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

Re: [TCML] measure VDG voltage



Hi Jim,

The ball gap used was 1.25" polished brass spheres (solid brass). The gap arced at a gap distance of 0.35" (I started at .5" and closed the distance in 0.1" increments). The gap arc voltage was 25 kV.

I hear what your saying and your right that it will vary. I'm not disputing that at all. The arc voltage will vary for many reasons. But in a situation with a gap of normal TC size, likely between in the 26kV range. My point is simple. Why use the upper end? Why not average the value for better precision.

I believe the arc voltage is based on electrode size and gap distance as the major factors as long as the electrodes themselves are not jagged or so small they are near point gaps. When you consider Tesla Coil sized gaps, that's when I'm saying 26kV is a better number on average to use (not the whole spectrum of gap sizes and gap distances). I'm thinking mainly about Tesla Coil gaps and the gap sizes we use on average.

Take care,
Bart

Jim Lux wrote:

For how long a gap?  If the gap is much more than 1/10th the size of the spheres, it's not a uniform gap anymore (for which that nominal 30 kV/cm number applies).  The actual calculation for spheres is a pain (no exact analytical solution that doesn't involve infinite series), and, there's a lot of stuff that can affect the exact voltage (surface finish, dust, etc.)

Most people use the tables (which can be scaled, to a point), and trust that you get around 5% accuracy, when you're looking at multiple tests averaged.


There is a paper I ran across recently in the context of high power RF breakdown at 100 kHz kinds of frequencies which looks like the breakdown for CW RF could be half that for DC or line frequencies.

I can't remember the author, title, or journal, so I'll have to wait til after the first when I get back to work.. It's laying on my desk.
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla