Method for manufacturing a magnetic core comprising an element enclosed in an arm

Abstract

A method for manufacturing a magnetic core (1) comprising: - printing a first part (73) of the magnetic core including an arm (75, 76, 77) having an open chamber (79, 110, 111), - printing a second part (74) of the magnetic core including a connecting piece (112, 113, 114), - inserting an element (115, 116, 117) into the chamber, and - fixing the first and second parts so that the connecting piece encloses the element in the chamber, the arm being included in a plurality of arms of the core, the plurality of arms being arranged along the edges of a convex polyhedron, and the connecting piece being included in a plurality of connecting pieces of the core, each connecting piece being located at a vertex of the polyhedron. Figure to be published for the abstract: Figure 1

Description

Description Title of the invention: Method for manufacturing a magnetic core comprising an element enclosed in an arm FIELD OF THE INVENTION

[0001] — The invention relates to magnetic cores, i.e. magnetic parts configured to channel magnetic field lines and couple magnetically tically with another magnetic core. STATE OF THE ART

[0002] — In aeronautics, magnetic cores are used in particular in electronics. power in multiphase couplers of a motor driving a turbine shaft- bomachine.

[0003] — The magnetic cores currently used can be designed by compression uniaxial or powder injection molding (also known as " powder injection molding" or PIM). These processes require a mold and therefore limit consequently the geometries of the cores manufactured to demouldable parts. These two processes require a mold for each geometry and are adapted to production in large series.

[0004] There is also a method of machining a core from a solid block, but this This process is not very suitable for magnetic materials which are fragile and whose magnetic properties are sensitive to the surface condition obtained. This process does not offer not a great precision or a great complexity in the shape of the piece produced, which can again have an impact on the magnetic performance of the part produced. Finally, machining results in material losses which are not always re- cyclable and, when they are, these require prior treatment.

[0005] These processes limit the achievable shapes of magnetic cores which limits optimization of core footprint, optimization of integration of cores and the implementation of functions additional to the magnetic function, such as example a core cooling function.

[0006] There is therefore a need for a method of manufacturing magnetic cores allowing greater freedom of shape of the manufactured cores. Statement of the invention

[0007] = An aim of the invention is to propose a method for manufacturing cores ma- genetics allowing greater freedom of shape of the manufactured cores.

[0008] — The aim is achieved within the framework of the present invention thanks to a method of fa- fabrication of a magnetic core, the method comprising:

[0009] — - an impression of a first part of the magnetic core comprising an arm, the arm with an open chamber, - an impression of a second part of the magnetic core comprising a junction piece, - an insertion of an element into the chamber, and - a fixing of the first part and the second part so that: -- the junction piece encloses the element in the chamber, -- the arm is comprised in a plurality of arms of the magnetic core, the plurality of arms being arranged along edges of a convex polyhedron, each arm forming a right cylinder, and -- the junction piece is included in a plurality of junction pieces of the magnetic core, each junction piece being located at a vertex of the polyhedron. Such a method is advantageously and optionally supplemented by the following different characteristics taken alone or in combination: - the element is chosen from an element made of phase change material, an element made of amorphous material, an element made of nanocrystalline material, and a hollow ferrite tube comprising an inner wall and at least one fin which extends radially inside the tube from the inner wall; - the printing of the first part and the second part is carried out from a spinel ferrite powder by binder jet or by deposition of molten material; - the printing of the first part and the second part is carried out by binder jet, the method comprising before the insertion step a step of removing powder grains separated from the core; - the element is chosen from an element made of phase change material, an element made of amorphous material, and an element made of nanocrystalline material, the method comprising, before the insertion step, a step of sintering the first part and the second part, the fixing being carried out by gluing or screwing; - the element is a hollow tube, the method comprising manufacturing by stereolithography of the hollow tube, and preferably the fixing is carried out by co-sintering; - the element is a hollow tube, the fixing being carried out by co-sintering and preferably the method comprises manufacturing by stereolithography of the hollow tube; - the printing of the first part is configured so that the arm of the first part is a first arm, the first part comprising a first portion of a second arm of the plurality of arms of the magnetic core, the first portion being solid, the printing of the second part is configured so that the second part comprises a second portion of the second arm, the second portion being solid, and the fixing of the first part and the second part is configured to place the first portion and the second portion opposite each other along two faces extending perpendicularly to an axis of the cylinder corresponding to the second arm, the material located between the two faces being non-magnetic; and - the material between the two sides contains glue. The invention also relates to a magnetic core obtained by a method such as has just been presented. DESCRIPTION OF FIGURES Other characteristics and advantages of the invention will emerge from the following description, which is purely illustrative and non-limiting, and must be read in conjunction with the appended drawings in which: [Fig.1] [Fig.1] is a schematic representation of a magnetic core obtained according to an embodiment of the invention, [Fig.2] [Fig.3] Figures 2 and 3 are schematic representations of a type of arm of a magnetic core according to a radial section in [Fig.2] and an axial section in [Fig.3], [Fig.4] [Fig.5] Figures 4 and 5 are schematic representations of a type of arm of a magnetic core according to a radial section in [Fig.4] and an axial section in [Fig.5], [Fig.6] [Fig.7] Figures 6 and 7 are schematic representations of a type of arm of a magnetic core according to a radial section in [Fig.6] and an axial section in [Fig.7], [Fig.8] [Fig.9] Figures 8 and 9 are schematic representations of a type of arm of a magnetic core according to a radial section in [Fig.8] and an axial section in [Fig.9], [Fig.10] [Fig.11] Figures 10 and 11 are schematic representations of a type of arm of a magnetic core according to a radial section in [Fig.10] and an axial section in [Fig.11], [Fig.12] [Fig.13] Figures 12 and 13 are schematic representations of one type of arm of a magnetic core in a radial section in [Fig.12] and an axial section in |Fig.13], [Fig.14] [Fig.15] Figures 14 and 15 are schematic representations of one type of arm of a magnetic core according to a radial section in [Fig.14] and an axial section in [Fig.15]. [Fig.16] [Fig.17] Figures 16 and 17 are schematic representations of one type of arm of a magnetic core in a radial section in [Fig.16] and an axial section in [Fig.17]. [Fig.18] [Fig.18] is a schematic representation of a magnetic core obtained according to an embodiment of the invention, DETAILED DESCRIPTION OF THE INVENTION Magnetic core In relation to [Fig. 1], a magnetic core 1 comprises a plurality of arms referenced 31 to 42. Each arm forms a right cylinder. This means that the smallest convex volume that contains the arm defines a surface that is the surface of a right cylinder. A volume is convex when, for any pair of two points contained in the volume, the segment joining these two points is entirely contained in the volume. Each cylinder has a principal axis A on which the cylinder is centered, this principal axis giving the direction of the generators of the cylinder. The cylinder can in particular have a circular section, that is to say that a section perpendicular to the principal axis has a circular external perimeter. The cylinder may have a radial dimension or an outside diameter greater than or equal to 5 millimeters and less than or equal to 30 millimeters. A radial dimension of the cylinder or arm is a length of the cylinder or arm measured in a radial direction orthogonal to the main axis λ. Each cylinder extends an axial length along the major axis between two faces forming the axial ends of the cylinder. A right cylinder means that these two faces are orthogonal to the major axis. The arms of the core can all be the same axial length or have different axial lengths. Each arm can be made of ferrite, particularly spinel ferrite. Materials known as "spinel ferrites" are magnetic ceramics. They are magnetic oxides of the MeFe:O type, the term "Me" designating an alloy metal such as MnZn, NiZn, etc. It is possible to add copper or other metals to this alloy. This addition allows for additional specificities. These ceramics have high melting temperatures. Each arm is configured to receive a winding of a conductive wire. The arm can receive two windings of conductive wire which can be magnetically coupled thanks to the material composing the magnetic core. The arms of the core are arranged along edges, referenced 51 to 62 in [Fig.1], of a convex polyhedron, referenced 7 in [Fig.1]. This means that each arm is associated with one and only one edge of the polyhedron and conversely that each edge of the polyhedron is associated with one and only one arm of the core. An edge is aligned with the principal axis of the cylinder formed by the arm associated with the edge. The arm extends along this principal axis over an axial length entirely contained within the edge associated with the arm. A polyhedron is convex when for any pair of two points contained in the polyhedron, the segment which joins these two points is entirely contained in the polyhedron. The polyhedron has different faces, each of which is a polygon. Each of these polygons can have the same number of edges and therefore the same number of vertices. In the example in [Fig.1], polyhedron 7 is a cube with a total of 12 edges and 8 vertices. The cube has 6 faces, all of which are squares, with each square having 4 edges and 4 vertices. The polyhedron can therefore have one and only one type of polygon on each of its sides: 3-sided triangle, 4-sided quadrilateral, 5-sided pentagon, etc. The number of faces of the polyhedron is greater than or equal to 4. This number does not depend on the type of polygon that forms the faces of the polyhedron. For example, the polyhedron can have triangular faces and include 4 faces as in the case of a tetrahedron or $ faces as in the case of an octahedron. The polyhedron is not necessarily regular. In other words, the polygons forming the faces of the polyhedron are not necessarily all identical. The arms of the nucleus do not necessarily all have the same axial length. The magnetic core further comprises connecting pieces, referenced 81 to 89 in [Fig.1], located at vertices of the polyhedron. More precisely, at each vertex of the polyhedron there is one and only one connecting piece. These connecting pieces are preferably made of ferrite, and in particular spinel ferrite. Each joining piece is contiguous with ends of at least three of the arms of the magnetic core. Each of these ends is an axial end, i.e., an end of the arm in the direction of the main axis of the arm cylinder. The joining piece is contiguous with an arm via an axial end of an arm means that the joining piece is monolithic with the arm or in direct contact with the arm along this end or else glued to the arm along this end. A polyhedron has at least four faces. A vertex of a polyhedron connects at least three different edges. For a polyhedron with more than four faces, a vertex of the polyhedron can connect more than three different edges. Each connecting piece at a vertex of the polyhedron is contiguous with the ends of each of the arms associated with the edges that are connected by the vertex, that is, it is contiguous at the ends of at least three arms. A core as just explained therefore has a three-dimensional shape corresponding to a polyhedron, each arm of the core corresponding to an edge of the polyhedron. The polyhedron structure allows for the definition of as many closed magnetic circuits as there are closed polygons in the polyhedron. Each of these polygons corresponds in the core to a succession of ferrite arms and ferrite connecting pieces adjacent to each other, this succession forming a closed mesh of magnetic material. Each arm may be the seat of a magnetic coupling between two conductive wires wound around the arm. More generally, it may be possible to couple two windings of conductive wire that are around different arms, such as arms connected by the same junction piece or even arms that are spatially more distant from each other in the core. The convex polyhedron structure of the magnetic cores presented here allows a wide variety of shapes for these cores. Such a core can couple many different electrical windings and according to different coupling strengths. It is thus possible to achieve within a single magnetic core different magnetic couplings which in the prior art would require a plurality of magnetic cores. Each arm can perform a particular function, such as for example a magnetic coupling between two coils wound around the arm or even a cooling of the arm and therefore more generally of the magnetic core. The core can thus perform a plurality of functions, these functions being performed in locations spatially separated from each other so that it is possible to implement these functions independently of each other. In particular, this implementation is carried out more independently than in the case of a two-dimensional core. The core presented here ultimately makes it possible to perform more functions than in the prior art and this in a smaller spatial volume and more independently of each other. Furthermore, a magnetic core in the form of a three-dimensional assembly of arms allows for topological optimization. Such a core defines a structure comprising empty areas while exhibiting significant absorption of mechanical vibrations, particularly mechanical vibrations related to magnetic resonance. The structure fulfills its magnetic function or functions, while exhibiting both less weight thanks to the empty areas and good mechanical strength. Types of magnetic core arms The arms of the magnetic core can be of different types. According to a first type and in relation to figures 2 and 3, an arm 301 may be a solid ferrite cylinder which extends along a main axis A. In other words, the arm 301 has no hollows or cavities so that the arm defines a cylinder volume entirely filled with magnetic material. [Fig.2] is a radial section of the arm, i.e. a section in a plane orthogonal to the main axis A. [Fig.3] is an axial section of the arm, i.e. a section along a plane which contains the main axis A. A solid arm of this first type is particularly suitable for achieving strong magnetic coupling between two electrical wires wound around this arm. A magnetic core as presented above may comprise one or more arms of this first type. All the arms of the core may be of this type. According to a second type and in relation to figures 4 and 5, an arm 303 may have a closed housing 9, the housing 9 being centered on the axis λ of the cylinder 303, the axis λ passing through the housing 9. The arm comprises an internal wall which defines the closed housing 9, that is to say that the closed housing 9 is not in fluid communication with the exterior of the arm. In other words, the housing 9 is included - that is to say enclosed - in a solid ferrite volume which defines the arm 303. The internal wall may have the shape of an internal cylinder centered on an axis which coincides with the main axis λ of the cylinder 303. The internal cylinder may have a circular section. The housing 9 is centered on the axis λ and the axis λ passes through the housing 9. In other words, there is in the arm a central zone surrounding the axis λ which is not occupied by the material of the arm. The housing 9 is surrounded by material of the arm.In the radial directions extending outwards from the axis λ, the thickness of the material of the arm surrounding the housing is substantially the same in all these different radial directions. The thickness of the arm defined between the inner wall and the outer wall of the cylinder is greater than or equal to 500 μm. When the arm and the inner wall each have the shape of a cylinder centered on the main axis λ of the cylinder, they define radii whose difference is greater than or equal to 500 μm. The housing 9 contains a material 10 chosen from a phase change material and an amorphous or nanocrystalline material. When the housing 9 comprises a phase change material, the arm of this second type is particularly suitable for filtering temperature peaks. The phase change material is chosen to be solid at the nominal operating temperature of the core, but liquid beyond. The latent heat of the material requires that significant thermal energy be supplied to or removed from the material to leave this temperature plateau, which imposes a constant temperature on the system during this phase change, lower than that of the peak. Therefore, rapid variations in temperature of the core can be absorbed locally and smoothed by changes in repeated phases of the material. When the housing comprises an amorphous or nanocrystalline material, the arm of this second type is particularly suitable for increasing the magnetic permeability of the arm. This is particularly relevant if the core is used with alternating signals over certain frequency ranges. Typically over the range of a few tens of kiloHertz to a few hundred kiloHertz, ferrite cores can have a magnetic induction of less than 0.5 Tesla. The presence of the amorphous or nanocrystalline material reinforces the magnetic induction of the “arm + material in the housing” assembly. It is also possible, by adding the material in the housing, to reduce the size of the core necessary to achieve a certain magnetic induction. This additional degree of freedom makes it possible to achieve an identical magnetic function with a smaller radial dimension of the arm. Certain space and / or integration constraints can thus be more easily respected. A magnetic core as presented above may comprise one or more arms of this second type. All the arms of the core may be of this second type. According to a third type and in relation to figures 6 and 7, an arm 302 can be crossed by a passage 11 opening out of the arm at each end of the arm. In other words, the passage 11 extends inside the arm along the axis λ from one axial end to the other. The passage 11 is centered on the axis λ of the cylinder, the axis passing through the passage 11. The arm 302 has a hollow cylinder structure, defining an inner wall 13 and an outer wall 14. These walls 13 and 14 are preferably concentric and of the same shape defining a uniform ferrite thickness. The passage 11 is defined by an internal wall 13. The internal wall 13 may have a cylindrical shape centered on the axis λ, and in particular a cylindrical shape with a circular section. The thickness of the arm defined between the internal wall 13 and the external wall 14 is greater than or equal to 500 μm. When the arm and the internal wall each have the shape of a cylinder centered on the main axis λ of the cylinder, they define radii whose difference is greater than or equal to 500 μm. A hollow arm of this third type is particularly suitable for passing a cooling fluid into the opening passage so as to cool the arm and more generally the core. A magnetic core as presented above may comprise one or more arms of this third type. All the arms of the core may be of this type. According to a fourth type and in relation to figures 8 and 9, an arm 304 can comprise two internal walls 17 and 19 facing each other, the two walls 17 and 19 defining between them a passage 15 opening out of the arm at each end. of the arm. In other words, the passage 15 extends inside the arm along the axis A from one axial end to the other. In this fourth type, the arm includes a single opening passage which is defined by two different walls facing each other. A first inner wall 17 is located radially outside the second inner wall 19, such that the first inner wall 17 surrounds the second inner wall 19. The passage 15 is centered on the axis λ of the cylinder, but the axis λ does not pass through the passage 15. In other words, the arm 304 has a central portion 16 which is located radially inside the second internal wall 19. The second internal wall 19 is included in the central portion 16 and delimits the central portion 16. The central portion 16 is solid and is crossed by the axis λ. The through passage 15 extends radially around this central portion 16. The through passage 15 separates the central portion 16 from a peripheral portion 18 of the arm 304. The peripheral portion 18 extends radially outward from the first internal wall 17. The first internal wall 17 is included in the peripheral portion 18 and delimits the peripheral portion 18.To fix the central part 16 and the peripheral part 18, the arm 304 may comprise at least one spacer E which extends in a radial direction from the central part 16 to the peripheral part 18 or in other words from the second internal wall 19 to the first internal wall 17. Preferably, the arm comprises at least two spacers E, distributed at different heights and at different angles in the radial plane. The spacers E have for example a rod or beam shape, the spacers not extending over the entire axial length of the arm 304. The spacers may be distributed over the axial length of the arm 304 and / or radially around the axis of the cylinder. The opening passage 15 allows fluid circulation between the peripheral part 18 and the central part 16. In this passage 15 the spacers do not hinder the fluid circulation. [Fig.8] represents a radial section of the arm 304 according to a plane of a magnetic core in which there are two spacers. [Fig.9] represents an axial section of the arm 304 along a plane not intersecting the spacer(s), the passage 15 thus appears as a continuous ring around the central part 16 of the arm. The first inner wall 17 and the second inner wall 19 may have a cylindrical shape centered on the axis A, and in particular a cylindrical shape with a circular section. The through passage 15 then has an annular section. The radial dimension of each of the central part 16, the peripheral part 18 and the through passage 15 is greater than or equal to 500 μm. When the arm and the walls internal 17 and 19 each have the shape of a cylinder centered on the main axis A of the cylinder, the arm and the internal wall 17 define radii whose difference is greater than or equal to 500 μm and the internal walls 17 and 19 define radii whose difference is greater than or equal to 500 μm. In this preferred embodiment the peripheral part 18 is an external hollow cylinder and the central part 16 is an internal cylinder. The through passage 15 allows fluid circulation between the outer hollow cylinder 18 and the inner cylinder 16. These cylinders are secured by at least one spacer E. The spacer E or the spacers E extend radially between the outer wall 19 of the inner cylinder 16 and the inner wall 17 of the outer cylinder 18. Structurally, the outer hollow cylinder 18, the inner cylinder 16 and the spacers E can be formed from a single block by additive manufacturing.The two walls 17 and 19 face each other and define the opening passage 15. In this passage the spacers do not hinder the fluid circulation. A hollow arm of this fourth type is particularly suitable for passing a cooling fluid into the opening passage 15 so as to cool the arm and more generally the core. A magnetic core as presented above may comprise one or more arms of this fourth type. All the arms of the core may be of this type. It should be noted that the characteristics of an arm of the second type and the characteristics of an arm of the fourth type can be combined in an arm of a fifth type. According to this fifth type and in relation to FIGS. 10 and 11, an arm 306 can comprise on the one hand the two internal walls 17 and 19 facing each other, the two walls 17 and 19 defining between them the opening passage 15 and on the other hand the closed housing 9, the housing 9 being centered on the axis A of the arm 306, the axis A passing through the housing 9. In other words the closed housing 9 is located inside the central part 16. The housing 9 is surrounded by the second internal wall 19. The thickness - measured in a radial direction - of material of the arm between the housing 9 and the second internal wall 19 is greater than or equal to 500 μm.When the arm and the housing 9 each have the shape of a cylinder centered on the main axis λ of the cylinder, the arm and the housing 9 define radii whose difference is greater than or equal to 500 μm. A hollow arm of this fifth type makes it possible to produce simultaneously or separately the technical effects associated with the arms of the second type and the fourth type. A magnetic core as presented above may comprise one or more arms of this fifth type. All the arms of the core may be of this type. According to a sixth type and in relation to figures 12 and 13, an arm 305 may comprise two internal walls 25 and 27, each internal wall defining a passage opening out of the arm at each end of the arm, the axis of the cylinder not passing through neither of the two passages. In other words: the first internal wall 25 defines a first passage 21 which extends inside the arm along the axis A from one axial end to the other, the second internal wall 27 defines a second passage 23 which extends inside the arm along the axis A from one axial end to the other, and the axis of the cylinder A does not pass through the first passage 21 nor through the second passage 93 In this sixth type, the arm comprises two different opening passages, each passage being through a single wall. The arm therefore comprises two different walls which each define a opening passage. The arm 305 has a central part 26 which is located radially between the first internal wall 25 and the second internal wall 27. The central part 26 is solid, that is to say made of ferrite, and it is crossed by the axis A. The first inner wall 25 and the second inner wall 27 extend radially around the central portion 26 over angular sectors of an extent less than or equal to 180°. The first inner wall 25 and the second inner wall 27 may be symmetrical with respect to each other with respect to the main axis λ of the arm 305. For example, each of the inner walls 25 and 27 has in radial section a circular portion centered on the axis of the cylinder and a rectilinear portion which connects the two ends of the circular portion, each of the inner walls being generated by translation of this radial section by a translation along the main axis λ of the arm 305. In another example, each of the inner walls 25 and 27 has in radial section an elliptical shape, each of the inner walls being generated by translation of this radial section by a translation along the main axis λ of the arm 305. In a preferred embodiment of the sixth type of arm, the first internal wall 25 and the second internal wall 27 each have two cylindrical-shaped parts centered on the axis λ, the two parts facing each other. The first passage 21 and the second passage 23 then have a radial section which has the shape of a portion of a ring. The arm 305 has in the preferred embodiment a peripheral portion 28 which is located radially around the first internal wall 25 and the second internal wall 27. In the preferred embodiment, the through passages 21 and 23 are located between the central portion 26 and the peripheral portion 28 of the arm 305. The central portion 26 and the peripheral portion 28 are connected to each other by two partitions C which extend in a radial direction from the central portion 26 to the peripheral portion 28. The partitions C extend over the entire axial length of the arm 304. The partitions C separate the through passages 21 and 23, so that over the entire axial length of the arm 304, a fluid cannot pass from one through passage to the other. The partitions C de- then end a part of the internal walls 25 and 27. Each internal wall in the preferred mode is formed from a part of the surface of each of the two partitions C and the two cylindrical-shaped parts centered on the axis A facing each other, the two parts being for each internal wall connected by the two partitions. The radial dimension of each of the central portion 16, the peripheral portion 18 and the opening passages 21 and 23 is greater than or equal to 500 μm. When the central portion 16 and the peripheral portion 18 each have the shape of a cylinder centered on the main axis λ of the cylinder, the central portion 16 and the peripheral portion 18 define radii whose difference is greater than or equal to 1 millimeter. A hollow arm of this sixth type is particularly suitable for passing two flows of cooling fluid through the two emerging passages 21 and 23 so that each emerging passage is crossed by a flow of fluid. This makes it possible to implement cooling of the arm using two separate fluids or a fluid which passes twice through the arm. A magnetic core as presented above may comprise one or more arms of this sixth type. All the arms of the core may be of this type. It should be noted that the characteristics of an arm of the second type and the characteristics of an arm of the sixth type can be combined in an arm of a seventh type. According to this seventh type and in relation to FIGS. 14 and 15, an arm 307 can comprise on the one hand the two internal walls 25 and 27, each internal wall defining a passage opening out of the arm at each end of the arm, the axis of the cylinder not passing through either of the two passages and on the other hand the closed housing 9, the housing 9 being centered on the axis A of the arm 307, the axis A passing through the housing 9. In other words the closed housing 9 is located inside the central part 26. The thickness - measured in a radial direction - of material of the arm between the housing 9 and the opening passages 21 and 23 is greater than or equal to 500 μm. A hollow arm of this seventh type allows the technical effects associated with the arms of the second and sixth types to be produced simultaneously or separately. A magnetic core as presented above may comprise one or more arms of this seventh type. All the arms of the core may be of this type. Cooling fin Optionally for arms of the third type, the arm may further include a fin extending from the inner wall into the passageway, the fin being configured to increase a heat exchange surface area between the arm and a fluid passing through the passageway. In the case where a fin is included in one of these arms, it extends into the passage emerging from the arm or at least into at least one of the two passages emerging from the arm, the fin extending from the wall or one of the two walls which defines the passage. The fin therefore protrudes from an internal wall of the arm towards the inside of the passage. In a first embodiment of the cooling fin, and with reference to figures 4 and 5, the arm 302 of the third type comprises a fin 12 which extends from the internal wall 13 inside the passage 11. Here the fin 12 takes the shape of a rectangular parallelepiped which extends in the direction of the axis of the cylinder over the entire axial length of the arm 302 and is located radially in the same angular sector of the cylinder over this entire length. In a second embodiment of the cooling fin, and in relation to figures 16 and 17 the through passage is a primary passage and the arm 302 of the third type houses inside this first passage a hollow sleeve 70. The hollow sleeve 70 has an inner wall 71 defining a secondary passage 72 narrower than the first passage. The secondary passage 72 opens out of the arm at each end of the arm 302. The secondary passage 72 extends inside the arm along the axis λ from one axial end to the other. The secondary passage 72 is centered on the axis λ of the cylinder, the axis passing through the secondary passage. The hollow sleeve 70 has an outer wall which is adjusted to the shape of the inner wall 13 of the arm 302. The inner wall 71 of the hollow sleeve is preferably concentric with the inner wall 13 and of the same shape defining a uniform thickness.The inner wall of the hollow sleeve may have a cylindrical shape centered on the axis A, and in particular a cylindrical shape with a circular section. The thickness of the hollow sleeve defined between the inner wall 13 and the inner wall of the hollow sleeve is greater than or equal to 500 μm. Preferably, the hollow sleeve 70 is made of ferrite. In this second embodiment of the cooling fin, the hollow sleeve comprises a fin 12 which extends from the inner wall inside the secondary passage. Here the fin 12 takes the shape of a rectangular parallelepiped which extends in the direction of the axis of the cylinder over the entire axial length of the arm 302 and is located radially in the same angular sector of the cylinder over this entire length. Common to both embodiments, the fin may take different shapes, it may extend in different directions in the primary or secondary passage, for example in a helical manner around the axis of the cylinder. The fin may also extend over only a portion of the axial length of the arm, for example the fin may extend over half or a quarter of the axial length of the primary or secondary passage. The fin is configured to increase a heat exchange surface between the arm and a fluid flowing in the primary or secondary passage. In this way, the cooling of the core by the fluid is greater. When the magnetic core comprises an arm of the third type, the core can be connected to an external cooling system so that a fluid allowing cooling circulates in the primary or secondary through passage(s). The junction parts are then chosen to allow the connection of the through passage(s) to the external cooling system. The fluid can be water or any other heat transfer fluid. Cooling water can optionally be loaded with nanoparticles (graphene, carbon nanotubes) in order to increase its thermal conductivity. The cooling fluid can be set in motion by an external pump connected to the core. The fluid can be cooled by at least one external heat exchanger. The external cooling system then includes the pump and the heat exchanger. Connecting pieces The connecting pieces located at the vertices of the polyhedron allow the different arms of the core to be connected and ensure that the core is held in place in three dimensions. For this purpose, a connecting piece has a shape adapted to the number of arms with which it is in contact. A connecting piece may have one or more collars whose dimensions are adjusted to the dimensions of the arms in contact with the connecting piece. Each collar defines a housing which allows one of the arms to be received. Each arm can thus be retained on the connecting piece. A junction room can be solid, that is to say, contain no housing or be crossed by any passage. In addition to this junction role, a junction piece can also fulfill fluid connection functions when the core contains one or more arms crossed by a through passage. In this case the junction piece is not solid. To fulfill a fluid connection function in a junction part, the magnetic core comprises an arm crossed by a through passage, or first through passage, the arm being contiguous to a junction part, the junction part comprising an internal conduit in fluid communication with the through passage. This internal conduit can open to the outside of the magnetic core. In this first case, the junction piece allows a fluid to circulate between the outside and the inside of the core. If the internal conduit does not open to the outside of the magnetic core, then the internal conduit is in fluid communication with a second opening passage different from the first opening passage, the second opening passage passing through the arm or another arm adjacent to the junction piece. In this second case, the junction piece allows a fluid to circulate inside the core between two different opening passages, Air gap A core as presented above may include an air gap, that is, one of the arms of the core and a junction piece of the core adjacent to the arm comprise two faces, the two faces extending perpendicularly to the axis of the cylinder, the material between the two faces being non-magnetic. The cylinder corresponds to the arm of the core mentioned above. The air gap is the space between the two faces and which does not include magnetic material. The air gap can in particular correspond to a bonding strip or an empty or air-filled space. The air gap allows the core to deform by thermal expansion or contraction due, for example, to a temperature variation during magnetic coupling within the core. The air gap, corresponding to a space that is more elastic than the rest of the core, allows the core to deform without breaking. The air gap corresponding to an empty space can make it easier to place a coil of electric wire around the arm. The two faces which define the air gap between them can both be included in the arm. Process for manufacturing a magnetic core Generally speaking, the core can be produced as just described using non-melting additive manufacturing techniques, particularly using binder jet printing and fused deposition modeling. Binder jetting is also known as binder jetting. This method involves spraying a liquid adhesive binder onto thin layers of powdered material, in this case spinel ferrite powder. Fused Filament Fabrication (FFF) is also known as Fused Filament Fabrication (FFF). This method involves mixing the powder with a polymer binder and then printing. The polymer must then be removed in a step called thermal debinding. The core as previously described can be produced by additive manufacturing in a single block or in several elements. When several elements are manufactured, they are then assembled in an appropriate manner using any known techniques, including gluing, screwing, sintering, co-sintering. Sintering a part increases its density. Sintering reduces porosities within sintered parts. For example, sintering can increase the density of parts by up to 90% of the material's density. the component, for example here 90% of the density of the spinel ferrite ceramic that makes up the parts. The sintering operation is accompanied by a contraction or shrinkage of the sintered part, that is to say that in all three dimensions the size of the part is reduced. This shrinkage can be taken into account in a preliminary step of computer-aided design of the part. Additive manufacturing techniques can be applied to manufacture in a single piece a core comprising an arm with a through passage, that is to say an arm of the third type, the fourth type or the sixth type, if this through passage is in communication with the exterior of the core via conduits located in junction parts. For arms of the fourth type, the spacers E can be produced by additive manufacturing. Structurally, the outer hollow cylinder 18, the inner cylinder 16 and the spacers E can be formed from a single block by this type of manufacturing. The two walls 17 and 19 face each other and define the through passage 15. In this passage the spacers do not hinder the fluid circulation. For arms of the sixth type, the C partitions can be produced by additive manufacturing. Pro ed of ae ig ion of a magnetic HOYQU p ÉSEHIGHLUNE CRAMBQre In connection with [Fig. 18], a method of manufacturing a magnetic core 1 is now presented. In a first step of manufacturing the core 1, a first part 73 of the core is printed, this first part 73 comprising an arm 75, 76 or 77, the arm having an open chamber 79, 110 or 111. The arm 75, 76 or 77 corresponds to the arms as described up to now. In particular, the arm forms a right cylinder and is intended to be included in a plurality of arms of the magnetic core, the plurality of arms being arranged along edges of a convex polyhedron. The term "chamber" is understood here as a niche or recess in the arm of the core. The chamber is open, meaning that the interior of the chamber is in communication with the exterior of the core. Compared to the types of arms described above, the chamber may correspond at least in part to: a passage emerging from an arm 75 or 76 of the third type, or a closed housing of an arm 77 of the second type, of the fifth type or of the seventh type. In a second step, a second part 74 of the magnetic core is printed, this second part 74 comprising a joining piece 112, 113 or 114. The connecting piece 112, 113 or 114 corresponds to the connecting pieces as described up to now. In particular, the connecting piece is intended to be comprised in a plurality of joining pieces of the magnetic core, each joining piece being located at a vertex of the polyhedron. The printing of the first part and the second part can be carried out using different additive manufacturing techniques. Preferably, they are made from spinel ferrite powder by binder jet or by fused deposition. Materials called ferrite or "spinel ferrites" are magnetic ceramics. They are magnetic oxides of the MeFe;O, type, the term "Me" designating an alloy metal such as MnZn, NiZn, etc. It is possible to add copper or other metals to this alloy. This addition allows for additional specifications. These ceramics have high melting temperatures. The advantage of using spinel manufacturing by binder jet or by molten material deposition is that it is not possible to use manufacturing processes with fusion in the case of ferrite, in particular because of the necessary surface conditions and the nature of the materials (volatilization of zinc in ferrites). Each of these processes is an additive manufacturing process. The use of this type of process has the following advantages: - a deposit of the magnetic material, and in particular spinel ferrite, layer by layer allowing the quantity of material used to be adjusted, - the application of the process to the field of signal electronics and power electronics, - the absence of machining: we speak of a “net-shape” process (also known by the English expression “net-shape”) in which all the - or almost all the - material used is part of the final part, - the absence of material melting, which prevents changes in the chemical composition of the ferrites and cracking of the part during printing, and - the possibility of producing cores with a “non-demouldable” shape, such as for example shapes with internal cavities; a “demouldable” shape is a part shape that can be obtained by a part molding process; a “non-demouldable” shape is a part shape that cannot be obtained by a part molding process. In a third step, an element 115, 116 or 117 is inserted into the chamber. In a fourth step, the first part 73 and the second part 74 are fixed so that the joining piece 112, 113 or 114 encloses the element 115, 116 or 117 in the chamber 79, 110 or 111. The connecting piece 112, 113 or 114 is attached to the arm 75, 76 or 77 containing the chamber 79, 110 or 111. The connecting piece 112, 113 or 114 and the arm 75, 76 or 77 are attached contiguous to each other. With this attachment, the magnetic core is formed more completely so that - the arm 75, 76 or 77 is included in a plurality of arms of the magnetic core, the plurality of arms being arranged along edges of a convex polyhedron, and - the junction piece 112, 113 or 114 is included in a plurality of junction pieces of the magnetic core, each junction piece being located at a vertex of the polyhedron. Inserting an element into the chamber, then enclosing the element in the chamber by fixing the second part to the first part allows the core to be functionalized in the chamber area. Depending on the nature of the inserted element, this allows greater freedom of shape of the core obtained, allowing in particular optimization of the size of the cores, optimization of the integration of the cores and the implementation of functions additional to the magnetic function, such as for example a function of cooling the core through the through passage. When the printing of the first part and the second part is carried out from a spinel ferrite powder by binder jet or by deposition of molten material, the manufacturing allows greater freedom in the shape of the core obtained because according to the prior art the magnetic materials forming the core, such as spinel ferrites, are difficult to machine, in particular on small dimensions. According to a first embodiment of the method, the element can be chosen from an element made of phase change material, an element made of amorphous material, and an element made of nanocrystalline material. This first mode corresponds in [Fig.18] to the arm 77 having the housing 111 in which is inserted a cartridge 117 of a phase change material or an amorphous material, or a nanocrystalline material. The arm 77 in the core once manufactured then corresponds to an arm of the second type, the fifth type or the seventh type. This arm 77 has a closed housing which corresponds to the chamber 111 once it has been closed by the junction piece 114 of the second part 74. The closed housing of the arm 77 enclosing a cartridge of a phase change material or an amorphous material, or a nanocrystalline material can fulfill different functions which have been presented previously in relation to the arms of the second type. According to a second embodiment of the method, the element is a hollow ferrite tube 115, 116 comprising an internal wall and at least one fin 118, 119 which extends radially inside the tube 115, 116 from the internal wall. This second mode corresponds in [Fig. 18] to the arms 75 and 76 respectively having the housings 79 and 110 in which the hollow tubes 115 and 116 are inserted. The hollow tube 115 is traversed on its internal wall by a fin 118 according to a helical trajectory around the axis of the arm 75. The hollow tube 116 is traversed on its internal wall by a fin 119 according to a rectilinear trajectory parallel to the axis of the arm 76. The arm 75 of the first part 73 is fixed to the connecting piece 112 of the second part 74. The arm 76 of the first part 73 is fixed to the connecting piece 113 of the second part 74. The arms 75 and 76 in the core once manufactured then correspond to arms of the third type. These arms 75 and 76 are crossed by a through passage extending along the axis of the cylinder of the arm from one end to the other of the arm. The initial chambers of the arms 75 and 76 are therefore open on both sides of the cylinders formed by the arm. The element inserted into each chamber of the arms 75 and 76 is a hollow ferrite tube which corresponds to a hollow sleeve. In this way, the arms 75 and 76 correspond to the second embodiment of fin as previously presented. The core 1 shown in [Fig. 18] allows the circulation of a fluid through a channel 123 which passes through the junction pieces 112, 120, 122 and 113 as well as the arms 75, 121 and 76. The junction pieces 112 and 113 are each crossed by a conduit which makes the channel 123 communicate with the exterior of the core. The secondary passages of the arms 75 and 76 in which the fins 118 and 119 extend are part of the channel 123. The junction pieces 120 and 122 are each crossed by an internal conduit which makes the secondary passage of one of the arms 75, 76 communicate with the passage opening the arm 121. Optionally, when the printing of the first part and the second part is carried out by binder jet, the method comprises, before the insertion step, a step of removing powder grains separated from the core. In particular, since the chamber is open, the powder grains separated from the core located in the chamber can be removed from the chamber to the outside of the core. This operation is called "depowdering" and makes it possible to obtain for the first part and the second part a single-piece part without residue that could detach therefrom. In the case where the printing of the first part and the second part is carried out from a spinel ferrite powder by binder jet or by deposition of molten material and the element 117 is chosen from an element made of phase change material, an element made of amorphous material, and an element made of nanocrystalline material, the method advantageously comprises, before the insertion step, a step of sintering the first part and the second part, the fixing then being carried out by gluing or screwing. Sintering the first part and the second part increases their density. Sintering reduces porosities within the sintered parts. For example, sintering can increase the density of the parts up to 90% of the density of the material that composes it, for example, here 90% of the density of the spinel ferrite ceramic that makes up the parts. The sintering operation is accompanied by a contraction or shrinkage of the parts, that is, in three dimensions the size of the first part and the second part is reduced. This shrinkage is taken into account in a preliminary step of computer-aided design. The first part and the second part are sintered separately. Once this densification is achieved, the element is introduced into the chamber and then the two parts are fixed so as to enclose the element. For this fixation, the two parts can be glued, screwed, or both. In the case where the printing of the first part and the second part is carried out from a spinel ferrite powder by binder jet and the element 116, 117 is a hollow tube, the method preferably comprises a manufacture by stereolithography of the hollow tube, and preferably the fixing is carried out by co-sintering. Stereolithography (SLA) manufacturing of the hollow tube makes it possible to achieve better resolution in the resulting hollow tube, particularly for the fin 118, 119. The co-sintering operation consists of placing the hollow tube in the chamber, the first part and the second part against each other and then carrying out a sintering operation on all of these parts arranged together. Alternatively, in this same case, the hollow tube can be produced using a technique other than stereolithography, for example by binder jet, and the fixing can be carried out by co-sintering. As an option to the various implementations of the process presented so far, the process may further comprise a flash sintering densification step (also known as "Spark Plasma Sintering" abbreviated as SPS) following one of the sintering steps or instead of one of them. This densification step allows the ferrite material to be densified to a higher density than with a sintering step. The core density can reach up to 98% of the density of the material that composes it, for example 98% of the density of the spinel ferrite ceramic that composes the material. Flash sintering also allows the ferrite to maintain a fine grain size. A suitable flash sintering or SPS cycle may be as follows: temperature between 1000°C and 1300°C maintained for 5 to 20 minutes under a uniaxial pressure of 30-70 MPa. Manufacturing of a core including an air gap During the process described above, the step of fixing the first part and the second part can be carried out by gluing. In this case, the arm of the first part and the joining piece of the second part are not necessarily in direct contact. The arm and the joining piece can be separated and held together by a thickness of glue. This thickness of glue corresponds then to an air gap in the form of a bonding strip as we were able to present previously. As an option to the various implementations of the process presented so far, the following steps can also be planned: - the printing of the first part 73 is configured so that the arm 75, 76, 77 of the first part is a first arm 75, 76, 77, the first part comprising a first portion 78A of a second arm 78 of the plurality of arms of the magnetic core, the first portion 78A being solid, - the printing of the second part 74 is configured so that the second part 78A comprises a second portion 78B of the second arm 78, the second portion 78B being solid, and - the fixing of the first part and the second part is configured to place the first portion and the second portion opposite each other along two faces 78C, 78D extending perpendicularly to an axis of the cylinder corresponding to the second arm, the material located between the two faces 78C, 78D being non-magnetic. These steps are illustrated in relation to the arm 78 of [Fig. 18]. The arm 78 is formed of a first portion 78A initially included in the first part 73 and a second portion 78B initially included in the second part 74. The axial length, that is to say along the axis of the cylinder corresponding to the arm 78, of each of the first portion 78A and the second portion 78B is less than the axial length of the arm 78, measured between the joining pieces located at the ends of the arm 78, the joining pieces 122 and 126 in |Fig. 18]. More precisely, the sum of the axial length of the first portion 78A and the axial length of the second portion 78B is also less than the axial length of the arm 78. In this way, a space separates the first portion 78A and the second portion 78B. The first portion 78A has a face 78C perpendicular to the axis of the cylinder of the arm 78. The second portion 78B has a face 78D perpendicular to the axis of the cylinder of the arm 78. The two faces 78C, 78D are located opposite each other, with a gap corresponding to the difference between, on the one hand, the axial length of the arm 78 and, on the other hand, the sum of the axial length of the first portion 78A and the axial length of the second portion 78B. The space between the faces can be left empty or glue can be placed there. In this method implementation option, it is possible to position an air gap within one or more arms depending on the desired magnetic circuit.

Claims

Claims

1. A method of manufacturing a magnetic core (1), the method including: - an impression of a first part (73) of the magnetic core (1) comprising an arm (75, 76, 77), the arm (73) having a chamber (79, 110, 111) open, - an impression of a second part (74) of the magnetic core comprising a connecting piece (112, 113, 114), - an insertion of an element (115, 116, 117) into the chamber, and - a fixing of the first part and the second part so that: -- the junction piece encloses the element in the chamber, -- the arm is included in a plurality of arms of the magnetic core, the plurality of arms being arranged along edges of a convex polyhedron, each arm forming a right cylinder, and -- the junction piece is included in a plurality of pieces of junction of the magnetic core, each junction piece being located at a vertex of the polyhedron.

2. A method according to claim ! in which the element is chosen from: - an element made of phase change material, - an element made of amorphous material, - an element made of nanocrystalline material. and - a hollow ferrite tube (115, 116) comprising an internal wall and at least one fin that extends radially inside the tube from the inner wall.

3. A method according to any one of claims 1 to 2 wherein the printing of the first part and the second part is carried out at from a spinel ferrite powder by binder jet or by deposition of molten material.

4. A method according to claim 3 wherein printing the first part and the second part is made by binder jet, the process comprising before the insertion step a step of removing grains of powder separated from the core.

5. A method according to claim 3 wherein the element (117) is chosen among a phase change material element, a phase change material element amorphous material, and an element made of nanocrystalline material, the process comprising before the insertion step a step of sintering the first part and the second part, the fixing being carried out by gluing or screwing.

6. A method according to claim 4 wherein the element (115, 116) is a hollow tube, the method comprising manufacturing by stereolithography of the hollow tube, and preferably the fixing is carried out by co-sintering.

7. A method according to claim 4 wherein the element (115, 116) is a hollow tube, the fixing being carried out by co-sintering and preferably the method comprises stereolithographic fabrication of the hollow tube.

8. A method according to any one of claims 1 to 7 wherein - the printing of the first part (73) is configured so that the arm (75, 76, 77) of the first part is a first arm (75, 76, 77), the first part comprising a first portion (78A) of a second arms (78) of the plurality of arms of the magnetic core, the first portion (78A) being full, - the printing of the second part (74) is configured so that the second part (78A) includes a second portion (78B) of the second arm (78), the second portion (78B) being full, and - the fixing of the first part and the second part is configured to compare the first portion and the second portion according to two faces (78C, 78D) extending perpendicular to an axis of the cylinder corresponding to the second arm, the material located between the two faces (78C, 78D) being non-magnetic.

9. The method of claim 8 wherein the material between the two sides include glue.

10. Magnetic core obtained by a method according to any one of the re- claims | at 9.