WO2000035039A1

WO2000035039A1 - Electric device, electrode for an electric device and a method of operating an electric device

Info

Publication number: WO2000035039A1
Application number: PCT/GB1999/004149
Authority: WO
Inventors: Philip Thomas Hughes
Original assignee: Philip Thomas Hughes
Priority date: 1998-12-09
Filing date: 1999-12-09
Publication date: 2000-06-15
Also published as: AU1669800A; GB9827144D0

Abstract

An electric device (1) has a first cell having a first anode (5) and a first cathode (6) and electrolyte (11) and a second cell (1) having a second anode (5) and a second cathode (1) and electrolyte. An insulator (10) is movable relative to the first anode (5) and the first cathode (6) and relative to the second anode (5) and the second cathode (6) such that movement of the insulator (10) varies the conductance of each of the first and second cells. The device has many applications, for example as a current source, a motor, an uninterruptible power supply, a switch, a rectifier and an inverter.

Description

ELECTRIC DEVICE, ELECTRODE FOR AN ELECTRIC DEVICE AND A METHOD OF OPERATING AN ELECTRIC DEVICE

The present invention relates to an electric device, an electrode for an electric device and a method of operating an electric device.

It is a well-known problem that electrochemical cells can suffer impairment of operation αue to polarisation (the build-up of gas on electrodes) and other effects such as passivation and dendπte formation. Several prior art schemes have been proposed which deal effectively with these problems. In particular, some schemes, for example CA-A-1219902, US-A-4684585 and EP-A-0207522 have used a combination of rotating components and electrochemical cells to help mitigate these effects. However, none of these schemes seeks to devise an electrochemical cell which can give rise to an alternating or otherwise time-varying current. US4306001 discloses an electric storage cell m which each electrode is encased by an enclosure which has tnrough holes. A movaole mask which has through holes corresponding to the through holes of the enclosures can be moved over the enclosures to vary the output current. However, the arrangements disclosed cannot simply produce alternating current and cannot simply produce multiphase alternating current.

Furthermore, whilst, electrical motors and generators are well established macnmes, in cases of which we are aware, all electric motors, whether AC or DC driven, have made use of an external power source. This means that power has to be fed into a rotating component, entailing the use of mechanical connections such as slip rings or an arrangement of commutator and brushes . Such arrangements are subject to wear and tend to produce copious electrical emissions. In addition, electric motors can only be used where there is either mains power available or where suitable batteries are provided. External batteries make the arrangement bulky and limit portability and ease of installation. Similarly, installations relying on mains power are very limited in their portability. For example, clean, quiet motive power is often required in remote areas where the appropriate power sources are not readily available .

According to a first aspect of the present invention, there is provided an electric device, the device comprising: a first cell having a first anode and a first cathode and electrolyte for conducting charge between the first anode and the first cathode; a second cell having a second anode and a second cathode and electrolyte for conducting charge between the second anode and the second cathode; and, an insulator which is movable relative to the first anode and the first cathode and relative to the second anode and the second cathode such that movement of the insulator varies the conductance of each of the first and second cells .

The same insulator can be used to vary the conductance of each cell. This allows for straightforward production of AC. It also facilitates the production of multi-phase AC if desired.

Typically, the cells will be arranged to be parallel to each other, with the anodes and/or cathodes on opposite sides of the respective cells being substantially coplanar with each other.

In one form, in which the electrodes and electrolyte comprise a primary or secondary cell, the device can give rise to an alternating current output by periodically varying the conductance of the cell or battery of electrochemical cells whilst: their voltage remains constant. In this way, time-varying current can be driven through an external load. This is in contrast for example to conventional alternating current generators which rely upon the generation of a time-varying voltage to drive a current through an external load. In a second form, m which the electrodes and electrolyte do not comprise an electrochemical cell, the device acts as a variable resistor, or switch, and can give r se to an alternating current by periodically varying its conductance when placed in series with an external current source which may be, for example, a conventional cell or battery of electrochemical cells. Again, in this way, time-varying current can be driven through an external load.

In either of the above cases, he electrode pair and electrolyte system will be referred to as a "cell" m the following. This will be qualified rfith the words "primary", "secondary" or "active" to indicate an electrochemical cell, or "passive" to indicate a non- electrochemical cell, or switch.

In a typical example of an active cell, the open- circuit voltage (whilst the electrodes are exposed) is substantially independent of the relative orientation of the electrodes and insulating plate, as this voltage is a property of the electrode materials. The output voltage of a passive cell is also typically suostantially independent of the relative orientation of the electrodes and insulating plate as this is dependent upon the connected current source. However, m eitner case, when the electrodes are rotating at a constant angular velocity with respect to the insulator, the conductance of each cell varies between a maximum value and substantially zero. Such varying conductance can then ce used to drive a varying current through an externa- load.

The electrolyte preferably nas a low viscosity, which minimises viscous drag in those erroodiments in which an electrode or an insulator is rotated relative to the electrolyte .

The device of the present invention can operate as a motor, which can effectively be self-powering. Two or more devices can be combined as discussed in more detail below, with for example one acting as a current source and one as a motor. Possible applications for the invention include static rotational power applications such as required by water pumping/irrigation stations, air-conditioning plants, cable-car winding plants, etc.

The insulator may be arranged to be rotatable relative to the first anode and the first cathode and relative to the second anode and the second cathode and to be selectively positionable between the first anode and the first cathode and between the second anode and the second cathode .

The insulator is preferably arranged such that selected rotation of the insulator relative to the first anode and the first cathode and relative to the second anode and the second cathode increases the conductance of the first cell whilst decreasing the conductance of the second cell. In the preferred embodiment, this variation is cyclic in the sense that after the conductance of the first cell reaches a maximum, and the conductance of the second cell reaches a minimum, continued rotation of the insulator causes the conductance of the first cell to decrease and the conductance of the second cell to increase (though not necessarily simultaneously) . Upon complete rotation of the insulator the conductances of both cells are equal to their original values .

In the preferred embodiment, the insulator has an open portion through which ions can move between an anode and a cathode when the open portion is positioned between said anode and said cathode .

The insulator may be stationary and the respective anode and cathode of each cell may be rotatable together.

Alternatively, the anode and the cathode of each cell are stationary and the insulator is rotatable.

The electrolyte may be common to the first and second cells .

Preferably, the insulator is a solid. A solid insulator provides a highly convenient way of preventing charge being conducted by the electrolyte between the cathode from the anode during other time intervals as required.

In another embodiment, the insulator is a gas or a mixture of gases .

In a preferred embodiment of either passive or active cells, the anodes of said two cells are arranged on one side of the device and the cathodes of said two cells are arranged on the opposite side of the device, and the insulator is arranged between the anodes and the cathodes to rotate relative to the anodes and cathodes selectively to expose more of the electrolyte to the anode and cathode of one of the cells whilst exposing less of the electrolyte to the anode and cathode of the other cell. Again, in the preferred embodiment, this variation is cyclic in the sense that after the conductance of the first cell reaches a maximum, and the conductance of the second cell reaches a minimum, continued rotation of the insulator causes the conductance of the first cell to decrease and the conductance of the second cell to increase (though not necessarily simultaneously) . Upon complete rotation of the insulator the conductances of both cells are equal to their original values .

In an alternative embodiment having just two cells, the anodes and the cathodes are substantially semi-circular plates and the anode of one cell is substantially coplanar with the cathode of the other cell and vice versa, the insulator being a substantially semi-circular plate which is rotatable between the anodes and the cathodes to progressively increase the conductance of one cell by progressively exposing more of the electrolyte to the anode and cathode of said one cell and simultaneously to progressively decrease the conductance of the other cell by progressively exposing less of the electrolyte to the anode and cathode of said other cell. This variation is cyclic in the sense that after the conductance of the first cell reaches a maximum, and the conductance of the second cell reaches a minimum, continued rotation of the insulator causes the conductance of the first cell to decrease and the conductance of the second cell to increase. Upon complete rotation of the insulator the conductances of both cells are equal to their original values.

The device may comprise one or more additional cells each having an anode and a cathode and electrolyte for conducting charge between said anode and said cathode, the insulator being movable relative to the anode and the cathode of said one or more additional cells to vary the conductance of said additional cells. The anodes of the cells are preferably arranged on one side of the device and the cathodes of the cells are arranged on the opposite side of the device. This arrangement helps to mitigate electrical leakage as well as being simpler to assemble. The insulator is arranged between the anodes and the cathodes to rotate relative to the anodes and cathodes selectively to expose more of the electrolyte to the anode and cathode of at least one of the cells whilst exposing less of the electrolyte to the anode and cathode of at least one of the other cells. Again, in the preferred embodiment, this variation is cyclic in the sense that after the conductance of the first cell reaches a maximum, and the conductance of the second cell reaches a minimum, continued rotation of the insulator causes the conductance of the first cell to decrease and the conductance of the second cell to increase (though not necessarily simultaneously) . Upon complete rotation of the insulator the conductances of both cells are equal to their original values .

Plural such devices may be assembled together to provide batteries of sets of such cells.

An external load may be connected across the anode and the cathode of at least one of the cells, the device being arranged to drive current through said external load. The device of this arrangement, when an active cell, acts as a current source.

The device may comprise inductive current extraction means for inductively extracting a current from said device. This is particularly useful for the embodiment m which at least some of the electrodes rotate and the insulator is fixed as it avoids the need for mechanical connections such as slip rings or an arrangement of commutator and brushes; such arrangements are subject to wear and tend to produce electrical emissions, which is avoided in this embodiment. The induced current may be an alternating or a direct current, depending on the arrangement of the windings as discussed in more detail below. Depending on the ratio of tne number of the first and second windings, tne device may act as a transformer, allowing a higher or lower voltage to be obtained from the second winding than is generated b the electrochemical potential of the device. Unlike a purely electromagnetic generator, the power output of the device when acting as a current source is not related to the power required to rotate the rotatable parts of the device. The power to rotate those parts may be obtained from any convenient source, including an internal combustion engine, or a wind or water powered mill, or even another device of the present invention operating as a motor as discussed further below .

The anodes and the cathodes may be segments of a circle. In the most preferred embodiment, the anode and the cathode are each substantially quadrant-shape plates. In this embodiment, the insulator is preferably a substantially triple quadrant plate positioned between the anode and the cathode plates, i.e. the insulator is a plate having three insulating quadrants and one quadrant through which ions can pass.

The device may comprise a second insulator which is operable to determine the maximum exposure of the anode to the cathode of at least one of the cells thereby to determine the maximum current available from that cell. Where a solid insulator is used as the first insulator, the second insulator is preferably designed to be a shape which is complementary to the shape of the first insulator. In an example, a maximum current is obtained when the second insulator is completely overlapped by the first insulator. To reduce the current output by the device, the second insulator is displaced with respect to the first insulator, thus screening off more and more of the through hole of the first insulator and thus increasing the maximum cell resistivity. In the embodiment where the anode, cathode, first insulator and second insulator are all substantially semi-circular plates, the electrode pairs are electrically isolated from each other and no current can flow in an external circuit when the second insulator is rotated such that it and the first insulator effectively form a circular disc. It will be appreciated that a second insulator baffle can be used with other electrode geometries.

The device may comprise a battery of cells each comprising an anode, a cathode and varying means. Such cells may be arranged in M groups, each being orientated at relative angles of 2π/M. In the case of active cells, the device itself can give rise to a M-phase alternating current. A passive cell device can also be used to produce an alternating current in a load in a similar fashion. However, each angular group of passive cells may be connected to a separate external current source.

In an active cell, or battery, windings may be connected across the anode and the cathode of at least one of the cells through which a current generated by the device can pass so as to generate a magnetic field which can interact with an externally applied magnetic field to drive the device as a motor. Such a motor is effectively self-powering. It will be appreciated that power dissipated in the windings may cause heating of the electrolyte. In such circumstance, the windings should be carefully designed to minimise Ohmic heating and some means of passive cooling to mitigate heating effects may be desirable.

When the device is used as a motor, each cell or group of cells is preferably positioned by a certain amount relative to their neighbouring cells or group of cells. Ideally, the relative angle between each cell, or group of cells, should be π/(N-l), where N is the total number of cells in the battery. This allows the peak current for each cell, or group of cells, to occur at uniformly spaced rotation angles, thus providing a more uniform torque on the armature. At least one of the anode and the cathode may have a plurality of through holes. This increases the effective surface area of the electrode. It also minimises the weight of the electrode, which is beneficial when the electrode is rotated. Increasing the effective surface area of the electrode can also be achieved by additionally or alternatively making the surface of the electrode relatively rough.

Preferably, at least one of the anode and cathode is substantially planar, at least some of said through holes being cylindrical, the cylindrical axes of said through holes being at an acute angle to the plane of said at least one of the anode and cathode. This increases further the gain in effective surface area and reduction in weight. It also helps to ensure that the electrolyte flows through the holes in the embodiments where the electrode rotates, thereby preventing the holes becoming blocked.

According to a second aspect of the present invention, there is provided an uninterruptible power supply for supplying power to a load on failure of a mains supply, comprising a device as described above acting as a current source to supply power to a load on failure of a mains supply.

According to a third aspect of the present invention, there is provided a rectifier for converting alternating current to direct current, the rectifier comprising a device as described above.

According to a fourth aspect cf the present invention, there is provided an inverter for converting direct current to alternating current, the inverter comprising a device as described above. According to a fifth aspect of the present invention, there is provided a method of operating an electric device which comprises a first cell having a first anode and a first cathode and electrolyte for conducting charge between the first anode and the first cathode and a second cell having a second anode and a second cathode and electrolyte for conducting charge between the second anode and the second cathode, the method comprising the step of moving an insulator relative to the first anode and the first cathode and relative to the second anode and the second cathode to vary the conductance of each of the first and second cells.

The method may comprise the step of rotating the insulator relative to the first anode and the first cathode and relative to the second anode and the second cathode to increase the conductance of the first cell whilst decreasing the conductance of the second cell.

According to a sixth aspect of the present invention, there is provided a unitary electrode for an electric device, the electrode being substantially planar and comprising at least a first electrode segment and a second electrode segment, and an insulator for insulating the first electrode segment from the second electrode segment .

The electrode may comprise two electrode segments each of which is substantially semi-circular.

Alternatively, the electrode may comprise four substantially quadrant shape electrode segments on a face of the electrode each insulated from each other by an insulator.

This electrode arrangement is particularly useful for the devices described above and further below. In a typical electric device employing such electrodes, there will be at least two such electrodes arranged with their planes parallel to each other. In the preferred embodiment of such a device, the arrangement will be such as to allow an anode portion of one electrode to oppose an cathode portion of the other electrode, there being means to modify the current being conducted by an electrolyte in the cells so formed during selected time intervals.

The anode and the cathode may each be described by the polar relation r(θ) discussed further below where r(θ) is the distance between the centre of the electrode and a point on the rim of the electrode, θ is the angular position of the point on the rim, and A is the total area of the electrode. In an active or passive cell using such electrodes m the arrangement discussed in more detail below, the short-circuit output current will be substantially sinusoidal. If other waveforms are required, the shape of the anode and the cathode can be altered accordingly.

At least one of the electrode segments preferably has a plurality of through holes. This increases the effective surface area of the electrode. It also minimises the weight of the electrode, which is beneficial when the electrode is rotated. Increasing the effective surface area of the electrode can also be achieved by additionally or alternatively making the surface of the electrode relatively rough.

At least some of said through holes may be cylindrical, the cylindrical axes of said tnrough holes being at an acute angle to the plane of said electrode. This increases further the gain m effective surface area and reduction in weight. It also helps to ensure that electrolyte flows into the holes in the emoodiments where the electrode rotates, thereby preventing the holes becoming blocked.

The present invention also includes a device as described above wherein at least one of the anode and the cathode is provided by an electrode as described above.

Embodiments of the present invention will now be described by way of example with reference to the accompanying drawings, in which:

Figures 1 to 3 are schematic transverse cross- sectional views of a first example of a device according to the present invention in different configurations and with some parts omitted for clarity;

Figures 4A and 4B are schematic longitudinal cross- sectional views of the first example of a device according to the present invention shown in Figures 1 to 3 in different configurations;

Figure 5 is a trace of the current output by the device of Figures 1 to 4;

Figure 6 shows schematically an arrangement for extracting current from the device of Figures 1 to 4 ;

Figures 7 to 10 show respectively for a second example of a device according to the present invention a transverse cross-sectional view of a composite electrode, an elevation of an electrode segment, an exploded perspective view of a composite electrode, and a longitudinal cross-sectional view of the second example;

Figures 11 and 12 show schematically elevations of further examples of an electrode and an insulating plate; Figure 13 is a schematic transverse cross-sectional view of a further example of a device according to the present invention;

Figure 14 is a schematic longitudinal cross-sectional view of a device according to the present invention having a battery of cells for producing a current;

Figure 15 is a trace of the current output by the device of Figure 14;

Figures 16A and 16B are a schematic longitudinal cross-sectional view and an end view respectively of an example of a device according to the present invention when used as a motor;

Figure 16C is a schematic drawing of apparatus for producing a uniform magnetic field;

Figures 17A and 17B show schematically examples of the connection of two devices according to the present invention, with one device acting as a current source and the other device acting as a motor;

Figure 18 is a perspective view of a composite electrode for showing electrical leakage;

Figures 19 and 20 are part cross-sectional views showing two ways of mitigating electrical leakage;

Figures 21 to 24 are diagrams for explaining use of the device as a switch;

Figures 25 and 26 are diagrams for explaining use of the device as a rectifier or an inverter; Figures 27 and 28 are schematic circuit diagrams showing examples of uninterruptible power supplies according to the present invention;

Figures 29 and 30 are partial perspective views of an embodiment of electrodes for and according to the present invention; and,

Figure 31 is an elevation of a" example of an electrode or an insulating plate for and according to the present invention.

Basic Device and Electrode Arranσements

Referring to Figures 1 to 4 of the drawings, there is shown a first example of a device 1 according to the present invention. The device 1 has a cylindrical outer wall or housing 2 of circular cross-section and made of electrically insulating material. It will be understood that an electrolyte is not shown IT. Figures 1 to 3 for reasons of clarity. The device 1 r.as a horizontal longitudinal axis X. An axle 3 is coaxial with the longitudinal axis of the device 1. A first composite electrode 4 in the form of a thin circular disc is coaxial with the longitudinal axis of the cevice 1, the plane of the electrode 4 being perpendicalar to the longitudinal axis of the device 1, and is positioned to one end of the device 1. A second substantially identical electrode 4 is positioned at the opposite end of tr.e device 1.

A first portion of each electrode 4, representing up to half of the disc-like electrode J, forms an anode 5 which is substantially a semi-circular disc. A second portion of each electrode 4, representing up to half of the electrode 4, forms a cathode 6 Λ/hicn is also substantially a semi-circular disc. The anode 5 and the cathode 6 of each electrode 4 are separated by an insulating strip 7. The insulating strip 7 may simply be a gap, containing no electrolyte, or more preferably may be some solid insulator. The anode 5, the cathode 6 and the electrolyte may be any suitable combination of materials to provide for example a primary cell or a secondary cell. For a secondary cell, the anode 5 may be lead oxide (Pb0₂) and the cathode 6 may be lead (Pb) and the electrolyte may be sulphuric acid. Any other suitable materials may be used for the anode 5, cathode 6 and electrolyte. Electrical connections are provided to the anode 5 and the cathode 6. In the example shown, the anode 5 has a radially projecting tag 8 which provides an anode terminal 8 projecting through the housing 2.

Similarly, the cathode 6 has a radially projecting tag 9 which provides a cathode terminal 9 projecting through the housing 2. Whilst a single anode 5 and a single cathode 6 are shown per electrode 4, more than one anode 5 and more than one cathode 6 may be provided per electrode 4 as will be discussed further below. Furthermore, one of the electrodes 4 could for example have anodes 5 only and the other of the electrodes 4 could have cathodes 6 only, again as will be discussed further below.

Referring to Figure 2, the same device 1 as shown in Figure 1 is illustrated with an insulating plate 10 in position. Because in this example the electrodes 4 have substantially semi-circular anodes 5 and cathodes 6, the insulating plate 10 is also substantially semi-circular . In the example shown, the insulating plate 10 covers completely the lower half of the electrode 4. However, the insulating plate 10 may be at a different angular orientation in the device 1. In this example, an electrolyte completely fills the device 1. In the example shown in Figures 4A and 4B, the anode 5 of one electrode 4 opposes the cathode 6 of the other electrode 5 and vice versa. In an alternative arrangement not shown in the drawings and briefly mentioned above, one of the electrodes 4 has two anodes 5 and the other electrode 4 has two cathodes 6, in each case each being approximately a semi-circle separated by an insulating strip 7. The anodes 5 of the one electrode 4 in this example oppose the cathodes 6 of the other electrode 4.

In the example shown in Figures 1 to 4, the insulating plate 10 is mounted between the electrodes 4 on the axle 3 for rotation with the axle 3 and the electrodes 4 are fixed relative to the housing 2. Thus, as the axle 3 is rotated, the insulating plate 10 rotates about the longitudinal axis of the device 1 relative to the electrodes 4. As the insulating plate 10 rotates, different parts of the surface of the electrodes 4 are progressively exposed to the electrolyte within the device 1. For example, in Figure 2, the anode 5 of the electrode disc 4 shown is fully exposed to the electrolyte whereas the cathode 6 is fully shielded from the electrolyte by the insulating plate 10. In the configuration shown in Figure 3 in which the insulating plate 10 has been rotated through 90° compared to the configuration shown in Figure 2, the anode 5 and the cathode 6 are both half exposed to the electrolyte and half shielded or occluded by the insulating plate 10. On rotation of the insulating plate 10 through a further 90°, the anode 5 will be completely shielded from the electrolyte and the cathode 6 will be fully exposed to the electrolyte .

It will be apparent to the person skilled in the art that the same results will occur if the electrodes 4 are rotated on the axle 3 and the insulating plate 10 is fixed. As noted above, in the example shown in Figures 1 to 4, the device 1 is entirely filled with electrolyte 11 and the semi-circular insulating plate 10 is used to successively and progressively expose and occlude the anode/cathode pairs. An alternative arrangement would be to omit the insulating plate 10 altogether and to only half-fill the device 1 with electrolyte 11. In this case, the electrolyte 11 will pool in the lower half of the device 1 with the upper space always being occupied by insulating gas such as, for example, air. In this example, the electrode discs 4 are rotated so that the anodes 5 and cathodes 6 are successively and progressively exposed to the electrolyte 11 in the lower half of the device 1 and to air in the upper half of the device 1. Electrically, this is equivalent to successively exposing and occluding the anode 5 and electrode 6 by means of the insulating plate 10.

Referring now in detail to Figures 4A and 4B, there is shown a cutaway side elevation of the first example of the device 1 in which the semi-circular insulating plate 10 is shown initially in the upper half of the device 1. In the example shown in Figures 4A and 4B, the two electrode discs 4 are rigidly mounted parallel to each other within the housing 2 and the insulating plate 10 is positioned in the space between the discs 4 for rotation with the axle. As shown in the drawings, the relative orientation of the anodes 5 and cathodes 6 of the electrodes 4 are reversed so that the anode 5 of one electrode disc 4 is opposite the cathode β of the other electrode disc 4 and vice versa. The cathode 6 of one electrode 4 and the anode 5 of the other electrode 4 effectively form one electric cell A and the anode 5 of the one electrode and the cathode 6 of the other electrode 4 effectively form a second electric cell B. The whole of the housing 2 is filled with electrolyte 11. Because of the presence of the insulating plate 10 between the electrodes 4, at any time instant only the fraction of the anode/cathode pairs 5,6 which are not occluded from each other by the insulating plate 10 are electrically active. Thus, in the configuration shown in Figure 4A, where the insulating plate 10 fully occludes the electrodes forming cell B, only cell A is active and able to generate current. In the complementary configuration shown in Figure 4B, only the cell B is active and able to generate current. The direction of the notional external current flow is indicated by arrows I in Figures 4A and 4B. The conductance of the cells A and B, and hence the short- circuit current, can be taken to be proportional to the lesser electrode area which is exposed to the electrolyte 11.

If external loads are connected across the terminal pairs 8-9,8-9 of the electrodes 4, the output current for each electric cell A,B will vary with the relative angular orientation of the insulating plate 10 as indicated in Figure 5. It will be appreciated that for a constant angular velocity of the insulating plate 10 on the throttle axle 3, the graph in Figure 5 shows current versus time. Figure 6 indicates schematically hew a combined AC output of the electric cells A,B can be obtained by using a primary coil 13A, 13B across each cell which are linked to a single common secondary coil 14.

In the examples shown in Figures 1 to 4 , the insulating plate 10 is mounted on the axle 3 which is free to be turned relative to the electrodes 4 and the housing 2 so that the insulating plate 10 rotates about the longitudinal axis X whereas the housing 2 and the electrodes 4 remain stationary. In an alternative but electrically equivalent arrangement, the electrodes 4 and the housing 2 rotate around the longitudinal axis X of the device 1 whilst the insulating plate 10 is held stationary. In the examples described above, there are two electrode segments 5,6 per electrode disc 4, each electrode segment 5,6 being substantially semi-circular and providing an anode 5 or a cathode 6 as the case may be. However, the number of electrode segments can be greater than two. By way of example, Figures 7 to 10 show an example of a device 1 according to the present invention having four electrode segments 5,6 per electrode 4. In this example, each electrode 4 is formed from a generally planar electrode frame 40 made of an insulating material and on each side of which are four quadrant-shape recesses 41 which are separated from each other by walls 7 of the frame 40, the walls 7 being insulating and corresponding to the insulating strip 7 of the first example described above. Each recess 41 receives a quadrant-shape electrode segment which may be an anode 5 or a cathode 6 as the case may be. Each electrode segment 5,6 has a connection tag 8,9 (not shown in Figure 9) which is displaced from the axis of symmetry Y of the electrode segment 5,6. This off-axis arrangement of the connection tags 8,9 allows the position of the tag 8,9 in the electrode frame 40 to be set as desired during assembly simply by reversing the electrode segment 5,6 m the recess 41. This assists in interconnection of the electrode segments 5,6. The tags 8,9 project through sealable holes provided in the rim of the electrode frame 40. Because each side of the electrode 4 has four electrode segments 5,6, the insulating plate 10 has a quadrant-shape opening 42 of substantially the same size as each of the electrode segments 5,6. It will be seen that in the example shown in Figure 9, the outer vessel is octagonal in cross-section. It w ll be appreciated this does not affect the principles of operation of the device and many other shapes are possible.

Referring now to Figure 10, which shows a device 1 having a battery of two sets 43 of four cells, a set 43 of four cells is formed by connecting a respective quadrant- type electrode 4 to each end of a generally cylindrical electrolyte container section 44. One of the electrodes 4 at one end has cathodes 6 on the side facing the electrolyte container section 44. The other electrode 4 at the other end has anodes 5 opposed to the cathodes 6 of the first electrode 4. The insulating plate 10 is mounted for rotation on an axle 3 and, in the example shown in Figure 10, is positioned very close to the cathodes 6. The electrolyte container section 44 has four longitudinal internal dividing walls 45 which correspond to the four insulating strips or walls 7 of the electrode frame 40 and which radiate from a central hub 46 through which the axle 3 passes. There is therefore formed a set 43 of four cells which are arranged around the axis X of the device 1. Rotation of the insulating plate 10 progressively exposes and occludes the cathodes 6 by virtue of the insulating plate 10 with its quadrant shape aperture 42 passing over the cathodes, in a manner similar to that described for the first example of a device 1 above. It is preferred that the insulating plate 10 be mounted close to the cathodes 6 (or anodes 5). Furtnermore, the surfaces of the insulating strips 7 of the electrode frame 4 are preferably flush with the surface of the electrode segments 5,6. The insulating plate 10 may have a keyed or otherwise non-circular through hole at its centre which engages on a corresponding feature on the axle 3 so that rotation of the axle 3 causes rotation of the insulating plate or plates 10. Suitable fixings (not shown) such as bolts may be used to compress the components togetner.

As will be appreciated from a study of Figure 10, because the electrode frame 40 carries electrode segments 5,6 on both sides, it is a simple matter to connect up plural sets 43 of cells to wnich appropriate electrical connections can be made. Sucn batteries of cells 43 can be assembled in a very simple manner by sliding the individual comoonents onto the axle 3. As will be discussed further below, alternating current can readily be obtained from one of the sets 43 of four cells shown in Figure 10. If the insulating plate 10 is rotated at 3,000 rpm (50 Hz), then the device produces a 50 Hz alternating current. Higher order arrangements for the electrode segments 5, 6 and the insulating plate 10 can be used. For example, an octant geometry is shown schematically in Figures 11 and 12. Figure 11 shows schematically an electrode 4 having eight electrode segments 5,6 and Figure 12 shows schematically an octant insulating plate 10 which has two octant-shape apertures 42 which correspond m size and shape to each of the electrode segments 5,6. With appropriate connections, a device 1 using such an octant geometry can give rise to an alternating current output which is twice the frequency of its rotation. Thus, in an example, an octant insulating plate 10 would have to be rotated at 1,500 rpm in order to obtain a 50 Hz output. In general, with an electrode arrangement having 2^n+~ electrode segments on each side and the insulating plate 10 having 2ⁿ apertures 42, the ratio of the axle rotational frequency to the output current frequency is 1/2" where n = 0,1, 2, etc. A lower rotational speed helps to mitigate the effect of friction and drag on the rotating component due to movement through the electrolyte 11. It should be noted that the total area of the apertures 42 is substantially one quarter of the area of the insulating plate 10 and, in general, for any n, the total area of the apertures 42 will be substantially one quarter of the total area of the insulating plate 10.

Accordingly, the total effective resistance of the set of cells is constant and in particular is independent of the sub-division (n) .

Referring to Figure 13, there s shown another example of a device 1. In this example, the housing 2 is again a hollow cylinder of circular cross-section. In this example, there is a first, outer anode plate 15 and a first, outer cathode plate 16 each of which is almost a semi-cylinder and which are concentrically arranged around the central longitudinal axis X of tne device 1. A longitudinal insulator 17, which may simply be a gap or which may be a solid insulator, is situated on each side of the first almost semi-cylindrical anode 15 and cathode 16 to separate the two electrodes 15,16. A second, inner anode plate 15 and a second, inner cathode plate 16 are concentrically arranged around the central longitudinal axis X of the device 1 within and spaced from the outer anode and cathode plates 15, 16. T e inner anode and cathode plates 15, 16 are again almost semi-cylinders and are separated from each other by a longitudinal insulator 17. The inner anode plate 15 opposes the outer cathode plate 16 and vice versa. The insulating plate 10 m this example is semi-cylindrical and is concentrically mounted with respect to the longitudinal ax s X of the device 1 to be positioned between the inner anooe and cathode plates 15,16 and the outer anode and cathode plates 15,16. The anodes 15 and cathodes 16 are arranged to be able to rotate relative to the insulating plate 1C (which may be achieved for example by having the anodes 15 and cathodes 16 rotate whilst the insulating plate 10 is stationary or vice versa) in order to give rise to an external alternating current in a manner similar to that described aoove . Further similar cylindrical concentric insulating p_ates and electrodes of diminishing radius can be provided to produce cells analogous to those described above.

It will be appreciated that otner geometries for the electrodes are possible, including for example annular arrangements of the electrodes. The following description will be restricted principally to tne arrangement in which the electrodes 4 are planar and typically disc-like of generally circular cross-section overall. Current Sources

There will now be described m detail various examples of devices 1 in accordance with the present invention which act as current sources.

It will be appreciated that for devices in which the electrodes 4 and housing 2 rotate with the axle 3 and in which the insulating plate 10 is fixed, extraction of the current from the cells must either be accomplished by means of brushes or slip rings, or through inductive coupling. However, the description of embodiments of the invention that follows assumes that the electrodes 4 and housing 2 are fixed and the insulating plate or plates 10 rotate. Such arrangements will normally be simpler to manufacture and operate, and have much less rotating mass.

A first example of a connected battery current source shown in Figure 14 uses the example of a device 1 shown in Figure 10 and described in detail above. The device 1 has a battery of sets 43 of four active electric cells. In this example, the electrodes 4 are constructed with anodes 5 on one face and cathodes 6 on tne other face. The anode and cathode terminals 8,9 of each electrode 4 project through the housing 2. As shown, one possible way of interconnecting the terminals 8,9 is to make parallel electrical connections 20 to each anode terminal 8. Similarly, parallel electrical connections 21 are made to each cathode terminal 9. Another possible way of interconnecting the terminals 8,9 is to make series electrical connections 20 to each anode terminal 8. Similarly, series electrical connections 21 may be made to each cathode terminal 9. A third possible way of interconnecting the terminals 8,9 is make use of a combination of series and parallel connections 20 to each anode terminal 8 and the same combination to each cathode terminal 9.

A drive shaft 22 is connected to one end of the axle 3 by means of which the axle 3 can be rotated about the longitudinal axis X. The insulating plates 10 have a quadrant shape aperture 42 to correspond to the quadrant- shape anodes 5 and cathodes 6 and are mounted on the axle 3 for rotation therewith. The insulating plates 10 are all mounted so that the apertures 42 are at the same angular position on the axle 3 and are spaced from each other so that an insulating plate 10 is interposed between each anode-cathode pair 5,6. The whole of the vessel 2 is filled with electrolyte 11. Thus, as the axle 3 and therefore the insulating plates 10 are rotated, the various opposed anode-cathode pairs 5,6 are successively and progressively activated by exposure to the electrolyte 11 and an alternating current can be arranged to flow through an external load connected across the terminals 24,25 of the electrical connections 20,21 to the anode and cathode terminals 8,9.

Considering one set 43 of four cells arranged around the axle 3, the cells output current as shown in Figure 15 in which cell 1/4 is diametrically opposite cell 3/4 and cell 2/4 is diametrically opposite cell 4/4 with the cells being arranged in the order 1/4, 2/4, 3/4, 4/4 in the direction of rotation of the insulating plate 10. The output of cell 1/4 is combined witn the inverse of the output of cell 3/4 to provide a first alternating current and the output of cell 2/4 is combined with the inverse of the output of cell 4/4 to provide a second alternating current which is out of phase with the first alternating current by π/2. It will be appreciated that other multiphase alternating current output can be obtained from the device 1 by having groups of insulating plates 10 displaced by a suitable angle with respect to each other or by having groups of electrodes 4 (fixed or otherwise) displaced by a suitable angle with respect to each other and making appropriate connections to pairs of electrodes 4 or a combination of both. For example, in order to obtain a 3-phase alternating current output, three groups of insulating plates 10 or pairs of electrodes 4 can be arranged with 120° displacement relative to each other and respective electrical connections made to the three groups of electrodes 4, or combinations of both.

As an alternative to the above arrangement in which the housing 2 and electrodes 4 are stationary and the insulating plates 10 are rotated, the insulating plates 10 may be held stationary and the housing 2 and the electrodes 4 rotated. As the housing 2 and the electrodes 4 are rotated, clearly the anode and cathode terminals 8,9 rotate as well. In this case, alternating or direct current can be obtained from the device 1 by for example conventional split slip rings or commutator mounted on an external axle which is concentric with the vessel 2. Power may be tapped off from the slip rings or commutator in a conventional manner .

Motors

A brief example of the device 1 used as a motor will now be described.

Referring to Figures I6A and 16B, a device 1 of active cells is placed in a uniform transverse magnetic field 60. In this example, the device I is of the general type shown in Figures 1 to 4 having semi-circular anodes 5 and cathodes 6 with a semi-circular insulating plate 10 though, in this example, the housing 2 and electrodes 4 rotate and the insulating plates 10 are stationary. The anode terminal 8 of one electrode 4 is connected by a first winding loom 61 which runs along the length of the device 1 parallel to the longitudinal axis of rotation X, down one end of the device 1, back along the other side of the device 1 parallel to the longitudinal axis X, across the other end face of the device 1, and again parallel to the longitudinal axis X of the device 1 to the adjacent cathode terminal 9. The other terminals 8,9 in the other half of the device 1 are similarly connected by a second winding loom 62. The looms 61,62 are connected to the terminals 8,9 such that the direction of current flow I on one side of the device 1 is always the same, e.g. to the right for the bottom half of the device 1 as shown. (Whilst the direction of current I is indicated in Figures 16A and 16B, it will be appreciated that current will only flow in the first loom 61 when the device 1 has been rotated through 180°.) As the device 1 rotates, the current in the windings 61,62 varies according to the angular position of the device 1. A maximum current flows in the windings 61,62 at the point where the angular velocity of the coil 61,62 is at right angles to the applied magnetic field 60. This is illustrated in Figure 10B where the current in the windings 61,62 at a particular angular position θ is indicated. In the particular example shown, the plane of the windings 61,62 is arranged to be at a right angle to the length of the insulating gap 7 between the anode 5 and cathode 6 of each electrode 4. With this arrangement, where the angular velocity of the windings 61,62 is parallel to the magnetic field such that no net force is acting on the windings 61,62, the current flowing is zero as indicated for θ = π/2 and θ = 3π/2 in Figure 16B. Where the magnetic field 60 is not uniform, the arrangement and geometry of the windings 61,62 can be arranged accordingly. It may be possible to use more than one anode 5 and one cathode 6 per electrode disc 4.

In Figure 16C, there is shown a practical way of realising the uniform magnetic field 70 described above. A split iron yoke 71 has a cylindrical cavity 72 containing a device 1 wound as described above. An external excitation coil 73 has connections 74 to a direct current power supply (not shown) to generate the required quasi-uniform field 70 transverse to the axis of rotation. As an alternative, permanent magnets could be used to generate the transverse magnetic field.

It will be appreciated that different arrangements of the device 1 are possible, for example using a paraxial rather than a radial field geometry.

Current Control

In the examples of the use of the device 1 as a motor or as a current source in which the electrodes rotate relative to a fixed insulating plate, means for controlling the current generated by the device 1 is preferably provided. This can be achieved for example by locating an electrically insulating throttle baffle adjacent to each insulating plate 10 which is complementary m shape to the associated insulating plate 10 so that the baffle and its associated insulating plate 10 together can form a substantially complete disc. In order to achieve maximum current output, the baffles are orientated so that they completely overlap the non-open portions of the insulating plates 10, thereby leaving the open portion of the insulating plates 10 open to the maximum extent. In order to reduce the output current, the baffles can be angularly displaced relative to the insulating plates 10 to cover as much of the open portions of the insulating plates 10 as required.

Combined Current Sources/Motors

Figure 17A shows schematically how two vessels 1 of active cells can be combined to provide a self-exciting DC motor. A small device 1 acting as a current source 100 shares a central axle 3 with a larger device 1 acting as a motor 101. The current source 100 is electrically connected to the motor 101 so that the current I generated by the current source 100 is used as the excitation current for the motor 101. The rotation of the motor 101 causes rotation of the rotating components of the current source

100 (whether the current source 100 is of the fixed or rotating housing type described above) which in turn provides the drive current I for the excitation coils of the motor 101. (It must be emphasised that this is not a so-called "perpetual motion" arrangement as, eventually, the current source 100 will discharge and there are also inevitably losses in the arrangement.) The motor 101 is provided with an output shaft 102 for driving other devices as required.

Another arrangement is shown in Figure 17B. In this example, a small device 1 acting as a motor 101 shares an axle 3 with a larger device 1 acting as a current source 100 so that the motor 101 drives the rotation in the current source 100. The current source 100 and the motor

101 are electrically connected so that a small portion i of the current generated by the current source 100 is fed back to the motor 101 as an excitation current. The bulk of the current I generated by the current source 100 is tapped for external use. Mitigation of Electrical Leakage

Where the insulating plate 10 has to rotate continuously, it is desirable that minimal power is consumed in rotating the insulating plate 10. A key factor in this power consumption is the degree of frictional force opposing this rotation. This frictional force is due to two possible causes, namely (1) mechanical contact between the rotating insulating plate 10 and stationary electrodes 5,6 or other stationary parts, and (2) the finite viscosity of the electrolyte 11 through wnich the insulating plate 10 has to pass. To effectively eliminate the mechanical frictional force, tne insulating plate 10 should either (l) at no time come into direct contact with either the electrodes 5,6 or the insulating material 7 separating them, or (ii) be made from a very low frictional material, such as for example PTFE. In the first case, however, this means that a certain amount of current leaks through the necessary gap, and hence past the insulating plate 10.

From the point of view of electrical sealing of the cell, there should be no gap between the insulating plate 10 and the electrodes/insulators. Therefore, a design compromise must be sought.

To determine tne magnitude of the effective leakage current due to a finite gap between the shutter and the electrodes, consider the resistance of a thin sheet of electrolyte of resistivity p of length L and cross- sectional area A. The resistance of this sheet is well approximated by the expression R = pL/A. The cross- sectional area of the sneet A is g m² per unit depth, where g is the thickness of the sheet. If g is interpreted as the clearance between tne electrodes/insulator and the insulating plate 10, then the leakage resistance due to this clearance is R = pL/g per unit depth. It is desired to make R - infinity, therefore g -> 0 or L -> infinity. In practice, since g must be finite, L must be made as large as possible.

The principal electrical leakages in the device 1 are indicated schematically in Figure 18. The cross electrode resistance Rx is increased by making the breadth of the insulating arms 7 on the composite electrode 4 as large as possible. Clearly, this breadth is only increased at the expense of the electrode area (and hence battery power to weight/volume ratio), and a compromise must be sought. An additional way of reducing the cross electrode current leakage is to make the polarities of adjacent electrodes the same. This means that on one face of a composite electrode 4, all of the electrode segments are preferably of the same polarity, as has been described above for the preferred embodiments.

The radial resistance Rr can be increased by making the leakage path over the rim of the insulating plate 10 as long as possible. One way of achieving this is to extend the insulating plate 10 past the peripheries of the electrodes 4 and, further, to provide a serpentine path for leakage by for example having a castellated surface 80 on the extended portion of the insulating plate 10 as indicated schematically m Figure 19 which can moves in a correspondingly shaped portion of tne housing 2 or other stationary part.

Another way to provide electrical sealing with a finite clearance between the insulating plate 10 and the electrodes 4 is to provide guard rings 81 on a portion of the insulating plate 10 wnicn extends past the peripheries of the electrodes 4 as indicated in Figure 20. The guard rings 81 provide an additional electric field in the vicinity of the electrodes 4 such that the field gradient between each electrode 4 and the nearby guard ring 81 is effectively zero. This means that current is inhibited from flowing from the electrodes 4 past the guard rings 81. Instead, a "sacrificial" current flows between the guard rings 81. If the guard rings 81 are made sufficiently small in exposed area, then this current is also small.

Charging of Cells

Where the device 1 acts as a secondary (i.e. rechargeable) cell, recharging of the cells can be effected by connecting a suitable external power supply across the electrode terminals. A DC or AC external supply can be used.

If a DC supply is used, then suitable reversed connections need to be made to opposite pairs of cells, and to ensure even exposure the insulating plates 10 must be rotated at a non-zero speed. Otherwise the one or two cells exposed by the insulating plates 10 will recharge, and the sealed one(s) will not.

Where an AC supply (of appropriate voltage) is used, this can be directly connected to the output terminals of the cells and the insulating plates 10 rotated at the supply frequency. In this way, each cell will receive a half-wave of charge of the correct polarity. With appropriate connections, a charge/discharge scheme can be set up whereby a cell is charged on one cycle and discharged on the next. Phase Switch

An example of the device 1 used as a phase switch will now be described with reference to Figures 21 to 24.

Referring first to Figure 21, there is shown an insulating plate 10' for an active quadrant cell 1 of the type having four quadrant-shape electrode segments 5,6 on a face of a composite electrode 4. The insulating plate 10 ' of this example is a double-hole quadrant insulating plate 10 ' as opposed to a single-hole plate 10 as shown in Figure 9 for example. This shape means that any electrode 4 of the device 1 is exposed to the electrolyte twice per revolution, instead of only once in the examples described above. In conjunction with appropriate switching capabilities, a charge/discharge regime can be set up in which a cell is charged on one half cycle 200 and discharged the next half cycle 201 of an alternating current as indicated schematically in Figure 22. Figure 23 shows schematically a switching arrangement 210 necessary to achieve the above charge/discharge regime in which the mains supply input 110 and AC output 112 from the device 1 are active simultaneously and there is minimal leakage from mains to load.

The switch functions 210 required for the charge- discharge scheme mentioned above can be realised using a passive quadrant cell arrangement of a device 1 according to the present invention in which each of the opposed faces of the composite electrodes have only two diametrically opposed inert electrode segments, the electrode segments of the opposed faces respectively opposing each other. A triple quadrant insulating plate 10 (i.e. with one quadrant -shape through hole) is interposed between these electrodes 4 as in the example described above. This arrangement has the required characteristic for each of the switches 210, etc. A full realisation of the required charge/discharge regime, in terms of these phase switches 210 and a double-quadrant insulating plate 10' device 1, is illustrated schematically in Figure 24. It should be noted that all of the cells, both passive and active, in this arrangement are driven from the same axle (not shown) and hence rotate strictly synchronously.

The other phases of the device 1 can be handled in the same way as above using either the other phases of the switches 210, or, alternatively, a duplicate set of switches 210 with appropriate angular orientation. In this way, four (or eight) switches 210 can, in principle, service a whole battery of cells 1.

Rectifier and Inverter

Using passive composite cells, an input AC current can be rectified to produce a (time-varying) DC in a load. As is well-known, full or half-wave rectified DC can be smoothed to produce a more-or-less ripple-free DC. Applications of rectifiers are, again, well-known, especially in terms of semiconductor (or thermionic valve) diodes. However, the present invention can be used in high-power applications, where the use of semi-conductor or other diodes is expensive or prohibitive. The present invention also provides a much more spatially compact solution than high power semi-conductor diodes.

The operation of a device 1 according to the present invention used as a full-wave rectifier 219 is shown in Figure 25 using conventional single-pole single-throw switches 220,221. If on the first half phase switches 220 A and D are closed and B and C 221 are open, DC will flow in the load 222 as indicated. On the next half-phase, if these switch states are reversed (i.e. A & D open, B & C closed) the current will still flow in the direction indicated through the load 222. Accordingly, a full-wave rectification of the input current 110 is achieved.

Figure 26 shows the conventional switches of Figure 25 replaced by two (passive) quadrant devices 225 of the present invention as described above, each mounted on the same axle (not shown) and driven by a low-power induction motor (not shown) which is therefore substantially synchronised to the input ma s AC. Thus, two quadrant devices 225 suitably interconnected and driven can act as a single phase full-wave rectifier 219' .

It should be noted that, because the full-wave rectifier 219' contains purely passive components, it can be driven in reverse to become a DC to AC converter (or "inverter") . Hence, not only can a high-power DC source can be built from the passive device acting as a full-wave rectifier, but also, an inverter can be made from the passive device acting as a DC to AC converter. Applications of this include a mams to DC battery charger, and a battery-powered UPS for mams (AC) backup.

More complex arrangements could be used to rectify multi-phase AC, using additional devices.

Uninterruptible Power Supply

A principal application of the present invention is as a current source in an uninterruptible power supply. Two examples of suitable arrangements are shown in Figures 27 and 28.

Referring first to the example shown in Figure 27, a device 1 acting as a current source is connected to a mams power supply 110 via a first isolating relay circuit 111. The mains power supply 110 is connected so as to be able to charge the device 1 and also has a mams output 112. The current output of the device 1 has a power extraction arrangement 113 (which may for example be wired or inductive coupling as described above) which in turn is connected via a second isolating relay 114 to the mams output 112. A first power quality sensor 115 monitors the quality of the mams power at the mams input 110 and operates the first isolating relay 111 accordingly. For example, if the mams power supply becomes faulty (for example if the voltage drops below a predetermined level), the first power quality sensor 115 causes the first isolating relay 111 to open to isolate the mams input 110 from the mams output 112. At the same time, assuming that the device 1 is adequately charged, a second power quality sensor 116 which detects the quality of the power output by the device 1 operates the second isolating relay 114 to close so that the device 1 can supply power to the ma s output 112. If the quality of the mams power supply at the mams input 110 is restored to an adequate level, the first power quality sensor 115 operates to close the first isolating relay 111, allowing the device 1 to be recharged and also directly supplying mams power to the mams output 112. If the device 1 is discharged at the time that the mams input fails, both power quality sensors 115,116 operate to open both relays 111,114 so that any equipment connected to the mams output 112 does not receive a power supply which is out of specification. Similarly, if the device 1 is supplying power to the mams output 112 but discharges below a predetermined minimum performance level, this is again detected by the second power quality sensor 116 wnich opens the second relay 114, again to protect the connected equipment from receiving a power supply which is out of specification. It w ll be appreciated that, in the arrangement shown, at all times when the mams input quality is at a sufficient level, the device 1 is charged by the input mains. Furthermore, the changeover between mains power supply and the back-up power supply from the device 1 can be effected, in principle, without any supply interruptions as far as any equipment connected to the mains output 112 is concerned. Figure 27 shows a motor 117 connected to drive the rotatable part or parts (whether the insulating plates 10 or the electrodes 4) of the device 1 in accordance with a motor speed regulation circuit 118 which itself is controlled by the second power quality sensor 116.

In the example of an uninterruptible power supply shown in Figure 28, the mains supply 110 is effectively isolated from the mains output 112, which is in contrast to the arrangement shown in Figure 27 in which unfiltered mains power is fed directly to the mains output 112 when the quality of the mains power supply is adequate.

The mains power supply input 110 of the example of Figure 28 is connected via a first isolating relay 120 to a phase switch 121 which is integrated with the device 1. The phase switch 121, which may be of the type discussed above, is rotated with the rotating part of the device 1 by a common drive motor 122. A first power quality sensor 123 detects the quality of the mains input power 110 and, when the quality drops below a predetermined level opens the first isolating relay 120 to isolate the mains power supply from the device 1. When the power supply at the mains input 110 is good, the first isolating relay 120 is opened so that mains power can pass into the phase switch 121 from where it is passed to the device 1.

A power extraction arrangement 124 is connected to tap power from the device 1. A second power quality sensor 125 detects the quality of the power extracted from the device 1. The extracted power passes via a second isolating relay 126 to the mains output 112. The second power quality sensor 125 operates to open the second isolating relay 126 if the quality of the power extracted from the device 1 falls below a predetermined threshold and otherwise operates to leave the second isolating relay closed in order to supply power to the mams output 112. A motor speed regulation circuit 118 is connected between the ma s output 112 and the drive motor 122 to drive the motor 122 appropriately. As will be appreciated, m this arrangement the device 1 constantly charges when the quality of the ma s power supply is above a predetermined level and constantly supplies power to the mams output 112 provided that the quality of the power output by the device 1 is above a predetermined level. In this way, equipment connected to the mams output not only has an uninterruptible power supply, but the power when supplied is normally of a good quality in that for example it will not have spurious voltage artefacts found in most mams power supplies .

In either of the examples described above, the power quality sensors 115,116,123,125 can operate in various ways. For example, the power quality sensors may simply measure the RMS input voltage and operate to open tne respective relays if the RMS input voltage is outside pre- set limits.

Miscellaneous

It is possible to increase the surface area of the electrodes 4 in any of the examples described above, for example by using rough surface electrodes 4 or foraminate electrodes 4. Examples of foraminate electrodes 4 are illustrated in Figures 29 and 30. As illustrated in Figure 29, the electrode 4 may have through holes 130 having a major axis which is perpendicular to the plane of the electrode 4. As well as increasing the effective surface area of the electrodes 4, the holes 130 also reduce the overall mass of the electrodes. In the examples of the vessels 1 where the electrodes 4 rotate, this allows for higher acceleration and deceleration of the rotating components. As illustrated in Figure 30, the holes 130 can be at an oblique angle to the plane of the electrodes 4, thereby increasing further the increase in surface area and decrease in volume and mass of the electrode 4. The oblique arrangement of the holes 130 also helps to ensure that electrolyte is forced to flow around the holes 130, thereby helping to prevent the holes 130 becoming blocked.

The semicircular anodes and cathodes 5,6 described above give rise to a time-varying output current which is dependent on the load resistance and the resistance of the cell. In short-circuit (i.e. where the load resistance is zero) , this output waveform is substantially triangular as shown for example in Figure 5. In all cases, the output waveform for finite loads has frequency components of odd multiples (3,5,7,9, etc.) of the rotation frequency. In the case of a triangular waveform, the amplitudes of these components are proportional to the inverse square of their relative frequency to the rotation frequency (1/9, 1/25, 1/49 etc) . The harmonic content of the output waveform varies with the applied load in general. To obtain a pure sinusoidal output for a wide-range of load values, it is possible to use a low-pass filter which removes all harmonics from the output waveform except the fundamental. To vary the harmonic content of the output, without recourse to filtering, it is possible to arrange for a more complex overlap relationship between the electrodes and the insulating plate. For example, to obtain a sinusoidal time-variation at the rotation frequency at short-circuit with a triple-quadrant insulating plate, the electrodes (anode and cathode) may have the shape indicated schematically in Figure 31 and the relation r(θ) given below, where r (θ) is the distance between the axis of rotation and a point on the rim of the electrode, θ is the angular position of the point on the rim, and A is the total area of the electrode. Alternatively, the electrodes may be substantially semi-circular and the insulating plate circular with a hole in one half described by r (θ) given below, where the r and θ now refer to the hole, and A is the area of the hole. The formal description of Figure 31 (neglecting the small central axle hole) is as follows:

It will be appreciated that the shape of the electrodes 4 and/or insulating plate (s) 10 can be tailored as required m accordance with these principles in order to give a desired output waveform.

An emoodiment of the present invention has been described with particular reference to the examples illustrated. However, it will be appreciated that variations and modifications may be made to the examples described withm the scope of the present invention.

For example, whilst in the embodiments shown where the insulators 10 are semi-circular solid plates, all of the insulators 10 are shown on the same side of the axle 3, in order to balance the axle 3, the insulators 10 may be on alternate sides along the axle 3. Other means for balancing the axle 3, such as by the use of balance weights on the axle 3, may be provided. By way of a further example of a variation, in order to reduce the viscous drag of a (substantially) semicircular insulator plate 10 moving through the electrolyte 11, especially because of the straight leading edge, the insulator plates 10 may be fully circular discs for example, with one semi-circular portion being insulating and the other semi-circular portion being ionically conductive (for example, by perforating the plate in that portion) .

Claims

1. An electric device, the device comprising: a first cell having a first anode and a first cathode and electrolyte for conducting charge between the first anode and the first cathode; a second cell having a second anode and a second cathode and electrolyte for conducting charge between the second anode and the second cathode; and, an insulator which is movable relative to the first anode and the first cathode and relative to the second anode and the second cathode such that movement of the insulator varies the conductance of each of the first and second cells.

2. A device according to claim 1, wherein the insulator is arranged to be rotatable relative to the first anode and the first cathode and relative to the second anode and the second cathode and to be selectively positionable between the first anode and the first cathode and between the second anode and the second cathode .

3. A device according to claim 1 or claim 2, wherein the insulator is arranged such that selected rotation of the insulator relative to the first anode and the first cathode and relative to the second anode and the second cathode increases the conductance of the first cell whilst decreasing the conductance of the second cell.

4. A device according to any of claims 1 to 3 , wherein the insulator has an open portion through which ions can move between an anode and a cathode when the open portion is positioned between said anode and said cathode.

5. A device according to any of claims 1 to 4 , wherein the insulator is stationary and the respective anode and cathode of each cell are rotatable together.

6. A device according to any of claims 1 to 4 , wherein the anode and the cathode of each cell are stationary and the insulator is rotatable.

7. A device according to any of claims 1 to 6 , wherein the electrolyte is common to the first and second cells.

8. A device according to any of claims 1 to 7, wherein the anodes of said two cells are arranged on one side of the device and the cathodes of said two cells are arranged on the opposite side of the device, and wherein the insulator is arranged between the anodes and the cathodes to rotate relative to the anodes and cathodes selectively to expose more of the electrolyte to the anode and cathode of one of the cells whilst exposing less of the electrolyte to the anode and cathode of the other cell.

9. A device according to any of claims 1 to 7 , comprising one or more additional cells each having an anode and a cathode and electrolyte for conducting charge between said anode and said cathode, the insulator being movable relative to the anode and the cathode of said one or more additional cells to vary the conductance of said additional cells.

10. A device according to claim 9, wherein the anodes of the cells are arranged on one side of the device and the cathodes of the cells are arranged on the opposite side of the device, and wherein the insulator is arranged between the anodes and the cathodes to rotate relative to the anodes and cathodes selectively to expose more of the electrolyte to the anode and cathode of at least one of the cells whilst exposing less of the electrolyte to the anode and cathode of at least one of the other cells.

11. A device according to any of claims 1 to 10, comprising an external load connected across the anode and the cathode of at least one of the cells, the device being arranged to drive current through said external load.

12. A device according to any of claims 1 to 11, comprising inductive current extraction means for inductively extracting a current from said device.

13. A device according to any of claims 1 to 12, comprising a second insulator which is operable to determine the maximum exposure of the anode to the cathode of at least one of the cells thereby to determine the maximum current available from that cell.

14. A device according to any of claims 1 to 13, comprising windings connected across the anode and the cathode of at least one of the cells through which a current generated by the device can pass so as to generate a magnetic field which can interact with an externally applied magnetic field to drive the device as a motor.

15. An uninterruptible power supply for supplying power to a load on failure of a mains supply, comprising a device according to any of claims 1 to 13 acting as a current source to supply power to a load on failure of a mains supply.

16. A rectifier for converting alternating current to direct current, the rectifier comprising a device according to any of claims 1 to 13.

17. An inverter for converting direct current to alternating current, the inverter comprising a device according to any of claims 1 to 13.

18. A method of operating an electric device which comprises a first cell having a first anode and a first cathode and electrolyte for conducting charge between the first anode and the first cathode and a second cell having a second anode and a second cathode and electrolyte for conducting charge between the second anode and the second cathode, the method comprising the step of moving an insulator relative to the first anode and the first cathode and relative to the second anode and the second cathode to vary the conductance of each of the first and second cells.

19. A method according to claim 18, comprising the step of rotating the insulator relative to the first anode and the first cathode and relative to the second anode and the second cathode to increase the conductance of the first cell whilst decreasing the conductance of the second cell.

20. A unitary electrode for an electric device, the electrode being substantially planar and comprising at least a first electrode segment and a second electrode segment, and an insulator for insulating the first electrode segment from the second electrode segment .

21. An electrode according to claim 20, comprising two electrode segments each of which is substantially semicircular.

22. An electrode according to claim 20, comprising four substantially quadrant shape electrode segments on a face of the electrode each insulated from each other by an insulator .

23. An electrode according to any of claims 20 to 22, wherein at least one of the electrode segments has a plurality of through holes.

24. An electrode according to claim 23, wherein at least some of said through holes are cylindrical, the cylindrical axes of said through holes being at an acute angle to the plane of said electrode.