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

Re: [TCML] Off grid, medium size, 12V battery powered TC



Steve,

Sorry for the late reply. I fill in at a small Fire Department sometimes and had a 72 hour shift.

I can look around for low input imedeance MOSFETS, but is the layout going to be super cranky?

I'd love to play with the IGBTs. I've never used 'em. I've built a couple of switchmode DC to DC converters that make around 200V, for nixie tubes, but nothing that can handle more than 50mA or so. Any links on makind a converter that has enough beef?

The thing for me is I am self taught. I;m good at building, have a couple of 'scopes and know how to use 'em. However, without a design to work from, I'm lost.

Going to go give your site a good look-over. Not only would I love to build one to play with, but it would be fun to take it to the local high school. I'm always worried that I'm going to backfeed nasties into their mains. It's never happened, but I'm always a little relieved when the show is done and everything at the school still works.

Jonathan

----- Original Message ----- From: "Steve Ward" <steve.ward@xxxxxxxxx>
To: "Tesla Coil Mailing List" <tesla@xxxxxxxxxx>
Sent: Thursday, December 30, 2010 10:35 AM
Subject: Re: [TCML] Off grid, medium size, 12V battery powered TC


Jonathan,

Without doing a serious investigation (via simulation and part hunting), i
suspect there is some viability in making a DRSSTC operate from 12V nominal bus. You'd need a tank impedance that is roughly 10-20X less than a typical
DRSSTC that might make 2 foot sparks.  For example, a "small" DRSSTC might
use something like a .05 to .1uF tank cap at 150khz Fres, which is a tank
surge impedance of 10-20 ohms, so for similar performance from 12V, you'd be
aiming for a ~1uF tank capacitor.  The current would be about 10X higher
than typical, so something in the 3000-5000A peak range.  The things to
investigate are 1) can you make a 1 turn primary to tune with a 1uF tank
cap?  2) can you parallel enough MOSFETS (note i specify MOSFETS and NOT
IGBTs due to the voltage loss) to handle the 3-5kA peak currents with
reasonably low loss?  Even at 1m ohm, that would be a 3-5V drop which is a
sizable fraction of the bus voltage, so its best to get it down to <100u
ohms or so per switch.  Should be feasible with big groups of low voltage
(40V?) high current MOSFETs, but is it reasonable for you??

A more realistic solution would involve stepping up the voltage to at least
150V i'd say, but if you are to build a solid-state converter, you may as
well step up to 350VDC where 600V IGBTs would be perfectly happy and deliver
plenty of power.  But this becomes a new project in itself.

Another, better option still, may be to design a 12V to 12kV converter.
Using a full-bridge of MOSFETS switching 12V into a ferrite step-up
transformer to get about 6kV peak. This could feed a voltage doubler to get 12kV, which should be plenty for a SGTC of the variety you seek. The thing
only has to convert a few hundred watts, so it shouldnt be terribly
difficult, though even at 12V this could be 50-100A range, so not trivial.
There are some special "topologies" (or types) of converters that are
specifically useful for charging capacitors, i discuss one of them (the SLR
converter) on my website, though its designed for 350V to 12kV to charge a
5kJ pulse cap at a few kW peak. The principals are similar and require some
scaling of components to achieve proper operation at just 12V input.

http://www.stevehv.4hv.org/ccps1.htm

I could offer more help if you thought this was a suitable route for you to
take.

Steve




It's a little discouraging though. A lot of work to make some sparks. Such
a bummer there is no way to use the battery bank, so much power there and
easy to charge back up. And I know that if Terry was here, he could figure
out a way. ;-)


Jonathan

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla