By expertly predicting the reliability of complete micro and nano technologycal systems, time-consuming interactions in development can be avoided even before they are completely assembled. This is made possible by estimating the characteristic system life from the reliability behavior of the critical components of the system. Checking the component behavior under system-relevant conditions provides the necessary information which is then linked and evaluated appropriately.
We work on this research task through experimentally dominated solution strategies. Based on more than 25 years of experience in testing the thermo-mechanical reliability of electronic components and systems, we are working in close cooperation with numerous partners from the industry to explore new methods for:
- accelerated testing of thermo-mechanical reliability for new technologies (e. g. SiC & GaN components, Ag & Cu sintering)
- multimodal loading (passive and active temperature changes, and vibration, humidity influence) for realistic determination of component life for the new applications with complex stress profiles (electric mobility, autonomous driving, industry 4.0, Smart Grid, Smart City, etc.).
In the field of lifetime modelling, we contribute to the enhancement of the method for:
- the determination of acceleration factors between different endurance tests as well as in relation to the operating behavior
- the aggregation and the summary evaluation of reliability under real conditions, meaning at very different loads
- estimating the system reliability from the behavior of the individual components, including irreversible failure mechanisms.
By means of analytical testing and characterization (group 'Analytics and characterization') of real components in connection with the determination of life time, degradation processes can be clarified phenomenologically and analyzed precisely in quantitative terms. Thus, it forms the basis for the work on modeling and simulation in the neighboring groups - both with regard to their alignment and for the validation of results.