Damage Tolerance of 3D Woven Composites
Topic(s) : Material and Structural Behavior - Simulation & Testing
Co-authors :
Christian STEWART (UNITED KINGDOM), Bassam EL SAID (UNITED KINGDOM), Stephen HALLETT (UNITED KINGDOM)Abstract :
3D woven composite materials are seeing increasingly wider use in engineering applications such as automotive, renewable energy, and aerospace structures. This category of composites provides several economic and performance advantages over their laminated counterparts and metal alloys. 3D woven composites are usually prepared in near net-shape preforms before being infused with resin and cured to form the final structure, eliminating the costs associated with the layup of laminated composites or with the use of expensive prepreg materials. Additionally, 3D woven composites provide some key mechanical performance advantages, such as higher resistance to out-of-plane loading, as well as higher damage tolerance. Consequently, the main area of application for these composites involves impact, where the presence of barely visible impact damage (BVID) can be detrimental when structures are subjected to subsequent static and fatigue loading. It is therefore necessary to investigate the effect of BVID on the mechanical performance of these composites. In the present work experimental investigations were carried out to determine how the presence of BVID affects the tensile performance of carbon/epoxy with a layer-to-layer 3D weave.
A level of damage, which is representative of BVID, had to first be identified and introduced into the specimen before subsequent testing. Several specimens were first loaded to failure in quasi-static indentation, where three distinct stages were identified in the load-deflection behaviour (see Figure 1): (1) linear up to damage onset, (2) damage propagation, and (3) ultimate failure. Some indentation tests were interrupted at different load levels (labelled A to E in Figure 1), followed by damage characterisation at each loading level. The extent of damage was first assessed using ultrasonic C-scanning, and then X-ray Computed Tomography (CT) scanning was employed to investigate the damage mechanisms created with increased loading. Figure 2 illustrates examples of CT scan slices taken through the damaged region. From these results, the indentation level most representative of BVID could be identified, which was necessary for subsequent testing.
To assess the influence of BVID on the tensile performance, tension tests were performed on both undamaged and post-BVID 3D woven composite samples. The effects of BVID on the global stress-strain behaviour and failure modes were assessed. The use of Digital Image Correlation (DIC) provided information on the strain distribution and localisation in both types of specimen. The experimental results presented in this work give further insight into the damage tolerance of 3D woven composites.
Acknowledgements:
The authors wish to acknowledge the support of Rolls-Royce plc through the Composites University Technology Centre (UTC) at the University of Bristol and the EPSRC through the CoSEM Centre for Doctoral Training grant, no. EP/S021728/1.
A level of damage, which is representative of BVID, had to first be identified and introduced into the specimen before subsequent testing. Several specimens were first loaded to failure in quasi-static indentation, where three distinct stages were identified in the load-deflection behaviour (see Figure 1): (1) linear up to damage onset, (2) damage propagation, and (3) ultimate failure. Some indentation tests were interrupted at different load levels (labelled A to E in Figure 1), followed by damage characterisation at each loading level. The extent of damage was first assessed using ultrasonic C-scanning, and then X-ray Computed Tomography (CT) scanning was employed to investigate the damage mechanisms created with increased loading. Figure 2 illustrates examples of CT scan slices taken through the damaged region. From these results, the indentation level most representative of BVID could be identified, which was necessary for subsequent testing.
To assess the influence of BVID on the tensile performance, tension tests were performed on both undamaged and post-BVID 3D woven composite samples. The effects of BVID on the global stress-strain behaviour and failure modes were assessed. The use of Digital Image Correlation (DIC) provided information on the strain distribution and localisation in both types of specimen. The experimental results presented in this work give further insight into the damage tolerance of 3D woven composites.
Acknowledgements:
The authors wish to acknowledge the support of Rolls-Royce plc through the Composites University Technology Centre (UTC) at the University of Bristol and the EPSRC through the CoSEM Centre for Doctoral Training grant, no. EP/S021728/1.