Speaker
Description
This study presents an extended infrared thermographic investigation of cold-sprayed 316L stainless steel coatings subjected to different post-spray heat-treatment temperatures. The primary objective of the work was to reliably differentiate coatings processed at distinct heat-treatment temperatures, namely as-sprayed (no heat treatment), 600 °C, 800 °C, and 1000 °C, which are known to induce systematic changes in coating porosity and microstructure. Flash-pulse thermography was employed as a non-contact and non-destructive technique to evaluate these thermally induced modifications. Particular attention was given to practical challenges such as non-uniform heating and surface curvature, which commonly occur in real industrial components and complicate thermographic interpretation.
A comparative evaluation of thermographic data-processing techniques was conducted, including Pulsed Phase Thermography (PPT), Thermographic Signal Reconstruction (TSR), Principal Component Analysis (PCA), and machine-learning–based classification models. Each method was assessed in terms of its ability to suppress artefacts caused by uneven thermal excitation and curved surfaces, while preserving informative signals related to coating condition. Phase-based analysis received special attention due to its inherent robustness against spatially non-uniform heating.
The experimental results demonstrate that phase responses obtained by pulsed phase thermography exhibit clear and systematic trends with respect to heat-treatment temperature. In particular, distinct peaks in the phase-frequency domain were observed, whose positions and amplitudes correlate with the applied heat-treatment temperature. These phase characteristics reflect changes in the thermal diffusivity of the coating, which are primarily governed by heat-treatment-induced variations in porosity and inter-particle bonding. Importantly, phase analysis was shown to effectively compensate for the influence of non-uniform heating and surface curvature, enabling reliable comparison between coatings with different geometries.
Additional thermographic indicators derived from TSR coefficients and PCA components were also analysed. Several of these metrics exhibited measurable sensitivity to heat-treatment temperature, although their robustness to heating artefacts varied across methods. Machine-learning approaches were employed as supervised classification models to automatically distinguish coatings processed at different temperatures based on thermographic features. The classification results demonstrate the potential of data-driven methods for automated assessment of coating condition, provided that sufficient and representative training data are available.
To validate the thermographic findings, the results were compared with metallographic analyses of coating cross-sections obtained by optical microscopy. Quantitative and qualitative porosity assessments derived from polished sections showed agreement with trends identified in the thermographic data. Coatings subjected to higher heat-treatment temperatures exhibited reduced porosity and more homogeneous microstructures, which corresponded to systematic shifts in thermographic phase characteristics. This comparison confirms that infrared thermography can indirectly capture microstructural changes in cold-sprayed coatings through their thermal response.
Overall, the study demonstrates that infrared thermography, and pulsed phase thermography in particular, is a robust and sensitive tool for the non-destructive characterization of cold-sprayed metallic coatings. Its ability to compensate for non-ideal heating conditions and to distinguish coatings with different heat-treatment histories highlights its strong potential for industrial quality control, process optimization, and in-line inspection of cold spray coatings.
