Hi,

I am modeling an asymmertrical system for laser line excitation. The symmetry is broken by a Powell lens to create a top-hat illumination on the sample. The last element is a microscope objective (paraxial surface) after which the beam somewhat evenly irradiates the image plane along x-plane and is close to diffraction limit along the y-plane.

I began tolerancing the system, includes a zoom triplet, and cylindrical afocal beam expansion and etc. I hope to achive diffractin limit along the y-plane, to do so i set two elements as compensators. When i set y-spot for tolerance criterion it seems to be working, however if i set the merit fucntion, that is defined for RMS y-spot, the compensator moves way out of its working limits and gives better (but false) results.

Anybody can explain what is the difference and what part am i missing?

Thank you,

Zsombor

Part of the result:

Compensator: Thickness 8, Min = -1,0000, Max = 1,0000

Compensator: Thickness 22, Min = -0,5000, Max = 0,5000

WARNING: Boundary constraints on compensators will be ignored.

Criterion : User defined merit function

Mode : Sensitivities

Optimization Cycles : 5

Nominal Criterion : 0,00117646

…

Estimated Performance Changes based upon Root-Sum-Square method:

Nominal Merit Function : 0,00117646

Estimated change : 0,00686855

Estimated Merit Function : 0,00804501

Compensator Statistics:

Thickness Surf 8:

Nominal : 9,924658

Minimum : -2,262563

Maximum : 29,530714

Mean : 9,800469

Standard Deviation : 3,409756

Thickness Surf 22:

Nominal : 12,660230

Minimum : 10,268682

Maximum : 80,473917

Mean : 13,678104

Standard Deviation : 6,776885

End of Run.