SUMMARYStiff, unattached structures are highly vulnerable to damage and failure during an earthquake, as evidenced following numerous past events. This class of structures encompasses a wide range of objects and systems such as electrical transformers, radiation shields, office furniture, and marble statues. The vulnerability of these objects is exacerbated when it is highly asymmetric and unattached. Although a number of studies have focused on rigid blocks, few have concentrated on blocks with asymmetric geometries. In an effort to better understand the implications of asymmetries, an extensive shake table testing campaign including more than 150 tests was conducted. These tests incorporate a systematic variation of the mass eccentricities of stiff, unattached structures. The primary modes of rocking, sliding, and twisting as well as interactive modes were recorded for the duration of numerous earthquake motions. The magnitude and direction of response are experimentally correlated with the geometric variations in the various models. These tests indicate that even for symmetric structures with uniaxial shaking, multiple modes and three-dimensional responses are probable. Furthermore, certain asymmetric geometries exhibited both increased rocking (and overturning) as well as increased sliding when compared with their symmetric counterparts. A final aspect of this study compared the free rocking response of symmetric and asymmetric structures to classical, two-dimensional rocking analysis. While the theoretical values for the coefficient of restitution yielded a significant overestimation in the simulation (up to ≈90%), reduced coefficients greatly improved the performance of the model.

No abstract

Summary Systems of unattached, or freestanding, structures are highly vulnerable to damage and/or collapse during an earthquake, as evidenced during numerous past events. This class of structural system includes statue–pedestal systems, multidrum columns, radiation shields, unreinforced masonry walls, and other mechanical and electrical equipment. While a number of studies have analyzed the response of the single rocking block, very few have tested the response of multiple block systems subjected to earthquakes. Therefore, this paper details an extensive shake table testing campaign in which the seismic response of a pair of stiff, unattached blocks, herein referred to as a dual‐body system, was evaluated. Experimental variables include the geometry, including asymmetry, of both top (tower) and bottom (pedestal) bodies, input motion, and the coefficient of friction beneath the system. Furthermore, the tower structures were tested both in dual‐body configurations as well as in single‐body configurations allowing an understanding of the effect of the pedestal. The tests indicate that the presence of a pedestal increases the likelihood of collapse and amplitude of rocking demands, in general. However, certain geometric and interface combinations yield a more stable tower in a dual‐body configuration compared to a single‐body configuration, because of the dependence of the pedestal response on the geometry of the tower. Furthermore, a low‐friction interface beneath the pedestal reduces demands on the tower. However, this low‐friction interface may still transfer long‐period contributions of the input motion to the tower, which may be detrimental to its response. Copyright © 2017 John Wiley & Sons, Ltd.

No abstract