is the AE hit rate per second, the horizontal axis is the fatigue testing time from 139.9 to 160.2 elapsed fatigue hours. Crack was initiated in about 513 thousand fatigue cycles or 142.3 hours after the fatigue test started, then the AE activity keeps relatively quite which implies that the crack was rest for a while after initiation. In about 6 hours later or 535 thousand overall fatigue cycles, significant crack activity or crack growth that was lasting for about 4 hours was observed. The AE amplitude in this period is shown in Figure 4(b), each dot in the figure represents the amplitude of each AE hit. Trending of the AE amplitudes around 40 dB is clearly seen in the figure, this indicates continuous crack growth in this period though occasionally the amplitude drops below 35 dB. By further zooming the instant signal around 12:00 am of August 3rd in Figure 4(b), the AE amplitudes in less than one minute span are shown in Figure 4(c). The amplitudes are scattered in fairly regular interval which coincides to the fatigue cycle, i.e. one per second. In other words, there is one AE hit or one crack event in every fatigue cycle in this period. Table 1. Remarks of crack monitoring alone with elapsed fatigue time & cycles at location C Date Time of the Day Elapsed Time (Hours) Fatigue Cycles (Thousands) Remarks 14:07 139.9 504 AE monitoring started. 16:32 142.3 513 AE activity that indicates crack initiation was observed in the first time. 20:17 146.1 526 AE activity lasting for about 10 minutes was observed. 2/Aug 22:42 148.5 535 Significant AE activity that indicates continuous fatigue crack development in about 4 hours was observed in the AE. 08:30 158.3 570 Visual inspection didn’t find crack. 14:24 164.2 592 Visual inspection didn’t find crack. 3/Aug 16:05 165.9 598 Stepwise AE activity that indicates crack growing and resting was observed. 05:50 179.6 647 Long lasted AE activities were observed. 4/Aug 08:50 182.6 658 Visual inspection found crack with 20mm long, in 40 hours later than that AE reported crack initiation. Besides the continuous cracking activity observed, a stepwise cracking activity was also observed in location C. Figure 5 shows the AE amplitude in about 598 thousand fatigue cycles or 165.9 elapsed fatigue hours. Not only the amplitude shows an increasing trend, but also the amplitude drops down to a lower level and then it increases again. This phenomenon implies crack growing and stress relief during the fatigue process. When micro cracking is build up, the AE amplitude is increased until the stress concentration is big enough to
RkJQdWJsaXNoZXIy MjM0NDE=