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+ is dead after voltage drops to 5V
#ATmega328
#QMXplus
|
Hi,
?
I was experimenting with a dev board and Arduino Nano on it powered through QMX+ Vdd. Everything was going on fabulously until I turned radio off while ardiono was plugged to PC through USB for programming. Radio turned off, then showed 5V and that was it. It does not come back anymore.
?
Is there a chance I can revive the board or it is gone?
--
73, Sardar ( AC8L), |
|
Some details:
?
Correction: Arduino on dev board was powered from Vcc (5V) on JP106.
?
Firmware was 27.
Can not go to firmware update mode even with the jumper wire (Aux2 to ground).
When plugged to PC - no COM port to try putty. So, board exhibits a brain death.
?
When 12V applied and PWR encoder is long pressed - voltage on JP106 from VCC and GND goes to 0.1V then gradually within few sec drops to 0V.
?
Any help would be appreciated.
?
--
73, Sardar ( AC8L). |
|
Sardar, did you measure Vdd when pressing the PWR On command. If you did and it was not there either, you might try a new 5 volt power regulator board. It may have been damaged by the backfeed, although I can't see the reason that might?happen. But it does provide for the initial startup sequence of events by seeing the Power On command from the switch and supplying 12v to the linear regulator which then boots up the processor which sends PWM back to both regulator boards and shuts down the linear regulator. Just a thought ... Ron On Tue, Sep 17, 2024 at 7:55?AM me via <me=[email protected]> wrote:
|
|
Not being critical, but ...
toggle quoted message
Show quoted text
I remembered seeing a pretty clear warning, highlighted in red, about not drawing any appreciable current from the VDD OR VCC rails when interfacing through the dev board. It's on page 68 of the assembly manual. Hopefully new SMPS board(s) will bring it back to life.? Greg
|
开云体育Hi Sardar;VDD in the QMX+ is 3.3 Volts. Is this what you had your Nano connected to? Can you explain what you mean by "showed 5V" - how was this voltage measured? 73; Steve, N3SB? ------- Original message -------- From: me@... Date: 9/17/24 1:26 AM (GMT-05:00) Subject: [QRPLabs] QMX+ is dead after voltage drops to 5V #ATmega328 #QMXplus Hi,
?
I was experimenting with a dev board and Arduino Nano on it powered through QMX+ Vdd. Everything was going on fabulously until I turned radio off while ardiono was plugged to PC through USB for programming. Radio turned off, then showed 5V and that was it. It does not come back anymore.
?
Is there a chance I can revive the board or it is gone?
--
73,
Sardar ( AC8L), |
开云体育Hi Sardar;Just saw your follow-up message. 1) With 9 - 12 Volts applied to the QMX+ power jack, what voltage do you see at the 3.3 Volt output of IC101 (78M33 3.3 Volt regulator) on plug-in power supply module pcb #2 when you press the power button? 2) With 9 - 12 Volts applied to the QMX+ power jack,?what voltage do you see at any of the VDD pins on the main board when you press the power button? 73, Steve, N3SB? -------- Original message -------- From: me@... Date: 9/17/24 10:55 AM (GMT-05:00) Subject: Re: [QRPLabs] QMX+ is dead after voltage drops to 5V #ATmega328 #QMXplus Some details:
?
Correction: Arduino on dev board was powered from Vcc (5V) on JP106.
?
Firmware was 27.
Can not go to firmware update mode even with the jumper wire (Aux2 to ground).
When plugged to PC - no COM port to try putty. So, board exhibits a brain death.
?
When 12V applied and PWR encoder is long pressed - voltage on JP106 from VCC and GND goes to 0.1V then gradually within few sec drops to 0V.
?
Any help would be appreciated.
?
--
73,
Sardar ( AC8L). |
|
Hi Greg!
?
I have read that portion of the manual too, since I bought QMX+ to integrate the ATU tuner and internal battery. And, as an experimenter, mentally I was ready to accept risks associated and bite the bullet.
?
So, I did not think that Nano would cause an appreciable current to be drawn while radio is off and it is powered through the usb.
?
Without turning radio off and arduino being powered from QMX+ - all has been working perfectly for couple of weeks. Until I plugged usb to arduino and turned radio off :)? And, I did that on purpose to see the serial debug messages from my code while radio is powering up.
?
And before powering arduino from QMX+ and USB on it - it also consistently worked from the protype board and arduino being powered with its own power supply. But grounds between itself and QMX+ were shared. Turning radio on/off did not cause any issue. So, these two bits of a success deceived me :)
?
I have a revision 2 board with SMPS integrated. But most of its components are big enough I can resolder them with my fat fingers. Now I will have to troubleshoot to find what is broken on SMPS. Will follow Steve's advise.
?
--
73, Sardar ( AC8L). |
开云体育Hi Sardar;Page 15 of the Rev 2.00 Assembly manual shows that IC101 is the 78M33 voltage regulator. It is the only 78M33 regulator in the parts list. Page 10 of the Rev 2.00 Assembly manual shows the location of IC101 - near the bottom right corner of the board next to JP201. 73, Steve, N3SB -------- Original message -------- From: me@... Date: 9/17/24 4:43 PM (GMT-05:00) Subject: Re: [QRPLabs] QMX+ is dead after voltage drops to 5V #ATmega328 #QMXplus Only rev1. schematics are available. But on Rev2 there are 4 voltage regulators in different parts of the board. It will be a little dangerous to try to repair the board without schematics.
?
--
73,
Sardar ( AC8L), |
|
Ah. Probably just a gotcha.
toggle quoted message
Show quoted text
Perhaps pull Arduino power right from the main power-in barrel connector or the very nearby JP using some header pins and include a drop down converter (if needed for the Arduino) on the dev board. Googled Arduino power specs just a few minutes ago. Sources say 7 to 12 volts. Others say beware of using 12 and convert down ... Hope the fix is easy.? 73 GREG KI4NVX?
|
|
Steve,
?
Many thanks for your guidance on locating the IC101.
?
I applied 12v, well, 11.5V to the barrel connector.
?
While pressing and holding the power button:
?
1. The output of IC101 gives 1.6V;
2. Both Vcc and Vdd at JP106 give 0.6V;
3. Both Vcc and Vdd at JP105's inner row also give 0.6V.
?
Does this mean IC101 is gone?
?
I checked - DigiKey has 78M33 in TO-252 package in stock:
This is the one, right?
?
--
73, Sardar ( AC8L). |
开云体育Hi Sardar;It's more likely that something on the 3.3 Volt VDD bus is drawing excessive current.? Does your power source have the ability to display current draw? The QMX+ should only draw maybe 100-200 milliamps during this bootstrap process. You could also use your DVM to measure the current draw by putting it in series with your power source. BTW - if you haven't already done so, remove your Arduino circuitry so that you are just troubleshooting the QMX+. If the current draw is less than this, then it could be IC101. They are pretty robust devices, and Hans has put a blocking diode at the output of U101 (at least on the SMPS daughterboard design) to isolate the U101 linear regulator when the high efficiency switching regulator starts up and the linear regulator shuts down. If the current draw is a lot higher than the 100-200 milliamps, then there is a short that will need to be found. 73, Steve, N3SB? -------- Original message -------- From: me@... Date: 9/17/24 9:50 PM (GMT-05:00) Subject: Re: [QRPLabs] QMX+ is dead after voltage drops to 5V #ATmega328 #QMXplus Addition: While pressing and holding the power button the IC101 INPUT gives 9V (normally 0V). So, 12v gets downconverter to 9V then fed to IC101 but guy is broken and does not produce the 3.3V?
?
--
73,
Sardar ( AC8L). |
|
Steve,
?
Immediately following the incident I have removed all homebrew circuits and mods from QMX+ board (all made pluggable). So, everything written in this thread from the beginning assumes the circuits caused the damage were removed.
?
While pressing and holding the power on button board draws 600 milliamps - 3x the normal. I have detached the front panel and connected jumper wires to pins of the JP302, connected to the power-on switch of the encoder. So, shorting that to the ground (according to schematics simulates the power-on action). That test also causes the 600 milliamp draw while shorted. At least eliminated the front panel circuit from troubleshooting.
?
Otherwise QMX+ main board is less dense and everything on both sides are pretty much is at the clear view. Although I have enjoyed more than a month of a full functionality and even made few FT8 contacts on multiple bands, I went ahead and reinspected the entire board using my microscope. Could not find any issues.
?
Now, if anything causes the short - that could be a result of a damage - internal element short (I hope very much not one of the small IC's I will not be able reshoulder).?
Any suggestions for further troubleshooting?
?
--
73, Sardar ( AC8L). |
|
Stan,
?
Tests done using diode mode of a multimeter. Readings are:
?
Q105 AOD403 p-channel mosfet, TO252 package, PCB top.
Drain neg probe, source positive - 0L.
Source negative probe, drain positive - 0.560 V drop
This looks healthy.
?
Q108 AO3415A p-channel mosfet. SOT23-3 package, PCB bottom.
Drain negative, source positive - 1.36V drop (should be 0)
Source negative, drain positive - 0.36 V drop
Looks like this one is fried.
?
Q103 - results are similar to Q108.
?
I did not check other mosfets of SMPS schematics yet.
?
--
73, Sardar ( AC8L), |
|
Interesting measurements.
In your second post that updated the statement of the failure, you said that you had the external device connected to Vcc (5V), not Vdd (3.3V).? Is that correct? Or was it the 3.3V rail?? or both?? The answer to this definitely guides the troubleshooting process.
Being on the 5V Vcc supply is why I said you should check Q108 and Q105, since a back-voltage on the 5V rail should hopefully only affect this part of the circuit.
And your test shows that Q108 is likely blown, which would prevent the 5V Vcc supply from coming up.? The failure could also have damaged other components in that same part of the schematic, to the right of Q105.? Only tests/measurements could say for sure.? If there is a blown component that shorts to ground (e.g. D106), it could cause significant current draw on the 12V line.
But your tests also show something holding down the output of the 3.3V Vdd supply.? So obviously something is wrong here, too.? The only thing connecting across these two supplies is the 12V line, and it is getting enough voltage to power IC101.? Note that normally, the IC101 3.3V is only 'on' for about 1/4 second right after power-on, while the firmware boots and enables the more efficient 3.3V supply.? This is obviously not happening, so the mcu is never getting enough voltage to boot.? I'm not sure exactly what to look at next, but with the output of IC101 lifted, and power_on set, I would check the voltage on the 12V line, and at the input and output of Q109, Q111, Q108, and test D103.? That will give some additional data points to guide the debugging...? Good luck. |
开云体育Hi Sardar;This test was a good idea - it will help to pinpoint the short. You could try lifting both leads of Q108 (leave its tab soldered in place) then reconnect U101 and see if the current draw is still high. Q108 is part of the 5 Volt VCC switching regulator. If the current draw is still high after this, then lift both legs of Q109, which is part of the 3.3 Volt VDD switching regulator, and measure the current draw. 73, Steve, N3SB? -------- Original message -------- From: me@... Date: 9/18/24 5:50 PM (GMT-05:00) Subject: Re: [QRPLabs] QMX+ is dead after voltage drops to 5V #ATmega328 #QMXplus I have unsoltdered the output of IC101 from the PCB and while testing, it now shows 3.3V when power on is pushed. Current draw is 0 amp.
?
So, problem with a short is further the schematic.
?
--
73,
Sardar ( AC8L), |
Ronald Taylor
to navigate to use esc to dismiss