Two sources of cross-field error in racetrack fluxgate

This contribution covers cross-field response of fluxgate sensors with a flat, single-sheet race-track core, which to our knowledge has not been studied yet. In the case of race-track shape, it was commonly believed that the error is automatically lower than for the ringcore due to the shape anisotropy. Instead we observed that the exact geometry of the race track (i.e., ratio between the longitudinal and upright section) and the position of the coils start to play a significant role, not only the core homogeneity—there is no more symmetry as in the case of a ring core. We found the cross-field response dependent on defining the perpendicular angle, either to the sensor’s core axis or to the axis of the feedback coil; these two axes were misaligned by up to 1°. When arranging the cross field to be perpendicular to the feedback coil axis, we were able to verify results previously published in literature: we obtained 60 nT response for 50 μT cross-field.