Crack tunneling effects on the elastic unloading compliance of C(T), SE(B) and clamped SE(T) specimens and correction methodology

Abstract

This paper covers the effects of crack tunneling on SE(B), C(T), and clamped SE(T) specimens and presents a correction methodology for this effect and is divided in two parts. Part one presents an investigation of how crack front curvature affects instantaneous crack size predictions based on the elastic unloading compliance technique. Relative crack depths (a/W) of 0.2, 0.5, and 0.7, were considered alongside five levels of crack curvature. Refined finite element models provided load-CMOD records in order to support compliance assessment. The crack front was modeled as a semi-ellipse, and the compliance results agreed with experimental data from the literature. It was shown that for the same equivalent physical straight crack standardized by ASTM, compliance generally decreases as tunneling increases. Since the maximum crack curvature allowed by the aforementioned standards is very restrictive, compliance did not meaningfully change within that limit, however, if violated, this paper shows that higher deviations may occur, leading to inaccurate crack depth estimations and invalid test results. These limits and deviations were clearly determined and, as a step to improve the techniques, this paper also presents – in part two – an exploration of a possible approach to mitigate this problem, which is based on the modification of how the equivalent straight crack of a curved crack front is determined. This new approach presents reduced errors in compliance-based crack size estimation as crack curvature increases when compared to current standardized protocols, and it can support further investigations in order to validate and standardize improved measuring techniques. Finally, it is important to state that even though the ASTM E1820 is used for the determination of crack driving forces, this study is based only on the study of the crack front curvature, the limit imposed by this standard and the deviations on crack size estimation when those limits are violated, while not focusing on determining errors directly on the J-integral. This paper is a further development on the studies published before by the research group.