Equivalent structural damping of drift-sensitive nonstructural building components

The consideration of nonstructural damage in seismic engineering has increased over recent decades as a direct result of two observations: 1) modern seismic design provisions have made great strides in preserving life safety as a primary objective, and 2) the resulting loss and consequences from nonstructural damage can be tremendous, even in well-designed structures. This problem has been addressed in design by applying some form of serviceability considerations to limit displacements and has been a main concern within numerous advancements of performance-based earthquake engineering. However, the treatment of the equivalent elastic damping provided by nonstructural components has, for the most part, been kept at widely accepted rule of thumb approximations (e.g. 5% damping). Although often assumed in practice, there is little evidence to suggest that this is appropriate in all cases, particularly at intensity levels relevant to the serviceability limit state. This study develops a procedure to quantify the equivalent structural damping offered by drift-sensitive nonstructural components in order to gauge how nonstructural inventory may affect seismic response. It is proposed that the damping values obtained could be used together with an elastic model of the nonstructural components to provide improved seismic response predictions. The work focuses on cold-formed steel framed gypsum partition walls and external façade systems, which are analyzed using mechanical models based on recent test results. To illustrate the procedure, an eight-story steel moment frame building designed according to Eurocode 8 with various nonstructural typologies is considered. The effects of the supplemental damping provided by the nonstructural components on the seismic performance of the building under serviceability limit ground motions are illustrated and are shown to be quite variable, in contrast to the assumptions typically made in practice. Furthermore, the simplified linear analysis approach is seen to perform well.