Full-Scale Testing of a Viscoelastic Coupling Damper for High-Rise Building Applications and Comparative Evaluation of Different Numerical Models

Abstract

Viscoelastic coupling dampers (VCDs) are used for the seismic and wind protection of tall buildings. In the past, several tests have been carried out on VCDs under severe loading conditions for concept validation purposes and different numerical models have been used for evaluating the performance of tall buildings with VCDs. Nonetheless, for practical applications understanding the performance of VCDs over a full range of demands including extremely small amplitudes of vibration, long-duration loading, and realistic well-defined design level seismic and wind events, as well as a comparative evaluation of different numerical models are needed. To address this research gap, a systematic study was carried out to understand the performance of VCDs based on full-scale tests and to assess the suitability of different numerical models in predicting their response. A range of displacement-controlled tests was carried out on a full-scale VCD specimen. Seismic loading was derived by considering both far-fault long-period long-duration ground motions as well as near-fault pulse-like ground motions that were developed based on a site-specific study and were scaled to represent the design earthquake (DE) and the risk-targeted maximum considered earthquake (MCER) levels for a real project. Long-duration wind loading of 6-h with 1-, 10-, 50-, and 500-year mean recurrence intervals (per Canadian practice) developed based on wind tunnel testing of another real project were used. The temperature rise in the specimen during the tests was measured using high-precision thermocouples embedded in the viscoelastic (VE) layers as well as an external thermal camera. The test results indicated well-defined force-deformation hystereses of the specimen at all levels of strain amplitudes including those at extremely small deformation amplitudes up to 2.5 μm of deformation. The temperature rise of the specimen was less than 1°C and 4°C, respectively for the earthquake and wind loadings representative of the real projects that were considered in this study. This temperature rise was found to be lower when compared with previous generations of VE materials, which were tested under loading representative of shorter buildings with higher fundamental frequencies. Finally, the accuracy of four different macroscopic numerical models with different degrees of complexities in simulating the test results was investigated. Different numerical models were found to be suitable for different loading conditions and recommendations are provided for practical nonlinear modeling of tall buildings with VE dampers.