Behaviour of beam lap splices under seismic loading

The results of an investigation into the performance of reinforced concrete beam-column subassemblages containing lap spliced reinforcement in the potential plastic hinge region of a beam are presented. Two specimens were tested with simulated seismic loading. One specimen complied with the New Zealand Concrete Design Code, NZS 3101:1982, except for the placement of the lap splices. The second specimen contained beam reinforcement details from a building constructed in the early 1960s. Current concrete design codes specify lap splices should not be placed in beam potential plastic hinge regions where inelastic reversing stresses are possible during seismic events. During testing the transverse steel specified for the confinement of the lap splices was unable to prevent bond deterioration between the spliced bars once inelastic bar strains had developed at one end of the splice. The failure of the lap splices led to a loss of lateral load capacity and a low level of ductility from the specimen. Reinforced concrete buildings designed to pre-1970s codes may be considered inadequate when viewed in light of the provisions in current codes for seismic design. The testing of beam details taken from one such building indicates insufficient anchorage existed for the plain longitudinal beam bars in the joint. The loss of bond for the plain bars began in the initial load cycles of the test and led to a lack of specimen stiffness and lateral load capacity. The presence of the lap splices is considered to have accelerated the loss of bond from the bars. Testing investigating the performance available from plain bar reinforced subassemblages should use anchorage for the bars that represent the conditions in the existing structure. The rapid loss of bond from the bars during cyclic loading can lead to the member end connections influencing the test results