Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-16T08:28:53.101Z Has data issue: false hasContentIssue false

Electro-vortex flows in a cylindrical cell under axial magnetic field

Published online by Cambridge University Press:  28 September 2022

Peter Frick*
Affiliation:
Institute of Continuous Media Mechanics UB RAS, Perm 614018, Russia
Sergei Mandrykin
Affiliation:
Institute of Continuous Media Mechanics UB RAS, Perm 614018, Russia
Vladislav Eltishchev
Affiliation:
Institute of Continuous Media Mechanics UB RAS, Perm 614018, Russia
Ilya Kolesnichenko
Affiliation:
Institute of Continuous Media Mechanics UB RAS, Perm 614018, Russia
*
†Email address for correspondence: frick@icmm.ru

Abstract

Liquid metal flows, generated in a cylindrical cell by a strong electric current flowing from the central bottom electrode to the sidewall, are studied experimentally and numerically. In the absence of an external magnetic field, the flow is caused by the interaction of the electric current with its own magnetic field and is known as electro-vortex flows (EVF), which has a pure poloidal configuration. It is also well known that even a relatively weak azimuthal force (which immediately arises if a vertical external magnetic field is applied) induces a swirl that completely changes the structure of the flow, suppressing the poloidal motion (Davidson et al., Eur. J. Mech. B Fluids, vol. 18, 1999, pp. 693–711). We study the steady-state modes and analyse the dynamics of transient modes, following the evolution of the flow structure at different ratios of azimuthal and poloidal Lorentz forces, governed by the applied current and vertical magnetic field. In the considered problem, the peculiarities of transient modes, leading to the suppression of the poloidal flow, are associated with the fact that the domain of acting forces is localized in the vicinity of the bottom electrode. The scenario of flow evolution strongly depends on the ratio of electro-vortex and rotational forces (whether the EVF has time to form before a noticeable swirl of the metal, or the swirl dominates from the beginning of flow evolution).

Information

Type
JFM Papers
Copyright
© The Author(s), 2022. Published by Cambridge University Press
Figure 0

Figure 1.Current, magnetic field and forces in the EVF cell with a central electrode.

Figure 1

Figure 2.Photograph and scheme of the experimental set-up (cut view): 1, sidewall; 2, Plexiglas bottom end face; 3, gallium alloy; 4, cathode; 5, anode; 6, Helmholtz coils; 7, UDV transducers. The beam direction of the UDV probe is indicated with a dashed magenta line.

Figure 2

Figure 3.Computational domain (a) and UDV probe positioning (b).

Figure 3

Figure 4.Instant (a–c) and time-averaged over 1000 s (d,e) and 500 s ( f) fields of poloidal velocity in the $xOz$ plane in a pure EVF flow (no external magnetic field) at the electric current $I=50$ A (a,d), $I=250$ A (b,e) and $I=750$ A (c, f), which correspond to the values of EVF parameter $S=4 \times 10^{5}$, $S= 10^{7}$ and $S=10^{8}$, respectively. The magenta line indicates the direction of the UDV beams.

Figure 4

Figure 5.Reynolds numbers $Re$ (points) and $Re_t$ (black crosses) versus the EVF parameter $S$. The experimental data for $Re$ are shown by green points for the solid upper boundary and by blue points for the free surface. Numerical data are indicated by red points ($Re$) and black crosses ($Re_t$).

Figure 5

Figure 6.Experimental profiles of the mean vertical velocity $v_z$ averaged over the UDV probes 1–8 for different values of the electric current $I$ for the EVF in the cell with a free surface (a) and a solid upper boundary (b). Here ${B_{ext}}=0$.

Figure 6

Figure 7.Experimental profiles of the mean vertical velocity $v_z$ averaged over UDV probes 1–8 for fixed electric current $I=1000$ A and different magnetic fields ${B_{ext}}$: free surface (a); solid upper boundary (b).

Figure 7

Figure 8.Experimental and numerical profiles (NS) of the mean vertical velocity $v_z$ under weak external magnetic field. Solid upper boundary.

Figure 8

Figure 9.Instant fields of the poloidal (a–c) and azimuthal (d–f) velocities in the $xOz$ plane in the absence of an external magnetic field (a,d), in the magnetic field ${B_{ext}} = 0.05$ mT (b,e) and in a strong vertical magnetic field ${B_{ext}} = 1$ mT (c, f).

Figure 9

Figure 10.Poloidal and azimuthal energies as a function of the external vertical magnetic field. $I=1000$ A. Numerical simulations. An extended version of figure 9 from Kolesnichenko et al. (2020).

Figure 10

Figure 11.Mean azimuthal velocity at $r=0.9R$ versus the external magnetic field. Here $I=1000$ A.

Figure 11

Figure 12.Standard deviation (SD) of the poloidal and azimuthal energies versus the external vertical magnetic field. Here $I=1000$ A.

Figure 12

Figure 13.Reynolds number defined through the maximal vertical velocity along the UDV beam vs the applied vertical magnetic field. Results of numerical simulations (red) and experiments with solid (green) and free (blue) upper boundary. Here $I=1000$ A.

Figure 13

Figure 14.Schematic of four kinds of poloidal flow in the cylindrical cell: a pure EVF (a); flows induced by the homogeneous radial current in a vertical magnetic field in the cell with a free surface (b) and with a solid cover (c); a flow associated with the velocity profile, observed in the EVF under a moderate vertical magnetic field (d). Corresponding sketches of vertical velocity profiles along the UDV beam (which is indicated by dashed magenta lines) are shown in the left-hand subpanel.

Figure 14

Figure 15.Evolution of the energy of poloidal and azimuthal flows under weak (${B_{ext}} = 0.01$ mT, panel a) and moderate (${B_{ext}} = 0.1$ mT, panel b) magnetic fields after the electric current $I=1000$ A is switched on.

Figure 15

Figure 16.Flow energy as a function of the external magnetic field, $I = 100$ A.

Figure 16

Figure 17.Evolution of the energy of poloidal and azimuthal flows without an external magnetic field (pure EVF, panel a), ${B_{ext}} = 0.003$ mT (b), ${B_{ext}} = 0.1$ mT (c) and ${B_{ext}} = 1$ mT (d) after the electric current $I=100$ A is switched on.

Figure 17

Figure 18.Wavelet spectrograms of the poloidal energy fluctuations at saturated state. Here (a) $I = 1000$ A, ${B_{ext}} = 0.01$ mT, (b) $I = 1000$ A, ${B_{ext}} = 0.1$ mT, (c) $I = 100$ A, ${B_{ext}} = 1$ mT.

Figure 18

Figure 19.Evolution of the EVF flow at $I=100$ A and ${B_{ext}} = 0.1$ mT. The poloidal (a,d,g,j,m,p) and azimuthal (b,e,h,k,n,q) velocity fields in the $x0z$ plane, and the angular velocity along the cell axis (c, f,i,l,o,r) for six time moments, $t=6.2$, $60$, $92$, $122.1$, $160.6$ and $478.6$ s.

Figure 19

Figure 20.Evolution of the EVF flow at $I=100$ A and ${B_{ext}} = 1$ mT. The poloidal (a,d,g) and azimuthal (b,e,h) velocity fields in the $x0z$ plane, and the angular velocity along the cell axis (c, f,i) for three time moments $t = 5$, $25$ and $300$ s.

Figure 20

Figure 21.Reynolds number $Re$ versus the external magnetic field for different inclination angles ($\alpha =0$ corresponds to the vertical magnetic field, $\alpha =90^{\circ }$ corresponds to the horizontal magnetic field).

Figure 21

Figure 22.Instant (a) and time-averaged (b) fields of poloidal velocity in the $xOz$ plane in a horizontal magnetic field ${B_{ext}} = 0.1$ mT. The averaging time interval is 200 s.