RE: Coil size to faraday cage size ratio

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Steve Conner" <steve.conner-at-optosci-dot-com> 

 >Oh good!!  The EMI is not serious :o)))

I think it's most likely that the EMI is getting conducted out through the
wires supplying power and trigger pulses to the coil.

Putting the 230v AC supply (for both the main power and the driver board)
through a commercial "high performance line filter" would help. The body of
the line filter needs to be screwed with several screws -or even soldered-
into the cage wall, so that the line terminals are outside the cage and the
load terminals are inside.

You will also need to filter the trigger pulse lead. I suggest using a
Faraday shielded pulse transformer. This is basically a 1:1 ferrite
transformer where one winding is plain wire but the other is miniature coax.
The plain wire winding goes to the driver board, and the coax one goes to
the trigger generator. The screen of the coax is grounded to the Faraday
cage at the point where it passes through (eg with a BNC bulkhead connector
screwed to the cage) If you want I can make one up for you and send it over
with the next lot of boards.

If this doesn't work you need to investigate another possibility. The
Faraday cage might have poor bonding between its sides, causing it to
radiate harmonics when a spark strikes the cage or when small sparks jump
between the sides. The sides should be connected to each other all along
their lengths, and if the cage has a door, this should have a conducting
seal that makes contact all around when the door is closed. You can buy a
sort of self-adhesive metal mesh "draught excluder" for just this purpose.

Also, dryer duct toroids, foil tape breakout bumps, etc. can all radiate
harmonics well into the UHF and microwave region, as tiny sparks jump
between regions that aren't solidly connected. A Faraday cage with poorly
bonded sides or large mesh might allow these UHF waves through.

Richie Burnett showed how putting a foil tape breakout bump onto a SSTC
caused it to interfere with TV reception in the 470 to 860 MHz band.


Steve C.