Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site petrus.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!bellcore!petrus!karn
From: karn@petrus.UUCP (Phil R. Karn)
Newsgroups: net.ham-radio
Subject: Re: "PL" and tweedles
Message-ID: <401@petrus.UUCP>
Date: Mon, 15-Jul-85 14:16:14 EDT
Article-I.D.: petrus.401
Posted: Mon Jul 15 14:16:14 1985
Date-Received: Thu, 18-Jul-85 03:33:30 EDT
References: <86@biomed.UUCP> <957@sdcsvax.UUCP> <11397@brl-tgr.ARPA> <204@redwood.UUCP>
Organization: Bell Communications Research, Inc
Lines: 21

> 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.

You're forgetting that FM is a nonlinear modulation method. An ideal
discriminator reads signal PHASE and is completely insensitive to RF level.
Moreover, there is limiting ahead of the discriminator, so the "gain" in an
FM receiver up to this point depends instantaneously on the signal level.
When the input signal to an "ideal" FM receiver goes away, the
pre-discriminator gain instantaneously goes to infinity. (In a real
receiver, it goes just high enough for front-end thermal noise to saturate
the limiters). Thus, you have squelch noise that starts immediately
upon loss of signal.

The AGC is there just as an engineering convenience to keep the
front-end intermod products down when receiving a very strong signal.
It plays no part in keeping the audio output level constant, as it does in
linear modulation reception such as SSB.

Phil