[GreenKeys] FRF FSK conv. (was - Two eBay Novelties)

[Duncan Brown]

Dave,

OK, that sounds like an older version. The AFSAV 133D (which became the 
AN/FRA-86 c1964) used miniature tubes, but was still large.

Yes, the DFS system probably would have some noise immunity (being FM) 
over a SSB system that is mostly AM.

As to how the system works, my derscription earlier was wrong. There is 
a good description on Nick's site: 
http://www.navy-radio.com/ct-equip/4a.PDF .  This shows that only one of 
the four tones was sent at a time, but each tone encodes a bit for each 
TTY channel ("canal").  eg. tone Freq 1 = MM, F2=MS, F3=SM, F4=SS.  
Apparently any combination could be used and the FRA-86 could be 
programmed for all of them with the little jumpers.

I think all the tone detectors were active all the time. You would have 
to detect something on one tone filter to decide that it was active, but 
it would be too late to shut the others off.  Today you might be able to 
do something with DSP, but not back in 1960. Due to the capture effect 
of the limiter ahead of the filters, there is very little energy at 
other frequencies, just the tone being sent at that moment. So the other 
filters see very little in-band energy (unless a fade takes out the 
desired tone, which is why the unit is designed to work diversity with 
two receivers).

Duncan


On 09-Feb-14 21:19, David Burns wrote:
> Duncan..
>     The NSA units (I pulled out of a sand pit!) had had their tags 
> removed.  I later learned that those monster units (which looked like 
> a rush-job from an engineering perspective using octal-based tubes) 
> were later redesigned (using miniature tubes) into a two-unit job that 
> took up about half the rack-space.  I saw those auctioned off at the 
> Navy Yard in D.C. (around 1970)  but never owned one. Was told they 
> were functionally identical to the older big units.
>
>     It would seem that a four-tone system (where each tone frequency 
> indicates the current mark/space state of -both- channels) gains 
> immunity to noise that would not be present in a "two-tones-at-once" 
> system, since you could design the decoder logic to be relatively 
> insensitive to the other three frequencies (for a short period of time 
> following a transition).  I don't know whether that was done in those 
> units.  I did have a circuit diagram (acquired from "a source"), but 
> it was wildly complicated for me to follow as a high-school person.  I 
> got the drift, but not the details.
>
>     Any thoughts?
>
> -Dave in Boston
>
>


---
This email is free from viruses and malware because avast! Antivirus protection is active.
http://www.avast.com