[GreenKeys] Crypto Sync Pulse on M28 machines

[Jim Haynes]

On Wed, 19 Feb 2020, Nick England wrote:
> 
> I wonder why it seems like the Navy made a big deal out of KW-7 syncing but
> the Army and Air Force evidently didn't. Was it a low-level signaling thing?
> Something related to long distance HF radioteletype?

The 3 racks of equipment I mentioned earlier came from an Air Foce surplus
sale and are covered in an Air Force technical order.

I think I mentioned earlier that there was a thing called a "printing
telegraph signal normalizer" covered in IRE Transactions on Communication
Systems 4:3, 1956, p. 18. and in Western Union Technical Review 11:1
January 1957 p. 10.  This was for some reason supposedly required on
receiving from time division multiplex (AN/FGC-5, AN/UGC-1, AN/UGC-3
because the multiplex delivered slightly shorter than normal stop pulses
and occasionally put in a one-character delay.  This was because the
mux had to run slightly faster than the incoming TTY signal to be sure
it stayed ahead if the incoming signal was slightly fast.  And for
some unknown to me reason the crypto gear didn't like the slightly fast
signal and the delays.