Comparison of the magnetization decay due to crossed or rotated magnetic field measured on a Ferromagnet disk/GdBCO pellet structure

Superconducting (SC) bulk materials are commonly used as permanent magnets in synchronous rotating machines. When the machine is in operation, the bulk materials that are placed in the machine rotor can experience a certain misalignment with the magnetic field generated by the stator. This situation is likely to happen during transients or non-ideal behavior of the machine: the angle between the stator field and the SC magnetization can change because of a sudden change of mechanical load or vibrations of the device. The consequence is that the SC pellet can be subjected to a transverse component of magnetic field which is likely to cause a significant demagnetization and therefore a decrease of the machine performances.

In previous works, we already have studied the problem of a “pure” transverse magnetic field (also called “crossed field”) as well as the beneficial influence of a ferromagnetic (FM) disk placed on the top of the SC pellet. To this aim, a bespoke insertion tool for a PPMS was designed and constructed. The tool can be used to characterize a SC sample whose orientation is changed from a parallel to a perpendicular direction with respect to the magnetic field direction. Here, this insertion tool was used to rotate the sample at various angles regarding to the field direction.

A GdBCO pellet of 9 mm in diameter and 5 mm in height was characterized at 77 K in various experimental conditions by measuring the magnetic induction below the centre of the sample surface. We compared the results obtained with the GdBCO pellet alone to those with a FM disk of various thicknesses attached on the top of the sample. A first set of experiments consisted in applying pure transverse magnetic field cycles of 50 mT and 100 mT at low frequency, the polarity of the field being either always the same (unipolar) or reversed after half of the period (bipolar). A second set of experiments was carried out by rotating the applied magnetic field (applied first in the pellet magnetization direction) in several steps to a given angle (up to 90°) and back to 0°. All these experiments were carried out after having fully magnetized the sample and recorded the magnetization decay due to the flux creep during 30 min.

The results show that (i) the FM disk reduces the magnetization decay in both pure transverse and rotated magnetic field experiments, (ii) rotated magnetic fields have a less detrimental effect than the “pure transverse” magnetic field. This phenomenon is likely to be caused by a partial remagnetization of the sample when the angle of the magnetic field is reduced back to 0°. Finite element modelling was carried out to sustain the measurements of the local magnetic induction and gives additional information on the supercurrent distribution inside the SC pellet during “pure transverse” and “rotated” applied fields.