MHD Simulation with Insulated Conducting Boundaries and Circuit Interactions
Author: George J Marklin
Submitted: 2005-12-21 17:46:29
Co-authors: T. R. Jarboe
Plasma Science and Innovation Center, University o
Room 120 AERB, Box 352250
Seattle, WA 98195
The advantages of insulating plasma from its surrounding conductors have been known since the early days of theta pinch research. In a theta pinch, this insulation allows flux to enter and spread out uniformly over the surface of the plasma, producing a symmetrical pinch. Modern ICC experiments, such as those using inductive helicity injection and rotating field current drive, use insulation to allow flux from multiple circuits to enter and form complex patterns of rotating E&M fields over the surface. The HIT-SI geometry is used to develop a new method of handling such boundary conditions for 3D simulations.
Previous MHD simulations of HIT-SI [V.A. Izzo and T. R. Jarboe, Phys. Plasma 12, 056109, (2005)] have assumed conducting walls with high plasma resistivity in a thin boundary layer to approximate the effect of the insulator. A more elegant method will be presented which implements the insulated conductor boundary condition exactly, without any boundary layers, by solving 2-D Poisson like equations for the electric and magnetic potentials on the surface. Circuit interactions are easily included by imposing appropriate jump conditions on the potentials to represent applied loop voltages or surface currents. Since the extra potential equations are only solved on the surface, which is a small subset of the total computational domain, the overall runtime is not significantly increased.
Resistive magnetic diffusion of a Taylor state shows self-similar decay with an insulated conducting wall, maintaining J parallel to B. With bare conducting walls the parallel current quickly decays away and is replaced by perpendicular current, taking the system out of equilibrium. Some initial low resolution simulations of HIT-SI injector formed spheromaks will also be presented if time permits.
E10 - How can MHD codes be made more predictive?
Please place with other PSI-Center talks or posters (at end).