Mirror log

This is a log of the mirror grinding I'm doing with my first 6" blank.

Date Entry
May 2, 2004 Mirror appears to be either 100" or 105" radius of curvature using wet-mirror-and-flashlight test. The 105" reading came after 20 minutes more grinding on the blank than the 100", so I suspect one of the tests was done wrong.
May 3, 2004 Mirror still appears to be about 105" radius of curvature using wet-mirror-and-flashlight test. This confirms that at least I'm testing consistently now.
May 5, 2004 Using the new paper-screen flashlight test, the mirror appears to have a radius of curvature of 109". The mirror is "cold" though. Hasn't been worked at all prior to this test.
May 16, 2004 Ground for around 1 hour with 120 grit. Mirror appears to have 105" radius of curvature with the flashlight test. Obviously my consistency with the flashlight test is still marginal.
June 6, 2004 Ground for about 1.5 hours with 120 grit. Mirror tests to 96", exactly where I wanted it to be. The next trick is to retest it tomorrow and see if it's still at 96" radius of curvature (since I still don't know whether being fresh from grinding makes a significant difference in this kind of rough testing).
June 29, 2004 Ground for about an hour on 220 grit, and found that I no longer hand any stationary pits. Moved to 30 micron and ground with that for about an hour. Flashlight test returned 93-94" radius of curvature before the last ~15 minute session, but no further testing tonight because it's too late. Pick up next time by testing radius and continuing with 30 micron.
June 30, 2004 Spent a good solid 3 hours with the 30 micron, wearing down some big pits left over from the 220. Finally, I've gotten past them. Next, test focal length again, and proceed to the 15 micron.
July 12, 2004 One and a half hours with 15 microns seems to have finished off the larger pits I could identify. The entire time with 15 micron was spent with the mirror on the bottom to correct an apparent 93" radius of curvature. Test before the next session to find the radius of curvature, and carefully inspect for any pits I may have missed.
July 24, 2004 One hour with 9 micron has resulted in no new big pits, and a 94" radius of curvature. It measured 93" at the start of this grinding session, so at least it's making some change. Continuing with 3 micron.
July 24, 2004 And 45 minutes with 3 micron got rid of all the pits. The mirror surface is indeed just cloudy at this point, and is easily seen through. Polishing is next, for which I'll need to contact the SAS ATM folks.
August 6, 2004 Spent an hour and 45 minutes polishing, and the difference is amazing. Even after 20 minutes, the mirror went from cloudy to visually perfectly clear. Testing with a laser shows an image still quite visible, so I still have a ways to go. However, it's impressive how quickly I was able to polish out the 3 micron grain. The flashlight test reveals the radius of curvature to be almost exactly 93.5", with a perfect reproduction of the flashlight.
August 14, 2004 About an hour polishing has resulted in no visible improvement in the mirror surface, but also a few quite substantial scratches. I believe a few more sessions of polishing will eliminate them, but the problem is I don't know how they got there. I can feel larger pieces in the polishing compound, but I also feel like I'm getting them out before I start polishing.
August 18, 2004 Spent a bit less than an hour polishing -- the lap has started closing up, and I wanted to stop before it has totally closed up. I'll have to figure out how to fix that next time I tackle it.
August 27, 2004 Spenet just over an hour polishing tonight. Surface is 95% clear, which is exciting -- I may be able to do final figuring and drop off the mirror to be coated tomorrow night. Of course, I haven't yet looked at the surface shape, so I may have significant errors to correct yet. But I'm getting close.
August 28, 2004 At the ATM meeting, I spent about an hour carefully polishing with an extra-long stroke to try and work down a hill in the middle of the mirror. Made progress, but there's still a hill.
September 11, 2004 Polished for about 3.5 hours after building a proper cleat system to hold the tool. Testing with the Foucault tester shows that I'm making progress, although I now have a ring more than a hill in the middle of the mirror. No more turned-down edges, though, so that's good at least. I suspect just more straight-up grinding will eventually fix it -- the hill is basically flat now.
October 13, 2004 Finally pulled the mirror out and did some more polishing. I think after 1.5 hours, I've finally got it into about the right shape. I need to pull it out and look at it again when I'm fresh, and compare my Foucault results to some diagrams of what it's supposed to look like. If there are any more faults in the figure of the mirror, they're probably focused around not having a deep enough curve yet. Very encouraging.
February 26, 2005 I took the mirror to an ATM meeting, and did about 15 minutes of polishing, once I finally got the lap warmed up enough. Did long, slow, higher pressure strokes, which helped bring down the hill in the middle. It also, surprisingly, took care of the last of the surface imperfection when viewed with a laser. I figured that would take an hour or two more, not 15 minutes.


Created by Ian Johnston. Questions? Please mail me at reaper at obairlann dot net.