DRSSTC frustration / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

DRSSTC frustration

first previous 2 3 4 5 next

Move Thread

LAN_403

Steve Conner

Sat Sept 29 2007, 07:44PM

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

Hey Marko, glad you got it working. To photograph single bangs I've used a fancy digital SLR that can be cranked up to 3200 ISO, and then post-processed with Photoshop or IrfanView's gamma adjustment.

Marko

Sat Sept 29 2007, 08:13PM

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

Hi steve

OK, sort of working.. I'm still very afraid about what would happen if I cranked it up?

I can't judge anything by waveforms I see. I'm already close to IGBT SOA, how can I know where will they blow up once I'm far above it?

I also can't see to what is my OCD set to, I didn't pull out any outputs I can hook a voltmeter to.

Now I'm unsure should I build a new dimmer or not. I would definitely go for a larger ST circuit because it's unbeatably rugged, and stand higher power dissipation. Without any fine circuitry it should be free of high-impedance nodes and bomb/tesla coil proof.

I need to get heatsinks for the SCR and resistors which I'm out off.. and would need to build some kind of box for it.

Marko

Danielle

Sun Sept 30 2007, 05:13AM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

For the IGBTs I was using the same 30N60s and pumping over 900A, 1200A is where it failed but that was because an arc came around from the top avoided the strike ring and hit the bridge directly! I would say you can safely run them if you have really good cooling up to 800A. run it at 140VAC in so 400VDC and you can get some 70+ inch sparks.

Steve Conner

Sun Sept 30 2007, 12:34PM

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

marko wrote ...
how can I know where will they blow up once I'm far above it?

The only way to know for sure is to destroy hundreds of IGBTs with overcurrent and do statistical analysis on the results, in conjunction with a computer model of the IGBT's guts. That was probably how the manufacturer did it, and it would need done all over again with a lower target lifetime to give you the answer you want with any degree of confidence.

No hobbyist that I know of has blown up enough devices under properly controlled conditions to give an answer that I would take seriously. The closest we have to an answer on IGBT overdriving probably comes from people like Greg Leyh at Stanford, who really did do some of the things I explained above and published papers on them. But his conclusions only hold for the same IGBTs that he investigated, and don't necessarily apply to other makes and models.

Hence I have always said that if you need guaranteed reliability, you shouldn't go outside the datasheet ratings. You just don't know what will happen when you exceed them: it could last 20 years or fail tomorrow.

Marko

Sun Sept 30 2007, 04:59PM

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

OK, OK, level of scariness is now exponentially increasing.

After messing with the coil for a while I felt it is time to hook it up to full mains voltage.

I lowered the on time to <100us, and carefully increased the OCD setting a bit. (A bit because I don't know what is it really set to.

The streamer was absolutely intimidating, like 60..70cm long, flailing and incredibly loud. It did not although hit the 1 meter target.

OCD was actually limiting the on time to just few cycles, current would hit like 500A before being turned off. When I wanted to photograph the scope camera's battery died, and coil was overwhelmed in massive flashovers some of which started more like racing arcs, from half of the secondary and then hit the primary.

I have never seen this kind of flashover before, and I don't know how to protect from it.

It would probably do it even if I had a conical primary. I raised the scondary for like 1,5cm, it helped a bit, but not enough.

What can I do to stop the flashovers?

Another bad thing I noticed only on the pic, is the arcing on the bridge again. It seemed to stop after I separated the secondary base ground, but now it's back.

I'l post more pics as soon as my camera charges.

And I need some consolation before continuing this...

Reaching

Sun Sept 30 2007, 06:16PM

Registered Member #76 Joined: Thu Feb 09 2006, 10:04AM
Location: Hemer, Germany
Posts: 458

mhh, where do you get the flahovers on the bridge? that would make me nervous. is it from a wire to your heatsink or inbetween traces?
You really have to solve it befor trying to crank up the whole thing. although looks great, nice work!

For racing sparks i would just try to raise the secondary another 1,5cm or more until they stop

Marko

Sun Sept 30 2007, 06:45PM

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

Now I thought worst part of my frustration is over, but then comes even worse.
I tried to run the coil again and something blew up as I turned it off. One UCC went short.
I have no clue what could it be yet, everything happened too fast. Some IGBT's may be gone although bridge is not short and caps hold the voltage, I can't tell until I disassemble the bridge COMPLETELY.

Failure happened as I turned the interrupter off, and it could be linked with bridge arcs or secondary flashovers. Current was mere 500A at that point.

Arcing between the bridge taces is a very old story.

It's all due to completely idiotic bridge design I'm so ashamed off. Although I have reduced RF impedance between those traces to near zero they still arc, I don't have clue why.

They arced much more while I used mains ground for my secondary base ground.

The worst thing is that I can't do ANYTHING on that part. Arcs between nodes connected with so low impedance (I've put mid-point grounded 0,68uf caps in between) between them.

Why these happen is a big mystery to this time, I would be very grateful if anyone could give explanation.
Arcs specifically happen between current transformer ground trace adn positive +325V plane.

I can't by any practical way fix or insulate the traces.

I would simply have to change the entire bridge layout and throw this one to bin.

By changing the layout I also assume putting new, proper electrolytic caps inside, and this requires completely changing all the coil's internals, driver board, faceplates, everything.
It is a ''bad plan that cannot be altered''.

My primary is whole other story. I have absolutely no clue why am I suffering such bad flashovers. They come from like half of the secondary, How can I help to stop these?

I would have to raise the secondary for like 10cm for that.

Stupidest thing is that I'm using only 2 turns of my original primary and rest serves to attract flashovers.

I need to build a new primary too if I really want to do anything.

So, what I need to do is practically throw this coil into trash and build new one. Save the driver board, secondary and toroid.

Where does this torture end? All I want now is to have this coil burst into flames and finally be gone forever.

Now this makes me want to cry

Marko

Finn Hammer

Mon Oct 01 2007, 06:25AM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Marko wrote ...

I have never seen this kind of flashover before, and I don't know how to protect from it.

It would probably do it even if I had a conical primary. I raised the scondary for like 1,5cm, it helped a bit, but not enough.

What can I do to stop the flashovers?

Marco,
I have 6 similar sized DRSSTC`s, all equipped with Steve Conners PLL driver.
_One_ of them repeatedly produced similar flashovers from the bottom 100-150 mm of the secondary, to the top of the conical primary. These flashovers are then 100 -150 mm long.

I might have a clue to the situation that produces them.

I tuned the primary to the low pole of the combined system.
Now, if the driver locked on to the low pole, it would arc.
If it locked on to the high pole, it would stop arching.

I have not read the full thread, so I don`t know how you have tuned your primary circuit, and I have no clue to how you can force a feedback system to operate at a desired pole, but at least here is something you can investigate.

Cheers, Finn Hammer

Steve Conner

Mon Oct 01 2007, 11:02AM

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

Marko, why so sad? You built a DRSSTC and it works, that is something to be proud of. The flashovers and exploding components are all part of the fun

About the mysterious arcs on your H-bridge, can you draw the schematic of your power board, draw a lightning bolt on it or something to indicate the two nodes that it's arcing between, and post it here?

Finn: Mjolnir does the same thing too, and I never quite figured out why one pole/mode made it flash over but the other didn't. My best guess is that the two resonant modes actually have different voltage profiles on the secondary.

In the upper mode, the induced voltage in the secondary due to the primary current is 180' out of phase with the voltage due to the secondary's own current in its self-inductance. This makes the voltage gradient shallower at the bottom of the secondary, where the primary's magnetic field is strongest.

In the lower mode, the induced voltage is in phase, so the voltage gradient is steeper at the bottom.

Don't even ask how the induced voltage can transfer power when it's 180' out of phase with the current. I don't know, but the math says it works, the experiments say that it works, and I have experimental evidence that the primary and secondary magnetic fields are out of phase on the upper pole.

The ability to choose one pole or the other is something that only my PLL driver can do, AFAIK. A self-oscillating driver will snap to whichever one oscillates most readily.

The poles/modes thing is really mostly of concern at startup when you're trying to build up voltage to ignite your streamers. Once a good streamer load has developed, the loading flattens everything out and the exact drive frequency doesn't seem to matter so much.

Tom540

Mon Oct 01 2007, 02:59PM

Banned on 3/17/2009.
Registered Member #487 Joined: Sun Jul 09 2006, 01:22AM
Location:
Posts: 617

Marko,

Whenever I tried raising my secondaries to get rid of flash over all it seemed to do was raise current and have bad coupling. I just started using a wider primary instead. I always make my primaries 2 inches wider than the secondary. Might be worth a try.

-Tom

first previous 2 3 4 5 next

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