This paper presents a study of a novel shear connector in a timber-concrete composite solution, focussing on the determination of an analytical expression that makes it possible to predict its behaviour and a numerical analysis that describes it accurately. The shear connector is composed of a perforated steel plate inserted into a slot within the timber rib and glued, in combination with reinforcing corrugated steel bars affixed to the top of the plate. Previous tests made it possible to establish failure mode in different T composite section plate-rebar configurations. These results determine the effectiveness of the system in terms of force-slip behaviour, with systematic failure in the timber section. A simple predictive model is proposed to determine the ultimate capacity of the joint, taking into account the mechanical properties of timber in relation with the fracture plane and the timber-adhesive interface. This model makes it possible to apply a design process that is able to predict the stiffness of the connection. FEM models were analysed for each configuration in a variable load process equal to that used in the test, according to the standard procedure. A variable friction coefficient in contact definition made it possible to achieve an accurate descriptive model in association with the test procedure.

Introduction and conceptual approach

Analytical studies of TCC systems

The efficiency of a timber-concrete composite solution (TCC) is directly related to the effectiveness of the joint used to connect timber and concrete. The effectiveness of the resulting joint depends on its capacity to transmit the shear loads developed in the contact plane between materials, and it has to be stiff enough to limit the slip between the timber-concrete sections. In any case the behaviour of a timber concrete composite element depends on the properties of its components (timber and concrete) and connector characteristics. The partial composite action definitively results from the rigidity of the shear connections. Thus structural efficiency and an adequate design process depend on the ability to predict the stiffness of connections. In behavioural terms the most effective connection is one that could satisfy the following conditions. Firstly, a connection between the timber section and concrete slab while undergoing stresses in an elastic range has to be stiff. Secondly, there has to be a ductile response when stresses develop in the plastic range. In conventional T beams with an upper concrete flange connected to a timber beam it is not possible to use the transformed section method. The Navier-Bernoulli hypothesis is not appropriate in situations where the material interlayer is not fully rigid, resulting in a relative slip between the lower surface of the concrete and the upper surface of the timber. Most common shear connections exhibit a partial composite action with some relative slip at the interlayer, and this leads to analytical complexity that is increased in the case of timber and concrete by the heterogeneous nature of their physical and mechanical properties. The calculations of stresses and deformations in TCC are usually performed on the basis of linear-elastic behaviour for all the materials, in accordance with simplified design rules. Some analytical methods have been developed to address the yielding of shear connectors. Some of them, like the Gamma Method proposed in Eurocode 5 [1], assume the case of mechanically jointed timber beams based on the theory of linear elasticity with a flexible connection, and they do not contemplate plastic deformations of connectors when yielding. An overestimation of the post-yielding loadcarrying capacity is therefore implicit in this method. Annex B of the Eurocode 5 proposes an analytical method for the case of mechanically jointed timber beams.