[GreenKeys] Military TUs

[jhhaynes at earthlink.net]

I was going to weigh in on this one, and the answer is that generally
they are not very good for ham use, especially the older ones.

In the beginning we used 850 Hz shift, because the FCC said so, and
because that was what the commercials and military were using.  I'm
convinced the main reason for using such a wide shift was to be able
to cope wth the drifty receivers and transmitters of the period.  As
a case in point, the manual for the AN/URA-17, which is a solid-state
T.U., suggests that it would be used with the RBB, RBC, or the SRR-13
family of receivers, both of which are ancient.  And often they used
things like the BC-312/342 and the Hammarlund Super-Pro.  Later military
stuff was often designed to accomodate a shift from 1 KHz down to 85 Hz
or so, which means a lot of compromise.

Another reason they were not very good is that they didn't have to be.
The military engineers circuits pretty conservatively, so for a ground
point-to-point operation they will use rhombic antennas and diversity
reception and all the power they need for reliable communication.  And
where they are stuck with mobile antennas they de-rate the expected
range of communication quite severely.  So they aren't particularly
interested in a T.U. that will dig a signal out of the noise or cope
with the kind of QRM you get in a RTTY contest.

Some of the military TUs operate at IF rather than AF.  The only reason
I can think of for this is to get BFO drift out of the picture.

Then as Don mentioned some of the military TUs operate on 85 Hz shift.
There's nothing wrong with 85 Hz except that the amateurs settled on
170 Hz back in the 1960s and it has become a de facto standard, wired
into some of the modern transceivers.

Of course the military has access to the best engineering talent there
is, but it has been the hams I think who have worked the hardest on
TU designs for signals that are weak and fading and QRMed and where
diversity reception is not available.  So we see things coming off
government surplus like the Dovetron and the Frederick that are basically
ham designs.  (Dovetron started out as a ham product.  Frederick was
not in the ham business but when Vic Poor was designing their stuff he
was an active participant in the discussions going on among Irv Hoff and
Keith Petersen and others back in the mid 1960s.)

Now of course it's a whole new ball game with computers and digital
signal processing.  Primarily we use the computer as a self-contained
modem and terminal.  I was able to persuade K6STI with his RITTY
program to provide a cleaned-up Baudot output when receiving so it can
be used to drive a printer.  I haven't had any luck with the authors
of other software, but then I haven't tried very hard.  With Linux
there is presently a problem of POSIX non-compliance that makes it
impossible to run a COM port at 45.45 baud; I hear this is going to be
fixed someday.


jhhaynes at earthlink dot net