Seismic performance of CLT core-wall systems and connections

The Ph.D. research reported explores the seismic performance of cross-laminated timber (CLT) core- wall systems and their connections. While there is a renewed interest and implementation of timber buildings globally, many of these structures are hybrid solutions with reinforced concrete or steel systems utilized to resist lateral loading. This in part motivated the research, with the main objective to quantify the increase in strength and stiffness when multiple CLT shear walls are connected together to transform a planar CLT lateral load resisting system (LLRS) to a CLT core-wall LLRS. The research first comprised of experimental and analytical investigations of three critical connections for a CLT core-wall LLRS. The testing of dowelled hold-downs found that increased row spacing and end distance increased connection displacement capacity and ductility when compared to current spacing recommendations in literature. To provide an orthogonal connection between CLT walls, screws were installed with mixed angles, i.e. different installation angles between the screw axis and the plane of the CLT surface. An optimum ratio of two inclined screwsto one screw installed at 90° to the CLT surface was found. The average experimental overstrength was 1.7 for these dowelled hold- downs and screwed orthogonal connections. Through testing, it was found that the shear strength and stiffness of castellated connections were 2.5 and 7 times greater than the specimens using commercial angle brackets. A simplified stiffness-based load sharing analytical model was developed to predict castellated connection strength. A three phase CLT shear wall testing programme was executed to study the contribution of each component (wall and joint) to a CLT core-wall system. The programme consisted of 4 post-tensioned (PT) single wall tests, 5 PT double wall tests, 7 PT core-wall tests and one conventional core-wall test for comparison purposes. The highest CLT core-wall composite action of approximatelytwo-thirds was achieved. The core-wall stiffness was greater than eight times a single CLT wall for an approximate 3.5 times increase in CLT wall area. Mixed angle screwed connections were implemented along the vertical joints to provide strong, stiff, and energy dissipative connections. Analytical investigations sought to modify and develop the existing sectional analysis method for PT wall systems using the Modified Monolithic Beam Analogy. Through the use of Particle Tracking Technology, it was found that a strain amplification factor of 1.3 was required for CLT walls which is non edge glued and whose lamella are not machine stress graded. Analytical models were developed for PT double wall and core-wall systems based on a nonlinear curve fitting screwed connection model. Different kinematic modes could occur such as Flange Wall uplift and simultaneous Web and Flange Wall uplift depending on the relative strength and stiffness of the screwed connections to the PT and dissipater elements. It was found that for the PT core-wall, the compression Flange Wall could be neglected during strong axis loading if screwed orthogonal joints are employed. The analytical model well predicted the instance when kinematic modes change due to the nonlinear behaviour of the screwed connections. The analytical model captured the behaviour of the tested PT double wall and PT core-wall systems within 10% error