ATM Date: Sun, 29 Oct 2000 12:44:45 -0800

["John Sherman" <shermj@netzero.net>]

Hello everybody!

This is my first post to this list.
I would like to describe a telescope I once built, my reasoning and my
failure.
I have for many years been aware that an off-axis reflector
(unobstructed aperture) can deliver the
very best images a telescope can offer. The problem is that most designs are
very hard to make and suffer a very long f/#. The Hershelian design seems to
give the best chance of a reasonable f/# (e.g. f/5 or f/7). The only way to
achieve this is by using a section of parabola. This is the same
configuraton used by many Newtonian owners when they put an off-axis mask on
the front of their entrance aperture. The off-axis mask is using only a
section of the full parabola. So why not just make that off-axis telescope
to start with? Sounds easy, but all of you can tell me it is VERY hard to
make the section of parabola on a piece of glass.
So it was with great interest in 1992 when I first read in S&T of Bill
Kelley's flexed mirror telescope. Could this technique be adjusted to make a
section of parabola? If the bolt was glued away from the center of the
mirror, would it approximate a section of parabola? I decided on an
experiment. I purchased an 8" X 3/4" piece of Pyrex and polished out a
sphere at f/5 (40" focal length). I glued the bolt 1 & 1/4" from the center
of the mirror. The line running thru the center of the mirror and also thru
the center of the bolt I will call the "axis of deflection", with Point A
being the center of the mirror and Point B being where the bolt is. I built
a wooden box for the "tube". I made a cell and mounted a focuser on my tube.
The cell holds Point B towards the focuser side of the tube. The mirror is
supported by a ring of carpet around it's outer 1".  This telescope is
incredibly easy to collimate--just aim the focuser at the glass, and aim the
glass at the sky. (I used a Tectron sight tube to aim the focuser at the
center dot on the mirror. I took a piece of string and tied a knot in the
middle and stretched it across the entrance aperture with the knot in the
center. I then aimed the glass so that the sight tube was now centered on
the dot and the knot.) Aimed at Polaris, a Ronchi grating (100 lpi)  in the
focuser showed the obvious tilt of the mirror. With the Ronchi parallel to
the "axis of deflection", the lines were something like this:   \|/
(ascii art). The challenge is to tighten the bolt to the point where the
lines become parallel. So I began to tighten my bolt, and sure enough the
lines did become parallel!!  ||||   (more art). How excited I was! I eagerly
threw in an eyepiece to get my first view, an was suddenly very
disappointed. No good focus. So I reaimed the mirror at my knot (this
alignment had changed slightly when I tightened the bolt), made a tiny
adjustment to make the Ronchi lines parallel again, and put the eyepiece in
again. Still no good. Testing revealed the problem to me. When the Ronchi is
parallel to the "axis of deflection", it showed four lines. When the Ronchi
is perpendicular to the "axis of deflection", it showed six lines. Either
way I pointed it, the lines remained straight. Only the number of lines
changed, showing me the astigmatism I created. The focal length was
different (shorter) perpendicular to the "axis of deflection" than it was
parallel to
it. However, the lines were straight and parallel in both directions. This
showed great promise. So what if I used two bolts? I could draw a line
perpendicular through the "axis of deflection" at Point B. I could guess
that if I put two bolts on this line, each about 3/4" from the "axis of
deflection" this might correct the astigmatism.


   "axis of deflection"  --->     (----B-A------)   first setup


                              C
                               |
               (------------B---A----------------)     new setup
                               |
                              C


I would call the new locations Point C. Each Point C is about 1 & 3/8" from
Point A. While drawing the new lines on the back of the mirror, I found
another problem. I had bought this glass from Newport Glass Works. I had to
pay extra to have the f/5 curve pregenerated, and also I paid more extra to
have
the blank diamond ground down to the 3/4" thickness (from the standard 1").
This grinding left a swirl pattern on the back of the glass. The center of
this pattern is not at the center of the glass!! It is off-center about
1/4". I wonder, does this contribute to my astigmatism problem? Was the
pregenerated f/5 curve also off-center? Since this is the only piece of
glass I have ever ground and polished, I do not have the expertise to learn
the answers to these questions. Bummer.
End of experiment. After all the investment of time and energy already put
in, I was not willing to go on with out being able to answer those
questions.
Years go by...
But now, in the Nov/00 S&T, here is a progression of the original BKD. Alan
Adler has made great progress. Also there are now computer programs which
might be adapted to help (such as PLOP and FLEX) by people who know how to
do such things. I can see that Adler's distribution of pull to the optimum
area improves the original BKD. Can an off-axis configuration be designed to
optimize for a Hershelian? I can estimate that the '3/8" puller sponge'
(illustration pg134 11/00 S&T) would take on a "happy face" configuration,
and if successful, would put a happy face on whoever builds it.
So, is anyone interested enough to make progress beyond mine???

John Sherman
john@johnspics.com
www.johnspics.com



.



____________NetZero Free Internet Access and Email_________
Download Now     http://www.netzero.net/download/index.html
Request a CDROM  1-800-333-3633
___________________________________________________________