Danger Den Water Box Plus Case (Part 5)ccokeman - September 9, 2007
» Discuss this article (15)
Testing this case and system components are really the only way to validate whether or not there was improvement in the results. Testing the CPU temperature and graphics card temperatures while under load and comparing the results to the same tests while the components were air cooled is the way we will verify any improvement. Danger Den states that an up to 30% increase in video card clock and memory speeds are possible while using the full coverage blocks. We shall see! We will be using Stressprime 2004 Orthos Edition for load testing the CPU and system memory, while at the same time running 3DMark06 and looping it 5 times to heat up the video cards, ensuring a fully loaded system. The fluid we are using for the project is MCT-5. This coolant is manufactured exclusively for Danger Den by Midwest Cooling Technologies. The reason for choosing this product for our project is that it is a non-conductive fluid. Being non-conductive means that should a leak develop, the system should not suffer a meltdown because of the leak. Electricity and liquids usually do not play nice together, so this bit of protection is a welcome bonus. We will test the fluid to see if it is conductive or not, as well as a couple other popular cooling fluids. Let's get on with the testing.
- Processor: Intel E6700 360x10
- Motherboard: DFI NF 680I LT
- System Memory: Mushkin HP6400 2x 2048MB
- Video Cards: 2x EVGA 8800GTS 640MB
- Hard drives: 2x 74GB Western Digital Raptors in RAID 0
- Power Supply: Mushkin XP650 Enhanced
- Opticals: Benq DW1655 Lightscribe DVD-RW, Sony DVD-ROM
- Operating system: Windows XP SP2
- Greenlee DM-300 Digital multimeter
- Kestral 4100 Pocket airflow tracker
Having a liquid be non-conductive is a concept that is tough to wrap your head around. Many of the liquids you would use as a coolant are conductive and leaks would be a disaster if they were to happen. I tested three liquids that are commonly recommended, MCT-5, distilled water and distilled water with Zerex Super Racing antifreeze. I will test for any continuity through the liquid to see if a charge will pass through the coolant. The results kind of surprised me. The MCT-5 really does not pass a charge through it in our testing. On the other hand, the distilled water and antifreeze mixture both allowed a charge through them.
The balance of our testing is based on the temperatures generated by the hardware in this case. While a comparison between air cooled and liquid cooled may seem a little unfair, it shows the capabilities and load capacity of the systems in question. The item we will test are the three items that are liquid cooled vs. the temperatures while air cooled. The voltage on the CPU is 1.61v set in the BIOS and no voltmods to the video cards. All temperatures are in degrees Celsius with the lower value being the winner in the tests
Airflow through the case was measured by checking the the airflow from each fan and using simple math. What goes in must come back out. The two Papst fans used on the radiator are rated at 55 CFM each. These are blowing into the case, while there are three 80mm fans pulling air out of the system. Two of the fans are rated at 27 CFM and the power supply fan is not rated, but is variable depending on heat load. The total intake was rated at 110 CFM but measured closer to 100 CFM. The two case fans mounted to the top measured above spec at 30 CFM each, while the power supply measured 31 CFM under load. The exhaust comes up a little short, but with a slight improvement in fan selection, the airflow will become balanced.