Presented are the features of a roof isolation system that is proposed as a
device to reduce the seismic response of buildings. Presented also are the
details of and results from analytical and experimental studies conducted
with a small-scale laboratory model to assess the feasibility and effective
ness of such a device. The roof isolation system entails the insertion of f
lexible laminated rubber bearings between a building's roof and the columns
that support this roof, and the installation of viscous dampers that are c
onnected to the roof and a structural element below the roof. It is based o
n the concept of a damped vibration absorber and on the idea of making the
roof, rubber bearings, and viscous dampers respectively constitute the mass
, spring, and dashpot of such an absorber. The model considered in the anal
ytical and experimental studies is a 2.44-m high, five-storey, moment-resis
ting steel frame, with a fundamental natural frequency of 2.0 Hz. In the ex
perimental study the frame is tested with and without the proposed roof iso
lation system on a pair of shaking tables under a truncated version of one
of the accelerograms from the 1985 Mexico City earthquake. In the analytica
l study, the frame is also analysed with and without such a system and unde
r the same ground motion except that the ground motion accelerations are pr
operly magnified to study the effectiveness of the roof isolation system wh
en the frame is stressed beyond its linear range of behavior. It is found t
hat the suggested device effectively reduces the seismic response of the fr
ame, although the extent of this reduction depends on how large its non-lin
ear deformations are. Based on these findings, it is concluded that the pro
posed roof isolation system has the potential to become a practical and eff
ective way to reduce earthquake damage in low- and medium-rise buildings. C
opyright (C) 1999 John Wiley & Sons, Ltd.