Eddy current testing (ECT) is a widely used non-destructive testing (NDT) method in many industrial applications, which has been developed and advanced significantly in the past few decades. However, there are still challenges to overcome, such as the need for scanning and imaging large surfaces, improving the defect detection and classification capability. In this work, these have been addressed using tunnelling magnetoresistance (TMR) sensor array and a swept-frequency chirp excitation signal, together with the pulse-compression algorithm. Conventionally, time domain analysis is used in pulsed eddy current to gather info about any defects. Here the time domain analysis is implemented on the output of the pulse-compression procedure applied to the swept-frequency chirp response, which is also processed via frequency domain analysis. The results showed larger SNR values for time domain images and a better clustering of the SNR values compared to the frequency domain, facilitating the defect depth estimation and classification. © 2021 Elsevier Ltd
Swept-Frequency eddy current excitation for TMR array sensor and Pulse-Compression: Feasibility study and quantitative comparison of time and frequency domains processing
Laureti, S.;Ricci, M.
2022-01-01
Abstract
Eddy current testing (ECT) is a widely used non-destructive testing (NDT) method in many industrial applications, which has been developed and advanced significantly in the past few decades. However, there are still challenges to overcome, such as the need for scanning and imaging large surfaces, improving the defect detection and classification capability. In this work, these have been addressed using tunnelling magnetoresistance (TMR) sensor array and a swept-frequency chirp excitation signal, together with the pulse-compression algorithm. Conventionally, time domain analysis is used in pulsed eddy current to gather info about any defects. Here the time domain analysis is implemented on the output of the pulse-compression procedure applied to the swept-frequency chirp response, which is also processed via frequency domain analysis. The results showed larger SNR values for time domain images and a better clustering of the SNR values compared to the frequency domain, facilitating the defect depth estimation and classification. © 2021 Elsevier LtdI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
