This PDF file contains the front matter associated with SPIE
Proceedings Volume 7103, including the Title Page, Copyright
information, Table of Contents, Introduction, and the
Conference Committee listing.
Response surface modeling (RSM) is described as a tool to perform optimizations and sensitivity analysis in optical
modeling. With this method, the behavior of the system is first mapped out using a limited set of ray-tracing simulations,
carefully spanning the full parameter space. This can already be done before the full merit function is known. The
relation between design parameters and system performance is then approximated by fitting the simulation results to
functional forms. All subsequent optimizations are then performed very time-efficiently on the functional descriptions of
the dependencies. This contrasts with direct optimization, where the computationally intensive optical modeling is in the
loop of the optimization algorithm, and where changes in merit function, mapping out trade-offs, and determining
sensitivities, are very time consuming. The paper will discuss the advantages of RSM with respect to direct optimization
and give recommendations for the type of problems that are preferentially addressed by RSM. The method will be
illustrated by a case: how optical simulations were used in the design of LumiramicTM phosphor conversion components
Illumination systems suffer from tolerances like all optical systems, including gross, process, roughness, and system
errors. A common system error in solid-state lighting (SSL) luminaires is source variation due to variation within
selected LED bins. Tolerances of LED bins include peak or dominant wavelength and flux output. These bin variations
lead to chromatic shifts in the resulting illumination distribution when discrete spectral emitters are mixed to provide a
desired color. Software tools are developed to provide ease in inspection of color mixing of discrete spectral LEDs
taking into account the selected bins for each LED. This tool is a graphical user interface (GUI) of standard LED
products from a series of LED manufacturers. Using this tool the color mixing of red, green, and blue LEDs are
presented, showing that there is a color shifts in the resulting distribution due to binning variation. Optics that have the
potential to alleviate the color shifts due to LED bin tolerances are presented.
High-Power-LEDs have reached a development stage that affords their reasonable application to general illumination.
Nonimaging total internal reflection optics (TIR optics) that generate non-rotationally symmetric light distributions are
proper components to preserve the advantages associated with this type of light source. Thus, high efficiency has to be
reconciled, e.g., with the use of freeform surfaces. This contribution investigates the development of TIR optics for LED-based
illumination applications. First, we consider rotationally symmetric TIR optics in order to illustrate their functional
principle and demonstrate some special design criteria. Second, we apply them to non-rotationally cases using the
tailoring technique. Finally, we illustrate various aspects of the design process with selected examples.
We propose an easy-to-handle method of surface parameterization which is valuable for enhancing the illumination
design process. Therefore, an additional normal vector function is provided defining the desired direction of the virtual
normal vector at each point of the surface. Hence, the reflection/refraction properties are seperated from the shape of the
surface. These surfaces are named Virtually Reflecting/Refracting Surfaces (VRS). This type of surface provides the
opportunity to alter the normal direction without changing the shape of the surface and vice versa. Therefore, the
designer can have a selective look at those quantities depending mainly on position or on direction. This means that, e.g.,
one can first prescribe the geometrical shape and adjust the surface's optical properties afterwards. Moreover, structured
surfaces, e.g., segmented reflectors, can be replaced in between by virtually reflecting surfaces in order to eliminate
discontinuities. This is apparently expedient for improving the convergence in automatic design. We investigate optical
layouts with VRS to demonstrate their impact on the design and the optimization process.
We developed a hollow light guide (HLG) model which allows to predict the best extractor combination for any given
HLG length and for typical constraints like mean illuminance values and uniformity. The HLG prototype we analysed
has a diameter of 30 cm, a 16 m length, 3M prismatic structures on the inner walls, curved light-reflecting surfaces with
different shapes working as light extractors. It is powered by a projector with a 400W HIT-lamp. Firstly, we
characterised extractors by their flux "consumption" and extraction. Then, we developed an analysis software to allow
real time acquisition of floor illuminance data by means of a cooled, calibrated photometric CCD camera. We thus
modeled the HLG as a lineal system of extractors. To evaluate both output flux and uniformity of the illuminance
distribution of an extensive set of solutions, we wrote a Simulation Software Extractor Position (SISEPO). SISEPO was
tested to optimise the extractor sequence of the initial HLG and a 15% improvement of the mean floor illuminance was
predicted. The corresponding measured values were in good agreement and also the measurements of the luminous
intensity distributions of the whole HLG-luminaire (EULUMDATA), carried out in individual segments of one meter
length, confirmed the improvement of the light extraction efficiency.
Mixing rods are used in illumination systems to provide a uniform output. The use of separate sources, especially RGB
LEDs, has renewed interest in the use of mixing rods. This paper describes some of the special issues that arise with the
use of RGB LEDs. Issues discussed include die layout for RGB LED arrays, straight round mixers with ripples and
RGB LEDs, the use of a diffusers with mixers, and the use of a square to rectangular mixer with ripples to circularize the
mixer's angular output.
Light emitting diodes (LED) have been increasingly used as light sources for projection display. LEDs have numerous
advantages as a light source for these applications especially when used with a digital micro-mirror spatial light
modulator such as the device offered by Texas Instruments. LEDs create an expanded color gamut, they can be
modulated at very high rates (thus, eliminating the need for a color wheel), and they last longer than other light sources.
One disadvantage they have is the luminous output is lower than would be desired for most front projection applications.
Smaller pocket projectors have used the LED sources successfully, but the luminous output is limited to between 25 and
100 lumens. One of the areas of light loss in the illumination system is in the multiplexer that combines the three
colored LEDs into a coaxial illumination beam. In this paper, this loss is quantified and an alternate multiplexer design
To compete with and to surpass the performance of traditional lighting systems, white LED development is still facing
the necessity of further improvements. An important topic that has to be addressed in this context is the spatial
homogeneity of the white light emitted, an issue that is directly associated with the geometry and the composition of the
color conversion elements (CCE) in phosphor converted LEDs. In order to avoid the need for experimental realization
and inspection of a large number of different configurations and compositions, optical simulation provides a time- and
cost saving alternative. In this contribution we discuss a simulation procedure which allows us to predict optimized
solutions for the CCEs in white LED light sources. The simulation process involves the set-up of a model for the blue
emitting LED chip and the implementation of a multitude of different geometries and compositions of individual CCEs
on top of the chip. Since the light is scattered within the CCEs, the respective scattering model, which considers the
phosphor particle size distribution and the phosphor weight fraction is of particular importance. In the final sequence of
the modeling procedure color uniformity is checked by monitoring the irradiance distributions both for the blue LED
light and the yellow converted light separately on a detector. From a comparison of the simulation results for a
significant number of different layouts we can deduce the impact of the individual materials parameters and predict
optimized CCEs which are finally compared with real device set-ups in order to verify the accuracy of the simulation
The importance of condenser optics is the fact that it is the bottleneck limiting efficiency in commercially available
projection systems. Conventional condensers use rotational symmetric devices, most of them being elliptic or parabolic
mirrors. They perform very far from the theoretical limits for sources such as arc lamps or halogen bulbs. Typical small
displays in the 5-15 mm2 etendue range have geometrical efficiencies about 40-50% for the best condensers; although
theory allows about 100% (no reflection nor absorption losses are considered). The problem is in the coma aberration of
the reflectors and the rotational symmetric image of the source making the source projected image to unfit with the
target. Thus, the only way to improve this performance is to generate a free form design that is able to control the shape
and rotation of the source projected images. As yet, this can only be done with the SMS3D design method. We present
here one of such designs achieving a collection efficiency 1.8 times that of an elliptical condenser for a 4:1 target aspect
ratio and for the range of target etendue with practical interest and 1.5 for 16:9 target. These designs use only 1
additional reflection, i.e., use a total of 2 reflections from the source to the target. A prototype of one type of free form
condenser has already been built.
As first generation extreme ultraviolet lithography (EUVL) systems go into the hands of lithographers, the obstacles to
full production are reducing in number and scope. Chief among these remaining obstacles, however, is the source
power. Although source manufacturers have made remarkable recent progress, power levels are still far below what is
needed for production tools. This makes the importance of an efficient illumination system even more so as Moore's
law pushes the industry to resolutions requiring EUVL. This paper describes some of the special issues for EUVL
illumination systems, discusses several modeling tools appropriate to EUVL illumination systems, with examples based
on fly's eye and debris mitigation systems found in the literature.
Discharge lamps serve a wide variety of applications and outperform novel light sources such as LEDs in terms of
luminous flux and luminance. Unfortunately, such lamps occasionally show arc movements (flicker) which change the
amount of light that is coupled into an optical system. A variety of measures in lamp design can suppress flicker
tendencies of a lamp but arc movement cannot be totally avoided.
In our contribution, we show that the way how the light is collected considerably influences the impact of flicker on the
collected luminous flux. We investigate light collection sensitivity of an illumination system as a function of the etendue
and of the particular realization of the illumination system. As a result, flicker sensitivity can be substantially reduced at
the expense of collection efficiency.
Plastic Optical Fibres (POF) were developed almost 3 decades ago. They are mainly used for short haul data
communications (up to 1 km with data rates up to 1 Gbps). Over the years, POF has found applications in many other
areas including solar energy transport for illumination. In such an application, light is collected from the sun and is
directed into a space which needs to be illuminated. The use of fibres and more specifically POF, in daylighting
systems, started only a few years ago. Several approaches have been investigated and we have seen the development of
a few commercial products. The market however, has not really taken off for these technologies simply because of their
enormous price tags. It is important to note that the use of POF in these designs has been limited to the function of POF
as the transmission medium only. We propose a novel solar illumination technique using POF as both the light
collecting/concentrating mechanism and the transmission medium. By modifying the structure of the fibre, solar light
can be directed into the fibre by using an analogous process to fibre side emission but, in the reverse. We shall report on
the solar light capturing efficiency of POF as modified by several types of external imperfections introduced onto the
fibre. One major advantage of our proposed approach lies in the fact that we aim to eliminate at least one of the two
axes of sun tracking that is currently used in existing solar illumination systems.
An innovative optical panel provided with an hexagonal array of refractive lenses having a properly optimized doublecurvature
profile has been simulated by ray-tracing and fabricated by injection-moulding. Such lenses are constituted by
a concave profile (having negative curvature radius) on their bottom and a convex profile (having positive curvature
radius) on their top. We demonstrate that, if compared to refractive elements with conventional geometry, bell-shaped
microlenses allow to collect incident rays within a wider angular range (so reducing the number of rays lost by TIR )
and to properly re-direct them. When installed on fluorescent tubes-based professional lighting systems, such refractive
elements allow to reduce undesirable glare as prescribed by EN12464-1 Interior Lighting Design Standards and to finely
control photometric outputs of luminaries.
Besides, bell-shaped microlenses-based films were also simulated to be applied onto a bottom-emitting lambertian
OLED. We demonstrate that, by properly tailoring both concave and convex profile shape, it is possible to increase the
outcoupling efficiency, as well as the luminous flux emitted by the exit surface, by a factor up to 1.95 and, at the same
time, to produce far-field photometric outputs characterized by uniform isocandela distribution maps with an aperture
angle up to 60°.
Light beam distortion from multiple reflections inside a luminaire is corrected by adding simple curvatures to different
reflective surfaces, separating the management of the vertical and horizontal components of the beam. The luminaire
directs the beam into the trajectory and convergence angle required for a digital ophthalmoscope, with a particular task
of maximizing the contrast ratio in the image by reducing visible corneal reflectance of the light source.
The Köhler illumination concept was originally invented to achieve uniform illumination in microscopy1. Köhler
integrators can also be formed by arrays of lenticulations that can be any combination of reflective and/or refractive
surfaces, organized in corresponding pairs. Arrays of integrating facets can be arranged not only on flat surfaces but on
rotationally symmetric and even freeform surfaces6. Currently flat lenslet arrays are widely applied as homogenizing
optics2 for lithography, machine vision illumination, and projection.
Adding Köhler facets onto already designed surfaces can improve the optical system performance, while respecting its
original function. In general, the optics output can be made somewhat independent of the source characteristics, although
at the expense of a slight ètendue dilution or efficiency losses.
This work revises the Köhler concept and its application to different kind of optics, ranging from photovoltaic
concentrators to automotive LED headlights. In the former, irradiance peaks on the solar cell can be avoided, while
preserving high aiming tolerance (acceptance) of the solar concentrator. In the latter, LEDs drawbacks like large source
image sizes, source misalignments, ill defined source edges, and low source radiance can be compensated.
Applications requiring high lumen packages are traditional the domain of light sources like discharge lamps. Currently,
LEDs make their way into such applications. LED street lighting projects, which are regularly covered in the press,
provide a case of point.
Life time and luminous efficacy are considered as being the main advantages of LEDs. Nonetheless, other current light
sources for street lighting have similar performance. Analysing a street-lamp as a complete system, we can show that
LED solutions have significant advantages if highly efficient optics are used. We present an example with tailored free-form
optics. These make efficient use of the valuable LED light by exactly redistributing it into the desired illuminance
Reflectors for prescribed intensity or illuminance distributions are commonly used in luminaires and automotive
headlights. However, their design has remained a challenge. Although reflector shapes are well known for a point
source, the point source approximation leads to significant errors in the output distribution with extended and nonuniform
source distributions. In practice, source non-idealities usually require the use of automated and manual
optimization in the design process. We propose a hybrid optimization scheme that generates reflector shapes for
rotationally symmetric systems and an arbitrary source luminance distribution. This method is based on an extension of
the traditional algorithm for a point source, and uses a combination of an iterative algorithm and an optimization
algorithm. Several case studies are presented.
A white light source, based on illumination of a yellow phosphor with a fibre-coupled blue-violet diode laser, has been
designed and built for use in endoscopic surgery. This narrow light probe can be integrated into a standard laparoscope
or inserted into the patient separately via a needle. We present a Monte Carlo model of light scattering and
phosphorescence within the phosphor/silicone matrix at the probe tip, and measurements of the colour, intensity, and
uniformity of the illumination. Images obtained under illumination with this light source are also presented,
demonstrating the improvement in illumination quality over existing endoscopic light sources. This new approach to
endoscopic lighting has the advantages of compact design, improved ergonomics, and more uniform illumination in
comparison with current technologies.
The purpose of the secondary optical design for light emitting diodes (LED) illumination is to rearrange the direct output
of LED and then achieve the desired illumination. Usually a lens is applied to realize this function in an LED
illumination system, and a freeform lens can do better than the traditional spherical ones. The method in this paper can
be used to design a freeform lens in short time, just less than 10 seconds. And this freeform lens is constructed using the
numerical solutions of a set of first-order partial differential equations, which are deduced from the Snell's law according
to the conservation of energy. Using an LED as the source while immersing it in the lens, a uniform rectangle can be
gotten through single refraction, and with uniformity near to 90%. The rectangular illumination also has a relatively clear
cut-off line with little blur at the edge. This method can shorten the designing time and improve the performance of LED
illumination system. Furthermore, not only rectangle, but other illumination figures can be achieved by freeform lens
designed by this method, which can broaden the scope of freeform lens's usage.
This paper develops a design method of the prism pattern for a LCD light guide plate to improve the uniformity of its exiting light. Firstly, the LGP prism surface is divided into several equal regions. With the aid of ASAP simulation, this method uses the mean light flux of all regions as a reference value to adjust the distribution density of the prism pattern for each region. Then curve fitting is performed to provide a smoothly changing distribution density for further improve-ment of the exiting light uniformity. ASAP results demon-strate that illuminance uniformity for the study case is sub-stantially improved from 45% to 90.9% by using this design method.