Sign in with
Sign up | Sign in
Your question

Custom Cooling - Tjunction is LOWER than my Tcase???????

Last response: in Overclocking
Share
May 6, 2008 10:31:46 PM

Idle/Load
Tcase: 9c/31c
Tjunction: 6c/15c
Case Ambient: 20c
Chipset: Intel Core 2 Quad Q6600(SLACR/G0)
C2Q/C2D: C2Q
Cpu Cooler: Swiftech Apogee GT Waterblock
Coolant: Feser One Non Conductive Cooling Fluid
Coolant Temp: 5c-15c (depending on how hot cpu gets)
Frequency: 380 (x9) - 3.4ghz(load) 2.3ghz(Idle - Speedstep)
Load: Prime95
Motherboard: Abit IP35-Pro
Stepping: G0
Vcore: 1.20 Idle - 1.30 Load
Cpu-Z and Hmonitor used to determine readings

I have reached higher speeds but Feel 3.4ghz is adequately high enough but completely stable w/o having to worry about issues involving the components itself (bios settings) and focus on cooling. I can lower the temperature even more but wish to hold out until I get some more feedback (online and in person) before I continue.

As you probably guessed. MY Tjunction is lower than my Tcase...
"* Tjunction is always higher than Tcase. * Tcase is always higher than Ambient." - In my case this statement is WRONG From Core 2 Quad and Duo Temperature Guide thread

I do not use a regular watercooling setup w/pump and aircooled radiator.
I currently use a different type of method to cool the coolant(Feser One Non Conductive Cooling Fluid) which I wish to not yet reveal as this is in its testing phases and I wish to hopefully build kits or custom build in the future for Computer Enthusiasts/HC OC'ers. (Sorry but won't show any pictures)

My question is basically why is my Tcase higher than my Tjunction?
I know it has to do with the non-typical cooling and low temps but I want to get other opinions who might know the abnormal temp differences in my situation.

Thank you.
May 6, 2008 10:47:46 PM

It is a computational error, more than likely its the BIOS revision. Have you updated your MoBo Bios to the newest release?

--Lupi!
May 6, 2008 10:54:31 PM

yes i have... the latest version: 16 for my motherboard
Related resources
May 6, 2008 10:59:03 PM

Interesting. That doesnt really mean it is not the Bios. Thats a common error. DL Core Temp and see if it records the temps the same as HWMonitor.

You should always disable speed step while testing. It just lists another set of values that are useless for testing. I'd search the web and see if others experience the same error while using the same bios!

--Lupi!
May 6, 2008 11:00:59 PM

Just tested.... Same readings as hmonitor and speedfan as well
May 6, 2008 11:05:01 PM

Wow! Speed fan usually comes with that error! That they all read the same means its being interpreted as that low. Which is impossible.

Tell me what Core Temp lists the VID as!!

I have a bunch of these chips.

Just looked it up. Looks like it is a Bios bug, hehe.

http://www.google.com/search?hl=en&q=Abit+IP35-Pro+CPU+...

--Lupi
May 6, 2008 11:08:47 PM

VID: 1.2750v
May 6, 2008 11:15:12 PM

Seems to be between uGuru(software) and cpu issues. I have it installed but not running. (reason is the uguru chip on the mobo shows up as a unknown device in device manager so I installed it but never run it)
May 6, 2008 11:20:30 PM

Yup, have an identical VID q 6600 over here if you ever need specific voltages for it at whatever speed.

If you can get your paws on CompuTronix, he is our master at temp calibration and such related to that!

As long as you know the cooler is applied properly to the chip, then you can assume that the hottest core will be about 12 c from the CPU temp itself. Or somewhere around there.

But you appear to know your stuff. And have good cooling! So you look well off until that bug is fixed!

--Lupi!
May 6, 2008 11:32:39 PM

Thanks for replying, if I only knew what specifically is causing the bug. I'll try to reach computronix but if you see him on when your around the forums please tell him to go to my thread and maybe we can point out the culprit.

Thanks for the input.
May 7, 2008 5:41:21 AM

:) 
a b à CPUs
a b K Overclocking
May 7, 2008 6:21:28 PM

[b said:

jhunoh wrote:

As you probably guessed. MY Tjunction is lower than my Tcase...
"* Tjunction is always higher than Tcase. * Tcase is always higher than Ambient." - In my case this statement is WRONG From Core 2 Quad and Duo Temperature Guide thread]
jhunoh wrote:


As you probably guessed. MY Tjunction is lower than my Tcase...
"* Tjunction is always higher than Tcase. * Tcase is always higher than Ambient." - In my case this statement is WRONG From Core 2 Quad and Duo Temperature Guide thread
[/b]

jhunoh, forgive me for failing to anticipate the need to include a disclaimer informing users "This Guide does not apply to Sub-Ambient CPU coolers", however, for the other 99.9999% of the systems out there, I stand by my statement "Tjunction is always higher than Tcase, and Tcase is always higher than Ambient".

Regarding your inverted temperatures; from the Guide:


Section 9: Calibrations

Note 1: PECI (Platform Environmental Control Interface) is a BIOS feature on some (not all) recent motherboards which determines the method by which chipsets interpret and manage CPU temperature. When enabled, thermal accuracy is enhanced, and if disabled, temperatures are typically inverted, where Tcase is higher than Tjunction.

Section 15: Troubleshooting

(E) If Tcase is higher than Tjunction, then enabling PECI (if equipped) in BIOS may correct inverted temps.


If PECI isn't the issue, and since you've updated to the latest BIOS, then the problem simply points to calibrations.

Also from the Guide:


Section 5: Findings

(C) Tcase and Tjunction are both acquired from Thermal Diodes. Tcase and Tjunction analog to digital (A to D) conversions are executed by separate devices in different locations. BIOS Calibrations from motherboard manufacturers, Factory Calibrations from Intel, and popular temperature utilities are frequently inaccurate.

Section 15: Troubleshooting

(C) Any hardware and / or software may misreport Tcase and / or Tjunction temps.

(D) BIOS updates will affect the accuracy of Tcase, but will have no affect on Tjunction.


On certain motherboards, I've noted several consecutive BIOS releases which each displayed different CPU temperatures; some that were well above Tjunction as well as some that were well below Ambient. Additionally, Intel's factory DTS calibrations aren't always accurate, and sometimes aren't even close. I've also noted Cores which were more than 30c above their neighbors and didn't throttle, as well as Cores which were nearly 10c below Ambient.

Since BIOS and DTS "canned" calibrations are frequently the source of inaccurate temperatures, I feature SpeedFan 4.34 in the Guide precisely because it can provide calibrated CPU temperature AND Core temperatures in 1c increments. The solution for your rig is to follow the calibration procedure in the Guide. Substitute "Ambient" temperature for "Coolant" temperature, and use "Y = 1" for cooler efficiency.

Also, remember that the sensors tend to be more accurate at higher temperatures, so you'll have the most accurate calibration results if you don't chill the coolant below Ambient, so as to provide an environment more consistent with high-end air cooling.


Hope this helps,

Comp :sol: 
May 7, 2008 6:28:50 PM

Wow, that 1.440 Loaded Vcore volts is a famous one! My 1.3000 and my 1.3250 both need it at 3.6, My 1.2625 needed 1.440 @ 3.7 and my 1.2375 Q6700 needs 1.440 for 3.8, lol! Thats funny!

Thanks for finding this thread, Compu!

--Lupi
a b à CPUs
a b K Overclocking
May 7, 2008 6:32:26 PM

No problem... always glad to help sort out these temperature problems. :D 
!