Fig. 4 Fatigue crack depth vs. number of fatigue cycles in sea water.
At 298K, cracks also grew steadily with increased N at any f in the range of 10-0.2 Hz indicating growth rate inversion with respect to f (at this temperature, maximum growth rate emerged at 1Hz) for SWRM10, but the growth rate tended to increase monotonically with decreasing f in the range of 10〜0.3 Hz for SWRH42A. While not shown in the figure, a similar trend was observed at 288K (growth rate maximum at f = 0.7 Hz in the range of 10〜0.2 Hz for AWRM10, and f = 0.5 Hz in the range of 10〜0.3 Hz for SWRH42A).
Figure 5 plots the observed aspect ratio against crack depth in 298K seawater for both specimens. As can be seen, with increasing crack depth, the semi-elliptical crack surface geometry approached a semicircular shape. It was also noticed that there was no significant f dependence of aspect ratio in seawater in the range of f = 0.15〜10 Hz for SWRM10, and f = 0.3〜10 Hz for SWRH42A. There was no temperature dependence of aspect ratio in seawater, and aspect ratio vs. crack depth relationships at any other temperature examined were virtually identical to that at 298K, exhibited in Figure 5.
Figure 6 plots the dl/dN vs. Δ K relations in air at f = 10 Hz determined for SWRM10 and SWRH42A from the crack growth curve in Figure 3. As can be seen, dl/dN was higher for SWRH42A than for SWRM10 in the higher Δ K region, but they were indistinguishable in the lower Δ K region, indicating that the crack growth behavior was virtually the same for the both specimens in the low Δ K range.
Similarly, the results concerning crack growth behavior in air showed that the crack growth rate was about the same in all the specimens: mild steel 41 and high-tensile-strength steel SB 50B [4] ; five steels, including SM50B, S45C, and HT80 [5] ; mild steel A36 and low-alloy high-tensile-strength alloy A588 [6] ; S15C, S35C, andS45C [7].
Figure 7 summarizes the dl/dN vs. Δ K relations observed in seawater at 298K as determined from the crack growth curves shown in Figure 4. There was a certain range of Δ K in which dl/dN varied linearly with Δ K in air and seawater, demonstrating the validity of the Paris rule in accordance with the available reports [8-10]. When compared in the range of validity of the Paris rule, dl/dN in seawater was greater than that in air. This trend seemed to be most pronounced in the low Δ K range. Similar trend was observed under other examined temperatures, but the results are not presented here.
