Timber joints with their simultaneous ductile and brittle failure modes still pose a major challenge when it comes to modelling. Wood is heterogeneous and highly anisotropic. It shows ductile behaviour in compression and brittle behaviour in tension and shear. A 3D constitutive model for wood based on continuum damage mechanics was developed and implemented via a subroutine into a standard FE framework. Embedment and joint tests using three different wood species (spruce, beech and azobé) were carried out, and the results were compared with modelling outcomes. The failure modes could be identified, and the general shape of the load–displacement curves agreed with the experimental outcomes.

Tests on double-shear steel-to-timber joints loaded parallel-to-grain were undertaken, using beech and azobé with one, three and five dowels in a row. The dowels were made of high strength steel (hss) and very high strength steel (vhss). The experiments have shown a significant difference in load-carrying capacity of joints with vhss and hss dowels. Joints with one dowel provided enough plastic deformation capacity to allow for ductile failure modes whilst this does not hold for joints with three or five dowels in a row. No correlation between load-carrying capacity and density within one wood species could be observed. Novel failure modes including steel failure of the dowels were observed. The observed effective number of fasteners was lower for the joints with vhss dowels. Also for the stiffness Kser, an effective number of fasteners could be observed. The tested timber joints with their simultaneous ductile and brittle failure modes still present a major challenge for modelling. Wood is heterogeneous, highly anisotropic and shows ductile behaviour in compression and brittle behaviour in tension and shear. A bespoke 3D constitutive model for wood based on continuum damage mechanics was used to capture these effects. Eight stress-based failure criteria were defined in order to formulate piecewise defined failure surfaces. The damage development was controlled by nine damage variables that were inserted in the damage operator. The joint test results were compared to modelling outcomes. The failure modes could be identified and the general shape of the load-displacement curves agreed with the experimental outcomes.

Tests on double-shear timber-to-timber joints and double-shear timber joints with slotted-in steel plates loaded parallel-to-grain were undertaken. The used species were spruce, beech, cumaru and azobé (ekki) with one, three and five dowels in a row. Two different steel qualities were used, high strength steel (hss) and very high strength steel (vhss) dowels. The experimental results have shown that the load carrying capacity of joints with vhss dowels is higher than for joints using hss dowels whilst still providing enough plastic deformation capacity to allow for ductile failure modes. No correlation between load carrying capacity and density within one wood species could be observed. The observed effective number of fasteners is lower for the joints with vhss dowels and depends also on the used wood species and the slenderness of dowels. Also for the stiffness K_ser, an effective number of fasteners for the joints with more than one dowel in a row could be observed. The well-established Johansen model can be used to design these types of joints.