Stabilising system for multi-storey beam and post timber buildings

Abstract: For multi-storey timber buildings, the stabilisation against horizontal loads is one of the key issues, which becomes more severe with height. There are a few main concepts utilised to resist lateral forces; shear wall action, diagonal bracing, and frame action. This doctoral thesis addresses the stabilising system of a beam and post system where the main load carrying members are connected in a pinned manner and horizontal stabilisation is provided by special stabilising walls. The system ‘trä8’ developed by Moelven Töreboda AB has been studied as an example of such a system. The stabilising elements are prefabricated panels made of a glued laminated timber frame and cross-laminated veneer lumber Kerto-Q sheathing glued-screwed to the frame. The stabilising elements were examined in full-scale racking monotonic experimental tests. It was shown that prefabricated stabilising walls are very strong and stiff structures and can be successfully used for stabilisation of multi-storey timber buildings; however they require a reliable connection to the foundation. Connections with glued-in steel rods offer high strength and stiffness and were proposed as anchorage devices. They were tested both in the full-scale racking test and in single joint tensile tests. The experiment showed that these connections possess large load-carrying capacity and stiffness, but they fail typically by brittle pull-out of the rod from the timber member. A literature study was conducted to review existing manufacturing methods, mechanisms and parameters governing the performance and strength of the joints, theoretical approaches for the analysis of glued-in rods connections and existing design recommendations. Based on this review, the most recent methods were chosen to determine the pull-out capacity of glued-in rods connections and evaluate the test results. Additionally, the transformed section method was also found to be applicable for analysing stabilising walls anchored with glued-in rods connections. Brittle failure modes should be avoided in the design; therefore findings of the literature review were utilized in an attempt to achieve ductile behaviour of connections with glued-in steel rods. Two configurations of timber-to-steel connections with multiple glued-in steel rods, with many small diameter rods of high strength and few large diameter mild steel rods were tested in tension up to failure. The results showed that ductile behaviour can be achieved with mild steel rods and the number of rods should be small to reduce the amount of work during production and to reduce risk for manufacturing errors. The rods should be equally distributed in the timber cross-section to optimize the load transfer. At the current state, the only thing that seems to prevent a design method for the design of glued-in rod connections in Eurocode 5 is a proper estimation of the characteristic shear strength value. The full-scale racking test also included nailed hold-downs. For large nailed connections loaded in tension parallel to the grain, an increased risk for plug shear failure exist, which was observed in the racking experiment. Nailed connections in general have a lower stiffness than glued-in rods connection provided they have the same connection area. The results of this test contributed, together to those of a previous study, to check the design proposal in Eurocode 5 for shear plug failure capacity of nailed connections, which was found to be non conservative.