PSI - Issue 33

R. Nobile et al. / Procedia Structural Integrity 33 (2021) 685–694 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

692

8

in bending tests in a flat position performed on the different batches of specimens. In all the specimens, the initial linear-elastic behaviour is followed by an elasto-plastic phase until a peak value is reached, after which there is a loss of pressure. The experimental data in the presence and in the absence of impact damage showed a variability of data different density of the foams, influencing the residual strength. a b

Fig. 9. (a) Load-displacement diagrams of AFS-20 and (b) AFS-30 specimens.

The different mechanical behavior in the absence or in the presence of damage is more evident in the case of panels with less thick foam, with the sudden drop in load after the peak. The perfect bond between skin and the core of all panels had been verified by pre-testing NDT controls. Sandwich beams can fail in several ways during flat position bending tests: core shear, skin yielding, skin recess and wrinkling and detachment [12]. AFS20 and AFS-30 specimens after three-point bending tests are shown in Figures 10 and 11. The partial detachment of the compressed skins and the breaking of the stretched GFRP skin occurs for almost all AFS-20 panels, however, in all three cases of flexural damage in the absence of impact, indentation damage is evident, not observed in the case of the pre impacted specimens. a b

Fig. 10. (a) AFS-20 specimens tested for flexion in the absence and (b) in the presence of impact damage.

Only the AFS-20_P6 specimen has a core fracture near the extreme side, which can be justified by the evident