A Reversed Spot Diagram: Image & Point Source Effects

[Emperor42]

If you have an optics system where you know the spot diagram of the image at on imaging on one plain and projecting onto another. Where you know the spot size for a point source at a given distance from the centre of the field.

Now if you reverse the system where the imaging plain now becomes the object plain and vice versa, can you extrapolate the new spot sizes from the old ones? Does it depend on the magnification?

 

[Simon Bridge]

If you know the transition matrix for the optical system, then yes.
This is exactly the same as just finding the image knowing the object and the optical system.

If you do not, instead you only only know what happens when experiments are done in the forward direction (ie spot pattern A turns into image spot pattern A'), then you will need to model the optical system using knowledge of the initial object and the initial image.
From there you can deduce the expected result of reversing the direction given the model and the assumptions the model is derived from.
This sort of thing occupies much of science and it is a foundational postulate of scientifiic investigation that this approach is useful.

 

[Emperor42]

If I know the magnification of the system, can I just extrapolate the size of the spots if I reverse the image and optical plain, by just multiplying their size by the magnification factor? Because of the reversibility factor of optics.

 

[Simon Bridge]

In general - no. There can be anything in the box marked "system".
Compare the result with a simple lens with that for a telescope.

 

[Emperor42]

Hmm... its just difficult because the way Zemax works you have to have the image and object plain inverted for it to work for some reason. I can't reverse the black box lens so I'm left in a situation where I know roughly the quality of the lens, but I can't measure exactly how the spots will look on my CCD.

 

[Simon Bridge]

Oh you are asking how to do it in the simulator??
No idea. Like you said, you'd have to reverse the order of the componets in the black box unless you can get hold of the transfer matrix directly, then you just invert it.