Alternative high voltage coupling - sparkgap to solid state. / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Alternative high voltage coupling - sparkgap to solid state.

1 2 3 4 next last

Move Thread

LAN_403

3l3ctrici7y

Fri Feb 11 2011, 07:49PM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

I am working to build a high frequency arc starter.
See here, here, here, here.
The first one is a terrible example, but has a nice picutre of the basic thing. The youtube vids, I think were based on the first link; running the capacitor discharge thru the mains ground.

Such devices are basically a spark gap Tesla coil, but with a very small number of secondary turns. In my case, I have 6 primary turns and 18 secondary turns with a ferrite core. It does work, but is very erratic and tends to lose performance as it is used. I don't think the capacitor is up to the task. After killing two doorknobs I made my own, and killed that too. The writing on the wall; it has been seen. While simple, it just doesn't work, and there are no prizes for the simplest failure.

Mine is based on a GM HEI ignition coil driven with a 555 and FET gate driver driving the gate on an IGBT to switch the iggy. I have a 100 or so nF cap on the iggy primary (this makes a huge difference on the output). The output is in series with four spark gaps in series at .008" each. The spark gaps are in series with the doorknob capacitor, a 5 (or so) ohm resistor, and the primary of the HF transformer where the primary is 6 turns and the secondary is 18 turns with a ferrite core.

Going to plan "B", I was thinking about some sort of a solid state Tesla coil version of this. Perhaps drive it using the ZVS FBT driver circuit. Ultimately the secondary is formed using thick welding wire with 4-6+mm diameter; so having a large number of turns is impractical.

I don't really have a specific question, other than being interested in whatever sage wisdom anyone would like to offer.

Thanks.

Conundrum

Fri Feb 11 2011, 07:55PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4059

homemade litz wire?
Tools required:- One drill, a large room and about 200 feet of 30 gauge wire would back and forth to make a 15 wire bundle.

-A

3l3ctrici7y

Fri Feb 11 2011, 07:58PM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

litz wire?? What is litz wire?
Oh, I see.. Wiki has the answer.

I get the idea I must be doing something wrong and litz wire will help me.
What is that??

Well, I assume it would be used on the primary since the secondary is what it is.
At the moment, I am using 16 or 14 gage stranded copper wire for the primary.
Is the skin effect causing problems for me here?

According to powerstream, the frequency for solid wire is between 6 and 11 kHz. I am not sure how much the stranding will change that. But that is far lower than the hundreds of kHz or few MHz that I expect this is oscillating at.

Ash Small

Fri Feb 11 2011, 08:10PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

3l3ctrici7y wrote ...

I am working to build a high frequency arc starter.
.

I built one using a single transistor flyback circuit, homemade caps, a spark gap tank circuit, and ten turns primary and secondary. It works really well.

You need to choke the HF off from getting back into the buzzbox, though, I used a core from a 50 Hz toriodal transformer (has to be large enough to take the thick diameter welding flex) and some home made caps (about 200nF total).

3l3ctrici7y

Fri Feb 11 2011, 08:18PM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

That's interesting. What I have seen done, and what was planning on doing, is using a large AC capacitor across the "buzz box" to allow the HF to bypass it.

What kind of capacitor did you use for your HF unit?

Ash Small

Fri Feb 11 2011, 09:10PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

3l3ctrici7y wrote ...

That's interesting. What I have seen done, and what was planning on doing, is using a large AC capacitor across the "buzz box" to allow the HF to bypass it.

What kind of capacitor did you use for your HF unit?

The cap across the buzz box will probably work on it's own, I just added the choke for extra protection. I think it was ten turns for each cable, wound so they oppose each other.

I used a home made cap for the HF tank circuit, 20nF if I remember correctly. I didn't really do any maths as I couldn't measure the inductance of the primary. I was going to do a bit of 'trial and error', but it worked so well I left it as it was.

I wound the air core transformer on a polypropylene tube that effervescent vitamin C tablets come in, after removing the paint with scotchbrite or wire wool (It was ten years ago so I can't remember which). I used multi-stranded wire for the primary, and welding cable for the secondary.

Polypropylene is supposed to be the best stuff for HF air core transformers.

3l3ctrici7y

Fri Feb 11 2011, 10:21PM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

Interesting :)

That was kind of my approach also; I chose the 4700pF since I was getting two I could put them in series to get half that, or in parallel for double, and I had seen something where 4700pF was used for this.

20nF huh? That's huge. The first link I posted uses 500pF, and the second uses 2500pF. I did have it working with 4700pF until the cap died (or so I presume)

If I try building a cap again or getting another doorknob, I think I am going to go smaller. The iggy runs out of breath and cannot charge 4700pF and break down four gaps at .008". It can only break two reliably. But hey, it stopped working all together as the caps died, so who knows.

On that page I read a while back where a 4700pF cap was used, the author said that two spark gaps are much better than one. This is my experience also. Those two projects I posted both use four gaps. As I have come to realize from reading, multi smaller gaps will quench sooner than a single equivalent gap and this causes a steeper dI/dt in the primary and thus a greater voltage spike. So using a smaller cap should (I hope) allow me to use all four gaps and perhaps increase the dI/dt and get better output. I can always crank up the drive frequency so whatever energy I lose using a smaller cap, I can make up for in frequency.

Here are a couple of youtube videos showing some impressive HF output.

The second one even has corona when it's not arcing :)

Ash Small

Fri Feb 11 2011, 10:45PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Yes, I realise that 20nF is large, but I thought bigger was better (more charge). It worked so I left it as it was.

I can't be certain that it worked as it is supposed to, and that the secondary isn't just running at the same frequency as the flyback, but the spark across the spark gap is much brighter than the arc from the flyback was during testing, so I assume that the tank circuit is resonating as expected.

I only used one spark gap, an unsuppressed spark plug (eg model T ford)

I think the bigger cap is more robust, I still plan to try different values but haven't got round to it yet.

I tried experimenting with multiple gaps but found one gave better results. I assume this is because my tank circuit oscillates at a lower frequency so quenching time isn't a problem. (I understand a bit more about this now than I did ten years ago)

I've not watched the videos as I'm on a mobile ISP that charges by the gigabyte.

3l3ctrici7y

Fri Feb 11 2011, 11:27PM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

I have some nice BC Components caps floating around that i had thought of making an MMC out of, but I thought that it would still be too much capacitance, but since 20nF is working along with a single gap, then maybe it's worth a try.

Also, in the first set of youtube vids I linked to, at least in the final version, it looks like he has the cap in across with the iggy rather than the single spark gap. I've seen this topology before, but it seemed improper, so I put the gap across the iggy. I wonder what effects this would have. He also used a ferrite yoke.

Ash Small

Fri Feb 11 2011, 11:51PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I had the cap across the flyback, with a resistor between the output of the flyback and the tank circuit. I can't remember the value offhand, but it was 33 something (approx) Either 33 k ohm, 330 k ohm, or 33 M ohm. I used about 12" of suppressed spark plug lead, (the carbon cored stuff) so that's a guide. (I think it was 33 k ohm)

I agree that 20 nF seems a bit high, but my reasoning was that it would be more robust than a smaller cap. I was planning to experiment, but it worked so I left it.

What was the ferrite yoke used for?

1 2 3 4 next last

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