Abstract | As proven by several past earthquakes, the seismic losses associated with nonstructural damage in contemporary buildings are likely to exceed those associated with structural damage by several orders of magnitude. Hence, for assessing satisfactorily the overall seismic performance of a building and consequently the associated losses, it is paramount to properly account for the nonstructural damage via estimating the acceleration and deformation demands that are imposed to its nonstructural elements and contents in any one floor level during an earthquake. Furthermore, being able to better understand the behavior of nonstructural components during a seismic event could have a direct impact on the pertinent design methodologies. To address such issues, we assess the validity of the dominant code/design approaches with reference to the evaluation of the component amplification factor, ap, which is essentially a measure of how much the acceleration of a component is amplified compared to the peak floor acceleration. The state-of-practice in the evaluation of ap is to account only for the component flexibility and period, an approach that essentially caps ap to a maximum value of 2.5. Instead, we investigate the extent to which this factor may be further amplified if we consider the vibration characteristics of the building that contains the component under consideration by employing actual recorded floor motions from instrumented buildings in the United States. At the same time, we evaluate the effect of yielding in the component or its anchorage in offering protection against such amplified acceleration demands.

Abstract | Nonstructural elements are typically at the end of a long chain of uncertainties making them particularly challenging to design or to estimate their seismic performance. Seismic design of nonstructural elements is typically based on over simplistic equations to estimate equivalent static forces that not only neglect the dynamic properties of the nonstructural element and of the building in which they are mounted on, but they also use large component force reduction factors that have no rational basis. Floor motions are characterized by being narrowband motions that produce extremely large accelerations on nonstructural elements whose frequencies of vibration coincide with those of the building in which they are mounted. For many years now, seismic design of structures relies on the identification of predetermined locations where non-linearities are expected to occur in the event of moderate or severe earthquake ground motions and other components and connections are designed to remain elastic. At present time, there is no equivalent approach in the design of nonstructural elements. The purpose of this paper is to introduce a new approach in which a new type of bracing elements of nonstructural elements are designed and detailed to work as fuses that limit forces acting not only in the nonstructural elements but also in the attachments to the structure and in the attachment(s) to the nonstructural element. It is shown that the proposed approach, which accounts for the narrow-band characteristics of floor motions, not only results in reductions in design forces but can also result in important reductions in deformation demands, especially for components that are tuned to modal frequencies of the building in which they are mounted on.

Abstract | Nonstructural components often represent the largest portion of the investment in new commercial buildings, typically accounting for 70 to 85% of the total construction cost. Furthermore, damage to nonstructural components can lead not only to serious injuries and even casualties, but can lead to partial or total temporary loss of use of the building and to large economic losses. Nonstructural components are often subjected to seismic motions that are completely different both in amplitude and in frequency content than ground motions. Floor motions are characterized by being narrowband motions that produce extremely large accelerations on nonstructural elements whose frequencies of vibration coincide with those of one of the modal frequencies of the building in which they are mounted. The purpose of this paper is to propose a new approach in which bracing elements of nonstructural elements are designed and detailed to work as fuses that limit forces acting not only in the nonstructural elements but also those acting on the attachments to the structure and on the attachment(s) to the nonstructural element. It is shown that the proposed approach not only results in reductions in design forces but can also result in important reductions in deformation demands, especially for components that are tuned to modal periods of the building in which they are mounted on.