Re: [TCML] Gate drive transformer help

[Tyler <tyler.c.c@xxxxxxxxx>]

The spike you describe at the beginning of the switching may be ringing of
leakage inductance in the secondary with the gate capacitance. Try adding a
few ohms of gate resistance if you haven't, this usually damps that out.
Also, I was working on some GDT's this weekend and also experienced the
height falling, and was using an 0.47uF blocking capacitor. I removed the
blocking capacitor entirely (don't do this unless you really understand
your driver and what the blocking cap is for!) and now I get a perfect
square waveform at the gate with zero ringing. A blocking cap is not needed
in a DRSSTC in principle, assuming proper operation and a perfectly tuned
duty cycle. But it is a good safety margin to have it there, especially for
initial testing.

Tyler


On Mon, Oct 31, 2011 at 12:17 PM, Scott Bogard <sdbogard@xxxxxxxxx> wrote:

> Hi Mike,
>     What I am seeing on the scope is the wave form is hardly square, it
> starts and bumps up high, then comes down; at higher frequencies this is
> where it ends, but at lower frequencies it then cuts out, then comes back
> up at about 3/4 height and ramps down, breaking a single pulse into two.  I
> have a lot of off axis inductance, the thing was gator clipped together,
> and subtracting 3 out of the 17 turns made no difference...  When I turn
> down the duty cycle, the two bumps merge, and form one distorted but stable
> bump, I can run really low frequencies at less than 10 percent duty
> cycle...  I need a gate drive transformer because I am going to be running
> this thing off of 300V DC once all is said and done (always though using a
> step down transformer to drive a step up transformer was ludicrous...)  I'm
> not using a resistor from the driver output to the transformer, and my dc
> blocking cap is .68uf I believe...  Not even sure if it is necessary for
> proper operation, tried doubling this and halving it, no appreciable
> difference.  Thanks for your input guys!  Oh and my turns ratio for the
> xformer is 1:1:1...
>
> Scott Bogard.
>
> On Mon, Oct 31, 2011 at 8:34 AM, Michael Twieg <mdt24@xxxxxxxx> wrote:
>
> > If you're seeing "distortion" only at lower frequencies then that
> probably
> > means you don't have enough magnetizing inductance.  Your driver may be
> > drooping due to excessive magnetizing current, or the core is saturating.
> > Though without a better description of the "distortion" it's hard to say.
> > Just try adding a few more turns and see if it helps.
> >
> > -Mike
> >
> > On Sun, Oct 30, 2011 at 6:26 PM, Scott Bogard <sdbogard@xxxxxxxxx>
> wrote:
> >
> > > Greetings all.
> > >     Built my gate drive transformer for my flyback driver, oddly enough
> > > when scoping it it works great at higher frequencies (30+kHz) but
> > distorts
> > > below that, except at low duty cycles.  So my questions is do I need a
> > > different core material, possibly a larger core, fewer turns?  I have
> one
> > > larger core, and also several smaller ones, the larger core is already
> > > wound but seems to only work at much higher frequencies in that
> > > configuration.  Also tried doubling and halving the dc blocking cap,
> > didn't
> > > change much...  Any advice would be greatly welcomed.  Also the
> waveforms
> > > are a bit distorted even when they are stable, up until very high
> > > frequencies (100kHz ish) I still think I could use them this way but
> > seems
> > > to me changing parameters should fix that as well, at least my my
> > intended
> > > operating range of 10-100 kHz...  Thanks.
> > >
> > > Scott Bogard.
> > > ______________________________**_________________
> > > Tesla mailing list
> > > Tesla@xxxxxxxxxx
> > > http://www.pupman.com/mailman/**listinfo/tesla<
> > http://www.pupman.com/mailman/listinfo/tesla>
> > >
> > _______________________________________________
> > Tesla mailing list
> > Tesla@xxxxxxxxxx
> > http://www.pupman.com/mailman/listinfo/tesla
> >
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla