Impact testing of pristine and repaired carbon fibre reinforced polymer composite materials for aircraft structures

Abstract

Aircraft technologies and materials have been developing and improving drastically over the last hundred years. Over the last three decades, an interest in the use of composites for the external structures has become prominent. For this to be possible, thorough research on the performance of composite materials, specifically the impact performance, is required. Previous research of impact testing for pristine carbon-reinforced epoxy composites describes matrix cracks, fibre fracture and delamination as the failure modes that require monitoring. An area of concern with the use of composites for aerostructures is their ability to be repaired and retain a suitable level of performance. Currently, since there are limitations in non-destructive testing (NDT) methods for adhesive bonding, adhesively joined or repaired composite materials are restricted to being used for secondary structures within the aircraft, unless another joining method – such as mechanical fasteners – are also implemented. Rigorous research and testing are required in this area because the current technique for metals, of bolting an undamaged piece of material over the damaged area, is not effective for composites as it introduces detrimental damage. There are two main repair techniques for composite materials: scarf and patch repair. Investigating the potential of the latter to restore the impact properties of carbon fibre reinforced polymer (CFRP) composite panels contributes to a large section of the research in this thesis. Two different types of lay-up (quasi-isotropic and cross-ply) were tested, with repair variables such as the patch diameter, patch thickness, inclusion of a plug and distance of the impact site from the centre of the patch being adjusted to see how each affects the overall performance of the repair. It was seen that the patch diameter has little to no effect on the repair performance, with both 55 and 65 mm patches giving similar load traces. The patch thickness and inclusion of a plug had a more significant effect on the impact properties of the repairs, with the two best performing repair configurations being a thick patch and a thin patch reinforced with a plug. The location of the impact site also greatly affected the performance of the panel, with impacting on the patch but not centrally giving the largest damage area of all the configurations tested. Two primary types of impact will be considered in this thesis: hard and soft. Hard impacts are defined as having limited deformation of the impactor upon impact, examples of which include metal debris hitting the aircraft. Soft impacts have a significant amount of deformation of the impactor and include hail stones or bird strikes. Considering the transition from a hard to a soft impact and the effect this has on the failure modes seen in the CFRP is investigated as part of the research in this thesis. To consider this, a rounded stainless-steel impactor, flat ended stainless-steel impactor and a flat ended stainless-steel impactor with different thicknesses of neoprene adhered to the end were used to impact pristine quasi-isotropic material. The results suggested that, although a lower damage initiation value was observed for the samples impacted with the round-nosed impactor in comparison, the samples impacted with the flat-ended impactor had a larger damage area once damage did initiate. The addition of rubber reduced the peak load and increased the displacement of the samples. It was also seen that the damage area reduced slightly as the thickness of rubber increased. Another area of interest is accurately predicting the impact performance of pristine panels under both hard and soft impact loading conditions through the use of numerical modelling. This allows for various scenarios and the transition between the two types of impact to be considered without the cost and time of experimental work. In this thesis, a model has been developed and compared with the experimental results for the hard and soft impact research to investigate the potential for it to be used to determine suitable scenarios to test experimentally. The numerical analysis results reflected the overall trends observed experimentally when increasing impact energy for round-nosed impacts and also when increasing the softness of the impact, with excellent agreement between the simulation and experimental testing for the damage area produced by a 7.5 J round-nosed impact. This demonstrates the potential to use a model similar to the one in this thesis to predict the performance of CFRP panels under impact loading conditions, but also highlights some drawbacks and a need for further development and refinement to improve the accuracy.