USB descriptor request failed - [SOLVED]

Generic boards that are not Maple or Maple mini clones, and don't contain the additional USB reset hardware
stevestrong
Posts: 630
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany

USB descriptor request failed - [SOLVED]

Postby stevestrong » Wed Jan 04, 2017 7:34 pm

Hi all,

3 pieces BP arrived yesterday, today I tried them out, I was able to flash all with the stm32duino bootloader over STlink.

- one had not-working power over USB, but blinking and seems to work over 3.3v direct supply. Solved by (re)soldering the USB pins.
- second board seems to execute correctly the loaded sketch over USB,
- the third one, however, has "Unknown USB device (Device descriptor request failed)". Seems to execute correctly the sketch loaded over STlink, but didn't check exactly the blink timing.

Anyone knows a hint what can I do with this? Can be fixed or I should just throw it out?
Last edited by stevestrong on Wed Jan 04, 2017 9:11 pm, edited 1 time in total.

victor_pv
Posts: 861
Joined: Mon Apr 27, 2015 12:12 pm

Re: USB descriptor request failed

Postby victor_pv » Wed Jan 04, 2017 7:46 pm

stevestrong wrote:Hi all,

3 pieces BP arrived yesterday, today I tried them out, I was able to flash all with the stm32duino bootloader over STlink.

- one had not-working power over USB, but blinking and seems to work over 3.3v direct supply. Solved by (re)soldering the USB pins.
- second board seems to execute correctly the loaded sketch over USB,
- the third one, however, has "Unknown USB device (Device descriptor request failed)". Seems to execute correctly the sketch loaded over STlink, but didn't check exactly the blink timing.

Anyone knows a hint what can I do with this? Can be fixed or I should just throw it out?


Did you check the USB pull up resistor value? (there are notes about the blue pill having 4k3 or 10k resistors, which are out of spec, USB spec is 1k5)
That may have something to do with it.
Also, when a diode blew up in my maple mini last year, it would do something similar. If I remember right the VCC voltage to the mcu was very low, which may the MCU unstable, replacing the diode solve the problem.

stevestrong
Posts: 630
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany

Re: USB descriptor request failed

Postby stevestrong » Wed Jan 04, 2017 8:16 pm

The USB pull-up is 10k, but all three boards have the same and the other two have working USB descriptor recognition, both DFU and CDCACM. But this one cannot get either DFU or CDCACM recognized by Win10.64b.
The 3.3V is correct and stable.

So I think it must have something to do with the USB peripheral, internal oscillator and/or internal clock for USB?
How could I evaluate whether it is good or not?

victor_pv
Posts: 861
Joined: Mon Apr 27, 2015 12:12 pm

Re: USB descriptor request failed

Postby victor_pv » Wed Jan 04, 2017 8:21 pm

stevestrong wrote:The USB pull-up is 10k, but all three are the same and the others have working USB descriptor recognition.
The 3.3V is correct and stable.

So I think it must have something to do with the USB peripheral, internal oscillator and/or internal clock for USB?
How could I evaluate whether it is good or not?


Can you output something from all three with the same code and compare the period? perhaps a pin toggling, or the SPI clock signal.

User avatar
RogerClark
Posts: 5272
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: USB descriptor request failed

Postby RogerClark » Wed Jan 04, 2017 8:43 pm

USB clock runs off the main 72MHz clock via a divider ( of 1.5 ) to bring it down to 48MHz

If the main clock is not running at 72MHz , for example if it was 71MHz, the USB will not work.

However the problem is more likely to be that the USB D- or D+ connection are not connected.

stevestrong
Posts: 630
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany

Re: USB descriptor request failed

Postby stevestrong » Wed Jan 04, 2017 9:11 pm

RogerClark wrote:However the problem is more likely to be that the USB D- or D+ connection are not connected.

That was it!
Although I have checked it via 5x magnifier, the soldering looked all right, but it wasn't.
Roger, you saved my evening!

User avatar
Rick Kimball
Posts: 706
Joined: Tue Apr 28, 2015 1:26 am
Location: Eastern NC, US
Contact:

Re: USB descriptor request failed - [SOLVED]

Postby Rick Kimball » Wed Jan 04, 2017 9:18 pm

stevestrong wrote:...
3 pieces BP arrived yesterday, today I tried them out, I was able to flash all with the stm32duino bootloader over STlink.
... many problems described ...


Can you provide a link to the seller so we can avoid them ?

-rick
-rick

stevestrong
Posts: 630
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany

Re: USB descriptor request failed - [SOLVED]

Postby stevestrong » Wed Jan 04, 2017 9:31 pm

Rick Kimball wrote:Can you provide a link to the seller so we can avoid them ?

Here it is, don't buy BluePills from this provider, unless you are ready to re-solder the USB connection lines:
https://www.aliexpress.com/item/1pcs-ST ... 95389.html
We should really make a black list of bad quality delivering sellers...

User avatar
mrburnette
Posts: 1759
Joined: Mon Apr 27, 2015 12:50 pm
Location: Greater Atlanta
Contact:

Re: USB descriptor request failed - [SOLVED]

Postby mrburnette » Thu Jan 05, 2017 1:26 am

stevestrong wrote:
Rick Kimball wrote:Can you provide a link to the seller so we can avoid them ?

Here it is, don't buy BluePills from this provider, unless you are ready to re-solder the USB connection lines:
https://www.aliexpress.com/item/1pcs-ST ... 95389.html
We should really make a black list of bad quality delivering sellers...



IMO, not wise. If members have a problem, it can be reported in a "I had an issue with..." thread. But to post a forum "black list" is not a proper forum function. Let's keep Roger out of jail legal hot water and just stick to experiences and specifics and avoid incriminating a specific seller.

Ray

User avatar
RogerClark
Posts: 5272
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: USB descriptor request failed - [SOLVED]

Postby RogerClark » Thu Jan 05, 2017 8:39 am

@stevstrong

I'm glad you fixed it.

I'm not sure if any vendor is any worse than the others. I doubt that any the Blue Pill vendors actually test the boards, so the problem probably lies with the manufacturer.

The Baite Maple mini's need to have the bootloader flashed somehow, which could be done before the MCU is soldered onto the board, but they may be flashing after...

Then at least they know the CPU must be getting power and also some of the GPIO connections, and they may go as far as checking that the LED flashes (but this is all conjecture)

However, I think most people's experience is that the Maple Mini from the official "Baite online store" is that they seem to be well made, and possibly tested before the leave the factory.


Return to “Generic STM32F103”

Who is online

Users browsing this forum: No registered users and 2 guests