This paper presents our practical experience with etendue and optical system design. We clarify several different commonly used definitions of etendue. Examples are given that illustrate the applicability of conservation of etendue.

The laws of Planck and Kirchhoff are fundamental to a physical model yielding the radiance distribution emitted from a filament light source. The filament is made of a wire coil, or a coiled coil, or even a coiled coiled coil. Some parts of the filaments face other parts. Some light is reflected or absorbed between parts of the filament. This effect is termed light recycling. Light recycling depends on the geometry of the filament, and its material properties. Our model is based on the thermodynamics, geometry, material and electrical properties of a filament lamp. Light recycling is integrated into the model. The model calculates the one-dimensional temperature distribution in the filament by solving the time dependent heat transfer equation. The results of the model are verified with absolute radiance measurements. Parameters are identified in order to increase the accuracy of the values used for material characterization. The source model may be integrated into optical software packages.

In Monte Carlo ray tracing, the efficacy of variance reduction techniques is often the subject of debate. One portion of the debate regards the use of ray-splitting in illumination analysis. While analysis results should be the same whether raysplitting or no ray-splitting is used, one approach might result in better precision for a given calculation time. Additionally, inexperienced illumination designers may perform analyses in such a way as to exacerbate the difference in precision between the two methods. This results in a very important decision for illumination designers: using ray-splitting or not can affect results and the time spent getting them. For this paper, common illumination applications are analyzed by ray tracing in TracePro1 (a non-sequential Monte Carlo ray tracing program) and the analysis results are compared. Both raysplitting and no ray-splitting methods are used to see if the analyses converge to the same results for simple setup conditions. The results will illustrate the factors to consider of before choosing to use ray-splitting or not, and show examples when one method may be better than another.

Conversely to the common way, geometrical optical in free-space and two dimensions is deduced from black body radiation transfer. This is obtained from the only hypothesis of energy conservation, reciprocity and that the emission of convex bodies is proportional to their perimeter. The analysis of an abstract relay element will lead to deduce the reflection law in an unconventional form and new properties of the elliptical mirror as an ideal radiation transfer device.

We derive a generalized functional method of nonimaging optical design, applicable in the case of axisymmetric or translationally symmetric geometries utilizing extended sources and multiple reflective and/or refractive optical surfaces. Design methods are described for generation of a specified intensity distribution as a function of angle or a specified irradiance distribution as a function of position along the shape profile of an aspheric target surface. A related method is presented for the design of two successive optical surfaces that transform a given source distribution into either two specified irradiance distributions on successive target surfaces or a specified irradiance distribution and a specified intensity distribution.

A novel waveguide-optical integrator is introduced for applications to LEDs. The concept is based upon a Kohler illuminator made of Luneburg lenses. Typical Kohler illuminators are formed by pairs of thin lenses, and perform badly when the paraxial approximation is rough, i.e., when the angular span of the incoming rays is wide. In contrast, the new illuminator performs ideally for angular spans up to 90^o (±45^o), and has only a 3% loss for a 180^o angular span. In general such an illuminator cannot be made in 3D, because adjacent Luneburg lenses overlap. It can, however, be implemented in planar optics, by using Rinehart geodesic lenses, which moreover do not use gradient index material. This waveguide device has application in illumination engineering as a light mixer, particularly for LEDs. Another light mixer using a combination of two kaleidoscopes with a geodesic lens is also presented. Irradiance at the exit of a kaleidoscope has good light mixing if the kaleidoscope is long enough, but the intensity is never well mixed, irrespective of the length. Inserting a Rinehart geodesic lens produces a 90-degree phase-space rotation of the rays, i.e., it exchanges irradiance and intensity. A further kaleidoscope assures complete mixing in both irradiance and intensity.

In the design of illumination lenses, there is a fundamental incompatibility between the spherical geometry of light radiating outwards and the rectangular geometry of typical illumination targets, analogous to trying to fit a round peg in a square hole. This amounts to establishing a rectangular grid on the sphere, the perennial problem of map-makers. Here we apply a new pseudo-rectangular spherical grid, originally developed for parallel-processor simulations of semiconductor devices, to establish correspondence between source-grid cells and the rectangular cells of a target grid. This correspondence establishes a grid of deflections, whereby source rays are deflected so as to impact the proper cell on the target grid. For a given lens refractive index, each deflection is implemented by the angles of inclination the ray encounters going into and out of the lens, resulting in two grids of surface gradient values, for the inside and outside lens-surfaces. Central spines are obtained for these surfaces by a linear integration, after which adjacent rows are successively obtained in a lawnmower fashion, so as to heal any imcompatible cross-derivatives. Example lenses are illustrated.

Aplanatic optics crafted from transparent dielectrics can approach the etendue limit for radiative transfer in pragmatic near-field systems. Illustrations are presented for the more demanding realm of high numerical aperture (NA) at the source and/or target. These light couplers can alleviate difficulties in aligning system components, and can achieve the fundamental compactness limit for optical devices that satisfy Fermat's principle.

Beam-shaping optics are used in various optical fields to change the luminous intensity distribution. In this paper a flexible method is presented to design beam-shaping optics with aspherical surfaces transforming the intensity profile of the light beam into any desired profile. The method is applied to a collimator lens that transforms a beam from a Lambertian emitter to a uniform light distribution.

The problem of generating a cut-off line with a carefully calculated reflector contour has been treated in detail by Spencer et al. for the case of a cylindrical source of light mounted perpendicular to the optic axis. Because this geometry does not properly represent the geometry in which standard light sources are used in the illumination systems which we study, the attempt was made to extend this theory to anisotropic light sources. This case of lower symmetry is closer to the geometry of light sources encountered in headlamp design. Spencer et al. were able to obtain an implicit algebraic equation for the problem of high symmetry that they analyzed. After adopting their method to the problem under investigation, the method of analysis used was different insofar as an algebraic equation was not obtained and the corresponding ordinary differential equation and the corresponding initial-value problem were solved instead and the solutions are visualized with the aid of a computer-algebra system. In this context, the concept of a so-called polar line or surface proved helpful. This describes a set of points that connect the tangent lines that link a given point of the reflector contour to a given extended lightsource of low symmetry. The extension of the lightsource is assumed to be elliptical in the plane that contains the optic axis and the plane perpendicular to the cut-off line. The analysis extended to the anisotropic case gave some insight into the underlying scaling laws and geometrical constraints.

How to design and calculate asymmetric reflectors using an equation method is presented here. Isotropic linear source used with trough reflector and side emitting cylindrical source used with revolution reflector are given as examples. All results are presented for side-illuminating configurations that may require converging optical plans.

The Carambola is an optical device designed to allow the deterministic and multiple recycling of light rays. The rays transit through the source a defined number of times before exiting in the same phase space as light directly emitted and not recycled. The brightness enhancement by light recycling (the optical light recycling factor) with the Carambola depends on the reflectivity of the reflecting walls of the Carambola, as well as on the size of the source and on the optical thickness of the source. The results of a ray-tracing simulation and an analytical model are promising an optical light recycling factor up to three for a Xenon high-pressure arc discharge lamp.

Wavien patented dual paraboloid reflector (DPR) system, while optimal in maximizing image brightness (equivalent to minimizing the image aberration) at the input of its light pipe, does not produce maximal lumen throughput at its output. The overall lumen throughput depends in large part on three factors: Fresnel losses and image aberration, both defined at the input face of the light pipe, and the light pipe's input dimensions. Fresnel losses can be reduced by narrowing the cone angles of the light cone incident on the light pipe, which in turn can be achieved by increasing the size and/or focal length of the second paraboloid reflector. Smaller cone angles also mean reduced tapering of the light pipe which translates into larger input sizes (as its output dimensions are fixed) and higher coupling efficiency. Unfortunately this gain in coupling efficiency comes at the expense of breaking the system symmetry, which destroys 1:1 imaging and leads to increased aberration and reduced brightness. Using a ray-tracing software an optimal point of operation can be reached and it is found that at least 10% increase in lumen throughput over the symmetric DPR system is achievable.

Solid State Lighting is becoming increasingly more advanced, both in terms of lumen output as well as energy efficiency. However, implementation in color consumer lighting products, such as the Philips Ambilight television sets, still requires improvements in both color reproduction as well as intensity uniformity. To build a lighting system capable of correctly reproducing a large color spectrum, 3 primary colored LEDs are required. However, this approach causes problems. In particular, the generation of a white color without color fringes is difficult to implement, as the total amount of light from each primary color should ideally be identical at each position within the light bundle. Our paper focuses on systems using a limited number of high power LEDs. The lumen output of these LEDs is such that even a single red, green and blue LED together can deliver the required lumen output for certain applications. To optimize performance for both luminance and color uniformity we investigated several design options. Ray tracing simulations are compared to the performance of real size prototypes, and recommendations are given for the design of color lighting systems.

Brightness enhancement of backlighting for displays is typically achieved via crossed micro prismatic films that are introduced between a backlight unit and a transmissive (LCD) display. Prismatic films let pass light only into a restricted angular range, while, in conjunction with other reflective elements below the backlight, all other light is recycled within the backlight unit, thereby increasing the backlight luminance. This design offers no free parameters to influence the resulting light distribution and suffers from insufficient stray light control. A novel strategy of light recycling is introduced, using a microlens array in conjunction with a hole array in a reflective surface, that can provide higher luminance, superior stray light control, and can be designed to meet almost any desired emission pattern. Similar strategies can be applied to mix light from different colored LEDs being mounted upside down to shine into a backlight unit.

Small-sized liquid crystal displays are widely used for mobile applications such as cell phones. Electronic control of a viewing angle range is desired in order to maintain privacy for viewing in public as well as to provide wide viewing angles for solitary viewing. Conventionally, a polymer-dispersed liquid crystal (PDLC) panel is inserted between a backlight and a liquid crystal panel. The PDLC layer either transmits or scatters the light from the backlight, thus providing an electronic control of viewing angles. However, such a display system is obviously thick and expensive. Here, we propose to place an electronically-controlled, light-deflecting device between an LED and a light-guide of a backlight. For example, a liquid crystal lens is investigated for other applications and its focal length is controlled electronically. A liquid crystal phase grating either transmits or diffracts an incoming light depending on whether or not a periodic phase distribution is formed inside its liquid crystal layer. A bias applied to such a device will control the angular distribution of the light propagating inside a light-guide. Output couplers built in the light-guide extract the propagating light to outside. They can be V-shaped grooves, pyramids, or any other structures that can refract, reflect or diffract light. When any of such interactions occur, the output couplers translate the changes in the propagation angles into the angular distribution of the output light. Hence the viewing-angle characteristic can be switched. The designs of the output couplers and the LC devices are important for such a backlight system.

This study develops a statistical prediction model for backlight systems based on a semi-analytical and experimental approach. The prediction model features an iteration algorithm which uses experimental measurements of the luminance, luminance cone angle and luminous efficiency to generate highly accurate luminance predictions. The prediction model allows the effects of manufacturing errors or uncertainties which cause a deviation of the luminance cone angle or an uneven luminance uniformity to be accessed. The results show that achieving an even luminance cone angle, i.e. a smaller mean and standard deviation of the luminance cone angle, is essential if the backlight luminance level required to achieve a high-brightness backlight is to be enhanced. It is shown that improving the luminance uniformity of the backlight is beneficial in increasing the luminance level. However, the influence of the backlight luminance uniformity is not as great as that of the luminance cone angle. Finally, a comparison between the analytical and experimental results shows that a good agreement exists between the results of the proposed statistical model and the experimental data.

In LED projection displays, total lumen-output equals the source-luminance multiplied by the etendue of the spatial light modulator, the latter being a bottleneck that cannot be overcome. In addition, the luminance of existing LED sources is still too low for many projection-display applications. This has spurred research into finding ways to increase their brightness without a significant loss in efficacy. Current techniques to increase LED source-luminance include: (a) a photonic lattice atop the LED, which partially collimates the exiting light though lowering the efficacy (by Luminus), (b) recycling the LED light through the chip via TIR on a flat cover rather than a dome (several LED suppliers), and (c) a light-confining box with an exit aperture smaller than the chip (GoldenEye). All the above mentioned existing approaches achieve an increase in luminance for the LEDs at the expense of a considerable drop in efficacy. In this paper we present four novel and different ways (patent pending) to considerably enhance LED luminance and offering the possibility of having relatively high efficacy.

Light from several LEDs or other light sources may be combined using light guides shaped as manifolds. These manifolds are composed of smaller elements such as CPCs, angle transformers, angle rotators, light shifters, light guides or others. Although some components, such as CPCs or angle transformers, have all-optical surfaces, other devices may be designed with non-optical surfaces. These may be used to place the injection gate in the case of injection-molded optics, to attach handles or holders and other non-optical components to the manifold without affecting the optical performance. Also, in some of these devices, the geometry can be changed by simple changes in the position of the curves that compose the optic profile. These optics may be applied in efficiently combining light from several LEDs into one single large source, changing the aspect ratio of a light source or in distributing light from one (or more) sources onto several targets.

The Defense Advanced Research Projects Agency has initiated the Very High Efficiency Solar Cell (VHESC) program to address the critical need of the soldier for power in the field. Very High Efficiency Solar Cells for portable applications that operate at greater than 55 percent efficiency in the laboratory and 50 percent in production are being developed. We are integrating the optical design with the solar cell design, and have entered previously unoccupied design space that leads to a new architecture paradigm. An integrated team effort is now underway that requires us to invent, develop and transfer to production these new solar cells. Our approach is driven by proven quantitative models for the solar cell design, the optical design and the integration of these designs. We start with a very high performance crystalline silicon solar cell platform. Examples will be presented. Initial solar cell device results are shown for devices fabricated in geometries designed for this VHESC Program.

A system exploiting solar energy, by means of optical collectors and fibres, has been applied for indoor illumination. The project has been called "The Sunflowers" for the property of solar collectors to track solar position during the day. Every "sunflower" contains several solar collectors, each of which is coupled to an optical fibre. The "Sunflower" is provided of mechanical systems and electric accessories for solar tracking. The light focused by the solar collector can be used in two possible ways: for internal illumination with direct solar light; otherwise it can be accumulated for lighting when the sun is not present. The first function is obtained coupling the optical collector to an optical fibre, which transports the solar light in selected points within the showcases. The second one consists in focusing solar light on a photovoltaic cell of the last generation type with high efficiency. In this configuration the photovoltaic cell converts the focused light into electric energy to be used for illumination in case of sun absence. A demonstrative installation has been realised applying this solar illumination system to museum lighting: a prototype has been tested in a prestigious museum in Florence.

The paper proposes a newly developed light source module mixing R.G.B. LED (Red, Green, Blue, Light Emitting Diode) for a modern LED backlight system, which might replace the white light LED and traditional CCFL (Cold Cathode Fluorescent Lamp). This system consists of three parts, a newly designed symmetric mirror for complete color mixing, a R.G. B LED (Red, Green, Blue Light Emitting Diode) light source and a diffusion plate with a newly designed microstructure for the improvement of light uniformity. The light module is designed and simulated by an ASAP (Advanced Systems Analysis Program), in which a sample with 160mm×90mm( around 7-inch panel) is presented with sixteen LEDs light source. Good color saturation and excellent light uniformity are reached in this research.