Seismic response of beam-column joints rehabilitated with FRP sheets and buckling restrained braces

2.1. Test specimens and materials

Six one-half scaled beam-column specimens were manufactured in order to evaluate seismic performance. All specimens have the same cross-sectional dimensions. Geometry and reinforcement details are given Fig. 1; height of column is 1500 mm and length of beam is 2700 mm with cross sections of (300×400) mm and (400×350) mm respectively. 28 day compressive strength of concrete was measured to be 24 MPa. The yield strength of longitudinal bars with 19 mm diameter and stirrups with 10 mm diameter was 400 MPa. In this study, the BRB system proposed by Shin et al. [11] was used for the test. The yield strength of steel for BRB system was 235 MPa and two U-shape channels were attached to the both sides of a central core to prevent buckling of the core. Each U-shaped channel was welded using eight steel plates. To maximize the stiffness of the core, the core was designed to resist only against the axial force. For this purpose, rubber sheets were attached between the core and the U-shaped cannels, so that the steel core and the channels would behave independently under the load. Dimensions and details of BRB system are given in Fig. 2. Aramid FRP (AFRP) and carbon FRP (CFRP) sheets were used, respectively. The FRP sheets were wrapped in two layers on 60° with respect to the longitudinal axis of the column. Material properties of AFRP and CFRP sheets are presented in Table 1. Notations used for various test specimens are listed in Table 2.

Fig. 1Geometric properties and reinforcement details

a) Beam-column front view

b) Beam section

c) Column Section

d) Beam reinforcement placing details

Fig. 2Dimensions of the BRB

a) Section

c) Front view

b) Welding Plate

d) Top view

2.2. Test setup

The experimental set-up is schematically illustrated in Fig. 3. The bottom of the column was connected to the base by a hinge support and the beam was connected to the base at its ends by fixed supports. Sustained axial load was applied to the column using a 300 kN capacity hydraulic jack. The magnitude of the axial load was 10 % of axial capacity of the column and kept constant during the test. Cyclic lateral loading was applied to the top of the column using a 1,000 kN capacity hydraulic actuator by displacement control. The loading protocol consisted of three cycles as shown in Fig. 4.

Fig. 3Test set-up

Fig. 4Loading protocol

3.1. Failure modes of specimens

Fig. 5 shows the failure modes. For the control specimen, the flexural cracks occurred on the beam during the initial loading stages. Cracks then developed outside and shear cracks observed in the beam-column joints region, then shear cracks developed wider as the load increased. No cracks were observed in the column faces. Concrete was crushed near the joints and longitudinal bars of column were yielded after the maximum load. The failure patterns for BRB specimen was similar as control specimen at initial loading stages. However, shear cracks with $X$-shape in the beam-column joints progressed along the direction of the column unlike control specimen. This is because energy dissipation had taken place in the beam-column joint by the BRB. After the maximum load, concrete spalling appeared in the column.

Fig. 5Failure modes of the specimens

a) Control

b) BRB

c) CFRP

d) AFRP

e) CFRP + BRB

f) AFRP + BRB

For CFRP and AFRP specimens, flexural cracks occurred in the beam at initial loading stage. Then, shear cracks developed in beam-column joints. As the load increased, concrete spalling occurred in the beam-column joints. For CFRP-BRB and AFRP-BRB specimens, shear cracks were observed in the joint then the cracks occurred adjacent to BRB. And rupture of FRP sheets was taken place. Both CFRP-BRB and AFRP-BRB specimens were failed by the pull-out of the anchors in the BRB. Also minor shear cracks in the joint were found.

3.2. Strength and stiffness

The load-displacement hysteresis curves of specimens are presented in Figs. 6 and 7. Also, the test results are summarized in Table 3. The maximum load capacity of the CFRP and AFRP specimens were increased by approximately 10 % and 9 % compared with the control specimen. Regardless of elasticity modulus and tensile strength of FRPs, the test results of CFRP and AFRP specimens showed similar load capacity in terms of maximum load. The maximum load capacities of BRB, CFRP-BRB, AFRP-BRB specimens were significantly higher than the control specimen. This is due to the increased strength of joints resulting from BRB system. However, for BRB specimen, the load capacity dropped dramatically after the maximum load. CFRP-BRB and AFRP-BRB specimens showed more ductile behaviors than that of BRB specimen. Test result indicated that the wrapping of FRP sheets effectively contributed to ductile behavior of beam-column joint.

Fig. 6Relationship of lateral load versus displacement

a) Control

b) BRB

c) CFRP

d) AFRP

e) CFRP+BRB

f) AFRP+BRB

To evaluate seismic performance of different rehabilitation methods, effective stiffness of the specimens are plotted in Fig. 8 and summarized in Table 4. The effective stiffness is defined as the slope of the line joining the highest load points in the load versus displacement curve attained at each displacement level [12]. It can be obviously seen that the initial stiffness of all specimens were almost the same except the control and AFRP specimens. The effective stiffness of BRB, AFRP-BRB and CFRP-BRB specimens increased by approximately 70 % compared with the control specimen. This result indicates that BRBs contributes to the effective stiffness of beam-column joints. The control and AFRP specimens showed similar effective stiffness while CFRP specimen exhibited an increase in effective stiffness.

Fig. 7Hysteresis curves for the specimens

Fig. 8Effective stiffness of the specimens

3.3. Energy dissipation and displacement ductility

Energy dissipation of specimens under cyclic loading has been calculated by summing enclosed area within the load versus displacement hysteresis curves at different loading cycles. The cumulative energy dissipation is plotted in Fig. 9 and summarized in Table 5. AFRP and CFRP specimens were increased by approximately 11 % and 15 % when compared to the control specimen, respectively. The energy dissipation capacity was significantly improved in the case of retrofitted using BRB system compared to the control, CFRP and AFRP specimens. For BRB, CFRP-BRB and AFRP-BRB specimen had almost same energy dissipation capacity at the initial loading stage. However, since BRB specimen was failed at the second loading cycle, it represents the lowest energy dissipation capacity. Specimens retrofitted FRP sheets and BRBs indicated the highest energy dissipation capacity.

Fig. 9Cumulative energy dissipation of the specimens

In this paper, the displacement ductility was computed by the following equation:

where $D$ is the ductility index, the ultimate displacement, ${\delta }_{ult}$, is defined as the displacement corresponding to a 80 % strength degradation of the maximum strength, ${\delta }_{yield}$, is defined as the displacement corresponding to the first yielding of the specimen. Fig. 10 and Table 7 present the displacement ductility index of the specimens. As shown in Fig. 10, all specimens exhibit higher displacement ductility index than the control specimen except BRB specimen. BRB specimen exhibited brittle failure at the anchorage of BRB to concrete, and the displacement remained relatively low at 27 mm. Hence, BRB specimen has less ductility than that of the specimens retrofitted with FRP sheets and BRB. The specimens retrofitted with FRP sheets increased approximately 24 % in average.

Fig. 10Displacement ductility index