The experimental work was done completely with a single tape-wound core of grain-oriented 50–50% NiFe. The hysteresis loop tracer employed feedback to the effect that the flux in the core was made to vary linearly in time. Under these conditions, Barkhausen noise (in the usual sense of the word) could not occur. It does appear, however, as noise on the mmf (magnetomotive force) signal. In this paper, it is this mmf noise which is referred to as Barkhausen noise. Barkhausen noise is analyzed in terms of the spring model, which explains the noise as being due to statistical fluctuations in the number and strengths of the defects with which a wall is interacting. An expression is derived for the autocorrelation function R(τ) of the noise. Comparison of theoretical and experimental R(τ)'s determines the value of two parameters of the model: S′ Z and N. The values of S′ Z found (S′ Z ≃ 1.7 × 10−3) agree well with values obtained earlier by six completely different methods. The value of the density parameter N is new and is ≃ 1.5 × 1012 m−3. It is concluded that the spring model not only explains the average shape of hysteresis loops of materials meeting the criteria for simplicity, but also explains statistical fluctuations from the average.

Baldwin, J. A., & Pickles, G. M. (1972). Autocorrelation Function of Barkhausen Noise in Simple Materials. Journal of Applied Physics, 43(3), 1263–1268.