View Single Post
Unread 11-27-2005, 04:57 PM   #25
RoboTech
Cooling Savant
 
RoboTech's Avatar
 
Join Date: Sep 2002
Location: Cincinnati, OH
Posts: 229
Default

No guarantees about the 70~80%… that’s just a typical value (but one I have verified during my own tinkering)

A nother huge variable that is sometimes overlooked is the water-cooling system. You can't expect a WC system that was designed for a single CPU application to work effectively with a TEC setup. To work well, it will need to be up-sized ~4X (CPU heat + TEC heat). A TEC creates a dT from one side to the other. The cooler you can keep the hot side, the cooler the cold side will run (other things being equal). So to get the most out of a TEC setup will require a very good (big radiator(s), multiple fans, lots of flow, etc) WC system (and low ambient temps don't hurt).

Edit: I also agree with what was just posted refrencing ls7's work... (for those, who think this sounds contradictory). Just to clarify: many ways to approach the task. (1) for max performance will require uber cooling system and finding the TEC's sweet spot, but (2) a very effective and much more efficient system can be made by undervolting the TEC, which will generate much less waste heat and won't require an ultimate water-cooling system. Just depends what your end goal is and how far you want to push things IMHO...

Last edited by RoboTech; 11-27-2005 at 05:07 PM.
RoboTech is offline   Reply With Quote