-1
\$\begingroup\$

The CP2102 was connected to the STM32 GPIO pins while USART1 was not configured (STM32 clean / not programmed). Is it possible that I destroyed my CP2102N via the GPIOs of the STM32 in that "initial state"? Would that be enough to damage the CP2102? Or why else doesn't my PC recognize the CP2102 at all?

schematic

\$\endgroup\$
11
  • 1
    \$\begingroup\$ It is possible as we don't know what you have done. Please show the schematics. But standard STM32 GPIO pins are inputs by default so it is very unlikely that it has damaged anything. Most likely the problem is at the USB chip or connections to PC. \$\endgroup\$ – Justme Oct 9 '20 at 16:06
  • 1
    \$\begingroup\$ I suspect the CP2102 won't work without VBus connected to something; spec-compliant USB devices basically have to at least monitor it. What did you hope to achieve by putting a resistor between the USB connector's VBus and ground? And of course with any QFN package you need to carefully inspect soldering while looking with magnification from an angle that lets you see into the inside corner where the metalization meets the board; looking from above alone is insufficient. \$\endgroup\$ – Chris Stratton Oct 9 '20 at 16:28
  • 1
    \$\begingroup\$ Don't STM32F4 chips have USB built-in? What is the reason to put an external USB serial chip onboard? \$\endgroup\$ – Justme Oct 9 '20 at 17:11
  • 1
    \$\begingroup\$ Restored the schematic you (accidentally I hope?) removed. Without it the question is meaningless. \$\endgroup\$ – Chris Stratton Oct 9 '20 at 19:31
  • 2
    \$\begingroup\$ Its great that you solved your problem, but stop deleting the schematic!!!. Without it your question is meaningless, so your repeated deletion of it is quite literally, sabotage. The schematic has now for the second time been restored to where it needs to be, further deletion attempts will be reported as abuse of the site. \$\endgroup\$ – Chris Stratton Oct 10 '20 at 20:52
2
\$\begingroup\$

You have not correctly wired the CP2102 for USB operation.

Among multiple possible issues:

  • the required connection of the VBus input is missing, which prevents the chip from placing the virtual enumeration resistor. To quote the data sheet "VBUS Sense Input. This pin should be connected to the VBUS signal of a USB network. A 5 V signal on this pin indicates a USB network connection."

  • you have feed REGIN with 3v3, but for a CP2102 the minimum to use that is 4 volts (the CP2109 apparently can accept down to 3 volts there)

  • the resistor you have on VBus of the USB connector likely accomplishes nothing. Hosts do not detect peripherals based on power consumption, but by the presence of a pullup resistor on a data line

Overall, you should review your design in comparison to manufacturer examples. Anything you do differently will need to be based on a very strong understanding of the chip, and you should consider if your belief that you could do that is actually mistaken.

As this is a QFN chip, also carefully inspect soldering under 10x magnification from an angle, so that you can see each of the solder fillets between the chip and the board. An eye loupe can work, or raise your microscope up and tilt the board at various angles in your hand. Also check trivial things such as that the orientation of the chip is correct, and that the pinout didn't get mixed up in the schematic design.

When the CP2102 part of the circuit is correct, you should get enumeration and ultimately (with correct drivers) a serial device, irrespective of just about anything but a power-to-ground short on the STM32 side of the board.

\$\endgroup\$
1
  • \$\begingroup\$ Thank you! I will try some of your points. Already checked the soldering, seems to be alright. \$\endgroup\$ – elian Oct 9 '20 at 21:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.