[Elecraft] [K3] Setting RF gain & AGC correctly

[Ian White]

To clarify what Don and Al have said, I don't take any position in the
"Bad Audio" discussion. I doubt if any of us truly understands what the
other guy means by "bad audio" anyway.

My comments were specifically about the so-called "pileup mush" that
results from the poorly chosen default AGC settings; and also about the
lack of follow-through from Elecraft on this topic.

[What follows is a re-post of my earlier message referenced by Al (with
minor edits). For clarity, I don't have any disagreement with the views
of Don, Al or Barry N1EU.

I'm out the door now for the GM DX Convention, so I won't be able to
enter into detailed discussions for the next few days.]

***************

The factory default AGC THReshold for the K3 and K3S is very, very low.
Measurements by KE7X [1] have shown that the default setting of THR=5
corresponds to an AGC onset at less than S3, whereas competing rigs have
a threshold of S7 or even higher. This very low AGC threshold is what
allows the AGC to be activated by band noise.

Elecraft's factory defaults of THR=5 and SLP=12 compress all signals
above the AGC threshold into a very narrow range of audio output levels
[see the measurements by KE7X]. This makes for comfortable
broadcast-style listening - but there is price to be paid. The default
settings greatly reduce the differences in the real-life strengths of
incoming signals - important differences that DXers and contesters
*need* to hear, in order to pick individual signals out of a pileup.

Following complaints from DXers, DXpeditioners and contesters about this
so-called "pileup mush", and following further series of measurements by
KE7X, Elecraft eventually did increase the available range of AGC
THReshold settings (the v4.51 firmware upgrade). 

Following the release of FW v4.51, the consensus within the
[K3-contesting] Yahoo group about the most important settings for DXing
and contesting has been as follows. 

* AGC THR: increase to at least 8 (from default = 5) and preferably to
12, and then take some time to become re-accustomed to the increased
range of audio levels [2]. (Some users recommend as high as 14, if your
ears can handle it.)

* AGC SLP: decrease to 8 or even less (from default = 12). Lower SLP
settings allow stronger signals to sound somewhat louder, and thus more
realistic, but the range of variation will be partly determined by the
THR setting already made; so always experiment with THR first, and with
SLP afterwards. 

* AGC DCY = SOFT (not default=NOR). The User Manual claims that "The
SOFT setting can reduce IMD caused by traditional AGC, and is especially
useful in 'pileup' conditions, in some cases making it unnecessary to
turn AGC off."
 
The settings recommended above have essentially solved the problem of
"pileup mush" for those who have tried them... but regrettably, Elecraft
has done almost nothing to make DXers and contesters  aware that such
changes are possible. 
 
More than 3 years after the firmware release notes boasted that v4.51
"greatly improves signal clarity in pile-ups and other high-noise or
dense-signal situations, especially with threshold set to 12 or higher",
that information is *still* missing from the User Manual. There is also
no information on how to create different 'profiles' of AGC settings for
different types of users.

Meanwhile, Elecraft's factory AGC defaults remain exactly as they were,
still optimized apparently for comfortable 'listening', while many DXers
and contesters remain unaware that their K3/K3S could actually be *so*
much better. 



[1] http://www.ke7x.com/home/miscellaneous-k3-information

[2] It is important to give any increased AGC THReshold setting an
extended trial. To anyone accustomed to an artificially restricted range
of audio signal levels, a higher setting will initially sound quite
'aggressive' - even though the new setting is much closer to real life. 

************

73 from Ian GM3SEK




73 from Ian GM3SEK


>-----Original Message-----
>From: Elecraft [mailto:elecraft-bounces at mailman.qth.net] On Behalf Of
>Don Wilhelm
>Sent: 09 April 2016 01:30
>To: elecraft at mailman.qth.net
>Subject: Re: [Elecraft] [K3] Setting RF gain & AGC correctly
>
>The K3/K3S AGC settings have a lot to do with the apparent "noisy K3"
>syndrome.
>IMHO, the K3 default AGC settings are not optimum.
>Look at my website www,w3fpr.com for information on how to customize
>your K3/K3S/KX3 AGC settings to optimum for your ambient noise and
>operating preferences.  You will find a 'different receiver' when you
>adjust the AGC parameters.
>
>Pay particular attention to the method of evaluating the results. You
>must do the evaluation on pauses in the SSB communications or
>word/sentence breaks in CW.  If you try to evaluate based on band noise
>alone, you will end up adjusting the parameters "backwards", and the
>result will be worse than when you started.
>
>73,
>Don W3FPR
>
>On 4/8/2016 8:04 PM, Al Lorona wrote:
>> Again and again, I have asked (privately) those reporting 'bad K3
audio' to
>furnish me with settings, measurements, recordings -- anything-- but I
>haven't ever received anything with which I could diagnose the problem.
>>
>> Bad. Noisy. Fatiguing. These adjectives have been used continually to
>describe K3 audio. I believe you, I just don't know exactly what you're
>hearing.
>>
>> In the meantime, the analysis and recommendations that Ian, GM3SEK
>made here on 23 March are excellent and what I would consider required
>reading for anyone plagued with 'bad audio'. I must second all that he
said.
>>
>> In conjunction with Ian's tips, if you're not in the habit of using
Bob,
>K4TAX's technique [see below] to set your RF gain correctly, please
consider
>doing so. I have come to believe that many hams may not want to turn
>down their RF Gain (or turn off their preamps, or turn on their
attenuators)
>perhaps for fear of not hearing very weak signals? However, you can't
hear
>anything below the atmospheric noise level... so once the receiver can
hear
>the external noise floor, any further sensitivity or gain doesn't buy
you
>anything and only increases the noisiness -- which is what many of the
>complainants complain about.
>>
>