Characterizing frictional behavior for use in predicting the seismic response of unattached equipment S. Ray Chaudhuri, T.C. Hutchinson * Department of Civil and Environmental Engineering, University of California at Irvine 92697-2175, USA Accepted 11 November 2004 Abstract Characterizing friction is a fundamentally important aspect of modeling the seismic response of a body resting on any mechanical interface. In the context of rigid bodies mounted within building structures, the numerous types of elements and interfaces encountered in practice require specific characterization before any modeling assumptions can be undertaken. To this end, in this paper, the specific problem of characterizing the frictional behavior for a variety of small equipment types typically found in Biological and Chemical Science laboratories is studied using two different methods of testing. A simple approach is presented to determine the instantaneous coefficients of friction considering inertial effects. Resulting coefficients of static friction m s and kinetic friction m k for these types of equipment and their plausible range of uncertainty are presented. Analytical comparisons with shake table experiments, using mean frictional resistance values, illustrate that reasonable estimations of time history response can be determined. The accuracy of the prediction increases as the effects of stick-slip are minimized. q 2005 Elsevier Ltd. All rights reserved. Keywords: Nonstructural elements; Frictional behavior; Unattached equipment; Sliding response; Shake table tests 1. Introduction Unattached scientific equipment such as analyzers, microscopes, computer workstations, and other such elements, are highly vulnerable to damage due to strong seismic motions. Following recent earthquakes these items have gained significant attention, largely due to their potential for significant economic losses and/or research downtime. These equipment are costly and when found in hospitals or other critical buildings, their failure may hinder emergency response efforts immediately after an earth- quake. Often these equipment are placed on ceramic laboratory benches and have varied supporting base properties. Base conditions observed include metallic or hard plastic bases, or hard to soft rubber pads. The behavior of these equipment is observed to be sliding-rather than rocking-dominated, where two predominant stages describe the movement: the initiation of sliding and the continuous sliding [1]. The initiation of sliding occurs when the absolute acceleration of the supporting bench overcomes the resistance due to friction between the two surfaces of contact. Subsequent continuous sliding occurs until the relative velocity of the equipment with respect to the bench becomes zero. Considering this movement dominating the response, it is apparent that the sliding response of unattached equipment is very sensitive to both the coefficients of static and kinetic friction (m s and m k respectively) between the two surface of contact. This has also been observed by previous researchers through numerical simulations, however few systematic studies have been carried out to characterize the interface friction properties in order to predict seismic response. This is very important since the behavior of the equipment depends on the interface properties. From shake table experiments it is observed that depending upon the contact surface properties, the response of bench-mounted equipment may vary significantly. Moreover, a large variation of the coefficient of friction inherently exists between even similar equipment types. Soil Dynamics and Earthquake Engineering 25 (2005) 591–604 www.elsevier.com/locate/soildyn 0267-7261/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.soildyn.2004.11.022 * Corresponding author. Tel: C1 949 824 2166; fax: C1 949 824 2117. E-mail address: thutchin@uci.edu (T.C. Hutchinson).