Bluepill: Will not power up via USB

Generic boards that are not Maple or Maple mini clones, and don't contain the additional USB reset hardware
johnwigley
Posts: 5
Joined: Mon Jan 23, 2017 10:15 am

Bluepill: Will not power up via USB

Postby johnwigley » Mon Jan 23, 2017 10:33 am

20170123_101108.jpg
20170123_101108.jpg (184.7 KiB) Viewed 251 times

20170123_101127.jpg
20170123_101127.jpg (221.61 KiB) Viewed 251 times


I've purchased 3x what appear to be generic STM32F103C8 Bluepill boards from two different Ebay vendors, all three of the boards work fine when powered up through the 4 pin debug/programming connector via an STlink programmer, I've loaded the both the generic STM32 USB bootloader, and hello world sketches and the boards run absolutely fine.

The issue however is that when powered via the micro USB connector the boards will not power up at all - this issue occurs across all three boards from two different vendors. Although probably suffering from the incorrect USB pull up resistor, this should not as far as I understand it prevent the boards from powering up via USB, but will prevent correct USB enumeration?

Having read through the forum postings I see repeated mention of the incorrect USB pull up resistor, but I cannot see that this will prevent power up, particularly when fed from a USB mains charger rather than a PC USB port?

Is this a known issue? It seems strange that it would affect 3 different boards from two different suppliers?

Thanks,

John

stevestrong
Posts: 882
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany

Re: Bluepill: Will not power up via USB

Postby stevestrong » Mon Jan 23, 2017 12:51 pm

I had 3 boards with not working USB connector, which worked after I soldered them...

User avatar
ahull
Posts: 1422
Joined: Mon Apr 27, 2015 11:04 pm
Location: Sunny Scotland
Contact:

Re: Bluepill: Will not power up via USB

Postby ahull » Mon Jan 23, 2017 1:11 pm

The USB connector would be my guess too. Also what on earth are those splashed blobs of solder next to the boot pins on the back of the board all about. :o
If it isn't the USB connector then my next guess would be the voltage regulator, or... you powered up using USB *and* had the 3v3 connected from the ST-Link at the same.. That seems to cause the voltage regulator to take a fit or even pop. Check to see if the regulator is getting 5V in, and spitting 3v3 out.
- Andy Hull -

johnwigley
Posts: 5
Joined: Mon Jan 23, 2017 10:15 am

Re: Bluepill: Will not power up via USB

Postby johnwigley » Mon Jan 23, 2017 4:52 pm

stevestrong wrote:I had 3 boards with not working USB connector, which worked after I soldered them...


Were they also these generic Blue pill boards?

I hold those who can hand solder SMD components in great respect, unfortunately I'm not one of them. I wouldn't have a hope in hell of being able to de/re solder that microUSB connector without bridging half the pins together :(

I suppose I could solder a USB cable directly to the broken out solder pads on the edge of the board though, that should work...

johnwigley
Posts: 5
Joined: Mon Jan 23, 2017 10:15 am

Re: Bluepill: Will not power up via USB

Postby johnwigley » Mon Jan 23, 2017 5:01 pm

ahull wrote:The USB connector would be my guess too. Also what on earth are those splashed blobs of solder next to the boot pins on the back of the board all about. :o
If it isn't the USB connector then my next guess would be the voltage regulator, or... you powered up using USB *and* had the 3v3 connected from the ST-Link at the same.. That seems to cause the voltage regulator to take a fit or even pop. Check to see if the regulator is getting 5V in, and spitting 3v3 out.


I hadn't noticed those errant solder splashes until you pointed it out, though it's the only one out of the three boards to have them, and they all exhibit the same issue.

I read up on the errors of simultaneously powering it through micro USB and via the ST Link, so no, thankfully I narrowly avoided that one...

What's bizarre and doesn't make any sense, is that when USB is applied, I measure around 0.5v on the input to the regulator, and around 0.2v on the 3v3 output, which presumably is the LDO part of the regulator in action only dropping around 0.3v across it, BUT how on earth it's getting 0.5v instead of 5v on the input is beyond me AND THIS IS ACROSS ALL THREE BOARDS! I assumed at first it was a measurement error, but it doesn't appear to be, directly measuring a USB connector reads 5v, and then I've re-read it across different boards.

Unless they've mistakenly stuck in a low value resistor instead of a bypass capacitor across the 5v input (as it works fine when powered from 3v via ST Link, so the fault has to be on the high 5v side), then I can't imagine what could cause all the boards to show 0.5v on the 5v input rail.

I've also tried the obvious of trying them with several different USB feeds, both hubs, and USB chargers.

racemaniac
Posts: 345
Joined: Sat Nov 07, 2015 9:09 am

Re: Bluepill: Will not power up via USB

Postby racemaniac » Mon Jan 23, 2017 6:34 pm

johnwigley wrote:
ahull wrote:The USB connector would be my guess too. Also what on earth are those splashed blobs of solder next to the boot pins on the back of the board all about. :o
If it isn't the USB connector then my next guess would be the voltage regulator, or... you powered up using USB *and* had the 3v3 connected from the ST-Link at the same.. That seems to cause the voltage regulator to take a fit or even pop. Check to see if the regulator is getting 5V in, and spitting 3v3 out.


I hadn't noticed those errant solder splashes until you pointed it out, though it's the only one out of the three boards to have them, and they all exhibit the same issue.

I read up on the errors of simultaneously powering it through micro USB and via the ST Link, so no, thankfully I narrowly avoided that one...

What's bizarre and doesn't make any sense, is that when USB is applied, I measure around 0.5v on the input to the regulator, and around 0.2v on the 3v3 output, which presumably is the LDO part of the regulator in action only dropping around 0.3v across it, BUT how on earth it's getting 0.5v instead of 5v on the input is beyond me AND THIS IS ACROSS ALL THREE BOARDS! I assumed at first it was a measurement error, but it doesn't appear to be, directly measuring a USB connector reads 5v, and then I've re-read it across different boards.

Unless they've mistakenly stuck in a low value resistor instead of a bypass capacitor across the 5v input (as it works fine when powered from 3v via ST Link, so the fault has to be on the high 5v side), then I can't imagine what could cause all the boards to show 0.5v on the 5v input rail.

I've also tried the obvious of trying them with several different USB feeds, both hubs, and USB chargers.

you should be able to find out where the voltage is getting lost. i assume at the pins of the usb socket you still have 5V, and then just follow the traces, it's just a 2 layer pcb, shouldn't be that hard to follow around :)

User avatar
ahull
Posts: 1422
Joined: Mon Apr 27, 2015 11:04 pm
Location: Sunny Scotland
Contact:

Re: Bluepill: Will not power up via USB

Postby ahull » Mon Jan 23, 2017 10:08 pm

Assuming the 5V doesn't even appear on the 5V pin adjacent to the LEDs, then the fault is either the connector or the via that passes the track from the top (USB connector side) to the bottom, and on to the 5V pin. Buzz it out with your continuity tester. My second bet however based on the symptoms is that the 0V side is not connected, hence the weird half a volt floating voltage. The soldering looks OK in the picture, so I suspect the USB connector or more likely your USB cable since the problem affects all three boards.

EDIT: .. and if your soldering skills are up to it, put some more solder on the tabs that hold the USB connectors to the board. This is a well known weakness. The USB connectors are easily parted from the board anyway, but the solder on those tabs on the one in the pictures is almost non-existent.
- Andy Hull -

User avatar
RogerClark
Posts: 5470
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Bluepill: Will not power up via USB

Postby RogerClark » Mon Jan 23, 2017 10:19 pm

ahull wrote:
EDIT: .. and if your soldering skills are up to it, put some more solder on the tabs that hold the USB connectors to the board. This is a well known weakness. The USB connectors are easily parted from the board anyway, but the solder on those tabs on the one in the pictures is almost non-existent.


Cover the whole connector in hot glue, or even better with epoxy, as the copper pads on the PCB can get pulled away even if you resolder them.

I have 2 Blue pill boards which are no longer work, due to usb connector failure
(I may eventually get around to resoldering them and perhaps that will fix it, but I now only buy Black pill's and Maple minis, because they have a much better usb connector)

johnwigley
Posts: 5
Joined: Mon Jan 23, 2017 10:15 am

Re: Bluepill: Will not power up via USB

Postby johnwigley » Tue Jan 24, 2017 12:04 am

Thank you all so much for the help and advice.

I took leave of my senses, and decided to attempt to resolder the micro usb connector, and by some miracle all three of the boards are now powering up, and two of the three successfully enumerated as Maple DFU devices, and I flashed a test ECC routine which they both ran successfully and re-enumerated as serial ports! Success, though now I have to figure out how you flash them a second time as they appear as serial ports rather than DFU devices, though I'm sure there's a FAQ on dealing with that.

The third board which is one of the batch of two from the same supplier, rather than the single one from a different supplier, is powering up and running fine BUT appears as an unknown USB device.The STM32duino boot loader flashes it's LEDs correctly, and the board runs code uploaded through the STlink fine.

I've checked continuity through a cut off microusb cable, and tested that the cut off microusb cable's wires are connected to PA12 & PA11, and that ground, and 5v are also connected. Further more, I've checked that the 10k (should be 1.5k according to USB specs, but well known Blue Pill issue) pull up resistor is connected correctly as well. The other two boards both have 10k pullups and they enumerate and program fine, so evidently the USB hub I've got isn't strict.

Lastly I've checked that there aren't any shorts between the USB pins. So it seems to me that the only thing left to check is whether or not PA11 & PA12 are actually connected to the MCU. For that I suppose I can write a test program that waggles the pins up and down and see if that appears on the PA11 & PA12 breakout.

Is there anything else I'm missing in terms of a diagnostic process on this one remaining board? I know the boards functional and working apart from the USB, so it seems to me that I've conducted all the USB connectivity tests possible already, with the exception of confirming that PA11/PA12 are actually attached to the MCU?

Thanks once again for the assistance, it's much appreciated.

User avatar
RogerClark
Posts: 5470
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Bluepill: Will not power up via USB

Postby RogerClark » Tue Jan 24, 2017 1:21 am

johnwigley wrote:
... now I have to figure out how you flash them a second time as they appear as serial ports rather than DFU devices, though I'm sure there's a FAQ on dealing with that.


If you select the com port in the IDE, it then the MapleUpload tool (jar) sends a reset sequence via serial data + control lines which resets the board into DFU for upload.


Return to “Generic STM32F103”

Who is online

Users browsing this forum: No registered users and 4 guests