CA2411245C

CA2411245C - Method for controlling the position of a permanent magnetically supported rotating component

Info

Publication number: CA2411245C
Application number: CA002411245A
Authority: CA
Inventors: Jan Hoffmann; Andreas Arndt; Tobias Merkel
Original assignee: Berlin Heart GmbH
Current assignee: Berlin Heart GmbH
Priority date: 2001-04-30
Filing date: 2002-04-29
Publication date: 2008-06-17
Anticipated expiration: 2022-04-29
Also published as: US20030187321A1; WO2002088548A1; CN1462344A; ATE337491T1; RU2277936C2; JP2004519994A; EP1386081B1; JP2006087298A; AU2002254996B2; JP3994343B2; CA2411245A1; DE10123138B4; DE10123138A1; CN1178006C; US7229474B2; EP1386081A1

Abstract

As especially at a pulsating flow through a pump with a magnetically supported rotor (permanent magnets and additional control current coils) continuous interference forces act on the rotor, the position control has to adjust the changed axial rotor position very quickly. On the other hand the control current should cause a low dissipation.

According to the present method the current through the control current coils is pulse width modulatedly controlled by a desired value, set in a controller arranged behind the position sensory analysis mechanism and the actual value of the position sensory analysis is stored for a defined time interval, respectively starting latest with the pulse edge of the control current, and the position sensory analysis is stopped for this time interval.

Description

Method for controlling the position of a permanent magnetically supported rotating component Description The invention relates to a method for controlling the position of a permanent magnetically supported rotating component, e.g. of the rotor of a synchronous motor without brushes, by means of a position determination of this component by means of a position sensory analysis and additional control current coils, influencing the magnetic field of the permanent magnetic support and which current value is determined by the position of the component. The synchronous motor can for example serve as a drive for an axial fluid delivery pump.
Multiphase fluids, e.g. emulsions and dispersions with a low stability can easily reach into instable areas during the delivery in corresponding delivery systems.
An especially sensitive fluid is blood. Blood is hermetically shielded in the natural recirculation system from the environment, so that no foreign interferences are acting upon it. If, however, the necessity exists, to substitute the heart by an artificial blood pump or to support the recirculation by an additional heart pump, reactions of the blood with the technical system are produced. The blood is subjected, then, easily to the haemolysis or the formation of thrombus with the corresponding disadvantageous effects for the patient. Therefore, recently large efforts were made, to form fluid delivery pumps in such a way, that the blood or other sensitive fluids are subjected to the lowest possible mechanical influences. One possibility for this is the magnetic support of the rotating element of a pump drive. The advantage of the magnetic support is not only, that no components mechanically in a frictional way are present any more, but that also the achievable rotational acceleration of the rotating element is increased and the controllability of the rotational speed and therewith, of the volume flow can be improved.

Such a fluid pump can be integrated in the known way in a synchronous motor without brushes. The fluid pump consists according to WO 00/640 30 essentially of a cylindrical tube, which can be connected at both sides to a fluid system. The tube is surrounded by the stator, consisting of the metal packet, the winding and the iron flux return hood. The rotor comprises permanent magnetic field exciters and has on its outer cover delivery devices for the fluid, so that the fluid can be axially delivered in the annular space between the tube and the rotor.
The rotor is magnetically supported. It carnes for the purpose on its both end sides cylindrical or annular permanent magnets attached thereon, which are magnetised in the axial direction. The permanent magnets of the rotor are opposed by counter magnetised permanent magnets, which e.g. can be arranged in the end sides of guiding devices, which themselves are mounted in the cylindrical tube.
Both magnet pairs act stabilisingly in radial direction, when they are orientated to attract each other, i.e., the radial support is passively stable. The rotor is, however, instable in the axial direction.
Without additional stabilisation the rotor would be attracted by one of the two pairs of permanent magnets. Therefore, control coils are arranged on the stator sides in such a way, that a current weakens by the in series connected control coils the magnetic field of one of the pairs of permanent magnets and increases the magnetic field of the other pair of permanent magnets. The control current has to be adjusted in dependency of the actual axial rotor position. For this, the rotor position has to be determined by means of position sensors.
The position sensors consist for example of two sensor coils, which can be arranged on the end sides of the guiding devices. The sensor coils are opposed on the ends of the rotor by aluminium bodies, in which eddy currents are formed, when the sensor coils are loaded by an alternating current. By the axial movement of the rotor a change in the inductance of the sensor coils is produced, which in an arrangement in a bridge connection can be evaluated as a measuring signal for the rotor position.

As especially in a pulsating flow through the pump disturbing forces act continuously on the rotor, the position control has to be able to quickly adjust a changed axial rotor position. On the other hand the control current should cause a low dissipation, which is especially important for blood pumps, as the produced heat energy should be kept S as small as possible. Furthermore, the drive energy has to be taken from implanted batteries, which operation time should be as long as possible.
The invention has the object to provide a method for controlling the position of a magnetically supported component, with which a dissipation of the control of the position can be kept small.
The object is solved according to the invention by the features of claim 1.
Suitable embodiments are subject of the dependent claims.
According to this the current through the control current coils is pulse width modulated according to a desired value, to which a controller arranged behind the position sensory analysis is set, wherein at a high desired value a switching to a higher voltage level takes place. This has the advantage, that the adjustment times can be kept very small and the necessary power can still be kept low.
The actual value of the position sensory analysis is stored for a defined time interval, respectively starting latest with the pulse edge of the control current and the position sensory analysis is stopped during this time interval.
Advantageously in the use of the position control in a synchronous motor without brushes, the actual value of the position sensory analysis is also temporarily stored in reference to the gatting impulse of the motor coils for a defined time interval, starting latest with the pulse edge of the gatting impulse, and the position sensory analysis is stopped for this time interval.
The interferences produced by the timing concerning the position determination are controlled by the stopping of the measuring during these time periods and by the storing of the measured values.

For specific applications it can be suitable, to take the square of the desired value of the controller arranged behind the position sensory analysis mechanism and to stop the position adjustment at a time-averaged overshooting of the threshold of this value until the next undershooting of the threshold. Thus, a rise in temperature of the control coils is reproduced and therefore, an overheating is prevented.
Appropriately, a PID-controller with an IZ-component is used for the controller arranged behind the position sensory analysis mechanism.
The invention is described in detail by means of an embodiment. In the corresponding drawings:
Fig. 1 shows a sectional view through a fluid delivery pump, which is suitable for the execution of the method according to the invention;
Fig. 2 shows a representation of the principal of the position control with the additional flow control according to the invention; and Fig. 3 shows a block circuit diagram of the position controller.
Fig. 1 shows such an axial pump suitable for the execution of the method. The drive of the blood pump works according to the principal of an electronic commutated synchronous motor. The motor has a stator, consisting of a metal sheet packet 31, of windings 33 and iron flux return hoods 2, 2a and a rotor 5 with a permanent magnetic core 32. The stator encloses a tubular hollow body 1, in which in axial direction a fluid, in the present case blood, is delivered. The rotor 5 is supported magnetically free of contact.
The magnetical support (bearing) consists of permanent magnets 42, 42a on the rotor end sides and permanent magnets 41, 41a on the end sides of the guiding devices 6 and 7. The guiding devices 6, 7 are mounted on the inner wall of the tubular hollow body 1.

To the magnetic support (bearing) further belong control coils 12, 12a. Sensor coils 43, 43a in the guiding devices 6, 7 and short circuit rings 80, 80a arranged opposed thereto, serve for measuring the actual rotor position.
The pairs of permanent magnets 41, 42; 41a, 42a are, respectively, polarised for attracting each other. Magnetically the pairs are arranged in series.
Without an additional stabilisation the rotor 5 would, however, be attracted to one side, therefore. An instable equilibrium exists in axial direction. In radial direction both magnet pairs act self centring, and, therefore, the radial position is passively stable.
The control coils 12, 12a are connected electrically in series and are magnetically arranged in such a way, that a current weakens the magnetic field of the one pair of magnets and increases the magnetic field of the other pair. The magnetic flux return path is produced via the iron flux return hoods 2, 2a and the metal sheet packet 31 of the stator.
The axial position of the rotor 5 can be determined by means of the sensor coils 43, 43a. The sensor coils 43, 43a are loaded by a higher frequent voltage. By the axial movement of the rotor 5 a change of the inductivity of the sensor coils 43, 43a is produced. By the arrangement of the sensor coils 43, 43a in a bridge connection a measuring signal for the axial position of the rotor 5 can be achieved.
As shown in Fig. 2, at the outlet of a controller, arranged behind the position sensory analysis mechanism, an operating value of the control current is produced by the control coils 12, 12a. The control current is transmitted by means of a current controller to the control coils 12, 12a. The current controller acts as a closed control circle, i.e. it measures the current by means of the control coils 12, 12a and compares the result with a default value (desired current) of the position controller.
By means of a pulse width modulation of a pulsed power stage the actual current is adjusted to the desired current. This process necessitates a specific time, which depends on the difference of the desired current and the actual current. The higher the voltage is, with which the power stage operates, the shorter is the adjustment time of the current controller. On the other hand the dissipation in the power stage increases with the voltage. To be able to achieve a quick reaction of the current controller and a lower dissipation, a higher voltage is additionally switched on, only when a large difference between the desired current and the actual current is present; otherwise it operates with a lower voltage.
By the excitation of the control coils 12, 12a by means of the pulsed power stage interferences, which can disturb the position determination of the rotor 5, are produced in the sensor coils 43, 43a. These interferences couple with each pulse edge on the control coils 12, 12a to the sensor coils 43, 43a and fade out after a defined time interval. Therefore, for the expected time interval of these interferences, the position signal, achieved directly beforehand, is temporarily stored and the position determination is stopped. The position controller works during this time interval with the stored value. When the interference is faded out, the position is again determined by means of the sensor coils 43, 43a. Similar interferences can also be produced by the excitation of the windings 33. Also for these the method of the temporary storing is used. The electronics of the interference suppression receive from the current controller and from the excitation electronics of the motor the exact time of the possible starting point of the interferences, so that they can store the position signals.
Fig. 3 shows the circuitry for the position control of the magnetic support.
From the measured position of the rotor, which loads the path 21, a set current for the control coils 12, 12a, which leads to a secure hovering of the rotor 5 in all operating conditions, is determined and transmitted to the outlet 22 of the position controller.
The position controller consists of a PID-controller, which is characterised by the time constants of the integrator Ti and of the differentiator Td as well as by the multiplication factor kr of a variable gain amplifier. To protect the control coils 12, 12a against a thermal overload the to be expected dissipation is additionally determined from the current square. During a threshold overshooting time averaged over a low-pass, the position control is switched-off, until the threshold is again undershot. The position controller has as an additional function, to keep the current through the control coils 12, 12a as low as possible. By means of an integrator (IZ-component) the set current is coupled back to the controller inlet. As a result the rotor is always positioned at the axial position in the pump, at which only a minimal current flows through the control coils 12, 12a.

' , ~ , 3623 0003 PCT/EP02/04737 Reference numerals list 1 Tubular hollow body 2 Iron flux return hood 2a Iron flux return hood 5 Rotor 6 Guiding device 7 Guiding device 12 Control coil 12a Control coil 31 Metal sheet packet 32 Permanent magnetic core 33 Windings 41 Permanent magnet 41 a Permanent magnet 42 Permanent magnet 42a Permanent magnet 43 Sensor coil 43a Sensor coil 21 Path 22 Outlet 80 Short circuit ring 80a Short circuit ring Ti Integrator Td Differentiator kr Multiplication factor

Claims

1. A method for controlling the position of a rotating component which is supported by a permanent magnetic support, comprising the steps of:
determining the position of the rotating component with a position sensor;
providing control current coils for influencing the magnetic field of the permanent magnetic support;
providing an adjustable control current through the control current coils, the strength of the control current being determined by the position of the rotating component, and being pulse width modulated according to a predefined value which is set in a controller coupled to the position sensor;
switching the control current to a higher voltage stage when a higher target value is required by the controller; and storing the actual value of the position sensor for a first defined time interval starting, at the latest, with a pulse edge of the control current, and stopping the position sensor for the first defined time interval.

2. The method according to claim 1, wherein the rotating component is driven by a synchronous brushless motor having motor windings excitable by a pulse width modulated excitation current having pulse edges, the method further comprising the step of:
temporarily storing the actual value of the position sensor for a second defined time interval starting, at the latest, with a pulse edge of the excitation current and stopping the position sensor for the second time interval.

3. The method according to claim 1, further comprising a step of:
providing a PID-controller with an 1 2-component for the controller coupled to the position sensor.

4. A method for controlling the position of a rotating component which is supported by a permanent magnetic support, comprising the steps of:
providing a position sensor to determine the position of the rotating component;

providing control current coils for influencing the magnetic field of the permanent magnetic support; and determining a control current by the position of the rotating component, by adjusting the pulse width of the control current through the control current coils, the pulse width being modulated according to a predefined value which is set in a controller coupled to the position sensor;
switching the control current to a higher voltage when the predefined value is high; and storing an actual value of the position sensor for a defined time interval starting, at the latest, with a pulse edge of the control current, and stopping the position sensor for said defined time interval.

5. A method for adjusting the position of a rotating component, which is supported by means of a permanent magnet and coils carrying a controlled current influencing the magnetic field of the permanent magnet, the method comprising the steps of:
determining the position of the rotating component with a position sensor;
controlling the controlled current delivered to the coils by varying the pulse width of the controlled current based on the position of the rotating component to achieve a target value;
switching the controlled current to a higher voltage if the target value is too high to be achieved by pulse width modulation;
storing the position of the rotating component for a defined period beginning, at the latest, with the pulse edge of the control current; and deactivating the position sensor during said defined period.

6. The method of claim 1 or 5 further comprising the steps of:
calculating the square of the target value specified by the controller coupled to the position sensor;
deactivating the position sensor whenever the time averaged value of the square of the target value exceeds a threshold value; and maintaining the deactivation of the position sensor until the square of the target value drops below the threshold value.