Active Filter for Reducing Electromagnetic Interference

Abstract

An active filter can reduce electromagnetic interference (EMI) generated by the current flowing through the power line. The active filter is connected to the power line at a single node through a coupling capacitor. When the power line current changes, an induced current flows through the coupling capacitor. This induced current is applied to a gain control circuit having cross-coupled PNP transistors, which drives the current to both ends of a variable capacitor. The variable capacitor converts these currents into a voltage, which is injected back into the power line as an injection compensation voltage through the coupling capacitor to compensate for the induced current.

Description

[0001]

Related Applications

[0002] This application is a Continuation-In-Part (CIP) of U.S. Application No. 17 / 091,089, filed on Nov. 6, 2020, titled “Active Filter for Reducing Electromagnetic Interference (EMI) Using a Single Connection Point and a Negative Impedance Converter”.

Technical Field

[0003] The present invention relates to active filter circuits, and more particularly to active filters for single-point induction and injection for reducing electromagnetic interference (EMI).

Background Art

[0004] Filters are used in various circuits. For example, the power transistors of a switched-mode power supply (SMPS) can be switched on and off rapidly, e.g., at several hundred kHz. These SMPSs can power household appliances, computers, medical devices, telecommunication systems, automotive systems, and many other applications.

[0005] However, the rapid switching of the transistors in an SMPS can generate electromagnetic interference (EMI) in other devices. Some standards have been established to limit this EMI, such as the EN55022 standard that many SMPSs must pass.

[0006] Traditionally, passive EMI filters are added to power supplies and other devices that generate EMI. Since these passive EMI filters rely on inductors and capacitors, they tend to be large and bulky. Although passive EMI filters are simple and effective in reducing electromagnetic interference, they can be large, bulky, and have power losses.

[0007] FIG. 1 shows a prior art active EMI filter that has separate induction and injection points. The active EMI filter can use an operational amplifier with a more complex circuit to reduce the size and volume of the EMI filter. The positive input POS_IN is filtered to produce a positive output POS_OUT with reduced EMI. When the load current changes, the capacitor 108 between POS_OUT and the ground GND_OUT helps to keep the supply voltage on POS_OUT constant.

[0008] The power current from POS_IN is induced by the inductive inductor 102. The current changes, such as an increase in the power current flowing through the primary winding of the sense inductor 102, will cause a current in the same direction in the secondary winding of the inductive inductor 102. The inductive inductor 102 increases the voltage at the inverting input terminal of the operational amplifier 110, thereby generating an opposite voltage at the bottom terminal of the capacitor 104. The charge on the capacitor 104 generates a positive voltage at the top terminal of the capacitor 104. Since the current sinks into the output of the operational amplifier through the capacitor 104, the voltage on the injection inductor 106 will inject a compensation voltage through the secondary winding of the injection inductor 106, thereby eliminating the noise voltage in the main circuit. Therefore, the noise voltage is reduced by the active filter.

[0009] The active filter has independent induction points and injection points. The inductive inductor 102 senses the current change in the power supply, while the injection inductor 106 injects a reverse current into the power line to compensate for the induced voltage. Although the active filter is smaller than the passive filter, having independent induction points and injection points still requires two bulky inductors. Some active filters replace one of the inductive inductor 102 and the injection inductor 106 with a capacitor, but the remaining inductor is still large in size and thus not advisable. The large inductor not only takes up a lot of space and is heavy, but also has bandwidth limitations due to the parasitics, non-ideal coupling, and self-resonance of the inductor.

[0010] Figure 2 An active filter with a single-point connection is highlighted. The parent application describes an active filter with a single-point connection. The node N0 is a single node on the line 10, and the line 10 can be a power line in the SMPS or a line filtered to reduce noise, such as the noise generated by EMI from the line 10. The connection capacitor 20 is connected to the node N0 on the line 10. The other end of the connection capacitor 20 is connected to the negative capacitance circuit 120.

[0011] The connection capacitor 20 acts as both a sensor and an injector of the active filter. The change in the current flowing along the line 10 causes an induced current I SENSE to flow from the line 10 through the connection capacitor 20 to the negative capacitance circuit 120. In response, the negative capacitance circuit 120 generates a voltage change V INJECT , which is applied to the connection capacitor 20 and injected back into the line 10. V INJECT compensates for I SENSE to reduce the current fluctuation in the line 10, thereby reducing the EMI generated by the line 10.

[0012] The impedance of the negative capacitance circuit 120 is Z=V / I=-1 / (sC), where V and I are the AC voltage drop and current on the negative capacitance circuit 120, s is a proportional constant, and C is an effective capacitance. In the frequency domain, for a normal positive capacitance, when the current I flows through the capacitor, the voltage drop on the capacitor will be (1 / sC)×I, where the coefficient (1 / sC) is the impedance of the capacitor, and in the time domain, the current will lead the voltage by 90 degrees. In the case of negative capacitance, the voltage drop on the capacitor will be (-1 / sC)×I, where the coefficient (-1 / sC) is the impedance of the negative capacitance circuit, and in the time domain, the current will lag the voltage by 90 degrees. In this way, the voltage drops of the positive and negative capacitances will be in anti-phase (180 degrees out of phase).

[0013] When the inductive current I SENSE according to Figure 2 When the current flows through the connection capacitor 20 and the negative capacitance circuit 120 to the ground in the direction shown, the negative capacitance circuit 120 acts as a negative capacitor, so the negative capacitance circuit 120 generates a negative voltage drop across its two ends instead of a positive voltage drop. The negative capacitance circuit 120 acts as a current-controlled-voltage source, and its input impedance is Z = -1 / (sC).

[0014] Since the bottom terminal of negative capacitance circuit 120 is grounded, the voltage at the top terminal of negative capacitance circuit 120 is negative. The negative voltage generated by negative capacitance circuit 120 pulls down the voltage of the bottom plate of connection capacitor 20. Since connection capacitor 20 is a conventional capacitor, the charge in it generates a voltage at V inject + and V inject - The voltage difference between inject - is pulled down by the negative capacitor 120, so V inject + is pulled low, helping to reduce the noise voltage V on line 10 inject +, thus reducing EMI.

[0015] Figure 3 A high level block diagram of the active filter of the parent application is shown. A connecting capacitor 20 is connected to node N0 of line 10. Changes in the current flowing through line 10 are sensed by the connecting capacitor 20 and applied to the input of a gain control circuit 112. The output of the gain control circuit is in phase with the input and is applied to a filter capacitor 56 which connects the gain control circuit 112 to the input of a power amplifier 114. The power amplifier 114 provides a relatively high current to drive the switching capacitor 30.

[0016] Gain control circuit 112 acts as a resistive voltage divider circuit. Capacitors 20, 30 act as a capacitive voltage divider. The voltage at node N1 between capacitors 20, 30 is driven below ground by the higher current in line 10, just like Figure 2As done by the negative capacitance circuit 120. The voltage at node N1 that goes below ground is coupled to the connection capacitor 20, injecting a compensating voltage into line 10 to compensate for the higher current in line 10.

[0017] Although the active filter of the parent application is useful and effective, the gain control circuit 112 may require an operational amplifier integrated circuit (IC). The performance of the active filter may be limited by the operational amplifier IC. Additionally, the power amplifier 114 may require a large current. Therefore, the active filter requires a different circuit to increase design flexibility and potential cost.

[0018] There is a desire for an active EMI filter without inductors. There is a desire for an active EMI filter with a single-point connection to the power line. There is a desire for an active EMI filter with a single-point connection through a capacitor. It is desired that the single-point connection capacitor can sense current variations and inject compensation into the power line. It is desired that an active filter using the negative capacitance concept can sense and drive the feedback single-point capacitor. There is also a desire for an active filter with variable elements to perform gain compensation for temperature and frequency bands.

BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIG. 1 shows a prior art active EMI filter with separate sensors and injection points.

[0020] Figure 2 Highlight an active filter with a single connection point.

[0021] Figure 3 Show a high-level block diagram of the active filter of the parent application.

[0022] Figure 4 Show a high-level block diagram of the improved active filter.

[0023] Figure 5 Show an improved circuit schematic of a single-point active filter with negative capacitance.

[0024] Figures 6A-6B Is an operating curve graph of the active filter.

[0025] Figure 7 Show a graph of reducing insertion loss using the active filter.

[0026] Figure 8 Show another active filter with multi-stage transistors.

[0027] Figure 9 Show yet another active filter with multi-stage transistors.

[0028] Figure 10Disclose an active filter with a simplified MOS circuit.

[0029] Figure 11 Disclose that the active filter is used as a differential and common-mode filter.

[0030] Figure 12 Disclose that the active filter is used as a differential and single-line filter.

Detailed implementation manners

[0031] The present invention relates to an improvement of an active filter. The following description is for enabling a person of ordinary skill in the art to make and use the present invention in the context of a specific application and its requirements. Various modifications to the preferred embodiments will be obvious to those skilled in the art, and the general principles defined herein can be applied to other embodiments. Therefore, the present invention is not intended to be limited to the specific embodiments shown and described, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0032] Figure 4 Disclose a high-level block diagram of the improved active filter. The power amplifier 114 in the active filter of the Figure 3 parent application case is deleted. Figure 3 The switching capacitor 30 and the filtering capacitor 156 of the

[0033] are also replaced by a single capacitor, i.e., the variable capacitor 40. The gain control circuit 121 uses a different circuit from the gain control circuit 112.

[0034] The coupling capacitor 20 is connected to the node N0 of the line 10. The change in the current flowing through the line 10 is sensed by the coupling capacitor 20 as an induced current flowing through the coupling capacitor 20, and this induced current is applied to the input end of the gain control circuit 121. The gain control circuit 121 is controlled by this induced current.

[0035] The gain control circuit 121 drives an internal current to the variable capacitor 40 according to the induced current. The variable capacitor 40 converts the internal current into an internal voltage. The gain control circuit 121 responds to the internal voltage on the variable capacitor 40 and injects a compensation voltage from the gain control circuit 121 to the line 10 through the coupling capacitor 20.

[0036] The gain control circuit 121 functions as a current-controlled voltage source and provides a transimpedance function. The gain control circuit 121 uses the variable capacitor 40 to convert an internal current into an internal voltage. Thus, the variable capacitor 40 serves as the conversion unit of the gain control circuit 121.

[0037] The current flowing through the connecting capacitor 20 and the variable capacitor 40 is in phase. Since the current and voltage differ by 90 degrees from each other when the variable capacitor 40 is used to convert the current into a voltage, the gain control unit 121 can reverse the polarity of the voltage drop across the variable capacitor 40 by 180 degrees. This reversed voltage is injected between the node N1 and the ground. Therefore, the active filter circuit is used to compensate for the instantaneous current variations or noise on the line 10.

[0038] The bias circuit 122 provides operating conditions by generating a bias voltage or a bias current for the internal nodes within the gain control circuit 121. The bias circuit 122 generates a bias voltage or current from a power supply and a ground separate from the line 10. The resistance of the variable capacitor 40 can be adjusted to adjust the performance of the active filter. The impedance of the variable capacitor 40 can be adjusted to adjust the performance of the active filter.

[0039] Figure 5 is an improved circuit schematic diagram of a single-point active filter with negative capacitance. The bias circuit 122 and the gain control circuit 121 are shown together as a combined circuit in Figure 5 The bias of the emitter node NE1 of the first PNP transistor 24 is provided by the resistor 51 grounded, while the bias of the base node NB1 of the first PNP transistor 24 is provided by the resistor 52 grounded and the resistor 55 connected between the base node NB1 and the collector node NC1 of the first PNP transistor 24. Similarly, the bias of the emitter node NE2 of the second PNP transistor 26 is provided by the resistor 54 grounded, while the bias of the base node NB2 of the second PNP transistor is provided by the resistor 53 grounded and the resistor 56 connected between the base node NB2 and the collector node NC2 of the second PNP transistor 26.

[0040] The resistor 57 connected to the power supply supplies current to the collector node NC1 of the first PNP transistor 24, and the resistor 58 connected to the power supply supplies current to the collector node NC2 of the second PNP transistor 26. These bias resistors establish a negative current from the ground to the emitter and the base, turning on the PNP transistors 24, 26 to conduct a larger collector current. The node NE2 serves as an AC ground for small-signal analysis, while the left terminals of the resistors 51, 52, 53, 54 are connected to a hard DC ground, such as a chassis ground or a ground pin.

[0041] The bases and collectors of the PNP transistors 24, 26 are cross-coupled, causing the polarity of the voltage drop across the variable capacitor 40 to reverse by 180 degrees. This reverse voltage is injected between the node N1 and the ground, so the active filter circuit is used to compensate for the instantaneous current variations or noise on the line 10. The collector node NC1 of the first PNP transistor 24 drives the base NB2 of the second PNP transistor 26, while the collector node NC2 of the second PNP transistor 26 drives the base NB1 of the first PNP transistor 24. The collector nodes NC1, NC2 are applied to the two terminals of the variable capacitor 40.

[0042] The induced current from the line 10 can be considered as a small-signal AC signal. For small-signal AC analysis, the resistors 51 - 58 can be ignored. The AC induced current from the line 10 flows through the coupling capacitor 20, PNP transistor 24, from NE1 to NC1, then through the variable capacitor 40 to the node NC2, and then through the PNP transistor 26, from NC2 to NE2, where NE2 is a virtual ground.

[0043] The coupling capacitor 20 drives the induced current to the first emitter node NE1. Since the base current is small compared to the collector current, most of the induced current flows to the collector node NC1. Similarly, the base current of the PNP transistor 26 is also small, so most of the collector current flows to the emitter NE2.

[0044] The base-emitter voltage Vbe is fixed by the PN emitter junction. So VNE1 = VNB1 + Vbe, and due to cross-coupling, VNC2 = VNB1, thus VNE1 = VNC2 + Vbe. Similarly, for the PNP transistor 26, VNE2 = VNC1 + Vbe.

[0045] The voltage drop across the variable capacitor 40 is:

[0046] VNC1 - VNC2 = I SENSE (1 / sC40),

[0047] where I SENSE is the induced current through the coupling capacitor 20, and sC40 is the equivalent capacitance of the variable capacitor 40 in the s-domain.

[0048] The total voltage drop across the gain control circuit 121 is VNE1 - VNE2. Substituting VNE1 = VNB1 + Vbe and VNE2 = VNB2 + Vbe:

[0049] VNE1 - VNE2 = (VNB1 + Vbe) – (VNB2 + Vbe) = VNB1 – VNB2

[0050] Due to cross-coupling, VNB1 = VNC2 and VNB2 = VNC1, so

[0051] VNE1 - VNE2 = VNB1 – VNB2 = VNC2 - VNC1

[0052] = -(VNC1 - VNC2) = -I SENSE (1 / sC40)

[0053] This is for the negative capacitance circuit. Therefore, the gain control circuit 121 acts as a negative capacitance circuit ( Figure 2 ).

[0054] Since the connection capacitor 20 and the gain control circuit 121 are connected in series, the equivalent capacitance of the series connection is:

[0055] C20 / / C40 = C20*C40 / (C20 + C40)

[0056] By adjusting the capacitance of the variable capacitor 40, the active filter circuit can be adjusted.

[0057] In addition, since VNE1 - VNE2 = I SENSE (1 / sC40), I SENSE is induced from line 10, so any change in the total capacitance of the variable capacitor 40 will not generate an additional change on line 10 and node N0. Therefore, the gain control circuit 121 acts as a current-controlled voltage source.

[0058] The capacitance of the variable capacitor 40 is variable, allowing adjustment or scaling of the injected voltage. The resistance values of the resistors 51 - 58 can be selected or adjusted to set the gain and other characteristics of the gain control circuit 121. The variable resistors can also be replaced. These variable resistors and capacitors can be adjusted to compensate for temperature or different operating conditions, such as when selecting different frequency bands or switching frequencies.

[0059] Figures 6A-6B is the operating curve graph of the active filter. At Figure 6A , the current flowing along line 10 oscillates, for example, at the switching frequency of the power transistor in a switched-mode power supply. This current is induced through the connection capacitor 20 and then passes through the gain control circuit 121 to the variable capacitor 40 and the AC ground. Figure 6A depicts the induced current I flowing through the connection capacitor 20 and into the gain control circuit 121 SENSE .

[0060] At Figure 6B , the voltage of line 10 at node N0, i.e., the noise voltage level, lags behind the line current and the induced current I flowing through the connection capacitor 20SENSE 90 degrees. The network is capacitive.

[0061] VNC1 - VNC2 is the voltage across the variable capacitor 40. The variable capacitor 40 is a conversion unit that transmits or converts the induced current I SENSE into voltage information (VNC1 - VNC2).

[0062] VNE1 is the voltage at node NE1, between the connection of capacitor 20 and the emitter of the first PNP transistor 24. VNE1 is 180 degrees out of phase with the line voltage VN0.

[0063] During operation, VNC1 is equal to VNE2 - Vbe, where VNE2 is 0 (AC GND, AC ground), so VNC1 is also close to 0. When the current I SENSE flows through the variable capacitor 40, the voltage drop VNC1 - VNC2 across the variable capacitor 40 is positive, which means VNC2 is below ground. At the same time, VNC2 is equal to VNE1 - Vbe, which means VNE1 is below VNE2 (AC GND, AC ground).

[0064] After the induced current I SENSE the voltage VNE1 is pulled below the AC ground by the transistor - based cross - connection circuit and the variable capacitor 40. Therefore, VN0 is also pulled down. By adjusting the capacitance value of the variable capacitor 40, the amount of noise reduction can be adjusted to compensate for current variations in line 10, as well as the amount of EMI reduction.

[0065] Figure 7 A graph showing the reduction of insertion loss using an active filter is presented. Curve 402 shows the insertion loss of a passive LC filter, where the capacitance value of C is the same as that of capacitor 20. Curve 400 shows the insertion loss of an active LC filter that has an active filter as shown Figure 5 and a theoretical series inductor L. Compared with a passive filter of the same capacitance size, the active filter used Figure 5 has an insertion loss reduction of approximately 10 dB at 2 MHz. This is only about 2 dB worse than the active filter of the parent application, but it is more likely to be a potential low - cost circuit.

[0066] The active filter can also reduce EMI by up to 75% at 2 MHz. The smaller capacitors in the active filter can be used to achieve the same EMI reduction as the passive filter. The size and volume of the filter can be reduced.

[0067] Figure 8is another active filter with multi-stage transistors. The induced current from the coupling capacitor 20 is applied to the base of the NPN transistor 64, which is also biased by the resistors 73, 75. The change in the induced current changes the emitter-base current through the NPN transistor 64, thereby changing the collector current from the resistor 71 to the node N3, and the node N3 is connected to the bases of the NPN transistor 66 and the PNP transistor 68 through the capacitor 80. The emitter node N5 of the NPN transistor 64 is connected to the ground through the parallel resistors 84 and capacitor 82.

[0068] The base node N4 of the output stage is biased by the resistors 77, 78. The emitters of the NPN transistor 66 and the PNP transistor 68 are connected together at the node N6, which is the first terminal of the variable capacitor 40. The current driven to the variable capacitor 40 by the NPN transistor 66 and the PNP transistor 68 is converted into a compensation voltage by the variable capacitor 40, and the variable capacitor 40 uses its second terminal, the node N1, to directly inject this compensation voltage back into the line 10 through the coupling capacitor 20.

[0069] Compared with the circuit using an operational amplifier, Figure 8 the circuit has higher design flexibility. Due to this design flexibility, the circuit can have stronger current capabilities, and the power amplifier 114 ( Figure 3 ) is removed, the number of components is also reduced, so the footprint may be smaller, and it can be integrated into the IC format. Since transistors are cheaper, the cost may be reduced by approximately 70%.

[0070] Figure 9 is another active filter with multi-stage transistors. The NPN transistor 62 is added between the collector of the NPN transistor 64 and the resistor 71. The base of the NPN transistor 62 is biased by the resistors 77, 78. The base node N3 of the NPN transistor 66 is directly driven by the collector of the NPN transistor 62, and the base node N4 of the PNP transistor 68 is directly driven by the emitter of the NPN transistor 62. The capacitor 80 is connected between the base nodes N3, N4 of the NPN transistor 66 and the PNP transistor 68.

[0071] Figure 10It is an active filter with a simplified MOS circuit. A series of resistors 90 - 94 form a voltage divider between the power supply and ground, generating a first gate voltage at node N3 and a second gate voltage at node N4. The gate of the pull-up transistor 96 is driven by node N3, and the gate of the pull-down transistor 98 is driven by node N4. The drains of transistors 96 and 98 are connected together at node N6 to drive one end of the variable capacitor 40, and the other end of the variable capacitor 40 is node N1, enabling the variable capacitor 40 to directly inject a compensation voltage back to node N1 and the coupling capacitor 20.

[0072] The induced current from the coupling capacitor 20 is used to select the resistance of the intermediate resistor 92, which can be a variable resistor like resistors 90 and 94. The pull-up transistor 96 can be a p-channel transistor, and the pull-down transistor 98 can be an n-channel transistor, using standard complementary metal-oxide-semiconductor (CMOS), gallium nitride (GaN), or other transistor devices.

[0073] Figure 11 Shows that the active filter is used as a differential and common-mode filter. The switched transistor power supply (STPS) 300 generates unwanted EMI. To reduce this EMI, the active filter 70 is connected between the positive line P1 and the negative line G1. The active filter 70 can be Figure 5 the active filter, where line 10 is line P1 and ground is the negative line G1.

[0074] The active filter 70’ is connected to a common-mode node CM between the capacitor 72 (which is connected to P1) and the capacitor 74 (to which it is connected to G1). The active filter 70’ can be Figure 5 the active filter, where line 10 is the common-mode node CM between capacitors 72 and 74 and ground is the negative line G1 or another ground node.

[0075] The inductor 312 filters POS_IN to generate node P1, and the inductor 314 filters NEG_IN to generate G1. The capacitor 76 between P1 and G1 filters the ripple in the power supply.

[0076] Figure 12 Shows that the active filter is used as a differential and single-line filter. The switched transistor power supply (STPS) 300 generates unwanted EMI. To reduce this EMI, the active filter 70 is connected between the positive line P1 and the negative line G1. The active filter 70 can be Figure 5 the active filter, where line 10 is line P1 and ground is the negative line G1.

[0077] The active filter 70’ is connected to the power line P1. The active filter 70’ can be Figure 5For the active filter, line 10 is the power line P1, and the ground is an independent grounding node, such as the ground in an AC system or COM in a DC-DC system.

[0078] The active filter 70” is connected to the negative power line G1. The active filter 70” can be Figure 4 an active filter, where line 10 is the negative power line G1 and the ground is an independent grounding node.

[0079]

Alternative Embodiments

[0080] The inventors also envision several other embodiments. For example, various circuits and configurations can be used. Transistors can be cross-coupled in some circuits but not in others. The feedback can be modified. Both ends of the variable capacitor 40 can be connected to the output of the gain control circuit 121 ( Figure 5 ), or only one end can be connected to the output of the gain control circuit 121 ( Figures 8-10 ), while the other end is directly connected to the node N1 and the coupling capacitor 20, such that the variable capacitor 40 directly provides an injection voltage to the coupling capacitor 20 instead of through feedback from the gain control circuit 121.

[0081] A single capacitor can be implemented as multiple parallel capacitors, and a variable capacitor can be implemented by a switched capacitor array, such as a binary weighted capacitor array and a decoder. The gain control circuit 121 can be replaced by other active circuits with an input impedance in the format of Z = -V / I, such as a power transistor circuit.

[0082] The variable capacitor 40 can be an electronically controlled capacitance, such as a voltage-tuned capacitor or a digitally tuned capacitor (but not limited to this). In the case where the equivalent impedance of the variable capacitor 40 changes due to different temperatures or frequencies, an active filter with a variable component is required to perform gain compensation under different temperature conditions and frequency bands.

[0083] Although reducing EMI on the power line has been described, EMI in downstream circuits powered by the power line can also be reduced. Active filters can be used for various purposes other than reducing EMI. Active filters can be applied to internal nodes on an integrated circuit (IC) to reduce internal crosstalk. Active filters can be connected to internal clock lines to reduce the EMI generated by those clock lines. Active filters can be connected to differential data lines in a telecommunications system to filter the noise on these lines. Application scenarios that require filters to reduce small signals can benefit from the present invention.

[0084] A controller or initializer or calibrator can be added to select the value of the variable capacitor 40 or the value of any of the resistors 51 - 58 (if variable resistors are used). The temperature compensator can use analog components such as resistors and negative temperature coefficient (NTC) devices such as thermistors to compensate for the bandwidth of the gain control unit. The controller can be digital, receiving temperature measurements and looking up values in a table to apply to the variable capacitor or variable resistor. The controller can also receive a mode bit indicating the operating frequency and adjust the variable capacitance or variable resistance according to the settings in the lookup table. Some of the variable elements can be fixed while others are variable. The coupling capacitor 20 can be a variable capacitor instead of a fixed capacitor.

[0085] The negative capacitance circuit 120 can be a negative impedance converter, a negative load, or another circuit that injects energy back into the system rather than storing energy from the system. Although a phase shift of 90 degrees between the input induced current and the output injected voltage has been shown in Figures 6A-6B , it can be replaced with a different phase shift such as 180 degrees, and the actual phase shift depends on the circuit delay. The variable capacitor 20 can be replaced with a conversion impedance, which can be an inductor instead of a capacitor. Similarly, the coupling capacitor 20 can be replaced by an inductor. It can also be replaced with a more complex network of capacitors, inductors, and / or resistors.

[0086] Values can be scaled or otherwise operated on. Parameters such as delay and target can be adjusted or scaled according to conditions such as the device temperature or the length of time between detected faults, or the statistics or properties of the specific load being driven. For example, a large load with a large capacitance can have a longer TDIP delay setting and a larger threshold TH compared to a load with a smaller capacitance or fewer load capacitance switches. When the load uses a higher frequency clock, TDIP and TH can be increased to allow more capacitor charging.

[0087] The current can be a positive current or a negative current and can flow in either direction depending on the polarity of the charge carriers. Various operating theories have been proposed to help understand the optimal operation of the system, but these theories are only approximations of the actual circuit behavior and may be incorrect.

[0088] Additional components such as resistors, capacitors, inductors, transistors, buffers, voltage dividers, etc. can be added at various nodes, and there can also be parasitic components. Enabling and disabling the circuit can be achieved using other transistors or other means. Transmission gate transistors or transmission gates can be added for isolation. Inverters or additional buffers can be added. Certain components can use separate power supplies and ground lines. Various filters can be added. Low-active signals can be used instead of high-active signals. Various reference voltages or virtual power supplies can be used instead of hard grounding.

[0089] The background section of the present invention may include background information on the problems or environment of the present invention, rather than describing the prior art of others. Therefore, the materials included in the background art section are not an admission by the applicant of the prior art.

[0090] Any method or process described herein is machine-implemented or computer-implemented and is intended to be executed by a machine, computer, or other device, rather than solely by a human without machine assistance. The tangible results produced can include reports or other machine-generated displays on display devices such as computer monitors, projection devices, audio generation devices, and related media devices, and can include hard-copy printouts that are also machine-generated. Computer control of other machines is another tangible result.

[0091] Any advantages and benefits described may not apply to all embodiments of the present invention. When the term "means" appears in a claim element, the applicant intends that the claim element falls within the provisions of 35 USC Section 112, Paragraph 6. Typically, there is one or more word labels before the term "means". The one or more words before the term "means" are a label for the purpose of facilitating reference to the claim element, rather than expressing a structural limitation. Such means-plus-function claims are intended to cover not only the structures described herein for performing the function and their structural equivalents, but also equivalent structures. For example, although nails and screws have different configurations, they are equivalent structures because they both perform the fastening function. Claims that do not use the term "means" do not fall within the provisions of 35 USC Section 112, Paragraph 6. Signals are typically electrical signals, but can also be optical signals, for example, which can be transmitted through fiber optic lines.

[0092] The above description of the embodiments of the present invention is presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the present invention to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. The intention is that the scope of the present invention is not limited by this detailed description, but rather by the appended claims.

Claims

1. An active filter, comprising: A connection capacitor connected between a current-carrying line and a first node; A gain control unit having a first input receiving the first node, the gain control unit generating a first output based on the first node; A bias circuit for biasing the gain control unit; A conversion capacitor connected to the first output; Wherein a change in the line current on the current-carrying line induces an induced current through the connection capacitor, the induced current is applied to the gain control unit to generate the first output, wherein the conversion capacitor converts the first output into an injection voltage, and the injection voltage is injected back into the current-carrying line from the first node through the connection capacitor to compensate for a change in the line voltage on the current-carrying line; Wherein the connection capacitor transmits the induced current from the current-carrying line to the first node, and the connection capacitor also transmits the injection voltage from the first node back to the current-carrying line, So that the connection capacitor transmits a bidirectional signal to the current-carrying line; Wherein the active filter is connected to the current-carrying line only through the connection capacitor; except for a single connection point through the connection capacitor, the current-carrying line is isolated from the active filter; Wherein the gain control unit further comprises: A first transistor having a first input connected to the first node, a first output terminal connected to the first output, and a first control node controlling a first current from the first output; A second transistor having a second input connected to a fixed node, a second output terminal, and a second control node controlling a second current from the second output; Wherein the first transistor and the second transistor are cross-coupled, wherein the first control node is connected to the second output, and the second control node is connected to the first output; wherein the conversion capacitor is connected between the first output and the second output, The conversion capacitor receives the first current from the first output terminal of the first transistor at a first terminal of the conversion capacitor, and the conversion capacitor receives the second current from the second output terminal of the second transistor at a second terminal of the conversion capacitor; Wherein the conversion capacitor generates a conversion voltage based on the first current and the second current, and the conversion voltage is fed back to the first node through the first transistor to generate the injection voltage through the connection capacitor.

2. The active filter according to claim 1, wherein the first transistor is a first bipolar transistor, and the first control node is a base terminal of the first bipolar transistor; Among them, The second transistor is a second bipolar transistor, and the second control node is a base terminal of the second bipolar transistor; Wherein the gain control unit has cross-coupled bipolar transistors driving the conversion capacitor.

3. The active filter according to claim 1, wherein the conversion capacitor further comprises a variable capacitor.

4. A single-connection-point active filter, comprising: A bidirectional connection point connected to a noisy line with a noisy current; A connecting capacitor connected from the bidirectional connection point to a first node; A transimpedance circuit with its first node as an input, and the transimpedance circuit generates a first current based on the first node; A conversion impedance connected to a first output node, and the conversion impedance converts the first current generated by the transimpedance circuit into an output voltage of the first output node; Wherein, the voltage of the first node causes the conversion impedance to inject a compensation voltage back into the noisy line of the bidirectional connection point through the connecting capacitor; Wherein, the compensation voltage compensates for the change in the noisy current of the noisy line; Wherein the transimpedance circuit further includes: A first bipolar transistor, whose first base is connected to a second output node, and the first base is used to control the first current between the first output node and the first node; A second bipolar transistor, whose second base is connected to the first output node, and the second base is used to control the second current between the second output node and a fixed node.

5. The single connection point active filter according to claim 4, wherein the transimpedance circuit further includes: A first bias resistor connected between the first node and a first fixed voltage source; A second bias resistor connected between the first base and the first fixed voltage source; A third bias resistor connected between the second base and the first fixed voltage source; A fourth bias resistor connected between the fixed node and the first fixed voltage source; A fifth bias resistor connected between the first base and the first output node; A sixth bias resistor connected between the second base and the second output node; A seventh bias resistor connected between a second fixed voltage source and the first output node; and An eighth bias resistor connected between the second fixed voltage source and the second output node.

6. The single connection point active filter according to claim 4, wherein the conversion impedance includes a variable capacitor.

7. An active noise filter, comprising: A connecting capacitor connected between a connection point of a current-carrying line and a first node, wherein a change in the current carried by the current-carrying line generates an induced current, and the induced current flows through the connecting capacitor to the first node; A gain control unit connected to the first node to receive the induced current from the connecting capacitor and generate a conversion current according to the induced current; A conversion impedance receiving the conversion current generated by the gain control unit, and the conversion impedance converts the conversion current into a conversion voltage; Wherein, the conversion voltage on the conversion impedance induces an injection voltage of the connecting capacitor to compensate the current-carrying line, Thereby, the active noise filter can reduce the electromagnetic interference (EMI) generated by the current-carrying line; Wherein the gain control unit further includes: A first transistor, whose emitter receives the induced current from the first node, and the first transistor generates a first collector current at a first collector according to a first base; A second transistor having an emitter, the second transistor generating a second collector current at a second collector according to a second base; wherein the first transistor and the second transistor are cross-coupled, wherein the first collector is connected to the second base and the second collector is connected to the first base; wherein the conversion impedance includes a conversion capacitor connected between the first collector and the second collector.

8. The active noise filter according to claim 7, wherein the conversion impedance includes a capacitor connected between a conversion output node of the gain control unit and the first node.

9. The active noise filter according to claim 8, wherein the gain control unit further comprises: A voltage divider network composed of resistors, which receives the induced current as an input from the first node, and the voltage divider network generates a first gate voltage and a second gate voltage that vary with the induced current; A first transistor, whose first gate receives the first gate voltage, and the first gate controls a first current to the conversion output node; A second transistor, whose second gate receives the second gate voltage, and the second gate controls a second current to the conversion output node.

10. The active noise filter according to claim 9, wherein the first transistor includes a metal oxide semiconductor field effect transistor (MOSFET); and wherein the second transistor includes a MOSFET.

11. The active noise filter according to claim 10, wherein the first transistor includes a p-channel MOSFET, whose source is connected to a power supply; wherein the second transistor includes an n-channel MOSFET, whose source is connected to ground; wherein the drain of the first transistor is connected to the conversion output node; wherein the drain of the second transistor is connected to the conversion output node.

12. The active noise filter according to claim 8, wherein the gain control unit further includes: A first input resistor connected between the first node and the power supply; A second input resistor connected between the first node and ground; A first transistor, whose control terminal is connected to the first node, and the control terminal is used to control a first current from an output terminal of the first transistor; A first output transistor, whose control terminal responds to the first current generated by the first transistor, and the control terminal is used to control a first output current from an output terminal of the first output transistor to the conversion output node; A second output transistor, whose control terminal responds to the first current generated by the first transistor, and the control terminal is used to control a second output current from an output terminal of the second output transistor to the conversion output node.

13. The active noise filter according to claim 12, wherein the first transistor is a bipolar transistor, whose base is the control terminal, whose collector is the output terminal, and the first transistor further has an emitter, and the emitter is biased by a first bias resistor; Among them, The first output transistor is an NPN bipolar transistor, whose base is the control terminal, whose emitter is connected to the conversion output node, and the first output transistor further has a collector; Among them, the second output transistor is a PNP bipolar transistor, whose base is the control terminal, the emitter is connected to the conversion output node, and the first output transistor also has a collector; It further includes: An intrastage capacitor connected between the collector of the first transistor and the base of the first output transistor.

14. The active noise filter according to claim 13, wherein the bases of the first output transistor and the second output transistor are connected together.

15. The active noise filter according to claim 13, wherein the base of the second output transistor is connected to the collector of the first transistor.

16. The active noise filter according to claim 13, wherein the conversion impedance includes a variable resistor.

Images