0

I know that anything more than 5v touches any GPIO then it's dead but I was changing the wiring and accidentally plugged a 12v 2A (for motor supply) connector to a female connector which leads to GPIO pins 20 and 21 for a very short amount of time, maybe a second; ofcourse the PI tripped and got reset immediately.

Now, I didn't let it power on for a minute and powered it on again. It crashed several times till the splashscreen showing it's logo and then booted to the desktop finally after few crashes.

I immediately noticed the cpu load percentage on top right being constantly above 25%(unusual) and the power(lightning) logo in the top right stayed there constantly as well so I touched the pi and it was too hot to touch so I immediately switched it off and haven't turned it on since.

I looked up the internet and there's mention of a fuse which resets over time but observing the behaviour, my question is, if it boot up to desktop, is it fine and will it reset?

  • Always power down if you meddle with cables and you know that some parts will be damaged if you improperly connect. Double / triple check before powering on.

    Also, always power down if working with voltages of 50 V and above.

    – Dr_Bunsen Nov 01 '19 at 12:54
  • @BLOODHOUND, Ah, let me see. If Rpi can still boot, but only with a yellow lightning bolt, that means it is only half dead. Yellow lightning means low power. My brainstorming wild guess #1 is (1) Rpi's poly fuse partially blown out (needs a couple of seconds to completely blown), so resistance changes from zero ohm to perhaps 0.2 ohm, causing a voltage drop from micro USB connector to Rpi 5V rail. You can wait one or more days and see if anything improves. In the mean time, you can try use a multi-meter to off power check the poly fuse resistance, is it self healing etc. / to continue, ... – tlfong01 Nov 02 '19 at 02:15
  • @BLOODHOUND, If you found polyfuse more than half dead, or you don't wish to wait, and are going to throw Rpi to the rubbish bin anyway, you can consider replacing polyfuse. But you need to be a ninja DIYer to do ployfuse surgery. A couple of days ago I made it and it took 10 min for the operation: "Rpi3B+ micro USB connector got 15V power. Is it dead? Can I save it?": https://raspberrypi.stackexchange.com/questions/104596/rpi3b-micro-usb-connector-got-15v-power-is-it-dead-can-i-save-it (Don't worry that much: 15V for one second is OK for the tougher than you think polyufuse guy! :). – tlfong01 Nov 02 '19 at 02:29
  • @BLOODHOUND, My polyfuse operation record: https://imgur.com/gallery/j4T6g2S. The Rpi polyfuse is hacker friendly. You don't need a sharp point soldering iron, just apply "plenty of" cheap, high tin, toxic solder, on both sides of the "big" polyfuse, and it will swim out of the molten solder pond by it self. You don't even need any magnifying glass to look closer. – tlfong01 Nov 02 '19 at 02:37
  • It wasn't the polyfuse as after letting it sit for 24 hours it gave the same response of overheating. After testing a bit more I found that GPIO wasn't functioning any good anymore. But I've found that I can still use this pi for I2C or such communications well and for other linux stuff too after adding a spare little heatsink! Guess I won't discard it yet, even if it dies I'll probably perform a surgery cause whats dead may never die – BLOODHOUND Nov 05 '19 at 12:49

1 Answers1

1

A second is not a short period of time for a computer which executes millions of instructions per second.

Neither is it a short period of time to damage hardware designed to handle no more than 3V3.

You have probably destroyed the directly connected GPIO. The damage will likely spread throughout the SoC over time.

Time to buy a new Pi.

The polyfuse isn't really there to protect the Pi (and would not have helped in this case anyhow as you bypassed the polyfuse). The polyfuse is there to prevent a current large enough to start a fire flowing through the Pi.

joan
  • 71,024
  • 5
  • 73
  • 106
  • You're right. It's just that the pi booted up fine except it was heating too much. I will still try and power it on see if the temperature rises again – BLOODHOUND Nov 01 '19 at 12:21
  • 1
    The SoC heating up very fast is a symptom of a failing SoC. – joan Nov 01 '19 at 12:24
  • You're right, Joan! I was just so curious after reading the stories of people frying their pis and not even getting anything else than the red led from the pi let alone a video signal and was curious that even after shorting a 12v line to one of the gpio the pi was working fine except the temperature. However, I've noticed increased errors and more heat and not all GPIOs work anymore consistently without restarts. My new Pi is on the way :D Thanks for your response! – BLOODHOUND Nov 05 '19 at 12:42