WO1996022630A1

WO1996022630A1 - Improvements relating to magnetic coupling systems

Info

Publication number: WO1996022630A1
Application number: PCT/GB1995/002854
Authority: WO
Inventors: Alan Keith Robinson
Original assignee: Alan Keith Robinson
Priority date: 1995-01-21
Filing date: 1995-12-07
Publication date: 1996-07-25
Also published as: AU4120596A; GB9501229D0

Abstract

The invention relates to a magnetic coupling system comprising three components wherein at least two of said components are adapted for relative movement and at least one of said components is provided, at least partially, with electrical windings so that mechanical energy is transferable by magnetic forces between at least two of the three components and electrical and mechanical energy is interconvertible by the provision of the electrical windings.

Description

IMPROVEMENTS RELATING TO MAGNETIC COUPLING SYSTEMS

The invention relates to magnetic coupling systems and in particular though not exclusively to such permanent magnetic and/or electromagnetic coupling systems and a method for the transfer of energy.

Conventionally, magnetic couplings comprise two parts, at least of one of which includes a permanent magnet or an electromagnet. The other part is made from material which can be magnetised or also includes a permanent magnet or an electromagnet. Magnetic couplings are used in many applications since they represent, for example a means of transmitting mechanical forces and/or electrical power between components and may be used for devices such as motors, friction free spacers, bearings and the like. In addition, magnetic coupling can take place through objects having little or no magnetic capability.

A common form of simple magnetic coupling is the magnetic stirrer device used to continuously stir liquids in a container. Typically, the device comprises a rotating permanent magnet mounted on a motor and a cooperating separate permanent magnet stirrer coated in a low friction and chemically resistant material such as PTFE. A beaker containing liquids to be mixed is placed above the rotating permanent magnet and the stirrer magnet is placed in the bottom of the beaker. When the motor is activated the permanent magnet rotates and the stirrer magnet which is magnetically coupled to the permanent magnet, also rotates thereby stirring and mixing the liquids. Energy is transferred from the motor to the liquid by the magnetic stirrer. The magnetic stirrer rotates at exactly the same speed as the rotating permanent magnet driven by the motor.

Another example of a magnetic coupling system is a stepper motor in which a rotating armature is provided having a plurality of arms and, at the distal end thereof, a series of spaced ridges perpendicular to the direction of travel. These ridges interact with opposing ridges at the distal end of arms supporting electromagnetic windings on an outer ring when the outer ring is energised so as to rotate the armature one step at a time. The speed of the rotor can be altered by varying the frequency of the applied voltages.

Other known magnetic coupling systems include devices in which movement of a first magnet, which magnet may be a permanent magnet or electromagnet, induces movement in a second magnet or piece of magnetisable material such as soft iron or the like. A simple example of this is movement of a permanent magnet adjacent the undersurface of a table so inducing movement in a paperclip or other magnetisable item on the exposed surface of the table. The table itself is not magnetic or magnetisable.

The above systems present a number of problems which have hitherto not been tackled. In particular, each of these systems is limited in the way and at what rate, movement, or rather energy, is transferred from one component to another. Whilst, the speed of the first two systems can be varied as a function of voltage, the relationship between speed and voltage, for a given system, is fixed. Similarly, movement of the paperclip is highly dependent upon the strength of the permanent magnet and this severely limits the range of uses to which a given system having a magnet of given strength can be put since different applications may require different responses in terms of speed of movement and strength of interaction with the inducing magnet. Furthermore, whilst it is known to provide a rotor and stator combination such as those in a motor or generator, these arrangements are reliant upon the use of mechanical apparatus for translating required rotational motion from a required motion or into a driving motion. Such apparatus which include mechanical gears even when well lubricated are never friction free and therefore incur a penalty in terms of efficiency, lifetime and maintenance.

A further problem of such mechanical apparatus is the need for the very precise mechanical alignment of rotating parts.

A further problem with electric motors is that, whilst these are more efficient to run at high speed, a low speed motor may be required.

It is therefore a first object of the invention to overcome the above disadvantages of the prior art.

It is a further object of the invention to provide a magnetic coupling system capable of operating at high speeds, having a long life span and requiring low maintenance.

It is a further object of the invention to provide a magnetic coupling system which is relatively friction free and which has a high efficiency.

It is a further object of the invention to provide a system in which precise mechanical alignment of components is not required.

It is yet a further object of the invention that lubrication requirements are In a preferred embodiment, said pole means comprise at least one pole member which pole member has magnetic properties. Ideally, said pole means comprises a pole member having a non-magnetic or non-magnetisable region or recess. Preferably, said pole means comprises a plurality of pole members, which pole members may be separate and/or spaced. Thus the magnetic properties of the pole means may vary with position. Preferably said pole members on any one component are of equal or equivalent size and/or shape and/or spacing and/or nature. Ideally, said pole members are distributed on or about at least a part of said member of said component.

In a preferred embodiment, the pole means of said first, second and third components each have different numbers and/or sizes and/or shapes and/or spacings and/or compositions of pole members. Ideally, said pole means each have different numbers and/or sizes and/or shapes and/or spacings of effective pole members, which effective pole members depend at least in part on the number and/or size and/or shape and/or nature of the physical pole members.

It will be understood by those skilled in the art from the above that, amongst other things, the above features of effective poles depend on the hysteresis properties of the physical pole members and the history of the system.

In a preferred embodiment, said pole members are equivalent to said effective pole members.

In a further preferred embodiment said pole member comprises a pole projection of magnetisable material attached thereto or integral therewith. Preferably said pole member comprises at least one pole projection at each end thereof. Ideally a plurality of spaced pole projections are provided at second components; wherein in the presence of a magnetic field, regions of variable flux coupling between said at least three components are created and in these circumstances energy is transferred from at least one component to at least a second of said three components.

In a preferred embodiment the transfer of energy represents a situation where movement of at least one component induces movement in at least a second of said three components.

In an alternative preferred embodiment said transfer of energy represents a situation where movement of at least one component induces current in or about at least one other component or vice versa.

In a preferred embodiment, said first, second and third pole means have variable magnetic properties. Preferably said magnetic properties vary with position and/or time. Advantageously, said magnetic properties may be periodic in position and/or time. Preferably, the variation with position of the magnetic properties of said pole means is determined at least in part by the size and/or shape and/or nature of said pole means. Ideally, the variation of the magnetic properties of said pole means of one component with time is determined at least in part by the size and/or shape and/or nature of said one component and the size and/or shape and/or nature of at least one other component.

It will be understood from the above that said regions of variable flux coupling result at least in part from said variable magnetic properties of said at least three components. In a preferred embodiment, said pole means comprise at least one pole member which pole member has magnetic properties. Ideally, said pole means comprises a pole member having a non-magnetic or non-magnetisable region or recess. Preferably, said pole means comprises a plurality of pole members, which pole members may be separate and/or spaced. Thus the magnetic properties of the pole means may vary with position. Preferably said pole members on any one component are of equal or equivalent size and/or shape and/or spacing and/or nature. Ideally, said pole members are distributed on or about at least a part of said member of said component.

In a further preferred embodiment said pole member comprises a pole projection of magnetisable material attached thereto or integral therewith. Preferably said pole member comprises at least one pole projection at each end thereof. Ideally a plurality of spaced pole projections are provided at each end. In this way, said pole member is of variable magnetic reluctance, the pole projections having low reluctance and the regions between the pole projections having high reluctance.

In a further preferred embodiment, at least one pole member is magnetised and comprises a magnetic north pole or a magnetic south pole of a magnet, which magnet may be a permanent magnet or an electromagnet. Preferably said pole member comprises a plurality of separate and, ideally spaced, magnetic north and/or south poles and an adjacent non-magnetic or non- magnetisable region and/or recess. Said magnetic pole may form a protrusion or projection, or may lie substantially within a smoothly continuous outline of said pole means.

In a further preferred embodiment, the intermediate member of the third component comprises at least one pole member having at least one pole projection arranged so as to magnetically couple said third component with said first component in the presence of a magnetic field and at least one second pole projection arranged so as to magnetically couple said third component with said second component in the presence of a magnetic field. Preferably said at least one pole projections are near or adjacent said respective first and second components. Ideally said at least one pole projections are each positioned at or about respective ends of said pole member. In this way said pole projections are arranged to magnetically couple said first and second components.

In a preferred embodiment, the pole means of said first and/or second components comprises magnetisable material such as silicon steel, soft iron or the like, which material may, ideally, be laminate. In a further preferred embodiment, the pole means of said first and/or second components comprises at least one magnetic pole.

In yet a further preferred embodiment, the pole means of said first component is magnetisable and the pole means of said second component is magnetised.

In a further preferred embodiment, the pole means of said third component comprises at least two, and preferably at least three, pole members of preselected size and shape so as to, in use, selectively and variably couple flux between said three components. Preferably, at least two and, preferably three, pole members of said third intermediate component are sized and shaped to variably couple flux from substantially the same percentage area of one pole member on said first component to substantially the same percentage area of one pole member on said second component. Ideally the adjacent ends of each of said at least two, and preferably three, pole members are of substantially the same size.

In a further preferred embodiment, the period of variation in magnetic property of said first and third components are similar but not identical.

In a further preferred embodiment, the size of the outermost and/or innermost faces of the pole projections of said third intermediate component are substantially equal to or less than the size of the pole members of said first and second components.

Preferably, said first and second components interact directly via said third component. Ideally, in the presence of a magnetic field, magnetic flux is coupled from said first component to said second component and vice versa substantially entirely via said third component. Alternatively, the flux may be coupled at least in part via a further external component preferably, said at least three components comprise substantially closed flux loops. Alternatively, or in addition, said at least three components and said further component form a substantially closed flux loop.

In a further preferred embodiment, the magnetic field is internal and, preferably, originates from said first and/or said second components of the systems.

In a further preferred embodiment, at least one of said first and second components is rotatably mounted. Preferably, at least one of said first and second components is substantially cylindrical and, ideally, rotates about a substantially central longitudinal axis. Preferably, at least part of the pole means of said component are peripherally located about said component.

In a preferred embodiment said first component is cylindrical and said second component is linear having its pole means positioned along an exposed side.

In a further preferred embodiment the pole means of the third component passes alternating flux or flux having an alternating component.

In a further preferred embodiment, said first and second components are substantially cylindrical and rotatably mounted about a substantially centrally located longitudinal axis. Ideally, at least two and preferably three of said components are substantially coaxially mounted. Advantageously at least two and preferably three of said components are substantially concentric, an intermediate component being sandwiched between a tubular first component having its pole means mounted on its innermost surface and a cylindrical second component having its pole means mounted on its outermost surface. Preferably, interaction of at least two and preferably three of said components is in a substantially axial direction. Alternatively, or in addition, said interaction is in a substantially radial direction.

In a preferred embodiment, periodic magnetic properties of at least one of said components are fixed in position relative to that component. Ideally, these magnetic properties are determined by the hysteresis properties and the history of the system. It will be understood from the above that magnetic flux passing between said three components in the presence of said pole means produces effective meshing of said pole means in an analogy with conventional mechanical gears wherein rotation of one component produces rotation in another.

In a further preferred embodiment, the interaction of said pole means of two of said three components in the presence of a magnetic field produces regions of variable magnetic flux which regions interact with the pole means of the further component. Preferably the number of such regions is equal to the number of pole members in said further component.

In a preferred embodiment the number of pole members in said first and third components differs by at least one and preferably by two or more and, this difference equals the number of pole members in said second component.

In a further preferred embodiment the pole means of said first, second and third components comprises at least one pole member connected to its respective component by a neck portion. Preferably, said pole members have a plurality of pole projections at the distal end thereof.

It will be understood by those skilled in the art that the provision of sections having neck portions enables the use of electrical windings about one or more of said neck portions to produce electromagnetic poles. Two or more neck members may be driven by the same electrical winding.

In a further preferred embodiment, said movement of said at least one component is generated by producing electromagnetic poles in said first, second or third components.

It will be understood that the above system can be used to produce devices or apparatus in which for example torque or forces are transferred. These devices include magnetic gears with fixed or variable transmission ratios; rack and pinion mechanisms; power generation and transmission devices such as gear and motor combinations; and power transmission devices having one or more outputs.

Thus the magnetic coupling system of the invention can be used in a variety of forms including magnetic gears, gear/motor combinations having multiple outputs, rack and pinion mechanisms and other power transmission devices and therefore, the applicant reserves the right to claim the magnetic coupling system in this manner.

According to a further aspect of the invention there is also provided a method for inducing movement in a system in which there are at least three components at least two of which are adapted for relative movement with respect to one another, comprising:

(i) magnetising or providing a first component with magnetised or magnetisable poles;

(ii) magnetising or providing a second component with magnetised or magnetisable poles;

(iii) providing a third component with magnetisable poles, which component is further arranged to magnetically couple said first and second components so that in the presence of a magnetic field, regions of variable flux, coupling between said components are created and in these circumstances energy is transferred from at least one component to at least a second of said three components.

Particular embodiments of the invention will now be described by way of example only with reference to and as illustrated in the accompanying Figures.

Figure 1 shows a plan view of a magnetic coupling system in accordance with the invention.

Figure 2 shows plan views of the system of Figure 1 at various stages of rotation as one component is being rotated.

Figure 3 shows a plan view of another system in accordance with the invention. Figure 4 shows a plan view of the system of Figure 3 at various stages of rotation.

Figure 5 shows a plan view of a rack and pinion arrangement in accordance with the invention.

Figure 6 shows the arrangement of Figure 5 at various stages of rotation.

Figure 7A shows plan views of a magnetic coupling system in accordance with the invention at two different starting positions.

Figure 7B shows a plan view of a similar system to the arrangement in Figure 7 A.

Figure 8 shows a plan view of magnetic coupling system in accordance with the invention.

Figure 9 shows a plan view of an alternative rack and pinion arrangement of a system in accordance with the invention.

Figure 10 shows a plan view of an alternative system in accordance with the invention.

Figure 11 shows a plan view of two components in accordance with the invention.

Figure 12 shows a plan view of two alternative components in accordance with the invention. Figure 13 shows a plan view of two alternative components in accordance with the invention.

Figure 14 shows a plan view of a magnetic gear in accordance with the invention with a two pole rotating core and a fixed intermediate component.

Figure 15 shows a plan view of an alternative magnetic gear in accordance with the invention with a two pole rotating core and a three pole rotating intermediate component.

Figure 16 shows a plan view of an alternative magnetic gear in accordance with the invention having a two pole rotating core and a ten pole intermediate rotor.

Figure 17 A shows a plan view of an alternative magnetic gear in accordance with the invention in which the intermediate poles are provided with pole projections on the outermost surface thereof.

Figure 17B shows an alternative magnetic gear in accordance with the invention in which the intermediate poles are spaced and are each provided with pole projections.

Figure 17C shows a further alternative magnetic gear in accordance with the invention.

Figure 17D shows a further alternative magnetic gear in accordance with the invention. Figure 18 shows a plan view of an alternative magnetic gear in accordance with the invention having a four pole central core and sixteen intermediate poles.

Figure 19A shows a plan view of an alternative magnetic gear having a two pole central core and a nine pole intermediate rotor.

Figure 19B shows a plan view of an alternative magnetic gear having a two pole central rotor and a ten pole intermediate rotor.

Figure 20 shows a plan view of an alternative magnetic gear having a single pole central rotor.

Figure 21 shows a perspective view of an alternative magnetic gear in which coupling between components is primarily along a longitudinal axis.

Figure 22 shows a perspective view of an alternative magnetic gear in which coupling between components is in both axial and radial directions.

Figure 23 shows a plan view of an alternative embodiment of a magnetic coupling system in accordance with the invention in which components are pivotable and rotatable with respect to each other.

Figure 24 shows perspective views of the components of a practical magnetic gear in accordance with the invention and a cross-sectional view of the three components when assembled.

Figure 25 shows a plan view of a known electromagnetic motor without the windings.

Figure 26 shows a plan view of a gear motor combination in accordance with the invention.

Figure 26A shows a first arrangement of windings for use in a magnetic gear motor in accordance with the invention.

Figure 26B shows a second arrangement of windings for use in a magnetic gear motor in accordance with the invention.

Figure 26C shows a third arrangement of windings for use in a magnetic gear motor in accordance with the invention.

Figure 27 shows a plan view of an alternative embodiment of the gear motor combination in accordance with the invention.

Figure 28 shows a plan view of an alternative embodiment of a gear motor combination in accordance with the invention.

Figure 29 shows a plan view of an alternative embodiment of a gear motor combination in accordance with the invention.

Figure 30 shows a sectional view of a magnetic gear motor in accordance with the invention.

Figure 31 shows a perspective view of part of the same magnetic gear motor shown in Figure 30. Figure 32 shows a side elevation view of an intermediate pole member for use in the magnetic gear motor shown in Figures 30 and 31.

Figure 33 shows a sectional view of a alternative embodiment of a magnetic gear motor.

Figure 34 shows a preferred arrangement of windings for use in the embodiment shown in Figure 33.

Referring now to Figure 1, the magnetic coupling system 1 has three components, namely, a rotor 2 pivotable about an axis 3, intermediate component 6 and a rotor 4 rotatable about axis 5.

Here, component 2 comprises magnetic south poles 2 A, 2C etc (shown in white) and magnetic north poles 2B, 2D (shown in black). Intermediate component 6 comprises four shaped poles 6A, 6B, 6C and 6D of soft iron, (shown in hatching). The end faces of which also completely enclose rotor 4. Indeed, the combined area of the end faces of components 6 A and 6B adjacent pole 4A is substantially the same as the area of magnetic north pole 4A of rotor 4. Similarly, the combined area of the opposite end faces of intermediate poles 6A and 6B is approximately equal to the area of magnetic south pole 2A of rotor 2. In other words, 50% of the surface area of pole 4C is mapped onto approximately 50% of the surface area of pole 2 A by each intermediate pole. The total percentage area mapped from pole 4C to pole 2 A need not be almost 100% as shown here, but may be less than this.

It will be appreciated by those skilled in the art that magnetic poles 4A and 4B, which are semicircular, will have greatest field strength at and around points 4C. Similarly, the field at points 4D, where point 4A and 4B meet, is significantly less. Also, in this particular embodiment, the combined end faces of intermediate poles 6C and 6D are of the same size as pole 2B. Similarly the opposite end faces of 6C and 6D in combination are of the same size as pole 4B.

Magnetic poles 2A and 2B and 4A and 4B are thus coupled in a one-to-one ratio. Furthermore, these magnetic poles induce magnetisation of intermediate poles 6A, 6B, 6C and 6D. The intermediate poles serve to pass net flux from component 2 to component 4 and vice versa.

In the following Figures, hatching will be used to indicate members comprising magnetisable material and like referenced numerals refer to afore like referenced features.

This can be seen more clearly in Figure 2 in which a number of views of system 1A at various points in the rotation of component 4 are shown. In view (i) the system is in equilibrium and flux passes from magnetic pole 4 A via intermediate poles 6A and 6B, magnetic poles 2A and 2B and intermediate 6C and 6D to magnetic pole 4B. In this way, the intermediate poles couple flux between the three components.

In this embodiment, component 4 is being rotated by external means (not shown) such as a motor for example. Rotor 2 is allowed to move freely.

When rotor 4 is moved to the position shown in (ii) poles 6A and 6B are adjacent portions 4C of rotor 4 and therefore these intermediate poles continue to couple flux between the three components. Intermediate poles 6B and 6D are opposite neutral portions 4B of rotor 4 and therefore do not contribute to the net passage of flux between the three components. Intermediate poles 6A and 6C are currently operatively active ie flux is coupled between the three components whereas poles 6B and 6D are operatively neutral ie flux is circulated about two components only.

When rotor 4 is in this position, pole 2A and 2B of rotor 2 are attracted towards the operatively active poles 6 A and 6C. Thus a net force is produced on rotor 2 in the direction of arrow B.

The position that rotor 2 would adopt if rotor 4 were to remain in place is shown in view (iii). However, rotation of rotor 4 is continuous so that, as shown in position (iii) all the intermediate poles are now operatively active. Here, the flux passes from pole 4B via intermediate member 6A and 6D, rotor 2 and poles 6B and 6C back to rotor 4. When rotor 4 is in this position poles 2A and 2B and, in addition, pole 2C of rotor 2 are attracted to the operatively active intermediate poles. Rotor 2 therefore again moves in the direction of arrow B.

Views (iv) and (v) indicate the situation as rotation of rotor 4 is continued. Half a revolution of rotor 4 produces one twelfth of a revolution of rotor 2 and therefore the gear ratio in this particular embodiment is six to one with the rotors moving in opposite directions to one another. Rotors 2 and 4 to do not interact directly but rather via magnetically soft intermediate poles. Rotors 2 and 4 are constructed so that their magnetic properties vary periodically with angle. Each 360° circuit of rotors 2 and 4 contains one or more complete periods of variation in magnetic property. The angles occupied by each such period in rotors 2 and 4 are called the pitch angles of rotors 2 and 4. For rotor 4, the pitch angle is 60° and for rotor 2 the pitch angle is 360° . The nature of the magnetic properties of rotors 2 and 4 may be the same as shown here in that rotors 2 and 4 comprise magnetic poles and the magnetic field strength about rotors 2 and 4 varies in a similar bipolar way. This is not necessarily the case however.

This is shown more clearly in Figure 3 in which a rotor 7 is made from magnetically soft ie magnetisable material such as silicon steel, soft iron or the like, which rotor 7 pivots about point 8. Rotor 7 comprises pole projections 7 A and 7C etc separated by recesses 7B and 7D etc. Rotor 7 is magnetically coupled to rotor 10 by intermediate component 9. Intermediate poles 9 A and 9B are sized and shaped to effectively map poles 10A and 10B onto rotor 7.

Here rotor 10 comprises 4 magnetic poles 10 A, 10B, 10C and 10D which rotate about point 1 1. The end faces of poles 9A and 9B adjacent rotor 7 are equal in size to LI the pole spacing of rotor 7 and again, the pitch angle of rotor 7 is 60° . The pitch angle of rotor 10 is 180° . In this embodiment, however, intermediate component 9 couples only half of poles 10 A, 10B, 10C and 10D of rotor 10 to a portion of rotor 7.

Referring now to Figure 4, the views of system IC at various stages of rotation of pole rotor 10 are shown.

In position (i) all four intermediate poles 9A, 9B, 9C and 9D are operatively active. Rotation of component 10 through 12Vz degrees renders poles 9B and 9D operatively inactive. Thus pole projections 7A and 7C of rotor 7 are attracted to operatively active poles 9A and 9C. In this particular embodiment, ¹Λ revolution of rotor 10 produces 1/6 revolution of rotor 7. The transmission ratio of the two gears is therefore 3:2.

Effectively, a position of low energy is defined for freely moving rotor 7 by a combination of rotor 10 and intermediate components 9. On rotation of rotor 10 this low energy position moves and rotor 7 responds accordingly in order to adopt this position. Typically rotor 7 will move so as to adopt the path of least reluctance through rotor 7 and components 9.

During the transfer of power, the intermediate poles pass alternating flux , or flux with an alternating component, between rotors 7 and 10. The flux through the intermediate poles oscillates on a common magnetic frequency, each pole having the same fundamental frequency but differing phases.

Each intermediate pole flux cycle corresponds to a rotation of two pitch angles of rotor 7 and one pitch angle of rotor 10. Thus, rotation of pitch angle for a rotor in a system depends upon the nature of the intermediate flux and the nature of the interaction between rotor 7 and the intermediate poles. The angle of a rotor corresponding to one intermediate flux cycle is called the cycle angle of the rotor. Therefore the cycle angle of rotor 10 which is magnetised is the same as the pitch angle of rotor 10, and the cycle angle of rotor 7 which is magnetisable is twice the pitch angle of rotor 7. Other embodiments can be envisaged in which the reverse is true or indeed in which the cycle angle and pitch angles of the two rotors to be the same.

The interaction between rotor 7 and the intermediate poles is independent of the polarity of the flux in the intermediate poles since rotor 7 and intermediate pole 9A, 9B, 9C and 9D comprises magnetically soft material such as silicon steel, soft iron or the like. This means that the forces on rotor 7 are dependant on the strength of the flux in each intermediate pole but not on its polarity.

The interaction between rotor 10 and the intermediate poles 9 is polarity dependant. This means that the interaction between poles 10A, 10B and the intermediate poles are dependent on both the strength and the polarity of the flux passing through each intermediate pole.

Rotor 7 may be the driving component whilst rotor 10 may be allowed to move freely. In this case, the forces acting on rotor 10 are dependant on both the strength and polarity of the flux passing through each intermediate pole.

The flux passing through the intermediate poles may then appear to be unipolar or bipolar depending upon the nature of the rotor with which it interacts. For example, for rotor 7 the effect of the flux is independent of the polarity of the flux. For rotor 10, the effect of the flux is dependant on its polarity in that flux alternates between peaks of similar magnitude but opposite polarity.

The flux strength in each of magnetic poles 10A, 10B, 10C and 10D is approximately equal. Other embodiments can be envisaged in which the magnetic flux strength in each of these poles is significantly different and these are intended to be covered by this application.

Referring now to Figure 5, a rack and pinion arrangement is shown in which a linear component 12 is coupled via intermediate component 13 to a two poled rotor 14. Intermediate poles 13A and 13B map magnetic pole 14A onto member 12. Member 12 and the end faces of poles 13A, 13B, 13C and 13D adjacent component 12 are each provided with a plurality of pole projections or teeth 12A and 13E. The spacing frequency of pole projections 12A and 13E are substantially the same so as to maximise reluctance change with relative movement. In other words, the positions of maximum and π nimum energy for component 12 with respect to component 13 are most effectively defined in this case.

As component 14 is rotated intermediate poles 13 A, 13B, 13C and 13D couple flux to linear component 12. This is shown more clearly in Figure 6. Clockwise rotation of rotor 14 produces movement of linear component 12 in the direction of the arrow as shown. The system is completely reversible in that movement in a linear direction of component 12 will produce rotational motion in rotor 14. Half a revolution of rotor 14 will produce a lateral movement equal to the spacing between adjacent pole projections 12 A.

Referring now to Figure 7 A, a driving rotor 15 is magnetically coupled to a rotor 17, having poles 17A and 17B, via intermediate poles 16A, 16B, 16C and 16D. In position (a), when rotor 15 is rotated in the direction of the arrow, pole 17B will be attracted towards poles 16B, 16C and 16D. However, if rotor 7 is started from position (b) in which pole 17A lies adjacent intermediate pole 16B and 16C, rotation of rotor 15 will cause pole 17A to be attracted at least initially towards, intermediate pole 1 C ie in a clockwise direction. Continued rotation of rotor 15 will cause pole 17A to become attracted to now operatively active poles 16B and 16C ie in an anticlockwise direction. Referring now to Figure 7B, a similar system is shown in which the starting position of the rotors is such that there is no net resultant torque on the magnetisable rotor (shown in hatching). Thus, there are positions of rotors 15 and 17 at which a sense of rotation is not adequately conveyed from one to the other.

In order to convey a sense of rotation at all positions of the two rotors where the interaction between at least one rotor and the intermediate poles is polarity independent, there should be enough phases per half intermediate flux cycle instead of per whole cycle. In particular, an arrangement in which there are four phases of flux present, each displaced by one quarter cycle from the next, will affectively be a situation in which there is only a phase and an anti-phase present so that the sense of rotation is not conveyed. Indeed, this is the situation in Figure 4 and Figure 6.

The phase difference between fluxes in adjacent intermediate poles is important in determining how the interaction between the intermediate poles and rotors operate. The phase difference should not be 360° or 0 since this means that all intermediate poles carry the same flux at the same time, which does not result in the generation of torque. The phase difference is preferably not 180° , since the fluxes are then in phase and anti-phase only and a complete sense of rotation is not conveyed at all positions of the rotors.

For example, in the case of bipolar flux and polarity independent interaction between the two rotors and the intermediate poles, the phase difference should not be 180 degrees, since the flux intensity is then the same in all intermediate poles and no torque is transmitted.

Referring now to Figure 8 a six pole polarity independent ie magnetisable rotor 18 is coupled to a four pole, polarity dependent ie magnetic rotor 20 by six intermediate poles 19. Intermediate poles 19A, 19B and 19C map one pole from rotor 18 to one pole of rotor 20. In this particular embodiment, the flux passing through each intermediate pole 19 A, 19B and 19C is out of phase by 60° or 120° thus the sense of rotation is conveyed from one rotor to another in all positions.

Referring now to Figure 9, a rack and pinion arrangement having a linear component 23 operatively coupled to a rotor 22 by intermediate component 21 as shown. Here, intermediate poles 21 A, 21B and 21C couple pole 22 A to component 23. The intermediate flux cycle through each intermediate pole is 60° or 120° out of phase and thus a complete sense of rotation is conveyed at all positions of 22 and 23.

Referring now to Figure 10, a six pole, polarity independent rotor 24 is magnetically coupled to a two pole polarity dependent rotor 26 via intermediate component 25 and a flux return component 27. Component 27 substantially surrounds the greater part of the perimeter of component 24. The poles of component 25 have slightly enlarged end portions to facilitate coupling of the magnetic flux from component 26 into component 24 and vice versa whilst limiting coupling of flux between intermediate poles. Flux return component 27 closes the flux loop via the south pole (in white) of component 26. Angle x of component 24 is the pitch angle between adjacent poles of pole 24 whereas angle y is the cycle angle of component 24, being the angle through which component 24 will rotate in one intermediate flux cycle of flux passing through the poles of component 25.

Referring now to Figure 11, two linear components 28 and 29 are shown having poles 28A and 29A respectively. The end faces 31 of poles 28A adjacent linear component 29 have a smaller linear dimension than pole spacing L3 of poles 29 A. This ensures unambiguous and efficient coupling of flux from component 29 to component 28 and vice versa. Poles 28A are spaced by a distance L4. Flux lines are indicated on the figure, which flux lines determine the position of effective poles 30. Spacing L5 of effective poles 30 is determined by the spacing of poles 28A and 29A. Poles 28A and 29A have a slightly different pitch.

When intermediate poles 28A and magnetic poles 29A are moved relative to one another, effective poles 30 move at a higher speed. This rapidly moving magnetic field comprising poles 30 can be made to interact with other magnetic components (not shown) so that the speed of this third component is higher, but the forces on it are lower, than the speed and forces on the intermediate poles 28 A and the magnetic poles 29 A. Operatively inactive poles 28B have flux passing into and out of an end face 31 forming a closed loop with the adjacent north and south magnetic poles 29 A.

Movement of component 29 in the direction of arrow D will produce movement in effective poles 30 in the direction of arrow E.

Referring now to Figure 12, a similar arrangement is shown using only polarity independent magnetically soft components 32 and 33. Component 32 has a series of pole projections 32 A. Component 33 comprises magnetically soft poles 33A having a trapezoidal cross-section. The pitches of the components are slightly different so that regions or fringes of high and low magnetic reluctance 34 are produced. In the presence of a magnetic field, poles 33B adjacent a region of low magnetic reluctance become operatively active and couple magnetic flux into or out of component 32. Again, relative movement of poles 32 and 33 results in faster movement of the regions or fringes 34 of variable reluctance.

Magnetic or magnetisable poles on the opposite sides of regions 30 and 34 in the two preceding figures can be made to interact with the regions or fringes 30 and 34 to produce a gearing effect.

Referring now to Figure 13, components 29 and 37 interact to produce a series of regions or fringes 34A having a relatively high magnetic field intensity or a relatively low magnetic field intensity. Flux is recirculated in the end faces of the poles of component 37 at end faces 36. Effective poles 35 comprising a series of alternating magnetic north and south effective poles result from the interaction of poles 37 A which couple with poles 29 A. A third component may interact with polarity independent regions or fringes 34A and/or with polarity dependent regions 35 depending upon the nature of the third component itself.

It would be apparent to a man skilled in the art that various combinations and arrangements of components can be envisaged in accordance with the invention such that varying effects are produced when the components are moved relative to one another.

Figure 14 shows a system having a two magnetic pole innermost rotor, a four pole intermediate stator 39 and a 10 magnetic pole outermost cylindrical rotor 38. The bearings separating these components may be magnetic so as to provide a substantially friction free system though in this case, the bearings will typically be remotely located from the three components. The arrangements necessary for doing this should be well understood by those skilled in the art. Alternatively, or in addition, any other suitable bearings may be used to space the three components as would be also understood by those skilled in the art.

The outer cylinder 38 is magnetised radially with five pole pairs 38 A and 38B etc, shown black (for north) and white (for south). These colours represent the polarity of the field at the inside surface of the cylinder. Between the core and outer ring is an arrangement of four intermediate poles 39B through which the magnets interact.

Figures 14a to 14h show the lowest energy positions of the inner core as the outer ring is made to rotate, or alternatively the lowest energy position of the outer ring as the inner core rotates. Every 9 degree rotation of the outer ring corresponds to a 45° rotation of the inner core. The gear ratio is thus 5 to 1.

In this example the numbers of pole pairs in the magnets, and the number of intermediate poles, are such that with the intermediate poles fixed, the inner and outer magnets rotate in the same direction. Fixing cylinder 38 and rotating the inner two components produces a gear ratio of 4 to 1 with the motion in opposing directions. It is also possible to make the inner core and outer ring rotate in opposite directions in another way by varying the numbers of pole pairs and intermediate poles. If, for example, there are the same numbers of pole pairs as in the example shown, but three intermediate poles instead of four the gear ratio remains the same, except that rotation is reversed.

Fixing the outer ring and allowing the inner core and pole assembly to rotate is illustrated in Figure 15. An outer ring 41 with five pole pairs 41 A and 4 IB etc, an inner core 43 with two poles, and an intermediate pole assembly 42 with three poles 42 A, 42B and 42C, are arranged so that the outer ring 41 is fixed and the intermediate poles 42 and inner core 43 rotate. This results in a gear ratio of six to one, with the intermediate poles and inner core rotating in the same direction. Fixing the intermediate poles 42 and allowing components 41 and 43 to rotate produces a gear ratio of 5 to 1, with the components rotating in opposite directions.

For every fixed position of inner rotor 43 with respect to stator 42 there is a position or set of positions of outer rotor 41 in which the overall magnetic energy is at a minimum, and in which rotor 41 will rest in the absence of any opposing applied torque.

In the following description the energy considerations will be described in detail with reference to Figure 15. However, it will be understood that these considerations have general application.

The two rotors 41 and 43 are referred to as A and B in the following paragraphs for clarity.

Each of the minimum energy positions of B is separated from the next by the pitch angle of B. Where there is only one minimum energy position, the angular separation between "adjacent" positions as one full rotation.

Similarly, for each fixed position of B with respect to a third stationary component, there is a position or number of positions of A in which the overall magnetic energy is at a minimum, and in which A will rest in the absence of any applied torque.

The rest position or positions of B are a function of the position of A so that each small movement of A results in a small movement in the rest position of B and vice versa. Continuous rotation of A results in continuous rotation of the rest position or positions of B and vice versa.

Energy transfer can be continuous, so that as A rotates ahead of its minimum energy position with respect to B, B rotates behind its minimum energy position with respect to A. The external torque on A pulls A forward, while the external torque on B pulls B back. There is thus a transfer of power from A to B. Power may similarly be transferred from B to A.

A three dimensional plot of magnetic field energy against the angular positions of A and B looks like a series of valleys and ridges. If A or B is rotated while the other is free, the combined positions of A and B will follow the bottom of the energy valley in which A and B are currently operating.

The maximum energy level possible, and torque that can be transmitted, is limited. Application of sufficient torque to A and B will result in the combined position of A and B moving over the energy ridge and into the next energy valley. This is equivalent to the slipping or cogging of mechanical gears from one meshing position of the cogs to a neighbouring meshing position.

The energy level at which maximum torque is transmitted is not the peak energy level, but the level at which the first derivative of energy with respect to angular displacement from rest position is highest - in other words, the point at which the sides of the energy valley are steepest.

It is not essential that the intermediate poles carry the same peak flux, or that the phases of the fluxes carried are regularly spaced, but this is preferable. In a typical practical gearbox, the intermediate poles will form a regular array, each one carrying the same peak flux, and the phases of the fluxes being evenly spaced. This does not necessarily mean that the intermediate poles are the same shape.

In order to achieve a gear ratio other than unity, techniques to interface different cycle angles of A and B to the same intermediate pole flux cycle are required.

One technique is to have a smaller cycle angle in A than in B, and directly map a number of cycle angles of A onto the same number of cycle angles of B via intermediate poles, leaving some of the magnetically active part of A unused. Some of the surface of B may also be unused, but necessarily a lesser proportion of it for example Figures 1-4, 7, 8 and 10. (The roles of A and B may of course be reversed).

Another technique is a fringing technique, as shown in Figures 11-13, whereby the intermediate poles have a pitch angle greater than half the pitch angle of A but less than the pitch angle of B. Frequency or pitch of intermediate poles closer to the pitch angle of A produce longer intermediate pole flux wavelengths and higher gear ratios.

Another, similar technique, is to place the intermediate poles on a pitch greater than the pitch angle of A. The important difference between the two techniques above, is that the sign of the phase angle difference is opposite, which leads to movement of the intermediate pole flux wave in opposite directions for the same direction of rotation of A. In other words the rotors move in opposite directions.

It is also possible to produce fringing effects by choosing intermediate pole pitches of greater than or close to twice the pitch angle of A, but this is less effective in practice and is not preferred.

Another fringing technique is to use polarity dependent interaction for A and polarity independent interaction for B, so that the wavelength of the intermediate flux cycle is effectively different for A and B. This relies on the effective doubling of the phase angle difference for polarity independent interaction compared with polarity dependent interaction.

Another technique is to change the effective pitch of the intermediate poles by giving them toothed faces on a pitch close to that of similar teeth or projections in A, so that the reluctance of the magnetic flux path from the intermediate poles varies cyclically as each tooth pitch of A passes. This technique is appropriate to polarity independent interaction.

Various combinations of these techniques can be used.

In practise we prefer to use a ratio of teeth: pitch of 0.4-0.45; or a tooth: notch ratio of 0.4:0.6, ideally 0.45:0.55. In some instances the ratio will be 0.5:0.5 i.e. 1.

Figure 16 shows an inner rotor 46, ten intermediate poles 45 in an intermediate pole assembly (not shown) and an outer rotor 44 with eight slots.

The reluctance of the magnetic path from inner rotor 46 via intermediate poles 45 to outer ring 44 depends on the position of intermediate poles 45 relative to the outer ring. Because the number of intermediate poles differs from the number of slots in the outer ring, the reluctance of the magnetic path from inner rotor to outer rotor varies with angular position. By making the difference between the number of intermediate poles and slots in the outer ring equal to two, there are two angular directions in which the reluctance is a minimum. A two pole magnet will rest with its poles pointing in these two opposite directions, as shown in Figures 16(a) to 16(h).

Given any position of the intermediate poles and outer ring, the inner rotor will tend to rest so that its two poles coincide with the path of least reluctance. Any displacement away from this position will cause torques to be exerted on the three components. By rotating two parts relative to one another, the third part will also tend to rotate.

With a large number of intermediate poles and outer slots, a very large gear ratio can be achieved in a single stage.

Except where there is only a small number of intermediate poles, or the angular pitches of the outer slots and intermediate poles differ widely, the magnetic fluxes in adjacent intermediate poles are almost in phase with one another, so that it is possible to combine two or more effective intermediate poles into a single physical pole with slots in it.

Figure 17A illustrates this. Eighteen intermediate poles 48 A are combined into six physical poles members 48B. For maximum effectiveness the angular pitch of the slots in the pole members 48B will be approximately the same as that of the slots in outer ring 47. Flux leakage, however, may occur between neighbouring physical pole members.

Where the intermediate poles are made of laminated iron stampings, this use of combined physical poles reduces the number of components in the assembly and will cut manufacturing costs.

Flux leakage can be reduced by missing out intermediate poles, as illustrated in Figure 17B. Here, only two of the three effective intermediate poles of each physical pole 48B are present, leaving a much larger spacing between physical poles 48B.

Referring now to Figure 17C, an alternative magnetic gear arrangement is shown in which the outer component 52C comprises pole projections 52A which are larger in size than recesses 52B. Similarly Figure 17D, shows another alternative gear arrangement having an outer stator 52D with relatively large recesses separating pole projections, an intermediate rotor 52E and an inner bar shaped magnetic rotor 52F.

Referring now to Figure 18, an inner rotor 58 is provided with four poles, intermediate rotor 57 with sixteen intermediate poles and outer stator 56 with twelve poles. The gear ratio is 4 to 1 and the two rotors move in the same direction.

Referring now to Figure 19 A, the system comprises a magnetisable inner rotor 59 having magnetisable poles 59A and 59B. Here flux is coupled from component 61 via component 60 internally across component 59 to the opposite pole before returning to component 60 and 61.

Referring to Figure 19B, an alternative magnetic gear is shown having a 12 pole outer rotor 62, a 10 pole intermediate rotor 63 and an inner rotor 64 having two magnetically uncoupled poles 64A and 64B of soft iron. Flux is again variably coupled between the three components when two of the components are moved with respect to one another inducing movement in a third. However, in this respect flux is recirculated into and out of the same face of each of the poles on the inner rotor. Thus, flux is coupled to pole 64B from magnetic pole 62 A via magnetisable pole 63 A. Flux is returned to magnetic pole 62B of outer rotor 62 from inner pole 64B via intermediate magnetisable pole 63B.

Referring now to Figure 20, a homopolar magnetic gear comprising components 65, 66 and 67 of magnetically soft material are shown. Inner rotor 65 has a single pole which, in the presence of a magnetic field, will tend to align itself with the path of least reluctance through members 66 and 67. Typically, members 65 and 66 will be coupled externally to provide a return flux mechanism and also a magnetic field by, for example, the provision a U-shaped magnet the poles of which lie adjacent components 65 and 66 respectively.

Figure 21 shows an exploded perspective view of a magnetic gear arrangement comprising end component 68, having poles 68A and 68B, intermediate component 69 and end component 70. In this particular embodiment, the three components rotate relative to one another about a longitudinal axis through their centres. Thus magnetic flux is coupled in a primarily axial direction.

Referring now to Figure 22, a perspective view of another magnetic gear comprising components 71, 72 and 73 is shown. Components 71 and 73 are mounted on respective rods whilst the support means for component 72 is not shown but will typically comprise a ring into which magnetisable poles have been stamped or inserted. Again the three components are relatively rotatable about a common longitudinal axis through their centres. In this case, flux coupling takes place primarily along both radial and longitudinal axes.

Referring now to Figure 23, a plan view of a three component system which is tolerant to misalignment is shown. Component 74 comprises a spherical head pivotally mounted on a rod at 75 A. Component 74 is also rotatable about its general longitudinal axis. Intermediate component 75 is cylindrical and comprises at least two poles. Component 76 is similarly provided with a semi-spherical head which is pivotal with respect to a supporting rod.

Thus, whilst the two gears in Figures 21 and 22 are capable of axial and radial misalignment, without adversely affecting the operation of the magnetic gears, the embodiment in Figure 23 is designed to tolerate primarily angular misalignment. Typically, the magnetic poles on the spherical heads have spherical surfaces with the same centre as self aligning bearing supporting the support rods.

Embodiments can be envisaged in which angular and axial and/or radial misalignment are tolerated and these are intended to be covered by this application. The breakdown torque of magnetic gears depends on the strength of the magnetic flux in the magnetic circuit or circuits. Most applications will require that this torque be higher than the maximum normally transmitted.

Some applications require adjustable torque limiting, however. Magnetic gears can be used in such applications, since the breakdown torque can be varied by adjusting the magnetic flux. The simplest way to do this would be to provide an additional variable reluctance which controls the flux.

The breakdown torque is substantially proportional to the square of the flux density. This allows the torque limit to be varied over a wide range with a smaller flux density adjustment range. Because the relationship between flux and breakdown torque does not depend on friction (unlike slipping clutches or gears), which is notoriously variable, the adjustment will be precise and repeatable.

Gears comprising three magnetisable components are well suited to this type of application, since they can be excited by external magnets. Including a variable gap in the circuit may also allow the flux to be varied. Alternatively, where automatic control is required, use of an electromagnet to provide the flux allows the breakdown torque to be controlled simply by varying the coil current.

Controlling the flux can also be used to engage and disengage the gears. Where the gears are excited by coils, the exciting current can simply be turned on and off to achieve this.

Referring now to Figure 24, a high speed rotor 78 mounted on a shaft and having two magnetisable pole projections 78A and 78B is shown. A non magnetic or non magnetisable support 78C isolates the shaft from the magnetic circuit.

Typically, this rotor will be a high speed rotor. When the torque transmitted is constant the distribution of flux in component 78 is substantially constant so that counter measures against iron losses are not taken. A design in which eddy current and hysteresis loss are present will tend to damp out any oscillations in the system so machined mild steel or cast dynamo steel would be appropriate.

Component 79 comprises laminated iron poles 79 A embedded in supporting non-magnetic and non-magnetisable matrix 79B. L_aminated construction is advantageous because the flux through poles 79A includes a substantial alternating component. Typically, component 79 rotates at a lower speed.

To avoid pole face loss in the high speed rotor, gap 79C between poles 79A are kept small closest to component 78, but are larger elsewhere to rninimise flux leakage between poles.

If the supporting matrix 79B is metallic (such as die cast aluminium), the iron stampings are insulated from it to prevent eddy currents. However, although poles 79A carry alternating flux, the total flux through the whole assembly is substantially constant, so it is acceptable for a conductive ring to be present around the outside. A metal ring around the outside would strengthen the assembly and prevent the pole pieces being thrown out by centrifugal forces.

Component 80 is an interference ring comprising pole projections 80A which interfere with poles 79A to create varying reluctance fringes. Component 80 can be made by winding silicon steel strip into a laminated ring and machining projections or teeth to one face of it.

When assembled, the shaft of component 79 passes through the centre of ring component 80 as shown in cross-section in view (d). In the angular positions in views (a), (b) and (c), pole 79A and 80A are out of phase at the top and bottom and in phase to the front and rear.

The gear box assembly in view (d) shows an outer case formed from two steel end plates 81 and a ring magnet 82 surrounding the inner gear component for providing a magnetic field. The ring magnet is axially charged in a similar way to loudspeaker magnets. A large proportion of the flux from ring magnet 82 passes through end plates 81 to return in the opposite direction through the gear components in the middle. The gear components are thus subjected to a strong axial field.

When in place, components 78 and 79 are separated by a small air gap and similarly components 79 and 80 are also separated by a small air gap. This gap may be filled or partially filled with a liquid or material of any suitable permeability or coercitivity, as may be required for particular applications. A non magnetic shim may be placed between ring component 80 and end plate 81 in order to reduce losses. The gear ratio in this particular embodiment is 9: 1 with the two shafts rotating in opposite directions.

Because of the strong axial fields used, the two rotors are subject to considerable axial forces as well as the torques transmitted through them. Bearings must be capable of handling these high axial forces. The gear box assembly comprises three main sub-assemblies namely: one end plate 81, a bearing housing and component 78; ring component 82 and possible associated alignment component and the other end plate 81, a bearing housing, low speed rotor 79 and ring component 80.

Instead of using a fixed ring component 80 and rotating intermediate component 79, the intermediate component can be fixed and the ring component can be rotated and furthermore coupled to an output shaft. This could be achieved by making the intermediate component 79 part of the magnet assembly.

Figure 25 shows a conventional known two component motor comprising an armature assembly 84 and a rotor 85 (for clarity windings are not shown).

Figures 26, 27, 28 and 29 show various embodiments of three component magnetic coupling systems which are adapted to be used with electromagnetic windings thus providing a novel motor gear combination.

Referring firstly to Figure 26, a fixed outer component 86 is provided with six pole members 86A having neck portions 86C and at the distal end thereof three pole projections 86B. Neck portions 86 facilitate the introduction of windings about neck projections 86C. Indeed one winding may span two more pole members 86 A. Many suitable arrangements of windings can be envisaged by those skilled in the art and Figures 26A, 26B and 26C show three such arrangements.

A function of the winding is to shift the angular position of the inner rotor and its associated flux pattern. The design of the windings may ignore the intermediate rotor but does need to take into account that the flux pattern provided by the inner rotor. The windings are designed so that they couple torque to the inner rotor with acceptable copper losses. Ideally the winding arrangement produces minimum torque ripple as well. This is equivalent to designing them to have a constant ratio of e.m.f. to speed.

By energizing windings (not shown) situated about pole members 86A of component 86, motion can be induced in components 87 and 88. Conversely, rotor 88 and/or 87 may be rotated with respect to stator 86 in order to induce electric current in the windings. Other arrangements in which for example a single input rotor produces electric current in the windings of one component and movement in another component or two inputs in the form of a rotor and energised windings of a stator produce movement or electric current in or about a third component can also be envisaged and these along with any obvious variants are intended to be covered by this application.

Referring now to Figure 27, neck portions 90A are provided between projections 90B and 90C of separate pole members 90D of intermediate component 90. Windings about said neck portions are also provided (not shown) which when energized result in induced movement in components 89 and 91 respectively. Alternatively, motion of 91 may be used to induce electric current in the winding and also respective relative movement of component 89.

Referring now to Figure 28, a further embodiment of a magnetic gear is shown having a two pole magnetic inner core 93, an intermediate pole rotor

94 and an outer, preferably fixed, ring component 92. Ring component 92 comprises pole projections 92A having enlarged distal ends for matching the size of pole projections in an adjacent component and, neck portions 92C for mounting windings there on. In this particular embodiment ring component 92 has the same number of pole projections 92A as slots 92D for windings. In addition, poles 94A have been combined in pairs so that there is one pole projection 94B on the inner side of each pole 94A and, there are two pole projections 94C on the outermost side of poles 94 A.

A motor designed in this manner would be particularly suitable to be driven by an electronic variable speed three phase inverter, driving two pole three phase coils in the stator slots 92D. The minimum number of slots required for three phase coils is three, but in practice a much larger number is generally used such as the eighteen illustrated here. Numbers of slots this high and higher are already common practice in polyphase cage motors. With 18 slots in the outer stator component 92 and 16 effective intermediate poles 94 A, the reduction ratio in this motor is 8:1. Use of a 24 slot stator and 26 intermediate poles would give a gear ratio of 13 to 1. Thus, relative pole speeds of the rotors may be 24,000 rpm to 3,000 rpm and 39,000 rpm to 3,000 rpm for these two arrangements respectively.

Referring now to Figure 29, a commutator motor with a three pole armature core is shown. In particular, a rotor 98 is provided with poles having neck portions 98A and enlarged distal ends. Intermediate component 97 comprises 15 intermediate magnetically soft poles. Outer component 95 is provided with a series of pole projections 95A on two sides thereof. Magnets 96 provide a magnetic field and complete a return flux loop for the variable flux passing between the three components.

Inner core 98 can be made into a three pole armature by the addition of coils around each of the neck portions and use of a three segment commutator.

Outer poles 95 A are drawn with teeth on a 20 degree pitch, equivalent to 18 with some missing. This, combined with 15 intermediate poles 97 A gives a reduction ratio of 1 :5 with the inner armature and intermediate poles running in opposite directions. A practical design like this may include a larger number of intermediate poles each one possibly comprising a single small strip of silicon steel compressed or cast into a plastic shell around the inner rotor. The inner rotor can be made with a hollow shaft through which the motor output shaft attached to the outer shell at the opposite end to the commutator, passes.

Although intermediate poles 97A carry flux of both polarities at different times, the field is still effectively unipolar. During one full rotation of intermediate poles 97 A, each pole undergoes a period of fluctuating unipolar flux when adjacent the left hand side of component 95, periods of inactivity at the top and bottom of component 95 and a period of activity with the unipolar flux in the opposite direction at the right hand side of component 95.

Other embodiments incorporating electromagnetic poles can be envisaged and these are intended to be covered by this application.

It will be apparent to those skilled in the art that the output from the three component system can be taken from any one of the three relatively moveable components. Thus, two rotors running at different speeds are made possible by using magnetic gears in combination with windings as described in the proceeding paragraphs and these will be particularly useful for specialised applications. The provision of a high speed rotor leads to a more efficient motor but at the same time an output shaft mnning at a lower more generally useful speed is also provided. The high speed shaft may be used for example to drive a cooling fan.

One of the three components in the system of the invention may be replaced by a squirrel cage, the rotating magnetic field set up by rotation of the rotor will tend to drag the squirrel cage around. Similarly, rotation of the squirrel cage will tend to rotate the ring.

The motor gear combination is particularly suitable for the provision of the three components as input or output ports. For example, one or two mechanical inputs in the form of rotating or otherwise moving components may be used to generate one or two electric current outputs or vice versa.

Alternatively, or in addition, one of the components may be provided with magnetic hysteresis properties similar to a hysteresis motor rotor. Again efficiency and torque coupling will be inferior to the use of, for example, permanent magnets, but some of the advantages of squirrel cages will be obtained. When the torque coupled is low enough not to shift the rotor polarisation, the speed ratio between the input and output will be precise when a component with hysteresis properties is used which is not the case with squirrel cage rotors. Squirrel cages can also be combined with other forms of rotors to damp out oscillations if these are a problem.

Figure 30 shows just one embodiment of a magnetic gear motor. In this embodiment coils are wound in slots 101 in a laminated iron stator 100. A central permanent magnet 102 is adapted to rotate at high speeds and a set of intermediate poles 103, between the central magnet 102 and the outer stator 100, is adapted to rotate at a lower speed.

In this particular embodiment of the invention three phase windings are used.

The intermediate pole assembly comprising intermediate poles 103 is a cage enclosing the inner core and fitting inside stator 100 so that the magnetic fields pass between the inner core 102 and the outer stator 100 via the intermediate poles 103.

The outer stator 100 is a stack of laminations with slots 101 therein adapted to accommodate the coils. The innermost face 100A of stator 100 is notched and comprises a pitch designed to interfere with the intermediate poles 103 and produce a required gearing effect so that the intermediate poles 103 are at a much lower speed than the inner magnetic core 102.

The inner core 102 is a permanent magnet mounted on rolling bearings and having four poles, however any number of poles may be used having regard to a user's requirements.

Figure 31 shows the intermediate pole assembly. A motor output shaft 104 runs through the centre of the assembly and is attached to the assembly at one end. The free end is open so that the inner core may slide along the shaft 104 into the assembly.

As shown in Figure 32 the intermediate poles 103 comprise individual strips which are made of silicon steel. Each strip has a small notch on each end which is adapted to fit into a groove in the intermediate pole assembly as will hereinafter be described. The intermediate assembly comprises a mounting disc 105 and a ring 106. A suitable groove is provided on one side of disc 105 and ring 106 and the ring and disc are sized and shaped so that when correctly orientated and positioned coaxially said grooves are aligned.

During assembly individual strips 103 are mounted in the said grooves and held in position using epoxy resin. Ideally the assembly is built by firstly placing the parts into a mould filled with resin after which the resin is left to cure and the assembly is removed.

It is envisaged that disc 105 and ring 106 hold the intermediate poles 103 rigidly in place. However, if there is any loss in rigidity an additional detachable mounting disc can be provided adjacent ring 106.

In order to safeguard against short circuits the ends of intermediate poles 103 and those parts of disc 105 and ring 106 in contact with said poles 103 are suitably coated.

Holes 107 are provided in disc 105 in order to facilitate positioning of the inner core 102 within the assembly.

In the embodiment shown in Figure 30 the magnetic gear motor has a gear ratio of 8:1, this is provided by a four pole permanent magnet 102, 32 intermediate poles 103, and 36 teeth on the inner surface lOOA of the outer stator 100.

Reducing the number of permanent magnetic poles to just two reduces the total magnet energy available, increases the flux path length, increases the bulk of the stator iron required to return the flux around the outside, and increases the amount of copper in the coil overhang, which are all disadvantages. It does, however, increase the gear ratio available for the same number and size of intermediate poles so that a higher output torque is available.

As previously mentioned, the number of stator pole teeth on surface 100 A differs from the number of intermediate poles 103 by the number of inner core magnet poles. In addition, the number of teeth is also a multiple of the number of stator pole slots assuming symmetry.

In so far as the windings are concerned coils have to be wound through the slots 101 between adjacent stator teeth. Where the pitch of teeth are small, it may be necessary to miss out one tooth so that the gap through which the coils are wound are about three times the size. It is possible to imagine a low voltage motor with a high gear ratio in which the coil wire is too thick to pass through the space available. Missing a tooth out like this is, however, best avoided because it may result in increased eddy current losses. Thus multifilar winding may be preferable.

In so far as the intermediate poles 103 are concerned their number is not constrained and therefore it is possible to choose the number of stator pole teeth on surface 100 A and inner core magnetic poles and then set the number of intermediate poles 103 to suit. The number of inner core magnetic poles can be either added to or subtracted from the number of stator pole teeth to give the number of the intermediate poles. If the number of intermediate poles is higher the inner core and intermediate pole assembly rotate in the same direction. If the number of intermediate poles is lower they rotate in opposite directions.

The gap tolerances between the components of the system are dependant upon two factors, firstly, the ability of the permanent magnet 102 to energise the gaps and secondly, the effectiveness of the interaction between intermediate poles 103 and stator teeth.

The first consideration requires that sum of the two gaps widths i.e. that between the inner magnet and the intermediate poles and also between the intermediate poles and the stator poles, is much less than the length of the permanent magnet if the full flux to be delivered by the magnet to the gaps.

The second consideration requires the gap between the intermediate poles and the stator pole teeth is much less than the pitch of the intermediate poles and the stator pole teeth. This becomes more critical the higher the gear ratio.

Figure 33 shows yet an alternative embodiment of a magnetic gear motor which is thought to be particularly suitable for machine tools. In this embodiment a smaller number of coils are used in order to exert the necessary torque on the particular rotor.

Briefly, the windings in the magnetic gear motor produce an output by applying a comparatively low torque to the inner rotor, which torque is then increased by the magnetic gear gearing effect. Since it is possible to make a motor so that the coils are capable of applying more torque to the inner rotor than is required in order to achieve maximum output torque it is possible to achieve maximum output torque by reducing the number of coils. If necessary the requisite number of coils can be enlarged in order to keep copper losses down. Depending upon the nature of the design construction the requisite coils may act on all or a part of the inner rotor periphery but in any case exerts sufficient torque on the inner rotor to achieve a high output torque.

There are a number of advantages associated with the embodiment shown in Figure 33. Firstly, the winding process is simplified and secondly, the shape of the motor can be changed so that the output shaft is now offset from the centre line of the motor or it can in someway be made to fit better into a constrained space.

In the embodiment shown in Figure 33, the outer stator 100 is provided with only five slots 101 adapted to accommodate coils. The middle slot 101 A is larger than the four outer slots because it is adapted to accommodate two coil sides.

The coils act on a part of the periphery of the rotors. The remaining part of the stator is provided with vernier teeth 100 A so that the gearing effect operates everywhere, thus making maximum output torque possible even though no coils are present.

In the embodiment shown the gearing ration is 8:1. This particular embodiment is suitable to even higher gear ratios because the torque exerted by the coils reduces with higher gear ratio. This is because the maximum limit to the output torque is essentially independent of vernier tooth pitch and depends primarily on the length and diameter of the active gap. In magnetic gear motors with very high gear ratios very little copper is needed and the torque only needs to be applied to a small part of the rotor periphery. Figure 34 shows an example of the winding arrangement suitable for use in the embodiment illustrated in Figure 33.

It will be appreciated from the above that magnetic gears and gear motor combinations in accordance with the invention have a wide range of possible applications including, but not limited to: fans and blowers; driving an impeller in a supercharger for an engine, for cutting tools and in particular high-speed tungsten carbide cutting tools since as the torque is limited, damage is limited if the cutting tools stall or seize; small DC motors; DC brushless motors; variable frequency driven cage motors; load protection using the torque limiting characteristics of the magnetic gears; motor protection, again using the torque limiting characteristics; torque coupling through seals, electric screwdrivers; reduction gears for small air turbines; and indeed in any application where mechanical gears are typically used to transfer energy from one component to another and also in the generation of power from motion and vice versa. Other applications will be apparent to those skilled and these are intended to be covered by this specification.

A three component system and method as herein described therefore provide for variable transmission of energy from one component to another in a variable and relatively friction free manner with a significant reduction in wear and the likelihood of seizure. The system and method enable the operation of components at very high speeds. In addition, the system according to the invention is highly tolerant to mechanical misalignment and will typically have a long life span.

Claims

1. A magnetic coupling system comprising at least three components, at least two of which are adapted for relative movement with respect to one another wherein; the first component comprises a member having magnetised or magnetisable pole means; the second component comprises a member having magnetised or magnetisable pole means and the third component comprises an intermediate member having magnetisable pole means which member is arranged to magnetically couple said first and second components and further wherein at least one of said three members is provided, at least partially, with electrical conducting means; the whole being arranged so that in the presence of a magnetic field, regions of variable flux coupling between said at least three components are created and in these circumstances mechanical energy is transferable by magnetic forces between at least two of said components and electrical and mechanical energy is interconvertible by the effect of electric currents in the said electrical conducting means.

2. A magnetic coupling system according to Claim 1 wherein at least one of said members is provided with at least one space adapted to receive windings of electrical conducting material.

3. A magnetic coupling system according to Claim 2 wherein a plurality of spaces are provided and positioned so as to accommodate a plurality of windings.

4. A magnetic coupling system according to Claims 2 or 3 wherein a plurality of spaces are provided and adapted to accommodate electrical conducting windings wherein two or more spaces accommodate windings from a single length of electrical conducting material.

5. A magnetic coupling system according to any preceding claim wherein at least one component is provided with a plurality of spaces for receiving electrically conducting windings and further wherein said spaces are symmetrically positioned having regard to the structure of the component.

6. A magnetic coupling system according to Claim 5 wherein said spaces are provided in ring-like formation.

7. A magnetic coupling system according to any preceding claim wherein, in the absence of excessive currents in said electrical conducting means and in the absence of an excessive force on any moving component, the relative movement of said components is constrained by a magnetic gearing effect that is determined by the geometry of the magnetic parts of said three components and the disposition of flux therein.

8. A magnetic coupling system according to Claim 7 comprising three interactive components wherein;

the first component is a magnetised component in which flux of different density and/or polarity appears at different points in its region of interaction with said other two components;

the second component is a magnetisable component with electric conductors placed so that flux generated by said first component couples with said conductors in a way which varies with said relative movement of said three components; and a third magnetisable component.

9. A magnetic coupling system according to Claim 8 wherein;

said first component is a high speed rotor;

said second component is a stator; and

said third component is a low speed intermediate rotor through which flux passing between said first and second components passes.

10. A magnetic coupling system according to Claim 8 wherein;

said first component is a high speed rotor;

said second component is a intermediate stator through which flux passing between said first and third components passes; and

said third component is a low speed rotor.

11. A magnetic coupling system according to Claim 8 wherein;

said first component is a stator;

said second component is a high speed rotor; and

said third component is a low speed intermediate rotor through which flux passing between said first and said second components passes.

12. A magnetic coupling system according to Claims 9, 10 or 11 in which said stator is one of at least two and possibly three components which rotate, but in which the relative movement of the said three components are otherwise the same.

>

13. A method for inducing movement in a system and/or generating electric current in which there are at least three components, at least two of which are adapted for relative movement with respect to one another, comprising; 0 i) magnetising or providing a first component with magnetised or magnetisable poles.

ii) magnetising or providing a second component with magnetised 5 or magnetisable poles.

iii) providing a third component with magnetisable poles which component is further arranged to magnetically couple said first and second components so that in presence of a magnetic field o regions of variable flux coupling between said components are created.

iv) providing electrical conducting means about a least a part of one of said components so that; 5 in the presence of a magnetic field regions of variable flux coupling between said at least three components are created and in these circumstances mechanical energy is transferable by magnetic forces between at least two of said components and electrical and mechanical energy interconvertible by the effect of the electrical currents in the said conducting means.