The purpose of this study is to determine how early man fastened wood together in order that it might be utilized to a greater extent and to trace the improvements and additions which have been made in these original fasteners of wood in the ensuing years.

With recent improvement in wood manufacturing technologies, taller and larger wooden structures are being constructed, thereby increasing the risk for potential damage due to a blast threat against such structures. Recent studies on the effects of high strain rate in wood have been undertaken, however the vast majority of these studies have focussed on structural elements with idealized boundary conditions. Some studies included realistic connections as the boundary conditions, but little progress has been made to date in order to quantify the behaviour of connections in isolation. The current study aims to investigate the response of steel-wood-steel bolted connections when subjected to blast loads. This includes determining the dynamic increase in resistance and stiffness for stocky and slender bolts in both the parallel and perpendicular to grain directions. The study also explores analytical solutions to predict the joint behaviour and discusses the validity of current blast design provisions. Bolted wood connections were investigated under both static and simulated blast loading using the University of Ottawa’s shock tube. The study found a dynamic increase in resistance and stiffness when the failure mode was dominated by wood crushing in both the parallel and perpendicular to grain directions. No increase in resistance or stiffness was observed when bolt yielding dominated the failure. A loss of ductility was observed under dynamic loading for the parallel to the grain connections designed to fail in wood crushing. It was found that the use of self-tapping screw reinforcement was an effective method of preventing premature splitting failures and enhancing the performance of a connection. The results showed that connections which engaged the fastener in bending exhibited more favourable behaviour than connections which engaged only in wood crushing. A two degree-of-freedom model was capable of modelling the connections even when the support frame system had some flexibility. The validated model was used to investigate cases where the connection could contribute to the energy dissipation. It was found that the performance of the assembly improved when the connections were considered. Recommended future work includes an investigation of brittle failure modes in bolted connections, exploring connections with more deformation capacities, and expanding the experimental component of the study to include full-scale structural assemblies with wood elements and boundary connections. Limited design recommendations have been proposed in the current study, however testing at the assembly level could shed more light on such an important topic.

The study of the ignition of wood and other cellulosic materials has resulted in the publication of a large amount of literature. As an aid in the planning and interpretation of future research, it was felt that a survey of a representative body of literature should be made. With this as its motivation, the present work was initiated.

Wood crossarms provide an efficient economical, structural, and sustainable solution to the distribution and transmission of electric utilities. The majority of these crossarms are made from two species of wood, Douglasir (Pseudotsuga menziesii) and southern pine (Pinus spp.). In this study 210 solid-sawn wood crossarms (120 pieces of Douglasir/ 90 pieces of southern pine) were evaluated with non-destructive (NDTs) and destructive testing to determine if NDT could be used to predict crossarm performance. Three NDTs were conducted. Subsequently, destructive testing in which each specimen was broken in accordance to ASTM D-198 for modulus of elasticity (MOE) and modulus of rupture (MOR) was performed. After testing each specimen the non-destructive data was compared to the destructive testing results by mean comparisons and correlations. In this study it was found that the E-computer (NDT) produced the strongest correlation in both species to destructive values, especially MOE.

This thesis introduces a new adjustment factor for the probability-based load and resistance factor (LRFD) design for wood structures. By investigating the empirical data of reaction forces for a wooden house built by the Forest Product Laboratory in 2001, it is found that the reaction values exhibit great variability. To explore the causes of this variability, a 3-D finite element model is built and analyzed using commercial software MSC/Nastran. It is found that differences in member geometry are a major cause of reaction variability. In examining the potential effect the reaction variability might have on the structural safety, reliability is assessed for two different types of wood products under several different situations. Finally, a new adjustment factor Ks, which accounts for the variability in load path, is obtained and validated based on structural reliability theory.

No description available.

No description available.

No description available.

No description available.

No description available.