Re: Sheppard-Talyor circuit up on my website/ low Z construction technique

["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>
> 
> Hi Rob,
> 
> > Now I bet that the FETs will explode with violence :) instead of just
> > burning out.  Gee, then I don't have to probe them with an ohmmeter.
> 
> Yup, they just might. This exploding thing must be genetical or something,
> it seems all mosfets have that behaviour preprogrammed. :-)
> 
> Anyway, there's a small but transistor-deadly mistake in your schematic!
> 
> You should have a ultrafast recover diode from the node between D24 and
> R36 pointing towards ground, that is, clamping positive voltages to
> ground. And that diode is missing, at least in your schematic.
> 
> (initially hard to believe maybe, but that's the right way round for the
> diode - the Sheppard Taylor is a bit odd as the second stage is fed from
> the storage cap with a negative voltage, not positive)
> 
> Without the diode, positive going voltage appearing on the transformer
> primary (secondary side fed by the TC) is not clamped to any rail. Via D24
> and D25 it may make the drain of M17 see an overvoltage, causing it to
> fail. I don't know why your simulation worked ok, though... maybe zener
> effect of the mosfet drain-source?
> 
> The diode works a bit like the flyback diode in flyback converters.
> 
> Hope this helps! And saves your expensive mosfets from further failures...
> 
> cheers,
>  - Jan
> 
> --
> *************************************************
>  high voltage at http://www.hut.fi/~jwagner/tesla

Hi Jan,

Yes, the normal DC-DC converter has this diode.  However, D24 isolates
isolates the new diode from the drain/L node.  i.e. it would be two back
to back diodes and I'm not certain if it will help keep FET voltages
down under mistuning/light loading. I will do the sim and implement it
if it helps.  Thanks.

I found yesterday that with only 30V input my converter produced 400V at
the drain/L junction :)   The FREDs got extremely hot and one blew out
(thermally).  Also, for {C} I am now using two 27uF 450V caps in series
and these got very hot as well.  I could go back to two 0.056uF 1600V
polypropylene caps in parellel for energy storage.  Or I could put one
poly cap in parallel with the electrolytics to reduce their dissipation.

I think I am going to test at 12V this time.  From what I've seen, this
converter has an extremely wide input voltage range.  It would be cool
to have a 12V Tesla coil driver.  I think this was the original novelty
of the Sheppard-Taylor converter-  that it could regulate under wide
variations in load and input voltage with closed loop PWM.  Rob.

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

www.pythonemproject-dot-com