Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site redwood.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!ihnp4!qantel!hplabs!hpda!fortune!redwood!rpw3
From: rpw3@redwood.UUCP (Rob Warnock)
Newsgroups: net.ham-radio
Subject: Re: "PL" and tweedles
Message-ID: <204@redwood.UUCP>
Date: Thu, 11-Jul-85 20:25:54 EDT
Article-I.D.: redwood.204
Posted: Thu Jul 11 20:25:54 1985
Date-Received: Wed, 17-Jul-85 05:55:59 EDT
References: <86@biomed.UUCP> <957@sdcsvax.UUCP> <11397@brl-tgr.ARPA>
Organization: [Consultant], Foster City, CA
Lines: 44

+---------------
| > (Note to ham radio newcomers: the ``squelch tail'' is the burst of white
| > noise heard as an FM receiver squelch delays closing for a short period...
|                                        ... Of course, there isn't a
| whole lot you can due about the squelch tail in the user's receiver.
+---------------

Actually, with some fancy coupling between the squelch and AGC circuits,
it should be possible to eliminate most or all of the "squelch tail".
The squelch tail occurs because the AGC gain opens up faster than the
squelch closes. By using a longer time-constant on AGC decay than on rise,
and by tying the time-constant to whether the squelch is open or closed
(you want a long decay if squelch is open), the "tail" will be heard as
QUIETER than the transmission, not louder. Of course, this means that the
squelch cannot be controlled simply by the AGC voltage (as is often the
case in simpler designs), or you'll get "motorboating" oscillations in
the squelch.

A possible algorithm is to use AGC voltage to OPEN the squelch (and switch
the AGC time-constant to "slow"), but then use some ratio of actual input
R.F. to "estimated" input (i.e. the AGC voltage) to trigger the CLOSING
of the squelch.

Careful design is needed in mobile rigs (or anywhere fast fading can occur)
to make this work, and of course impulse noise has to be treated carefully
as well. Note that the "slow" AGC decay (with squelch open) need not be
"slow" by human standards, only "slow" compared to the time-constant of
the squelch-closing detector.

Note that a squelch tail will still exist -- it just won't be so annoying.

p.s. Hmmm... maybe the problem is not the AGC, but the squelch open/close
detector. Mayebe all you need to do is make the squelch OPEN slow (compared
to impulse-noise rejection time) and CLOSE fast (compared to AGC recovery time).
That would get rid of the "tail" entirely.


Rob Warnock
Systems Architecture Consultant

UUCP:	{ihnp4,ucbvax!dual}!fortune!redwood!rpw3
DDD:	(415)572-2607
USPS:	510 Trinidad Lane, Foster City, CA  94404
require