Re: ATM Cooling fans - blow sideways like the pro's

[Aplanatic@aol.com]

> One factor worth mentioning (it was mentioned, but I can't find the post, 
> sorry) is that the rate of cooling is proportional to the inverse of the 
> thickness squared - and thus, cooling from both sides simultaneously, would 
> take 1/4 of the time of cooling from one side only. So, if forcible cooling 
> helps (yet to be decided) by directed air flow, you may as well put 
something 
> in front of the mirror while it cools (such as a lid with a center fan) - 
> unless you are anxious to begin observing *before* the mirror is near 
ambient.

Nils Olof, here's my original post on thermal diffusion through glass:

http://www.system.missouri.edu/ics/staff/andy/ATM/ARCHIVES/FEB98/msg00370.html
 

To measure the temperature of a mirror you must make the 
thermal resistance of the thermometer to the glass much lower 
than its thermal resistance to air or other heat sinks/sources.  
This can be accomplished by bonding (ouch!) a thermocouple 
probe to the center of the mirror with a dab of super glue or RTV 
and then insulating the probe with some foam taped over it.  We 
use this method all the time to estimate the temperature of various 
components in an electronic assembly.  If you don't like the glue 
idea, then just tape the probe to the mirror and cover the probe 
with a decent layer of insulating foam.  The back side of the 
mirror may not come to the same temperature as the front side 
due to the large difference in emissivity of the reflective coating.
In fact, I once tried to calculate this and got about a degree C
difference.

Dave Rowe.