1

Today, I've experienced kernel panics every time I'd execute a certain python script. I was a little bit surprised, since I've executed more heavy scripts before, but doing my research, I found out, that the Pi detected an undervoltage at boot time (at every boot). The warning would go away (no yellow flash in the corner, which I'd experience with another microUSB cable), but to be sure, that the Python script doesn't trigger undervoltage again, I set out to measure 5V-GND with my digital multimeter.

My multimeter has some very small measurement tips, which don't stick well to the GPIO pins, so I think, I might have broken something. In the beginning, not understanding the pin assignment correcctly, I measured between random pins. Then I found the correct pins 2 and 6. By that time, the Pi had already rebooted a few times, but it still came on again every time.

At some point during measurement, the Pi shut down and wouldn't come on again. Only the red LED is constantly on (but the ACT LED shows no activity at all). I've tried disconnecting any peripherals (other tghan the power supply, obviously) from the Pi and tested several other SD cards (including a backup that I've just made yesterday from my main SD card). No success. :-(

If I interpret the schematics correctly, the red PWR LED shouldn't light up at all, if the fuse F1 were triggered, right? However, the controller doesn't get hot when powering the PI, either. I read that this could be the case, if 5V were supplied to a 3.3V GPIO pin, and the controller had consequently been fried.

Any ideas what else I could try or measure? I'll set up as replacement Pi tomorrow, but I'd be really glad, if my original Pi can be revived.

6
  • Do you have a specific question? It would help to make it more obvious. I suspect that your Pi is broken from a power short. In that area of the GPIO header, there is 5v, 3v3, ground, and gpio's. It's a bad idea to short them. Rather than put a probe directly on the pins, it would have been safer to use short jumper wires to bring the two pins out to the meter so they can be separated and it's less likely to short things out. And to double-check on the pin assignments.
    – NomadMaker
    Commented Jul 24, 2018 at 13:46
  • I'd like to know if there is an easy way to find out what is the specific problem, so that I could then figure out if it can be fixed or not. Aside from the question, shorting 5V and GND by a measurement probe shouldn't do anything to the Pi, since the high current will (practically) only flow though the probe, right?
    – Hemanti
    Commented Jul 24, 2018 at 14:35
  • Where do you think the high current is before it passes between the probes? It has to travel through small PCB traces which are not designed for high current and may have melted.
    – joan
    Commented Jul 24, 2018 at 14:37
  • Didn't think of that. :-( I guess, that's what the fuse is for, just it's not always quick enough?
    – Hemanti
    Commented Jul 24, 2018 at 14:39
  • The polyfuse may have triggered. If you are lucky it will recover after a day or so of being cool. The fuse is really to prevent enough current flowing to start a fire, its aim is not to protect the Pi itself.
    – joan
    Commented Jul 24, 2018 at 14:49

1 Answer 1

0

I did this to my Raspberry Pi last week, actually. I found forums which discuss the reasons and causes, but this is what happened to my board, and I am confident, that by your description, happened to your board too. If you accidentally short the 3.3V rail to the 5.0V, it will essentially place 5.0V across all of the computer components which run off of 3.3V, frying them. I'm not sure why they are located right next to each other on the pins.

Sorry to say, the polyfuse does not guard against this nature of a short, and there is no way to fix it, that I know of, short of replacing the fired components.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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