Predicting failure in sliding isolation bearings under long-period motions

For the majority of structural systems, there has been a push to understand collapse behaviour and to quantify collapse margin ratios. However, for seismic isolation there is still significant work to be done in this area, especially for sliding bearings, for which little research on failure has been investigated. While it may be the goal of the designer that isolation bearing capacity is not reached, for performance based design it is essential to understand how and under what levels the bearing will fail. To investigate failure, a model based on the theory of rigid body kinematics, rigid body dynamics and contact mechanics is employed with an added parallel non-linear damper to explicitly consider the energy dissipation. Ricker pulses are extracted from long period motions and used to predict whether the bearing impacts and if that impact results in failure. The research finds that these pulses are good predictors providing that the extracted pulse periods are sufficiently long (greater than 0.5 s); otherwise the pulse does not dominate the response of the bearing. Generalized graphs are produced for use in predicting bearing performance and collapse margin ratio (assuming a rigid superstructure) at initial stages of design.