Increasing Switching Frequency Research Articles

Bridge-arm current reduction in DC-AC inverter

ABSTRACTWhen building single-phase inverter with power Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET), switching action may cause poor reverse recovery characteristic due to body parasitic diode of MOSFET, which can produce peak current in the circuit loop and the high transient voltage and current (dv/dt, di/dt) during the turning-on period. In this article, a novel method to reduce the bridge arm current spike in DC-AC inverter is proposed. The presented method uses the improved and simplified coupled inductor which is connected between the upper and lower power devices. The parasitic capacitors of MOSFET are charged and discharged by the coupled inductor and the energy is released in the new loop; therefore, the bridge peak current is diminished. The time-domain model of transient-state analyses is given in detail. The current spike of the main switch is clamped efficiently. By increasing switching frequency, the volume of the magnetic core can be further reduced which is resulted from reduction in the reverse recovery current in parasitic diode. Because of the suppression of the spike current via the device, the switch-on loss of the power loss is reduced, and low on-state resistor of the power device can be adopted to suppress the conduction loss. The proposed approaches are validated with experimental results.

Comparative Analysis of Si- and GaN-Based Single-Phase Transformer-Less PV Grid-Tied Inverter

Recently, the interest in grid-tied PV transformer-less inverters has increased rapidly, because of their higher efficiency and lower cost compared to traditional line transformer inverters. This paper presents a new modified transformer-less topology derived from H5 inverter, and provides a detailed comparison between the use of GaN and Si devices for the proposed topology. Detailed operation modes, inverter structure and switching strategy of the proposed topology are presented. Datasheet information, conduction losses, switching losses, and heat sink requirements are studied and analyzed to provide an accurate comparison between GaN and Si power devices for the proposed topology operating at unity power factor. The results show that, GaN power devices significantly reduce the power losses in the system, which consequently allow a significant increase in either inverter power rating or switching frequency. Thus, the use of GaN power devices for the proposed inverter can be more appealing and cost-effective approach.

A 60-kW 3-kW/kg Five-Level T-Type SiC PV Inverter With 99.2% Peak Efficiency

A silicon carbide (SiC) T-type LCL inverter can achieve smaller device loss than two-level topology, however, its improvement on power density is limited by current ripple loss on magnetic components as switching frequency increases. This paper presents a five-level T-type (5LT2) photovoltaic (PV) inverter that achieves better utilization of SiC devices than the traditional three-level T-type LCL topology at higher switching frequency. The operation principle of the SiC 5LT2 PV inverter has been presented. The key design aspects including magnetic balancing, short-circuit protection, and digital controller computation time have been discussed and methods are developed. A 60-kW PV converter including boost stage and inverter stage has been built in the laboratory, which achieves a power density of 27 W/in3 and 3 kW/kg, and measured peak efficiency of 99.2%.

Numerical modelling of PCB planar inductors: impact of 3D modelling on high‐frequency copper loss evaluation

Loss values are key parameters for designing high-performance high-frequency magnetic components for power electronics (PE) converters. With the increase of PE switching frequencies, copper losses have to be precisely quantified, ideally until some megahertz. In the literature, many 2D numerical simulations based on finite element analysis (FEA) are performed for such computations. 3D FEA studies of planar components are still limited because of modelling problems, computational resources and computing time. In this study, quantitative comparisons between 2D and 3D simulation results for planar inductors are achieved focusing on copper loss computation. Results are compared in terms of simulation performances and accuracy. The aim of the study is to highlight benefits of 2D and 3D FEA simulations in order to choose the appropriate model according to the studied problem.

Single-Step Current Control for Voltage Source Inverters With Fast Transient Response and High Convergence Speed

Current control loop response in voltage source inverters impacts the quality of output current and output voltage waveforms. Parabolic current control provides a fast transient response with approximate-constant switching frequency, solving the frequency variation problem of hysteresis current control. This makes it a good candidate for the current control loop of voltage source inverters to achieve a good system performance. Yet, parabolic current control is often implemented with digital-to-analog converters, analog comparators, and field-programmable gate array circuits where increasing switching frequency pushes update speed and bandwidth requirements. Concurrently, even if the transient response of parabolic current control is fast, it can still take up several switching cycles converging to steady state. In order to solve both problems, a new current control strategy, motivated by the convergence analysis of parabolic current control but with a convergence process that takes just one switching operation, is proposed: single-step current control. Since single-step current control samples two data points in a switching cycle instead of using continuous parabolic carriers, it can be easily implemented in a digital microcontroller then high switching frequency can be achieved. The small signal model, dead-time compensation, and stability analysis are also studied in this paper. The convergence speed of the algorithm is verified with experimental hardware prototype and the transient performance of the designed voltage source inverter meets expectation.

Method for Diffusing Carrier Noise under Zero‐Speed Condition using a Zero‐Sequence Voltage

SUMMARYIn this paper, a technique is proposed to reduce the carrier noise of pulse width modulation (PWM) inverters for a permanent magnet motor drive. The carrier noise is generated by the motor and the reactor driven by the inverter and affected by the switching frequency. When a motor rotates and drives a machine, the carrier noise is hidden by the machine noise. However, when the motor speed is approximately zero, the carrier noise becomes dominant and raucous. The typical method to reduce the noise employs a switching frequency higher than the audible frequency of human hearing. However, the increase in switching frequency results in the inverter suffering a higher switching loss and lower efficiency. In particular, when the permanent magnet motor operates at approximately zero‐speed and full‐load, for example, in the hill‐start conditions of electric vehicles and the start and stop conditions of elevators, the current flows in specific power devices and the switching loss further increases. The proposed technique uses a zero‐sequence voltage, which is generated randomly with the M‐sequence signal, and diffuses the frequency components of the ripple contained in the current. The technique is able to reduce the noise without increasing the switching frequency of the inverter when the motor speed is almost zero. Simulation and experimental results show that the proposed technique can diffuse the carrier noise and the cycle of the M‐sequence signal changes the diffusion effect of the carrier noise.

Design Considerations for Gan-Based Microinverter for Energy Storage Integration Into Ac Grid

Abstract A full bridge converter with electrolytic capacitor on the dc bus is a widely used approach for a single phase interface for renewable energy source generation or energy storage integration in the utility grid. New wide bandgap devices enable higher switching frequency, higher efficiency and higher power density. In the paper, the authors introduce the challenges associated with an increase in switching frequency of a single phase inverter and implementation of wide bandgap GaN-based transistors instead of traditional Si-based transistors. The low gate threshold voltage of GaN transistor and unique reverse conduction behaviour require different driving circuit. The design of the driver circuit and other practical issues are analysed in the paper. The paper also presents some practical results. The research results can be useful to avoid mistakes by designing GaN-based power converters as these devices become increasingly interesting for commercial applications.

Hybrid modulation strategy for eliminating low‐frequency NP voltage oscillations in NPC using redistribution of NTV duty ratios

Neutral point (NP) clamped (NPC) inverters produce low-frequency NP voltage oscillations when modulated with nearest three-vector (NTV) methods for certain loading conditions. Non-NTV modulations can address these oscillations at the expense of increased switching frequency, output voltage distortion and complexity. This study presents a new generalised algorithm to implement hybrid modulation strategies which can also eliminate NP voltage oscillations using a combination of the aforementioned methods. The proposed algorithm avoids abrupt changeover between the two modulations and ensures minimum contribution of non-NTV operation and hence its drawbacks. Moreover, the algorithm is computationally efficient compared with its counterparts, since it only uses redistributed NTV duty ratios to generate NTV and non-NTV modulations instead of using two separate complex algorithms which result in increased computational burden. A non-NTV method called selected three-vector (STV) modulation is chosen to perform the detailed analysis of above proposed algorithm for hybrid NTV–STV modulation. Additionally, a simplified STV (SSTV) is derived to develop a new hybrid NTV–SSTV modulation strategy which further reduces losses and complexity compared with the hybrid NTV–STV modulation, without affecting the performance of NPC. The proposed algorithm and the derived hybrid modulation strategy are validated using simulation and experimental results.

Potential energy savings from high-resolution sensor controls for LED lighting

LED lighting systems paired with high-resolution sensor systems offer great potential for substantial energy savings via enhanced control options, and have other value-added features. We used modelling and demonstrations in a full-scale office test bed to illustrate this potential. First, we propose that such a system enables much shorter timeout periods for occupancy sensing: the higher density of sensors provides more reliable occupancy detection and LEDs are impervious to lifetime degradation from the consequent increased switching frequency. We used occupancy data measured in real office environments to estimate that a reduction in timeout period from the current code maximum of 20min to 1min would yield 26% additional energy savings. Further, we installed 72 LED luminaires in an office test bed with windows; each luminaire featured a co-located motion sensor and light sensor. The localized light sensor enabled daylight harvesting at each fixture to be optimized to local spatial conditions, yielding an extra 35% energy savings compared to a single photosensor controlling all luminaires, and ensuring better delivery of target horizontal illuminance values across the space. Finally, we demonstrated a combination of local occupancy and daylight harvesting features that resulted in 79% energy savings compared to typical prevailing energy code provisions for open-plan offices.

Study on the Application of Gallium Nitride Transistors in Power Electronics


 Wide bandgap semiconductors have emerged as an attractive option for silicon (Si) replacement in the recent years. Among the new materials, gallium nitride (GaN) has been considered as the most promising candidate. This paper presents an overview of the GaN technology in power electronics. The review focuses on the main aspects of GaN transistors, such as electrical, thermal and economical characteristics. A comparison between Si and GaN switching devices in a family of synchronous buck converters designed for LED lighting applications is also presented. This comparison was performed using synchronous buck converters, designed under same parameters, at five different switching frequencies, ranging from 100 kHz to 1 MHz. Efficiency and temperatures were recorded. GaN based converters presented higher efficiency and lower operating temperatures in all cases, with a maximum efficiency of 96.8% and a minimum of 94.5%. Besides, Si-based converters exhibited a higher performance degradation as switching frequency and dead time increase.

Half-Bridge GaN Power ICs: Performance and Application

For a variety of switching power topologies, especially modern soft-switching or resonant designs, half-bridge circuits are essential, ubiquitous building blocks. Operating these half-bridge circuits-i.e., providing power and signal to a floating high-side switch at very high frequencies-can shrink magnetics and enable a dramatic reduction in size, cost, and weight while delivering faster charging. However, such frequency increases have eluded the industry for decades as silicon (Si) devices have been too slow and suffered from parasitic impedances between the driver, high-capacitance Si field-effect transistors (FETs), and poorly performing level-shifters or isolators. As a result, most converters still run at 65-100 kHz. This article describes the exploitation of monolithic, laterally integrated, high-voltage, gallium nitride (GaN) technology to create high-efficiency, high-performance, half-bridge GaN power integrated circuits (ICs), and it shows practical achievements in application efficiency, simultaneously with increased switching frequency, which deliver higher power density.

Mutual inhibition circuit as underlying mechanism for bi-stable perception and non-linear responses in vision

It has been hypothesized that mutual inhibition circuit underlies bi-stable perception. It is possible that this circuit functions to resolve ambiguity of input image by quickly shifting the balance of competing signals in response to conflicting features. Furthermore, it is possible that feedback signals influence this "biased competition" to reflect global properties of the image. I developed a system to record from two pyramidal cells in primary visual cortex in vitro by double patch clamp recording technique combined with so-called "dynamic clamp" system. In this system, two neurons can be connected by modeled synapses. When a connection between the two neurons are established by mutually inhibiting model synapses, simultaneous activation of the two neurons caused bi-stable activities: dominance of neural activities alternated between the two neurons in about 1~2 seconds interval. Next, I ran paradigms analogous to various experimental paradigms used for behavioral studies of bi-stable perception. Levelt's 4th law states that increasing the strength of input contrast causes increase of switching frequency. Similarly, increasing the depolarizing current injected to both neurons showed increase of switching frequencies at the higher range of injected current. Flash suppression paradigm was modeled by injecting depolarizing current to one neuron for few seconds followed by simultaneous injections of the current to the two neurons. The simultaneous injection evoked bi-stability that mostly started with the dominant activity of the non-adapted neuron. Intermittent presentation of bi-stable figure is known to prolong the percept duration. When this paradigm was modeled by injecting depolarizing current intermittently, the prolongation effect was not observed. Finally, by placing a stimulus electrode to V2, feedback excitation was evoked. The stimulation often caused the switch of dominance. Hence, feedback signals can influence the outcome of the competition. With this system, the details of neural processes that undergo bi-stable activities can be investigated. Meeting abstract presented at VSS 2017

Dimensioning and shaping of inductors for power converters

The increase in active component switching frequency, via GaN technology, allows downscaling the inductor size as soon as the magnetic core-losses are controlled. However frequency induced dissipations in the magnetic material will create thermal heating, consequently complex core shapes allowing a better thermal exchange are needed. Here we propose a specific dimensioning methodology to determine the best form factor to optimise the exchange surface and verify constraints on the inductor. An inductor made of NiZn spinel ferrite powder has been manufactured by tape casting using results of calculations based on magnetic core-loss measurements.

An Accurate Subcircuit Model of SiC Half-Bridge Module for Switching-Loss Optimization

The increasing demand for high power density requires the power converter to operate in high switching frequency. Silicon carbide (SiC) power module is regarded as one of the most promising candidates for high-frequency applications due to the superior switching speed and low switching loss. With the increase of switching frequency, the switching loss will be the limiting factor of efficiency. Hence, it should be minimized during each switching transition. The optimization of switching loss is normally achieved by the repetitive double pulse test experiments. It is time-consuming to find an optimum gate resistance to achieve the tradeoff between switching loss and electromagnetic interface. In this paper, an accurate subcircuit model for SiC power module is proposed to assist optimization of switching loss in converter design. By considering the device physics and structure, an accurate Miller capacitance model is obtained. Moreover, a parameter extraction procedure is presented, which is based on the datasheet. Good agreements are achieved between the PSpice simulation and experiment.

SiC Power Devices in Impedance Source Converters

This paper discusses issues related to application of SiC power devices to new family of power converters. Impedance source converters show unique feature, buck boost characteristics due to specific impedance network. Passive elements of this network may be seriously reduced with the switching frequency increase, possible with fast-switching SiC transistors. On the other hand, switching conditions of the power devices are more severe than in traditional voltage-source or current-source converters. These issues are discussed on the base of the 6kVA/100kHz quasi-Z-source inverter example.

Detailed Study on Dynamic Characteristics of a High-Performance SGT-MOSFET With Under-the-Trench Floating P-Pillar

We proposed a 150 V shielded-gate trench (SGT) power MOSFET with floating P-pillars under the trench and studied its static and dynamic characteristics, especially the transient capacitance and dynamic ${\mathrm {R}}_{{\mathrm {ds}}({\mathrm {on}})}$ at high switching frequency. TCAD simulation demonstrated over 33% ${\mathrm {R}}_{{\mathrm {ds}}({\mathrm {on}})}$ reduction compared with regular SGT-MOSFET. Up to MHz switching frequency, simulation indicates almost no ${\mathrm {R}}_{{\mathrm {ds}}({\mathrm {on}})}$ increase in this device, meaning negligible dynamic ${\mathrm {R}}_{{\mathrm {ds}}({\mathrm {on}})}$ penalty. On the other hand, elevated ${\mathrm {C}}_{{\mathrm {oss}}}$ is observed within a particular ${\mathrm {V}}_{{\mathrm {ds}}}$ range (40–90 V) as switching frequency increases, which translates to $\text{Q}_{{\mathrm {oss}}}$ penalty. We investigated the underlying physical process for this ${\mathrm {Q}}_{{\mathrm {oss}}}$ change, and attributed it to the hole pilings in the undepleted region of the P-pillar. The overall parameters of this device, including ${\mathrm {R}}_{{\mathrm {ds}}({\mathrm {on}})}$ and ${\mathrm {R}}_{{\mathrm {ds}}({\mathrm {on}})} \ast {\mathrm {Q}}_{{\mathrm {oss}}}$ , are superior to regular SGT-MOSFET and are close to superjunction MOSFET, which is more expensive and challenging in manufacturing. Although a 150 V MOSFET is used as our study vehicle, the analyses and conclusions apply to any power devices that employ the proposed device architecture.

Self‐capacitor voltage balancing method for optimally hybrid modulated cascaded H‐bridge D‐STATCOM

A hybrid modulation (HM) for cascaded H-bridge converters is suggested which is a fusion of selective harmonic elimination (SHE) and carrier based bipolar pulse width modulation. It differs from its counterparts in the sense that only one triangular carrier for all H-bridges is used as opposed to multi-triangle waveform set and only one level of capacitor voltage and single type of semiconductor are considered. HM eliminates calculating and storing offline delay angles for each modulation index as in case of SHE. A novel self-voltage balancing mechanism is also proposed for HM which is based on strategically swapping redundant switching states. This mechanism is free from voltage sorting, current measurement, extra circuitry, or additional controllers. The comparison with phase and level shift pulse width modulation schemes reveal that the HM has 15% more DC voltage utilisation, and lower harmonic content for wide interval of modulation index. Moreover, HM has low semiconductor losses, as low as 0.55%. The tests on 1500 kVAR distribution static synchronous compensator shows that HM generates only 4% voltage ripple at the cost of few increase in average switching frequency. Voltage and reactive power balancing in each phase are also established with high power efficiency.

A Three-Level Dodecagonal Space Vector-Based Harmonic Suppression Scheme for Open-End Winding IM Drives With Single-DC Supply

This paper presents a harmonic elimination technique based on a three-level dodecagonal space vector structure for open-end winding induction motor (IM) drives with a single-dc supply. Advantages of dodecagonal space vector switching and multilevel inverters are achieved with a single-dc supply. A dc supply-fed three-level flying capacitor (FC) inverter feeds active power to one end of the IM winding terminals and H-bridge connected capacitors eliminate fifth- and seventh-order harmonics from the other end. A pulsewidth modulation (PWM) technique is developed to switch the three-level dodecagonal space vectors and simultaneously control the H-bridge capacitors at $0.1445V_{\text{dc}}$ . The fifth- and seventh-order harmonics are eliminated for the full modulation range of the three-level FC inverter, including the extreme 6-step operation. An increase in the linear modulation range has been achieved, resulting in improved dc bus utilization. Harmonic elimination is achieved by increasing switching frequency of the low-voltage capacitor connected H-bridge inverter; thus, switching frequency for the high-voltage dc supply-fed FC inverter is not increased. Using the proposed scheme, the instantaneous phase voltage never exceeds $\frac{2}{3}V_{\text{dc}}$ (maximum phase voltage applied for hexagonal space vector structure) for which the machine windings are rated. Additionally, the proposed inverter has also been shown to operate for field oriented control of the open-end winding IM drive.

Development and Optimization of Thin-Film Technology Based Micro Inductors and Transformers

In order to follow the current trends of electronic device development like miniaturization and increase of switching frequencies especially for magnetic passive components e.g. inductors and transformers innovative fabrication technology is expedient. In general switching frequencies in the range of 15 MHz to 50 MHz are intended. Some researcher teams focus on manufacturing micro transformer and inductors with thin-film technology in order to satisfy the market demands [1,2]. The design, the fabrication technology and the characteristics of the manufactured micro inductor and transformer were previous published. The 1st and 2nd generation were equipped with 4 contact pads on each side, primary and secondary side. The 5 µm thick soft magnetic closed oval core was deposited electrochemically. It consists of NiFe 80/20 [3] or CoFe 45/55 [4] and a multilayer coil with non-copper seed layer. The 3rd generation was equipped with 3 contact pads on each side and 10 µm [5] or 5 µm [6]thick electrochemically deposited CoFe 45/55 closed oval core with an electroplated multilayer coil with non-copper seed layers. The 4th generation of this solenoid-core transformer is developed and presented in this work. The dies have 3 contact pads on each side, 5 µm thick electroplated CoFe 45/55 core is enclosed in pure copper coils. The impetus to further development of the devices is the electrical and magnetic performance optimization. The process flow is modified in order to eliminate the interplies of chromium and gold -which were used as seed layers for electrochemical deposition of copper on copper. The 4th generation of the micro inductor and transformer with Electronic Industries Alliance Standard 1008 footprint is based on thin-film technology namely electrochemical deposition of copper and soft magnetic alloy CoFe 45/55 , physical vapour deposition, photolithography and encapsulating with copper compatible, photosensitive polyimide. The optimization of the process flow shows lower electrical resistance of the coils in comparison to earlier generations. Whole multilayer coil structures are made of pure copper and have no delamination defects. The closed magnetic core is built of electrochemically deposited soft magnetic alloy CoFe 45/55 . In order to reduce the residual low electrical conductivity of the polyimide surface which will be brought by ion beam etching substrate cooled sputter etching is used. Another step of preparing the copper surface after wet chemical development of photosensitive polyimide is presented. The surface cleaning is necessary for successful ultrasonic wire bonding. The final embedding of the device into a QFN-package is needed for the characterization. It is fulfilled with two component epoxy which has a wide temperature working range. The functionality of fabricated micro magnetic device was tested in high frequency DC-DC step down converter. For this test a specific integrated circuit MDCD073 a dual 7V PN half bridge with pre-drivers in 180 nm SOI technology is used. The frequency of the convertor was from 15 MHz to 30 MHz. The micro device shows the efficiency of about 80 % for output currents between 300 mA to 800 mA. The electrical resistance of the devices, with all four coils in serial connection, is 250 mΩ. The measured inductivity is above 50 nH which is stable up to very high frequencies. Yet the breakthrough voltage has to be determined as well as the impedance parameter which will be calculated from the s-parameters by matrix transformation. [1] Xing, Xing; Sun, Nian X;Chen, Baoxing; High-Bandwidth Low-Insertion Loss Solenoid Transformers Using FeCoB Multilayers; IEE Transactions on Power Electronics 2012 Vol 28 4395-4401 [2] Feeney, Ciaran; Wang, Nigning; Mathúna, Seán Cian; Duffy, Maeve; A 20-MHz 1.8-W DC–DC Converter With Parallel Microinductors and Improved Light-Load Efficiency; IEEE Transactions on Power Electronics 2015 Vol 30 771-779 [3] Dinulovic, Dragan; Gerfer, Alexander; Opitz, Oliver; Kaiser, Matthias; Wurz, Marc C., Rissing, Lutz; Thin-Film Microtransformer for High Frequency Power Application; Joint European Magnetic Symposia 2014 Vol 75 [4] Dinulovic, Dragan; Kaiser, Matthias; Gerfer, Optiz, Oliver, Martin; Wurz, Marc C., Rissing, Lutz; Microtransformer with closed Fe-Co magnetic core for high frequency power applications; Journal of Applied Physics 2014 Vol 115 17A317-1 – 17A317-3 [5] Dinulovic, Dragan; Gerfer, Alexander; Haug, Martin; Kaiser, Matthias; Wurz, Marc C., Rissing, Lutz; Improved Microtransformer Design Utilizing Fe-Co Magnetic Core; Physics Procedia 2015 Vol 75 1252-1258 [6] Dinulovic, Dragan; Shousha, Mahmoud; Haug, Martin; Beringer, Sebastian; Wurz, Marc C.; Thin-Film Based Microtransformer Suitbale for High Switching Frequency Power Applications, in press Figure 1

The High-efficiency LED Driver for Visible Light Communication Applications.

This paper presents a LED driver for VLC. The main purpose is to solve the low data rate problem used to be in switching type LED driver. The GaN power device is proposed to replace the traditional silicon power device of switching LED driver for the purpose of increasing switching frequency of converter, thereby increasing the bandwidth of data transmission. To achieve high efficiency, the diode-connected GaN power transistor is utilized to replace the traditional ultrafast recovery diode used to be in switching type LED driver. This work has been experimentally evaluated on 350-mA output current. The results demonstrate that it supports the data of PWM dimming level encoded in the PPM scheme for VLC application. The experimental results also show that system’s efficiency of 80.8% can be achieved at 1-Mb/s data rate.