What's new
What's new

checking flatness with laser

dian

Titanium
Joined
Feb 22, 2010
Location
ch
there is a method where a laser beam is projected parallel to a surface. if i recollect correctly, the distance between two projected lines is used to make a judgement on flatness. i tried it a few years ago and it sort of worked, but the set up was very crude. now im not able to find any info on it.

1. what is this method called?
2. is something based on this being used in professional measurement?

thanks
 
There is inferometer. Basically, it projects light onto a surface. It diverts part of that light beam. The difference between the diverted and the straight is the distance. I use one at work. They can measure angles, spheres etc.
Another machine I use does use a laser on the upgraded model, that I don't have. Its a CMM, but more like an Optical comparator.
 
there is a method where a laser beam is projected parallel to a surface. if i recollect correctly, the distance between two projected lines is used to make a judgement on flatness. i tried it a few years ago and it sort of worked, but the set up was very crude. now im not able to find any info on it.

1. what is this method called?
2. is something based on this being used in professional measurement?

thanks

I don't recognize this as a real method for checking flatness of a surface, as I think of that concept. There are interferometric systems that employ "grazing-incidence" illumination, which can be used for flatness evaluation of test objects that are not highly reflective, such as lapped and ground surfaces. One company that makes (made) that sort of system is Tropel.
The problem with optical interferometry is always getting light back into the instrument from the test object; if the test part is non-reflective, you're hosed.
 
my setup was as follows: a laser parallel to the surface of a granite plate. a piece of paper on the other end. there were two projection dots. the vertical distance between them was an indication of the flatness. i believe i did the same with a line laser as well.

now i dont remember the math behind it and cant find anything because i dont know what to search for.
 
If you are looking for crude options you can use an optical flat, a laser beam and a tennis ball. Cut a hole in the tennis ball and put the tip of the laser pointer inside. The tennis ball will diffuse the light which can then be projected on the optical flat (which is on top of the surface you are checking) and the light bands you see through the flat will give you a good indication of how flat the surface is.
 
thanks for that. interesting reading. this is probably the closest to what i had in mind:

http://www.aspe.net/publications/Annual_1999/PAPERS/OPTICMET/KULAWIEC.PDF

as most procedures i found use mirrors. however the beam always is described as at a flat angle, whereas my beam was parallel, so no idea if its really the same. what i learned is that the angled beam is stretching the waves (instead of using infrared?) and therefore i guess it must decrease resolution.

corednc, i sincerely wish i had an optical flat the size of the surface plate (and a crane to handle it) and that it would work on granite.
 
It sounds like an interesting project if you just have time and want another hobby. I've got serious reservations about whether it could ever be developed into a reliable quantitative assessment of the surface. Too much is up to judgement and opinion. So theoretically you've got a reflected band whose width is a representation of the amount of error and lighter or darker areas within that give you the approximate position of the errors.

To me, it's just a lot of chasing a subject around a tree with a lot of sturm und drang but never concluding the pursuit. Witness a similar discussion on that forum about why an autocollimator and a Repeat-O-Meter can utterly fail to find any problem with a surface. That argument hinges around regular, periodic waves in the surface that fall exactly in synchronicity with the spacing of the mirror and ROM feet such that errors are never apparent. One could spend the remainder of a lifetime spinning out hypotheses of what about a surface that has XXX, or it just happens that the instrument has this characteristic which couples with the whatever. If a person wants to spend their time with it, that's their business. If they suck you into their demands for an answer to every hypothetical, that you're issue.

If there is a paper from a reliable source describing how to generate an error report on a reasonable sized surface plate like an autocollimator, ROM, and level traverse would do but using just a glancing laser, I'll say sorry, I was mistaken in my judgement.
 
I don't have a lot of experience with lasers, but it seems to me that they are typically used when you are trying to figure a dimension between 2 points. When you are figuring flatness however, you are not measuring between two points, you are seeing how 3 or more points relate to each other. If those 3 points fall in line, they are flat. You could use a laser to shoot between the two points, but for however accurate the laser is, you are still limited to how accurate you are finding the other two points. That's why stuff like autocollimators and optical comparators work to such high accuracy with simple light projection, because it's the other components that are really making it work.
 
well, looking at the document i mentioned in post #7 it seem to work somehow. i wish i could find more about it.
 
well, looking at the document i mentioned in post #7 it seem to work somehow. i wish i could find more about it.


Yes, I've read the article. It doesn't correlate well with Evan's work noted on HSM because it's different apparatus other than using a laser. The paper uses a variety of lenses and prisms to project an interference pattern that's the difference between the direct beam and the reflected one and from which they derive deviation numbers. All well and good. It's a self contained apparatus that measures and area about 1" by 15". SO, do they have a good control for what happens when you move the apparatus? You've mapped a small section of a larger surface, but to really know the whole surface you have to stitch together all the 1" and 15" moves. How do you do that and control the whole thing? AND, they show a chart of what they read the surface error to be but have they verified that with another measurement system? In other words, what do we really know about the calibration of their machine other than their assertion that this is an accurate measurement?

It's an interesting article but it doesn't actually give me confidence in a way to map a larger surface or a machine way longer than 15" assuming you trust them, with all that presumes.
 
A few points:
The illumination source is a laser with substantial coherence length, so the interference is created (not projected) as a function of recombining the two beams, the direct (reference) path beam, and the test beam that is reflected from the test surface at a low angle. The diffraction gratings are what makes this method work as a fundamental principle. The lenses do some beam shaping, and prisms are simply items of convenience for beam handling to reduce the physical size of the instrument.

The calibration of the instrument is demonstrated by the measurement of the 18" optical flat (stated to be flat to 0.1 um, or about 1/6th wave), and then the subsequent subtraction of that calibration measurement as a reference map for comparison, thereby removing the anomalies that may be artifacts of the measurement system when processing the data acquisition from the object under test. This is a universal method for this sort of interferometric phase measurement map.

Stitching of a series of 1" x 15" maps can be done reasonably well by competent scientific programmers, based on matching pixels and phase information in overlapped map areas, and this is fairly well understood also. The shape of the data area in this particular instrument looks to me like a minor disadvantage in stitching, as the areas are small in one dimension, and may require a larger number of measurements to get a high-confidence stitched map in which all the sub-maps are leveled and fitted properly.

Interesting paper about trying to make a portable piece of grazing-incidence hardware; Tropel's work known to me was in fixed instruments that can measure 200mm diameter wafers and similar items.
 
It seems like an autocollimator and known 90 degree reference block is easier and more straightforward. Best method depends on how large a surface you need to measure. For some things the old taut wire method and a microscope work well.
 
Page 28 of Moore's fma book shows this Screenshot_20200607-062027_Hancom Office Editor.jpg

Someday I will build one except I will use three tooling balls that are adjustable in height and can be reached by a indicator from either side
The tool will have 3 point contact on the plate on each side
This is probably the hardest part about building this tool because these need to be the SAME height!

To (calibrate) the tool measure one of the balls then invert the tool and measure the same ball now adjust 1/2 the error out invert the tool again and check repeat as necessary for all the ball's
At the end you will have 3 balls in a straight line minus the effects of heat gravity friction and any size difference in the balls and tool.

Ps it really only needs two balls one in the middle of the tool and one at the support pads I would still use three just to keep things symmetrical but it wouldn't give you any information about the plate.
Hope this makes sense
 








 
Back
Top