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+ Strange Audio Artifact - Help, Please...
|
Hi All,
?
Just completed building my third QMX+ (the first two were quite successful) and I've come upon an issue that I need help with. The rig appears to work as advertised: receives well, GPS works, RF output levels are acceptable, everything works great except.... when I key down I hear what sounds like a machine gun. It is not there during receive - only when I transmit. Not real loud, but loud enough to be annoying. Happens on all bands, with AGC on or off.
?
Any idea what might be the cause, or where to start looking for it?
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
On Sat, Mar 22, 2025 at 08:14 PM, Jim Bennett / K7TXA wrote:
It sounds like it is possible you are hitting a current limit on your power supply and/or the voltage is dropping below the rigs TX voltage limit, possibly rapidly cycling above and below a voltage or current limit.? Can you monitor the input voltage to the QMX when you transmit?
Brien - KE7WB |
|
toggle quoted message
Show quoted text
|
|
OK folks, did some measurements a few minutes ago and have interesting results. But first, to answer a couple questions: The power supply is not current limited, and the voltage is a pretty solid 11.7-12.0 volts. This happens ONLY in CW - I am at firmware level xx.027 - not the beta SSB version. And this happens on either a real or dummy 50 ohm load.
?
Looking at the PuTTY diagnostics panel, it appears that there is a problem in the "Bias SMPS" area. On a known, working QMX+ when I press "T" the Current in that section is 30.5 ma, Duty Max is 50%, and Duty Cycle is 16%. On my new QMX+, pressing "T" results in the current flopping between 0 and 41 ma, Duty Max is 50%, and the Duty Cycle is 0%. Clearly a problem in the Bias SMPS area. And that is where my electronics knowledge hits me in the face. I have no idea where that Bias SMPS is coming from or what component(s) might be the ones to look into. Bad solder joint - possibly? Yeah, I could re-heat solder connections, but it would be nice to zero in on a specific area. Bad component - maybe? But where does this inexperienced Sherlock Holmes start?
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
Jim, the bias 'smps' is the isolated circuitry at the upper left of page 5 of the schematics.? It feeds the TX_BIAS to the input and output sides of the LPFs at the top right of page 5.
?
From an assembly point of view, I would look carefully at L501 (the surface-mount 330uH choke) and make sure it didn't get damaged.
Then I would also look at two of the leaded 47uH inductors you soldered in, L504 and L514, which carry the bias current to the LPFs.? Make sure they are soldered correctly, and since they are somewhat fragile, make sure neither of them is cracked.? These are the primary assembly-related components on the bias current.
Stan KC7XE |
|
Hi Stan,
?
Sorry for the delayed response to your post - busy weekend. Solder joints for L504 and L514 looked OK, but I re-heated them anyway - no change, still sounds like a 30 wpm machine gun on power down. Neither of them are cracked or messed up in any way. Every past kit I got from Hans had superficially damaged surface-mounted 330uH chokes. This kit 'appeared' to have no damage when I got it. However, near the end of construction, I noticed that the one nearest the PA transistors was also slightly damaged. It looked like the wires were not broken so I put a dab of CA glue on the plastic to keep it in place. Now, I'm wondering if in fact there was wire damage.
?
Are there any voltage measurements I can do to confirm the status of that little thing? If it is hosed, I'll probably have to send the rig off to Jeff Moore for repair, as there is no way I can fix one of those things where it is located. I've done some minor SMD work in the past, but if that component is broken - not a chance.
?
Regards, Jim
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
Too bad - I was hoping that L504 or L514 was broken, since they are so easy to replace.
To test L501, the 330uH choke, I would first test it for proper continuity through the circuit card by probing the devices it connects to.? That would be the top end of D502 and C504.? Those are the large-ish diode and capacitor along the sides of L501.? The top end of the diode is toward the edge of the circuit card, and it is the stripe on the capacitor.? This should show continuity (a few tenths of an ohm or so), and I would touch the coil while measuring, and apply a little pressure to move it around a bit to see if there is an intermittent connection.? Likewise with power on and in transmit mode, the voltage to ground from either of those points should be about the same.? If you look at the voltage on those points with an oscilloscope, you will see some saw-tooth-like variations at the frequency of the switching, around the DC value, with the capacitor side being somewhat more filtered.
I hope that shows something.? You can also check for continuity across L504 and L514 in case something is hidden there.
Stan |
|
Hi Jim,
?
L501 or C504 failure can cause these symptoms. ?Make sure to have about 2 ohms continuity on l501 at the pcb pads. ?It’s a little difficult to test C504 in circuit. ?These big tantalum’s fail though. ?You can remove it and test it on the bench.
It’s hard to diagnose the exact problems in the bias circuitry. ?There is interaction everywhere. ?But if you look around with a scope, you can often get a clue to what’s failing. ?Most often it’s one of the transistors, the coil being broken or the big cap failing.
?
if you can’t find the culprit in that bias circuit, certainly send it along to me. ?
--
73 Jeff Moore W1NC |
开云体育Back from the trout trip. Sunburned, and tail between legs - didn’t catch a thing. But my two buddies were able to net three very nice 3-4 pound rainbows. Yeah, that’s BIG! Maybe next time….Anyway, just got done messing with the QMX+ and I wish I could see something conclusive. Interestingly, the “331” inductor that I thought was broken was L101. Per your suggestions, I have continuity between the + of C504 to the + of D502, so L501 is not broken. Since I thought L101 was hosed, I also checked the + of C107 to the + of D106, and the neighbor of L101 (L102) from the + of C106 to the + of D107. Continuity there also.? I applied power and measured the voltages to ground from the + of C504 and D502 on key down. Both showed about 2.9vac. I also measured it as DC and it jumped all over the place from 1 v to 5 v. I hesitate to remove C505 to check it but if you think at this point it is worth doing, I can give it a shot. What’s the worst that can happen? :-) Any other suggestions? Jim
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
What input supply voltage were you using when you made those measurements?
D502 should have a square pulsed signal from about 0V to almost the supply voltage, which should average to a small constant DC voltage when measured with a DMM.? C504 should have a relatively clean constant DC voltage, at about 2V.
So something is likely broken here.? Since L501 has continuity, it is likely the diode or capacitor.? Jeff would know better than I.
Stan KC7XE |
|
toggle quoted message
Show quoted text
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
Jeff - you meant C504, right? Not C604?? :-)
?
I can handle replacing C504 and D502, I'm pretty sure. Looks like I can get a C504 from Mouser for $0.29 providing I have the right part number (512-SS14), plus shipping - ughhh. If that's not the right part number, please let me know. As far as the tantalum capacitor C504, not sure about a Mouser part number on that one. If you have a number then I could get both from the same supplier and save a bit on shipping.
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
Took a deep breath and removed C504. It measures about 500 uf so this cap is fine. No problem re-installing it. Other than my shaky hands!
?
I did find an interesting thing while doing the testing. I had been using an Astron VS-35M p.s. in the shack into a boost-buck device set to exactly 12 v. It also shows current but I've found that to be completely bogus. Today I took the QMX+ to the garage workbench and connected it to my benchtop p.s. which is a lot more accurate than the boost-buck and is capable of current and voltage limiting. With it set at 12.0 v I did a key down on the QMX+. The output voltage stayed at 12.0 or so, but the constant current light and constant voltage light both flickered - and at the exact tempo as that machine gun sound on my headphones. Strange. I unhooked the radio and replaced it with another QMX+ that I built last year that works perfectly. No machine gun audio on key down and no flashing lights on that bench p.s.
?
At this point I may have reached the upper limit of my debugging ability. Perhaps the only other thing I could try would be to remove D502 and test it, although it is a lot smaller than that tantalum cap and might prove tricky for me to attack.
?
If you guys think it's worth testing that thing I'll give it a try. Or if there is something else that is within my capabilities... Otherwise it might be headed to Jeff for a New England vacation.
--
Jim / K7TXA Eagle, ID SKCC 10447T BUG 301 |
|
toggle quoted message
Show quoted text
|
Brien - KE7WB
Donald S Brant Jr
to navigate to use esc to dismiss