Development of verified FE simulation method on crimp joint with gasket : A combined experimental and numerical study

Sammanfattning: When developing drivetrain cooling modules for commercial vehicles, the top priority is the strength of the products. The reason lays within the structural change that companies are undergoing where the warehouses are minimized and the companies are relying on the transportation of commercial vehicles. If a drivetrain cooling module would fail to maintain a targeted temperature range of the drivetrain, there is not long until the vehicle is forced to a stop. At worst, the downtime of commercial vehicles can cause companies to come to a halt as well. Therefore, developing verified FE simulation methods together with strength verification tests, are some of the core activities to ensure the strength of the cooling modules before implementation in vehicles. One of the methods that have not been sufficiently verified regards how to perform structural FEA on a crimp joint with a gasket. A crimp joint is established through a hemming process, where the aluminium header tabs are crimped over the polyamide tank, creating a watertight seal together with a gasket. The objective of this master’s thesis is to verify the existing simulation methods of the crimp joint and determine the margin of error with respect to strains. The objective of this thesis is also to develop a verified FE simulation method with a lower margin of error compared to the existing methods. The verification of the methods has been conducted through a comparison between numerical results and an experimental stress analysis, where the strain located on the header is measured with strain gauges in a pressure pulsation rig. The average margin of error found on the three existing methods of simulating the crimp joint was 39,7%, 13,4%, and 11,5% located on the outer bottom dimensioning radius of the crimp joint at a pressure of 100 kPa. However, the method used to determine the margin of error was found to be invalid, but the margin of error is only slightly affected by the verification method where the existing methods still are insufficiently representing the crimp joint. The method of determining the margin of error was then corrected for the development of the new method to represent the crimp joint in the FE model. The developed method in this thesis has an average margin of error of 6,9% at 100, 160, 220, and 260 kPa.