The user interface for optical design software is explored. Interactive methods, menus, and the use of graphics have evolved. Friendlier future systems may use interactive graphics and artificial intelligence techniques.

We are developing a lens design program intended to operate without user intervention, and to improve its performance with repeated usage. The methodology and current status will be discussed in this paper.

Analytical intelligent optical design program and its techniques include solving problems in analytic and symbolic form, interactive communication with microcomputer by definealanguage, high speed input by analog data scanning and lens spacing, bending, lens classification, search for the lens design methods, learning by doing and learning by being told, integral with real world lens design system, etc. are all discussed.

Higher-order aberrations are due to intrinsic and induced components at each optical surface. This paper shows that the induced components are mainly due to pupil aberration, and are generally much more important than the intrinsic components.

The combination of an optical ray-tracing program with a solid modeling C.A.D. (computer-aided-design) system creates a very flexible tool for optical system analysis and evaluation. The program uses the CAD data-structure and user-friendly menus for creation, manipulation and visualization of the optical system. Furthermore, it is capable of dealing with problems which are impossible or difficult to handle by existing optical design programs, such as calculations of three-dimensional sensitivities, multiple reflections, multiple-surface apertures, specular stray radiation, image rotation and complex-prism design. It can also be used as an efficient tool for error-budget and error-analysis, and can be fully interfaced with a finite-elements analysis program, thus enabling the evaluation of the effects of mechanical or thermal loads on the optical performance.

Optical coating parameters are introduced in the lens design phase. Using ray tracing, the range of angles of incidence have been determined. Optimized coatings were defined. The influence of those optimized coatings on the performance of an optical system is discussed. The linkage between lens and coating design is demonstrated on a wide field night vision objective by a comparison between the old and this new method of design.

A new method of automtic lens design is described, representing an improved version of the Suzuki and Yonezawa method using simultaneous linear inequalities. The new method does away with the need for setting parameters that are arbitrary, a requirement associated with the conventional methods. The results obtained with the use of the program based on this new method in actual lens design work have shown that the convergence speed is significantly faster as compared with previous methods.

Merit functions used in optical design are almost always extremely nonlinear in their behavior with respect to optical system parameters. This prevents the existing optimization algorithms from necessarily converging on the global minimum. They converge to the nearest local minimum which may, or may not, be the global mini- mum. It is possible that the global minimum can be achieved by successively minimizing a sequence of increasingly nonlinear merit "subfunctions" that converge to the complete merit function. The complete merit function is constructed in the usual way: Ф = X23 + X25 + X27+ . . . . . (1) However, each aberration Xi is composed of a special linear combination of aberration coefficients. The sub-script i refers to the lowest order of aberration coefficient present in the aberration Xi. The complete merit function can be either mean-square spot size or wavefront variance by a proper choice of the linear combinations of aberration coefficients. Starting with the highest order aberration coefficients considered significant to the design problem (say, order nine), the sequence of merit subfunctions is constructed as Ф = Ф3 = X23 + X25 + X27 + X29 Ф5 = X25 + X27 + X29 Ф7 = X27 + X29 Ф9 = X29. (2) Note that Ф9 behaves much more simply with respect to changes in the parameters of the optical system than does Ф. Also, Ф9 must be small at the global minimum of Ф so that minimizing (09 may result in the optical system being in the region of the global minimum of Ф. By minimizing sequentially Ф9, Ф7, etc., the optical system may be sequentially moved to the global minimum of the complete merit function Ф.

The Brixner lens optimization procedure, in which a lens design moves steadily toward diffraction-limited performance, samples lens performance with bundles of precisely traced skew rays, analyzes performance by calculating the image-spot sizes and positions, and optimizes performance in a least squares system that minimizes the lateral ray deviations from their ideal image points. Minimizing the rms image-spot size minimizes the rms optical path differences (OPDs). Minimizing the rms OPDs also optimizes the diffraction modulation transfer function (DMTF). Minimizing the image-spot size and position errors also minimizes and balances the Seidel and higher-order aberrations.

A new method of selecting glass combinations for the correction of paraxial chromatic aberration in optical systems has been developed. This new method corrects axial color for at least three wavelengths using two different types of glass and at least four wavelengths using three different types of glass. Certain combinations can be found that are corrected for as many as five wavelengths, using either two or three different types of glass. The Abbe number and relative partial dispersions are not utilized in this new approach; instead, the dispersion of the glass is characterized using Buchdahl's chromatic coordinate. The Buchdahl model and the mathematical methods employed are described briefly. Thick air-spaced doublet and triplet lenses are presented as examples of designs that demonstrate color correction at three, four, and five wavelengths and are corrected for spherical aberration and spherochromatism as well. A more complex lens design is illustrated by a Petzval objective with a 24-in. focal length, a focal ratio of f/3.5, and a 12-deg field, which is corrected for both focal shift and spherochromatism and is diffraction-limited over the spectral region from 0.44 to 0.9 μm.

A simple aberration theory is described for systems containing tilted and decentered surfaces. Design principles are developed and applied to two and three mirror systems.

Polarization aberrations are variations of intensity and polarization of an optical beam in the exit pupil of an optical system, and the dependence of these variations on the position of the object in the field. A set of functions has been derived to describe the low-order polarization aberrations of a symmetric optical system. A method is given for calculating polarization aberration coefficients for a refracting system from a paraxial raytrace. A model of these aberrations as weak polarizers that vary across the pupil will be discussed.

The design of optical systems can be aided by general guidelines to the ways in which the most highly-corrected systems are assembled. We suggest that the designer can employ some of the more effective methods of controlling aberrations in the form of sub-systems which are internally corrected. Present knowledge of this is documented in only a fragmentary way. In this work we attempt to identify the useful principles. We list six general approaches to the reduction of aberrations. Two of these are described as "avoiding" and "negating" aberrations and these are the ones which are selected for further study. They require the knowledge or discovery of particular systems, and these often lead to the best results. Avoiding aberrations requires certain specific types of component and surface which we call the "special cases" of aberration correction. We discuss why they are useful. Negating or cancelling aberrations implies the use of two or more separated elements, where an aberration is large in an intermediate space but can sum effectively to zero. The systems in which good cancellation is achieved are further examples of the special cases: they may themselves be used as parts of more complicated systems. We give examples from our own work in which such methods are employed.

Spherical transversal aberrations at the best focus is essential to determine the third order aberration coefficient of coma and astigmatismus. We discuss above, including distortion.

The background, theory and application of aberration fields in nonsymmetric systems is discussed. The usefulness of this concept, in the context of a new aberration display, is highlighted with an example.

Optical designers have used the aspheric expansion given in equation 1) for many years. z= CQ2 + 44 + eQ6 + fQ8 + g Q10 1) 1 +0 â€" (k + 1)C2Q2 This surface is rotationally symmetric about the optical axis which runs along the z direction in cylindrical coordinates. The surface sagitta, "sag", is given by z and i)=N/xa + ya is the radial distance from the optical axis.

Spline surfaces can be formed at will. They are always twice continuously dif-ferentiable, i. e. they have no jumps, bends or ripples. In the automobile industry they are used in the design of car bodies for the minimization of air resistance. In the field of optics splines are employed for the spectacle lens, Gradal HS. The form of the spline surface is defined by optical requirements. Splines have contributed to the successful reduction to a minimum of the unavoidable aberrations present in progressive addition lenses and to the binocular equilibrium of the remaining aberrations. Conventional surface structures will be used as a comparison.

A new approach describing the third-order aberration fields of simple tilted and decentered optical systems (plane-parallel plate, wedge, mirror, including aspherics, and thin lens) is presented.

An "unusual" optical surface is the plane symmetric equivalent of the general rotationally symmetric aspheric familiar to optical designers. This surface was introduced to facilitate the design of fast, unobstructed aperture, large field-of-view, reflecting telescopes, and has resulted in a number of successful designs of such systems. The fabrication of these systems is inhibited primarily by the difficulty of testing such surfaces, so in this paper we describe an appropriate test procedure. This is an interferometric method that makes use of a hybrid setup in the test arm where the geometry, test optics, and a computer-generated hologram are all used to attain zero aberration.

An optical design program for the Apple Macintosh microcomputer is described. The program features ease of data entry and graphic display of results.

In order to take full advantage of CCD detector arrays for high dispersion astronomical spectroscopy, an echelle spectrograph is being designed and built for the Palomar sixty inch reflecting telescope. Critical to the performance of the spectrograph is the design of a seven element lens for the spectrograph camera, which has proved capable of 20 micron spot diameters over the CCD field and for wavelengths ranging from 320 nm to 700 nm.

A ray-matrix method for the analysis of resonators with image rotation is derived and used to determine alignment sensitivities and other properties.

The design of a three-mirror all-reflective, flat-field objective with a field-of-view of 1° by 6° is presented. The objective was intended for use in an orbiting push-broom linear-array scanner for spaceborne land mapping. The objective is telecentric and employs the ring-field principle.

In order to develope the three mirror telescope, a formula group of third order aberrations and a program had been developed. Two families of the aplanat with zero astigmatism or zero distortion were found. One of them is a new result. Based on the family 2, a 10 me-ter diameter telescope with a spherical primary mirror had been designed. The primary's F-number is 7/1.5 and the full field angle of view is 20 arc minutes. The RMS spot diameter (5 colours) is less than 0.313 arc second.

An interactive computer lens design program has been developed. It has capabilities for editing lens data, optimizing zoom lens, evaluating image qualities, etc.. A Tessar lens and an IR zoom telescope designed by using this program are discussed.

This paper is concerned with the development of lens design software for personal computers, and specifically relates to the programs developed by Kidger Optics over the period from 1980, (the Mills College conference) until mid-1985. Over this period we have seen enormous advances in the development of personal computers, and consequent developments in optical design programs. This paper discusses both the hardware developments and also the software advances that have been made possible.

Exit pupil correction of the Large Deployable Reflector's (a proposed IR to sub-mm space telescope) segmented primary can be done by reimaging it onto a like segmented surface at the exit pupil. This allows the primary to be more flexible, the adaptive element to be smaller, and the supporting structure to be cheaper than if all correction were performed at a stiffly supported primary. We consider piston, tilt, and decenter errors of an annulus of the primary and show the equations for required corrections. To verify these, we simulate the perturbations with spline functions in the lens design program ACCOSV. Strehl ratios used to measure image quality show that a piston error of 1 mm is fully corrected over a 5 arcmin field for an f/10 system with a 0.7 n.a. primary at 30 micrometers. Limits of correction are also shown for tilt and decenter errors of segments. Tolerances are given for tilt and decenter errors of the reimaging optics also.

Several refinements to the PSD optimization algorithm have yielded improved convergence and enhanced ability to cope with unusual nonlinearities encountered in the course of optimization.

In Japan, plastic aspheric lenses have been greatly increased on two fields mainly since autumn of 1981. There is one method of aspheric lens design by spherical geometry.

A multi-image plane merit function is used for improved aberration control and depth of focus in automatic lens design. The effectiveness of the optimization technique is demonstrated with two examples.

Optical surface deformations are spline fitted using several algorithms including that from the ACCOS V optical analysis program (Scientific Calculations, Inc., Fishers, New York). The effects of spline fitting errors are investigated. The effect on image quality of deformation of the primary mirror in the Solar Optical Telescope (SOT) must be investigated. NASA's NASTRAN finite element thermal-mechanical structural analysis program is used to simulate the deformations due to various types of thermal and mechanical loading. Several simulated NASTRAN outputs are used to generate a deformed surface for spline fitting. The ACCOS V spline fitting algorithm is compared to other spline fitting alogrithms. The effects on the optical analysis of spline fitting errors are presented.

Criteria for color correction derived from the dispersion formulas of Hartmann, Cauchy, Schmidt, Conrady, and Buchdahl will be discussed, and their application to the design of apochromatic optical systems will be presented.

The use of the zero power corrector concept has been extended to the design of microscope objectives. Several four and five-element designs are described which include a flat field 10x design of 0.25 numerical aperture and a 40x design of 0.65 numerical aperture.

In this article a systematic attempt has been made to design diffraction limited space camera lenses with minimum number of elements. In this attempt, instead of starting from complicated n-element/component configuration, simple air-spaced 'Cooke triplet' is choosen as starting system and is then expanded to its derivatives, using the concept of fictitious glass constants, until optimized construction with desired specifications is achieved. It has been found that air-spaced pentac, with alternate positive and negative elements in it opened up a new region with low powers, promising the corresponding optimized pentac and its derivatives to cover wide specifications. Also, it has been proved that simultaneous doubling of the focal length of the optimized pentac , sextuplet and half reduction in the spectral-band width coverage have practically no effect on image quality. Thus in camera lenses for each resources survey, pentac and sextuplet con-figurations are optimized for low,medium and high resolution on the same lines Sky-lab lens design was worked out to cover various bands by changing air spaces in the basic design.

A simple method of correcting the cylindrically curved field caused by the motion of the scan mirror in a post-objective optical scanner without introducing bow distortion is presented.

For at least the last 12 years, half lenses have been used in small copiers. The half lens is desirable because of its compactness and because it folds the optical path and shortens the overall track. A half lens is less costly than a normal lens since it contains only half the necessary components.

A coherent fibre-optic Cartesian oval 'lens' is described which is used to convert linear positional to angular information, thereby increasing the look angle of a small quadrant detector.

Design, fabrication, and test results are discussed for a compact two-mirror collimator that was diamond turned from a single piece of material.

The Nova and Novette lasers' focusing lenses, with added cylinder elements, produce line foci for irradiating x-ray laser targets. Design, implementation, and results are discussed. An analysis of the focusing properties of a symmetric lens and cylinder lens combination is presented. It is shown that spherical aberration affects the line width and the intensity distribution in the line focus. In addition, expressions are derived which show that the line foci lie on curved surfaces.

Strict apianatism is required in optical disk systems. Biaspheric design yields superior single aplanatic lenses, and can tolerate decentering.

Aspheric lenses of conventional optical glasses have been molded. A 16 mm diameter photographic lens with aspheric figure error of 0.6)1m, accuracy error of 0.04 μm, decenter of 75" and surface roughness of 200 Å Rmax has been obtained. For an aspheric pickup lens in a compact disc system, a wave-front aberration within 0.04 μm rms has been obtained.

We reported the diffracton-limited plastic aspheric singlet for CD(Compact Disc) use last year,1,2 and it is now under mass-production. We investigated the off-axis performance of bi-aspheric singlets as well as mono-aspheric singlets, and achievable size of image field from a point of view of optical design. In this study, the specifications of singlets are f=4.5mm, 4mm, NA 0.45 and NA 0.6. Optical performance is evaluated by the value of max-imum image field defined by the area of which the rms wave front error is less than 0.07X (Marechal Criterion). The performance of the design results of bi-aspheric singlets and mono-aspheric singlets under several conditions are investigated. From these results, we concluded that bi-aspheric singlets are most suitable for not only CD but also general optical disc use maintaining enough back focal length and image field.

A radial gradient index aplanatic singlet pickup objective lens for the compact disk (CD) system has been designed, manufactured, and evaluated. We have realized a piano-convex aplanatic singlet by optimizing the refractive index coefficients of the GRIN medium,lens thickness and one curvature. The GRIN singlet was mounted in a CD player and tests confirmed that the lens meets the system requirements.

A comprehensive review is given, of the lens options available to the designer, for coupling semi-conductor lasers to single mode fiber. Various arrangements are considered, including microlenses formed directly on tapered fibers, and designs based on one, or more discrete lenses. The optical and mechanical advantages of the different approaches are compared. The highest coupling efficiencies are shown to be obtained, when high index materials are used in the first lens. A new design, based on a Si lens, in a confocal arrangement, is presented. A coupling efficiency of 70% is reported. Different means of bonding the critical alignments, including cementing, soldering, and laser welding are discussed.

A. 2-D arrayed planar microlens with a diameter of 180 μm was monolithically integrated by an ion exchange technique. The focal length of the lens was 0.45 mm and N.A. was as large as 0.2. Distance between each lens is about 190 pm in the longitudinal direction and 300 μm in the vertical direction. Aberration of the lens can be controlled by applied voltage at the electromigration.

Design problems associated with the design of Luneburg-type waveguide lenses fabricated by a waveguide index of refraction gradient are investigated. In particular, the control of polarization, modal, and chromatic aberration and focal length control are discussed. A design example of a lens to be fabricated by non-resist electron beam lithography which is corrected for polarziation aberration is presented.

Optics for laser diode based communication must be designed with the laser source in mind. Heterodyne detection is possible if the optical design allows efficient mode matching between the signal and local oscillator field.

This informal evening lecture covered briefly the careers and accomplishments of some of the more interesting lens designers and opticians of the past. Starting in the 1600s with the early astronomers and microscope makers, this was followed by the makers of photographic lenses after the introduction of photography in 1839. Numerous portraits of early workers were shown. The names of people specifically covered in this talk are listed below. Some other names were mentioned without comment; these are not included here.

It has become traditional[1,2] at these International Lens Design Conferences to pose a problem for designers to tackle, on their own before the conference and during the conference using available computer facilities. The problems provide amusement and challenge to the participants, but in addition, discussion and comparison of the designs submitted can teach us about different approaches to the design process. In the recent past, progress has been made in several areas affecting design. New systematic approaches to the design process are being proposed. Technolo(xies such as the use of aspherics and gradient index materials are now available to the designer, requiring new approaches to optical system design. For the problem this time. the participant.: were asked to make all decisions about design form. They were asked how the designs were initiated and what factors were important to the design form decision. A first-order problem was posed with no suggested starting solution and no performance requirement. It was intended to be a problem with no well known "best" solution -- something sufficiently out of the ordinary to pique interest and encourage original thought. The participants were asked to design a finite conjugate lens with pupil-image symmetry: a reversible lens. This paper is intended to introduce the problem and, of course, to present the solutions submitted before the conference. Forty solutions were received from twenty-eight designers from eight countries, representing a wide variety of approaches to the problem. Two companion papers can be found elsewhere in this Proceedings: Walther[3] discusses the fundamental performance limitations of symmetric reversible systems and Hopkins presents his personal approach to the problem as a designer with many years of experience.

The limitations on the performance of a reversible lens are analysed by mock ray tracing. To al the design of reversible lenses is a theme marred by false notes. Maxwell proved proved that "it is impossible, by means of any combination of reflections and refractions, to produce a perfect image of an object at two different distances, unless the instrument be a telescope," a telescope which must have unit angular magnification. T. Smith 2 as well as Herzberger3 found flaws in Maxwell's proof, but showed that his results were correct. Maxwell's proof does not depend on the symmetries of the system; lenses with particular symmetries were studied in detail by Buchdahl. He also comes to the conclusion that a lens such as the conference lens cannot be free from aberrations. From a practical standpoint this is not the end of the matter. How serious are the imperfections? The residual aberrations might be smaller than the unavoidable diffraction effects and therefore of no consequence. To the best of our knowledge the existing literature does not provide numerical estimates for the best image quality that can be realized. In this paper we begin to address this question; we show how a judicious mixture of analytic and computational techniques reduces the complexity of the problem to a manageable size. It turns out that for the conference lens the theoretical limitations are inconsequential as long as we allow the distortion to be large. If, however, the lens specifications are made more demanding, the fundamental restrictions become at times a serious matter, as we shall show by an example. One approach to the calculation of the residual aberrations is a frontal attack by algebraic methods in the style of Smith and Buchdahl. This method has the advantage that the relations between the aberrations and the design parameters can be expressed in analytical form. However, it usually remains unclear how far the series expansions should be extended, and the amount of algebra required is formidable because we must not only determine the unavoidable aberrations, but also balance them as best we can by choosing the optimum values for other aberrations that we do have under our control. We have chosen an alternative technique, mock ray tracing, 5 which consists of tracing rays through a lens specified by one of its eikonal functions rather than by its curvatures, thicknesses, and indices. As the physical symmetries of a lens can be easily translated into mathematical symmetries of the eikonal, this method is well suited to the problem at hand.

Peter Clark asked me to say a few words on how a designer undertakes the design of an unfamiliar problem like the one he presented for 0SA85 design study. The first thing an experienced designer usually does is draw analogies with previous well known designs and concludes:

Published biocular magnifiers are reviewed. Design methods and permissible levels of residual visual aberrations are proposed. Recent developments are mentioned.

It is proposed a novel configuration of zoom lens, which has a function to modify the focal length range due to zooming without any changes of the final image position and the overall length of the system. By employing Gaussian brackets, it is presented the analysis on the thin lens configuration of a novel zoom lens having range extender together with a designed example.

Wide angle to telephoto zoom lenses having zoom ratios of up to 7.5:1 have been designed and successfully developed for 35mm photography.

Large rear-screen TV projection systems are typically enclosed in a large box. To reduce the volume of the box two mirror folds have been used. A new wide angle lens covering a 70-degree field has been developed which further reduces the volume, yet only requires one mirror. The design evolved from the earlier design types, having a relatively weak first group, a strong positive second group, and a negative field flattener. It is dependent upon the use of aspherical surfaces, having no analogous all-spherical type.

Projection TV cabinets are bulky. Increasing the projection angle from ±25° to ±400° significantly reduces the size of the cabinet, but introduces severe requirements on the lens design to correct the large aberrations and poor illumination at the corners of the picture. Simpler designs with improved performance are possible if the cathode ray tube faceplate is curved.

This report describes the basic concepts for the design of high power long working distanct-: (L.W.D.) microscope objectives by comparing with a high power oil immersion objective. A 100X objective designed after the concepts is shown.

We studied two types of projectors for a liquid-crystal-display breadboard and designed and used an on-axis f/4 lens.

A new lens design concept, based on the use of a zero (or near zero) power corrector, will be described. The logical development of the design, based on the work of Schmidt', Houghton' and others will be discussed and examples will be given of moderate field of view lenses with apertures ranging from f/0.35 to f/2. It will also be shown that the lens configuration is relatively insensitive to the aperture stop location and that for less demanding applications only very basic optical glass types need be used.

The use of the zero power corrector concept has been extended to the design of objective lenses and magnifiers suitable for use in night vision goggles. A novel design which can be used as either an f/1.2 objective or an f/2 magnifier is also described.

Techniques are explored for enhancing the communication link between the optical designer and the engineering groups involved in the design and implementation of high performance optical hardware.

There is a growing requirement for precision lenses that can image millions of pixels. Our present methods of optical fabrication are not adequate for achieving design performance. There is need for more in process testing and less need for expensive testing. The mount design and assembly procedures are vitally important. There is need for development of new mounting techniques.

Serial scan thermal imaging systems require high speed scanning in order to generate a standard TV display of a scene. This is commonly done with a high speed rotating polygon mirror and a lower speed framing mirror. The scanner which is the subject of this paper uses a high speed rotating disc which embodies a circular array of spherical concave reflectors. Framing of the scene is done with an oscillating plane mirror, as with the polygon scanner. Use of the disc as the high speed line scan element allows a more com-pact scanner which runs at lower speed for a given data rate.

A tutorial is given on the development of spinning polygon laser scanners. A new approach is introduced, eliminating both wobble and bow without cylinder optics.

The concept of an afocal IR scanner is presented, showing how two distinct fields of view are obtained simultaneously, using only a single detector and sharing the same scan mechanism. The resulting system may view two separate directions at the same time, or may examine the same scene with different magnifications, but is more compact than two individual scanners performing the same task. The commonality of scanning and imaging components in both optical paths, and the requirement for efficient optical switching, place unique demands on the design of the imaging optics within the scanner.

This paper develops and illustrates the geometrical relationships characterizing the behavior of polygonal scan wheels used in scanning optical systems.

The constraints placed on the optical design by the other subsystems of a complicated system can be more difficult to satisfy than the imaging requirements. The space limitations inherent in a satellite optical system force many restrictions on the locations of elements and distances between these elements. The restrictions placed on the optical system by the thermal environment also serve to make the design more challenging. A meteorological satellite must operate in many wavelength bands over a wide region of the spectrum (from 0.55 to 15 μm). The arrangement of common and separate beam paths must be done carefully to ensure adequate transmission. Manufacturability at a reasonable cost is another important factor in the design. An optical design which satisfies the requirements for an operational meteorological satellite system is presented, and the requirements which most influenced the design are described.

As with other 3D projection systems, polarized images are projected onto a metalized screen and then viewed with polarizing glasses. In order to obtain a wide,bright picture, both images are placed on 70 mm film. These over-under images(.661X1.5inch and separated by a bar of .15 inch) are projected with a single lens. By means of mirrors,these images are separated,then polarized and registered on the screen. A novel camera relay-prism* creates the over-under images on the film with the proper orientation. It consists of a matched pair of front objective lenses; each is followed by a field lens, relay lens assembly, and finally a rear penta prism assembly.

The purpose of an analytical orthoprojector is the computer-controlled rectification of aerial photographs into a parallel perspective orthophoto. For the optical transformation the aerial photographs are scanned with a small slit. Each picture element must be exposed onto a specific position in the orthophoto at a specific magnification and orientation (rotation). The density of the orthophoto must be uniform without visible strips. Hence the basic requirements for the optical system are: large range zoom, uniform slit illumination independent of zoom position, image rotation, high boresight stability for projecting beam. The optical design of the Zeiss orthoprojector ORTHOCOMP Z 2 is discussed.

Various design concepts for exposure of transmission and reflection HOEs are described. A method for modeling these systems on optical design programs is discussed.

Criteria for evaluating the quality of optical concentrators are first discussed. Since nearly any optical system can behave as a concentrator, the suitability of several imaging and nonimaging optical systems for this purpose is examined, and their relative merits are evaluated. The basic design tool for nonimaging concentrators, namely the "edge-ray principle", is explained. An example is given in which it is shown how the concentrator system of choice can be selected nearly exclusively on the basis of the absolute and relative values of the required concentration ratios.

An HOE and refractive element create a hologlass doublet for correction of longitudinal chromatic aberration, using HOE negative dispersion. Cemented and separated type of hologlass combinations are discussed.

An attempt has been made to modify Bergstein's iteration design procedure for the four element optcal-compensation varifocal system. Here all available degrees of freedom are made use of, to satisfy the desired conditions in practical systems under consideration. It has been concluded that most of the procedures claimed after Bergstein, can be derived as special cases of the proposed procedure. A minor mechanical movement to the central stationery element of the zoom unit has been proposed to reduce the magnitude of image oscillations further, resulting in opto-mechanical compensation system, and this minor movement also has been envisaged to provide explanation for the different possible profiles of the movement followed in other solutions of four element opto-mechanical compensation zoom systems.