Elecraft

K3 receiver desensing on CW during contest

Classic

List

Threaded

65 messages Options

1234

Peter Wollan-2

Re: K3 receiver desensing on CW during contest

Aha. It sounds like the "desense" problem arises when the DSP filter
is narrower than the roofing filter. A strong station within the
roofing passband but outside the DSP passband will shut down the
receiver through AGC.

I think we all misled ourselves by the impression that desense was
impossible in the K3. It's working just as it ought -- but operators
need to widen out the DSP to the full width of the roofing filter,
when there are strong stations every few Hz. A bit counter-intuitive,
perhaps.

Peter N8MHD

On 2/23/09, K2MK <[hidden email]> wrote:

> Hi Wayne and all:
>
> My filters are 200, 400, 1.0, 1.8 and 2.7 and they are all CW enabled. I
> was using headphones and my sub RX was off. My serial number is 2208 and the
> radio was factory assembled (including the sub RX).
>
> I did not hear key clicks when this was happening. The off frequency
> transmissions were clean. I also made a point of observing the signal
> strength of the offending stations and they were usually S8 or S9. Not the
> 20db or 40db over S9 that I was expecting.
>
> I thought it was interesting that recovery was noticeably slow on the DX
> station's frequency when this was happening. However, under normal
> conditions when US stations were calling directly on top of the DX station,
> recovery was fast. Hence the ability to copy the DX station through the dits
> when the QRM was on frequency.
>
> Maybe a passband shift could of helped. I would like to see the shift limits
> changed to something less than 50Hz. Maybe 10Hz on CW and 50Hz on other
> modes.
>
> I appreciate the interesting ideas from K6LL and W4ZV. Unfortunately I'll
> have to wait for the next busy contest to try them out. This just isn't the
> type of condition you can experience under normal band conditions.
>
> 73,
> Mike K2MK
>
>
>
> wayne burdick
> Sun, 22 Feb 2009 18:26:03 -0800
>
> Mike,
>
> The K3 is virtually desense-proof, with a BDR of ~140 dB. But to take
> advantage of this, you need a narrow crystal filter -- the closer to
> the DSP bandwidth the better. This is exactly the situation that we had
> in mind when we designed the 200-Hz 5-pole filter. For CW pileups, you
> can't beat it.
>
> What crystal filter were you using at the time?
>
> Of course if the transmitting stations are "wide" due to key clicks,
> there may be situations where no amount of filtering can help (for any
> receiver). The DSP noise blanker and NR may be useful sometimes -- you
> might give this a try.
>
> Wayne
> N6KR
>
> On Feb 22, 2009, at 4:45 PM, K2MK wrote:
>
> > I had a great time with my K3 during the ARRL DX contest. I do S&P and
> > I was
> > trolling around with my filter width at 50Hz. Absolutely outstanding.
> > The
> > auto spot is equally outstanding.
> >
> > At 50Hz width it was quite clear that many stations call off frequency.
> > Using RIT, I could see that it was typical for them to be 70Hz or more
> > away
> > from the DX station but I could not hear them in my 50Hz passband. The
> > real
> > problem was when one of them was S9 or greater. They completely swamped
> > weaker DX stations.
>
>
> ______________________________________________________________
> 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
>

Barry N1EU

Re: K3 receiver desensing on CW during contest

Peter Wollan-2 wrote

Aha. It sounds like the "desense" problem arises when the DSP filter
is narrower than the roofing filter. A strong station within the
roofing passband but outside the DSP passband will shut down the
receiver through AGC.

I think we all misled ourselves by the impression that desense was
impossible in the K3. It's working just as it ought -- but operators
need to widen out the DSP to the full width of the roofing filter,
when there are strong stations every few Hz. A bit counter-intuitive,
perhaps.

Correct me if I'm wrong, but a strong station within the
roofing passband but outside the DSP passband will NOT shut down the
receiver through AGC unless the strength exceeds s9+25dB (when hardware AGC kicks in due to ADC dynamic range limits).

73,
Barry N1EU

Bill W4ZV

Re: K3 receiver desensing on CW during contest

In reply to this post by Mike K2MK

Hi Mike,

On Mon, Feb 23, 2009 at 11:00 AM, K2MK <[hidden email]> wrote:

> Here are my CONFIG settings related to the AGC. The only ones I played with
> during the contest were SLP and THR. I always use AGC-F but did occasionally
> try AGC-S.
>
> AGC-HOLD 000
> AGC-PLS NOR
> AGC-SLP 010
> AGC-THR 005
> AGC-F 120
> AGC-S 020

These all look reasonable. I set my AGC-S to the max 040 but I
actually never use AGC-S. If I were an AMer or 75m SSB ragchewer, I
might.

> The differing conditions I described were the situation of everyone calling
> the DX close to his frequency versus one strong station calling 70Hz off
> frequency. When even strong stations called the DX right on frequency I was
> able to hear signals through the QRM. When strong stations called 70Hz off
> of the DX frequency the receiver went quiet and took about 1/4 to 1/2 second
> to recover after he stopped sending.

This is what I'm having problems understanding:

"When even strong stations called the DX right on frequency I was able
to hear signals through the QRM."

1. If two stations are exactly on frequency, the best ears in the
world will not be able to separate them. I can distinguish pitch
*differences* (i.e. one after the other) of about 0.4 Hz repeatably
(see link below), but there's no way I could copy two simultaneous CW
signals dead zero beat, so I don't understand your comment. The human
ear's DSP has an effective bandwidth of ~50 Hz, so I could probably
distinguish two signals 70 Hz apart if their amplitude was not too
different (the human ear has AGC limits also!)

http://tonometric.com/adaptivepitch/ (a fun test to take!)

2. The way the K3 and similar front ends (i.e. Orion, FT2000/9000, IC
7700/7800) work, strong signals inside the roofing filter passband
will desense *all* signals within the passband...i.e. they will affect
zero beat signals just as much as those 70 Hz away (which would be the
case even if you were using the 200 Hz filter). So I cannot
understand how signals off zero beat would be affected more than those
zero beat.

Anyhow, I'm clearly not understanding what you're saying.

73, Bill
______________________________________________________________
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

K7TV

Re: K3 receiver desensing on CW during contest

In reply to this post by Barry N1EU

I believe the hardware AGC would be activated if the *combined* strength of
all signals within the crystal filter passband gets above the threshold of
S9+25 dB or so. This means that multiple unwanted signals don't have to
reach the threshold strength *individually* for AGC pumping to occur, but
there would have to be *a lot* of closely spaced unwanted signals at S8 to
cause a problem if a narrow roofing filter is used. Also under those
conditions, the operator would be unlikely to report that individual
unwanted signals are no more than S8. I suppose one could also have a case
where the wanted signal is almost strong enough to set off the hardware
AGC, and a single, weaker signal gets through the roofing filter (but not
the DSP) and adds just enough to the combined signal to cause hardware AGC
action, resulting in an audible effect on the stronger wanted signal.

73,
Erik K7TV

----- Original Message -----
From: "Barry N1EU" <[hidden email]>
To: <[hidden email]>
Sent: Monday, February 23, 2009 9:59 AM
Subject: Re: [Elecraft] K3 receiver desensing on CW during contest

>
>
> Peter Wollan-2 wrote:
>>
>> Aha. It sounds like the "desense" problem arises when the DSP filter
>> is narrower than the roofing filter. A strong station within the
>> roofing passband but outside the DSP passband will shut down the
>> receiver through AGC.
>>
>> I think we all misled ourselves by the impression that desense was
>> impossible in the K3. It's working just as it ought -- but operators
>> need to widen out the DSP to the full width of the roofing filter,
>> when there are strong stations every few Hz. A bit counter-intuitive,
>> perhaps.
>>
>
> Correct me if I'm wrong, but a strong station within the
> roofing passband but outside the DSP passband will NOT shut down the
> receiver through AGC unless the strength exceeds s9+25dB (when hardware
> AGC
> kicks in due to ADC dynamic range limits).
>
> 73,
> Barry N1EU
> --
> View this message in context:
> http://n2.nabble.com/K3-receiver-desensing-on-CW-during-contest-tp2369819p2373077.html
> Sent from the Elecraft 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
>

Mike K2MK

Re: K3 receiver desensing on CW during contest

In reply to this post by Mike K2MK

Hi Bill:

You stated it correctly. Here's an example.

There are two US stations calling the DX. Me and you both within my 50Hz
passband. I send my call once and you send your call twice. You are S9 at my
receiver and the DX is S7. When I finish calling I can hear you and I can
also hear the DX station. (If there is no pitch difference there are still
differences in strength and speed). I may not hear every character he is
sending but I might hear him come back immediately after I stop sending and
I might hear parts of a number 2 and parts of a letter K in his response. So
when I no longer hear him I take a chance and assume he called me and I
respond with my info. This all happened while you were still sending your
call. This was not an unusual situation this weekend. I had a good signal
and was getting most stations in one or two calls so it was not really a big
leap for me to respond. But I was only able to do it because I could still
hear him through the QRM.

Now take the same situation except you were calling 70Hz off of his
frequency. My receiver was dead until you stopped sending. Once you stopped
I might hear the tail end of his message but I wouldn't know if it was me
unless he called me a second time. And of those 20 occurrences that I
described earlier, maybe half of those times he was calling me but I had to
wait for the re-call.

Someone else asked if I had the text decode on and I did not. As for NR, I
usually had it on. But I always try to keep the aggressiveness low. And I
did experiment and I still had the same problem with NR off.

73,
Mike K2MK

Bill Tippett
Mon, 23 Feb 2009 09:20:14 -0800

Hi Mike,

On Mon, Feb 23, 2009 at 11:00 AM, K2MK <[hidden email]> wrote:

> Here are my CONFIG settings related to the AGC. The only ones I played
> with
> during the contest were SLP and THR. I always use AGC-F but did
> occasionally
> try AGC-S.
>
> AGC-HOLD 000
> AGC-PLS NOR
> AGC-SLP 010
> AGC-THR 005
> AGC-F 120
> AGC-S 020

These all look reasonable. I set my AGC-S to the max 040 but I
actually never use AGC-S. If I were an AMer or 75m SSB ragchewer, I
might.

> The differing conditions I described were the situation of everyone
> calling
> the DX close to his frequency versus one strong station calling 70Hz off
> frequency. When even strong stations called the DX right on frequency I
> was
> able to hear signals through the QRM. When strong stations called 70Hz off
> of the DX frequency the receiver went quiet and took about 1/4 to 1/2
> second
> to recover after he stopped sending.

This is what I'm having problems understanding:

"When even strong stations called the DX right on frequency I was able
to hear signals through the QRM."

1. If two stations are exactly on frequency, the best ears in the
world will not be able to separate them. I can distinguish pitch
*differences* (i.e. one after the other) of about 0.4 Hz repeatably
(see link below), but there's no way I could copy two simultaneous CW
signals dead zero beat, so I don't understand your comment. The human
ear's DSP has an effective bandwidth of ~50 Hz, so I could probably
distinguish two signals 70 Hz apart if their amplitude was not too
different (the human ear has AGC limits also!)

http://tonometric.com/adaptivepitch/ (a fun test to take!)

2. The way the K3 and similar front ends (i.e. Orion, FT2000/9000, IC
7700/7800) work, strong signals inside the roofing filter passband
will desense *all* signals within the passband...i.e. they will affect
zero beat signals just as much as those 70 Hz away (which would be the
case even if you were using the 200 Hz filter). So I cannot
understand how signals off zero beat would be affected more than those
zero beat.

Anyhow, I'm clearly not understanding what you're saying.

73, Bill

______________________________________________________________
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

John W2XS

Re: K3 receiver desensing on CW during contest

Is my summary of the issue correct?

If both signals are within both the DSP and roofing filter bandwidths, then the receiver recovers instantly after each character.

If the stronger signal is outside the DSP filter window (and you can’t hear him) but inside the roofing filter window, then there is de-sensing.

Would passband tuning (shift) help to shift his signal outside of the roofing filter bandpass?

John W2XS

Jim Brown-10

Re: K3 receiver desensing on CW during contest

In reply to this post by Bill W4ZV

It seems that this thread is VERY confused about how AGC works to
control the RF gain. I also suspect that some are using the radio
with too much RF gain. The settings that W4ZV posted are pretty
much what I use, except that I regularly also run the RF gains
(both main and sub) at about 2 o'clock when I hear signals getting
buried in the background noise.

I also think many do not understand the definition of desense.
Desense occurs when a gain stage is driven into clip so badly that
the bias point shifts and reduces the gain (worst case, to
cutoff). So far, I've never heard my K3s desense. I have heard a
strong station pump the AGC when I'm copying a weak station. I've
also heard massive phase noise and clicks from a dirty transmitter
(most recently, a JA who was only about 5dB over S9 when I was
trying to run JAs 1 kHz away from him on 40M). And there's a JA
who made it into my logbook on 80M with a power exchange of 1
watt! He wouldn't have with any other radio.

One expansion of W4ZV's comments regarding perception of pitch.
Human sight and hearing are logarithmic. Human hearing works on
the basis of what acousticians call "critical bands," which are
logarithmic, and which are tied to the physical construction of
the human hearing system itself. The filter between our ears is
better able to separate a 50 Hz pitch difference at 300 Hz than at
900 Hz! That's one reason why some CW operators like to work at
lower pitches when the going gets tough.

73,

Jim Brown K9YC
(Member Acoustical Society of America)

______________________________________________________________
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

Steve Ellington

Re: K3 receiver desensing on CW during contest

Is this correct?
If a cw signal falls within the DSP passband it should and will pump the
AGC. If a CW signal falls outside the DSP passband it should not pump the
DSP's AGC. The problem here is defining what the DSP's passband it. If a
signal is just outside the audio range of the DSP (can't be heard) then I
would consider it "outside the DSP passband" and it should not pump the AGC
however this is never the case. I see moderate signals just slightly outside
the audio passband that pump the AGC. This is partly what is confusing us.
Signals that we can't hear pumping the AGC worry us. What we hear coming
from the speaker doesn't match how the AGC is responding. If my WIDTH is set
for 100hz and a signal is at 110hz, I won't hear him but my S-meter responds
to him as well as my AGC "desenses" (reduces gain).
All of this has nothing to do with the roofing filter.
Steve Ellington
[hidden email]
----- Original Message -----
From: "Jim Brown" <[hidden email]>
To: "Elecraft List" <[hidden email]>
Sent: Monday, February 23, 2009 1:46 PM
Subject: Re: [Elecraft] K3 receiver desensing on CW during contest

> It seems that this thread is VERY confused about how AGC works to
> control the RF gain. I also suspect that some are using the radio
> with too much RF gain. The settings that W4ZV posted are pretty
> much what I use, except that I regularly also run the RF gains
> (both main and sub) at about 2 o'clock when I hear signals getting
> buried in the background noise.
>
> I also think many do not understand the definition of desense.
> Desense occurs when a gain stage is driven into clip so badly that
> the bias point shifts and reduces the gain (worst case, to
> cutoff). So far, I've never heard my K3s desense. I have heard a
> strong station pump the AGC when I'm copying a weak station. I've
> also heard massive phase noise and clicks from a dirty transmitter
> (most recently, a JA who was only about 5dB over S9 when I was
> trying to run JAs 1 kHz away from him on 40M). And there's a JA
> who made it into my logbook on 80M with a power exchange of 1
> watt! He wouldn't have with any other radio.
>
> One expansion of W4ZV's comments regarding perception of pitch.
> Human sight and hearing are logarithmic. Human hearing works on
> the basis of what acousticians call "critical bands," which are
> logarithmic, and which are tied to the physical construction of
> the human hearing system itself. The filter between our ears is
> better able to separate a 50 Hz pitch difference at 300 Hz than at
> 900 Hz! That's one reason why some CW operators like to work at
> lower pitches when the going gets tough.
>
> 73,
>
> Jim Brown K9YC
> (Member Acoustical Society of America)
>
>
> ______________________________________________________________
> 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

--------------------------------------------------------------------------------

No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.0.237 / Virus Database: 270.11.3/1967 - Release Date: 02/23/09
07:17:00

______________________________________________________________
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

Merv Schweigert

Re: K3 receiver desensing on CW during contest

In reply to this post by Jim Brown-10

From what I though the original post was about the trouble copying signals
in a pile up, it appears to me that thread has become at least 3 different
topics.
Seems there are people talking desense, AGC settings, and the pile up
problem.
To make mine clear, I do not have any desense problems with the K3,
I can work within hundreds of HZ from very strong stations and have
never had it blank the receiver when using narrow filters.
AGC, I dont think it is the problem as I have it turned off most of the
time and use the RF gain.
The third problem is what I have had and apparently others, when there
are many stations calling in a pile up that are a sig strength of say s4
to s5,
the audio turns to mush and they all sound the same, no difference in
strength, like they are all clipped somehow to the same level and
distorted.
Not a pure audio tone any longer, does not happen with just a couple
stations in the pile up, only when maybe over 5 or 6 at a time are calling.
Turns to mush as some one said. Have not tried turning the radio on and
off when that happens, but will next time to see it that resets something.
In an case dont lump all these comments that have come under the desense
heading as one problem, it appears to be at least 3.. Perhaps the
subject is
incorrectly titled.
Merv KH7C

Darwin, Keith

Re: K3 receiver desensing on CW during contest

In reply to this post by Steve Ellington

I'm at work and don't have time to check any of this ... But I was under
the impression that it was not this way. Signals outside of the DSP
passband but inside the roofing filter passband would activate the DSP.
So, with a 2 KHz roofing filter and a 200 Hz DSP, you'd have very few
signals being heard by the OP but lots of signals affecting the AGC.

- Keith N1AS -
- K3 711 -

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Steve Ellington
Sent: Monday, February 23, 2009 1:58 PM
To: Jim Brown; Elecraft List
Subject: Re: [Elecraft] K3 receiver desensing on CW during contest

Is this correct?
If a cw signal falls within the DSP passband it should and will pump the
AGC. If a CW signal falls outside the DSP passband it should not pump
the DSP's AGC. Steve Ellington
[hidden email]
______________________________________________________________
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

Mike K2MK

Re: K3 receiver desensing on CW during contest

In reply to this post by Mike K2MK

Hi John:

You are correct. When there was no strong off center signal but two or more
strong signals within the 50Hz passband it was possible to copy one signal,
as we all do, in the presence of QRM. There was no noticeable AGC action. It
was just a normal situation of trying to copy one signal out of several.
Even at 50Hz you can't remove 100% of all QRM.

When an S8 or S9 signal was around 70Hz away the signals inside the 50Hz
passband were quiet until he stopped sending. I wanted to hear what was
causing the problem so I used RIT to tune up or down until I could hear the
off center signal. When I did I found that the offset was about 70Hz as read
on the lower display. I could then also see his signal strength on the S
meter. I wish I could say what happened when I opened up the passband but I
don't recall.

I thought a passband shift might help but the shift is limited to 50Hz
increments. So one touch of the shift knob caused me to lose the signal of
interest.

73,
Mike K2MK

John W2XS
Mon, 23 Feb 2009 10:18:02 -0800

Is my summary of the issue correct?

If both signals are within both the DSP and roofing filter bandwidths, then
the receiver recovers instantly after each character.

If the stronger signal is outside the DSP filter window (and you can't hear
him) but inside the roofing filter window, then there is de-sensing.

Would passband tuning (shift) help to shift his signal outside of the
roofing filter bandpass?

John W2XS

______________________________________________________________
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

K7TV

Re: K3 receiver desensing on CW during contest

In reply to this post by Darwin, Keith

There are two separate AGC functions:

1. Hardware AGC is activated when the combined signals that get through the
roofing filter reach a certain level, about S9+25.
When it is activated, all those signals, including the wanted signal, are
attenuated before the DSP.
AFAIK the hardware AGC cannot be turned off.

2. Software AGC implemented within the DSP. It can be turned on or off, and
its parameters can be adjusted. I am not sure whether it is ever totally
turned off, even when "turned off". It reacts to the strenght of all that
has passed through the DSP bandwidth.

Signals outside the DSP passband but inside the roofing filter certainly
enter the DSP, but the DSP bandpass filter should eliminate them before they
can do anything to the software AGC. They can affect the hardware AGC, but
only if they together (including the wanted signal) are so strong that the
hardware AGC does anything at all.

Erik K7TV

----- Original Message -----
From: "Darwin, Keith" <[hidden email]>
To: "Elecraft List" <[hidden email]>
Sent: Monday, February 23, 2009 12:22 PM
Subject: Re: [Elecraft] K3 receiver desensing on CW during contest

> I'm at work and don't have time to check any of this ... But I was under
> the impression that it was not this way. Signals outside of the DSP
> passband but inside the roofing filter passband would activate the DSP.
> So, with a 2 KHz roofing filter and a 200 Hz DSP, you'd have very few
> signals being heard by the OP but lots of signals affecting the AGC.
>
> - Keith N1AS -
> - K3 711 -
>
> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Steve Ellington
> Sent: Monday, February 23, 2009 1:58 PM
> To: Jim Brown; Elecraft List
> Subject: Re: [Elecraft] K3 receiver desensing on CW during contest
>
> Is this correct?
> If a cw signal falls within the DSP passband it should and will pump the
> AGC. If a CW signal falls outside the DSP passband it should not pump
> the DSP's AGC. Steve Ellington
> [hidden email]
> ______________________________________________________________
> 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
>

Arie Kleingeld PA3A-2

Re: K3 receiver desensing on CW during contest

This is an interesting subject.
I've read several opinions on how the K3-AGC works here and MORE
important: how it behaves.

Can somebody from Elecraft shine a light on the AGC behaviour? I'm very
curious now.

73,
Arie PA3A

______________________________________________________________
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

w0mu

Re: K3 receiver desensing on CW during contest

In reply to this post by Merv Schweigert

This is what I was seeing too. I never had any issues with wide or clicky
stations causing any issues. The radio is great to use. The K3 produced
the least amount of listening fatigue that I have experienced. A good
thing!

Not being able to pull out callsigns out of a big pile is not a good thing.

"A slip of the foot you may soon recover, but a slip of the tongue you may
never get over." Ben Franklin
-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Merv Schweigert
Sent: Monday, February 23, 2009 12:15 PM
Cc: Elecraft List
Subject: Re: [Elecraft] K3 receiver desensing on CW during contest

From what I though the original post was about the trouble copying signals
in a pile up, it appears to me that thread has become at least 3 different
topics.
Seems there are people talking desense, AGC settings, and the pile up
problem.
To make mine clear, I do not have any desense problems with the K3, I can
work within hundreds of HZ from very strong stations and have never had it
blank the receiver when using narrow filters.
AGC, I dont think it is the problem as I have it turned off most of the
time and use the RF gain.
The third problem is what I have had and apparently others, when there are
many stations calling in a pile up that are a sig strength of say s4 to s5,
the audio turns to mush and they all sound the same, no difference in
strength, like they are all clipped somehow to the same level and
distorted.
Not a pure audio tone any longer, does not happen with just a couple
stations in the pile up, only when maybe over 5 or 6 at a time are calling.
Turns to mush as some one said. Have not tried turning the radio on and off
when that happens, but will next time to see it that resets something.
In an case dont lump all these comments that have come under the desense
heading as one problem, it appears to be at least 3.. Perhaps the subject
is incorrectly titled.
Merv KH7C

> It seems that this thread is VERY confused about how AGC works to
> control the RF gain. I also suspect that some are using the radio with
> too much RF gain. The settings that W4ZV posted are pretty much what I
> use, except that I regularly also run the RF gains (both main and sub)
> at about 2 o'clock when I hear signals getting buried in the
> background noise.
>
> I also think many do not understand the definition of desense.
> Desense occurs when a gain stage is driven into clip so badly that the
> bias point shifts and reduces the gain (worst case, to cutoff). So
> far, I've never heard my K3s desense. I have heard a strong station
> pump the AGC when I'm copying a weak station. I've also heard massive
> phase noise and clicks from a dirty transmitter (most recently, a JA
> who was only about 5dB over S9 when I was trying to run JAs 1 kHz away
> from him on 40M). And there's a JA who made it into my logbook on 80M
> with a power exchange of 1 watt! He wouldn't have with any other
> radio.
>
> One expansion of W4ZV's comments regarding perception of pitch.
> Human sight and hearing are logarithmic. Human hearing works on the
> basis of what acousticians call "critical bands," which are
> logarithmic, and which are tied to the physical construction of the
> human hearing system itself. The filter between our ears is better
> able to separate a 50 Hz pitch difference at 300 Hz than at 900 Hz!
> That's one reason why some CW operators like to work at lower pitches
> when the going gets tough.
>
> 73,
>
> Jim Brown K9YC
> (Member Acoustical Society of America)
>
>
> ______________________________________________________________
> 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

______________________________________________________________
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

N8LP

Re: K3 receiver desensing on CW during contest

In reply to this post by Bill W4ZV

When I first received my K3, I too was disappointed with the BDR until I understood what was happening.

In an effort to quantify the BDR of signals between the xtal filter bandpass and the DSP bandpass, I set up the following scenario...

Two generators feeding the K3 through a hybrid combiner.
One set to about -85dBm (S7 on my K3), and the other variable.
I tuned in the -85dBm signal, with the xtal filter set to 500Hz, and DSP to 50Hz.

With the "interfering" signal set to 200Hz offset (within the xtal filter passband) I started to see desense of the weaker signal at -34dBm output level for the stronger signal. With the interfering signal set to 2kHz offset, I started to see desense at +15dBm. The difference of 49dB shows the importance of the xtal filter to this design. Of course, different AGC settings would affect this test. The promised variable roofing filters would reduce the problem quite a bit, but that opens up other issues.

I think the days of receivers with xtal filters are numbered. High speed ADCs capable of 140dB dynamic range without xtal filtering are on the horizon. A 20-bit ADC with enough processing gain would do it.

73,
Larry N8LP

Bill W4ZV wrote

K2MK wrote

I thought it was interesting that recovery was noticeably slow on the DX
station's frequency when this was happening. However, under normal
conditions when US stations were calling directly on top of the DX station,
recovery was fast. Hence the ability to copy the DX station through the dits
when the QRM was on frequency.

Mike this definitely sounds like desense but I'm surprised you experienced it with only S8-9 signals. I'm not sure what you mean about differing conditions on and off the center frequency. Desense will affect everything inside the roofing filter.

What speed AGC were you using (F or S) and what were your AGC-F and AGC-S CONFIG settings? I NEVER use AGC-S on either CW or SSB and like at least 120 dB/sec for AGC-F. Also, help your AGC do its job by not running RF GAIN fully clockwise (as mentioned in my previous message).

73, Bill

P.S. As I said before my K3 worked extremely well coping weak signals in the midst of massive ones on 160 this weekend. I smiled when I saw the following comment from OL0W(OK1DSZ):

"Only for fun, when the band was already closing, I decided to try what I will
be able to work with QRP. Using longer callsign OK1DSZ and the power reduced to
5W I spend about 10 minutes calling several east coast big guns. The award for
the best ears goes to W4ZV, who was worked as easily as with the kilowatt
earlier in the night. K8PO and W2FU were worked with several repeats. The
others just kept CQing without any reaction with exception of K1TTT, who
apparently heard something, but not enough for the QSO."

Alan Bloom

Re: K3 receiver desensing on CW during contest

On Mon, 2009-02-23 at 18:40, N8LP wrote:
...
> I think the days of receivers with xtal filters are numbered. High speed
> ADCs capable of 140dB dynamic range without xtal filtering are on the
> horizon. A 20-bit ADC with enough processing gain would do it.

I don't think you'll find a 20-bit ADC with a high enough sample rate to
digitize the 3-30 MHz HF band (i.e. >65-70 MHz or so). At least not at
a reasonable cost.

I believe the best suitable, reasonable-cost ADCs available these days
are able to achieve a 500-Hz blocking dynamic range in the low 120's dB,
maybe 15-20 dB worse than the K3. That's significantly better than the
previous generation of ADCs could achieve, and no doubt someday we'll
get even better parts that are good enough to challenge the traditional
superhet/crystal filter architecture. But I don't believe we're close
to that level of performance today.

Another issue, of course, is spurious responses. I'm pretty sure that
current ADCs don't have good enough spurious-free dynamic range to
challenge a state-of-the-art receiver like the K3.

Al N1AL

______________________________________________________________
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

Arie Kleingeld PA3A-2

Re: K3 receiver desensing on CW during contest

In reply to this post by N8LP

Thanks Larry,

These measurements explain a lot of what's happening.

73,
Arie PA3A

-----Oorspronkelijk bericht-----
Van: [hidden email]
[mailto:[hidden email]] Namens N8LP
Verzonden: dinsdag 24 februari 2009 3:41
Aan: [hidden email]
Onderwerp: Re: [Elecraft] K3 receiver desensing on CW during contest

When I first received my K3, I too was disappointed with the BDR until I
understood what was happening.

In an effort to quantify the BDR of signals between the xtal filter
bandpass and the DSP bandpass, I set up the following scenario...

Two generators feeding the K3 through a hybrid combiner.
One set to about -85dBm (S7 on my K3), and the other variable. I tuned
in the -85dBm signal, with the xtal filter set to 500Hz, and DSP to
50Hz.

With the "interfering" signal set to 200Hz offset (within the xtal
filter
passband) I started to see desense of the weaker signal at -34dBm output
level for the stronger signal. With the interfering signal set to 2kHz
offset, I started to see desense at +15dBm. The difference of 49dB
shows the importance of the xtal filter to this design. Of course,
different AGC settings would affect this test. The promised variable
roofing filters would reduce the problem quite a bit, but that opens up
other issues.

I think the days of receivers with xtal filters are numbered. High speed
ADCs capable of 140dB dynamic range without xtal filtering are on the
horizon. A 20-bit ADC with enough processing gain would do it.

73,
Larry N8LP

Bill W4ZV wrote:
>
>
>
> K2MK wrote:
>>
>> I thought it was interesting that recovery was noticeably slow on the

>> DX
>> station's frequency when this was happening. However, under normal
>> conditions when US stations were calling directly on top of the DX
>> station,
>> recovery was fast. Hence the ability to copy the DX station through
the

>> dits
>> when the QRM was on frequency.
>>
>
> Mike this definitely sounds like desense but I'm surprised you
> experienced it with only S8-9 signals. I'm not sure what you mean
> about differing conditions on and off the center frequency. Desense
> will affect everything inside the roofing filter.
>
> What speed AGC were you using (F or S) and what were your AGC-F and
> AGC-S CONFIG settings? I NEVER use AGC-S on either CW or SSB and like

> at least 120 dB/sec for AGC-F. Also, help your AGC do its job by not
> running RF GAIN fully clockwise (as mentioned in my previous message).
>
> 73, Bill
>
> P.S. As I said before my K3 worked extremely well coping weak signals

> in the midst of massive ones on 160 this weekend. I smiled when I saw

> the following comment from OL0W(OK1DSZ):
>
> "Only for fun, when the band was already closing, I decided to try
> what I will be able to work with QRP. Using longer callsign OK1DSZ and

> the power reduced to
> 5W I spend about 10 minutes calling several east coast big guns. The
award
> for
> the best ears goes to W4ZV, who was worked as easily as with the
kilowatt
> earlier in the night. K8PO and W2FU were worked with several repeats.
The
> others just kept CQing without any reaction with exception of K1TTT,
who
> apparently heard something, but not enough for the QSO."
>
>

--
View this message in context:
http://n2.nabble.com/K3-receiver-desensing-on-CW-during-contest-tp236981
9p2375849.html
Sent from the Elecraft 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

Geoffrey Mackenzie-Kennedy

Re: K3 receiver desensing on CW during contest

In reply to this post by Mike K2MK

Larry N8LP wrote on 24 Februay 2009, at 02:40.

> I think the days of receivers with xtal filters are numbered. High speed
> ADCs capable of 140dB dynamic range without xtal filtering are on the
> horizon. A 20-bit ADC with enough processing gain would do it.

------------------------------
Larry,

You could be right, given an adequate amount of R&D funding and time.

When the performance of a state of the art HF receiver using xtal filters
will be surpassed by one digitizing at signal frequency is an open question
I think.. The performance standard in terms of Spurious Free 3rd
Order dynamic range vs. time set in recent years by real HF superhet
receivers using xtal filters, as I know it to be, is:

Year 3rd Order SFDR 3rd Order SFDR Filter Noise
@ 2 kHz In Passband BW Figure

1993 110 db 95 db 400 Hz 8 db

2007 122 db approx 115 db 500 Hz 7 db

So don't put the old horse out to pasture just yet :-)

Sorry to have digressed from the subject matter.

73,
Geoff
GM4ESD

______________________________________________________________
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

Re: K3 receiver desensing on CW during contest

Let's see: 122 to 140dB: only 15% increase! a walk in the park...
; - )

David
G3UNA

---- Geoffrey Mackenzie-Kennedy <[hidden email]> wrote:

> Larry N8LP wrote on 24 Februay 2009, at 02:40.
>
> > I think the days of receivers with xtal filters are numbered. High speed
> > ADCs capable of 140dB dynamic range without xtal filtering are on the
> > horizon. A 20-bit ADC with enough processing gain would do it.
>
> ------------------------------
> Larry,
>
> You could be right, given an adequate amount of R&D funding and time.
>
> When the performance of a state of the art HF receiver using xtal filters
> will be surpassed by one digitizing at signal frequency is an open question
> I think.. The performance standard in terms of Spurious Free 3rd
> Order dynamic range vs. time set in recent years by real HF superhet
> receivers using xtal filters, as I know it to be, is:
>
> Year 3rd Order SFDR 3rd Order SFDR Filter Noise
> @ 2 kHz In Passband BW Figure
>
> 1993 110 db 95 db 400 Hz 8 db
>
> 2007 122 db approx 115 db 500 Hz 7 db
>
>
> So don't put the old horse out to pasture just yet :-)
>
> Sorry to have digressed from the subject matter.
>
> 73,
> Geoff
> GM4ESD
>
> ______________________________________________________________
> 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

Bill W4ZV

Re: K3 receiver desensing on CW during contest

In reply to this post by Geoffrey Mackenzie-Kennedy

Geoffrey Mackenzie-Kennedy wrote

Larry N8LP wrote on 24 Februay 2009, at 02:40.

> I think the days of receivers with xtal filters are numbered.

So don't put the old horse out to pasture just yet :-)

I recall Icom's President Mr. Inouye saying something similar to Larry's comment just after the IC-756PRO was introduced (~10 years ago). Then Icom added even more crystal filters to keep the 7600/7700/7800 in the performance ballpark with Orion...and then the K3. I bet he wishes he had never said that.

73, Bill

1234