Compression after impact fatigue damage growth in CFRP - What does no-growth really mean?

icaf2023 Tracking Number 51

icaf2023 presentation

Impacts on carbon fiber reinforced composites (CFRP) can produce internal damage comprising of matrix cracks, multiple delaminations and possibly fiber fracture, which is hard to detect from simple surface inspection. This situation is known as barely visible impact damage (BVID). Considering that every airplane faces impacts of various severities during its operational life and that the majority of exposed surfaces in new-generation aircraft are made of CFRP, there is a high chance that aircraft will be flying with unnoticed impact damage. For this reason, BVID fatigue propagation must be taken into account in design. The EASA and the FAA dictate a no-growth design philosophy for BVID. This is usually verified by tracking the projected delaminated area using ultrasound inspections. However ultrasound inspection can only detect delamination and suffers from a well-known shadowing phenomenon. Up to what extend does a no-growth in the detected damage (using ultrasound inspection) correspond to a no-growth in the real damage? To investigate this question, fatigue compression after impact (CAI) tests were conducted on a quasi-isotropic layup of CFRP. The following fatigue damage propagation was monitored by combining through thickness attenuation scan, echo-pulse C-scan and acoustic emission. The delamination growth was precisely reconstructed and combined with acoustic emission data leading to the following critical considerations: - Projected delaminated area is not sufficient to describe the fatigue damage propagation of BVID. Through thickness attenuation scan showed growth inside the impact cone, while echo pulse scan showed the outward growth of short delamination falling inside the projected area of larger ones (see figure). Both types of growth are not captured if only the overall projected area is considered. - Mechanisms of damage accumulation different from delamination cannot be excluded a priori. Acoustic activity was detected also when no delamination propagation was observed using ultrasound. The conclusion is that current damage descriptions are insufficient, and that a no-growth in the detected damage (using the current procedure) does not necessarily correspond to a no-growth in the real damage.