Re: [ATM] Thermal mirror deformation -fan on both sides

["tony gondola" <acgna@comcast.net>]

> Tony wrote:
>> I don't think it's a matter of how important the work is but is it a
> useful,
>> validated tool? It's interesting but I don't think anyone really knows
> what
>> to do with it. Everyone knows that performance will be degraded while the
>> mirror is equalizing to ambient. My personal feeling is that actual field
>> measurements would be more useful.
>
> I know that most of the ATMs are pragmatic, and are more inclined to
> building than theorizing.
> Personally, I believe that not only the observation that methods work is
> important, but also the reasons why they work. Understanding the 
> underlying
> mechanisms can open up completely new ways of doing things, which you 
> would
> not think of while building by intuition. (Think of another FEM based 
> tool:
> PLOP)

I don't disagree with that at all.

> There have been many discussions on the list about thermal behaviour of
> mirrors, most were driven by qualitative arguments and subjective 
> evidence.
> What I have tried is to quantisize behaviour, to give at least a feeling 
> for
> the magnitude of the effects of convection and radiation. Of course the
> model is only as good as it is, but at least it gives you a vehicle to
> analyze the effect of changes you make to it. For example, I have been
> playing with non-uniform convective cooling profiles.

The thing that bothers me is when ATMs take a vary preliminary result and 
run with it without really understanding the limits of the modeling. I've 
already seen glimmers of that in some of the responses to the paper in other 
groups.


> A corroboration of the results can be found in the afore mentioned
> observations by Wolfgang Rohr (www.astro-foren.de/showthread.php?t=6454).
> This leads me to believe that the results of the FEM model at least are in
> the ballpark.


I'm sure it is based on the limits of the model.


>> a flow of air that surrounds the mirror, front back and sides. The actual
>> airflow is much more complicated then Jan's simple infinite diameter disk
>> model suggests and is also subject to turbulence and dead areas induced 
>> by
>
> Yes, of course it is, but as a first approximation of cooling times it 
> still
> gives a good estimate of time. The simple model assumes a medium with very
> high heat capacity, which free-flowing air certainly is not. Lots of 
> forced
> airflow approach it though.

I can believe the basic cooling times but I'm not sure about the rest of the 
conclusions. The biggest problems I see are:

Assuming that the front surface is only cooled by radiation and natural 
convection.

Absence of edge effects.

Unless you have a seal all around the edge of the mirror a forward facing 
rear fan will still induce a turbulent airflow over the front face, 
especially at the edge zones. I think this could give a significantly 
different result to the final gradient, something perhaps closer to the 
non-forced cooling result where the final gradient is in fact, very small. 
It would certainly have a large impact on the actual deformation of the 
surface and it's associated effects on the image. Unless you can find a way 
to approximate those effects then I think it's hard to put much weight into 
the results so far. Where I think the model might be the most useful is in 
predicting the effects of raditional cooling on a naturally stabilized 
mirror. That was pretty much the core question behind the threads on this 
subject. I'd love to see you run that case especially if it's possible to 
define a range of unobstructed sky angles as an aide in optimizing the 
design of radiative cooling blocking shrouds.




> The FlexPDE model discussed later in the article gives you a more 
> elaborate
> picture. Just try it, and play with it: FlexPDE is not so hard to
> understand, especially when you have an example to work with:
> home.hetnet.nl/~artm/atm/resource/cooling.zip
> You can download FlexPDE from: www.pdesolutions.com/ along with the free
> student license you will need to run it.

I'll take a look at that.

Tony

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