cancel
Showing results for 
Search instead for 
Did you mean: 

Overheating Issue and Power Failure on STM32 When Connected via USB with Peripheral Power Pins

Kannn1chtpennen
Associate II

Hello community,

I’m encountering a challenging issue with my STM32 setup that I’m hoping to get some insights on. When I mount my microcontroller onto another PCB and connect it via USB, the board overheats near the USB connection and the LD3 (power LED) does not light up. I only soldered the 3V3, 5V, and GND pins on this secondary board, where 3V3 and 5V are used to power peripherals.

Here are the symptoms:

  • When connected to a PC via USB, the area near the USB connection on the extension board becomes very hot.
  • The power LED (LD3) on the microcontroller does not turn on.
  • Removing the extension board and powering the microcontroller directly via PC resolves the issue, and everything functions normally.

My initial guess is that there might be a power supply issue or a short circuit involving the 3V3 and 5V pins used to power peripherals, possibly causing insufficient power delivery or overheating due to improper load handling.

Could anyone provide guidance on how to properly manage power distribution in such a setup? Any suggestions on troubleshooting steps to isolate and resolve this overheating and power failure issue would be highly appreciated.

Thank you in advance for your time and help!


1.jpg

1 ACCEPTED SOLUTION

Accepted Solutions
Peter BENSCH
ST Employee

The picture is slowly becoming clearer.

  • Have you looked at the schematics of your NUCLEO-F334R8?
  • Do you understand where the 5V comes from that you are current drawing from at CN7.18 (+5V)?

Since you haven't specified the type for the DC-DC converter, no one can estimate the current it draws - please check it.

By the way, your zero-crossing detection with the LM2903 will not provide any pulses. The LM2903 is a comparator with an open collector output. Therefore, in your design the LED in the optocoupler will receive either GNDI (0V) or floating = 0V as input voltage from the LM2903.

You might be able to:

  • disconnect R3 from IC1A.1 and connect it to +5V_I,
  • then connect OK1.2 to IC1A.1
  • and connect OK1.1 to +5V_I, too.

However, such optocouplers do not want to see just 1mA of current, but 10mA, so you should change R3 from 2k to 180...220 ohms to provide sufficient current for the optocoupler.

In order to give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.

View solution in original post

5 REPLIES 5
Peter BENSCH
ST Employee

Welcome @Kannn1chtpennen, to the community!

At least I have no idea what you have connected where and where your LEDs are located. Why don't you make the problem clearer with a schematic?

Regards
/Peter

In order to give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.

Hallo Peter,

thank you for your suggestion. The circuit diagram of the PCB board is as follows, with 5V and 3.3V labeled as labelEIN and 3V3 respectively. The 5V will serve as the input voltage for the DC-DC converter, which will ultimately supply 5V DC to power the comparator. The 3.3V is connected to the right side of the optocoupler to output zero-crossing detection voltage at the VO point.

When the extension board is on the PCB board, I connected the CN1 of the extension board, which is the STLink port, to my computer. You can understand it this way: the PC powers the extension board, and the 3.3V and 5V ports on the extension board supply power to other components. The LED is the one lit up at the bottom of the first picture.

 

2.PNG

Sounds like you've already caused catastrophic damage..

Would have thought E5V would have been the better injection point, should probably review the NUCLEO schematic (under CAD Resources tab)

Tips, Buy me a coffee, or three.. PayPal Venmo
Up vote any posts that you find helpful, it shows what's working..
Peter BENSCH
ST Employee

The picture is slowly becoming clearer.

  • Have you looked at the schematics of your NUCLEO-F334R8?
  • Do you understand where the 5V comes from that you are current drawing from at CN7.18 (+5V)?

Since you haven't specified the type for the DC-DC converter, no one can estimate the current it draws - please check it.

By the way, your zero-crossing detection with the LM2903 will not provide any pulses. The LM2903 is a comparator with an open collector output. Therefore, in your design the LED in the optocoupler will receive either GNDI (0V) or floating = 0V as input voltage from the LM2903.

You might be able to:

  • disconnect R3 from IC1A.1 and connect it to +5V_I,
  • then connect OK1.2 to IC1A.1
  • and connect OK1.1 to +5V_I, too.

However, such optocouplers do not want to see just 1mA of current, but 10mA, so you should change R3 from 2k to 180...220 ohms to provide sufficient current for the optocoupler.

In order to give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.

Hallo Peter,

thank you for your valuable advice. Yes, the zero-cross detection circuit was missing a very crucial resistor between pins 1 and 8 of the comparator. It wasn't shown on the circuit diagram, but I did solder a 1.8k ohm resistor on my own board. I improved the current flowing into the optocoupler based on your suggestion, and the test results were good. The overheating issue was due to the wiring on my PCB being too close, leading to a short circuit, which is now resolved. Anyway, thank you very much for your help.