[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

[ATM] Removing tube currents, improving images with fans,was Tees instead of triangles



Alan Henderson asked several questions related to fans.  My answers are 
not in the same order he asked them.  Sorry if the message seems a bit 
disjointed.  The subjects come from Alan's questions, and I haven't 
tried to put them together seamlessly.  Alan's message is copied at the 
bottom.

I think a baffle intended to keep out light from behind the mirror can 
go in several places and still be effective.  The important air flow 
direction is across the mirror.  Your baffle will be parallel to this 
direction, so it can be arranged to not interfere without much trouble. 
  It could be behind the mirror, at the rear edge of the mirror, at the 
side of the mirror, at it's front edge, as you propose, or a bit in 
front, and still do a good job of keeping out light from behind.

Fan size is a function of mirror size.  You want a good stream of air 
across nearly all of the face of the mirror.  Unfortunately front fans 
have to be perpendicular to the mirror, so that a single fan large 
enough to cover the whole mirror also blows a lot of air that never gets 
anywhere near the mirror.  (One atm has mounted his fan on a spider, a 
couple inches above the mirror with it's air flow toward the mirror. 
The fan is in the shadow of the diagonal and he aligned the fan spider 
with the diagonal spider.  It is a neat idea, I don't know how well it 
works for boundary layer removal and mirror cooling.)

The ready availability of 12 volt computer cooling fans in various sizes 
is a great asset.  Probably the 80 and 90 mm sizes are the most 
applicable to medium size and smaller scopes.  Since you need a layer of 
moving air, two or three of these mounted next to each other will be 
more effective than a single larger one.  There are also 120 mm cooling 
fans readily available for the bigger scopes.  Experiment with a 
variable resistor (you will need one rated for a couple of watts and 
perhaps 0 - 100 ohms) to control fan speed.  High speed for 30 minutes 
or so, followed by a lower speed for the rest of the night has been 
reported by more than one observer to be an effective plan.  Once you 
have the needed resistance worked out, it may be more convenient to do 
the speed regulation with a switch and fixed value resistors.

In most cases, you want the front fan just ahead of the mirror, so that 
the stream of air from the fan will blow across the face, but will 
alsoactually impinge on the face.  The greatest optical damage comes 
from a warmed layer of air only a few millimeters thick hugging the 
front face of the mirror.  The natural viscosity of air tends to make 
this layer hug the mirror as it gently flows up due to bouyancy.  This 
warm boundary layer is now known to cause the great majority of what 
have long been known as "tube currents".  Blowing the boundary layer 
away with a fan achieves two valuable purposes simultaneously.

1.  Breaks up and moves away the warm air layer.  This greatly reduces 
the optical effect of the warm air, even when the mirror is still warm. 
  This is not a theoretical statement.  Brian Greer and Alan Adler both 
have done practical experiments showing that it works.

Many ATM's have been confused by the idea that turbulence is the cause 
of bad seeing through air.  This is not true.  Turbulence has a much 
smaller effect than temperature differences.  Adding fans adds 
turbulence, but, because thermal effects are much worse, the net effect 
is a great improvement.

2.  The boundary layer also acts as thermal insulation.  Breaking it up 
greatly increases the cooling rate of the mirror.  This is also not a 
theoretical statement.  Brian and Alan have shown it with direct 
measurements.  It is also very much in line with conventional 
engineering knowledge.

With your 3 inch thick mirror, it is very likely that both front and 
rear fans will be useful.  The mirror can be cooled through both faces 
and the boundary layer at the rear face is just as much a hinderance to 
heat flow as the one at the front.  Breaking up the rear boundary layer 
won't immediately help image quality, but will contribute significantly 
to the mirror getting near air temperature quickly.  When the mirror 
gets close to air temperature, much of the bad optical effects of the 
front boundary layer go away.

With a thick mirror, cooling at both front and back will also help get 
the glass to a more uniform internal thermal state.  Although the effect 
of internal mirror temperature inhomogeneity is probably not as big a 
contributor to bad images as was once commonly thought, it is a real 
effect none the less.  Reducing it can only help.

As to whether a fan can be helpful at the diagonal, the most important 
variable is the thickness.  A 3 inch diagonal is probably only 1/2 to 
3/4 inch thick.  This may cool fast enough to not be much of a practical 
problem.  I am not aware of experiments using a fan for the diagonal. 
Alan Adler's cool.exe program can model this situation for you to give 
some guidance.  There may well be a benefit to blowing the boundary 
layer away on larger diagonals.  I doublt that the air flow from a fan 
near the primary will still be foreceful enough to disrupt the boundary 
layer at the diagonal.  The required air velocity isn't terribly high, 
but air slows down pretty rapidly.

If your tube is not quite open, you should probably consider a vent 
opposite the primary front fan.  This vent will let much of the warm air 
from the mirror blow out of the light path so that it can do no further 
harm to your images.

Do try to get and read the articles I cited.  They are very well written 
and illustrated and will give you a better understanding than what I 
have written.  Here are the citations again.

If you read Alber Highe's article in the most recent S&T, you will see 
that he has put fans in the mirror box of his ultralight 12.5 inch 
scope.  The position of the fans indicates that they blow across the 
front of the mirror.  (The fans are not mentioned in the text, but are 
clearly visible in the pictures.)

  Alder, Alan, Thermal Management in Newtonian Reflectors, Sky &
  Telescope, Vol. 103, No. 1, January 2002, pages 132 - 136

  Greer, Bryan, Understanding Thermal Behavior in Newtonian Reflectors,
  Sky & Telescope, Vol. 100, No. 3, September 2000, Pages 125 - 133

  Greer, Bryan, Improving the Thermal Properties of Newtonian Reflectors 
  - Part 1, Sky & Telescope, Vol. 107, No. 5, May 2004, Pages 128 - 132

  Greer, Bryan, Improving the Thermal Properties of Newtonian Reflectors 
- Part 2, Sky & Telescope, Vol. 107, No. 6, June 2004, Pages ? - ?

  Video examples from Bryan Greer's article are available at
  http://skyandtelescope.com/howto/scopes/article_1182_1.asp
  or at
  http://www.fpi-protostar.com/bgreer/may2004s%26t.htm This second link
  has more content than the one at S & T.

  Alan, in particular gives a very graphic account of the improvement 
from a front blowing fan, even before the mirror has enough time to cool 
off.   While the mirror is still well away from thermal equilibrium, a 
front blowing fan gives significantly better images.


Alan K. Henderson wrote:
> I've seen a few sites that say the front fan should blow across the top.
> Where should the mirror be mounted, and how big a fan?
> 
> I've considered installing a baffle, similar to the one for this scope, that
> blocks ambient light from behind the mirror. (See last two images.)
> 
> http://home.planet.nl/~asuyker/nvws/frames/my_ultralight_dobsonian.htm
> 
> Due to the mirror's quarter-inch beveled edge, I can attach the baffle
> directly to the mirror without obstructing the reflecting surface. Will this
> have any detrimental effect on the air flow from the two fans?
> 
> Should I consider a small fan to blow across the secondary? (The primary is
> 16" f/4; I calculate a 3.1" secondary.) Or will the breeze from the powerful
> rear fan be strong enough to go up the length of the scope and disrupt the
> secondary thermal layer?
> 
> Alan


-- 
Mark Holm
mdholm@telerama.com

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