LT COMPONENT WITH SHARED WINDING

Abstract

A component (1) includes a planar transformer (3) formed on a PCB (2) by interlacing a primary winding (33) and a secondary winding (34) within a winding window formed by a first magnetic core (31, 32). It also includes a planar inductor (4) formed by extending at least one of the windings with two connecting tracks (340) forming connection terminals (53, 54) and by arranging a second magnetic core (41, 42) around a portion (44) of each of the two connecting tracks. Preferably, the planar inductor (4) is formed at the terminals of the low-voltage winding. The connection tracks (340) are formed on the same PCB to obtain a one-piece planar component, or formed on a separate PCB to obtain a rigid-flex component where the flexible part consists of a portion (60) of the connection tracks (340). Figure for the abbreviation: [Fig 5c]

Description

Title of the invention: LT COMPONENT WITH SHARED WINDING technical field

[0001] The present invention relates to passive electrical components incorporating a planar transformer which are used to interface a primary network and a secondary network in direct current, which can equip certain aircraft onboard systems.

[0002] The present invention aims to efficiently integrate a line inductance necessary for the proper functioning of the planar transformer. Previous techniques

[0003] It is known to use planar transformers in power converters, when galvanic isolation is required.

[0004] This is particularly the case for aircraft systems, where the converter is used to connect a high-voltage electrical network, for example, a high-voltage direct current (HVDC) aeronautical network operating at 270 volts, to a low-voltage, high-current electrical network, for example, a low-voltage direct current (LVDC) network operating at 28 volts. More generally, the invention applies to HVDC power distribution.

[0005] Silicon carbide or gallium nitride components, known as "Big Gap" components, are used in power converters embedded in aircraft systems, allowing high switching frequencies with less loss, which allows a significant gain in terms of mass and volume of these embedded converters.

[0006] Planar transformers, such as the one described in application FR3129244, are designed without an inductive function in order to minimize copper losses.

[0007] These planar transformers use a primary winding and a secondary winding made in a printed circuit board (or PCB for "printed circuit board"), which are interlaced in a winding window formed by a magnetic core, to ensure an almost uniform current distribution.

[0008] The inductive function absent from the planar transformer but necessary for power transfer in the DC / DC converter is implemented using a discrete component.

[0009] In practice, a Litz wire wound inductance is often used, positioned at the terminals of the high voltage winding, where the stresses are the least.

[0010] However, this configuration introduces parasitic resonances, mainly due to the higher parasitic capacitances in a planar transformer compared to a conventional wound transformer. Voltage oscillations are critical for certain non-resonant converters, particularly those with a high voltage ratio, such as the Dual Active Bridge (DAB), which allows interfacing electrical networks, for example, 270 V HVDC and 28 V LVDC aeronautical networks.

[0011] These resonances can in particular lead to a decrease in conversion efficiency, to overvoltages on the transformer insulators and to a risk of partial discharge.

[0012] Therefore, it may be considered to place a line inductor on the low-voltage side of the transformer. However, this solution is difficult to design and offers poor performance, particularly due to the very high current flowing and the resulting copper losses, but also due to the complex interconnection required with, for example, the low-voltage (LV) bridge of the converter.

[0013] There is therefore a need to improve the integration of inductances with a planar transformer. Description of the invention

[0014] The invention aims to overcome at least some of the aforementioned drawbacks and to provide a dual-function passive component that efficiently integrates planar transformation and induction functions, capable of combining advantages of compactness, performance and simplicity of design / assembly.

[0015] In view of the foregoing, one aspect of the invention relates to a component comprising a planar transformer formed on a printed circuit board by the interlacing of a primary winding and a secondary winding in a winding window formed by a first magnetic core, at least one of the windings extending by two connection tracks forming connection terminals.

[0016] The component includes a second magnetic core arranged around a portion of each of the two connection tracks, so as to form a planar inductance at the terminals of said winding.

[0017] Thus, each connection track contributes to an additional induction function, notably by forming a half-turn in the winding window of the second magnetic core. By combining the windings of the transformer and the inductor, the design of both functions is simplified by eliminating the need for coil heads and connectors (terminations and electrical connectors) between the transformer and the inductor. Compactness is improved.

[0018] This results in a compact passive cell of type LT (inductance L, transformer T).

[0019] Optional features of embodiments are defined in the appended claims. Some of these features are explained below with reference to a system, while they can be transposed into process features.

[0020] In one embodiment, the two connection tracks forming the connection terminals are extended on the same printed circuit board as the primary and secondary windings. A single-piece planar LT component is thus obtained. "On the same printed circuit board" means that the windings and the two connection tracks consist of conductive traces formed on the same insulating substrate and separated from each other by an insulating layer. The resulting assembly forms a monolithic block.

[0021] In one embodiment, the two connection tracks forming the connection terminals are extended onto a printed circuit board separate from that of the primary and secondary windings. The resulting component is of the flex-rigid type. The connection tracks can form the mechanical linking elements, typically pivots, between the two PCBs (the planar transformer and the planar inductor). Their flexibility allows adjustment of the positioning of one PCB relative to the other. This configuration therefore offers the advantage of easier integration of the planar component into equipment, for example, aircraft equipment.

[0022] In one embodiment, the portions of the two connecting tracks are arranged on either side of a central branch of the second magnetic core. This arrangement allows for the design of relatively simple connecting track portions, for example straight or curved depending on the shape of the second magnetic core.

[0023] In one embodiment, the ends of the two connection tracks extending the winding form connection terminals of the planar component, in particular for connecting a bridge in a DC / DC converter.

[0024] In one embodiment, the winding across which the planar inductance is formed is a low-voltage winding. A transformer component thus exhibits reduced or even eliminated parasitic oscillations.

[0025] Alternatively, the winding across which the planar inductance is formed is a high-voltage winding. This eliminates the need for a conventional Litz wire winding.

[0026] In a variant combining these two configurations, the component comprises a third magnetic core arranged around a portion of each of two connection tracks extending from the other winding and forming connection terminals, so as to form a planar inductance across the terminals of said other winding. As mentioned above, the two connection traces forming the connection terminals can be located on the same PCB or on a separate PCB, in which case they can serve as mechanical linking elements between the two PCBs.

[0027] In one embodiment, the connecting tracks comprise a plurality of conductive tracks formed respectively on a plurality of layers composing the corresponding winding. This configuration simplifies the design since the layers in the winding and in the inductor can be made simultaneously.

[0028] Preferably, the layers of said corresponding winding comprise single conducting turns connected in parallel with respect to each other, the ends of each single turn of a layer being extended to form a layer of the two connection tracks.

[0029] In one embodiment, the second magnetic core is formed of a magnetic material with low relative permeability, between 10 and 100, without an air gap. This configuration limits iron losses without fringe effect.

[0030] In one embodiment, the second magnetic core is formed of a half-core in E with high relative permeability, between 20 and 5000, and a half-core in E or I with low relative permeability, between 10 and 100, without an air gap. This configuration allows the use of off-the-shelf magnetic elements.

[0031] According to another embodiment, the second magnetic core is formed of a magnetic material with high relative permeability, between 20 and 5000, with an air gap perpendicular to a principal plane of the two connection tracks, i.e., perpendicular to the principal plane of the printed circuit board and to the plane of the layers forming the windings. One or more air gaps may be provided.

[0032] The invention also relates to a bidirectional DC-DC electrical converter comprising a component such as the one described above. The invention also relates to an aircraft comprising such an electrical converter or a component as described above.

[0033] A second aspect of the invention relates to a method for manufacturing a component comprising a step of forming a planar transformer on a printed circuit board, the planar transformer comprising the interlacing of a primary winding and a secondary winding in a winding window formed by a first magnetic core, at least one of the windings extending into two connection tracks forming connection terminals.

[0034] The method further includes a step of arranging a second magnetic core around a portion of each of the two connecting tracks, so as to form a planar inductance at the terminals of said winding. Brief description of the drawings

[0035] The invention will be better understood upon a detailed study of two embodiments taken by way of non-limiting examples and illustrated by the accompanying drawings, in which:

[0036] [Fig-1] represents a one-piece planar component;

[0037] [Fig.2] represents the one-piece planar component without the upper element of the magnetic cores;

[0038] [Fig.3a] [Fig.3b] [Fig.3c] [Fig.3d] [Fig.3e] represents different layer configurations in a planar transformer;

[0039] [Fig.4] illustrates, in top view, the primary and secondary windings of the one-piece planar component according to one embodiment;

[0040] [Fig.5a] [Fig.5b] [Fig.5c] represent several sections of the one-piece planar component according to an embodiment;

[0041] [Fig.6a] [Fig.6b] [Fig.6c] represent magnetic cores of the planar inductance according to several embodiments;

[0042] [Fig.7] illustrates, in top view, a secondary winding of the one-piece planar component according to another embodiment;

[0043] [Fig.8] illustrates a rigid-flexible component according to another embodiment; and

[0044] [Fig.9] represents the one-piece planar component integrated into an electrical circuit “dual active bridge” type.

[0045] For the sake of clarity, the same elements are designated by the same reference numerals in the different figures. Furthermore, the various figures are not drawn to scale, as is customary in the representation of integrated circuits. Detailed description

[0046] The [Fig.1] is a general cavalier perspective view of a one-piece passive planar component 1, in a finished state.

[0047] It comprises, on the same substrate or printed circuit board 2 (also known as a PCB), a planar transformer 3, a planar inductor 4 and terminals or pads 53, 54 for connecting the planar component to external circuits. Such a component integrating the functions of transformer and inductor can be called an "LT component".

[0048] Fig. 2 illustrates, in cavalier perspective, the same planar monobloc component without the upper element of the magnetic cores forming the planar transformer 3 and the planar inductance 4.

[0049] Figures [Fig.3a], [Fig.3b], [Fig.3c], [Fig.3d] and [Fig.3e] illustrate, in section, different planar transformer configurations usable for the planar transformer 3. In these figures, only the left winding window is shown, the right one being identical.

[0050] The planar transformer 3 conventionally comprises a magnetic core 31 forming a winding window and surrounded by a winding. The core 31 may comprise aluminum and ferrite comprising a manganese-zinc alloy suitable for operating between one hundred and five hundred kilohertz, such as those known under the references "3C95", "N87" and "N49".

[0051] The shape defined by the core 31 can be of type “El”, “EE”, “EQ” or “ER”, typically with a central branch 32 surrounded by the winding.

[0052] The winding is formed by a stack of primary layers 33 along the X axis, dedicated to the highest voltage, and of secondary layers 34 also along the X axis, dedicated to the lowest voltage, separated by a dielectric 39.

[0053] Since the primary layers wind around the central branch 32, the entire assembly of these is also called the "primary winding" and is designated by the same reference numeral 33 for simplicity of description. Similarly, a secondary winding 34 is made up of the secondary layers. Typically, the primary and secondary windings comprise the same number of layers. By way of illustration only, five layers are shown in the figures to form said primary and secondary windings.

[0054] The primary layers 33 and the secondary layers 34 comprise respectively primary conducting turns and secondary conducting turns, i.e. windings of conductive material, typically copper deposited in thicknesses of between seventy and three hundred micrometers, preferably on the order of one hundred and seventy-five micrometers.

[0055] The primary layers 33 and the secondary layers 34 are covered with an insulating dielectric material 39 which separates them from each other.

[0056] In one embodiment, the primary layers 33 are superimposed and electrically connected in series.

[0057] In one embodiment, the secondary layers 34 are superimposed and electrically connected in series. Alternatively, the secondary layers are electrically connected in parallel.

[0058] Fig. 3a represents a so-called "non-interlaced" configuration where the two windings are superimposed, without interlacing of layers.

[0059] Figure 3b illustrates a so-called "full interleaving" configuration. The interleaving is such that a primary layer 33 alternates successively with a secondary layer 34 in their stacking along the X-axis.

[0060] Figure 3c illustrates a so-called "partially interlaced" configuration. The interlacing is such that two (or another number) primary layers 33 alternate successively with two (or another number) secondary layers 34 in their stacking along the X-axis, with the exception of the two extreme layers.

[0061] The schematic representation of the tracks in these figures indicates that each primary or secondary layer may comprise one or more conducting turns.

[0062] Fig. 3d illustrates a fully interlaced configuration described in application FR3129244 where each primary layer 33 has two conducting turns in series while each secondary layer 34 has a single conducting turn.

[0063] Similarly, [Fig.3e] illustrates a partially interlaced configuration described in application FR3129244 where each primary layer 33 also has two conducting turns in series and each secondary layer 34 has a single conducting turn.

[0064] The number of layers, the number of turns per layer as well as the interlacing of the primary and secondary windings can be determined according to the desired characteristics of the converter.

[0065] The primary winding 33 is electrically connected to the primary terminals 53, typically by extending the tracks of the first and last layers of this winding onto the PCB support, to form the primary terminals 53.

[0066] The secondary winding 34 is also electrically connected to the secondary terminals 54, typically by extending the track of each of the layers of this winding into connecting tracks 340. The ends of the connecting tracks 340 form the secondary connecting terminals 54.

[0067] As illustrated in [Fig.1] and [Fig.2], a second magnetic core 41 is arranged around a portion 44 of each of the two connecting tracks 340, so as to form the planar inductance 4 at the terminals 54 of the secondary winding 34.

[0068] The portion 44 of the connection tracks 340 dedicated to the planar inductance 4 extends between the planar transformer 3 and the secondary connection terminals 54 formed by the ends of the connection tracks 340.

[0069] The shape defined by the core 41 can be of various types, including but not limited to the types "El", "EE", "EQ", "ER", "PM", "RM", typically with a central branch 42 surrounded by the portion 44 of the connecting tracks 340, thus forming a winding in the winding window defined by the core 4L. The stacking of the portions 44 of connecting tracks 340 (illustrated for example in [Fig.5b] described later) forms an induction winding, also referred to as 44 for simplicity of explanation.

[0070] The portion 44 has a width D, corresponding to the width of the core 41 constituting the planar inductance 4.

[0071] The width D can be determined, by simulation methods, taking into account the desired inductance value as well as the number of layers in the inductance winding 4.

[0072] The secondary winding tracks 34 are extended by a length at least equal to the sum of D and the desired length of the secondary terminals 54, so as to form the induction winding 44 in addition to the connection terminals. Compared to conventional methods of designing a planar transformer, the connection tracks 340 are lengthened by a distance D (and the PCB 2 as well). Minor modifications are therefore only necessary to allow the realization of the planar LT component 1.

[0073] Figure 4 illustrates a top view of the primary and secondary windings. The primary winding 33 of the transformer 3 is wound only within the winding window of the transformer's magnetic core 31. The secondary winding 34, on the other hand, is wound within the winding window of the transformer's magnetic core 31, and its connecting tracks 340 pass through the winding window defined by the magnetic core 41 of the inductor 4 over a length D, thus creating the induction winding 44 and the inductive function.

[0074] The secondary layers 34 each comprise a single conducting turn.

[0075] The secondary winding may consist solely of conductive turns mounted (connected) in parallel with each other, using for example vias allowing the through passage of conductive materials from one secondary layer to the next.

[0076] The connection tracks 340 terminating the single turns of the secondary layers are, in the example, formed straight and parallel on the PCB 2, for a simplified design. The magnetic core 41 can advantageously be placed perpendicular to the principal Y axis of the connection tracks.

[0077] The portions 44 of these two connection tracks 340 are arranged on either side of the central branch 42 of the magnetic core 4L. This gives a coupled inductance which benefits from two half-turns corresponding to the two elongated connection tracks 340.

[0078] Sharing the winding of the planar transformer 3 and the planar inductance 4 makes it possible to do away with coil heads and connections between the transformer and the inductance.

[0079] Figure 5a is a view along section AA of the planar transformer 3, illustrating another example of a fully interlaced configuration with a single turn for each of the four primary and four secondary layers. Of course, any other configuration as mentioned above may be used as an alternative.

[0080] Fig. 5b is a view along section BB of the planar inductance 4, comprising an induction winding 44 also formed of four layers of conductive tracks.

[0081] Figure 5c is a view along the CC section of the planar monoblock LT component 1, illustrating the sharing of layers between the planar transformer 3 and the inductor planar 4. By this sharing, the connection tracks 340 comprise a plurality of conductive tracks which are formed respectively on a plurality of layers composing the secondary winding 34. The secondary winding 34 and the induction winding 44 therefore share the same layers separated by the dielectric insulator 39.

[0082] During the design of the planar LT component 1, each layer 44 of the planar inductance 4 can therefore be deposited at the same time as the corresponding layer of the secondary winding 34. As can be seen from [Fig.4], a single deposition of the same track is possible to simultaneously produce a secondary layer of the transformer 3 and a layer of the planar inductance 4.

[0083] Fig. 5c also illustrates the lengthening of the 340 connection tracks over a length D compared to a planar transformer alone.

[0084] In particular, the two connection tracks 340 are extended on the same PCB 2 as the primary and secondary windings 33, 34, so as to obtain a one-piece planar LT component 1.

[0085] In design variants, some secondary layers of the transformer 3 are not extended into connecting tracks 340. This results in an induction winding 44 formed of fewer layers than the secondary winding 34 of the transformer, while retaining the sharing of layers in the manufacturing process.

[0086] Fig. 6a illustrates a magnetic core 41 according to a first embodiment.

[0087] The magnetic core 41 is made of a magnetic material with low relative permeability, between 10 and 100, without an air gap allowing to limit iron losses and to eliminate any fringe effect which causes copper losses.

[0088] Figure 6b illustrates a magnetic core 41 according to a second embodiment in which the core is partially low permeability.

[0089] In this example, the element carrying the central branch 42 (here the half-core in E) of the core 41 is formed of a magnetic material with high relative permeability, between 20 and 5000 and the other element of the core (here the half-core in I) is formed of a magnetic material with low relative permeability, between 10 and 100. Again, the magnetic core 41 is formed without an air gap, making it possible to eliminate any fringe effect.

[0090] Figure [Fig. 6c] illustrates a magnetic core 41 according to a third embodiment.

[0091] The magnetic core 41 is here made of a highly magnetic material relative permeability, between 20 and 5000, with air gaps 43. These air gaps 43 are arranged perpendicular to the layers of the induction winding 44, and therefore perpendicular to a principal plane of the PCB 2.

[0092] The air gaps allowing the magnetic field developing there to be tangent to the conducting turns and thus to induce a current in the direction of the power current flowing in the primary and secondary windings, the fringe effect is limited.

[0093] If these figures represent a magnetic core of type "El", they also apply to any other form of magnetic core having a central branch 42, the portions 44 of the two connecting tracks 340 being arranged on either side of said central branch 42.

[0094] For example, [Fig.7] illustrates a planar inductance 4 whose shape defined by the second magnetic core 41 is of type “PM”, “RM” or “TS”.

[0095] In this [Fig. 7], the size of the PM pot has been deliberately enlarged to better illustrate that in this example, the connection tracks 340 terminating the turns of the secondary layers are no longer straight and parallel on the PCB 2, but curved in the "PM" pot. In practice, the transverse width (along the AA axis) of the central branch 32 is greater than the transverse width of the central branch 42 of the PM pot.

[0096] This example also shows an induction winding 44 formed of turns mounted (connected) electrically in parallel. Only one end of the lower layer is extended into a connecting track 340 to form one of the secondary terminals 54; the opposite end of the upper layer is extended into a connecting track 340 to form the other secondary terminal 54.

[0097] Fig. 8 illustrates a planar flex-rigid LT component 1' in which the primary windings 33 and secondary windings 34 are made on a first PCB 2, the tracks of the secondary winding being extended into connection tracks 340 on a separate second PCB 2'.

[0098] Section 60 of the extension of the tracks between the planar transformer 3 and the inductor 4 is made using a flexible PCB or "flex" technology. The flex section 60 thus constitutes a mechanical link between the two PCBs 2, 2', offering at least one degree of freedom, such as that of a pivot joint.

[0099] The planar flex-rigid LT component 1' can thus be positioned in non-rectilinear locations, with constrained geometry.

[0100] As illustrated by [Fig. 9], the planar LT component 1 can be integrated into a bidirectional DC-DC electrical converter, such as the "dual active bridge" type electrical circuit. A transformation ratio of approximately ten is obtained between an aircraft circuit 90 incorporating a high-voltage (HV) bridge 91 and the primary winding 33, and a secondary winding 34 combined with the induction winding 44 which are connected to a low-voltage (LV) bridge 92.

[0101] The low-voltage secondary winding 34, formed of parallel turns, makes it possible to obtain an efficiency greater than ninety-eight percent, by exploiting of a larger equivalent copper cross-section to carry the high current to which this winding is subjected, and by improving the distribution and amplitude of the magnetic field in the winding window which limits the proximity effects between the different copper layers.

[0102] The proposed planar LT component 1 optimizes size, weight, and efficiency because the inductance 4 results from the coupling between two half-turns extending the turns of the secondary winding 34, thus eliminating the need for coil heads and connections between the transformer and the inductance. This reduces losses and size.

[0103] This sharing of layers between transformer 3 and inductor 4 simplifies the design and manufacturing process, contributing to a reduction in manufacturing and assembly costs.

[0104] Finally, such a planar LT component 1 integrating the inductance 4 on the low voltage side makes it possible to reduce or even eliminate parasitic oscillations at the terminals of the transformer 3.

[0105] As previously stated, the manufacturing process remains simple. The primary and secondary winding layers 33, 34 are conventionally formed on a PCB 2, which is extended by a distance D (for a monoblock version) and already has an opening to receive the central branch 42 of the magnetic core 4L. Specifically, the tracks of the secondary winding 34 are extended into connection tracks 340. Their electrical connection in series or parallel is then made. The magnetic core 41 is then positioned perpendicular to the Y-axis, its central branch 42 being inserted into the opening provided for this purpose.

[0106] Of course, the present invention is not limited to the embodiments described above by way of example; it extends to other variants.

[0107] If, in the examples above, the winding across which the planar inductance 4 is located is the low voltage winding, variants may provide that the planar inductance 4 is formed across the high voltage winding according to the preceding teachings.

[0108] The line inductance required for the proper operation of the converter can also be divided between a planar inductance on the high-voltage side and a planar inductance on the low-voltage side. In this configuration, the LT 1 component has a third magnetic core arranged around a portion of each of two connection tracks extending from the other winding and forming connection terminals, so as to form a planar inductance at the terminals of said other winding.

[0109] The assembly can be made as a single unit, i.e., the planar transformer 3 and the two planar inductors on the same PCB 2. Alternatively, the planar LT component 1 can be flex-rigid in two PCBs (one containing the planar transformer 3 and one of the two planar inductors, the other containing the other planar inductor) or three PCBs (one for each element).

Claims

Demands

1. Component (1) comprising a planar transformer (3) formed on a printed circuit board (2) by the interlacing of a primary winding (33) and a secondary winding (34) in a winding window formed by a first magnetic core (31, 32), at least one of the windings extending into two connection tracks (340) forming connection terminals (53, 54), characterized in that the component (1) comprises a second magnetic core (41, 42) arranged around a portion (44) of each of the two connection tracks, so as to form a planar inductance (4) at the terminals of said winding.

2. Component (1) according to claim 1, wherein the two connection tracks (340) forming connection terminals are extended on the same printed circuit board (2) as the primary and secondary windings.

3. Component (1) according to claim 1, wherein the two connection tracks forming connection terminals are extended on a printed circuit board separate from that of the primary and secondary windings.

4. Component (1) according to any one of claims 1 to 3, wherein portions (44) of the two connecting tracks (340) are arranged on either side of a central branch (42) of the second magnetic core (41).

5. Component (1) according to any one of claims 1 to 4, wherein the ends of the two connecting tracks (340) extending the winding form connecting terminals (53, 54) of the planar component.

6. Component (1) according to any one of claims 1 to 5, wherein the winding (33, 34) across which the planar inductance (4) is formed is a low voltage winding.

7. Component (1) according to any one of claims 1 to 6, wherein the connecting tracks (340) comprise a plurality of conductive tracks (44) which are formed respectively on a plurality of layers composing the corresponding winding.

8. Component (1) according to claim 7, wherein the layers of said corresponding winding comprise single conducting turns connected in parallel with respect to each other, the ends of each single turn of a layer being extended to form a layer of the two connecting tracks.

9. Component (1) according to any one of claims 1 to 8, wherein the second magnetic core is one of: - a magnetic core formed of a magnetic material with low relative permeability, between 10 and 100, without an air gap, - a magnetic core formed of a half-core in E with high relative permeability, between 20 and 5000 and a half-core in E or I with low relative permeability, between 10 and 100, without an air gap, and - a magnetic core formed of a magnetic material with high relative permeability, between 20 and 5000, with an air gap perpendicular to a principal plane of the two connection tracks.

10. Bidirectional DC-DC electrical converter (29) comprising a component (1) according to any one of the preceding claims.

11. Aircraft comprising at least one of a component (1) according to any one of claims 1 to 9 or a converter according to claim 10.

12. A method for manufacturing a component (1) comprising a step of forming a planar transformer (3) on a printed circuit board (2), the planar transformer comprising the interlacing of a primary winding (33) and a secondary winding (34) in a winding window formed by a first magnetic core (31, 32), at least one of the windings extending into two connection tracks (340) forming connection terminals (53, 54), characterized in that it further comprises a step of arranging a second magnetic core (41, 42) around a portion (44) of each of the two connection tracks (340), so as to form a planar inductance (4) at the terminals of said winding.

Images