开云体育

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

QMX(+) Build experience, Poor Audio Filter Sweep/RF Filter Sweep/Image Sweep/ADC IQ test and resolution with new PCM1804


 

So maybe I have an IC407/PCM1804 failure on a QMX+ REV1 board? Audio, RF, image and Test ADC I/Q sweeps are all static, random noise plots while LPF is better than any of the three or four QRP Labs transceivers I have built yet. SWR is kinda okay and receive is loud noise.
?
What say you? I will try the voltage and mayne the scope checks...
--
73
Karl
KI4ZUQ


 

Karl, Audio, RF, image and Test ADC I/Q sweeps on the one hand and LPF sweep on the other hand? are using different ways. After reading QMX Terminal Hardware Sweeps you may have a clue where to debug. Maybe it's a broken PCM1804 (see QMX QMX_plus Audio PCM1804 ) or a failure in the section from Q508 up to IC407 (PCM1804).
?
73 Ludwig


 

Thanks, Ludwig! Appreciate your input!
?
Somewhere? on this forum and on this topic of PCM1804 failures is a voltage check on pin pairs of the 1804. I saw it, read it and thought I could find it...but...
?
I think it was different? pairs would see about 2.5 vdc on certain bands..where was that? I will delay replacing the IC407/PCM1804 until this simple test is passed.
?
By the way, my 1804 looks like authentic Texas Instruments logo but then a fake would, too.
--
73
Karl
KI4ZUQ


 

On Sun, Mar 16, 2025 at 10:31 PM, Karl wrote:
would see about 2.5 vdc on certain bands
Karl, I guess you mean IC402 and IC403 and for QMX+ IC408. If one or more is/ are broken it will interrupt the RX signal path.
?
A good way to check if the problem is inside the PCM1804 or before this IC is using the method No 2 from QMX QMX_plus Audio PCM1804 section Diagnostics.
?
73 Ludwig


 

So I am preparing to inject a signal and see the path through IC405 and IC406 makes it to IC407, the PCM1804 and what do I find? A complete continuity break from C422 to the junction of IC405, R413 and R416!
--
73
Karl
KI4ZUQ


 

Further continuity testing shows a break between C423 and the junction of IC405B pin 6, R413 and R417.
?
I venture to write this would be enough to render IC407 with no signal to process.
--
73
Karl
KI4ZUQ


 

My experience with assembled board (and repaird boards) at a really larger computer
manufacturer is all ESD.
?
Example board fails or deemed failed (board swapping) returned ot Depot to test and repair.
Tested board is repackaged and eventually to the field for use.... only to be returned again
now failed.? With thousands of said boards repair costs were a factor so investigation found
ESD (we are talking 74xxxx class TTL)? was the cause.? Adding tracking to the system? found
where and even who so inquiry was done to find out.??
?
End result was ESD at the site end point, in dry locations, and Field Engineer not using?
ESD handling technique.? Or the external system had poor grounding and introducing?
transients (traceable to board that did external interfacing to the machines.? Generally
reinforcing ESD procedure in the field and insuring they had needed kit had a marked?
improvement in dead on arrival (or shortly after) boards.
?
Likely cause would be during last parts stuffing (end user) the board suffered ESD hits
likely un-noticed while handling.? So that might answer shipped good failed in field.
Those would not be guarenteed immune as its cmos in a cmos environment unpowered.
?
Also that would be seasonal, aka worse in dry weather (desert southwest usa)?
or during winter (heated dry air).
?
I'd add ESD warning during assembly with suggesting mats and all.
?
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.


 

Well said, Allison, excellent reminder!? Very similar to my experience.
?
I definitely would encourage using good ESD practices during handling and assembly of QRP-Labs kits, including use of a good ESD mat and wrist strap.? It would likely significantly decrease failures that happen for "unkown reasons", sometimes after the assembled kit has been working and in use for a time.?
?
Invisible ESD events can weaken a chip internally such that it later fails during normal operation.?
?
Stan KC7XE?


 

Karl, how are you testing for a continuity break between those capacitors and the IC inputs?? It would require a board fabrication error, or faulty soldering on the capacitors or ICs to cause such a fault, so while certainly possible, it seems unlikely.? It seems more likely a measurement error, perhaps a layer of flux on the capacitor terminal or IC pin preventing good contact of the probe?
?
And no, such a fault would not render IC407 with no signal to process.? Each of these is a feedback capacitor, which helps stabilize and improve the performance of the amplifier - so yes, it is an important part of the circuit to keep performance high, but it would not inhibit the signal from going through and being amplified.
?
Stan KC7XE


 

I have some pin-sharp probe adapters for DMM normal probes that poke through that durable film that seems to cover all the SMD joints. The connection, for example between C422 and pin 1 of IC405 can be verified as good. The other end of C422 should do the same for R413 and R416 ends towards the rear of the pcb..but the connection is open. Numerous probings yield no connection. It is either a pass through or a missing/broken/incomplete trace.
--
73
Karl
KI4ZUQ


 

Additionally, there is a second REV 1 board with the same problem. I have two case numbers with Hans dealing with this.
--
73
Karl
KI4ZUQ


 

On Wed, Mar 19, 2025 at 01:43 PM, Karl wrote:
The other end of C422 should do the same for R413 and R416 ends towards the rear of the pcb..but the connection is open
Karl, exactly where are you probing for R413/R416?? You say "towards the rear of the PCB", but R413 and R416 are located on the bottom of the pcb, directly under IC405, as are the resistors for the other section of that op-amp.?? So I'm wondering if you are mistaking R513 or R516 for the correct point - that would of course have an open circuit.
An easier place to probe for continuity of that same trace is at pin 2 of IC405, adjacent to where you probed for the other side of C422.
Stan KC7XE


 

--
73
Karl
KI4ZUQ


 

I have been probing R213 and R216! Talk about blinders!
--
73
Karl
KI4ZUQ


 

Just to see how far up my head has been during this 'cranial-rectal inversion' I did a continuity test C422 to R413/R416 which was, of course, successfull!
?
Thank you, Stan, for being the lighthouse in the storm!
--
73
Karl
KI4ZUQ


 

Thank you, Stan, for being the lighthouse in the storm!
Sometimes we all need someone else to shine a light into our rabbit-hole and help us back out.?
Stan KC7XE?


 

Finally got a 14.074 mhz signal into the QMX+ at about 17 mv. My Rigol DS1102E scope may be insufficient for this task but I can display the 17 mv signal at the BNC? but nothing at pins 1 and 7 on either IC405 or 406. So my problem? is likely back towards the BNC...
--
73
Karl
KI4ZUQ


 

The signal at IC405/406 should be about 10x larger, so easily visible.
If your scope really lets you see that smaller 17mV signal (my cheap scope gives up well higher that level), you should be able to trace the smaller signal through the receiver section.? Look at IC 408 pin 5, the signal after the t/r switch, and then IC402 pin9, the output of the band selector mux.? Those two points will help see where the signal disappears.


 

I was kinda counting on seeing that 10x signal! I will take you test sequence and report back. Thanks!
--
73
Karl
KI4ZUQ


 

Stan, signal present at BNC, pin 5 IC408 and not present at pin9 IC402.
The signal is also present on pins 1&2 IC408 but not on pins 3,4, 12 thru 15. The signal pins 1&2 IC408 are same phase. The signal thru L403 arrives at pin 11 of IC402 a mess about 4 x the base freq of 14.074 mhz, reduced amplitude.
?
Might be bad joints in L403 and probably L401, 404 & 405....
--
73
Karl
KI4ZUQ