Tesla Coil (TinyTesla) Kit Experiences and Questions

[motocoder]

Some time back, I funded a Kickstarter campaign whose product was a small Tesla Coil kit. The kit is the "Tiny Tesla", and you can find out more info about it on their web site:

http://onetesla.com/

I have some questions about problems I've had with the kit, and also about Tesla coils in general. While the OneTesla site does have a forum, they tightly control posts, and it sometimes takes them as long as a week for the moderator to approve a post. That just makes it impossible to carry on a conversation. So for that reason, I thought I might have better luck with my questions here.

I received the kit a few weeks ago, and assembled it over the weekend. The active electronics portion of the kit seem to work pretty well, but I had a really bad experience with the arcing within the secondary coil and between the secondary coil and the primary. The instructions, and what little helpful posts there are on the OneTesla forum, repeatedly stress that you need to coat the secondary with at least 3 or 4 coats of some sort of varnish - they recommend polyurethane. This seems to be the only solution they propose to peoples' problems with arcing within the secondary coil. Well, I put 6 very thick coats of MG Chemicals Super Corona Dope on the coil before testing it out. This is a product specifically designed for high voltages, so I figured it would work better than polyurethane . It didn't prevent arcing. Reading posts on their forum, it seems many coats of polyurethane didn't help either.

So I am wondering if the design of the coil is just bad - maybe the dimensions are such that the voltage gradient across the coil is just too high? Or maybe the top load needs to be higher above the secondary coil (i.e. farther away from the primary)?

Anyway, if you have any experience with these things, I would love to get some advice. Despite the problems, I did have some success with the coil before it eventually burst into flames, and I'd really like to try again - doesn't have to be a kit. But I'd like to learn more about how to be successful with the design before spending more effort on it.

[JimRemington]

From this up front suggestion on the Tiny Tesla order page:

We highly recommend that you get a spare set of IGBTs, as they are typically the first components that need replacement.

it appears that the design has many serious failure points. Very sad.

[LaserSteve]

It's big brother  has a huge array of failure points. Buy multiples of spare everything.
I've been mentoring a second year EE student who bought one of the larger ones.

The coil itself was good, but the driver lacks shielding, surge protection, and back EMF protection.

I don't know about the tiny, but the large one needs some serious improvements. When it works its cool, but it is a hanger queen most of the time.

Steve

[T3sl4co1l]

Their circuit is at best a toy...

Heh... I know whose circuit that comes from, too... but it's probably much older than that.

I've mentioned this before, how kits and communities tend to be predominantly social rather than technical affairs.  Just because the same bad circuit is repeated, time and time again, doesn't mean it's a good circuit, technically.  It means it's a good meme, which has nothing to do with its technical value.

It would seem all the value added by the KickStarter was exclusively for appearances and marketing only.  Flashy website, cute product; technically deficient.  They should've taken the wise route and not had a forum at all; that way there's fewer complaints about a defective product.  (Excuse the cynicism dripping from this paragraph; but to be fair, these are real kinds of decisions that might be made, for various good and bad reasons, by top management.)

Also, I wouldn't expect corona dope to do much.  You're talking quite a lot of voltage, and at frequencies where it's more than happy to DBD (dielectric barrier discharge) on the surface of the stuff, until it breaks down and arcs directly through.  Your only choices are to reduce power (reduce primary turns, coupling?) or lengthen the secondary.

Tim

[motocoder]

it appears that the design has many serious failure points. Very sad.

I think that is fairly typical for these tesla coil designs. In many cases to build something that is reliable the cost gets out of range for the hobbyist. So it's fairly common to operate components out of spec.

I'm not saying that more could not be done on the TinyTesla for a reasonable cost. Thermal shutdown and back-EMF protection on the IGBTs seems like it could be added without much additional cost.

[motocoder]

Their circuit is at best a toy...

Heh... I know whose circuit that comes from, too... but it's probably much older than that.

I've mentioned this before, how kits and communities tend to be predominantly social rather than technical affairs.  Just because the same bad circuit is repeated, time and time again, doesn't mean it's a good circuit, technically.  It means it's a good meme, which has nothing to do with its technical value.

It would seem all the value added by the KickStarter was exclusively for appearances and marketing only.  Flashy website, cute product; technically deficient.  They should've taken the wise route and not had a forum at all; that way there's fewer complaints about a defective product.  (Excuse the cynicism dripping from this paragraph; but to be fair, these are real kinds of decisions that might be made, for various good and bad reasons, by top management.)

Also, I wouldn't expect corona dope to do much.  You're talking quite a lot of voltage, and at frequencies where it's more than happy to DBD (dielectric barrier discharge) on the surface of the stuff, until it breaks down and arcs directly through.  Your only choices are to reduce power (reduce primary turns, coupling?) or lengthen the secondary.

Tim

Thanks, Tim. I forgot about your username, or I would have PM-ed you to the thread. Are there any web sites you would recommend to learn more about Tesla coil design? Despite the problems with the Tiny, I did enjoy the project, and I'd like to learn more on the topic with a goal of eventually building my own. I'd rather just buy a pre-wound secondary, though - winding 1000 turns of thin wire isn't my idea of fun

[T3sl4co1l]

Ironically, I haven't played seriously with TCs in *years*... this nickname is quite old.

It's not a bad idea in principle, but it lacks subtlety, control and protection.  As with induction heaters (which I have been more involved with in recent years), it often comes down to having a PLL-VCO-driver-inverter model.  The VCO ensures that the circuit is always oscillating smoothly (even during startup, whereas the transformer and flip-flop in the present circuit could be picking up any random noise on the coil, causing erratic switching at who knows what frequencies, modes or subharmonics!), and the PLL selects the dominant (fundamental) mode with relative insensitivity to noise and harmonics.  The drivers must be transient protected; the best way to handle IGBTs is with a desat circuit, which triggers a start/stop latch on the oscillator side.

All of these concepts are included in the (rather dated) schematics on my website, http://seventransistorlabs.com/tmoranwms/Elec_IndHeat6.html but there are better/more convenient ways to do pretty much everything.  One warning: the XOR type phase detector was a mistake; a type 2 detector is required.

Tim

[Fungus]

Try moving the primary up and down slightly. It makes a huge difference to coupling, which can cause secondary arcing like you describe.

Once you find the optimum height for the primary you should gradually add/remove turns on the primary (not complete turns, 10 or 20 degrees of turn at a time). Finding the optimum number of turns on the primary can really increase transfer efficiency and keep everything cool.

Every coil is slightly different, you have to tune them...

[motocoder]

Ironically, I haven't played seriously with TCs in *years*... this nickname is quite old.

It's not a bad idea in principle, but it lacks subtlety, control and protection.  As with induction heaters (which I have been more involved with in recent years), it often comes down to having a PLL-VCO-driver-inverter model.  The VCO ensures that the circuit is always oscillating smoothly (even during startup, whereas the transformer and flip-flop in the present circuit could be picking up any random noise on the coil, causing erratic switching at who knows what frequencies, modes or subharmonics!), and the PLL selects the dominant (fundamental) mode with relative insensitivity to noise and harmonics.  The drivers must be transient protected; the best way to handle IGBTs is with a desat circuit, which triggers a start/stop latch on the oscillator side.

All of these concepts are included in the (rather dated) schematics on my website, http://seventransistorlabs.com/tmoranwms/Elec_IndHeat6.html but there are better/more convenient ways to do pretty much everything.  One warning: the XOR type phase detector was a mistake; a type 2 detector is required.

Tim

What didn't make sense to me is that the "transformer and flip-flop", as near as I can tell from their incomplete schematic and what I see on the PCB, is actually an antenna, not a transformer per-se. So it's for sure picking up the radiated emissions from the secondary. They are then using this to control the switching of the current through the primary. It seems like this means the switching is NOT going to be during zero crossings, which they go at some lengths to explain is the purpose of this circuit. This might explain why so many people burn out their IGBTs.

[onetesla]

Hi everyone, Heidi from oneTesla here.

motocoder, we don't have experience with the varnish you were using, but we have had good success with Minwax polyurethane. I also recommend putting on a thick layer of packing tape as shown in this album: https://imgur.com/a/cqq3I?gallery#0. It's possible to completely eliminate the arcing if you do a good varnish and/or taping job. If you have a lot of trouble with your secondary, we have a number of pre-varnished coils we can offer. Feel free to shoot us an email at nikola@onetesla.com.

Many of you have valid comments on features that our coils don't have, but as designers we need to balance a simple, accessible kit design with supporting a lot of features.

If you have any other questions about our kits we're always here to help, and you can reach us by email as well as our own forum.

Happy coiling everyone!

[motocoder]

Hi everyone, Heidi from oneTesla here.

motocoder, we don't have experience with the varnish you were using, but we have had good success with Minwax polyurethane. I also recommend putting on a thick layer of packing tape as shown in this album: https://imgur.com/a/cqq3I?gallery#0. It's possible to completely eliminate the arcing if you do a good varnish and/or taping job. If you have a lot of trouble with your secondary, we have a number of pre-varnished coils we can offer. Feel free to shoot us an email at nikola@onetesla.com.

Many of you have valid comments on features that our coils don't have, but as designers we need to balance a simple, accessible kit design with supporting a lot of features.

If you have any other questions about our kits we're always here to help, and you can reach us by email as well as our own forum.

Happy coiling everyone!

Hi Heidi -

Thanks for the reply. I did actually have kapton tape on there (plus a layer of vinyl electrical tape around the primary). It did help, but didn't completely eliminate the problem. I ran the coil a fair amount without issues, but then when I set it up at work to show my coworkers, it arced. The antenna was in a different location, and I've read that can have an impact.

Also, based on the posts I see on your forums, I am extremely skeptical that polyurethane, no matter how many coats, eliminates the problem. I've spoken with several people knowledgeable in TC design,  and the universal answer I've gotten is that the polyurethane is there to keep things in place and provide some mechanical protection for the coil, and can't by itself provide protection from arcing.

But I do understand that you have to make some trade-offs to keep the price down, and like I said, I have had some fun with the kit.

By the way, I strongly urge you to move away from the moderated comment system on your forum. Week-long delays for posts to be approved is totally unacceptable, and it really kills any hope of users being able to help each other out. It also says something not very complementary about your products that you feel the need to limit posts.

[onetesla]

I've spoken with several people knowledgeable in TC design,  and the universal answer I've gotten is that the polyurethane is there to keep things in place and provide some mechanical protection for the coil, and can't by itself provide protection from arcing.

All of our coils have only ever used polyurethane varnish. However, the confusion may come from the fact that at the first sign of arc-over, the varnish and wire enamel become compromised and need patching immediately, using a drop of varnish, hot glue, epoxy, or even tape. Continuing to run the coil after it arcs will always exacerbate the problem unless you patch the burned spots.

Moderation at onetesla.com/forum is an anti-spam measure and goes away after three posts. We typically approve posts within a day, but some may fall through the cracks. If you continue experiencing issues please do not hesitate to contact us at nikola@onetesla.com.