Idle Temp Core 2 Duo

[bloodandsoil]

Hi I just put together my new system last night. It's a C2D E6600 with a Tuniq Tower. Motherboard is an Asus P5b-Deluxe inside an Antec P180g case.

I was sitting at the hardware monitor in the BIOS and the CPU temp was at about 46 degrees celsius. Is this a normal idle temp?

 

[Assman]

don't think it's normal, my e6400 @3.2 idles 41-42C, make sure you've mounted tuniq correctly, also make sure you haven't used too much thermal paste

 

[bloodandsoil]

Thanks!

From the instructions:

3) Break-In Period
Due to the unique shape and sizes of the particles in Arctic Silver 5 conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop 2C to 5C over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.

Could be that this is the case with my system, since it's brand-new? I just put it together last night.

 

[Assman]

don't think so, 46 idle with tuniq is a little too high (@stock)

 

[Tostino]

Thanks!

From the instructions:

3) Break-In Period
Due to the unique shape and sizes of the particles in Arctic Silver 5 conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop 2C to 5C over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.

Could be that this is the case with my system, since it's brand-new? I just put it together last night.No that's not it.
Your heatsink is most likely either not mounted correctly (not making full contact with the cpu surface), or you used too much/little thermal paste.

 

[bloodandsoil]

I didn't apply my Arctic Silver 5 according to these instructions.

I just smeared an even layer of paste over the entire surface of the CPU heatspreader and then plopped the heatsink on top of it and mounted it down. I never knew that was the wrong way to do it!

So let me get this right. According to these instructions, thermal paste would NOT be applied on the entire heatspreader. It would just be smeared in a vertical line mostly in the center of the heatspreader?

That seems so counter-intuitive to me! Sure that's right?

Maybe I should let it set the way it is now for a couple days and get a solid idea of the idle temp, and then after that redo the paste according to these instructions and see if it actually improves the temp?

 

[Tostino]

It will improve, but it could be a lot better if you just re-mounted it with the proper amount of paste. From the way you said you did it, it seems like you put a bit more then you really need on the cpu. All the paste is for is to fill in the microscopic lines in the metal of the heatsink, and heat spreader so you get full contact.

What most people don't realize is that metal to metal contact is always better then metal to thermal paste to metal contact for heat transfer. So if you had a PERFECTLY flat heatsink, and cpu heat spreader, you would not want ANY thermal paste, because it would only constrict heat flow.

But alas, you will never get two perfectly flat pieces of metal, so the paste will just fill in all contact area that would not be touching otherwise.

 

[bloodandsoil]

That makes perfect sense. And when you put it that way, I now realize that I put way too much thermal paste. I really put it on thick! :lol:

I'll fix it when I get home tonight.

 

[CompuTronix]

Although the proper application of Thermal Interface Material is an important detail, your Idle temp in BIOS of 46c is not normal. It's not even close. The answer may simply be the temp is "offset" due to incorrect thermal tables stored in BIOS.

For info regarding correct temps, links, testing and results, check out the following Sticky:

Core 2 Duo Temperature Guide

Hope this helps,

Comp 8)

 

[bloodandsoil]

From the Core 2 Duo Temperature Guide:

(E) 965 chipsets may misreport Tcase and Tjunction temperatures with +/-15c offsets.

I have an Asus P5B-Deluxe (P965 chipset)...

Just to verify; the BIOS temp monitor is reporting Tcase correct?

 

[CompuTronix]

Yes, that's correct. BIOS is Tcase Idle, and should typically be a few degrees above ambient, unless Vcore is high and cooling is poor.

Comp 8)

 

[randomizer]

Or the BIOS is screwed.
@OP: Make sure you got the latest version just in case its wrong.

 

[bloodandsoil]

Since I have a P965 chipset, it's possible that my BIOS is reporting the Tcase up to an offset of +15c (meaning that my actual Tcase could be 46c - 15c = 31c) right?

Is SpeedFan reading from the same sensor? That is, using SpeedFan, would I have the same +15c offset as my BIOS hardware monitor? Or would SpeedFan show the actual Tcase temp?

 

[CompuTronix]

Since I have a P965 chipset, it's possible that my BIOS is reporting the Tcase up to an offset of +15c, right (meaning that my actual Tcase could be 46c - 15c = 31c)?

That's probably correct, however, it's possible that your Tcase is offset by +20c, and Idle is actually 26c. Without knowing what your variables are, I can't provide an accurate answer, and this still says nothing regarding Tjunction (core) temps.

Is SpeedFan reading from the same sensor? That is, using SpeedFan, would I have the same +15c offset as my BIOS hardware monitor? Or would SpeedFan show the actual Tcase temp?

Yes, but just as any software may misread temps, SpeedFan may not agree with BIOS. Fortunately, offsets can be configured in SpeedFan for corrections, so you can finally have a single monitoring program which will show accurate temps.

Comp 8)

 

[bloodandsoil]

OK I reapplied the thermal paste according to the instructions posted here (single vertical line, smeared 2 degrees in each direction by the heatsink)...

I'll report tomorrow if it changed my idle temp in my BIOS hardware monitor...

 

[Grimmy]

don't think it's normal, my e6400 @3.2 idles 41-42C, make sure you've mounted tuniq correctly, also make sure you haven't used too much thermal paste

When you say Idle.. in Windows?

When your in the Bios, it has a load on the CPU, so it's not running idle. The usage is going to be somewhat high, and the bios runs like a DOS app. So without Windows power management, your going to have higher temp reading when your in the bios settings.

So I'd say his 46C would prolly be okay, and should be lower in Windows once he has it installed.

 

[endyen]

What most people don't realize is that metal to metal contact is always better then metal to thermal paste to metal contact for heat transfer. So if you had a PERFECTLY flat heatsink, and cpu heat spreader, you would not want ANY thermal paste, because it would only constrict heat flow

Well said, and well worth repeating.

 

[bloodandsoil]

Updated my BIOS and now it's saying 37 degrees while sitting at the BIOS hardware monitor. Is 37 acceptable?

 

[CompuTronix]

To whom are you replying? 37c in BIOS is too high for Idle.

Comp 8)

 

[Spidubic]

My C2D idles between 28-30. Under full gaming load it might hit 41.

 

[CompuTronix]

To which temps are you refering, at what clock, at what Vcore, at what ambient, using what cooler, and how do you know if it's accurate? It's important to be specific. From the Core 2 Duo Temperature Guide:

Background

...Some users also may not know that C2D's feature 3 sensors at 2 different temperatures; a single on-die Tcase sensor (between the cores), and dual Tjunction sensors (one within each core). Consequently, there is much confusion regarding specifications, calibration offsets and test methods, so results can be difficult to decipher and compare. Therefore, when expressing Idle & Load test results, it's also necessary to define the variables such as:

Tcase = Idle & Load
Tjunction = Idle & Load, Hottest Core

Ambient = Room Temp
Chipset = Model
C2D = Model
CPU Cooler = Model
Frequency = CPU Clock
Load = Test Program
Motherboard = Model
Vcore = CPU Voltage...

Hope this helps,

Comp 8)

 

[bloodandsoil]

I'll have to look into that more. I'm just using the "Hardware Monitor" from within the BIOS of my P5B-Deluxe motherboard. From what I understand, this is reporting TCase Idle.

I don't know if it's accurate! How do I find out if it's accurate?

 

[CompuTronix]

Just read the Core 2 Duo Temperature Guide, download TAT and SpeedFan 4.32 from the links in the "Tools" section, then test your rig as shown under the "Testing" section, and post your "Results" and "Variables" so I can analyze your temps.

Hope this helps,

Comp 8)

 

[bloodandsoil]

Alright, this is probably not in the format that you exactly want, but I'll give it a shot anyhow.

Using TAT with 0% workload, CPU 0 and CPU 1 both report a Digital Temp that fluctuates between 25°C and 27°C.

Using TAT with 100% workload, CPU 0 reports a Digital Temp of 45°C and CPU 1 reports a Digital Temp of 43°C.

SpeedFan shows the following at idle:

System 33C
CPU 36C
AUX 127C
HD0 34C
Core 0 25C
Core 1 25C

Using 100% workload with TAT, SpeedFan shows the following:

System 33C
CPU 54C
AUX 127C
HD0 34C
Core 0 47C
Core 1 45C

 

[CompuTronix]

bloodandsoil, thanks for the temp info, but I also need your rig info, (variables), as shown below:

Background

...when expressing Idle & Load test results, it's also necessary to define the variables such as:

Tcase = Idle & Load
Tjunction = Idle & Load, Hottest Core

Ambient = Room Temp
Chipset = Model
C2D = Model
CPU Cooler = Model
Frequency = CPU Clock
Load = Test Program
Motherboard = Model
Vcore = CPU Voltage

Thanks,

Comp 8)