On Thu, 03 Dec 2009 10:29:53 -0800, Richard Clark
wrote:

This small, resonant plate load, is quite specifically designed for RF
with low in resistive loss - and yet it is miserable as a propagator
of that same RF. The physical size compared to the wavelength size
dominates that efficiency with a fourth power law.

To extend this to Art's misinterpretation of Faraday Shields:

In the old days, breadboard design was exactly that - your rig was
built on (hammered to) a breadboard. It was open wiring with open
components. It radiated well with an antenna, and poorly without one.

However, as poorly as it radiated without an antenna, if you had a
separate receiver, you would hear yourself. This was sometimes useful
and gave us what is called "side tone."

The monitor was born.

Of course, with antennas connected, the receiver was bound to get more
than enough of that anyway and if the two were closely spaced,
feedback could drive all circuits into saturation. Not a good thing.
The Faraday shield for the transmitter was born.

It, as many can witness from simple observation, was composed of a
fine grid mesh of wire either tied to ground, or to a heavily AC/RF
filtered DC potential. As with all Faraday shields that came before
it (indeed since Faraday invented it), it completely encapsulated the
RF power source. The screen or mesh was simply a contrivance to allow
cool air to move in and hot air to move out. Modern implementations
use finned constructions and heat wicks - but this is topic drift.

With this added to the breadboard, other circuits also came to be
shielded, and generally so with the appearance of sheet metal chassis
with suitably wavelength small openings for access and heat transfer.
As the breadboard went into this RF impenetrable shell for both
receivers and transmitters (and with even more care for transceivers),
there arose a problem: What about the wires that go in and out?

Yes indeed. If those wires were not, in themselves, decoupled; then
they became radiators. The lesson to be learned was that those wires
had to be held at the same potential as the Faraday shield. This
could be accomplished by a simple connection, but with more than one
wire this leads to dead shorts between wires. Not a good thing.

The solution was to use AC/RF shorts (capacitors) to the shield from
the wire and the wire could only penetrate the shield through a very
small (in proportion to wavelength) opening. This was not always a
good thing.

A capacitor could be good, but it exhibits a roll-off of only 6dB per
octave, or 10dB per decade isolation. If your line going in and out
was a DC control line, and your principle frequency was 1MHz (talking
about the old days now); then you had 6 decades of separation between
1Hz and 1MHz - pretty good. If in the intervening years you pushed
the technology envelope and added voice modulation and that came
through the same wire; then your system shrunk to 3 decades of
separation between 10,000Hz and 1MHz. This might work, sometimes it
didn't.

As the years spun on, more wires penetrated that RF barrier, and they
needed to not only be isolated from the RF, but each other; and often
they contained very small signals that needed suitable signal to noise
ratio (noise being that soup of RF that was stewing inside the
shield).

Inline bypass filters were born.

The lines that penetrate a Faraday shield now appear to be more
multi-stage low pass filters with repeating sections of shunt
capacitors and series inductors. Their common (ground to the old
brass pounder) was the shield which was RF free (as it was decoupled
to a sanctioned earth ground). And lest we forget the principle
penetration of that old time Faraday shield:

The coaxial transmission line was born.

By all appearances, this line satisfies the convention of a small
opening through the Faraday shield. It's diameter is easily very
small in relation to the wavelength of the RF power it reaches into
the shield to tap. In a sense, it extends that hole in the shield to
some very remote area that is far from the operating position, and
then allows a wire(s) to emerge without regard for further shielding:

The antenna is born.

Funny thing, however, is that presumption of the shield of the coax
being inert, un-perturbing, quiescent, invisible, benign - for that
presumption is an illusion, a grand delusion. The line is very long
with respect to wavelength, it is in the field of excitation that has
been drawn out of the soup within the cage, and it is as much an
antenna as the wire that emerged from its end. Many familiar problems
rise from the ashes of this illusion. The exterior of the coaxial
cable appears to the field to be a very long, grounded radiator.
However, at any appreciable length (wavelength raises its familiar
visage with an ironic grin), this exterior surface ceases to be the
familiar DC grounding strap material, and becomes a full-fledge
radiator according to its physical length vs. wavelength relationship.
Not a very good thing, untill:

The transmission line choke is born.

To decouple the OUTSIDE of the coaxial line, the convention that has
been observed (to widespread validation) is to either wind some
sections of the line into Inductive chokes, or to add ferrites which
serve the same purpose. These chokes, to be fully useful to their
purpose, should be found at not only one point along the line, but at
several so as to suppress (wavelength based) couplings along the line,
by the line and by the field.

When the combination of all these methods are employed, then the
Faraday shield does what it has done for these several hundred years
while allowing the migration of RF power to a remote drive point, and
without allowing that RF power to re-intrude into the shield, nor
along the coaxial cable. Thus, the only evidence of RF from inside
the Faraday shield is that which arrives over-the-air from the remote
antenna.

Any other claim is a profanation of Faraday.

73's
Richard Clark, KB7QHC