A detailed analysis of the crosstalk in an uncompensated gapped-core current transducer is presented in this paper. A cut-core current transducer with a magnetic field sensor in the airgap is widely used to measure current in industry and in laboratories. Crosstalk is the effect of a nearby current-carrying conductor on the reading of this type of sensor. We present a study of the dependence of the crosstalk on the position of the external conductor, the core material, and the core geometry, including the number of airgaps. A 3D Finite Element Method (FEM) based model is used to analyze the crosstalk, and the results are compared with measurements.Using a low-permeability ferrite core with two 2 mm airgaps and a single Hall sensor results in a maximum crosstalk error of 18 %. This error can be reduced to 1 % either by using a differential Hall sensor pair, or surprisingly by using a single 4 mm airgap. This error can be further reduced to 0.15 % by using an FeSi core with larger permeability. However, this type of sensor is very sensitive to the position of the Hall sensor in the center of the airgap. Displacement or an angular mismatch can increase the error to 1.5 %, as demonstrated on a commercial sensor.
Read full abstract