[GreenKeys] on-off radio teletype (not FSK)

[Jim Haynes]

Some of this has been touched on already.  Waaay back in the day there was
on-off keyed RTTY used commercially or experimentally.

E. H. Armstrong published in 1928 a suggestion to use FSK for telegraphy,
at the time Morse telegraphy, with the intent of reducing atmospheric
disturbances (Proc. IRE, January 1928, p. 15)  This particular incidence
did not involve using a limiter, and was in advance of his invention or
discovery of wide band FM for high quality radio broadcasting.

In the same year Lawrence Schmitt of Teletype patented a very similar
scheme with the same purpose, Patent #1,705,211  I'm reasonably sure that
neither man knew of the other's work.

I don't know much about it in the 1930s, but apparently FSK was the
established way of transmitting radioteletype before and during WW-II.

In the late 1940s hams around New York City were using AFSK for RTTY on
VHF.  FSK was not allowed to hams on HF except in the 11-meter band (which
was later taken away and turned into CB.)  There was some amateur work
with make and break keying radioteletype in those days.  Then the FCC
was petitioned to allow FSK on HF and granted that privilege in February
of 1953.

You mentioned a detector for 60-600 wpm make-and-break telegraphy, and
that would have been mainly for high speed Morse operation, which was
done commercially and in the military.  There was also use of FSK for
high speed Morse.  And example is the Boehme demodulator; the AWA museum
has one of those.  This stuff was transmitted from punched paper take and
recorded in ink on paper tape.  Then operators at typewriters transcribed
the inked slip into hard copy.  In my childhood we could turn to the
shortwave bands on the family radio receivers and hear tremendously strong
droning sounds that we thought were airplanes.  I was told that this was
high speed Morse.

Then there has been considerable use of single tone on-off keying in
conditons where signal quality is not an issue, such as carrying TTY
signals over phone lines between an operating site and a remote 
transmitter site a few miles away.  And for handling signals within
an operating site.

There was also use of single tone on-off keying in carrier systems
used by the telephone company to transmit several telegraph channels
over one voice circuit.  Later they went to FSK since it was insensitive
to variations in signal level.

For a long time the standard shift for RTTY was 850 Hz.  It's my belief
that this wide shift was chosen to allow considerable tolerance for
frequency drift in transmitters and receivers.  I remember operating
amateur RTTY in the late 1950s using pre-WW-II transmitters and receivers,
and it was necessary almost to keep a hand on the receiver tuning knob
to keep the signal from drifting out of the range of the demodulator.
850 Hz shift was one reason the ARRL was initially very unenthusiastic
about RTTY - noting that it took up almost a KHz of bandwidth to transmit
at 60 wpm, while a CW Morse signal almost as fast was much narrower.  And
in those days there were problems with CW operators jamming RTTY signals
because of anger about the wide bandwidth.  Another reason for 850 Hz 
shift, realized later, is that the two tones can be detected separately
and the detector outputs combined, giving the effect of frequency 
diversity.  It was unlikely that both frequencies would fade at the
same time.

As the frequency stability of amateur equipment improved, with the 
widespread adoption of SSB for voice, amateurs realized that narrower
shifts for RTTY were perfectly feasible.  After some experimenting the
ham community settled on 170 Hz as the standard narrow shift, and thus
it has been ever since.  (Except for AMTOR and its followers which chose
200 Hz because that was the shift used by standard telephone modems.)

A standard for carrier systems used over voice circuits has been 85 Hz
shift.  This doesn't seem to have ever seen much use in ham radio, though
some of the MARS systems required it at one time.