Need Help! Overheating CPU at Boot 100C

[pjo6493]

I've spent the last 3 days trying to fix this and I can not figure it out for the life of me. I've list my system specs bellow.

The overheating heating issue is not consistent and only arise after I change a few setting in the BIOS in order to prepare the CPU to begin overclocking. I have not yet adjusted the Core Clock Speed Multiplier nor the Core Voltage, everything is still at "Auto". The only settings I change are EIST, Turbo Boost, and Intel Enhanced Turbo by setting them to "Disabled". From what I understand about these setting it should be fine to disable them and the CPU should just stay all base clock speed (4.0 GHz).

After I disable those setting and boot in to windows I immediately open Core Temp and watch the temps. As Windows continues to boot and launch startup apps I see the CPU hit upward to 90-100C and periodically hits 80+% CPU load, where normally the hottest core only hits 60-65C at boot. Once all the startup application are running I open Corsair Link (Software to control the H115i cooler) I see water temps between 45-50C, where normally it is as high as 36C after boot. Even after hard rebooting the system, the problem still persists.

If I go back in to the BIOS and re-enable those setting the boot temps improve slightly 70-80C, even after letting the system sit to cool off with it completely shutdown for about 15-20 minutes.

The only solution I have found that brings it back to normal is to: restore factory defaults in BIOS, shutdown the system, cut power to the PSU with switch on the back, and wait ~20 minutes. However this only brings me back to where I started and gives me no answers as to the cause or solution.

I have even tried re-flashing the BIOS to the most current version (7A11v2B) to no avail.

For all I know I could just be doing something wrong and just not know it. If anyone has any ideas or solutions it would be most appreciated.

System Specs:
CPU: Intel Core-i7 6700K
CPU Cooler: Corsair Hydro Series H115i AIO
MB: MSI Z170A Krait Gaming 3X (MS-7A11)
RAM: 16 GB (4x4GB) G Skill Ripjaws V DDR4 3200
GPU: EVGA GeForce GTX 1080 FTW Gaming Hybrid
PSU: EVGA Supernova 750 G2
OS: Windows 10 Pro x64


I just want to clarify that I have checked, double checked, and triple checked the following possible causes and solutions:

• 
  ■ Is the CPU cooler correctly and firmly mounted? Yes!
  ■ Is the AIO's pump working? Yes, as far as I can tell. I can feel some vibrations.
  ■ Reapply thermal paste? Yes, used Arctic Sliver 5 and used the pea size method.
  ■ Clean the old thermal paste off before reapplying? Yes, used isopropyl.

 

[terry4536]

There are only two scenarios that seem likely. The first is that the CPU cooler isn't making good contact with the processor. The cooler bracket may be damaged or faulty. The surface of the cooler or processor may be deformed (not flat).

The other scenario is that too much voltage is being applied to the processor. This could be a mistake in the settings or a faulty motherboard. Recheck the BIOS overclock settings.

 

[terry4536]

The thermal paste probably isn't the problem. Even poorly applied thermal paste wouldn't result in 100 C temperatures that quickly. This is more like the CPU cooler wasn't installed or wasn't in contact with the processor.

 

[terry4536]

Reset the the BIOS to the default settings. Then see if the CPU temperature climbs to 100 C again. That reset should delete any overclock.

( I see that you tried this.)

 

[terry4536]

One thing that does come to mind is the situation involving I7-8700K and the 300 series motherboards. The combination of multi-core enhancement and XMP resulted in an automatic overclock. To be honest, I don't know if that is even a possibility on this motherboard / processor combination. But it something to consider.

 

[terry4536]

It sounds like there may be a problem with the Corsair H115i AIO. Is this a new build? Do you have experience with this cooler to know if is plugged or if the pump is running normally?

 

[terry4536]

You can use a straight edge (use different angles and look for light under the straight edge) on the two surfaces to determine one of the surfaces is deformed.

 

[J_E_D_70]

If I read your post right, you hit 60C on a cpu core on boot? That’s nuts. Sitting in the bios, you should be about 32C unless you are operating the pc in an oven.

What is the vcore showing in bios?

I’d bet that if not a warped block, the impeller snapped off the motor so it vibrates but doesn’t do anything since it can’t move liquid.

 

[terry4536]

Do you have another CPU cooler that you can install? Another operational cooler would work as a test on the status of the Corsair H115i. If it keeps the CPU temperature under control, it would confirm a problem with the Corsair H115i.

 

[pjo6493]

Thank you for responding and I'm sorry for my delayed response.

There are only two scenarios that seem likely. The first is that the CPU cooler isn't making good contact with the processor. The cooler bracket may be damaged or faulty. The surface of the cooler or processor may be deformed (not flat).

The other scenario is that too much voltage is being applied to the processor. This could be a mistake in the settings or a faulty motherboard. Recheck the BIOS overclock settings.

The thermal paste probably isn't the problem. Even poorly applied thermal paste wouldn't result in 100 C temperatures that quickly. This is more like the CPU cooler wasn't installed or wasn't in contact with the processor.

For the most part I have eliminated the CPU cool as the issue because my temps are normal if I don't change the settings in BIOS that I have described and will return to normal if I use the method I also described. As for the motherboard be faulty I think that may be the case.

Reset the the BIOS to the default settings. Then see if the CPU temperature climbs to 100 C again. That reset should delete any overclock.

( I see that you tried this.)

Yes, I have done this multiple times and it does fix the issue, but as I stated it just brings me back to where I started. Plus I haven't even started to overclock yet all I did was disable EIST, Turbo Boost and Intel Enhanced Turbe, everything else is at default.

One thing that does come to mind is the situation involving I7-8700K and the 300 series motherboards. The combination of multi-core enhancement and XMP resulted in an automatic overclock. To be honest, I don't know if that is even a possibility on this motherboard / processor combination. But it something to consider.

I think this could be possible but unlikely because I normally have XMP enabled and the system run fine if I dont mess with the BIOS Settings.

It sounds like there may be a problem with the Corsair H115i AIO. Is this a new build? Do you have experience with this cooler to know if is plugged or if the pump is running normally?

As I said before, I'm sure the cooler is not the issue. No this is not a new system, I built it back in November 2016. This is the second PC I've built myself, so I'm not new to it. My first system I disabled these same settings and had no issues.

You can use a straight edge (use different angles and look for light under the straight edge) on the two surfaces to determine one of the surfaces is deformed.

I had originally thought that the cooler maybe at fault so I unmounted it. I noticed the that back plate/mounting bracket (the part where the standoffs screw into) stuck through the motherboard a bit far and thought it might not be getting good contact with the CPU. So I sanded it down about a 1mm, I used a caliper to measure insuring I sanded them down evenly.

If I read your post right, you hit 60C on a cpu core on boot? That’s nuts. Sitting in the bios, you should be about 32C unless you are operating the pc in an oven.

What is the vcore showing in bios?

I’d bet that if not a warped block, the impeller snapped off the motor so it vibrates but doesn’t do anything since it can’t move liquid.

Yes, under normal conditions while in BIOS the temps read in the 30s, but after I change the disable the EIST, Turbo Boost and Intel Enhanced Turbe then save setting and reboot, temps in BIOS read 60c and slow increases to 70+. I see about 1.200v in BIOS. I don't think that the pump is bad because the outlet hose gets quite warm as does the radiator.

Do you have another CPU cooler that you can install? Another operational cooler would work as a test on the status of the Corsair H115i. If it keeps the CPU temperature under control, it would confirm a problem with the Corsair H115i.

All I have is the Intel stock cooler and I have already tried that. It didn't help much because under normal system conditions the temperature difference was to much to conclude anything.

 

[terry4536]

One thing you might try is to use the automated overclock. First select one of the lower overclocks. Then after the overclock is applied, monitor the CPU temperatures. If the temperatures remain under control, then you will know that an overclock can be applied. Then you can work on the issue going forward.

If it the temperature problem returns, then it is just another data point.

after I change the disable the EIST, Turbo Boost and Intel Enhanced Turbe then save setting and reboot, temps in BIOS read 60c and slow increases to 70+

Then stop doing that. Leave all of the settings at the default except setting the multiplier. Try that as above, and monitor the temperatures.