Hi Kevin,

Thanks for posting on the MyZemax forums!

This depends on where you are looking to place this detector and how you are looking to use it. Non-sequential analysis features, non-sequential rays, and detectors cannot be used outside of the defined non-sequential component. However, detectors can be placed inside of the non-sequential component. One major caveat with this is that the detector will not model any rays traced from sequential mode. It can only view non-sequential rays that have been traced from a source placed in the non-sequential component. Another way to think about this is that you can use a non-sequential source/detector in mixed mode, but only inside the non-sequential component as diagnostic tools to help optimize only the non-sequential component. In summary, you cannot use a detector in mixed mode to measure sequential rays that enter the non-sequential component.

In mixed mode, you can open the 'Non-Sequential Component Editor' under 'Setup' to edit sources/detectors in the non-sequential component itself. Under 'Analyze', there is a button called 'NSC Raytracing' which is an easy way to access NSC features. Again, these features will only use rays traced from non-sequential sources and can be analyzed by detectors in the non-sequential component. These traced rays will not leave the exit port of the non-sequential component; they will remain 'inside' the NSC. The help file page, 'The Analyze Tab (sequential ui mode) > Applications Group (the analyze tab, sequential ui mode) > NSC RayTracing' shows what this button looks like.

Regarding your second question, the Bitmap optimization tool is only available in pure Non-sequential mode. In mixed mode, there is not a way to optimize your entire system based on an illumination image target metric. However, in pure non-sequential mode, you could optimize the non-sequential components in your system via the Bitmap Optimization with a properly placed source and detector. Once optimized to your liking, you could pull the optimized non-sequential components into your mixed mode system without the source and detector.  The sequential components can't be optimized to the same metric, but optimizing the non-sequential components in this fashion could help you get closer to your final system goals!

If you have any additional questions, please let us know!

Best,

Jordan Teich

Hi Jordan,

thanks for the answers, now I understand that any rays from SQ mode source can't be detected by detecors in NSC and there is no bitmap optimization metric in SQ mode.

Actually, all the questions I asked come from the consideration of  trying to use TreeFreeForm function in SQ mode to optimize lens structure to get a specific target illumination.

May I have any suggestions if I would like to use TreeFreeForm function in SQ mode to optimize a lens to get a specific target illumination in current substitution version?

Thanks

Kevin

Hi Kevin, 

Are you referring to the True FreeForm surface in the lens data editor? I'm assuming that you are looking to optimize different propreties of this freeform surface to meet your illumination target of interest.

What kind of target illumination pattern are you looking to optimize too? And what kind of source are you using in your setup? Generally, non-sequential mode is more suited to tasks involving illumination analysis (including optimizing to an illumination target via bitmap optimization).  If you let me know what illumination pattern you are looking to optimize to, I'll be able to better answer your question! 

In the MyZemax knowledgebase, I was able to find an article that details how to transform a Gaussian illumination pattern to a top hat profile in sequential mode.  I've linked this article below.  However, this is a limited application and only involves using one field point in the system.  I thought you would at least be interested in taking a look at this article to help get you started in terms of thinking about what is possible in sequential mode.

How to design a Gaussian to Top Hat beam shaper · MyZemax

Looking forward to hearing back from you soon! 

Best, 

Jordan Teich

Hi Jordan,

I mentioned the surface as Tree Free Form 

the target image is an irregular shape and there is a up down symetrry about x axis.

thanks

Kevin

Hi Kevin,

Thanks for your post here on the forums. I just wanted to touch base here as I noticed that you had a case not too long ago on this topic as well. Did you have any outstanding questions on this topic?

In short, as Jordan said, I think the best route to using the TrueFreeForm surface to get the distribution you want will come down to the Merit Function Editor definition. For TrueFreeForm, the strength of that surface is that you have different surface definitions to expose as variables -- Zernike Standard Sag, Even Asphere, and discrete point-by-point control, to name a few. So, while you can set different parameters as variables, what it will come down to is constructing a merit function that targets your desired distribution. Since Bitmap Optimization is not supported in Sequential Mode, you will need to construct a function using operands like REAY/REAX to target landing coordinates of specific rays in your pupil at certain locations. This is what the article that Jordan shared demonstrates. If you have an idea about the shape you're looking for, a good place to start might be thinking about how rays that fill the pupil need to land on the image plane to generate your desired shape. This is really the most direct way to get a 'pattern' target in Sequential Mode. Depending on the pattern you want to achieve, your merit function may need to become quite granular in the targets:

Thanks again for your post!

~ Angel