Column-Slab Connection In Timber Flat Slabs

Abstract

With the development of large cross laminated timber (CLT) plates, a new field of possibilities for timber construction has opened. Combined with the invention of a rigid, momentresisting connection between CLT plates, it is possible to create structures like it was so far only achievable with reinforced concrete. The CLT elements can be placed next to each other on the construction site and be connected with the newly developed bonding system of the company Timber Structures 3.0 AG. In the end, the connected plates are vertically only supported by columns and thus create a timber flat slab. Since no additional vertical load carrying elements, such as beams, are needed in flat slabs, this structural system is highly desired in office or industrial buildings, as it adds flexibility to a building and therefore increases its value. Besides the necessary rigid connection between the CLT elements, a column-slab connection which is able to support one slab, while at the same time transferring vertical loads from upper storeys through the slab, has to be analysed as well, since the knowledge about this connection is still limited. The vertical load transfer is achieved by cutting an opening in the slab and connecting the columns with the corresponding counterparts right through this opening. The present thesis focuses on the column-slab connection in timber flat slabs made of CLT. The aim is to provide the basis of design, so that engineers are able to safely design such a connection. The topics of compression perpendicular to the grain, rolling shear, bending stress concentration and calculation methods are the crucial points in the design process. Since the rolling shear strength of timber often limits the load carrying capacity of column-slab connections and a requirement in the research project was to refrain from using metallic fasteners, novel solutions to achieve a sufficient load carrying capacity had to be developed. Within this thesis four experimental campaigns were conducted, accompanied by numerical investigations. In the first experimental campaign the material properties of beech plywood in compression perpendicular to the grain were determined. This material was then used in the second campaign to increase the support area and to locally reinforce the CLT plates, which were tested in biaxial punching to determine the influence of the opening in terms of rolling shear resistance and bending stress concentration. A third experimental campaign was performed on single boards subjected to rolling shear. The aim of this campaign was to analyse the influence of locally stressed, continuous boards. The fourth and final experimental campaign focused on the stress concentration around openings. Twelve CLT plates were tested in three-point bending tests. To determine the influence on the stress concentration, each specimen was tested nondestructively several times with different opening and support widths. Additional numerical simulations were then performed on full-scale column-slab connections to widen the application of the experimental findings. The biaxial punching tests showed that an opening in the plate does not decrease the load carrying capacity in terms of rolling shear. Previous studies showed that the rolling shear resistance of CLT plates subjected to concentrated loads is higher than the rolling shear strength of single boards would allow. An increase of 40% in rolling shear resistance between shear tests on single boards and punching tests on CLT plates was observed in the present work, which is similar to previous studies. The local reinforcement of the column-slab connection with beech plywood plates, either as column capitals or drop panels, showed its efficiency by increasing the load carrying capacity up to 69%, compared to plates without reinforcement. The third campaign with continuous boards showed that boards only locally subjected to rolling shear stress have an up to 20% higher load carrying capacity, compared to uniformly loaded boards. Together with the simultaneously acting compressive stress perpendicular to the grain, this effect is responsible for the increase in rolling shear resistance from shear tests to punching tests. In the fourth campaign, an approach to describe the stress concentration around interior and edge column was established. The basic idea is to find a factor that correlates the bending stress at a support without an opening to the actual bending stress if a certain opening is present. This was done in perspective to develop an approach applicable for engineers, since simulated bending stresses next to an opening do not represent the true stresses needed for the determination of the load carrying capacity and are therefore difficult to interpret. The main influencing parameters, listed in order of decreasing importance, are the opening width, the support width and the thickness of the plate. The accompanying, as well as the additional, numerical simulations helped to understand and support the findings from the experimental investigations. By combining the experimental, numerical and analytical investigations, it was possible to establish and simplify the methods to design a column-slab connection.