From: "Dwight K. Elvey" 
To: atm{at}shore.net
Reply-To: "Dwight K. Elvey" 


>From: DWilliams1128{at}aol.com
>To: atm{at}shore.net
>Subject: ATM mirrors and lenses
>MIME-Version: 1.0
>X-Report-Errors-to: mikell{at}optonline.net
>X-WSS-ID: 121BFAF7590109-01-02
>Content-Transfer-Encoding: 8bit
>
>
>Hello:
>
>I was wondering how you get any accuracy in a Foucalt tester.  I understand
that it has a light source, a knife edge (to cut incoming light or outgoing
light), an eyepiece,

Hi
 I'm not sure what the eyepiece is for? One could use
a small low power telescope but that is with an eyepiece and and an objective.

 and a micrometer.  Also if you put the light source at 2 times the focal
length and close to the axis line and the look at the same length from the
mirror, but next to the light source you will see the to any degree of
magnification.  As the distance to the radius of curvature approaches zero
the magnification approaches infinity.  You can use this setup to measure
for sphericity, parobolic curve, hyperbolic, or eliptical or another words
the change in focal length or curvature in different zones or how far from
the light cone axis you are.
>
>  There are a couple of problems.  The fixed or moving light source,
whether the micrometer screw is left handed or right handed.  And to me it
seames that you would have to know the distance from the mirror to the
Foucalt tester.  And how would you do that without using a tape measure and
that seems awfuly inaccurate.

 We are measuring the difference in slope from a sphere. Since
the length of the light ray to the mirror is so large, a small change in
slope will be magnified by that distance. Since we are measuring a
difference, the absolute value isn't as important. Try to imagine you had a
really long square. You moved it over an uneven surface. As you moved
along, the free end would tend to wobble. We can see even tiny wobble,
clearly, as the beam of light missing the knife edge and multiplied in
effect by the length.
 As Bob mentioned in another post, + is away from the mirror,
regardless of the micrometer scale. My tester has the micrometer reversed.
It is just another math operation to reverse the scale.
Dwight


>
>   Do anybody know how it works?  And how to get accuracies of 1/100,000
inch or wavelengths of light?
>
>   I haven't built one yet, but probably will have to.  After my 20" mirror
I might build a 40" mirror and get it from Newport Glass.  I might get
the curve pre-generated from there.  Either that or I might buy their
8" refractor telescope lens kit with tool.  If I do that I'll need
dial deflector readout guage that machinists use to determine out of round
of parts.  I'd need to use that to make sure there's no wedge of the lens.
>
>  Has anyone here made lenses before?
>
>  Thank you for your help.
>  David Williams
>  dwilliams1128{at}aol.com
>
>

--- BBBS/NT v4.00 MP