Optical Flat Light

I understand that optical flats require a single wavelength light source. I see micrometer checking sets advertised but never the light source. What am I missing?

I think generally the people who make the test gage sets (Mitutoyo etc) don't make lights for use with the flats. They expect the QC department already has one.

You can find them via Google, from several different companies. Edmund Optical www.edmundoptics.com makes some, as do Nikon, Zeiss, van Keuren, LapMaster,...

They're not cheap.

- Leigh

Not all micrometer checking sets will contain optical flats/parallels.

Many light emitting diodes (LEDs) put out a pretty narrow band of wavelengths. You might try that.

Cal

Cal Haines said:

Many light emitting diodes (LEDs) put out a pretty narrow band of wavelengths. You might try that.

Click to expand...

Good idea.

LED technology has advanced dramatically in the last couple of years. You should be able to find good monochromatic emitters with sufficient brightness for this application.

- Leigh

Our Van Keuren light is Helium, and looks a bit pinkish. A sodium light will give a real narrow spectrum yellow, and Argon glow lamps are a nice blue.
All will work well, as would a nice LED, as long as you put a frosted glass diffuser in front of the light.

A red incandescent light has a peaky spectrum. I've used them in an ordinary desk lamp many times for readings with optical flats. If you need to quantitify readings with this lash-up you have to remember to adjust for the wavelength. Red light is about 650 nM and the neon-helium used in traditional monochromatic light is a bit shorter I think.

A filter from a photographic supply store would work to narrow the spectrum.
Red light at 650nM would have a half-wavelength of 12.53uin.
A Van Keuren monochromatic half-wavelength is 11.56uin, and a blue light would be about 9.25uin.
Blue argon lights are useful when measuring very flat optical reference flats.
Half-wavelength is what you see as fringe lines when measuring flatness. The thickness of the air wedge between the optical flat and the surface being measured is one half wavelength between two adjacent fringe lines.
If you see NO lines (or circles, etc) either there are too many for you to resolve with your eye, meaning a really un-flat surface, OR there are no lines, meaning the surface is dead flat.
Usually, it will be somewhere between the two...
I once went insane and lapped the spindle and anvil of our Pratt & Whitney Standard Measuring Machine (SMM) to <.000001" in flatness and parallelism. Good thing we had two of them, as I put about 80+ hours into this project, which I did on my own time, as some know-it-all said it couldn't be done...

Some optics shops use a mercury vapor street light such as used as a night light at a farmer's barn. All you need is one color light. The street light is handy because it casts enough light to cover a big work bench so opticians can just put a test plate on a lens or a test flat on flat work and check progress on the work without having to get in a hood or wait in line to get under a smaller light. Fairly inexpensive and it is just another light in the shop which is generally under lite in most cases anyway.

Plain ol' ordinary red light bulb works pretty good. I've used one for years. Get the glass kind that's hard core red. The color coated lear glass kind suck. Tail loghts work good too.

Some optics shops use a mercury vapor street light such as used as a night light at a farmer's barn

Click to expand...

That's what I use, although it's a Sodium lamp. Been hand lapping to 1/20th wavelength on 12" flats for years.