Optical Design Software

 (Note: This article is adapted from the chapter "Optical Design Software" by Douglas C. Sinclair in the Optical Society of America's "Handbook of Optics, Second Edition ", Mc-Graw-Hill, ISBN: 0-07-047740-X. The section "Whither Optical Design" was published in Optics and Photonics News, June 2000.)

The first practical computer software for optical design was developed in the 1950's and 1960's.[1][2][3][4] Several commercially available programs were introduced during the 1970's, and development of these programs has continued through the 1980's to the present time. Despite the fact that there has now been more than thirty year's development of optical design software, there are still substantial improvements to be made in optimization algorithms, evaluation methods, and user interfaces.

This article attempts to describe a typical optical design program. It is intended for readers that have a general background in optics, but who are not familiar with the capabilities of optical design software. We present a brief description of some of the most important mathematical concepts, but make no attempt to give a detailed development. We hope that this approach will give readers enough understanding to know whether an optical design program will be a useful tool for their own work.

Of course, many different programs are available, each with its own advantages and disadvantages. Our purpose is not to review or explain specific programs, but to concentrate on the basic capabilities. Some programs work better than others, but we make no quality judgment. In fact, we avoid reference, either explicit or implicit, to any particular program.

The figure is a flow chart of a typical optical design project. Usually, the designer not only must enter the starting design and initial optimization data, but also must continually monitor the progress of the computer, modifying either the lens data or the optimization data as the design progresses to achieve the best solution. Even when the performance requirements are tightly specified, it is often necessary to change the error function during the design process. This occurs when the error function does not correlate with the desired performance sufficiently well, either because it is ill-conceived, or because the designer has purposefully chosen a simple error function to achieve improved speed.

The fact that there are alternate choices of action to be taken when the design is not good enough has led to two schools of thought concerning the design of an optical design program. The first school tries to make the interface between the designer and the program as smooth as possible, emphasizing the interactive side of the process. The second school tries to make the error function comprehensive, and the iteration procedure powerful, minimizing the need for the designer to intervene.

This article is divided into four technical sections. Data Entry deals with issues that concern the representation of lens data in a computer. Evaluation describes the principal techniques used to compute the performance of an optical system. Optimization concerns the techniques commonly used to improve performance. Software considerations discusses programming issues.

A final section, Whither Optical Design, comments on the general state of the art in the field of optical design.