DESIGN OF PRESTRESSED CONCRETE STRUCTURES THIRD EDITION NED H, BURNS, Prefer of Cal Eger, The Ur JOHN WILEY & SONS New York Chichester Brisbane Toronto SingaporeANOTE TO THE READER ‘Tris book has been electronically reproduced from igital information stored at John Wiley & Sons, In. We are pleased thatthe use of this new technology will enable us to keep works of enducing scholarly value in print as long as there's a reasonable demand for them. The content ofthis book is identical to previous printings. Copyright © 1955, 1963, 1961, by John Wiley & Sons, Inc ‘Al rights reverved, Published simultaneously ia Canada Reproduction or translation of any part of {his work beyond that permitted by Sections 107 and 108 of the 1975 United States Copyright ‘Act without the permission of the eopyriaht ‘owner is unlawful. Requests for permission ‘of further information should be addressed to the Permissions Department, Joha Wiley & Sons. Library of Congress Cataloging in Publication Data Lin, Tung yeu, 1911+ : Design of prestressed concrete Structures Includes bibliographies and indexes. 1. Prestressed concrete construction 1. Burns, Ned Hamilton, 1932- joint author. IL. Title. TASESSLS 1980 693.542 80.20619 ISBN 0-471-018988, Printed in the United States of America 9 Printed and bound by Quinn - Woodbine, Inc. To engineers who, rather than blindly following the codes of practice, seek to apply the laws of nature“PREFACE > The development of prestressed concrete can perhaps be best described by this parody, which Lin presented before the World Conference on Prestressed Conerete in San Francisco, 1957. All the world’s a stage, ‘And all engineering techniques merely players: ‘They have their exits and their entrances. Prestressed concrete, like others, plays a part, Its acts being seven ages. At first the infant, ‘Stressing and compressing in the inventor's arms. Then the curious schoolboy, fondly created By imaginative engineers for wealthy customers; Successfully built but costs a lot of dough. Then the lover, ‘Whose course never runs smooth, embraced by some, ‘Shunned by others, especially building officials. Now the soldier, Produced en masse the world over, quick to fight ‘Against any material, not only in strength, but in economy as well. ‘Soon the justice—codes and specifications set up.to abide, Formulas and tables to help you decide. No more fun to the pioneers, But so prestressed concrete plays its part. The sixth age Shifts to refined research and yet bolder designs, ‘Undreamed of by predecessors and men in ivory towers. Last scene of all, in common use and hence in oblivion, Ends this eventful history of prestressed concrete {As one of engineering methods and materials Like timber, like steel, like reinforced concrete, Like everything else. When Lin wrote the first edition of this book, prestressed concrete in the United States had barely entered its fourth stage—the beginning of mass production. Now it has emerged from the sixth stage and into the last. This rapid advancement has made possible a rather thorough revision of the previous two editions This edition presents the basic theory in prestressed concrete similar to the second edition. The method of load balancing is discussed side by side with the wih Frelave working-load and the ultimate-load methods, forming a tripod for prestressed concrete’ design. The chapters on materials, prestressing systems, and loss of prestress have been brought up to date. An important section on moment-curvature relation- ships has been added for flexural analysis. The problems of shear and bond, having been codified since the second edition was written, are now based on substantial experimental findings as well as theoretical interpretations. Camber and deflection are more thoroughly discussed with respect to time-dependent properties of concrete, The coverage of analysis and design of continuous beams with draped posttensioned tendons is more complete. Load-balancing with idealized tendons is covered along with the analysis using actual tendon layout in numerical example problems. Since building codes and bridge specifications on prestressed concrete are now pretty well standardized, thé ACI Code with its latest revisions is used in design examples. Bridge designers should refer to AASHTO Bridge Specifica- tions for some minor adjustments concerning allowable stresses and load fac- Some selected problems, which the authors have used for their undergraduate and graduate classes, are presented in Appendix E. Solutions for these are available to faculty members upon request. We believe that this revised edition, with its up-to-date contents, will continue to be used both as a text and a reference for engineers interested in prestressed concrete. ‘The assistance of Martha Burns and Stephanie Burns with manuscript typin, ea Chang and Sung L. Lam with SI. units is graeully acknowl Berkeley, California and Austin, Texas 1981, TY. Lin Ned H. BurnsCONTENTS INTRODUCTION MATERIALS, PRESTRESSING SYSTEMS; END ANCHORAGES LOSS OF PRESTRESS; FRICTION ANALYSIS OF SECTIONS FOR FLEXURE DESIGN OF SECTIONS FOR FLEXURE ‘SHEAR; BOND; BEARING. CAMBER, DEFLECTIONS; CABLE LAYOUTS PARTIAL PRESTRESS AND NONPRESTRESSED REINFORCEMENTS 10 continuous BEAMS 11 LOAD-BALANCING METHOD 12 sass 13. TENSION MEMBERS; CIRCULAR PRESTRESSING 14. COMPRESSION MEMBERS; PILES 15 ECONOMICS; STRUCTURAL TYPES AND LAYOUTS 16 DESIGN EXAMPLES APPENDIX A DEFINITIONS, NOTATIONS, ABBREVIATIONS APPENDIX B_ DATA FOR SOME PRESTRESSING SYSTEMS APPENDIX GC CONSTANTS FOR BEAM SECTIONS APPENDIX D PRESTRESSED CONCRETE LOSS OF PRESTRESS CALCULATIONS APPENDIX E PROBLEM STATEMENTS INDEX COVE KAONH 0 or 27 126 188 at 290 325 302 428 461 488 517 aS 597 610 ou 6281 INTRODUCTION 1-1 Development of Prestressed Concrete ‘The development of structural materials can be described along three different columns, as in Fig, 1-1. Column 1 shows materials resisting compression, starting with stones and bricks, ttn developing into concrete and more recently high-strength concrete. For materials resisting tension, people used bamboo, and ropes, then iron bars and steel, then high-strength steel. Column 3 indicates materials which resist both tension and compression, namely, bending. Timber was utilized, then structural steel, reinforced concrete and finally prestressed conerete was developed. The main difference between reinforced and prestressed concrete is the fact that reinforced concrete combines concrete and steel bars by simply putting them together and letting them act together as they may wish. Prestressed concrete, on the other hand, combines high-strength concrete with high-strength steel in an “active” manner. This is achieved by tensioning the steel and holding it against the concrete, thus putting the concrete into compression. This active combination results in a much better behavior of the two materials. Steel is ductile and now is made to act in high tension by prestressing. Concrete is a brittle material with its tensile capacity now improved by being compressed, while its compressive capacity is not really harmed. Thus prestressed concrete is ‘an ideal combination of two modern high strength materials. The historical development of prestressed concrete actually started in a different manner when prestressing was only intended to create permanent vompression in voncrete to improve its tensile strength. Later it became clear that prestressing the steel was also essential 10 the efficient utilization of hhigh-tensile steel. Prestressing means the intentional creation of permanent stresses in a structure or assembly, for the purpose of improving its behavior and strength under various service conditions. Throughout this chapter, photographs of significant structures designed in prestressed concrete using this fundamental concept will be shown. Note that the basic idea and the high-strength materials have now become a vital part of ‘modern structural engineering practice. The basic principle of prestressing was applied to construction perhaps centuries ago, when ropes or metal bands were wound around wooden staves to form barrels (Fig. 1-2). When the bands were tightened, they were under tensile2 Introduction MATERIALS, MATERIALS —__ MATERIALS RESISTING. RESISTING RESISTING TENSION COMPRESSION TENSION AND COMPRESSION ‘STONES Tanigoos am BRICKS mores | [_TMee TmoNsaas | [StaucToRAL cononere | | sreEL WIRES STEEL, REINFORCES PASSIVE Once oNaINATION ‘CONCRETE faiGa-stREncTH] [wicH—sraencTa] rconcrere | [STEEL per _-[PRe-sraesse9| conination eoncnere OEVELOPMENT OF BUILDING MATERIALS Development of building materials. Fig. 1-1 prestress which in turn created compressive prestress between the staves and thus enabled them to resist hoop tension produced by internal liquid pressure. In other words, the bands and the staves were both prestressed before they were subjected to any service loads. The same principle, however, was not applied to concrete until about 1886, when P, H. Jackson, an engineer of San Francisco, California, obtained patents for tightening steel tie rods in artificial stones and concrete arches to.serve as floor slabs. Around 1888, C. E. W. Doehring of Germany independently secured 2 patent for concrete reinforced with metal that had tensile stress applied to it “ft 7 compressive [OT staes E CTT 50 ksi maximum from Table 48 (section 4-12) Use $0 ksi losses for design in this case. 202.5 —50= 152.5 ksi (1051 N/mam?)(vs. 1534 ksi= 1058 N/mm? from [PBEAM analysis at 3 years or 152.5 ksi= 1051 N/mm? at 5 years, Table 49) (6) Shown in the following summary table are the losses computed by PCI Commit! tee General Method (see Appendix D for detailed calculations for each of the time steps * Comestion fr dy 631 ka, which changes our ctimate of, We would etimatej, 65161524 bs The ACLASCE Comantes woulé consider the maximum los of 30 Bees almost exacy th sme y= 2025-50 1525 4-13. Elongation of Tendons Itis often necessary to compute the elongation of a tendon caused by prestress- ing. When fabricating the anchorage parts in some systems, the expected amount of elongation must be known approximately. For the Prescon and other button-head, wire-type systems it must be known rather accurately. For all systems the measured elongation is compared to the expected value, thus serving as a check on the accuracy of the gage readings or on the magnitude of frictional loss along the length of the tendon. The computation of such elonga- tion is discussed in two parts as follows. Neglecting Frictional Loss along Tendon. If a tendon has uniform stress f, along its entire length L, the amount of elongation is given by A,=8,L=f,L/E,FL/E,A, (423) For prestress exceeding the proportional limit of the tendon, this formula may not be applicable, and it may be necessary to refer to the stress-strain diagram for the corresponding value of 8.122 Loss of Prestress; Friction Before any tendon is tensioned, there almost always exists in it a certain amount of slack. For systems requiring shim plates, such as the Prescon system, that slack must be allowed for when computing the length of the shims. In addition, it may be desirable to allow for the shrinkage and elastic shortening of concrete at the time of tensioning. Hence the length of the shims must equal the elastic elongation of the tendon plus the slack in the tendon plus the shortening of concrete at transfer. Conversely, the elastic elongation of the tendon must be computed by deducting the initial slack and the elastic shortening of concrete from the apparent elongation. tis not easy to determine the slack in a tendon accurately, hence the usual practice is to give the tendon some initial tension f,, and measure the elongation A, thereafter. Then, neglecting any shortening of the concrete, the total elastic clongation of the tendon can be computed by Elastic elongation= (4-24) a Sha? EXAMPLE 46 A Prescon cable, 6 ft long, Fig, 412, is to be tensioned from one end to an initial prestress of 150,000 psi immediately after transfer. Assume that there is no slack in the cable, that the shrinkage of concrete is'0.0002 at time of transfer, and that the average ‘compression in concrete is 800 psi along the length of the tendon. E,=3,800,000 psi; E,=29,000,000 psi. Compute the length of shims required, neglecting any elastic shorten ing of the shims and any friction along the tendon (span= 18.3 m, initial prestress= 1034 1N /mnm?, £,=262 N/mm? and E,=200 KN/mm*). ‘Solution’ From equation #10, the elastic elongation of steel is A, =f.L/,= 150,000% 60 12/29,000,000=3.72 in. (94.5 mm) Shortening of concrete due to shrinkage is (0.0002 60% 12=0.14 ia. (3.6 mm) Elastic shortening of concrete is £80060 12/3,800,000=0.15 in. (3.8 mm) Length of shims required is 3.7240.1440.15=401 in, (101.9 mm) If shims of 401 in, (101.9 mm) are inserted in the anchorage, there should remain an initial prestress of 150,000 pai (1034 N/mm?) in the stel immediately after transfer. Shortening of eee rene Tendon i Benetton of steel = 372° ta oem alr] of beam etre prestesing wo" Fig. 4.12, Example 4-6. Elongation of Tendons 123 EXAMPLE 4.7 Eighteen 0.196-in. wires ia a Freyssinetexble, 80 {¢ 244 m) long, are tensioned initially to a total stress of 3000 Ib. What additional elongation of the wires as measured therefrom is required to obtain an initial prestess of 160,000 pai (1103 N/mm")? E, = 28,000,000 psi (193 KN mm). Assume no shortening of concrete during the tension- ing process and neglect friction, ‘Solution A, =18X0.03 0.54 sq in. (348 mm?) fa =3000/0.54 = 5500 psi (37.9 N/mm*) Total elastic elongation of tendon from 0 to 160,000 psi (1103 N/maz) is SL /E, = 60,000 80% 12/28 000,000 = 5.48 ia. (139 mm) From equation 4-11, 4,=5.28in. (134mm) ‘Thus, with zero reading taken at a total stress of 3000 Ib (13.3 KN), an elongation of 5.28 in, (134 mm) must be obtained for a prestress of 160,000 psi (1103 N/mm?), Considering Frictional Loss along Tendon. It was shown in section 4-11 that, for a curved tendon with a constant radius R, the stress at any point away from the jacking end is Fea enteorks) ‘The average stress F, for the entire length of curve with stress varying from F, to Fycan be shown to be EBERLE This equation is solved graphically in Fig. 4-14, where the dotted lines give the values of f, = F,/A,. “The total lengthening for length L is given by BL eoht EA, wa*KL (4-25) (4-26) If only an approximate solution is desired, the medium value of F, and F, can be used in computing the elongation; thus AtAL 7 Ee (427)124 Loss of Prestress; Friction - ote end A 8 & \as/* or Vv Fig. 4.13. Example 4.8 EXAMPLE 4-8 ‘A teadon 80 ft 244 m) long is tensioned along a circular curve with Re 102 ft 311m), Fig. 413. For a unit stress of 180,000 psi (1.241 N/mm) applied atthe jacking end, & toa elongation of 480 in. (122 mam) is obaied £,=20,00)000 psi GOT KN), Compute the ses fy a the far end ofthe teadon. hg ean popes eermgen fu=B,E,/L=4.80% 30,000,000 /(80% 12) ‘= 150,000 psi (1034 N/mm?) Sincé the maxizium stress is 180,000 psi (1241 N/mm?), the minimum stress fg must be 120,000 psi (827 N/mm), assuming uniform decrease inthe stress Exact solution can be shown to give fy =125 ksi (862 N/mm). References 1. “Recommendations for Estimating Prestress Losses,” Report of PCI Committee on Prestress Losses, J. Prestressed Conc. Inst, Vol. 20, No. 4, July-August 1975, pp. 43-75. 2. P. Zia, H. K. Preston, N. L, Scott, and E. B. Workman, “Estimating Prestress Losses,” (ACI-ASCE Comm. on Prestressed Concrete Recommended Procedure), Cone. Int, Vol. 1, No. 6, June 1979, pp. 32-38. 3. R. Sinno and H. L. Fur, “Computer Program for Predicting Prestress Loss and Camber," J. Prestressed Conc. Inst, Vol. 17, No. 5, September-October 1972, pp. 21-38, 4, C. Suttikan, “A Generalized Solution for Time-Dependent Response and Strength of Non-Composite and Composite Prestressed Concrete Beams,” Ph.D. Disserta- tion, The University of Texas at Austin, August 1978 5. H. D. Hemandez and W. L. Gamble, “Time-Dependent Losses in Prestressed Concrete Coastruction,” Structural Research Series No. 417, University of Ilinois, Urbana, May 1975. “Tentative Recommendations for: Prestressed Concrete,” Report by ACI-ASCE Committee 423, J. Am. Conc. Inst, Wol. 54, No. 7, January 1958, pp. 548~578. 7. AASHTO Interim Specifications— Bridges—1975, Subcommittee on Bridges and Structures, American Association of Highway and Transportation Officials, Washington, 1975, pp. 41-79. Elongation of Tendons 125 Posttensioning Manual, Posttensioning Institute, Phoenix, Arizona, 1976, p. 189. D. D. Magura, M. A. Sozen, and C. P. Siess, “A Study of Suess Relaxation in Pressing Reinforcement,” J. Preniesed Cone, Insts Vol. 9, No.2, Apri 1964, pp. E.H, Cooley, “Friction in Posttensioned Prestressing Systems,” and “Estimation of Friction in Prestressed Concrete,” Cementand Concrete Assn., London, 1953. T. Y. Lin, “Cable Friction in Posttensioning,” J. Structural Dio, Am, Soe. of Civil Engineers, November 1956. F. Leonhardt, “Continuous Prestressed Concrete Beams," J. Am. Conc. Inst, March 1953 (Proc., Vol. 49), p. 6175 ANALYSIS OF SECTIONS FOR FLEXURE 5-1 Introduction and Sign Conventions Differentiation can be made between the analysis and design of prestressed sections for flexure. By analysis is meant the determination of stresses in the steel and concrete when the form and size of a section are already given or assumed. This is obviously a simpler operation than the design of the section, which involves the choice of a suitable section out of many possible shapes and dimensions. In actual practice, itis often necessary to first perform the process of design when assuming a section, and then to analyze that assumed section. But, for the purpose of study, it is easier to learn first the methods of analysis and then those of design. This reversal of order is desirable in the study of prestressed as well as reinforced concrete. This chapter will be devoted to the first part, the analysis; the next chapter will deal with design. The discussion is limited to the analysis of sections for flexure, meaning members under bending, such as beams and slabs. Only the effect of moment is considered here; that of shear and bond is treated in ‘Chapter 7. ‘A rather controversial point in the analysis of prestressed-concrete beams bas been the choice of a proper system of sign conventions. Many authors have used positive sign (+) for compressive stresses and negative sign (—) for tensile stresses, basing their convention on the idea that presiressed-concrete beams are normally under compression and hence the plus sign should be employed to denote that state of stress. The author prefers to maintain the common sign convention as used for the design of other structures; that is, minus for compressive and plus for tensile stresses. Throughout this treatise, plus will stand for tension and minus for compression, whether we are talking of stresses in steel or concrete, prestressed or reinforced. When the sense of the stress is self-evident, signs will be omitted. 5-2 Stresses in Concrete Due to Prestress Some of the basic principles of stress computation for prestressed concréte have already been mentioned in section 1-2. They will be discussed in greater detail here, First of all, let us consider the effect of prestress. According to present Stresses in Concrete Due to Prestress 127 Practice, stresses in concrete due to prestress are always computed by the elastic theory. Consider the prestress F existing at the time under discussion, whether it be the inital or the final value. If Fis applied at the centroid of the concrete section, and if the section under consideration is sufficiently far from the point of application of the prestres, then, by St. Venant's principe, the unit stress in concrete is uniform across that section and is given by the usual formula, F oa where 4 isthe area of that concrete section For a pretensioned member, when the prestress in the steel is transferred from the bulkheads to the concrete, Fig. 5-1, the force that was resisted by the bulkheads is now transfered to both the stel and the concrete in the member The release of the resistance from the bulkheads is equivalent to the application of an opposite force F; to the member. Using the transformed section method, and with A, =net sectional area of concrete, the compressive stress produced in the coneretis sz (5-1) while that induced in the ste! is nF _ nk, homage 62) which represents the immediate reduction of the prestress in the steel as a result of the transfer. Although this method of computation is correct according to the elastic theory, the usual practice is not to follow such a procedure, but rather to consider the prestress in the steel being reduced by a loss resulting from elastic Section of During Transfer Member Fig. 5-1. Transfer of concentric prestress in a pretensioned member.