A few questions reguarding my first NSTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

A few questions reguarding my first NSTC

1 2 3 next

Move Thread

LAN_403

Jrz126

Wed Apr 19 2006, 03:03PM

Registered Member #242 Joined: Thu Feb 23 2006, 11:37PM
Location: Erie PA
Posts: 210

I'm building my first tesla coil and I have a few questions...
First off, I'm running a 15kV 30mA NST with a cap bank of 10.7nF
My secondary consists of ~1000 turns of #26 wire on a 4"pvc, and my primary is 50 ft of 10 gauge copper at a 30 degree angle. I'm also running an asnyc spark gap.

Now for the Q's:

NST related
I have a noise filter for the low voltage side of the NST. Is there a polarity to it? (the connections are just numbered 1-4 and a ground).
Where would I connect the grounds of this filter and the nst?
What is a good cutoff frequency to use for the 'terry filter'? (RC combination). I cant seem to locate the schematic for it.

Spark Gap
What break rate should I be going for? I'm using a DC servo motor with an analog encoder. I figure its spinning at about 6000rpm at full 24Vdc. (with 4 electrodes, its roughly 400 BPS).
Also, I'm using tungsten for the stationary electrodes and some 1/4-20 brass bolts for the flying electrodes. Is this a good idea? I think the brass might get too hot.

Topload
I'm still working on the final design of the topload, but in general, is it better to have a larger or smaller topload?
I'm using some 4" aluminum duct. should I stick another 6" duct on top of that?
What should I use to cover it to get it nice and smooth? I was thinking of aluminum foil.

Wow, I guess I had quite a few questions there...I cant wait to start making some sparks

vasil

Wed Apr 19 2006, 08:03PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

1. I think it has to be used in reverse. The ground of the filter goes to the mains ground. The case of the NST and the middle electrode in the safety gap go to the secondary grounding (RF grounding). It has to be large range freq (the SGTC emittes a large band of freq, especially when arcing). Terry filter is here:

http://hot-streamer.com/temp/NSTFilt.jpg

2.Do not use low BPS because the larger energy bang can produce racing sparks. Start from 120 BPS (if 60 Hz mains frequency) and go higher. It seems that NSTs are sensible going with an asincronous spark gap (but maibe is a simple mith, i dont know for sure). I would use tungsten for all the electrodes. The brass can work for flying electyrodes (some cooling by speed rotation). Use it first for low run times and verify the temp of the electrodes. Depending by that you will decide how much time can run the TC at once.

3.You have to try different setups. Use the 4 " first, then 6 " and look for the best output. Using the smaller one you will see the tuning option you have (if have enough primary turns for bigger toploads). Alu foil works fine for nice, smooth toploads.

vasil

Hazmatt_(The Underdog)

Wed Apr 19 2006, 10:28PM

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

I also run an async gap and its a mother of a job to sync it up. I have to figure out a controller to do that yet.

Why I'm commenting is that I have been doing a LOT of looking into best performance capacitence matches, and 2xCres is not a good match. I have verified this in simulation with transformer losses included of course. You can even verify this if you want. Pspice is pretty accurate when it comes to this.

I have had best performance with my 12/60 at 1.2xCself while testing ALL of this out! I did a lot of testing here. I was able to do this with my laser caps. I went as high as 2xCself, which I strongly believe caused the death of my NST. You can look around, my posts are all over.

Anyways, what I am getting at is after the death of my 12/60 I investigated this as much as I could, then went back and calculated for my 12/30, ran simulations and found the best match for the 12/30 was 9nF, 1.35xCres! Tested this on my system, and had beautiful results! about 2 feet of arcs from this small transformer.

So whats the point?

I am finding that high capacitences do not give good results at all!
12/30 from my calculations should be 9nF, I am running 10nF because my caps incriment at 2000pF each
12/60 should be 18nF for optimal, which maintains terminal voltage and current at rated values
15/30 would be 7nF
15/60 would be 14nF

You don't have to take my word for this, but that's what I am finding in simulation. I have written Terry and want to start a correspondence about this for much further investigation!

Steve Conner

Thu Apr 20 2006, 09:24AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hi,

If you're using an asynchronous rotary gap with an NST, you want to keep it running well above 120bps at all times. If it runs too slow, the voltage might surge dangerously high and maybe destroy the insulation on your NST. I would be especially worried in your case, since the cap bank value you chose will be near to 60Hz resonance with the NST's inductance.

A breakrate of 300-400 is probably about right, although with the power supply you have, probably not all of the electrode presentations will fire, as it can't charge the capacitor quickly enough.

You should have a properly set safety gap in parallel with the rotary anyway, in case the rotary loses power and slows down.

GimpyJoe

Thu Apr 20 2006, 12:42PM

Registered Member #316 Joined: Mon Mar 13 2006, 01:30PM
Location: Marietta, GA
Posts: 212

A while back I read one of hazmatt's posts about async. gaps and he had the idea to use a static gap in parallel with the rotary gap so you don't have to worry about missed firings.
In my coil with a 12/30, the best capacitor size I found was 6.3nf. Faster break-rates seem to be better for smaller coils.
I find the easiest way to make a smooth toroid is to use drier ducting and form the toroid, then wrap it with two or three layers of masking tape before covering it with al. foil. Use skinnier strips of al. foil to avoid wrinkles. For a break-out point, use a brass drawer pull to get long streamers.

HV Enthusiast

Thu Apr 20 2006, 03:15PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

I wouldn't recommend an asynchronous gap with an NST powered tesla coil. Stay with a synchronous gap.

Hazmatt_(The Underdog)

Thu Apr 20 2006, 07:15PM

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

I would have to recommend a sync. gap as well. Async's are just too much trouble unless you have a thoughtfully designed speed controller, which is a project I have yet to design.

Go with a synchronous motor and you'll be set. Turn it on, wait a few moments, done. No worries! It's rpm will change a bit on you at like .5 Hz or so. I noticed this with my sync. gap a long time ago. But no worries.

HV Enthusiast

Thu Apr 20 2006, 10:27PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Its not that they are too much trouble - in fact they are easier in my opinion since universal motors are a dime a dozen while synch motors are difficult to find.

Asynchs just don't work well with NSTs and have their own set of problems which was already mentioned.

Jrz126

Fri Apr 21 2006, 02:18AM

Registered Member #242 Joined: Thu Feb 23 2006, 11:37PM
Location: Erie PA
Posts: 210

I double checked my cap bank, its 9.4 nF using 4 strings of 20 .047uF 1600V caps...If I knock it down to 3 strings, I'll be running 7.05 nF. Would there still be a risk involved with this cap bank?

How close does it have to be to the 60Hz? I wouldnt think it'd be all that hard to build a controller, as long as there is alittle error is acceptable. I might have to dig out my pic microcontrollers and give it a shot

Hazmatt_(The Underdog)

Fri Apr 21 2006, 03:00AM

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

Well, It can be difficult because the controller I envision is a PWM motor control with PID compensator and optical feedback. The PWM shouldn't be too much trouble, but getting the right PID is difficult because, as my controls instructor told me, sometimes one works when many others have not.

I wish I had the time, I would do it and give it out to everyone. The only luxury I have been given is the time to come here and talk, and I really want to do projects for you guys.

1 2 3 next

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