Qualification of standardized long-term tests on copper materials for the economic determination of material parameters for CAE applications

Project description

The long-term tests used in practice to characterize the materials behavior of copper materials are only of limited suitability for determining reliable input data for CAE applications (e.g., finite element method, FEM). However, due to increasing demands on components made of copper materials, e.g., in electrical connectors, more precise knowledge of this input data is essential for safe and resource-efficient component design.

The project aimed to obtain more accurate information about the long-term behavior of copper materials from established standard tests without increasing the experimental effort in order to provide reliable input data for CAE applications.

To this end, a new test bench was set up at fem to enable precise instrument measurements in cantilever tests. At the same time, comparative measurements were carried out in uniaxial tests in the universal testing machine. Using numerical methods and machine learning, a clear correlation was demonstrated between the measured variables of the ASTM tests and the time and temperature-dependent properties of copper materials. This makes it possible to determine the time and temperature-dependent materials behavior or the material parameters for a selected materials model directly and cost-effectively from the measured variables of an ASTM test.

Fraunhofer IWM subproject:

As part of the subproject at Fraunhofer IWM, a novel materials model was developed based on the results of two previous projects (IGF 17278 N, IGF 18597 N). Previous materials parameters were derived from uniaxial relaxation tests, which are costly and resource intensive. Machine learning methods were used to make better use of alternative test methods such as the cantilever test (ASTM Standard E328 Standard Test Methods for Stress Relaxation for Materials and Structures) and to convert the characteristic values from these tests into materials parameters.