
We studied a C22E/C22R-grade structural steel sample with a minimum cross-section of 7.6 mm and a gauge of 12 mm. We applied a cyclical load of 230 MPa at a frequency of up to 5 Hz until failure. To monitor variations
in magnetic permeability, eddy-current measurements were taken at different excitation frequencies (50 kHz, 100 kHz, 250 kHz, and 500 kHz).
A semi-analytical model was used to solve the direct problem: the calculation of impedance as a function of magnetic permeability. The experimental results were compared to simulations to validate the model. The material’s relative permeability, μ, was estimated in the range 150-180, with permeability decreasing as a function of time over the course of the experiment.

The results showed a clear difference between the low-frequency (below 3,000 fatigue cycles) and high-frequency portions of the measurements. This difference can be attributed to different microstructural mechanisms occurring at different times during the experiment. The high-frequency portion of the measurements revealed almost linear behavior, confirmed by an analysis of the actual voltage values.
We also took magnetic Barkhausen noise (MBN) measurements to round out the eddy-current measurements. Thermographic data was used for correlation. We also factored in temperature variations to correct for any drift caused by the sample heating up.

According to our results, eddy-current measurements are suitable for monitoring fatigue-induced degradation in metal. Variations in magnetic permeability can be correlated with microstructural changes, enabling the
non-destructive monitoring of the material’s condition.
Now that eddy-current measurements have been demonstrated on the monitoring of fatigue-induced degradation, the door is open to new industrial solutions for real-time monitoring to help improve the safety and
durability of metal structures.