SPARK LENGHT: 14 inches streamers, 16 inches arcs to ground.
Wound on a 11 cm diameter PVC pipe, 28 cm wound lenght, 966 turns
L = 21978.21 microH, Cself = 6.0514 pF
Fres without toroid = 436.63 kHz
Little toroid with C = 11.60 pF
Flat primary, 7 turns with solid copper wire 3 mm dia,
inner radius = 8 cm, outer radius = 16 cm, 8 plastic supports glued with poxipol on a round glass base
L = 8.09 microH, taped at turn 6
2 strings in parallel, 8 caps (100 nF @ 1600 V each) in serie per string, for a total of 25 nF @ 12.8 kV
Made with two inox bowls. Not quenched, the secret is to use large bowls for proper cooling. If the border is coroded after some time, you can rotate them and get a new shiny border :))
2 MOTs with secondaries in serie and primaries in antiparallel, ballasted with 2 ballast coil (from high pressure Hg lamps)
Ballast coil: P = 250 VA, I = 2.15 A, cos fi = 0.56 (for normal use)
measured R = 3.7 ohms
inductive reactance (Xl)= (P*cos fi)/ I^2 = 30.2 ohmi
total impedance = sqrt(R^2 + Xl^2) = 30.42 ohms
Unfortunatelly I cant measure the input current but I guess that is around 40-50 mA
The charging current of the primary cap can be calculated from the primary circuit impedance at mains frequency (50Hz)- the sparks gap is in parallel with the PSU output
Zpr = sqrt (R^2 + (Xc - Xl)^2) = 127 kohms
So, for an output voltage = 4200 V, the charging current is around 33 mA.
THE MULTI-GAP vs SIMPLE GAP
I built a new spark gap with 2 gaps, introducing an intermediary electrode ( a big aluminium heatsink), between the first initial electrodes. RESULTS: 18 inches streamers and 21-22 inches arcs to ground. The TC arcs to primary more often, I have to build a strike ring. It is runing more smooth, and the use of the alu electrode decreased somehow the noice made by the gap (easy quenching)
-2 MOTs, even with low voltage output, can be a good choice for NSTs, sometime hard to find (especially in Eastern Europe)
-the ballast coils, used for Hg lamps, are easy to find, and match for the power level you wanted. I tryed first with only one ballast in serie with the antiparalleled MOTs, but the TC works eratically, so it needs at least 2 ballasts in parallel to mantain the arc in the spark gap.
-the primary cap is cheaper because the lower voltage rating.
-the air flow in the spark gap is not necessary, I recomend just a multiple gap, with big metal electrodes, to get proper cooling and quenching. Use at least 3 gaps and make the third gap variable. The total gap in my setup is around 4-5 mm. Start the spark gap with the third gap closed (only with 2 gaps), then manually set the third gap for the best output.
-the strike ring is a must. Without it, when the arcs strike the primary, I had sparks on MOTs between the core and the primary windings. With the strike ring in place, the problems vanished.
-if you want longer streamers, use one more ballast in parallel with the first two, increase the primary cap value, increase the number of gaps and get it tuned.
