RC beam-column joints seismically retrofitted with selective beam weakening and local FRP strengthening

Abstract

The capacity design method has been widely accepted in the design of reinforced concrete (RC) frames to ensure that they have sufficient plastic deformation capacities when subjected to a seismic attack. The key of this method is to make columns stronger than the connected beams at a joint, thus realizing the so-called Strong-Column-Weak-Beam (SCWB) hierarchy. However, many existing RC frames, especially those designed according to outdated design codes; do not meet the SCWB principle. Inadequate consideration of the contribution of a cast-in-place slab to the hogging moment capacity of a beam is a common cause. Conventional seismic retrofit methods have limited effects in enhancing the seismic safety of these RC frames. Against the above background, this thesis presents a systematic study into a new seismic retrofit method that involves beam weakening and FRP (fibre reinforced polymer) strengthening (the BWFS method). This method was proposed by the author's research group. In the experimental part of the study, three different beam weakening techniques were studied by testing 9 full-scale RC beam-column joints (under a combination of constant axial loading and cyclic lateral loading), including: (a) the slab slit (SS) technique, in which a transverse slit is cut in the slab at each beam end; (b) the beam web opening (BO) technique, in which an opening is cut in the beam web; and (c) the beam section reduction (SR) technique, in which a deep transverse groove is cut on the soffit of the beam near the joint. The test results show that (a) the SR technique reduces both the strength and ductility of the specimen; (b) the BO technique leads to a ductile failure mode if the opening size is sufficiently large; and (c) the SS technique has a small negative effect on specimen ductility although it can effectively reduce the beam strength. A combination of the SS and the BO techniques is shown to be an efficient retrofit method in reducing the strength of a T-beam and enhancing the ductility of the joint. In the finite element (FE) analysis part, accurate three-dimensional (3D) FE models for predicting the behaviour of T-beams with an opening and FRP strengthening were developed. 3D FE models using either solid or shell elements were both developed. The static analysis problem was regarded as a dynamic problem and solved using the explicit centre-difference-method (CDM). A few significant issues, such as the loading time, the damping scheme, the computational time and the accuracy associated with the explicit dynamic method, are discussed in depth. The 3D FE model built using shell elements was then applied to study the issue of effective slab width of T-beams in a hogging moment zone. Based on the results of parametric studies, new models for the effective slab widths of T-beams of both interior and exterior beam-column joints were proposed. Finally, 3D FE models for retrofitted RC joints were developed and substantiated with test results; the verified FE model can be used in further investigations of such seismically retrofitted RC beam-column joints.