PSI - Issue 52

Minori Isozaki et al. / Procedia Structural Integrity 52 (2024) 176–186 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

183

8

Table 6. Failure Load / Apparent tensile strength (Flatwise Test, Experiment) Specimen No. Welding time ( s ) Failure load ( N )

Apparent tensile strength ( MPa )

5

5.5

3320

36.5

The resulting displacement-load diagram is shown in Fig. 9. Table 6 shows the failure load, and the apparent tensile strength calculated from the area of the weld zone obtained based on VHX 8000 (KEYENCE Co.) image and the failure load. As in the compression shear test, the load decreased at a certain displacement, at which point the welded joint failed (Fig. 9). The tensile strength calculated from the experimental results was 36.5 MPa. 3.2. Analysis As with the numerical simulation of the compression shear test, Abaqus 2020 was used to reproduce the test. This model consists of two CFRTP laminates and a circular weld zone (PEEK), shown in Fig. 10. The diameter of the weld zone, α mm, was calculated based on the condition of the weld zone ob served by VHX8000 after the test, assuming that the shape of the weld zone is a perfect circle (Table 7). The material properties assumed for the CFRTP, and welded area are as shown in Table3 and Table 4, just as in 2.2. In this case, the flatwise test was reproduced by fixing the CFRTP on one side with PIN and applying displacement in the z direction to the CFRTP on the other side (Fig. 10). This analysis was performed in a secondary analysis.

Fig. 10. Flatwise Test FEM model.

Table 7. Diameter of welding area α ( mm ) Specimen No. Area ( mm2 )

Diameter α ( mm )

5

90.7

10.7

Using the models constructed in this way, numerical simulations of flatwise tests were performed. In this model, the stress state when the load applied to the displaced surface (surface indicated in red in Fig. 10) is the same as the experimental value is discussed. It was showed the stress in the normal direction (Fig. 11(a)) and the stress in the shear direction (Fig. 11(b)) at the weld interface to discuss the stress distribution at the adhesive interface. Table 8 shows the stresses in the normal and shear directions at the interface when the load is the same as the experimental values, along with the experimental results. Fig. 11 shows that in the flatwise test, stress was concentrated at the edge of the weld, especially stress in the normal direction, which was uniformly concentrated at the edge. Considering this stress distribution and the fact that the failure of the adhesive interface is based on the parabolic criterion (Koyanagi et al. 2014, Koyanagi et al. 2010, Ogihara et al. 2010), it can be concluded that the failure of the weld interface in the compression shear test is due to normal and shear stresses. The stresses listed in Table 8 are the interfacial stresses at the initiation point of rupture. As described above, in the flatwise test with tensile force applied, failure was attributed to a mixed mode of tensile