[mythtv-users] Interesting cooling approach

[Michael T. Dean]

On 02/29/2008 05:23 PM, Robin Hill wrote:
> On Fri Feb 29, 2008 at 05:00:31PM -0500, Josh White wrote:
>> On Fri, Feb 29, 2008 at 4:43 PM, Larry Sanderson wrote:
>>> On Friday 29 February 2008 03:36:41 pm Michael T. Dean wrote:
>>>       
>>>> I--a user who burned out 3 heat pipes on a laptop--will stay away
>>>> from /any/ design that uses a heat pipe.  (Especially since my Myth
>>>> box is asked to do a /lot/ more work than my laptop ever was.)
>>> Burned out a heat-pipe?  I had no idea that was a common problem.  Did
>>> the pipe just burst, or did it spring a leak?  Do you know if this is a
>>> problem with a lot of designs?
>>>
>>> I ask, because a lot of the high-end CPU coolers today employ a
>>> heat-pipe design, and I've never heard of one failing.
>> What would they leak?  Unless it's liquid cooled, my understanding was that
>> a heat pipe was simply a copper pipe, and that heat simply transfered
>> through the copper to the heat sink.  Is that assumption wrong?
> Yup - see http://en.wikipedia.org/wiki/Heat_pipe - the heat pipe relies
> on a fluid changing state between a liquid and a vapour.  The vapour
> moves away from the heated end, and the liquid flows back toward it (via
> either gravity or capillary action).  The phase changes then take place
> at either end, capturing and releasing the heat.
>
> Oh, and please don't top post.

Yep.  And given enough heat strain/time, that liquid/vapor will leak.  
Mother Nature doesn't like it when we humans try to trap her molecules.

Though, typically, because of the design, a failure of the containment 
would result in air leaking into the heat pipe (as it's generally 
pressurized at less than 1 atmosphere) rather than liquid leaking out of 
the heat pipe.  Then, the liquid would evaporate as it reaches its 
boiling point at atmospheric pressure.

Relying on the metal alone to transfer the heat would be far less 
efficient.  For example, copper (a good conductor of heat) tends to 
transfer heat at about 1/80 the efficiency of a heat pipe.

And the fact that I went through three before my laptop was toast (less 
than 2 years after purchase) indicates it's not just a "tried to use the 
heat pipe for 15 years" kind of thing (most heat pipes are advertised as 
having MTTF > 100,000 hours =~ 11.5 years).  It's possible that the heat 
pipes used on this particular laptop design were faulty, but I wasn't 
willing to test this theory with my new laptop, so I only use it for 
web/e-mail/watching video at 480p60 or less/ssh'ing to my "workhorse" 
systems.  Generally, the laptop CPU remains underclocked to its lowest 
frequency (half its rated freq).

Mike