Home Page

Valley Skies

Newsletter Home Page

The Telescope Nut
by Jeff Baldwin

Shaping the Mirror for Grinding Preparations

Before grinding a large, fast, thin mirror, it must be prepped to prevent astigmatism, print-through, and other nasty errors. The back must be ground flat, the front must have the edge fairly well coplanar with the back, and it must be somewhat curve-generated. For very large mirrors, this is serious work.

Grinding the Back Flat

If the back of the mirror is not flat, it might rock. If it does rock, it might rock better one way than the other. If this is true, even a little, the mirror will resist your work one direction more than another, and this will impress a different radius of curvature on one axis from the other axis. We call this astigmatism, and it is the most easily seen image-ruining aberration in large, fast, thin telescope mirrors today. (If you don't believe me, go look through a 13.1" Coulter Odyssey at high power.) The back flattening routine is the best way to prevent this problem.

For the 40" mirror I used 3/4" green lime based plate glass with pulverized granite/silicon carbide (sandblasting sand). Simply using a straight-edge to check for flatness is enough precision. It needs to be ground smooth and flat over the entire surface of the back. I like to grind all the way up through 500 grit. You don't need to check for pits between grits. Since it is not an optical surface, just go until it looks good. Once it is ground flat, sign and date the mirror's back for pride, future reference, and kicks. I use an electric pencil for this.

Jeff grinding the back of the 40" mirror with 3/4" plate glass donated by Mat Hutchings. 120# of iron is Velcroed to the plate for vertical force.

Leveling the Edge of the Mirror in Reference to the Back

If a large, thin, fast mirror has the front surface tilted with respect to the back surface, funny things tend to happen. With thick and small mirrors, it's no big deal; the front is unaware of the back. With the large and thin mirrors, the difference in thickness can be especially frustrating when the telescope changes temperature while observing--which is all the time. Thick glass will change shape and size at a different rate from thin glass. The middle of a paraboloidal mirror is thinner than the outer parts, so when the mirror cools and shrinks, the middle beats the outer parts because it cools more efficiently. This causes over-correction for the time it takes the outer part to catch up to the inner part. However, when the mirror is thicker on one side than the other, because the front edge is not coplanar with the back, then the mirror will form an astigmatism at any time it is changing size and shape, which is a non-stop thing during the night. This means that the mirror will be permanently astigmatic even if the front is a perfect paraboloid during testing. Also, being thicker on one side than on the other means it is heavier on one side, producing a larger gravitational moment which pulls the mirror to asymmetry. This stuff is nonsense with typical mirrors, but serious issues with the big, fast, thin mirrors.

One of the problems encountered with this 40" blank was that it was very irregular in thickness. On one axis it was 1/4" or more thicker than on the opposing axis. Grinding the edge coplanar with the back was a lot of work using a high speed grinder.

Here is a graph showing the thickness of the blank in thousandths of an inch in 16 intervals. The high graph is the original condition, the middle graph is after 5 hours of repair, and the low graph is after 7 hours of repair.

Curve Generating and Glass Removal

The 40" mirror needed 436 cubic inches of glass hogged out of it to make the paraboloidal surface from a flat surface. We sandblasted it to get most of it out, but that left an irregular curve which had to be shaped up. Using conventional mirror-tool grinding techniques would have taken a few million years to achieve. So, out came Jeff Draper's 7" High Speed Makita Grinder, half a dozen masonry wheels, dozens of hours, and Jeff Draper's laser cut mirror curve template. The mirror was then ground into a fairly nice sphere, close enough so that a cement tool could be cast against it. I placed the template against the mirror, and wherever the two were in contact were the high spots. That's where I ground away. Eventually the entire mirror mated with the template and the cement tool could then be cast.

Since the back was not flat to begin with, and since I flattened it with the front not perfect and wobbling while I worked the back, it is possible that the back has a print-through of the irregular front. That means that the back really isn't flat. I remedy this by grinding the back flat, then making the front spherical and the edge coplanar to the back, then once again grinding the back flat and finishing it off to 500 grit. Now the front is worked.

This seems like a lot of extra stuff to do, but I plan on using this telescope until my eyes no longer work, so I want it to be right, and astigmatism is not part of my plan.

Casting a tool

Most ATM-ers cast plaster tools. A typical plaster tool for this size would be 5000 cubic inches, and this would make the plaster tool very expensive, and possibly weak. I decided to cast my tool out of cement. It is 2.5" thick, 36" in diameter, and weighs about 240 #. I bought a 36" Sonotube one foot long, and when set onto the spherical blank makes a great cast mold for the tool. We only used 2.5" of the Sonotube. Visquine on the mirror kept the cement from bonding to the mirror (bad). Metal reinforcement keeps the tool from breaking, bending and cracking; handles and bolts within it will allow us to pick it up, move it and hang it.

Eric Reichenbach made the tool, and it is a beautiful piece of work. When the tool was cured, ceramic tiles were mortared to its curved surface to create the grinding surface.

The mirror was checked for areas of contact after an hour (or so) of grinding. It would be spherical wherever the mirror had been ground smooth by the tiles, and the smooth areas would also be the highest protrusions above the non-spherical low spots. I take the high speed grinder and grind on these smooth areas to bring them down, since masonry wheels are cheaper than grit. Then I grind with the tool again. This will be repeated until the entire mirror is mated with the tiled tool, and spherical contact is declared.

Friends

I've had so many people help me so far, and the telescope project isn't even in full swing yet. Thanking them is important to me, and missing somebody would really be bad.

So far there has been at least a dozen people in on this: Glenda, Glen, Roger and Leo, Mat, Eric, Lloyd, Trevor, Doug, and many others. When I say "I'm building a 40" scope", that's code for "My friends are building a 40" scope with me". Most important of them are Glen and Glenda.

Glenda puts up with a lot of astrocrap from me. Can you imagine being married to me with my insane projects and ideas? Whoa.

Also, Glen Youman makes stuff for me. I told Glen that we'll try to put this in one or both of the astronomy magazines when we're done, he as the builder and I as the optician.

My current time-line is as follows:

We'll see.

Clear Skies...Jeff Baldwin
For more information on Telescope Making jump to the ATM page.

Copyright © 2001 by Jeff Baldwin
Last Updated: 3/5/2001
http://www.stocktonastro.org/Newsletter/TTN_Shaping.html