Mike Kaliski wrote:

Haines,

The versions I used years ago relied purely on capacitive coupling. The
braid was broken and only the centre lead of the coax was connected to
the plates. The earth return was via the automobile chassis for vehicle
use, or via earth connections for domestic use. The metal frame of the
window might work if you have any means of getting a good connection,
otherwise you will need some form of reliable (low impedence) earth
connection inside and outside the building for the system to work properly.

The coupling through the glass should be purely capacitive although some
inductance effects will inevitably be introduced due to the size of the
plates.

You'll have a considerable amount of inductance for another reason if
you follow this approach.

Consider what happens when you separate the two conductors of the coax
as Mike suggests. One conductor continues more-or-less directly through
the glass. The other conductor becomes the "ground" path from wherever
the current leaves the inside of the coax shield on one side of the
glass to where it again enters on the other side. These are the two
conductors of a transmission line which you've inserted in series with
your normal line. If you'll do a little rough calculating, you'll find
that this line has a much higher impedance than 50 ohms. So, if
electrically short, its effect will be that of a series inductance. If
it's not so short, the impedance transformation will be a lot more
interesting. Another thing that happens is that the length of the
"ground" conductor is greater than the length of the center conductor.
This will probably give rise to common mode currents, which will involve
the outside of the coax in the active feedline.

As an alternative you might consider making up two matched tuned
circuits (coil and capacitor) set for the 10 metre band and try
inductively coupling the coils through the glass. To my mind that would
be smaller and easier to set up without the need to worry about earth
feeds. A kind of through glass balun if you like. Obviously you will
need to keep a reasonable distance away from the metal window frame, but
a couple of turns of wire and a couple of capacitors may be all you need.

That might have a better chance of producing a smaller disturbance to
the whole feed system. If you use direct capacitive coupling through the
glass, however, you should remember that you have two conductors to deal
with -- calling one "ground" doesn't give it magic properties. I'd use
two coupling plates, one for each conductor, as close together as
possible. A bit of intentional shunt capacitance would help compensate
for the inevitable series inductance which would be created by even that
setup.

Roy Lewallen, W7EL