No abstract

SUMMARYTwo interrelated issues related to the design of non-linear viscous dampers are considered in this paper: structural velocities and equivalent viscous damping. As the e!ectiveness of non-linear viscous dampers is highly dependent on operating velocities, it is important to have reliable estimates of the true velocity in the device. This should be based on the actual relative structural velocity and not the commonly misused spectral pseudo-velocity. This is because if spectral pseudo-velocities (PSV) are used, they are based on design displacements (S " S ) and are thus fundamentally di!erent from the actual relative structural velocity. This paper examines the di!erence between these two velocities, and based on an extensive study of historical earthquake motions proposes empirical relations that permit the designer to transform the well-known spectral pseudo-velocity to an actual relative structural velocity for use in design. Non-linear static analysis procedures recommended in current guidelines for the design of structural systems with supplement damping devices are based on converting rate-dependent device properties into equivalent viscous damping properties based on an equivalent energy consumption approach. Owing to the non-linear velocity dependence of supplemental devices, an alternative approach for converting energy dissipation into equivalent viscous damping is advanced in this paper that is based upon power consumption considerations. The concept of a normalized damper capacity ( ) is then introduced and a simple design procedure which incorporates power equivalent linear damping based on actual structural velocities is presented.

In this experimental study, elastomeric spring dampers, which have a distinct re-centering characteristic, are used to retrofit a non-ductile, previously damaged 1/3 scale model reinforced concrete building frame structure which is subjected to a variety of ground motions in shaking table tests. A velocity dependent analytical model is developed and verified for the elastomeric spring dampers. This model is implemented in the widely available non-linear dynamic time history analysis computer program DRAIN-2DX to produce response predictions which are in good agreement with experimental observations. The elastomeric spring damper devices significantly attenuate the seismic response of the structure and provide a considerable amount of energy dissipation while the main non-ductile reinforced concrete structural load carrying elements remain elastic. The effect of varying the damper configuration on the structural response was also investigated.

No abstract