Plagued by sudden restarts and power issue


Recommended Posts

I have been plagued by sudden restarts of my system for over a year and it is just getting worse. No blue screen and in event viewer i just get a generic Kernel-Power error (all BugcheckCode values are 0) it just restarts It used to happen with mostly very GPU demanding games so i suspected it was just my GPU overheating. And the temps on my 780 under full load got in the high 80s and if not for my system restarting i was sure it would go even higher (My new 1070 has never gone over 70 during full load still restarts so not a overheating issue). after capping my frame rate with RTSS lowering settings and keeping and eye at all times on my GPU temps i was able to reduce the problem at the time but it would still happen from time to time sometimes even when i was just watching a video. It also often happens 5 minutes after i am done playing a game and just browsing the web which seems to indicate it has trouble adjusting the power down or something?

 

Usually, it restarts in a second but sometimes it takes up to 30 mins for it to reboot again. (no fans spinning just seems dead then it eventually wakes from the dead) I suspect it's waiting for the capacitors to discharge. After the pc has been off for the night it can also takes 15-30 mins for it to power on (usually not when it's hibernating just when i turn it off) so I'm not so sure what that's about.

 

It has gotten worse recently and I'm pretty clueless about this stuff and i dont trust my own assessment if i buy a new PSU it will be a downgrade from what i had so i really would greatly appreciate a second opinion. Could it be caused by a faulty motherboard? Or do you guys think its pretty safe to assume its the PSU? I dont have a spare one to try.

 

The pc is also pretty clean as i dust it off often. I also have checked that there is nothing loose.

 

Some info from HWmonitor taken mid match in PUBG (it rebooted 10-15 mins later during the end game)

 

3.36 Volts [0xD2] (AVCC)

3.34 Volts [0xD1] (3VCC)

1.17 Volts [0x92] (CPU VCORE)

1.76 Volts [0x6E] (VCCIN)

1.49 Volts [0x5D] (DRAM)

10.91 Volts [0x7C] (+12V) Seems way too low not sure if accurate 

4.96 Volts [0x7C] (+5V)

26 degC (78 degF) [0x1A] (SYSTIN)

32 degC (89 degF) [0x20] (TMPIN3)

127 degC (260 degF) [0x7F] (TMPIN5)  This never changes its always 127 im 

guessing its a incorrect reading

26 degC (78 degF) [0x1A] (TMPIN6)

1.05 Volts [0x41A] (GPU)

35 degC (95 degF) (Package)

29 degC (84 degF) (Core #0)

33 degC (91 degF) (Core #2)

31 degC (87 degF) (Core #4)

30 degC (86 degF) (Core #6)

1.05 Volts [0x41A] (GPU)

101.28 pc (GPU)

48 degC (118 degF) [0x30] (GPU)

 

 

 

System:

MSI gaming z97

i5

gtx 1070

Corsair RM850 ( psu)

16 gigs of ram

haf xm (case)

NH-D14 (cpu cooler)

Link to comment
Share on other sites

I would at first blame your PSU. Do you have any spares that you can test this?

Link to comment
Share on other sites

17 hours ago, Mindovermaster said:

I would at first blame your PSU. Do you have any spares that you can test this?

No, unfortunately, also cant RMA it since i bought it used. going to be buying a new one. Thinking of the Corsair TX750m but unsure about compatibility. It should be no problem right? 

Link to comment
Share on other sites

No issues with compatibility.  I buy strictly Corsair for PSUs the last 5 or 6 years.  They've been pretty reliable.  Mainly CX/CS, so the TX should be good.  That's probably the one component I wouldn't ever buy used.  It's also the most common one I have had problems with besides hard drives.  

Link to comment
Share on other sites

37 minutes ago, radical_enzyme said:

No, unfortunately, also cant RMA it since i bought it used. going to be buying a new one. Thinking of the Corsair TX750m but unsure about compatibility. It should be no problem right? 

No, unless you get an old one, you should be OK. TX750m sounds good. :)

Link to comment
Share on other sites

If you say that your PC reboots often, and sometimes it doesn't boot at all for some time after it reset, I would rather think something temperature wise. I would check the following:

- Did you check if all your fans are spinning properly? (CPU, GPU, case fans, etc...)

- When your PC restarts, try going straight into the BIOS and check the temperature values of the various sensors. Maybe some are higher than they should (especially CPU)

- Older GPU's run hotter than newer ones (some do anyway), so that doesn't nessessarily mean it's  a bad thing if your 1070 is on lower temps than your older card

- re-seat your RAM

- Maybe remove the CPU heatsink and check the thermal paste. Could be by re-applying new  paste  and re-seating the heatsink and fan you get better temps

- I don't think the event logs will show anything, as with sudden reboots the system won't be able to write anything into the logs in time

- If your hardware monitoring tool can log the data from the sensors, try doing that. Wait until your PC reboots again and then check the last temps you had. A CPU can cool really quick after it stops being under load.

- Does your PC only reboot when or after you played games? What if you turn on your PC, don't play any games, but just do other stuff?

Link to comment
Share on other sites

Your +12v seems too low, and I'm afraid these Z97 boards are bad for having dodgy VRM's. I worked for an internet PC builder last year that recently went bust, and at least 30% Z97 Gaming board systems that came back seemed to have VRM issues with the voltages being too low, causing crashes and reboots under load.

 

They're digital on newer boards, the transistors seem to fail prematurely and don't switch on and off correctly to the timings they should. Do you have a dedicated PSU tester? This is how I used to verify the Z97's, it'd look like PSU, but when tested out of the system the PSU's were fine so I knew it was board.

 

My Gigabyte Z170-SLI's 12V never goes below 11.808v. I recommend you get a PSU tester with digital LCD screen. Mine is my best friend at work.

Edited by Tidosho
Link to comment
Share on other sites

It is the PSU.

 

Quote

 


Usually, it restarts in a second but sometimes it takes up to 30 mins for it to reboot again. (no fans spinning just seems dead then it eventually wakes from the dead) I suspect it's waiting for the capacitors to discharge. After the pc has been off for the night it can also takes 15-30 mins for it to power on (usually not when it's hibernating just when i turn it off) so I'm not so sure what that's about.
 

 

 

Although it never got more than a few minutes to boot I have RMA'd 2 AX760 with this problem. If you search google this is a pretty common issue with Corsair PSUs.

 

You should remove the PSU and do the self test, you can find instructions on the Corsair website.

Link to comment
Share on other sites

2 hours ago, Odom said:

If you say that your PC reboots often, and sometimes it doesn't boot at all for some time after it reset, I would rather think something temperature wise. I would check the following:

- Did you check if all your fans are spinning properly? (CPU, GPU, case fans, etc...)

- When your PC restarts, try going straight into the BIOS and check the temperature values of the various sensors. Maybe some are higher than they should (especially CPU)

- Older GPU's run hotter than newer ones (some do anyway), so that doesn't nessessarily mean it's  a bad thing if your 1070 is on lower temps than your older card

- re-seat your RAM

- Maybe remove the CPU heatsink and check the thermal paste. Could be by re-applying new  paste  and re-seating the heatsink and fan you get better temps

- I don't think the event logs will show anything, as with sudden reboots the system won't be able to write anything into the logs in time

- If your hardware monitoring tool can log the data from the sensors, try doing that. Wait until your PC reboots again and then check the last temps you had. A CPU can cool really quick after it stops being under load.

- Does your PC only reboot when or after you played games? What if you turn on your PC, don't play any games, but just do other stuff?

I can only get it to reliably reboot when playing games or just after i have played a game (depends on the game)  i have had it reboot a few times when watching videos. and a few times when i have left the room with just chrome open i come back to it having rebooted but that is very rare its very stable when not playing games. 

Link to comment
Share on other sites

1 hour ago, Tidosho said:

Your +12v seems too low, and I'm afraid these Z97 boards are bad for having dodgy VRM's. I worked for an internet PC builder last year that recently went bust, and at least 30% Z97 Gaming board systems that came back seemed to have VRM issues with the voltages being too low, causing crashes and reboots under load.

 

They're digital on newer boards, the transistors seem to fail prematurely and don't switch on and off correctly to the timings they should. Do you have a dedicated PSU tester? This is how I used to verify the Z97's, it'd look like PSU, but when tested out of the system the PSU's were fine so I knew it was board.

 

My Gigabyte Z170-SLI's 12V never goes below 11.808v. I recommend you get a PSU tester with digital LCD screen. Mine is my best friend at work.

1

Could the low +12v be caused by the PSU? Would all my issues be possible with a bad motherboard?

Link to comment
Share on other sites

Well, the PSU would make sense. When playing games your graphics card is under load, with also taxes the PSU and motherboard. I would tend to agree with the others in that case.

Link to comment
Share on other sites

48 minutes ago, radical_enzyme said:

Would all my issues be possible with a bad motherboard?

Yes. I would have said RAM, but your too-low 12v reading rules that out. Like I said, disconnect the PSU from the motherboard, and use a dedicated tester. HWMonitor takes readings from the VRM's and power control unit chip on the motherboard. Testing the PSU on its own will tell you if it is board or PSU. If you test it with a dedicated tester and the +12v is higher this time, the board is at fault.

Link to comment
Share on other sites

2 hours ago, Tidosho said:

Yes. I would have said RAM, but your too-low 12v reading rules that out. Like I said, disconnect the PSU from the motherboard, and use a dedicated tester. HWMonitor takes readings from the VRM's and power control unit chip on the motherboard. Testing the PSU on its own will tell you if it is board or PSU. If you test it with a dedicated tester and the +12v is higher this time, the board is at fault.

In the MSI bios the +12 voltage is a little over 12 so seems HWmonitor is wrong. 

Link to comment
Share on other sites

19 hours ago, radical_enzyme said:

In the MSI bios the +12 voltage is a little over 12 so seems HWmonitor is wrong. 

Update to the latest version and retry. My Z170's BIOS reading is in line with HWMonitor at 11.880

Link to comment
Share on other sites

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.