Tesla Coil Build Thread

[Speedy78]

Well the secondary is up at 1100Khz and I like the little topload. I'm sure I can get better output if I knew what my cap was for sure. I don't have any way of measuring.

 

[Sigurthr]

...Test run was the ity bity coil sitting on it. Decided to make a toroid for it. ...

Haha. I love small TCs. Won't be good for anything solid state outside of a class-e single ended coil, and those tend to blow up.

Edit: Built a small cap from aluminum foil and used the clear top from a binder as dielectric.

For sure the smallest SGTC I have ever seen though I'm sure one of you will link something smaller and more successful. ... Anyone have any recommendations for making a new cap that can withstand out of household materials?

Just use the MMC, lol. If you really want to make a cap I can instruct you how to make one out of beer bottles, a plastic bucket, aluminum foil, and salt water.

Oh and there was a contest a few years back on the TCML about smallest SGTC; I think the winner had one which was the size of a film canister.

 

[Speedy78]

Oh and there was a contest a few years back on the TCML about smallest SGTC; I think the winner had one which was the size of a film canister.

I figured someone smart had done a lot better than I. I figured the MMC would be too much. The whole thing is run with speaker wire and the gap some bent copper from past TC projects. I would be able to half ass tune it if I had the tank value using java. I'm looking for a small style cap to keep everything tiny. I'll run the MMC on it in a bit and see what catches fire!

 

[Sigurthr]

Lol, ahh I see, you just want a tiny cap. All home made caps have non standard capacitance values, so you'd need a faradometer (capacitance meter) either way.

Tune her up with the mmc and post video!

 

[Speedy78]

If I can get this thing to work decently I may purchase a single cap at whatever value I think works the best. Whats the smallest solid state driver out there and do they make one that only runs from a small DC input, say 19V? Would love to have this on my desk and throwing sparks!

 

[Sigurthr]

Whats the smallest solid state driver out there...

Mine.

...and do they make one that only runs from a small DC input, say 19V? Would love to have this on my desk and throwing sparks!

Again, Mine. Haha.

Though it isn't complete, to be technical, you'd still need to either
1)build a (half) bridge, or
2)use a 1:1 GDT to run my board single ended, or
3)leave off one of the UCC chips and only use one of the GDT output pins to drive single ended (least viable option).

If you wanted an on-board FET you'd have to consider EVR's SSTC1 or plasmaspeaker as the true smallest. On-board implies poor heatsinking though and that is the main killer of HF (<750KHz) solid state TCs. That is provided you can drive the gate fast enough with enough power.

I have an unfinished (read: shelved) design for an "all in one single ended SSTC driver" that I can dust off for you if needed. Never tested it above 750KHz though. That's kind of the limit for what hard switched mosfets can do. I can probably throw in some revisions to get it to work around 1MHz. The trick is converting to a TO-220 based mosfet driver so you can heatsink it as well and selecting a mosfet robust enough to hard switch at 1MHz. I can whip up the schematic for you rather fast if you like, but at 1MHz you need to know how to do SSTC layout when it comes to actually making it.

One of my next projects will be a 1MHz DRSSTC. Yep, you heard right. (There are TC experts shaking their heads right now. It will rely heavily on the MCU based interrupter I'm working on now though. It all hinges on a theory I have that it can be done as long as pulse widths are precisely controlled and that you need to use an oscillator driven topology, no feedback.

 

[Speedy78]

All that sounds way too advanced for me . I would need something plug and play and that 1Mhz speed will make anything solid state and small far too advanced for me lol

 

[Sigurthr]

I -could- make a plug and play driver for ya, but I'd have to charge for labor too. 1MHz coils tend to eat mosfets periodically too.

 

[Speedy78]

What kind of price?

New primary to match the MMC, not as much output with this. Increased coupling and wayy to much flashover. Went back to 7 turns instead of 3 with not much improvement. I think my homemade capacitor was a better match lol. Oh well, this one doesn't catch fire lol. My scope will be no good at this frequency so there's no reason to even try.

Can I pull the breakout off this? May also try to blow across the gap to see if I can get any other output.

 

[Sigurthr]

At 1.1MHz, with an unquenched gap, and only 225W of input the 4cm you're getting isn't too bad. The higher the frequency the shorter the arcs as the air appears more capacitive than resistive, so there is less drive to grow streamers. Yes you can remove the breakout point. Try to use the calculated turns ratio and then adjust coupling by placing an insulating stand under the secondary to elevate it above the plane of the primary (in gradual increments).

I'm not sure, I'd have to pull up my old plans and look it all over. Off the top of my head it would only be ~$25 labor + parts, I don't see more than 3 hours total labor. Actual parts count is low: 1x LM7812, 1x CD40106BEE, 1x TO-220 TC4420, 3 through hole 1x2 rising cage clamp connectors, 1 through hole 1x3 rising cage clamp connector, and a handful of capacitors. I'll have to do a bit of research for a good mosfet for 1.1MHz. I think I know a few but I wasn't looking into hardswitching when I recently poured through my catalogue for a 1Mhz mosfet (since my 1MHz coil will be Class-E or DE).

As much as I could use the work/money it honestly isn't worth the trouble imo. Secondaries above 750KHz aren't worth using unless you're doing something special or experimental. Besides, soon my controller will be available and you'll be occupied with it, heh.

Speaking of my boards; OSHpark emailed me today they received my boards from fab and are getting ready to ship them out to me.

Also I had the insane(ly brilliant) idea to upgrade my RF grounding network by using the left over 5kW 500MHz rated 7/8" coax line. Been doing that most of the time today. It's nearly done, just gotta make the junction to my dedicated ground line.

 

[Speedy78]

My only negative thing I have to say about this mini is the noise generated by the spark gap. If you ever get bored and have nothing better to do feel free to work up that driver for this. It would be really neat to have a desktop coil running on DC. If I could have similar output as my EVR SSTC at the breakout that would be awesome and even more compact. Its amazing that at this scale and frequency I get such small output but go full scale and that increases drastically. Its like damn magic to me.

Cant wait to start playing with your driver and get some e field going on!

Changed it up to a helical coil to help with the over coupling issue. Moved it up until I experienced too much flashover and went back down a nudge. Added a little more topload. (I guess this coil is named Jenny now after my wife seeing as how it has her name on the top) Strike rod grounded to ground. Was using a screw driver but that little f@#ker wasn't as insulated as I thought. Still getting minor flashover but new record! Also forced air across the gap. I notice a small increase in corona on the breakout or sharp point of the topload without the breakout but cant tell if I get longer arcs or not.

 

[Sigurthr]

Its like damn magic to me. ... Was using a screw driver but that little f@#ker wasn't as insulated as I thought.

Lol, that made me chucle pretty good. Yeah, pulsed coils have a bit to them, hehe. Secondaries can accumulate a DC voltage between them an ground as well, stored in the interwinding capacitance. Real nasty static shock when you go to touch it (when it is powered down).

Edit: oh and helical refers to a vertical solenoid type of coil. You have what is known as a "pancake" primary coil there.

It's funny that you refer to it as magic, because in a lot of ways that's the best way I can describe it to the layman. I recently had the demoralizingly uncomfortable task of trying to explain to my family exactly what my projects are. I couldn't even come close to getting them to understand. They literally would have understood better if I said "it is a bunch of electrical bits that make magic.". Unfortunately most people don't really even understand how a light bulb works, at least not up here.

It wasn't until a while later that I was thinking about how I'm a rather good teacher of subjects I know well, in general, but when it comes to electrical engineering, trying to teach the truly layperson is like trying to explain to someone how magic works. The very language we have to describe the fundamentals of what is happening is beyond the grasp of the average person.

The old example holds true; two people look at a scene of a rainbow. The average person sees pretty colors in the sky in an outdoor scene. The physicist/engineer sees a dispersion of atmospherically filtered light due to microscopic droplets of water in the atmosphere held together by surface tension and hydrogen bonds, and that the droplet aggregation that permits the phenomena is only possible due to the relatively higher pressure at that elevation which prevents the water from dissociating into a gaseous state. In my physicist's description most people would only understand water and light out of all the key words. Never mind them trying to understand the ideas behind the words, the words themselves are arcane to them!

 

[Fiddy]

Nice work Speedy!

Its getting close to my new years eve party where im deploying the Onetesla DRSSTC for musical arcs Wish you all a Happy New year from Down Under!

 

[Speedy78]

Good idea! I might bring mine to the party we are going too!

 

[Fiddy]

Hehe do it! Amaze the people in its mystery.

Sizing up a pancake coil with mah SGTC.

Should i raise the primary higher to engage more of the secondary? what type of distance should i need between primary and secondary?

Sigurthr i choose you! if your not already munted in tonight's festivities

 

[Sigurthr]

Spacing looks good! But for the height of the primary you'll have to try it out and see, or use JavaTC to model it. There isn't a finite yes or no.