Korneel Wijnands

Modular multilevel converters with non-sinusoidal ac voltage output can reduce cost and volume in medium-voltage-connected electric vehicle battery charging applications. The use of full-bridge submodules in such converters enables single-stage ac/ac voltage conversion, allowing a medium-voltage grid to be directly connected to a medium-frequency isolation transformer. The application of a square wave voltage at the medium-frequency transformer’s single-phase port enhances the converter’s efficiency and power density in comparison to a sinusoidal voltage. This paper presents the analysis and modelling of a modular multilevel converter, comparing its operation with sinusoidal and square wave output voltages. A single control scheme for both output voltage waveforms is proposed for the three-phase and single-phase ac currents, circulating currents, and the energy stored in the submodule capacitors. The control strategy of the three-phase and single-phase port currents is verified through simulation and experiments using a scaled-down prototype, thereby validating its suitability for high-power bidirectional battery chargers.

Accurate calculation of semiconductor losses and temperature is the foundation of any design methodology for a power electronic converter. Computation accuracy and speed play a vital role if a large set of parameters needs to be considered. Averaged loss models often neglect the temperature dependence of transistors, leading to fast, but inaccurate results. In contrast, iterative methods and simulation tools, which can include temperature dependence, take significantly longer to compute, but yield more precise results. This paper presents a best of both worlds approach, by using the harmonic balance method to obtain the steady-state solution for any inverter topology including temperature dependent conduction and switching losses. The proposed method solves for the discrete Fourier series of the device temperature, by expressing the temperature dependence and operating parameters in the frequency domain. The set of equations for each coefficient is solved by a single matrix inversion, resulting in very fast computation for steady-state temperature cycles. The steady-state operation of over one thousand possible inverter designs is calculated within less than one minute, matching iterative simulation in device temperature, conduction and switching losses, at a fraction of the computation time. In addition, the method shows good agreement with temperature measurements of a three-phase silicon-carbide inverter.

This work presents a closed-form analytic approach to calculate the distortion introduced in resistive current measurements due to thermal properties of shunt resistors. As a shunt’s resistance varies with temperature, the sensed voltage becomes distorted. This non-linear effect depends on the thermal interface, dissipated power and material properties of the shunt resistor. It is presented, how to simplify the non-linear equation of the shunt temperature and calculate the resulting voltage spectrum, without the use of computationally extensive time-to-frequency-domain transforms. The results of the presented method match simulations including the non-linear behaviour, enabling a quick search through a solution space. The method is experimentally verified by use of a linear amplifier with an output distortion of -90 dB at low frequencies. In the experiment, a total harmonic distortion of up to -70 dB is observed in several shunt resistors covering a range of different thermal properties. Based on the presented method, shunt resistors can be chosen according to the minimum distortion required in a given application. Depending on the frequencies of interest, shunt resistors with a high temperature coefficient might still offer sufficient distortion performance.

For atomic scale plasma processing involving precise, (an)isotropic and selective etching and deposition, it is required to precisely control the energy of the plasma ions. Tailored waveforms have been employed to bias the substrate table to accurately control this ion energy. Recent research has shown that switched-mode power converters can be used to generate this kind of waveform, with the benefit of increased energy efficiency and flexibility compared to the traditionally used linear amplifiers. In this article, an improved equivalent electric circuit model of the plasma reactor is proposed to allow simulation and bias waveform optimization. The equivalent electric circuit is analysed for different process phases, including the charge, discharge, and post-discharge phase. The proposed model is suitable for electric circuit simulation and can be used for predicting the electric waveforms and ion energy distributions. Plasma parameters are required as input for the model, thus an empirical parameter identification method based on the electrical measurements of the bias voltage and output current waveforms is introduced. Since these electrical measurements do not interact with the plasma process, the proposed parameter identification method is nonintrusive. Experiments have been carried out, which demonstrate that the proposed model and parameter identification method provide the expected accuracy.

The distortion of an H-bridge inverter can originate from many nonlinear factors, of which the most dominant one is typically the dead-time. In previous research, analytical models have been proposed to calculate the output spectrum of the H-bridge inverter. However, the existing models have not fully explored the effects of soft-switching and discontinuous conduction mode, which might severely degrade their accuracy. In this paper, the characteristics and the corresponding effects of these switching modes are analysed. Constraint functions are proposed in order to determine the time position of each switching mode. Based on this analysis, a numerical model of the output spectrum of the H-bridge inverter is then introduced. The model is shown to have a significantly improved accuracy compared to the previous analytical models especially for large inductor current ripples. The accuracy improvement is confirmed by simulation and measurements on an experimental prototype.