Re Recommended AGC settings for separation of signals in CW pile-ups

[Bob Henderson-2]

Barry N1EU wrote:

Bob, this topic has come up a few times in the past and is something that
has my full interest during/after each cw contest.  I hate to say it, but I
don't think there's much of a solution via AGC settings.  I typically run
with AGC OFF and minimal RF Gain to try and maximize transfer of RF to AF rx
dynamic range, but it honestly doesn't help much with the pileups.  Perhaps
it's just the nature of K3's A/D/A.  I've used both the K3 and Orion in

ARRL160 this weekend and my sense is that the similarly rx gain optimized
Orion does a bit better in those situations where there are multiple callers
on frequency.

73, Barry N1EU

Hello Barry,

Comments from others indicate a perception of within passband IMD as
the cause of the problem.  In response to this I made some
measurements this morning.

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 following 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 on an HP8568B is recorded
alongside.

-73dBm  Better than -55dBc  (Input S9)
-63dBm  -55dBc
-53dBm  -50dBc
-42dBm  -45dBc
-33dBm  -42dBc
-23dBm  -37dBc  (Input S9+50)

These figures are good and IMHO entirely acceptable.  Some older
transceivers would struggle to provide -35dBc at S9.

I made a similar set of measurements with my old K2 S/N 997

-73dBm  -50dBc
-63dBm  -48dBc
-53dBm  -45dBc
-43dBm  -48dBc
-33dBm  -45dBc
-23dBm  -15dBc

Broadly speaking the K3 is as good or better than the K2, which is
itself good. It is possible later mods to the K2 may have made it even
better.  The most recent mods to my K2 were carried out over 5 years
ago.

These figures appear to confirm within passband IMD is not creating
the mud.  Perhaps AGC artefacts generated in a more dynamic (true to
life) scenario might be the cause.

The addition of Bandwidth Adaptive Stereocode to the K3s DSP armoury
might provide dramatic benefit.  The current AFX facility doesn't seem
all that useful.

The term Bandwidth Adaptive Stereocode likely won't mean much to many.
 You will recall Orion offered Stereocode for CW but its effectiveness
in narrow b/w was very limited due to the code having been optimised
for wider bandwidths.  Listening to CW on Orion in an SSB b/w with
stereocode enabled provided an impressively wide sound stage which
significantly aided signal separation.  Unfortunately the fixed b/w
code resulted in a sound stage which shrank with reducing b/w
similarly reducing its usefulness.  In my time between Orion and K3 I
have been fortunate enough to become involved in G3XJP's PicaSTAR
project.  Peter implemented Bandwidth Adaptive Stereocode in Picastar
DSP.  The results have been dramatic.  Preferred pitch appears centre
stage with lower tones moving progressively off in one direction and
higher in the other.  As the bandwidth narrows so tonal separation
increases across the sound stage.

I'm not aware of any commercially produced Ham transceiver which
offers this capability.  Perhaps Elecraft will be the first.

What say Lyle?

73 Bob, 5B4AGN
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html