Omega Thermocouple Vs. Onboard Sensors

[MaxxxRacer]

Please Read the following.. You will now understand why I so emphatically state that the onboard sensors are totaly useless.


Test System
DFI nF4 Ulta-D w/ Vcore, Vdimm and Vtt mod
AMD A64 3000+ Winchester with T-type thermocouple mod to IHS (lapped IHS)
2x512mb OCZ PC4800 EL plat
ATI X800GTO
Omega HH506RA Multilogger Thermometer w/2 T-type thermocouples

Cooling System
Cooltechnica MP-05 SP Pro
Danger Den Maze4 GPU
Cooltechnica AquaExtreme 50Z 12VDC Pump
Danger Den Black Ice Extreme 2
2x Yate Loon 120mm fans @12v
Daul 5inch Bay Res
Clearflex 60 1/2" ID 3/4"OD tubing
Loop Order - Pump -> GPU -> CPU -> Radiator -> Reservoir


Test Method
Ambient temp taken with 36gague T-type thermocouple mounted in front of intake fan of radiator
Actual CPU temp taken with 36gauge T-type thermocouple embedded in CPU IHS with Omega bond OB-101 thermal epoxy
CPU Load created with S&M v1.7.6 beta with FPU test
Temperatures were allowed to settle for approximately 30 minutes before readings were taken.




Thermocouple

Ambient
---------------------
23.2 +-0.1C

Idle
---------------------
27.7C (with 22.7 ambient)

Load
---------------------
32.2C

dT (CPU Load to Ambient)
---------------------
9.0C



Onboard Sensor

Idle
---------------------
27-29C (ambient between 22.7-23.2)

Load
---------------------
40 +-1C (jumps around)

[[XC] Lead Head]

It got idle right, but the load is way off. Why is the idle near actual but load is off?

[MaxxxRacer]

lol.. posted in wrong section.. but i suppose it will get more views here.

the onboard sensors getting the idle right was a mere fluke... (no pun intended)