Question Random power cycling ?

[Samuel hut]

Hey, so my PC has been randomly power cycling since the past two months.
I have replaced everything including the PSU, RAM, MOBO, SSD.

PC specs:
i7-8700
Mobo : Gigabyte B365 DS3h
Gskill Ripjaws 3600mhz DDR4 X1
WD black 512 M.2
Seagate 256 SSD
PSU: Coolermaster GX 750 cmstorm 80 plus bronze

I still encountered the same issue. The only thing left to replace is the processor itself. Before I do so, could it be causing the power cycling as everyone I asked say it's rare and moreover, I have no more money left to shell for a new processor.

Requesting answers as I'm exhausted trying to resolve this issue since the past 2 months and want to get it solved asap.
TIA

 

[Ralston18]

By "this" do you mean the SSD?

In any case I do not know if the SSD is the problem.

We have to have some knowledge of what SSD temperatures are allowable/acceptable and accurate measurements of the temperatures that are actually occurring.

If the SSD is performing within the specified temperature ranges then the SSD, on that basis, may not be the problem.

The WD diagnostic software tool may find other non-temperature related SSD problems .

 

[Samuel hut]

By "this" do you mean the SSD?

In any case I do not know if the SSD is the problem.

We have to have some knowledge of what SSD temperatures are allowable/acceptable and accurate measurements of the temperatures that are actually occurring.

If the SSD is performing within the specified temperature ranges then the SSD, on that basis, may not be the problem.

The WD diagnostic software tool may find other non-temperature related SSD problems .

Yes, I have shared the temperatures and the WD diagnostic software screenshot, it is reaching a maximum of 80°C, after 15 minutes of reading and writing, does this tell you anything?
The health of the m.2 is at 72%, could this be a problem?

 

[Ralston18]

72% Health - yes that would be a problem in my mind. If the measurement is truly health.

However, in my mind I would also be asking how was that 72% value determined? What measurements or other quantifiable values where used to calculate 72%?

Will that green line turn orange at some lower %? And then red at some even lower %?

How were the calculations made, weighted, averaged, and so forth? Specifics and details are important.

That said the 72% is not health per se. It is, per Post #49, "Life Remaining" which could be some simplistic measurement. Sort of like miles on a tire.

A tire might be good for 100 miles and has been driven for 28 miles so the remaining "life" could be 72 miles. That sort of measurement neglects what the road was like for the first 28 miles. Tire could be shredded.

Are you able to install that m.2 to into another known working computer without making any other changes and then run that health test?

Or try another known working (100% healthy) m.2 in your computer without making any other changes and run the health test?

 

[Samuel hut]

72% Health - yes that would be a problem in my mind. If the measurement is truly health.

However, in my mind I would also be asking how was that 72% value determined? What measurements or other quantifiable values where used to calculate 72%?

Will that green line turn orange at some lower %? And then red at some even lower %?

How were the calculations made, weighted, averaged, and so forth? Specifics and details are important.

That said the 72% is not health per se. It is, per Post #49, "Life Remaining" which could be some simplistic measurement. Sort of like miles on a tire.

A tire might be good for 100 miles and has been driven for 28 miles so the remaining "life" could be 72 miles. That sort of measurement neglects what the road was like for the first 28 miles. Tire could be shredded.

Are you able to install that m.2 to into another known working computer without making any other changes and then run that health test?

Or try another known working (100% healthy) m.2 in your computer without making any other changes and run the health test?

Hey, will do that. Something really weird happened today.
While I was checking the insides of the PC, I rested my arm on top of the cabinet where the AIO fans and radiator has been screwed onto.
As soon as I rested my arm, I felt a sharp shock, like an electric shock, I did this multiple times to be sure, it isn't harmful but it definitely stings like someone pierced your skin.
Here is the photo of the cabinet:

https://imgur.com/Jk85Xho

View: https://i.imgur.com/Jk85Xho.jpg

Could this be causing the power offs?
Side note: I have checked the standoffs as previously mentioned, they are correctly installed.

 

[Ralston18]

Shocks are a sign that something is very wrong.

Very likely a short circuit: if not the standoffs then maybe the I/O panel, or some case connection, around the edges of the fans/radiator.

Do not do any work inside or on that computer with the computer turned on or even plugged in.

You need to find someone who has a multi-meter and knows how to use it to find the short.

It could be intermittent. As in "rested my arm" you get shocked because the weight of your arm (or other circumstances) presses on the cabinet allowing current to flow where it should not be going.

I would look for pinched or kinked wires. Maybe bare conductor showing somewhere or a screw dug into a wire.

All around the fans/radiator shown in the photograph. Carefully inspect all cables and wires.

Again power off, unplugged.

Take a couple of additional photographs of the inside area on the other side of the photograph above.

Post accordingly.

Maybe someone will spot something astray. Some error of omission or commission.

 

[Samuel hut]

Shocks are a sign that something is very wrong.

Very likely a short circuit: if not the standoffs then maybe the I/O panel, or some case connection, around the edges of the fans/radiator.

Do not do any work inside or on that computer with the computer turned on or even plugged in.

You need to find someone who has a multi-meter and knows how to use it to find the short.

It could be intermittent. As in "rested my arm" you get shocked because the weight of your arm (or other circumstances) presses on the cabinet allowing current to flow where it should not be going.

I would look for pinched or kinked wires. Maybe bare conductor showing somewhere or a screw dug into a wire.

All around the fans/radiator shown in the photograph. Carefully inspect all cables and wires.

Again power off, unplugged.

Take a couple of additional photographs of the inside area on the other side of the photograph above.

Post accordingly.

Maybe someone will spot something astray. Some error of omission or commission.

Hey, so this electrocution problem was solved, turns out the socket in my room wasn't grounded.

It haven't found anyone of my friends with a m.2 supported Mobo, so it will take a while.
Could you please have someone take a look at the log files? Would be immensely helpful.

 

[Ralston18]

The log files are there for anyone to look at and read.

For the most part anyone who reads the logs and thus discovers or notices something would likely post accordingly.

Again, there have been so many changes made in the meantime tha the log recordings are overall now moot. Especially with Windows having been reinstalled twice.

= = = =

That said, now that the socket is hopefully properly grounded, my thought is a bit of a "do over".

Electrical problems can wreak all sorts of havoc.

Try running "sfc /scannow" and "dism" again.

Premise being that the electrical problem outlet problem continually caused recurring and varying file corruption problems.

Could be Windows files, driver files, applications, utilities, configuration files, and even data files. (You have backups - correct?)

Maintain the system (for now) as bare bones as possible.

Make no unnecessary changes or install any new apps.

Minimal software installed and minimal running in the background.

I recommend that that you clear out the logs in Reliability History and Event Viewer completely.

Then check the logs each and every time a problem occurs. Determine what the logs indicate with respect to error codes, patterns, etc..

When too many changes are made then it becomes difficult to establish what caused what and why.

You may need to just let some error continue a bit. Could resolve itself, reveal some pattern, or manifest in another manner.