Frequency division inductor structure for improving filtering effect of EMI (Electro-Magnetic Interference) filter and overall design method

Abstract

The invention belongs to the technical field of EMI filters, and relates to a frequency division inductor structure for improving the filtering effect of an EMI filter and an overall design method.The frequency division inductor structure comprises a primary frequency division inductor arranged close to the EUT side, a secondary frequency division inductor arranged close to the LISN side, and a decoupling module connected to a pin of a differential mode capacitor in series; the overall design method comprises the steps that a traditional passive filter is designed according to EUT bare noise, a primary frequency division inductor and a secondary frequency division inductor are designed according to EUT high-frequency noise and the high-frequency impedance characteristic of a traditional common mode inductor, and a decoupling module is connected into the filter in series; according to the invention, the suppression effect of the filter on low-frequency noise and high-frequency noise can be greatly improved only by increasing the volume by less than 10% while the design of a traditional passive EMI filter is not influenced, so that the cost and the volume of a filtering module are greatly reduced while the suppression capability of a high-frequency band is effectively improved.

Description

Technical Field

[0001] The invention belongs to the technical field of EMI filters, and relates to a frequency-dividing inductor structure and an overall design method for improving the filtering effect of an EMI filter. Background Art

[0002] With the rapid development of the electronics industry, the increase in switching frequency has become an unavoidable trend. Although this progress has brought higher efficiency and smaller component size, it is also accompanied by a series of new challenges. Among them, the problem of electromagnetic interference (EMI) is particularly prominent. High-frequency switching operations will generate stronger electromagnetic radiation, which not only threatens the normal operation of the electronic equipment itself, but also may interfere with other surrounding electronic equipment and even affect the stability of the communication system. Therefore, how to effectively suppress EMI has become an unavoidable technical problem in the electronics industry. In the process of dealing with EMI problems, the design and optimization of EMI filters are particularly important. As the main means of suppressing electromagnetic interference, EMI filters can effectively reduce the propagation of high-frequency signals and reduce the impact of interference on sensitive circuits and external devices. Traditional EMI filters are divided into passive EMI filters and active EMI filters. Since the high-frequency filtering effect of active EMI filters is limited by the gain-bandwidth product of the op amp, and the cost is high, the power supply system is also more complicated, so more stable and cheaper passive EMI filters are generally used in the industry.

[0003] However, with the further increase in switching frequency, traditional passive EMI filtering designs can no longer meet more stringent EMI suppression requirements. Traditional passive EMI filters have the following technical defects: first, the low-frequency suppression effect is limited by volume. In order to achieve effective low-frequency interference suppression, larger inductors and capacitors are often required. However, with the miniaturization of electronic equipment, the volume of the filter is limited; second, the high-frequency suppression effect is limited by the high-frequency parasitic parameters of the filter device. In the high-frequency band, non-ideal factors such as parasitic inductance and capacitance of the filter device will significantly affect the performance of the filter, seriously weakening its ability to suppress high-frequency noise.

[0004] Therefore, a smaller, lower-cost, and more effective EMI filter is needed to solve this technical problem. Summary of the invention

[0005] The technical solution adopted by the present invention to solve the technical problem is: a frequency division inductor structure for improving the filtering effect of an EMI filter, comprising: a primary frequency division inductor, a secondary frequency division inductor and a decoupling module;

[0006] The primary frequency division inductor is used to filter out the high-frequency noise generated by the EUT;

[0007] The secondary frequency division inductor is used to correct the high frequency characteristics of the common mode inductor; the frequency point which is greater than the first resonant frequency of the common mode inductor and at which the impedance is reduced to 70% of the peak impedance is taken as the design frequency point of the secondary frequency division inductor;

[0008] The decoupling module is used to decouple the coupling problem between the frequency division inductor and the common mode inductor in the low frequency band by utilizing the damping characteristics of the magnetic beads, correct the quality factor of the filter, eliminate the resonance peak of the filter insertion loss curve, and thus improve the noise suppression effect of the filter in the low frequency band;

[0009] The primary frequency division inductor is arranged close to the EUT side, the secondary frequency division inductor is arranged close to the LISN side, and the decoupling module is connected in series to the pins of the differential mode capacitor;

[0010] The secondary frequency division inductor of the present invention is a high-frequency inductor added near the LISN side to correct the impedance characteristics of the traditional common-mode inductor. It is mainly used to correct the high-frequency impedance characteristics of the traditional common-mode inductor and filter out the high-frequency noise generated by the coupling of the traditional common-mode inductor with the differential mode capacitor and the common mode capacitor. After the first resonant frequency, before the impedance of the traditional common-mode inductor drops to 70% of the impedance at the first resonant frequency, the traditional common-mode inductor still plays a good role in filtering out common-mode noise. After that, the filtering effect of the traditional common-mode inductor is greatly reduced, and it is necessary to add a secondary frequency division inductor to correct the impedance of its high-frequency band. Therefore, the frequency at which the impedance of the common-mode inductor of the present invention drops to 70% of the peak impedance after the first resonant frequency is the design frequency of the secondary frequency division inductor;

[0011] The present invention uses a decoupling module to decouple the coupling problem between the frequency division inductor and the traditional common mode inductor in the low frequency band. The turning frequency of the insertion loss curve of the traditional filter is around 10kHz, and after adding the frequency division inductor, a turning frequency will be added in the test frequency band. At this frequency point, a resonance peak will be generated due to the excessive quality factor, thereby deteriorating the effect of the filter at low frequencies. The present invention uses the damping characteristics of the magnetic beads to correct the quality factor, thereby eliminating the resonance peak of the filter insertion loss curve and further improving the noise suppression effect of the filter in the low frequency band.

[0012] Preferably, the primary frequency division inductor is used to filter out high-frequency noise above 5MHz generated by the EUT; the first resonant frequency of the impedance characteristics of traditional common-mode inductors such as common-mode inductors of manganese-zinc ferrite cores is mostly in the hundreds of kHz, and even the common-mode inductors of amorphous cores with greater magnetic permeability have a first resonant frequency of impedance characteristics around 2 to 3MHz. The impedance characteristic curve of traditional common-mode inductors will begin to decline after the first resonant frequency, and the impedance decreases as the frequency increases, showing capacitance. Therefore, it is generally believed that the filtering benefits brought by traditional common-mode inductors after 5MHz are very small, and may even worsen the noise. Moreover, 5MHz is also the turning frequency of EMI test standards EN55022 and CISPR22. Therefore, the present invention uses 5MHz as the main frequency point of the primary frequency division inductor design, and the primary frequency division inductor mainly filters out high-frequency noise above 5MHz generated by the EUT.

[0013] Preferably, the secondary frequency division inductor is also used to filter out high-frequency noise caused by coupling between the common-mode inductor and the differential-mode capacitor or the common-mode capacitor.

[0014] The present invention also discloses a general design method for improving the filtering effect of an EMI filter. The general design method is used for the above-mentioned frequency-dividing inductor structure. The general design method comprises the following steps:

[0015] Step S1, designing a traditional passive filter according to the bare noise of the EUT;

[0016] Step S2, designing a primary frequency division inductor and a secondary frequency division inductor respectively according to the high frequency noise of the EUT and the high frequency impedance characteristics of the traditional common mode inductor;

[0017] Step S3: inserting a decoupling module in series into the filter.

[0018] Preferably, in step S2, the step of designing a primary frequency division inductor includes:

[0019] Step S2-A1, determining a target insertion loss according to a noise spectrum of a noise source;

[0020] Step S2-A2, filtering out the high frequency noise of the EUT through a primary frequency division inductor;

[0021] Step S2-A3, obtaining the primary frequency division inductance value according to the noise source impedance and the load impedance.

[0022] More preferably, in step S2-A1, the noise spectrum of the noise source is 5 MHz to 30 MHz.

[0023] Preferably, in S2-A2, the common mode insertion loss calculation formula is:

[0024]

[0025] In formula (1), L df1 is the first-order frequency division inductance, C y is the common mode capacitor value, R s is the noise source impedance, R L is the load impedance and w is the angular frequency.

[0026] Preferably, in step S2, the step of designing a secondary frequency division inductor includes:

[0027] Step S2-B1, obtaining a high-frequency equivalent model of the common-mode inductor according to its impedance characteristic curve;

[0028] Step S2-B2, correcting the high frequency characteristics of the common mode inductor by using the secondary frequency division inductor;

[0029] Step S2-B3, obtaining the secondary frequency division inductance value according to the high frequency parasitic parameters of the common mode inductor.

[0030] More preferably, in step S2-B1, the high-frequency equivalent model is an inductor in parallel with a capacitor in parallel with a resistor, and the impedance formula of the high-frequency equivalent model is:

[0031]

[0032] In formula (2), Z L is the common mode inductor impedance, C eq is the high frequency parasitic capacitance of the common mode inductor, R eq

[0033] is the high frequency equivalent resistance of the common mode inductor.

[0034] More preferably, in step S2-B2, the value of the secondary frequency division inductor complies with the following formula:

[0035]

[0036] In formula (3), Z is the impedance value of the common mode inductor and the secondary frequency division inductor in series, L df2 is the secondary frequency division inductor value, L is the common mode inductor value, C eq is the high frequency parasitic capacitance of the common mode inductor, R eq is the high frequency equivalent resistance of the common mode inductor, w f is the first resonant frequency of the common mode inductor.

[0037] The beneficial effects of the present invention are:

[0038] 1. The present invention improves the filter design process to address the common phenomenon that the filtering effect of conducted frequency band electromagnetic interference deteriorates in the high frequency band. Compared with the traditional EMI filter design, the high frequency band suppression capability is effectively improved while the cost and volume of the filter module are greatly reduced.

[0039] 2. The present invention takes better account of the high-frequency characteristics of the traditional common-mode inductor, corrects the impedance characteristics of the traditional common-mode inductor through the secondary frequency division inductor, and ensures the effectiveness of its suppression effect in the high-frequency band.

[0040] 3. The present invention implements a decoupling design to solve the coupling problem between the added frequency division inductor and the traditional common mode inductor in the low frequency band, thereby ensuring the effective suppression of low frequency band noise. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a method flow chart of the frequency division inductor structure and overall design method for improving the filtering effect of the EMI filter of the present invention;

[0042] Figure 2 It is a circuit topology diagram of the present invention with a two-frequency division inductor;

[0043] Figure 3 It is a schematic diagram of the EMI filter circuit using a two-way frequency inductor decoupling design of the present invention;

[0044] Figure 4 It is a schematic diagram of an EMI filter with a two-frequency inductor according to the present invention;

[0045] Figure 5 It is a rendering of the EMI filter with a primary frequency division inductor according to the present invention;

[0046] Figure 6 It is a rendering of the EMI filter with a secondary frequency division inductor according to the present invention;

[0047] Figure 7 is the bare noise spectrum of the EUT in the embodiment of the present invention;

[0048] Figure 8 is the noise spectrum after adding a traditional passive filter in the embodiment of the present invention;

[0049] Fig. 9 is the noise spectrum after adopting the two-way inductor decoupling design in the embodiment of the present invention;

[0050] Fig.10 is a comparison spectrum of the EMI filter with and without a primary frequency division inductor in the embodiment of the present invention;

[0051] Fig.11 is a comparison spectrum of the EMI filter with and without a secondary frequency division inductor in the embodiment of the present invention;

[0052] Fig.12 is the insertion loss curve of the filter with the decoupling module added in the present invention;

[0053] Fig.13 is a comparison spectrum of the EMI filter with and without a decoupling module in the embodiment of the present invention;

[0054] Fig.14 It is a physical picture of the two-way inductor decoupling design adopted in the embodiment of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the relevant technologies in the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] refer to Figures 1 to 14 In this embodiment, the two-way inductor decoupling design method applied to the EMI filter includes the following steps: first, a traditional passive filter is designed according to the bare noise of the EUT, then, the primary and secondary frequency inductors are designed according to the high-frequency noise of the EUT and the high-frequency impedance characteristics of the traditional common-mode inductor, and finally, a decoupling module is connected in series in the filter. Figure 2 The circuit topology diagram with the two-way frequency inductor can be obtained.

[0057] In this embodiment, the EUT is a flyback charger with a full load of 135W, with an AC input of 220V and a DC output of 19V under working conditions, and a topology of PFC+LLC, which is a typical AC-DC converter. Figure 7 is the bare noise spectrum of the EUT. Based on the raw noise, the insertion loss of the required EMI filter can be confirmed. After measuring the noise source impedance, the main topology of the filter used can be confirmed based on the impedance mismatch principle. Figure 3 The circuit topology in the middle is the traditional passive filter topology selected in the embodiment. Then the capacitance can be determined according to the safety requirements, and the inductance value can be determined according to the insertion loss and cutoff frequency, and finally Figure 8 The noise spectrum of adding a traditional passive filter is shown. Figure 8It can be seen that today's traditional passive filters have two major technical defects. First, the low-frequency suppression effect is limited by the volume of the filter. If you want to make the low-frequency suppression effect better, you need to use larger capacitors and inductors or increase the number of filter stages, which will inevitably greatly increase the volume of the filter; second, the high-frequency suppression effect is limited by the high-frequency parasitic parameters of the device. High-frequency noise is mainly provided by common-mode noise. When the frequency increases to the first resonant frequency of the traditional common-mode inductor, the traditional common-mode inductor becomes capacitive, and the impedance begins to decrease, which greatly reduces the filtering effect and makes it difficult to meet the standard requirements. At this time, it is necessary to seek a common-mode inductor with better high-frequency characteristics, which will inevitably increase the design cost of the filter. In order to solve the above problems, the EMI filter design method based on the two-way inductor decoupling design of this embodiment only needs to increase the volume of the original filter by 10% to solve the two major technical problems mentioned above. Reference Fig. 9 As shown, it can be seen that the low-frequency and high-frequency noise spectrum are greatly reduced after adopting the two-way inductor decoupling design.

[0058] refer to Figure 5 It can be seen that the frequency band and effect of the primary frequency division inductor. The first resonant frequency of traditional common-mode inductors is mostly around several hundred kHz. The first resonant frequency of common-mode inductors with amorphous cores that have better high-frequency characteristics is also 2-3MHz. After exceeding the first resonant frequency, the traditional common-mode inductor becomes capacitive and the impedance becomes smaller. Therefore, this implementation method takes into account the high-frequency characteristics of traditional common-mode inductors and assumes that traditional common-mode inductors do not work in the frequency band after 5MHz. Therefore, a primary frequency division inductor is added on the EUT side to filter out high-frequency noise. It is known from previous measurements that the high-frequency common-mode noise of the EUT is around 80-90dB. The primary frequency division inductor and the common-mode capacitor are regarded as a first-order common-mode filter. The specific common-mode insertion loss calculation formula is as follows:

[0059]

[0060] Where L df1 is the first-order frequency division inductance, C y is the common mode capacitor value, R s is the noise source impedance, R L is the load impedance and w is the angular frequency.

[0061] According to the required insertion loss at different frequencies, the corresponding primary frequency division inductor values ​​can be obtained as shown in Table 1:

[0062] Table 1 Relationship between primary frequency division inductance values

[0063]

[0064] The load impedance is set to 25Ω, the noise source impedance is the measured value, and considering the need to leave a margin, the primary frequency division inductor is finally set to 300μH. Fig.10 It can be seen that adding a primary frequency division inductor on the EUT side can have a very high benefit at high frequencies above 2MHz.

[0065] refer to Figure 6 It can be seen that the frequency band and effect of the secondary frequency inductor are used. The secondary frequency inductor is mainly used to correct the high-frequency characteristics of the traditional common-mode inductor and filter out the high-frequency noise generated by the coupling of the traditional common-mode inductor with the differential mode capacitor and the common mode capacitor. The common-mode high-frequency equivalent model of the traditional common-mode inductor is an inductor in parallel with a capacitor in parallel with a resistor, so its impedance formula is:

[0066]

[0067] Where Z L is the traditional common mode inductor impedance value, C eq is the high frequency parasitic capacitance of the traditional common mode inductor, R eq It is the high frequency equivalent resistance of the traditional common mode inductor.

[0068] Considering that the traditional common-mode inductor still plays a good filtering role after the first resonant frequency and before its impedance drops to 70% of the peak impedance, the present invention defines the frequency point at which the impedance of the traditional common-mode inductor drops to 70% of the peak impedance as the design frequency point of the secondary frequency division inductor, and the value of the secondary frequency division inductor complies with the following formula:

[0069]

[0070] Where Z is the impedance value of the traditional common mode inductor and the secondary frequency division inductor in series, L df2 is the inductance of the secondary frequency division inductor, L is the inductance of the traditional common mode inductor, C eq is the high frequency parasitic capacitance of the traditional common mode inductor, R eq is the high frequency equivalent resistance of the traditional common mode inductor, w f It is the first resonant frequency of the traditional common mode inductor.

[0071] Based on the measured high-frequency parasitic parameters of the traditional common-mode inductor, and with appropriate margin, the final secondary frequency division inductor value is 100μH. Fig.11 It can be seen that adding a secondary frequency division inductor close to the LISN side can further improve the filtering effect in the high frequency band.

[0072] refer to Fig.12It can be seen that after adding the decoupling module, the resonance peak of the filter insertion loss curve at 200kHz is eliminated. This is because the decoupling module uses the damping characteristics of the magnetic beads to decouple the coupling of the traditional common-mode inductor and the frequency division inductor at low frequencies, correcting the quality factor, thereby eliminating the resonance peak of the filter insertion loss curve caused by the excessive quality factor, further improving the noise suppression effect of the filter in the low-frequency band. Fig.13 It can be seen that the filtering effects of both low-frequency and high-frequency bands are significantly improved after adding the decoupling module.

[0073] In summary, the present invention can greatly improve the filter's suppression effect on low-frequency noise and high-frequency noise without affecting the design of traditional passive EMI filters, and only increases the volume by less than 10%, thereby bringing higher benefits to the filter at high and low frequencies. Therefore, the present invention has broad application prospects.

[0074] It should be emphasized that the above are only preferred embodiments of the present invention and do not limit the present invention in any form. Any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A frequency-dividing inductor structure for improving the filtering effect of an EMI filter, characterized in that: include: Primary frequency division inductor, secondary frequency division inductor and decoupling module; The primary frequency division inductor is used to filter out the high frequency noise generated by the EUT; The secondary frequency division inductor is used to correct the high frequency characteristics of the common mode inductor; The decoupling module is used to decouple the coupling problem between the frequency division inductor and the common mode inductor in the low frequency band by utilizing the damping characteristics of the magnetic beads, correct the quality factor of the filter, and eliminate the resonance peak of the filter insertion loss curve; The primary frequency division inductor is arranged close to the EUT side, the secondary frequency division inductor is arranged close to the LISN side, and the decoupling module is connected in series to the pins of the differential mode capacitor.

2. The frequency division inductor structure for improving the filtering effect of the EMI filter according to claim 1, characterized in that: The primary frequency division inductor is used to filter out high frequency noise above 5 MHz generated by the EUT.

3. The frequency division inductor structure for improving the filtering effect of the EMI filter according to claim 1, characterized in that: The secondary frequency division inductor is also used to filter out high frequency noise caused by coupling of the common mode inductor and the differential mode capacitor or the common mode capacitor.

4. A general design method for improving the filtering effect of an EMI filter, characterized in that: The decoupling method is used for the frequency division inductor structure according to any one of claims 1 to 3, and the overall design method comprises the following steps: Step S1, designing a traditional passive filter according to the bare noise of the EUT; Step S2, designing a primary frequency division inductor and a secondary frequency division inductor respectively according to the high frequency noise of the EUT and the high frequency impedance characteristics of the traditional common mode inductor; Step S3: inserting a decoupling module in series into the filter.

5. A general design method for improving the filtering effect of an EMI filter according to claim 4, characterized in that: In step S2, the step of designing a primary frequency division inductor includes: Step S2-A1, determining a target insertion loss according to a noise spectrum of a noise source; Step S2-A2, filtering out the high frequency noise of the EUT through a primary frequency division inductor; Step S2-A3, obtaining the primary frequency division inductance value according to the noise source impedance and the load impedance.

6. A general design method for improving the filtering effect of an EMI filter according to claim 5, characterized in that: In the step S2-A1, the noise spectrum of the noise source is 5 MHz to 30 MHz.

7. The overall design method for improving the filtering effect of an EMI filter according to claim 5, characterized in that: In S2-A2, the common mode insertion loss calculation formula is: In formula (1), L df1 is the first-order frequency division inductor value, C y is the common mode capacitor value, R s is the noise source impedance, R L is the load impedance and w is the angular frequency.

8. The overall design method for improving the filtering effect of an EMI filter according to claim 4, characterized in that: In step S2, the step of designing the secondary frequency division inductor includes: Step S2-B1, obtaining a high-frequency equivalent model of the common-mode inductor according to its impedance characteristic curve; Step S2-B2, correcting the high frequency characteristics of the common mode inductor by using the secondary frequency division inductor; Step S2-B3, obtaining the secondary frequency division inductance value according to the high frequency parasitic parameters of the common mode inductor.

9. A general design method for improving the filtering effect of an EMI filter according to claim 8, characterized in that: In step S2-B1, the high-frequency equivalent model is an inductor in parallel with a capacitor in parallel with a resistor, and the impedance formula of the high-frequency equivalent model is: In formula (2), Z L is the common mode inductor impedance, C eq is the high frequency parasitic capacitance of the common mode inductor, R eq is the high frequency equivalent resistance of the common mode inductor.

10. The overall design method for improving the filtering effect of an EMI filter according to claim 8, characterized in that: In step S2-B2, the value of the secondary frequency division inductor complies with the following formula: In formula (3), Z is the impedance value of the common mode inductor and the secondary frequency division inductor in series, L df2 is the secondary frequency division inductor value, L is the common mode inductor value, C eq is the high frequency parasitic capacitance of the common mode inductor, R eq is the high frequency equivalent resistance of the common mode inductor, w f is the first resonant frequency of the common mode inductor.

Images