Archived from groups: microsoft.public.windowsxp.hardware (
More info?)
"w_tom1" wrote:
> If a processor heats in a 70 degree F room to cause reboots,
> then semiconductor or other hardware is defective. In fact,
> Intel CPUs even back to some 486s simply slow down if too hot;
> no reboots and no damage.
>
> Posted are other myths about heat such as a 'need' for
> thermal compound. Thermal compound typically results in a
> less than 10 degree temperature reduction in CPU heat - a
> reduction that should be totally irrelevant to whether a CPU
> overheats and reboots. More important reasons for using
> thermal compounds are heatsinks not properly machined. If
> thermal compound results in greater temperature reduction,
> then the computer assembler should first start with himself as
> reasons for the problem. We recommend thermal compound because
> too many computer assemblers buy defective heatsinks. They
> buy on price rather than degrees C per watt.
>
> Yes a stuck fan can cause problems. But will more cooling
> solve a hardware problem? Of course not. More fans cure
> symptoms.
>
> Will a hotter processor (still within manufacturers spec
> range) fail faster? Yes. And then we apply numbers to the
> conclusion. So it will fail in 100 years instead of 150
> years. Who cares? Those who make claims without numbers
> promote half truths. When numbers are applied, claims about
> 'heat causing premature failure' becomes irrelevant.
>
> Heat is a problem when IC temperature exceeds manufacturer
> limits. An IC operating within manufacturer numbers that
> fails when warmer (and still within specs) is defective and
> must be replaced. Heat is the diagnostic tool to find
> defective hardware - including a stuck fan or a CPU heatsink
> that was not properly machined.
>
> Applying thermal compound to fix a defective heatsink does
> not solve the real problem- a human who buys heatsinks on
> price rather than first obtaining thermal conductivity
> numbers. Heat is a tool to find defects. Too many computer
> assemblers would rather cool their mistakes with too many fans
> rather than deal with THE reasons for that failure. If the
> processor gets so hot as to cause reboots, then we have a
> defect that must be fixed. Too much heat did not cause that
> processor reboot. Something defective causes too much heat to
> cause a reboot. Use heat to find and replace defective
> hardware - such as heatsinks machined so poorly as to require
> thermal compound or too much thermal compound between CPU and
> heatsink.
>
> What is too much heat? Temperatures that exceed those
> manufacturer numbers. Old engineer's guideline for
> semiconductors: touch it and it does not leave skin, then it
> is not too hot.
>
>
> Kold wrote:
> > That all sounds good, but processors do get heated up and
> they do
> > cause a reboot. Things like clogged up heatsink fans, lack
> of
> > lubrication in the fan, too little or ineffective conductive
> grease on
> > the processor. All these problems can crop up on a perfectly
> built PC.
> > This is why PC are "cooled", because they fail when there is
> too
> > much "heat".
w_tom,
either you’ve never used AMD processors, or you read to
many books. A fan failure will kill a modern AMD processor in about 30
minutes or less, if you don’t believe that, go ahead and try. Also,
have you noticed that AMD puts thermal compound on it’s heatsinks? Go
ahead and peel that off and see how long that lasts.
--
http://www.WindowsForumz.com/ This article was posted by author's request
Articles individually checked for conformance to usenet standards
Topic URL:
http://www.WindowsForumz.com/Hardware-computer-shuts-sudden-ftopict235752.html
Visit Topic URL to contact author (reg. req'd). Report abuse:
http://www.WindowsForumz.com/eform.php?p=718038