2

I tried running a Pi Zero W on a 10000 mAh powerbank with qualcomm quickcharger capability (https://www.biltema.se/kontor---teknik/mobiltillbehor/powerpacks/powerpack-10000-mah-2000041816) . The first few times it worked ok, it ran for more than 24 hours but then the Pi suddenly stopped working. It appears to be fried:

  • No response when plugged into PC in usb cable (see link https://www.raspberrypi.org/forums/viewtopic.php?t=58151 bottom post on Zero)

  • When I plugg the pi into a regular pi-powersupply it smells of burning circuits and either the chip or the components between the chip and GPIO-pins quickly gets hot (I use a thermal camera but it is hard to judge exactly what component gets hot, its only a pinprick).

  • No reaction on entering a fresh SD card.

  • Finally, when I tried the PC trick again I used a thinner data USB cable and it got very hot, along with something sounding like a dying mouse. I think I may have fried the USB outlet on the PC too. If the damn pi wasn´t fried before it probably is now.

SO, probably quite fried!

Now the questions:

  1. Does anyone have similar experience?

  2. Could there be another reason why the Pi fried, e.g. bad power cable, short circuit, just a bad Pi?

Regarding #2: I dont think qualcom will jack upp power to 9 or 12 v without communication with the receiving device, then again, this is a cheaper powerbank.

I also checked the cable and it does not appear to be broken though it could still be an internal break and short circuit.

Also I used the old SD-card in a new Pi and it seems to be working. Will check again in a few hours.

Carl-Fredrik von Essen
  • 127
  • 2
  • 14
  • USB outlets in computers like your PC usually have a polyfuse that prevents permanent damage in the event of overcurrent. So it should recover by itself after several hours. – kwasmich Feb 05 '19 at 14:44
  • Thanks, your are right. The PC usb outlet was back to normal after a rest and restart. :)

    The Pi is supposed to have polyfuse too (thanks for giving me the term I was looking for) but in this case it was not enough to protect the pi.

     – Carl-Fredrik von Essen Feb 09 '19 at 08:13

5 Answers5

1

wished I had read your post before. I had one Zero W on a QC 3.0 powerbank for several months. Worked well, even for hours recording video. Then it abruptly died. I bought a replacement. That one lasted only a day. It was only then I started thinking and measuring. Seems like the Switching Regulator IC is gone on the first one. I haven't tested the second yet. I second your recommendation: no QC-powerbank if you want to keep your Pi. Too risky.

H

Henkki
  • 29
  • 2
  • Yup, that is sadly the conclusion. I have run Pi's on regular powerbanks (no quickcharge) for a couple of months now and experienced no problems. In fact they seem to have been more stable than regular power supplies but this may have to do with the programs I am using. Different versions seem to do different things to the pi's. – Carl-Fredrik von Essen Aug 28 '19 at 06:09
1

You seem to overgeneralize the problem you have with one particular power bank to all Quick Charge / Power Delivery supplies. The implementations of these standards differ, but they all follow the same train of thought:

  1. Any device plugged into a USB port initially receives 5V.

  2. If the device supports higher voltages, it can request them from a compatible supply.

With a correct implementation, it's not possible to fry a 5V device by plugging it into a Quick Charge supply. If there's a bug in the implementation, it's quite possible that even a 5V only supply delivers a higher voltage by mistake: guess why many cheap power supplies use capacitor on the output rated for 10 or even 16V? That's because they can go way over 5V for a short time under just the wrong conditions.

Indeed, a multi-voltage supply / power bank is more complex than a single-voltage one, so there's more chances that it might misbehave, especially if you get a really cheap one. Considering the fact that no Raspberry can benefit from a higher voltage, there is no good reason to use such device to power the Pi, even if it normally should be safe.

Dmitry Grigoryev
  • 27,928
  • 6
  • 53
  • 144
  • Well, thank you for clarifying that! It seems that this has happened to more than one person. Any idea what fried the Pi(s). That would probably be quite helpful to others considering to use such a powerbank for their projects. – Carl-Fredrik von Essen Apr 16 '20 at 14:39
1

Looking at the specs on the link above I'd say that it's gone over 5v for some reason and overloaded the pi. If you can smell burning when it's plugged in I'm afraid to say your Pi has died :(

I'd be wary of anything that can vary the voltage over 5v with mobile devices and especially Raspberry Pi's. Varying the amps won't affect it as the Pi will only take as many amps as it can use but the volts will fry it if it goes over 5v. In my experience chargers that can vary their amps/volts aren't very reliable at getting it correct with certain devices. They work fine with things that have matching quick charge capabilities etc. But with a generic device like a Pi they're very hit and miss.

rohtua
  • 459
  • 1
  • 4
  • 8
  • 1
    "...chargers that can vary their amps/volts aren't very reliable at getting it correct with certain devices" -> I believe this will cover just about all powerbanks, which are generally (explicitly or implicitly) sold for charging and not as a power supply. The problem is the voltage may drop too low, especially when a sudden current draw occurs (as with active live devices that are not simply recharging their own battery). However, it should not go over 5V unless it malfunctions. – goldilocks Feb 01 '19 at 16:09
  • :) I thought I was airing on the side of caution there. I was sure someone would correct me that a power bank was fine. Looking at the one listed at the above link it can provide a range of voltages apparently up to 12v 1.5a if it's burning it must have hit 12v or over 5 at least. If it was undervolt the pi would likely still work (although the sd card could be corrupted if it's switched during use a few times) – rohtua Feb 01 '19 at 16:23
  • I have gathered from various forums that the regular pi has a fuse that will reset in a few days, though not sure if it's true for the zero. Will check in a few days and post result. – Carl-Fredrik von Essen Feb 02 '19 at 23:00
  • I think I will play it safe and not use this powerbank again. I have used others before, 2.1 A but not quick charge. I will try again with a fresh sd card and make sure the cable is good.

    I did try the same sd card in another pi and it booted ok but did then freeze and give me trouble. Tests will continue.

     – Carl-Fredrik von Essen Feb 02 '19 at 23:03
0

The specs of the power banks write something like this: 5V/2.4A, 9V/2A, 12V/1.5A 9V and 12V fries any raspberry pi (Zero/1/2/3/4). Some power banks support low current usb gadgets, but they have quick charge mode too. Also the problem with the usb-c power for the raspberry pi 4 was that the raspberry foundation doesn't follow the standards of the USB. The power bank can't recognize the raspberry pi right.

Δημήτρης Δημητριάδης
  • 1
0

Update: I tried the Pi Zero with a regular powersupply and a fresh SD card but still nothing, except the processor gets very hot, 50+ degrees and rising fast.

That is one fried Pi!

With the comments above and some other stuff I have gathered on various forums I would say that the conclusion is to avoid powerbanks with quickcharge and variable voltage features.

Please let me know if anyone has tried and come upp with other results / conclusions.

Carl-Fredrik von Essen
  • 127
  • 2
  • 14
  • Please accept your own answer with a click on the tick on its left side. Only this will finish the question and it will not pop up again year for year. – Ingo Apr 15 '20 at 15:50