Mayank Tripathi
Title: Buckling of reinforcing bars and its effects on seismic performance of slender reinforced concrete walls
Student: Mayank Tripathi
Senior Supervisor: Prof. Rajesh Dhakal
Co-Supervisor(s): Dr. Farhad Dashti and Dr. Leonardo Massone
Observations from past earthquakes in Chile (2010) and New Zealand (2010-11) have demonstrated that slender RC walls can be susceptible to brittle compression-dominated failure, rather than the desired ductile tension-governed response. Buckling of longitudinal reinforcing bars was a commonly observed failure mode that restricted the deformation capacity of RC walls. In addition to buckling, fracture of buckled reinforcing bars was also observed. In the current methodology for the design of anti-buckling transverse reinforcement, international design standards emphasize only on the spacing of transverse reinforcement as a critical parameter to restrict bar buckling in RC structures, presuming that the area of transverse reinforcement determined for the concrete confinement is adequate to restrict buckling. Furthermore, the probability of reinforcing bars to buckle while unloading from a large inelastic tensile strain is not incorporated in the design framework. Therefore, this research project has addressed these key issues pertaining to bar buckling in ductile RC walls.
Bar buckling causes reduction in the compressive stress carried by reinforcing bars. Therefore, to better understand and quantify the ill-effects of buckling on the hysteretic behaviour of reinforcing bars, monotonic and cyclic tests on reinforcing bars with different yield strengths and slenderness ratios are conducted in this study. In addition to bar buckling, fracturing of buckled reinforcing bars also adversely affects the seismic performance of RC structures. Therefore, the effect of buckling on the low-cycle fatigue life of reinforcing bars is also quantified in this study by conducting fatigue tests on bars with different yield strengths and slenderness ratios. The fatigue test data is utilised to develop fatigue life models relating the total-strain amplitude and total-dissipated energy with the fatigue life of reinforcing bars as a function of buckling parameter (Tripathi et al. 2018). Additionally, the experimental test data from monotonic, cyclic and fatigue tests on reinforcing bars are used to develop a path-dependent cyclic stress-strain model that incorporates the effect of bar buckling and low-cycle fatigue.
Next, the inadequacy of current design procedures to restrict bar buckling in the boundary elements of slender RC walls is evaluated by conducting tests on three large-scale code-compliant rectangular walls with different boundary zone transverse reinforcement detailing (Tripathi et al. 2019a). Two of the test specimens satisfied the anti-buckling transverse reinforcement requirement prescribed in NZS3101:2006, while in the third test specimen the arrangement of transverse reinforcement was modified to limit the buckling of boundary zone longitudinal reinforcing bars to single tie spacing. The test results reaffirm the susceptibility of RC walls designed according to the current New Zealand Concrete Standard to failure associated with bar buckling. In addition, the results from the test on the wall specimen with improved detailing indicate that bar buckling can be limited to single tie spacing by designing transverse reinforcement using a mechanics-based approach. The overall hysteretic response of RC walls can be improved by enhancing the local-compression response of wall boundaries. Therefore, in this research experimental and numerical studies are also carried out to scrutinise the effect of concrete compressive strength on the seismic response of slender RC walls (Tripathi et al. 2019b). The following figure displays the test setup and the different modes of bar buckling observed in this study.
Test setup used for experimental investigation on development of bar buckling in RC walls and the observed modes of bar buckling (Tripathi et al. 2019a)
The performance of slender RC walls under seismic excitation depends on the efficacy of the wall boundary elements to resist axial strain demands. Therefore, axial tests on rectangular prisms idealised as the boundary elements of slender RC walls are carried out in this study to investigate the effect of key parameters on performance of wall boundaries. The key parameters identified in this study include: (i) loading history; (ii) transverse reinforcement detailing (arrangement, spacing and diameter); (iii) longitudinal reinforcement detailing (yield strength and area); and (iv) cover concrete. A comparative evaluation of the test results is carried out and the effect of the identified parameters on the performance of the prism specimens is scrutinised. Additionally, the reliability of testing RC prisms to simulate the failure mechanisms in wall boundaries is also assessed by testing prism specimens identical to the boundary elements of the tested RC walls under axial strain histories measured at the wall boundaries. The wall and prism responses are then compared to evaluate the limitations of testing prism specimens to simulate failure mechanisms in wall boundaries.
Finally, the results from the tests on reinforcing bars, RC walls and RC prisms are consolidated and guidelines are proposed for the design of anti-buckling transverse reinforcement. The proposed design procedure recognises the importance of explicitly considering the tie leg stiffness on the buckling-resisting capacity of transverse reinforcement. Additionally, recommendations are proposed to limit the spacing of transverse reinforcement such that the compressive stress deterioration in longitudinal reinforcing bars is controlled until the limiting curvature prescribed in the NZS3101:2006 is achieved.