The Ronchi Test

I highly recommend building a Ronchi tester. The Ronchi test is a quick, intuitive and useful test, and the tester is pretty straightforward to build. Most people find interpreting the Ronchi pattern easier than interpreting knife edge shadows or star testing, and it can get you a long way to finishing the mirror.

The parts

At its most simple, the Ronchi tester consists of the following parts:

A light source: You want a small fairly bright light source. Today it is very common to use bright LEDs for the light source. You can buy LEDs from Radio Shack or any competent electronics supply place. Other possibilities are small light bulbs, the ones that power tiny mag lights are ideal. In fact, using one of the mini maglights as a light source is a perfectly reasonable thing.
A diffuser: It is important that the light emitted by your tester not be highly directional. You can diffuse the light source in many ways. I've taken #220 grit and rubbed it on a microscope coverslip. I've similarly frosted a tiny prism with #220 grit. Or you could go low tech and merely wrap the light source with a few layers of sticky translucent tape. Or bore a hole, place the light source at one end and stuff some cotton in to diffuse it. It isn't rocket science.
A Ronchi grating: The one specialized bit of kit that you will need is a Ronchi screen. It is nothing more than a grating which has opaque and clear lines on it. You can get them with any line spacing you like, but 100 lines per inch are the most common. If you are making mirrors that are sixteen inches or bigger, you might want to get a coarser grating, say 80 lines per inch. You can buy these from places like Edmund Scientific. A 1" square grating will be just fine. You can also make them yourself. Check the bottom of this page for instructions and links.
Batteries, switches, and a small box.: I picked up all these goodies from Radio Shack.

Assembly

It isn't tricky to make a Ronchi tester that works. There are only a few things to remember:

Light from the bulb must be diffused.
Light leaving the bulb must pass through the Ronchi grating on the way to the mirror.
The light which returns from the mirror must pass through the grating on the way to your eye.
The distance between the light source and the return path must be as small as is practical, say 1/2 or so.

My first Ronchi tester was made with project box. I drilled a 1/4 inch hole near the top of the box. Directly behind the hole I placed my scuffed up microscope slide, which I glued in place with some silicone glue. I soldered a couple of leads onto the maglight bulb, and then to a switch, and then to a battery holder. I then screwed the lid to the project box back on and taped the Ronchi grating so that it overlapped both the hole and the edge of the box. Voila!

Other testers

Most other pages describe combination Foucault/Ronchi testers. These are usually a bit more complex, because they need a way of measuring the precise distance that the knife edge moves for the Focault test, but if you get serious about testing you probably want to make one of these sooner or later.


A Foucault Tester: Nicely done instructions for a combination Foucault/Ronchi tester

Foucault/Ronchi Tester: A nicely done home made tester with lots of pictures.

Stellafane's Build a Foucault and Ronchi Tester: Stellafane is the mother of all telescope making conferences, so you might expect they'd know something about telescope testing. Good explanations and descriptions on how to build a combination tester.

Performing the test

The test is actually quite simple. You merely position the tester at the center of curvature of the mirror, allow the light to bounce off the mirror and return through the grating to your eye.

This is quite a bit easier if you take the time to build a stand to hold your mirror. At the Chabot Telescope Making Workshop, we have a nice metal mirror stand that we've used for decades. I decided I needed one for my home workshop, and build my own interpretation. The most useful adjustment is a screw adjustment so you can adjust the tilt of the mirror up and down to get it centered on your Ronchi grating.

Normally I pick up the reflection of the mirror by moving fairly close, and then backing up. When you get close to the center of curvature (which is twice the focal length from your mirror), you'll begin to see fewer and fewer lines. When a single line covers the entire mirror, then you are at the center of curvature. Move the grating inward (toward the center of your mirror) until you can view five dark bands, with one of them centered on the mirror.

How do I interpret what I see?

This table will help you interpret the patterns that you see on the mirror.

The easiest pattern to recognize is a sphere. It has nice straight lines which run all the way accross the mirror. By pushing the mirror in and out, you see more or fewer lines, but they remain straight both inside and outside of the radius of curvature.

If the lines bow outward like this when viewed inside the radius of curvature, then you are somewhere in the range of a parabola. The exact amount of bowing that you get will depend on the focal ratio of the mirror. Shorter focal lengths have more steeply curved lines, while longer focal lengths may be indistinguishable from the spherical straight line case.

If the lines bow inward like this when viewed inside the radius of curvature, then you have a curve which is called an oblate spheroid. I will sometimes shorten this to saying you are oblate or too flat. To correct this to a sphere, you need to remove more material from the center of the mirror. WARNING: Be careful to remember whether you are inside or outside of focus. If you look at the paraboloid outside of focus, it will look like the oblate and vice versa. This is why I normally standardize on the figure inside the radius of curvature.

Another common malady for mirrors is a turned down edge. This is a parabolic mirror, but as the mirror gets closed to the edge, the lines hook in rather sharply. This indicates that the mirror has a turned down edge.

The Matching Ronchi Test

Normally we say that the Ronchi test is qualitative. It can tell you when you are getting close, but it doesn't spit out a number which represents the wave rating of your mirror. Mel Bartels uses Ronchi testing to get a lot closer to the desired figure by using the Matching Ronchi Test.

The basic idea is to use a computer to predict precisely what each Ronchi pattern should look like at various offsets in and outside of focus. Then actually perform the test by trying to see if the figure matches the predicted patterns precisely.

As an example, the patterns he usually draws up are done at .2 inch intervals, beginning .3 inches inside focus and ending .7 inches outside. Below are some of the patterns that you might expect for common sizes and focal lengths.

You can read up more on the Matching Ronchi test at Mel's webpage.

Example Ronchi Patterns for the Matching Ronchi Test
Mirror Size -0.3 inches -0.1 inches 0.1 inches 0.3 inches 0.5 inches 0.7 inches

8"f/6

12.5"f/5

How to make your own Ronchi grating

If you are on a budget or just want to make a single telescope, spending money on a Ronchi screen may not seem to be the most economical thing to do. There are several ways you can actually make your own without a great deal of expense or difficulty.

Wind one using monofilament line

An excellent description of making your own Ronchi screen from monofilament line was written up by Clint Bach. The basic idea is to make two frames out of some fairly stiff material, bore matching holes through both of them, and then wrap a pair of monofilament lines around the frame parallel to each other. You then unwind one of the strands, glue the remaining strands in place, and then split the frames apart and you have two matching Ronchi screens.

I've used a screen like this, and they work great. You need to carefully measure the thickness of the strands to figure out the lines per inch.

Print one on transparency film

If you have access to a good printer, you can actually use some kind of drafting program to print these out directly. This can be a little tricky, because to get good results you need to make sure each line has the same width. For instance, a 300dpi printer can only make 150 or 75 line screens. If you try to make a 100lpi screen, you get black lines which alternate being one or two pixels wide. If you can get a 600dpi or higher printer, then 100lpi screens are doable.

Here is a semi-useful chunk of PostScript that I used to print out screens for our workshop. I printed these on normal transparency film using a 600dpi HP LaserJet. It should work on any 600dpi Postscript printer.

Reduce a coarser grating photographically

If you can't print one, you could make a coarser grid and then photograph it onto black and white film. The developed negative will be a Ronchi screen. Is this cheaper than buying one? Perhaps if you made a bunch of them for a club.

Mark T. VandeWettering <markv@telescopemaking.org>

The easiest pattern to recognize is a sphere. It has nice straight lines which run all the way accross the mirror. By pushing the mirror in and out, you see more or fewer lines, but they remain straight both inside and outside of the radius of curvature.
If the lines bow outward like this when viewed inside the radius of curvature, then you are somewhere in the range of a parabola. The exact amount of bowing that you get will depend on the focal ratio of the mirror. Shorter focal lengths have more steeply curved lines, while longer focal lengths may be indistinguishable from the spherical straight line case.
If the lines bow inward like this when viewed inside the radius of curvature, then you have a curve which is called an oblate spheroid. I will sometimes shorten this to saying you are oblate or too flat. To correct this to a sphere, you need to remove more material from the center of the mirror. WARNING: Be careful to remember whether you are inside or outside of focus. If you look at the paraboloid outside of focus, it will look like the oblate and vice versa. This is why I normally standardize on the figure inside the radius of curvature.
Another common malady for mirrors is a turned down edge. This is a parabolic mirror, but as the mirror gets closed to the edge, the lines hook in rather sharply. This indicates that the mirror has a turned down edge.