Defense and illustration of time-resolved pulsed thermography for NDE

After defining a good usage of pulse thermograms, in particular their ‘rectification’ thanks to log-log plot, and fitting by the TSR (thermographic signal reconstruction) technique, it is demonstrated that a simple and efficient normalisation is possible. The optimisation of pulse thermography is proposed based on two complementary approaches: the early detection of contrasts and the use of the TSR technique with logarithmic derivation. It is shown that in these conditions the quality of thermal images is greatly enhanced (reduction of blurring) and the accuracy of the identified defect parameters (depth and thermal resistance) is better. Based on experimental and simulated results it is shown that to optimise pulse stimulated thermography it is necessary to consider two parameters: the signal-to-noise ratio of the defect contrast and the precociousness of its exploitation. It is shown that this second parameter, which has been neglected by most of the users of pulse thermography who identified the defects parameters from the maximum contrast, can be improved by the logarithmic derivation.