The development of load-bearing timber systems have led to the realization of an increasing number of solid wall (X-Lam) or framed (platform frame and heavy frame) timber constructions, demonstrating the validity and competitiveness of wood in house building also in terms of comfort, sustainability and energy efficiency. The growing spread of the use of timber structures has led to the development of numerous innovative construction systems but at the same time remains a lack of norms in seismic field, in particular about the q-factor to be used for the design of different timber system. The standards uniquely allow to classify construction systems according to ductility classes (Low, Medium, High) on the basis of ductility value and on the degradation of resistance under cyclic loading but do not provide an unique value of the q-factor.

The paper analyzes the definition of the q-factor in the literature and its relevance in the design of structures in seismic areas. The available methods for estimating the q-factor are based on the verification of the non-linear seismic response of the entire building by means of experimental full scale shaking table tests or numerical simulations. These procedure results very expansive both from the economic and computational point of view . Furthermore the estimation of the q-factor performed using these methods are specifically of the examined building and should not be considered representative for the constructive system.

This work proposed a new procedure for the q-factor evaluation based on a proper application of the pushover method to the load-displacement curve obtained through experimental quasi static test on representative wall elements. The devised procedure was validated referring to the q-factor estimation performed by means of shaking table test and numerical simulations on CLT and Timber Frame building system.

An extensively application of the proposed procedure to a number of wooden building system tested in the CNR IVALSA laboratory (TN) is presented in order to carry out a reliable estimation of the q -factors. Finally the main advantages and limitations of the proposed procedure are presented and critically discussed.