Sticky
#qmx #Firmware release 1_02_000
64
#qmx
#firmware
Hi all SSB firmware was released into beta on 17-Mar-2025 after 9+ months of development. Many thanks to everyone who bravely helped test it and patiently provided their feedback and observations. There are now quite a few YouTube videos, forum reports and other social media mentions about on-air successes with QMX SSB. Now that it has been at least 6 weeks in beta and I believe most of the issues have been resolved, it's time to promote it to an official version. I have called this version 1_02_000 and published it on the QMX page http://qrp-labs.com/qmx where I have also published the updated operating manual, which grew from 111 pages to 134 pages. The main changes are: SSB (obviously!) including CESSB, TX Equalization, Mic AGC, VOX, Pre-distortion and and and... Addition of up to two more USB Virtual COM serial ports (up to three total). All of the hardware tests such as RF and SWR sweep, and Diagnostics, are now available on the LCD too. New CAT commands for LCD contents, S-meter, AGC-meter, Power meter, SWR meter, SSB source Complete list (relative to 1_00_027 the previous official version): SSB modes added (USB and LCB), with CESSB, TX Equalization, Mic AGC, VOX, Phase pre-distortion and more Rationalization of menus, grouped into CW, Digi and SSB submenus (now Keyer and Decoder are submenus of CW) Added up to two more USB Virtual COM serial ports (configurable) LCD versions of AF, RF, Image, LPF, Test ADC I/Q sweeps (miniature graph on LCD) New function PA mod. test to check operation of amplitude modulator Diagnostics screens are now available on the LCD interface too Now remembers previous menu location on ALL menus regardless of nesting level Add way of forcing factory reset, press Menu (left) button during power-up Change TX color in diagnostics screen from red to blue, to avoid confusion since red things are errors Added CAT LC command to retrieve LCD contents Added CAT commands SA (AGC meter), SM (S meter), PC (Power meter), SW (SWR meter) Added CAT command SS (get/set SSB input source) Bug fix: PTT P did not work properly Bug fix: Incorrect PA voltage display on the Diagnostics screen 73 Hans G0UPL http://qrp-labs.com https://www.buymeacoffee.com/g0upl
|
QMX+ issues
My QMX+ was working but when I powered it on this morning the display backlight is not on and the display shows a row of black blocks. Does anyone have any thoughts on where to start troubleshooting. +V or +12v has 12v, VCC has 5.1v, and VDD has 3.3v. I have had some issues with this unit. The 78M33 failed. When I tried to remove it the two small pads lifted. I currently have the 3.3v being supplied by a small DROC 3.3v buck converter. It was working well as of yesterday, I completed several FT8 and SSB QSOs.
|
#qmx plus assembly one LP and Band filter at the time
6
#qmx
Hi Everyone I was wondering if is possible to assemble only one band filter at the time on a qmx plux rev3, this means I have assembled the all QMX leaving out only the toroids of all the filters, the LP and BP ones. Now next I have assembled only the 160m LP and BP and test the filters. The swipes look normal on the 160m. Next is to test transiting to a dummy load only this band, but I get the swr protection kicking in and no power out. voltages look normal. running a swr plot on 160m shows swr 10 on all graph. The main idea is to tune filters one by one with space for the soldering iron making it easy. I have reviewed soldering of transformer and swr brigde plus the jumper to the bnc is in place. My question is, Is it expectable that the SWR is High when only the filters for one band are on the board or this means something else is already wrong from the beginning. Thank you a lot 73 Azgael
|
Solder pot for toroid leads
3
Does anyone use Solder pots for tinning and removing the enamel from the toroid leads? Any hints and tips? I see a cheap one on Amazon for about $20. Any suggestions if I should buy special solder for this, or use my normal Kester 63/37 solder in it?
|
QMX PLUS
5
Hi all I have a qmx plus and I ran 13.6 volts into it on a first time power up it worked for abt 5 min then it shut dow. Then i read about the max of 12v vcc. I am assuming that I lost the buck converter. I have replaced all the sbject transitors that showed bad via the ohm meter still no joy. does anyone of you smart ones have test point diagram fot the buck converter so i can trace it and trouble shoot . thanks you a bunch Joe
|
QMX+ RF output fluctuates wildly
22
#QMXplus
#troubleshooting
Assembled my QMX+ and it's working. I had a couple of issues related to enameled wire not making good connections that were resolved. I know it's working, I've made contacts on several bands successfully, including some decent DX, all on FT8. What bothers me is that the output power seems to jump around wildly. If I go to the diagnostics screen and key the transmitter, the Bias SMPS shows fluctuations in current that correspond to the fluctuations in the output power. The Duty cycle on the Bias SMPS always shows 0%, even as the current jumps around in the red. I suspect I've managed to damage one (or more) of the surface mount semiconductors. At least that's what my reading on the subject would seem to suggest. What I need is a test that can be done with a multi-meter that would shed some light on the problem. I can probably resurrect an oscilloscope if I need it, but I'm hoping someone can lead me to a simple voltage or resistance test that could help isolate the problem. My last problem was related to soldering and enameled wire which was easily resolved. I'm hoping this issue can be resolved without major surgery as well. Thanks in advance for all of the good advice I'm about to receive. Andy WB9YXA
|
R520 Wrong component value.
When testing my newly built QMX+ I was getting high SWR readings while running the diagnostics SWR test. Even when connected to a dummy load the readings were around 2.7 to 1 across all bands. I removed and re-wound T507 (twice) but could not resolve the problem. So I removed T507 for a third time and started to check component values around the bridge circuit. I found that R520, although marked as 68X (49.9 R) , actually measured as 124.5 K. Has anyone else seen this problem. Regards Stephen
|
New QMX+ Calibration victim
17
Hi, I¡¯ve had my QMX+ for a number of months and had tested the then beta now regular SSB firmware with good results except pretty low output power in SSB. As I had never gotten around to fine tune the LPFs when building it I decided this was the logical next thing. So I fiddled with spreading and compressing turns on the toroids and managed to get all bands to between 4 and 5w output for CW except 160m that was already at 6w. Running 12.0v input and build for 12v Being happy with results I then first tested resetting to factory defaults, and then tested SSB output power. Still max 2-2.5w according to the Bird. I then decided to run the SSB calibration again. Worked fine til perhaps 12m or 10m phase. I then heard some terrible noise from the speaker and saw the QMX crashing with all blocks in the display and turning itself off. I tried powering on again, but won¡¯t start now. Just blank display and current protection on PSU kicks in due to over current. I suspect maybe finals are fried with internal short as result? Jonas 7 maj 2025 kl. 23:12 skrev Stan Dye via groups.io <standye@...>: ? On Wed, May 7, 2025 at 01:38 PM, Wiggle Pig wrote: It¡¯s D503 and D504 These are placeholders for diodes that could be placed across the BS170 outputs, as was commonly done with QDX. Hans included them in the design, but has not encouraged installing the diodes, which he believes are unnecessary for QMX. They may suppress spikes, but they also add additional capacitance to the PA, which may affect it adversely in other ways. There were many posts on this list regarding this topic a year or so ago. Stan KC7XE
|
50 Watt Power Amplifier - Low Power, High Harmonics
5
#50w
#50wpa
Hello. I'm assembling a 50 watt power amplifier for use with a QCX+ that I assembled last month, for the 80 metre band. I have very high harmonics at the output of T3 - as a result, I get only about 9.3 watts out when running at 13.8 V. Output from the LPF looks OK on the 'scope. I found the excellent videos by NA5Y which shows the output of T3 should be much closer to a sinusoid than what I get. I also found the discussion with KO1KR in August of 2020, but unfortunately I can't see his scope photos so I don't know if he had the same problem. I've since skimmed through the many discussions on this forum but haven't found the same issue exactly, though I did check out some of the suggestions given here. I found the amplifier took 67 mA with R5 fully CCW - I adjusted R5 till the amplifier drew about 70 mA with the QCX+ key-down in practice mode. Here are my scope traces: 0092 is the gate drive, yellow is Q1 gate, purple is Q2 gate - all is well here. 0090 is the drain voltage, Q1 in yellow and Q2 in purple - these look like there is oscillation ? 0091 is the voltage at the output of T3. This has a lot of third harmonic and doesn't look like the waveform NA5Y showed at this point. (QCX+ was set for 3.520 MHZ , amplifier connected to a dummy load.) I thought that perhaps my bench power supply is doing something odd when the amp is drawing power, but I probed the 12 volt line right at the barrel connector and the 'scope showed almost no disturbance of the +12 v supply. In frustration, I observed the output of T3 with the 'scope set for a long capture - it almost looks as if the distortion grows with time, for the first few tens of microsceconds the scope showed a small but sine-ish waveform, which over the span of a few hundred microseconds got wigglier and wigglier till it developed into the mess seen in my scope capture 0091. I've checked DC continuity to the windings of all the transformers and coils. I did find a bad joint on the primary of T1 but I cleaned it up and can now get power out of the amplifier. I haven't yet checked bias on all the 1N4007 diodes but the input looks quite sinusoidal at R3 closest to the input. The few watts the amplifier does produce look OK at its output. The QCX+ makes about 4.3 wats into a dummy load (43 V peak to peak) at 13.8 V DC. I'm assembling these units for a friend, and I had really hoped to have them ready for Field Day this year. The instructions are excellent and I have followed them closely. What can I look at next? Bill VE4STW
|
My 20M band QCX+
2
After a few days of work I completed my new QRP-Labs Xcvr after having built several years ago a Hi Band QDX. It was a real pleasure, no surprise due to an excellent manual and the adjustement and alignment were done today without dificulty. Next step will be to connect it to my Alexloop antenna. To be sure the 6 aft connectors whould slide nicely into the holes of the back plate I had screwed this plate to the lower enclosure and slid the main PCB each time I added a connector to check its position before finishing to solder it. The first difficulties I faced were to fit the LCD module and the front panel PCB to the Front panel of the enclosure : I had to cut slightly the nylon nut close to the potentiometer and I could not put in place the other nylon nut close to the encoder, I just left it near the hole in order to keep the good distance between the LCD board and the front PCB. The Last difficulties happened when I wanted to slid the main PCB inside the enclosure after having put in place the four 12MM spacers which hold the dev kit, the nylon screws below did not slide easily and I had to push quite strongly the PCB inside the enclosure. WELL I'M HAPPY WITH THIS WONDERFUL KIT AND I'M GRATEFUL TO HANS FOR THIS NICE DESIGN.
|
Digikey order
12
I need to order some parts from Digikey. I though it would be good to order some backup parts for the QMX+. Besides BS170's what are the components that fail most frequently? Also if you have specific digikey part numbers or names that would be great, ss some of the part names match multiple items.
|
QMX 1n4007 diodes
9
I needed to replace one of the 1n4007 diodes in my QMX (I cut the lead for an experiment) so I went looking for a replacement. None of the 1n4007 diodes at Mouser have the small diameter leads like the ones from QRP Labs. They are all about .7mm diameter where the QRPLabs diodes are more like .3mm. I ended up ordering Vishay diodes with the fat leads. I had cleared the holes with the solder sucker but the leads still did not fit so I used some PCB drill bits to clear the holes and finally got the diode in there. Hopefully the via barrel is still intact. I am waiting for some caps to put it back together and try it. Any discussion about these diodes? My experience is that 1N400x diodes all had fat leads. Thanks, Tony AD0VC
|
Why no AGC for digital modes, as other modes?
6
To Da Wiz; thank you, and a request, to make it better. Personally, I would appreciate the choice to use AGC, with my QMX+, or not, for all modes. I have used fast or slow AGC with my FT-990 and TS-790A for years, for different modes, and have found no noticeable performance penalty, except that the sound from no AGC was harsh and undesirable, and the S meter unusable. Some folks will not RTFMs to turn off the AGC, or properly tweak other software or hardware, and complain about imagined "bugs" - but for this ol' geezer, the choice would be appreciated, and I promise I will not whine about the consequences ;-) Thanks for a hearing, DA Ol' Dog, wa4hei
|
QMX+ firmware feature request
I'll start by saying I think if this as a "nice to have" change: I use a manual antenna tuner and the built in SWR bridge is great for checking the state-of-tune. In a (my) "perfect world", the "Tune SWR" would present a full width, bottom line, bar graph of the VSWR, rather than a single number. It's hard (for me) to spin the tuning knobs looking a flickering numbers. Just a thought. Thank you Hans for producing such a great radio! -- 73, Brent DeWitt, AB1LF Milford, MA
|
Release of version 2.0 of ABS solution for QMX+ (Automatic internal antenna tuner, internal battery and internal speaker with audio amp)
30
#PSU
#QMXplus
#audio
#ABS
#ATU
Hi, I have released a version 2 of ABS solution. Please refer to the readme front page file at the git repository for full details and release notes: https://github.com/AC8L/QMX_Plus_Autotuner_Internal_Battery_Speaker/ The release includes new PCB fabrication files, BOM and detailed build instructions. This release took into consideration lessons learned and a feedback from first group buy participants fearlessly organized by Cal (AD8Q) and Stan (KC7XE). -- 73, Sardar (AC8L, 4K6SA, VA3DUA).
|
OoT-RADIO HIFI TAMAGUCHI
2
Hello there, I found something amazing is possible. We need intelligent friends and we want to speak with them on many different topics. Do you remember that Tamaguchi baby robots need 24 hours of attention ? I found that it's possible to train large language models lot smaller than chat gpt , free , open source and expandable and than load that model in to 90 dollars worth of raspberry pi 5 and add to that chip an oled touch sensitive screen and a ram and than you have a box you can carry around and chat with it I am thinking of training and expanding the mini llama llm model 1GB- or a 7GB one with stereophile magazines - around a 60 years collection - and with some large amateur radio magazines collection , arrl books in 80 years - not all of them , only I can find free pdf - and electronics , physics books. I can reach hundreds of them But I have two ways to walk. The first one is to translate every book and magazine into Turkish and train the mini llama OR other LLM with 800 000 pages of Turkish and the same amount of English data. It would take 2 weeks to translate the data into Turkish and another 72 hours to train the LLAMA or larger model and 72 hours to optimize it with a GPU card. BUT I found that it's possible to hire a GTX 1080 12 GB GPU WINDOWS 10-11 Computer from Shadow PC company for 29 dollars per month. Computer memory is 256 GB. Every Turkish or foreign radio or hifi amateur can use this . I will publish all the data free at GITHUB. Mustafa Umut Sara? Istanbul TA2JCO , 73
|
QRP Labs kit shipping status
Hi all Well I got back to Dayton and after some work to release my suitcases full of electronics, which were impounded by customs at our local airport, all is well. I have a large backlog of sleep and emails to catch up on, as well as the jet lag... Since the SSB firmware beta release at the end of March, sales volume doubled compared to the normal monthly average. The waiting list for assembled radios also doubled (see attached chart which includes Dayton booth orders for assembled transceivers). Unfortunately I am informed by my colleagues that some kits have run out before the new batches arrived. It takes several months to arrange manufacturing of a batch of 1,000 QMX or QMX+ kits, including component procurement from various parts of the planet, PCB manufacture and SMD assembly, preparing the kits of through-hole parts etc. It's a little difficult to judge the timing exactly. So: QMX: no stock. All through hole parts are either on hand, or on the way. The PCB assembly is completed and just pending shipment. All QMX kit and assembled radios orders are waiting for this new batch of QMX boards to arrive. QLG3 GPS: no stock. There was an unexpected lead time on the E108 module but they were delivered yesterday. Now pending the SMD factory to assemble the boards, they will be shipped (along with lots of other board batches I ordered). Any QMX+ orders that include the QLG3 option are pending the arrival of the new batch of QLG3 boards. I updated the shipping status page: https://qrp-labs.com/status We have several other kits that are low on stock (including QMX+) but the new boards and parts are expected to arrive before the stock reaches zero. Thank you for your continued support. 73 Hans G0UPL http://qrp-labs.com https://www.buymeacoffee.com/g0upl
|
QMX+: low RX sensitifity @ 6m / 22 dB noise floor / 1500 Hz wistle
14
Hello, I recently build QMX+ and noticed a huge noise floor and wistling a 6 m. Therfore I measured sensitifity using a signal generator. Band [MHz] Sens [dBm]@Noise+10dB Noise [dB] 1,85 -100,0 0 3,65 -106,0 0 5,35 -107,0 6 7,07 -107,0 0 10,14 -110,0 2 14,23 -107,0 1 18,1 -103,0 4 21,04 -99,0 9 24,9 -97,0 7 28,1 -103,0 5 50,5 -79,0 20 Method: injection of signal Rx frequency + 1,1 kHz Mode: Digi Noise: signal detected by WSJT-X Sens: Noise floor + 10 dB Noise floor is increasing by frequency, but at 6 m wistling and digital noise make RX unusable. Output at 6m is about 2 W at 9V Vbat (12V Version) 1,5 kHz wistle starts at 20 m and is increasing by frequency Using wsjt-X i evalating waterfall diagram: (QMX+ at 50 Ohm) AVE:1, cumulative Noise [dB] : signal by WSJT-X signal in waterfall diagram AU QMX+ freq [Hz] freq [Hz] freq [Hz] freq [Hz] freq [Hz] freq [Hz] Frequency [MHz] Noise [db] Noisefloor 500 1000 1500 2000 2500 3000 1,84 0 0,2 0,2 0,2 0,8 0,2 0,2 0,2 3,57 0 0,3 0,3 0,3 0,7 0,3 0,3 0,3 5,3 1 0,5 0,5 0,5 1 0,5 0,5 0,5 7,04 5 0,8 0,8 0,8 1,1 0,8 0,8 0,8 10,1 1 0,5 0,5 0,5 0,5 0,5 0,5 0,5 14,1 1 0,5 0,5 1,3 2,8 2,1 0,5 1,5 18,1 5 0,8 1 1,8 2,3 2,1 1,1 2 21,1 7 1 1 2,1 2,6 2,1 1,1 2,1 24,9 7 1 1,2 2,7 2,8 2,7 1,3 2,5 28,1 7 1 1,2 2 1,5 1,5 1,2 2,2 50,3 22 1,8 2,5 3,8 4,2 3,9 2,8 3,8 At higher bands 1,5kHz signal is constant and independent from RX freuqency. At 6m additional audio signals are appearing. 6m RF Filter seems to be ok. Any suggestions to detect cause? vy 73 Michael (DF6TV)
|
Tune SWR on QMX+ always shows SWR <1.05
13
Hi, I wanted to tune my magloop antenna with the "Tune SWR" tool in the QMX+ menus, and no matter what I do with the antenna, the SWR never goes above 1.04 or so. That cannot be correct, as I'm using a magloop. Any tips on what I'm doing wrong?
|
Audio level request
10
I would like to block Digi mode audio from going to the earphone jack. The audio levels are painfully loud when adjusted for correct usb output to WSJT-X. I am afraid to switch between sidebands or to CW without removing headphones first. Any Suggestions ? QMX+ with 1.02 . de VK6XT
|