PSI - Issue 5

Jürgen Bär et al. / Procedia Structural Integrity 5 (2017) 793–800 Jürgen Bär et al. / Structural Integrity Procedia 00 (2017) 000 – 000

799

7

8

4

stress concentration factor without crack

4 stress concentration factor  loc /  nom 5 6 7

2 stress concentration factor  loc /  nom 3

stress concentration factor without crack

1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 3

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

distance from crack tip [mm]

distance from crack tip [mm]

Fig. 6. Stress concentration factor in the notch root (red) and at the side of the specimen (green) for a quarter elliptical crack with a length of a = 0.5 mm and c = 1 mm.

5. Discussion

The experiments have shown that the potential drop method can be improved by using a LOESS filter algorithm. This smoothing method reduces the detection limit for short fatigue cracks and leads to a distinct reduction of the crack initiation lifetime, provided that the first increase of the potential is taken as the criterion for the beginning of crack propagation instead of a defined crack length of a=250 µm. A quantitative measurement of the length of short fatigue cracks with the potential drop method is problematic. The potential drop method gives just an integral signal that is not suitable for a determination of the crack geometry. Furthermore, the Johnson ’s formula is not sufficient for quantitative measurements of crack lengths until a through-the-thickness crack is formed. For long cracks with a continuous crack front, the LOESS filtering provides a significant improvement of the DC potential drop method, but in case of short crack growth, optical measurements are recommended. The cyclic lifetime of notched specimen of 7475-T761 is determined by the propagation of short fatigue cracks. These cracks were formed at the edges of the notch and propagate in the form of quarter elliptical cracks with the large half-axis along the notch root and the short axis in the direction of the specimen surface. The crack propagation rate along the notch root was found to be higher compared to the crack propagation on the specimen surface. It could be shown by FEM calculations that the higher crack propagation rate in the notch root is caused by an elevated stress concentration between the two corner crack tips. In addition, the interdependency between the two cracks effects an additional acceleration of the crack propagation in the notch root. Assuming that the crack propagation in the notch root can be determined by a long crack propagation law, calculated crack propagation rates are significantly higher than the correspondent experimental values. An approach to explain this discrepancy could be a 2-dimensional crack propagation as illustrated in figure 7. The sketch shows a quarter elliptical crack emanating from the corner of the notch. It is supposed that a crack can only propagate when a critical opening at the crack tip is exceeded and that the crack propagation in the notch root (z-direction) is coupled with the crack propagation on the specimen surface (x-direction). Consequently, an extension of the crack along the