GB2376720A

GB2376720A - Fluid-propelling device

Info

Publication number: GB2376720A
Application number: GB0115065A
Authority: GB
Inventors: Anthony Hooley; Ursula Ruth Lenel; Mark Richard Shepherd; David Pearce; Gareth Mckevitt
Original assignee: 1 Ltd
Current assignee: 1 Ltd
Priority date: 2001-06-20
Filing date: 2001-06-20
Publication date: 2002-12-24
Anticipated expiration: 2021-06-20
Also published as: GB0115065D0; GB2405256A; GB2405256B; GB0425037D0; GB2376720B

Abstract

A fluid propelling device comprises a fan blade 2 mounted in a support body such as a duct 1. The blade 2 is linked to an actuation means which is capable of causing ends 3,4 of the blade 2 to move in a sequence which propels a fluid in a consistent direction 5,6. There may be a number of actuation means, and a number of blades 2 may be arranged in series along the support body or one above another ( see figures 10-11B). The actuation means may be an electro-active device ( see figures 3-5 ) or may be a means which is capable of driving the blade 2 with both linear and rotational displacement (see figure 9). A device is also discussed in which the blade is annular ( see figures 12A, 12B ) with means for causing the inner and outer peripheries to move in sequence.

Description

Fluid-propelling device This invention is concerned with fluid-propelling devices-that is, devices for propelling gases and liquids-and relates in particular to novel varieties of fan. A further aspect of the invention is concerned with an electro-active device which is capable of both linear and rotational displacement.

Devices for propelling fluids can take a number of forms, and can have a number of uses. Most seem to be of what might be called the"propeller"variety ; they consist of a number of angled (sometimes aerofoil-section) blades-from two up to as many as 12 or more, mounted symmetrically on and around, and extending radially from, a central boss itself mounted for rotation on an axle, and they work in rotational mode, that is, they are spun round in a chosen direction, at some appropriate speed, so as to deflect the ambient fluid as they turn, and so push it "forwards"to achieve the desired purpose. Thus, such air-propelling devices can be relatively slow-speed stationary fans, typically for blowing cooling air into an environment or over an object, to control its temperature, or they can be relatively high speed"propellers"as used in aircraft to blow air rearwards with sufficient speed to move the aircraft forwards. Of course, liquid-propelling fans also exist, moving coolant fluid around some heated volume, as do propellers, as found in water-borne craft like boats and submarines.

Other types of fluid-propelling devices also exist, typified, perhaps, by a simple hand-held fan, as might be waved back and forth (with a reciprocatory or oscillatory motion, rather than a rotary one) by a 19th-Century lady to waft air across her face, to cool her down. Another example of such a reciprocatory fan is the much larger device found in, say, India (and called a punkah: these tend to be large fans mounted on the end of a pole and either held and operated by a servant or pivotally suspended from a ceiling and operated by a servant pulling a cord suitably attached thereto).

For the most part the devices presently employed to move fluids such as air and water are of the rotary"propeller"variety ; there seem to be few modem uses for

devices, like the punkah, in the form of reciprocating fans, although there are some.

The present invention proposes a novel form of such a device.

Fluid-propelling devices-for the most part referred to hereinafter for convenience as"fans", which latter term therefore includes objects like propellersthat are of the rotary type tend to be noisy, for to move sufficient amounts of fluid they have to be either quite large or they have to spin fairly fast, and all the resulting turbulence wastes energy and causes noise. The reciprocatory versions, on the other hand, tend to move amounts of air more efficiently whilst being either smaller or moving more slowly, and thus they are relatively quiet and energy-efficient.

Basically, the reciprocatory type comprises an elongate area-extensive blade pivoted at one end and operatively linked to a mechanical actuator that in use causes the blade to pivot back and forth in a reciprocatory fashion, so wafting away the relevant ambient fluid (in which the fan is placed) just like a punkah or a lady's fan. The actuator could in principle take any form capable of providing the correct driving force to the blade; it could, for example, be a rotary device operating some sort of blade-thrusting cam, or it might alternatively be a solenoid-type device magnetically pushing a blade-contacting drive rod back and forth. However, it has relatively recently been proposed that the actuator should instead be a piezoelectric transducer converting electric signals into a shape or dimension change in a simple block of material, and so being in effect a solid-state device having essentially no moving parts.

However, such a simple fan device, whilst being undoubtedly useful, is unable to produce any really substantial fluid flow, and often suffers from the problem that the flow tends to be oscillatory, the fluid being blown in one direction with one stroke and, at least partially, sucked back on the reverse stroke. As a result attempts have been made to improve the design to achieve greater and more directionally-consistent flow rates. In one device, for example, there have been deployed two such blades side by side and operating 180 degrees out of phase so that as one blade waves in one direction the other waves in the opposite direction; in theory this should provide approximately twice the flow as well as a more consistent

direction of flow.

Even these devices, though, are not entirely satisfactory, and the invention seeks to provide a yet further improved version. More specifically, the invention suggests a device, and most preferably a ducted device (one in which the fan blade waves back and forth within a duct, or passageway, along which the fluid is blown), wherein the blade moves back and forth with a phase difference between the ends of the blade so they do not move in synchronism. For example, instead of being pivoted at one end and moving fan-wise at the other, the blade can be moved fan-wise at each end in sequence while at the other end it is temporarily pivotally mounted. Thus, first End 1 moves up, say, while End 2 remains down as though pivoted, then End 2 moves up while End 1 remains up as though pivoted, then End 1 moves down while End 2 stays up, and finally End 2 moves down while End 1 remains down, and the four-part cycle can begin again.

Such a fan is referred to hereinafter as a"Quadrature"fan, because one end moves out of phase with the other, the timing typically being 90 , or one quarter of a cycle, behind the other. The tenn "quadrature" is common in electronics, where signals take on a sine wave shape, and if two are 90 degrees apart they are said to be "in quadrature". It should be noted, though, that in the case of the present invention the term"Quadrature"does not imply anything about the"shape"of the waveform nor the precise degree of phase difference. More specifically, it does not mean, as it would in an electronics setting, that either end of the fan necessarily moves in a manner reminiscent of a sine wave or necessarily moves with a precise 900 phase difference. In the case of the present invention the term"Quadrature"is generalised to mean simply that the two ends of the fan blade move with different phases, so that they move at different times.

The great, and perhaps surprising, advantage of such a quadrature system is that the"fanning"end of the blade is always moving the fluid being fanned in the same direction, pushing it out at one end of the duct while dragging it in at the other.

Moreover, the fan blade need not move particularly fast, so it can be efficient and almost noiseless.

In one aspect, therefore, this invention provides a fluid-propelling device comprising an elongate area-extensive fan blade mounted along a support body for either free or pivotally-restrained movement at either end, which blade is operatively linked to a mechanical actuator that in use causes the two ends of the blade to move back and forth in Quadrature such that the relevant ambient fluid moves at all times predominantly in the same direction.

As is discussed in more detail hereinafter, the actuator or each actuator-there may be more than one-is most preferably a piezoelectric transducer converting suitable electric signals into a shape or dimension change which results in the desired movement of the blade.

The fluid-propelling device (the fan) of the invention may have many uses, though to some extent its form, the details of its size, its construction and its operation, will depend upon what its intended use is. Indications of the value of these parameters suitable for some chosen purposes are given hereinafter. Possible uses range from relatively small specialised air-moving cooling fans, for controlling the temperature of electronic components such as semiconductor chips inside a computer, for instance, to drive units for water-borne vessels. The fan is particularly suited for causing quiet, smooth-flow (low-turbulence) low-pressure fluid movement, and such uses include: electronic-component cooling fans; domestic and industrial room (air) cooling fans; fish-tank or fish-pond water circulation units; and stirrers for chemical reaction vessels.

In overall size and fluid-moving capability the fan of the invention can range from the relatively tiny, a computer-chip cooling fan might be anywhere from 50mm (2in) down to 10mm (0. 4in) across, and move around 7 litre/min of air, to the relatively large, a domestic fan might be 30cm (12in) across and move 90 l/min. A fish tank pump unit might be 1cm (0. 4in) across and move 0. 1 1/min water, while a small pleasure boat drive unit might be 25cm (lOin) across, and move 500 l/min water. An ocean liner might use several individual drive units each 3m (10ft) across moving water at 90,000 1/min.

The invention's fan uses an elongate area-extensive fan blade, or vane. This

is mounted for movement at either end so it can pivot or move back and forth in a reciprocatory fashion, wafting the relevant ambient fluid in an appropriate direction.

The two ends of the blade are driven to move back and forth in Quadrature such that the relevant ambient fluid moves at all times in the same direction. The blade is area-extensive in the sense that it has significant length and breadth, so that it has a substantial area relative to the overall size of the device and is capable of propelling fluid when moved.

For example, for a small device like a chip-cooling fan the blade might be a rectangular object 30mm (1.2in) long in the direction of fluid flow and 25mm (lin) wide in the direction across the flow, whereas for a domestic room cooling fan it might be the same general shape but ten times as big. In general a rectangular blade is likely to be most convenient, but of course the blade need not be rectangular and can be any suitable shape. The use in a device of a quite different blade shape is discussed hereinafter.

The fan blade may be any suitable material, and may be constructed in any appropriate manner. However, the blade must be sufficiently rigid to move the relevant ambient fluid. High stiffness in the blade is advantageous because it allows a wide spacing of actuators (see below). On the other hand, some flexibility or elasticity may be desirable to allow lateral displacements (as also discussed further below) or to allow displacement of the blade to propagate along the blade like a wave. Further, it is desirable that the blade be of minimum weight, so that the energy required to accelerate the blade, rather than the fluid, be minimised. In any event, the blade is preferably much thinner than it is broad or long (probably by at least an order of magnitude). The blade may be constructed from a simple sheet of low cost, light weight material, such as a suitable plastic, or it may advantageously be of a composite construction, for instance a fairly rigid frame supporting a flexible diaphragm (conveniently a fabric or a polymer film).

The invention's fan utilises a fan blade mounted along a support body for either free or pivotally-restrained movement at either end. This body can notionally take any form (so long as the actuator can be secured to it), and therefore could be a

simple plain surface of a wall, casing or the like. However, when mounted on such a flat wall-like surface, and operated so that it waves back and forth in space, as a lady's fan or a punkah waves freely in the air (although, of course, in the device of the invention the ends of the blade are moving in Quadrature), the ambient fluid, though being moved predominantly from one end of the blade to the other, will also move in other directions, thus, from side to side and to and fro. To prevent such unwanted movement of air, which could represent a considerable input energy loss, the blade is very preferably situated in a duct. The blade may be mounted along the duct in the sense that it extends along the duct and pivots on a line across the duct; as it operates, then, it pushes the ambient fluid along the duct.

The duct is advantageously as long as, or longer than, the blade, while the duct width and height are chosen such that in operation the blade makes a sufficiently close fit to largely prevent movement of air in unwanted directions. Other than that, the duct may be any suitable size and shape, and its aperture may be constant or varying along its length-for instance, the duct may flare towards one end, typically the input end. A convenient form, however, is a rectangular duct with constant aperture.

A particularly interesting form of duct, namely an annular duct with a fan blade to match, is discussed in more detail further hereinafter.

At each of its ends the fan blade can be either free or allowed some pivotally-restrained movement. Though there are a number of ways of achieving this, that preferred for the invention involves using at least two actuators, preferably one at each end or spaced along the blade, to which the blade is linked in a pivotal manner preferably loosely. In this way pivotability may be attained by one actuator slightly (and temporarily) restraining movement of the blade at the relevant one end while the second actuator moves the second end in the manner desired (at the beginning of a cycle, starting with the blade parallel to and against the support body surface, this would be away from that surface), and then having the first actuator drive its end in the required fashion (away from the support surface) while the second actuator similarly restrains its end. Thus, either end of the blade can act as the

"pivot"end while the other acts as the driven end, and to achieve the desired Quadrature effect each can so act sequentially.

In operation the blade lies along the support body, the beginning of a cycle starting with the blade parallel to and against the support body surface. The blade first moves pivotally from that position to a second position where it extends "diagonally"away from that surface. Next, the end of the blade that was"pivoted"to the surface follows the end that first moved, so leaving the blade approximately parallel to but spaced from the support body surface. That is halfway through the cycle-which is then completed by the first end moving back to the surface wall, followed by the second end.

Where the device is a ducted one, the duct having walls with inner surfaces, the blade moves in a pivotal fashion first between a position where it lies against the inner surface of one wall of the duct and a position where it extends from that one wall into the central volume of the duct and across into near contact with the opposed wall's inner surface. Next, the end of the blade that was"pivoted"to the one wall surface follows the end that first moved, so resulting in the blade lying right against the inner surface of the opposed wall. That is halfway through the cycle-which is then completed by the first end moving back to the first wall, followed by the second end.

During operation of the device of the invention the fan blade's position varies, as discussed above (and to the extent that it is rigid), from lying parallel to the support body surface to extending at an angle thereto. It will be appreciated that this necessarily requires the blade ends to move in arcs, which may give rise to a mechanical problem. A possible solution to this is to use one or more actuator whose motion is not purely linear, but follows the desired path. It is preferable, however, to utilise simple linear actuators, and for each actuator to allow for the lateral movement by incorporating some flexibility in one or more of : the actuator fixing; the actuator itself ; the connection between actuator and blade (in the case where the actuator is connected directly to the blade); the strut or its connections (in the case where the actuator is connected to the blade by a strut); or the blade itself. For example,

connections may be made by means of a flexible adhesive which allow a little movement between the connecting parts, or the blade or connecting strut may be made from a flexible or elastic material.

The same problem can also be tackled by minimising the degree of "longitudinal"movement of the blade ends-which can be achieved by selecting the blade dimensions such that the blade length is very much larger than the distance through which the end moves. The longitudinal displacement of one end relative to the other on going from the"support surface parallel"orientation to the"diagonal" orientation can be calculated by simple geometry; thus the lateral displacement of, for example, a 20mm (0. 8in) long blade with 10mm (0. 4in) vertical displacement is 2. 4mm (0. 1 in), while that for a 30mm (1.2in) long blade with 3mm (O. lin) vertical displacement is less than 0. 2mm (O. Olin). In order to minimise longitudinal displacements, and thus the degree of flexibility required, it is therefore preferred that the blade length is considerably greater than the (vertical) displacement, preferably by at least a factor of 3.

In the invention the fan blade is operatively linked to a mechanical actuator that in use causes the blade to move or pivot back and forth in a reciprocatory fashion. The positioning and operative linking of the actuator can be effected in any suitable fashion, and how this is done will depend to a considerable extent on the nature of the actuator itself, about which more is said below.

The movement of the blade is effected by the actuator means in any suitable manner. There can be, and preferably is, more than one actuator. As already described, in operation the two ends of the blade follow the position cycle in-in, in-out, out-out, out-in; for a horizontally-mounted fan this would be down-down, up-down, up-up, down-up; other mountings are possible, with appropriate cycles). In this terminology, the or each actuator needs to move the blade, at different times, to horizontal lower and upper positions, and to both diagonals (like a forward and a backward slash). Such movement can be effected by a single complex actuator capable of both translational and rotational movement, preferably mounted approximately centrally on the blade, and this forms one preferred device of the

invention. However, other preferred devices of the invention use multiple actuators.

Where there are two or more actuators, each actuator need only produce simple translational movement, for instance up and down. Such actuators are commonplace, examples being electromagnetic linear actuators, hydraulic and pneumatic actuators, and piezoelectric actuators.

In principle, however, a preferred structure of actuator is a device that becomes longer or shorter as it is operated, and is placed between the blade and its mounting, to push the blade to and fro as its (the actuator's) effective dimensions change. In the preferred two-actuator design mentioned above each actuator could be a length-changing device, and in use either end would act as the"pivot"end, loosely held by the quiescent actuator mounted at that end, while the other acted as the driven end, being moved by the active actuator at that end.

For relatively small devices of the invention in particular, e. g. electronic chip cooling fans, the actuator is most preferably a piezoelectric or other electro-active transducer converting suitable electric signals into a shape or dimension change which results in the desired movement of the blade. Electro-active materials are materials which deform in response to applied electrical conditions or, vice versa, have electrical properties which change in response to applied deformation. The best known and most developed type of electro-active material is piezoelectric material, but other types of electro-active material include electrostrictive material and piezoresistive material. Many devices which make use of electro-active properties are known.

The most simple type of piezoelectric device is a block of piezoelectric material actuated in an expansion-contraction mode by applying an actuation voltage in the direction of poling. However, as the piezoelectric effect is small, of the order 10-10 m/V, the change in dimensions is relatively small, typically less than a micron.

Therefore, more complicated electro-active structures have been developed to achieve larger displacements.

A known electro-active structure is the bender construction, for example a bimorph bender construction. With a bender construction, the electro-active

structure comprises a plurality of layers at least one of which is of electro-active material. On actuation, the layers deform with a differential change in length between the layers for example one layer expanding and another layer contracting.

Due to the layers being constrained by being coupled to one another, the differential change in length causes the bender to bend perpendicular to the layers. Accordingly there is a relative displacement of the ends of the structure. However, the relative displacement does not follow a linear path in space. As the structure bends and the degree of curvature increases, the relative displacement of the ends follows a curve in space. Furthermore, to achieve relatively large displacement, it is necessary to increase the length of the structure which therefore becomes inconvenient. For example, to achieve a displacement of the order of 0. 1mm with a bimorph bender construction, a structure of length around 5cm is typically needed.

Such known electro-active devices as a bender may be used as actuators in fluid propelling devices in accordance with the present invention but suffer from the problems discussed above. Therefore preferably the actuators in the fluid propelling devices are each an electro-active device comprising an electro-active structure extending along a curved minor axis and arranged, on actuation, for the structure to twist around the minor axis and concomitantly for relative displacement of the ends of the structure to occur.

The operation of such an electro-active device may be understood as follows.

The relative displacement between the ends of the device occurs concomitantly with the twist of the structure around the minor axis on actuation, because of the fact that the device extends along a curved minor axis. The electro-active device uses the physical principal that twisting of a curved object causes displacement perpendicular to the local curve, and vice versa displacement of the ends of a curved object causes twisting along its length. The displacement is equivalent to a change in the orientation of the minor axis of the structure relative to its original orientation.

The device uses an electro-active structure which twists on actuation.

Considering any given small section of the structure along the curved minor axis it is easy to visualise how twist of that given section rotates adjacent sections and hence

relatively displaces them in opposite directions perpendicular to the local curve of the given section, because the adjacent sections extend at an angle to the given section as a result of the curve of the minor axis. Therefore twisting of the given section is concomitant with a relative displacement of the adjacent sections perpendicular to the plane of the curve. The degree of relative displacement is proportional to the degree of curvature in the given section and the magnitude of the twisting. The overall displacement of the device is the combination of the displacement of each section.

Thus the overall displacement on actuation is a relative displacement of the ends of the structure.

For minor axes which extend along a regular curve around a major axis, such as along an arc of a circle or a helix, on actuation each section produces displacement in the same direction parallel to the major axis. Therefore, the overall relative displacement of the end of the structure is a linear displacement parallel to the major axis. Therefore an electro-active device in accordance to the present invention can produce displacement which is linear in space.

The degree of displacement is proportional to the length of the structure along the minor axis, because each section of the structure contributes to the overall displacement. Therefore any desired degree of displacement may be achieved by suitable design of the device, in particular by selection of the length of the structure along the minor axis and of the type of structure which controls the magnitude of the twisting-field response. As a result of the structure extending along a minor axis which is curved, a relatively compact device may be produced. In general, the curve along which the minor axis extends may be of any shape whatsoever.

Such electro-active devices are capable of producing tip movements of many millimetres, and are therefore suitable for use in fans in which the major blade dimensions (length and width) are of the order of tens of millimetres.

One possibility is for the curve along which the minor axis extends to be planar, for example as the arc of a circle or a spiral. In this case, the displacement on actuation occurs perpendicular to the plane of the curve. The thickness of the device in the direction in which relative displacement occurs is merely the thickness of the

electro-active structure so a relatively thin device may be produced.

Another possibility is for the curve along which the minor axis extends to be a helix. In this case, each helical turn of the structure contributes towards displacement in the direction along the geometrical major axis around which the helix is formed. Therefore a large degree of displacement may be achieved proportional to the number of helical turns, therefore producing relatively high displacement for a relatively compact device.

Preferably, the electro-active structure comprises electro-active portions disposed successively along the minor axis and arranged to bend, on actuation, around the minor axis. concomitantly with twisting of the structure around the minor axis. As a result, the electro-active portion may have any construction which bends on actuation. The preferred construction is the known bender construction comprising a plurality of layers including at least one layer of electro-active material, preferably a bimorph bender construction having two layers. Such a construction is well known and understood as applied to a straight bender and particularly easy to manufacture. The same benefits are obtained when the bender construction is applied to the portions of the present invention. However, any other construction which provides bending on actuation may be used.

Preferably, the electro-active structure comprises a continuous electro-active member curving around the minor axis, said electro-active portions being adjacent finite portions of the continuous member.

This structure is particularly easy to manufacture, for example by winding a deformable continuous electro-active member into shape.

Preferably, the continuous electro-active member curves in a helix around the minor.

By using a continuous electro-active member which curves in a helix around the minor axis a number of advantages are achieved. Firstly, it is easy to provides a structure which is regular along the length of the minor axis and hence provide the same degree of twisting along the entire length of the minor axis. Secondly, the helix is easy to manufacture, for example by winding a deformable continuous member

into shape or by making a helical cut in a tubular electro-active member. Thirdly, the device is compact as the helical turns of the member around the minor axis may be packed closely together.

However the electro-active structure may alternatively comprise a continuous electro-active member having a different shape which provides for bending around the minor axis concomitantly with twisting around the minor axis. For example it may comprise a continuous member having the shape of a flat member twisted around the minor axis. Furthermore, instead of comprising a continuous electroactive member, the electro-active structure may comprise a plurality of electro-active portions coupled together.

Several configurations of actuator and blade are possible. For example, with a basically rectangular blade, four actuators, placed one at or near each comer of the blade, provide considerable force and control. However, a device using only two actuators acting at points spaced along the fan blade, preferably near the ends of the blade, is less complex and less costly to manufacture (two actuators instead of four), and forms a preferred device of the invention.

The actuators may be situated acting on either face of the blade-for a horizontally-disposed blade they may be either under or above the blade (and indeed could be one on each face), and for balanced operation are preferably placed half-way across the width of the blade. However, such actuators are necessarily in the path of air movement and are likely to cause turbulence and inefficiency. Therefore, in a further preferred device of the invention, the actuators are situated outside of the air-flow path, preferably within (or beyond) the support body (in the duct walls, say) and are coupled to the actuation point (on the blade) with a simple strut or tie or other linkage means that extends from the tip of the actuator through a suitable gap in the support body to the actuation point, which is preferably at or near the end of the blade. The strut shape is chosen to offer minimum resistance to the movement of air flowing past; for instance the strut may be thin in the blade-width direction, and free of sharp, turbulence-inducing edges.

As an alternative to the use of plural actuators spaced along the fan blade as

described above, the actuator means may comprise at least one actuator capable of driving the fan blade with both a linear and a rotational displacement with a predetermined phase relationship between the linear and the rotational displacement.

By the use of such an actuator to drive linear and rotational displacement with a predetermined phase relationship between the linear and rotational displacement, it is possible to drive the desired motion of the fan blade in which the ends of the fan blade move back and forth in quadrature. The desired movement of the ends of the fan blade results from the use of a predetermined phase relationship between the linear and rotational displacement driven by the actuator. In general terms it may be thought of as the actuator moving the fan blade back and forth with a linear displacement combined with a rotational displacement so that the fan blade is rotated to tilt in opposite directions during linear displacement in each of the respective directions back and forth. The actual phase relationship depends upon the desired motion of the ends of the fan blade and the positioning of the actuator along the fan blade. For example, if the actuator is coupled to the fan blade at an actuation point at the centre of the fan blade along its length in the fluid flow direction, the desired quadrature motion of the ends of the fan blade may be achieved by use of a 90 (or quarter cycle) phase a difference between the linear and rotational displacement, so that the rotational displacement is zero when the linear displacement is at a maximum magnitude from its central position and the rotational displacement is at a maximum magnitude when the centre of the fan blade is at the middle of its linear motion.

As for the"quadrature"motion of the ends of the fan blade, the predetermined phase relationship between the linear and rotational displacement does not imply anything about the shape of the waveform of the linear and rotational displacements.

Whilst a sinusoidal linear displacement and a sinusoidal rotational displacement may be preferred, either or both waveforms may have any other shape. For example, the linear displacement and/or the rotational displacement might be performed intermittently with no displacement in between. For example, the linear displacement might be smoothly driven back and forth, but with the rotational

displacement occurring in a short period of time when the linear displacement has a maximum magnitude. This would have the effect of achieving a maximum tilt of the fan blade for the entire linear displacement in each direction, although may be less desirable when the blade is disposed in a duct.

The advantage of using an actuator capable of both linear and rotational displacement is that the number of actuators used to drive the fan blade may be reduced. For example a single actuator may be used, in contrast to the use of two linear actuators spaced along the fan blade. Furthermore, the use of an actuator providing both linear and rotational displacement of the fan blade allows a high degree of control over the motion of the fan blade.

Preferably, the actuator comprises an electro-active structure extending along a curved minor axis to form a linear displacement portion and arranged, on actuation, for the structure to twist around the minor axis and concomitantly for relative linear displacement of the ends of the structure to occur.

In this case, the linear displacement portion of the actuator has the same structure as the electro-active device described above. Therefore, the same comments on the structure and advantages apply. Preferably, the actuator further comprises an electro-active structure extending along a straight minor axis to form a rotational displacement portion and arranged, on actuation, for the structure to twist around the minor axis for relative rotational displacement of the ends of the structure.

This is particularly advantageous because it allows the electro-active structure of the rotational displacement portion to be a continuation of the electro-active structure of the linear displacement portion. In this case, the actuator is particularly easy to manufacture, because the same electro-active structure may be formed to extend in one part along a curved minor axis to form the linear displacement portion and along another part to extend along a straight minor axis to form the rotational displacement portion.

In fact, the linear actuator described above is of general application as an actuator to provide linear and rotational displacement in systems other than a fluidpropelling device. Therefore, in accordance with a second aspect of the present

invention there is provided an electro-active device comprising: an electro-active structure extending along a curved minor axis to form a linear displacement portion and arranged, on actuation, for the structure to twist around the minor axis and concomitantly for relative linear displacement of the ends of the structure to occur; and a rotational displacement portion coupled to one end of the linear displacement portion and arranged, on actuation, to generate relative rotation between the ends of the rotational displacement portion.

The linear displacement portion has the same structure as the electro-active device described above, so the same comments about the structure and advantages thereof made above apply also to the linear displacement portion of the electro-active device in accordance with the second aspect of the present invention.

In addition, a rotational displacement portion is coupled to one end of the linear displacement portion and on actuation drives relative rotation of the ends of the rotational displacement portion. Thus an electro-active device in accordance with the second aspect of the present invention is advantageous because the same device may be used to generate linear displacement and rotational displacement. It therefore allows a degree of control in two dimensions.

In the same way that the magnitude of the linear displacement may be controlled by selecting the length of the curved minor axis along which the linear displacement portion extends, the degree of rotation generated between the ends of the rotational displacement portion may be controlled by selecting the length of the rotational displacement portion. This is because each section of the rotational displacement portion contributes to the overall relative rotation between the ends.

Preferably, the rotational displacement portion comprises an electro-active structure extending along a straight minor axis and arranged, on actuation, for the structure to twist around the minor axis for relative rotational displacement of the ends of the structure.

This is particularly advantageous because it allows the electro-active structure of the rotational displacement portion to be a continuation of the electro-active

structure of the linear displacement portion. This greatly simplifies manufacture, because the same electro-active structure may be formed to extend along a curved minor axis to form the linear displacement portion and also along a straight minor axis to form the rotational displacement portion.

Furthermore, as the rotational displacement portion may have the same electro-active structure as the linear displacement portion, the various comments made above about the structure of the linear displacement portion and its advantages apply also to the rotational displacement portion, except of course that it extends along a straight rather than a curved minor axis.

Considering again a fluid propelling device according to the first aspect of the present invention, during operation the fan blade is caused to wave back and forth in a reciprocatory fashion. The rate, and the frequency, at which this waving motion occurs can be whatever is suitable, but will generally depend on the size of the device and on its intended purpose. Small devices e. g. computer-chip cooling fans, preferably move back and forth both quickly and frequently (about 20 times a second; higher frequencies-possibly lOOHz-produce a higher output but make the fan's operation rather noisier), while large devices, e. g. drivers of boats, move less quickly and less frequently (about once every second or so).

The volume rate of ambient fluid swept away by the fan of the invention also depends upon purpose and size. In a general sense the volume (V: in cubic metres) depends upon the vane length (L: in metres), the vane width (W: in metres), the peak-to-peak displacement (d: in metres) and the frequency of the complete cycle (f : in Hertz), and can be approximated from the formula V = W L d f m3/sec For a chip-cooler fan, where W=25mm, L=30mm, d=8mm and f=20Hz, the volume displaced is 1. 2xl0-4m3 per sec (or about 7. 2 litres a minute). At a frequency f of 100Hz the volume rate would be more like 35 litres a minute. It will be seen that this compares favourably with the small conventional fans presently used for cooling chips (these are pretty noisy, and have an air volume throughput of around 50 litres a minute).

The peak-to-peak displacement naturally depends on the size of the device and the space in the environment in which it has to operate. Computer system boxes tend to be somewhat crowded, especially in the region of the main board carrying the processor chips, and in such cases the displacement needs to be fairly small-say, in the range 5-lOmm (0. 2-0. 4in). The fan will work rather better when the displacement d is smaller, and preferably much smaller, than the blade length L so the blade quivers rather than waving vigorously.

The invention's fluid-propelling device has a fan blade mounted within a duct.

There may, of course, be more than one fan blade, conveniently mounted as a sequence along the duct, and the several blades can be driven in or out of step.

Sequential multiple blades do not increase the amount of air flow but may serve to damp any unwanted resonances, and can ensure entirely uni-directional flow.

As already discussed, the air flow depends on the area swept by the blade.

Where space permits a particularly wide device, several fan blades may be mounted side-by-side (within a single wide duct, say). The advantage over a single blade device lies in the potential to use a number of less powerful actuators rather than a single powerful one, and to use small lightweight blades rather than a larger more rigid one.

A further arrangement involves stacking a number of blades one on top of the other. This may be particularly appropriate for a ducted device where each blade is within its own, separate, duct. The advantage here is that the movement of all the blades may be coupled together so that they are driven by a single pair of actuators.

The arrangement is not unlike that of a Venetian blind, in which all slats are caused to tilt together.

As noted hereinbefore, a particularly interesting form of duct is one that is annular and extends radially of the annulus, with a fan blade to match. The configuration is what might be achieved if a short length of plain rectangular-section ducted device of the invention were able to be fanned out sideways-including its blade-into a circular object in much the same way that a pack of cards is fanned out into a circular arrangement. What is obtained is an annular duct rather like a hollow

POLO mint, or the sort of flat can that is used to hold a pancake air filter for a motorcar engine's carburettor, and inside that annular duct is a correspondingly-annular fan blade. The annular duct is without any actual curved-side walls (which is where the original duct's input has been"smeared"out into a circumferential aperture), but it has two apertures centrally located one on each of its two flat sides (which is where the original duct's output has been smeared around). If one of these central apertures is then closed off, the device can be operated to take in ambient fluid through the circumferential aperture and pump it out through the remaining central aperture.

In an annular device of this type there is an annular fan blade, and in operation the inner and outer edges of that annulus move sequentially up and down just as an ordinary rectangular blade does. Of course, for this to work the blade will need to be able temporarily to change its apparent dimensions and/or area. Such a change can be accommodated either by making the blade of an elastic material or, and perhaps preferably, of forming the blade as a suitable multiplicity of overlapping trapezium shapes (much like the cards in the fanned-out pack mentioned above) the edge areas of each of which can slide back and forth over those of its neighbours.

As described above a device of the"annular"type is in fact circular in overall shape. However, the principle discussed may equally be applied to devices which have other outline shapes, typically rectangular (imagine the pack of cards partially smeared out sideways in a straight line, then further smeared out in a second line at right angles to the first, and then yet further smeared orthogonally once, and then again, to end up back at the start position, to make a"square"annulus rather like a cathedral's cloisters).

An annular device is discussed further hereinafter with reference to the accompanying drawings.

Several embodiments of the invention are now described, though by way of illustration only, with reference to the accompanying drawings in which: Figs. 1A-D are schematic side views illustrating the Quadrature operation of a fluid-propelling fan device;

Figs. 2A-C are perspective and side views of a fan device; Fig. 3 is a plan view of a first electro-active device; Fig. 4 is a side view of a second electro-active device ; Fig. 5 is a perspective view of a portion of either the first electro-active device of Fig. 3 or the second electro-active device of Fig. 4; Figs. 6 and 7 are side views of two fan devices using electro-active devices as actuators; Figs. 8A, B are side and end views of a further fan device using externally-mounted electro-active devices as actuators; Fig. 9 is a side view of a fan device using a non-linear actuator; Fig. 10 is a perspective view of the end of a fan device using stacked multiple ducts; Figs. 11A, B are side views of another stacked multiple-duct fan device; Figs. 12A, B are perspective and side views of an annular fan device; Fig. 13 is a perspective view of an electro-active device capable of rotational and linear displacement; Fig. 14 is a perspective view of a further electro-active device capable of linear and rotational displacement; Fig. 15 is a perspective view of a fluid propelling device using the electroactive device of Fig. 13; and Fig. 16 is a perspective view of a fluid propelling device using the electroactive device of Fig. 14.

The following embodiments have much structure in common for which the same reference numerals will be used and a description of which will not be repeated, for brevity.

Fig. 1 shows a schematic side view illustrating an example of the Quadrature operation of an air-moving device of the invention.

Fig. 1A shows a parallel-sided duct 1, with lower and upper (as viewed) walls 7. The duct 1 acts as a support body along which a fan blade 2 is mounted. The fan blade 2 moves relative to the duct 1 with the ends of the fan blade moving in

quadrature with different phases as follows. As shown, in the first phase of operation the fan blade 2 lies horizontally at the bottom of the duct 1. The second phase is shown in Fig. 1B, in which the left hand (as viewed) end 3 of the fan blade 2 is raised to the top of the duct 1, while the right hand end 4 remains on the bottom of the duct.

As the fan blade 2 moves to this position, air is drawn into the duct 1 in the direction of the arrow 5 and is expelled from the duct in the same direction, the direction of the arrow 6.

Fig. 1 C shows the third phase of operation, in which the right hand end 4 moves to the top of the duct 1 while the left hand end 3 remains stationary. Air is again moved in the direction of the arrows 5 and 6. Fig. ID shows the device in the fourth phase of operation, in which the left hand end 3 moves to the bottom of the duct 1 while the right hand end 4 remains stationary. As before, air movement is in the direction of the arrows 5,6.

The position shown in Fig. 1A follows that in Fig. ID and is reached by movement of the right-hand end 4 of the fan blade 2 from the top (Fig. ID) to the bottom (Fig. 1A) of the duct, entraining air in the direction of the arrows 5,6 as in the other phases.

In operation, this four-phase cycle is repeated.

To summarise, in Quadrature operation the ends of the fan blade 2 are respectively (left to right) down-down, up-down, up-up, down-up, and this cycle is repeated indefinitely. In each phase, air is moved left to right. It can readily be envisaged that air could instead be made to move right to left simply by reversing the up-down and down-up phases. Other orientations of the device are clearly also possible.

The fan blade 2 is described above as though it is a rigid element. This is the case for many embodiments where a stiff fan blade 2 is used. For other embodiments the fan blade 2 will have sufficient flexibility to cause the displacement of the fan blade 2 to propagate along the fan blade 2 as a wave, propelling fluid along the fan blade 2 in a similar manner to the fluid flow achieved by swimming fishes.

The motion of the fan blade 2 has been described above as a four-part cycle,

but it should be noted that this is merely to illustrate an example of the quadrature motion of the ends of the fan blade 2. In fact, the shape of the waveform of the motion of each of the ends may vary considerably. For example, instead of the intermittent motion described above in which each end 3,4 of the fan blade 2 moves whilst the other end 3,4 is stationary and acts as a pivot, the motion of each end may be continuous, for example having a displacement which varies over time with a sinusoidal or saw-tooth waveform. Furthermore, whilst the movement of the left hand end 3 is a quarter-cycle ahead of the movement of the right hand end 4, this phase relationship between the movement of the ends 3,4 of the fan blade 2 is not necessary. Other predetermined phase differences between the motion of the end 3,4 of the fan blade 2 may be employed, provided that the ends 3,4 do not move in synchronism. Similarly, the phase relationship between the motion of the ends 3,4 of the fan blade 2 may vary during a single cycle or over several cycles.

In the following embodiments the actuator means used to move the fan blade 2 in the manner illustrated in Fig. 1 will be described.

Fig. 2 shows perspective and side views of a fan device of the invention.

Fig. 2A shows a perspective view of a rectangular fan blade 2 mounted along a support body 13 by actuators 21,22 positioned underneath the blade towards each end 3 and 4, half-way across the width of the blades. The fan blade 2 is a sheet of lightweight material with sufficient stiffness that one actuator 21,22 at each end 3,4 serves to raise or lower the end of the blade across its entire width. The fan blade 2 is thinner than it is broad or long by at least an order of magnitude.

Fig. 2B shows a side view. There can be seen the fan blade 2 with the actuators 21,22 spaced along the fan blade 2 and coupled at one end to the underside of the blade near each end 3,4 by flexible joints 12. The actuators 21,22 are coupled at the other end to the support body 13. In operation the actuators 21,22 extend and contract vertically, serving to raise the blade ends 3,4.

This is shown in Fig. 2C, which depicts a side view similar to that in Fig. 2B but with the actuator 21 at the left hand end 3 of the fan blade 2 shown in the extended state, such that the left hand end 3 of the blade is raised, and the blade rests

diagonally. It can readily be seen that activating the other actuator (22) instead of the actuator 21 gives the opposite diagonal state, and that activating both actuators 21,22 causes the blade 2 to rest in a horizontal, but raised, position. In practice some lateral tilting movement of the actuators 11, or 21,22, is required to accommodate the changes in geometry. This can be allowed for by providing compliant mountings between the actuators 21,22 and the support body 13.

In use, the actuators 21,22 are operated in quadrature with a phase difference between the two actuators 21,22 to move the fan blade 2 as described above. If the actuators 21,22 are at respective ends 3,4 of the fan blade 2, the actuators 21,22 directly control the motion of a respective end 3,4. In general the actuators 21,22 may be at any position spaced along the fan blade 2, the motion of the actuators 21, 22 controlled to drive appropriate motion of the ends 3,4 of the fan blade 2 taking into account the dynamics of the fan blade 2.

The actuators 21,22 are illustrated in Fig. 2 schematically. In fact, the actuators 21,22 may have a wide range of different forms, as discussed above. In the following description, the actuators 21,22 are exemplified as electro-active devices of a particular structure, these being the preferred form of actuators. To assist understanding, the electro-active devices will first be described. In the following description, the electro-active devices are described with reference to minor and major axes which are imaginary, but are nonetheless useful for visualising and defining the devices.

A first electro-active device 1 is illustrated in Fig. 3. The device 1 comprises a structure consisting of a continuous electro-active member 2 curving in a helix around a minor axis 3 so that the structure extends along the minor axis 3. The minor axis 3 is curved, extending in a curve which is an arc of a circle around a geometrical major axis 4 perpendicular to the plane of the minor axis 3, i. e out of the plane of the paper in Fig. 3. As the minor curve 3 is planar, the thickness of the device parallel to the major axis 4 is merely the thickness of the helical structure of the electro-active member 2.

A second electro-active device 11 is illustrated in Fig. 4. The device 2

comprises a structure consisting of a continuous electro-active member 12 to curving in a helix around a minor axis 13 so that the structure extends along the minor axis 13 The minor axis 13 is curved, extending in a curve which is a helix around a geometrical major axis 14. The electro-active device 11 is illustrated in Fig. 4 with a minor axis which extends along of a helix of three turns merely for illustration, any number of turns being possible.

Fig. 5 illustrates a portion 20 of either the continuous member 2 of the first device 1 of Fig. 3 or the continuous member 12 of the second device 11 of Fig. 4.

The construction of the portion 20 being the same for both the first device 1 and the second device 2 the electro-active portion 20 is a finite portion of the continuous member 2 or 12 and hence the electro-active member 2 or 12 may be considered as a plurality of adjacent portions 20 as illustrated in Fig. 5 disposed successively along the minor axis 3 or 13. Hence, the portion 20 extends along part of a helical curve around the minor axis 3 or 13 as shown in Fig. 5.

Fig. 5 illustrates the construction of the electro-active portion 20. This construction is preferably uniform along the entire length of the minor axis 3 or 13 in order to provide uniform properties on actuation. Alternatively, the device 1 or 11 may be designed with some variation along the length of the minor axis 3 or 13, either in the construction of the continuous member 2 or 20 or in the shape of the curve of the continuous member 2 or 20 around the minor axis 3 or 13.

The electro-active portion 20 has a bimorph bender construction comprising two layers 21,22 of electro-active material extending along the length of the portion 20. The layers 21,22 of electro-active material both face the minor axis 3 or 13. The electro-active layers 21 or 22 preferably extend, across the width of the portion 20, parallel to the minor axis 3 or 13, although there may be some distortion of the electro-active portion 20 of the continuous member 2 or 12 due to the nature of the curve around the minor axis 3 or 13. Alternatively, the layers 21 or 22 may extend, across the width of the portion 20, at an angle to the minor axis 3 or 13 so that one edge along the electro-active portion 20 is closer to the minor axis 3 or 13 than the opposite edge.

The material of the electro-active layers 21 or 22 is preferably piezoelectric material. The piezoelectric material may be any suitable material, for example a piezoelectric ceramic such as lead zirconate titanate (PZT) or a piezoelectric polymer such as polyvinylidenefluoride (PVDF). However, the material of the electro-active layers 21, 22 may be any other type of electro-active material, for example piezoresistive material, in which the electrical resistance changes as the material is deformed or strained, or electrostrictive material, which constricts on application of an electric field.

The electro-active portion 20 further comprises electrodes 23 to 25 extending parallel to the layers 21, 22 of piezoelectric material. Outer electrodes 23, 24 are provided outside the electro-active layers 21, 22 on opposite sides of the electricactive portion 20. A centre electrode 25 is provided between the electro-active layers 21 and 22. The electrodes 23 to 25 are used to apply poling voltages and to operate electro-active portion 20 in a bending mode. On electrical actuation, actuation voltages are applied to the electrodes 23 to 25 and conversely on mechanical actuation voltages are developed on the electrodes 23 to 25. On actuation, the electroactive layers 21 and 22 undergo a differential change in length concomitant with bending of the portion 20 due to the constraint of the layers being coupled together at their interface formed by the centre electrode 25. For maximum displacement, on actuation one of the electro-active layers 21 or 22 expands and the other one of the electro-active layers 21 and 22 contracts The relative direction and magnitude of the actuation and poling voltages may be selected in the same manner as for known linear electro-active devices having a bender construction. For example, poling voltages of sufficient magnitude to pole the electro-active layers 21 and 22 may be applied in opposite directions across the electro-active layers 21 and 22 by grounding the centre electrode 25 and applying poling voltages of the same polarity to both the outer electrodes 23, 24. In this case, the electro-active portion 20 is electrically actuated by applying actuation voltages in the same direction across the electroactive layers 21 and 22 by applying voltages of opposite polarity to the two outer electrodes 23 and 24.

On actuation the electro-active portion 20 bends around the minor axis 3 or 13, either towards or away from the minor axis 3,13 depending on the polarity of the actuation voltages. On electrical actuation the actuation voltages are applied from a circuit 26 through external terminals 27 electrically connected to the electrodes 23 to 25 in the manner known for known straight piezoelectric devices having a bender construction. On mechanical actuation, the actuation voltages developed at the electrodes 23 to 25 are fed to the circuit 26.

Electrical connection to the electrodes 23 to 25 may be made in the same way as is known for known straight devices having a bender construction, in principle at any point along the length of the device of which the portion 20 forms part but preferably at the end. The preferred technique is to provide the electrodes with fingers (not shown) extending at the end of the device at different lateral positions across the width of the device as known for straight devices having a bender construction.

It will be appreciated that other bender constructions could equally be applied to the portion 20, for example a unimorph bender construction comprising a layer of electro-active material and an inactive layer or a multimorph bender construction comprising a plurality of layers of electro-active material.

Whilst the bender construction illustrated in Fig. 5 is preferred for simplicity and ease of manufacture, it will be appreciated that the continuous numbers 2 or 12 could in fact have any construction which bends around the minor axis 3 or 13 on actuation. For example, the continuous members could be electro-active elements of the type described in the application being filed simultaneously with this application entitled"Electro-Active Elements and Devices"in which the elements have two pairs of electrodes extending along the length of the member for bending across the width on activation.

On actuation, the electro-active portions 20 of the continuous member 2 or 12 bend around the minor axis 3 or 13. As a result of the continuous electro-active member 2 or 12 curving around the minor axis 3 or 13, in particular in a helix, such bending is concomitant with twisting of the continuous member 2 or 12 around the

minor axis 3 or 13. This may be visualised as the turns of the continuous member 2 or 12 as the bending tightening or loosening causing a twist of the structure of the member 2 or 12 along the minor axis 3 or 13. The twist of the continuous member 2 or 12 occurs along the entire length of the minor axis 3 or 13 causing a relative rotation of the ends of the structure labelled 5 and 6 in the first device 1 of Fig. 3 and 15 and 16 in the second device 11 of Fig. 4.

It will be appreciated that the continuous member 2 or 12 could curve around the minor axis 3 or 13 in curves other than a helix to produce such twisting, for example by having the shape as though formed by twisting a flat member round the minor axis. It will also be appreciated that other structures other than a continuous member could be applied to produce twisting around the minor axis. For example the electro-active structure could consist of a plurality of electro-active portion disposed successively along the minor axis and coupled together so that the bending of each individual portion twists the adjacent portion around the minor axis causing twisting of the structure as a whole. Alternatively the electro-active structure could be a device of the type described in the application being filed simultaneously with this application entitled"Piezoelectric Devices"which comprises a plurality of electro-active torsional actuators which may comprise electro-active elements activated in shear mode.

Considering the first device 1 of Fig. 3, the twisting of the continuous member 2 around the minor axis 3 is concomitant with relative displacement of the ends of the device 5 and 6 perpendicular to the curve of the minor axis 3, that is parallel to the major axis 4. The relative displacement of the ends 5 and 6 derives from the twisting of the continuous member 2 around the minor axis 3 in combination with the curve of the minor axis 3. It is an inevitable result that twisting of a curved object causes relative displacement of the ends of that object perpendicular to the local curve of the object.

In a similar manner, on actuation of the second device of Fig. 4, the twisting of the continuous member 12 around the minor axis 13 is concomitant with displacement of the ends of the device 15 and 16 parallel to the major axis 14.

Again, this relative displacement derives from the rotation of the continuous member 12 around the minor axis 13 in combination with the curve of the minor axis 13. In this case, the relative displacement caused by any given'small section of the structure along the minor axis 13 causes relative displacement of the ends of that section perpendicular to the local curve of the minor axis 13. The overall displacement of the ends 15,16 of the device 11 is the sum of the displacements of all the sections which results in an overall relative displacement parallel to the major axis 14.

The exact construction and dimensions of the member 2 or 12 and the form of the electro-active structure may be freely varied to produce the desired response. A suitable member 2 or 12 has a 0.5 mm thickness tape wound as a 4 mm diameter minor helix around the minor axis 3 or 13. When this forms the first device 1 in which the minor curve extends around about three quarters of a circle of 30 mm diameter the observed displacement is about 6mm. Similarly if this structure was used to form the second device 11 in which the minor curve extends along a 20 turn helix of diameter 30mm, this would produce displacement of around 120mm.

In general, the minor axis, along which the structure of devices in accordance with the present invention extend, may follow any curve and the resultant displacement of the ends of the structure will be the sum of the displacement caused by each section of the structure along the curve. However, curve which are regular such as the curve of the minor axis of the first and second devices 1 and 11 are preferred so that all sections of the device caused relative displacement in a common direction and also because design and manufacture are thereby simplified.

The first and second devices 1 and 11 may be electrically actuated to create mechanical displacement between the ends 5 and 6 or 15 and 16, or may be mechanically actuated in which case relative displacement of the ends 5 and 6 or 15 and 16 causes an electrical voltage to be developed across the electrodes 23 to 25. In the case of electrical actuation, the ends 5 and 6 or 15 and 16 of the electrical device 1 or 11 are coupled to further elements to be relatively displaced similarly in the case of mechanical actuation the ends 5 and 6 or 15 and 16 are coupled to elements which drive deformation of the device 1 or 11.

Manufacture of the electro-active devices 1 and 11 will now be described.

The preferred method of manufacture is to initially form the electro-active structure extending along a straight minor axis and subsequently to bend the straight electro-active structure so that the minor axis along which it extends becomes curved.

To form the continuous member 2 or 12 as an electro-active structure along a straight minor axis there are two preferred techniques.

The first preferred technique is to initially form the continuous member 2 or 12 as a straight member and subsequently to deform it to curve around the straight minor axis. The bender construction of the continuous member 2 or 12 is in itself known and the continuous member 2 or 12 may be formed by applying any of the known techniques for manufacturing a device having a bender construction. For example, the continuous member 12 may be initially manufactured by co-extrusion of the layers 21 and 22 of plasticised material or by co-calendering of the layers 21 and 22. Alternatively, the continuous member 2 or 12 may be made through lamination of thin layers 21 and 22. These thinner layers may be made by any suitable route, such as high shear mixing of a ceramic powder, polymer and solvent mixer followed by co-extrusion and calendering. Alternatively, techniques such as tape casting or the process known as the Solutech process) known in the field of ceramics may be used.

The electrodes may be formed as an integral part of the manufacture of the continuous member 2 or 12, for example by being in co-extruded or co-calendered.

Further electrodes, which may be activation layers 23 to 25 or may be terminal electrodes to allow access to the electrodes 23 to 25, may be applied by printing, by electro-less plating, through fired-on silver past or by any other appropriate technique.

The second preferred technique is to initially manufacture the continuous member as a cylinder or other tube with a multi-layered bender construction of electro-active layers 21 and 22 and electrodes 23 to 25 and subsequently to cut the member along the helical line to leave the continuous member 2 or 12 extending in a helix around the axis of the cylinder or tube which then constitutes the minor axis.

Subsequently the straight structure is bent to curve the minor axis along which the structure extends.

-here mnst exiq in thi-initiqllv fnrmpil To deform the member and structure, there must exist in the initially formed member a sufficient degree of flexibility. Suitably deformable electro-active materials are known, typically including constituent polymers which enhance the defonnability. With such materials after shaping, the constituent polymers are burnt

out, typically at up to 600 C and the material is then densifie through further sintering at higher temperature, typically 1000 C to 1200 C. In this case, the electro- active structure is initially formed with enlarged dimensions to allow for linear shrinkage which occurs during sintering, typically of around 12 to 25%.

The curving of the straight member and the bending of the structure may be performed around formers. The formers are subsequently removed either physically or by destruction of the former for example by melting, burning or dissolving.

Embodiments of the present invention which are fluid propelling devices will now be described.

Fig. 6 illustrates a fluid propelling device in accordance with the present invention in which the actuators 21,22 are exemplified as electro-active devices of the type described above. In particular, the electro-active devices 21,22 are shown as extending along a minor axis which curves in a helix, as in the second electroactive device 11 illustrated in Fig. 2, in particular with a helix of 3 turns. However, this structure is merely for illustration and the actuators 21,22 may have any of the types of structure described above. The actuators 21,22 are each coupled at one end to the fan blade 2 and at the other end to the support body 13, preferably by couplings which have flexibility sufficient to accommodate lateral movements caused by the stiffness of the fan blade 2 as the fluid propelling device is operated.

Fig. 3 illustrates the point in operation in which the first actuator 21 is in its extended (fully activated) position and the second actuator 22 is in its contracted (inactivated) position.

Fig. 7 illustrates a further embodiment of a fluid propelling device similar to that illustrated in Fig. 3 but in which the support body 13 has walls defining a duct 1

in which the fan blade 2 is mounted extending along the duct 1, but with the actuators 21,22 themselves outside of the duct 1. The actuators 21,22 are coupled to the fan blade 2 by struts 61 coupled at one end to the blade 2 via flexible couplings 12 and at the other end to actuators 21,22. The struts 61 extend through gaps 62 in the lower duct wall 7. The actuators 21,22 are mounted outside the duct 1, so do not interfere with the air flow.

Fig. 8 shows another embodiment of the invention in which the actuators 21,22 are of the type described above. The actuators 21,22 are illustrated as having structures extending along a minor axis which is curved along an arc of a circle as in the first device of Fig. 3 but this is merely for illustration and any of the structures described above may be used.

Fig. 8A is a plan view showing the actuators 21,22 mounted in the side-walls 8 of the duct 1 forming the support body, so that the actuators lie outside the duct 11.

In this case the fan blade 2 is illustrated as a rectangular frame 71 with a light-weight membrane 72 stretched across it. The frame member 73 at the left-hand end 3 of the blade is shown continuing through a gap 74 in the side-wall 8, where it is attached to the actuator 21.

The benefit of side-mounting the actuators is that the actuator mechanism is virtually entirely outside the path of the air-flow, which is in the direction of the arrows 5,6, minimising any unwanted turbulence and drag.

The blade 2 is of a light-weight construction with the end-member 73 sufficiently rigid to transmit the actuator drive across the whole width of the blade 2.

Fig. 8B is an end-on view of the device of Fig. 8A showing one end 4 of the fan blade 2 lying in the duct 1 with actuators 21,22 mounted at 13 in the side-walls 8 of the duct 1, halfway up the duct. Both actuators 21,22 are shown activated in the down position, such that the fan blade 2 is lying flat at the base of the duct 1.

Fig. 9 shows that with suitable couplings, non-linear actuators can be used.

Fig. 9 is a side view showing rotary actuators 21,22 mounted to the side of the blade 2, which lies in a duct 1. In this case the actuators 21,22 are in the form of a piezoelectric bender curved into an arc of a circle. One end 81 of each actuator is

fixed such that on activation the other end 82 rotates by a few degrees. As shown, the end of the actuator 82 is connected to a radial flange 83 which is itself rotatably mounted to the ends 3,4 of the blade 2.

Fig. 10 shows a perspective end view of a framework 91 of ducts 1 stacked vertically, each duct containing a fan blade 2. Such an arrangement of several blades 2 (in this case five) allows a greater area to be swept than by a single blade (in this case by a factor of five) for the same amount of actuator movement. The several blades can be made to operate in tandem by a suitable arrangement of actuators and connectors, as shown in Fig. 8.

Fig. 11A shows schematically a side view of an arrangement of five vertically-stacked fan blades 2 lying in ducts (which for clarity are not shown).

Actuators 21,22 lie beneath the stack near each end 3,4. Vertical connecting posts 101 run up from the top of the actuators through holes in the blades 2. Above and below each hole is a flange 102,103 on the post, which flange is more extensive than the hole.

Fig. lIB shows the same arrangement when the actuator 22 near the end of the blade 4 is extended. As the post 101 lifts, the lower flanges 103 lift the end 4 of each blade 2. Similarly, when the actuator is activated such that it contracts, the upper flanges 102 would lower the blade ends 4. The actuator 21 operates in a similar manner on the other ends 3 of the blades 2.

The direction of air movement is shown by the arrows 5,6. Such an arrangement allows a large volume of air to be swept by a number of blades operated by a small number of actuators of relatively low travel. It is particularly appropriate where the extension and contraction of the actuators is small, as is the case for most piezoelectric actuators.

Fig. 12 shows an embodiment of the device of the invention in which the fan blade and the duct are annular. Fig. 13A is a perspective view of the arrangement in its start (or first phase) configuration, while Fig. 13B is a vertical cross-section showing the penultimate (or fourth phase) configuration.

The blade 112 is a flat annular disc lying in a circular duct 111, whose upper

and lower walls 117 are also discs, one or both of which has an opening 118 to act as an exhaust port, shown attached to a chimney 119. The annular blade 112 is shown in Fig. 9A lying flat at the base of the duct 111, and in Fig. 9B with its outer rim 113 at the bottom of the duct 111 and its inner rim 114 raised to the top of the duct 111.

The rims 113,114 are driven by electromechanical actuators (not shown) which raise and lower the rims in Quadrature, causing air to be drawn in radially from the outside, in the direction of the arrows 5, towards the centre and out through the exhaust port 118 in the direction of the arrow 6.

It will be appreciated that there may in addition be a second exhaust port 118 in the base of the duct 111, and that it is possible to operate the fan in reverse, drawing air in through the chimney (s) 119 and exhausting it at the outer circumference of the duct 111.

Next there will described an embodiment of the invention in which the motion of the fan blade 2 is driven by an actuator which provides both linear and rotational displacement. However, for clarity, suitable actuators will first be described. The actuators also constitute embodiments of the second aspect of the present invention.

A first electro-active device 50 is illustrated in Fig. 13. The device 50 comprises a structure which has two portion 51 and 52 separated by the (imaginary) dotted line 53. The first portion 51 is identical to the first electro-active device 1 illustrated in Fig. 3. Therefore, a description thereof will not be repeated and the same reference numerals will be used. The first portion 51, on actuation, creates linear displacement between its ends 5,6, and so acts as a linear displacement portion.

The second portion 52 is a continuation of the first portion 51 in which the same continuous electro-active member 2 which constitutes the first portion 51 curves in a helix around the minor axis 3, but with the difference that in the second portion 52 the minor axis 3 is straight. The structure and construction of the continuous member 2 in the second portion 52 is the same as in the first portion 51 and the first electro-active device 1 of Fig. 3 (as described above with reference to

Fig. 5). Therefore, on actuation of the second portion 52, the continuous member 2 bends around the minor axis 3. Concomitantly there is a relative rotation of the ends 54 and 55 of the second portion 52 which therefore constitutes a rotational displacement portion.

The first and second portions 51 and 52 are independently actuatable by providing a split in the electrodes 23 to 25 of the continuous member 2 between the first and second portions 51 and 52. By applying appropriate activation voltages to the continuous member 2 in the first portion 51 to electrically activate it, linear displacement of the ends 5,6 of the first portion 51 is achieved out of the plane of the curve of the minor axis 3 in the first portion 51. Relative to the end 5 of the first portion 51 (e. g. if that end 5 is held fixed), the other end 6 of the first portion moves linearly in the direction of arrows A in Fig. 13. This displacement also displaces the second portion 52 coupled to the end 6 of the first portion 51 relative to the opposite end 5 of the first portion 51. Similarly, by applying appropriate activation voltages to the continuous member 2 of the second portion 52 to electrically activate it, relative rotation between the ends 54 and 55 of the second portion 52 is achieved.

Relative to the end 54 of the second portion 52 (e. g. if the end 5 of the first portion 51 is held fixed), the other end 55 of the second portion 52 rotates in the direction of arrows B in Fig. 13.

In use, the free ends 5,55 of the electro-active device 50 (that is the end 5 of the first portion 51 and the end 55 of the second portion 52 which are not coupled together) are coupled to objects to be relatively displaced. By activation of the first and second portions 51 and 52, the two objects may be relatively displaced linearly and/or rotationally. The electro-active device 50 can therefore drive relative displacement of the two objects in two dimensions.

A second electro-active device 60 is illustrated in Fig. 14. The device 60 comprises a structure which has two portion 61 and 62 separated by the (imaginary) dotted line 63. The first portion 61 is identical to the second electro-active device 11 illustrated in Fig. 4. Therefore, a description thereof will not be repeated and the same reference numerals will be used. The first portion 61, on actuation, creates

linear displacement between its ends 15,16, in the direction of arrows A in Fig. 14 and so acts as a linear displacement portion.

The second portion 62 is a continuation of the first portion 61 in which the same continuous electro-active member 2 which constitutes the first portion 61 curves in a helix around the minor axis 3, but with the difference that in the second portion 62 the minor axis 3 is straight. The structure and construction of the continuous member 2 in the second portion 62 is the same as in the first portion 61 and the second electro-active device 11 of Fig. 4 (as described above with reference to Fig. 5). Therefore, on actuation of the second portion 62, the continuous member 2 bends around the minor axis 3. Concomitantly there is a relative rotation in the direction of arrows B in Fig. 14 of the ends 64 and 65 of the second portion 62 which therefore constitutes a rotational displacement portion.

The first and second portions 61 and 62 are independently actuatable by providing a split in the electrodes 23 to 25 of the continuous member 2 between the first and second portions 61 and 62. By applying appropriate activation voltages to the continuous member 2 in the first portion 61 to electrically activate it, linear displacement of the ends 15,16 of the first portion 61 is achieved out of the plane of the curve of the minor axis 3 in the first portion 61. This displacement also displaces the second portion 62 coupled to the end 16 of the first portion 61 relative to the opposite end 15 of the first portion 61. Similarly, by applying appropriate activation voltages to the continuous member 2 of the second portion 62 to electrically activate it, relative rotation between the ends 64 and 65 of the second portion 62 is achieved.

In use, the free ends 15,65 of the electro-active device 60 (that is the end 15 of the first portion 61 and the end 65 of the second portion 62 which are not coupled together) are coupled to objects to be relatively displaced. By activation of the first and second portions 61 and 62, the two objects may be relatively displaced linearly and/or rotationally. The electro-active device 60 can therefore drive relative displacement of the two objects in two dimensions.

There will now be described fluid propelling devices in accordance with the present invention in which the electro-active devices 50,60 described above are used

to drive displacement of the fan blade 2. Fig. 15 illustrates a fluid propelling device 56 in which the electro-active device 50 of Fig. 13 is employed, whereas Fig. 16 illustrates a fluid propelling device 66 in which the electro-active device 60 of Fig.

14 is used. These two fluid propelling devices 56 and 66 will now be described in common.

In the fluid propelling devices 56 or 66, the fan blade 2 is mounted along a support body 13 by the electro-active device 50 or 60. In particular, the free end 5 or 15 of the first portion 51 or 61 is coupled to the support body 13, and the free end 55 of the second portion 52 or 62 is coupled to the centre of the fan blade 2.

Accordingly, activation of the first portion 51 or 61 displaces the fan blade 2 linearly back and forth with respect to the support body 13, whereas activation of the second portion 52 or 62 rotationally displaces the fan blade 2, thereby changing its orientation with respect to the support body. Thus by actuating the first and second portions 51 and 52 or 61 and 62 with a predetermined phase difference therebetween, the desired motion of the fan blade 2 with its ends moving in quadrature may be achieved.

In use, appropriate activation voltages are applied by a central circuit to the first and second portions 51 and 52 to achieve the desired motion. Any desired motion of the fan blade 2 in which the ends 3 and 4 move back and forth in quadrature may be achieved by appropriate control of the waveforms of the actuation voltages applied to the first and second portions 51, 52 and 61 and 62, with a predetermined phase difference therebetween.

Claims

1. A fluid-propelling device comprising an area-extensive fan blade mounted along a support body for either free or pivotally-restrained movement at either end, which blade is operatively linked to a mechanical actuator means that is capable of causing the two ends of the blade to move back and forth in quadrature to propel the relevant ambient fluid predominantly in the same direction throughout the movement of the fan blade.

2. A device as claimed in Claim 1, wherein the elongate area-extensive fan blade is generally rectangular.

3. A device as claimed in either one of the preceding Claims, wherein the fan blade is thinner than it is broad or long by at least an order of magnitude.

4. A device as claimed in any one of the preceding Claims, wherein the fan blade is a sheet of light weight material, or a rigid frame supporting a diaphragm.

5. A device as claimed in any one of the preceding Claims, wherein the support body along which the fan blade is mounted is a duct, the blade is so mounted extending along the duct.

6. A device as claimed in Claim 5, wherein the duct is longer than the blade, while the duct's cross-section is such that in operation the blade makes a close fit therein.

7. A device as claimed in any one of the preceding Claims, wherein the actuator means is situated outside of the fluid-flow path, and is coupled to the actuation point on the blade with a linkage means that extends from the actuator through a suitable gap in the support body to the actuation point on the blade.

8. A device as claimed in any one of the preceding Claims, wherein there are two or more fan blades mounted along the support body so as to form a sequence.

9. A device as claimed in any one of the preceding Claims, wherein there are two or more fan blades mounted alongside each other on the support body to form a wide blade.

10. A device as claimed in any one of the preceding Claims, wherein there are two or more fan blades mounted above each other on the support body to form a stack of blades.

11. A device as claimed in any one of the preceding Claims, wherein the support body is an annular duct, and the fan blade is annular.

12. A device as claimed in Claim 11, wherein the blade is made either of an elastic material or is formed from a suitable multiplicity of overlapping trapezium shapes.

13. A device as claimed in any of the preceding Claims, wherein the actuator means comprises at least two actuators, spaced along the fan blade, and capable of being actuated in quadrature.

14. A device as claimed in claim 13, wherein the blade is linked to the actuators loosely in a pivotal manner.

15. A device as claimed in claim 13 or 14, wherein the actuators are linked to the ends of the fan blade.

16. A device as claimed in any one of claims 13 to 15, wherein the actuators are each an electro-active device comprising an electro-active structure

extending along a curved minor axis and arranged, on actuation, for the structure to twist around the minor axis and concomitantly for relative displacement of the ends ofthe structure to occur.

17. A device as claimed in any one of claims 1 to 12, wherein the actuator means comprises at least one actuator capable of driving the fan blade with both a linear and a rotational displacement with a predetermined phase relationship between the linear and the rotational displacement.

18. A device as claimed in claim 17, wherein the at least one actuator comprises an electro-active structure extending along a curved minor axis to form a linear displacement portion and arranged, on actuation, for the structure to twist around the minor axis and concomitantly for relative linear displacement of the ends of the structure to occur.

19. A device as claimed in claim 18, wherein the at least one actuator further comprises an electro-active structure extending along a straight minor axis to form a rotational displacement portion and arranged, on actuation, for the structure to twist around the minor axis for relative rotational displacement of the ends of the structure.

20. A device as claimed in claim 19, wherein the electro-active structure of the rotational displacement portion is a continuation of the electro-active structure of the linear displacement portion.

21. A device as claimed in any one of claims 17 to 20, wherein the linear displacement portion and the rotational displacement portion are independently actuatable.

22. A device as claimed in any one of claims 16 or 18 to 21, wherein the

electro-active structure comprises electro-active portions disposed successively along the minor axis and arranged to bend, on actuation, around the minor axis.

23. A device as claimed in claim 22, wherein the electro-active structure comprises a continuous electro-active member curving around the minor axis, said electro-active portions being adjacent finite portions of the continuous member.

24. A device as claimed in claim 23, wherein the continuous electroactive member curves in a helix around the minor axis.

25. A device as claimed in any one of claims 22 to 24, wherein the successive electro-active portions have a bender construction of a plurality of layers including at least one layer of electro-active material.

26. A device as claimed in claim 25, wherein the electro-active portions have a bimorph bender construction of two layers of electro-active material or a multimorph bender construction of more than two layers of electro-active material.

27. A device as claimed in any one of claims 16 or 18 to 26, wherein the electro-active structure includes electrodes for development of an electric potential thereacross on actuation of the electro-active structure.

28. A device as claimed in any one of claims 16 or 18 to 27, wherein the minor axis extends in curve which is a helix.

29. A device as claimed in any one of claims 16 or 18 to 28, wherein the minor axis extends in curve which is planar.

30. A device as claimed in any one of claims 16 or 18 to 29, wherein the electro-active structure includes piezoelectric material.

31. A device as claimed in claim 30, wherein the piezoelectric material is a piezoelectric ceramic or a piezoelectric polymer.

32. A device as claimed in claim 31, wherein the piezoelectric material is lead zirconate titanate (PZT) or polyvinylidenefluoride (PVDF).

33. A fluid propelling device constructed and arranged to operate substantially as hereinbefore described with reference to the accompanying drawings.

34. An electro-active device comprising: an electro-active structure extending along a curved minor axis to form a linear displacement portion and arranged, on actuation, for the structure to twist around the minor axis and concomitantly for relative linear displacement of the ends of the structure to occur; and a rotational displacement portion coupled to one end of the linear displacement portion and arranged, on actuation, to generate relative rotation between the ends of the rotational displacement portion.

35. A device as claimed in claim 34, wherein the rotational displacement portion comprises an electro-active structure extending along a straight minor axis and arranged, on actuation, for the structure to twist around the minor axis for relative rotational displacement of the ends of the structure.

36. A device as claimed in claim 35, wherein the electro-active structure of the rotational displacement portion is a continuation of the electro-active structure of the linear displacement portion.

37. A device as claimed in any one of claims 34 to 36, wherein the electroactive structure comprises electro-active portions disposed successively along the minor axis and arranged to bend, on actuation, around the minor axis.

38. A device as claimed in claim 37, wherein the electro-active structure comprises a continuous electro-active member curving around the minor axis, said electro-active portions being adjacent finite portions of the continuous member.

39. A device as claimed in claim 38, wherein the continuous electro-active member curves in a helix around the minor axis.

40. A device as claimed in any one of claims 34 to 37, wherein the successive electro-active portions have a bender construction of a plurality of layers including at least one layer of electro-active material.

41. A device as claimed in claim 40, wherein the electro-active portions have a bimorph bender construction of two layers of electro-active material or a multimorph bender construction of more than two layers of electro-active material.

42. A device as claimed in any one of claims 34 to 41, wherein the electroactive structure includes electrodes for development of an electric potential thereacross on actuation of the electro-active structure.

43. A device as claimed in any one of claims 34 to 42, wherein the minor axis extends in curve which is a helix.

44. A device as claimed in any one of claims 34 to 43, wherein the minor axis extends in curve which is planar.

45. A device as claimed in any one of claims 34 to 44, wherein the electroactive structure includes piezoelectric material.

46. A device as claimed in claim 45, wherein the piezoelectric material is a piezoelectric ceramic or a piezoelectric polymer.

47. A device as claimed in claim 46, wherein the piezoelectric material is lead zirconate titanate (PZT) or polyvinylidenefluoride (PVDF).

48. A device as claimed in any one of claims 34 to 47, wherein the linear displacement portion and the rotational displacement portion are independently actuatable.

49. An electro active device constructed and arranged substantially as hereinbefore described with reference to the accompanying drawings.