Experimental Evaluation of Seismic Performance of Bridge Restrainers

Abstract

The 1971 San Fernando Earthquake showed that many highway bridges are vulnerable to collapse due to excessive longitudinal movements at expansion joints and supports. The California Department of Transportation has since initiated a retrofit scheme, which includes tying bridge spans together with restrainers at their hinges. This experimental study investigates the impact between adjacent bridge spans at in-span hinges, evaluates the efficacy of restrainers in reducing relative displacements across the hinges and examines the effects of restrainer stiffness and gap on the response of the hinge-restrainer system. A specimen consisting of two box girder reinforced concrete blocks representing the adjacent spans of a reference bridge at in-span hinge was constructed and tested under incrementally increasing worst case scenario input excitation. The blocks were supported by elastomeric bearings simulating the stiffness of the substructure. Cable restrainers were used to connect the two blocks, the number and gap of which were varied during the experiments. Experimental results indicated that impacts between the adjacent frames produce acceleration levels significantly higher than what is typically assumed in design. Restrainers were capable of reducing hinge relative displacements and preventing span unseating. However, restrainer yielding occurred under strong input motions, especially when the restrainer gap was set to zero. Only pounding at the hinge location was included in this study, while pounding elsewhere in the superstructure as well as the detailed response behavior of substructure elements were neglected.