Building a heatsink testbed...

[Hani]

Hello,

I was wondering about building a heatsink testbed to use when reviewing CPU Heatsink/fans instead of using an actual CPU. Now, I don't have a lot of money to begin with, so this has to be as cheap as possible. Can someone tell me how to build one from scratch? What should I use as a heat source? I've heard about using several 25w resistors to make the neccessary heat. I have just found an article over at 2CoolTek where the guy used a TEC as a heater. I am really beggining to like this better than resistor. Anyway, I would appreciate any information, as well as links and advice. Also, how many watts would a 1/1.4GHz Athlon & AXP 1500+/1800+ produce? I need to know this to get the right TEC, that is, if I didn't find a better heat source.

Thanks!

 

[Sockmonkey4]

Check out what this guy used... I think it's one of the better testers I've seen:

http://www.dansdata.com/coolercomp.htm

 

[Hani]

Yeah, I've read that. But where do you get that resistor? Problem is that I live outside the US, and Radio Shack is the only electronic store available where I live. That made me think of using peltiers, since I know a company that has 40w TEC with some of their heatsinks. They've sent me heatsinks before, and I think I can get this one too.

 

[Mikewarrior2]

The DansData model is nice, but a better model is the overclockers.com artificial test bed...


Overclockers Simulated DIE test bed

Not only are they measuring "die" temp at the heat source, but also heatsink-thermocouple temperatures. Great way of showing how heatsink-thermocouples aren't as accurate as everyone believes.



Mike