A Tale of Four Flats

fixaflatIf only flat making was as easy as using the item pictured to the right here....

If you've been through the telescope section of my web page, you've probably read about our 30" F/3.8 scope.  The primary is in good shape, but after using the scope a few times we discovered that the secondary mirror (flat #1) was introducing astigmatism.  Mike Conron made the secondary - he did a wonderful job - he cored it from a 10" blank, ground and polished and figured it according to a borrowed reference flat (flat #2).  The astigmatism was evidence that the "reference flat" did not deserve to be called that (it wasn't flat).

Recently, I bought a surplus 10" reference flat (flat #3).  We tested it against several other surfaces, and confirmed that it was indeed quite flat on BOTH sides, confirming the documentation that came with it.  I decided to refigure one 10" pyrex flat (flat #4) against my reference flat.  Then I could use it for pressing to press a lap to work on the 5.5" minor-axis diagonal mirror and fix it.

10" Pyrex Flat Refigure

First, we checked the figure of the other "reference flat", which was 7" in diameter and made of fused quartz.  A photograph of it under test, along with a smaller 2" flat, sitting on my flat, is shown below on the left.  (The 2" flat had just been cleaned and handled, which is why its fringes were curved, due to thermal distortion.)

Additionally, a test of the 10" Pyrex mirror that was used in the figuring of the diagonal revealed about 12 fringes of concavity (!), pictured below at right (smaller photo).  This 10" mirror mated with the 6-7 fringe-convex 7" "reference flat", producing fairly straight fringes.  Unfortunately neither is flat.

Other flats 10in test

These pieces were so far from flat that careful work with a modest spherometer could easily show the deviation from a flat surface.  Here are the spherometer readings for my 10" reference flat, the 10" Pyrex mirror (12 fringes concave) and another 10" surface that I intended to use to grind against the Pyrex mirror to flatten it.  Note the needle centered on the 10 index in the first photo, and reading 9.8 and 9.7 respectively in the next two photos.  This indicates concavity.  (While the accuracy of the indicator is admittedly not perfect, the relative readings were more consistent than I expected.)

Master 10 in pyrex 3rd surface

Due to the departure from flatness, I decided to go back to 12 micron and grind the 10" Pyrex mirror flatter against another 10" Pyrex blank. The third surface, intended for grinding against the bad 10" Pyrex flat, was slightly more concave than the bad flat, as shown in the spherometer readings above.  So, grinding the two together should make them both reasonably flat, and this of course could be adjusted during grinding to make them both as flat as possible.

The picture below shows the polishing still remaining in the center of the 10" Pyrex mirror surface after 20 minutes of grinding with 12 micron..

Partial grind

An hour gave a good uniform surface grind on both blanks, and then about 45 minutes at 5 micron produced reasonably flat, ready to polish surfaces.  The positions of the pieces were adjusted to further refine the flatness.

Below several pictures of the fringes at various stages in the refiguring are shown.  Polishing produced a somewhat concave 10" mirror, and that was removed and the figure refined.  At the end, a fairly good 10" flat was obtained, not perfect at the edge.  Tool on top polishing was used to remove the concavity, and the stubborn raised edge was worked down with a small local polisher and the full-sized lap.  Lots of pressing is important in such work.

Fringe 2 fringe 3 fringe 4

fringe 5         fringe 6

All in all, a successful refiguring.  And now we have a good surface to press a lap against for the refiguring of the 30" diagonal mirror.  (I prefer not to press laps against nice reference flats that I have purchased!)

Refiguring a 5.5" minor-axis secondary mirror

Careful work is required so that the edge of such an elliptical secondary mirror is not turned.  Pressing the lap on a known good flat will help minimize this, and avoiding touching the edges (especially on smaller flats) is advisable.

Here's a picture of the diagonal mirror before refiguring, under test on my 10" reference flat.  The mirror is approximately 4 fringes concave, with many of the fringes bunched up at the edge, indicating a severe slope error in the outer areas of the mirror.  The picture on the right is of the mirror sitting on top of the 10" lap that was used to refigure it.  (This lap was first used to make the 10" Pyrex flat described above.)

Before\Diagonal on lap

Refiguring was accomplished by polishing the mirror on the lap with strokes of varying length and width.  Care was taken not to heat up the mirror with heat from my hands, and pressure was kept light.  Longer strokes lead to a more concave mirror, shorter to a more convex mirror, but things can act a little differently on the major and minor axes given the strange dimensions of the mirror.

The concavity pictured above is astigmatic - the fringes follow the elliptical shape of the mirror.  This shape gave way to overall convexity, but of the non-astigmatic type, with nice round fringes.  From here, the mirror was gradually worked back to somewhat concave, and then flat over the course of about 10 figuring sessions.

Here are pictures of the mirror and test setup in the midst of refiguring, and of a test where very few fringes are visible, and the test is quite sensitive.  Both pictures show concavity in the center of the mirror.  The test setup is my interference tester with a new light source - two 8-watt black lights (F8T5-BL - black lights without the normal blue filters on the glass) shining through a green filter (Lee Filters - I'll check the number) to isolate the 546 nm emission line.  I like this setup - the lights don't get hot, and I can leave it on for long periods of time without cooking the ballasts or using much power.

Setup Sensitive test

Testing this mirror revealed interesting cooling characteristics.  For the scope (30" F/3.8) the cooling of a diagonal is nothing compared to the 2.125" thick primary, but it's still interesting to watch.  Below is a collection of six photos taken at various times after the mirror was worked, rinsed and dried.  It appears the figure stabilizes fairly well after 30 minutes.  The figure goes from about two fringes of curvature (1 wave on the glass) to less than 1/4 fringe (1/8th wave on the glass) after 35 minutes.  There is a slight depression in the center of the mirror.

Here are the final fringe photos of the finished mirror.  The surface is quite smooth.  There is a very slight depression in the center, but I'm not going to risk the rest of the figure to remove it.  It's a whopping 1/10th wave deep.  Can't wait to look at Jupiter and Saturn with our "new" scope!

Finished 1 Finished 2

After making the two flats this February and washing/drying my hands countless times, I will be using these:


Now it's time to finish the folding flat for my travel scope....

Back to home page....