Elecraft › [K3]

Re: New K3 Firmware: Improved AGC (great for pile-ups as well as general operation)

Classic

List

Threaded

10 messages Options

Thomas Horsten

Re: New K3 Firmware: Improved AGC (great for pile-ups as well as general operation)

Wayne,

It's hard for me to do a before/after side by side comparison since I only
have one K3 and the band conditions change all the time, but after using
the 4.51 firmware for a couple of days, it seems to me that APF is maybe
slightly less effective than it was before. Maybe it's because the overall
clarity of the weak signals is generally improved, or else the AGC changes
somehow interact with the APF in a detrimental way?

I'm generally using the 400Hz filter for CW and adjust the passband down to
about 250Hz for best readability on weak signals, engaging APF at that
setting previously made most weak signals much louder, now it seems to make
that much difference only on a smaller subset of signals and only when the
band is relatively quiet.

It could well be down to random band condition changes before and after the
upgrade, but I thought I'd mention it anyway.

73, Thomas M0TRN

On 14 May 2012 05:14, Wayne Burdick <[hidden email]> wrote:

> K3 beta-test firmware revision 4.51 (with DSP rev. 2.76) is now
> available. We'd like to thank the 100+ K3 owners who have put this
> code through its paces over the past week.
>
> The most important change is that the K3's AGC threshold (CONFIG:AGC
> THR) can now be set much higher. In addition, the AGC now has very
> precise response on attack/decay (monotonic, not "scalloped"). The
> combination of these changes can greatly improve signal clarity in
> pile-ups and other high-noise or dense-signal situations. Many
> operators have also reported improved effectiveness of noise blanking.
>
> With this release, NR and Auto-Notch can be used at all of the new
> threshold settings.
>
> The firmware can be found on our K3 software download page:
>
> http://www.elecraft.com/K3/k3_software.htm
>
> 73,
> Wayne
> N6KR
>
> ______________________________________________________________
> 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

wayne burdick

Re: New K3 Firmware: Improved AGC (great for pile-ups as well as general operation)

Administrator

Hi Thomas,

I haven't noticed this myself, and no one else has reported it that I
know of, but I'll listen for it. We didn't change the APF algorithm in
any way.

73
Wayne
N6KR

On May 14, 2012, at 2:31 AM, Thomas Horsten wrote:

> Wayne,
>
> It's hard for me to do a before/after side by side comparison since
> I only have one K3 and the band conditions change all the time, but
> after using the 4.51 firmware for a couple of days, it seems to me
> that APF is maybe slightly less effective than it was before. Maybe
> it's because the overall clarity of the weak signals is generally
> improved, or else the AGC changes somehow interact with the APF in a
> detrimental way?

______________________________________________________________
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

Rick WA6NHC

Re: New K3 Firmware: Improved AGC (great for pile-ups aswell as general operation)

I think the new AGC sounds great. However, auto-notch is no longer working.
I can manually notch, but auto mode does nothing (LSB).

Firmware is 4.51 and I've also installed the latest DSP firmware (in both
DSP). I've also cycled the power, no change.

AGC DCY Nor
AGC Hold 0.00
AGC PLS Nor
AGC SLP 000 (tried 001, same results)
AGC THR 016 (tried different values, no change)
AGC F 120
AGC S 020

Rick wa6nhc

______________________________________________________________
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

Re: New K3 Firmware: Improved AGC (great for pile-ups aswell as general operation)

I have 4.51 in and the auto notch is working fine.

Did your log on the utility show the DSP transferring as well? You
need to use the transfer all files. On my CONFIG: FW REVS do I have
uC=04.51, d1=02.76, d2=02.76, FL=01.15, dr=01.03

I checked the contents of the zip in the beta directory with the zip
of 4.51 distributed to pre-beta testers and they look the same.

73, Guy

On Mon, May 14, 2012 at 2:03 PM, Rick Bates <[hidden email]> wrote:

> I think the new AGC sounds great. However, auto-notch is no longer working.
> I can manually notch, but auto mode does nothing (LSB).
>
> Firmware is 4.51 and I've also installed the latest DSP firmware (in both
> DSP). I've also cycled the power, no change.
>
> AGC DCY Nor
> AGC Hold 0.00
> AGC PLS Nor
> AGC SLP 000 (tried 001, same results)
> AGC THR 016 (tried different values, no change)
> AGC F 120
> AGC S 020
>
> Rick wa6nhc
>
>
> ______________________________________________________________
> 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

Rick WA6NHC

Re: New K3 Firmware: Improved AGC (great for pile-ups aswell as general operation)

Hi Guy,

Lyle also contacted me and it's appearing that AGC THR at 16 is not allowing
a 15 db signal (P3) to not get notched. It works beautifully at THR 7,
completely gone.

I sat and watched as both DSP were updated and it shows the appropriate
codes on the display.

It's likely along the lines of me not fully understanding what I'm doing.
;-)

73,
Rick

-----Original Message-----
From: Guy Olinger K2AV

I have 4.51 in and the auto notch is working fine.

Did your log on the utility show the DSP transferring as well? You
need to use the transfer all files. On my CONFIG: FW REVS do I have
uC=04.51, d1=02.76, d2=02.76, FL=01.15, dr=01.03

I checked the contents of the zip in the beta directory with the zip
of 4.51 distributed to pre-beta testers and they look the same.

73, Guy

On Mon, May 14, 2012 at 2:03 PM, Rick Bates <[hidden email]>
wrote:
> I think the new AGC sounds great. However, auto-notch is no longer
working.

> I can manually notch, but auto mode does nothing (LSB).
>
> Firmware is 4.51 and I've also installed the latest DSP firmware (in both
> DSP). I've also cycled the power, no change.
>
> AGC DCY Nor
> AGC Hold 0.00
> AGC PLS Nor
> AGC SLP 000 (tried 001, same results)
> AGC THR 016 (tried different values, no change)
> AGC F 120
> AGC S 020
>
> Rick wa6nhc
>
>
> ______________________________________________________________
> 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

Pierfrancesco Caci

Re: New K3 Firmware: Improved AGC (great for pile-ups aswell as general operation)

It nicely notches away a broadcasting carrier at 7200 kHz here, tried
with THR 12, 16 and 20.

Pf

>>>>> "Rick" == Rick Bates <[hidden email]> writes:

Rick> Hi Guy,
Rick> Lyle also contacted me and it's appearing that AGC THR at 16 is not allowing
Rick> a 15 db signal (P3) to not get notched. It works beautifully at THR 7,
Rick> completely gone.

Rick> I sat and watched as both DSP were updated and it shows the appropriate
Rick> codes on the display.

Rick> It's likely along the lines of me not fully understanding what I'm doing.
Rick> ;-)

Rick> 73,
Rick> Rick

Rick> -----Original Message-----
Rick> From: Guy Olinger K2AV

Rick> I have 4.51 in and the auto notch is working fine.

Rick> Did your log on the utility show the DSP transferring as well? You
Rick> need to use the transfer all files. On my CONFIG: FW REVS do I have
Rick> uC=04.51, d1=02.76, d2=02.76, FL=01.15, dr=01.03

Rick> I checked the contents of the zip in the beta directory with the zip
Rick> of 4.51 distributed to pre-beta testers and they look the same.

Rick> 73, Guy

Rick> On Mon, May 14, 2012 at 2:03 PM, Rick Bates <[hidden email]>
Rick> wrote:
>> I think the new AGC sounds great. However, auto-notch is no longer
Rick> working.
>> I can manually notch, but auto mode does nothing (LSB).
>>
>> Firmware is 4.51 and I've also installed the latest DSP firmware (in both
>> DSP). I've also cycled the power, no change.
>>
>> AGC DCY Nor
>> AGC Hold 0.00
>> AGC PLS Nor
>> AGC SLP 000 (tried 001, same results)
>> AGC THR 016 (tried different values, no change)
>> AGC F 120
>> AGC S 020
>>
>> Rick wa6nhc
>>
>>
>> ______________________________________________________________
>> 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

Rick> ______________________________________________________________
Rick> Elecraft mailing list
Rick> Home: http://mailman.qth.net/mailman/listinfo/elecraft
Rick> Help: http://mailman.qth.net/mmfaq.htm
Rick> Post: mailto:[hidden email]

Rick> This list hosted by: http://www.qsl.net
Rick> Please help support this email list: http://www.qsl.net/donate.html

--
Pierfrancesco Caci, ik5pvx
______________________________________________________________
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

Re: New K3 Firmware: Improved AGC (great for pile-ups aswell as general operation)

In reply to this post by Rick WA6NHC

HMM. I moved the AGC slope to 20 and then went listening. Was
working somewhat, but not well. HOWEVER, when I cycled the notch to
MANUAL, then OFF and then ON, it worked just fine. ????

73, Guy.

On Mon, May 14, 2012 at 2:59 PM, Rick Bates <[hidden email]> wrote:

> Hi Guy,
>
> Lyle also contacted me and it's appearing that AGC THR at 16 is not allowing
> a 15 db signal (P3) to not get notched. It works beautifully at THR 7,
> completely gone.
>
> I sat and watched as both DSP were updated and it shows the appropriate
> codes on the display.
>
> It's likely along the lines of me not fully understanding what I'm doing.
> ;-)
>
> 73,
> Rick
>
> -----Original Message-----
> From: Guy Olinger K2AV
>
> I have 4.51 in and the auto notch is working fine.
>
> Did your log on the utility show the DSP transferring as well? You
> need to use the transfer all files. On my CONFIG: FW REVS do I have
> uC=04.51, d1=02.76, d2=02.76, FL=01.15, dr=01.03
>
> I checked the contents of the zip in the beta directory with the zip
> of 4.51 distributed to pre-beta testers and they look the same.
>
> 73, Guy
>
> On Mon, May 14, 2012 at 2:03 PM, Rick Bates <[hidden email]>
> wrote:
>> I think the new AGC sounds great. However, auto-notch is no longer
> working.
>> I can manually notch, but auto mode does nothing (LSB).
>>
>> Firmware is 4.51 and I've also installed the latest DSP firmware (in both
>> DSP). I've also cycled the power, no change.
>>
>> AGC DCY Nor
>> AGC Hold 0.00
>> AGC PLS Nor
>> AGC SLP 000 (tried 001, same results)
>> AGC THR 016 (tried different values, no change)
>> AGC F 120
>> AGC S 020
>>
>> Rick wa6nhc
>>
>>
>> ______________________________________________________________
>> 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
>

Rick WA6NHC

Re: New K3 Firmware: Improved AGC (great for pile-upsaswell as general operation)

In reply to this post by Pierfrancesco Caci

Thanks guys for the input. I had an off list conversation with Lyle and got
my answer, Wayne was included too.

The scenario was the morning 75M 'old guy' net. During the lulls, there was
a ~15 db above the -110 db noise floor signal (I LOVE the P3!). Auto notch
would not take it completely out until I lowered AGC THR to 7 (from 16).

I restarted, cycled through auto and manual notch, no joy with auto until I
dropped the THR.

If I understand it correctly, as you raise THR and/or SLP, it gets harder
for the auto notch to remove weaker signals (which this apparently is).

So now my goal is to find a reasonable compromise or define two macros to
make simple and quick changes, one for ragchewing and another for DXing.
But first I want to better understand what THR and SLP do and how they
interact. (NOT tonight!)

Thanks again all (grazie e ciao!),

Rick wa6nhc

-----Original Message-----
From: Pierfrancesco Caci

It nicely notches away a broadcasting carrier at 7200 kHz here, tried
with THR 12, 16 and 20.

Pf

______________________________________________________________
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

wayne burdick

Re: New K3 Firmware: Improved AGC (great for pile-upsaswell as general operation)

Administrator

We'll eventually be making AGC THR and AGC SLP per-mode, which might
help with such situations. You could optimize these settings
differently for SSB (so that auto-notch worked optimally) and for CW
(to allow for the case where many signals are nearly on top of each
other).

73,
Wayne
N6KR

On May 14, 2012, at 8:56 PM, Rick Bates wrote:

> Thanks guys for the input. I had an off list conversation with Lyle
> and got
> my answer, Wayne was included too.
>
> The scenario was the morning 75M 'old guy' net. During the lulls,
> there was
> a ~15 db above the -110 db noise floor signal (I LOVE the P3!).
> Auto notch
> would not take it completely out until I lowered AGC THR to 7 (from
> 16).
>
> I restarted, cycled through auto and manual notch, no joy with auto
> until I
> dropped the THR.
>
> If I understand it correctly, as you raise THR and/or SLP, it gets
> harder
> for the auto notch to remove weaker signals (which this apparently
> is).
>
> So now my goal is to find a reasonable compromise or define two
> macros to
> make simple and quick changes, one for ragchewing and another for
> DXing.
> But first I want to better understand what THR and SLP do and how they
> interact. (NOT tonight!)
>
> Thanks again all (grazie e ciao!),
>
> Rick wa6nhc
>
> -----Original Message-----
> From: Pierfrancesco Caci
>
>
> It nicely notches away a broadcasting carrier at 7200 kHz here, tried
> with THR 12, 16 and 20.
>
> Pf
>
> ______________________________________________________________
> 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

yo2liw

Re: New K3 Firmware: Improved AGC (great for pile-ups as well as general operation)

In reply to this post by Thomas Horsten

Hello

couple of my observations:

I also noticed that APF action it's somehow softer than it was. Due to my
modest antenna conditions APF was a tremendous help in working some really
weak signals dxpeditions. Now it's action it is not the same, it helps but
not that much like on the older firmware when actually without the APF it
was impossible to copy something.

Now I use AGC:
THR is on 14
SLP is on 10

SLP bellow 10 act somehow like AGC OFF, for every quite strong signal I
have to turn down the AF gain so it was quite uncomfortable until I found
SLP=10 was a good value, at least for the moment, hi.

CWT it seems it's not so sensitive using the same settings like in the old
version.

Autonotch is working fine, Noise reduction, even on F1-1 (on CW) introduce
a small "delay" like muting the signal for 1-2seconds but I guess it was
the same with the older firmware.

The new K3 Utility software has a small minus from my point of view:

working CW with the K3 Utility 1.4.4.25 it was easy to use memories on the
terminal tab because I set the button on transmit and just clicking the
callsign box(memory) until it was copied by Dx station; now on version
1.12.3.28 the button doesn't stay on transmit it goes on stop very fast
requiring an additional click on transmit every time after
callsign/control, etc. Not very practical when working DX in this manner.

Overall the AGC improvement is obvious.

73! Adi
Yo2liw
K3#791

On Mon, May 14, 2012 at 12:31 PM, Thomas Horsten <[hidden email]> wrote:

> Wayne,
>
> It's hard for me to do a before/after side by side comparison since I only
> have one K3 and the band conditions change all the time, but after using
> the 4.51 firmware for a couple of days, it seems to me that APF is maybe
> slightly less effective than it was before. Maybe it's because the overall
> clarity of the weak signals is generally improved, or else the AGC changes
> somehow interact with the APF in a detrimental way?
>
> I'm generally using the 400Hz filter for CW and adjust the passband down to
> about 250Hz for best readability on weak signals, engaging APF at that
> setting previously made most weak signals much louder, now it seems to make
> that much difference only on a smaller subset of signals and only when the
> band is relatively quiet.
>
> It could well be down to random band condition changes before and after the
> upgrade, but I thought I'd mention it anyway.
>
> 73, Thomas M0TRN
>
>