Robotic Sequential Ultrasonic Welding of Thermoplastic Composites: process Development and Testing

Abstract

Multi-spot sequential ultrasonic welding is a promising joining technique for fibre-reinforced thermoplastic composites structures (TPC). In existing research on the multi-spot sequential ultrasonic welding process, welds are produced through the use of a static table-top welding machine, at a coupon level. However, in order to apply this joining technology to large structures, the welding process needs to be up-scaled through the use of a robotic platform. At the Smart Advanced Manufacturing (SAM|XL) automation field lab and TU Delft Aerospace Engineering, a robotic sequential ultrasonic welding system has been developed. The system consists of a welding end-effector (EEF) equipped with various sensors that enable online process monitoring and control, which can be mounted on an industrial robot arm to perform sequential multi-spot welds. The goal of this study was to assess the welding performance of the ultrasonic welding EEF, which was mounted on an industrial KUKA KR210 R2700 Extra 10-axis robot arm, by comparing it to the performance of welds produced through the static table-top machine. In this study, single and multi-spot welds were produced on thermoplastic composite coupons, based on welding conditions which were defined in a preliminary study. The robot and EEF deflections observed during the welding process were analysed to assess the deviation of the robotic process from the static one. The feedback obtained from the welding equipment in terms of consumed power and tool displacement in both processes was also compared. The weld quality was assessed though single lap shear testing of the welded joints as well as fractography of the failure surface. The results of this study indicate that the developed robotic welding process is quite robust and is capable of producing high-quality sequential welded joints despite significant system deflections observed during the welding process. Slightly lower welded area and weld strength was obtained which can be attributed to the system deflections. Finally, the results indicate that the use of a stiffer robotic platform as well as a stiffer EEF construction will result in better system rigidity and weld spot positioning accuracy, and through this the welding process shows promise for large-scale industrial applications.