Re: Re:ATM astatic supports

["Frédéric Géa" <geaf@unionfinancieredefrance.fr>]

This cell and with its support system has been developed over many years and
are the result of work from many people.  As I’ve mentioned on the web page,
I haven't created anything of my own but the great performance of this cell
has convinced me to try it on my own scope, a 22-inch Dobsonian on an
equatorial platform. Astatic means always in balance. This cell is
interesting for use with large thin mirrors. Not long ago, the design used
levers that were necessary for each point, needless to this "forest" of
levers under the cell made it really hard to “tune” the cell for optimal
performance. This hybrid astatic cell is a combination of triangles like
that, which is used in a conventional flotation cell. The astatic cell has
existed for quite a long time but that "forest of levers" was really
complicated. With this hybrid, it's no longer true.



The usual design for a mirror cell used in a large DOB is the classical
flotation system. This cell is a system of balanced "layers" of triangles
and a joining bar. When the ‘scope moves up and down, these triangles and
joining bars also need to move to maintain their balance. When these
"layers" move, there is friction between them. Everywhere there is a contact
point with movement; it's a potential friction. This friction is never the
same and it can give a random position of the points and cause a distortion
to the optic surface. These different positions give an "always changing"
support of the mirror. This changing force produces tension.  One can see
how a large mirror blank can act like rubber and if the glass is thin
enough, it is noted that the cell can easily deform the glass.  The weight
of a large mirror puts enough force to cause enough friction in the cell
flotation system. If the number of supports increases, this can give even
more problems. In order to properly support larger and thinner mirrors, some
people have made a 54-point mirror cell or even more. That means even more
chances for the friction to cause problems. When the OTA is moved down, the
triangles are "unloaded" and the sling carries the weight of the mirror.
When the ‘scope is move up, the weight goes back to this "random" cell
without any control. With astatic levers, it's different. There is only one
layer of triangles, like for an ordinary 9-point cell. With fewer layers in
the system it means less effect from friction. Also, triangles are of course
not free of friction but they are very much reduced. The levers axis of
rotation is achieved using a female cone and a matching conical point, so
the possible frictional surface area is very small. Using a “chrysocal”
blade (called flexural pivots), it's even less cause for friction, it's an
"elastic" lever made with a metal that always comes back to the same
position. For the transmission of the push of the counterweight, it is the
same thing. Some astatic cells use a transmission part (a "piston")  that
can give some friction but it's possible to use a chrysocal blade or other
metal to transmit the effort with very minimal problems, so the push that
was given to levers sometime to release tension is very reduced, a simple
movement at the begining of the night is enough.  Astatic levers only
transmit a pressure, a strength, not a movement. Parts don't move. It's not
a question of moving parts but a question of effort applied with little
friction in order not to hinder motion. The triangles always apply the same
force to the glass. When the ‘scope is moved down, the counterweight
regularly gives less and less push and lets the lateral support carry the
weight. When the scope is point up, the counterweight has more and more
"power" to carry the load, always the same way because the position is still
the same. This difference, compared to a classical cell is that it is
totally reproducible and always gives the same thing. On the cell, there are
3 "fixed" points (or 3 triangle on sphere to allow movement for a 18 points
cell) that are evenly located around the triangle. They maintain the mirror
on is position in "space" and can't move vertically or horizontally. The
other triangles with levers simply apply a force that is proportional to the
amount of glass they have to carry. Their is no "piston" movement that
raises the mirror from it’s position on the 3 fixed points.



Lateral support made with piano wire works in compression. It's also a
transmission of an effort. When the blank is carried more and more by the
lateral support, they are always in close contact with the Teflon tipped
pads on top of that support. The piano wires slowly carry the load and
received the effort. As it is a highly elastic material used longitudinally,
it resist to the pressure laterally while still allowing the mirror to
“float” on the levers. With 3 of these supports around the circumference,
and the astatic levers of the cell, there is no longer a lateral movement
problem of the mirror if used on an equatorial platform for example. No foam
or anything else. The cell keeps the glass accurately in only one position.
There is no movement; it's simply a question of transmission of weight
without friction.



This cell and lateral supports are not very hard to make. I have made mine
with the same tools need to make a standard flotation cell. If a lathe is
available it's of course easier to make some parts but it’s possible to make
them with a saw and a drill press. The price of the different elements is
about the same than for a classical cell too. Counterweights can be made
using a heavy metal that doesn't cost a fortune (lead comes to mind). The
added weight of the counterweights can slightly increase the weight of the
total scope, for about 1/5 of the weight of the glass. Due to some poor OTA
layout, it is often necessary to add some counterweight at the bottom of the
OTA. I prefer see it as something useful than simply dead weight. It's
probably possible to reduce the weight of counterweights but this ratio
works quite well. Experimenting is of course possible.



I have used this cell and have really been pleased with it. I must confess
that the first time I saw one; it was really a strange feeling, what a
complicated thing. It looks complicated but really, it's not. Thanks to
software like PLOP, calculating the points is no longer hard to do. Of
course, a classical cell could have some enhancements, like a special effort
to reduce the friction of the flotation triangles, the astatic cell doesn't
require complicated tools to make and I prefer to choose this much forgiving
system.  The Lateral Supports are also quite strange but now really love
them. They enable an accurate positioning of the mirror in the OTA. Like the
saying, “Holding a mirror is like holding your girlfriend, hold her close
and secure but NO PINCHING”! When pointed at a very low position they can
give astigmatism, but as it is not the way large scopes are used, that's not
a big problem. The collimation is much better kept over the course of few
nights and the glass is in perfect shape, no matter where the ‘scope is
moved. All that may look strange but I will not make a large Dob using a
conventional cell again because it's easier not to bother with mirror
distortions caused by binding flotation triangles or improperly positioned
slings when those rare moments of good seeing occur. It's of course some
more works but use a large thin mirror as it full potential worth the
effort.




Frédéric Géa
http://www.astrosurf.com/altaz
-----Message d'origine-----
De : Michael Koch <astro.electronic@t-online.de>
À : martti.koskimo@kolumbus.fi <martti.koskimo@kolumbus.fi>
Cc : atm@shore.net <atm@shore.net>
Date : mardi 27 mars 2001 06:44
Objet : Re:ATM astatic supports



Hi Martti,

> extra care in making the joints sensitive and all contact points to the
> mirror from Teflon

That's exactly what I wanted to do: Teflon / glass contact points.
In my case, the weight per contact point is about 5 kg.
What do you think is better:
Should the back of the mirror be rough ground, fine ground or polished, or
is
there no difference?

Another question:
My mirror will have a central hole. For the lateral support, it would be
possible to share the load to the edge and to the hole.
For example, one could support 70% of the lateral load by a sling (or
multiple
points) and 30% by a bolt in the central hole.
Has anybody tested this?

Michael

*******************************************
ASTRO ELECTRONIC  Dipl.-Ing. Michael Koch
      http://www.astro-electronic.de
*******************************************