Fatigue crack growth (fcg) is one of the most challenging damage events for structural components subjected to non-constant loads. In particular for lightweight structures like airplanes, which are designed along the fail-safe concept, fatigue cracks are inherent to their design. Consequently, profound knowledge of the crack growth behaviour of structural materials is crucial for safety critical applications.

The methodology for fcg experiments has not changed for decades and usually still relies on the concept that a theoretical stress intensity factor is calculated with respect to a standardized specimen geometry, applied load, and projected crack length measured by integral methods like direct current potential drop. Such conventional fcg experiments need a lot of time and, thus, are very expensive. On the other hand, the experiments only result in a single material curve, i.e. a-N, which is hard to reproduce even under identical testing conditions. Consequently, the experimental outcome-to-cost ratio is relatively low.

In our work, we introduce a modern methodology for fcg experiments combining digital image correlation (DIC), AI-enhanced crack detection and robotics. The experimental hardware is built around a global DIC system. During experiments we use this system to monitor the specimen, detect the growing fatigue crack and send information to a robot which carries a light optical microscope for local high resolution DIC-measurements. In a nutshell, our method enables continuous access to the global and local displacement and strain field of the specimen throughout the whole experiment. Data acquisition, storage and analysis is automatically controlled by our open source Python software “Toolkit for Mechanical Testing (ToM)”. We show that the proposed method vastly increases the experimental outcome compared to the conventional method.