Non-destructive defect detection in additive manufacturing using the CTT technique
摘要
Additive Manufacturing (AM) processes, particularly Fused Deposition Modeling (FDM), enable the fabrication of complex components through layer-by-layer material deposition. However, process instabilities may introduce structural defects that compromise the quality and reliability of printed parts. This study evaluates, for the first time, the applicability of the Chirp Through-Transmission (CTT) technique for monitoring structural conditions in FDM-manufactured components. Controlled volumetric discontinuities were introduced into the first printed layer to create four defect severity levels. Two piezoelectric diaphragms acting as transmitter and receiver were attached to the print bed, and broadband chirp signals ranging from 0 to 250 kHz were transmitted through the printed specimens. The acquired signals were analyzed using Power Spectral Density (PSD) to identify frequency bands sensitive to structural variations. The repeatability of the excitation system was verified through correlation and statistical analyses of consecutive chirp acquisitions. Subsequently, three statistical descriptors, Root Mean Square (RMS), Constant False Alarm Rate (CFAR) and Mean Value Deviation (MVD) were evaluated for defect characterization. The results showed that frequency-selective RMS analysis provided the most consistent discrimination among the investigated structural conditions, with the frequency band between 169.4 and 170.4 kHz exhibiting the highest sensitivity to defect severity. In contrast, CFAR and MVD failed to consistently distinguish the conditions investigated. These findings demonstrate the potential of applying the CTT technique to FDM process monitoring and highlight the importance of frequency-selective analysis for detecting structural variations in additively manufactured components.