Hazmatt Tesla Coil / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Hazmatt Tesla Coil

Move Thread

LAN_403

Hazmatt_(The Underdog)

Tue Feb 14 2006, 06:59AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Okay, I know this is kinda a continuation of a lame thread, but what I'm trying to do is help!

Progress Update:

I used my voltage divider for the first time tonight and I am happy that there were no ground issues, smoked scopes, shocks, pops, or fizzles. However, upon using the divider set for 1000:1 division, the pulses went right of the screen. Now I only have 5V/Div. max, so I recalibrated the divider to 1000000:1 with 1000V being 1mV and that improved my situation.

So I can see every so often a massive ringdown on my scope, but there are problems!
Whichever one of you guys, I can't remember now, said "You don't need a tachometer for your rotary" or some such, well, I'm the kinda guy whose not so good at synchronizing things, so the next step is to sync things up. Major delay! So now I get to make some kind of stupid triac trigger at 120/sec for my motor! AARRGG.

Also my stationary gap has to be set properly. I'm having some words with it!

Things aren't going so well. Performance is down even though I've set things up with better leading and positioning. One secondary arcs down the length of the tube I know because when I exchange it for my other secondary which has a 4" more pipe at the top I get better output.

My laser cap bank has some problems with having an arc shoot through the primary coil platform to the plate of the cap. Varnished it an hour ago. The hasty primary construction does not pay off!

I have to go back and do some current determinations when the system doesn't fire to make sure the charging current isn't too high for the transformer. I hope to see 35mA A.C. at my half power point. I think I can calibrate for this. Should be 45V rms through my 1500R, and .045 V on my divider once I recalibrate.

This was so much easier when I used resonant charging. But if I can size it larger and pull a little more current, I can get more energy into the system which means longer sparks.

Wish I had more knowlege in simulation, then it would be easy!

>>>>>>>>>Update for 2-22-06<<<<<<<<<<<<<<

I have moved on forward deciding to use what is commonly recommended as the LTR capacitor for the rotary system. I was terribly hung-up on this because I was worried about the secondary inductances of the source and its associated losses.

I will consolidate my figures here when I can. I tried to find my calculations on the archive but I can't remember where I put all of that, so I'll do it later.

Anyway, what I can say is that I am using the simple calculation to derive my resonant capacitor like WinTesla does, then for a static gap I have found that 1.4xCres works with simple simulation and is also where Terry sizes his capacitor for a static LTR.

For my present system its a rotary so I'm using his figures which is 2.8xCres.

My NST is a 12KV 60mA so my Cres=13.26nf
Static LTR Csys=18.56nf
Rotary LTR Csys=26.0nf

Having the extra capacitence is important because it adds to the system resources. More Joules in the system results in longer arcs.

Mon Feb 27 2006, 06:17AM

Registered Member #256 Joined:
Location:
Posts: 124

hope you get this runnin full efficiancy cuz this is lookin really cool

Kolas

Mon Feb 27 2006, 06:48AM

Registered Member #102 Joined: Thu Feb 09 2006, 08:15PM
Location: Raleigh, NC
Posts: 169

heh i really admire how well you keep your shop
mine is a total mess
i need to clean up i guessi better wait till spring

kolas

Hazmatt_(The Underdog)

Mon Feb 27 2006, 09:00AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

well, I have to push everything to the sides because we have to park the cars in at night.
There's barely enough room as it is! Work benches on the one side, chemical cabinet there with all the other junk pushed to the side in the pic. I have about 18" to open my door. But I can make it allright. I guess that proves I design thoughtfully. (I made the cabinet)

>>>Update for Design considerations<<<

In case anyone looks at this in the future, the resolved value for my capacitence was determined after several factors.

>>>>Update for progress<<<<

Here's the new primary coil base out of plexi. I had divverent plans but had to change my plans. All worked out in an acceptable way.

1. I wanted to know more about the system concerning parasitic inductances in the secondary of the transformer and how that relates to its impedance transformation. In doing so, you find there is a large discrepency between the calculated impedance and the source impedance. Right now I do not have the skills to simulate and clarify further, but the calculated versus apparent can be off by 30%. This leads to quite a bit of confusion for me in which number to use, but I settled on the common calculation and based my numbers around that.
2. Common impedance calculation Z=E/I
This is misleading because there are inductive lags 'in' the transformer which are not accounted for. I settled on this calculation because it was the easy way to deal with the complex situation.
3. Taking Z we find that Cresonant (Cres) to be a particular value. Using Cres in a system is a bad idea because the 'resonant rise' of the system will cause a failure. Simulations will carry voltage rises out to hundreds of thousands of volts, but you never actually get there because a fault occurs much sooner then that.
4. Proper choice of C. The wise choice I found was 1.6xCres period. You can try different values if you must, but I investigated this to the point where my transformer failed. The cause of death was 2.8xCres.

If you are not sure about your calculations I have some verification for you here:

12000VAC 30mA Csystem=.010uF Proven 3-10-06, not just calculated
12000VAC 60mA Csystem=.020uF Never had the chance, NST dead.
9000VAC 120mA Csystem=.056uF Not tested yet.

**********Video Woot **********

Hazmatt_(The Underdog)

Thu Jul 05 2007, 11:27PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

***** HOMECOMING *********

After more then 1 year at college, and no A.S. degree(joke), my coil has finally come home with some improvements and some changes. These changes were to test some ideas about Tesla coils and to ask further questions.

My coil rewind was a fairly quick job so it wasn't really coated well, I had to prepare for the 4th and had to get it going, so the coating is pretty shabby.

The coil features some odd things to test ideas.

The only parabolic primary TC in existance that I know of. This was for fun.

Corona ring to help supress breakouts below the toroid.

Segmented static gap of pipe nipples. So I could adjust the robust gap in steps.

Horrendously overwound secondary coil. #26 AWG x 29" winding length, for 720VA input. Odd H/D ratio.

Secondary winding start .5" below primary winding start. I was hoping to really overcouple this one.

Ground patch glued to the bottom of the coil. NO wall intrusions.

#10 Silver Teflon wire interconnects for high di/dt handling.

There were some issues with the coil in that the toroid was much too large for the coil. I was 10% out of tune which probably lost me a good amount of spark length. I was getting approximately 3 foot arcs from the system and I hope to recover my foot after I fabricate a new toroid.

Discussion:

The corona ring I thought might help supress some streamer breakouts, but it really didn't help much because the ring is not varnished. It helped promote some ~ 6" arcs from the point of termination, but I couldn't help this because I needed something that was not so permanant. I may put a small rubber cover over the screw terminal to prevent breakouts, but I think when I construct the new toroid I will make it seat lower so as to be closer to the ring. I think the ring will help supress breakouts if it is closer to the toroid.

The static gap setting had some problems because the maximum gapping would fire at 105V input to the 12/60. This would cause a lot of power arcing if I used the 20nF cap my optimizations suggest. So I chose to lower the terminal voltage by going to 22nf. This kept most of the power arcing under control and limited the firing of the static gap to the majority of the rotary. I like this pipe nipple setup, but more thought has to be given to the exact spacing of each gap. I'll have to go back sometime and space them out further.

The base of the coil has a ground patch glued in place. The ground lead is attached by nylon screws that thread into the form. I like this because it limits metal hardware intrusions. However, this may be seen to the form as a capacitive terminal, so there may be some significant charging on the form and dielectric stress associated with this method. I would have to investigate that a lot further down the road sometime.

The secondary coil is overwound for this setup on purpose. I wanted to see if there would be a significant improvement of spark length due to secondary inductance. I'll have to report on that later because the system is not in total resonance yet.
What I can say though is that I expected horrendous racing arcs and pri-sec strikes at the base of the coil, and there were none. I did not use any special varnish or corona dope at the bottom end, and yet this coil is the only one I have ever made without racing arc problems. I'm not sure what accounts for this at the moment, maybe they'll show up under a resonance condition, but for now there were no lower arcing problems.

The tightly twisted silver teflon wire improved the situation greatly as there is no more heating of the lead wiring. Previously Belden 40KV rubber wire was used, and as some of you guys know its not a tight twist wire, and its only 18 AWG. Not really suited for 300A pulses and TC duty. It was still used for the supply-to-gap which has no real issues, but I don't use it anywhere else now. The wire exhibited very low inductance, which is exactly what we want. We want all the inductance to be in the primary tank and not with the lead ins. The wire exhibited about 1uH/foot inductance. I still have to go back and trim down my tap selector wire, but it concerns me less now.
As before with the Belden wire I would have significant heating and inductance throughout the lead-ins. I think my old tap selector wire had 7uH of inductance, which would be detremental to the effectiveness of the primary tank. Going to heavy conductors really helps push that current to where its supposed to be.

I am very curios as to what my coupling is at this time. I will have to investigate that when I have a little more energy. As I said I wanted to really overcouple this coil and see if it would break down under the stresses, but we didn't see any racing arcs or strikes like before.
Some have suggested that coils operate best above the primary inductor, but I tested this before and found that if you want to overcouple the coil it must be somewhat below the primary. There is a limit to this coupling. I found that after about -4" I had no improvement in the voltage transfer with a steady sinusiod input.
Initially the winding was about -1", but during the process of removing all the tiny gaps in the coil I lost about .5", so the result was about -.5" from the primary inductor. I didn't try to go any lower then this because I was afraid the racing arcs would light up the bottom like a christmas tree.

I will try to post some spark pictures after I get the properly sized toroid constructed, and hopefully I will get back to 4' arcs. If this system compares to the old system with the same spark length, then the conclusion will be that the increased sec. inductance has no favorable impact on the system. I just wanted to be sure and test how the secondary performance influences the mutual inductance and voltage transfer in the system.

**************************

Thu Jul 05 2007, 11:34PM

Registered Member #809 Joined: Wed May 30 2007, 08:59AM
Location: Melbourne, victoria
Posts: 114

great work, now i really want to build a bigger tesla coil!!!

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