Abstract: In this paper, a novel dead time compensation method is presented which produces inverter output voltages equal to reference voltages. The experimental result is also presented to demonstrate the validity of the proposed method. It shows that the compensation of the distorted voltage is possible up to sub-microsecond range. The reference voltage can be used as a feedback value, which is essential for sensorless vector control and flux estimation. The method is based on space vector PWM strategy and it can be carried out automatically by an inverter controller for initial set-up without any extra hardware. >

Abstract: An electric drive apparatus and method for controlling medium-voltage alternating current motors wherein a multi-phase power transformer having multiple secondary windings provides multi-phase power to multiple power cells, each of which have a single-phase output and are controllable by a modulation controller. The primary and secondary windings in the power transformer may be star- or mesh-connected; the secondary windings may be shifted in phase. Because the power cells are connected in series, the maximum output voltage for each cell may be less than the maximum line-to-line voltage. The power cells can have a tri-state output which can be controlled using pulse-width modulation techniques. AC input power is converted to DC output power for each output phase. Output power modulation can be produced by interdigitating a selected number of carrier signals so that harmonic components reflected back to the input are attenuated.

Abstract: A novel zero-voltage and zero-current-switching (ZVZCS) full-bridge (FB) pulse-width modulated (PWM) converter is proposed. The new converter overcomes the limitations of the zero-voltage-switching (ZVS)-FB-PWM converter, such as high circulating energy, loss of duty cycle, and limited ZVS load range for the lagging-leg switches. By using the DC blocking capacitor and adding a saturable inductor, the primary current during the freewheeling period is reduced to zero, allowing the lagging-leg switches to be operated with zero-current-switching (ZCS). Meanwhile, the leading-leg switches are still operated with ZVS. The new converter is attractive for high-voltage (400-800 V), high-power (2-10 kW) applications where IGBTs are predominantly used as the power switches. The principle of operation, features, and design considerations of the new converter are described and verified on a 2-kW, 100-kHz, IGBT-based experimental circuit.

Abstract: This paper investigates the power losses for two different IGBT technologies (nonpunch through and punch through) for use in PWM-VSI inverters in order to choose the right device technology for a given application. A loss model of the inverter is developed based on experimental determination of the power losses. The loss model is used on two different modulation strategies which are a sine wave with a third harmonic added and a 60/spl deg/-PWM modulation where only two inverter legs are active at the same time. The two IGBT technologies are characterized on an advanced measurement system which is described. The total power losses in the inverter are estimated by simulation at different conditions and it is concluded that the nonpunch through (NPT) technology is most useful for higher switching frequencies, while the punch through (PT) technology is especially useful at lower switching frequencies and high load currents. It is also concluded that the 60/spl deg/-PWM modulation has the lowest power losses and the power losses are almost independent of phase angle cos(/spl phi/) for normal motor operation. >

Abstract: The authors comment on the paper by F. Casanellas titled 'Losses in PWM inverters using IGBTs' (see ibid., vol.141, p.235-9, 1994). They suggest that Casanellas only gives a relatively coarse estimate concerning the reduction of the switching losses of discontinuous modulation in comparison to continuous modulation and expand on this subject. The original author replies to the comments.

Abstract: A novel vector PWM method for three-phase voltage-controlled inverters is described. The so-called minimum-loss vector PWM (MLVPWM) strategy is characterized by the minimum amount of switching losses incurred in the inverter switches. Comparative analysis proving superiority of the MLVPWM technique over the existing regular-sampling PWM methods, and results of experimental investigation of a prototype modulator are presented. >

Abstract: The paper presents a systematic approach to small-signal modeling, and control design of three-phase PWM converters. The well established design procedures used in DC-DC converter control design, are adapted for the three-phase converter control based on the similarity in the small-signal dynamics of the three-phase converters and their DC-DC counterparts. The approach is especially beneficial in three-phase rectifier control, which is reduced to a single-input single-output system after closing the current loops. The modeling approach is verified on a 10 kW three-phase boost rectifier switching at 15.6 kHz with DSP control. A wide-bandwidth output voltage control loop is then designed based on the verified small-signal transfer functions. >

Abstract: This paper presents a new PWM method for the three-level GTO inverters based on the space voltage vectors. With the proposed PWM method, we can minimize the harmonic components of the output voltage by avoiding the minimum pulse width limitation problem of the GTO thyristors and keeping the voltage balancing of the DC-link capacitors. The principle of the proposed PWM method and the voltage vector output sequences are described in detail. Computer simulation and experimental results verify that the proposed PWM method is suitable for high voltage and large capacity three-level GTO inverters applied to induction motor drives.

Abstract: A real-time current controller for PWM inverter-fed permanent-magnet synchronous motor drives is presented and analyzed. The proposed current control scheme is based on predictive control with a parallel integral loop added to compensate for the inaccuracy of the motor model and for the variations of motor parameters and DC voltage source. The proposed current control scheme is analyzed and its performance is evaluated by computer simulation. An EPROM-based implementation is presented in which calculations and pulsewidth modulation are executed by lookup tables resulting in high-speed operation. The controller performance is evaluated using a prototype l kW PM synchronous servo drive. Experimental results are given and discussed. >

Abstract: The PWM control scheme is illustrated in conjunction with a power distribution control circuit (30), wherein the duty cycle of convverter (32) is suitably controlled by a number of feedback loops embodying various parameters, for example a voltage error signal, an average (eg. RMS) value of the input voltage (Vin). Circuit (30) comprises a rectifying circuit (42), a converter circuit (32), for example an AC-DC converter, a capacitor (C), an oscillator and PWM circuit (44), an error signal (46), a multiplier circuit (48), a differencing amplifier (50), and a synthesizing circuit (52). Circuit (30) appears as a resistive load to the line for feed PWM, a fixed load, and a fixed RMS input voltage. To stabilize the output voltage (Vout), the pulse width signal (44a) applied to converter suitably varies in proportion to both changes in the RMS value of the input voltage and the output voltage. The output voltage error signal applied to multiplier circuit by error circuit provides the correction value for Vout variations. The line RMS signal applied to multiplier by the line RMS filter circuit (52) compensates the PWM signal for variations in the RMS value of Vin. In this regard, the RMS loop modulator control signal may be advantageously normalized by Vrms (square) since Pin varies as Vin (square). Decreases in either the output error signal or the RMS signal increase the PWM value.

Abstract: This paper presents a review of recently used current regulation techniques for voltage source pulse width modulated (VS-PWM) inverters. A variety of techniques, different in concept, are described as follows: on-off hysteresis free running and fixed frequency regulators (phase independent, look-up table based, space vector based); linear regulators (carrier based, working in stationary and rotating coordinates, PI and state feedback); predictive (minimum and constant switching frequency); and dead beat regulators. Also, recent trends in the current regulation-neural networks and fuzzy logic based regulators-are presented. Some oscillograms which illustrate the properties of the presented regulator groups are shown. >

Abstract: By adding a simple external commutating aid circuit to the full-bridge DC/DC converter with phase-shift control and by reducing the magnetizing inductance, optimum performance (i.e. ZVS operation and high conversion efficiency) can be achieved from full load down to almost zero load. An analysis is presented to determine the required and available commutation energies. The analysis is complemented with design considerations and experimental results taken from a 3-kW converter operating at 200-kHz clock frequency. >

Abstract: There is provided a display-integrated type tablet device which can achieve a high coordinate detection accuracy even when an electrode density of electrodes is greater than a electrode density of the other electrodes in a display panel. A segment electrode scanning clock signal for scanning segment electrodes having an electrode density three times as high as that of common electrodes has a frequency three times as high as a common electrode scanning clock signal. Shift data input to a segment electrode drive circuit has the same pulse width as that of the shift data input to a common electrode drive circuit. As a result, a scanning speed in an x-direction (shift speed of a segment electrode scanning signal) can be made approximately equal to a scanning speed in a y-direction (shift speed of a common electrode scanning signal). Furthermore, the width of the segment electrodes in an active state is made equal to the width of the common electrode. Thus a waveform of a voltage induced at a detection electrode of a detection pen in an x-coordinate detection period and a waveform of a voltage induced in a y-coordinate detection period are made approximately equal to each other to obtain a high coordinate detection accuracy.

Abstract: This paper presents a new topology of PWM rectifier which can achieve unity-power-factor on the AC supply side and ripple-reduction on the DC output side. The main circuit of this rectifier is accomplished by adding only a pair of switches to a conventional PWM rectifier. And it does not need the large DC capacitor or passive L-C resonant circuit. These additional switches and PWM rectifier are controlled not only to make a unity power factor but also to reduce the ripple current. The effectiveness of this circuit is confirmed by the experiments and analysis. This rectifier is useful for UPS and DC power supply, especially in case that the batteries are connected to the DC line. >

Abstract: A novel full-bridge (FB) zero-voltage-transition (ZVT) PWM DC/DC converter for high power applications is proposed. An auxiliary network which consists of two small switches and one small inductor provides zero-voltage-switching (ZVS) for entire line and load ranges without increasing device voltage and current stresses. Since the ZVS range is independent of the switch capacitance, IGBTs can be used instead of MOSFETs by adding external capacitors to the switches without increasing switching losses, which allows the proposed converter to handle higher power. Operation, analysis, and features are described and verified experimentally. A 1.8 kW ZVT PWM converter prototype is compared with a ZVS PWM converter prototype of the same power rating. >

Abstract: This paper studies all-digital discrete-time vector current control loops containing a voltage fed invertor in series with a permanent-magnet synchronous motor. Normal operation of the typical pulse width modulating voltage fed inverter introduces nonlinear distortion into the voltage and the current waveforms of the motor. This paper focuses on both the causes and the attenuation of these disturbances using all-digital control. An effective, fully-digital current controller is developed and experimental results from a Motorola 68020 microprocessor based system are presented. >

Abstract: First and second dc linked, preferably voltage sourced, inverters are connected to an electric power transmission line in shunt and in series, respectively, by coupling transformers. Firing of the GTO-thyristors of the first inverter is controlled to regulate reactive power on the transmission line and to supply real power requirements of the second inverter. The second inverter is controlled to inject into the transmission line in series a voltage which can be adjusted in magnitude and from zero through 360 degrees in phase with respect to the transmission line voltage to selectively adjust any one or more of transmission line voltage magnitude, transmission line impedance and transmission line voltage phase angle. In a high power configuration of the controller each of the inverters includes a plurality of six-pulse inverter modules connected by interstage transformers to form balanced higher order pulse groups controlled through pulse width modulation or through adjustment of the phase between opposite groups to generate the appropriate ac voltages.

Abstract: State-of-the-art pulsewidth modulation techniques fail to give satisfactory results when high-bandwidth torque control is required at the low switching frequency of modern high-power AC machine drives. A novel current control method refers to precalculated optimal synchronous pulse patterns. These are used to generate specific current reference trajectories to be adapted on-line to the actual dynamic torque command. A fast tracking controller, operated in parallel to a conventional PI controller, minimizes the trajectory tracking error. The optimized current trajectories ensure maximum response and minimum harmonic distortion. >

Abstract: The present invention provides a method and circuit for controlling the flow of current through a load. In a preferred embodiment, an oscillator generates a pulse signal of constant frequency. A pulse width modulator adjusts the duty cycle of the pulse signal in response to a dimming level signal input indicative of the desired level of current flow through the load. A converter receives the pulse signal as an input and converts it into an AC signal, the frequency of which follows the frequency of the pulse signal and the symmetry of which varies with the duty cycle of the pulse signal. The load is connected into a resonant circuit tuned such that a change in the symmetry of the AC signal changes the level of current flowing through the load.

Abstract: An apparatus (10) and method to find leakage due to tissue load or transients at the start or end of electrosurgery. Active and return electrodes (11 and 16) between a patient and an ESU (13) pass RF energy sensed by inductive transformers (17 and 19). A circuit finds leakage at more than two thousand times per second; that frequency depends on the phase shift (23) between voltage and current. Software and feedback (26) manage RMS voltage to reduce the peak voltage of the output wave or increase the crest factor by pulse width modulation of the RF drive. Phase shift (23) changes of the active and return current signals (18 and 20) and peak voltage and current are found. The phase angle υ is compared to a threshold and if greater than the frequency at which the differences (22) between the active and return current signals (18 and 20) are examined is increased. The differences (22) between the active and return current signals (18 and 20) are examined and if greater than a maximum for leakage while the mode selected is in coagulation then the pulse width of the RF drive is reduced to hold voltage wave-form peaks at a predetermined value while the RMS voltage is reduced to lower the leakage to a maximum or the frequency at which the leakage current is calculated is held to a maximum level until the phase angle υ is smaller than the threshold. If the difference (22) is greater than a maximum for leakage while the mode is cut or bipolar then the Vrms is reduced to a maximum level or the level remains high until the phase angle υ is smaller than a threshold. The crest factor is increased by reducing the duty cycle or the pulse width of the output wave shape.

Abstract: This paper presents the hybrid pulsewidth modulation (HPWM) method which requires only two of the four switches in a full-bridge inverter to be pulsewidth-modulated at high frequency, thus effectively reducing the switching loss by a factor of two. For triangular carriers, the HPWM frequency spectrum is identical to the conventional unipolar PWM (UPWM) spectrum. A low-frequency model for a fast-switching HPWM full-bridge inverter with high-quality output is described, and is substantiated by experimental data. >

Abstract: A high power IGBT tester, based on the typical clamped inductive load circuit and incorporating zero-voltage-transition and zero-current-transition soft-switching schemes, has been developed. Both turn-on losses, including the losses associated with reverse recovery of the clamp (free-wheeling) diode, and turn-off losses are measured and compared for slow and fast IGBT devices. Benefits of zero-voltage-switching (ZVS) and zero-current-switching (ZCS) operation of IGBTs are demonstrated. The ZVS operation can reduce the total switching losses by more than half, while the ZCS reduces them by about 25% when compared with hard-switching. The ZCS turn-off also relieves the voltage stress of the IGBTs. >

Abstract: The paper presents the implementation of a DSP-based controller for three-phase, space-vector modulated converters. The implementation is illustrated for the control of a 2 kW, ZVS matrix power converter-based three-phase PWM rectifier. The controller features very high data processing speed (converter switching frequency of 100 kHz), and provides high-quality, low-distortion power converter input currents and output voltages. The controller can be implemented using only a few standard integrated circuits, providing high reliability and low cost. >

Abstract: Operation and implementation of the novel, space vector modulated, zero-voltage transition, three phase voltage source inverter/boost rectifier is presented. The converter is intended for high performance, medium power applications requiring bidirectional power flow. The proposed modified space-vector modulation allows all switches to be operated with soft-switching and constant frequency. The modulation algorithm also eliminates any low frequency distortion caused by the zero-voltage transitions and can be applied to any soft switching PWM three-phase converters. A simple, digital signal processor based implementation of the modulator and current regulators, operating with 30 /spl mu/s sampling time, is described. Measured efficiency of the 30 kHz, IGBT prototype is around 95%, and distortion of the three-phase currents is extremely small. >

Abstract: This paper describes how to improve the performance of open loop, low cost PWM-VSI controlled AC machines by the use of one current sensor. The performance is especially improved at low speed where nonlinearities like blanking-time and voltage drop are dominant. An often used modulation technique is described and basic compensation techniques are presented. Only one current sensor in the DC link is used for reconstruction of all three phase currents and the currents are used for compensation. Different limitations are discussed. The compensation techniques are tested in a 16-bit microcontroller based inverter. Tests show that the phase currents can be reconstructed by measurement of the DC link current both at low and high speed. They also show that a highly improved torque-speed characteristic can be obtained by using the compensation techniques and an almost ideal inverter is obtained. >

Abstract: This paper explores the origin of the DC current-sharing problem of parallel-converter systems and the dual problem of voltage sharing in series-converter systems. Both problems may be studied by examining the output plane (output current versus output voltage) of a particular converter. It is shown that strict current source behavior is unnecessary for good current sharing in parallel-converter systems. Furthermore, a broad class of converters whose output voltage is load-dependent, i.e., those that have a moderate value of output resistance, all exhibit good voltage- and current-sharing characteristics. Such converters are often suitable for a/spl times/b arrays of converters that can meet a large range of power-conversion requirements. The output planes of discontinuous mode PWM converters as well as conventional and clamped series resonant converters are examined in detail. A simple small-signal model of the modular converter system is developed. Experimental confirmation of load sharing and the small-signal model is given for the clamped series resonant converter and the series resonant converter for various configurations of four converters. >

Abstract: On-line PWM pattern generators for current source converters offer a number of control advantages over off-line optimized patterns. However, when implemented using the principles applying to voltage source inverter PWM pattern generators, the switching frequency is equal to: (a) the carrier frequency in standard carrier-based implementations and to (b) 2/3 the cycle frequency in space vector implementations. This paper shows that this frequency can be reduced to 1/2 of the respective frequencies. Two pattern generators are investigated: (a) a carrier based technique, namely the modified dead-band technique; (b) a space vector based technique, where the advantage is taken of the extra zero state available in current source converters. It is shown that a significant reduction of AC line current distortion is obtained with the modified dead-band technique for modulation indices greater than 0.4. Furthermore, with the two proposed pattern generators, there is no penalty on the distortion of the DC bus voltage. The principles of operation of the proposed schemes are explained. DSP implementation algorithms of the space vector scheme are given. Experimental results on a 5 kVA current rectifier and a 5 kVA current source inverter confirm the feasibility of the proposed pattern generators, and illustrate the input and output frequency spectra obtained. >

Abstract: A DC-to-DC (buck) converter comprises a PWM regulation loop and a hysteretic control loop, which are alternatively enabled by a mode selection circuit of the converter in function of the load level. When the level of load drops below a preset limit related to a design load level, the converter passes from a PWM control mode to a hysteretic control mode, thus eliminating switching losses during periods of operation at relatively low load level.

Abstract: Obtaining soft-switching operation for PWM power converters, except for a few cases, has required either high switch voltage stresses or high switch current stresses, or both. This paper presents a new class of commutation self-resonant PWM power convertor that overcomes this great disadvantage as well as being able to operate without switching losses in high switching frequencies for a wide line and load range. The circuit diagram and the phase-plane of each power converter of a family of such converters are also presented. In order to emphasize the principal characteristics of these new converters, a study, including experimental results, is carried out in detail for the buck and zeta converters. >