SGTC (Spark gap tesla coil) help needed.

[viliuks]

Hello, geekies, I built a tesla coil not long ago with a flyback and some saltwater capacitors.

So lets get to the point, I have saltwater capacitors which I want to change to normal capacitors that I need help determining which ones I need and what would their capacity be.

I got a crummy home made driver for the flyback which I'm going to change out to a ZVS driver.

Flyback is running on 12v DC and will be running the same voltage when I will get the ZVS driver.

Tesla coil specs:

Secondary:
height: 17cm
width: 5cm
wire: 28-32 awg (0.32mm - 0.20 mm) can't remember which one.

Primary is a tad wider about 6-7cm with 3wraps of 3mm wire.

So I need to get how many capacitors I need and what their capacity would be, with the type of capacitor too

If I need to provide anything else tell me. :thanks:

 

[ped]

If I need to provide anything else tell me. :thanks:

 

[viliuks]

Messy, I know

 

[Cyparagon]

One google hit says "Each beer bottle has a voltage rating of over 75kV, a capacitance of 3nF". This doesn't mean you need to replace them with this, though. A flyback's output voltage isn't typically anywhere near 75kV. The output voltage depends a LOT on your specific driver component values, and the flyback itself.

Since you're using beer bottles as caps, I'm going to assume you don't have access to a high voltage volt meter. You can instead approximate the voltage by the distance it jumps (strikes an arc). A rule of thumb is 1mm per kV, but it depends on electrode shape, and air conditions. Without knowing anything else about your setup, my best guess would be 12nF at 30kV.

 

[RedCowboy]

Yes electrode shape makes a big difference, round balls are the worse, but electrodes that come to sharp point fizzle into the air easily because electrons will spread out over the surface of a big round ball, but they will collect at a point.

Needles ionize a path fizzling to start, but balls will CRACK when they arc.

I had thought that at low humidity from 2 smooth surfaces, stationary surfaces that it took closer to 100,000 volts per inch or 4kv per mm, from smooth stationary surfaces, apparently not.

A quick search came back 25-30 kv between needles, 50-55 kv between 2.5 cm diameter spheres per 1 inch in dry air

http://www.kronjaeger.com/hv/hv/msr/spk/index.html

For anyone who does not already know there are 25.4 mm in an inch.

This is an old NST

 

[Down with Umbrella]

Heyy, have you thought about using an NST? They are quite common at neon sign shops as new ones come out the old models tend to pile up on shelves.
Must not have a GFCI**

What material is your SG made from?

Salt water caps. On the TC thread (daunting as it may be), salt water caps are discussed here on page 82, post #1297
http://laserpointerforums.com/f60/tesla-coil-build-thread-73702.html
(I forgot how to link specific posts to a link back)
Ahh...hmm...USAbro, *gasp* no longer around. But he asked questions and the folks here at LPF had patients to answer most of them.
Anyways check it out.

Keep us updated.
DWU

 

[viliuks]

Heyy, have you thought about using an NST? They are quite common at neon sign shops as new ones come out the old models tend to pile up on shelves.
Must not have a GFCI**

What material is your SG made from?

Salt water caps. On the TC thread (daunting as it may be), salt water caps are discussed here on page 82, post #1297
http://laserpointerforums.com/f60/tesla-coil-build-thread-73702.html
(I forgot how to link specific posts to a link back)
Ahh...hmm...USAbro, *gasp* no longer around. But he asked questions and the folks here at LPF had patients to answer most of them.
Anyways check it out.

Keep us updated.
DWU

Spark gap is just some flathead hex bolts.

And I'm not sure if they have NST's here, but I will look for them.

 

[Cyparagon]

(I forgot how to link specific posts to a link back)

Click the number on the top-right portion of the post. Your post is "#6" for example. The address bar of your browser will change to this post specifically. Copy the new address from your address bar.

http://laserpointerforums.com/f60/sgtc-spark-gap-tesla-coil-help-needed-95839.html#post1393147

Or in the Tesla coil post:

http://laserpointerforums.com/f60/tesla-coil-build-thread-73702-82.html#post1341071

 

[Sigurthr]

Hello, geekies, I built a tesla coil not long ago with a flyback and some saltwater capacitors.

I think the single biggest improvement you'll find will be in the matter of tuning. You've got what most would call a "junk pile flyback TC". Correct me if I'm wrong:

-You didn't calculate resonant frequency (aka. "f0") of the secondary, which means you didn't select the capacitance value of the primary capacitor or target turns of the primary coil.

So, what you need to do first off is learn the basics of Tesla Coil theory and build procedure. This section of wiki does a decent job of the basic theory: https://en.wikipedia.org/wiki/Tesla_coil#Resonant_transformer .

Tesla coil specs:

Secondary:
height: 17cm
width: 5cm
wire: 28-32 awg (0.32mm - 0.20 mm) can't remember which one.

Primary is a tad wider about 6-7cm with 3wraps of 3mm wire.

Essentially, you need to go to JavaTC Javascript Tesla Coil Designer and plug in the dimensions of your secondary and topload. Then play around with the dimensions of a primary coil and value of primary capacitor. You want the results section to say the primary coil is resonant at as close to the secondary coil resonant frequency as possible. Now you know what capacitance you need, and how many turns of a primary coil you need to tune that capacitance.

From here if you want you can look around for better capacitors (mind you, they are expensive, especially so for using a higher voltage source like a flyback, most builders spend between $60 and $200 on the capacitors alone). However, I would suggest you simply build better saltwater capacitors. "The Geek Group" had some videos up on youtube about how to build salt water capacitors using beer bottles, and it was really well done. Basically you use a plastic bucket to contain the bottles, and fill the bottles and bucket with saturated salt water - no aluminium foil needed, and much better performance.

If you want a major performance boost you'll need an AC high voltage, low frequency transformer like a NST (neon sign transformer) or OBIT (oil burner ignition transformer). DC flybacks work too but to get equivalent performance out of them takes a very high power well designed driver running in ZVS mode, and a few more parts like a charging inductor and filter capacitor that are hard to make effectively or expensive. A DC flyback on a zvs driver will work without these, but performance will always be far below that of a low frequency ac transformer. If you do get such an ac transformer, you'll have to size your primary capacitor according to the specific transformer you get; the capacitor must be matched to the transformer as well.

So you see the biggest issue isn't your salt water caps per se, but rather the haphazard selection of parts used. By no means should you think I am denigrating your coil or work, we all started with very similar builds. TCs that perform well are precision engineered devices, even the simplest spark gap type ones.

 

[viliuks]

I think the single biggest improvement you'll find will be in the matter of tuning. You've got what most would call a "junk pile flyback TC". Correct me if I'm wrong:

-You didn't calculate resonant frequency (aka. "f0") of the secondary, which means you didn't select the capacitance value of the primary capacitor or target turns of the primary coil.

So, what you need to do first off is learn the basics of Tesla Coil theory and build procedure. This section of wiki does a decent job of the basic theory: https://en.wikipedia.org/wiki/Tesla_coil#Resonant_transformer .



Essentially, you need to go to JavaTC Javascript Tesla Coil Designer and plug in the dimensions of your secondary and topload. Then play around with the dimensions of a primary coil and value of primary capacitor. You want the results section to say the primary coil is resonant at as close to the secondary coil resonant frequency as possible. Now you know what capacitance you need, and how many turns of a primary coil you need to tune that capacitance.

From here if you want you can look around for better capacitors (mind you, they are expensive, especially so for using a higher voltage source like a flyback, most builders spend between $60 and $200 on the capacitors alone). However, I would suggest you simply build better saltwater capacitors. "The Geek Group" had some videos up on youtube about how to build salt water capacitors using beer bottles, and it was really well done. Basically you use a plastic bucket to contain the bottles, and fill the bottles and bucket with saturated salt water - no aluminium foil needed, and much better performance.

If you want a major performance boost you'll need an AC high voltage, low frequency transformer like a NST (neon sign transformer) or OBIT (oil burner ignition transformer). DC flybacks work too but to get equivalent performance out of them takes a very high power well designed driver running in ZVS mode, and a few more parts like a charging inductor and filter capacitor that are hard to make effectively or expensive. A DC flyback on a zvs driver will work without these, but performance will always be far below that of a low frequency ac transformer. If you do get such an ac transformer, you'll have to size your primary capacitor according to the specific transformer you get; the capacitor must be matched to the transformer as well.

So you see the biggest issue isn't your salt water caps per se, but rather the haphazard selection of parts used. By no means should you think I am denigrating your coil or work, we all started with very similar builds. TCs that perform well are precision engineered devices, even the simplest spark gap type ones.

I'm not thinking you are denigrating my work, I'm just taking this as some tips I need to follow, now I'm going to go and try to tune things properly and expect something better than what I have now. Thanks for the response

 

[Seoul_lasers]

I'm not thinking you are denigrating my work, I'm just taking this as some tips I need to follow, now I'm going to go and try to tune things properly and expect something better than what I have now. Thanks for the response

How far are you along on the SGTC so far now? I am also an experienced builder that has built for galleries and schools.
I can probably help as well. I second the opinion that you want a 60Hz Iron core XMFR like a Neon sign transformer to get better performance.


There are a bunch of us on here that have experience with these intriguing devices. :can:

 

[viliuks]

How far are you along on the SGTC so far now? I am also an experienced builder that has built for galleries and schools.
I can probably help as well. I second the opinion that you want a 60Hz Iron core XMFR like a Neon sign transformer to get better performance.


There are a bunch of us on here that have experience with these intriguing devices. :can:

Well its for a school project and I wouldn't want to take an NST to there

I got the ZVS driver hooked it up, works real good.
Right now I just need to remake my capacitors and all should be good

 

[Seoul_lasers]

Well its for a school project and I wouldn't want to take an NST to there

I got the ZVS driver hooked it up, works real good.
Right now I just need to remake my capacitors and all should be good

I strongly suggest you use PE Polyethylene or PP polypropylene for your caps..


DO NOT USE GLASS. Glass becomes very hot and lossy at the high frequencies that ZVS drivers run at. Typically 100's of Khz - Mhz levels.

 

[Davidx]

Plus Rep for you. Admire that fortitude

 

[Seoul_lasers]

Plus Rep for you. Admire that fortitude

What's happening now with the project?
Using glass on ZVS driver powered TCs isn't recommended as it can heat up quite quickly. Also you loose capacitance as well. (in some cases if the freq is high enough, acting as a dead short)

Optimally Go for plastic (PP or PE will work well.) poly bucket caps or roll caps can be easily homemade.

 

[viliuks]

What's happening now with the project?
Using glass on ZVS driver powered TCs isn't recommended as it can heat up quite quickly. Also you loose capacitance as well. (in some cases if the freq is high enough, acting as a dead short)

Optimally Go for plastic (PP or PE will work well.) poly bucket caps or roll caps can be easily homemade.

Hey, the project is working fine, everything is nice..

I told my physics teacher about it and she wanted to see it, then brought me into a few conventions with it, didn't want to go but it went swell and I won several awards :beer:

Seems like no one has presented this kind of thing.

Now everyone knows me in school as a crazy kid with HV :undecided: