[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

New SISG Coil - Piranha

Original poster: Vardan <vardan01@xxxxxxxxxxxxxxxxxxxxxxx>


These are all untested preliminary plans so beware!

I have written up the plans for my next SISG coil at:


It will use one of Mark Dunn's nice SISG boards and the new charging circuit topology. It will have a coil power of 365 watts. Burning about 250 watts in the resistors and the AC input current is just over 7 amps at 93 volts input. It has very good power factor but the resistor loss is a pain. But it has a big fan :-))

It is designed for 240 BPS but should do 120, 240, "more" too if you crank it up >:)) At 240 BPS it should hit 640 amps peak in the primary at either 24 or 30 amps RMS. ScanTesla and MicroSim don't agree there and I have not figured out why yet.

It is not a giant coil. The secondary will be about 12 inches of #30 wire. I was thinking of one of John's big toroids for the top. About 18 inches high from the base of the secondary to top of the toroid. 150nF at 3600V can jump 2 feet but I am using 235nF, so I would guess about 30 inches to ground. One can jump SISG sections or individual SIDACS to lower that as one wishes.

It is interesting that the primary "Q" is lower now too!!! Before with conventional gaps, the Q was devastated by the gap loss. No I am pushing the primary Q with giant primary capacitance >:-)) Big early power to the sparks seems like a winner according to vast history... Coupling can be anything I want from 0.25 down. I am thinking of having the primary coil on the "underside" of a Lexan sheet this time so there are no real exposed dangerous things. I have plenty of primary to secondary distance to play with!

In other news... The SISG was tested for voltage and frequency effects. It certainly has somewhat lower loss as the voltage is raised from 300, 600, 900 V as expected. Oddly it also does a bit better at higher frequency. That might be due to less IGBT instant heating or the IGBT and diode are hanging "on". The SISG "loves" cross conduction :o)) But bottom line is, it has no problem with primary frequency.

There is just no hope of doing first notch quenching just by playing with R4. The later currents required and the gate voltage to support those currents just don't line up. This is a little more of an issue with BIG primary caps pushing the primary "Q" down...

So it goes on...