Re: [ATM] ATM Digest, Vol 28, Issue 20

[Mark Holm <mdholm@telerama.com>]

Arjan te Marvelde wrote:

> Before considering the response of the control electronics, it might be
> worthwhile to decide what temperature difference to measure.
> For minimum mirror deformation you would like to keep the gradient in the
> glass down to zero.
A couple of years ago, on this list, somebody wise in the mysteries of
heat transfer pointed out that the thermal conductivity of glass, while
low compared to metal, is still lots higher than the thermal
conductivity of air.  Therefore the mirror will have a lot less internal
gradient than the mirror surface to air temp.  Conclusion: worry about
the nasty effects of thermal boundary layer in the air and don't worry
about gradients in the glass.  Once you get the air to glass gradient
knocked down, the glass will take care of itself.  OK, if you have 5
inch thick glass, you might want to get a good heat transfer engineer to
run the numbers for you.

 The convection however remains even when these gradients
> are under an acceptable level.
> Hence the ultimate temperature difference to measure would be between the
> mirror surface and the ambient air.
That is my suggestion, though for obvious practical reasons, you
probably want to measure the temperature of the back surface.

> Would it not be better to just leave the fan on? Or is the circuit really
> meant for knowing when the initial cooldown is close enough?

Even good fans tend to shake a bit, so slowing them down to limit
vibration is a good thing.  Also, batteries are heavy.  Minimizing power
use is a good thing.  Run fast at the start to bring down an initially
high gradient, then slow down to reduce vibes and battery draw while
keeping temperature gradient under control.  Only thing is, humans are
pretty lousy thermometers, and you don't want to keep touching the
mirror to gauge it's temperature anyway.  The necessary electronics do
not appear to be too complex or expensive for many atm's, so the only
strong reason for not doing so that I can think of is that a two speed
switch may already do the job so well that even a little more complexity
and cost is not worth it.

With a controller, if your air temperature has dropped slowly early in
the evening, and then starts dropping rapidly, or if the overall cooling
rate is faster than your fixed low speed setting can deal with, then
your fan controller would notice and speed up the fan, possibly even
before you noticed the image degradation.  I know perfectly well that
this is a bit hypothetical, but the beneficial effects of front blowing
fans for eliminating boundary layer and minimizing mirror to air
temperature gradient are not even a little hypothetical.  They have been
tested and proven by careful and thorough atm's.

Mark Holm
mdholm@telerama.com

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