How Grazing Incidence Works



  Why Grazing Incidence?

  Other Fizeau 

  Infrared Interferometers

  Dual Wavelength

  How Phase Works


So how do you measure the flatness of a surface with a matte or semi-matte finish?

Ordinary normal incidence interferometers are not very helpful because there is no specular reflection from the test surface, and therefore no interference fringes to analyze.

Some typical samples with low surface reflectance are shown at the left -- square of silicon wafer and a molded part of silicon carbide -- both with only a semi-matte finish.  Some dark materials such as silicon carbide or graphite present a further problem.  Even if showing a moderate polish, their dark color makes it difficult to see any interference fringes.

A number of ceramic parts and stainless steel seal plates are also very difficult to measure.  This can create a problem for the customer who is required to produce flat-lapped parts that are not fully polished.  The perforated stainless steel seal plate at the right is an example of a rather difficult  test piece -- particularly if you are making millions of them per year!. 

A typical specification of about "3 light bands" but not fully polished can result in a very difficult measurement problem for the manufacturer and his customer alike. 

Some parts have sufficient polish to use the contact method with an optical flat, but this procedure is fraught with peril, both from incorrect contacting methods and from misinterpretation. The possible damage to both the optical flat and the test piece are also very real possibilities.

The steel ring at the left is an example of a very matte fine-ground finish, which shows absolutely no specular reflection at normal incidence.  In order to circumvent the measurement problem, over the years, a number of "tricks" have been developed to give the surface a pseudo-polish --  things like rubbing the surface on a piece of paper that has been scribbled with a china marker -- a messy operation at best, and not very accurate.  Although in some cases it would permit the measurement with a normal incidence interferometer, there were inherent inaccuracies -- and you still had to clean off the china marker!

Since such low surface reflectivity effectively prohibits the use of a normal incidence interferometer for measuring flatness, another solution had to be developed.  That was the grazing incidence interferometer.

The Grazing Incidence Interferometer

Grazing incidence interferometers have been around for some time.  They are wonderful instruments, but until now, there were some serious disadvantages.

(1) The Systems are High Cost.

(2) The Systems required close contact between test piece and reference surface, causing various problems:

  • The very precise and delicate reference surface can be scratched or otherwise damages by the requirement to make close contact with the surface being tested (with or without a protective mesh).
  • Some substrates are so easily damaged that they can not safely be brought into contact with the reference surface.
BUT GRAHAM's Grazing Incidence Interferometers ... 
the "LAZER GRAZER" product line solve both of the above problems.

(1) The LAZER GRAZERS are low cost -- only slightly higher cost than our standard Fizeau Interferometers.

(2) Our "LAZER GRAZERS" are non-contact ... no delicate surface to damage, no problems with finger prints and other smudges 

(3) And no need to refinish the reference flat ... EVER!

So how does a LAZER GRAZER measure flatness of matte or semi-matte finished parts?

The drawing below illustrates the concept of Grazing Incidence reflection.  Even though the surface shown has a matte finish, at a low grazing angle the observer will be able to see a reflection of the light bulb.

A little experiment helps to understand how a grazing incidence interferometer works.  Take any unpolished part made of  metal, glass, plastic, or whatever, and hold it at an angle to the line of sight so the eye skims along the surface.  It is immediately apparent that the surface looks much more polished -- you can even see a reflection in a surface that showed no reflection at normal incidence.

At the right are two pieces of silicon wafer -- one fully polished and one with a matte finish.  Notice in the polished sample that you can see the reflection of the camera tripod used to position the camera.

These same two silicon wafers are shown in the image below. Since they are being viewed at a low grazing angle, they appear foreshortened. Note that silicon samples -- both the polished and the matte finished show clear reflections of the text shown to the right of the image.

With the camera placed to give a grazing angle view, 
note the reflection of the text in both surfaces.

These are just some simple examples of what grazing angle optics can do.  The Grazing Incidence Interferometer is a bit more complicated than that, but you get the idea.  The important concept, is that by examining a surface at a low grazing angle, we are able to produce interference fringes that will permit us to evaluate the surface configuration of the part being tested.

For rougher, less flat surfaces, longer wavelength infrared laser sources can be used. Such instruments provide the ability to check flatness of surfaces which would present so may interference fringes that analysis would be difficult.

In particular, instruments using a 1.55 micron wavelength laser still represent an economical solution to the problem of measuring parts with large departures from flat -- in this case with departures up to .003 inches -- putting the instrument well into the area where it can be useful for measuring standard machined parts! 


For more information on GRAHAM's Grazing Incidence Interferometers, click on Model 2GRP and Model 4 GRP
Call Gordon Graham today at (818) 700-1263      E-mail:

GRAHAM OPTICAL SYSTEMS, 9530 Topanga Canyon Blvd., Chatsworth, California 91311
Copyright © 2014 Graham Optical Systems  All Right Reserved
Durango is a trademark of Diffraction International            This page last updated July 16, 2014