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 bugs
|
As I play more on the air with the QMX, I'm enjoying it more and more. Multi bands is great in a rig this size. I have been keeping track of little bugs. I am mainly a CW guy so my observation will have to do with CW.? Hans do with what you like:
?
1. Once you do a swap (A<>B) it doesn't seem to matter what else you do (push other buttons, twist knobs) the next time you single click the second button it sets the vfo's equal. Like it's never timing out of the long press/short press sequence.
?
2. Changing bands does not remember the last freq on previous bands Seems to restore from configuration. (Power off and power on does remember the freq last used on the current band.)
?
3. Single click on second button. I've mentioned this before but changing VFO mode (normal/split) should only have two states, i.e. normal and split. Having an extra swap (two normals) to get to split makes no sense at all. A single click on the second button should toggle VFO mode from normal to split and back. The extra state in there is redundant. Long press give swap any time it's needed.
?
4. If you have the 19m (or 21m too I suppose) then changing bands by double clicking the left encoder give strange results when at 20m. It does not seem to like moving back to 80m. I did a factory reset to fix this. Probably could have just changed the band config as well.
?
I'll be slowly checking out other features and will list them as I find issues.
?
-mike/w1mt
|
|
Hi Mike, all Thanks, very useful; some of these are known to me, others are new.? I made great progress on the firmware last week and should be expecting to release a new version this week which fixes a lot of things. I got stuck for an entire day on a nasty I2C issue, reading the EEPROM, which locked up the I2C bus and could only be resolved by a power cycle; I was very happy to find the cause, and be able to resolve it by re-writing the EEPROM process as interrupt driven so the timing could not be interrupted?by any other processing. All great now!? Until I get the next firmware version going - I have, initially, been testing the assembled QMX myself rather than passing it over to my colleague to handle the testing in her lab; I observe the following, on about the last 10 assembled QMX I looked at, which will be useful to others powering up their QMX for the first time: 1) On first power up, as expected (since the application firmware isn't loaded), the unit goes into bootloader mode and appears as a USB Flash drive on the connected PC. I copy in the firmware file. It reboots however it does not reboot properly; so I have to cycle the power to get it to reboot properly after that first firmware load.? 2) A proportion of the time, more than half, for some reason on first power up, the "Factory Reset" doesn't happen. Then the display has junk frequencies and everything is no use. So you have to do a factory reset (from the buttons/rotary encoder or the terminal) to get things sorted out.? 3) All the time, the Tx/Rx Param 1 and 2 are both set to zero, rather than 0 and 1 which is what they are supposed to be. Your sidetone will CLICK horribly. I should say BOOM. So you must go into the Band Configuration screen on the terminal, and manually set that Tx/Rx Param 2 to "1". On the next firmware version I will remove these parameters. I had thought that it would be a delicate optimization matter and the parameter values would need to be adjusted per band by experiment; however this does NOT appear to be the case, 0 and 1 seem to work beautifully ALL the time; so there is no need for these to be a configuration parameter. 4) Just an observation: of course, the output power varies a bit by band. All these QMX were assembled with the 20m LPF correctly placed as per the latest assembly manual. I did NOT note particularly low power on 20m. Though of course, the output power varies a bit by band, and from unit to unit.? 5) On all the units I had, the 29?+ 9 turn L401 (as per current manual), the turns had to be fairly much fully spread out, to get the 20, 30 and 40m BPF peaks correct. It may even be worth using 29?+ 8 turns for L401. We'll see.? 6) Watch out for shorts between the positive (rear) pin of the 2.1mm barrel power connector, and the body of the rotary encoder; they are in very close proximity. It can be a good idea to put a bit of insulating tape on the back of the power connector. I only had a short on one of the units I checked but it made an alarming spark; of course I thought something else was wrong and the unit had fried but this was not the case, it's just a direct?+12V to Ground short at the input and nothing has touched the rest of the circuit. Note that the metal body of the rotary encoder is NOT grounded, except if: a) the metal happens to touch the metal enclosure and make electrical connection through the anodized coating OR b) when soldering the ground pin of the two pins (for the rotary encoder shaft button), which requires plenty of heat because of the 6-layer board thermally conducting heat away, perhaps you made a solder bridge to the encoder body. Which is no big problem UNLESS you have the short to the power connector.? I also have a couple of units built by my colleague, which did NOT work... so when I have time I will investigate those more thoroughly and start writing up a troubleshooting guide as on the other QRP Labs transceiver web pages.? 73 Hans G0UPL On Sun, Jul 16, 2023 at 3:26?PM mike/w1mt <w1mt.qrp@...> wrote:
|
|
Comments on Hans' comments on QMX: 5: Did you mean 19 + 8? ? I used 17 squeezed + 9 squeezed.?
|
¿ªÔÆÌåÓýIf I can add one more bug to the list, please.On the QMX the RF Sweep usually works after a hard reset, but after that it will hand on the 40M band. I can run the other bands, but not 40M. Not a big deal, but hopefully this is already on the list. This is with Firmware 002 or 003. 73, Cliff, AE5ZA
|
|
On Mon, Jul 17, 2023 at 02:41 AM, Hans Summers wrote:
5) On all the units I had, the 29?+ 9 turn L401 (as per current manual), the turns had to be fairly much fully spread out, to get the 20, 30 and 40m BPF peaks correct. It may even be worth using 29?+ 8 turns for L401. We'll see.?Hi Hans. This is great information to have.? Thank you.? One question, however.? The photo in the manual shows the 9 turns to be spread out but the 19 turns to be not so spread out.? Do you mean that you had to spread out both sections?? I found that when I spread out the 19 turns, the RF sweeps got worse (but I was battling a second simultaneous issue at the time).? Just to be clear, do you mean to spread out both or just the 9 turns? 73, Mike KK7ER |
|
Hi Mike? 5) On all the units I had, the 29?+ 9 turn L401 (as per current manual), the turns had to be fairly much fully spread out, to get the 20, 30 and 40m BPF peaks correct. It may even be worth using 29?+ 8 turns for L401. We'll see.?Hi Hans. This is great information to have.? Thank you.? One question, however.? The photo in the manual shows the 9 turns to be spread out but the 19 turns to be not so spread out.? Do you mean that you had to spread out both sections?? I found that when I spread out the 19 turns, the RF sweeps got worse (but I was battling a second simultaneous issue at the time).? Just to be clear, do you mean to spread out both or just the 9 turns? I had to spread out the 9 turns a lot, I actually removed one turn; the 19 turns weren't fully spread out, I adjusted them to put the peak on the 20m band. I think everyone has a different winding style and tightness, and there are component tolerances too.? 73 Hans G0UPL |
|
Here's another bug...
Single left button click to keyer speed mode. If you click the right encoder once the screen with take to you out of keyer speed mode but left encoder and neither button work any more. Can only turn off by removing power. Workaround: use the right button to exit keyer mode and not the right encoder. -mike/w1mt |
|
Hi Mike ? Single left button click to keyer speed mode. If you click the right encoder once the screen with take to you out of keyer speed mode but left encoder and neither button work any more. Can only turn off by removing power. Thanks, I found the cause and fixed it. This bug is in QCX too! For a little under 6 years, and 20,000+ QCX-series kits, nobody?has noticed... how about we keep it quiet? ?:-D 73 Hans G0UPL |
|
On Tue, Aug 1, 2023 at 08:33 AM, Hans Summers wrote:
I noticed yesterday evening on my QCX mini. 73 Adrian yo3hjv |
|
Still in 1_00_009
Hold right encoder down long to swap A<->B. This works but now try to change VFO with a short click of right encoder and it does not work. Instead it transfers A to B (Combo long/short should do this) It's still remembering the long press for too long. Turn rit on. Turn decode RX on. Watch screen scramble.... Needs better boundaries for decode writing when Rit is on. -mike/w1mt |
|
Hans, I encountered the dreaded left encoder/power jack short. My encoder was definitely grounded. Did you know the ground pin on the control board makes a great fuse? Lol
I desoldered the power jack and pushed out as far as possible. I also filed the corner of the encoder. Lastly, I stuffed some cardboard between the encoder and the buck converter socket. These steps seem to provide a suitable gap.? I must have locked out with L401. I seem to get 4.5 to 5 watts on every band.? I will pray and light votive candles until the AGC arrives. I just had a 30 over call come into my headphones. I haven¡¯t felt like this since my last Black Sabbath concert.? |
F Adrian
K9NUD-Steve
to navigate to use esc to dismiss