Fatigue damage and residual strength of composite materials for COPVs

This thesis investigates the effect of fatigue-induced matrix damage on the residual fibre-direction strength of composite laminates relevant to composite overwrapped pressure vessels (COPVs). A plate-to-coupon procedure was developed in which large parent plates were cyclically loaded before daughter specimens were extracted and tested against virgin references to determine the residual ultimate tensile strength (UTS) in the fibre direction. Two loading routes were considered: [±45°]₂s laminates to generate shear-dominated damage and [90°]₈ laminates to promote transverse matrix cracking. Stereo digital image correlation (DIC) was used to monitor strain-field evolution, spatial uniformity and stiffness degradation during fatigue. The procedure was established on glass-fibre/epoxy (GFRP) plates and then applied in a preliminary carbon-fibre/epoxy (CFRP) campaign. In the GFRP shear route, moderate apparent shear-stiffness reductions of 20–28% closely matched residual UTS losses of 20–28%, whereas at the highest damage level the stiffness loss of 64% exceeded the 43% reduction in UTS. The transverse route produced a different response: stiffness degradation reached approximately 24%, while residual UTS losses remained below 9%. LB02 instead showed neither measurable stiffness degradation nor residual UTS loss despite a comparable stress range, suggesting a possible energy threshold for sustained transverse-damage accumulation. In CFRP, both routes showed little distributed stiffness degradation at the investigated loads. However, the transverse-route plate failed suddenly through a local transverse crack without a preceding reduction in spatially averaged stiffness. Inconsistencies in the virgin reference tests prevented a quantitative residual-strength comparison, although the high post-fatigue UTS values suggested that most fibre-direction capacity was retained. Overall, a clear relationship between stiffness degradation and residual strength was observed only for the GFRP shear route, particularly at moderate damage levels. The relationship therefore depends on the damage mechanism, severity and material. The validated procedure provides an experimental basis for fatigue-strength modelling and further CFRP testing.