"Electric flame" HF VTTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

"Electric flame" HF VTTC

previous 1 2 3 next

Move Thread

LAN_403

Tonskulus

Sat Aug 29 2009, 06:21PM

Registered Member #1223 Joined: Thu Jan 10 2008, 04:32PM
Location:
Posts: 133

Uzzors wrote ...

This VTTC might interest you.

Yes its cool looking indeed but we had to use 3-phase input to get enough power. 5-10kW...

For candle looking flame, 0.5-1kW should be enough. I suggest tubes like PL519, 811A, 4-125A or similar power tubes.
Microwave oven transformer is also good choise as plate power supply. Plasmatweeter circuit is the best one for that purpose.

Arcstarter

Sun Aug 30 2009, 01:27AM

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

Tonskulus wrote ...

Uzzors wrote ...

This VTTC might interest you.

Wait, how could you use an 811A? How do you get enough feedback for a lower gain triode? I have tried that before, and got nothing. No hot plate, nothing got warm. That would be very cool.

Also, i found some sweep tubes on ebay for a good price, going to bid tomorrow. Good specs and everything.

EDIT: Disregard everything i said above. I remembered i *did* one of these things before with an 811A and 833C. About an inch flame on each, i used the 811 with halfwave and took it to the 833C to try fullwave. Thanks to Myke for reminding me :D. I used magnetic, not capacitave feedback, a tiny bit of grid leak which despite never seeing a schematic using that, worked fine. It was base fed, and around 30 or 40 turns of 16 gauge magnet wire. I am not sure if that is right, i calculated 30 something Mhz. I am going to make one tonight, and i will report back.

Oh yea, and the 'magnetic' (it may have still been capacitive, just more capacitance) coupled feedback was around 6 turns of 28 gauge, about 1/2 inch under the secondary. That was definitely not ideal, and the grid leak was something like 6nf and 100 ohms. Anything different made the tube heat with no output, this circuit does not need grid leak. Just experimentation, and i now have a bit more knowledge. The 'plasma flame vttc' is a strange creature. I know nothing really about them...

Tonskulus

Sun Aug 30 2009, 09:56AM

Registered Member #1223 Joined: Thu Jan 10 2008, 04:32PM
Location:
Posts: 133

811A just came to my mind but yes, it is triode so it may not be the best solution for pt.cuircuit. So lets forget that, tetrodes and pentodes are maybe better to be used with capacitive feedback.

However, I made that system using PL519 and it was just ok. It gave maybe 20mm flame arc...

Arcstarter

Sun Aug 30 2009, 05:51PM

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

Tonskulus wrote ...

811A just came to my mind but yes, it is triode so it may not be the best solution for pt.cuircuit. So lets forget that, tetrodes and pentodes are maybe better to be used with capacitive feedback.

However, I made that system using PL519 and it was just ok. It gave maybe 20mm flame arc...

Yes, but you can sure use a triode. But pentodes have greater power/size ratio anyway, and they will work with no tinkering with feedback. But it seems to be easier to find a powerful triode than pentode, until you get up to the multi KW levels with radiators for heatsinking. Then it is about even :P. 811A's are very cheap, i got two for 24 bucks, and they work fine.

I am trying to take the filament winding out of this MOT now to rewind it with some 24 gauge for the 811A filament. May heat a bit, but it is rated at 3.5 amps.

UPDATE: I made a 10mhz res freq resonator, and made a ghetto 811A socket (Important tip, on the cheap 3 dollar harbor freight meters, the tiny PCB with the 3 input jack on them are spaced perfect for the 811A filament sockets (mine, at least), i cut one of and used the two on the board for the filament/cathode, and an RCA female as the grid). I started with 5 turns feedback over a 2 inch PVC tube around the 1.5 inch 40 turns of 24 gauge 10mhz resonator. I took one turn of the feedback away, the tube heat 4 times less with 2 times bigger sparks, took another away, and even bigger sparks and less heating, which is where i am not. ~1 inch really cool and 'smooth' sparks. The sound is also strange. I use a proper filament transformer for the filament supply, and a MOT on 80 volts for B+ halfwave rectified by 4 1n4007's, and RF block inductor with around as much inductance as the resonator.

Video soon, a bit more tweaking to do.

Arcstarter

Tue Sept 01 2009, 05:43AM

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

I know that with a high enough frequency (several MHz?) and CW operation, I can get this. So the questions remain, how high frequency and how much power to create a candle-like flame. I saw a "lighter" VTTC which created just that, a HF plasma flame.

Well, 10Mhz would pretty much do it. Mine does not do it so well, the sparks are a little more unstable and occasionally it is like streamers instead of a lighter. Not alot of work to make them though.

A few turns (20-40) of 24 gauge will get you suitable freq. I would go for like 24 gauge and a 1 inch by 1 inch resonator. That will get you about 24Mhz, plenty enough for the candle like spark, and lower than the frequency that many popular tubes are rated for at full ratings.

Here is my plasma torch 'thing' (we need a permanent name! Not a Tesla coil at all!)

UPDATE: I took another turn off, so i now have a single feedback turn. Less heating, so now i use 120v input. Much bigger sparks.

Good luck.

Dr Hankenstein

Sun Sept 06 2009, 06:14AM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

Does anyone know what kind of stinger one would use for the discharge point on these HF coils. Looks to me like it's just the end of the wire.

Steve Conner

Sun Sept 06 2009, 09:59AM

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

May I suggest "fire tweeter" as a name

The stinger needs to be something that won't melt.

Plasma Lover

Sun Sept 06 2009, 06:13PM

Registered Member #1911 Joined: Mon Jan 05 2009, 06:30PM
Location: Salem, Oregon, USA
Posts: 165

Steve McConner wrote ...

May I suggest "fire tweeter" as a name

The stinger needs to be something that won't melt.

Fire tweeter would be a wonderful name, as long as the setup used a triode and had no audio modulation (in my opinion), since this would not give the hiss that is so closely related to that of fire.

So would a tungsten welding rod that had been sharpened to a fine point on one end be a suitable stinger?

Arcstarter

Mon Sept 07 2009, 06:23AM

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

I would say tungsten is the only thing that would work. My small tungsten welding rod (i think it is tungsten, it takes forever to cut, and it sure does not melt as easily as steel, but it might not be tungsten) i used as a breakout point gets red in maybe 5 seconds, and looses the sharp tip with 80v input to the MOT and smoothed fullwave. Copper would melt into nothing in a matter of seconds...

If you look at that last video with halfwave and 120v input on the MOT you can see the breakout point getting red. That is why i am so into these high frequency 'things' now, because of such huge current density.

To more accurately answer Jan's question, 10Mhz will do it, but just barely. Any lower will start to flail around. Mine flails also, when smoothed. It is usually a flame looking discharge, but sometimes it will violently flail (too fast for the camcorder to catch in detail), and once the electrode is hot it has a very bright intense light at the breakout point, and it flails even more, and you cannot even look because of how bright it is.

This kind of discharge with semiconductors is hard and takes alot of work. Getting a semiconductor to run the required power (i am sure you could get a 2 inch flame with 400 watts) at the required freq would likely need a scope and tuning, someone correct me if i'm wrong.

GeordieBoy

Tue Sept 08 2009, 04:33PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

400 watts at 10MHz is achievable with a single switchmode MOSFET like the IRFP450LC running in Class E from a -150VDC supply. But, yes it does require some RF engineering skills, a scope and some tuning to get it to work.

If you want more power (several kW but still below 20MHz) you can use several IRFP460LC devices. Drive the gates in series and combine all the outputs in parallel through seperate drain matching networks.

As for whether it's easier to use vacuum tubes: It depends if you know your toobs better than your FETs. For me, I would reach for the trusty MOSFET first because that's what I am most familiar with designing for.

-Richie,

previous 1 2 3 next

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