Obsidian Optics production: nothing left to chance
   We test throughout our production process to assure the highest quality results:
  • A fine figure of revolution
       All mirrors are ground and polished using machine turntables built flat and mechanically stable, laser trued. Glass is supported on full flotation and always worked face up.
       Regular high power inspection at center of curvature (COC) of a LED illuminated 50 micron optical aperture (artificial star). This test is applied both to the original machine generated spherical figure and to the developing paraboloidal figure during work.
       This is an exquisitely sensitive test for errors of astigmatism of all degrees, and the best result is complete freedom from such errors. If the mirror fails this test at any point during parabolization it goes back to the machine for reworking.

  • Perfect edges
       All our mirrors show a balanced diffraction edge under Foucault knife-edge at center of curvature - the most stringent test of edge quality. Ronchi grating inspection at focus show no "hooks" or irregularities.
       This is the result of using optimal techniques during polishing. The edge is the most important area of the mirror and should never be compromised.

  • "Best polish we've ever seen"
       User reports of our mirrors exhibiting no - zero - scatter under green laser illumination are not a surprise. We extend initial polishing well beyond ordinary levels and regularly observe that the result is outstanding contrast performance.
       Polish is monitored with laser during machine use, inspected at the edge of the mirror with a loupe under diffuse illumination, and evaluated over the full mirror at center of curvature with halogen illumination. A final check is run after completion and final cleaning just prior to packing for coating.

  • Smoothness = contrast
       Optical smoothness is measured in fractional wavelengths. The Foucault test can reveal roughness well below 1/100th wavelength. Our typical results on Pyrex™ show none. Plate may show barely detectable levels.
       Our production method utilizes soft laps for parabolization and smoothing as needed. Precise indexing and stroke control generates incredibly smooth and regular paraboloidal correction. The mirror is parabolized as a whole, not as a collection of zones to be blended.
       All our production laps bear a special faceting pattern that eliminates periodic effects (primary ripple). Slow figuring controls irregular heating that can lead to irregular defects with larger patterns ("dog biscuit") and smaller roughness ("orange peel").

  • Testing throughout
       We use an optimized Foucault knife-edge tester utilizing a precision slit and bright orange 620nm LED light source. Couder masks are used for paraboloidal mirrors down to around f/4. Mirrors are tested multiple times during figuring and final testing is done on two axes after thermal equalization. Spherical mirrors are evaluated for extinction smoothness under the Foucault knife. Mirrors are supported on an adjustable test stand that eliminates errors due to gravity induced sag.
       Testing masks are optimized for specific f/ratios with equal deltas between correction zones. Longitudinal zonal measurements are typically repeatable to well below 1/50th wavelength level. Excellent mirrors that have no defects of revolution or surface are capable of being Foucault tested with great accuracy and the results may be reduced to Strehl estimates with a high degree of confidence.
       On request we will target the correction to your choice of effective mask radii - currently we utilize both the traditional arithmetic mean for Couder masks and recent diffraction modeling corrections, the difference in most cases is negligible.
       Our bench tests have been crosschecked repeatedly with star testing and field performance - as well as replicated by customers with high agreement. Interferometric certification is available on request for $300 additional and all interferometric results are posted along with bench testing.

  • Rating the results
       We provide a guarantee of wavefront rating with each mirror derived from the bench testing results on multiple axes, and an estimate of Strehl for those axes, as well as RMS surface error.
       We also rate the following qualities separately, a score below fine in any of the four calls for further work on the mirror:

Rating Figure of Revolution Edge Polish Smoothness
Excellent No visible error Diffraction edges identical at COC Brilliant polish with very high contrast No visible roughness
Fine Minor deviation well below mirror's wave rating Slight imbalance between left and right sides at COC No visible defects May exhibit minor burnish or barely visible roughness over small area
       Please note that the rated defects on any released mirror will never affect the at-focus performance measurably - the sole reason for rating is to continually improve our standard practice in delivering the best mirrors possible.


© 2005 Obsidian Optics