Pro/Forums - View Single Post

Cathar · 05-27-2003, 07:52 PM

Anyone notice how the C/W values are tied to the heat load being distributed evenly over a 33x33mm area that is intended to simulate a P4?

A P4 heat-spreader is actually 30x30mm (go read the specs over at www.intel.com) and as most everyone should know by know, the P4 CPU is actually an 11.3x11.3mm silicon die under that heat spreader, so the heat really only has a ~127mm^2 area to conduct through at the narrowest point, rather than the ~1100mm^2 that was tested with.

All this results in is an entire test that is greatly skewed towards basically measuring the heat dissipation ability of the radiator, as it forms a predominant factor in the final temperature of the heat die.

05-27-2003, 07:52 PM	#6
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	Anyone notice how the C/W values are tied to the heat load being distributed evenly over a 33x33mm area that is intended to simulate a P4? A P4 heat-spreader is actually 30x30mm (go read the specs over at www.intel.com) and as most everyone should know by know, the P4 CPU is actually an 11.3x11.3mm silicon die under that heat spreader, so the heat really only has a ~127mm^2 area to conduct through at the narrowest point, rather than the ~1100mm^2 that was tested with. All this results in is an entire test that is greatly skewed towards basically measuring the heat dissipation ability of the radiator, as it forms a predominant factor in the final temperature of the heat die.