UittrekselThe Solid State Lighting market shows these days a trend to the usage of high flux packages with higher efficacy and higher Color Rendering Index, realized by means of multiple color chips and phosphors. The resulting white LEDs can be several centimeters in size. When it comes to the design of optics for such sources, the demand for compact and efficient luminaires implies that well-known design methods employing a point-source approximation are often not applicable. Thus, today’s research in illumination is strongly focused on design techniques for extended emitters. In my PhD work, I have focused on new illumination optical design methods for applications in which the spatial extent and non-uniformity of LEDs play an integral role.
I first concentrated on the problem of dealing with non-uniform sources. In 2012 Light Prescriptions Innovators (LPI) developed a specific concept, called “Shell-Mixer”, for performing color mixing of extended multi-color LEDs. The Shell-Mixer solution consists or arrays of Köhler integration lenses arranged on a hemispherical base cap, meant to be located on top of the multi-color chip array; this configuration creates a virtual source which replaces the original LED, ith the same size and position but with uniform color. A first version of this technology (“spherical Shell-Mixer”) was designed by LPI in 2012, using spherical surfaces designed in paraxial approximation. The resulting color mixing and efficiency are statisfactory, but the overall size (three times the size of the light source, in diameter) needs to be reduced in order to git into more types of luminaires. I have addressed the problem of designing a new Shell-Mixer with reduced size, equalling the performance of the original sphericall shell. I have implemented advanced concepts of Nonimaging and Freeform Optics to design a new freeform Shell-Mixer, whose size is just twice the diameter of the source to which it is applied. The efficiency and color mixing performance equal the previous spherical shell, with a volume reduction of about 70%. An actual prototype, developed within the framework of an FP7 EC-funded project, has undergone preliminary test which suggest that the freeform Shell-Mixer performs as expected from ray tracing simulations. The design techniques explored in this study can be applied to generic Köhler integration optics, yielding improved efficiency and reduced size.
In the second part of my research, I focused on an evolution of the Simultaneous Multiple Surface design method (SMS). This is one of the most effective direct design methods developed in the context of Nonimaging Optics. The SMS method has proven particularly effective in designing optics for extended emitters. In its three dimensional version (SMS3D) it allows to calculate N freeform surfaces connecting the input and output components of N freeform wavefronts. So far, SMS has been used to design two-surface optical devices, for applications like illumination and concentration photovoltaics. I have worked on the extension of SMS3D to the case of three simultaneously calculated surfaces, in the context of the so-called 3D RXI optic. This is a very shallow optic, which can be used for generating prescribed intensity patterns using extended sources. The result presented in the thesis is a novel RXI with three freeform surfaces controlling three freeform wavefronts at once, for the first time. The enhanced beam control capabilities of the so-called 3-SMS-surface RXI allow better control of extended sources and provide improved accuracy in the creation of prescribed illumination patterns. The test designs conducted during my analysis were inspired by real cases (like automotive lighting).
In the final part of my PhD, I concentrated on a promising method for designing beam-shaping optics for extended sources. In this approach, the optical design is achieved by paying attention to the wavefronts of an optical system. The extended source is described in terms of some input “edge” wavefronts; the intensity pattern is characterized by means of output edge wavefronts. The practical optical design is performed by requiring the optic to connect the input and output wavefronts. A crucial step in this approach is the description of intensity patterns in terms of the output wavefronts (“wavefront tailoring”). Only an approximate procedure for this task was available so far, which worked well for a restricted range of amission angles. I have worked on a generalization of the wavefront tailoring technique, which has no limitations in the emission angle range. A first proof-of-concept example proves the validity of the generalized wavefront tailoring approach in creating illuminance patterns which were difficult, if not impossible, to achieve with the previous approximate version. This first result is very promising and clearly indicates that the generalized wavefront tailoring method deserves further study.
|Datum Prijs||16 apr 2018|
|Begeleider||Fabian Duerr (Promotor), Peter Schelkens (Jury), Roger Vounckx (Jury), Wendy Meulebroeck (Jury) & Hugo Thienpont (Jury)|