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

[Arie Kleingeld PA3A-2]

The K3 has its limits :-)

Thanks for the explanation Don,

73, Arie PA3A

>
> If the ADC is overloaded, the digital output will be garbage, and
> there will be nothing to resemble any form of a signal out of the
> receiver.  In a contest like CQWW where there are lots of extremely
> strong signals, you would work fewer stations if the ADC were allowed
> to overload.  Lots of lost points while waiting for the digital bits
> to settle down and recover.
>
> 73,
> Don W3FPR
>
>

______________________________________________________________
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

[Guy, K2AV]

I believe the parameters were that people hearing a diff at 3 dB were
the first statistically significant group, and half/most at 6 dB.  And
this was a defined task without distractions.  Would guess that same
on a noisy 160m is more difficult, yes.

73, Guy.

On Sun, Dec 5, 2010 at 1:54 PM, Barry N1EU <[hidden email]> wrote:

>
>
> Guy, K2AV wrote:
>>
>> Or stated differently, your ear probably needs something an S-unit
>> difference to discern two signals exactly on frequency with no other
>> separation information in the signal blend.
>>
>
> My experience, which seems to be shared by several others, is that even
> signals of greater than an S-unit difference in strength cannot be
> discerned.  Weak and strong signals are all blended indiscernibly together.
>
> Barry N1EU
>
> (bottom posted to first show context of response)
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/Recommended-AGC-settings-for-separation-of-signals-in-CW-pile-ups-tp5804756p5805683.html
> Sent from the [K3] mailing list archive at Nabble.com.
> ______________________________________________________________
> 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
>

______________________________________________________________
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

[P.B. Christensen]

> The K3's AF Limiter is a HARD limiter that has an abrupt onset
> and causes extreme distortion.

Dave,

I don't contest (last and only one was Novice Roundup, 1973) , so I am not
familiar with the "mushiness" that's being discussed here.  However, I do
like to disengage AGC and I also find the extreme distortion to be as
objectionable as the harmful high audio level it's designed to prevent.

It may be beyond the ability of the K3's DSP, but commercial digital audio
limiters, like those found on DSP-based mastering processors, function like
a very fast analog peak limiter.  As level increases, it simply drives the
analog circuit into further limiting.

What I would like to see when the K3's AGC is OFF, is to briefly
re-establish AGC-F when the AF LIM threshold is reached and remains active
just long enough to overcome the objectionable level burst -- then it goes
back to AGC OFF.  If the K3's limiter is already trying to do this now with
brief AGC activation, it's not doing a very good job.  It would also be
helpful to have the display indicate when the AGC-F is coming active in this
mode with something like a "*" or some other special display character that
functions as a one-shot in duration.  IOW, the special character displays
for a minimum of say 0.25 sec so that the action can be seen on very brief
peaks.

Paul, W9AC

______________________________________________________________
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

[k6dgw]

My K3 has an AF Limiter?

I must be doing something wrong or perhaps I'm more deaf than I thought
but I've never heard any of these problems.  I can usually pick out one
of two callers exactly zero beat with each other, I'm not sure exactly
how I do it but more often than not, I do.  Three or more zero beat
starts to get pretty hard on any receiver I've ever used.

I run with AGC on essentially all the time and every now and then I
rotate the RF Gain down and back up just so it doesn't freeze in the
full on position :-)  I'll give the ancient "AF full, ride the RF, AGC
off" a try today on ECN and see how it sounds, Kevin is usually at the
noise level on 20 this time of year.

73,

Fred K6DGW
- Northern California Contest Club
- CU in the 2011 Cal QSO Party 1-2 Oct 2011
- www.cqp.org

On 12/5/2010 11:23 AM, Dave Hachadorian wrote:
> Hi Lyle,
>
> Thanks for entering the discussion. It's great that the company
> guys monitor these reflectors.
>
> Even at its maximum 030 setting, the AF Limiter is problematic.
______________________________________________________________
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

[Dave Hachadorian]

I have read K3NA's article very closely, and tried very hard to
make it work for me.

I should have mentioned that the headset I am using is based on
this hearing protector:
http://www.amazon.com/Peltor-H10A-Optime-Over-Earmuff/dp/B00009LI4K/ref=pd_cp_hi_1


Dave Hachadorian, K6LL
Yuma, AZ





-----Original Message-----
From: Barry N1EU
Sent: Sunday, December 05, 2010 12:45 PM
To: [hidden email]
Subject: Re: [Elecraft] AF Limiter (was Recommended AGC settings
for separation of signalsin CW pile-ups)


It is fundamentally essential to use very high isolation
headphones to
achieve K3NA's dynamic range goal (i.e., to hear the weakest and
strongest
signals with AGC OFF without kicking in the limiter).  It IS
achievable.
It's not a matter of sensitivity.

Barry N1EU



Dave Hachadorian wrote:
>
> There is no way, despite K3NA's excellent attempt
> http://wiki.contesting.com/index.php/Setting_receiver_gain_controls
> to set the K3's RF and AF gain controls to accommodate the
> dynamic range of signals encountered on the bands, without
> riding
> the RF and/or AF gain controls.  Maybe it is possible with
> super-sensitive headphones, but not with any that I have tried.
>

--
View this message in context:
http://elecraft.365791.n2.nabble.com/Recommended-AGC-settings-for-separation-of-signals-in-CW-pile-ups-tp5804756p5805798.html
Sent from the [K3] mailing list archive at Nabble.com.
______________________________________________________________
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 

______________________________________________________________
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

[P.B. Christensen]

> It would also be helpful to have the display indicate when the AGC-F is
> coming active in this
> mode with something like a "*" or some other special display character
> that
> functions as a one-shot in duration..."

If I can back-track -- A special display character isn't even needed.  Just
briefly flashing the "AGC" display text when the AF LIM threshold is reached
and AGC becomes active would suffice...

Paul, W9AC

______________________________________________________________
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

[daleputnam]

In this particular case, and pretty much only this case, which is NOT detectable by the tx control operator, is the ONLY
situation that key clicks may be of some assist to the rx operator, and how much help the clicks are will depend
entirely upon the experience of the operator,  AND the condition or quality of the performance of his ears at that particular moment.
AND THAT is discernable only by the RX control operator.

--... ...-- Dale - WC7S in Wy


 

>
>
> Guy, K2AV wrote:
> >
> > Or stated differently, your ear probably needs something an S-unit
> > difference to discern two signals exactly on frequency with no other
> > separation information in the signal blend.
> >
>
> My experience, which seems to be shared by several others, is that even
> signals of greater than an S-unit difference in strength cannot be
> discerned. Weak and strong signals are all blended indiscernibly together.
>
> Barry N1EU
>
> (bottom posted to first show context of response)
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/Recommended-AGC-settings-for-separation-of-signals-in-CW-pile-ups-tp5804756p5805683.html
> Sent from the [K3] mailing list archive at Nabble.com.
> ______________________________________________________________
> 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

     
______________________________________________________________
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

[David Cutter]

I recall someone on this list predicting this behaviour and said that he
used a tx shift of 50Hz so that he stood out from the crowd to get in first
shout.

David
G3UNA



Guy Olinger said:
It's not NEARLY on frequency.  It's EXACTLY ON FREQUENCY.  I know what
I'm hearing.  I can hear a hertz in the area around my usual 450 Hz
spot.

This problem is getting worse as frequency accuracy is getting to half
a hertz on modern rigs (half a hertz measured on my K3), and people
use packet spots to arrive on frequency.
73, Guy.


______________________________________________________________
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

[KK7P]

Answers interspersed to reduce confusion from top posting.
>
> The K3's AF Limiter is a HARD limiter that has an abrupt onset
> and causes extreme distortion.

Precisely. It's sole purpose is to PROTECT your ears, not help you copy
weak signals in the midst of strong ones.

> ...
> Once the limiter is hit, there is absolutely no hope of pulling a
> signal out of the pileup.  One absolutely must reduce AF or RF
> gain to pull anything out at that point.

Correct. Note that *any* kind of limiter is going to cause huge IMD
issues if multiple signals are present, and it is exactly this condition
that is a primary driver for people not using AGC, at least in the
context of this thread.  Using AGC is much better than any kind of
limiter if the intent is to still copy the weak signal in the presence
of the strong one.  The only solution is to reduce gain and get things
linear again.

You have to turn the knob CCW :-)

> A soft limiter on the other hand (such as a pair of back-to-back
> Schottky Diodes) doesn't cause that extreme distortion, and
> allows a much wider usable dynamic range, albeit with some minor
> fuzz on very loud signals.

The Fuzz (harmonic distortion on a single signal) or IMD (if multiple
signals) is the condition that is the problem in the first place.  Any
limiter is nonlinear and will produce IMD.

The extreme IMD from the hard limiter is telling you to back off the
gain until things get linear again.
73,

Lyle KK7P
______________________________________________________________
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

[ac0h]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Exactly!

It continues today.

A couple weeks ago there was a dust up on the list about "Variable Q"
for the APF filter. Can you imagine what's going to happen the first
time someone actually starts monkeying around with that and can't get it
set back to default?

Said user will post ad nausea, "if it was done right in the first place"
he/she wouldn't have been forced to play with it and muck it up beyond
recognition, requiring a call to Elecraft support, the hard reset, and 6
hours lost in the CQWW CW.

I guarantee said user will be posting all sorts of vitriol on the 3830
list. By then it will be the radio and thus Elecraft's fault and no
mention will be made of the budding DSP engineer's attempt to "do it
right" or "their way" or "make the K3 sound like my FT-950".

On 12/5/2010 11:42 AM, Guy Olinger K2AV wrote:

> As to the AGC, all the options are what everybody was hugely clamoring
> for so everybody could "do it their way."  So you got what you wanted.
> The flip side of that is that all these variations have to have a
> value set BY YOU. Not by Elecraft, BY YOU.  All you guys that wanted
> all this flexibility, stand up now and be counted and defend yourself.


- --
R. Kevin Stover

ACØH
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJM/OtPAAoJEAu8BkU0m9MnKgEIAM0E4rr9btFfmNXzwObNDq5R
sRuE5LC0Q/Z8Fq8fzZfl5/Ps8q7tFKtxm3kbPvk5tOISOA2jUJK+KDIIgpk9mECe
jeLiUIlFZUhNyHYjaNXz5vTbXfRhfn1aQTLwOC1LV5Zz+fXwocmttNYYDaVmIDX9
MnBhR6GireBTpS4Jroe64MTbAoAloDGcOR5fhvfdnCFBRbHn4Qe4CU3jSm5rKM3c
1DnfVzmBP4oE4+ifVdgcPr2UW3zheehHf6S85JWotyu+Kyb7wfhRY4e+kbiU7psb
VSwmNKv0dEsAsBT/JjClpiONa8XtpV+I2wNRPxoe4CAI7Yif3cL3/+YSOY6Kfb8=
=1tIJ
-----END PGP SIGNATURE-----

______________________________________________________________
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

[Chris GM3WOJ]

Hello Bob et al

I guessed you might raise this subject, Bob!  I used my 10-week old K3 in CQ WW SSB with the default AGC settings and was a bit surprised by the trouble I had copying any signal less than about S3 when multiple stations around that signal strength were calling. I have local power-line QRN (smooth - similar to white noise) which is about S1-2 and I felt that this might be a factor in the DSP behaviour ?  Just a guess which may be completely wrong.

For my CQ WW CW DXpedition, I set AGC-SLP = 0 and AGC-THR = 8 (as per Dave K6LL's posting) and this made things seem slightly better. Although I was now in a low noise location, I still felt that when a bunch of stations were of similar (but not the same) signal strength and almost on the same frequency, they did become an unreadable mush.

The front-end performance of the K3 is excellent - I had no problem finding a space on any band in either big contest, but the AGC does seem to have some problems, which I hope can be addressed.

Off-topic, I also wonder if the K3 RX has too much gain (I've tried the default RX calibration and also borrowed an XG1 and done the calibration using that) - maybe e.g. IF gain could be made an adjustable parameter ? Or maybe it is the distribution of RX gain ?

73
Chris
GM3WOJ / ZL1CT

Sent from my ancient PIII shack PC