Disturbances produced by an electric potential probe on MHD flows in rectangular ducts

Abstract

Experimental data for local velocity in liquid metal duct flows exposed to an external magnetic field can be obtained from measurements of electric potential differences recorded by probes that are moved along the channel width. These instruments, known as conduction anernometers or Liquid-metal Electromagnetic Velocity Instruments (LEVI), have been used in the past preferentially for investigating almost fully developed magnetohydrodynamic (MHD) flows in poorly conducting ducts and flows with smooth variations along the channel axis. For such applications, where electric current density is negligible, the probe was assumed to give reliable results and the potential gradient signal was directly interpreted as a velocity measure. If the flow varies along its path on very short length scales, like in ducts with abrupt change of cross section or in manifolds, non-negligible 3D electric currents occur so that the LEVI readings may become inaccurate. Moreover, the presence of the probe itself may perturb significantly the flow field due to the formation of internal layers that develop along magnetic field lines tangential to the shaft and the insulating body of the instrument. Experiments showed an asymmetry in the distribution of the measured transverse potential gradient and its underestimation compared with the one expected from flow rate measurements and from theoretical predictions for fully developed MHD flows. A numerical analysis of MHD flows around a probe inserted in a rectangular duct has been performed to support the physical interpretation of potential measurements and to study and quantify the influence of the instrument itself on the readings. A calibration procedure is suggested, which allows using measurements of potential differences to get reliable data for velocities in the duct.