What's it about?

The importance of understanding the design and use of Fizeau transmission spheres, for accurate evaluation of surface irregularities, cannot be understated. There are several characteristics that have a significant impact on the interpretation of the resulting data. Specific tests can be performed on a routine basis to quantify and maintain calibration of the precision masters and as a result, errors attributed to the transmission sphere can be isolated. The overall goal is to obtain accurate evaluations of surface irregularities of the actual surface under test.


When optical components are tested interferometrically with transmission spheres, we assume that the resulting data output is the profile and/or error of the surface under test. This assumption may not always hold true. Based upon the level of test accuracy required, there are certain characteristics needed to evaluate the peak to valley test data.{Note: We have chosen to exclude the details of the interferometer accuracy / precision as well as transmission sphere alignment, which has been covered in subsequent writings.)


1.1 Characteristics of importance

Each of the below characteristics can be spoken to in great detail and length. However the purpose of this writing is to briefly identify the important characteristics of the transmission spheres so that one can accurately evaluate surface irregularities.

  • Optical design considerations
  • Aplanatic conditions, slope and mapping errors (distortion)
  • Wavefront - lack of focus and its effects (autocollimation)
  • Mechanical design considerations
  • Reference surface stability
  • Glass homogeniety and possible effects
  • Anti- Reflection coatings- hot spots and ghost images
  • Cosmetics -scratch/dig & cleanliness and their effects
  • Storage and handling considerations
  • Certification and calibration


1.2 Peak to valley considerations

Commercial manufacturers certify the quality of the transmission spheres to be typically on the order of λ/10 peak to valley. What does this mean? Is this sufficient? What should we be concerned about if we are evaluating λ/4 surfaces? What should we consider if we are evaluating λ/8 surfaces, λ/10 and smaller?


In the past, a peak to valley specification only was used to rate the quality of the transmission sphere (ex. λ,/10). This specification was the quality of the reference or aplanatic surface of the transmission sphere. Today's advances in manufacturing have demanded additional constraints. Specifications identifying localized surface slope errors of the reference surface as well as wavefront errors as measured in auto -collimation in the form of density functions (spot diagrams) and slope distribution function (geometric encircled energy) have been requested. These new specifications have been necessary to improve the quality of the transmission spheres for evaluation of aspheric surfaces and short convex spherical radii.

In evaluating surfaces up to λ/8, new commercially available transmission spheres are optically designed and mechanically robust enough to maintain the necessary wavefront accuracy needed to perform this level of test. Maintenance in the form of calibration is relatively benign and about the only consideration that needs to be performed on a regular bases is a "blow off' of the outer surfaces to remove dust and particulate.

For surface tests of λ /10 many considerations have to be rationalized. Special attention to reference surface stability, the transmitted wavefront as measured in auto -collimation, the cosmetic -scratch/dig & cleanliness of the transmission sphere all of which will have effects on the measurement test which can be misinterpreted as surface error.

1.3 Review of test data

We will capture and show with interferometric plots some of the resulting effects that can be seen with the Fizeau transmission spheres. We will present data to understand the effects of a defocused transmission spheres and its impact on measurements made on short convex radius surfaces. We will give advice on how to use and how to maintain the quality of these precision reference spheres.


Many optics facilities have several transmission spheres in their metrology test area. Given the substantial investment in capital attributed to each instrument, careful consideration must be taken to calibrate and care for these precision reference tools.


With the need for self -calibrating transmission spheres, technology is advancing towards achieving absolute test data from transmission spheres without the effects of contributing errors.




Joseph Pinto, QA Manager | Optimax Systems, Inc., 6367 Dean Parkway, Ontario, NY 14519


Optifab 2003: Technical Digest, edited by Walter C. Czajkowski, Toshihide Dohi, Hans Lauth, Harvey M. Pollicove, Proc. of SPIE Vol. 10314 (Vol. TD02), 1031415 · © (2003) 2017 SPIE

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 10/25/2018
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use