EP4042453A1 - Electromagnetic induction device - Google Patents

Abstract

The invention relates to an electromagnetic induction device provided with a variable air gap equipped with heat dissipation means. In particular, the device according to the present invention comprises a core in which the variable air gap is housed. Said air gap further comprises first ferromagnetic plates intended to guide a magnetic flux that can originate in the core and can collectively have a saturation magnetic field less than that of the core. The variable air gap according to the terms of the present invention further comprises lateral protuberances which form the heat dissipation means and extend from lateral faces of the first plates.

Description

Description

Title: ELECTROMAGNETIC INDUCTION DEVICE TECHNICAL FIELD

The present invention belongs to the field of electronics and electricity. In particular, the present invention relates to a magnetic inductance with variable inductance and for which the magnetic and thermal losses are reduced with regard to the devices known from the state of the art.

The magnetic inductance according to the present invention is advantageously implemented in an AC / DC or DC / DC power converter, and in particular a DAB (“Dual Active Bridge”) converter.

STATE OF THE PRIOR ART

Magnetic inductors are devices well known to those skilled in the art, and implemented in a number of applications.

A magnetic inductor generally comprises a core, made of a ferromagnetic material, and a coil formed around a section of the core. The core can also include an air gap. This device is characterized by a characteristic quantity, called magnetizing inductance L, which depends on the ferromagnetic material, on the geometry of the core (and of its air gap), and of the winding, in particular on the number of turns forming this winding.

Certain applications, in particular electronic converters, may require a high magnetizing inductance value L for nominal operation and a low magnetizing inductance for certain operating points.

This is particularly the case for topologies of so-called resonant converters.

LLC.

Indeed, the magnetizing inductance of the transformer of an LLC converter must be raised to the nominal operating voltage, in order to limit switching losses and guarantee good efficiency, but must be able to be greatly reduced in order to ensure the continuity of the power delivered to the load, when the input voltage drops (see the case of a server power supply system that must ensure data backup in the event of a power failure such as as described in document [1] cited at the end of the description). This problem can lead to dimensioning converters of this type with low inductance values to ensure the backup function to the detriment of efficiency.

In the case of DAB converter topologies, the so-called “series” inductance value because it is placed in series with the magnetizing inductance of the transformer determines the operating range of the converter.

The transmitted power is inversely proportional to the value of series inductance L and to the phase shift between the input and output voltages set by the control.

In some cases, the series inductance function can be implemented by a component separate from the transformer. However, in other cases, the series inductance function is performed by the leakage inductance of the same transformer.

In these two cases, a single value of series inductance restricts the operating range and proves to be less flexible for controlling the DAB converter as described in document [2] cited at the end of the description. In order to alleviate these problems, it may be considered to implement a magnetic inductor with magnetizing inductance L or series variable as a function of the magnetic flux (and consequently of the current I flowing through the winding). More particularly, it may be required to have a magnetic inductor which has a high magnetizing L or series inductance at low current I and lower at high current I.

However, a magnetic inductor is generally sized to operate in a current range I flowing in the winding less than a saturation current I _sat for which the ferromagnetic core is not saturated. In this range, as long as the current I is less than the saturation current l _sat , the magnetizing inductance L remains independent of said current I. However, as soon as the current I exceeds the value of the saturation current l _sat , a magnetic saturation of the nucleus intervenes and causes a rapid decrease in its magnetic permeability and consequently of the magnetizing inductance L.

Thus, when a variability of the magnetizing inductance L is required during the operation of the magnetic inductance, different solutions can be considered. In general, these solutions propose to extend the operating range of the magnetic inductance to a nonlinear regime preceding the complete saturation of the ferromagnetic material. For this purpose, it is therefore possible to consider a lower magnetic permeability and / or a resizing of the air gap.

As soon as one and / or the other of these two solutions is implemented, saturation occurs uniformly throughout the core for a value of the current I greater than the saturation current I _sat .

However, these solutions are not satisfactory.

Indeed, saturation of the entire core, when the inductor is subjected to a current I at high frequency (typically greater than 10 kHz), can be a source of significant volume magnetic losses in the inductor, and in the entire component into which it is integrated.

Moreover, these losses can lead to heating of the core.

In addition, in the saturation regime, the magnetic flux lines, which are no longer confined in the core, are liable to disturb the components placed near the magnetic inductance. These disturbances can in particular be the source of electromagnetic incompatibilities and / or of eddy current losses.

It has therefore been possible to propose, in documents [3] to [7] cited at the end of the description, dimensioning of the core, and in particular of the air gap, making it possible to locate the saturation of the core in said air gap.

However, the proposed sizing is not satisfactory. Indeed, even though the saturation of the core remains localized in the vicinity of the air gap, the latter remains the site of magnetic losses and overheating liable to disturb the operation of the entire component.

In addition, these magnetic and heating losses limit the use of such magnetic inductors at currents I of high frequencies and in particular up to 500 kHz.

An aim of the present invention is therefore to provide an electromagnetic induction device with a variable magnetizing inductance L and for which the magnetic and heating losses are reduced with regard to the devices known from the state of the art.

Another aim of the present invention is also to provide an electromagnetic induction device which can operate at higher frequencies than the devices known from the state of the art.

DISCLOSURE OF THE INVENTION

The aims of the present invention are, at least in part, achieved by an electromagnetic induction device comprising:

- a ferromagnetic core;

- at least one air gap, called a variable air gap, and defining in the core a volume V, in which are arranged main ferromagnetic plates, essentially parallel and arranged in a direction parallel to field lines capable of circulating in the core, the plates main having a cross section configured so that all of said plates have a saturation magnetic field lower than that of the ferromagnetic core, the main plates also being provided with lateral protuberances intended to diffuse a heat capable of being produced within the main plates when the latter are traversed by a magnetic field greater than their saturation magnetic field, said lateral protuberances extending from lateral faces of said plates in a direction essentially orthogonal to said lateral faces. According to one embodiment, the lateral protuberances interconnect the main plates in pairs so as to form secondary plates perpendicular to said main plates.

According to one embodiment, the ferromagnetic core, the first plates and the protuberances are made of the same ferromagnetic material.

According to one embodiment, the main plates are all identical.

According to one embodiment, a vacant volume V _v of the volume V left vacant by the first plates and the protuberances is filled, at least in part, by a heat dissipating material which has a thermal conductivity greater than 10 W / m / K , advantageously the heat dissipation material comprises alumina.

According to one embodiment, the ferromagnetic core comprises a ferromagnetic material chosen from: a metal alloy of the FeX type where X comprises one of the elements chosen from Si, Al, Co, Ni, ferrite oxide of type A spinel structure (Fe , B) 0 ₄ with A = (Mn, Ni) B = (Co, Cu, Al, ..).

According to one embodiment, the ferromagnetic core comprises two plane ends, essentially parallel to each other, facing one another and a surface S, the two ends delimiting the variable air gap, the main plates being arranged perpendicularly. at the ends, the ferromagnetic core advantageously comprises a frame of polygonal shape, and even more advantageously of rectangular shape.

According to one embodiment, the main plates have a cross section of a transverse area S _t , the sum of the transverse areas of all the main plates being less than the area S.

According to one embodiment, the ferromagnetic core comprises two bases each provided with two main faces, called the inner face and outer face respectively, essentially parallel, the bases each face each other along their inner face, the variable air gap being formed in the one of the bases, the core further comprises a plurality of legs, substantially parallel to each other and which extend between the two internal faces, the plurality of legs comprises at least one main leg, at least one side leg and at least two trailing legs; the device further comprises at least one primary winding and at least one secondary winding, each comprising a main section, wound around the main leg, and a leakage section, called respectively primary leakage section and secondary leakage section each wound on a different leaking leg.

According to one embodiment, the variable air gap being arranged between the two leakage legs, the primary plates being in the form of fins.

According to one embodiment, the fins are oriented in a direction defined by an axis joining the two leakage legs between which the air gap is arranged.

According to one embodiment, the recess opens onto the internal face of the base considered.

According to one embodiment, the recess opens onto the external face of the base considered.

According to one embodiment, the at least one main leg comprises a single main leg, the at least two leakage legs include four leakage legs, in which the primary leakage section comprises two primary leakage sections so that the coil primary comprises in order one of the primary leakage sections, the main section, and the other primary leakage section, the primary leakage sections each being wound around a different leakage leg, and in which the leakage section secondary comprises two secondary leakage sections so that the secondary winding comprises in order one of the secondary leakage sections, the main section, and the other secondary leakage section, the secondary leakage sections each being wound around a different leakage leg.

According to one embodiment, the at least one side leg comprises four side legs, the four side legs and the four trailing legs describing a circle centered on the main leg, and in which alternate, and in a regular manner, the side legs and the trailing legs, each primary trailing section being arranged diametrically opposed to a secondary trailing section with respect to the main leg.

According to one embodiment, the at least one lateral leg comprises two lateral legs, the at least two leakage legs comprise four leakage legs, forming two groups of two leakage legs, called the first group and second group respectively, the four trailing legs and the two side legs describing a circle centered on the main leg, and in which the side legs and the groups alternate in a regular manner.

According to one embodiment, the at least one air gap comprising a first air gap and a second air gap arranged midway between the leakage legs, respectively, of the first group and of the second group.

According to one embodiment, the primary leakage sections are each formed around, respectively, one and the other of the trailing legs (104) of the first group (106), and the secondary leakage sections are each formed around , respectively, of one or the other of the leakage legs (105) of the second group (107).

According to one embodiment, a groove is formed on one and the other of the internal faces, at a distance from and around each leakage leg, the groove being interposed between the leakage leg and the main leg.

BRIEF DESCRIPTION OF THE DRAWINGS

Other characteristics and advantages will appear in the following description of an electromagnetic induction device according to the invention, given by way of nonlimiting examples, with reference to the appended drawings in which:

FIG. 1 is a schematic perspective representation of a variable air gap according to the present invention;

FIG. 2 is a schematic perspective representation of another configuration of a variable air gap according to the present invention;

FIG. 3 is a representation of a ferromagnetic core capable of being implemented within the framework of the present invention; FIG. 4 is a graphic representation of the evolution of the inductance L (vertical axis, unit "H") of the electromagnetic induction device as a function of an electric current I (horizontal axis, unit "A") flowing through coil;

FIG. 5 is a schematic representation of the electromagnetic induction device according to a first embodiment of the present invention;

FIGS. 6a, 6b are schematic representations of a half-core according to a side view (FIG. 6a) and a top view (FIG. 6b) capable of being implemented within the framework of the present invention;

FIG. 7 is a schematic representation of the electromagnetic induction device according to a first embodiment of the present invention and implementing two half-cores as illustrated in FIGS. 6a and 6b;

FIG. 8 is a schematic representation of the electromagnetic induction device according to a first variant of the second embodiment of the present invention;

FIG. 9 is a schematic representation of a half-core in section, along the internal face and according to a second variant of the second embodiment of the present invention;

FIG. 10 is a schematic perspective representation of a half-core according to a second variant of the second embodiment of the present invention;

FIG. 11 is a schematic representation of a sectional half-core provided with the coils, along the internal face and according to a second variant of the second embodiment of the present invention;

FIG. 12 is a schematic representation of an air gap opening at the level of the external face of a base in contact with a cold source.

DETAILED PRESENTATION OF PARTICULAR EMBODIMENTS

The present invention relates to an electromagnetic induction device provided with a variable air gap provided with heat dissipation means. In particular, the device according to the present invention comprises a core at which the variable air gap is housed. Said air gap also comprises first ferromagnetic plates intended to guide a magnetic flux liable to originate in the core and intended to operate in a saturation regime for a magnetic flux value lower than the value required for the saturation of the core, the flux being conservative throughout the magnetic circuit.

The variable air gap according to the terms of the present invention further comprises lateral protuberances forming the heat dissipation means and which extend from lateral faces of the first plates.

More particularly, the invention relates to an electromagnetic induction device 100 (FIGS. 1 to 3).

The electromagnetic induction device 100 can be a magnetic inductor of inductance L integrated in an AC / DC or DC / DC power transformer, and in particular a DAB converter.

The electromagnetic induction device 100 comprises a ferromagnetic core 200 (illustrated in Figures 1, 2 and 3).

The ferromagnetic core 200 may comprise at least one ferromagnetic material chosen from: a metal alloy of the FeX type where X comprises one of the elements chosen from Si, Al, Co, Ni, a ferrite oxide of spinel structure of the type A (Fe, B) 0 ₄ with A = (Mn, Ni) B = (Co, Cu, Al, ..).

The ferromagnetic core 200 is capable of being traversed by field lines induced by an electric current I flowing in at least one conductive coil 300 (or winding) formed around a section of the ferromagnetic core 200 and which extends along a main axis XX '.

By “main axis” is meant an axis of symmetry of the conductive coil.

The conductive coil 300 is in particular made of a winding of a conductive wire, for example a copper wire, around a section of the ferromagnetic core 200. The ferromagnetic core 200 also comprises an air gap (“Air Gap” according to Anglo-Saxon terminology), and more particularly a variable air gap 400 (FIGS. 1 and 2).

The air gap 400 is in particular formed by a recess or an absence of material in the ferromagnetic core 400.

The recess or the absence of material results in a rupture of continuity of the ferromagnetic material forming the ferromagnetic core 400.

The variable air gap 400 defines in the ferromagnetic core 200 a volume V corresponding to the volume of material hollowed out or absent.

Main plates 500, made of a ferromagnetic material, are arranged in the volume V defined by the variable air gap 400.

The term “plate” is understood to mean an element of generally planar shape and not very thick. In particular, a plate comprises two side faces 501, essentially parallel to each other and connected by a contour.

The main plates 500 are moreover essentially parallel to each other and arranged in a direction parallel to field lines capable of circulating in the ferromagnetic core 200.

According to the present invention, the orientation of a plate, and in particular of a main plate, is defined by the orientation of its side faces 501. In other words, said field lines are parallel to a direction of the planes. formed by the side faces of the main plates 500.

The main plates 500 also have a cross section of surface S _t adapted so that all of said main plates 500 have a saturation magnetic field B _sati , called first magnetic field B _sati , less than that of the ferromagnetic core 200, called second field magnetic B _sat 2.

The term “cross section” is understood to mean a section along a section plane perpendicular to the field lines crossing the main plates.

The amplitude of the magnetic field B passing through the ferromagnetic core 200 depends on the electric current I flowing in the coil 300. In particular, the first magnetic field B _sati and the second magnetic field B _sat 2 are reached when the electric current I flowing in the coil 300 is equal, respectively, to a first saturation current I _sati , and to a second saturation current I _sat 2.

Thus, the behavior of the electromagnetic induction device 100, and more particularly its magnetizing inductance L, will depend on the electric current I flowing in the coil 300.

In this regard, FIG. 4 is a graphical representation of the different operating regimes of an electromagnetic induction device 100, the main plates 500 of which are all identical.

By “identical main plates” is meant plates of the same shape, same dimensions and same material.

Such a device 100 has three operating modes or stages “A”, “B” and “C” associated with an electric current I flowing in the coil 300, respectively, less than the first saturation current I _sati , between the first current saturation current I _sati and the second saturation current I _sat 2, and greater than the second saturation current I _sat 2.

More particularly, the “A” regime corresponding to a linear regime for which the ferromagnetic core 200 and the main plates 500 are unsaturated. In this regime, the permeabilities of the ferromagnetic core 200 and of the main plates 500 are little or not dependent on the magnetic field circulating in the core so that the induction L is also essentially constant and equal to a first induction Li.

The "B" regime is characterized by a drop in inductance L to a second induction L2.

This drop is in particular due to saturation of the main plates 500 which, under the effect of a magnetic field greater than the first magnetic field B _sati , see their permeability decrease sharply to reach a value close to 1.

Finally, the "C" regime corresponds to a saturation regime of the ferromagnetic core 200 and of the main plates 500 caused by an electric current. circulating in the coil 300 greater than the second saturation current I _sat 2. In this regime, the inductance L drops again to a value L3.

According to the present invention, the main plates 500 are also provided with lateral ferromagnetic protrusions 600.

The term “protuberance” is understood to mean members projecting at the level of the surface on which they are placed.

The lateral protuberances 600 are intended in particular to diffuse a heat capable of being produced within the main plates 500 when the latter are traversed by a magnetic field greater than their magnetic field of saturation B _sati .

The lateral protuberances 600 extend in particular from the lateral faces 501 of said main plates 500 in a direction essentially orthogonal to said lateral faces 501 (FIGS. 1 and 2).

The lateral protuberances 600 can have a rectangular, circular, square, triangular section.

The lateral protrusions 600 can interconnect the main plates 500 two by two so as to form secondary plates 700 perpendicular to said main plates 600 (FIG. 2).

Particularly advantageously, the secondary plates 700 are dimensioned so as not to saturate when the magnetic field circulating in the core is less than the second magnetic field B _sat 2. Thus, when the main plates 500 saturate, the secondary plates 700 limit the overflow of the core. magnetic flux around the air gap area. In other words, the secondary plates 700 ensure the guidance of the magnetic flux in the air gap, and de facto limit any lateral radiation of the magnetic field.

Furthermore, the lateral protuberances can be limited to the volume V defined by the variable air gap.

Advantageously, the ferromagnetic core 200, the first plates 500 and the lateral protuberances 600 are made of the same ferromagnetic material. Still advantageously, the vacant volume V _v of the volume V left vacant by the first plates 500 and the protuberances 600 can be filled, at least in part, by a heat dissipating material 601 (FIG. 5) which has a higher thermal conductivity. at 10 W / m / K, advantageously the heat dissipating material comprises alumina.

The presence of the heat dissipation material makes it possible to assist the cooling by the lateral protuberances 600 by draining the heat produced in the air gap towards a heat sink.

According to a first embodiment of the electromagnetic induction device 100, the ferromagnetic core 200 comprises two flat ends 200a and 202b of surface S, essentially parallel to each other, facing each other.

The ends 200a and 200b define the variable air gap 400, and the main plates 500 are arranged perpendicular to the latter.

The main plates 500 have a cross section of a transverse area S _t , the sum of the transverse areas of all the main plates 500 being less than the area S.

The ferromagnetic core can comprise a frame of polygonal shape, and even more advantageously of rectangular shape.

For example, as illustrated in FIG. 5, the magnetic core 200 comprises five parallelepipedal sections 201-205 composed of ferromagnetic materials joined two by two by their ends in order to form a rectangular frame. Two sections 204 and 205 form one side of the rectangular frame and are spaced at their ends 200a and 200b by the air gap 400 (spacing g).

The sizing principle of a device according to the present invention is presented below on the basis of the core forming a side square frame / and illustrated in FIG. 5. This sizing principle is not however limited to this single configuration, and those skilled in the art can easily adapt it to other types of core geometries. In this example, the primary plates 500, identical, connect the ends 200a and 200b which have a spacing g. The fraction of the end surface 200a and 200b covered by the sections of the primary plates is noted /.

The reluctance R _e of the air gap structure provided only with the primary plates made of a ferromagnetic material of permeability p _s is then expressed as follows:

As soon as N _p secondary plates are considered, the reluctance of the air gap structure is calculated simply by applying a reluctance network method to each of the constituents of the structure. The thickness of these plates, denoted e _p , is small compared to the spacing g (e _p “g).

The secondary plates furthermore comprise a ferromagnetic material of permeability m _r and divide the air gap into several secondary air gaps placed in series. The reluctance R _es of each secondary air gap is therefore expressed as follows:

The total reluctance of the air gap structure comprising the primary and secondary plates is the sum of the N _p secondary reluctances separating the plates:

These expressions make it possible to deduce the inductance L (p _C m ₅ ) of the device illustrated in FIG. 5 and which is expressed as follows:

This expression of the inductance is identical to that obtained with a constant air gap whose spacing g is weighted by the term F (^ _s , f).

This term therefore makes it possible to modulate the air gap distance which intervenes in the value of the inductance without modifying the geometry of the magnetic circuit. For this, it is necessary to produce, as a function of the applied current, a variation in the permeability p _s of the primary plates.

The primary plates, which have a _{relatively small cross section of surface S t} compared to surface S, are traversed by a magnetic induction greater than that traversing the core according to the principle of conservation of magnetic flux. Such a consideration makes it possible to produce locally, in particular at the level of the primary plates 500, a saturation effect.

The magnetic induction in the primary plates B _st corresponds to an amplification of the magnetic induction B _c in the ferromagnetic core. This amplification is a function of the surface fraction f and is given by the following relation:

When the current I flowing through the coil increases, the magnetic induction also increases. However, as soon as the current I flowing in the coil reaches the saturation current l _sat , the magnetic induction in the primary plates reaches the saturation value B _sat , while the ferromagnetic core remains in linear regime.

The value of the saturation current is in this regard given by the following relation: For a current I greater than the saturation current l _sat , the permeability of the primary plates p _s is equal to 1. In order to determine the amplitude of the variation in inductance attainable by current control, nominal operation at low induction (l "l _sat ), excluding saturation, followed by high induction operation (l" l _sat ) where saturation occurs in the primary plates (p _s = l). The variation of inductance between these two borderline cases is given by:

Figures 6a, 6b, and 7 propose another configuration of the ferromagnetic core in connection with the first embodiment of the electromagnetic induction device 100. In this other configuration, the ferromagnetic core comprises two half-cores 200i and 200 ₂ is of the ETD type (double E with cylindrical central leg), well known to those skilled in the art.

The dimensions of the ferromagnetic half-cores are given in relation to figures 6a and 6b and repeated in the following table:

The two identical half-cores are mounted opposite each other with an air gap formed at a central column 207 (FIG. 7). The spacing g of the air gap is in this example equal to 5 mm. The air gap structure comprises 5 primary plates (21.65 mm x 5mm) 0.41mm thick and 2 secondary plates (21.65mm x 21.65mm) 1mm thick evenly spaced.

The permeability of the ferromagnetic material is 1500 and the saturation induction is 430 mT. The central column 207 is wound with a winding of 5 turns of conductive wire.

Under these conditions, the saturation current is 6 A. For a current less than 1 _sat , the core inductance is 16 mH and decreases to 3 mH after saturation of the primary plates. The heat exchange surface developed by the secondary plates improves cooling by natural air convection and limits heating to 100 ° C in the structure.

The remainder of the description relates to a second embodiment of the electromagnetic induction device 100.

The electromagnetic induction device 100 corresponding to this second mode can in particular be implemented as a component of a power converter of the “Dual Active Bridge” (DAB) type, and essentially uses the elements described above.

Figures 8 and 9-11 are in this regard schematic representations, in top view, of a half-cores 2OO ₃ and 2OO ₄ capable of being implemented according respectively to a first variant and a second variant of this second of achievement.

According to this second embodiment, the ferromagnetic core 200 comprises an assembly of the two half-cores 2OO ₃ and 2OO ₄ .

In this regard, the ferromagnetic core 200 comprises two bases 101 each provided with two main faces, called the inner face 101a and outer face 101b, respectively, essentially parallel.

The bases each face each other according to their internal face 101a, and the variable air gap 400 is in one of the bases, more particularly in its volume.

The core further comprises a plurality of legs, substantially parallel to each other and which extend between the two internal faces 101a. The plurality of legs include at least one main leg 102, at least one side leg 103, and at least two leakage legs 104 and 105.

The device further comprises at least one primary winding 301 and at least one secondary winding 302.

Each of the primary 301 and secondary 302 coils comprises a main section, wound around the main leg 102, and a trailing section, called the primary trailing section and the secondary trailing section, respectively, each wound on a different leakage leg 104 and 105.

Advantageously, the variable air gap 400 being disposed between the two leakage legs, and the primary plates are in the form of fins.

Still advantageously, the fins are oriented in a direction defined by an axis joining the two leakage legs between which the air gap is arranged.

Furthermore, the recess may open onto one and / or the other of the internal face and the external face of the base which includes the air gap.

In this regard, FIG. 12 is a schematic representation of an air gap 400 opening out at the level of the external face 101b of a base 101 in contact with a cold source.

The ferromagnetic core can include a single main leg 102 and four leak legs 104 and 105.

In this regard, the primary leakage section comprises two primary leakage sections so that the primary coil 301 comprises in order one of the primary leakage sections 301a, the main section 301b, and the other primary leakage section 301c, the primary leak sections each being wrapped around a different leak leg.

Equivalently, the secondary leakage section comprises two secondary leakage sections so that the secondary winding 302 comprises in order one of the secondary leakage sections 302a, the main section 302b, and the other secondary leakage section 302c, the secondary leakage sections each being wound around a different leakage leg. According to the first variant (Figure 8), at least one side leg 103 comprises four side legs 103.

More particularly, the four lateral legs 103 and the four trailing legs 104, 105 describe a circle centered on the main leg 102 and in which the side legs and the trailing legs alternate in a regular manner. Each primary leakage section is further arranged diametrically opposed to one of the secondary leakage sections with respect to the main leg.

The device thus described comprises a transformer function and a series inductance function.

The transformer function is performed by the main sections 301b and 302b, respectively, of the primary winding 301 and the secondary winding 302, surrounded around the main leg.

The series inductances created at the primary and secondary windings are provided by the primary leakage sections 301a and 301c and by the secondary leakage sections 302a and 302c.

Thus, a magnetic flux, called "transformer flux", created at the level of the main leg by the passage of a current in the primary winding follows a loop path successively crossing the base, the side legs, the other base, and crosses the main leg again.

Equivalently, a primary “leakage” flow follows a different contour connecting two trailing legs of the primary circuit and crossing the cylindrical base in its thickness along a line connecting the base of the two primary leakage legs. The secondary leakage flow follows a similar contour described by the two secondary leakage legs.

The implementation of the variable air gap between two primary trailing legs makes it possible to confer a variable leak inductance character on the device.

In an equivalent manner, the implementation of the variable air gap between two secondary trailing legs makes it possible to confer a variable leak inductance character on the device. According to the second variant (FIG. 9 to 11), the ferromagnetic core comprises two lateral legs 103, and four leakage legs 104 and 105.

The two leakage legs 104 and the two leakage legs 105 form two groups of two leakage legs, called the first group 106 and second group 107, respectively.

Furthermore, the four trailing legs and the two side legs describe a circle centered on the main leg, and in which the side legs and the groups alternate in a regular manner.

Advantageously, the at least one air gap 400 comprises a first air gap 401 and a second air gap 402 arranged midway between the leakage legs, respectively, of the first group 106 and of the second group 107.

In particular, each of the primary trailing sections is formed around, respectively, one of the trailing legs and the other of the trailing legs of the first group.

Equivalently, each of the secondary leakage sections is formed around, respectively, one of the leakage legs and the other of the leakage legs of the second group 107.

According to this second variant, the proximity between the leakage legs of the same group of leakage legs allows more precise control of the leaks.

A flux barrier 800 can also be formed in the bases 101 so as to limit the magnetic flux between the leakage legs and the side legs. These flow barriers 800 may in particular include a recessed area between each of the elements of the first group 106 and those of the second group 107 and the side legs.

The hollowed out zone can in particular extend from the edge and along a radius of the base considered.

Finally, whatever the variant considered, a bleeding can be formed on one and the other of the internal faces, at a distance from and around each leakage leg, and which is interposed between the leakage leg and the main leg. The process for manufacturing the core according to the present invention may make use of an injection molding technique (“PIM” or “Powder Injection Molding” according to Anglo-Saxon terminology). This technique is particularly well suited for the production of large series parts of complex geometry. Injection molding initially implements a step of forming a masterbatch (“feedstock” according to Anglo-Saxon terminology).

The masterbatch comprises in particular a mixture of organic material (or polymeric binder) and inorganic powders (metallic or ceramic) intended to form the final part.

The masterbatch is injected into an injection molding machine, the technology of which is known to those skilled in the art. The injection press makes it possible to melt the polymers injected with the powder in a cavity, and to give said powder the desired shape. The masterbatch, thus shaped and melted, is subjected to cooling so as to solidify it and fix it in a shape imposed by the injection molding machine.

The part formed by the masterbatch is then unmolded and unbound in order to eliminate the organic matter. The part can then be consolidated by sintering.

REFERENCES

[1] Jeong et al., "Analysis on Half-Bridge LLC Resonant Converter by Using Varaible Inductance for High Efficiency and Power Density Server Power Supply", 2017 IEEE Applied Power Electronics Conference and Exposition, March 26-30, 2017, [2] Saeed et al., "Extended Operational Range of Dual-Active-Bridge

Converters by using Variable Magnetic Devices ", 2019 IEEE Applied Power Electronics Conference and Exhibition, March 17-21, 2019,

[3] US3603864,

[4] US5440225, [5] US4728918,

[6] US2015 / 0109086,

[7] US2010 / 0085138.

Claims

Claims 1. Electromagnetic induction device (100) comprising: -a ferromagnetic core (200); - at least one air gap, called a variable air gap (400), and defining in the core a volume V, in which are arranged the main ferromagnetic plates (500), essentially parallel and arranged in a direction parallel to the field lines capable of circulating in the core, the main plates (500) having a cross section configured so that all of said plates have a saturation magnetic field lower than that of the ferromagnetic core (200), the main plates (500) also being provided with lateral protuberances (600) intended to diffuse a heat capable of being produced within the main plates (500) when the latter are traversed by a magnetic field greater than their saturation magnetic field, said lateral protuberances (600) extending from side faces (501) of said plates in a direction essentially orthogonal to said side faces (501). 2. Device according to claim 1, wherein the lateral protrusions (600) interconnect the main plates (500) in pairs so as to form secondary plates (700) perpendicular to said main plates (500). 3. Device according to claim 1 or 2, wherein the ferromagnetic core (200), the main plates (500) and the side protrusions are made of the same ferromagnetic material. 4. Device according to one of claims 1 to 3, wherein the main plates (500) are all identical. 5. Device according to one of claims 1 to 4, wherein a vacant volume V _v of the volume V left vacant by the main plates (500) and the protuberances sides is filled, at least in part, with a heat dissipation material which has a thermal conductivity greater than 10 W / m / K, advantageously the heat dissipation material comprises alumina. 6. Device according to one of claims 1 to 5, wherein the ferromagnetic core (200) comprises a ferromagnetic material chosen from: a metal alloy of the FeX type where X comprises one of the elements chosen from Si, Al, Co, Ni, ferrite oxide of spinel structure of the type A (Fe, B) 2O4 with A = (Mn, Ni) B = (Co, Cu, Al, ..). 7. Device according to one of claims 1 to 6, wherein the ferromagnetic core (200) comprises two ends (200a, 200b) planar, substantially parallel to each other, facing each other and a surface. S, the two ends define the variable air gap (400), the main plates (500) being arranged perpendicular to the ends, advantageously the ferromagnetic core (200) comprises a frame of polygonal shape, and even more advantageously of rectangular shape. 8. Device according to claim 7, wherein the main plates (500) have a cross section of a transverse area S _t , the sum of the transverse areas of all the main plates (500) being less than the area S. 9. Device according to one of claims 1 to 6, wherein the ferromagnetic core (200) comprises two bases (101) each provided with two main faces, said respectively inner face (101a) and outer face (101b), essentially parallel. , the bases (101) each face each other along their internal face (101a), the variable air gap (400) being formed in one of the bases, the core (200) further comprises a plurality of legs, essentially parallel between them and which extend between the two internal faces, the plurality of legs comprises at least one main leg (102), at least one side leg (103) and at least two trailing legs (104, 105); the device further comprises at least one primary coil (301) and at least one secondary coil (302), each comprising a main section (301b, 302b), wound around the main leg (102), and a trailing section, said respectively primary leakage section (301a, 301c) and secondary leakage section (302a, 302c) each wound on a different leakage leg. 10. Device according to claim 9, wherein the variable air gap (400) being disposed between the two leakage legs, the primary plates being in the form of fins. 11. Device according to claim 10, wherein the fins are oriented in a direction defined by an axis joining the two leakage legs between which the air gap is disposed. 12. Device according to one of claims 9 to 11, wherein the recess opens onto the internal face (101a) of the base considered. 13. Device according to one of claims 9 to 12, wherein the recess opens onto the outer face (101b) of the base considered. 14. Device according to one of claims 9 to 13, wherein the at least one main leg (102) comprises a single main leg (102), the at least two leakage legs comprise four leakage legs, wherein the primary leakage section comprises two primary leakage sections so that the primary winding comprises in order one of the primary leakage sections (301a), the main section (301b), and the other primary leakage section (301c), the primary leakage sections each being wrapped around a different leakage leg, and wherein the secondary leakage section comprises two secondary leakage sections of so that the secondary winding comprises in order one of the secondary leakage sections (302a), the main section (302b), and the other secondary leakage section (302c), the secondary leakage sections each being wound around a different leaking leg. 15. Device according to claim 14, wherein the at least one side leg comprises four side legs, the four side legs and the four trailing legs describing a circle centered on the main leg (102), and in which alternate , and regularly, the side legs and the trailing legs, each section being disposed diametrically opposed to a secondary trailing section with respect to the main leg (102). 16. Device according to claim 14, wherein the at least one lateral leg comprises two lateral legs, the at least two leakage legs comprise four leakage legs, forming two groups of two leakage legs, respectively called the first group (106 ) and second group (107), the four trailing legs and the two side legs describing a circle centered on the main leg (102), and in which alternate, in a regular manner, the side legs and the groups (106, 107). 17. Device according to claim 16, wherein the at least one air gap comprising a first air gap (401) and a second air gap (402) arranged midway between the leakage legs, respectively, of the first group (106) and of the second group (107). 18. Device according to claim 17, wherein the primary leakage sections are each formed around, respectively, one and the other of the trailing legs (104) of the first group (106), and the secondary leakage sections. are each formed around, respectively, one or the other of the leakage legs (105) of the second group (107).