Re: Poor SGTC results Problem
Nik, Thu Jun 22 2006, 04:47PM

Try raising the primary a bit, it will change the coupleing and could help.

Re: Poor SGTC results Problem
Michael W., Thu Jun 22 2006, 05:01PM

I tried that and it increased the output a bit but not anything substantial....

Re: Poor SGTC results Problem
Terry Fritz, Thu Jun 22 2006, 06:48PM

Hi,

I think I have heard of this type of problem before Yeah!! like hundreds of times!!! (literally)...

There are two problems.

First and biggest is that the resonant frequency of the secondary coil and top load does not match the resonant frequency of the primary coil and cap. That is 80% of the problem. Having the primary and secondary matched to the same frequency is "important"!! Otherwise the coil is a space heater.

There are many Tesla coil design programs out there and lists of formulas:





I would be happy to work the numbers for you so you can see how to do it. Let me know, and I can do it latter. I should just make a web page on this...

Parts of the above problem but a problem too is that beer bottle are far better for beer than caps. Modern MMC type polypropylene caps will perform vastly better. But for now, fix the tuning. Bottle caps like yours are about 800pF each.

I assume you are putting the caps in salt water too when you run it. If you run it out of water as in you picture, that is another problem

Cheers,

Terry

Re: Poor SGTC results Problem
Marko, Thu Jun 22 2006, 07:58PM

How's your spark-gap?

Single stage gaps usually quench poorly even on small power levels, and you get more of arcing than regular banging there.

Try something simple like nut gap (with flyback of course, you don't need 3kW for this one)

Try using a topload, maybe a chrismas ball wrapped in foil or something. SGTC's generaly don't do well without a topload.

I assume you are putting the caps in salt water too when you run it. If you run it out of water as in you picture, that is another problem

If you look closer bottles are wrapped in foil and then with plastic tape around.

Re: Poor SGTC results Problem
Hazmatt_(The Underdog), Fri Jun 23 2006, 12:00AM

You will also notice improved performance with a somewhat large sheet attached to the RF return from the secondary. This adds some capacitence referenced to the secondary, which helps control your frequency. For my ~700W coil I used a 4' x 4' with pretty good results, but I would think you could get away with a 1'x 1' or 2' x 2' sheet no problem.

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 12:47AM

I actually have "RF Ground" made of Coiled chicken wire for that part, but I did suspect the fact thats its horribly out of resonance. How would I go about Transformer/Capacitor matching with bottle caps? and then taking on the daunting task of trying to tune the primary? (I do have a Func. Generator and an O-Scope)...

Re: Poor SGTC results Problem
Marko, Fri Jun 23 2006, 01:21AM

Use wintesla to calculate your coil, you'l usually be about the range but tuning is best done by trial-and-error.

You can use mains ground for coil of such size, and it is necessary.

Topload is also going to pump the performance up.

Spark gap is pretty important thing but you said nothing about it.
Give some more info about the coil.

Re: Poor SGTC results Problem
Desmogod, Fri Jun 23 2006, 01:30AM

Yeah, make sure you ground the coil properly as stated, I would also say that it's way out of tune, as you are only tapped on the second turn of your primary.
You also might not have enough capacitance in there.
Can you measure the C of your cap at all?
Definately use wintesla and go from there. Can you post specs? Secondary specs? C, spark gap etc etc?

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 01:58AM

Secondary Specs are as previously stated : 1 3/4 " X 6.5" Secondary, 26 Gauge wire. For a spark gap I've tried many variations; Pointed Electrodes, Pipes, nuts etc. No Change. I've always wondered about the Capacitance though. For two MOT's would 4-5 bottles be enough? I've tried trial by error tunning and at about 3 turns I get the best performance but its still only like 1 inch arcs.

Re: Poor SGTC results Problem
EDY19, Fri Jun 23 2006, 03:01AM

Ahh, 4 bottles are not quite enough. I tried it on wintesla, and for 4000VAC (two MOTs in series) and .5A (~2kW, quite large) the suggested capacitance is .33uf. Also, if you want to use those two transformers, you WILL need a bigger secondary. Even rating the transformers for .25A gives a capacitance value of .165uf, and when each bottle is about 800pf, your total is only .0032uf. So either make a LOT of bottles (412.5 ) or make an MMC .

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 03:11AM

I've also used a level shifter on the MOT's (to produce about 8KV?) with an MOT to ballast the lot...Mabye I need an MMC but that doesn't solve the resonance problem. If the coils out of sync it could blow the MMC, could it not?

Re: Poor SGTC results Problem
Coronafix, Fri Jun 23 2006, 03:45AM

I'd suggest making a poly rolled cap, or stacking foil and poly vertically and immersing in mineral oil.
I made a salt water cap and it measures at 600pF, that means for my static gap 15kv60ma NST, I need 30 bottles!
I don't think your 6 will be anywhere near enough when you're using a dual MOT.

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 04:15AM

Well I'm sure as not ready to take on the daunting task of making 30 to 412 bottle Caps. I'm thinking more of a stacked cap...how about plastic plates and foil somthing along the line of this....How many plates do you think I'd need? (that is, if it works.. )

Re: Poor SGTC results Problem
Desmogod, Fri Jun 23 2006, 04:37AM

Re: Poor SGTC results Problem
Coronafix, Fri Jun 23 2006, 07:00AM

Here is another good one.

Re: Poor SGTC results Problem
EDY19, Fri Jun 23 2006, 01:03PM

"I need an MMC but that doesn't solve the resonance problem. If the coils out of sync it could blow the MMC, could it not?"

A MMC would solve the resonance problem. I'm guessing you have come nowhere near the correct tuning for this coil. Since MOTs have a High current, to get even close to 1kW from the coil, you will need bigger capacitors, and the easiest and most efficient way is with a MMC. Just think about it- if the capacitors are two small, and the the MOT can supply a high current, the capacitors are holding nowhere near the charge that the MOT could supply. And, it wouldn't set you back too far price wise. They make .15uf 2kV capacitors that would be about perfect, and you could definatly make longer sparks! I would suggest putting 4 caps in series, and then 4 of these in parallel to get ~.33uf at about 8kV. This would do a lot of good, and if you only wanted one MOT (still large enough for BIG sparks) you couls just use 2 parallel strings, and your output would be greatly increased!

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 04:34PM

42L's .15uf 2KV, 4 Bucks a pop, thats 4 strings of 4, 16, is $64 plus shipping, and I'm canadian... Anywhere cheaper or anyone have any they want to sell off?

Re: Poor SGTC results Problem
Marko, Fri Jun 23 2006, 06:02PM

I think you need to improve much more things over there before making monster 150nF 8kV capacitor for such small coil.

Even TDU's 18 inch coil uses 'just' 90nF at about same voltage.

Powering it with two MOT's is just silly for 5cm wide secondary.
You'l be just about enough with small 100W mazzili driver (by freau's formula this would be a 17 inch spark, I guess half of it would be pretty good for this coil).

Bottle caps are fine, especially because they allow you high-ish charging voltage without blowing up (enough bang energy with small capacitance).

Current 4-5nF is absolutely fine, if you really feel like such you can add some more bottles there.

Some more apsolutely necessary things are simple multi-static gap, for this power level few closely-placed metal nuts can be a good example, then a good ground (mains ground or water pipe), and invetable, topload.
Topload can be a foil covered sphere or toroid, two welded round bowls, or anything without sharp edges you can find.

This is especially important for such a small coil with hign no-topload resonant frequency.

Use a fuse clip to tune the coil.

All aproximate calculations you can do using wintesla, TCCAD, or any program you like. You have entire primary to tap.


Good luck, and don't get dissapointed or nervous just at the beggining.

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 06:08PM

I'll use my flyback driver (produces up to 3 inch arcs) to power my coil, I'll make a new spark gap, and I'm going to try and tune the LC Circuit with an O-scope/Sig. Gen. this weekend...I have a top load suitable but I thought since with out a topload the max arc was about an inch, what would a larger topload do...I'll try it

Re: Poor SGTC results Problem
GimpyJoe, Fri Jun 23 2006, 06:38PM

Also try a primary with WAY more turns. I recommend going to Home Depot or some similar store and buying two rolls of spring bronze weather stripping and two rolls of 1/4" foam adhesive weather stripping tape. All of these should be seventeen feet long each, at least that's what they carry at my store. Solder the ends of the two pieces of bronze weather stripping together to form one long strip, then remove the adhesive backing from the foam strips and stick them along the length of one side of the bronze. Wind this into a spiral and secure it with hot glue. Voila! A perfect bronze primary with 1/4" turn spacing. To tap it, use a flat piece of copper wedged between the bronze and the foam. While you're at it, make a bigger secondary. 3" diameter by 18" should be fine.
Your bottle caps will work fine with your MOT's, but you're going to have an insanely high break rate. To compensate you need excellent gap cooling. I like the inside-out sucker gap design from Greg Hunter's Hot Streamer site. Use the biggest vacuum or blower you can find.
With all those primary turns you'll be able to put a decent sized topload on your coil. Make a 4" aluminum duct toroid maybe 15"-18" diameter and use a break point.
With these improvements I wouldn't be surprised to see you get 18"-24" sparks at least from your MOT's and bottle caps.

Re: Poor SGTC results Problem
Terry Fritz, Fri Jun 23 2006, 09:31PM

Hi,

Let's see what we have now...

I have a 1 3/4 " X 6.5" Secondary, 26 Gauge wire, 4 Beer Bottle Capacitors

The secondary has 377 turns of wire on it. The inductance is 1.49mH.

I'll guess that the top terminal is about 2.5 x 7 inches so the resonant frequency is 1.35MHz.

You have six bottles (the picture looks like six) at say 800pF each so 4.8nF or primary capacitance. Thus you need a primary coil of 2.89uH. If your primary is 3 inches ID with a pitch of 0.5 inches, it should tune at about 4.5 turns. The wiring from the caps and gap will lower that some to maybe 4 turns.

You caps seem to be made fairly well so that is probably not a big problem.

If you can fire the coil 120 times per second at 8kV, it should put out 7.3 inch sparks "theoretically".

The very high frequency will add losses especially in any ferromagnetic metals and aluminum has poor skin depth properties. You are loosing some there for sure. Copper only conducts in the outer 0.002 inches at that frequency. Steel might as well be zero...

But I think it should be able to do 3,4...5 inch sparks. If you can fire faster at say 240...480 BPS that will help some. That should be the case anyway since two voltage doubled MOTs will charge the little cap very fast.

I'll make a new spark gap, and I'm going to try and tune the LC Circuit with an O-scope/Sig. Gen. this weekend...

That sounds like a good plan!

I have a top load suitable but I thought since with out a topload the max arc was about an inch, what would a larger topload do...I'll try it

You always want a topload! The one you have should be fine.

Fiddle with the gap and tuning and it should work.

Hope this helps Your first coil is 10,000 times better than "my" first coil was

Cheers,

Terry

Re: Poor SGTC results Problem
Michael W., Fri Jun 23 2006, 11:14PM

UPDATE: 3 inch arcs to screw driver; Door Knob Topload, 5 bottle Capacitors, Eyeball Tuning, Flyback Power Supply and to top it off, a rotary Spark gap. I tried a gap using 4 brass nuts as suggested, but the rotary one gave 2X the results, hopefully I can pump more out of this hodgepodged mess....

Re: Poor SGTC results Problem
Marko, Fri Jun 23 2006, 11:37PM

I think now you just need a much bigger topload.
Make yourself a nice sphere or toroid, that one from avatar looks good.

Rotary is an overkill for coil of this size but you'l have a good quenching at least.
Can you post a pic of it?

Re: Poor SGTC results Problem
Michael W., Sat Jun 24 2006, 12:35AM

I've tried a 3X7" and a 3X10" Toroid with horrible results, >1" arcs opposed to the 3 inch arcs I get off of a doorknob. My spark gap is just thrown together....:

Re: Poor SGTC results Problem
Hazmatt_(The Underdog), Sat Jun 24 2006, 04:43AM

thats because you have to retune when you put a topload on top!

when in tune with your topload, you should have 4-5" sparks if you have 3" sparks with a free terminal.

My system gives me about 2' without topload and almost 4' with, but its all in tune. I also use a rotary with safety gapping. The system power is small in comparison to what you're using if you can believe that! I'm using a 12/60 NST and 16nF (should be 18nF from my calculations and simulation), and that gives me a 4' spark. So I would strongly suggest go back and do some calculations and some work insted of throwing everything together.

Re: Poor SGTC results Problem
Steve Ward, Sat Jun 24 2006, 06:22AM

Well, a topload can in fact hurt performance on very small coils if the topload is too big. Remember, energy is conserved, so you can only charge that toroid so much, and if its a bigger C, then V will be lower. Bigger coils have enough power to form streamers, so top voltage isnt as useful as having more capacitance there, but on little coils that can hardly break out, you dont want to load it down too far (or else the voltage is too low to get good streamers forming). I would think a 2"x8" toroid or maybe a 4" sphere would be about the right size for this coil.

Re: Poor SGTC results Problem
Terry Fritz, Sat Jun 24 2006, 11:17PM

Steve brings up a good point! Let's figure it out...

The top terminal, and all, capacitance is 9.33pF. The primary is 4.8nF at 8kV.

So the primary energy is 1/2 x C x V^2 or 0.1536 joules.

Only 1/2 that will make it to the secondary with a high frequqncy spark gap coil so:

1/2 x 0.1536 = 1/2 x 9.33pF x V^2 V = 128kV

A 2.5 inch diameter top terminal should break out at 95kV. So if the losses are a little higher than I calculate here (easily), it might not have enough voltage to break out. But simply using a pin or sharp point on the toroid will fix that.

So if it is not making sparks with the toroid, add a sharp needle or pin preakout point to insure the sparks can get out. Also, the primary will certainly have to be retuned for any change in the top terminal. Having the toroid on the top, should make better sparks than no toroid at all. But your 3 inches off the door knob is pretty good!! There is a saying, "if it works, don't fix it" :o)

Cheers,

Terry

Re: Poor SGTC results Problem
Michael W., Sun Jun 25 2006, 01:01AM

Now that its working, what is a reasonable expectation of when I can stop hoping for longer arcs? Right now I'm using a flyback to power, rotary gap and 6.5 Salt water Caps and I can get a "Good" 3.5-4 inches...Primarys tapped at 5 which seems to get the best results.

Re: Poor SGTC results Problem
..., Sun Jun 25 2006, 01:54AM

how much power are you running into the flyback? There is sorta a standard formula of spark length=[square root (power in)] *1.2

Re: Poor SGTC results Problem
Michael W., Sun Jun 25 2006, 02:04AM

I'm putting 45 Volts into the circuit, Its a 2N3055 driven one, I'm not sure how I would measure the amperage to get wattage.

Re: Poor SGTC results Problem
GimpyJoe, Sun Jun 25 2006, 03:46AM

You will see a huge improvement in spark length if you remove those gator clips and hook the whole thing up with very short wires. In my old nst coil this brought my spark length from 4" to 10". You will have to re-tune the primary since you'll be reducing inductance in the tank circuit.

Re: Poor SGTC results Problem
Hazmatt_(The Underdog), Sun Jun 25 2006, 07:12AM

you guys do realize that spark length is a function of ENERGY right?!

Re: Poor SGTC results Problem
Steve Conner, Sun Jun 25 2006, 12:12PM

Yes, and replacing crappy Radio Shack clip leads with short, thick wires will waste less energy, leaving more to make sparks.

How strong is the salt water in your beer bottle caps out of interest? It needs to have as much salt in it as possible to make it conduct well.

Re: Poor SGTC results Problem
Michael W., Sun Jun 25 2006, 01:49PM


Hey! No making fun of my obsolete equipment...
I changed the "Crappy" Wire to short. thick cable and there "might" have been some improvement but It wasn't very noticible and I did try to retune. The Salt Solution is Saturated, so theres as much salt in them as possible while still being in liquid form...