System Powers Off Between a Few Seconds and 10 Mins of Starting

J

jlaw

n00b
Joined
Aug 3, 2015
Messages
24
Got a build that runs fine in UEFI.

It passes the Intel UEFI CPU check tool, Dell UEFI diags and Memcheck86 UEFI. (Temps are great during tests ~32C)

Problem is, it powers off shortly after starting Windows or Linux (usually within a minute).

I have swapped motherboard, power supply, hard drive and video card. I have reduced RDIMM count to 1.

I suspect memory at this point, but would like some proof. Asrock says there are no UEFI logs that would indicate the source of the problem.

Strange thing is this system ran fine for several weeks before it started exhibiting this behavior. Please help me find the cause.

Intel - Xeon E5-1650 V3 3.5 GHz 6-Core Processor
Corsair - H150i PRO 47.3 CFM Liquid CPU Cooler
ASRock - X99 OC Formula/3.1 EATX LGA2011-3 Motherboard
8 x MTA9ASF51272PZ / Micron 4GB 1RX8 PC4-2400 ECC
ADATA - XPG SX8200 960 GB M.2-2280 Solid State Drive
Asus - GeForce GTX 1070 8 GB Video Card
EVGA - 850 W 80+ Gold Certified Semi-Modular ATX Power Supply
 
Does it make it all the way through the boot process (drivers loaded, etc), or is it rebooting even while it's still strapping up?
 
  • Like
Reactions: jlaw
like this
Brian_B said:
Does it make it all the way through the boot process (drivers loaded, etc), or is it rebooting even while it's still strapping up?
Click to expand...

It makes it through the boot process. I have not seen shutdowns happen during the boot process or in the UEFI. I had some sessions last long enough to reinstall Windows. It will sometimes freeze / lock up in Windows as well. I was also able to install Ubuntu on another drive successfully after a few tries were unsuccessful due to shutdowns.

auntjemima said:
Powers off completely or reboots?
Click to expand...

It does not reboot. It just shuts down. I can hit the power switch and it will boot back up again for a random amount of time.
 
Last edited:
Okay, how about:

Test your outlet/UPS, might be faulty

Unplug all mobo USB connections along with front audio connection, then start your PC, computer case connections to mobo might be faulty

Check power button, might be faulty

Remove your video card and try running using only mobo video, PCI-E slot might be faulty

I hope you find the cause of your problem.
 
  • Like
Reactions: jlaw
like this
Nanogrip said:
Check power button, might be faulty
Click to expand...
Go one step further since the mobo has power/reset buttons on it, turn the system on using that and leaving the other switch controls unhooked (and make sure those buttons on the board aren't sticking).
When you traded out the PSU did you use a different power cable as well?
Nanogrip said:
Remove your video card and try running using only mobo video, PCI-E slot might be faulty
Click to expand...
Try a different slot since theres no integrated graphics.
 
  • Like
Reactions: jlaw
like this
Turn off all power save options in BIOS and in Windows.
Look for a BIOS update.

vCore and vMem in spec?

.
 
  • Like
Reactions: jlaw
like this
Good thoughts, please keep them coming.

Nanogrip said:
Okay, how about:

Test your outlet/UPS, might be faulty

Unplug all mobo USB connections along with front audio connection, then start your PC, computer case connections to mobo might be faulty

Check power button, might be faulty

Remove your video card and try running using only mobo video, PCI-E slot might be faulty

I hope you find the cause of your problem.
Click to expand...



Spartacus09 said:
Go one step further since the mobo has power/reset buttons on it, turn the system on using that and leaving the other switch controls unhooked (and make sure those buttons on the board aren't sticking).
When you traded out the PSU did you use a different power cable as well?
Try a different slot since theres no integrated graphics.
Click to expand...

I tried direct to wall power and through my UPS. I tried with several cables. I pulled the motherboard out of the case and had the same issue - no usb connections. I tried different PCIE slots for the different video cards. All the case connections such as USB & power switch are disconnected.

I was recalling when I first put the memory in, the board didn't show all the dimms in the UEFI. At the time I just pulled them out and rearranged them and put them back in. Then they all showed up. This is what makes me suspicious of the RAM even though there are no failures in memcheck86.
 
Spartacus said:
Turn off all power save options in BIOS and in Windows.
Look for a BIOS update.

vCore and vMem in spec?

.
Click to expand...

BIOS is latest for this board - v3.80
Windows Power plan is set to High performance with PCIE & USB power save disabled. No sleep or drive spin down.
C States disabled in BIOS
vcore is auto (UEFI default) - 1.256 V
since I am suspicious of the memory, your question about vmem got me pondering - It was set to 1.2 by default so I set it 1.5. System has been up for about an hour now which is unusual. HWinfo shows temps @ 35C-39C.

hmmm...
 
jlaw said:
BIOS is latest for this board - v3.80
Windows Power plan is set to High performance with PCIE & USB power save disabled. No sleep or drive spin down.
C States disabled in BIOS
vcore is auto (UEFI default) - 1.256 V
since I am suspicious of the memory, your question about vmem got me pondering - It was set to 1.2 by default so I set it 1.5. System has been up for about an hour now which is unusual. HWinfo shows temps @ 35C-39C.

hmmm...
Click to expand...

Hmmm, maybe your RAM slots need inspecting too? Try cleaning them out with canned air, then spray a bit of silicon lube onto a clean and durable paper towel, then wipe the edges of your RAM sticks before you stick them back in.
 
Nanogrip said:
Hmmm, maybe your RAM slots need inspecting too? Try cleaning them out with canned air, then spray a bit of silicon lube onto a clean and durable paper towel, then wipe the edges of your RAM sticks before you stick them back in.
Click to expand...

well, asrock replaced this board with the same model RMA already. If memory is the issue, I would think it would have to be the DIMMs. No issues since upping the voltage. Anyone have a thought on why I would have to do that at stock speeds? I guess I can safely leave it there as long as the temps are ok?
 
Sounds like you fixed it by bumping up the vMem to 1.5.
It's not unusual to need to do that.

.
 
Spartacus said:
Sounds like you fixed it by bumping up the vMem to 1.5.
It's not unusual to need to do that..
Click to expand...

Well, I have read 1.5 v is the upper limit so I was not expecting to have to start there (with no overclock), but the thing that is most surprising to me is that it worked fine for a while but then started needing more voltage a few weeks later. Also, my previous power supply was not able to deliver 1.5 v reliably, so having the seasonic also helped.
 
Setting vMem at 1.5v should be fine, I know my memory is SPD rated at 1.35v, but I had it at 1.5v for over a year on my old mobo.
From what I remember, I had very occasional BSODs until I bumped it up to 1.5v.

Just checked my new mobo and the same memory is running the same speed and timings at the XMP auto setting (3200) with 1.36v vMem.
Just differences in the two mobos (two different MSI models).

So anyway.... along with your old PSU not giving you stable voltages, you can have some variance on what it takes to stabilize the memory
on a particular mobo.

You are correct that I would not run vMem over 1.5v though.

Is it still stable?

.
 
Spartacus said:
Setting vMem at 1.5v should be fine, I know my memory is SPD rated at 1.35v, but I had it at 1.5v for over a year on my old mobo.
From what I remember, I had very occasional BSODs until I bumped it up to 1.5v.

Just checked my new mobo and the same memory is running the same speed and timings at the XMP auto setting (3200) with 1.36v vMem.
Just differences in the two mobos (two different MSI models).

So anyway.... along with your old PSU not giving you stable voltages, you can have some variance on what it takes to stabilize the memory
on a particular mobo.

You are correct that I would not run vMem over 1.5v though.

Is it still stable?

.
Click to expand...

Yep yep, so far it's made me a happy guy again. Thanks for the kind assistance.
 
So after these 5 months the system was solid - no issues then bam... wont post. Motherboard gives memory related debug codes but also random other debug codes sometimes. I believe the IMC was weak to begin with and finally gave out. So now I need to find a e5 1691 v3 to replace the CPU. Donations gladly accepted :)

Is this a known issue with 16xx v3 xeon? I mean did I just get a bad one? Should I worry about getting another? I'm think I'm not ready to leave x99 quite yet, even though the AMD 3600 looks really appealing. I don't like the PCI lanes limit there - but I'd be lucky to get a 1660 v3 for the same $200.
 
Last edited:
You must log in or register to reply here.
Back
Top