PC Hard Restarts (youtube/chrome)

Discussion in 'Troubleshooting Help' started by xHardz, Jan 11, 2017.

  1. xHardz

    xHardz Member

    Joined:
    Aug 22, 2008
    Messages:
    343
    Hi Guys,

    I7 6700K (Just over 12 mths, purchased from forums)
    Asrock Z170 Extreme4 (PCCG 12mths)
    Corsair LPX 2400MHZ C14 16GB (PCCG 12mths)
    SilverStone Strider Plus 600W ST60F-PS (PCCG 24mths)
    Noctau 9UB-SE2
    WF3 R9 290X (Forums - 2.5 years?)
    CM690 Orig.

    Windows 10

    G110 + G502

    After pulling my hair out for the last couple of weeks I am seeking some help/ideas as to what could be going wrong with my PC.

    The restarts are a hard restart... No Bluescreen or PC freeze. Power gone, then instantly starting back up again.

    When I first rebuilt the PC and finally got Windows 10 installed it seemed fine but these random restarts happened. Never while gaming, mostly while it appeared to sit idle or when I was just watching some YouTube. After the usual drivers/bios/updates etc they seemed to just fade away with no issues for months and months, even with my PC running for weeks at a time.

    Now I know this is going to sound like an OVERHEATING issue, as I built it in January last year and this is when I was having the issues and during the cooler months it was fine. I hope it is something so simple but just to try and make this happen i have run;

    Intel Burn Test w/ 12+ GB for 3 hours - Max temp was around 70C
    MemTest (from windows) x 5 for 3 hours - No errors or temp issues
    R6:Siege / ARMA3 - Can crack out a 3-4 hour gaming session without issue. - After gaming sometimes it is fine till the next day, sometimes I open chrome to start browsing and bam, restart.

    • I have taken both sides off the PC, while it is cooler, it doesn't appear any more stable.
    • After the return of these issues just over a week ago I updated all the MoBo drivers as there were some newer ones released since I had last check, Updated my BIOS from 2.x to 7.0, used AMD driver cleaner and reinstalled the newest drivers WITHOUT ReLive.
    • Stock settings on CPU, MoBo settings for RAM (2133) all volts at stock.
    • Unplugged my 2 extra HDD's
    • Unplugged USB hub + Controllers
    • Disabled hardware acceleration within chrome
    • Tested Firefox
    • Tested Chrome Canary
    • Manually reinstalled flash in chrome

    I have read forum post after forum post with 100's of different 'solutions' but so far haven't found anything that works. Sometimes I feel as if I have had a breakthrough only for it to happen as soon as I get a few minutes into a video.

    Nearly all the parts are still in warranty, but without ordering replacements it will be hard to discover if something is faulty or if its a software issue.
    Its not high on my list, but reinstalling windows 10 or even windows 7 could be something to test.
    Considering I am nearly to the point where I am willing to order a new PC and throw this one outside I need some help <3

    Any more info let me know.

    Thanks,
    Matt.
     
  2. ae00711

    ae00711 Member

    Joined:
    Apr 9, 2013
    Messages:
    1,123
    few other things to try:

    - no/different dGPU
    - re-install browser(s)
    - re-install windows: start from scratch/fresh
     
  3. cesario

    cesario Member

    Joined:
    Jun 15, 2009
    Messages:
    283
    Have you taken a look at the event viewer yet? It may of record the restart instances and could provide some useful info in there.
    I would also suggest disabling automatic restarts to see if any messages appear before it restarts thats too quick for the human eye.
     
  4. OP
    OP
    xHardz

    xHardz Member

    Joined:
    Aug 22, 2008
    Messages:
    343

    Yeah, I diagnosed a few things from there but the restart is only listed as 'Kernel Power'.

    The automatic restart was disabled from the get go.

    Different browsers doesn't help.
    Haven't tried the iGPU yet. May have to give it a go.
    A fresh install was on the cards, but I think it might be more of a hardware issue as the restarts are as if someone flicked the power switch. No bluesrceen/error events
     
  5. pwlm1975

    pwlm1975 Member

    Joined:
    Apr 18, 2016
    Messages:
    102
    Location:
    Logan, Qld
    If you haven't already, remove any overclock and see if it fixes it. These errors apply to hard resets, which genrally happen with a power issue. Coild be an undervolt by the PSU or not enough voltage in your overclock setting.

    Also check that any BCLK overclock hasn't jacked the RAM to an unsustainable speed.
     
  6. kEv_

    kEv_ Member

    Joined:
    May 5, 2012
    Messages:
    497
    Location:
    Adelaide
    error 41?

    i had exactly the same issue, would never crash under load, but surf the web or even leave it alone for 10 minutes and the sucker would restart

    never found a definitive fix, like you tried pretty much everything, including full clean installs, of win 7, i dont think its done it since i've been on 10

    stupid thing is, it just stopped doing it all by itself, i'm sorry that doesnt really help, but at least you know you're not alone...

    GL!!
     
  7. zero_velocity

    zero_velocity Member

    Joined:
    Sep 16, 2010
    Messages:
    2,398
    Location:
    QLD
    Check in your BIOS for your PSU voltage settings, specifically your 12V, 5V and 3.3V rails.

    I just (today) finally fixed the 'random' lock ups and every time I enter a game it would go to a black screen (but PC still running) requiring a hard reset.

    Every issue gone with a new PSU.

    3.3V line was showing as 2.9V
     
  8. sanousie

    sanousie Member

    Joined:
    Apr 12, 2003
    Messages:
    2,128
    Location:
    Sydney
    I still dont get why people purchase new tech and expect it to run flawlessly on older gen psu's.

    The new gen psu's have different amps on their rails compared to older gen high quality psu's.

    I think there should be a sticky about this, in the last couple of months there have been multiple threads of users having restart/hard lock issue/soundlooping locks etc and all have been solved with a newer gen psu's.
     
  9. zero_velocity

    zero_velocity Member

    Joined:
    Sep 16, 2010
    Messages:
    2,398
    Location:
    QLD
    24 mths old PSU is still considered 'new gen' by that logic. Old gen PSUs would be what... 10 yr old PSU's now? Older?
     
  10. pwlm1975

    pwlm1975 Member

    Joined:
    Apr 18, 2016
    Messages:
    102
    Location:
    Logan, Qld
    Sometimes you just get a dud part, too. Still no guarantee that the PSU is the culprit, but it's a prime suspect in my simple mind.
     
  11. sanousie

    sanousie Member

    Joined:
    Apr 12, 2003
    Messages:
    2,128
    Location:
    Sydney
    New gen have multiple rails as opposed to the old gen mate.

    Cannot compare a 10 year old psu to a psu made in say 2016, they have multiple rails for a reason. E.g. with multiple rails you get less noise on different components, which equates to a higher overall bandwidth, hence new gen.
    cheers.
     
    Last edited: Jan 13, 2017
  12. pwlm1975

    pwlm1975 Member

    Joined:
    Apr 18, 2016
    Messages:
    102
    Location:
    Logan, Qld
    Pointless, off topic discussion as the PSU is 2 years old, not 10, and I don't think the design of the 12v rails is at fault, rather the PSU itself.
     
  13. zero_velocity

    zero_velocity Member

    Joined:
    Sep 16, 2010
    Messages:
    2,398
    Location:
    QLD
    Yeah, I know exactly what Im talking about. You seem to be under the impression a 2 year old PSU is 'old gen'. As I clarified, old gen would be in the vicinity of 10+ years, where rails were common.

    My 4 year old corsair HX1050 has 6x 12V rails, except, its a common 1000W 12V supply.... This technology hasn't changed even by todays standards, so before you go flapping your gums on something you seem to know little about, do some research.

    This ^ OP needs to check the Voltage outputs of his rails. Components die, its a fact of life. Age considered or not the whole PC is no more than 3 years old, so all relatively 'new tech'.

    Start with the fundamentals. Check voltages, Check plugs/connections of everything, see if that improves. Remove any connection extension/braided cables etc
     
  14. [SweN]

    [SweN] Member

    Joined:
    Sep 2, 2002
    Messages:
    2,417
    +1 from me.

    Nearly every issue I've had over the years was came down to PSU.
     
  15. sanousie

    sanousie Member

    Joined:
    Apr 12, 2003
    Messages:
    2,128
    Location:
    Sydney
    Dam son, take it easy mate, no need to go insulting, Please refrain from the hostility mate, we are trying to help out this dude.

    Cannot compare a high end psu to a Generic/case psu, thats what i meant mate.

    Generic/low end psu say made in 2013 wont be able to keep up with a psu made in late 2016, again talking about low end affordable PSU's.

    When it comes to psu's, the gen has changed over the past five years with how many rails/amps/and noise on each rail (completely took my post out of context), Again im talking about lower end/case psu's not higher end 1000+ watt psu's.

    Op has a strider 600watt, seems sufficient, as zero said, check all psu plugs/connections, Check temps when stress testing along with voltages, post up your results here mate, best of luck.

    OP should run a stress program (prime95/linx64) and monitor voltage with hwmonitor.
     
    Last edited: Jan 14, 2017
  16. i7Baby

    i7Baby Member

    Joined:
    Jul 3, 2012
    Messages:
    290
    Can you install HWInfo64 and run Sensors? Then post the results - it'll show voltages, temps etc. Can be of good help with hardware issues.
     
  17. OP
    OP
    xHardz

    xHardz Member

    Joined:
    Aug 22, 2008
    Messages:
    343
    [​IMG]temporary image hosting

    The report was massive, was that the section you were after?

    Other Temps;
    GPU: 40-45
    HDD: 34

    Its probably mid 20's in the room atm.

    Thanks for all the replies! Starting to look like a PSU issue. I posted on the Asrock forum too and they seemed to have similar thoughts.
     
  18. sanousie

    sanousie Member

    Joined:
    Apr 12, 2003
    Messages:
    2,128
    Location:
    Sydney
    The 12v rails dont look to good, the min and max have to much of a difference. E.g. whats considered stable is 12.10v and it dropping down to 12.05 - 11.95v , some peeps allow for more of a drop, but the less of a voltage drop, would equate to less noise, better stability etc. The 5v rail dont seem to bad.

    Run the system in idle for a while with HW monitor open for about 5mins, run a stress test program to see what the voltage drop is (take a screen shot as you did in the above post and post it here, after running the stress program for 5 mins or so). Or can do the above, but instead of a Prime95/Linx play a game, that way you'd get max current draw on the psu and it should show its true colours.

    Best of luck mate.
     
    Last edited: Jan 17, 2017
  19. OP
    OP
    xHardz

    xHardz Member

    Joined:
    Aug 22, 2008
    Messages:
    343
    Ok.

    I was away for the weekend and the PC went into sleep mode and came out fine. Sunday arvo/monday were ALOT cooler than it had been and there were no restarts.

    Today is a return to the heat and I have just had a restart while browsing/idle. Ran R6:Siege for about 15 minutes to get the below results.
    [​IMG]

    [​IMG]


    imgurl
     
  20. sanousie

    sanousie Member

    Joined:
    Apr 12, 2003
    Messages:
    2,128
    Location:
    Sydney
    Seems like its a heat and current problem, more heat, more voltage/current noise which in turn will cause a system to be unstable at some point.

    12v psu rail is dropping atleast 0.2v and your 12v GFX VRM rail is dropping atleast 0.18v, all these could be the culprit.

    Best thing would be, Remove your GFX (less strain on the PSU) use onboard GFX and stress test using LINX/prime95 (i'd go linx64 and stress all available ram). This way we know for sure that your board, CPU, RAM is stable and we can use the process of elimination on the psu/GFX or vice versa.

    Post your results.

    Cheers.
     
    Last edited: Feb 2, 2017

Share This Page