Modified CoolIT Eliminator vs Black Ice Extreme II
 

Many of you might have heard of the CoolIT watercooling products, especially those where air cooled TEC units are used to chill the water loop instead of a classical radiator. We at Madshrimps also had the change to compare the CoolIT products with plain Intel boxed heatsinks aswell as with high-end aftermarket heatsink, here is link to the review made my Jmke:
http://www.madshrimps.be/?action=get...&articID=6 25

I now made an adjustment so that I can hook up the CoolIT Eliminator to my own 1/2" watercooling setup. Here is an original picture of my watercooling setup:



Now modified with CoolIT Eliminator in-line with pump and cpu block, BIX2 tubing is disconnected and sticking out on top of my selfmade reservoir:



A closer look at the modification, I got rid of the pump and added 1/2" fittings:



Test setup:
Reservoir: selfmade with 1/2" fittings
pump: Laing D4 12V
tubing: clear PVC 1/2" ID
cpu block: TTIC Micro Flow
power: dedicated 10y old PC PSU, jumper started

heat exchangers:
- Black Ice Extreme + Enermax B12025112M-3M 12V 120mm + Papst 12V 120mm fan
- CoolIT Eliminator (modified)

Test Setup:
E8600 @ 4500MHz (450x10) 1,55V (BIOS setting)
MSI p35 based mainboard, sis videocard, DVD-ROM, HD, Silverstone PSU

Software used:
IntelBurn, RealTemp3

Testing methodology:
2 system states have been tested using IntelBurn: idle and load. Temperature was measured using RealTemp3. We stressed the system for 1h and then read the highest temperatures measured by RealTemp, idle temperature was read after 30m running the desktop screen in Windows, it's the minimal reached value's we wrote down. Results in our charts are the average temperate measured over the two CPU cores. Power consumption was measured using a Brennenstuhl PM 230 electrical energy meter. In real life the power consumption is not measured, it is being calculated by measuring the AC net voltage (while using the notebook's power adapter) multiplied with the measured current flowing through the mains cable and multiplied with the power factor which occurs when using capacitive or inductive loads on alternating current (AC). In any case, our device is no professional equipment, our results can be off by 5%, if not more, but at least we are not left guessing. Do understand that this is the total power consumption: cooling system + dedicated power supply. I jump wired the PSU with nothing connected and red 26 Watt, the results below are actually recalculated results of the total system usage minus the PSU power usage. No actual results as you have the changing efficiency of the PSU at different load levels, but more then decent numbers instead.



The modified CoolIT Eliminator setup uses more then double the amount of electrical power compared to a ordinary watercooling setup, we're guessing at roughly 41 Watt extra power created by those 3 TEC devices. During a long period of stress testing the CoolIT heatsink barely felt warm too.



In idle modus the cpu doesn't use too much energy so it's normal we see the Eliminator score better then an ordinary watercooling setup because of the TECs which force sub ambient temperatures for the cooling flued. At system LOAD status however the Black Ice Extreme wins it from the Eliminator by large, yet I'm quite surprised to see the system is not overheating at all. The TEC heat pump uses only 40~45 Watt (for all 3 units together) but can effectively cool a 100~150W cpu! The Eliminator might have scored better with wider channels and fittings, it has without doubt a lot more flow resistance then the Black Ice Extreme II and in the 1/2" tubing loop cooling fluid was running rather slowly.

I hope I can give more in-depth info later on...

the CoolIT unit is more compact, right?

Yes, do you think it would beat single 120mm radiator?

on par I think, but extra power draw is not worth it from your first results

I'm more then ever curious what results I can obtain once I can complete my own TEC device, the costs of CNC'ing is probable way to high though for just a test sample.

if you could have the CNC'ing done at a school the costs would be lower.
I seem to remember that it was possible in VTI @ Waregem and the price was really low there.

Ah thx good tip.

From the results above also keep in mind that without radiator I got cpu temps somewhere halfway 80's (1h test), not that bad.

wel you know what people say :)

Its not bad but it always could be beter. (damn this doesnt sound like it should xD tis nie slecht maar tkan altijd beter :-p sounds much better ^^)

First ever, a look at the thermal layer in between the TEC and cold side water blocks, they forgot to cover 20% of the TEC:

hehe :)