CHAPTER 6 ———— DESIGN FOR SHEAR AND TORSION as 6.1 INTRODUCTION The effects of external loadings on structures seldom lead to pure flexure alone. Other actions are simultaneously induced and include shear, torsion, and axial forces. Shear is most commonly encountered in combination with flexure. Its consideration in design logically follows that of flexure and represents a major step, as shown in Figs. 4.1 to 4.3. Torsion is dealt with only occasionally in everyday design. Basically, shear and torsion are different in nature: shear is a force and torsion is a twisting moment. However, they both lead to similar shearing stresses in the structure. Such stresses can be resolved into a principal tensile stress, also called diagonal tension, and a principal compressive stress. The diagonal tension is of the greatest concern, because it induces cracking in the concrete. To ensure that such acking does not lead to failure, transverse reinforcement resisting shear and/or torsion is generally provided in the form of stirrups or ties. Additional longitudinal reinforcement is also needed since both shear and torsion influence the demand on longitudinal reinforcement. However, the ACI code requires additional longitudinal reinforcement only in the design for torsion. Failure due to shear and/or torsion is brittle in nature and occurs with little warning: thus appropriate measures have to be taken to design for sufficient shear and torsion resistance. There is considerable body of research and analysis of the behavior of reinforced concrete in shear has taken place [Refs. 6.5 to 6.10, 6.20, 6.21, 6.27 to 6.30, 6.44, 6.45], and general theories have been established to cover the whole range of 303304 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN reinforcement, prestressed and non-prestressed. Examples include the truss model [Refs. 6.22 to 6.26, 6.33], the softened truss model developed by Hsu [Ref’. 6.15 to 6.19], the compression field theory developed by Mitchell and Collins and the modified compression field theory developed by Vecchio and Collins [Refs. 6.7 to 6.9, 6.37 to 6.39, and 6.46, 6.47], and generally strut and tie models [Refs. 6.42, 6.43]. Some details are given in Section 6.15.5. Based on these theories, special provisions exist in the ACI code and the AASHTO LRFD specifications for the design of reinforced and prestressed concrete members subjected to shear and/or torsion, in combination with bending. Information is given in this chapter to cover most common design problems involving shear and torsion of prestressed concrete. No attempt is made to use a sign convention for shear or torsional stresses. They are assumed positive throughout as treated in the code and various publications. Also assumed positive are values of specified factored strength and nominal resistance. 6.2 SHEAR DESIGN In the earliest codes for reinforced and prestressed concrete the design for shear was based on limiting the magnitude of diagonal tension under working loads, thus providing a safety factor against cracking. However, in prestressed concrete an overload may induce substantial changes in compressive stresses, thus leading to disproportionately high increases in diagonal tension at some points of the section and seriously jeopardizing the corresponding margin of safety. Moreover, the magnitude and direction of principal tensile stresses theoretically derived lose their applicability once cracking occurs. After cracking, considerable changes in stresses and stress distribution take place. A drastic increase in crack widths and decrease in stiffness is observed. This explains why, although principal tension stresses remain a good measure for evaluation, shear design is now considered at the ultimate limit state in both the ACI and AASHTO codes [Refs. 6.1, 6.2], and shear reinforcement is determined for factored loads. Because concrete retains a significant shear resistance after cracking (shear carried in compression zone, crack roughness and aggregate interlocking), the full contribution of transverse reinforcement takes place mostly after cracking, 6.3 PRESTRESSED VERSUS REINFORCED CONCRETE IN SHEAR It is often said that, with respect to shear, prestressed concrete offers two major advantages over reinforced concrete, namely: 1. For the same external loading, everything else being equal, the shear force in prestressed concrete is often smaller (because of the slope of the prestressing force) than that in reinforced concrete. This is illustrated in Fig. 6.1 where theChapter 6 - DESIGN FOR SHEAR AND TORSION 305 sign convention for shear adopted in this text is also given. It is observed that at any section x, the difference in shear between a prestressed concrete beam and an otherwise identical reinforced concrete beam is essentially due to the vertical component of the prestressing force V,=Fsina. V, generally acts in a direction opposite to the external loads, thus reducing the shear force in prestressed concrete, which at any section x is given by: V(x) = Vioads (2)— F sin (a) 6.1) (a) Sign convention for shear forces (+) + f a / pall | 7 External forces Internal forces, (b) Reinforced concrete | Reaction (0) Prestressed concrete NS Shear due to extemal load i os ane Vioads(*) HZ +Vp Wt Vpc(x)=Vioads(x)-F sin(ax) Vpc =Vrc -Fsin(a) Reaction Figure 6.1 Shear in reinforced and prestressed concrete.306 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN 2. Due to the compression induced by prestressing, the diagonal tension is smaller in prestressed concrete than in reinforced concrete. Furthermore, its angle of inclination @, with respect to the beam’s axis, is reduced. This implies that if cracking occurs and if, for safe design, the inclined crack is assumed to cross at least one stirrup, the stirrups’ spacing in prestressed concrete would be larger and their required area smaller than in reinforced concrete. Thus, a more economical solution (for shear) is obtained. The determination of principal tension is clarified in the next section. 6.4 DIAGONAL TENSION IN UNCRACKED SECTIONS Let us analyze, assuming elastic behavior, the stresses on a square element of main axis XX and YY taken along a concrete beam subjected to transverse loading. If the beam is plain or reinforced, an element taken along the neutral axis, such as element A (Fig. 6.2), will not be subjected to any axial stresses, but only to shear stresses. If the beam is prestressed, an axial stress will be exerted on 4 in addition to the shear stress. The analysis, in this case, is similar to that of an element B taken above the neutral axis of a plain or reinforced concrete beam subjected to a positive moment. 7 A Shear Bending stress stress profile profile Figure 6.2 Typical shear and bending stress profil along vertical plane YY. To find the magnitude of principal stresses along an element of beam, it is convenient to construct the Mohr’s circle of the element. The following general notation is adopted: o for normal axial stress, 0 (or v) for shear stress, 0) for principal tensile stress, o for principal compressive stress, XX for the horizontal plane, YY for the vertical plane, CC for the plane normal to the principal compression, and 77 for the plane normal to the principal tension.Chapter 6 - DESIGN FOR SHEAR AND TORSION 307 The Mohr’s circle of an element, taken along the neutral axis of a nonprestressed beam, is shown in Fig. 6.3a. The point of the circle representing each plane of interest is indicated. Note that XX and YY are diametrically opposed, thus bounding an are of angle 180° on the circle (that is, twice their actual angle of 90°). It can be observed that the magnitudes of principal stresses (points with zero shear ordinate) represented by points CC and 7T on the circle are equal to the magnitude of the shear stress, and their angles of inclination are at 45° with respect to the horizontal and vertical axes. (The actual angle of a plane is half its angle on the Mohr’s circle.) Also, the radius of the circle is equal in magnitude to the shear stress v. a te a 7 Principal | Principal \. tension compression aes plane); Axial () stress | x Cc (a) Nonprestressed element at centroid (RC) Shear stress A (v) Axial () stress Figure 6,3 Mohr’s circle for an element taken along the neutral axis.308 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Figure 6.3b gives the Mohr’s circle for an element prestressed along its x axis, It can be observed that the magnitude of principal tension represented by the distance from the origin O to point TT is substantially smaller than in the previous case. Assuming v and o, given, the radius of the circle can be determined from: 2 = 2 a; Ray +( ‘ ) (6.2) The principal stresses are then computed from: (6.3) (6.4) The angle of inclination y of the plane of principal tension with respect to the XX plane is obtained from (Fig. 6.35): 2 O_o ox tany = TL = Ox fy 4) Ox 65 ny (%) a It could also be obtained from: tan 2y = 22 (6.6) oy Note that, as tany from Eq. (6.5) is smaller than 1, y is smaller than 45°, There is one convenient method to determine the inclination of different planes from the Mohr’s circle. First, a pole point is determined: the pole is a point of the circle representing the intersection of a line parallel to a known plane and passing by its representative point on the circle. For instance, a line parallel to the horizontal plane and passing by point XX intersects the circle at the pole. The pole has the following interesting property: any line passing by the pole and another point of the circle is parallel to the plane of the element represented by that point. For instance, the line joining P to point TT in Fig. 6.36 gives the actual direction of the 77 plane. In comparing the results obtained in Figs. 6.3a and 5, it can be observed that both the principal tension and its inclination are smaller for a prestressed element than for a nonprestressed element. For the same shear stress v, we have:Chapter 6 - DESIGN FOR SHEAR AND TORSION 309 (01 prestressed = (tan ¥)(O} )nonprestressed (6.7) Mohr’s circle representation can also be very useful when biaxial prestressing is applied. This occurs, for instance, when, in addition to horizontal prestressing, vertical prestressing is used near the supports of long bridges. Such technique can substantially reduce or eliminate diagonal tension. A typical representation is shown in Fig. 6.4 for an element subjected to shear stresses and compressive stresses 0, and oy with o, > oy. The following expressions can easily be derived: ae Vertical prestressing Longitudinal prestressing Shear stress @) (6.8) (6.9) (6.10) (6.11) (6.12)310 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN It is interesting to note that the principal tension can be eliminated by properly selecting the value of oy, the vertical prestress. Setting oj = 0 in Eq. (6.10) and replacing R by its value from Eq. (6.8) leads to: (6.13) cx oy Note that, to compute the principal stresses, the values of v and o, (shear and axial stress) are needed. The determination of o, under the combined effects of prestressing and external loads was covered in Chapter 4. The determination of shear stress is treated next. 6.5 SHEAR STRESSES IN UNCRACKED SECTIONS The shear stress along any section of a flexural beam, assumed elastic uncracked, is given by (Fig. 6.5): (6.14) where: O = first static moment about the neutral axis of the portion of section outside the shear plane considered area of the portion of section outside the shear plane distance from neutral axis to the centroid of Ay, width of section at level y of shear plane considered J = moment of inertia of the cross section. Note that V is the shear force at the section of beam considered and, therefore, generally varies along the span. For a prestressed beam, V is the net shear (Eq. 6.1) due to the combination of applied load and prestressing. For a given value of V and section dimensions, the shear stress v, at a given section varies along the section. Typical shear stress diagrams for common sections are shown in Fig. 6.5. It is observed that the maximum value of v, occurs when the shear plane considered is taken along the neutral axis of the beam. The magnitude ofChapter 6 - DESIGN FOR SHEAR AND TORSION 311 maximum shear stress, also given in Fig. 6.5, can be put in the following form for rectangular and flanged sections: Parabola For by=ab and hp=ah Parabola hy =(h-hy)/2 wy 3] OUP ae +buhe |p, Vmax = 5) 3 _ 73 3 | 2bw| BCA — hp) + bh od So a. Voom ={ 28 = (32D? Parabola - S2 a a Vinay =| LOD? + DyD + DP “ ; oe Seat Dy Figure 6.5 Typical shear stress diagrams and maximum shear stresses for various sections.312 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Vv Umax = 4 (6.15) Duh where / is a factor larger than one. In computing an average shear stress from the shear force, the ACI code essentially adopts a value of 2 equal to one and uses d (dp or d,) instead of h. If cracking occurs, the shear stress distribution along the section will change. According to the analysis, it appears that the shear stress will be constant in the cracked region of the section. 6.6 SHEAR CRACKING BEHAVIOR Structural cracking occurs in concrete when its tensile strength is exceeded. As any external loading generally leads to combined effects such as flexure and shear, it is likely that more than one type of cracking will be critical depending on the variables at hand. An extensive number of investigations [Refs. 6.5, 6.20, 6.27 to 6.30, 6.44, 6.45] have shown or confirmed that two types of shear related cracks can develop in reinforced and prestressed concrete beams: flexure-shear cracks and web-shear cracks (Fig. 6.6a). The manner in which these cracks develop and grow strongly depends on the relative magnitude of shearing and flexural stresses. Flexure-shear cracking is due to a combined effect of flexure and shear. The corresponding cracks start as flexural cracks (normal to the beam’s axis). Then, due to the increased effect of diagonal tension at the tip of the crack, they deviate and propagate at an inclined direction corresponding essentially to the inclination of the diagonal tension plane. Typical flexure-shear cracks are shown in Fig. 6.6a. Flexure-shear cracking can lead to several types of failures [Refs. 6.4, 6.5, 6.12], schematically illustrated in Fig. 6.6). Very slender beams generally fail in flexure either by their tensile reinforcement or by the concrete compressive zone. However, in beams with smaller shear span-to- depth ratio, failure may occur by flexure-shear cracking before the flexural capacity is developed. In moderately slender beams, one of the cracks may continue to propagate until it becomes unstable, reaching throughout the depth of the beam and leading to what is described as diagonal tension failure. In relatively deep beams, a secondary crack, triggered by a flexure-shear crack, may propagate horizontally along the longitudinal reinforcement, leading to a loss of bond followed by a loss of anchorage near the support and subsequent failure described as shear tension failure. Alternatively, the concrete at the upper end of the crack may fail by crushing in what is described as shear compression failure.Chapter 6 - DESIGN FOR SHEAR AND TORSION 313 WA --—I Flexure-shear Web-shear cracks — v,; cracks — voy (a) Types of shear cracking Diagonal tension Web-crushing failure failure Loss of bond (Splitting failure) Crushing Example Shear tension Shear compression failure failure (b) Major types of shear failures Figure 6.6 Shear cracking and shear failures. Web-shear cracking (Fig. 6.6a) occurs when the magnitude of principal tension is relatively high in comparison to flexural stresses. It is characteristic of beams with narrow webs, such as I-beams, where cracking due to diagonal tension develops before flexural cracking. Web-shear cracking may lead to the same types of failures described for flexure-shear cracking, namely, diagonal tension, shear tension, and shear compression failures. In addition (Fig. 6.65), crushing under diagonal compression may occur within the web, leading to web-crushing failure. Figure 6.7 illustrates the above cracking classification and its schematic location within a typical beam. Understanding this figure is important, since the provisions of the ACI code for shear implicitly refer to this type of behavior.314 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Several models have been proposed to simulate shear cracking and failure mechanisms in reinforced concrete beams and include limit analysis mechanisms and analogies with arches, trusses or frames. The arch and truss analogies are illustrated in Fig, 6.8. The truss analogy is at the basis of the ACI code provisions to determine shear reinforcement. A more comprehensive treatment of each mechanism and their comparison can be found in Refs. [6.6, 6.10, 6.20, and 6.32]. It is much preferable to avoid shear failure, since it is substantially more brittle than flexural failure. To supplement the shear resistance of concrete members and to ensure flexural failure prior to shear failure, shear reinforcement (also called web or transverse reinforcement) in the form of stirrups is generally provided. The function of shear reinforcement is explained in more detail in Ref. [6.30 and 6.32]. i: Applied load _ PIATTTT + 5 Web- Web- Simple Continuous ties i Support _ shee a Flexure ‘and flexure-shear shear support Flexure and “flexure-shear | Figure 6.7 Typical cracking due to bending and shear in a reinforced concrete beam. (a) Arch analogy (0) Truss analogy Figure 6.8 Typical analogies for shear failure mechanisms. 6.7 SHEAR REINFORCEMENT AFTER CRACKING The approach used to determine shear reinforcement in prestressed concrete is identical to that used for reinforced concrete. It is assumed that a shear crack extends at 45° to the beam’s axis, as shown in Fig. 6.9. The force carried by the webChapter 6 - DESIGN FOR SHEAR AND TORSION 315 reinforcement is assumed to balance a principal tension across the crack equal in magnitude to the shear stress v. According to the ACI code, the portion of v, balanced by the shear reinforcement, is the shear stress in excess of the shear strength contribution of the concrete. Let us call it v,. It is also assumed that, at the onset of shear failure, the shear reinforcement will be stressed to its yield strength ty: @ Any such line should cross at least one stirrup (b) (c) Figure 6.9 Shear resistance provided by (a) Vertical stirrups, (b) Inclined stirrups, (c) Truss analogy model (see also Chapter 15).316 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Let us first consider the case of shear reinforcement placed normally to the beam’s axis (vertical stirrups) with spacing s (Fig. 6.9a). The crack can be divided in equal segments, each traversed by a single stirrup. A free body diagram of a typical segment is also shown in Fig. 6.9a. The length of the crack segment along its plane is given by (s/cos 45°). Projecting the forces on the vertical axis leads to the following equilibrium equation: Ayfy = buPsS | o95 45° = by bys (6.16) cos 45° from which the required area of vertical reinforcement (stirrups) is obtained: _ dyvys Ay ty (6.17) If the stirrups were placed at an angle @ to the beam’s axis (inclined stirrups), an approach similar to the above would lead to the area of inclined reinforcement given by (Fig. 6.95): Bybgs sin@ +cosa@) Ay (6.18) Although inclined stirrups are more efficient than vertical stirrups, the increased labor cost associated with their placing seldom justifies their use. Note that in designing the shear reinforcement according to the ACI code, the value of v, in Eqs. (6.17) and (6.18) is replaced by the shear demand needed at ultimate (v, /¢—v,) as explained in Sec. 6.8. Equations (6.17) and (6.18) were derived assuming a 45° shear crack. It is expected that a smaller angle will develop in prestressed beams, thus leading to a conservative design. ACI consider that an angle of 37.5 degrees is generally acceptable for prestressed concrete. The above results can also be simply derived using the truss analogy model of the web reinforced cracked beam in which the stirrups act as tension members of the analogous truss, and the concrete struts running parallel to the diagonal cracks act as compression members (Fig. 6.9¢). Assuming vertical stirrups, the shear at the section shown is given by: V, =T, s (6.19)Chapter 6 - DESIGN FOR SHEAR AND TORSION 317 where 7, is the tensile force in the stirrup (not to be confused with torsional moment), The web steel force per unit length is given by T,/s in which = jd, cot. Hence, ss jd, cot 6.20 ss jd, cord (6.20) Assuming, as a first approximation, jd, =d, and @ = 45° leads to: (6.21) 1A Arn i fee] Welded wire fabric, with at least two Primary | horizontal wires reinforcement pend aon T2in > 6dz >12dy — 4 ( 2 6dy 26d, 4 yy | £ 2.5 in goo» o> 135° No. 5 bar No. 6, 7, No. 8 bar and smaller or8bar and smaller Figure 6.10 ‘Typical shear reinforcement and standard hook dimensions for stirrups according to the ACI code.318 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Since stirrups are assumed to yield at ultimate, 7, is replaced by 4, fy. thus: 4 (6.22) in which: Ve =[Link] (6.23) Eq. (6.22) is identical to Eq. (6.17). Shear reinforcement in prestressed beams consists mostly of stirrups. A stirrup has generally two legs, leading to an area of vertical reinforcement twice the cross- sectional area of the stirrup bar. Typical shapes of stirrups and standard hooks and ties, as recommended by the ACI code, are shown in Fig. 6.10. Stirrups used for shear can be “open” stirrups but must be appropriately anchored in the compressive zone of the member. Because of relatively high labor costs associated with the placing of stirrups, an alternative solution consisting of welded wire fabric is often used with precast prestressed concrete products (Fig. 6.10). Commercial welded wire fabrics used for shear reinforcement have few longitudinal wires and selected sets of transverse wire spacing covering a practical range. It is common practice to assume that the vertical wires are properly anchored in the tensile zone if at least two welded junctions are provided. 6.8 ACI CODE DESIGN CRITERIA FOR SHEAR 6.8.1 Basic Approach As mentioned earlier, the ACI design approach is based on ultimate strength requirements (factored loads). A design shear strength at each section of a member is required. It is achieved by the combined contributions of concrete and shear reinforcement. The concrete is assumed to provide a prescribed shear resistance; the shear reinforcement (stirrups) provides the excess resistance, if any, needed to satisfy the required strength. In the 1971 edition of the code, shear related equations were given in stress units. From 1977 on, including the 2002 code, they are given in force units where the force is obtained by multiplying the shear stress by bd. Although the approaches used and the results obtained are identical, the author prefers the use of stresses, because they are independent of the size of the member and, thus, can be easily correlated when different solutions are compared. They also have the advantage of providing a warning or flag easily recognizable by the designer when aChapter 6 — DESIGN FOR SHEAR AND TORSION 319 limit is exceeded. However, each important equation is written in terms of both stresses and forces. The required factored shear at each section of the member should be less than or equal to its shear resistance, that is: {r Earthquake : ¥ ae ‘Tv, Ww ‘Spandrel . on ; Sail soction’| Figure 6.12 Typical support conditions for locating first critical section in shear design. For nonprestressed members the first critical section for shear is taken at a distance d, from the face of support, and the shear reinforcement required at that section is extended to the support. Figure 6.11 provides a justification for this assumption by showing that, for a typical shear crack inclined at 45 deg, loads placedChapter 6 - DESIGN FOR SHEAR AND TORSION 327 at a distance smaller than d, transfer directly to the support. For prestressed members, the first critical section for shear is to be taken at 4/2 from the face of the support and the same shear reinforcement required at //2 is to be used at sections located at a distance less than #/2. However, in some particular cases, the first critical section must be taken at less than 4/2, as illustrated in the lower part of Fig. 6.12 [Ref. 6.2]. Indeed these are sections where the shear load must be considered at the face of the support. In the design of stabs and footings, shear reinforcement is not required when Vy <$ve. Similarly to what is done for beams, the first critical section in each direction is to be taken at A/2 from the face of the support (or column). Simultaneously, punching shear must be checked along a critical section perpendicular to the plane of the slab with perimeter h, (Fig. 6.13) located a distance dy /2 from the perimeter of the concentrated load or reaction area. The value of vu, to be used for punching shear in prestressed and nonprestressed slabs is explained in Section 11.1.1. Perimeter b, 4 4 4¢-| — Figure 6.13 Critical section for punching shear in slabs. 6.9 DESIGN EXPEDIENTS Design for shear can be very time consuming, since one does not know a priori the locations of the critical sections which require more than the minimum amount of reinforcement. It is not uncommon to find out (assuming uniform loading) that excess reinforcement is needed at two sections along the span and is not needed at a section in between.328 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN ‘Select section abeissa x along beam 1 for normal weight concrete 0.85 for sand lightweight conerete 0.75 for all lightweight conerete (generally starting atx = Wi? fron Elaborate design method: = lesser of | Vow =3:52 “Alternate conservative method suggested by ACI when fe > 04 fy and for uniform loading only: d, 4, 2a, ssalh ad r= 0627 + 70074 where Yydp /M, <1 reinforcement not required Gaz] ti E oud, ‘Change concrete where d, 20.88 ie, larger b,,4,) Yes} Given 5 50.754 or 24 in Method 1 Method 2 Compute: (Av nin = 506,81 Fy ApsSous [dy 80fsdp Y bw where dj, is taken 20.8 and Din = A Assume a value ofs such that Consider 4, for a given type of 4/9 MD compute required 4, from > minimum then check if A, >(A wins ot repeat this step using (min larger of (40min Use s= 9/2 for same A, computed above, or double 4, for same s Repeat for \ yes /<> minimum \. No ‘Choose larger next section practical size stirrup Figure 6.14 Flow chart for the design of shear reinforcement in prestressed beams.Chapter 6 - DESIGN FOR SHEAR AND TORSION 329 No. 3 U stirrup fy = 60,000 psi Ay = 0.22 in? luylo~ vel. psi 24 10 Figure 6.15 Design chart for stirrup spacing. In most common design problems involving prestressed beams, it is very likely that the minimum amount of shear reinforcement (4,)miq at the maximum allowable spacing of 0.75h or 24 in (= 60 cm) will prevail throughout the span. If the alternative method to determine v, is considered, its lower bound value of 2) f2 can be used as a first approximation in design and checked against the value of v,/¢ at several sections. Only if significant differences exist, the more accurate analysis will be pursued. In order to ease the design procedure, a design flow chart for shear is proposed in Fig. 6.14. It applies to normal- and light-weight prestressed concrete beams and synthesizes most of the steps encountered if the ACI code approach is followed. As in each complete iteration only one section is considered, it is appropriate to organize the numerical results in a table, such as in Tables 6.1 and 6.2 of the example in Section 6.10. Once the computations are completed, the results can be directly used to determine stirrups’ areas and spacing. However, a graphical representation of the variation of shear stresses and strengths may give a better visual presentation of stirrup distribution along the span. Such a typical graph is shown in Fig. 6.17 for the330 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN example of Section 6.10. As only a discrete number of sections is analyzed, the graph can help isolate graphically the segments of the span where same shear reinforcement and spacing prevail. Often, in practice, a type of stirrup is selected for a given beam, say a #3 U stirrup, and the spacing is selected at each section according to the design requirements. However, only a discrete number of spacings is considered, such as 4, 6, 8, 12, 18 and 24 in (10, 15, 20, 30, 45 and 60 cm). Thus, one can go backward and, given 4, and a spacing, find graphically the value of (v, /—v,) resisted by the shear reinforcement and the corresponding segment of beam where such spacing applies. Charts can be developed to determine the value of (vu, /¢—v,) that can be resisted by a certain stirrup at a certain spacing. Using Eq. (6.35) for vertical stirrups, we have: (-».) Aly (61) é bys which can be plotted versus b,, for different values of s, assuming 4, and fy are given. Such a chart for a #3 U stirrup with f, = 60 ksi (414 MPa) is given in Fig 6.15. 6.10 EXAMPLE: DESIGN OF SHEAR REINFORCEMENT The T beam described in Sections 4.9.1, 4.12.3, and 5.18 is considered (Fig. 6.16). The design characteristics of interest to this example are: h = 40 in, by =8 in, 4, = 550 in, Ap. =1.53 in’, ¥%, =12.9 in, Z, =3028 in®, f= 5000 psi (normal weight concrete), F= 229.5 Kips, fie > 0-4 Spy» = 70 fi, live load = 0.4 Klf, superimposed dead load = 0.04 kif, self-weight wg = 0.573 kif For normal weight concrete, The final eccentricity of the prestressing force along the half span was arrived at in Sec. 4.12.3 and is given by: ey = 21.7 in for x2 28 ft 9413.8 for 0 LLTAY A, = 120 psi \ x In order to determine v,,,, we need the following: F _ 229,500 = 417.3 psi (Note: The centroid is in the web.) (@a)28 ~ (0 )support _ 21.7-7.9 28x12 28x12 0.04107 sina = tana = sin ar = 229,500 0.04107 = 9426 Ib Voy = 3.5K 4 030g + wd p 9426 8x32 The shear resistance v, is the smaller of uy. and Uy. Thus at.x= 2 ft = 3.55000 +0.3% 417.34 = 247.49-+125.19 + 36,82 = 409.5 psi 1b, = 409.5 psi In order to determine v,, we need V,,: u(S-9] = 1.3756(35—2) = 45.395 kips Thus: 45,395 @ Gb,d, 0.1Sx8x32 The value of v,/¢ is to be compared te v,/2andv,. As u,/G =236.43 psi is more than v,/2 but less than v, = 409.5 psi, only minimum reinforcement is needed at this section. Let us select a stirrup with f, =60 ksi, at a maximum spacing s = 24 in < 0.75h. As Spe > 94 fpy> the minimum area of shear reinforcement is obtained from: 236.43 psi333 Chapter 6 - DESIGN FOR SHEAR AND TORSION aaSiny st sonaypiyas yg'9.20 “p= 4p 7 - 1 (noynonp (pounbas ; Buiveds quias : , wy uw) : : : : : poptroad a< 16 sost a0 CROLEY | BRBRT | LLT rl Our ate) eed 0< sul 410 | er9sts | coves | a'sz | Or out (paunbas |"? 0 00 90 | pe6e9e | cise |scez] oc | ome} ay £10) 0< DALTESON LST Iso | wolzel | recep | eSzT] SE oct : 0< aaedon | 9c sorsp| siz] cz 40 %p 10) 5 sdiy | ut ae | os “M4 | “4p | aoueisiq | uonses, sao Tee] 0) oct wa (@uoeds (paninbas © | eae) oll oct | 87 = 078 “juan + | seor| szre| oct | . uw) : + | opt | seor] ozro | ici] : : 5 wiezo)| unos] 9< sost | sss'az| ori | coor] ozro| ori | ze | osze | sevece | sprit | ceozi | tt =ozr pauasn | 9< ssi | couse] var | sor} seve] rai | ze | core | corsuse | oswo1 | ocort | sot ~ ocr paxnbas | o< | sanetau | g00c | tise | esse | oor | ozve| ssc | ce | seve | srozet | oszet | prot |b =ocuz | o< |} eanetou | vszz | tee'er | soor | soor | ozro | oorr| ce | sso | oxose | eiz| sosi | se=or | o< | sanefou | pez | soesr | sor | soor| ozre | reve | ze | ezip | ivigs | vee | tot ca 4so A a0) | wo sd] asd | ody | asa diy u 510) : wa | sta | ta +84 | +x uonoag “poypau areroqeye ay) Buysn '9 ajdurexg 40) suoneyndmos uitsap 28ays yo Cammutng 1°9 998.334 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN 4 [dy 32 2 (nin = bos SpuS dp _1.53x270x24 [32 0.13 in? 80f,d, Yb, 80x60x32 V8 Note, a #3 U stirrup at a spacing of 24 in would provide 0.22 in? and thus would be adequate. The other sections covered in Table 6.1 lead to the same minimum reinforcement. If, instead of selecting the spacing in the above equation, we would have selected the area of a #3 U stirrup, the required spacing would have been s = 40.9 in. Of course, then the maximum allowable spacing of 24 in would control, leading to the s It The alternate equation for minimum shear reinforcement would have given: Pres _ 5 8*24 916 in? soe = 0.16 in fy 60,000 res Aven = larger of (A) ger [rb = gg 8x4 0.15 5 vie f, 60,000 It is also smaller than the reinforcement area of 0.22 in? provided. Note that for a given area of reinforcement at a given spacing the maximum shear stress to be resisted by the reinforcement can be determined. In this case, we have: (u-a)Js Arty _ 0.22% 60,000 69 py; bys 8x24 = 0.17 in? 6.10.2 Alternate Conservative Approach to Determine v, In order to compute v,, we need: ¥,, = 45.395 kips (calculated above) 1, XE=2) _ 13756268 _ 95 541 kipset= 1122.5 kips-in 21.8 in Ip Vudy = 0.8816 <1 OK Thus, for 2=1 0, = 0.64. fe +700 = 42.43 + 7000.8816 = 659.5 psi is larger than 22/7 =141.4 psi but should be taken fess than 52/72 use u, = 353.5 psi. Results for the other sections are summarized in Table 6.2. If we compare v,/¢ and v,, we find out that, similarly to the previous case, a minimum reinforcement is sufficient. \ 353.5 psi. Thus, 6.10.3 Design for Increased Live Load: Partially Prestressed Beam To illustrate a case where more than the minimum reinforcement is needed, let us assume that the specified live load for the previous example is increased from 0.4 kif to 1 KIf. Let us also assume that the beam is designed as a partially prestressed beam, whereas the amount and profile of the prestressing steel are the same as described above but, in order to balance the inerease in the required resistance, nonprestressed reinforcement is added. It amounts to four #9 bars with 4, = 4 in’, fy = 60 ksi, and d, = 36 in at midspan, For simplicity, use d, for the computation of v,.Chapter 6 - DESIGN FOR SHEAR AND TORSION 335 Table 6.3 Summary of shear design computations for Example 6.10.3. Section, [ AV, | AM, | Var | Vom | Ye | Var | Ou/& | oy Required x() | ips | kips-in | psi | psi | psi | kips | psi | @ ©? | Sfor psi Ay = 0.22 in? 0.5) 54.384 767.1 | 409.5 | 409.5 | 77.075 | 401.4 a 24 max. 120 51.912 446.6 | 409.5 | 409.5 | 73.571 | 383.2 ve | 24 max. 220-7 | 46.144 255 | 409.5] 255 | 65.397 | 340.6 | 86 19.2 31/20 40.376 184 | 409.5] 183 | 57.222 | 298 1s 143 41/20 =14 | 34,608 146 | 409.5] 146 | 49.048 | 2555 | 110 51/20 28.840 121 | 4095] 121 | 40873] 2129 | 92 18 61/20 =21 | 23.072 | 10,174.8 | 120. | 409.5 | 120 | 32.698] 1703 | 50 | 24 max. 70720 17.304 | 11,022.6 | 120 | 408.5 | 120 | 24.524] 1243 4 24 max. 81/20 =28 | 11.536 | 11,6283 | 120 | 372.7| 120 | 16349] 787 |... 24 max. 91/20 5.768 | 11,992.7 | 120 Rye ae a 39.4 24 max. w=35__| 0 12.1128 | 120_| 372.7| 120_| 0 0 24 max. 400 + 8 T Shear strass, psi 1 Stirrup spacing, in 8 T 100 + ~~ Sf Range of by alternative alternative method one on 0a Oa ’ Required s for (AJ) = 0.22 in? Spray = 24 it ob ae #24 in ab ' eae 2 e514 L L \ L 0 ov oa oa oa Ost Figure 6.17 Shear stress distribution and stirrups spacing for example beam.336 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN For this case, the values of Vg, dy, Mz, and V;, are the same as in the example of Section 6.10.1 and are given in Table 6.1. Due to the increase in live load, the following values of factored loads apply: = 1.2(0.573+ 0.04) +1.6x1 = 2.3356 kIf Avy, = 1.20.04 + 1.6% 1 = 1.648 kIf The corresponding computations AV,, AM,, Ucjs Voy Ver and the required spacing (Eq. (6.37)) of #3 U stirrups are summarized in Table 6.3. It can be observed that, although the live load has been substantially increased, the required amount of shear reinforcement is still relatively small. If the alternative conservative approach was used, the value of u, would have been the same as shown in Table 6.2. This is because the ratio ¥,, /M,, remains the same with an increase in uniform live load, The values of v,;, Ugys and vz, as well as the required stirrup spacing, are plotted versus x in Fig. 6.17. The shaded area corresponds to the part of the beam where stirrups are theoretically needed. It can be seen that the greatest need for stirrups is not near the support but instead in a region where flexure-shear cracking may control, here between 0.10/ and 0.257. 6 illustrate design of stirrups, consider the section at x = 31/20 from the left support. Using Eq (6.37) and No. 3 U-shaped stirrups leads to: Ay 0.22 x 60,000 gS 21435 in (vy /P=0,)by Sx Table 6.3 gives the results for the other sections. Figure 6.17 provides a graphical illustration of shear resistance by the concrete, shear demand due to loading, and required stirrup spacing along the span, Finally, note that, although shear may not be critical near the support, additional stirrups are needed to contain tensile splitting cracks in the end zone (see Section 4.17). 6.11 DERIVATION OF CONCRETE NOMINAL SHEAR STRENGTH EQUATIONS (ACI CODE) The derivation of Eqs. (6.29) and (6.31), on which the ACI shear design provisions for prestressed concrete are based, is given in the commentary of the 1963 ACI code. It is essentially based on the work of MacGregor, Sozen, and Seiss [Refs. 6.28 and 6.29]. For flexure-shear cracking, their tests have indicated that, in order to reduce the capacity of a beam, a diagonal crack must have a projection on the longitudinal axis of the beam at least equal to the depth of the beam d. (dis used here instead of d, to comply with the original notation; use d, for fully prestressed beams.) Flexure-Shear Cracking. Let us consider a section along the beam of abscissa x with respect to the support (Fig. 6.18). A flexural crack distant d from x in the direction of decreasing moments may lead to a diagonal crack which could be critical for section x. Tests have indicated that failure becomes imminent when a second flexural crack occurs at a distance d/2 from x and extends to the centroid of the section where the increase in the principal tensile stress triggers diagonal cracking. Collapse occurs for an additional load, the effects of which are described later. LetChapter 6 - DESIGN FOR SHEAR AND TORSION 337 us evaluate the load that led to the second flexural crack. The change in moment from section x to section(x—d/2) is equal to the area under the shear diagram between these two sections. That is: V4+Vey \d M-M,,=|——" |= 6.42, ‘er ( 2 jg (6.42) where M and V are the moment and shear at section x and M,, and V,, are the cracking moment and corresponding shear at section (x-d/2). The difference between Vand V,,. over d/2 is generally small. Thus, as a first approximation: M-M.,= “ (6.43) or M_ Me _4 (6.44) VoVve2 from which: — Mer (6.45) M/V -d/2 where V is the shear at section x. Note that V appears on the two sides of the equation. However, V can still be computed, because the ratio M/V remains constant at a given section when the applied load increases proportionately. — sy Diagonal shear | crack | —-—|-.- A Ze | | ! y \ i ! 1 i 21 Flexural >} cracks di J : i 1 : 1 Mer M Vor v Figure 6.18 Flexure-shear cracking model,338 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN It can be assumed that, in a beam test, the shear given by Eq. (6.45) is generated by external loads. In design, these loads are represented by superimposed dead loads and live loads. The corresponding shear acts in addition to the dead load shear Vg. Moreover, as mentioned earlier, tests have indicated that, in order for collapse to occur after development of the second flexural crack, a shear force increment estimated at 0.6/fb,d is needed. Hence, the total shear force V,;, which will produce a flexure-shear failure, can be estimated from: ORR. re (6.46) Ve M/IV-di2 The above relationship is compared to actual data in Fig. 6.19 [taken from ACI SP10, 1965]. To simplify the above equation, the term d/2 is deleted, leading to a safer design limit. As M, V, and M,, are due to additional external loads (superimposed dead load and live load), and as nominal shear resistance at ultimate is sought, they are replaced, using the notation of this text, by AM,, AV,,, and AM,,. Hence, Eq. (6.46) becomes: AM, een 6.47 a G (6.47) Voi = 0.64) fobyd + fe M,,/ AV, Vei — Vo bai NO aN @ © © ; ‘AM, _ n; ~[o. soa + am, [tt 3 Figure 6.19 Comparison of Eq. (6.46) with experimental data [Ref. ACI SP10, 1965].Chapter 6 - DESIGN FOR SHEAR AND TORSION 339 Dividing the two sides of Eq. (6.47) by byd leads essentially to the flexure-shear strength (stress) v,;, given by Eq. (6.29). Web-Shear Cracking. Web-shear cracking occurs when the maximum principal tensile stress becomes equal to the tensile strength of the concrete, f,.. The principal tensile stress is assumed maximum at the centroid of the section. Using Eq. (6.3) and assuming all stresses are positive (to keep up with the original derivation), the following relationship is derived to represent the onset of web-shear cracking: 2, Tie= fa + (*) ao (6.48) where o, is the compressive stress at the concrete centroid due to the effective prestressing force and v,,, is the critical shear stress at web-shear cracking. Solving for Vey leads to: og Yow = tefl + F, (6.49) tc 10 ae le “|e ele Vow 5 : JE 3 =>? fe + 0.30 a a at LK in psi units aa I l | | i 1 | 1 0 2 4 6 8 10 12 14 16 18 ogiNhe Figure 6.20 Nominal shear stress at web-shear cracking [Ref. ACI SP10, 1965].340 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN The magnitude of fj. indicated by tests appears to be about 4/7. It is conservatively set equal to 3.5\/f. Hence, Eq. (6.49) becomes: Vow = 3.5) fe, [1+ (6.50) The curve representing Eq. (6.50) is plotted in Fig. 6.20 and compared to that representing a simplified form given by: Vow =35Y fe +030 (6.51) As the two curves are close, the simplified form is selected for design, In prestressed concrete, the shear stress value given by Eq. (6.51) is augmented by the shear stress generated by the vertical component of the prestressing force. It essentially leads to the nominal web-shear cracking resistance Uv, given by Eq. (6.31). 6.12 AASHTO GENERAL PROCEDURE FOR SHEAR DESIGN The AASHTO LRED code recommends two possible approaches to design for shear: 1. Strut-and-tie model which applies at the level of the member and thus covers. simultaneously many sections or regions along the member. A brief introduction to strut-and-tie modeling is given in Chapter 15. 2. Modified compression field theory which applies at the level of the section, that is, the design is carried out section by section in a manner similar to the procedure followed by the ACI code and explained in Section 6.8. In the following treatment, only the sectional model is briefly summarized. It is based on the modified compression field theory which applies to shear as well as shear combined with torsion. The theory was initially developed by Mitchell and Collins (Refs. 6.7, 6.8, 6.9, 6.37, 6.38, 6.39], as the compression field theory, and later modified by Vecchio and Collins [6.46, 6.47]; it is an extension of the general truss model with the modified assumption that cracked concrete provides some resistance to shear. The main basis of the theory as reduced to code implementation is summarized in Fig. 6.21. The truss model, the struts at angle @, equilibrium of forces in the longitudinal and transverse directions, and stresses in the Moht’s circle are all illustrated in the figure, which will help understand the equations used in the code. Note that the figure relates to a section with zero moment. and describes the tensile force N, that develops in the longitudinal reinforcement due to shear only. The modified compression field theory leads to a procedure and equations totally different from those used by the ACI code. Unlike the truss model for shear,Chapter 6 - DESIGN FOR SHEAR AND TORSION 341 the modified compression field theory assumes that the concrete contributes some resistance in tension through concrete struts in between diagonal shear cracks. Since it was first adopted in the 1993 draft of the AASHTO LRED code, the procedure has seen several refinements and adjustments which are simply reported here without explanation, and in accordance with the AASHTO LRFD code, Interim 2002, Article 5.8.3 [Ref. 6.1]. To minimize confusion, the notation and sign convention of AASHTO is adopted in this section. Variation of tensile 0.5N, (a) Cross section (om Principal stresses and longitudinal equilibrium: N= tensile fores in Shear longitucinalreinforcement Stress ¥ ldue to shear sing Normal Stress (c) Average concrete stresses (@) Force in stirups Figure 6.21 Equilibrium in concrete strut and steel tie according to the modified compression field theory. (Adapted from Collins and Mitchell, 1997) 6.12.1 General Sectional Procedure for Shear Design In line with the general approach for LRFD or USD, the following equation must be satisfied at each section: Ve = GV 2 Vy (6.52) where: V, = design shear resistance V,, = nominal shear resistance342 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN ", = factored shear force = capacity reduction factor = 0.9 for normal weight concrete, and 0.7 for structural lightweight concrete. The nominal shear resistance at a given section is expressed as the sum of contributions from the concrete, the transverse reinforcement (i.e., shear reinforcement or strirrups), and the transverse component of the prestressing force, as follows: Ve +V,+Vy (6.53) contribution of concrete to member shear strength contribution of steel (that is stirrups or transverse steel) to member shear strength; in the modified compression field theory, this contribution results from the combined action of a concrete strut and steel tie, but is simply referred to as steel contribution = component of prestressing force in the direction of applied shear; it is assumed positive if it resists the applied shear and negative otherwise. The nominal shear resistance V;, given by Eq. (6.53) is constrained according to the following equation which sets an upper limit on the combined contribution of concrete and transverse reinforcement: Vn 0.25 ff by dy +Vp (6.54) where: ft: = design compressive strength of concrete at 28 days (ksi), taken less than 10 ksi b, = effective web width (in) taken as the minimum web width within the effective depth, d,, modified for the presence of ducts, where applicable. For posttensioned members, 4, can be computed from the web width minus half the diameter of ungrouted ducts, or one quarter the diameter of grouted ducts. d, = effective shear depth (in); it is taken approximately equal (d, ~ a/2) in which a is the depth of the stress block at nominal bending resistance of the midspan or critical section. The effective depth d, is also calculated from:Chapter 6 - DESIGN FOR SHEAR AND TORSION 343 (0.9d, d,, = larger of {or (6.55) 72h where: d, = effective depth for bending, that is, distance from extreme compression fiber to centroid of the tensile force in the tensile reinforcement at nominal bending resistance h = overall member depth, (in) Assuming vertical shear in a typical beam, the vertical or transverse component of the prestressing force is obtained from: Vp= Fsin@ (6.56) where q@ is the angle of inclination of the prestressing force with respect to the longitudinal axis of the beam. A similar approach can be followed if the prestressed member is not horizontal such as in the case of a transversally loaded column. The contribution of the concrete and transverse steel to the nominal shear resistance is estimated from: V, = 0.0316 B [fe bd, (6.57) Ay fy dy cord s (6.58) factor indicating the ability of diagonally cracked concrete to transmit tension between cracks A, = area of transverse reinforcing steel (in) within a distance s (in) fy = yield strength of the transverse steel reinforcement (ksi) 6 = angle of inclination of diagonal compressive struts (Fig. 6.21). Equation 6.58 implies stirrups at 90 degrees to the member longitudinal axis. The angle @ is assumed to be a variable with a possible range from about 24 to 45 degrees. In comparison, typically ACI assumes a constant value of @=45 degrees for reinforced concrete, and 37.5 degrees for prestressed concrete. The factor 0.0316 is equal 1/1000 and converts the numerical results from psi to ksi. Thus, expressed in psi, the shear stress resistance provided by the concrete is Bf. The coefficient # should be compared with the equivalent value used in the ACI code for reinforced concrete (#=2) and for prestressed concrete. If we consider the limits set by the ACI code for the expressions of v,;, and v, (Eqs. 6.29344 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN and 6.32) then we note that their equivalent # ranges from 1.7 (lower limit for v,;) to 5 (upper limit for v, by the conservative method). Comparatively in the AASHTO LRED code, the value of f ranges from 1.29 to 6.32 (Table 6.4) for members with at least minimum transverse reinforcement. In order to determine V. and V,, and Qare needed. For nonprestressed concrete sections not subjected to axial tension and containing at least the minimum amount of transverse reinforcement, or having an overall depth of less than 16 in (400 mm), the values of # and @ can be conservatively taken as 2 and 45 degrees, respectively, Otherwise, and for prestressed members, 8 and @are calculated from the following variables: ¢ v/ f@, that is, the ratio between the shear stress at ultimate, v (Eq. 6.59), and the concrete compressive strength, 2 © &,, that is, the longitudinal strain in the web, assumed positive for tension © sy, that is, the crack spacing; this parameter is needed only for members without transverse reinforcement. Physically, the longitudinal strain ¢, can be regarded as an average (smeared) strain in a reinforced concrete tensile tie (see also Remarks in Section 6.12.2). The shear stress v is determined from: (6.59) Assuming first an estimated value of the angle @ the longitudinal strain & is calculated from: ¢ For sections with at least minimum transverse reinforcement: M et OSN, +05 (Vy Vp) C00 Aps Spo : ry ee (6.60a) 2 (BsAy + Ep Ans) For sections with less than minimum transverse reinforcement: Me 050, +0.5 Vy Vp) cot = AnsSpo B Scena ayy (6.60b) Epps where: M,, = factored moment at section considered (kip-in); it should be taken 2VydyChapter 6 - DESIGN FOR SHEAR AND TORSION 345 N,, = factored axial force taken positive if tensile (kips) Spo = Stress in the prestressing steel (ksi) which can be conservatively taken equal to the effective prestress f¢ . (See discussion in the remarks at the end of this section.) E,,Eps = modulus of elasticity (ksi) of reinforcing steel and prestressing steel, respectively Ag, Aps = cross sectional area (in’) of non prestressed reinforcement, and prestressing steel on the flexural tension side of a member, respectively. If the calculated longitudinal strain & from Eq. (6.60) is found negative, its value should then be determined from: Mu 4.0.5Ny 40.5 Vy —Vp)C0t8 ~ Aps fn v (6.60) 2 (Ep Ay + By Ay + Ey Aps) where: ep = area of concrete (in”) on the flexural tension side of the member up to half its depth (Fig. 6.22) E,, = modulus of elasticity of concrete (ksi) Zz For members with at least minimum, transverse reinforcement | Area of tensile zone: Ag |- For members without or with less than minimum, 7 transverse reinforcement Figure 6.22. Definition of area of tensile zone. Compared to Eq. (6.60a), Eq. (6.60c) introduces in the denominator the contribution of concrete area, dy. A, is defined as the area of concrete on the flexural tensile side of the section, below half the depth of the section, as shown in Fig. 6.22. Thus for a section symmetrical with respect to the x axis, Ay =0.54.. For a composite beam, the / is that of the precast section.346 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Table 6.4 Values of @ and for concrete sections with at least minimum transverse steel reinforcement (from AASHTO LRED, Interim 2002). » &, 1,000 fe | $020] <0.10 | =0.05 [<0] <0.125 | £0.25 | <0.50 | <0.75 | =1.0] <15[ 20 =0.075 | 223 | 204 | 21.0 [218] 243 | 266 | 305 | 337 [364] 408) B9 6.32 4.75 4.10 | 3.75 3.24 2.94 2.59 2.23 | 1.95 | 1.67 0.100} 18.1 [204 | 214 [225] 249 | 27.1 | 30.8 | 34.0 | 36.7] 408 | 43.1 3.79 | 3.38 | 3.24 |3.14| 291 | 2.75 | 2.50 | 2.32 | 2.18 | 1.93 | 1.69 20.125 19.9 21.9 22.8 | 23.7 = ee eee 3.18 | 299 | 294 |297| 2.74 | 2.62 | 2.42 | | 2.13 | 1.90 | 1.67 =0.150] 21.6 | 233 | 242 [250] 269 | 288 | 32, 9 | 37.3 | 40.5 | 42.8 288 | 279 | 2.78 | 2.72| 2.60 | 2.52 | 2.36 | 2.21 | 2.08 | 1.82 | 1.61 0175 | 23.2 [247 | 255 | 262] 28.0 | 29.7 | 32.7 | 35.2 | 36.8 | 39.7 | 42.2 2.73 | 2.66 | 2.65 | 260| 252 | 244 | 2.28 | 2.14 | 1.96 | 1.71 | 1.54 £0200 | 24.7 | 261 | 26.7 | 274] 29.0 | 30.6 | 32.8 | 345 [36.1 | 392 | 417 2.63 | 259 | 252 |251| 243 | 237 | 2.14 | 1.94 | 1.79 | 161 | 1.47 26.1 27.3 27.9 | 28.5 30.0 30.8 ke ee ee | 414 2.53 2.45 2.42 | 2.40 | 2.34 2.14 1.86 ee <0250| 275 | 286 | 29.1 [297] 306 | 31.3 | 328 | 34.3 [35.8] 386 | 412 239 | 239 | 233 | 233] 212 | 1.93 | 1.70 | 1.58 | 1:50 | 1.38 | 1.29 Table 6.5 Values of @ and f for concrete set steel reinforcement (from AASHTO LRFD, Interim 2002). ns without or with less than minimum transver Sxe x1,000 (in) [<0.20 | <0.10 [<-0.05 [<0] <0.125 | <0.25 [<0.50 | <0.75 [=1.00 | <1.50 | =2.00 <5 [254 [25.5 [25.9 [264]27.7 [289 [309 [324 [337 [35.6 [37.2 6.36 | 6.06 | 556 [515/441 [3.91 | 3.26 | 2.86 | 258 | 2.21 | 1.96 10/276 [27.6 [283 | 29.3] 316 36.3 [384 [40.1 [42.7 [447 5.78_|5.78 | 5.38 | 4.89 | 4.05, 2.88 | 2.50 | 2.23 | 1.88 _| 1.65 <15]295 [29.5 [29.7 [311/341 399 [424 [444 [47.4 [49.7 534 | 5.34 [5.27 | 4.73 | 3.82 2.64 | 2.26 | 2.01 | 1.68 | 1.46 $20] 31.2 [31.2 [31.2 732.3 )36.0 42.7 [455 [478 [509 | 534 499 | 4.99 | 4.99 | 4.61 | 3.65 246 | 2.09 | 1.85 | 1.52 | 131 30/341 [341 [341 [34.2 [389 46.9 | 30.1 | 52.6 | 363 | 59.0 445 | 445 | 445 | 4.43 | 3.39 219 | 1.84 | 1.60 | 1.30 | 1.10 =407366 [366 [366 | 366] 41.2 50.2 | $3.7 | 56.3 | 60.2 | 63.0 4.06 | 4.06 | 4.06 _| 4.06 } 3.20 2.00 | 1.66 | 1.43 | 1.14 | 0.95 =60/408 [408 [408 | 40.8] 445 55.1 | 589 [618 | 65.8 | 68.6 3.50 | 3.50__| 3.50 _| 3.50 | 2.92 1.72 | 140 | 118 10.92 | 0.75 80/443 [443 [443° [443/471 58.7 | 628 | 65.7 [69.7 [724 3.10 [3.10 | 3.10 | 3.10 | 2.71 1s2_ [121 | Lor | 0.78 | 0.62 Once & is obtained for the assumed angle @, the shear factor # and a new value for the angle @can be read from Tables 6.4 and 6.5 (Tables [Link].2-1 and [Link].2- 2 of the AASHTO LRFD). However, in order to verify the validity of the selectedChapter 6 — DESIGN FOR SHEAR AND TORSION 347 angle @ and corresponding factor £, the new angle @ must be compared with the previously assumed value. If they differ, the procedure should then be repeated until the new and previously assumed angles @are about equal. Alternatively, the design procedure can be started by assuming a value of &. Then, values of @ and f are read from Tables 6.4 or 6.5 and a new value of & is determined using Eq. (6.60). If the calculated value of &, is equal to or smaller than the previously assumed value of &, then the values of @ and f could be used for design. Otherwise, a new value of & must be assumed. This process is illustrated in the flow chart Fig. 6.23 which is adapted from the AASHTO LRED code. Typically, to start the iteration a value of €, = 0.0005 can be selected. Note that Table 6.4 applies to concrete sections with at least minimum transverse steel reinforcement, while Table 6.5 applies to sections without transverse steel reinforcement or with transverse reinforcement less than the minimum required. In order to determine @ and in members with no transverse steel reinforcement or with web reinforcement less than the minimum required, the crack spacing parameter in Table 6.5, yo, is needed; it can be estimated from: (6.61) = maximum aggregate size (in) sy = the lesser of either of the following two values, namely: © dy for members without stirrups and with conventional longitudinal reinforcement, or * for members without stirrups but with well distributed longitudinal reinforcement, the maximum distance between layers of longitudinal reinforcement, provided the area of reinforcement in each layer is not less than 0.003b,s. (see AASHTO Fig. C5.[Link]-2). With Band @known, the concrete contribution to shear strength, V, is calculated from Eq. (6.57), and the required steel contribution to shear strength can then be determined from: vy -Vy- a -V,,-0.0316 B J f'e by dy (6.62) Finally, for a given area of transverse reinforcement, 4,, assumed normal to the longitudinal reinforcement, the required stirrup spacing can be determined from Eq. (6.58) as follows:348 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Given materials and sectional properties, and: My ViusTu Ss Assume value of e, and p| take @ and f from corresponding cell of Determine b, and d, - Table 6.4 | Calculate V, Calculate &, from — Eq, (6.60) Calculate shear stress ratio: v/ fi Ifthe section is within the | transfer length of any is calculated e, less than assumed value ? Is assumed €, too strands, calculate the } Cen ects average value of frye | high? . i Ld ———— If the section is the | development length of any| . | reinforcement to ensure: bars, calculate the effective value of Ay | Kus Wn Can longitudinal Determine transverse f nt reinforcement re required tension? ‘Can excess shear capacity be used to reduce the longitudinal steel requirements ? = Yes ‘q End nd ¥ Provide additional longitudinal reinforcement Yes Calculate values of @ and f corresponding to larger ©, (Table 6.4) Figure 6.23 Flowchart for shear design (adapted from AASHTO 1998, Interim 2002)Chapter 6 - DESIGN FOR SHEAR AND TORSION 349 A fy d, cote s< Aedy frost’ (6.63) s A flowchart summarizing the general sectional procedure for shear design is given in Fig. 6.23; it is adapted from the 2003 AASHTO LRFD code. Two examples illustrating the procedure as applied to bridge beams are given in Sections 14.10 and 14.11. 6.12.2 Special Considerations 1. First Critical Section for Shear If the reaction force in the direction of applied shear introduces compression into the end region of a member (that is generally the case), the distance between the critical shear section and the internal face of the support shall be taken as the larger of 0.5d, cot or dy. 2. Minimum Transverse Reinforcement Except for slabs, footings and culverts, transverse reinforcement must be provided if V.>0.5 6 Ve + Vp) (6.64) and the reinforcement should satisfy the following: bys fy Note that the minimum reinforcement specified in the right term of Eq. (6.65) leads to a value of =1. This should be compared to the factor 0.75 given by ACI in Eq. (6.40) where fis in psi. A, 20.0316 ff", (6.65) 3. Maximum Spacing of Transverse Reinforcement The spacing of transverse reinforcement shall satisfy the following conditions: For V, <0.1fibydy: §<0.8d, <24 in For Vi, 20.1 f{bydy: 5 <0.4d, <12 in This approach_is somewhat similar to the ACI approach where, for (%/-Ve) = 4 Fc in psi, the stirrup spacing is halved (Section 6.8.4).350 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN 4, Minimum Longitudinal Reinforcement at Each Section For sections not subjected to torsion, the longitudinal reinforcement shall be proportioned so that at each section the following equation is satisfied: y, Ag fy + Ans Fs 2 + 05%, - Vp)cotd (6.67) My 4o ¢ where ¢ values are the appropriate resistance factors corresponding to moment, shear, and axial resistance, as suitable. The reinforcement in Eq. (6.67) is assumed to be fully developed at the section considered. If direct compression is introduced into the flexural compression zone of the member by the reaction force or applied load at the location of maximum moment, the area of longitudinal reinforcement on the flexural tension side of the member need not to be larger than that required to resist the maximum moment acting alone. The longitudinal reinforcement on the flexural tension side of the member shall resist at least a tensile force 7 at the inside edge of the bearing area at simple end supports, given by: T= Cx-o5r, -Vy)cotO (6.68) in which V,,,¥,,¥p,and Ocan be calculated at a distance taken as the larger of 0.5d, cot @ or d,, from the face of the support. 5. Modifications for Lightweight Concrete When lightweight aggregate concrete is used, AASHTO Article [Link] recommends to replace the square root of the concrete compressive strength in Eq. (6.57) by: AT fa S Vf'e (6.69) where /, is the splitting tensile strength of lightweight concrete (see Table 2.8). If fy is not known, Wit shall be replaced by 0.754/f.. for all-lightweight concrete or by 0.85, f% for sand-lightweight concrete. The unit stress is in ksi. 6. Remarks Since the modified compression field theory was first introduced in the 1993 AASHTO LRED code, the procedure has seen several adjustments. Among them are the values given in Table 6.4 and 6.5 and their limitations, the value of ¢, from Eqs. (6.60a) or (6.60c), the definition of 4,y and the definition of f,,. The last information available was reported above. However, at time of this writing,Chapter 6 - DESIGN FOR SHEAR AND TORSION 351 conflicting information can be found in the code about fp, . In particular, it can be interpreted as the effective prestress, the stress at decompression at the level of the prestressing steel (fyy =fpe+fpcEp/ Ec), oF the initial prestress just before transfer (0.70 fy, to 0.75fpu). Using the effective prestress is on the safe side, since it leads to an increase ing, , thus a decrease in f (Table 6.4), a decrease in V,. and an increase in shear reinforcement; however, either value can be used until further clarification is released from AASHTO. Physically, the longitudinal strain ¢, can be regarded as an average (smeared) strain in a reinforced concrete tensile tie with centroid assumed located at mid-depth of the member for members with at least minimum transverse reinforcement, and at d, for members without or with Jess than minimum transverse reinforcement (Fig. 6.22 and Eqs. 6.60a and 6.60b). Note that it is not recommended to fictitiously increase the tensile reinforcement (which is present in the denominator of Eqs. (6.60a to 6.60c)), in order to reduce ¢, and thus reduce the required shear reinforcement. Although the ACI and AASHTO methods for shear differ widely, it is observed that the limitations set on several equations in effect lead to similar design for most routine problems. An example is treated next. Additional examples of bridge beam design for shear can be found in Sections 14.10.2 and 14.11. ——— 6.12.3 Example: Shear Design by AASHTO LRFD Code Consider the same example as in Section 6.10. The AASHTO LRED specifications and notation will be used whenever possible. Articles of the code are identified by the letter A, followed by the code number in brackets, Let us check shear at f/2 from the center of support and three other sections defined in Table 6.6. It can be shown that for the first section: distance from center of support 67 ft ~(on safe side since cot 0 is not known yet) e=79+ TT 58.72 in 28 b, =8in d, = effective shear depth or lever arm of couple from Eq. (6.55) [A5.8.2.7] y+ ¥ =8.72+12.9 = 21.62 in d, = 21.62 -0,5x1,98 = 20.63 in > 0.9 d, = 0.9 x21.62 =19.46 in OK. > 0.72h =0.72x40= 28.8 in N.G. Hence, d,, = 28.8 in at this section. Values of d, for other sections are given in Table 6.7. Shear forces at section x= Ve =19.1 kips Vp =3.33 kips Vy, =11.33 kips .Sh = 1.67 flare given by (Table 6.6)352 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Factored shear force assuming 17 = 0.95 [A3.4.1] V,, = 0.95 (1.25%19.1 + 1.5%3,33+1.75%1 1.33) = 46.263 kips Note that the AASHTO load factors are also given in Tables 14.1 and 14.2. Factored moments assuming the superimposed dead load due to asphalt is included [A3.4.1]: My, =0.95x(1.25x32.64-+1.5x5,70+1.75x19,36) = 79.069 kips-ft = 948,822 kips-in Assuming a draped tendon profile (see Fig. 6.16): V, =F xsin a = 229.5 xsin(a) = 229.5 x 0.04107 = 9.426 kips From Eq. (6.59) [A5.[Link]-1} Vu p _ 46.263-0.9x9.426 = 0.18 ksi ob,d, 0.98 28.8 O18 oro fe § Therefore, s<0.8d, <24 in (Section 6.12.2-3) [A5.8.2.7-1] According to AASHTO [[Link].2], the shear resistance can be determined as follows: First estimate: fing = Spe + S; = 0.70 f py =189 ksi [C5.[Link]-1 pe = Spe + fpe ‘pu I" iteration. Assume 0 = 21,8° ; thus: cot =2.5 From Eq. (6.60a) [A5.[Link] interim 2000] and assuming at least minimum transverse reinforcement Check —M,, = 948.822 kip-in Converged, From Table 6.4: for ot 04 and 0=21.8°, cord =2.50 and f = 3.75 Therefore: V, =0.0316x Bf fib,d, (Eq. 6.57) [A5.83.3-3] 0.0316%3.75 x V5 x8x 28.8 = 61.05 kips 46.263 =9,426-61.05= -19.07 (negative) (Eq. 6.62) Check minimum required transverse reinforcement (Eq. 6.64) [A5.8.2.5}: O.56(V,. +V))=0.5x0.9 x(61.05+9.426) = 31.71 kips < V, Minimum reinforcement is needed. Consider #3 U-shaped stirrups with 4, = 0.22 in? 16.26 kips Check minimum reinforcement and spacing (Eq. 6.65) [A3.8.2.5-1]: AySy 0.2260 0.0316) f'. 5, 0.0316x V5 x8 Select s~23 in For ss = 23.35 in Cheek maximum spacing (Section 6,12.2-3) [AS.8.2.7] For: Vy Stirrup in tension Stirrup anchorage Z a End Corner Diagonal Longitudinal Longitudinal distribution distribution compression steel in steel plate plate in concrete tension anchorage Figure 6.27 Idealized functions of concrete and reinforcement after torsional cracking. (Ref. 6.7, courtesy American Concrete Institute.) If the concrete element is reinforced against torsion, its behavior before cracking is essentially identical to a plain concrete element. Cracking would occur at about the same torque as for the unreinforced element. Prior to cracking, the contribution of the torsional reinforcement is small, as evidenced by the low stresses recorded during testing. After cracking, the stiffness of the member drops suddenly to a358 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN fraction of its precracking value and the stress in the reinforcement increases sharply to maintain equilibrium. With increasing torque, multiple inclined cracks develop at regular spacing and take the form of S shapes (Fig. 6.26). The principal compressive strain measured along elements of concrete bound by two cracks seems to increase at a rate faster than predicted by the theory. Similarly to the behavior in bending, redistribution of torsional stresses can occur in statically indeterminate structures. In general, three types of failure by torsion can be observed [Ref. 6.17] depending on the amount of reinforcement, and they are comparable to the flexural types of failure. Beams underreinforced for torsion will fail by yielding of the steel, while overreinforced beams will fail by crushing of the concrete, mostly on the wider face of the member. A third mixed mode type of failure occurs in partially overreinforced members in which a combination of stirrup yielding or longitudinal steel yielding and concrete crushing occurs. The functions after cracking of the different elements of a cracked concrete member reinforced for torsion can be visualized by referring to Fig. 6.27, taken from Ref. [6.7]. The figure is also useful in summarizing the information on the types of torsional reinforcement and the details that must be addressed in designing them. 6.15 BACKGROUND TO STRESS ANALYSIS AND DESIGN FOR TORSION 6.15.1 Torsional Stresses This section provides background information. Its content was used in prior versions of the ACI code between 1971 to 1989. The reader is referred to the first edition of this book and the cited references for details of its application. While the following treatment is useful for understanding the shear and torsional behavior of structural concrete members, it is not necessary for applying the provisions of the 2002 ACI code covered in Section 6.16. Using either the mathematical theory of elasticity, the plastic theory, or the skew bending theory, it can be shown that the magnitude of the maximum torsional shear stress in a rectangular section subjected to a torque T is given by: T my t= (6.70) shorter overall dimension of rectangular section y= longer overall dimension of rectangular section n = torsional coefficient The value of 7 varies with y/x from 0.208 to 1/3 in the elastic theory, and from 1/3 to 1/2 in the plastic theory. For simplicity, an overall value of 1/3 can be adoptedChapter 6 - DESIGN FOR SHEAR AND TORSION 359 for nonprestressed members. For prestressed members, a value of 77 is suggested in Eq. (6.73). For a plain concrete section at the onset of cracking, Eq. (6.70) can be written as: bop =—— (6.71) where: ty = torsional shear stress at cracking T.» = cracking torque For flanged sections such as 7; I, or L sections, the denominator of Eqs. (6.70) and (6.71) is replaced by Ex7y, in which © applies to the various rectangular parts of the section: ler _ (6.72) fer shorter overall dimension of rectangular part of cross section longer overall dimension of rectangular part of cross section Znx*y = the sum to be chosen as the largest value of several alternatives, if any (safer design) The above equation is not exactly in accordance with the theory, but simplifies the results enormously. For rectangular box sections, the code prescribed that an equivalent solid section may be used provided the wall thickness ’ is at least x/4. If the wall thickness h’ is less than x/4 but greater than x/10, an equivalent solid section may be taken, provided the term x?y is multiplied 4h’/, When h’ is less than x/10, the stiffness of the wall must be considered, since possible buckling and crushing may occur. The denominator of Eqs. (6.70) to (6.72) is generally called the torsional constant, @. A summary of the torsional constants recommended for various sections in earlier versions of the ACI code (from 1971 to 1989), is given in Fig. 6.28. It is generally accepted, that Eqs. (6.70 to 6.72) would remain the same for prestressed concrete, except for the value of 7. Based on 218 test results, Zia and McGee [Ref. 6.49] have shown that the torsion coefficient 7 for prestressed members can be estimated as a first approximation from:360 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Torsional constant @: O=ny 1 =1/3 (ACI, non-prestressed) 1 =0.35/(0.75+x/y) (Zia, prestressed) S o-(20°y) Po Example: © =| + xy 4 for section at left ' ‘ ACI 1971-1989: For: h'>x/4, 0=Dyx"y x 4h a For: —sh' fl = 0.6 for nonprestressed members, in which case 7, =0 > " 0.133/7 for prestressed members [Ref. 6.49] in which 77 is given by Eq. (6.73). Hence k = (0.75—0.4x/7)/1.05. The value of x/y is determined for the largest component rectangle where the web reinforcement is usually placed. Therefore, for nonprestressed members, i, =0.48,Jf.. 7 =1/3,and the value of f; is as discussed in Eq. (6.74). These values are useful to explain some equations found in various codes. 6.15.4 Combined Loading Seldom do pure torsional loadings apply to structures. It is more likely that a combination of shear torsion and bending exists. Their interaction is generally represented by an interaction surface and, when taken two at a time, by an interaction curve. Circular, elliptic, parabolic, or linear interaction curves have been used to model the interaction of bending and torsion or torsion and shear [Refs. 6.13, 6.14, 6.22, 6.23, 6.25, 6.36, 6.49]. In the ACI code approach, design for bending is achieved separately (thus it does not need to be reconsidered), while that of torsion and shear is combined. A \2 2 te & =1 (axis with citer 10 scales) + 98 a iqeeenen 1 1 gz fu aan i go te | z o4 1 ns Hl fc L shear 0.2 ' reinforcement 1 yu 4 - I ! > 02 04 06 08 1.0 VG W!P Vy Shear stress @ O) Figure 6.30 (a) Typical interaction curve for combined torsion and shear. (6) ‘Typical representation of concrete and reinforcement contribution after cracking.Chapter 6 - DESIGN FOR SHEAR AND TORSION 365 Tests have shown [Refs. 6.14 and 6.36] that the interaction between torsion and shear in prestressed and nonprestressed members can be adequately represented by a circular curve (Fig. 6.304) of the form: (é] (2) =1 (6.82) le Ue in which: * ‘ . . 1, = torsional strength in presence of flexural shear, t. < tf {, = torsional strength when member is subjected to torsion alone v2 = shear strength in presence of torsion, 02 < Ug U, = shear strength when member is subjected to flexural shear alone (v, for prestressed members is lesser of v,; and V.,, given by Eqs. (6.29) and (6.31) Equation (6.82) also applies to plain concrete where the subscript cr (for cracking) replaces c. Note that, because of the circular shape of the curve, little reduction in strength is observed for one effect, unless the other effect is significant. For instance, about 90 percent of the shear resistance acting alone can be counted on if the torsion is less than 45 percent of the torsion resistance acting alone, and vice versa. The design approach followed in the 1971 to 1989 versions of the ACI code for combined shear and torsional loading was the same as for shear loading, that is, the torsional strength of concrete in combined loading is considered first and then the web reinforcement, if needed, is proportioned to carry the excess torsion necessary to balance the ultimate required or factored torque (Fig. 6.305). Thus, the following equations applies: Ty , is zero, and thus the nominal torsional strength, Ty, comes from torsional reinforcement only. This is a simplifying assumption, but the code equations are calibrated to make up for the shortfall. The shear resisted by concrete after cracking, V;, is assumed unchanged by the presence of torsion. ¢ For solid sections, the code assumes a circular interaction diagram between shear and torsional stresses, while for thin-walled hollow sections, it assumes a linear diagram. The provisions of the code reflect two conditions under which torsion is considered in reinforced concrete members: 1. Equilibrium torsion: this condition prevails when the torsional moment cannot be reduced by redistribution of internal forces. For this condition torsional370 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN reinforcement, when needed, must be provided to resist the full factored torsional moment. 2. Compatibility torsion: this condition prevails when redistribution of internal forces after cracking occurs to maintain compatibility of deformations leading to a reduction in the resulting torsional shearing stresses. Redistribution can happen in statically indeterminate structures. When redistribution can happen, the condition is described as compatibility torsion, and the code provides an upper design limit to the factored torsional moment taken equal to the cracking torque, Top Thus if T,>@T», the code allows to design for only @7,, provided redistribution of internal forces is accounted for in the design of other members of the structure, 6.16.3 Condition for Consideration of Torsion in Design Torsion effects can be neglected when the factored torsional moment 7, at the section satisfies the following conditions: (a) For non-prestressed members: a Ty S(T )min = OV Fe (6.85) Pep (b) For prestressed members . ze | Spe Ty $ Tu)min = OL Fe eae (6.86) cp AK (c) For non-prestressed members subjected to axial tension or compression force: 2 Ag N, Ty Ty min = ON Fo I+ (6.87) eT min Wier adh where: T, = factored torsional moment at section, Ib-in ¢ = strength reduction factor for torsion, equal 0.75 by 2002 ACI code Agp = area enclosed by outside perimeter of concrete cross section, in’ (Fig. 631) Pcp = Outside perimeter of the concrete cross section, in (Fig. 6.31)Chapter 6 - DESIGN FOR SHEAR AND TORSION 371 > " , = gross area of concrete section, in’. For a hollow section, Ag is the area of concrete only and does not include the area of the void(s). Soe = compressive stress in concrete (after all losses) at centroid of cross section resisting externally applied loads or at junction of web and flange when the centroid lies within the flange, psi. In a composite member, fy is the resultant compressive stress at centroid of composite section or at junction of web and flange when the centroid lies within the flange, due to both prestressing and moments resisted by precast member acting alone. factored axial load (Ib) normal to cross section, occurring simultaneously with V,, and/or T,; to be taken positive if compression. If the section is hollow and is not statically indeterminate, Ay shall be used instead of Ac, in Eq. (6.87). Note that in the above equations (T,)min is taken equal to 25% of the assumed cracking torque T, (Eq. 6.74). 6.16.4 Critical Section for Torsion For non-prestressed members, the critical section for torsion can be taken at a distance d, from the face of the support, unless a concentrated torque occurs within that distance; in such a case, the critical section is taken at the face of the support. For prestressed members, the critical section for torsion can be taken at a distance h/2 from the face of the support, unless a concentrated torque occurs within that distance; in such a case, the critical section is taken at the face of the support. The code does not address the case for partially prestressed members. The author recommends to follow the provisions for prestressed members with a value of d, as suggested in Section 6.18. 6.16.5 Maximum Allowable Torsional Moment Strength Even if torsional reinforcement is provided, the cross sectional dimensions and properties of the section shall be such that: (a) For solid sections: 2 2 \ Vu TPh ( 7 + <¢ +8ff | (6.88) (zi) (= byde ~*°)372 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN The above equation reflects a elliptical interaction diagram between shear and torsion. (b) For hollow sections: Vu }. TuPh_\< (2 + ) ae | | SuPh /< gf “c_ ag lf (6.89) (z, 1.74, ‘ Dyydle : If the wall thickness varies around the perimeter of a hollow section, Eq. (6.89) shall be evaluated at the location where the left hand side of the equation is a maximum. If the wall thickness is less than 4,,/ py, the second term on the left side of Eq. (6.89) shall be taken as: ( Ty } (6.90) LT Appt \ where f is the thickness of the wall of the hollow section at the location where the stresses are being checked. Equation (6.89) reflects a linear interaction diagram between shear and torsion. The notation is as follows: V, = factored shear force at section, Ib V, = nominal shear strength (force) provided by concrete, Ib (Section 6.8.2) Agn = area enclosed by the centerline of the outermost closed torsional reinforcement Ph = perimeter of centerline of outermost closed transverse torsional reinforcement, in d, = as defined in this text; it is equal d, for reinforced concrete and d, for prestressed concrete. In Eqs. (6.88) and (6.89), the maximum design limit of 7,,, that is, #7,,., can be used when compatibility torsion conditions prevail. Also one should use Vi, Kbyde)=Vy and V,/(byde)=veas calculated for shear (Fig. 6.14), that is, for prestressed concrete the value of d,=d, is adjusted according to the code provisions. Examining the right hand side of Eqs. (6.88) and (6.89) one can observe that a section shall not be allowed if the combined stress due to shear and torsion exceedsChapter 6 - DESIGN FOR SHEAR AND TORSION 373 #(v.+8)/72), where ve is the contribution of conerete to shear resistance as defined in Eqs. (6.29, 6.31 or 6.32) or Figure 6.14. 6.16.6 Transverse Reinforcement Design The reinforcement required for torsion shall be determined to satisfy the general condition: Ty Spy (691) Ty =To+Tis =0+T; (6.92) n= Te +Ts is and T,, is the factored torsional moment, T» is the design torsional resistance, and Tig is the torsional resistance provided by the torsional reinforcement. As mentioned above, it is assumed that concrete does not contribute any resistance to torsion (thus, T. = 0) and that the torsional resistance is provided by the reinforcement alone, which includes stirrups and longitudinal steel. 7,, can be estimated from: (6.93) = gross area of concrete enclosed by the shear flow path [Ref. 6.15]; it can be taken as a first approximation equal to 0.8544 (defined in Fig.6.31). The use of 4, reflects the fact that the cover of concrete may spall off under large deformations. Ash = atea enclosed by centerline of the outermost closed transverse torsional reinforcement, in’ (see Fig.6.31) A, = area of one leg of closed stirrup resisting torsion within a distance s;, pr) in’ Jyy = yield strength of closed transverse torsional reinforcement, psi, not to exceed 60,000 psi (414 MPa) 8, = spacing of torsion reinforcement measured in a direction parallel to the longitudinal reinforcement, in 9 = angle of compression diagonals in the truss analogy for torsion Equation (6.93) is based on the space truss analogy with compression diagonals at an angle @ assuming the concrete carries no tension and the reinforcement yields; 6 shall not be taken smaller than 30 degrees nor larger than 60 degrees. It shall be permitted to take @ equal to 45 degrees for nonprestressed members and 37.5374 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN degrees for prestressed members with an effective prestress not less than 0.4f,,. If the effective prestress is less than 0.40fp,, 8 can be taken equal to 45 degrees. For a required nominal resistance, the transverse torsional reinforcement can be computed from Eq. (6.93): = __4» __ (6.94) 5) G2A,fyy COO For a given factored torsional force, 7,, the transverse reinforcement, 4, can be calculated assuming a practical value of stirrup spacing, s,, which can be estimated a priori from the limitations on spacing given in Sec. 6.16.10. 6.16.7 Longitudinal Torsion Reinforcement In addition to the transverse reinforcement needed for torsion (Eq. 6.94), longitudinal reinforcement along the axis of the member is also required, and is a function of the transverse reinforcement. The additional longitudinal reinforcement required for torsion shall not be less than: (6.95) where the ratio 4, /s, shall be taken as the amount computed from Eq. (6.94), and fy is the yield strength of longitudinal torsional reinforcement, not to exceed 60,000 psi (414 MPa). In prestressed members, the longitudinal reinforcement required for torsion may consist of unstressed tendons with a tensile yield force Apy {py = Arty) 6.16.8 Combining Shear and Torsion Reinforcement Reinforcement required for torsion shall be added to that required for shear, moment, and axial force that act in combination with torsion. The most restrictive requirements for reinforcement spacing and placement shall be met. Since the cross sectional area of stirrup for shear is defined in terms of all the legs of a given stirrup while the stirrup area for torsion is defined in terms of one leg only, the addition of stirrups to satisfy both shear and torsion is carried out as follows: Total stirrup requirement = +24 (6.96)Chapter 6 - DESIGN FOR SHEAR AND TORSION 375 where s, and s, are the stirrup spacing for shear and torsion, respectively. In prestressed beams, the total longitudinal reinforcement including prestressing steel at each section shall resist the factored bending moment at that section plus an additional concentric longitudinal tensile force equal 4)f,; based on the factored torsion at that section (Eq. 6.95); the spacing of the longitudinal reinforcement shall satisfy the code requirements explained below. 6.16.9 Minimum Torsion Reinforcement When the factored torsional moment exceeds the minimum value given by Eqs. (6.85) to (6.87), a minimum area of transverse and longitudinal torsional reinforcement is required by the ACI code. * Minimum transverse closed stirrups shall be provided according to: 0.75 no 4 - fw +229 nin = larger of os (6.97) tv * Minimum total area of longitudinal reinforcement shall be provided according to: Acy (4) fy eee a sce — 6.98 Toi a Ph a (6.98) in which: 4, /s, shall not be taken less than 25b,,/ fy. we A Armin ( 6.16.10 Spacing and Detailing The spacing of transverse torsion reinforcement shall not exceed the smaller of Pp/8or 12 in. All stirrups must be closed stirrups. Typically a 135 deg standard hook around a longitudinal bar is required; additional details for minimum leg length and the case of welded wire fabric can be found in the code. Note that 90 deg hooks and lapped U-shaped stirrups are inadequate to resist torsion since they become ineffective when the cover of concrete spalls off. The longitudinal reinforcement required for torsion shall be distributed around the inside perimeter of the closed stirrups, with a maximum spacing of 12 in (30 cm). There shall be at least one longitudinal bar or tendon at each corner of the stirrups.376 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN To minimize buckling, bars shall have a diameter at least 0.042 (that is, 1/24) times the stirrup spacing, but not less than that of a No. 3 bar (9.5 mm). 6.16.11 Type of Torsion Reinforcement As mentioned above, torsion reinforcement consists of both longitudinal and transverse reinforcement. e Longitudinal reinforcement shall consist of longitudinal bars or tendons. e Transverse reinforcement shall consist of: —» Closed stirrups or closed ties perpendicular to the axis of the member — Closed cage of welded wire fabric with transverse wire perpendicular to the axis of the member — In nonprestressed beams, spiral reinforcement. The transverse reinforcement shall be anchored with standard hooks detailed according to code requirements. Longitudinal torsion reinforcement shall be developed at both ends according to code recommendation for development length. Figure 6.32 Typical layout of torsion reinforcement in different sections. 6.16.12 Design Steps for Combined Torsion and Shear A flow chart summarizing the main design steps for torsion reinforcement is given in Fig. 6.33. While using the chart, care should be also taken to check details and exceptions given in the code prior to finalizing the design. Note that the flow chart accommodates the design for both torsion and shear. To carry out the computations, the contribution of concrete to shear resistance, V,. is needed and can be obtained from Section 6.8.2 or the flow chart Fig. 6.14.Chapter 6 - DESIGN FOR SHEAR AND TORSION 377 Given material and seetional properties, Ty, Vas¥us Nuv Ves Yer and Ay, [Link] for vertical shear, select critical section at d, for reinforced member or hi/2 for prestressed member. No need for torsional reinforcement Ap fpe_9 Ty < (Ty min = Oy Se Ae aoe Wy, positive if compression, No ‘Statieally indeterminate structure allowing redistribution of torsional moment after crackin Use: Ty =Ty oF 4(Tidmin Whichever is smaller Section is inadequate. Increase dimensions of the section and go back to start, wall thickness limitations Yes fv 2 [4¢o, L2 ooo? 1h hy and Apé larger of ( SYR: Ap\_(41), Lo where Oe (2) ees RC: 0 = 45°; cord ors fee PC: 8 = 37.5%; cot = 1.303 fw oy Reduce s; or increase 4, to satisfy previous limitation, Provide details of torsional reinforcement according to code. 18. Note: 3, = salt v8 $5)!24 in S12 in; diameter of 4) | Repeat for another section Figure 6.33 Proposed design flow chart for torsion (pure or combined) for prestressed and nonprestressed members.378 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN Thus the dp value used to compute V, applies. This is also true for V,. For uniform loading the approximation given in Eq. (6.32) for V; can be used. Compared to the ACI code, some notations are expanded such as for sy and s; to differentiate stirrup spacing, s, used for shear and torsion. Same spacing value s could be used for both. Also some equations are written in a slightly different form than given in the code to simplify the computations. 6.17 EXAMPLE: TORSION DESIGN OF A PRESTRESSED BEAM Because a new example involving a combination of bending, shear, and torsion can be very lengthy, let us consider, for simplicity, the same beam (Fig. 6.34a) treated in the Examples of Sections4.9, and 6.10. Let us also assume that the beam is subjected to the same magnitude of dead and live loads. However, the design calls for a special loading condition in which a line load of magnitude equal to the live load of 0.40 kif can be applied at an eccentricity e = 1.5 ft from the axis of the beam, Restraint against rotation is provided at the ends of the beam by a transverse diaphragm. Thus, a torsional moment results at each section of the beam in addition to the bending moment and shear. According to the ACI code procedure, the design for bending remains unchanged, while special design considerations must be taken for combined shear and torsion. ‘The first critical section, at a distance x = 2 fi from the face of the support, will be covered in detail. A similar procedure can be followed for the other sections along the span, The ultimate design torsional moment at the first critical section is given by: 7,=(1 6 )(S-s)en 6x0.4x(35—2)x1.5 = 0.64%33x1.5 = 31.68 kips-ft = 380160 Ib-in sign ultimate shear force was determined in Example 6.10 (Tables 6.1 and 6.2). Its value is V, = 45.395 kips. The corresponding value of », is given by: yy = Le = 45395 _ 177.32 psi, where dy 20.8h=32 in Dud, 8X32 ? The design flow chart for torsion (Fig. 6.33) will be followed for the main steps. The beam cross section is first idealized to an equivalent T section with a constant depth flange of 5.75 in, Then it is separated into two rectangular parts, the web of dimensions & by 40 in, and the overhanging flange of dimensions 5.75 by 40 in (Fig. 6.340). Sectional parameters are computed for the idealized section. In determining the dimensions x and y of the closed stirrups, the cover to the centroid of first layer of reinforcement (0.5-in diameter strands) was taken equal 2 in, from all sides, Assuming No. 3 stirrups, leads to a cover to the centroid of stirrups equal: 2-0.5x0.5-0.5x3/8= 1.5625 in The dimensions of closed loop stirrups are obtained as follows (Fig. 6.34c): ay =8- 21,5625 = 4.875 in J =40-2x1,5625 = 36.875 in 75~2x 1.5625 = 2.625 in 0—2x1.5625 = 36.875 in v2Chapter 6 - DESIGN FOR SHEAR AND TORSION 379 in ein t at I 5.15 ciel 5.75, ++} 20 40 sas a 8 18 kK Led ie 5 (a) () 8 \~—* +! Bigs | x, =4875 36.875 ¥ ila a © Figure 6.34 Let us compute first the torsional parameters of the section: Agy = Ag = 550 in? Pop = 48+ 2%5.75 +2x204234.5+8=176.5 in Phy = 203) +41) +202 +49) = 2(36.875 + 4.875) + 2(36.875 + 2.625) = 162.5. in Ap = ix, +¥9Xp = 36.875 x 4.875 + 36.875 x 2.625 = 276.563 in A, = 0.854), = 235.08 in? ‘The average prestress at the centroid of the section is given by: F _ 229,500 Tye A 550 17.27 psi Following the steps in the flow chart of Fig. 6.33 (also Eq. 6.86):380 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN a fy 3507 {417.27 T. - PP 4 PZ = 0,75,/5000 f+ Tudmin = O4 Se Per Vf 176.5) 4/5000 = 143,000 Ib-in = 143 kip-in = 11.92 kip-ft Check: Ty, = 31.68 kip-ft > (L,)yyin = 11.92 kip-ft_ > Torsional reinforcement is needed Check if section dimensions are adequate. From shear design: d, =d, > 0.8h =32 in R409 5 pai indy vy =o = 177.32 psi dy For a solid section (Eq. 6.88): 2 2 ( Fe } (4) oo as 7} =02ss9. 815000) = 73139 psi de) (1.743, byde 2 (array [ene =r ‘ 1.7276.563? } Thus the section dimensions and properties are adequate. 7.32) +(475.1)° =507.11 < 731.39 psi O.K. From Eqs. (6.94) and (6.98): Zi 31.68%12000 one 4 2A, fyy COLO 0.752 235.08 60000%1.30. t= larger of A a = 0.0138 i =v *S- = 0.0033 fr 60,000 A py 2c 6)? =0.0138%162.5(1.303)" = 3.807 in? > controls St yh Ay = larger of ' Yh e| (4), Siw _{ $5000 x550 Ty a 60000 Assuming a No. 3 stirrup size, of section 0.11 in’, the following stirrup s AON poss + 5, =O . eS 0.0138 Check stirrup spacing requirements: Ph _ 162.5 ; 442 =2031 5, = 8=smallerthan 1B” 8 ™ Ly OK. 12 in 0.0138 x 162.5 = 0.998 in? acing is needed: ‘The shear reinforcement provided at this section is No. 3 U-stirrup at 24 in (see Section 6.10). Thus:Chapter 6 - DESIGN FOR SHEAR AND TORSION 381 A022, = 0.0092 Shear and torsion reinforcement must satisfy the following minimum reinforcement condition: 0.75 fe * = 0.0071 2 wv 822 4 210.0138 = 0.0368 = larger than — > OK. = = 0.0067 Following is a summary of the reinforcement needed for torsion and shear. Since for shear one stirrup is needed at 24 in spacing, and for torsion one stirrup is needed at 8 in spacing, it is advisable to select one closed stirrup at 6 in spacing to cover both requirements in the web. Indeed we can verity that the transverse reinforcement provided is sufficient: fs 2 = 0.367 (provided) ~ 0.368 (required) 5 In the flange, No. 3 closed irrups are needed only for torsion. ¢ longitudinal reinforcement required for torsion can be satisfied using 19 No. 4 bars having cach a cross section of 0.20 in’, placed around the periphery of the section. Because of symmetry Fig 6.34a shows 20 such bars. Their average spacing along the sides can be about 8 in and should satisfy the maximum spacing of 12 in allowed by the code. Longitudinal bars for torsion should also have a diameter larger than 0.042 times the stirrup spacing. Here the No. 4 bar has a diameter of 0.5 in > 0,042x6 = 0.25 in and is satisfactory. These longitudinal bars are in addition to the longitudinal reinforcement required for bending. 6.18 SHEAR AND TORSION IN PARTIALLY PRESTRESSED MEMBERS A distinction was made throughout this chapter between a prestressed concrete member and a nonprestressed or reinforced concrete member. A review of the design approaches used for shear and torsion strongly suggests that partially prestressed members, which contain both prestressed and nonprestressed flexural reinforcement, can be designed essentially as prestressed members. The only difference between them is a lower value of the average prestress. In designing for shear and in order to be consistent with the procedures developed earlier for bending, the following value of d, can be used: PPRd, +(1~PPR)d, d, = the larger of } or (6.99) PPR(0.8h) + (1~ PPR)d, whet listance from extreme compressive fiber to centroid of tensile force in the reinforcement382 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN = distance from extreme compressive fiber to centroid of prestressing force d, = distance from extreme compressive fiber to centroid of tensile reinforcing steel PPR = Partial Prestressing Ratio defined in Refs. [1.30 and 1.48] As a first approximation, the value of PPR for Eq. (6.99) can be estimated from: ppp ~—“0s/ov = ——psiey (6.100) Apstpy + Ash, REFERENCES 6.1 AASHTO LRFD Bridge Design Specifications, 2nd Edition, American Association of State and Highway and Transportation Officials, Washington, DC, 1998. Also Interim 2002. 62 ACI Committee 318, Building Code and Commentary. ACI 318-02/318R-02, American Concrete Institute, Farmington Hills, MI, 2002. 6.3 ACI Special Publication SP-18, Torsion of Structural Concrete, American Concrete Institute, Detroit, 1968. 64 — ACI Special Publication SP-35, Analysis of Structural Systems for Torsion. American Concrete Institute, Detroit, 1973. 65 ASCE-ACI Joint Committee 426 Report, “The Shear Strength of Reinforced Conerete Members,” Journal of the Structural Division, ASCE, 99(ST6): June, 1973, Chaps. 1-4, pp. 1091-1197. (Also reproduced in ACI Manual of Concrete Practice.) 66 Bresler, B., and J. G. MacGregor, “Review of Concrete Beams Failing in Shear,” Journal of the Structural Division, ASCE, 93(55.2), 1967. 67 Collins, M. P., and D. Mitchell, “Shear and Torsion Design of Prestressed and Nonprestressed Concrete Beams,” PCI Journal, 25(5): 32-100, 1980. 68 Collins, M. P., and D. Mitchell, “Design Proposals for Shear and Torsion,” PCI Journal, 25(5), 1980, 70 pp. 69 Collins, M. P., and D. Mitchell, Prestressed Concrete Structures. Englewood Cliffs, NI: Prentice Hall, 1991, 766 pp. Also 2" Edition, Response Publications. Canada, 1997. 6.10 Cuevas, O. G., F. Robles, and R. Diaz de Cosio, “Strength and Deformation of Reinforced Concrete Elements,” In Reinforced Concrete Engineering, Chap. 5, Vol. |, Boris Bresler, ed. New York: John Wiley & Sons, 1974, pp. 194-301. 6.11 Fanella, D, A., and B. G. Rabbat, Design of Concrete Beams for Torsion. Engineering Bulletin EB106.02D, Portland Cement Association, Skokie, IL, 1997. 6.12 Fanella, D. A., B. G. and Rabat, Notes on ACT 318-02, Building Code Requirements for Structural Concrete with Design Applications, Portland Cement Association, Skokie, IL, 2002. 6.13 Gangarao, H. V. S., and P, Zia, “Rectangular Prestressed Beams in Torsion and Bending,” Journal of the Structural Division, ASCE, 99(St. 1): 183-98, 1973. 6.14 Henry, R. L., and P. Zia, “Prestressed Beams in Torsion, Bending, and Shear,” Journal of the Structural Division, ASCE, (100(ST.5): 933-52, 1974, 6.15 Hsu, T. T. C., Unified Theory of Reinforced Concrete, Boca Raton: CRC Press, 1993, 313 pp.6.16 6.17 6.18 6.19 6.20 6.21 6.22 6.23 6.24 6.25 6.26 627 6.28 6.29 6.30 631 6.32 6.33 6.34 6.35 6.36 637 6.38 Chapter 6 - DESIGN FOR SHEAR AND TORSION 383 Hsu, T. T. C., “Torsion of Structural Concrete — A Summary on Pure Torsion,” in Ref. 6.3, pp. 165-78 and Hsu, T. T, C., “Torsion of Structural Conerete ~ Plain Conerete Rectangular Sections,” in Ref. 6.3, pp. 203-38, Hsu, T. T. C., “Torsion of Structural Concrete - Behavior of Reinforced Concrete Rectangular Members,” in Ref. 6.3, pp. 261-306. Hsu, T. T. C., “Torsion of Structural Concrete ~ Uniformly Prestressed Rectangular Sections without Web Reinforcement,” PCI Journal, 13(2): 34-44, 1968. Hsu, T. T. C., “Shear Flow Zone in Torsion of Reinforced Concrete,” Journal of Structural Engineering, ASCE, \16(11): 3206-26, 1990. Kani, G. N. J., “Basic Facts Concerning Shear Failure,” ACI Journal, 63(6): 675-92, 1966. Kani, G.N. J., “A Rational Theory for the Function of Web Reinforcement,” ACI Journal, 66(3): 185-97, 1969. . Lampert, P., Torsion und Biegung von Stahlbetonbalken (Torsion and Bending of Reinforced Concrete Beams), Bericht 27, Institute fir Baustatik, Zurich, January 1970, Lampert, P., “Torsion and Bending in Reinforced Concrete and Prestressed Concrete Members,” Proceedings of the Institution of Civil Engineers, Vol. 50, December 1971, pp. 487-505, Lampert, P., and B. Thiirlimann, “Ultimate Strength and Design of Reinforced Concrete Beams in Torsion and Bending,” Publications, International Association for Bridge and Structural Engineering, Zurich, Vol. 31-1, 1971, pp. 107-131 Lampert, P., and M. P. Collins, “Torsion, Bending, and Confusion: An Attempt to Establish the Facts,” ACI Journal, Proceedings, 69(8): 500-04, 1972. Lessig, N. N., Determination of the Load Carrying Capacity of Reinforced Concrete Elements with Rectangular Cross-Section Subjected to Flexure with Torsion, Work 5, Institute Betona 1 Zhelezobetona, Moscow, 1959, pp. 4-28. Also available as Foreign Literature Study 371, PCA Research and Development Labs, Skokie, IL. Lorentsen, M., “Theory of the Combined Action of Bending Moment and Shear in Reinforced and Prestressed Conerete Beams,” ACI Journal, 62(4), 1965 MacGregor, J. G., M. A. Sozen, and C. P. “Strength and Behavior of Prestressed Concrete Beams with Web Reinforcement,” y of Illinois Civil Engineering Studies, Structural Research Series 210, Urbana, August, 1960, MacGregor, J. G., M. A. Sozen, and C. P. Siess, “Strength of Concrete Beams with Web Reinforcement,” ACI Journal, 62(12): 1503-19, 1965. MacGregor, J. G., and J. M. Hanson, “Proposed Changes in Shear Provisions for Reinforced and Prestressed Concrete Beams,” ACI Journal, (66(4): 276-88, 1969. MacGregor, J. G., and M. G. Ghoneim, “Design for Torsion,” ACI Structural, 92(2): 211-18, 1995, MacGregor, J. G., Reinforced Conerete Mechanics and Design, 3" Edition. Englewood Cliffs, NI: Prentice Hall, 1997, 939 pp. Marti, P., “Truss Models in Detailing,” Concrete International, ACI, December 1985, pp. 66-73. Mattock, A. H., and A. N. Wyss, “Full Scale Torsion, Shear and Bending Tests of Prestressed I-Girders,” PCI Journal, 23(2): 22-40, 1978. Mattock, M. A., and N. M. Hawkins, “Research on Shear Transfer in Reinforced Concrete,” PCI Journal, V7(2): 55-75, 1972. McGree, D., and P. Zia, “Prestressed Concrete under Torsion, Shear, and Bending,” ACI Journal, 73(1): 26-32, 1976. Mitchell, D., and M. P. Collins, “Diagonal Compression Field Theory ~ A Rational Model for Structural Concrete in Pure Torsion,” ACI Journal, 71(8): 396-408, 1974. Mitchell, D., and M. P. Collins “Detailing for Torsion,” ACI Journal, 73(9): 506-11, 1976.384 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN 6.39 Mitchell, D., and M. P. Collins, “Influence of Prestressing on Torsional Response of Conerete Beams,” PCI Journal, 23(3): 54-73, 1978 640 Park, R., and T. Paulay, Reinforced Concrete Structures. New York: Wiley-Interscience, 1975, 769 pp. 641 Rangan, B. V., and A. S, Hall, “Strength of Prestressed Concrete I Beams in Combined Torsion and Bending,” ACI Journal, 73(11): 612-18, 1978, 6.42 Reineck, K. H., “Editor, Examples for the Design of Structural Conerete with Strut-and-Tie Models,” Special Publication-208, American Concrete Institute, Farmington Hills, MI, 2003 6.43 Schlaich, J., K. Schafer, and M. Jennewein, “Toward a Consistent Design of Structural Conerete,” PCI Journal, 32(3): 74-150, 1987. 6.44 Sozen, M. A., E. M. Swoyer, and C. P. Siess, “Strength in Shear of Beams without Web Reinforcement,” Engineering Experiment Station Bulletin No. 452, University of Illinois, Urbana, April, 1959. 645 Taylor, H. P. J., “The Fundamental Behavior of Reinforced Concrete Beams in Bending and Shear,” Shear in Reinforced Conerete, Special Publication SP-42, Vol. 1, American Concrete Institute, Detroit, 1972, pp. 43-77, 646 Vecchio, F. J., and M. P. Collins, “The Modified Compression Field Theory for Reinforced Concrete Blements Subjected to Shear,” ACI Journal, Proceedings, 83(2): 219-31, 1986, 647 Vecchio, F. J., and Collins, M. P., s of Cracked Reinforced Conerete,” ASCE Journal of Structural Engineering, 119(12): 3590-3610, 1993, 648 Zia, P., “Torsion Theories for Concrete Members.” In Torsion of Structural Concrete, SP-18, American Conerete Institute, Detroit, 1968, pp. 103-132, 649 Zia, P., and W. D. McGee, “Torsion Design of Prestressed Conerete,” PCI Journal, 19(2): 46-65, 1974, Also Discussion in PC/ Journal, November/December, 1974. 6.50 Zia, P., and T. T. C. Hsu, “Design for Torsion and Shear in Prestressed Concrete,” ASCE ‘Annual Convention, Chicago, 1978, Preprint No. 3423, 17 pp. PROBLEMS 6.1 Because of high shear stresses (or principal tension) in the web of a beam near the supports, you propose to use vertical prestressing in combination with horizontal prestressing. Assuming 9, =700 psi and v = 400 psi, determine the magnitude of vertical prestress so that the principal tension is reduced at the centroid of the section to: (a) ~ 100 psi, or (b) zero, 4 2 17 a & _ 40 ft | igke Figure P6.2 6.2 Go back to Prob. 4.7g, where a steel profile has been selected, and check shear requirements along the span (Fig. P6.2). What can you conelude if you are told that the beam is part of a joist slab system?Chapter 6 - DESIGN FOR SHEAR AND TORSION 385 6.3 Go back to Prob. 4.8 and check shear requirements near the supports and at sections located 5 ft from the support on the cantilever side, and 5 and 10 ft on the span side. 6.4 The precast prestressed beam of Fig. P6.4 has been selected for the roof of a stadium. The beam is simply supported with, on one side, a cantilever of 20 fi. The following information is given: Live load = 40 psf Steel: 0.5 in diameter strands Section properties: =615 in? Su = 270 ksi A, = 615 in 7 4 Spy = 245 ksi 1 = 59,720 in Yp = 21.98 in diameter = 1/2 in 2 yy = 10.0: area per strand = 0.153 in’ : fp; at transfer = 175 ksi 2y = 2717 in Frye = 150 ksi Z, = 5960 in® 641 kif Conerete #2 = 5000 psi fy = 4000 psi Gj =-190 psi Fj =2400 psi Ty =—424 psi Foguy = 2250 psi to n= 120in A 8 cb Bor 2on Figure P6.4 (a) Determine the position of point B at which maximum positive moment occurs. Build for sections Band C the two feasible domains of ¢, versus I/F; and select a common satisfactory value of F corresponding to an even integer number of strands. (b) Determine the strands’ layout and their centroid at sections B and C and check the ultimate moment requirements according to ACI specifications at both sections. Is the ratio of ultimate to cracking moment at section C satisfactory? (c) Assuming F constant along the beam, construct the limit zone and the steel envelopes. Suggest a satisfactory steel profile along the beam. (d) Check shear requirements along the beam and determine the required stirrups. Plot graphically shear stresses versus abscissa and show the selected stirrup spacing. 6.5 An elevated guideway for a mobile lifting crane is made out of a series of consecutive simple span prestressed concrete beams. The maximum reaction from the wheel of the crane is P = 30 kips, and the span, center to center of supports, is 40 ft. The beam cross-section is a symmetrical / section with: b = 18 in., by = 6 in., hy= 4 in, and h = 30 in, The following information is provided:386 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN &-j = 2400 psi; Gj = -189 psi; Toquy = 2000 psi; Fg = 3000 psi B, = -422 psi Fi = 50000 psi; f= 4000 psi: fy, = 270 ksi; fy = 243 ksi; fpe = 150 ksi 4286 ksi; E,, 3834 ksi 1 = F/F, = 0.85; unit weight of concrete ~ 150 pef, fy = 60 ksi; dy =28 in Maximum practical eccentricity (¢))yp = 13 in Stress relieved low relaxation strands with area per stand = 0.216 in? 1. Design the beam as a fully prestressed beam for service load and check ultimate strength requirements for bending according to the ACI code, Use computer programs, if available, and summarize the results obtained. 2. Assume that the beam is designed as a partially prestressed beam with the following final prestressing force: F = 129.6 kips which corresponds to four strands with 4p, = 0.864 i eccentricity at midspan = maximum practical eccentricity; eccentricity at support = 6 in; tendons profile = single draping at midspan, Use the ACI code. (a) Using ultimate strength design, determine the area of reinforeing steel 4, needed in addition to Apsto satisfy nominal bending resistance criteria, () Determine the required shear reinforcement (i.c., spacing of No, 3 U stirrups) at the following sections along the span: 20 in, 5 ft, 10 fi, 15 ft, and 20 ft from the center of the left support. From the information obtained, plot a figure similar to Fig. 6.17 in this book. Use the elaborate method of analysis for shear. 3. Repeat question (2) using the AASHTO LRFD code. 18in 4 Figure P6.5 6.6 An L-shaped spandrel beam (also called ledger beam) is to be designed as part of a precast prestressed concrete parking structure. The spandrel beam supports, in addition to its own weight, the reactions from four double-T beams that are part of the floor system, The spandrel beam spans 29 feet between the columns, and the double-T beams span 44 feet. The floor surface is topped with a 2 in thick conerete topping leading to composite double-T beams. The topping adds 25 psf to the load, that is, 200 pif to each beam. The double-T beams are made of lightweight concrete. Each T-beam weighs 320 plf (8LD124 from PCI Handbook). Figure P6.6 provides some details. The purpose of this problem is to design the spandrel beam for bending, shear, and torsion according to the ACI code. The spandrel beam is subjected to concentrated loads (or reactions) imposed by each leg of the double-T beams. The following information is given: Live load ~ 120 psf on floor, which is equivalent to 960 pif for the double-T beams, 3150 psi; & = 4200 psi; d, = —502 psiChapter 6 - DESIGN FOR SHEAR AND TORSION 387 f= 7000 psis ff; = 5000 psis fy, = 270 sis fry, = 243 ksi; fine = 150 ksi F'/F; = 0.83; unit weight of concrete = 150 pet E, = 5072 ksis Fy; = 4286 ksi fy = 60 ks ‘The compressive strength of the concrete topping is 3500 psi The strands used are 0.6 inch in diameter. for stirrups; demin ~3 iN: (€, support = 6-5 in. i 32h ‘| | aft 16 el 2 | cxmmosggey =A TopPINO eeteee_42in FT Yee 2) A, aLpT24 6 le 3.75 A, =401 in?s/y = 20,985 in*s y, = 6.85 in; Seo L-shaped qi 3 f+ spangie! beam ¥p =17.15 inZ, = 3063 in; Z, = 1224 in or edget bean Beam 8LDT24 (lightweight concrete) 20 pit, Wiss = 25 pst = 200 pit —, >! Figure P6.6 1, Check that the prestressing force needed for the spandrel beam ean be provided by seven bonded strands with a maximum practical eccentricity at midspan, Five strands are taken straight and 2 strands are draped with a single draping section at midspan. Nonprestressed reinforcement is placed near the top fiber to account for tension under initial loading. 2. Show a detailed cross section with the location of each prestressing strand at midspan and at support so as to achieve the design eccentricity at each section: (€o)support = 6.5 in. 3. Design the vertical shear reinforcement for the spandrel beam assuming no torsion is to be considered. Use the ACI code. It is suggested to check shear requirements at the location of each leg of the double-T beams, using the maximum shear force computed at that section (that is, prior to subtracting the reaction of the leg). This will reduce the number of sections analyzed to four for half the span. Use No. 3 closed stirrups (two legs). 4. Assuming the ends of the spandrel beam are restrained against torsional rotation, design the transverse and longitudinal torsion reinforcement. Draw a typical section showing the details of all reinforcements used.388 Naaman - PRESTRESSED CONCRETE ANALYSIS AND DESIGN (teenies mt ol. Full-seale test of prestressed concrete bridge beam under combined bending, shear and torsion, (Courtesy Portland Cement Association.)