开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育

#qmx #firmware 1_00_017 AGC, S-meter, LPF sweep #qmx #firmware


 

Hans, FYI:

I just tried to set Paul's "good start" settings (post below), not through Terminal, but through the QMX menu. After finishing, selecting the right push-button to step out, my QMX locked-up with just "Audio" showing. No buttons worked including power off, so had to power off by pulling the plug.

On re-start (plugging back into power and starting), QMX came up just normally, and all settings I made were still present.

I was able to reproduce the issue (again, Menu settings via QMX, not terminal) by trying a few things. This is what did it in:

  • moving "Sample blocks" back up to 5 (your originally published settings to try)
  • backing all the way out of menu. (QMX working OK here)
  • returning to Menu -> Audio -> AGC Settings -> Sample blocks
  • reducing the Sample blocks to "2".
  • back -> back (to AGC Settings) -> back (to Audio)
  • QMX totally hung now
Something to do with reducing the value of Sample blocks appears to make QMX grumpy.

Roy - KI0ER


 

The DX contest is proving to be a great way to test the AGC.
And it is working quite nicely.? Thanks, Hans!
Just hopped on 20m and snagged two Spanish stations and a Hawaiian station.
Amazing what 4W and a 15ft dipole can do when the band is open.

73, Mike KK7ER


 

Hans,

Thanks so much for this, a great improvement on CW!? Paul, thanks for thes ettungs, I've been playing around with them a bit with QSOs on 30M and litening to the 'test on 40M.? Think I'll end up somewheer between youe settings and Hans LOL.? Owning an Elecraft K3 and K4 means I've done a lot of playing around with AGC settings!

73, Bill NZ0T


 

Very pleased with the new features.? The QMX is growing up now. Well done Hans!
One thing for my wish list would be a variabel filter setting for CW. I'm working a lot of guys using homebrewed tx or ww-2 gear. They are not spot on most of the time and VFO' s tend to drift. This morning I had a QSO? with a om who returned to my cq about 150 Hz lf and I first did'nt hear him.
Now I'm using RIT and tune around my calling frequency after a cq.

73 Durk PA3BYW?

Op za 17 feb 2024 03:09 schreef <nz0tham@...>:

Hans,

Thanks so much for this, a great improvement on CW!? Paul, thanks for thes ettungs, I've been playing around with them a bit with QSOs on 30M and litening to the 'test on 40M.? Think I'll end up somewheer between youe settings and Hans LOL.? Owning an Elecraft K3 and K4 means I've done a lot of playing around with AGC settings!

73, Bill NZ0T


 

Hi Roy

> Something to do with reducing
> the value of Sample blocks
> appears to make QMX grumpy.

Thanks... I'll fix that. No problem. It's just that pesky AGC sample buffer again... It's in use all the time audio is running. When you go in and change the size while it's in use things can get confused. You'll find probably it crashes sometimes not others, when you change sample size. I can fix it easily don't worry. For now, restarting QMX will recover it.

73 Hans G0UPL


On Sat, Feb 17, 2024, 4:50?AM Roy - KI0ER via <ki0er=[email protected]> wrote:
Hans, FYI:

I just tried to set Paul's "good start" settings (post below), not through Terminal, but through the QMX menu. After finishing, selecting the right push-button to step out, my QMX locked-up with just "Audio" showing. No buttons worked including power off, so had to power off by pulling the plug.

On re-start (plugging back into power and starting), QMX came up just normally, and all settings I made were still present.

I was able to reproduce the issue (again, Menu settings via QMX, not terminal) by trying a few things. This is what did it in:

  • moving "Sample blocks" back up to 5 (your originally published settings to try)
  • backing all the way out of menu. (QMX working OK here)
  • returning to Menu -> Audio -> AGC Settings -> Sample blocks
  • reducing the Sample blocks to "2".
  • back -> back (to AGC Settings) -> back (to Audio)
  • QMX totally hung now
Something to do with reducing the value of Sample blocks appears to make QMX grumpy.

Roy - KI0ER


 

I've listened to this weekend contest in 20 and 40m bands and tinkered with AGC settings some more. I've also tried Paul (W9AC) suggestion to increase "Smooth samples", and it seems to alleviate the clicks on TX.
My goal was to keep reasonable good dynamics and make RX as less tiresome as possible. Here's what worked for me:



It's nice to have so many options =) =)

Regards,
YO3GFH
op. Adrian


 

Thanks, Hans, this update (AGC) really makes a huge difference! ?As a feature request, could the S meter have a peak hold function? ?


 

On Sat, Feb 17, 2024 at 01:13 AM, Paul Christensen wrote:
See below.? The high "Smooth Samples" setting seems to eliminate much of the audio clicks.? The remaining settings give reasonably good dynamic range while limiting strong signals. I'm still experimenting with values but this is a good start.

Paul, W9AC
I have just tried this and smoothing makes no difference to the clicking for me.? It seems much more related to the AF gain with the clicking being very much reduced with the AF gain turned down and seems to be related to the clunking that is sometimes present when keying on older firmware versions.? With S9=S9 the clunking is very pronounced and makes keying barely possible.? With S9=S0 then this is reduced to the spark like sound and with the AF gain dialled down becomes perfectly manageable but present.? However I have not tried it on air to see if the incoming signal will be loud enough to hear.


 

开云体育

Danny,

?

I failed to mention that the high Smooth Samples setting related to receive only to minimize the background “frying” noise.? During Tx, I still hear some clicks, mostly on a crowded band like this weekend’s DX contest.?

?

Paul, W9AC

?

From: [email protected] <[email protected]> On Behalf Of Danny Bower
Sent: Saturday, February 17, 2024 6:17 AM
To: [email protected]
Subject: Re: [QRPLabs] #qmx #firmware 1_00_017 AGC, S-meter, LPF sweep

?

On Sat, Feb 17, 2024 at 01:13 AM, Paul Christensen wrote:

See below.? The high "Smooth Samples" setting seems to eliminate much of the audio clicks.? The remaining settings give reasonably good dynamic range while limiting strong signals. I'm still experimenting with values but this is a good start.

Paul, W9AC

I have just tried this and smoothing makes no difference to the clicking for me.? It seems much more related to the AF gain with the clicking being very much reduced with the AF gain turned down and seems to be related to the clunking that is sometimes present when keying on older firmware versions.? With S9=S9 the clunking is very pronounced and makes keying barely possible.? With S9=S0 then this is reduced to the spark like sound and with the AF gain dialled down becomes perfectly manageable but present.? However I have not tried it on air to see if the incoming signal will be loud enough to hear.


 

Hans,

Just a quick note: while changing "Audio gain" in band config. menu from the terminal, I observed that the S-meter and the AGC are heavily affected by it, i.e. if I set 0dB of "Audio gain", there's barely any S-meter or AGC action displayed on the screen.
This can be also be observed on "The Lab" (see what you have done? :-)) )

OTOH, if I change the "Audio atten." value, S-meter and AGC level seem to be unaffected by it.
Is this how it's supposed to be? I'm thinking that, at this momemnt, if you're changing "Audio gain" to anything other than the default value of 54, the S-meter is losing it's calibration. Or so it appears to me, I'm known to have been wrong before :-))

And one other thing, if I set the "Audio gain" to, say, 30dB on one band and 0dB on all other bands, the audio seems to honor the 30dB for all bands. The reverse is also true, If I set 0dB on one band and 30 on all the others, the 0dB seem to take over.

Regards,
YO3GFH
op. Adrian


 

Awesome work Hans, when I watched the YouTube video last night I thought wow is the QRPLabs group going to be going wild tomorrow!!
Thanks for your incredible commitment to satisfying your supporters!

John?
K7KTM?


 

Hello Adrian?

Just a quick note: while changing "Audio gain" in band config. menu from the terminal, I observed that the S-meter and the AGC are heavily affected by it, i.e. if I set 0dB of "Audio gain", there's barely any S-meter or AGC action displayed on the screen.
This can be also be observed on "The Lab" (see what you have done? :-)) )

OTOH, if I change the "Audio atten." value, S-meter and AGC level seem to be unaffected by it.
Is this how it's supposed to be? I'm thinking that, at this momemnt, if you're changing "Audio gain" to anything other than the default value of 54, the S-meter is losing it's calibration. Or so it appears to me, I'm known to have been wrong before :-))

Yes, you are correct. But this is also how I have declared that the system works. If you look in the operating manual section on the AGC I described all the stages the signal goes through from the antenna BNC port through to the 3.5mm earphone output. You'll see that I said the band configuration gain value comes before the AGC action. The audio attenuator and volume control gain come later. This is why they don't affect the S-meter. So the intention was that the band configuration gain can be used to equalize and calibrate, to cancel out any differences in BPF attenuation etc between the bands.?

On my unit a band configuration gain of 54 on 40m calibrates the S-meter correctly. Note that the S-meter is logarithmic with 6dB/bar. This doesn't depend on anything in band configuration, it's linearity as 6dB/bar is always accurate. But the number of bars for a given level such as S9 would vary and that's what needs calibration by adjusting the gain in the band configuration screen.

So what you observed is the correct behavior for how the system is written now. But if you think I should improve it somehow please let me know.?

And one other thing, if I set the "Audio gain" to, say, 30dB on one band and 0dB on all other bands, the audio seems to honor the 30dB for all bands. The reverse is also true, If I set 0dB on one band and 30 on all the others, the 0dB seem to take over

Yeah that really doesn't seem right, I should have to look into that.?

73 Hans G0UPL



 

On Sat, Feb 17, 2024 at 07:58 PM, Hans Summers wrote:
Yes, you are correct. But this is also how I have declared that the system works. If you look in the operating manual section on the AGC I described all the stages the signal goes through from the antenna BNC port through to the 3.5mm earphone output. You'll see that I said the band configuration gain value comes before the AGC action. The audio attenuator and volume control gain come later. This is why they don't affect the S-meter. So the intention was that the band configuration gain can be used to equalize and calibrate, to cancel out any differences in BPF attenuation etc between the bands.?
?
On my unit a band configuration gain of 54 on 40m calibrates the S-meter correctly. Note that the S-meter is logarithmic with 6dB/bar. This doesn't depend on anything in band configuration, it's linearity as 6dB/bar is always accurate. But the number of bars for a given level such as S9 would vary and that's what needs calibration by adjusting the gain in the band configuration screen.
Thanks, it's clear to me now. I read the AGC section in the manual (very nice, btw) but I was under the (wrong) impression that the system somehow would compensate for the audio gain setting in order to keep the calibration. Reading you post, I think you have a good point, with a sig. gen. one can manually set gain per band and end up with a nice S-meter all across, unlike some top brands which shall remain unamed :-)

So thanks again for lifting the fog :-)

Regards,
YO3GFH
op. Adrian


 

Wow, superb, thanks Hans.

73 Simon
G0FCU.


 

AGC is really nice addition! Been playing with the settings - here are my current ones:
Thresh 9
Slope? 40
Noise? 10
Hang? ?30
Smooth 50
Recovery 40
Blocks? 4
S9? ? ? ? ?9
AGC? ? ?ON
dB/Bar? 6

Can't wait to take it on an outing!
--
73, Dan? NM3A


 

Thank you. I just blew out the votive candle I had lit since finishing my kit!
--
72/73 de Steve, K9NUD


 

Target: QMX firmware 1_00_017 (second release)

Bug: QMX display and control and serial console frees when using VOL encoder when QMX is in menu.?

Reproduction steps: entering menu on QMX, select any item (for example CW Keyer). Then use serial console and go to diagnostics screens for testing the encoders/buttons, all work but when turning the VOL encoder, diagnostics screens stops working, also controls / display on QMX stop working (frees state).

Fix: Pressing Ctrl-Q in serial console, serial console jumps back to Hardware test menu menu and controls/display on QMX are working again. Leave menu on QMX and then diagnostics for VOL encoder doesn't frees (don't be in a QMX menu when using diagnostics).

--

73 Lex PH2LB?


 

开云体育

I believe this is "Works as Designed." (If I recall correctly...)

73, Paul -- AI7JR

On 2/18/24 08:59, Lex PH2LB wrote:

Target: QMX firmware 1_00_017 (second release)

Bug: QMX display and control and serial console frees when using VOL encoder when QMX is in menu.?

Reproduction steps: entering menu on QMX, select any item (for example CW Keyer). Then use serial console and go to diagnostics screens for testing the encoders/buttons, all work but when turning the VOL encoder, diagnostics screens stops working, also controls / display on QMX stop working (frees state).

Fix: Pressing Ctrl-Q in serial console, serial console jumps back to Hardware test menu menu and controls/display on QMX are working again. Leave menu on QMX and then diagnostics for VOL encoder doesn't frees (don't be in a QMX menu when using diagnostics).

--

73 Lex PH2LB?

-- 
Paul -- AI7JR


 

I installed the new firmware with no issues. Unfortunately I could not give it a try, yet. Can't wait!!?

If I recall correctly, Hans said, a few days ago, that the priorities were:

1. Fix the longer dits issue. Done!
2. Implement AGC. Done!
3. SSB mode... ??

Not meant to do any pressure on you Hans... ?hihi

Can't wait for that either!!

Best,

Joan - EA3IKB


 

Thanks Hans. Fantastic update - again!
The AGC works really well after a bit of tweaking to get rid of the rice krispies and get the required amount of squish and release. So many parameters to play with!? :)

I tried the radio out from home (near Edinburgh, Scotland) this evening - running into an EFHW cut for 40m, up about 7m. Calling CQ on 20m brought in 6 contest stations from the US within 10 minutes! Not bad at all. The radio, and its AGC performed admirably in the crowded band. Now to try it out on the hills...

Cheers,
Ronan
MM0IVR