Salt Water Capacitor
This is my saltwater capacitor. I built it because I blew up my 47nF MMC and I'm too poor to fix it, and because I've always been curious about saltwater caps. Tesla coil builders love to hate SWCs because they are bulky, heavy, and inefficient. On the plus side, they are cheap to build and quite sturdy. Some coilers like to point out that Nikola Tesla himself used banks of champagne bottles in the tank circuits of his giant lightning bolt machines. Of course, Tesla was a perfectionist (except at business finance?), and he selected a very specific kind of Champagne bottle with an unusual glass formula having superior RF qualities.
My SWC is made of 52, twelve ounce (355ml) Miller Genuine Draft beer bottles tightly packed onto a 19" x 19" x .125" square of scrap Aluminum plate that happened to be in my junk collection. The bottles are wrapped with heavy-duty Aluminum foil stuck on with spray glue. I covered as much glass as possible, including the bottom and sides of the bottles right up to the junction of the neck and shoulder. My cheap, Chinese DMM says this thing is 42.2nF total. That makes the bottles about 811pF each.
The outermost bottles are cemented in place with hot glue. The inner bottles are just sitting there. The bottles are wired in parallel with 17 gauge bare Aluminum electric fence wire. The strings are then tied to a common bus strip. High voltage connections to the bus strip and the Aluminum base plate are made via large copper grounding lugs. The electrolyte is 5 pounds of rock salt dissolved in 5 gallons of hot water, and I used almost the whole 5 gallons! This cap is heavy--the water alone weighs 40 pounds! I floated 1/2" of motor oil on top of the brine in each bottle. Construction of this cap was tedious and time-consuming, but not especially difficult.
Performance is most satisfactory. I've pounded this thing with 4400VA at 11.5KVAC for several longish runs, and it didn't even get warm. Output streamers are long and intense. Notice the very thick strike rail hit. I think spark production is comparable to my 47nF polypropylene & foil MMC. For years I've read that glass capacitors are poor performers, but I'm just not seeing it. Also, I get an occasional streamer hit directly into the top of the SWC with no apparent injury. I dare you to try that with a plastic film cap!
I've read other web literature claiming that salt water caps tend to overheat so quickly that the glass can warp or crack. I've never had one warm above room temperature, even on long runs at pole transformer power levels. I've also read that glass is fragile and that high voltage punctures are common. Again, I've never had a bottle fail in service, and I've built over 150 bottle caps. I used bottle caps with my pole pig coil system under the harsh punishment of an asynchronous rotary spark gap, and experienced no punctures--ever. I even used bottle caps in resonant charging mode with the junk box coil powered by a 15kvac NST, and the bottles never failed. In fact, I've had such good luck with salt water caps, I decided to build abigger one.