Sure, that's all fine... if you don't care about ever getting to net power. If you do, the grid is a major problem. You cannot produce megawatts of power with a hunk of metal in the center.
I'm still not sure why you think Coulomb scattering is a bigger problem. Where are you saying the energy goes?
It goes into heating up the background matter. Plain, and simple. That's what all the 'glow' is about.
The question is whether you really ever could end up with a situation in which there is so little 'slow' stuff in a Polywell that all the bouncing around is all radial. Thing is, you've got ions moving in and
out and if you get them colliding just nanometers aways from the dead centre then they will start scattering into a Maxwellian distribution, and that's presuming you've got the impossible 'perfect' vacuum, into which you never get any
sputtered products off wall bombardment, &c., &c..
But that's to confuse the specific question of the grid.
You should still be able to get net power at mW (not counting support equipment, of course), if the device works as intended. The grid is a bar to very high input power - agreed, no challenge - but it is not the
problem for getting net power, it's about 10% of that problem, according to my understanding of Wisconsin/simulations. I would've thought it was less than that myself, but will happily pay due respect to those who have doen the experiment.