Fig. 12: Crack propagation process for an unpatched under fatigue loading
Monitoring the metallic crack growth with passive IrT shows that the longer the crack grows, the better it can be made out on the thermal image. For small crack lengths on the other hand, crack tip detection can be difficult. One reason might be, that the movement of the lower piston heats up the specimen during testing. Temperature thus increases from the bottom to the top and the crack acts as a heat barrier. The bigger the temperature differences become over time, the easier cracks can be made out. Moreover, a circular area around the crack tip can be identified which builds up during loading and vanishes while unloading the specimen. This area helps to identify the crack tip and thereby to measure the crack length.
As for the quasi-static test, differences between the images from the unpatched specimen and the ones with patch monitored from the metallic side are neglectable. Fig. 13 shows the temperature evaluation for two patched specimens, one monitored from the metallic side and the other from the patched one.