Search

Results of tests performed on joints used in timber-frame construction allow characterizing the variability of their mechanical behavior, which differs substantially from one joint to the next. The parameters of a constitutive model of the joints and their variability are identified. Finite element (FE) models of a shear wall and a timber-frame house are used in nonlinear dynamic calculations to study the propagation of uncertainty through the structure. It demonstrates that for a single-story 6 m × 6 m house, the variations in mechanical strength of each connection do not significantly affect the structural behavior of the house. Both the numerical and experimental results (on a shaking table) are quite similar, proving the model accuracy, its ability to study the propagation of uncertainty and its relevance for future development (non-regular, multi-story buildings…). Moreover, a sensitivity analysis performed on a FE wall model under uncertain seismic loads reveals the importance of earthquake motion modeling.

A numerical model based on the finite element method is presented for prediction of the cyclic response of wood frame structures. The model predicts the cyclic response of shear walls. Nonlinear phenomena are assumed to be concentrated in the connections that are modelled through elements linking the structural elements including the posts, beams, and sheathing panels. Identification of model parameters relies on tests on individual connections. Connection tests on different nail lengths were conducted under monotonic and cyclic lateral loads. Based on the results from past studies that indicate the pull-through failure is an important failure mode in common nail connections with lumber and oriented strand board (OSB), washers were considered as a means to reinforce the connection. The influence of reinforced nailing on the static and dynamic performance of full-size wood frame shear walls with large openings, sheathed with OSB panels, was evaluated experimentally. Combinations of parameters were studied, such as the number of hold-downs, the panel shapes, the nail distribution, and the bracing systems. Comparisons of the dissipated energy per cycle revealed a higher capacity for walls using nails with washer reinforcement than without. Results from numerical simulations of the monotonic and cyclic tests performed on the walls are presented.Key words: timber shear wall, connections, finite element, dissipated energy.