Rotating machinery

Abstract

A rotary machine has a stator (13) and a rotatably mounted rotor (11), wherein one or more magnetic field sensors (12) are fixedly arranged at a radial distance with respect to the stator (13) from an axis (20) fixed with respect to the stator (13), and at least one measuring device (1) is designed to detect a change in the magnetic field by means of the magnetic field sensors (12), the rotor (11) being designed to generate one or more electrical signals by means of one or more constant magnetic source voltages and one or more of the magnetic field sensors (12), respectively, which have signal components corresponding to the rotor rotation frequency (171) and the respective distance between the magnetic field sensors (12) and the rotor (11), characterized in that a demodulator unit (4) is provided, which is designed to demodulate the signals generated by the magnetic field sensors (12) or derived therefrom, so as to generate signals (140) corresponding to the distance between the rotor (11) and the magnetic field sensors (12) corresponding to the respective signals.

Description

TECHNICAL FIELD

[0001] The present application relates to a rotary machine. The present application also relates to a method of operating a rotary machine and to a system having a rotary machine. BACKGROUND

[0002] Rotary machines are characterized by having at least one part, generally referred to as a stator, which is fixedly arranged with respect to a coordinate system, and at least one rotating part, generally referred to as a rotor. Examples of rotary machines are, inter alia, electric machines, compressors, turbines. The field of application is manifold. For example, an electric machine or a turbine can be used as a drive device, which in turn drives other work machines, such as, for example, a compressor or a pump. The rotary machine can correspondingly be used as a drive device itself, but also as a driven machine. Combinations of drive machines and work machines are also possible, for example in electric machine compressors or pumps, as they are used, for example, in heart assist systems, in which an electric machine and a pump are integrated into one system. It is also possible here that the pump and the electric machine function are realized with only one rotor.

[0003] Generally, bearings, such as, for example, rolling bearings, sliding bearings or magnetic bearings, are used as intermediaries between the moving rotor and the fixedly arranged stator. Forces acting on the rotor, such as, for example, unbalance forces during rotation or external forces during use, can cause the rotor to vibrate. This vibration generally comprises, here, not only components corresponding to the rotational frequency of the rotor, but also harmonics of the rotational frequency in addition to the actual rotational frequency. However, it is also possible that the rotor vibrates at other frequencies that are not synchronized with the rotational frequency, for example at a resonance frequency, the vibration frequency of which is not only related to the rotor itself, but also to the rotor support. Whether and which resonance frequencies are excited, inter alia, also depends on the forces acting on the rotor.

[0004] In principle, this vibration is transmitted through the bearing to the stator, so that the vibration can also be detected on the stator with a vibration sensor in the form of an eddy current sensor by means of measurement technology in the form of an eddy current sensor. For long-term use, generally, limit values are set for permissible vibration values.

[0005] Generally, one is interested in this respect in keeping the amplitudes small both in the rotor and in the stator in order to suppress wear and function-affected. In particular in magnetic bearing machines, i.e. machines in which the rotor should be held in place in a theoretically defined position relative to the stator by means of magnetic forces, there is generally a large rotor mobility, which is set by the gap between the fixed stator and the rotatably mounted rotor. In order to detect the vibration level and, perhaps, also to effectively counteract, the rotor vibration is generally continuously measured, in particular with a distance sensor fixedly arranged on the stator, which measures the relative distance between the sensor and the rotor and, in turn, between the stator and the rotor, in order to derive the rotor vibration from the temporal profile of the distance.

[0006] Although this measurement method is widespread, a major disadvantage is that a complete measurement chain is required for the rotor position detection, and in addition, space must be provided for the sensor system in the mechanical structure. This measurement chain can typically comprise components such as sensors, signal processors, amplifiers and also analog-digital converters, for example. It can therefore be advantageous in some embodiments to perform the rotor position detection or additional sensor technology, i.e. the rotor position is at least sometimes performed using mechanisms that are already present in the mechanical structure. Sensorless rotor position detection is proposed in this case. Such a system is described, inter alia, in the patent documents US 9 506 475 B2 and US 8 226 373 B2. SUMMARY

[0007] The task in the rotary machine disclosed herein is to provide a rotor position detection device with which one or more components that are typically used in a measurement chain for rotor position detection can be dispensed with. This task is accomplished by a device and a method.

[0008] The rotary machine disclosed comprises, in addition to a stator, a rotatably mounted rotor which is designed to move relative to the stator. For the rotary machine, an axis fixed relative to the stator is also defined. One or more magnetic field sensors are fixed immovably relative to the stator at a radial distance from the axis. The magnetic field sensors can here expressly have the specific ability to perform measurements for rotor position detection. They can also be part of other components whose main field of application is not the performance of measurements for rotor position detection, but for example for driving the rotor. They can thus be motor coils, for example, whose main field of application is to generate a magnetic field to generate a force, thereby rotating the rotor. The motor coils must here be designed to generate a magnetic field suitable for driving the rotor, respectively. The axis fixed immovably relative to the stator can extend substantially parallel to the axis of rotation, for example, and can be used to define a theoretical rotor position, for example.

[0009] In addition, the rotary machine comprises at least one measurement device which is designed to measure the change in the magnetic field with the aid of the magnetic field sensors described above. The measurement device here also has the property of transforming the signals provided by the magnetic field sensors in the form of voltages or currents in such a way that they are suitable for subsequent processing steps. The measurement device can thus comprise impedance matching, conversion of voltages to currents or vice versa, reduction or amplification of the amplitude of the currents or voltages, or also analog-digital conversion, for example.

[0010] The rotating machine further comprises a rotor which is designed to generate one or more electrical signals with one or more constant magnetic source voltages and one or more of the above-mentioned magnetic field sensors, respectively, which have signal components corresponding to the rotor rotation frequency and the respective distance between the magnetic field sensor and the rotor. This means that a component which generates a magnetic field, such as a permanent magnet or an electromagnet, is arranged on the rotor. When rotating, this magnet generates a variable magnetic field. The above-mentioned magnetic field sensor is arranged in such a way that it encounters this variable magnetic field and generates a signal which corresponds to the strength of the magnetic field present at the respective moment in time or its temporal development and the spatial orientation. This arrangement is typical for synchronous machines, but also for brushless DC machines or axial machines, and is also common for machines in which, for example, magnets, such as permanent magnets, are arranged on the rotor. In this case, the magnetic field sensor can be formed, for example, by a machine coil which is generally arranged in the magnetic field of the magnet on the rotor. In particular, an alternating magnetic field is obtained when the rotor is rotating, which leads to an induced voltage in the machine coil.

[0011] The assembly consisting of the rotor equipped with the magnet and the magnetic field sensor can also be interpreted as an amplitude modulator. The alternating magnetic field generated by the rotation of the rotor generates an alternating voltage in the magnetic field sensor which is designed, for example, as a machine coil, which is proportional to the change in the magnetic flux in the coil in the case of a coil. The magnetic flux depends here on the current position of the magnet with respect to the respective coil on the one hand. But on the other hand, the magnetic flux also depends on the configuration of the magnetic circuit, in particular the materials used and the gap present in the magnetic circuit, which contains non-magnetic or only slightly magnetic materials. The magnetic flux is in particular related to the distance between the respective coil and the respective rotor magnet. The magnetic flux is thus modulated by the rotor position relative to the coil and accordingly also the voltage in the respective coil by the distance between the rotor and the coil. The resulting voltage signal thus has the characteristics of an amplitude modulation signal, which is characterized by a carrier oscillation and a modulation signal which modulates the carrier oscillation. The frequency of the carrier oscillation, i.e. the carrier oscillation frequency, in this case is the frequency which results from the product of the rotor rotation frequency and the even number of poles of the rotor and which modulation signal is the respective distance between the coil and the rotor. The carrier oscillation frequency is thus in a range which overlaps or is adjacent to the range of the motor rotation speed. The even number of poles relates to the number of rotor poles arranged in the vicinity of the respective magnetic field sensor and whose magnetic field, in which the magnetic field sensor induces a voltage, is formed when rotating. The even number of poles is calculated by dividing the number of poles by two.

[0012] The rotor with the described arrangement is designed to generate a distance modulation signal with its magnet and the distance to the magnetic field sensor. This is particularly meaningful in machines, such as, for example, synchronous machines or brushless DC machines or axial machines, in which the magnet is originally arranged on the rotor.

[0013] The rotary machine further comprises a demodulator unit which is designed to perform a demodulation in the case of a signal generated by the magnetic field sensor or derived therefrom having signal components corresponding to the rotational frequency of the rotor and to the respective distance between the magnetic field sensor and the rotor, so as to generate a signal corresponding to the distance between the rotor and the magnetic field sensor which is assigned to the respective signal.

[0014] The demodulator unit is a component which, for example, performs an amplitude demodulation. Amplitude demodulation of a signal generally comprises a conversion of a signal band which is defined around a carrier oscillation frequency into a range around 0 hertz. The actual conversion can be carried out, for example, using known envelope demodulation, but can also be carried out by multiplication by a sinusoidal signal having the carrier oscillation frequency and subsequent low-pass filtering. Other demodulation methods are also known, for example multiplication by a periodic signal instead of multiplication by a sinusoidal signal, wherein the periodic frequency or the harmonic corresponds to the carrier oscillation frequency. The multiplication described can be carried out, for example, using electronic hybrid circuits. In addition, the band conversion can be carried out by means of a Fourier transform. This method is preferably used in the digital range, or in the time and amplitude discrete range, since here fast and efficient algorithms are available for calculating the Fourier transform (FFT).

[0015] It is also possible for at least one magnetic field sensor to be designed as a coil in the rotary machine. If the coil, which can also be referred to as a sensor coil, is used as a magnetic field sensor, the voltage which is induced by the coil and which is measurable at the coil terminals is proportional to the change in the magnetic flux in the coil.

[0016] It is clear from the previous embodiments and should be expressly mentioned here that the rotary machine can be an electric machine.

[0017] It is also conceivable for at least one magnetic field sensor to be designed as an electric machine coil and to be designed for detecting a change in the magnetic field and for generating a magnetic field which is suitable for driving the rotor. It is possible here for a plurality of electric machine coils to have to generate a magnetic field which is suitable for driving the rotor in order to actually move the rotor. If an electrically energized electric machine coil or bearing coil is used as a magnetic field sensor, the voltage which is measurable at the coil terminals is the sum of the induced voltage due to the change in the magnetic flux in the coil, the self-induced voltage due to the inductance of the coil and the ohmic voltage drop over the resistance of the coil wire. A greater alternating magnetic flux amplitude thus leads to a proportionally greater alternating voltage component at the coil terminals. The magnetic field sensor can also be designed as a differential Hall sensor. The signal generated by the Hall sensor is proportional to the magnetic flux.

[0018] The coils that generate the magnetic field suitable for driving the rotor in the electric motor, i.e. the motor coils, can at the same time be used to detect the magnetic field variations. In this case, components already present in the electric motor can take over the additional sensor task, so that no separate sensor is required. But instead, other sensors than magnetic field sensors, for example differential Hall sensors or coils that are not primarily assigned to the driving motor coils, can also be used.

[0019] The rotating machine can also be equipped with a device that provides the rotor rotation frequency. It is possible here that the rotor rotation frequency is provided by the electric motor controller, since this information is available there anyway. But it is also possible to measure the rotor rotation frequency, for example in that one or more markings are applied to the rotor and are detected by a suitable sensor during operation. One possible example of a device for measuring the rotor rotation frequency is in particular an elongated slot as a marking and a distance sensor aligned to the rotor, i.e. a so-called keyphasor, which is set up to output a voltage signal when passing through the slot, also called keyphasor slot, which differs from the voltage signal output thereby in the phase of the rotor cycle in which the keyphasor and the keyphasor slot are not opposite. With a comparison device, the voltage signal generated by the keyphasor can be converted into short voltage pulses per rotor cycle, which can then be converted into a rotation speed signal by a processing unit, for example a counter assembly. Similar methods can also be implemented, for example, on an optical basis. The rotation speed signal can finally exist in different forms, for example as a voltage, a current, a numerical value, a pulse / cycle form or a periodic signal with a frequency corresponding to the rotor rotation frequency. In principle, the rotor rotation frequency can also be performed by frequency analysis, such as a Fourier transform. Here, the magnetic sensor signal is transformed into the frequency range and the rotational frequency is detected by means of peak recognition.

[0020] Optionally, the demodulator unit is designed to use the rotor rotation frequency in the demodulation. This is in particular the case when the demodulation in the digital range is performed by means of a computer or microcontroller using a Fourier transform, in particular a discrete Fourier transform, with its efficient implementation, the fast Fourier transform, or for example also with the Goertzel algorithm, other than envelope demodulation.

[0021] To this end, the signal to be demodulated is transformed into the frequency domain by means of a Fourier transformation. In the discrete Fourier transformation, the Fourier transform of the signal exists in the form of discrete values at frequency reference points, i.e. in a scanned state. For the demodulation, all values at frequency reference points not belonging to the carrier oscillation frequency and the amplitude modulation signal corresponding to the rotor rotation frequency are first masked out, i.e. reduced to zero. The remaining unmasked frequency reference points are shifted with the carrier oscillation frequency value towards the 0 hertz frequency. All signal portions shifted towards the region around the 0 hertz frequency are added together, and the resulting signal is transformed back into the time domain. The advantage of this approach is its simplicity. The disadvantage is the loss of the phase of the distance sensor, so that information which can be important in some embodiments for generating the control signal is lost. It is also possible to determine at least the power of the rotor vibrations. Since this power can in principle also be determined in the frequency domain according to the Parseval theorem, the rotor vibration power can also be determined directly in the frequency domain. The vibration power can be used to assess the respective current vibration level and to generate a control signal on the basis thereof, for example like an emergency shutdown.

[0022] The rotating machine can also be equipped with a device for providing a rotor rotation angle. The rotor rotation angle is the angle that occurs as a result of the rotor rotation, which is defined in the angular plane perpendicular to the rotor axis of the rotor and which results from the current position of a reference point on the rotor, the reference position with respect to the stator and the penetration point of the rotor rotation axis through the angular plane as the vertex of the rotor rotation angle. As a reference point, for example, a key phase slot can be employed, which is also used for the rotational speed determination. The position of the key phase is suitable, for example, as the reference position with respect to the stator. The current rotor rotation angle is ultimately determined by means of the current rotational speed, where:

[0023] rotor rotation angle = (rotor rotation frequency * dt * 360 mod 360)°,

[0024] where dt is the elapsed time since the reference point of the rotor last passed the reference position with respect to the stator. This time can be measured, for example, with a clock or counter component.

[0025] With the known rotor rotation angle, the aforementioned demodulation method can be supplemented by a phase correction in the frequency domain. The phase correction in this case means adding a deviation related to the rotor rotation angle to the phase angle of each Fourier coefficient, so that the phase of the signal corresponds to the phase of the rotor vibration signal when transformed back into the time domain.

[0026] Furthermore, the rotor rotation angle can be used to perform signal demodulation by means of an electronic circuit by multiplication with a sinusoidal oscillation. For this purpose, for example, an oscillator is used to generate a frequency which corresponds to the carrier oscillation frequency and the phase of the generated frequency is adjusted, for example, with a phase-locked loop, such that it corresponds to the phase of the carrier oscillation. The phase can be taken directly from the carrier oscillation phase here or, alternatively, can also be calculated from the rotor rotation angle.

[0027] Optionally or additionally, the rotating machine comprises a first processing unit which is designed to superimpose and / or filter one or more electrical signals of the aforementioned magnetic field sensors into one or more signals such that a signal portion of the respective resulting signal which contains information about the distance between the rotor and the respective magnetic field sensor is amplified in comparison to other signal portions.

[0028] The first processing unit can be designed as an analog circuit and can comprise, for example, an addition circuit or a subtraction circuit or a filter circuit which can be implemented, for example, with an operational amplifier. However, the first processing unit can also be designed partly or completely digitally, for example, based on one or more microcontrollers, processors, special circuits or also field-programmable gate arrays or alternatively digital computing units with discrete components.

[0029] In the case of a signal superimposition of the signals measured by the magnetic field sensors possibly within the first processing unit, for example, a plurality of measurement signals can be added with the correct phase such that signal components which do not contain rotor position information are eliminated. This possibility can advantageously be used in some embodiments when the rotating machine is, for example, an electric motor and the magnetic field portion associated with the drive represents a considerable portion of the measured magnetic field power or when signal components which are independent of the rotor axis position are superimposed on the rotor position signal.

[0030] If signal filtering of the signals measured by the magnetic field sensors can be carried out in the first processing unit, it can be selected in some embodiments, for example, to advantageously use a frequency-selective filter, for example, a low-pass filter, such that all signal components which lie outside a frequency band containing rotor position information in the measurement signals are suppressed, i.e. damped. Such signal portions to be suppressed can be, for example, signal portions originating from the motor controller. The motor control signals can be, for example, pulse-width-modulated, wherein the switching frequency of the pulse-width-modulation can be, for example, several kilohertz. By means of the low-pass filter, the switching frequency and also its harmonics are suppressed.

[0031] Furthermore, the rotary machine can also comprise a second evaluation unit which is arranged downstream of the demodulator and is designed to generate one or more rotor position signals from the demodulated signals. The demodulated signals comprise signals which correspond to the distance between the rotor and the respective magnetic field sensor. Since the magnetic flux variation is proportional to the magnetic flux alternating frequency, the amplitude of the distance signal corresponds not only to the distance between the magnetic field sensor and the rotor, but also to the carrier oscillation frequency, which in turn is associated with the rotor rotation frequency. In order to eliminate this correlation from the signal, a rotor speed-dependent scaling of the distance signal is first carried out, and then the rotor position in relation to a coordinate system fixed to the stator is determined with the aid of the known position of the magnetic field sensor and the determined distance information between the magnetic field sensor and the rotor. Here, the coordinate system is usually, but not necessarily, a Cartesian coordinate system. On the basis of this coordinate system, a rotor position signal is generated which represents the rotor coordinates in relation to the coordinate system for each measurement instant. This is preferably a coordinate which describes the rotor position perpendicular to the axis of rotation or also parallel to the axis of rotation.

[0032] The rotary machine can also preferably comprise a control unit which is designed to generate control signals from the rotor position signals. The control unit can for example be a monitoring unit which, for example in the event of excessive rotor vibrations, effects an emergency shutdown of the entire system, causes a change in the rotation speed, operates a control valve or causes a warning signal to be output, for example in an optical or acoustic manner. In addition, a recording of the rotor position signals can be carried out. Alternatively, the control unit can also be designed as an adjustment device which generates control signals which exert one or more forces on the rotor by means of one or more actuators, which influence the rotor vibrations, in particular the position or the speed of the rotor relative to the stationary magnetic field sensors. The actuators can for example be vibration elements which for example transmit vibrations to the stator, whereby for example forces can be transmitted to the rotor by means of the bearings, which can lead to a damping of the rotor vibrations. A prerequisite for the damping action is that the vibrations of the vibration elements are generated in a phase-correct manner. Other actuators for exerting forces on the rotor are for example electromagnets or also piezoelectric actuators.

[0033] It is meaningful for the monitoring, the later signal evaluation or also the error finding that the rotary machine is optionally equipped with or connected to a data collection unit which is designed to store one or more determined rotor position values. The data collection unit can for this purpose be arranged in or at the rotary machine. But it can also be arranged spatially separate from the rotary machine, for example on a remote server.

[0034] The application also relates to a method for a rotating machine using the aforementioned assembly. The main core of the method is that in a first step an electrical signal is measured at the magnetic field sensor and in a second step the electrical signal or a signal derived therefrom is demodulated. The electrical signal can be a current or a voltage generated by the magnetic field sensor. The demodulation can be performed using various methods, for example as envelope demodulation, but also using the rotor rotational frequency in the time domain or in the frequency domain, optionally also using the rotor rotational angle.

[0035] Before the demodulation, an optional intermediate step can be provided in which one or more electrical signals of the aforementioned magnetic field sensors are processed into one or more signals in such a way that the signal portions in the respectively resulting signals which contain information about the distance between the rotor and the respective magnetic field sensor are amplified relative to the other signal portions. This task can be performed, for example, by using a frequency-selective filter, for example like a low-pass filter, or also by a linear combination of the one or more signals of the magnetic field sensors.

[0036] In a further method step it is also possible to determine the rotor position and / or the linear movement speed and / or the linear acceleration of the rotor axis from the demodulated signals. In order to determine the rotor position, the rotor position is preferably determined in the form of coordinates of a coordinate system from the known position of the magnetic field sensors and by means of the determined distance of the rotor to the magnetic field sensors.

[0037] In an optional method step, a control signal can be generated from the rotor position signal. The control signal can be used for monitoring in order to, for example, cause an emergency shutdown or to change certain operating parameters if the vibration values are too high. For example, the control device can cause a change in the rotational speed, control a control valve or cause an output of a warning signal, for example in an optical or acoustic manner. In addition, a recording of the rotor position signal can be initiated.

[0038] In addition, the control signal can also be used to influence the position or the speed of the rotor relative to the stationary magnetic field sensors, for example by means of electromagnetic actuators, piezoelectric actuators or vibration elements, in order to actively counteract the vibrations. The actuators can be, for example, electromagnets in the form of motor coils or also electromagnets of active magnetic bearings. The generation of the control signal can take place, for example, by means of a controller, which has, for example, PID characteristics or which can optionally be supplemented by further filters. The controller can also optionally be designed as a multivariate controller in a state space representation in which the controller parameters are determined by means of an optimization method, for example an H∞-method. The influencing of the position and the speed is intended to influence the center position of the rotor and to suppress the tendency of the rotor to oscillate.

[0039] When the magnetic field sensor is designed as a motor coil with a center tap, the disclosed method is optionally designed to generate one or more control signals for symmetric or asymmetric control of one or more motor coils and to influence the position or the speed of the rotor relative to the stationary magnetic field sensor in this path. In the case of symmetric control, the control current is added to the motor coil of one phase in such a way that the control is at the phase terminals and thus not noticeable for motor control. In the case of asymmetric control, for example when only one motor coil of the phase divider is controlled, the current at the phase terminals caused by the control does not cancel out.

[0040] In another embodiment, a controllable bypass is used which shorts a motor phase or a part of a motor phase by means of a switching element, for example a transistor. The controllable bypass bypasses a part of the motor coil current around the motor coil and thus attenuates the force generated by the stator or the generated torque. The controllable bypass has a control input which specifies the strength of the attenuation. The controllable bypass is an energy-passive system and thus enables the integration of force generation, for example for vibration damping, into the motor without a power supply for the actuator.

[0041] A particular advantage is obtained when a rotor position sensor element which is demodulated with a rotational frequency is combined with an actuator element which is modulated with a rotational frequency. The demodulation and the modulation cancel each other out in this case and can be dispensed with. This effect is referred to as self-mixing. Figures 1 to 15 Sensor elements of the type described in DE 10 2004 019 553 A1 are particularly suitable for this. Sensor elements which are loaded with an external signal are not suitable for this. By dispensing with the demodulator and the modulator, a very compact damping system is obtained which can also be constructed in an energy-passive manner in combination with the controllable bypass Figure 22 ). The price of compactness is that the position signal is not available and cannot be measured or monitored.

[0042] The disclosed method is optionally designed to determine the force or the torque acting on the rotor from the rotor position signal. The rotor position not only describes the position relative to a coordinate system fixed to the stator, but also the position relative to a bearing which is used to support the rotor. For example, rolling bearings, sliding bearings or magnetic bearings can be used. These bearings have a known stiffness, respectively, and thus enable a direct determination of the force with which the rotor is pressed into the bearing by means of a stiffness scaling factor. It is also possible to determine the torque relative to a point fixed to the rotor using the acting force. When an actuator is used to control the rotor position, the force exerted by the actuator must be taken into account when determining the force or the torque acting on the rotor.

[0043] Optionally, the disclosed method can be designed to derive flow characteristic parameters, such as pressure profile, pressure generation or flow, from the rotor position and the torque acting on the rotor, depending on tables stored in the pump controller or simple multi-dimensional approximations with n-stage polynomials, preferably not larger than 4. In addition to this, this can also be done in combination with the speed and power consumption of the motor and the estimated viscosity. Furthermore, the movement of the rotor, in particular its movement frequency, can be used to detect thrombi in the pump range and to estimate the blood viscosity.

[0044] In addition, the force acting on the rotor or the torque acting on the rotor can optionally be evaluated to derive system parameters, such as aging, wear, corrosion or biological growth.

[0045] In addition, it is possible within the scope of the disclosed method to store one or more determined rotor position values in the data collection unit. It is also possible that further secondary data, such as determined forces, torques and further estimated parameters on aging, wear, corrosion or biological growth, are stored in the data collection unit. BRIEF DESCRIPTION OF DRAWINGS

[0046] In the following, embodiments are shown in connection with the enclosed drawings and are explained in the following, in which:

[0047] Figure 1 An overview of the system is shown;

[0048] Figure 2 An example of a rotor and a magnetic field sensor is shown in radial direction;

[0049] Figure 3 An example of a rotor and a magnetic field sensor is shown in axial direction;

[0050] Figure 3 An example of a definition with respect to an axis fixed immobile to the stator is shown;

[0051] Figure 4 An example of a magnetic field sensor with a two-pole rotor designed as motor coil is shown;

[0052] Figure 4 An example of a magnetic field sensor with a four-pole rotor designed as motor coil is shown;

[0053] Figure 4 An example of a definition of an immobile coordinate system is shown;

[0054] Figure 5 An example of a motor coil wiring connection and its use for magnetic field measurement is shown;

[0055] Figure 6 An example of a pre-processing of the magnetic field sensor signals with frequency-selective filtering is shown;

[0056] Figure 6 b shows an example of pre-processing of magnetic field sensor signals with frequency selective filtering and signal combination;

[0057] Figure 6 c shows an example of pre-processing of magnetic field sensor signals with signal combination;

[0058] Figure 7 a shows an example of determining the rotor rotation frequency and the rotor rotation angle by means of a counter component;

[0059] Figure 7 b shows an example of determining the rotor rotation frequency and the rotor rotation angle by means of a motor controller;

[0060] Figure 8 a shows an example of a signal with amplitude modulation in the time domain;

[0061] Figure 8 b shows an example of a signal with amplitude modulation in the frequency domain;

[0062] Figure 9 a shows an example of a signal with amplitude demodulation in the time domain;

[0063] Figure 9 b shows an example of a signal with amplitude demodulation in the frequency domain;

[0064] Figure 10 a shows an example of shielding in the frequency domain;

[0065] Figure 10 b shows an example of frequency shifting in the frequency domain;

[0066] Figure 10 c shows an example of a demodulated signal in the frequency domain;

[0067] Figure 11 shows an example of carrier synthesis by shielding in the frequency domain;

[0068] Figure 12 a shows an example of carrier synthesis by means of an oscillator circuit for amplitude demodulation;

[0069] Figure 12 b shows an example of carrier synthesis by means of a Fourier transform for amplitude demodulation;

[0070] Figure 13 a shows an example of demodulation with the rotor rotation frequency but without the rotor rotation angle in the frequency domain;

[0071] Figure 13b shows an example of demodulation with the rotor rotation frequency in the frequency domain and with the rotor rotation angle;

[0072] Figure 13 c shows an example of demodulation with the rotor rotation frequency in the time domain and with the rotor rotation angle;

[0073] Figure 14 Mapping of demodulated magnetic field sensor signals into a coordinate system is shown;

[0074] Figure 15 An example of using the rotor position in a controllable rotor bearing system is shown;

[0075] Figure 16 a shows coupling of actuator signals into the motor coils;

[0076] Figure 16 b shows modulation of the actuator force in the motor coils with controllable shunts;

[0077] Figure 16 c shows loading of the actuator signals into the center tap of the motor phases by means of additional half bridges;

[0078] Figure 16 d shows loading of the actuator signals into the center tap of the motor phases and loading of compensation signals to suppress the influence of the actuator signals on the motor operation;

[0079] Figure 17 A general control circuit for compensating rotor position disturbances is shown;

[0080] Figure 18 A model for compensating disturbing forces by means of a PID controller and a phase shifter feedback is shown;

[0081] Figure 19 A control circuit for a pump with a synchronous motor is shown, for example, which Figure 17 measures the rotor position by means of Hall sensors and couples the actuator signals itself via a transformer;

[0082] Figure 20 Compensation of disturbances by means of a vibrator as an actuator is shown;

[0083] Figure 21 A damping control circuit with self-mixing function is shown;

[0084] Figure 22 A damping control circuit with self-mixing function is shown, which is realized in an energy-passive manner;

[0085] Figure 23 Amplitude characteristics of dynamic systems with and without active damping are shown, for example. DETAILED DESCRIPTION

[0086] Figure 1 The basic system components, i.e. the rotor position detection device 300, the control unit 6 and the data collection unit 8 are shown. The measuring device 1 is designed, inter alia, to also detect one or more magnetic fields or changes in the magnetic field superimposed into the motor signal, to convert this into a current signal or a voltage signal and to bring the voltage signal into a form in which it can be further processed. The measuring device 1 for this purpose comprises magnetic field sensors 12, for example like electrical connectors and terminals and optionally a measuring amplifier, a level converter or an impedance converter. At the output of the measuring device 1 the electrical signals 101, 102, 103, 104, 105, 106, 107 or 110 are provided for use by the optional signal processor 2. The signal processor 2 performs a signal pre-processing, in the course of which the signal portions containing the rotor position information or information about the distance between the magnetic field sensors and the rotor are relatively amplified with respect to other signal portions. This can be achieved, for example, by frequency-selective filtering, by leaving only the relevant information in the signal, or also by linearly combining a plurality of measurement signals. The output signal 120 of the signal processor 2 is optionally converted into a digital signal using the analog-digital converter 3. It is possible in principle to integrate the analog-digital converter 3 into the signal processor 2, so that further signal processing methods can optionally also be added after the analog-digital converter 3. The output signal 130 of the analog-digital converter 3 serves as an input signal for the demodulator 4, which is designed to perform amplitude demodulation. Depending on whether the optional analog-digital converter 3 is used, the demodulation is performed in analog or digital fashion. It is also optionally possible to use the rotor rotation frequency 171 and / or the rotor rotation angle 172 in the amplitude demodulation by means of the device 7 for providing the rotor rotation frequency 171 and optionally the rotor rotation angle 172. At least one component of the rotor position, preferably a plurality of components, in particular at least an adjustable rotor position component, is determined from the demodulated signal 140 in the device 5. The output signal 150 is used in the optional control unit 6 to generate a control signal 160. Optionally, the measured and / or calculated signals and data can be stored in the data collection unit 8.

[0087] Figure 2A component is shown, which comprises a rotor 11, magnetic field sensors 12 in a Cartesian coordinate system 14. In principle, the coordinate system 14 is defined fixed with respect to the stator 13, in particular in a plane perpendicular to an axis 20 fixed with respect to the stator. The rotor axis of rotation theoretical position is generally placed at the coordinate origin of the coordinate system 14. The magnetic field sensors 12 are arranged in the coordinate system 14 axes in this example, which can be traversed by the magnetic field generated by the rotor. The rotor 11 contains at least in the axial section of the rotating machine, where the magnetic field sensors 12 are also located, a magnet, which is designed as a permanent magnet with a north pole 26 and a south pole 27 in this example, but can also be an electromagnet. Correspondingly, an even number of magnetic poles are distributed over the circumference 18 of the rotor 11, so that the magnetic field sensors 12 encounter a periodically alternating magnetic field when the rotor rotates. The periodic frequency of the alternating magnetic field corresponds to the product of the rotor rotation frequency 171 and the even number of rotor poles. In this example, the magnetic field sensors 12 are designed as coils, so that a voltage Vx or Vy is induced in the coils 12, respectively, due to the alternating magnetic flux, which is proportional to the magnetic flux change. The induced voltage Vx or Vy thus depends not only on the rotor rotation frequency 171, but also on the distance 141 of the rotor 11 to the respective coil 12, which represents the radial distance of the rotor 11 with respect to the fixed axis 20 in this example.

[0088] Figure 3 a shows a component with a rotor 11 and magnetic field sensors 12. The rotor 11 contains at least on one end side a magnet 25 with at least a north pole 26 and a south pole 27, so that a magnetic field sensor 12 spaced axially from the rotor 11 can be traversed by the alternating magnetic field of the magnet 25 when rotating. The magnetic field sensors 12 are designed as coils 12 in this view. The coils 12 can also be designed as motor coils. According to the induction law, a voltage proportional to the magnetic flux change is induced in the coils 12. The induced voltage thus depends not only on the rotor rotation frequency 171, but also on the distance 140 of the rotor 11 from the coils 12, which indicates the axial position of the rotor with respect to the fixed position of the coils 12. Figure 3 a also shows an axis 20 defined fixed with respect to the stator 13. The rotor axis 19' is located in the geometric center of the rotor cross section. The axis of rotation 19 can deviate from the rotor axis 19' and the axis 20 defined fixed with respect to the stator, in particular when rotating, caused by external forces or unbalances.

[0089] Figure 3 b shows a partial view of the rotor 11 in the area of a possible bearing 21. The bearing can be, for example, a rolling bearing, a sliding bearing or a magnetic bearing. The position of the axis 20 fixed with respect to the stator 13 is defined, for example, by the angle between the bearing plane 22 and the axis 20 and by the penetration point 23 of the axis 20 through the bearing plane 22 in the center of the bearing, wherein the center of the bearing is defined by the same spacing r24 from diametrically opposite elements of the bearing 21.

[0090] Figure 4 a shows an axial view of an assembly consisting of a rotor 11 and a stator 13 typical for a motor configuration. A magnetic field sensor 12 is shown, which can be used in motor applications simultaneously as a drive coil or motor coil, which are designed to generate one or more magnetic fields to generate a torque. The coils are characterized by the pairs A1-A2, B1-B2 and C1-C3 in opposition. In motor applications, each pair can be electrically connected in series. The rotor 11 contains an assembly of magnets 25 with pole pairs consisting of a north pole 26 and a south pole 27.

[0091] Figure 4 b shows an axial view of an assembly consisting of a rotor 11 and a stator 13 typical for a motor configuration similar to a. Different from Figure 4 a, the rotor 11 contains an assembly of magnets with two pole pairs, i.e. two north poles 26 and two south poles 27. In principle, more than two rotor pole pairs can also be employed, which consist of one north pole 26 and one south pole 27, respectively, or more or less than the three coil pairs depicted. Figure 4

[0092] Figure 4 c shows an axial view of an assembly consisting of a rotor 11 and a stator 13 and a coordinate system 14 with two mutually orthogonal axes defined. In addition, the axes defined by the coil pairs A1-A2, B1-B2 and C1-C2 are drawn, which deviate from the x-axis of the coordinate system 14 by the angles a 15, β 17 and γ 16. The angles 15, 16 and 17 can be used to convert the distance 141 measured with respect to the sensor 12 into the coordinates of the coordinate system 14.

[0093] Figure 5 ​A wiring connection of a motor coil and its use for magnetic field measurement (i.e. as magnetic field sensor 12) is shown. A star circuit is shown, where the motor is electrically connected to phase terminals 2101, 2102 and 2103 and to a center tap terminal 2104 for motor operation. The individual phase terminals 2101, 2102 and 2103 are terminals for one electrical phase 1101, 1102 and 1103 of the motor, which are composed of the coil pairs Al-A2, Bl-B2 and Cl-C2, respectively. The phases 1101, 1102 and 1103 are electrically connected to each other at the center tap terminal 2104. Further, center tap terminals 2105, 2106 and 2107 are defined. Further, voltages 101, 102, 103, 104, 105, 106 and 107 are measured, for example, to ground, respectively. In addition to the voltage or current applied by the motor controller via the phase terminals 1101, 1102 and 1103, a voltage is induced in the coils 12 upon rotation of the rotor 11, which decreases across the coils 12 Al, A2, Bl, B2 and Cl and C2, respectively. The diagram thus shows an example of an electrical wiring connection for a magnetic field sensor 12 when designed as a motor coil. The motor coil is thus part of the measuring device 1. The voltages 101, 102, 103, 104, 105, 106, 107 and the voltages 110, Vai, Va2, Vbi, Vb2, Vci and Vc2, which decrease across the coils 12, respectively, can be subsequently processed, for example, by electrical connectors and terminals and optionally by measurement amplifiers, level converters or impedance converters.

[0094] The signal processor 2 is shown in Figure 6 part of the diagram.

[0095] Figure 6a The signal processor 2 is shown here with a filter 31 which filters the coil voltage 110, which filter is designed as a low-pass filter. By low-pass filtering, the high-frequency signal portions which are not relevant for the rotor position determination are removed from the coil voltage signal 110. Irrelevant signal portions can be, for example, the pulse-width modulation (PWM) signal of the motor controller, which can have a high switching frequency of, for example, 4 kHz or 8 kHz and whose harmonic components can extend up to the megahertz range. The low-pass filter which is designed to suppress the pulse-width modulation signal portions of the motor controller can have, for example, a cut-off limit frequency of 3.9 kHz, where the cut-off limit frequency is the frequency above which signal components in the signal are suppressed. The pass limit frequency of such a low-pass filter can be determined, for example, such that the signal components relevant for the rotor position determination can pass the filter undamped or with a weak damping. The minimum pass limit frequency in this example results from the sum of the carrier oscillation frequency 176 and the maximum frequency of the modulation signal 179. Alternatively, other filters 31 can also be employed here, for example band-pass filters, which can have, for example, the cut-off limit frequency of the low-pass filter as upper pass limit frequency and upper cut-off limit frequency. The lower pass limit frequency of the band-pass results in this example from the carrier oscillation frequency 176 minus the maximum frequency of the modulation signal 179. The lower cut-off frequency must be smaller than the lower pass limit frequency, but can otherwise be freely chosen, for example.

[0096] Figure 6 b The signal processing 2 is shown here as being achieved by a combination of a plurality of measurement signals. The voltages 101 at phase A, 104 at the junction and 105 at the center tap A, for example, are combined. All three voltages 101, 104, 105 are measured to ground, for example. First, all three voltage signals 101, 104 and 105 are filtered with a filter 31 which is designed as a low-pass filter to remove the PWM signal portions, after which they are then added together with a weighting. With this circuit, for example, the motor signal portions can be removed from the voltage signals, since with this circuit the difference of the voltages Va1-Va2 applied in the motor is determined. The two voltages are in the case of a motor with two phases A1 and A2 which are assumed to be identical, back to the voltage Vm of the motor controller and to the voltage Vp of the modulated rotor position signal, but they have different signs because of the opposite arrangement of the coils A1 and A2. Figure 5 Figure 6 b The circuit shown in b thus calculates the voltage as follows:

[0097] (Va1-Va2) / 2 = ((Vm+Vp)-(Vm-Vp)) / 2 = 2Vp / 2 = Vp.

[0098] The formula result is also observed for a two-pole rotor.

[0099] ​The example shows that the combination of the plurality of coil voltages 110 can be advantageously used in some embodiments to reduce signal portions that are not relevant for the rotor position determination. Signal components that are attributed to the motor controller are suppressed in the processed signal 120.

[0100] Figure 6 c shows Figure 6 b shows a possible way of actual conversion of the signal processor 2 with transformer. Active computing circuits with operational amplifiers known from the literature provide an alternative to the transformer circuit.

[0101] Figure 7 a shows one possible implementation of the device 7 for providing the rotor rotation frequency 171 and optionally the rotor rotation angle 172. As main component of the device 7 a counter assembly 174 is shown, for example, which increments an internal counter in regular cycles. At the input the counter assembly 174 obtains a signal, for example a voltage pulse, generated by the key phase 173. The voltage pulse generated by the key phase 173 can be converted by further components, for example, to a voltage pulse that has a uniform predetermined voltage, for example 5 V, and that also has the same length, for example 50 μs. A voltage pulse is always generated by the key phase 173 when a key phase slot passes the key phase 173 in one revolution of the rotor 11. The number of voltage pulses generated per revolution of the rotor is therefore as many as the number of key phase slots arranged on the rotor. The further processing of the voltage pulses in the case of one key phase slot exactly provided on the rotor 11 is explained by way of example. In the counter assembly 174 the internal counter is reset to zero, for example, at the rising signal edge of the voltage pulse, and the counting process including the clock of the counter is started. The clock frequency is set in such a way that even at maximum speed a large number of increments is performed per revolution of the rotor. When the same edge of the subsequent voltage pulse arrives, the currently present counter value is stored, the counter is called back to zero and the counting process is restarted. The rotor rotation frequency 171 is determined by dividing the clock frequency by the stored counter value. In order to estimate the rotation angle of the rotor during the subsequent rotation, for example, the current counter value is divided by the stored counter value and multiplied by 360°. The rotor rotation frequency 171 and the rotor rotation angle 172 are provided at the output of the device 7.

[0102] Figure 7 b shows an alternative device 7 for providing the rotor rotation frequency 171 and optionally the rotor rotation angle 172, which can be implemented in some motors. The rotor rotation frequency 171 and the rotor rotation angle 172 are present in the motor in this case, for example, and are provided by the motor controller 175.

[0103] Figure 8a shows an example of a signal for generating an amplitude modulated signal 180 in the time domain. The amplitude modulated signal 180 results from a carrier oscillation 178 multiplied by a modulation signal 179. The frequency of the carrier oscillation 178 is called carrier oscillation frequency 176. The carrier oscillation frequency 176 is visible in the amplitude modulated signal 180. In addition, the amplitude of the amplitude modulated signal 180 fluctuates synchronously with the modulation signal 179. In the disclosed rotary machine, the modulation signal 179 corresponds to the distance 141 between the magnetic field sensor 12 and the rotor 11. The varying magnetic flux within the magnetic field sensor 12 caused by the rotor magnetic poles 26, 27 results in a carrier oscillation signal in the magnetic field sensor 12.

[0104] Figure 8 b shows an example of a signal for generating an amplitude modulated signal 180 in the frequency domain. The spectrum (i.e. the Fourier transform) of the modulation signal 179 is convoluted with the spectrum (i.e. the Fourier transform) of the carrier oscillation 178. The result is the spectrum, i.e. the Fourier transform, of the amplitude modulated signal 180. The spectrum of the amplitude modulated signal 180 shows that the spectrum of the modulation signal 179 is shifted to the left, i.e. towards negative frequencies, by the carrier oscillation frequency 176 and to the right, i.e. towards positive frequencies, by the carrier oscillation frequency 176, respectively.

[0105] Figure 9 a shows an example of a signal for amplitude demodulation of the amplitude modulated signal 180 in the time domain. The amplitude modulated signal 180 is multiplied by the carrier oscillation 178 and then filtered with a low pass filter. As a result, the demodulated signal 179 appears. The necessity of low pass filtering becomes more clear in Figure 9 b.

[0106] Figure 9 b shows an example of a signal for amplitude demodulation of the amplitude modulated signal 180 in the frequency domain. Its spectral components, which are distributed around the carrier oscillation frequency 176, are convoluted with the spectrum (i.e. the Fourier transform) of the carrier oscillation 178. As a result, the modulation signal (see Figure 8 b) spectrum can be seen in three ranges. In order to suppress the two spectral regions not at 0 Hz and to restore the spectrum of the modulation signal 179, the signal is filtered with a low pass filter (dashed area).

[0107] Figure 10 An example of a demodulation method in the frequency domain is shown.

[0108] Figure 10 a shows the spectrum of the amplitude modulated signal 180. It is distributed around the carrier oscillation frequency 176, respectively. For demodulation in the frequency domain, all signal components (i.e. Fourier coefficients) not belonging to the amplitude modulated signal are shielded, i.e. brought to zero, by means for shielding 43.

[0109] The true demodulation isFigure 10 The signal part remaining after the shielding is shifted from the initial position to the 0 Hz frequency, i.e. copied and cancelled at the initial point. The shift takes place with a frequency value corresponding to the carrier oscillation frequency 176. The coefficients located at the same bit point after the shift are added.

[0110] Figure 10 c shows the spectrum occurring by the described approach. It corresponds to the spectrum of the modulation signal 179.

[0111] Figure 11 An example of a software-based carrier synthesis 201 by shielding in the frequency domain is shown. To this end, the amplitude modulation signal 180 is shielded in the frequency domain such that all signal parts not belonging to the carrier oscillation 178, i.e. signal parts having no frequency component containing the carrier oscillation frequency 176, are reduced to zero by means for peak identification and shielding 204. The signal shielded in this way is transformed into the time domain in the scope of the software-based carrier synthesis 201. If the carrier oscillation frequency 176 is not known, it can be estimated by means 204, for example by peak identification.

[0112] Figure 12 a shows a possible way in which a carrier oscillation can be generated with an electronic component for carrier synthesis 200. Such a component 200 can be, for example, an electronic oscillator circuit, in which the carrier oscillation 178 uses the set conditions for the frequency and phase resulting from the rotor rotation frequency 171 and the rotor rotation angle 172. As an alternative, the carrier synthesis 200 can take place by means of a microcontroller or computer, i.e. a signal stored in a memory is called up and converted into a voltage signal using a digital-to-analogue converter. In some embodiments, this method has the advantage of higher flexibility and easier configurability. The carrier oscillation frequency 176 generated by the carrier synthesis 200 here comes from the product of the rotor rotation frequency and the rotor pole evenness. It is calculated in a component 205 for calculating the carrier oscillation frequency 176. The phase 177 to be generated is calculated in a component 206 for calculating the phase 177. It has to be adjusted individually for each magnetic field sensor signal or signal to be demodulated. It depends on the position of the magnetic field sensor and the rotor rotation angle reference point fixed to the stator, the current position of the rotor rotation angle reference point fixed to the rotor and the rotor pole evenness.

[0113] Figure 12b illustrates an example of software-based carrier synthesis 201 using Discrete Fourier Transform (DFT). For this purpose, a signal 130 converted to the digital range is transformed to the frequency domain, for example, by an FFT transform unit 202 designed to perform a Fast Fourier Transform (FFT). The transformed signal is then transferred to a unit 204 for peak identification and masking, designed to identify peaks within the Fourier transform and mask frequency bands or individual frequencies, reducing them to zero. Unit 204 masks all regions of the Fourier transform that are not located within a predetermined range around the carrier oscillation frequency 176. The carrier oscillation frequency 176 is derived from the product of the rotor rotation frequency 171 and the rotor pole evenness. The masked signal is then transformed to the time domain in an iFFT unit designed to perform an Inverse Fourier Transform (iFFT). The result is a carrier oscillation signal with a carrier oscillation frequency 176 and a phase 177 suitable for demodulation.

[0114] Figure 13 Different examples of demodulators in the time and frequency domains are shown.

[0115] Figure 13 Example 4 of demodulation in the frequency domain is shown, where the rotor rotation frequency 171 is used but the rotor rotation angle 172 is not. Here, the signal 130, which exists in digital form, is transformed to the frequency domain by means of a Fast Fourier Transform (FFT) 41. Masking 43 is performed in the Fourier transform, i.e., all frequency components not located around the carrier oscillation frequency 176 are reduced to zero. The carrier oscillation frequency 176 is provided by unit 205 for calculation. It is derived from the product of the rotor rotation frequency 171 and the rotor pole evenness. It is important to note that masking 43 is performed not only for the frequency reference corresponding to the positive frequency but also for the frequency reference corresponding to the negative frequency. Additionally, the remaining frequency components are shifted 44 towards 0Hz by the magnitude of the carrier oscillation frequency 176. Here, the shifted frequencies are respectively compounded to the frequency portions already present around 0Hz. The resulting signal is transformed back to the time domain by means of an inverse Fast Fourier Transform. The resulting signal 140 does not contain correct phase information and therefore cannot be used to calculate the rotor position. However, its power can be used to estimate the current vibration level of the rotor.

[0116] Apart from Figure 13 In addition to the processing level shown in a, there is also Figure 13 In example b, phase correction is performed. Here, phase correction is performed for each Fourier coefficient shifted toward 0Hz, such that the value of the corrected phase is derived from the difference between the uncorrected phase and the phase value at the carrier oscillation frequency of 176.

[0117] Figure 13c shows an example of the demodulation in the time domain by means of the carrier oscillation frequency 176 and the phase 177. In this case, the signal 130 present in digital form is demodulated by means of a multiplier 46 and a subsequent low-pass filter using a cosine signal with the carrier oscillation frequency 176 and the estimated phase 177 synthesized by the components for the carrier synthesis 201. This low-pass filter can be realized not only as a digital filter in the time domain, but also in the frequency domain by masking the frequency components to be removed.

[0118] Figure 14 An example of an apparatus 5 for calculating the rotor position is shown. This apparatus 5 comprises as a first processing stage a rotational speed-dependent scaling 51 which calculates a rotational speed-dependent correction factor using the rotor rotation frequency 171 and the rotor pole evenness information. This correction factor takes into account that the induced voltage is proportional to the change in magnetic flux when, for example, the magnetic field sensor is designed as a coil. But the change in magnetic flux is directly related to the rotor rotation frequency 171, so this influence has to be eliminated when calculating the distance between the magnetic field sensor and the rotor. The rotational speed-dependent scaling 51 can be designed here, for example, as a table or a family of characteristic curves and implemented in a computer or microprocessor.

[0119] The distance values corrected in this way are subsequently used to determine the rotor position. For this purpose, the deviation from a predetermined theoretical value is calculated for each distance value by subtracting the predetermined theoretical value from the corrected distance value. Since the angles a 15, β 17 and γ 16 which indicate the angular misalignment of the axes defined by the magnetic field sensors 12 with respect to the coordinate system 14 are known, the rotor position can be mapped from the respective magnetic field sensor axes onto the coordinate system axes of the coordinate system 14 by means of trigonometric relationships. The coordinates in this coordinate system 14 obtained from the different magnetic field sensors 12 can be combined, for example, by averaging.

[0120] Figure 15 An example is shown in accordance with which the rotor position signal 150 can be used to control an actuator. Of interest in this example is that the motor coil which is needed anyway for generating the drive magnetic field is used as a magnetic field sensor 12 and also as an actuator. Figure 15The detection of the rotor position is shown for example in the phase branch C 1103 with the voltages measured on the coils C1 and C2. A controller 400, which can optionally be supplemented by further transmission elements, for example a PID controller, calculates a control signal for holding the rotor in a certain position or for damping the oscillating behavior of the rotor. This control signal is transformed with the amplitude modulator 260 in such a way that they have a phase with respect to the rotor rotation angle 172 which leads to a stable regulation loop. The modulation signals are applied to the primary winding 242 of the transformer by means of controllable current sources 243, which together with the secondary windings 240, 241 form the actuator-coupling unit 210. By means of the secondary windings 240 and 241, a current 230 is applied to the coil Al and a current 231 to the coil A2, which in this example are oriented in such a way that they cancel each other out at the phase connection A 2101 and thus also have no influence on the voltages and currents at the other phase connections 2102 and 2103. The symmetrical manner in which the currents are added to the phase A 1101 can alternatively be replaced by an asymmetrical manner for loading the currents, in that for example the actuator-coupling unit 210 is only coupled to one of the coils 12 Al or A2.

[0121] The following solutions are inter alia contained herein:

[0122] 1. A rotary machine, having

[0123] - a stator (13) and a rotatably mounted rotor (11) designed to move relative to the stator (13), wherein one or more magnetic field sensors (12) are fixedly arranged relative to the stator (13) at a radial distance from an axis (20) fixed relative to the stator (13),

[0124] - at least one measuring device (1) designed to detect magnetic field changes by means of the magnetic field sensors (12),

[0125] - a rotor (11) designed to generate one or more electrical signals (101, 102, 103, 104, 105, 106, 107, 110) with one or more constant magnetic source voltages and one or more of the magnetic field sensors (12) respectively, the electrical signals having signal components corresponding to a rotor rotation frequency (171) and a respective distance between the magnetic field sensor (12) and the rotor (11),

[0126] characterized in that

[0127] - a demodulator unit (4) is provided, which is designed to demodulate a signal (101, 102, 103, 104, 105, 106, 107, 110, 120, 130) having signal components corresponding to the rotor rotation frequency (171) and to the respective distance between the magnetic field sensor (12) and the rotor (11) generated by the magnetic field sensor (12) or derived therefrom, so as to generate a signal (140) corresponding to the distance between the rotor (11) and the magnetic field sensor (12) corresponding to the respective signal.

[0128] 2. Rotating machine according to scheme 1, characterized in that the rotating machine is an electric machine.

[0129] 3. Rotating machine according to one of the preceding schemes, characterized in that the magnetic field sensor (12)

[0130] - is designed as a coil, and / or

[0131] - is designed as a machine coil and is designed for detecting a magnetic field change and generating a magnetic field suitable for driving the rotor.

[0132] 4. Rotating machine according to one of the preceding schemes, characterized in that:

[0133] - a device (7) is provided which is designed to provide the rotor rotation frequency (171),

[0134] - the demodulator unit (4) is designed to use the rotor rotation frequency (171) in the demodulation.

[0135] 6. Rotating machine according to one of the preceding schemes, characterized in that a first processing unit (2) is provided, which is designed to superimpose and / or filter one or more electrical signals (101, 102, 103, 104, 105, 106, 107, 110) of the aforementioned magnetic field sensors (12) into one or more signals (120) in such a way that a signal portion in the respective resulting signal containing information about the distance (141) between the rotor (11) and the respective magnetic field sensor (12) is amplified compared to other signal portions.

[0136] 6. Rotating machine according to one of the preceding schemes, characterized in that:

[0137] - a second processing unit (5) is provided downstream of the demodulator (4) and is designed to generate one or more rotor position signals (150) from the demodulated signal (140), and

[0138] - a control unit (6) is preferably provided, which is designed to generate a control signal (160) from the rotor position signal (150).

[0139] 7. Rotary machine according to one of the preceding aspects, characterized in that a data collection unit (8) is provided, which is designed to store one or more determined position values of the rotor (150).

[0140] 8. Method in the case of use of a device according to one of the preceding aspects, characterized in that

[0141] - one or more electrical signals (101, 102, 103, 104, 105, 106, 107, 110) are measured at the magnetic field sensors (12), and

[0142] - one or more signals (101, 102, 103, 104, 105, 106, 107, 110, 120, 130) measured at the magnetic field sensors (12) or derived therefrom are demodulated.

[0143] 9. Method according to aspect 8, characterized in that the rotor rotation frequency (171) is used for the demodulation and preferably the rotor rotation angle (172) is used for the demodulation.

[0144] 10. Method according to one of aspects 8 to 9, characterized in that one or more electrical signals (101, 102, 103, 104, 105, 106, 107, 110) of the magnetic field sensors (12) are processed to one or more signals (120) in such a way that a signal portion of the respective resulting signal, which contains information about the distance (141) between the rotor (11) and the respective magnetic field sensor (12), is amplified compared to other signal portions.

[0145] 11. Method according to one of aspects 8 to 10, characterized in that at least one or more components of the rotor position (150) and / or the linear movement speed and / or the linear acceleration of the rotor axis (19') are determined from the demodulated signals (140).

[0146] 12. Method according to aspect 11, characterized in that a control signal (160) is generated from the rotor position signal (150).

[0147] 13. Method according to one of aspects 11 to 12, characterized in that a force acting on the rotor and / or a torque acting on the rotor is determined from the rotor position signal (150).

[0148] 14. Method according to one of aspects 11 to 13, characterized in that one or more determined rotor position values (150) are stored in a data collection unit (8).

[0149] 15. Blood pump system with a rotary machine according to one of aspects 1 to 7.

[0150] The several features shown in Figures 16 to 23 the drawings are also embodied in the following claims:

[0151] 1. A rotating machine, having

[0152] - a stator (13), and

[0153] - a rotatably mounted rotor (11) designed to move relative to the stator (13), wherein one or more magnetic field sensors (12) are fixedly arranged relative to the stator (13) at a radial distance from an axis (20) fixed relative to the stator (13), and

[0154] - at least one rotor position detection device (300) designed to provide one or more signals (101, 102, 103, 104, 105, 106, 107, 110, 150) depicting the spatial position of the rotor (11) relative to one or more spatial reference points (10) of the stator (13),

[0155] characterized in that

[0156] - at least one actuator-coupling unit (210) is provided, which is designed to apply one coil current (162) to at least one of the coils (12), respectively, wherein the respective applied coil current (162) contains at least one amplitude-modulated signal component (161), whose carrier oscillation frequency (176) corresponds to the product of the even number of poles of the rotor (11) and the rotor rotation frequency (171), and whose modulation signal (179) is composed of one or more rotor position signals (150) and / or control signals (160) designed to influence the spatial position of the rotor (11),

[0157] - coils (12) are provided, which are designed to generate one or more magnetic fields acting on the rotor (11) corresponding to the respective applied coil current (162), and

[0158] - the rotor (11) is designed to demodulate the magnetic fields corresponding to the applied coil currents (162) into one or more translational forces acting on the rotor (11) using one or more magnetic fields having a constant magnetic source voltage fixed relative to the rotor (11), respectively.

[0159] 2. The rotating machine according to claim 1, characterized in that the rotating machine is an electric machine.

[0160] 3. The rotating machine according to claim 1, characterized in that the rotating machine is a driven rotating machine.

[0161] 4. Rotary machine according to solution 1 or 2, characterized in that the coils (12) are designed as motor coils.

[0162] 5. Rotary machine according to solution 4, characterized in that one or more of the actuator-coupling units (210) are designed to apply a coil current (162) into at least one motor coil (12) such that the current uniformity in the motor coils (12) belonging to the respective motor phase (1101, 1102, 1103) is cancelled, whereby at least one force is applied to the rotor (11).

[0163] 6. Rotary machine according to solution 4 or 5, characterized in that one or more of the actuator-coupling units (210) are designed to apply a coil current (162) into at least two coils (12) of one phase such that the applied coil currents (162) cancel each other at the corresponding phase connection (2101, 2102, 2103) and at the junction point (2104).

[0164] 7. Rotary machine according to one of the preceding solutions, characterized in that one or more of the spatial reference points (10) used for the rotor position detection (300) are arranged at other locations in space than one or more of the coils (12).

[0165] 8. Rotary machine according to one of the solutions 4 to 6, characterized in that the rotor position detection (300) is performed on the basis of one or more of the signals (101, 102, 103, 104, 105, 106, 107, 110) at the motor coils (12) of a first motor phase (1101, 1102, 1103) and that the actuator-coupling unit (210) acts on the motor coils (12) of a second motor phase (1101, 1102, 1103).

[0166] 9. Rotary machine according to one of the solutions 1 to 8, characterized in that an actuator-coupling unit (210) is provided which is designed to inductively couple a coil current (162) into one or more coils (12) with a coil (240, 241, 242).

[0167] 10. Rotary machine according to one of the solutions 1 to 8, characterized in that an actuator-coupling unit (210) is provided which is designed to couple a coil current (162) into one or more coils (12) with a controllable bypass (245, 246).

[0168] 11. Rotary machine according to one of the preceding solutions 1 to 8, characterized in that an actuator-coupling unit (210) is provided, which is designed to couple a coil current (162) into one or more coils (12) with one or more controllable current sources (247, 248).

[0169] 12. Rotary machine according to one of the preceding solutions, characterized in that

[0170] - a control device (6) is provided, which is designed to generate one or more control signals (160) from one or more signals (101, 102, 103, 104, 105, 106, 107, 110, 150) depicting the spatial position of the rotor (11) relative to one or more spatial reference points (10) of the stator (13),

[0171] - a modulator unit (260) is provided, which is designed to generate one or more amplitude-modulated signal components (161).

[0172] 13. Rotary machine according to one of the preceding solutions 1 to 11, characterized in that a transmission member (290) is provided, which is designed to change one or more electrical signals (101, 102, 103, 104, 105, 106, 107, 110) measurable at the coils (12) into amplitude-modulated signal components (161) with a linear transmission function and / or to superimpose them.

[0173] 14. Rotary machine according to one of the preceding solutions, characterized in that a data collection unit (8) is provided, which is designed to store one or more signals (101, 102, 103, 104, 105, 106, 107, 110, 150) depicting the spatial position of the rotor (11) relative to one or more spatial reference points (10) of the stator (13) and / or the corresponding control signals (160).

[0174] 15. Method in the use of a device according to one of the preceding solutions, characterized in that the rotor resonance is damped.

[0175] 16. Blood pump system, having a rotary machine according to one of the preceding solutions 1 to 15.

Claims

1. A rotary machine, having: - a stator (13) and a rotatably mounted rotor (11), which is designed to move relative to the stator (13), wherein one or more magnetic field sensors (12) fixed immovably at a radial distance from an axis (20) fixed immovably with respect to a stator (13), - at least one measuring device (1) designed to detect magnetic field changes by means of the magnetic field sensors (12), - a rotor (11) designed to generate one or more electrical signals with one or more constant magnetic source voltages and one or more of the magnetic field sensors (12) respectively, the one or more electrical signals having signal components corresponding to a rotor rotation frequency (171) and the respective distance between the magnetic field sensor (12) and the rotor (11), characterized in that - a device (7) is provided which is designed to provide the rotor rotation frequency (171), - a demodulator unit (4) is provided which is designed to demodulate signals generated by the magnetic field sensors (12) or derived therefrom having signal components corresponding to the rotor rotation frequency (171) and the respective distance between the magnetic field sensor (12) and the rotor (11) using the rotor rotation frequency (171), so as to generate signals (140) corresponding to the distance between the rotor (11) and the magnetic field sensor (12) corresponding to the respective signal.

2. The rotary machine of claim 1, wherein, The at least one magnetic field sensor (12) is designed as a motor coil and is designed to detect magnetic field changes and to generate a magnetic field suitable for driving the rotor.

3. The rotary machine according to claim 1 or 2, characterized in that, The at least one magnetic field sensor (12) is designed as a coil.

4. The rotary machine of claim 1, wherein, The rotary machine is a motor.

5. The rotary machine of claim 1, wherein, A first processing unit (2) is provided which is designed to superimpose and / or filter the one or more electrical signals of the magnetic field sensors (12) into one or more signals (120) such that signal portions containing information about the distance (141) between the rotor (11) and the respective magnetic field sensor (12) are amplified in the respective resulting signal compared to other signal portions.

6. The rotary machine of claim 1, wherein There is provided: - a second processing unit (5) downstream of the demodulator unit (4) and designed to generate one or more rotor position signals from the demodulated signals (140), and - a control unit (6) designed to generate a control signal (160) from the rotor position signals.

7. The rotary machine of claim 1, wherein A data collection unit (8) is provided which is designed to store one or more determined position values of the rotor (11).

8. A method in the case of use of a rotary machine according to any one of claims 1 to 7, characterized in that - one or more electrical signals are measured at the magnetic field sensors (12), and - one or more signals measured at the magnetic field sensors (12) or derived therefrom are demodulated.

9. The method of claim 8, wherein, The rotor rotation frequency (171) is used for demodulation and the rotor rotation angle (172) is used for demodulation.

10. The method according to claim 8 or 9, characterized in that, The one or more electrical signals of the magnetic field sensors (12) are processed into one or more signals (120) such that signal portions containing information about the distance (141) between the rotor (11) and the respective magnetic field sensor (12) are amplified in the respective resulting signal compared to other signal portions.

11. The method of claim 8, wherein, At least one or more components of the rotor position and / or the linear movement speed and / or the linear acceleration of the rotor axis (19') are determined from the demodulated signals (140).

12. The method of claim 11, wherein, A control signal (160) is generated from the rotor position signals.

13. The method according to claim 11 or 12, characterized in that, A force acting on the rotor and / or a torque acting on the rotor is determined from the rotor position signals.

14. The method of claim 11, wherein, One or more determined rotor position values are stored in a data collection unit (8).

15. A blood pump system having a rotary machine according to any one of claims 1 to 7.

Images