Re: Sheppard-Talyor circuit up on my website/ latest

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

Original poster: "rob by way of Terry Fritz <twftesla-at-qwest-dot-net>" <rob-at-pythonemproject-dot-com>

Tesla list wrote:
> 
> Original poster: "rob by way of Terry Fritz <twftesla-at-qwest-dot-net>"
<rob-at-pythonemproject-dot-com>
> 
> > Just an observation: it looks like using a sledgehammer to crack a
> > nut. There must be something fundamentally wrong, either with the
> > circuit or the way it's being implemented. No?
> >
> > Regards,
> > Malcolm
> 
> Yes, actually these supplies are designed to be run under closed loop
> SMPS control.  With light loading the duty cycle is very small and
> stresses are decreased.  When the load is increased the controller ramps
> up the duty cycle.
> 
> I am trying to get a design that works open loop.  I've found that the
> diodes are the culprit.  The FETS dissipate little power. Using real IR
> HexFreds has made a substantial difference.  IR has an app note on how
> not all 35nS turn off diodes are alike.  I now believe it.  One diode in
> particular went from too hot to touch to just warm after putting in the
> IR model.
> 
> I'm doing a lot of spice simulation to see if I can reduce diode
> losses.  Its a challenge.
> 
> Rob.
> 
> --
> -----------------------------
> The Numeric Python EM Project
> 
> www.pythonemproject-dot-com

In the sim world, I can put a small value inductor 1uH to 10uH in series
with the energy storage cap, and it drastically reduces diode current
spikes.  The penalty is higher drain-source voltages on the FETs. 

Unfortunately, I blew out my last UC2710 FET driver.  I ordered some
more plus some of the Microchip 9A CMOS drivers.  What happened is that
via my own stupidity I disconnected the gate drive wire, so my
unregulated supply went up to over 20V which blew the device. I am going
to put a 18V zener in there.  My TLC555 has its own 5V regulator, but I
didn't have enough drop out margin under load for an 18V regulator.

These 1kV FET's have a lot of gate capacitance.  One thing I didn't know
before, is that operating them at very low voltages drastically
increases the gate-source capacitance.  The IR datasheet has lots of
nice plots on the subject.

Rob.

-- 
-----------------------------
The Numeric Python EM Project

www.pythonemproject-dot-com