[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: OLTC maggy
Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>
At 10:10 PM 4/5/2003 +0100, you wrote:
> The kickback problem still needs work but I suspect a bit of PSpice
> simulation might help.
Two things may help this. Quenching 180 degrees earlier or later may tend
to meet the kick back with open IGBTs rather than the antiparallel
diodes. If you can set the quenching time, a little adjustment may help a
lot. I really don't seem to have any problem with kick back. I am a
little careful about quenching when the primary coil is full of current
since the transorbes then get hit with the stored energy as a voltage
spike. So far, the transorbes seem to take these spikes easily, but if
they did fail, the damage may be dramatic.
Another thing that helps is big secondary streamers that eat up the energy
before it wants to go back into the primary circuit. The energy has to go
somewhere and if it does not go into streamers, it will go back into the
>However, I can't be sure whether I would have the time/money/amount of
>2000 amp IGBT bricks required to actually build my creation.
Be carful of the big brick IGBTs. Their internal design is very poor for
high frequency use. They may be fine at 60 Hz, but the poor internal
layout will fry at say 250kHz. You are far better off in using many small
devices that are connected together with high frequency operation in mind.