Sensors and Actuators A 106 (2003) 104–107 Demonstration of a new method for magnetic flux measurement in the interior of a magnetic material George Loisos , Anthony J. Moses Wolfson Centre for Magnetics Technology, Cardiff University, P.O. Box 925, Cardiff CF24 0YF, UK Abstract A method is proposed for magnetic flux measurement in localised regions inside a magnetic material using a novel surface probe sensor. The principle of operation is described and its potential use in steel products is experimentally verified. © 2003 Elsevier B.V. All rights reserved. Keywords: Magnetic flux density measurements; Magnetic material characterisation 1. Introduction The extreme non-linearity between B and H in a ferro- magnetic material has prevented the exact calculation of the internal eddy currents or flux density. Despite the good ap- proximations by assuming linear BH relationships it has always been desirable to have a method or sensor, which could give an indication of the actual flux density variation inside a material. This would provide a better understanding of eddy current distribution inside materials, which for ex- ample is necessary for evaluating and attempting to reduce losses in magnetic materials. The proposed method is applicable to plates, strips and laminations and also to any other shape of ferromagnetic ma- terial but for convenience the theory and experiment refers to a typical lamination of electrical steel. The well-known search coil (Fig. 1a), the conventional probe sensor [1] and the modified probe sensor (MPS) (Fig. 1b) [2] are the three most common sensors for global and more specifically localised flux density measurements in laminations. However, these sensors only give an indica- tion of the total flux in the material enclosed by the search coils or the needle probe wirings whereas it is useful to have a measure of the localised internal variation. The modified probe sensor is more accurate than the con- ventional probe sensor [2]. It can further be modified to include diagonal positioning of the conductive probes (i.e. PCB contact probes) (Fig. 1c) and it will be proved theoret- ically and shown experimentally that it is possible to have Corresponding author. E-mail addresses: loisosg1@cardiff.ac.uk (G. Loisos), mosesaj@cf.ac.uk (A.J. Moses). an induced voltage proportional to the flux density rate of change in a triangular cross-sectional area such as shown in Fig. 1c instead of the traditional orthogonal cross-sectional areas. 2. Principle of modified probe sensor A schematic diagram of a modified probe configuration is shown in Fig. 2. The output voltage V A from the set of probes at positions A and A is given by V A =- B ∂t d A 2 - B air ∂t d A air + V AA (1) Similarly, the output voltage V B from the set of probes at positions B and B is given by V B =- B ∂t d A 3 - B air ∂t d A 3 - B air ∂t d A air + V BB (2) where A 3 = A 1 + A 2 ; A 1 is the area enclosed by the path ABB A , A 2 the area enclosed by OAA O and A 3 the area enclosed by OBB O . If we subtract Eq. (1) from (2) (|V A | < |V B |) the resulting voltage e output is equivalent to a voltage induced in a one-turn search coil enwrapping the cross-sectional area ABB A (that is, area A 1 ) plus an error due to air flux. This is given by, e output = V B - V A =- B ∂t d A 1 - B air ∂t d A air (3) since it can be assumed that V AA = V BB . 0924-4247/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0924-4247(03)00144-4