Search

At present, the existing electromagnetic interference (EMI) analysis methods for the cavity with transmission lines (TLs) often need a large number of meshes to satisfy certain accuracy, which would seriously affect the calculation time, memory, and efficiency. Therefore, this paper presents a time-domain parallel method consisting of finite-difference time-domain method, TL equations, and message passing interface library. This method has two obvious advantages: one is that it does not need to mesh the TLs, which can reduce a lot of computational memory; and the other is that it can run on multiple processors or computers to save a lot of computational time. In order to verify the accuracy and validity of this method, the intracavity transmission of multiconductor TLs and coaxial cable under ambient wave excitation is simulated numerically and compared with the traditional method. The results show that the method can be well applied for the EMI analysis of the TLs in cavity under ambient excitation. [ABSTRACT FROM AUTHOR]

This paper presents an efficient hybrid method consisting of finite-difference time-domain (FDTD) method, transmission line (TL) equations, and a novel time-domain S-parameter cascade (TD-SC) technique, which can be directly applied in time domain for the EMI analysis of transmission line network (TLN) in shielded electronic device excited by ambient electromagnetic wave. The FDTD method combined with an auto mesh generator is utilized to model the shielded structure of an arbitrary electronic device rapidly, and obtain the excitation fields of the TLN exactly. Then, the TD-SC technique based on TL equations, FDTD method, and S parameter of circuits is proposed for the further EMI analysis of the TLN consisting of multistage TLs and circuits. It has the advantage of without modeling the fine structures of TLs and circuits directly. Numerous examinations exhibit that this hybrid method is an efficient method for the EMI analysis of TLN by comparing with a three-dimensional FDTD method. The computation time and memory cost by this hybrid method is much less than that of the FDTD method, because the hybrid method provides a fast way to compute the transient responses on TLN in which the TLs and circuits do not need to be meshed. [ABSTRACT FROM AUTHOR]

This paper presents a novel time domain hybrid method consisting of finite-difference time domain (FDTD) method, transmission line (TL) equations, and circuit-based method, which can be directly applied in time domain for the electromagnetic interference (EMI) analysis of the lumped circuits in a shielded device excited by ambient electromagnetic wave. The FDTD method coupled with the TL equations is utilized to model the electromagnetic properties of arbitrary structures containing transmission lines and shielded enclosure of the electronic device with any ambient excitation. With numerical simulation, an equivalent circuit model considering the external EMI effects is extracted as an excitation source for the internal lumped circuit according to Thevenin's theorem. Then, the state variable approach is used for the further EMI analysis of the internal lumped circuit in time domain. Two cases, in which a transmission line and/or a transmission line network are connected to a lumped circuit within a shielded enclosure excited by an intense electromagnetic pulse, are computed with the proposed method. The numerical results are compared with those obtained by using the FDTD method, and they are almost in perfect agreements. Meanwhile, at last, an example of this method applied for the EMI protection of an electronic device is given. [ABSTRACT FROM AUTHOR]

The interwinding capacitance of the transformer and the primary-side circuit nodes to ground capacitance are critical coupling path for common-mode (CM) noise in isolated power converters. In order to analyze the CM noise, the transformer interwinding capacitance is usually equivalent to lumped capacitance. In this paper, a generalized lumped capacitance model for constructing the CM noise equivalent circuit that is valid at low frequencies is derived. By simplifying the CM noise equivalent circuit based on Thevenin’s theorem, the equivalent noise source (ENS) for CM noise is proposed. The ENS is expressed by a linear combination of voltages in the converter, where the coefficients of these voltages are determined by transformer winding structure and the circuit configuration. With the help of ENS, the CM noise of the converter under different circuit configuration can be systematically analyzed and conveniently compared, which develops an effective reference for selecting a proper circuit configuration. Finally, the lumped capacitance model and the ENS for CM noise are verified by the experimental results. [ABSTRACT FROM PUBLISHER]

In an isolated power converter, the interwinding capacitance of the transformer is the critical coupling path for the generating of common-mode (CM) noise. The shielding technique, for the purpose of blocking the electric coupling between the adjacent windings, has been widely used to deal with the CM noise. For the transformer with a single shielding layer, the displacement current that flows between the shielding layer and its adjacent windings still causes CM noise. In this paper, the shielding-cancelation (SC) technique is proposed by combining the shielding technique and the winding cancelation method together: The shielding layer is adjusted to have the same turns number and winding direction with its adjacent windings. In this way, the $dv/dt$ on the interwinding capacitance is always zero; thus, the CM noise can be effectively suppressed. A flyback converter prototype is built in the laboratory, and the performance of the SC technique on suppressing the CM noise is verified by experiments. [ABSTRACT FROM PUBLISHER]

The average-current-controlled boost power factor correction converter may operate in all-continuous current mode (CCM), part-CCM/discontinuous current mode (DCM), or all-DCM mode in a half line cycle under different input voltage and load conditions, and the conducted electromagnetic interference (EMI) spectrum of the converter exhibits differences in these operation modes. In this paper, by analyzing the characteristic of the harmonics of the drain-to-source voltage of the power switch, the characteristics of the common-mode and differential-mode noise of the converter are revealed in different operation modes. The analyzed results can be used to guide the EMI filter design for the converter and ensure the conducted EMI satisfying the standard under all working conditions. [ABSTRACT FROM PUBLISHER]

In this paper, the differential-mode (DM) electromagnetic interference (EMI) noise of a single-phase boost power factor correction converter operating in critical current mode was analyzed. The DM noise spectra are calculated based on the mathematical model of EMI receiver and the required corner frequencies of DM filter are obtained. It can be seen that the minimum corner frequencies are determined by the maximum noises at 150 kHz. With the relation between the magnitude of the inductor current ripple and the DM noise, the characteristics of noise at 150 kHz are obtained by analyzing the current ripple magnitude at 150 kHz; thus, the worst conditions which have the maximum noise value are figured out. Meanwhile, the maximum noises at 150 kHz for different input voltages are identical, so the DM filter can be designed based on one worst spectrum at one input voltage without testing the spectra in other conditions. [ABSTRACT FROM AUTHOR]

A discontinuous-current-mode (DCM) boost power factor correction (PFC) converter features zero-current turn-on for the switch, no reverse recovery in diode, and constant–frequency operation. However, the input power factor (PF) is relatively low when the duty cycle is constant in a half line cycle. This paper derives the expressions of the input current and PF of the DCM boost PFC converter, and based on that, variable-duty-cycle control is proposed so as to improve the PF to nearly unity in the whole input–voltage range. A method of fitting the duty cycle is further proposed for simplifying the circuit implementation. Other than a higher PF, the proposed variable-duty-cycle control achieves a lower output–voltage ripple and a higher efficiency over constant-duty-cycle control. The experimental results from a 120-W universal input prototype are presented to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]