RE: Shottky Diodes in SSTC ---> Not A Necessity

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

Original poster: "Jan Wagner by way of Terry Fritz <teslalist-at-qwest-dot-net>" <jwagner-at-cc.hut.fi>

Hi,

On Mon, 17 Mar 2003, Tesla list wrote:
 > Original poster: "Stephen Conner by way of Terry Fritz 
<teslalist-at-qwest-dot-net>" <steve-at-scopeboy-dot-com>
 >
 > At 08:08 17/03/03 -0700, you wrote:
 >
 > > > 3. MOSFET body diodes are EXTREMELY SLOW. An IRFP460 MOSFET body
 > > > diode has a rated reverse recovery time of: TYP=580nS and MAX=1200nS.
 > > > This is like hours in an SSTC circuit. Most ultrafast diodes blow
 > > > body diodes away in Trr (rev. rec. time), hence no need for a series
 > > > shottky to isolate the two diodes.
 > >
 > >Current doesn't choose the fastest lane, but divides into paths
 > >proportionally to their impedance. You'll probably have current flowing
 > >in both body and external diodes.
 >
 > I would say the crux of the matter is the IV characteristics of the diodes
 > in question. I think this might have been mentioned before, but what the
 > heck. Purpose-made fast diodes tend to have a lower forward voltage than
 > the body-drain diode. The difference might not be very great, but the
 > dynamic impedance of both diodes is very low, so the external diode will
 > take the vast majority of the current.

The internal diode won't initially conduct any current (except some
microamps of leakage) as it won't get forward biased. If you're lucky.

I'd agree about I-V characteristics - if the current through the
parallel diode is high enough and the forward drop of the diode thus
exceeds the internal/body diode "band gap" (510mV?), then the body diode
will start conducting.

So you'd have to over-rate the parallel diode a lot, current wise, to
prevent it going into >500mV forward voltage. (1V..3V is typical at _full_
current rating)

With closer to 50% duty cycles for the drive pulses, the
diode currents will be smaller => decent forward voltages

OTOH when the interrupter kicks in and the current through the diodes
rises to t.ex. a couple of 10A, then it still doesn't matter if the slow
internal diodes go into forward conduction - with the interruptor on
i.e. SSTC off, those diodes now have plenty of time to recover. Again, if
you're lucky. ;-)


 > You could help to promote this by
 > choosing your add-on diode to have a low forward voltage and a high current
 > rating. However, you can't guarantee that it'll work unless you use the
 > series Schottky.

Oops, seems i just re-stated what you said...  ;-)


 > P.S. If you do use series Schottkys then they need to be rated and
 > heatsinked to take the full MOSFET drain current. A 2kW SSTC like Justin
 > and Aron's would need TO-220 diodes on small heatsinks. I'm not surprised
 > those 1N5822s are blowing out.

Another possibility is that the ultrafast diodes are a bit slow to turn on
and the schottkys exceed their reverse voltage rating (1n5822 was 40V or
so)

Ok, 3A diodes => 6A H-bridge input => 1.4kW (230vac) or 660W (110Vac)
<=> 2kW/230v=>9A/2=4.5A rms or 2kW/110v=>18A/2=9A rms per diode, that's
pushing those poor 3A rms schottkys quite hard as well... :)

cheers,
  - jfw

--
*************************************************
  high voltage at http://www.hut.fi/~jwagner/tesla
  Jan OH2GHR