Trevor,

The Gunnplexer also needs an FM receiver at the chosen IF frequency.
Unless path loss is too much of a problem, wideband FM with the IF set
on a clear spot in the FM BC band works well, and it's fairly easy to do
high fidelity full duplex voice communications.

But Gunnplexers drift around like crazy with temperature and other factors,
so it is very desirable to include a frequency tracking loop that tunes one
of the Gunnplexers (not the receiver!) so as to maintain a constant IF
frequency at a clear spot in the FM BC band.

It's very nice if you can scare up an old-fashiond FM BC receiver that
has AFC, to provide access to a DC-coupled signal from the FM
discriminator to provide the error signal for the AFC system.

You may actually find that you don't need a mic amplifier, rather that
you'll want to use a mic attenuator (depending on the mic).

I once did a bunch of this kind of stuff circa 1980, including a link
between two cars traveling in "very loose formation" from Ann Arbor
to Dayton.  There an additional complication was that there were a
lot of complete signal dropouts, and I had to make the AFC loop
switch into "hold" mode during such times, else the AFC would
usually fail to re-acquire lock when coming out of the signal drop.

All this was done entirely with analog technology, with the "track
and hold" AFC being hastily cobbled together in the last few days
before Hamvention start after a highway test revealed the need for
it.  Part of this was also a "quieting detector" so that the loop would
know when to go into hold mode.  The FM BC receiver was a kluge
built out of parts bought from Heath Co for their (then) top of the
line Hi-Fi FM tuner, originally built to serve as a "laboratory" receiver
of sorts.

Sadly I no longer have that receiver, else I'd offer it for your use.  But
when I was packing for my move from Puerto Rico, I looked at that
pretty carefully and concluded that the poor thing had been eaten
alive by mold and fungus, so I tossed it out instead of moving it.

No matter what you've got, if you want the link to keep working
without fairly constant attention, you will need to position the
antennas at both ends so that people aren't moving around near
the antennas.  Also, doing so will more or less automatically
assure RF exposure safety.

Dana

On Thu, Mar 31, 2022 at 6:53 PM Trevor Clarke via mvus-list <
mvus-list@lists.febo.com> wrote:

Trevor,

Assuming that you're talking about an IF at 2m and the
usual 2m FM deviations, then you'll be using only about
1/10 the BW, which will indeed help sensitivity.  But in so
doing, frequency maintenance will be at least 10X the
headache!  For a short range like across the hamvention
site, you should have no real problem with SNR even with
the wider BW.  You'll probably want to point the two
Gunnplexers at some elevated scattering object, perhaps
one of the lighting towers around the oval track.

A single Gunnplexer has no set IF frequency- that is
defined entirely by the difference in RF frequencies
to which the two Gunnplexers are tuned.
In my work in Mich I took the trouble to put a tuned
matching network in the path between the Gunnplexer's
detector diode and the receiver, in the interest of highest
practical performance.  These two networks were
dirt simple, each comprising a small air-core inductor
and a pair of mica compression trimmer capacitors.
Tuning these is a snap if you have a directional coupler
and either a SA or an o'scope with sufficient vertical BW
to allow viewing the IF frequency of interest.  Note
that it is necessary to have the Gunnplexer running
into the intended antenna load at the time you tune the
matching network, else the measurement will be
skewed.

Re Kent's comments on IF frequency choice:  I was
lucky in that it was not difficult to find a good hole in
the FM BC band, even at the top of a 200 ft bldg near
campus in Ann Arbor.  However, that was 40(ish) years
ago, and things are surely different now.  I was also
going for maximum RF simplicity and didn't want to
muck around with the IF interface stuff in the tuner I
was using.  Kent is undoubtedly correct about the
present-day situation.

Dana

On Fri, Apr 1, 2022 at 8:29 AM Trevor Clarke via mvus-list <
mvus-list@lists.febo.com> wrote:

repurpose

Richardson's

do

receiver

to

and

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

Luckily these days it's much easier to get a working receiver in another
frequency range. An RTL-SDR and a tablet or phone is an option but there
are numerous FM receiver chips around such as the Si4702..you just need
watch XO, a couple of passives, a basic audio amp, and a microcontroller
that can handle SPI. At about $4, it's an afternoon to build a wideband FM
receiver that'll do 70-108MHz or so. They're designed for broadcast band
but have worldwide range so that 70-80MHz region might make the difference
when it comes to finding an open frequence.

Trevor R.H. Clarke
Computer Science House
Rochester Institute of Technology
retrev@csh.rit.edu
http://www.csh.rit.edu/~retrev/

On Fri, Apr 1, 2022 at 10:20 AM Dana Whitlow via mvus-list <
mvus-list@lists.febo.com> wrote:

broadcast

newer

AFC

set

tunes

IF

that

that

would

of

close

point

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

Trevor,

The Gunnplexer also needs an FM receiver at the chosen IF frequency.
Unless path loss is too much of a problem, wideband FM with the IF set
on a clear spot in the FM BC band works well, and it's fairly easy to do
high fidelity full duplex voice communications.

But Gunnplexers drift around like crazy with temperature and other factors,
so it is very desirable to include a frequency tracking loop that tunes one
of the Gunnplexers (not the receiver!) so as to maintain a constant IF
frequency at a clear spot in the FM BC band.

It's very nice if you can scare up an old-fashiond FM BC receiver that
has AFC, to provide access to a DC-coupled signal from the FM
discriminator to provide the error signal for the AFC system.

You may actually find that you don't need a mic amplifier, rather that
you'll want to use a mic attenuator (depending on the mic).

I once did a bunch of this kind of stuff circa 1980, including a link
between two cars traveling in "very loose formation" from Ann Arbor
to Dayton.  There an additional complication was that there were a
lot of complete signal dropouts, and I had to make the AFC loop
switch into "hold" mode during such times, else the AFC would
usually fail to re-acquire lock when coming out of the signal drop.

All this was done entirely with analog technology, with the "track
and hold" AFC being hastily cobbled together in the last few days
before Hamvention start after a highway test revealed the need for
it.  Part of this was also a "quieting detector" so that the loop would
know when to go into hold mode.  The FM BC receiver was a kluge
built out of parts bought from Heath Co for their (then) top of the
line Hi-Fi FM tuner, originally built to serve as a "laboratory" receiver
of sorts.

Sadly I no longer have that receiver, else I'd offer it for your use.  But
when I was packing for my move from Puerto Rico, I looked at that
pretty carefully and concluded that the poor thing had been eaten
alive by mold and fungus, so I tossed it out instead of moving it.

No matter what you've got, if you want the link to keep working
without fairly constant attention, you will need to position the
antennas at both ends so that people aren't moving around near
the antennas.  Also, doing so will more or less automatically
assure RF exposure safety.

Dana

On Thu, Mar 31, 2022 at 6:53 PM Trevor Clarke via mvus-list <
mvus-list@lists.febo.com> wrote:

I also did some mobile to mobile Gunnplexer QSO's with W5ETG.
Doppler is also an AM effect as you are moving though a diffraction
pattern of multiple paths in an urban area.
With wideband FM, the FM limiters took out this AM effect.
BUT, just as we would go over a hill or something, just as the signal
level was about to drop out, you hear that Doppler tone.  Not enough RF
for the limiters to work.
Now, on SSB, you hear every reflection.  Some reflections have a positive
frequency shift, some negative.It was fun working W5DBY on 3 cm SSB while
driving though downtown Fort Worth.  Sounds like a 20 meter pile up, I
would hear W5DBY's signal on a dozen different frequencies spread out.
Kent WA5VJB

 On Friday, April 1, 2022, 09:36:02 AM CDT, Dana Whitlow via mvus-list <

mvus-list@lists.febo.com> wrote:

I forgot to mention one of the more amusing aspects of the Gunnplexer
link between two cars on the way to Hamvention.  Quite frequently
some car would pass the following car in the "formation", then suddenly
slow down.  This was in the days when automotive radar detectors were
basically crystal sets with no real selectivity.  It did not take very long
to figure out what was going on  :-)

IIRC, the two-car link was usually pretty solid with the two cars up to
a mile or so apart, maybe a bit more.  This was using the 10 mW
(10 GHz) Gunnplexers with the supplied horn antennas.  We also
experimented with the two cars' positions interchanged, so that
signal coupling between cars depended largely upon double scattering
by "stuff" along the highway as well as by nearby cars.  Under these
conditions the useful range was appreciably shortened, to perhaps
a quarter mile or so, depending.

I also remember that one of the problems we had was with severe
organ-pipe resonances between the Gunplexers and their respective
antennas due to the cars' motion through the air.  Apparently the
periodic air density change in the waveguide was enough to pull
the Gunnplexers' around in frequency.  This was easily solved by
the addition of a couple slugs of polystyrene foam at strategic
locations.

Those were fun days.  I wish I still had the energy to do that kind
of stuff.

Dana

On Fri, Apr 1, 2022 at 8:21 AM Dana Whitlow k8yumdoober@gmail.com wrote:

factors,

one

But

a

someone

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

Thanks for the info Dana! I need to figure out what IF the devices are
currently tuned to but if it's in the proper range I was going to repurpose
my baofeng HT as a receiver, should do a bit better than an FM broadcast
receiver as far as sensitivity. I've got a scanned copy of Bob Richardson's
Gunnplexer Cookbook. While it's a few decades old, the concepts are the
same, might just need to adjust some of the sample circuits to use newer
parts. It discusses some frequency stabilization techniques including AFC
and ovens.

Trevor R.H. Clarke
Computer Science House
Rochester Institute of Technology
retrev@csh.rit.edu
http://www.csh.rit.edu/~retrev/

On Fri, Apr 1, 2022 at 9:21 AM Dana Whitlow via mvus-list <
mvus-list@lists.febo.com> wrote:

factors,

one

But

here a

someone

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

But Trevor, is your $4 SDR going to provide either an analog
DC tuning error voltage output or a SW AFC loop?  Can you
do either of those in an afternoon, including debugging and
fixing?

And how about an RF preselector ahead of the first mixer
(or sampler, as the case may be)?
Considering the environment you'll be in (a metro area whose
spectrum is loaded with strong RF signals all over the place,
complicated by attendees' keying up HT's within a short
distance of your setup at all manner of frequencies, you're
very likely to need a front end a lot stronger than a wide-open
mixer or sampler will manage.

Good luck!

Dana

On Fri, Apr 1, 2022 at 9:37 AM Trevor Clarke via mvus-list <
mvus-list@lists.febo.com> wrote:

FM

difference

are

the

frequency.

to

link

use.

couple

discussion

site

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

  I also did some mobile to mobile Gunnplexer QSO's with W5ETG.
Doppler is also an AM effect as you are moving though a diffraction
pattern of multiple paths in an urban area.
With wideband FM, the FM limiters took out this AM effect.
BUT, just as we would go over a hill or something, just as the signal
level was about to drop out, you hear that Doppler tone.  Not enough RF
for the limiters to work.
Now, on SSB, you hear every reflection.  Some reflections have a positive
frequency shift, some negative.It was fun working W5DBY on 3 cm SSB while
driving though downtown Fort Worth.  Sounds like a 20 meter pile up, I
would hear W5DBY's signal on a dozen different frequencies spread out.
  Kent WA5VJB

    On Friday, April 1, 2022, 09:36:02 AM CDT, Dana Whitlow via mvus-list <
mvus-list@lists.febo.com> wrote:

  I forgot to mention one of the more amusing aspects of the Gunnplexer
link between two cars on the way to Hamvention.  Quite frequently
some car would pass the following car in the "formation", then suddenly
slow down.  This was in the days when automotive radar detectors were
basically crystal sets with no real selectivity.  It did not take very long
to figure out what was going on  :-)

IIRC, the two-car link was usually pretty solid with the two cars up to
a mile or so apart, maybe a bit more.  This was using the 10 mW
(10 GHz) Gunnplexers with the supplied horn antennas.  We also
experimented with the two cars' positions interchanged, so that
signal coupling between cars depended largely upon double scattering
by "stuff" along the highway as well as by nearby cars.  Under these
conditions the useful range was appreciably shortened, to perhaps
a quarter mile or so, depending.

I also remember that one of the problems we had was with severe
organ-pipe resonances between the Gunplexers and their respective
antennas due to the cars' motion through the air.  Apparently the
periodic air density change in the waveguide was enough to pull
the Gunnplexers' around in frequency.  This was easily solved by
the addition of a couple slugs of polystyrene foam at strategic
locations.

Those were fun days.  I wish I still had the energy to do that kind
of stuff.

Dana

On Fri, Apr 1, 2022 at 8:21 AM Dana Whitlow k8yumdoober@gmail.com wrote:

factors,

one

But

a

someone

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com

mvus-list mailing list -- mvus-list@lists.febo.com
To unsubscribe send an email to mvus-list-leave@lists.febo.com