Hi Zhenfeng,

The reason you're seeing this message appear is because the analysis depends on the assumption that your model is rotationally symmetric. The following is from the Help Files at 'The Analyze Tab (sequential ui mode) > Image Quality Group > Aberrations (Image Quality Group) > Seidel Coefficients':

I haven't tested this out, but I think generally, any User Defined Surface will likely display this messge, since OpticStudio doesn't know if the DLL used to generate the sag profile for the custom surface will be rotationally symmetric. In any case, if you're using a physical array of lenses with the US_ARRAYEVEN DLL, then the resulting paraxial ray trace to compute the Seidel coefficients is not likely to be accurate.

Would you be able to use something like a Zernike Standard/Fringe Coefficients analysis to get at the data you're looking for?


 

These analyses will look at real ray trace data to generate a Wavefront Map, and subsequently fit Zernike coefficients (using the convention selected in the menu shown above) to that data array.

Let us know if this helps clarify things and gives you a path forward!

~ Angel

Only the central element of an array of lenses would have a paraxial equivalent and hence give meaningful Seidel coefficients. Think of the difference between the ray trace of a lens array and of a single lens with the same power...they're entirely different.

The lens array breaks the incoming beam into multiple separate beams, so most wavefront calculations will fail, as they assume all rays share a common exit pupil. The geoemetric calculations like spot size, Geometric Image Analysis and so on will work: but don't try the Ray Fan, OPD plot, Wavefront map etc.

If you really need wavefront analysis of a system using Lenslet arrays, try this KB article : https://support.zemax.com/hc/en-us/articles/1500005575542-How-to-use-POP-with-lenslet-arrays

- Mark