[TowerTalk] Narrow Band Filters

[Jim Lux]

----- Original Message -----
From: "Richard (Rick) Karlquist (N6RK)" <richard at karlquist.com>
To: "Jim Brown" <jim at audiosystemsgroup.com>; "Tower Talk List"
<towertalk at contesting.com>
Sent: Sunday, July 04, 2004 10:36 PM
Subject: RE: [TowerTalk] Narrow Band Filters


> The key problem is dirty transmitters, as opposed to receiver
> overload.  You can afford an extra 6 or 10 dB of receiver front end loss
> due to filtering to fix receiver overload.  On the transmit side,
> any significant loss produces a lot of heat, plus you need extra
> transmitter power to make up for the loss.  Before resorting to
> this, it is better to set up separate receive antennas adjusted
> to null the transmit antenna with directivity and/or cross polarization.
> Typically, the best unloaded Q is around 1000, which means that
> you lose 1 dB per resonator if the separation is 10% (rough
> rule of thumb).  A tunable notch at the receive frequency in the
> transmitter feed line gives best rejection vs loss.  The problem
> is that station A needs to control station B's notch by remote
> control, and vice versa.

Good point.. Well... FD operations are getting more networked, so it's not
inconceivable. But you're not going to see with clunky DOS laptops any time
soon.

I can see this, though...

Say you have something like a computer controlled autotuner (aka controlled
pi or t network) set up as a phase shifter.  You run a pass before starting
up to precalculate the L and C settings to null for all frequencies, so,
when you QSY the Tx, the nuller tracks right along.   It's not much more
complex than the logic to store the settings for autotuning.  You DO need a
couple extra Ls and Cs, though, to set phase and amplitude independently, as
well as match the R and X of the antenna.  It could be done manually, but it
would be a royal pain.

You could do the same to trap/notch out the Rx frequency in the Tx side (to
get rid of icky phase/far out sideband noise from Tx amplifiers).

I would think that for the typical FD setup (especially if you don't have a
rotator on the beams) the coupling between antennas stays relatively
constant. (barring the usual FD stuff like wires falling down, etc.)

Lots of complexity for probably not much gain (other than tech gee whiz).
Realistically, you're looking at thousands of dollars in hardware
(especially if you're running QRO).  Consider that a single computer
controlled LC to handle hundreds of watts runs in the kilobuck range, and
you're probably going to need 4 of them (two radios, 1 Tx trap, 1 Rx trap
for each).  For FD, you've got 1000 ft of separation potentially available,
so, investing in coax might be a better deal.

On the other hand, if you're building a gee whiz active phased array, that's
just another design requirement to throw into the mix.
>
> 40 years ago, using vacuum tube equipment, and no filters, we
> somehow managed to operate on phone and cw in the same band
> at the same time on field day with antennas 100 feet apart.

Maybe you were lucky, and it just happened to be a good nearfield null...

I seem to recall that some boatanchor mil surplus receiver had a capability
for an auxiliary antenna to cancel adjacent transmitters. A tunable LC of
some sort.

And, of course, this is standard stuff in the adaptive canceller business.