CHAPTER 2 EQUIVALENT CIRCUIT MODELING OF CONDUCTED EMI BASED ON NOISE SOURCES AND IMPEDANCES

29 CHAPTER 2 EQUIVALENT CIRCUIT MODELING OF CONDUCTED EMI BASED ON NOISE SOURCES AND IMPEDANCES A simple equivalent circuit modeling approach to describe Conducted EMI coupling system for the SPC is described in this chapter. The resulting model assumes a minimum number of noise sources and contains essential coupling paths that permit simple substantial interpretations. The three modes of Conducted EMI noise are: Mixed-Mode (MM), Intrinsic- Differential-Mode (IDM), and Common-Mode (CM) are acknowledged by measurements connected with an isolated half-bridge AC DC converter. 2.1 INTRODUCTION The analytical noise model is first investigated to get a full understanding of the EMI mechanism. The EMI characteristic of the power converter is analytically realized from a circuit theoretical point of view. The procedure of parameters extraction for the noise models consists of simple measurements and is suitable to be implemented. Experimental results are included to verify the validity of this method. Based on an equivalent circuit approach, the proposed model is easy to apply in practice for understanding, diagnosing and approximating EMI behaviors. Accurate modeling of EMI noise generation and circulation in power converters is the first step to predict and manage the EMI noise in a system. The diagnosis and modeling method of noise sources and coupling

30 paths are helpful for designers in improving the converters EMC performance. Various methods have been proposed for parasitic modeling such as the three-dimensional (3-D) Finite Element Analysis (FEA), Time-Domain Reflectometry (TDR), and the Partial Element Equivalent Circuit (PEEC) method. Due to limitations of the existing EMI source modeling methods, a practical and accurate EMI noise source modeling method is needed. The established equivalent circuit models are found to be sufficient to analyze and predict EMI behaviors up to 30 MHz. This model contains all essential coupling paths that should be taken into consideration when evaluating the level of EMI that can be picked up by the standard LISN. The CM and DM excitation sources are measured online and Thevenin impedances are measured offline. 2.2 EMI GENERATION AND ESSENTIAL COUPLING PATHS The EMI modeling techniques for power converters need to be extended to include noise coupling phenomena in SPC. In power converters, the major EMI source is associated with high dv/dt and di/dt during the switching instant. 2.2.1 Conducted EMI Measurement Figure 2.1 shows the configuration of the Conducted EMI measurement for the EUT. The power source, provided through a LISN, is required for the measurement of Conducted EMI. The LISN contains two 50µH inductors, two 0.1µF capacitors and two 50 resistors. For power-line frequency, the inductors are essentially short, the capacitors are essentially open, and the power is passed through LISN to supply the EUT. According to

31 conventional theory, DM noise is defined as the voltage difference between two LISN resistors, and CM noise is defined as the average voltage of two LISN resistors. L N 50µH L 1 50µH L 2 A LISN B C 1 C 2 0.1µF 0.1µF D1 D4 D2 C 3 C 4 D3 S 1 D 01 S 2 C P L 0 C 0 R 0 R 1 R 2 50 50 V Y V X D 02 DM Current CM Current Figure 2.1 Half-bridge Isolated AC DC Converter with EMI Test Setup For Conducted EMI noise frequency, two 50 H inductors present high impedances and two 0.1 F capacitors present small impedances. The voltages measured across the two 50 impedances are defined by the Conducted EMI. DM noise voltage and CM noise voltage are expressed as Equations (2.1) and (2.2). V CM = (V X V Y )/2 (2.1) V DM = (V X V Y )/2 (2.2) Where V X is line-side EMI and V Y is neutral-side EMI

32 During normal operation, DC link is clamped at a fixed voltage by the capacitances C 3 and C 4. When AC side line voltage is larger than the capacitance voltage, the diode bridge is ON, and when line voltage is smaller than the capacitance voltage, the diode bridge is OFF. The measured DM noise fluctuates with time because of the rectifier diodes which are ON and OFF during half a supply cycle. But CM noise is independent of the conduction state of the rectifier. The DM noise is higher when the rectifier diodes are OFF than the rectifier diodes are ON. The hardware detail of the converter is given in Appendix 1. 2.2.2 Simplification of the Converter Circuit Equivalent circuit model contains essential noise sources and coupling paths. They are adequate for the analysis and prediction of the Conducted EMI behaviors upto 30 MHz. The parasitic plays an important role in the generation of MM noise and CM noise. The stray capacitance existing between MOSFET and heat sink, and the heat sink is normally connected to the ground for safety reasons. The MOSFET is mounted in the heat sink with an electrical insulating material. To reduce thermal resistance, the insulating layer is normally made as thin as possible which results in the formation of a fairly large capacitance between the switching device and the ground. While considering the ground current path, the parasitic capacitance C p is placed between the middle point of the switching cell and the ground, as shown in Figure 2.2. During normal operation, the capacitances C 3 and C 4 clamp the DC link voltage at a fixed value, so that they can be modeled as two DC voltage sources, since C 3 = C 4 in Figure 2.2, so V 1 = V 2. In the following analysis, the EMI measurement ground is taken as the reference point. When S 1 and S 2 are in OFF state, V F = V D = 0, V C = V 1, V E = V 2, V A = V S / 2 and V B = V S / 2.

33 The overall noise is equal to the sum of all components from individual analysis for all harmonics for a linear model. C A D1 D2 V 1 S 1 off Vs D F B S 2 off C 1 C 2 D4 D3 V 2 C P R 1 R 2 E Figure 2.2 Simplified Circuit Model of the Converter 2.2.3 MM Noise Coupling Mode This mode of Conducted EMI is produced during the positive half cycle of the line voltage. This means diodes D 1 and D 3 are reverse biased and the rectifier is OFF. In Figure 2.3, when S 1 and S 2 are both at OFF positions, V F is originally at zero potential. When S 1 goes to the ON position, node F is clamped to V C instantly, resulting in charging of C p through R 1, D 1, S 1 and C p as indicated by thick line in Figure 2.3. Now V C V A = V S /2, so V E is equal to V C (V 1 V 2 ), which is equal to V A > (V 1 V 2 ), and is less than -V S /2. This means V E > V B < V A, D 2 and D 4 are reverse biased. Since only D 1 is conducting, there is noise current flowing through one branch of LISN.

34 C D1 D2 V 1 A S 1 on Vs D F B S 2 off C 1 C 2 D4 D3 V 2 C P V X R 1 R 2 V Y E Figure 2.3 MM Noise Current Path with S 1 turned ON When S 2 goes to the ON position, node F is clamped to V E instantly, resulting in discharging of C p through C p, S 2, D 3, and R 2, as indicated by thick line in Figure 2.4. Now V E V B = V S /2, so V C is equal to V E (V 1 V 2 ), which is now equal to V B (V 1 V 2 ), and is more than V S /2. Thus V C > V A > V B and D 1 and D 3 are reverse biased. Now diode D3 alone is conducting. Switching operation of S 1 generates high-voltage change rates (dv/dt), causing negative (V X is negative) noise current flowing through R 1. However, no current flows through R 2 and V Y = 0. Now DM noise V DM = V X /2 and CM noise V CM = V X /2.

35 C D1 D2 V 1 A S 1 off Vs D F B S 2 on C 1 C 2 D4 D3 V 2 C P V X R 1 R 2 V Y E Figure 2.4 MM Noise Current Path with S 2 turned ON Switching operation of S 2 generates high-voltage change rates (dv/dt), causing positive (V Y is positive) noise current flowing through R 2, and no current flows through R 1, V X = 0. DM noise V DM = V Y / 2, CM noise V CM = V Y / 2. The measured voltage V S1 does not stay at a fixed level when both the switches are in OFF state that is V S1 changes as a slow slope before S 1 or S 2 is turned ON. Because of the unbalanced current flow through two branches of the LISN, DM noise, and CM noise are both measured. This DM is different from the conventional DM coupling because it is not related to the input power current flow, but related to the charging and discharging current of C P. This mode of EMI is called MM.

36 2.2.4 IDM Noise Coupling Mode A pulsating or harmonic rich input current causes IDM Noise. The pure DM noise, which is generated during ON state of the rectifier is called IDM noise. The forward bias of D 1 and D 3 diodes represent the ON state of the rectifier. When S 1 is in ON position, switching operation generates highcurrent slew rates on the primary side of the output transformer, causing noise current flowing through R 1, D 1, C 1, C 2, D 3, R 2 and switch S 1, as indicated by thick line in Figure 2.5. When S 2 goes to ON position, switching operation generates highcurrent slew rates (di/dt) on the primary side of the output transformer, causing noise current flowing through R 1, D 1, C 1, C 2, D 3, R 2 and switch S 2, as indicated by thick line in Figure 2.6. C D1 D2 V 1 A S 1 on Vs D F B S 2 off C 1 C 2 D4 D3 V 2 C P V X R 1 R 2 V Y E Figure 2.5 IDM Noise Current Path with S 1 turned ON

37 C D1 D2 V 1 A S 1 off Vs D F B S 2 on C 1 C 2 D4 D3 V 2 C P V X R 1 R 2 V Y E Figure 2.6 IDM Noise Current Path with S 2 turned ON 2.2.5 CM Noise Coupling Mode In this mode diodes D 1 and D 3 are forward biased, and rectifier is ON state. When S 1 goes to the ON position, node F is clamped to V C instantly; V C finds two paths to charge: (i) C p, R 1, D 1, S 1, C p and (ii) R 2, D 3, C 2, C 1, S 1, C p as indicated by thick line in Figure 2.7. When S 2 goes to ON position, node F is clamped to V E instantly, resulting in the discharge of C p through two paths: (i) C p, S 2, D 3, R 2 and (ii) C p, S 2, C 2, C 1, D 1, R 1 as indicated by thick line in Figure 2.8.

38 C D1 D2 V 1 A S 1 on Vs D F B S 2 off C 1 C 2 D4 D3 V 2 C P V X R 1 R 2 V Y E Figure 2.7 CM Noise Current Path with S 1 turned ON C D1 D2 V 1 A S 1 off Vs D F B S 2 on C 1 C 2 D4 D3 V 2 C P V X R 1 R 2 V Y E Figure 2.8 CM Noise Current Path with S 2 turned ON

39 When S 1 or S 2 switches ON, there exist two paths with almost identical impedances. The two flowing currents through the LISN branches are almost same. This constitutes a pure CM noise coupling and DM noise is not generated. When S 1 is turned ON, V CM = V X /2 = V Y /2, when S 2 is turned ON, V CM = V X /2 = V Y /2. It is similar to MM noise in the realization of CM noise. This phenomenon also confirms that CM noise is independent of the conduction state of the rectifier. From the above, it is observed that the conduction states of rectifier diodes affect the EMI noise propagation path balance. In Continuous Conduction Mode (CCM), the DM noise only has IDM noise, because there are always two diodes conducting simultaneously. The CM noise acts as a more serious problem because the DM noise is attenuated by the dc link capacitors (C 3 and C 4 ). However, for Discontinuous Conduction Mode (DCM), the DM noise is dominated by MM noise because of the unbalanced diode-bridge conduction. The MM noise also acts as a serious problem because of the similar CM generation mechanism. 2.3 MODEL IMPLEMENTATION Figure 2.9 shows an equivalent circuit model for description of the essential coupling paths between the converter and the three physical terminals L, N, and G of LISN. The linear model for the converter physical circuit consists of simple combination of three impedance elements together with three voltage sources.

40 L Z scm1 R N R N V X - - V Y LISN G N V - scm1 - - V scm2 Z scm2 Z sdm V sdm Converter noise equivalent model Figure 2.9 Noise Model for Essential Coupling path The overall noise is equal to the sum of all components that arise out of the individual analysis of a linear model. This helps in maintaining the external responses at the LISN input terminals. Each phase of the LISN can be represented by 50 impedances. I scm represents the SPC noise source current and Z scm represents the noise source impedance. Z sdm is the DM noise source impedance of the SPC. EMI is coupled with three voltage sources through three impedances: V sdm and Z sdm for DM noise and V scm1, V scm2, Z scm1, and Z scm2 for CM noise. LISN is represented by two resistors R N. These voltage sources depend on the high dv/dt and high di/dt slew rates and circuit parasitic parameters, device package, and layout. Normally Z scm1 and Z scm2 are high source impedances because they are associated with parasitic capacitance to the earth. Assume that Z scm1 >>50 and Z scm2 >>50, obtain the simplified noise model in Figure 2.10.

41 L R N V X I scm1 - G - I sdm R N V Y LISN N I scm2 Converter noise equivalent model Figure 2.10 Simplified Noise Model of Figure 2.9 The CM and DM currents are expressed in Equations (2.3) to (2.5). V scm1 I scm1 = (2.3) Z scm1 V scm2 I scm2 = (2.4) Z scm2 V sdm I sdm = (2.5) Z sdm The effectiveness of an EMI filter depends not only on the filter itself but also on the noise source impedance. For CM noise, the source is modeled by a current source in parallel with high source impedance Z s. For DM noise, the source is modeled by a voltage source in series with low impedance or a current source in parallel with high impedance depending on the state of the input rectifier diodes. When two of the rectifier diodes are conducting, the

42 noise source is modeled by a voltage source in series with a low impedance source Z s. When all the four diodes are cut off, the noise is modeled by a current source in parallel with a high source impedance Z p. This results in the higher fluctuation of DM noise which is twice than its fluctuation in case of line frequency. Z s is associated with wire inductance and resistance, and Z p is associated with diode parasitic capacitance. These source impedances depend on parasitic parameters and are therefore package dependent. 2.3.1 EMI Model for Symmetrical Circuit The model in Figure 2.9 contains all essential coupling paths and circuit parameters that can provide a full picture of the EMI conduction and coupling mechanism. The DM voltage (with R N = 50 ) can be written in Equation (2.6). 100 VsDM 50 ZsDM V DM = (IsCM1 - I scm2) Z sdm 100 Z sdm 100 (2.6) As per the Equation (2.6), total DM noise has two parts: (1) the first part is determined by DM noise source V sdm and impedance Z sdm called as Intrinsic Differential Mode (IDM) noise and (2) the second part is the difference of two CM current sources through the DM impedance and LISN, called as Non-Intrinsic Differential Mode (NIDM) or Mixed Mode (MM) noise. The MM noise is caused by unbalanced CM current which flows through the two LISN branches. By following the above deducing method, the DM and CM noise source model can be reduced to a simple two-port lumped circuit model, as shown in Figure 2.11.

43 Z N Z S I S LISN Converter Figure 2.11 Simplified EMI Noise Equivalent Circuit Model Z N is 2R N in case of DM and 0.5R N in case of CM, Z S is Z sdm in case of DM and 0.5Z scm2 in case of CM, I S is I sdm in case of DM and I scm1 in case of CM. For reducing the complexity, the EMI issues using very simple models always consider the combination of impedance Z S and source V S. 2.3.2 Equivalent Circuit Model for Conducted EMI Figure 2.12 shows equivalent circuit models for description of MM, IDM, and CM noise coupling paths between the converter s and the LISN s physical terminals. 50 V MM Z 1 V 1 - Figure 2.12 (a) MM Noise Equivalent Circuit Model

44 50 V IDM - Z 2 V2 50 Figure 2.12 (b) IDM Noise Equivalent Circuit Model L CM 25 - V CM-LCM Z 3 V 3 Figure 2.12 (c) CM Noise Equivalent Circuit Model EMI is coupled with three voltage sources through three impedances, i.e., V 1 and Z 1 for MM coupling, V 2 and Z 2 for IDM coupling, and V 3 and Z 3 for CM coupling. There are three coupling impedances associated with the different modes of the EMI, i.e., Z 1 Z 2, and Z 3. They can be approximated by the corresponding RLC circuits. The noise excitation sources are modeled as Thevenin equivalent voltages V 1, V 2 and V 3, which are terminal voltages measured at the corresponding open points of the converter. From the models shown in Figure 2.12, the noise voltages of the three modes can be written in terms of the equivalent noise voltage sources and coupling impedances as in Equations (2.7) to (2.9).

45 50 V ( ) 1 V MM ( ) = (2.7) 50 Z 1 ( ) 50 V ( ) 2 V IDM ( ) = (2.8) 100 Z 2 ( ) 25 V ( ) 3 V CM ( ) = (2.9) 25 Z 3 ( ) 2.4 PARAMETER ESTIMATION AND EMI PREDICTION The estimation of parameters in the models obtained by measurements, and the significance of each coupling path in producing Conducted EMI are described here. MM noise is generated when charging and discharging current of C p flows through one LISN branch. Equivalent noise voltage sources, V 1 and V 2 are the same when fast dv/dt is impacted on C p. The only difference between the two modes is that their coupling impedances, Z 1 and Z 3, are different. Figure 2.13 (a) Measured Impedance of MM Noise

46 Figure 2.13 (b) Measured Impedance of IDM Noise Figure 2.13 (c) Measured Impedance of CM Noise

47 Impedance Z 1 is the impedance between point A and the ground by shorting D 1 and S 1. DM impedance Z 2 is measured between point A and B by shorting D 1, D 3 and S 1. For the CM noise coupling, the two input phases of the converter are effectively in parallel, and Z 3 is the impedance between AB and the ground by shorting D 1, D 3 and S 1. An Agilent 4294A (40 110 MHz) precision impedance analyzer is used to perform the impedance measurements. The measured Z 1, Z 2 and Z 3 for the test converter are shown in Figure 2.13. From measured impedances, Z 1 and Z 3 behave like two capacitances and Z 2 like an inductance. The component values of the RLC circuits are given in Table 2.1. Table 2.1 Parameter Values of the Equivalent Circuit Models Component Values of the RLC Circuits Symbol R L C Z 1 15 900nH 150pF Z 2 800nH 120µF Z 3 13 600nH 165pF C x 20m 60nH 120nF L CM 50 10mH 10pF EMI in a Thevenin is the terminal voltage measured when the corresponding external part of the system is an open circuit. The CM equivalent noise sources V 1 and V 3 require the disconnection of the ground conductor from heat sink to LISN. The voltage is then measured using Tektronix P5205 (1300 V/100 MHz) differential voltage probe connected between F and the middle point of the LISN, is shown in Figure 2.14. Figure 2.15 shows the equivalent noise current I SDM flowing through the

48 primary side of the output transformer measured by Tektronix TCP305/A300 (50 A/50 MHz) current probe. Then, DM equivalent noise source V 2 is measured using Equation (2.10). V 2( ) = I SDM( ) / (j C 1) (2.10) Figure 2.14 Measured Noise Voltages V 1 and V 3 Figure 2.15 Measured Noise Current I sdm

49 C D1 D2 V 1 Vs A C X B LCM D S 1 S 2 F C 1 C 2 D4 D3 V 2 C P R 1 R 2 E Figure 2.16 Filter Setup of X capacitance and CM chokes Once the noise equivalent circuits for the converter system are known, it is also possible to predict the EMI, when filters are introduced between the converter and LISN. To investigate this, a simple filter element is included in the experimental setup. The conventional filtering design using X capacitance to suppress DM (MM and IDM) noise, and CM choke to suppress CM noise setup is shown in Figure 2.16. Based on the models in Figure 2.12, the MM, IDM and CM equivalent circuits with filter in place are shown in Figure 2.17. Cx Z 1 V 1 50 V MM-Cx - 50 Figure 2.17 (a) MM Noise Equivalent Circuit Model with EMI Filter

50 50 50 V Z 2 IDM-Cx - V 2 Cx Figure 2.17 (b) IDM Noise Equivalent Circuit Model with EMI Filter 50 50 V CM Z 3 V 3 - Figure 2.17 (c) CM Noise Equivalent Circuit Model with EMI Filter Impedance characteristics of X capacitance and choke are measured and represented by lumped parameters circuits. Equations (2.7) to (2.9) are modified as Equations (2.11) to (2.13) to compute noise voltages. 50 Cx( ) V ( ) 1 V MM - Cx ( ) = Z 1 ( ) (100 Cx( )) 50 (50Cx( )) (2.11) 50 Cx( ) V ( ) 2 V IDM - Cx ( ) = Z 2 ( ) (100 Cx( )) 100 Cx( ) (2.12) 25 V ( ) 3 V CM - L ( ) = CM Z 3 ( ) L CM ( ) 25 (2.13)

51 Figure 2.18 (a) Measured MM Noise Spectrum with Filter Figure 2.18 (b) Measured IDM Noise Spectrum with Filter

52 Figure 2.18 (c) Measured CM Noise Spectrum with Filter Figure 2.18 shows the measured spectrum of MM, IDM and CM noise voltages with EMI filters. All EMI spectrum displayed on the spectrum analyzer are employed by a peak detector and a 10 KHz resolution bandwidth. The measured results illustrate that the information of impedances and voltage sources for a particular power converter is valid in predicting the actual EMI filter s effect. The proposed model provides a simple way to make an initial estimation of the interference. The parameter values can never be estimated accurately. These results show that once the noise model is determined, the proposed method is used to investigate the effectiveness of filtering techniques.

53 2.5 CONCLUSION Equivalent circuit modeling method is proposed to represent the Conducted EMI coupling of a SPC. Three dominant modes of EMI noise coupling are analyzed and investigated and the essential coupling models have been described for the noise coupling based on the measurements. Once the parameters of the EMI model are identified, the process can predict the actual attenuation of a particular EMI filter. The equivalent noise voltage sources as well as the coupling impedances are measured separately. Experimental Results show that the proposed method is very effective and accurate in identifying and capturing EMI features in power converters. The method presented is not only limited to half-bridge converters, but it can also be applied to many different converter topologies, such as buck, fly back, boost, with single-phase diode bridge frontend. This model is convenient to use, contains the salient features of Conducted EMI, and gives adequate prediction of EMI behavior in switching converters. Table 2.2 shows the hardware results. The results obtained satisfy the Federal Communications Commission (FCC) class B regulations. Table 2.2 Comparison of Hardware Results using Equivalent Circuit Modeling Method Sl. No. Frequency in MHz Without Filter (dbµv) With Filter (dbµv) MM IDM CM 1 10 83 40 37 38 2 20 80 30 28 27 3 30 76 25 23 20