Redundent cooling system for CPUs.

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Redundent cooling system for CPUs.

Post by Bluefront » Tue Jan 28, 2003 3:34 am

This system works....running it now on a Celeron2.0 sitting on a Gigabyte 667 Ultra, with an Alpha 8942 heatsink. Should the main cooling system fail, or the CPU start to overheat for any reason, the secondary system starts up and the CPU is saved. Here's the specs.

The main system cooler is actually the rear case fan (a PCToys CrystalMaxx, variable speed with an external temp probe at the base of the heatsink). The airflow is directed onto the heatsink by means of a shroud arrangement (the shroud is off a Zalman 5700 cooler). This system is so efficient, the CPU usually runs about 4c over ambient. Right now at an idle it's sitting at 27c. At 100% cpu usage it barely reaches 38.6C.

Here's the redundent system....a 60mm fan mounted directly on the top of the Alpha heatsink. It normally just sits there, not running. The rear case fan blows through it. I'm using a DigitalDoc5....one of it's temp probes is at the base of the Alpha's cooling pins. Should that probe sense a temp over 39c (that temp is adjustable), the little 60mm fan starts up, and since it's also drawing outside air through the duct, by itself it is capable of cooling off the CPU.

You could configure this fan to turn on at high CPU usage...such as intensive game playing....so both fans would run at once. Since I don't usually play games, this secondary fan rarely turns on for me. Neat setup.....provides an extra margin of safety without adding to the noise. :D

Oh, heh. This is an original design of my own. I swear I never saw it used anywhere else. No patents on it so feel free to build one yourself........

TheMuffinMan
Posts: 146
Joined: Sat Jan 04, 2003 3:45 am
Location: New York, NY
Contact:

Post by TheMuffinMan » Tue Jan 28, 2003 7:15 am

A Celeron 2.0 should save itself - I know the P4s do. They automatically slow (underclock &/or undervolt) themselves down until they hit a safe temperature (this is, again, P4s).

Does anyone know for sure if the Celeron's save themselves?

Post Reply