“Numerical modeling of a strategic timber building in L'Aquila with and without passive base isolation.”

Light-frame construction is used extensively in the European, American and Australasian market for low and medium rise timber buildings. These buildings are light-weight and with high dissipative capacity due to the ductile behaviour of the shear walls, which are made of plywood or other types of sheathing (OSB, gypsum board, etc.) nailed on light timber frames [2]. The system is therefore very convenient for use in earthquake-prone areas as the behaviour factor to use in design may be fairly high (up to 5 according to the Eurocode 8). On the other hand the use of high behaviour factors may imply significant structural and non-structural damage at the end of the design earthquake, leading to potentially high economical losses, also due to downtime.

According to the modern design philosophy of the Damage Avoidance Design, a structure should be designed also to minimize the structural and non-structural damage after an earthquake. Several construction systems have recently been developed to comply with this new performance requirement. Passive base isolation is by far the most effective way to reduce the effect of an earthquake on a structure. Its use has been somewhat limited by the cost, which is often believed to be high. However, significant progress has been made in this decade from when passive base isolation system was first proposed, and nowadays the cost does not seem to be an issue anymore. For these reasons, passive base isolation system was recently used also for residential applications, for example in L’Aquila (Italy) where both reinforced concrete and timber buildings were isolated [3]. Modern seismic design should therefore investigate this option also for timber buildings,
where the reduced weight compared to reinforced concrete allows the use of smaller and, therefore,
potentially cheaper base devices.

In this paper, the use of passive base isolation is investigated for light-frame multi-storey timber buildings. Two designs of a three-storey light-frame building located in L’Aquila (see Figure 1) were carried out according to Eurocode 5 and the Italian regulation for construction (NTC 2008): one with (see Figures 2 and
Figure 3) and the other without Friction Pendulum System isolators [1]. In the structure without seismic
isolation a behaviour factor q of 4 was used (high ductility), whilst the isolated building was designed
elastically assuming q=1.5. The seismic performance and cost of both solutions are compared, demonstrating
the convenience of using passive base isolation.

The main elements resisting the lateral loads in a light-frame building are the shear walls. The behaviour of timber shear walls is fairly complex as they are composed by different elements: timber frame, sheathing and metal connections (panel to frame connection and anchoring of the shear wall to the foundation or to the floor below). The seismic behaviour of the whole light-frame building, if not isolated, markedly depends on the dynamic performance of the shear walls. The shear walls were modelled using two nonlinear diagonal springs implemented in the SAP 2000 software package. The spring mechanical properties were assessed using the formulas provided by the New Zealand Timber Standard for the lateral deflection of a shear wall [4], and by calibration on former experimental tests performed at the University of Trieste. The model was
then used to carry out a linear static, a modal, and a non-linear time-history analysis of the entire 3D
building, with and without passive base isolation. The isolation devices were modelled using a non-linear
link available in the SAP 2000 library.

The behaviour of the two buildings was compared in terms of stresses in the elements and inter-story drifts, demonstrating the superior performance of t