"Satanic cross" indicates strained glass
A little extra trouble now can save you from big headaches in the future.
|
"Satanic cross" indicates strained glass |
Sources of polarized light: Camera polarizing filters can be used for viewing and photographing the strain patterns in glass, but you must be sure to use the LINEAR polarizing filter, not the circular polarizing filter that the camera store salesman will try to push onto you. Cheap large polaroid filters have been available from Edmund Scientific, but that may no longer be the case. You can make a large polarizer from several sheets of window glass. Light that passes through several layers of glass at about a 45 degree angle will be strongly polarized. Use a second polarizer to minimize the light transmitted through both polarizers, then fine tune the exact tilt angle of the glass for total darkening. Similarly, the light that is reflected from glass at this critical angle ("Brewster's angle") will be polarized. Sometimes you can capture useful polarized light from natural sources like the grazing angle reflection from a car body.
If the inside of your glass is not visible due to an unpolished exterior, use an aquarium tank for this test.
During grinding you should check for surface quality, depth of the curve (or focal length), and the overall shape of the surface. Here are some tests you can do:
During the last grade of abrasive, you should be able to look through the mirror. Although the view will be foggy, you should clearly see an illuminated lamp. The view may be tinted red (because blue light is scattered more than red light). You are not ready to polish until you get this sharp but foggy view through the glass. (This inspection applies only to transparent materials, of course!)
You should check the bevel on the edge of your mirror. Run your thumb over the edge, and if it feels sharp, re-grind the bevel until it does not feel sharp. More than 0.5 mm of bevel is a waste of aperture.
Visual Inspection
Check the surface quality of your mirror during grinding. The ground surface should cover the entire front surface of the mirror blank (when you first start grinding, the mirror edges and the tool center will be untouched for a while). The texture of the grind should be uniform, from center to edge, on both the mirror and the tool. There should be no large pits left over from previous grades of abrasive. There should be no large pits due to excessive pressure during grinding. While grinding the last two grades of abrasive, you should use a magnifying lens or a cheap microscope for inspecting the texture.
Feel
The feel test checks the surface quality of your ground mirror and tool and provides confidence that you can go to the next grit. During grinding stop occasionally to check your progress. Clean and dry the mirror and tool. Stroke the surface of both mirror and tool; the surface should feel the same all over. With very clean hands, scratch the surface of the mirror and tool with your fingernail. (Don't worry, glass is harder than fingernails.) The sound and feel of this should be the same at the center as at the edge of the pieces. If the edge zone is rougher, you have to grind some more with your present grade of abrasive.
Red-Out Test
The Red-Out test of surface quality is based on the fact that even unpolished surfaces can form virtual images (mirror-like reflections) at grazing angles. Starting with #220 grit, you should be able to view the reflection of a lamp in a ground mirror, if you are looking at the mirror edge. If you gradually raise your viewing angle, you will find a critical angle where the reflected image reddens, and then a slightly greater angle where the image vanishes. Near this angle, you can clearly see differences in the average texture of the ground surface as differences in the color and brightness of the reflected image. The goal in grinding is to have a surface that is the same from center to edge. I have found that a surface that is fine-ground with three micron grit and is ready to polish has a red-out angle of about 45 degrees. For more on the red-out test check Bob May's web page.
 Sagitta measurement with a ruler and straight edge |
Direct sagitta measurement can also provide a crude check that the shape of the curve you are grinding is approximately correct. Measure the sagitta at the 50% zone (halfway between center and edge). The measured sagitta should be 3/4 the sag at the center. Then measure the sag at the 71% zone; the measured sag should be 1/2 the sag at the center.
Sunshine Test
This check measures the focal length of your mirror. If you are grinding with a coarse abrasive, be sure that your mirror and tool are in good contact all over, and grind a quick wet with #400 or finer grit. Rinse off the mirror, and with it wet, take it out into the sun. The wet surface will focus the sun's rays. Try to get the angle of incidence as low as possible, and focus the sun on a wall or other object. Then have a friend measure the focal length directly. This test depends on getting a sharp focus, so it becomes more accurate as you get to the finer grits.
Flashlight Test
The flashlight test is similar to the sunshine test. It was developed for telescope makers who live in the Seatle, Washington area, where there is no sunshine. Obtain a powerful flashlight and put a fat rubber band around its lens. Set it in a doorway with the beam perpendicular to the adjacent wall. If you are coarse grinding, prepare the surface as with the sunshine test. Put the wet mirror in the beam and focus a spot of light on the wall adjacent to the flashlight. When you have perfect focus with the finer grits, the image of the rubber band will be visible. Measure the distance from the mirror to the wall; that is your radius of curvature. The focal length is half this radius of curvature value. Check at Bob May's site for another way to do the flashlight test. [note: in telescope literature, R means the radius of curvature, the BIG R, and r means the distance from the center of the mirror to the edge, the little r.]
Spherometer
You can use a spherometer to measure your focal length and to provide a rough check on the overall shape of your surface. I prefer the in-line variant of the spherometer because it can check for astigmatism; three-foot and ring spherometers cannot. If your spherometer is smaller than the mirror, check for constant readings at all possible positions and orientations; any differences indicate that the surface is not spherical.
A spherometer need not be complex or expensive, nor does it require a lot of fancy machining. To learn more, click Spherometer Ideas.
 Test pattern applied to large mirror for Sharpié test |
 Blind Test: convex surface |
 Blind Test: concave surface with seriously turned down edge |
 Blind Test using the edge of the display case for William Herschel's 48" mirror. It looks pretty good after 200 years. The curve is inverted because we are outside the focal point. |
Sharpié Test
This checks that the tool and mirror are in uniform contact, and therefore verifies that the shape of surfaces is perfectly spherical. The test is done by drawing a grid on the mirror blank and the tool and grinding a wet. Make the grid with a "Sharpie" marker pen on a clean, dry mirror, and use a clean Sharpie for each grade of grit (to avoid scratches). If there is good contact between mirror and tool, the marks will be worn away equally over all zones of the mirror and tool. For more about the history of this test, see History of the Sharpié test. We are unable to verify if Dr. Auguste DuMont Sharpié is related to the creator of the famous blue screen oscilloscope.
The Blind Test
The blind test, like the red-out test, uses mirror-like reflection from an unpolished surface at a grazing angle. The grazing angle greatly magnifies distortions in the mirror surface. To do the test, use vertical blinds. The horizontal bands you see reflected in the mirror represent a greatly magnified cross section of the mirror. You are looking for a smooth curve here. If the mirror is concave, your eye must be close to the mirror to see the reflected vertical blind. Because of the grazing angle, the focal length is greatly forshortened. The 48" mirror shown here was made by William Herschel about 200 years ago. I was able to use the edge of the museum display case to do a crude blind test on this relic.
Visual Inspection During Polishing
Visual inspection of the surface quality during polishing is based on the fact that an unpolished surface will scatter light in all directions, while a completely polished surface will not. There are two ways to do it:
Note for beginners: after you pass these tests, your polishing is HALF DONE. That is because imperfect polishing that you don't see becomes much more prominent when the mirror is coated with aluminum.
Friction Feel During Polishing
If you are polishing by hand, you should feel a lot of drag as you push your glass over the pitch lap. It should be a steady, even, viscous drag. This tells you that your pitch lap is doing work on the glass. Of course, if you are polishing on a machine you don't care. You will just let the machine chug until the glass is polished or until oblivion, whichever comes first.