[Elecraft] Attenuator and RF Gain Settings

[Don Wilhelm]

Brian,

That is an illusion - yes the K3 gets more quiet (on noise) if the AGC 
Threshold is reduced.
BUT what that indicates is that the AGC is being activated by the noise 
and the K3 is reducing the gain due to AGC action.
The lowest AGC Threshold that is usable for any given level of 
atmospheric noise is the threshold setting just higher than the setting 
which begins to reduce the noise.  In other words, listen to a spot on 
the band where there are no signals, then start at the highest Threshold 
setting of 008.
Reduce the value of the AGC Threshold until you sense that the noise 
level starts to go down - then move the Threshold up to the next level.

That setting will produce the most usable sensitivity for that band, 
that antenna, and at that particular point in time (atmospheric noise 
levels do change with time).  If the resulting noise level is 
bothersome, then take other steps to reduce that response - Preamp off, 
ATT ON, and reduce the RF Gain (in that order) until you can just barely 
hear the noise.  That will result in the most sensitive setting for the 
receiver.
Remember (as Jim Brown just posted) - that atmospheric noise is just 
another 'signal' to the receiver that is to be amplified (it is coming 
in on the antenna port).  One must condition the receiver to place that 
noise level at (or just above) your threshold of perception to reduce 
your fatigue level when listening to any band.  That is what the preamp, 
attenuator, and RF Gain controls are for.

After a bit of experimenting, you may come to the point where you have 
achieved a compromise set of AGC Threshold and Slope setting that work 
well in all but the most demanding situations.  Leave them set at that 
point and control the receiver response with the normal controls of 
preamp, ATT, and RF Gain.  For those digging for signals beneath the 
noise level (yes, some ops have very good ears for that), they will have 
to suffer with hearing some of the noise as well - just how much is an 
individual decision.

Again, as Jim Brown pointed out - noise arriving from the antenna is 
just another signal source - some signals will be above that noise level 
and others will be below it.  Normally we can only copy signals that are 
above the noise level, so why do some insist on setting "all controls to 
the right" and having the receiver amplify the noise as well as the 
signals - it is all in the quantity called "(signal + noise) to noise 
ratio".

The only way to further reduce noise is to use noise reduction 
techniques which search through the signal plus noise to find something 
that appears to be coherent and amplify it.  That is what the K3 NR 
algorithms do.  How well it works depends on the type of noise present 
at the time, and the K3 offers several settings to handle the differing 
situations - do expect distortion with the more aggressive settings, but 
if they are able to dig a signal out of the noise for us, perhaps we can 
put up with the distortion to complete the contact.  This is for use in 
extreme situations, and not for normal use when we want more casual copy.

73,
Don W3FPR

Brian Machesney wrote:
> Wow, Don, the effect of setting the AGC THR is dramatic!
>
> I live in a very rural location that seems relatively free of man-made
> noise. With antenna connected, the K3 S-meter shows about S-5 background
> noise. (S-0 with no antenna!)
>
> Surprising to me, turning *down* the AGC THR to 002 (was set to 008) really
> quieted things down and made signals "pop" out of the noise the way you
> describe. (Note: NF and NB are OFF) I could hear the background noise level
> rise monotonically as I increased the THR setting and fall as I decreased
> it.
>
> With seven AGC variables to adjust -- DCY, HLD, PLS, SLP, THR, -F and -S  --
> setting AGC up for a given operating scenario in a given location seems like
> a very complex topic. Is there any way this could be reduced to a tutorial
> for the K3 Wiki (like K3NA's entry at
> http://www.zerobeat.net/mediawiki/index.php/K3_Receiver_Gain_Configurations,
> which addresses the "no AGC" case)?
>
> This might create a run on N-GENs...
>
>   
> ------------------------------------------------------------------------
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com 
> Version: 9.0.733 / Virus Database: 271.1.1/2663 - Release Date: 02/02/10 02:35:00
>
>