Presented in this paper are the results of an experimental program investigating the out-of-plane behavior of CLT panels under static and blast loading. A total of 18 CLT panels, with panel thicknesses of 105 and 175 mm corresponding to a 3-ply and 5-ply panel, respectively, were investigated with the aim to determine the dynamic increase factor (DIF). An average dynamic increase factor of 1.28 on the resistance and no apparent increase in stiffness from static to dynamic loading were observed. Two resistance material predictive models that account for high strain-rate effects and the experimentally observed post-peak residual behavior were developed. A single-degree-of-freedom model was validated using full-scale simulated blast load tests, and the predictions were found to match well with the experimental displacement-time histories.

An understanding of the behavior of wood stud walls in a region corresponding to hazardous-blowout damage levels is currently lacking. The focus of the current study is on investigating the flexural response of full-scale wood stud walls subjected to pressure-impulse combinations that would yield severe damage levels. Static material properties were determined and used as input in an analytical model that considers the nonlinear behavior of stud-to-sheathing joints, as well as high strain-rate effects. Experimental results showed that premature sheathing failure could occur prior to full flexural stud response, and that sheathing panel debris could be generated. The use of thicker sheathing and screws shifted the failures to the studs, while decreasing the amount of debris. Welded wire mesh was successfully used as a reinforcement of the sheathing or as a catcher system while maintaining the residual axial capacity of the studs. A single-degree-of-freedom material-predictive model was successfully...

An experimental program investigating the behavior of glued laminated timber (glulam) assemblies with various bolted connections subjected to simulated blast loading was undertaken. A total of 14 full-scale tests on $137\times 267\text{-}\mathrm{mm}$ glulam members with idealized and realistic boundary conditions were carried out using a shock tube apparatus capable of simulating the effects of far-field blast explosions. Full-scale glulam specimens with bolted connections designed to yield in bolt bending performed better than those that were overdesigned. Proper detailing of the bolt group geometry was found to be sufficient to achieve the desired failure sequence. Reinforcement with self-tapping screws changed the failure mode from that of splitting to a combination of bolt yielding and wood crushing and provided additional ductility in the assembly. A two-degree-of-freedom blast analysis was found to adequately capture the system response with reasonable accuracy. An investigation of the current Canadian blast design provisions showed that the design approach does not allow for energy dissipation in the assembly. An approach is proposed that requires more stringent provisions for the design of brittle failure modes in connections and ensures that ultimate failure will occur in load-bearing timber elements.

The performance of light-frame wood stud walls under simulated blast loading has so far been limited to investigating the behaviour of structural elements with idealized boundary conditions. The current study investigates, experimentally and analytically, whether walls with prescriptive connection detailing for low and high seismic and wind regions are capable of resisting blast loadings such that the walls’ ultimate capacity can be reached. The study also investigates the behaviour of different connections with various design capacity levels to develop failure in the stud wall system rather than in the connection. A total of ten full-scale walls with different boundary conditions were tested dynamically. The results showed that typical prescriptive connection detailing did not perform adequately. Designed connections performed well, but the findings show that basing the connection design solely on capacity may be inadequate. Single degree-of-freedom modelling may only be utilized if damage in the connections is limited.