Date: Fri, 03 Nov 2000 17:53:37 -0600

Author: brian whatcott

Subject: Re: Measuring Ball Bearings


I hold that it is unwise to rule out the precision
achievable by a determined worker. I hear that folks
have been known to produce parabolic surfaces to a
surface tolerance of 200 nanometers over much of
the real estate measuring 12 cm. or more.
By hand.
That's about 8 microinches surface variation of a surface
more difficult to produce than a spherical one.

Then they look at the starry sky with it.

I was hoping some one could remember the ball
bearing process better than I could - as a grinding process
between planes. An off the shelf ball (grade#25) from
Stock Drive Products has 25 microinches tolerance on the
diameter by the way.

At 13:36 11/3/00 -0500, you wrote:
>A careful apprentice can machine an object on a lathe to one part in
>10,000 if the object is cylindrically symmetric. A ball bearing's
>spherical symmetry taxes a common lathe way beyond the lathe's
>* * * * * * * *
>Mike McFarland
>Resources for Instructional Labs and Lectures
>Physics, University of Notre Dame

brian whatcott Altus OK