A novel method for forming realistic pre-preg wrinkles in an AFP representative setup and their characterisation
Topic(s) : Special Sessions
Co-authors :
Ege ARABUL (UNITED KINGDOM), Robert HUGHES (UNITED KINGDOM), Vincent MAES (UNITED KINGDOM), James KRATZ (UNITED KINGDOM)Abstract :
Automated Fibre Placement (AFP) is an automated manufacturing technique to deposit pre-preg tape onto a surface using a gantry head, heater, and compaction roller [1]. Due to the automated nature of AFP, defect occurrence is inevitable during the manufacturing process. These defects can be costly for manufacturers, where defect inspection and re-work can result in 42% of the total build time [2]. Out-of-plane wrinkling is a notable example where the pre-preg tape lifts locally and distorts the part geometry [3]. This distortion can lead to a significant decrease in the strength of the overall part, which can be up to 50-70% [4].
While there is some existing literature regarding the formation of wrinkles for specific loading cases, such as the fibre steering, a better understanding of the forces and mechanisms that lead to their formation is needed, specifically after the first layer is deposited. It is established that wrinkle formation is correlated to AFP process parameters, such as material temperature, compaction force, layup rate, and material tack. Tow tension is a lesser investigated parameter, contributing to the wrinkle formation.
A novel experiment setup representing the process was developed to investigate the formation of wrinkles in AFP. The experiment involved applying tension on a longer strip of pre-prep tape using a tensile testing machine, representing the tension when the material is deposited. While the first layer is under tension, a second, shorter pre-preg tape is applied. The tension is then released, leading to wrinkling on the second layer. The experimental process was further improved by applying heat to the ends of the second prepreg layer, which increased the adhesion and attachment between the layers. Another improvement was depositing the second layer over a backing sheet, which ensured the second layer did not stick to the first layer before the tension was released. These additions improved the method's reliability, allowing for the generation of realistic wrinkles of varying fibre orientations and amplitudes. By combining the wrinkled layers, a quasi-isotropic composite laminate was formed.
During deposition, wrinkles in individual layers were quantified using a laser scanner. To characterise the wrinkles in the cured sample, eddy current sensors were utilised, which exploit the conductive properties of the carbon fibre material to detect individual fibres and subsurface defects. The sample was also characterised via sectioning, where clear indications of post-cure fibre waviness were observed, verifying the experimental methodology for generating realistic wrinkles. Furthermore, the results allowed for a direct correlation of laser scan data, eddy current, and microscopic measurements. Future work would include varying different tension levels to establish a relationship between tension and wrinkling and applying the theory established in an AFP deposition scenario.
While there is some existing literature regarding the formation of wrinkles for specific loading cases, such as the fibre steering, a better understanding of the forces and mechanisms that lead to their formation is needed, specifically after the first layer is deposited. It is established that wrinkle formation is correlated to AFP process parameters, such as material temperature, compaction force, layup rate, and material tack. Tow tension is a lesser investigated parameter, contributing to the wrinkle formation.
A novel experiment setup representing the process was developed to investigate the formation of wrinkles in AFP. The experiment involved applying tension on a longer strip of pre-prep tape using a tensile testing machine, representing the tension when the material is deposited. While the first layer is under tension, a second, shorter pre-preg tape is applied. The tension is then released, leading to wrinkling on the second layer. The experimental process was further improved by applying heat to the ends of the second prepreg layer, which increased the adhesion and attachment between the layers. Another improvement was depositing the second layer over a backing sheet, which ensured the second layer did not stick to the first layer before the tension was released. These additions improved the method's reliability, allowing for the generation of realistic wrinkles of varying fibre orientations and amplitudes. By combining the wrinkled layers, a quasi-isotropic composite laminate was formed.
During deposition, wrinkles in individual layers were quantified using a laser scanner. To characterise the wrinkles in the cured sample, eddy current sensors were utilised, which exploit the conductive properties of the carbon fibre material to detect individual fibres and subsurface defects. The sample was also characterised via sectioning, where clear indications of post-cure fibre waviness were observed, verifying the experimental methodology for generating realistic wrinkles. Furthermore, the results allowed for a direct correlation of laser scan data, eddy current, and microscopic measurements. Future work would include varying different tension levels to establish a relationship between tension and wrinkling and applying the theory established in an AFP deposition scenario.