EP3224840A1 - Basic module for magnetic core of an electrical transformer, magnetic core comprising said basic module, method for manufacturing said magnetic core, and transformer comprising said magnetic core - Google Patents

Abstract

The invention relates to a basic module of a magnetic core of a wound electrical transformer. Said basic module is characterized in that it consists of a first (1, 2) and second (3, 4) winding that are placed one on top of the other and made of a first and second material, respectively. Said first material is a crystal material having a saturation magnetization (Js) greater than or equal to 1.5 T and magnetic losses less than 20 W/kg in sine waves having a frequency of 400 Hz, for maximum induction of 1 T, and said second material is a material having an apparent saturation magnetostriction (λ_sat) less than or equal to 5 ppm and magnetic losses less than 20 W/kg in sine waves having a frequency of 400 Hz, for maximum induction of 1 T. The cross-sections (S₁, S₂) of the first winding (1, 2) and cross-sections (S₃, S₄) of the second winding (3, 4) are such that the proportion (S₁/(S₁ + S₃); S₂/(S₂ + S₄)) of the first material, having a high saturation magnetization (Js), compared to the cross-section of both materials together, is between 2% and 50%, preferably between 4% and 40%. The invention also relates to a magnetic core of an electrical transformer, comprising at least one such basic module, to a method for manufacturing said magnetic core, and to a transformer comprising said magnetic core.

Description

 Electrical Transformer Magnetic Core Elementary Module, Magnetic Core Comprising It And Its Manufacturing Process, And Transformer Including It

 The invention relates to the field of electrical transformers that can be carried on board aircraft. Their function is the galvanic isolation between the source network and the on-board electrical and electronic systems, as well as the transformation of voltage between the primary circuit (power supply side by the generator (s) on board) and one or several secondary circuits. Moreover these transformers can be "rectifiers" by a swallowing function based on electronic components, in order to deliver a constant voltage to certain aircraft devices.

 Low-frequency on-board transformers (≤ 1 kHz) consist mainly of a soft magnet magnetic core, laminated, stacked or wound according to the construction constraints, and primary and secondary windings of copper. The primary supply currents are variable in time, periodic but not necessarily purely sinusoidal, which does not fundamentally change the needs of the transformer.

 The constraints on these transformers are multiple.

 They must have a volume and / or a mass (in general the two are very connected) as small as possible, therefore a density of power volume or mass as high as possible. The lower the operating frequency, the greater the section of the magnetic yoke and the volume (thus also the mass) of this cylinder head, which exacerbates the interest of miniaturizing it in low frequency applications. Since the fundamental frequency is very often imposed, this amounts to obtaining the highest possible magnetic flux of work or, if the electric power delivered is imposed, to reduce as much as possible the section of passage of the magnetic flux (and therefore the mass of materials), always to increase the mass power by reducing the masses on board.

 They must have sufficient longevity (at least 10 to 20 years depending on the application) to make them profitable. Therefore, the thermal operating regime must be taken into account vis-à-vis the aging of the transformer. Typically a minimum life of 100,000 h at 200 ° C is desired.

The transformer shall operate on a roughly sinusoidal frequency power supply network, with an amplitude of the output voltage that may vary transiently by up to 60% from one moment to the next, and in particular during switching on the transformer or when suddenly switching on an electromagnetic actuator. This has the consequence, and by construction, a current draw to the primary of the transformer through the nonlinear magnetization curve of the magnetic core. The elements of the transformer (insulators and electronic components) must be able to withstand, without damage, large variations of this inrush current, the so-called "inrush effect".

 The noise emitted by the transformer due to electromagnetic forces and magnetostriction must be low enough to comply with the standards in force or to meet the requirements of users and personnel posted near the transformer. Increasingly, pilots and co-pilots wish to be able to communicate not with the help of helmets but by direct voice.

 The thermal efficiency of the transformer is also very important to consider, since it sets both its internal operating temperature and the heat flows that must be discharged, for example by means of an oil bath surrounding the windings and the cylinder head, associated with oil pumps dimensioned accordingly. Thermal power sources are mainly Joule losses from primary and secondary windings, and magnetic losses from changes in magnetic flux in ddVdt time and magnetic material. In industrial practice, the volume thermal power to be extracted is limited to a certain threshold imposed by the size and power of the oil pumps, and the internal operating limit temperature of the transformer.

 Finally, the cost of the transformer must be kept as low as possible in order to ensure the best technical-economic compromise between material costs, design, manufacturing and maintenance, and optimization of the electrical power density (mass or volume). ) of the device through taking into account the thermal regime of the transformer.

In general, it is advantageous to look for the highest density of mass / volume power possible. The criteria to be considered in order to appreciate it are mainly the saturation magnetization Js and the magnetic induction at 800 A / m B ₈₀₀ .

 Two technologies for manufacturing low-frequency on-board transformers are currently used.

According to a first of these techniques, the transformer comprises a magnetic circuit wound when the power supply is single-phase. When the power supply is three-phase, the core structure of the transformer is made by two ring cores of the preceding type contiguous, and surrounded by a third torus wound and forming an "8" around the two previous toric cores. This form of circuit in practice imposes a small thickness of the magnetic sheet (typically 0.1 mm). In fact, this technology is used only when the supply frequency constrains, taking into account the currents induced, to use strips of this thickness, that is to say typically for frequencies of a few hundred Hz.

 According to the second of these techniques, a stacked magnetic circuit is used, regardless of the thicknesses of magnetic sheets envisaged. This technology is therefore valid for any frequency below a few kHz. However, particular care must be taken in deburring, juxtaposing or even performing electrical insulation of the sheets, in order to reduce both the parasitic air gaps (and thus optimize the apparent power) and to limit the induced currents between sheets. .

 In either of these technologies, a soft magnetic material with high permeability is used in embedded power transformers, and whatever the band thickness envisaged. Two families of these materials exist in thicknesses of 0.35 mm to 0.1 or even 0.05 mm, and are clearly distinguished by their chemical compositions:

 Fe-3% Si alloys (the compositions of the alloys are given throughout the text in% by weight, with the exception of that of nanocrystalline alloys which will be discussed later) whose fragility and electrical resistivity are mainly controlled by the Si content; their magnetic losses are quite low (non-oriented N.O. grain alloys) to low (G.O. grain oriented alloys), their saturation magnetization Js is high (of the order of 2T), their cost is very moderate; There are two sub-families of Fe-3% Si used either for an embedded transformer core technology or for another:

 o Fe-3% Si with Oriented Grains (GO), used for embedded transformer structures of "wrapped" type: their high permeability (B800 = 1 .8 - 1 .9 T) is related to their texture {1 10 } <001> very pronounced; these alloys have the advantage of being inexpensive, easy to form, of high permeability, but their saturation is limited to 2 T, and they have a very marked non-linearity of the magnetization curve which can cause very important harmonics;

o Non-Oriented Grain Fe-3% Si (NO), used for "cut-stack" type embedded transformer structures; their permeability is smaller, their saturation magnetization is similar to that of GOs;

 Fe-48% Co-2% V alloys, whose brittleness and electrical resistivity are mainly controlled by vanadium; they owe their high magnetic permeabilities not only to their physical characteristics (weak magnetocrystalline anisotropy K1) but also to the cooling after final annealing which adjusts K1 to a very low value; because of their fragility as soon as they stay a few seconds between 400 and 700 ° C, these alloys must be shaped in the hardened state (by cutting, stamping, folding ...), and only once that the piece has its final shape (rotor or stator of rotating machine, profile E or I transformer) the material is then annealed in the last step; moreover, because of the presence of V, the quality of the annealing atmosphere must be perfectly controlled so as not to be oxidizing; finally the price of this material, very high (20 to 50 times that of Fe-3% Si - G.O.), is related to the presence of Co and is roughly proportional to the content of Co.

 In addition to these two families of materials with high permeability (Fe-3% Si G.O. and Fe-

48% Co-2% V) currently mainly used in embedded low-frequency power transformers, Iron-based amorphous materials are sometimes encountered when the requirement on the thermal (dissipation, magnetic losses) is very strong, which then requires degrading much power density (Js = 1, 88 T). Amorphous materials are only used in coiled circuits.

 It is also known for a long time that the additions of Co in iron increase the magnetic saturation of the alloy, until reaching 2.4 T towards 35 to 50% of Co, and one could thus have expected to see use other materials based on FeCo and containing less cobalt than Fe-48% Co-2% V in embedded transformers.

Unfortunately, it turns out that these alloys with a lower Co content have a magnetocrystalline anisotropy of several tens of kJ / m ³ , which does not allow them to have a high permeability in the case of a random distribution of the final crystallographic orientations. . In the case of magnetic plates less than 48% Co for medium-frequency on-board transformers, it has long been known that the chances of success necessarily pass through an acute texture characterized by the fact that in each grain, an axis <100> is very close to the rolling direction. The texture {1 10} <001> obtained by Goss in 1946 in Fe-3% Si by secondary recrystallization is an illustrative case: however the sheet should not contain cobalt.

More recently, it has been shown in US-A-3 881 967 that with additions of 4 to 6% Co and 1 to 1.5% Si, and also using a recrystallization secondary, high permeabilities could also be obtained: B800 ~ 1, 98 T, a gain of 0.02 T /% Co at 800 A / m compared to the best sheet Fe-3% Si GO current (B10 ~ 1, 90 T ). However, it is clear that an increase of only 4% of the B ₈₀₀ is not enough to significantly reduce a transformer. By way of comparison, a transformer-optimized Fe-48% Co-2% V alloy has a B800 of approximately 2.15 T ± 0.05 T, which makes it possible to increase magnetic flux at 800 A / m for the same yoke section from about 13% ± 3%, to 2500 A / m from about 15%, to about 5000 A / m from about 16%.

Note also the presence in Fe-3% Si GB of coarse grains due to secondary recrystallization, and a very low disorientation between crystals allowing a B ₈₀₀ of 1, 9 T, coupled with the presence of a coefficient magnetostriction λ ₁₀₀ very clearly greater than 0. This makes this material very sensitive to mounting and operating constraints, which brings the B800 of an Fe-3% Si GO back into industrial practice in operation in a transformer on board about This is also the case for the alloys of US Pat. No. 3,881,967. Furthermore, Fe-48% Co-2% V has amplitude magnetostriction coefficients still 4 to 5 times more. high as Fe-3% Si, but a random distribution of crystallographic orientations and a small average grain size (a few tens of microns), which makes it much less sensitive to low stresses, and therefore does not significantly decrease the B800 in operation.

 In operation, it must therefore be considered that the replacement of a Fe 3% Si G.O. by an Fe-48% Co-2% V brings an increase in the magnetic flux constant section of the on-board transformer of the order of 20 to 25% for operating field amplitudes of 800 to 5000 A / m, so about 0.5% magnetic flux increase by 1% Co. The alloy of US-A-3 881 967 allows a 1% increase in magnetic flux by 1% Co, but as said, this total increase (4%) was considered too low to justify the development of this material.

It has also been proposed, particularly in document US Pat. No. 3,843,424, to use a 35% Co Fe-5 alloy having less than 2% Cr and less than 3% Si and having a Goss texture obtained. by primary recrystallization and normal grain growth. Compositions Fe-27% Co-0.6% Cr or Fe-18% Co-0.6% Cr are cited as making it possible to reach 2.08 T at 800 A / m and 2.3 T at 8000 A / m. These values would allow in operation, compared to a sheet Fe-3% Si-GO operating at 1 .8 T at 800 A / m, and 1 .95 T at 5000 A / m, to increase by 15% to 800 A / m and 18% at 5000 A / m magnetic flux in a given breech section, and therefore reduce the volume accordingly or the mass of the transformer. Thus, several compositions and processes for manufacturing Fe-low Co alloys (with possible additions of alloying elements) have been proposed, generally making it possible to obtain magnetic inductions at 800 A / m close to those accessible with the alloys. commercial Fe-48% Co-2% V, but with Co levels (and therefore cost) significantly lower (18 to 25%).

 In summary, the different issues facing aeronautical transformer designers can arise as well.

 In the absence of a strong requirement for noise due to magnetostriction, the compromise between the requirements on a low inrush effect, a high mass density of the transformer, a good efficiency and low magnetic losses lead to the use of solutions magnetic cores wound in Fe-Si GO, in Fe-Co or in iron-based amorphous materials, or solutions involving magnetic cores in cut and stacked pieces made of Fe-Si NO or Fe-Co.

But these requirements on a low noise of magnetostriction being more and more widespread, it is not possible to satisfy them with the previous technologies other than by increasing the volume and the mass of the transformer, because it is not known how to lower the noise other than by reducing the average work induction B _t , thus increasing the core section and the total mass to maintain the same magnetic flux of work. B should be reduced to about 1 T instead of 1.4 to 1.7 T for Fe-Si or Fe-Co in the absence of noise requirements. It is also often necessary to pad the transformer, resulting in an increase in weight and bulk.

 Only a material with zero magnetostriction would, at first glance, solve the problem, and provided that it has a higher induction of work than current solutions. Only alloys Fe-80% Ni which have a saturation induction Js of about 0.75 T and nanocrystalline alloys known as "recumbent or cut cycle" of which Js is about 1.26 T have such a low magnetostriction. But Fe-80% Ni alloys have a working induction B, too low to provide lighter transformers than traditional transformers. Only nanocrystalline would allow this relief with the low noise demanded.

 It is recalled that a material with a hysteresis cycle lying or cut is a material whose hysteresis cycle B = f (H) is such that its slope is relatively low, up to possibly intersect the abscissa axis. H.

But these nanocrystallines pose a major problem in the case of a solution "embedded transformer". Their thickness is about 20 μηι and they are wound in amorphous soft torus around a rigid support, so that the shape of the torus is preserved throughout the heat treatment resulting in the nanocrystallization. And this support can not be removed after the heat treatment, always so that the shape of the torus can be preserved, and also because the torus is then often cut in half to allow a better compactness of the transformer by using the technology of the previously wound circuit described. Only impregnating resins of the wound core can maintain it in the same form in the absence of the support which is removed after polymerization of the resin. But after a cut in C nanocrystalline core impregnated and hardened, there is a deformation of the C which prevents the two parts to be put back exactly face to face to reconstitute the closed torus, once the windings inserted. The constraints of fixing the C within the transformer can also lead to their deformation. It is therefore preferable to keep the support, which weighs down the transformer.

 The object of the invention is to provide a low-frequency electrical transformer design, suitable for use in aircraft, and to better solve the technical problems that we have just talked about, and at the lowest cost.

For this purpose, the subject of the invention is an elementary module of magnetic core of a wound-type electric transformer, characterized in that it is composed of a first and a second superimposed winding, respectively made in a first and a second second material, said first material being a crystalline material with saturation magnetization greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better still greater than or equal to 2.2 T and magnetic losses less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg and said second material being a material with apparent magnetostriction at saturation (A _sat ) less than or equal to 5 ppm, preferably less than or equal to 3 ppm, more preferably less than or equal to 1 ppm, and magnetic losses of less than 20 W / kg in sinusoidal waves of equation 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg, the sections (Si; S ₂ ) of the first winding and (S ₃ ; S ₄ ) of the second winding being such that the ratio + S ₃ ); S ₂ / (S ₂ + S ₄ )) of each section of first material with high saturation magnetization (Js) compared to the section of the set of two materials of the elementary module is between 2 and 50%, preferably between 4 and 40%.

Said first material may be chosen from grain oriented Fe-3% Si alloys, Fe-6.5% Si alloys, Fe-15 alloys with a total of 55% Co, V, Ta, Cr, Si, Al. , Mn, Mo, Ni, W textured or not, soft iron and ferrous steels and alloys at least 90% of Fe and having Hc <500 A / m, ferritic stainless Fe-Cr at 5 to 22% Cr, 0 to 10% in total of Mo, Mn, Nb, Si, Al, V and more than 60% Fe, non-oriented Fe-Si-Al electric steels, Fe-Ni alloys at 40 to 60% Ni with not more than 5% total additions of other elements, Fe-based magnetic amorphous Fe 5 to 25% in total of B, C, Si, P and more than 60% of Fe, 0 to 20% of Ni + Co and 0 to 10% of other elements, all these contents being given in percentages by weight.

 Said second material may be selected from 82% Ni-2 to 8% Fe-75 alloys (Mo, Cu, Cr, V), amorphous cobalt base alloys, and FeCuNbSiB nanocrystalline alloys.

 Said second material may be a nanocrystalline material of composition: with a <0.3; 0.3 <x <3; 3 <y <17, 5 <z <20, 0 <a <6, 0≤ β <7, 0≤γ <8, Μ 'being at least one of the elements V, Cr, Al and Zn, M " being at least one of the elements C, Ge, P, Ga, Sb, In and Be.

 It may comprise an air gap (17) dividing it into two parts.

 The air gap separating the two parts of the first windings may be different from the air gap separating the two parts of the second windings.

 Said two parts may be symmetrical.

 The subject of the invention is also a magnetic core of a single-phase electrical transformer, characterized in that it consists of an elementary module of the above type.

 The invention also relates to a single-phase electrical transformer, comprising a magnetic core and primary and secondary windings, characterized in that the magnetic core is of the preceding type.

 The invention also relates to a magnetic core of a three-phase electrical transformer, characterized in that it comprises:

 an internal magnetic sub-core composed of two elementary modules according to one of the claims 1 to 6 contiguous;

 and an external magnetic sub-core composed of two additional superimposed windings arranged in this order around the internal magnetic sub-core:

^■ a first wrap formed from a strip of material with low magnetic losses less than 20 W / kg sine wave frequency of 400 Hz for a maximum magnetic flux of 1 T, preferably less than 15 W / kg preferably less than 10 W / kg, and with apparent magnetostriction at saturation less than or equal to 5 ppm, preferably less than or equal to 3 ppm, more preferably less than or equal to 1 ppm;

 A second winding made from a strip of a high saturation magnetization material greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better still greater than or equal to 2.2 T, and low magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg;

the section (S ₁₃ ) of the first winding of the outer magnetic sub-core and the section (S ₁₄ ) of the second winding of the outer magnetic sub-core being such that the ratio (S ₁₄ / (S ₁₃ + S ₁₄ )) of the section of the material with high saturation magnetization and the section of the set of two materials of the external magnetic sub-core is between 2 and 50%, preferably between 4 and 40% and the section of high saturation magnetization material (Js) in the whole nucleus, in terms of section ratios, with respect to the total of the sections of the two types of materials in the whole core ((S ₃ + S ₄ + S ₁₄ ) ^ of between 2 and 50%, preferably between 4 and 40%.

 Said first winding of the outer magnetic sub-core may be of a material selected from 82% Ni-2-8% Fe-75 alloys (Mo, Cu, Cr, V), amorphous cobalt-base alloys, and nanocrystalline alloys. FeCuNbSiB.

 Said first winding (13) of the external magnetic sub-core may be made of a nanocrystalline material of composition:

[Fe _! - _a Ni _a l _! oo ^^^ CU _x If _y B _z Nb _a MM ^

 with a <0.3; 0.3 <x <3; 3 <y <17, 5 <z <20, 0 <a <6, 0≤ β <7, 0≤γ <8, Μ 'being at least one of the elements V, Cr, Al and Zn, M " being at least one of the elements C, Ge, P, Ga, Sb, In and Be.

Said second winding of the outer magnetic sub-core may be of a material selected from grain-oriented Fe-3% Si alloys, Fe-6.5% Si alloys, Fe-15 alloys at 50% Co total, V, Ta, Cr, Si, Al, Mn, Mo, Ni, W textured or not, soft iron and steels and ferrous alloys made of at least 90% Fe and having Hc <500 A / m, stainless steel ferritic Fe-Cr at 5 to 22% Cr, 0 to 10% total Mo, Mn, Nb, Si, Al, V and at more than 60% Fe, non-oriented Fe-Si-Al electrical steels, alloys Fe-Ni at 40 to 60% Ni with at most 5% of total additions of other elements, amorphous Magnetic Fe base at 5 to 25% total of B, C, Si, P and more than 60% Fe, 0 to 20% Ni + Co and 0 to 10% other elements.

 The said core may comprise an air gap dividing it into two parts.

 The air gap separating the two parts of the first windings of the inner magnetic sub-core and the two parts of the second winding of the external magnetic sub-core may be different from the air gap separating the two parts of the second windings from the internal magnetic sub-core and the two parts of the first winding of the external magnetic sub-core.

 The various air gaps separating the two parts of the various windings may not all be identical between the internal magnetic sub-core and the external magnetic sub-core.

The ratio between the section (S ₁₃ ) of the first winding of the external magnetic sub-core and the section (S ₃ ; S ₄ ) of each of the second windings of the internal magnetic sub-core can be between 0.8 and 1.2. .

The ratio between the section (S ₁₄ ) of the second winding of the external magnetic sub-core and the section (Si; S ₂ ) of each of the first windings of the internal magnetic sub-core can be between 0.3 and 3.

 Said two parts may be symmetrical.

 The subject of the invention is also a three-phase electrical transformer, comprising a magnetic core and primary and secondary windings, characterized in that the magnetic core is of the preceding type.

 The subject of the invention is also a method for manufacturing a single-phase electrical transformer core of the above type, characterized in that it comprises the following steps:

 a magnetic metal support is manufactured in the form of a first winding made of a first material, said first material being a crystalline material having a saturation magnetization greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T better than or equal to 2.2 T and low magnetic losses of less than 20 W / kg at a frequency of 400 Hz in sinusoidal waves, for a maximum induction of 1 T;

a second winding made of a material having, or being intended to have, after a nanocrystallization annealing, an apparent magnetostriction at saturation less than or equal to 5 ppm, preferably less than or equal to 3 ppm, better lower, is wound on said metal support; or equal to 1 ppm and magnetic losses less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for an induction maximum of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg and 2 to 50% in proportion of section of high saturation magnetized material;

 - An annealing of nanocrystallization and contraction of said second winding is carried out on said support;

 - And the two windings are secured, for example by hooping, or by bonding, or by impregnation with a resin and polymerization of said resin.

 It can include the following steps:

 an internal magnetic sub-core composed of two elementary modules is produced, each elementary module being produced as follows:

 A magnetic metal support is manufactured in the form of a first winding made of a first material, said first material being a crystalline material with a high saturation magnetization greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better than or equal to 2.2 T and low magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T;

A second winding made of a material having, or being intended to have, after a nanocrystallization annealing, an apparent magnetostriction at saturation less than or equal to 5 ppm, preferably less than or equal to 3 ppm, better lower than said metal support. or equal to 1 ppm and magnetic losses less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg, the ratio of the high saturation magnetization material section (Js) with respect to the total of the material sections of the first and second windings being 2 to 50%, preferably 4 to 40%;

Optionally annealing of nanocrystallization and contraction of said second winding is carried out on said support;

 said elementary modules are joined along one of their sides to form said internal magnetic sub-core;

 an external magnetic sub-core is produced as follows:

A third winding formed around a strip of material having, or intended to have, after a nanocrystallization annealing, an apparent magnetostriction at saturation of less than or equal to 5 ppm, preferably less than or equal to 5 ppm, is arranged around the inner magnetic sub-core; or equal to 3 ppm, better less than or equal to 1 ppm and losses magnets less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg;

 An annealing of nanocrystallization and contraction of said third winding is carried out on the internal magnetic sub-core;

 A fourth winding is arranged around said third winding in a magnetization material having a saturation greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better still greater than or equal to 2.2 T, and low magnetic losses. from less than 20 W / kg to 400 Hz sine waves, for a maximum induction of 1 T, the ratio of the section of high saturation magnetized material to the total of the material sections of the third and fourth windings being from 2 to 50%, preferably from 4 to 40%, and the proportion of high saturation magnetization material in the whole core, in terms of sectional ratios, with respect to the total of the sections of the two types of materials, being between 2 and 50%, preferably between 4 and 40%;

And said windings are secured, for example by shrinking, or by bonding, or by impregnation with a resin and polymerization of said resin.

 Said magnetic transformer core is cut to form two elementary cores, said elementary cores then being intended to be reassembled so as to define between them an air gap.

 The two elementary nuclei can be symmetrical.

 The surfaces of the elementary cores for defining the gap can be shaped and surfaced before the elementary cores are reassembled.

 The shaping and surfacing can be carried out so that the surfaces intended to define the air gap separating the first windings of the two elementary cores define an air gap different from the air gap separating the second windings of the two elementary cores.

The two elementary cores can be reassembled by shrinking by means of a crystalline material having a saturation magnetization of greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better still greater than or equal to 2.2 T, and low magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T. The inventors were surprised to note that, with a view to transforming electrical energy at frequencies of the order of a few hundred Hz or even a few kHz, for example in aeronautical transformers, where it is required as well. a high density of power density and / or mass, a low to very low emitted noise, low magnetic losses in sine waves from the magnetic core (less than 20 W / kg at 400 Hz, preferably less than 15 W / kg and preferably less than 10 W / kg, for a maximum induction of 1 T) and Joule effect (resulting from the conductors) and sufficient damping of the inrush effect (inrush current of a transformer ignition) , the "composite" wound magnetic core configuration, ie consisting of a wound magnetic core using at least two materials of clearly different natures by the composition or the properties and such that one at least of these materials is at the same time majority in volume and has a low apparent saturation magnetostriction (typically A _sat ≤ 5ppm, preferably ≤3ppm, and better ≤1ppm) with low magnetic losses at 40Hz and that at least one of these materials has a high saturation magnetization, typically Js≥ 1.5 T, preferably≥ 2.0 T, and better 2,2 2.2 T), has the following advantages (in particular with reference to the current solution the most efficient and using 100% nanocrystalline material):

 - Good mechanical strength of the entire composite core, under the effect of winding constraints, thermal stresses during annealing, maintenance constraints during cutting C core (which is only optional) but is preferred), holding stresses during operations of surfacing the cut areas, constraints for maintaining the C in a stable position under regulated airgap;

 a significant reduction in the number of manufacturing operations and in the overall cost of manufacturing, in particular by the lower consumption of nanocrystalline material

(all things being equal), and by the use of the winding support of the invention not only as a mechanical support, but also as an inrush effect damper and as a steady-state energy transformer. , in addition to the nanocrystalline circuit;

a density of power density and / or mass equivalent, or even slightly better, with respect to the solution using 100% nanocrystalline, and much higher than other mono-material solutions still widely used based on FeCo or FeSi wound, and the sufficiently low noise emitted is obtained by degrading the induction of work, and therefore necessarily weighing down the transformer. The invention will be better understood on reading the description which follows, with reference to the following appended figures:

 - Figure 1 which shows schematically an example of a three-phase transformer core according to the invention, with the windings of the transformer;

 FIG. 2 which schematically shows an example of the sub-core of the three-phase transformer of FIG. 1, which can also be used to constitute a single-phase transformer core;

 FIG. 3 which shows the relationships between noise, index of inrush and mass of the core in the reference examples and the examples according to the invention presented in the description.

 It has been said that one of the main problems posed by the usual transformers used in aircraft is its noise level which interferes with conversations between crew members.

 Transformer noise comes from two sources: magnetic forces and magnetostriction of magnetic materials used in the cores of these transformers.

 The noise resulting from the magnetic forces can be reduced quite easily in a closed magnetic circuit with very small distributed air gaps, by suitable mechanical systems for holding the various elements in electromagnetic materials (conductors and magnetic sheets).

 On the other hand, the noise of magnetostrictive origin is based on the very often non-zero and anisotropic magnetostrictive characteristics of the ferromagnetic crystal, and also on the magnetic flux which often changes direction in these crystals. Logically, to reduce or even cancel this type of noise it is necessary:

 - either choose a material with weak or no magnetostrictive characteristics

(example: the alloy FeNi80 called "Mumetal");

 or have a magnetic material and a transformer structure for which the magnetic flux will propagate only in the same crystallographic direction.

Magnetostrictive phenomena must be considered with several deformation quantities (λ ₁₀₀ , λ _{11 ΐ 5} A _sat ) or energy.

The magnetostriction constants λ ₁₀₀ and A ,,, represent the amplitude of the coupling between local magnetization and deformation of the lattice along the <100> crystallographic axes, respectively <1 1 1>. This coupling is therefore also anisotropic with respect to the crystallographic reference mark, so that for a magnetization supposed uniform of the metal (and, therefore, of direction given in the reference of the sample, and therefore also of specific direction in each of the crystals considered), each crystal would tend to deform differently from its neighbor (the crystallographic orientations being necessarily different ), but will be prevented by intergranular mechanical cohesion. The resulting elastic stresses, which can be represented in a simplified way by a magnitude σ ,,, generate a magnetoelastic energy, of order of magnitude (3/2) λσ, which partially demagnetises the material (in this expression, λ is an approximate average magnetostriction of the same order of magnitude as the constants λ ₁₀₀ and λ ₍₁₁ ) Except in some cases (for example a pull on the FeSi-GO alloys) the application of an external stress also degrades the performances: this is the inverse effect of magnetostriction, these magnetostriction constants λ ₁₀₀ and λ ₍₁₁ depend very much on the composition, and also on the fraction crystallized in the case of a nanocrystalline material, and they are known for a number of materials.

A _sat is the apparent magnetostriction at saturation. The magnitudes λ ₁₀₀ and Am refer to the magnetostriction deformations along the <100> and <1 1 1> axes of a monocrystal that is free to deform. The behavior of an industrial material (thus generally polycrystalline) introduces the internal elastic stress σ, because of the different crystallographic orientations in the presence, which amounts to hinder the deformation of each of the crystals. This results in a global magnetostriction, called "apparent magnetostriction" of the material, measured since the demagnetized state, and having no rigorous explicit relationship with the constants λ ₁₀₀ and Am, other than the same order of magnitude. This apparent magnetostriction A _sat is determined after saturation, and therefore represents the maximum amplitude of deformation of the material when magnetized, relative to its initial state "demagnetized" or not, which is in all cases an initial state of deformation unknown. A _sat is therefore a variation of state of deformation between two badly identified states. A _sat is thus a use value which occurs in the first order in the vibration of the magnetic sheets, the noise emitted or the deformation compatibility between the magnetic material and its immediate vicinity (for example the packing of a component magnetic core passive, field sensor, signal transformer ...).

In a material without pronounced texture (we shall see here the effect of a texture) and having magnetostriction coefficients very different from 0, such as an Fe3% Si-NO electric steel which is devoid of texture or has no that a not very pronounced texture, then according to the phases of excitation of the material in the transformer, the magnetic magnetization will periodically alternate at any point of the material between its direction of easy magnetization (no or little field of excitation) and a local direction more or less close to the DL Rolling Direction. This alternation, which is different from one grain to another in the metal, associated with magnetostriction coefficients λ ₁₀₀ and Am different, generates cyclic deformations of the metal, which are at the origin of the acoustic noise emitted by these vibrations.

 Concerning the low magnetic losses at medium frequency, it is necessary to know that two sizes influence the choice of the most adapted material:

 - the accessible induction B (Hm) which is located towards 90% of the saturation in order to make maximum use of the material while limiting the magnetizing A.trs and the harmonics generated by the non-linearity B-H;

 - and magnetic losses.

 In aeronautics, the on-board network has been for a long time at a fixed frequency of 400 Hz, but the variable frequency (typically 300 Hz to a few kHz) supplied directly by the generators is being used more and more. In these relatively low "medium frequencies", it is advantageous to have a high induction material with low losses (the thermal dimensioning also conditions the volume and the mass of the transformer), such as thin Fe-Co alloys, GO Fe-Si Thin Electric Steels or N.O. high saturation amorphides, possibly Fe-6.5% Si. This frequency range corresponds to skin thicknesses of less than 1/10 mm, which is entirely compatible with the need for thicknesses of this type in the case of a wound-type magnetic core technology according to the invention. . Above 0.1 mm, it is more and more difficult to wind the metal in toric form.

 Also, if only the magnetic losses of material at high Js are considered in order to reduce the mass and the volume of the magnetic core, the choice of the main known accessible materials corresponds to Table 1 below. The materials high Js are used in the invention to operate very mainly in transient mode to dampen the effect of inrush. As a result, it is mainly the low magnetostriction materials, ensuring the essential of the steady-state operation of the transformer, which will emit the magnetic losses.

Due to the thermal confinement of the transformer cores, the magnetic losses must remain low as well as the Joule losses of the conductors, in order to maintain an ambient temperature of the internal transformer of less than 150 ° C, in a cooling regime without forced convection. Typically, it is customary to consider that the magnetic losses of an embedded transformer core should not exceed 20 W / kg of magnetic material installed, preferably less than W / kg, and better still less than 10 W / kg, for a maximum induction of 1 T under sinusoidal field at a frequency of 400 Hz (this corresponds to 2 T / 400 Hz at respectively less than 80 W / kg, and preferably less than 60 W / kg and better still less than 40 W / kg). This condition must be fulfilled by the materials of all transformer core windings.

 It can be seen from Table 1 below that the amorphous or nanocrystalline materials respect the most severe limitations on magnetic losses (<5 W / kg).

The nanocrystalline material FeCuNbSiB given as an example in the various tables has the typical composition

with pei: electrical resistivity at 20 ° C and ρ _νο,: density at 20 ° C Table 1: Technical characteristics of different magnetic materials for embedded transformers

The induction of work B serves to size the magnetic circuits (FeSi, FeCo) when the frequency does not exceed 1 kHz, because the magnetic losses remain modest, therefore easy to evacuate. Above 1 kHz, the losses necessitate the use of a larger cooling system or a reduction of B, (because the losses are bound to the square of B _t ): the iron-base amorphos then appear as a interesting alternative (B, lower but much lower losses): indeed the saturation magnetization Js weaker amorphous is then no longer a disadvantage, while their low magnetic losses are a strong advantage.

 The trend in civil aeronautics is to design onboard transformers with acoustic noise emitted more and more low or very low when it is located next to the cockpit and that pilots work without a helmet to communicate . Like any embedded component, the transformer must be the lightest and the least bulky possible, consume as little power as possible and heat as little as possible, and also be able to cash without damage to its integrity (its insulators, its electronic components) large variations charge, ie large variations in the inrush current of the transformer. This inrush current, called "Inrush current," should be as low as possible.

It is established in the recent literature that the maximum current of inrush (magnetizing transient current of a transformer) is proportional to (2B, + B _r - B _s ) where B, is the nominal work induction (resulting from the design of the magnetic circuit), B _r is the remanent induction of the magnetic circuit (that is to say of the assembly consisting of the ferromagnetic core and air gaps localized or distributed according to the structure of construction of the core), and B _s is the saturation induction of the nucleus.

 To get a low maximum current of Inrush, you have to:

 a material with high saturation magnetization (FeSi or FeCo, preferably

FeNi and nanocrystalline);

 a low-remanence magnetic circuit, which can be obtained either directly by the choice of the material which constitutes it (example of the hysteresis cycle coated with nanocrystalline alloys), or by a construction effect of the cylinder head (distributed or localized gaps) , producing sufficient demagnetizing field);

 a labor induction B, weak; but this is antinomic with the high power density, miniaturization and lightening of transformers, and therefore does not constitute a satisfactory solution to the problem;

 a small section of magnetic core which would lead to using a high saturation material;

 - a large area of the coils.

In short, if we consider only the question of inrush, the ideal magnetic circuit comprises an alloy with high saturation magnetization (FeSi, FeCo) and low remanence, used with reduced induction: it passes by a design and a dimensioning optimized magnetic circuit and adequate calibration of the air gap (s) from these materials with high saturation magnetization Js.

If we combine the constraints of low space and low mass, low magnetic losses, low to very low acoustic noise and low inrush effect in an aeronautical embedded transformer, it remains to cross the most interesting solutions to optimize each size binding previously seen. Table 2 summarizes this in the case of a magnetic core structure wound and cut into two C-shaped elements, with a small and calibrated gap (hence a low B _r ) and for the same core mass magnetic, in the different cases where a single material is used to form the core. The characteristics of some materials are given for different values of B, and / or Hc.

Table 2: Expected properties of the materials that can be used to form a monomaterial nucleus (Decreasing interest ratings: excellent> very good>good>weak>poor> bad)

 It appears that with such monomaterial solutions thus known from the prior art, the types of choice are the following three:

or one sets in conditions of material with low magnetic losses associated with low thicknesses and low inductions (Fe-3% Si-GO to B, from 0.5 T, Fe-50% Co to B, from 0 , 5 T, Fe-50% Ni {100} <001> at B, 0.7 T, nanocrystalline (the index numbers corresponding to atomic percentages as is customary in the definition of such materials) to B, from 0.6 T, amorphous cobalt base to B, of 0.3 T), and then we reach good to very good performance in dissipated losses, acoustic noise emitted, A.tr, conductive losses and inrush effect, but the power density is then strongly degraded;

 - or we place high induction (1, 5 to 2 T) in different materials and we reach good to very good power densities, but then the effect of inrush and acoustic noise are significantly increased, and in all case well beyond what is now accepted;

 either a nanocrystalline material of the specified type is used, these being distinguished by a work induction of approximately 1 T and making it possible to satisfy at least all the basic needs with acceptable inrush, low noise, losses magnetic weak, A.tr (and therefore conductive losses) weak, but with a power density average.

 In wound core, the nanocrystalline known for this use is therefore the best compromise solution. But to make it even more interesting, we should find a way to do without the conservation of the winding support to reduce the total mass. Also an even better compromise between the mass and the different usage values required for an aeronautical aeronautic transformer with magnetic core with a wound core, subjected to a medium frequency of a few hundred Hz to a few kHz, whether single-phase or three-phase, would be desirable.

This objective can be achieved by the following general solution according to the invention, developed here in the most restrictive case of a three-phase transformer, illustrated in FIG. This figure is a schematic diagram, and does not represent the mechanical support parts and assembly for maintaining the various functional parts. But the skilled person can easily design these parts by adapting them to the specific environment in which the transformer according to the invention is intended to be placed. The elementary module of the invention is a magnetic core, wound type known per se, but achieved by the combination of two different soft magnetic materials, in different proportions. One, predominant in cross section (in other words in volume since all the elements of the module have the same depth), is distinguished by a weak magnetostriction, the other, minority in cross section, is distinguished by a strong saturation magnetization Js and serves as a mechanical support for the first material, an inrush limiter, and has a minor but non-negligible participation in the transformation of energy in steady state. These materials may possibly be present with identical sections / volumes, but the high saturation magnetization material Js must not exceed in section / volume the low magnetostriction material.

 The inventors were, in fact, surprised to find that in such a configuration, the nanocrystalline nuclei (low magnetostriction materials) wound around the first wound core and previously made of crystalline material with high saturation magnetization (Fe, Fe-Si , Fe-Co ...) not only were held mechanically since the support is here preserved (not only as a mechanically useful part, but especially as an essential part of the electromagnetic operation of the transformer), but that the power density obtained remained at the same level as that of a nanocrystalline core without support. Of course, we do not have, here, the disadvantages that would be related to an absence of support, namely the geometric instability of the nanocrystalline core, and possible alterations in the operation of the transformer that would result. If one chooses the material of the crystalline core, one obtains, in addition to the support function of the nanocrystalline core, important advantages on the overall operation of the transformer. These advantages are a limitation of the inrush effect during the transient regime and, in steady state, a good transformation of the energy under a medium frequency alternative, so that the power density of the transformer is not degraded by compared to what it would be with a "nanocrystalline material alone" solution, assuming that, in the latter case, it is possible to maintain a good geometric stability under stress of the two half-cores in C.

We will now describe, in the order of manufacture of a three-phase magnetic core according to the invention (combination of three elementary modules), the various possible constituents and the characteristics of a transformer structure according to the invention resulting from this invention. manufacturing. This structure is illustrated schematically in FIG. We start by making a wound composite structure of internal magnetic sub-core, this sub-core being composed of two adjoining elementary modules. The term "composite structure" means that the structure uses several magnetic materials of different natures. It is constituted as follows, and assembled in the order to be exposed.

 The structure comprises firstly a winding 1, 2 of two magnetic sub-cores each made from a strip of material consisting of a high saturation magnetization material Js and low losses, such as Fe-3% alloys. If grain oriented, Fe-6.5% Si alloys, Fe-15 alloys to 55% total Co, V, Ta, Cr, Si, Al, Mo, Ni, Mn, W textured or not, the soft iron and ferrous steels and alloys consisting of at least 90% Fe and having a coercive Hc field of less than 500A / m, Fe-Cr ferritic stainless containing 5 to 22% Cr, 0 to 10% total Mo, Mn, Nb, Si, Al, V and more than 60% Fe, non-oriented Fe-Si-Al electric steels, Fe-Ni alloys containing 40 to 60% Ni with not more than 5% additions total of other elements, Fe-based magnetic amorphides containing 5 to 25% total B, C, Si, P and more than 60% Fe, 0 to 20% total Ni and Co and 0 to 10% d other elements.

 These two windings 1, 2 each constitute the (inner) winding support of one of the two internal magnetic sub-cores of the transformer. Preferably, this winding is self-supported after extraction from the winder, but it can itself be wound on a more rigid support as light as possible so as not to overload the transformer, this support being in any type of material , magnetic or not.

The function of these windings 1, 2 of the inner magnetic sub-core is to dimensionally stabilize the final magnetic circuit in C, and also to absorb the very important A.trs and the transients that occur during the power up, during the connection of the transformer to the network, when the sudden power demand of a load ... and which causes a high inrush current in the transformer (inrush effect). This sub-part 1, 2 in high-Js material, in a transformer sized for a much lower nanocrystalline induction of work (a little below the Js of a low magnetostriction material, ie ≤1.2 T) will be then magnetized to saturation during the duration of inrush (which varies from a few seconds to 1 to 2 min.) since B _t . This makes it possible to store a lot of magnetization energy in this form in these high-Js materials, and prevents this energy from being reflected on a hypersaturation of the section of low magnetostrictive material and low Js, which would cause fields of excitations and huge currents of call. Materials with high Js are desirable, because if the requirement was only to collect the transient A.trs by a large energy storage, it would be enough to have a minimum permeability μ _Γ of at least 10 to 100 in the transient H field period during the inrush phenomenon, which will soon become greater than the inrush permeability of materials with high permeability, low magnetostriction and low Js, falling from very high values (μ _Γ > 100,000) to a value close to unity in hypersaturation zone BH.

 However, the requirement is not only to withstand transient A.s for these high Js materials, but also not to shield the internal materials of the transformer magnetic yoke in steady state. Indeed, for variable frequencies ranging from 300 Hz to 1 kHz (or more) which are more and more encountered on the aeronautical edge networks, the skin thickness is from 0.05 to 0.2 mm (depending on the material, frequency and permeability of the medium). Therefore, a high material winding Js having a thickness insufficiently small compared to the skin thickness would shield the outside field from the windings, and all the more so since there would be a large number of metal turns at high Js in the winding. It is therefore preferable to use a high Js material of small thickness (0.05 to 0.1 mm).

In addition, we want to remain with a very low acoustic noise during operation of the transformer in steady state, despite the presence of a portion of the magnetic yoke material high Js and magnetostriction ranging from "medium" to "strong". It is therefore necessary that these latter materials are not magnetically active in the steady state of the transformer, or at least they operate at an induction operating point low enough that the acoustic noise emitted is very low. This requires that the permeability of materials with low magnetostriction be much higher (1 to 2 orders of magnitude) at 300 Hz-1 kHz than the permeability of materials with high Js. This is achieved by using nanocrystalline or cobalt-based amorphous on the one hand (μ _Γ at 1 kHz> 50,000 - 100,000) and thin FeSi or FeCo alloys (μ _Γ at 1 kHz <3000), or also alloy Fe-80% Ni by reducing their thickness sufficiently (≤ 0.07mm) on the other hand.

 The high-J materials may be, for example, all the Fe-3% Si alloys with {1 10} <001> so-called Goss texture, known in the "electrical steels" under the names of the two sub-families:

- FeSi-GO for Grain Oriented; - and FeSi-HiB for High Induction, whose textures are the tightest and the performance of μ _Γ βί losses are the best.

 These performances are obtained only in the rolling direction of the materials, which is very suitable for wound magnetic cores, whereas when one deviates from this direction, the performance decreases very quickly.

 It is also possible to use the alloy Fe-49% Co-2% V-0 at 0.1% Nb, the V being partially or totally replaced by Ta and / or Zr. The performances, unlike the previous FeSi, are not related to the texture but to the optimization composition and heat treatment, and their performances are approximately isotropic in the plane of the sheet. The performance is largely preserved when the strip thickness is lowered to 0.05-0.1 mm

 It is also possible to use a Fe-10 alloy 30% Co slightly textured or with a Goss texture such as Fe-3% Si precedents. In the case of a Goss texture, which makes it possible to increase the permeability and to reduce the magnetic losses (but it is not particularly required for the high Js magnetic yoke part operating mainly transiently or at very low permanent induction ), the following materials may in particular be used:

 Fe-10 at 30% Co, preferably 14 to 27% Co, preferably 15 to 20% Co, also containing:

 0 to 2% (Si, Al, Cr, V), preferably 0 to 1% (Si, Al, Cr, V);

 0 to 0.5% Mn, preferably 0 to 0.3% Mn.

 - 0 to 300 ppm C, preferably 0 to 100 ppm C;

 0 to 300 ppm each of S, O, N, B, P, preferably 0 to 200 ppm each of S, O, N, P, B;

 The rest is Fe, accompanied by impurities resulting from the elaboration. These materials can be shaped and processed by:

 - Hot rolling ending in ferritic phase, preferably at a temperature of less than 900 ° C;

 - then two cold rolling sequences: the first passes with a reduction rate of 50 to 80%, the second passes with a reduction rate of 60 to 80%

 - ferritic phase annealing after hot rolling, and rapid temperature decrease after annealing (> 200 ° C / hr between Ad and 300 ° C)

- intermediate annealing (between the two cold rolling sequences) in ferritic phase, with a slow rise in temperature (<200 ° C / h between 300 ° C and Ac1). The different ferrous materials with high Js, previously described, are illustrated by examples in the following Table 3. When a content in one of the elements mentioned is not specified, it means that this element is only present in the state of traces, or at a relatively low content which makes it without very significant influence on the Js of the material. The possible contents of the elements other than Co, Si, Cr and V present in the alloys have not been specified, since these elements have little influence on the magnetic properties in question.

 Induction here at 800 A / m (B800), because in this type of material high Js, the application of a field of 800 A / m achieves an induction B located towards the elbow of the curve B = f (H). However, it is around the elbow of the curve B = f (H) that one reaches the best compromise between reduction of volume (high B) and low consumption of the transformer (low A.tr). The B8000 (induction at 8000 A / m) reports, on the contrary, the induction of saturation approach, exploited not only in the potential of power density (B, <B8000) but also in the reduction of the effect of inrush.

Table 3: Examples of high materials usable in the invention The structure then comprises two additional windings 3, 4. They are each superimposed on one of the windings 1, 2 high material Js previously described, "superimposed Meaning that the additional winding 3, 4 is arranged around the corresponding winding 1, 2 of high material Js which has been previously made. These additional windings 3, 4 are made with a strip of a material having both low magnetic losses and low magnetostriction, such as Fe-75 82% Ni-2 8% polycrystalline alloys (Mo, Cu, Cr, V), cobalt-based amorphous alloys, and, most preferably, FeCuNbSiB nanocrystalline alloys and the like.

 A particularly recommended polycrystalline material with about 80% Ni is also known as Mumetal. It reaches a very low magnetostriction for a composition 81% Ni, 6% Mo, 0.2 to 0.7% Mn, 0.05 to 0.4% Si, the remainder being iron, and for an appropriate heat treatment. optimization of magnetic performance, well known to those skilled in the art.

 A particularly recommended nanocrystalline material, known to those skilled in the art since the 1990s, is known for its very low magnetic losses from low frequencies up to 50-100kHz and for its ability to adjust its magnetostriction, via compositions adequate and adequate thermal treatments, at a value of zero or very close to 0. Its composition is given by the formula (the index numbers corresponding to atomic percentages as is customary in the definition of such materials):

[Fei-aNia] I OO-xy-ζ-α-β- _T Cu _x Si _y B _z Nb "M'pM" Y

with a <0.3; 0.3 <x <3; 3 <y <17, 5 <z <20, 0 <a <6, 0≤ β <7, 0≤γ <8, M 'being at least one of V, Cr, Al and Zn, M " being at least one of the elements C, Ge, P, Ga, Sb, In and Be, having a relative permeability μ _Γ between 30,000 and 2,000,000, a saturation of more than 1 T, and even 1, 25 When the composition is optimized to achieve zero magnetostriction.

 During annealing, the nanocrystalline material shrinks by about 1% from its initial amorphous strip state. This phenomenon must therefore be taken into account in anticipation in the winding of the amorphous strip around the first portion 1, 2 of the inner sub-core of high Js material, before the annealing of nanocrystallization. Otherwise the 1% retraction on the first core part can cause very strong internal stresses on the two core materials, which makes the whole fragile to the point of risking breaking and increases the magnetic losses. Conversely, this retraction favors the mechanical joining of the two types of materials, and thus, if not excessive, favors better dimensional stability of the C parts after impregnation and cutting.

Each of these bi-material windings (1, 3; 2, 4) constitutes an internal magnetic sub-core (called "elementary module"), defining a space 5, 6 in which two primary windings 7, 8, 9 will be inserted. three phases of the transformer and two of the secondary windings 10, 1 1, 12 of the three phases of the transformer. Note that if the transformer is single-phase, only one of these elementary modules alone constitutes the magnetic core of the transformer.

 The structure then comprises a winding 13, which is arranged around the assembly formed by said two internal magnetic sub-cores tightly attached along one of their sides. The winding 13 is formed from a strip of material with low magnetic losses and low magnetostriction, such as alloys Fe-75 at 82% Ni - 2 at 8% (Mo, Cu, Cr, V), alloys amorphous base cobalt, and very preferably FeCuNbSiB nanocrystalline alloys and related as defined above. This winding 13 constitutes a part of the external magnetic sub-core.

 Until this step included, it is preferable to maintain all the materials integral with each other only by added metal parts, mechanically resistant to annealing at 600 ° C. It is in fact the maximum nanocrystallization temperature that will have to be applied, preferably at the end of this step, to the entire constituting transformer core, when the materials of the windings 3, 4, 13 require it. If resins or glues are used previously to immobilize the magnetic tapes wrapped with respect to each other, then they will likely be degraded during nanocrystallization annealing. Their use should therefore preferably be postponed to a post-annealing stage of nanocrystallization.

For reasons of conservation of the magnetic flux, it is preferable to wind in this step a section of material 13, denoted Si ₃ , approximately identical to each of sections S ₃ or S ₄ which have been wound in low magnetostriction material in the inner sub-nuclei. It is also preferable to minimize the vacuum zones located between the three windings of low magnetostriction material. We will take as recommended S ₃ / Si ₃ or S ₄ / Si ₃ ratios a value of 0.8 to 1, 2 to compensate for the differences in winding perimeter and the possible differences in the air gap between the different materials we will talk about more far.

The structure then comprises a new winding 14 superimposed (in the sense seen previously about internal magnetic sub-cores) around this portion 13 with low magnetic losses and low magnetostriction of the external magnetic sub-core. This new winding 14, the section of which will be marked S ₁₄ , is formed from a band of material with high Js and low losses, such as alloys Fe-3% Si-GO, Fe-6.5% Si, Fe-15 at 55% (Co, V, Ta, Cr, Si, Al, Mn, Mo, Ni, W) textured or not, soft iron and various steels, ferritic stainless Fe-Cr at 5 to 22% Cr, 0 to 10% in total of Mo, Mn, Nb, Si, Al, V and at more than 60% Fe, Fe-Si-AI NO electric steels (no Oriented), Fe-Ni alloys close to 50% Ni, iron-based magnetic amorphs. This final winding 14 completes the supply of magnetic material in what constitutes the wound cylinder head of the transformer.

It is preferable to wind in this step a section Su of material 14 with high Js and low losses not too different from those Si or S ₂ , which are themselves close to each other or identical, and which have were wound in material 1, 2 high Js in the inner sub-cores, in order to have the same inrush attenuation effect in the three phases of the transformer. We will take 0.3≤ ~ S ₁₄ / S ₂ ≤ 3 because the material of the winding 14 high Js and low losses has a course (perimeter) of winding can be very different from that of the materials of the windings 1 or 2 placed in the center of the sub- sets, and this must be taken into account in the sizing of the composite core (this results from the application of Ampère's theorem).

 Thus the parts 3, 4 and 13 with low magnetic losses and low magnetostriction will have identical sections, or of the same order of magnitude, whereas the sections of materials with high Js and low losses of the first windings of the two sub-cores, 1 and 2 on the one hand, and the final winding 14 on the other hand, can be quite substantially different within the limits that are specified.

 The nanocrystallization heat treatment of the windings 3, 4, 13 with low magnetic losses and low magnetostriction, if necessary, can be carried out at the end of this step, all the metal materials having been assembled. But because of the contraction of the material 3, 4, 13 during the nanocrystallization, it is exposed after annealing to a detachment of the second winding 14 of the outer sub-core relative to the first winding 13 of the outer sub-core, making many more difficult the "joining" of the assembly before cutting. It is therefore preferable to apply this annealing at the end of the previous step, as previously said.

 At the end of this step of setting up the winding 14 with low magnetic losses and low magnetostriction of the external sub-core, it is, on the other hand, recommended to apply by depositing, or by prior gluing of the strips, or by impregnation. under vacuum (or any other suitable method) a resin, glue, polymer, or other comparable substance, which will transform the entire wound magnetic yoke into a resistive one-piece body with high dimensional stability under stress. A hooping may possibly replace this bonding or this impregnation, or precede it.

Then the magnetic yoke thus formed is cut so as to divide the different sub-cores into two parts 15, 16 to form two "half-circuits" elementary, after using the different technologies of immobilization of the strips of material and sub-nuclei previously mentioned. These two parts 15, 16 are intended to be separated by an air gap 17 as shown in FIG. Cutting must be done by firmly holding the magnetic yoke, within the strength of the solidified core, and by any cutting process such as wire abrasion, cutting, water jet, laser, etc. . It is preferable to divide the yoke into two symmetrical parts as shown, but an asymmetry would not be contrary to the invention.

 Then the surfaces of the air gap 17 are shaped and surfaced, followed by the repositioning of the two portions 15, 16 of the magnetic yoke cut (to find the starting structure) after a possible wedging of the air gap 17, and after insertion of the primary windings 7, 8, 9 and secondary 10, 1 1, 12 pre-realized transformer.

The air gap 17 has the function of naturally désaimanter any part of the magnetic core at the times of the electric period when the magnetic excitation becomes low or zero. Thus, if the transformer is initially stopped and, therefore, the whole magnetic yoke is demagnetized by the air gap (B _r = 0), the inrush effect that is observed when the transformer is brutally reloaded will be reduced.

 The surfacing or the calibration of the air gap 17 are not absolutely necessary for the invention, but they allow a better adjustment of the performance of the transformer. This increases the inrush performance, and makes more reproducible the characteristics of the transformers of a series of production.

 The "replacement" or "assembly" of the two parts 15, 16 of the magnetic circuit cut, and possibly surfaced and wedged, can in particular be achieved by means of a shrink-fit also using a high-Js material having properties comparable to those of the material used in the winding 14, and thus also participating (but without gap) attenuation of the inrush effect like other high-Js materials. This option is particularly interesting because it allows to further lighten the magnetic circuit, while giving it a strong mechanical cohesion.

The section of high material Js with respect to the total section, on the one hand for each sub-core taken alone, and, on the other hand, for the magnetic core taken as a whole, is from 2 to 50%, and preferably 4 to 40%. Therefore, this section is generally the minority, and in any case not the majority, in the elementary module defined externally by the winding 14 of high-material web Js superimposed on the winding 13 of low magnetostrictive tape and in each of the elementary modules of the inner sub-core.

In other words, the ratio of the winding sections between materials with high Js (SS ₂ , S ₁₄ ) and low magnetostriction materials (S ₃ , S ₄ , S ₁₃ ) must be maintained for each elementary module in a determined range so that the invention is implemented satisfactorily. The proportion of material with high Js (in terms of section ratios), relative to the total of the sections of the two types of materials, must be between 2 and 50%, and preferably between 4 and 40%. This can result in the following inequalities:

2 < ¹ °° - ^Sl <50, preferably 4 < ^{1 00} - ^Sl <40

S-, + S ₃ S-, + S ₃

2 ^{<1 00} - ^S <50, preferably 4 ^{<1 00} - ^S2 <40 2 ^{<1 00} - ^{Sl 4} <50, preferably 4 ^{<1 00} - ^Sl4 <40

S ₁₃ + S

S _! + s ₂ + s ₁₃ + s ₃ + s ₄ + s

4 100. (S ₃ + S ₄ + S ₁₄ ) For a good operation of the transformer, passing through a good balance of the masses of the different materials between the different magnetic circuits, and not to weigh too much while enjoying the benefits of the invention that provides the presence of the high Js material in all the sub-cores, it is necessary to respect the proportion in high material section Js of 2 to 50%, better 2 to 40%, both for the core of the transformer taken as a whole, which the last inequality expresses, than for each of its subsets (the two internal sub-nuclei (1, 2, 3, 4) and the external sub-core (13, 14)) in isolation, as reflected in the first three inequalities.

 Since the different elements of the transformer normally have the same depth p, these section ratios are equivalent to volume ratios of the different materials.

For the invention to function as required, it must be possible to form a "mandrel" of winding 1, 2 high Js material for low magnetostriction material 3, 4, and therefore a minimum of high material Js is necessary . The participation in the amortization of the inrush effect also requires a section minimum of material at high Js. For these two reasons, the minimum value of the high material section Js with respect to the total section of material is set at 2%, preferably 4%, for each of the sub-cores and for the core taken as a whole. .

 If the material with high Js becomes predominant in section in the sub-cores and / or the core (≥ 50%), then its mass unnecessarily increases the structure. As has been said, it only actively participates significantly in the damping of the inrush effect, whereas in the steady state of the transformer, it is desired that the high-Js material is only weakly magnetized. not to make noise (it inevitably has an apparent magnetostriction from medium to strong). Thus, the dimensioning of the transformer to achieve the desired power relies essentially on the low magnetostriction material λ. If we had less than 50% of low λ material (50% or more of high Js material), there would be essentially only this minority section that would participate in the electrical transformation. Consequently, the high Js material is limited to at most 50% of the total section of magnetic materials present in the sub-cores and the transformer core, as mentioned above.

 The following examples, which will be detailed later in Table 4, and the comments thereon, illustrate this point:

 By taking, for example, Fe49Co49V2 material as a high-grade material: - If 100% Fe49Co49V2 (Examples 2 to 5) is used to form the core of the transformer, B (transformer working induction in steady state) must be lowered. ) at less than 0.3 T to obtain a noise of 55-60 dB (whereas it will be seen that a noise of 55 dB maximum is desirable) which corresponds to a mass of more than 18.7 kg to be able to transform the requested electrical power; in this example the mass power density of the transformer core can be evaluated at a rate of 46 kV A / 18.7 kg = 2.46 kVA / kg of magnetic core, which is a power density too low to be acceptable ;

- In Example 21 with 53.3% Fe49Co49V2 section (thus 46.7% nanocrystalline material section), the noise (58 dB) is still too high to comply with the specifications; the total mass is 6.4 kg, 28% larger than that of Example 12 entirely in nanocrystalline, which would be acceptable, and the index of inrush is -0.35, which is good; Examples 19 and 20 show that an acceptable noise can be obtained with more than 50% of Fe49Co49V2, but for a total mass too high, which is of respectively 7.4 and 7.1 kg (thus 40 to 50% higher than with the nanocrystalline solution alone of Example 12);

 - Conversely, in Examples 18 and 18B, at respectively 23.6 and 39% FeCo27 section, the noise is a little too strong (56 and 58 dB) while the masses have been reduced to a level suitable; thus, having less

50% of the magnetic section made of high-grade material is a necessary but not sufficient condition for a satisfactory implementation of the invention; for example Examples 15 and 18C with respectively 23.6 and 39% section of FeCo27 emit a sufficiently low noise, for small masses of respectively 5.1 and 5.8 kg, only 2 and 16% of section more than the nanocrystalline solution alone of Example 12, but allowing to benefit from all the advantages of the invention.

 The elementary half-circuits formed by the parts 15, 16 are very dimensionally stable, especially after impregnation with a varnish and polymerization, even under the constraints of maintaining the two C parts of the elementary magnetic core. This would not be the case if we removed the high parts Js 1, 2 which serve as mechanical supports to windings 3, 4 low magnetostriction, and stiffen each elementary core.

 Magnetic alloys with low magnetostriction and low magnetic losses of the windings 3, 4 make it possible to satisfy most of the requirements, in particular the very low acoustic noise emitted, even when working induction B, close to saturation. This allows in this case to maximize the power density, particularly in the case of nanocrystalline materials where it can work up to 1, 2 T. This is the other material, high Js, winding 14 the outermost of the core that contributes the most to the damping of the inrush effect.

 Surprisingly, however, thanks to the high magnetic support material Js of the inner windings 1, 2 of the sub-cores, the inrush effect is distributed over the two types of material. Thus, the operating induction of the majority nanocrystalline material can be increased almost to saturation, which makes it possible to lighten the transformer accordingly.

The high Js alloys are characterized by a magnetostriction of medium amplitude (FeSi, FeNi, amorphous iron base) to significant (FeCo), which forces to reduce very significantly the induction of work B, (typically at most 0.7 T) to obtain a low acoustic noise. It has been realized that by using, in a clever way, the alloys with low magnetostriction and low magnetic losses and the high-alloys Js, especially preferably by the differentiated adjustment of the air gap 17 which is provided, advantageously but not necessarily, between the materials of each pair of C, so as to give it a value ε1 at the level of the first material and a value ε2 at the level of the second material, and also by the respective proportions of the materials, one could at the same time time on the one hand set a high induction of work in the low magnetostrictive part, and on the other hand set a low work induction in the high Js part. By proceeding in this way, the inrush effect is sufficiently damped and distributed over the two types of material, and the noise emitted by each of the materials remains low, while allowing a fairly high power density, in any case better than what is known in the state of the art for solutions in which a low magnetostriction noise is primarily sought.

 Examples of application of the invention and reference examples will now be described, based on FIGS. 1 and 2 and on the experimental results of Table 4 as shown in FIG.

 FIG. 2 shows a core 18 of a single-phase transformer, characterized by a rectangular-oblong shape of height h, of width I and of depth p, on which the winding of the main active material of the transformer is based: low magnetostriction. This elementary core 18 can also be integrated into a three-phase transformer circuit as shown in FIG. 1 as an elementary module.

This single-phase oblong circuit transformer module is made of a first high-Js material, of ep1 winding thickness, and with a second low-magnetostriction material wound around the first material itself wound beforehand, and having a thickness of winding ep2. The inner small and large sides of the winding 3 (second material), and which are also the outer small and large sides of the winding 1 (first material) when present (as in the examples according to the invention and in some of the reference examples), respectively, are denoted "a" and "c", and are respectively, for all the examples tested, a = 50 mm and c = 125 mm. a and c are also the dimensions of the inner sides of the windings 3, 4 of the second material, with low magnetostriction, arranged around the windings 1, 2 of the high material Js. For all tests, ep2 is equal to 20 mm. and ep1 is included, according to the tests, between 0 (absence of material at high Js) and 20 mm. The depth p is variable according to the tests, because it is designed so that the power transferred is substantially the same in all the tests (of the order of 46 kVA), considering that the values of a and c are also the same in all tests. It should be noted (see Table 4) that p can reach values as high as 265 mm for a reference test 4 using a Fe49Co49V2 alloy alone and 1 76 mm for the reference test 8 making use of a FeSi3 alloy alone. The reference solutions using a nanocrystalline alone and the solutions of the invention which make use of a nanocrystalline and a high-Js material have a much lower depth p. In the examples according to the invention, it is of the order of 60 to 80 mm.

The transformer is supplied with electrical current of nominal frequency 360 Hz. The primary supply current has an intensity of 1 15 A with a number of turns Ni generally equal to 1 turn, but being 5 turns in reference example 1 and 2 turns in the reference examples 2, 3 and 4, taking into account the considered gaps of each winding 1 and 2 on the one hand, 3 and 4 on the other hand, also taking into account the material considered for each winding (therefore of its permeability), in order to reach the induction of work B _t . A voltage of 230 V is applied to the primary. The secondary winding has, in all the examples described, a number N ₂ = 64 turns, and the rated voltage expected at the secondary is 230 V. In any case the energy conversion system in which the transformer is integrated forces the one This is to provide a constant voltage variation of 230 V. This also amounts to providing a constant three-phase power of 46 kVA.

 The magnetic core is thus made from a wound structure consisting of:

 a first material with high saturation;

 and, in addition, a second low magnetostriction material wound around the first material.

In order to always deliver the same secondary voltage of 230 V, we play on the section of the magnetic core, via the depth p of the core, while the wound thickness ep2 of the second material is kept identical for all tests, equal to 20 mm and corresponds to a constant magnetic circuit length of 430 mm. By difference, the magnetic circuit length of the first material, with a variable thickness according to the examples, ranges from 270 to 343 mm in all the examples according to the invention and also in all the reference examples with a bi-material elementary module. If we consider that P is the transformed power, like P = l.fem (intensity the primary current multiplied by the electromotive force fem generated at the secondary) is a design stress (P = constant), and that the electromagnetic force is imposed by the electrical circuit and since "fem = N ₂ .B, .section of the core. 27i.frquence ", then we must increase the section when we have to reduce B, to lower the noise.

 It is recalled that it is the second low magnetostriction material which operates very mainly in steady state, and therefore provides the voltage and output power of the transformer. On the other hand, the inrush effect comes from the combination of the magnetic behaviors of the two materials, and in order to appreciate the innovative contribution of the presence of another magnetic material (the first material) in the core, the wound thickness ep1 of this first material varies from 0 (which corresponds to an absence of the first material) to 20 mm according to the tests. This corresponds to a magnetic circuit length that varies from 0 to 343.2 mm.

The noise comes from the magnetostriction of the materials and their level of magnetization, and therefore the noise will be mainly related in steady state to the magnetic behavior of the second material. The index of inrush is given by the known formula: l _n = 2.B, + B _r - B _s for a magnetic core with a single magnetic material. This formula is generalized in the case of two materials according to:

(Si + S2) -ln = S2.B _{r 2} + Si. (2B _t , 1- _s , 1) + S2- (2B _t , 2-J _s , 2)

where Si and S ₂ are the winding sections of the first and second materials respectively, B _{r 2} is the residual induction of the second material, which is only active at the end of the steady-state period when the transformer cutoff and the passage of the core occur in the residual state, B _{t 1} and B _{1 , 2} are the working inductions, J _{s 1} and J _{s 2} are the saturation magnetizations of the first and second materials respectively. The formula can be easily adapted to the case where more than two materials are used.

 The induced voltage (in other words, the electomotrice force fem) by the transformer is called ddVdt. It is thanks to it that we transform the electrical power P requested: P = fem.l where I is the intensity of the magnetizing current of the transformer.

The noise emitted by the various examples made of transformer wound cores is measured by a set of microphones arranged around the transformer, in the median plane of the magnetic yoke. The various examples of magnetic cores use only one (references) or two (some references and the invention) materials, namely soft magnetic materials (FeCo27, Fe49Co49V2, Fe-3% Si-GO, FeSi grain-oriented electrical steel, _! nanocrystalline FeCuNbSiB the type [Fe - _{Ni has} the oo- _x - _y - _z - _a ^ CU _x Si _y B _z Nb _has the TpM ^ with a = 0;! x = 1; y = 15, z = 7 , 5; a = 3; β = γ = 0. The material (s) is (are) wound (s) according to the basic structure defined above.

The examples in Table 4 below are dimensioned and powered to always transform substantially the same power, namely about 46kVA. This three-phase power being given by ^ dO / dt with ddVdt =. (^ S S ^ B _{2 2 2} ) ₂ ). (OJ = 230V, where h = 1 15A, N ₂ (number of turns of the secondary ) equal to 64, ω (pulsation) = 2.TT.f, f being the frequency, here equal to 360 Hz, Si and S ₂ (magnetic yoke sections of the first and second materials respectively) respectively equal to (H.epl ) and (H.ep2), and B, _j is the work induction of material i.

Another possibility is to precisely adjust the air gaps (after cutting) ε1 and ε2 between the half-circuits of the windings of the first and second materials respectively, giving them, if necessary, different values during the shaping of the cutting areas, so to be able to limit the magnetization of one material with respect to the other. Otherwise some uncontrolled magnetization levels of material 1 could increase the magnetostriction or the inrush effect too much. It must, however, be remembered that increasing an air gap increases the current required for magnetization at B _T , and therefore degrades the efficiency of the transformer. A balance must therefore be found between the advantages and disadvantages of the practical use of this solution.

For example, in example 13 of the invention, the minimum residual air gap ε2 between the two half-circuits of the second material (the nanocrystalline material) is evaluated at 10 μηι, and then the equivalent relative magnetic permeability of the magnetic circuit "material 2 + gap" makes the intrinsic permeability μ _Γ , ₃ ΐ 2 of the material 2 increase from 30,000 to 1 7670 in the case of the example (by applying the formula - - ~ air gap + - -) . If the air gap ε2 had been ten times larger (100 μηι),

would have an intrinsic permeability μ _Γ , _θς , η ₃ ΐ2 = 3760, four times less than previously. Or (according to Ampère's theorem), HL = H (L being the average length of magnetic circuit) and H = Β / μ _{Γ θς} as long as the material works with an approximately linear curve B = f (H) (case of transformer). So by keeping B, constant (to maintain the constant electromagnetic force and transferred power, as said before), it is necessary to compensate for an increase in the gap (and therefore a decrease of μ _{Γ, θς} ) by an increase in the intensity I of the magnetizing current, which leads to a degradation of the efficiency of the transformer. If we consider in the same example 13 the air gap ε1 magnetic circuits high material Js, we conclude that a gap ε1 3.5 mm limits the equivalent permeability of the first material (here FeCo) to 0.05 T (see the formula μ _{Γ, θς} above), and therefore a low noise of 43 dB. If the air gap ε1 is reduced to 10 μηι, thus to a value equal to that of ε2, then the high Js FeCo material greatly exceeds the induction of 1 T in steady state of the transformer, and the noise of the FeCo then becomes predominant. and unsatisfactory (well above 55 dB), but may be eligible for the duration of the Inrush effect (a few fractions of a second to a few seconds).

 The general rule of limiting the effects of inrush and noise is that, like the induction of work B, has a degrading influence on both the inrush effect and the magnetostriction noise, it is necessary to lower B to mitigate these effects. But we have to compensate for this drop in B, by increasing the magnetic section to keep ddVdt and the power transformed to the same level.

 The specification of this aeronautical transformer is that the noise must be at most 55 dB, at least outside the periods during which the inrush effect is felt, and the inrush factor less than or equal to 1, with a magnetic core mass as low as possible. In addition, the total mass of magnetic materials should not exceed about 6.5 kg. It will be seen that in order for this latter condition to be fulfilled at the same time as the other two, the total cross-section of material at high Js with respect to the total section of magnetic materials in the core must not exceed 50%. This condition must also be respected if one reasons on each of the internal and external sub-nuclei taken in isolation. In order not to weigh down Table 4, only the ratio of the total sections has been specified, but it should be understood that all the examples according to the invention also respect the condition for each of their sub-nuclei.

 The examples in Table 4 show the following. Those denoted "ref" are reference examples, those denoted "inv" are examples according to the invention.

 Examples 1 to 12, 18, 18B, 19 to 21 inclusive of Table 4 are therefore reference examples, and Examples 13 to 17 inclusive, 18C, 22 to 24 inclusive are examples according to the invention which respond to all criteria of the specifications as defined above.

It will be noted that for the reference examples 1 to 12, no gap was provided in the second material. For all the other examples, whether they are reference or according to the invention, an air gap ε2 of 10 μηι has been provided in the second material. For examples 13 to 24, whether reference or according to the invention, there is provided both an air gap ε2 of 10 μηι in the second material and an air gap ε1 in the first material, ε1 can take various values according to the tests, and ε1 being different from ε2, except for example 24 where ε1 = ε2 = 10 μηι. It should be understood that in these examples, ε1 and ε2 are the same for all elements of the core: the two inner sub-nuclei and the outer sub-core.

To calculate the volumes and deduce the sections of the different materials, the density was 7900 kg / m ³ for FeCo27, 8200 kg / m ³ for FeCo50V2, 7650 kg / m ³ for FeSi3, 7350 kg / m ³ for the nanocrystalline.

 The Js of the different materials are 2.00 T for FeCo27, 2.35 T for

FeCo50V2, 2.03 T for FeSi3, 1, 25 T for the nanocrystalline.

Second Regime Mass and section

 permanent material material

 Ex. Materials ep2 ep1 P B, Br B, άΦ / dt Noise Index Mass mass Mass% wt% Power Air gap 2 + 1 mm mm mm material material material (V) (dB) inrush total material latia of three-phase section material

 2 2 1 1 (kg) 2 (kg) material of (kVA) 2

B (T) Br (T) Bu (T) (kg) 1 (at material z2 (xm) high Js 1 (at

 high

 js)

 ref FeCo27 20 0 40 2 0.95 - 231, 62 105 2.95 0 2.7 2.7 0 0 46.14 0 ref Fe49Co49V2 20 0 40 2 1, 3 - 231, 62 115 2.95 0 2, 8 2.8 0 0 46.14 0 ref Fe49Co49V2 20 0 53 1, 5 0.975 - 230.18 96 1, 63 0 3.7 3.7 0 0 45.85 0 ref Fe49Co49V2 20 0 80 1 0.65 - 231, 62 75 0.3 0 5.6 0.6 0 0 46.14 0 ref Fe49Co49V2 20 0 265 0.3 0.195 - 230.16 61 -1, 56 0 18.7 18.7 0 0 45.85 0 ref FeSi3 20 0 53 1, 5 1, 05 - 230.18 92 2.02 0 3.5 3.5 0 0 45.85 0 ref FeSi3 20 0 72 1, 1 0.77 - 229.31 85 0 , 94 0 4,7 4,7 0 0 45,67 0 ref FeSi3 20 0 176 0,3 0,21 - 229,31 57 -1, 22 0 19,0 19,0 0 0 45,67 0 ref FeSi3 20 0 72 1, 1 0 - 229.31 85 0.17 0 4.7 4.7 0 0 45.67 10

Nanocrystalline 20 0 72.3 1, 1 0.055 230.26 40 1, 01 0 4.6 4.6 0 0 45.87 10 ref cycle lying down

 or cut

 1 1 Nanocrystalline 20 0 94 0.85 0.0425 231, 33 40 0.49 0 5.9 5.9 0 0 46.06 10 ref cycle

 or cut

 12 Nanocrystalline 20 0 79.5 1 0.05 230.18 41 0.8 0 5.0 5.0 0 0 45.85 10 ref cycle

 or cut

 13 Nanocrystalline 20 2 72 1, 10 0.06 0.05 230.35 43 0.71 0.40 4.6 5.0 8.0 10.8 45.88 10 inv cycle lying down

 or cut +

Fe49Co49V2 14 Nanocrystalline 20 1, 7 70 1, 1 0.055 0.4 229.83 46 0.8 0.33 4.4 4.7 7.0 9.3 45.78 10 cycle lying down

inv

 or cut +

 FeCo27

 Nanocrystalline 20 5 66.3 1, 1 0.055 0.4 230.35 49 0.49 0.90 4.2 5.1 17.6 23.6 45.88 10 cycle lying down

inv

 or cut +

 Fe49Co49V2

 16 Nanocrystalline 20 2.1 68.3 1, 1 0.055 0.6 229.98 48 0.8 0.44 4.3 4.7 8.5 11, 2 45.81 10 cycles lying down

inv

 or cut +

 Fe49Co49V2

 17 Nanocrystalline 20 2.5 66.5 1, 1 0.055 0.75 229.84 52 0.8 0.46 4.2 4.7 9.8 13.1 45.78 10 recumbent cycle

inv

 or cut +

 Fe49Co49V2

 18 Nanocrystalline 20 5 60 1, 1 0.055 0.9 230.18 56 0.69 0.81 3.8 4.6 17.6 23.6 45.85 10 recumbent cycle

ref

 or cut +

 FeCo18

8B Nanocrystalline 20 10 56.7 1, 1 0.055 0.6 229.83 58 0.29 1, 44 3.6 5.0 28.8 39 45.78 10 recumbent cycle

ref

 or cut +

 FeCo18

8C Nanocrystalline 20 10 66 1, 1 0.055 0.2 229.31 53 0.02 1, 68 4.2 5.88 28.6 39 45.67 10 recumbent cycle

inv

 or cut +

 FeCo18

 19 Nanocrystalline 20 20 69 1, 1 0.055 0.05 229.74 53 -0.62 3.06 4.4 7.4 41, 4 57.8 45.76 10 recumbent cycle

ref

 or cut +

 Fe49Co49V2

 Nanocrystalline 20 17 69.5 1, 1 0.055 0.05 229.90 52 -0.49 2.73 4.4 7.1 38.5 53.3 45.79 10 recumbent cycle

ref

 or cut +

 FeCo27

 21 Nanocrystalline 20 17 69.5 1, 1 0.055 0.2 229.90 58 -0.35 2.46 4.0 6.4 38.4 53.3 45.78 10 ref cycle lying down

 or cut +

 FeCo27

2inv Nanocrystalline 20 5 71, 5 1, 1 0.055 0.05 230.30 47 0.42 0.90 4.5 5.4 16.7 23.6 45.87 10 cycle lying

 or cut +

 FeSi3

 23 Nanocrystalline 20 5 71, 5 1, 1 0.055 0.05 230.30 50 0.42 0.90 4.5 5.4 16.7 23.6 45.87 10 recumbent cycle

inv

 or cut +

 FeSi3

 24 Nanocrystalline 20 5 59 1, 1 0.055 1 230.61 52 0.8 0.74 3.7 4.5 16.4 23.6 45.93 10 cycles lying down

inv

 or cut +

 FeSi3

Table 4: Performances of the different configurations of tested nuclei

An entirely nanocrystalline circuit (reference examples 10 to 12) makes it possible, of course, to meet the requirements of the noise and inrush specifications, for a single mass of magnetic circuit that can be as low as 4.6 kg, which would be satisfactory. At first glance. However, this mass does not include non-magnetic media of the magnetic circuit, which can be made, for example, of wood, teflon or aluminum, and which can be a mass of several hundred grams. The nanocrystalline solution alone necessarily requires the use of a temporary or permanent winding support. In the case where it is permanent, it increases the mass of the nanocrystalline circuit as we have just said.

 In all cases (permanent support or not), this support must be realized, while it does not participate anyway in the electrical operation of the transformer, contrary to the cases of the invention. The cost of producing the support is therefore not profitable in the design of the transformer, contrary to the cases of the invention. Examples 10 to 12 are therefore not considered to fully meet the specification of the invention, and are classified as references.

 To clarify this important point, we can compare the reference examples

(Nanocrystalline alone) and 17 according to the invention (nanocrystalline composite core recumbent or cut cycle + FeCo27). These two examples are chosen because they can be considered to be the best performers for their respective technological choices, since they have the same Inrush index. The noise emitted is lower for the 100% nanocrystalline solution (41 dB compared with 52 dB for the nanocrystalline composite core recumbent or cut + FeCo27 solution), but in both cases the noise is below the allowable threshold of 55 dB.

 Example 12 uses a mass of nanocrystalline material of 5.0 kg, to which must be added a minimum mass of 200 to 300 g of teflon, aluminum or nonmagnetic stainless steel. The two possible cases for this example were considered: permanent support and non-permanent support.

 Table 5 lists the successive operations in these three embodiments, and compares the orders of magnitude of the costs of each step (from +: inexpensive to +++: expensive; 0: step absent from the embodiment) solutions in the case of producing a functional sub-assembly of a single torus (single-phase transformer type):

Solution 100% Solution 100% Cost

 Cost of Cost

Nanocrystalline Step (No. 12), Nanocrystalline (No. 12), Solution According to the Invention (No. 20)

 the stage the step support non permanent permanent support the stage

 Realization of a

 Realization of a support

 support non Realization of a support

1 core permanent

 core +++ magnetic metal magnetic FeCo ++ permanent

 magnetic

 Winding the tape

 Wrapping the tape Wrapping the tape

 nanocrystalline on the support, amorphous on an amorphous on a support

 2 + with a contraction wedge of + metal support + extractable metal

 nanocrystallization (about 1%) extractible (ferrule) ("ferrule")

removed at the end of winding Annealing Annealing

 nanocrystallization and nanocrystallization and Annealing of nanocrystallization and contraction of the ribbon + contraction of the ribbon + contraction on the amorphous FeCo + support on the amorphous ferrule on the ferrule

Core extraction

 Core extraction

 nanocrystalline from the

 + nanocrystalline ferrule + 0 0 ferrule

 Nanocrystalline core

 Nanocrystalline core placed

 placed on the support

 on permanent support

 non-permanent with + + 0 0 with minimal clearance

 minimal game

 Impregnation impregnation of

 Impregnating the whole set "kernel + ++ the whole" kernel + ++ ++

 "Core + support"

 support »support»

 Polymerization of Polymerization of

 Polymerization of the whole set + set + +

Extraction of the support

 0 0 non-adherent +

 Cut in two C Cut in two C equal

 equal to the nucleus of the core + support + Cut in two C equal +

 ++

 impregnated impregnated

 Surfacing Surfaces Surfacing Face Surfacing Surfaces

 + cut C + cut C + C

 Assembly Mechanical assembly at

 Mechanical mechanical assembly with gap air gap set (hold ++ ++

 +++ air gap adjusted (possible wedge) set (possible wedge) possible)

Table 5: Comparison of the costs of solutions 12 (reference) and 17 (invention)

Table 5 shows that there are fewer operations in the case of the invention, and, moreover, some of the operations common to the various solutions are less expensive in the case of the invention. Indeed, when cutting and assembling C parts in 100% nanocrystalline material (Example 12 without permanent mechanical support), the lack of rigidifying mechanical support (case "without permanent support") requires maintaining the C with care, so using suitable clamping jigs so as not to deform and damage the parts.

In the case of the reference example 12 with permanent support, the precautions are the same as for the invention, but in this case, it weighs the final core, and the cost of the support is added to each magnetic core product. In the case of example 17 according to the invention, the support FeCo constitutes a mechanical core avoiding irreversible mechanical deformations, and is at the same time used functionally electromagnetically and electrically.

 Finally, compared to the invention, the 100% nanocrystalline solution of the prior art (Example 12) is either a little more expensive because of the greater number of operations and heavier because of the mass of the support ( case of permanent support), or (case of non-permanent support) of equal or slightly higher mass, but in any case significantly more expensive to achieve. It does not therefore, globally, a satisfactory solution to the problems that the invention sought to solve.

 Returning to Table 4, we see that a mainly nanocrystalline circuit with an additional Fe-27% Co alloy circuit in certain limited proportions makes it possible to achieve equivalent or slightly better mass performances (final mass close to 4, 5 kg in the best case), while also respecting the specifications in inrush and noise, if it is compared to a 100% nanocrystalline solution with non-permanent support (see above). This optimum sizing corresponds, in the case of the examples according to the invention, to a proportion in FeCo or FeSi section of 9 to 40% approximately, and 7 to 29% by weight approximately, with respect to all the materials magnetic core. This optimum is also valid for each of the sub-nuclei taken alone.

 By further increasing the proportion of FeCo, and thus increasing the magnetic circuit (more than 30% by weight and more than 50% in FeCo section, Examples 19, 20 and 21), it can be seen that the effect of Inrush can be drastically reduced to a negative index. In this case, the magnetic circuit reaches a mass of the order of 7 kg (for a zero inrush index). This mass can, however, be considered a little too high for this technical solution to be fully satisfactory, especially since, moreover, the noise is only relatively below the acceptable maximum of 55 dB (Examples 19 and 20). ) or is above this acceptable maximum (Example 21). A mass of the order of 6.5 kg would generally be considered acceptable, but only if, on the other hand, the conditions on noise and inrush are respected. This explains why Example 21 is not considered to fall within the scope of the invention.

The use of FeSi-GO (electric steel Fe-3% Si with Oriented Grains), instead of FeCo, in the previous case allows to observe the same trend results as the previous case, but by adding some weight to the magnetic circuit if you want to get a comparable inrush index. The use alone and without localized air gap (that is to say with an uncut magnetic circuit), and high induction, traditional materials aeronautical embedded transformers (FeCo27, Fe49Co49V2, FeSi3) leads to very low masses magnetic circuit (examples 1, 2, 3, 6), but also to a very large noise (92 to 1 15 dB), well above the allowable limit of 55 dB, and to a very important inrush effect (index from 1.63 to 2.95) which will degrade certain electronic components in the on-board network. Note that if we cut the circuit to obtain a localized air gap and a very low remanence Br, then the effect of inrush would be much less important. But the noise would remain as strong and the cost of implementation would be much higher.

 The use of these same crystalline materials alone, but at a significantly lower induction, makes it possible to substantially reduce the inrush effect and the noise (examples 4, 5, 7, 8, 9) until approaching ( noise) or reach (inrush) the permissible limits of the specifications. However, when this situation is obtained (examples 5 and 8), the mass of the magnetic circuit is of the order of 18-19 kg, which is three times greater than that of the reference solutions based on nanocrystalline alone and induction high, or solutions according to the invention wherein the nanocrystalline is associated with FeCo or FeSi.

 FIG. 3 summarizes the performances of different possible solutions of magnetic circuit in an inrush-noise index diagram where the transformer masses corresponding to the different points are also specified.

 The maximum noise values of 55 dB and inrush index of 1 required by the specifications mentioned above are shown in dashed lines. Surrounding the area in which are the examples that meet these points of the specification and have, in addition, a section of high material Js referred to the total section of magnetic materials of not more than 50%, and sections of high material Js reported to the total sections of the magnetic materials of each sub-core of 50% maximum. This last point, which is also part of the specifications, allows, in addition, to guarantee that the core of the transformer is of a very reduced weight, of the order of 6.5 kg or less.

It clearly appears that the invention makes it possible, by the use of a nanocrystalline circuit combined with FeCo or FeSi, to respect the noise and inrush limits by using magnetic circuits that are much lighter than the solutions in question. traditional crystalline materials (FeSi, comparable FeCo) used alone. As for the solutions using a nanocrystalline alone, their performances, with equal mass, are fairly comparable to those of the invention in terms of noise and inrush index, but we saw in Table 5 that the cost of realization of these solutions was substantially higher than that of the embodiments according to the invention.

 The inrush index is always a strictly decreasing function of the mass of the magnetic yoke. But this curve is not linear, and it allows in the case of the analyzed example to determine fairly low mass magnetic core solutions (4 to 6.5 kg) for an already very low inrush index. In a different way the noise depends not only on the mass, but also on the choice of the material (s) used (via their magnetostrictive properties).

 It thus clearly appears that the solutions of the invention based on nanocrystalline associated with another material (FeCo or FeSi in particular) make it possible to combine a low mass (4 to 6.5 kg), a low noise and a low 'inrush, and for a manufacturing cost and complexity as moderate as possible.

 Variations of the invention can be envisaged.

 Several high Js materials can be used in the same magnetic core, for example a Goss-textured Fe-3% Si alloy in the inner coil of the inner sub-cores and an Fe-50% Co alloy in the outer coil of the sub-core. - external core.

Several low magnetostrictive materials can be used in the same magnetic core, such as, for example, a nanocrystalline FeCuNbSiB alloy of the composition specified above, in the inner coil of the inner sub-cores and an amorphous cobalt base in the coil. outside of the outer sub-core. It is best to use the same material for both inner sub-cores. It is preferable to keep the "J _s .Section" magnetic flux conservation rule between the three sub-parts involved in low magnetostriction materials.

 According to the invention, the use of nanocrystalline materials is recommended with respect to the use of other types of low magnetostriction materials.

 Indeed, the nanocrystalline materials of composition FeCuNbSiB cited, which are preferred but not exclusive examples of materials usable for the implementation of the invention, are known to allow their magnetostriction to be adjusted to 0 by a suitable heat treatment, while their saturation magnetization remains relatively high (1, 25 T), so conducive not to weigh too much on the transformer (see principles of sizing already recalled affecting d () / dt and the inrush).

The invention is not valid for a three-phase structure with two sub-cores placed side by side and nested in a third sub-core, but is also applicable to a simple single-phase transformer magnetic core, or any other nesting of a higher number of magnetic sub-cores, for example in the case of polyphase transformers with more than three phases. The skilled person can easily adapt the design of the transformer according to the invention to the latter case.

 The cutting of the finished magnetic core, forming the gap 17, so as to better fill the winding window and thus reduce the mass / volume of the magnetic core, is not essential, but it is very preferable both for the previous reason since we increase the power density, via the optimal filling of the winding window, but also to lower the remanent induction of the magnetic circuit. An additional advantage of the cutting is to possibly be able to differentiate the air gaps ε1 and ε2 of the two materials, in order to better control the maximum level of magnetization of the first material with high Js and high magnetostriction.

 The adjustment of the air gap can therefore be different between low magnetostriction materials and high Js materials, as has been seen in most of the examples according to the invention of Table 4 and as shown in FIGS. 1 and 2. the magnetostriction is very low, the cyclical deformation of the materials will be very weak and the wedging of the air gap will propagate and amplify only little noise. On the other hand, for high Js materials, which are very magnetostrictive, even for low steady-state working inductions (less than 0.8 T or even less than 0.4 T), the vibrations may still be sufficient to generate a noise greater than highest demands. In this case it may be preferable to machine a slight air gap, greater than that of the low magnetostriction material, so that the high Js materials are not in contact with the wedge, which reduces the noise emission.

 If we see an interest, we can also provide different values of ε1 and / or ε2 for the various parts of the core, in other words. that the air gaps (ε1, ε2) separating the two parts of the various windings (1, 2, 3, 4, 13, 14) are not all identical between the internal magnetic sub-core and the external magnetic sub-core.

 The surfacing of the cutting faces of the magnetic core is not essential, but it is preferable because it allows a better sizing of the performance of the transformer. This makes it possible to increase the inrush performance, and to make the transformers more reproducible during an industrial production.

Calibration of the air gap by a shim is not essential but it is preferable to precisely adjust the residual induction (linked in particular to the inrush effect) and the maximum level of magnetization accessible in each material, and to make transformers more reproducible in an industrial production. The symmetry of cutting of the magnetic core is not essential.

 In case of non-cutting, it is not essential to glue, impregnate, fix the different metal parts of the cylinder head, more rigidly and tightly than do the different tight windings and the heat treatment (s) (s) ).

 The different materials do not necessarily have the same width. For example, three FeCuNbSiB nanocrystallizable amorphous strips of width I each may be wound around a pre-wound corona of FeSi or FeCo internal sub-core of width 31. This brings the advantage of providing the same mechanical support of winding for FeCuNbSiB strips which are especially easy to produce and use when their width is less than 20-25 mm, whereas the needs for the magnetic cores of embedded transformers can largely exceed such widths.

 As an alternative to the previous solution, it is also possible to stack different magnetic cores with the same widths of material, in order to also obtain a larger macro-torus before gluing, fixing, impregnation, mechanical setting or other, and then cutting, surfacing then assembly of prefabricated windings.

 All materials, or only some of them, can be wound in the amorphous state or hardened or partially crystallized (as appropriate), or be wound in the nanocrystallized state (FeCuNbSiB), relaxed (amorphous iron base or cobalt base) or crystallized (Fe-80% Ni, FeCo, FeSi, other polycrystalline materials).

Claims

1. - Elementary module of magnetic core of wound type electric transformer, characterized in that it is composed of a first (1; 2) and a second (3; 4) superimposed windings, respectively made in a first and a second material, said first material being a crystalline material with saturation magnetization (Js) greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better still greater than or equal to 2.2 T and lower magnetic losses at 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg and said second material being a material with apparent magnetostriction at saturation ( A _sat ) less than or equal to 5 ppm, preferably less than or equal to 3 ppm, better less than or equal to 1 ppm, and magnetic losses less than 20 W / kg in sinusoidal waves of frequency 400 Hz, p for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg, the sections (Si; S ₂ ) of the first winding (1; 2) and (S ₃ ; S ₄ ) of the second winding (3; 4) being such that the ratio (Si / (Si + S ₃ ); S ₂ / (S ₂ + S ₄ )) of each section of first material with high saturation magnetization (Js) compared to the section of the set of two materials of the elementary module is between 2 and 50%, preferably between 4 and 40%.

 2. - elementary module according to claim 1, characterized in that said first material is selected from oriented grain oriented Fe-3% Si alloys, Fe-6.5% Si alloys, Fe-15 alloys at 55% by weight. total Co, V, Ta, Cr, Si, Al, Mn, Mo, Ni, W whether textured or not, soft iron and ferrous steels and alloys consisting of at least 90% Fe and having Hc <500 A / m, Ferritic stainless Fe-Cr 5-22% Cr, 0-10% total Mo, Mn, Nb, Si, Al, V and more than 60% Fe, electric steels Fe-Si-AI undirected, Fe-Ni alloys at 40 to 60% Ni with at most 5% total additions of other elements, Fe-based magnetic amorphous at 5 to 25% total of B, C, Si, P and more than 60% of Fe, 0 to 20% of Ni + Co and 0 to 10% of other elements, all these contents being given in percentages by weight.

 3. - elementary module according to claim 1 or 2, characterized in that said second material is selected from alloys Fe-75 to 82% Ni - 2 to 8% (Mo, Cu, Cr, V), amorphous alloys base cobalt, and FeCuNbSiB nanocrystalline alloys.

 4. - elementary module according to claim 3, characterized in that said second material is a nanocrystalline alloy of composition:

[Fe ₁ . _{a Ni] 10} oxyza ^ - _T Cu _x Si _y B _z Nb _a M _p M "Y

 with a <0.3; 0.3 <x <3; 3 <y <17, 5 <z <20, 0 <a <6, 0 <β <7, 0 <γ <8, M 'being at least one of the elements V, Cr, Al and Zn, M " being at least one of the elements C, Ge, P, Ga, Sb, In and Be.

5. Elementary module according to one of claims 1 to 4, characterized in that it comprises a gap (17) dividing it into two parts.

6. - Elementary module according to claim 5, characterized in that the air gap (ε1) separating the two parts of the first windings (1; 2) is different from the air gap (ε2) separating the two parts of the second windings (3). 4).

 7. - elementary module according to claim 5 or 6, characterized in that said two parts are symmetrical.

 8. - Magnetic core of single-phase electrical transformer, characterized in that it consists of an elementary module according to one of claims 1 to 7.

 9. - Single-phase electrical transformer, comprising a magnetic core and primary and secondary windings, characterized in that the magnetic core is of the type according to one of claims 1 to 8.

 10. Magnetic core of three-phase electrical transformer, characterized in that it comprises:

 an internal magnetic sub-core composed of two elementary modules according to one of the claims 1 to 6 contiguous;

 and an external magnetic sub-core composed of two superposed additional windings (13, 17) arranged in this order around the internal magnetic sub-core:

^■ a first winding (13) made from a strip of material with low magnetic losses less than 20 W / kg sine wave frequency of 400 Hz for a maximum magnetic flux of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg, and with apparent magnetostriction at saturation (A _sat ) of less than or equal to 5 ppm, preferably less than or equal to 3 ppm, better still less than or equal to 1 ppm;

^■ a second winding (14) made from a strip of high saturation magnetization material (Js) greater than or equal to 1, T 5, preferably greater than or equal to 2.0 T, preferably greater than or equal at 2.2 T, and low magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 1 W / kg, preferably less than 10 W / kg ;

the section (Si ₃ ) of the first winding (13) of the outer magnetic sub-core and the section (Si ₄ ) of the second winding (14) of the outer magnetic sub-core being such that the ratio (Si ₄ / (Si ₃ + SH)) of the section of the high saturation magnetization material and the section of the set of two materials of the external magnetic sub-core is between 2 and 50%, preferably between 4 and 40% and the material section high saturation magnetization (Js) in the whole nucleus, in terms of section ratios, relative to the total of the sections of the two types of materials in the whole nucleus ( being

 ^ 4

between 2 and 50%, preferably between 4 and 40%.

1 1. Three-phase magnetic transformer magnetic core according to claim 10, characterized in that said first winding (13) of the external magnetic sub-core is in one material selected from Fe-75 82% Ni - 2 to 8% alloys (Mo, Cu, Cr, V), amorphous cobalt base alloys, and FeCuNbSiB nanocrystalline alloys.

 12. Magnetic core of three-phase electrical transformer according to claim 1 1, characterized in that said first winding (13) of the outer magnetic sub-core is a nanocrystalline alloy of composition:

 [ANialioo Fei-x-y-z-a-p ^ ^ M ^ UxSiyBzNbaM

 with a <0.3; 0.3 <x <3; 3 <y <1 7, 5 <z <20, 0 <a <6, 0 <β <7, 0 <γ <8, M 'being at least one of the elements V, Cr, Al and Zn, M "being at least one of the elements C, Ge, P, Ga, Sb, In and Be.

 13. Magnetic core of three-phase electrical transformer according to one of the claims

1 0 to 12, characterized in that said second winding (14) of the outer magnetic sub-core is made of a material selected from grain-oriented Fe-3% Si alloys, Fe-6.5% Si alloys, alloys Fe-1 5 to 50% in total Co, V, Ta, Cr, Si, Al, Mn, Mo, Ni, W textured or not, soft iron and steels and ferrous alloys made of at least 90% of Fe and having Hc <500 A / m, ferritic stainless Fe-Cr at 5 to 22% Cr, 0 to 10% in total of Mo, Mn, Nb, Si, Al, V and more than 60% Fe, unoriented Fe-Si-Al electric steels, Fe-Ni alloys at 40 to 60% Ni with at most 5% total additions of other elements, Fe-based magnetic amorphous at 5 to 25% total of B , C, Si, P and more than 60% Fe, 0 to 20% Ni + Co and 0 to 10% other elements.

 14. - Magnetic core according to one of claims 10 to 13, characterized in that it comprises a gap (1 7) dividing it into two parts.

 5. Magnetic core according to claim 1, characterized in that the gap (ε1) separating the two parts of the first windings (1; 2) from the inner magnetic sub-core and the two parts of the second winding (14). the external magnetic sub-core is different from the gap (ε2) between the two parts of the second windings (3; 4) of the inner magnetic sub-core and the two parts of the first winding (13) of the outer magnetic sub-core.

 6. A magnetic core according to claim 14 or 15 characterized in that the various air gaps (ε1, ε2) separating the two parts of the various windings (1, 2, 3, 4, 1 3, 14) are not all identical. between the inner magnetic sub-core and the outer magnetic sub-core.

7. Magnetic core according to one of claims 1 0 to 1 6, characterized in that the ratio between the section (S ₁₃ ) of the first winding (13) of the external magnetic sub-core and the section (S ₃ ; S ₄ ) of each of the second windings (3, 4) of the inner magnetic sub-core is between 0.8 and 1, 2.

8. Magnetic core according to one of claims 1 0 to 1 7, characterized in that the ratio between the section (S) of the second winding (14) of the external magnetic sub-core and the section (S!; S ₂ ) of each of the first windings (1, 2) of the inner magnetic sub-core is between 0.3 and 3.

9. Magnetic core according to one of claims 14 to 18, characterized in that said two parts are symmetrical.

20. Three-phase electrical transformer, comprising a magnetic core and primary and secondary windings, characterized in that the magnetic core is of the type according to one of claims 10 to 19.

 21. - A method of manufacturing a single-phase electrical transformer core according to claim 8, characterized in that it comprises the following steps:

 a magnetic metal support is manufactured in the form of a first winding (1) made of a first material, said first material being a crystalline material having a saturation magnetization (Js) of greater than or equal to 1.5 T, preferably greater than or equal to equal to 2.0 T better than or equal to 2.2 T and low magnetic losses of less than 20 W / kg at a frequency of 400 Hz in sinusoidal waves, for a maximum induction of 1 T;

a second winding (3) made of a material having, or being intended to have, after a nanocrystallization annealing, an apparent saturation magnetostriction (A _sat ) of less than or equal to 5 ppm, preferably less than or equal to 5 ppm, is wound on said metal support; equal to 3 ppm, better less than or equal to 1 ppm and magnetic losses less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg and 2 to 50% in proportion of section of high saturation magnetization material;

 - An annealing of nanocrystallization and contraction of said second winding (3) is carried out on said support;

 - And the two windings (1, 3) are secured, for example by shrinking, or by bonding, or by impregnation with a resin and polymerization of said resin.

 22. - A method of manufacturing a three-phase electrical transformer core according to claim 1 0, characterized in that it comprises the following steps:

 an internal magnetic sub-core composed of two elementary modules is produced, each elementary module being produced as follows:

 A magnetic metal support is produced in the form of a first winding (1; 2) made of a first material, said first material being a crystalline material with a high saturation magnetization (Js) greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better than or equal to 2.2 T and low magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T;

A second winding (3; 4) made of a material having, or being intended to have, after a nanocrystallization annealing, an apparent saturation magnetostriction (A _sat ) of less than or equal to 5 ppm, preferably a second winding (3; less than or equal to 3 ppm, more preferably less than or equal to 1 ppm and magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W / kg, preferably less than 10 W / kg, the ratio of the section of high saturation magnetization material (Js) to the total of the material sections of the materials. first (1; 2) and second (3; 4) windings being 2 to 50%, preferably 4 to 40%;

 Optionally annealing of nanocrystallization and contraction of said second winding (3; 4) on said support;

 said elementary modules are joined along one of their sides to form said internal magnetic sub-core;

 an external magnetic sub-core is produced as follows:

A third winding (13) formed around a strip of material having, or intended to have, after a nanocrystallization annealing, an apparent saturation magnetostriction (A _sat ) of less than or equal to at 5 ppm, preferably less than or equal to 3 ppm, better than or equal to 1 ppm and magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T, preferably less than 15 W kg, preferably less than 10 W / kg;

 · Nanocrystallization annealing and contraction of said third winding (13) is optionally performed on the inner magnetic sub-core;

A fourth winding (14) is placed around said third winding (13) in a saturation magnetization material (Js) greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better than or equal to at 2.2 T and low magnetic losses from less than 20 W / kg at 400Hz sine waves, for a maximum induction of 1 T, the ratio of the section of high saturation magnetization (Js) material to the total of the sections materials of the third (13) and fourth (14) windings being 2 to 50%, preferably 4 to 40%, and the proportion of high saturation magnetizing material (Js) throughout the core, in terms of section ratios, in relation to the total of the sections of the two types of materials, being between 2 and 50%, preferably between 4 and 40%;

 And said windings (1, 2, 3, 4, 13, 14) are secured, for example by shrinking, or by gluing, or by impregnation with a resin and polymerization of said resin.

 23.- Method according to claim 21 or 22, characterized in that said transformer magnetic core is cut to form two elementary cores, said elementary cores then being intended to be reassembled so as to define between them a gap (17) .

 24. The method of claim 23, characterized in that the two elementary nuclei are symmetrical.

25. The method of claim 23 or 24, characterized in that the surfaces of the elementary cores for defining the air gap (17) are shaped and surfaced before the elementary cores are reassembled.

26.- Process according to claim 25, characterized in that the shaping and surfacing is carried out so that the surfaces intended to define the air gap (17) separating the first windings (1; 2) of the two elementary cores define an air gap (ε1) different from the gap (ε2) between the second windings (3; 4) of the two elementary cores.

 27. The process as claimed in one of claims 23 to 25, characterized in that the two elementary cores are reassembled by shrinking by means of a crystalline material with saturation magnetization (Js) greater than or equal to 1.5 T, preferably greater than or equal to 2.0 T, better than or equal to 2.2 T and low magnetic losses of less than 20 W / kg in sinusoidal waves of frequency 400 Hz, for a maximum induction of 1 T.