WHEA_UNCORRECTABLE_ERROR - Diagnosing Fault Hardware - New build - i7 4770k - ASUS Z87-PRO

wardler

Distinguished
Jun 28, 2011
91
0
18,630
I have done zero overclocking and BIOS settings are at defaults.

Components that have been added since instability occurred:
Intel Core i7 4770k
ASUS Z87-PRO
Corsair Vengeance Pro Series 8GB 2 x 4GB DDR3 1600 MHZ
CORSAIR AX series AX760
Seagate Barracuda 7200.14 ST3000DM001 3TB

Other installed components:
SAMSUNG 830 Series 2.5-Inch 128GB SSD
EVGA "Superclocked" GeForce GTX 770 2GB (new, but stable in previous Sandy Bridge build)
CM Hyper 212 EVO

OS:
Windows 8.1 64-bit (installed to SAMSUNG SSD)

The problem:
WHEA_UNCORRECTABLE_ERROR (blue screen)
Computer crash

When does the problem occur:
The problem occurs frequently anywhere between 10 and 60 minutes. It appears to happen when streaming video (Google Chrome) and playing games (WoW), although it apparently restarts during the day, because I can't access it remotely from work sometimes. This problem began after installing the new components above. I am having trouble diagnosing the problem. I work in the IT field, but I do less and less hardware maintenance every year.

At first I suspected drivers at fault as I did not do a clean install. I broke down and did a clean install last night hoping that would fix the problem, but not long after installing the drivers and starting up a game (WoW) the machine crashed again with the same error. I refused to believe that it was hardware, because I hate the hassle of returns and I hate being without my main machine at home for very long, but I have accepted the fact that this is a hardware problem. Working in IT all day makes me just despise coming home and working on my own computers.

What I have done:
Updated BIOS
Reinstalled Windows w/ correct and most up to date drivers
Memtest 86 (when I left home it was about 88% complete and I expect it to finish without errors - I will update when I get home)
Ran Prime95 to see if I could make the CPU crash. It ran through about 3 different tests on blend over about 2 hours and passed all tests. I saw temps reaching 96c during some phases of the test according to Piriform's Speccy.
Prayed

Any help in diagnosing this so I can return the least number of parts as possible and be on my way are very much appreciated.
 
Solution


Ok, those temperatures sound appropriate for CM 212 EVO, we may be able to rule the bad heatsink install out.

Next step is to manually set voltages, and see if your problems continue. I recommend that you stay away from Prime95 until voltages are manually set.

If manually setting voltages does not fix the issue, you will probably need to update BIOS to rule that out.

Beyond that, I'm gonna say you have either a faulty CPU or Motherboard - but I would start with the above because I give it a 80% it will fix your problems once you...

Adroid

Distinguished
WHEA errors are the result of inadequate voltage. Based on your temperatures, you need to back your overclock down. If you aren't overclocked, something is horribly wrong, either 1)You are using the stock cooler and running prime 95 which is a really bad idea, or 2) you didn't install your heatsink and/or thermal paste properly.

You should never run Prime95 with CPU on automatic voltages, you run the risk of damaging your CPU. Please stop whatever you are doing and give us BIOS and/or overclock settings.
 
Since it's a hardware error, reinstalling the OS shouldn't help. Since it crashes quite often, it should be easier to determine which component is causing the hardware fault. Start by removing the GPU and use the on-board video. That will at least determine if the video card could be causing the issue. If that doesn't help, then disconnect the SSD and boot from a different device because a defective hard disk or SSD could also cause that error.
 

wardler

Distinguished
Jun 28, 2011
91
0
18,630
Thank you for that quick reply, guys. I forgot to mentioned that I have a CM Hyper 212 EVO air cooler and artic silver thermal paste and I am fairly certain that it is seated properly. I have done zero overclocking and BIOS settings are at defaults.
 

wardler

Distinguished
Jun 28, 2011
91
0
18,630
Also, another possibility as far as incorrect voltages. I am not sure what configuration the PSU should be set up in as it is completely modular and to be honest once I couldn't find any type of diagram explaining what rails were in which slots I hooked everything up in a row. Maybe I have put everything on one rail? The motherboard connected is keyed to one connection point, but the PCIe and the rest seemed to go into the same type of connectors.

This may be why when I stressed the CPU that the computer did not shut off. Maybe it is exceeding voltage when the CPU and GPU are kicking in at the same time.
 

Adroid

Distinguished




Your PSU has a single 12V rail, so how you hooked it up is a non issue, provided you got all the parts and pieces plugged in.

Now, what are your idle and load temperatures? You should be running around 30° C during idle, and 55-60° under load. Please use core temp for temperatures, and make sure you have CPU-Z to check voltages.

Your voltage should not exceed 1.25v until we figure out what's going on here.

I may be wrong, but at first glance I think something is wrong with the heatsink install, or your CPU has something going wrong. You need to normalize your voltages and then do some testing to make sure the CPU is not damaged.
 

wardler

Distinguished
Jun 28, 2011
91
0
18,630
According to Piriform's Speccy my idle temps are around 24-26c and around 50-60c while gaming. When I get home this evening I will try the software you suggested to measure the temp and voltage.
 

Adroid

Distinguished


Maybe so, if you set your voltages manually in the BIOS that is the best way. But, if the offset voltages etc. are not manually fixxed, BIOS will not give an accurate read because you can't read under full load in BIOS. Most of us don't have a DVM - so CPU-Z is the next best thing...
 

Adroid

Distinguished


Ok, those temperatures sound appropriate for CM 212 EVO, we may be able to rule the bad heatsink install out.

Next step is to manually set voltages, and see if your problems continue. I recommend that you stay away from Prime95 until voltages are manually set.

If manually setting voltages does not fix the issue, you will probably need to update BIOS to rule that out.

Beyond that, I'm gonna say you have either a faulty CPU or Motherboard - but I would start with the above because I give it a 80% it will fix your problems once you manually set voltages and/or update BIOS.
 
Solution

wardler

Distinguished
Jun 28, 2011
91
0
18,630
Just an update here. I reset BIOS to default settings (although I have never changed any settings) and I had no shutdowns last night. I will give it a few days and post back again.
 

wardler

Distinguished
Jun 28, 2011
91
0
18,630
One last update - resetting the BIOS to defaults seems to have my system stable. I have not had a crash since my last update and I have been performing a wide variety of tasks since changing the settings. Thank you very much for the help.