Way to much on: Overheating problems...

[Ted Frederick <tedf@cadence.com>]

OK, here we go... ;]

Note I said 'seem', due to the fact that through natural convection there
will be a more uniformly distributed temperature pattern across the the
bottom of the case as opposed to a high-temperature concentration under the
processor when there is an established conductive cooling path to the 'heat
sink' (desk, stone table, etc.)

>If the processor is cooled by conduction to the case, *all* that matters 
>to the processer is the temperature of the case.  If the case is cooler, 
>it's better...period.

True, if the thermal system is one-dimensional.  It's not.  The conduction
equation is also not isotropic (unlike rocks). 

Your theory about it melting on a stone table because it 'conduct(s) *very*
slowly' is also flawed.  You are confusing the first-order thermal response
with the thermal conductivity.  Thermal conductivity determines the rate of
heat flux through a material and the thermal mass (density*specific heat)
determines the rate at which a unit volume of material will reach a
steady-state temperature.

Stone (like granite) actually has a rather good thermal conductivity (2-5x
better than wood or formica), and a good size table (due to it's long
thermal time constant) would actually behave much like an 'infinte heat
sink' (constant surface temperature).

>One experimentalist is worth ten theorists :).

One governing equation is worth ten misinterpreted observations. ;]

Regards,
-Ted


At 07:19 PM 3/9/96 +0100, Victor Kress wrote:
>>At 11:13 AM 3/8/96 -0800, you wrote:
>>The case may seem cooler when it is up on it's feet, however, the processor
>>will be hotter.
>>
>Now wait a  minute.  If the processor is cooled by conduction to the case,
>*all* that matters to the processer is the temperature of the case.  If the
>case is cooler, it's better...period.  It doesn't matter if the case was
>cooled by conduction or convection.  As someone who does calculations of
>heat flow in rocks, I can tell you that some materials conduct *very*
>slowly.  For example, one could, in theory, melt the case running a tp on a
>stone table with no convection.  At least your lap has both a temperature
>sensor, and a boiling-point governor :).  Go for whatever keeps the case
>the coolest.  One experimentalist is worth ten theorists :).
>
>In my experience (using my lap sensor), most of the heat is generated in
>the charging battery.  If you want to keep the heat load down, the best
>strategy would be to remove the battery and run on AC.
>
>----------------------------
>Victor Kress
>Geophysical Laboratory
>5251 Broad Branch Road N.W.
>Washington, D.C. 20015-1305
>(202) 686-2410 x2489
>web page: http://granite.ciw.edu/~kress
>---------------------------

*******************************
 Ted Frederick

 Strategic Programs Manager
 Worldwide Practice Management
 Cadence Design Systems, Inc.

 T:408.894.2275
 F:408.894.3484
*******************************
"The idea that information can be stored in 
 a changing world without an overwhelming 
 depreciation of its value... is false."

-Norbert Weiner (the father of 'Cybernetics'), 1950