Simple Rules for Producing Longest Sparks Using Minimum Power Input  ! 
 ( John Freau -1998  )
More about TC efficiency theory at John's homepage:  http://hometown.aol.com/FutureT/page5.html
From:  FutureT@aol.com [SMTP:FutureT@aol.com]
Sent:  Thursday, August 13, 1998 5:49 PM
To:  tesla@pupman.com
These rules may be helpful for some who may wish to experiment with sync-gaps at low break rates.
In a recent posting, I showed the better efficiency I obtained using these methods. It is important to have 
 a well matched system when using sync-gaps. These rules are based on my 42" and 64" spark TC's. 
It should be possible to scale the results up or down, by following these rules. These rules are just a
 rough guide, and some slight non-linearities may be seen with up or down scaling. Results will be
 affected by transformer losses, design techniques, etc. These rules are based upon a square law for
 power versus spark-length scaling, which seems to fit well with empirical results. 
1, Quelles est la long, d'étincelle voulue 1,22 m
2, La long, d'étincelle/2,5 donne approx, le dia du tore 0,49 m
3, Diviser la long, d'étincelle par 10  0,12 m
donne la grosseur du tore (Approximative)
4, Puissance de l'alimentation 798 watts
5'.  Mains Power frequency f = 50 Hz,      BPS = 2*f =  100 breaks per second
5. (Watts/BPS)/1.5= Joules/bang (the1.5 factor covers losses)  5,32 Joules/bang
6. Minimum rating for transformer = power input (watts) from (4) 
    above, for non-shunted transformer. (approx)  798 watts
7. Voltage on capacitor, Vc = input voltage times 2.22 times
     transformer nameplate rms voltage rating  ( =   8 kV  ), 
     for resonant charging with some losses thrown in. (approx)  17,76 kV
9. Capacitor size; C = 2 J/Vc^2 (joules from (5) above) (approx)  0,033748 uF        = 33,75 nF
10. Spark gap; synchronous rotary, 100 BPS
11. Primary and secondary coils; use traditional designs.  ...but learn more at John's homepage:
 http://hometown.aol.com/FutureT/page5.html
12. Inductive ballast; adjustable as needed. 
Let's try an example, and aim for an 80" spark. 80/2.5 = 32" toroid dia.  80/10 = 8" = toroid thickness.
 (80/1.7)^2 = 2215 watts. (2215/120)/1.5 = 12.3 Joules/bang. Minimum non-shunted transformer rating = 2215watts.
 Voltage on capacitor, (assuming 14.4kV transformer), Vc = 14.4kV * 2.22  = ~32kV.
 Cap size = (2 * 12.3)/32^2 = 0.024uF.
 For a small coil, aiming for a 20" spark: 20/2.5 = 8" toroid, 20/10 = 2" toroid thickness.
 (20/1.7)^2 = 138 watts. (138/120)/1.5 = .77 Joulesper bang. Minimum non-shunted transformer rating = 138 watts.
 Voltage on capacitor, (assuming 10kV transformer), Vc = 10kV * 2.22 = 22.2kV.
 Cap size = (2 * .77)/22.2^2 = 0.003uF.
 For a tiny TC, aiming for a 5" spark: 1/2" by 2" toroid, 8.65 watts, .048 joules/bang, assuming a 6kV transformer,
 Vc = 13.32kV, C = .00054uF. It is unknown if these rules will track "linearly" down to the small TC levels above.
Enjoy!  ( John Freau 1998)
New spark length formulas, as of April 27th, 2000:
inches cm
Spark length (inches) = 1.8*sqrt power input (wallplug watts) 51 129
Spark length = 5.8*sqrt power input / 4th root of the bps 52 132
Spark length = (3.9*16th root of the power input*sqrt power input) / 4th root of the bps 53 134