WO2003070469A1

WO2003070469A1 - Actuation device and use thereof in fluid pumping and droplet deposition

Info

Publication number: WO2003070469A1
Application number: PCT/GB2003/000748
Authority: WO
Inventors: Stephen Temple; Robert Harvey; Ronald Zmood; Robert Jonathan Lowe; Paul Raymond Drury
Original assignee: Xaar Technology Limited
Priority date: 2002-02-20
Filing date: 2003-02-20
Publication date: 2003-08-28
Also published as: AU2003208434A1; GB0204009D0

Abstract

In fluid pumping apparatus suitable for use in drop on demand ink jet printing, an electromagnetic actuator operating on fluid in the chamber causes a change in pressure, the actuator comprising an armature displaced through a modulation in the distribution of a flux. Modulation of flux is achieved through constructive and destructive interference, between a first magnetic field and a second switchable magnetic field. A combination of permanent magnets and electromagnets are used to provide flux in the arrangement. The invention is preferably of planar construction, manufactured using MEMS techniques.

Description

ACTUATION DEVICE AND USE THEREOF IN FLUID PUMPING AND

DROPLET DEPOSTION

The present invention relates to fluid pumping devices and in a particular example to actuation devices of a form suitable for use in drop on demand ink jet printing and other droplet deposition apparatus. Fluid pumping apparatus, and in particular miniature fluid pumping apparatus has a number of commercially important applications including the dispensing of drugs, and in a particular example apparatus for producing an aerosol.

It is an object of the present invention to seek to provide an improved fluid pumping apparatus.

A fluid pumping application of particular relevance to the present application is printing. Digital printing and particularly inkjet printing is quickly becoming an important technique in a number of the global printing markets. It is envisaged that pagewide printers, capable of printing over 100 sheets a minute, will soon be commercially available.

Inkjet printers today typically use one of two actuation methods. In the first, a heater is used to boil the ink thereby creating a bubble of sufficient size to eject a corresponding droplet of ink. The inks for bubble jet printers are typically aqueous and thus a large amount of energy is required to vapourise the ink and create a sufficient bubble. This tends to increase the cost of the drive circuits and also reduces the life time of the printhead. The second actuation method uses a piezoelectric component that deforms upon actuation of an electric field. This deformation causes ejection either by a pressure increase in a chamber or through creation of an acoustic wave in the channel. The choice of ink is significantly wider for piezoelectric printheads as solvent, aqueous, hot melt and oil based inks are acceptable. It is an object of the present invention to seek to provide an improved droplet deposition actuator.

According to one aspect of the present invention there is provided fluid pumping apparatus comprising a fluid chamber with an inlet and an outlet arranged so that fluctuating pressure within the chamber causes a flow of fluid; and an electromagnetic actuator operating on fluid in the chamber to cause a change in pressure; wherein the actuator comprises an armature displaced through a modulation in the distribution of a flux.

In a second aspect the invention consists of droplet deposition apparatus comprising a liquid chamber; a droplet ejection nozzle communicating with the liquid chamber for the ejection of a droplet; and an electromagnetic actuator operating on liquid in the chamber to effect droplet ejection; wherein the actuator comprises an armature displaced through modulation in distribution of a flux.

Preferably, the total flux is of substantially constant magnitude. The action of modulating the flux therefore has the effect of redistributing the path of the flux, but a force can be applied to, and result in displacement of the armature without substantially increasing the magnitude of flux. In this way the flux in at least parts of the actuator arrangement can advantageously be kept at or close to saturation at all times. In certain embodiments it is desirable that the flux in the armature be maintained at or close to saturation level. Conversely the armature can be returned to its original position without decreasing the total amount of flux.

Preferably, the path of the flux includes two or more flux carrying air gaps and modulation of the flux comprises modulation of the flux density in said two or more air gaps. This modulation may take the form of an increase in flux density at a first air gap location, and a decrease in flux density at a second air gap location, preferably achieved through constructive and destructive interference respectively between a first magnetic field and a second, switchable magnetic field. Suitably, the electromagnetic actuator comprises a primary magnet, which is preferably a permanent magnet, establishing a flux and secondary magnet, which is preferably an electromagnet, serving to modulate said flux.

In an alternative aspect the invention comprises fluid pumping apparatus comprising a fluid chamber; a chamber outlet; and an electromagnetic actuator operating on fluid in the chamber to cause fluid flow in the chamber outlet; wherein said electromagnetic actuator comprises: a primary magnet establishing a primary flux; a primary flux path including at least two parallel flux paths, between which said primary flux is divided; a secondary magnet operable to increase the flux in at least one of said parallel flux paths, and decrease the flux in at least one other of said parallel flux paths; and an armature arranged to experience an actuation force in response to variations in distribution of flux between said parallel flux paths.

Preferably said secondary magnet is arranged to produce a flux in a first direction in a first of said parallel flux paths, and flux in a second direction, substantially opposite said first direction, in a second parallel flux path.

Advantageously one of said parallel flux paths includes an air gap, and said armature forms part of said primary flux path.

It is preferred that the actuator is formed via a MEMS technique in that it is a laminate manufactured through the repeated formation and selective removal of layers. It is further preferred that the layers are formed in the plane orthogonal to the direction of actuation of the actuator.

It is an object of a further aspect of the invention, to provide an actuator having wider application. Accordingly, the present invention consists in another aspect in an actuation device manufactured by MEMS technology and having an electromagnetic actuator; wherein the actuator comprises an armature displaced through modulation in distribution of a flux.

The invention will now be described, by way of example only, with respect to the following drawings in which:

Figure 1 depicts in perspective a view from underneath a channelled component according to one embodiment of the present invention;

Figure 2 depicts in sectional view a printhead according to a second embodiment of the present invention; Figure 3 shows in perspective under view printhead according to a further embodiment of the present invention;

Figures 4 to 11 depict in respective sectional views steps in the manufacture of the printhead shown in Figure 3;

Figure 12 depicts in sectional view the actuation of the printhead shown in Figure 3;

Figure 13 is a flux modulation actuator in a printhead according to an embodiment ofthe present invention;

Figure 14 is an expanded view of the flux modulation actuator of Figure 13 showing field lines;

Figures 15 to17 are views similar to Figure 14 respective orientations adopted by the actuator in use;

Figure 18 depicts key dimensions in the arrangement of the bias flux actuator;

Figure 19 is a graph showing F_x vs x for the bias flux actuator with i=0;

Figure 20 is a graph of F_x vs i for the range -kg < x < +kg;

Figure 21 depicts a flux modulation actuator coupled to an ejection chamber via a push-rod spacer plate;

Figure 22 illustrates a generic planar construction of a fluid pumping apparatus according to one embodiment of the invention; Figure 23 shows a view of a channelled construction for use in a fluid pumping apparatus according to one embodiment of the invention;

Figure 24 shows a variable reluctance type magnetic actuator in a printhead according to an embodiment of the present invention;

Figure 25 depicts in a similar view an alternative type variable reluctance type magnetic actuator;

Figure 26 shows a Lorenz force actuator in a printhead according to an embodiment of the present invention;

Figure 27 depicts an alternative actuator arrangement;

Figures 28 to 31 illustrate further alternative actuator arrangements; and

Figures 32 to 40 depict steps in the manufacture of the actuator shown in Figure 21.

One of the benefits of certain aspects of the present invention is that the printhead itself can be formed from a number of individually manufactured components. The first component comprises the actuator element whilst a second component comprises the channel structure. Other features may be manufactured as separate components or may be formed as part of the components above.

Figure 1 depicts the channelled component in one embodiment of the invention. A sheet of silicon, ceramic or metallic material 1 is etched, machined or electroformed as appropriate to form a plurality channels, separated by walls 2, extending the length ofthe component. The component comprises a resiliently deformable wall 4 that extends part oH e way along the channel. The wall forms the base of the ejection chamber and is deformed by an actuator (not shown), remote from the channel, acting on its reverse side. At either end of the resiliently deformable wall through ports 6 are provided that act to supply ejection fluid to the completed actuator. A cover component 8 of a Nickel / Iron alloy, such as Nilo42, is attached to the top surface of the channelled component and comprises through ports for alignment with nozzle orifices 12 located in a nozzle plate 10.

The width W_c, Height H_c, and Length L_c of the ejection chamber have dimensions that satisfy the conditions W_c, H_c « L_c. The acoustic length L_c being determined from the operating frequency and the speed of sound in the chamber and is typically of the order 2mm. The nozzle is positioned mid-way along the chamber and each end ofthe chamber opens into the manifold formed by the through ports 6.

In operation, the manifolds can either both supply ink to the chamber or the supply arrangement can be such that ink can continually be circulated through the chamber, one of the manifolds returning the excess and unprinted fluid to a reservoir.

The open ends ofthe chamber provide an acoustic boundary that negatively reflect the acoustic waves in the channel. These reflected waves converge at the nozzle and cause droplet ejection. Thus, the manifolds must have a large cross-sectional area with respect to the size ofthe channel in order to achieve an appropriate boundary.

The resiliently deformable wall 4 comprises a directly or indirectly attached actuator element. The actuator element is positioned on the opposite side of the resiliently deformable wall to that facing the nozzle and is thus located remote from the ejection chamber. The actuator moves in a straight line to cause the deformable wall to deflect orthogonally with respect to the direction of chamber length to generate the acoustic waves. The initial direction of movement can be either towards or away from the nozzle. By repeatedly actuating the deformable wall in quick succession it becomes possible to eject a number of droplets in a single ejection train. These droplets can combine either in flight or on the paper to form printed dots of different sizes depending on the number of droplets ejected.

In Figure 2, a more complex silicon floor plate 20 is used to transmit the force of the actuator element 22 to the ejection chamber 24 rather than the simple flat diaphragm 4 of Figure 1. The plate 20 is formed from two etched silicon wafers bonded together by adhesive or other standard silicon wafer bonding methods and performs two functions. In the first instance it needs to support the actuator and provides a restoring force to bring the actuator back to its steady state rest position as well as to prevent bending forces and moments on the plate from being transmitted to the actuator.

In the second instance the floor plate must be sufficiently stiff so that the volumetric compliance due to changes in ink pressure is low otherwise the acoustic velocity in the ink will be adversely affected.

The floor plate can be seen as effectively forming a parallelogram linkage comprising flexure elements 26 with respect to a rigid element 21 , the actuator acting directly onto the rigid element.

The usefulness and benefits of such a floor plate will later be described in greater detail with regard to Figure 21.

Whilst, in the example of Figure 2, the floor plate is considered to be a separate plate, it is equally possible to form it as part of the channelled component as will be described with reference to Figure 3.

The channels are at the underside of the component as seen in Figure 3 and are not visible.

Push-rods 30 are formed integrally with the floor 34 of the ejection chamber. A base plate 38 is attached to the component such that it extends over the upstanding walls 32 and isolates the push-rods and the push-rod chamber 36. This base plate is flexible, thus providing a flexible linkage for the end of the push-rod remote from the ejection chamber.

The manufacture ofthe channelled component of Figure 3 is preferably achieved by a mixture of wet etching and deep reactive ion etching (DRIE). A silicon plate is provided and, as shown in Figure 4, is etched from one surface using DRIE to form the ejection chambers 24 and walls dividing the ejection chambers 33.

At a predetermined depth etching is halted and an etch stop layer 34 of silicon dioxide and / or silicon nitride is deposited over the surface of the ejection chamber as depicted in Figure 5. From the opposite side, by DRIE, the pusher rod 30 and dividing walls 31 are formed with the etchant removing silicon to the previously formed SiO₂and / or SiN layer 34. Because this layer is not removed a thin flexible membrane, as in Figure 6, remains to separate the ejection chamber from the pusher rod chamber 36. In Figure 7, a second silicon plate 33 is bonded to the side of the first plate comprising the pusher rod chamber 36. This second plate has a two layer coating, namely SiO₂35 overlaid with a coating of SiN 37, with the SiN preferably extending over a greater area of the second plate than the SiO₂. The second silicon plate 33 is a sacrificial layer that is subsequently removed by wet etching to leave a flexible membrane of SiN and SiO₂as depicted in Figure 8.

As in Figure 9, an actuator (depicted schematically through armature 39) can then be formed on the SiN and SiO₂ membrane using MEMS fabrication techniques. (This process is later described in greater detail with respect to Figures 32 to 40.) The final steps are to remove the SiN or SiO₂ layer that remains in the ink supply ports 6 and to apply cover and nozzle plates.

Figure 10 is a view along line B-B of Figure 3 before the membranes 34 and 35,37 within the ink supply ports 6 are removed. These are removed, preferably by wet etching, to open up the supply ports and allow ink to flow along the ejection chamber. A cover plate is added in Figure 11. Figure 12 shows the cross sectional view across line A-A of Figure 3. The ink channel 24 is bounded on one side by the resiliently deformable channel wall 34, a nozzle plate 31 forming the wall opposed the resiliently deformable channel wall and two rigid non-deformable walls 33.

The pusher-rod 30 is positioned in a chamber located between the resiliently deformable wall and the resiliently deformable base plate 35,37. An actuator is positioned such that an armature 39 acts on the opposite side of the resiliently deformable base plate to the pusher rod.

As the actuator acts on the pusher-rod, both the resiliently deformable floor plate and the resiliently deformable base plate are deformed. In certain circumstances it is desirable that the stiffness of the two resiliently deformable plates is chosen to be different. However, it is equally sufficient that the two resiliently deformable plates are of the same stiffness.

It has also been depicted that the walls 33 bounding the ejection chambers 24 and the walls 35 bounding the pusher-rod 36 chamber are of equal thickness. However, according to particular resiliency of the deformable walls it is sometimes desirable to alter the thicknesses of the walls 33, 35 such that one is thicker than the other.

The actuator, which may include the resiliently deformable base plate, is preferably attached as a plate structure. A preferred method of construction is described later with respect to Figures 32 to 40. As mentioned earlier, the actuator is formed distinct from the channelled component and therefore a number of different types of actuator are appropriate for use with the above described channelled component. The present invention is in certain embodiments particularly concerned with electromagnetic actuators and with new types of electromagnetic actuators preferably manufactured by a MEMS technique.

The preferred magnetic actuator is described with respect to Figure 13. This actuator can be defined as a slotted stator actuator that is deflected by modulating the air gap magnetic bias flux field distribution. The actuator armature 98 moves in the direction of arrow F and pushes against a diaphragm 100 to induce a pressure disturbance, and hence an acoustic wave, in the ink within the ink chamber 102.

The actuator component consists of a permanent magnet 92 that lies between a slotted stator plate 94 and the flux actuator plate 90. The slot of the slotted stator plate contains a multi-turn excitation coil 96. This coil, when excited with a DC current, generates a constant axial force F on the shaped armature 98. Beneficially, the magnitude of the force F is directly proportional to the magnitude of the current i.

Figures 14 to 17 depict the actuating principle of the actuator. Figure 14 shows the path of the field lines from the permanent magnet. As shown in Figure 15, when no current is flowing through the coil the field strengths 120a, 120b are similar at both pole faces of the slotted stator 94. This is achieved by making the armature pole face 'ab' shorter than the stator pole face 'cd'.

When a DC current is passed through the coil the flux lines and field strength are distorted as depicted in Figure 16. Using the equation:

where W is the total energy of the system, B is the flux density in the air gap, μ₀ is the magnetic permeability of free space and V is airgap volume, it can be seen that, because B is squared, the total energy in the system is greater in Figure 16 than in Figure 15.

By the principle of least action, the system attempts to revert to the lowest energy state. The armature is therefore moved down in relation to the stator poles in order to minimise the active height Y₁ as depicted in Figure 17. By reversing the current, it is possible to deflect the armature in the opposite direction thus pushing the diaphragm and decreasing the volume of the ejection chamber.

The dimensions of the actuator are dimensioned with regard to the air- gap g and the required travel t as shown in Figure 18.

In this arrangement, the travel t of the armature defines the height of the stator pole faces x₅, x₆. Preferably, the distance x., is a half of x₅ as this serves to provide an equal linear movement in both of the actuation directions. It is desirable that x., remains within the range g < x₁ < (x₅- g) as field edge effects begin to apply stress to the coil and reduce actuator efficiency outside this range. A clearly defined shoulder 91 serves to define the air gap spacing g and the air gap volume v. The air gap between the flux actuator and the flux actuator plate 90 is also important, hence the overhang 93. This air gap is also of the order g.

Typical dimensions are:

x₅ = t + 2kg y > 2g x₃ > t/2 + kg

where k will typically lie in the range 1 to 3.

It is important that the shape of the armature and the geometry of the air gap are such that the armature has a minimum energy position on excitation of the coil and that this minimum energy position is displaced in the actuation direction from the rest position. This is achieved in the described arrangement essentially through shoulder 91. A wide variety of other orientations are of course possible.

One advantage that the slotted stator or bias field magnetic actuator has over the Lorentz forms of magnetic actuator is that the force acting on the coils is weak. The coils themselves are formed as multiple coils in multiple layers and the limited size of the actuators makes the coils susceptible to damage. Thus, it is important to reduce the force acting on them.

A second advantage is that the armature mass is minimised compared to the Lorenz force types. Minimising the armature mass results in maximising the operational frequency of the droplet deposition device. Advantageously, when compared with a variable reluctance actuator, the force developed is substantially linearly dependent on current regardless of the polarity ofthe current. With variable reluctance type actuators, the force is a function ofthe air gap and is therefore very sensitive to manufacturing tolerances. This requirement for high tolerance is reduced in the flux modulation actuator.

Looking in greater detail at the armature force, it has been found that the armature force F_x can be plotted as a function of the armature position. The graph for the situation where no current is flowing in the coil is given in Figure 19.

It has been noted that there is a dead band lying approximately in the range -kg < x < +kg where the armature force F_x is close to zero. A field from the permanent magnet is, however, continually present but force is only applied to the armature when a current is applied to the coil. When a non zero coil current i is applied to the excitation coil, the magnetic field in the air gap 'ab' is distorted with the field in the slot remaining relatively weak. This field distortion generates a force on the armature.

In the case where the flux density in the air gap due to the permanent magnet is B, the coil length L and the coil has N turns, the flux linkages with the coil is 2BΔxLN when the armature moves upwards by a distance Δx in time Δt. By the conservation of energy and the principle virtual work, the force F acting on the armature is given by

FΔX = (2BΔXLN / Δt)iΔt

So that F = 2BLNi

The force of the actuator plotted as a function of the coil current is given in Figure 20. The linear nature of the force makes this type of actuator easily controllable simply by varying the current through the coils.

Figure 21 depicts the bias flux actuator attached to an ejection chamber through a pre-described push-rod plate. As mentioned earlier it is a requirement that the push-rod plate does not transmit rotational and bending forces from the floor of the ejection chamber to the actuator.

In the bias field actuator, the air gap spacing is important in defining the dimensions of the armature element. It is noted that, in this embodiment, the armature is fixed only at one point, namely to the channelled or push-rod components. Since the opposite end is free to move within the stator any rotational and bending forces will be transmitted to the armature. This will have a bearing on the air gap and thus the flux density within the air gap. The push- rod component serves to prevent this error.

The actuator plate component can be formed through the repeated formation and selective removal of layers. Appropriate techniques include those known as MEMS fabrication techniques.

Figure 22 illustrates an embodiment of a planar construction of a fluid pumping apparatus. A first planar layer 302 is arranged parallel to a second planar layer 304. An actuator layer separates the two layers 302 & 304, and maintains structural integrity between them. Located in the actuator layer between layers 302 & 304 is an actuator assembly 306 and a push rod 308, which in this case serves as the armature for actuator assembly 306. The push rod is attached to layers 302 and 304 and is thereby constrained to move in an actuation direction 314. The layered construction described so far with respect to Figure 22 is supported on substrate 310 to form a planar component generally designated by numeral 311 Substrate 310 includes a hollow 312 to allow free movement of push rod 308 in the actuation direction (indicated by arrow 314. In order that this motion may occur it can be seen that portions 303 of layer 302 are resiliently deformable. Corresponding portions 305 of layer 304 are also resiliently deformable. Also shown in Figure 22 is a walled component 316 defining an open channel generally designated by numeral 318. Component 316 further includes a channel outlet 319, and has attached a nozzle plate 320. It can be seen from Figure 22 that walled component 316 can be mated with planar component 311 to form a fluid pumping apparatus. Such a pumping apparatus can be operated to cause a flow of fluid from channel 318 through said outlet 319. Channel 318 may be supplied with fluid from a fluid supply (not shown).

In a preferred arrangement the armature 308, which is constrained to straight line movement by the flexible portions 303, 305 functioning as a parallelogram linkage, is subject to an electromagnetic force provided, for example, by the arrangement of Figure 13. Figure 23 is a view of a channelled construction forming part of a fluid pumping apparatus. A first planar component 352 comprises a first resiliently deformable layer 354; a second resiliently deformable layer 358; and an actuator arrangement 360. Actuator arrangement 360 includes a number of armatures 362 bonded to and carried between the layers 354 and 358. The regions 356 of the layer 354 overlying the armature 352 will remain stiff, and - on actuation - will move in translation as shown on the right hand side of the figure in an actuation direction perpendicular to the plane of layer 354.

A second component 364 having channel walls 366 defining a channel 370, is arranged to be mated with component 352. In this way, the first layer 354 forms one of the channel walls of channel 370. It can be seen that channel 370 may comprise a number of regions 356 which may be acted upon by actuator arrangement 360 via armatures 362. Each armature may act upon one or more regions 356 of layer 354, and may be individually addressable. In this way a fluctuating pressure distribution may be produced in channel 370. In one embodiment it may be desirable to set up a peristaltic wave in channel 370 through sequential operation of armatures 362. In Figure 23 the armatures are operated by a single multiply addressable actuator assembly 360, however a number or discrete actuators could also be employed in a similar fashion. Regions 356 may be arranged in a wide variety of patterns with respect to channel 370. In Figure 23, there is shown two rows of elongate regions (arranged parallel to the length of the channel) operable by elongate armatures running the length of the portions, and each row having two separately operable regions. In an alternative arrangement there might be provided a series of elongate regions having an elongation direction perpendicular to the channel length, the series extending along the length of the channel. Further possible patterns of regions are included in the scope ofthe claims.

Although a flux modulation actuator has been described as a preferred magnetic actuator, it should be understood that a number of different types of magnetic actuator could be employed in conjunction with the present invention. Figure 24 depicts a magnetic actuator operating according to variable reluctance force. The channelled component 42, and nozzle 44 are formed as described with reference to Figures 1 to 3 above.

An armature 46, is formed from an electroformed, soft magnetic material such as Nickel/Iron or a Nickel/lron/Cobolt Alloy. The armature is designed to provide an element of spring to aid deformation and recoil.

An electroformed stator component 48 of a soft magnetic material is provided with a copper coil 50 encircling the stator core 52. In operation, a DC current is passed through the coil to generate a magnetic field that attracts the armature. The volume of the ink channel is thus increased in order to initiate an acoustic wave. At an appropriate timing, equal to Vi Jc, (where L_cis the effective channel length and c is the speed of sound in the ink) the current is removed to allow the armature to recoil. The recoil reinforces the reflected acoustic wave in the channel and causes a droplet to be ejected from the nozzle 44.

An alternative form of variable reluctance type actuator is depicted in Figure 25. The spring element 56 is formed as a diaphragm of etched silicon or some other other non-magnetic material. A stator 58 forms a central area through which a portion 64 of the armature 62 extends in order to be in contact with the diaphragm. A coil 60 is provided within the stator adjacent to a portion of the armature 62 having a large surface area. Upon actuation, the armature is attracted towards the stator and thus deflects the diaphragm into the channel and causes droplet ejection from the nozzle.

Figure 26, depicts an actuator capable of deflecting using a Lorentz force. A channelled component is formed as described earlier and the actuator component is formed as a separate component and attached to it. An etched silicon actuator plate 74 is formed with a number of holes through which a moveable armature structure is posted. A stationary coil 78 is attached to the underside (or in an alternative embodiment to the upper-side) of the etched silicon plate between the plate and the diaphragm 100. The movable armature structure consists of two metallic extensions 76,

77 joined by a permanent magnet 84. The middle extension is posted through the annulus defined by the coil and is joined to the diaphragm 100. The outer extension extends around the coil and is shorter than the middle extension.

Application of a current to the coil interacts with the permanent magnetic field according to the Lorentz force equation and has the effect of moving the middle extension to deflect the diaphragm. This deflection results in ejection of a droplet from the nozzle.

Whilst all the previous bias flux actuators have been depicted using only a single coil layer it is possible to use two layers of coils as shown in Figure 27. The flux from the magnet is the same whether there is one coil or two. However, the force generated by the armature can be increased by adding a second bias field from the second coil positioned on the opposite side of the magnet to the first coil.

Further preferred actuator embodiments are shown in Figures 28 to 31. Figure 28 illustrates a further alternative actuator arrangement. An armature is provided comprising a central magnetic portion 1504 and two non magnetic rigid portions 1506. The armature is constrained to move in the (generally vertical as viewed in Figure 28) actuation direction at one end by a first planar layer 1508, and at the other end by a second layer 1510. The actuator arrangement includes a supporting substrate 1512. A permanent magnet 1514 is located beneath the substrate with polarity as indicated in the Figure. A magnetic yoke is provided to channel flux from magnet 1514, through magnetic portion 1504 of the armature, and back to the opposite pole of magnet 1514. In the region of the armature, the yoke providing flux to the armature comprises two magnetic portions 1516 and 1518, separated magnetically in the actuation direction. A similar yoke arrangement is provided to return flux passing from the armature back to permanent magnet 1514. In this way it can be seen that a permanent magnetic flux is established which, in the region of the armature, is divided into two substantially parallel flux paths, spaced apart in the actuation direction. These flux paths include air gaps 1520 and 1522 adjacent to the armature. A channel component 1524 is also shown.

Figure 29 depicts substantially the same actuator arrangement as in Figure 28 but now illustrates lines of flux. It can be seen that in this arrangement the flux from the permanent magnet (shown solid line) passes through the armature substantially in a single direction, perpendicular to the direction of actuation (indicated by arrow 1552). Figure 29 also shows excitation coils 1550, and the flux produced from said coils (shown broken line). It can be seen that this secondary flux reinforces the primary flux at flux carrying air gaps 1554 and 1556, and that it acts to reduce primary flux density at air gaps 1558 and 1560. Although the flux passing through the armature remains substantially constant, an unbalanced acts on the armature in the direction of actuation. In Figure 29 the secondary flux has been shown forming a continuous path around both sets of coil windings 1550. Secondary flux may however also be considered to form a closed circuit around a single set of windings as shown in Figure 31. This does not alter the principle of flux modulation providing a force in the actuation direction. The embodiments of Figures 28 and 29 can advantageously be used as the basis for an actuator having multiple armatures with multiple flux carrying air gaps.

Figures 30 and 31 illustrate still further alternative actuator arrangements. Figure 30 shows an actuator arrangement with two armatures 1602 and 1604, each armature having two magnetic portions 1606, and a plurality of non magnetic, supporting portions. A single primary magnet 1608 provides a primary flux (shown solid line) in two flux paths separated in the actuation direction, for each of the magnetic armature portions 1606 of the two armatures. Excitation coils 1610 are provided for each armature, arranged with the coil axis perpendicular to the actuation direction. In this way the secondary flux (shown broken line) for each armature acts to reinforce and cancel the primary flux respectively at corresponding pairs of air gaps to provide a force acting on each magnetic portion of a given armature in the actuation direction. Whilst both armatures in the figure share a permanent magnet providing primary flux, the excitation coils for each armature may be independently actuated to allow each armature to be separately operable. Although Figure 30 shows the two actuators acting on separate channels, they could of course operate on the same channel, spaced in the width, or in the length of channel, operating in unison or in a peristaltic or other cooperative manner.

Figure 31 illustrates a variation on the embodiment of Figure 30. There is again shown an actuator arrangement with two armatures 1602 and 1604, each armature having two magnetic portions 1606, and a plurality of non magnetic portions. Here however, the magnetic portions of the armatures extend and laterally overlap with the yoke in regions surrounding the flux carrying air gaps

1620 (only two such air gaps are shown in the figure). This results in primary flux (shown solid line) in the air gaps having a direction substantially parallel to the actuation direction. The same is true also for the secondary flux (shown broken line) caused by the excitation coils (only one part of the secondary coils has been shown for simplicity). This embodiment is advantageous in that the area of the flux carrying air gaps perpendicular to the flux direction can be greater than in a corresponding embodiment having air gap flux passing in a direction perpendicular to the actuation direction. This enables a greater actuation force to be generated. This embodiment has further advantage in an actuator arrangement formed of a series of parallel layers, each layer being orthogonal to the direction of actuation of the actuation device. In this case, the thickness of the air gap is controlled by layer deposition thickness. The thickness of an air gap formed in this orientation can therefore be more accurately defined than that of an air gap in an orientation as shown in Figure 28 for example, in which the air gap tolerance would be controlled by mask registration.

It should be understood that embodiments of the invention wherein the magnetic portion of the armatures laterally overlap with the yoke in the regions surrounding the flux carrying air gaps, are not limited to the particular example described above. Such a feature could equally be usefully applied to other embodiments of actuator arrangements.

There will now be described an example of a MEMS manufacturing process, with reference to Figures 32 to 40. The example is taken of the manufacture of the structure shown in Figure 21 ln Figure 32, a patterned photo resist 120 is deposited onto the resiliently deformable pusher-rod plate 100 of Figure 21. Subsequently a layer of electroformed nickel alloy 122 is deposited. The nickel alloy will form the first part ofthe armature and a support for the stator. The photoresist, once removed will form an air gap.

Once the first layer of Figure 32 is completed, a subsequent layer of photoresist and metal alloy is similarly deposited as shown in Figure 33. These steps may repeated a number of times until the desired structure is achieved. In Figure 34, a layer is formed in which a permanent magnet 124 is deposited along with the photoresist 120 and the electroformed alloy 122. Further layers of alloy and photoresist are deposited in Figures 35 and 36. It can be seen that in Figures 35 and 36 the profile of a flux carrying air gap is developed. In this particular example the width of the air gap W shown in Figure 36, is controlled by mask registration in the deposition process. At a certain depth, a layer comprising electrical coils 126 is deposited as shown in Figure 37. As multiple layer coils are preferred, this layer may be repeated a number of times. A number of connections and vias may be incorporated into some or all of the layers to allow for electrical connection of the coils. More layers of photoresist and metal alloy are deposited in Figures 38 and 39.

Finally, in Figure 40, the photoresist is removed from the whole construction separating the armature from the remainder of the structure.

Some of the particular embodiments described refer to drop on demand ink jet apparatus, however the invention may find application in a wide variety of fluid pumping applications. Particularly suitable applications include so called "lab-on-chip" applications and drug delivery systems. The invention is also applicable to other droplet deposition applications such as apparatus to create aerosols.

Micro-Electro-Mechanical-System techniques have been discussed as suitable for manufacture of apparatus according to the present invention. MEMS techniques include Deep Reactive Ion Etching (DRIE), electroplating, electrophoresis and Chemical-Metal Polishing (CMP). Examples of general MEMS techniques are discussed in textbooks of which the following are examples:

P. Rai-Choudhury, ed., Handbook of Microlithography, Micromachining, and Microfabrication, Vol 1 and Vol 2, SPIE Press and IEE Press 1997, ISBN 0-8529-6906-6 (Vol 1) and 0-8529-6911-2 (Vol 2)

Mohamed Gad-el-Hak, ed., The MEMS Handbook, CRC Press 2001 , ISBN 0-8493-0077-0

Both magnetic and non magnetic materials are used in the present invention. Suitable materials for use in construction include Si-based compounds, Nickel and Iron based metals including Ni-Fe-Co-Bo alloys, Polyimide, Silicone rubber, and Copper and Copper alloys. A useful review of magnetic materials suitable for use with MEMS techniques (and incorporated herein by reference) is to be found in: J. W. Judy, N. Myung, "Magnetic Materials for MEMS", MRS workshop on MEMS materials, San Francisco, Calif. (Apr. 5-6, 2002) pp. 23-26.

Although embodiments have been shown having particular numbers of channels, actuators and armatures, it should be understood that large arrays of channels and actuators can be manufactured on a single substrate, and that arrays of channels can be butted together.

Whilst embodiments have been described with respect to linear channels. It would be equally possible to utilise other chamber architectures including, but not exclusively, architectures where the acoustic wave travels radially of the nozzle as described with regard to WO 99/01284 the contents of which are incorporated herein.

Each feature disclosed in this specification (which term includes the claims) and / or shown in the drawings may be incorporated in the invention independently of other disclosed and / or illustrated features.

Claims

1. Fluid pumping apparatus comprising a fluid chamber with an inlet and an outlet arranged so that fluctuating pressure within the chamber causes a flow of fluid; and an electromagnetic actuator operating on fluid in the chamber to cause a change in pressure; wherein the actuator comprises an armature displaced through a modulation in the distribution of a flux.

2. Droplet deposition apparatus comprising a liquid chamber; a droplet ejection nozzle communicating with the liquid chamber for the ejection of a droplet; and an electromagnetic actuator operating on liquid in the chamber to effect droplet ejection; wherein the actuator comprises an armature displaced through modulation in distribution of a flux.

3. Apparatus according to Claim 1 or Claim 2, wherein the total flux is of substantially constant magnitude

4. Apparatus according to any one of Claims 1 to 3, wherein the path of the flux includes two or more flux carrying air gaps, and wherein said modulation comprises the modulation of flux density in said two or more flux carrying air gaps.

5. Apparatus according to Claim 4, wherein the modulation of the distribution of flux density in said one or more air gaps comprises the increase in flux density at a first air gap and a decrease in flux density at a second air gap.

6. Apparatus according to any preceding claim, wherein said modulation of flux, is achieved through constructive and destructive interference, between a first magnetic field and a second switchable magnetic field.

7. Apparatus according to any preceding claim, wherein the electromagnetic actuator comprises a primary magnet establishing a flux and secondary electromagnet serving to modulate said flux.

8. Apparatus according to Claim 7, wherein the primary magnet comprises a permanent magnet.

9. Apparatus according to Claim 7 or Claim 8, comprising one or more further actuators, wherein said primary magnet can establish a flux for more than one actuator.

10. Apparatus according to any preceding claim, wherein the armature is shaped such that the minimum energy position of the armature is changed by modulation of the flux distribution about the armature.

11. Fluid pumping apparatus comprising a fluid chamber; a chamber outlet; and an electromagnetic actuator operating on fluid in the chamber to cause fluid flow in the chamber outlet; wherein said electromagnetic actuator comprises: a primary magnet establishing a primary flux; a primary flux path including at least two parallel flux paths, between which said primary flux is divided; a secondary magnet operable to increase the flux in at least one of said parallel flux paths, and decrease the flux in at least one other of said parallel flux paths; and an armature arranged to experience an actuation force in response to variations in distribution of flux between said parallel flux paths.

12. Apparatus according to Claim 11 , wherein said secondary magnet is arranged to produce a flux in a first direction in a first of said parallel flux paths, and flux in a second direction, substantially opposite said first direction, in a second parallel flux path.

13. Apparatus according to Claim 11 or 12, wherein at least one of said parallel flux paths includes an air gap.

14. Apparatus according to any one of Claims 11 to 13, wherein said armature forms part of said primary flux path.

15. Droplet deposition apparatus according to any one of Claims 2 to 14, wherein said fluid chamber comprises an elongate liquid channel and the actuator operates in an actuation direction orthogonal to the channel length.

16. Droplet deposition apparatus according to Claim 15, wherein the liquid channel is capable of sustaining acoustic waves travelling in the liquid along the length of the channel and the electromagnetic actuator serves to create an acoustic wave in the channel and thereby effect droplet ejection.

17. Droplet deposition apparatus according to Claim 15 or Claim 16, wherein the actuator extends along substantially the length of the channel.

18. Droplet deposition apparatus according to any one of Claims 15 to 17, wherein the actuator is remote from the channel.

19. Droplet deposition apparatus according to any one of Claims 15 to 18, further comprising acoustic boundaries at respective opposing ends ofthe channel serving to reflect acoustic waves in the liquid of the channel.

20. Droplet deposition apparatus according to Claim 19, wherein said acoustic boundaries serve to reflect acoustic waves negatively.

21. Droplet deposition apparatus according to any one of Claims 15 to 20, wherein the ejection nozzle is connected with the channel at a point intermediate its length.

22. Droplet deposition apparatus according to any one of Claims 15 to 21 , further comprising a liquid supply providing for continuous flow of liquid along the channel.

23. Droplet deposition apparatus according to any one of Claims 15 to 22, wherein the channel is defined by elongate channel walls, one of said channel walls being resiliently deformable in the actuation direction under the action of said actuator.

24. Droplet deposition apparatus according to Claim 23, wherein said resiliently deformable channel wall forms a liquid seal isolating the actuator from liquid in the channel.

25. Droplet deposition apparatus according to Claim 23 or Claim 24, wherein said resiliently deformable channel wall comprises a substantially rigid element capable of transmitting force from the actuator to liquid in the channel and at least one flexure element.

26. Droplet deposition apparatus according to Claim 25, wherein said rigid element extends along the length of the channel.

27. Droplet deposition apparatus according to Claim 25 or Claim 26, wherein said resiliently deformable channel wall comprises a plurality of flexure elements arranged to constrain movement ofthe rigid element to said actuation direction.

28. Droplet deposition apparatus according to Claim 27, wherein at least one of said flexure elements contacts liquid in the channel and is stiff with respect to liquid pressure.

29. Droplet deposition apparatus according to Claim 27 or Claim 28, wherein said flexure elements are arranged in a parallelogram linkage with respect to the rigid element.

30. Droplet deposition apparatus according to any one of Claims 25 to 29, wherein the armature acts on said rigid element.

31. Droplet deposition apparatus according to Claim 30, wherein the armature is carried on said rigid element.

32. A method of fluid pumping utilising apparatus comprising a fluid chamber with an inlet and an outlet arranged so that fluctuating pressure within the chamber causes a flow of fluid, and an electromagnetic actuator operating on fluid in the chamber to cause a change in pressure, said electromagnetic actuator comprising an armature; said method comprising the steps of: establishing a flux in two or more air gaps abutting said armature; and varying the distribution of flux in said two or more air gaps so as to apply a force to said armature in an actuation direction.

33. A method according to Claim 32, wherein varying the distribution of flux does not substantially alter its total magnitude.

34. A method according to Claim 32 or Claim 33, wherein varying the distribution of flux comprises increasing the flux density at a first air gap and decreasing the flux density at a second air gap.

35. Actuation device manufactured by MEMS technology and having an electromagnetic actuator; wherein the actuator comprises an armature displaced through modulation in distribution of a flux.

36. Actuation device according to Claim 35, wherein the total flux is of substantially constant magnitude

37. Actuation device according to Claim 35 or Claim 36, wherein the path of the flux includes two or more flux carrying air gaps, and wherein said modulation comprises the modulation of flux density in said two or more flux carrying air gaps.

38. Actuation device according to Claim 37, wherein the modulation of the distribution of flux density in said one or more air gaps comprises the increase in flux density at a first air gap and a decrease in flux density at a second air gap.

39. Actuation device according to any one of Claims 35 to 38, wherein said modulation of flux, is achieved through constructive and destructive interference, between a first magnetic field and a second switchable magnetic field.

40. Actuation device according to any one of Claims 35 to 39, wherein the electromagnetic actuator comprises a primary magnet establishing a flux and secondary electromagnet serving to modulate said flux.

41. Actuation device according to Claim 40, wherein the primary magnet comprises a permanent magnet.

42. Actuation device according to Claim 40 or Claim 41 , comprising one or more further actuators, wherein said primary magnet can establish a flux for more than one actuator.

43. Actuation device according to any one of Claims 35 to 42, wherein the armature is shaped such that the minimum energy position of the armature is changed by modulation of the flux distribution about the armature.

44. Actuation device according to any one of Claims 35 to 43, manufactured by the selective deposition and removal of successive layers of material.

45. Actuation device according to Claim 44, wherein each layer of material may comprise magnetic and non magnetic materials.

46. Actuation device according to any one of claims 32 to 45, formed of a series of parallel layers, each layer being orthogonal to the direction of actuation of the actuation device.

47. Actuation device according to any one of Claims 32 to 46, wherein manufacturing includes an electroforming process.

48. Actuation device according to any one of Claims 32 to 47, wherein manufacturing includes an electroless deposition process.

49. Actuation device according to any one of claims 32 to 48, manufactured on a silicon wafer.