Re: QMX - smoke - another C107/Q108 failure
I'm curious about what you propose, but that link doesn't show
anything. :-(
Paul -- AI7JR
On 8/30/23 18:17, Stephan Ahonen KE0WVA
wrote:
toggle quoted message
Show quoted text
>In the end, it has to be acknowledged that you can never
please all the people all the time, and have to try and strike a
balance somewhere. Which is often very difficult indeed.
I suppose you know your target market better than I do, and it's
presumptuous of me to think otherwise, I apologize.
>I think with my firmware changes I was able to very
substantially mitigate that vulnerability but the risk is still
there if extreme voltage fluctuations suddenly take place.
I think I remember reading somewhere that the control loop
operates at a frequency of 1 khz?
If this is the case then the software physically cannot respond
quickly enough. According to SPICE, if input voltage steps from 6
to 12 volts, with a 330u series inductor and 470u output
capacitor, 1 ms is enough time for the output voltage of the 5v
regulator to rise from 5v to 11v before the CPU even has an
opportunity to figure out that something is wrong.
to the power supply circuit would help. At a cost of
three extra parts (including a series resistor on the PWM_5V line,
not pictured), a voltage excursion now immediately shuts down PWM,
much faster than the CPU can react. Total BOM cost of $.10 or so?
|
Re: QMX -- first try, first problems (bad sweeps, no output power)
Sorry folks - found the short - yes, it was at the BNC socket.
Steve
|
Re: QMX - smoke - another C107/Q108 failure
>In the end, it has to be acknowledged that you can never please all the people all the time, and have to try and strike a balance somewhere. Which is often very difficult indeed.
I suppose you know your target market better than I do, and it's presumptuous of me to think otherwise, I apologize.
>I think with my firmware changes I was able to very substantially mitigate that vulnerability but the risk is still there if extreme voltage fluctuations suddenly take place.
I think I remember reading somewhere that the control loop operates at a frequency of 1 khz?
If this is the case then the software physically cannot respond quickly enough. According to SPICE, if input voltage steps from 6 to 12 volts, with a 330u series inductor and 470u output capacitor, 1 ms is enough time for the output voltage of the 5v regulator to rise from 5v to 11v before the CPU even has an opportunity to figure out that something is wrong.
to the power supply circuit would help. At a cost of three extra parts (including a series resistor on the PWM_5V line, not pictured), a voltage excursion now immediately shuts down PWM, much faster than the CPU can react. Total BOM cost of $.10 or so?
|
Re: QMX won't power off and odd voltages on Q103
Phil,
You also mentioned that the tabs on the rotary encoder broke at one time and you installed wires to repair them. Please check that area for shorts to the rotary encoder housing and that the switch works correctly. Check at the 8 pin connector to make sure you are also checking continuity through the solder joints.
73 Kees K5BCQ
|
Re: QMX -- first try, first problems (bad sweeps, no output power)
Hi Adam - I have the same symptoms as you - my rf sweeps are very similar with no receive or rf out. I've been stuck on this for hours - where was the short you fixed on yours? Did you mean the BNC socket?
I'm clutching at straws here!?
Many thanks,
Steve MW0KST
|
Re: QMX won't power off and odd voltages on Q103
Phil,
?LIN_REG_EN (the middle pin on the top of PCB #2) should be high when you want the 78M33 regulator switched ON and you should have ~3.3V at D103. What does the pin measure ? Ground that pin and the 78M33 will turn OFF.
73 Kees K5BCQ
|
Sure, right after making the side tone volume adjustment actually adjust the side tone volume.?
That's something that materially affects basic operation [lest it cause ear damage while trying to work someone else who's under an S-5 to S-9).? When it can be worked into the schedule, of course.
Darned nice engineering, though.? I like it.
K3RTA
|
Re: QMX won't power off and odd voltages on Q103
It occurs to me that PWR_HOLD runs through Q104--have you checked for the Q103/Q104 short?
Jonathan KN6LFB
|
Re: QMX won't power off and odd voltages on Q103
Of all your measured voltages, 2.16V on the rightmost pin of the 5V connector is the most bewildering. ?That should be the PWR_HOLD signal and either 0V or 3.3V as it's driven by the CPU. Hans has stated that the bootloader does not enable the SMPS and only runs off the linear regulator ( /g/QRPLabs/message/106572).??Have you looked over the 3v3 board to make sure the linear regulator is running? ?LIN_REG_EN (the middle pin on the top) should be higher than 0V when switched on and you should have ~3.3V at D103. It certainly seems like there's a short or bad connection somewhere that might be pulling the 3.3V rail down and preventing the processor from booting. Jonathan KN6LFB
|
Re: QMX won't power off and odd voltages on Q103
I put the bench supply up to 9volts and 250ma. Still couldn't shut it down, so the MCU isn't starting.?
I've looked at the board, and it looks fine visually. There wasn't much I could do to hurt it, since the only thing the builder does is solder the pins of the 2X4 female connector to it. I've checked those and freshened them up several times. Also, I've touched up the pins of the male connector 2X4 where they connect with the main PCB. And I checked those pins for shorting to the Source and Gate of Q103.?
Can someone point out some specific steps I should take next?
Phil K3UT
|
Using a QDX with AndFLmsg for Olivia
Hi folks,
I am trying to use my QDX-M with a Galaxy S-5 phone, using the AndFLmsg app to run Olivia. I am able to transmit Olivia messages OK, using "VOX enabled", as I can receive the transmission on another SSB receiver and decode the message with a second Galaxy S-5 listening to the audio. What I can't seem to do, is use a QDX-M to pass received audio by USB (and OTG connector) to a Galaxy phone for decoding.
I would be very happy to hear from people who have managed to get the Phone/QDX interface working both on Tx and Rx.
Thanks for any comments
Ian
VE6IXD
|
Re: U3S No Power Troubleshooting
Just a bit more background:? everything else seems to be working fine...start-up screen (v3.12), gps reporting, transmissions starting and stopping appropriately.
|
U3S No Power Troubleshooting
I have had my U3S running (nearly) non-stop for a couple of years with no issues.? WSPR spots reported from Japan to Germany.
Recently...nothing.
The QRP-Labs Dummy Load reports:
? 80 meters: 0.022 volts (yes, that's two one-hundredths of a volt)
? 40m: 0.03
? 30m: 0.09
? 20m: 0.05
? 10m: 0.04
I will appreciate hearing from others who might have experienced something similar.? Where did you start troubleshooting? Were you successful in returning your U3S to operation?? What was the fault?
I was running at 5 volts, with a single BS170, 5x switched filter board, GPS mounted in the top of the standard enclosure (patch antenna fed through).
Thanks.
|
On Wed, Aug 30, 2023 at 10:35 PM, Da Amazin' man G0FTD wrote:
Now then, don't get me started about variable caps.
Linear wavelength types and linear capacitance types, mostly forgotten knowledge these days ;-)
1.33pf per meter of wavelength and then make the inductance to suit for optimum Q. ? -- - 73 de Andy -
|
On Thu, Aug 24, 2023 at 02:39 AM, Daniel Walter wrote:
Recommendation is not to switch inductors while transmitting. Really pretty easy to peak up noise or signal on Rx, then just use Cin & Cout to fine tune. Rarely do you need to change inductor, but if you do, turn off Tx during switching.?
Good advice. Unless you know how the switch works then this advice should precedence. But just one small note, it can depend upon the switch type. Switches can be break and then make (potentially leaving an open circuit before the switch has made) or can be the opposite (make before break) Break before make or make before break. It can make a difference to expected operation and is often overlooked when purchasing a simple switch in other circuits too. Now then, don't get me started about variable caps. Linear wavelength types and linear capacitance types, mostly forgotten knowledge these days ;-) -- - 73 de Andy -
|
Re: Changing USB port on QMX
I haven’t fully sorted this out but I think that if you assign a com port it survives to any physical port and through restarts. Give this a try:
- Open the Device Manager by pressing the?Windows Key?+?R. Type “devmgmt.msc” and press?Enter.
- Expand the?Ports (COM & LPT)?section.
- Right-click the COM port and select?Properties.
- Click the?Port Settings?tab and click?Advanced.
- Change the COM port number to an available COM port number.
- Reboot
-- Colin - K6JTH?
|
Re: QMX won't power off and odd voltages on Q103
I am unable to see the QMX on my computer.?
|
I've just inadvertently? connected my QMX to a different USB port on my W11 laptop and it failed to connect the WSJT-X.? What happens is that the different port means that a different COM port is assigned and different USB audio ports.
Is there a way for an individual QMX to retain it's serial and audio port designations if it's moved to a different USB port on the same PC?
What I've found is that a FTDI serial port will do this but most others won't.? It makes things far easier - for the user at least.
And the developer, they stop getting "I can't find my QMX" messages.
Chris G5CTH
|
Anyone tried this 4sqrp tuner above 10w spec.
|
Purchased and assembled by Hans 2021. Worked flawlessly for a few months and then the tuning encoder started locking up Have replace the encoder three times and once again the new encoder will not even change frequencies.
All other functions work.
The towel is flying through the air
$75.00, includes shipping within continental U.S.
Tony
|
Paul - AI7JR
Colin Kaminski