Seemingly random system shutdowns

[EpsilonLima]

Hi All, I'm having an issue with my pc where it will shut off all power and reboot, usually between 5 and 10 minutes of being on. I'm thinking its something hardware related as I have upgrade to windows 10 (via fresh install to format C:\ drive in the process) to try and fix this, however the issue is still apparent. So far the fixes I've tried are as follows;

• 
  Reinstall OS - formatted C:\ drive.
  Looked for errors in event viewer - there was only DNS errors (number 1004 or something similar) - and changed DNS settings to google's DNS server to resolve these.
  Enabled XMP profile in BIOS - Had this issue before about a year ago and this was the fix due to having 2 sticks of one RAM and two sticks of a another
  Checked all temps - all seem to be running at a fine temperature

There has not been any changes to my system's hardware recently, apart from an upgrade to my GPU to a 980ti 2 weeks ago, and no changes have been made to the BIOS.

This issue happens without the system being under load.

When I get back from work tonight I plan to swap test my hardware to see if I can isolate the issue, however other suggestions as to what the issue could be and a possibly resolution would be much appreciated.

System specs:
CPU: i5-4670k - factory OC'd to 4.20GHz - coolded with corsair h100i
GPU: Asus GTX 980ti - No OC
Mobo: Asus z87-k
RAM: 16Gb - 2 sticks (4Gb each) of corsair CMY8GX3M2A1600C9 in channel 2 and 2 sticks of (4Gb each) of Corsair CMX8GX3M2A1333C9 in channel 1 - XMP enabled but no settings changed
Storage: 1 SSD 120Gb for boot drive, 1Tb HDD and 500Gb HDD
PSU: corsair semi modular 750w

 

[MasterDell]

Running mismatched RAM is pretty sketchy. Remove a pair. If the issue persists, swap in the other pair. Don't try running in XMP with none-matching kits. The odds that it is going to work is quite low.

I recommend swapping in the previous GPU to see if the issue persists.

 

[EpsilonLima]

Running mismatched RAM is pretty sketchy. Remove a pair. If the issue persists, swap in the other pair. Don't try running in XMP with none-matching kits. The odds that it is going to work is quite low.

I recommend swapping in the previous GPU to see if the issue persists.

Hi, thanks for the quick response. This was going to be the first thing I tried later as I suspected this as the issue. The only thing is I've been running it for quite some time now without issues (I know its bad but as it worked so I always left it). I'll drop a response later as to the results of it.

 

[MasterDell]

Okay let me know

 

[EpsilonLima]

I've just got in from work and have removed one of the RAM kits. I'll run this all night and let you know if the issue returns.

 

[EpsilonLima]

Okay let me know

Right, I've just tested it once with one RAM kit, once with the other, and another test where I removed the GPU and used the onboard graphics. Unfortunately the same issue occurred each time.

Any ideas on what to try next?

 

[MasterDell]

Have you disabled all overclocks?

 

[EpsilonLima]

Have you disabled all overclocks?

I disabled the CPU turbo in the BIOS and its now been on for an hour with incident I haven't tested it under load yet, but all seems to be ok. - Thanks so much!

Any ideas why this could have just started happening when I've had the CPU for for a long time and run a very similar setup (only change being the GPU upgrade 2.5 weeks ago to a 980ti from 2 7970's in crossfire) ? Just interested so I could look for a resolution to get the cpu back to 4.20GHz instead of 3.40GHz using that turbo.

 

[MasterDell]

I honestly have no idea.. That's pretty weird if you ask me and if the symptoms come back I wouldn't be surprised.

 

[EpsilonLima]

I honestly have no idea.. That's pretty weird if you ask me and if the symptoms come back I wouldn't be surprised.

Yeah it is weird. I'll run another test tomorrow turning turbo back on to see if the causes the issue to return as to ensure it is that. I'll also try it under load to see how it handles that.

 

[MasterDell]

The 4670k only turbos up to 3.8GHz stock. Are you sure it wasn't overclocked? If it was actually overclocked, that would make a lot more sense.

 

[EpsilonLima]

I never OC'd it myself, but its possible it was factory over clocked.

 

[MasterDell]

I never OC'd it myself, but its possible it was factory over clocked.

Did you get this computer used or was it a prebuilt or something? That would definitely explain it

 

[EpsilonLima]

Nope, got all parts new and built myself.

 

[EpsilonLima]

Seem to have found the solution. I was speaking to a friend of mind about it today, and he told me about a similar issue where his motherboards surge protection had been doing a similar thing. I disabled the surge protection on my motherboard and re-enabled the turbo and it all seems good so far. I'm guessing as disabling turbo yesterday was causing less power to be drawn it might have also achieved the same thing.

Anyway it seems that the issue is down to some power surges, probably from the extension lead I'm using. I've got an order for a surge protected one arriving in a few days so hopefully that should sort it all. Thank you for your help yesterday.

 

[MasterDell]

That's actually really interesting.. I'm really glad you found your solution

 

[EpsilonLima]

That's actually really interesting.. I'm really glad you found your solution

Thanks, cheers for your help too, it helped to narrow down the issue for me.

 

[MasterDell]

Glad I could help