Seismic retrofitting of RC beam-column joints using composite materials

Abstract

Many of the existing reinforced concrete structures built prior to 1999 have deficient design details due to their non-seismic design or construction flaws. These structures are at risk of collapse under severe earthquakes. Especially, the beam-column joints would experience high shear forces during such events due to transverse reinforcements and lap splice inadequate design details resulting in damage or failure of structures. To enhance the seismic performance of such joint systems with insufficient reinforcement details, several strengthening techniques have been proposed by various researchers. In this study, new strengthening techniques were developed using carbon fiber reinforced polymers (CFRP) to retrofit the deficient beam-column joints and the effects were investigated on the specimens having structural deficiencies. One control specimen was constructed according to detailing provisions specified by the 1975 design code of Turkey, whereas the other four control specimens were constructed with deficiencies observed in the real applications. Moreover, three additional specimens were constructed to develop an alternative strengthening technique. The tests were carried out by applying a constant axial load and a reversed cyclic lateral load to the top of the column. The tested control specimens were repaired and/or strengthened by CFRP sheets and retested. Comparative analysis of control and CFRP-strengthened joint specimens’ results showed that, significant improvements in the lateral load and the energy dissipation capacities were achieved by using the proposed CFRP-strengthening techniques. Furthermore, an analytical model was proposed to predict the lateral load capacities of the shear critical joint specimens. The predicted results were compared with the experimental results of this research. All the predictions were in good agreement with the experimental results. On the other hand, the proposed model was found to reproduce 11 test results from the literature with reasonable accuracy.