By "not successful" do you mean that the calculations do not explain reality very well? Should you then consider both inelastic and elastic collisions?
IIRC the way electrostatic cooling works is a bit different, and it would negate your attempt to model it as based on gas collision rates anyway. In fact there are apparently two kinds, and neither is understood. One of them is written up on Rex: http://www.rexresearch.com/blomgren/blomgren.htm This does not answer David's question, however - not that I really understand it. That answer, as to why there could be less dissipation in practice than calculations would indicate, be may involve a percentage of inelastic collisions i.e. collisions in which kinetic energy is not conserved. But anyway, the subject of a second kind of electrostatic cooling is possibly a niche that only works in space. Thus, it was "black" for a good reason. The general idea would be to "force" the Edison effect onto what can be called a "reluctant cold cathode" which can be extremely cold to begin with but you need it to be extremely cold. This could be done via a high positive charge grid effect - so that you get the heat removal via electrons "boiling off" of an already cold material. Of course, it is possible to view the other kind of polarity effect (Blomgren) with a high negative charge being directed as a 'wind', as an effective variant of the Edison effect where there is a novel heat transfer of some kind to the external HV electron wind ... but needless to say it will not work well if there is no ground :) Jones
