[TCML] Tesla coil using ferroresonant transformer

[Chris <spamsandwich@xxxxxxxxxx>]

My old post on the home-made 5.6 kV ferroresonant transformer is updated
here. Photos show the practical application of this design in my first
working Tesla coil.

It was tuned using a storage oscilloscope, initially with an AA battery
shorting to the Tesla primary then later by connecting a NiMH car racing
pack to the iron-core transformer primary. The core alone stores enough
energy to operate a 0.5 mm gap. 

http://tinypic.com/r/whi44n/6

http://tinypic.com/r/35ip3jt/6

http://tinypic.com/r/8x5wdv/6


The high voltage transformer is wound on four ungapped steel C-cores.
Each pair has 20.5 x 40.5 mm cross-section and will support 1.9T. 

V = 4.44nAfB = 4.44 x 8000 x 0.0205 x 0.0405 x 2 x 50 x 1.9 = 5603 V RMS

This gives 8 kV peak configured with a center tap to ground to keep
creepage and clearance rules in line with the IEC60950-1 safety
standard.

A sphere gap is set to this voltage; 2 mm is indicated by the literature
on voltage measurement using spark gaps.   

Twelve 942CD12P22K-F (1200 VDC 0.22uF) parts in series form the main
capacitor bank. Their datasheet AC voltage rating is observed first of
all with their DC rating being adequate considering the spark gap
setting. Using screw terminals for joints; solder is too ohmic.

Low pass RC filters at the transformer output shunt the 280 kHz signal
to ground. The main bank is 18.3 nF while the filters are 3.3 nF. The R
element is provided by pairs of long 1k 10W wirewound resistors in
series. These also choke the RF output from the spark gap.  

Self-resonance between magnetizing inductance and the capacitor banks is
40 Hz. Saturation of the core raises the resonance to line frequency to
provide the magnetically regulated voltage with little reactive drive
power under no spark conditions.

An external gapped shunt inductor is designed to be just below
saturation at normal line voltage. LC voltage doubling at the main bank
is now possible to help a fast recharge. Operation of the spark gap
drives the shunt into saturation giving unlimited current to recharge
the bank several times within the half-cycle. Unlike a NST, there is no
magnetic shunt and no current limit.

The system needs 48 VAC power at 20 A. Half of the power is wasted as
heat in the white PVC insulated windings. Copper becomes expensive these
days. Four kilos of 1.8mm wire would be needed to wind the bobbins
properly for direct AC line input. (Anyone have some spare in Europe?) 

Spark output is pretty much as expected based on the system VA minus
obvious Ohmic loss; a good 40 cm flame but the system has never been
tuned or optimized. 

Our vacuum cleaner sucks the gap and more significantly masks the noise
so the neighbours don't notice the Tesla coil. Do dirty vacuum cleaner
bags work as a catalytic converter to control ozone production?

Also note the cool yellow "thick Ethernet" cable from the scrap yard.
This stuff has about 6mm^2 of copper cross section with double braiding.



_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla