Non-Destructive Plastic Deformation Estimation of Metals Using GMR Sensor-Based Eddy-Current Measurement

This study presents the development of a non-destructive plastic deformation estimation technique for paramagnetic metals, such as aluminum alloy and stainless steel, using Giant-Magneto Resistive (GMR) sensor-based eddy-current measurement. Plastic deformations in metallic structures can occur when the loading exceeding the yield strength of the material is applied. In addition, plastic deformation can induce the initiation and growth of fatigue cracks, leading to the catastrophic failure of metallic structures. Therefore, it is crucial to detect and evaluate the severity of plastic deformation in advance to ensure structural integrity and safety. Plastic deformation in metallic materials causes perturbation (or dislocation) in the material lattice, resulting in changes in electrical resistance (or electrical conductivity). Thus, in this study, the plastic deformation-induced electric resistance variation was estimated using eddy- current measurement. The sensitivity of eddy-current measurement was enhanced by replacing the conventional pick-up (sensing) coil with a GMR sensor. Test specimens using aluminum alloy (Al6061-T6) and stainless steel (SUS304) were fabricated, and various tensile plastic strain levels were applied using a hydraulic loading machine. The phase values were extracted from the eddy-current raw signal, and the relationship with the plastic strain level was investigated. Additionally, the influence of the external magnetic fields was minimized by applying a constant weak magnetic field using a permanent magnet. The experimental results indicate that the phase values of the eddy- current signal increase as the extent of plastic deformation. The uniqueness of this study lies in 1) the fabrication of the eddy-current probe based on the GMR sensor, 2) the minimization of the external magnetic field to enhance the eddy-current measurement, and 3) the experimental validation of the plastic deformation estimation using the eddycurrent phase values.