Induction devices such as reactors are used in power systems, for example, in order to compensate for the Ferranti effect from long overhead lines or extended cable systems causing high voltages in the open circuit or lightly loaded lines. Reactors are sometimes required to provide stability to long line systems. They may also be used for voltage control and switched into and out of the system during light load conditions. In a like manner, transformers are used in power systems to step up and step down voltages to useful levels.

Such devices are manufactured from similar components. Typically, one or more coils are wrapped around a laminated core to form windings, which may be coupled to the line or load and switched in and out of the circuit in a desirable manner. The equivalent magnetic circuit of a static inductive device comprises a source of magnetomotive force, which is a function of the number turns of the winding, in series with the reluctance of the core, which may include iron and, if provided, an air gap. While the air gap is not strictly speaking necessary, reactors and transformers without air gaps tend to saturate at high magnetic field densities. Thus, control is less precise and fault currents may produce catastrophic failures.

The core may be visualized as a body having a closed magnetic circuit, for example, a pair of legs and

interconnecting yokes. One of the legs may be cut through to form the air gap. The core may support the windings which, when energized by a current, produce a magnetic field in the core which extends across the air gap. At high current densities the magnetic field is intense.

Although useful and desirable, the gap represents a weak link in the structure of the core. The core tends to vibrate at a frequency twice that of the alternating input current. This is the source of vibrational noise and stress in such devices.

Another problem associated with the air gap is that the magnetic flux 4,"fringes", spreads out and is less confined. Thus, field lines tend to enter and leave the core with a non-zero component transverse to the core laminations which can cause unwanted eddy currents and hot spots in the core.

These problems are somewhat alleviated by the use of one or more inserts in the gap designed to stabilize the structure and thereby reduce vibrations. However, these devices are difficult to manufacture and are expensive.

A typical known insert comprises a cylindrical segment of radially laminated core steel plates arranged in a wedge shaped pattern. The laminated segments are moulded in an epoxy resin as a solid piece or module. Ceramic spacers are placed on the surface of the module to space it from the core or, when multiple modules are used, from an adjacent module. In the latter case, the modules and ceramic spacers are accurately stacked and cemented together to make a solid core limb for the device.

The magnetic field in the core creates pulsating forces across all air gaps which, in the case of devices used in power systems, can amount to hundreds of kilo-newtons (kN). The core must be stiff to eliminate these objectionable vibrations. The radial laminations in

the modules reduce fringing flux entering flat surfaces of core steel which thereby reduce current overheating and hot spots.

These structures are difficult to build and require precise alignment of a number of specially designed laminated wedge shaped pieces to form the circular module.

The machining must be precise and the ceramic spacers are likewise difficult to size and position accurately. As a result, such devices are relatively expensive. Accordingly, it is desirable to produce an air gap spacer which is of unitary construction and substantially less expensive than the described prior arrangements.

SUMMARY OF THE INVENTION An aim of the present invention is to provide an improved method of producing a magnetic product, for example a block of distributed air gap material, which has magnetic material distributed within dielectric plastics material.

According to one aspect of the present invention there is provided a method as claimed in the ensuing claim 1.

Conveniently the magnetic product produced comprises a unified solid block of distributed air. gap material.

Preferably the block has a plurality of regions of magnetic material interleaved with a plurality of regions of dielectric plastics material.

If the magnetic product comprises a block of distributed air gap material, the block is intended to be fitted within an air gap of an inductor of a power system.

The block so formed is intended to be easily shaped, e. g. by machining, to achieve the desired shape to fit within the air gap. The magnetic material may comprise magnetic particles, the magnetic particles having a particle size and volume fraction sufficient to provide a magnetic

permeability similar to an air gap with reduced fringing effects. Alternatively, the magnetic material may comprise short lengths of magnetic wire.

Typically the magnetic product comprises at least two dielectric layers positioned on top of each other, each dielectric layer having its own layer of magnetic material associated therewith.

The plastics material may include all kinds of polymers, copolymers, polymer blends or elastomers.

Conveniently, however, the plastics material comprises thermoplastics material. In this case the layers may be joined together by heat and pressure, the thermoplastics material being softened by the heat and the application of the pressure causing the layers of the layered device to be pressed together before the heat is removed to allow the softened thermoplastics material to solidify.

Alternatively the plastics material may comprise a settable or curable synthetic resin, e. g. a thermosetting resin used, for example, with an accelerator for controlling the curing rate of the resin. In this case the arranged layers are joined together by the application of pressure and the settable synthetic resin is set, e. g. by the application of heat if the plastics material is thermosetting. In this case, the or each dielectric layer may comprise a first layer of curable sheet plastics material supporting a second layer of curable plastics material, the associated layer of magnetic material being positioned either between the first and second layers or over the second layer, the application of said pressure causing the magnetic material to be pressed into the second layer prior to setting of the latter. The curable synthetic resin material may include a catalyst which reacts to heat or electromagnetic radiation, e. g. ultraviolet radiation.

The magnetic product is suitable for magnetic cores of low voltage and/or low power devices, and/or high

frequency devices such as electric power converters. The magnetic product may also be used, for example, as an air gap material for a magnetic shunt in a controlled-leakage inductance transformer or a rotating electric machine.

If the magnetic material is particulate it suitably has a particle size of about 1 nm to about 1 mm, preferably about 0.1 gm to about 200 ym, and a volume fraction of up to about 60%. The magnetic permeability of the particulate magnetic material is suitably from 0 to 20.

The at least one layered product may be formed by feeding separate first and second sheets of plastics material to a uniting station and uniting the first and second sheets together at the uniting station, particulate magnetic material being applied to one side of the first sheet of plastics material upstream of the uniting station so that the particulate material is sandwiched between the first and second sheets when the latter are subsequently united together at the uniting station.

According to other aspects of the invention there is provided a solid block of unified air gap material as claimed in claim 41; an inductor core as claimed in claim 44; a method as claimed in claims 45; and a method as claimed in claim 46.

BRIEF DESCRIPTION OF THE DRAWINGS Embodiments of the invention will now be described, by way of example only, with particular reference to the accompanying drawings, in which: Fig. 1 shows the electric field distribution around a winding of a inductive device for a power transformer or reactor having a distributed air gap;

Fig. 2 is a perspective fragmentary view of a cable which may be used in the winding of a high power static inductive device for a power system; Fig. 3 is a cross section of the cable shown in Fig.

2; Fig. 4 is a schematic perspective view of a high power inductive device having a distributed air gap; Fig. 5 is a fragmentary cross section of the distributed air gap of the inductive device of Fig.

4; Figs. 6A to 6D show schematically stages in a method according to the invention of producing a unified magnetic product; Fig. 7 is a schematic cross sectional view of an alternative magnetic product; Fig. 8 shows schematically another embodiment of apparatus for producing magnetic material in sheet form for producing a magnetic product by a method according to the invention; and Fig. 9 shows schematically a further embodiment of apparatus for producing magnetic material in sheet form for producing a magnetic product by a method according to the invention.

DESCRIPTION OF THE INVENTION Fig. 1 shows an induction device 1, such as a power transformer or reactor, having at least one winding 2 and a core 3. Fig. 1 also shows a simplified view of the electric field distribution around the turns of the winding 2, with lines of equipotential designated E and indicating where the electric field has the same magnitude. The lower part of

the winding is assumed to be at earth potential. The core 3 has a distributed air gap 4 and a window 5. The core is typically formed of laminated sheets of magnetically permeable material, e. g. electrical steel, but may, alternatively, be formed of magnetic wire, ribbon or powder metallurgy material. The direction of the magnetic flux 4 is shown by the arrow in Figs. 1 and 2 and, in general, is confined, or is at least nearly confined, within the core 3.

The potential distribution determines the composition of the insulation system, especially in high power systems, because it is necessary to have sufficient insulation both between adjacent turns of the winding and between each turn and earth. In Fig. 1, the upper part of the winding is subjected to the highest dielectric stress. The design and location of a winding relative to the core 3 are in this way determined substantially by the electric field distribution in the core window 5. The windings 2 may be formed of a conventional multiturn insulated wire, as shown, or the windings 2 may be in the form of a high power transmission line cable discussed below. In the former case, the device may be operated at power levels typical for such devices in known power generating systems. In the latter case, the device may be operated at much high power levels not typical for such devices.

Figs. 2 and 3 illustrate an exemplary cable 6 for manufacturing windings 2 useful in high voltage, high current and high power induction devices. Such cable 6 comprises at least one conductor 7 which may include a number of strands 8 with a cover 9 surrounding the conductor 7. In the exemplary embodiment shown, the cover 9 includes a semiconducting inner layer 10 disposed around the strands 8, a solid main electrically insulating layer 11 surrounding the semiconducting inner layer 10, and a semiconducting outer layer 12 surrounding the main electrically insulating layer 11 as shown. The inner and outer layers 10 and 12 have a similar coefficient of thermal expansion as the main electrically insulating layer 11. The cable 6 may be

provided with additional layers (not shown) for special purposes. In a high power static conductor device, for example, the cable 6 may have a conductor area which is between about 30 and 3000 mm and the outer cable diameter may be between about 20 and 250 mm. Depending upon the application, the individual strands 8 may be individually insulated. A small number of the strands near the interface between the conductor 7 and the semiconducting inner layer 10 may be uninsulated for establishing good electrical contact therewith. As a result, no harmful potential differences arise in the boundary layer between the innermost part of the solid insulation and the surrounding inner semiconducting layer along the length of the conductor. The cable 7 may typically be as described in WO 97/45931 and such cable is incorporated herein by way of reference.

Devices for use in high power applications may have a power rating ranging from 10 kVA up to over 1000 MVA with a greater voltage ranging from about 3-4 kV and up to very high transmission voltages, such as 400 kV to 800 kV or higher.

The similar thermal properties of the various layers, results in a structure which may be integrated so that semiconducting layers in the adjoining insulation layer exhibit good contact independently of variations and temperatures which arise in different parts of the cable.

The insulating layer and the semiconducting layers form a monolithic structure and defects caused by different temperature expansion of the insulation and the surrounding layers do not arise.

Referring to Fig. 4, there is shown a simplified view of an exemplary induction device 20 including a core 22 and at least one winding 24 having N turns. The core 22 is in the form of a rectangular body which may be formed of insulated laminated sheet 26 having a window 28. The core may also be formed of a magnetically permeable ribbon, wire

or a powder metallurgy substance. The core 22 has limbs or legs 30 and 32 joined by opposite yoke portions 34. The winding 24 may, for example be wrapped around the solid leg or limb 30. Limb 32 is formed with a gap 36 and a relatively high reluctance distributed air gap insert 38 is located in the air gap as shown.

The arrangement of Fig 4 may also operate as a transformer when the second winding 25 is employed. As illustrated, the winding 25 may be wound around the core 22.

In the arrangement illustrated, the winding 25 is wound concentrically with the winding 24.

The core limb 32 exhibits a relatively high reluctance to the flux cD produced when either of the windings 24-25 are energized. The insert 38 acts as a distributed air gap and is generally non-saturated thereby allowing the device 20 to act as a controller, reactor or transformer device in a variety of power applications.

Fig. 5 illustrates the distributed air gap insert 38 in fragmentary schematic cross-section. The insert 38 comprises a matrix of dielectric material 40 containing magnetically permeable particles 42.

The dielectric material 40 may be an epoxy resin, polyester, polyamide, polyimide, polyethylene, cross-linked poly-ethylene, polytetrafluoroethylene (PTFE) and poly- formaldehyde (PFA) sold under the trademark Teflon by Dupont ; rubber, ethylene propylene rubber (EPR), acrylonitrile-butadiene-styrene (ABS), polyacetal, polycarbonate, polymethyl methacrylate (PMMA), polyphenylene sulphone (PPS), PSU, polyetherimide (PEI), PEEK, silicone rubber, polymers containing silicone and the like. The magnetic particles 42 may be formed of iron, amorphous iron based materials, or alloys of Ni-Fe, Co-Fe, Fe-Si, or ferrites based on at least one of manganese, zinc, nickel and magnesium and the like. The magnetic particles may optionally be coated with a dielectric material which may be

organic, such as some of the dielectrics referred to above, or may be an inorganic compound.

In the exemplary embodiment shown in Fig. 5, opposing faces 45 of the air gap 36 and the corresponding confronting surfaces 46 of the insert 38 may be formed with planar or curvilinear confronting surfaces. The insert 38 may have convex surfaces and the confronting surfaces 45 of the core may be concave to stabilize the structure mechanically.

Alternatively, the surfaces 45 of the core may be concave and the surface of the insert may be convex to modify field fringing. Generally however, in the arrangement illustrated, the flux in the core 22 tends to be better confined within the distributed air gap insert or region 38.

This occurs because the particles 42 provide an insulated magnetic path through the insert 38 for the flux s which tends to minimize fringing effects at the interfaces 45 and thereby reduce eddy currents in the core 22 and the insert 38.

Figs. 6A to 6D illustrate schematically one method of producing a magnetic product, in particular a unified solid block of distributed air gap material. Typically a thin sheet 50, typically about 1 mm thick, of thermoplastics material (Fig. 6A) is coated with particulate magnetic material 51 (Fig. 6B). Several layers (three in Fig. 6C) of the sheets 50 coated with the particulate magnetic material 51 are positioned or arranged one on top of the other to form a layered product 52 and are positioned in a press 53.

The actual number of layers of the sheets 50 and the particulate magnetic material 51 that are used will depend on the thickness of the finished unitary magnetic product, e. g. the air gap insert, that it is desired to produce. The layered product 52 is heated to an elevated temperature so as to soften the thermoplastics material and is pressed together in between upper and lower press parts 53a and 53b.

As shown, the upper press part 53a applies heat (H) and pressure (P) and the lower press part 53b applies pressure (P) only (although it may apply heat in addition). The

thermoplastics material is then allowed to cool so that it hardens and the pressure exerted by the press 53 is released. The magnetic material 51 is pressed into adjacent sheets 50 to form unitary magnetic material 55 (see Fig. 6D) having regions of magnetic material interleaved with regions of dielectric plastics material. Typical examples of thermoplastics material for the sheets 50 are low cost polyethylene, LDPE, HDPE or PP, possibly even recycled grades of such materials. It is also possible to use high performance plastics material such as PTFE, but such material is relatively expensive. It will be appreciated that, by the application of magnetic fields, the magnetic moments of each magnetic layer may be arranged to line up parallel to each other. In this way the magnetic moments of one magnetic layer may be aligned in a different direction, e. g. perpendicular to, the magnetic moments in another magnetic layer. This alignment may be carried out before consolidating the layers together, after pressing the layers together and even both before and after consolidation. In this way the magnetised material may be created with specific properties and may exhibit magnetic anisotropy.

The application of pressure may also be used to align the magnetic moments in a layer in a predetermined direction.

It will be apparent that unified magnetic material may be arranged to provide magnetic anisotropy in more than one direction, for example anisotropy in the plane of a layer and anisotropy in a direction through the unified material perpendicular to the layer.

Anisotropy may additionally be provided by distributing the particulate material of a layer in varying concentrations. in particular, areas of low and high concentrations of particulate material may be arranged to provide a desired pattern of varied magnetic permeability and thus magnetic anisotropy in one or more planes.

Instead of using"particulate"magnetic material, the magnetic layers may be formed from magnetic wires or

ribbons. The wires or ribbons in each layer may be arranged parallel to each other with typically the wires or ribbons of one magnetic layer being arranged at an angle to, e. g. perpendicular to, the wires or ribbons of another magnetic layer. Again, this allows magnetised material of the desired magnetic properties to be created.

As an alternative to using sheets 50 of thermo- plastics material, it is possible to use settable or curable resin material. For example heat curable thermosetting resins used with an accelerator for regulating the rate of curing may be employed. Alternatively, synthetic resins may be used which cure with the application of other forms of electromagnetic radiation, e. g. ultraviolet radiation, may be employed. In this case the sheet 50 typically comprises a set or partially set sheet on which a"gel"coating or other preparation of a non fully cross-linked settable or curable material is arranged. The magnetic material is then distributed over the gel sheet. As with the thermoplastics sheets, the curable resin sheets may be stacked to provide a multi-layered article. The layered product is then subjected to electromagnetic radiation, e. g. heat or W radiation, and pressure, the settable material cross-linking under the action of the electromagnetic radiation. In this way set or cured panels or boards are produced with the magnetic material embedded in the plastics material.

Typical settable or curable plastics materials are polyester or epoxy resins. High performance or high temperature plastics materials which may be used are polyamide, polyimide or polycarbonate.

Fig. 7 shows an alternative magnetic product 56 for use, for example, as a reactor core with a winding 57. The product comprises sheets 58 formed from magnetic powder (e. g. more than 90% magnetic powder). These magnetic sheets are interleaved between polymer sheets 50 and all the sheets 50,58 are laminated together. The polymer sheets can be reinforced with non-magnetic powder or fibre to improve their mechanical and thermal properties. A product made in

this way can carry approximately twice the flux density of a core in which magnetic particles are evenly distributed, and thus the cross sectional area of the product 56 can be halved with respect to such a core.

Figs. 8 and 9 illustrate further methods of producing a layered magnetic product, e. g. a unified block of distributed air gap material.

Fig. 8 illustrates schematically how a sheet of plastics material is coated with particulate magnetic material (or magnetic powder 82) and also how sheet form magnetic material is produced which can subsequently be assembled to form magnetic material for an insert of a distributed air gap. Thin sheet thermoplastics material, typically about 1 mm thick, in the form of a continuous first film or sheet 80a from a previous process step or from a finished roll (not shown) is fed through the nip of a pair of rollers 81 and is subsequently supported by the upper flight of a belt conveyor 83 or the like. Between the rollers 81 and the conveyor 83, a unit 87 electrostatically charges the sheet 80a across the width of the sheet as the latter passes the unit 87. Conveniently a substantially uniform charge is applied to the sheet 80a although it is possible, if required, to apply a charge pattern to the sheet 80a. This may be of importance if it is required to optimise the magnetic circuit of the magnetic product being produced.

. The magnetic powder 82 is applied on the upper surface of the sheet 80a from an applying device (not shown), e. g. a brush or roller, at the bottom of the hopper 84. The magnetic powder has an opposite electrostatic charge to that of the sheet 80a and thus the magnetic powder is attracted to and is held in contact with the sheet 80a.

In this manner, the magnetic powder 82 is applied evenly in the desired surface pattern on the sheet 80a. Typically, for example, the charge pattern applied to the sheet 80a may cause magnetic moments of the magnetic powder 82 to be

arranged generally parallel to each other. An alternative method of application of the magnetic powder is to apply an electrostatic charge, or even a magnetic charge, to the magnetic powder and to direct the charged particles into a desired position or predetermined pattern on the sheet 80a by using one or more discharge apertures, vanes or nozzles.

The desired positioning on the sheet may represent an even distribution of particles or patterns of varying concentrations of magnetic particles.

An alternative method of applying or retaining charged particles on the sheet 80a is to position a charged metal plate 89 beneath the sheet 80a so as to attract the magnetic powder towards the sheet 80a as the latter passes over the plate 89. The charge on the plate may be evenly distributed or in a predetermined pattern of varying charge intensity.

Depending on the desired pattern, the magnetic powder may be retained on the sheet 80a by the use of a suitable adhesive as well as, or independently from, the use of magnetic or electrostatic means.

Further thin sheet thermoplastics material, typically about 1 mm thick, from a separate source and in the form of a continuous second film or sheet 80b is positioned on top of the sheet 80a at, or before, being fed through a further pair of rolls 88.

Although not shown, particulate magnetic material may also be'applied to the sheet 80b prior to the sheets being united under heat and pressure at a uniting station with the particulate magnetic material sandwiched between the sheets.

The rollers 88 may be heated rollers arranged to heat soften, and to press together, the sheets 80a and 80b with the magnetic material sandwiched between the sheets. The unified sheet material 85 so formed is led away on the upper flight of a conveyor 86. As an alternative to having heated

rollers 88, the layered assembly of the sheets 80a and 80b supported on the conveyor 83 with the magnetic material positioned between the sheets, may be passed through a heating station (not shown) prior to being pressed together by passage through the nip of the rollers 88. The heating station may comprise, for example, an oven, infra red heating means or focused microwave heating means.

Typical examples of thermoplastics material for the sheets 80a and 80b are low cost polyethylene, LDPE, HDPE or PP, possibly even recycled grades of such materials. It is also possible to use high performance plastics material such as PTFE, but such material is relatively expensive.

Fig. 9 illustrates schematically another embodiment of apparatus for producing sheet form magnetic material which can subsequently be assembled to form magnetic material for a distributed air gap insert. Thin sheet thermoplastics material, typically about 1 mm thick, in the form of a continuous first film or sheet 110a from a previous process step or from a finished roll (not shown) is fed through the nip of a pair of rolls 111 and is supported by the upper flight of a belt conveyor 113 or the like.

Particulate magnetic material 102, in powder form, is applied on the upper surface of the sheet 110a. Further thin sheet thermoplastics material, typically about 1 mm thick, from a separate source and in the form of a continuous second film or sheet 10b is positioned on top of the sheet 110a at, or before, being fed through a further pair of rolls 112.

The magnetic material 102 is shown being applied from a hopper 114. It will be appreciated, however, that the magnetic powder may be applied in different ways. For example the sheet 110a may be pre-coated with an adhesive or the like, e. g. Scotch tape, to fix the magnetic powder when it is applied. However, a preferred method of applying the magnetic material 102 is to apply an electrostatic charge to the sheet 110a prior to the magnetic powder, which is

oppositely charged, being applied to the sheet 110a. The electrostatic charge may be applied to the sheet 110a by designing the rollers 111 as charging rollers. Although not shown, particulate magnetic material may also be applied to the sheet 110b prior to the sheets being united under heat and pressure with the particulate magnetic material sandwiched between the sheets.

The rollers 112 may be heated rollers arranged to heat soften, and to press together, the sheets 110a and 110b with the magnetic material sandwiched between the sheets.

The unified sheet material 115 so formed is led away on the upper flight of a conveyor 116. As an alternative to having heated rollers 112, the layered assembly of the sheets 110a and 110b supported on the conveyor 113 with the magnetic material positioned between the sheets, may be passed through a heating station (not shown) prior to being pressed together by passage through the nip of the rollers 112. The heating station may comprise, for example, an oven, infra red heating means or focused microwave heating means.

As an alternative to sandwiching particulate magnetic material between layers of thermoplastics materials, it is possible to sandwich particulate magnetic material between layers of curable resin material, e. g. a thermosetting material or curable sheet form plastics material, the curing rate of which can be adjusted by the application of appropriate electromagnetic radiation, e. g. ultraviolet radiation or heat. Heat may optionally be applied by means of suitably adapted microwave heating means.

A continuous strip or sheet of unitary magnetic material made as described above with reference to Fig. 8 or Fig. 9 can subsequently be cut into relatively short lengths or sheets and stacked or arranged on top of each other to form a product of the desired thickness. Alternatively, the cut lengths of the strip or sheet unitary magnetic material may be wound in a tight roll to form a cylindrical shape.

If thermoplastic material is used, the stacked or

cylindrical product can be heated (to join the layers together) and shaped to the desired shape and size prior to being cooled. If used as distributed air gap material, the shaped product can be suitably arranged in the air gap. If the shaped product is formed from a stack of strips or sheets, the product is conveniently oriented in the air gap so that the planes of the strips or sheets are either perpendicular to, or parallel with, the lines of flux. If the shaped product is formed from wound roll, the product is conveniently oriented in the air gap so that the edge of the sheet structure is presented to the lines of flux. The particulate magnetic material can be considered, relative to the lines of flux, as being a series of coaxial cylinders (in reality a helical arrangement) which may be more effective in reducing fringing effects than a more random arrangement of a stack of sheets arranged with the sheets containing magnetic powder evenly distributed in their planes perpendicular to the lines of flux. By arranging the particulate magnetic material as required in the various layers, a magnetic material having desired magnetic properties can be produced. In particular the magnetic permeability of the magnetic material can be tailored as required, for example the magnetic permeabilities y of the different magnetic layers may be deliberately made different to influence the magnetic permeability of the magnetic material throughout its volume.

Unitary magnetic material made according to the invention can subsequently be cut or machined to the desired shape and size.