Nonlinear Ferromagnetic Effects on Magnetic Barkhausen Noise Measurements of Steels
Magnetic Barkhausen Noise , Ferromagnetism , Electromagnetism , Non-Destructive Testing , Nonlinear Superposition
MBN probes measure Magnetic Barkhausen Noise (MBN), a magnetic signal produced when the magnetization of a ferromagnetic material changes state. Usually a dipole (two-pole) MBN probe is employed, which needs to be physically raised, and re-oriented, in order to induce and observe MBN signals from different directions. A tetrapole (four-pole) MBN probe, however, can use the principle of superposition to accomplish the same. Superposition allows tetrapole probe measurements to be performed more rapidly and without the repositioning errors associated with the dipole probe. However, past results from tetrapole MBN probes were inexplicably different from those obtained using a dipole MBN probe. The physics underlying a tetrapole MBN probe was the focus of the present thesis research. It was found that flux superposition is nonlinear in ferromagnetic material, whereas theory originally developed for tetrapole probe superposition had assumed linearity. In-sample flux density was modelled incorporating the effects of nonlinearity and anisotropy. It was also found that, 1) The MBN probe’s measure of in-sample flux density is usually inaccurate due to nonlinearity and flux spread in the sample, however the measure is accurate for a specific range of flux densities. 2) In-sample surface flux density is proportional to the square-root of the sum square of the MBN signal, and hence, may be inferred from the MBN. 3) There is a fundamental connection between MBN and ferromagnetic processes such as hysteresis and initial magnetization. An empirical magnetization model was developed, and derivation of a first-principles micromagnetics model was initiated. This work has explained the formerly inexplicable data obtained from performing superposition measurements using the tetrapole MBN probe, making the probe useable at low applied fields in a sample having a single easy axis. To simplify confounding factors, however, it is recommended that future material studies using MBN be done using the dipole MBN probe.