Re: Hvguy-dot-com Feedback SSTC and New Stuff!!!

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "K. C. Herrick by way of Terry Fritz <twftesla-at-qwest-dot-net>" <kchdlh-at-juno-dot-com>

It's a good point, about the symmetric gate drive:  With tight
drive-coupling all-around, via just 1 transformer, turning-on FETs can't
start to do so until turning-off ones are off (gates near 0 V).  But that
needs to be a good transformer.

Ken Herrick

On Wed, 13 Nov 2002 16:55:56 -0700 "Tesla list" <tesla-at-pupman-dot-com>
writes:
 > Original poster: "Jan Wagner by way of Terry Fritz
 > <twftesla-at-qwest-dot-net>" <jwagner-at-cc.hut.fi>
 >
 > Hi Justin,
 >
 > and congrats again! :)
 >
 > Ken C Herrick wrote:
 >  >  > And another thing: be wary of having both FETs in each half of
 > the H
 >  >  > being on at the same time!  You might want a bit of
 > crossover-
 >  >  > control in the circuit to avoid that.
 >  >
 >  > We keep this in the back of our mind, for sure! A commercial
 > design
 >  > would never use a single gate transformer to directly control
 > four
 >  > (or two) MOSFET's in a bridge, and we realize that.
 >
 > Well, actually... ;)
 > Main criteria for commercial xfmr mass manufacture is the usual one
 > which
 > discerns "a good, working design" from a not so good one, i.e. total
 > cost.
 >
 > All windings on one toroid is fine, and has some minor benefits in
 > this
 > SSTC case, but the more challenging work to build these (properly,
 > anyway
 > :) isn't necessarily worth it.
 >
 >
 >  > But, at the same
 >  > time, it works! I imagine there is some degree of shoot-through
 >  > current (we should measure it) but it just isn't hurting
 > anything.
 >
 > As the FBSSTC has this symmetric gate drive (-x V .. +x V), and
 > drive
 > windings are well coupled, and as there's just one gate resistor,
 > there
 > shouldn't be any shoot through. That's because the mosfet threshold
 > voltage together with the far-negative bias adds plenty of dead
 > time, and switching hasn't any chance to overlap.
 >   => it isn't a problem
 >   => adding dead time is not required, way simpler circuit :)
 >
 > At least, my experiments give around ~50mA current draw from mains
 > when
 > switching at no-load. That was for a 300 kHz 350VDC IRF840
 > fullbridge.
 > 50mA can be explained by all the mosfet output capacitances being
 > switched.
 >
[snipped]