Karl, I think you misunderstand me.

My understanding of AGC [my college days were almost 50 years ago] is briefly
described as follows.

An AGC [Automatic Gain Control] curve is a plot of the negative feedback to
control [reduce] the gain of the rf amplifier and/or first i.f. stage necessary
to maintain a constant audio level. Typical AGC dynamic ranges may easily exceed
90 dB of attenuation or more. In a modern receiver the MDS is typically -134
dBm. An average audio output may be +16 dBm [earphone]. With an input variation
from -134 dBm [S0] to S9+40 [typically -34 dBm] the AGC dynamic range has to be
100 dB for 'constant audio output'.

In FM radios the constant audio level is typically determined by the saturation
level of the discriminator/limiter. Any further increase in the input signal
cannot produce more audio/output because the discriminator/limiter have
saturated [assuming a TX/RX matched deviation system]. In amateur radio
equipment a relatively low level rf signal will fully saturate the limiter
circuit [10 to 15 uV]. This will indicate S9++ on a typical amateur receiver.
[Note: 10 to 15 uV will produce an average S meter reading of S7 +/- in amateur
SSB/AM/CW modes.]

I understand from your description that the measure of i.f. voltage amplitude
is simply a measure of the open circuit gain and associated normal non
linearities of the rf/i.f string before limiter saturation. Is this correct? Is
there a negative feedback system to reduce the gain of the rf stages in your
description?

Are you using the term in a manner that I am not familiar with? I am trying to
understand exactly what you are saying.

/s/ DD, W1MCE

Karl wrote:

Dave,

I guess I didn't explain myslef. Acually all FM receivers can and may
use the the IF to generat a DC
voltage depending on the strength of the signal coming down the IF.
This voltage can be used to drive
an "S" meter or other measuring device. If you calibrate this voltage
by plotting on a graph for the Y axis
with a sig gen in the into for the X axis you can plot an AGC curve.
Real handy for determing RSLs.
We do it in microwave work all the time.

Regards,
Karl