 |
 |
EPE Journal Volume 21-2 - Papers
- Using the Chip as a Temperature Sensor - The Influence of Steep Lateral Temperature Gradients on the VCE(T)-Measurement
- COTS Circuit Predicting Coil Zero-Current Crossing for Early Thyristor Triggering in a Fast, High-Power Magnetic Generator
- Experimental Study of the Shoot-Through Boost Control Methods for the Z-Source Inverter
- Generalized Algorithm for Pulse Width Modulation using a Two-Vectors Based Technique
- A Novel Hysteresis Direct Torque Control for Matrix Converter Drives
EPE Journal Volume 21-2 - Editorial
By B. Sneyers
EPE 2011 welcomed over 1000 participants in Birmingham from 30 August to 1 September. As the result of the long negotiation process that ended with the contract’s signature reported in EPE Journal Vol. 20 no 4, this conference was the first to be organized in the new framework established between EPE and IEEE-PELS.
EPE Journal Volume 21-2 - Papers
By Ralf Schmidt; Uwe Scheuermann
During operation steep lateral temperature gradients evolve in IGBT power semiconductor chips. The influence of these lateral gradients on the measurement of the virtual junction temperature by means of the widely used VCE(T)-method was investigated. In particular we address the question, how the obtained single temperature value is related to the temperature distribution of the chip. A combination of electrical and thermal measurements together with thermal simulations was performed to understand the implicit averaging mechanisms of the VCE(T)-measurement. It is found that the lateral temperature gradient in the chip results in an inhomogeneous sense current distribution during the measurement. This current distribution is responsible for the formation of the measurement value and its corresponding temperature T*. A comparison of experimental and simulation results shows that for currently existing IGBTs, T* corresponds to the areaweighted average of the active chip area. The maximum imbalance in sense current density during the VCE(T)-measurement was determined for maximum load condition on a high performance cooling system to be 150 % and 50 % of the average current density for the central and the corner area of the chip, respectively. Furthermore, the temporal evolution of the temperature profile and its influence on the thermal impedance measurement are discussed. It is shown that the temperature at the chip center evolves with a smaller thermal time constant (i.e., faster evolution) than at the chip corners.
By R. Charlet de Sauvage; Ph. Dondon; I. Lagroye; B.Veyret
This paper describes the design and construction of a zero-current crossing detection circuit for a high-power (6000 A, 650 V) magnetic generator delivering short (340 μs), high-intensity (2 Tesla) pulses, used to stimulate living tissues. The magnetic pulses are produced by discharging a capacitor Cd into a coil in resonant mode. Half-sine or full-sine waveforms may be selected. When full-sine current into the coil is selected, a smooth transition between the 2 half-sines is required to ensure continuity in tissue excitation, i.e. the second thyristor must be triggered shortly before the first one switches off. The operating theory is to compute the period of the resonant circuit in advance, irrespective of interchangeable coil inductance and damping factors. The system core consists of a toroid sensor encircling the high-current main wire of the coil and an integrator circuit that is switched on or off by three events: the starting time and zero-crossing time of the sensor voltage and the zero-crossing time of the potential across Cd. This makes it possible to detect the pulse period right from the end of the first quarter period. Details of the robust COTS (commercial off-the-shelf) electronic design are given. Electric tests fully validated this design. The final section discusses the results and possible improvements.
By Omar Ellabban; Joeri Van Mierlo; Philippe Lataire
This paper presents a simulation and experimental comparative analysis of the Z-source inverter (ZSI) with four different shoot-through (ST) control methods, namely: the simple boost control, the maximum boost control, the maximum constant boost control and the modified space vector modulation boost control methods. A review of these methods is presented with a summary of all expressions. A prototype of a 30 kW ZSI is designed and implemented. The eZdspâ„¢ F2808 evaluation board is used for the realization of the shoot-through control methods and the real time workshop (RTW) is used for automatic code generation. The paper compares between the different four shoot-through control methods in terms of: the line voltage harmonic, the phase current harmonic, the d.c. link voltage ripples, the switch voltage stress, the inductor current ripples, the obtainable a.c.output phase voltage and the overall efficiency with detailed simulation and experimental results. In addition, the paper presents the effect of varying the d.c. input voltage and the shoot-through duty ratio on the overall inverter efficiency. The maximum constant shoot-through boost control method seems to be the most suitable shoot-through control method for the ZSI. Also, it is shown that the efficiency of the ZSI improves with increasing the d.c. input voltage and degrades with increasing the shoot-through duty ratio.
By José Restrepo; José M. Aller; Alexander Bueno; VÃctor M. Guzmán; MarÃa I. Giménez
This paper presents a generalized and compact space vector modulation algorithm valid for both (a, b, c) and (x, y) coordinates, suitable for triangle comparison modulators. The proposed algorithm uses a two base-vectors duty cycle computation by mapping the solution space into only three parallelogram shaped zones (Fig. A). This generalized algorithm is especially suited for vector control applications requiring high dynamic response and for applications where changes in demand may happen at a frequency comparable with the modulator's carrier frequency. Several carrier base modulation methods extensively described in the literature are obtained from the proposed generalized space vector modulation algorithm with the introduction of the null vector ratio d. The proposed method has been experimentally tested, and a practical implementation of the generalized algorithm with low computational requirements is presented.
By D. Xiao; M. F. Rahman
In this paper, the application of classical hysteresis direct torque control (HDTC) to the IPM synchronous machine fed by a matrix converter is investigated and one of the problems associated with this scheme is highlighted. A modified HDTC scheme for matrix converter drives is proposed to solve this problem by applying two switching combinations at each sampling period, which allows significant reduction of input current harmonics without any adverse effect on the drive performance or increasing the complexity of the system. The fast dynamics is preserved in the proposed scheme by applying a single current vector in the rectifier stage and thus the maximum line-to-line input voltage is utilized to generate the voltage vector in the inverter stage. A novel input power factor correction strategy is proposed to compensate the effect of input capacitors on the displacement angle by introducing a power-related displacement angle in the duty cycle calculations. The joint stator flux and rotor speed estimation is implemented with an adaptive sliding mode flux observer in the estimated rotor reference frame to enhance the robustness and reliability of the drive system. The comparisons between the classical and proposed HDTC schemes for matrix converter fed IPMSM machine have been carried out by both simulation and experiments.
EPE Journal Volume 21-2: Other
By B. Sneyers
A description of the Awards Ceremony at EPE 2011 – ECCE Europe