Re: 4 Inch tesla coil - Inquiries

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "Barton B. Anderson" <bartb@xxxxxxxxxxxxxxxx>

Hello,

Lot's of detailed information and annotation regarding your coil. 
Thanks for that. I think most would agree that the spark gap is the 
biggest are of concern, but I'd like to intersperse comments. See below.


Tesla list wrote:

> Original poster: "Hydrogen18" <hydrogen18@xxxxxxxxxxxxx>
>
> I've got a 4 inch tesla coil that I've pulled out of storage and set 
> about improving and trying to make into a more logical setup. Specs 
> are as follows
>
> Secondary - 4 inch Schedule 80 pipe, 1500 turns 28 awg, 18 inches long

If it's 28 awg magnet wire, it's about a maximum of 1230 turns if 
closewound with no winding error for an 18 inch winding length. That 
is based on thin insulation as well. If it were double or triple 
insulation, the turns would be even less. Not a big deal, but I do 
like to note discrepancies. To verify, mark off 1 inch of winding and 
count the turns in that 1 inch space (I like to simply use masking 
tape to mark off the 1" area). You should count about 68 turns for 
1230 turn coil.

> Primary - Coil of 1/4 inch copper pipe, about 13 turns worth. laid 
> out flat and zip tied to plexiglass. 1/4 spacing between turns. 
> about 2 foot overall diameter.

Wow, that must be quite a large inner primary hole (3.25 inch spacing 
between primary and secondary). Yields a low coupling. If you even 
rebuild the primary, you could certainly tighten that up to about 1" 
to 1.5" spacing).

> Spark gap - Static style, 1 inch spherical brass drawer pulls 
> mounted to 1/2 inch copper pipe. This can be adjusted really easily. 
> A cheap blower supposedly running at 25000 rpm blows alot of air in 
> an uncontrolled fashion across these. They get pretty dang hot, but 
> the arc seems to be quenched pretty well, although I have no first 
> hand experience with spark gaps outside of my own design.

Biggest problem is the spark gap. The brass drawer pulls are often 
solid brass. Once they heat up your discharge voltage across the two 
sphere's will decrease and output will suffer dramatically, even with 
a fan across them. You would probably gain the best performance 
rebuilding the spark gap to maybe a heavy duty static gap such as a 
RQ/TCBOR style (pipes mounted inside a PVC tubing with a goodly fan 
either sucking or blowing across the pipes). I recommend large 1.25 
inch diameter pipe segments (available ready to go in 3" or 4" 
segments at most hardware stores). They will cool much faster and 
will maintain a steady state temp with proper air flow.

> Topload - 12 inch toroidish thing with a section of 4 inch dryer 
> duct stacked around it. Seems to discourage primary strikes pretty 
> well, although it was designed for such

Should be fine.

> Tank Cap - 942C20P15K's(2000 VDC 0.15MFD) in strings of 8, up to 3 
> strings of these in paralell(plenty more strings ready to be 
> assembled). Mounted in a wooden box but the capacitors themselves 
> are in 180 degree sections of PVC pipe for insulation. Connection is 
> made at all junctions to the strings by means of 1/4-20 brass bolt.

This is fine to. The full cap bank capacitance is 0.0563uF. Based on 
your system, you are probably tuning in about 5 turns. The cap size 
explains the narrow tuning range as well. I'll comment later on that.

> Power supply - 4 MOTS producing around 9500 volts RMC AC, current is 
> limited by 10 lbs 10 AWG wrapped around a 3 inch piece of Schedule 
> 80 with an adjustable iron core(I have also left the shunts in the 
> MOTS, as I see no need to drive them as hard as some people do; mine 
> are not under oil but epoxy potted). Driven by a 30 Amp 240 VAC 
> outlet, I do not feel this is the limitation in my setup.

Should be good for lots of current. No problems there.

> So far my recent attempts have yielded point to point strikes of 28 
> inches or more; longer arcs accounting for the bend. That was with 2 
> capacitor strings. Attempting to use 3 capacitor strings seems to 
> make finding a tuning point impossible, all I suceeded in doing was 
> getting the primary so hot I burned myself(this was in just a second 
> or so of having the spark gap firing). I am dumb tuning this coil, 
> just starting at turn one and moving out until I get a large 
> response, and then making smaller movements from there. I've made 
> the following conclusions:

Yes, less capacitance will increase the range of tuning. There's a 
ratio of C and L for any given resonant frequency. The more C you 
have, the less L you have to play with.

> - Steel anywhere in the primary circuit basically destroys 
> performance and likes to get really hot.

It can if it's close enough to the field.

> - Tuning seems to be very pecuilar on a lower coupling coil(I had 3 
> inch coil in the past with 'high' coupling between the primary and 
> secondary, you could make turns of adjustment to the primary without 
> noticeable output in spark).

It's not the lower coupling, it's the fact that the tank cap value is 
so large compared to the inductance required to tune into the 
resonant frequency of the secondary. If the cap were 10 times 
smaller, your tuning sensitivity would decrease by the same amount.

> - The charging current available to the coil doesnt seem to affect 
> things much, other than the spark gap getting louder.

The charging current, as long as it is beefy enough to charge the 
caps within a decent time frame will never be problem. But if a weak 
tranny were used, you would then experience charging problems.

> - This spark gap is better than the last, because I don't have to 
> wait for it to cool down between runs. It takes a very long time run 
> time for the initial breakdown voltage to change due to tempature of 
> the electrodes. They are hotter than a car brake rotor at the parts 
> you can observe, but I suspect where it matters they are remaining 
> cool. They also dont require a resurfacing every 30 seconds of runtime.

Still a problem. Solid brass will not perform as well as properly 
designed copper pipe gap unless something special were added to them 
to allow quick dissipation of the heat. Of course, it's very easy to 
build a poor copper pipe gap as well. Large radius, proper spacing, 
and well designed air flow are the main ingredients for a nice 
running pipe gap.

> And I've got these concerns:
>
> - My primary is very poor. The spacing is nonuniform. I think tuning 
> is changing because the turns are not held very securely.

Nope, tuning is not changing. A primary with a lot of gap to gap 
differences will only look ugly, it will still perform just as well a 
nicely wound primary. The only time it might cause a problem is if 
two segments were close enough to arc together. Tuning will not 
change unless the caps were heating up and changing their value. Most 
likely, primary and gap heating is simply changing the arc voltage 
and decreasing performance. Easy to identify: basically, it starts 
out good and after a little while of running, sparks start 
diminishing in length.

> - Should I add another gap to my static spark gap? Are there any 
> telltale signs of an overloaded static power gap? I know that the 
> signs of a 'power arc'(the deep orange-reddish arc that does not 
> quench) or rapid electrode erosion are two to look for, and I have neither.

Look for heat. If their getting hot, your definitely going to have 
problems. I would rebuild this part of the system first. A lot of 
people here have built a lot of different gap types that work well. 
Almost anything would beat the brass sphere's on a coil like yours 
which has a big tank cap and lots of energy.

> - I have a safety gap on my transformer, so I'm not concerned about 
> opening my spark gap up too much. But is there a primary side side 
> RMS VAC that I shouid aim to have the spark gap breakdown at?

Opening the gap will increase the voltage required to arc the gap. 
This won't help and only puts the MOT stack at risk. For 1" brass 
sphere's, about 0.18" is about right for 10500V output. I wouldn't go 
beyond 0.2 myself.

> My questions are from here:
>
> - Should I bother improving the primary? I'm not getting turn to 
> turn arcing, it's the inconsistency of the shape that's my concern.

I would. Mainly for raising the coupling. Again, the shape is not a 
concern, but low coupling can be if it's too low and your energy is 
large (which it is).

> - What's the maximum size capacitor for this coil I can reasonably 
> expect to run?(I know my limitation is the spark gap, and probably 
> will be for all my coiling years given the easy availability of 
> large dangerous HV transformers)

Well, actually, depending on the MOT's, you could probably go to a 
larger cap size. The problem is that the coil is small for such 
energy and a larger cap is unnecessary for spark lengths. Your better 
working on reducing the thermal losses.

> - Why is my primary getting so darn hot? Is off resonance driving of 
> the coil the cause of this?

Well, first, the primary tank energy is high. You are dumping a lot 
of current into the primary. Secondly, the coupling is quite low. 
This can be similar to running the primary without a secondary in 
place (which is a big no no). You need a load to dump the energy to, 
otherwise, the primary has to get rid of all that energy 
somehow.  Heat is natures backup mechanism. I would lower the 
capacitance from your two strings of 8 to two strings of 16 for 
.0188uF. This will increase the primary tap requirement to about 8 
turns with your primary as is. A few reasons is you will put more L 
into the circuit, decrease that energy to something the coil would be 
happy with, and your peak voltage out the tranny is nearing 15kV (the 
8 string is good for 16kV which is a little close). Adding the extra 
caps per string will help the voltage rating of the cap bank.

> - Is there a better(intelligent, mathematical) method of tuning the 
> coil I can employ?

Use a design program would be beneficial so you know at least the 
ball park for any given configuration.
http://www.classictesla.com/java/javatc.html

Also, you could employ a generator and oscilloscope to do the same.

> - What kind of spark length performance should I expect from this coil?

Far better than your getting. There are just too many losses 
throughout the system robbing all your power. For reference, here is 
a similar coil in size but only uses a 12kV 60mA NST. This one also 
uses a .0188uF MMC.
http://www.classictesla.com/photos/ba45/ba45.html

> - Any other obvious mistakes?

Heat is the obvious problem. That's the thing to work on. Also, if 
you do use a smaller cap size and you want to rewind your primary, 
remember 1 to 1.5" clearance.

Take care,
Bart