[ATM] Ball bearing for triangles,spherical aberration caused by thermal gradient in the mirror: the results...(long)

[Raphaël GUINAMARD <rguinamard@infonie.fr>]

I don't know if you remember the thread I posted in september about the
spherical aberration I have on my 24" f/3.3 scope.
Here is the story, the results, and one question:
My mirror is very thin: 40 mm at the edge => 28 mm at the center. I have a
27 point classical Kriege style (calculated with plop)
I have approx (according the star test...) 1/4 wave of spherical aberration.
I suspected 3 things could produce spherical aberration:
1) it is in the glass itself => nothing to do except redo the mirror or use
an astatic mirror cell that could correct some spherical ab.
2) it is caused by a temperature difference between the top of the mirror
(cold because it sees the 3K temp of the sky) and the bottom of the mirror
(warm because it is at the outside temperature if you wait enough so that it
equilibrates the ambiant air). The gradient temp between the 2 faces of the
mirror would then deform it. A french professional mirror 28mm thick
(meniscus), 1.5 m diam, that had this problem (which lead to 10 wave of
spherical ab...) solved the problem by heating the aluminium layer using
electricity joule effect.
3) it is caused by my mirror cell because my 3 big (bottom) triangles didn't
seems equilibrate because of the friction in the acorn nut I use as the
pivot center for the triangle (Kriege design)

I have now some answers of the points 2 and 3:
Point 2: I put the scope outside at the beginning of the night with its
primary dust cover on it so that the top faces dosesn't sees the 3K sky.
I waited 3 hours or so so that I'm quiet sure the mirror is at a uniform
temp - the same as the outside tem-. Them I put off the dust cover and make
the star test immediately.
Nothing had changed with the spherical aberation: it was the same as usual.
So I think that if the gradient temp could have a big effect on huge
professional very thin mirror, it has certainly a non detectable effect on
my 24" thin mirror.

Point 3 : To be sure the friction in the pivot center on the big triangle of
my cell was not the problem, I put ball bearings like in Bruce Sayre' 22"
bino
(http://www.foothill.net/~sayre/images/22-in.%20spherical%20bearing.jpg )
It changed nothing neither, but it showed that the 9 little triangles are
not really coplanar.
They are not coplanar certainly because I put 3 little shanks between the
big triangles to avoid rotation and these shanks doesn't seems to have
exactly the good lenght => they bend a bit and so does the now-flexible
triangle. It could also be that I drill the triangle at the COG given by
plop, forgetting that the outside part (not shown in plop) of the real
triangle moves the COG a bit.
Anyway I have to change little things on my cell to put them coplanar and
say definitively that the mirror cell is not the cause, but i'm nearly sure
now that it's not the cause because I'm near a perfect cell.

According the deformation drawing of spherical aberation I found in
D.Suiter's star testing astronomical telescope, I also tried to deform my
primary by putting washers under the 3 inner (upper) triangles and add
counterweigh under the external part of the 6 external (upper) triangles.
Nothing changed neither about the star test pattern showing spherical
aberation.

The only thing I have to test now is to deform the secondary so that it
gives an "inverse" spherical aberration => I hope the resulting image would
be free of spherical ab.
My 130 mm secondary (20 mm thickness) mirror is glued with silicone on my
aluminium secondary holder (wich is smaller than the secondary, 80mm * 140
mm in lengh).
 I think I could easily deform my mirror by puting some (the less the
better!) small screws through my secondary holder acting like fixed
"actuators" for my secondary mirror. I could then adjust the pressure of the
screw on the sec mirror to have the finest star test -aberation free image-.
Not too much because I don't wan't the mirror to unglue from the silicone of
course...
My question is :
Knowing that my primary spherical aberation is undercorrected (so the
contrary shape of the fig 10-3 p175 of Suiter's book), do you know if I have
to put screws near the center of the sec mirror, on the 2/3 part of the sec
mirror or on the outer part?. If it's on the outer part, I can't because my
secondary holder doesn't cover all the secondary....
I think to give my secondary an inverse spherical ab of my undercorrected
primary, I certainly have to give my secondary an overcorrected form (so
"actuators" screws on the 2/3 part of the ellipse mirror), but I'm not
sure...
What do you think, guys?
Thanks in advance for your help. I hope my english is understandable
Raphaël Guinamard

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