Experimental Study on Seismic Performance of Reinforced Concrete Frame Columns Supported with Post-build Wing Walls

2.1.. Specimen Design and Fabrication

Six specimens were designed and manufactured. Axial compression was determined, and the minimum longitudinal reinforcement ratio corresponding to axial compression ratio and anti-seismic grade was determined [6]. The reinforcements in wing wall not only has effect of resisting shear force, but also has effect of connecting existing columns and form skeleton of mesh reinforcement. In addition, the horizon reinforcements should also meet the requirement of resisting shear strength between wing walls and existing column. Three axial compression ratio including 0.31, 0.36, 0.40 were designated for exiting frame columns. The ratio of longitudinal reinforcement was 1.23%, 1.05%, 0.82%, respectively, and stirrup ratio was 0.6%. The test basic parameters are shown in Fig. (1) Table 1.

The existing frame columns of the 6 strengthened specimens had cross-sectional dimensions of 250mm×250mm with a 20mm thick concrete cover. All of the columns had a height of 1200mm. The area of post-build wing walls is larger than the area of exist frame columns, thus the two phase back wing wall length is 350mm, the thickness of the post-build wing walls were all 100mm. There were beams respectively placed on the top and bottom of the specimens. The top beam was 1350mm×450mm×400mm, while the bottom beam was 1350mm×450mm×600mm. The standard cubic specimens (150mm×150mm×150mm) were measured to have a average compressive strength of 38.5MPa. The longitudinal reinforcements were HRB400 and the lateral reinforcements were HPB300. Table 2 presents the basic mechanical property indices of the reinforcements. The large-diameter longitudinal reinforcement of the original column was arranged in the corner.

The specimens were made in two phases. First, the concrete columns, top beams and bottom beams were produced. The columns were put on pressure by screw tension to consider the actual reinforce situation. Then, the post-build wing walls were placed on the columns under loading. The specimens were produced in two stages based on the method of horizontal concrete casting.

In the test, four externally pre-stressed tendons were used to anchor both the top and bottom beams in order to reinforce the concrete column under loading. The post-build wing walls were connected to the sides of the column after the vertical loading became stable, as shown in Fig. (2). Before the concrete of the wing walls was cast, the horizontal reinforcement was planted in the column, while the vertical reinforcement was planted in the top and bottom beams. The horizontal and vertical reinforcements had lengths of 15d.

2.2. Test Device and Loading System

The lateral reverse load was exerted by an MTS horizontal hydraulic servo-actuator fixed to the reaction wall. The loading point was located at the centre of the top beam through the vertical centroid line of the original column. A vertical load was applied by a hydraulic jack, which was installed on the distribution beam and remained unchanged during the test process. A rubber hinged support was placed between the vertical hydraulic jack and the specimen. The bottom of the specimen was fixed by vertical ground anchors and horizontal reaction frames on both sides of the channel flow. Fig. (1) shows the loading device, and Fig. (4) shows the force diagram of the specimen. The hydraulic jack was used to replace the four external pre-stressed reinforcements at the top of each specimen, which applied a predetermined axial force on the original column, in order to avoid them influencing the test results. The replacement process included applying an axial force from zero to the predetermined value by the hydraulic jack and unloading the resultant force of the four bars from the predetermined axial force to zero and then removing them. The horizontal loading tests complied with the loading regimes provided by the current standard [7] (Fig. 5). Linear variable differential transformers (LVDTs) (Fig. 3) were used as horizontal displacement sensors to measure the lateral displacements of relevant parts of the specimens during the loading process. The cracking load was defined as the point at which the specimen began to crack. As shown in Fig. (6), the yield load P_y was determined by the energy method. The ultimate load P_u was the peak load of the skeleton curve, and the displacement corresponding to 85% of the peak load in the descending stage was regarded as the ultimate displacement Δ_u.

3.1. Test Procedure and Phenomena

The direction of the horizontal load and the name and location of the wing wall are marked in Fig. (1)

The test results of the six specimens showed that the maximum tensile and compressive stresses and strains of the cross-section were all on the edges of the post-build wing walls because these walls were on the opposite faces of the existing columns (Table 3). The normal stresses and strains of the existing columns due to the horizontal load were not very prominent, the columns played a relatively large role in resisting the horizontal shear force. Not only was the seismic bearing capacity of the specimens clearly improved but also the damage to the original column under horizontal and vertical loads was relatively low because of the effective energy dissipation by the post-build wing walls on both sides.

Under low reverse cyclic loading, the wing walls experienced the process of cracking, cracking development, yielding of the longitudinal reinforcement close to the edge of the cross-section, and crushing of the concrete at the edge of compression zone. The existing columns were relatively lightly damaged. Thus, the post-build wing walls provided resistance against horizontal earthquake action, achieved effective energy consumption, and guaranteed relatively light damage to the original frame columns under vertical and horizontal loadings. Figs. (7 and 8) were shown the damage and cracks of the specimens.

3.2. Test Procedure and Phenomena

Hysteretic and skeleton curves of specimens under a horizontal load are both synthetic reflections of seismic behavior. Figs. (9 and 10) present the hysteretic and skeleton curves of each specimen.

1. When the horizontal load was less than the cracking load, the specimens were basically elastic. The load displacement curves rose along a straight line during loading, while the deformation was restored after unloading. As the load was increased, the load displacement curves gradually deviated from a straight line, and deformations accelerated so that residual deformations began to appear and reached a certain degree under unloading. The specimens began to assume elastoplastic properties. When the load was less than the yield load, the loading stiffness and unloading stiffness did not change significantly. When the load reached the yield load, the loading stiffness and unloading stiffness decreased with increasing load or lateral displacement.
2. The specimens showed better deformability with a low axial compression ratio than with a high axial compression ratio. After the maximum load was reached, the decline in the skeleton curve was relatively slow, which indicates that the ductility was relatively good. When the axial compression ratio of the columns was increased, the skeleton curves showed a relatively steep decline after the peak value, and the bearing capacity degraded relatively rapidly.

3.3. Experimental Results

The cracking load P_cr, the cracking displacement ∆_cr, the yield load P_y, the yield displacement ∆_y, the ultimate load P_u, the displacement corresponding to the ultimate load ∆_max, the ultimate displacement ∆_u, the load corresponding to the ultimate displacement P’_u, elastic plastic story drift angle θ_p and ductility factor μ of each specimen are shown in Table 4. In the current code, the limit values for the elastoplastic inter-story drift are 0.02 and 0.008 for a frame structure and shear wall structure [6], respectively. The data were between 0.021 to 0.033, which were all greater than the limits. The elastic plastic interstory drift index of reinforced concrete frame structures, structure failure was 0.016 [8], so the specimens were failure. The specimens ductility coefficient were satisfied with seismic fortification requirements in basically [9].

3.4. Stiffness Degradation

Under horizontal low-cyclic reverse loading, the deformations of the specimens developed constantly, and the stiffness gradually decreased with the increasing loads and displacements of the test pieces. The secant stiffness K_i can be used to represent the stiffness under a horizontal low-cyclic reverse loading:

where P_i represents the load under the control displacement and ∆i represents the displacement value corresponding to P_i. Fig. (11) shows the secant stiffness curves at different axial compression ratios of the specimens changing with the displacement. As the displacement increased, the secant stiffness of the test pieces gradually decreased.

3.5. Comparison Between Bearing Capacities Before and after Reinforcement

Before reinforcement, the ultimate horizontal load of each column was estimated according to the current code [10] by using the measured strengths of the concrete and steel. After reinforcement, the test values of the horizontal ultimate load adopted the ultimate loads of the skeleton curves under forward and reverse loadings.

Table 4 indicates that all of the specimens that showed large increases in the seismic capacity.