LAUSR

Abstract Inertia Friction Welding (IFW) is commonly approached by considering ideal conditions in which the two workpieces are in perfect contact, subjected to the nominal loads and the machine reacts ideally to the process loads. These conditions, however, are not representative of a real weld, where fixturing issues, non-ideal interaction between the workpieces and compliance of the system under the process loads could significantly affect the quality and the repeatability of the weld. To fill this research gap, a novel monitoring system able to collect in-process data and a methodology for their analysis was developed. A set of rundown tests and steel w elds were performed on an industrial inertia welder to validate the rig. Then, the data extracted were used to study the interaction conditions between the spindle and fixture side of the machine and build dynamics models to understand the physical implication of specific events connected to the impact at part contact and the flywheel deceleration. The results showed a significant influence of the machine in the alignment of the workpieces, with the runout between spindle and fixture that became larger and irregular during welding when the workpieces interact in non-ideal conditions. The quantitative comparison between the runout magnitude of rundown tests and welds showed an increase of more than four times that can be justified with the compliance of the machine, in which the spindle bearings representing the weakest element. The comparison of the outputs of the different sensors installed allowed to obtain, for the first time, a holistic view of the macroscopic phenomena occurring during the welding phase and observe how these could affect the final weld geometry and the dynamically evolving thermo-mechanical conditions of the weld.