Homebuilt plasma arc speaker

Maybe, I dunno. I'll be buying another one and ripping it apart.

It was a dimming electronic NST, and operated at 20kHz for the bright and 40kHz for the dim. Splitting the difference, I guess. I forget the drive circuitry but IIRC it had two FETs and I think the low voltage secondary wires were for a feedback type thing. I'll but another and rip it open. I just ordered some 320W 24V and 12V switching supplies from Meanwell for this project, time to get some flybacks and ignition coils (though I think I will have much more luck with the flyback types).

FWIW I observed the behavior of the transformer as described above just driving it with the frequency generator directly. Its output voltage sagged a lot when I hit resonance even though the secondary voltage went way high. I don't know how it would behave with a low Z high current capable source.

Q: how could the transformer alone exhibit resonance like that? Could the inter-winding capacitance have been big enough to make it behave like that?

I would have thought that they added a capacitor but it might be tacked on to the transformer in this case.

Also a transformer will have two resonances: the primary and secondary.

Yeah ignition coils do resonate at somehere from 5 to 10 Khz.And they have a pretty small output when not in resonance.They should work well FM modulated but there is going to be a very loud sqealing sound from it(The carier)

So i would recomend useing a flyback they normaly resonate at about 20 to 30 Khz.I guess even better would be a air core transformer since that would resonate at 100s of Khz.Thats basicly a tesla coil.You can also thurn a ignition coil in to a air core transformer by takeing it apart and removeing the iron core.Note that this is a very messy job since ignition coils are filled with oil for insulation.

Oh and recently i built a IGBT high freqency driver.And tested it out on a ignition coil.It made white hot and long arcs....for 5 seconds and then the electrical tape(my temoprary fix for arcovers) arced over and made a pretty big fire.I should have not thrusted electrical tape in the first place.Now im solving the problem with a garden hose and lacquer.Videos of it on youtube if anyone wants to see il give a link

speakerguy79 said:

I've worked with ignition coils before. I could get a specialized drag racing coil to do 10kHz but the primary wasn't being fully charged. It was $250 for that single ignition coil, so I doubt there's much better out there. The more common ignition coil I had on hand was limited to a few kHz.

Click to expand...

Wish I saw your post before I built mine. I got a nice (noisy) arc using an ignition coil, but the arc always died when I pushed the frequency over 1.5kHz.

As for frequency modulation, I used a 555 timer in astable mode and then fed the audio signal into pin 5 to modulate the frequency. It looks fairly clean on the oscilloscope, but I haven't been able to test it since I never got a usable arc.

Any recommendations for a cheap flyback transformer that I can get on ebay? (link?) Thanks in advance. Otherwise I might try taking apart that ignition coil and removing the core.

You can just get the cheapest flyback.

There is also a nother way, you can wind your own high freqency transformer.Since they dont realy need a lot of thurns. like 5 thurn primary and a 500 thurn secundary will work.The trick is to wind i on a plastic tube but you have to wind it from one side slowly to the other, NOT going from one to the other side because it will arc over.Then you can paint it with varnish or like some clear laqour for wood to further keep from arcovers.Stick a ferite core inside with a small thick primary on it and your done.

Winding my own coil doesn't sound too fun. I'll try to grab one from ebay first and see how it goes. Thanks for the help.

actually i found a site with a working plasma speaker and complete schematic with instructions to wind the proper coil for it aswell.

https://www.hvlabs.com/plasmasonic.html

i just built it using a TV flyback as coil, and it does work and produces clear audio without distortion. only problem is the TV flyback coil must be of the wrong windings to work with the secondary windings he suggests at his site, so my plasma speaker would love to use moar then 11 amps at 24 volts but the powersupply i use will only output 11 amps at 30 volts on 2 terminals so i suppose if i parallell couple them id get 30 volts and 22 amps. but i have too small heatsinks on the mosfets so they will burn for sure

so i think i will increase the number of windings on the secondary of the flyback coil from 8 windings to 16, which should lower the current used and increase the voltage? and put some larger heatsinks on the mosfets

the proper HV coil for this project is a total of 6400 windings on a ferrite core custom built by 2 separate ferrite cores from tv flybacks :O

but anyway this speaker design works

Boy oh boy it downloads slowly. I have waited for about 15 minutes for it to download. Ho hum.

Guess what? It sounds horrible!
The plazma tweeter produces 5kHz to who knows how high, then the sub-woofer produces from about 200Hz down to who knows what. All the frequencies between 200Hz and 5kHz are missing.

Boom boom and tweet tweet. It ain't musical.

Thanks for the link.. guess I'll give that one a shot