Conclusions

The advantages of the curved focal plane array (CFPA) have been evaluated in this report by comparing the optical properties of three lens systems. The results state that CFPA has better MTF, field curvature, relative illumination, point spread function. Still inconclusive is the field distortion, which requires more software research. The results of the analysis also show that there needs to be modifications to both Zemax and ISET. Both optical programs are designed for planar FPA and were not designed with the possiblity of a curved FPA. These problems also appeared during the field distortion analysis. As a last addition this project showed that there is a wide tolerance for curvature misalignment while still taking advantage of the curvature properties. All these results demonstrate that there is sufficient advantages to using a CFPA that there should be more emphasis on this are of research.

Better: MTF, field curvature, relative illumination, point spread function, geometric robustness

Inconclusive: field distortion

Future Works

The scope of this project has been focused on the confirmation of previous simulations and analysis. There is a great deal of promise shown in these results but the project is still at the very begining of the process. There are several areas that would be interesting as well as important to analyze for future progress of CFPA

Field distortion in Zemax

Field distortion in Zemax was not analyzed. Futher investigations are needed into how Zemax actually differs from CodeV in field distortion analysis. And the raytrace techniques in general. While Zemax is inexpensive compared to a CodeV license, is the difference in quality worth it?

Is Field Distortion a valid descriptor for CFPA?

One instresting point is that field distortion is an accurate discriptor for a planar FPA because this is the actual field height that the sensors use. While on a CFPA the field distortion height is actually a translated height, and not that of the sensors. The sensors will already be going through a series of transormations to reach the "field height" and so is it a fair comparison?

Bugs in ISET

Continual changes in ISET are required if it is to be used as an accurate analyzer for the CFPA. There has been no serious simulations done in the system before for CPFA and some reconsideration of CFPA parameters should be researched.

FDTD – cross-talk, efficiency, resolution

Electromagnetic simulations in FDTD would be interesting. Since the incident rays on the CFPA should always be normal to the surface, it would be expected that the CFPA has minimal cross-talk, as opposed to the planar CFPA. But before FDTD is used for simulations, an acceptable CFPA must be designed. While NASA was able to merely curve the surface of the lens still utilizing a single wafer substrate, Rim e al proposes the use of a curved substrate. The use of a curved substrate has with it many implicit manufacturing and assembly issues. For example, how will the micro lenses and color filters be adapted to the curved surface? These questions need to be answered before a EM FDTD analysis is useful.

Geometric resolution limit for array technique

Rim et al proposed inexpensive fabrication technique requires the stretching of silicon arrays. Because these arrays are stretched it would be interesting to see what the "real" resolution of the system ends up being after transformation from the CFPA to a digital image. Will there be higher density in the peripheral? Will there be a scarcity in the center?