A Help for Lifetime Management: Defect Characterisation Through Signal Processing of Ultrasonic Data

So as to detect and to characterize potential defects on cast components, a program of in-service inspections is carried out for the non destructive evaluation (NDE) of the component. A first step of inspection usually consists in detection. Still, should a defect be detected, one can be asked to prove the component still stands the mechanical constraints. Then, it is very important to be able to characterize precisely the size and the position of the defect. In the studied case, we concentrate on planar defects which are to be characterized from ultrasonic. The estimation of the height consists in measuring the time of flight between two reflectors: the diffraction and the corner echoes. But in case of small defects, the two echoes overlap and it is very difficult to give a height estimation of the defect. Then, the core of our contribution deals with the signal processing method we developed for the estimation of the height of planar defects. The use of our so-called deconvolution process now enables to obtain the height estimation of the defect, when it was impossible without processing. Processing results of real data sets are given for the inspection of representative mock-ups of homogeneous austenitic welds. In conclusion, the signal processing step turns out to be an important step in the characterization of defects and can thus contributes to maintain some components in service.