Ralph Shumaker wrote:
Gregory K. Ruiz-Ade wrote:
On Oct 3, 2007, at 7:46 PM, Richard Reynolds wrote:
you didnt say here what type of video hardware you have but when my
beyondtv(widndows variant) started rebooting i learned that the fan
on my video card had failed while it appeared to spin, and was clean
it was not working took a while to track down, but it was a fast fix
once i figured it out
I very purposefully purchased a fan-less graphics card with a big fat
heatsink on it. While doing a physical inspection of the system
during diagnosis, I noticed there was no dust on any of the internal
components, just on the intake grills and the fans themselves.
Does anyone make a case that does not pull air in from the outside?
This is the source of most of the dust.
For some specialized systems such as mini-ITX, yes. Also some industrial
PCs.
With all the cooling aids that I've seen now, I don't think there is
much need to continue to pull air in from the outside. The one cover
that comes of to give access to the guts of the PC could have a large
version of that radiator-like cooler with fins on the interior for
gathering heat and fins on the outside for expelling heat with the
coolant tubes making the transfer even more efficient. A fan inside
(mounted near the top, but not mounted to the fins of the cover) could
blow hot air down the interior fins while a fan outside (below the
exterior fins) could blow cool air up the exterior fins. The power
supply could do something similar. The only dust would be on the
exterior fins and fans (*away* from the electronics). And if done well,
the exterior fins could be easy to clean. Am I way off? Is this just
wishful thinking?
It's doable. Things like water cooling and heat pipes show the trend
toward passively cooled systems. The real problem is ensuring good
mixing of the air inside the case and having enough surface area. A
couple useful equations I've used:
delta T (degrees F) = 3.16 x Watts / Cubic Feet Minute air flow
Thanks to <http://www.chassis-plans.com>
Power (watts) = k x Area (meters squared) x delta T (degrees kelvin) /
thickness (meters)
where k is thermal conductivity in Watts/meter-kelvin
A sample k value for steel (AISI 1020) is 65, aluminum (alloy 1100) is
220 and copper is 380. Diamond (type IIa) is 1850. You can see this is
why heat sinks are made mostly of aluminum and copper until diamond
becomes more affordable.
Gus
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
