[GreenKeys] Northern Radio 152/2

[David I. Emery]

On Sun, Jun 21, 2015 at 08:26:08PM -0400, Duncan Brown wrote:

> I think you would need a strong & stable (frequency) signal for 85 Hz 
> shift to work well.

	Certainly stability was needed - thus things like the CV-157 and
the R-1051x family.   The sharp filters in many commonly deployed 85 Hz
shift FSK converters required better than 3-5 Hz end to end frequency
accuracy and stability  before teletype signal distortion from the
effects of tuning error and frequency drift becomes insignificant enough
to ignore.

	Wider shift FSK had also some advantage with more sophisticated
modems in that the mark and space tones tended to fade from multipath
(selective fading) more and more independently of each the greater the
frequency separation.   This meant that wide shift signals (850 Hz and
sometimes beyond) when processed with suitable techniques had some
inherent frequency diversity advantage from uncorrelated fading not
present in very narrow shift signals as the mark tone or space tone of a
wide signal might fade into the noise at any give moment but the other
tone might not allowing some continuing copy.

	Not all HF modems in that era were sophisticated enough to take
advantage of this mode of operation, however... many simply had a
bandpass filter as wide as the shift ahead of a hard limiter... 

	When much of the military and government fixed station HF plant
was converted to "tone packs" of 16 channel VFT 85 Hz on ISB SSB
transmissions in the 50s it was not uncommon to see pairs of channels
(tones) - often 8 channels apart - in frequency diversity mode with the
same bit stream on both of them to get back the frequency diversity
advantage given up when going to narrow shift.   Much of the gear the
Navy used for its VFT fleet broadcasts also had this capability with
circuity in the demods selecting the stronger signal at each moment -
and sometimes both frequency and space diversity being used with 4
signals combined to get one channel of information.

	On the other hand a 170 Hz channel is plenty wide enough for a
45, 50 or 75 baud FSK signal in near MSK mode with shifts of 85 or 60 Hz
or less ... and anything wider lets in more noise and no more signal -
so it is not clear one needs a lot more actual signal energy for good
copy with a 85 Hz shift signal in any way at all.

> I've never heard of 240Hz shift, either.  The AWA Museum has a Collins 
> diversity FSK converter that will operate on 425, 600,  or 850 Hz shift. 
> I had never heard of 600 Hz shift before.

	For whatever reason there was a certain use of 600 Hz shift by
some folks... 425 and 400 were not uncommon either.   There were of
course pressures to use less spectrum than 850 Hz shift signals did
(nominally 1 KHz channels), but the narrower the shift the more critical
drift and tuning accuracy became so there was some inclination to
compromise... especially with older non synthesized HF receivers and
transmitters where maintaining frequency accuracy and stability in the
range of a few Hz was pushing the gear beyond its design limits. 

	There was quite a bit of wireline (microwave and cable) use of
120 Hz spaced VFT which got more TTY channels into a voice channel; that
was usually done with 60 Hz shift - so the 240 Hz spaced channels may
have been based on using a 240 Hz spaced channel for two channel time
division multiplex of two 50 baud TTY signals... making a channel bit
rate around 100 (or 96) baud.


-- 
  Dave Emery N1PRE/AE, die at dieconsulting.com  DIE Consulting, Weston, Mass 02493
"An empty zombie mind with a forlorn barely readable weatherbeaten
'For Rent' sign still vainly flapping outside on the weed encrusted pole - in 
celebration of what could have been, but wasn't and is not to be now either."