The dynamic response of timber shear walls

This thesis describes three numerical models, developed by the author, that predict the behavior of timber shear walls. Two of the models have been implemented in finite element programs. One program predicts the static behavior of shear walls and the other predicts the dynamic response to earthquakes. Both programs incorporate 1) the ability to predict the ultimate load capacity of the walls, 2) the effects of bearing between adjacent sheathing panels, 3) the effects of bending in the sheathing, and 4) the effect of bearing and gap formation between framing elements. The third model is a closed form mathematical model that was developed to predict the steady state response of shear walls to harmonic base excitations. A series of experimental tests were performed to determine the load-deflection characteristics of single nail connections between the solid wood, used for framing, and sheathing materials. The load-deflection characteristics for these connections were used to predict the behavior of timber shear walls using the finite element models. An extensive experimental program, consisting of five different tests and forty-two full size shear wall specimens, was conducted to verify the three numerical models. The experimental program included a new test that is representative of the earthquake loading for a ground floor wall of a three-storey North American apartment building. The test results were also used to: 1) compare the performance of waferboard with that of plywood sheathing, 2) investigate the dynamic behavior of shear walls, 3) investigate effects of out-of-plane deflections of the sheathing, and 4) examine the anchoring connection that resists the overturning moment. The merits and shortcomings of the five shear wall tests are discussed, along with their future usefulness in determining the effects of changes in the construction techniques used for timber shear walls. The dynamic model and shear wall test results are then used to investigate four design codes to see if shear walls, designed according to the various codes, adequately resist the loads experienced during earthquakes. A resistance factor for the design of shear walls is recommended for inclusion in the 1990 Canadian timber design code. The recommended resistance factor will result in the design loads specified by the code being more representative of the loads generated during an earthquake. Three important construction details are also discussed to inform the reader of possible problems that can be expected if these details are neglected during either design or construction. The details are: the hold-down anchor, framing corner connection, and sheathing connector. Finally, recommendations are made as to the type of research that is required to develop a design procedure for timber shear walls, based on the actual dynamic characteristics of shear walls as well as reliability theory.Applied Science, Faculty ofCivil Engineering, Department ofGraduat