So I bought a Spitfire 12" lapping machine... optical flat selection?

I have been having a good time playing with a Spitfire Lapping machine I bought a while back. I have conditioned the plate so that no light gap is visible with a Rahn granite parallel. I have lapped a few things to the point where I believe the only real way to measure them is via Optical Flat.

I bought a DoAll flat that has .000001 written on the side. It is a 3" flat. Ebay item number 371240178473

I believe the .000001 is in inches, as DoAll is a us brand. that and if it were in metric, the flatness would be almost unbelieveable.

According to this website- Tutorial on Fringe Interpretation. .000001" is less than 1/20 wavelength of Helium Neon laser of 632.8 nanometers.

This leads me to believe that I have bought a very accurate optical flat.

Anyone have any input? I would love to hear about measurement with optical flats. I have been reading as much as I can, but would love some "practical" advice.

Use of an optical flat requires a monochromatic light source.
There are several different types available, as discussed in the Optical Flat Manual mentioned below.
Mercury gives the finest resolution, but you can make a cheapy from a sodium lamp if you prefer.

One major manufacturer of optical flats is Edmund Optics at www.edmundoptics.com

I believe they have some application literature available.

And of course Google will find others. Just search for: optical flat filetype:pdf
The "filetype" will limit the hits to actual documents.

I just tried that and got 934,000 hits, the first of which is the Optical Flat Manual from Edmund Optics.

- Leigh

leigh is right - when you buy an optical flat from edmunds it ships with a little booklet showing the basics of how to use it and how to interpret interference fringes.
i'm pretty sure you can find it on the website too.

1/20 of a wave is pretty good, but you must remember that it does not hold that form under all circumstances. for example, if you are holding the glass while you do your measurement it will distort due to the heat from your fingers. you can also distort it under load fairly easily. the nice thing about glass is that it is tough to do permanent damage short of breaking it. a few scratches generally dont effect a measurement.

also, if you don't have a monochromatic light source you'll need to get one.
you can sort of see fringes under a fluorescent light, and they improve with a yellow filter, but for my eyes at least its not worth dicking around with a crappy light.
the fringe clarity with a proper light source makes everything easier.

-mike

A HeNe laser works well if you add some beam expanding optics. Very easy to get fringes with that monochromatic source.

Make sure the work and the test glass are clean before putting them together. When you do wish to evaluate the work, Place a new optical tissue between work and test glass, then slide the tissue out, letting the glass surfaces come together. Let things set a moment then observe the fringe pattern. Pressing lightly with the fingers at the edges of the test flat is OK if fringes do not form right away.

Lift the test flat off the work, don't slide it EVER.

For most work, let the part and the test flat sit together at the observing station for at least 10 minutes to stabilize temps. Longer if the work is large.

Don't get excited about measuring flatness of thin workpieces. Glass is quite flexible.

Regarding the spitfire. It's better to keep the platen flat by constant monitoring and the use of conditioning plates, than to let it go until it needs major correction.
Also, if 1/4 wave per 25 mm flatness is "good enough" for your work. Use pads on the platens. The work will go more quickly, and with less "drama" if things go south. (crashes are a fact of life.)

I love planetary lapping machines! You can make things as flat as you have the patience for. Make your self a flatness gage from a piece of stout flat stock, two contact points on one end, one on the other. Mount a sensitive dial indicator about 1/3 in. The contact points should hit the outer diameter of the plate, and the indicator should be near the inner edge. Zero it on a decent surface plate. Use it frequently. If your plate becomes slightly (a couple tenths) concave, no biggie. Move the conditioning rings out a tad. If it starts to go even slightly convex, correct immediately. A convex plate is hellish to get back to flat. For fine work you'll know what's happening just from the optical flat results.