First results from 2-D simulation, 3-D simulation plans
Lawrenceville Plasma Physics, Inc. started an ambitious large-scale computer simulation project at the beginning of January, in collaboration with Dr. David Rose of Voss Scientific and Dr. John Guillory, emeritus professor at George Mason University. This project will also use open-source software developed by Dr. Robert Terry. First results are in…
The aim of the first year of the simulation is to use an innovative approach to simulating the run-down phase of the plasma focus process and the formation of the plasmoid-- the dense, self contained ball of plasma where the fusion reactions take place. No simulation so far has shown exactly how the plasmoid is formed and what conditions are optimal for transferring energy efficiently into the plasmoid. If the simulation can succeed in doing this, researchers can use it to guide the design of electrodes in future experiments. Once an experiment is started, the simulation will as well be useful in understanding the experimental results.
After the first year, a further goal will be to simulate the plasmoid during its brief lifetime as it burns the fusion fuel. Other goals may include simulating the ion beam that the plasmoid emits and seeing how to optimally transfer energy from the beam into a circuit for energy output.
The first results have already been obtained from a 2-D particle-in-cell (PIC) simulation. The initial simulation covered the first 15 ns of the plasma focus pulse. The rapid rise in ionization and the formation of an ionized sheath along the insulator between the cathode and anode is clearly seen in the simulations. (See attached graphic and brief film. In the simulation outputs, the density of ions is symbolized by color, with red the highest density. )
The simulation team has made plans on the basis of the first successful run for a more extensive 2-D simulation of the Speed-2 dense plasma focus, which LPP hopes will soon be operational in Chile. Changes to the software will allow a roughly five-fold speed up allowing us to run about 80 ns per month.
This is still slow, as the full run-down and compression lasts 400 ns, so LPP is actively seeking additional processing power to speed the insulations. Currently we are using a 2.8 GHz Intel Xeon “quad core”, in other words four processors each running at 2.8 GHz. If anyone has access to a faster machine, LPP would appreciate learning about it. (send an email to ) The processors must work in close synchronization, so this is unfortunately not the sort of program that can be farmed out to many unlinked processors.
The simulation team is now planning the initial phase of the full 3-D simulation. This phase will simulate the formation of the plasma vortex filaments, tiny whirlwinds of plasma that start to concentrate the plasma and its magnetic fields. The initial 3-D simulation will model as short section of a pair of filaments as they form, using the conditions from the 2-D simulation as initial input.



