Community
cancel
Showing results for 
Search instead for 
Did you mean: 
IBaro
Beginner
548 Views

DH67BL motherboard and Ivy Bridge CPUs compability, overheating, surpassing CPU T. case, updating BIOS and Intel Management Engine

Hello! I have motherboard DH67BL. Recently I’ve decided to upgrade my CPU and changed i5-2300 (Sandy Bridge) to Xeon E3 1230 V2 (Ivy Bridge). Although this processor is officially supported by this motherboard, I have some problems. Xeon has very high temperature with any cooler (50 C in stock and 85 C in stress test), although on the other board it had only 30C in stock and 55C in stress test. T. case of this CPU is 65C only!

I’ve updated my BIOS with the latest version using “F7 to update BIOS”, but it seems that not all components have been properly updated. Particularly, version of the Intel Management Engine in BIOS is 7.1, although 3rd generation CPUs require 8.1 at least. After I’ve put my new CPU I received a message during the first load: “This CPU and Intel Management Engine 7.x have not been validated together and are not supported. Proceed at your own risk”. I’ve tried to install the latest versions of Intel ME (8.1 and 11), but in BIOS the 7.1 version remains unchanged.

Maybe the cooling system doesn’t work properly because not all components of BIOS have been updated? Or maybe 3rd Gen processors aren’t fully supported by this motherboard, despite the official information? What will be with my Xeon 1230 V2, if it will have to work on 85 C, surpassing by 20 C its T. case (65 C)?

0 Kudos
10 Replies
AlHill
Super User
258 Views

n_scott_pearson
Super User Retired Employee
258 Views

How do you know that it is 7.1? Are you seeing this in BIOS Setup? If not, you should check there (go into Additional System Information and scroll to bottom of display). If no version number is displayed or it is displayed as 0.0.0.0, this means that the ME was misconfigured and is not operating properly.

 

What you need to do is reinstall the latest BIOS using the jumper-based BIOS Recovery Process. During this process, follow these additional rules:

 

  1. Use only USB 2.0 flash disks. DO NOT USE USB 3.0 flash disks!
  2. This must be done at least once to every USB flash disk used: On a Windows-based PC, use the standard Format tool to reformat this flash disk with the FAT32 file system and do so with the Quick Format option disabled. DO NOT USE Linux- or MACOS-based PC to do this! Resulting File System MUST BE FAT32. If FAT32 is not offered as a choice in the format tool, then the flash drive is too big and cannot be used. Quick Format option MUST BE disabled. If you are unsure if this has been done to a particular flash disk, then do it (again)! I apologize for this taking so long to run, but it is absolutely necessary!
  3. Plug the flash disk into a black USB 2.0 port on the rear panel of the PC. DO NOT USE blue USB 3.0 ports! DO NOT USE yellow charging ports! DO NOT USE front panel USB ports!

 

Let me know how it goes,

...S

 

P.S. Either you did a poor job of applying the TIM or the heatsink-fan (or water-cooling exchanger) is improperly attached to the processor.

IBaro
Beginner
258 Views

Thanks for your advice! After updating BIOS by jumper I have ME 8.1 in BIOS Additional System Information. Sadly, it didn't help to reduce CPU temperature, which remains very high (50 C in stock and 85 C in stress test). It's confusing, because I've tried to use different coolers, but the result remains the same. The same processor had significally lower T on the other board (30 C / 55 C). Besides, I've tried to use another same CPU (Xeon 1230 V2) on my board, and the problem with overheating repeated. So it seems to me that the problem still is in the motherboard. If nothing else can be done, how do you think - is it dangerous to use this CPU with surpassing its T. case (65 C) by 20 C, or any T below 100 C is acceptable? Throttling hasn't been yet.

n_scott_pearson
Super User Retired Employee
258 Views

So have you looked at the configuration of the fan speed control in the BIOS? This can certainly affect what temperatures you see at these times. These settings often have to be modified to suit the processor, the cooling solution used and the temperature vs. acoustics preference of the user.

...S

IBaro
Beginner
258 Views

In fan control settings I have "control temp" 83 C. After reducing it to 65 C nothing had changed. In stress test T began to rise above 86 C and at that moment I had stopped the test. Maybe I had to change another option in BIOS? If not, how do you think - is it safe to use this CPU if its T is very high, but doesn't reach 100 C and it doesn't throttle? I wouldn't like to change the motherboard at this moment.

AlHill
Super User
258 Views

Stop running stress tests.

 

Doc

n_scott_pearson
Super User Retired Employee
258 Views

Ok, let's be clear about a few things before anything else.

 

First of all, your processor's Control Temperature, Tcontrol, is somewhere in the vicinity of, on average, 83c. By definition, the processor can run at this temperature, constantly, for its full warranted lifetime, and it will *never* suffer any thermal degradation. Also by definition, in the absence of any mitigating information, if the temperature is at or above the Tcontrol temperature, then the fan speed control solution is required to have the processor cooling fan running at full speed.

 

In your system, the BIOS defaults the upper limit for its processor fan speed control range to the processor's Tcontrol temperature. This ensures that, at temperatures above this level, the fan will be at full speed as is required by the Control Temperature specification. The BIOS gives you the ability to lower this limit if you want your processor to remain cooler.

 

Ok, let's look at the information you have provided. If the temperature is reaching or is above the Tcontrol limit, the processor fan should be operating at its maximum speed. It is actually doing this? If not, something is wrong. Rather than spend a considerable amount of my time discussing possibilities regarding what we should do about this, let's have you answer the question first: Do you hear the processor fan screaming along at full speed?

 

...S

IBaro
Beginner
258 Views

Thanks once again for your advices! My previous fan worked at full speed, but it still could't reduce the temp. I've tried my friend's fan, bit it didn't helped too. Now I 've bought another fan, which works better. In BIOS I have 2 settings: "All-on temperature" (on which fan begins to work at full speed) and "control temp" (which fan tries to maintain). I'm confused why "All-on temperature" by default is 93 C, while my CPU case temp is 65 C only. How do you think - which values should I set for these 2 parameters? I set 55 C for "Control temp" and 65 C for "All-on". Now I have 65-68 С in stress tests, which is a huge advance! Are you sure, that surpassing case temperature (65.8 C) is not critical? It's the last thing, that bothers me. Here are my CPU specs:

https://ark.intel.com/content/www/us/en/ark/products/65732/intel-xeon-processor-e3-1230-v2-8m-cache-...photo_2019-06-15_20-46-10.jpg

n_scott_pearson
Super User Retired Employee
258 Views

First of all, forget CPU Case Temperature. There are no sensors that can properly read CPU Case Temperature so it has no bearing on this situation.

 

From a design standpoint, the All-on temperature is a threshold where the fan speed controller is supposed to take all fans (chassis fans as well as processor fan) to full speed in order to try and address the temperature overrun that is occurring. Unfortunately, the Super I/O IC used in these 6 Series board designs does not support this capability, so the parameter really accomplishes nothing (only the processor fan is affected, but this fan is already at full speed (per the Processor Control Temperature specification) at all temperature levels above the Control Temperature).

 

I explained what the Control Temperature parameter is and does. There are two additional parameters that need to be discussed, namely Responsiveness and Damping. The Damping parameter is supposed to act as a filter, smoothing temperature readings over time to prevent spurious readings from over-affecting fan speed. Unfortunately, this capability is also not supported by the Super I/O IC used on this board, so the setting of this parameter does not matter and you shouldn't waste your time trying to do anything with it. The Responsiveness parameter is the really important one. It determines how the fan responds to temperatures below the Control temperature:

 

  • Setting this parameter to Slow prioritizes acoustics over temperature. It delays (minimizes) the fan's response to processor temperature increases until the temperature is close to the Control temperature and only then does it aggressively address it. This allows a higher average processor temperature and a lower average cooling level
  • Setting this parameter to Aggressive prioritizes temperature over acoustics. It aggressively responds to processor temperature increases earlier on the control range in order to keep temperatures at lower average levels (allowing average fan speed to be higher).
  • Setting this parameter to Normal equalizes the priority of acoustics and temperature. It is more aggressive than the Slow setting but not as aggressive as the Aggressive setting.

 

In a situation like yours, it is obviously best to use the Aggressive setting.

 

But let's talk about your situation. You should be hearing the fan speed running at full speed at all temperatures above the Control Temperature setting. If this is not the case, something is physically wrong (you are using a proper 4-wire, PWM-controlled fan, right?). If, on the other hand, the fan is running at full speed and the temperature is still rising above the Control Temperature, then the overall cooling solution is insufficient. That is, either the mass of the heatsink is not high enough, the airflow of the fan is not high enough or thermal transfer from the processor to the heatsink is too slow (i.e. the TIM is bad, an insufficient amount of TIM is provided or this TIM is not properly installed).

 

So that's everything. My conclusion is there is an issue with your cooling solution. Please fully describe exactly what you are using.

...S

IBaro
Beginner
258 Views

Initially I was confused by the fact, that expensive fan of my friend didn't help - that's why I thought, that I have some problems with motherboard, but now I see why it has happened - because of BIOS settings (control temp was 85 C).

At this moment I have "Aerocool Verkho 2" cooler (TDP=110W), while "Zeon 1230 v2" TDP is 69 W.

https://www.aerocool.com.tw/en/cooler/verkho2

I set control temp to 55 C, and when the temp surpasses this value, fan begins to work at full speed (about 2000 RPM), but the temp still rises up to 63-68 C. But anyway it's better, than it was before (80-87 C), so your advices really helped a lot! Thanks once again!

If it's not critical to surpass CPU "case temp" (65.8 C), then the problem is solved, I think.

 

Reply