Reading an optical flat correctly

[NBR90]

So I have three optical flats of 1/20, 1/10 wave and 1/4 wave accuracy.

The surface test piece (A) that is being measure is flatter than 1/4 wave. Let's assume it has accuracy of ~1/8 wave.

So from my understanding, the 1/4 wave optics should show all flat parallel lines when places above the (A) piece under monochromatic sodium light. All this means is that the test piece (A) is as flat or flatter than 1/4 wave.

My question is, how would the 1/8 wave flat piece appear under 1/20 wave and 1/10 wave optics?

I assume it would appear more curvy (fringes) under 1/20 and less fringes under 1/10.

Since one light band wave under sodium light is 0.00029 mm or 0.0000116 inches and two fringes equals 1 light band;

Therefore, I would assume I would see few fringes under 1/10 optics and even more fringes under 1/20 wave?

Without actually knowing the test piece (A) was 1/5 wave flat, how would I estimate the flatness of the test piece (A)? By counting the number of fringes and figure out its ratio to the 1/10 and 1/20 optics?

 

[NBR90]

Correction for last sentence, it should read "Without knowing the test piece (A) was [highlight #CC0000]1/8[/highlight] wave flat..."

 

[Ceramicguy]

Go to this link at Edmunds Optics for a basic education on optical flats:

http://www.edmundoptics.com/technic...g-literature/files/eo-optical-flat-manual.pdf

Basically, you are measuring the thickness of the gap between your reference flat and your test surface. The potential exists for the measurement to be the sum of the two flatnesses. Your flatness with a quarter wave reference may appear worse on your test surface than it is. I suggest you use the rule of ten in your measurement methodology where your "instrument" should be at least ten times more precise than the tolerance of the feature you are trying to measure.

Bruce

http://accuratus.com