PSI - Issue 18

Nikolas Baak et al. / Procedia Structural Integrity 18 (2019) 274–279 Author name / Structural Integrity Procedia 00 (2019) 000–000

278

5

(a)

(b)

Fig. 6. Comparison of coercive field strength for three different feed rates for (a) S110 (Baak et al. 2018) and (b) S280 melts.

Throughout the fatigue tests the coercive field strength Phi cm decreases for S110, except an initial drop for f = 0.05 mm, with the same slope for all investigated feed rates. This behavior cannot be observed for the S280 specimens. For this melt Phi cm decreases just for f = 0.05 mm; for f = 0.10 mm and 0.15 mm Phi cm remains, after an initial increase, on a constant level. Figure 7 displays the maximumMBN amplitude M max versus the normalised number of cycles for both investigated melts and the three investigated feed rates. It can be noticed that the initial value of the maximum magnetic Barkhausen noise (MBN) amplitude M max for f = 0.15 mm for both melts is comparable, whereas the values for f = 0.05 mm and 0.10 mm is significantly higher for melt S280 compared to melt S110. For S110 after a levelling during the first cycles M max remains constant throughout the fatigue tests. M max shows for the feed rate f = 0.10 mm in S280 a sharp decrease after the first 10 4 cycles and remains constant afterwards, but shows a considerable big scatter. For f = 0.05 mm and 0.15 mm M max shows a smooth asymptotic development.

(a)

(b)

Fig. 7. Comparison of maximum magnetic Barkhausen noise (MBN) amplitude for three different feed rates for (a) S110 and (b) S280 melts.

These results show that the measurement of the micro-magnetic parameter in the initial state of a deep-drilled specimen do not allow conclusions on the fatigue performance of the specimen. A consideration of at least 20% of the expected fatigue life has to be done, to create a reliable fatigue life prediction model. In most cases a significant difference in the development of the micro-magnetic parameter between the first 10 4 cycles and the further development can be noticed. The observed leveling is not correlated to the general development of the material parameter, but can be interpreted as an indicator of the stability of the induced surface characteristics. Moreover it can be concluded that for every material, as shown for slight changes in the sulpher content, and manufacturing parameter combination a separate consideration of the micro-magnetic parameters has to be done.