Scribd Logo
Upload

Welcome to Scribd!

  • 93 S05 PDF

    Uploaded by

    bllld
    14 views12 pages

    Original Title

    93-S05.pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    14 views12 pages

    93 S05 PDF

    Uploaded by

    bllld

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 12
    Title no. 93-85 TECHNICAL PAPER A General Shear Design Method wes by Michael P. Collins, Denis Mitchell, Perry Adebar, and Frank J. Vecchio A simple, unified method is presented for the shear design of both pre stressed concrete members and nonpresirssed concrete members. The ‘method can treat members subjected 10 axial tension or axial compression ‘and treats members with and without web reinforcement. The derivation of the method is summarized and the predictions ofthe method are compared with those of the current ACI Code Keywords: aggregate interlock: axial loads: building codes; erack width and spacing; reinforced concrete; shear stength; structural design. ‘The shear design provisions of the 1995 ACI Code! con- sist of about 43 empirical equations for different types of ‘members and different types of loading, some of which are illustrated in Fig. 1. In 1973, the ACI-ASCE Shear Committee? expressed the hope that these “design regula- tions for shear strength can be integrated, simplified, and given a physical significance.” As shown by the growth in the number of ACI shear design equations (see Fig. 2), the ‘code has not met this desirable goal. Iti interesting to note that, prior to 1963, the ACI shear design procedure was so simple that only four equations were required. Most of the shear design equations given in Fig. 1 were in troduced in either the 1963 or 1971 edition of the ACI Code." ‘These design equations were developed in the period follow- ing the 1955 air-force warehouse shear failures® and rely on the traditional concept of adding a concrete contribution V, to the shear reinforcement contribution V, calculated on the basis, of the 45 deg truss equation. Since 1971 there has been an intensive research effort aimed at improving design methods for shear (see Fig. 3), The re~ search has shown that, in general, the angle of inclination of the concrete compression is not 45 deg, and that equations based on a variable angle truss provide a more realistic basis for shear design. In addition, tests of reinforced concrete pan- els subjected to pure shear improved the understanding of the stress-strain characteristics of diagonally cracked concrete, ‘These stress-strain relationships made it possible to develop ‘an analytical model, called the modified compression field theory, that proved capable of accurately predicting the re- sponse of reinforced concrete subjected to shear. 36 The objective of this paper is to present briefly a simple, general shear design method based on the modified compres- sion field theory. This design method, recently introduced by Collins and Mitchell,’ has been adopted by the Ontario Highway Bridge Design Code,* the Canadian Standards As- sociation Concrete Design Code,’ and the AASHTO LRFD specifications." The method is summarized in Fig. 1 SHEAR RESPONSE OF CRACKED CONCRETE ‘Tests of reinforced concrete panels subjected to pure shear (see Fig. 4) demonstrated that even after cracking, tensile stresses exist in the concrete and that these stresses can sig nificantly increase the ability of reinforced concrete to resist shear stresses, Cracked reinforced concrete transmits load in a relatively complex manner involving opening or closing of pre-exist- ing cracks, formation of new cracks, interface shear transfer fat rough crack surfaces, and significant variation of the stresses in reinforcing bars due to bond, with the highest steel stresses occurring at crack locations. The modified compression field model attempts to capture the essential features of this behavior without considering all of the de- tails. The crack pattem is idealized as a series of parallel cracks all occurring at angle @ to the longitudinal direction In lieu of following the complex stress variations in the cracked concrete, only the average stress state and the stress state at a crack are considered [see Fig. 4(b) and 4(c)]. As these two states of stress are statically equivalent, the loss of tensile stresses in the concrete at the crack must be replaced by increased steel stresses or, after yielding of some of the reinforcement at the crack, by shear stresses on the crack in- terface. The shear stress that can be transmitted across the crack will be a function of the crack width, Note that shear stress on the crack implies that the direction of principal stresses in the concrete changes at the crack location, ACI Sractaral Journal ¥.93, No.1 Januaey-ebuary 1996 Reveved June 17, 198 and revived under Institute publication polices. Copy. right © 1993, American Concrete Institue. Allright eseved.tcludg he Mak. ‘oFeopes unless permission band from te copyright propetors.Ferinent ds ‘ston wil be publsd in the November December 1996 ACT Srcturol Journal ‘vce by Jay 196, ACI Structural Journal / January-Febuary 1996 ‘Michael PColing, FAC. ie Ban Tanenbaum Professor of Ci Engineering ot he Univers of Toronto, Toronto, Canada. He is a member of ACI Commitee 388. Con ‘rete Guida the ACI Technical Actisties Commie subcommice on High Per formance Concrete and joint ACEASCE Commitee 445 Shear and Torsion He is member ofthe Canadien Standards Arsociation Commie forthe Design of Com free Scares Denis Michell, ACI, is a profesor in the Department of Cit Enginering and Applied Mechance at McGill University He i @ member of ACI Commer 408, Bond and Developmen of Reinforcement. and ACFASCE Commitee 445, Shear and Torsion He is Chairman of the Canadian Standards Association Commie forthe Design of Concrete Sucre ACI memberBerry Adebar i an axsociete professor inthe Deparment of iil Engi reering othe University of British Columbia, Voncowver Canada, He i Secretary of JACK Commie 341, Earths Resistant Concrete Bridges and joint ACLASCE ‘computraded design of wnforced concrete ACI member Frank J. Vecchio is a professor in the Department of Ci Enginering ‘atthe Univers Tron, Hes a member of ACI Commatees 441 Reinforced Con ‘rete Cols and 47, Finite Element Aalyis. and ofthe CEB Comniter on Com stave Modeling. ‘The average principal tensile strain e, in the cracked con- crete is used as a “damage indicator” that controls the aver age tensile stress fin the cracked concrete, the ability of the diagonally cracked concrete to carry compressive stresses f;, and the shear stress v,; that can be transmitted across a crack. ‘The principal compressive stress in the concrete fis relat- ed to both the principal compressive strain €, and the princi- pal tensile strain e, in the following manner [see Fig. 5(a)] h ay where Pamax = 4.7 (08 + 1108,) $f.’ S ACI Method General Method ¥, naVet% y, ve, +¥, ¥, (191 + 2500, % ¥, 538 0E ba ot ¥, ¥ < 81h bya o vent = 08[Fibyd + y+ yA ol na Ana oa 2\E ba (00 = MoM oe on 2 r0 9, « IE ce oe ™, and, < Yeu = (980 + 03%,) te ¥, ons, Sh ae where 8 and 8 are functions of the strain, ¢,, shear stress, », and crack spacing s, where M, = a = pu vy AT bd 1.9 fff + 2500p, Axial Tension and Shear : : Detailing Rules © Feintorcement shall extend © shear at cutoff < 2/3 shear permitted, or 8 tirup a 4, 60 f, 8 § d/8By, OF teinforcemen shall be inited so that es eitle beyond the pont at which its no longer required to resist exure fora distance equal fo the eflecive depth ofthe member or 124, ahichis greater, (© Flexural reinforcement shall not be terminated in a tension zone unless in excess of that required for shear and torsion, is provided © for #11 bars or smaller: shear atthe cutofl 3/4 shear permitted and continuing reinforcement provices double the area required for flexure a the cutot, © At simple supports and points of inflection, the diameter ofthe positive moment tension Detailing Rules Longitudinal steo! must be Setalled so that + -05u- soso Fig. I Comparison of ACI and proposed shear design approaches ACI Structural Journal / January-Febuary 1996 37 50 of 10 4900 191010201000 [NUMBER OF EQUATIONS FOR SHEAR DESIGN IN ACI CODE 1940 Fig. 2—Number of ACI shear design equations 1950 1960 197019801980 seo 1810 a0 100 180 1950 1960 197019001000 Fig. 3—Research into shear design methods (©) Catcutated average stresses __(¢) Local stresses at crack Fig. 4—Reinforced concrete panels subjected 10 shear From Eq, (I) the principal compressive strain for the loading portion ofthe stress-strain relationship is 0.002 (1 = {1 2/famax) 3) where €;” has been taken as -0.002. ‘After cracking, the principal tensile stress in the conerete fis related to the principal tensile strain ¢, as follows [see Fig. 5(b)] Ser 1+ /500e, where the cracking stress f,, can be taken as 4,/7," psi (0.33 ff,’ MPa). For large values of €,, the cracks will become and the magnitude of f, wll be controlled by the yielding, of the reinforcement atthe crack and by the ability to transmit shear stresses v; across the cracked interface (see Fig. 5(b)] ‘The shear stress that can be transmitted across the crack is a function of the crack width w and the aggregate size a [see Fig. 4()}, as given by 216K" ~ aw oT 08 fi “ psi and in, For MPa and mm units, replace the 2.16 by 0.18 and the 0.63 by 16. ACI Structural Journal / January-Febuary 1996 vo transverse tensile strain eo T,aterackalip (0) Average tensile stresses in cracked concrete a a function of, Fig. 5—Stress-strain relationships for cracked conerete If the stirrups have reached their yield stress and the crack begins to slip, the average tensile stress in the con- crete fj is limited to f, tan © The previous stress-strain relationships, together with equilib- rium and compatibitity, can be used to predict the load-deforma- tion response of reinforced concrete beams subjected to shear."! In addition, these relationships can be used as the basis for non- linear finite element formutations."2" DESIGN OF STIRRUPS FOR SHEAR In applying the modified compression field theory to the de- sign of beams, itis appropriate to make a number of simplifying assumptions. As illustrated in Fig. 6, the shear stresses are assumed to be uniform over the effective shear area b,d,, The highest longitudinal strain e, occurring within the web is used to calculate the principal tensile strain €,, For design, €, can be approximated as the strain in the flexural tension re- inforcement. The determination of , for a nonprestressed beam is illustrated in Fig. 7. Fora prestressed concrete mem- ber, the concrete surrounding the reinforcement will remain in compression until the applied tension exceeds the pre- stress force Ap, fyyr WheTe fn isthe stress in the tendon when the surrounding concrete is at zero stress. In lieu of more ac- curate calculations, f,, can be taken as 1.10 times f,. Hence, for design ACI Structural Journal / January-Febuary 1996 o (o) (o Cross-section sheer — Longitudinal Sess strains bs (9) ) Biaxial strains Tension in inweb web reinforcement Fig, 6—Beam subjected to shear, moment, and axial load ‘Axial Load Fig. 7—Determination of strain €, for nonprestressed beam (M,/d,) + 05N, + 05V, C0 ~ Apdoo 5 4 7 o EA,+EA;, atibility, the principal tensile strain e; can be related to the longitudinal strain e,, the direction of the principal compressive stress @, and the magnitude of the principal compressive strain &) in the following manner 1 = E+ (E,-8,) core 8) Hence, as the longitudinal strain e, becomes larger and the inclination 6 of the principal compressive stresses becomes smaller, the “damage indicator” e, becomes larger. ‘The nominal shear strength V,, of a member can be expressed as Ah V, = V.+V,+V, =f,b,d,cot0 +d, cot + V, 39 A = Bhi, +d, cout sv, From the expressions for the average tensile stress in the cracked concrete [Eq, (4) and (6)], the tensile stress factor B. can be determined as dco ¢ 2.16 1+ ,/500€, 103406 psiandin. (10) For MPa and mm units, replace the 4 by 0.33, the 2.16 by 0.18, and the 0.63 by 16. The crack width w is taken as the crack spacing times the principal tensile strain €). It can be seen from the previous expressions for B that as the tensile straining of the concrete increases (i.e., € increases), the shear that can be resisted by tensile stress- sin the concrete V, decreases, The value of the principal tensile strain e, will depend on the magnitudes of the lon- gitudinal strain e,, the principal compressive strain €, and the inclination @ of the principal stresses [see Eq. (8)] Strain e; can be found from Eq. (3). In using thi the principal compressive stress f can be conservatively tak~ enas v(tan@ + cotd) ay where (12) Table 1—Values of @ and { for members with web reinforcement ; Congitadinal sain €, > 1000 =0_| 5025 [S050 | <1.00 | 21.50 | 2200 coos) Sue | 270 | 285 [290 | 360 [a0 | #0 B | sas | 349 | 251 | 223 | 195 | 172 ~coors| Sse | 270 [275 [300 | 360 [400 | 320 B | 488 | 301 [247 | 216 | 190 | 16s we | 235 3x 3 ooo | 288 265 | 905 | 360 | 380 | 390 B_ | 326 | 258 [241 | 200 | 172] 14s ~050(V,+V,) (16) ‘where the minimum requirement is "Nese Tor vals 2 MPa nits divide piven vals By 12 40 Ash : eh 20 fF wi ACI Structural Journal / January-Febuary 1996 A = BYf’b,d, +L cot, (9) From the expressions for the average tensile stress in the cracked concrete [Eq. (4) and (6)], the tensile stress factor B can be determined as Sea 2 psiandin. (10) + [5008 in 1034206 For MPa and mm units, replace the 4 by 0.33, the 2.16 by 0.18, and the 0.63 by 16. The crack width w is taken as the crack spacing times the principal tensile strain It can be seen from the previous expressions for B that as the tensile straining of the concrete increases (i.e., increases), the shear that can be resisted by tensile stress- es in the concrete V, decreases. The value of the principal tensile strain €; will depend on the magnitudes of the lon- gitudinal strain €,, the principal compressive strain €, and the inclination @ of the principal stresses [see Eq. (8)]. Strain €) can be found from Eq, (3). In using this equation, the principal compressive stress f; can be conservatively tak- enas (tan + cot8) ay where «2y Table 1—Values of 0 and (i for members with web reinforcement ; Longitudinal sain €, > 1000 =0_| 2025 [5050 | £1.00] 31.30 | 2200 Geez | 270 | 285 | 290 | 360 | 410 | 430 B | 438 [349 [2st | 223 | 195 | 172 Gace | 270 | 275 | 300 | 360 | 400 | 420 B | 488 | 301 | 247 | 216 | 190 | 165 Ouez | 235 | 265 | 305 | 360 | 380 | 390 B_| 326 | 258 [241 [209 [iz [14s Beez | 250 | 290 | 320 | 360 | 365 | 370 B | 25s | 245 [228 | 193 | 150 | 126 Ouez | 275 | 10 | 330] 45 | 350 | 360 50.030] <0.075 0.100] <0.130] 0.200] B | 24s [233 [210 | tse | 121 | 100 oaso| See | 300 [320 [330 | 355 [385 | as B_ | 230 | 201 | to | 140 | 130 | 135 Touse Eq. (9) to determine the required stinups, the designer needs to determine appropriate values of @ and . For this pur- pose, Table | gives suitable values of 6 and B as functions of the longitudinal strain €, and the shear stress level vif’. While the values in Table 1 were calculated assuming a diagonal crack spacing of 12 in, (305 mm) and a maximum aggregate size of 3/ 4 in, (19 mm), itis believed that these values are appropriate for the full range of beams containing stirups. ‘The 0 values given in Table I have been chosen to insure that the stirrup strain g, is at least equal to 0.002 and to insure that, for highly stressed members, the principal compressive stress fy in the concrete does not exceed the crushing strength Fomaxe Within the range of values of @ that satisfy these re- ‘quirements, the values given in Table 1 will result in close to the smallest amount of shear reinforcement. While the values in Table 1 can be applied to a range of values of e, and vif.’ (e.g., 8 = 36 deg and B= 2.09 can be uused provided that e, is not greater than 1 x 10° and v/f.’ is not greater than 0.10), they were calculated for the up- per limits of the range. Linear interpolation between the values given in Table | could be used, but it is usually not worth the effort. ‘At a particular section of a member subjected to V,, My. and N,, the required shear strength is determined from V,sov, aay where the strength reduction factor 6 can be taken as 0.85. ‘The amount of stirrups required at the section can then be found from Eq. (9) as -V,-V, : as) While this calculation is performed for a particular sec~ tion, a shear failure caused by yielding of the stirrups in- volves yielding the reinforcement over a length of beam about d,cot® long. Hence, the calculations for one section can be taken as representing a length of beam, d,cot® long, with the calculated section being in the middle of this length. Thus, near a support, the first section to be checked is the section 0.5d,cot® from the face of the sup- port. Near concentrated loads, sections closer than 0.5d,cot9 to the load need not be checked. As a simplifi cation, the term 0.5d,cot® may be taken as d,. Since 1963, the ACI Code has required that at least a minimum area of stirrups be provided whenever V, exceeds one- half of the shear strength provided by the concrete. For the design method presented in this paper, it is recom- mended that a minimum area of stirrups be provided if V,>050(V,+V,) 16) ‘where the minimum requirement is Ad, ‘Noe for values in MPa nits divide piven valves By 12 40 pazon fps ACI Structural Journal / January-Febuary 1996 For MPa units replace the 0.72 by 0.06. DESIGN OF LONGITUDINAL REINFORCEMENT Fig. 8 illustrates the influence of shear on the tensile forces, in the longitudinal reinforcement. While the moment is zero at the simple support B, there still needs to be considerable tension in the longitudinal reinforcement near the support. ‘The required tension in the bottom reinforcement at Support B can be determined from the free body diagram in Fig. 8(b) by taking moments about Point C and assuming that the ag- _gregate interlock force in the crack that contributes to V, has ‘a negligible moment about Point C. For this nonprestressed ‘beam, the tensile force required at the inner edge of the bear- ing area is T= (Z-osy, Jota ayy Eq, (17) gives the additional tension due to shear. Hence, at a section subjected to a shear V,,,a moment M,, and an ax- ial force Nj, the longitudinal reinforcement on the flexural tension side of the member must satisfy My Ne (Vu AL tApsdys2 ap tO +( G-OSV,- V, Joot0 (18) oa," 8 ‘At maximum moment locations, the shear force changes sign and hence, the inclination of the diagonal compressive stresses changes. At direct supports and point loads, this change of inclination is associated with a fan-shaped pattern of compressive stresses radiating from the point load or the direct support, as shown in Fig. 8(a). This fanning of the di- agonal stresses reduces the tension in the longitudinal forcement caused by the shear (i., angle @ becomes steeper). Due to this effect, tension in the reinforcement does not exceed that due to the maximum moment alone. MEMBERS WITHOUT WEB REINFORCEMENT In evaluating the B factors given in Table I, it was assumed that the diagonal cracks in webs containing stirrups would be spaced about 12 in, (305 mm) apart. For members not contain- ing web reinforcement, this assumption may be unconservative; hence, itis inappropriate to use the B factors in Table 1 to eval- uate the shear strength of members without web reinforcement, For members without stirrups, the ability of the cracked con- crete to transmit shears primarily govemed by the width of the diagonal cracks {see Eq, (10)]. The crack width can be taken as, the principal tensile strain €, multiplied by the crack spacing. Hence, fora given value of ey, the shear strength will be a func~ tion of the crack spacing, with more widely spaced cracks re- sulting in lower shear capaci Fig. 9 illustrates the assumptions made in this design method concerning the crack spacings. For members without stirups, the diagonal cracks will become more widely spaced as 0 ap- proaches zero. The crack spacing when @ = 90 deg is called 5, ‘and this spacing is primarily a function of the maximum dis- tance between reinforcing bars or between reinforcing bars and the flexural compression zone. ACI Structural Journal / January-Febuary 1996

    You might also like