[Amps] Water cooled amp question

David Kirkby david.kirkby at onetel.net
Sat Mar 5 07:22:32 EST 2005


Japerlick at aol.com wrote:

>Sure, you could use Glycol, but why would you want to?
>
>Glycol has very high "Enthalpy" which means it can hold a lot of heat per 
>unit volume.  
>
Which is what you want for a good coolant!!  You want a small amount of 
the liquid to be able to absorb a lot of heat, with the minimum change 
in temperature.

However, I think water has one of the highest specific heat capacities, 
which is why it is a good coolant.

>It is also really poinsonious. 
>
I don't think there is such a chemical as 'glycol', but there are many 
'glycols'.

http://www.thermalfluidsinc.biz/generic2.html

has a coolant based on ethylene glycol (I can't be bothered to look at 
their data, as it is in Microsoft Word format). but ethylene glycol is 
dangerous as you say

But 'Thermasafe'

http://www.thermalfluidsinc.biz/generic3.html

is based on propylene glycol, and is said to be okay if there is 
accidental contact with food.

But I doubt any have the specific heat capacity of water, which why it 
is one of the best liquids for removing heat,

> There are LOTS of other good fluids 
>one could use, but why?  Water is great!  
>
It has its downsides, as have been referred too.

>Cray Research used Flourinert, made by 3M, in their computers.  The entire 
>computer, circuit boards and all, were inside a flourinert solution which was 
>continously circulated to draw heat out of the ECL logic.
>

Just thought I'd mention that if anyone wants to use a Cray 
supercomputer for free, you can get an account here

http://www.cray-cyber.org/general/start.php

Just don't expect much performance, as the machines are very old.

It sure checks the portability of your code, as sizeof(short)=8 !!!

A paper describing the original Cray and its cooling can be found there 
too.

-- 
Dr. David Kirkby, 
G8WRB

Please check out http://www.g8wrb.org/ 
of if you live in Essex http://www.southminster-branch-line.org.uk/





More information about the Amps mailing list