Raspberry Pi Forums

I got an rpi 4 B kit for my birthday recently, but so far it's been a rather frustrating journey.
The first rpi I got proved defective, so I had to return it. The second one turned on, then fried itself.
For the record, for each successive pi I'm only using the components of the kit I got: the rpi itself, an rpi-branded wall wart capable of delivering 5.1V / 3A, and the 16GB microSD card.

Combined with the first rpi, I must say that I already have quite a sour taste in my mouth: something that should have been < 30 minutes to set up has already taken up 2 weeks of my time.
So the electrical design is perhaps an important area to consider a lot more when designing the next rpi version.

Which brings me to the 3rd rpi 4 B I've received.
I turn it on, and it boots, but it will simply turn itself off after a little while (and leave the red/power LED blinking).
Here "a little while" is defined as "I can't even complete the config wizard that starts after the rpi has booted".

As-is, the device is unpredictable and nonfunctional, and so it's worse than useless.
Therefore I'd like to know why it turns off, and what I can do about it so I can actually start using it.

jjpe wrote: ↑

Thu Jun 18, 2020 4:09 pm

...
Which brings me to the 3rd rpi 4 B I've received.
I turn it on, and it boots, but it will simply turn itself off after a little while (and leave the red/power LED blinking).
...

This suggests (to me) that it is/was the (supplied) power supply that is/was faulty. The red PoWeR led blinking is one of the (transient) low-voltage warning mechanisms (the other is a lightning bolt icon on the display). Was anything, apart from a keyboard & mouse, plugged into the USB ports?
Trev.

Was anything, apart from a keyboard & mouse, plugged into the USB ports?

This is the list of things that were plugged in:
0. The microSD card
1. The microHDMI plug to see what I'm doing
2. keyboard (not even a mouse)
3. USB3.1 stick
4. the power plug

Note that the results are exactly the same if the USB key is not plugged in, so I can't really fault that for the results so far.

This suggests (to me) that it is/was the (supplied) power supply that is/was faulty.

Indeed, that has occurred to me. What trips me up though is that I (presumably) got a new adapter with each kit, and if this is the case then the question becomes: how is it that 3 adapters in a row can be that busted up without it being a bad batch (unlikely but possible) or design flaw?

I'm pretty convinced now that it is the PSU causing issues here.

I hooked up the rpi4B to my 5V / 2.1A powerbrick, and while some things are clearly still off limits since in total that's just 10.2W rather than 15.3W e.g. wifi @ 5GHz, the rpi as a whole is humming along now. I haven't seen the rpi lose power either while doing that.

So I guess I'll have to get a new PSU.

Were you plugging all 3 PSU into the same wall socket and is that different that the 5V / 2.1 A powerbrick is plugged into? It is possible you have a power socket that is bad. Are you rural or urban, old or new house/apt, and does your area suffer from brownouts and power failures?

While bad Pis occur (I have had one), they are tested in the factory and initial failure rate is very very low. 3 in a row does point to an external factor that is killing them.

jjpe wrote: ↑

Fri Jun 19, 2020 5:26 pm

Was anything, apart from a keyboard & mouse, plugged into the USB ports?

This is the list of things that were plugged in:
0. The microSD card
1. The microHDMI plug to see what I'm doing
2. keyboard (not even a mouse)
3. USB3.1 stick
4. the power plug

Note that the results are exactly the same if the USB key is not plugged in, so I can't really fault that for the results so far.

This suggests (to me) that it is/was the (supplied) power supply that is/was faulty.

Indeed, that has occurred to me. What trips me up though is that I (presumably) got a new adapter with each kit, and if this is the case then the question becomes: how is it that 3 adapters in a row can be that busted up without it being a bad batch (unlikely but possible) or design flaw?

OOC, were all 3 (kit supplied?) psu's of the same type/style? Were they "official" ones like those I've tested here: https://www.cpmspectrepi.uk/raspberry_p ... _Models.29 ?
(IIRC, some kits may include a PSU of similar spec. to the official ones but provided/branded by the kit supplier depending upon where one is. )
Trev.

No I know for sure now it's the PSU.

The powerbrick I mentioned is a 20000mAh battery essentially, so there's not wall socket to speak of there.
Thing is, it works with that without seeing symptoms of undervolting, and it also works with a socket brick from IKEA that has 4 regular sockets and 2 USB sockets that can output 5V/2.6A DC. In fact on both the rpi4B now runs as stable as a brick (pun intended, but I do mean the stable part).

The only variable left is the original, rpi branded PSU that came with the rpi4B itself.
My hypothesis on the previously burned-out rpi's is that the PSU was already bad and thus burned them, rather than the other way around.

OOC, were all 3 (kit supplied?) psu's of the same type/style?

The PSU is definitely rpi-branded. It's the model that delivers 5.1V/3A, is black with the rpi logo on top, and has a non-detachable usb-c cable.
In fact it is perfectly possible it is not just the same model, but the very same physical device. I sent an email to inquire about that, and expect an answer early next week.