.

Hi Laurent,

The nominal criterion is simply the Merit Function Value without any perturbations or compensations, but with a rebuilt MFE (in this case, RMS Spot Size).  The RMS Spot Size MFE will always be:

• Rings = sampling
• Arms = 2 * sampling
• Gaussian Quadrature
• Centroid reference

It’s hard to tell from the image, but it looks like the Spot Diagram is dithered and the reference is set to Chief ray.  You will need to use hexapolar and centroid and ensure the ray density matches the sampling.

To see how the nominal criterion is calculated, you can temporarily delete your Tolerance Data Editor and run a tolerance while saving 1 Monte Carlo file.  This file will have the rebuilt MFE which you can use to ensure the nominal value matches your spot diagram:

Thank you very much @MichaelH ! Indeed, when I choose the reference Centroid in the Spot Diagram instead of Chief Ray the RMS radius is very close (1 µm difference) from the nominal criterion.

The radius of the image spot is decreased by 15 µm RMS radius and 30 µm in GEO radius in the axis.

A last question: what is the real size of the image spot? This variation seems to me to be very large to be related to a calculation reference. I have read in Zemax manual that : “ Note the centroid reference is generally superior
because it is based upon the aggregate effect of all the rays that actually illuminate the image surface, and not
on the arbitrary selection of one ray which is "special" “.

Thank you in advance.

Hi Laurent,

The reason the manual states “the centroid reference is generally superior” is because the centroid reference (for all analysis) is the most accurate way to measure something like the spot diagram in a lab setting.  When your vendor/manufacturer tests the performance of the lens, they will always report the PV/rms for spot size or PSF with respect to the centroid.  So the “real” size of the spot according to any measurement device would be with a centroid reference.