Diffractiongrating

While perfect imaging of the source would be smaller perfect circles of light, this shows the smearing of the light by diffraction into the bullseye patterns.

Have you had a chance to consider David’s points. I think he gets to the heart of the matter. In normal usage, the definitions of the chief ray as passing through the center of the aperture stop or one of the pupils are absolutely identical. OpticStudio does expand on this a bit, in that if vignetting factors are applied, then the vignetted pupil may not be the aperture stop exactly. You can find this discussion in the help page ‘Chief Ray’ in the section Conventions and Definitions, as well as the link to the vignetting factors help page.

Rayleigh criterion

This shows the intensity curves for the radial distribution of the diffracted light for different separations. Your eye sees the characteristic bullseye distribution of light as illustrated below. While perfect imaging of the source would be smaller perfect circles of light, this shows the smearing of the light by diffraction into the bullseye patterns. For modern digital photography where the images are projected onto a CCD, the information is collected on pixels of the digital detector. At left is an attempt to show the effect of diffraction on such imaging in cases where the diffraction is the phenomenon that limits the resolution. If the image is in focus and free of visible affects of lens aberrations, then it may be that it will fit on one pixel. But if the aperture is small enough, then diffraction can spread the image onto neighboring pixels and constitute the limit on the resolution of the image. References: The Cambridge in Colour site has calculations and demonstrations of diffraction spreading at different f-numbers.

For systems with signification pupil aberrations (Analyze > Aberrations > Pupil Aberrations), the ray in the center of the EP will not go through the center of the Stop.

References: The Cambridge in Colour site has calculations and demonstrations of diffraction spreading at different f-numbers.

Fraunhoferdiffraction

I tried to figured out the difference between the chief ray definition by the classical books and the convention by ZEMAX.

I don’t think the definitions are different between ZEMAX and classical textbooks. I can see why it can be confusing sometimes. For example, in the University of Arizona figure, the entrance pupil is on the image side, and this might not be so intuitive at first glance (perhaps think of the entrance pupil as a virtual image of the aperture stop in this instance).

Diffractionresolution

If an image is made through a small aperture, there is a point at which the resolution of the image is limited by the aperture diffraction. As a matter of general practice in photographic optics, the use of a smaller aperture (larger f-number) will give greater depth of field and a generally sharper image. But if the aperture is made too small, the effects of the diffraction will be large enough to begin to reduce that sharpness, and you have reached the point of diffraction-limited imaging.

I’m thinking that because OpticStudio is a piece of software rather than a century-old definition from the days of paper-and-pencil design, it made sense to have a more expansive definition. I don’t know this for sure, but it seems a likely explanation. The CR as ray through center of AS is going to be a subset of the possible CRs when vignetting is applied.

If you are imaging two points of light, then the smallest separation at which you could discern that there are two could reasonably be used as the limit of resolution of the imaging process. Presuming that diffraction is the determining factor, then the generally accepted criterion for the minimum resolvable detail is the Rayleigh criterion. This shows the intensity curves for the radial distribution of the diffracted light for different separations. Your eye sees the characteristic bullseye distribution of light as illustrated below. While perfect imaging of the source would be smaller perfect circles of light, this shows the smearing of the light by diffraction into the bullseye patterns. For modern digital photography where the images are projected onto a CCD, the information is collected on pixels of the digital detector. At left is an attempt to show the effect of diffraction on such imaging in cases where the diffraction is the phenomenon that limits the resolution. If the image is in focus and free of visible affects of lens aberrations, then it may be that it will fit on one pixel. But if the aperture is small enough, then diffraction can spread the image onto neighboring pixels and constitute the limit on the resolution of the image. References: The Cambridge in Colour site has calculations and demonstrations of diffraction spreading at different f-numbers.

Abbediffraction limit

For modern digital photography where the images are projected onto a CCD, the information is collected on pixels of the digital detector. At left is an attempt to show the effect of diffraction on such imaging in cases where the diffraction is the phenomenon that limits the resolution. If the image is in focus and free of visible affects of lens aberrations, then it may be that it will fit on one pixel. But if the aperture is small enough, then diffraction can spread the image onto neighboring pixels and constitute the limit on the resolution of the image.

Do not provide any information or data that is restricted by applicable law, including by the People’s Republic of China’s Cybersecurity and Data Security Laws ( e.g., Important Data, National Core Data, etc.).

Diffraction limitcalculator

the chief ray, by SPIE and Arizona University, is defined to pass through the center of the aperture stop. whereas, is defined to pass through the center of the entrance pupil. the difference causes a lot of troubles using ZEMAX. Can anybody explain this to me ?

Ray aiming should be turned on by default in 2023. Back in the 90s when Ken first wrote the code, he defaulted it off because users were running on 286, 386 processors and the like. There’s no reason for that now.

In the first figure that you show from Arizona University, if you loolk at the initial trajectory of the chief ray, it is indeed aiming at the center of the entrance pupil. I’ve tried to highlight this with a blue (dash-dot-dot) arrow in my screenshot below (sorry for colorblind people, its all I could do with the original figure colorcode):

If the system is paraxial, the entrance/exit pupils are images of the aperture stop. That means, if a ray goes through the center of the entrance pupil. Then, that same ray will also go through the center of the aperture stop, and the exit pupil.

If you have any examples of how the differing definitions can cause further difficulties, please don’t hesitate to let us know.