Do you mean diffraction or refraction?  If diffraction is a concern, can you provide more detail about your problem and/or concern?  Otherwise it’s difficult to make a recommendation.

Hi Charlotte,

Assuming you mean diffraction, OpticStudio cannot model diffraction in non-sequential mode. The reason for this is that it is a ray tracing application. Rays travel in straight lines and obstructions cast infinitely sharp shadows. The ‘bending’ around obstructions created by the wave nature of light is not modeled in ray tracing.

Ray tracing can, however, model interference. In systems well modeled by ray tracing and interference -- like interferometers -- OpticStudio can provide useful simulations.

Sequential mode does offer tools for modeling the wave nature of light, principally Physical Optics Propagation (POP). POP is complicated, computationally expensive, and limited. But it can be quite useful. There are several good articles on POP in the knowledge base, like here.

Hey Charlotte

I need to slightly disagree with David here. Although Non-sequential mode is purely ray-tracing and has no equivalent of POP, it does support Huygens diffraction so you can see a full diffraction PSF.

Check out the sample file Documents\Zemax\Samples\Non-sequential\Coherence Interference and Diffraction\Interference Example 4- Diffraction Limited Imaging.zos. This shows a Cooke triplet that was optimized to prove a diffraction limited spot on-axis. There are two detectors at the image plane. The first is set up as most Detector objects are, to look at the incoherent spot formed, which is an unphysical point:

The second detector is showing the coherent sum of the rays landing, and has its PSF_Wave# parameter (in the NSC editor) set to the ray’s wavelength, which (see docs) tells OS to do a Huygens integration of the rays. With Log scaling added, you see:

As a result, you can see the diffraction structure in image points that go below the Airy disk resolution.

• Mark

That’s really cool, Mark. 
Would it be possible in non sequential to model the diffraction spikes produced by the spider in a Newtonian telescope?

Yes. It’s just a coherent addition of plane waves on the detector. You should get the same results in seq and NS modes, as long as the NS mode has only the seq path. Of course, the NS PSF can be washed out by scattering, stray light etc...as it should be.

Wow, Mark!  Here’s an 8 inch f/8 Newtonian with a 4-vane spider. I didn’t model the secondary, just the spider as an obstruction in the beam.

I had no idea!

It is totally cool, isn’t it? And all done with rays...

Thank you so much Mark, this is really helpful!