serviceable and that the a=vcipated creep-induced deflections will fall within accepted limits. Case 2. In prestressed concrete construction, members are often stressed by anchoring heavily tensioned cables, which frequently pass through ducts in the member, to the ends of members. As the compressed concreie shortens because of creep, the force in the cable reduces as the ends of the member move toward each other. Loss of prestress can lead to both increased cracking and larger deffections under service loads, which can be prevented by using high- strength steel in which final strains—even after creep losses (as well as other losses)—are still significant.1600 1400 200 cep in compression, millions & Water cement ratio 0.62 by weight Uni FIGURE 2.11 Rate of creep as 0 100200 307407500600 affected by the intensity of sustained Time after application of oad, days load. (Adapted from Ref. 4.) Case 3. In a reinforced concrete column supporting a constant load, the deformations due to creep can cause the initial stress in the steel (o double or triple with time, assuming that the steel remains elastic. As the creep deformation of the concrete takes place over time, the force carried by the noncreeping steel, which is fully bonded to the concrete, gradually increases while the total force carried by the concrete reduces by an equal amount. Since experimental studies have shown that elastic design underestimates the stress in the steel, the design of axially loaded reinforced concrete columns has been based on the strength of the section rather than on limiting the level of stress in the materials to specified allowable values. Similarly, when reinforcement is added to the compression zone of a beam to increase bending capacity, an empirical adjustment must be made to account for the additional stresses induced over time in the steel by creep. Although creep deflections are often undesirable, creep is beneficial when a concrete mem- ber is heavily stressed in compression locally. As the load is applied and stresses build up, rapid creep reduces the highest stresses and increases the stresses in more lightly stressed adjacent sections, thereby relieving high local stress concentrations that might produce failure. As long as stresses in the concrete do not exceed approximately 50 percent of the concrete strength, studies indicate that creep is nearly proportional to the magnitude of the stress (see Fig. 2.11), The ultimate value of creep is also a function of the type of aggregate, the strength of the concrete, the method of curing, the age of the concrete when initially loaded, the surface- to-volume ratio of the member, and the relative humidity. Creep can be reduced by . Deferring the application of load until the concrete gains . Using high-strength concrete strength . Keeping the volume of cement paste low relative to the volume of the aggregate 1 2. 3 4, Steam curing under pressure 5. Adding reinforcement 6 . Using limestone aggregates Shrinkage In order to produce a workable concrete mix that will flow readily between reinforcement to fill all parts of the formwork and can be effectively compacted to eliminate voids, nearly twice as much water as is theoretically required to hydrate the cement must be added to the concrete mix. After the concrete has been cured and begins to dry, the excessive water that has not reacted 2526 ENFORCED CONCRETE DEsioN FIGURE 2.12 The interrelation of 200250300 350400450 shrinkage, cement content, and water ‘Water content per unit volume of sonerete 1b/yd? ‘content. (Adapted from Ref. 3.) with the cement will begin to migrate from the interior of the concrete mass to the surface, As the moisture evaporates, the concrete volume shrinks. The loss of moisture from the concrete varies with distance from the surface. Drying occurs most rapidly near the surface because of the short distance the water must travel to escape and more slowly from the interior because of the increased distance from the surface. The shortening per unit length associated with the reduction in volume due to moisture loss is termed the shrinkage strain or shrinkage. ‘The magnitude of the ultimate shrinkage strain is primarily a function of both the initial water contert of the concrete and the relative humidity of the surrounding environment. As in- dicated in Fig. 2.12, a nearly linear relationship exists between the magnitude of the shrinkage and the water content of the mix for a particular value of relative humidity. Ifthe relative humid ity is increased, the shrinkage of the concrete drops. When concrete is exposed to 100 percent relative humidity or submerged in water, it will actually increase in volume slightly zs the gel continues to form because of the ideal conditions for hydration. Both the rate at which shrinkage occurs and the magnitude of the total shrinkage (a function of the member’s size and shape) increase as the ratio of surface to volume increases. The larger the surface area relative to the cross section, the more rapidly moisture can evaporate, For un- reinforced concrete members exposed to average values of relative humidity (about 70 percent in the United States) ultimate values of shrinkage strain reach 0.0004 to 0.0007. If members are reinforced, the resistance to shrinkage deformation provided by the steel reduces the ultimate shrinkage strain for standard conditions to values* between 0.0002 and 0.0003. ‘Although shrinkage continues for many years, particularly for thick members, approxi- mately 90% of the ultimate shrinkage occurs during the first year. Typical plots of shrinkage over time for normal and lightweight coneretes are shown in Fig. 2.13, indicating that shrink- age is higher for lightweight concretes than for normal-weight concretes. The increased shrink age is caused by the initial swelling of the more porous lightweight ageregates, which absorb somewhat more water during concrete mixing. As the concrete dries, the larger shrinkage strains reflect the combined shrinkage of the aggregates and of the cement paste. Stresses Created by Shrinkage Because concrete adjacent to the surface of a member dries more rapidly than concrete in the interior, shrinkage strains are initially larger near the surface than in the interior. As a result of the differential shrinkage, a set of internal self-balancing forces, ic., compression in the interior and tension on the outside, is set up in the concrete. The approximate distribution of the longitudinalLightweight concrete 0.0004 Normal-weight concrete Decrease in length, iin 0 30 100 130 200 250 300880 Time, days FIGURE 2.13 Rate of shrinkage of lightweight and normal-weight concrete.® Aer shrinkage pil dimension = i ort om Hf 0 @ ” © FIGURE 2.14 Longitudinal stresses created by differential shrinkage: (@ stresses in a concrete cylinder; (b) shear stresses required for compatibility; (c) free-body diagrams of the upper half of the cylinder. stresses associated with these factors is shown in Fig. 2.14a, The stresses induced by shrinkage can be explained by imagining that the cylindrical core of a concrete cylinder is separated from its outer shell and that the two sections are then free to shrink independently in proportion to their existing water content (see Fig. 2.14). Since deflections must be compatible at the junction between the core and the shell, shear stresses must be created between the core and the outer shell. If free-body diagrams of the upper half of the cylinder are considered (Fig. 2.14), it is clear that vertical equilibrium requires the shear stresses to induce compression in the core and tension in the exterior shell. With time, as the interior dries and the shrinkage strains become ‘uniform throughout the cross section, these stresses will dissipate. In addition to the self-balancing stresses set up by differential shrinkage, the overall shrink- age creates stresses if members are restrained in the direction in which shrinkage occurs. For example, consider a beam that is a component of a continuous frame. If the beam were free to shrink after it was poured, its length would reduce to L’ (see Fig. 2.15), which is less than & eo TLE —————_1————} FIGURE 2.15 Tension created in @ t as restrained member by shrinkage. 27 MATERIALS2B Desi Compressible pads. Pour? Expansion oi iter cach ae ——= [ mat _4 ‘ Dovel ne en greased Dowel to prevent bond Section A @ ® © FIGURE 2.16 Details of (a) construction and (6) expansion joints, (c) square dowel with compressible side pads. the distance L between column faces. Since the beam is connected rigidly to the joints, tensile forces will be induced in the beam by the joint to resist the shortening. Because these stresses normally exceed the tensile strength of the concrete, random cracking develops. This cracking, which initiates at points of local weakness, will occur along the length of the beam. Tensile cracking due to shrinkage will take place in any structural element (wall, a slab on grade, or a framed slab) restrained at its boundaries. It must be controlled since it pe-mits the Passage of water, is detrimental to appearance, reduces shear strength, and exposes the rein- forcement to the atmosphere. To minimize cracking, the designer should act to Minimize the water content. Use dense nonporous aggregates. Cure the concrete well, Limit the area or length of concrete poured ata given time. (If walls or floor slabs are poured in small sections, a portion of the shrinkage will occur in the segments poured and will reduce the overall shrinkage.) Use construction and expansion joints to control the location of cracks. (Figure 2.16 shows two commonly used joint details for walls and slabs. If construction joints are used to reduce shrinkage in beams or framed slabs, they should be located in regions of low shear near midspan.)’ To reduce shrinkage cracking created in slabs on grade, square steel dowels with compressible pads bonded to the sides are recommended for expansion Joints (see Fig. 2.16c), ‘Add reinforcement to limit the width of cracks. A well-distributed grid of bars or welded- wire mesh, called shrinkage steel, reduces crack width and results in the development of ‘many small hairline cracks that are often invisible to the naked eye. ‘Temperature Change Concrete expands as the temperature increases and contracts as the temperature decreases. If a ‘member is restrained, a drop in temperature produces tensile stresses and an increase in tem- perature produces compression stresses. The effects of temperature change are handled very much like those of shrinkage. The designer must either make provision for the deformations to occur so that stresses are not induced in the structure, or the members must be designed for the induced forces. The coefficient of temperature expansion for concrete is equal approximately to 5.5 X 10° in/iny“F, a value only slightly less than that of steel.* High-Strength Concrete During the past twenty years, advances in materials technology have enabled concrete man- ufacturers to steadily raise the compressive strength of concrete that can be delivered to thejob site. For example, in the 1950s, structural designers knowledgeable about construction prac tices in the field typically believed that contractors could not reliably produce concrete with a compressive strength greater than 4000 or 5000 psi (28 or 35 MPa). Concretes of greater strength—produced without admixtures—were often stiff’ and difficult to place and compact in forms, However, in 1989 concrete with a 56-day compressive strength of 14,000 psi (97 MPa) was used to construct the heavily loaded lower sections of the columns in the high-rise Chicago Mercantile Exchange building.° Currently (in 1997) the record for the highest strength ready-mix concrete delivered to a construction site is approximately 19,000 psi (130 MPa). This concrete was used to construct the lower columns of the Two Union Square building in Seattle, Washington. Workable high-strength concretes [i.e., those with a 28-day compressive strength of 10,000 psi (69 MPa) or more} are now routinely produced by adding to portland-cement concretes admixtures of silica fume (a finely divided form of SiO.) and high-range water reducers, also called superplasticizers. Superplasticizers are sulfonated melamine formaldehyde or sulfonated napthaline formaldehyde condensates that act as surface-active agents. These compounds sharply reduce the amount of water required to produce a workable mix so that concretes with water—cement ratios on the order of 0.25 or less flow easily but are not subject to exces- sive bleeding or segregation, Although high-strength coneretes have been used to construct prestressed concrete beams and certain types of off-shore structures, the primary use of this ‘material is for the construction of columns supporting large axial compression forces. Members of this type particularly are able to utilize most efficiently concretes with high compressive strengths. Although a 14,000-psi (97-MPa) concrete costs approximately three times as much as a 3000-psi (20-MPa) concrete, its compressive strength is nearly five times greater; thus itis economical. Further, high-strength concrete permits the designer to reduce the cross-sectional area of columns thereby increasing the rentable floor area. Uniaxial compression tests using standard cylinders show that high-strength concretes are ‘more brittle than lower strength concretes. Upon reaching a strain of approximately 0,003, cylin- ders of these concretes disintegrate. Although the modulus of elasticity of high-strength con- crete is greater than that of the lower strength coneretes, tests indicate that the current ACI equation for modulus of elasticity, which is based on testing concrete cylinders whose compres- sive strengths did not exceed 6,000 psi, overestimates the magnitude of this property. Based on experimental studies at Comell University, Carrasquillo, Nilson, and Slate" recommend that the ‘modulus of elasticity of normal weight high-strength concretes with f values between 3,000 to 12,000 tb/in? be predicted by the following equation 15 Ex, = 40,000 Vf + 1.0 (is) 4) where E, and /7! are expressed in Ib/in? and win Ib/ft, Summary The discussion of concrete properties in this chapter should provide the designer with the back- ground required to prepare the concrete specifications that are a part of each project. For a broader treatment of the properties of concrete see Ref. 11. Although care is obviously required at each step in the manufacture and placing of concrete, attention to curing is probably the single most important requirement for the production of strong, durable, watertight concrete. 2.3 REINFORCEM! The low tensile strength of unreinforced concrete, a brittle material, results in limited struc~ tural applications since most structural elements carry forces, e.g., shear and moment, that create 2930 REINFORCED CONCRETE tensile stresses of significant magnitude. The addition of high-strength ductile reinforcement that bonds strongly to concrete produces a tough ductile material capable of transmitting tension and suitable for constructing many types of structural elements, e.g., slabs, beams, and columns, Although a variety of materials such as glass fibers and plastic filaments have been used as reinforcement, most concrete members are reinforced with steel, in the form of bars, wire mesh, of strand, Thanks to its high strength, ductility, and stiffness, steel reinforcement imparts great strength and toughness to concrete. Reinforcement also reduces creep and minimizes the width of cracks. Slender reinforcing bars are easily fabricated into a variety of shapes that can be used to construct members of any configuration. Characteristics of Reinforcing Bars ‘The largest volume of steel reinforcement is supplied in the form of circular bars whose surface has been imprinted with protruding ribs to improve the bond between the steel and concrete. Several examples of deformation patterns are shown in Fig. 2.17. Although manufacturers are hot restricted to a specific pattern of ribs, the height, length, shape, and spacing of the ribs are controlled by ASTM specifications, Deformed bars are manufactured in diameters from 3 to 2} inches (metric bars from 11.3 to 56.4 mm) Bar size is designated by the number of eighths of an inch contained in the bar diameter, For example, a j-in-diameter bar, which has six £-inches in its diameter, is a no.6 bar. The standard bars and their properties are shown in Table 2.2 (metric bars in Table 2.4) Metric bars are designated by their diameter in millimeters rounded to the nearest multiple of 5. For example, a no. 45 bar has a diameter of 43.7 mm. Areas of groups of standard bars are tabulated in Table B.1 in Appendix B. Bars are supplied in lengths up to 60 ft (18 m). Bars of greater length would bend excessively when picked up because of their flexibility. Although most bars are rolled from new billet steel, bars are also fabricated from worn rails of train tracks or from axles of old locomotives, termed rail and axle steel, respectively. The properties of bars made from various ASTM grades of steel are listed in Table 2.3. As might be expected, rail and axle steels, heavily cold-worked by the repeated impact of train wheels, are the least ductile. If reinforcement must be welded, low-alloy steels, which have a limit on carbon content and other embrittling elements, are specified. Since the cost of stee! with a yield point of 60 kips/in? (414 MPa) is about the same as that of. steel with a yield point of 40 kips/in? (276 MPa), the 60-yield-point steel is most commonly used in the manufacture of reinforcing bars. For certain structures in which stresses are to be limited to reduce crack width orto limit deflections, the lower-yield-point st els may be specified because of their increased ductility. While the majority of reinforced cor erete structures are designed using reinforcing bars with a yield strength of 60 kips/in? (400 MPa), ACI Code §9.4 permits reinforcement with yield strengths as high as 80,000 Ib/in? (550 MPa). This limit does rot apply to prestressed tendons. ‘To identify the characteristics of reinforcing bars, all bars are stamped with a code that identifies the type of steel, the bar size, the yield point, and the manufacturer (Fig. 2.18). Although suppliers producing the nine sizes of reinforcing bars rolled to metric dimensions are currently limited in the United States, in future years we can anticipate that metric bars will eventually replace the eleven bar sizes shown in Table 2.2. Properties of metric bars and the strength and availability of the various grades of steel are given in Tables 2.4 and 2.5. Physical Properties ‘The strength and stiffiness of reinforcing bars are determined from a stress-strain curve of a reinforcing bar specimen loaded to failure in uniaxial tension. For steels whose yield point does not exceed 60 kips/in? (414 MPa), the stress-strain curve is assumed to be composed of twoFIGURE 2.17 Examples of deformed bars.!? (Concrete Reinforcing Steel Institute.) TABLE 22 Properties of standard U.S. reinforcing bars Nominal Cross-sectional diameter Perimeter Nominal weight Bar ———. — not in mm in mm Tht —_ke/m 3 0375 9.52 ou 7 Lig 299 0376 0.560 4 0500 12.70 020 129 1571 399 0681 1.013 5 0.625 15.88 031 200 1963 49.9 1.043 1.552 6 0.750 19.05, 044 284 2356 508 1502 2.235 7 0875 22.22 060 387 2749 698 20044 3.042 8 1,000 25.40 079 510 3142 798 2670 3973 9 1128 28.65 1.00 645 3544 900 3.400 5.060 10 1270 32.26 127 819 3.990 101.4 4303 6404 n 1410 35.81 156 1006 44301125 5313 7.907 4 1.693 43,00 225 1452 532 135.1 7650 1138 18 2257 97.33 4.00 2581 7.09 180.1 13.600 20.24 ‘Based on the numberof eighths of an inch included in the nominal diameter of the bas. BLE 23 Physical requirements of standard ASTM deformed reinforcing bars Minimum yield Minimum point or yield Tyosalsted J tensile strength strength and ASTM, Bar size Bem ore mera specification no, no. Grade __kipvin? MPa __kipsin? MPa Billet, A615-81 40 70 482.7 40 2158 oo 9 6206 60 4137 Rail, A616-79 50 80 5516 so 3447 0 90 6206 6 4137 Axle, A617-79 40 70 482.7 40 2758 oo 90 6206 6 4137 Low.alloy, A706-80 oo 80 5516 6 4137 ‘Bars marked with the symbol S satisfy more restrictive supplementary requirements of ASTM 615-79; ‘other bars marked with an N, Source: Adapted from Ref. 12. 3132 ‘REINFORCED CONCRETE ¢ ¢ GRADE 40 GRADE 50 Main Ribs Letter or Symbol for Producing Mill Bar Size #6 One Line ‘Type Steel ‘New Billet) "per SD, for New Billet for Supplemental Requirements A615 for Axle for Rail for Low Alloy ghee @ symbol for, Producing Mill Bar Size #11 ‘Type Stee! (per st) (-new billed (W-low alloy) FIGURE 2.18 Identification marks rolled into the surface of standard bars to denote the producer's mill, bar size, type of steel, and yield strength: Yield Designation (a) types of steel; (8) grade GRADES GRADE6O marks by line or number AND A706 AND A706 system. (Adapted from Ref. 12.) TABLE 2.4 Properties of ASTM metric rebars Nominal Cross-sectional diameter rea Perimeter Nominal weight Bar no. mm in mm in kg/m Ibyft 10 0.44 100 0.155 35 14 079 053 15 0.66, 200 031 503 1.98 156 1.05 20 077 300 0.465 613241 235 158 25 0.99 500 0.775 792 3.12 393 264 30 118 700 1.08) 9393 37 548 3.68 35 138 1000 1.55 1122 4.42 786 5.28 45 172 1500 2.32 1373 541 1787.92 58 222 2500 3.87 1772 698 1964 132 TABLE 2.5 ASTM specification AGISM Billet steel AG1OM Rail steel 617M Axle steel ATO6M Low-alloy Physical requirements of ASTM metric reinforcing bars Grade minimum 1d point, f, Bar size a ‘ipo? 10, 15,20 435 10-55 38 35,45, 55 ns 10-35 350 508 10-35 400 38 10-35 300 25 1035 +400 38 10-55 400 38Start of strain hardening Z Experimental curve & —— ts Tetizedseseatain cone 4 ¢ psi] MPa & 00 kisi? (200 GPa) ao | 276 | cose jai range so | 345 | cooi72 co io | aie | ‘00207 | 4s | 517 | 00259 ote — Plas range ° Strain FIGURE 2.19 Idealized stress-strain curve for reinforcing bars. Srrese-strain curve h siress FIGURE 2.20 Stress-strain curve of a high-strength steel Sirin that lacks @ horizontal yield range. straight lines (see Fig. 2.19). This assumption neglects the curvature of the stress-strain curve near the yield point and the slope of the strain-hardening region (see the dashed lines in Fig. 2.19). The modulus of elasticity of the steel E, (the slope of the stress-strain curve in the elastic region) measures 29,000 kips/in? (200 GPa) for all grades of steel. For higher-yield-point steels, the stress-strain curve may not develop a horizontal yield region, continuing instead to slope ‘upward at a reduced slope when the steel is strained beyond the elastic region. For stress-strain curves of this shape the yield point is defined as the stress associated with a strain of 0.35 percent (ACI Code §3.5.3.2) (see Fig. 2.20). ‘ACI Code §9.4 permits the use of reinforcing bars with yield points as high as 80 kips/in? (550 MPa). When the yield point of a steel exceeds 60 kips/in? (400 MPa), its properties must satisfy one of the provisions of ACI Code §3.5.3.1 and §3.5.3.2. Fabrication of Bars Once detailed plans and specifications for a reinforced concrete structure have been prepared, the builder will contract with a steel fabricator to supply the reinforcing bars, which must be cut and bent to fit inside the forms. Fabricators are normally able to deliver bars to the site within several weeks. Asa first step in the fabrication of the reinforcing steel, detailed drawings (called shop drawings) are prepared of every reinforcing bar that will go into the structure. Examples of common bar configurations produced by fabricators are shown in Fig. 2.21. After the design engineer has reviewed and approved the fabricator’s shop drawings, the bars are cut and bent to the specified dimensions. The fabricated bars, tagged to identify their location in the structure, are then shipped to the construction site in bundles. To ensure that bars will not be knocked out of position when concrete is poured into forms and vibrated, ironworkers tie the reinforcement together securely with steel wire to form rigid[REINFORCED CONCRETE DESIGN FIGURE 2.21 Typical bar bends. (Adapted from Ref. 12.) 1 sand 2n heights in $- 7 and 10+ engths 11 $+.2:,and over 2-10 5-in heights ss in increments of} in ia lengths of 5 ft PUL ttm Lina © Nn 2-10 15-in heights in increments of w FIGURE 2.22 Examples of standard bar supports: @) slab bolster, (b) beam bolster (¢) individual bar chait, (d) individual high chair, (Adapted from Ref. 12.) ‘mats or cages, which are positioned and secured in the forms with a variety of steel accessories. Figure 2.22 shows several types of supports for reinforcement in slabs and beams. If bars are exposed to rain, dew, or snow, rust develops rapidly on the surface of the bar. ‘Normally the rusting is light and need not be removed (in fact, light ust, which pits and roughens the reinforcement surface, improves bond). However, if steel is exposed to moisture for long periods, heavy surface rust that produces flaking may develop. To ensure a proper bond the excess rust must be removed. Before concrete is poured, the reinforcement is given a final inspection by the design engi- neer or the field inspector to make certain that the steel is positioned as required by the plans and specifications. This final check is very important. Many structural failures can be traced to the omission or incorrect positioning of reinforcement. Epoxy-Coated Bars Reinforced concrete structures (¢.g., bridge decks, marine structures, floors of parking garages, water treatment plants, and so forth) exposed to deicing salts, seawater, or other fluids containing chlorides may deteriorate rapidly—particularly if the reinforcement corrodes. Since the oxides of steel produced by corrosion occupy a greater volume (estimates indicate as much as 600 per- cent) than the original steel, they exert lateral pressures that can produce cracking and spalling of the concrete surrounding the bars. As the concrete deteriorates over time, the concrete encase- ‘ment that protects the reinforcement is no longer effective and the rate of corrosion of the steel increases. With time the reinforcement will debond from the concrete weakening the structure and increasing the likelihood of a structural failure. To reduce corrosion, reinforcement can be protected by 2oating its surface with an epoxy resin. Although the epoxy coating increases costs‘moderately and bars must be handled with care during construction to prevent flaking or cracking of the epoxy coating, the service life of a structure may be significantly extended and mainte- nance costs reduced. At the present time, the results of studies on the effectiveness of epoxy coating in reducing corrosion of bars in chloride-contaminated concrete are inconclusive.!? On the one hand, certain studies indicate the life of a bridge deck can be extended from approxi- ‘mately 5 to as long as 50 years by using epoxy-coated bars instead of uncoated bars. On the other hand, if the epoxy coating is pitted, has flaked off, or has worn off, the ability of the coating to resist corrosion is compromised and it will have little or no effect in reducing corro: ‘As we will discuss in Chapter 6, the epoxy coating produces a slippery surface to which cconerete does not bond well. Thus the resistance to slip between a coated, deformed bar and the concrete will depend primarily on the bearing stresses that develop between the bar deformations and the concrete, As a result, the ACI Code specifies longer anchorage lengths for epoxy-coated bars than for uncoated bars. REFERENCES 1. T.C.Hsu, F.O. Slate, G. M. Sturman, and G. Winter: “Microcracking of Plain Concrete and the Shape of the Stress-Strain Curve,” J, ACI, Vol. 60, p. 209, February 1963. S.A. Mirza, M. Harzinkolas, and J. MacGregor: “Statistical Description of Strength of Concrete,” Proc. ASCE, vol. 105, vol. ST6, p. 1021, June 1979. 3. Concrete Manual, Tth ed., U.S. Bureau of Reclamation, Denver, Colorado, 1963. 4. Design and Control of Concrete Mixtures, 11th ed., Portland Cement Association, Skokie, IL, July 1968. 5. H. Kupfer, H. K. Hilsdorf, and H, Rusch: “Behavior of Concrete under Biaxial Stresses,” J. ACI, vol 66, p. 656, August 1969. 6, PCI Design Handbook, Prestressed Concrete Institute, Chicago, 1971 “ACI Committee 504: Guide to Joint Sealants for Concrete Structures,” J. ACI, vol. 67, p. 489, July 1970. 8, A. M. Neville: “Hardened Concrete: Physical and Mechanical Aspects,” ACI Monogr. 6, Detroit, 1971 9. J. Moreno: “225 W. Wacker Drive,” Concrete International, vol. 12, no. 1, p. 38, January 1990. 10, R, Carrasquillo, A. Nilson, and F. Slate: “Properties of High-strength Concrete Subject to Short-term Loads,” J. ACI Proceeding, vol. 78, no.3, May-June 1981 11. G. E. Troxell, H. E. Davis, and J. W. Kelly, Composition and Properties of Concrete, 2nd ed., MoGraw-Hill, New York, 1968 12. Manual of Standard Practice. 23rd ed., Concrete Reinforcing Steel Institute, Chicago, 1981 13. K. Babaci and N. Hawkins: “Evaluation of Bridge Deck Protective Strategies,” Transportation Re- search Board, Program Report 297, 1987. 35 MATERIALS QUESTIONS 2. Define hydration. What is bleeding of concrete? 2.7, Define curing. How is it done? 2.2, What accounts for the large difference between tensile 2.8 What is the field test used to measure workability? 23. 24. 25, 26. ‘and compressive strengths of concrete? Which size of aggregates divides coarse from fine ag- gregates? ‘What characteristics define structural lightweight con- crete? What is the major parameter that determines the strength of concrete? What benefits are produced by using air-entraining for concrete? 29. 2.10. 2a. 212, Which type of concrete is required for high durability? Which type of cement should be used to manufacture concrete exposed to sulfates? Define f;. How is it established? What is the modulus of elasticity of a lightweight concrete that weighs 110 Ibift® and has an f= 4000 Kevin?36 _eanFoRceD CONCRETE DESIGN 243, 244, 248. 216. 247. 218, 219. 2.20, 2.21, ‘What is the modulus of rupture? How is it measured? Define creep. List two negative effects of creep. ‘What steps can be taken to reduce creep? What is the major factor that influences the magnitude of shrinkage? ‘What are the major problems produced by shrinkage? How can they be mitigated’? What isa contol joint? ‘Whats the purpose of using reinforcing bars with sur face deformations? Ifasteel has a yield point of 70 kipsfin?, at what strain wilt yield? A reinforced concrete parking garage is to be con- structed in northern New England during the winter ‘months. Except fora continucus concrete wall, three feet high, that extends aroun¢ the perimeter of each floor, the Completed garage will be open tothe weather. When the garage is in operation during the winter, the concrete floor will be exposed to snow and ice containing large quantities of dissolved deicing salts (chlorides) that will fall from the bottom surfaces of 2.22, cars. Discuss the type and strength of concrete you ‘would recommend for the construction of the garage. Describe any details of construction you feel should bbe included in the contract specifications to produce @ serviceable structure. Loads are moderate and spans are relatively short, In Jubail, Saudi Arabia, large quantities of water are required for the cooling system of an industial com- plex. Since a natural source of freshwater isnot avail- able, seawater will be used, The salty water is to be ‘transported from the ocean through reinforcedconcrete canals and stored in large reinforced concrete tanks. ‘The groundwater—located approximately 3 ft below ‘grade—is highly saline and contains a high concentra- tion of sulfates. Describe the type and strength of con- crete you would recommend for this project. In your discussion include the types of cement, additives, and any other details you feel should be considered to pro- duce durable high-quality concrete. Because the days are often hot (temperatures rise to well over 10°F during the day) and windy, discuss any problems that might arise during the casting of concrete in the field. ‘How might these problems be resolved?CHAPTER 3 DESIGN OF BEAMS FOR FLEXURE Olympic Park Montes. (Phorograph supplied by Régie des Installations Olympigques) 3.1 INTRODUCTION In this chapter we examine in detail the behavior of shallow beams and their response to moment. Shallow beams are one of the most common elements in reinforced concrete design. The same principles that apply to shallow beams can be readily extended and applied to the design of more complicated elements, such as slabs, footings, and the girders of continuous frames. If the designer is thoroughly familiar with the bending behavior of shallow beams, these more complex structural elements should present no unfamiliar problems. Since the flexural strength normally controls the dimensions of beams, the beam is initially designed for moment and only later checked for shear and reinforced additionally if required, ‘The check for shear occurs at a later stage and is independent of the design for moment. The effects of shear will be covered in detail in Chap. 4. Only in the most extreme cases (short spans and heavy loads) are the proportions of the beam controlled by sheat. Before the final selection of reinforcing steel, the designer must also check the bond between the reinforcement and the concrete to ensure that no slippage will take place. In this chapter 3738 REINFORCED CONCRETE FIGURE 3.1 Beams as supporting elements of a beam-and-slab floor system. on shallow beams the bond will be assumed to be adequate; the subject of bond will be covered fully in Chap. 6. Usually the designer wants beams to be as shallow as possible to maximize headroom, even though deeper beams are more rigid and structurally efficient. The ACI Code defines a shallow beam as one whose depth-to-span ratio is less than 3 for continuous spans and less than $ for simple spans, With few exceptions, beams used in buildings and bridges belong in this category. Although equations for proportioning beams will be derived for cross sections of any shape, considerable emphasis will be placed on beams with rectangular compression zones since most beams actually constructed have rectangular proportions. Special attention will be paid to the area of steel reinforcement required to produce ductile flexural members so that structures will have a high degree of toughness. Since proportioning by strength design accounts only for the ultimate moment capacity of the cross section and does not consider the state of stress in the materials or the behavior under service loads, the ACI Code has also established criteria to limit crack width and to control deflections. Beams that are excessively flexible (even if their strength is adequate) tend to vibrate, present an unsatisfactory sagging appearance, and may damage attached or supported nonstructural elements Although concrete beams are typically constructed as part of a continuous floor system (Fig. 3.1), to simplify the analysis, this chapter will focus for the most part on the design of statically determinate beams. Once the design of a determinate beam is understood, continuous beams can be designed with no difficulty. The analysis and design of continuous beams that are an integral part of a floor system of a building frame will be covered in Chaps. 10 and 11 Since most beams receive continuous lateral support from the floor system, they never fail in lateral torsional buckling. Should a beam not be rigidly connected to a slab but exist as a separate unit, lateral torsional buckling will not occur as Tong as the distance betweea Tateral supports does not exceed 50 times the width of the compression flange, Because the designer will probably never encounter a situation in which lateral torsional buckling controls the moment capacity of a beam, only failure of a cross section by overstress in flexure will be considered. ‘Organization of Chapter Understanding the flexural behavior of reinforced concrete beams at all significant stages of loading is essential to provide the engineer with the background needed to design beams that have an adequate factor of safety and behave well under service loads. The sections that follow ccxamine in detail the behavior of a concrete beam at progressively higher levels of load. At the first level of loading to be considered, the reinforced concrete beam is loaded just to the verge of cracking but remains uncracked. Information gathered from this stage of loading will have application in the derivation of minimum steel requirements and in deflection compu- tations. The second level of loading will be at the stage where service loads are applied. With service loads in place, serviceability requirements such as magnitude of deflections and width of cracks rather than strength are of primary concernAt the third level the beam is loaded to the verge of failure, i.c., to the point where its capacity to carry moment is fully mobilized. ‘The proportioning of beams for moment is based on ensuring that at all sections the flexural design strength gM, of the beam will equal or exceed the moment M, produced by factored service loads. Once reinforced concrete behavior has been examined at each of these significant levels of moment, the basis of the ACI design procedures, which are developed in later sections of this chapter, will be clarified. 3.2. THE UNCRACKED CROSS SECTION; COMPUTATION OF THE CRACKING MOME! T The cross-sectional area of a reinforced concrete beam that is effective in carrying flexural stresses varies with the magnitude of the moment. If the moment is small, the tensile bend- ing stresses in the concrete are low (less than the flexural tensile strength of the concrete) and no cracking of the concrete occurs (Fig. 3.2a). For this condition the entire cross section carries flexural stresses. On the other hand. if the moment is large, the maximum tensile stresses in the concrete exceed the tensile strength of the concrete and cracking occurs. Once the beam cracks, only the longitudinal steel in the tension zone carries tensile stresses (see Fig. 3.2b). Since the area of flexural reinforcement is small, on the order of 1 percent of the gross cross- sectional area, the influence of the stee! area on the flexural stresses in the uncracked concrete is small and may be neglected. In other words, the analysis for bending stresses in the uncracked beam can be based on the properties of the gross cross-sectional area using elastic equations, Stresses in the concrete can then be predicted by the standard beam equation foe G1) where M = y ending moment on the cross section distance from centroid of the gross cross-sectional area to point at which stress f is to be evaluated J = moment of inertia of gross section about its centroidal axis, {f= stress at distance y from the centroidal axis, Equation (3.1) is valid as long as the maximum tensile stress in the concrete does not exceed the modulus of rupture j,. Ifa moment is applied that causes the maximum tensile stress in the concrete just to reach the modulus of rupture, the cross section will be on the verge of cracking. The moment that produces a tensile stress just equal to the modulus of rupture is termed the cracking moment Mc. This moment (Mc) can be evaluated by setting f equal to f, in Eq. (3.1), In ACT notation SN fn e fo FIGURE 3.2 The effective cross section: ‘moment (a) less than Mc, and (b) greater @ © than Mey. 3940 RENO DesiCn FIGURE 33 therefore Me, = 18h (2) M where y; is the distance from the neutral axis (N.A.), which passes through the centroid of the uncracked cross section, to the outside tension surface, and f, is given by either &g, (2.3), Eg. (2.3a), or Eg, (2.36). Since the magnitude of f, can deviate by 50 percent or more from the value given by Eq. (2.3) (see Fig. 2.6), the actual magnitude of Mc, can deviate by the same amount (50 percent or more) from the value predicted by Eq, (3.2). Note that cracking of a reinforced concrete cross section does not imply failure 2s long as the area of steel in the tension zone has the capacity to carry the tension force formerly carried by the concrete. Examples 3.1 and 3.2 illustrate the use of Eq. (3.2) to compute the cracking moment of a reinforced cross section. EXAMPLE 3.1, A reinforced concrete beam with the rectangular cross section shown in Fig. 3.3, is constructed of stone concrete. If f= 20 MPa and f, = 300 MPa, estimate the magnitude of the cracking moment using Eq. (3.2) Solution. Neglect the area of steel and base computations on the gross area of the cross section bi 300 x 450° 2” i2 h = 2.278 x 10? mm* Using Eq, 2.36), compute f, $= 010 R = 0730 = 3.13 MPa = 3.13 x 108 X i 12: v= b= 225mm Subsiuing into Eq, (32) N 2 P mmé x 3.13 x 105%, x 2.278 10 mm x 3.13 10° x ay a leh oy 25 mm = 3.17 107 N-mm = 31.7 KN «m (23.4 ft kips) Moy EXAMPLE 3.2. Compute the cracking moment Me, for the cross section in Fig. 34 if f= ‘3600 Ibvin® (24.82 MPa), f, = 60 kips/in® (413.7 MPa), and stone concrete is used. Solution. Locate the neutral axis ofthe cross section. Neglect 4,; break the cross section into three rectangular areas (Fig. 3.4); and sum the moment of areas about a horizontal axis through the base of Viacomaal a “7 in-m f Ax 140in? A=10 sented . NAL h=20in vt A=Soin? f © » FIGURE 3.4 (a) Cross section; (b) distribution of flexural stresses as, cracking impends.41 DESIGN OF BEANS ao 1 ®t mle? “aE C 2 SS" FIGURE 3.5 Influence of cracking on cross-sectional © properties: (a) moment curve, (b) effective beam, the cross section Aut? = EAnye [720)(2) + 6(6)1F = 140(10)(2) + 36(3) Y = 9.2in (233.68 mm) Compute . bh sad? i2 = 220) 5 rao, 10 - 9.292 + | 2&2 + 36¢6.22) 2 12 = 11,0044 int Then use Eg. (2.3) 0 compute f, f= 15. Jf. = 7.5 3600 = 450 Ibvin? = 0.45 kips/in® (3.1 MPa) From Eq, (3.2) 0.45(11,004.4) . Ma = Hb = ACIS 538.30 ips = 44.85 ft kips (60.8 KN» m) When the total service loads are applied to a concrete beam, moments develop that are considerably larger than the cracking moment; therefore Eq. (3.2) is of no use in sizing reinforced concrete members since the cross section will be cracked throughout most of the beam’s length. In deflection computations, however, use is made of the cracking moment in calculating the effective moment of inertia for use in elastic-deflection equations. A concrete beam of constant in fact, behave as if it were composed of segments with two different moments of . 3.5). In regions where the moment is less than Mey, the cross section is uncracked, and the moment of inertia is based on the gross area. On the other hand, in regions where the ‘moment exceeds Mc;, the beam is cracked, and properties of the cross section are a function of the properties of the cracked section (see Example 3.4), EXAMPLE 33. The concrete beam in Fig. 3.6 is made of an all-lightweight concrete with a unit weight y = 110 Ib/ft’(1762 kg/m’). (a) Compute the value of the cracking moment and (b) estimate the Value ofthe concentrated load P at midspan that causes the beam to crack (include the dead weight of the beam). f= 3000 Ib/in? (20.68 MPa) and f, = 320 Ibvin’ (2.2 MPa).42 ps? y= 0.1345 Kips 4 y= 9.68 pee be i tat | e To masse3se [a ~| Moment Tepe FIGURE 3.6 Solution. (a) The weight ofthe beam per foot is! /A. * 176 14 Ha yx volume = ho( is 34.4 lb/ft (1.96 kN) Compute f, with Eq. (2:30): @ -15f ~ 15d but not ore than 15 Fe ~ 411 tin? 2.83 MP3) f= 71520 = 358 bin <411 yin? OK From Eq. (3.2) JSily, _ 0.358 kipslin?(4051) Mem, 964 = 150.4 in kips = 1255 ft- kips (16.95 kN - m) (6) Compute the value of P by equating the midspan moment due to extemal loads (Fig. 36) t0 Ma, ce 33435P = 125 f-kips P = 2.63 kips(11.70kN) 3.3 WORKING-STRESS DESIGN; SERVICE-LOAD STRESSES, Introduction Before strength-design provisions were introduced into the ACI Code in 1963, beams were proportioned by working stress or elastic design. This method is based on the designer's holding computed stresses produced by service loads below specified values of allowable stress. The derivation of design equations for the working-stress method is based on the assumption that reinforced concrete behaves like a linear-elastic material and cracks uniformly across the length of the beam whenever service moments exceed cracking moments. However, the intensity of the stresses predicted by working-stress design does not account for initial stresses produced by shrinkage or the redistribution of stresses that occurs with time as a result of concrete creep. As strength design replaced working-stress design as the preferred method, elastic design was steadily dcemphasized in the ACI Code. Although the 1963 ACI Code gave equal cover- age to both design methods, the 1971 Code was devoted almost exclusively to strength design. Inthe 1971 Code less than one page was devoted to working-stress design, now termed the Al- ternate Design Method. In the 1983 edition of the Code, working-stress design was moved out of the main body of the Code and relegated to an appendix; this move by the Code commit- tee reflects the present obsolescence of the method. Now only the design of sanitary structures holding fiuids is based on working-stress design, since keeping stresses low is a logical way to limit cracking and prevent leakage. Even for this application many concrete designers feel that ultimate-strength design with the proper controls on cracking would be superior to working- stress design. ‘Although limited use is made of working-stress design today, the method is introduced here to develop an understanding of behavior when service loads are applied. Service loads cause the development of extensive flexural cracking in sections where the service moments exceed the cracking moment of the cross section. Once the cross section cracks, the steel alone must carry the tensile stress produced by moment, ‘The derivation of expressions that relate stresses in the steel and the concrete to the service- load moments and to the properties of the cracked cross section will introduce the concept of the transformed cracked section. Once the beam develops flexural cracks, its bending stiffness reduces significantly. For problems where the bending stiffness influences behavior, such as defiection computations, the influence of cracking must be considered, Equations Relating Flexural Stresses and Service-Load Moments To derive equations that relate the bending stresses at a particular point in a reinforced concrete beam to the dimensions of the cross section and the service-load moment, we shall consider the flexural deformations of a short segment of the simply supported beam shown in Fig. 3.7a, ‘The segment, cut from the unstressed beam by two vertical planes a short distance Ax apart, is shown by dashed lines in Fig. 3.7a. As the service load is applied, the beam bends and deflects and the faces on each side of the segment rotate about a horizontal axis in the plane of the cross — o amr (b) compen ee 1 IS + ATT + Je h 4 © @ © o ® FIGURE 3.7 Flexural stresses and strains produced by service-load moments: (a) simply supported beam with service load, (b) stress-strain curves of steel and conerete, (c) cross section of beam, (d) flexural deformations of a small beam segment, (e) strain distribution, (f) stress distribution, and (g) intemal couple. 43 DESIGN OF BEAMS FOR FLEXURE44 ‘REINFORCED CONCRETE esto section. This axis, located a distance kd from the top surface (Fig. 3.7¢), is called the neutral axis (N.A.). Since the longitudinal deformations at the level of the neutral axis are zero, no ongitudinal stresses develop at that depth. Above the neutral axis, longitudinal elements of the segment shorten, an indication of compression. Below the neutral axis longitudinal elements elongate, an indication of tension, These deformations create the longitudinal strains and stresses shown in Fig. 3.7e and f. The variation of the stresses and strains is based on the following assumptions given in appendix A of the ACI Code: 1. Plane sections before bending remain plane after bending; therefore, strains vary linearly from the neutral axis, 2. Hooke’s law, that stress is proportional to strain, applies. . The concrete in the tension zone is fully cracked, and only the steel carries ten 4. No slip occurs between the reinforcing steel and the concrete, To relate f., the maximum stress in the concrete atthe top of the compression surface (Fig. 3.7), and f,, the stress in the steel, to the internal service-load moment M,, we first locate the posi- tion of the neutral axis. Then the position of the resultant of the compressive stresses (Fig. 3.72) can be established (C of course is equal to T since the two forces constitute the equal and op- positely directed components of the internal couple). Once the arm jd of the internal couple is known, the components of the internal couple can be evaluated by equating the internal moment jd or Tid to the known value of the external service-load moment at the section. The position of the neutral axis can be located by equating the components of the internal couple to give C=T G3) Expressing the forces in Eq. (3.3) in terms of the stresses (Fig. 3.78) gives fe kdb = Ashe G4) where 6 = width of cross section d = distance from compression surface to centroid of tension steel fe = maximum stress in concrete A, = total area of tension steel fy = average stress in steel Equation (3.4) can be simplified by using Hooke’s law to eliminate the stresses. Substituting Se = Es€s and fo = Ere. into Eq. (3.4) gives kdbEcSS = AsEsts G5) where E, = modulus of elasticity of concrete E, = modulus of elasticity of steel €e = strain in concrete in top fiber of compression zone 5 = strain in steel ince elastic behavior is assumed, the stresses in the stecl and concrete are assumed to be in the elastic region of the stress-strain curves of the two materials (Fig. 3.76). From the geometry of the strain distribution (Fig. 3.7e), the strains in the steel and concrete can be related by & __& kd © d—kdand 6) Substituting Eq. (3.6) into the left side of Eq. (3.5), dividing both sides by E,, and simplifying gives Eg - = Avesee(d ~ kd) BD Then set G8) where n, called the modular ratio, may be rounded off to the nearest whole number but must not be less than 6, Substituting Eq. (3.8) into Eq. (3.7) and cancelling e, produces, ravi nAg(d ~ kd) G9) which can be further simplified to yield an expression for k by dividing each side by bd to give kdbkd _ Ajnd(1 ~ k) ‘bd(2) bd G.10) Now set GB) where p, called the reinforcement ratio, represents the area of steel per unit area of concrete (a large value of p, for example, indicates a heavily reinforced cross section). Substituting Eq, (3.11) into Eq, (3.10) and solving for k gives k= —pn+ Jpn? + 2pn (3.12) where p = Ay/bd and n = E,/E¢. It would be emphasized that Eq. (3.12) can be used to estab- lish the position of the neutral axis only for a beam with a rectangular compression zone. Equation (3.9) shows that the area of steel A, influences the position of the neutral axis: on the left side of Eq. (3.9) the quantity kdb equals the area of the compression zone, and kd/2 equals the distance from the centroid of the compression zone to the neutral axis. Thus, the left side of Eq, (3.9) represents the moment of the area of the compression zone about the neutral axis. On the right side of Eq. (3.9) the term nA,, called the transformed area of steel, is multiplied by (d ~ kd), which represents the distance of the centroid of the steel from the neutral axis. Thus, the right-hand side of Eq. (3.9) represents the moment of the transformed steel area about the neutral axis. Ifthe area of steel used to reinforce a particular cross section is reduced, the neutral axis must move away from the centroid of the steel area, reducing the size of the compression zone in order to maintain the equality of the moments of the respective areas about the neutral axis. Since the magnitudes of the strains in the steel and the concrete vary linearly from the neutral axis (Fig. 3.7e), the strain in the steel increases and the strain in the concrete decreases as the neutral axis moves upward and away from the steel. If the neutral axis is positioned so that the strains in the steel created by flexure are large relative to those in the concrete, the beam will behave in a ductile manner as the beam bends under increasing load because the concrete, which fails at a low value of strain (0,003 is usually assumed), will not crush until the steel has undergone large deformations. 45 DESIGN OF BEANSDesicn Although Eq, (3.9) was derived for a rectangular cross section, the basic concept of locating, the position of the neutral axis by summing moments of the appropriate areas about the neutral axis is applicable to cross sections of any shape. To locate the neutral axis the designer defines the distance Ad of the neutral axis from the compression surface by a single variable, say Y, and then expresses the moments of both the transformed steel and the compression zone about the neutral axis in terms of this variable. This procedure results in a quadratic equation in terms of the single variable ¥, which can be solved by any convenient method. The application of this principle to a nonrectangular section is illustrated in Example 3.4, where the quadratic equation is solved by completing the squares. To derive an expression for f,, the average stress in the steel produced by the service-load ‘moment M,, we can sum moments of the internal forces on the cross section about C (Fig. 3.7). The internal moment is then equated to M, M, = TX arm = Ayfijd and solving for f, leads to fi (3.13) © Asid Since the compressive stresses are assumed to have a triangular distribution, C is located a distance kd’3 from the top of the compression zone at the centroid of the compressive stresses; therefore, the arm of the internal couple jd is jana wali (3.14) Similarly, an expression for the maximum compressive stress at the top of the compression zone ‘Je can be established by summing moments of the internal forces about T to give M, = CX arm = (Gao) ia Solving for fe, we have 2M, f= Fai G.15) where all terms in Eqs. (3.13) and (3.15) are illustrated in Fig. 3.7 and M, represents the value of service moment at the section where stresses are being computed. These equations apply only to a beam witha rectangular compression zone. The use of Eqs. (3.13) and (3.15) will be illustrated in Example 3.6. EXAMPLE 3.4. For the T:shaped section in Fig. 38, (a) locate the positon of the neutral axis using, the fact that, with respect to the neutral axis, the moment ofthe area of the compression zone equals, the moment of the area of the transformed steel, (b) Next compute the moment of inertia of the cracked transformed cross section Iz. which equals the sum of the moment of inertia ofthe compression zone and the moment of inertia ofthe transformed steel, both with respect to the neutral axis. The moment of inertia of the transformed steel about its own axis is insignificant and may be neglected; f 3.5 kipsin? and n = 8.6. Solution. (a) Assume that the position of the neutral axis isin the web. The distance of the neutral axis from the top of the compression zone is denoted by ¥ (Fig. 3.86). Sum moments of areas about the neutral axis (24x ay = 2) + (F = 412. 32621.5~¥) ‘Simplifying gives Y? + 13.43Y = 132.82as-¥ L n= 3.79 (86) = 32.6 in Tia @ wo FIGURE 3.8 (a) Cross section, (b) transformed section, Solve by completing the square, to give (Y +672) = 13282 +67 ¥ = 662in The neutral axis lies in the web, as assumed. ® nAd ~ YP + Kooy + Ihacge 2) 4, 122.62") | [24 s2.o(14.ss?) + PES) | APO + acaaya.s2y = 9467.11 in* Properties of a Doubly Reinforced Section ‘The compression zone of a reinforced concrete beam is occasionally reinforced with steel to raise the member's bending strength, to reduce long-term deflections produced by creep, or to increase ductility, Stee! located in the compression zone, or compression steel, is denoted by A’, Beams reinforced with both tension and compression steel are doubly reinforced. To compute the properties of the cracked transformed section of a doubly reinforced beam the area of the tension steel is multiplied by the modular ratio n = E,/E, to transform the steel to an equivalent area of concrete. On the other hand, the compression stee! area A’ is multiplied by the factor (n ~ 1). Reducing n by | permits the designer to account for the area of concrete that has been displaced by the compression steel. When this adjustment is made, the designer can treat the compression zone as a solid area and does not have to deduct the space occupied by the compression steel. If flexural stresses are to be evaluated, ACI Code § A.5.5 specifies that the area of the compression steel be multiplied by 2n, Doubling the modular ratio of the compression steel accounts for the increase in stress that occurs with time as the concrete in the compression zone creeps. The creep deformation of the concrete produces additional strain in the compression stee! and gradually raises the level of stress to approximately twice that of the initial value, EXAMPLE 35. (a) Locate the neutral axis and (b) compute the moment of inertia of the cracked transformed section of the doubly reinforced beam in Fig. 3.9. f: = 3600 Ibn’, and stone concrete is used Solution. (a) Compute the modular ratio n E, _ 29,000,000 Ibn? © 57,000,360 Locate the neutral axis. Equate moments of compression area and transformed compression steel to ‘the moment of the transformed tension steel. Refer moments to the neutral axis. rao! +e Dae =) = naar ¥) sy? +70, PY ~ 3) = 83 in? V7 ~ ¥) ¥ = 662in 47 DESIGN OF BEANS FOR FLEXURE48 ain pesan Astin) eye Wi J -DA, it: mE : a=1in PSE NA : vey A woe} ces ttn, pa 10% —— ey ® FIGURE 3.9 (a) Double reinforced cross section, (b) cracked transformed section, () Compute fe = Oe gn yaler — 3) + maT) = 106.62). 941.2y6.62 - 3)? + 8G)(17 - 662)? 3663 in* EXAMPLE 36. The rectangular eross section in Fig 3.19 is 12 by 24 in (30S by 610 mm) deep and carries a service-load moment M, of 150 ft -kips (203.4 KN m), Determine (a) the maximum compressive stress in the concrete and (b) the stress in the steel iff? ~ 3 kipvin® (20.68 MPa) and Jy = 60 kipsin® (413.7 MPa) Solution. (a) Establish proper es of the cross section: 29,000,000 na = OO = 9.25 used E. 57,000 /3000 Ap 5.06 _ ynios = ba ~ Ta1.6 ~ 90 Use Eg. 3.12) to compute k k= -pnt Spm? + 2pm 0.019510) + \{OOIOSO)P + GIVEN) = 0.442 use 0.44 a( and j = 18497216 = 0.8. (@) Use Eq, (3.13) to compute the stress inthe ste! M, _ _(isoy12 Aujd ~ 30601843) ~ By Eq. 3.14) = 18.43 in (467.36 mm) fm 19.3 Kipsfin® (133.1 MPa) Use Eg. (3.15) to compute f,, the stress at the top of the compression zone 2M, 25012) . 0 ~ Fab? ~ TS(0.44) 12.21.68) ~ 7 MPsie’ (11.65 MPa) 3.4_ LIMIT OF ELASTIC BEHAVIOR FOR AN UNDERREINFORCED BEAM Reinforced conerete beams are not strictly elastic because of creep. random patterns of cracking, and the nonlinear relationship between stress and strain in concrete (even at moderate levels of4 5 k rl 7 Cc ta 1 ae i ul t D 4 @ _ > Th ToAL a beac! “=F @ © © ny © FIGURE 3.10 Limit of elastic behavior, tension steel strain reaches €,. stress). Nevertheless, certain types of engineering studies require that we establish the approx- imate value of moment My at which initial flexural yielding of the tension steel occurs. This value of moment represents the limit of elastic behavior. For an underreinforced beam, flexural failure initiates by the yielding of the tension steel while the compressive strains in the concrete are relatively small, that is, the maximum compression stress fe is less than If we assume that the compressive stresses vary linearly from the neutral axis (see Fig. 3.10d), we can establish the value of M, for a rectangular cross section by setting f; = f, in Eq. (3.13) to give My = frAsid (3.16) Flexural ductility represents the ability of a beam to undergo bending deformations without rupture of the tension steel or crushing of the compression zone concrete. To quantify the flexu- ral ductility of a section, we often plot a moment versus curvature diagram (see Fig. 3.11). This diagram shows the variation of curvature as the moment that acts on a section increases from zero to the ultimate flexural strength of the section (the moment producing failure). Typically the points associated with (1) initial yielding of the tension steel and (2) the ultimate flexural strength are sufficient to define the shape of the diagram. The value of curvature yy associated with moment My can be established by considering the longitudinal deformations and strains of Moment (Curvature, y radians / unit length 49 [DESIGN OF BEAMS50 Desion a beam segment of infinitesimal length dx (see Fig. 3.10b). From the geometry of the deformed element, we can express the deformation of the top concrete fiber dA. as dS. = kd(d8) @.17) Dividing both sides of Eq, 3.17 by dx gives ae do FE = ae 3.18) concrete strain at the top surface (see Fig. 3.10c) eon Be a G.19) and that for shallow slopes - curvature wy (3.20) we can write Eq, (3.18) as ay = os G21) kd As shown in Fig. 3.10c, we can represent yy given by Eq. (3.21) graphically as the slope or gradient of the strain curve. Considering the lower portion of the strain diagram, we see that Wy can also be expressed in terms of the steel strain as eee 2 Waa (3.22) Setting €, = f,/E, in Eq. (3.22), we can express yy alternately as Se WEG 8.23) Initial yielding ofthe tension steel does not mean theultimate flexural strength ofthe cross section has been reached since a small reserve of bending strength is still available. As the mo- ‘ment increases above M,, cracks extend further upward (reducing the depth ofthe compression zone) and the concrete stresses in the upper region of the compression zone increase. Now the stress distribution becomes nonlinear (see Fig. 3.33/). As result ofthe redistribution of stresses in the compression zone, the resultant C of the compression stresses moves upward a small dis- tance, inereasing the arm of the intemal couple. The increase in arm produces a small addtional increase in fexural capacity above M, (approximately 4 or 5 percent), The final flexural failure occurs when the conerete crushes, ic, the strain inthe concrete at the top surface reaches 0.003, The value of moment that develops as crushing ofthe concrete impends represents theultimate flexural strength M,, of the cross section (see Fig. 3.11). Computation of M, and the curvature at failure y, will be discussed in Sec. 3.8 (see Example 3.22). A measure of a cross section's ductility—the ability to deform (rotate) inelastically without failing—is given by the curvature ductility ratio, ie, the ratio of the curvatures associated with M and M,, that is, Curvature ductility ratio (CDR) a (3.24) Curvature ductility ratios of 4 or more are typically considered desirable for reinforced concrete members that are subject tothe large displacements and forces created by earthquakes Members with this degree of ductility can be deformed significantly into the inelastic region before a loss of flexural capacity occurs. To reduce the probability of total collapse of reinforcedDESION OF BEAMS FoR FLEXURE S11 = Rin 7 fe TO | [ade atin NA. = Na. NA. v = 1 in + 1298} A,=091 in? - «= 6, = 0.00207 @ ” © @ FIGURE 3.12 (a) Cross section, (b) strain variation produced by M,, (c) stress distribution, and (d) internal couple. concrete structures from earthquake-induced oscillations, the ability of individual members and joints to sustain large inelastic flexural deformations without breaking down is essential. EXAMPLE 3.7. For an underreinforced beam with the cross section shown in Fig. 3.12a determine (a) the magnitude of the moment M, that produces initial failure of the eross section (tension steel strained 10 €,), (6) the associated curvature yy, and (c) the stress at the top surface of the beam. fy = OOkipslin’, f! = 3.6 kipsfin?, and kal = 4.1 in Solution, (a) See Fig. 3.12d. ‘M, = T (arm) = 54.6 kips (15.63 in) = 853.4 in kips = 71.1 ft- kips (b) See Fig. 3.12b. 0.00207 y= Tag = Fao pe 7 0.00016 cadianstin Altemately use Eq. (3.23) to produce the same result te 60 > Bea aay 7 moos; ~ 200016 raansin (©) Compute stress f, at op surface by equating C = T (see Fig. 3.12c). oI 54.6 kips 2 54.620) _ 4 55 Fe = “pay, 7 222s Note: f- = 2.22 kips/in? is well below the concrete compressive strength f 3.6 kips/in? 3.5 DEFLECTIONS UNDER SERVICE LOADS Introduction To be designed properly, reinforced concrete beams must have adequate stiffness as well as strength. Under service loads, deflections must be limited so that attached nonstructural ele- ments, e.g., partitions, pipes, plaster ceilings, and windows, will not be damaged or rendered inoperative by large deflections (Fig. 3.13). Obviously floor beams that sag excessively or vibrate as live loads are applied are not satisfactory. A study of deflections of reinforced concrete beams must account for the instantaneous clastic deflections as loads are first applied, as well as for the long-term deflections that develop due to creep and shrinkage and continue to increase over a period of several years. Under a constant value of load, by the time long-term deflections reach their maximum size they are generally of the order of twice the magnitude of the initial elastic deflections.52 REINFORCED CONCRETE Defected position of beam Folding patton @ FIGURE 3.13 Examples of damage or loss of function to nonstructural elements attached to beams that undergo large deflections: (a) deflecting beam buckles window; (b) excessive deflections of beam cause attached plaster ceiling to sag and crack and the folding partition cannot be closed. Although the research engineer in the laboratory is able to carry out carefully controlled loading tests in which measured instantaneous elastic deflections are within 20 or 30 percent of those predicted by empirical equations for deflection," the practicing engineer must expect devi- ations greater than 30 percent between predicted and measured deflections of beams constructed under actual field conditions. Deflections are minimized when beams are carefully constructed out of high-strength, low-slump concretes that are well compacted and effectively cured. In the field the engineer has a certain limited control over construction methods and procedures by means of the plans and specifications covering the design of the conerete mix and details of placing steel and concrete; however, what the designer specifies and what the construction crews produce can differ widely. Water content may be increased at the job site, incomplete compaction may leave voids and honeycombing, and reinforcing bars may be improperly po- sitioned, By reducing the quality of the concrete, these and other construction procedures can produce members that will undergo larger than expected deflections. For example, if forms are removed from a beam before the concrete has developed its full design strength, the dead weight of the beam and superimposed construction loads can increase the amount of initial creep deflection by several hundred percent over that of a beam whose concrete has fully hardened before the removal of forms. In addition, reinforced conerete beams loaded at an early curing age, while the modulus of rupture and the modulus of elasticity remain low, may crack and deflect extensively in regions that would normally not be cracked when a properly cured beam is put into service and carries its designated design loads. To speed construction of floors or bridge decks, formwork supporting reinforced concrete members is typically removed before the concrete reaches its 28-day strength. To ensure that the unsupported, understrength concrete is not overstressed (and damaged) by its own dead weight or by short-term construction loads, builders often support members with shores, i.e. light, temporary columns that are typically fabricated from small-diameter steel pipes or timber posts. The common practice of storing construction materials on unshored floors mey subject beams and slabs to short-term loads far in excess of anticipated service loads. Although the beams may not fail, the temporary loads can produce large moments that cause more extensive cracking than the normal service loads would have caused. This premature cracking can reduce the bending stiffness of a cross section permanently by 50 percent or more. In spite of the many expedient construction practices that can reduce the stiffness of rein- forced concrete beams and increase the creep deflections, few serious problems develop that are attributable to excessive deflections if the well-considered recommendations of the ACI Code are followed. Concrete beams that are part of continuous frames are naturally stiff because of negative end moments, which provide substantial resistance to deflections. When shallow beams are required whose deflections are difficult to control, the use of pre- stressed corcrete should be considered. The prestressing of concrete beams eliminates crackingand makes the moment of inertia of the entire cross section effective in resisting bending deflec- tions. Moreover, by proper positioning of prestressed steel, deflections—opposite in direction to, and of the same magnitude as those produced by the applied loads—can be induced to produce deflection-free beams for one specific loading condition. In the next sections the ACI Code procedures for controlling and predicting deflections are examined, and a number of examples are worked. Computation of Immediate Deflections Elastic equations (Fig. 3.14) are used to compute the immediate deflections of a reinforced con- crete beam, Other useful deflection equations are tabulated in engineering textbooks and design manuals, e.g., the AISC Steel Construction Manual. Effective Moment of Inertia In areinforved concrete beam, the effective cross section varies along the length of the mem- ber. In regions of low moment, where no cracks exist, the effective moment of inertia should logically be based on the gross transformed area of the cross section. At sections of high mo- ment, where cracking is extensive, the effective moment of inertia is more properly based on the properties of the cracked transformed cross section. To account for the variation of the moment 0 = BET bent} o @ By anf ae pee 33 ‘SION OF BEANS54 REINFORCED CONCRETE FIGURE 3.15 Variation of effective moment of inertia with ‘maximum moment, of inertia along the beam axis, ACI §9.5.2.3 requires the use in elastic-deflection equations of an effective moment of inertia 7., which is computed by Eq. (3.25). This empirical equation is based on a statistical analysis of deflection measurements from extensive beam tests. In Eq, (3.25) the moment of inertia of the gross cross-sectional area is used to approximate the moment of inertia of the uncracked transformed cross section ‘Mex Ma\ |) = t= (He) ofi-(He) Jes 625 where Me, = cracking moment = frlg/y, (Eq. 3.2)] modulus of rupture = 7.5 Jf Ib/in? or 0.7 Jf: MPa for stone or gravel concrete {see Eq. (2.3)] distance from centroid of gross section to extreme fiber in tension ‘maximum moment in member at stage for which deflection is being computed moment of inertia of gross section neglecting area of tension steel Ic, = moment of inertia of transformed cracked cross section Equation (3.25) should be used when 1 Mo/Mzy = 3. If Ma/Mzr > 3, the cracking will be extensive and /, can be taken equal to Ic, with no significant error. If Ma/Mer < 1, no cracking is likely and 7, can be taken as equal to I, (Fig. 3.15). To compute defiections in a span of a continuous beam, the effective moment of inertia may be taken as the average of the values computed for the sections of maximum negative and maximum positive moment. If a continuous beam is prismatic (has a constant cross section), 1, may be computed by Eq, (3.25) considering only the properties of the midspan cross section (see ACI Code §9.5.2.4). ite EXAMPLE 3.8. Compute the initial deflection of the beam at midspan under service loads shown in Fig. 3.16; stone concrete f = 36001b/in? and f, = 60 kips/in? Pekin skip | EE |. = NA nA © Moment FIGURE 3.16Solution, Compute Mz, using Eq. (3.2) 55 om _ 1208) DESIGN OF BEAMS Be = py = yo 13 24 i Fo LEKURE Mg = file = 75.,/3600013,824) "ye 2 in) (12,000) Ms _ 84 en ag Use Eq, (3.25) to compute /,- ‘Compute He (Fig. 3.166). Locate the neutral axis using Eq. (3.12) E, _ 29,000,000 2 =H 200.00 gs oe E. 57,000 /3600 ReD en = 0.0079 k = -pn + Jpn? ¥Ipn = 0.0672 + J(O.0672}? + 210.0672) = 0.306 ¥ = kd = 0.306(21) = 643in Ter + 8.5221 ~ 6.43)? = 4672 in* } |. = (hg) wagons [i = (LY Jaora = sons int ~ \r94) 194, ~ 92518 Using equations from Figs. 3.14 a and b, the deflection at midspan is Swit | PL __ 505)(24491728)_ |, 8#)1728) BRET, ~ 3843.42 x 10*VSIIS) * 49E.42 x 10VS92S) = 0.184 in + 0.196 in 0.38 in EXAMPLE 3.9. The 10-ft-long cantilever beam in Fig. 3.17 carries service loads consisting of a uniform dead load of 0.3 kips/ft and a 4.5-kip concentrated live load atthe tip. Compute the immediate deflection atthe tip for (a) uniform load and (b) total load if 1g = 3600 in’, xg = 2000 in", Mer = 25 R- Kips, and E = 3000 kipsin® weO3kips/t — B My = 18 hips 1M, = 60 f-kipe @ © FIGURE 3.17 (a) Dead load only; (b) dead and live loads.56 FRERFORCED CONCRETE Solution. (a) For deflection at point B due to uniform dead load (Fig. 3.17a) My _ 15 t+ kips Me * 25it-kips ~ 08S! Since the beam is not cracked, use Ie = Iy Compute deflection using Fig. 3.14d wt (03)104)1728) BET, * 8300033600) (2) For deflection at point B due to dead and live loads (Fig. 3.17b) Me _ 60%-kips 4 Me ~ Biekips = 24>! Since the beam is cracked, use Eq. (3.25) to compute Je +(e |- ty ame (ame = 2116in* As = 0.06 in Use Fig. 3.14 and ¢ to compute deflection atthe tip _ wit PP © BEd, * 3E, hy = 23001728), (4.5)10°01728) © 830002116) * “3(3000)(2116) Once a beam has been cracked by a large moment, it can ne 4 0.51 in er return to its original un- cracked state; therefore, the effective moment of inertia Z. that should be used in deffection computations must always be equal to the effective moment of inertia associated with the max- imum past moment to which the beam has been subjected. Often this moment is impossible to determine accurately for most beams. EXAMPLE 3.10. Consider the beam in Fig. 3.17 again. Computations indicated that the effective ‘moment of inertia for the beam is J. = 2116 in* when the beam is subject to a maximum moment ‘M, = 60 ft: kips. If the live load is now removed, so that only the dead load acts, M, decreases to 15 ft- kips. What value of Z, should be used to compute the deflection atthe top of the cantilever? Solution, Use f, = 2116 in‘ for all computations in which M, is equal to or less than 60 ft- ips. Long-term Deflections Due to Creep and Shrinkage Deflections of concrete beams consist of two components, an initial deflection A; that occurs simultaneously with the application of load and a long-term or additional inerement of deflec- tion A, (produced by creep and shrinkage) that takes place over time. This latter component of deflection, which occurs rapidly at years. The increase with time of the long-term deflection is shown in Fig. 3.18. st and then slows, is largely complete after four or five To estimate the magnitude of the additional deflection Ag that occurs with time, the ACI Code specifies that the instantaneous deflection A; produced by the sustained portion of the applied load is to be multiplied by the empirical factor A, Ay = AA; where é a= 3.26)37 DESIGN OF BEANS FOR FLEXURE FIGURE 3.18 Variation of deftection with time. A, = intial elastic deflection, Time, years Ag = long-term deflection. In Eq, (3.26) p’ is the compression steel ratio (A!/bud) at midspan for simple and continuous beams and at the support for a cantilever. The factor € that accounts for both the rate and mag- nitude of the additional deflection, equals 1 at 3 months, 1.2 at 6 months, 1.4 at 1 year, and 2 at 5 years or more. The variation of £ with time indicates that approximately 50 percent of the long-term de- flection occurs after three months. If no compression steel is used, p’ = 0, and A = &. For this case, which results in the greatest long-term deflection, A, = 2A. The additional deflection predicted by Eq. (3.26) is based on average humidity conditions. Ifa member is continually submerged in water or exposed to a moist environment, the additional deflection will be small because no shrinkage occurs. On the other hand long-term deflections of structural members constructed in hot dry regions may be as much as twice as large as the values predicted by the ACI Code because both shrinkage and creep strains are large. Ifthe long-term deflection exceeds the value permitted by Table 3.1, the designer may either increase the depth of members or add additional compression steel. If the sag produced by the long-term deflections is objectionable from an architectural or functional point of view, forms may be raised (cambered) a distance equal to that of the anticipated deflection. Example 3.11 illustrates the use of Eq. (3.26) in deflection computations. EXAMPLE3.11. Investigation of Deflections. The beam in Fig. 3.19 carries uniform service loads of | kiprft dead load and 1.5 kips/ft (21.89 kN/m) live load, The dead load, which includes the weight of the beam, is always in place, but only an average of 60 percent of the live load acts continuously. (@) Compute the initial deflection due to tral load and (estimate the additional ong term deflection 3000 Ibn? (20,68 MPa), f, = 50 kipslin? (334.7 MPa), J, = 13,824 in‘ (5752 x 10° mm), Tez = 10;300 in* (4287 x 10° ram), and f = 7.5 Jf = 411 lbvin® (2.83 MPa) Solution, (a) SOP) asp = 500 in -kips (169.5 kN m) 0.41103,824) 2 My _ 1500 feta TBIT>3 473 in kips (53.45 kN m) Therefore use Ty = Tog = 10,300 in* 57,000 3000 7000 Swit _5(2.5)(208(1728) 3B4ET. ~ 384(3120)(10,300) E 3120 kiin’ (21,500 MPa) 4 0.28 in (6.6 mm)sowowa poyoene 40 pauoddns 0) aeuep wasaid o} uayr axv samsvau arenbape 4. popaaoxo aq A “pasaprsuoo B39 2604) 01 Jo sonsuargeys uonsoyep-oumn ayp 01 Sunejas wep Suud.u/BuD pardaode Jo siseq a4p uo om - Saonsogap aT ra q poteurp 2q 01 Sex| 108 sour [eAMnsUoU 0 potanne 20 Zunuodns wortonnsto0 4004 © 400% ‘Spry. any euoRIPpS uv 0 onp woRD9yop a¥epoum om PAE “speo| pautesns fe 01 np wondagop oumn-But ap 30 soon yop or tums ou) “soutop9 eanionisuot if Jo wou; aim &q poScurep 0q 1 Soy stuowayo jemmonnsuo ff sage simoo0 apa wonaoyap jeioh 3xp Jo 78d HL, 0 poysene 30 Suuoddns 109200] 30 J00% suonsogop afin £q poleaep oq or Ajax sua 1 poo any 2G 01 np wonsoyop areypoUN Teimannsuow oy poyoene 10 Humoddns tou sioouy suonogap aie Aq paleuzep 29 01 Aim poo ant an 01 arp wonsoy9p aripoLa resmonnsvow of pajgene 20 Suto Pavapysuod aq oF woRS=HFC saquiew jo ad (4S°6 *148F, 9POD IDV) SuoNDayep panduiod ajquaoye uNUIXeW 1'¢ TTAVL 58b=12in 59 1 srerea taki ee 21.6ia| a: fa) 0) FIGURE 3.19 (®) Base computations on the sustained portion of the load: dead load plus 60 percent ofthe live load, ‘The sustained load is L kip/ft + 0.6(1.5 kipift) = 1.9 kipsit (27.7 kNim) Since, except for the value of load, all terms in the deflection equation are identical to those in the ‘computation of total load deflection, deflections are proportional to load and the initial deflection is 4 = 122028) 9219 53am Substituting € = 2 and p' = Ointo Eq (3.26) the addtional longs-ter deflection say after 5 years) isthen a, = (2), = 20.21) = 042 in 10.7 mm) rap) = 0.42 im (10.7 mi EXAMPLE 3.12. Estimate (a) the instantaneous deflection at midspan of the continuous beam ABC in Fig. 3.20a produced by the uniformly distributed service load of 2 kipsift, and (b) estimate the midspan deflection after one year assuming the entire load acts continuously. Base your analysis ‘on the properties of the midspan section. The beam is constructed of stone concrete with 1600 Ibn’, E, = 3420 kips/in’, n= 9, fee = 2204.2 int, I, = 5832 in* Solution. (a) Because the beam is prismatic (.¢., of constant cross section) the ACI Code permits us to base deflections onthe properties ofthe midspan cross section. Therefore, the value of M, used in the computation of wll be the positive moment of $4 ft kips at midspan ‘Compute the cracKing moment using Eq, 3.2) Mg = lehe = 589275 V5600_1 = 201.6 in kips = 24.3 ft- kips % 9 000 Me 34 M, 243 a Since the ratio of M, to Mg; is less than 3, use Eq. (3.25) to compute J, Mo» fy (Mery |p (i) =[- Ge] ~ (2) + [+ (2n) oe Is 8.25) = 2535.7 ‘Compute the initial deflection at midspan by superimposing the deflections at that point produced by the uniform load and the end moments. For this computation we can treat the continuous beam as equivalent to a simply supported beam loaded by a uniformly distributed load and moments at each end (see Fig. 3.14a and f for deflection equations; Fig. 3.20b shows the three loading cases superimposed). The multiplier of 2 in the second term of the following equation accounts for the contribution of the 90 ft - kip moments acting at each end of the span. We use a negative sign toREINFORCED CONCRETE DESIGN Kipvtt Pr F200, dears coninaous tw.7f __ 200.9 bas amiss" | | |S + (Section 1-1) Lath Sa -kips Service load moment diagram 90 8 kips 90 hips FIGURE 3.20 indicate a downward deflection and a posit ive sign an upward deflection Sw! ME « ~ 3646, * 6Er” SQx24y1728 , 901728 3eaaaa02535.7 * T6ea2025357 > 0.44 in downward () Total deflection A, after one year: é Te pO) La T+ Goo0m 4) = ‘The additional deflection A, = Ad, 0 o4in where the one-year time factor € = 1.4 and AL‘TABLE 3.2 Minimum thickness of beams or one-way slabs unless deflections are 61 computed (ACI Code Table 9.5a)t DESION OF BEAMS ‘Members not supporting or attached to partitions or other construction likely to be damaged by large deflections ‘Minimum thickness ft Simply Oneend —_Both ends supported continuous continuous Cantilever Solid one-way slabs 120 ma ns mo Beams or ribbed one-way slabs 6 mss 121 18. ++Span length Jin inches. Values in the table apply to normal-weight concrete reinforced with steel of f = 60,000 Ibvin? (413.7 MPa. For structural lightweight concrete with a unit weight between 90 and 120 Ib/ft? (1440 ‘and 1920 kg/m?) multiply the table values by 1.65 ~ 0.005w (1.65 — 0.0003w) respectively, but not less than 1.09; the unit weight w is in IBM (kg/m). For reinforcement having a yield point other than 60,000 Ivin? (400 MPs), multiply the table values by 0.4 + f,/100,000 (0.4 + f,/700) with f, in Ibfin® (MPa. Control of Deflections Since deflection computations can take some time to carry out, the ACI Code has established a procedure to limit deflections by placing restrictions on the minimum depth of a beam.? This method eliminates the need to compute deflections and controls deflections by requiring that ‘beam depths notbe less than a specified fraction ofthe span length. Minimum depths fora variety of common support conditions are given in Table 3.2. The theoretical basis for this method of controlling deflections can be established by considering the calculations for the midspan deflection of a simply supported, uniformly loaded beam that behaves elastically (Fig. 3.21) — Kwlt 3 Wr Where Ki = 355 S a where w = uniform load L = span length E = modulus of elasticity effective moment of inertia Equation (1) can be expressed in terms of the maximum moment M at midspan by substi- tuting wL? = 8M into Eq. (1). _ Ki8ME _ KML? EL EL A where 8K; = Kp @ ‘When a linear variation of compressive stress is assumed, the maximum concrete stress can be expressed as Me > 3 fF @ where c is the distance from the neutral axis to the outside compression surface and f is the maximum service stress at a distance c from the neutral axis. FIGURE 3.21Solving Eq. (3) for M = fI/c and substituting into Eq. (2) gives — Kft rr 7 f and E in Eq. (4) can be expressed as functions of /,. (In earlier editions of the ACI Code Eq., E was expressed as 1000f..) f= Ksfp and E=Kifl (4) and £_ Ks ees 6) Using the relationship in (5) to eliminate f and E in (4) gives 2 a = KoKsE 6 c ‘When c is expressed as a fraction of the overall depth h by ¢ = Kh, Eq. (6) can be written as KoKsl? _ KL? _ KoKs Koh here K = Re Based on a study of the maximum deflections reinforced concrete beams can safely undergo without producing damage to attached construction, maximum values of deflection have been established as a fraction of the span length L for a variety of design situations. For example, if plaster ceilings are not to crack, experience indicates that the maximum live-load deflection of beams must aot exceed 1/360. In general terms L As @) 1 where A is the maximum permitted deflection and C; isa constant that depends on the maximum deflection the beam can tolerate if it is to serve its intended purpose satisfactorily. Using Eq, (7) to eliminate A in Eq, (8) gives el 7 a or h= KGL ° Tocliminate a decimal, the product of the two constants K and C; is set equal to 1/N, where ‘Nis a number greater than 1, Then Eq, (9) can be written L hey 0) Values of N that ensure the serviceability of beams and one-way slabs are tabulated in ‘Table 3.2. By the proper choice of N values, provision can be made for the effects of creep and shrinkage on long-term deflections. Since the method is approximate, the ACI Code limits its use to beams that support or are attached to nonstructural elements capable of tolerating relatively large deflections without incurring damage. If (for architectural or other functional reasons) the designer must use a beam whose depth is less than the values required by Table 3.2, the zesulting deflections must be computed and compared with the maximum permitted values of deflection tabulated in Table 3.1.<>? ee 2086.19 —4 [20 (6.1 m fei 2.6m] FIGURE 3.22 FIGURE 3.23 1 a Ta me ssa@om | 2006.1m | 18nsm FIGURE 3.24 EXAMPLE 3.13. Determine the minimum depth required by the ACI Code if deflection computa- tions are to be avoided for the lightweight-concrete beam in Fig. 3.22; f = 3 kips/in® (20.68 MPa), ‘fy = 50 kips/in? (344.7 MPa), and w = 110 Tb/ft? (1762 kg/m). Solution, From Table 3.2, under “Simply supported” we read Iria = F065 — 00050) (04+ fs ) 168 —aotse = 108 2) 50,000 tan ~ 2822145 agosto (04 + $02) = a in 377.2 mm) EXAMPLE 3.14. Determine the minimum depth of the two-span continuous beam in Fig. 3.23 if deflection computations are to be avoided. f = 4 kips/in? (27.58 MPa), f, = 60 kips/in® (413.7 MPa), and w = 145 Ib/ft? (2320 kg/m’) Solution. Since the conerete is normal weight and f, = 60 kipsfin®, no modifiers are required. From Table 3.2 under “One end continuous” we read 1 _ 20(12) 185 ~ 185 agin = = 13 in (330.2 mm) EXAMPLE 3.15. What isthe minimum depth required to control deflections of the thee-span con- tinuous beam in Fig. 3.24 made of stone conerete with J, = 60 kipslin®? Solution. Span BC for two ends continuous (see Table 3.2) _ L202) agin = 5 = AS = 11.43 in 2903 mm) ‘Span CD for one end continuous (shorter span AB not critical) 1 182) | Innis = yg 7 “GES. 7 1-68in(296.7 mm) controls 3.6 CONTROL OF CRACKING UNDER SERVICE LOADS As a reinforced concrete beam deflects, the tension side of the beam cracks wherever the low tensile strength of the concrete is exceeded. The more the beam deflects, the greater the length and width of cracks. Although cracking cannot be prevented, it is possible by careful detailing of the steel to produce beams that develop a large number of narrow, closely spaced cracks in preference to a few wide cracks.64 REINFORCED CONCRETE esicn | Excessive cracking of the concrete that covers the reinforcement is of considerable concern: the protection of the reinforcement from the environment depends on the integrity of the con- crete cover. If members have been constructed of high-quality, well-compacted concrete, studies indicate that the presence of narrow cracks in the cover will not significantly reduce the effec- tiveness of the concrete to cover and to protect the reinforcement from deterioration by weather or by other corrosive elements, stich as seawater. On the other hand, if wide cracks develop in the cover, the reinforcement will be exposed to the atmosphere and be vulnerable to continuing deterioration by corrosion. The maximum crack width the designer should permit depends on exposure conditions.* If concrete is exposed to seawater or cycles of wetting and drying, the ‘maximum width of any crack should not exceed 0.006 in (0.15 mm) or at the far limit 0,008 in (0.2 mm). For members protected against weather, crack widths up to 0.016 in (0.41 mm) are permitted by the ACI Code. Where concrete surfaces are exposed to view, excessive cracking looks unsightly, the integrity and solidity of the member is compromised, and the architects intention of a smooth monolithic concrete surface is not achieved. In past years, when concrete design was based on working- 145 no good If the bars are separated, A = 18 and 2=136<145 OK 36 EXAMPLE 3.18. If the cross section in Fig. 3.28 is reinforced with 2 no. 35 metric bars, does it satisfy ACI Code requirements for crack control? Exterior exposure, f, = 500 MPa, Solution. Use Eq. (3.29) to compute z area of tension concrete __2(75)280 = 2 mm? number of bars oa f, = 06f, = 0.6500) = 300 MN 2 = F&A = 300:/75(21,000) = 34,905 MNO _ 34.9 MN > 25MN oy good Skin Reinforcement for Deep Beams Experimental studies as well as field observations show that at or near midheight of deep beams the maximum width of flexural cracks may be as much as two or three times larger than the width of the same crack at the level of the tension surface from which the crack initially originates (see Fig. 3.29a). Because crack widths of this magnitude detract from the surface appearance of a beam and also expose the interior of the beam to corrosion, ACI Code §10.6.7 requires that in the tension zone of beams whose depth, d, of web exceeds 3 ft (900 mm) the designer add ‘additional longitudinal reinforcement to both sides of the beam. Since the purpose of this skin reinforcement is to control the width of cracks on the exterior surfaces of the web, the Code requires it to be uniformly distributed along both sides of the tension zone for a distance d/2 ‘measured from the main flexural steel ‘The minimum required area of the skin reinforcement A,, per foot of height on each side of the web, is given as Age = 0.012(d — 30) where d equals the effective depth (3.30)+ 4 Flexural i a ack reinforeement | 4 3) for erack contro! JH», xt Main} * Main flexural 7 reinforcement reinforcement @ o FIGURE 3.29 (a) Side elevation of flexural crack in a deep beam with no skin reinforcement, (b) skin reinforcement; see black dots. In addition, the maximum spacing between bars is not to exceed the smaller of d/6 or 12 in (300 mm); moreover, the total area of the longitudinal skin reinforcement in both faces need not exceed one half of the required flexural steel (see Fig. 3.296). ‘Although the main function of the skin reinforcement is to limit the width of flexural cracks, the Code permits the designer to include the contribution of the skin reinforcement to the flexural strength of the cross section. As part of this computation, the designer must determine the stress in the steel by considering the variation of the flexural strains over the depth of the cross sec- tion (for example, see Fig. 3.33c). Since the flexural strains vary linearly from the neutral axis (where they equal zero), the stress in bars located near the neutral axis may be well below thei yield-point stress. Because the full depth of the cross sections of prestressed concrete beams are normally in compression (and therefore uncracked), they do not require skin reinforcement. EXAMPLE 3.19, Selecta suitable se of bars to serve as skin reinforcement forthe beam in Fig. 3.30 To supply a required flexural steel area of A, = 7.2 in®, the beam is reinforced with 5 no. 11 bars. Compt the required area of skin reinforcement using Ea (3.30): Au = 0012(4 ~ 30) 0.012 (44 ~ 30) = 0.168 inte ‘Area of steel required for each face = Ajxd/2 = 0.168(]) = 0.308 in* ‘Use 3 no. 3 bars each side, which supply A, = 0.33 in’. Verify that the total area of skin reinforcement does not exceed half the required flexural stel a 0.33 in*(2) < 7.2/2 in? OK |poe atten A = Ay=7.2in? FIGURE 3.30 6768 REINFORCED CONCRETE Desien Space bars uniformly over adistance d/2 = 22 inches; 2 dy s 24 = 5Sin< f= E = 7h 3.7_ FAILURE MODES AND FLEXURAL STRENGTH OF REINFORCED CONCRETE BEAMS Introduction ‘When a reinforced concrete beam is loaded to failure, three modes of bending failure ere possi- ble. The particular mode of failure is determined by the percentage of steel located in the tension zone. Two of these modes are brittle and one is ductile. Since the designer’s prime conzem is to produce duc:ile beams with a high capacity for energy absorption, beams must be proportioned to ensure that only the ductile failure mode is possible. Case 1. The beam is overreinforced and the failure mode is a sudden, brittle failure, which the engineer must carefully guard against in design, When the overreinforced beam is loaded to failure, the fuilure is initiated by the crushing of the concrete followed by the sudden disintegra- tion of the compression zone while the stress in the relatively large area of steel has not reached its yield point. To prevent a brittle failure, the reinforcement must yield while the strain in the concrete is ess than the failure strain of 0.003. Case 2. The beam has a moderate percentage (on the order of one percent) of stee!, and the failure mode is initiated by a yielding of the steel while the strains in the concrete are relatively low. Such beams can continue to carry loads and are able to undergo large deflections before final collapse occurs; this ductile mode of failure is the only acceptable mode. Case 3. The beam is lightly reinforced with a very small percentage of steel, and the failure ‘mode is also brittle. When the tensile stress in the concrete exceeds the modulus of rupture (the tensile strength of the concrete), the concrete cracks and immediately releases the tensile force it carries; the lightly stressed steel must then absorb this increment of load. If the area of steel provided is wo small to carty this added force, the steel will snap and total rupture ofthe section will occur suddenly. To ensure ductile failures, upper and lower limits on the permitted area of reinforcing steel are established by the ACT Code. The lower limit ensures that enough steel will be used to prevent the steel from snapping suddenly and causing the beam to split. The upper limit on steel area prevents the design of overreinforced beams. Since the presence of shear force has little influence on the moment capacity of a cross section, shear is not considered in the design of members for bending. Failure of an Overreinforced Concrete Beam To study the bending failure of a simply supported overreinforced concrete beam, the behavior of the beam in Fig. 3.31 is observed as it is loaded to failure by progressively increasing load increments at midspan. In a beam of constant cross section, failure will occur at the section of maximum moment where stresses are greatest. Deformations, strains, and stresses discussed here are all associated with the section of maximum moment. As discussed in Section 3.3, the position of the neutral axis in a reinforced concrete beam is a function of the area of the rein- forcing steel. In an overreinforced beam, the large area of steel positions the neutral axis close to the tension steel (Fig. 3.31). As load is applied, the beam bends and cross sections rotate about the neutral axis, The deformations and, correspondingly, the bending strains are zero at the neu- tral axis and vary linearly with distance from the neutral axis. The maximum compressive strainP Concrete crushes & = =a eA tone i @ 0 « % 00m Sein ° «=o —-h na fon A fi © w © FIGURE 3.31 Failure of an overreinforved beam: (a) beam with failure load, (b) stress-strain curves showing stress in the reinforcing steel and ‘maximum stress in the concrete at failure, (¢) section A-A, (d) strain distribution at failure, (e) stresses. in the concrete occurs at the top surface of the compression zone. The maximum compressive strain is much greater than the tensile strain in the steel since the upper surface of the concrete is farther from the neutral axis than the steel. When the strain in the concrete at the top surface reaches 0,003, failure occurs. It is characterized by a large depth of the concrete compression zone crushing rapidly while the steel, strained below its yield point, is still elastic, as shown in Fig. 3.316, When steel is not designed to yield at failure, designs are uneconomical because the potential strength of the steel is not used. To sum up, failure in an overreinforced beam is initiated by crushing of the concrete while deflections are relatively small and before the appearance of extensive cracks in the tension zone. Before failure, deflections and cracking are small because of the low strains in both the concrete and the steel. Since concrete is brittle, failure will be sudden and will occur with limited visual warning. When the concrete crushes and falls out, the beam’s capacity to carry load is destroyed. Such local failure can be dangerous if the falling concrete overloads other portions of the structure and precipitates failure of other members. To guard against the design of brittle overreinforced beams, the ACI Code establishes a limit on the maximum amount of steel that can be inserted in a beam. Limiting the percentage of reinforcing steel moves the neutral axis toward the compression zone. As a result, the strain in the steel will increase more rapidly than the strain in the concrete, and the beam will fail in a ductile manner by yielding of the steel before the conerete crushes. Such beams are termed underreinforced. The difference in behavior between an underreinforced and an overreinforced beam is shown by the load-deflection curves in Fig. 3.32. Except for the area of reinforcing steel, both simply supported beams in Fig. 3.32 have the same dimensions. Notice that at low values of load, before the concrete has cracked, both beams deflect approximately the same amount since behavior is governed primarily by the properties of the gross area of the concrete. Flexural Strength of an Underreinforced Beam ‘To understand the flexural behavior of an underreinforced beam, we shall study the variation of the longitudinal stresses on a vertical section at the center of a simply supported beam with 0 DESIGN OF BEAMS0 REINFORCED CONCRETE Deston Sudden faite y ‘ompresion zone disintegrates i i -Overreinforced Load ; o © FIGURE 332 (a) Comparison of behavior between an underreinforced (b) and overteinforced (c) beam with identical dimensions 4 concentrated load at midspan (Fig. 3.334). This study will cover the changes in stress at the section of maximum moment as the internal bending moment increases from the service-load level to the ultimate flexural strength of the section. The stress distribution associated with failure is of particular importance since this distribution is needed in the derivation of design equations relating the ultimate flexural strength, which the ACI Code calls the nominal moment capacity, to the cross-sectional dimensions, the area of reinforcement, and the strength of the materials (as measured by f! and f,). Although the beam is assumed to have a rectangular cross section, the most common shape in practice, the procedure for evaluating the nominal moment capacity can easily be extended to cross sections of any shape. Experimental studies verify that flexural strains in a reinforced concrete beam vary linearly from the neutral axis even when the beam is heavily cracked and the materials are strained into the inclastic region, Because of the relatively small area of steel in an underreinforced beam, the neutral axis is positioned closer to the outside compression surface than to the tension steel. AS aresiilt, when the beam bends, the tensile strain in the stee! is greater than the maximum com- pressive strain in the concrete at the top surface. If the concentrated load at midspan is initially assumed to be equal to the service load, the stress distribution shown in Fig. 3.33d develops on the cross section at midspan. As indicated in this figure, the concrete stress, proportional to the strain, varies linearly from zero at the neutral axis to its maximum value at the top compression zone. In the tension zone, the concrete is assumed to be cracked so that all the tension force is carried by the steel, which is stressed to f, typically about half the yield-point stress. The mag- nitude of the stresses induced in the steel and in the top surface of the concrete by the service load is noted (Fig. 3.33i) on the stress-strain curves of the materials by points labeled I. At this level of load, deflections are small and tension cracks in the concrete are typically invisible to the naked eye. If the concentrated load at midspan is now increased, the beam will bend and deflect farther, ue to the increase in moment, which is proportional to the load. Strain and stresses on verticalnm 0.003 ‘DESIGN OF BEANE FOR FLEXURE 6 0008 rm Swain @ FIGURE 3.33 The variations of bending stress with moment in an underreinforced beam loaded progressively to failure: (a) beam and moment curve: (b) cross section; (6) stain; (2) service-load stresses; (c) steel yields, initial failure; (f) secondary compression failure; (g) equivalent stress block, a = 8c; (A) intemal couple; (j stress-strain curves for steel and concrete. cross sections will increase. The new distribution of strain is denoted by line 2 in Fig. 3.33c, and the corresponding stress state by the distribution in Fig. 3.33e. This figure indicates that the total load has the particular value needed to raise the steel stress just to the yield point. The bending moment My produced by this load can be evaluated by Eq. (3.16). Points labeled 2 in Fig. 3.331 show the level of stress and strain in the steel and in the concrete at the top surface. Since the strains vary linearly from the neutral axis, the strain in the steel is greater than the strain at the compression surface of the concrete; therefore when the stee! is strained just to its yield point, the maximum concrete strain (at the top surface) is still below the value of 0.003, the strain at failure. Since the forces in the steel and in the concrete constitute the components of the internal ‘moment, they must be equal in magnitude under all stress conditions (ie., T = C). Yield- ing of the tension steel does not produce total failure of the beam since a small additional amount of moment capacity can still develop; this additional moment capacity is created by the ability of the arm between the steel force T and the resultant compression force C in the concrete to increase in length a small amount as the beam is further deflected and strained by additional load. The increased strain causes the concrete located away from the neutral axis to be stressed into the nonlinear range as shown in Fig. 3.33f. The additional strain cannot increase the steel2 Desion stress, which is)in the horizontal portion of the stress-strain curve, where stress remains con- stant with ‘ncreasing strain; the tension force, equal to A, f,, therefore remains constant, For this reason the compression force C, the other component of the internal couple, must also re~ main constant in magnitude. With the stresses in the concrete increasing, the equality of internal forces can be maintained only by a decrease in the area of the compression zone. For the area to reduce in size, the neutral axis must shift upward. The effect of this move is to increase the distance between the neutral axis and the steel. As a'resull, thé strain in the steel increases at an accelerating rate. In the same way, the reduced distance between the neutral axis and the compression surface causes the concrete strain to increase at a decreasing rate. Associated with the shift of the neutral axis upward are an increase in the length and width of tension eracks and a substantial increase in the magnitude of beam deflection. Total failure of the beam is assumed to occur when the moment is large enough to produce a strain of 0.003 in the concrete at the top surface of the beam. At this level of strain the concrete crushes and the beam fails. Figure 3.33/ shows the stress distribution on the beam cross section just before failure occurs; points in Fig. 3.33i marked with a 3 denote the state of stress at final failure. The variation of concrete stress in the compression zone between the neutral axis and the outside surface of the concrete is identical to that of the stress-strain curve of the concrete in Fig. 3.337, Since a nonlinear stress distribution complicates the derivation of design equations, the state of stress is approximated by a uniform stress of 0.85 f acting over the upper portion of the compression zone (Fig. 3.33g). The area on which the uniform stress acts is called the stress block. Whitney* established the magnitude of the uniform stress and the depth of the stress block required to produce a resultant compression force equal in magnitude and located at the same depth as that produced by the actual nonlinear stress distribution in Fig. 3.33f. From analyses of test results, the depth of the stress block a was related to c, the distance between the outside ‘compression surface and the neutral axis, by a= Bic G3) where 0.85 if ff = 4 kipsin? GOMPa) —(3.32a) Bi = 9 0.85 ~ 0.05[f.(kips/in) ~ 4] if f > 4kipsfint USCU 3.32) 0.85 — 0.008[f;MPa~ 30] iff >30MPa SI. (3.320) but 8 not to be less than 0.65. ‘These expressions apply to compression zones of any shape. The variation of By with i is shown in Fig. 3.34, As Fig. 3.34 shows, 8) reduces to 0.65 for values of J.’ equal to or greater than 8 kips'in? or 55 MPa, The resultant compression force in the concrete C acts at the centroid of the stress block. For a rectangular compression zone, C is located a distance of a/2 from the top surface. With the intemal forces established, the nominal bending strength of the section at failure M,, can be evaluated by summing moments about T to give M, = (a 5 5) os a(t 5 5) (3.33a) or summing moments about C to give My = re 5) = Asfy (« 5) (3.336) If the components of the internal couple are equated, that is, C = 7, the depth of the stress block @ can be expressed in terms of the dimensions of the cross section and the strength of the materials as 34)85—0.008L;-30] St oss Bix 0.65 a 2085-05144] USCU 4 8 ipwin? ms 55 MPa & FIGURE 3.34 Variation of f with 28-day compressive strength, {i ‘Multiplying the top and the bottom of the right-hand side of Eq. (3.34) by d and setting Ay/bd peives Afd_ _ phd _ ad BaO8SAD — O85F 0.85 where q = pylf!. The tension reinforcement index q is a measure of the strength of the rein- forcement. Substituting Eq, (3.35) into Eq. (3.334) and simplifying gives My = fiibd?q(1 = 0.599) (3.36) Experimental studies verify that the equations derived in this section predict the ultimate flexural strength of underreinforced beams accurately. In Fig. 3.35 the nominal flexural strength represented by Eq. (3.36) and plotted as a solid line is compared with the flexural strength at fail- ure measured in 364 beam tests. To permit test results of beams constructed of different strength concretes and with a variety of steel yield points to be compared with values of moment pre- dicted by Eq, (3.36), nondimensional variables are established by dividing the flexural strength by fibd?. Figure 3.35 indicates that the actual flexural strength, which rarely deviates more than 10 percent from the calculated value, will usually be within 4 or 5 percent of the predicted value. (3.35) 038 os a1-0594) 02s NS 10005 010 O15 020 025 0.30 0.35 O40 OAS 4 = phi FIGURE 3.35 Variations of flexural strength with reinforcement index 3 DESIGN OF BEANS"4 ENFORCED CONCRETE DESIGN 2 n0.20 EXAMPLE 3.20. Determine the theoretical value of the concentrated load P, that will cause the underreinforced beam in Fig. 3.36 to fail in flexure, Base computations on the direct evaluation of the intemal couple from the assumed stress condition at failure. f = 3 kipsfin® (20.68 MPa) f, = {60 kips/n® (413.7 MPs). Solution. ‘The midspan moment due to the applied load at midspan is Pal _ Pa(l6) _ T = Ayfy = (2 in?)(60 kips/in?) = 120 kips (533.76 kN) Equate T 120 kips 1H(0.85 f’) 7 in(119.4 mm) Evaluate the nominal moment capacity by summing moments about C (Fig. 3.364), a 16-3) ~1m(i $2) = s6seucr22i-m e=47in > (119.4 mm) b= 10 in (254 mm) 986 Cn 08S fab € * Pe) isin Ft anal (4372. mm) Feats eum) 4 ee lt = — 7 b 2 in? (12903 mm’) o @ © FIGURE 3.36 (a) Beam and moment diagram; (b) cross section; (c) stresses st failure; (d) internal couple. quate moment at midspan due to applied loads to M, = 156.5 ft-kips P, = 39.1 kips(173.9 kN) EXAMPLE 3.21. For the cross section in Fig. 3.37a determine (a) the nominal flexural strength 'M, of the cross section reinforced with 2 no, 20 metric bars, Also verify that the cross section is 250 am +] ose: ae c= ab.85 eT) 4 =360 mm * at 3 ane | > z ~ mTHAS, rf - F @ © © @ FIGURE 3.37 (a) Cross section, (6) strain variation, (c) state of stress as failure impends, (d) intemal couple M,underreinforced; ic. the strain in the steel exceeds the yield strain when failure impends. fy 400 MPa = 400. N/mm,’ = 40 MPa. Solution, (a) Compute T = Af, = 600 mm? (400 N/mm?) = 240,000 N Determine the depth a of the stress block by equating T = C (see Fig. 3370) T=c 240,000 N= ab0.85, = a250(0.85}40 a= 282mm My = T(d ~ af2) = 240 KN(360 ~ 28.2/2) = 83.02 KN-m (b) Use the strain diagram in Fig. 337% to establish e, by considering similar triangles. By = 0.85 ~ 0.0081! ~ 30] (&q.3.320) B= 0.77 Using Eq, (3.31) we compute c, the distance of the neutral axis below the compression surface: _ a _ 82 Be o77 = 3662 mm By similar triangles (see Fig. 3.376), 0.003 _ _¢, ¢ d= dmc 360 — 36.62, e, = 0.003 = “SE 0.003 = 0.026 ‘The strain required to yield the ste! is e, & . 400 © = ~ 300,000 ~ 900 Since €, = 0026 exceeds €, = 0.002, the beam is underreinforced. This check is not required in practice if the normal design procedure is followed. 3.8 THE PLASTIC HINGE AND MOME! -CURVATURE DIAGRAMS ‘A beam on the verge of failure from the strain of its ultimate load is said to have developed a plastic hinge when the beam can be displaced up or down by the application of an infinitesimal force. Such behavior implies that the application of ultimate load has reduced the bending stiff- ness of a beam to zero with respect to its response to any additional load. The stress distribution in Fig. 3.33g constitutes the state of stress associated with the formation of a plastic hinge. A method of analysis for indeterminate structures, called plastic design, makes use of the plastic- hinge concept. Although plastic design is not currently permitted for the analysis of reinforced concrete structures, ACI Code §8.4 in its provisions on moment redistribution recognizes that indeterminate, ductile concrete structures that are overstressed at one section can use the reserve bending capacity of adjacent sections to carry additional loads. To illustrate the concept of a plastic hinge, a load-deflection curve of a simply supported, underreinforced beam with a concentrated load at its center is shown in Fig. 3,38a. Deflection at midspan is plotted against applied load P or against the internal moment M at midspan in Fig. 3.38b. Of course, M, a direct function of P, is equal to PL/4. As the load increases in size from zero to its ultimate Value P,, the load-deflection curve initially rises at a steep slope. When the load is large enough to create bending moments greater than the cracking moment of the cross section, the slope of the load-deflection curve reduces slightly, due to the reduced stiffness of the cracked beam. Point 1 in Fig. 3.380 is the point at which the slope changes. Above point 1 the load-deflection curve continues to rise almost linearly to point B, where the load P,, has created sufficient flexural deformation of the beam to strain the steel to its yield point. Between points A and B an increment of deflection, such as As, requires a substantial increment of load 86 REINFORCED CONCRETE esion RS rece hee Midspan deflection A @ o FIGURE 3.38 Behavior of an underreinforced beam: (a) moment curve and deflected shape, (b) load-deflection curve. ‘AP2. In this region of the curve the beam can be considered to have a stiffness Ky = AP2/A2. ‘As shown in Fig. 3.38b, Kz is equal to the slope of the load-deflection curve. ‘At point B on the curve, the load Py has just raised the stress in the steel to the yield point, Additional load above P, causes the neutral axis to shift toward the compression surface of the beam. Once the steel yields, little additional moment capacity is available and the slope of the load-deflection curve (still a measure of bending stiffness) reduces rapidly and approaches zero. ‘When the beam is loaded into the region between points C and D where the stiffness reduces to nearly zero, a plastic hinge is said to have formed. In the horizontal region between C and D in Fig. 3.38 an increment of deflection As is produced by an infinitesimally small increment of load APs, which is essentially zero in magnitude. If we again define stiffness of the member by the slope of the load-deflection curve, we can see that the stiffness is approximately zero. In other words, with the failure load in place, the beam loses all resistance to deflection by additional load. Such a beam, which may be supporting many tons of load, can be moved up ot down by the lightest pressure. The term plastic hinge is applied to this physical state s the behavior of a hinged joint, which also is characterized by zero resistance to rot Moment-Curvature Diagrams Revisited In Section 3.4 we examined the response of a beam to all levels of moment by constructing a ‘moment versus curvature diagram (see Fig. 3.11). At that point in our study of beams we had not developed either the procedure or the equations required to evaluate the moment M, associated with the failure of an underreinforced beam. In Section 3.7 we discussed the behavior of a beam loaded to feilure (see Fig. 3.33), and developed several methods to evaluate M,, the moment associated with the ultimate flexural strength. For an underreinforced beam with a rectangular compression zone, M, can be calculated either by Eq. (3.336) or by working directly with the internal couple as illustrated in Example 3.21. In Section 3.4 we showed that the slope of the strain diagram represents curvature. When ‘M, acts, the curvature at failure, yy, can be computed by considering the portion of the strain diagram above the neutral axis (see Fig. 3.376). Observing that the strain at the compression surface of the concrete equals 0,003, we can write (37) where ¢ = the distance from the neutral axis to the compression surface.1 ‘SIGN OF BEAMS Moment (hips) FIGURE 339 Simplified 0 ‘001 0.002 plot showing variation of Curvature y radians in) moment versus curvature, We can now evaluate the two critical points (initial yielding of the tension steel and final failure when the strain in the compression zone reaches 0,003) that define the simplified moment- curvature diagram. In addition we can compute the curvature-ductility ratio (CDR) given by Eq. (3.24). These computations are illustrated in Example 3.22. EXAMPLE 3.22. Construct the moment-curvature diagram and compute the CDR for the rectangu- lar cross section previously analyzed in Example 3.7. 6 = 12in,d = 17 in, Ay = 0.91 in’, fy = 60 kipvin®, f! = 3.6 kips/in’, Previously established in Example 3.7: My = 71.1 f-kips, iy = 0.00016 radians Solution. Compute M, using Eq. (3.34) to establish a = Ad oR 034 0.91(60) in 0.85(3.6)12 ed M, = A,fld ~ a2) 3.338) ‘My = 0.91 x 60(17 ~ 1.49/2) = 887.5 in kips = 73.96 ft kips Compute, using Eq. (3.37) _ 0.003 _ 0,003 9.00% = 0.0017 radiansfin | LIS Ye where c L73 in B and by Eq. (3.324), Br = 0.85 since f Using Eg. (3.24) 3.6 kips/in® is less than 4 kips/in® Yu co = $= aogTe 7 Note: A beam with a curvature ductility ratio of 10.7 can undergo large deformations before failure and is considered to be very ductile. The plot of moment versus curvature is shown in Fig. 3.39. 3.9 MINIMUM REINFORCEMENT TO PREVENT A BRITTLE FAILURE For most beams the cracking moment, whose magnitude is normally much smaller than that of the maximum moment produced by service loads, is a small fraction (in the 15 to 20 percent range) of the nominal flexural strength. As long as the maximum tensile stress in the concrete is8 RNFORGED CONCRETE less than the tensile strength of the concrete, given by the modulus of rupture, the beam remains uuncracked and the stress in the tension steel is low, of the order of 2 or 3 kips/in® (14 or 20 MPa), Although the stress in the reinforcement will increase after a crack forms because the steel must also carry the tension formerly carried by the concrete in addition to its initial stress, the steel stress is typically well below the yield-point stress; therefore, the initial cracking of the concrete and the subsequent redistribution of tensile stress to the steel normally occur without overstressing the reinforcement. n occasion, architectural or functional considerations may require beam dimensions to be set much larger than those required for flexural strength or for deflection control. Because of the large arm between the components of the internal couple, a beam of this type may require a very small area of reinforcement. As a result its nominal flexural strength may be less than the cracking moment of the cross section. If the cracking moment in a beam of this type is ever exceeded, ¢.., by accidental overload, the beam will fail immediately by rupture of the steel. To prevent this type of brittle failure, ACI Code § 10.5.1 requires that all beams be reinforced with a minimum area of steel to ensure that the nominal flexural strength will exceed the cracking moment by a safe margin (see exception on p. 80). Although the ACT equation for minimum steel is applied to cross sections of all shapes, the form of the variables and the approximate magnitude of the factor of safety can be established bby equating the cracking moment to the nominal flexural strength fora rectangular cross section (see Fig. 3.406). Using the internal-couple method, we can express the cracking moment as Mg = TeZh) .38) where 7. is the tension force carried by the concrete. Expressing T- in terms of the maximum tensile stress f, in the concrete gives fe liby (2 Mex i (by) 639) If the area of the steel is small relative to the gross area of the section, the depth of the stress block at failure will also be small; therefore, to evaluate the nominal flexural capacity of the cracked cross section, the arm of the internal couple can be taken equal to d approximately (see Fig. 3.40c), and My = Af (3.40) Equatng the racking moment given by Eq, (3.39) to the nominal flexural strength given by Eq. (3.40), expressing J; as7.5 J/f:, and noting that h ~ d gives 15 J Fildby 2 ee yd = Afid Solving for A, produces Fi A, = 12S vlna aan », é é 2 i © © © FIGURE 3.40 State of stress in a rectangular beam: (a) cross section, (6) cracking impends,(c) at failure.To establish a minimum area of steel for a concrete of average strength, assume that f= 4000 Ib/in? in Eg, (3.41); then 9b ud fr Ifthe right-hand side of Eq. (3.42) is multiplied by a factor of safety equal to approximately 2.5, Ay (now renamed A, nin) equals (3.42) 200b.d Again = SOOO 3.43) sis = Fa?) (43) Dividing both sides of Eq. (3.43) by byd gives the ACI code Eq, for the minimum reinforcement ratio as 2 2 uscu nin = 4 3.44 p La 44) Since the square root of f rather than f; is used in the computation for minimum steel, the factor of safety is not as sensitive to concrete strength {, as it would be if it were a direct function of /. Had a smaller value of f? been used instead of 4 kips/in? (27.58 MPa), a factor of safety slightly larger than 2.5 would be required to produce the coefficient 200 in the nu- merator of Eq. (3.44). Keeping the coefficient in Eq. (3.43) constant means that the factor of safety will increase as the compressive strength J.’ of the concrete decreases and decrease as the compressive strength of the concrete increases. Because the tension force T (see Fig. 3.40b) carried by the uncracked concrete is a function of ff! (see Eq. (2.3)], the minimum area of steel for high-strength concrete beams will be con- trolled by /. If the factor of safety of approximately 2.5 is to be maintained for high-strength concrete beams, we must develop an equation for minimum steel that is based on the concrete’s compressive strength f.. For the higher-strength concretes—those with a value of f.! exceeding 4440 Ib/in?—the ACI Code establishes an equation for minimum steel by multiplying Eq. (3.41) by a factor of safety of 2.5, which rounded off to the nearest whole number gives the ACI Code equation for Asin 8S i x SYS ef but not tess than 2% yscu 45a) nel S Fi Lobe ue pot tess than Se gy 3.450) Ay fy where f= 28-day compressive strength, Ib/in? (MPa) by, = the width of the web, in (mm) d = effective depth, in (mm) Table 3.3 shows the variation of paia, with J.) as required by Eqs. (3.45a) and (3.456). ‘When flexure produces tension in the flange of a statically determinate T-beam, which un- like continuous members lacks the ability to redistribute moment to adjoining sections whose flexural capacity is not fully utilized, a larger value of Asin than supplied by Eqs. (3.45) is required. For this case the ACI Code requires that A, nix must be equal to or greater than the smaller of the values given by the following two conditions. 6 Vfebwd fy a) Again = uscU st (3.46) 0 DESIGN OF BEANSENFORCED CONCRETE TABLE 3.3 Variation of pmin with f; as required by Eqs. (3.48a) and (3.456) uscU St A a i 4400 or less 2001, 306orless —LA/f, 5000 212/f, 35 LaBif, 6000 232/f, 40 LS8/f, 7000 -251/f, 50 17h, 8000 268/f, Oo L94/f, where by = the width of web, in (mm) or 2 Agmin is evaluated by Eqs. (3.45) with the value of b, in that equation set equal to by, the width of the tension flange, in (mm) The application of the ACI Eqs. for minimum steel is illustrated in Example 3.23. The provision for minimum steel does not have to be followed if the area of steel provided at every section, both positive and negative, is at least one-third greater than that required by analysis. Itis felt this extra area of steel provides sufficient extra capacity to ensure a safe de- sign. For beams with large cross sections, even the minimum steel requirement will require an extremely large area of flexural steel EXAMPLE 3.23. For the determinate T-beam in Fig. 3.41 determine the minimum area of flexurat steel required at the sections of (a) maximum positive moment, and (b) maximum negative moment. ff = 5000 Ibfin®, f, = 60,000 Ibvin?, and d = 15.5 in. Solution. (a) At the section of maximum positive moment, tension develops in the bottom of the web; therefore, use Eq. (3.454) BV Fibed _ 3. /500012(15.5) _ Asin ie 60,000 = 0:66 in? but not less than >, », { A 40in a} ee = 12in min ast — aeissin aM, Section A-A Moment = diagram FIGURE 341(©) At the section of maximum negative moment, tension develops in the top flange and Asai i established using the smaller value given by either Eqs. (3.46) or (3.45a) 6 Jfebed _ 6.5000 1215.5) Asai = Fe $000 = 1.32in? controls 3.46) _ 3vFibed _ 3V50004005.5) _, 5 Pa Oo 228 G.450) where b, = by = 40in 3.10 BALANCED FAILURE OF A BEAM WITH A RE CROSS SECTION CTANGULAR Introduction To ensure ductile behavior and controlled failures, the ACI Code permits only the design of underreinforced beams. The maximum area of steel permitted in a cross section can be developed from a consideration of the level of strain in the concrete at the instant the steel initially yields. Ductility depends primarily on the magnitude of the maximum strain in the concrete as the tension steel is stressed just to its yield point. If the strain in the concrete is small when the steel begins to yield, the beam can undergo considerable additional bending deformation before the concrete is strained to failure. As shown in Fig. 3.33c, after the tension steel yields, additional moment applied to an underreinforced beam causes the neutral axis to shift toward the compression surface. This movement of the neutral axis toward the compression surfaces causes the concrete strains to increase less rapidly while the strains in the steel increase more rapidly. Under these conditions, the beam behaves in a ductile manner and is able to undergo considerable deflection before failure of the concrete in the compression zone. As the percentage of steel used for reinforcement increases, the neutral axis moves toward the centroid of the tension steel and away from the surface of the compression zone. Since strains vary directly with distance from the neutral axis, the greater the distance between the neutral axis and the compression surface, the greater the strains in the concrete as the beam bends and the cross section rotates about the neutral axis (Fig. 3.42). Ductility disappears when the percentage of steel, called balanced steel Ay, is large enough to position the neutral axis so that the strain in the concrete reaches 0.003 just as the steel is strained to its yield point (Fig. 3.42c). With the concrete reaching its ultimate strain and crushing just as the steel yields, the moment capacity of the section is destroyed, and a brittle failure occurs In the next section the area of balanced steel will be established for a rectangular cross section. Computation of balanced stee! for a rectangular cross section is important, since most beams have this shape. Balanced Steel for a Rectangular Cross Section With balanced steel as reinforcement, the neutral axis is positioned so that the strain distribution in Fig. 3.43b results; the strain at the surface of the compression zone is 0.003, and the steel strain is €,. The position of the neutral axis can be established from the geometry of the strain diagram, By similar triangles in Fig. 3.435 0.003 _ ey | de a 81 DESIGN OF BEANS FOR FLEXURE{82__REINFORCED CONCRETE. ESIGN a a 60.008 dna a -—bna tenet — oo oo é& oe 7 = @ ® © G6 = 0003 Stain @ FIGURE 3.42 Influence of steel area on the magnitude of the strain in the top concrete fiber at the inital yielding of steel: (a) small A,; (6) moderate Ay; (©) balanced Ay; (d) stress-strain curves, s os Tt | ae - ona eco t text a 5 @ » FIGURE 3.43 Balanced failure of a rectangular beam: (a) cross section, (b) strains, (c) stresses, (d) intemal couple, (e) stress-strain curves; dots indicate the level of stress at failure in steel and in the top concrete fiber. Solve for cy @ ch Substitute e, = fy/E, _ 0.003 003 + ATE, d 8 &Multiply the top and bottom of Eq. (3) by Bs “ BE, + fe" w Letting E, = 29,000,000 Ib/in? in Eq. (4) gives a 87,000 47) 87,000 + f, (Ib/in?) Equation (3.47), which applies only to the case of a balanced failure, must not be used to establish the position of the neutral axis when less than balanced steel is used. The stress distribution shown in Fig. 3.43c follows from the strain in Fig. 3.43b. Multiply- ing the areas of steel and concrete by the stresses acting on them gives the equal and opposite components of internal moment, C and 7 (Fig. 3.434). By equating C and T, Ay, can be estab- lished T=C Acsfy = 0.85 flasb ee SBS hae (3.48) To eliminate ag, the depth of the stress block from Eq. (3.48), use a, = cp and the value of ¢» from Eq. (3.47) 0 give O.85Biflbd 87,000 Aw = "5 87,000 + fy (bln?) G49) Equation (3.49) can be put into a more useful form by dividing each side by bd 0.8581 f! 87,000. EE DE uscU - (50) asspiss 600 gy t +h where By is given by Eqs. (3.322), (3.32b), or (3.32c). ‘The term pp, the balanced reinforcement ratio, represents the area of steel per unit area of the concrete cross section. By using py instead of A,y, the dimensions of the cross section are eliminated. To ensure that concrete beams fail in a ductile manner, ACI Code §10.3.3 requires that beams in regions of low seismic activity be reinforced with an area of steel Ay that is not to exceed $Ay, As = YA (3.51) or, dividing both sides by bd, As. Aw bd ~ bd ex ip 52) where pp is given by Eq. (3.50). Values of 3p» for various grades of steel and strengths of con- crete are tabulated in Table 3.4. In zones of high seismic activity see ACI §21.3.2 for limit on p. 83 DESIGN OF RFAWSTABLE 3.4 Values of maximum reinforcement ratio }p5 for rectangular cross sections, no compression steel [see Eq. (3.50)] uscU fe lbsin?) Si (lbvin®) 3000 4000 5000 6000 40,000 0.028 0.037 0.044 0.049 50,000 0.021 0.028 0.032 0.036 60,000 0.016 0.021 0.028 (MPa) 4 (MPa) 40 50 300 0.024 0.036 o.044 0.049 350 0.020 0,029 0.036 0.040 400 0016 0.024 0.030 0.033 500 0012 0.018 0.022 0.024 3.11 STRENGTH DESIGN OF RECTANGULAR BEAMS FOR MOMENT From the basic principles and equations established in the preceding sections we now develop a procedure for designing a beam with a rectangular cross section. Since most reinforced con- crete beams used in construction are rectangular, this procedure will be used repeatedly by the designer. All steps are consistent with the requirements of the current ACI Code 318-95. All beams are designed to ensure that the moment produced by factored loads does not exceed the available flexural design strength of the cross section at any point along the length of the beam. If the flexural design strength ¢M, just equals the required flexural strength My (which ensures the most economical design), the criterion for design can be stated as = 6M, 3.53) where ¢ = 0.9 and M, is the nominal moment capacity of the cross section. This criterion can be developed into a design equation if we express M, in terms of the material and the geometric properties of a rectangular cross section (Fig. 3.44d). If we sum moments about the centroid of the tension steel, Mf, can be expressed as My = (a 3.54) where Cis the resultant of the compressive stresses and a is the depth of the rectangular stress block. As indicated in Fig. 3.44d, C = 0.85,fab. Substituting this value of C into Eq, (3.54) ap a comes FIGURE 3.44 (a) State of stress in an underreinforced ‘beam at failure: (b) strains, (©) stresses, (d) internal couple. o ©gives M, = 0. ss ao(a - 5) (3.55) ‘To express a in Eq. (3.55) in terms of the dimensions of the cross section and the properties of the material, f! and f,, we set T = C and solve for a, to give Ashy “= Faas) ee Multiplying both top and bottom of Eq, (3.56) by d and setting A,/bd = p leads to Aid _ phd “= FaO.85f) ~ 08572 657 Substituting Eq. (3.57) into (3.55) and simplifying gives - 2 Pfs = phybd (1 ay ) (3.58) Finally, Eq, (3.58) is substituted into Eq, (3.53) to give the basic beam design equation _ a(, phe My = dp fybd' ( He) (3.59) where p must not be greater than 3p or less than Pmiq associated with Ay min given by Eqs. 3.45) oF (3.46). The first requirement ensures that the beam will be underreinforced and will fail in a ductile manner; the second requirement prevents a brittle failure, ic., the rupture of the steel when the beam cracks initially Equation (3.59) can be used either to investigate the capacity of a cross section if the dimen- sions and material properties are known or to design a cross section (i.c., to establish the width , the depth d, and the area of steel A,) if the value of the factored moment M, is specified. Although Eq, (3.59) can be used to establish the flexural design strength of a cross section since all terms on the right side of the equation are known, the designer may prefer to work directly with the internal forces on the rectangular cross section to evaluate M,, because of the simplicity of the calculations. In the latter procedure, 7 = A, y is first evaluated, then the depth of the stress block a is computed by equating 7 = C, and finally the internal couple is evaluated by multiplying 7 by the arm d ~ a/2 between T and C. Both procedures are illustrated in Exam- ple 3.24. EXAMPLE 3.24. Investigation: Determining the Flexural Strength of a Reinforced Cross Section. Determine the maximum value of factored moment M, that the cross section in Fig. 3.45 ccan support. f? = 20.68 MPa, and f, = 344.7 MPa, b= 300mm FP ¢es00mm assy: o © FIGURE 3.45 (a) Cross section; (b) stresses at failure; (c) internal couple. 85 DESIGN OF BEAMS FOR FLEXURE.86 esto Solution, Compute Ay __ 2280 Bd ~ 3000400) ~ °°1° From Ei. (3.59) My = eonioe (1 - He) _ 0.019(344.7) = 0,9(0.019)(334.7 10° Nim? V0.3 my(Q.4 [1 STTEOEa = 0.230 x 10° N-m = 230 KN-m ‘As a check, recompute M, based on the statics ofthe intemal forces (see Fig. 3.45) T = Agfy = (2280 x 10°° m?)(344.7 x 10° Nim?) = 785.9 KN Since T = € 785.9 x 10° N = ab(0.85f,) = a(0.3 m)[0.85(20.68 x 10° N/m?)] a= 0149m M, = or(- s)- acres 9409 (04 - oe) 230 KN m ‘The Design Procedure ‘The procedure for establishing the proportions of a cross section for a particular value of My requires that the designer make a number of sizing decisions in certain variables of Eq. (3.59) at various stages of the design. (Additional discussion of the factors that influence the choice of materials and the dimensions follows this brief discussion of the design procedure.) First, the designer selects / and f,. On the right side of Eq. (3.59) the properties b, d, and p cf the cross section are unknown. With one equation involving three unknowns, the designer must specify two of the variables to achieve a solution, First, the designer selects a value of p—any value of p between Pmia and 3p, is satisfactory. The larger the value of p selected, the smaller the cross section will be. After p has been selected, Eq. (3.59) is solved for bd”, To produce specific values of b and d, the designer then specifies b and solves for d. Typically, an even value of b is selected, such as 10 or 12 in (250 or 300 mm). The width must provide enough clearance for the insertion ofthe specified reinforcing bars. For many floor systems of moderate span—say 15 0 25 fi (4.6 to 7.6 m)—an economical beam results when the depth equals 14 to 2 times the width, For long spans, deep narrow beams are often most efficient. If no limit exists on the depth, the depth may be made 3 or 4 times the width of the beam. ‘After d has been selected, the overall depth of the cross section is established by adding about 24 in (64 mm) to the effective depth d. This additional depth provides approximately 2 in (50 mm) of cover to protect the steel from fire and corrosion, The final dimension for the depth should be rounded off to the nearest whole inch (25 mm) to simplify fabrication of tke forms. If the resulting values of width and depth seem reasonable, the design is completed by computing A, = pbd.in which p represents the value of the reinforcement ratio previously selected by the designer. If the proportions of the cross section are not satisfactory (the beam is too shallow to control deflections, or the width of the web is too narrow to fit in all the steel), the designer must repeat the compntations after making appropriate modifications to the dimensions. After the first cycle of the design, the designer will know the approximate dimensions of the final cross section. In the second cycle, the designer may either specify the values of b and d and solve for p or select a new value of p. In either case, the designer should be able to arrive at acceptable proportions in the second cycle of design. The basic procedure for propor- tioning a cross section for a given moment is illustrated in Example 3.25. Example 3.26 shows the complete design procedure of a beam with the required checks for serviceability. A checklistfor the steps required to design a rectangular beam is given in the summary following Example 87 3.25. The designer should also keep in mind that bond stresses must be checked to verify that @ —sesqvor sous particular size of bar is properly anchored. Bond is covered fully in Chap. 6 and will be neglected pee at this stage, but it must be checked before the design is complete. EXAMPLE 3.25. Proportioning a Rectangular Cross Section for Moment. Use Eq, (3.59) to establish the width, depth, and area of steel fora rectangular eross section that must carry a moment M, = 150ft-kips (203.4KN- 1m). f = 3 kipsin? (20.68 MPa), f, = 60 kipsin? (413.7 MPa), Ji» = 0.016, and prin = 200/f, = 0.0033. Solution. The designer is free to select any value of p between pai and 34; try p = 0.012. From Eq. 3.59), Ma = ofybi°ll — (of JIL 150(12)in ips = 0.9(0 o2y60ed)[1 we] bd? = 32344 in? Try b = 10 in;then d? = 3234.4/10, and d = 18.0in Az = phd = 0.012(10)(18.0) = 2.16 in? The total depth is h +24 in = 1804 20.5 in use 21 in (533.4 mm) As another possible design, try b = 12 in; then d? = 3234,4/12 and d = 16.4 in Ay = phd = 0012(12\16.4) = 2.36in? (1522.7 mm?) ‘The total depth is head+2hin = 164+ 4 = 189in use 19 in(482.6 mm) If, after reviewing the dimensions of the previous cross sections, the designer decides to use a beam with b = 12 in 05 mm) and h = 18 in (457 mm), what area of steel is required? Estimated by deducting 24 in from hyd = 18-24 = 15.5 in, Use Eq. (3.61) to solve for p - 2} 150112) = o9peoyr2x158[1 ao} p=0.0138 OK. p 388 a33in 33 3.68 asco(is5~ 268)» stint = 15378 2084 “Br $M, = eae oM, From the strain diagram in Fig. 3.48 by similar triangles, we compute €, 0003 __e. Oe = gf. and €, = 0.0077 CONCLUSIONS. For a 60 percent increase in concrete strength, the computations lead to two con- clusions: (1) Increasing the strength ofthe concrete produces only an 8.8 percent increase in moment capacity. Since the tension force is the same in both beams because the area of steel and J, is the same, the difference in moment capacity is due to the small increase in length of the arm between 7 and C. (2) The reduction in depth of the compression zone for the cross section made of higher- strength concrete results in a 108 percent increase in steel strain at failure. The larger strains result in a more ductile beam that can undergo much larger deflections before the secondary compression failure of the concrete occurs. Since ductility is especially important in seismic zones, where struc tural members can be subjected to large strains, ACI Code §21.2.4 requres that / not be less than 3000 Ibvin® for stone concrete or less than 4000 Ib/in? for lightweight concrete. 3.12 TRIAL METHOD The trial method, a procedure for determining the area of steel required (o reinforce a cross section for a particular value of moment, is based on estimating the arm of the internal couple.96 ‘REINFORCED CONCRETE Desicn ‘The procedare, which can be used for cross sections of any shape, produces approximate values of the required steel area. Since the procedure also permits the designer to compute an improved value of arm, an improved estimate of the required steel area can be made if the enalysis is repeated In the trial method the designer estimates the length of the arm between the internal couple (see, for example, Fig. 3.47) and then solves for the tension force T by using the basic relation- ship that the applied moment equals the flexural design strengi M, = 6 X arm and Ma : 2 T° Sxanm ed) where = 0.9, M, equals the factored moment, and the designer estimates the arm between the components of the internal couple T and C. To start the procedure the arm may be estimated as some fraction of d, the effective depth. For the initial estimate of the arm, 0.85d for a rectangular section and 0.95d for a T-shaped section are recommended. After Eq. (3.62) has been solved for TT, the approximate area of steel can be computed by dividing T by f,. To procuce a more accurate value of steel area, the components of the internal couple can be equated to provide a close estimate of the area A. of the stress block. In this step, C is expressed as the product of the area of the stress block and 0.85 f! Cc=T O85f[Ac = T Ac = oe 63) Oasz where A. = approximate area of stress block T. = approximate value of tension force predicted by Eq. (3.62) AJ = compressive strength of concrete ‘Once A; has been evaluated, the designer can locate the position of C (it passes through the cen- troid of A.) and recompute the distance between T and C. If the beam has a rectangular section, the designer can express the arm as d ~ a/2 and select values of the arm by guessing values of 4G. An initial value of a = 0.2d is recommended for a rectangular cross section, Regardless of the first assamption for the arm, two cycles of the trial method should be adequate to produce a close estimate of the required steel area. Computations that result in the area of steel increasing significantly with each cycle are an indication taat the beam is overreinforced. In this situation, the designer should verify that Ay does not exceed the value given by Eq. (3.51). Examples 3.28 and 3.29 illustrate the method for a nonrectangular and a rectangular cross section. For these problems it will be assumed that the area of stee! from the analysis i less than three-fourtks of the balanced stee! area. If this information is not given, the designer will also be required to compute the area of balanced steel to ensure that the cross section is underreinforced as required by the ACI Code (see Table 3.4), EXAMPLE 3.28. Use of the Trial Method to Determine the Required Area of Flexural Steel. If the cross section in Fig. 3.49 carries a factored moment M, of 115 ft kips (155.94 kN m), what area of steel is required? f) = 3 kips/in? (20.68 MPa), and f, = 60 kipsiin? (413.7 MPa). Solution. In the initial analysis we estimate the arm of the internal couple. Try arm = 085d Arm = 0,85(15.5) = 13.2 in (335.3 mm)44 inn in ” FI O85 [DESIGN OF BEAMS a= 4.8inl Tera oases FOR FLEXURE Am=d-¥ t b= iin @ o o FIGURE 3.49 (a) Cross section, (b) stresses, (c) internal couple, ‘Compute the approximate value of T using Eq. (3.62) My 115(12) 1 xan ~ 09032) 116.2 kips (516.86 kN) Approximate value of A, Ty _ M162kips Be ~ okipslin? ‘To improve the value of A,, repeat the analysis with a revised value of arm. Use the distance ‘between the centroid of the tension steel and the centroid ofthe compression zone. Compute A- using, Eq, (3.63) with T = 7) from the initial analysis Ty_ _ _1162kips O.85f! — 0.85(3 kips/in*) Establish the depth ofthe stress block from the geometry of the compression zone. The distance that the stress block extends below the notch equals. 45.6 ~ (3(4)(2) 2 a=3+18= A= 1.94 in? (1251.7 mm?) Ac = 45.6 in* (29,421 mm?) 18 in(45.7 mm) 8 in (242 mm) Locate the centroid of the stress block by summing moments of the stress-block area about a horizontal axis through the top of the cross section AY = BAnyy 3(4)(1.5)2) + 21.63 +09) 456 Repeat the analysis for A, using the new value of arm, where the arm is 2.64 in (67 mm) d—Y = 155 — 2.64 = 12.86 in(326.6mm) M._ __115(12) 1 Dixasin ~ T9cia86) ~ 1193 Kips A, = 2 = 11823 _ | 99 in? (1284 mm?) OK; litle change from value of % 0 Use two n0.9 bars; Aysup = 2 in® 1,94 in? from initial trial EXAMPLE 3.29. (a) Use the trial method to determine the area of tension steel required by the cross section in Fig. 3.50 to carry a factored moment My = 180 ft-kips (244.1 kN m). (6) Compare results to an exact solution in which the nominal flexural strength M, is expressed by Eq, (3.33a); <= 3 kipslin? 20.68 MPa), and f, = 40 kips/in? (275.8 MPa).