Pulsed thermography is a commonly used infrared thermal technique for non-destructive evaluation of engineering materials and components. The quality of the obtained results, in terms of sizes and depths of the researched defects depends mostly on the data processing methods and the observed time intervals. This work is focused on the algorithms used for processing the thermal data after a pulsed test: Pulsed Phase Thermography (PPT), Principal Component Thermography (PCT), Thermographic Signal Reconstruction<sup>®</sup> (TSR<sup>®</sup>), Slope and R<sup>2</sup>. The work focuses on an aluminium sample with shallow imposed defects and regards the post-processing analysis with different algorithms by considering different lengths of the cooling sequence (time interval or number of frames) and the investigation of the correlation between the signal contrast and the aspect ratio of defects. This correlation represents a first attempt for estimating the size and the depth of the defects, with a new empirical approach. Results show as the influence of the truncation window size changes according to the algorithm used for data analysis and the depth and the size of the detected defects. Moreover, each algorithm has its own peculiarities and capabilities and a synergic action in defects detection and characterization can be obtained if more algorithms are applied on the same thermal sequence.
Resistance Spot Welding is one of most used method to weld two sheet material thanks to its reliability and rapid production that makes it economical. The quality of the produced joints is related to the process parameters such as current, time, electrode pressure and surface condition and generally, destructive tensile tests are used to investigate the effect of each parameter on the quality of joints. However, these latter are time consuming and do not allow for a complete check (100%) of joints. In this work, a non-destructive thermographic technique and a new procedure was used to assess the quality of joints. In particular, different steel joints were obtained by varying two process parameters: current and cycle time. These joints were tested with the pulsed thermography technique adopting a transmission set-up. Thermographic data were acquired by means of a cooled infrared camera capable to acquire thermographic sequences at 400 Hz with a geometrical resolution equal to 0.062 mm/pixel. A flash lamp has been used to produce a thermal pulse of 3000 J with duration 5 ms. The thermal signal was investigated in the time domain with the aim to obtain different features related to the investigated process parameters. Proposed approach allows for detecting the typical defects that affect the joints such as stick and expulsion of material.