E. Della Torre

Parameter identification of the complete-moving-hysteresis model using major loop data

The complete-moving-hysteresis (CMH) model computes both the reversible and the irreversible components of the magnetization and the relationship between these components. An important question associated with a given model is the corresponding identification strategy which provides a method to experimentally obtain the parameters of the model. In order to be experimentally applicable, such an identification strategy has to be robust in the presence of experimental error. This paper presents a complete parametric identification strategy for the CMH model that is applicable to a wide range of recording materials. The method is robust in the presence of experimental errors and requires the measurement of the major loop, the major remanence loop, the virgin magnetization curve and the virgin remanence magnetization curve only. The method is experimentally verified on various commercial recording media. It is found that the model and the identification strategy are in excellent agreement with measured higher-order asymmetrical reversal curves for commercial samples with markedly different characteristics. The limitations of the identification strategy are discussed. A generalized Preisach function for the CMH model is suggested that is applicable to soft magnetic materials, as well. >

Preisach modeling and reversible magnetization

Methods of dealing with reversible magnetization in the context of Preisach modeling are examined. A generalization of the vector product model which includes reversible effects, as well as the usual irreversible effects, is presented. >

A Preisach model for accommodation

A statistical interpretation of the Preisach model led to the moving model that relaxed the congruency property and to the complete moving hysteresis model (CMH) which included a state-dependent locally-reversible magnetization. Since, like all Preisach models, these models possess the deletion property, minor hysteresis loops calculated by them are immediately stable. In real media successive traversals of minor loops are not superimposed. Accommodation describes this approach to equilibrium of static minor hysteresis loops, which may require many cycles for some materials. This paper presents a further statistical modification to the Preisach model to include accommodation, and it discusses the properties of such a model and the identification of its parameters. >

Measurements of output-dependent Preisach functions

The interiors of hysteresis loops were examined for commercial recording tape samples of gamma -Fe/sub 2/O/sub 3/ and Co-modified gamma -Fe/sub 2/O/sub 3/. It was found that the minor loops did not satisfy the congruency condition of the classical Preisach model. Since both the moving model and the product model are capable of describing minor loops whose height varies with the magnetization, they were used to analyze this data. It was found that the results were consistent with the expectations of the moving model and not with the product model. An identification method for the moving parameter is given. Two models for the reversible component of the magnetization that are compatible with the moving model are presented. >

Theoretical aspects of demagnetization tensors

A magnetometric demagnetization tensor is defined for uniformly-magnetized samples of arbitrary geometry. It is shown that this is a real symmetric tensor with non-negative diagonal elements and unit trace. Diagonalization of the magnetometric demagnetization tensor yields the Brown-Morrish equivalent ellipsoid and the conclusion that Stoner-Wohlfarth particles of any surface geometry have only two easy directions of magnetization. Transformation of the tensor, subject to periodicity constraints, leads to the conclusion that uniformly-magnetized samples with rotational periodicity, of order n \geq 3 about a single axis, have isotropic behavior perpendicular to that axis. It is also shown that if a sample has spherical periodicity with four or more vertices it does not exhibit shape anisotropy. Various methods for determining the numerical values of the tensor elements are presented.

Measurements of Interaction in an Assembly of γ‐Iron Oxide Particles

An experimental study of the statistical stability of the Preisach function is performed. The shape of the sample is varied in order to introduce an adjustable demagnetizing field to simulate various degrees of statistical instability. By choosing the proper shape it is possible to obtain a statistically stable distribution of interaction fields. Measurements of the standard deviation of interaction field show, just as calculations of the critical fields of the particles have shown, that the particles do not appear to switch as signal units.

Particle Interaction in Magnetic Recording Tapes

The magnetic coating of recording tape is assumed to be composed of an assemblage of small, single‐domain particles. Each particle is assumed to have a symmetrical, square hysteresis loop when the reversible component of magnetization is neglected and when the particle is not influenced by the fields of neighboring particles. When influenced by the fields of its neighbors, the particle may exhibit an asymmetrical loop when the loop is plotted relative to an external applied field. In this case the positive and negative switching fields for the particle are not equal, and their difference gives an indication of the particle interaction. While it is not possible to measure the switching fields of a single particle on the recording tape, the distribution of switching fields in the assemblage of particles and the associated magnetic moment can be measured. The two switching fields and the magnetic moment define a 3‐dimensional distribution function which describes the magnetic properties of the tape, and in t...

A variable variance Preisach model (garnet film)

The analysis of an artificial medium has led to a new type of Preisach model where the variance of the interaction field is a function of the magnetization. The ideal particles of this medium have no reversible magnetization and only interact magnetostatically with their neighbors. In this model, the line separating the positively and the negatively magnetized regions of the Preisach plane is not a staircase. >

Efficient numerical implementation of complete-moving-hysteresis models

An efficient numerical technique for the implementation of Preisach-derived models is presented. It is shown that, when efficiently implemented, all Preisach-derived models require only a few table lookups at each field value to compute the magnetization for both locally reversible and irreversible components. In particular the complete-moving-hysteresis (CMH) model accurately describes the magnetization processes. Using this implementation technique, the computation of the magnetization using the classical Preisach model has the same order of complexity as that of a single-valued nonlinearity. The incorporation of these models into field problems using a boundary-element method (BEM) a finite-difference method (FDM), and a finite-element method (FEM) is discussed. >

Magnetization calculation of fine particles

The switching mechanism of a distribution of fine particles is not understood quantitatively. A computational model that calculates the switching field of a magnetic fine particle of arbitrary shape, accurately accounting for particle interaction, has yet to be developed. This paper discusses some of the problems in constructing numerical models that address this problem from first principles.