Lens design

  • 10 May 2021
  • 1 reply
  • 176 views

Hi,


newbie question here, and I realise it is possibly not an Optics studio question, strictly speaking.


I am working with a nonsequential model for an optical system in which I have a meniscus lens (ZnSe) collimating an MIR laser. I am in fact not really collimating it but focussing it ~20cm away from the lens and trying to minimise the spot size on that plane.


That works well, but now I am trying to design a new meniscus lens (current design was of a former colleague) that would have shorter working distance and try to achieve a smaller spot radius on my detector plane. I am using the merit function on the lens and universal plot to try and find the lens-laser separation to minimise the spot size. I am manually varying the two meniscus radii and the central thickness to try and achieve smaller spot size but getting nowhere. Every time the spot RMS on the merit function finds a minimum within the lens Z-position range I explore, but this ends up always larger than what I had originally.


Does anyone have a lead of any kind for me? whether it be exisitng content in the forum, more effective use of the merit function, lens design criteria resources? anything would be appreciated, thank you.


Regards


Giuseppe


1 reply

Userlevel 6
Badge +4

Hi Giuseppe,

 

It is possible this task is more complicated than it first appears.

To begin with, I suggest that this should be done in sequential mode, not nonsequential. There are several reasons for this. In general, optimizations for image quality are best done in sequential because the optimization methods are better at choosing rays and are more stable and robust. It is also easier to specify constraints. Also, in nonsequential, the calculation of the spot radius requires that the spot cover a reasonable group of pixels, and the method cannot determine the radius of a spot smaller than a pixel. This creates difficulties when a very small spot is anticipated.

And there is another very significant reason to use sequential. In most laser designs, the goal is to achieve diffraction limited performance. In the focusing case, the geometric spot radius determined by ray tracing will be significantly smaller than the actual spot radius achieved, which is limited by diffraction effects. Sequential mode has an analysis method called Physical Optics Propagation (POP) which simulates diffraction and can be used to optimize for the diffraction limited spot.

The general method in sequential mode is to first optimize the design for minimum spot radius using ray tracing, and then refine the design by setting up a POP analysis and the using the POPD operand instead of ray tracing to optimize for minimum spot radius. (This can resut in a design a slightly different from the one that minimized the geometric radius.)

Using POP is complicated and requires study. There are several good resources on the Zemax site. I especially recommend all three parts of “Using Physical Optics Propagation (POP).”

Kind regards,

David

https://support.zemax.com/hc/en-us/articles/1500005488601-Using-Physical-Optics-Propagation-POP-Part-1-Inspecting-the-beams

https://support.zemax.com/hc/en-us/articles/1500005490261-How-to-model-laser-beam-propagation-in-OpticStudio-Part-1-Gaussian-beam-theory-and-ray-based-approach

https://support.zemax.com/hc/en-us/articles/1500005490541-Simulating-lasers-webinar

https://support.zemax.com/hc/en-us/articles/1500005489641-Laser-applications-webinar

https://support.zemax.com/hc/en-us/articles/1500005488421-Exploring-Physical-Optics-Propagation-in-OpticStudio

Reply