[TowerTalk] Faraday Loops???

Jim Lux jimlux at earthlink.net
Sun Nov 6 10:33:47 EST 2005


At 07:13 AM 11/6/2005, Tom Rauch wrote:
> > If the perimeter of the hole is < 1/2 wavelength (or so),
>then it will
> > shield fairly well. The real challenge for building a
>shielded room would
> > not be propagation through holes, but signals carried on
>wires penetrating
> > the shielding wall.
>
>I don't think "shield fairly well" paints an accurate
>picture of what it could do.
>"Might reduce magnetic flux in a small area around the loop
>axis " would be a better description.


I was assuming that the loops went around the entire room.  Imagine a giant 
bird cage, albeit with widely spaced bars.

And, of course, the shielding effectiveness will potentially be greatly 
compromised by any of a number of construction details (like how the bars 
are attached to each other).


>I can't imagine anyone doing that for lightning protection.

Perhaps not for lightning protection, but to establish an equipotential 
surface around a room, I can see.

This kind of thing is moderately common in HV impulse testing work and 
Tesla coil demos, where you create a Faraday cage to put test equipment or 
an observer inside.  The pulses are "lightning like" in that they have <1 
microsecond rise times with RF spectra well up into the HF region.  Even 
fairly high power discharges directly to the cage bars have almost no 
effect.  I say almost because although I've never actually seen any damage 
or hiccups, even on fairly low level signals (video), I imagine that one 
could set up an experiment to detect the fields.

Naturally, nobody is stupid enough to actually stick anything out between 
the bars!


Which brings up the whole problem of how useful this might be for shielding 
a shack.  When working with these sorts of things, I've always used battery 
powered equipment, so there was no need to bring any wire into the cage 
from outside.



>You'd be better off burying the copper and using it for a
>ground if lightning protection is the goal.
>
>73 Tom



More information about the TowerTalk mailing list