[R-390] AGC/MGC

[David Wise]

> From: fraserb at quasc.com [mailto:fraserb at quasc.com]
>
> RF gain at 10, AGC at FAST, on a local AM station, audio is 
> absolutely excellent.
> 
> I switch to MGC and audio becomes very badly distorted.  If I 
> back off RF gain to about 5 or 6, audio gets nice again.
> 
> Normal?

<Laugh> I should hope so!  Note the reading on the
Carrier meter when you're in AGC mode.  It should
be up the dial a ways, indicating that the Automatic
Gain Control system is lowering the radio's gain.
When you switch to Manual Gain Control, this is
no longer done for you, and that strong local signal
overloads the IF amplifier with attendant distortion,
unless you back if off by hand.  By the way, the
Carrier meter circuit is designed to go upscale on
overload as well as on AGC action, so in your scenario
it should be off 0 in both modes.

Related topic: I'm an SWL and most of my radio time
is spent listening to international broadcasts.  I
have found that the SLOW position gives the most
uniform and listenable audio across fades, and I
have gradually become convinced that the receiver's
slow-attack AGC design contributes to this.  This
strikes me as counterintuitive, but I believe what
I hear.  Can anyone who's applied the Lankford or
any other fast-attack AGC mod compare notes?
Remember, I'm talking about AM voice/music, not SSB or CW.
Does fast attack have a downside?  Why?  Discuss.

On the other hand, FAST is better when I'm cruising
looking for signals; then when I find a good one I
go to SLOW.  Because of this habit, I repeatedly
experience the blast when entering SLOW and the
Moment Of Silence when exiting.  If you have replaced
C551, it can be avoided to some extent by turning the
knob as fast as possible, minimizing the time spent in
MEDIUM, but this kind of clumsy workaround annoys me
no end.  Now and then I wonder if it would be possible
to eliminate or reduce it without too much hacking.
A quick glance at the Y2K suggests an interesting
possibility, to wit:

Disconnect S107 (AGC) pin 9 (MEDIUM) from ground, and
instead connect a new capacitor between pin 9 and pin 8 (SLOW),
sized to make a series combination with C551 that
gives the same time constant operating in the V608A
feedback loop, as was afforded with C551 in its
original connection to ground and not using V608A.
I haven't done the math, but it's probably between
.1uF and .5uF .  Also, connect a 22 meg resistor
between pins 7 (rotor) and 8 (SLOW) to charge C551
to the average V608A plate voltage.  The IF deck
does not need to be modified, only the front panel.

This would be very easy to test out with jumper leads
clipped onto the switch (using the FAST position),
no soldering or cutting.

73,
Dave Wise (SWL in Portland OR)