[GreenKeys] Watkins Johnson FSK converter on 'bay...

Dave Emery die at dieconsulting.com
Mon Jan 3 18:21:28 EST 2005


On Sun, Dec 19, 2004 at 07:30:00PM -0500, Roy Morgan wrote:
> At 01:52 AM 12/31/2004 -0500, Dave Emery wrote:
> >        I have nothing to do with the seller, but noticed a 80s SIGINT
> >grade WJ HF FSK converter with a fancy scope display as Item 3863916556.
> >
> >        Dual diversity...
> 
> Dave,
> 
> Mighty fancy but, being ignorant about most W-J, I wonder:
> 
> - What does it accept as inputs?  RF? IF - what frequency?
> - Does it drive RS-232 type output, or 60 ma loop?  What?

	Guys it is a bit late, but this thing takes 600 ohm unbalanced
audio on a BNC - not IF.   Nobody built IF type converters after the
mid to late 60s and this is an early to mid 80s design.

	Notice particularly the Mark and Space tone frequencies set in
hertz ?   No IF converter would have any such provisions at all... and
of course such would be essential for SIGINT/COMINT work on VFT
multiplexes both from microwave and satellite and HF SSB receivers where
there might be several sets of mark and space tones in one 3.2khz audio
channel (ISB from an ISB receiver commonly)  - very very common on HF
through the early 90s or so.

	The Space IF and Mark IF outputs are audio - probably around 16
khz - intended to feed a XY scope for the classic cross tuning display.
They can be also used to feed a counter to determine tuning error
for the tones.

	Internally this WJ unit and the somewhat similar Plantronics
1280A family are audio superhets which upconvert 300-3200 audio (actually
they may go higher and work up to 6 khz or so)  to a fixed IF at around
15 or 16 khz where a variable bandpass bandpass filter (bw depends on
baud rate and perhaps shift) filters the mark tone or the space tone.
(There are two LOs and two upconversions - one for mark and one for
space so the unit at auction is actually 4 separate audio superhets
for two mark tones and two space tones).

	The LOs that do this mixing and upconversion are generated in a
classic PLL divide by N synthesizer setup so they can be tuned in 1 hz
or maybe .5 hz steps.

	This arrangement allows any shift at all from very narrow to
very wide (5 khz or so potentially) and allows use of filter technology 
for the IF working at one frequency allowing variable bandwidth to be
easily and accurately obtained.   The Plantronics 1280As I have the
manuals for (and 4 of the units) use switched capacitor filters for the
IF, I am not certain what this WJ uses.

	All of this stuff is controlled by a microprocessor (or two)
allowing flexible setup of tone frequencies and modes - both from the
front panel and the 488 bus.

	Output is very likely RS-232 levels (or Mil 188) on a BNC.

	And yes, confusing as it is, much NSA WJ intercept gear of the
70s and 80s used BNC connectors for 600 ohm unbalanced 300-3200 hz
audio.   Apparently they found using 50 or 75 ohm patchbays and cabling
more convenient for distribution of audio in intercept bays than using
the balanced 600 ohm audio on twisted pairs much more common in
broadcast and other communications plant.   I suspect the fact the same
cabling and patchbays used at audio could be used for video or IF or
even RF was the driver here...  I have also seen audio matrix switchers
from this era of NSA stuff on eBay... with gazillions of BNC inputs and
outputs.

-- 
   Dave Emery N1PRE,  die at dieconsulting.com  DIE Consulting, Weston, Mass 02493




More information about the GreenKeys mailing list