I’m working on a system in sequential mode where a beam is launched from a multimode fiber (0.12 NA) goes through some lenses, and is focused on an image plane (see my other post on the topic). I had previously asked about how to evaluate the size of the focus after the beam. The suggestion was to put a field point originating from the edge of the fiber (in addition to one launching from the middle of the fiber) and to look at where it lands after the lens - that worked well for that purpose.
Now I would like to know how to evaluate the spot size at a location where the beam is not focused. Using the field height doesn’t work since the rays from all the fields are spread out and partially overlapped. So how would you do this? Is it just a matter of looking at the locations of the marginal rays or is there a better way?
Thanks!
Page 1 / 1
Hi Ryan,
In my opinion, this is best done in non-sequential. However, geometric image analysis in sequential mode will do a pretty good job. In the attached file, the fields are set up for a 200um diameter fiber in object height coordinates. The aperture is NA 0.12 in object space. The lens is a paraxial lens with the focal length optimized for spot size on the image plane. Geometric image analysis is performed using a disk 200um diameter as the source image, and the result is evaluated at the image plane and at an intermediate surface.
Kind regards,
David
One thing slipped my mind: In the example file I should have checked Telecentric Object Space under aperture so that the emissions from the fiber are directed axially. It makes little difference in this case, but could be significant in some designs.
Thanks David! This is really helpful.
I ran geometric image analysis and, in the defocused region, I got a spot diameter of a few millimeters. This seems to be consistent with the spot diagrams, as rays from a point were spread out over a similar diameter (and since the object is much smaller than the defocused spot, it should approximate to a point source fairly well).
I did have a question however: I tried to do a parallel simulation using physical optics propagation with a top hat of the same width as the disk image and the results were completely different. Instead of a spot diameter spread over a few millimeters, I got something like an Airy disc with a much smaller total diameter (~7X smaller). Maybe there is something about one of these analysis methods that I’m missing. Do you have some thought on the reason for the discrepancy?
Also, I noticed that if I turn telecentric object space on and then off, it doesn’t seem to revert unless I undo. Do you know if checking that option changes some other parameter?
Thanks again!
Hi Ryan,
I don’t have any problem moving back and forth to telecentric space. In my file, I judge that by the automatic clear aperture of the paraxial lens, which switches back and forth, and also by the GIA results.
Regarding the comparison with POP, you are really simulating two different sources. In the GIA ray tracing case there is an extended source, where each point on the extended sources emits over the 0.12 NA aperture, with no phase coherence. In the POP case, the top hat beam has uniform intensity and phase over the region. Since phase is constant, the wave front is perpendicular to the axis. A ray tracing analog to that would be a source that emits a collimated beam. In that case, the beam could be focused to a point. In the POP case you get an Airy disk due to diffraction. But in the extended source case, you get an image of the extended source, with a diameter determined by the system magnification.
