5

I have a Model B Raspi and - shame on me - overpowered it with about 9V. And to make it worse I powered it using the test points TP1 and TP2, avoiding the polyfuse. See Raspberry Pi Model B Schematics.

It booted up fine. I think my power supply wasn't that powerful enough to blow the SoC while it was doing work and the rest of the power was going through the suppressor diode D17 but this all is speculation.

When I was done for the day I was too lazy to pull the power cord and only shut down the Pi via command line.

The next morning the diode D17 was so hot that the case above deformed. After letting it cool off the Pi didn't boot any more. The red PWR LED is on while the green ACT LED is glowing dimly.

I removed the diode to be sure that it didn't short the power flow. It was broken anyway. A replacement is ordered but if the Pi is broken there is no need to solder a new diode on the board.

I already tested the polyfuse F3 as described here: Troubleshooting power problems.The voltage drop is less than 0.1V. But as the power isn't supplied through the MicroUSB that has nothing to say.

I tested my SD card with a friends Pi to exclude it from the list of possible errors. I also put a fresh image on it (Arch but that doesn't matter).

Now the question: Is there a way to

a) find out if the SoC is definitively broken?

b) load the GPU firmware code onto the GPU again (as it seems to me that the first stage bootloader is not executed as described here: How does Raspberry Pi boot?)

c) revive the Pi in any other way? Perhaps using a JTAG device?

EDIT: I checked all power regulators: RG1 is providing 1.80V, RG2 is providing 3.29V, RG3 is providing 2.50V. So it seems that it is not a power issue.

Community
  • 1
Uwe
  • 51
  • 1
  • 3
  • 1
    Might be cheaper to just buy a new one – nagyben Feb 26 '14 at 00:04
  • 1
    From an economical point of view you might be right. I just don't like it to throw away a piece of hardware from which I don't know definitively that it is broken. And where is the learning effect in buying a new one only because I don't understand why it is not working? – Uwe Feb 26 '14 at 06:51
  • I know how you feel - I always prefer to repair equipment vs. replacing it. Normally it's cheaper this way and that's one of the reasons I do it but in your case it might be more trouble than it's worth – nagyben Feb 26 '14 at 09:40
  • Looking at the schematic, 5V doesn't go directly to the SoC. You should check the 3.3v and 1.8v voltage regulators. – Gerben Feb 26 '14 at 11:24
  • On the first page of the schematics the SoC BCM2835 is printed four times. On the first (upper middle) and on the second (upper right corner) it is connected to 1.8V, on the third (on the right) it is connected to 3.3V. But on the fourth (lower middle) it is also connected to 5.0V directly (and to 3.3V). On the second page the SoC is connected to 1.8V, 2.5V, 3.3V and to 5V through a 1800 Ohm pull-up resistor. – Uwe Feb 26 '14 at 21:47

1 Answers1

5

Hoarding is bad.

This is most definitely a bricked pi. The heat generated indicates this has a short, and is totally bricked.

I've found that repairing even the power capacitor, or attempting to revive a dead pi, can have disastrous results. For example, multiple bricked/burned SD cards. Another example, I've had multiple bad power supplies ruin brand new pi in under a month. Hence, the new indicators for power issues in the Rev2.

In the end trying to 'repair' a dirt cheap computer, or it's power supply issues is the ultimate in hoarding exercises.

I recommend marking it as defective with a sharpie/paint and taking it to work/school/bus station for show and tell. Then leave it there and forget about it.

  • 3
    +1 Also, this RPi is clearly a fire hazard. If it got hot enough to deform the case it was housed in, the RPi needs to be recycled. End of story. – Jacobm001 Jul 15 '15 at 19:33
  • Good point, added that to the answer. –  Jul 17 '15 at 00:52