PERMANENTLY EXCEEDED ROTOR WITH IMPROVED MAGNETIC GEOMETRY

Description

[0001] The present invention relates generally to the field of permanent magnet rotors.

[0002] In particular, the invention relates to a rotor for a rotating electrical machine, in particular for an electric motor, preferably of a centrifugal pump, comprising a number of radially magnetized permanent magnets arranged along the outer circumference of the rotor in surface recesses of a flux-conducting core which is connected in a rotationally fixed manner to a rotor shaft or is formed by a part of a rotor shaft, wherein the permanent magnets, viewed in cross section, are defined by a circular-arc-shaped outer contour with a radius corresponding to the outer radius of the rotor, and an inner contour such that the radial thickness of the permanent magnets is maximum in their center and decreases in the circumferential direction towards the sides, wherein the maximum radial thickness of the permanent magnets is 65-95% of the maximum radial thickness of the core (3).

[0003] A rotor of this type is known from European patent application EP3451500A1. Furthermore, the prior art includes a wide variety of permanent magnet rotors, in particular, a wide variety of geometries and sizes of isotropic or anisotropic permanent magnets in the rotor. Nevertheless, no geometry is known in the prior art that optimally utilizes the magnetic material used or minimizes the amount of magnetic material used in relation to its utilization.

[0004] For example, rotors with buried permanent magnets are known. These are characterized by the permanent magnets being axially inserted into closed recesses within the core, so that the core material is present between the permanent magnet and the air gap. The geometry of these permanent magnets can, for example, be cuboidal or trapezoidal in cross-section, or have a convex outer and inner contour. In contrast, there are surface-mounted magnets with similar geometries, and in the field of electric motors for centrifugal pump units, such as those used in heating, cooling, or drinking water systems as circulating, charging, or booster pumps, rotors with annular magnets with or without an iron core are frequently used.

[0005] Permanent magnets necessarily have a certain thickness to develop sufficient torque to drive the load. At the same time, the space available in the electric motor is limited, partly by the stator bore and partly by the rotor shaft. This is even more true when replacing an existing rotor, and its dimensions must therefore be taken into account. In rotors with an iron core, the radial core thickness is therefore usually too small to absorb the entire magnetic flux or to conduct the magnetic return path, meaning the amount of magnetic material is not fully utilized.

[0006] To achieve a high efficiency of the electrical machine, a high magnetic flux density B in the iron core is necessary. Due to the saturation of the iron, the flux density B has a physical maximum, which also results in the magnetic flux ϕ = B · A by a cross-sectional area A is limited.A is the cross-sectional area in m 2< whose normal vector is parallel to the flux density vector A or perpendicular to the magnetic flux ϕ. This is in Fig. 1a which shows the distribution of the magnetic field lines for a solid iron core 3 without permanent magnets. The area A results from the radial thickness of the rotor core 3 (outer radius R2 - inner radius R1) multiplied by the axial rotor length, see Fig. 1b . In the case of the solid iron core 3 according to Fig. 1a is the area A = A max maximum. Fig. 1c shows, in contrast, that for buried block magnets 2b the area effectively used for the magnetic return A is smaller than in Fig. 1a , since only the area A 1 between circumferentially adjacent permanent magnets 2b, which lies radially below the permanent magnets 2b, serves for the return path. Thus, the maximum radial thickness of the rotor core 3 is not utilized. Fig. 1d shows that in a ring-shaped magnet 2a with a coaxial iron core 3, the entire radial thickness of the core 3 is defined as an area A 1 is used for the magnetic return, however, this thickness is due to the radial thickness of the ring magnet 2a with the same outer diameter R2 of the rotor as in Fig. 1a und 1c inevitably smaller. The smaller area A 1 < A max The problem with the magnetic return path is particularly noticeable in smaller motors. As a result, the magnetic material, which usually contains rare earth elements and is therefore expensive, is not fully utilized.

[0007] Rotors with ring magnets are also known, which have an internal return path for the magnetic flux and are called Halbach arrays or Halbbach cylinders. These ring magnets do not require an iron core, but are comparatively thick, both to provide the internal return path and to perform the mechanical functions of the iron core. They therefore require a lot of magnetic material, which is why Halbach arrays or Halbbach cylinders are very expensive.

[0008] Another disadvantage of the state of the art is that changing from isotropic ring magnets to anisotropic magnets requires a completely new rotor design due to the different magnetic properties.

[0009] The object of the present invention is to provide a rotor with permanent magnets that has a maximum flux-conducting cross-sectional area A for the magnetic return path in the core and optimally utilizes the magnetic material used for the permanent magnets, while simultaneously saving magnetic material while maintaining the same or even increasing torque. Furthermore, the object of the invention is to provide a rotor with permanent magnets that can replace an existing rotor with a ring magnet while maintaining identical or even better magnetic properties.

[0010] This object is achieved by a rotor having the features of claim 1. Advantageous further developments are mentioned in the subclaims and / or are explained below.

[0011] According to the invention, a rotor for a rotating electrical machine, in particular for an electric motor, preferably of a centrifugal pump, is proposed, comprising a number of radially magnetized permanent magnets arranged along the outer circumference of the rotor in surface recesses of a flux-conducting core which is connected in a rotationally fixed manner to a rotor shaft or is formed by a part of a rotor shaft.

[0012] Viewed in cross-section, the permanent magnets are defined by a circular outer contour with a radius corresponding to the outer radius of the rotor, and an inner contour such that the radial thickness of the permanent magnets is maximum at their center and decreases circumferentially toward the sides. The maximum radial thickness of the permanent magnets is 65-95% of the maximum radial thickness of the core, with the inner contour essentially described by a magnetic equipotential line.

[0013] This ensures that 100% of the flux-conducting surface of the core between its outer and inner radius is usable, meaning the magnetic flux is not limited by the rotor structure. The optimized contour of the permanent magnets fully utilizes the available magnetic material, ensuring no magnetic material is wasted. Overall, magnetic material can be saved compared to the state of the art, allowing even higher torque and thus higher machine performance. Another advantage is that both isotropic and anisotropic magnets can be used.

[0014] Compared to plastic-bonded ring magnets, the invention enables savings of up to 50% in magnetic material. Furthermore, due to the contours of the magnets and the core, demagnetization is eliminated.

[0015] The arrangement of the permanent magnets on the surface of the core also allows them to be easily magnetized in a magnetic field while they are in the rotor.

[0016] Radially magnetized permanent magnets are permanent magnets whose poles lie outside the outer and inner contours. This can involve a purely radial orientation of the magnetization, in which the magnetic moments are aligned radially with respect to the rotor axis. This type of magnetization can be achieved, for example, by hot extrusion. Alternatively, it can involve a quasi-radial orientation of the magnetization, in which the magnetic moments in the center of the permanent magnets are aligned radially with respect to the rotor axis, but the magnetic moments are otherwise parallel to one another. This type of magnetization is achieved, for example, when permanent magnets are magnetized in the rotor.

[0017] Along a magnetic equipotential line, all points of the magnetic field have the same magnetic vector potential A(r).An equipotential line thus connects the geometric locus of all points with the same magnetic vector potential, which a particular point p1 possesses in the nucleus. For example, the equipotential line can be described by the following equation: φ = 1 p arccos 1 − x max D max 1 − R 2 − r D max where φ is the angle of the point in polar coordinates, r is the radius of the point in polar coordinates, p is the number of pole pairs of the rotor, x max is the maximum radial thickness of the permanent magnets, D max is the maximum radial thickness of the core and R2 is the outer radius of the rotor.

[0018] This point is determined by the maximum radial thickness of the permanent magnets. It is thus located on a radius through the center of the permanent magnets at the height where the core and magnet meet. The center of the permanent magnets forms the pole center. Equipotential lines are perpendicular to the magnetic field lines, which describe the direction and course of the magnetic flux. For this reason, the inner contour of the permanent magnets according to the invention is essentially perpendicular to the magnetic field lines, which makes it possible to use both isotropic and anisotropic permanent magnets.

[0019] According to the invention, the outer contour is understood to be the geometry of the outer boundary surface of the permanent magnets that faces the stator of the electric motor surrounding the rotor, or that is opposite this stator. Accordingly, the inner contour is understood to be the geometry of the outer boundary surface of the permanent magnets that lies within the rotor. It faces away from the stator and borders the core.

[0020] The surface recesses of the core ideally have a concave contour that corresponds in shape and size to the inner contour of the permanent magnets in order to achieve a positive fit of the permanent magnets in the surface recesses.

[0021] The core has an outer radius interrupted by surface recesses. In one design variant, the core also has an inner radius, which is determined by a central recess in the core through which the rotor shaft extends.

[0022] In another embodiment, the core is formed by part of a rotor shaft, so that no central recess is required in the core. Here, the core and shaft form a single part. In this embodiment, the surface recesses are therefore formed directly in the rotor shaft. The permanent magnets are therefore also arranged in the rotor shaft. The rotor shaft simultaneously forms the rotor, with the axial section of the rotor shaft lying inside the stator being understood as the rotor within the meaning of the invention. The rotor shaft can have a larger diameter in this section than outside this section or outside the stator. However, it is also possible for the rotor shaft to have the same diameter inside and outside the stator.

[0023] When referring to the core below, it should be noted that in one embodiment, the core may be formed by a part, in particular the section of the rotor shaft located in the stator, so that the terms "core" and "rotor shaft" are interchangeable. In the case of a solid rotor shaft forming the core, the maximum radial thickness of the core is understood to be the (outer) radius of the rotor shaft, i.e., the distance from the shaft axis to the outer circumference of the rotor shaft, so that D max = R 2.

[0024] Due to the inventive geometry of the permanent magnets, the core is cross-shaped or star-shaped in cross-section. Its maximum radial thickness is in the area of ​​the arms extending outwards in a star shape, which corresponds to the difference between the outer and inner radius. This creates the maximum available cross-sectional area for the magnetic flux. Aused in the core. The arms separate two circumferentially adjacent permanent magnets. Two circumferentially adjacent arms are in turn connected by legs.

[0025] The minimum radial thickness of the core is at the center of the limbs, where the permanent magnets have their maximum thickness. The circumferential center of a permanent magnet is considered the pole center.

[0026] Preferably, the permanent magnets are symmetrical, in particular mirror-symmetrical to a central axis of the magnets, which corresponds to the radius in the pole center.

[0027] In a design variant in which the core is connected to the rotor shaft in a rotationally fixed manner, the core can suitably be made of a ferromagnetic material, for example iron. It can be formed in a known manner as a package from a stack of individual sheets, for example from electrical sheets. Alternatively, the core can be made solid, for example in a pressing process from ferromagnetic powder. It is not itself permanently magnetic and serves merely to conduct flux. The rotor shaft can be made of stainless steel or ceramic. In a design variant in which the rotor shaft also forms the core, the rotor shaft or the core can be made of magnetic steel.

[0028] Since the magnetic flux density of a permanent magnet, and thus the torque developed by the rotor, increases with radial thickness, the aim is to make their radial thickness as large as possible. This is certainly possible in the case of a new rotor or electric motor design. However, the minimum core thickness must not be too small to allow for easy manufacturing, for example, by stamping the individual laminations. For a magnet thickness of 98%, for example, the minimum core thickness is only 2%, which may be more difficult to manufacture or could lead to stability problems. In practice, the minimum radial core thickness for a laminated core stack should be at least 1 mm. A compromise must therefore be found between the maximum magnet thickness and the minimum core thickness.To accommodate this compromise, the maximum radial thickness of the permanent magnets can preferably be between 85% and 90% of the maximum radial thickness of the core. The minimum radial thickness of the core is then correspondingly between 10% and 15% of the maximum core thickness.

[0029] If, in another case, the rotor according to the invention is to replace another prior-art rotor, its electromagnetic properties must be taken into account. For example, the maximum radial thickness of the permanent magnets can be dimensioned just so that a rotor equipped with these permanent magnets generates the same electromagnetic force (EMF) or induces the same voltage in the stator as the other prior-art rotor. In such a case, it is quite possible that permanent magnets with a maximum radial thickness of only 65% ​​to 70% will be sufficient.

[0030] By specifying the maximum radial thickness of the permanent magnets, the magnetic potential at the boundary between the permanent magnet and the core is determined, thus also determining the potential on the equipotential line and thus its position and course. This also determines the width of the permanent magnets, i.e., the circumferential angle over which they extend, as well as the corresponding distance between adjacent permanent magnets at their mutually facing circumferential ends, or the width of the arms.

[0031] According to one design variant, the magnetization of adjacent permanent magnets in the circumferential direction can be opposite. This means that north and south poles alternate along the circumference of the rotor. A pole is the geometric location where the magnetic field is strongest. Poles thus only exist at the permanent magnets, so that no pseudopoles form between adjacent permanent magnets. In this case, the magnetic field lines also close within the core between adjacent permanent magnets.

[0032] According to another embodiment, the magnetization of adjacent permanent magnets in the circumferential direction can be identical, so that all magnets have the same pole on the outside and the same opposite pole on the inside. This creates pseudopoles between adjacent permanent magnets with opposite magnetization directions to those of the permanent magnets.

[0033] In particular, the rotor according to the invention has a high efficiency and savings potential for electric motors with a power of less than 1 kW, which is why its use in these electric motors is preferred.

[0034] The outer radius of the rotor is preferably less than or equal to 22.5 mm, since the advantages of simple block magnets prevail for larger rotors. For example, the outer radius can be between 10 mm and 15 mm, preferably approximately 12 mm. Ideally, the maximum radial thickness of the core is between 4 mm and 10 mm, preferably between 5 mm and 7 mm, more preferably approximately 6 mm. The inner radius can also be, for example, between 4 mm and 10 mm, preferably between 5 mm and 7 mm. In particular, the specified dimensions can also be used in any combination.

[0035] To simplify the mechanical production of the inner contour, the equipotential line can be approximated by sections of straight lines and / or circular arcs, so that the inner contour of the permanent magnets does not correspond exactly, but only essentially. A straight line has a correspondingly defined length, a circular arc a defined radius and center point, the latter usually lying on the center axis of the magnets. For example, on each half-side of a permanent magnet, two, three, four or more of the aforementioned sections can approximate one half of the equipotential line. In total, the equipotential line can thus be approximated by three or more sections of straight lines and / or circular arcs, which preferably merge into one another without kinks or offsets in order to obtain a smooth inner contour.

[0036] According to one embodiment, the equipotential line can be approximated from a total of three sections, with a circular arc-shaped section with a radius on the outside in each case and these identical circular arc-shaped sections being connected by a central straight section into which they merge without kinks or offsets. According to another embodiment, the equipotential line can be approximated from a total of five sections, with a first circular arc-shaped section with a first radius on the outside in each case, to which first sections are each adjoined further inwards towards the pole center by a second circular arc-shaped section with a second radius, and these second circular arc-shaped sections being connected by a central straight section, with all sections merging into one another without kinks or offsets.

[0037] According to the invention, the outer radius of the core corresponds to the outer radius of the rotor, so that the periphery of the arms forms a circumferential section of the rotor circumference. In other words, the outer contour of the permanent magnets is then aligned with the circumferential section of the arms of the core.

[0038] To prevent the circumferential ends of the permanent magnets from breaking off during handling, it is advantageous for the permanent magnets to have flattened portions at their circumferential ends. This prevents the inner and outer contours from tapering to a point at the circumferential ends, which could easily cause them to break off. As a result of the flattened portion, the inner contour at the circumferential ends deviates slightly from the equipotential line. Suitably, the circumferential end of the arms can be widened in the circumferential direction by the space created by the flattened portion of the permanent magnets, so that the magnets continue to completely fill the surface recesses with a positive fit.

[0039] Preferably, a cylindrical sleeve can externally encapsulate the core and the permanent magnets, in particular, encapsulate them in a sealed manner. The sleeve holds the surface-mounted permanent magnets in position, preventing them from becoming detached due to centrifugal forces. If the sleeve tightly encapsulates the core and the permanent magnets, it prevents moisture or water from entering the core or between the core and the permanent magnets and causing corrosion. This enables the rotor to be used as a wet rotor. The sleeve can suitably be made of sheet metal, for example, with a thickness of 0.1-0.5 mm, preferably approximately 0.2 mm.

[0040] In one design variant, an annular cavity can exist between the core and the rotor shaft, whereby the core is only connected to the rotor shaft in a rotationally fixed manner via end caps arranged on either side of the axial ends of the core. In the radial direction outwards, the annular cavity is limited by the inside of the core, and inwards by the rotor shaft. This saves weight on the rotor. The power is transmitted between the core and shaft solely via the end caps, which are firmly connected to the shaft and the core, for example by soldering or welding. The annular cavity is preferably sealed so that it forms a closed internal volume. Such a structure is advantageous for a wet-running rotor, particularly in a circulating pump for heating, cooling, or drinking water systems, as it enables leak testing.

[0041] The rotor can preferably have four or six poles. However, other pole numbers are also possible.

[0042] The magnetic material from which the permanent magnets are made can preferably contain rare earth elements to generate a high magnetic flux density. Preferably, plastic-bonded magnetic material can be used and pressed into the magnet contour according to the invention. Alternatively, hot-pressed magnets can be used and formed into the proposed contour by hot extrusion.

[0043] Preferably, the core has no enclosed cavities, thus being free of flow barriers.

[0044] The rotor according to the invention is intended for an electric motor, for example, a synchronous motor. This motor can be suitably controlled in terms of its speed by means of a frequency converter. In one embodiment, the electric motor can be an electronically commutated direct current (BLDC) motor.

[0045] Furthermore, the electric motor can be designed as a wet-running motor, with a can hermetically separating the rotor chamber from the stator. This allows the rotor chamber to be flooded and the rotor to rotate in a liquid. This lubricates the plain bearings supporting the shaft and cools the motor. In this variant, the rotor is a wet-running motor.

[0046] According to a preferred embodiment, the electric motor is part of a centrifugal pump assembly, in which the electric motor drives a centrifugal pump. The present invention therefore also relates to a centrifugal pump, in particular for a heating, cooling, or drinking water system, with an electric motor driving it, wherein the electric motor has a rotor of the type described above.

[0047] If the electric motor is designed as a wet-running motor, the pumped medium of the pump can flow through the rotor chamber.

[0048] Further properties, advantages and features of the invention are explained below using examples and the attached figures.

[0049] It should be noted that, in the context of this description, the terms "have," "comprise," or "include" in no way exclude the presence of other features. Furthermore, the use of the indefinite article for an object does not preclude its plural.

[0050] They show: Fig. 1a: the distribution of the magnetic field lines in an electric motor with an iron core without permanent magnets Fig. 1b: a simplified cross-sectional view of a rotor without permanent magnets Fig. 1c: a simplified cross-sectional view of a rotor with buried permanent magnets Fig. 1d: a simplified cross-sectional view of a rotor with ring magnet and core Fig. 2a: a simplified principle view of a rotor according to the invention in cross section Fig. 2b: perspective view of the rotor according to Fig. 2b with a removed permanent magnet Fig. 3a: Cross-sectional view of a four-pole reference rotor with equipotential lines Fig. 3b: Course of the magnetic vector potential along a radius in the core Fig. 3c: Course of the magnetic vector potential along the rotor circumference Fig. 4a-4d: Cross-sectional views of reference rotors with different numbers of pole pairs with equipotential lines Fig. 5a-5d: Cross-sectional views of four-pole reference rotors with equipotential lines at different inner radii Fig. 6a, 6b: Comparison of the field line distribution for a rotor with a thin ring magnet Fig. 7a, 7b: Comparison of the field line distribution for a rotor with a thick ring magnet Fig. 8a-8d: First embodiment of a rotor according to the invention Fig. 9a-9d: Second embodiment of a rotor according to the invention

[0051] Fig. 2a shows a cross-sectional view of a rotor 1 according to the invention for a rotating electrical machine, in particular for an electric motor 13. This rotor 1 is preferably part of a centrifugal pump assembly, such as is used, for example, as a heating circulation pump. For example, the rotor 1 can be wet-running and thus form the rotor 1 of a canned motor, in particular a centrifugal pump assembly.

[0052] The rotor 1 comprises a number of permanent magnets 2 arranged along the outer circumference of the rotor 1 in surface recesses 7 of corresponding shape and size within a flux-conducting iron core 3. The permanent magnets are thus surface-mounted and form a circumferential section of the rotor 3. This simplifies the magnetization of the magnets 2 when, after installation in the rotor, this is caused by an external electromagnetic field, which, due to its low long-range effect, generally does not penetrate very deeply into the rotor 1.

[0053] In the example after Fig. 2a the number of permanent magnets 2 is four, resulting in a 4-pole rotor 1. The permanent magnets 2 are radially magnetized, so that each permanent magnet 2 has one pole radially outward and the other pole radially inward. The outer poles of circumferentially adjacent permanent magnets 2 are opposite, so that along the circumference at the location of the permanent magnets 2 there is an alternating north pole N and south pole S. There are no pseudopoles between the permanent magnets 2. The permanent magnets 2 are symmetrical with respect to their central axis, with them being thickest in the center and their thickness decreasing towards the sides.

[0054] The core 3 forms a cross-shaped flux guide and can be made from a stack of sheets or solid.

[0055] The permanent magnets 2 can, for example, be glued into the surface recesses 7. To prevent them from detaching from the core 3 due to the high centrifugal forces during the rotation of the rotor 1, they are held in position by a non-magnetic sleeve 10 that surrounds the rotor core 3 and the permanent magnets 2 on the outside. While the sleeve 10 eliminates the need for gluing or other pre-attachment of the permanent magnets, such pre-attachment is useful when handling the rotor, provided the sleeve 10 has not yet been pushed on.

[0056] When the rotor 1 is used in a wet-running electric motor, such as those commonly used in circulation pumps, for example, in heating pumps, the sleeve 10 also serves the purpose of hermetically encapsulating the core 3 and the magnets 2, particularly protecting them from water ingress. If the rotor 1 is used in a dry-running electric motor, such as in dry-running pumps, the sleeve 10 can be omitted. The permanent magnets 2 are then held exclusively by adhesive attachment to the core 3 or by means of a positive fit.

[0057] It should be noted that the sleeve 10 is not taken into account when referring to dimensions within the meaning of the invention, in particular the outer diameter R2 of the rotor, unless otherwise stated. The outer diameter R2 therefore does not include the thickness of the sleeve 10. The sleeve 10 has a thickness of 0.2 mm, for example.

[0058] The core 3 has a central recess 6 that extends coaxially through the core 3. A rotor shaft 11 is located in the recess 6 and is connected to it in a rotationally fixed manner to transmit torque. This can be achieved directly by the core 3 resting with the inner surface of the recess 6 on the outer surface of the shaft 11, for example, in a force-fitting and / or form-fitting manner.

[0059] Alternatively, this can be done indirectly. For this purpose, an air-filled annular space 22 (see Fig. 8d , 9d ), wherein the rotationally fixed connection between shaft 11 and rotor 1 is then effected exclusively via end caps which cover the rotor 1 on its axial end faces and which are firmly connected to the shaft 11 on the one hand and to the core 3 on the other.

[0060] The end caps can be formed by perforated disks, with the shaft 11 passing through a central hole in these end caps. They can, for example, be welded or soldered to the shaft 11, or alternatively, pressed onto it. The end caps also cover the rotor 1 at its axial ends such that the core 3 is hermetically sealed. For this purpose, the end caps can be welded or soldered to the sleeve 10. The cavity 22 has the advantage that the tightness of the rotor 1 can be tested. For this purpose, the rotor is immersed in water, for example, and a negative pressure is created outside the water. Tightness is ensured when no bubbles rise in the water. If this is the case, however, water penetrates into the interior of the rotor and displaces the air in the cavity 22, which can lead to corrosion of the core 3 and the permanent magnets 2.

[0061] The permanent magnets 2 are defined in cross-section by an outer contour 8 and an inner contour 9. The outer contour 8 is convex and described only by a circular arc with a radius R2, which corresponds to the outer radius R2 of the rotor 1, and here also of the core 3. The inner contour 9 is selected according to the invention such that it runs along a magnetic equipotential line 12. Depending on the selection of this equipotential line 12, which is determined by the desired radial thickness x max of the permanent magnets 2 and by the number of poles of the core 3, the inner contour 9 can vary.

[0062] The inner contour 9 is in the design variant according to Fig. 2a also convex, although it can also be completely or partially concave (see e.g. Fig. 4a , 5a, 5b ).

[0063] The inner contour 9 can be simplified by several sections of different geometries that merge seamlessly and preferably without kinks. The more sections used, the more precise the description becomes. The sections can be straight sections of a specific length and / or arc sections of a specific radius and angle. Naturally, the goal is to describe the equipotential line 12 through the sections as precisely as possible in order to achieve this optimal contour as precisely as possible and not deviate from it.

[0064] Thus, the inner contour 9 of a permanent magnet 2 can be Fig. 2a For example, it can be described by a circular arc that is flattened in the central region. Each magnet half is defined in this case by two sections, namely an outer arc section and an inner straight section, wherein the straight sections of the magnet halves merge into one another and become a single straight section that forms the flattened section. Alternatively, each magnet half can also be described by three sections, wherein an outer first section forms a straight line that merges into an arc-shaped second section, which in turn merges into a third, straight-line section that forms one half of the central region of the inner contour 9. The transitions are continuous, i.e. without offset and free of kinks. The description of an equipotential line 12 by individual sections is described below using the Figuren 8a-8c and 9a-9c even deeper.

[0065] The middle area can also be slightly indented with a different equipotential line 12, so that the inner contour 9 is approximately kidney-shaped overall.

[0066] The permanent magnets 2 always have their maximum radial thickness x max at the pole center, where the pole center is the center of the permanent magnets 2 in the circumferential direction. The permanent magnets 2 are mirror-symmetrical to the pole center, so that the central axis 26 of the permanent magnets 2 also coincides with the pole center. The thickness of the permanent magnets 2 decreases continuously in the circumferential direction on both sides. The cross section of the permanent magnets 7, in particular also of the rotor 1, is constant over the axial length, as Fig. 2b to recognize.

[0067] The surface recesses 7 correspond in shape and size to the inner contour 9, so that a permanent magnet 2 fits positively in a surface recess 7, as Fig. 2b illustrated. In this figure, the permanent magnet 2 at the top right has been removed, allowing a clear view of the outer contour of the core 3 in the area of ​​the surface recesses 7.

[0068] The width of the permanent magnets 2 in the circumferential direction results from the course of the equipotential line 12, in particular from where it intersects the outer radius R2. As in Fig. 4a As shown, the permanent magnets 2 extend over a width corresponding to a circumferential angle of 2 φ 2 . Permanent magnets 2 adjacent in the circumferential direction are therefore spaced apart from one another, with a part of the flux-conducting core 3, which is regarded as arm 5, lying between them. In cross-section, the core 3 is cross-shaped or star-shaped due to the geometry of the permanent magnets 2, with the arms 5 extending radially outwards and adjacent arms 5 being connected to one another by a radially inner leg 4 in the approximate form of a ring segment.

[0069] The core 3 has its maximum radial thickness D max in the area of ​​the arms 5, which corresponds to the difference between the outer radius R2 and an inner radius R1, which defines the central recess 6 in the core 3. The radial thickness D max of the arms 5 is thus D max = R 2 − R 1 .

[0070] This means that between adjacent permanent magnets 2 the maximum cross-sectional area A max is available for the magnetic flux, meaning this flux is not limited. The only limiting factor is the outer diameter R2 of rotor 2 and the thickness D max of core 3.

[0071] The permanent magnets 2 are magnetized radially. This means that the magnetic north and south poles are located radially outside and radially inside, with the maximum field strength in the center of the poles, since the magnets 2 are thickest here. Neighboring permanent magnets 2 in the circumferential direction are magnetized in opposite directions. Thus, the north and south poles alternate along the circumference from magnet to magnet. The direction of the magnetic flux ϕ is in Fig. 2a simplified by arrows. The magnetic field lines close almost homogeneously in the core 3 by passing from one permanent magnet 2 to the adjacent permanent magnet 2 through the arm 5 and perpendicular to the cross-sectional area A max get lost.

[0072] To define the inner contour 9, the magnetic vector potential A(r) of the field in rotor 1. This is determined by Fig. 3a-3c This is a mathematical auxiliary quantity in the form of a vector field A(r) whose rotation generates a second vector field that determines the magnetic flux density B(r) For this purpose, a reference rotor 1 with an iron core 3 without permanent magnets 2 (solid iron rotor) is considered, which is penetrated by the magnetic field of the stator of the electric motor 14, within which the rotor 1 is located, see Fig 3a . This Fig. 3a also shows the course of the field lines 19 of the magnetic field. Although a solid iron rotor 1 is used for the analysis, the course of the field lines 19 does not change due to the integration of permanent magnets 2 according to the invention into this rotor 1, as Fig. 6b and 7b show.

[0073] Along a radius r of the core 3, the potential increases A(r) from the inner radius R1 to the outer radius R2 linearly, as Fig. 3b This can be described mathematically as follows: A r , φ = 0 = A max ⋅ 1 − R 2 − r R 2 − R 1 , mit r ∈ R 1 , R 2 .

[0074] The magnetic vector potential A(r) is thus zero at the inner radius R1 and within the recess 6, A(r≤R1)=0. Radially outside the core 3, ie at the level of the outer radius R2, the potential is, however, maximum, A(r=R2)=A max .

[0075] Furthermore, the magnetic vector potential A(r) along the outer circumference of the rotor 1 sinusoidal, as Fig. 3c using a four-pole stator field. Thus, the potential A(r=R2) along the circumference with increasing circumferential angle φ sinusoidal between the positive and negative maximum value A max , at which the actual poles S, N are located. The zero angle φ For simplicity, = 0 is assumed to be the positive maximum value A max. The potential distribution at the outer radius R 2 can be mathematically described for a rotor with any number of pole pairs p as follows: A r = R 2 , φ = A max ⋅ cos p ⋅ φ ,

[0076] Where p is the number of pole pairs, i.e., 2p is the number of poles of the stator field. In the case of a four-pole magnetic field, p = 2.

[0077] The magnetic vector potential A ( r , φ ) of a point p at any geometric location with the coordinates p(r, φ ) can be given according to equation Gl.4. A r φ = A max ⋅ 1 − R 2 − r R 2 − R 1 ⋅ cos p ⋅ φ ,

[0078] If we now consider a permanent magnet 2 with the maximum radial thickness x max , it extends from the outer radius R2 on the radius r to the point p1, see Fig. 3a . In polar coordinates, this point p1 has the coordinates p1(r = r1, φ = φ 1 = 0), where r1 = R 2 - x max Furthermore, this point p1 can be assigned a magnetic vector potential A(p1) be assigned as follows: A p 1 = A max ⋅ 1 − R 2 − r 1 R 2 − R 1 = A max ⋅ 1 − x max D max

[0079] Since the distance between outer radius R2 and inner radius R1 corresponds to the maximum core thickness D max =R2-R1, the rational expression R 2 − r 1 R 2 − R 1 = x max D max the ratio of the magnet thickness x max to the core thickness D max . For example, if the magnet thickness x max is 90% of the core thickness D max , the potential at point p1 is 0.1· A max . At 80% of the core thickness D max , the potential at point p1 is 0.2· A max .

[0080] Since the potential A at the circumference fluctuates sinusoidally, a point p2 on the circumference can be found where the magnetic vector potential is identical to the vector potential at point p1. This is shown in Fig. 3a at a circumferential angle φ 2 the case. This point has the polar coordinates p2(r = R2, φ = φ 2 ) and can be derived from the requirement A(p1) = A(p2): A max ⋅ 1 − x max R 2 − R 1 = A max ⋅ cos p ⋅ φ 2 φ 2 = 1 p ⋅ arccos 1 − x max R 2 − R 1

[0081] In the case of a magnet thickness x max of, for example, 90% of the core thickness D max , the circumferential angle φ 2 with a four-pole stator field thus φ 2 = 1 2 ⋅ arccos 0 1 = 42 , 13 ° . Consequently, the width of a permanent magnet 2 corresponds to a circumferential angle of 2 · φ 2 = 84.26°, which in turn corresponds to a circumferential length U PM of U PM = 2 φ 2 ⋅ 2 πR 2 360 ° = 1 , 47 R 2 This also determines the distance between two adjacent permanent magnets 2, ie, the width Us of an arm 5 in the circumferential direction, and in the case under consideration is 5.74°. In general, the width Us of the arms 5 can be specified as follows: U s = 2 ⋅ 360 ° 2 ⋅ 2 p − φ 2 ⋅ 2 πR 2 360 ° = 1 p − φ 2 90 ° πR 2 .

[0082] For the considered case of a magnet thickness x max of, for example, 90% of the core thickness D max, the arm width Us = 0.1R 2 .

[0083] The two points p1 and p2 can be connected by a line along which all points have the same magnetic vector potential as the points p1 and p2. This line forms one half of the equipotential line 12a associated with the magnet thickness x max, which is shown in Fig. 3a for one half of the sector section under consideration of a pole of the rotor 1. A corresponding second half of the equipotential line 12a lies mirror-symmetrically on the opposite half of the sector section under consideration and is shown there in dashed lines. In this example, this equipotential line 12a forms the inner contour 9 of the permanent magnet 2, which is shown for illustration purposes in the clockwise adjacent sector section of the next pole.

[0084] From the requirement that the potential A(r)on the line 12a is identical everywhere, the following equation 11 follows, which mathematically describes the course of the equipotential line 12a: A p 1 = A r φ A max ⋅ 1 − x max R 2 − R 1 = A max ⋅ 1 − R 2 − r R 2 − R 1 ⋅ cos p ⋅ φ φ = 1 p arccos 1 − x max R 2 − R 1 1 − R 2 − r R 2 − R 1 = 1 p arccos 1 − x max D max 1 − R 2 − r D max where φ is the circumferential angle for any point on the equipotential line r is the radius for this arbitrary point on the equipotential line p is the number of pole pairs R2 is the outer radius of the rotor R1 is the inner radius of the rotor D max is the maximum radial thickness of the core and x max is the maximum radial thickness of the permanent magnet.

[0085] How Fig. 3a As can be seen, mathematically there are a multitude of equipotential lines 12, 12a running alongside one another. The equipotential line 12a used for the inner contour 9 according to the invention results from the determination of the thickness x max of the permanent magnets 2 for given rotor dimensions.

[0086] When designing a new electric motor, the rotor dimensions can generally be chosen freely. However, if the rotor 1 according to the invention replaces an existing rotor with a different magnet geometry, the rotor dimensions are inevitably predetermined by the rotor. Since the magnetic flux density of a permanent magnet 2 increases with its thickness in the direction of magnetization, the aim is to select the largest possible maximum thickness x max. This enables maximum torque to be maximized.

[0087] Nevertheless, a certain minimum thickness of the legs 4 must be taken into account in order to be able to produce the core 3, since the core 3 has its minimum radial thickness D min =R2-R1-x max in the area of ​​the legs 5.

[0088] Furthermore, when replacing an existing prior art rotor with a rotor 1 according to the invention, it is advantageous if the voltage induced by the rotor in the stator 14 is identical, so that in the case of vector control of the motor, this voltage does not need to be changed. In this case, the maximum thickness x max of the permanent magnets 2 can be selected such that the rotor 1 induces an identical voltage in the stator 14 as the rotor to be replaced.

[0089] To cover all applications, according to the invention, the maximum radial thickness x max of the permanent magnets 2 is selected to be between 65% and 95% of the maximum radial thickness D max of the core 3. Preferably, it is between 80% and 90% to achieve a compromise between magnet thickness and core thickness.

[0090] Depending on the number 2p of poles of the magnetic rotor field to be realized and the ratio of outer radius R2 to inner radius R1, the equipotential lines 12 extend along a wider (with few poles) or narrower (with many poles) bolt circle segment. Furthermore, the shape of the equipotential lines 12 also varies depending on the poles and the core thickness.

[0091] Furthermore, the example of the four-pole rotor in Fig. 3a It can be seen that the equipotential lines 12 become increasingly flattened in their central region with increasing proximity to the inner radius R1 and even change from a convex shape radially further out to a concave shape radially further in, this concave shape being characterized by the fact that the central region is indented.

[0092] Fig. 4a bis 4d and 5a bis 5d show different rotor designs and their equipotential lines. Fig. 4a a two-pole rotor 1, Fig. 4b the four-pole rotor according to Fig. 3a , Fig. 4c a six-pin and Fig. 4a an eight-pole rotor. Inner radius R1 and outer radius R2 are shown in the Fig. 4a-4d identical.

[0093] Fig. 5a bis 5d Using the example of four-pole rotors, show the influence of the rotor radii on the equipotential line. Fig. 5a is the inner radius R1 compared to Fig. 3a , 4b reduced, in Fig. 5b It can be seen that the equipotential lines 12 in the direction from the outer radius R2 to the inner radius R1 exhibit a concave central region much earlier. They are compressed by the enlargement of the inner radius R1. Fig. 5c is the inner radius R1 compared to Fig. 3a , 4b reduced, in Fig. 5d further reduced. The equipotential lines 12 are stretched towards the rotor axis 21 by reducing the inner radius R1. Regardless of the compression or expansion, however, the point at which a respective equipotential line 12 intersects the circumference remains the same, so that the distance Us (cf. Fig. 3a ) between adjacent permanent magnets 2 is independent of the inner and outer radius R1, R2 in the angular dimension, but depends solely on the magnet thickness x max.

[0094] Fig. 6a und 6b as well as Fig. 7a und 7b show the field line pattern and the flux density distribution in grayscale in a three-phase electric motor 13 with a rotor 1 with ring magnet 2a according to the prior art ( Fig. 6a , 7b ) in comparison to a rotor 1 with magnets 2 according to the invention, the inner contour 9 of which lies essentially on an equipotential line 12a or is essentially described by such a line.

[0095] The electric motor 13 is an electronically commutated synchronous motor controlled by a frequency converter and, in a conventional manner, has a stator 14 with six coils 15 made of copper wire, which are wound on coil carriers 15a and each surround a pole piece 18 that extends radially through a coil 15. Radially outward, all pole pieces 18 are connected to a return ring 16. At the radially inner end of the pole pieces 18, they widen in the circumferential direction on both sides to form a pole piece 17, which is radially opposite the rotor 1, with an air gap 20 existing between the rotor 1 and the pole piece 17. Field lines 19 illustrate the course of the magnetic field for an exemplary moment of stator current supply.

[0096] Fig. 6a shows an embodiment with a comparatively thin ring magnet 2a without iron core, which is formed by the Fig. 6b shown rotor 1 can be replaced with the magnet geometry according to the invention. Fig. 7a shows an embodiment with a comparatively thick ring magnet 2a without iron core, which is formed by the Fig. 7b The rotor 1 shown can be replaced with the magnet geometry according to the invention. As a comparison of Fig. 6a with Fig. 6b as well as Fig. 7a with Fig. 7b As shown in each case, the courses of the field lines 19 in the rotor 1 with magnets 2 according to the invention are essentially identical to the field line course in a rotor with a ring magnet 2a. This is achieved precisely by the special inner contour 9 of the permanent magnets 2, because it is selected so that it lies perpendicular to the field lines of the magnetic flux. At the same time, however, between 29% (in the case of the thin ring magnet 2a) and 50% (in the case of the thick ring magnet) less magnetic material is required for the rotor 1. This is due to the fact that flux-conducting regions inside the core 3 replace magnetic material of the ring magnet 2a in regions that only serve the magnetic return path in the rotor 1. The invention thus enables the use of a minimal amount of magnetic material with identical or even better magnetic properties, in particular with the same or even increased torque.

[0097] As examples of specific dimensions of a rotor 1 according to the invention, according to a first embodiment, for example, an inner radius R1 of 5.615 mm and an outer radius R2 of 11.615 mm can be used, resulting in a core thickness Dmax of 6 mm. According to a second embodiment, the rotor 1 can be manufactured with an inner radius R1 of 6.5 mm and an outer radius R2 of 12.4 mm, resulting in a core thickness Dmax of 5.9 mm. According to a third embodiment, the rotor 1 can be manufactured with an inner radius R1 of approximately 10 mm and an outer radius R2 of approximately 19.5 mm, resulting in a core thickness Dmax of 9.5 mm.

[0098] Fig. 8a-8d and 9a-9e each show an embodiment of a rotor 1 according to the invention according to the first embodiment, wherein Fig. 8a , 9a the core 3 in cross section, Fig. 8b , 9b the core 3 in perspective view, Fig. 8c , 9ca permanent magnet 2 and Fig. 8d , 9d show the entire rotor 1 in cross section. Fig. 9a' shows that in Fig. 9a circled detail B in enlarged view.

[0099] As shown by the Figuren 8d and 9dAs can be seen, the rotor 1 has four permanent magnets 2 which are arranged in a form-fitting manner in surface recesses 7 of the core 3, which has a cross-shaped or star-shaped cross-section, so that the outer contour 8 of the magnets 2 is aligned with the outer circumferential section of the arms 5 of the core 3. For this purpose, the magnets 2 have an outer contour 8 which corresponds to an arc section with the outer radius R2 of the core 2. A sleeve 10 tightly encapsulates the magnets 2 and the core 3. A shaft 11 extends through a central recess 6 with the inner radius R1 of the core 3, with an annular gap 20 existing between the shaft 11 and the core 3. End caps (not shown) connect the rotor 1 to the shaft 11 in a rotationally fixed manner. Such a rotor 1 is specially designed for wet running, in particular in a centrifugal pump unit such as a circulating pump for heating or cooling systems.

[0100] In the first embodiment, the permanent magnets 2 have a maximum axial thickness x max , which corresponds to 2 / 3 or approximately 66% of the maximum radial core thickness D max = R2-R1. The inner contour 9 of the permanent magnets 2 runs essentially along an equipotential line 12 of the magnetic field in the core 3, which connects the geometric location of all points of the same magnetic vector potential, which the point p1 has at 1 / 3 of the core thickness in the pole center, i.e. along the central axis 26 of the magnets 1. In other words, the equipotential line 12 connects all points with a potential A(r) of 1 / 3· A max .

[0101] In the second embodiment, the permanent magnets 2 have a maximum axial thickness x max of approximately 85.25% of the maximum radial core thickness D max . The inner contour 9 of the permanent magnets 2 thus runs essentially along an equipotential line 12 of the magnetic field in the core 3, which connects the geometric location of all points that form the magnetic vector potential A(r) of 0.1475· A max have.

[0102] In practice, a mathematically exact geometry of the equipotential line 12 is not easy to achieve. To generate the inner contour 9, an approximation of the equipotential line 12 is used, consisting of segments of straight lines of a specific length and arcs of specific radii and angles, which merge seamlessly into one another without offset or kinks.

[0103] How Fig. 8c As shown, in the first exemplary embodiment, the inner contour 9 is formed by two sections 23b, 24b relative to one half of the permanent magnet 2, with a first, laterally outer section 23b being formed by a circular arc with a radius R3 and a center point beyond the outer contour of the magnet 2. The radius R3 is, for example, 8.3 mm. Adjoining this, without offset or kinks, is a second section 24b in the form of a straight line with a length L1. The length L1 is 1.5 mm. This second section 24b is part of the central region of the permanent magnet 2.Since the magnet 2 is seal-symmetrical to its central axis 26, the other half of the magnet 2 also has a corresponding first and second section 23b, 24b, wherein the second, straight section 24b of one half-side merges into the second, straight section of the other half-side without offset and without kinks, so that the inner contour 9 as a whole can be regarded as being described by three sections which approximate the corresponding equipotential line 12a.

[0104] Out of Fig. 8a It can be seen that the surface recesses 7 in the core 3 have a shape corresponding to the inner contour 9 of the magnets 2 according to Fig. 8c corresponding shape, which is formed per half side from two sections 23a, 24a, which fit well in Fig. 8b can be seen. For this purpose, the core 3 has an outer contour in the area of ​​the surface recesses 7 with a first, laterally outer section 23a in the shape of a circular arc with radius R3 and a center point beyond the core 3, and with a second section 24a in the shape of a straight line with a length L1, which adjoins the first section 23a without offset or kinks. Here, too, the two straight sections 24a of the half-sides of the surface recesses 7 merge into one another without offset or kinks, so that a continuous contour is present.

[0105] How Fig. 9c shows, in the case of the second embodiment, the inner contour 9 is formed by three sections 23b, 24b, 25b with respect to one half-side of the permanent magnet 2. A first, laterally outer section 23b is formed by a circular arc with a radius R5 and a center point beyond the outer contour of the magnet 2. The radius R5 is, for example, 17.5 mm. Furthermore, here too there is a second section 24b in the form of a straight line with a length L2, which defines the central region of the inner contour 9 of the permanent magnet 2. The length L2 is 2.85 mm. Between the first and second sections 23b, 24b there is a third section 25b, which is also formed by a circular arc, but with a smaller radius R4 than the first section 23b, and with a specific angle α. The radius R4 is, for example, 4.45 mm. The angle α is 32°.The circular arc of the third section 25b also has a center point beyond the outer contour of the magnet 2. All three sections 23b, 24b, 25b merge into one another without offset or kinks. Since the magnet 2 is also seal-symmetrical to its central axis 26 here, the other half of the magnet also has a corresponding first, second, and third section 23b, 24b, 25b, with the second, straight section 24b of one half merging into the second, straight section of the other half merging without offset or kinks, so that the inner contour 9 as a whole can be considered as described by five sections that approximate the corresponding equipotential line 12a.

[0106] Out of Fig. 9a It can be seen that the surface recesses 7 in the core 3 have a shape corresponding to the inner contour 9 of the magnets 2 according to Fig. 9c corresponding shape, which is formed per half side from three sections 23a, 24a, 25a, which fit well in Fig. 9b can be recognized. Fig. 9a' shows these sections 23a, 24a, 25a in cross section by showing the detail B in Fig. 9a enlarged. The core 3 has, in the area of ​​the surface recesses 7, an outer contour with a first, laterally outer section 23a in the shape of a circular arc with the radius R5 and a center point beyond the core 3, with an inner second section 24a in the shape of a straight line with the length L2 and with a third section 25a lying between these sections 23a, 24a in the shape of a circular arc with the radius R4, the angle α and a center point beyond the core 3. The first section 23a merges into the third section 25a without offset or kinks and this merges into the second section 24a without offset or kinks. Furthermore, here too the two straight sections 24a of the two half-sides of the surface recesses 7 merge into one another without offset or kinks, so that a continuous contour is present.

[0107] Due to their geometry, the permanent magnets 2 can be either isotropic or anisotropic. A known material can be used as the magnet material for the permanent magnets 2, for example, ferrite or a metal alloy containing rare earth elements, such as a neodymium-iron-boron (NdFeB) alloy.

[0108] The permanent magnets can be manufactured using a process known to those skilled in the art. For example, the magnets can be produced by sintering or forming (hot extrusion) hot-pressed magnets into the inventive shape, which is easily possible. Alternatively, plastic-bonded magnets can be used, which are either also mechanically formed into the inventive shape, or extruded or injection-molded directly into this inventive shape.

[0109] It should be understood that the foregoing description is provided merely by way of example for illustrative purposes and in no way limits the scope of the invention. Features of the invention stated as "may," "exemplary," "preferred," "optional," "ideal," "advantageous," "optionally," or "suitable" are to be considered purely optional and do not limit the scope of protection, which is defined exclusively by the claims. To the extent that the foregoing description mentions elements, components, process steps, values, or information that have known, obvious, or foreseeable equivalents, these equivalents are encompassed by the invention.Likewise, the invention includes any changes, variations or modifications of embodiments that involve the replacement, addition, change or omission of elements, components, method steps, values ​​or information, as long as the basic idea of ​​the invention is retained, regardless of whether the change, variation or modifications lead to an improvement or deterioration of an embodiment.

[0110] Although the above description of the invention mentions a multitude of physical, non-physical, or method-related features in relation to one or more specific embodiments, these features can also be used in isolation from the specific embodiment, at least as long as they do not require the mandatory presence of further features. Conversely, these features mentioned in relation to one or more specific embodiments can be combined with each other as desired, as well as with other disclosed or undisclosed features of shown or not shown embodiments, at least as long as the features correspond to the scope of the appended claims. Bezugszeichenliste

[0111] 1 Rotor 2 Permanent magnets 2a Ring magnet 2b Block magnet 3 Core 4 Leg 5 Arm 6 Recess 7 Surface recess 8 Outer contour 9 Inner contour 10 Sleeve 11 Shaft 12 Equipotential line 12a Selected equipotential line 13 Electric motor 14 Stator 15 Coils 15a Coil carrier 16 Return ring 17 Pole shoe 18 Pole web 19 Field lines 20 Air gap 21 Rotor axis 22 Hollow annular space 23a First section of the inner contour of the surface recess 23b First section of the inner contour of the permanent magnet 24a Second section of the inner contour of the surface recess 24b Second section of the inner contour of the permanent magnet 25a Third section of the inner contour of the surface recess 25b Third section of the inner contour of the permanent magnet 26 Center axis of the magnets

Claims

1. Rotor (1) for a rotating electric machine, in particular for an electric motor (13), comprising a number (2p) of radially magnetized permanent magnets (2), which are arranged along the outer circumference of the rotor (1) in surface recesses (7) of a flux-conducting core (3), which is non-rotatably connected to a rotor shaft (11) or is formed by a part of a rotor shaft, wherein the permanent magnets (2), viewed in cross-section, are defined by a circular arc-shaped outer contour (8) with a radius corresponding to the outer radius (R2) of the rotor (1), and an inner contour (9) such that the radial thickness (xmax) of the permanent magnets (2) is maximal in their centre and decreases in circumferential direction towards the sides, wherein the maximum radial thickness (xmax) of the permanent magnets amounts to 65-95% of the maximum radial thickness (Dmax) of the core (3), characterized in that the inner contour (9) is essentially described by a magnetic equipotential line (12) and the outer radius of the core (3) corresponds to the outer radius (R2) of the rotor (1).

2. Rotor (1) according to claim 1, characterized in that the maximum radial thickness (xmax) of the permanent magnets (2) amounts to 65-70% or 85-90% of the maximum radial thickness (Dmax) of the core (3).

3. Rotor (1) according to claim 1 or 2, characterized in that the magnetization of adjacent permanent magnets (2) is opposite.

4. Rotor (1) according to one of the preceding claims, characterized in that its outer radius (R2) is less than or equal to 22.5mm, preferably between 10mm and 15mm, in particular approximately 12mm.

5. Rotor (1) according to one of the preceding claims, characterized in that the maximum radial thickness (Dmax) of the core (3) is between 4mm and 10mm, preferably between 5mm and 7mm, in particular approximately 6mm.

6. Rotor (1) according to one of the preceding claims, characterized in that the equipotential line (12) is approximated by sections of straight lines and / or circular arcs.

7. Rotor (1) according to one of the preceding claims, characterized in that the equipotential line (12) is approximated by three or more sections of straight lines and / or circular arcs, which transition into each other without a kink or offset.

8. Rotor (1) according to one of the preceding claims, characterized in that the equipotential line (12) is described by the following equation, φ = 1 p arccos 1 − x max D max 1 − R 2 − r D max where φ is the angle of a point in polar coordinates r is the radius of a point in polar coordinates p is the pole pair number of the rotor xmax is the maximum radial thickness of the permanent magnets Dmax is the maximum radial thickness of the core, and R2 is the outer radius of the rotor.

9. Rotor (1) according to one of the preceding claims, characterized in that the permanent magnets (2) have flattenings at their circumferential ends.

10. Rotor (1) according to one of the preceding claims, characterized in that a cylindrical sleeve (10) encapsulates the core (3) and the permanent magnets (2) externally, in particular in a sealed manner.

11. Rotor (1) according to one of the preceding claims, characterized in that it is a wet runner.

12. Rotor (1) according to one of the preceding claims, characterized in that there is an annular, in particular sealed, cavity (20) between the core (3) and the rotor shaft (6), and the core (3) is fixed in rotation to the rotor shaft (11) only via end caps arranged on both sides of the axial ends of the core (3).

13. Rotor (1) according to one of the preceding claims, characterized in that it has four-poles or six-poles.

14. Centrifugal pump, in particular for a heating, cooling or drinking water installation, with an electric motor (13) driving it, characterized in that the electric motor (13) comprises a rotor (1) according to one of the preceding claims.