In this paper a new illumination system with high power light emitting diode (LED) sources for projection displays is proposed. The prototype of a rear projection system has been developed by using red, green and blue (RGB) LEDs and three LCD panels. Four LEDs were used for each primary color. Parabola reflectors were used for collimating the LED lights and a new array style of LEDs and collimators were used. Integration rods were directly used between collimators and LCD panels without relay lens for uniform light distribution. A 40" projection display system was made with a light output about 25 lm on the screen and the projection engine was very small comparing to the original engine which uses an arc source.

We consider the problem of designing illumination optics to transfer flux from an LED source into a rectangular target aperture. An important constraint on this design was that it provide a large working distance between the source and the target aperture. An optimized compact nonimaging system consisting of an aspheric totally-internally-reflecting (TIR) lens along with an aspheric singlet was developed to provide high flux-transfer efficiency with the required working distance. A global optimization procedure was used to search a 65-dimensional parameter space for the set of optical-component shape and positioning parameters providing maximum performance, while satisfying the design constraints. A total source-to-target flux-transfer efficiency of 85.0% was achieved by the design, assuming ideal AR coatings on all refractive surfaces.

We present a new type of optical engine for projection displays. The optical engine is based on a light guide with embedded color filters. It is intended for three-panel projection displays with micro-display panels of the transmissive type. The light guide serves the purpose of integrating the light and guiding the light to each of the three panels. Proximity illumination is used to illuminate the micro-display panels: the exits of the light guide are in close contact with the entrance of the panel. The optical design considerations underlying the principle of using light guides are discussed. Among these considerations are measures required to prevent light leakage. We also discuss light guide based optical engines relying on the principle of color recycling and polarization recycling. The results of simulations and experiments on a prototype are discussed. It is shown that the use of light guides enables a very compact design. The lumen output of such a projector can be comparable to, or even better than that of conventional systems.

In this paper the etendue efficient illuminator having retro-reflecting aperture with a slit is proposed. Such an illuminator allows to increase light efficiency of the projection system which has an etendue much smaller than the arc lamp used typically as the light source. In modeling and analysis results, it is evaluated that retro-reflecting aperture with a slit make it possible to increase light efficiency of single panel color scrolling projection system by about 15%.

The Simultaneous Multiple Surfaces design method (SMS), proprietary technology of Light Prescription Innovators (LPI), was developed in the early 1990's as a two dimensional method. The first embodiments had either linear or rotational symmetry and found applications in photovoltaic concentrators, illumination optics and optical communications. SMS designed devices perform close to the thermodynamic limit and are compact and simple; features that are especially beneficial in applications with today's high brightness LEDs. The method was extended to 3D "free form" geometries in 1999 that perfectly couple two incoming with two outgoing wavefronts. SMS 3D controls the light emitted by an extended light source much better than single free form surface designs, while reaching very high efficiencies. This has enabled the SMS method to be applied to automotive head lamps, one of the toughest lighting tasks in any application, where high efficiency and small size are required. This article will briefly review the characteristics of both the 2D and 3D methods and will present novel optical solutions that have been developed and manufactured to meet real world problems. These include various ultra compact LED collimators, solar concentrators and highly efficient LED low and high beam headlamp designs.

In illumination systems for projection display, it is often necessary to tailor the aperture size and angle of light to suit the system requirements such as an imager panel. One way to do that is to use lenses, but depending on the angle and area this may not be achieved economically in terms of space and costs. Non-imaging optics is often used to perform such function. A compound parabolic concentrator (CPC), which is simple to use and small in size, is one example of such approach, but it has the disadvantage of non-uniform output intensity profile and not conserving brightness. A tapered light pipe (TLP) is often used instead to alleviate such shortcomings, but such light pipe would need to be infinitely long to conserve brightness, which makes it quite impractical. In this paper, a lensed tapered light pipe is described that change the area and angle of light with minimal loss of brightness and provide a very uniform output intensity profile at the output. Numerical modeling using ray tracing computer program is used to optimize the TLP for each applications.

An analytical model of flux propagation in light pipes, termed the flux confinement diagram (FCD), is further developed and applied. The construction of the FCD is reviewed. Non-planar surface geometries, non-constant cross-sectional geometries, and non-rectangular cross sectional geometries are examined with the FCD. It is shown that in the limit of a circular cross section the predictions of the FCD match the theory for large core fibers. Additionally, the angular propagation space defined by the FCD is explored further to describe the redistribution of propagating flux after a sudden change in geometry, such as a bend. This analysis is used to explore total internal reflection (TIR) at light pipe output surfaces. The implications of analytical modeling with the FCD on light pipe design are discussed.

One of the important properties to the emission of light from a side lighting waveguide using a notched cladding and solid core is the numerical aperture of the source at the input of the fiber core. Reducing the emission angle of light entering the fiber allows light transmission further down the fiber so that it becomes available for side emission at an increased distance from the source. This allows side-lighting fibers to be made longer and more uniform. Utilizing this effect, the fiber can be notched in a controlled manner that efficiently converts the source light accepted into the waveguide into a side-emitted light that is distributed uniformly over the length of the fiber. A detailed optical analysis of this system will be presented with evaluation of the effect on the resulting side light emission as the source light parameters are adjusted. The analysis will include discussions on the trade-off associated with the reduced source efficiencies at lower etendue values and the optimization of the system as a whole. Experimental results using a metal halide arc lamp with dual parabolic reflector as the source and solid-core polymer fiber of 12-18 mm diameter with customized notched cladding will be presented and discussed.

We review the theory of the radiative transport equation governing the radiance in a random medium. Using symmetry and orthogonal properties of plane wave solutions, we can compute readily the Green's function for a uniform medium. We use this Green's function to develop a general theory for inhomogeneous media analogous to scattering theory for classical wave propagation.

Application of the Stokes theorem to the conservation of the 2D etendue of a one-parameter bundle of rays leads to the Lagrange's integral invariant, one consequence of which establishes that the integral &sh;p∙dx between any two points is independent of the path of integration (p is the ray vector field and x is the vector position), and more generally, the integral &sh;p∙dx between two wavefronts is independent of the path of integration. This integral is called the optical path length. This is another way to see Fermat's principle. The conservation of 2D etendue is a property of any Hamiltonian system. Using the Hamiltonian formulation, there is no difference between the configuration variables x and their conjugates p. Thus an integral invariant &sh;x∙dp can also be established similar to the Lagrange invariant. We show how its application to simple cases leads to Cartesian-oval designs through an unconventional method. The 2D etendue conservation is connected with Levi-Civita's anormalita function and with the ray equation. In this connection we found that the equation p×(∇×p)=0 suffices for a vector field to be a ray vector field.

We demonstrate a consistent approach to incorporating the diffractin properties of the instrument in optical measurements.

The purpose of this paper is to present an overview of the Edge-Ray Theorem in 2D geometry, covering the different optical systems treated up today, including some cases (as the refraction/reflection of sequential optical surfaces) which have not been formally discussed previously, and also analyzing the role of slope discontinuities in the creation/annihilation on edge rays. Also illustrative novel examples are given. In section 2, a simple device that seems to beat either the Edge Ray Theorem or the Second Law of Thermodynamics is presented. In section 6, it is proven that there exits perfect solution to the theoretical problem of achieving maximum concentration on a circular receiver from a source at infinity with a single slope discontinuity and a sizeable gap between optics and receiver. At the end (section 7), the explanation of the device of section 2 offending the Edge Ray Theorem will be given.

Conventional incandescent light bulbs have a wire filament acting as an extended light source with nearly constant intensity throughout its quasi-spherical emission pattern. Here we present a novel family of optical devices that make use of commercially available Lambertian or near-Lambertian LED light sources, in conjunction with tailored optical element bonded to the top surface of the LED. These hybrid devices can emulate the output of traditional incandescent filaments, or can be designed to produce a wide range of light output beam patterns. We call these new devices Virtual Filaments, as they can be designed to appear the same as an incandescent filament, with a similar light output pattern, and having a similar focal position above the base. These new lamps can then be used in the same applications as those they replace, thus eliminating the need to redesign or replace the original luminaire. We present several possible optical designs that can be used with a number of standard LEDs to replace standard incandescent bulbs. In one example we show a design that provides an output with near-uniform intensity across a full beam angle of 300 degrees, from a focal position 20 mm above an LED. Other major advantages of these new devices include their ability to be given sharp cutoffs, to homogenize non-uniform LED light sources and to color-mix the output of RGB LEDs.

Purely imaging strategies can offer ultra-compact concentrators and illuminators both of which approach the thermodynamic limit to optical performance. The tailored optical surfaces are monotonic functions that can be expressed analytically, which can facilitate optimization studies as well as practical fabrication. Also referred to as aplanatic designs, devices with two smoothly contoured surfaces can compete with, and even exceed, the performance of high-flux nonimaging systems. We develop and analyze tailored imagin devices for two-stage reflector systems of varying numerical aperture, motivated by applications in solar concentration as well as light collimation.

To characterize CMOS imagers an LED-based multi-spectral optical source system was designed and tested which is capable of illuminating a 15-mm field at a conjugate distance of 50 mm with 98 percent uniformity. The calibration source is comprised of an array of RGB semiconductor LEDs, an IR cutooff filter and a diffusing lens. The system is integrated into an anodized aluminum housing. The spatial uniformity of the LED optical source was compared with an optical integrating sphere and an Optoliner projection system.

Liquid-crystal-on-silicon (LCOS) technology enables affordable, true high definition resolution televisions for the growing rear projection display market. In this review, we will present a taxonomy of LCOS optical architectures and survey some recent advances in LCOS panels and the optical systems that support them.

We investigate a class of illumination systems, which consist of a light source, a conical (elliptical or parabolic) reflector and an optical device that collects the radiation. The collection efficiency of such systems is limited for conventional means of light collection such as rod and fly's eye integrators. The distribution of the collecting etendue in phase space differs from that of the light, which comes from the reflector. The mismatch is caused by a series of optical effects such as the actual shape of the source, a variable magnification of the conical section, and by consequences from skewness conservation. Nevertheless, this situation is accepted as in many commercial devices, and one may at least ask what conditions yield best efficiency. We show, for example, how to calculate optimum eccentricity of an elliptical reflector to be used with a given source and integrator. To go further and to reduce the etendue mismatch, one may either redesign the lamp for a better fit to the collection system or modify the collection method. Based on the analysis of the conical reflector, we discuss options to achieve better collection efficiency and show examples. For a simultaneous treatment of both elliptical and parabolic reflectors, we introduce a new method that describes the lamp by a luminance distribution on the director circle or plane of the conical section.

Simple optics composed of a spherical lens and a conic mirror are described and the relation between the radius of the lens and height of the cone on far field illuminance performance is analyzed for a fixed exit aperture dimension. Ray sets for real LEDs were used to simulate the performance of the hybrid optics and it is shown that there are combinations of values for the lens radius and cone height for which the optic produces an approximately constant illuminance pattern on a distant target. The effects of varying the lens radius while keeping the cone height constant, and of varying the cone height while keeping the lens radius constant, are also presented, as these variations result in beams of varying angular spread. It is shown that a relatively course two parameter optimization can find near optimum solutions, where the optimization is carried out using ray sets of commercially available LEDs and the merit function is constant illuminance.

In this contribution we consider the problem of designing coupling optics to optimally transfer light from a metal-halide arc lamp to a large core polymethyl methacrylate (PMMA) fiber optic cable. We investigate a refractive optical coupling concept comprising a non-axisymmetric bifurcated refractive glass TIR lens (TIRL) set. The design goal is to maximize the photometric flux incident onto two 19-mm-diameter apertures within an acceptance half angle of 38º. The light source is an 80-Watt metal halide arc lamp, characterized by means of photometric data measured by Radiant Imaging. The lenses each comprise a central refractive section combined with an annular TIR section. The exit pupil of each TIR lens is separated from the entrance aperture of the target by a short air gap. The refractive section of the TIR lens utilizes two aspheric surfaces to collect source flux over a central solid-angular region and redirect it into the target. The TIR section utilizes a total of three aspheric surfaces, two of which are refractive and one of which is a TIR surface. To account for right-left asymmetries in the optical source, the TIR lenses were independently globally optimized for both sides. Our TIRL design has the advantage of being able to collect and control radiation emanating at both large and small angles from the source, with little overall loss. The TIRL, with an ideal AR coating, has a predicted coupling efficiency of 89.6%. This was accomplished even though the target to source etendue ratio is only 63.7%. This is possible due to the ability of this design to preferentially transfer radiation from the higher radiance portions of the source phase space to the target. The work described above was funded by the Defense Advanced Research Projects Agency's High Efficiency Distributed Lighting Program known as HEDLight.

The study of general microstructures in 2D geometry and rotational 3D microstructures is presented. The study is based on infinitesimal microstructures for some calculations and the macro-profile of the surface can be treated as a new type of optical surface with a certain deflection law, which will be different of the reflection law or the Snell law. In two dimensions, we discuss the propagation of wavefronts by general microstructured surfaces (which do not fulfill the Fermat principle) and the discontinuity of the eikonal function at the microstructure. Naturally, a classification of the microstructures is obtained (regular and anomalous) and the concept of 2D ideal microstructures is also introduced, as those that perfectly couple two macroscopic extended bundles in 2D geometry. In 3D, after classifying the rotational optical systems into point-spot and ring-spot types, the first-order properties of both regular and anomalous rotational microstructured surfaces are discussed. Finally, an application of anomalous rotational microstructured surfaces to the problem of mixing the light from three RGB LED chips is introduced.

Lighting plays an important role in many applications of computer vision, machine vision and computer graphics. Often, an object needs to be photographed multiple times, in each of which lighting comes from a different direction. Lighting which uses a single source per image is prone to dynamic range problems, especially in dark areas and in specular highlights. In addition, it becomes a practical problem to use an increasingly larger number of discrete sources (say, hundreds). To counter these problems we develop a novel illumination strategy. In each image, multiple light sources irradiate the scene simultaneously. The set of light sources is different in each image, but not mutually exclusive. Then, the contribution of each individual source is extracted in computational post-processing. The number of acquired images using this approach is the same as the number used in single-source images. However, thanks to the multiplexing of light in the raw images, more light is used from a variety of directions, diminishing problems of dynamic range. We derive the optimal illumination multiplexing scheme, which increases the SNR of the images by (√n}/2, where {n} is the number of sources. This lighting strategy is complemented by a novel illumination setup. The setup is easily built and scaled to a huge number of sources, and is controllable by the computer. These advantages are obtained since the apparatus is based on indirect lighting originating from an LCD projector.

This paper focuses on the facets of illumination system optimization, in particular parameterization of objects, the number of rays that must be traced to sample properly its properties, and the optimization algorithm with the associated merit function designation. Non-interference ensures that the parameterized objects do not erroneously intersect each other or leave gaps during the steps of the optimization procedure. The required number of rays is based on a model developed for television cameras during their initial days of development. Using signal to noise ratio, it provides the number of rays based on the desired contrast, feature size, and allowed error probability. A lightpipe is used to highlight the nuances of this model. The utility of using system symmetry to increase ray count is also discussed. A modified simplex method of optimization is described. This algorithm provides quicker convergence than the standard simplex method, while it is also robust, accurate, and convergent. A previous example using a compound parabolic concentrator highlights the utility of this improvement.

A new optical system is presented for a flat-plate micro-concentrator. A single unit is made of an off-axis portion of a paraboloid mirror that focuses sunlight on a photovoltaic cell hidden behind the adjacent unit. This unit can be repeated in a closed-packed two-dimensional array to form a module as large as a conventional flat-plate module, only slightly thicker. Theoretical study shows that an entire module could be molded as a single piece. Tiny solar cells, a few millimeters a side, packaged inside a Secondary Optical Element would be inserted in the module and interconnected from the backside. Each cell would operate at a concentration ratio of several hundreds suns but only require passive cooling because of the small power involved.

Experimental results generated with novel miniature fiber-optic concentrators and commercial tandem III-V concentrator solar cells are reported, including (1) measured power densities up to 10,000 suns, (2) solar cell efficiencies in excess of 30% and (3) totally passive cooling. Mini-dish concentrators (a) generate uniform and individualized cell illumination, (b) allow assembly from readily available elements, and (c) are devoid of chromatic aberration. Measurements include the sensitivity of conversion efficiency to (i) power input, (ii) extreme flux inhomogeneities and (iii) the modified spectrum from fiber-optic concentrators. The weak sensitivity of cell performance to acute non-uniformities in flux map is addressed with a relatively simple model that regards the cell as an effective parallel connection of its uniformly irradiated areal elements. Our findings bode favorably for the feasibility of such concentrator designs at concentration levels as high as thousands of suns.

We report results of a study our group has undertaken under NREL/DOE auspices to design a solar concentrator with uniform irradiance on a planar target. This attribute is especially important for photovoltaic concentrators.

Filters are employed in optical systems to reduce stray light and block unwanted spectra. However, common filter materials are susceptible to solarization, a bulk degradation wherein the material properties change when exposed to UV radiation. A new decay model is developed to describe this behavior. Solarization effects for various kinds of filters are measured and recommendations for mitigation are made.

Array illuminator based on Talbot effect is an important optical element that has wide applications in optical interconnection, optical communication, and optical computing. This paper summarizes our recent results on this subject. Symmetry of the Talbot effect, that was reported in Optics Communication 115, 40 (1995), is now realized as the first step to revealing other rules for explanation of the Talbot effect for array illumination. The prime-number decomposing rule (Applied Optics 40, 607 (2001)) shows that the number of phase level of a Talbot array illuminator is related with the prime number. Along with the study of the characteristics of the Talbot array illuminator, the applications of the Talbot effect are also in progress. Talbot phase codes are the orthogonal codes that can be used for phase coding of holographic storage. A new optical scanner based on the phase codes for Talbot array illumination has unique advantages over the previous Fresnel encoding method. Furthermore, the hexagonal array illumination based on the Talbot effect was reported in Optics Letters 27, 228 (2002). Recently, a novel two-layered multifunctional computer generated hologram based on the fractional Talbot effect was proposed and implemented (Optics Letters 28, 1513 (2003)), which can be multifunctional for potential use in secure system technology. We believe that the results reported in this paper should be useful tools for further exploration of the Talbot effect and design of novel optical devices that should benefit the practical applications.

At application of "pillow optics" for various lenses a problem of non-uniform distribution of output light can be faced. Even if input light from the source (i.e. before the lens) is a uniform parallel beam, the output light will not be uniform: Intensity is higher for central angles than for outlying angles. It takes place due to geometry of spherical pillows. To improve angular uniformity of output light, the other kind of pillow geometry can be applied instead of spherical geometry.

Light sources consisting of multiple Light-Emitting Diodes (LEDs) are becoming the preferred choice for many lighting applications that require uniform illumination distribution. However, packaging density of LED arrays is limited by cost, available space, and particularly by thermal problems. Therefore low density or diluted arrays of LEDs are becoming the option for many applications. This paper presents an investigation of the effects on illumination uniformity due to dilution on light sources consisting of multiple LEDs. The characteristics of illumination uniformity are obtained for several diluted-array configurations and degrees of dilution. This analysis is performed using a radiometric analysis, and considering each LED as an imperfect Lambertian emitter. Analytical expressions are derived for the maximum degree of dilution for different configurations of LED arrays.

In recent years, several architectures have been proposed for projection systems with an improved light efficiency by means of color recycling and/or polarization recycling. The recycling of light takes place in a rod integrator where light is coupled in from the lamp through a small hole in an entrance mirror. At the exit of the integrator, light of the wrong polarization state and/or wrong color is reflected back such that, after a round trip in the integrator, the light has a second chance of passing through the exit with a different polarization state or through a different color filter. Besides for recycling light of the wrong color or polarization, the integrator may also be used for recycling the unused light of pixels that are in a dark state. This allows for an increased brightness of bright parts in a dark scene, the so-called sparkling effect known from CRTs. We analyze the combined effects of color, polarization, and dark pixel recycling, extending the models previously proposed by Duelli et al. and by Zwanenburg.

Long UV radiation exposure can result in damages of biological tissues, as burns, skin aging, erythema and even melanoma cancer. In the past years an increase of melanoma cancer has been observed and associated to the atmospheric ozone deployment. Attendance of sun tanning unit centers has become a huge social phenomena, and the maximum UV radiation dose that a human being can receive is regulated by law. On the other side, UV radiation is largely used for therapeutic and germicidal purposes. In all these areas, spectroradiometer and radiomenter are needed for monitoring UVA (315-400 nm), UVB (280-315 nm) and UVC (100-280 nm) irradiance. We have selected some commercial photodiodes which can be used as solid state detectors in these instruments. We have characterized them by measuring their absolute spectral response in the 200 - 400 nm spectral range.