Blood pump system with a magnetically levitated rotor

Abstract

This application relates to a blood pump system, particularly a ventricular assist device (VAD). The system includes a blood pump, which comprises: a housing that includes an inlet and an outlet, preferably axial inflow and tangential outflow; a motor actuator, wherein the motor includes a plurality of motor coils (for driving an impeller); and an impeller, wherein the impeller is located within the housing and includes a plurality of rotor magnets. The system further includes: a drive circuit; and a control unit configured to control the operation of the pump. The control unit is configured to: operate the motor such that the impeller rotates about an axis; and measure the position of the rotor in a direction along the axis using at least one of the plurality of motor coils.

Description

Technical Field

[0001] The present invention relates to the field of blood pumps, and more particularly, to a blood pump system having a magnetically levitated rotor. Background Art

[0002] Currently, the ventricular assist market does not provide a good implantable cardiac assist solution for children with a body surface area less than 1.3 m 2 . Due to the lack of a more suitable implantable ventricular assist device, large ventricular assist devices (VADs) planned and approved for adults are sometimes used in children. The results of off-label use of adult VADs in children are significantly worse than the treatments approved in adults.

[0003] In order to make the VAD suitable for children, the size and fluid performance must be optimized without compromising blood compatibility.

[0004] Currently, blood pumps with magnetically levitated rotors are considered to be the prior art for implementing blood-compatible blood pumps. However, the miniaturization of magnetic bearings is limited by their complexity and the number of components. Summary of the Invention

[0005] The object of the proposed invention is to provide an improved VAD that includes a blood pump suitable for use in or within a child. In particular, a new solution is provided to reduce the complexity and the number of components of a blood pump having a magnetically supported rotor. The present disclosure also relates to aspects of system safety, operating system safety, and system redundancy of the blood pump and aspects of connecting the blood pump to the blood circulation.

[0006] This object is achieved by the devices and methods described herein. Further advantageous aspects of the present invention are provided below. Exemplary embodiments are shown in the drawings.

[0007] The pumps, motor drivers, and methods described herein allow the operation of a motor for driving a rotor, measuring the rotor position using the motor, and influencing the rotor position.

[0008] To drive the rotor using a motor and measure the rotor position in at least one degree of freedom (DOF), e.g., along the direction of the axis of rotation, several embodiments and improvements to the known prior art are proposed.

[0009] The disclosed blood pump system, particularly a ventricular assist device (VAD), the system comprising a blood pump which includes: a housing having an inlet and an outlet, preferably axial inflow and tangential outflow; a motor actuator, wherein the motor includes a plurality of motor coils (for driving an impeller); and a rotor including the impeller, wherein the impeller is located within the housing and includes a plurality of rotor magnets. Additionally, the system includes: drive circuitry; a control unit configured to control the operation of the pump, the control unit being configured to: operate the motor such that the impeller rotates about an axis; and measure the position of the rotor in a direction along the axis using at least one of the plurality of motor coils. The rotor and the impeller are rigidly connected to each other. They may also be integrally connected.

[0010] The motor coils are configured to provide a magnetic field for applying torque to the rotor. The motor coils may also be regarded as actuator coils. The coils of an active magnetic bearing may also be regarded as actuator coils. The motor coils may be used as the coils of an active magnetic bearing.

[0011] In one embodiment, the control unit may be configured to reduce or eliminate switching noise from the motor driver. The motor driver may generate a pulse width modulation (PWM) signal to control the motor coils. The PWM signal is generated by switching a switch, for example, a transistor may be used for switching, which has the side effect of switching noise. For example, switching noise including high-frequency current components and voltage components (especially harmonics of the switching frequency) may be reduced by applying a power filter (low-pass), which is configured to suppress the harmonic components in the switching signal. Another method of reducing switching noise may be to not switch at all, for example, by using a class AB amplifier, or a tracking DC-DC controller with a smoother output than a PWM-based motor driver may also be applied.

[0012] Optionally, the output stage of the motor driver may further include filter elements for filtering out high-frequency signal components. As an example, filter elements having low-pass characteristics may be used. However, in order to remove harmonics in the motor driver signal, a band-stop filter or a band-pass filter may also be used. Filters having multiple stop bands may also be used. Filters having multiple pass bands may also be considered.

[0013] It may be considered to add a high-frequency signal to the filtered motor driver output. For example, this high-frequency signal can be used to determine the impedance of one or more motor coils. Thus, the measurement of the motor current can include the measurement of the motor coil impedance, preferably the measurement of the high-frequency motor coil impedance. The frequency of the high-frequency signal can be greater than the rotational frequency of the rotor. It is conceivable that the high-frequency signal has a frequency greater than, for example, 1 MHz or 3 MHz, but it can also have a frequency between 100 kHz and 1 MHz, for example, between 100 kHz and 300 kHz. The frequency of the high-frequency signal can be inconsistent with the harmonic frequencies of the motor driver output signal. In addition, the high-frequency signal can be close to or equal to the resonant frequency of the motor coil. The resonant frequency of the motor coil can result from the inductance of the motor coil and the parasitic capacitance or stray capacitance of the motor coil. The capacitance of the motor coil can be adjusted, for example, by connecting a capacitor in parallel.

[0014] In one embodiment, it is also possible to, for example, use an inductive shunt voltage measurement to replicate the back electromotive force (BEMF) inside the motor outside the motor. Based on the physical law of induction, the BEMF is generated by moving a permanent magnet close to the motor coil, and these permanent magnets can be fixed to the rotor. Then, the BEMF is generated within the windings of the motor coil. If the inductance and resistance of the motor coil are known, it is possible to replicate (i.e., estimate) the BEMF generated within the motor coil windings. Then, by measuring the current value passing through the coil, the BEMF can be estimated. The accuracy of the estimate may depend on the accuracy of the impedance and resistance values used in the calculation. To measure the current in the motor coil, a shunt resistor can be used; a shunt inductor can also be used.

[0015] It is also conceivable that, preferably using back electromotive force replication and a pair of matched high-pass filter elements and low-pass filter elements, the magnetic field strength is replicated outside the motor in the form of an electrical signal or a digital signal or a combination thereof. As an example, the low-pass and high-pass can be complementary filters, and their transfer functions add up to a constant value. By separately processing the BEMF signal in the high-pass and low-pass signal flow paths, the accuracy of position estimation (for example, position estimation in the axial direction of the rotor) can be improved. The axial direction refers to the direction parallel to the main rotation axis of the rotor, and this main rotation axis actually connects the axial center of the inlet of the blood pump and the center of the back plate of the blood pump.

[0016] In some embodiments, the control unit can reduce voltage transients in the drive line. The control unit can be configured to reduce a trapezoidal or triangular current waveform relative to a sinusoidal current waveform.

[0017] As an option, the control unit may include a DC-DC converter. The control unit may include one or more class-AB amplifiers. It is also conceivable that the control unit includes passive filter elements. This option can be used to reduce and / or attenuate harmonics in the drive line signal, thereby reducing high-frequency switching noise that may contaminate the position measurement, for example. The class-AB amplifier can be used to amplify the drive line signal generated by the control unit, which has more than two amplitude levels (such as a class-D amplifier). The increase in the amplitude level of the drive line signal can significantly reduce the generation of harmonics and switching noise. However, the efficiency of the class-AB amplifier depends on the power supply voltage level related to the signal amplitude. For this reason, an adjustable DC-DC converter can be used to adjust the power supply voltage of the class-AB amplifier, thereby improving its efficiency. The remaining amplitude steps after amplification of the drive line signal can be smoothed by a passive filter (such as a low-pass filter). The passive filtering can be performed before or after the amplification of the drive line signal.

[0018] In one embodiment, the drive line may include no more than four wires, preferably three wires and one redundant wire.

[0019] The blood pump may include a passive magnetic radial bearing, wherein the passive magnetic bearing supports a rotor with one or more permanent magnets. Optionally, the blood pump may include a passive magnetic tilting bearing. In some embodiments, the blood pump may include an active axial magnetic bearing configured to actively control the position of the rotor with respect to the axial degree of freedom. The active magnetic bearing supports the rotor with electromagnetic force, and the electromagnetic force can be adjusted by closed-loop control. According to Earnshaw's theorem, it is impossible to fully support the rotor with passive magnetic bearings. It is also conceivable that, in addition to the axial degree of freedom, another degree of freedom is actively controlled. In this case, the axial degree of freedom can be supported by a passive magnetic bearing.

[0020] The electric motor can be, for example, a brushless DC (BLDC) motor. Other types of motors can be used, such as synchronous motors or induction motors or other types of motors (such as DC motors). It is also conceivable that the motor is an axial-flux motor, preferably an ironless axial-flux motor.

[0021] The motor of the blood pump system may include a capacitor electrically connected in parallel with the motor coil, wherein the motor coil and the capacitor form a resonant circuit, and the resonant circuit has an impedance including amplitude and phase and a resonant frequency. In a high-frequency equivalent circuit, the motor coil can be regarded as a series connection of an inductor and a resistor and a parasitic or stray capacitance in parallel with the series connection of the resistor and the inductor. The capacitance of the capacitor and the stray capacitance are added together. The capacitance and the inductor form a resonant circuit with a resonant frequency. The resistor introduces attenuation to the resonant circuit. The resonant frequency of the resonant circuit can be adjusted by using an additional capacitor.

[0022] In addition, the motor coil may include a first coil, and a first capacitor may be electrically connected in parallel with the first coil, and the two form a first resonant circuit. In addition, the motor coil may include a second coil, and a second capacitor may be electrically connected in parallel with the second coil, and the two form a second resonant circuit. The capacitance of the first capacitor may be different from the capacitance of the second capacitor, and the resonant frequency of the first resonant circuit is different from the resonant frequency of the second resonant circuit. By making the first capacitor and the second capacitor have different capacitances, different resonant frequencies can be achieved in the resonant circuits of different motor coils arranged in different motor phases. This can be used to assign resonances to motor phases or motor coils and determine the position of the rotor based thereon (e.g., relative to a spatial coordinate system fixed to the motor stator).

[0023] The blood pump system may further include a measurement unit configured to determine the impedance of one or more of the resonant circuits. The impedance is a quantity that may depend on the rotor position. The combined impedance of the two motor coils can be determined based on measurements at the respective phase terminals of the motor. If the capacitances of the parallel-connected capacitors for each motor coil have different values, such that the resonant frequencies of the resonant circuits in each motor phase are different, then the contribution of the impedance of each motor coil to the combined impedance can be determined.

[0024] It is conceivable that the blood pump system includes an estimation unit configured to estimate the translational and / or rotational position of the rotor based on the impedance of one or more of the resonant circuits. For example, when different materials are used along the periphery of the rotor, the impedance of the motor coil varies with the angular position of the rotor, i.e., the resonant frequency of the motor coil varies with the angular position of the rotor. By evaluating the impedance of each motor coil relative to a reference resonant frequency (e.g., the average resonant frequency), the angular position of the rotor can be determined by analyzing the impedance variations of one or more of the motor coils.

[0025] All measurements affected by the impedance of the motor coil or its resonant behavior are also considered impedance measurements. For example, by using the resonant circuit as a frequency-selective element in an oscillator circuit, the resonant frequency can be measured without directly measuring the impedance. However, the frequency behavior is fully described by the impedance characteristics of the resonator. Other embodiments of indirect impedance measurements include resonant decay measurements, oscillator quality measurements, or phase shift measurements.

[0026] As an option, a test signal can be fed into the motor coil, where the test signal can include components that have been modulated by at least one of amplitude modulation, frequency modulation, phase modulation, and code modulation, and where the code modulation component preferably includes a random code modulation component or a pseudo-random code modulation component. Using this modulation, the test signal can be detected in a more robust manner. Also as an option, the blood pump system can further include a detector unit, which preferably includes a correlator or a synchronous detector, and which is configured to detect the test signal in the voltage measured across the motor coil and / or in a signal (such as BEMF) derived from the voltage. In one embodiment, the detector unit is configured to estimate the impedance of the motor coil based on the detected test signal. For example, based on the measurement of the voltage across the motor coil and the measurement of the current in the coil (such as using a shunt resistor, and by using the estimated impedance of the motor coil that includes the estimated resistance and the estimated inductance), the BEMF generated in the coil can be determined by estimating the voltage across the estimated impedance of the motor coil and subtracting the estimated voltage from the voltage measured across the coil. If the estimated impedance of the motor coil is accurate, the modulated test signal will disappear from the estimated BEMF. Otherwise, within the range of the estimated BEMF, the modulated test signal can still be detected by the detector unit.

[0027] Additionally, in one embodiment, during operation of the blood pump system, the impedance of the motor coil can be continuously estimated. The estimated impedance of the motor coil can also be used to calculate a replica of the BEMF, and the replica of the BEMF can be calculated continuously.

[0028] As a conceived option, the estimated impedance of the motor coil can be estimated by minimizing the test signal component within the BEMF replica. The amplitude of the detected test signal can be used to adjust the estimated impedance of the motor coil, for example, by minimizing the amplitude of the detected test signal with respect to the estimated impedance of the motor coil. The minimization of the test component can be achieved by a digital minimization or optimization program (such as a gradient-based algorithm or a similar algorithm, or a control algorithm like a PI-controller or an I-controller or a similar one).

[0029] As a possibility, the magnetic field strength can preferably be replicated externally to the motor in the form of an electrical or digital signal by integrating the back electromotive force replica with an integrator, where the integrator is numerically stabilized by feeding the output signal of the integrator back to the input of the integrator via a moving average filter that produces an average signal. The stabilization of the integrator means preventing the output signal of the integrator from drifting, i.e., making the average value of the integrator output signal reach or remain at zero.

[0030] By feeding back the average value of the integrator output to the integrator input, in a stationary situation, it is possible to prevent a non-zero average value of the integrator output signal. Thus, in one embodiment, the BEMF replication can be the input signal of the integrator, and the average signal can be subtracted from the input signal of the integrator. In certain embodiments, it may be advantageous that the averaging time of the moving average filter is one rotation period of the rotor or an integer multiple of one rotation period. The average signal can also be low-pass filtered before being subtracted from the input signal of the integrator.

[0031] The blood pump system may further include a connection system for use in medical applications, the connection system including:

[0032] - A cannula made of a flexible material, with a claw ring arranged on the cannula and having at least two claws, wherein the claw ring surrounds the outer surface of the cannula and is arranged at the cannula end of the cannula to rotate and axially displace on the cannula to a stop, the stop including a collar located on the outer surface of the cannula at the cannula end; and

[0033] - A tube, the tube including a locking ring attached to the tube end and a threaded joint attached to the tube, wherein the claw ring can engage with the locking ring by axially moving the claw ring relative to the cannula towards the locking ring and locking at a position where the axial movement is restricted by the stop through the at least two claws on the locking ring.

[0034] Regarding the advantages of this connection system, the following points can be considered:

[0035] - A simple, rapid and safe connection between a flexible hollow tube and a metal tube can be implemented under the implantation conditions for a blood pump, wherein the connection can be released by rotating the claw ring and pulling the claw ring axially out of the cannula;

[0036] - The outer diameter of the connection position between the cannula and the tube increases only a small amount relative to the outer diameter of the cannula or the tube, which reduces the weight;

[0037] - The connection conditions are reproducible;

[0038] - Different from the connection using a union nut, the axial retention and the radial and / or axial sealing are separate;

[0039] - The claw ring ensures axial retention and prevents separation of the connection. Radial and / or axial sealing is achieved by optimizing the diameter ratio and / or thickness ratio at the threaded joint (hose coupling) and the cannula. Thus, no indeterminate axial or radial forces are applied to the cannula, i.e., the material of the cannula is not adversely affected. Additionally, the additional rotary lock is eliminated because the claw ring must be rotated to release the connection, which requires torque that can unfold the claw through the sliding of the inclined plane of the locking ring. If necessary, this torque can be supplied by the operator. Thus, this represents a self-locking arrangement;

[0040] - Through the snap connection, an ideal and gentle transition from the cannula to the tube can be achieved. The connection between the threaded joint in the form of a hose coupling (usually made of titanium) and the cannula actually has zero flow resistance;

[0041] - This connection system can be operated with any known cannula material without substantial changes to the design. According to this connection system, no special tools are required when coupling and disconnecting the device;

[0042] - By cutting off unnecessary strengthening elements, the length of the inlet cannula is adjusted at the pump end. Then the claw ring is pushed onto the cannula again, and the spacer element is inserted into the groove closest to the cannula end. Then, the outlet end of the cannula can be designed without constraints.

[0043] Furthermore, the blood pump system may include means for connecting the cannula to a hollow organ, particularly the heart, characterized in that the cannula tip of the cannula has an opening, which is wavy at its upper edge and provided with notches in order to prevent complete occlusion and retain the blood flow from the hollow organ into the cannula.

[0044] In one embodiment, the cannula can be combined with a suture ring that can be sutured to the heart. It is conceivable that the cannula has a suture flange.

[0045] With the cannula proposed here, the jet flow can be reduced, thereby reducing the incidence of thromboembolism. Compared with the rigid design, the device for connecting the cannula to the hollow organ has two particular advantages: because the apex of the heart has lateral and rotational movements during the cardiac cycle (due to the helical arrangement of the myocardial fibers), the flexible cannula can absorb these movements, thus preventing the formation of forces acting on the myocardial interface. These forces are potentially dangerous because they may cause bleeding or myocardial injury. Additionally, due to the flexible elbow of the device, the surgeon has the opportunity to adapt the position of the blood pump to the anatomical features. Description of the Drawings

[0046] Now turn to the drawings. The drawings described herein illustrate embodiments of the subject matter of the present disclosure and illustrate selected principles and teachings of the present disclosure. However, these drawings do not illustrate all possible embodiments of the subject matter of the present disclosure and are not intended to limit the scope of the present disclosure in any way.

[0047] In the drawings:

[0048] Figure 1 A concept of a magnetically levitated ventricular assist device is shown;

[0049] Figure 2 A blood pump is shown, in which two axial flux motors are used to compensate for magnetic drag (source: US2016 / 0281728A1);

[0050] Figure 3 A VAD with a radial motor is shown, which balances the radial magnetic drag between the rotor and the stator (source: US5588812A);

[0051] Figure 4 A pump that compensates for magnetic drag by using a magnetic compensation bearing is shown (source: US2016 / 0281728A1);

[0052] Figure 5 A VAD with an ironless axial flux motor is shown (source: US006071093A);

[0053] Figure 6 The principle of simultaneously generating axial force and torque with an axial flux motor is shown (source: US006071093A);

[0054] Figure 7 Typical current and voltage waveforms for self-sensing operation according to US6302661B1 are shown;

[0055] Figure 8 A block diagram of rotor position estimation disclosed in US6302661B1 is shown;

[0056] Figure 9 A block diagram of a VAD with sensorless magnetic levitation is shown;

[0057] Figure 10 A block diagram of a VAD with sensorless magnetic levitation is shown, which has additional components to improve system redundancy and the signal-to-noise ratio of rotor position measurement;

[0058] Figure 11 An electrical equivalent circuit of a DC motor is shown;

[0059] Figure 12a An electrical equivalent circuit of a brushless DC motor is shown;

[0060] Figure 12b Shows the electrical equivalent circuit of a brushless DC motor, where in each motor phase, a capacitor is in parallel with the resistance and inductance of the motor coil;

[0061] Figure 13 Shows the electrical and mathematical structure of a prior art electromotive force measurement system for sensorless commutation of a brushless DC (BLDC) motor;

[0062] Figure 14 Shows a method of using an inductive shunt to estimate the BEMF;

[0063] Figure 15 Shows a replica of the simulated BEMF;

[0064] Figure 16 Shows sampling the current using a combined resistive-inductive shunt;

[0065] Figure 17 Shows a method of simulating an inductive shunt using an inductive shunt and filtering;

[0066] Figure 18 Shows a filter structure for generating a magnetic field strength estimate from a BEMF estimate, which uses a pair of matched high-pass and low-pass filters;

[0067] Figure 19 Shows an arrangement of motor coils that can be used to measure the axial and / or rotational position of the rotor using eddy current measurements of the motor coils;

[0068] Figure 20 Shows the equivalent circuit diagram of a motor coil with eddy current measurement and capacitance tuning;

[0069] Figure 21 Shows various implementations of magnetosensitive tuning capacitors;

[0070] Figure 22 Shows how to modify the resonant frequency of a motor coil using a tuning network through a dedicated sensor element with a voltage signal output;

[0071] Figure 23 Shows the voltage and current waveforms in a motor driver with and without a power filter;

[0072] Figure 24 Shows the energy flow and spectrum on the drive line when connecting a radio frequency source to the drive line conductor;

[0073] Figure 25 Shows the energy flow and spectrum of a controller and pump system with integrated radio frequency motor coil impedance measurement and radio frequency filter;

[0074] Figure 26 Shows a block diagram of a noise filter in a current measurement chain;

[0075] Figure 27 Shows a block diagram of a class AB motor driver with a dynamic power supply;

[0076] Figure 28 Shows an AC inverter motor controller;

[0077] Figure 29 Shows the spectra of the motor driver PWM signal and radio frequency impedance measurement without the influence of jitter;

[0078] Figure 30 Shows the influence of clock jitter on the PWM spectrum and the radio frequency impedance measurement spectrum;

[0079] Figure 31 Shows a block diagram of a motor driver and impedance measurement using a shared clock source;

[0080] Figure 32 Shows the electrical structure of a power connector with additional sensing contacts;

[0081] Figure 33 Shows a motor connection scheme using a fourth drive line conductor to increase the redundancy of the VAD system;

[0082] Figure 34 Shows a motor connection scheme connecting a dedicated bearing coil to the star point and the fourth drive line conductor;

[0083] Figure 35 Shows a motor connection scheme with six drive line conductors, a motor, and a dedicated bearing coil, where each drive line can fail in an open connection and the pump can still operate;

[0084] Figure 36 Shows a motor connection scheme without a fail-safe feature with four drive line conductors, a motor, and a dedicated bearing coil;

[0085] Figure 37 Shows a motor connection scheme with seven drive line conductors, a motor, and a dedicated bearing coil, where the pump can operate without reducing the motor efficiency even after an open connection failure of the drive line;

[0086] Figure 38 Shows a cross-section of an embodiment of a blood pump used as a VAD, which uses a combined rotary and linear motor;

[0087] Figure 39a and 39b show a) a cross-section of one embodiment of a blood pump used as a VAD according to this document, and b) a top view of a part of the blood pump rotor;

[0088] Figure 40 Shows a cross-section of one embodiment of a blood pump to be used as a VAD according to the present disclosure;

[0089] Figure 41 Shows a cross-section of one embodiment of a blood pump to be used as a VAD according to the present disclosure;

[0090] Figure 42 Shows a cross-section of one embodiment of a blood pump to be used as a VAD according to the present disclosure;

[0091] Figure 43 Shows a cross-section of one embodiment of a blood pump to be used as a VAD according to the present disclosure;

[0092] Figure 44 Shows a cross-section of one embodiment of a blood pump to be used as a VAD according to the present disclosure;

[0093] Figure 45 Shows a system for coefficient tuning in the context of a BEMF-based rotor position estimation method;

[0094] Figure 46a Shows how integration error is generated when integrating BEMF to obtain the B-field;

[0095] Figure 46b Shows a low-pass filter in a self-regulating servo loop;

[0096] Figure 46c Shows the disadvantages of the servo loop and how a moving average filter compensates for this disadvantage;

[0097] Figure 47 a shows a graph depicting the frequency and impedance curves of a resonant circuit connected in series, where the resonant circuit can be used as an eddy current sensor to measure the rotor position, such as the rotor angle;

[0098] Figure 47 b is similar to Figure 47 a and shows an impedance curve, where in one of the resonant circuits, there is an additional capacitor in parallel with the parasitic capacitance;

[0099] Figure 47 c is similar to Figure 47 b and shows an impedance curve, where there are additional capacitors in both resonant circuits, and each capacitor has a different capacitance;

[0100] Figure 47 d is similar to Figure 47Similarly, an impedance curve is shown, where the rotor presents an inclined position;

[0101] Figure 48 A cross-section of an embodiment of a blood pump to be used as a VAD according to the present disclosure is shown;

[0102] Figure 49 A cross-section of an embodiment of a blood pump to be used as a VAD according to the present disclosure is shown;

[0103] Figure 50 A cross-section of an embodiment of a blood pump to be used as a VAD according to the present disclosure is shown;

[0104] Figure 51 A cross-section of an embodiment of a blood pump to be used as a VAD according to the present disclosure is shown;

[0105] Figure 52 A cross-section of an embodiment of a blood pump to be used as a VAD according to the present disclosure is shown;

[0106] Figure 53 a schematically shows an axial cross-section of the cannula end of the connection system before engagement;

[0107] Figure 53 b schematically shows an axial cross-section of the tube end of the connection system before engagement;

[0108] Figure 53 c schematically shows an axial cross-section of the joint between the cannula end and the tube end of the connection system;

[0109] Figure 54 A perspective view of the snap ring of the connection system is shown;

[0110] Figure 55 a shows a perspective view of the locking ring as viewed towards the tube end of the connection system;

[0111] Figure 55 b shows a perspective view of the locking ring as viewed towards the cannula end of the connection system;

[0112] Figure 56 a shows an axial cross-section of the snap ring of the connection system;

[0113] Figure 56 b shows an axial cross-section of the locking ring of the connection system;

[0114] Figure 56 c shows an axial cross-section of the snap ring and the locking ring of the connection system in the locked state;

[0115] Figure 57 a shows the arrangement when the claw is pushed onto the locking ring of the connection system (locked position);

[0116] Figure 57 Figure b shows the arrangement of the claws on the locking ring after the claw ring has rotated to the unlocked position, where one claw is not deployed and contacts the rotation stop, thereby defining the unlocked position;

[0117] Figure 58 Figure a shows the cannula of the device for connecting the cannula;

[0118] Figure 58 Figure b shows the suture ring of the device for connecting the cannula;

[0119] Figure 58 Figure c shows the suture ring of the device for connecting the suture to the cannula of the hollow organ. Detailed Description

[0120] In order to levitate the rotor, all the forces acting on the rotor must be counteracted by equal reaction forces. A common method is to use passive magnets to generate reaction forces when the rotor is displaced.

[0121] According to Earnshaw's theorem, it is not possible to simultaneously stabilize all degrees of freedom using passive magnets. A permanent magnet arrangement that stabilizes one axis will always destabilize another axis by at least the same amount.

[0122] Earnshaw's theorem also states that for an arrangement of passive magnets, the sum of all stabilities and instabilities is always zero. A common method is to use magnets to stabilize one axis and create some unstable degrees of freedom (DOFs) with unstable equilibrium points. Then, active control is applied to the DOFs with unstable equilibrium. A common control strategy is zero-force control, which balances the DOF at the unstable equilibrium to minimize the necessary power for levitation.

[0123] At the zero-force position or unstable equilibrium, the power consumption can be close to zero and is only limited by the noise and time delay in the control loop. For an implantable blood pump, the common power consumption for levitation control is 500 mW (e.g., Berlin Heart).

[0124] When zero-force control is not operating, the rotor will move away from the equilibrium point and accelerate until it contacts the pump housing. When the rotor remains on the pump housing, the force acting on the rotor is equal to the maximum force the rotor can withstand under zero-force control. This force is also equal to the force required to disengage the rotor from the pump housing. Therefore, it is only possible to reduce the disengagement force by reducing the maximum load that can be borne by the rotor. The peak disengagement power can be several orders of magnitude higher than the zero-force suspension power. In INCORVAD, it is three orders of magnitude (suspension power of 0.5 W, disengagement power of 500 W), and in the HeartMate 3 VAD, it is approximately two orders of magnitude (suspension power of 0.5 W, disengagement power of 50 W). Reducing the disengagement force is one way to reduce the disengagement power.

[0125] The force between the rotor magnet and the motor stator (usually magnetic drag) also contributes significantly to the load on the rotor and the disengagement force. Most brushless DC (BLDC) motors have a magnetic rotor and iron in the stator. This results in a continuous attractive force between the rotor and the stator. This load force can be reduced by using two stator components that act on the rotor with forces that cancel each other out, as can be seen in the stator configurations of Figure 2 (US20160281728A1) and Figure 4 (US20160281728A1). Another option for balancing the static magnetic drag is to use a radial flux motor, as can be found in the HeartMate 2 VAD (US5588812A, Figure 3 ). However, although the load force can be balanced, the magnetic stiffness is always negatively affected due to the introduction of more iron as part of the motor stator. This results in disturbing forces when the rotor is not fully balanced in its equilibrium position.

[0126] By using an ironless stator configuration, the magnetic drag load force can be completely eliminated. The effect of an ironless stator on the magnetic levitation stiffness of the rotor is also zero.

[0127] An ironless axial flux disc motor can generate torque as well as positive and negative pulling forces. Patent document US6071093A ( Figure 5 、 Figure 6 ) shows a radial blood pump with an ironless axial flux motor, where the motor coils are used to generate axial thrust and torque.

[0128] So far, ironless (coreless and yoke - less) axial - flux motors have not been used in VAD systems. The reason may be that the motor efficiency of ironless motors is reduced compared to motors with iron yokes or cores. However, in an actively - levitated VAD system, the drawback of lower efficiency is countered by the gain in levitation stability. As long as the motor can be sufficiently cooled without warming the blood too much, ironless or reduced - iron axial - flux motors can miniaturize the VAD device because no yoke or magnetic compensation bearings are required. Thermal management implementing the features proposed in patent application EP19159286.4 can make it feasible to use ironless axial - flux motors in implantable VADs.

[0129] The pump proposed in patent document US6071093A uses dedicated sensors to detect the axial rotor position. These sensors must be read out either through additional drive - line wires or an electronic system integrated into the pump. Both solutions increase the size of the implantable VAD system and thus reduce the feasibility of pediatric VAD systems.

[0130] Sensorless operation of brushless DC (BLDC) motors is commonly used in VAD systems. Sensorless BLDC operation generally refers to achieving sensorless commutation of the motor phases by measuring the induced back - electromotive force (BEMF) or by measuring the current in the motor phases and estimating the BEMF based on the current waveform. The BEMF signal is proportional to the rotor speed. This means that the signal is very low at low rotational speeds and zero at rest. Below a certain speed, the noise in the current or voltage measurement endangers the detection of the BEMF, and thus sensorless BEMF commutation is not possible.

[0131] For commutation and motor operation, only the rotational position of the rotor needs to be known and measured. For magnetic levitation, the position of the rotor in the unstable degrees of freedom (DOF), which is usually the linear DOF, must be measured.

[0132] Patent document US6302661B1 describes a method that generates an electronic signal related to the rotor position in the linear DOF through a combination of the current and voltage provided by an active coil. Figure 7 shows the measurable voltage and current waveforms in the active coil (as shown in patent document US6302661B1). The expected method for extracting the rotor position (in this case, the bearing clearance) (see Figure 8 ; also taken from US6302661B1) uses an envelope filter to measure the current to condition the signal from the sensor and a control loop with an electromagnetic - body simulation model to estimate the bearing clearance.

[0133] Efforts to replicate the proposed rotor position measurement described in US6302661B1 have shown that the rotor position signal can in fact be detected. However, signal noise and measurement latency must meet certain requirements in order to levitate the rotor based on the measurement. These two requirements are more difficult to meet for smaller and lighter rotors. Thus, the levitation requirements for an implantable pediatric VAD system are particularly difficult to meet. The measurement method described in US6302661B1 produces too much noise.

[0134] The present disclosure proposes new methods for reducing measurement noise at various stages in the measurement chain.

[0135] In addition, VAD systems (especially when operating outside a strictly controlled hospital environment) are constantly subject to various external interferences, which in turn are not conducive to the operational safety of the VAD system.

[0136] These external interferences are mainly electromagnetic in nature and may thus particularly include electromagnetic interferences originating from sources such as mobile phones, RFID, CT scanners, etc. Thus, these interferences may interfere with the sensors involved in various stages of the measurement chain and their associated sensor signals.

[0137] The present disclosure also proposes new methods for improving the operational safety of the VAD system.

[0138] Each levitation control loop requires a rotor position signal at its input. Typically, dedicated sensors such as magnetic field sensors (HeartMate3, Ab-bott) or eddy current sensors (INCOR, BerlinHeartGmbH) are used to measure the position of the rotor. The sensor signal is either evaluated within the pump using the integrated pump electronics or transmitted via additional wires in the drive line to the control unit. Both of these options reduce the feasibility of the VAD system for use in children.

[0139] The proposed sensor solution does not require dedicated sensors, complex electronics within the pump, or additional wires to the drive line. In most cases, the motor structure is directly or indirectly used to measure the position of the rotor.

[0140] The proposed concept was developed for a disk-shaped axial flux motor. However, the sensing method can be used for other types of brushless DC motors and even linear actuators (such as voice coils).

[0141] Figure 9 An exemplary block diagram of a VAD with sensorless magnetic levitation is shown. A power supply 1 supplies power to a switched-mode motor driver 2, such as an inverter that supplies a PWM signal to an actuator coil 3. The signal provided by the switched-mode motor driver 2 corresponds to the voltage across the actuator coil 3 and the current within the actuator coil 3, thereby generating a magnetic field that induces a force on the rotor 4. AsFigure 9 As depicted in, the current waveform in the actuator coil 3 and / or the voltage waveform across the actuator coil 3 can be sampled by the waveform sampling unit 5 and then filtered by a filter 6 (such as a low-pass filter). Based on the filtered waveform (e.g., by evaluating the phase deviation between the current and the voltage), the rotor position can be estimated by the rotor position estimator 7. Based on the estimated rotor position, the levitation controller 8 can calculate an appropriate control signal to control the switched-mode motor driver 2, thereby providing a signal suitable for levitating the rotor 4 to the actuator coil 3.

[0142] Now refer to Figure 10 , which provides an overview of the possible processing blocks of the VAD.

[0143] The actuator coil 3 (which can be a motor coil) is located inside the VAD and is magnetically coupled to the rotor magnet arranged on the rotor 4. The rotor magnet can be arranged in a Halbach array. As the rotor 4 rotates, a BEMF is induced in the actuator coil 3. The amplitude of the induced BEMF depends on the rotational speed and the distance between the rotor magnet and the actuator coil 3, i.e., on the position of the rotor in the Z direction, which is also referred to as the axial direction and may be an unstable DOF (degree of freedom). Therefore, the position of the rotor can be estimated from the BEMF. Since the BEMF cannot be directly measured during motor operation, generally, the time-dependent value of the BEMF needs to be estimated.

[0144] The position of the rotor 4 along the unstable DOF modulates the shape of the BEMF, and in addition to the BEMF corresponding to the rotor rotation, a further BEMF in the motor coil 3 is also induced. The BEMF from the rotor rotation and translation in the Z direction modifies the phase current and terminal voltage of the motor. Both can be estimated by the BEMF duplicator 12, which generates a digital or analog duplicate of the induced BEMF. The BEMF duplicator 12 can further be used to generate an electrical or digital signal having the magnetic field shape (i.e., waveform) inside the actuator coil 3. As Figure 9 illustrates, the duplicated waveform is fed via the waveform sampler 5 and the filter 6 to the rotor position estimator 7, which estimates the linear and / or rotational position of the rotor. The method used by the rotor position estimator 7 can include an arithmetic formula for reversing the induction law, a Kalman filter based on a motor model, or a controller for controlling a motor simulation model. The rotor position is used by the levitation controller 8 to control the motor driver 2 (see Figure 9 or Figure 10 ) to adjust the force on the rotor to levitate the rotor inside the VAD.

[0145] It is also possible to consider the position of the rotor 4 as the input to the coil impedance modulator 13, which is configured to modulate the impedance of the actuator coil 3. Additionally, the actuator coil 3 can be connected to a motor driver (e.g., Figure 10 the switchless motor driver 10 in

[0146] ) via a corrosion-resistant connector 15 and optionally via a power filter 11. The motor driver (e.g., the switchless motor driver 10) can receive electrical power from a power supply 1 via an optional power modulator 9.

[0147] A function generator 16 can impress a current signal (e.g., a sinusoidal signal) into the motor coil. This signal impressed by the function generator 16 will only be picked up by the BEMF replicator 12 if the BEMF parameters (e.g., motor coil impedance estimation) are not perfectly selected. Therefore, if the output of the BEMF replicator 12 is correlated with the impressed signal, the BEMF parameters must be adjusted. The phase and amplitude of the correlation indicate in which direction the parameters must be adjusted. The impressed signal can be a high-frequency signal, e.g., its frequency is greater than the rotational frequency of the rotor. The impressed signal can also be modulated. The modulated signal can be used as a test signal.

[0148] However, even in the absence of BEMF (i.e., when the rotor is not rotating), it is feasible to estimate the BEMF parameters using an external impressed signal impressed by the function generator 16.

[0149] When the distance between the rotor-magnet and the motor coil arranged within the motor stator increases, the magnetic flux reaching the motor coil decreases. This affects the characteristics of the motor setup. Characteristic motor parameters are, for example, the no-load speed and the stall torque. A weaker magnetic field or a larger rotor-stator distance will increase the no-load speed and decrease the stall torque. During operation, these parameters cannot be measured directly. However, both define the current, voltage, and speed behavior of the motor. Measuring any of these values can be used to calculate an estimate of the BEMF.

[0150] The BEMF is related to the magnetic field passing through the motor coil by Faraday's law of induction: (1)

[0152]

[0153] where N is the number of windings and Φ is the magnetic flux passing through the coil.

[0154] The field outside a linear Halbach array rotor can be estimated as: (2)

[0156] B(x, z) = B0e ikx e -kz

[0157] B0 is the magnetic field strength at the surface of the magnet, k is the wavenumber of the magnet, and z is the distance from the magnet surface, where in the Halbach array, the magnets are specifically arranged along a line, and where x represents the displacement of the rotor in the direction of the Halbach array line.

[0158] When the magnets of the Halbach array are arranged in a ring, by inserting (2) into (1), in the case where the rotor rotates at the rotor speed ω, the BEMF is substantially proportional to the field strength and the electrical rotor speed ω: (3)

[0160]

[0161] c geo is the geometric constant of the coil, and t is the time.

[0162] The actually measurable BEMF is:

[0163] (4) = re(Equation 3)

[0164]

[0165] where re(·) represents the real part of a complex number.

[0166] With the known speed and the known (e.g., estimated) BEMF value, the rotor position in the z direction can be calculated. However, when the motor operates under field-oriented control, the BEMF cannot be directly measured. The only measurable motor voltage is the sum of the BEMF, the voltage across the resistance coil, and the voltage across the inductance coil ( Figure 11 the V in Mot 、V BEMF 、V R 、V L ). By using the measurement of the motor current I Mot , the voltages V L and V R inside the motor can be estimated by calculation. After measuring the voltage V Mot across the motor coil, an estimate of V BEMF can be calculated. Figure 11 shows the equivalent circuit of the motor coil, which includes a series connection of a coil inductance L with a corresponding voltage V L , a coil resistance R with a corresponding voltage V R , and a BEMF voltage source V BEMF , and includes a shunt resistor R with a corresponding voltage V Rshunt connected in parallel with the motor coil.shunt and a drive voltage source V Drive The voltage V across the motor coil Mot is V L , V R and V BEMF sum, and the current through the motor coil is denoted as I Mot .

[0167] The speed of the motor can be estimated using, for example, frequency analysis of the BEMF. From these measurements, (5)

[0169]

[0170] c zest is the z - position estimation constant.

[0171] Each phase of the three - phase brushless DC motor can be modeled according to Figure 11 . Measuring only one phase is sufficient to obtain the z - position measurement. However, by measuring all three phases, more accurate measurements can be made. For example, Equation (5) can be evaluated for each phase of the motor and combined into a space phasor with an angle and amplitude from which the z - position of the impeller can be estimated. Equations (3), (4), and (5) are approximations of the quasi - static variation of the axial rotor position z. To estimate the z - position more dynamically, the error introduced by the approximation can be compensated. Known compensation methods include: numerically integrating the BEMF to produce a coil flux estimate; a sliding - mode observer or a PLL - based observer to directly estimate the rotor position.

[0172] The DC equivalent motor current of a brushless DC motor can be accurately measured by measuring the current in the supply of the commutation circuit (see, for example, Figure 12a and Figure 12b ).

[0173] These methods are accurate enough to commutate the motor using the estimated BEMF and are also accurate enough to obtain an average z - position measurement. However, if a fast and accurate measurement of the BEMF is required (e.g., for a position control loop in a levitation device), improvements may be needed to increase the signal - to - noise ratio.

[0174] The proposed improvement in measuring the BEMF reduces the system noise in current measurement, voltage measurement, subsequent filtering, and position estimation. In particular, high - frequency noise reduces the accuracy of the inductance voltage estimate. This is due to the high - pass characteristic of the differentiation operation in the inductance voltage estimate.

[0175] Figure 13 The circuit for calculating the BEMF is shown in Mot(Here represented as the phase voltage V Ph ) waveform is sampled, and thus converted into a digital signal V Ph,s . Similarly, the voltage across the shunt resistor R shunt connected in series with the motor coil (proportional to the current I Ph passing through the shunt resistor and the motor coil) is sampled, and thus converted into a digital signal I Ph,s . Before sampling the voltage across the shunt resistor, the voltage can be amplified, where the amplifier can have a high input impedance. I Ph,s is processed in the first processing path and the second processing path. In the first path, I Ph,s is derived with respect to time and multiplied by the (estimated) value of the coil inductance to calculate the estimated V L of the voltage across the coil inductance L L,s . In the second path, I Ph,s is multiplied by the value R to obtain the estimated V R of the voltage across the coil resistance V R,s . Finally, the estimated V BEMF of the BEMF in the motor coil is obtained through V Ph,s = V R,s - V L,s - V BEMF .

[0176] As Figure 13 shows, the circuit for calculating the BEMF has limitations. The embodiments for overcoming Figure 13 the limitations of the circuit will be explained below.

[0177] Figure 14 The circuit of Figure 13 is similar to the circuit of Figure 14 . However, in shunt , the inductive shunt L shunt is connected in series with the resistive shunt R Figure 13 . Different from shunt , the voltages across R shunt and L R,s are amplified, where each amplifier can have a high input impedance and are then respectively converted into digital signals V L,s . The motor current and its transients also flow through the resistive shunt and the inductive shunt. The voltage across the resistive shunt is a scaled image of the voltage across the motor coil resistance. The voltage across the inductive shunt is a scaled image of the voltage across the motor inductive coil. Therefore, the analog voltage samples of V R can be generated by amplifying (rescaling) the voltage across the resistive shunt. V LThe analog voltage samples can be generated by amplifying (rescaling) the voltage across the inductive shunt. Thus, with appropriate amplification, Figure 13 the first and second paths of Figure 14 can be simplified in R,s That is, V L,s and V shunt and L shunt can be directly used as the signals provided by the corresponding analog-to-digital converters that sample the amplified (rescaled) voltages across R Figure 13 As shown in BEMF V Ph,s = V R,s - V L,s The BEMF estimate can be calculated. In this embodiment, the appropriate amplification factor for the inductive shunt voltage can be L / L shunt and the appropriate amplification factor for the resistive shunt voltage is R / R shunt .

[0178] Unlike Figure 13 or Figure 14 where the BEMF voltage V BEMF is calculated in the digital domain, the calculation of V BEMF can also be implemented in the analog domain, as shown in Figure 15 In Figure 15 , the mathematical operation of V BEMF = V Ph,s - V R,s - V L,s is performed in the analog domain (e.g., using an analog adder with an operational amplifier), and then it is converted to the digital domain using an analog-to-digital converter.

[0179] Due to the limited sampling frequency of the analog-to-digital conversion, the digital implementation of the BEMF estimate ( Figure 13 and Figure 14 ), i.e., the generation of the BEMF replica, may be affected by interference above the Nyquist frequency being convolved into the signal frequency range (aliasing). Subsequent operations will amplify these convolved interferences. The analog generation of the BEMF replica ( Figure 15 ) does not have this problem.

[0180] Figure 16 Shows the pair of Figure 15Circuit for further simplifying the circuit in: If the ratio between the resistor shunt and the inductor shunt is equal to the ratio between the motor resistance and the motor inductance, a single combined resistor-inductor shunt can be used. To accurately match the impedance, resistance, and stray capacitance, a physical replica of the motor coil can be used. The size of this replica can be exact or scaled. By using only one complex impedance shunt instead of a resistor shunt and an inductor shunt, fewer amplification parameters need to be correctly set to replicate the BEMF. In Figure 16 In the illustrated embodiment, the amplification factor can be R / R shunt .

[0181] Figure 17 Shows a further simplification compared to Figure 15 and Figure 16 . It is not necessary to capture the voltage across the resistor shunt and the inductor shunt or their combination. Only the voltage across the resistor shunt needs to be captured. Then, the inductive part can be emulated with a high-pass filter (such as a first-order high-pass filter), which can have a transfer function similar to that of the inductor shunt and is equal to analog differentiation. By amplifying the output signal of the high-pass filter, the scaling differences can be equalized, such that the amplified high-pass filtered signal becomes an estimate V L of the voltage across the motor coil inductance V L,s . The captured voltage across the resistor shunt can be used as an estimate V R of the motor coil resistance voltage V R,s . Therefore, the estimated voltage across the motor coil resistance and inductance can be obtained using V RL,s = V R,s + V L,s . Then, the BEMF voltage can be estimated according to V BEMF = V Ph,s - V RL,s . Omitting the inductor shunt is advantageous because the inductor shunt requires physical space and generates additional losses.

[0182] Before adding all the voltage components to calculate the BEMF, the values of L, R_shunt, and L_shunt must be known in order to select the optimal value for the amplification factor (represented as "constant" in Figure 17 ).

[0183] Obviously, the above operations shown in Figures 15 to 17 can also be implemented in the digital domain using a circuit similar to Figure 13 and Figure 14 or a portion of these circuits as appropriate.

[0184] If the amplification factor is not optimal, the phase current and the phase current derivative will crosstalk into the BEMF signal. The optimal amplification factor can be determined during factory calibration. However, temperature dependence or aging may change the value of R, L, R shunt or L shunt over time.

[0185] In a VAD system, it is not possible to stop the pump just for calibration. Therefore, the components R, L, R shunt and L shunt or the optimal amplification factor must be calibrated or re - calibrated or measured while the blood pump is operating.

[0186] Now refer to Figure 45 . In the proposed solution, an additional test signal is added to the phase current I Ph . The added signal needs to be a signal that can be separated from the normal motor operating current. At the same time, this signal should not interfere with motor operation or levitation control. The selection of this signal includes a repetitive waveform with a specific frequency, preferably significantly higher than the necessary levitation control frequency, or a pseudo - random signal with a significantly lower amplitude below the levitation control frequency.

[0187] The test signal can be added using the existing hardware of the motor driver (i.e., no further hardware modification is required).

[0188] If the amplification factor of the BEMF is not optimal, the phase current will crosstalk into the BEMF estimation. The added test signal can be separated from the BEMF estimation, for example, using frequency filtering, such as lock - in detection, or using correlation.

[0189] Based on the measurement of the crosstalk, the negative - feedback controller adjusts the amplification constant, thereby reducing the crosstalk. The two necessary amplification factors can be distinguished by the phase of the crosstalk. The crosstalk signal regarding the amplification factor related to R / R shunt has no phase shift relative to the phase current, and the crosstalk signal regarding the amplification factor related to L / L shunt has a 90 - degree phase shift for an ideal inductor and a smaller phase shift for a real inductor.

[0190] The proposed amplification - factor control method does not require a physical inductive shunt. It can also be used to adjust the amplification factor in the BEMF estimation method as shown in Figure 13 (the amplification factors L and R in the digital domain).

[0191] The estimation according to formulas (1) to (5) is only valid for slow movement in the z - direction. The z - axis of magnetic destabilization can reach a speed such that the change in the magnetic field also induces a significant voltage in the motor windings.

[0192] This results in an incorrect estimation of the rotor angle and the z-axis distance when evaluating the traditional BEMF vector.

[0193] To detect rapid movement in the z-direction, it is advantageous to know the actual magnetic field strength within the coil windings of the motor phases.

[0194] Due to the relationship between BEMF and magnetic flux: (6)

[0196]

[0197] The magnetic flux can be calculated from the BEMF: (7)

[0199]

[0200] Φ is the magnetic flux, N is the number of windings, A is the coil area, and B is the B-field (magnetic flux density) of the coil.

[0201] Due to the unknown integration constant, the integration introduces a deviation. Therefore, only the high-pass characteristic of the magnetic flux can be determined with high accuracy. However, the low-pass characteristic of the z-axis movement can be directly extracted from the BEMF measurement.

[0202] To obtain the high-pass characteristic of the magnetic flux, the magnetic flux calculated by formula (7) can be filtered with a high-pass filter. Additionally, to obtain the low-pass characteristic of the position signal, the estimated BEMF can be filtered with a low-pass filter and combined additively with the high-pass characteristic of the magnetic flux. Through this combination, a fast and accurate rotor position estimation can be obtained. As an option, the low-pass and high-pass filters can be designed to match each other, i.e., form a matching pair. As an example, the low-pass and high-pass filters can be complementary filters whose transfer functions add up to a constant, also known as a complementary filter pair. The integration and filtering using mutually matching filters can be implemented in both the analog domain and the digital domain.

[0203] Figure 18 A flowchart and a possible conversion from the analog domain to the digital domain are shown. As an example, Figure 18 The circuit in Figure 16 shows the estimation of the BEMF voltage V BEMF,s using the method according to Figures 13 to 17 The BEMF voltage can also be determined by another method, such as any of the methods described above or the method depicted in Figure 18(Exemplary) can be implemented by a first signal path including an integrator and a high-pass filter, and a second path including a low-pass filter. The filter parameters of the high-pass filter and the low-pass filter can be matched such that the high-pass filter and the low-pass filter are a pair of matched filters. The output signals of the high-pass filter and the low-pass filter are added together to obtain an estimate B of the coil B-field s .

[0204] The filtering method described combines the advantages of a BEMF sensor (i.e., an integrator and an output signal without integrator drift) and the advantages of B-field calculation (i.e., a good low-noise distance signal even for fast axial movement).

[0205] The position of the rotor can be directly calculated from the magnetic field strength using Equation (8): (8)

[0207] B(x, z) = B0e iωt e -kz

[0208] Now referring further to Figure 45 . To be able to accurately replicate the BEMF, the motor parameters coil inductance L and winding resistance R need to be known precisely. They can be determined during factory calibration. However, their values may change afterwards, e.g., depending on temperature or due to aging. Therefore, it is necessary to redetermine and / or track these parameters before implanting the blood pump and / or during operation of the blood pump.

[0209] The method described subsequently determines the estimation error of the motor parameters during operation or at rest and uses a control loop (servo loop) to adjust the motor parameters. For this purpose, the BEMF and the magnetic flux are continuously determined according to Equation (1) and Equation (7), respectively. In addition, a test signal can be injected (see Figure 45 ). The test signal can include frequency components that are high enough relative to the rotor moment of inertia to prevent or at least strongly attenuate the reaction of the rotor to the test signal due to the magnetic field induced by the current corresponding to the test signal. The test signal can be generated, for example, by Figure 10 a function generator 16. In Figure 45In this case, the test signal is fed into the motor coil via a resistor shunt and into a test signal processing block, which includes a detector, a correlator, or a filter. The BEMF calculation that can implement the above method or apparatus for BEMF estimation is fed with the voltage values across the motor coil and the resistor shunt. The test signal processing block is configured to estimate the accuracy of the motor parameter (i.e., the impedance of the motor coil) from the amplitude of the test signal within the estimated BEMF, for example, by way of correlation. The principle of this method is that a poor estimate of the motor parameter corresponds to a high correlation value, while a good estimate corresponds to a low correlation value between the estimated BEMF and the test signal. The test signal processing block provides information to the controller on how to adjust the motor parameter estimate. Thus, the accuracy of the motor parameter estimate is continuously adjusted such that the BEMF estimate is continuously adapted, with the aim of improving the accuracy of the BEMF estimate, which may vary with the operating conditions of the blood pump.

[0210] Since the rotor may not follow the test signal, the induced BEMF ( Figure 45 the B-BEMF in ) does not contain spectral components within the test signal frequency range. Thus, if the BEMF replica is estimated with the correct motor parameters (such as the impedance of the motor coil), the estimated BEMF replica will also not contain spectral components of the test signal, or will contain a known reduced amount of spectral components. Although in this case, V L 、V R and V Ph will include components of the test signal, these components will be canceled to zero when calculating the BEMF replica. However, in the case of inaccurate motor parameters, these components of the test signal will not be completely canceled, so that the components of the test signal will remain in the estimated BEMF replica. The effect of the test signal on V L and V R ( Figure 45 ) is different from the inductance phase deviation. This phase deviation still exists in the estimated BEMF replica. The corresponding detector (such as a correlator or a filter) can detect from the BEMF replica an error signal indicating the parameter L and an error signal indicating the parameter R. Each error signal is passed to the controller, which is configured to adjust the respective motor parameters to reduce the respective error signals. The controller can be implemented as, for example, an I controller.

[0211] The test signal needs to prevent the rotor from following the test current and allow the test signal to be well detected from the BEMF copy and well distinguished from external interference. Possible test signals are, for example, sine wave signals above the motor operating frequency. To better distinguish the BEMF copy from external interference, they can be amplitude modulated, phase modulated, or frequency modulated. Combinations of multiple frequencies are also conceivable. Alternatively, random and / or pseudo-random signals can be used for modulation (e.g., Gold codes, maximum length sequences), as long as they include sufficiently high frequency components. Such signals can be better filtered out from the BEMF copy using a correlator or a synchronous detector.

[0212] If the frequency range of the test signal overlaps with the frequency range of the rotor movement, then a mechanical model can be used to estimate and compensate for the BEMF copy caused by the test signal. The mechanical model includes model parameters, such as, the rotor mass, moment of inertia, spring constant, and / or friction coefficient. These parameters can be stored in the control unit and can be verified or corrected during the startup process or another dynamic process.

[0213] The advantages of this method are:

[0214] - Calibration can be performed during operation;

[0215] - Calibration can be performed without rotor movement;

[0216] - Continuous compensation for drift (shunt, amplifier, current sensor, analog-to-digital converter, etc.);

[0217] - Temperature compensation for, for example, coil resistance, drive line resistance, plug resistance, shunt resistance, shunt amplifier, analog-to-digital converter;

[0218] - No additional hardware is required;

[0219] - Detection of wear / corrosion based on parameter drift, such as wear / corrosion of drive lines, plugs, or coil insulation layers.

[0220] When integrating the BEMF to obtain the B-field, an integration error ( Figure 46a )) may occur, for example, resulting in a deviation (drift) that may increase over time.

[0221] Using the pair of matched high-pass and low-pass filters mentioned above, this error can be minimized. However, the fact that the B-field oscillates around zero over a long period of time can be utilized. Therefore, any integration error can be filtered out from the output using a high-pass filter.

[0222] This method is itself numerically unstable. The high-pass filter behind the integrator can only operate if the integration constant is within a certain range. However, without further measures, the integration constant can increase indefinitely. Instead, a low-pass filter is typically used to determine the integration error, which is then used in a self-regulating servo loop( Figure 46b ). The servo loop prevents the accumulator within the integrator from increasing to a value much larger than the input sample, as increasing to a value much larger than the input sample would lead to large rounding and integration errors. Thus, in Figure 46b , the integrator output is fed into the low-pass, and the low-pass output is fed back to the integrator input, where this low-pass output is subtracted from the BEMF, which is also the input signal to the integrator. The integrator output is an estimate of the B-field.

[0223] According to Figure 46b ) the disadvantage of the servo loop is that the low-pass filter must have a high enough cut-off frequency to be able to follow the integration error. However, this may limit the bandwidth of the B-field estimate at the integrator output, i.e., the estimated B-field at the integrator output may not follow the actual B-field in the motor coil. To increase the bandwidth of the estimated B-field (i.e., improve the dynamic behavior of the servo loop), another property of the BEMF can be used to estimate the integration error more quickly.

[0224] Thus, as another specific property of the BEMF signal, it can be observed that: the integral of the BEMF is zero not only when calculated over time but also when calculated over the rotation angle of an entire electrical and / or mechanical rotation. A moving average filter can be used to determine the average integrated BEMF of the electrical and / or mechanical rotation. The output of the moving average filter is an approximation of the integration error with low latency. The subsequent low-pass filter can have a cut-off frequency much lower than the low-pass filter in Figure 46b , which results in Figure 46c a better dynamic response of the B-field in

[0225] When floating-point arithmetic is used in combination with an accumulator, instability may also occur in the moving average filter in some computationally efficient implementations. A numerically stable implementation of the moving average filter is to use fixed-point arithmetic or to reset the accumulator periodically or even at each time step.

[0226] Referring again to Figure 10 . An additional or alternative measurement method for BEMF duplication can be used to determine the rotor position. The following method is also capable of achieving detaching at rest, where detaching means moving the rotor, for example, from an eccentric position within a blood pump to a central position. Now regarding Figure 10Describe the rotor position estimation 7 based on signals provided by the impedance analyzer 14, waveform sampling 5, and filter 6. The rotor position estimated using the method described subsequently can also be used as an input signal to the levitation controller 8. The impedance analyzer 14 is fed with voltage and / or current signals corresponding to the respective signals supplied to the actuator coils 3. These signals can include components generated by the function generator 16, such as high-frequency signals and / or test signals including modulated high-frequency signals.

[0227] In a first method, the high-frequency impedance of the actuator coil 3 (which can be a motor coil) is affected by the rotor position. The actuator coil 3 has inductance, but also has a certain amount of capacitance between the turns. This results in an RLC circuit forming a resonant circuit with a resonant frequency. The resonant frequency is typically in the range of several MHz and is much higher than the common PWM frequency. Due to magnetic coupling, the rotor position affects the resonant frequency and the quality factor (corresponding to the 3dB bandwidth of the resonance). Any high-frequency magnetic field generated by the actuator coil 3 can induce eddy currents in the conductive parts of the rotor 4. The induced eddy currents generate their own magnetic field, which opposes the magnetic field generated by the actuator coil 3, thereby changing their impedance. This mechanism is generally described as an eddy current sensor and is prior art when used with a dedicated sensing coil. However, in some aspects, the actuator coil 3 is used as an eddy current sensor.

[0228] Coreless and / or yoke-less motor windings are particularly suitable for this application because eddy currents would be introduced into the core and yoke. This effect may reduce the signal-to-noise ratio.

[0229] The eddy current sensor detects the presence of a conductive target in a high-frequency coil field. Generally, motor coils are not suitable for eddy current measurements because they mainly detect the presence of the iron core. On the other hand, it has been found that ironless (without stator core and without yoke) motors do not have this problem but still have sufficient efficiency to drive a pump. In this case, the windings of the ironless motor can be used to obtain eddy current readings. The target can be a copper plate in the rotor, the titanium housing of the rotor, or a magnetically conductive material.

[0230] The negative effects of the iron core can be reduced by laminating the iron core from several isolated thin metal sheets or by using a sintered ferrite core. In operation, high-frequency currents are imprinted on the motor coil currents. The motor driver and the eddy current sensing circuit can be isolated from each other using passive filters. A large gap between the eddy current frequency and the motor operating frequency or the PWM frequency is beneficial for filter design.

[0231] Active filters (such as a lock-in amplifier) can also be used for the extraction of eddy current signals. One of the largest sources of noise is the harmonic components of the PWM signal. Motor drivers without or with reduced switching noise are particularly suitable for simultaneous eddy current measurements. These drivers will be further explained below.

[0232] BEMF sensing cannot be used at very low speeds. This means that below a certain speed, the axial rotor position cannot be measured using BEMF. A magnetic levitation pump that uses only BEMF to measure the rotor angle and axial position needs to spin the rotor before takeoff. A good enough backup bearing is necessary to support the rotor when the magnetic bearings do not support the rotor, to allow this type of operation.

[0233] Advantageously, eddy current sensors do not depend on the rotation of the rotor and produce a signal even at 0 rpm. The rotational rotor position can be measured by comparing multiple eddy current sensors in multiple motor phases ( Figure 19 ). Rotation detection can even be implemented in a shrouded rotor. A dedicated copper target can be placed on top of some magnets to make them better targets. Alternatively or in addition, the conductivity of some magnets can be reduced by segmenting and isolating the magnets. This method is overused in high-performance BLDC motors to reduce eddy current losses.

[0234] To improve the signal-to-noise ratio, commutation of a brushless DC motor can be operated in a six-step mode, where one of the phases does not carry motor current. In each step, two phases are energized by the motor driver and the third phase is floating. This phase (i.e., the motor coil of this phase) is then eddy current measured by imprinting a high-frequency current. Only the measurement current flows in the third phase. The benefit of this is that there is no need to filter the measurement current of the eddy current sensor from the motor current. Many eddy current sensing circuits that usually require dedicated sensor coils can now be used with the motor coils while also operating the motor. Examples of these sensing circuits are balanced impedance bridges or resonant circuits.

[0235] Figure 12a An electrical equivalent circuit of a brushless DC (BLDC) motor is shown, which includes a DC power supply and a commutator, and three substantially equal phases A, B, and C, where each phase includes a motor coil as depicted in Figure 11 and described above, having an inductance L, a resistance R, and a back electromotive force (BEMF) voltage source. Each of the phases A, B, and C can respectively include a phase current I A , I B and I C . One end of each motor coil is connected to the commutator, and the other ends of the motor coils are electrically connected to each other (Y connection). The commutator is powered by a DC voltage source having a voltage V DC and a current I DC .

[0236] Similar to Figure 12a , Figure 12bshows the electrical equivalent circuit of a brushless DC motor, where the parasitic capacitance is in parallel with the series connection of the resistor, inductor, and BEMF voltage source (i.e., the motor coil) in each motor phase (phase), and the current I A In phase A, the current I B In phase B, the current I C In phase C. The resistor, inductor, BEMF voltage source, and parasitic capacitance can be considered as the equivalent circuit of the motor coil of one phase, where the resistor, inductor, and capacitor form a resonant circuit with a resonant frequency. In addition to the parasitic capacitance, an additional capacitor in parallel with the parasitic capacitance of each phase can increase the capacitance present in each phase, thereby reducing the resonant frequency of the corresponding resonant circuit. The benefit of reducing the resonant frequency is that the resonant frequency moves to a frequency range where resonant detection and processing can be carried out more effectively.

[0237] Furthermore, by selecting different additional capacitors in parallel with the parasitic capacitances of different phases, the resonant frequency of the resonant circuit in one phase may be different from the resonant frequency of the resonant circuit in another phase, so that a specific resonant frequency can be assigned to a specific motor phase. Since each motor phase can be implemented with a motor coil, and the spatial position of each motor coil is known according to the motor design, the resonance of each resonant circuit corresponds to a spatial position. Therefore, by observing the resonance, translational and / or rotational changes in the rotor position can be detected because the rotor position may affect the inductance and / or parasitic capacitance of the motor coil forming the resonant circuit. Due to eddy currents, the position of the rotor may also affect the losses and thus the effective resistance of the resonant circuit.

[0238] Each of the aforementioned resonant circuits has an impedance, which is generally complex and can be interpreted as the impedance of an eddy current sensor formed by the parallel connection of the motor coil and the capacitor. In Figure 12b two of the three motor phases are connected in series, such as phase A and B, phase A and C, or phase B and C. Due to this series connection, the impedances of the resonant circuits are added.

[0239] In addition, the impedance of the resonant circuit exhibits a local maximum at the resonant frequency. Figure 47 a shows a graph depicting the frequency versus impedance curve of the series connection of the resonant circuits of, for example, phase A and phase B, where there are no additional parallel capacitors in these phases. Since the motor is designed such that the values of the resistor, inductor, and parasitic capacitance are substantially the same, the two resonant circuits have the same resonant frequency and thus the maximum impedance at the same frequency. Therefore, there is only one local maximum on the frequency versus impedance curve.

[0240] Figure 47b shows a curve similar to Figure 47 a for the case where there is an additional capacitor in parallel with the parasitic capacitance of phase A. The curve includes two local maxima: one maximum at the same frequency as Figure 47 a, i.e., at the resonance frequency of phase B without an additional capacitor (resonance of the unmodified coil), and for the resonance related to phase A, there is an additional capacitor in parallel with the parasitic capacitance. Since the total capacitance in phase A is larger, the resonance frequency of the resonance circuit in phase A is less than the resonance frequency of the resonance circuit in phase B.

[0241] Figure 47 c shows a curve similar to Figure 47 b for the case where there are additional capacitors in both phase A and phase B, where the capacitance of the additional parallel capacitor in phase A is different from the capacitance of the additional parallel capacitor in phase B. Again, two local maxima of the impedance appear, but both maxima are arranged at the resonance frequency, which is less than the resonance frequency of the phase without additional capacitors (unmodified coil). If the rotor moves, then generally, the inductance and capacitance of the resonance circuit may change. In particular, in an axial flux rotor, this effect can be seen. Therefore, the resonance frequency of one or more resonance circuits may shift to another frequency, the phase angle of the complex impedance may change at a specific frequency, and the local maximum of the resonance frequency depicted in the graph of frequency versus impedance may change, for example, become larger or smaller.

[0242] The specific change in impedance after the rotor moves may depend on the type of the movement itself and may depend on the design of the rotor and / or stator. For example, in a rotationally symmetric design of an axial flux rotor, such as the rotationally symmetric design of an eddy current target, when the rotor moves axially, all the impedances may change in the same way. This effect is depicted by the dashed line in Figure 47 c, where the values of the resonance peaks change in the same way.

[0243] If the rotor is tilted, the impedances of the spatially opposite resonance circuits may change in opposite directions. Additionally, in the case of an asymmetric rotor rotation (e.g., as depicted in Figure 19 ) or in the case of an asymmetric application of copper on the rotor, the angular position of the rotor can be detected by evaluating the impedance of the resonance circuit. Figure 47 d is similar to the curve depicted in Figure 47 c, but for the latter two cases, it shows a dashed line curve, which is an embodiment of the impedance change when the rotor is tilted or the rotor rotates asymmetrically. In this case, the values of the resonance peaks may change differently. For example, the value of one resonance peak may increase, and the value of one resonance peak may decrease.

[0244] In addition to isolating the eddy current sensor signal from the PWM harmonics, one of the PWM harmonic components can be used to excite the motor coil resonance. Here, high-frequency PWM with short switching times is actually advantageous. High-frequency PWM motor drives are also the best choice for ironless BLDC motors. The low inductance of ironless motors requires additional inductance in the motor drive. In high-frequency motor drives, these inductances are smaller or can be completely omitted.

[0245] The resonance signal can be extracted from the motor signal using a phase-locked filter, which is directly or indirectly timed by the same clock source that also times the PWM period.

[0246] Highly integrated motor drive integrated circuits achieve high switching speeds in a small package, enabling a small control unit.

[0247] The resonance of the eddy current sensor can be characterized by the resonance frequency and the resonance quality factor. In most cases, the frequency is evaluated. However, in some applications, the quality of the resonance may be more sensitive to rotor movement. The switching instants of the PWM period can usually be approximated as Dirac pulses. These Dirac pulses excite the resonance frequency of the eddy current sensor. The resonance then decays until the next switching instant. The quality factor is directly related to the decay time.

[0248] To amplify the excitation, the PWM switching can include a series of transients instead of a single transient. This can also shape the harmonic spectrum to concentrate the energy of the PWM harmonics close to the eddy current resonance.

[0249] By using the PWM frequency, its harmonics, or a PWM switching pulse train, the coil resonance can be excited without an additional amplifier, or without using filter elements to isolate the amplifier from the motor drive, thus reducing the size of the VAD control unit and improving its reliability.

[0250] Additional or alternative measurement methods using BEMF replication can be used. The following methods can also be carried out during standstill disengagement.

[0251] In a second method, the impedance of the actuator coil 3 (i.e., the motor coil) can be additionally or alternatively modified by a coil impedance modulator 13. The coil impedance modulator 13 can be implemented as: a magneto-sensitive capacitor connected to the actuator coil 3; a magnetic saturation component that saturates due to the magnetic field generated by the rotor magnet, thereby changing the magnetic resistance of the magnetic flux circuit passing through the actuator coil 3; or a dedicated coil or an active electronic circuit using a magnetic field sensor.

[0252] If a magnetic saturation component is used inside the stator, an additional force acts on the rotor. Then, a compromise must be made between a high sensor signal and a low rotor force by using the optimal material and amount of magnetic saturation material. High-frequency current must be imprinted on the phase current to measure the phase impedance at that frequency. The commonality of all methods is that they modulate the impedance of the actuator coil 3, so there is no dedicated wire for reading out its signal other than the motor wire.

[0253] The impedance of the actuator coil 3 is measured in the control unit with an impedance analyzer 14. The impedance analyzer excites a resonance or another high frequency inside the actuator coil 3 and observes the phase current or terminal voltage to determine the impedance. This method is also limited by noise and benefits from a low-noise motor driver and filtering of switching noise. Alternatively, the switching action of the motor driver can excite the actuator coil 3 with the harmonic components of RF current or PWM. The impedance signal is fed into another rotor position estimator 7 and also provides a rotor position signal to the suspension controller.

[0254] A third method of rotor position measurement (not shown) uses a dedicated sensor (such as but not limited to a Hall effect sensor, eddy current sensor, fluxgate sensor, or ultrasonic sensor) to measure the rotor position. The sensor signal is imprinted on the drive line signal. The signal can be frequency, amplitude, or code modulated before being imprinted on the drive line. A radio frequency receiver similar to the impedance analyzer detects the signal inside the control unit to provide a rotor position signal. In one embodiment, the actuator coil 3 (motor coil) is used to oscillate the rotor in the axial, radial, rotational, or tilting direction at an audible or ultrasonic frequency. A microphone or the actuator coil 3 picks up the sound originating from the rotor. The time delay or phase shift between the acoustic transmitter and receiver can be used to measure the rotor position or the volumetric flow rate through the VAD.

[0255] The suspension controller 8 uses one or more available rotor position signals to close the suspension control loop. Based on the motor and suspension operation, different weightings or thrusts can be applied to the position measurement. To assist with the weighting, the rotor position estimator 7 can optionally prove a signal quality index. If any signal deviates from another signal or an analog pump model, an alarm or logging can be optionally triggered.

[0256] Figure 10 The VAD system in Figure 9 and Figure 10 can be implemented by using any combination of the boxes in

[0257] The sensorless motor VAD described in this application can also be used in combination with a dedicated bearing coil to isolate sensor and actuator signals from each other or to improve the efficiency of suspension control.

[0258] Figure 19 The arrangement of the motor coils is shown, which can be used to measure the axial and / or rotational position of the rotor using eddy current measurement with the motor coils. The coils are arranged in a circular pattern in a plane. In Figure 19 , only 3 of the 12 coils are depicted, and 3 coils can be considered to be arranged on a quarter circle. Figure 19 The impeller of the blood pump is also depicted. The impeller can rotate in a plane parallel to the plane of the coils. During rotation, the inductance of each coil may change with the non-uniform structure of the impeller, as Figure 19 depicted. Thus, the angular position of the rotor can be estimated based on the coil impedance. The distance between the impeller rotation plane and the coil plane also determines the coil impedance. Therefore, the coil impedance can also be used to estimate the axial impeller position, which is a function of the distance between the impeller rotation plane and the coil plane.

[0259] At high frequencies, the stray capacitance of the motor winding forms a resonant circuit (resonator) with the winding inductance. Most commonly, the eddy current sensor operates at a frequency close to the self-resonant frequency of the sensing coil. Eddy currents modify the inductance value, thus changing the resonant frequency ( Figure 20 , right frame: rotor). Another option to modify the resonant frequency is to tune the stray capacitance. Due to the high frequency, only a very small capacitance is required to significantly shift the resonant frequency by adding a tuning capacitor ( Figure 20 , left frame: stator).

[0260] The tuning capacitor will be constructed such that the capacitance changes according to the rotor position. The magnetic field strength or BEMF can be measured. Possible implementations include capacitors filled with magnetodielectrics ( Figure 21 , left).

[0261] Alternatively, when a ferrofluid moves under the contacts, the tuning capacitor can be partially filled with ferrofluid to change its dielectric constant ( Figure 21 , center left).

[0262] Alternatively, particles can be suspended in an emulsion between the capacitor plates ( Figure 21 , center right). These particles are magnetic, so they form chains along the magnetic field lines. These chains oriented parallel and perpendicular to the capacitor plates result in different complex impedances of the capacitor. Thus, the orientation of the magnetic field can be sensed.

[0263] Due to magnetic drag, the geometry of the tuning capacitor can be modified by the magnetic field instead of changing the dielectric ( Figure 21 , right).

[0264] With more available space, any electronic sensor can be used to tune the varicap with its output signal. Then the capacitance of the varicap can tune the resonant frequency ( Figure 22 ). Possible sensors include magnetic field sensors, capacitive distance sensors, radio frequency (RF) transceivers, ultrasonic transceivers or dedicated BEMF sensing coils, as well as flow or pressure sensors. The active sensor circuit can be parasitically powered by the motor line and does not require additional drive line leads. The signal of an electric field sensor or another rotor distance sensor can be read out by measuring the motor coil impedance.

[0265] A tuning network as shown in Figure 22 is required to convert the electrical sensor signal into an impedance adjustment. The tuning network can be composed of a variable capacitance diode (as shown in Figure 22 ), a metal-insulator-semiconductor capacitor with voltage-dependent capacitance, a transistor that couples additional impedance to an LCR tank circuit, or an active circuit made of an operational amplifier that emulates an additional inductor or capacitor coupled to the LCR tank circuit.

[0266] The coupling mechanism mentioned can also be used to modify an LCR tank circuit whose inductance is not the motor winding. A dedicated LCR series resonant tank circuit can also be used to amplify the effect of the tuning network. Preferably, the series resonant frequency of the LCR tank circuit is close to or significantly higher than the parallel resonant frequency of the motor coil tank circuit.

[0267] The harmonic components of the PWM are placed at n*f_PWM, where n {1,2,3,4,...}. This means that the switching noise is confined to certain regions of the spectrum. The noise level between these regions may be much lower.

[0268] The narrowband RF sensor signal can be placed in such a low-noise frequency range. However, the input filter of the sensor needs to be able to suppress adjacent PWM peaks (see Figure 29 ). This frequency placement increases the requirements for the frequency stability of the PWM and the sensor signal.

[0269] Figure 30 Shows the origin of these requirements. If the PWM jitter at the fundamental PWM frequency (PWM period) is a certain amount, the jitter at the nth harmonic is n times wider. At high frequencies, the PWM spectrum merges into a continuous frequency band without gaps. This is because the distance between the harmonics is constant, but as the frequency increases, the jitter becomes wider and overlaps at a certain point. At this point, it is impossible to extract the sensor signal without also picking up the switching noise.

[0270] Jitter is a mathematical way to describe frequency variations. Even with jitter, at any instant, the PWM spectrum consists of a single frequency peak and is similar to the spectrum in Figure 29 . Only after averaging over time does the spectrum appear as shown in Figure 30 . Therefore, the sensor frequency can always be placed between the PWM harmonics. However, the sensor frequency and filter characteristics must be adjustable.

[0271] The PWM frequency, sensor frequency, and switching filter can be synchronized to the same clock source. This can keep the sensor frequency always between the PWM harmonics. The sensor filter can be implemented using a phase-locked filter, which can be easily tuned by the clock source. A phase-locked loop (PLL) and a frequency divider can be used to keep the PWM and sensor frequencies at a specific ratio (see Figure 31 ). The motor driver can be synchronized with the sensor frequency instead of the PLL timing the sensor frequency. The important feature is that both are directly or indirectly timed by the same frequency generator, so they see the same jitter. This reduces the requirement for additional filters, thus saving space and improving reliability.

[0272] Now turning to the safety aspects of the VAD.

[0273] The voltage and current signals at the output of the power filter 11 are mostly sinusoidal, and the main frequency component is at the electrical rotational speed of the motor. The output of the power filter is optionally connected to the motor through a corrosion-resistant connector 15 (e.g., implemented as a direct feedback connector), which can measure the voltage across the motor without being affected by the contact resistance of the connector.

[0274] Power supply, power line communication, and motor line communication on the motor lines are ways to read and supply sensors within the VAD without adding additional wires to the drive lines. The method disclosed in patent application WO2018206754A1 can be used to transmit the signals of dedicated sensor signals from the implanted VAD to the controller. Possible sensors include but are not limited to rotor position sensors, acceleration sensors, gyroscope sensors, blood flow sensors, and blood pressure sensors.

[0275] Drive line defects and drive line infections constitute an important part of VAD treatment failure. A common method to reduce drive line defects is to use spare wires inside the drive line. If any wire breaks, the corresponding spare wire can take over. Alarms usually notify the user of the emergency state.

[0276] A common method is to double the number of wires in the drive line. Due to redundancy, a pure motor VAD like the HVAD has 6 wires instead of 3 in its drive line. However, this increases the cross-section of the drive line and thus increases the risk of drive line infection.

[0277] Other VAD system wiring. The HeartMate 3 reduces the number of drive line wires by providing direct current to the pump and placing the motor driver or bearing driver inside the pump. Using this method, the fail-safe drive line only requires one or two additional wires. The additional electronic components inside the pump increase its size and reduce the number of possible patients.

[0278] The present disclosure also relates to a fail-safe method for an external motor driver VAD system.

[0279] The disclosed fail-safe method uses a three-phase BLDC motor. The drive line contains three wires, each of which connects a motor phase to a phase leg. The motor coils of the motor phases must be connected in a star (rather than delta) configuration. The star point is typically connected inside the motor and is not accessible from the outside. An additional spare wire in the drive line connects the star point to an additional spare phase leg inside the drive line (see Figure 33 ).

[0280] The operating requirements for the motor, bearing, or sensor of the BLDC motor are that at least two motor windings must be supplied with independent current. For example, if in Figure 33 one of the coils L3 or wires R3 or switches M5 or M6 in the non-redundant VAD system fails due to an open connection, the current in L3 is zero and cannot be set from the outside. The remaining coils L1 and L2 are now connected in series and thus carry the same amount of current. Using this configuration, only an oscillating magnetic field can be generated, rather than a rotating magnetic field, and it is generally impossible to achieve efficient motor operation. It is impossible to achieve reliable motor startup with this configuration.

[0281] By utilizing the additional drive line wires R4 and the additional phase legs M7 and M8, the currents in L1 and L2 are again independent of each other. This allows a rotating magnetic field to be created and the motor to be operated in a normal two-phase manner.

[0282] If a dedicated bearing coil is required, the bearing coil can be connected between the BLDC star point and the fourth drive line wire. As shown in Figure 34 , this configuration can maintain motor operation when an open connection fault occurs in any of the wires, motor coils, or phase legs in the drive line. The main drawback is that the bearing can no longer operate independently of the motor in case of a fault. If the VAD has a complex backup bearing, this may be an option.

[0283] A major drawback of the proposed configuration is that all bearing currents pass through the motor wires, resulting in additional losses and magnetic fields. Compared to the motor current, the bearing current should be small under zero force control, thus reducing the impact of this drawback.

[0284] Using six wires, the bearing and motor can continue to operate after an open connection fault occurs in the drive line (see Figure 35 ).

[0285] When the risk to the patient from a large drive line or low bearing efficiency is greater than the risk of an open connection fault, a four-wire BLDC motor with a dedicated active bearing configuration can be used ( Figure 36 ).

[0286] A seven-wire BLDC motor with a dedicated bearing structure ( Figure 37 ) ensures that the motor can operate and the active bearing does not lose any efficiency even if an open connection fault occurs.

[0287] To accurately measure the BEMF as described above, it may be advantageous to accurately know the motor phase voltage. Between the motor and the control unit where the voltage can be measured, there are drive lines and connectors. In particular, the connectors are subject to gradual and sudden resistance changes. The movement of the contacts suddenly changes the contact points and contact pressures, thus changing the contact resistance. The contacts in the connectors are also more exposed to the environment compared to the wires in the drive line. This causes corrosion, which gradually changes the contact resistance and increases the sudden resistance changes due to non-uniform corrosion.

[0288] The proposed solution uses a connector with additional contacts to measure the phase voltage without being affected by the connector resistance (see Figure 32 ). The number of wires in the drive line is not affected by the direct feedback connector.

[0289] One of the largest contributions to noise stems from the switching events in the motor driver. The switching is clearly visible in the voltage and current waveforms (e.g., see Figure 7 ).

[0290] Therefore, the switching noise from the motor driver can be reduced or eliminated at the source by changing the motor driver (isolating the motor driver from the measurement hardware using a filter), or the measurement hardware or method can be made insensitive to the switching noise. The main noise source in the prior art VAD motor driver ( Figure 9 ) is the switched-mode motor driver 2.

[0291] As shown in Figure 10 , using a non-switching motor driver 10 can significantly reduce the switching noise. This driver uses a class-AB power stage instead of a class-D PWM power stage, even though the class-AB efficiency is lower than that of the PWM power stage. The class-AB efficiency is at most 78.5% at peak power output and decreases as the output power decreases.

[0292] Peak efficiency can be achieved at any operating point by inserting an optional power modulator 9 between the power supply 1 and the switchless motor driver 10. Alternatively, three (or more) power modulators (such as tracking DC-DC converters or DC-AC converters) can be used as low-noise motor drivers to replace Figure 10 the switchless motor driver 10 in

[0293] The purpose of the motor driver is to generate the required amount of torque in the motor. Torque depends only on the rotor position and the current in the phase coils. The most common method of controlling the current is pulse width modulation (PWM). PWM rapidly switches the phase voltage between multiple voltage levels. Due to the motor inductance, the current cannot follow the rapid changes in the applied voltage and generates a triangular current waveform. The exact amount of current is controlled by the PWM duty cycle. An additional inductance in series with the motor is typically used to smooth the current waveform.

[0294] The main advantage of PWM is its energy efficiency. While a single voltage level is applied to the phase, only very small resistive losses are generated in the switching elements. Some additional losses are generated during the transients from the low phase voltage to the high phase voltage and from the high phase voltage to the low phase voltage. These losses can be reduced by keeping the transient times as short as possible.

[0295] Small transient times generate high-frequency components. These frequency components can be a problem for electromagnetic radiation (EMI) and additional sensor devices. Common measures against high radiation are passive filters that either conduct the high-frequency current components to ground or burn them off as resistive losses. A small capacitor combined with a common-mode choke is usually sufficient to make the motor driver EMI compliant.

[0296] The high-frequency components not only radiate from the drive lines but also conduct to the sensor electronics attached to the motor driver. The most common form of sensors in motor drivers is phase current sensors. They are used to generate a current control feedback loop to accurately control the phase current. In sensorless BLDC motor drivers, these current sensors are also used to estimate the rotor angle. Common rotor angle estimation methods include zero-crossing (or over-zero) detection, BEMF estimation, or model-based estimation methods.

[0297] To reduce the impact of fast switching times on current measurements, it is common to synchronize the current sampling with the PWM switching so that the time between sampling and switching is maximized or constant. Using low-pass filtering on the sensor signal to suppress the switching noise is also prior art.

[0298] However, current motor drive and current sensor concepts are not sufficient to simultaneously perform motor operation and eddy current measurements, nor can they provide high signal-to-noise ratio current or voltage measurements that can be used to levitate a pediatric VAD rotor.

[0299] In the absence of switching noise, the BEMF can be estimated more accurately. BEMF detection is typically only sufficient when the motor is operating above a certain minimum motor speed, and this BEMF detection can also be used for fast rotor position measurement as described above.

[0300] Figure 25 The effect of adding a power filter between the motor drive and the motor coil on the current and voltage waveforms is shown. Here, a passive second-order filter is used to suppress harmonic components. Compared to the setup without a power filter, the switching noise can be reduced to less than 1%. This reduces the noise seen by the sensing hardware also attached to the motor coil, thus increasing the signal-to-noise ratio.

[0301] The switching noise of the motor drive can be significantly reduced by not switching the output transistors at all. Instead, the output stage uses a class-AB topology to linearly control the current in the motor phase. The drawback of the class-AB output stage is its limited efficiency of up to 78% at full output voltage swing. If the maximum output voltage of the motor drive is much lower than the DC rail voltage, the efficiency will be much lower.

[0302] To keep the efficiency of the motor drive at an acceptable level, a DC / DC converter can be used to control the DC rail voltage just above the maximum voltage swing (see Figure 27 ).

[0303] Multiple tracking DC / DC converters can be combined into an AC inverter (see Figure 28 ). The AC inverter can directly control the motor phase current. The difference between a push-pull motor drive and an AC inverter or a tracking DC controller is that the voltage waveform at the output of the AC inverter can be much smoother than the output of a PWM motor drive.

[0304] The output switching noise of a motor drive with a "class-AB" or "tracking DC / AC" topology is reduced, thus reducing the high-frequency components in the motor phase current and voltage. The absence of high-frequency components reduces interference in motor-based measurement systems, such as BEMF or motor coil eddy current sensors. The larger inherent signal-to-noise ratio increases the tolerance of the VAD system to electromagnetic radiation or conducted interference.

[0305] As described above, a high-frequency current can be imprinted on the motor coil for eddy current measurement with the motor coil.

[0306] When a high-frequency source is connected to the drive line (see Figure 24), the RF energy will flow into the motor driver. This is due to the high impedance of the motor during motor coil resonance and the relatively low input impedance of the motor driver at the motor coil resonance frequency of a few MHz. The load on the RF source is high, but there is no RF current flowing in the motor coil. Since the motor coil resonance is not excited, the RF current does not depend on the specific impedance characteristics of the motor coil. Therefore, it is impossible to perform rotor position sensing.

[0307] To direct the RF current to the motor, the RF input impedance of the motor driver must be high, at least at the resonance frequency of the motor coil. A resonance band-stop filter and / or a low-pass filter can be used to increase the impedance at a specific frequency (see Figure 25 ). Using this method, a high filter quality can be achieved, with only two passive components per phase.

[0308] Now the RF current can excite the motor coil resonance. To accurately measure the impedance, other RF sources should not excite the resonance. However, the harmonic components of the PWM can reach several MHz. This can greatly reduce the signal-to-noise ratio of the sensitive rotor position measurement system.

[0309] To prevent the PWM voltage harmonics from being converted into harmonic motor currents, a low-pass filter can be used (see Figure 25 ). A wide range of frequencies (10 Hz - 2 kHz) must pass through this filter, while suppressing the wide spectrum of PWM harmonics. A passive low-pass filter (with multiple stages) can at least reduce the harmonic PWM components.

[0310] More filter components may be required to couple the RF source to the drive line or to prevent the voltage on the drive line from damaging the RF source.

[0311] A common method to suppress the PWM noise on the shunt current sensor is to synchronize the sampling with the PWM frequency.

[0312] The shunt signal is usually also filtered using a low-pass filter, such as a passive low-pass filter. This filter can be more effective than the filter in series with the motor phase because only the low sensing current rather than the motor phase current enters the filter.

[0313] Greater noise attenuation can be achieved using digital filtering (see Figure 26 ). The frequency components of the signal to be attenuated must be within the Nyquist frequency range, i.e., the frequency components are less than half of the sampling frequency. Therefore, the sampling frequency must be significantly higher than the PWM frequency. To effectively suppress the signal frequencies above the Nyquist frequency range, a Nyquist filter can be applied before the analog-to-digital conversion, where the Nyquist filter is configured to attenuate the frequencies above the Nyquist frequency range.

[0314] All of the measurement methods for estimating the position and orientation of the pump rotor and safety-related technical aspects related to the safe operation of the blood pump outlined in the above section apply to various alternative pump designs. The following section outlines various exemplary designs of blood pumps, namely, in particular, designs of blood pumps according to Figure 1 and Figure 38 and so on.

[0315] Figure 1 An exemplary blood pump is shown. The blood pump 100 includes a housing 102 having an axial inlet 104 and an outlet chamber 106. The outlet chamber 106 includes an outlet 108, which can be connected to a graft or tube to connect to a vascular system such as an artery. The inlet 104 can be inserted into the apex of the ventricle or can also be attached to a graft or tube attached to the vascular system.

[0316] The outlet chamber 106, which can be designed as a volute, includes a back plate 110 remote from the inlet. In the illustrated embodiment, the back plate 110 includes a central spire that extends in the axial inlet direction and houses a permanent magnet 112. The chamber 106 houses a magnetically levitable impeller 114, which includes four blades 116 (some of which are shown in cross-section to show the interior of the blades), and these blades are interconnected via webs 118. The impeller includes a plurality of permanent magnets. Each blade includes a permanent magnet 120 as a counterpart to the permanent magnet 112. This magnet system is part of a passive radial magnetic bearing. The impeller further includes optional tilting bearing magnets 122 (preferably also located in each blade), which interact with magnetic rings 124 placed on the housing to form a tilting bearing.

[0317] In addition, each blade includes a rotor magnet 126, which interacts with a motor coil 128. The motor coil is placed on the far side of the back plate and is ironless, preferably a copper winding. The power (i.e., current and voltage) within the motor coil is controlled via a control unit ( Figure 1 not shown in the figure), which is preferably placed outside the human body. The coil and the control unit are connected via a drive line 130. The drive line 130 includes four wires, and one of the wires can be used as a redundant wire. These wires are used to control the motor coil so that the impeller can rotate and the position of the rotor (at least the rotor position in the axial direction) can be measured by one of the control schemes, methods, or circuits outlined in this application. Since the blood pump does not have a separate rotor position sensor, the number of wires required within the drive line can be reduced compared to traditional blood pumps.

[0318] In other embodiments, the pump can have other sensors; however, the motor coil is used to sense and measure the position of the rotor.

[0319] Figure 38A blood pump is shown that can utilize the proposed improved bearing concept. The blood pump 50 includes an inlet 58 and at least one outlet 59. The rotor 66 is driven by an actuator that includes a rotor magnet 51 and an actuator coil 52, and additionally or alternatively includes an actuator coil (motor coil) 69 (in some embodiments, constituted by these components).

[0320] In operation, the rotor rotates mainly about the pump axis 67. The rotor 66 is magnetically levitated within the blood pump 50. Passive magnetic bearings including at least two of the magnetic components 53, 54, 55, or 56 limit the radial movement of the rotor. The radial bearings are unstable in the axial direction and can have an unstable equilibrium position. The actuator components 51, 52, or 69 can be used to control the axial position of the rotor to the unstable equilibrium position or another predetermined axial rotor position. The tilt of the rotor about the tipping point 68 is limited by one or more passive magnetic bearings including at least two of the magnetic components 60, 61, 62, 63, or 64. The actuator components 51, 52, or 69 can be alternatively or additionally used to control the tilt. The blood pump can include a central hub or apex 57 to hold the radial bearing component 54 in place. If the radial bearing does not rely on the magnetic component 54, the hub can be minimized or omitted. The transition from the hub 57 to the backplate 65 can be implemented gradually to improve fluid flow and hemodynamics. The actuator coil 52 or 69 can be positioned near the outlet, near the inlet, or at both locations.

[0321] Figure 39 In a sectional view ( Figure 39 a) and a top view of a part of the rotor ( Figure 39 b), another embodiment of the blood pump is shown. The rotor blades 201 are connected by a central rotating hub 202 that can extend to the inlet 203. The rotating hub includes a radial bearing 204, and the rotor blades include additional passive magnetic radial bearings 205. These two passive magnetic bearings stabilize the rotor in the radial and tilt directions. The unstable axial position is controlled by an actuator 206. The magnets of the actuator can be arranged in a Halbach array. The Halbach array can also be applied to the embodiments described subsequently.

[0322] According to Figure 39 The advantages of the blood pump are that the magnetic levitation of the pump rotor is achieved without the need for complex pump-integrated electronics, which is beneficial for drive lines with a very small number of wires, for example, three wires are sufficient. In addition, the blood pump is characterized by a favorably reduced design height, which is one of the most critical design parameters in pediatric VADs. Additionally, compared to the shrouded pump design, the blood pump according to Figure 39 for example promotes the washing away of thrombi from above and below the blades.

[0323] Figure 40 A blood pump 300 is shown, which has a central hub 301 and two axially separated passive magnetic bearings 302 and 303. The magnet component of the motor or actuator 304 is placed within a disk 305 or within the rotor blades 306 in the volute near the fluid outlet. The magnet disk is connected to the central hub via the pump blades 306. An active magnetic actuator 307 can also be placed at the fluid inlet and within the central hub, and the magnetic actuator can also be arranged as a Halbach array. To improve the efficiency of the pump, the motor 304 and the dedicated active magnetic bearing 307 can be used together to control the axial position of the rotor and / or the rotor speed and / or the tilting movement.

[0324] According to Figure 40 The advantages of the blood pump produced are that efficient axial force and torque generation are obtained, which is beneficial for a more robust magnetic bearing, and the robust magnetic bearing in turn results in higher bearing robustness to, for example, resist external forces and accelerations applied to the pump rotor. Additionally, the advantage of high efficiency is that it reduces the heating of the blood pumped by the pump, thereby reducing the risk of thrombus formation.

[0325] Figure 41 Shown is Figure 40 a variant of the blood pump 300 shown in

[0326] According to Figure 41 The advantages of the blood pump produced are that the magnetic levitation of the pump rotor is achieved without the need for complex pump-integrated electronics, which in turn promotes a drive line with a minimum number of wires. For example, three wires are sufficient. Additionally, since the pump motor and the rear magnetic bearing are the same component, the blood pump according to Figure 41 has a simpler design.

[0327] Figure 42 Shown is Figure 1 or Figure 38 a variant of the blood pump in

[0328] According to Figure 42 The advantages of the blood pump produced are that the magnetic levitation of the pump rotor is achieved without the need for complex pump-integrated electronics, which promotes a drive line with a minimum number of wires. For example, three wires are sufficient. Additionally, by using according toFigure 42 The blood pump has obtained a flexible inlet geometry, which provides multiple implantation positions for this blood pump.

[0329] Figure 43 The blood pump in [reference] shows Figure 42 A variant of the blood pump shown in [reference]. The motor magnet 601 is located in a disc 602, and the bearing magnet 603 is located in another disc 604. The magnetic field of the bearing magnet 603 can be used for an active axial bearing. The motor 605 and the active axial magnetic bearing 606 can share the load of rotor axial position control to increase the maximum force and / or efficiency.

[0330] Similar to the blood pump according to Figure 40 The advantage of the blood pump according to Figure 43 is that high efficiency is obtained in the generation of axial force and torque, which is beneficial for a more robust magnetic bearing, and this robust magnetic bearing in turn results in higher bearing robustness to resist, for example, external forces and accelerations applied to the pump rotor. In addition, the advantage of high efficiency is that it reduces the heating of the blood pumped through the pump, thereby reducing the risk of thrombus formation. In addition, with the blood pump according to Figure 43 a flexible inlet geometry can be obtained, which provides multiple implantation positions for this blood pump.

[0331] Figure 44 shows the load sharing of the motor 701 and the active axial bearing 702 implemented in the shroudless rotor centrifugal pump 700. The motor 701 and the active bearing 702 can be axially separated or radially separated as shown. The blood pump according to Figure 44 generally produces similar advantages to the blood pump according to Figure 43

[0332] Figure 48 shows another embodiment of the blood pump 800. The blood pump 800 is similar in technical design to the blood pump shown in Figure 39 The passive magnetic bearing of the pump rotor 810 is realized by two axially displaced radial bearings 820, 830. Each radial bearing exhibits a radially stable and axially destabilizing effect. Combined, these two axially displaced radial bearings 820, 830 produce an inclination stabilizing effect on the pump rotor. In addition, each of these two passive radial bearings 820, 830 is composed of two radially repulsive magnetic elements 821, 822, 831, 832. In addition, the motor stator 840 is realized as an ironless axial flux motor and is arranged at the bottom of the pump housing 850.

[0333] Figure 49 The blood pump 900 in [reference] shows Figure 42 ​Variants of the blood pump shown in. Regarding this embodiment of the blood pump 900, the passive magnetic bearing of the pump rotor 910 is realized by two magnetic bearings, which are arranged on a plane at the bottom of the pump housing 920.

[0334] The first magnetic bearing located radially inward is implemented as a radial bearing 930. This bearing 930 consists of two radially repulsive magnetic elements 931, 932, which further produce a radially stabilizing effect but produce an axially and tilt-de-stabilizing effect.

[0335] In addition, the second magnetic bearing located radially outward is implemented as an axially and tilt-stable bearing 940. Here, this bearing consists of two radially attractive magnetic elements 941, 942, which produce an axially and tilt-stabilizing effect but produce a radially de-stabilizing effect.

[0336] The tilt stability of the pump rotor 910 is achieved through the radial position of the axial bearing 940 and the moment arm relative to the rotation axis of the pump rotor 910. Here, the motor stator 950 is also implemented as an ironless axial flux motor and is now arranged on the other side of the pump housing (i.e., the pump cover 960).

[0337] Figure 50 The blood pump 1000 in shows Figure 48 Variants of the blood pump shown in. Regarding this embodiment of the blood pump, the passive magnetic bearing of the pump rotor is implemented in the same way as Figure 48 the blood pump embodiment shown in.

[0338] However, regarding Figure 50 the blood pump, the axial positioning of the pump rotor is achieved solely through the active bearing winding 1010, which is arranged at the inlet of the pump. During pump operation, the axial flux motor here only generates torque to rotate the pump rotor.

[0339] Figure 51 The blood pump 1100 in shows Figure 49 Variants of the blood pump shown in. Regarding this embodiment of the blood pump, the passive magnetic bearing of the pump rotor is implemented in the same way as Figure 49 the blood pump embodiment shown in.

[0340] However, regarding Figure 51 the blood pump, the axial positioning of the pump rotor is achieved solely through the active bearing winding 1110, which is now arranged at the bottom of the pump housing. During pump operation, the axial flux motor here only generates torque to rotate the rotor of the pump.

[0341] Figure 52 shows another embodiment of the blood pump 1200. This blood pump 1200 is the same asFigure 48 The technical designs of the blood pumps shown are similar.

[0342] The passive magnetic bearing of the pump rotor 1220 is achieved by two axially displaced axial bearings 1230, 1240. Each axially displaced axial bearing 1230, 1240 exhibits a radially stabilizing and axially destabilizing effect. Combined, these two axially displaced axial bearings 1230, 1240 produce a tilting stabilizing effect on the pump rotor. In addition, each of these two passive radial bearings 1230, 1240 consists of two axially attracting magnetic elements 1231, 1232, 1241, 1242.

[0343] In addition, a Halbach configuration is achieved using passive magnetic elements associated with the pump motor 1270 and the passive magnetic bearing. The advantage of this is that the pump rotor has maximum controllability, which in turn results in higher bearing robustness, for example to withstand external forces and accelerations applied to the pump rotor.

[0344] In addition, with respect to Figure 52 the blood pump, the axial positioning of the pump rotor is achieved solely by the active bearing winding 1260, which is arranged at the inlet of the pump. During pump operation, the pump motor here only generates torque to rotate the pump rotor 1220. Alternatively or additionally, in another embodiment, the control of the axial position of the pump rotor 1220 is achieved by a combination of magnetic bearing forces and axial motor forces, which results in the highest overall axial stability of the system (i.e., the pump rotor 1220).

[0345] Thus, through the active bearing winding 1260, in combination with the Halbach magnet and optionally in combination with axial motor forces, high operating efficiency is obtained, which has the advantage of reducing the heating of the blood pumped through the pump, thereby reducing the risk of, for example, thrombus formation.

[0346] The blood pump system may also include a connection system for medical applications, the connection system including: a cannula c7 made of a flexible material; a claw ring c1, which is arranged on the cannula c7 and has at least two claws c11, wherein the claw ring c1 surrounds the outer surface of the cannula c7 and is arranged on the cannula end c71 of the cannula c7 for rotating on the cannula c7 and axially displacing to a stop, the stop including a collar on the outer surface of the cannula c7 at the cannula end c71; and a tube c5, the tube c5 including a locking ring c3 attached to the tube end and a threaded joint attached to the tube c5, wherein the claw ring c1 can engage with the locking ring c3 by the axial movement of the claw ring c1 relative to the cannula c7 towards the locking ring c3 and by locking at least two claws c11 on the locking ring c3 in a position where this axial movement is restricted by the stop.

[0347] Figures 53 to 57An exemplary device for a connection system for medical applications is shown, which provides a connection between an intubation tube c7 and a tube c5. A threaded joint (shown here as a hose coupling c4) is inserted into the intubation tube end c71 (see Figure 53 b). A reinforcing element c6 is arranged on the intubation tube end c71 and fixedly connected to the intubation tube end c71. A spacer ring c2 is arranged in a groove c61 of the reinforcing element c6. A claw ring c1 locks with its claws c11 onto a locking ring c3, where the claws c11 of the claw ring c1 are attached to a base ring c15 (see Figure 54 ). The claw ring c1 thus presses onto the spacer ring c2, and the spacer ring c2 in turn presses onto a collar c62 of the reinforcing element c6, thereby forming a tight connection with an annular end face c36 of the locking ring c3. The seal between the intubation tube c7 and the hose coupling c4 is formed by elastic expansion of the intubation tube end c71 when the intubation tube end c71 is pushed onto the hose coupling c4. This generates a radial sealing force that presses the inner surface of the intubation tube end c71 onto the hose coupling c4. A tension ring c63 of the reinforcing element c6 further increases the radial sealing force. The device is essentially a snap connection. The claws c11 arranged on the base ring c15 are pushed onto the locking ring c3, where they lock releasably.

[0348] The end face c36 transitions radially outwards to an inclined surface c31 and radially inwards to a gap c35 that receives an intubation tube shoulder that is generated when the intubation tube is shortened during implantation and is arranged in front of the collar. A retaining surface c32 and an inclined surface c33 are arranged behind the end face c36, where the claws c11 of the claw ring c1 arranged on the intubation tube snap into the surfaces c32 and c33, thereby connecting the intubation tube c7 to the tube c5. The spacer ring c2 and the claw ring c1 for connecting the intubation tube c7 can already be pre-mounted on the intubation tube c7 when the intubation tube is supplied, or alternatively can be mounted on the intubation tube c7 during the operation. For connection, the intubation tube c7 is pushed onto the hose coupling c4, thereby elastically expanding the inner diameter of the intubation tube c7. The claw ring c1 is rotated until it is positioned in the locked position relative to the locking ring c3 ( Figure 57 a), and then axially moved towards the locking ring c3. When the claws c11 are pushed onto the inclined surface c31 of the locking ring c3, the claws c11 unfold and lock with a claw support surface c12 on a retaining surface c32 of the locking ring ( Figure 56 a, Figure 56 b, Figure 56 c). In a perspective view of the claw ring c1 according to Figure 54 , six claws c11 are arranged on the base ring c15. A claw retaining surface c12 and a claw inclined surface c14 are arranged at both ends of the claw c11. As Figure 56 a, Figure 56 b, Figure 56As shown in c, these are all set to be able to engage securely with the locking ring c3. The elastic joint c13 on the claw ring c1 allows the claw c11 to expand when snap-engaging with the locking ring c3.

[0349] The blood pump system may further include means for connecting the cannula to a hollow organ (in particular to the heart), wherein the cannula tip of the cannula has an opening which is wavy at its upper edge and provided with notches in order to prevent complete occlusion and to retain the blood flow from the hollow organ into the cannula.

[0350] In Figure 58 In an exemplary embodiment of the means for connecting the cannula to the hollow organ shown, the upper edge of the cannula tip ca13 which can project into the left ventricle of the heart ca3 is wavy and provided with deep semi-circular notches. In this case, these notches prevent the opening from being completely occluded, so that blood can continue to flow from the ventricle into the inlet cannula.

[0351] The cannula can be combined with a suture ring ca1 which can be sutured at the heart ca3. For example, in order to connect the inlet cannula to the left ventricle ca3, a circular opening is first cut out at the apex of the heart, and later the cannula tip ca13 is inserted therein. Before inserting the cannula tip ca13, the suture ring ca1 is sutured around the circular opening.

[0352] The cannula can have a suture flange ca14. In one embodiment, the suture ring ca13 has the same diameter dimension as the suture flange ca14 at the inlet cannula ca2. In one embodiment, the suture ring ca1 can consist of a five-layer silicone core which consists of an un-reinforced silicone inner layer ca4 and reinforced silicone layers ca5, ca6 on both sides; Dacron flannel is adhered to both the top and the bottom.

[0353] All the pump arrangements shown ( Figure 1 、 Figures 38 to 44 、 Figures 48 to 52) are designed to achieve suspension through BEMF-based rotor position detection and rotor position detection based on motor coil impedance, are compatible with these detections, and achieve sensor redundancy without additional implanted components. By utilizing the axial force generated by the axial flux motor, dedicated bearing coils can be minimized or completely omitted, thereby further reducing the size of the pump. The disclosed pump does not require active suspension electronics within the pump, thus maintaining a small size and high reliability. Some reliability measures (e.g., fully redundant DC link of the pump) cannot be achieved without the electronics of the pump. Therefore, alternative measures are disclosed that increase redundancy and are compatible with the disclosed suspension structure. When optimizing the size of the pump, it is important to make the implanted peripheral devices as small as possible without compromising blood compatibility. Therefore, the disclosed connection system is particularly suitable for the disclosed pump arrangement.

[0354] The disclosed pump systems combine small size, reliability, and blood compatibility, making them particularly suitable for pediatric VADs.

Claims

1. A blood pump system, the blood pump system comprising: - A blood pump, the blood pump comprising: - A housing, the housing comprising an inlet and an outlet; - A motor, wherein the motor comprises a plurality of motor coils for driving an impeller; - A rotor comprising the impeller, wherein the impeller is located within the housing and comprises a plurality of rotor magnets; The blood pump system further comprises: - A drive circuit; - A control unit for controlling the operation of the blood pump, the control unit being configured to: - Operate the motor such that the impeller rotates about an axis; - Imprint a high-frequency sine signal into at least one of the plurality of motor coils; and - Measure the position of the rotor in a direction along the axis using at least one of the plurality of motor coils based on the interaction of the high-frequency sine signal with the rotor.

2. The blood pump system according to claim 1, wherein, The control unit is configured to reduce or eliminate switching noise from the motor driver.

3. The blood pump system according to claim 2, wherein The output stage of the motor driver comprises filter elements for filtering out high-frequency signals.

4. The blood pump system according to claim 3, wherein, The high-frequency sine signal is added to the filtered motor driver output.

5. The blood pump system according to claim 1, wherein, The measurement of the motor current includes the measurement of the motor coil impedance.

6. The blood pump system according to claim 1, wherein The back electromotive force inside the motor is replicated outside the motor using inductive shunt voltage measurement.

7. The blood pump system according to claim 1, wherein The magnetic field strength is replicated outside the motor in the form of an electrical signal or a digital signal using back electromotive force replication and a pair of matched high-pass filter elements and low-pass filter elements.

8. The blood pump system according to claim 1, wherein, The control unit is configured to reduce voltage transients in the drive circuit, or is configured to reduce trapezoidal or triangular current waveforms relative to a sinusoidal current waveform.

9. The blood pump system according to claim 8, wherein, The control unit comprises a DC-DC converter, or comprises a class AB amplifier and / or passive filter elements.

10. The blood pump system according to claim 1, wherein, The drive circuit comprises no more than four wires.

11. The blood pump system according to claim 1, wherein, The blood pump comprises passive magnetic radial bearings and / or passive magnetic tilting bearings.

12. The blood pump system according to claim 1, wherein, The blood pump comprises active axial magnetic bearings.

13. The blood pump system according to claim 1, wherein, The motor is an axial flux motor.

14. The blood pump system according to claim 1, the blood pump system comprising a capacitor electrically connected in parallel with the motor coil, wherein the motor coil and the capacitor form a resonant circuit, the resonant circuit having an impedance and a resonant frequency comprising an amplitude and a phase.

15. The blood pump system according to claim 14, wherein, The motor coil comprises a first coil, wherein a first capacitor is electrically connected in parallel with the first coil and forms a first resonant circuit, wherein the motor coil comprises a second coil, wherein a second capacitor is electrically connected in parallel with the second coil and forms a second resonant circuit, wherein the capacitance of the first capacitor is different from the capacitance of the second capacitor, and the resonant frequency of the first resonant circuit is different from the resonant frequency of the second resonant circuit.

16. The blood pump system according to claim 15, the blood pump system further comprising a measurement unit configured to determine the impedance of one or more of the first resonant circuit and the second resonant circuit.

17. The blood pump system according to claim 16, wherein the blood pump system further comprises an estimation unit configured to estimate the translational and / or rotational position of the rotor based on the impedance of one or more of the first resonant circuit and the second resonant circuit.

18. The blood pump system according to any one of claims 1 to 17, wherein, A test signal is fed into the motor coil, wherein the test signal comprises a component that has been modulated by at least one of amplitude modulation, frequency modulation, phase modulation, and code modulation, and wherein the component of the code modulation comprises a random code modulation component or a pseudo-random code modulation component.

19. The blood pump system according to claim 18, wherein the blood pump system further comprises a detector unit configured to detect the test signal in the voltage measured across the motor coil and / or in a signal derived from the voltage.

20. The blood pump system according to claim 19, wherein, The detector unit is configured to estimate the motor coil impedance based on the detected test signal.

21. The blood pump system according to claim 20, wherein, The motor coil impedance is continuously estimated during operation of the blood pump system.

22. The blood pump system according to claim 20, wherein, The motor internal back electromotive force replica is calculated using the estimated motor coil impedance.

23. The blood pump system according to claim 22, wherein, The estimated motor coil impedance is estimated by minimizing the high-frequency signal component in the back electromotive force replica.

24. The blood pump system according to claim 1, wherein, The magnetic field intensity is replicated outside the motor in the form of an electrical signal or a digital signal by integrating the back electromotive force replica with an integrator, wherein the integrator is numerically stabilized by feeding back the output signal of the integrator to the input of the integrator via a moving average filter that generates an average signal.

25. The blood pump system according to claim 24, wherein, The average time of the moving average filter is one rotation period of the rotor or an integer multiple of one rotation period.

26. The blood pump system according to claim 24, wherein, The back electromotive force replica is the input signal of the integrator, and wherein the average signal is subtracted from the input signal of the integrator.

27. The blood pump system according to claim 26, wherein, The average signal is low-pass filtered before being subtracted from the input signal of the integrator.

28. The blood pump system according to claim 1, wherein the blood pump system further comprises a connection system for use in a medical application, the connection system comprising: - a cannula made of a flexible material, a claw ring being arranged on the cannula and having at least two claws, wherein the claw ring surrounds the outer surface of the cannula and is arranged at the cannula end of the cannula to rotate and axially displace on the cannula to a stop, the stop comprising a collar on the outer surface of the cannula at the cannula end; and - a tube, the tube comprising a locking ring attached to a tube end and a threaded joint attached to the tube, wherein the claw ring can engage with the locking ring by axially moving the claw ring relative to the cannula towards the locking ring and locking at a position where the axial movement is restricted by the stop by the at least two claws on the locking ring.

29. The blood pump system according to claim 1, wherein the blood pump system further comprises means for connecting the cannula (ca2) to the hollow organ (ca3), characterized in that, The cannula tip (ca13) of the cannula (ca2) has an opening, the opening having a wavy upper edge and being provided with notches in order to prevent complete occlusion and retain the blood flow flowing from the hollow organ (ca3) into the cannula (ca2).

30. The blood pump system according to claim 29, wherein, The cannula (ca2) is combined with a suture ring (ca1) that can be sutured to the hollow organ (ca3).

31. The blood pump system according to claim 29 or 30, wherein, The cannula (ca2) has a suture flange (ca14).

32. The blood pump system according to claim 1, wherein, The blood pump system is a ventricular assist device.

33. The blood pump system according to claim 1, wherein The inlet is an axial inflow inlet, and the outlet is a tangential outflow outlet.

34. The blood pump system according to claim 5, wherein, The measurement of the motor coil impedance is the measurement of the high-frequency motor coil impedance.

35. The blood pump system according to claim 1, wherein, The drive line includes three wires and one redundant wire.

36. The blood pump system according to claim 13, wherein, The axial flux motor is an ironless axial flux motor.

37. The blood pump system according to claim 19, wherein, The detector unit includes a correlator or a synchronous detector.

38. The blood pump system according to claim 29, wherein, The hollow organ is the heart.

39. The blood pump system according to any one of claims 1 to 17, wherein, The rotor includes a conductive target, and wherein the control unit is further configured to measure the rotor position using eddy currents measured by detecting the conductive target through at least one of the plurality of motor coils.

Images