RE: [ATM] Temperature-changes

["Arjan te Marvelde" <arjan.te.marvelde@hetnet.nl>]

In most mirrors the edge thickness is only 15% or so more than the center. I
don't think this weighs up to the effect of additional heatloss through the
mirror edges.
In a simulation of convection cooling I did for a 250x19mm glassplate, the
temp dsitribution roughly has a cosine profile, both in lateral and paraxial
direction.

Mirror start temp is 293K, environment temp is 273K, results after an hour
of cooling:
  Mirror edge temp: 275.50K
  Mirror center temp: 277.75K
  Center-surface temp: 277.57K
Note that the cooling is most effective at the edges. I leave it as an
excercise to the reader to calculate the resulting deformation. Gut feeling
tells me it's gonna be undercorrected.

/Arjan


> Since the edges of the mirror are thicker than the
> center, all else equal I'd think they would
> equilibrate with the air more slowly.  As the evening
> cools, the delayed shrinkage of the edges would lead
> to a degree of overcorrection.  Most of us in trying
> to speed things along have been blasting the center of
> the mirror with a back fan.  Seems to me that would
> temporarily make the situation worse and lead to
> greater overcorrection.  Would it make more sense to
> ventillate the edges rather than the center?

_______________________________________________
ATM mailing list http://www.atmlist.net/