Sudden PC problem.

Shamrock · 13 Oct 2017 at 10:20

Last night my PC suddenly decided to die and I'm trying to determine what exactly has failed. The components are about 2.5 years old.

I was browsing when Chrome suddenly became unresponsive, closely followed by the OS. I had to use the power button to turn it off. When I powered it back on it would turn on for about 2 seconds then immediately switch off. It kept trying to do this but never managing to post.

My flat mate had a spare PSU so I swapped in the connectors for the motherboard and gpu and it's still the same, so I think I'm ruling out the PSU being duff.

I removed all but one DIMM and the PC booted all the way into Windows for about 30s before powering off again. I tried all 4 of the DIMMs individually without much luck. The PC now sometimes stays on but with no display, sometimes posts and then dies and other times will power off after a couple of seconds without posting like before.

At this point I'm thinking dodgy mobo, but could it be the CPU? It doesn't seem like an overheating issue. It just powers off like someone pulled the plug.

Tonight I'm going to try resetting CMOS, disconnecting the front panel switches and using the onboard power button. Anything else I've not thought of?

Spec:
Intel i7 5820K (I think)
Motherboard is MSI X99S SLI Plus
Teamgroup 16GB 2400MHz DDR4

I've noticed some B-grade MSI X99A mobos on Overclockers so if it is the mobo I can get a relatively cheap replacement. I just wanted to be certain I'm replacing the right part!

orbitalwalsh · 13 Oct 2017 at 10:52

Shamrock said:
Last night my PC suddenly decided to die and I'm trying to determine what exactly has failed. The components are about 2.5 years old.

I was browsing when Chrome suddenly became unresponsive, closely followed by the OS. I had to use the power button to turn it off. When I powered it back on it would turn on for about 2 seconds then immediately switch off. It kept trying to do this but never managing to post.

My flat mate had a spare PSU so I swapped in the connectors for the motherboard and gpu and it's still the same, so I think I'm ruling out the PSU being duff.

I removed all but one DIMM and the PC booted all the way into Windows for about 30s before powering off again. I tried all 4 of the DIMMs individually without much luck. The PC now sometimes stays on but with no display, sometimes posts and then dies and other times will power off after a couple of seconds without posting like before.

At this point I'm thinking dodgy mobo, but could it be the CPU? It doesn't seem like an overheating issue. It just powers off like someone pulled the plug.

Tonight I'm going to try resetting CMOS, disconnecting the front panel switches and using the onboard power button. Anything else I've not thought of?

Spec:
Intel i7 5820K (I think)
Motherboard is MSI X99S SLI Plus
Teamgroup 16GB 2400MHz DDR4

I've noticed some B-grade MSI X99A mobos on Overclockers so if it is the mobo I can get a relatively cheap replacement. I just wanted to be certain I'm replacing the right part!

board should still be within RMA limit right ? 3 years ?

Shamrock · 13 Oct 2017 at 10:56

orbitalwalsh said:
board should still be within RMA limit right ? 3 years ?

You are right, I never realised MSI had a 3 year motherboard warranty.

orbitalwalsh · 13 Oct 2017 at 11:01

Shamrock said:
You are right, I never realised MSI had a 3 year motherboard warranty.

i think they do, every one else does - yep 3 years

If you bought from OCUK should be easy , or the other big place - dont think any MSI reps for mobo's here or if you can send board directly into them- think RMA center is EU based - but yeah as always try the store you got it from first

Shamrock · 13 Oct 2017 at 11:33

Yeah I got it from OcUK so I will contact them. Before I proceed though, does anything I've said suggest it may not be the mobo?

I've borrowed a spare gpu from work I can try just to confirm it's not the graphics card.

orbitalwalsh · 13 Oct 2017 at 13:12

Shamrock said:
Yeah I got it from OcUK so I will contact them. Before I proceed though, does anything I've said suggest it may not be the mobo?

I've borrowed a spare gpu from work I can try just to confirm it's not the graphics card.

i think if it was CPU it wouldnt even post -

they might go through a check list with you anyways to confirm steps then get you to send it in

Danny75 · 13 Oct 2017 at 14:05

There was a big Windows update this week. Not sure it sounds like your issue to be honest but perhaps worth mentioning.

http://www.pcgamer.com/windows-10-update-sends-some-pcs-into-a-bsod-boot-loop/

Have you tried disconnecting all drives and seeing if BIOS can be accessed like that?

Shamrock · 13 Oct 2017 at 14:15

I did wonder but I'm running Windows 8.1 and I haven't experienced an actual BSOD.

I haven't physically disconnected all SATA devices yet but when I tried the second PSU the only things getting power were GPU and mobo so I thought that was as good as disconnecting the other devices (SSD, HDD, sound card and blu-ray drive).

Shamrock · 16 Oct 2017 at 08:31

I've been able to use the PC over the weekend, I'm not sure why.

It's now sort of functioning with 2 DIMMs but if I put the other two in it fails to boot again. Before it was not working even with one stick installed! Whilst it's nice that I could use it, it's frustrating not knowing what/where exactly the problem is. It still stinks of the mobo to me. :confused:

ED209 · 16 Oct 2017 at 22:55

Could be faulty ram, do you have another PC you could test the ram in or run memtest on them?

Born2BSlow · 16 Oct 2017 at 23:17

I’ve been having problems too since the weekend when it kept crashing. I reseated the AIO cooler, reset the bios to default then re-applied my overclock. Tonight I got the WHEA UNCORRECTABLE ERROR, now I have to check all my bloody drivers and hardware to find out what’s causing the problem. I don’t really want take off the cpu overclock because performance is pretty pants without it...

Danny75 · 16 Oct 2017 at 23:52

Born2BSlow said:
WHEA UNCORRECTABLE ERROR

99% that is due to OC. If you temporarily remove it, you'll find out soon enough anyway? Instead of spending a lot of time checking drivers and what's not.

Quartz · 17 Oct 2017 at 00:00

Shamrock said:
At this point I'm thinking dodgy mobo, but could it be the CPU?

Tell us about your CPU cooler. Is it an AIO? If so, have you checked that there's still fluid in it?

Shamrock · 17 Oct 2017 at 08:13

Quartz said:
Tell us about your CPU cooler. Is it an AIO? If so, have you checked that there's still fluid in it?

It's one of the Corsair Hydro series. CPU is running at stock.

I've not really checked the cooler but assumed it was all fine based on the temps seeming normal. I've actually been playing games the last couple of days without issue but soon as I put the full 16GB in I'm sure it'll break again. It's odd how nothing's really changed but I'm not getting the instant power cut style issues from last week.

Part of me just wants to ignore it as it's working for now, a month long RMA process is unappealing and I can live with 8GB instead of 16GB. The other part of me can't cope with the thought of an unresolved hardware issue somewhere in my system. :mad:

Born2BSlow · 17 Oct 2017 at 10:19

Danny75 said:
99% that is due to OC. If you temporarily remove it, you'll find out soon enough anyway? Instead of spending a lot of time checking drivers and what's not.

I know...I'm gutted because it has been running so smoothly for months (since I rescued it from my own incompetence). Any idea what I should be looking out for, is there a chance it could be a driver issue? I'd like to overclock, I do have the other paired chip I could put in and see if it's the CPU that's fried. I was running 1.3v, but had reduced it to 1.275v after the crashes at the weekend. It seemed to be stable again until last night.

I don't know if this means anything, but it hasn't crashed under load, it just seems to be when it's idling...