Automated crack length measurement for mixed mode fatigue cracks using digital image correlation

Abstract

It is well known that mode II leads to a kinking and mode III to a twisting of a fatigue crack. On the surfaces of an SEN specimen, an overlapping of these two crack loading modes occurs. Superimposed with mode I, this results in very complex fatigue crack paths and, in some cases, multiple crack branches. Thus, established methods for mode I crack growth measurement cannot be utilized. Therefore, the digital image correlation (DIC) technique is used to determine the crack length of mixed mode cracks on the surface of a specimen by means of the maximum principal strain. By Gehri et al. (2020) a concept is described to obtain the crack path by morphological thinning of the area on the surface, where the maximum principal strain exceeds a threshold value. In this paper, an extension of the provided MATLAB script is proposed that includes two criteria for measuring the crack length versus the number of load cycles (a-N-curve). For the first criterion, the crack opening is used to determine the current crack tip position and to obtain the crack length as a function of the number of load cycles. In the second criterion, a threshold value of the maximum principal strain is used for determining the number of cycles at the detected crack length. The new evaluation routine with both criteria is initially calibrated on mode I experiments using the direct current potential drop method. Moreover, mixed mode experiments under tension-torsion loading are conducted, to validate the method and compare the two proposed criteria. The experimental results in terms of the a-N-curves obtained with the marker load technique are in good agreement with the results using the presented method. Thus, the extended tool provides an evaluation routine to determine a-N-data for mixed mode fatigue crack growth tests independently from the crack orientation and possible branching.