4 Month Old P.C shut down and rebooted twice today

Sarz91

4 month old pc that I built up last year. In the 4 months I've had it it's shut down and rebooted of its own accord 2 or 3 times. Today however, in the space of about 2-3 hours it shut down and booted back up twice. There are no warnings displayed or error messages. The BIOS displays the correct date and time. Apart from it possibly being PSU related I'm unsure as to what the problem is. It's not my CPU or mobo (it might be but I really don't want it to be. Ignorance is bliss right?).

The specs for the pc are as follows:
RAM-8GB-Kit G.Skill PC3-10667U CL9
Mobo-ASRock Z77 Pro3
CPU-Intel Core i5-3570K
HDD-Seagate Barracuda 7200 1000GB
SSD-Samsung 830
GPU-None
PSU-Super-Flower Amazon 80Plus 550W

Any help would be greatly appreciated.

Thanks in advance.

Deano12345

Try run memtest on the memory, HDTune on the HDD, and get RealTemp and let me know what your CPU temps are like. Do the crashes happen at random or when your doing something in particular (gaming etc)

klub44

Hi

Do few check and post results:
Enter BIOS -> Hardware monitor and checks voltages on lines 3.3v, 5v and 12v.
Download OCCT and test each components separately:
CPU
RAM
GPU,
min. 30 min each under full stress.
Check also voltages in OCCT under full stress, and CPU and GPU temp.

Mostly faulty component will cause crash during heavy load, if you fail on first test, check other components.

CPU, RAM and GPU malfunctions likes to post blue screen before crash. Power problems won't.

Guffy

This type of thing happened to my old pc. It must have gotten a bang during a house move and the fan had slipped off the CPU causing it to overheat and shut off. If its still under warranty I advise you to just bring it back to the shop in case opening it voids it. If not open her up and check all your fans are fitted correctly.


Edit: it used to happen to me playing fm which wasn't too taxing on my specs

Sarz91

Deano12345 wrote: »

Try run memtest on the memory, HDTune on the HDD, and get RealTemp and let me know what your CPU temps are like. Do the crashes happen at random or when your doing something in particular (gaming etc)

I was watching a few stream when it crashed. Memtest reported no errors. Idle CPU temps 24 degrees.

Deano12345

Sarz91 wrote: »

I was watching a few stream when it crashed. Memtest reported no errors. Idle CPU temps 24 degrees.

Really does sound like the PSU could be on the way out, do you have another you could try ?

Sarz91

Deano12345 wrote: »

Really does sound like the PSU could be on the way out, do you have another you could try ?

No. I'll be ordering a new GPU in the next month or so I'll prob just pick one up then. Cheers for the help. I ran OCCT for 30 minutes as well and everything seemed fine.

Arciphel

You could also try something like running the torture test on Prime 95 or check the graphics card by running MSI Kombustor. Good luck.

ED E

Yeah, try furmark it(Or kombustor, same thing).

Whats your case airlfow like? The integrated graphics are likely passively cooled.

Sarz91

ED E wrote: »

Yeah, try furmark it(Or kombustor, same thing).

Whats your case airlfow like? The integrated graphics are likely passively cooled.

Two 120mm exhaust fans on top. One 200mm intake at the front. One 120mm intake on the side. The mistake that I never bothered fixing was the CPU fan pulls air in from the rear of the case (no fan sitting in the rear) and pushes air into the case. I haven't really been bothered trying to fix it as it shouldn't make a huge difference to the airflow.

ED E

You should be ok, but if your CPU flow was going against case flow there could be an "eye of the storm" over the mobo.

Run the test for 20 mins in any case and see.


Had a bug with the aul fellas machine there a few months back. Was running an old green team card powering two panels for just basic work stuff. Nothing graphics intensive. He had regular BSODs but passed all stress tests. Eventually replaced the card and no problems since, so even if it passes furmark it might still be at fault :S

Sarz91

ED E wrote: »

You should be ok, but if your CPU flow was going against case flow there could be an "eye of the storm" over the mobo.

Run the test for 20 mins in any case and see.


Had a bug with the aul fellas machine there a few months back. Was running an old green team card powering two panels for just basic work stuff. Nothing graphics intensive. He had regular BSODs but passed all stress tests. Eventually replaced the card and no problems since, so even if it passes furmark it might still be at fault :S

Should I run it even if I don't actually have a stand alone card? I sold my gtx 670 a while back. Should I still use Furmark on the integrated intel hd 4000 graphics?

Lu Tze

Those Amazon PSUs are supposed to be quite reliable. I have a spare PSU here (Ballsbridge) if you are anywhere nearby you are welcome to borrow to test. its a 550W High Power (not the most reputable) but should do to test if the PSU is the cause. Was working fine the last time it was used (powering core 2 duo and a HD4870)

ED E

Yep, just watch the temps as you go and dont let it get excessively hot.

Sarz91

Lu Tze wrote: »

Those Amazon PSUs are supposed to be quite reliable. I have a spare PSU here (Ballsbridge) if you are anywhere nearby you are welcome to borrow to test. its a 550W High Power (not the most reputable) but should do to test if the PSU is the cause. Was working fine the last time it was used (powering core 2 duo and a HD4870)

Thanks very much for the offer but I honestly don't have the time with college. It's a pretty hectic final year (making me wonder why I just didn't drop out a year ago). Also, its an excuse to get a modular PSU haha.

Lu Tze

Sarz91 wrote: »

Thanks very much for the offer but I honestly don't have the time with college. It's a pretty hectic final year (making me wonder why I just didn't drop out a year ago). Also, its an excuse to get a modular PSU haha.

Fair enough, it will be a real pain in the ass if its not a problem with the PSU though! :pac:

Sarz91

ED E wrote: »

Yep, just watch the temps as you go and dont let it get excessively hot.

Ok cool it might be worth noting that there are no blue screens or any screen for that matter when the computer shuts down. It just as if I've shut it down and a few seconds later turned it back on.

klub44

Sarz91 wrote: »

Ok cool it might be worth noting that there are no blue screens or any screen for that matter when the computer shuts down. It just as if I've shut it down and a few seconds later turned it back on.

Did you checked voltages in idle and under full load?
Give it a go, better check now, then fry your motherboard someday
Cheers

Sarz91

My CPU voltages are in the test below. I just took a picture of the occt test and posted it.

klub44

Sarz91 wrote: »

My CPU voltages are in the test below. I just took a picture of the occt test and posted it.

That's only cpu and 3.3v line voltage. They are ok. Older versions of occt I've used also showed 5v and 12v lines. You may check them in BIOS (idle only), any outstanding values (mostly over 0.4v and more) may cause computer shutdowns.

Sarz91

klub44 wrote: »

That's only cpu and 3.3v line voltage. They are ok. Older versions of occt I've used also showed 5v and 12v lines. You may check them in BIOS (idle only), any outstanding values (mostly over 0.4v and more) may cause computer shutdowns.

5V was at 5.112V and 12V was at 12.302V. They seem ok?

klub44

These values are ok. No idea what to check next. All components works fine. Maybe that some Motherboard issue? Have you got latest BIOS upgraded?