I've wound a number of Tesla coils and a 1 KW induction coil with ~13
miles of wire in the secondary... According to the DC resistance, I
nailed the length of wire to within 300 feet of my design goal - not
too shabby out of 13 miles.
First few TC's I did were by holding the PVC tube and guiding wire on
while rotating the tube in my hands. That got old fast.
I built a coil winder with a large PM DC motor. Fixed the motor
horizontally, to a 2X6 plank ~50" long.
I used a reel drive motor from a computer (1" X 15" tape reels), but
there's lots of "treadmill motors" these days with similar power and
size on the surplus market.
For the mandrels, I laminated a pair of spruce 2X4's together, drilled
a hole through the center and mounted a short length on the motor by
just drilling the hole undersize and beating it on with a hammer
(healthy motor with a 3/8" X 2" shaft).
I turned the mandrels using a variable power supply (24 volt battery
charger with a variac to control voltage) and some wood chisels, rasps
and sandpaper to make a friction fit to the 1-1/2" to 3" PVC coil
forms. Turning wood is a lot of fun....
The fixed center of my winding lathe, was a construction out of wood
using a 3/8" bolt as the lathe "center." Turn two mandrels and leave
one on the motor and one on the center, with the PVC tube between.
Winding is a snap - took less time to build the, admittedly crude,
lathe, and wind the 3" wide long coil, than it would have taken to
wind the coil alone.
I wound a total of 12 coils in the 2" to 4" diameter range and had
loud noisy sparks 8 feet long out of some of them.
They can easily be coated with thick layer of epoxy or varnish by
leaving the coil on the lathe, rotating it slowly, while using a
credit card to spread the epoxy evenly. It rotates slowly while the
epoxy sets up so there are no sags and you get beautiful results.
I just set the power supply so it is tensioning the wire then hold the
wire back with one hand and guide it on with the other as the motor
turns. Precut masking tape holds the wire in place long enough to
cement the end down (under tension).
The induction coil was the same deal, but I made four large bobbins
out of 3" PVC for the cores, and 1/8" thick Plexiglas donuts for the
end plates. They are 3-1/3" wide, since I had plain adding machine
paper on rolls to use as layer insulation.
I'd wind a layer using the TC winder with one mandrel to hold the
bobbin (and a shorter bed on the lathe for this project). A layer
took about 2 minutes using #32 AWG wire. Then tape it in place long
enough to paint it with quick drying varnish to hold the turns, wrap
it with a layer of paper and do another. Took about a week of spare
time to finish 4 bobbins with some 24 layers on the outside coils and
30 on the inside ones. Took about a month total to build the
induction coil. I wound the primary in four layers and brought out
the starts and finishes so I could wire them in series or parallel to
use as an AC HV transformer or DC induction coil.
AC performance is ~10,000 volts and DC unknown (but jumps a 5" spark
with no problems) The whole thing got vacuum impregnated with wax
(using my pressure cooker on the range top with a vacuum pump attached
to it filled with molten wax - the wound bobbins soaked in the oven
overnight on "warm" to get the wax to flow easily)
My stove is electric - I wouldn't do it that way with gas since the
paraffin is flammable and the varnish is hot and out gassing residual
solvents when the vacuum pump is running. Hotplate outdoors with a
fire extinguisher standing by makes more sense....
Oak box, porcelain conical turret insulators, with brass wood screws,
and a largish 1930's X-ray machine ammeter on the primary, and switch
box to serve different primary combinations for different voltages.
The coils are potted in wax, the iron protrudes out either side of the
box. Weight is ~35 pounds and it should have handles or wheels on it.
Another, perhaps easier, way to do it is with "pies." Bobbins are
only 1/8" wide with plastic film side plates, and the wire is jumble
wound on - depending on the large number of bobbins to provide
insulation between layers. I was working from a 1930's book on the
subject and just adapted what they had to modern materials.
I use the induction coil to excite the Tesla coils - I found that up
to around 200 hertz the TC performs much better. They resonate at
~100 KHZ. Higher and either the iron loses are too great or the waves
aren't damped out before the next spark comes along.