PSI - Issue 41

P.M.D. Carvalho et al. / Procedia Structural Integrity 41 (2022) 24–35 Carvalho et al. / Structural Integrity Procedia 00 (2019) 000 – 000

29

6

Fig. 3. Example of mesh close to the adhesive layer for the model with t P2 =1 mm and cohesive elements; mesh biases shown.

The models using the CZM had the adhesive layer meshed with a unique layer of four-node cohesive elements (COH2D4) with their sweep direction as the through-thickness of the adhesive layer, as proposed by Barbosa et al. (2018), and their approximate element size was 0.2×0.2 mm 2 (Fig. 3). The cohesive law was imposed through the material properties. Regarding the models to obtain stress distributions, the adhesive layer was modelled with 10 layers of continuum elements (CPE4R), resulting in finer meshes. The boundary conditions imposed represent the experimental setup. Symmetry (Ux=0) conditions were imposed to the left vertical edges (Fig. 4). The lower adherend’s right edge was constrained in both vertical and horizontal directions (Ux=Uy=0), representing the grip exerted by the testing jig. Finally, the models were displacement driven, i.e., a displacement  was applied to the upper edge of the L-shaped adherend, being in these cases larger than the experimental measurements (  =6 mm) to allow the prediction of P m . Similarly to the experimental cases, the reaction forces were calculated, and the maximum value was regarded as the numerical P m .

Fig. 4. Boundary conditions imposed to the model and detail of the mesh at the end of the bond-line for the models to extract stress distributions.