[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

ATM Lumicon coma corrector



I've made two types spherometers, calibrated against three mirrors of 
known radius of curvature, calibrated on a flat, run numbers through 
two different raytrace software packages and I'm getting good cross 
checks and agreement.  I'm pretty confident of the numbers I'm 
getting so this is probably the last post of raw data:

Lumicon Coma Corrector - Reverse Engineering Data

OVERVIEW:  Two lenses, air spaced, both meniscus.  See diagram below
for lens and surface numbering scheme I'll use.  All four surfaces
bulge out toward the main mirror (away from the film plane).  The lens
closest to the main mirror (lens 1) is a positive meniscus lens (forms
a real image of distant objects if you hold it in front of paper) the
lens closest to the film plane (lens 2) is a negative meniscus lens.
Both elements are BK-7 crown glass.

CRUDE ASCII DIAGRAM OF LENS LAYOUT:

<html><pre>
F P      +---+     +----+       M  M
I L       +   +     +    +      A  I
L A        +   +    +     +     I  R
M N        +   +    +     +     N  R
  E       +   +     +    +         O
         +---+     +----+          R

         |   |     |    |
SURFACE: 4   3     2    1

         LENS      LENS
           2         1
</pre></html>


PHYSICAL DIMENSIONS:  Both lenses are 1.870 inch diameter.  With the
chamfer the clear aperture is approx 1.84 inch for both lenses. Spacer
ring between lenses is 0.078 inch.  Lens 1 is 0.433 inch thick *at the
edge* and lens 2 is 0.305 inch thick *at the edge.*  Working backwards
from these values and using new spherometer data:  lens 1 is .514 inch
thick *at the center* and lens 2 is .197 inch thick *at the center*.  
NOTE.  Although the spacer is .078 inch thick, the stronger curve of 
surface 3 compared to surface 2 means the center of these two 
surfaces are only .003 inch apart!

Radii of curvature:

Surface 1:  4.12 inch radius of curvature
Surface 2:  18.40 inch radius of curvature
Surface 3:  4.12 inch radius of curvature
Surface 4:  2.10 inch radius of curvature

NOTE:  Lumicon says that recommended spacing is rear of cell 3.57 
inches in front of film plane.  Add the cell wall thickness and 
curvature of surface 4 and this means the middle of surface 4 is 
3.822 inches in front of the film plane.

With the revised data above, the performance in raytrace software is 
even better - for the 17 inch f/4.5 paraboloid it was designed for.  
Also, spacing of corrector from the film is even more critical that 
with my previous lens data - move the cell .05 inch and off axis spot 
diagrams are quite different.  This improved performance in the 
software and greater criticality in spacing gives me more confidence 
that I've done a reasonably good job reverse engineering this animal.

Want to refigure your 17 inch f/4.5?  Go for about 1.115 
eccentricity.

What about my 6 inch f/3.75?  The corrector, with this better data, 
is not nearly as good for this mirror.  I need an eccentricity of about 1.35 
and it appears the surface of best focus is pretty strongly curved - 
no surprise I suppose.  It appears best focus .76 inch off axis is 
about .016 inch behind the best focus on axis (i.e. .016 inch farther 
from the mirror).  If I could curve the pressure plate (convex) in a 35mm 
camera by this amount I can keep the spot diagrams from getting above 
about .004 inch diameter.  I know this isn't so good, but without the 
corrector it would be far worse.

QUESTION:  Is it too difficult for me to take the camera pressure 
plate and grind it against a tool of the proper radius of curvature 
(the pressure plate will be convex) and get the film to behave in the 
camera without a vacuum system?  Any other ideas?

Capt Tom Krajci
B-52 Intelligence Officer
"In God we trust, all others we monitor!"
http://spur.barksdale.af.mil