#5151
belbear
Participant

Henning wrote: I believe you’re thinking of inverters (DC->AC), not of rectifiers (AC->DC). That’s actually the beauty of FF, because you already get a inverter for free. Output capacitors are already required, otherwise you’re getting only impulses of a few milliseconds.
Or are you’re thinking of eliminating those output capacitors? And replacing them with what? Inductors of the same size? Not much gained (but might be cheaper).

Why not eliminating a separate output storage alltogether? Why not let the beam and X-ray pulses recharge the input capacitors directly using a passive high-power transmission line toward an additional set of switches on the input capacitor bank?

This way you only need to siphon off and convert to AC the net power surplus (which hopefully will be there) from the pulse-recharged input capacitor before initiating the next shot, and not having to deal with all the recycled power through your DC converter. You may need to convert the pulse to a higher voltage using some sort of pulse transformer, which can be a passive device.

When running “under-unity”, the recharger (needed for startup anyway, but can be much less powerful) needs to add only the deficit to the input capacitor, not the entire charge. This can be useful when fusion energy (beam or X-ray) is to be used for something else than electricity generation.

It will be a complicated mechanism to design, so probably only possible for a second-generation machine, but it can save a great deal of $ and bulk on power electronics when commercial competition starts to heat up.

In any case, you need a type of inverter that can take a sawtooth-shaped input voltage (>40kV, at FF pulse frequency) and block-shaped input current (constant current, briefly interrupted by the firing sequence), and produce tri-phase sinusoïdal output voltage and current, synchronized with the grid frequency (50 or 60Hz)