You've actually seen the glass flow? Really? You were there when it was
installed, and verified that it was of uniform thickness?
This is, I repeat, an old wives tale. All of the supposed observations of
glass flow has been reliably debunked. This includes the window thicker at
the bottom, both in windows and in old stained glass. The very simple
explanation for this is that craftsman installed the glass that way, since
it was wedged when they got it. Occasionally an old window is found with
the wedge to the top. Window glass at one time was made by blowing a large
sphere, rolling it flat to make a cylinder, then cutting it open (all done
while soft) and pulling it out flat. This doesn't make a very flat product.
Glass is an extremely stable material. If it flowed as much as the wedge in
the window indicates, then it would flow a detectable amount in a matter of
weeks. These flows are not detected. A very simple way of measuring the
wedge of glass is by looking at the Fizeau fringes of glass flat stood on
edge. Every optical shop with interferometers has these things around, and
they are standards of flatness. Nothing changes over very long periods of
time. I've used flats that were parallel, and had been so for at least 20
years.
Interferometric measure of tilt grossest unit is the fringe, which for
visible light is about .25 microns. If in 100 years the glass had an
increase of say 2.5 mm thickness, in one year this would be .025 mm, or 25
microns. Thats about 100 fringe/year, or 2 fringes/week. A flat that
changed this much (or 1% this much) would be impossibly useless- people
would notice it right away.


I know its hard to give up ideas that have been repeated so often, but the
scientific evidence is completely against flowing glass. The idea of glass
fluidity came into being when people had very limited ideas about the
different forms that matter could take, and classified it as liquid when
they were unable to explain it lack of apparent crystaline structure.

On the subject of telescope mirrors;
Telescope mirrors of the thin sort have become popular in the last few
decades for financial and operational reasons. There are many ways of
supporting them, many of which use active controls and servos(not EMC, to my
knowledge). There are some good passive ways of supporting them.
1. bubble wrap
2. wiffle trees- a generalization of the common 2-d dimensional sort, it
consists of tetrahedrons with 3 points in contact with the mirror, and one
point facing down. Three tetrahedrons are joined with a fourth tetrahedron,
and its lowest vertex also points down. By continuing this, the entire
mirror load can be reduced to three points. 3 points->9 points->27
points..... The tetrahedrons are free to rotate where they touch eachother,
so no torques are transmitted to the mirror.

Of course the two techniques above will only work if the glass is uniformly
loaded.

Elliot Burke

>Message: 12
> Date: Wed, 17 Jan 2001 00:14:56 -0000
> From: ballendo@...
>Subject: re:re: flat glass



>I read what you're writing, but my EYES have SEEN the results of
>glass "flow!?! And as a previous posted reply siad, mine is NOT a
>singular experience!

>More to the point of this list and the original idea, I would just
>say that using 1/4 plate glass of such a large area will need
>EXTENSIVE, ADJUSTABLE, support to be reliably flat. Any telescope
>makers wish to respond...?

>Hope this helps.