Minix Z83-4 - red light

[shianne]

Hello,

some day ago I have received a Minix Z83-4 (Intel Atom x5 Z8350-based nettop) that should be disposed of.
Reason: Only the red light lights up when a power supply is connected.
Hm, this seems to be a known problem: https://theminixforum.com/index.php?threads/what-do-the-colors-of-the-led-light-stand-for.21/
OK, I have no invoice and the device is out of warranty. And I am curious 

So I took the device apart and let my eyeballs do a optical inspection.

Well, an attempt at heat dissipation in the past.

-> U16-ausgebrannt-Beschaltung.jpg
It could be that it was too much heat 

After firing up the microburner  (Ersa H25 is too weak) I can see a little bit more.
But nothing what I did not know before by probing with the multimeter.

-> U16-entfernt-Beschaltung.jpg

Let's sketch a little bit:

-> Skizze-Eingangsschaltung_LQ.jpg

So U16 seems to be a main power switch controlled by - whatever...

And yes, I replaced it with a wire.

But I'm interested to know: What is the idea behind U16?

My first guess: It is the "main power switch" to reduce power consumption during power-off state.
But if U16 opens, the red light comes up (red light = The voltage outputs by the power adapter is not right.). So this can't be the idea behind.


Here a bad photo of the whole PCB top side (if someone is interested in):

-> PCB-BS-Foto_k.jpg


Edit: I'm too stupid to use the inline attachments, sorry. I've added the image names.

[shianne]

After doing some tests today I would say: Using a bridgewire instead of U16 works fine.

[shianne]

A short follow-up: The Minix Neo Z83-4 is working 24/7 as datalogger since my repair. So I guess whatever let burn U16 did not damage other parts.

The UEFI seems not to have an option of powering up the device after power is connected. So I built in a own circuit yesterday. First time I used 555 
The shown circuit may not work for short power failures because the C for time delay is not discharged quick enough.