I recently came across a bunch of 590pf 30kv doorknob capacitors on ebay for a great price. I bought 30 such capacitors to be used for this project. The idea is to build a twin Marx generator, capable of 900kv total output! The total energy when both units are completed will be just under 8J, but I'm not looking to break any energy records here.
As of 11/13/04 I have constructed and tested one of the units. I was surprised to see that I could coax a 16" spark from a 15 stage device!
Construction was very straight forward, and I actually built the first tower in a matter of hours (built it in a day). The upright supports are HDPE, 1/4" thick, 4x20". I taped the 2 pieces together so that when I drilled them out, the holes would match exactly. The charging resistors were found on ebay. They are 2W carbon composition, 4.7MEG, and were manufactured in 1968! They are great at standing off the high voltages involved despite their 700v rating. 4.7MEG is a bit large in my opinion, 1MEG would have been preferable, but the price was right on these. The spark gaps are simply 8/32" acorn nuts. The spark gaps are set to 1/2", though at the moment I may need to space them farther apart (11/16") due to premature firing issues. In any case, more experimentation is needed at this point to optimize the design and perhaps achieve even greater voltages.
See the construction photos below, they explain a lot:
First light (max spark 16"):
After some experimentation I found something somewhat remarkable (at least to me). Depending on the polarity of the output, spark length was limited to 11" or 16". I found that using a negative charging supply produced the 16" sparks, while charging positive only gave 11" sparks at best!
I also made an attempt to get the twin Marx bank to fire, but this only resulted in an 18" spark and a dead power supply. I'm still trying to figure out exactly what the problem is, when I figure it out details will be posted here as to what it take to get a twin Marx generator to work.