Magnetic bearings

【0001】<u style="single">Field of invention</u>The present invention relates to passive and active magnetic bearings, and more particularly to, but is not limited to, small, highly rigid, and large load capacity magnetic bearings. [0002]<u style="single">Background of the invention</u>Magnetic bearings --- Purpose and characteristics The purpose of magnetic bearings is to provide force between two main bearing members without causing contact. Therefore, this force is called bearing capacity. Continuing with the usual definition of any bearing, magnetic bearings also allow free movement in one or more orientations, while providing the ability to exert bearing capacity in at least one other orientation. Many magnetic bearings are used in rotating machines that separate the rotor from the stator. Magnetic bearings have the advantages of extremely low rate of energy loss (when properly designed), potentially extremely long life due to no contact between parts, and the ability to withstand relatively high temperatures. doing. [0003] Magnetic bearings may be active or passive. The active magnetic bearing senses the relative position of the two main bearing members and adjusts the current flowing through the coil so that the net force between the two main bearing members has the appropriate magnitude and direction. Passive magnetic bearings usually involve a magnetic field from a permanent magnet, but can instead be constructed using a coil of conductor that provides a magnetomotive force (MMF). However, the current through these coils is not a strong function of the relative positions of the two bearing members. Passive magnetic bearings often operate on the basis of the repulsion of the same pole. [0004] From a simple point of view, active magnetic bearings can be of any stiffness in the sense that a finite amount of force can be generated by the least amount of relative motion between the two main bearing members. For this, there are limitations associated with the ability to detect extremely small motions and the need to stabilize the closed-loop control system. However, it is widely accepted that the stiffness of active bearings is generally several orders of magnitude greater than the stiffness of equivalent passive bearings. The rigidity of the bearing is also due to the acceptable dynamic properties of the bearing and so that the relative positions of the two main bearing members are not affected by the externally applied load existing between the two main bearing members. It is extremely important to make it. [0005] Another vital property of any bearing is reliability. Active magnetic bearings are complex systems that require sensing, control and power current. Therefore, there are numerous possible failure modes other than simple mechanical failure. In contrast, passive bearings tend to be extremely rugged and reliable, with very few possible failure modes other than mechanical failure. [0006] The main characteristic of all magnetic bearings is size. A second property related to size and equally important is gross weight. It is widely accepted that for a given force rating, the size of a radial magnetic bearing is many times larger than its rotating element equivalent. [0007] Basic bearing area and central surface Consider a bearing that has two main bearing members and provides some free relative motion between the two main bearing members. In many cases, at least one of the free relative motions is rotation. Rotation of any physical object on a scale beyond the atomic scale involves translation of the particles on the surface of the object. In order to provide a bearing that produces a force that resists the relative motion of the main bearing member in some orientations but can tolerate free rotation around any axis, the translation must be along at least one axis. It is necessary and sufficient to provide a region in the bearing that is opposite and free along at least one other axis. Such a region is called a basic bearing region. [0008] For the sake of explanation, conventional ball bearings are useful. Ball bearings are a finite number of basic bearings in which the relative translations of the inner and outer races are strongly resisted along one direction and the translations of the inner and outer races are free along the other two directions. There is one area for each ball. The resistance direction of an individual ball at a given moment is along the diameter (of the ball) between the contacts. Obviously, this simple conceptual model ignores friction and viscous shear forces at the contacts. FIG. 1 shows the basic bearing area of ball bearings. [0009] The collective action of all these basic bearing regions in the case of ball bearings provides free rotation around one axis, but all net translations and (angular contacts) between the two main bearing members. (In the case of ball bearings) there are bearings that repel rotation around the other two orthogonal axis of rotation. [0010] The same view can be applied to cylindrical roller bearings. Each roller can generate extremely strong resistance to relative translation of the internal and external races in one direction, allowing extremely free movement in the direction of rotation. The rollers provide some axial resistance to the relative motion of the inner and outer races, but this resistance is usually not used. In the case of cylindrical roller bearings, it can be considered that there is one basic bearing area for each roller. In the case of roller bearings with conical rollers, each roller is considered to have a very large number of disc-shaped slices and represents the basic bearing region. Figure 2 shows the basic bearing area from tapered roller bearings. [0011] It is easy to extend this view of all bearings that facilitate rotation to hydrostatic and hydrodynamic bearings. In the case of hydrostatic bearings, the basic bearing area can be considered as a separate location where the pressurized fluid is supplied to the cavity between the two main bearing members. FIG. 3 shows the basic bearing region from the hydrostatic bearing, and shows the pressure distribution for such individual positions in an overlapping manner. In the case of hydrodynamic bearings, the lubricant intermediate layer between the two main bearing members can be disassembled into patches that exert the force to maintain the distance between the two main bearing members. FIG. 4 shows the direction of relative motion between one such patch and bearing member. [0012] According to the above logic, all bearings have a plurality of sets of basic bearing regions having at least one direction of relatively free relative translation and at least one direction that strongly repels (or can strongly repel) the translation. Can be disassembled into. [0013] In each of the above bearing examples, the basic bearing region includes a portion of the surface of each of the two main bearing members. There is a central surface between these two surfaces. This central surface is on any smooth surface so that the action of the bearing region in providing the direction of free translation can be considered equivalent to slipping of one surface of this central surface against the other. is there. The term central surface is used many times throughout the rest of the specification. [0014] Often, the basic bearing area is used only (or primarily) to provide free translation in one direction. This direction is in the plane of the central surface. Therefore, according to FIG. 5, a set of axes in the main direction with respect to the basic bearing region can be established. In Figure 5, the three axes are (1) (Main) Axis of free relative translation. For clear practical reasons, free relative translation is similar to pure discrete shearing action on the central surface. This direction is arbitrarily designated by the symbol x in FIG. [0015] (2) Axis perpendicular to the central surface. This direction is arbitrarily designated by the symbol z in FIG. (3) The remaining orthogonal axes with the direction y in Fig. 5. Is In all of the above cases, the force acting between the two surfaces of the two main bearing members is primarily the perpendicular to the central surface, that is, in the direction along the z direction of FIG. [0016] No practical bearing on a scale larger than the atomic scale is truly lossless. Ball bearings and roller bearings have some rotational resistance. There is some viscous resistance in the bearing fluids of hydrostatic and hydrodynamic bearings. The magnetic bearing has an eddy current loss and a hysteresis loss. Therefore, in all cases, there is always some component of force acting to oppose the relative translation of the two surfaces in the "free" direction x. [0017] Magnetic stress in magnetic bearings Many of the existing designs of magnetic bearings rely heavily on the fact that tensile Maxwell stresses are effectively present in the air in the direction of the magnetic flux lines when the magnetic flux is passed through the air. Many, but not all, active magnetic bearings currently available operate directly on this tensile stress. [0018] Figure 6 shows perhaps the simplest example of the action of tensile Maxwell stress, in which a horseshoe-shaped permanent magnet drives a magnetic field through itself and the air gap (twice) and some second object. Since the magnetic flux lines in this example are mainly perpendicular to the surface of the horseshoe-shaped permanent magnet and the surface of the second object, the net force generated at each of the two air gap intersections should be approximated using a simple mathematical formula. Can be done. By combining these two discrete forces using basic trigonometry, we can then generate an equation for the resulting total attractive force between the magnet and the second object. [0019] The oldest designs of active magnetic bearings include multiple individually urged horseshoe electromagnets arranged around the circumference of an air gap centered on a solid (or hollow) cylindrical rotor. Each horseshoe-shaped electromagnet has its own complete magnetic circuit, with little interaction between the individual electromagnets. In normal operation, each electromagnet has a bias magnetic field such that some magnetic flux is always present through the horseshoe-shaped electromagnet. The bias magnetic field may be provided by a DC component of the current flowing through the electromagnet, but can be provided by a permanent magnet in a magnetic circuit. The forces generated by the bias magnetic field are usually nearly zero in total. Therefore, a relatively small amount of (additional) current is passed through one horseshoe-shaped electromagnet, and a negative current of this (additional) current is passed through the horseshoe-shaped electromagnets facing in the radial direction thereof, thereby causing the bearing stator and the bearing rotor. A net lateral force is generated between and. [0020] Modern designs of magnetic bearings utilize a stator shape similar to that of a switchable reluctance machine in that inwardly projecting stator poles are mounted on a continuous cylinder of back irons. There is a coil on each stator pole, or the coil is connected to two or more poles. Instead, a coil is formed around the rear of the core according to the older Gram ring winding method that was common in electromechanical machines until a few years ago. Permanent magnets may be provided on the stator poles or on the back iron cylinders to generate a bias magnetic field. In such cases, the relationship between the individual currents flowing through the coil (or phase) and the amount of magnetic flux passing through the individual stator poles is more complex than a simple array of multiple independent horseshoe electromagnets. However, the basic principle of operation is the same, and the attractive force per magnetic pole is (generally) proportional to the square of the total magnetic flux passing through the magnetic pole surface. [0021] [0021] Many magnetic-mechanical devices are basically limited by magnetic flux density. The magnetic flux density of any machine containing iron rarely exceeds 2 Tesla in any part of iron due to saturation. (The term iron is used here to include any ferromagnetic material.) The maximum magnetic flux density of a ferromagnetic material is an important parameter for selecting a material for its application, but that's it. is not it. Mechanical strength, stiffness, resistivity (against eddy current loss) and low magnetic hysteresis effect are other properties that designers must keep in mind when selecting materials for use in magnetic-mechanical devices. After all, it is not surprising that there is no maximum magnetic flux density in iron or any other material, but the (incremental) relative permeability to iron is from greater than 1000 to more than 2 tesla at low flux levels. It falls within the range up to a value slightly above 1 at the level. [0022] The magnetic flux density of iron in iron-containing magnetic-mechanical devices is always higher than the magnetic flux density in the air gap where the magnetic flux effectively produces force. The term air gap is used herein to mean a region of space filled or unfilled with ferrofluid. This usage is consistent with the interpretation of terms in the context of electromechanical. Most commonly, the gaps between the relative movable parts of the device are occupied by air. [0023] If the magnetic flux density of the air gap is limited, then the magnitude of the achievable Maxwell stress is also limited. The net force or torque acting through the air gap can be calculated by selecting any surface through the air gap and integrating the magnetic stress with respect to the surface. If this were done, the average effective air gap stress would be the result of dividing the total force by the total air gap area, or the total torque divided by the total first moment of the air gap area around the axis of rotation. Can be derived. The average air gap stress is limited to about 0.4 MPa. [0024] In relation to the design of any magnetic bearing, an important requirement is the ability to generate some nominal force that can withstand unidirectional motion. If iron saturation inherently limits the effective air gap stress in any magnetic-mechanical device, then there will be a minimum operating area of the air gap for a given rated load. One approach taken by magnetic bearing designers is to use a relatively large, flat bearing surface area through which magnetic flux passes. Another approach taken by magnetic bearing designers is to use relatively large and long bearing surfaces so that the required air gap area can be achieved with a finite length shaft. [0025] Assuming that a given magnetic flux density at the air gap is B, the tensile Maxwell stress in the direction "r" of the magnetic flux lines is [0026] [Number 1]<img file="JP5113313B2_D0001.tif" />Given by. [0027] What is ignored in the design of magnetic bearings is in the two directions "s" and "t" perpendicular to r. [0028] [Number 2]<img file="JP5113313B2_D0002.tif" />The compressive stress given by is effectively present. [0029] FIG. 7a shows a set of magnetic flux lines in a plane of constant t. The square box depicted in Figure 7a shows the tension σ acting on the two opposing surfaces._rrAnd the compression (negative tension) σ that acts on the other two surfaces facing each other_ssCan be considered to have. FIG. 7b shows the same set of magnetic flux lines in the same plane of constant t. Figure 7b also depicts a square box of the same size as in Figure 7a, but the orientation of the square box in Figure 7b is 45 ° with respect to the orientation of the box in Figure 7a. In this figure, the axes "u" and "v" are defined to be at an angle of 45 ° with respect to the direction of the magnetic flux. On the sides of this box, it can be seen that practically pure shear stress acts without a component of normal stress. The magnitude of this pure shear stress "τ_uv(Fig. 7b) is the same as the magnitude of normal stress on the sides of the box in Fig. 7a. That is, [0030] [Number 3]<img file="JP5113313B2_D0003.tif" />Is. [0031] Returning to the consideration of the basic bearing region, the magnetic flux lines are perpendicular (at least approximately) to the x direction (the direction in which free relative motion of the two interface is desired). Consider that it passes between the two interface between the basic bearing regions of. If this condition is met, there will be a force component between the two principal components in the x direction. When these flux lines are parallel to the z direction (perpendicular to the central surface), the force between the two interface is the product of stress and area, or B.²A / 2μ₀Is equal to. However, B is the magnetic flux density, and A is the area of the central surface. [0032] As shown in FIG. 8, when all the magnetic flux lines are perpendicular to x and form an angle α with respect to the perpendicular line z, between the two boundary surfaces of the basic bearing region in the y direction and the z direction, [0033] [Number 4]<img file="JP5113313B2_D0004.tif" />There is a component of the force given by. [0034] In Figure 8, a positive F_yActs to pull the upper interface in the -y direction and to pull the lower interface in the + y direction. Positive F_zActs to pull the upper interface in the -z direction and to pull the lower interface in the + z direction. [0035]<u style="single">Description of the invention</u>According to the present invention, each of the two bearing members carries a set of bearing elements, and the bearing elements supported by one bearing member and the bearing elements supported by the other bearing member alternate with each other. Positioning and defining three or more substantially parallel interleaf gaps between successive elements, thereby generating bearing capacity as a result of magnetic shear stress acting across these gaps. Magnetic bearings that can be provided. [0036] A distinguishing feature of the most preferred embodiments of the present invention is that the magnetic bearings described herein achieve bearing capacity as the sum of the contributions of forces from a number of parallel (or generally parallel) air gaps. Yes, the contribution of the forces of the individual air gaps occurs as an integral of the magnetic shear stress over the air gap area resulting from the magnetic flux lines crossing the air gap at an angle to the perpendicular. A substantial portion of the flux lines present in the bearing at any time is effective in generating useful air gap shear stresses in three or more parallel air gaps. [0037] The present invention First and second bearing members provided with at least two protruding elements, respectively, bearing members in which protruding elements are alternately arranged so as to define at least three gaps between consecutive elements of the two bearing members. When, With a magnetomotive force (MMF) source that the flux lines cross the intermediate gap at an angle to the perpendicular to generate magnetic shear stress across each gap, thereby generating bearing forces or forces between bearing members. , Magnetic bearings are provided. [0038] The present invention provides a bearing having a sturdy structure, and a passive magnetic bearing having high rigidity can be obtained. Advantageously, at least one magnetomotive force source is arranged so that there is at least one set of magnetic flux lines across the three or more intermediate gaps, and at least most bearing capacity is magnetism acting across the intermediate gaps. It occurs as a result of shear stress. [0039] In a preferred embodiment of the invention, the magnetomotive force sources are arranged such that there is a set of magnetic flux lines across a set of three or more intermediate gaps, and at least most bearing capacity is a set of intermediate gaps. It is generated as a result of magnetic shear stress acting across. To produce this effect, the magnetic flux is made to follow a zigzag pattern as it passes through a stack of alternating bearing elements. It is particularly preferred that substantially all intermediate gaps be included in the set. FIG. 9 schematically shows this zigzag path. [0040] As a result of this arrangement, significant weight and material savings are possible, thus promoting low cost and large specific load capacity. A major factor in determining the weight of any magnetic bearing is the material needed to complete the flux circuit, i.e. from one side of the set of gaps if useful in producing bearing contributions. It is a material necessary to guide the magnetic flux to the other side. An array of flux line pairs across all (or many) gaps allows the weight associated with the magnetic return path for a given maximum bearing capacity to be minimized. [0041] The magnetic bearing according to the present invention achieves a high load capacitance by having an appropriate effective shear stress in each of a large number of (nearly) parallel air gaps. In order for the magnetic shear stress to exist in the air gap, it is necessary to provide a magnetomotive force source that provides the magnetic flux, and to allow the magnetic flux to cross the air gap at an angle. The maximum shear force for a given magnetic flux density occurs when the angle is 45 °. Therefore, it is advantageous to be able to control how the magnetic flux crosses the various intermediate gaps in the bearing. Many different configurations can be considered in which some shear stress can be present. However, there are essentially only three clear ways in which the path of magnetic flux can be changed from the path that magnetic flux naturally takes through free space. The three methods are (a) placing some ferromagnet in the magnetic path, (b) placing some permanent magnet material in the magnetic path, and (c) placing some current in the magnetic path. Is. [0042] In some preferred embodiments of the invention, the conductive material is placed in one or more alternating bearing elements to allow current to flow to affect the magnetic path across at least one intermediate gap. Be arranged. Alternatively or additionally, permanent magnet materials may be distributed within the alternating bearing elements to affect the magnetic path across at least one intermediate gap. [0043] In yet another preferred embodiment of the invention, materials with different magnetic permeability are distributed within the alternating bearing elements to affect the magnetic paths across at least one intermediate gap. In such cases, for this purpose, the ferromagnetic material is properly distributed within the alternating bearing elements. Therefore, in at least one bearing element in which the ferromagnetic materials are alternately arranged so that the magnetoresistance received by the magnetic flux lines passing from one side to the other side of the bearing element is a strong function of the position of the magnetic flux lines. It is distributed in a pattern. The dependence of the reluctance on the position acts to affect the magnetic path across at least one intermediate gap. [0044] All or any of these methods affecting the magnetic flux across at least one intermediate gap may be incorporated into one embodiment of the invention. In FIGS. 10, 11 and 12, three different effects (uneven magnetic permeability in the bearing element, permanent magnet material in the bearing element, and current in the surface of the bearing element) affect the magnetic path across the air gap. It outlines what to do. [0045] In embodiments that include some permanent magnet material or some current distribution in one or more bearing elements, it may or may not be necessary to provide separate magnetomotive force sources. In some preferred embodiments of the invention, the bearing is configured as an active bearing. In another preferred embodiment of the invention, the bearing is configured as a passive bearing. In yet another preferred embodiment of the invention, the magnetomotive force source comprises a single coil for selectively adjusting the total flux interlinkage, in which case the bearing is configured as a semi-active bearing. To. [0046] In some preferred embodiments of the invention, the bearing is configured as a linear bearing. In another preferred embodiment of the invention, the bearing of the invention is configured as a rotary bearing. When configured as a rotary bearing, the bearings are arranged to generate radial or axial bearing capacity. [0047] One of the bearing members preferably has one more alternating element than the other bearing member. This promotes bearing symmetry and also means that at least four such gaps are present. Shear stress is achieved in each of these gaps. By providing a plurality of gaps, the total surface area acting to generate a force between the bearing members is increased, and a bearing having a high load capacity can be realized. The width of the gap present in the bearing has a minimum value determined by the achievable alignment of the two bearing members. [0048] The number of alternating elements can be conveniently increased to increase the number of intermediate gaps and thus the total surface area acting to generate forces between the bearing members. There are preferably at least 6 such intermediate gaps, more preferably at least 8 or 10 such intermediate gaps, and 26 or more such gaps may be present. [0049] The gaps between the alternating elements may optionally be filled with non-magnetic fluid. The gap may be filled with air or evacuated. In embodiments involving a non-uniform distribution of ferromagnetic materials, the high magnetic permeability region is composed of one or more ferromagnetic materials and the low magnetic permeability region is any non-ferromagnetic material, eg, a fiber reinforced resin material. Etc. may be composed of compounds such as. Carbon fiber compounds are particularly suitable. [0050] Forces between bearing members are generated in one direction or in both directions. The direction of the force between the bearing members is preferably parallel to the central surface of the gap. The force between the bearing members acts in a direction approximately parallel to the central surface of the gap. [0051] The elements of the first bearing member are mounted together, and the elements of the second bearing member are also mounted together. These elements are mounted on a mechanical platform. The mechanical platform may be a shaft or a sleeve attached to the shaft. The mechanical platform can be a shell that holds the elements together at its outer edge. The mechanical platform is preferably made of a non-magnetic material so that it does not magnetically short-circuit. [0052] The alternating element of the first and second bearing members can be an annular disc. In rotary bearings, such discs are mounted perpendicular to the axis of rotation. Such alternating discs are arranged to generate bearing forces that act perpendicular to the axis of rotation of the bearing. [0053] The alternating element of the first and second bearing members can be a cylinder. In rotary bearings, such cylinders are mounted concentrically with the rotating shaft of the bearing. Such alternating cylinders are adapted to generate bearing forces that act coaxially or perpendicularly to the axis of rotation of the bearing, depending on how the magnetic flux is zigzag between the alternating cylinders. [0054] In yet another embodiment of the invention, the alternating elements of the first and second bearing members can be conical. In a rotary bearing, such a conical element is mounted coaxially with the rotating shaft of the bearing. Such alternating conical elements are adapted to generate bearing forces that act coaxially or perpendicularly to the axis of rotation of the bearing, depending on how the magnetic flux is zigzag between the alternating conical elements. There is. [0055] The alternating elements can be composed of laminated steel, whether it is an annular disc, a cylindrical or conical element, or a linear bearing element. Other construction methods or materials are possible, including the use of high resistivity powder metallurgy compounds and the use of composite materials that are embedded in a matrix and include a portion of appropriately oriented magnetic wire. [0056] The MMF source is equipped with a series of permanent magnets, two concentric coils, or four identical pieces reminiscent of the shape of a G-clamp. The MMF source also serves to complete the magnetic path through the bearing. The MMF source has a return path only at one end of the bearing. [0057] The first bearing member can be the rotor and the second bearing member can be the stator of the rotating machine. Alternatively, the first bearing member may be the stator and the second bearing member may be the rotor of the rotating machine. The MMF source, when provided, is preferably in the same motion frame as the bearing stator. [0058] [0058] The magnetic bearing may be a passive magnetic bearing or an active magnetic bearing. The magnetic bearing can be an active magnetic bearing that is small, lightweight, and has a large force capacity. The magnetic bearing can be a small, highly rigid passive magnetic bearing. The present invention provides passive bearings with stiffness per unit volume that is many times greater than provided by passive bearings of conventional design. The magnetic bearing can be a small, highly rigid passive magnetic bearing that allows relative rotation by generating restoring thrust but prevents relative axial motion. The magnetic bearing can be a passive thrust bearing having an axial thrust capacitance per unit volume / mass that is many times greater than the thrust per unit volume provided by passive magnetic thrust bearings of other designs. Magnetic bearings can provide substantially greater axial stiffness per unit volume / mass, and their axial stiffness is much higher than provided by passive magnetic thrust bearings of other designs. A magnetic bearing is a magnetic bearing that provides a substantial axial thrust that provides a relatively small amount of relative axial motion with little or no resistance to relative rotation. [0059] Hereinafter, some preferred embodiments of the present invention will be described with reference to the accompanying drawings, but this is merely an example. Figures 1 to 12 have already been referenced. [0060] Specific Embodiment "A" -Passive Radial Magnetic Bearing 13 to 18 show the passive radial magnetic bearing according to the first embodiment of the present invention. The magnetic bearing has three main components: a bearing rotor member 1, a bearing stator member 2, and a magnetomotive force (MMF) source 3. [0061] Recognizing that the function of a so-called radial bearing is to maintain a given relative lateral position of two bearing members that rotate relative to each other, one of the bearing members is a fixed member and the other. It is clear that the bearing member of is a movable member. Here, and in any part of the description, the terms bearing rotor and bearing stator are used only to distinguish between the two bearing members. At least in the case of rotary bearings, it is assumed that the MMF source is in the same motion frame as the bearing stator member. [0062] FIG. 13 shows a cross section of one side of the bearing rotor 1, the bearing stator 2, and the MMF source 3. The figure also shows that the MMF source works to complete the magnetic path. The dotted line 12 in FIG. 13 indicates the direction of the magnetic flux flow. The MMF source provides a reasonably homogeneous MMF via the bearing rotor and bearing stator. That is, any cross section of the bearing exhibits the same pattern and density of magnetic flux through the rotor and stator. [0063] The bearing rotor 1 comprises a number of circular rotor plate bearing elements 4 mounted on a single mechanical platform 5, as shown in FIG. The central plane of the rotor plate bearing element 4 is perpendicular to the axis of rotation. The mechanical platform 5 is a sleeve that can be fitted to the shaft. The mechanical platform of the bearing rotor 1 is made of non-magnetic material. This allows the mechanical platform to magnetically short a set of flux lines 12 through the rotor plate bearing element 4 and the stator plate bearing element 6 through the intermediate gap 11 defined by these bearing elements. This is to prevent it. [0064] The bearing stator 2 comprises a number of circular stator plate bearing elements 6 mounted on a single mechanical platform 7, as shown in FIG. Like the rotor plate bearing element, the central plane of the stator plate bearing element is perpendicular to the axis of rotation. The mechanical platform 7 is a shell that integrally holds the stator plate bearing elements at its outer diameter. The mechanical platform 7 of the bearing stator 2 is made of a non-magnetic material. This is to prevent the mechanical platform 7 from magnetically shorting the magnetic flux intended to pass through the rotor plate bearing element and the stator plate bearing element. [0065] All rotor plate bearing elements 4 are similar to each other, and all stator plate bearing elements 6 are also similar to each other except that the two end plate bearing elements differ in the orientation in which they are integrated with the MMF source 3. ing. Any single rotor plate bearing element 4 appears to be approximately identical to the stator plate bearing element 6. The main difference is that the innermost and outermost diameters of the rotor plate bearing element 4 are slightly smaller than the innermost and outermost diameters of the stator plate bearing element 6. FIG. 16 shows the rotor plate bearing element 4, and FIG. 17 shows the stator plate bearing element 6. Each of these plate bearing elements comprises a set of annular high magnetic permeability regions 8 separated by a set of relatively low magnetic permeability annular regions 9. The radial spacing of the high magnetic permeability region 8 is the same for the rotor plate bearing element and the stator plate bearing element, and the high magnetic permeability region 8 is very roughly approximated to that of the stator plate bearing element and the rotor plate bearing element. It has the same radial depth as the low magnetic permeability region 9. For simplicity, the dimensions of this radial spacing are exaggerated in the figure. In fact, the radial pitch of the high magnetic permeability region 8 is roughly on the order of three times the maximum relative lateral motion allowed between the bearing rotor 1 and the bearing stator 2. The size of the air gap between the rotor plate bearing element and the stator plate bearing element is typically about half this radial depth. [0066] FIG. 18 shows a cross section of the combination of the bearing rotor and the bearing stator, and there is some degree of lateral misalignment between the bearing rotor and the bearing stator in terms of cross section. Only the rotor plate bearing element 4 and the stator plate bearing element 6 are shown together with the high magnetic permeability region 8 and the low magnetic permeability region 9 in these plate bearing elements. The high permeability region 8 is clearly not aligned. The effect of the MMF source is to try to drive the magnetic flux axially through the stack of rotor plate bearing elements 4 and stator plate bearing elements 6 and the air gap between them. When the high magnetic permeability region portions are aligned, a relatively high density magnetic flux passes through, but the force generated between the rotor and the stator is extremely small. When the parts of the high magnetic permeability region are not aligned, the total reluctance of the axial magnetic path between both ends is large, and the magnetic flux flowing along this magnetic path must follow a "zigzag" orbit. I don't get it. FIG. 18 includes a line 10 that roughly represents this zigzag trajectory, in which the magnetic flux attempts to axially pass through a stack of alternating rotor plate bearing elements 4 and stator plate bearing elements 6. Because it does. The fact that the magnetic flux passes through each air gap at a constant angle means that there is some useful shear stress, and the effect of this shear stress is to pull the rotor and stator back into lateral (radial) alignment. It is to be. [0067] FIG. 19 shows a typical graph of static restoring force as a function of the relative lateral displacement between the bearing rotor and the bearing stator with respect to a constant axial MMF at the MMF source. Obviously some relative displacement δ_maxExists and the relative displacement δ_maxBeyond, the additional force is largely unavailable as the displacement increases. This deflection δ_maxIs approximately equal to 1/4 of the average radial distance between the centers of adjacent high magnetic permeability regions on the rotor plate bearing element or stator plate bearing element. Therefore, the maximum predicted relative deflection between the bearing rotor and the bearing stator provides the lowest limit for the radial gap in the rotor plate bearing element and the stator plate bearing element in the high magnetic permeability region. [0068] The thickness of the air gap present in the bearing has a minimum value determined by the achievable axial alignment of the bearing rotor with respect to the bearing stator. For the bearing to be effective, the radial spacing between the high permeability regions must be substantially greater than the thickness of the average air gap, typically 2-10 times greater. Therefore, achievable axial alignment effectively places another small area in the radial spacing of the high permeability regions. [0069] The axial thickness of the rotor and stator discs is the two separate lower bounds, namely the shear stress in the disc (τ)._rθ) and these axial thicknesses must be substantially thicker than the axial thicknesses of the air gaps. The thickness of the rotor and stator discs varies with respect to radius. The rotor plate bearing element has a particularly thick inner diameter portion because it withstands shear stress, and becomes, for example, uniformly gradually thinned toward the outer diameter. The thickness of the stator plate bearing elements is largely determined by the requirements for the minimum thickness at small radii and the requirements for maintaining a productive air gap at other radii. The thickness of the stator plate bearing elements varies to provide a side gap parallel to the alternating rotor plate bearing elements. [0070] The actual MMF present in the MMF source can take any one of many forms. MMF sources often include a series of permanent magnets stacked in the return path, as shown in FIG. Figure 21 shows how two concentric coils are used to achieve very similar effects. FIG. 22 shows how an MMF source is constructed from four identical members recalled in a G-clamp arrangement for a given bearing. FIG. 22 shows only one side of one G-clamp. [0071] The high magnetic permeability region 8 itself in the rotor plate bearing element, the stator plate bearing element, or both may be a permanent magnet material. The low magnetic permeability region 9 of the rotor plate bearing element 4 is often created as a composite of carbon fibers (or other fibers) to promote stability and mechanical integrity in each rotor plate bearing element. .. [0072] Specific Embodiment "B" -Passive Axial Magnetic Bearing 23 to 27 show a passive axial magnetic bearing according to a second embodiment of the present invention. The magnetic bearing described herein is a magnetic bearing that provides a substantial axial thrust that provides a relatively small amount of relative axial motion with little or no resistance to relative rotation. The bearing has three main components: bearing rotor member 21, bearing stator member 22 and MMF source 23. [0073] FIG. 23 shows a cross-sectional view of one side of the bearing rotor, bearing stator and MMF source. The figure also shows that the MMF source works to complete the magnetic path. The dotted line in FIG. 23 indicates the direction of the magnetic flux flow. The MMF source provides a reasonably homogeneous MMF via the bearing rotor 21 and the bearing stator 22. That is, any cross section of the bearing, including the axis of rotation in the face of the cross section, shows more or less the same pattern and density of magnetic flux through the rotor and stator. [0074] The bearing rotor 21 comprises a number of concentric cylindrical rotor plate bearing elements 24 mounted on a single mechanical platform 25, as shown in FIG. FIG. 24 is a cross section of the bearing rotor 21, and the cross-sectional surface includes the rotation axis. The mechanical platform 25 is generally disc-shaped and is made of a non-magnetic material so as not to magnetically short-circuit the magnetic flux intended to pass through the rotor plate bearing element and the stator plate bearing element. [0075] Each of the rotor plate bearing elements 24 comprises a set of ring-shaped regions 28 having a high relative magnetic permeability, separated by a set of ring-shaped regions 29 having a low relative magnetic permeability at regular intervals. [0076] The bearing stator 22 comprises a number of concentric cylindrical stator plate bearing elements 26 mounted on a single mechanical platform 27, as shown in FIG. FIG. 25 shows a cross section of the bearing stator 22 on one side, and the cross-sectional surface includes the rotation axis. The mechanical platform 27 shown in FIG. 25 is generally disk-shaped and is made of a non-magnetic material so as not to magnetically short-circuit the magnetic flux intended to pass through the rotor and stator plate bearing elements. [0077] Each of the stator plate bearing elements 26 comprises a set of ring-shaped regions 28 having a high relative magnetic permeability, separated by a set of ring-shaped regions 29 having a low relative magnetic permeability at regular intervals. [0078] FIG. 26 shows a cross section of one side of the combination of the bearing rotor 21 and the bearing stator 22, and there is some degree of axial misalignment between the bearing rotor 21 and the bearing stator 22 in terms of cross section. There is. Only the rotor plate bearing element 24 and the stator plate bearing element 26 are shown together with the high magnetic permeability region 28 and the low magnetic permeability region 29 in these plate bearing elements. The high permeability region 28 is clearly not aligned. The effect of the MMF source is to drive the magnetic flux radially through the stack of stator plate bearing elements 24, rotor plate bearing elements 26 and the air gaps between them. When the high magnetic permeability region portions are aligned, the alternating arrangement of the rotor plate bearing elements 24 and the stator plate bearing elements 26 provides a path for the magnetic flux with relatively low reluctance to provide the rotor and stator. Little net force is generated in between. If the high permeability regions are not aligned, the magnetic flux must follow a "zigzag" trajectory due to the high reluctance of the axial reluctance between both ends. FIG. 26 includes a line 30 that roughly represents this zigzag trajectory. This is because the magnetic flux tends to pass in the radial direction through the stack of the alternately arranged rotor plate bearing elements and the stator plate bearing elements. The fact that the magnetic flux passes through each air gap at an angle means that some useful shear stress exists and the effect of this shear stress tends to pull the rotor and stator back into axial alignment. [0079] The thickness of the air gap present in the bearing has a minimum value determined by the achievable lateral alignment of the bearing rotor 21 with respect to the bearing stator 22. For the bearing to be effective, the axial spacing between the high permeability regions must be substantially greater than the average thickness of the air gap. The achievable lateral alignment then effectively reduces the lower bound on the axial spacing of the high permeability regions. By using the higher magnetic permeability regions arranged closer together, a stiffer axial bearing can be made, but without significantly affecting the amount of axial force available. If the axial spacing is too narrow with respect to the air gap, the available axial force is weakened. [0080] [0080] The radial thickness of the cylindrical rotor plate bearing element 24 and the cylindrical stator plate bearing element 26 is two separate lower limits, namely the shear stress in the cylinder (τ)._zθ) and the fact that the radial thickness of the cylindrical plate bearing element must be substantially thicker than the radial thickness of the air gap. The thickness of the stator plate bearing element and the rotor plate bearing element varies with respect to the axial position. The rotor plate bearing element is particularly thick at its root, that is, near the mechanical platform 25 of the bearing rotor 21, because it withstands shear stresses. The same applies to the stator plate bearing element 26 for the same reason. [0081] The actual MMF present in the MMF source 23 can take any one of many forms. Often, the MMF source was stacked in an axially oriented return path in the "cylindrical" portion of the MMF source 23 or in a radially oriented return path in the "disc-shaped" portion of the MMF source 23. It has a series of permanent magnets. FIG. 27 shows how MMFs are successfully produced by providing a coil 31 wound to drive magnetic fluxes of the same pole. [0082] The low magnetic permeability region of the rotor is a composite of carbon fibers (or other continuous fibers) to improve the mechanical integrity of the rotor plate bearing element 24 and the stator plate bearing element 26, especially with respect to high speed rotation resistance. Often produced as a thing. [0083] In FIG. 23, the MMF source is shown as a two-sided configuration as long as there is a radial return path for the magnetic flux through the MMF source at both ends of the bearing. In many examples, there may be a return path only at one end of the bearing. The cross sections of the rotor plate bearing element 24 and the stator plate bearing element 26 in the high magnetic permeability region are shown as rectangles in the figure. The shape of these cross sections can vary depending on the desired shape of the force deflection curve. In FIGS. 24 and 25, the low magnetic permeability regions of the rotor plate bearing elements and the stator plate bearing elements appear to have the same dimensions as the high magnetic permeability regions. In general, this does not necessarily have to be the same. At least in some cases, the view is that the axial length of the high permeability region relative to the low permeability region is reduced in order to alleviate the requirement for a very thick return path in the MMF source. In FIG. 26, the shaft is part of a magnetic circuit. In general, the shaft may or may not form part of a magnetic circuit. If the shaft is non-magnetic, an inner sleeve must be provided to complete the magnetic circuit. The sleeve is physically connected to the rotor or stator. Obviously, it is wise to connect the sleeve to the shaft, at least in some cases. The high magnetic permeability regions themselves in the rotor plate bearing element, the stator plate bearing element, or both can be made from a permanent magnet material magnetized by radial equipolar magnetization. In such cases, it is necessary or unnecessary for an element called the MMF source to contribute to some net MMF to the magnetic circuit, in which case its role is simply to complete the magnetic circuit. [0084] Specific Embodiment "C" --- Active Radial Magnetic Bearing 28 to 35 show an active radial magnetic bearing according to a third embodiment of the present invention. The magnetic bearing has four main components: bearing rotor member 33, bearing stator member 34, external MMF source 35 and internal MMF source 36. In this embodiment, the internal MMF source and the external MMF source are in the same motion frame as the bearing stator. [0085] FIG. 28 shows cross sections of the bearing rotor, bearing stator, external MMF source and internal MMF source. This cross section is taken parallel to the axis of rotation. The dotted line in FIG. 28 indicates the direction of the magnetic flux flow. The MMF source produces a two-pole MMF pattern through the bearings. That is, given an arbitrary diameter line of angle θ, the net MMF across the bearing along this diameter line changes according to cos (θ + φ (t)). However, φ (t) is some kind of time-dependent phase shift. In this figure, it is also shown that the MMF sources 35, 36 complete the magnetic path, that is, act to guide the magnetic flux in the circumferential direction. [0086] FIG. 29 shows a cross section perpendicular to the axis of rotation of the bearing rotor, bearing stator, external MMF source 35 and internal MMF source 36. A set of concentric rings is apparent, with each ring alternating between regions 37 with low relative permeability and regions 38 with high relative permeability (ferromagnetic material / composite). Except for the innermost ring and the outermost ring, the alternating rings in FIG. 29 belong to the bearing stator and bearing rotor, respectively. [0087] The innermost ring shows the cross section of the inner MMF source 36 and the outermost ring shows the cross section of the outer MMF source 35. The details of the windings in these MMF sources are omitted in the figure for simplicity. The task of designing MMF source windings to produce two-pole MMFs is a completely standard part of the design of rotating electromechanical machines, and virtually all options available from the electromechanical industry are now available. Applicable to the situation of. [0088] Also, FIG. 29 shows several paths 39 that the magnetic flux passing through the bearing takes in a given moment. The strong zigzag pattern in this flux clearly means that the substantial air gap shear stress across each individual air gap acts to pull down the bearing rotor and pull up the bearing stator (in this example). ing. (The magnetic flux lines try to be effectively straight lines to minimize the reluctance of the magnetic path.) As is clear from FIG. 29, at this moment there is a very large reluctance that blocks any significant passage of magnetic flux through the upper half of the bearing. To pass through this region, the magnetic flux must cross all of several low permeability regions. At this moment, there are several paths with moderately low magnetoresistance through the lower half of the bearing, and a finite amount of magnetic flux passes through these paths. In the lower half of the bearing, the net amount of force generated by this magnetic flux between the bearing rotor and the bearing stator at this moment is small. [0089] FIG. 30 shows a cross section perpendicular to the axis of rotation of the bearing stator 34 and the two MMF sources 35, 36. The radial alignment of the high permeability regions 38 of all stator elements is shown. FIG. 31 shows a cross section of the bearing stator 34 parallel to the rotation axis and including the rotation axis. This clearly shows how the bearing stator consists of multiple stator cylinders 40 mounted on a single mechanical platform 41. The mechanical platform 41 of the bearing stator 34 is made of a non-magnetic material so as not to magnetically short-circuit the magnetic flux intended to pass through the rotor element and the stator element. [0090] FIG. 32 shows a cross section perpendicular to the rotation axis of the bearing rotor 33. The radial alignment of the high permeability regions 38 of all rotor elements is shown. FIG. 33 shows a cross section parallel to the rotation axis of the bearing stator 33. This clearly shows that the bearing rotor 33 comprises a large number of rotor cylinders 42 mounted on a single mechanical platform 43. The mechanical platform 43 is also made of a non-magnetic material so as not to magnetically short-circuit the magnetic flux intended to pass through the rotor and stator elements. [0091] The number of high permeability regions 38 on each stator cylinder 40 is the same, and these regions are separated by even angular increments. Generally, this number is N_SIt is represented by. The number of high permeability regions 38 on each rotor cylinder 42 is the same, and these regions are also separated by even angular increments. Generally, this number is N_RIt is represented by. Two numbers N_SAnd N_RIs different by 1 and N in the current example_S= 20, N_R= 21. [0092] At any given moment, there is one direction between the bearing stator 34 and the bearing rotor 33 that can generate a substantial force. When viewed from the frame of the stator in this direction, this direction sets the relative rotation speed of the stator and rotor to N._RIt rotates at a doubled frequency. When this "direction" is viewed from the rotor frame, this direction sets the relative rotational speed of the stator and rotor to N._SIt rotates at a doubled frequency. So, for example, if the stator of this embodiment is stationary and the rotor is rotating at 100 cycles per second, there are individual opportunities to apply impulses to the rotor in any given direction 2100 times per second. To do. By changing the magnitude and direction of the applied MMF magnetic field, a force of extremely strong frequency components can be achieved in any direction with respect to a frequency of up to 1050 Hz in this example. Generally, the frequency limit (before aliasing) is N_RΩ / 2, where Ω is the rotational speed of the shaft. [0093] Obviously, since this bearing operates by supplying a set of impulses instead of a steady force, there is some possibility that the rotor resonance or stator resonance will be excited. Therefore, the position of the bearing and the bearing support characteristics must be carefully selected. By properly shaping the current waveform at the MMF source, N_RHarmonic components of net relative forces above Ω / 2 can be reduced to any low level. [0094] The thickness of the air gap existing between the rotor cylinder and the stator cylinder of the bearing clearly has a minimum value determined by the acceptable lateral misalignment expected to be accepted by the bearing. The minimum air gap is usually many times larger than this minimum. In this embodiment, the air gap between adjacent rotor cylinders and stator cylinders increases approximately proportionally with increasing radius. [0095] 29, 30 and 31 show that there are only seven active air gaps. In a practical embodiment, the number of air gaps is substantially higher. In some cases, there may be fewer air gaps. However, in that case, the more commonly used bearing design can have higher force capacity. [0096] In the above description, the internal MMF source 36 is active in providing some MMF to drive the magnetic flux through the rotor and stator cylinders and is fixed to the same motion frame as the external MMF source 35. There is. The internal MMF source does not necessarily have to provide a net MMF for the magnetic field, in which case the internal MMF source is free to rotate with the rotor. The internal MMF source can then include a mere stack of annular laminates that only serve to direct the magnetic flux across the central portion of the bearing. Alternatively, if the internal MMF source provides sufficient MMF, the external MMF source does not need to provide a net MMF, in which case the external MMF source can rotate freely with the rotor. The external MMF source can then include a simple stack of annular laminates that only serve to direct the magnetic flux around the outer circumference of the bearing. [0097] 29, 30 and 31 show the radial thickness of the rotor cylinder 40 and the stator cylinder 42 as a constant thickness, but in some optimal cases the radial thickness. The radius can vary along the axial length. Since bearing capacity is accumulated along the length, in some cases the radial thickness of the base of the cylinder above the rotor and stator (near the mechanical platform in any case) is greater than the tip. Also needs to be increased. [0098] If the external MMF source 35 is designed to provide a net MMF to the magnetic circuit, this component is integrated with the outermost stator cylinder and the low magnetic permeability region of this stator cylinder is occupied by windings. It is wise to do so. Similarly, if the internal MMF source 36 is designed to provide a net MMF to the magnetic circuit, this component should be integrated with the innermost stator cylinder and the low magnetic permeability region of this stator cylinder should be wound. It is wise to make it occupied. Figures 34 and 35 show these situations. [0099] In FIG. 28, the rotor cylinder, not the stator cylinder, is adjacent to the internal MMF source 36. If the internal MMF source and the innermost stator cylinder are one integrated unit, it is clear that there is no rotor cylinder between them. [0100] Specific Embodiment "D" --- Active Radial Magnetic Bearing 36 to 40 show an active radial magnetic bearing according to a fourth embodiment of the present invention. The magnetic bearing comprises four main components: bearing rotor member 50, bearing stator member 51 and two MMF sources 52. [0101] In this embodiment, the net direction of the magnetic flux passing through the bearing is axial. This is in contrast to the previously described embodiment in which the net direction of magnetic flux is perpendicular to the axis. In the embodiments described herein, the two MMF sources are the same and reside in the same motion frame as the bearing stator. [0102] FIG. 36 shows a cross section of the bearing rotor 50, the bearing stator 51 and the two MMF sources 52. This cross section is taken parallel to the axis of rotation. The line in FIG. 36 shows the direction of the magnetic flux flow. The magnetic circuit is completed internally within the MMF source 52. The MMF source produces a two-pole MMF pattern through the bearings. That is, the magnetic flux is pushed axially along one surface of the bearing and returns along the other surface. [0103] The bearing stator 51 includes a set of stator plate bearing elements 53 (FIG. 38) whose central surfaces are parallel to each other and perpendicular to the axis of rotation. These stator plate bearing elements 53 have a disk shape, and as shown in FIG. 37, sectors 54 having a high magnetic permeability and sectors 55 having a low relative magnetic permeability are alternately arranged. The stator plate bearing elements are mechanically combined into one by a common mechanical platform 56 (Fig. 38). The mechanical platform consists of a non-magnetic material to prevent short circuits in the magnetic circuit. Figure 37 shows N_SPieces (in this case, N_SA single stator plate bearing element with a high relative permeability sector 54 of = 20) is shown. [0104] The bearing rotor 50 includes a set of rotor plate bearing elements 57 (FIG. 40) whose central surfaces are parallel to each other and perpendicular to the axis of rotation. These rotor plate bearing elements 57 have a disk shape, and as shown in FIG. 39, high magnetic permeability sectors 54 and low relative magnetic permeability sectors 55 are alternately arranged. Rotor plate bearings are mechanically coupled together by a common mechanical platform 58, which can be a sleeve for fitting to the shaft or the rotor shaft of the rotating machine itself (Fig. 40). The mechanical platform consists of a non-magnetic material to prevent short circuits in the magnetic circuit. Figure 39 shows N_RPieces (in this case, N_RA single rotor plate bearing element with a high relative permeability sector 54 of = 21) is shown. [0105] The operating principle of this bearing is the same as the operating principle of the bearing described above in connection with the third embodiment. At any given moment, there is one direction between the stator and the rotor that can generate substantial force. When viewed from the frame of the stator in this direction, this direction sets the relative rotational speed between the stator and rotor to N._RIt rotates at a doubled frequency. When this direction is viewed from the rotor frame, this direction sets the relative rotational speed between the stator and rotor to N._SIt rotates at a doubled frequency. So, for example, if the stator of this embodiment is stationary and the rotor is rotating at 100 cycles per second, there are individual opportunities to apply impulses to the rotor in any given direction 2100 times per second. To do. By varying the magnitude and direction of the applied MMF magnetic field, a force of extremely strong frequency components can be achieved in this case in any direction with respect to frequencies up to 1050 Hz. Generally, the frequency limit (before aliasing) is N_RΩ / 2, where Ω is the rotational speed of the shaft. [0106] Obviously, since this bearing operates by supplying a set of impulses instead of a steady force, there is some possibility that the rotor resonance or stator resonance will be excited. Therefore, the position of the bearing and the bearing support characteristics must be carefully selected. Correctly shaping the current waveform at the MMF source can minimize the harmonic content of the net relative force. [0107] The thickness of the air gap present between the rotor plate bearing element and the stator plate bearing element in this embodiment of the bearing is clearly determined by the acceptable axial misalignment that the bearing is expected to accept. Has a minimum value. The minimum air gap is usually many times larger than this minimum. In this embodiment, the air gap between adjacent rotor plate bearing elements and stator plate bearing elements increases approximately proportionally with increasing radius. This increase in air gap is primarily adapted by the corresponding decrease in the axial thickness of the rotor plate bearing elements. The axial thickness of the stator plate bearing elements can also change with radius. [0108] Specific Embodiment "E" --- Active Radial Magnetic Bearing 41 to 45 show a magnetic bearing according to a fifth embodiment of the present invention. This magnetic bearing also achieves a high load capacitance by having a moderate number of effective shear stresses in each of the (nearly) parallel air gaps, similar to the magnetic bearings in the above embodiments. [0109] The bearing of this embodiment distributes the current in at least one layer of each air gap to give the magnetic flux in the air gap the required slope. On the other side of each air gap is provided with another layer in which there is another current distribution or distribution of permanent magnet material. Whether there is a current distribution or a permanent magnet material distribution in a given layer, the net effect is to provide an axial MMF in the layer that varies with position within the layer. [0110] This bearing is similar in configuration to the bearing described in the fourth embodiment, but the means for directing the magnetic flux is completely different. That is, while it is based on the distribution of permanent magnets and the distribution of current, in the fourth embodiment, the magnetic flux is directed based on the high ferromagnetic permeability region. [0111] The magnetic bearing of this embodiment comprises three main components: a bearing rotor member 60, a bearing stator member 61 and two external MMF sources 62 (FIG. 41). The two external MMF sources 62 are again present in the same motion frame as the bearing stator. [0112] FIG. 41 shows a cross section of the bearing rotor 60, the bearing stator 61 and the external MMF source 62. This cross section is taken parallel to the axis of rotation. The net direction of magnetic flux flow is shown by arrows in FIG. The external MMF source 62 produces a two-pole MMF pattern through the bearings. That is, given an arbitrary diameter line of angle θ, the net MMF generated across the bearing at any moment along this arbitrary diameter line is cos (θ + φ) with respect to the phase angle φ. It is proportional, but irrelevant to its position along the diameter line. The external MMF source also works to complete the magnetic path, that is, to guide the magnetic flux in the circumferential direction at both ends of the bearing. [0113] The bearing rotor 60 has a large number of rotor disks 63 in the form of a stack, and the bearing stator 61 also has a large number of stator disks 64 in the form of a stack (FIG. 42). The rotor disc 63 and the stator disc 64 are "layers" in the sense applied above and are therefore all adapted to provide an axial MMF. [0114] FIG. 42 schematically shows how a bearing achieves a net lateral force. The magnetic flux lines alternately pass in the axial direction along the stator disc 64 and the rotor disc 63 that cut the bearing. The individual stator discs 64 provide an axial MMF pattern that varies (approximately) according to cos (θ + φ). However, φ is a phase angle. The individual rotor disks 63 provide an axial MMF pattern that varies (approximately) according to cos (2θ + Ψ). However, Ψ is the phase angle. In FIG. 42, φ is set to zero and Ψ is set to -45 °. [0115] The magnetic flux line passing through the center of the stator disk 64 in the axial direction at θ = 0 ° naturally tries to pass through the center of the adjacent rotor disk 63 (or its vicinity) at θ = 45 °, and of the stator disk 64. When passing through the center again, it returns to the line of θ = 0 °. Similarly, a magnetic flux line passing through the center of the stator disk 64 at θ = 180 ° naturally tries to pass through the center of the adjacent rotor disk 63 (or its vicinity) at θ = 135 °, and the stator disk 64 When passing through the center again, it returns to the line of θ = 180 °. Due to the symmetry, there is no net axial magnetic flux on the plane θ = ± 90 °. Looking at the magnetic flux pattern from one side, it can be seen that all the magnetic flux lines increase so as to enter the rotor disk 63, and then decrease when leaving the other side of the rotor disk 63. This behavior provides the flux with the angle required to generate a substantial average shear stress that acts to pull all rotor disks downwards with respect to the stator disks. [0116] In this embodiment, the distribution of the permanent magnet material is used for the rotor disk 63, and the distribution of the radial current is used for the stator disk 64. Magnetic iron is used to give the structure rigidity and strength without substantially obstructing the axial passage of magnetic flux. Magnetic iron does not substantially contribute to the direction of the magnetic flux in this case. This point is different from the conventional embodiments. [0117] Each external MMF source 62 has a toothed disc 65 (FIG. 43) and a set of windings. The toothed disc has a laminated structure with a single coil of thin laminated magnetic iron wound flat or a pair of thin concentric cylinders with an extremely thin insulating layer between adjacent cylinders. .. This structure ensures that the alternating magnetic field can pass through the toothed disc 65 in the axial and circumferential directions with minimal eddy current loss. Since no radial magnetic flux component is required, it is not a problem that the magnetic flux trying to pass radially through the toothed disk 65 causes a much larger eddy current loss. FIG. 43 shows a front view and a side view of the toothed disk 65. [0118] FIG. 43 shows a single MMF source coil 66 connecting multiple teeth. The number of conductors in this coil and the insulation thickness of these conductors are governed by the expected voltage drop and current load on this coil. The MMF source coils 66 are connected in series to form multiple groups, which are connected in parallel to form multiple phases according to standard conventions in the construction of disc electromachines. The winding set of each external MMF source 62 has at least two independent phases to generate a rotating bipolar axial magnetic field. That is, given an arbitrary straight line parallel to the axis of the magnetic bearing, located at an angle θ, and located at a radius r reaching between the two external MMF sources, the pair of external MMF sources contributes. The net MMF along the straight line is approximately determined by cos (θ + φ) and is independent of radius. In this case, the angle φ is a phase angle that can be controlled to any value between 0 and 2π depending on the value of the phase current in the winding set. [0119] FIG. 44 shows a single stator disc 67. Each stator disc 67 carries a set of stator disc coils 68 formed as stator disc windings. The stator disc is relatively thin in the axial direction. Due to its structure, the alternating magnetic flux can pass axially with minimal eddy current loss. In this embodiment, the configuration is made of thin laminated magnetic iron that is rolled flat to create a dense spiral. Teeth are machined on each side of the stator disc 67, the stator disc coil 68 is fitted into these teeth, and the set of stator disc coils 68 on one side of the stator disc 67 is a stator circle. It is a mirror image of the stator disk coil 68 on the other side of the plate 67. The stator disc coils are connected in series to form multiple groups, which are connected in parallel to form multiple phases in the same pattern used for windings formed from MMF source coils. [0120] The phases of each stator disc 67 are a plurality of external MMF sources 62 so that when these phases are urged, the axial component of the magnetic flux density is reasonably uniform with the axial position along the magnetic bearing. It is electrically connected to the phase. When the entire bearing rotor 60 is magnetically inert, the magnetic flux in the bearing is axially dominant at almost all positions and its distribution is approximately represented by cos (θ + φ). However, the angle φ is a phase angle that can be controlled to an arbitrary value between 0 and 2π depending on the value of the phase current in the set of windings. [0121] FIG. 45 shows the axial magnetization of a single rotor disk 69 and a 4-pole pattern. In this embodiment, the rotor disk is such that the net MMF contributing to any magnetic flux lines passing axially from one side to the other side of the rotor disk 69 changes according to cos (2θ + Ψ). , With a distribution of axially magnetized permanent magnet materials. However, the angle Ψ is a phase angle controlled by the rotation angle of the rotor. Ψ = 0 at the reference position of the rotor. [0122] It can be seen from FIG. 42 that two orthogonal forces can be generated by urging the MMF source coil 66 and the stator disc coil 68 in an appropriate manner at any given angle position of the rotor. It's natural. [0123] Specific Embodiment "F" --- Passive radial magnetic bearing Here, the sixth embodiment will be described. In the sixth embodiment, in contrast to the previous embodiment using the permanent magnet material, the required pattern of axial MMF in the rotor disc 63 is in this case using a set of windings. It is the same as the fifth embodiment described above, except that it is generated. [0124] The shape of the winding of the rotor disk 63 is very similar to that of the stator disk 64, except that the number of magnetic poles of the rotor disk 63 must always differ from the number of poles (5) of the stator disk by ± 2. .. As in the previous embodiment, the preferred number of poles is 2 for the stator magnetic field and 4 for the rotor magnetic field. One of the important differences between the windings of the rotor disc 63 and the stator disc 64 is that the distribution of axial MMFs on the rotor disc does not need to be rotated relative to the rotor disc, therefore. , The rotor does not require more than one electrical phase. The windings of all rotor discs 63 are electrically connected together so that each rotor disc always produces a similar distribution of axial MMFs. [0125] In the fifth and sixth embodiments described above, the predominant direction of effective magnetic flux through the bearing is axial, so a parallel air gap is between the rotor disc 63 and the stator disc 64. It is assumed that the air gap is in the shape of a disk. In fact, it is conceptually easy to apply the same idea to a system in which a parallel air gap is located between a parallel rotor and a stator cylinder and is cylindrical. The predominant direction of magnetic flux in that case is radial, with the two MMF sources having one (possibly solid) cylinder inside the rotor cylinder with the smallest diameter and one outside the rotor with the largest diameter. It will be equipped with two hollow cylinders. This shape change conceptually makes one of the external MMF sources 62 conical to increase its average diameter, and the rotor disc 63 and stator disc 64 housed inside it are similarly conical. It is started by considering making it the other external MMF source. [0126] All rotor disks 63 have the same number N_RThese magnetic poles are oriented at the same angle with respect to each rotor disk. Similarly, all stator disks 64 have the same number N._SThese magnetic poles are oriented at the same angle with respect to each stator disc and external MMF source 62. In the case of the above embodiment, N_S= 2, N_R= 4. The following constraints, ie | N_R-N_S| = 2, N_R.N_S 0 Any pair of magnetic poles {N_S, N_R} Can generate the desired net lateral force. [0127] In general, it is attractive to reduce the number of magnetic poles of the stator disk 64 in order to minimize the frequency of the alternating magnetic flux in the stator disk and thus minimize the loss. Increasing the number of magnetic poles increases the proportion of active copper and reduces the axial depth required for the toothed disk 65, but increases the alternating frequency of the magnetic flux for a given shaft velocity and is active. It is necessary to increase the update rate of the controller. [0128] The stator disc 64 needs to allow the alternating magnetic flux to pass axially without substantial loss. In the particular embodiment described above, the stator disc can be configured as a roll of laminated steel, as described. Other construction methods are possible, including the use of high resistivity powder metallurgy composites and the use of composites consisting of the majority of axially oriented magnetic wires. [0129] FIG. 41 should not be construed as implying that the inner diameter of the rotor is always small. In fact, there is no limit to the inner diameter, and it works well even with a large inner diameter. Specific Embodiment "G" --- Linear Magnetic Bearing FIG. 46 shows a linear bearing in which a pedestal 70 with a first bearing member 71 made of a low magnetic permeability material carries a pair of spaced rectangular plate elements 72. Each plate element 72 formed of multiple alternating strips of ferromagnetic and non-ferromagnetic material is an elongated ribbon formed of multiple alternating strips of ferromagnetic and non-ferromagnetic material and spaced apart. Alternating with element 75. The elongated ribbon elements 75 that are spaced apart are supported on the second bearing member 74 of the low magnetic permeability material, and the ribbon element 75 and the second bearing member 74 form a rail that supports the base 70. [0130] The permanent magnets 73 that form part of the pedestal 70 generate magnetic flux lines 76 that zigzag across the gaps between the alternating elements 72 and 75, as well as the magnetic shear stress that creates the forces that support the pedestal on the rails. Produces. The bearing capacity between the sides is generated in another way. [0131] In this embodiment, it is convenient to mount the magnet 73 and the MMF source on a "rotor" or base 70 rather than on the stator or rails 74, 75. For linear bearings, it is generally convenient to mount the main MMF source on the shorter of the two bearing members. [Simple explanation of drawings] FIG. 1 is a diagram showing a basic bearing region from a ball bearing. FIG. 2 is a diagram showing a basic bearing region from a tapered roller bearing. FIG. 3 is a diagram showing a basic bearing region from a hydrostatic bearing. FIG. 4 is a diagram showing a fluid wedge and relative motion in the basic bearing region from a hydrodynamic bearing. FIG. 5 is a diagram showing a main (orthogonal) direction for each basic bearing surface region in which the axis z is perpendicular to the central surface. FIG. 6 shows the simplest tensile Maxwell stress action in which a horseshoe-shaped permanent magnet drives a magnetic field through itself and an air gap (twice) and some second object. FIG. 7a is a diagram showing normal stress in a given plane of a magnetic field in air, and FIG. 7b is a diagram showing shear stress in a given plane of a magnetic field in air. FIG. 8 is a diagram showing a magnetic flux passing through an angle α with respect to the z-axis. FIG. 9 shows magnetic flux passing through parallel air gaps in a zigzag pattern. FIG. 10 is a diagram showing a bundle of magnetic flux guided by a high magnetic permeability region. FIG. 11 is a diagram showing a bundle of magnetic flux guided by a permanent magnet. [Fig. 12] It is a figure which shows the bundle of the magnetic flux guided by the electric current line. FIG. 13 is a cross-sectional view of one side of a passive radial bearing according to the first embodiment of the present invention. 14 is a cross-sectional view of one side of the bearing rotor of FIG. 13. FIG. 15 is a cross-sectional view of one side of the bearing stator of FIG. 13. FIG. 16 is a plan view of the rotor plate bearing element of FIG. 13. FIG. 17 is a plan view of the stator plate bearing element of FIG. 13. FIG. 18 is a cross-sectional view of the entire bearing of FIGS. 13 to 17. FIG. FIG. 19 is a plot of static restoring force as a function of relative lateral displacement between a bearing rotor and a bearing stator. 20 is a schematic diagram showing a first embodiment of the MMF source of FIGS. 13 to 19. FIG. 21 is a schematic diagram showing a second embodiment of the MMF source of FIGS. 13 to 19. FIG. 22 is a schematic diagram showing a third embodiment of the MMF source of FIGS. 13 to 19. FIG. FIG. 23 is a cross-sectional view of one side of a passive axial thrust bearing according to a second embodiment of the present invention. FIG. 24 is a cross-sectional view of a bearing rotor of the bearing shown in FIG. 23, showing a high magnetic permeability region (ring) as a dark rectangle. [Fig. 25] It is sectional drawing of the bearing stator of the bearing shown in FIG. 23 which shows a high magnetic permeability region (ring) by a dark rectangle. FIG. 26 is a cross-sectional view of the bearing stator and bearing rotor of the bearing shown in FIG. 23, having some axial displacement, showing how the reaction occurs. FIG. 27 is a schematic showing the three main components contributing to the isopolar MMF of the bearing shown in FIG. 23, in particular the coil (8). FIG. 28 is a cross-sectional view parallel to the rotation axis of the embodiment of the cylinder in which the active radial bearings are alternately arranged according to the third embodiment of the present invention, in which the vertical dotted line having an arrow indicates the magnetic flux line. .. FIG. 29 is a cross-sectional view perpendicular to the rotation axis of the embodiment of FIG. 28, showing a partial path of magnetic flux with a zigzag dotted line. FIG. 30 is a cross-sectional view perpendicular to the rotation axis of the bearing stator and MMF source of the embodiment of FIG. 28. 31 is a cross-sectional view parallel to the rotation axis of the bearing stator of the embodiment of FIG. 28. FIG. 32 is a cross-sectional view perpendicular to the rotation axis of the bearing rotor according to the embodiment of FIG. 28. FIG. FIG. 33 is a cross-sectional view parallel to the rotation axis of the bearing rotor according to the embodiment of FIG. 28. [Fig. 34] It is a figure which shows the external MMF source integrated with the outermost stator plate bearing element of the bearing shown in FIG. 28. FIG. 35 shows an internal MMF source integrated with the innermost stator plate bearing element of the bearing shown in FIG. 28. FIG. 36 is a cross-sectional view of a disk in which active radial bearings are alternately arranged according to a fourth embodiment of the present invention, which is parallel to the rotation axis of the embodiment. 37 is a front view of a single stator plate bearing element of the bearing of FIG. 36. FIG. 38 is a cross-sectional view of the bearing stator of FIG. 36. FIG. 39 is a front view of a single rotor plate bearing element of the bearing of FIG. 36. FIG. 40 is a cross-sectional view of the rotor of the bearing of FIG. 36. FIG. FIG. 41 is a schematic diagram showing the main components of an embodiment of alternating discs of active radial bearings according to a fifth embodiment of the present invention. FIG. 42 is a schematic diagram showing the operating principle of the bearing of FIG. 41 showing how the zigzag magnetic flux lines generate good practical shear stress in each gap between the discs. 43 is a front view and a side view of an embodiment of an MMF source for use with the bearing of FIG. 41. FIG. [Fig. 44] FIG. 4 is a front view and a side view of a single stator disk for use with the bearing of FIG. 41. 45 is a front view and a side view of a single rotor disk for use with the bearing of FIG. 41 showing quadrupole axial magnetization. FIG. 46 is a cross-sectional view of an embodiment of a linear bearing configured according to the present invention, which is the sixth embodiment. FIG. 47 is a cross-sectional view of an embodiment of a linear bearing configured according to the present invention.