Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
QMX - Any progress on key clicks please?
|
I just came up with a possible theory for the key clicks.? My theory is that the output op amp, an NE5532, may be going into slew rate limiting under certain conditions.? The T.I. data sheet for the NE5532 under Features on the first page indicates that the slew rate is?9 V/
¦Ìs Typ. That's pretty fast. In paragraph 7.6, it states that the unity gain slew rate is 9V/uS still consistent with the first statement.? Paragraph 8.33 slew rate says that the slew rate is 9V/m Sec. That's pretty slow? Note that is says m Sec or milliseconds, not 9V/ uS.? I don't know which is right.??
At any rate, I recall that in an older software version before AGC, the rise and fall time of the audio tone was 5 milliseconds, same as the rise and fall rate of the RF output.? In that version, I did not hear any clicks on start and stop of a dot or dash, only a thump at the end of a dot or dash.? The rise time and fall time of the audio tone was reduced (I forget the value) but it was much smaller to make the tone sound more snappy.? If this is the problem, it should be testable with a special software version to go back to 5 milliseconds or perhaps a bit slower. The clicks appear to get more noticeable at high amplitudes, again indication a possible slew rate problem. In addition, other data sheet specs were for a load of 600 ohms.? The data sheet doesn't specify how low an impedance you can use, but headphones have a relatively low impedance which may have an effect on the closed loop frequency response of the op amp.? I know it can put screamingly loud audio to headphones, but this is separate from any slew rate or potential stability issues. -Steve K1RF? ------ Original Message ------
From "Axel Kohlgrueber" <axelkohlgrueber@...>
Date 3/15/2024 12:44:52 PM
Subject Re: [QRPLabs] QMX - Any progress on key clicks please? QMX mp3 |
|
Re: slew rate,? The Onsemi datasheet for NE5532 is self consistent showing a 9V/uS? slew rate and a large power BW much greater than this application requires. I think the TI datasheet may have a typo l. Also, slew rate limiting is easily observed on a scope. Someone should have spotted it and posted a screenshot by now. Net: I think the problem is elsewhere. JZ KJ4A? On Fri, Mar 15, 2024, 6:31?PM Steven Dick, K1RF <sbdick@...> wrote:
|
|
Thanks for checking the data sheet John. There's a big clue though in the fact that the glitch occurs both at the start and the end of a tone. More info could possibly be obtained by looking at the glitches on a scope to gain more insight about their nature. Under a set of conditions that produce the glitches, It might be a good experiment to slow down the rise and fall times of the sidetone and see if the glitches are affected.? Another admittedly remote possibly is the data rate of the serial data going into IC401, a CS4334 being too slow under conditions of high data rate digitized data being under-sampled at high speed audio amplitude changes.
-Steve K1RF ------ Original Message ------
From "John Z" <jdzbrozek@...>
Date 3/16/2024 12:00:35 PM
Subject Re: [QRPLabs] QMX - Any progress on key clicks please?
|
|
Glad to, Steve! Yes, I would really like to see some scope traces! In particular, I would like to see the stereo DAC outputs. That is sort of a fold-and-tear-here point in that if the signal looks aberrant at that point the post-DAC amplifiers are absolved of guilt and the problem is upstream. I am building up a QMX now. It is a fun project! Having one available to probe and test will make me a better helper here on the forum! 73 JZ KJ4A? On Sat, Mar 16, 2024, 12:20?PM Steven Dick, K1RF <sbdick@...> wrote:
|
|
I placed an order for a QMX 5 months ago and, based on my position in the waiting list, I still have 2-3 months left to wait. Having read all the posts about key clicks and thumps I am concerned I may have made a big mistake. I am no longer looking forward to receiving my unit.? I hope the situation gets resolved before I get to the top of the waiting list otherwise, like others have posted, my QMX will be going straight to the junk drawer. It is time for Hans to put our minds at ease and tell us what he is going to do about this problem. John Corby, VA3KOT |
|
Just stumbled across this on YouTube comapring the QMX with SW3b - interesting to hear the key clicks on QMX against the purer sound of the SW3b. Of course, as a number of us have pointied out, it seems worse on headphones, presumably due to less ambient noise, as opposed to the problem actually getting worse. Worth a quick look. I think my QMX clicks are slightly worse than the one on the video, but again, that may be down to using headphones.
Daimon. G4USI |
|
Hi John, all > It is time for Hans to put our > minds at ease and tell us what > he is going to do about this > problem. Obviously... I'm going to FIX IT! It is now my highest priority item.? Over the last few months there have been several high profile issues to solve, starting with lack of sidetone volume, then the CW keyer glitches, and then lack of AGC. Each drew some critics ("my QMX is thrown in a drawer"). I solved them one by one and I shall this one too.? I would remind everyone that right at the top of the QMX shop page, in bold red font on a yellow background (not hidden in some small print at the bottom!) is a notice: "Beta firmware supports CW and Digi (all functionality of both QCX-mini and QDX)." In other words it should be clear that firmware is still beta.? There will no doubt be some people who would insist that I am gravely mistaken and I should not have released QMX until the firmware had reached a greater state of perfection. However, I've been making QRP Labs kits since 2010 and I believe I know enough about you guys, the kit constructors, to make a judgement; in this case it was my opinion that many of you would prefer to have such an exciting kit as QMX in your hands earlier despite the firmware defects, than wait in the dark. QMX already offers a huge and unbeaten level of functionality and performance for the price.? In other words I think the majority of you prefer to be along for the journey, then cherish and enjoy all the more, the fine taste of the eventual end result.? Another reason is knowing myself... QMX is ambitious and complex to develop. By releasing it for sale as soon as the hardware was ready, even if the firmware still needed further work, I pushed my own boundaries to make it actually happen. It's very easy otherwise to get stuck in an endless development rut that drains the life out of you, then you end up taking a lot longer and a lot less happily. The QMX way - many of us are already enjoying this great transceiver and the momentum grew quickly, it'll keep pushing forward until the firmware releases resolve all current issues and further.? 73 Hans G0UPL On Sat, Mar 16, 2024, 7:50?PM John VA3KOT <va3kot@...> wrote:
|
|
Many thanks for the update Hans
Actually for me the key clunks are better in 17 than they have been from 14 onwards so far and I have not played around the AGC settings much. Just to add an observation - it is worse on some bands than others.? 20m has been pretty much unusable oftentimes, 30m - barely any issues at all. I at least have faith in you fixing it :).? And yes, I love my QMX.? Need to order another high band kit as I've ended up messing the last one up beyond the point of repair... 73 Danny M0SDB |
|
I use my QMX almost every day (I'm cw only).? The thumps and clicks don't bother me very much.? I do use an external powered speaker whether on the bench or in the field. I guess the filter between my ears helps too.? Thump, clicks or not, the cw is still there. No worries here Hans.? I love mine.? Thank you!
Ken, W4KAC |
|
And one more thing, Hans! I am always anxiously waiting for the next upgrade to see this little gem performing better. Sometimes compulsively checking this group to see if a new upgrade is released. If QMX would have been perfect from day one, all this ?excitement would have not existed.
I bought a QCX 20m band in August 2022 (arrived a few months later) when I did not know CW. But I wanted to have that little jewel in front of me to remind me of my commitment with devoting daily time to learning CW. So QCX has been my companion during this journey and when QMX was launched, that was the logical follow up! Multiband cw and digimodes!! That was even more reason to push keep learning cw. And here I am!! Only doing SOTAs, with no base station yet, but enjoying infinitely that small footprint yet powerful (not necessarily in terms of watts ?) rig!! Thanks Hans for your great job!! Joan - EA3IKB |
|
Bravo, Hans! So many of us are having a great experience building and getting on the air with QMX that it validates your approach. We QMX users get to participate in refining and optimizing the radio.? Even if some of us have scant electronics background we can contribute by asking questions and reporting observations. I appreciate your strenuous effort to make QMX ownership and use a learning and rewarding? experience.?? Thanks for your reassurance message that may help anyone who is frustrated to realize this is some advanced amateur radio that can be mastered but does not follow a predictable timeline. Charles K5KXJ |
Good day Hans.
The right lines at the right time.
I think I speak for most people when I say that 99% of the reports here |
|
Thanks Hans,
As the person starting this particular thread, I am reassured that: a) We have now identified this is a widespread problem, for many of us making the rig unusable on CW. b) It has now been brought to your attention, whereas before I don't think you were fully aware of the extent of the problem c) We have your reassurance is is #1 priority. That works for me! Many thanks Daimon. G4USI |
|
Me too! Mike Krieger On Sun, Mar 17, 2024 at 2:23?AM Axel Kohlgrueber <axelkohlgrueber@...> wrote: Good day Hans. The right lines at the right time. I think I speak for most people when I say that 99% of the reports here |
|
We are with you Hans!
We all know what we are signing up for, or at least we should, when we take part in QRP-Labs. Yes you have helpers in the business, but from an engineering standpoint you are a one man band. Furthermore, you are not resting your back on 25+ year old technology. If I want a multi band, multi mode, extremely modern radio that fits in the palm of my hand, backed by a phalanx of engineers and a highly refined manufacturing process then I¡¯ll be paying easily 10x the cost of a QMX. QRP-Labs is not Icom or Yaesu, or even Elecraft, and it¡¯s a mistake to categorize it as such. 72 and peace to all. |
|
You are so right. Guys, you paid $100.00. Yes Hans we are with you.? Mike Krieger On Sun, Mar 17, 2024 at 9:25?AM <w6csn@...> wrote: We are with you Hans! |
|
Hans,
Excellent ongoing work by you and your team. You are an asset to the Ham Community in the area of innovation and QRP kits. Keep up the good work. Some of the "detractors" ought to try doing what you do. I used to supply kits for Hams for 20 years just because I liked providing educational building opportunities (like in the old days) and I thought/know others charged too much. it was FAR from simple and easy, yet I regret none of it. 73 Kees K5BCQ (slowing down a bit after 80) |
|
A little more info on the problem beginning from the settings I previously posted. I simulated the audio output stage in LTSpice including a lumped element headphone model using a tone which was ramped up and down where I could change the rise and fall times of the tone.? I did not see any problems in the simulation with either a 600 ohm load or nominal headphones.? Doing a little more playing, I tried using a bluetooth speaker with a line in and compared the clocis? to those I hear in my headphones.? I still hear the clicks on the speaker, though they have less highs than the headphones but are still there.? Then I made some setting adjustments.? I found that changing the parameter sidetone relative/absolute from absolute to relative.made a huge improvement in reducing the clicks for me.? I also reduced my sidetone volume from 83 down to 64.? ?With these changed settings, the clicks are at a very low level now. When I crank up the volume on a typical band, the sidetone sounds quite useable now. -Steve K1RF |
to navigate to use esc to dismiss