Laser fiber imaged onto QSI camera through foreoptics system
The FWHM for this cut is about two or three of the sensor's 5.4µm pixels. Through the magic of ImageJ we have a 3D version of it:
The through-focus image is not a simple, round, fuzz-out as we might like...it has a lot of busy structure that might indicate a recognizable misalignment with reference to Zemax models:
Series of frames taken with camera moving through focus,
3mm change in focus per frame.
Clearly there is a lot of structure in the near-focus images, but the best-focus image looks pretty sharp....not sure how this would translate to the infrared.
For completeness, here is a best-focus image for the image after the hyperboloid, without the window in the system:
Laser fiber imaged onto QSI camera through ellipsoid + hyperboloid system
The FWHM in the post-hyperboloid image is about 9 pixels....is this the right scale factor based on the magnification of the window?? Without the window it's not nearly as tight, but still about 50µm and roundish. It's not the cleanest image, but we can try tweaking the elements next week to see what we can learn. I have not been able as yet to intentionally misalign any of the elements to see what happens.
For the complete system, the best-focus image in visible light is obtained with the sensor about 105mm behind the back (plano) surface of the window...I would have thought it should be father away, and maybe it is in the IR...I haven't looked at the index yet.
2 comments:
I have to say, that's a pretty darn good spot size. In the infrared, I would think it should be diffraction limited.
The focal length shouldn't change that much in the infrared. The KBr could be causing us a problem.
Are you sure the laser isn't saturating the detector when it is tightly focused?
Dana and I re-checked the measurement between the lens/window and the camera sensor today, and I had made a mistake. The actual number is very close to what Dana's model predicts (~120mm). My bad.
Post a Comment