"Robert Baer" wrote in message
...


SNIP

"Faraday shield" to some degree is a myth.
I have seen radars inside quonset huts track a *bird* flying a few
miles away (thru the metal wall)!

You must have some strange buddies. Who in the world would set up a
radar
within a metal hut? And even if they did, who would think it's a good
idea
to stay inside with it if it were on?

There's nothing mythical about the Faraday shield; it works really
well,
so
long as there are no discontinuities (apertures) and sufficient
thickness
and conductivity. Under real-world conditions, steel works pretty
good,
and
any thickness sufficient to support itself will yield great shielding
effectiveness. So the only real performance variable left is the holes
in
the conductive surface. How many, maximum dimension, proximity of
radiating
source to the shield, etc.

While I would expect a Quonset hut to really mess up the accuracy of a
radar, it likely wouldn't be a good shield, as the floor isn't metal,
I
don't think the ends are metal, and the various skin panels are rather
poorly RF bonded.

Ed
wb6wsn


I do not think your objections concerning the floor or the bonding of
the panels are too relevant.
The ends are metal and not relevant either.
The radar was pointing right at the wall (no windows nearby); any
presumed leakage via remote holes that you assumed might allow the
transmitted signal to leak, but would then not be focused on the
bird(s)
and the path lengths would vary.
But the reflected signal from the bird or birds would be rather weak
and could not possibly be received via the same wild path(s) to a very
directional antenna.

My point is that a Farady shield is a good attenuator, but not
"perfect" as ASSuMEd.
And it sure is not "flat" in attenuation characteristic as a function
of frequency.

Those weren't objections, they were speculations on my part as to how you
boys could have been finessing the generally applicable laws of physics.

But truly, the story stinks. So you and your army buddies are in this
metal
hut, with a fairly high-power radar, and somebody comes up with the
bright
idea to turn the thing on. Apparently no thought about RF personnel
hazards
and no concern about strong reflections cooking your detector. Did you
test
your M16's in a Quonset hut too?

Next point. "The radar was pointing right at the wall..." Now tell me, in
a
semi-circular Quonset hut, how do you point anything "right at the wall"?
Maybe straight up?

Now, a bird doesn't have a very big radar cross section, maybe only about
0.01 square meters, so the return loss is really big. And to resolve a
single bird, I'm gonna guess that you had an X or K band radar. So let's
run
some numbers. Let's say you had a 100kW radar, with a 30 dBi antenna of 1
square meter aperture. At 1500 meters, your detector power would be about
1
picowatt, or -60 dBm. Well hey, that's pretty decent, I'll bet you could
see
a bird at one mile.

But that's assuming no loss at all due to the metal hut skin. Let's see
what
happens if we say that the metal hut walls give us only 40 dB of
shielding
(by absorption or reflection, it doesn't matter). That bites 80 dB out of
your path budget, putting your detector signal down to -140 dBm. I think
your story just ran out of luck.

Now you can argue about the 40 dB shielding effectiveness of the metal
wall,
but I'll say that I was being very generous about that. At 10 GHz, I know
(How? Easy, I do it everyday. Just 3 days ago, I was keeping some 1.3 GHz
from radiating off of some cables, and it was common old Reynolds Wrap to
the rescue.), I can get 100 dB out of a sheet of aluminum foil. The SE
is
so damn high from the material that the only significant factor is when
the
energy finds a path around the shield.

Don't try to argue that a Faraday cage leaks; you appear to be trying to
build a general case based on your experience of always having observed
leaky structures. Sure, I know that shielding varies with lots of
factors,
conductivity, permeability, thickness, frequency, angle of incidence,
distance from source, and then there's the problem of apertures. But your
hut, with plain old galvanized steel about 1/16" thick, would make a
great
shielded enclosure, as long as the joints didn't leak.

BTW, I don't like using the term "Faraday cage". Despite all due respect
to
Mr. Faraday, calling it a shielded enclosure is a clearer description.

Ed
wb6wsn

I was not alluding to leakage; a more accurate term would be
re-radiation.
Take an ordinary transformer; it radiates a magnetic field, despite
the fact that the core is a closed loop.
In fact, one could get nasty and say the same thing about a toroid
transformer.
Now add a shorted copper turn around the outside of the ordinary
transformer's core (i have seen this on many TV power transformers and
others as well).
What happens? That magnetic field induces a current in that shorted
turn, making an opposing magnetic field - thereby reducing the net
radiated magnetic field greatly - but not to zero.
Now, instead of using that closely wrapped copper shoted turn, put
that transformer inside that shielded room you love.


My shielded enclosure only asks that I respect it; I don't think it would
provide better SE even if I told it that I loved it.


Results: great reduction, but not to zero.


Nothing ever goes to zero; I'll usually settle for "great" reductions.


Increase the frequency to something one might consider RF.
Now one has an RF transmitter inside that shielded room, inducing
currents in the wall(s).
Those currents create opposing fields, and greatly attenuate the
signal outside the walls.


You're getting a little fuzzy here. The propagating wave induces surface
currents on the metal barrier. The currents "sink" into the metal,
decreasing to about 37% (1/e) in what's called a "skin depth". At 10 GHz, a
"skin depth" in steel is really thin. After even 10 skin depths, the current
is down to only about 1/100,000 of what was on the surface. And there's a
whole lot of more skin depths to go before the current is presented to the
far surface of the steel barrier. And only then does the surface current on
the far side of the barrier get to launch a propagating wave.

Note that the "opposing fields" you mentioned are on the INSIDE, the near
surface, of the barrier. The reflected field is 180 degrees out of phase
with the incident field, so, real close to the metal surface, the E-field
nulls. OTOH, that reflected wave now goes marching back at you, creating
lots of fun with out-of-phase energy pumped back into the original radiating
element. Everybody sees bad, bad VSWR. And, since you guys were inside a
metal hut, there's even more fun in store for you. Not all that energy goes
back into the originating antenna. A lot of it just keeps bouncing around
inside the hut, creating 3-D variations in power density. Think of yourself
as a potato, slowly cooking.

So, to keep this straight, the current that survives Ohmic losses to make it
to the far side of the barrier doesn't "greatly attenuate the signal outside
the walls". It actually creates the signal (the propagating wave) on the far
side of the barrier.

We can talk about aperture leakage and re-radiation from barrier impedance
discontinuities some other time.


But they are not zero.
BW, radar is usually pulsed, and in the megawatt to multi-megawatt
region for the pulse.


Multi-megawatt? Hmm, 10 MW? OK, and maybe a duty cycle of 0.01%? Isn't that
1 kW average? I own a 250 kW X-band radar that will do up to 0.1% duty
cycle. I sure wouldn't sit in a metal box with that thing running! I
wouldn't even want to be in the boresight of the antenna within a few
hundred feet. I was trying to be charitable in assuming that nobody would be
so dumb as to fire up a multi-megawatt radar INSIDE a metal hut. Looks like
you guys proved me wrong!

Ed
wb6wsn


Ed
wb6wsn