zapkitty wrote:
errrrrr…. if multiple coils are to be considered part of the learning process then that would seem to imply that we are already sure about that much regarding the structure of the alpha recovery system. .. are we?

Thus far I have only scratched the surface of the design criteria. The coil designs need to be modeled with a magnetics simulator to hammer out number of coils, number of turns, number of windings per coil, wire diameters, etc. Particularly, the question of whether this transformer would benefit from a high-permeability, low-reluctance core has not been studied at all. It is possible that it is not necessary to keep the coil inductance down to nanohenry range, since the production reactor will be doing ~1 k shots/s.

I am pretty sure that each coil must be high quality, implying very low series-resistance, implying fat wires, or possibly even superconducting. I am expecting that it will be high voltage, relatively low current, in comparison to the input pulse. (ie: megavolts), but not certain.

Electrical engineers assure me that transformer coils are regularly >95% efficient. But they warn that, to design a transformer without solid understanding of the physics will fail. We will probably end up running thousands of simulations before construction. And maybe make small-scale mock-ups.