EP3329027B1 - Feco- or fesi- or fe-alloy sheet or strip and its manufacturing process, transformer magnetic core made from it and transformer containing it - Google Patents

Description

The present invention relates to alloys of iron and cobalt, particularly those which have a content of the order of 10 to 35% Co, and also pure iron and alloys of iron and silicon which have a content of around 3% of Si. These materials are used to constitute magnetic parts such as transformer cores, in particular intended for aeronautics.

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

The constraints weighing on these transformers are manifold.

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

They must have a sufficient longevity (10 to 20 years minimum depending on the applications) to allow them to be profitable. Therefore, the thermal operating regime must be taken into account with regard to the aging of the transformer. Typically a minimum life of 100,000 hrs at 200 ° C is desired.

The transformer must operate on a power supply network with a roughly sinusoidal frequency, with an amplitude of the output rms voltage which may vary transiently up to 60% from one moment to another, and in particular when switching on. when the transformer is energized or when an electromagnetic actuator is suddenly engaged. This has the consequence, and by construction, a current inrush at the primary of the transformer through the nonlinear magnetization curve of the core. magnetic. The transformer elements (insulators and electronic components) must be able to withstand strong variations in this inrush current without damage, which is called the “inrush effect”.

This inrush effect can be quantified by an "inrush index" In which is calculated by the formula In = 2.Bt + Br - Bsat, where Bt is the nominal working induction of the magnetic core of the transformer, Bsat is the nucleus saturation induction and Br is its remanent induction

The noise emitted by the transformer due to electromagnetic forces and magnetostriction must be low enough to comply with current standards or to meet the requirements of users and personnel stationed near the transformer. More and more, pilots and co-pilots of aircraft want to be able to communicate no longer using headsets but by direct means.

The thermal efficiency of the transformer is also very important to consider, since it fixes both its internal operating temperature and the heat flows that must be removed, for example by means of an oil bath surrounding the windings and the cylinder head, associated with oil pumps sized accordingly. The sources of thermal power are mainly the losses by the Joule effect resulting from the primary and secondary windings, and the magnetic losses resulting from the variations of the magnetic flux over time and in the magnetic material. In industrial practice, the volumetric 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 the cost of materials, design, manufacture and maintenance, and optimization of the electrical power density (mass or volume). ) of the device by taking into account the thermal regime of the transformer.

In general, it is advantageous to seek the highest possible specific / volume power density. The criteria to be taken into consideration to assess it are mainly the saturation magnetization Js and the magnetic induction at 800 A / m B ₈₀₀ .

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

According to a first of these techniques (called “wound core”), the transformer comprises a wound magnetic circuit when the power supply is single phase. When the power supply is three-phase, the structure of the transformer core is made by two toroidal cores of the previous type placed side by side, and surrounded by a third toroid wound and forming an “eight” around the two previous toric cores. This form of circuit imposes in practice a small thickness of the magnetic sheet (typically 0.1mm). In fact, this technology is used only when the supply frequency forces, given the induced currents, to use bands of this thickness, that is to say typically for frequencies of a few hundred Hz.

According to the second of these techniques (called “cut-stacked core”), a stacked magnetic circuit is used, whatever the thicknesses of the magnetic sheets envisaged. This technology is therefore valid for any frequency lower than a few kHz. However, special care must be taken in deburring, juxtaposition, or even electrical insulation of the sheets, in order to both reduce parasitic air gaps (and therefore optimize the apparent power) and limit the currents induced between sheets.

• les alliages Fe-3% Si (les compositions des alliages sont, dans tout le texte données en % pondéraux) dont la fragilité et la résistivité électrique sont principalement contrôlées par la teneur en Si ; leurs pertes magnétiques sont assez faibles (alliages à grains non orientés N.O.) à faibles (alliages à grains orientés G.O.), leur aimantation à saturation Js est élevée (de l'ordre de 2T), leur coût est très modéré ; il existe deux sous-familles de Fe-3% Si utilisées soit pour une technologie de noyau de transformateur embarqué, soit pour une autre :
  • ∘ les Fe-3%Si à Grains Orientés (G.O.),utilisés pour les structures de transformateur embarqué de type « enroulé » : leur perméabilité élevée (B800 = 1.8 - 1.9 T) est liée à leur texture {110} <001> très prononcée ; ces alliages ont l'avantage d'être peu coûteux, faciles à mettre en forme, de grande perméabilité, mais leur saturation est limitée à 2 T, et ils présentent une non-linéarité très marquée de la courbe d'aimantation qui peut provoquer des harmoniques très importantes ;
  • ∘ les Fe-3%Si à grains Non Orientés (N.O.), utilisés pour les structures de transformateur embarqué de type « découpé-empilé » ; leur perméabilité est plus réduite, leur aimantation à saturation est similaire à celle des G.O ;
• les alliages Fe-48% Co-2% V, dont la fragilité et la résistivité électrique sont principalement contrôlées par le vanadium ; ils doivent leurs perméabilités magnétiques élevées non seulement à leurs caractéristiques physiques (K1 faible) mais aussi au refroidissement après recuit final qui règle K1 à une valeur très basse ; du fait de leur fragilité, ces alliages doivent être mis en forme à l'état écroui (par découpe, estampage, pliage...), et une fois seulement que la pièce possède sa forme finale (rotor ou stator de machine tournante, profile en E ou I de transformateur) le matériau est alors recuit en dernière étape ; de plus, à cause de la présence de V, la qualité de l'atmosphère de recuit doit être parfaitement contrôlée pour ne pas être oxydante ; enfin le prix de ce matériau, très élevé (20 à 50 fois celui du Fe-3% Si - G.O.), est lié à la présence de Co et est grossièrement proportionnel à la teneur en Co ; des alliages Fe-Co à plus basses teneurs en Co (typiquement 18 ou 27%) existent aussi ; ils ont l'avantage d'être moins chers que les précédents, comme ils contiennent moins de Co, tout en apportant une aimantation à saturation aussi bonne, voire dans certain cas encore un peu plus élevée, que celle de l'alliage FeCo48V2 précédent ; cependant leur perméabilité magnétique et leurs pertes magnétiques sont significativement plus élevées que celles des alliages FeCo équiatomiques.

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

• Fe-3% Si alloys (the alloy compositions are, throughout the text given in% by weight), the brittleness and electrical resistivity of which are mainly controlled by the Si content; their magnetic losses are quite low (non-oriented grain alloys NO) to low (oriented grain alloys GO), 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 one on-board transformer core technology or for another:
  • ∘ Oriented Grain Fe-3% Si (GO), used for on-board transformer structures of the "coiled" type: their high permeability (B800 = 1.8 - 1.9 T) is linked to their texture {110} <001> very pronounced; these alloys have the advantage of being inexpensive, easy to shape, of high permeability, but their saturation is limited to 2 T, and they exhibit a very marked non-linearity of the magnetization curve which can cause very important harmonics;
  • ∘ Fe-3% Si with Non Oriented grains (NO), used for on-board transformer structures of the “cut-stacked” type; their permeability is lower, 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 (low K1) but also to the cooling after final annealing which sets K1 to a very low value; due to their fragility, these alloys must be shaped in the work-hardened state (by cutting, stamping, bending, etc.), and only once the part has its final shape (rotor or stator of a rotating machine, profile in transformer E or I) 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 - GO), is linked to the presence of Co and is roughly proportional to the Co content; Fe-Co alloys with lower Co contents (typically 18 or 27%) also exist; they have the advantage of being less expensive than the previous ones, as they contain less Co, while providing a saturation magnetization as good, or even in certain cases a little higher, than that of the previous FeCo48V2 alloy; however, their magnetic permeability and magnetic losses are significantly higher than those of equiatomic FeCo alloys.

Only these two families of high permeability materials are currently used in on-board power transformers.

With the exception of the equiatomic FeCo alloy, high saturation materials (pure Fe, Fe-Si or Fe-Co at less than 40% Co) 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 sheets with less than 48% Co for on-board medium-frequency transformers, it has therefore been known for a long time that the chances of success necessarily depend on an acute texture characterized by the fact that in each grain, an axis <100> is very close to the direction of rolling. The so-called “Goss” texture {110} <001> obtained in Fe-Si by secondary recrystallization is a case in point. However, according to these bibliographical works the sheet should not contain cobalt.

More recently, we have shown in the document US-A-3,881,967 with additions of 4 to 6% Co and 1 to 1.5% Si, and also using recrystallization secondary, high permeabilities could also be obtained: B ₈₀₀ ≈ 1.98 T, i.e. a gain of 0.02 T /% Co at 800 A / m compared to the best current Fe 3% Si GO sheets (B ₁₀ ≈ 1.90 T). However, it is obvious that an increase of only 4% of the B ₈₀₀ is not sufficient to significantly lighten a transformer. By way of comparison, an Fe-48% Co-2% V alloy optimized for a transformer has a B ₈₀₀ of approximately 2.15 T ± 0.05 T, which allows an increase in magnetic flux to 800 A / m for a same cylinder head section of about 13% ± 3%, at 2500 A / m about 15%, at 5000 A / m about 16%.

It should also be noted the presence, in Fe 3% Si -GO, of large grains due to secondary recrystallization, and of a very slight disorientation between crystals allowing a B ₈₀₀ of 1.9 T, coupled with the presence of a magnetostriction coefficient λ ₁₀₀ very clearly greater than 0. This makes this material very sensitive to assembly and operating constraints, which brings back to industrial practice the B ₈₀₀ of a Fe 3% Si GO in operation in an on-board transformer to about 1.8 T. This is also the case for alloys of US-A-3,881,967 . In addition, Fe-48% Co-2% V has magnetostriction coefficients of amplitude still 4 to 5 times higher than Fe-3% Si, and a random distribution of crystallographic orientations as well as a small average size. grains (a few tens of microns), which makes it very sensitive to low stresses in particular, which lead to very strong variations in the magnetization characteristic J (H), and therefore also of B (H). These variations go in the direction of improvement when the stress is unidirectional and in tension, in the direction of degradation when the stress is unidirectional and in compression.

In operation, due to the increase in magnetization and induction at saturation, it must be considered that the replacement of an Fe 3% Si GO by an Fe-48% Co-2% V leads to an increase in the magnetic flux at constant section of the on-board transformer of the order of 20 to 25% for operating field amplitudes of 800 to 5000 A / m, i.e. approximately 0.5% increase in magnetic flux per% Co. The alloy of US-A-3,881,967 allows an increase of 1% of the magnetic flux per 1% of Co, but as we said, this total increase (4%) was considered much too low to justify the development of this material.

It has also been proposed, in particular in the document US-A-3,843,424 to use an Fe-5 alloy at 35% Co, comprising less than 2% Cr and less than 3% Si, and exhibiting 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 an Fe-3% Si-GO sheet operating at 1.8 T at 800 A / m, and at 1.95 T at 5000 A / m, to increase the magnetic flux in a given cylinder head section by 15% at 800 A / m and from 18% to 5000 A / m, and therefore to reduce the volume or the mass of the transformer accordingly. Thus, several compositions and methods of manufacturing Fe-low Co alloys have been proposed (with possible additions of alloying elements) generally making it possible to obtain magnetic inductions at 10 Oe close to those accessible with commercial alloys. Fe-48% Co-2% V but with significantly lower Co contents (and therefore cost prices) (18 to 25%).

JP 2001181803 A discloses for a sheet a maximum difference between the magnetostriction deformation amplitudes of less than 25 ppm for a recrystallized microstructure.

However, experience shows that all these materials, when obtained and processed by the usual methods, exhibit high magnetostrictions, at least in some of their directions (taking, for example, the rolling direction DL as a reference) . However, as the direction of the magnetic excitation can vary greatly from one place to another of the magnetic circuit and at the same time, this inhomogeneity of the magnetostriction according to the different directions can very well lead to the generation of a magnetostriction noise very significant, even if the magnetostriction in a given direction turns out to be weak.

In cut-stacked core technology it is not known that Fe-Ni alloys are used in aircraft transformers. Indeed these materials have a magnetization at saturation Js (1.6 T maximum for Fe-Ni50) much lower than Fe-Si (2 T) or Fe-Co (> 2.3 T) above and moreover have magnetostriction coefficients for FeNi50 of λ ₁₁₁ = 7 ppm and λ ₁₀₀ = 27 ppm. This results in an apparent magnetostriction at saturation λ _sat = 27 ppm for a polycrystalline Fe-Ni50 material of the “non-oriented” type (that is to say not having a pronounced texture). Such a level of magnetostriction is at the origin of a high noise, and this, added to a rather moderate Js saturation magnetization, explains why this material is not used.

In summary, the various problems with which the designers of aeronautical transformers are confronted can arise in this way.

In the absence of strong requirements on the 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 involving involved magnetic cores wound in Fe-Si GO, Fe-Co or iron-based amorphous, or solutions involving magnetic cores in pieces cut and stacked in Fe-Si NO or Fe-Co.

• soit réduire l'induction de travail, mais il faut alors augmenter dans le même rapport la section du noyau, donc son volume et sa masse pour conserver une même puissance transférée ;
• soit blinder acoustiquement le transformateur, d'où un surcoût et une augmentation de la masse et du volume du transformateur.

In the latter case, E or I cut-stacked cores are frequently used in FeSi NO or GO electric steel, or in FeCo alloys such as Fe49Co49V2. But since these materials have a significant magnetostriction and the magnetization direction does not always remain in the same crystallographic direction in an E-structure, these transformer structures deform a lot and emit a significant noise if their dimensioning is carried out with a level induction of usual labor (about 70% of Js). To reduce noise emission, you must:

• either reduce the work induction, but it is then necessary to increase in the same ratio the section of the nucleus, therefore its volume and its mass in order to maintain the same transferred power;
• or acoustically shield the transformer, resulting in an additional cost and an increase in the mass and volume of the transformer.

Under these conditions, it is far from always possible to design a transformer simultaneously meeting the weight and noise constraints of the specifications.

The requirements on a low magnetostriction noise being more and more widespread, it is not possible to meet them with the previous technologies other than by increasing the volume and the mass of the transformer, because we do not know how to reduce the noise other than by reducing the average work induction Bt, therefore by increasing the core section and the total mass to maintain the same magnetic work flux. _{B 1} should be lowered 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 its weight and bulk.

Only a material with zero magnetostriction would make it possible, at first sight, to solve the problem, and on condition of having a work induction greater than that of the current solutions. Only Fe-80% Ni alloys which exhibit a saturation induction Js of approximately 0.75 T and nanocrystallines whose Js is approximately 1.26 T exhibit such a low magnetostriction. But Fe-80% Ni alloys have too low a Bt work induction to provide lighter transformers than traditional transformers. Only nanocrystallines would allow this reduction in the case of a very low noise required. When the need for noise reduction is less important, nanocrystallines prove to be a relatively silent solution, but requiring too much weighting compared to the solution consisting of lowering the work induction in traditional solutions and / or padding the transformer.

However, nanocrystallines pose a major problem in the case of an “on-board transformer” solution: their thickness is around 20 μm and they are wound in a torus in the flexible amorphous state around a rigid support, so that the shape of the torus either retained throughout the heat treatment resulting in nanocrystallization. And this support cannot 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 using the technology of the previously wound circuit. described. Only resins for impregnating the wound toroid can maintain it in the same shape in the absence of the support which is removed after polymerization of the resin. But after a C cut of the impregnated and hardened nanocrystalline torus, a deformation of the C is observed which prevents the two parts from being placed exactly face to face to reconstitute the closed torus, once the windings have been inserted. The stresses of fixing C within the transformer can also lead to their deformation. It is therefore preferable to keep the support, which makes the transformer heavier. In addition, nanocrystallines have a magnetization at saturation Js which is markedly lower than other soft materials (Iron, FeSi3%, Fe-Ni50%, FeCo, amorphous iron base), which necessitates significantly increasing the transformer weight, since the increase of section of the magnetic core will have to compensate for the drop in work induction imposed by Js. Also the “nanocrystalline” solution would only be used as a last resort, if the maximum noise level required is low and if another lighter and quieter solution does not appear.

The aim of the invention is to provide a material for forming transformer cores exhibiting only a very low magnetostriction, including when they are subjected to a strong work induction which would make it possible not to use a magnetic core mass. too high, therefore to provide transformers having a high specific (or volume) power density. In this way, the transformers that they would make it possible to produce could advantageously be used in environments such as an aircraft cockpit where low magnetostriction noise would be advantageous for the comfort of the users.

To this end, the invention relates to a sheet or strip of cold-rolled and annealed ferrous alloy, characterized by the claims.

As will be understood, the invention is based on the use as a material intended to constitute magnetic parts, such as elements of a transformer core, of an iron-cobalt or iron-silicon or iron type alloy. -silicon-aluminum, on which well-defined thermal and mechanical treatments have been carried out, the heat treatments all being in the ferritic range of the alloy. The use of pure or very low alloyed iron is also envisaged.

Completely unexpectedly, and which the inventors cannot, for the moment, explain in a definite way, the result is a magnetostriction which, in the first place, is very low even in magnetic fields of high intensity which can range from up to 1.8 T. This result is surprising in particular for the case of FeCo type materials concerned by the invention, as FeCo alloys have long been known to usually exhibit high apparent magnetostriction.

But above all, which was particularly unexpected, this magnetostriction presents a remarkable isotropy, even for these high fields. It remains, in fact, almost zero both in the rolling direction of the sheet, in the transverse direction (perpendicular to the rolling direction) and in the direction forming an angle of 45 ° with these two directions, up to 'at an ambient magnetic field of at least 1 T. Beyond 1 T, the difference between the magnetostrictions observed in these three directions remains remarkably small up to a field of at least 1.8 T, or even 2 T.

Thus, transformers are obtained having a low magnetostriction noise in all directions of the sheets constituting their cores, therefore a particularly low overall magnetostriction noise, making them suitable for constituting, in particular, on-board transformers for aircraft which can be placed in the substation. piloting without interfering with direct conversations between its occupants.

• la figure 1 qui montre comment ont été prélevés et testés les échantillons de tôle qui ont été utilisés lors des essais décrivant l'invention et des essais de référence ;
• les figures 2, 3, 10, 11 et 12 qui montrent les courbes de magnétostriction, en fonction de l'intensité du champ magnétique selon diverses directions, d'échantillons d'un alliage FeCo27 obtenus par des procédés non conformes à l'invention ;
• les figures 4 à 9 qui montrent les courbes de magnétostriction, en fonction de l'intensité du champ magnétique selon diverses directions, d'échantillons d'un alliage FeCo27 obtenus par des procédés conformes à l'invention.

The invention will be better understood with the aid of the following description, given with reference to the following appended figures:

• the figure 1 which shows how the samples of sheet metal which were used in the tests describing the invention and the reference tests were taken and tested;
• the figures 2, 3 , 10 , 11 and 12 which show the magnetostriction curves, as a function of the intensity of the magnetic field in various directions, of samples of an FeCo27 alloy obtained by methods not in accordance with the invention;
• the figures 4 to 9 which show the magnetostriction curves, as a function of the intensity of the magnetic field in various directions, of samples of an FeCo27 alloy obtained by methods in accordance with the invention.

The metals and alloys to which the invention applies are iron and ferrous alloys with a ferritic structure, containing, in addition to iron and the impurities and residual elements resulting from their production, the following chemical elements. All percentages are weight percentages.

It should be understood that when we speak of "traces" to define the lower limit of a range of contents of a given element, it should be understood, as is usual for metallurgists, as meaning that the element in question is present at most only at a very low content, without influence on the properties of the material, but which cannot be said with certainty that it would always be strictly zero. Usually, a small amount of the item in question is detected by devices. analysis in the final alloy, because of its almost inevitable presence in some of the raw materials used or because of the pollution introduced during the production of the liquid metal. This pollution can be due, for example, to the wear of refractory materials, containing in particular magnesia and / or alumina and / or silica, which coat the receptacles (melting furnace, ladle, etc. .) where the liquid metal stays. The contact of the liquid metal with the atmosphere can also lead to the absorption of nitrogen, and also of oxygen which can combine with the most deoxidizing elements (Al, Si, Mn, Ti, Zr ...) to form non-metallic inclusions, some of which will remain in the final metal. The precision of the analysis apparatus for the detection and measurement of the content of the element in question is also to be taken into account. In general, it is considered that when an element is said to be able to be present in the form of “traces”, this includes all cases where its content is simply suffered, that it was not added voluntarily during the development, and that it is not necessary to maintain this content above a precise limit. In particular, if an element is not explicitly mentioned in the definition of the alloy used in the invention, it must be considered that its possible presence would be limited to “traces” as has just been defined.

• soit une limite supérieure du niveau d'impureté à ne pas dépasser, car au-delà de celle-ci, certaines propriétés de l'alliage seraient insuffisantes, et il faut alors veiller à ce que lesdites impuretés n'excèdent pas cette limite, en sélectionnant soigneusement les matières premières et/ou en évitant autant que possible la pollution du métal liquide lors de l'élaboration, et/ou en effectuant lors de l'élaboration des opérations spécifiquement destinées à abaisser la teneur de l'impureté lorsque cela est nécessaire et possible (désulfuration, déphosphoration...) ;
• soit une limite supérieure qui correspond à celle d'un ajout volontaire de l'élément en question pour conférer à l'alliage final des propriétés avantageuses, cet ajout étant donc facultatif.

For the elements which are said to be present at a content between “traces” and a defined upper limit, this means that said limit is:

• either an upper limit of the level of impurity not to be exceeded, because beyond this, certain properties of the alloy would be insufficient, and it is then necessary to ensure that said impurities do not exceed this limit, in carefully selecting the raw materials and / or avoiding as much as possible the pollution of the liquid metal during the preparation, and / or carrying out during the preparation operations specifically intended to lower the content of the impurity when necessary and possible (desulfurization, dephosphorization, etc.);
• or an upper limit which corresponds to that of a voluntary addition of the element in question to confer advantageous properties on the final alloy, this addition therefore being optional.

In the tables showing the composition of the different alloys tested, it should be understood that when a content is noted as being "less than ...", this amounts to saying that the element in question is only present in the form of traces in the sense seen above, in that the analysis devices are not able to determine very reliably whether the element would really be completely absent, or if it would be present but at a level below the given low limit In the picture.

The alloys making up the sheets or strips according to the invention contain C at a content between traces resulting from the production, without C having been added to the raw materials, and 0.2%, preferably between traces and 0.05%, better between traces and 0.015%.

The FeCo27 and FeSi3 type alloys to which certain possible variants of the invention fall typically have C contents of 0.005 to 0.15%, which result much more from the conditions of deoxidation of the liquid metal (in particular from the formation of CO within liquid metal during vacuum passages) than a deliberate desire to find these C contents in the final product for reasons related to the mechanical or magnetic properties of the alloy.

In fact, it is not desirable to find a very significant C content in the final alloy used in the invention, because beyond a threshold which may be between 0.05 and 0.2%, there is a risk of observing a precipitation of carbides which would tend to degrade the magnetic properties, a content of more than 0.2% being unacceptable in all cases for this reason. Also, it is known that above 0.01% of C, it is possible that we observe an aging precipitation of clusters or clusters of C, after the transformer has been operated, for months or years, above room temperature. Magnetic properties (magnetic losses, permeability, etc.) can be affected. For these reasons, it will be preferable to keep the C content within the aforementioned optimum limits.

They contain between traces and 40% of Co. The maximum of 40% is determined by the desire not to have a too rapid and acute order-disorder transition during heat treatments. This would prevent multiple annealing after hot rolling, and it will be seen that two anneals, preferably three, preceding or following a cold rolling are necessary to practice the invention. Performing more cold rollings with their corresponding intermediate anneals is also possible when it is desired to obtain particularly thin strips suitable for use in a coiled-type core transformer.

The Co can be present in limited quantity, only in the state of traces resulting from the production, therefore not to be added voluntarily, but if Co <35% it is necessary to Si + 0.6% Al ≤ 4.5 - 0 , 1% Co and also Si ≤ 3.5%. Thus, for example, in the absence of cobalt, a content of traces at 3.5% of Si, and of traces at 1% of Al is required to remain within the scope of the invention. We are then in the case of an alloy of the class of iron-silicon or iron-silicon-aluminum alloys, or even a pure or very low-alloyed iron, to which the invention can also be applied.

In the case of an iron-cobalt alloy proper (which would therefore contain less than 3.5% Si), a Co content of 10 to 35% is preferred.

The invention is most typically applicable to Fe-Co alloys of a conventional type containing approximately 27% Co and to Fe-Si alloys containing approximately 3% Si.

• de traces à 1,0% si la teneur en Co est d'au moins 35% ;
• si la teneur en Co est de moins de 35% : Si + 0,6 %Al ≤ 3,5 - 0,1 %Co.

The alloy to which the invention is applied contains an Si content which is:

• 1.0% traces if the Co content is at least 35%;
• if the Co content is less than 35%: Si + 0.6% Al ≤ 3.5 - 0.1% Co.

However, an Si content + 0.6% Al ≤ 4.5 - 0.1% Co can be accepted if the rolling operations are carried out not strictly cold, but "lukewarm", that is to say at a temperature ranging up to 350 ° C., this rolling temperature preferably being obtained by steaming, that is to say heating in a static chamber at a low temperature. This lukewarm rolling (which it is agreed that it is fully comparable to cold rolling in the context of the invention; the term "cold rolling", when no further details are given on the temperature of its execution, should be understood, in the present text, to also include lukewarm rolling carried out up to 350 ° C. It is employed as opposed to the "hot" rolling mills well known to metallurgists, which are carried out at temperatures considerably higher. high, of several hundred degrees, or even 1000 ° C or more) allows, compared to a rolling carried out at room temperature or at a temperature close to this one, that the material rolls better, is more ductile in deformation and less risk of cracking during rolling. Static heating, in an oven, of the hot-rolled and coiled strip or of the hot-rolled sheet makes it possible to leave the coil or the sheet at the target temperature for a few hours for a few hours, so that the temperature becomes homogeneous throughout the material. before lukewarm rolling is performed. A parade annealing furnace is less suitable than an oven for this purpose, as it is generally not sized to operate at such low temperatures. This baking can be carried out in air, the maximum desirable temperature generally not being high enough to cause strong oxidation of the surface of the strip or of the sheet, at which the annealing (s) in a hydrogenated atmosphere which will follow will not follow. could not remedy.

The reheating temperature is also to be determined as a function of the cooling that the strip or sheet will undergo, predictably, during its transfer between the reheating installation and the rolling mill. The reheating temperature must be sufficient so that the actual temperature of the strip or sheet at the time of warm rolling is that targeted, but it must not exceed 400 ° C to avoid significant oxidation of the material during reheating, and even also during transfer to the rolling mill.

Of course, the use of a neutral or reducing atmosphere during steaming, or reheating in general, is not excluded.

The limitation of the Si content, linked to the Al content, taking into account the Co content, is due to the concern to keep the material good cold rolling ability, or at a temperature significantly above ambient but nevertheless not very high (case of lukewarm rolling up to 350 ° C, see above).

The Si content is also governed by the desire to permanently retain a ferritic structure during the manufacture of the material, which proves to be important for obtaining the low and isotropic magnetostriction on which the invention is based.

The inventors believe that it might be possible that an explanation for the remarkable isotropy of the magnetostriction of the sheets according to the invention lies in the fact that during heat treatments and cold rolling, the "filiation" or the The texture “heredity” is total, so it is imperative to remain constantly in the ferritic domain.

By using the terms of “filiation” or “inheritance” of texture, one refers to the phenomena which naturally lead to a progressive transformation of the texture of the material during metallurgical operations. In the case of the invention, it turns out that it could be important for this transformation not to be disturbed by phase changes which would occur during processing, so as to keep a “memory” of the initial texture of hot rolling in the material. This is what motivates the inventors' desire to ensure that all the treatments take place entirely in the ferritic field of the alloy. It is nevertheless surprising, from a theoretical point of view, that this texture filiation seems to be important for obtaining the low magnetostriction and the magnetostriction isotropy which characterize the invention, even though the process according to the invention does not lead to , at most, at a low texturing of the material, as will be seen in the examples.

The Cr content can range from traces to 10%. An addition of Cr only slightly modifies the Fe stacking fault energy, and therefore does not greatly modify the texture filiations during the treatments carried out according to the invention. It lowers the magnetization to saturation J _sat , and it is undesirable to add an amount exceeding 10% for this reason. On the other hand, just like Si, it appreciably increases the electrical resistivity, therefore advantageously decreases the magnetic losses. Cooling of the transformer allows, however, to tolerate more magnetic losses, and a low Cr content, even in the state of traces, may be acceptable in this case.

The total of the contents of V, W, Mo and Ni is between traces and 4%, preferably between traces and 2%. These elements increase the electrical resistivity, but they lower the saturation magnetization, which is generally not desired.

The Mn content is between traces and 4%, preferably between traces and 2%. The reason for this relatively low maximum content is that Mn reduces the saturation magnetization which is one of the major contributions of FeCo. Mn only slightly increases the electrical resistivity. Above all, it is a gammagenic element, which therefore reduces the temperature range allowing ferritic annealing. We have seen that for questions linked to the inheritance of ferritic microstructures, it was not desirable to leave the ferritic range during the treatments, and an excessive presence of Mn would increase the risks of such an exit. The Al content is between traces and 3%, preferably between traces and 1%. Al reduces saturation magnetization and is much less effective than Si or Cr at increasing electrical resistivity. But Al can be used to extend the cold-rollability range of high alloy FeCo grades when reaching the limits of silicon additions, as previously discussed.

The S content is between traces and 0.005%. Indeed, S tends to form sulphides with manganese, and oxysulphides with Ca and Mg, which greatly degrades the magnetic performance and in particular the magnetic losses.

The P content is between traces and 0.007%. In fact, P can form phosphides of metallic elements which are harmful to the magnetic properties and to the development of the microstructure.

The Ni content is between traces and 3%, and preferably less than 0.5%. Indeed, Ni does not increase the electrical resistivity, reduces the saturation magnetization and therefore degrades the power density and the electrical efficiency of the transformer. Its addition is therefore not necessary.

The Cu content is between traces and 0.5%, preferably less than 0.05%. Cu is very poorly miscible with Fe, Fe-Si or Fe-Co, and therefore forms copper-rich, non-magnetic phases, significantly degrading the magnetic performance of the material as well as greatly hindering the development of its microstructure.

The contents of Nb and Zr are each between traces and 0.1%, preferably less than 0.01% since Nb and Zr are well known to be potent inhibitors of grain growth, and therefore will interfere strongly and unfavorably. with the metallurgical mechanism of texture filiation which one suspects to be at the origin of the good results obtained thanks to the invention.

The Ti content is between traces and 0.2% in order to limit the harmful formation of nitrides, which would significantly degrade the magnetic properties (increase in losses) and could interfere with the texture transformation mechanisms during rolling-annealing. .

The N content is between traces and 0.01%, again to avoid excessive formation of nitrides of all kinds.

The Ca content is between traces and 0.01% to avoid the formation of oxides and oxysulphides which would be harmful for the same reasons as Ti nitrides.

The Mg content is between traces and 0.01% for the same reasons as Ca.

The Ta content is between traces and 0.01% because it can greatly hinder the growth of the grain.

The B content is between traces and 0.005% to avoid the formation of boron nitrides which would have the same effects as Ti nitrides.

The O content is between traces and 0.01% to prevent oxidized inclusions formed in too large quantities from having the same harmful effects as nitrides.

These maximum contents for S, P, Ni, Cu, Nb, Zr Ti, N, Ca, Mg, Ta, B, O often correspond to simple impurities resulting from the production of the alloy, and are common in alloys Fe-Co and Fe-Si of the types concerned by the invention. If necessary, a rigorous choice of raw materials and careful processing allow them to be achieved.

Regarding the manufacturing process which leads to the products according to the invention, it is as follows.

An ingot or a continuously cast semi-finished product is prepared, having the composition described above. For this purpose, all production and casting methods making it possible to obtain this composition can be used. In the case where the aim is to obtain an ingot, processes such as arc melting processes under slag, induction melting under slag or under vacuum (VIM for Vacuum Induction Melting) are recommended. They are preferably followed by remelting processes to obtain a secondary ingot. In particular, ESR (Electroslag Remelting) or VAR (Vacuum Arc Remelting) type processes are particularly suitable for obtaining alloys with a purity optimum and low fractions of precipitates for the preferred applications of the invention.

In the most general case of obtaining an ingot of non-parallelepipedal shape, a first hot forming by forging or rolling (blooming) to give it this parallelepipedal shape is conventionally carried out. An ingot is thus obtained which often has a thickness of the order of 10 cm.

The ingot optionally shaped beforehand, or the continuous casting semi-finished product, is hot-rolled, in the usual way, until a sheet or strip with a thickness of 2 to 5 mm is obtained, preferably between 2 and 3.5 mm, for example of a thickness of the order of 2.5 mm. This hot rolling therefore constitutes the last step (or the only one) in the hot forming of the process according to the invention.

Then, preferably, an annealing is carried out, static or in scrolling, of said sheet or strip, in the ferritic range, therefore at a temperature between 650, preferably 700 ° C, and a temperature which guarantees that no will not leave the purely ferritic range and which therefore depends on the composition of the alloy, for 1 minute to 10 hours.

If the Si content is greater than or equal to a limit noted (% Si) _α-lim which depends on the Co and C contents, then the heat _{treatment temperature T tth} of this annealing can go up to 1400 ° C.

This limit is (% Si) _α-lim = 1.92 + 0.07% Co + 58% C

If the Si content is less than (% Si) _α-lim then the temperature T _tth of heat treatment of this annealing is such that T _tth <T _α-lim upper limit temperature for the presence of the ferrite, with $${\mathrm{T}}_{\mathrm{\alpha }\mathrm{-lim}}={\mathrm{<figure-callout\; id="0"\; label="T"\; filenames="imgf0002.png,imgf0003.png"\; state="\{\{state\}\}">T</figure-callout>}}_0+\mathrm{k}\%\mathrm{Yes}\mathrm{or}{\mathrm{<figure-callout\; id="0"\; label="T"\; filenames="imgf0002.png,imgf0003.png"\; state="\{\{state\}\}">T</figure-callout>}}_0=900+2\%\mathrm{Co}-2833\%\mathrm{VS}\mathrm{and}\mathrm{k}=112-1250\%\mathrm{VS}$$

These conditions result from a study carried out by the inventors on the phase diagrams of Fe-Co alloys comprising various other alloy elements.

This annealing should be carried out in a dry hydrogenated atmosphere. The atmosphere should contain between 5% and ideally 100% hydrogen, the remainder being one or more neutral gases such as argon or nitrogen. Such an atmosphere can result from the use of cracked ammonia. A maximum content of 1% in total of oxidizing gaseous species for the alloy (oxygen, CO ₂ , water vapor, etc.) may be present, preferably less than 100 ppm. The dew point of the atmosphere is at most + 20 ° C, preferably at most 0 ° C, better at maximum -40 ° C, optimally at maximum -60 ° C.

• d'empêcher une oxydation de la surface de la tôle ou bande et des joints de grains ; une telle oxydation des joints de grains est très défavorable à la filiation de la texture, et s'il se confirmait que l'une des raisons du succès de l'invention était cette très bonne filiation de texture lors des traitements thermiques et des laminages à froid, on tiendrait là une condition importante pour la mise en œuvre de l'invention ;
• d'assurer une bonne transmission de la chaleur lors du recuit, notamment si celui-ci est effectué au défilé ; H₂ est de loin le gaz le plus caloporteur, et il permet d'obtenir des bandes laminables sans risques de casse en sortie de recuit, par évitement d'une mise en ordre fragilisante, grâce à une efficace extraction de la chaleur de la bande recuite dans la zone de mise en ordre (soit entre 500 et 700°C).

This hydrogenated atmosphere, therefore reducing, has the following functions, compared to an atmosphere which would be simply neutral, a fortiori oxidizing:

• to prevent oxidation of the surface of the sheet or strip and of the grain boundaries; such an oxidation of the grain boundaries is very unfavorable to the texture filiation, and if it is confirmed that one of the reasons for the success of the invention was this very good texture filiation during heat treatments and cold, this would be an important condition for the implementation of the invention;
• to ensure good heat transmission during annealing, in particular if this is carried out in the process; H ₂ is by far the most heat-transfer gas, and it makes it possible to obtain laminable bands without risk of breakage at the end of annealing, by avoiding a weakening ordering, thanks to efficient extraction of heat from the band annealed in the ordering zone (ie between 500 and 700 ° C).

After this optional but preferred annealing, a natural or forced cooling of the sheet or strip is carried out, under conditions which prevent excessive weakening of the strip. For a Co content of more than 20%, this cooling rate must be at least 1000 ° C / h. For a Co content of 20% and less, therefore including the case of FeSi alloys of the types concerned by the invention, it is not necessary to set a minimum cooling rate.

Is carried out (either after the optional annealing above, or after the hot rolling) then a first cold rolling at a reduction rate of 50 to 80%, preferably 60 to 75%, and at a temperature between ambient temperature (eg 20 ° C) and 350 ° C. The upper limit of 350 ° C. corresponds to the case where, as we have seen, a “warm” rolling is carried out, the reheating being preferably carried out by a baking, for the alloys relatively rich in Si. Generally, the temperature of cold rolling is between room temperature and 100 ° C.

Too low a reduction rate (less than 50%) during at least one of the cold or “warm” rolling operations does not make it possible, as will be seen, to obtain the low and isotropic magnetostriction sought. Too high a reduction rate (greater than 80%) would be liable to modify the texture of the material too strongly, so that the magnetostriction will be degraded.

An annealing is then carried out, static or by scrolling, in the ferritic range, at a bearing temperature of between 650 and 930 ° C, preferably between 800 and 900 ° C, and for 1 min to 24 hours, preferably 2 min at 1 h, in a dry hydrogenated atmosphere (partial or total) as defined above, for the reasons seen in connection with the optional annealing following the hot rolling, followed by cooling to perform under conditions similar to those described for the optional annealing and for the same reasons.

A second cold rolling is then carried out, the characteristics of which are located in the same ranges as those already described for the first cold rolling.

Finally, a final recrystallization annealing, static or in flow, is carried out under a preferably hydrogenated atmosphere (partial or total) like the atmospheres of the previous anneals. But this final annealing can also be carried out under vacuum, under neutral gas (argon for example) or even in air, in the ferritic range, at a temperature of 650 to [900 + (2 x% Co)] ° C, for a period of 1 minute to 48 hours. A hydrogenated atmosphere is no longer necessarily necessary for this last annealing, because at this stage the metal may have already reached its final dimensions, in particular in thickness, or even also with regard to its perimeter, in particular if a cutting has already had place to give the pieces of the future stack their final shapes and dimensions. In this case, even if an absence of hydrogen led to an embrittlement of the metal during this recrystallization annealing, this would be without consequences if it only remained to stack the pieces to form the core.

If the final annealing is prolonged too long, there is a risk of already obtaining at 900-930 ° C for an Fe-Co alloy a hollowing of the grain boundaries on the surface of the material which will degrade the magnetic losses, and also an oxidation at the grain boundaries. , even if a reducing and dry atmosphere is used, which will have the same effect. Under these conditions, there would be a degradation of the magnetic losses, and the low and isotropic magnetostriction targeted by the invention would also be degraded. A final recrystallization rate of 100% is preferred but is not mandatory, since it will be seen in the examples that recrystallization rates of 90% may already be sufficient to obtain satisfactory results in terms of weakness and isotropy of the magnetostriction. . The minimum necessary recrystallization rate is estimated at 80%.

The precise conditions for carrying out this final annealing making it possible to achieve such recrystallization, for a material of given composition and thickness, can be determined experimentally by those skilled in the art by means of routine tests. Static annealing, whose temperature rise rate is lower than for step annealing and which lasts longer, has the advantage of increasing the ferritic grain better than step annealing, which is favorable to the process. obtaining low magnetic losses.

This final annealing ends with relatively slow cooling such as natural cooling in air, or cooling under a hood or other device limiting heat loss by radiation. A speed less than or equal to 2000 ° C / h, preferably less than or equal to 600 ° C / h is typically recommended. A faster cooling would be likely to introduce internal stresses by establishing a thermal gradient in the material, which would degrade the magnetic losses.

These conditions guaranteeing sufficiently slow cooling are most easily fulfilled, in particular when the final annealing is a static annealing, that is to say carried out in a closed vessel, and the material is quite simply left in the treatment chamber during its cooling.

Coolings following annealing other than final annealing need not be particularly advantageous at low speed. Too slow cooling would even risk reducing the laminability of the material in the following step.

This relatively slow cooling is coupled with a rate of temperature rise for the purpose of annealing which is itself less than or equal to 2000 ° C./h, better still less than or equal to 600 ° C./h.

In addition, in general, the inventors believe that in order not to obtain an excessively marked texture, of Goss or other, and a good texture lineage, the rate of temperature rise for the final annealing and the rate of the cooling which follows this. final annealing are among the parameters on which one can play to achieve the desired objectives in terms of low and isotropic magnetostiction of the alloys used in the invention, in addition to the composition of the alloy and the conditions of its heat and thermomechanical treatments during cold or warm rolling and annealing.

The inventors obtain on the final product no more than 30% of Goss texture component or of {111} <110> texture component (these are the orientations which are found to be most present in the sheets and strips according to the invention. ) and, in general, not more than 30% of any marked {hkl} <uvw> texture component, that is to say a component characterized by the fact that at most 30% volume fraction of the grains of the material have orientation {hkl} <uvw> less than 15 ° in disorientation of a specific orientation {h ₀ k ₀ l ₀ } <u ₀ v ₀ w ₀ > ..

After the final recrystallization annealing which makes it possible to obtain the final magnetic properties of the material, an additional oxidation annealing of the material can be added, at a temperature between 400 and 700 ° C, preferably between 400 and 550 ° C, allowing strong but superficial oxidation of the material on at least one of its faces, without risking intergranular oxidation since this is known to occur at higher temperatures. This oxidation layer has a thickness of 0.5 to 10 µm and guarantees electrical insulation between the stacked parts of the transformer magnetic core, which makes it possible to substantially reduce the induced currents and therefore the magnetic losses of the transformer. The precise conditions for obtaining this oxidation layer can easily be determined by a person skilled in the art using conventional experiments, depending on the precise composition of the material and the oxidizing power of the treatment atmosphere chosen. (air, pure oxygen, oxygen-neutral gas mixture ...) vis-à-vis this material. Conventional analyzes of the composition of the oxidation layer and of its thickness make it possible to determine for which conditions of treatment of a given material (temperature, time, atmosphere) the desired oxidation layer can be obtained.

A manufacturing process has been described comprising two stages of cold rolling and two or three anneals. But it would still be in accordance with the invention to carry out more cold rolling steps similar to those which have been described, which can be separated by intermediate anneals similar to the first of the compulsory anneals which have been described.

It should be understood that each of the cold rolling with a reduction rate of 50 to 80%, preferably 60 to 75%, which has been mentioned, can be carried out gradually, in several successive passes not separated by an intermediate annealing.

The end result is a cold rolled and annealed sheet or strip whose thickness is typically 0.05 to 0.3 mm, preferably at most 0.25 mm, better at most 0.22 mm to limit magnetic losses, which has the particularity of exhibiting very low magnetostrictions λ in the three directions DL (rolling direction), DT (transverse direction) and 45 ° (middle direction between DL and DT), measured both parallel and perpendicular to the direction of the applied field, and above all a very small difference between the highest and lowest magnetostrictions of those measured, and this for different inductions from 1.2 T to 1.8 T. These inductions are those for which it is It is often desirable to operate on-board aeronautical transformers using Fe-Co or Fe-Si cores to obtain, in addition to the low magnetostriction and low inrush effect, a transformer mass as small as possible. 1.8 T, in particular, is an interesting induction to obtain a transformer as light and little noisy as possible.

It is well understood that to obtain a low magnetostriction noise of the transformer, it would hardly be useful to obtain a weak magnetostriction only in one or a few direction (s) which one would define with respect to the rolling direction and to the direction of the rolling. field, maintaining a relatively strong magnetostriction in the other directions. The user satisfaction criterion is therefore taken as the maximum deviation “Max Δλ” between the magnetostriction amplitudes observed during the measurements carried out on three types of sample coming from the same material and represented on the figure 1 . The following examples will be based on this evaluation method.

• Type 1 : échantillons 2 rectangulaires allongés (par exemple 120x15 mm) découpés de telle sorte que la direction LONG de l'échantillon 2 soit parallèle à DL. Le champ magnétique Ha sera appliqué durant la mesure de déformation, par une bobine d'excitation de même axe que la direction LONG de l'échantillon 2, donc également selon la direction LONG de l'échantillon 2. Les mesures de déformation ε, nommées λ^H//DL, sont effectuées aussi bien selon la direction du champ (λ^H//DL _ε//H), que perpendiculairement à celle-ci (λ^H//DL _ε⊥H) et il en résulte donc deux valeurs de magnétostriction pour l'échantillon 2 de type 1.
• Type 2 : échantillons 3 rectangulaires allongés (par exemple 120x15 mm) découpés de telle sorte que la direction LONG de l'échantillon 3 soit parallèle à l'axe situé à 45° de DL et DT. Le champ magnétique Ha sera appliqué durant la mesure de déformation, par une bobine d'excitation de même axe que la direction LONG de l'échantillon 3, également selon la direction LONG de l'échantillon 3. Les mesures de déformation, nommées λ^H//45°, sont effectuées aussi bien selon la direction du champ (λ^H//45° _ε//H), que perpendiculairement à celle-ci (λ^H//45° _ε⊥H) et il en résulte donc deux valeurs de magnétostriction pour l'échantillon 3 de type 2.
• Type 3 : échantillons 4 rectangulaires allongés (par exemple 120x15 mm) découpés de telle sorte que la direction LONG de l'échantillon 4 soit parallèle à DT. Le champ magnétique Ha sera appliqué durant la mesure de déformation, par une bobine d'excitation de même axe que la direction LONG de l'échantillon 4, également selon la direction LONG de l'échantillon 4. Les mesures de déformation, nommées λ^H//DL, sont effectuées aussi bien selon la direction du champ (λ^H//DT _ε//H), que perpendiculairement à celle-ci (λ^H//DT _ε⊥H) et il en résulte donc deux valeurs de magnétostriction pour l'échantillon 4 de type 3.

These samples are taken from a strip 1 prepared according to the invention or according to a reference method, depending on the example. Its rolling direction DL, its direction through DT and its middle direction 45 ° are represented by arrows. Three types of samples are taken from the sheet 1 for carrying out the magnetostriction tests.

• Type 1: elongated rectangular samples 2 (eg 120x15 mm) cut so that the LONG direction of sample 2 is parallel to DL. The magnetic field Ha will be applied during the deformation measurement, by an excitation coil having the same axis as the LONG direction of sample 2, therefore also along the LONG direction of sample 2. The deformation measurements ε, named λ ^{H // DL} , are performed both according to the direction of the field (λ ^{H // DL} _{ε // H} ), and perpendicular to it (λ ^{H // DL} _ε⊥H ) and it therefore results in two values magnetostriction for sample 2 type 1.
• Type 2: elongated rectangular samples 3 (for example 120x15 mm) cut so that the LONG direction of sample 3 is parallel to the axis located at 45 ° from DL and DT. The magnetic field Ha will be applied during the deformation measurement, by an excitation coil having the same axis as the LONG direction of sample 3, also according to the LONG direction of sample 3. The deformation measurements, called λ ^{H // 45 °} , are carried out as well according to the direction of the field (λ ^{H // 45 °} _{ε // H} ), as perpendicular to this one (λ ^{H // 45 °} _ε⊥H ) and it therefore results in two magnetostriction values for sample 3 type 2.
• Type 3: elongated rectangular samples 4 (for example 120x15 mm) cut so that the LONG direction of sample 4 is parallel to DT. The magnetic field Ha will be applied during the deformation measurement, by an excitation coil having the same axis as the LONG direction of sample 4, also according to the LONG direction of sample 4. The deformation measurements, called λ ^{H // DL} , are performed both according to the direction of the field (λ ^{H // DT} _{ε // H} ), and perpendicular to this (λ ^{H // DT} _ε⊥H ) and this therefore results in two magnetostriction values for sample 4 of type 3.

In total therefore six different strain measurements are measured at each induction level B (measured) of each of the three types of sample. To know the magnetoctrictive behavior of the material, not only three directions (types) of sampling are used (DL, DT and the direction making an angle of 45 ° with DI and DT) but also several levels of induction such as by example 1T, 1.5T, 1.8T.

The quantity Max Δλ, measured for an induction amplitude B in the material and which can also be noted Max Δλ (B), is representative of the isotropy of magnetostriction. It is therefore calculated by taking into account the highest value and the lowest value among these six values of λ measured on samples 2, 3, 4 from the same strip 1 of material as indicated on the figure 1 . This consideration is the highest value that can be found among the six absolute values of the algebraic differences between each possible pair of magnetostriction measurements described above. In other words : $$\mathrm{Max}\mathrm{\Delta \lambda }\left(\mathrm{B}\right)=\underset{\mathrm{i},\mathrm{j}=\mathrm{DL},45°\mathrm{Where}\mathrm{DT}}{\mathrm{Max}\left|{\mathrm{\lambda }}_{\mathrm{\epsilon }//\mathrm{H}}^{\mathrm{H}//\mathrm{i}}\left(\mathrm{B}\right)-{\mathrm{\lambda }}_{\mathrm{\epsilon }\perp \mathrm{H}}^{\mathrm{H\; //\; j}}\left(\mathrm{B}\right)\right|}$$

For a sheet or a strip to be declared in accordance with the invention, it is agreed that the maximum value Max Δλ measured for an induction of 1.8 T must be at most 25 ppm.

The ten tests which will be described were carried out in particular on samples of an FeCo27 type alloy, the detailed compositions of which will be given. However, it will be seen that the invention would be applicable in a completely comparable manner to all the alloys falling within this category which is known per se and which is commonly used in transformer cores, without however the advantage of very texturing. weak but not zero which will be described, with the means to obtain it, has so far been identified. Table 1 shows the compositions of various alloys according to the invention and of reference alloys, used during the tests.

In particular, two FeCo27 alloys obtained from different castings, but of very similar compositions so that the test results are directly comparable, were tested. Alloy A was used for reference tests 1 and 2, alloy B was used for tests according to the invention 3 to 9 and for reference tests 10 to 12. Table 1: Compositions of the alloys of the tests Element (%) A Invention B Invention C Invention D Invention E Invention F Invention G Invention H Invention I invention J Reference K Reference L Reference M Invention N Invention VS 0.010 0.009 0.007 0.023 0.012 0.013 0.011 0.012 0.010 0.008 0.009 0.009 0.012 0.015 Mn 0.261 0.256 0.195 0.234 0.248 0.421 0.532 0.810 0.167 0.208 0.520 0.289 0.368 <0.010 Yes 0.142 0.153 0.330 0.720 0.031 2,730 0.070 0.013 3.020 0.023 3.07 1.53 0.640 0.083 S 0.0023 0.0042 0.0033 0.0021 0.0048 0.0008 0.0006 0.0028 0.0005 0.0015 0.0007 0.0044 0.0008 <0.0005 P 0.0025 0.0055 0.0031 0.0029 0.0029 0.0032 0.0047 0.0037 0.0053 0.0031 0.0043 0.0049 0.0041 <0.0005 Or 0.030 0.030 0.100 <0.01 0.130 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.080 <0.01 Cr 0.514 0.498 1.00 0.200 0.011 0.008 0.048 6.06 0.047 0.089 0.007 0.038 0.072 <0.01 Mo <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 0.170 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cu 0.009 0.010 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Co 27.09 27.32 18.35 10.07 4.21 0.020 <0.01 27.11 <0.01 49.0 18.20 38.15 38.82 15.10 V 0.01 0.01 <0.005 0.51 <0.005 <0.005 <0.005 <0.005 <0.005 2.03 <0.005 <0.005 <0.005 <0.005 Al <0.001 <0.001 0.14 <0.001 <0.001 0.60 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Nb <0.001 <0.001 <0.001 <0.001 0.005 <0.001 <0.001 <0.001 <0.001 0.040 <0.001 <0.001 <0.001 <0.001 Ti <0.001 <0.001 <0.001 0.080 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 NOT 0.0015 0.0044 0.0023 0.0036 0.0043 0.0027 0.0041 0.0045 0.0048 0.0018 0.0021 0.0019 0.0027 0.0012 That <0.0003 <0.0003 <0.0003 0.0013 <0.0003 <0.0003 0.0009 <0.0003 <0.0003 0.0007 0.0015 <0.0003 0.0009 <0.0003 Mg <0.0002 <0.0002 0.0006 <0.0002 <0.0002 0.0005 0.0004 <0.0002 <0.0002 0.0004 <0.0002 0.0004 <0.0004 <0.0002 Your <0.002 <0.002 0.0025 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 B <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 0.0007 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 W <0.005 <0.005 <0.005 <0.005 0.28 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.010 <0.010 Fe 71.93 71.70 79.87 88.15 95.06 96.20 99.33 65.81 96.75 48.59 78.19 59.97 60.00 84.80

Samples of Alloys A and B were prepared as follows.

The alloy was developed in a vacuum induction furnace, then it was cast in the form of an ingot of 30 to 50 kg, frustoconical, with a diameter ranging from 12 cm to 15 cm, with a height of 20 to 30 cm, which it was then rolled on a coarse rolling mill to a thickness of 80 mm, then hot rolled at a temperature of about 1000 ° C to give it a thickness of 2.5 mm.

• échantillon 1 : LAF 1 à taux de réduction 84% ; recuit 1 au défilé à 1100°C durant 3 min; LAF 2 à taux de réduction de 50% ; recuit 2 statique à 900°C, 1 h ;
• échantillon 2 : LAF 1 à taux de réduction 84% ; recuit 1 au défilé à 1100°C durant 3 min; LAF 2 à taux de réduction de 50% ; recuit 2 statique à 700°C, 1 h ;
• échantillon 3 : recuit 1 au défilé à 900°C durant 8 min; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé durant 8 min à 900°C; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 660°C, 1 h ;
• échantillon 4 : recuit 1 à 900°C au défilé durant 8 min; LAF 1 à 70% de taux de réduction ; recuit 2 à 900°C au défilé durant 8 min; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 680°C, 1 h ;
• échantillon 5 : recuit 1 à 900°C au défilé durant 8 min; LAF 1 à 70% de taux de réduction ; recuit 2 à 900°C durant 8 min; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 700°C, 1 h ;
• échantillon 6 : recuit 1 au défilé à 900°C durant 8 min; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé à 900°C durant 8 min; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 720°C, 1 h ;
• échantillon 7 : recuit 1 au défilé durant 8 min à 900°C; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé durant 8 min à 900°C; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 750°C, 1 h ;
• échantillon 8 : recuit 1 au défilé durant 8 min à 900°C; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé durant 8 min à 900°C; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 810°C, 1 h.
• échantillon 9 : recuit 1 au défilé durant 8 min à 900°C; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé durant 8 min à 900°C; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 900°C, 1 h.
• échantillon 10 : recuit 1 au défilé durant 8 min à 900°C; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé durant 8 min à 900°C; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 1100°C, 1 h.
• échantillon 11 : recuit 1 au défilé durant 8 min à 900°C ; LAF 1 à 80% de taux de réduction ; recuit 2 au défilé durant 8 min à 900°C ; LAF 2 à 40% de taux de réduction ; recuit 3 statique à 700°C, 1 h.
• échantillon 12 : recuit 1 au défilé durant 8 min à 900°C ; LAF 1 à 70% de taux de réduction ; recuit 2 au défilé à 1100°C durant 8 min; LAF 2 à 70% de taux de réduction ; recuit 3 statique à 700°C, 1 h.

Then, on these hot-rolled products, annealing and cold-rolling (LAF) suites were carried out at less than 100 ° C under the following conditions:

• sample 1: LAF 1 at 84% reduction rate; annealing 1 in line at 1100 ° C for 3 min; LAF 2 at a reduction rate of 50%; static annealing 2 at 900 ° C, 1 h;
• sample 2: LAF 1 at 84% reduction rate; annealing 1 in line at 1100 ° C for 3 min; LAF 2 at a reduction rate of 50%; static annealing 2 at 700 ° C, 1 h;
• sample 3: annealing 1 in line at 900 ° C for 8 min; LAF 1 to 70% reduction rate; annealing 2 in the process for 8 min at 900 ° C; LAF 2 at 70% reduction rate; static annealing 3 at 660 ° C, 1 h;
• sample 4: annealing at 1 at 900 ° C. in line for 8 min; LAF 1 to 70% reduction rate; annealing at 2 to 900 ° C in line for 8 min; LAF 2 at 70% reduction rate; static annealing 3 at 680 ° C, 1 h;
• sample 5: annealing 1 to 900 ° C in the process for 8 min; LAF 1 to 70% reduction rate; annealed at 2 at 900 ° C for 8 min; LAF 2 at 70% reduction rate; static annealing at 700 ° C, 1 h;
• sample 6: annealing 1 in line at 900 ° C for 8 min; LAF 1 to 70% reduction rate; annealing 2 in line at 900 ° C for 8 min; LAF 2 at 70% reduction rate; static annealing at 720 ° C, 1 h;
• sample 7: annealing 1 in line for 8 min at 900 ° C; LAF 1 to 70% reduction rate; annealing 2 in the process for 8 min at 900 ° C; LAF 2 at 70% reduction rate; static annealing 3 at 750 ° C, 1 h;
• sample 8: annealing 1 in line for 8 min at 900 ° C; LAF 1 to 70% reduction rate; annealing 2 in the process for 8 min at 900 ° C; LAF 2 at 70% reduction rate; static annealing 3 at 810 ° C, 1 h.
• sample 9: annealing 1 in line for 8 min at 900 ° C; LAF 1 to 70% reduction rate; annealing 2 in the process for 8 min at 900 ° C; LAF 2 at 70% reduction rate; static annealing 3 at 900 ° C, 1 h.
• sample 10: annealing 1 in line for 8 min at 900 ° C; LAF 1 to 70% reduction rate; annealing 2 in the process for 8 min at 900 ° C; LAF 2 at 70% reduction rate; static annealing 3 at 1100 ° C, 1 h.
• sample 11: annealing 1 in line for 8 min at 900 ° C; LAF 1 to 80% reduction rate; annealing 2 in the process for 8 min at 900 ° C; LAF 2 at 40% reduction rate; static annealing 3 at 700 ° C, 1 h.
• sample 12: annealing 1 in line for 8 min at 900 ° C; LAF 1 to 70% reduction rate; annealing 2 in line at 1100 ° C for 8 min; LAF 2 at 70% reduction rate; static annealing 3 at 700 ° C, 1 h.

The static annealing concluding the preparation were, for all the samples, preceded by a rise in temperature at a rate of 300 ° C / h and followed by cooling at a rate of the order of 200 ° C / h, carried out simply by leaving the samples in the annealing oven. The temperature rise rates before the final annealing and cooling after the final annealing were therefore relatively moderate, which in all cases contributed to obtaining a relatively low-textured final product, as will be seen in the Table 2. The differences in magnetostriction and its isotropy observed for the samples according to the invention and the reference samples will therefore be attributable to other factors, and in particular to the fact that, for the reference samples, there was a passage in the austenitic domain during annealing.

Note that the final annealing tests carried out at 850 ° C for 3 h in another static furnace, under a hydrogen atmosphere, with parameters comparable to those of the tests described here, but with a cooling rate after final annealing even lower (60 ° C / h), gave very similar results concerning the level of magnetostriction and its isotropy. The cooling after final annealing can therefore be particularly slow without drawbacks.

All annealing of all samples was carried out under pure, dry hydrogen atmosphere with a dew point below -40 ° C. No other gaseous species was present at more than 3 ppm.

Thus, the reference samples 1 and 2 underwent cold rolling directly after the hot treatments, then annealing at high temperature (1100 ° C) in the austenitic domain, then a second cold rolling, then a final annealing at 900 ° C (test 1) or 700 ° C (test 2) in the ferritic range.

The samples according to the invention 3 to 9 began, after the hot treatments, by undergoing annealing at 900 ° C, then a first cold rolling, then a second annealing at 900 ° C, then a second cold rolling, then a final annealing at a variable temperature according to the tests, from 660 to 900 ° C. All the annealing operations therefore took place in the ferritic field, in accordance with the invention, and were three in number, against two for the first two reference samples 1 and 2. All the cold rolling were carried out with a rate 70% reduction.

The reference sample 10 first underwent a ferritic annealing at 900 ° C just like the samples according to the invention and unlike the other two reference samples, then a first cold rolling, then an intermediate annealing at 900 ° C. , therefore in the ferritic field, then a second cold rolling, then a final annealing at a temperature of 1100 ° C, therefore in the austenitic field. It thus underwent a treatment comparable to that of samples 3 to 9 according to the invention, apart from the fact that the final annealing took place in the austenitic domain. All of its cold rolling was carried out at 70% reduction rate, as for the samples according to the invention.

Reference sample 11, after the hot treatments, underwent annealing at 900 ° C, then a first cold rolling at 80% instead of 70% like all samples 3 to 10 (which remains in accordance with l invention), then a second annealing at 900 ° C, then a second cold rolling at 40%, therefore in a manner not in accordance with the invention, instead of 70% like all samples 3 to 10, then a final annealing at a temperature of 700 ° C, therefore in the ferritic range.

The reference sample 12 is quite similar to the sample 10, by virtue of its passage through the austenitic domain, which however takes place at a different stage of the treatment. It first underwent a ferritic annealing at 900 ° C, just like the samples according to the invention and unlike the first two reference samples, then a first cold rolling, then an intermediate annealing in the austenitic range at 1100 ° C. , therefore in a manner not in accordance with the invention, then a second cold rolling, then a final annealing at a temperature of 700 ° C., therefore in the ferritic range. It thus underwent a treatment comparable to that of samples 3 to 9 according to the invention, apart from the fact that the intermediate annealing took place in the austenitic range. All of its cold rolling was carried out at 70% reduction rate, as for the samples according to the invention.

The characteristics of the various samples thus obtained, in terms of the presence of a Goss or {111} <110> texture measured by X-ray, of average grain diameter measured by image analysis of the samples, characterized by backscattered electron diffraction ( EBSD), and recrystallized fraction, measured at the surface area by the same EBSD technique, and assuming that the surface fraction is the volume fraction) are summarized in Table 2. Table 2: Texture, grain diameter and recrystallization rate of the samples tested according to their processing conditions Test Cold rolling reduction rate Final annealing temperature (° C) Alloy % Goss texture % texture {111} <110> Grain diameter (µm) Recrystallized fraction 1 Reference 84/50% 900 (but annealed 1 to 1100 ° C) TO 10 10 150 100% 2 Reference 84/50% 700 (but annealed 1 to 1100 ° C) TO 7 10 15 100% 3 Invention 70/70% 660 B 10 10 16 90% 4 Invention 70/70% 680 B 9 11 18 95% 5 Invention 70/70% 700 B 10 12 20 100% 6 Invention 70/70% 720 B 10 11 23 100% 7 Invention 70/70% 750 B 12 10 26 100% 8 Invention 70/70% 810 B 13 11 44 100% 9 Invention 70/70% 900 B 12 15 95 100% 10 Reference 70/70% 1100 (annealed 1 and 2 at 900 ° C) B 4 7 285 100% 11 Reference 80/40% 700 B 17 8 22 100% 12 Reference 70/70% 700 (but annealed 2 to 1100 ° C) B 6 11 21 100%

The different ranges of metallurgical treatments applied have led to substantially identical final grain sizes between the references and the tests according to the invention, that is to say a grain size range of approximately 300 to 15 μm: more precisely from 16 to 95 μm for the tests according to the invention, ie when all the annealing operations are carried out in the ferritic range; from 15 to 285 µm for the references, ie when at least one step of the process takes place outside the ferritic range. It can thus be seen that the grain size range is similar and has no connection with the low magnetostrictions obtained. But test 2, the final annealing of which was carried out at 700 ° C, has leads to a grain size markedly smaller than that of the reference tests 1 and 10 and 9 according to the invention, and which is of the same order of magnitude as those of the tests according to the invention 3 to 8 which were also carried out at temperatures around 700 ° C. In general, the metallurgical ranges of the tests according to the invention provide a grain size (between 16 and 95 μm depending on the tests) relatively close to that of the reference tests, and in any case fairly consistent with what could be expected. wait a priori, in particular in view of the conditions of the final annealing. It will be noted that the execution of an annealing at 900 ° C before the first cold rolling in the tests according to the invention and the reference test does not appreciably affect, by itself, the size of the grains obtained at the outcome of the entire process compared to reference tests 1 and 2 where the cold rolling was performed directly on the hot rolled sample.

More surprisingly, the noticeable differences between the treatment ranges of the different tests did not lead to very significant differences in the final textures of the materials, from the point of view of the proportion of Goss texture and the proportion of texture. {111} <110>.

Then the magnetostrictions (measured in ppm) on the different samples 1 to 3, 5, 7 to 12 cut, in different directions DL, DT and at 45 ° from DL and DT as indicated on the figure 1 (the direction mentioned is the direction of the sheet in which the large side of the rectangular sample is located), were observed, measured either parallel to the large side of the sample (therefore also parallel to the direction of the applied magnetic field and of the magnetic flux of the generated induction B) and noted "// H", i.e. perpendicular to the large side of the sample (therefore perpendicular to the direction of the applied magnetic field and of the magnetic flux of the generated B induction) and noted "⊥ H". The measurements were carried out continuously over a wide range of B and exploited precisely for three amplitudes of magnetic induction B: 1.2 T, 1.5 T and 1.8 T. The results are summarized in Table 3, where the different samples are designated by their composition A or B and by the temperature of their final annealing. The measurements were not carried out on samples 4 and 6, but it is certain that they would have been very comparable to those of the samples according to the invention treated at final annealing temperatures close to their own. Table 3: Results of magnetostriction tests Test Sample (composition, final annealing temperature) Magnetostriction strain measurement direction ε B = 1.2T B = 1.5 T B = 1.8 T Max Δλ 1.2 T Max Δλ 1.5 T Max Δλ 1.8 T λ ^{H // DL} (ppm) λ ^{H // 45 °} (ppm) λ ^{H // DT} (ppm) λ ^{H // DL} (ppm) λ ^{H // 45 °} (ppm) λ ^{H // DT} (ppm) λ ^{H // DL} (ppm) λ ^{H // 45 °} (ppm) λ ^{H // DT} (ppm) 1 A, 900 ° C // H +4.5 +3 +11 +9 +5 +18 +12 +10 +22 31 44 66.5 ^⊥ H -1.5 -4 -20 -5 -10.5 -35 -11 -17.5 -44.5 2 A, 700 ° C // H +1.2 +7 +8 +21 +13 +14 +30 +21 +21.5 22 38.5 54 ^⊥ H -10 -4 -4.5 -17.5 -8 -9 -24 -14 -14 3 B, 660 ° C // H 0 +2 0 +5 +9 +2.5 +10 +12.5 +8 4 15 20.5 ^⊥ H 0 -2 0 -2 -6 -2 -5.5 -8 -6 5 B, 700 ° C // H 0 0 0 +0.5 0 0 +5 +4.5 +3 0 2.5 10 ^⊥ H 0 0 0 -0.5 -2 0 -5 -5 -2.5 7 B, 750 ° C // H 0 0 +1 0 +1 +1.5 +2 +5 +6 1 2.5 9 ^⊥ H 0 0 0 0 -1 -0.4 -0.5 -3 -2 8 B, 810 ° C // H 0 0 0 0.5 +2 +3 +4.5 +5.5 +7.5 0 6 15 ^⊥ H 0 0 0 0 -1 -3 -3 -4.5 -7.5 9 B, 900 ° C // H 0 -1 0 0 -2 0 -1 -1.5 +2.5 1.5 3 5 ^⊥ H 0 0.5 0 0.5 +1 0 +1 0 -2.5 10 B, 1100 ° C // H +10 +13 +9.5 +17 +22.5 +17 22.5 +31 +25.5 20.5 34.5 47 ^⊥ H -7.50 -7 -6.50 -12 -10 -10 -16 -14 -14.5 11 B, 700 ° C // H 15 8.5 13 25 14 21.5 38 23 27 25 38.5 57.5 ^⊥ H -8 -3 -10 -12 -7 -17 -18.5 -12.5 -19.5 12 B, 700 ° C // H 8 9 10 14.5 15 15.5 22 22 22.5 15.5 25.5 36.5 ^⊥ H -4.5 -5 -5.5 -9 -9.5 -10 -14 -14 -14

There are very strong differences in magnetostriction measurements, in terms of absolute value and isotropy, between the reference tests 1, 2 for which the first annealing was carried out in the austenitic field, and the tests according to the invention 3 at 9 where all the anneals have been carried out in the ferritic range, including the optional annealing preceding the first cold rolling, not carried out in tests 1 and 2 of reference.

It can also be seen, from test 10, that by escaping only at the end of the process from the ferritic phase, by a final annealing carried out in the austenitic domain, the low and isotropic magnetostriction aimed at is not obtained either, although here too a ferritic annealing was carried out preceding the first cold rolling.

Benchmark Test 11 shows that the target low and isotropic magnetostriction is also not obtained when one of the cold rollings is performed at a low reduction rate, even though all annealing takes place in the field. ferritic.

Reference run 12 shows that the target low and isotropic magnetostriction is also not obtained when the second of the three anneals is carried out in the austenitic range. Reference Examples 1 and 2 had austenitic annealing performed at the start of processing, after the first cold rolling, and Reference Example 10 had austenitic annealing performed at the very end of processing. Example 12 therefore completes the demonstration of the harmfulness of austenitic annealing whatever its position in the treatment.

The figures 2 to 12 highlight these differences.

The figure 2 translates the magnetostriction results observed during the reference test 1. It can be seen that even for weak inductions of the order, in absolute value, of 0.5T, the magnetostriction according to DT begins to become significant and increases very quickly with induction. For DL and for the 45 ° steering of DT and DL, it is from around 1 T that the magnetostriction begins to increase noticeably and rapidly. This leads to significant magnetostriction deformations which can reach several tens of ppm in certain directions at inductions of the order of 2 T, and to a strong anisotropy of these deformations, all this going in the direction of the creation of a noise of magnetostriction too intense for the privileged applications of the invention envisaged.

The figure 3 reflects the magnetostriction results observed during the reference test 2. It is observed there that, compared to test 1, the isotropy of the magnetostriction is slightly improved, and certain extreme values of the magnetostriction are a little lower . But from an induction of 1 T, the magnetostriction begins to become important in the three directions considered. The material thus obtained would therefore not be well suited to the privileged applications of the invention either. The significantly smaller grain size in the test 2 sample than in the test 1 sample therefore did not very fundamentally improve the magnetostriction results.

The figure 4 reflects the magnetostriction results observed during test 3 according to the invention. In this case, the shape of the curves changes radically. On the one hand, we observe a magnetostriction which remains almost zero in all the directions considered up to values of the induction exceeding a little 1 T. And when this magnetostriction begins to increase for higher fields, its value remains very significantly lower than during reference tests 1 and 2. In addition, the magnetostriction differences between the different directions remain relatively small, even for high fields. At 2 or -2 T, we have a magnetostriction which does not reach 15 ppm or -10 ppm, and this for all directions considered. These results are therefore very significantly better than for the reference tests, and they are sufficient to make the materials thus prepared suitable for constituting, in particular, the cores of low noise on-board aeronautical transformers.

The figure 5 reflects the magnetostriction results observed during test 7 according to the invention. We find qualitatively magnetostriction curves very comparable to those of test 3 ( figure 4 ), with, moreover, a magnetostriction which begins to become significant only for inductions of at least ± 1.5 T. At ± 2 T, the magnetostriction can be less than 5 ppm and never exceeds 10 ppm. We therefore have excellent results for this test, which differs from test 3 only by its final annealing temperature of 750 ° C, instead of 660 ° C, which led to total recrystallization while it did not was only 90% in trial 3.

The figure 6 reflects the magnetostriction results observed during test 8 according to the invention, which had a final annealing temperature of 810 ° C. We find qualitatively magnetostriction curves very comparable to those of test 3 ( figure 4 ) and test 7 ( figure 5 ). Quantitatively, the results are good, with maximum values of the magnetostriction which remain of the order of ± 10 ppm even for inductions of ± 2 T, and a Max Δλ of 15 ppm at 1.8T.

The figures 7 to 9 compare the magnetostriction measurements recorded for tests 5 and 9 according to the invention. The figure 7 shows the tests carried out according to the DT direction, the figure 8 shows the tests carried out in the 45 ° direction and the figure 9 shows the tests carried out according to the DT direction. The results are very comparable and excellent for the two tests according to the DL and DT directions up to inductions of ± 1.8 T. For the 45 ° direction, the magnetostriction begins to no longer be completely negligible from 1 , 8 T approximately in the case of test 5, while in test 9 it remains very low even beyond 2 T. In general, a final annealing temperature of 900 ° C therefore gives results of magnetostriction better than final annealing at 700 ° C. But already at 700 ° C the magnetostriction at 1.8T does not exceed ± 5 ppm in the three measurement directions, which is very significantly better than for the reference tests, both for the absolute value of the magnetostriction and for its isotropy.

The results of test 9 are particularly remarkable at strong inductions of 1.8 T or even slightly beyond, both on the weakness of the magnetostriction and on its isotropy.

The figure 10 shows the results of the reference test 10 in which the final annealing was carried out at 1100 ° C, therefore in the austenitic range, while the two previous anneals 1 and 2, carried out at 900 ° C. like all anneals 1 and 2 of the tests according to the invention, had been carried out in the ferritic field. We find magnetostriction curves in the various directions comparable, qualitatively and quantitatively, to those of the other reference tests 1 and 2, seen on the figures 3 and 4 . It can be concluded from this that the passage of the alloy in the austenitic domain during one of its anneals, even if it only occurs at the end of the treatment, constitutes a very important factor in the failure to obtain a weak and isotropic magnetostriction.

Test 11, in which the second cold rolling was carried out with a reduction rate of only 40%, shows, according to the figure 11 , a conventional parabolic and low isotropic behavior of the magnetostriction as a function of the induction, therefore a behavior outside the invention, with for example a magnetostriction according to DL of more than 35ppm at 1.5T, of nearly 60ppm at 1, 8T. It can be concluded that the texture filiation, modulated by the cold rolling reduction rates, is effectively well controlled by the texture transformations during cold rolling, which restricts the invention to certain ranges of reduction rates. .

The figure 12 shows the results of the reference test 12 in which the intermediate annealing was carried out at 1100 ° C, therefore in the austenitic range, while the two anneals 1 and 3 were carried out at 900 ° C like all the anneals 1 and 3 tests according to the invention, therefore in the ferritic field. We find magnetostriction curves in the various directions comparable to those of the other reference tests 1, 2 and 10, seen on the figures 3 , 4 and 10 , however with a fairly large isotropy of the magnetostriction. But the level of magnetostriction remains too high, even for relatively low inductions. It can be concluded, in conjunction with test 10, that the passage of the alloy in the austenitic domain during any of its anneals constitutes a very important factor in the failure to obtain magnetostriction at the both weak and isotropic

It was also noted with surprise that the magnetic losses at 400 Hz for different inductions (1, 1.2 and 1.5 T) were notably lower in the case of the materials obtained according to the invention than they are for the materials. of non-oriented grain reference. One might have thought that the examples according to the invention could exhibit unacceptable magnetic losses by induced currents, either because of their structure not entirely recrystallized (tests 3 and 4), or of their fine-grained microstructures. However, the results presented in Table 4 demonstrate the opposite. They were obtained on samples 0.2 mm thick, 100 mm long and 20 mm wide cut according to DL, immersed in a magnetic field of frequency fundamental 400 Hz and by slaving the magnetic induction according to a sinusoidal temporal form. The measurements were made for maximum amplitudes of induction B of intensity equal to 1, 1.2, 1.5 or 1.8 T. The magnetic losses are expressed in W / kg. Table 4: Magnetic losses at 400 Hz measured on different samples Composition / final annealing B = 1 T B = 1.2 T B = 1.5 T B = 1.8 T Test 1 (reference) A / 900 ° C 40 50 78 113 Test 2 (reference) A / 700 ° C 47 61 120 156 Test 3 (invention) B / 660 ° C 48 62 90 130 Test 5 (invention) B / 700 ° C 48 62 90 113 Test 7 (invention) B / 750 ° C 32 44 65 96 Test 8 (invention) B / 810 ° C 27 38 56 80 Test 9 (invention) B / 900 ° C 22 30 45 63 Test 10 (reference) B / 1100 ° C 35 48 75 101

As can be seen, the magnetic losses of the samples produced according to the invention and having grains of reduced size and a structure not completely recrystallized (tests 3 and 4) or completely recrystallized thanks to a final annealing of 700 ° C or more are not not particularly high, and remain competitive with that obtained on the reference samples. Above all, the samples according to the invention that are 100% recrystallized and produced with a final annealing at 720 ° C and more (up to 810 ° C, test 8 or better 900 ° C test 9) exhibit magnetic losses which are still significantly improved compared to to the reference samples, including that of test 1 which has a large grain size and a 100% recrystallized structure. This advantage over magnetic losses is, for the moment, not clearly explained by the inventors. It is all the more remarkable when one places oneself at inductions higher than 1.5 T, such as 1.8 T (see table 4), since the magnetic losses vary according to the square of the induction. Here again, this is an advantage for use in on-board aeronautical transformers, the sizing of which is strongly linked to the evacuation of the various losses (by Joule effect and magnetic).

It should be noted that, surprisingly, while the large grain size of the reference test 10 went a priori in the direction of obtaining the magnetic losses the lowest, it is test 9 of the invention which exhibits the lowest magnetic losses.

In general, the results are all the more favorable in terms of magnetic losses as the temperature of the final ferritic annealing is higher, the best results being obtained for the sample of test 9 which was annealed at 900 ° C. .

For magnetostriction, the ferritic annealing temperatures between 800 and 900 ° C show a weak to very weakly marked deformation anisotropy and Max Δλ amplitude deviations of magnetostriction not exceeding, in all cases, not 6 ppm at 1.5T , 15 ppm at 1.8T, therefore significantly better than those of the samples of the reference tests.

In general, the invention is defined by saying, in particular, that all annealing must take place in the ferritic range, at a minimum temperature of 650 ° C and at a maximum temperature which, taking into account the effective composition of the l The alloy is indeed in the purely ferritic range, without a transformation of at least part of the ferrite into austenite occurring. We saw above what this maximum temperature was as a function of the Si, Co and C contents of the alloy.

The strips obtained according to the invention can be used to constitute transformer cores which are both of the “cut-stacked” type and of the “wound” type as defined above. In the latter case, to carry out the winding, it is necessary to use very thin strips of the order of 0.1 to 0.05 mm thick, for example.

As has been said, annealing carried out before the first cold rolling is preferably carried out within the framework of the invention. However, this annealing is not essential, in particular in the case where the hot-rolled strip has remained in the coiled state for a long time during its natural cooling. In this case, the winding temperature often being of the order of 850-900 ° C, the duration of this stay can be quite sufficient to obtain very comparable effects on the microstructure of the strip at this stage. to those which would be provided by a true annealing in the ferritic range carried out under the conditions which have been said for the optional annealing before the first cold rolling.

Table 5 recalls the results obtained during tests 1 and 9 previously described on the isotropy of magnetostriction and the magnetic losses at 1.5 T, 400 Hz, and it adds information on the suitability for cold rolling or lukewarm samples before being applied to a treatment according to the method of the invention, and the saturation magnetization Js of the final product. These results are also compared with those obtained during tests numbered 13 to 24, in which alloys of conforming compositions (13 to 19 and 23, 24) or not (20 to 22) to the invention were also tested. The compositions of these new alloys are also specified, with those of tests 1 and 9 as a reminder. As samples K and L of tests 21 and 22 were found to be unsuitable for cold or warm rolling (breakage due to brittleness, starting from the middle of the strip towards the edges), these tests were not continued at -beyond the rolling attempt, hence the absence of results concerning them in Table 5.

For all these samples, the final thickness is 0.2 mm. Table 5: Conditions and results of tests 1, 9, 13-24 Alloy A (%) Ref. Alloy B (%) Inv. Alloy C (%) Inv. Alloy D (%) Inv. Alloy E (%) Inv. Alloy F (%) Inv. Alloy G (%) Inv. Alloy H (%) Inv. Alloy I (%) Inv. Alloy J (%) Ref. Alloy K (%) Ref. Alloy L (%) Ref. Alloy M (%) Inv. Alloy N (%) Inv. Test 1 9 13 14 15 16 17 18 19 20 21 22 23 24 VS 0.010 0.009 0.007 0.023 0.012 0.013 0.011 0.012 0.010 0.008 0.009 0.009 0.012 0.015 Mn 0.261 0.256 0.195 0.234 0.248 0.421 0.532 0.810 0.167 0.208 0.520 0.289 0.368 <0.010 Yes 0.142 0.153 0.330 0.720 0.031 2.73 0.070 0.013 3.50 0.023 3.07 1.53 0.640 0.083 S 0.0023 0.0042 0.0033 0.0021 0.0048 0.0008 0.0006 0.0028 0.0005 0.0015 0.0007 0.0044 0.0008 <0.0005 P 0.0025 0.0055 0.0031 0.0029 0.0029 0.0032 0.0047 0.0037 0.0053 0.0031 0.0043 0.0049 0.0041 <0.0005 Or 0.030 0.030 0.100 <0.01 0.130 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.080 <0.01 Cr 0.514 0.498 1.00 0.200 0.011 0.008 0.048 6.06 0.047 0.089 0.007 0.038 0.072 <0.01 Mo <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 0.17 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cu 0.009 0.010 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Co 27.09 27.32 18.35 10.07 4.21 0.02 <0.01 27.11 <0.01 49.00 18.20 38.15 38.82 15.10 V 0.01 0.01 <0.005 0.51 <0.005 <0.005 <0.005 <0.005 <0.005 2.03 <0.005 <0.005 <0.005 <0.005 Al <0.001 <0.001 0.14 <0.001 <0.001 0.60 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Nb <0.001 <0.001 <0.001 <0.001 0.005 <0.001 <0.001 <0.001 <0.001 0.040 <0.001 <0.001 <0.001 <0.001 Ti <0.001 <0.001 <0.001 0.08 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 NOT 0.0015 0.0044 0.0023 0.0036 0.0043 0.0027 0.0041 0.0045 0.0048 0.0018 0.0021 0.0019 0.0027 0.0012 That <0.0003 <0.0003 <0.0003 0.0013 <0.0003 <0.0003 0.0009 <0.0003 <0.0003 0.0007 0.0015 <0.0003 0.0009 <0.0003 Mg <0.0002 <0.0002 0.0006 <0.0002 <0.0002 0.0005 0.0004 <0.0002 <0.0002 0.0004 <0.0002 0.0004 <0.0004 <0.0002 Your <0.002 <0.002 0.0025 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 B <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 0.0007 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 W <0.005 <0.005 <0.005 <0.005 0.28 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 <0.010 <0.010 Fe 71.93 71.7 79.87 88.15 95.06 96.20 99.33 65.81 96.75 48.59 78.19 59.97 60.00 84.80 Suitable for cold or warm rolling YES YES YES YES YES YES YES YES YES YES NO NO YES YES Final annealing temperature R3 (° C) 900 900 900 900 850 850 850 900 850 880 900 900 Final annealing time R3 (min) 60 60 600 600 300 120 120 300 30 180 60 60 Max Δλ at 1.2 T (ppm) 31 1.5 2 5 6 5 10 2.5 4 28 9 2.7 Max Δλ at 1.5 T (ppm) 44 3 6 8 9 8 13 4 7 46 15 3.5 Max Δλ at 1.8 T (ppm) 66.5 5 8 11 13 12 18 7.5 10 73 23 5 Magnetic losses 1.5 T / 400 Hz (W / kg) 78 45 49 52 53 38 59 38 36 42 48 51 Js (T) 2.35 2.35 2.25 2.20 2.16 1.97 2.14 2.12 2.02 2.35 2.34 2.25

As we have seen, sample A (test 1) underwent, without prior annealing, an LAF 1 at a reduction rate of 84%, then an R1 annealing at 1100 ° C for 3 min, then an LAF 2 at a reduction rate of 50%, then static R2 annealing at 900 ° C. for 1 h.

Samples B to H (tests 2 to 18) underwent R1 annealing at 900 ° C for 8 min, then LAF 1 at a reduction rate of 70%, then R2 annealing at 900 ° C for 8 min. min at 900 ° C, then an LAF 2 at a reduction rate of 70%, then a static R3 annealing at different temperatures and times, noted in Table 5.

Sample I (test 19) underwent R1 annealing at 900 ° C for 8 min, then lukewarm rolling 1 at 150 ° C with a reduction rate of 70%, then R2 annealing at 900 ° C ° C for 8 min, then lukewarm rolling 2 at 150 ° C with a reduction rate of 70% and static R3 annealing at 850 ° C for 30 min.

Sample J (test 20) underwent static R1 annealing at 935 ° C for 1 h, then an LAF 1 at 70% reduction rate, then R2 annealing at 900 ° C for 8 min, then a LAF 2 at 70% reduction rate, then static R3 annealing at 880 ° C for 1 h.

As we have seen, the reference test 1 carried out on the FeCo27 type alloy A did not give satisfactory results, from the point of view of the isotropy of the magnetostriction: see the high values of Max Δλ observed. This is, apparently, to be linked to the fact that one of its anneals (the R1) was carried out at a high temperature (1100 ° C) located in the austenitic range.

The test according to the invention 9, carried out on the alloy B which is also an FeCo27, for which all the annealing took place in the ferritic range, on the other hand, led to excellent isotropy of the magnetostriction.

We find this good isotropy of the magnetostriction on tests 13 and 14 which concern FeCo alloys having Co contents lower than 27%: respectively 18 and 10% approximately, and of which the composition and the treatments are, moreover, in conformity. to the other requirements of the invention. Example 13 also exhibits relatively significant Si, Cr, Al, Ca, Ta contents. Example 14 also exhibits significant Si, V and Ti contents. But all these contents remain within the limits defined for the invention.

Likewise, a good isotropy of the magnetostriction is present on test 23 which concerns an FeCo alloy having a Co content of nearly 39%, therefore significantly higher than 27% but remaining within the limit of 40% at the maximum set. for the invention, and an Si content which is significant, but not so high as to compromise cold or warm rollability. The loss magnetic fields and the saturation magnetization are of the same order of magnitude as for the other samples treated according to the invention.

Regarding test 24, it relates to an alloy containing 15% Co and devoid of significant contents of other alloying elements, in particular Cr. It also exhibits a particularly weak and isotropic magnetostriction. The magnetic losses and the saturation magnetization are of the same order of magnitude as for the other samples treated according to the invention. In particular, compared to test 13, the absence of Cr in test 24, this absence tending to increase the saturation magnetization, is compensated for by a slightly less presence of Co which, for its part, goes into the sense of a decrease in saturation magnetization. Likewise, the absence of Cr in test 24 goes in the direction of an increase in magnetic losses compared to test 13, but the lower Co content in test 24 goes in the direction of a reduction of these same magnetic losses. Therefore, the differences in the composition of the alloy between tests 13 and 24 tend to compensate for each other, from the point of view of the magnetic losses and Js.

Regarding the reference test 20, it was carried out on an FeCo alloy containing 49% Co, therefore above the upper limit of 40% accepted by the invention. All of its anneals have been carried out in the ferritic field. Its magnetic losses are very good, but its magnetostriction does not have the desired isotropy. As has been said, at these excessively high Co contents, the order-disorder transition during heat treatments is undoubtedly too rapid and acute, and the number of anneals required by the invention is not compatible with this composition. of the alloy. The presence of 0.04% Nb, although still below the maximum limit tolerated by the invention, may also have contributed to hampering the mechanism of texture parentage, which has been said to be an explanation for the isotropy of the magnetostriction observed when the method according to the invention is applied.

Concerning the reference test 21, its Si content is too high compared to the Co content, and the condition "Si + 0.6% Al ≤ 4.5 - 0.1% Co if Co <35%" required by the invention is not satisfied. The consequence is, as explained previously, that the alloy is not suitable for being cold or warm rolled, as experience confirms.

Regarding the reference test 22, we find ourselves in the case where Co is ≥ 35% and where Si, according to the invention, should therefore not exceed 1% to ensure good suitability for cold or warm rolling. However, the Si content in this test is 1.53%: here again it is confirmed that the good cold or warm laminability of the alloy is not obtained only under certain compositional conditions, which must be included in the definition of the invention.

Test 15 according to the invention shows that a relatively low Co content (4.21%) is not inconsistent with obtaining the desired good magnetostriction isotropy, if the Si and Al contents are sufficiently low. . The presence of 0.005% of Nb does not interfere with obtaining the desired results.

Test 16 according to the invention relates to an Fe-Si-Al alloy with a very low Co content. In its case, the desired isotropic magnetostriction is also obtained, together with low magnetic losses.

Test 17 according to the invention relates to an alloy which is practically 99% pure Fe, with relatively low presences of Mn, Ca, Mg. The isotropy of the magnetostriction is less than in the other tests according to the invention, but it is nevertheless very good in absolute terms, as Max Δλ at 1.8 T remains ≤ 25 ppm as required on the sheets or strips according to l 'invention. The magnetic losses are also a little higher than for the other tests according to the invention, but remain at a good level, and are lower than those observed on the reference test 1.

Test 18 according to the invention relates to an FeCo27 type alloy with a high Cr content (6%) and also containing Mn (0.81%) and a little Mo and B. The good isotropy of the magnetostriction is confirmed, and the magnetic losses are as low as for test 16 despite the presence of 7 ppm of B. The saturation magnetization remains of the order of that observed during the other tests, such as the contents of Cr, Mn and Mo is not so high that it deteriorates undesirably.

Test 19 according to the invention relates to an Fe-Si alloy containing 3.5% Si and not containing Al, and shows that the operating conditions of the process according to the invention are also applicable with profit to this type of FeSi3 alloys to achieve the desired magnetostriction isotropy. In addition, this example exhibits particularly low magnetic losses.

Table 6 presents experimental results obtained by varying the treatment conditions, the composition of the alloy treated and the final thickness of the sample. The results of the previous tests 1 and 9 were taken over, and new tests 25 to 31 carried out on alloys having the compositions B (FeCo27), I (FeSi3) and C (FeCo18) explained in Table 5 were added. Table 6: Influence of the treatment conditions on the isotropy of the magnetostriction for different alloy compositions and final thicknesses of the sample Test number Alloy Final thickness (mm) R1 annealing time (min) LAF 1 reduction rate (%) R2 annealing time (min) LAF 2 reduction rate (%) Annealing temperature R3 (° C) Duration R3 (min) Max Δλ at 1.8 T (ppm) 1 TO 0.2 0 84 3 50 900 60 66.5 Reference 9 B 0.2 8 70 8 70 900 60 6 Invention 25 B 0.2 8 70 8 70 900 240 7 Invention 26 B 0.2 8 70 8 70 900 1440 5 Invention 27 B 0.2 8 70 8 70 920 60 2.7 Invention 28 B 0.2 8 70 8 70 920 240 5.4 Invention 29 B 0.2 8 70 8 70 920 1440 6 Invention 30 I 0.2 60 70 60 70 850 180 16 Invention 31 VS 0.5 5 60 5 50 900 60 18.5 Invention

If we compare the results of the various tests according to the invention, carried out on samples of the same composition, we see that varying the parameters of the LAFs and annealing within the limits of the definition of the invention still makes it possible to obtain a unusually good magnetostriction isotropy in all cases.

It can be noted that concerning alloy I (of the FeSi3 type), a comparison between tests 19 and 30 makes it possible to deduce that the increase in the temperature and the duration of the final annealing R3 in test 30 caused a certain degradation of this isotropy, which nevertheless remains within the limits of the objectives set. It is believed to be possible to relate this degradation to the fact that the Goss texture component was arguably stronger in run 30 and near the preferred upper limit of 30%, also due to differences in the hot rolling process.

It can also be noted that, concerning the alloy C (of the FeCo18 type), a final thickness of 0.5 mm obtained before the final annealing R3 leads, for identical final annealing conditions R3, to a certain degradation of the isotropy. magnetostriction (see test 31). This could be remedied by increasing, for this thickness, the duration and / or the temperature of the final annealing while remaining within the limits set by the definition of the invention.

In general, it can be seen, in the light of the various tests carried out, that the magnetic properties of the samples (magnetic losses and magnetostriction in particular) are relatively little dependent on the precise conditions of the final annealing, contrary to what has often been observed in prior art. The use of multiple rolling with an intermediate annealing between each rolling, and a final annealing after the last cold rolling (and not a single cold rolling followed by a final annealing), combined with obtaining a very strongly, or even completely, recrystallized final product, could be one of the factors favorable to this wide tolerance in the manufacturing conditions, which is obviously very advantageous. The persistence, over the course of manufacture, of at most low proportions of Goss textures and {111} <110> (or, generally, less than 30% of any texture component {hkl} <uvw> defined by a disorientation of less than 15 ° around a crystallographic orientation defined {h ₀ k ₀ l ₀ } <u ₀ v ₀ w ₀ >), which the process according to the invention makes it possible to obtain, could also contribute to this result. The inventors are, however, for the moment only at the stage of hypotheses to explain the remarkable properties obtained at the same time on the isotropy of the magnetostriction and the magnetic characteristics thanks to the application of the method of the invention. .

The strips and sheets according to the invention make it possible to manufacture, in particular, after their cutting, transformer cores composed of sheets stacked or wound up, without requiring modifications to the general design of the cores of these types usually used. It is thus possible to take advantage of the properties of these sheets to produce transformers producing only a low magnetostriction noise compared to existing transformers of similar design and sizing. Transformers for aircraft intended to be installed in a cockpit are a typical application of the invention. These sheets can also be used to form transformer cores of higher mass, therefore intended for transformers of particularly high power, while retaining a magnetostriction noise remaining within acceptable limits. The transformer cores according to the invention can consist entirely of sheets made from strips or sheets according to the invention, or only partially in cases where it is considered that their association with other materials would be technically or financially advantageous.

Claims (9)

1. Sheet or strip of cold rolled and annealed ferrous alloy (1), its composition consists of, in weight percentages:

   - traces ≤ C ≤ 0.2%, preferably traces ≤ C ≤ 0.05%, more preferably traces ≤ C ≤ 0.015%;

   - traces ≤ Co ≤ 40%;

   - if Co ≥ 35%, traces ≤ Si ≤ 1.0%;

   - if traces ≤ Co < 35%, traces ≤ Si ≤ 3.5%;

   - if traces ≤ Co < 35%, Si + 0.6% Al ≤ 4.5 - 0.1% Co, preferably Si + 0.6% Al ≤ 3.5 - 0.1% Co;

   - traces ≤ Cr ≤ 10%;

   - traces ≤ V + W + Mo + Ni ≤ 4%, preferably ≤ 2%;

   - traces ≤ Mn ≤ 4%, preferably ≤ 2%;

   - traces ≤ Al ≤ 3%, preferably ≤ 1%;

   - traces ≤ S ≤ 0.005%;

   - traces ≤ P ≤ 0.007%;

   - traces ≤ Ni ≤ 3%, preferably ≤ 0.3%;

   - traces ≤ Cu ≤ 0.5%, preferably ≤ 0.05%;

   - traces ≤ Nb ≤ 0.1%, preferably ≤ 0.01%;

   - traces ≤ Zr ≤ 0.1%, preferably ≤ 0.01%;

   - traces ≤ Ti ≤ 0.2%;

   - traces ≤ N ≤ 0.01%;

   - traces ≤ Ca ≤ 0.01%;

   - traces ≤ Mg ≤ 0.01%;

   - traces ≤ Ta ≤ 0.01 %;

   - traces ≤ B ≤ 0.005%;

   - traces ≤ O ≤ 0.01%;

   the remainder being iron and impurities resulting from the preparation, characterized in that, for an induction of 1.8 T, the maximum difference (Max Δλ) between the magnetostriction deformation amplitudes λ, measured parallel to the magnetic field (Ha) applied (λ//H) and perpendicular to the magnetic field (Ha) applied (λ^⊥H) on three rectangular samples (2, 3, 4) of the said sheet or strip, whose long sides are respectively parallel to the direction of rolling (DL) of the said sheet or strip, parallel to the transverse direction (DT) of the said sheet or strip, and parallel to the direction forming an angle of 45° with the said rolling direction (DL), and with the said transverse direction (DT), is at most 25ppm, in that its recrystallization rate is 80 to 100% and in that it comprises not more than 30% of any {hkl}<uvw> texture component defined by a disorientation of less than 15° from a defined crystallographic orientation {h₀k₀l₀}<u₀v₀w₀>, the texture component and the difference between the magnetostriction deformation amplitudes being measured according to the description.
2. Sheet or strip according to claim 1, characterized in that 10% ≤ Co ≤ 35%.
3. Method of manufacturing a ferrous alloy strip or sheet (1) according to one of claims 1 or 2, in that:

   - a ferrous alloy is prepared, the composition of which consists of:

   - traces ≤ C ≤ 0.2%, preferably traces ≤ C ≤ 0.05%, more preferably traces ≤ C ≤ 0.015%;

   - traces ≤ Co ≤ 40%;

   - if Co ≥ 35%, traces ≤ Si ≤ 1.0%;

   - if traces ≤ Co <35%, traces ≤ Si ≤ 3.5%;

   - if traces ≤ Co <35%, Si + 0.6% Al ≤ 4.5 - 0.1% Co, preferably Si + 0.6% Al ≤ 3.5 - 0.1% Co;

   - traces ≤ Cr ≤ 10%;

   - traces ≤ V + W + Mo + Ni ≤ 4%, preferably ≤ 2%;

   - traces ≤ Mn ≤ 4%, preferably ≤ 2%;

   - traces ≤ Al ≤ 3%, preferably ≤ 1%;

   - traces ≤ S ≤ 0.005%;

   - traces ≤ P ≤ 0.007%;

   - traces ≤ Ni ≤ 3%, preferably ≤ 0.3%;

   - traces ≤ Cu ≤ 0.5%, preferably ≤ 0.05%;

   - traces ≤ Nb ≤ 0.1%, preferably ≤ 0.01%;

   - traces ≤ Zr ≤ 0.1%, preferably ≤ 0.01%;

   - traces ≤ Ti ≤ 0.2%;

   - traces ≤ N ≤ 0.01%;

   - traces ≤ Ca ≤ 0.01%;

   - traces ≤ Mg ≤ 0.01%;

   - traces ≤ Ta ≤ 0.01%;

   - traces ≤ B ≤ 0.005%;

   - traces ≤ O ≤ 0.01%;

   the remainder being iron and impurities resulting from the preparation;

   - it is cast in the form of an ingot or a semi-finished continuously cast product;

   - the said ingot or semi-finished continuously cast product is hot-shaped in the form of a strip or a sheet 2 to 5 mm thick, preferably 2 to 3.5 mm thick;

   - at least two cold rolling operations of the said strip or sheet are performed, each having a reduction ratio of 50 to 80%, preferably 60 to 75%, at a temperature which is:

   - from ambient temperature to 350°C if the alloy has an Si content such that 3.5 - 0.1% Co ≤ Si + 0.6% Al ≤ 4.5 - 0.1% Co and Co < 35%, or if the alloy contains Co ≥ 35% and Si ≤ 1%; and if the cold rolling is preceded by reheating, preferably stoving, for a period of 1h to 10h at a temperature of less than or equal to 400°C;

   - from room temperature to 100°C in other cases;

   - wherein the said cold rollings are each separated by static or continuous annealing in the ferritic range of the alloy for 1 min to 24 hours, preferably for 2 min to 1 hour, at a temperature of at least 650°C. C, preferably at least 750°C, and at most:

   - 1400°C, if the Si content of the alloy is greater than or equal to (% Si)_α-lim = 1.92 + 0.07% Co + 58% C;

   - T _α-lim = T₀ + k% Si, where T₀ = 900 + 2% Co - 2833% C and k = 112 - 1250% C, if the Si content is less than (% Si)_α-lim;

   - the said annealing separating two cold rolling operations taking place in an atmosphere containing at least 5% of hydrogen, preferably 100% of hydrogen, and less than 1% in total of gaseous oxidizing species for the alloy, preferably less than 100 ppm, and having a dew point below + 20°C, preferably below 0°C, more preferably below -40°C, optimally below -60°C;

   - and wherein a final static or continuous recrystallization annealing is carried out in the ferritic range of the alloy for 1 min to 48 h, at a temperature of 650 to (900 ± 2% Co)°C, in order to obtain a recrystallization rate of the strip or the sheet from 80 to 100%

   characterized in that the final recrystallization annealing is preceded by a heating at a speed less than or equal to 2000°C/h, preferably less than or equal to 600°C/h and is followed by cooling performed at a speed less than or equal to 2000°C/h, preferably less than or equal to 600°C/h.
4. Process according to claim 3, characterized in that the final recrystallization annealing is carried out under vacuum, or in a non-oxidizing atmosphere for the alloy, or in a hydrogenated atmosphere.
5. Process according to claim 4, characterized in that the final recrystallization annealing is carried out in an atmosphere containing at least 5% hydrogen, preferably 100% hydrogen, and less than 1% in total of gaseous oxidizing species for the alloy, preferably less than 100 ppm, and having a dew point below + 20°C, preferably below 0°C, more preferably below -40°C, optimally below -60°C .
6. Process according to one of claims 3 to 5, characterized in that the first cold rolling is preceded by a static or continuous annealing in the ferritic range of the alloy, for 1 min to 24 h, preferably for 2 min to 10 h, at a temperature of at least 650°C, preferably at least 700°C, and at most:

   - 1400°C, if the Si content of the alloy is greater than or equal to (% Si)_α-lim = 1.92 + 0.07% Co + 58% C;

   - T_α-lim = T₀ + k% Si, where T₀ = 900 + 2% Co - 2833% C and k = 112 - 1250% C, if the Si content is less than (% Si)_α-lim;

   - wherein the said annealing takes place in an atmosphere containing at least 5% hydrogen, preferably 100% hydrogen, and less than 1% in total of gaseous oxidizing species for the alloy, preferably less than 100 ppm, and having a dew point below + 20°C, preferably below 0°C, more preferably below -40°C, optimally below -60°C.
7. Process according to one of claims 3 to 6, characterized in that, after the final recrystallization annealing, an oxidation annealing is carried out at a temperature between 400 and 700°C, preferably between 400 and 550°C, for a time that is sufficient to obtain an insulating oxidized layer with a thickness of 0.5 to 10 µm on the surface of the sheet or strip.
8. Transformer magnetic core, characterized in that it is composed of stacked or rolled-up sheets, at least some of which have been prepared from a sheet or strip according to one of claims 1 or 2.
9. Transformer comprising a magnetic core, characterized in that the said core is of the type according to claim 8.

Images