Mirror testing at the Telescope Makers Workshop


test-set.jpg (118808 bytes)

Here is the mirror testing set at the telescope makers workshop. The mirror is on an adjustable platform to allow for easy adjustment. A small bungee holds the mirror.

comp-pat.jpg (56980 bytes)

After measuring the focal length the mirror parameters have been entered into a computer program to generate a simulated ronchi-gram. Always using the same setup with the grating inside of focus allows us to become experienced at reading mirrors. Normally the computer printout would have a dark line in the center with a white line on each side. During mirror testing moving the grating left or right causes the lines to zip by at lightning speed.
gratslit.jpg (543568 bytes) In this setup the same grating produces both the slit and the filter for light returning from the mirror. A small flashlight bulb with a dimmer is behind the grating and three slits are actually produced due to the size of the light bulb. The three slits cause no problems while visually testing but seem to cause a triple image when using the camera. All the picture below have this triple image effect to some degree.

Here are the ronchi-grams that we compare to the computer simulation.
The mirror has a turned down edge at the outer 1/4 inch of aperture.
Try to look for this in the ronchi-grams.

norm-00b.gif (36760 bytes) norm-45b.gif (43071 bytes) norm-90b.gif (49642 bytes)
12 o'clock 1:30 3:00 o'clock

Here the grating is further inside of focus. The test becomes less sensitive further inside of focus.
Too bad the triple image effect washes out the contrast.
To keep from becoming confused we usually test inside of focus.
One exception is when trying to detect a turned down edge.
Outside of focus the lines would spread rather than shrink at the mirror edge.
For some reason testing for turned edge outside of focus seems to be more sensitive.

in-00b.gif (50922 bytes) in-45b.gif (46558 bytes) in-90b.gif (47184 bytes)
12 o'clock 1:30 3:00 o'clock

 Same mirror further inside of focus. 



The Knife Edge and Ronchi Test Equipment

jan-67-01.gif (835226 bytes) The knife edge setup is an adaptation of a set described in Sky and Telescope. This set is extremely sensitive and relatively easy to use. When the article was written bright LED's were not available.
knife-edge-01-gif.gif (43836 bytes) Here we test my 12 1/2 inch mirror after I attempted to reduce the turned edge by polishing just inside the edge. The zonal polishing didn't reduce the 1/4 inch wide turned edge by much but made quite a zone on the mirror. The knife edge would be on the right side and the imagined rising sun shinning from the left. Think of the rising sun lighting the sharply turned left edge. The knife edge is closer to the mirror than the focal plane of the mirror edge.
Design for a home built Knife Edge / Ronchi Tester

The Final Mirror Testing

chc01gif.gif (240633 bytes) Eventually the mirror will look exactly the same as the computer simulation even to Paul's experienced eyes.
chd01gif.gif (440407 bytes) Paul then takes the knife edge reading for several percent zones and enters the data in the datasheet.
che02gif.gif (136920 bytes) He then enters the readings into the other computer program to get the surface printout. The surface printout shows a measure of the mirrors surface in waves of light. Mirror by Paul Harris, August 1992.
Paul-Chabot.jpg (142259 bytes) Here Paul Zurakowsky interprets the computer results for another anxious mirror maker.
tele-top-thumb-nail.gif (21290 bytes) Paul Zurakowsky used an article from Tele-Topics to help in developing the interpretive graphing of mirror aberrations.
The article also provides an explanation as to why the mirror surface must be quite a bit better than 1/4 wave to meet the classic Rayleigh limit.



Experimenting with heat effects

hand_a.jpg (34988 bytes) When the mirror was first brought in from the warm car the entire mirror showed these heat waves. My hand is actually a foot or so in front of the mirror.
hand_b.jpg (34171 bytes) Cool air settles to the lower part of the room. As people walk by stirring the air heat waves are produced.
iron_a.jpg (36971 bytes) Here is a soldering iron we just plugged in.
iron_b.jpg (40155 bytes) And now its getting hot. At one time optical setups like this were used to see density waves in aeronautical wind tunnels.

Experimenting with mirror warping

A spherical mirror can be warped to be a parabola using a special harness. Also sometimes the corners of square mirrors are warped slightly to bring them into tolerance. In this case we are starting with a reasonably good mirror and warping it out of shape to see what happens to the ronchi lines. Our harness lifts at two points at the edge of the back and presses down with a compressed spring in the center of the mirror face. We used foam tape to keep the spring from damaging the glass.

Here the spring pressed down 20 pounds in the center and lifts up 10 pounds at two points on the back edge. The mirror only weighs 20 pounds so its hard to imagine a mirror cell that would support the mirror this poorly.

20-00.gif (51388 bytes) 20-45.gif (38348 bytes) 20-90.gif (49275 bytes)
20 pounds 12 o'clock 20 pounds 1:30 20 pounds 3 o'clock
Here I further compress the spring to produce 40 pounds pressure at the center and 20 pounds lift at each edge. With this much warping force I was beginning to have a bit of panic attack. At 12 o'clock the turned down edge has become a turned up edge. Amazingly with the harness at 3 o'clock you can hardly see a thing. This much astigmatism would show up easily in an eyepiece test. If you are doing your final testing using the ronchi test you should be careful to maintain mirror symmetry otherwise you might miss astigmatism during the ronchi test.
40-00.gif (53110 bytes) 40-45.gif (49286 bytes) 40-90.gif (42513 bytes)
40 pounds 12 o'clock 40 pounds 1:30 40 pounds 3 o'clock


Return to calling page