Nulling QRM

[AD6XY]

I know this is not possible right now. But, is there any way one can use the DSP on SSB to allow adjacent stations to be actively rejected and thereby arranged to not pump the AGC? For unknown reasons in contests the weak DX stations always like to call within a few kHz of the strongest station on the band. As far as I can tell at the moment the AGC is found from the entire bandwidth, not the wanted bandwidth.

My SSB filter is nominally 2.7kHz but it is much wider than that if you go further down the skirts and when signals are 60-80 dB stronger than the weak ones you need to worry about the skirts. When the going is really tough I will have 2kHz or less DSP bandwidth. A tighter filter would help of course as would some form of tunable IF null like we used to use before DSP.  

Mike

[AD6XY]

Thanks Guy

Yes the latter was what I was wondering too - and even if you could use the 2nd RX to determine the metrics for the DSP in the main RX. For example, centre the sub RX on the unwanted signal, or use a wide filter on the sub RX to get sample the splatter. Model this to generate an inverse splatter function and subtract it from the main receiver samples to get just the wanted signal. Far too difficult I suspect but if you could do it you would sell a lot of receivers.

Mike


"It's pretty impressive what the K3 does with no more smarts than bandwidth and not driving the hardware IF into hardware AGC.

It's what's left after the above (splatter, next door neighbor up 5 khz, etc) that's interesting.  Does a splattering signal have a signature in the complex waveform that will allow it to be nulled to some degree? I'm sure that Lyle has a permanent inventor's agenda on that problem.
"

[KK7P]

AD6XY wrote:
> I know this is not possible right now. But, is there any way one can use the
> DSP on SSB to allow adjacent stations to be actively rejected and thereby
> arranged to not pump the AGC?

You have several tools at your disposal, but perhaps an explanation of
what is happening might help.

Any signal within the roofing filter passband that results in a certain
strength (nominally -50 dBm after the roofing filter selectivity) will
activate the hardware AGC to prevent over-range of the analog to digital
converter.

The MDS of the radio is near -140 dBm, so this provides some 90 dB of
dynamic range before the hardware AGC activates.  I find people often
turn the preamp on even when it isn't needed.  This adds noise and
reduces the effective in-band dynamic range. So I suggest turning off
the preamp.  It's easy to tell if you need it: tuned to a dead spot on
the band, disconnect the antenna.  If the noise level change is
detectable without the preamp, you don't need it on and it isn't helping
even with the weakest of stations.  Similarly, turn on the ATTenuator
and see if you can still detect a difference in noise level. If you can,
leave the ATT on.  Turning it off won't help even with the weakest signal.

These two tools: PRE and ATT: allow you to increase the effective
in-band dynamic range by as much as 20 dB!  And misadjusting them can
cost you as much as 20 dB in resistance to strong signals pumping the AGC.

If you are still bothered by very strong signals inside the roofing
filter passband but not in the passband of the signal of interest, it's
time to consider a narrower roofing filter.

The rest of the AGC is done by the DSP after additional filtering.  It
will only respond to energy inside the passband set up by use of the
SHIFT/WIDTH (HI/LO CUT) controls.

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

[AD6XY]

Thanks for the interesting reply Lyle,

I am mostly interested in operating on HF. QRM is not such an issue on HF as the signal to noise is so much worse. I know what you mean about minimizing the gain before the selectivity but at VHF one has a low noise floor to work with so that does not really work. You need a low MDS.

I operate with the XV144 and with the pre-amp off and with a 15 dB attenuator in the 28MHz IF. This does nothing good for the noise figure of course but I have a switchable pre-amp at the front to deal with that if needed. The XV144 has a bit too much gain for my situation and I am looking forward to the internal 144MHz option arriving on my doorstep next month.

I don't think the limit is the hardware AGC - I thought it was the DSP one responding to the very high signal energy within the filter bandwidth but outside the main passband. I agree a tighter filter would help. I would like a variable crystal filter please! I am prepared to build my own if the K3 can be made to tune it.

The K3 deals with this situation very much better than anything else, especially now I have fixed the VCO spurii problem. It is a monthly inter-club sprint contest and I have 2 other stations within a mile and line of sight.

In the interim, I am thinking of perhaps using a second antenna and seeing if I can do this nulling at RF.

Mike

[Elliott Lawrence-3]

My new K3 is complete except for installing the some remaining filters and
the KRX3.  I am in the process of understanding the functionality of all of
the menu items and finding the various buttons!!

I will be using the K3 with the Steppir controller to automatically change
bands as the radio is retuned.  I see that the AUTOINF entry needs to be
changed to AUTO 1 to provide band data information on the RS232 port.  My
question:  When using the RS232 port with the K3 Utility, does this menu
item need to be returned to NOR for proper interface compatibility?

For anyone using the K3 with the Alpha 87A.  Is any setting for the TX DLY
other than NOR 008 more optimum for proper QSK operation?

Thanks and 73
Elliott WA6TLA

______________________________________________________________
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

[Brett Howard]

I had a thought about your comments on the preamp.  Being that many
simply leave it on (personally I don't think I've ever used mine) but
anyway wouldn't the hardware AGC being at max gain be another indicator
that it may be useful to use?  Perhaps the PRE icon could flash if use
of that feature was not helping you much like the NB icon does if one
has high settings and tunes to an area where its no longer needed.

~BTH

On Thu, 2009-09-03 at 06:41 -0700, Lyle Johnson wrote:

> AD6XY wrote:
> > I know this is not possible right now. But, is there any way one can use the
> > DSP on SSB to allow adjacent stations to be actively rejected and thereby
> > arranged to not pump the AGC?
>
> You have several tools at your disposal, but perhaps an explanation of
> what is happening might help.
>
> Any signal within the roofing filter passband that results in a certain
> strength (nominally -50 dBm after the roofing filter selectivity) will
> activate the hardware AGC to prevent over-range of the analog to digital
> converter.
>
> The MDS of the radio is near -140 dBm, so this provides some 90 dB of
> dynamic range before the hardware AGC activates.  I find people often
> turn the preamp on even when it isn't needed.  This adds noise and
> reduces the effective in-band dynamic range. So I suggest turning off
> the preamp.  It's easy to tell if you need it: tuned to a dead spot on
> the band, disconnect the antenna.  If the noise level change is
> detectable without the preamp, you don't need it on and it isn't helping
> even with the weakest of stations.  Similarly, turn on the ATTenuator
> and see if you can still detect a difference in noise level. If you can,
> leave the ATT on.  Turning it off won't help even with the weakest signal.
>
> These two tools: PRE and ATT: allow you to increase the effective
> in-band dynamic range by as much as 20 dB!  And misadjusting them can
> cost you as much as 20 dB in resistance to strong signals pumping the AGC.
>
> If you are still bothered by very strong signals inside the roofing
> filter passband but not in the passband of the signal of interest, it's
> time to consider a narrower roofing filter.
>
> The rest of the AGC is done by the DSP after additional filtering.  It
> will only respond to energy inside the passband set up by use of the
> SHIFT/WIDTH (HI/LO CUT) controls.
>
> 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

______________________________________________________________
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

[Dick Dievendorff]

Hi, Elliot.

The K3 Utility turns AutoInfo to "none" when it needs to, and my intent is
to restore it whenever this is done.

Dick, K6KR


-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Elliott Lawrence
Sent: Thursday, September 03, 2009 7:26 AM
To: [hidden email]
Subject: [Elecraft] New K3 Menu Setting

My new K3 is complete except for installing the some remaining filters and
the KRX3.  I am in the process of understanding the functionality of all of
the menu items and finding the various buttons!!

I will be using the K3 with the Steppir controller to automatically change
bands as the radio is retuned.  I see that the AUTOINF entry needs to be
changed to AUTO 1 to provide band data information on the RS232 port.  My
question:  When using the RS232 port with the K3 Utility, does this menu
item need to be returned to NOR for proper interface compatibility?

For anyone using the K3 with the Alpha 87A.  Is any setting for the TX DLY
other than NOR 008 more optimum for proper QSK operation?

Thanks and 73
Elliott WA6TLA

______________________________________________________________
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

[N5GE]

On Thu, 3 Sep 2009 08:02:48 -0700, "Dick Dievendorff"
<[hidden email]> wrote:

>Hi, Elliot.
>
>The K3 Utility turns AutoInfo to "none" when it needs to, and my intent is
>to restore it whenever this is done.
>
>Dick, K6KR
>
>
>-----Original Message-----
[snip]
>For anyone using the K3 with the Alpha 87A.  Is any setting for the TX DLY
>other than NOR 008 more optimum for proper QSK operation?
>
>Thanks and 73
>Elliott WA6TLA
>

I have my 87A keying and RF inputs connected normally, using the
default settings of the K3 and NO ALC CONNECTION, as recommended by
Elecraft AND Alpha Power.  I find it works very well indeed.

Tom, N5GE

[hidden email]
K3 #806, K3 #1055, PR6,
XV144, XV432, KRC2,
W1 and other small kits.
http://www.n5ge.com
http://www.swotrc.net
 
[snip]

______________________________________________________________
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

[Eric Swartz - WA6HHQ]

I leave my K3 at its default TX Delay for my 87a. It works beautifully.

73, Eric

_..._



Elliott Lawrence wrote:

> My new K3 is complete except for installing the some remaining filters and
> the KRX3.  I am in the process of understanding the functionality of all of
> the menu items and finding the various buttons!!
>
> I will be using the K3 with the Steppir controller to automatically change
> bands as the radio is retuned.  I see that the AUTOINF entry needs to be
> changed to AUTO 1 to provide band data information on the RS232 port.  My
> question:  When using the RS232 port with the K3 Utility, does this menu
> item need to be returned to NOR for proper interface compatibility?
>
> For anyone using the K3 with the Alpha 87A.  Is any setting for the TX DLY
> other than NOR 008 more optimum for proper QSK operation?
>
> Thanks and 73
> Elliott WA6TLA
>
> ______________________________________________________________
> 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

[Lu Romero - W4LT]

Lyle speaks the truth!

Also, dont forget that handy little knob labled "RF GAIN"!  

I find that, in combination with the HI setting in the AF
output, the sdjustable AGC attack and decay settings and the
new DSP parameters since the release of V3r25, amazing
things can be dug out from 200hz above and below a 30 over
S9 CQ'er on 40 meters (using the 500hz filter).

It also works well when 1.3-1.5kHz above or below the same
on SSB (using the 1.8kHz filter).

Reducing the RF Gain also reduces the inherent atmospheric
noise, which the DSP is now quite adept at masking.

It always amazes me how many people never touch the RF gain
control on their receivers, leaving them wide open all the
time.  On 40 and below, most all rigs have way too much
sensitivity, which is actually a hinderance in many
instances, in my opinion.

Note the Kenwood "AIP" button that effectively "deafens" a
TS 450/850/950 rig by shutting off the first IF amplifier.
They should have labeled that button "Contest Receiver
Optimizer".  

-lu-W4LT-
K3 S# 3182
______________________________________________________________
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