Experimental investigation of cross-laminated timber panels with realistic boundary conditions subjected to simulated blast loads

Recent research efforts have established key dynamic characteristics of cross-laminated timber (CLT) panels and developed parameters that are suitable for analysis and design. Despite such effort, no information is available on the critical role connections play in the behaviour of CLT when subjected to blast loads. The current study investigates a total of thirteen full-scale CLT wall panels with varying connection detailing by simulating the blast loading through a shock tube apparatus. The study found that typical connections with end grain screws into wall elements as well as angled double-threaded screws performed poorly and failed prematurely in tension perpendicular to grain. Reinforcing the connection with screws applied on the panel face did not mitigate the brittle failure. It was found that connection types where the wood is bearing on steel angles or directly on other wood members performed well even when under-designed according to the blast design standard. Whereas thin steel angles provided significant energy dissipation in bending of the angle resulting in reduced panel deflection and damage, thicker steel angles provided the energy dissipation through yielding in the screws as well rotational restraint of the panel ends. Using balloon type assemblies provided significant rotational restraint at the panel ends. The study also found that simplified models developed for idealized simply-supported conditions were not adequate to describe the behaviour of the system.