[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: SSTC, xfmr gate drive oddity



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

Hi all,

What I tried now was switching places of mosfets in the old compl. source
follower circuit, as shown in the CD4049 datasheet "internal schematic".
Now the p channel pulls up to 13.6V. The n channel pulls down to GND.
Common-drain connection. This did help _hugely_ to make overcome the
mosfets V_th threshold voltage problem and make the output rail-to-rail
and work like a charm waveform wise, but now there's new grief with
cross-conduction. About 20A peak during the ~50ns switching time =>
mosfets rather hot. A small series inductor would restrict this current,
but sadly the gate xfmr peak output current too => not a very usable
solution. :o(

>Looking at your drawing I dont see any thing to drive the gates 
>out of phase. the voltage at the capacitor remains the same with 
>out regards to current. The circuit requires the drive signal 
>turn one on as the othes is turned off. I only see both driven 
>at the same time. No voltage swing, only current change. The 
>capacitor sees no change.

Umm... not sure... my garbled-ASCII schematic is trying to depict a
"complementary source follower" (as in "emitter follower") which isn't a
"totem pole drive" and doesn't require an additional inverted signal.
Until a year ago I too thought that "totem pole" is just a synonym for
"complementary emitter follower". Ain't so. :)

But anyway, you did give a good idea here! 
Totem pole seems the way to go.

So then... My 50%-duty-always drive signal has to be split into two
signals, inverted and non-inverted, with additional dead time in between
to prevent this cross-conduction. That calls for an el-cheapo LM3525 or
other PWM IC clocked by the existing circuit, right? And then finally let
the PWM IC output drive two n channel 10A / 60A-pulsed mosfets in a totem
pole setup. 

Yeah talk about complicated... 

Is there a simpler way? 
Any advice'd be highly welcome!

(Already considered push-pull drive for the gate drive xfmr but lack of
dead-time would be a problem there too. Driver ICs with 6A or 9A peak
"only" seem still poor wrt a 60A peak discrete driver... well maybe after 
a few hours more of trouble and pain such ICs seem a good deal after
all... oh well let's see ;o)

many thanks,

 - Jan

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