Is there a built-in method to subtract defocus from wavefront map ?

  • 8 October 2019
  • 2 replies
  • 350 views

Is there any build-in option to remove the defocus share from the wavefront error in the wavefront map analysis ? I am thinking of something similar to the checkbox 'remove tilt'.

2 replies

Userlevel 4
Badge +3
Hi Christian, there is no built-in option to remove defocus from the wavefront map. However, we might have a trick that can help you achieve what you're looking for. If you place a dummy surface co-located with your image plane (place your cursor on the image surface and press insert), then look at that dummy surface in the Wavefront Map, this should remove defocus. There's more information about this in the KBA "Why is the Optical Performance Sometimes Different at the Image Surface Versus a Co-Located Surface?" and in the help file excerpt below. Let me know if this performs as you were hoping!





Help file excerpt

Optical Path Difference - Evaluating results at intermediate surfaces


OpticStudio can compute analysis results for surfaces other than the image surface by applying assumptions. These assumptions work in most cases, however, there may be cases where the methods described here for analyzing intermediate surfaces are not appropriate, including systems which require ray aiming. All of the changes made to the lens described are made on a copy of the original lens data made for analysis purposes so no changes are made to the original lens data.


If the field type is either real or paraxial image height, the field type is changed to angle or object height for infinite or finite conjugate systems, respectively. The angles and heights used correspond to the primary wavelength chief ray angles and heights as computed for the unaltered system.


If the selected surface precedes the stop surface, OpticStudio moves the stop surface to a (possibly virtual) dummy space prior to the existing surface 1. Unless the system aperture is object space numerical aperture or cone angle, the system aperture is changed to entrance pupil diameter, and the aperture value is set equal to the original paraxial entrance pupil diameter computed for the original stop position. Note this assumption might not be valid for systems that require ray aiming. If the selected surface follows the stop surface, no changes are made to the system aperture or stop definitions.


Surfaces which follow the selected surface are then deleted. The glass of the new image surface is set to be the same as the selected surface.


Most analysis features that compute results for focal mode systems make more sense if the rays are allowed to come to a focus after refraction from the desired surface. For example, the OPD plot is generally a useful diagnostic only when the OPD is measured at the (possibly virtual) focus of that surface. Other features which require a temporary image to be formed include PSF, MTF, and diffraction encircled energy. For these features requiring a temporary image, the new image surface is set to be a standard plane surface. A paraxial marginal ray height solve is placed on the selected surface thickness to place the image surface at paraxial focus for the selected surface. The analysis computation then proceeds at this newly created intermediate image surface. Note the analysis occurs at the paraxial focus formed by the rays after refracting through the selected surface. This shift to paraxial focus is not performed if the intermediate system is in afocal mode.


Hello Alissa, i found that your suggestion removes some of the defocus (Zernike Coefficient Z4) from the wavefront map, but I can not specify the pupil zone for the marginal ray height solve mentioned in the help file excerpt, it is fixed to 0 (paraxial).



Even though this will move the image surface, i now use a (real) marginal ray height solve on the last surface with pupil zone = 0.7, As suggested in:



Help file excerpt

Global Optimisation - Suggestions for Using Global Optimizers



3) Use a marginal ray height solve (for zero ray height) on the last thickness before the image surface. Most lenses are well corrected for the 0.7 pupil zone on axis. You may want to use another pupil zone if your intuition guides you to. This solve will ensure that every design generated by Global Optimization is in focus, just as the curvature solve ensures the correct focal length. These two solves together can improve the performance of the Global Optimization algorithm by several orders of magnitude. For Hammer optimization, replace the solve with a variable to allow for optimal defocus.



This achieves less remaining defocus (Z4) in the wavefront than your suggested solution, but of course moves the image surface.

As i will introduce element decenters and tilts, i try using a seperate merit function with solely the thickness of the last surface variable to decrease Z4 across the field.



Thank you for your help !

Reply