Tiny twin Tesla coil

The scrapped remains of my very first TV-flyback driven, single shot TC included  two 1.2" diameter #30 AWG wound secondaries. It would have been a waste to just throw them in the dumpster, thus I recycled them in an even smaller TC... Got some 3kV 50mA (150VA / 75 Watt) non-centertapped NSTs for free from a local scrapyard, again... ;o).

This coil is "split" i.e. a twin coil.

It has two primaries wound in opposite direction and two secondaries with a common ground and wound in the same direction - giving a 180 degree phase shift between the TC top polarities, thus making the streamers attract.

The primary is tappable so one can "switch off" one of the primary coils. This lets one secondary & primary run solo as the TC, and the other secondary act as a counterpoise (sort of, anyway).

It seems that the other secondary is getting some sort of RF base feed through the common ground connector as it gives off little streamers, too. Go figure. Bad grounding, maybe.

Considerations for optimizing: everything needs improvement.


mains side:  
controller: Solid state relais, optically switched by 12V input pulse from a timed controller.


PSU: a 220V 50Hz neon transformer, rated 150VA, that gives 50mA current at 3kV. Has only one ceramic HV terminal, the other is mains ground (I left it floating).

NST has a 16uF PFC caps (for no real reason, as there is: )

a small mains isolation transformer in between, just for fun. Maybe damps some of the RF power.

In the picture on the right, you see the mains side with isolation transformer, PFC, and small transformer and rectifier for driving the SG fan.


grounding: - isolation transformer is grounded
- NST not on mains ground (not grounded at all, really)



primary side:

spark gap: Typical static SG with copper pipes. Small and compact. Three gaps.

Has a 12V hairdryer(?) fan blowing accross. Fan supplied by a little xfmr, with rectification and all.

A quick-and-dirty job, but at that 80W power level of this coil the SG works nice.


tank cap: Self made 8.5nF capacitor

Two smoke-grey (wow...) plexiglass plates as cover. The capacitor itself is about A5 sized and 5mm thick.

If I remember right it was 3 sheets in "parallel" (foil only between), and about 5 of them in "series" (foil out from one end).

Same vegetable oil treatment during assembly.

It gets a bit warm after some minutes of operation.

Shown in the right hand pic is, from bottom up:

1 - at the bottom is the tank cap with plexi covers

2 - on top of two .3" high spacers comes a thicker plexi plate to which the secondary base is screwed onto. This plate has a broad aluminium foil strip on it for RF ground.

3 - the sandwiched and encapsulated primary coil



primary coil: It's two primary coils wound in opposite direction and placed about 10" apart, connected in series.

The inner dia 2", each has 9 turns of thicker transformer enameled copper wire (like in MOT primary) with no spacing.

The whole primary is packed/sandwiched between acardboard and transparency sheet layers. The individual turns were taped over with adhesive tape to prevent arcing. Whole thing painted with white paint. From farther away it looks pretty neat. A strike rail is immersed into the cover (that snake on the top)


connectors: betweem cap and SG: short piece of AWG#10 audio (audio speaker) cable.

primary to cap: 2cm x 2cm strip of folded-together alu foil
connectors: none, wiring permanent


grounding: shared common RF ground (bases of both coils connected to that alu strip), and a short piece of AWG#10 wire leading to some bigger metal object (the protective caging of my other, "small TC" coil).


RF filter: One 470pF+4x15kOhm symmetric RC filter for testing.

Two 15kOhm on input, 470pF@6kV filter cap in between, two 15kOhm on output.

It dramatically reduces performance. Needs redesign. Resistors dissipate most of the power.





secondary side:

coil form:
  • both secondaries are 1.2" diameter cardboard roll, ~5" high
  • polyurethane varnish cover (soaked), wax paper layer, pu varnish again
  • turns  (AWG#30 or so) coated a few times with urethane varnish
wire: AWG is close to #30, or even finer. A bit thicker than in a small audio transformer.


winding: approx. 700 turns on each secondary


coil ends: bottom: small plexi piece with connector screw so coils can be fastened on ground.
top: about 0.3" clearance between uppermost turn and an non-closed metal ring.


topload: anything.

For left secondary, a pistachios nuts box plus ugly toroid. For the right secondary, it's approximately the same assembly for in-tune, and just a little toroid for not-too-much-out-of-tune.