Re: [ATM] Mirror cell ramblings re: thin plate glass 3/4-inch x 24-inch

["Bob May" <rmay@nethere.com>]

The aluminum plate is really a bad idea!
The difference between what PLOP produces for a 23" vs. a 24"
piece of glass is going to be very minimal.  Run the numbers and
see.  Odds are that you won't build the support that accurately!
Better to just build a proper support system than try the heavy
plate.
For one, the plate is heavy.
For two, the plate will will expand/contract a lot more than the
glass will so you'll have all kinds of stress in the interface
from thermal problems if you glue the glass to the aluminum.
Third, the support for the glass won't be even so you'll have
printthrough of the spots that the glass is sitting on (that's
why PLOP is so nice these days) and that will destroy your figure
that you worked so hard to achieve.
For a fourth, that thickness of aluminum really isn't going to be
any stronger than the glass is if even that strong so the flexing
will really get bad in the mirror's surface with only three
supports.
Bob May

rmay at nethere.com
http: slash /nav.to slash bobmay
http: slash /bobmay dot astronomy.net

----- Original Message -----
From: Ted Cohen <tcohen@blakeglobal.com>
To: <atm@atmlist.net>
Sent: Monday, May 26, 2008 5:14 PM
Subject: [ATM] Mirror cell ramblings re: thin plate glass
3/4-inch x 24-inch


Warning - I'm not an expert in this field, although my writing
style
sometimes comes across as though I am - I'm just a newbie and my
ideas are
usually counter to the tried and true techniques.

Also, note that PLOP is not applicable to my situation since the
diameter of
the flat back of my glass (24") is not the same as diameter of
the optical
surface (23").

1) Structural Rigidity - To gain structural rigidity and minimize
deformation of the optical surface due to gravity as well as
forces entering
from the tube assembly a thick, solid , ¾-inch piece of 6061-T6
aluminum is
cut into a circle of slightly larger diameter than the back of
the glass.
The back of the glass is then lapped against the aluminum cell
substrate
until good contact is achieved between the back of the glass and
the front
of the aluminum.
a. Benefit - simple
b. With good contact between mirror and cell substrate, fine
grinding,
polishing and figuring can be performed with mirror resting on
cell
substrate.
c. Drawback - The cell is heavy (40-50lbs) - heavier than the
glass.

2) Application of axial and radial support - one continuous
membrane
approximately 2 - 3mm thick between mirror back and cell
consisting of an
elastomeric resin or polymer adhesive.

a. Benefits - Support forces are distributed evenly. There are no
discrete pressure points. Axial support forces counteracting
gravity are
reactionary forces acting equally and opposite to the force of
gravity. This
is a simple and low-cost design. Lateral support does not have to
be
designed into the mirror cell. Finite element analysis is not
required.
Manufacture of 27-point or 36-point whiffletree is not required.
Polishing,
and figuring can be performed with the mirror in its final
position,
fastened to the cell
b. Drawback - Induced "moments" due to gravity as a result of
radial
support off the center of gravity. This is offset by the
non-localized
application of these torques - they are spread out over the whole
glass (as
opposed to localized torques at a set of support points). Also,
the thinness
of my glass means that the center of gravity is fairly close to
the back of
the mirror. In the horizontal position, adhesive forces should
somewhat
counteract these torques. For now, I'm not yet sure whether the
deformation
caused by this effect will be acceptable.
c. Drawback - deformation due to weight gradient acting on
elastomeric
membrane. This will be mitigated by performing polishing and
figuring of the
mirror while it as attached to the cell. Optionally, the mirror
cell can be
held at an angle while figuring to approximate typical viewing
angles.
d. Drawback - mirror will have to be aluminized while attached to
its
cell backing.

3) Thermal considerations.

a) The elastomeric membrane material will be chosen for low
thermal
conductivity. This will help isolate the mirror from non-uniform
thermal
gradients originating from the cell, as well as from the effects
of thermal
sinks in the cell and tube.
b) Thermal equilibrium will be achieved via air flow over the
optical
surface of the glass. The thinness of the glass is an asset in
this case.
c) Lateral expansion and contraction forces generated in the
mirror
substrate and cell are partially absorbed by the elastomeric
membrane, which
flexes under these forces. Also the CTE of the membrane material
can be
chosen to lie between those of the glass and the cell substrates.
In this
regard, there will have to be a tradeoff when determining the
thickness of
the membrane - a thin membrane minimizes axial deformation, while
a thick
membrane that minimizes stress due to thermal expansion. I think
this will
be the most challenging issue. It may require a composite
membrane comprised
of more than one resin/cement to optimize this tradeoff.
According to
ATM-1, the CTE of plate glass is 8.9 x 10^-6 per degree Celsius.
Approximating to 1 x 10^-5 and multiplying by my radius of 19 cm
gives
approximately a 0.057 mm maximum displacement (at the perimeter
of the
glass) over a variation in temperature of 30-degrees Celsius. If
the
membrane is 2.0 mm thick, then the material will have to
accommodate a
lateral deformation of 0.057 / 2 = 0.285% of its thickness, which
seems
reasonable.

Thoughts Welcome

Ted.

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


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