Multiphase Switching Regulators

Abstract

A switching regulator is disclosed, which includes a high voltage port, a low voltage port, n switch poles, a magnetic element and a controller. Furthermore, each switch pole is connected across the high voltage port and can be composed of a switch and a diode or two switches and two diodes. Furthermore, the magnetic element includes a ferromagnetic core with n magnetic branches, each magnetic branch including a winding. Each winding start is connected to the phase node of the corresponding switch pole, and each winding end is commonly connected to one side of the low voltage port. The n+1th magnetic branch establishes a defined common mode inductance, which is combined with the transformer action to limit the current ripple. The transformer action is used to exchange ripple power between phases, thereby greatly reducing the need for inductance.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of U.S. Provisional Application No. 62 / 824,931, filed on March 27, 2019, entitled “POLYPHASE SWITCHING REGULATOR,” the entire contents of which are incorporated herein by reference. Technical Field

[0003] One or more aspects of embodiments according to the present disclosure relate to regulators, and more particularly to multi-phase switching regulators. Background Art

[0004] Figure 1a A single-phase regulator is shown. During the on-time, a portion of the energy applied to the high voltage 102 is transferred to the low voltage port 116, and the remainder is stored in the inductor 195. During the off-time, the energy previously stored in the inductor is transferred to the low voltage port 116. In the steady state, the output voltage is equal to D times the input voltage, and the output current is equal to 1 / D of the input current.

[0005] like Figure 1b As shown, the inductor current includes a DC component plus an AC ripple component. As the product of inductance and switching frequency increases, the AC ripple current decreases. This in turn means that a smaller output capacitor can be used and the AC losses in the inductor 195 are reduced. Unfortunately, both the inductor size and the DC conductor losses increase with the inductance. In addition, as the switching frequency increases, the switching losses in the switch pole 106 increase proportionally. Therefore, there is a trade-off between the inductor size and DC losses and the AC losses and capacitor size. In some embodiments, the compromise solution is that the rms ripple current is somewhere between 10% and 20% of the rated DC current. As used herein, a "switch pole" is a circuit element that operates as a single-pole double-throw switch.

[0006] like Figure 2a As shown, the individual switching regulators can be connected in parallel to provide a multi-phase regulator 103, which includes a high voltage port 102, a high voltage capacitor 104, switch poles 106-1 to 106-n, inductors 195-1 to 195-n, a low voltage capacitor 114, and a low voltage port 116. Figure 2b As shown, the switching waveforms of each of the n switching poles can be staggered.

[0007] In this way, the ripple current applied to capacitors 104 and 114 is significantly reduced due to the cancellation of the AC current harmonics. This allows the size of both capacitors to be significantly reduced. However, the overall size and mass associated with inductors 195-1 to 195-n remain similar to that of a single-phase regulator. Multiphase regulators have the additional advantage of reduced currents in each switch compared to single-phase; in some cases, this helps reduce electromagnetic interference and reduces problems associated with stray inductance. Multiphase switching regulators may make sense in situations where current or power levels are relatively high and where capacitor size is an issue. Their disadvantage is increased complexity.

[0008] like Figure 2c As shown, by integrating the inductors 195-1 to 195-n into a single unit, the multiphase regulator concept can be improved and simplified. Although there are multiple versions of this concept, the version disclosed herein significantly reduces the overall magnetic component size, mass, and winding losses compared to conventional switching regulators. With this new magnetic component, both the transformer action and the inductive energy storage effect are utilized, so that the current ripple associated with each phase is smaller than the corresponding flux density swing. In addition, the design of the magnetic component can achieve efficient heat transfer for both the windings and the ferromagnetic core.

[0009] It should be noted that similar to Figure 1a The topology, Figure 2a The topology can realize buck, boost and buck-boost operation modes according to the selection of high-side semiconductor elements and low-side semiconductor elements. For example, when the high-side semiconductor element 108 is a semiconductor switch and the low-side semiconductor element 110 is a diode, the buck mode operation is enabled so that the high voltage port 102 is the input and the low voltage port 116 is the output. In the buck mode operation and the boost mode operation, the power flow can be unidirectional. When both the high-side semiconductor element and the low-side semiconductor element are semiconductor switches (with anti-parallel diodes), the buck-boost mode operation is enabled, and DC power can be transmitted bidirectionally between the high voltage port and the low voltage port.

[0010] In the following, a new concept is disclosed, in which: Figure 2a The individual inductors of the MOSFET are integrated into a single unit so that both the transformer action and energy storage are utilized, thereby enabling reductions in magnetic component size, mass and DC losses. Summary of the invention

[0011] According to an embodiment of the present invention, a circuit having a first input conductor, a second input conductor, a first output conductor and a second output conductor is provided, the circuit comprising: a first switching pole; a second switching pole; and a magnetic element, the magnetic element comprising a plurality of windings, the plurality of windings comprising a first winding connected to the first switching pole and a second winding connected to the second switching pole, the mutual inductance of the first winding and the second winding being at least 50% of the self-inductance of the first winding.

[0012] In some embodiments, the magnetic element includes a magnetic core constructed at least in part from a solid material, the magnetic core having at least two prongs, a prong of the at least two prongs extending through the first winding.

[0013] In some embodiments, the magnetic core has a gap.

[0014] In some embodiments, the magnetic path through the first winding extends through the gap.

[0015] In some embodiments, the magnetic path through the first winding does not extend through a gap in the magnetic core.

[0016] In some embodiments, the magnetic path through the first winding does not pass through any other winding of the plurality of windings.

[0017] In some embodiments, the magnetic core is comprised of a material selected from the group consisting of magnetic laminations, iron powder, ferrite, and combinations thereof.

[0018] In some embodiments, the circuit further includes an inductor connected in series with the magnetic element.

[0019] In some embodiments, the first winding is a surface-wound or edge-wound single-layer winding.

[0020] In some embodiments, the first winding is multi-layered.

[0021] In some embodiments, the power flow is unidirectional.

[0022] In some embodiments, power flow may be bidirectional.

[0023] In some implementations, a circuit includes a high voltage port and a low voltage port.

[0024] In some implementations, the circuit includes a capacitor connected across the high voltage port or a capacitor connected across the low voltage port.

[0025] In some embodiments, each switching pole includes a high-side semiconductor switch and a low-side diode.

[0026] In some embodiments, each switching pole includes a low-side semiconductor switch and a high-side diode.

[0027] In some embodiments, each switching pole includes a high-side semiconductor switch connected in parallel with a diode and a low-side semiconductor switch connected in parallel with the diode.

[0028] In some embodiments, the circuit includes multiple switching poles, including a first switching pole and a second switching pole. The circuit also includes a control circuit, which controls the switching poles using corresponding control waveforms, and the duty cycles of the control waveforms are symmetrically spaced and equal.

[0029] In some embodiments, all coils have the same number of turns.

[0030] In some embodiments: the magnetic element includes a magnetic core, the first winding is in close contact with the first heat removal surface, and at least a portion of the magnetic core is in contact with the second heat removal surface.

[0031] In some embodiments, the magnetic element further includes: a thermally conductive material between the first winding and the first heat removing surface, and a thermally conductive material between at least a portion of the magnetic core and the second heat removing surface.

[0032] In some embodiments, the coolant is forced into direct contact with at least a portion of at least one winding.

[0033] According to an embodiment of the present invention, there is provided a multi-phase inverter including a plurality of circuits.

[0034] In some embodiments, a multi-phase inverter is configured to power a motor or a utility load. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] These and other features and advantages of the present disclosure will be appreciated and understood with reference to the specification, claims and accompanying drawings, in which:

[0036] Figure 1a A power topology of a general DC to DC regulator according to an embodiment of the present disclosure is shown.

[0037] Figure 1b The embodiment according to the present disclosure is shown. Figure 1a Associated pole voltage and inductor current waveforms.

[0038] Figure 2a is an n-phase universal DC-to-DC regulator according to an embodiment of the present disclosure.

[0039] Figure 2b The embodiment according to the present disclosure is shown. Figure 2a The associated pole voltage.

[0040] Figure 2cA general n-phase DC to DC regulator using integrated magnetics according to an embodiment of the present disclosure is shown.

[0041] Figure 2d A generic n-phase DC to DC regulator using integrated magnetics plus an external inductor is shown according to an embodiment of the present disclosure.

[0042] Figure 2e According to the embodiment of the present disclosure, for each phase use Figure 2a The regulator of the three-phase inverter.

[0043] Figure 2f is a block diagram of an inverter connected to a motor.

[0044] Figure 3a According to the embodiments of the present disclosure, Figure 2a Schematic representation of a first embodiment of an n-phase magnetic element used in a regulator.

[0045] Figure 3b According to the embodiments of the present disclosure, Figure 2a Schematic representation of a second embodiment of an n-phase magnetic element used in a regulator.

[0046] Figure 3c According to the embodiments of the present disclosure, Figure 2a Schematic representation of a third embodiment of an n-phase magnetic element for use in a regulator.

[0047] Figure 4 According to the embodiments of the present disclosure Figure 3b A perspective view of a six-phase version of the magnetic components.

[0048] Figure 5 The embodiment according to the present disclosure is shown. Figure 4 The voltage and current waveforms associated with the magnetic components, where the gap g I is negligible.

[0049] Figure 6 The embodiment according to the present disclosure is shown. Figure 4 The voltage and current waveforms associated with the magnetic components, where the gap g I is limited.

[0050] Figure 7 is a perspective view of a three-phase implementation of a magnetic element including multiple layers of conductors according to an embodiment of the present disclosure.

[0051] Figure 8is a perspective view of an embodiment of three-phase liquid cooling of magnetic elements using coils connected in parallel according to an embodiment of the present disclosure.

[0052] Fig. 9 is a perspective view of an embodiment of three-phase liquid cooling of magnetic elements using helical coils according to an embodiment of the present disclosure.

[0053] Fig.10 According to the embodiments of the present disclosure, Figure 8 A perspective view of a coil used in an embodiment.

[0054] Fig.11 According to the embodiments of the present disclosure, Fig. 9 A perspective view of a coil used in an embodiment. DETAILED DESCRIPTION

[0055] The detailed description set forth below in conjunction with the accompanying drawings is intended to be a description of an exemplary embodiment of a multiphase switching regulator provided according to the present disclosure, and is not intended to represent the only form in which the present disclosure can be constructed or utilized. This description sets forth the features of the present disclosure in conjunction with the embodiments shown. However, it should be understood that the same or equivalent functions and structures can be implemented by different embodiments, which are also intended to be included within the scope of the present disclosure. As indicated elsewhere herein, similar reference numerals are intended to indicate similar elements or features.

[0056] Figure 1a and Figure 1b is the background involving single-phase general-purpose switching regulators, and Figure 2a and Figure 2b This is the background on multiphase general purpose DC to DC regulators. Figure 1a, the switch pole 106 and the high voltage capacitor 104 are connected across the high voltage port 102. The inductor 195 is connected between the phase node of the switch pole 106 and one side of the low voltage port 116. The switch pole 106 is in turn composed of a high-side semiconductor element 108 and a low-side semiconductor element 110 connected in series. In the case where the semiconductor element 108 is a semiconductor switch (e.g., an IGBT or a MOSFET) and the device 110 is a diode, buck mode operation is possible, and the input voltage applied at the port 102 is "stepped down" to a lower voltage provided as an output at the port 116. In the case where the low-side semiconductor element 110 is a semiconductor switch and the high-side semiconductor element 108 is a diode, boost mode operation is possible, where the port 116 is the input and the port 102 is the output. In the case where both the semiconductor elements 108 and 110 are semiconductor switches with anti-parallel diodes, buck-boost mode operation is possible. As used herein, the terms "high voltage" and "low voltage" describing the high voltage port and the low voltage port are used only to distinguish the two ports; in operation, the voltage at the high voltage port 102 can be greater than the voltage at the low voltage port 116. Each of these ports can have two conductors, which can be referred to as the "positive side" and the "negative side", respectively. In this usage, the modifiers "positive" and "negative" are used only to distinguish the conductors, and, for example, the voltage on the positive side can be negative relative to the voltage on the negative side. Figure 1a In FIG. 1 , the common conductor 148 connects one conductor 102 b of the high voltage port 102 to one conductor 116 b of the low voltage port 116 .

[0057] exist Figure 2a , the operation is similar in all cases to Figure 1a The difference is that, as opposed to a single phase, n individual phases are operated in parallel. Since a parallel path is established that behaves as a voltage source, steps must be taken to ensure that the DC current components remain balanced. This can be achieved by sensing the individual phase currents and controlling the corresponding duty cycle (e.g., by Figure 2a to Figure 2f The controller 193 shown in the figure is implemented so that the desired current is achieved for each phase. Balancing can also be achieved by keeping the duty cycles exactly equal (which most digital controllers can provide). Figure 2b As shown, by symmetrically spacing the duty cycles, optimal current harmonic cancellation can be achieved at both ports. Figure 2b As shown, the duty cycles can be symmetrically spaced and equal.

[0058] like Figure 2c As shown, Figure 2b The n inductors of can be replaced by a single magnetic element 112. There are various choices for the magnetic element, ranging from cases where only the packaging activity is represented to cases where the magnetic topology is changed. Figure 2cIn FIG. 1 , the common conductor 148 connects one conductor 102 b of the high voltage port 102 to one conductor 116 b of the low voltage port 116 . Figure 3a and Figure 3b In , novel magnetic structures are shown which enable the magnetic elements to have reduced size and mass due to the fact that there is a transformer role in addition to the energy storage role. Each of these structures acts as a summing transformer in combination with a series inductor. Figure 3a In the case of a MOSFET structure, leakage inductance is increased by adding core pins that provide a defined path for the magnetic flux. Figure 3a In the case of the structure, the series inductance can be completely provided by the leakage inductance. Figure 3a In a magnetic element, for each winding, there is a magnetic path through the winding that does not pass through any other winding but instead passes through one of the auxiliary core pins, such as through core pin 122-a or core pin 122-b.

[0059] A "ripple suppression inductor" effectively connected in series with the output may have the effect of reducing the ripple. This inductance may be the leakage inductance of the magnetic element 112, or it may be the inductance of a separate inductor connected in series with the output (as described below). If Figure 2c If the magnetic element described does not have sufficient leakage inductance to limit the ripple current to the desired value, an external inductor 113 can be connected between the summing node 118 of the magnetic element 112 and the node formed by the terminal 116a in parallel with the capacitor 114 (see Figure 2d ).exist Figure 2d In FIG. 1 , the common conductor 148 connects one conductor 102 b of the high voltage port 102 to one conductor 116 b of the low voltage port 116 .

[0060] Figure 2e Shows three Figure 2c The regulators are combined to form an application of a three-phase inverter. Each regulator contains n phases. (Because the number of regulator phases may be confused with the number of inverter phases, the regulator phases are designated as "sub-phases" and the term "phase" will be applied to the inverter stage.) The DC power at port 102 is exchanged with the AC power at AC port 116 in a manner similar to a conventional three-phase inverter. Figure 2e The advantages of the scheme compared with conventional inverters are that the AC port switching ripple can be almost eliminated while maintaining low mass and high energy efficiency. When a motor such as an induction, permanent magnet or reluctance machine 191 is driven (see Figure 2f), the new inverter can reduce machine losses, resulting in improved machine power rating and machine efficiency. In addition, by eliminating high frequency switching components, winding capacitance no longer affects machine efficiency. This in turn allows for system designs in which both system voltage and winding capacitance can be relatively high without the usual impact on system efficiency.

[0061] The DC bus capacitors 104 may be combined into a single unit. In case the number of sub-phases is larger (eg, 6), the AC port capacitors 114 may be deleted without significantly affecting losses and efficiency. Figure 2e The scheme can be applied to single-phase inverters (where only two regulators are used); it can also be applied to systems with more than three phases and to systems driving multiple loads. Figure 2f is a block diagram of an inverter connected to the motor 191.

[0062] Figure 3a is instead of Figure 2a Schematic representation 105 of a first embodiment of an n-phase magnetic element of n inductors. The core includes n phase pins 120-1 to 120-n, which have corresponding coils 124-1 to 124-n, each coil including m turns, and each pin has an effective gap g1 as shown. The outer pins 122-a and 122-b provide two parallel flux return paths at the ends of the core; each outer pin has a gap g2 as shown. The magnetic circuit is completed by core elements 126 and 128. The core pins 122-a and 122-b can be replaced by a single pin; two pins are selected for reasons of symmetry and minimized total core mass.

[0063] When applied to Figure 2a When the topology is , each winding has a voltage with a magnitude of V in , a pulse voltage with a duration of D*T and a frequency of 1 / T. The n phase voltages are symmetrically spaced so that positive transitions occur at times 0, T / n, 2*T / n, ..., (n-1)*T / n, where D is the duty cycle (D ranges from 0 to 1). Each winding end is connected to node 116a. In the steady state, the voltage at port 116 is equal to the voltage at port 102 multiplied by D.

[0064] Eliminating the magnetic effects associated with the core pins 122-a and 122-b, Figure 3a The structure acts as a transformer in which the sum of the magnetic flux through the pins 120-1 to 120-n is zero. This in turn means that the sum of the instantaneous voltages across the windings 124-1 to 124-n must also be zero, which in turn means that AC power is exchanged between the phases. When applied to Figure 2a When the topology Figure 3aThe structure is used as an averaging device. The amplitude of the AC ripple voltage is reduced to one-nth, while the ripple frequency is increased to n times. In this way, the inductance required to limit the current ripple is reduced to n 2 One percent.

[0065] It can be seen that Figure 3a The magnetic components of are equivalent to an "averaging transformer" connected in series with the inductor. The voltage at the transformer summing node 118 is then equal to the instantaneous average of the voltages applied to the beginning of the windings (each voltage is referenced to ground). Thus, it can be further seen that the transformer output voltage (referenced to ground) has a value equal to V in * The DC component of D and the AC peak-to-peak pulse component V with a frequency of n / T in / n. The duty cycle D′ of this AC component is given as D′=n*D-{n*D}, where {n*D} specifies the largest integer less than or equal to n*D. When D′ is zero, the switching ripple current approaches zero, and when D′ is 0.5, the switching ripple current is maximum. Therefore, for the general n-phase case, as D approaches 0 and 1, the ripple current becomes zero, and for the other n-1 cases of D=1 / n, 2 / n, …, (n-1) / n, the ripple current also approaches zero. Similarly, for the n-1 cases of D=1 / (2*n), 3 / (2*n), …, (2*n-1) / (2*n), the ripple current reaches a maximum value. At these worst-case points, the volt-second applied to the series inductor is V in *T / (4*n 2 ).

[0066] As with any transformer, the winding DC voltage component can be small enough so that core saturation does not occur, and to this end, each of the N switch pole duty cycles can be precisely matched to each other. In many cases, this can be achieved on an open loop basis if the n signals are digitally generated. In the presence of core gap g1, feedback adjustment of the duty cycle can be achieved because the flux density at the jth core pin can be observed as the weighted difference between the jth phase current and the average current of all n phases (see Figure 6 ). This information can be used to adjust the corresponding duty cycle so that the core element does not saturate. This approach solves problems such as gate driver and power device offset timing.

[0067] and Figure 3a The additional series inductance (in microhenries, L) associated with the magnetic components s ) is approximately

[0068] L s =2.52*N t 2 *A2 / (n 2 *g2) (1)

[0069] Among them, N t is the number of turns of each of the core pins 120-1 to 120-n, and A2 is the cross-sectional area (in m) of the two outer pins 122-a and 122-b. 2 The peak value of the total ripple current I p is equal to 0.5 volt-seconds divided by the inductance in Henrys. Therefore,

[0070] I p =49,600*V in *T*(g2 / A2) / n t 2 (2)

[0071] From equation (2), it should be noted that the ripple current is independent of the number of phases. This is due to the fact that the AC volt-seconds and inductance decrease equally fast as n increases. It should also be noted that the ripple current decreases as g2 / A2 decreases.

[0072] As g2 / A2 decreases, the DC component of the flux density increases for all parts of the ferromagnetic core. This sets a lower limit on g2 / A2, which is given by:

[0073] g2 / A2=2.51*(E-6)*n t *I DC / n 2 * / A1 / B jDC , (3)

[0074] Among them, I DC is the total DC current (sum of n phases), and B jDC is the DC component of the flux density within any core pin (all n have the same value). In some embodiments, B jDC is chosen to be approximately one quarter of the core saturation flux density.

[0075] Equation 3 shows that as DC current increases, g2 can increase to prevent magnetic saturation. In turn, the steady-state DC current is limited by the heat transfer and temperature ratings of the materials used in the magnetic components. As heat transfer improves, the maximum allowed DC current increases, which in turn means that g2 can increase. At some point, g2 becomes equal to h (the window height), and Figure 3a The structure can be simplified to Figure 3b The structure of Figure 3b The core pins 122-a and 122-b are removed from the structure.

[0076] for Figure 3a and Figure 3bThe core gap g1 can be increased to accommodate both configurations. As the gap g1 increases, the ripple current amplitude increases, while the flux saturation sensitivity to DC current imbalance decreases. In many cases, g1 should be set to zero because the DC current imbalance is small in nature. The core gap can be placed inside the pin (g1), such as Figure 3a and Figure 3b as shown, or they can be placed on the horizontal element of the core (g × ) or placed in two locations (see Figure 3c ). One advantage of locating the gaps within the horizontal core elements is that fringing flux entering the windings is virtually eliminated, thereby reducing winding eddy current losses. A second advantage is that the individual gaps can assume different values ​​from each other, which can provide benefits in certain circumstances. The core halves 126 and 128 can be designed to be the same shape, or they can be designed to be different shapes.

[0077] Figure 4 Shown as Figure 3b and Figure 3c The magnetic element 109 of the six-phase embodiment of the concept. Figure 4 In an embodiment, a six-pin ferromagnetic core 132 is joined to six single-layer windings 134. An electrically insulating spacer 136 is coaxially placed between each core pin and the corresponding winding so that electrical breakdown does not occur between the two. The winding-core assembly is contained by an encapsulating housing (or "encapsulating cover") 138. A first portion of an inner surface 140 of the encapsulating housing 138 fits closely with a portion of the outer surface of the winding 134 so that a small gap is formed between the two when assembled. Similarly, a second inner surface 142 of the housing 138 fits closely with a portion of the outer surface of the ferromagnetic core 132 so that a small gap is formed between the two when assembled. Both the winding-housing gap and the core-housing gap can be filled with a thermally conductive encapsulating material so that effective heat transfer is achieved for both the winding and the ferromagnetic core. In most cases, the heat transfer is high enough to allow winding current densities in excess of 2000 A / cm 2 It should be noted that two or more phases may be used; the ripple current decreases as the number of phases increases. In the case of a low number of phases and at low current densities, the leakage inductance may not be sufficient to limit the current ripple to the desired value. In such cases, it may be necessary to add an external inductor in series with the magnetics to adequately limit the ripple current (see Figure 2d ).

[0078] The windings may be formed of copper or aluminum conductor material of rectangular cross-section, although round conductors may be used in some cases. The windings may be edge wound or surface wound, depending on the design requirements. The winding start 144 is connected to Figure 2aThe winding end 146 is joined together to form a node 116a of the port 116 (see Figure 2a ).

[0079] Figure 5 Shows Figure 4 The three-phase version of the components is used for Figure 2a Electrical waveforms for the case of a topology where the gap g1 is negligible and where the duty cycle is 0.50. (All voltage waveforms are referenced to ground.) Due to the transformer action, all three phase currents are balanced and therefore equal. The ripple frequency is 3f (three times the switching pole frequency), and the ripple amplitude is proportional to the amplitude of the average voltage (the average of phases 1, 2, and 3) divided by the product of the inductance and the frequency. The inductance is in turn proportional to the number of turns per phase multiplied by the area associated with the leakage flux divided by the square of the gap associated with the leakage flux. By controlling the aspect ratio, the inductance can be controlled as desired. (In most cases, the inductance should be such that the rms ripple current is no more than 10% to 20% of the rated DC current.)

[0080] When the gap g1 is increased, such as Figure 6 As g1 increases, a 1f ripple current component is gradually added to each phase current. This added current signal can be isolated by subtracting a signal proportional to the sum of the other n-1 phases. The remaining signal can then be used as a measure of the instantaneous flux density of the jth phase. Each of the n such signals can be applied to the controller 193 (e.g. Figure 2a and Figure 2c see) so that the peak flux density within each pin remains equal, which in turn serves to prevent magnetic saturation within any part of the core.

[0081] Figure 7 is a perspective view of a three-phase embodiment of a magnetic element 111 including multiple layers of conductors. The magnetic element includes coils 134, a common core 132, and winding spacers 136. Coils 134 may in turn be formed from multiple layers of conductors 145; each coil having a coil start 144 and a coil end 146. The core 132 may be formed from a sintered powder such as iron powder or ferrite.

[0082] By using multiple conductor layers to form each coil, two goals can be met - reducing high frequency eddy current losses and improving winding ease. To maximize eddy current loss reduction, a transposition (not shown) can be included at the center point of each coil. By using multiple layers of conductor, coil manufacturing can be easier in some cases due to reduced material stretching and deformation; in some cases, this can also improve the packing factor. It can include something like Figure 4 package shell to achieve effective heat transfer.

[0083] Figure 8is a perspective view of an embodiment of providing three-phase liquid cooling of the magnetic element 113 of the winding using multiple coils connected in parallel. It should be noted that Figure 8 The structure is magnetically Figure 3b The magnetic element includes a core 132 and axially stacked coils 156a and 156b to form an active structure. The core 132 is three-pronged and can be made from two identical "E" members. As discussed above, gaps can be included between the core halves, making it possible to estimate the flux within each core leg. Alternatively, the same result can be achieved by using a core material with reduced permeability.

[0084] The spiral (eg, surface-wound) coils 156a and 156b may be connected in series, such as Fig.10 As shown, to form a coil assembly 156. The interconnection between the two coils can be achieved through an "S" bend 184. In turn, the coil assembly can be connected in series or in parallel to achieve the desired electrical parameters. By using a coil pair interconnected at ID, the coil start point 180 and the coil end point 182 are both at OD, which achieves effective electrical termination.

[0085] Adjacent coils 156a and 156b are separated by electrically insulating separators 158. Separators 158 serve to provide enhanced insulation between adjacent coils while establishing a thin axial gap, which in turn allows liquid coolant to flow across each coil surface. The axial gap is established by ribs 160, which are integral features of the spacer surface.

[0086] Feed plate 162 is used to introduce coolant into the annular gap (not shown) between core 132 and coil pair 156. End plate 164 is used to axially clamp the stack consisting of coils, spacers and feed plate. Axial compression of core 132 is provided by band 174. Electrical termination is provided by termination structure 168 including termination bar 170 and termination post 172.

[0087] Fig. 9 is a perspective view of an embodiment of three-phase liquid cooling of a magnetic element 115 with Figure 3b The magnetic components are consistent and include a magnetic core 132, a spiral coil 119 (see Fig.11 ), feed plate 162 and manifold 176. Adjacent turns of coil 119 (see Fig.11 ) are separated by spacers 181 so that a precise gap 183 is formed. Fig. 9 ) is directed via manifold 176 to flow within the annular gap between coil 119 and the magnetic core, so that the coolant then flows radially outward through gap 183 (see Fig.11 ). It should be noted that Fig. 9The implementation method is electromagnetically similar to Figure 3b The layout is consistent.

[0088] Each winding 119 includes a winding start point 121 and a winding end point 123, each of which is connected to the winding start point 121 and the winding end point 123 via a terminating element 168 (see Fig. 9 ) termination, the termination element 168 again includes an electrically interconnected termination strip 170 and a termination post 172. A band 174 holds the various components together under compression.

[0089] Some embodiments of the switching regulator described herein can provide buck, boost, and buck-boost functions associated with DC power processing. Some embodiments of the switching regulator described herein can be used as a replacement for the switching pole in a conventional inverter (e.g., a multi-phase inverter), and such a replacement can have the effect of producing a smooth output waveform (or one or more smooth output waveforms) instead of one or more square waves. Some embodiments of the switching regulator described herein can also replace the individual poles associated with a multi-phase inverter, such as those used for motor drives, solar inverters, etc.

[0090] As used herein, the word "or" is inclusive; thus, for example, "A or B" means any of (i) A, (ii) B, and (iii) A and B.

[0091] Any numerical range described herein is intended to include all sub-ranges of the same numerical precision contained in the range. For example, the range of "1.0 to 10.0" or "between 1.0 and 10.0" is intended to include all sub-ranges between the described minimum value 1.0 and the described maximum value 10.0 (and including 1.0 and 10.0), that is, having a minimum value equal to or greater than 1.0 and a maximum value equal to or less than 10.0, such as 2.4 to 7.6. Any maximum numerical limit described herein is intended to include all lower numerical limits contained therein, and any minimum numerical limit described in this specification is intended to include all higher numerical limits contained therein.

[0092] Although exemplary embodiments of multiphase switching regulators have been specifically described and illustrated herein, many modifications and variations will be apparent to those skilled in the art. Therefore, it should be understood that a multiphase switching regulator constructed according to the principles of the present disclosure may be implemented in a manner different from that specifically described herein. The present invention is also defined in the appended claims and their equivalents.

Claims

1. A circuit comprising: a magnetic element and n switching poles, each of the switching poles comprising a phase node; a high voltage port having a first node and a second node; a low voltage port having a first node and a second node; and High voltage capacitors, in: The switch pole phase node is connected to the corresponding first port of the magnetic element, The magnetic element includes n first ports and a summing port, where n is a positive integer. The instantaneous current at the summing port is equal to the sum of the currents at the first ports, the instantaneous voltage at the summing port is substantially equal to the average value of the instantaneous voltage at the first port, The high voltage capacitor is connected across the high voltage port, The first node of the high voltage port is connected to the first node of one of the n switch poles, The second node of the high voltage port is connected to the second node of one of the n switch poles and the second node of the low voltage port, and A summing port of the magnetic element is connected to a first node of the low voltage port.

2. The circuit according to claim 1, wherein The high voltage port is an input and the low voltage port is an output, and the circuit is configured to operate in a buck mode.

3. The circuit according to claim 1, wherein: The high voltage port is an output and the low voltage port is an input, and the circuit is configured to operate in a boost mode.

4. The circuit of claim 1, wherein: The high voltage port is both an input and an output. The low voltage port is both an input and an output, and The circuit is configured to operate in both a buck mode and a boost mode.

5. The circuit of claim 1 further comprising a low voltage capacitor connected across the low voltage port. 6 . The circuit of claim 1 , further comprising a controller configured to provide respective switching commands having substantially equal duty cycles to the n switching poles.

7. The circuit according to claim 6, wherein: Each of the duty cycles is selected such that magnetic saturation does not occur within the magnetic element.

8. The circuit according to claim 6, wherein: At least one phase current is sensed.

9. The circuit according to claim 6, wherein: The n corresponding switching commands are spaced symmetrically or approximately symmetrically within one switching cycle.

10. The circuit of claim 1 , excluding an external inductor connected in series with the summing port of the magnetic element.

11. The circuit of claim 1 further comprising an external inductor connected in series with the summing port of the magnetic element.

12. The circuit of claim 1, wherein: n is an integer greater than 2.

13. The circuit of claim 1, wherein: The magnetic element includes N phases, N being a positive integer greater than 2, wherein two or more windings are associated with a first phase of the N phases.

14. A circuit comprising: a magnetic element and n switching poles, each of the switching poles comprising a phase node; a controller configured to provide a switching command for at least one of the n switching poles; in: The switch pole phase node is connected to the corresponding first port of the magnetic element, The magnetic element includes n first ports and a summing port, where n is a positive integer. The instantaneous current at the summing port is equal to the sum of the currents at the first ports; The instantaneous voltage at the summing port is substantially equal to the average value of the instantaneous voltage at the first port; and The controller is configured to provide respective switching commands having substantially equal duty cycles to the n switching poles.

15. The circuit of claim 14, wherein: Each of the duty cycles is selected such that magnetic saturation does not occur within the magnetic element.

16. The circuit of claim 14, wherein: At least one phase current is sensed.

17. The circuit of claim 14, wherein: The n corresponding switching commands are spaced symmetrically or approximately symmetrically within one switching cycle.

18. The circuit of claim 14, wherein: The magnetic element comprises a magnetic core, the magnetic core comprises a plurality of magnetic core elements, the magnetic core has a gap; and The controller is configured to adjust the duty cycle based on the feedback signal to avoid saturation of each of the core elements.

19. The circuit of claim 14, wherein: The magnetic element includes a magnetic core that lacks a gap.

20. A magnetic element, comprising: n first ports and summing ports, n is a positive integer; magnetic core and conductive windings, in: The instantaneous current at the summing port is equal to the sum of the currents at the first ports; The instantaneous voltage at the summing port is substantially equal to the average value of the instantaneous voltage at the first port; The magnetic core of the magnetic element comprises: The first guard iron member, a second back iron member, and Multiple core pins, and A first one of the core pins is bridged between the first back iron member and the second back iron member. 21 . The magnetic element according to claim 20 , further comprising a thermally conductive encapsulation cover, wherein the thermally conductive encapsulation cover has a surface that is in contact with a surface of the winding or a surface of the magnetic core.

22. The magnetic element of claim 20, configured to enable direct cooling of the windings by a cooling liquid.

Images