[Beowulf] Innovative liquid cooling

Lux, Jim (337C) james.p.lux at jpl.nasa.gov
Fri Mar 1 14:19:17 PST 2013

On 3/1/13 8:16 AM, "Joe Landman" <landman at scalableinformatics.com> wrote:

>On 03/01/2013 10:45 AM, Prentice Bisbal wrote:
>>  Using this numbers, Water is 24x better than air Mineral oil is 6.75x
>> better than air 3M Novec 7000 is only 3.125x better What is clearly
>> much better than the other alternatives for heat transfer. If only it
>> wasn't conductive, and corrosive (rust, anyone?) I hope this makes up
>> for my earlier exaggerated numbers. Prentice
>Technically, pure water is not a good conductor.  More of a dielectric
>insulator.   Its when you get those nasty impurities and ions in it,
>that it becomes an ok conductor.  Since its such a good solvent, its
>rather hard to keep those nasty ions out.
>As far as I remember, one of the hardest parts of dealing with cooling
>liquids are the impurity buildup issues.  Unless you are dealing with
>Flourinert (see www.sisweb.com/referenc/msds/fc70.pdf ) which is
>goodness embodied ... unless you burn it ... then its pure eeeevuuulll.

FC70 is pretty pricey, as I recall... $250-300/liter or something like

Compare mineral oil at $1/liter


More information about the Beowulf mailing list