Re: Coil size to faraday cage size ratio

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

Original poster: "Jim Lux" <jimlux-at-earthlink-dot-net> 

At 04:18 PM 11/18/2004 -0700, you wrote:
>Original poster: "Finn Hammer" <f-h-at-c.dk>
>All,
>
>I`ve had my OLTC coil tested for EMI, and since it exceeds the legal limit 
>a 100 times, inside a faradays cage with 50db damping. This corresponds to 
>a 4Km (3miles)radius inside which the legal limit is violated.
>I am going to need a better cage.


Exceeds 100 times the power limit? or the magnetic/e field limit?



>Question: For the same coil, will a big faraday cage offer better damping 
>than a small one.
>My reasoning says yes, since the longer distance to the cage, in the case 
>of a big cage, will result in a weaker field than as in a small cage.
>If the coil produces 600kV and the distance to the cage is 1meter in the 
>small cage and 2 meters in the big cage, then the small cage has to damp 
>from 600kV/meter, whereas the big cage only sees 300kV/meter.
>
>Is this correct?, and will it matter? If yes, by which factor 1/2? :-|
>1/4? :-\    1/16? ;-)
>
>Cheers, Finn Hammer


EMI shielding is a tricky business.  You need to find out why the cage 
isn't attenuating enough.  Is it a radiated limit or a conducted limit 
you're exceeding?

For low frequencies and very much less than a wavelength, the odds are that 
the signal is being coupled out by the magnetic field, not the E field.

What frequencies and standards are you being tested to?  Usually below 
30MHz the test is done with a magnetic probe, and some calibration factor 
is used to determine compliance.  Is the calibration and test distance 
being properly applied?

Are you busting the limit for the fundamental or at a harmonic?

How is power getting into the coil through the cage wall?  What sort of 
filter is on the power line?  Is there a way to make DC outside the cage 
and feed it into the cage to run the coil?  It's a lot easier to filter to 
DC than to filter AC.

There's also a bit of specsmanship to consider, which is why I ask which 
standard you're being asked to comply with.  For example, ANSI C95.1 lets 
you time average over various periods, so even though your peak power might 
be high, the average power might be quite low.  (As you can imagine, with 
radars, this is a big deal....)