Characterization of the damage induced by machining of fibre-reinforced composites is usually performed by measuring surface roughness. Contact-based surface profilometers are the most used equipment in industry; however, it has been found that there are performance limitations which may result when used to measure machined heterogeneous composite surfaces. In this research, surface roughness is characterised using a commercial non-contact optical method, and compared with a conventional stylus profilometer. Unidirectional and multidirectional carbon fibre laminates were edge trimmed and slot milled. The variation in surface roughness was compared using different tool types, fibre orientations and cutting parameters. Surface damage and cutting mechanisms were assessed by using scanning electron microscope images, and the suitability of roughness parameters were also analysed including: Sa, Skewness and Kurtosis. Using the optical system allowed accurate roughness calculation of individual plies on a multidirectional laminate with different fibre orientations. The research has also shown that the optical system, including the use of areal roughness parameters, can increase the accuracy of roughness measurement for machined fibrous composite surfaces and is less sensitive to measurement position than the stylus.

Many carbon fibre reinforced polymer composite parts need to be edged trimmed before use to ensure both geometry and mechanical performance of the part edge matches the design intent. Measurement and control of machining induced surface damage of composite material is key to ensuring the part retains its strength and fatigue properties. Typically, the overall surface roughness of the machined face is taken to be an indicator of the amount of damage to the surface, and it is important that the measurement and prediction of surface roughness is completed reliably. It is known that the surface damage is heavily dependent on the fibre orientation of the composite and cutting tool edge condition. This research has developed a new ply-by-ply surface roughness measurement methods using optical focus variation surface analysis and image segmentation for calculating areal surface roughness parameters of a machined carbon fibre composite laminate. Machining experiments have been completed using a polycrystalline diamond edge trimming tool at increasing levels of cutting edge radius. Optical surface measurement and µ-CT scanning have been used to assess machining induced surface and sub-surface defects on individual fibre orientations. Statistical analysis has been used to assess the significance of machining parameters on Sa (arithmetic mean height of area) and Sv (areal magnitude of maximum valley depth) areal roughness parameters, on both overall roughness and ply-by-ply fibre orientations. Empirical models have been developed to predict surface roughness parameters using statistical methods. It has been shown that cutting edge degradation, fibre orientation and feed rate will significantly affect the cutting mechanism, machining induced surface defects and surface roughness parameters.

No abstract