PSI - Issue 77

D.C. Gonçalves et al. / Procedia Structural Integrity 77 (2026) 79–86 Gonçalves et al. / Structural Integrity Procedia 00 (2026) 000 – 000

83

5

Fig. 4. (a) DCB ready for testing; (b) ENF specimen ready for testing.

(a) (b) Fig. 5. (a) Experimental DCB − curves; (b) Experimental ENF − curves. Table 1. Experimental and [ ⁄ ] obtained with DCB and ENF test. DBT 0.1762 0.4663 CBT 0.2321 0.4756 CBBM 0.2381 0.5127

3.2. Numerical Results Fig. 6 shows the nodal discretizations of the DCB and ENF numerical models. The adhesive layer is very thin in both adhesive joints, which requires an extremely refined nodal density in this region compared with the rest of the model. The DCB and ENF models were discretized with 7231 and 10513 field nodes, respectively. In the ENF case, the specimen is more slender, demanding a higher number of field nodes for the same nodal refinement in the adhesive layer. Moreover, as Fig. 6 shows, in the ENF case, the crack tip region is refined, whilst nodal density is decreased along the adhesive layer to reduce computational cost whilst maintaining accuracy. The boundary conditions shown in Fig. 3 were imposed in the numerical analysis. Also, =6 and =9 were considered in the construction of influence domains in the DCB and ENF cases, and two-dimensional plane strain conditions were assumed.