This thesis discusses the design, construction and testing of a highefficiency, 100 MHz, 1 kW, Class-E solid state power amplifier. The design was performed with the aid of computer simulations using electronic design software (ADS). The amplifier was constructed around Ampleon's BLF188XR LDMOS transistor in a single ended design. The results for 100 MHz operation show a power added efficiency of 82% at 1200 W pulsed power output. For operation at 102 MHz results show a power added efficiency of 86% at 1050 W pulsed power output. Measurements of the drain- and gate voltage waveforms provide validation of Class-E operation.

In this paper a prototype spectrum-splitting photovoltaic system based on volume holographic lenses (VHL) is designed, fabricated and tested. In spectrum-splitting systems, incident sunlight is divided in spectral bands for optimal conversion by a set of single-junction PV cells that are laterally separated. The VHL spectrum-splitting system in this paper has a form factor similar to conventional silicon PV modules but with higher efficiencies (>30%). Unlike many other spectrum-splitting systems that have been proposed in the past, the system in this work converts both direct and diffuse sunlight while using inexpensive 1-axis tracking systems. The VHL system uses holographic lenses that focus light at a transition wavelength to the boundary between two PV cells. Longer wavelength light is dispersed to the narrow bandgap cell and shorter wavelength light to the wide bandgap cell. A prototype system is designed with silicon and GaAs PV cells. The holographic lenses are fabricated in Covestro Bayfol HX photopolymer by 'stitching' together lens segments through sequential masked exposures. The PV cells and holographic lenses were characterized and the data was used in a raytrace simulation and predicts an improvement in total power output of 15.2% compared to a non-spectrum-splitting reference. A laboratory measurement yielded an improvement in power output of 8.5%.

Techniques for adapting metamaterials for the improvement of RF energy harvesting and infrared photodetection are demonstrated using experimental and computer simulation methods. Two methods for RF energy harvesting are experimentally demonstrated and supported by computer simulation. In the first method, a metamaterial perfect absorber (MPA) is made into a rectenna capable of harvesting RF energy and delivering power to a load by soldering Schottky diodes onto connected split ring resonator (SRR) structures composing the planar metasurface of the perfect absorber. The metamaterial rectenna is accompanied by a ground plane placed parallel to it, which forms a Fabry-Perot cavity between the metasurface and the ground plane. The Fabry-Perot cavity stores energy in the form of standing waves which is transferred to the SRR structures of the metasurface as AC currents that are rectified by the diodes to create DC power. This type of design enables highly efficient energy harvesting for low input power, creates a large antenna capture area, and uses elements with small electrical size, such that 100 uW of power (enough to operate simple devices) can be captured at ambient intensities ~ 1 - 2 uW/cm2. Two designs using this method are presented, one that operates for linear polarizations at 0.9 GHz and a smaller polarization-independent design that operates around 1.5 GHz. In the second method, the energy stored in the standing waves of an MPA Fabry-Perot cavity is instead harvested by placing a separate energy harvesting antenna within the cavity. The cavity shapes and enhances the incident electric field, and then the separate energy harvesting antenna is designed to be inserted into the cavity so that its shape and/or radiation pattern matches the electric field lines within the cavity and maximally extracts the stored energy. This method allows for great customization of antenna design parameters, such as operating frequency, polarization dependence, and directionality, by swapping out different metasurface and antenna designs. Using this method, the amount of power harvested by a simple dipole rectenna placed within a cavity is improved by a factor of 18 as compared to what it would harvest by itself at an ambient intensity of 35 nW/cm2. Lastly, the addition of plasmonic structures to DWELL (quantum dot-in-a-well) infrared photodetectors is investigated by computer simulation. DWELL photodetectors have the potential to one day replace standard mercury cadmium telluride detectors by being cheaper alternatives with a higher operating temperature. The inclusion of gold plasmonic structure arrays into DWELL detectors enables excitation of surface plasmon polariton modes that increase the responsivity of the detector to incident infrared radiation. The peak responsivity of a DWELL detector is demonstrated to improve by a factor of 8 for a 1 um thick layer of plasmonic structures and by a factor of 15 for a 2 um thick layer. These works are steps forward in making RF energy harvesting practically useful and for improving infrared photodetector performance.

The efficiency of an internal combustion (IC) engine is governed by the thermodynamic cycle underpinning its operation. The thermodynamic efficiency of these devices is primarily determined by the temperature gradient created during the compression process. The final conversion efficiency also known as brake thermal efficiency (BTE) of IC engines, however, also depend on other processes associated with its operation. BTE is a product of the combustion, thermodynamic, gas-exchange, and mechanical efficiencies. The improvement of BTE through maximation of any one of the four efficiencies is reduced by its implication of the other three. Split-cycle engine provides an alternative method of improving the engine efficiency through over-expansion of combustion gases by transferring it to a cylinder of greater volume. The operation of split-cycle engines is based on either the Brayton or the Atkinson Cycles. Atkinson Cycle has been demonstrated in IC engines without the split-cycle architecture but is limited by the reduced energy density. Double Compression Expansion Engine (DCEE) provides a method of accomplishing the Atkinson Cycle without the constraints faced in conventional engine architectures. DCEE splits the compression and expansion processes in a vertical manner that enables the use of larger cylinder volumes for over-expansion as well as first-stage compression without much friction penalties. The present thesis explores the thermodynamic cycle of this novel engine architecture using well-validated 1-dimensional engine models solving for gas-exchange, real gas properties, and heat transfer provided in the GT-Power software tool. The effect of compression ratio, rate of heat addition, sensitivity to design and modeling parameters was assessed and contrasted against conventional engine architecture. The synergies of combining low-temperature combustion (LTC) concepts with DCEE was investigated using simulation and experimental data. DCEE relaxes many constraints placed the operation of an engine in Homogenous Charge Compression Ignition (HCCI) mode. The limitations of adopting Partially Premixed Combustion (PPC) concept is also alleviated by the DCEE concept. BTE improvement of above 10% points is achievable through the DCEE concept along with possibility to achieve very low emissions through use of LTC concepts and new after-treatment methods uniquely available to the DCEE.

The topic of master thesis is the thermal calculation and proportion design of gas-fired steam boiler. The first part describe technical characteristics and parameters of the boiler. The main part of this thesis is the thermal calculation of the boiler.

Diese Arbeit behandelt neue Konzepte zur Realisierung hocheffizienter Weißlicht emittierender organischer Leuchtdioden (OLEDs), wobei blaue fluoreszierende Emitter mit grünen und roten phosphoreszierenden Emittern kombiniert werden. Bisherige Ansätze zur Erreichung höchster Quantenausbeuten basieren auf der ausschließlichen Verwendung phosphoreszierender Emitter, da diese prinzipiell 100% der elektrisch erzeugten Exzitonen in Licht umwandeln können. Allerdings sind speziell OLEDs mit phosphoreszierenden tiefblauen Emittern heutzutage nach wie vor nicht langzeitstabil. Andererseits gibt es zwar sehr stabile fluoreszierende Emitter auch im tiefblauen Spektralbereich, jedoch kann eine rein fluoreszierende OLED aus spinstatistischen Gründen maximal nur ein Viertel der erzeugten Exzitonen in Licht umwandeln. Für eine ernsthafte Verwendung von OLEDs als Lichtquellen sind sowohl die Umwandlungseffizienz elektrischer Leistung in Lichtleistung im sichtbaren Spektralbereich, als auch ihre Langzeitstabilität entscheidend. Ein Kompromiss lässt sich daher mit der Kombination von blauen fluoreszierenden Emittern mit grünen und roten phosphoreszierenden Emittern erzielen. Die beiden in dieser Arbeit entwickelten Konzepte unterscheiden sich in der energetischen Lage des Triplettniveaus des jeweils verwendeten fluoreszierenden blauen Emitters relativ zu den verwendeten phosphoreszierenden Emittern. Das erste Konzept verwendet einen fluoreszierenden blauen Emitter mit niedriger Triplettenergie, weshalb er bei direktem Kontakt mit den phosphoreszierenden Emittern deren Phosphoreszenz löscht. Eine Exzitonen blockierende Zwischenschicht unterdrückt diesen Verlustmechanismus. Dies wird sowohl in Photolumineszenzexperimenten als auch in OLEDs nachgewiesen. Weiterhin muss die Zwischenschicht gleichzeitig die Exzitonengeneration auf beiden Seiten gewährleisten, sie muss also bipolare Transporteigenschaften haben. Mischschichten aus einem Elektronen transportierenden und einem Löcher transportierenden Material werden mit der Methode der raumladungsbegrenzten Ströme in unipolaren Strukturen untersucht, um ihren Einfluss auf die Ladungsträger- und Exzitonenbalance in OLEDs zu erklären. Das zweite Konzept verwendet einen fluoreszierenden blauen Emitter mit hoher Triplettenergie. Dadurch ergeben sich einige Vorteile. Phosphoreszenz wird nicht mehr gelöscht, weshalb keine Zwischenschicht mehr notwendig ist. Zusätzlich können außerdem die auf dem blauen fluoreszierenden Emitter erzeugten Triplettexzitonen für die Lichtemission verwendet werden, indem man sie auf die phosphoreszierenden Emitter überträgt. Damit ist es grundsätzlich möglich, 100% der elektrisch erzeugten Exzitonen für die Lichtemission zu verwenden, obwohl ein fluoreszierender Emitter verwendet wird. Allerdings ist dabei darauf zu achten, dass die Singulettexzitonen nicht ebenfalls übertragen werden, da sonst kein Weißlicht mehr erzeugt werden kann. Es werden verschiedene OLED-Strukturen untersucht, um Singulett- und Triplettexzitonen so auf die jeweiligen Emitter zu verteilen, dass eine ausgewogene spektrale Balance der Emission erreicht wird. Ein zentraler Punkt ist dabei die Ausnutzung der unterschiedlich großen Diffusionslängen von Singulett- und Triplettexzitonen.

This project proposes a new hybrid voltage divider DC-DC converter that utilizes switching capacitors and inductors to produce zero voltage switching (ZVS) at the turn on state of its switches. By achieving ZVS, the switching losses are significantly reduced; thus, increasing the overall efficiency of the converter at various loads. The goal for this thesis is to perform analysis of the operation of the converter, derive equations for sizing the main components, and demonstrate its functionality through computer simulation and hardware prototype. Results of the simulation and hardware testing show that the proposed converter produces the desired output voltage while providing the zero voltage switching benefits. The converter’s efficiency reaches above 92% starting from 1A load and continues to increase to 97.6% at 4A load. Overall, results from this thesis verifies the potential of the proposed converter as an alternative solution to achieve a very efficient DC-DC solution when half of the input voltage is required at the output without the use of complex feedback control circuitry.

No description available.

No description available.

En esta Tesis se estudian las etapas que componen un amplificador de audio conmutado o clase D. El enfoque abarca desde el acondicionamiento y procesamiento digital de la señal hasta el transductor pasando por la etapa de potencia. La investigacion se centra en el analisis y/o compensación de la distorsión generada en las diferentes etapas que atra-viesa la señal. El desarrollo es teórico-práctico. Varios esque-mas de modulación por ancho de pulso son estudiados en el dominio frecuencial. Se analiza además el efecto de los tiempos muertos necesarios entre el encendido y el apagado de los semiconductores de potencia en el contenido espectral de la señal. Se muestra que establecen un límite en la distor-sión que no puede reducirse incrementando la frecuencia de la portadora, o cambiando la técnica de modulación. Se estudian e implementan una serie de algoritmos de procesamiento digital de señales para la reducción de la distorsion generada por la modulación y por la etapa digital debido a la utilización de un procesador de punto fijo. Se ensayan técnicas de sobremues-treo, decimación, interpolación, moldeo del ruido de cuantiza-ción y esquemas de modulación digital por ancho de pulso, y se reportan resultados experimentales medidos con un anali-zador dinámico de espectros. Estas herramientas se aplican en el diseno de un amplificador conmutado a lazo cerrado que mantenga una presión acústica constante en un determi-nado rango de frecuencias. Se discuten distintos modelos del parlante que vinculan las variables acústicas con las eléctricas, y se diseña el lazo de realimentación lineal discreto tomando como variable de salida la aceleración del cono del parlante. / In this thesis a study of the stages comprising a class D or switching amplifier is made. The approach ranges from the signal conditioning and digital signal processing stages up to the transducer, also covering the power stage. The research focuses on the analysis and compensation of the distortion generated in the different stages through which the signal passes. The development is theoretical/practical. Various pulse width modulation (PWM) schemes are studied in the frequency domain. The effects on the spectral content of the PWM signal with dead times, necessary between the on and off states of the power semiconductor devices, are analysed. A bound in the total harmonic distortion, that cannot be reduced by increasing the carrier frequency or changing the modulation technique, is shown. Some digital signal processing algorithms for the reduction of distortion, generated by the modulation and the digital stage due to the use of a fixed point processor, are studied. Oversampling, decimation, inter-polation, cuantization noise shaping and digital pulse width modulation schemes are investigated and experimental results obtained with a dynamic signal analizer are reported. These tools are applied to the design of a closed loop switching amplifier that holds the acoustic pressure constant in a determined frequency range. Different models of the louds-peaker which link the electric and acoustic variables are discussed. Finally a linear, digital control, feedback loop that uses the acceleration of the speaker cone as ouput is designed.