PSI - Issue 28

Levke Wiehler et al. / Procedia Structural Integrity 28 (2020) 925–932 Author name / Structural Integrity Procedia 00 (2019) 000–000

931

7

Fig. 8. (a) Run of the quotients, specimen 2; (b) visualization of overloads on the fracture surface of specimen 2.

4. Discussion The investigations have shown that the multiple potential drop measurement enables the early detection of cracks combined with a determination of the crack`s location. Due to the formation of quotients, potential changes caused by temperature fluctuations are eliminated, improving the accuracy of the potential measurement. The measurement of multiple potentials is inexpensive and easy to implement by using amplifiers to measure the potential. It offers itself as a simple method for crack detection during fatigue tests. A detection of crack initiation in a running test is possible by integrating suitable conditions into the control software. In the case of single-edge notched specimens, the detection of the crack location can be determined by simply forming the quotient between the relative potential measured on the front-/backside and on the narrow side of the specimen. The crack initiation can be identified by changes in the quotient of the potentials, which depend on the location of the crack formation. The formation of secondary cracks can also be detected using this method, but there are some challenges. The determination of the cracking of subsequent cracks always depends on the first cracking. From the potential curves, information on the geometry or inclination of the crack front in long cracks can be obtained. These are not easy to interpret due to the non-linear relationship between potential change and crack length (Johnson (1965)). No reliable statements can be made about the minima and maxima of the crack front in relation to the mean crack length, e.g. an ‘S’-shaped crack must be considered as linear. However, a distinction can be made between a straight, linear, or curved crack front. An optimization of this method is quite possible with the help of further tests and a calibration of the potential differences. The presented experiments show different opportunities multiple potential drop measurements offer in fatigue tests. These are to be expanded and refined in further studies. In addition to the development of clear criteria for crack detection, the focus is on the determination of the crack front’s geometry. Knowledge of the crack front’s geometry offers, e.g. the ability to detect inadmissible skewing of the crack front at an early stage in crack generation for fracture toughness measurements without visual inspection of the samples and to take appropriate remedial measures. Furthermore, attempts should be made to determine the geometry of the crack front during the propagation of short cracks, thus creating the possibility of enabling a comprehensive investigation of the spread of short cracks without complex optical measurements. For this, however, further experimental and numerical investigations are necessary in