Experimental Investigation on Cyclic Behavior of Reinforced Concrete Coupling Beams Under Quasi-static Loading


The current status of the reinforcement layouts stated in the codes and frequently used in new construction areas needs to be evaluated and determining its effects on the behavior of structural elements may provide essential contributions to the field. For this purpose, conventionally reinforced, diagonally reinforced (confinement of individual diagonals),and diagonally reinforced with full confined section concrete coupling beam specimens were designed according to ACI 318-14 and Turkish Earthquake Code 2018 in this study. ½-scale test specimens with aspect ratio two were tested under quasi-static cyclic loading using a new complex experimental setup. According to the results obtained, while for conventionally reinforced coupling beams ductile hysteretic behavior remained at 2.0% drift, for diagonal reinforced coupling beams it sustained up to 4.0% drift. Although the diagonal reinforcement enhanced ductility capacity of coupling beams compared to conventionally reinforced coupling beams having a comparably low post-elastic behavior, high ductility demands in the code requirements did not meet. For the same reinforcement ratio and layout, individual confinement around diagonal bars retarded bar buckling and ultimate strength and ductility capacity increased slightly compared to diagonally reinforced coupling beam with full confined section. After all, bar buckling that occurred after the life safety performance level given in the codes was seen at 3.5–4.0% drift. Diagonal reinforcing bars ensured life safety performance level up to 4.0% drift, but it remained at 3.0% drift for conventional reinforcement.


High-rise buildings are important technological innovations among the buildings constructed since the beginning of the twentieth century. In high-rise buildings, as the number of stories and height increase, the sensitivity increases to horizontal loads caused by natural events such as earthquakes and winds. The most functional bearing element for the preferred lateral load-bearing systems in high-rise buildings is reinforced concrete shear walls carrying both horizontal and vertical loads. In some cases, shear walls that are built side by side on the same axis are connected to each other by a coupling element such as floor and beam. Coupling elements with high shear strength should be used to compensate for the loss of stiffness and strength caused by the opening and to improve the interaction of walls placed side-by-side on the same axis. This frame-like system is called as reinforced concrete coupled shear wall system. In the reinforced concrete coupled shear wall system, single shear wall piers are often connected to each other by short and deep coupling beams.

The energy entering the system with the effect of lateral load can be distributed to the coupling beams along with the height of the building instead of concentrating only on the shear wall piers. To ensure obtaining the optimum performance from the system, the energy dissipation mechanism must be provided by the plastic hinge mechanism at the base of the shear walls and in all of the coupling beams. In order to achieve the expected performance from coupled shear walls, the coupling beams must have acceptable strength, rigidity, and ductility. Additionally, the hysteretic behavior of coupling beams significantly changes and improves the behavior of coupled shear wall systems. The only way to understand the behavior of reinforced concrete coupled shear walls is to fully understand the hysteretic behavior of the coupling beams [1].

In order to study the effect of coupling beams between the coupled shear walls systems, very large test platforms are needed. Because of inadequate laboratory conditions and difficulties in the construction of a multi-stories shear wall, researchers idealize the system. For this reason, many studies have been focused on improving the behavior of the coupling beams. As all researchers in the field know, the improvement of the coupling beams, which play a dominant role in the behavior of coupled shear walls, would improve the shear wall piers at the same rate.

In the 1964 Alaska earthquake, it was determined that the conventionally reinforced concrete coupling beams suffered major damages that could not be recovered and repaired. Observed damages demonstrated the vulnerability of conventionally reinforced coupling beams. Moreover, the studies on the conventionally coupling beams carried out after the earthquake revealed that the coupling beams were subjected to large cyclic shear deformations during the earthquake [2, 3]. Especially, conventionally coupling beams with small clear span to depth ratios (ln/h) behaved as brittle under large shear forces and sliding shear failure occurred at the beam ends of conventionally reinforced coupling beams. Transverse reinforcement and stirrups did not prevent shear failure and sliding shear failure [2]. Moreover, the experimental ductility of conventionally coupling beams is significantly lower than the theoretical ductility [4]. To improve the performance and ductility capacity of deep coupling beams, diagonal reinforcement was suggested by Paulay and Binney [4]. They found that diagonal reinforcement prevented the sliding shear failure of the coupling beam and ensured superior ductility and energy dissipation capacity. This beam achieved a larger fraction of their theoretical strength capacity, but the cause of failure in the diagonally reinforced concrete beam was buckling of the compression steel bars. Buckling of steel bars may be suppressed with transverse reinforcement enclosing diagonal bars.

If the aspect ratio of the coupling beam is greater than four, the efficiency of the diagonal reinforcement decreases as the inclination angle and cross bracing truss action of the diagonal reinforcement decrease and the behavior of coupling beams is dominated by flexure [5]. However, diagonal reinforcement has encountered construction difficulties with the placement of the diagonal bars and transverse reinforcement enclosing entire structure of the beam and diagonal bars. Barney et al. [6] introduced the diagonal reinforcement in the hinging region of the coupling beam to improve the performance of short-span beams. Diagonal bars were placed to eliminate the sliding shear and diagonal reinforcement was designed to provide an internal truss system to move the location of critical plastic hinge away from the face of the wall. Since the concrete within the region of the diagonals deteriorated by spalling and crushing, diagonal truss softened and efficient truss action could not be developed. Therefore, diagonal bars were not effective to carry the shear forces. Tassios et al. [7] investigated the performance of conventionally reinforced beam, diagonally reinforced beam, conventional beams containing long dowels and short dowels at the ends, and conventional beam with additional bent-up bars intersecting at the mid-height of the beam. Bent-up bars increased the ultimate capacity of the conventional reinforced beams by 30 percent and improved the overall behavior of the conventionally reinforced beams. Coupling beams reinforced with short and long dowel bars did not exhibit sliding shear at the beam ends but showed the most brittle behavior. Although the poor behavior of the conventional reinforced beam improved, they could not reach the performance of the diagonally reinforced beam. Galano and Vignoli [8] conducted tests on the coupling beams consisted of four different reinforcement configurations (conventional, diagonal with and without confining and rhombic) with an aspect ratio of 1.5. Although rhombic layout produced lower strength, it displayed the highest rotational ductility. A comparably high energy dissipation capacity was achieved with the rhombic layout. That study demonstrated that the rhombic layout is easier to fabricate than the diagonal layout involving confinement ties. Kwan and Zhao [9] carried out an experimental study including six experimental elements with different reinforcement layouts and reinforcement ratios to determine the behavior of conventionally reinforced beams preferred nowadays, although there are alternative reinforcement layouts. It was found that diagonally reinforced coupling beams did not provide a great contribution in terms of ultimate displacement capacity or deformability although energy consumption capacity and ductility levels are high. Fortney et al. [10] investigated the effect of confinement bars around individual diagonals on the performance of diagonally reinforced coupling beams. Even though it is more difficult in terms of construction difficulties, it is concluded that increasing the confinement bars play an active role in the behavior of diagonally reinforced coupling beams as it delays the damage of core concrete and buckling of diagonal reinforcement. Brena and Ihtiyar [11] studied the behavior of conventionally reinforced coupling beams with different aspect ratios and different shear-bending reinforcement ratios. The results showed that diagonal cracks continued from corner to corner in deep beams of aspect ratio although diagonal cracks are concentrated near the beam ends in large-span beams. They also found that the sliding shear displacement was quite high for beams with high aspect ratio and low bending reinforcement ratio. Han et al. [12] proposed a bundled diagonal reinforcement as the diagonal reinforcement is difficult to fabricate on site although it allows desirable inelastic damages during an earthquake. Bundled reinforcement provided the desired performance to allow the arrangement of diagonal reinforcement in long-span beams with high aspect ratio and simplified construction on the site.

As a new trend of literature with regard to the coupling beams, fuse or damper coupling beams have been especially tested by some researchers in the numerical and experimental studies instead of the reinforced concrete coupling beams [13,14,15]. In addition, the behavior of coupled shear walls has been investigated analytically and experimentally depending on the developments related to coupling beams [16,17,18]. The fact that the behavior of reinforced concrete coupling beams designed according to the current code requirements is known contributes to the development of such studies.

Three different reinforcement layouts are proposed in ACI 318-14 [19] and the Turkish Earthquake Code 2018 [20] for reinforced concrete coupling beams. The contributions of all reinforcement layouts on the coupling beams in the above-mentioned studies were not analyzed in the same studies. Additionally, there were no studies examining the damage level on shear wall piers and coupling beams with critical sections depending on the current codes, and only the coupling beam performance was examined. In this study, conventionally reinforced, diagonally reinforced (confinement of individual diagonals) and diagonally reinforced (with full confined section) concrete coupling beam specimens were designed according to ACI 318-14 and Turkish Earthquake Code 2018. ½-scale coupling beam specimens with an aspect ratio of two produced in the laboratory were tested under quasi-static cyclic loading using a new experimental setup which displays the behavior of coupling beams and beam–wall interactions. The aspect ratio of two and factored shear stress of 0.33 \(\sqrt {f_{c}^{\prime } }\) were considered as critical parameters of design and section. The results of experimental tests were discussed in terms of their compatibility with literature studies. The test setup and behavior of specimens designed according to ACI 318-14 and the Turkish Earthquake Code 2018 were evaluated.

Test Setup, Instrumentation and Loading Protocol

As seen in the published papers, different test setups, especially such as direct and indirect load paths are considered for reinforced concrete coupling beams and coupled shear walls [2, 6,7,8, 11, 13, 21,22,23,24,25]. Although the existing test setups used in the studies do not have any disadvantages in terms of material and reinforcement arrangement, they may have some disadvantages regarding the determination of ductility, strength, rotation, displacement, and local deformations at the beam–wall joints and end blocks (wall piers); therefore, the behavior of coupled shear wall systems cannot be demonstrated completely in the relevant studies [23].

In this study, a new experimental setup was designed to provide equal end rotations, double curvature bending, and local deformations at beam–wall joints. Base and the top surface of wall piers were connected to pin assemblies. Although base pin assemblies were fixed on the floor, top pin assemblies were fixed on a stiff loading beam. If the lateral loads of the wall piers vary due to the tributary inertial mass, large axial forces may develop in the beam, but the floor slabs and in-plane stiffness of the coupled shear walls may produce axial restraint [26]. The axial restraint was compensated with the stiff loading beam. In the designed test setup, lateral loads are transferred to the loading beam connected to the wall piers with pin assemblies using a 100-kN actuator. Hence, half of the lateral load transferred to the loading beam is directly applied to the wall piers by means of each pin support as a shear force. The pin assemblies were connected to wall piers with steel bolted connection plates having a thickness of 30 mm. Steel bolts were produced by threading the ends of longitudinal reinforcement bars in the laboratory. This was applied to reduce local stresses on the shear wall piers. Instead of extra bolts for connecting, longitudinal reinforcement bars of shear walls to top and bottom plates were used to satisfy code requirements and also to reflect the local deformations at beam-wall joints. The fundamental difference of designed setup from other indirect load path setups is that stress intensity on connection point or zone between pin supports and reinforced concrete wall piers were reduced. Stress intensity on local joints was minimized or distributed on wall piers using steel bolts produced by threading the ends of longitudinal reinforcement bars. Moreover, the setup optimized equal rotations of wall piers and it may be used for the development of new coupling or link elements (damper coupling beams, prefabricated coupling beams, etc.). The distance from the base pin to the top pin was chosen as 1600 mm. The general view of the setup is given in Fig. 1.

Fig. 1

Test setup: a installation of instruments and b fabrication in laboratory

Quasi-static displacement-controlled cyclic loading history prescribed in Chapter 2.9.1 of FEMA-461 [27] was used for the tests. In FEMA-461, the targeted smallest displacement of the loading history is recommended as 0.0015H (H is the distance from base pin to top pin) if no data exist regarding what amplitude of deformation is likely to initiate damage. The initial displacement was chosen as 1.20 mm to reduce the effects of connection gaps in the setup for all specimens. The amplitude of the steps in the loading history was 1.40 times the amplitude of preceding steps and each step was applied in two cycles according to FEMA-461. The Quasi-static displacement-controlled cyclic loading history used in the tests was given in Fig. 2. In this study, shear wall piers were designed according to code requirements, but the vertical load on shear wall piers was neglected.

Fig. 2

Quasi-static displacement-controlled cyclic loading history

Linear Potentiometric Displacement Transducers (LPDTs) were used to measure the displacements. An LPDT (LPDT-1) was placed horizontally in direction of movement of the loading beam to measure the lateral displacement of the loading beam. For sliding shear, two LPDTs (LPDT-2 and LPDT-3) were installed vertically at the beam–wall joints. The load corresponding to the displacement at each loading step was determined by the loadcell connected to the actuator.

Specimen Geometry and Reinforcement Layout

If coupling beams have an aspect ratio of (ln/h) ≥ 2 and factored shear force Vu exceeding 0.33 \(\sqrt{{f}_{c}^{^{\prime}}}\) Acw (Acw = area of concrete section; \({f}_{c}^{^{\prime}}\)=specified compressive strength of concrete), coupling beams can be reinforced with conventional reinforcement according to ACI 318–14. As stated in the Turkish Earthquake Building Code 2018, factored shear force Vu will not be higher than 1.5bhfct (b = width of beam section; h = depth of beam section; fct = tensile strength of concrete) for conventional reinforcement. Therefore, the aspect ratio of two and factored shear stress of 0.33 \(\sqrt{{f}_{c}^{^{\prime}}}\) are critical values for shear intensity in beam sections and the application of conventional reinforcement for these codes.

To evaluate the performance of RC coupling beams designed according to ACI 318–14, three ½-scale coupling beam specimens were tested. The aspect ratio of two and factored shear stress of 0.33 \(\sqrt{{f}_{c}^{^{\prime}}}\) were considered as design parameters (contribution of the concrete and transverse reinforcement bars was ignored). In order to compare the results of the experimental studies on coupling beams, the longitudinal reinforcement ratio and transverse reinforcement ratio were taken into account at a similar rate for all specimens. Since diagonal reinforcement arrangement is required for coupling beams according to codes, the amount of longitudinal reinforcement was determined by the calculation of coupling beams with diagonal reinforcement. Dimensions and reinforcement details of coupling beams aare illustrated in Fig. 3.

Fig. 3

Dimensions and reinforcement details of coupling beam specimens: a conventional reinforcement (CRCB); b diagonal reinforcement with confinement of individual diagonals (DRCB), and c diagonal reinforcement with full confined section (DRCB-FC)

The span and depth of the coupling beams were 900 mm and 450 mm, respectively; therefore, the aspect ratio was chosen as two. D14 bar with 14 mm diameter was selected for longitudinal or diagonal reinforcements. The length of the D14 bars threaded at two ends used in shear wall piers is 1500 mm. As a code requirement, additional horizontal reinforcement of 12 mm-diameter (D12) along the depth was used in spacing 80 mm for all specimens. The spacing of transverse reinforcement of 8 mm (D8) was 80 mm because it shall not exceed six times the diagonal bar diameter according to ACI 318-14.

Specimens were tested to assess the effects of the design made under code requirements on the behavior. Test specimens were designed using concrete with a 28-day nominal compressive strength (\({f}_{c}^{^{\prime}})\) of 30 MPa and steel bars with a nominal yield strength (\({f}_{y}\)) of 420 MPa. The inclination angle of diagonal reinforcement was determined as 18.66° for diagonally reinforced specimens according to concrete cover and transverse reinforcement. The summary of specimen geometry and measured material properties is given in Table 1. Three 150 × 300 mm cylinder specimens fabricated and cured at the same conditions were tested at 28 days age and the average strength of cylinder specimens were used for the estimation of the strength of specimens. The compressive stress–strain curves of concrete using in specimens for 28 days aging is shown in Fig. 4a. Reinforcement steel bars came from the same bundle for all specimens and three tension tests were conducted to determine the strength of each steel bar size. Tensile stress–strain curves of steel bars used in all specimens are shown in Fig. 4b. The mechanical properties of steel bars used in the specimens are given in Table 2.

Table 1 Coupling beam specimens and concrete
Fig. 4

Stress–strain curves of concrete and steel bars: a concrete and b steel

Table 2 Mechanical properties of steel bars

In the design process, it is desirable that the coupled shear walls meet adequate stiffness and ductility demands. Coupling beams were the main elements determining the ductility of the coupled shear walls. The idealization of the structural interaction between the coupling beams and the wall piers in the laboratory depends not only on the appropriate test setup but also on the selection of the appropriate dimensions of beam and wall piers. Even though providing equal end rotations of coupling beams was one of the main features of the test setup, optimum stiffness and local wall deformation at the beam-wall joint should also be obtained from the test setup. Kwan and Zhao [23] suggested that the wall piers should be designed to have a width greater than the beam depth and a height greater than two times the beam depth. Therefore, the wall piers in our study had 1350 mm height and 750 mm width. The thickness of the wall piers was 200 mm. Regarding the wall piers, diameters of the vertical reinforcement and horizontal reinforcement were preferred as 14 mm and 8 mm, respectively. Vertical reinforcement bars were threaded at the ends to use as steel bolts. The pin assemblies were connected to wall piers with using threaded bars. These threaded vertical bars satisfied the tensile forces and shear forces transferred from pin assemblies to steel bolted connection plates. Transverse reinforcement of 8 mm was located in spacing 70 mm in the wall piers to provide a factored shear stress value Vu approximately 0.55 \(\sqrt{{f}_{c}^{^{\prime}}}\).

Test Results

Cracking and Failure Modes

Crack development and propagation on the specimens are shown in Figs. 5, 6, 7 depending on the drift ratio (θ) defined as a lateral displacement (δ) of the loading beam divided by the distance between the top and bottom pins (1610 mm). For Specimen CRCB (conventionally reinforced coupling beam specimen), vertical flexural cracks occurred at the corners of the interface between coupling beam and wall piers during a loading step 0.25%, but diagonal tension cracks with an increased inclination developed after 0.5% drift level in Fig. 5a. Diagonal tension crack caused the diagonal splitting failure of the beam was observed and first shear cracks on shear wall piers were seen at 0.75% drift level in Fig. 5b. Although shear cracks propagated from the beam–wall interface to pin supports and the centerline of piers, diagonal shear and vertical flexural cracks on the beam and connection intensified at the maximum shear crack width of 0.28 mm and drift level of 1.0% (Fig. 5c). Flexural cracks on tension corners and diagonal cracks perpendicular to tension axis in the beam reached up to 1.2 mm width at 1.5% drift as seen in Fig. 5d. CRCB specimen reached up to ultimate strength at 1.8% drift and lost about 35% of its strength around 2.0% drift. Figure 5e depicted that CRCB specimen were subjected to diagonal splitting failure and it did not achieve the 2.0% drift level. At this stress level, diagonal crack width grew up to about 4.0 mm and concrete in compression end zones of the beam were partially spalled and crushed.

Fig. 5

Failure modes of CRCB specimen

Fig. 6

Failure modes of the DRCB specimen

Fig. 7

Failure modes of the DRCB-FC specimen

For Specimen DRCB (diagonally reinforced coupling beam specimen), flexural cracks were observed at corners of the coupling beam during a loading step of 0.20%. Although corner flexural cracks formed in the fifth loading cycle sequence for CRCB specimen, the first vertical cracks appeared in the fourth loading cycle sequence for DRCB. After vertical cracks propagated towards to center of the beam, the first diagonal tension cracks developed with an increased inclination after a 0.5% drift level in Fig. 6a then, the first shear cracks on shear wall piers were seen at 0.75% drift level at the tensile loading direction in Fig. 6b. At 1.0% drift level, diagonal tension cracks were quite concentrated on the shear walls and beam center; however, the width of these cracks increased to 0.1 mm (Fig. 6c). Vertical cracks forming at the beam–wall interface reached up to 1.2 mm width and they continued up to central axis of the beam at 1.5% drift as seen in Fig. 6d. In the second loading cycle of this loading step, interface cracks peaked at approximately 1.4 mm width and slightly spalling of concrete was seen on the tension corners of the beam. At 2.0% drift level, vertical cracks on beam span expanded up to 0.4 mm and minor diagonal cracks (0.3–0.4 mm) on the beam were elongated, and cracks on walls were concentrated within a distance of about the depth of the beam as seen in Fig. 6e. Crack width and spalling of concrete on the corners under tension and compression became apparent. As seen in Fig. 6f, when lateral drift ratio was 3.0%, the diagonal cracks on the beam center were elongated to the corners of the beam and flexural cracks on the interfaces and through the beam span extended up to 3.6 mm but width of the diagonal cracks on beam span peaked up to 0.7 mm. Figure 6g exhibits that DRCB specimen could reach up to 4.0% drift level and it had deformations associated with flexural and shear failure. A large fraction of corner concrete was crushed and spalled under compression and tension. Widths of diagonal cracks that elongated to corners of the beam reached up to 5.4 mm.

Diagonal reinforcement with a full confined section was first involved in ACI 318-08 [28] and the Turkish Earthquake Code 2018 [20]. In comparison to diagonally reinforcement (confinement of individual diagonals), confinement and transverse reinforcement were used in full section instead of in individual diagonals. Although the contribution of two type diagonal reinforcement layouts is theoretically accepted as similar, the contribution of this reinforcement layout to failure mode and structural behavior of coupling beams should be investigated. For this purpose, the DRCB and DRCB-FC specimens examined in our study showed similarity in terms of failure mode. At a loading step of 0.20% drift, the first vertical cracks were observed at the tension corners under bending. Unlike the CRCB and DRCB specimens, flexural cracks formed in the third loading cycle sequence for the DRCB-FC specimen. Although flexural cracks increased towards to beam center, after that, diagonal tension cracks were seen close to the beam center at 0.5% drift level in Fig. 7a and the first shear cracks on shear wall piers were seen at 0.75% drift level along the tensile loading direction. As seen in DRCB specimen, diagonal tension cracks were quite concentrated on the shear walls and beam center but the width of these cracks increased up to 0.15 mm at 1.0% drift level (Fig. 7b). As displayed in Fig. 7c, vertical cracks formed at interface reached up to 1.4 mm width and continued central axis of the beam at 1.5% drift but unlike the DRCB, the flexural crack growths on the interface were not fast and deep. In the second loading cycle of this loading step, interface cracks peaked at approximately 1.7 mm width but concrete spalling was not seen on the tension corners. At 2.0% drift level, slightly crushing of concrete was seen on the compression corner and vertical cracks on interface expanded up to 2.4 mm and minor diagonal cracks (0.35–0.4 mm) throughout the beam span were elongated. Cracks on walls concentrated within a distance of about a depth of the beam as seen in Fig. 7d. The diagonal cracks on the beam center elongated to the corners of the beam and flexural cracks on the interfaces and through the beam span extended up to 3.9 mm as displayed in Fig. 7e, but the width of the diagonal cracks throughout the beam span peaked up to 0.6–0.8 mm. As can be seen from Fig. 6f, DRCB-FC specimen could reach up to a 3.5% drift level and it had deformations associated with flexural and shear failure. Similar to DRCB specimen, a large fraction of corner concrete was crushed and spalled under compression and tension. Widths of diagonal crack elongated to corners of the beam reached up to 6 mm.

Minor hairline cracking of walls must be less than 0.16 mm for immediate occupancy but coupling beam experience cracking must be less than 0.32 mm according to ASCE/SEI 41-06 [29]. ASCE/SEI 41-17 [30] considers only diagonal crack development on the shear wall and coupling beam. For the life safety performance level, coupling beams had extensive shear and flexural cracks; some crushing, but the concrete generally remains in place according to ASCE/SEI 41-06 and ASCE/SEI 41-17. However, some boundary element stress, including limited buckling of reinforcement, some sliding at joints, some crushing, and flexural cracking are allowed for shear walls. In our study, crack width was about 0.15 mm at the maximum load level (attained for DRCBC specimen). Crushing and spalling of concrete on the wall piers were not observed and all damages were collected on the coupling beam span and beam-wall interface. Except for CRCB specimen, crack widths were maximum 0.4 mm throughout beam span up to 2.0% drift level despite extensive crushing of corner concrete. CRCB, DRCB and DRCB-FC specimens ensured life safety conditions up to 3.0, 3.5 and 4.0% drift level, respectively. Experimental observations pointed out that designed shear wall piers did not display significant crack development and damage. Considerations put forward by Kwan and Zhao [23] about the design of two wall piers were supported with this study. Considering the specimen dimensions, cracks, and damages on the shear walls can be monitored during the cyclic loading. Coupling beam-wall interaction and coupling action can be exhibited in accordance with the considerations.

Strength degradation for CRCB specimen started at a loading level in which the specimen was subjected to diagonal tension failure (diagonal splitting) as seen in Fig. 8a. Under cyclic loading sequence applied after the diagonal tension failure, crushing and spalling of concrete occurred in beam midspan of CRCB specimen but no crack development was observed on the beam-wall interface. Ultimate damage development did not depend on the rupture and buckling of bars for CRCB specimen. For DRCB and DRCB-FC specimens, reinforcing diagonal bars buckled following the crushing of the compression corner concrete and expansion of vertical tension cracks in tension corners. Bar buckling occurred at about 3.5% drift level for DRCB-FC specimen despite the 4.0% drift level of the DRCB specimen. Although buckling occurred after a 4.0% drift for DRCB specimen, the first signs of rupture of reinforcing bars were observed up to about 5% drift as displayed in Fig. 8b, c.

Fig. 8

The condition of specimens after failure stage: a CRCB specimen; b DRCB specimen; c DRCB-FC specimen

Load–Displacement Responses

Load–displacement responses of specimens under lateral cyclic loading are shown in Fig. 9a–c. According to structural analysis and considering the geometry and boundary conditions of the experimental setup, coupling beams and wall piers were respectively subjected to shear loads which are 0.982 and 0.50 times of the lateral load applied throughout the central axis of stiff loading beam by the actuator. FEMA-273 [31] limited chord rotation (θb) of coupling beams in such a way that maximum allowable chord rotations are 0.025 and 0.030 for conventionally and diagonally reinforced coupling beams, respectively. Chord rotation values of coupling beams were computed considering the collapse mechanism or lateral behavior of coupled systems using Eq. 1:

$$ \theta_{b} = \theta_{w} \left( {\frac{{L_{w} }}{{L_{b} }}} \right) $$

where Lw is the distance between the central axis of wall piers, Lb is the coupling beam length, θw is the wall pier rotation, and θb is the chord rotation of the coupling beam.

Fig. 9

Load–displacement relationship: a CRCB specimen; b DRCB specimen; c DRCB-FC specimen

Many researchers defined three different ductility measurements such as static ductility, energy ductility, and cumulative ductility [7, 8]. Static ductility (µ) was estimated using Eq. 2:

$$ \mu \, = \frac{{\theta_{u} }}{{\theta_{y} }} $$

where θy is the drift ratio at yielding and θu is the maximum drift ratio. According to FEMA 306 [32], ductility capacity is classified as low for µ < 2, moderate for 2 ≤ µ ≤ 5, and high for 5 < µ. Drift ratio at yielding (θy) corresponds to an intersection point of a line drawn from the origin to a point of 3/4 the ultimate load and horizontal line at the ultimate strength [8]. The maximum drift ratio (θu) corresponds to a point when strength is reduced to 0.85 times the ultimate strength as seen in Fig. 10.

Fig. 10

Definition of ductility on load–displacement envelope curve [7, 8, 30]

In this study, factored shear stress 0.33 \(\sqrt{{f}_{c}^{\mathrm{^{\prime}}}}\) was considered as a design parameter (contribution of the concrete and transverse reinforcement bars was ignored in design), but transverse reinforcement bars and concrete contributed to the shear strength. Therefore, the technique in ACI 318-14 (Chapter 18) was specifically used for coupling beams instead of shear walls. Shear strength of a conventionally reinforced coupling beam subjected to diagonal tension (Vu) was determined using Eq. 3 in ACI 318-14 (Equations in Chapter 18):

$$ V_{u} = A_{cv} \left( {\alpha_{c} \sqrt {f_{c}^{\prime } } + \rho f_{y} } \right) $$

where Acv = gross-sectional area of coupling beam; αc = 0.17 (using 2 for inch) for an aspect ratio > 2; ρ = transverse reinforcement content. For diagonal reinforced concrete coupling beams, the contribution of diagonal reinforced bars was added as a shear strength 0.33 \(\sqrt{{f}_{c}^{\mathrm{^{\prime}}}}\) (design criteria for this study) to Eq. 3 in the calculation of theoretical \(\mathrm{s}\) hear strength in diagonal tension (\({V}_{u}^{*})\).

Bilinear idealized pushover curves were given from load–displacement envelopes as seen in Fig. 11. Ultimate strength (Fu), drift ratio at yielding (θy), maximum drift ratio (θu), and static ductility (µ) for all specimens were given in Table 3.

Fig. 11

Bilinear idealized response curves: a CRCB specimen; b DRCB specimen; c DRCB-FC specimen

Table 3 Summary of the calculated and experimental results

Diagonally reinforced coupling beam specimens, DRCB and DRCB-FC exhibited the expected stable hysteretic load–displacement behavior up to 4.0% and 3.50% drift levels, respectively. However, conventionally reinforced coupling beam specimen CRCB showed brittle behavior and sudden loss of strength. Pinching effect defined as stiffness increase due to close of concrete cracks under compressive forces were seen specifically on CRCB specimen in comparison with DRCB and DRCB-FC specimens, but intense pinching effects were observed in many studies [6, 7, 11] for conventionally reinforced coupling beams. Diagonally reinforced coupling beam specimen, DRCB clearly displayed more ductile and stable behavior compared to DRCB-FC specimen since buckling of the diagonal bar in diagonal compression axis of beam occurred earlier than diagonal bars of DRCB specimen. Ultimate loads for DRCB and DRCB-FC specimens were obtained at 2.90% and 2.65% drift levels, respectively. CRCB specimen reached the ultimate load at a drift of 1.80%. The ultimate load for CRCB dramatically dropped at 2.11% drift. It was proved that CRCB specimen was very brittle. Strength degradation for DRCB and DRCB-FC was clearly seen at 4.10% and 3.72%, respectively. Due to diagonal bar buckling and crushing of corner concrete, the load-bearing capacity of beams dropped suddenly. While DRCB specimen sustained approximately 62% of the ultimate load at 4.10% drift, DRCB-FC specimen sustained approximately 66% of it at a 3.72% drift. As per FEMA-461, the ratio of initial displacement amplitude to ultimate displacement amplitude shall be equal to or less than 0.05. This ratio was obtained as 0.037 for CRCB which had a minimal ultimate displacement amplitude.

Strength and Stiffness Degradation

Load–displacement envelopes were obtained from peak load and displacement at the end of the first cycle of every loading step. Load–displacement envelopes for all specimens are given in Fig. 12. DRCB specimen achieved greater peak load at ultimate displacement. On the other hand, CRCB specimen had the least ultimate strength. Although material properties and design acceptations were the same, confinement of individual diagonals mainly affected the ultimate strength. Confinement of individual diagonals increased the strength of about 11%. At the push and pull directions, ultimate strengths achieved as approximately equal for DRCB specimen but the ultimate strengths of CRCB and DRCB decreased about 10% and 12% at pull cycles. At the first loading step following the maximum loading step, which caused the peak load, beam ultimate strengths dropped 65, 30, and 42% for CRCB, DRCB, and DRCB-FC specimens, respectively.

Fig. 12

Load–displacement envelopes

Stiffness degradation-versus-displacement curves were illustrated in Fig. 13. Secant stiffness (normalized stiffness) is very useful in the assessment of stiffness degradation for coupling beams. In this study, secant stiffness was determined separately by dividing stiffness in every cycle to the stiffness in the first cycle for push and pull directions as seen in Fig. 13. Stiffness of CRCB specimen was rapidly degraded up to 1.50% drift. At this drift level, stiffness degradation of the specimen was reduced up to about 50% for push and pull loading cycles. At 2.0% drift level which is the permissible value for structures according to general code requirements, CRCB, DRCB, and DRCB-FC specimens lost stiffness by 42%, 38%, and 15%, respectively. DRCB specimen exhibited stable behavior and stiffness degradation gradually up to a 3.0% drift. It was found that DRCB displayed great ductility but CRCB specimen were too brittle.

Fig. 13

Normalized stiffness-displacement relationship

Energy Dissipation

Energy dissipation capacity is one of the most important criteria for understanding the performance of structural elements. As a result of experimental studies, the cumulative energy dissipation-lateral displacement curve obtained by the cumulative sum of the areas covered by each cycle in the lateral load-lateral displacement curves are given in Fig. 14. While CRCB specimen achieved ultimate strength before 2.0%, all experimental specimens showed similar energy dissipation at permissible drift level of 2.0%. Since DRCB had a greater drift capacity compared to DRCB-FC, DRCB dissipated approximately 25% energy. Although DRCB and DRCB-FC specimens displayed elastic behavior in 2.0% drift, they dissipated more energy than CRCB specimen. CRCB had low energy dissipation capacity as expected. However, although CRCB has behaved post-elastic behavior before DRCB and DRCB-FC, it did not reach the energy capacity of both DRCB and DRCB-FC. Additionally, DRCB did not show better energy dissipation than DRCB-FC, since DRCB-FC behaved post-elastic behavior before DRCB.

Fig. 14

Cumulative energy dissipation

Sliding Shear Behavior

Sliding shear failure at the interface between beam and wall cannot be adequately prevented and eliminated. This failure mode causes inadequate ductility and therefore, ductility demand is not satisfied for coupled shear walls. For sliding shear measurements, 2 LPDTs were used vertically at the beam-wall joints. It is quite difficult to take realistic measurements in this interface because of the crushing and spalling of concrete. In our study, LPDTs were stiffly connected to lower steel bolted connection plates since LPDTs move with shear wall piers. When LPDTs connect parallelly to shear wall piers, little movement of coupling beams at the interface can be read with LPDTs as a negative or positive value according to the loading direction. To obtain LPDT measurements precisely, a flat glass support was placed between the LPDT and bottom surface of the beam so as not to be affected by the beam end rotations.

Vertical displacements were measured and recorded during the loading cycles. Change in measured vertical displacements depending on the lateral displacement is shown in Fig. 15 for all specimens. As seen from the cracking and failure pattern of CRCB specimen, flexural or vertical cracks on the beam-wall interface were not expended or developed throughout the interface. Vertical displacement was measured as about 1.40 mm at 2.0% drift level which CRCB specimen were failed and lost its strength. However, DRCB and DRCB specimens had higher sliding displacements compared to CRCB specimen. DRCB and DRCB-FC specimens moved vertically 1.90 mm and 2.32 mm at 2.0% drift, respectively. At this drift level, crack width and spalling of concrete on the corners under tension and compression were seen clearly. Sliding displacements increased by 3.90 mm and 3.55 mm for DRCB and DRCB-FC specimens, respectively. After a 3.0% drift level, measurements became unstable due to crushing and spalling of corner concrete.

Fig. 15

Sliding shear displacement

Brena and Ihtiyar [11] stated that shear-sliding behavior would take place in coupling beams as the aspect ratio and shear strength throughout the coupling beam span increases. In this study, shear reinforcement on the span of conventionally reinforced coupling beam had not met high shear demand and subjected to diagonal splitting failure. As shown in Figs. 5, 6, 7, visible cracks appeared at a 2.0% drift level. Since high shear strength demands were met by diagonal reinforced bars on the beam span, cracks and spalling of the concrete were concentrated on the corners under tension and compression were became apparent for DRCB and DRCB-FC specimens at 2.0% drift level but any spalling and crack width on the corners had not seen for CRCB specimen. Therefore, DRCB and DRCB-FC specimens behaved in sliding shear compared to CRCB specimen. The sliding mechanisms exhibited by Brena and Ihtiyar [11] for conventionally coupling beams were presented for diagonally reinforced coupling beams under higher shear demands.


In this study, conventionally reinforced, diagonally reinforced (confinement of individual diagonals) and diagonally reinforced (with full confined section) concrete coupling beam specimens were designed according to ACI 318-14 and the Turkish Earthquake Code 2018. The specimens were fabricated in the laboratory using a new complex experimental setup which was designed to represent the behavior of coupling beams in RC coupled shear walls. The coupling beam specimens were tested under quasi-static cyclic loading. Three ½-scale coupling beam specimens were tested. The aspect ratio of two and factored shear stress of 0.33 \(\sqrt {f_{c}^{\prime } }\) were considered as design parameters. The behavior of specimens designed according to ACI 318-14 and Turkish Earthquake Code 2018 under given design parameters was evaluated. Following results and recommendations are obtained:

  • Our experimental setup provided equal local deformations at wall piers and beam-wall joints. The crack pattern on wall piers showed that the length and height of wall piers must be at least two and three times of beam depth, respectively. The proposed and designed test setup had presented real-like wall-beam interaction for crack development on coupling beam and shear walls.

  • Threading the end of longitudinal reinforcement bars had eliminated the need for additional bolts in shear wall piers to investigate and satisfy code requirements.

  • The crack development on shear wall piers designed according to the code requirements remained in immediate occupancy. While diagonally reinforced coupling beams ensured life safety conditions up to 4.0% drift level, it remained at 3.0% for conventionally reinforced coupling beams. The coupling beams and shear walls which were designed according to ACI 318-14 and Turkish Earthquake Code 2018 ensured high structural performance demand in ASCE/SEI 41-17.

  • Coupling beams designed according to ACI 318-14 and Turkish Earthquake Code 2018 presented a satisfactory performance in terms of theoretical diagonal shear strength. However, further research should be conducted to verify the use of Eq. 3 given in ACI 318-14 for conventionally coupling beams which have different aspect ratios diagonal tension.

  • Ultimate shear stress values of the coupling beams obtained from cyclic tests were 0.73 \(\sqrt {f_{c}^{\prime } }\), 1.18 \(\sqrt {f_{c}^{\prime } }\), and 1.07 \(\sqrt {f_{c}^{\prime } }\). Although allowable shear stresses for coupling beams in ACI 318–14 and Turkish Earthquake Code 2018 are 0.83 \(\sqrt {f_{c}^{\prime } }\) and 0.85 \(\sqrt {f_{c}^{\prime } }\), respectively, diagonally reinforced coupling beams withstood higher stress.

  • Diagonal reinforcement in beams with an aspect ratio of two enhanced the ductility capacity compared to conventionally reinforced coupling beams, but high ductility demand stated in code requirements has not been met.

  • Coupling beams that were designed with an aspect ratio of two according to minimal code requirements exhibited good performance for beam chord rotations in allowable code requirements.

  • Diagonal reinforcing bars in every diagonally reinforced coupling beam specimen buckled but reinforcing bars in diagonally reinforced coupling beam specimens with full confined section buckled at earlier drift levels. Confinement around diagonal bars retarded bar buckling and enhanced ultimate drift capacity. Especially, diagonal bars buckled (diagonal compression) or ruptured (diagonal tension) after exceeding the life safety performance level given in the codes.

  • As a result of post-elastic behavior, the energy dissipation of specimens was similar up to 2.0% drift but the energy dissipation of diagonally reinforced beams increased considerably after 2.0% drift. In addition, diagonally reinforced (with full confined section) concrete coupling beam specimens exhibit a better energy dissipation capacity than diagonally reinforced (confinement of individual diagonals) concrete coupling beam specimen as a consequence of the limitations of our study. If more specimens are tested for different aspect ratios, more satisfactory results can be obtained.

  • Sliding shear displacement was observed in all specimens but it was lower than expected for conventionally reinforced coupling beam. Diagonal tension failure occurred in earlier drift decreased sliding shear displacement for conventionally reinforced coupling beams. As a result of meeting high shear demand throughout beam opening, cracks and spalling of the concrete were especially concentrated on the corners under diagonal tension and diagonal compression. Therefore, sliding shear was measured in diagonal reinforced coupling beams after a 2.0% drift ratio.

Further research is needed to satisfy minimal chord rotation capacity and minimize instantaneous brittle failure of conventionally coupling beams by using additional confinement and transverse bars. Moreover, the produce of diagonal reinforcement layout can be simplified by new reinforcement designs instead of individual confinement around diagonal bars. This study highlights the general behavior of coupling beams with an aspect ratio of two for minimal current code requirements;s however, future studies should be extended to include coupling beams with different aspect ratios in the future.


  1. 1.

    Sesli H (2018) Investigation of coupling beam behavior in reinforced concrete coupled shear walls and recommendation of novel mechanical systems. Ph.D. Dissertation, Karadeniz Technical University

  2. 2.

    Paulay T (1969) The coupling of shear walls. Ph.D. Dissertation, University of Canterbury

  3. 3.

    Paulay T (1971) Coupling beams of reinforced concrete shear walls. J Struct Div 97(3):843–862

    Google Scholar 

  4. 4.

    Paulay T, Binney JR (1974) Diagonally reinforced coupling beams of shear walls. ACI Spec Publ 42:579–598

    Google Scholar 

  5. 5.

    Harries KA, Gong B, Shahrooz BM (2000) Behavior and design of reinforced concrete, steel and steel-concrete coupling beams. Earthq Spectra 16(4):775–799. https://doi.org/10.1193/1.1586139

    Article  Google Scholar 

  6. 6.

    Barney GB, Shiu KN, Rabbat BG, Fiorato AE, Russel HG, Corley WG (1980) Behavior of coupling beams under load reversals. Portland Cement Association

  7. 7.

    Tassios TP, Moretti M, Bezas A (1996) On the behavior and ductility of reinforced concrete coupling beams of shear walls. ACI Struct J 93(6):711–719. https://doi.org/10.14359/518

    Article  Google Scholar 

  8. 8.

    Galano L, Vignoli A (2000) Seismic behavior of short coupling beams with different reinforcement layouts. ACI Struct J 97(6):876–885. https://doi.org/10.14359/9633

    Article  Google Scholar 

  9. 9.

    Kwan AKH, Zhao ZZ (2002) Cyclic behavior of deep reinforced concrete coupling beams. Struct Build 152(3):283–293. https://doi.org/10.1680/stbu.

    Article  Google Scholar 

  10. 10.

    Fortney PJ, Rassati GA, Shahrooz BM (2008) Investigation on effect on transverse reinforcement on performance of diagonally reinforced coupling beams. ACI Struct J 105(6):781–788. https://doi.org/10.14359/20106

    Article  Google Scholar 

  11. 11.

    Brena SF, Ihtiyar O (2011) Performance of conventionally reinforced coupling beams subjected to cycling loading. J Struct Eng 137(6):665–676. https://doi.org/10.1061/(ASCE)ST.1943-541X.0000316

    Article  Google Scholar 

  12. 12.

    Han SW, Lee CS, Shin M (2015) Cyclic performance of precast coupling beams with bundled diagonal reinforcement. Eng Struct 93:142–151. https://doi.org/10.1016/j.engstruct.2015.03.034

    Article  Google Scholar 

  13. 13.

    Montgomery M (2011) Fork configuration dampers for enhanced dynamic performance of high-rise buildings. Ph.D. Dissertation, University of Toronto

  14. 14.

    Ahn T-S, Kim Y-J, Kim S-D (2013) Large-scale testing of coupled shear wall structures with damping devices. Adv Struct Eng 16(11):1943–1955. https://doi.org/10.1260/13694332.16.11.1943

    Article  Google Scholar 

  15. 15.

    Zarei D, Tasnimi AA (2019) New fused steel-coupling beam with optimized shear panel damper. Int J Civ Eng 17:1513–1526. https://doi.org/10.1007/s40999-019-00430-5

    Article  Google Scholar 

  16. 16.

    Lequesne RD, Parra-Montesinos GJ, Wight JK (2013) Seismic behavior and detailing of high-performance fiber-reinforced concrete coupling beams and coupled wall systems. J Struct Eng ASCE 139:1362–1370. https://doi.org/10.1061/(ASCE)ST.1943-541X.0000687

    Article  Google Scholar 

  17. 17.

    Cheng MY, Fikri R, Chen CC (2015) Experimental study of reinforced concrete and hybrid coupled shear wall systems. Eng Struct 82:214–225. https://doi.org/10.1016/j.engstruct.2014.10.039

    Article  Google Scholar 

  18. 18.

    Hung C, Lu W (2018) Tall hybrid coupled structural walls: seismic behavior and design suggestions. Int J Civ Eng 16:567–582. https://doi.org/10.1007/s40999-017-0162-2

    Article  Google Scholar 

  19. 19.

    ACI Committee 318 (2014) Building code requirements for structural concrete (ACI 318-14) and commentary on building code requirements for structural concrete (ACI 318R-14). Farmington Hills

  20. 20.

    Disaster and Emergency Management Presidency (2018) Turkish Building Earthquake Code (TBEC 2018). Ankara

  21. 21.

    Igarashi H, Izumita S, Takeuchi M, Arai M, Fujisawa M, Murakami H (1992) Deformation capacity of reinforced concrete beams used height strength concrete and reinforcement. In: Proceedings of the 10th World Conference on Earthquake Engineering

  22. 22.

    Harries KA, Cook WD, Rewood RG, Mitchell D (1993) Concrete walls coupled by ductile steel link beams. In: Proceedings of Earthquake Engineering Tenth World Conference, pp 3205–3210.

  23. 23.

    Kwan AKH, Zhao ZZ (2002) Testing of coupling beams with equal end rotations maintained and local joint deformation allowed. Struct Build 152(1):67–78. https://doi.org/10.1680/stbu.

    Article  Google Scholar 

  24. 24.

    Canbolat BA, Parra-Montesinos GJ, Wight JK (2005) Experimental study on seismic behavior of high-performance fiber-reinforced cement composite coupling beams. ACI Struct J 102:159–166. https://doi.org/10.14359/13541

    Article  Google Scholar 

  25. 25.

    Kuang JS, Baczkowski BJ (2006) Shear capacity of steel fibre reinforced concrete coupling beams. In: Proceedings of International Conference on Computing and Decision Making in Civil and Building Engineering

  26. 26.

    Barbachyn SM, Kurama YC, Novak LC (2011) Analytical evaluation of diagonally reinforced concrete coupling beams under lateral loads. ACI Struct J 109(4):497–507. https://doi.org/10.14359/51683869

    Article  Google Scholar 

  27. 27.

    Federal Emergency Management Agency (2007) Interim testing protocols for determining the seismic performance characteristics of structural and nonstructural components (FEMA 461). Washington

  28. 28.

    ACI Committee 318 (2008) Building code requirements for structural concrete (ACI 318-08) and commentary. Farmington Hills

  29. 29.

    ASCE (2006) Seismic rehabilitation of existing buildings (ASCE/SEI 41-06). Virginia

  30. 30.

    ASCE (2017) Seismic Evaluation and Retrofit of Existing Buildings (ASCE/SEI 41-17). Virginia

  31. 31.

    Building Seismic Safety Council (1997) NEHRP Guidelines for the Seismic Rehabilitation of Buildings (FEMA 273). Washington

  32. 32.

    Federal Emergency Management Agency (1998) Evaluation of earthquake damaged concrete and masonry wall buildings (FEMA 306). Washington

Download references


This research with project No. FBA-2017-7037 was supported by Scientific Research Projects Unit of Karadeniz Technical University.

Author information



Corresponding author

Correspondence to Hasan Sesli.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sesli, H., Husem, M. Experimental Investigation on Cyclic Behavior of Reinforced Concrete Coupling Beams Under Quasi-static Loading. Int J Civ Eng (2020). https://doi.org/10.1007/s40999-020-00573-w

Download citation


  • Coupling beams
  • Coupled shear walls
  • Quasi-static cyclic loading
  • Failure mode
  • Aspect ratio