WO2006053384A1

WO2006053384A1 - Fluid pump

Info

Publication number: WO2006053384A1
Application number: PCT/AU2005/001748
Authority: WO
Inventors: Daniel Lee Timms
Original assignee: Queensland University Of Technology
Priority date: 2004-11-17
Filing date: 2005-11-17
Publication date: 2006-05-26

Abstract

A fluid pump including a housing (101) defining a cavity (102) having a cavity axis. The housing (101) has two fluid inlets (105C, 105D) and outlets (106A, 106B) and a set of coils (11 IA, 11 IB) for generating a magnetic field. An impeller (103) is positioned in the cavity (102), the impeller (103) having two sets of vanes (104A, 104B), each set of vanes (104A, 104B) being adapted to urge fluid from a respective inlet (105C, 105D) to a respective outlet (106A, 106B), in use. A drive system is also provided for rotating the impeller (103) about the cavity axis and controlling an axial position of the impeller (103) in a direction parallel to the cavity axis to thereby control the relative flows from each inlet (105C, 105D) to each respective outlet (106A, 106B).

Description

FLUID PUMP

Background of the Invention

The present invention relates to a motor, and in particular a fluid pump using an impeller which is suitable for use as a heart assist or heart replacement device, such as a BiVAD (Bi-ventricular device).

Description of the Art

The reference to any prior art in this specification is not, and should not be taken as, an acknowledgment or any form of suggestion that the prior art forms part of the common general knowledge.

Due to their non-contact support, magnetic bearings have been gradually introduced into high speed and maintenance free machines. However, initial magnetic bearing systems required a mechanically coupled driving motor. Therefore, when applying this technology to fluid pumps, a seal is required to prevent fluid contacting the driving motor.

To overcome this problem, self-bearing motors have been developed. Self bearing motors create magnetic fields that act to suspend and rotate the rotor, thus allowing the capacity for bearing and motor functions from either the same set of coils, or two coil sets occupying the same circumferential location. A number of magnetic principles can achieve this goal, based on so called Reluctance and Lorentz theories. When applied to centrifugal pump technology, self bearing motors eliminate the need for a shaft or seal, as the rotor is completely encapsulated within the pump cavity.

This is particularly relevant in the area of heart transplant surgery, which was developed as a sophisticated response to the problem of life threatening damage to the human heart.

However, with an increasing average age of the general population, the absolute numbers of severe cardiac malfunction are increasing, providing a large number of patients who are candidates for heart transplant surgery. Unfortunately, the availability of suitable replacement organs has always been limited and is becoming increasingly inadequate for general application to patients at serious risk of death from cardiac causes.

One alternative response to the shortage of available heart donors has resulted in various attempts to develop artificial pumps to augment the action of a patient's heart or, ideally, to be suitable for entirely replacing an incompetent heart.

One of the challenges of providing an effective form of such a pump is the need to efficiently harness an available power supply. Given the critical need for such a pump to reliably operate without cessation, it is necessary to supply peak efficiency in the use of a power source. To function other than as an experimental prototype, such a pump must provide a patient with the option to remove themselves from a refuelling or recharging device for considerable periods of time.

Pumps should also minimise the damaging effect of the pumping mechanism on the cells of the blood. Any pump segment generating excessive shear stress will result in traumatic blood cell lyses in that region. Additionally, segments promoting blood stagnation can result in dangerous clots forming within the pump, which can dislodge and travel through the body, potentially blocking vital arteries within the peripheral circulation.

One of the major points of energy loss and blood trauma arises in relation to the method of rotation of a pump rotor or impeller. The use of conventional bearings to support a rotor may lead to increased power consumption and device lifetime issues relating to frictional resistance and consequently wear respectively. Furthermore, this technique promotes blood cell trauma due to excessive heat generation and high shear stress at the contacting bearing site, coupled with stagnation around the rotating shaft. This requires the need for maintenance procedures to eliminate these adverse affects.

Various attempts have been made to provide magnetic or hydrodynamic suspension of a pump rotor or impeller to provide a frictionless and seal-less bearing.

US Patent No. 6,609,883 entitled "Rotary Pump with Hydrodynamically Suspended Impeller" describes a seal-less, shaft-less pump featuring open or closed impeller blades with their edges used as hydrodynamic thrust bearings. Rotational torque is provided by the interaction between magnets embedded in the blades and a rotating electromagnetic field generated in coils fixed to the pump housing.

Hydrodynamics bearings require relatively small clearances to provide sufficient bearing force. High shear stress is found in such clearances and is a possible site of haemolysis should blood cells enter this region. Hydrodynamic bearings rely on a passive response to forces imposed on the impeller, that is bearing stiffness cannot be altered for a given condition. Should forces exceed the predetermined bearing stiffness, touchdown is imminent. Additionally, rotors suspended hydrodynamically generate bearing force stiffness in direct proportion to rotor rotational speed. Ventricular assist devices have the potential to collapse the ventricle wall and thus block the inlet canula to the pump when excessive flow is demanded by the circulation system. To prevent this affliction, the rotational speed of the pump must be reduced for a period of time to allow the ventricular wall to detach, a situation not afforded to pumps suspended via hydrodynamic bearings, since a reduction in rotational speed results in reduced bearing stiffness and potential impeller touchdown.

In light of the potential difficulties of passive hydrodynamic suspension techniques, other configurations centred on magnetic bearing technology to control the position of the rotor within the pumping cavity, have been proposed, as shown for example if US Patent No. 5,326,344 and International Publication No. WO 99/01663.

US Patent No. 5,326,344 entitled "Magnetically Suspended and Rotated Rotor" describes an impeller for a blood pump which is supported by permanent magnets located in the impeller and pump housing respectively and stabilised by an electromagnet on the housing. The configuration of the magnetic suspension system lends itself to low efficiency due to relatively large magnetic "Air Gaps". That is, the distance from rotor magnets to housing magnets is relatively large resulting in excess magnetic flux leakage and thus poor efficiency. Additionally, the magnetic suspension system is separated from the drive mechanism leading to increased size and electronics to provide both functions independently. International Publication No. WO 99/01663 describes an improved rotor for a blood pump which relies on positioning of an impeller to give dynamic balance which is achieved through a combination of hydrodynamic and buoyant forces. The rotor is driven electromagnetically by means of an electromagnetic drive system operating in conjunction with one or more arrays of permanent magnets. The permanent magnets may be enclosed within radial vanes on the surface of the rotor with the drive components being arranged in a brushless motor configuration. In this device rotational movement of the rotor may occur around an axis which varies relative to the pump housing.

The arrangement was provided to attempt to eliminate the need for highly precise axis control and design fabrication. Part of the problem addressed is the need to avoid the introduction of large forces which otherwise are created when the axis of the rotor shifts away from its normally centrally disposed position in a magnetic levitation system. The rotor is fashioned around an internal bore for accommodating inlet fluid and it is provided with a series of external shrouds providing multiple annular flow channels. This arrangement provides a larger surface area to contact blood during normal pumping operation. Erythrocyte fragility becomes a very relevant concern in such a device as shear force and frictional force interaction between the moving components and the blood volume may tend to lysis of red blood cells.

In addition to this, most of the centrifugal heart pumps used for left ventricular assistance to date, feature a conventional single sided impeller. That is, a rotating impeller with one set of vanes used to urge blood from one inlet to one outlet. Howeverj this technique can result in stagnation zones beneath the impeller, as well as unbalanced hydraulic forces experienced by the impeller which must be countered by forces generated from the suspension technique, thus leading to increased bearing power requirements and an overall loss of efficiency. Placement of magnetic material within the rotor required for suspension and/or drive is limited due to the miniature rotor size. Furthermore, when these pumps are used to address the application of Bi-ventricular assistance, a current technique is to implant and operate two separate pumps, thus resulting in increased implantation size as well as increased control complexity arising from the need to control two independent pumps for left and right heart assistance. Attempts to create a single rotary centrifugal device BiVAD are troubled by difficulties in altering the output flow of each cavity independently, since the impeller has a common rotational speed, as well as preventing leakage from the high pressure left cavity to the low pressure right cavity. The present invention provides techniques to address these limitations.

Summary of the Present Invention In a first broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; and, iii) a set of coils for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having two sets of vanes, each set of vanes being adapted to urge fluid from a respective inlet to a respective outlet, in use; and, c) a drive system for: i) rotating the impeller about the cavity axis; and, ii) controlling an axial position of the impeller in a direction parallel to the cavity axis to thereby control the relative flows from each inlet to each respective outlet.

Preferably reducing the physical separation between a respective set of vanes and the housing increases the efficiency of the vanes thereby increasing the flow of fluid to the respective outlet.

Typically the impeller cooperates with the cavity to define first and second cavity portions, each cavity portion including a respective inlet, outlet, and the impeller and cavity being configured to substantially prevent fluid transfer between the cavity portions.

Typically movement of the impeller towards the first cavity portion increases the flow of fluid from the first inlet to the first outlet through the first cavity, and decreases the flow of fluid from the second inlet to the second outlet through the second cavity.

Typically the drive system includes: a) a set of coils for generating a magnetic field; b) magnetic material provided in the impeller and positioned in the magnetic field in use; and, c) a controller for controlling the magnetic field generated by the coils to thereby control: i) rotation of the impeller about the cavity axis; and, ii) the axial position of the impeller.

Typically the impeller includes a shroud coupled to the vanes, the magnetic material being provided in the shroud.

Typically the magnetic material includes at least one of: a) magnets; and, b) iron.

Typically the drive system includes at least two sensors positioned in the housing, the sensors being adapted to detect the position of the impeller within the cavity, the controller being coupled to the sensors and controlling the axial position of the impeller in accordance with signals from the sensors. Typically the impeller includes a shroud coupled to the vanes, the sensors being adapted to determine the position of the shroud.

Typically the impeller includes position magnets, the sensors being formed from Hall effect sensors adapted to determine the magnetic field generated by the position magnets.

Typically the controller is formed from a processing system having a memory and a processor, the processing system being adapted to generate control signals to control the magnetic field generated by the coils in accordance with a predetermined algorithm stored in the memory.

Typically the processing system is coupled to a signal generator, the control signal being formed by having the processing system cause the signal generator to apply a predetermined current to the coils.

Typically the fluid pump includes a radial bearing for controlling the position of the impeller in a direction orthogonal to the cavity axis.

Typically the radial bearing includes: a) an inner surface of the housing; and, b) at least a portion of an outer surface of the impeller, the outer surface portion being shaped so as to cooperate with the inner surface such that when fluid is pumped in use, a boundary layer is formed between the inner and outer surface portions, to thereby restrict movement of the impeller in a direction orthogonal to the cavity axis.

Typically the impeller includes one or more wedges positioned on an outer circumference of the body to thereby cooperate with the inner surface to hydrodynamically control the position of the impeller in a direction orthogonal to the cavity axis. Typically the dimensions of the vanes are selected to control at least one of the fluid pressure and the flow rate at the outlet.

Typically the dimensions include at least one of: a) The height of the vanes; b) The diameters of the impellers; c) The length of the vanes; d) The width of the vanes; e) The inlet and outlet vane angles; f) The shape of the vanes; and, g) The number of vanes.

Typically the impeller and the cavity have a substantially cylindrical shape.

Typically the drive system is adapted to move the impeller in a direction parallel to the cavity axis in a reciprocating manner to thereby control induce a pulsatile flow from the outlets.

In a second broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; and, iii) a set of coils for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having first and second sets of vanes, each set of vanes being adapted to urge fluid from a respective inlet to a respective outlet, in use, wherein the impeller cooperates with a shaped portion of the housing to define first and second cavity portions, each cavity portion including a respective inlet, outlet, and the impeller and shaped portion being configured to substantially prevent fluid transfer between the cavity portions; and, c) a drive system for: i) rotating the impeller about the cavity axis; and, ii) controlling the position of the impeller within the cavity. Typically the shaped portion is in the form of an annular ridge extending radially inwardly toward the cavity axis.

Typically the shaped portion cooperates with a substantially cylindrically shaped vane mounting.

Typically the vane mounting is used to mount the second set of vanes to the impeller, the second set of vanes having a smaller diameter than the first set of vanes to thereby minimise the cross sectional area of a gap between the impeller and the housing.

Typically the fluid pump is a fluid pump according to the first broad form of the invention.

In a third broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; and, iii) a set of coils for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion, each set of vanes being adapted to urge fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the magnetic field in use; c) at least two sensors, the sensors being adapted to detect the position of the impeller within the cavity; d) a radial bearing for controlling the position of the impeller in a direction orthogonal to the cavity axis; e) an axial coupling for controlling the position of the impeller in a direction parallel to the cavity axis; and, f) a controller coupled to the sensors and the first and second sets of coils, wherein in use the controller to control the magnetic field to thereby control rotation of the impeller about the impeller axis. fourth broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; and, ii) two fluid outlets; iii) first and second sets of coils for generating respective first and second magnetic fields; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the first and second magnetic fields in use; c) at least two sensors, the sensors being adapted to detect the position of the impeller within the cavity; and, d) a controller coupled to the sensors and the first and second sets of coils, wherein in use the controller: i) controls the first magnetic field to thereby control rotation of the impeller about the impeller axis; and, ii) controls the second magnetic field to thereby control movement of the impeller orthogonal to the cavity axis.

In a fifth broad form the present invention provides a fluid pump including: a) a housing having an inner surface defining a cavity having a cavity axis, the housing having: i) two fluid inlets; and, ii) two fluid outlets; iii) a set of coils coupled the housing for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the magnetic field in use; and, iv) an outer surface, at least a portion of the outer surface being shaped so as to cooperate with a correspondingly shaped portion of the inner surface such that when fluid is pumped in use, a boundary layer is formed between the inner and outer surface portions, to thereby restrict movement of the impeller in a direction orthogonal to the cavity axis; c) an axial coupling for restricting movement of the impeller in a direction parallel to the cavity axis; and, d) a controller for controlling the first magnetic field, to thereby cause rotation of the impeller about the impeller axis.

In a sixth broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; and, ii) two fluid outlet; iii) a set of coils for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the first magnetic field in use; c) at least two sensors, the sensors being adapted to detect the position of the impeller within the cavity; and, d) a controller coupled to the sensors and the set of coils, wherein in use the controller controls the magnetic field to thereby control: i) rotation of the impeller about the impeller axis; and, ii) movement of the impeller in a direction parallel to the cavity axis. In a seventh broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; and, ii) two fluid outlet; iii) a set of coils to generate a first magnetic field; iv) a set of magnets to generate an axial control magnetic field; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the first magnetic field in use; and, iv) a number of support magnets positioned in the axial control magnetic field to thereby restrict movement of the impeller in a direction parallel to the cavity axis; c) a controller coupled to the set of coils to control the first magnetic field to thereby control rotation of the driver about the driver axis.

In an eighth broad form the present invention provides a fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; and, iii) a first bearing member; iv) a set of coils for generating a respective magnetic field; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets; and, iv) a second bearing member, the first and second bearing members being in contact in use to thereby restrict movement of the impeller in a direction parallel to the cavity axis; c) a controller coupled to the set of coils, the controller being adapted to control the magnetic field generated by the first set of coils to thereby control rotation of the driver about the driver axis.

In a ninth broad form the present invention provides a fluid pump including: a) a housing having an inner surface defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; iii) a set of coils coupled the housing for generating a first magnetic field; and, iv) a set of magnets coupled to the housing to generate an axial control magnetic field; b) an impeller positioned in the cavity for urging fluid from the inlet to the outlet, wherein the impeller has: i) two body portions defining an impeller axis, ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the first magnetic field in use; iv) an outer surface, at least a portion of the outer surface being shaped so as to cooperate with a correspondingly shaped portion of the inner surface such that when fluid is pumped in use, a boundary layer is formed between the inner and outer surface portions, to thereby restrict movement of the impeller in a direction orthogonal to the cavity axis; v) a number of axial support magnets positioned in the axial control magnetic field to thereby restrict movement of the impeller in a direction parallel to the cavity axis; c) a controller for controlling the magnetic field, to thereby cause rotation of the impeller about the impeller axis.

In a tenth broad form the present invention provides a method of pumping fluid, using a fluid pump having: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; and, ii) two fluid outlets; iii) first and second sets of coils for generating respective first and second magnetic fields; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the first and second magnetic fields in use; and, c) at least two sensors, the sensors being adapted to detect the position of the impeller within the cavity, wherein the method includes: i) controlling the first magnetic field to thereby control rotation of the impeller about the impeller axis; and, ii) controlling the second magnetic field to thereby control movement of the impeller orthogonal to the cavity axis.

In an eleventh broad form the present invention provides a method of pumping fluid using a fluid pump having: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; and, ii) two fluid outlets; iii) a set of coils for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having: i) two body portions defining an impeller axis, the impeller axis being substantially parallel to the cavity axis in use; ii) a set of vanes positioned on each body portion for urging fluid from a respective inlet to a respective outlet, in use; iii) a number of driver magnets positioned in the first magnetic field in use; c) at least two sensors, the sensors being adapted to detect the position of the impeller within the cavity, wherein the method includes controlling the magnetic field to thereby control: i) rotation of the impeller about the impeller axis; and, ii) movement of the impeller in a direction parallel to the cavity axis.

In a twelfth form the present invention provides a computer program product including computer executable code which when executed on a suitable processing system causes the processing system to perform the method of tenth or eleventh broad forms of the invention.

In a thirteenth broad form the present invention provides a impeller for use in a fluid pump, the impeller being mounted in a pump cavity and being adapted to urge fluid from first and second inlets to corresponding first and second outlets, wherein the impeller includes: a) a body defining an impeller axis, the body including first and second body end portions; b) a set of vanes positioned on each body end portion for urging fluid from a respective inlet to a corresponding outlet, in use; c) a number of driver magnets positioned in the first and second magnetic fields in use; and, d) an outer surface portion shaped so as to cooperate with a correspondingly shaped portion of the inner surface such that when fluid is pumped in use, a boundary layer is formed between the inner and outer surface portions, to thereby restrict movement of the impeller in a direction orthogonal to the cavity axis. In a fourteenth broad form the present invention provides a impeller for use in a fluid pump, the impeller being mounted in a pump cavity and being adapted to urge fluid from first and second inlets to corresponding first and second outlets, wherein the impeller includes: a) a body defining an impeller axis, the body including first and second body end portions; b) a set of vanes positioned on each body end portion for urging fluid from a respective inlet to a corresponding outlet, in use; c) a number of driver magnets positioned in the first and second magnetic fields in use; and, d) at least a first part of an axial coupling for restricting movement of the impeller in a direction parallel to the impeller axis.

Typically the axial coupling includes: a) a second set of coils for generating a second magnetic field; and, b) a number of axial support magnets provided in the body and positioned in the second magnetic field in use, wherein the controller controls the second magnetic field to thereby control movement of the impeller in a direction parallel to the cavity axis.

Typically the axial coupling includes the first set of coils and a number of axial support magnets provided in the body and positioned in the magnetic field in use, wherein the controller controls the magnetic field to thereby control movement of the impeller in a direction parallel to the cavity axis.

Typically the axial coupling includes at least one of: a) a set of magnets coupled to the housing to generate an axial control field; and b) a number of axial support magnets provided in the body and positioned in the axial control magnetic field in use to thereby restrict movement of the impeller in a direction parallel to the cavity axis.

Typically the axial support magnets are the driver magnets.

Typically the axial coupling includes: a) a first bearing member coupled to the housing; and, b) a second bearing member coupled to the impeller, the first and second bearing members cooperating to thereby restrict movement of the impeller in a direction parallel to the cavity axis. Typically the radial bearing includes a second set of coils for generating a second magnetic field, wherein the controller controls the third magnetic field using signals from the sensors to thereby control movement of the impeller orthogonal to the cavity axis.

Typically the fluid pump includes at least one sensor positioned at a first end of the cavity and at least two sensors positioned at a second opposing end of the cavity.

Typically the dimensions of the vanes are selected to control at least one of the fluid pressure and the flow rate at the outlet.

Each set of vanes may have substantially identical dimensions to thereby provide fluid at substantially equal pressures and flow rates at the outlets or may have different dimensions to thereby provide fluid at respective first and second pressures at the respective outlet.

Typically a first outlet is coupled to a second inlet. Typically the cavity is substantially rotational symmetric around the cavity axis, the coils being arranged circumferentially spaced around the cavity, in which case the body is typically substantially rotational symmetric around the impeller axis, with the driver magnets being positioned radially outwardly from and circumferentially spaced around the impeller axis.

Typically the driver magnets are positioned radially inwardly of the coils. Typically the coils are mounted on a yoke that includes slots for receiving the coils.

Typically the set of coils includes three pairs of first coils, the first coils in each pair being positioned in circumferential opposition, the controller being adapted to cause current to flow through the coils in corresponding directions, to thereby cause rotation of the impeller.

Typically the controller is adapted to: a) determine a required impeller rotation rate; b) determine a control signal sequence for each pair of first coils; and, c) apply the respective control signal sequence to the respective pair of first coils to urge the driver in a direction predetermined direction thereby causing rotation of the impeller.

Typically the second set of coils include three pairs of second coils, the second coils in each pair being positioned in circumferential opposition, the controller being adapted to cause current to flow through the coils in counter-corresponding directions, to thereby cause radial movement of the driver member with respect to the cavity axis.

Typically the controller is adapted to: a) monitor the sensors; b) determine any movement of the driver axis in a direction radial to the cavity axis; c) select one or more pairs of second magnets; d) generate a control signal for each selected pair; and, e) apply the control signal to the respective selected pair to urge the driver in a direction

Typically the driver magnets include eight circumferentially spaced permanent magnets, the poles of the magnets being aligned perpendicularly to the driver axis, and arranged such that the direction of pole alignment is reversed in adjacent magnets.

Typically the vanes extend from a surface of the body end portions in a direction substantially parallel to the driver axis, and being positioned radially outwardly from and circumferentially spaced around the driver axis. Typically the body includes a substantially cylindrical central body portion positioned between the two body portions, the driver magnets being positioned in the cylindrical central body portion.

Typically the cavity includes first and second cavity end portions each cavity end portion having: a) an inlet positioned along the cavity axis; and, b) an outlet directed orthogonally with respect to the cavity axis, and arranged radially offset from the cavity axis; and, c) a shape for cooperating with the impeller to allow the vanes to urge fluid from the inlet to the outlet in response to rotation of the impeller.

Typically the impeller and the cavity cooperate to substantially prevent fluid flow from one cavity end portion to the other, the vanes on each body end portion having a respective height or length to thereby control the pressure or rate of fluid flow at the outlet in the respective end portion.

Typically the pump includes at least three sensors positioned radially outwardly from and circumferentially spaced about the cavity axis.

Typically the sensors include a radiation source adapted to emit radiation towards the surface of the respective body end portion and a detector for detecting reflected radiation. Typically the impeller includes position magnets, the sensors being formed from Hall effect sensors adapted to determine the magnetic field generated by the position magnets. Typically the position magnets are the driver magnets. Typically the impeller and the cavity have a substantially cylindrical shape. Typically the wedges are provided in a central body portion.

Typically the axial support magnets are formed from a magnetic material positioned in a magnetic field, such as soft iron.

Typically the impeller includes a shroud on each body portion, in which case each shroud can include driver magnets such as axial support magnets.

Typically the pump includes a set of second magnets provided in each of two cavity end portions, in which case the controller controls the second magnetic field to thereby control movement of the impeller parallel to the cavity axis.

Typically the fluid pump includes an axial coupling for restricting movement of the impeller in a direction parallel to the cavity axis.

It will be appreciated that the broad forms of the invention may be used interchangeably and or in conjunction with one another and that the above typical features may also be used in any of the broad forms of the invention.

Brief Description of the Drawings

An example of the present invention will now be described with reference to the accompanying drawings, in which: -

Figures IA to IG are schematic views of an example of a fluid pump incorporating an impeller; Figures 2A to 2F are schematic plan views of examples of inlet and outlet volutes for use in the pump of Figures IA to IE;

Figures 3A to 3H are schematic perspective and plan views of examples of impellers for use in the pump of Figures IA to IE;

Figure 4A is a schematic diagram of an example of a control system; Figures 4B and 4C are schematic diagrams of the positioning of the sensors of the control system of

Figure 4A; and,

Figures 5A to 5D are schematic diagrams of the use of impeller position to control the relative flows provided by the fluid pump.

Detailed Description of the Preferred Embodiments An example of a fluid pump incorporating a double impeller is shown in Figures IA to IF. As shown the pump includes a housing 101 having two end portions 101 A, 101B, defining a cavity 102, containing an impeller 103. The impeller has two set of vanes 104A, 104B, mounted thereon.

The housing includes a clearance gap 108A between a rim 108 and the impeller 103, which cooperate to effectively split the cavity 102, into two cavities 102A, 102B, and reduce fluid flow between the two cavities. Thus in this example, the pump effectively includes two pumps defined by the cavities 102A, 102B.

In use, the impeller 103 is rotated about an axis shown generally at 107. This causes fluid at the respective inlets 105A, 105B to be urged towards respective outlets 106A, 106B, via volutes 137, as will be appreciated by persons skilled in the art. Position and rotation of the impeller 103 are controlled by coupling magnetic fields generated by two sets of coils H lA, H lB, wound on respective stators 115A, 115B, to driver magnets 109 embedded in a shroud 103D provided on the impeller 103.

Operation of the fluid pump and variations thereof is described in more detail in copending International Patent Application number PCT/AU2004/000600, the contents of which is incorporated herein by reference.

In use, this form of pump assembly can provide either assistance or replacement of both ventricles of a mammalian heart. This is achieved by connecting the inlet 105 A, of the pump to either the left ventricle or atrium, imparting energy to the blood via the double sided centrifugal impeller 103, and connecting the outlet 106A, to the aorta, thus assisting the natural function of the left heart. Additionally, the inlet 105B may be connected to the right ventricle/atrium, with the outlet 106B connected to the pulmonary artery, thus providing right heart assistance. In combination, the device functions to assist both ventricles of a failing heart.

An example of the operation of the pump to act as a BiVAD will now be described.

In this example, fluid is drawn from the left ventricle into the cavity 102 A via a canula connected to the inlet 105 A. Fluid passes onto the cylindrical impeller 103, whose vanes 104A urge fluid via the volute

137A to the outlet 106A, returning to the circulatory system via a canula connected to the aorta. This imparts sufficient a flow rate and pressure required by the systemic circulatory system, which in one example includes a pressure rise of lOOmmHg at a flow rate of 5 L/min. Simultaneously, fluid is drawn from the right ventricle into the cavity 102B via the canula connected to the inlet 105B. Fluid passes onto the side 103B of impeller 103, which has vanes 104B that urge the fluid, via the volute

137B, to the outlet 106B. In this case, the vanes 104B are designed to provide a flow rate and pressure suitable for the pulmonary circulatory system, which in one example is a pressure of 20 mmHg at a flow rate of 5 L/min. Fluid is transferred via a suitable canula connected to the pulmonary artery.

Volutes

To improve anatomical compatibility in the heart pump applications, the fluid should enter each cavity from a direction substantially perpendicular to the cavity axis. To achieve this, the inlet includes a 90° a turn to align entry of the fluid into the cavity 102 along the cavity axis 107.

This may be achieved using inlet volutes 105C, 105D, examples of which are shown in Figures 2A and 2B. Flow enters the inlets volutes 105C, 105D via respective ports 105E, 105F and is provided with a rotational component by the shape of the volute shown at 105G, and the lip 105H at the impeller eye. This lip 5H also reduces recirculation and potential stagnation in the region. The inlet volutes 105C, 105D provide additional anatomical compatibility when implanting the fluid pump beneath a subject's heart. In one example the pump, the inlet, and outlet flow should preferably be in the same plane. Additionally, the axial size of the fluid may be reduced by employing a flattened 'pancake' style inlet volute. In use, the inlet volutes 105C, 105D direct blood onto the revolving impeller 103. The vanes 104 have the effect of urging blood via respective outlet volutes 137A, 137B, which describe a spiral channel, which leads to a respective one of the outlets 106A, 106B.

It should be noted that despite the anatomical suitability of this arrangement, inlet volutes 105 A, 105B may be replaced with an elbow bend, or straight canula parallel to the axis of impeller rotation 107. Preferably, fluid leaving the impeller 103 is collected by split volutes 137A, 137B and transmitted to the outlets 106A, 106B, with the angle of spiral corresponding to the relative angle of flow off the vanes 104 at design conditions.

Examples of a number of alternative outlet volutes are shown in Figures 2C to 2F. As shown in Figure 2C, the volutes 137A, 137B commence with a narrow space between the vanes 104 and an inner surface of the cavity 102, known as the 'cutwater' 138A. The volute increases in volume and flares out to be continuous with the respective outlet 106A, 106B, or alternatively can further flare out into a diffuser 138C, as shown in Figure 2E. This volute structure can help decrease turbulent flow and wall friction during flow as the volume of blood is urged towards the outlet, since fluid velocity is kept uniform, thereby reducing shear stress development and preserving the components of the fluid being pumped.

The embodiment shown in Figure 2D implements a split volute. In this example, the radial clearance gap or first volute area 137A between the rotor and the housing increases from the initial point (cutwater) 138A, until it encompasses half or approximately 180° of the impeller. A second cutwater 138B is implemented at 180° as the beginning to the second volute 137B. The split, or 'double' volute encounters a reduced radial thrust and improved efficiency over a larger operating range, which is encountered by a patient with varying levels of heart failure and thus assistance required during the course of their support.

Alternatively, Figure 2F shows a concentric volute. Radial hydraulic thrust is reduced only at the best efficiency point of the single volute, whereas concentric volute encounters the most radial thrust over all expected operating conditions.

Impellers A number of example impeller configurations are described by Figures 3A to 3H.

In particular, as shown in Figures 3A to 3C, the impeller 103 includes respective sets of vanes 104A, 104B, each set of vanes 104 having a shroud 103D mounted thereon.

The shroud 103D improves hydraulic efficiency as well as providing closer proximity for the magnetic material used in the drive system. Thus, whilst, the shroud can lend itself to potentially higher shear values between the shroud 103D and the wall of the cavity 102, it does increase magnetic flux in the drive and bearing systems, as will be described in more detail below.

An alternative impeller arrangement incorporating a shroud is shown in Figure 3 G. A further example of an un-shrouded or semi-open type impeller is shown in Figures 3D, 3E and 3F. This configuration improves the ability to alter fluid output characteristics from each of the outlets 106A, 106B, as will be described in more detail later.

Figure 3A shows a 'backward inclined' vane embodiment, with outer and inner blade diameter 4D, 4C, outer vane angle 4E, inlet vane angle 4F and vane height 4G, which can be used to produce the desired flow characteristics. Alternatively, the vanes 4 can be straight radial in nature as shown in Figure 3H, or forward inclined (not shown). However, difficulties arise in efficiently recapturing the high velocity fluid jets within the volute casings of forward vane impellers.

The development of differential pressures and flow characteristics is achieved by utilising respective dimensions for the sets of vanes 104A, 104B. In particular, since the impeller operates at a single rotational speed, the difference in pressure is achieved by providing different impeller vanes 104A, 104B on the high pressure impeller side 103 A and low pressure side 103B. In this case, the vanes 104B on the impeller portion 103B are of a reduced length, as shown in Figure 3 C, when compared to the vanes 104A on the impeller portion 103 A. In one example, with the impeller 103 rotating at 2400rpm, the required flow rates and pressures can be achieved using the following dimensions:

• Vanes 104A Ll = 22 mm;

• Impeller end 103 A radius Rl = 25mm • Vane lO4B Lr = 11 mm;

• Impeller end 103B radius R = 12.5mm;

It will be appreciated that in this example, the pressure difference may result in minimal blood flow from the cavity 102 A to the cavity 102B. This flow is reduced using the technique described later. In any case, this represents a flow of oxygenated blood into the deoxygenated blood, and therefore does not represent a problem from a medical perspective.

The nature of the vane profile (backward, radial, forward) dictates the performance characteristic curve. That is, the output pressure and flow rate for a particular system resistance. As such, using a combination of different profiles for each side will alter the output characteristics for each possible operating resistance. This may be advantageous when different output flow rates are required from each outlet port, and dealing with changing systemic and pulmonary resistances, as described in more detail later.

Drive System

The drive system is formed from a magnet system formed from the two sets of coils 11 IA, 11 IB, wound on respective stators 115A, 115B, and corresponding driver magnets 109, and which is used to provide both axial suspension and rotation of the impeller 103, with the cavity 102.

In use, when the impeller 103 is centrally positioned in the cavity 102, the force required to maintain the position is minimal. However the rotor position is inherently unstable due to the negative stiffness, or attractive force of the permanent magnets to the stator. Therefore, the active magnetic bearing provides additional magnetic flux to counteract any disturbance forces. This practice is common to magnetic motor-bearing systems derived from so called Lorentz and Reluctance theory, as described in "Active Magnetic Bearings : Basics, Properties and Applications of Active Magnetic Bearings", Schweitzer, G. Bleuler, H.,Traxler, A., (1994).

Operation of the system to position and rotate the impeller 103 may be achieved in a number of manners. For example, the stator 115B can be adapted to act solely as a magnetic bearing, coupling to iron 151 on the first end 103B of the impeller, as shown in Figure 4B. In this case, the rotational torque is only provided by the stator 115A, which has more iron cores, coupled to a suitable number of driver magnets 109 embedded in the end 103 A. Furthermore, Stator 115A may be configured as a magnetic motor bearing.

It is also possible, with suitable geometrical arrangement for pump outlet ports, to switch the stator 115 A with 115B and a corresponding switch of magnets 109 with iron target 151.

Rotation

Rotational torque can be provided by coupling a three phase motor to magnets 109 embedded in the impeller 103. To produce motor drive, the coils are arranged on a common yoke 115 A, which acts as a stator. The yoke is preferably constructed from a laser or water jet cut laminated or sintered iron core, for the purpose of eddy current loss reduction. The magnetic field produced by energising these coils with electrical current is coupled to the rotor via a number of permanent magnets embedded in the rotor.

The magnetic material can be rare-earth, neodynimum, or any material available to provide sufficient magnetic flux. In one example, twelve coils are coupled to eight permanent magnets, but other numbers amounts could be used. For example, in use as a magnetic motor bearing, the stator may have six slots coupled to four permanent magnets embedded in the impeller. This arrangement allows the bottom stator to produce rotational torque and axial restoring force toward stator 115 A.

The coils 11 IA may be provided flush on the surface of the yoke 115A (known as a "slotless" arrangement. Compared to other arrangements this results in a lower production of torque, though significantly reduces the degree of negative stiffness instability as well as reducing the inductance and therefore allows for greater rotational speed generation as the EMF generation is lower.

As an alternative the coils can be located in slots 115A in the yoke (known as a "slot-type"), so that the yoke extends axially as shown in Figures IA to IG. Greater torque generation is possible for this stator type, however speed is limited due to the relatively larger inductance, although sufficient rotational speed is possible for this application.

In one example, the impeller 103 is cylindrical and employs a shroud 103D to improve hydraulic efficiency as well as providing a region for magnet material placement. Alternatively, by employing impeller vanes of small height, the magnetic material may be placed beneath the vanes, thus eliminating the shroud requirement. Additionally, or alternatively magnets can be configured in a Halbach array to improve the flux density. This involves constructing each separate magnet from a plurality of smaller magnets each with slightly changing axial polarity direction. Li use, the driver magnets are arranged with their poles aligned parallel to the axis of the impeller 3, with adjacent magnets having poles aligned in opposing directions. This provides alternating magnetic fields around the impeller.

Axial Positioning Axial position and tilt can be controlled using a pure axial magnetic bearing located at the housing 101B.

The stator 115B is constructed with at least four slots with coils wound around each respective pole. The magnetic field generated by energising these coils with electrical current, couples with an iron only target embedded in the impeller 103 beneath the flat face 151. The magnetic fields generated by the coils 11 IB couple directly to iron embedded within the impeller 103, at end 103B. This provides axial bearing control only, and can therefore be used to control the position of the impeller 103 along the cavity axis 107.

Additionally, permanent magnetic material may be included in the stator. This arrangement produces bias magnetic flux to attract the iron embedded in the impeller axially toward the stator 115B. The magnitude of this attraction is determined to oppose the axial magnetic attraction of the driver magnets 109 to the stator 115 A. Thus, the impeller has zero resultant attractive permanent magnetic force when located in an axially neutral position, which will be discussed in more detail below with respect to Figure 5D.

It will be appreciated by persons skilled in the art that some form of active magnetic suspension must be implemented in at least one degree of freedom as described by EARNSHAWS Theorem. As such, the impeller is inherently unstable in the neutral position, and thus requires active magnetic control, provided by energising the coils 11 IB and 11 IA, to maintain impeller position.

Radial Positioning

The radial degrees of freedom are restricted by a combination of passive magnetic forces caused by the active axial bearing, coupled with a hydrodynamic journal bearing located at 108.

The combination of a concentrically offset impeller relative to the pump casing 108 and relative rotational speed cause a thrust force due to the squeezing of fluid through the gap, as described by Reynolds' theory of lubrication. This radial thrust force is used to balance any radial forces encountered within the pump. The hydrodynamic bearing requires a minimal clearance gap 108 A, which acts to significantly reduce the fluid flow from the high pressure cavity 102 A to the low pressure side 102B therefore minor pollution is experienced.

In another case, the cylindrical circumference of the impeller is constructed from a number of small tapered wedges 130, examples of which are shown in Figure 3B. The combination of these wedges and a relative rotational speed cause a thrust force due to the squeezing of fluid through the tapered gap. This radial thrust force is used to balance any radial forces encountered within the pump.

Hydrodynamic bearing stability is increase by the incorporation of a bias force. In use, due to the geometrical configuration and placement in the body, this may be provided by the weight of the rotor. Furthermore, the choice of volute casing dictates the magnitude and direction of radial force, and thus can be tuned to act in the desired direction. Additionally, appropriate magnetic forces may be implemented to attract the rotor to the desired radial direction.

These techniques all act to improve rotor stability and reduce the incidence of impeller whirl.

As an alternative, radial control can be provided by a set of coils that generate a magnetic field to thereby control movement of the impeller orthogonal to the cavity axis.

Control

An example of the manner in which control of the impeller rotation and position may be achieved will now be described with reference to Figures 4A, 4B and 4C.

In particular, operation of the drive system is typically achieved using a controller, an example of which is shown in Figure 4A. In this example, the controller includes a processing system 20 having a processor 21, a memory 22, and an external interface 23 coupled together via a bus 24. An optional input/output device 25 may also be provided.

In one example, the processing system 20 is coupled to four sensors, 26A, 26B, 26C, 26D and a signal generator 27 via the external interface 23. The signal generator 27 is coupled to the motor and bearing coils, to allow pairs of opposing coils to be selectively activated as shown generally for the coils 11 IA, H lB.

In use, the processor 21 determines from data stored in the memory 22 in response to the sensor input, the current that must be applied to the bearing coils to maintain axial position in response to disturbance loads. In the case of desired axial impeller movement to alter output flow rates, a new "zero" set point is determined by the controller, and the impeller axial position is maintained about this new point. The processor 21 also determines the current to be supplied to the motor coils in order to provide a required rotation speed. In the case of a heart pump, a single speed may be pre-set before the device is implanted. It will be appreciated that this may be achieved by storing information defining the pre-set speed in the memory 22, using either the I/O device 25, or a remote processing system coupled to the external interface 23.

Alternatively different speeds may be set whilst the pump is in use. This may be achieved for example using the I/O device 25, or via a wired or wireless connection to the external interface 23, or in response to external impetus, such as an increase in latent heart rate, or the like, through the use of appropriate detection systems. Detection of this heart rate is described later. In any event, signals representing the currents to be supplied are provided to the signal generator 27, which operates to generate appropriate currents, which are then supplied to the motor coils H lA, 11 IB. This will typically involve activating the motor coil pairs in a predetermined sequence in order to create the desired movement of the impeller.

During rotation, the pairs of motor coils H lA, 11 IB are driven by three phase currents. Exact rotational speed can be monitored by a suitably placed Hall effect sensor mounted to detect the change in driver magnet position with respect to the sensor fixed to a housing position. Alternatively, the back EMF encountered in the 1st set of magnets can be related to rotational speed. Once this rate has been determined, it may be fed back to the controller in order for the maintenance of a set rotational speed.

The sensors 26A, 26B, 26C, 26D located in casing 101B operate by determining the distance "d" between the respective sensor and a surface 113 of the impeller 103, as shown in Figure 4B.

The determination of the distance between the sensors 26A, 26B, 26C, 26D and the impeller surface 113 can be achieved in a number of ways. Thus, for example, the sensors 26A, 26B, 26C, 26D can be eddy current sensors adapted to detect the location of the iron 151. The magnitude of the current generated in the eddy current sensors can then be used by the processor 21 to derive the position of the impeller 103.

Alternatively, sensors 26A, 26B, 26C, 26D would be of hall effect type and located in the housing 101A to target the magnets 109 either embedded within the impeller 103 or in the shroud 103D.

Sensing the position of the impeller using the shroud 103D has the advantage that the distance "d" between the sensor targets (the iron 115 or the magnets 109) and the sensors 26A, 26B, 26C, 26D is less than if the sensor targets 109,115 are positioned in the surface 113 of the impeller. In addition to this, when measuring the distance between the surface 113 and the sensors 26A, 26B, 26C, 26D, the vanes 104 will periodically pass between the sensors 26A, 26B, 26C, 26D and the surface 113, which can affect the measurements. As a result, the use of the shroud 103D tends to allow the processor to determine the position of the impeller faster and more accurately, with a higher degree of resolution.

However, the presence of the shroud 103D can reduce the effectiveness of the impeller 103, by generating higher shear stresses within the respective cavity 102 A, 102B. This is undesirable for some applications, such as use of the system as a heart pump.

Alternatively, each sensor may be formed from a radiation source such as an LED laser, or the like and corresponding optical detector, as will be appreciated by persons skilled in the art. Other techniques may also be used.

An alternative method of sensing the position of the impeller within the cavity 102 relates to the so called 'self sensing' technique, whereby the voltage and current waveforms of each motor and/or bearing coil can be monitored by the processing system 20 to sense the position of the impeller. A component of the current waveform is related to the circuit inductance, which in turn varies inversely with axial movement of the permanent driver magnets 109 in the air gap. Recording and analysing the back EMF voltage generated by the motor achieves rotational position sensing, with signals from the bearing coils determining the position. It will be appreciated that in this example, the bearing coils and the motor coils 11 IA, 11 IB form the sensors 26.

In any event, if the processing system 20 detects movement of the impeller axially, the processor will use a predetermined algorithm to determine the current that needs to be applied to selected pairs of bearing coils H lB. An indication of this is transferred to the signal generator 27, which in turn responds by generating appropriate currents and applying these to the selected ones of the bearing coils 11 IB. The current in the coils 11 IB causes the impeller body to move in the axial direction.

It will be appreciated that the sensors 26 can also be adapted to determine the rotational position of the impeller 1033, thereby allowing the rate of rotation to be monitored and adjusted by the processing system 20. The sensors 26 can be mounted to target onto the iron target beneath the flat impeller surface 113, to directly sense motion in the axial direction only, as required for the axial type magnetic bearing.

In use, the impeller may experience some degree of tilt about the radial x and y axis. Under these conditions, the axial magnetic bearing 11 IB can be used to reduce the effect. For example, this may be achieved by increasing the electrical current to the coil 11 IB located perpendicularly to the axis of tilt whilst reducing the current to the symmetrically opposite coil.

The impeller may also encounter hydraulic or external forces causing displacement of the impeller axially. Additional forces resulting in movement toward the stator 115A can be opposed by increasing the current supplied to the coils H lB of stator 115B. Similarly, forces causing impeller displacement toward stator 115B can be reduced by reducing the current supplied to the coils H lB. Additionally, when employing a magnetic motor bearing at 11 IA, current may be increased to the bearing coils in response to the latter condition to provide additional restoring force. It will be appreciated from the above that a number of different axial and radial bearings can be provided, together with a number of different rotational drives. These may be used in any combination, together with an impeller having a cylindrical shape.

Thus, for example, the system can use varying combinations of:

• active axial suspension provided by magnetically coupling coils in the pump housing with permanent magnets or iron in the impeller;

• passive radial suspension provided by magnetically coupling permanent magnets in the pump housing with permanent magnets in the impeller;

• passive radial suspension provided by hydrodynamic effects between the impeller and the housing; and, • rotating the impeller by magnetically coupling coils in the pump housing with permanent magnets in the impeller.

Relative Flow Control

This axial magnetic bearing configuration enables control of the physical axial "zero" position within the pump cavity. This technique provides the unique ability to alter the respective outflow from the outlets 106A, 106B, which will be described in further detail later.

As described earlier, the profiles of the sets of vanes 104 A, 104B may be of different shape (backward, radial, forward). This enables the single impeller device to change relative flow rates by altering rotational speed. Take for example a backward facing profile for the vanes 104 A, which correspond to the left side of the heart, and a radial/forward facing profile for the vanes 104B, which correspond to the right side of the heart. Operating the pump at a set speed produces a balanced flow from the outlets 106A, 106B. With unchanged systemic and pulmonary resistances, reducing the speed reduces the flow from the outlet 106A, while the flow from the outlet 106B is reduced to a lesser degree. Similarly, increasing the impeller speed increases the flow from the outlet 106A while the flow from the outlet 106B increases to a lesser degree. It is therefore possible to alter relative flow rates from either outlet 106A, 106B by altering rotational speed.

Additionally or alternatively, the left and right impeller vanes may be un-shrouded or semi-open. The efficiency of this impeller type is extremely sensitive to the axial clearance above the impeller vanes. This property can be used to provide altering flow rates from the outlets 106A, 106B. Thus, when the impeller 103 is situated in the axially neutral position as shown in Figure 5C, flows to the outlets 106A, 106B are in balance at 5 L/min. Urging the impeller 103 in the direction of arrow 115 in Figure 5B reduces the clearance between the impeller 103A and housing 101A. This improves the efficiency of the pump defined by the cavity 102A, thereby increasing the blood flow to the outlet 106A. This corresponds to an increase in flow for the left side of the heart. Simultaneously, the clearance in the cavity 102B increases and efficiency is reduced, thereby decreasing the blood flow to the outlet 106B. This effectively increases the output for the left side of the heart, while decreasing output for the right side of the heart at a constant rotational speed.

It will be appreciated that urging the impeller in the direction of the arrow 114 will have the opposite effect

The magnetic bearing system employed has a limit to the range of axial actuation. The efficiency and thus output flow rates of each impeller is dependant on the relative ratio of axial clearance and impeller vane height. Therefore, impeller vane heights must be small to obtain a sufficient change in efficiency and consequent range of output flow rates for small actuations. The combination of these techniques enables the single BVAD device to output varying flow rates from the left and right outlets in response to the requirements of the systemic and pulmonary circulatory systems.

In one example, appropriate axial movement of the impeller 103 can be used to induce a pulsatile flow. That is, reciprocal movement of the impeller 103 along the axis 107 within the cavity 102 will repeatedly improve and then reduce the efficiency of the pumps defined by the cavities 102 A, 102B. This creates periodic fluctuations in blood flow to each of the outlets 106A, 106B, thereby resulting in a pulsatile flow for the left and right sides of the heart accordingly.

The ability to alter the relative flows from the outlets 106A, 106B is still possible. In this case, the controller operates to increase the average the time the impeller spends close to the cavity 102 A, 102B for the pump side requiring more flow, whilst still maintaining a degree of axial reciprocation.

Leakage

In order to reduce the effects of leakage between the two cavities 102 A, 102B a small clearance gap 150 is provided as shown in Figure 5D. This gap is defined between an annular ridge 151, which extends radially inwardly of the housing 10 IB, and a vane mounting 152 on which the second set of vanes 104B are mounted.

The vane mounting 152 has a substantially cylindrical shape, as shown. This helps ensure that the gap between the vane mounting 152 and the ridge 151 is minimised. Additionally however, this also ensures that there is no variation in the size of the gap 150 when the impeller 103 is moved in a direction parallel to the cavity axis 107, as described above.

This arrangement is also used as the second vanes 104B have a smaller diameter than the first vanes 104A. As a result, this effective cross sectional area of the clearance gap is smaller than if it were provided at the perphery of vanes 104 A, and therefore reduces the effective gap area and thus the overall amount of leakage from left to right cavities considerably.

Heart Rate Detection

Physiological control of assist devices is desirable to accommodate changes in patient activity. A parameter must be identified to be indicative of changing physiological requirements. This parameter may be Heart Rate. Heart rate can be inferred in this device from magnetic bearing current. In the assist environment, the failing heart has some latent ability to produce pressure and flow. This reduces the output performance requirements of the device during the systolic period, as the heart assumes some of the load. Therefore, a variable load will be placed on the axial magnetic bearing and motor. The power requirements to provide rotation and sustain axial suspension will vary with each heart beat, and this can be recorded to determine the heart rate. For example, an increase in heart rate may indicate increased activity, therefore pump speed may be increased accordingly.

It will be appreciated by persons skilled in the art that numerous variations and modifications will become apparent. All such variations and modifications which become apparent to persons skilled in the art, should be considered to fall within the spirit and scope of the invention as broadly hereinbefore described.

Thus, whilst the above description focuses on the use of the fluid pump as a heart assist device, it is possible for the fluid pump to be used for other purposes, and indeed any fluid pumping application where the pressure and flow rates fall within the operating parameters of the device.

Additional features and in particular, specific axial and radial supports, as well as drive coil arrangements are shown in more detail in copending International Patent Application number PCT/AU2004/000600, the contents of which is incorporated herein by reference.

Claims

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1) A fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; and, iii) a set of coils for generating a magnetic field; b) an impeller positioned in the cavity, the impeller having two sets of vanes, each set of vanes being adapted to urge fluid from a respective inlet to a respective outlet, in use; and, c) a drive system for: i) rotating the impeller about the cavity axis; and, ii) controlling an axial position of the impeller in a direction parallel to the cavity axis to thereby control the relative flows from each inlet to each respective outlet.

2) A fluid pump according to claim 1, wherein reducing the physical separation between a respective set of vanes and the housing increases the efficiency of the vanes thereby increasing the flow of fluid to the respective outlet.

3) A fluid pump according to claim 1 or claim 2, wherein the impeller cooperates with the cavity to define first and second cavity portions, each cavity portion including a respective inlet, outlet, and the impeller and cavity being configured to substantially prevent fluid transfer between the cavity portions. 4) A fluid pump according to claim 3, wherein movement of the impeller towards the first cavity portion increases the flow of fluid from the first inlet to the first outlet through the first cavity, and decreases the flow of fluid from the second inlet to the second outlet through the second cavity.

5) A fluid pump according to any one of the claims 1 to 4, the drive system including: a) a set of coils for generating a magnetic field; b) magnetic material provided in the impeller and positioned in the magnetic field in use; and, c) a controller for controlling the magnetic field generated by the coils to thereby control: i) rotation of the impeller about the cavity axis; and, ii) the axial position of the impeller.

6) A fluid pump according to claim 5, wherein the impeller includes a shroud coupled to the vanes, the magnetic material being provided in the shroud.

7) A fluid pump according to claim 5 or claim 6, the magnetic material including at least one of: a) magnets; and, b) iron.

8) A fluid pump according to any one of the claims 5 to 7, the drive system including at least two sensors positioned in the housing, the sensors being adapted to detect the position of the impeller within the cavity, the controller being coupled to the sensors and controlling the axial position of the impeller in accordance with signals from the sensors. 9) A fluid pump according to claim 8, wherein the impeller includes a shroud coupled to the vanes, the sensors being adapted to determine the position of the shroud. 10) A fluid pump according to claim 8 or claim 9, wherein the impeller includes position magnets, the sensors being formed from Hall effect sensors adapted to determine the magnetic field generated by the position magnets.

11) A fluid pump according to any one of claims 5 to 10, wherein the controller is formed from a processing system having a memory and a processor, the processing system being adapted to generate control signals to control the magnetic field generated by the coils in accordance with a predetermined algorithm stored in the memory.

12) A fluid pump according to claim 11, wherein the processing system is coupled to a signal generator, the control signal being formed by having the processing system cause the signal generator to apply a predetermined current to the coils. 13) A fluid pump according to any one of the claims 1 to 12, wherein the fluid pump includes a radial bearing for controlling the position of the impeller in a direction orthogonal to the cavity axis.

14) A fluid pump according to claim 13, wherein the radial bearing includes: a) an inner surface of the housing; and, b) at least a portion of an outer surface of the impeller, the outer surface portion being shaped so as to cooperate with the inner surface such that when fluid is pumped in use, a boundary layer is formed between the inner and outer surface portions, to thereby restrict movement of the impeller in a direction orthogonal to the cavity axis.

15) A fluid pump according to claim 14, wherein the impeller includes one or more wedges positioned on an outer circumference of the body to thereby cooperate with the inner surface to hydrodynamically control the position of the impeller in a direction orthogonal to the cavity axis.

16) A fluid pump according to any one of the claims 1 to 15, wherein the dimensions of the vanes are selected to control at least one of the fluid pressure and the flow rate at the outlet.

17) A fluid pump according to claim 16, wherein the dimensions include at least one of: a) The height of the vanes; b) The diameters of the impellers; c) The length of the vanes; d) The width of the vanes; e) The inlet and outlet vane angles; f) The shape of the vanes; and, g) The number of vanes. 18) A fluid pump according to any one of the claims 1 to 17, wherein the impeller and the cavity have a substantially cylindrical shape.

19) A fluid pump according to any one of claims 1 to 18, wherein the drive system is adapted to move the impeller in a direction parallel to the cavity axis in a reciprocating manner to thereby control induce a pulsatile flow from the outlets.

20) A fluid pump including: a) a housing defining a cavity having a cavity axis, the housing having: i) two fluid inlets; ii) two fluid outlets; and, iii) a set of coils for generating a magnetic field; , b) an impeller positioned in the cavity, the impeller having first and second sets of vanes, each set of vanes being adapted to urge fluid from a respective inlet to a respective outlet, in use, wherein the impeller cooperates with a shaped portion of the housing to define first and second cavity portions, each cavity portion including a respective inlet, outlet, and the impeller and shaped portion being configured to substantially prevent fluid transfer between the cavity portions; and, c) a drive system for: i) rotating the impeller about the cavity axis; and, ii) controlling the position of the impeller within the cavity. 21) A fluid pump according to claim 20, wherein the shaped portion is in the form of an annular ridge extending radially inwardly toward the cavity axis.

22) A fluid pump according to claim 20 or claim 21, wherein the shaped portion cooperates with a substantially cylindrically shaped vane mounting.

23) A fluid pump according to claim 22, wherein the vane mounting is used to mount the second set of vanes to the impeller, the second set of vanes having a smaller diameter than the first set of vanes to thereby minimise the cross sectional area of a gap between the impeller and the housing.

24) A fluid pump according to any one of the claims 20 to 23, the fluid pump being a fluid pump according to any one of the claims 1 to 19.