Induction welding of high performance thermoplastic composites

Abstract

Mass reduction is one of the main drivers for aircraft design. A big potential for saving mass can be found in part joining. Currently, most parts are joined by mechanical fastening. However, mechanical fastening requires labor intensive work and results in a weight increase especially for composite parts that, due to the mass saving they allow, are increasingly used in the aerospace industry. Induction welding of thermoplastic composites shows great potential to limit this mass increase and it can vastly reduce the amount of work that is needed to join parts while improving the working conditions on the shop-floor. A characteristic of induction welding is that the heating rate goes down with increasing distance to the inductor. This often leads to fully melting the top plate across its thickness. This results in an imperfect weld as pressure is required for proper consolidation which leads to a damaged surface. Therefore, the challenge is to find a way to selectively heat the weld zone, without using a metallic susceptor that can lead to an unreliable joint. Accordingly, the aim of this thesis is to find how induction heating can be used to selectively heat the weld zone of multidirectional carbon fiber laminates with and without using a carbon fiber susceptor material. To this end, three sub-questions are answered. Can specific plies in multidirectional CFRP laminates be heated by manipulation of the magnetic field? Can a carbon fiber material be used as susceptor material in order to increase the temperature at the weld interface? Is it possible to weld parts with lightning strike protection? Using an infrared camera and fine wire thermocouples of type E to measure the temperature profile across the thickness of the test parts, a dependency of the heating rate on the orientation of the magnetic field was discovered for all parts that were tested. This dependency can indeed be used to selectively heat plies in a specific orientation within the laminate. Also, the possibility of using susceptors made of carbon fiber to increase the temperature at the weld zone was shown. To this end, susceptors were made of fabrics and of thin-ply material. Further research has to be done to improve these susceptor materials. An increase of the heating benefit they deliver could enable them to be used for more efficient welds in the future. Finally, by clamping lightning strike protection (LSP) onto the test parts, the temperature profile of parts with LSP was imitated. Striking was that the heat generated in the two CFRP parts went down when the LSP was added. No peak temperature was measured at the side of the LSP but this could probably be explained by the poor contact between the thermocouple in the test part and the LSP. It is recommended to further investigate this topic by consolidation of the LSP onto the test part.