ZVS flyback driver SSTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

ZVS flyback driver SSTC

previous 1 2 3 next

Move Thread

LAN_403

Matt Edwards

Wed Nov 17 2010, 02:18PM

Registered Member #2838 Joined: Fri Apr 30 2010, 07:55PM
Location: tehachapi, CA
Posts: 333

What are you using for the power supply?

Arcstarter

Wed Nov 17 2010, 08:37PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

LittleVenture wrote ...

I am really confused... My ZVS flyback driver works awesome for my flybacks. My ZVS flyback driver has 3 out puts of course i was wondering why i need a capaictor i thought it would be just like a big flyback basically. I just wound 3+3 around the base of the secondary and nothing happened i will try some more and if i get it working i will post info

Could someone make a simple schematic please thanks

Follow the ZVS driver schematic... All ZVS flyback drivers require a tank cap, so why would you not need a capacitor? Remember what i said, if it is not in tune, you wont see an output because it is too off resonance. Guessing values won't work well. Also, you will need more than 12v to make any visible discharge most likely.

I know that Jovicakluk on youtube made a ZVS driver running with 15v drive a secondary WAY off resonance and still got discharge, somehow. I guess the secondary was running like a transformer rather than a Tesla coil. His tank cap was .1uf so i know the frequency was way lower than the small secondary with no topload.

dillon

Sun May 17 2015, 05:02PM

Registered Member #54416 Joined: Wed Feb 04 2015, 01:30AM
Location:
Posts: 13

i think zvs flyback driver topolgy is great for a tesla coil. i have 500 nanohenery+ 500 nanohenery primary with a 50 nano farad capacitor in parallel. I get 23" sword sparks from mine.

Wolfram

Sun May 17 2015, 06:21PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Very nice result. Do you have any more information on your setup? Is it half wave rectified, and what voltage are you running at?

dillon

Mon May 18 2015, 01:46PM

Registered Member #54416 Joined: Wed Feb 04 2015, 01:30AM
Location:
Posts: 13

running straight off the wall,and yes it is half wave rectified.the mosfet is FQA19N60. and i have a ltspice file at

Eleccentric

Tue May 19 2015, 01:13AM

Registered Member #33460 Joined: Tue Aug 27 2013, 06:23PM
Location: Seattle
Posts: 46

That is certainly the simplest schematic I have ever seen for such a well-performing coil. What did you use to generate the negative pulse on one of the gates at start-up?

dillon

Wed May 20 2015, 03:38PM

Registered Member #54416 Joined: Wed Feb 04 2015, 01:30AM
Location:
Posts: 13

the startup pluse is just to get ltspice to self oscillate. In reality it oscillates by its self. the mosfets are paralleled with 3 sets, and they are frozen in the freezer before operation. other than that the circuit is 100% accurate. even the coupleing. It does not work well close coupleing. The secondary is 3.5 " and the primary is 6" in diameter.

Eleccentric

Wed May 20 2015, 04:47PM

Registered Member #33460 Joined: Tue Aug 27 2013, 06:23PM
Location: Seattle
Posts: 46

*hearty laugh* You freeze the mosfets before running? How long are your run times before needing a refreeze?

The LTSpice simulation oscillates on its own without the startup pulse, but takes a few milliseconds to start, during which time massive current passes through the fets. Simulating using a more realistic mosfet negates the utility of the pulse - even with it things still take a few milliseconds to run properly.

I'm not surprised you need paralleled devices - the simulation predicts much larger currents than single FQA19N60s could handle.

I have a stack of C2M0080120D fets.. ..I imagine they might work well if I decide to cook up a version of this thing over the weekend.

dillon

Wed May 20 2015, 05:15PM

Registered Member #54416 Joined: Wed Feb 04 2015, 01:30AM
Location:
Posts: 13

it's funny the mosfet never heat up, but I figure the junction temperature is even though the case does not. I can run straight for about 7 seconds , then pause for about 15 seconds and repeat up to about 5 minutes. When paralleling the mosfets you will need about 6-8" of wire length separating each mosfet to the resonant capacitor, or else parasitic oscillations will destroy each mosfets immediately. and your mosfets will need an output capacitance lower than 500 pf , because the primary coil inductance is so low. Oh, I forgot the ballast resistor is at 1.6 ohms in for that picture

Eleccentric

Wed May 20 2015, 11:16PM

Registered Member #33460 Joined: Tue Aug 27 2013, 06:23PM
Location: Seattle
Posts: 46

Interestingly, LTSpice doesn't show parasitic oscillation with paralleled devices, but if I add 150nH in series with each drain, then there is terrible UHF oscillation along with voltage spikes approaching 2kV.

*spends time with the simulation*

Ah, of course. The gate turn-off diodes should run directly to the tank cap, since we want THAT oscillation to control the gate, not some local UHF crud. Is that what you did?

previous 1 2 3 next

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