Robthebob wrote:The bird talked to Charles Chase, and he was very very realistic about the machine. Their major losses are electrons out of open field lines, into the support structure, etc.
To me that suggests the machine uses a polywell style potential well to contain ions. Otherwise ions would be escaping out the same ways electrons are. I will note that several of the simulations I've run based on the skunkworks configuration have developed a net negative charge on the plasma with nothing specific to introduce extra electrons. The ions just escaped faster until the charge differential developed.
I have vacillated between thinking the Lockheed effort was a straight forward adaptation of core Polywell concepts, to seemingly confounding manipulations along lines that have proven failures in the past. The idea of Loffle bars, baseball seams, etc. The basic three ring narrow line cusp single center high Beta, potential well hints are all Polywell derived (either derived or developed in parallel), with the possible deviation from the polyhedral magnet arrangement. The effects are the same. The exchange for perhaps greater line cusp losses if there are significantly fewer point cusps available for losses may give similar final performance in the final volume, density versus loss picture. The additional external magnets to change the shape of the confined plasma and the line cusp escaped plasma seem to introduce poor edge stability conditions (concave fields), and offer substantial increased oppertunities for ExB diffusion issues. If these external flows are dedicated primarily to modestly improving recirculation of electrons, fine and good. If the additional magnets are to confine high density plasma, and flow it around magnets, then it seems fraught with know flaws.
If it is intended to function very similar to a Polywell, then significant proof of concept studies have already bene done. Perhaps the two most important is the deep potential well, and high Beta Wiffleball performance, with decoupling of ion and electron confinement modalities. There may be different design advantages- direct conversion may be simpler, and the choices for electron and ion/plasma injection may be modified. This last consideration seems to be the main challenge facing EMC2 currently.
Basically, most of the physics is the same but engineering and operational issues may be different. The interesting effect of converting the central B field null/ minimum from a dumbell to a point to a torus configuration might allow for some interesting (and useful?) oscillations in the core plasma. This combined with POPS considerations opens further windows and complexity- good for job security
on plasma manipulations.
To error is human... and I'm very human.