Behavior and Design of Reinforced Concrete, Steel, and Steel-Concrete Coupling Beams

Kent A. Harries , M.EERI, Bingnian Gong , and Bahram M. Shahrooz

1 2 3

Corresponding (first) author: Phone: Fax: 803 777 0671 803 777 0670

Kent A. Harries

Mailing address: USC-Civil Engineering, 300 Main Street, Columbia, SC 29208

E-mail address:

harries@engr.sc.edu

Submission date for review copies: February 7, 2000 Submission date for camera-ready copy: June 21, 2000

(KAH) Dept. of Civil and Environmental Engineering, University of South Carolina, Columbia, SC 29208 (BG) CBM Engineers, Inc., 1700 W. Loop So., Suite 830, Houston, TX 77027 (BMS) Dept. of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH 45221-0071

Behavior and Design of Reinforced Concrete, Steel, and Steel-Concrete Coupling Beams
Kent A. Harries , M.EERI, Bingnian Gong , and Bahram M. Shahrooz
4 5 6

The efficiency of coupled wall systems to resist lateral loads is well known. In order for the desired behavior of the coupled wall system to be attained, the coupling beams must be sufficiently strong and stiff. The coupling beams, however, must also yield before the wall piers, behave in a ductile manner, and exhibit significant energy absorbing characteristics. This paper reviews the current state of the art for the design of conventional reinforced concrete, diagonally reinforced concrete, steel and composite steelconcrete coupling beams. Although not exhaustive, critical aspects of the design of these systems are presented. INTRODUCTION An efficient structural system for seismic resistance can be achieved if openings in structural walls can be arranged in a regular pattern as shown in Figure 1. In this manner, a number of individual wall piers are coupled together to produce a system having large lateral stiffness and strength. By coupling individual flexural walls, the lateral load resisting behavior changes to one where overturning moments are resisted partially by an axial compression-tension couple across the wall system rather than by the individual flexural action of the walls. As a result, a significantly stiffer lateral load resisting system is achieved. The beams that connect individual wall piers are referred to as coupling beams. Coupling beams are analogous to and serve the same structural role as link beams in eccentrically braced frames. Well-proportioned coupling beams above the second floor generally develop plastic hinges simultaneously, and are subjected to similar end rotations over the height of the structure (Aktan and Bertero 1981, Aristizabal-Ochoa 1982, Shiu et al. 1981). Hence, dissipation of input energy can be distributed over the height of the building in the coupling beams rather than concentrating it predominantly in the first-story wall piers (Aktan and Bertero 1984, Fintel and Ghosh 1982, Park and Paulay 1975). For optimum performance, the energy dissipating mechanism should involve formation of plastic hinges in most of the coupling beams and at the base of each wall. This mechanism is similar to the strong column-weak girder design philosophy for ductile moment resisting frames (Paulay 1971, Park and Paulay 1975). In order for the desired behavior of the coupled wall system to be attained, the coupling beams must be sufficiently strong and stiff. The coupling beams, however, must also yield before the wall piers, behave in a ductile manner, and exhibit significant energy absorbing characteristics.

(KAH) Dept. of Civil and Environmental Engineering, University of South Carolina, Columbia, SC 29208 (BG) CBM Engineers, Inc., 1700 W. Loop So., Suite 830, Houston, TX 77027 (BMS) Dept. of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH 45221-0071

Coupling Beams

(a)

(b)

(c)

Figure 1. Examples of coupled wall geometry.

The degree of coupling of a coupled wall system is defined as the ratio of the total overturning moment resisted by the coupling action (i.e., the axial load couple formed in the walls) to the total overturning moment. The coupling action between wall piers is developed through shears in the coupling beams. The hysteretic characteristics of coupling beams, therefore, may substantially affect the overall response of a coupled wall system particularly for structures having a high degree of coupling (Aktan and Bertero 1981, Harries 1999, Paulay 1986, Paulay and Taylor 1981, Paulay 1980). Coupling beams should be designed to avoid over coupling, which causes the system to act as a single pierced wall with little frame action. Similarly, light coupling should also be avoided as it causes the system to behave like two isolated walls (Aktan and Bertero 1987, 1984, 1981, Aristizabal-Ochoa 1987, 1982, Aristizabal-Ochoa et al. 1979, Lybas and Sozen 1977, Shiu et al. 1981, 1984). Reinforced concrete coupling beams may have conventional longitudinal reinforcement, diagonal reinforcement or a hybrid reinforcing steel arrangement. Coupling beams may also be fabricated from rolled or built-up steel shapes, or have a composite steel-concrete design. Additionally, recent investigations have proposed alternative hybrid designs. REINFORCED CONCRETE COUPLING BEAMS
BEHAVIOR

It has been shown that the ductility capacity of a coupled wall system increases with increasing degrees of coupling (Harries 1998). The degree of coupling is a function of the relative stiffness and strength of the beams and walls. Saatcioglu et al. (1987) showed that coupling beams must be capable of developing displacement ductilities in excess of 6 for system ductilities of about 4 to be achieved. Several researchers have investigated novel approaches for improving the ductility and energy absorption capacity of reinforced concrete coupling beams. For span-to-depth ratios less than about 2, a system using specially detailed diagonal reinforcement, developed by Paulay (e.g., Paulay and Binney 1974), has been shown to significantly improve the reversed cyclic loading response. Shiu et al. (1978) confirmed the improved behavior of diagonally reinforced beams over conventionally reinforced beams. However, these tests also demonstrated that for larger span to depth ratios (2.5 and 5) the diagonal reinforcement was not as efficient due to its lower angle of inclination and hence its reduced contribution in resisting shear. The stiffness required of "efficient" coupling beams usually results in moment arm to effective depth ratios (a/d ratio) of less than 2 (Aktan and Bertero 1981). Indeed, a/d ratios of

1 or less are common (Aktan and Bertero 1981, Bertero 1980, Bertero et al. 1974, Paulay 1977, Paulay and Binney 1974, Paulay 1971). As the a/d ratio decreases, arching action contributes more significantly to shear strength than flexural behavior (Aktan and Bertero 1981, Paulay 1977, Paulay and Binney 1974, Paulay 1971). In conventionally reinforced coupling beams, both top and bottom reinforcement may simultaneously undergo tension as a result of diagonal cracking. Thus, it is recommended that the increased ductility due to compression bars not be accounted for (Park and Paulay 1975, Paulay 1971). Coupling beams require special detailing particularly if v u > 0.5 f c ' ( v u > 6 f c ' in psi units) and a/d<3.0.
8 - No 25 No 10 hoops @ 90 mm 2 - No 15 midside 8 - No 25 400 mm wall steel not shown 3000 mm 1.3ld = 1200 mm

700 mm

(a) Conventionally reinforced coupling beam

4 - No 30 in No 10 hoops @ 100 mm No 10 hoops @ 300 mm 4 - No 30 in No 10 hoops @ 100 mm 400 mm

700 mm

1.3ld = 1450 mm wall steel not shown 1200 mm

(b) Diagonally reinforced coupling beam

Figure 2. Examples of reinforced concrete coupling beams (Harries et al. 1998).

Conventional reinforcement, i.e., longitudinal bars with transverse steel (see Figure 2a), performs satisfactorily at nominal shear stresses of 0.25 fc ' ( 3 f c ' ) or less when the flexural mode controls the failure mechanism (Aktan and Bertero 1981, Aristizabal-Ochoa 1987). Sliding shear at the face of the wall begins to affect the response of the beams in the range of 0.3 f c ' ( 3.5 fc ' ) to 0.5 fc ' ( 6 f c ' ). By providing intermediate midheight longitudinal bars, the hysteretic response is improved and strength deterioration due to shear is delayed (Aktan and Bertero 1984, Scriber and Wight 1978, Tassios et al. 1996). Beams with intermediate bars do not perform well when the shear stress is greater than 0.5 fc ' ( 6 f c ' ). Providing cranked diagonal reinforcement near the beam ends has been shown to improve the hysteretic behavior by preventing sliding shear, and by spreading the hinging regions away from the wall face (Paparoni 1972, Aristizabal-Ochoa 1987). This detail, however, poses construction difficulties and results in extra cost.

Full-length diagonal reinforcement, as shown in Figure 2b, is a relatively simpler detailing alternative, and has been found to maintain stable hysteretic behavior provided the bars are adequately confined by spirals or hoops, and sufficient concrete cover is provided (Aktan and Bertero 1981, Paulay and Binney 1974). This confinement provides resistance against buckling to the individual bars and the bundled diagonal bar group. Such diagonal reinforcement is perhaps the only successful solution for reducing the potential for sliding shear, and enhancing the hysteretic characteristics of coupling beams having a/d ratios as large as 3.33 (Aktan and Bertero 1981). Considering that higher modes contribute to the response of coupling beams more than to walls (Aktan and Bertero 1981, Mahin and Bertero 1975), these elements are subjected to large inelastic end rotations and reversals. As a result, adequate anchorage of the coupling beam reinforcement in the walls is necessary. The possibility of buckling of compression bars at the beam-wall interface (where large cracks are anticipated) is also minimized by providing effective confinement (Aktan and Bertero 1981, Barney et al. 1978, Paulay 1980, Paulay and Santhakumar 1976, Paulay and Binney 1974). It is noted that anchorage and confinement requirements often make these diagonally reinforced coupling beams very difficult to construct due to congestion at the center of the beam and at the wall faces. (Recall that there will often be concentrated vertical wall steel at these interfaces.) Figure 3 shows a typical example of a diagonally reinforced coupling beam and its wall anchorage.

concentrated vertical reinforcing at face of wall

diagonal primary reinforcement

Figure 3. Diagonally reinforced concrete coupling beam (photo courtesy of Reinforcing Steel Institute of Canada).

Table 1 summarizes the guidelines for the selection and detailing of reinforced concrete coupling beams for seismic design given by the ACI 318-95 and ACI 318-99 Building Code Requirements for Structural Concrete (ACI 1995, 1999), CSA A23.3-94 Design of Concrete Structures (CSA 1994), and NZS 3101 Concrete Structures Standard (NZS 1995). When no seismic design is required, coupling beams may be detailed as regular reinforced concrete beams. It is noted, nonetheless, that these beams do typically exhibit very high shear to moment ratios requiring special attention.

Table 1. Guidelines for selecting reinforced concrete coupling beams

Governing Limits L/d shear stress, v (psi)

v 0.167 f c' + ?n f y v ac f '+? f c n y

Type of Coupling Beam ACI 318-95 Clause 21.6

Design Criteria

L/d 2 L/d 2

ACI 318-95 does not have provisions for diagonally reinforced concrete members

where 0.167 < ac 0.25

ACI 318-99 Clause 21.6.7

L/d 4

v ac

f '+? f c n y

conventionally reinforced

where 0.167 < ac 0.25

beams may be designed as flexural members according to Clause 21.3, however the commentary appears to invoke Clause 21.6.4

v 0.83 f c '
L/d 4

diagonally or conventionally reinforced diagonally reinforced

see above or below diagonal reinforcement to carry shear (Avdfy Vn/sina)

L/d 2

v 0.33 f ' c v 0.83 f ' c

CSA A23.3-94 Clauses 21.3 and 21.5

L/d 4

v 0.1( L / d ) f ' c

conventionally reinforced

i) beams capable of significant flexural hinging (Vr 2Mp /L) ii) provide confinement and antibuckling reinforcement

L/d 4

fc '

diagonally reinforced

i) diagonal reinforcement to carry shear (fsAsfy Vu/sina) ii) provide adequate confinement and anchorage for diagonal bars

NZS 3101-1995 Clauses 12.4.6.2 and 9.4.4.2 conventionally reinforced

i) beams capable of significant flexural hinging (Vr 2Mp /L) ii) provide confinement and antibuckling reinforcement

not specified

v 0.1( L / d ) f ' c

v 0.25 f ' c

some diagonal reinforcement required

i) diagonal reinforcement to carry shear (fsAsfy Vu/sina) ii) provide adequate confinement and anchorage for diagonal bars

The ACI 318-95 seismic provisions do not specifically address details for coupling beams. Clause 21.6.5.7 limits the nominal shear strength of a horizontal wall segment to
0.83A cp fc' ( 10A cp f c' ). The accompanying commentary states that the coupling beam is in effect, a pier rotated through 90 deg. Applying this interpretation allows the nominal shear strength of a coupling beam to be determined from Clauses 21.6.5.2 or 21.6.5.3, depending on the span-to-depth ratio of the beam. If this interpretation is applied, Clause 21.6.5.3 permits the shear stress in a very short coupling beam (L/d < 2.0) to be greater than that in a longer beam. From the previous discussion of coupling beam behavior, it is clear that this interpretation runs counter to observed beam behavior. ACI 318-95 does not address diagonally reinforced concrete beams in any way.

ACI 318-99 has adopted specific provisions for coupling beams. For beams having spanto-depth ratios, L/d, greater than 4.0, coupling beams are designed as flexural members satisfying the requirements of Clause 21.3. It is noted, however, that Clause 21.6.4.5 (formerly 21.6.5.7 in 318-95) and the accompanying commentary continues to suggest that horizontal wall segments may be treated as pier[s] rotated through 90 deg. as discussed above. For coupling beams having span-to-depth ratios between 2 and 4, Clause 21.6.7.2 allows the use of diagonal or conventional reinforcement. For beams having span-to-depth ratios less than 2 and a factored shear force exceeding 0.33 f c' ( 4 f c' ), diagonal reinforcement is required. In Table 1, it is noted that the Canadian standard CSA A23.3 provides the clearest direction as to the use of diagonal reinforcement: diagonal reinforcement requirements are triggered by both span-to-depth ratios and shear stresses in the beam. The New Zealand standard NZS 3101, on the other hand, combines these triggers into one: shear stress, normalized by the span-to-depth ratio. NZS 3101 does, however, require diagonal reinforcement if the shear stress exceeds an absolute threshold value of 0.25 f c' ( 3 f c' ). Furthermore, NZS 3101 Commentary Clause C9.4.4.2 appears to permit shear to be resisted by a combination of diagonal and conventional reinforcement providing that the contribution of the conventional reinforcement remains below the threshold value. For seismic design, the New Zealand Standard does recommend, however, that 100% of the shear forces be carried by diagonal reinforcement in any case where diagonal reinforcement is required. The design and detailing requirements for diagonally reinforced concrete coupling beams are essentially identical in both the Canadian and New Zealand standards.
ANALYSIS OF WALLS COUPLED WITH REINFORCED CONCRETE COUPLING BEAMS

Geometric Modeling A simple two-dimensional equivalent frame model suitable for elastic analysis of coupled wall systems is illustrated in Figure 4. Such a model can be used to compute beam and wall design forces. This model has several drawbacks; the most significant of which is that it does not simulate the shift of the neutral axis in the wall piers beyond elastic limit. Other models appropriate for inelastic analysis are available (Kunnath and Reinhorn 1994, Kabeyasawa et al. 1983, Teshigawara et al. 1996, Milev and Kabeyasawa 1996, Vulcano and Bertero 1987). Multiple pier walls may be similarly modeled or may be reduced to a simpler two-pier model using further analytic tools (Stafford-Smith and Coull 1991).

Frame Elements

Rigid Elements

L
Figure 4. Two dimensional equivalent frame model of coupled wall system.

Modeling Beam and Wall Properties Elastic analysis is sufficient to determine a theoretical degree of coupling for a coupled wall structure and thus the distribution of internal forces in the walls and beams. Local inelasticities at the beam-wall interfaces, redistribution of forces between coupling beams and from the tension wall to the compression wall and strain hardening effects all contribute to reducing the theoretical degrees of coupling. Reduced section properties, accounting for cracking and loss of stiffness due to cycling can be used to determine a more appropriate value for the theoretical degree of coupling. Table 2 shows the reduced stiffness values suggested by the current ACI, CSA and NZS standards. ACI 318-99 recommends the use of any set of reasonable assumptions[which are] consistent throughout the analysis and recommends using one half of the gross section properties for flexural members. It is interesting to note that, using the reduction factors recommended by the New Zealand Standard, one computes effectively stiffer coupling beams and more flexible wall piers, resulting in a larger degree of coupling, and hence greater beam forces, than one computes with the ACI or Canadian recommendations.
Table 2. Recommended reduced member stiffnesses

Member compression wall in flexure tension wall in flexure compression wall axial tension wall axial average wall axial2 conventionally reinforced coupling beam diagonally reinforced coupling beam

ACI 318-99 0.70EI (uncracked) 0.35EI (cracked) 1.00EA 0.35EI (inferred) not applicable 0.35EI not applicable

CSA A23.3-94 0.80EI 0.50EI 1.00EA 0.50EA 0.75EA 0.20 EI 1 + 3(h / L ) 2 0.40 EI 1 + 3(h / L) 2

NZS 3101-19951 0.45EI 0.25EI 0.80EA 0.50EA not applicable 0.40 EI 1 + 8(h / L) 2 0.40EI 1.7 + 2.7(h / L) 2

1. NZS 3101 has different recommendations for different limit states. The values corresponding to the most critical limit state are given here. 2. CSA A23.3 suggests that an average axial wall stiffness is appropriate to simplify analyses.

Design As shown in Table 1, depending on the shear span, coupling beams are designed differently. Flexural Failure Mode Coupling beams are designed and detailed as standard beams if both L/d 4 and Vu 0.33 f c 'b w d ( Vu 4 f c 'b w d ). If the structure is located in high seismic regions (UBC Zone 4, NEHRP Seismic Design Category D or E), provisions from ACI 318-99 Section 21.3 are used. ACI 318-99 Section 21.9 is followed for structures in moderate seismic regions (UBC Zone 3, NEHRP Seismic Design Category C). It is noted that ACI 318-99 provisions for minimum beam width (Clauses 21.3.1.3 and 21.3.1.4) may be waived if rational analysis shows that lateral stability of the beam is adequate, or lateral stability is provided by alternative means. Often the provision of an integral slab will be sufficient to provide lateral stability. The longitudinal flexural bars in the beam should be anchored adequately in the wall piers. For high seismic regions (UBC Zone 4, NEHRP Seismic Design Category D or E), development length requirements in ACI 318-99 Clause 21.5.4 must be met in addition to those in ACI 318-99 Chapter 12. For moderate seismic regions (UBC Zone 3, NEHRP Seismic Design Category C), development lengths for the longitudinal bars based on ACI 318-99 Chapter 12 are considered sufficient. Shear Failure Mode If L/d < 4 and Vu > 0.33 fc 'b wd ( Vu > 4 f c 'b w d ), two intersecting groups of diagonal bars are to be used to reinforce coupling beams. Although it is preferable for each intersecting group to contain the same number and size of reinforcing bar, this is often not practical. Where this is the case, it is necessary to ensure that the amount of steel provided in each diagonal is the same. If L/d < 4 and Vu < 0.33 f c ' b w d ( Vu < 4 fc 'b w d ), diagonal reinforcement may also be used. Each group of diagonal bars must have a minimum of four bars placed in a core with a minimum depth and width equal to the greater of bw/5 in the plane of the diagonal bars or bw/2 in the plane of the beam (ACI 318-99 Clause 21.6.7.4 (a)). The diagonal bars must be confined according to ACI 318-99 clauses 21.4.4.1 through 21.4.4.3. The required area of one leg of the diagonal reinforcement, A vd is (ACI 318-99 Clause 21.6.7.4): Vn = 2 Avd f y sin 0.83 f c 'bwd (1)

Although the diagonal reinforcement is intended to prevent sliding shear, the value of coupling beam shear is limited to 0.83 f c 'bw d ( 10 f c 'bw d ). Minimum horizontal and vertical reinforcement requirements for deep beams (ACI 318-99 clauses 11.8.9 and 11.8.10) must also be respected. Considering the importance of adequate anchorage for the diagonal bars, it has been recommended that the development length, calculated based on Chapter 21 of ACI 318-99, inside the wall piers be increased by as much as 50%, if possible (Paulay 1986). Although there are no provisions in the Standards for increasing the development length, it is recommended that these bars be considered as top cast bars and that appropriate factors be

applied as shown in Figure 2 (Harries et al. 1998). Often geometric constraints within the wall will require that these bars be provided with bent or hooked anchorages (see Figure 3). As previously stated, this detail may be very difficult to construct. STEEL OR STEEL-CONCRETE COMPOSITE COUPLING BEAMS Structural steel members provide a viable alternative to reinforced concrete coupling beams. Furthermore, a steel member may be encased with varying levels of longitudinal and transverse reinforcement. The resulting steel-concrete coupling beams are referred to as composite coupling beams herein. The advantages of steel or composite coupling beams become particularly apparent in cases where height restrictions do not permit the use of deep reinforced concrete beams, or where the required capacities and stiffness cannot be developed economically by a concrete beam. Similar to reinforced concrete coupling beams, energy dissipation characteristics of steel or composite coupling beams play an important role in the overall performance. Depending on the coupling beam length, steel coupling beams may be detailed to dissipate a major portion of the input energy by flexure or by shear. For most coupling beams, however, it is more advantageous to design the coupling beams as shear critical, or shear yielding members since such members exhibit a more desirable mode of energy dissipation. Such a choice is not possible for reinforced concrete members. The coupling beam-wall connection depends on whether the wall boundary element is reinforced conventionally or contains embedded structural steel columns. In the latter case, the connection is effectively a joint between a steel beam and column, as shown in Figure 5. Standard design methods for steel structures can be used for this connection (AISC 1995, 1997); however, proper measures must be taken to prevent failures similar to those observed in steel moment-resisting frames during recent earthquakes. Furthermore, such a composite wall system will not require significant amounts of boundary element reinforcement.

steel boundary column

wall steel not shown shear studs

steel coupling beam

holes in beam

moment connection

face of wall
Figure 5. Coupling beam connection to embedded steel column.

If the wall boundary element is reinforced with longitudinal and transverse reinforcing bars, a typical connection involves embedding the coupling beam into the wall and

interfacing it with the boundary element, as shown in Figure 6. In addition to boundary element reinforcing, embedded steel members may also be provided with welded vertical reinforcing passing through their flanges or shear studs to resist horizontal loads (PCI 1992). The vertical bars may be attached to the flanges through mechanical half couplers which are welded to the flanges. These bars must have adequate development length, and must be large enough to provide enough stiffness to the coupling beam. Shahrooz et al. (1992, 1993) have found 25.4 mm diameter (#8) bars to be sufficient for this purpose. Such details result in greater congestion in the boundary element. It has been shown that these additional welded appendages are not necessary provided that the embedment region is sufficiently long and that there is adequate boundary element reinforcing to control cracking at the beam-wall interface (Harries 1995, Harries et al. 1997, Shahrooz et al. 1993). Harries et al. recommend that two-thirds of the required vertical boundary element steel be located within a distance of one half the embedment length from the face of the wall. Furthermore, the width of the boundary element steel should not exceed 2.5 times the coupling beam flange width. Satisfying these requirements will provide adequate control of the gap that opens at the beam flanges upon cycling. With typical boundary element designs, these requirements are easily met.

No. 10 hoops and hairpins No. 10 hoops and hairpins SECTION A-A 900 1000 900 12 - No. 25 No. 10 hoops @ 150 No. 10 @ 150 vertical e.f. No. 10 @ 100 horizontal e.f. SECTION B-B (b) shear critical steel coupling beam SECTION A-A

14 - No. 35 No. 10 hoops @ 200 No. 10 @ 150 vertical e.f. No. 10 @ 100 horizontal e.f. SECTION B-B (a) flexure critical steel coupling beam

Figure 6. Steel coupling beam embedment details (Harries et al. 1998).

As is illustrated in Figure 6a, lighter flexure critical coupling beams require more boundary element steel in the wall since a greater proportion of the overturning moment is carried by the wall. This may also result in the need to provide barbells at the toes of the walls, further complicating construction. It is not necessary, nor is it practical, to pass boundary element reinforcing through the web of the embedded coupling beam. It is however necessary to provide good confinement in the region of the embedded web. This may be accomplished using hairpins and cross ties parallel to the web as shown in Figure 6 (section A -A). Additionally, vertical boundary element reinforcement in this region may also be relied upon to provide significant confinement to the embedded region (Harries et al. 1997).
STEEL-CONCRETE COMPOSITE COUPLING BEAMS

The selection of the embedded steel beam in a composite coupling beam typically neglects the contribution of the encasing concrete. The contribution of the encasing concrete, however, must be accounted for in subsequent stiffness and embedment length calculations. The presence of the encasing concrete does permit the detailing requirements of the embedded member to be relaxed. Gong and Shahrooz (1998) have shown that stiffeners are not required in the case of a composite coupling beam. The surrounding concrete is sufficient to provide adequate stability and thus ductility to the embedded steel beam.
ANALYSIS OF WALLS COUPLED WITH STEEL OR STEEL-CONCRETE COMPOSITE COUPLING BEAMS

Stiffness For steel coupling beams, standard methods are used to calculate the stiffness. The stiffness of steel-concrete composite coupling beams needs to account for the increased stiffness due to encasement. Stiffness based on gross transformed sections should be used to calculate the upper bound values of demands in the walls, most notably wall axial force. The cracked transformed section moment of inertia may be used when deflection limits or coupling beam shear angles are checked. A previous study indicates that the additional stiffness due to the presence of a composite floor slab is lost shortly after coupling beams undergo small deformations (Shahrooz and Gong 1998). Until additional experimental data become available, it is recommended to include the participation of the floor slab when the stiffness of steel-concrete composite coupling beams are computed. Effective flange width for T beams as specified in ACI 8.10.2 may be used for this purpose. Deflection limits should be checked neglecting the contribution of the floor slab. Geometric Modeling Previous studies (Shahrooz et al. 1992, 1993, Gong et al. 1997, 1998, Harries et al. 1997) suggest that steel or steel-concrete composite coupling beams are not fixed at the face of the wall. As part of design calculations, the additional flexibility needs to be taken into account to ensure that wall forces and lateral deflections are computed with reasonable accuracy. Based on experimental data (Shahrooz et al. 1992, 1993, Gong et al. 1997, 1998), the effective fixed point of steel or steel-concrete composite coupling beams may be taken as approximately one-third of the embedment length from the face of the wall. Thus, the effective clear span, L of the frame element representing the coupling beam in the model shown in Figure 4 is:

L = Lclear + 0.6 Le

(2)

The aforementioned method assumes that the walls have been modeled by column elements located at the centroid of wall piers as shown in Figure 4. Finite element modeling of structures have become more common in recent years where shell elements are used to model the wall piers. These elements are typically connected to nodes along fictitious column lines which are placed at the wall boundaries. One may attempt to account for the additional flexibility of steel/steel-concrete composite coupling beams by shifting the column lines, which define the walls, to a distance equal to 1/3 of the embedment length from the face of the wall. However, this approach will alter the wall properties, and other methods are necessary. A possible approach is to obtain equivalent values for the coupling beam axial area and moment of inertia in order to incorporate the additional beam flexibility. Equations to obtain the lower and upper bound values of the equivalent area and moment of inertia have been derived (Tunc et al. 2000). Another alternative is to use an effective stiffness. Approaching the same decrease in stiffness from another point of view, Harries et al. (1998) propose the use of an effective stiffness of the steel coupling beam of 0.60kEI. The factor k represents the reduction in flexural stiffness due to shear deformations:

12 EI k = 1 + 2 L GAw

-1

(3)

If the modeling technique used includes shear deformations, k = 1. With this approach, the effective length of the beam in increased only to account for spalling at the face of the wall: L = L clear + 2 c (4)

This approach does not require a preliminary design to determine the coupling beam embedment length. The effective stiffness of 0.60kEI has been observed to agree very well with experimental observations of beams having significantly different embedment details (Harries 1995)
DESIGN

Steel Coupling Beams Well established guidelines for shear links in eccentrically-braced frames (AISC 1997) may be used to design and detail steel coupling beams. The expected coupling beam rotation angle plays an important role in the required beam details such as the provision of stiffeners. This angle is computed with reference to the collapse mechanism shown in Figure 7 which corresponds to the expected behavior of coupled wall systems, i.e., plastic hinges at the base of walls and at the ends of coupling beams. The value of qp is taken as Cd qe, where the elastic interstory drift angle, qe, is computed under code level lateral loads (e.g., ICBO 1994; NEHRP 1997). Knowing the value of qp, shear angle, gp is calculated as:

p =

p Lwall
L

(5)

Note that in this equation the additional flexibility of the steel or composite coupling beams may be taken into account by increasing the length of coupling beam as indicated in

Equations 2 or 4. This approach provides a reasonable approximation even if shell elements have been used to model wall piers. Steel-Concrete Composite Coupling Beams Previous research on steel-concrete composite coupling beams (Shahrooz et al. 1992, 1993, Gong et al. 1997, 1998) indicates that nominal encasement around steel beams provides adequate resistance against flange and web buckling. Therefore, composite coupling beams may be detailed without web stiffeners. Due to inadequate data regarding the influence of encasement on local buckling, minimum flange and web thicknesses similar to steel coupling beams should still be used.

plastic hinges

qp gp Lwall gp = qp L
L Lwa ll
Figure 7. Determination of coupling beam angle of rotation.

Coupling Beam Embedment The coupling beam is embedded in the wall such that its capacity can be developed. A number of methods may be used to calculate the necessary embedment length. Either method can be used although the equations proposed by Marcakis and Mitchell (1980) generally result in slightly longer embedment lengths. Mattock and Gaafar (1982) This model is based on mobilizing an internal moment arm between bearing forces Cf and Cb as shown in Figure 8a. A parabolic distribution of bearing stresses is assumed for Cb, and Cf is computed by a uniform stress equal to 0.85f' c. The bearing stresses are distributed over the beam flange width, bf. Following these assumptions and calibrating against experimental data, the required embedment length, Le, may be determined from:

twall Vu = 4.05 f c ' bf

0.66

0.58 - 0.22 1 1b f Le 0.88 + a Le

(6)

Marcakis and Mitchell (1980) Using slightly different assumed stress distributions shown in Figure 8b, and assuming a rigid-body motion of the embedded element, the following expression may be used for determining the required embedment length, Le. This expression has also been calibrated against experimental data.

Vu =

0.85c f c ' b' ( Le - c ) 3.6e 1+ ( Le - c)

(7)

Where e is the eccentricity from the midspan of the coupling beam to the center of the effective embedment as shown in Figure 8b. In either method, the value of the shear span, a is taken as one half of coupling beam length if the inflection point is assumed to be at the midspan. It is suggested that the shear span be increased to account for spalling of the walls as calculated in Equation 4. The significant difference between these methods is that Marcakis and Mitchell account for spreading of the compression forces within the wall. As a result, the width of the effective compression zone, b, may be greater than the width of the beam flange, bf.
e Le - c Cb c L/2 = a spalled cover concrete Vu Le - c 2 Cb c L/2 = a spalled cover load spreading concrete bf V
u

Cf Le eb

C L
eb assumed strain e f = 0.003 distribution xb

Cf Le xf ef = 0.003

C L

b < 2.5bf

Cb 1/3(L e-c) 0.85f c 3/4(L e -c) (a) Mattock and Gaafar (1982) assumed afc stress distribution b1x f b xb 0.85fc

(b) Marcakis and Mitchell (1980)

Figure 8. Methods for determining embedment capacity.

Calculation of Embedment Length The embedment length, Le, required to develop the coupling beam shear, V u , may be calculated by either of the two aforementioned methods. For steel coupling beams, Vu is taken as 1.5 times the plastic shear capacity of the steel member, Vp:

Vu = 1 .5V p = 1.5 x 0.6F y ( h - 2t f )tw

(8)

For flexure critical steel beams, the shear capacity may be taken as the shear corresponding to the development of the plastic moment capacity:

Vu = 1.35

2M r 2M r = 1.5 Ls L

(9)

The contribution of concrete encasement to the shear capacity of composite coupling beams needs to be taken into account when the embedment length is calculated (Gong et al., 1997 and 1998):

As f y d Vu = 1.56(V p + VRC ) = 1.56(0.6 Fy ( h - 2t f )tw ) + 0.167 fc 'bw d + s

(10)

Additional transfer bars attached to the beam flanges (as discussed above) can contribute to the capacity of the embedment. The embedment length computed by Mattock and Gaafar (1982) or Marcakis and Mitchell (1980) can be modified to account for the additional strength. To ensure that the calculated embedment length is sufficiently large to avoid excessive inelastic damage in the connection region, it is recommended that the contribution of transfer bars be neglected. DESIGN EXAMPLES Examples included in Appendix A illustrate the designs of coupling beams discussed in this paper. Details of the beam and wall designs contained in Appendix A may be found in Harries (1996) and Harries et al. (1998). These examples are intended for illustrative purposes. CONCLUSIONS The efficiency of coupled wall systems to resist lateral loads is well known. In order for the desired behavior of the coupled wall system to be attained, the coupling beams must be sufficiently strong and stiff. The coupling beams, however, must also yield before the wall piers, behave in a ductile manner and exhibit significant energy absorbing characteristics. This paper reviews the current state of the art for the design of conventional reinforced concrete, diagonally reinforced concrete, steel and composite steel-concrete coupling beams. Although not exhaustive, critical aspects of the design of these systems are presented. In the case of diagonally reinforced concrete coupling beams, only included in ACI provisions for the first time in ACI 318-99, considerable additional guidance in design and detailing may be found in following Canadian (CSA) or New Zealand Standards (NZS).

Steel coupling beams may be efficiently designed and detailed as link beams in eccentrically braced frames. When encased in concrete, considerable stability is imparted to the beam allowing simplified details to be used. The proper design of any coupled system must ensure the desired hierarchy of hinging. In addition it is critical that hinges in the coupling beam remain in their clear span and do not propagate into the walls. As such proper design of the embedment region is critical. There are a number of methods of determining the required embedment length and details of steel or composite coupling beams. All of these methods are based on an assumed rigid body rotation of the embedment region of the coupling beam.

APPENDIX A COUPLING BEAM DESIGN EXAMPLES Salient points of the design of four types of coupling beam are presented. The beams are located in eighteen storey prototype structures located in Vancouver, Canada. The plan of these prototype structures is shown in Figure A1. Seismic design forces were determined using the 1995 National Building Code of Canada equivalent lateral force method. Base shears were determined to be in the range of 0.036W to 0.041W depending on the prototype structure. These same base shears would be found by applying the NEHRP equivalent lateral force method using R=5.8 to 6.7. NEHRP recommends a value of R=6 for either prototype structure.
900 mm x 900 mm "barbells" (storys 1 to 5) 800 mm x 800 mm "barbells" (storys 6 to 18)

Note: wall thickness = 500 mm (storys 1 to 5) = 400 mm (storys 6 to 18) 550 x 550 column (typ.) 6000 (typ.)

500

9000

5500 3000 5500 14 000 Partially (lightly) coupled base shear = 0.041W

200 mm thick flat plate 500 6000 (typ.) 42 000 5500 end bays 6400 9000

6400 1200 14 000 Typical floor plan Fully (heavily) coupled base shear = 0.036W Prototype cores Figure A1. Plan of prototype structures.

Each of the 3000 mm long prototype beams are designed as either conventionally reinforced concrete coupling beams or flexure critical steel coupling beams. Similarly, the 1200 mm beams are designed as both diagonally reinforced concrete coupling beams and shear critical steel beams.
CONVENTIONALLY REINFORCED CONCRETE COUPLING BEAMS

The final details of the 3000 mm long prototype conventionally reinforced concrete coupling beam are shown in Figure 2a. The design forces on these beams were determined to be 467 kN in shear and 700 kNm in flexure. The span-to-depth ratio, L/d, for these beams is 4.3 and the shear stress is approximately 0.42 f c ' , thus diagonal reinforcement is required. The design progressed according to standard beam design methods of ACI 318-99 or CSA A23.3-94.

30 000

DIAGONALLY REINFORCED CONCRETE COUPLING BEAMS

The final details of the 1200 mm long prototype diagonally reinforced concrete coupling beam are shown in Figure 2b. The design shear force for these beams was determined to be 770 kN. The span-to-depth ratio is 1.7 and the shear stress is approximately 0.25 fc ' , thus conventional reinforcement is permitted. The design progressed according to standard beam design methods of ACI 318-99 or CSA A23.3-94: Determine required dimensions in diagonally reinforced beam Assume diagonal bar bundles consisting of 2 layers of No. 30 bars. Bar bundles, including ties are 100 mm deep and 200 mm wide (see Fig. 2b). d = 700 - 40 - 100/2 = 610 mm d = 90 mm tan a = (h - 2d)/Lu = 700 - (2 x 90)/1200 therefore: a = 24o Selection of main diagonal reinforcement Diagonal reinforcement is determined from Equation A1:
As = Vn 770000 = = 2783mm 2 2 s f y sin 2 x 0.85 x 400 x sin 24

(A1)

therefore use 4 - No. 30 bars in each diagonal (As = 2800 mm2) Embedment and confinement of diagonal reinforcement Recommended embedment length = 1.5Ld = 1.5 x 1480 = 2220 mm Seismic hoops are required over entire length of diagonal with spacing not exceeding: 6db = 6 x 30 = 180 mm 24db (hoops) = 24 x 10 = 240 mm 100 mm Thus, provide No. 10 closed hoops at a spacing of 100 mm over the length of the diagonals.
FLEXURE CRITICAL STEEL COUPLING BEAMS

Ideally, the selection of a steel coupling beam must provide adequate stiffness for the desired degree of coupling, while limiting overstrength. Furthermore, the steel section must be narrow enough to fit within the concentrated wall reinforcement. For longer, flexure critical coupling beams, these criteria can often be satisfied with standard rolled shapes. For shorter spans, however, built-up sections may be required. Details of the flexure critical coupling beam are shown in Figure A2a. For the 3000 mm prototype steel coupling beam the design shear was determined to be 597 kN, corresponding to a moment of 895 kNm. For these forces, a W610 x 140 (W24 x 94; inchpound units) section with a 915 mm embedment region was designed.

At this point, it is important to verify that the section will indeed behave as expected. In this case, that it is flexure critical. The factored shear capacity is determined to be 1340 kN while the shear corresponding to the unfactored probable moment capacity is: 2M p L + 2c = 2 x1.5F y Z L + 2c = 2 x1.5 x300 x2510 = 734kN 3000 + 2 x 40 (A2)

Thus the beam is flexure critical. Required embedment length The required embedment length is determined from the following relation, (based on Marcakis and Mitchell, 1980), where the factored shear resistance of the embedment, Vu, is taken as 1.35Mr/2Lfs = 1.35 x 727 / 0.9 = 1090 kN (see Figure 8b).

Vu =

0.85c f c ' b' ( Le - c ) 3.6 e 1+ ( Le - c )

(A3)

Because of the presence of a barbell at the wall toe, b may be taken as 2.5b = 2.5 x 230 = 575 mm. Taking b = 575 mm, e = (3000 + Le + 40) / 2, and c = 40 mm. Solving Equation A3 yields: Le = 915 mm. It is necessary to ensure that there is adequate concentrated boundary element reinforcement, Asc, to develop Ve:
Asc = Ve 1090000 = = 3206 mm 2 s f y 0.85 x 400

(A4)

This requirement is essentially satisfied, as this steel represents a small portion of what is required in the region of concentrated reinforcement, and there will typically be some overstrength provided. Details of the embedment region of this prototype are shown in Figure 6a. At the beam-wall interface, double-sided stiffeners are provided to simplify the formwork and to aid in the force transfer between the steel beam and the concrete wall. Due to the high moment to shear ratio and the need to ensure lateral stability, single-sided stiffeners are located 335 mm from the face of the wall. The required embedment length may also be determined using the equation proposed by Mattock and Gaafar (1982): t wall Vu = 4.05 f c ' b f
0.66

0.58 - 0.22 1 1b f Le 0.88 + a Le

(A5)

If the bearing area width, twall, is assumed to be equal to the entire 900 mm width of the barbell, the required embedment length is found to be 725 mm. It is noted that if the width of the effective bearing area is limited to 2.5bf (Marcakis and Mitchell 1980), the required embedment length is increased to 870 mm.

3000 mm d = 617 mm b = 230 mm t f = 22 mm t w = 13 mm W610 x 140

915 mm

335 mm

(a) flexure critical prototype

d = 600 mm b = 225 mm t f = 25 mm t w = 9 mm built-up section

50 mm

1200 mm

1000 mm

300 mm typ.

(b) shear critical prototype

Figure A2. Prototype steel coupling beams.

SHEAR CRITICAL STEEL COUPLING BEAM

For the prototype structure having the 1200 mm long steel coupling beam and a high degree of coupling, the coupling beam was designed to yield in shear while remaining elastic in flexure. A 600 mm deep built-up section having an embedment length of 1000 mm was selected. The high shear demands on this beam necessitated the provision of single-sided stiffeners spaced at 300 mm over the entire span, these were continued into the embedment to provide stability in the event of excessive spalling. Figure A2b shows details for this shear critical steel coupling beam. Details of the embedment region are shown in Figure 6b. The design forces for this beam were 764 kN and 458 kNm for shear and moment, respectively. The design criteria called for a depth of no more than 600 mm (similar to the flexure critical and concrete beams). The resulting built-up section has dimensions shown in Figure A2. The required embedment length is determined from Equation A3 to be 1000 mm, similar to that required for the longer flexure critical beam. Similarly, using Equation A5, the required embedment length is determined to be 780 mm.

COMPOSITE COUPLING BEAMS The previously designed steel coupling beams are assumed to be encased as shown in Figure A3. The longitudinal reinforcement consists of four 10M (diameter = 11.3 mm) corner bars and two 10M midside bars. The transverse reinforcement consists of 10M closed hoops spaced at 390 mm. For simplicity, it is assumed that the steel beams are identical to those used previously, i.e., the contribution of concrete is ignored when the steel coupling beams are sized. This would often be the case for composite coupling beam design. The dimensions of the steel beams are given in Figure A2.
No. 10 corner and midside bars 760 mm No. 10 closed hoops or hairpins @ 390 mm

390 mm
Figure A3. Cross section of composite coupling beam.

The overall dimensions for both the 3000 mm and 1200 mm long coupling beams are h = 780 mm and bw = 390 mm. These dimensions were obtained based on assuming that the longitudinal bars would be at about 25 mm from the corners of the flanges, and by providing 40 mm of cover to the transverse reinforcement. The concrete compressive strength is taken as 35 MPa, and the yield strength of the longitudinal bars is taken as 420 MPa. The required embedment length for the flexural and shear critical beam is calculated in the following. Flexural Critical Beam The shear capacity of the coupling beam is determined as the sum the contribution from the steel beam and that of the of the concrete encasement (Equation 9):

As f y d Vu = 1.56(V p + VRC ) = 1.56(0.6 Fy ( h - 2t f )tw ) + 0.167 fc 'bw d + s

(A6)

Using an effective depth, d = 780 81.5 = 698.5 mm, the ultimate shear capacity of the coupling beam is found to be 2747 kN. The required embedment length may be calculated using Equation A5 (Mattock and Gaafar 1982). Similar to the previous example, t wall is taken as 900 mm. The value of bf is taken as the flange width of the steel coupling beam. The value of a is taken as half of the 3000 mm span. By applying Equation A5, the required embedment length, Le, is found to be 1285 mm. The required embedment length determined using the method proposed by Marcakis and Mitchell (1980), given in Equation A3 is found to be 1690 mm. Similar to the previous example, this assumes that b is taken as 2.5b = 2.5 (230) = 575 and e = 0.5L + c+ 0.5(Le-c) = 1520 + 0.5Le (see Figure 8).

It is suggested that since the beam is flexure critical the value of Vp may be replaced with the value of the shear required to achieve the ultimate flexural capacity of the beam, shown above to be equal to 1.35 x 2Mr/Lfs = 1.35 x 727 / 0.9 = 1090 kN. In this case, the required shear capacity of the embedment becomes:
Vu = 1.35 As f y d 2M r 2M r + 1.56VRC = 1.5 + 1.56 0.167 f c 'bw d + L s L s

(A7)

For the flexure critical beam designed above, the required shear capacity of the embedment region is V u = 1090 + 1.56(420) = 1745 kN. The require embedment length is found to be 970 mm (Equation A5) or 1280 mm (Equation A3). Shear Critical Beam For the shear critical beam, the method is identical to that used for the flexural critical composite beam although only Equation A6 is considered. The design shear force for the embedment, V u is 2,026 kN. Based on Mattock and Gaafars model (Equation A5), setting twall = 500 mm, the necessary embedment length is 1025 mm. When using the model proposed by Marcakis and Mitchell (Equation A3), the value of b is approximately 400 mm, limited by the width of the confining steel (see Figure 8b), and e = 620 + 0.5Le. Based on this method, the value of Le is 1350 mm. The preceding calculations demonstrate the effect of including the capacity of the concrete encasement. Although the embedments are longer, it has been shown that encased steel coupling beams do not require the significant stiffener details (see Figure A2) in order to ensure adequate ductility and stability. DESIGN OF WALL PIERS The design of the wall piers is then carried out for flexure using axial loads resulting from full plastification of all coupling beams. Figure 6 gives some indication as to the resulting boundary element reinforcement required for these prototype structures. ACKNOWLEDGEMENTS The authors would like to gratefully acknowledge the input and support of members of American Concrete Institute (ACI) Committee 335 Composite and Hybrid Structures and Professor Denis Mitchell of McGill University. The design methods and recommendations are based on several projects funded by the U.S. National Science Foundation (Grants No. BCS-8922777 and BCS-9319838 with Dr. Shih Chi Liu as the program director) and the Natural Science and Engineering Research Council of Canada. Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the sponsors. NOTATION a A As Asc Avd = = = = = shear span of coupling beam, typically taken as L/2 cross sectional area of wall pier area of transverse reinforcement in concrete encasement area of concentrated reinforcement in boundary element of wall pier area of reinforcement in each group of diagonal bars

Aw b' bf bw c Cd Cf d

= = = = = = = =

e = E = fc = fy = Fy = G = h = I = k = L = Lclear = Le = Lwall = Mr = s = tf = tw = twall = v = vu = Vc = Vn = Vp = Vr = VRC = Vu = xb = xf = a = ac = b or b1= gp = l = qp = qe = rn = fc fs = =

area of web of steel beam effective width of the concrete compression block coupling beam flange width, also the bearing width of the embedment web width of concrete beam or concrete encasement depth of concrete cover deflection amplification factor defined by NEHRP (BSSC 1997) resultant bearing (compressive) force in embedment distance from the extreme compression fiber to centroid of longitudinal tension reinforcement in concrete section eccentricity from midspan of beam to center of embedment Youngs modulus of coupling beam or wall pier specified concrete compressive strength yield strength of reinforcing steel yield strength of steel beam shear modulus overall depth of coupling beam moment of inertia of coupling beam or wall pier stiffness factor accounting for shear deformations in flexural members effective length of coupling beam length of coupling beam measured from face of wall piers embedment length of the steel coupling beams inside the wall piers distance between centroidal axes of the wall piers moment capacity of steel beam spacing of transverse reinforcement flange thickness of steel beam web thickness of steel beam thickness of wall pier shear stress factored shear stress factored shear force resisted by concrete nominal shear capacity of coupling beam plastic shear capacity of coupling beam factored shear capacity of coupling beam shear force resisted by reinforced concrete encasement factored shear force on section length of compression block at back of embedment length of compression block at front of embedment angle of inclination between diagonal reinforcement and longitudinal reinforcement coefficient defining relative contribution of concrete strength to wall strength ratio of the average concrete compressive strength to the maximum stress shear angle shape factor plastic interstory drift angle elastic interstory drift angle computed under code level lateral loads ratio of area of distributed reinforcement parallel to plane of shear resistance to the concrete area perpendicular to that plane material resistance factor for concrete = 0.60 strength reduction factor for steel = 0.90

REFERENCES CITED
American Concrete Institute (ACI) Committee 318, 1995, Building Code Requirements for Reinforced Concrete and Commentary (ACI 318-95/ACI 318R-95), Farmington Hills, MI, 353 pp. American Concrete Institute (ACI) Committee 318, 1999, Building Code Requirements for Reinforced Concrete and Commentary (ACI 318-99/ACI 318R-99), Farmington Hills, MI. American Institute of Steel Construction (AISC), 1995, Manual of Steel Construction, Load and Resistance Factor Design, Chicago, IL. American Institute of Steel Construction (AISC), 1997, Seismic Provisions for Structural Steel Buildings, Chicago, IL. Aktan, A. E. and Bertero, V. V., 1981, The seismic resistant design of R/C coupled structural walls, Report No. UCB/EERC-81/07, Earthquake Engineering Research Center, University of California, Berkeley. Aktan, A. E. and Bertero, V. V., 1984, Seismic response of R/C frame-wall structures, ASCE Journal of the Structural Division, 110 (ST 8), 1803-1821. Aktan, A. E. and Bertero, V. V., 1987, Evaluation of seismic response of RC building loaded to failure, ASCE Journal of the Structural Division, 113 (ST 5), 1092- 1108. Aristizabal-Ochoa, J. D., 1982, Dynamic response of coupled wall systems, ASCE Journal of the Structural Division, 108 (ST 8), 1846-1857. Aristizabal-Ochoa, J. D., 1987, Seismic behavior of slender coupled wall systems, ASCE Journal of the Structural Division, 113 (ST 10), 2221-2234. Barney, G. B. et al., 1978, Earthquake Resistant Structural Walls - Test of Coupling Beams, Report to NSF, submitted by Portland Cement Association, Research and Development, Skokie, IL. Bertero, V. V., 1980, Seismic behavior of R/C structural walls, Proceedings of the Seventh World Conference of Earthquake Engineering, Istanbul, Turkey, 1980. Bertero, V. V., et al., 1974, Hysteretic behavior of reinforced concrete flexural members with web reinforcement, Report No. UCB/EERC 74-9, Earthquake Engineering Research Center, University of California, Berkeley. Building Seismic Safety Council (BSSC) 1998, NEHRP Recommended Provisions for Seismic Regulations for New Buildings and Other Structures (FEMA 302) and Commentary (FEMA 303), 1997 edition, Washington, DC. Canadian Standards Association (CSA), 1994, CSA A23.3-94 Design of Concrete Structures, Rexdale, Canada. Fintel, M. and Ghosh, S. K., 1982, Case study of aseismic design of a 16-story coupled wall using inelastic dynamic analysis, ACI Journal, 79 (3), 171-179. Gong B., Shahrooz B. M., 1998, Seismic behavior and design of composite coupled wall systems, Report No. UC-CII 98/01, Cincinnati Infrastructure Institute, Cincinnati, OH. Gong B. Shahrooz B. M., and Gillum A. J., 1997, Seismic behavior and design of composite coupling beams, Composite Construction in Steel and Concrete III, ASCE, New York, 258-271. Gong B., Shahrooz B. M., and Gillum A. J., 1998, Cyclic response of composite coupling beams, ACI Special Publication 174 Hybrid and Composite Structures, Farmington Hills, MI, 89-112. Harries, K. A., 1995, Seismic Design and Retrofit of Coupled Walls Using Structural Steel, Ph.D. Thesis, McGill University, 229 pp. Harries, K. A., 1996, Design and analysis of four prototype eighteen storey coupled wall structures, McGill University Department of Civil Engineering and Applied Mechanics Structural Engineering Series Report No. 96-01, 122 pp.

Harries, K. A., 1998, Ductility and deformability of coupling beams in reinforced concrete coupled walls, Proceedings of the Eighth Canadian Conference on Earthquake Engineering, Vancouver, June 1999, 475-481. Harries, K. A., Mitchell, D., Redwood, R. G. and Cook, W. D., 1997, Seismic design of coupling beams - A case for mixed construction, Canadian Journal of Civil Engineering, 24(3), 448-459. Harries, K. A., Mitchell, D., Redwood, R. G. and Cook, W. D., 1998, Seismic design and analysis of prototype coupled wall structures, Canadian Journal of Civil Engineering, 25 (5), 808-818. International Council of Building Officials (ICBO), 1994, Uniform Building Code, Volume 2, Structural Engineering Design Provisions, Whitier, CA. Kabeyasawa T., Shiohara H., Otani S., and Aoyama H., 1983, Analysis of the full-scale seven-story reinforced concrete test structure, Journal of the Faculty of Engineering, University of Tokyo, XXXVII (2), 431-478. Kunnath, S. K. and Reinhorn, A. M., 1994, IDARC-2D Version 3.1: Inelastic Damage Analysis of Reinforced Concrete Building Structures, State University of New York at Buffalo. Lybas, J. M. and Sozen, M. A., 1977, Effect of beam strength and stiffness on dynamic behavior of reinforced concrete walls, SRS No. 444, University of Illinois, Urbana-Champaign. Mahin, S. A. and Bertero, V. V., 1975, An evaluation of some methods for predicting seismic behavior of reinforced concrete buildings, Report No. UCB/EERC 75-15, Earthquake Engineering Research Center, University of California, Berkeley. Marcakis K. and Mitchell D., 1980, Precast concrete connections with embedded steel members. Prestressed Concrete Institute Journal, 25 (4), 88-116. Mattock A. H. and Gaafar G. H., 1982, Strength of embedded steel sections as brackets. ACI Journal, 79 (2). Milev J. I. and Kabeyasawa T., 1996, Modeling of reinforced concrete shear walls in RC frame-wall structures, personal communication, Tokyo. New Zealand Standards Association (NZS), 1995, NZS 3101:1995 Concrete Structures Standard. Paparoni, M., 1972, Model studies of coupling beams, Proceedings of the International Conference on Tall Concrete and Masonry Buildings, 3, 671-681. Park, R. and Paulay, T., 1975 Reinforced Concrete Structures, John Wiley and Sons, Inc., New York. Paulay, T., 1971, Coupling beams of reinforced concrete shear walls, ASCE Journal of the Structural Division, 97 (ST 3), 843-862. Paulay, T., 1977, Coupling beams of reinforced concrete shear walls, Proceeding of a Workshop on Earthquake-Resistant Reinforced Concrete Building Construction, University of California, Berkeley, July 11-15, 1977, 1452-1462. Paulay, T., 1980, Earthquake resisting shear walls - New Zealand design trends, ACI Journal, 77 (3), 144-152. Paulay, T., 1986, The design of ductile reinforced concrete structural walls for earthquake resistance, Earthquake Spectra, 2 (4), 783-823. Paulay, T. and Binney, J. R., 1974, Diagonally reinforced concrete beams for shear walls, ACI Special Publication SP 42 - Shear in Reinforced Concrete, 579-598. Paulay, T. and Santhakumar, A. R., 1976 Ductile behavior of coupled shear walls, ASCE Journal of the Structural Division, 102 (ST 1), 93-108. Paulay, T. and Taylor, R. G., 1981, Slab coupling of earthquake-resisting shear walls, ACI Journal, 78 (2), 130-140. Precast/Prestressed Concrete Institute (PCI), 1992, PCI Design Handbook, fourth edition. Saatcioglu, M., Derecho, A. T. and Corley, W. G., 1987, Parametric study of earthquake-resistant coupled walls, ASCE Journal of the Structural Division, 113 (1), 141-157.

Scriber, C. F. and Wight, J. K., 1978, Delaying shear strength decay in reinforced concrete flexural members under large load reversals, Report UMEE 78 R2, Department of Civil Engineering, University of Michigan. Shahrooz B. M. and Gong, B., 1998, Steel-concrete coupling beams: a critical overview of design guidelines, Structural Engineers World Wide 1998, paper T186-3. Shahrooz B. M., Remmetter M. A. and Qin F., 1992, Seismic response of composite coupled walls, Composite Construction in Steel and Concrete II, ASCE, 429-441. Shahrooz B. M., Remmetter M. A. and Qin F., 1993, Seismic design and performance of composite coupled walls, ASCE Journal of the Structural Division, 119 (11), 3291-3309. Shiu, N. K., Barney, G. B., Fiorato, A. E. and Corley, W. G., 1978, Reversed load tests of reinforced concrete coupling beams, Proceedings of the Central American Conference on Earthquake Engineering, El Salvador, 239-249. Shiu, N. K., Barney, G. B., Fiorato, A. E. and Corley, W. G., 1981, Earthquake Resistant Walls Coupled Wall Test, Report to NSF submitted by Portland Cement Association, Research and Development, Skokie, Illinois. Shiu, N. K., Takayangi, T. and Corley, W.G., 1984, Seismic behavior of coupled wall systems, ASCE Journal of the Structural Division, 110 (ST 5), 1051-1066. Stafford-Smith, B. and Coull, A., 1991, Tall Building Structures, Wiley Interscience. Tassios, T. P., Moretti, M., and Bezas, A., 1996, On the coupling behavior and ductility of reinforced concrete coupling beams of shear walls, ACI Structural Journal 93 (6), 711-720. Teshigawara M., Kato M., Sugaya K.I., 1996, Seismic test on 12-story T-shaped coupled shear wall, Proceedings of the Third Joint Technical Coordinating Committee Meeting, Hong Kong, December 12-14, 1996. Tunc G., Deason J., and Shahrooz B. M., 2000, Seismic response of R.C. core wall-steel frame hybrid structures, Report No. UC-CII 00/01, Cincinnati Infrastructure Institute, Cincinnati, OH. Vulcano A. and Bertero V. V., 1987, Analytical models for predicting the lateral response of RC shear walls, Report No. UCB/EERC-87/19, Earthquake Engineering Research Center, University of California, Berkeley.