[Prev][Next][Index][Thread]

Re: A coil design (fwd)



=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
 Chip Atkinson 
 http://bhs.broo.k12.wv.us/homepage/chip/info.htm
 --- Tighten it 'till it strips and back off half a turn ---
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

---------- Forwarded message ----------
Date: Mon, 3 Feb 1997 14:39:04 +1200
From: Malcolm Watts <MALCOLM-at-directorate.wnp.ac.nz>
To: Chip-at-pupman-dot-com
Subject: Re: A coil design

Hi Chip,
           I sent a very similar post to this a bit earlier. I wonder 
whether you would toss it out and post this one instead please. There
was a minor error in it.

Thanks,
Malcolm
------- Forwarded Message Follows - - - - - - -


Hello Tom (McGahee) and all,
                               Tom asked me off-list.....

> Malcolm, if you had to give advice to someone getting ready to
> build their first DECENT Tesla coil, what would that advice be?
> Bottom line: Assume the person can get their hands on 12KV 60 ma
> worth of neon transformer, 3/8 inch copper tubing, a Richard Quick
> style spark gap, PVC pipe in any standard size, and any size double
> formvar magnet wire. Assume they can make any size Toroid. What
> would you recommend? Operating frequency? Height/Diameter of form?
> Size of wire? (also length of wire or # of turns) Close or
> spacewound? Primary type, size, inner and outer diameter, # of
> turns. Type and size of interconnect wire. Capacitance? Toroid
> specifications? Mounting of toroid?  What size spark should they
> be able to expect from the Toroid? What would be its approximate
> maximum terminal voltage?
> 
> If you could distill out some practical guidelines for such a
> simple beginner/intermediate Tesla coil, that would greatly help me
> in one aspect of my writing the Guides. I would like to be able to
> work a person all the way through the design procedure for an
> actual, DECENT Tesla coil that would make them feel good about
> having made it. Many times it is helpful to not just show a
> formula, but to actually work a person through the application of
> that formula in a real-world case. If you could help me here I
> would greatly appreciate it. I am sure you are very busy. It would
> help even if all you did was gave me a decent operating frequency
> to begin with.

I decided to stick my neck out and post the reply to the list despite 
coming up with this one off the cuff. It shows the sort of approach I
take to a specific design knowing what I do now. So here goes......

Primary Cap : 13nF or so for a static gap (based on US mains 
frequency and transformer).
Wire Size : 3 skin depths at final resonant frequency (1 skin depth 
in copper is about 66/SQRT(f) mm. ).
H/D ratio 2.5 - 3.5.
Secondary Height : about 18"
    One person on the list has scored 4 feet from the transformer you 
name running a synchronous gap system. I have scored 3 feet using a 
static gap. H=18" allows about 3 times the secondary length in output.
If that were so, I'd go for a coil around 6" diameter or more bearing 
in mind that it has a significant effect on Cs which in turn is going 
to affect Vo. The other side of the balance demands the lowest 
frequency you can run at for efficiency. Cself for those figures is 
around 9.3pF. You would want around the same amount or more in the 
top hat since you want top storage and a decent ROC for long sparks.
Turns : around 1000 or so (too few yields a higher than optimum 
operating frequency and too many yields unacceptably small wire size
compromising secondary performance). 
   Using Wheeler's inductance formula, this gives Ls around 43mH. If 
Ctot is about 18pF, this gives fr about 180kHz. Check to see whether
the wiresize is suitable: Using the skindepth formula -
dwire = 200/SQRT(180,000)mm = 0.47mm/turn. The insulation for this 
size wire would add around .05mm/turn, so I would use 0.45mm (copper 
diameter) wire and accept a few less turns which brings dwire within 
cooee.
      You can lower fr still further by using an even larger top hat
than the one suggested below without serious degradation of secondary 
losses.
      This yields Lp = 60uH which seems like a nice value. Zsurge is
around SQRT(Lp/Cp) which is about 68 Ohms. This gives a crude peak 
current estimation of around 176 Amps ( you really need to quantify 
this using the DE's). I'd use a flat spiral (max clearance from top 
of secondary) with a guard ring above it and an inner diameter around 
that of the coil (I'd be surprised if k was less than 0.15 for such a 
scheme and it might well be too much for this coil). I assume 
secondary to be sitting 1/2" or so above the primary which means an 
acrylic slab can separate the two. The secondary earth wire is a 
problem though. It would have to run over the top of the slab and be 
extremely well insulated. Perhaps a primary of larger diameter with 
the earth wire dropping through it might be better. 

     Using 3/8" pipe would seem a good choice: with the high surge 
impedance we want circuit reactances rather than resistances to 
dictate the gap current (throw the onus on the gap). This coil would 
demand a professional pulse cap for best results.  
     Terminal: I'd use a 6" ring, 12" OD toroid to get near the chosen 
Ctop and give a good ROC. You could probably use a larger one to even 
better effect. The coil would look nicely proportioned like this IMO.
    Well, there's an off-the-cuff design. Estimated Vo assuming 
60% losses due to the gap and k chosen would be around 210kV. That 
amounts to about 300V/turn worst case (at the bottom) because the 
voltage distribution is sinusoidal to a first approximation.

     Re the rest, it really depends on where the person wants to end 
up. Is this to be the only power supply this coil will be run with?

If you try it, let me know. I haven't built it but I'm tempted. It 
sounds rather nice. You may want to check the arithmetic. Comments
from anybody welcome on this one.

Malcolm