Jump to content

WHEA_UNCORRECTABLE_ERROR new cpu

Hey Guys! 

 

Yesterday I installed my new i7 8700k and a new cooler (dark rock pro 4). For the cooler installation I had to remove the motherboard and install it externally. Therefore I had to disconnect all my cables. Now after installing everything back together I boot up my pc and everything runs fine as before. But I start noticing that when I use prime95 my cpu receives voltages of 1.2-1.21v which is kinda high since the base voltage according to my MB should be 1,144 v think. Shortly after that I started receiving blue screens with the WHEA_UNCORRECTABLE_ERROR. I didn't notice any high temps. Even under full load I only saw cpu voltages of 68 degrees celcius max.

So I took a closer look to my MB and noticed that I plugged in the cpu power cable in the wrong slot. The gigabyte aorus z390 elite has the standard 2x4 12v 8 pin connector and an additional 2x2 12v 4 pin connector. My psu (straight power 11 550w) has two 4 pin connectors labeled cpu 1 -P4 and cpu2 - P8. Since the pin layout of the 2x2 and the closest to the cpu from the 2x4 is the same I accidentally plugged in the cpu1 - P4 connector in the 2x2 connection spot. After unplugging in and moving it to the right place my voltages are definitely lower. 1.15v under prime95 full load. 

 

Now I couldn't test it any more but I didn't receive any blue screens after that. Could my mistake with the power connector cause the high voltages and blue screen? Or do I have to worry about a possible cpu hardware error? 

 

Greetings and sorry for the long post 

Link to comment
Share on other sites

Link to post
Share on other sites

most electronics are designed with at least 5% tolerance if not 10% which 5% would put max voltage at 1.258v so you should be good

Link to comment
Share on other sites

Link to post
Share on other sites

9 minutes ago, yaboistar said:

the voltages are going high because p95 is very stressful and it's trying to compensate. BUT: you shouldn't trust anything that software reports back to you about voltages - the only accurate voltage measurements come from a meter on the back of the socket.

 

WHEA happens most commonly under overclocking where there is insufficient voltage to properly support the CPU.

 

i don't know how that board is actually wired up but even a 4 pin CPU header should support a stock 8700k and i can't imagine they've gone to the effort of making it only workable on one of those power sockets

 

you won't know if switching to the 8pin EPS header has worked until you test it again

I'm just curious why the voltages are that much lower after switching slots. From 1,21v under load to 1,15v. 

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, cole0622 said:

most electronics are designed with at least 5% tolerance if not 10% which 5% would put max voltage at 1.258v so you should be good

But my pc kept crashing. After switching slots my voltages under load don't exceed 1,15v now

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, cole0622 said:

at what point does it keep crashing

On different occasions. Surprisingly not when running prime95 but when having chrome open and starting a game. Or then starting a download in Chrome. So it seems a bit random to me. I had hw monitor open all the time and never had any high temps

Link to comment
Share on other sites

Link to post
Share on other sites

No I havent. I only used "scannow" to scan for corupted files. It actually found and repaired them but this was before switching pins and i had another blue screen after it.

 

I'm going to look up the event log once im at home but im starting to believe this is actually a hardware error. i bought the cpu used and it seems i hit the jackpot......

Link to comment
Share on other sites

Link to post
Share on other sites

you might also look into a memory dump. when it crashes is it a bsod or just die? if bsod is it still the same error?

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, cole0622 said:

you might also look into a memory dump. when it crashes is it a bsod or just die? if bsod is it still the same error?

yeah its a bsod with the WHEA error code. it then shows a percentage where it collects error info and restarts.

Link to comment
Share on other sites

Link to post
Share on other sites

you might delete the memory dump or move it to your desktop you might try and reset bios, last if you have a boot-able flash drive or a disk you can put windows or linux on to see if you keep running into issues if not then some software in your original windows could be causing the problem

Link to comment
Share on other sites

Link to post
Share on other sites

So is this something that CAN generally happen when switching out the cpu? 

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, SwarlesBrkly said:

So is this something that CAN generally happen when switching out the cpu? 

 

Your post was extremely confusing and hard to follow.

Are you saying that your PSU has a single EPS12v connector that is split into two 4+4 pin connectors (which combine), and you plugged in the plugs "split up" into the solo (secondary) 4 pin connector on the motherboard, and the second (that you split from PSU cable) 4 pin plug into the eight pin motherboard connector, with only 4 of the 8 pins plugged in?

 

And then after that, you combined the two 4+4 pin PSU plugs and plugged them both together into the motherboard 8 pin plug, leaving the solo 4 pin on the motherboard unconnected, right?

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, Falkentyne said:

 

Your post was extremely confusing and hard to follow.

Are you saying that your PSU has a single EPS12v connector that is split into two 4+4 pin connectors (which combine), and you plugged in the plugs "split up" into the solo (secondary) 4 pin connector on the motherboard, and the second (that you split from PSU cable) 4 pin plug into the eight pin motherboard connector, with only 4 of the 8 pins plugged in?

 

And then after that, you combined the two 4+4 pin PSU plugs and plugged them both together into the motherboard 8 pin plug, leaving the solo 4 pin on the motherboard unconnected, right?

Almost. Right now only the one matching 4 pin connector from the psu is plugged in the 8 pin motherboard plug. The single 4 pin plug in the motherboard is unused yes

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, SwarlesBrkly said:

Almost. Right now only the one matching 4 pin connector from the psu is plugged in the 8 pin motherboard plug. The single 4 pin plug in the motherboard is unused yes

So are you using a 4 pin only?  Where is the other 4 pin on the PSU?  

Link to comment
Share on other sites

Link to post
Share on other sites

5 minutes ago, Falkentyne said:

So are you using a 4 pin only?  Where is the other 4 pin on the PSU?  

Its not used since the pins don't fit in the remaining pin slots on the motherboard. I was told and am pretty sure I'm not supposed to force any pins into slots they don't fit it.

 

I was troubke shooting for a while now. Updating drivers, resetting and updating bios, checking windows registry files and updates, monitoring cpu temps and voltages which seem to run fine. I was even able to run Jedi fallen order for 20 minutes with good performance and low temps. 

Right now I'm checking my hard drive for any corruption. After that my last ideas is to just install windows again

IMG_20191206_145344.jpg

Edited by SwarlesBrkly
Link to comment
Share on other sites

Link to post
Share on other sites

My suggestion is to buy a new, modern PSU asap and get rid of that old clunker.

You are using an unsupported PSU configuration.  You should get a new PSU as soon as possible and not use the system in the current state except for basic work.  You can potentially burn out the connector (or even cause a fire!).

 

Can you please download HWinfo64 and look in the VRM section (Intersil or IR 35201) and look for the VR VIN value?  Record that value, as well as the VR VOUT, and post that here.  Idle is good enough.

 

Then, power off and unplug, and do that configuration that gave you a BSOD before, where you said your voltages were higher--- P4 into the motherboard 8 pin, and P8 into the 4 pin, then boot back to windows and then tell me the VR VOUT and VR VIN values again.

 

Note: if you see a VR VOUT of 0.004v if your iGPU is disabled, use the other field.

 

Thank you.

I am asking this because I have an idea what the motherboard is doing, but it's bugging out because 4+4 in separate pins is not a supported configuration.

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, Falkentyne said:

My suggestion is to buy a new, modern PSU asap and get rid of that old clunker.

You are using an unsupported PSU configuration.  You should get a new PSU as soon as possible and not use the system in the current state except for basic work.  You can potentially burn out the connector (or even cause a fire!).

 

Can you please download HWinfo64 and look in the VRM section (Intersil or IR 35201) and look for the VR VIN value?  Record that value, as well as the VR VOUT, and post that here.  Idle is good enough.

 

Then, power off and unplug, and do that configuration that gave you a BSOD before, where you said your voltages were higher--- P4 into the motherboard 8 pin, and P8 into the 4 pin, then boot back to windows and then tell me the VR VOUT and VR VIN values again.

 

Note: if you see a VR VOUT of 0.004v if your iGPU is disabled, use the other field.

 

Thank you.

I am asking this because I have an idea what the motherboard is doing, but it's bugging out because 4+4 in separate pins is not a supported configuration.

Mh OK. I bought the psu a year ago. It's the be quiet straight power 11 550w. Seems a bit crazy to me that it's outdated. 

I've been using it for a year now with a i5 8600k without any problem. Gaming daily 

Link to comment
Share on other sites

Link to post
Share on other sites

Wait a minute.

If you bought it a year ago and it was manufactured a year or two ago, then you should be able to plug in *both* 4 pins together into the 8 pin slot as a grouped 8 pin.

The two 4 pins are keyed exactly the same way.  Just 'attach' them as one and then push them in at the exact same time.  I am 100% sure this is how it's supposed to work.

 

There is no such configuration as a motherboard that has two, separate broken up 4 pin connectors, except for those boards that use a separate onboard PCIE connector for extra video card power for multiple PCIE slots.  There is only single 4 pin (ancient), single 8 pin, 8+4 pin, and 8+8 pin.


My Seasonic has two 8 pin connectors.  The first is a "Fused" 8 pin (cannot be split).  The second is a 4+4 pin, which can be split.  I have them both plugged into both 8 pin connectors (Z390 Aorus Master) completely.

 

The reason your voltage was reported low with only one 4 pin plugged in was your CPU +12v reading was probably horribly low because you are --required-- to plug in the entire 8 pin connector on these Z390 boards.  Solo 4 pin is not supported, IIRC.  It may work for 4 core non K processors however but this is still not a supported configuration.  So your board probably adjusted the voltages to accommodate for the solo 4 pin by itself (which is not a standard configuration for Z390).  It probably has a failsafe for this, as Gigabyte has low end boards that only have a 4 pin motherboard connector.

 

When you had them plugged into both 8 pin (as 4 pin) and other 4 pin solo, the voltage may have been higher due to current balancing (load coming from both PSU pins but not doubled load).  But you were unstable because you were still drawing WAY too low +12v and the motherboard cannot handle this configuration (4 pin and 4 pin in two different PSU plug groups), if that makes sense, thus you crashed.

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Falkentyne said:

Wait a minute.

If you bought it a year ago and it was manufactured a year or two ago, then you should be able to plug in *both* 4 pins together into the 8 pin slot as a grouped 8 pin.

The two 4 pins are keyed exactly the same way.  Just 'attach' them as one and then push them in at the exact same time.  I am 100% sure this is how it's supposed to work.

 

There is no such configuration as a motherboard that has two, separate broken up 4 pin connectors, except for those boards that use a separate onboard PCIE connector for extra video card power for multiple PCIE slots.  There is only single 4 pin (ancient), single 8 pin, 8+4 pin, and 8+8 pin.


My Seasonic has two 8 pin connectors.  The first is a "Fused" 8 pin (cannot be split).  The second is a 4+4 pin, which can be split.  I have them both plugged into both 8 pin connectors (Z390 Aorus Master) completely.

Yeah that's what I'm starting to think too. I just took another look into the psu manual and it seems you are 100 percent right. I will plug it in the right way as soon as the hdd repair is done. 

 

It could also make sense that it causes the whea error due to unintentionally "undervolting" the cpu? 

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, SwarlesBrkly said:

Yeah that's what I'm starting to think too. I just took another look into the psu manual and it seems you are 100 percent right. I will plug it in the right way as soon as the hdd repair is done. 

 

It could also make sense that it causes the whea error due to unintentionally "undervolting" the cpu? 

I'm not sure, but I think when you had the 4+4 pin separated into two banks, the motherboard got confused and expected an 8+4 configuration, but was unable to draw power since only a 4 pin was connected.  That's why I was wondering what your VR VIN and VR VOUT was showing in HWInfo64.  HWmonitor is garbage and doesn't support VRM monitoring (That's why everyone should use HWinfo64 these days).

Link to comment
Share on other sites

Link to post
Share on other sites

I will correctly plug it in and then. When using my old i5 8600k I had it also plugged in the wrong way. One 4 pin in the single 4 pin connector and 1 4 pin in the 8 pin connector. It was probably still sufficient to run the i5 without crashing it. I will run hwinfo64 and let you know

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Falkentyne said:

Wait a minute.

If you bought it a year ago and it was manufactured a year or two ago, then you should be able to plug in *both* 4 pins together into the 8 pin slot as a grouped 8 pin.

The two 4 pins are keyed exactly the same way.  Just 'attach' them as one and then push them in at the exact same time.  I am 100% sure this is how it's supposed to work.

 

There is no such configuration as a motherboard that has two, separate broken up 4 pin connectors, except for those boards that use a separate onboard PCIE connector for extra video card power for multiple PCIE slots.  There is only single 4 pin (ancient), single 8 pin, 8+4 pin, and 8+8 pin.


My Seasonic has two 8 pin connectors.  The first is a "Fused" 8 pin (cannot be split).  The second is a 4+4 pin, which can be split.  I have them both plugged into both 8 pin connectors (Z390 Aorus Master) completely.

 

The reason your voltage was reported low with only one 4 pin plugged in was your CPU +12v reading was probably horribly low because you are --required-- to plug in the entire 8 pin connector on these Z390 boards.  Solo 4 pin is not supported, IIRC.  It may work for 4 core non K processors however but this is still not a supported configuration.  So your board probably adjusted the voltages to accommodate for the solo 4 pin by itself (which is not a standard configuration for Z390).  It probably has a failsafe for this, as Gigabyte has low end boards that only have a 4 pin motherboard connector.

 

When you had them plugged into both 8 pin (as 4 pin) and other 4 pin solo, the voltage may have been higher due to current balancing (load coming from both PSU pins but not doubled load).  But you were unstable because you were still drawing WAY too low +12v and the motherboard cannot handle this configuration (4 pin and 4 pin in two different PSU plug groups), if that makes sense, thus you crashed.

While winning the hdd repair I just noticed that my gpu backplate (rtx 2060 super msi gaming x) is fairly hot... shouldn't it be not even running at all while basically being in a startup mode?

Fans are not running

15756457797677021877399323759996.jpg

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×