SSTC question / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

SSTC question

1 2 next

Move Thread

LAN_403

vasil

Sat Aug 05 2006, 02:31PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

It would be possible to have , for a SSTC, only 2 IGBTs, as for a half bridge, but still full voltage on the primary (not V/2 as for halfbridge), something like this:

http://www.geocities.com/gemaria_ro/sstcex.html

It would use two primaries, oposite wounded ( a bifilar cable).

Marko

Sat Aug 05 2006, 03:17PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

I tried such a thing not so long ago. It won't behave as a normal bridge SSTC, but rather as double ended flyback SSTC and will suffer from big voltage spikes on primaries.

WIthout class E operation you probably won't be able to increase input voltage to more than some 50V or etc. for 600V igbt's (with best adjustment of pprimary turns and impedance match).

vasil

Sat Aug 05 2006, 03:31PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

Pitty...i would like to have some good results with only two IGBTs

second variant?

http://www.geocities.com/gemaria_ro/sstcex1.html

Each primary tuned with a paralleled C, which will short the unwanted primary spikes....?

Marko

Sat Aug 05 2006, 03:42PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

As I said, this topology can work very well in class E. Jus tput your resonant caps across the switch instead having them across the primaries.

For higher frequencies (>1mhz) IGBT's are bad choice because they are usually slow at turnoff.
Prolonged current tailing can cause them to suffer extreme losses (death) due to some energy from shunt cap getting discharged trough them while they are supposed to be 'off'.

vasil

Sat Aug 05 2006, 03:50PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

I dont make high frequency SSTCs (usually under 200 kHz), so it would not be a problem. Maibe at lower freq, the voltage spikes will be lower (your coils are high freq Firkragg).

Its hard to match a class E when you have no tools. Hmm...will see....

Steve Conner

Sat Aug 05 2006, 10:53PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Your circuit looks like a variation of the centre tapped circuit used in the Mazzilli flyback driver. (except without the resonant capacitor, so it's a bit more like the original Royer oscillator)

So I'm going to guess that the same conditions apply, and the switches see 2*Vdc plus spikes if you don't use a tuning capacitor, and roughly pi*Vdc if you do use one.

I guess what I'm saying is that the switches end up seeing twice the voltage applied to the primary, so you might as well use a half-bridge instead and double the DC bus voltage. I think the phasing of one of your primaries is wrong too.

...

Sat Aug 05 2006, 10:59PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

Couln't you buid it like a push-pull vttc, replacing the tubes with your igbt's? Use steves vttc faq's eqations to get a decent Q for your primary, then tune for resonance, and viola

Steve Conner

Sat Aug 05 2006, 11:03PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

It worked for Vladimiro Mazzilli (the guy who invented our "ZVS Flyback Driver") back in 2003

Marko

Sat Aug 05 2006, 11:38PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Your circuit looks like a variation of the centre tapped circuit used in the Mazzilli flyback driver. (except without the resonant capacitor, so it's a bit more like the original Royer oscillator)

So I'm going to guess that the same conditions apply, and the switches see 2*Vdc plus spikes if you don't use a tuning capacitor, and roughly pi*Vdc if you do use one.

Yeah, this variation of 'double-ended' drive can also work in ZVS by putting a cap across the primary aka. mazzili driver.
Without any caps, with that topology I had IRFP450's (500V) starting to avalanche at some 30V input. Not good..
Due to fast switching very high voltages develop across the primary and there's nothing to drain that energy out.

(vasil already owns one coil based on it).

I saw his schematic quite long ago, and it is actually interesting how it works on mains voltage with just 5W gate resistors.

I tought this could make nice little unregulated SMPS (with mini mosfet's requiring much less gate drive power than IGBT's). Only if peak voltage could be reduced to some 2x Vin

vasil

Sun Aug 06 2006, 06:50AM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

The resistors are boiling hot after 1 min runing....
I run out of IGBTs stock. It would be nice to have only 2 IGBTs...Seems that would not work.

The primaries phasing is wrong yes.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob