Common-mode Filter Research Articles

Effect of Semiconductor Parasitic Capacitances on Ground Leakage Current in Three-Phase Current Source Inverters

This paper investigates the influence of power semiconductor parasitic components on the ground leakage current in the three-phase Current Source Inverter topology, in the literature called H7 or CSI7. This topology allows reducing converter conduction losses with respect to the classic CSI, but at the same time makes the topology more susceptible to the parasitic capacitances of the semiconductors devices. In the present work, a grid-connected converter for photovoltaic power systems is considered as a case study, to investigate the equivalent circuit for ground leakage current. The same analysis can be extended to applications regarding electric drives, since the HF model of electric machines is characterized by stray capacitance between windings and the stator slots/motor frame. Simulation results proved the correctness of the proposed simplified common-mode circuit and highlighted the need of an additional common-mode inductor filter in case of resonance frequencies of the common-mode circuit close to harmonics of the power converter switching frequency. Experimental results are in agreement with the theoretical analysis.

An alternating high‐low impedance common‐mode filter for differential striplines

The conventional quarter/half-wavelength common-mode filter (CMF) cannot work well for a differential stripline when there is no ground via. A new and effective via-free CMF design for the differential striplines is proposed in this study. An alternating high-low impedance structure is developed to increase the coupling between the stripline and the CMF, thus achieving a good filtering effect. A cascaded CMF design is further studied to enhance the stopband without deterioration of the differential signals. Both simulation and measurement results demonstrate the effectiveness of the proposed high-low impedance CMF. The significant advantage of the proposed CMF is its compactness and simplicity, which has almost no impact on the original layout of differential striplines.

PWM Strategy to Alleviate Common-Mode Voltage with Minimized Output Harmonic Distortion for Five-Level Cascaded H-Bridge Converters

High-frequency components of common-mode voltage (CMV) induce the shaft voltage and bearing current, which lead to premature failures in motors. In addition, due to non-zero average CMV, the low-frequency components of CMV, particularly the third-order harmonic component, have been reported to cause difficulties in common-mode filter design. Furthermore, the utilization of distant voltage vectors in the pulse-width modulation (PWM) with reduced CMV magnitudes gives rise to high output harmonic distortion compared to PWM ones without CMV reduction. In an attempt to solve the aforementioned issues, this article presents a PWM strategy that features reduced CMV magnitudes, zero average CMV, and improved output harmonic distortion for a five-level cascaded H-bridge (CHB) converter. In addition, the carrier rotation technique based on the phase-leg redundancy of the CHB topology is also combined with the proposed scheme to achieve equal power loss distribution among power switching devices. Both simulation and experimental results confirm that the proposed strategy produces better output harmonic distortion than that of POD-SPWM and APOD-SPWM under the condition of reduced CMV magnitudes, zero average CMV, and equal power loss distribution.

Comparison of Two Filter Design Methods for Conducted EMI Suppression of PMSM Drive System for Electric Vehicle

In order to solve the problem that the conducted voltage of a motor drive system for electric vehicle (EV) exceeds the standard limits of CISPR25-2016, two different design methods of electromagnetic interference (EMI) filter for HV power supply were proposed. The first method is to design a wide-band EMI filter based on engineering method. It can achieve the insertion loss of 60 dB within 150 kHz--108 MHz. Another proposed method is to design the EMI filter based on the resonance peaks suppression. During the second filter design process, by analyzing the common mode (CM) and differential mode (DM) interference paths at two key frequency points 1 MHz and 30 MHz, the dominated parameters responsible for the over-standard points were determined, and the CM and DM filter circuit was designed accordingly. Compared with the filter designed by the first method, the filter designed based on resonance peaks suppression can better meet the design goals, without further rectification according to the tests, and it is smaller in size, only 1/4 of the filter size designed by the first method, lower in cost, and easy to be engineering.

Optimized Filter Design for Common-Mode Current Reduction in Four-Wire Inverter-Fed Motors

Two optimized filter topologies for four-wire electric drive systems (EDSs) are proposed with smaller size, lower power loss, and larger common-mode (CM) current reduction than the usual filter topologies. It has been previously shown that using a four-wire topology reduces the CM current in a wide frequency range, when compared with the three-wire one. However, there still exist low- and high-frequency resonances in the transfer function between neutral and inverter output voltage, resulting in the increase of CM current. By means of the transfer function analysis, two types of optimized filter topology specified for the four-wire EDS are proposed. Unlike conventional CM filter (CMMF), the design of these two new filters also requires a serious consideration of the characteristics of motor and cable impedance. Based on the design procedure, these two filters are built with excellent performance of CM current reduction in the whole frequency range. Besides, compared with conventional CMMF used for three-wire EDS, they also show benefits, such as lower power loss and less implementation space. All theoretical analyses are validated by the experimental measurements and can benefit further studies.

A Simplified Frequency Model for Industrial Common-Mode Chocks Used in High-Power Converters

This paper proposes a simplified analytical model for electromagnetic interference (EMI) filters used in high-power converters. Highpower converters produce radio frequency conducted noise because they use high-frequency switching in the range of a few kHz to tens of kHz. The noise propagates into the power grid, which disturbs the functionality of the radio frequency apparatuses. Well-known standards, such as CISPR, provide the measurement and assessment methodologies for these devices. Moreover, the emission level of the noise is restricted at the source side. Using EMI filters is the most effective approach for dealing with this issue. However, due to the nonlinear nature of the common-mode (CM) cores, the modeling of the cores is a complicated task, which makes the selection and design of the filters less than optimal. In this paper, an analytical modeling of the CM filters is provided to suit the nonlinear frequency-dependent behavior of the CM cores. The simplicity of the proposed model makes it a very suitable choice for inclusion in the design procedures, which results in a more accurate and optimum filter selection among the many available commercial industrial filters. To validate the proposed model, the frequency-dependent model obtained is verified by experimental tests with a commercial CM choke. According to the results, the proposed model accurately describes the actual EMI filter behavior.

Design of a Differential Coupled-Line Directional Coupler Using a Double-Side Coplanar Waveguide Structure With Common-Signal Suppression

In this work, a rigorous approach for the design of a differential coupled-line directional coupler (DCLDC) implemented in double-side coplanar-waveguide (CPW) technology is presented. Due to the double symmetry of the structure, it is possible to obtain the coupling level and port characteristic impedances of the coupler in a simple and efficient way. In addition, due to the broadside coupling of the involved differential CPW lines, a wide range of coupling levels can be covered. Design curves relating the geometrical parameters of the structure to the electrical parameters of the DCLDC are provided, and a 3-dB coupler with an impedance of $50~\Omega $ at the terminal ports has been designed, fabricated, and measured. Common-mode (CM) transmission to both through and coupled output ports has been reduced by adding a simple and compact CM filter at the input port. The good agreement between the results obtained with EM simulation, transmission line models, and measurements confirms the benefits of the proposed structure and the suitability of the design procedure.

Model Predictive Control With Reduced Common-Mode Current for Transformerless Current-Source PMSM Drives

The integrated dc-link choke can replace the bulky transformer in current-source converter (CSC)-fed motor drive systems to bear the common-mode voltage (CMV). However, the common-mode (CM) resonance will be excited by the specific harmonics in the CMV, which causes excessive CM current in the loop. Moreover, due to the high-order filter, the $LC$ / $CL$ resonance is prone to be induced at the rectifier and inverter sides, respectively. In this article, the model predictive control (MPC) scheme is proposed to tackle the CM resonance and filter resonance simultaneously. In the low-speed region, besides restraining the peak-to-peak (PTP) magnitude of the CMV at the rectifier side, the third-order harmonic of the CMV generated by the inverter is extracted and then penalized in the cost function to further suppress the CM current. With the increase of the motor speed, the control objective of the inverter side controller switches to the PTP magnitude of CMV suppression because of the alleviation of the CM resonance. The capacitor voltage of both rectifier and inverter sides are regulated via the cost function as well, which can mitigate the current harmonic distortion and improve system stability. The simulation with 1MVA rated power and the scaled-down experiment shows that the proposed scheme can suppress both the CM resonance and the $LC$ / $CL$ resonance effectively with a low switching frequency.

ОСОБЛИВОСТІ ВИМІРЮВАННЯ ТА ЕФЕКТИВНОГО ПРИТЛУМЛЕННЯ НАПРУГИ КОНДУКТИВНИХ ЕЛЕКТРОМАГНІТНИХ ЗАВАД ВІД ТРАНЗИСТОРНИХ ПЕРЕТВОРЮВАЧІВ НАПРУГИ

The paper considers voltage converters as sources of conductive electromagnetic interference (EMI). Some features of the standard measurement method of converters EMI are considered and it is shown that it measures only the total level of noise, which can be the same at completely different relations between the differential and common components of EMI. This leads to inefficient use of means reducing the total converter noise to the permitted level, because the different ratio between the above components of this noise requires different means to reduce them. The paper proposes to supplement the standard noise measurement method by two additional measurements at frequencies where the total noise exceeds the permitted level, using during the first measurement the additional RFI common mode filter which effectively reduces only the common noise, and the additional RFI differential mode filter which effectively reduces only the differential noise during the second measurement. It is shown that these two additional measurements make it possible to determine the differential and common components of the total noise. This, in turn, makes it possible to reduce the total noise to the permitted level at the minimal cost. References 6.

A Novel Ultrawideband Absorptive Common-Mode Filter Design Using a Miniaturized and Resistive Defected Ground Structure

An ultrawideband absorptive common-mode filter (A-CMF) with a newly developed defected ground structure (DGS) is proposed in this article. The DGS miniaturized to a subwavelength size is able to totally suppress the common-mode noise while maintaining the signal integrity of the differential mode. In addition, the reflection and radiation of the common-mode noise are negligible in the stopband. For quick design, a simplified equivalent circuit model is also proposed. This model can be employed to estimate the resistances of the lumped resistors embedded in the DGS to obtain a very wide absorption band. In the demonstrated example, the fractional bandwidth determined by a 90% absorption efficiency can achieve as large as 104% around 3.55 GHz for an A-CMF realized on a simple two-layer printed circuit board. Such a widely absorptive bandwidth is never achieved before to our knowledge. The design methodology is validated by full-wave simulation. Finally, real measurement of the design is also carried out to confirm the design and analysis.

EBG Common-Mode Filter Design Using Uncoupled Coplanar Waveguide to Microstrip Transitions

This letter investigates the effects of an Electromagnetic Bandgap (EBG) common-mode filter design using coplanar waveguide (CPW) to microstrip transitions on differential signals. Specifically, the improvements in the frequency and time domains are studied by s-parameters and eye diagrams. EBG structures are based on a periodic alteration of even-mode impedances to create stop bands in the common-mode, while not affecting the differential mode as much as possible. The proposed common-mode filter is unique in the way that a broad bandwidth is achieved by choosing one of the even-mode characteristic impedances as small as possible. This EBG structure is implemented on a printed circuit board (PCB) and characterized using a 4-port vector network analyzer (VNA).

Rapid Geodetic Observations of Spatiotemporally Varying Postseismic Deformation Following the Ridgecrest Earthquake Sequence: The U.S. Geological Survey Response

Abstract The U.S. Geological Survey’s geodetic response to the 4–5 July 2019 (Pacific time) Ridgecrest earthquake sequence comprised primarily the installation and/or reoccupation of Global Navigation Satellite System (GNSS) monumentation. Our response focused primarily on the United States’ Navy’s China Lake Naval Air Weapons Station base (NAWSCL). This focus was because much of the surface rupture occurred on the NAWSCL and because of NAWSCL access restrictions only permitting Federal and State of California personnel. In total, we measured or are still measuring at 24 sites, 14 of which were on the NAWSCL and, as of this writing, operational. The majority of sites were set up as continuous stations logging at either 1 sample per second or 1 sample per 15 s. Two stations were recording a 200 m cross-rupture aperture starting ∼10 hr after the M 6.4 event, and they recorded the coseismic displacements of the M 7.1. Approximately, 1 hr after the M 7.1 event, two new stations were recording a ∼200 m cross-rupture aperture of the surface rupture. In the days following, we established the rest of the stations ranging to a distance of ∼15 km from the M 7.1 principal rupture trace. The lack of differential displacement across the M 6.4 rupture during the M 7.1 event suggests that it did not reactivate the M 6.4 plane. The lack of differential cross-fault displacement for both events suggests that rapid shallow afterslip did not occur at those two locations. The postseismic time series from these stations shows centimeters of horizontal displacement over periods of a few months. They record a mixture of fault-parallel and fault-normal displacements that, in conjunction with analysis of more spatially complete Interferometric Synthetic Aperture Radar displacement fields, suggest that both poroelastic and afterslip phenomena occur along the M 6.4 and 7.1 rupture planes. Using preliminary data from these and other regional stations, we also explore the Ridgecrest sequence’s effect on regional GNSS time series and the differentiation of long-term postseismic motions and secular deformation rates. We find that redefining a common-mode noise filter using different GNSS stations that are assumed to be unaffected by the earthquakes results in small but systematic differences in the regional velocity field estimate.

Balanced Bandpass Filter With Common-Mode Reflectionless Feature by Terminated Coupled Lines

Based on the terminated coupled-line pairs, this article reports a novel design method to synthesize a common-mode (CM) reflectionless all-stop filter with differential mode (DM) bandpass characteristic for balanced radio-frequency (RF) transmitter. With the adoption of symmetric property, the equivalent circuit models for both DM and CM are developed for the design purpose. The main contribution of this article is the development of a systematic impedance matching method to CM noise. Compared with conventional approach like adjusting the transmission zeros, the CM noise power would be dissipated on the terminators (resistors) rather than reflected. For practical validation, two butterworth microstrip prototypes (2nd-order/ 4th-order) were manufactured and measured in the 0–5.0 GHz range. It would be found that the measured CM absorption rate is over 88% from 0 to 5 GHz for both samples. Besides, the DM center frequency locates at 1.0 GHz with the measured 3-dB fractional bandwidth of 56% and 65%, respectively. To author's best knowledge, this work is the first RF balanced bandpass filter that is capable of absorbing the CM noise at all frequencies.

Mismatch in EBG Common-Mode Filters Implemented on PCBs

In this article, impedance mismatch effects on the characteristics of common-mode (CM) electromagnetic bandgap (EBG) filters are studied using three-dimensional (3-D) full-wave numerical simulations. The CM EBG filters considered herein are designed using standard printed circuit board technology and contain either microstrip (MS) differential pairs running above the EBG plane, or stripline (SL) differential pairs running on one of the layers next to the EBG plane. First, the terminations are fixed at 50 Ω, and the effects of the differential line impedance variations on the EBG filter parameters are studied. Overall, the considered variations in the widths of the traces and edge-to-edge separation distances in both the MS and SL structures do not drastically deteriorate the performance of the EBG CM filters. Then impedances of the ports are set different from 50 Ω, and the effects of this mismatch on the baseline SL and MS EBG structures are studied. It is shown that lower port impedances may have significant effect on the characteristics of filters. In addition, baseline EBG filters are cascaded with unmatched four-port load, and this may also deteriorate performance of filters.

EBG Common-Mode 20-GHz Microstrip and Stripline Filters: Sensitivity to Design Parameters

In this article, design and characteristics of 20-GHz electromagnetic bandgap (EBG) common-mode (CM) filters for microstrip (MS) and stripline (SL) differential pairs are considered. CM filter notch (resonance) frequency, depth, and width at the -15 dB level are calculated from the mixed-mode S-parameters. The agreement between the measured and modeled S-parameters of the baseline structures validates the models and allows for further numerical experiments. The systematic analysis of the trends for both MS and SL filter characteristics as functions of various geometrical and material design parameters is presented. Sensitivity to various technological features (conductor surface roughness, trapezoid shape of the traces, and line length imbalance) is also analyzed. It is shown that the MS EBG structure is more sensitive and less monotonic to most of the design parameters than those for the SL. This is attributed to the less electromagnetic field containment and more complex mode structure in the MS structures as compared to the SL. Systematic sensitivity analysis allows for an optimal design of the EBG CM notch filters for any printed circuit board layer where a differential pair runs. Though the analysis is presented for 20-GHz EBG filters, most of the trends would be applicable to the filters for the other frequencies.

Compact differential signal balancer embedded in metal wiring layers of silicon LSI for common mode noise filtering

The differential signal balancer (DSB) is a discrete common mode filter component using a low temperature co-fired ceramic (LTCC) process in which non-magnetic materials are employed. In this paper, we have developed the new DSB embedded in metal wiring layers of silicon LSI for achieving the multiple ultra-high-speed differential transmission exceeding 10 Gb s−1 between LSIs. While the spiral inductors in the silicon CMOS process are affected by the substrate loss and conductor resistance with the thinner thicknesses for both metal and dielectric layers, its lumped circuit parameters are optimized well, and the obtained characteristics are almost comparable with that of the conventional DSB using LTCC. It has approximately −10 dB attenuation of common mode noise at 2.4 GHz with very small size of 120 μm × 250 μm and volume size of about 1/6000 compared to the conventional DSB.

Reduced Common-Mode Voltage PWM Scheme for Full-SiC Three-Level Uninterruptible Power Supply With Small DC-Link Capacitors

In this article, a pulsewidth modulation (PWM) scheme for three-level full-SiC uninterruptible power supplies is developed to achieve a high power density. Two key passive components are selected for size reduction of the ac–ac stage: common-mode (CM) EMI filter, and dc-link capacitors. To reduce the CM noise, a new vector combination is proposed based on synchronous switching among three-phases. The proposed combinations align CM voltage (CMV) to be a single pulse per switching period. Owing to the simple shape, CMV cancellation between a three-level rectifier and inverter can be maximally utilized. A transition between the three combinations can control the drift of neutral point voltage. An equivalent carrier-based implementation is developed. Second, a simple algorithm to compensate neutral point voltage fluctuation is proposed both for differential mode (DM) and CM output voltage. Low-order harmonics on three-phase currents and an additional high-frequency CM noise by misaligned switching instant can be eliminated. The proposed compensation can be implemented by a simple correction on carrier slopes and injected zero-sequence voltage. The proposed PWM scheme is verified with 20-kW full-SiC UPS switching at 60 kHz with 140 μ F dc-link capacitors.

Eigenvalue Analysis and Behavioural Common-Mode Equivalent Circuit for an Asymmetric Active Phase Converter

A common-mode (CM) filter for a reduced switch active phase converter (APC) is proposed in this article for mitigation of switching frequency, and higher order CM current (CMC) injected into the grid. Due to asymmetry of the APC topology, a mixed-mode analysis of the differential-mode (DM) and CM components becomes essential. The presented state-space approach for the converter analysis captures the mixed-mode behavior of the APC. The eigenvalue analysis of the system shows that the inclusion of the proposed CM filter does not affect the existing DM APC system poles. Hence, the CM filter does not affect the APC control structure. This work also introduces an equivalent behavioral circuit which simplifies the CM filter component selection. This article provides a design methodology for the selection of the CM filter components by identifying the range of CM resonance frequencies. This design method results in active damping of the CM resonance without any additional control loop. The simulation results validate the behavioral CM equivalent circuit. The experimental validation is carried out on a laboratory APC prototype feeding a three-phase 3 hp induction motor from a single-phase grid. The measured CM voltages and currents validate the efficacy of the proposed CM filter for the APC.

A Common-Mode Noise Suppression Filter for Gigahertz Differential Signals Based on Substrate Integrated Waveguide Resonators

A common-mode noise suppression filter for microstrip differential lines based on substrate integrated waveguide resonator (SIWR) is proposed in this article. The proposed common-mode filter (CMF) consists of cascaded SIWR units with different size. The equivalent characteristic impedance circuit model of the proposed CMF is given to predict the common-mode suppression characteristics by introducing the spectral domain approach (SDA). Simulated results show that the common-mode noise is decreased more than 35 dB from 4.5 GHz to 11.4 GHz, and −30 dB rejection band is obtained in the frequency band of 4.3-15 GHz. Moreover, the insertion loss and the group delay in the frequency domain show that there is little degradation for differential signals. In the time domain, 86.5% amplitude suppression of common-mode noise is achieved. In addition, the eye diagrams in the time domain are investigated to explain the effectiveness of the proposed CMF and indicate that the CMF can maintain the differential signal integrity. The prototype is fabricated to verify the validity of the proposed structure, and the measurement results show good agreement with the simulation results.

A Bidirectional Absorptive Common-Mode Filter Based on Interdigitated Microstrip Coupled Lines for 5G “Green” Communications

This paper proposes a bidirectional absorptive common-mode filter (A-CMF) for 5G “green” communication systems. The A-CMF is a balanced-to-balanced structure using interdigitated coupled lines to replace normal double parallel-coupled lines for enhanced coupling and easy manufacturing. The resistors are introduced to dissipate the common-mode (CM) noises into heat and thus to avoid the noises being reflected and still existing in the communication system. This novel A-CMF features an intrinsic CM noises absorption while maintaining its differential-mode (DM) filtering characteristics. As a proof-of-concept demonstration, one microstrip prototype is fabricated in a two-layer printed circuit board (PCB) and the measurements are consistent with the simulations. The DM signals can pass through this A-CMF without being attenuated from 1.38 GHz to 5.19 GHz but the CM noises are suppressed throughout the broad frequency range between 0.72 GHz and 8 GHz. It is worth noting that this A-CMF realizes a wide band with 90% absorption efficiency of CM noises from 2.18 GHz to 4.97 GHz.