Re:ATM Beyond the fan

["Arjan te Marvelde (ELN)" <Arjan.te.Marvelde@eln.ericsson.se>]

> come out colder than ambient.  As Alan
> Alder noted in his S&T article, the biggest thermal gradient, 
> once a fan has been running for a few
> minutes, is not likely to be between the mirror and the air, 
> but internal to the mirror.  An

If this were true, mirror cool-down times would be much shorter. Unfortunately the majority of temp drop is in the boundary layer of air, and this is mainly caused by the low heat capacity of the gas. 
For my 10 by 3/4" plate mirror i've calculated a cool-down time down to 1% of the original T difference to be only 8 minutes ....... 
...... provided that the glass surface is kept at ambient T. This is in fact only possible with something like running water at Tamb, air just cannot contain enough energy.

Some rule of thumb figures for air cooling:
  free convection : 5-10 W/m^2
  forced airflow  : 10-100 W/m^2

If you want the formulae, link to:
 http://home.hetnet.nl/~artm/atm/primer/primer.html#mechanical

/AtM