I hope he doesnt mind me posting this, but here is what he sent me in PM so you get an idea of what this guy is driving towards:
Quote:
|
Originally Posted by laxman
Well Ill do it if you want and if the members of this site accept it to be good enough. Right now I am look at a .001 res and .01 accuracy for the thermometer (not including the .1% error for the probe), probably around .75% for Dp, and 2% error for the flow meter over .2 to 20 GPM. Overall the percent error will be around 3% error or +- .0033 on a .1 C/W water block. To me this acceptable but we have a tough crowd here. Also for the heat die I will be using this silicon heater: http://www.minco.com/heater_config/model.aspx?PN=HR5293 capable of 145 watts. The good thing about this heater is that the effective heating area in only about a mm bigger then 14mm^2 die. So this allow for lower heat loss but I will measure and account for this too. Insulation will by 1/8 phenofilic resin (2 layers) and spray insulation. I will also try to create some kind of real cpu simulation, I have to talk to PH about that. Hows that sound? |
he does seem open to suggestions and info, so feel free to speak up
(not that procooling has ever had issue with that
hehe )