Re: Sheppard-Taylor SSTC driver

["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: "Jan Wagner by way of Terry Fritz <twftesla-at-qwest-dot-net>"
<jwagner-at-cc.hut.fi>
> 
> > I was wondering if anyone else has ever tried this type of driver for
> > SSTC use.  I am trying here.  The Sheppard Taylor circuit is unique in
> > that is doesn't require any charge reservoir on the DC supply.
> 
> As in the following?
>  http://www.eie.polyu.edu.hk/~cktse/pdf-paper/PE-9809.pdf
>  http://www.en.polyu.edu.hk/~cktse/pdf-paper/PE-0001.pdf
> 
> A rather odd form of a boost converter, but it looks very interesting
> as a single-switch converter or PFC pre-reg...  :o)
> PFC would be simple with that one.
> 
> The continuous mode buck converter following that PFC section probably is
> the TC pri itself?  I.e. where did you hook up your TC primary coil? (or
> base drive xfmr?)
> 
> Figure 2 doesn't seem applicable for TC sec drive, as the output section
> is in discontinuos mode...? Also, the TC sec will be driven only for half,
> or shorter, of one cycle of f_res. Or am I missing something there?
> 
> > That is how my spice simulaton looks.  However, in real life I keep
> > blowing out just one of the FETS.  I think I need to make a new board
> > with much better grounding :)    Rob.
> 
> Quite likely. :) Both switches should see the same stress so if just one
> fails always then you've an layout issue there. The switches aren't
> soft-switched but switched under rising current and voltage so the
> turn-off might cause some nasty ringing and stuff.
> Well, at least if I interpreted the schematic and figure 13 correctly,
> anyway... ;)
> 
> good luck!
> 
> cheers,
> 
>  - Jan
> 
> --
> *************************************************
>  high voltage at http://www.hut.fi/~jwagner/tesla

Hi Jan,

I found through exhaustive simulations that the FETS are only under low
stress during low duty cycle and/or heavy loading of the output.  With
170VDC in, the FETS, cap, and diodes can see as much as -> 800V.  Of
course in a real converter there is an IC translating loading to duty
cycle.  I am just using 50% duty cycle which blows up one 600V FET and
nearly blows the other.  I just ordered another supply of various 600 to
1000V FETS,1000V FREDS, non-inductive resistors, plus lots of
polypropylene caps, so I should be in a position to study this some more
in a day or so.  Gee, experimentation can be expensive :)

Right now I'm using L1=160uH on powdered iron core (will not saturate),
and an energy storage cap (Cs) of 200uF.  I can move the 200uF cap to
the power supply and go with a 1200V 0.1uF energy storage cap.  And it
looks like a snubber will easily lower the FET d-s voltage down to 600V
or less.  Once the SSTC is loading down the isolation transformer, then
voltage will drop to safe levels and the snubber doesn't have to do much
work.

Anyway, I started this project as I got bored with half-bridges,
full-bridges and the normal stuff.  Other people have perfected those
designs and I've built them before, so I wanted to try some different
converter types.

Rob.

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

www.pythonemproject-dot-com