Design Validation of Automotive Inverter Components: the importance of Fatigue Simulations

Vibration qualification tests are currently indispensable for vehicle manufacturers and their suppliers. Carmakers’ specifications are therefore conceived to challenge the mechanical endurance of car components in the face of numerous in-service detrimental phenomena: fatigue damage, shocks or accidents, and mechanical degradation. In the automotive industry, components are commonly qualified by means of a test without failure, the goal being to determine whether it will or not "pass" customer requirements. Validation of new or newly designed components is obtained via bench test and structural simulation. Simulation has gained traction in recent years because it represents the first step of the design validation process. In particular, Finite Element Analysis simulations are so powerful that they not only justify physical testing (shaker tests) on prototypes but enable engineers to optimize the design and predict the durability of the part. This paper illustrates how FEA simulations were applied to product validation in the pre-serial phase, i.e. during the optimization of the manufacturing process. In particular, we will focus on the printed circuit board (PCB) of the inverter of an electrical driven compressor, undergoing the final vibration validation test. The FEA simulation has permitted to optimize the initial assembly process that was not compliant to the vibration test. Two failure modes were identified and fixed thanks to the simulation's iterations. The first failure mode regarded one of the PCB components: the simulation results lead to a more reliable manufacturing process with particular regards to the additional damping element (glue) and its optimal position on the critical PCB elements. The second failure mode concerned the optimization of the position of the screws that hold the PCB inside the inverter case. Simulating various screws angles and torque values sensibly ameliorated the mechanical endurance of the PCB during the shaker test. The novelty of this approach lies in using FEA simulation to control the PCB's dynamic response (e.g., optimizing the position and geometry of applied glue to avoid stress concentration) and optimizing the PCB assembly components (the screws). In addition to the simulation, controlling the screw angle in testing is essential for achieving a uniform assembly process, thereby reducing the risk of variability-induced stress concentrations.