Page 1 of 3 123 LastLast
Results 1 to 10 of 21

Thread: Is my PSU going bad?

  1. #1
    Join Date
    Dec 2011
    Posts
    9
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default Is my PSU going bad?

    Starting 2 days ago, my system will run for about 30 minutes, just surfing the web, and will then reboot itself and continue to reboot until the system is powered off. There have been no changes to the system. I do not have another PSU to swap and test with. I have considered buying a new PSU, but before I do I would like to try and determine if the PSU is the problem.

    Any help with determining the cause is appreciated. Recommendations for possible replacement PSU's is also appreciated.




    System specs are as follows:

    Antec Nine Hundred Two Case
    Intel Core i7 920 CPU
    Arctic Silver 5 Thermal Compound
    Noctua NH-U12P CPU Cooler
    3 - WD Caviar Black 500GB HDD
    EVGA E758 X58 Motherboard
    EVGA GeForce GTX 285
    Mushkin Enhanced Blackline 4GB DDR3 1600 RAM
    ZALMAN ZM750-HP 750W PSU
    LITE-ON 22X DVD Burner

  2. #2
    Join Date
    Aug 2011
    Posts
    526
    Thanks Thanks Given 
    30
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    If you don't have any backup components, a multimeter can help test the PSU.

  3. #3
    Join Date
    Dec 2011
    Posts
    27
    Thanks Thanks Given 
    6
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    You should monitor the temps and also the voltages, with something like HWinfo32, just to make sure you aren't overheating.

    Before your computer starts rebooting, are you able to run a 3D game, or something like Prime 95 blend ? That would catch PSU voltage issues, I think.

    You can also try HCI Memtest to test the memory as well.

  4. #4
    Join Date
    Dec 2011
    Posts
    9
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by ssshjp View Post
    If you don't have any backup components, a multimeter can help test the PSU.
    I don't have and have never used a multimeter before. I will do some research and see if this is something I would be comfortable doing.


    Quote Originally Posted by Falkentyne View Post
    You should monitor the temps and also the voltages, with something like HWinfo32, just to make sure you aren't overheating.

    Before your computer starts rebooting, are you able to run a 3D game, or something like Prime 95 blend ? That would catch PSU voltage issues, I think.

    You can also try HCI Memtest to test the memory as well.
    I will try these when I get home. I always have speedfan running so I can monitor my temps. When the system reboots, the temps for my CPU have never been above 24C and my GPU temps are between 47C-50C.

  5. #5
    Join Date
    Aug 2011
    Location
    PlAnEtCrAzY!
    Posts
    7
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Excessive heat from inside the case, CPU, PSU, drives, etc, is always a good place to start! If it's heat related you want to nail this down first otherwise you will destroy hardware. Software and other issues are annoying but less costly, well, unless you have to donate more money to the Bill and Melinda Gates foundation.

    Random reboots could be any number of issues from failing hardware to buggy software including the OS, usually a device driver, not to mention malware as a last resort.

    Always start with the obvious - look inside your case, any obstructions blocking airflow, a bad fan, dust buildup, RAM that needs to be reseated?

    I would then find out if your hardware offers any system info and testing utilities, HWINFO32, Speccy, PCWizard and others are good too. You could also go into your BIOS when the system reboots and have a look at the temps, voltages etc.

    Is there an Event Viewer log created at the time of a reboot, anything flagged, a memory dump file? Is your system set to automatically restart upon a system failure? What does Device Manager have to say?

    Point being - leave no stone unturned, and keep track of your findings, solutions you've tried, their outcome, possibilities eliminated.
    Mike

  6. #6
    Join Date
    Dec 2011
    Posts
    9
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    All the fans are working fine, airflow is fine, and there is no dust build up, system is cleaned on a regular basis. No event log entries at time of reboot, no flags, or memory dumps. No issues in device manager. Prime 95 ran for 6 minutes before system rebooted. Below are temps and voltages from idle, during Prime 95, and BIOS after failure. I will run HCI Memtest in a little while and post results.


    EVGA E-LEET temps and voltages at idle:

    CPU temp: 19C
    System temp: 21C

    Voltages:
    3.38
    5.13
    12.28


    HWinfo temps and voltages during Prime 95:

    CPU temp: min 19C, max 42C
    Core temps: min 33C, max 58C
    GPU temp: 47C

    Min Voltages:
    3.28
    5.087
    12.184

    Max Voltages:
    3.408
    5.171
    12.258


    BIOS temps and voltages after failure:

    CPU temp: 26C
    System temp: 23C

    Voltages:
    3.40
    5.12
    12.18




    HCI Memtest information:

    Idle HWinfo temps and voltages:

    CPU temp: min 18 max 27
    Sys temp: min 18 max 20
    Core temps: min 31 max 41

    Voltages:
    min 3.328 max 3.360
    min 5.129 max 5.129
    min 12.184 max 12.258


    During Memtest HWinfo temps and voltages:

    CPU temp: min 18 max 30
    Sys temp: min 18 max 23
    Core temps: min 31 max 45

    Voltages:
    min 3.280 max 3.408
    min 5.087 max 5.171
    min 12.184 max 12.258

    After running the memtest for 21 minutes, the system had completed 25% of memtest with no errors and then the system failed and rebooted.
    Last edited by azrael; 01-03-2012 at 01:26 AM. Reason: Added HCI Memtest info

  7. #7
    Join Date
    Dec 2011
    Posts
    27
    Thanks Thanks Given 
    6
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Ok I'm going to admit I'm sort of over my head here, and step out.
    The system rebooted in just Memtest, also?

    What QPI/vtt voltage and DDR voltage are you supplying?
    If setting the VTT to 1.3v and DDR to 1.65v doesn't fix the problem, then it could be the motherboard or PSU. I guees the only thing you can do is try replacing the PSU before dismantling the system to replace the motherboard... but again I'm probably over my head here. Hope you manage to find a good solution...

  8. #8
    Join Date
    Dec 2011
    Posts
    9
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by Falkentyne View Post
    Ok I'm going to admit I'm sort of over my head here, and step out.
    The system rebooted in just Memtest, also?

    What QPI/vtt voltage and DDR voltage are you supplying?
    If setting the VTT to 1.3v and DDR to 1.65v doesn't fix the problem, then it could be the motherboard or PSU. I guees the only thing you can do is try replacing the PSU before dismantling the system to replace the motherboard... but again I'm probably over my head here. Hope you manage to find a good solution...
    Yes, the system rebooted during the memtest. The QPI/vtt voltage and DDR voltage are all stock. None of my hardware is OC'ed.

  9. #9
    Join Date
    Jul 2009
    Location
    North Carolina
    Posts
    259
    Thanks Thanks Given 
    3
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    A few things I would try:
    1. Update your BIOS.
    2. Make sure the command rate on your RAM is 2T, not 1T. Set it manually if you have to.
    3. If those don't work, set your CPU to stock clocks.
    4. If you still aren't having any luck, you may need to take a hard look at the PSU being the culprit.

    Also, on a side note. Your statement about your CPU temps never being above 24C is completely incorrect. This is likely an incorrect reading provided by the motherboard. The CPU readings ("Core temps: min 33C, max 58C") you got from HWinfo are closer to what you would actually expect. Basically your idle is around 33C and your max temps will be around 58C as reported by the CPU itself.

    Your other issue could be a bad or damaged motherboard due to your heavy heatsink. Massive "big air" heatsinks can strain your motherboard due to the sheer weight of the thing. The other thing could be the combination of that and your CPU getting really hot on a regular basis. 58C during prime95 with no overclock on a Noctua NH-U12P is honestly not too good, especially if you are running dual fans.

    I would also try to see if the problem occurs more often the longer the computer is left on. If you give it a 1-2 hour rest, does it take longer for the problem to occur? If so, then you likely have a simple overheating issue. There are a number of things on the motherboard that can overheat and cause reboots.

    Let me know if this helps.
    PSU: Rosewill FORTRESS-650 || Mobo: MSI Z97M-G43
    CPU: Intel i7-4770K || Corsair H60 (New Version)
    RAM: 8GB G. Skill Ripjaws 2133MHz @11-11-11-30-2T || SSD: Intel 530 240GB
    HDDs: 1TB WD Black, 2x 2TB Hitachi || GPU: AMD Radeon HD7870 @ 1150/1400

  10. #10
    Join Date
    Dec 2011
    Posts
    9
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by p4l1ndr0m3 View Post
    A few things I would try:
    1. Update your BIOS.
    2. Make sure the command rate on your RAM is 2T, not 1T. Set it manually if you have to.
    3. If those don't work, set your CPU to stock clocks.
    4. If you still aren't having any luck, you may need to take a hard look at the PSU being the culprit.

    Also, on a side note. Your statement about your CPU temps never being above 24C is completely incorrect. This is likely an incorrect reading provided by the motherboard. The CPU readings ("Core temps: min 33C, max 58C") you got from HWinfo are closer to what you would actually expect. Basically your idle is around 33C and your max temps will be around 58C as reported by the CPU itself.

    Your other issue could be a bad or damaged motherboard due to your heavy heatsink. Massive "big air" heatsinks can strain your motherboard due to the sheer weight of the thing. The other thing could be the combination of that and your CPU getting really hot on a regular basis. 58C during prime95 with no overclock on a Noctua NH-U12P is honestly not too good, especially if you are running dual fans.

    I would also try to see if the problem occurs more often the longer the computer is left on. If you give it a 1-2 hour rest, does it take longer for the problem to occur? If so, then you likely have a simple overheating issue. There are a number of things on the motherboard that can overheat and cause reboots.

    Let me know if this helps.
    When I said the CPU temps were never over 24C, that was the temps being displayed in my system tray by Speedfan.

    My CPU is at stock and I am using a single fan on the Noctua as there is an exhaust fan directly behind the heatsink and can't fit the second fan on the heatsink.

    The system has run fine for the last 2 years on the current BIOS and with the system rebooting, I am not going to risk trying to update the BIOS.

    The computer will not stay on for more than 30 minutes, when just surfing the web. If anything else is done, e.g. memtest, prime95, or gaming, it will reboot sooner. Between the Prime95 and Memtest, the system was off for 3 hours.

    Also changed the command rate on RAM from T1 to T2, but no difference, still reboots after 30 minutes.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •