[ATM] Fan locations & function for a 32

[Dr. Wm. D. Hanagan, Jr.]

Bob Drew wrote:
> I accept the concept of sweeping away the boundary layer from the 
> primary mirror surface. And I’ll use some fans under the mirror back.
......
> 1. Should the flow be across the mirror front be from one side to the 
> other side? Or should it be vertically? (From the lower side of mirror 
> box, exiting on the top? or vice versa? ) The scope will have a full 
> shroud as a minimum or possibly a rigid tube later.
>
> 2. Since its operating in conjunction with the above (and without 
> starting a debate) should the fans (imbedded in the cell underneath 
> the mirror) suck or draw the air down and vent out the back of the 
> box, or blow on the back of the mirror.
>
> 3.Are there 3” diameter fans that would be able to maintain flow 
> across a box about 40” wide, or would a 4” be nec.? Brand names and 
> models also appreciated
>
> The mirror is a 32” f4, 3” thick weighing about 200 lbs. 
.......

Bob,

Here are a few recommendations.

Crossflow Fans

If you use a rectangular rather than a tube shaped mirror box, or you 
can otherwise accommodate them, I would strongly recommend using one or 
more "crossflow" fans. Crossflow fans are designed to produce an airflow 
pattern that's almost exactly what we need to sweep the thermal boundary 
layer from in front of a telescope mirror. Here's a link to a ball 
bearing crossflow fan that is relatively inexpensive when you consider 
the width of the air stream it produces. The speed of this fan is 
controlled by the DC voltage applied (between 6V and 12V).

http://www.coolermaster.com/index.php?LT=english&Language_s=2&url_place=product&p_serial=STF-B01-E1

Much longer models are available from other manufacturers that produce a 
wider swath of air flow. A Google search on "Crossflow Fans" will turn 
up a large number of other options to consider, many of which are 
available in 12 Volt models.

Square Fans

If you must use a tube, especially a smaller diameter tube, the 
straight-line crossflow fan can be difficult to integrate into your 
design. The more common square fans can be used in that situation to 
sweep the surface of the mirror. Here's a link to an 80mm ball-bearing 
fan (Sanyo Denki San-Ace 80 X 15 mm) which is only 15 mm deep from front 
to back, so it will fit into tight spaces.

http://www.sidewindercomputers.com/sandenpetace.html

The 15 mm depth allows these fans to be mounted on the INSIDE of a 10" 
I.D. tube which houses an 8" diameter mirror without blocking the 
incoming light path for objects near the center of the field of view. As 
the primary diameter increases, the space constraints on an inside 
mounted square fan ease up as you go to larger tube and mirror 
diameters, provided the typical 1" clear space exists between the mirror 
and the inside of the tube wall. Even with inside mounted fans, the tube 
itself will still need vent holes in the tube wall for each fan. Those 
vent holes need to be at least as large as the inside opening of the fan 
cage in order not to seriously diminish the volume of air moved by the fan.

An inside mounted fan can be useful to avoid a variety of mechanical 
conflicts and space constraints, and it makes it more convenient to 
mount a filter on the outside of the tube. However, I would consider the 
use of a filter as experimental.

I wouldn't expect square fans to sweep a path much wider than roughly 
1.5 times the diameter of the fan blade. Of course, the factor you use 
depends greatly on how low a minimum airflow you can tolerate over the 
surface of the mirror. You might want as many as 6 inlet and 6 outlet 
fans for your 32" mirror. At $18.95, using 12 of these to ventilate a 
32" mirror would be a little expensive compared to using crossflow fans. 
Of course, the cost for the fans would be nothing compared to the 
overall cost of a 32" telescope.

The Airflow Pattern Across the Face of the Mirror

You raised the question of whether the air coming out of a square fan 
will travel the full distance across the face of a large mirror. The 
airflow pattern of square fans is cone shaped and the air speed 
diminishes fairly quickly as you move away from the fan. The airflow 
pattern of crossflow fans is more linear and holds together over a much 
greater distance out away from the fan. However, airflow takes the path 
of least resistance. The combination of inflow and outflow fans that you 
mentioned will put the path of least resistance directly across the face 
of the mirror. If you keep that arrangement, I think you can expect most 
of the air coming in to flow past the face of the mirror to the outflow 
fans, even if you use square fans.

Direction of Airflow and the Dust Problem

Using fans introduces so much airflow through the telescope tube that 
natural convection is likely to make a trivial contribution to cooling 
the mirror and destroying the boundary layer, if indeed it contributes 
at all. One concern introduced by using fans is the increased amount of 
dust that passes through the telescope. A lot of dust can get kicked up 
near ground level on a dry day. To draw as little of that dust into the 
telescope as possible, I would recommend arranging for the air to flow 
from top to bottom. I would duct the airflow from the vent fans out the 
back of the box only if a minimum duct cross section of 1.5-2" X D can 
be accommodated. Otherwise I'd run the air directly out the bottom side 
of the mirror box.

Filtration

The airflow that you get with most crossflow and square fans drops 
precipitously if you use a filter having the same area as the fan 
intake. However, a box housing a filter with a vastly larger surface 
area, or even a serpentine air inlet, might be usable for filtering dust 
out of the incoming air stream. I'm still experimenting with approaches 
for filtering the air and I can't give you any specific instructions on 
this except to say that if you use a filter, the intake area must be 
very large to have any hope of maintaining a high airflow across the 
primary mirror.

I hope all of that helps.

Clear Skies,

Bill Hanagan