Re: electronic PWM variac

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: Brady Hauth <bhauth@xxxxxxxxx>

On further thought, you can put a high current low inductance inductor
in series with the pole pig, and then put a small PP snubber cap in
series with a high current low inductance inductor across that
assembly, with the IGBT connecting between the two high current low
inductance inductors. Properly tuned, the IGBT will charge the
capacitor through one inductor and run the pole pig through the other.
 The capacitor will then push the charge back through its charging
inductor, according to the standard rules of LC resonance. The current
through the inductor in series with the pole pig then comes from the
capacitor for a little bit, at which point you can do a soft switch
with the IGBT. The two inductors then start charging the cap in an LC
resonance with the inductors and the pole pig in series, charging the
cap and dropping the current in the inductors. The IGBT then turns on
again, and since everything in front of it is through an inductor
there's no significant energy loss.

One thing I'm not sure about is how much of a problem core losses in
the inductors would be.

>  > Putting in a small capacitor across the pole pig will prevent heat
>  > buildup in the IGBT.
>
> That would be true, but then you get heat in the cap instead, which is
> probably worse. Stupid me. If you were using enough power to have a
> problem with hot IGBTs, you'd need to put a inductive element in with
> a resistor of something that would give you a voltage differential
> less than the IGBT's rating with your current draw when pulsing on
> running through it across it that can handle dissipating that much
> energy.