[Elecraft] An AGC Story

[Dick Green WC1M]

Speaking of criticism, I've done my share of complaining about AGC pumping
and pileup mush in the K3. That said, it's still the best contest receiver
I've ever use, by a wide margin. It's sensitive, quiet and incredibly
selective. In fact, I can get *too* close to adjacent stations without
realizing it! 

In actual contest conditions, the AGC pumping can be reduced to more-or-less
acceptable levels by using the 8-pole 400 Hz filter. It becomes a
non-problem with the 5-pole 200 Hz filter. I suspect it could be reduced
even more by increasing the HAGC threshhold, but evidently not without
changing other components in the chain to avoid increased IMD. 

IMHO, the pileup mush is no worse than any other receiver I've used, and
most of the time it's better (that sounds strange, but you know what I
mean.) Considering that I had quite a few hours with my rate meter near
200/hr in CQ WW CW, I'd have to say the mush isn't reducing my rate a whole
lot. One clue is that the mush effect is most noticable when there's a
pileup of relatively weak signals and there's not a nearby strong station --
i.e., when the HAGC is not activated. If most of the signals are at the same
pitch, it's mush. A bunch of really weak signals like that are simply darned
hard to copy, and I'm beginning to think that no receiver can separate them
unless AGC is turned off, with the attendant problem of ears being blown out
when a strong station drops by. I have found, however, that the situation
improves considerably with diversity reception using two antennas with
opposite polarity. There's enough difference in the ever-changing arrival
angles to be able to distinguish weak signals at the same pitch. No other
receiver can do that for me.

73, Dick WC1M

-----Original Message-----
From: Geoffrey Mackenzie-Kennedy [mailto:gm4esd at btinternet.com] 
Sent: Sunday, December 12, 2010 6:30 AM
To: Bob Henderson
Cc: Elecraft Discussion List
Subject: Re: [Elecraft] An AGC Story

Bob,

Yes I did see your post thanks, but have not yet had the opportunity to 
digest your results - but I will.

When you made these measurements, what settings did you use for the 'audio' 
Hi-cut and Lo-cut?

My thinking was that I would first look at the SA612's output wile injecting

multiple signals, and then do the same at suitable points along the signal 
path as it made its way to the audio output. By doing this it should be 
possible to find the weakest link in the chain (in terms of IMD), and then 
attempt to do something to improve the 'weak link's' performance.

May I take this opportunity to stress that my comments were not intended to 
be any kind of criticism of the K3.

73,
Geoff
GM4ESD

 .
"Bob Henderson" <bob at 5b4agn.net> wrote on Sunday, December 12, 2010 4:11 AM:

<snip>

> I made within DSP b/w IMD measurements on my K3 and comparable
> measurements on my K2. These were included in a post dated 5 December
> but you may have missed it.
>
> The following measurements relate to K3 S/N 4904 on 7MHz with 2.7kHz
> roofing filter and 2.7kHz DSP b/w.  Default slow AGC was selected.
> The table shows input level at the K3 antenna connector for
> each of two carriers spaced 500Hz.  At each input level, the level of
> the strongest IMD product observed at the AF output using an HP8568B is 
> recorded
> alongside.

<snip>